Calculus Concepts and Applications
Uploaded by
AnkurCalculus Concepts and Applications
Uploaded by
AnkurConcepts and Applications
PAUL A. FOERSTER
Editor: Bill Medigovich
Mathematical Reviewers: Cavan Fang, Leslie ielsen, Loyce Collenback
Art Development : Casey FitzSimons
Editorial Assistant: Romy Snyder
Copyeditors: Greer Lleuad, Luanna Richards
Production Editors : Deborah Cogan, Joe Todaro
Production Service: Greg Hubit Bookworks
Production Manager: Luis Shein
Text Design: Terry Lockman, Lumina Designworks
Cover Design: Maryann Ohki
Technical Art: Jason Luz, Ann Rothenbuhler
Photo Research: Ellen Hayes
Composition: Peter Vacek, Eigentype Compositors
Cover Photograph: Images 1995 PhotoDisc, Inc.
Publisher: Steve Rasmussen
Editorial Director: John Bergez
1998 by Key Curriculum Press
All rights reserved. No part of this publication may be reproduced, stored
in a retrieval system, or transmitted, in any form or by any means, electronic,
photocopying, recording, or otherwise, without the prior written permission
of the publisher. Requests for permission should be made in writing to:
Key Curriculum Press, 1150 65 th Street, Emeryville, California 94608.
ISBN 1-55953-117-7
Printed in the United States of America
10987654 02 01 00 99
The graphs in this text were created using PSMathGraphs II. PSMathGraphs is a
trademark of MaryAnn Software.
Photo credits appear on the last page of the book.
Consultants
Donald J. Albers, Mathematical Association of America, Washington, D.C.
Judith Broadwin, Jericho High School, Jericho, New York
Joan Ferrini-Mundy, University of New Hampshire, Durham, New Hampshire
Gregory D. Foley, Sam Houston State University, Huntsville, Texas
John Kenelly, Clemson University, Clemson, South Carolina
Dan Kennedy, Baylor School, Chattanooga, Tennessee
Deborah B. Preston, Keystone School, San Antonio, Texas
Field Testers
Betty Baker, Bogan High School, Chicago, Illinois
Glenn C. Ballard, William Herny Harrison High School, Evansville, Indiana
Bruce Cohen, Lick-Wilmerding High School, San Francisco, California
Christine J. Comins, Pueblo County High School, Pueblo, Colorado
Deborah Davies, University School of Nashville, Nashville, Tennessee
Linda E. de Sola, Plano Senior High School, Plano, Texas
Paul A. Foerster, Alamo Heights High School, San Antonio, Texas
Joan M. Gell, Palos Verdes Peninsula High School, Rolling Hills Estates, California
Valmore E. Guernon, Lincoln Junior / Senior High School, Lincoln, Rhode Island
David S. Heckman, Monmouth Academy, Monmouth, Maine
Don W. Hight, Pittsburg State University, Pittsburg, Kansas
Edgar Hood, Dawson High School, Dawson, Texas
Ann Joyce, Issaquah High School, Issaquah, Washington
John G. Kelly, Arroyo High School, San Lorenzo, California
Linda Klett, San Domenico School, San Anselmo, California
George Lai, George Washington High School, San Francisco, California
Katherine P. Layton, Beverly Hills High School, Beverly Hills, California
Debbie Lindow, Reynolds High School, Troutdale, Oregon
Robert Maass, International Studies Academy, San Francisco, California
Guy R. Mauldin, Science Hill High School, Johnson City, Tennessee
Windle McKenzie, Brookstone School, Columbus, Georgia
Bill Medigovich, Redwood High School, Larkspur, California
Sandy Minkler, Redlands High School, Redlands, California
Deborah B. Preston, Keystone School, San Antonio, Texas
Susan M. Smith, Ysleta Independent School District, El Paso, Texas
Sanford Siegel, School of the Arts, San Francisco, California
Gary D. Starr, Girard High School, Girard, Kansas
Tom Swartz, George Washington High School, San Francisco, California
Tim Trapp, Mountain View High School, Mesa, Arizona
Dixie Trollinger, Mainland High School, Daytona Beach, Florida
David Weimeich, Queen Anne School, Upper Marlboro, Maryland
John P. Wojtowicz, Saint Joseph's High School, South Bend, Indiana
Tim Yee, Malibu High School, Malibu, California
iii
Author 's Acknowledgments
This text was written during the period when graphing calculator technology was
making radical changes in the teaching and learning of calculus . The fundamental
differences embodied in the text have arisen from teaching my own students using
this technology . In addition, the text has been thoroughly revised to incorporate
comments and suggestions from the many consultants and field testers listed on
the previous page .
Thanks in particular to the original field test people - Betty Baker, Chris Comins,
Debbie Davies, Val Guernon, David Heckman, Don Hight, Kathy Layton, Guy Mauldin,
Windle McKenzie, Debbie Preston, Gary Starr, and John Wojtowicz . These instructors
were enterprising enough to venture into a new approach to teaching calcu lu s and to
put up with the difficulties of receiving materials at the last minute.
Special thanks to Bill Medigovich for editing the book, coordinating the field test
program, and organizing the first two summer institutes for instructors . Special
thanks also to Debbie Preston for drafting the major part of the Instructor's Guide
and parts of the Solutions Manual, and for working ,vith the summer institutes for
instructors. By serving as both instructors and consultants, these two have given
this text an added dimension of clarity and teachability .
Thanks also to my students for enduring all those handouts, and for finding things
to be changed! Special thanks to my students Craig Browning, Meredith Fast, William
Fisher, Brad Wier, and Matthew Willis for taking good class notes so that the text
materials could include classroom-tested examples .
Finally, thanks to the late Richard V. Andree and his wife, Josephine, for allowing
their children, Phoebe Small and Calvin Butterball, to make occasional appearances
in my texts.
Paul A. Foerster
Dedication
To people from the past, including James H. Marable of Oak Ridge
National Laboratory, from whom I first understood the concepts of
calculus; Edmund Eickenroht, my former student, whose desire it was
to write his own calculus text; and my late wife, Jo Ann.
To my wife Peggy, who shares my zest for life and accomplishment .
iv
Foreword
by John Kenelly , Clemson University
In the era of calculus reform initiated in January 1986 at the Sloan Conference and
fueled by the explosion of technology in mathematics instruction, we have all had to
deal increasingl y with the question "When machines do mathematics, then what do
mathematicians do?" Many feel that our historical role has not changed, but that the
emphasis is now clearly on selection and interpretation rather than manipulation
and methods . As teachers, we sense th e ne ed for a major shift in the instructional
means we employ to impart mathematical und erstanding to our students. At the
same time, we recognize that behind any technology there must be human insight.
In a world of change, we must build on the past and take advantage of the future.
Applications and carefully chosen examples still guide u s through what works.
Challenges and orderly investigations still develop matur e thinking and insights.
As much as the instructional environment might chang e, quality education remains
our goal. What we need are authors and texts that brid ge the transition. It is in this
regard that Paul Foerster and his text provide an outstanding answer.
In Calculus: Concepts and Applications, Paul is at his famous best. The material is
presented in an easily understood fashion, with ample technology-based examples
and exercises. The wealth of applications are intimately connected with the topics
and amplify the key elements in each section. The material is loade d vvith both fresh
items and ancient insights that hav e stood the test of time . For exampl e, you will
find both Escalante's "cross hatch " m ethod of repeated int egratio n by parts right
alongside Heaviside's thumb trick for solving partial fractions! The stu dents are
repeatedly sent to their "graph er." Early on, when differentiation is introduced, local
linearity is discussed, and lat er the zoom features in calculators are explo ited in the
coverage of !'Hospital's rule. That's fresh . Later on, the logisti c curve and slope fields
in differential equations are discussed. All of these are beautiful examples of how
comp uting technology has changed the calculus course.
Throughout the book you will see how comprehensive Paul is in his study of th e
historical role of calculus and yet how curr ent he is in his und erstanding of the AP
community and collegiate "calculus reform." Brilliant, timely, solid, and loaded with
tons of novel applications-your typica l Foerster!
John Kenelly has been involved with th e Adva nced Placement Calculus program for the past
30 years. He was Chi ef Reader and lat er on Chairman of th e AP Calculu s Committee whe n
Paul Foerster was grading th e AP exams in the 1970s, and was instrum enta l in gett ing the
reading sessions moved to Clemson Univ ersity when they outgrew the prior facilities. He is
a leader in developm ent of th e graphing calculator and in pion eerin g its use in college and
school classrooms. Hi s organi zation of TI CAP sessions following recent AP readings has
allowed calculus instructors to share ideas for implem enting the chang es in calculus that
hav e been made inevitabl e by the advent of technology.
V
Contents
Experiments and Projects xi
A Note to the Student xiii
~
1 Limits, Derivatives, Integrals, and Integrals
1-1 The Concept of Instantaneous Rate 3
1-2 Rate of Change by Equation , Graph , or Table 6
1-3 One Type of Integral of a Function 13
1-4 Defin ite Integrals by Trapezoids , from Equations and Data 18
1-5 Limit of a Function 25
1-6 Calc ulus Journal 31
1-7 Chapter Review and Test 33
2
[jJ Properties of Limits
2-1
2-2
Numerica l Approach to the Definition of Limit
Graphical and Algebraic Approaches to the Defin ition of Limit
37
39
40
2-3 The Limit Theorems 45
2-4 Continuity 52
2-5 Limits Involving Infinity 60
2-6 The Intermediate Value Theorem and Its Consequences 67
2-7 Chapter Review and Test 71
lt;kl 3 Derivatives, Antiderivatives, and Indefinite Integrals
3-1
3-2
Graphical Interpretation of Derivative
Difference Quotients and One Definition of Derivative
77
79
80
3-3 Derivative Functions, Numerically and Graphically 84
3-4 Derivative of the Power Function and Ano ther Definition of Derivat ive 90
3-5 Displacement , Velocity , and Acce lerat ion 98
3-6 Introduction to Sine, Cosine, and Composite Functions 105
3-7 Derivatives of Composite Functions-T he Cha in Rule 107
3-8 Proof and Applica tion of Sine and Cosine Derivat ives 111
3-9 Antiderivatives an d Indefinite Integra ls 119
3-10 Chapter Review and Test 123
4
ls
.
Products, Quotients, and Parametric Functions 129
)
4-1 Combinat ions of Two Functions 131
4-2 Derivative of a Product of Two Functions 132
4-3 Derivative of a Quotient of Two Functions 136
4-4 Derivatives of the Other Trigonometric Functions 141
4-5 Derivatives of Inverse Trigonometric Functions 145
4-6 Differentiability and Continuity 153
4-7 Derivative of a Parametric Function 160
4-8 Graphs and Derivatives of Implicit Relations 168
4-9 Chapter Review and Test 172
vii
5
[ID_] Definite and Indefinite Integrals
5-1
5-2
A Definite Integral Problem
Review of Antiderivatives
179
181
182
5-3 Linear Approximations and Differentials 183
5-4 Formal Definition of Antiderivative and Indefinite Integral 189
5-5 Riemann Sums and the Definition of Definite Integral 195
5-6 The Mean Value Theorem and Rolle's Theorem 202
5-7 Some Very Special Riemann Sums 213
5-8 The Fundamental Theorem of Calculu s 215
5-9 Definite Integral Properties and Practice 219
5-10 A Way to Apply Definite Integrals 225
5-11 Numerical Integration by Simpson's Rule and a Grapher 230
5-12 Chapter Review and Test 240
6 The Calculus of Exponential
and Logarithmic Functions 249
6-1 Integral of the Reciprocal Function: A Population Growth Problem 251
6-2 Antiderivative of the Reciprocal Function 252
6-3 Natural Logarithms, and Another Form of the Fundamental Theorem 253
6-4 1n x Really Is a Logarithmic Function 263
6-5 Derivatives of Exponential Functions-Logarithmic Differentiation 268
6-6 The Number e, and the Derivative of Base b Logarithm Functions 272
6-7 The Natural Exponential Function: The Inverse of ln 279
6-8 Limits of Indeterminate Forms: !' Hospital 's Rule 285
6-9 Derivative and Integral Practice for Transcendental Functions 290
6-10 Chapter Review and Test 296
6-11 Cumulative Review: Chapters 1-6 302
~
7 The Calculus of Growth and Decay 307
7-1 Direct Proportion Property of Exponential Functions 309
7-2 Exponential Growth and Decay 310
7-3 Other Differential Equations for Real-World Applications 316
7-4 Graphical Solution of Differential Equations by Using Slope Fields 326
7-5 Numerical Solution of Differential Equations by Using Euler's Method 333
7-6 Predator-Prey Population Problems 338
7-7 Chapter Review and Test 341
7-8 Cumulative Review: Chapters 1-7 346
~
8 The Calculus of Plane and Solid Figures 351
?
8-1 Cubic Functions and Their Derivat ives 353
8-2 Critical Points and Points of Inflection 354
8-3 Ma x ima and Minima in Plane and Solid Figures 369
8-4 Area of a Plane Region 380
8-5 Volume of a Solid by Plane Slicing 385
8-6 Volume of a Solid of Revolution by Cylindrical Shells 396
8-7 Length of a Plane Curve-Arc Length 403
viii
8-8 Area of a Surface of Revolution 410
8-9 Lengths and Areas for Polar Coordinates 417
8-10 Chapter Review and Test 426
~
9 Algebraic Calculus Techniques
for the Elementary Functions 433
9-1 Introduction to the Integral of a Product of Two Functions 435
9-2 Integration by Parts-A Way to Integrate Products 436
9-3 Rapid Repeated Integration by Parts 440
9-4 Reduction Formulas and Computer Software 446
9-5 Integrating Special Powers of Trigonometric Functions 451
9-6 Integration by Trigonometric Substitution 456
9-7 Integration of Rational Functions by Partial Fractions 462
9-8 Integrals of the Inverse Trigonometric Functions 469
9-9 Calculus of the Hyperbolic and Inverse Hyperbolic Functions 473
9-10 Improper Integrals 485
9-11 Miscellaneous Integrals and Derivatives 493
9-12 Integrals in Journal 498
9-13 Chapter Review and Test 499
10
~
The Calculus of Motion-
Averages, Extremes, and Vectors 503
10-1 Introduction to Distance and Displacement for Motion Along a Line 505
10-2 Distance, Displacement, and Acceleration For Linear Motion 506
10-3 Average Value Problems in Motion and Elsewhere 513
10-4 Related Rates 517
10-5 Minimal Path Problems 523
10-6 Maximum and Minimum Problems in Motion and Elsewhere 529
10-7 Vector Functions for Motion in a Plane 532
10-8 Chapter Review and Test 548
~
11 The Calculus of Variable-Factor Products 555
11-1 Review of Work-Force Times Displacement 557
11-2 Work Done by a Variable Force 557
11-3 Mass of a Variable -Density Object 563
11-4 Moments, Centroids, Center of Mass , and the Theorem of Pappus 568
11-5 Force Exerted by a Variable Pressure-Center of Pressure 578
11-6 Other Variable-Factor Products 584
11-7 Chapter Review and Test 592
12
[R!J The Calculus of Functions Defined by Power Series
12-1
12-2
Introduction to Power Series
Geometric Sequences and Series as Mathematical Models
597
599
600
12-3 Power Series For an Exponential Function 607
12-4 Power Series for Other Elementary Functions 609
12-5 Taylor and Maclaurin Series, and Operations on These Series 615
ix
12-6 Interval of Convergence for a Series-The Ratio Technique 622
12-7 Convergence of Series at the Ends of the Convergence Interval 630
12-8 Error Analysis for Series 642
12-9 Chapter Review and Test 649
12-10 Cumulative Reviews 654
Final Examination: " A Guided Tour Through Calculus" 661
Appendix A: Summary of Properties of Trigonometric Functions 665
Appendix B: Answers to Selected Problems 667
Glossary 761
Index of Problem Titles 7 65
General Index 7 69
X
Experiments and Proiects
Problem Set 3 -8
5. Pendulum Experiment 117
6 . Daylight Research Project 117
Problem Set 4-7
11 . Pendulum Project 166
Problem Set S-S
12 . Exact Integral of the Square Function by Brute Force Project 201
13 . Exact Integral of the Cube Function Project 202
Problem Set 7-3
6. Burrette Experiment 323
8. Advertising Project 324
9. Water Heater Project 324
10 . Vapor Pressure Project 325
Problem Set 7 -S
7 . U.S. Population Project 336
Problem Set 8 -3
6. Open Box II (Project) 373
8. Open Box IV (Project) 374
16 . Tin Can Generalization Project 376
30. Submarine Pressure Hull Project 379
32 . Corral with Short Wall Project 380
Problem Set 8-9
19 . LP Record Project 424
20 . Kepler 's Law Project 424
22 . Project-The Angle Between the Radius and the Tangent Line 425
Section 8-10 Concepts Problems
C5 . Area by Planimeter Project 430
C6 . Hole in the Cylinder Project 431
Cl . Three-Hole Project 431
Problem Set 9-9
28 . Hanging Chain Experiment 482
Problem Set 10-2
16 . Elevator Project 512
Problem Set 10-7
16 . Curvature Project 54 7
Problem Set 11 -4
7 . Centroid of a Triangle Experiment 574
8 . Centroid Cut-Out Experiment 574
xi
A Note to the Student
Calculus, since its invention over 300 years ago, has been the culmination of elemen-
tary mathematics and the springboard from which much of higher mathematics gets
its start. The factoring, algebraic fractions, equation solving, and graphing you have
experienced were done at least in part because "You will need them in calculus."
The advent of the computer and hand-held graphing calculator (the "grapher") has
changed much of this, both for how mathematics is learned before you reach calcu-
lus and for how it is used after calculus.
Nevertheless, in this course you will learn that the underlying concepts of calculus
are still important . Calculus deals with functions that behave "continuously" or
"smoothly ." Digital computers and calculators work "discretely," by small steps.
The discrete operations of calculators and computers can help you understand the
idea of limit, on which the continuous behavior of functions in calculus is based.
As you watch the calculator plot a graph you will get a feeling for what the rate of
change of a function means. Perhaps the most significant thing you will find is that
the calculator gives you ways to solve real-world calculus problems approximately,
using graphs and tables of values, before you have developed all of the algebraic
techniques needed for their exact solution.
You will, of course, learn how to do calculus on paper or in your head. If you are
standing in front of the Board of Directors presenting your proposed project, you
can't afford to lose their attention by saying, "Just a minute . Let me find my calcula-
tor." The ability to do things yourself will give you more confidence in answers that
come from a computer. Fortunately, you will not have to make a career out of doing
difficult computations on paper.
The time saved by using technology for solving problems and learning concepts can
be used to develop your ability to write about mathematics. You will be asked to
keep a written journal recording the concepts and techniques you have been
learning, and to verbalize about things you may not yet have mastered .
At times you will feel you are becoming submerged in details. When that happens,
just remember that calculus involves only four concepts:
limits,
derivatives,
integrals, and
integrals.
Ask yourself, "Which of these concepts does my present work apply to?" That way,
you will better see the big picture. Best wishes as you venture into the world of
higher mathematics!
Paul A. Foerster
Alamo Heights High School
San Antonio, Texas
xiii
CHAPTER
1 Limits, Derivatives,
Integrals, and Integrals
In the design of a new car model, it is possible to predict its
performance characteristics even before the first prototype is built.
From information about the acceleration, designers can calculate the
car's velocity as a function of time. From the velocity they can predict
the distance it will go while it is accelerating. Calculus provides the
ma thematical tools to analyze quantities that change at variable rates.
1
Mathematical Overview
Calculus deals with calculating things that change at variable rates.
The four concepts invented to do this are
limits
derivatives
integrals (one kind)
integrals (another kind)
In Chapter 1 you will study three of these concepts in four ways.
Graphically The logo atop each even-numbered
page of this chapter illustrates a limit,
a derivative, and one kind of integral.
Numerically X x- d slope
2.1 0.1 1.071666 .. .
2.01 0.01 1.007466 .. .
d a b
2.001 0.001 1.000749 .. . C
f(x) - f(2)
Algebraically Average rate of change =
x- 2
Verbally I have learned that a definite integral is used to measure the product
of x and f(x). For instance, velocity multiplied by time gives the
distance traveled by an object. The definite integral is used to find
this distance if the velocity varies.
2
1-1 The Concept of Instantaneous Rate
If you push open a door that has an automatic closer, it opens fast at first, slows
down, stops, starts closing, then slams shut. As the door moves, the number of
degrees it is from its closed position depends on how many seconds it has been
since you pu shed it. Figure 1-la shows such a door from above.
The questions to be answered are, "At any particu lar instant in time, is the door
opening or closing?" and "How fast is it moving?" As you progress through this
course, you will learn to write equations express ing the rate of change of one
Figure 1-la variable quantity in terms of another. For the time being, you will answer such
questions graphically and numer ically.
OBJECTIVE Given th e equation for a fun ction relatin g two variables , estimate the inst an tan eous rate
of change of th e depend ent variabl e with resp ect to th e independ ent variable at a given
point.
Sup pose that a door is pushed open at time t = 0 sec and slams shut again at time
t = 7 sec. While the door is in motion, assume that the number of degrees, d, from
its close d position is mode led by the following equation.
d = 200t 2- c for O~ t ~ 7
How fast is the door moving at the instant when t = 1 sec? Figure 1-1b shows this
equation on a grapher (graphing calculator or computer) . When tis 1, the graph is
going up as t increases from left to right . So the angle is increasing and the door is
opening. You can estimate the rate numerically by calculating values of d for values
Fig ure 1-1b
oft close to 1.
t = 1: d = 200(1) . 2- 1 = 100
t = 1.1 : d = 200(1.1) 2- ll = 102.633629 ... 0
The door's angle increased by 2.633 . .. in 0.1 sec, meaning that it moved at a rate of
0
about (2.633 ... )/0 .1, or 26.33 .. . deg/sec . However, this rate is an average rate , and
the question was about an instantaneous rate. In an "instant" that is O sec long, the
door moves 0. Thu s, the rate would be 0/ 0, which is awkward because of division
by zero .
To get closer to the instantaneo u s rate at t = 1 sec, find d at t = 1.01 sec and at
t = 1.001 sec .
t = 1.01 : d = 200(1.01) 2- 1.0J = 100.30234 .. . , a change of 0.30234 ... 0
t = 1.001 : d = 200(1.001) 2- 1.00! = 100 .03064 .. . , a change of 0.03064 ... 0
Here are the average rates for the time interva ls 1 sec to 1.01 sec and 1 sec to
1.001 sec.
1sec to 1.01sec: average rate = 0.30234
O.Ol
.. . = 30 234
.
d
. . . eg/sec
1sec to 1.001sec: average rate = 0.03064...
O.OOl =
30 - d
.64.. . eg/sec
Section1-1: The Conc
ept of InstantaneousRate 3
The important thing for you to notice is that as the time interval gets smaller and
smaller, the number of degrees per second doesn't change much . Figure 1-lc shows
why. As you zoom in on the point (1, 100), the graph appears to be straighter, so the
change in d divided by the change in t becomes closer to the slope of a straight line.
If you list the average rates in a table, another interesting feature appears. The
values stay the same for more and more decimal places .
sec averag e rate
1 to 1.01 30.234 20 ...
1 to 1.001 30.64000 ...
1 to 1.0001 30.6807 5 ...
1 to 1.00001 30.68482 ...
1 to 1.000001 30.68524 ...
Figure 1-l c
There seems to be a limiting number that the values are approaching.
Estimating the instantaneous rate at t = 3 gives the following results.
t = 3: d = 200(3) 2- 3 = 75
t = 3.1: d = 200(3.1) 2- 3 1 = 72.310056 ... 0
t = 3.01: d = 200(3 .01) 2- 3 .oi = 74.730210 . .. 0
t = 3.001 : d = 200(3.001) . 2- 3 001 = 74.973014 . . . 0
Here are the corresponding average rates.
3 sec to 3 .1 sec : average rate _ _ ... - 75 = - 26 .899 . .. d eg/sec
72.310056
31 3
3 sec to 3.01 sec: average rate 74.730210 .. . - 75 = - 26 978 d
3.01 - 3 . . . . eg/sec
74 973014 75
3 sec to 3.001 sec: average rate = 3.001 - 3 - = - 26 .985 ... deg/sec
Again, the rates seem to be approaching some limiting number, this time around - 27.
So the instantaneous rate at t = 3sec should be somewhere close to -27 deg/ sec .
The negative sign tells you that the number of degrees, d, is decreas ing as time goes
on . Thus, the door is closing when t = 3. It is opening when t = 1 because the rate
of change is positive.
For the door example shown above, the angle is said to be a function of time . Time is
the inde pen dent var iable and angle is the dependent variable. These names make
sense, because the number of degrees the door is open depends on the number of
seconds since it was pushed. The instantaneous rate of change of the dependent
variable is said to be the limit of the average rates as the time interval gets closer
to zero . This limiting value is called the derivative of the dependent variable with
respect to the independent variable.
4 Chapter
1: Limits, Derivatives,
Integrals,
andIntegrals
Problem Set 1 1
l. Pendulum Problem: A pendulum hangs from the ceiling
(Figure 1-ld). As the pendulum swings, its distance, d cm,
from one wall of the room depends on the number of seconds,
t, since it was set in motion. Assume that the equation ford as
a function of t is
d = 80 + 30cos-'ft, t <':0.
It is desired to find out how fast the pendulum is moving at a
given instant, t, and whether it is approaching or going away
from the wall.
a. Find d when t = 5. If you don't get 95 for the answer, make
sure your calculator is in radian mode.
b. Estimate the instantaneous rate of change of d at t = 5 by
finding the average rates fort = 5 to 5.1, t = 5 to 5.01, and Figure 1-ld
t = 5 to 5.001.
c. Why can't the actual instantaneous rate of change of d with respect tot be calculated
using the method in lb ?
d. Estimate the instantaneous rate of change of d with respect to tat t = 1.5. At that
time is the pendulum approaching the wall or going away from it? Explain.
e. How is the instantaneous rate of change related to the average rates? What name is
given to the instantaneous rate?
f. What is the reason for the domain restriction t <': O? Can you think of any reason
that there would be an upper bound to the domain?
2. Board Price Problem: If you check the prices of various lengths two -by-six
of lumber, you will find that a board twice as long as another
of the same type does not necessarily cost twice as much . Let x
/
I
A2"
~ "
be the number of feet long a 2" x 6" board is (Figure 1-le) and l-xf eet---1
let y be the number of cents you pay for the board. Assume Figure 1-1e
that y is given by
y = 0.2x 3 -4.8x 2 + 80x.
a. Find the price of 2" x 6" boards that are 5 ft long, 10 ft long, and 20 ft long.
b. Find the average rate of change of the price in cents per foot for 5 ft to 5.1 ft, 5 ft
to 5.01 ft, and 5 ft to 5.001 ft .
c. The average number of cents per foot in 2b is approaching an integer as the change
in x gets smaller and smaller. What integer ? What is the name given to this rate of
change?
d. Estimate the instantaneous rate of change in price if xis 10 ft and if xis 20 ft . You
should find that each of these rates is an integer.
e. One of the principles of marketing is that when you buy in larger quantities, you
usually pay less per unit. Explain how the numbers in Problem 2 show that this
principle does not apply to buying longer boards. Think of a reason why it does not
apply.
Sectionl l: TheConcept
of Instantaneous
Rate 5
1-2 Rate of Change by Equation, Graph, or Table
In Section 1-1, you explored functions for which an equation related two variable
quantities. You estimated the rate of change of one variable as the other increased.
In this section you will do the same for functions specified not only by an equation
but also by a graph or by a table of values.
OBJECTIVE Given a function specified by a graph, by a table of rnlues, or by an equation, and given a
value of x, tell whether the y-value is increasing or decreasing as x increases through that
value, and estimate how fast the y-value is changing.
Background:
Function
Terminology
andTypesof Functions
The price you pay for a certain kind of board depends on how long it is. In
mathematics the symbol f(x ) (pronounced "f of x" or "f at x") is often used for
the dependent variable. The letter f is the nam e of the function, and the number
in parentheses is either a value of the independ ent variable or the variable itself. If
f (x) = 3x + 7, then f(S) is 3(5) + 7, or 22.
The equation f(x) = 3x + 7 is the particular equation for a linear function . The
general equation for a linear equation is written y = mx + b, or f(x) = mx + b,
where m and b represent the constants . The following box shows the names of
some types of functions and their general equations .
Definitions:Typesof Functions
Linear:
f(x) = mx + b; m and b stand for constants, m I=0
Quadratic:
f(x) = ax 2 + bx+ c; a, b, and c stand for constants, a I=0
Polynomial: f(x) = a 0 + a 1 x + a 2 x 2 + a3x 3 + a4x 4 + + anxn; ao, a 1 , . . . stand for
constants, n is a positive integer, an I=0 (nth degree polynomial function)
Power:f(x) = axn; a and n stand for constants
Exponential:
f(x) = abx; a and b stand for constants, a I=0, b > 0, b I=1
Rational
Algebraic:
f(x) = (polynomial)/(polynomial)
Absolute
value:f(x) contains !(expression)!
Trigonometric
orCircular:
f(x) contains cosx, sinx, tanx, cotx, secx, or csc x
Example 1 Figure l-2a shows the graph of a function. At x = a, x = b, and x = c, tell if y is
increasing, decreasing, or neither as x increases, and tell if the rate of change is fast
or slow.
Solution At x = a, y is increasing quickly as you go from left to right.
At x = b, y is decreasing slowly because y is dropping as x goes from left to
right, but it's not dropping very quickly.
6 l: Limits,Derivatives,
Chapter Integrals
, andIntegrals
X
Figure l-2a
At x = c, y is neither increasing nor decreasing, as shown by the fact that the
graph has leveled off at x = c.
Example 2 A mass is bouncing up and down on a spring hanging from the ceiling (Figure
l -2b). Its distance, y feet, from the ceiling is measured by strobe photography each
1/10 sec, giving the adjacent table of values, in which tis time in seconds.
a. Tell how fast y is changing at each time.
i. t = 0.3
ti. t = 0.6
tii. t = 1.0
b. At time t = 0.3sec, is the mass going up or down? Justify your answer.
y
0.2 3.99
0.3 5.84
0.4 7.37 y
0.5
0.6
8.00
7.48 l y
0.7
0.8
0.9
6.01
4.16
2.63
_j
1.0 2.00
1.1 2.52
Figure l -2b
Solution a. If data is given in numerical form, you cannot get better estimates of the
rate by taking values of t closer and clos er to 0.3. However, you can get a
better estimate by using the t-values on both sides of the given value. A
time-efficient way to do the computations is shown below. If you like, do the
computations mentally and writ e only the final answ er.
Section1-2: Rateof Change
byEquation
, Graph,
orTable 7
~
t y diff erence rate average rat e
0.2 3.99 :::=-
1.85 1.85/ 0.1 18. 5
0.3 5.84 :::=-
1.53 1.53/ 0.1 15.3 :::=- 16.9
0.4 7.37
0.5 8.00 :::=-
7.48 -0.52 -0 .52/ 0.1 = - 5.2
0.6 -9.95
0.7 6.01 :::=--1. 47 -1. 47 / 0.1 = -14.7 :::=-
0.8 4.16
0.9
1.0
2.63
2.00
===-
-0 63
. -0 .63/ 0.1 -6.3
-0.55
1.1 2.52 ===-
0.52 0.52 / 0.1 5.2 :::=-
All you need to write on your paper are the results, as shown below.
i. t = 0.3 : increasing at about 16.9 ft/sec
ii. t = 0.6 : decreasing at about 9.95 ft/sec
iii. t = 1.0: decreasing at about 0.55 ft/sec Writ e real-world answers with unit s.
b. At t = 0.3, the rate is about 16.9 ft/ sec, a positive number. This fact implies
that y is increas ing. As y incr ease s, the mass goes downward. ,.
y
. .. Note that although a graph is not asked for in Example 2, plotting the data either
. . .. .
i
on graph paper or by scatter plot on the grapher will help you understand what is
5
. Ii happ ening. Figure l-2c shows such a scatter plot.
. t
1,
The technique in Example 2 for estimating instantaneous rates by going forward
l and backward from th e given value of x can also be applied to functions specified
by an equation. The result is usually more accurate than th e rate es timated by going
Figure l-2c just forward as you did in the last section. Example 3 shows how this computation
can be done.
Example 3 An inflated toy ba lloon is tied to a small rock, then th e ro ck and the balloon are
thrown into the air . While it's moving, the rock's height , f (x) feet above the ground,
f( x)
20 Not chan ging is given by the quadratic function
f(x) = - x 2 +Bx+ 2,
15 where x is time in seconds since the rock was thrown. The graph of this function is
shown in Figure l-2d. At approximatel y what rate is f (x ) increasing or decreasing if
x equals the following times?
10
a. X = 3
b. X =7
Domain
S vb egins C. X =4
Domain
ends
X
3 4 7
Figure l-2d
Solution a. Find out how much f(x) changes from x = 2.99 to x = 3 and divide by the
change in x. Do the same as x goes from 3 to 3.01. Using the trace feature
8 l: Limits,Derivatives,
Chapter Integrals,
andIntegrals
on the grapher is a time-efficient way to do this. Figure 1-Ze shows the
graph drawn in a "frien dly" window for which x = 3 is a grid point and the
x-increm ent is 0.01.
f(Z.99 ) = 16.9799
X
f( 3) = 17 difference = 17 - 16.9799 = 0.0201
y=l7 .0199
f( 3.01) = 17.0199 difference = 17.0199 - 17 = 0.0 199
Figure l -2e 0.0201
From f(Z.99) to f (3), rate;:::, ~ = 2.01 ft /sec .
0.01 99
From f (3) to f (3.01) rate;:::, ~ = 1.99 ft/sec.
Averaging these values gives 2.00.
:. f (x) is increasing by about 2 ft/sec when x = 3.
b. Choose a friendly window for which x = 7 is a grid point and the x-increment
is 0.01.
f (6.99) = 9.0599
f(7) = 9 difference = 9 - 9.0599 = - 0.0599
f(7 .01) = 8.9399 difference = 8.9399 - 9 = - 0.0601
- 0.0 599
From f (6.99) to f (7), rate ;:::, O.Ol = - 5.99 ft/ sec.
- 0.0601
From f (7) to f(7.01), rate;:::, O.Ol = -6 .01 ft/sec .
Averaging these values gives -6 .00
: . f(x ) is decreasin g by about 6 y-units per x unit (6 ft/ sec) when x = 7.
c. Choose a friend ly window for which x = 4 is a grid point and the x-increment
is 0.01.
f (3.99 ) = 17.9999
f (4) = 18 difference= 18 - 17.9999 = 0.0001
f (4.01 ) = 17.9999 diff erence = 17.9999 - 18 = - 0.000 1
0.0001
From f( 3.99) to f (4) , rate;:::,~ = 0.01 ft/ sec
- 0.0001
From f (4) to f (4.0 1), rate ;:::, O.Ol = -0.01 ft/sec
Averaging th ese valu es gives 0.
:. f(x) is not changing when x = 4.
As you learned in Section 1-1, the instantaneous rate of change of a function at a
given value of x is called the derivative of the function at that point . The following
describ es th e meaning of the word derivative. You will learn the pr ecise defini tion
when it is time to calculat e derivatives exactly.
Section12: RoteofChange
byEquation, Graph,
orTobie 9
Meaningof Derivative
The derivative of a function at a particular value of the independent variab le is the
instantaneous rate of change of the dependent variable with respect to the independent
variable.
Note that "with respect to the independent variable" implies that you are finding
how fast the dependent variable changes as the indep end ent variab le changes.
Preview:Definition
of Limit
In Section 1-1, you saw that the average rate of change of the y-value of a function
got closer and closer to some fixed number as the change in th e x-value got closer
and closer to zero. That fixed number is called the limit of the average rate as the
change in x approaches zero . The following is a verbal definition of limit. The full
meaning will become clearer to you as the course progresses.
VerbalDefinitionof Limit
Lis the limit of f(x) as x approaches c
if and only if
Lis the one number you can keep f(x) arbitrarily close to
just by keeping x close enough to c, but not equal to c.
Problem Set 1 2
DoThese
Quickly
Starting here, there will be ten short problems for you to work at the beginning of most
problem sets. Some of the problems are intended for review of skills from previous
sections or chapters. Others are to test your general knowledge. Speed is the key here, not
detailed work. You should be able to do all ten problems in less than five minutes.
Q1. Name the type of function: f(x) = x3.
Q2. Find f(2) for the function in Problem Ql.
Q3. Name the type of function: g(x) = JX.
Q4. Find g(2) for the function in Problem Q3.
QS. Sketch the graph: h(x ) = x 2 .
Q6. Find h (S) for the function in Problem QS.
Ql. Write th e genera l equation for a quadratic function.
QB. Write th e particular equation for the function in Figure l-2f.
Q9. Write the particular equation for the function in Figure l-2g.
Q10. What name is given to the instantaneous rate of change of a function?
10 l : Limits,Derivatives
Chapter , Integrals,andIntegrals
y y
X X
Figure l-2f Figure 1-29
Problems 1-10 show graphs of functions with values of x marked a, b, etc . At each marked
value, tell whether the function is increasing, decreasing, or neither as x increases from
left to right, and tell whether the rate of increase or decrease is fast or slow.
1. 2. 3. 4.
f(x) f (x)
5. 6. 7. 8.
9. 10.
f(x)
'
' X ~ I X
d a b ,c d
Section1-2:RateofChange
byEquation
, Graph,
orTable 11
For Probl ems 11 and 12, x is in minut es and y is in centim eters. Find , approximat ely, th e
rate of chan ge of y at th e given valu es of x , and tell wheth er y is in creas ing or decreas ing.
11. a. X = 1.5 X min yc m 12. a. X = 1.0 X min yc m
b. X = 3.0 0.0 10.0 b. X = 3.0 0.5 8.24
C. X = 4.0 0.5 7.6 c.x = 4.5 1.0 8.30
1.0 8.0 1.5 8.26
1.5 10.4 2.0 8.20
2.0 14.0 2.5 8.19
2.5 18.1 3.0 8.30
3.0 22.0 3.5 8.61
3.5 24 .9 4.0 9.20
4 .0 26 .0 4 .5 10.14
4 .5 24 .6 5.0 11. 50
13. Rolling Tir e Problem: A pebbl e is stuck in t sec y in.
the tread of a car tire (Figure l -2h). As th e 1.2 0.63
whe el turns , the distan ce, y inch es, betw een 1.3 0.54
th e pebbl e and th e ro ad at various tim es, 1.4 0.45
t seconds , is given by th e adjac ent chart . 1.5 0.34
a. About ho w fast is y changin g at each tim e? 1.6 0.22
i. t = 1.4 1.7 0.00
ii . t = l. 7 1.8 0.22
iii. t = 1.9 1.9 0.34
b. At what tim e do es th e ston e strike the 2.0 0.45
pavement ? Ju stify your answer. Figure 1-2h
14. Flat Tir e Probl em: A tir e is pun ctur ed t min y in .
by a nail. As th e air leaks out, th e distanc e, 6.00
0
y inch es , betw een th e rim and th e pavem ent
2 4.88
(Figure l -2i) depends on the tim e, t minut es, 4 -l.42
sin ce the tir e was pun ctur ed. Values of t 6 -l.06
and y are given in the adja cent chart . 8 3.76
a. About ho w fast is y changing at each tim e? 10 3.50
i. t = 2 12 3.26
ii . t = 8 14 3.04
iii . t = 14 16 2.84
b . How do you int erpr et th e sign of th e rat e Figure 1-2i
at which y is changin g?
For Probl ems 15-24 , do th e following .
a. Tell th e typ e of fun ction (linear , quadr atic, etc.),
b. Find f (c)
c. Tell whether f (x) is incr easing or decreasin g at x = c, and at approximat ely what rate.
15. f (x) = x 2 + 5x + 6, c = 3 16. f (x) = - x2 + 8x + 5, c = 1
17. f (x ) = 3x, C = 2 18. f (x) = 2X,C = - 3
1 1
19. f (x ) = - -, C = 4 20. f (x) = - - , C = - 2
x-5 X
12 Chap
ter 1: Limits, Derivative
s, Integrals
, andIntegrals
21. f(x) = - 3x+7,c= 5 22. f(x) = 0.2x - 5, c = 8
23. f (x ) = sinx, c = 2 (Radi an mode!) 24. f(x) = COSX, C =1 (Radian mod e!)
25. A ccurate Graph of a Cubic Function Problem : Plot on the grapher the graph of
f(x) = 0.004x 3 - 0.02x 2 - 0 .2x + 4
and sketch the result. Answer the following questions.
a. How can you tell when f(x ) is increasing quickly?
b. In what part of the domain is f(x ) decreasing ?
c. True or false : When f(x) is positive, the function is increasing.
d. At approximately what value of xis f (x) decreasing the most quickly?
26. Accurate Graph of a Rational Function Problem : Plot this function on your grapher.
2
f (x ) = 0.2x - 1
x- 3
Pick a window for which x = 3 is a grid point. Sketch the result , then answer the
following questions .
a. Describe how f(x) is changing when xis far away from 3.
b. Describe how f(x ) is changing when xis close to 3.
c. At approximately what value of x does f(x) stop increasing and start decreasing?
d. At approximately what value of x does f(x) stop decreasing and start increasing?
e. In what domain is the value of f(x ) a real number?
f. Based on your work in 26a-e , what would you say the range of f is?
27. Definition of Limit Problem: Write the verbal definition of limit. Compare it with the
definition in this text. If you did not state all parts of the definition correctly, read it
again. Then try writing the definition again until you get it completely correct.
28. Discussion Problem : Meaning of Limit: Finding a derivative involves finding a ratio:
the change in y divided by the change in x when the change in xis zero. You don't
get an answer when you divide by zero. But you can keep the change in x very close
to zero. When you do, th e valu e of th e ratio seems to stay close to some particular
number. Based on the verbal definition of limit, tell why the derivative of a function
is the same as the limit of the ratio
change in y
change in x
1-3 One Type of Integral of a Function
The title of this chapter is "Limits, Derivatives, Integrals, and Integrals." In Section
1-2 you estimated the derivative of a function, which is the instantaneous rate
of change of y with respect to x. In this section you will learn about one type of
integral, the definite integral.
Section
1-3:OneTypeof Integralofa Function 13
Suppos e you start off in your car. The velocity incr eas es for a while, then levels off.
Figur e l- 3a shows the velocity in creasin g from zero , th en approa ching and leveling
off at 60 ft/ sec .
Velocity, ft/ sec
Area=
dis tance
traveled
Time,
sec
30 70 100
Figure l-3a
In th e 30 sec between tim e t = 70 and t = 100, th e velocity is a constant 60 ft / sec.
Becaus e distan ce = rate x tim e, th e distan ce you go in this time interval is
60 ft/ sec x 30 sec = 1800 ft.
Geom etri cally, 1800 is th e ar ea of th e rectan gle shown in Figure l -3a. The width is
30 and th e length is 60. Between O sec and 30 sec, where the velocity is changing,
th e area of the region und er th e graph also equals the distance traveled. Becaus e
th e length vari es, the ar ea cannot be found simpl y by multiplyin g two numbers .
The proc ess of evaluatin g a product in which one fa ctor varies is called finding a
definite integral. Definit e inte grals can be evaluated by findin g the corr esponding
ar ea. In thi s section you will find th e approximate ar ea by countin g squares on
graph pap er (by "brut e for ce"!). Later, you will apply th e concept of limit to calculat e
definit e int egrals exactly.
OBJECTIVE Given the equation or the graph for a function , estimate on a graph the definite integra l of
the function between x = a and x = b by counting squares.
If you are given only th e equation, you can plot it with your graph er 's grid on
featur e, estim ating th e numb er of squ ares in this way. However, it is mor e accurat e
to use a plot on graph pap er to count squar es. You can get plotting data by using
your graph er's tr ace or tabl e featur e.
Example 1 Estimat e th e definit e int egral of th e exponential fun ction f (x ) = 8( 0.7)Xfrom x = 1
to X = 7.
Solution You can get rea sonabl e accurac y by plottin g f (x ) at each int eger valu e of x
(Figur e l -3b).
The int egral equals th e ar ea und er th e graph from x = l to x = 7. "Und er" the
graph m eans "between the graph and th e x- axis ." To find the ar ea, first count
the whol e squar es. Put a dot in each squar e as you count it to keep tra ck, then
estimat e the ar ea of each parti al squar e to th e nearest 0.1 unit . For instance,
less than half a squar e is 0.1, 0.2, 0.3, or 0.4. You be the judge . You should get
14 r l : Limits,Der
Chapte ivative
s, Integrals
, andIntegra
ls
&~fflMi ?Us t Wl
X f(x ) f (x)
8
0 f(O) = 8
1 f(l ) = 5.6
2 f (2) "' 3.9
3 f (3) "' 2.7
4 f(4) "' 1.9
5 f (S) "' 1.3
6 f (6) "' 0.9
7 f (7) "' 0.7
7
Figure l -3b
about 13.9 squar e units for th e area, so th e definit e integral is approximately
13.9. Answers anywhere from 13.5 to 14.3 are reasonabl e.
If the graph is already given, you ne ed only count the squares . Be sure you know
how much area each square repr ese nts' Example 2 shows you how to do this .
Example 2 Figure l- 3c shows th e grap h of the velocity function v(t) = -1 00t 2 + 90t + 11,
where tis in seconds and v(t) is in feet per secon d. Estimate the definite int egral of
v(t) with respect to t for the tim e interval from t = 0 sec to t = 1 sec.
Solution Notice that each space in the t dir ection is 0.1 sec and each space in the v(t)
direction is 2 ft / sec. Thus, each squar e represents (0.1)(2), or 0.2 ft. You should
coun t about 113.4 squar es for the area. Thus, th e definite integral will be about
(113.4)(0.2) ""2 2.7 ft .
Figure l-3 c
The following is th e meaning of definit e integral. The precise definition is in
Chapter 5, where you will learn an algebraic technique for calculating exact values
of definite integrals .
Section
1-3: OneTypeof Integralof a Function 15
f(x) Integral = area,
repre se ntin g
f(x)- (b - a) Meaningof DefiniteIntegral
The definite integral of the function f from x = a to x = b gives a way to find the product
of (b - a ) and f(x ) , even if f(x) is not a constant. See Figure l-3d.
X
Figure l-3d
Problem Set 1 - 3
DoThese
Quickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Q1. Find f(S) if f(x) = x - 1.
Q2. Find the area of the trapezoid in Figure l -3e.
Q3. Sketch the graph of a linear function with positive y-intercept
and negative slope.
Q4. Sketch the graph of a quadratic function opening downward.
QS. Sketch the graph of an increasing exponential function.
Q6. At what value(s) of x is f(x) = (x - 4) / (x - 3) undefined ?
Ql. Write the particular equation for the function graphed in Figure l-3f.
QB. Write the particular equat ion for the function grap hed in Figure l-3g.
Q9. Write th e particular equation for the function graphed in Figure l -3h.
Q 10. Write the particular equation for the function graphed in Figure l -3i.
10 ft
[" 14 ft
Figure l-3e
\
y y
y y
-~;
. .
.. 1
X
~ 1
.
:-... . .~-
. '
- . :- . . . ~. X
!i:---+-+
..'.+i ., ;.. : :;.\
Figure l -3f Figure 1-39 Figure l-3h Figure l -3i
16 l : Limits,Derivatives,
Chapter Integrals,
andIntegrals
For Problems 1-4, estimate the definite integral by counting squares on a graph .
2+7 2. f (x) = - 0.2x 2 + 8
1. f(x) = -O.lx
a. X = 0 to X = 5 a. X = 0 to X = 3
b. X = - 1 tO X = 6 b. X = - 2 tO X = 5
3. h(x) = sinx 4. g(x) = 2x + 5
a. X = 0 to X = TT a. X = 1 to X = 2
b. X = 0 to X = TT / 2 b. X = - 1 to X = 1
5. In Figure l-3j, a car is slowing down from a speed of v = 60 ft/sec. Estimate the
distance it goes from time t = 5 sec to t = 25 sec by finding the definite integral.
6. In Figure l-3k, a car speeds up slowly from v = 55 mi/ hr during a long trip . Estimate
the distance it goes from time t = 0 hr to t = 4 hr by finding the definite integral.
V, mi/hr
70 .. < .. f-~:;...
< ;;..;
~------'""
~~--:--,
..
--~---(r
....~--
- ~-1 ~0 ; -:; HO.,: -
-~ - . ,:.-...; . ' .. -. ' ... ' ..
.. ... .... .....
. _.
--;,--~-~~~~- r, sec
..........
~
5 25 2 3 4
Figu re 1-3j Figure l-3k
For Problems 7 an d 8, estimate the derivative of the function at the given value of x.
7. f(x ) = tanx, x = 1 8. h(x) =- 7x + l00,x = 5
9. Sports Car Prob lem: You have been hired by an automobile manufacturer to analyze
the predicted motion of a new sports car they are building . When accelerated hard from
a standing start, the velocity of the car, v(t) ft / sec, is expected to vary exponentially
with time, t seconds, according to the equation
v(t) = 100(1 - 0.9 1 ) .
a. Draw the graph of function v in the domain [O, 10].
b. What is the range of the velocity function?
c. Approximately how many seconds will it take the car to reach
60 ft/ sec?
d. Approximate ly how far will the car have traveled when it
reaches 60 ft / sec?
e. At approximately what rate is the velocity changing when
t = 5?
f. What special name is given to the rate of change of velocity?
Section1-3:OneTypeof Integra
l of a Function 17
10. Slide Prob lem: Phoebe sits atop the swimming pool slide
(Figure 1-31). At time t = 0 sec she pushes off. Calvin
ascertains that her velocity, v(t), is given by
v(t) = 10sin0 .3t,
where v(t) is in feet per second. Phoebe splashes into the
water at time t = 4 sec.
a. Plot the graph of function v. (Don't forget to set your
calculator to radian mode 1)
b . What are the domain and range of the velocity function?
- Figu re 1-31
c. How fast was she going when she hit the water?
d. Approximately how long is the slide?
e. At approximate ly what rate was her velocity changing at t = 3? Velocity, v(t )
5 ..;.. , ;- : ......,.
f. What special name is given to the rate of change of velocity?
11. Negative Velocity Prob lem: Velocity differs from speed in that it
can be negative . If the velocity of a moving object is negat ive,
then its distance from its starting point is decreasing as time
increases . The graph in Figure l-3m shows v(t)centimeters per
second as a function of t seconds after its motion started . How ....
far is the object from its starting point when t = 9 sec? i }-~.-r.--.....
~
- 5 .... - - . ---
12. Write the meaning of derivative .
Figure l-3m
13. Write the meaning of definite integral.
14. Write the verbal definition of limit.
1-4 Definite Integrals by Trapezoids,
from Equations and Data
In Section 1-3, you learned that the definite integra l of a function is the product of
x- and y-values, where the y-values may be different for various values of x. Because
the integral is represented by the area of a region under the graph, you were able to
f"(x)
estimate it by counting squares . In this section you will learn a more efficient way
of estimating definite integrals .
5
Figure l -4a shows the graph of
f(x) = 8(0.7)X,
which was the function in Example 1 of Section 1-3. Instead of counting squares,
divide the region into vertical strips and connect the boundaries to form trapezoids .
3 5 7
Although the trapezoids have areas slightly different from the region under the
Figure l -4a graph, their areas are easy to calculate and add. From geometry you recall that the
area of a trapezoid is the average of the parallel sides multiplied by the altitude.
18 l: Limits,Derivatives,
Chapter Integrals
, andIntegrals
Figure l-4b shows the trapezoid between x = 1 and x = 3. Its parallel sides are
f(l) = 5.6 and f(3) = 2.744, and its altitude is (3 - 1) = 2. Thus its area is
(5.6 + 2.744)(2) = 8.344.
((3) = 2.744
The areas of the other two trapezoids in Figure l -4a can be found the same way.
"Altitude"
= 2 spaces
_________. x (2.744 + 1.34456) (2) = 4.08856
3 0.3 4456 + 0.6588344) (2) = 2.00 3394 4
Figure 1-4b The total area of the trapezoids is approximately equal to the definite integral.
Integra l "" 8.344 + 4.08856 + 2.0033944 = 14.4359544 "" 14.4
The answer is slightly larger than the 13.9 that was estimated in Examp le 1 of
Section 1-3. This result is reasonable, because the trapezoids inclu de slightly mo re
area than the region under the graph.
OBJECTIVE Estimate the value of a definite integr al by dividin g the region und er the graph
into trapezoids .
To accomplish th e objective in a time-efficient way, observe that each y-value in the
sum appears twice, except for the first and the last values.
(5.6 + 2.744)(2) + (2.744 + 1.34456)(2) + (1.34456 + 0.6588344)(2)
The sum can be rearranged as shown .
[ (5.6) + 2.744 + 1.344 56 + (0.6588344)] (2)
The four terms inside the brackets are the y-values at the boundar ies of the four
vertical strips. (There is one more boundary than there are strips .) To find the
area, you simply add the y-valu es, taking half of the first one and half of the last
one . The answer is this sum mu ltiplied by th e width of each strip (2, in this case).
This procedure for finding an approximate value of a definite integral is called the
trapez oidal rule .
Example 1 Use the trapezoidal rule to estimate the definite integral of f(x) = 8(0.7) x from
x = 1 to x = 7. Use ten increments .
Solution From x = 1 to x = 7 there are 6 x-units, so th e width of each str ip will be 6/ 10 =
0.6 units . An efficient way to do the computation is to list the x-values in table
form, then compute and add the corresponding y-values, multiplying the first
and last by 1/2. Finally, multip ly the sum by 0.6 to get the answer .
Section 1-4: Definite Integrals byTrap
ezoids, from Equations
andData 19
:1ffi
X y = f(x)
1 5.6 - Use half of this one .
1.6 4.5211 ...
2.2 3.6501 ...
2.8 2.9468 ...
3.4 2.3791 ...
4 1.9208
4.6 1.5 507 ...
5.2 1.2519 ...
5.8 1.0107 .. .
6.4 0.8160 ...
7 0.6588 ... - Use half of this one .
Sum= 23.1770 ... Without rounding, add th ey-va lues as you calc u late them.
Integral "' (23.1770 . . .)(0.6) = 13.9062452 ...
Note that this answer is closer to 13.9, the answer to Example 1 in Section 1-3.
This result is to be expected, because the smaller trapezoids fit the region under
the graph better, as shown in Figure l-4c. If you were to increase the number of
strips you drew in, the value you'd get for the integral would get closer and closer
to 13.8534138 . . .. This number is the limit of the areas of the trapezoids as their
widths approach zero. In Chapter 5, you will learn how to calculate exact values of
integrals.
0.6 3 5 7
The trapezoidal rule is advantageous if you must find the definite integral of a
Figure l -4c function specified in table form, rather than by equation . Example 2 shows you how
to do this.
Example 2 On a ship at sea, it is easier to measure how fast you are going than it is to measure
how far you have gone. Suppose you are navigator aboard a supertanker. The speed
of the ship is measured each 15 min and recorded in the table shown. Estimate the
distance the ship has gone between 7:30 p .m . and 9:15 p.m .
time mi / hr time mi/hr
7:30 28 8:30 7
7:45 25 8:45 10
8:00 20 9:00 21
8:15 22 9:15 26
Solution Figure l -4d shows the given points. Because no informa tion is known for times
between the given ones, the simplest thing to assume is that the graph is
Speed, mi/ hr
a sequence of line segments. Because miles equals (miles/hour)(hours), the
30 answer will equal a definite integral. The integral can be found from the area of
20 the shaded region in Figure l-4d, using the trapezoidal rule .
10 area= ( + 25 + 20 + 22 + 7 + 10 + 21 + ) = 132
Tim e . integral = 0.25 x 132 = 33 Why 0 .257
7:30
Figure l-4d
9:15
. distance is about 33 mi.
20 Chapter
1: Limits, Derivatives,Integrals, and Integrals
Problem Set 1 -4
DoTheseQuic
kly
The following problems are intended to refresh your skills. You shou ld be able to do all
ten problems in less than five minutes.
QI. The value of y changes by 3 units when x changes by 0.1 unit. About how fast is y
changing?
Q2. The value of y changes by - 5 units when x changes by 0.01 unit. Approximately what
does the derivative equal?
Q3. Sketch the graph of the absolute value function, y = Ix I.
Q4. Find f(3) if f (x) = x 2
Q5. What is 50 divided by 1/2?
Q6. Evaluate: sin (rr / 2)
Ql. How many days are there in a leap year?
QB.The instantaneous rate of change of a function is called the - ?- of the function.
Q9. The product of x and y for a function is called the-? - of the function.
QIO. At what value(s) of xis f(x) = (x - 4) / (x - 3) equal to zero?
1. Spaceship Problem: A spaceship is being launched from Cape Canaveral . As the last
stage of the rocket motor is firing, the velocity is given by
v(t) = 1600 x 1.1 1 ,
where v(t) is in feet per second and tis the number of seconds since the last stage
started .
a. Plot the graph of v(t) versus t, from t = 0 to t = 30. Sketch the result .
b. Tell why the area of the region under the graph represents the distance the
spaceship went in this 30-sec interval.
c. Find, approximately, the distance traveled between t = 0 and t = 30 by using
trapezoids of width corresponding to 5 sec . Sketch these trapezoids on your graph.
d. What mathematical term is used for the product of velocity and time found in this
way?
e. To go into orbit around the earth, the spaceship must be going at least 27,000
ft/ sec. Will it be going this fast when t = 30? Justify your answer.
2. Walking Prob lem: Pace Walker enters a walkathon . She starts off at 4 mi/ hr, speeds
up as she warms up, then slows down again as she gets tired. She estimates that her
speed is given by
v(t) = 4 + sin l.4t,
where tis the number of hours since she started and v(t) is in miles per hour.
a. Pace walks for 3 hr . Draw the graph of v(t) as a function oft for these three hours.
Sketch the result on your paper. (Be sure your calculator is in radian mode 1)
b. Tell why a definite integral would be used to find the distance Pace has gone in 3 hr .
c. Estimate the integral in 2b, using six trapezoids. Show these trapezoids on your
graph . About how far did Pace walk in the 3 hr?
Section1-4: DefiniteIntegralsbyTrapezoids,
fromEquat
ionsandData 21
d. How fast was Pace walking at the end of the 3 hr? When did her maximum speed
occur? What was her maximum speed?
3. A ircraft Carrier Landing Problem: In 1993, Kara Hultgreen became one of the first
female pilots authorized to fly navy planes in combat. Assume that as she comes in
for a landing on th e carr ier, her speed in feet per second takes on the values shown in
the table. Find, approximately, how far her plane travels as it comes to a stop. Is her
plane in dang er of running off the other end of the 800-ft-long flight deck?
sec ft/sec
0.0 300
0.6 230
1.2 150
1.8 90
2.4 40
3.0 0
4. Water over the Dam Problem: The amount of water that has flowed over the spillway
on a dam can be estimated from the flow rate and the length of time the water has been
flowing. Suppose that the flow rate has been recorded every 3 hr for a 24-hr period,
as shown in th e table . Estimate the number of cubic feet of water that has flowed over
the dam in this period.
time ft 3/ hr time ft 3/ hr
12:00 a.m. 5,000 12:00 p.m. 11,000
3:00 a.m. 8,000 3:00 p.m. 7,000
6:00 a.m. 12,000 6:00 p.m. 4,000
9:00 a.m. 13,000 9:00 p.m. 6,000
12:00 a.m. 9,000
If a graph is already drawn accurately, it may be more efficient to estimate a definite
integral by counting squares . For Problems 5 and 6, count squares to estimate the definite
int egral for the function shown.
22 r l : Limits,Derivatives
Chapte , Integrals,
andIntegrals
5. Integral from x = 1 to x = 6 6. Integral from x = 0.4 to x = 2
f(x)
10 : t"'i 1'~1-t
. : : : : : :
X X
6 0.4 2
7. Program for Trapezoidal Rule Problem: For us e now and later, it is advantageous to
have a program for your grapher that will evaluate integrals by the trap ezoidal rule.
Write and save such a program. A clever way to writ e it for a grapher is to specify
the function on they = menu - as y 1 , for instance - then set up a loop that changes
x by the appropriate amount, evaluates Y1 for the current value of x, and adds the
result to the sum of the previous y 1 valu es . The input should be the values of a
and b, the lowest and highest x-values for the region under the graph, and n, the
number of increments (trapezoids) to be used. The output shoul d be the approximate
value of the integral. Be sur e that the program uses only half of the first and last
y-values.
8. Debug the program in Problem 7 by using it for the integral in Example l. If the
program does not give the correct answer, go back and fix it.
For Problems 9 and 10, find the definite integrals indi cated . These are the same integrals
you found by countin g squares in Problems 1 and 4 in Problem Set 1-3.
9. Integral of f(x) = - O.lx 2 + 7 from: 10. Integral of g(x) = 2x + 5 from:
a. x = 0 to x = S, 10 increments a. x = 1 to x = 2, 10 increments
b. x = - 1 to x = 6, 10 increments b. x = - 1 to 1, 10 increments
c. x = - 1 to x = 6, 100 increments c. x = - 1 to 1, 100 increments
11. Trapezoidal Rule Error Problem: The trapezoidal rule overestima te s the integral in
one of Problems 9 and 10, and underestimates it in the other. Which is which ? How
do you tell?
12. Elliptical Table Problem: Figure l-4e shows the top of a coffee y
table in the shape of an ellipse. The ellipse has the equation
X
C:of
+(:of= 1
where x and y are in centimeters . Use the trapezoidal rule to
estimate the area of the table. Will this estimate be too high or too Figure l-4e
low? Explain. What is the exact area of the ellipse?
s byTrapezoids,
Section1-4: DefiniteIntegral fromEquations
andData 23
13. Football Problem: The table shows the cross-sectional area, A, of a football at various
distances, d, from one end . The distances are in inches and the areas are in square
inches. Use the trapezoidal rule to find, approximately, the integral of area with
respect to distance. What are the units of this integral? What, then, do you suppose
the integral represents?
d (in.) A (in 2 ) d (in.) A (in 2 )
0 0.0 7 30.3
1 7.0 8 27.2
2 10.5 9 23.0
3 23.0 10 10.5
4 27.2 11 7.0
5 30.3 12 0.0
6 31.8
14. Integra l as a Limit Problem: Now that you have a program to calculate definite
integrals approximately, you can see what happens to the value of the integral
as you use narrower trapezoids. Estimate the definite integral of f(x) = x 2 from
x = 1 to x = 4, using 10, 100, and 1000 trapezoids. What number do the values
seem to be approaching as the number of trapezoids gets larger and larger? Make a
conjecture about the exact value of the definite integral as the width of each trapezoid
approaches zero. This number is the- ?- of the areas of the trapezoids as the limit
of their widths approaches zero. What word goes in the blank?
15. Derivat ive from Graph Problem: You recall that the derivative
of a function is the instantaneous rate of change of the y-
value as the x-value increases. If the graph of a function
is already drawn, you can estimate its derivative at a given
point by drawing a line tangent to the graph at that point. If
you put a ruler on the concave side of the graph (see Figure
l -4f), with a bit of the graph projecting beyond the ruler, you
can draw a reasonably accurate tangent . Pick a convenient
run and measure the rise to find the slope .
a. Find the derivative of f(x) at x = 2 for Problem 5 of this Figure l-4f
problem set.
b. Find the derivative of f(x) at x = 0.6 for Problem 6 of this
prob lem set.
16. Exact Integra l Conjecture Problem: The exact definite integral of g(x) = x 3 from x = 1
to x = 5 is an integer. Make a conjecture about what this integer equals. Justify your
answer.
'' 17. Meaning of Limits Problem: In this problem you will learn something about the
meaning of the word limit and how this meaning relates to the verbal definition. Let
2
f(x) = 2x - 50.
x- 5
a. Explain why f(5) is undefined.
*This problem prepares you for the next section.
24 Chapter
1: Limits,Derivatives,
Integrals,
andIntegrals
b. Find f(4 .9) and {(5.1). If you like, you may simplify the express ion for f(x) first.
c. What number are f(4.9) and f(5 .l) both close to?
d. Find f(4 .99) and f(5.01 ). Are these both close to the number you wrote in 17c?
e. Let L be the num ber in 17c. Write the following sentence, filling the appropriate
number into the blank : "If xis within 0.01 units of 5 (but not equal to 5), then f(x)
is within -7- units of L."
1-5 Limit of a Function
Calculu s involves four concepts : limits, derivatives, integrals, and integrals. You
have learned that a derivative is the rate of change of a function, and that a definite
integral gives a way of calculating a product - such as (rate)(time) - in which one of
the factors varies.
The formal definitions of derivative and definite integral involve the concept of the
limit of a function . In Section 1-2 you learned a verbal definition of limit . In this
sect ion you will learn a formal definition, and see how this relates to the meaning
of limit .
OBJECTIVE Given the graph or the equation of a function , tell whether or not the function ha s a limit
as x approa ches the given value and tell how your answer relat es to the definition of limit.
Consider the rationa l algebraic function
2
f(x) = 0.4x- 10.
x- 5
y
If you plot the function on the grapher, you get a straight line with a gap in it, as
shown in Figure l-5a. The gap occurs where x = 5, and is due to the fact that the
denominator, x - 5, is zero when xis 5. It will show up if you use a window that
includes x = 5 as a grid point.
X An in teresting thing shows up if you use the trace or the table feature of your
X= 4.6 5 y=3.84 grapher . Using a window with an x-increment of 0.1, you get the values shown . No
y-valu e appears where x = 5, b ut following the pattern suggests that y should be
Figure l -5a exact ly 4 when x = 5.
X = 4.6 y = 3.84
X = 4.7 y = 3.88
X = 4.8 y = 3.92
X = 4.9 y = 3.96
x= 5 y =
X = 5.1 y = 4.04
X = 5.2 y = 4.08
Sec
tion 1-5: Limitof a Function 25
Try to find f(5) by direct substitution.
52
f(5) = 0.4( _) - lO = O, whi ch 1s. un d eflne db ecause ofd 1v1s10n
' . . b y zero .
5 5 0
The fraction 0/ 0 is called an indeterminate form . Algebra shows you what is
going on.
f(x) = 0.4(x + 5) (x - 5)
x- 5
f(x) = 0.4x + 2, provided that x f=5.
Substituting 5 for x in the simplified expression gives 0.4(5) + 2 = 4, which is the
value you would get by following the pattern in the table.
The graph in Figure l-5b is said to have a removable discontinuity at x = 5. The
Figure l-5b function is discontinuous because of the gap, but the gap can be removed simply
by defining f(5 ) to be 4. When you draw such a graph on your paper, it is customary
to show an open circle at the discontinuity, indicating that there is no value of y for
that one value of x.
The numb er 4, which f(x) is close to when x is close to 5, is the limit of f(x) as x
_____________3,99 approaches 5. You should begin to see how the parts of the verbal definition of limit
are coming together. You can make f (x ) as close as you like to 4 just by keeping x
X
close enough to 5 (but not equal to 5). Suppose someone tells you, "Keep f (x ) within
X=4.98 5 5 Y = 3.994 0.01 unit of 4." To find approximately how close you must keep x to 5, you can trace
on the graph (Figure l-5c) or make a table.
Figure l-5c
X = 4.97 y = 3.988
X = 4.975 y = 3.99
X = 4.98 y = 3.992
X = 4.985 y = 3.994
X = 4 .99 y = 3.99 6
X = 4 .995 y = 3.998
x within 0.025 unit of 5 x= 5 y= y within 0.01 unit of 4
X = 5.005 y = 4.002
X = 5.01 y= 4 .004
X = 5.015 y = 4 .006
X = 5.02 y = 4.008
X = 5.025 y = 4.01
X = 5.03 y = 4.01 2
{ (x) 41~
3.99
lm
/-~~~:'.
~-------
4.01
:
---
--
~----
''
:
{(x) wind s up
:
''
:
:
''
:
You can also use algebra to find out exactly how close to keep x to 5.
3.99 < f(x) < 4.01
3.99 < 0.4x + 2 < 4.01
1.99 < 0.4x. < 2.01
4.975 < x < 5.025 (and x f=5)
'
4.9:75
Pick x in her e.
v~
' '
5:025
The last inequality says that x is within 0.02 5 unit of 5. By reversing the steps,
you can see that if xis within 0.025 unit of 5, then f(x) is within 0.01 unit of 4
(Figure l-5d).
Figure l-5d
The above work leads to the following formal definition of limit. You should commit
this definition to memory so that you will be sure to remember its various parts.
26 l: Limits,Derivatives,
Chapter Integrals,
andIntegrals
FormalDefinitionof Limit
L is the limit of f (x) as x approaches c
if and only if
for any positive number epsilon, no matter how small,
there is a positive number delta such that
if x is within delta units of c (but not equal to c),
then f(x) is within epsilon units of L.
You should be able to make the following connections:
4 is the value of L. It is the one number f(x) stays arbitrarily close to if x is close
to 5.
5 is the value of c. It is the numb er you keep x close to in order for f(x) to stay
close to 4.
0.01 is a value of epsilon (quite small!). It is picked "arbitrarily" to tell how close
to 4 you are supposed to keep f(x ) .
0.025 is the corresponding value of delta. It tells how close to 5 is "clos e enough"
to keep x in order that f(x) will wind up somewhere within 0.01 unit of 4.
Note that f(x) would not be within 0.01 unit of 4 if x were equal to 5 because there
is no value of f(5) to begin with!
Example 1 Tell whether or not the function graphed in Figure f(x )Step
l-5e has a limit at the given x-value, and tell why or discontinui1 y
at X= 3
why not. If there is a limit, give its value . 10 -------------- ,
8 -------------
a. X = 1 b. X =2 C. X =3
X
I I
3 4
Figure l -5e
Solution a. As x approaches 1, the limit is 2. If x is close to 1 but not equal to 1, f(x) is
close to 2.
b. As x approaches 2, the limit is 5. If xis close to 2 but not equal to 2, f(x) is
close to 5. The fact that x can equal 2 is of no consequence in this problem.
c. As x approaches 3, there is no limit. If x is close to 3 on the left, f(x) is close
to 10. If x is close to 3 on the right, f(x) is close to 8. So there is no one
number f(x) can be kept close to just by keeping x close to 3 but not equal
ro3.
Note : The discontinuity shown in Figure l-5e is called a step discontinuity.
Problems 11-14 in Problem Set 1-5 will show you in a "learn as you go" manner how
to find a limit if an equation is given.
Section1-5: Limit
ofa Function 27
~
Problem Set 1 5
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Sketch the graph of a function that is positive and increasing fast at x = 2.
Q2. Sketch a graph showjng the meaning of definite integral.
Q3. Sketch the graph of y = sinx.
Q4. Evaluate: tan (rr / 4)
Q5. Evaluate: 3- 2
Q6. Simplify: 24/ 36 X
Ql. Find 30% of 600.
QB.How many weeks are there in a year? Figure l-5f
Q9. What type of function has a graph like that shown in Figure l-5f?
Q10. What is the name for the instantaneous rate of change of a function?
For Problems 1-10, tell whether or not the function has a limit as x approac hes c; if so,
tell what the limit equals.
1. f (x) 2. grx) 3. f (x) 4. f(x)
8 - - - - - -~
~~
''
X '' X
C 4 C
5. f(x) 6. 7. y 8. f(x)
:--1:2
2- ~
-
~ -- 5
X
' X X X
C
9. 10.
f(x ) ,
rx):
:\
I
:c
X
~ -L
'
:c
'
X
28 Chapterl : Limits,Derivatives,Integrals, and Integra
ls
Problems 11-14 are intended for you to work in groups. The goal is to find out how
various parts of the definition of limit apply when a function is specified by an equation.
11. Definition of Limit I:
a. Write the formal definition of limit.
b. The graph of f(x) = 3x - 7 is shovm in Figure l-5g.
Show that f(4) = 5.
c. You can keep f(x) close to 5 just by keeping x close to X
4. How close to 4 must you keep x in order for f(x) to
stay within 0.6 unit of 5?
d. The 0.6 in llc is a value of epsilon in the definition -7 r
of limit. The answer to l lc is a value of delta. Sketch
how epsilon and delta are related to the graph. Figure l -5g
e. Pick a value of x that is within delta units of 4 but not equal to 4. Show that f(x)
really is within epsilon units of 5.
f. The number 5 fits the definition of limit because you could find a value for delta
no matter how small epsilon is . Show that you understand the meaning of this
statement by finding a value of delta if epsilon is 0.00012. Tell how you found this
value of delta.
12. Definition of Limit II: The function
2
f(x) = 4x - 7x - 2
x-2
is undefined when x = 2. In this problem you will show that f(x) does have a limit as
x approaches 2.
a. Plot the graph of f. Use a friendly window that includes x = 2 as a grid point .
Sketch the graph and name the feature that seems to be present at x = 2.
b. From the graph, tell what you think the limit of f(x) is as x approaches 2.
c. Try to evaluate f(2) by direct substitution. What form does the answer take? What
name is given to a form such as this?
ct. Factor the numerator and simplify the expression by canceling the common factor .
Although the simplified expression does not equal f(2), you can substitu te 2 for
x and get an answer . What is this answer and what does it represent?
e. How close to 2 would you have to keep x in order for f(x) to be between 8.9 and
9.1?
f. How close to 2 would you have to keep x in order for f(x) to be within 0.001 unit
of the limit in 12b? Answer in the form" x must be within-?- units of 2."
g. Four constants appear in the definition of limit: L, c, epsilon, and delta. What are
the values of these four constants in l 2f?
h . Explain how you could find a suitable value of delta no matter how small epsilon
is.
i. What is the reason for the restriction" ... but not equal to c" that appears in the
definition of limit?
Section1-5:Limit
ofa Function 29
13. Definition of Limit III: The function
f(x) = (x 2 - 6x + 13)(x - 2)
(x - 2 )
is undefined when x = 2. However, if you cancel the (x - 2)
factors, the equation becomes f(x ) = x 2 - 6x + 13 (x /, 2). So f
would be a quadratic function, except that there is a removable 5 -----
discontinuity where x = 2 (Figure l-5h). The y-value of this
missing point is the limit of f(x) as x approaches 2. X
2
a. Show that f(2) has the indet erminate form 0/ 0. Do an
appropriate calculation to show that 5 is the limit Figure 1-5h
of f(x) as x approaches 2.
b. Plot the graph close to the discontinuity. Use a friendly window that includes
x = 2 as a grid point and has an x-increment of 0.001. Then use your grapher's
trace or table feature to mak e a table of values of f (x ) for each value of x from
1.990 through 2.010. For which values of x in the table is f(x) within 0.01 unit of
5? Complete th e statement "If xis within-?- units of 2, then f (x ) is within 0.01
unit of 5."
c. Find the largest interval of values of x for which f (x ) is within 0.01 unit of 5. You
can do this by setting f (x ) = 4.99 and solving to find the value of x nearest 2.
Repeat for f(x) = 5.01. Keep as much precision as your calculator will give you.
d. Sketch the part of the graph close to x = 2. Show how the numbers in 13c relate
to the graph.
e. Find the largest number you could put in th e blank of the statement in 13b. Take
into account that the interval in 13c is of a different width on one side of 2 than
on the other .
f. Write the values of the constants L, c, epsilon, and delta for 13a-e.
14. Definition of Limit IV: Answer the following questions for the function
x -2
f(x) = x - Ix - 2 I
a. Plot the graph off. Sketch the result, showing the discontinuity at x = 2.
b. Find f(l.99) and f(2.01). Based on these numbers and the graph, explain why
there is no one number L for which f (x ) is very close to L when xis close, but not
equal, to 2.
15. One-Sided Limit Prob lem: The following function has a step
discontinuity at x = 2.
1 Ix - 21
f(x) = 3 + - x + --
2 x- 2
5~(/~~ : Step
,-dis continuity
a. Figure l-5i shows the graph off. Explain why the graph
takes a jump at x = 2.
b. What is the limit of f(x) as x approaches 2 from the left 2
side ?
c. What is the limit of f(x) as x approaches 2 from the right Figure l-5i
side?
d. Explain why there is no single number that can be the limit of f(x ) as x
approaches 2.
30 Chapter
1: Limits,Derivatives
, Integrals,
andIntegrals
16. Piecewise-Defined Function Problem: A function may be defined by different equations
in different parts of its domain. If this is the case, a single brace is us ed to bracket
together the different equations and to indicate the x-values for which they apply.
Consider the following function.
x+l, ifx < 2 Graph is called the left bran ch.
f(x) = 1, ifx = 2 Graph is called th e middl e branch.
{ x 2 - 6x+ll, ifx > 2 Graph is called th e right branch.
a. Plot the graph off. Sketch the result, showing the behavior around x = 2.
b. Because the equation y = x + 1 is used only for x < 2 and the equation
y = x 2 - 6x + 11 only for x > 2, there is a discontinuity at the end of each branch.
Are these two discontinuities at the same point7
c. Does f(x) have a limit as x approaches 2? If so, what is that limit? If not, tell why
not.
d. Is f(2) equal to the limit of f(x ) as x approaches 2? Explain.
e. The graph off is discontinuous at x = 2. What number would f(2) have to equal in
order for f to be continuous at x = 27
17. For r (x) = 1/ x, estimate the derivative of rat x = -3 .
18. For j(x) = 10 - 2x, estimate the definite integral for x = 0 to x = 3.
19. Don't Believe Everything You See Problem! Ima B. Leaver sets
her grapher's window to [-0. 1, 0.1] for x and [ -0 .07, 0.07]
for y. Then she plots the graphs of
X
Y1 = sinx and Y2 = x .
Because the graphs seem to coincide (Figure 1-Sj), Ima
believes the two functions are identical. Explain to Ima how
she reached this wrong conclusion and some ways in which
she could quickly show that there are really two different Figure l-5j
graphs on the screen.
1-6 Calculus Journal
You have been learning calculus by reading, by listening, by discussing, and by
working problems. An important ability you should develop for any subject you
study is the ability to writ e about it. To gain practice in this technique you will be
asked to keep a journal recording what you have been learning . Uourna l comes
from the same source as the French word )our, meaning "day." Journey comes from
the same source and means "a day's travel.")
Section1-6:Calculus
Journa
l 31
OBJECTIVE Start writing a journal in which you can record things you've learned about calculus and
what questions you still have about certain concepts . In doing so, you'll gain practice in
writing about mathematics, and you'll have a source of reference in your O\.\'ll words to
re,iew before tests.
Rath er than buy a spiral or loo se-leaf not ebook, you might find it us eful to invest
in a bound notebo ok that will hold up und er daily u se. Bound not ebo oks ar e oft en
used by researcher s as th ey rec ord th eir findin gs in th e laborator y. You should
record both th e dat e on which you make a parti cular journal entr y and th e general
headin g und er which th e entry is class ified. A typi cal entr y migh t look som ethin g
like th e one sh own here.
Topio:L~ 9/15
to- 2, for ~Le,, I CI
0-1.erurnea,
tha,t CV@'f'etph,doey VI.at"need, to- hwve, cv "hol,e;' """i,t
fort;hue,to-l,e,cvl.i,mu: The,,ne,,,nbe,,-i,of""?' m,.,fy,voup to-ld,me,tha,t,1,i,nce,the,defi,ru,t'um,
I CIMO-/.erurned, tha,t I CAMV fi,t,d, l.iAni,n, of /retet'..o-rw
Li,k,e,
(2>v+ 7)( >v-3) tha,t ~ to-Q Cv.Ythe,de,,,w,'l'UA'\.CLto-,r
etppvo-o..cJ,,,ey
(>v-3) 0
by ~the,~/U\d,l,y }~..tut'~3 for "'"""whacwl.eft. I'm,;t'.,U,V\.ot";uve,
pvo-d.uce,;, the, vi,gAf CLl'\<l<W.( 0/0 wca,lled, Mv i.ndetum.i,r\a,te, fornu)
P vo-hl.e.m,5 (,rv the, hom.e,wo-,rk.,~ d.id.n!t mo..ke, ~ to- "1< I c;c,.,n.:tu.nde.nt/U\d,
wh,y the, oo.d<, of the, &o-o-k,~l' thue, WCv.Y~ u.w..u: i,t l.oo-to- me, Li,k,e,the, U-m.i.tw
wheA'11
2. /-{ewe, to-Ci<,the, i.n,.\,tvuct"o-vcwout thw n.e.,<.tclct.w.
The journ al should not simpl y be a transc rip tion of your class not es. Nor should
you take not es dir ectly in the journal. The purpos e of the journal is to writ e thin gs
down after you' ve had a chanc e to think abou t th em for a while.
32 l : Limits, Derivatives, Integra
Chapter ls, andIntegrals
Problem Set I -6
1. Start a journal in which to record your understandings about calculus. The first entry
should include such things as those listed here.
The four concepts of calculus
The distinctions among derivative, definite integral, and limit
The fact that you still don't know what the other kind of integral is
The techniques you know for calculating derivatives, definite integrals, and limits
Any questions that still aren't clear in your mind
1-7 Chapter Review and Test
In this chapter you have had a brief introduction to the major concepts of calculus.
Limits
Derivatives
Definite integrals
Another type of integrals
The derivative of a function is its instantaneous rate of change . A definite integral
of a function involves a product of the dependent and independent variables, such
as (rate)(time). A limit is a number that y can be kept close to, just by keeping x
suitably restricted. The other kind of integral is called an indefinite integral, also
known as an antiderivative. You will see why the word integral is used twice when
you learn the fundamental theorem of calculus in Chapter 5.
You have learned how to calculate approximate values of derivatives by dividing
small changes in y by the corresponding change in x. Definite integrals can be found
using areas under graphs and can thus be estimated by counting squares. Limits
of functions can be calculated by finding the y-value of a removable discontinuity
in the graph. Along the way you have refreshed your memory about the shapes of
certain graphs.
The Review Problems below are numbered according to the six preceding sections
of this chapter. The Concepts Problems allow you to apply your knowledge to new
situations . The Chapter Test is more like a typical classroom test your instructor
might give you.
Review Problems
Rl. Bungee Prob lem : Lee Per attaches himself to a strong bungee cord and jumps off
a bridge . At time t = 3 sec the cord first becomes taut. From that time on, Lee's
distance , d feet, from the river below the bridge is given by the following equation.
d = 90 - 80sin l.2(t - 3)
a. How far is Lee from the water when t = 47
Sectio
n 1-7: Chapter
Review
andTest 33
b. Find the average rate of change of d with respect to t for the interval t = 3.9 to
t = 4, and for the interval t = 4 tot = 4.1. Approximately what is the instantaneous
rate of change at t = 4? Is Lee going up or going down at time t = 4? Explain.
c. Estimate the instantaneous rate of change of d with f(xl
respect to t when t = 5.
d. Is Lee going up or down when t = 57 How fast?
e. What mathematical name is given to the instantaneous rate X
of change? 3 5
R2. a. What is the physical meaning of the derivative of a function7
b. For the function in Figure l-7a, tell how f(x ) is changing
(increasing or decreasing, quickly or slowly) when x equals Figure 1-7a
- 4, 1, 3, and 5.
c. If f(x) = 5", find f(2) and the value of the derivative, approximately, at x = 2.
d. Mary Thon runs 200 m in 26 sec! Her distance from the start at various times is
given in the table. Estimate her instantaneous speed in meters per second when
t = 2, when t = 18, and when t = 24. For what interval of times did her speed
stay relatively constant? Why is the speed at t = 24 reasonable in relation to the
speeds at other times 7
t sec m t sec m
0 0 14 89
2 7 16 103
4 13 18 119
6 33 20 138
8 47 22 154
10 61 24 176
12 75 26 200
R3. a. Izzy Sinkin winds up his toy boat and lets it run on the pond. Its speed is given by
v(t) = (5t)(0.8)1,
where v(t) is in feet per second and tis time in seconds since he let it go. Draw
the graph of this function.
b. Find, approximately, the distance the boat travels between t = 2 and t = 10.
c. Name the concept of calculus used in finding this quantity.
R4. The graph in Figure l-7b is for
2
f(x) = -0.5x + l.8x + 4. f(x)
a. Confirm on your grapher that the graph is correct. Use a
friendly interval that includes values of x at the grid lines.
b. Find the definite integral of f (x ) from x = 1 to x = 4 by
counting squares.
X
c. Make a table of values of f(x) for each 0.5 unit, from x = 1 4
to x = 4. Based on the results, use trapezoids to estimate
the integral. Figure 1-7b
34 Chapter
1: Limits,Derivatives,
Integrals,
andIntegrals
RS. a. Write the formal definition of limit.
.
b . G1ven t h e function
. f(x ) = (x 2 - Bxx +_ 5) (x - 3) , find th e limi' t of f(x ) as x
3
approaches 3.
c. What one number tells how far x can be from 3 so that f(x) will be within 0.01
unit of the limit in RSb?
d. Pick a value of x that is within the number of units of 3 you found in RSc, and
show that f(x) really is within 0.01 unit of the limit.
e. Sketch the graph of a function for which the limit of f(x) is 5 as x approaches - 1.
R6. What is the purpose of keeping a journal in calculus? What type of information
should be in the journaJ7 Must you know the answer to everything you write in your
journal? Explain.
Conce pts Problems
Cl. Exact Value of a Derivative Problem: You have been calculating approximate values
of derivatives by finding the change in y for a given change in x, then by dividing. In
this problem you will apply the concept of limit to the concept of derivative to find
the exact value of a derivative. Let y = f(x) = x 2 - 7x + 11.
a. Find f (3).
b. Suppose that x is slightly different from 3. Find an expression in terms of x for
the amount by which y changes, f(x ) - f(3).
c. Divide the answer to Clb by x - 3 to get an expression for the approximate rate
of change of y. Simplify.
d. Find the limit of the fraction in Cle as x approaches 3. The answer is the exact
rate of change at x = 3.
C2. Tangent to a Graph Problem: If you worked Problem Cl correctly, you found that the
instantaneous rate of change of f(x) at x = 3 is exactly - 1 y-unit per x-unit. Plot the
graph of function f. On the same screen, plot a line through the point (3, f(3)) having
slope -1. What do you notice about the line and the curve as you zoom in on the
point (3,f(3))7
Cha pter Test
Tl. Sketch the graph of a function that is increasing quickly at (2, 3) and decreasing
slowly at (5, 6). Make the function "continuous" (no discontinuities).
T2. Sketch the graph of a function with a removable discontinuity at (4, 1).
T3. If f(x) = 2 cos x, find, approximately, the instantaneous rate of change of f(x)
with respect to x when x is 1. Name the concept of calculus that represents the
instantaneous rate.
T4. Given f(x) = 1.2X,find, approximately, the definite integral off from x = 1 to x = 5,
using trapezoids of width 0.5 x-unit.
Section1-7:Chapter
Review
andTest 35
_ 5x2 + l 7x + 14
T5. Let g (X ) - .
x +2
a. Plot the graph, using a ,vindow that includes x = - 2. Sketch the graph on your
paper.
b. What is the limit of g(x) as x approaches - 2? Show this feature on the graph.
c. Show on the graph the interval of x-values for which g(x) will be vvithin 0.4 unit
of the limit in T5b.
d . For T5c, what are the values of c, L, epsilon, and delta in the definition of limit?
e. What is the reason for the restriction" . .. but not equal to c" in the definition of
limit?
T6 . The velocity of a moving object is given by v(t) = 2.718 1 in./ min.
a. How fast is the object moving when t = 1 min ?
b. At approximately what rate is the speed changing when t = 1 min?
c. Approximately how far does the object go between t = 0 min and t = 1 min7
Tl. Cy Kling coasts down the hill on his bicycle, then up the next hill . Every 4 sec, he
notes the speed the bike is going, recording the values in the table. How far did he go
in the 28 sec?
sec ft/ sec sec ft/ sec
0 16 16 37
-t 26 20 36
8 30 24 32
12 34 28 25
36 Chapter
1: Limits,Derivatives,
Integrals,
andIntegra
ls
CHAPTER
2 Properties of Limits
Finding the average speed of a moving vehicle requires dividing
the distance it goes by the time taken to go that distance. The
instantaneous velocity can be calculated by taking the limit of the
average velocity as the interval of time approaches zero. Limits are
also used to find exact values of definite integrals . Thus limit is the
foun dation for the other three concepts of calculus.
37
Mathematical Overview
Informally, the limit of a function fas x approaches c is they-value
f(x) stays close to when xis kept close enough to c but not equal to c.
In Chapter 2 you will formalize the concept of limit by studying it in
four ways.
Graphically The logo on each even -numbered f (x)
page of this chapter shows that
f(x) is close to L when xis close
enough to c but not equal to c. --- ----------r----
' ' --------
: :
L t---'--------a '
------------- r- '
--,--------
,
Numerically x f(x) '
'
'
3.01 3.262015 '
3.001 3.251200 .. . o:'~
3.0001 3.250120 .. . X
3.00001 3.250012 .. . C
Algebraically O < Ix - c I< 6 => I f(x) - l I< E, part of the limit definition.
Verbally I have learned that a limit is a y-value that f(x) can be kept arbitrarily
close to just by keeping x close enough to c but not equal to c.
I remember 6 and E go with x and y, respectively, because they come
in the same alphabetical order. Limits are used to find exact values
of derivatives . Limits as x approaches infinity are related to
horizontal asymptotes.
38
2-1 Numerical Approach to the Definition of Limit
In Chap ter 1, you were ask ed to memorize the definition of limit. Then you got an
intuitive idea that a limit is a number that y stays close to when x is close, but not
equa l, to c. In this section you will explore the function shown in Figure 2-la not
only to discover something about the meanings of th e various parts of the definition
of limit , but also to see how those meanings relate to the grap h .
OBJECTIVE For the cube root function f(x) = (x - 2) 113 + 3, make tables of values that show how close
x must be kept to 2 in ord er for f (x ) to be within given ranges of f( 2).
The pro bl ems in this section are meant for you to do on your own or with your
stu dy group as an assignment after your test on Chapter l.
Exploratory Problem Set 2-1
l. Let f(x) = (x - 2) 113 + 3, as in Figure 2-la.
a. Plot this graph on your grapher. Does your graph agree with
Figure 2-la?
b . Make a table of values of f(x ) for each 0.0001 unit of x from
1.9995 uni ts to 2.0005 units . Use a time-efficient method .
X
c. For which values of x in your table (x * 2) is f(x)
within 0.07 2
unit of 3? That is, for which values of xis 2.93 < f(x) < 3.07?
d . For which values of x in your tab le (x * 2) is f(x) within 0.05 Figure 2-1a
unit of 3?
e. Would it be possib le to keep f(x) within 0.01 unit of 3 without h aving x equ a l 2? If
so, find out how close . If not, te ll why not.
f. The numbers L, c, epsilon , and delta in the definitio n of limit all appear in le. Which
nu mber is which?
g. How could you calculate a value of delta for any given value of epsilon, no matter
h ow small the value of epsilon is?
2. Let g (x) = (x - 3) sin ( x ~ 3 ) + 2. Plot the graph of g by using a window that inclu des
y = 2 and for which x = 3 is a grid point. Then zoom in on the point (3, 2) by a factor
of 10 in both the x - and y -directions . Sketch the resu lting graph. Does g(x) seem to be
approaching a limit as x approac h es 37 If so, what does the limit equal? If not, tell why
not.
3. Let h(x) = sin (x~ J + 2. Plot the graph of h by using a window with a y-range of
about O to 3 and for which x = 3 is a grid point . Then zoom in on the point (3, 2) by a
factor of 10 in the x-direction . Leave the y-scale the same . Sketch th e resu lting graph.
Does h (x ) seem to be approaching a limit as x approaches 3? If so, what does the limit
equ al? If not , tell why no t.
Section2-1: Numerical Approach to the Definitionof Limti 39
2-2 Graphical and Algebraic Approaches
to the Definition of Limit
The definition of limit you memori ze d in Chapt er 1 can be shorten ed som ewhat by
u sing E for ep silon and 6 for delt a. (Epsilon and delta ar e th e English words for th e
lower cas e Greek lett ers E and 6.) The definition can be short en ed more by using
algebrai c symbols and by leaving out word s th at were add ed for clarity . Th e word
limit is abbr eviat ed lim, and th e words x approa ches c ar e expr ess ed as x - c. Th e
short en ed ver sion is shown in the bo x.
Definitionof Limit(shortenedversion}
L = limf (x ) if and only if
x- c
for any number > 0,
there is a numb er o > 0 such that
if xis within 15units of c (but x * c ),
then f(x) is within units of L.
Th e words no matt er how small th at des crib e E hav e be en left out b ecaus e the y are
unne cessar y. However, you should rememb er th at E will usuall y b e very clos e to
ze ro.
OBJECTIVE Gi\en values of the three numbers L, c, and , calculate the correspo ndin g valu e of o in
the defini tion of limit an d show by graph that you understand the significance of these
numbers .
Figur e 2-2a sh ows th e graph of f (x) = (x - 2) 1n + 3, which you work ed with in
Probl em Set 2-1. The limit of f (x ) as x approach es 2 is 3, th e sam e as the value of
f (2) . To see how the definition of limit relat es to this function, let E = 0.8. Th e figure
on th e left shows what happ ens if 6 = 0.2. All th e corr esponding values of f (x ) (the
solid part of th e graph) will b e b etween th e horizontal lin es at y = 2.2 and y = 3.8.
f(x) __
i~o:s
:
( r
((x) -- - ( (x)
3+ C ----.--- ----
c=0.81;
L=3 - - - _t_- - - - -
''
/- -----
:
3+ E
L=3
_____
-;JJ
_____
::i:
~o
:n - 3+ E
L= 3
---
3 -E - - - - -;. :,/.. ~
'' ~ - - - - - -
-:- 3-c __ , I
T __
t 3 -E -----
;;/ 1---,---
--- : : :
--
-- - I I I
--
~ : :~ 8 = 0.2
''' X
- :o= 1.0:---
X
--
'
'
'
'
'
:o=o.s12
,__
X
'''
C=2 c =2 C=2
o is small enough . ois too large. Some ois as large as po ssible.
All of th e gra ph is within of the gra ph is beyond The graph just fits within
th e hori zo ntal lines. the hori zont al lines. the horizont al lines .
Fig ure 2-2a
40 Chapter
2: Properties
of Limits
If 8 = 1.0 as shown in the middle figure, some of the corresponding values of f(x)
will be above or below the horizontal lines . Thus, 1.0 is too big to be a suitable value
of 8 in this case . Example 1 shows you how to make the value of /5 "just right," as
shown in the figure on the right.
Example 1 For the point marked on the graph in Figure 2-2b, f(x) .
write the limit of f(x) as x approaches 2 using proper 7
. . .
r
v:0:
limit terminology. If E = 0.6, estimate to one decimal 6
place the largest possible value of /5 that can be used 5
to keep f(x) within E units of the marked point when 4
3 .: :
x is within /5units of 2.
2 ..
X
1 2 3 4 5 6 7
Figure 2-2b
Solution limf(x) = 4, as shown on the graph.
x- 2
Because E = 0.6, draw lines 0.6 unit above and f(x)
7 ..
below y = 4, as in Figure 2-2c. Where these
lines cross the graph, go down to the x-axis
5
and estimate the corresponding x-values to get
4
X ~ 1.6 and X ~ 2.8.
3
Thus, x can go as far as 0.4 unit to the left 2
of x = 2 and 0.8 unit to the right. The smaller of
X
these units, 0.4, is the value of /5.
l 2 3 ~ 5 6 7
Figure 2-2c
Example 2 For the function f(x) = (x - 2) 113 + 3, estimate the largest possible value of /5 that
can be used to keep f(x) within E = 0.8 unit of 3 when xis within /5 units of 2.
Solution Figure 2-2d shows the graph off, zooming in on the point (2, 3). Because E = 0.8,
f(x) must be between 2.2 and 3.8. Tracing the graph shows that for x = 2.51,
f(x) ~ 3.798957, which is okay. However, if x = 2.52, then f(x) ~ 3.8041452,
4J(x2 3
Ly _ - - )L ------- which is too big. Similarly, x can go as low as 1.49 and still have f(x) above 2.2.
I
I Thus, the largest possible value of 8 is about 0.51.
I
3~
I
: y= 2.2 If the value of E is very small, it may be easier to find /5 algebraically than it would
2 r -- - __ ----,:---------- ,x
1
1 2 3 be to find a suitable friendly window. The algebra will also let you show that there is
1
X= 2.51 = 3.798957 always a positive value of /5 that will do the job, no matter how small E is. Example
3 shows how.
Figure 2-2d
Example 3 For f(x) = (x - 2) 113 + 3, calculate algebraically the largest possible value of /5 that
can be used for a given value of E > 0 to keep f(x) within E units of 3 when x is
within /5 units of 2.
Sec
tion2-2: GraphicalandAlgebraic
Approaches
to theDefinitionof Limit 41
L~+
-,-{-
l;;ft:-
C
Solution The function f(x) must be between 3 - E and 3 + E. Write an inequality an d
solve it to find x.
3- E< (X - 2)113 + 3 < 3 + E f(x) must be between 3 - E and 3 + E.
-E < (X - 2) t / 3 < E Subtract 3 from all thre e members of the inequality.
-E 3 < X - 2 < E3
Cube all three member s of the inequality.
3 3
2- E <X < 2+E
.. largest possible value of 8 is E3 .
Because the steps are reversible, f(x) will be within E units of 3 whenever x is
within E 3 units of 2.
Example 4 For the function in Example 3, show that the value of 8 you get when E = 0.007
really does keep f(x) within 0.007 unit of 3 when xis within 8 units of 2.
Solution 8= E3 = 0.007 3 = 0.000000343
Keeping x within 0.000000343 unit of 2 is equivalent to making x satisfy the
inequality
l.999999657 < X < 2.000000343.
- 0.000000343 < X - 2 < 0.000000343 Subtra ct 2 from all thr ee memb ers.
-0.007 < (x - 2) 113 < 0.007 Take the cube root of all three members.
113
2.993 < (x - 2) + 3 < 3.007
2.99 3 < f (x) < 3.007
.. if xis within 0.000000343 unit of 2, then f(x) is within 0.007 unit of 3.
Example 5 Let f(x) = 0.2(2x).
a. Plot the grap h on your grapher.
b. Find the limit of f(x) as x approaches 3.
c. Calculate the maximum valu e of 8 that can be used for E = 0.5 at x = 3.
d. Show how to calculate a positive value of 8 for any E > 0, no matter how small .
Solution a. Figure 2-2e shows the graph.
b. Because the graph has no discontinuities, the limit is 1.6, the same as f(3).
c. Figure 2-2f shows how you might draw the graph on your paper. The
horizontal lines at y = 1.1 and y = 2.1 show that you want to keep the value
of f(x) within 0.5 unit of 1.6. To find the largest possible value of 8, you
must find out where these horizontal lines cross the gra ph .
X 0.2(r) = 1.1 ~ 2x = 5.5 ~
X=3 )'= 1.6 log2 x = log5.5 ~ xlog2 = log5.5 ~
log 5.5
Figure 2-2e X = log = 2.4594 ... "=' 2.46
2
42 Chapter
2: Properties
of Limits
f(x) Graph is within the ,
hori zo ntal lines. ~ /
:::::::
::::}o;::::: )!
1.1 --- - --- - --- - --- - --- - ___;!. 1,. - --L---- -- -- -
,,-/ ' ' ' :rick 6 on the
,,"' : : 6 : 6 :stee~ side.
.,,.- -----
- :use 1t ~n
: :~ :..........
____ , .both sid es.
-r, 1 ~. 54 ~1~1:
Figure2-2f
The value is rounded up because x. must be closer to 3 than 2.4594 is . . ..
Similarly, if f(x.) = 2.1, then
log 10.5
x. = log = 3.3923 ... ~ 3.39.
2
This value is rounded down, again to make x. closer to 3 than 3.3923 is . . ..
Thus, 6 could be
3 - 2.46 = 0.54 on the left side or
3.39 - 3 = 0.39 on the right side .
The one value of b in the definition of limit will be the smaller of these two,
b = 0.39 . Note that this value occurs on whichever side of x. = c the graph is
steeper.
d. To shovv that there is a value of b for any E > 0, no matter how small, repeat
the calculations for step c, using E instead of 0.5. Picking b on the right
(steeper) side of x. = 3 means that f(x.) will equal 1.6 + E.
0.2 (2") = 1.6 + E =>2" = 8 + SE =>
_ log (8 + SE)
log2 " = log(8 + :>E)=>x.log2 = log (8 +S E) =>x. = log2
. s: = log (8 + 5E) _
u log 2 3
You should check that b > 0. Becaus e (log 8) / (log 2) = 3, log (8 + SE)/log 2 will
be greater than 3. Thus, b will be positiv e, which was to be shown.
Sectio
n 2-2: Gra
phicalandAlgebra
ic Approaches
to theDefin
itionof Limit 43
,-
,-
I~ c
Problem Set 2-2
DoTheseQuickly
The following problems are intended to refresh your skills. You should b e able to do all
ten problems in less than five minutes.
01. Sketch an ellipse.
02. Sketch th e graph of a function with a removabl e discontinuity at (2, 5).
Q3. Evaluat e: 1-491
04. Evaluate: ./ -25
Q5. Evalu ate: 17 - 31
06. Fill in the blank: log (5 x 3) = log 5 -? - log 3.
Ql. Fill in the blank: log 53 = -7- log 5.
QB.Which axiom is illustrated by the equ ation 3(x + 2) = 3(2 + x )?
09. Sixty is 30% of what numb er7
Q10. Is f(x ) = sinx increasing or decr easing at x = 2?
For Problems 1-6, photocop y or sk etch the graph. For the point marked on the graph, us e
proper limit terminolog y to write the limit of f(x ). For the given valu e of E, esti mat e to
one decimal plac e the lar gest possible value of b that can be us ed to keep f (x) within E
units of the mark ed point when xis within b units of th e value shown.
2. X= 2,=0 .5 3. X = 6, E = 0.7
f(x ) f(x)
7 7 ~-t :
6
. .;.
5
-!
3
2 2
1 l ---
X X ' X
1 2 3 4 5 6 7 5 6 7 1 2 3 4 5 6 7
4. X = 4, E = 0.8 5. X = 5, E = 0. 3 6. X = 3, E = 0.4
f(x) f(x) f(x)
7 7 7 .. , ' ...... , .... , . ; ... ,.
6
5
)
f\ Li;
4
3
2
3/ :::
2 .
~ '-' ..... : ........ , ...;.
X X X
1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7
44 Chapter
2: Properties
of Limits
!ill
For Problems 7-12, do the following .
a. Plot the graph on your grapher (what interesting thing do you notice?)
b . Find the limit of the function as x approaches the given value.
c. Calculate the maximum value of i5 that can be used for the given value of E at the
point.
d. Calculate algebraically a positive value of i5 for any E > 0, no matter how small.
7. f(x) = 5 - 2 sin (x - 3) 8. f(x) = (x - 2) 3 + 3
X = 3, E = 0.5 X = 2,E = 0.5
9. f(X) = 1 + 3 "!)7 - X 10. f(x) = 1 + 2~- x
X = 6, E = 0.7 X = 4,E = 0.8
2
if X < 5
11. f(x) = { 0.25(x 2- 5) + 2, 12. f(x) = 6 - 2(x - 3) 213
(x - 5) + 2, if X ~ 5
X = 3, E = 0.4
X = 5, E = 0.3
13. Limits Applied to Derivatives Problem: Suppose you start driving off from a traffic
light. Your distance, d(t) feet, from where you started is given by
d(t) = 3t 2 ,
where t is time in seconds since you started.
a. Figure 2-2g shows d(t) versus t. Write as an algebraic fraction
the average speed, m(t), for the time interval from 4 tot.
b . Plot the graph of function m on your grapher. Use a friendly d (t)
window that includes t = 4. What does this graph have at the 48
point t = 4? Sketch the graph.
4 t
c. The speed you are going at the instant t = 4 is the limit of the
average speed as t approaches 4. What does this limit appear Figure 2-2g
to equal? What are the units of this limit?
d . How close to 4 would you have to keep t for m(t) to be within 0.12 unit of the
limit? This is an easy problem if you simplify the algebraic fraction first.
e. Explain why the results of this problem give the exact value for a derivative.
2-3 The Limit Theorems
Suppose that f(x) is given by the algebraic fraction
2
f(x ) = 3x
- 48 .
x- 4
There is no value for f (4) because of division by zero. Substituting 4 for x gives
f(4) - 3 .442__443 -- 20. f(J, ) is und efined becau se it has an indet erminat e form .
Section 2-3: TheLimitTheorems 45
k,
-, ' ,-
c
Because th e numerator is also zero, there may be a limit of f (x ) as x approaches 4.
Limits such as this arise when you try to find exact values of deriv atives . You may
ha ve seen thi s fra ction alr ead y in Probl em 13 of Problem Set 2-2. Simplifying the
fraction befor e substitutin g 4 for x gives
f (x) = (3x + 12) (x - 4 )
x- 4
= 3x + 12, pro vid ed x * 4.
Surprisin gly, th e limit can be found by sub stitutin g 4 for x in th e simp lified
expr ession .
lim = 3 (4) + 12 = 24
x- 4
From Section 1-5, re call that 0/ 0 is called an indeterminate form. Its limit can
be diff erent numb ers dep endin g on ju st what expr ess ions went to zero in the
num erato r and denominator. Fortun ately, there are prop erti es (called the limi t
theorems) that allow you to find su ch limit s by m akin g substitutions, as shown
abov e. 1n this section you will learn th ese prop erti es so tha t you can find exact
values of der ivatives and int egral s th e way Newton and Leibniz did mo re than 300
year s ago.
OBJECTIVE For th e prop erti es list ed in the table in thi s section, be able to state them, use them in a
proof, and explain why they are tru e.
Limitof a Product
ora Sumof TwoFunctions
Suppos e th at g (x ) = 2x + 1 and h (x) = 5 - x . Let f (x ) be th e produ ct: f (x ) =
g(x) h (x) , or
fl\'
y f (x) = (2x + 1)(5 - x) .
,c;r
10 You ar e to find the limit of f (x) as x appro aches 3. Figur e 2-3a shows the grap hs of
"
..' .
....- :
funct ions f, g, and h. Direct substituti on gives
..:....~-
........ .... X f (3) = (2 3 + 1)(5 - 3) = (7) (2) = 14.
X = 3.1 3 y = 13.68
The import ant idea concernin g limit s is that f (x ) st ays clos e to 14 when xis kept
Fig ure 2-3a close to 3. You can demon strat e thi s fa ct by makin g a table of valu es of x, g(x ) , h (x ) ,
and f (x).
X g(x) h (x ) f (x) = g(x ) h (x )
2.95 6.9 2.0 5 14.145
2.96 6.92 2.04 14.1168
2.97 6.94 2.03 14 .0882 When g (x) and h (x ) are close
2.98 6.96 2.02 14.0592 to 7 an d 2, respectively,
2.99 6.98 2.01 14.0298 f(x) is close to 14.
3.01 7.02 1.99 13.9698
3.02 7.04 1.98 13.9392
3.03 7.06 1.97 13.9082
3.04 7.08 1.96 13.8768
46 Chapter 2: Propertiesof Lim
its
You can keep the product as close to 14 as you like by keeping x close enough to
3, even if xis not allowed to equal 3. From this information you should be able to
see that the limit of a product of two functions is the product of the two limits. A
similar property applies to sums of two functions . By adding the values of g(x) and
h(x) in the table above, you can see that the sum g(x) + h(x) is close to 7 + 2, or 9,
when xis close, but not equal, to 3.
Limitof a Quotient
of TwoFunctions
The limit of a quotient of two functions is equal to the quotient of the two limits,
provided that the denominator does not approach zero . Suppose that function f is
defined by
f(x) = g (x) 2x + 1
=
h(x) 5- X '
and you want to find the limit of f(x) as x approaches 3. The values of g(3) and
h(3) are 7 and 2, respectively. By graphing or by compiling a table of values, you
should be able to see that if xis close to 3, then f(x) is close to 7/2 = 3.5. You can
keep f(x) as close as you like to 3.5 by keeping x close enough to 3. (When xis equal
to 3, f(x) happens to equal 3.5, but that fact is of no concern when you are dealing
with limits.)
There is no limit of f(x) as x approaches 5. The denominator goes to zero, but the
numerator does not. Thus, the absolute value of the quotient becomes infinitely
large, as shown in the table below. Figure 2-3b shows that the graph off has a
vertical asymptote at x = 5.
X g(x) h(x) f(x) = g(x) / h(x)
4.96 10.92 0.04 273
((x) 4.97 10.94 0.0 3 364 .6 ...
so 4.98 10.96 0.02 548
4.99 10.98 0.01 1098
5.00 11.00 0.00 None (infinite)
5.01 11.02 -0 .01 - 1102
5.02 11.04 - 0.02 - 552
5.03 11.06 -0.03 - 368.6 ...
Figure 2-3b 5.04 11.08 -0.04 - 277
The following box summarizes the important properties of limits . Proofs of these
properties involve some exciting algebra ! They all revolve around finding a value of
o that is small enough to make f(x) stay within E units of the limit whenever xis
within o units of c.
Section 2-3:The Lim
it Theorems 47
SomePropertiesof Limits:TheLimitTheorems
limitof a Productof TwoFunctions: Iflimg(x)
x- c
= L1 and limh(x)
x-c
= L2,
then lim[g(x) h(x)] = limg(x) limh(x) = L1 L2.
x- c x-c x-c
Informal words: Limit distributes over multiplication; or, the limit of a product equals the
product of the limits.
Limit
of a SumofTwoFunctions : If limg(x)
x-c
= L1 and lim h(x)
x- c
= L2,
+ h(x)] = limg(x) + limh(x) = L 1 + L 2.
then lim[g(x)
x-c x- c x-c
Informal words: Limit distributes over addition; or, the limit of a sum equals the sum of
the limits.
limitof a Quot If Jim g(x) = L 1 and lim h(x) = L2, where L2 * 0,
ientof TwoFunctions: x-c x -c
. g(x ) V...11Jg(x) Li
x- c -h (X ) = 1Im
thenhm h()X = L2 .
x-c
Informal words: Limit distributes over division, except for division by zero; or, the limit of
a quotient equals the quotient of the limits .
Limitof a Constant Timesa Function : If limg(x) = L,
x-c
then lim[k g (x)] = k lim g (x) = kL.
x-c x-c
Informal words: The limit of a constant times a function equals the constant times the
limit.
Limit
of theIdentity
Function
: limx
x- c
=c
Informal words: The limit of x as x approaches c is simply c.
Limit
of a Constant
Function
: If f(x) = k, where k is a constant, then limf(x) = k.
x- c
Informal words: The limit of a constant is that constant.
Limit
of a Composite
Function:
If x - c ~ u - k, then limf (u) = limf (u) .
x-c u- k
Informal words: If f(u) is close to L when u is close to k, and if u is close to k when xis
close to c, then f (u) = f(g(x)) is close to L when xis close to c; or, you can replace u - k
with x - c in a limit provided x - c implies u - k.
48 Chapter
2: Properties
of Limits
2
Example 1 . properties
Use t h e 1mut 3xx _- 48 = 24 . Justl'f y eac h step.
to prove t h at 1,;!!}
4
Solution . 3x 2 -48 . (3x+12)(x-4)
11m --- = 11m ------- Algebra.
x--1 X - 4 x--1 X - 4
The canceling can be don e beca use the def-
= lim(3x + 12) inition of limit keeps x from equaling -t.
x - -1
= lim 3x + lim 12 Limit of a sum of two functions.
x- 4 x - -1
=3 limx + liml2 Limit of a consta nt tim es a function.
x- 4 x--1
= 3(4) + Jim
x--1
12 Limit of x as x - c equals c.
=3( 4) + 12 Limit of a constant is that constant.
= 24,Q.E.D. The abbreviation Q.E.D. repr esents
the Latin quad erat demonstratum,
meaning "which was to be demon-
strated."
Problem Set 2-3
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Find the limit of 13x/x as x approaches zero.
Q2. Sketch the graph of a function if 3 is the limit as x approaches 2 but f(2) is undefined.
Q3. Sketch the graph of a function that is decreasing slowly when x = - 4.
Q4. Sketch the graph of a quadratic function.
QS. Sketch the graph of y = x 3 .
Q6. Factor: x 2 - l 00
Ql. Thirty is what percentage of -10?
QB.How far do you go in 20 min at 30 mi/hr?
Q9. Simplify: (12x 30 )/(3x 10)
Q10. What is meant by definite integral?
l. Limit of a Function Plus a Function Problem: Let g(x) = x 2 , and let h(x) = 12 / x. Plot
the two graphs on your grapher, along with the graph of f(x) = g(x) + h(x). Sketch
the result, showing that the limit of f(x) as x approaches 2 is equal to the sum of the
limits of g(x) and h(x) as x approaches 2. Make a table of values which shows that
f(x) is close to the limit when xis close, but not equal, to 2.
2. Limit of a Constant Times a Function Problem : Plot g(x) = x 2 and f(x) = 0.2x 2 on your
grapher. Sketch the result. Find the limit of f(x) as x approaches 3 and the limit of
g(x) as x approaches 3. Show that the limit of f(x) is 0.2 multiplied by the limit of
g(x) . Make a table which shows that f(x) is close to the limit when xis close, but not
equal, to 3.
3. Limit of a Constant Problem : Let f(x) = 7 . Sketch the graph off. (Don't waste time
using your grapher!) Show on the graph that the limit of f(x) as x approaches 3 is
equal to 7. Does it bother you that f(x) equals 7, even if xis not equal to 3?
Sec
tion2-3:TheLimitTheorems 49
,-,
',-
~ c
4. Limit of x Problem: Let f(x) = x. Sketch the graph off . (Don't waste time using your
grapher!) Then explain why the limit of f(x ) as x approaches 6 must be equal to 6.
5. Limit of a Product Problem: Let f(x) = x 2 tan x. Write the values of 1.52 and tan 1.5,
then multiply them to find f(l.5). By experimenting on your calculator, find a value
of D that keeps f(x) within 0.01 unit of f(l.5) when x is within [J units of 1.5. When
x = 1.5 + D,how close are x 2 and tan x to 1.52 and tan 1.5, respectively?
6. Limit of a Quotient Problem: Let r(x ) = 2" / sinx. Write the values of 23 and sin 3. Then
divide them to find r(3). By experimenting with your calculator, find a value of [J that
keeps r(x) within 0.1 unit of r(3) when xis within [J units of 3. When x = 3 + D,how
close are 2" and sin x to 23 and sin 3, resp ectively?
For Problems 7-17, complete the following.
a. Plot the graph on your grapher, using a friendly window that includes the given
x-value . Sketch the result.
b. Find the limit of f(x) as x approaches the given value.
c. Prove that your answer is right by using the appropriate limit properties.
7. f(x) = 3x - 7, x-4 8. h(x) = - 5x + 23, x- 2
9. f(x) = x 2 - 9x + 5, x-3 10. p(x) = x 2 + 3x - 6, x- - 1
x2 - 4x - 12 x2 + 3x - 40
11. r(x) = x- - 2 12. f(x) = X - 5
x+2 ' x- 5 '
(For Problems 13-17, you will have to recall how to factor a higher degree polynomial or
how to long divide or synthetically divide the numerator by the denominator.)
x3 - 3x 2 - 4x - 30 x3 + x 2 - 5x - 21
13. f(x) = , x - 5 14. f(x) = , x - 3
x- 5 x- 3
x 3 - 4x 2 - 2x + 3
15. f(x) = l , X - - 1 16. f(x) = x 3 - l0x 2
+ 7x - 11, X - 2
x +
x-i - llx 3 + 2lx 2 - x - 10
17. f(X) = , X- 2
x- 2
18. Check the An swer by Tabl e Problem: For Problem 13, let E = 0.1. Calculate the
corresponding value of [J. (Recall the quadratic formula 1) Then make a table containing
six conveniently-spaced x-values, each within [J unit of 5, and show that f(x) is really
within 0.1 unit of the limit for all these value s.
19. Limit of a Composite Function Problem: Let g (x ) = cos x and let f(u) = tan u. Then
f(x) is the composite function
f(g(x)) = tan (cosx).
In this problem you will show that the limit of f(g(x)) as x approaches O is the same
as the limit of f(u) as u approaches 1.
a. Sketch a graph that shows cos x is close to 1 when x is clos e to 0.
b. Plot the graph of y = tan x on your grapher. Use a friendly window for which x = 1
is a grid point . Show that y stays close to tan 1 when x is kept close to 1.
c. Plot the graph of y = tan (cos x) on your grapher and sketch the result. What value
does y stay close to when x is clos e to O?
50 Chapter
2: Properties
of Limits
d. Complete the equations .
lim x-o cosx = -?- limu - i tanu = -?- limx-o tan(cosx) = -?-
20. Pizza Delivery Prob lem: Ida Livermore starts off on her route. She records her truck's
speed, v(t) miles per hour, at various numbers of seconds, t, since she started.
a. Show Ida that these data fit the equation v(t) = 5t 112 .
v ( t)
0 0
1 5
4 10
9 15
16 20
b. The truck's acceleration, a(t), is the instantaneous rate of change of v(t). Estimate
a(9) by using v(9) and v(9.001). Make a conjecture about the exact value of a(9).
What are the units of a(t)?
c. a(9) is exactly equal to the limit of [v(t) - v(9)] / (t - 9) as t approaches 9.
Factor the denominator as a difference of two "squares ." Then find the limit as
t approaches 9 by applying the limit properties . Does this limit agree with your
conjecture in 20b?
d. Approximately how far did Ida's truck go between t = 1 and t = 9?
21. Exact Derivative Prob lem: Let f(x) = x 3 .
a. Find, approximately, the derivative of f at x = 2 by dividing the change in f(x)
from x = 2 to x = 2.1 by the corresponding change in x.
b. In 21a you evaluated the fraction [f(x) - f(2) J/ (x - 2) to get an approximate value
of the derivative . The exact value is the limit of this fraction as x approaches 2.
Find this limit by first simplifying the fraction . Prove that your answer is right by
citing limit properties.
c. Plot the graph off. Through the point (2, 8), construct a line whose slope is the
value of the derivative in 21 b. What relationship does the line seem to have to the
graph7
22. Find, approximately, the derivative of f(x ) = o.7 x when x = 5.
23. Find, approximately, the definite integral of f(x) = 1.4'' from x = 1 to x = 5.
24. Mathematical Induction Problem-The Limit of a Power : You recall that x 2 = x x, so
the limit of a product property can be used to prove that
2
limx = c2
x-c
Section2-3: TheLimit Theorems 51
I! ~w-
L -1
'
1
Prove by mathematical induction that
limx" = c"
x- c
for any positive inte ger value of n. Th e recursive definition of x", which is x" = x . x 11- 1,
should be helpful in doing the induction part of the proof.
25. Journal Problem: Updat e your calculus journal. You should consider the following.
The one most important thing you have learned since your last journal entry
What you now understand more fully about the definition of limit
How the shortened definition of limit corr esponds to the definition you learned in
Chapter 1
Why the limit properties for sums, products, and quotients are so obviously true
The meaning of the limit of a composite function property
What may still bother you about the definition of limit
2-4 Continuity
A function such as
x2 - 9
g (x ) = - -3
x-
has a discontinuity at x = 3 because the denominator is zero there. It seems
reasonable to say that the function is "continuous" everywhere else because the
graph seems to have no other "gaps" or "jumps" (Figure 2-4a). In this section you
will use limits to define precis ely the property of continuity.
g(x)
Figure 2-4a
OBJECTIVE Define continuity. Learn the definition by using it in several ways .
Figures 2-4b through 2-4g show graphs of six functions, some of which are
continuous at x = c and some of which are not.
52 Chapter
2: Properties
of Limits
The first two functions have a limit as x approaches c. In Figure 2-4b, f is discon-
tinuous at c because there is no value of f(c). In Figure 2-4c, f is discontinuous at
c because f(c) =t=L. Both are removable discontinuities. The value of f(c) can be
defined or redefined to make f continuous there.
"/
f(x) f(x)
((c) ; Step
: di scon tinuity
X X X
Not continuous. No f(c) Not continuous. f(c) ct L Not continuous. No L
Figure 2-4b Figure 2-4c Figure 2-4d
In Figure 2-4d, f has a step discontinuity at x = c. Although there is a value of f(c),
f(x) approaches different values from the left of c and the right of c. Thus, there is
no limit of f(x) as x approaches c. A step discontinuity cannot be removed simply
by redefining f(c).
In Figure 2-4e, f is discontinuous at x = c because it has a vertical asymptote at c.
There is neither a value of f(c) nor a finite limit of f(x) as x approaches c. Again,
the discontinuity is not removable just by redefining f(c). In Section 2-5, you will
study such infinite limits.
Figures 2-4f and 2-4g show graphs of functions that are continuous at x = c. The
value of f(c) equals the limit of f(x) as x approaches c. The branches of the graph
are "connected" by f(c).
f(x) f(x) f(x)
Vertical
asymptote
X X X
Not continuous . No f(c) , no L Continuous. L = ((c) Continuous. L = f(c)
Figure 2-4e Figure 2-4f Figure 2-49
The above examples lead to a formal definition of continuity.
I
Section 2-4: Con
tinuity 53
Definition:Continuity
: Function f is continuous at x = c if and only if:
Continuityal a Point
1. f(c) exists,
2. limf(x) exists, and
x- c
3. limf(x)
x- c
= f(c) .
Continuity Function f is continuous on an interval of x-values if and only if
onan Interval:
it is continuous at each value of x in that interval.
Not e that the graph can have a cus p (an abrupt change in direction) at x = c and still
be continuous there (Figure 2-4f). The word bicuspid in relation to a tooth comes
from the same root word .
Definition: Cusp
A cusp is a point on the graph at which the function is contin uous but the derivative is
discontinuous.
Informal words: A cusp is a sharp point or an abrupt change in direction.
Figures 2-4h, 2-4i, and 2-4j illustrate why a function must satisfy all three parts of
the continuity definition. The graph in Figure 2-4h has a limit as x approaches c but
it has no function value . The graph in Figure 2-4i has a function value, f(c), but no
limit as x approaches c. The graph in Figure 2-4j has both a function value and a
limit, but they are not equal.
"'' /
f(x) f(x)
f(c) = 7 ~
: discontinuit y
4 -
X X X
C
Limit but no value for f(c). Value f( c) but no Unequal values for
value for limit. limit and f(c).
Figure 2-4h Figure 2-4i Figure 2-4j
The graph in Figure 2-4i illustrates the concept of a one-sid ed limit. If x is close to
c on the left side, f(x) is close to 4. If xis close to c on the right, f(x) is close to 7.
The symbols shown in the following box are us ed for left- and right-sided limits .
54 Chapter
2: Proper
tiesof Limits
!M&F & ii t@Uttt@
Symbols:One-SidedLimits
lim f(x) x - c from the left (through values of x on the negative side of c)
x-c -
lim f(x) x - c from the right (through values of x on the positive side of c)
x -c+
. f {0.Sx + 3, if x < 2
Example 1 Lett h e function (x) = - x2 + Gx _ 2 , l'f X 2. 2.
a. Draw the graph .
b. Find lim f(x) and lim f(x).
x-2 - x-2 +
c. Tell whether or not f(x) is continuous at x = 2.
Solution a. Figure 2-4k shows the graph. You can probably sketch it without the grapher.
f(x)
There is a step discontinuity at x = 2 because the two branches do not
connect.
6
b . lim f(x) = 4, because f(x) is close to 4 when xis close to, but less than, 2.
x-2 -
lim f(x) = 6, because f(x) is close to 6 when xis close to, but greater than, 2.
x -2+
c. The function f is not continuous at x = 2 because f(x) has no limit as x - 2.
The value of f(x) is close to no one number when x is near 2 but x * 2.
X
Example 1 illustrates the relationship between limits and one-sided limits. A function
has a (two-sided) limit at x = c if and only if both one-sided limits are equal.
Figure 2-4k
Property:EqualLeft and RightLimits
L = limf(x) if and only if L = lim f(x) and L = lim f(x) .
x-c .v.-c- x-c +
. { lx - 3 1+ 4, ifx z. 2
Example 2 Let the function h(x) = k X 2 , .f
l X < 2.
a. Find the value of k that makes the function continuous at x = 2.
b. Plot and sketch the graph.
Solution a. For h to be continuous at x = 2, the left and right limits must be equa l. Set
the expressions for the left and right branches equa l to each other at x = 2
and solve for k .
12 ~ 3 1+ 4 = k(2 2 ) ~ 5 = 4k ~ k = 1.25
b. To make the grapher plot only in a given domain, you can use the logic test
menu to get inequality signs. For instance, for the second branch of the
function, enter the following equation.
2
l.25x
Y2= x<2
Section24:Continuity 55
L~
+-,1:--
!7.t
C
If x is not less than 2, the quantity (x < 2) is zero. Dividing by zero makes
y2 undefined, and the grapher plots nothing. Figure 2-41 shows that the two
branches "link up" to each other at x = 2.
Figure 2-41
Problem Set 2-4
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
QI. What is meant by the derivative of a function?
Q2. What is meant by the definite integral of a function?
Q3. If f(x) = 200x + 17, what is the maximum value of 8 that will ensure f(x) is within 0.1
unit of f(3) when xis within 8 units of 37
Q4. Draw a pair of alternate interior angles. f(x)
QS. What type of function has a graph like that in Figure 2-4m7
Q6. Sketch the graph of y = cos x.
Ql. Factor: x 2 + 5x - 6
QB.Evaluate: 53 200 1 / 5 3 2000 X
Q9. Evaluate: 5!
QIO. Quick! Divide 50 by 1/ 2 and add 3. Figure 2-4m
For Prob lems 1-20, tell whether or not the graph illustrates a function that
a. has left and right limits at the marked value of x,
b. has a limit at the marked value of x,
c. is continuous at the marked value of x. If it is not continuous there, tell why not.
4.
f (x)
56 Chapter
2: Propert
iesof Limits
5. 6. f (x) 7. 8.
X
I
{
9. f (x) 10. f (x) 11. f (x) 12. f(x)
\ -------
r
X X
-2 3
13. 14. 15. 16.
-----~-,__ 1
X
17. f (x) 18. f (x) 19. f(x) 20. f(x)
----er-
X X
c X
C
C
For Proble m s 21-42, sketc h the grap h of a function that has th e indicat ed featur es .
21. Is continuous at x. = 3 bu t has a cusp there .
22. Is con tinu ous at x. = 4 and is "smooth" there .
23. Has no valu e of f(5) but has a limit as x. approaches 5.
Section2-4: Continuity 57
24. Has a value of f (-2) but has no limit as x approaches -2 .
25. Has a vertical asymptote at x = -3 .
26. Has a vertical asymptote at x = 6.
27. Has a value of f(2 ), a limit as x approaches 2, but is not continuous at x = 2.
28. Has a negative value of f (5) and a positive limit as x approaches 5.
29. Has a step dis continuity at x = - 2, and f (-2) = 10.
30. Has a step discontinuit y at x = - 4 and no valu e of f( - 4) .
31. The value of f(2) = 3, the function is increasin g slowly at x = 2, and the limit of f(x)
as x approaches 5 is equal to 6, although there is no value off( 5).
32. The value of f (3) = 10, the function is decreasing ra pidl y at x = 3, and the limit of
f (x) as x approaches 7 is 1, although there is no value of f (7) .
33. The value of f( 4) is negativ e, but the function is incr easing rapidly at x = 4.
34. The value of f (-2) is positi ve, but the function is decreasing slowly at x = - 2.
35. The limit of f(x ) as x approa ches - 4 is 8, but there is no valu e of f (- 4).
36. The limit of f (x ) as x approaches 5 is -2, and the value of f (5) is also -2 .
37. The limit of f (x) as x approach es 1 is 4, but f(l ) = 6.
38. The limit of f (x) as x approach es 7 is 2, but f (7) = -3.
39. The value of f(3 ) = 5, but th ere is no limit of f(x) as x approaches 3 and no vertical
asymptote there.
40. The value of f (-3) = 2, but there is no limit of f(x) as x approaches -3 and no vertical
asymptote there .
41. The valu e of f (x) ha s no limit as x approaches -9 , no valu e of f( -9) , and no vertical
asymptote at x = - 9, yet f (x ) is define d for all values of x in a neighborhood of -9,
not including - 9 itself.
42. The value of f(x) is increasin g if x < 6 and decr easing if x > 6, but there is no limit of
f (x) as x - 6.
For Problems 43-46, tell where, if anywhere, the fun ction is dis continuous.
5
43. f (x) = : : : 44 . f(x) = xx:
11
45. g(x) = tanx 46. g(x ) = cosx
For Problems 47 - 52, the function is discontinuous at x = 2. Tell which part of the
definition of continuity is not met at x = 2. You may plot the graph on your grapher.
(Note: The symbol int (n) indicates the greatest int eger less than or equal to n. Graph
in dot mode .) Sketch the graph.
47. f(x) = x + int(cos rr x) 48. g(x) = x + int (sin rr x)
49 . s(x) = 3+~ 50. p(x) = int (x2 - 6x + 9)
58 Chapter
2: Properties
of Limits
if x * 2
1
5 1. h(x) = sin (x; 2) 52 _ f(x) = {x ,+ (2 - x) - ,
x- 3 if X = 2
For Problems 53-58 do the following .
a. Sketch the graph of the function . Use the grapher only if necessary. Tell whether
or not the function is continuous at the value of x where the rule for the function
changes .
b . Tell what the left- and right-limits are and whether or not the function has a limit
at the value of x where the rule for the function changes .
2 2
53. d(x) = {7 - x , ifx < 2, 54 _ h(x ) = {4 -x , ifx < l
5 - X, if X > 2 X + l, if X ~ l
{93x if X < 2 {2-x'3 if X .:5 - 1
r-,
55. m (x ) = ~ x, if X ~ 2
56. q (x) =
if X > - l
X + '
57. T(x) = 1 x -2
3, l
X + l,
if X < 2
if X = 2
if X > 2
58. Z(x)= x 2- x
3,
2,
if x
if X = 0
if X = l
* 0 and x * l
For Problems 59-66, find the value of th e constant k that will make the function continuous
where the defining rule changes by making the left limit of the left branch equal the right
limit of the right branch . Sketch th e graph, showing that the two branches "link up ."
59. f(x ) = {o-03.4x+k 2, iffx .s; 11 60. h (x) = {xk2, ~ffx < 22
. X + , 1 X > - X, 1 X ~
61. g(x )= {~x~x
2
, ~~: : ~ 62.
4
f(x)= {~~: ; ,
1
!~:
:~
kx 2 if X .:5 3 + 5, i_fx <- 1
63 u(x ) = { kx ~ 3, if x > 3 64 . V(x) = {kx
kx 2 , 1f X ~ - l
2
65 . f(x) = {0.4x + k , if x < l 66. f(x ) = {k 2 - x2,_ if x < 2
kx + 2.4, if X ~ l l.5kx, 1fx ~ 2
67. Two Constants Prob lem: Let a and b stand for constants and let
b - X, if X .:5 1
f(x ) = { a (x - 2) 2 , if x > 1.
a. Find an equation relating a and b if f is to be continuous at x = 1.
b . Find b if a = -1. Show by graphing that f is continuous at x = l for these values
of a and b.
c. Pick another value of a and find b . Show that f is continuous for th ese values of a
and b.
10-20
68. Surprise Function Problem! Let f(x) = x + 3 + --.
X - 1
a. Plot the graph on the grapher.
b. What appears to be the limit of f(x ) as x approaches 1?
Section2-4: Continuity 59
c. Show that f(x) is very close to the number in 68b when x = 1.0000001.
d. Function f is not continuous at x = 1 because there is no value for f(l). What type
of discontinuity is there at x = 1? Be careful!
69. Continuity of Polynomial Functions: The general polynomial function of degree n has
an equation of the form
P(x ) = ao + a1x + a 2x 2 + a 3x 3 + + anx n.
Based on the closure axioms for real numbers and the properties of limits you have
learned, explain why any polynomial function is continuous for all real values of x.
70. The Signum Function: Figure 2-4n shows the graph of the
signum function, f(x ) = sgn x. The value of the function is 1
when xis positive, - 1 when xis negative, and O when x is zero. f(x) = sgn x
This function is useful in computing for testing a value of x
to see what sign it has (hence the name signum). The formal
definition is given below .
1, if X >0 X
1
sgn x = 0, if X =0 -----0 -1
{
- 1, if X <0
In this problem you will explore various compositions of the Figure 2-4n
signum function.
a. Does r(x) = lsgn x i have a limit as x approaches zero? Does it have a function
value at x = O? Is it continuous at x = O?
b. Sketch the graph of g (x ) = 3 sgn (x - 2).
c. Sketch the graph of h(x) = x2 - sgn x.
d. Show that the function a (x ) = lx l/ x is equal to sgn x for all x except zero.
e. Sketch the graph of f(x) = cos x + sgn x.
f. Suppose you want to plot p(x) = x 317 on a calculator or a computer that will not
handle powers with negative bases. With the help of the signum function, write
x 317 as a single expression in a form that will give the correct answer for both
positive and negative x-values, but that does not involve raising a negative number
to a power. Sketch the resulting graph on your paper.
2-5 Limits Involving Infinity
Nwnb er of doctors
~!~ ~
Suppose that there is an increase in the demand for doctors in a particular
community. The number of people who choose to pursue that career will increase
Numb er of doctors to meet the demand. After a while there may be too many doctors, causing the
stablizes at 1000 .
number of people who want to enter the medical profession to decrease. Eventually
Time
the number of doctors stabilizes somewhere, say at 1000 (Figure 2-5a). This steady-
state value is called the limit of the number of doctors as time approaches infinity.
Figure 2-5a
60 Chapter
2: Properties
of Limits
Another type of limit involving infinity can be visualized by imagining you are
pointing a flashlight straight at a wall (Figure 2-Sb). If you begin to turn with the
flashlight in your hand, the length of the light beam increases. When the angle is
rr / 2 radians (90), the beam is parallel to the wall, so its length becomes infinite .
tL engt h
i"[b
"m :Infinit
t e
:limit
I
1
Ang]C, X
I ,E
2
Figure 2-5b
In this section you will see how the definition of limit must be modified to take into
account cases where x approaches infinity or where the limit is infinite.
OBJECTIVE Find limits of functions where either x goes to infinity, the limit is infinite , or both .
The definition you have learned for limit tells what happens to the y-value of a
function when xis kept close to a certain number, c. If x increases without bound,
there is no such number it can be kept close to. Suppose the function value can
be made to stay arbitrarily close to a given number L, like the 1000 in Figure 2-Sa,
just by making x large enough. Then Lis the limit of the function as x approaches
infinity. The following definition records this fact.
Definition:Limitas x Approaches
Infinity
L = x-oo
Jim f(x) if and only if for any number E > 0, there is a number D > 0 such that if
x > D, then f(x) is within E units of L.
A similar definition holds for limx--oo f(x), and the same type of reasoning can be
applied if the function value becomes infinite as x approaches c. There is no number
L that f(x) can be kept close to, so the following definition is used.
/'(x)
JOO
((x) become s
infinite as Definition:InfiniteLimit
x becomes
infinite .
limf(x) is infinite if and only if for any number E > 0 there is a number 6 > 0 such that if
x-c
xis within 6 units of, but not equal to, c, then lf(x) I is greater than .
X
10
Figure 2-5c If f(x) becomes infinite as x becomes infinite, both modifications to the usual
definition of limit are used. Figure 2-Sc shows the graph of f(x) = x 2 , which
Section
2-5: LimitsInvolving
Infinity 61
' L:af_
1 .-
, :-
c
illustrates this property. ln plain English, this property says "If you can make y
as big as you like by making x big enough, then the limit of f(x) is infinite as x
approaches infinity ."
Definition:InfiniteLimitas x Approaches
Infinity
lim f(x) is infinite if and only if for any number E > 0, there is a number D > 0 such that
x-oo
if x > D, then lf(x)I > E.
A similar definition holds for limx- - oof(x) .
A NoteonInfinity,#Undefined,"
andtheProper
Useof the= Sign
The = in mathematics is used to connect two equal numbers. Because infinity is not
a real number, it is more appropriate to say "the limit of f(x) is infinite" rather than
saying it "equals infinity ." Sometimes, when it is important to describe briefly the
direction in which a function goes to infinity, the following statements are used.
limf(x) = oo or limf(x) = -oo
x-c x-c
The symbols oo and - oo are convenient shorthand for describing the behavior of the
function at x = c. You should realize that the limits still do not exist. The notation
simply explains why the limits do not exist . They are undefined because they are
infinite. Note that a limit may be undefined without being infinite. For instance, the
limit f(x) = lx l / x as x approaches zero is un defined because of a step disc ontinuity
there.
The statement "x - oo" is also somewhat misleading. Although pronounced "x
approaches infinity," the statement really means "x gets infinitely far away from
zero." No matter how large x gets, it can never be close to infinity 1
Properties
of Limits
Involving
Infinity
Two properties come directly from the above definitions . They concern what
happens to the reciprocal of a function if the value of the function approaches
f(x) either zero or infinity . For instance, suppose
5 1
f(X) = X - 3 .
As x approaches 3, the denominator goes to zero . When you use a grapher (Figure
2-Sd) you can see that the fraction itself becomes infinitely large in the positive
direction if x is to the right of 3, and infinitely large in the negative direction if x
is to the left of 3. On the other hand, if x approaches infinity or negative infinity,
Figure 2-5d the denominator becomes infinite and the fraction itself approaches zero . These
properties are summarized in the following box .
62 Chapter
2: Propert
iesof Limits
Property:Reciprocalsol ZeroandInfinity
1
If f(x) = - ( ) and limg(x) = 0, then limf(x) is infinite.
g X x -c x-c
1
If f(x) = -g (X ) and limg(x)
x- c
is infinite, then limf(x)
x-c
= 0.
The same properties app ly if x approaches infinity.
1 1
Informal words: The form - is infinite. The forms _!_and - - have a zero limit.
0 00 -00
Note that the above property holds if c is replaced by oo or - oo.
Example 1 If f(x) = tan - 1 x, find limx-oof(x). Find a number D such that f(x) is within E = 0.1
unit of the limit whenever x > D . Illustrate by graph the meaning of D and E.
Solution You recall that the inverse tangent graph has a horizontal asymptote at y = rr / 2.
Thus,
y limf(x) = !I._
2 0.1 y=~ ,v.-oo 2
__ j _____ _ 2__ _ _____ _
--r ------- To find D, let
tan - 1D = f - 0.1 = 1.470796 3 ...
X :. D = tan 1.4707963 . .. = 9.966644 ...
9.966 ...
As indicated in Figure 2-Se, tan - 1x is closer than 0.1 unit to rr / 2 whenever xis
Figure 2-5e greater than 9.966 . . .. Because this calculation could be done for any number
E > 0, rr / 2 is the limit of f(x) as x approaches infinity.
4
Example 2 For f(x) = x - , show that limx- z+ f(x) = -oo . Find a value of c5such that lf(x) I > 10
x- 2
whenever x is within c5units of 2 on the positive side but not equal to 2.
Solution Figure 2-Sf shows that there is a vertical asymptote at x = 2 because of division
by zero, so the limit is infinite . Because f(x) has the form (positive) 7 (negative)
when x is close to 2 on the positive side, the limit is - oo. To find the value of c5,
set f(x) = - 10.
- lO = x - 4
x- 2
- lOx + 20 = x - 4 =>x = }= 2.1818 .. .
... c5= 2.1818 ... - 2 = 0.1818 .. .
Figure 2-5f Thus, lf(x) I > 10 whenever xis within 0.1818 ... unit on the right of 2.
Sec
tion2-5: LimitsInvolving
Infinity 63
~
Problem Set 2-5
DoTheseQuickly
Th e follovving probl ems ar e int ended to refresh your skills. You should be able to do all
ten probl em s in les s than five minut es. Refer to Figur e 2-Sg.
QI. lim x-1 f (x) = - ?-
Q2. lim x- 2 f (x ) = - 7 -
f (x)
Q3. limx-3 f (x) = -?- -I
~
Q4. lim x-4 f (x) = -?- 3
QS. lim x- s f (x) = - 7 - 2
Q6. Is f continuous at x = 1? X
Ql. Is f continuous at x = Z? 1 2 3 4 5 6
QB. is f continuous at x = 3? Figure 2-5g
Q9. Is f continuous at x = 4?
Q10. Is f continuous at x = S?
For Probl ems 1-4, sk etch the graph of a function that has th e given features.
1. lim x-2- f (x ) = oo and lim x-2+ f (x ) = oo
2. lim x-2- f (x ) = oo and lim x-2+ f (x) = -oo
3. lim x-oof (x ) = -5 and lim x--oo f (x ) = 7
4 . lim x-co f (x ) = 00 and lim x- - oof (x ) = 00
1
S. Let f (x ) = 2 +- - .
x-3
a. Sketch the gr aph of f
b . Find lim f (x ), lim f (x ), lim f (x ), lim f (x ) , and lim f (x ).
x - 3+ x-3- x-3 x-oo x - - oo
c. Find a number D > 0 such that f (x ) > 100 if x is within D units of 3 on the
positive sid e.
d. Find a number D > 0 such that f (x ) is vvithin 0.001 unit of the limit as x - oo
when ever x > D.
6. Let g (x ) = sec x .
a. Sketch th e graph of g.
b . Explain why lim x-rr/2g (x ) is infinit e.
c. Find a numb er D > 0 such that g( x ) > 1000 if x is within D units of rr/2 on the
negativ e side.
d. What does lim x-rrri~g (x) equal ? Using the valu e of D from 6c, what can you say
about the value of g (x ) if x is kept within D units of rr /2 on the positive side?
sinx
7. Let r (x) = 2 + -- .
X
a. Plot the graph of r. Use a friendl y window with an x-range of about - 20 to 20 for
which x = 0 is a grid point. Sketch th e r esult.
b. Find th e limit of r (x ) as x approaches infinit y.
64 Chapter ofLimits
2: Properties
c. You realize that sin x ranges between x = - l and x = l. What, then, could you
pick for D so that r(x) would be within E = 0.001 unit of the limit found in 7b
whenever xis greater than D?
d. If in 7b you draw a horizontal line at y = Jim, will it be an asymptote 7 Explain.
e. Make a conjecture about the limit of r (x) as x approaches zero . Give evidence to
support your conjecture.
1 '
8. Leth(x) = (1 + x) '
a. Plot the graph of h. Use a friendly window with an x-rang e of Oto about 100. You
will have to explore to find a suitable y-range . Sketch the result.
b. As x becomes large, 1 / x approaches zero, so h (x) takes on the form 1 You realize
00
that 1 to any power is 1, but the base is always greater than 1, and a number
greater than 1 raised to a large positive power becomes infinite .
Which phenomenon "wins" as x approaches infinity,
Y _ Is there a finit e limit ? _
1, infinity, or some "compromise" number in between7
9. Figure 2-5h shows the graph of X
y = logx.
Does the graph level off and approach a (finite) limit as x
approaches infinity, or is the limit infinite ? Justify your answer.
You might find that the definition of logarithm is helpful Figure 2-5h
(y = log x if and only if lQ Y = x).
10. Wanda Wye wonders why the form 1/ 0 is infinite and why th e form 1/oo is zero.
Explain to her what happens to the size of fractions such as 1/0 .1, 1/0 .0001, etc.,
as the denominator gets close to zero. Explain what happens as the denominator
becomes very large.
11. Limits Applied to Integrals Problem: Rhoda Huffy starts riding down the driveway on
her tricycle. Being quit e precocious, she figures her velocity, v, in feet per second is
V = .jt ft/sec
where tis time in seconds since she started. Figure 2-5i shows v 3
as a fw1etion of t . 2
a. Explain why the definite integral from t = 0 to t = 9 repr e-
sents the distance Rhoda rode in the first 9 sec.
1 2 3 ~ 5 6 7 8 9
b. Estimate by trapezoids the integral in l la, using 9, 45, 90,
and 450 trapezoids . Record all the decimal places your Figure 2-5i
program gives you.
c. What number (an integer in this case) do you think is the exact value of the
integral? Explain why this number is a limit . Why are the approximate answers by
trapezoids all smaller than this number ?
d. Figure out how many trapezoids are needed so that the approximation of the
integral is within 0.01 unit of the limit. Tell how you go about getting the answer.
Section 2-5: Limits InvolvingInfinity 65
,-,
' ,-
~ c
12. Work Problem: The work done as you drag a box across the floor is equal to the
product of the force you exert on the box and the distance the box moves . Suppose
that the force varies, and is equal to
F(x) = 10 - 3.Jx, Force, lbs
10
where F(x) is force in pounds and xis the number of feet the box
is from its starting point. Figure 2-5j shows the graph of F .
a. Explain why a definite integral is used to calculate the
amount of work done.
4
b. Use the trapezoidal rule with n = 10 and n = 100 increments
to estimate the value of the integral. What are the units of Figure 2-5j
work in this problem?
c. The exact amount of work is the limit of the trapezoidal sums as n approaches
infinity . In this case the answer is an integer. What do you suppose that the
integer is?
d. What is the minimum number, D, such that the trapezoidal sums are closer than
0.01 unit to the limit in 12c whenever n > D?
13. Searchlight Problem : A searchlight shines on a wall as shown in Figure 2-Sk. The
perpendicular distance from the light to the wall is 100 ft. Write an equation for
the length of the beam of light as a function of the angle (in radians) between the
perpendicular and the beam. How close to rr/2 must the angle be in order for the
length of the beam to be at least 1000 ft, assuming that the wall is long enough?
Length
of beam
:: Infinit
t e
: limit
Ang le, x
I 7r
Searchlight I -
I 2
Figure 2-5k
14 . Zero Times Infinity Problem: You have learned that 0/ 0 is called an indeterminate
form . You can't determine what it equals just by looking at it. Similarly, 0 oo is
an indeterminate form. In this problem you will see three possibilities for the limit
of a function whose form goes to O oo. Let f, g, and h be functions defined as
follows.
1 l 1
f(x ) = Sx(x - 2) -- g(x) = 5x(x - 2) (x _ )2 h(x) = Sx(x - 2) 2 - -
x- 2 2 x - 2
a. Show that each of the three functions takes the form O oo as x approaches 2.
b. Find the limit of each function as x approaches 2.
c. Describe three things that the indeterminate form O oo could approach.
66 Chapter
2: Propert
iesof Limits
2-6 The Intermediate Value Theorem
and Its Consequences
Suppose you try to find a solution of the equation x 3 = 6 by tracing the graph of
y
y = x 3 (Figure 2-6a). The cursor never quite hits a value of x that makes y equ al
exactly 6. That's because the grap her plots discrete poi nts that represent only
approximate ly the continuou s graph. However, because y = x 3 is continuo u s, there
X
real ly is a value of x (an irrational number) which, when cubed, gives exact ly 6.
y= 5.929741
y= 6.028568 The property of continuous functions that guarantees there is an exact valu e is
called the interm ediate v alue theor em. Informally, it says that if you pick a valu e
Figure 2-6a of y betwee n any two values of f(x), there is an x-value in the domain that gives
exact ly that y-value for f(x) . Because 1.81 3 = 5.929741 and 1.82 3 = 6.028568, and
because 6 is between 5.929741 and 6.028568, there must be a number x between
1.81 and 1.82 for which f (x) = 6 exactly . One must show that y = x 3 is contin uou s
in order for the property to app ly.
Property:TheIntermediateValueTheorem
If function f is continuous for all x in th e closed int erval [a, b], and y is a numb er betvveen
f (a ) and f (b ), then there is a numb er x = c in (a, b ) for which f (c) = y .
Figure 2-6b illustrates the meaning of the theorem. Picky between f(a) an d f(b)_ If
Pick{~aJ
~(:
~-b, f is continuo us, you can go over to the graph, then go down to th e x-axis and find c,
number , a corresponding value of x. The value of f(c) will thus equal y exactly. The proof of
in here. : this theorem relies on the complet eness axiom. This axiom, which comes in several
y - - - - ' ' - -- - ' '
f(b ) - - - - H -- - -f --s forms, says that there is a real number corresponding to every point on the number
: ' i ': X line, and vice versa . Thus, the set of real numbers is "comp lete." It has no "holes,"
a C b as does the set of rational numbers .
Get at least one
x-value in here.
A formal proof of the intermediate value theorem usually appears in later courses on
Figure 2-6b ana lysis of rea l numbers. The gist of the proof is that for any y-value you pick in the
interval, there will be a point on the graph because the grap h is cont inuo us. Going
vertically to the x-axis gives a point on the number line . This point corres ponds to a
rea l number x = c, because the set of real numbers is comp lete. Reversing the steps
shows that f(c) really does equal y .
OBJECTIVE Given an equation for a continuous function f and a value of y between f(a ) and f (b ) ,
find a value of x = c between a and b for which f( c) = y.
In addition, you will be exposed to a corollary of the interme diat e value theorem ,
called the image theorem , which re lies for its proof on the extreme v alue theorem.
Example 1 If f(x) = x 3 - 4x 2 + 2x + 7, find, approximate ly, a value of x = c between 1 and 3 for
which f(c) = 5. Prove that there is a value of c for which f(c) is exactly 5.
ion2-6: The Intermediate ValueTheoremandIts Conseque
Sect nces 67
-,-
tA 1-
~ c
Solution First plot the graph (Figure 2-Gc).Use a friendly window. A line at y = 5 intersects
the graph at three points. The one between 1 and 3 is approximately 1.3.
To show that f(c) can be exactly 5, first observe that f is a polynomial function.
Thus, f is continuous. (See Problem 69 in Problem Set 2-4.) Therefore, one
hypothesis of the intermediate value theorem is true for x in [1, 3].
By tracing or by direct substitution, f(l) = 6 and f(3) = 4.
t 3 The number 5 is between 4 and 6, so the other hypothesis of the intermediate
X = 1.3 y = 5.037
value theorem is true in the interval [1, 3].
Figure 2-6c :. there is a number x = c between l and 3 for which f(c) = 5 exactly, Q.E.D.
If desired, you can find a better approximation of c by zooming in on your grapher
or by using the solve feature of your calculator. The answer is c = 1.31110781 ....
Problem Set 2-6
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Evaluate f(2) if f(x) = 3x 4 + 5.
Q2. Find limx-2 f(x) if f(x) = 3x 4 + 5.
Q3. Evaluate h(3) if h(x) = S(x - 3) / (x - 3).
Q4. Find limx-J h(x) if h(x) = S(x - 3) / (x - 3).
QS. Evaluate s(O) if s(x) = lx l/x.
Q6. Find limx-o s(x) if s(x) = lx l/x.
Ql. Evaluate: sin (rr/2)
QB. Evaluate: I13 - 71
Q9. Fill in the blank with the correct operation: log (xy) = logx -?- logy.
Q10. Solve: - 3x < 12
For Problems 1 and 2, find, approximately, a value of x = c in the given interval for which
f(c) equals the given y-value. Tell why each function is continuous, then prove that there
is a value of x = c for which f(c) is exactly equal to the given y-value. Illustrate by graph.
l. f(x) = (x - 3) 4 + 2, [1,4), y =8 2. f(x) = 0.00lx 5 - 8, [O, 6), y = - 1
3. Converse of the Intermediate Value Theorem? The intermediate value theorem is not
an "if and only if" theorem. The conclusion can be true even if the hypotheses are
not met.
a. The left figure of Figure 2-6d shows
Ix - 21
f(x) = 2 +x + -- - .
x- 2
Explain why the conclusion of the intermediate value theorem could be true or false
for the interval [l, SJ, depending on the value of y you pick in [2, 8].
68 Chapter
2: Properties
of Limits
b. The right figure of Figure 2-6d shows the graph of
g(x) = 2 + x - Ix - 221.
x-
Explain why the conclusion of the intermediate value theorem is always true for
the interval [1, 5], no matter what value of y you pick between g(l) and g(5), even
though the function is discontinuous at x = 2.
g(x )
f (x) ------ /
8 ; :
2 --
X X
1 2 5 1 2 5
Figure 2-6d
f(x) .
4. Figure 2-6e shows the graph of the following . .!:! ..... .... ......... /...... ... .
2x, if x is rational
f(x) = { 8, if x is irrational
a. Find f(2), f(3), f(0.5), and f( -/5 ).
b. Is f continuous at x = 3? Explain.
c. Where else is f continuous? Surprising 7 ! X
l 2 3 4
d. Because f(O) = 1 and f(2) = 4, is the conclusion of the
intermediate value theorem true for all values of y between Figure 2-6e
1 and 4? Explain .
5. Use the intermediate value theorem to prove that there is a real number equal to J3.
That is, prove that there is a number c such that c 2 = 3.
6. Use the intermediate value theorem to prove that if f is continuous, and if f(a) is
positive and f(b) is negative, then there is at least one zero of f(x) between x = a and
x = b . (Recall that a zero of a function is a value of x that makes f(x) = 0.)
7. The intermediate value theorem is an example of an exis tence theorem. Why do you
suppose this term is used? What does an existence theorem not tell you how to do?
8. Sweetheart Problem: You wish to visit your sweetheart, but you don't want to go all
the way over to his or her house if your sweetheart isn't home. What sort of "existence
proof" could you do beforehand to tell whether or not it is worthwhile to make the
trip? What sort of information will your proof not give you about making the trip? Why
do you suppose mathematicians are so interested in doing existence proofs before
they spend a lot of time searching for solutions?
Sec
tion 2-6:TheIntermediate
ValueTheorem
andIts Consequences 69
I L~-,-,
',-
c
9. Foot Race Pr oblem: Jesse and Kay run the 1000-m race . One minute after the race
begins, Jesse is running 20 km/ hr and Kay is running 15 km / hr . Three minutes after
the race begins, Jesse has slowed to 17 km/ hr and Kay has speeded up to 19 km/ hr.
Assume that each runner's speed is a continuous function of time. Prove that there
is a time between 1 min and 3 min after the race began at which each one is running
exact ly the same speed. Is it possible to tell what that speed is? Is it possible to tell
when that speed occurred? Explain.
10. Postage Stamp Prob lem: United States postage rates in 1996 for
Nwnb er of dollars
first-class letters was 32<tfor the first ounce and 23<tper ounce
thereafter. Figure 2-6f, the classical example of a step function, 2.00
shows the cost of mailing a first-class letter versus the number
of ounces it weighs . Does the function me et the hypotheses of
the intermediate value theorem? Is there a weight of letter that J.00
can be mailed for exactly $1.00? Justify your answ ers.
()-4
11. Cosine Function Problem: Figure 2-6g shows f(x) = cosx . From
trigonometry you recall that cos O = 1 and cos TT = - 1. As -,1
r 23456
Number of oz
7 89
shown in the figure, 0.6 is between - 1 and 1. Can you use the
intermediate value theorem to show that there really is a value Figure 2-6f
of x = c between O and TT for which cos c = 0.6? If so, tell how
you know and calculate it as accurately as possible. If not, tell
why not.
12. Exponentia l Funct ion Prob lem: Figure 2-6h shows f (x) = 2x.
Why does f(O) = 1? Because f(2 ) = 4 and 3 is between 1 and 4,
X
does it necessarily follow that there is a value of x = c between
0 and 2 for which 2c = 3? If so, tell how you know and find as
accurate a value of c as possible . If not, tell why not.
13. The Extreme Value Theorem : The extreme value theorem
expresses the property that if f is continuous on the closed
interval [a, b ], then there are numbers c 1 and c2 in [a, b] for Figure 2-6g
which f(c 1 ) and f(c 2 ) are the maximum and minimum values of
f(x ) for that interval . Think about what this means, then draw
a graph showing what you have thought. Draw a graph showing
why the conclusion might not be true for a function that has a
discontinuity somewhere in [a, b].
14. The Image Theorem : The image theorem is a corollary of the
intermediate value theorem, expressing the property that if f is
continuous on [a, b], then the image (the set of y-values) is all
real numbers between the minimum of f(x ) and the maximum
of f(x), inclusive. Use the extreme value theorem as a lemma (a
preliminary result) to prove the image theorem. X
7 2
Figure 2-6h
70 Chapter
2: Properties
of Limits
2-7 Chapter Review and Test
In this chapter you have gained further insight into the meaning of limit. You have
broadened the idea of limit to include right- and left-hand limits and infinite limits,
and you have applied these ideas to make precise the concept of a continuous
function . By now you should realize that a limit is a number f(x) stays close to when
x is kept close enough, but not equal, to c. You should be able both to find a limit
of a function given by graph or by equation and to show that f(x) really does stay
within E units of the limit when xis within 6 units of c(x * c). You should be able
to use the limit properties to prove that the number you've found is the limit. If
you think of "lim" as an operation that acts on functions, many of these properties
can be thought of as distributive properties. The following box summarizes the
properties .
LimitPropertySummary
Limit distributes over addition, subtraction, multiplication, and division
(denominator * 0) with any finite number of terms or factors.
Finally, you learned another major theorem of calculus, the intermediate value
theorem, which expresses a property of continuous functions.
The Review Problems below are numbered according to the sections of this chapter.
The Concepts Problems allow you to apply your knowledge to new situations. The
Chapter Test is more like a normal classroom test your instructor might give you.
Review Problems
RO. You have learned that calculus involves four concepts. Your goal for the course is to
be able to do four things with each of these concepts.
Define it. Understand it. Doit. Apply it.
Limit
Derivative
Integral
Integral
Make a table like that shown and put it into your journal. Check each box you have
worked on as you have studied this chapter. Make journal entries for such things as
those listed below.
The one most important thing you have learned in studying Chapter 2
A statement telling what you now understand a limit to be
How limits apply to derivatives and definite integrals
Your understanding of continuity and the intermediat e value theorem
Anything you need to ask about in class before your test on Chapter 2
Rl. a. Write the definition of limit.
b. Plot the graph of f(x ) = (x - 2) 115 + 3. Show that f(2) = 3. Make a table of values
of f(x) for each 0.0001 unit of x from 1.9995 through 2.0005. Based on the table,
find out how close to 2 you must keep x in order for f(x) to be within 0.2 unit of 3.
Section
2-7: Chapter
Review
andTest 71
wnur nma:zrn rmmmw r ..
c. Plot and sketch th e graph of g(x) = (x - 2) 21s + 3. Tell what feature this graph has
that the graph off in Rlb doesn't have. Is there a limit as x approaches 2?
R2. a. For f(x) = (x - 2) 115 + 3 from Problem Rl, calculate the maximum value that b can
be so that f(x) is within 0.2 unit of 3 when x is within b units of 2. Sketch on a
graph.
b. Show that ther e is a positive number b for any positive
value of E, no matter how small, such that f(x) is within E f(x)
units of 3 whenever xis within b units of 2. 5
c. For the function in Figure 2-7a, write the limit of f(x ) as x
approaches 2. Estimate from the graph the largest possible
3
value of b that can be used to keep f(x) 'vVithin0.--!unit of
the limit when x is within b units of 2. 2
d. In R2c, f(x) = 2 + Fx"=l . Calculate the maximum value of
b that you estimated in R2c. X
2 3 4 5
e. For the function in R2d, show that there is a positive value
of b for any E > 0 for which f(x) will be within E units of the Figure 2-7a
limit when xis within b units of 2.
R3. a. State the following limit properties.
i. Limit of a sum of two functions
ii. Limit of a constant times a function
iii. Limit of a quotient of two functions
b . For g (x ) = x 3 - 13x2
x _+ 32x - 6 , p 1ot t h e grap h , usmg
. a fnen
. dly wm
. d ow t h at
3
includes x = 3, then sketch. Use algebra and the appropriate limit properties to
find limx_ 3 g(x). Show that it agrees with your graph.
c. What part of the definition of limit allows you to cance l the (x - 3) factor in the
denominator in R3b without having to worry about dividing by zero?
(x+3)(x-2) . . . ..
d. Let m(x) = --- -- . Fmd llmx_6 m(x). Prove that you are nght by citmg the
x- 4
appropriate limit properties. Explain why you cannot use the limit of a quotient
property for limx- 4 m(x).
e. Chuck's Rock Prob lem: Chuck throws a rock high into the air. Its distance, d(t)
meters, above the ground is given by d (t) = 3 5t - 5t 2 , where t is the number of
seconds since he threw it. Find the average velocity of the rock from time 5 sec to
5.1 sec . Write an equation for the average velocity from 5 sec to t sec . By taking
the limit of the expression in this equation, find the instantaneous velocity of the
rock at t = 5. Was the rock going up or down at t = 5? How do you tell? What
mathematical quantity is this instantaneous velocity?
R4. a. Write the definition of continuity at a point.
b. Sketch graphs of the functions described.
i. Has a removable discontinuity at x = l.
ii. Has a step discontinuity at x = 2.
iii. Has a vertical asymptote at x = 3.
iv. Has a cusp at x = 4.
v. Is continuous at x = 5.
vi. Has a limit as x - 6, a value of f(6), but is discontinuous at x = 6.
vii. Has left-hand limit of -2 and right-hand limit of 5 as x approaches l.
72 Chapter
2: Properties
of Limits
x2, if X ='>2
c. Let f( x ) = { x 2 - 6x + k, if x > 2.
i. Sketch the graph off if k = 10.
ii. Show that f is discontinuous at x = 2 if k = 10.
iii. Find the valu e of k that makes f continuous at x = 2.
RS. a. Write the definition: lim x-oof(x) is infinite.
b. Sketch a single graph for whlch lim x_2 - f(x) = oo, lim x-z+ f(x) = - oo, and
limx-oof(x) = 3 .
c. Let f(x) = 6 - 2- x. Find the limit of f(x) as x approaches infinity. Sketch a graph
showing that f(x) can be kept within 0.1 unit of thls limit by keeping x large
enough. What values of x will ensure that f(x) is within 0.001 unit of the limit ?
d. The distance you travel at a variable velocity, v (t), is the definite integral of v (t)
with respect to t . Suppose that your car's velocity is given by v(t) = 40 + 6-/f ,
where tis in seconds and v(t) is in feet per second. Use the trapezoidal rule with
varying numbers, n, of increments to estimate the distance traveled from t = 0 to
t = 9. What limit do these sums seem to be approaching as n approaches infinity?
Find a number D for whlch the trapezoidal sum is withln 0.01 unit of thls limit
when n > D.
R6. a. State the intermediate value theorem. What axiom forms the basis for the proof
of the intermediate value theorem? State the extreme value theorem . What word
tells how the extreme valu e theorem is related to the intermediate value theorem?
b . For f(x) = -x 3 + Sx2 - lOx + 20, find f(3) and f(4) . Based on these two numbers,
how can you tell immediately that there is a zero of f(x) between x = 3 and x = 4?
What property of polynomial functions allows you to make thls conclusion? Find
as accurate a valu e of thls zero as possible.
2 28
c. Plot the function f(x) = x + l lx/ . Use a friendly window that includes x = - 4 .
x+
Show that f( - 6) = 1 and f( - 2) = S. Based on the intermediate value theorem, if
you pick a number y between 1 and 5, will you always get a value of x = c between
- 6 and - 2 for whlch f (c) = y? If so, tell why. If not, give a counterexample.
Concepts Problems
Cl. Squeeze Theorem Introduction Problem : Suppose that g(x) and h (x) both approach
7 as x approaches 4, but that g(x) < h(x) for all other values of x. Suppose another
function , f, has a graph that is bounded above by the graph of h, and bounded below
by the graph of g. That is, g(x) ='>f(x) ='>h(x) for all values of x. Sketch possible
graphs of the three functions on the same set of axes. Make a conjecture about the
limit of f(x) as x - 4.
C2. Der ivat ives and Continuity Prob lem: Figure 2-7b shows the graph of
2
X + 3, if X < 1
f(x) = { x2 - 6x + 9, if x ~ 1.
1
Find the valu e off( 1). Is f continuous at x = 1? Find the limit of f(x) - f( ) as x - 1-
x- l
Section2-7: Chapter
Review
andTest 73
k-, , ,-
c
and as x - 1+ . Based on your work, explain how a function can be continuous at a
point but not have a derivative there.
X X
-5 5
Figure 2-7b Figure 2-7c
C3. Equation from Graph Problem : Figure 2-7c is the graph of a discontinuous function .
Write a single equation whose graph could be that shown in the figure .
C4. Abso lute Value Definition of Limit: Later in your mathematical career, you may
encounter the definition of limit written in the form shown in the box .
Algebraic (AbsoluteValue}Definitionof Limit
L = limf(x) if and only if , for any
x- c
E > 0, there is a o > 0 such that if O < Ix - c l < o,
then lf(x) - LI < .
Explain how this algebraic definition of limit is equi valent to the "within" definition you
have learned.
Chapter Test
Tl. State the Limit of a Quotient Property.
For the functions graphed in Problems T2-T5, tell th e following for x = c.
a. Left- and right-hand limits if th ey exist
b. The limit if it exists
c. Whether or not the function is continuous
T2 . f(x) T3.
2 --- - ~
X X
:c IC
I
I
74 Chapter
2: Propertie
s of Limits
T4. f (x) T5. f{x)
T6. Sketch graphs showing that you understand the difference in behavior of the following func-
tions at x = 0.
c. h(x) = ~
1 X
a. f(x) = 2 b. g(x) = - d. s(x) =s in.!.
X X X X
4x 2 + l 7x + 15
T7. Let g (X ) =
x+ 3
a. Plot the graph, using a friendly window for which x = -3 is a grid point . Sketch.
b. Find the limit of g(x) as x approaches -3 . Show the limit on your graph.
c. Make a table of values of g(x ) for each 0.01 unit of x from x = -3. 05 to x = -2 .95.
d. Based on your table, for which values of xis g(x) within 0.1 unit of the limit?
e. What is the maximum value of 8 that could be used to insure that g (x) is within
0.1 unit of the limit whenever xis within 8 units of -3?
f. Show how to calculate 8 for any E > 0, no matter how small, for which keeping x
within 8 units of -3 will make g(x) stay within E units of the limit.
T8. Let h(x) = x j - lOx ' + 3x + 31.
a. Prove that h(2) is the limit of h(x) as x - 2 by citing the appropriate limit
properties.
b. Based on the definition of continuity, explain why his continuous at x = 2.
c. Show that h(3) has the opposite sign from h (2) . Use the result to explain why
h(x) has at least one zero between x = 2 and x = 3. Find this zero as accurately
as possible .
T9. Glacier Problem: To determine how far a glacier has traveled in a given amount of
time, naturalists drive a metal stake into the surface of the glacier. From a point not
on the glacier, they measur e the distance, d(t)
centimeters, the stake has moved in t days from
its original position. Every ten days they record
this distance, getting the values shown in the
table.
t days d(t) cm
0 0
10 6
20 14
30 24
40 36
50 50
Section
2-7: Chapter
Review
andTest 75
,-,
-
tJL ' ,-
c
a. Show that the equation d(t ) = O.Olt 2 + O.St fits all the data points in the table.
Use the most time-efficient way you can think of to do this problem .
b . Use the equation to find the average rat e the glacier is moving for 20 days to 20.1
days.
c. Write an equation for the average rate from 20 days to t days. Do the appropriate
algebra, then find the limit of the average rat e as t approaches 20. What is the
instantaneous rate the glacier is mo ving at t = 207 What mathematical nam e is
given to this rate?
d. Based on the table, does the glacier seem to be speeding up or slowing down as
time goes on? How do you reach this conclusion?
TIO. Calvin and Phoebe's Acceleration Problem: Calvin and Phoeb e are running side by side
along the jogging trail. At time t = 0, each one starts speeding up . Their velocities are
given by the following, where p(t) and c(t) are in fee t per second and tis in seconds.
c( t ) = 16 - 6(2 - 1 ) For Calvi n.
p (t) = 10 + Jf For Pho ebe.
Show that each is going the same speed when t = 0. What are the limits of th eir speeds
as t approaches infinit y? Surprising?!
kx 2 if X < 2
Tll. Let f (x) = { 10 ~ kx, if x ~ 2 _ What valu e of k makes f continuous at x = 2? What
feature will the graph of f have at this point ?
Tl 2. Let h (x) = x 3 Show that the numb er 7 is between h ( 1) and h ( 2). Since h is continuous
on [ 1, 2] what theor em allows you to conclude that ther e is a real numb er 1/7between
1 and 2?
76 Chapter
2: Properties
of Limits
CHAPTER
3 Derivatives, Antiderivatives,
and Indefinite Integrals
During the free fall part of a skydiver's descent, her downward
acceleration is influenced by gravity and by air resistance. The
velocity increases at first, then approaches a limit called the terminal
velocity. At any instant in time the acceleration is the derivative of
the velocity. The distance she has fallen is the antiderivative or
indefinite integral of the velocity. Velocity, acceleration, and
displacement all vary, and depend on time.
77
Mathematical Overview
In Chapter 3 you will apply the concept of limit to find formulas with
which you can calculate exact values of derivatives. Then you will go
backward to find the function equation if the derivative equation is
given. The antiderivative you get will give you a clue as to why there
are two concepts of calculus, both with the name "integral." You will
work with these concepts four ways.
Graphically The logo on each even-numbered
page of this chapter shows the run,
Lix, and the rise, Liy, from one point
on a graph to another. The limit of
Liy/ Lix is the instantaneous rate, or
derivative of f(x).
Numerically Lix Liy/ Lix X
1.3
0.1 0.705
0.01 0.6915
0.001 0.69015
0.0001 0.690015
3
. IIy
Al ge b ra,ca f'(l .3) = 1
x~
f(x)x -_ f(l.
1. ) , t h e d e fini. t10n
3
. o f d envat1ve.
. .
Verbally I had learned about limits, derivatives, and definite integrals already.
But it wasn't until I learned how to find an algebraic equation for a
derivative that I could understand the meaning of indefinite integral,
which is a name for antiderivative.
78
3-1 Graphical Interpretation of Derivative
You have learned the physical meaning of derivative: an instantaneous rate of
change. In this section you will be exposed to a graphical meaning of derivative .
OBJECTIVE Given the equation of a function, find the value of the derivative at a given point, and learn
the meaning of the derivative as it relates to the graph of the function.
The problems in this section will allow you to accomplish this objective as an
ass igm11ent after your test on Chapter 2, either on your own or with your stu dy
group.
Ex p loratory Problem Set 3- 1
Spaceship Problem: A spaceship approaches a far-off planet. At time x minutes after its
retrorockets fire, its distance , f(x) kilometers, from the surface of the planet is given by
f(x) = x 2 - Bx + 18 .
f(x)km
1. Figure 3-la shows the graph off. Confirm by grapher that this 10 1-1--'-'--'--+- '--f -'-"'-+
graph is correct .
2. Find the average rate of change of f'(x) with respect to x from
x = 5 to 5.1. What are the units of this rate of change?
3. The average rate of change, m(x), of f(x) from 5 to xis
_ f(x) - f(5)
mx( ) - x - 5. 5 10
By appropriate substitution , express m(x) in terms of x . Figure 3-1a
4. Find the limit of m(x) in Prob lem 3 as x approaches 5. This num ber is the derivative
off at x = 5. What physical quantity does this number represent? What are its units?
5. What form does m(5) take if you substitute 5 for x without having canceled? What
word describes this form? Why is m(5) undefined?
6. Photocopy the graph or plot it accurately on graph paper. On your graph, plot a line
with slope equal to the deriva tive you calculated in Problem 4, passing through the
point on the graph off where x = 5.
7. What words can you use to describe the relationship between the line in Problem 6
and the graph of function f?
8. Plot the graph off and the line in Problem 6. Use a friendly window for which x = 5
is a grid point, then zoom in repeatedly on the point (5, 3). Describe the relationship
between the line and the graph of f as you zoom in.
9. Record in your journal what you learned as a result of doing this problem set .
Section
3-1: Graphica
l Interpretat
ionof Derivative 79
b4~:
C
3-2 Difference Quotients and
One Definition of Derivative
You have been finding derivatives by taking a change in x, dividing it into the
corresponding change in y, and taking the limit of the resulting fraction as the
change in x approaches zero. Figure 3-2a illustrates what you have been doing . The
y change in they-value is equa l to f(x ) - f (c) . The change in the x-value is x - c. Thus,
the derivative is approximately equa l to
f (x) - f(c)
x -c
This fraction is called a difference quotient .
f{c)
The derivative is the function you get by taking the limit of the difference quotient
!in
, Change
x-cxis
X
as the denominator approaches zero. If the function's name is f, then the symbol
f' (pronounced "f prime") is often used for the derivative function. The symbol
' shows that there is a relationship, yet a difference, between the original function
C X
and the function "derived" from it (hence the name derivative), which indicates its
Figure 3-2a rate of change. One way to write the definition of derivative is shown in the box. An
alternative way to state the definition is shown in Section 3-4.
Definitionof Derivative(derivativeat x = c form)
('{c ) = lim f(x ) - f(c )
x-c X - C
Meaning: The instantaneous rate of change of f (x) with respect to x at x = c.
OBJECTIVE Given the equation of a function and a value of x, use the definition of derivative to
calculate the value of the derivative at that point, and confirm your answer numerically
and graphically.
Example 1 shows how the definition of derivative can be us ed to accomplish this
objective.
Example 1 If f(x) = x 2 - 3x - 4, find f'(5), the value of the derivativ e if x = 5. Check by graphing
the difference quotient and finding the limit.
Solution f(5 ) = 52 - 3(5) - 4 = 6
. ~--- (x 2 -3 x - 4)-6
. ------ 2
. x-----
... f '( 5 ) = 11m f(x )- f( 5)
= 11m = 11m -3x-10
x-s 5
X - x-5 X - 5 x-5 X - 5
. -----
= 11m (x - 5)(x+2) = ]'(
!ill X + 2) Why ca n you ca ncel without
x- 5 X - 5 x-5
dividing by zero?
=5+2= 7 Use the limit of a sum and the limit of x prop erti es from Chapter 2.
f(x) - f(5 ) x2 - 3x - 10
As a check, plot y = ----- , the difference quotient
x- 5 x-5
(Figure 3-2b).
80 Chapter
3: Deri
vatives, Antigerivat
ives,andIndef
inite Integrals
Difference quoti/ X quotient
7 ---------- - -t.97 6.97
4.98 6.98
-t.99 6.99
X 5.00 (none)
X= 5 y= 5.01 7.01
5.02 7.02
Figure 3-2b 5.03 7.03
Use a friendly window for which x = 5 is a grid point. There is a removable
discontinuity at x = 5.
Using the trace or table feature gives the values shown for x on either side of 5.
You should be able to see from the pattern that the limit of the quotient is 7 as
x approaches 5.
Another way to check the answer in Example 1 is to graph the original function,
then draw a line with a slope of 7 (the derivative) through the point on the graph at
x = 5 (that is, (5, 6)). To plot the line on the grapher, find its equation.
y = 7x + b => 6 = 7(5) + b => - 29 = b
. y = 7x - 29
Figure 3-2c shows that the line will be tangent' to the graph off.
You've already discovered this property from Problem Set 3-1. The fact that the
Figure 3-2c line is tangent to the graph is a geometrical interpretation of the meaning of the
derivative.
GeometricalInterpretational Derivative:Slopeal a TangentLine
The derivative of a function at a point equals the slope of the tangent line to the graph of
the function at that point. Both equal the instantaneo us rate of change .
Problem Set 3-2
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
QI. Quick! What does derivativ e mean ?
Q2. Simplify: (x 2 - 81 ) / (x - 9)
03. Find lim x_9 (x 2 - 81) / (x - 9).
04. Sketch the graph of y = 2x.
QS. Do the squaring: (3x - 7) 2
06. Fill in the blank with the appropriate operation: log(x / y ) = logx -?- logy.
Sec
tion3-2: Difference
Quotients
andOneDefinition
of Derivative 81
t /4v
I~
'-
Ql. Sketch a graph with a step discontinuity at x = 3.
QB. Sketch a graph with a cusp at the point (5, 2).
Q9. Sketch: y = 3x + Ix - 21
Q10. Who invented calculus ?
l. Write the definition of derivative.
2. What are th e physical and the geometrical meanings of the derivative of a function?
For Problems 3 and 4, do the following.
a. Use the definition of derivative to calculate th e value of f '( c) exact ly.
b. Plot the difference quotient in a neighborhood of c and sketch the result.
c. Plot the graph of the function in a neighborhood of c.
d. Draw a line through (c,f(c) ) with slope f '(c). Sketch the graph and the line.
3. f(x) = 0.6X 2 , C =3 4. f(x ) = -0.2x 2, C =6
For Problems 5-12, use the definition of derivative to calculate f'(c) exactly.
5. f(x) = x 2 + 5x + 1, c = - 2 6. f(x) = x 2 + 6x - 2, c = -4
7. f(x ) = x 3 - 4x 2 + x + 8, c = 1 8. f (x ) = x 3 - x2 - 4x + 6, c = - 1
9. f(x ) = -0.7x + 2, c = 3 10. f (x ) = l.3x - 3, c = 4
11. f(x) = 5, C =-1 12. f(x) = - 2, C =3
13. From the results of Problems 9 and 10, what can you conclude about the derivative of
a linear function? How does this conclusion relate to derivatives and tangent lines?
14. From the results of Problems 11 and 12, what can you conclude about the derivative
of a constant function ? How does this conclusion relate to derivatives and tangent
lines?
15. Local Linearity Problem: Figure 3-2d shows the graph of f(x) = x 2 , along with a line
of slop e f '( 1) passing through the point on the graph of f where x = l.
a. Reproduce this graph on your grapher . Use a friendly
window that includes the point (1, 1). Tell how you plotted
the tangent line.
b. Zoom in on the point (1, 1). What do you notice about the
line and the curve?
c. Zoom in several more times. How do the line and the curve X
seem to be related now? /
/
d. The graph off possesses a property called local linearity at
x = l. Why do you suppose these words are used to describe Figure 3-2d
this property?
e. Explain why you could say that the value of the derivative
at a point equal s the "slope of the graph" at that point if the
graph has local linearity.
82
I
Chapter
3: Derivative
s, Antid
erivatives,
andIndefiniteIntegrals
16. Local Nonlinearity Problem: Figure 3-2e shows the graph of
2
f(x ) = x 2 + 0.1 ( ~ x - l)
a. Show that the point (1, 1) is on the graph off. Does the
graph seem to possess local linearity at that point? (See X
Problem 15.)
b. Zoom in on the point (1, 1) several times . Sketch what you
see. Figure 3-2e
c. Explain why the graph of f does not have local linearity at
X = l.
d. Explain why f does not have a value for the derivative at x = l.
17. Let
X2 - X - 6
f(x) = X - 3 ' if X * 3
{ 7, if X = 3
a. Plot the graph on your grapher, using a friendly window for which x = 3 is a grid
point. Sketch the result, showing clearly what happens at x = 3.
b . Write the difference quotient for f'(3) and plot it on the grapher. (See if you can
find a time-efficient way to enter the equation!) Sketch the result.
c. Make a short table of values of the difference quotient for values of x close to 3
on both sides of 3. Based on your work, tell why the function has no derivative at
X = 3.
18. Let s(x) = 2 + lsin(x - 1) 1.
a. Plot the graph of s. Sketch the result.
b. Plot the difference quotient for s'(l). Sketch the graph.
c. Explain whys does not have a value for the derivative at x = l.
19. Tangent Lines as Limits of Secant Lines: Figure 3-2f shows the graph of
f(x) = 0.25x 2 - 2.5x + 7.25 .
f(x )
a. Show that the tangent line on the diagram at x = 3 has
slope equal to f'(3).
b . On a photocopy of Figure 3-2f, draw secant lines
starting at the point (3, 2) and going through the
points on the graph where x = 9, 8, 7, 6, 5, and
4. Tell what happens to the secant lines as the
x-distance between those points and the point
X
(3, 2) decreases.
3
c. Does the same thing happen with the secant lines
from the point (3, 2) to the points on the graph where Figure 3-2f
x = O,l, and 27
Section
3-2: Difference
Quotients
andOneDefinition
of Derivative 83
~y
+-#4
C
d. Figure 3-2g shows the graph of g (x)
4
g(x ) = 4 - 6 lcos %xi.
On a photocopy of Figure 3-2g, draw secant lines from the
cusp at the point (3, 4) through the points on the graph
where x = 8, 7, 6, 5, and 4 and where x = 0, 1, and 2.
e. The slopes of the lines in 19d approach a different limit as x
approaches 3 from the left than they do when x approaches
3 from the right. Explain why g has no derivative at x = 3.
Figure 3-29
f. Make a conjecture about what numbers the two limits in 19e equal.
20. Based on your work in Problem 19, explain why the following property is true .
Property:TangentLineas a Limitof SecantLines
The tangent to a graph at (c,f(c)) is the limit of the secant lines from (c,f(c)) to (x,f(x))
as x approaches c. The slope of the tangent line equals f'(c).
3-3 Derivative Functions,
Numerically and Graphically
You have been calculating the derivative of a given function at one particular point,
x = c. Now it is time to turn your attention to finding the derivative for all values
of x. That is, you seek a new function whose values are the derivatives of the given
function. In this section you will use th e numerical derivative feature of your grapher
to do this. In the next sections you will find formulas with which you can simply
write down the derivative function for various types of given functions.
OBJECTIVE Given the equation for a function, graph the function and its (numerical) derivative
fw1etion on the same set of axes, and make conjectures about the relationship between
the derivative function and the original function.
Graphers calculate numerical derivatives by using difference quotients, just as you
have been doing. Often they use a symmetric difference quotient . As illustrated
in the third figure in Figure 3-3a, points on the graph h (meaning "horizontal"
distance) units from x are found . The corresponding difference in y-values, t.y
(pronounced "delta y," meaning difference between y-values), is divided by the
difference in x-values, 2h, to get an estimate of the rate of change of the function .
You have used a forward difference quotient or a backwards difference quoti ent,
as shown in the first and second figures of Figure 3-3a. A symmetric difference
quotient usually gives a more accurate answer.
84 Chapter
3: Derivatives,
Antiderivatives,
andIndefinite
Integrals
f(x) /(x) f(x)
f(x + h) f (x + h)
f (x) f(x) / (x) - - - - - - - - - - -
f(x-h)
'
' X
X X+ h X- h X x- h x x+h
Forward differ en ce qu o tient Backward difference quotient Symmet ri c diff erence quotient
tly / h tly / h tly / (2h)
Figure 3-3a
To see how to calculate forward, backw ard, and symm etric diff erence quoti ents,
suppose you want to estimat e the der ivative of f(x) = x 3 at x = 2 (abbreviated f '(2)
and pronounced "f-prime of 2"). The x-increment, h, is oft en called the tolerance. If
a toleranc e of 0.01 is specified, for exampl e, th e calculation s are as shown .
F w d f (2.01 ) - f (2) = 2.01 3 - 23 = 8.120601 - 8 = 12
or ar 2.01 - 2 0.01 0.01 060 1
Bac kwar ct f (l.1.99
99) - f(2) = 1.99 3 - 23 = 7.880599 - 8 = 11 9401
- 2 - 0.01 - 0.01
S . . {(2.01) - f(l.99) _ 2.0l3 - 1.99 3 _ 8.120601 - 7.880599 _ 12 0001
ymmetnc. 2.01 - 1.99 - 0.02 - 0.02 -
As you might gu ess from the answers, the exact value of the derivative is 12. The
symmetri c diff ere nce quotient really does give a mor e accurate answer.
The num erical deri vative instruction on a typical grapher is shown here.
nDeriv (x3 , x, 2)
It says to tak e the numeri cal derivative of x3 with respect to x, evaluat ed where
x = 2. You will need to check your grap her 's manual to find out exactly how to write
th e instru ction and how to specify th e desired tolerance.
Because ther e will be a value of the derivative for each value of x, th ere is a function
whose indep endent variable is x and whos e dependent variable is the value of the
derivative . This function is called (obviously!) the derivative function and is usuall y
abbreviated f'(x). The grapher plot s the (numerical) derivative function by using an
instruction such as
Y2 = nDeriv (y 1 , x, x).
The instruction tells th e grapher to find the num erical deriva tive of y 1 with respect
to x (the first x in the par enthes es) and evaluate it at whatever th e value of x happens
to be (the second x in the parenth eses) . Example 1 shows what you can learn.
Example 1 Find f (x) = x 3 - 5x2 - 8x + 70.
a. Plot the graphs off and f ' (the num erical derivative off) on the same scree n.
Sec
tion3-3:DerivativeFunctions,
Numerically
andGraphically 85
b. Function f is a cubi c fun ction . What type of function does f' appear to be ?
c. Trace to find values of f (3) and f '(3) . Describe how f'(3) relat es to th e graph of
fat X = 3.
d. For what values of x does f'(x) = O? What feature does the f graph have at th ese
x-values?
e. Plot g(x) = f (x) + 10. On the sam e screen, plot g ' . How is the gra ph of g relat ed
to the graph off? How is the graph of g ' r elated to th e graph off 7
Solution a. Figure 3-3b shows the graphs off and f' (dotted) on the same screen . Type
in instructions such as thos e shown (dep endin g on your grapher) .
y1 = x3 - 5x 2 - Bx + 70
Y2 = nDeriv (y 1 , x, x )
b. The f ' graph looks lik e a parabola. Conjectur e: f' is quadratic.
c. f (3) = 28 and the value of f '(3) = - 11. The negat ive value of th e derivative
says that f (x ) is decr easing wh en x = 3, which the graph shows.
d. f '(x) = 0 for x = 4 and for x ""' -0 . 7. Th e gr aph off appears to have a high
point or a low point when f' (x) = 0.
e. Figur e 3-3c shows y 3 = y 1 + 10 along with the f and f' graphs. The graphs of
f and g ar e congruent to each other, separat ed vertically by ten units space .
The gr aphs of f' and g ' are identical. This is to be exp ected b ecause f and g
ar e changing at the same rate.
' ', r
' X
......__......
3/
Figure 3-3b Figure 3-3c
Example 2 Explore on the grapher th e cubi c fun ction f(x) = - x 3 + 3x 2 + 9x + 20 and its numer-
ical deriv ative, f'. Make some conclusions about how certain features on the graph
off', such as high and low points and x-int ercepts, ar e r elated to features on the
original function's graph. Write in your journal the conclusions you reach.
Solution In this problem you have considerable freedom to explore, to conjecture, to
discuss, and to write. Probl ems such as this are best don e with your study
group so that you may share id eas. Here is a typical (correct!) response to such
a problem as it might appear in your journal.
86 Chapter
3: Derivatives,
Antiderivatives
, andIndefiniteIntegrals
I plotted, the, g,vcq>rw off cuui-f' /!'UL,ff<)'t~ Uke, t:h.w(f L.iJ'W'e,,3 -3 cl,).
The,~ cuui-the, Low po{,y\,Q,of the, cubic, g,vaph,come, wh<we,,the, deYwat'we,,
g,vaph,cv 0'0;\,'the, )&-~ I w,ppow,t;h,w w v~ Wh<we,,the, deYwat'we,,
g,vaph,CV0'0;\,' the, !U-/M< the, deYwat'we,, w po, ~ the, vc;i;te,, of~
of the, ftMl.ctto-vvg,vaph,wpo-. 1riw wou.l.d, he,, tvue, c;i;tev ~ or Lowpoi,n,t. <9YU1/
of the, g,voup ~.I' the, deYwat'we,, g,vaph,wpo-:ru"w~ the,
,l,CIM/ thcu: whe.Ye>ve.+'
fu.,n..ct'wvv g,vaph,w ~ u;p.A Yl.Oth.et-- one, ,1,C1M1thcu: the,~ poi,vit' o-vvthe,
deYwat'we,, g,vaph,w c;i;t!U =1, wh<we,,the, fu.,n..ct'wvv g,vaph,~ U;y :.t'eep~ u;p~&
Figure 3-3d I :,t"Jl, d.o-n:t'I.AAIUie.v:,t'cuui-why the, deYwat'we,, g,vaph,w ~ u;pc;i;t!U =-2,
for ~C\,Yl,C', but' the, fu.,n..ct'wvv g,vaph,w clecv~
Problem Set 3.3
DoTheseQuickly
The following prob lems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
QI. Find lim x- o 3x / x .
Q2. Sketch a graph with a removable discontinuity at the point (- 2, 5).
Q3. Sketch the graph of y = tanx .
Q4. What percentage of 70 is 14?
QS. Multiply: (3x + 4 ) (x - 2)
Q6. Do the squaring: (5x - 7) 2
Ql. Write as the log of a single argument: log 36 :_ log 12 + log 2.
. Factor: x 3
QB - 8
Q9. Draw a tangent lin e to a circle.
Q10. Draw a secant line to a circle.
l. Cubic Function Prob lem I: Let f(x ) = - x 3 + 12x + 25.
a. Plot the graphs off and f ' (the numerical derivative off) on the sam e screen.
b. For what values of x is f'(x) positive? What is th e graph off doing for these values
of x ?
c. For what values of x is f(x) decreasing? What is true about f' (x ) for these values
of x ?
d. What does the graph off do at values of x wher e the f' graph crosses the x-axis?
e. Sketch the graphs off and f' , showing clearly the relationships you have stated in
la through ld.
f. Make a conjecture about what type of fun ction f' is.
2. Cubi c Func tion Problem II: Explor e the cubi c function g(x) = x 3 - 2x 2 + 2x - 15 and its
numerical derivative, g ' . Does th e graph ha ve a high point and a low point such as is
typical for cubic functions ? How does the graph of the derivative function rev eal this
behavior? When your study group has com e to a consensus, write the conclusions in
your journal.
Section
3-3: DerivativeFunctions,
Numerically
andGraph
ically 87
MWawtmtwf,aeee,emetttm
et ! AZJ
{y
~C
3. Quartic Fu nct ion Problem I: Let h(x) = x ~ - 2x3 - 9x 2 + 20x + 80.
a. Plot the graphs of h and h' on the same screen.
b. What typ e of function has a graph that is the shape of the h' graph? Make a
conjecture about what type of function the derivative of a seventh-degree function
would b e.
c. What do the zeros of h'(x) equal? That is, what values of x make h'(x) = O?
d. What features does the graph of h have if h'(x) = O? Based on the meaning of
derivative, explain why this observation is reasonable .
e. Sketch the graphs of h and h' , consistent with your answers above .
4. Quartic Function Problem II: Explore the function q(x) = - x 4 + 8x 3 - 24x 2 + 32x - 25
and its numerical derivative, q'. Does the function have the shape you expect of a
fourth-degree function (that is, three high or low points) ? How does the derivative
graph compare with the derivative graph for the quartic function in Problem 3? When
your study group has come to a consensus, write the conclus ions in your journal.
5. Sinusoid Problem I: Let f(x) = 4 + sinx.
a. Plot the graphs of f and f ' on the same screen. Be sure your calculator is set in
radian mode.
b . The graph off is called a sinusoid. You recall that the amplitude of a sinusoid is
the distance from the middle to a high point, and th e period is the distance along
the x-axis from one high point to the next. What are the amplitude and the period
of the f graph 7
c. The f ' graph is also a sinusoid. What are its amplitude and period?
d. Let g(x) = 3 + sinx. Plot the graphs of g and g ' on the same screen, along with the
graphs of f and f'. How are the f and g graphs related to each other? How do you
explain the relationship between the f ' and the g' graphs?
6. Sinusoid Problem II: Let f (x ) = 4 + sinx, as in Problem 5. Make a conjecture about which
one of the functions available on your grapher the numerical derivative turns out to
be. Give both graphical and numerical evidence to support your conjecture. Record th e
results in your journal. You will be using this information later in this chapter.
7. Exponent ial Function Problem: Figure 3-3e shows th e graphs of th e
exponential function f(x) = zx and its numerical derivativ e f'. They y
have similar shape. For instance, each has y-values that are four
times as large at x = 2 as they are at x = 0. By experimenting with
bases other than 2, see if you can find an exponential function f'_ - -
that is identical to its derivative. Write a paragraph in your journal, - - - - ------ X
telling what you have learned .
8. How the Grapher Works Problem: The text in this section shows that Figure 3-3e
by symmetric difference quotient, if f(x ) = x 3 , then f'(2) "" 12.0001.
Find f'(2) by using the numerical derivative feature on your grapher. Set the tolerance
for x at 0.01, as you did with the text. Based on the result , tell whether your calculator
seems to find num erica l derivatives by symmetric differ ence quotient or by some other
method .
88 Chapter
3: Derivatives,
Antiderivatives,
andIndefiniteIntegrals
9. Tolerance Problem (Epsilon and Delta): Figure 3-3f depicts a
squar e floor tile that is to be manufactured to a certain tolerance.
Suppose the ctimensions are 12" x 12" .
a. If the tolerances on the dimensions are 0.01 " , what range
of areas can the tile have? Within how many square inches of
the nominal area (144 in 2 ) is the tile?
b . If it is desired to manufacture tiles that are within 0.02 sq in .
of the nominal area, to what tolerance must the dimensions
of the tile be kept?
c. How does your work in this problem relate to the epsilon and
delta definition of limit? Figure 3-31
10. Symmetric Difference Quotient Problem: In the box are formulas for forward, backward,
and symmetric difference quotients. Show algebraically that the symmetric difference
quotient is always the average of the forward and backward difference quotients .
Formulasfor DifferenceQuotients
. . f (x + h ) - f (x )
Forward difference quotient = h
f (x) - f(x - h )
Backward difference quotient = h
Symmetric difference quotient = f (x + h ) -/ (x - h)
2
The tolerance , h, is positive in each case .
11. Difference Quot ient Accuracy Problem: Figure 3-3g shows the
graph of
f(x) = x 3 - x + 1.
a. By the method of Section 3-2, find the exact value of f'(l) .
b. Using a tolerance of h = 0.1, find the forward, backward, and X
symmetric differenc e quotients for f'( 1). Explain why the
symmetric difference quotient is so much closer to the actual
derivative than either the forward or backward difference Figure 3-3g
quotient .
c. Find the exact value of f' (O).
d. Using a tolerance of h = 0.1, find the forward, backward, and symmetric difference
quotients for f'(O). How do you explain the fact that all three are equal, and thus
that the symmetric clifference quotient is not more accurate than the other two?
e. Write a paragraph in your journal, telling what you have learned about the relative
accuracy of forward, backward, and symmetric clifference quotients .
Section
3-3:Derivative
Functions,Numerically ondGraphically 89
~
C
12. Numerica l Derivativ e Error Problem! Figure 3-3h shows the graph of
2
f (x ) = ( 1/x - 2) + x - 1.
y
Use various toleranc es, h, to explor e th e forward, backwar d, and
symmetric diff erence quotients at x = 2. Do all three values
seem to be approaching the same number as h approaches
zero? What do you conclude about f '(2)? Does the num erical X
derivative fun ction on your grapher give a value for f '(2)? 2
Write a paragraph in your journal, describing what you have
learn ed in this probl em about symmetric difference quoti ents Figure 3-3h
and numerical derivati ves .
13. Journa l Problem: Upd ate your journal with things you've learn ed since your last ent ry.
Include such things as thos e listed here.
The one most important thing you have learne d since the last journal entr y
What you now bett er und ers tand about th e derivative of a function
How the grapher calculates numerical derivatives
How the grapher could give a wrong answer for a numeri cal derivative
Any conjectures you have made concerning derivatives of various types of
functions
Anything you're still unsur e of concerning derivatives and want to ask about during
the next class period
3-4 Derivative of the Power Function and
Another Definition of Derivative
In Section 3-2 you learned a definition of derivative. Another form of this definition
lets you find an equation that will give th e values of the derivativ e without ha ving
to resort to the definition each time . In this sec tion you will find the derivative of a
power function, such as
f (X) = X 5 ,
or a linear combination of power functions, such as
g(x ) = 7x 415 - 11x - 2 + 13.
In each term the exponent is a constant. (Similar functions with variable exponents
are called exponential functions . You will deal with these in Chapter 6.) If each expo-
nent is a nonn egativ e int eger consta nt, th e function is called a polynomial function .
OBJECTIVE Given a power function, f (x) = xn, where n stands for a constant , or given a linear
combination of power functions, find an equation express ing f'(x) in terms of x .
90 Chapter
3: Deriva
tives,Antiderivat
ives, andIndefiniteIntegrals
~----------------------------------=1
The derivative of a function at a point x = c is
f'(c) = lim f (x) - f (c) .
x-c X - C
This form of th e definition is easy to rem emb er because it ties in with the slope
formula for a line you will recall from algebra, namely,
rise
slope = - .
run
It also calls attention to th e fact that you are finding the derivative at the one fixed
point where x = c (Figure 3-4a).
Another form of the definition leads more dire ctly to an equation for a derivative . As
shown in Figure 3-4b, c is replac ed by x and x is replaced by x + /::;.x.The change in x
thus takes the simpler form !::;.x(pronounced "delta x," meaning "difference between
y = f (x + /::;.x
x-values"). The rise, /::;.y,can be written /::;. ) - f(x). As x approaches c in
Figure 3-4a, !::;.xapproaches zero in Figure 3-4b. The definition of derivative reduces
to that shown in the box.
f(x) /
f(c)
)
Rise is
f(x)- f(c)
--------:
-------;
Y':
'
'
Find
derivat ive
here
: Run is
: x- c
X
l
f(x + t,x) /
: se is
f(x + t,x ) - f (x)
t,y
-- -- ----;
f(x) --------
:
:/).X
I
'
Y:
'
Find
derivative
her e
': Run is t,x
' X
Fixed_...,...c x ......_Variable Fixed _...,...x x+ t,x......_Variable
Figure 3-40 Figure 3-4b
Definitionol Derivative(!).xor h form}
f'(x) = lim 6y = lim f(x + 6x) - f(x) = lim f(x + h) - f(x)
t.x-0 6x t.x-0 6x h -0 h
Note that/::;. is the uppercase form of the Greek letter delta, while r5,which you have
been using in notation for limits, is the lowercase form. Sometimes th e single letter
h (for horizontal) is used in place of !::;.xto mak e the algebra easier to write.
Example 1 Given f (x) = x 5 , us e the definition of derivativ e to find an equation for f '(x).
Solution f'(x) = lim (x + h) s - xs By th e definition of derivative.
h- 0 h
(x 5 + 5x 4 h + 10x 3 h 2 + 10x 2 h 3 + 5xh 4 + h 5 ) - x5
f'(x) = lim h Expand.
/1- 0
5x 4 h + 10x 3 h 2 + 10x2 h 3 + 5x h4 + h 5
f'(x) = lim h Com bine like terms.
h- 0
f'(x) = lim (5x 4 + 10x 3 h + 10x 2 h 2 + 5xh 3 + h 4 ) Distribute 1/ h to each term.
h-0
f'(x) = 5x4 + 0 + 0 + 0 + 0 = 5x4 Use the limit properties .
Section3-4:Derivative
of the PowerFunctionandAnotherDefinitionof Derivative 91
t /1',,-.v
~C
You should see both the source of the answer and a pattern.
If f(x) = x 5 , then f'(x) = 5x4 .
The answer, 5x4 , is what remains of the second term in the binomial expansion
of (x + h ) 5 .
/ Source:Answer comes from second term.
,..,_,
x5 + 5x4 h + 10x 3 h 2 + 10x 2 h 3 + 5xh 4 + h 5
Patterns: \ \One less than th e original expon ent.
The original expon ent.
Generalizing these patterns gives a property for finding the derivative of any power
function.
Property:Derivativeof the PowerFunction
If f(x) = xn, then f'(x) = nxn- 1 . Restriction: The exponent n is a constant.
In Problem 28 of Problem Set 3-4 you will demonstrate that this property works for
powers with constant exponents but not for variable exponents. In Chapter 4 you
will prove that it works for negative and fractional exponents. In Chapter 6 you will
deal with variable exponents .
The process of finding an equation for the derivative of a function is called
differentiation. The word reflects the fact that t:.y / t:.x is a difference quotient. The
corresponding verb is differentiate. It may be easier to remember this prop erty by
what it allows you to do.
Procedure:Differentiatingthe PowerFunction
To differentiate the power function, f(x ) = xn, you multiply by the old exponent, n, then
reduce the exponent by 1 to get the new exponent.
To find a formula for differentiating a lin ear combination of powers, you must be
able to differentiate a sum of two functions, and a constant times a function. An
algebraic proof for a sum property is shown here. You will prove a prop erty about
the derivative of a constant times a function in Problem 35 of Problem Set 3-4.
Theorem:
If f(x) = g(x) + h (x), where g and h are differentiable functions, then f'(x) =
g '( x) + h '( x).
Algebraic Proof:
By the definition of deri vat ive,
f'(x) = lim f(x + t:.x) - f(x)
Definition of derivativ e.
L1x-O t:.x
f'(x) = lim [g (x + t:.x ) + h (x + t:.x)] - [g(x) + h (x)]
Substin1te for f(x ) .
11x-o t:.x
92 Chapter
3: Derivatives,
Antiderivatives,
andIndefinite
Integrals
f'(x) = lim [g(x + tlx ) - g (x)] + [h (x + tlx ) - h (x)]
Assoc iate the g's and
t,x-0 tlx the h's.
f'(x) = lim (g (x + 6x ) - g(x ) + h(x + 6x) - h(x)) Reas on?
t>x-0 6x 6x
'( ) . g (x + tl x) - g(x) . h (x + 6x ) - h (x )
f x = 11m ~----~- + 11m ------- Reason?
t>x-o tlx C>x-0 tlx
f'(x) = g '(x) + h'(x) Definition of derivative
(backward).
The derivati ve of a constant function is zero. Figure 3-4c shows the function
f(x) = 4.
As you can see, the function value is always 4. It doesn 't change, so its rate of
change is zero. Don't confuse the derivative of a constant function with the limit
Figure 3-4c of a constant function. The limit of f(x) as x approaches 6, for instance, is still 4,
not zero.
ThreePropertiesof Differentiation
Derivativeof a Sumof TwoFunctions: If f(x) = g(x) + h (x), where g and hare differentiable
functions of x, then f '(x) = g '( x) + h '(x).
Words: The derivative of a sum equa ls the sum of the derivatives.
Differentiation distributes over addition.
Derivative ofa Constant Times a Function:If f(x) = kg(x), where g is a differentiable function
of x, then f '(x) = kg '(x), provided k is a constant.
Words: The derivative of a constant times a function equals the constant times the
derivative of the function.
Derivative
of a Constant
Function:
If f(x) = C, where C stands for a constant, then f '( x ) = O
for all values of x.
Words: Constants don't change, so their rate of change is zero.
Combining these properti es with the derivativ e of a power function allows you to
differentiat e any linear combination of power functions in on e step!
Example 2 If f(x) = 5x 7 - llx 3 + 12x - 47, find f'(x ) .
Solution f' (x) = 5(7 x 6 ) - 11 (3x 2 ) + 12(1 ) = 35x 6 - 33x 2 + 12
The answer to Example 2 is called the algebraic derivative, which distinguishes it
from the numerical derivative found by difference quoti ents. The algebraic derivative
gives the exact values. The values by numerical derivative are only approximate .
Section
3-4: Derivative
of the PowerFunction
andAnother
Definition
of Derivative 93
I. j I
1~6: 1
c_ J
Example 3 Check the answer to Example 2 by graphing both the algebraic derivative and the
numerical derivative on the same screen.
Solution Enter: y1 = 5x 7 - llx 3
+ 12x - 47
6 2
Y2 = 35x - 33x + 12
y 3 = numerical derivative of Yi
Plot only y 2 and y 3 One graph appears on the screen. By using trace, you can
I
find small differences between y 2 and y 3 . Another option is to separate the
I
graphs by adding a constant to y 3 , as shown below.
X
y3 = nDeriv(y 1, x, x) + 5
1
Algebraic
Figure 3-4d shows the two graphs, using a window of [ - 1.2, 1.2] for x and
Figure 3-4d [- 10,30] for y. The check comes by observing that the two graphs have the
same shape.
There are other widely used symbols for the derivative, with which you should
become familiar. Each has advantages and disadvantages, depending on where it is
used.
dy/dx andy' Terminology
If y = f(x), then instead of writing f '(x) you can write any of the following.
y', pronounced "y prime" (a short form of f'(x))
!~, pronounced "dee y, dee x" (a single symbol, not a fraction)
:x (y), pronounced "dee, dee x,of y" (an operation done on y)
The symbol dy / dx comes from the difference quotient 6y / 6x. It means that the
limit is to be taken as both 6y and 6x go to zero. Mathematician Gottfried Leibniz
(1646-1716) started using this terminology in about 1675. Later in this book you
will learn that dy and dx are called differentials. For the time being, regard dy I dx as
a single symbol that cannot be taken apart - avoid saying "dy over dx."
The symbol !!.._comes from rearrangmg the letters in dy. It is an operator that acts
dx dx
on the expression y, similar to sin y or log y. It tells you to take the derivative of
y with respect to x. It is useful if you want to show the expression for the function
rather than just y. For example,
d
dx (xs) = 5x4.
Example 4 If y = 7x- 4 15 , write an equation for the derivative function, dy / dx. Assume the
power rule works for negative and rational exponents.
Solution Assume the power rule works for negative and fractional exponents.
:: = 7(- ~x - 915 ) = - 5.6x - 915 Subtra ct 1 from - 4/5 to get -9 /5 .
94 Chapter
3: Derivat
ives,Antiderivatives,
andIndefiniteIntegrals
Example 5 If y = 75 , find y' .
Solution y' =0 Becaus e 75 is a constant and th e derivativ e of a constant is zero .
It is possible to sketch the graph of a derivative function just by looking at the
graph of the function. For example, at a high or a low point in the function graph,
the derivative will be zero. If the graph is going up, the derivative will be positive,
and so forth . Example 6 shows how to sketch such a graph.
Example 6 For the function in Figure 3-4e, sketch a reasonable graph of the derivative function.
f (x)
Figure 3-4e
Solution Photocopy Figure 3-4e or sketch the graph on graph pap er. At x = 1 and at
x = 4, the function has leveled off. Thus, it is not changing, and its derivative is
zero. Mark these two points (Figure 3-4f).
((x)
\3 ,.....,. Between x = 1 and x = 4, the function is decreasi ng. Its gre atest downward
:\
~ ,. slope comes at about x = 2.5. Mark a point below the x-axis at x = 2.5.
Above x = 4, the graph slopes up at an increasing rate. Draw the derivative
graph as being positive and increasing.
Below x = 1, the graph is also sloping up but at a decreasing rate. Draw the
derivative graph as being positive but coming down toward y = 0 at x = l.
Figure 3-4f
Problem Set 3-4
DoTheseQuickly
The following problems are int ended to refresh your skills. You should be able to do all
ten problems in less than five minutes .
Ql. Do the squaring: (3x - 4) 2
Q2. Do the cubing : (a+ b ) 3
Q3. Write the definition of derivative (either form).
Q4. Write th e formula for a symmetric differenc e quotient.
QS. Find limx--3 (x - 5) / (x + 3).
Section
3-4:Derivative
of thePower
Function
andAnother
Definition
of Derivative 95
'hf:
C
Q6. Write as the log of a power: 3 log 7.
Ql. Write the exact valu e of tan (rr / 3) .
. Name the theorem whose conclusion is r 2 = s 2 + t 2 .
QB
Q9. Evaluate: 100 112
Q10. Sketch the graph of a function that is continuous at x = 4.
For Problems 1-18, find the equation of the derivative function .
l. f(x) = 5x4 2. y = llx 8
x-9
3. v = o.oo?t
-83 4. v(x) =
18
5. M(x) = 1215 6. f(x) = 4.77 23
7. y = 0.3x 2 - Bx + 4 8. r = 0.2x 2 + 6x - 1
d
9. dx(l3-x) 10. f(x) = 4.Sx2 - x
11 . y = x 23 + 5x- 2 - lOOx + 4 12. :x (x 2 15 - 4x 2 - 3x- 1 + 14)
13. v = (3x - 4) 2 (Do the squaring first.) 14. u = (5x - 7) 2 (Do the squa ring first.)
15. f(x) = (2x + 5) 3 (Be clever 1) 16. f(x) = (4x - 1) 3 (Be clever!)
x2 x3 x 2
17. P(x)= - x +4 18. Q(x) = +
3 -X+1 2
2
For Problems 19-22, use the definition of derivative in the f(x + h) form to find the
equation for the derivative function . In each case, show that the answer is consiste n t with
the answer you would get if you used the formula for the derivative of the power function.
19. f(x) = 7x 4 20. g(x) = 5x 3
21. v(t) = l0t 2
- 5t + 7 22. s(t ) = r' - 6t 2 + 3.7
23. Misconception Prob lem: Mae Danerror needs to find f '(3), where f(x) = x 4 . She
substitutes 3 for x, gets f(3) = 81, differentiates 81, and gets zero for the an swer.
Explain why she also gets zero for her grade.
24. Higher Math Prob lem: Chuck stands atop a cliff (Figure 3-4g). He
throws his math book into the air with an initial upward velocity of .....
20 m/ sec. As a result, its height above where he threw it, h(x)
meters, after x seconds is
h(x) = - 5x2 + 20x.
a. The upward velocity of th e book is the derivative of heig ht with
respect to time. Write an equation for the upward velocity.
b. How fast was the book going at time x = 3? Was it going
up or going down ? Explain .
\
c. At time x = 3, was the book above or below where
Chuck threw it? How far ?
d. At what time is the book at its highest point? How do you know? Figure 3-4g
96 Chapte
r 3: Derivatives,Antiderivatives,
and Indefinite Integral
s
For Problems 25 and 26, photocopy or sketch the graph of the function. On the same set
of axes, sketch a reasonable graph of the derivative function.
25. 26.
27. Numerical Versus Exact Derivative Problem: Figure 3-4h shows the graph of
f(x) = 0.4x 3 - 7x + 4,
along with two graphs of f'(x). One derivative graph is found
numerically as in Section 3-3. The other derivative graph is found
algebraically using this section's technique.
a. Plot these graphs on your grapher.
b. Tell which graph is f and which is f'.
c. Why do there appear to be only two graphs on the screen, not three? Figure 3-4h
d. Trace to x = 3. Write the y-value you get for each of the three graphs. How closely
does the numerical derivative fit the algebraic derivative at x = 3?
28. Power Formu la for Various Types of Exponents: The derivation of the power formula,
f(x) = x" implies f'(x) = nx" - 1 ,
was based on an example (Example 1 of this section) from which a pattern was
found. In Example 1, the binomial formula was used to expand (x + h) 5 as a sum.
Unfortunately, such sums will have an infinite number of terms unless the exponent
is a nonnegative integer. Your objective in this problem is to show by example that
the power formula for derivatives works if the exponent is a negative or fractional
constant but clearly does not work if the exponent is a variable . To accomplish the
objective, predict the derivative for each function given below, assuming that the
formula works for the type of exponent shown. Then show that the answer is correct
or incorrect by plotting three graphs on the same screen.
Yi = the given function
y 2 = your answer from using the derivative of a power formula
y3 = the numerical derivative of y 1
a. g(x) = x- 1 b. h(x) = x 112 c. e(x) = 2"
For Problems 29-32, tell quickly whether f (x) is increasing or decreasing at x = c, and
how fast.
29 . f(x) = x 112 + 2x - 13,c = 4 30. f(x) = x - 2 - 3x + 11, c = 1
31. f(x) = x1.s - 6x + 30, c = 9 32. f(x) = -3 .jx + X + l, C =2
Sect
ion3-4: Derivat
iveof thePower
Func
tionandAnother
Definition
of Derivative 97
For Problems 33 and 34, plot the graphs off and f ' on the same screen. Show that each
place where the f ' graph crosses the x-axis corresponds to a high or low point on the f
graph .
x3 x3
33. f(x) = - x2 - 3x + 5 34. f(x) = - 2x 2 + 3x + 9
3 3
35. Formula Proof Problem I: Use the definition of derivative to prove that if g is a
differentiable function of x, and f(x) = k g(x), then f '(x) = k g'(x).
36. Formula Proof Problem II: Given f(x) = x 5, derive the formula for f '(c) directly from
the x = c form of the definition of derivative,
f'(c) = lim f(x) - f(c).
x-c X - C
It will be helpful to remember how to factor a difference of like, odd powers.
4
a5 - b5 = (a - b)(a + a 3 b + a 2 b 2 + ab 3 + b~)
37. Derivative of a Power Formula: Show that the formula for the derivative of a power,
namely, f'(x) = nxn - 1 , comes from the second term in the binomial series expansion
of (x + 6.x)n.
38. Derivative of a Sum of n Functions Problem: Prove by mathematical induction that the
derivative of a sum of n functions is equal to the sum of the derivatives.
39. Introduction to Antiderivatives: Suppose you know the answer to a derivative problem,
and you want to know just what function has been differentiated . For example,
suppose that
f'(x) = 3x 2 - lOx + 5.
a. Figure out an equation for a function f for which f'(x ) is the function shown
above.
b . If f(x) is the answer you found in 39a, explain why g(x) = f(x) + 13 is also an
answer to 39a.
c. The functions f and g in 39a and b are called antiderivatives off'. Why do you
suppose they are called antiderivatives?
3-5 Displacement, Velocity, and Acceleration
You have found the velocity or acceleration of a moving object several times in this
course. Now that you know an algebraic method to find derivatives, you can get an
equation for the velocity or acceleration if you know an equation for the position of
the object.
98 Chapter
3: Derivatives,
Antiderivatives,
andIndefiniteIntegra
ls
OBJECTIVE Given an equation for the displacement of a moving object, find an equation for its velocity
and an equation for its acceleration, and use the equations to analyze the motion.
y ft Suppose that a football is punted into the air . As it rises and falls, its distance above
the ground (its displacement) is a function of the number of seconds since it was
punted. Suppose that by experime nt it is found that
2
y = -l6t + 37t + 3,
tsec
where y is its displacement in feet and t is the number of seconds since it was
punted. Figure 3-Sa shows the graph of this function.
Figure 3-5a The velocity of the ball tells how fast it is going and in which direction . Because
velocity is an instantaneous rate of change, it is a derivative. For the above equation,
velocity = !~ = - 32t + 37. Differenti ate y = - 16t 2 + 37t + 3.
To find the velocity at any particular time, all you need to do is substitute for t. Here
are some examp les .
t = l: !~ =- 32(1) + 37 = 5f t / sec
t = 2: !~ = - 32(2) + 37 = - 27 ft / sec
The dy / dt symbol for the derivative helps you remember the units of velocity .
Because y is in feet an d tis in seconds, dy / dt is in feet / second.
Speed is the absolute value of the velocity. Speed tells how fast an object is going,
without regard to its dir ection. When t = l, the ball is traveling 5 ft / sec. When t = 2,
it is traveling faster at 2 7 ft / sec. The negative velocity when t = 2 indicat es that
the displacement is decreasing, which means that the ball is coming back down .
Although the ball is traveling faster when t = 2, its velocity is less b ecaus e - 27 is
less than 5.
The velocity cha nged from t = l to t = 2. The instantaneous rate of change of
velocity is called acceleration. It is the derivative of the velocity. Using v for
velocity, v = - 32t + 37. Thus,
acceleration = !~ = -32. Differentiate v = -32 t + 37.
The dv / dt symbol for the derivativ e tells you the units of acceleration. Because vis
in feet / seco nd and tis in seconds, dv / dt is in (feet / second) / second. This quantity
is pronounced, "feet per secon d, per second," with a pause at the comma. It is often
pronounced "feet per second squared" and written "ft / sec 2 ."
The ne gat ive acceleration means that the velocity is decreasing. Be sure you know
how to interpret this fact . In this instance, v decr eased from 5 ft / sec when t = l to
-27 ft/sec when t = 2, even though the ball is traveling faster at t = 2. Note that the
acceleration is constant, -32 (ft / sec)/ sec, for an object acted on only by gravity.
Section
3-5: Displacement
, Velocity,
andAcceleration 99
t A,,,,v
~
C
The acceleration of a moving object is the derivative of a derivative. The words
second derivative are used for the derivative of a derivative. The symbol for the
second derivative of y with respect to t is
d 2y
dt 2
pronounced "dee squared y, dee tee squared." The symbol comes from performing
d I dt on dy I dt, then using "algebra" on the symbols .
2
d (dy) d y
dt dt becomes dt 2 .
If x = f(t), then the symbol f " (t), pronounced "f double prime oft," is used for
the second derivative. Sometimes y " is used if it is clear what the independent
variable is.
Example 1 An object moves in the x-direction in such a way that its displacement from the
y-axis is
X = 3t 3 - 30t 2 + 64t + 57, fort ~ 0,
where x is in miles and t is in hours. With your grapher in parametric mode, plot
the path of the object. Hold or repeatedl y press the trace key to find out how the
object moves as time goes on . Write a paragraph describing the motion. Include the
approximate times and places at which the object reverses direction, and the time
intervals during which the object is traveling to the right and to the left. A sketch
may help.
Solution Enter the equation above for x, then enter some convenient value for y, such
as l. Use a friendly range from Oto about 10 and an x window from Oto about
200. The graph is a straight line (Figure 3-Sb). It is interesting to watch the line
get generated 1 Your vvrite-up might look something like this.
The, obj~ ~CM"t}' o-ut" {A;(: )i./ = 5 7 wwwhe.vt = 0 . It fJ-0%vlfjh:t, .OWLA-'11f' cl.owVII
UNl,t""'vHw cwout" I. 3 "fw-,{A;(: whidv t'une, u
~op:,,wrieve, )i./ w cwout" 96 WU:,,Cv.YC,vv
u
f ifyuve, 3 - 5 Cl. -rhe'v tw,-Vl,.y Cl,VO"t,Ul\,Cl/
evvui-~(M'fy off~ to-the, Left, _qJ~
LA;pfor ev whae,,, the,vv .;low LA-'1%' whe.vt w cthout" 5. 3 "fw-,{A;(:
cl.owVII evvui-~oppLA-'11f'
whidvt'une, )i./ w cwout" 0. I nw. A~ev that u
~CM"t}' off to-the, v~ ~ evvui-
cont'~~~ LA;p.
Y Tum s around near
y I = 5.3, X= 0.1 3 1
t = 1.3
X = 96.091 y= l X
100
Figure 3-5b Figure 3-Sc
In Example 2, you will use y versus t graphs to analyze the same motion.
100 Chapter
3: Derivatives,
Antiderivatives,
andIndefinite Integrals
Example2 The object in Example 1 moves in the x-direction with displacement from the y-axis
given by
X = 3t 3 - 30t 2 + 64t + 57, fort ?c 0,
where x is in miles and t is in hours .
a. Find equ ations for its velocity and acceleration.
b. At time t = 2, is x increasing or decreasing? How fast?
c. At time t = 2, is the object speeding up or slowing down? At what rate ?
d. At what times in the interval [O, 8) is x at a maximum? Justify your answer.
e. Is x ever negative if tis in the interval [O, 8]? Justify your answer.
dx
Solution a. V = dt = 9t 2 - 60t + 64 Velocity equation .
dv d 2x
a =- =- = 18t - 60 Acceleration equation.
dt dt 2
b. At t = 2, v = 9(2 2 ) - 60(2) + 64 = -2 0. Thus, xis decreasing at 20 mi / hr.
The fact that x is decreasing can be seen from a graph like that shown in
X
Figure 3-5d. Note that x is plotted vertically because it is the dependent
variable.
c. At t = 2, a= 18(2) - 60 = - 24, which means that the velocity is decreasing.
However, the object is speeding up. As shown in Figure 3-5e, v is negative at
2 t = 2. If t increases a bit, v becomes larger in the negative direction . To check
for speeding up or slowing down, ask yourself if v is getting closer to the
Figure 3-Sd t-axis (slowing down) or farther from it (speeding up) as time increases . The
object is speeding up in the negative direction at 24 (mi/ hr) / hr when t = 2.
Velocity, v d. The maximum value of a function occurs where it stops increasing and
starts decreasing, or an endpoint of the domain. The closed interval [O, 8]
100 means that O :,; t :,; 8. From th e graph in Figure 3-5d, you can see that x is
a maximum in this doma in when t = 8. However, there is another relative
maximum somewhere between t = 1 and t = 2. To find exactly this value of
8 t, realize that the velocity will be zero then .
9t 2 - 60t + 64 = 0 Set the velocity equa l to zero.
Figure 3-Se . 60 .J3600 - 4(9)(64)
.. t = 2(9) Use the quadrati c formula .
t = 5.3333 ... or 1.3333 . ..
A relative maximum occurs at t "" 1.33 hr . Pick th e valu e between 1 and 2.
The absolute maximum occurs at t = 8 hr . An endpoint maximum.
e. From Figure 3-5d it appears that x is zero som ewhere between t = 5 and
t = 6. The lowest value of x will occur where the object stops traveling in
the negative direction and starts back in the positive dir ection. That is, the
minimum will occur where the velocity is zero. In step d it was found that
v = 0 when t = 5.333 .... Substituting this value for t gives
X = 3(5.333 .. . )3 - 30(5.333 .. .)2 + 64(5.333 ... ) + 57 = 0.111 . . ..
Because x is still positive at the lowest point in the interval [O,8], it is never
negat ive in that interval.
Sectio
n 3-5: Displacement,
Velocity,
andAcceleration 101
The following box summarizes what you should know about velocity and
acceleration.
Properties:VelocityandAcceleration
If x is the displacement of a moving object from a fixed plane (such as the ground), and t
is time, then the fol10vvingare true .
. , dx
Vel OClty = V = X = dt
. dv ,, d2x
Aceeleranon = a = - = x = -. 2
dt dt
Problem Set 3-5
DoTheseQuickly
The following problems are int ended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
QI. Quick! For what values oft does 9t 2 - 30 t + 64 equal zero? (Be clever!)
Q2. Find an equation for dy / dx: y = 5x 2 .
QJ. Find an equation for y' : y = ux - 3 .
Q4. Find an equation for f'(x) : f(x ) = x L7 .
QS. Find (d/dx)(3x + 5).
Q6. Find f(3 ): f(x ) = 5x 2 .
Ql. Findf '(3) : f(x ) = 5x2 .
QB. Find lim x- 3 5x 2 .
Q9. In the definition of limit, which goes with f(x ) , E or o?
QIO. Which concept of calculus is used to find the product of x and y if y varies?
For Problems 1 and 2, find equations for the velocity, v, and the acceleration, a, of a
moving objectify is its displacement.
l. y = 5t 4 - 3t 24 + 7t 2. y = 0.3t - 4 - St
For Problems 3 and 4, an object is moving in th e x-direction with displacement, x feet,
from the y-axis given by the function of t seconds. With your grapher in parametric
mode, plot the path of the object . Hold or repeatedl y pre ss the trace key to find out
how the object moves as time goes on . Write a paragraph describing the motion . Include
the approximate times and places at which the object reverses direction, and the time
intervals during which th e object is traveling to the right and to the left. A sketch may
help .
3. X = - t 3 + 13t 2 - 35t + 27, t <::.0 4. X = t4 - l lt 3
+ 38t 2 - 48t + 50, t <::.0
5. An object moves as in Problem 3, with displacem ent, x, given by
3 2
X = -t + l3t - 35t + 27,
where x is in feet and t is in seconds.
10 2 Chapter
3: Derivatives,
Antiderivatives,
andIndef
inite Integrals
a. Find equations for the velocity and the acceleration.
b. At t = l, is x increasing or decreasing? How fast?
c. At t = 1, is the object speeding up or slowing down? At what rate?
d. At what times in the interval [O, 9] is x a relative maximum ? Justify your answer.
e. Is x ever negative if tis in the interval (0, 9]? Justify your answer.
6. A particle (a small object) moves as in Problem 4, with displacement, x, given by
X = t4 - l lt 3
+ 38t 2 - 48t + 50,
where x is in meters and t is in minutes.
a. Write equations for the velocity, v, and acceleration, a.
b . Find all values oft for which v = 0. You may use the solve feature of your grapher .
c. Plot the graphs of x and v on the same screen, then sketch the results. Tell what is
true about the displacement whenever v = 0.
d. Plot the graph of a on the same screen you used in 6c. At each time the acceleration
is zero, what is true about v7 About x?
7. Car Prob lem: Calvin's car runs out of gas as it is going up a hill. The car rolls to a stop,
then starts rolling backward. As it rolls, its disp lacement, d(t) feet, from the bottom
of the hill at t seconds since Calvin's car ran out of gas is given by
d(t) = 99 + 30t - t 2 .
a. Plot graphs of d and d' on the same screen. Use a window large enough to include
the point where the d graph crosses the positive t-axis. Sketch the result.
b. For what range of times is the velocity positive? How do you interpret this answer
in terms of Calvin's motion up the hill?
c. At what time did Calvin's car stop rolling up and start rolling back? How far was it
from the bottom of the hill at this time?
d. If Calvin doesn't put on the brakes, when will he be back down at the bottom of the
hill?
e. How far was Calvin from the bottom of the hill when the car ran out of gas?
8. Sky Diver's Acceleration Prob lem: Phoebe jumps
from an airplane . While she free-fa lls, her down-
ward velocity, v(t) feet per second, as a function
of t seconds since the jump, is
v(t) = 251(1 - 0.88 1 ).
a. Plot the velocity, v, and acceleration, a, on
the same screen. Use an x-window (actually
a t-window) of O sec to 30 sec. Sketch the
results.
b. What is Phoebe's acceleration when she first
jumps? Why do you suppose the acceleration
decreases as she moves faster and faster?
c. What does the limit of v (t) seem to beast approaches infinity? This limit is called
the termi nal velocity.
Section3-5:Displacement,
Velocity,andAcceleration 103
kt
C
d. How many seconds does it take Phoebe to reach 90% of her terminal velocity?
Explain how you got your answer.
e. When Phoebe reaches 90% of her terminal velocity, is her acceleration equal to 10%
of its initial value? Justify your answer.
9. Velocity from Displacement Problem: If you place-kick a football, its displacement
above the ground, d(t) meters, is given by
2
d(t) = 18t - 4.9t ,
where t is time in seconds since it was punted (Figure 3-Sf).
Displacement, m
(The coefficient 18 is the initial upward velocity in meters per second.)
20
a. Find d'(l) and d '(3). What name from physics is given to d'?
b. At times t = 1 and t = 3, is the football going up or down?
How fast? How does the derivative tell you this? How do you
know this from the graph? Time, sec
5
c. Use the d' equation to find the velocity at time t = 4. Explain
why the answer has a meaning in the mathematical world but Figure 3-5f
not in the real world.
10. Displacement from Velocity Problem: A sports car accelerates in such a way that its
velocity as a function of time is given by
v(t) = 15t 0 6,
where v(t) is in feet per second and tis in seconds. Figure out an equation for x(t), the
displacement of the car from a fixed point. Assume that the car is 50 ft from the fixed
point at time t = 0. What concept of calculus have you used to find the displacement
equation? Where is the car when t = 1O?How far does it travel between t = 0 and t = 1O?
11. Average Versus Instantaneous Velocity Problem: Suppose that f(t) is the displacement
of a moving object from a fixed plane. Explain why the difference quotient,
f(b) - f(a)
b- a
is the average velocity from t = a to t = b. How is the instantaneous velocity at t = a
related to this average velocity ? The definition of derivative should help you answer
this question.
12. Figure 3-5g shows the displacement, y, in centimeters, from a tabletop for a mass
bouncing up and down on a spring. Time t is in seconds.
a. Sketch or photocopy the graph. On the same axes, sketch the y
graph of the velocity.
b. For what values oft is y a relative maximum? For what values
oft is y a relative minimum?
c. At what times is the velocity a relative maximum ? Make some
observations about the displacement graph at these times.
,, .. ,.
d. Just for fun, see if you can duplicate the graph in Figure 3-5g ..
on your grapher. If you can, then check your answer to 12a Figure 3-5g
by plotting the numerical derivative.
104 Chapter
3: Derivatives,
Antiderivatives,
andIndefinite
Integrals
3-6 Introduction to Sine, Cosine,
and Composite Functions
You have learned how to write down a formula for the derivative of a linear
comb ination of power functions. Now you will do this for the transcendental
functions sine and cosine .
Figure 3-6a shows the graph of y = sin x and its numerical derivative as they
might appear on your grapher . ote that the derivative graph is congruent to the
sine graph, just shifted sideways . Each graph is called a sinusoi d. Recall from
trigonometry that the cosine function has a graph that looks just like the derivative
graph. A conjecture is as follows.
If f(x) = sinx, then f'(x) = cosx.
Figure 3-6b shows similar relationships for the graphs of g(x) = cosx and its
instantaneous rate of change. This time, the pattern seems to be the opposite of sin
x, so the conjecture is as follows.
If g(x) = cosx, then g '(x) = -s inx.
These two conjectures turn out to be true, as you will prove in Section 3-8.
y
y =C OS X
f I y 2 = derivative
i
'
' 1
~
;
,~',L __
....._ .........
'
...........,,....
'\ Looks like the,
ne gative of the
sine graph!
Figure 3-60 Figure 3-6b
Properties:Derivativesol Sineand CosineFunctions
:x (sinx) = cosx !!_(cosx) = - sinx
dx
In this section you will try to discover how to find the derivative of a compos ite
function, such as
f(x) = sin (x 5 ) or g(x) = (cosx)4,
where an operation is performed on the answer to another function . The function
performed first is called the insi de function (the fifth power function inf and the
cosine function in g, above). The function performed second, on the answer to the
inside function, is called the outside function (the sine in f and the fourth power
in g, above).
Section
3-6: Introductionto Sine, Cos
ine, andComposite
Functions 105
OBJECTIVE Work with your study group to form conjectures on how to differe nti ate a composite
function.
Problem Set 3-6
1. Plot the graph of y = sinx on your grapher. Use a friendly
window starting at x = 0 and ending at some convenient place
around x. = 10. Draw the resulting graph on graph paper, using
scales such as those shown in Figure 3-6c. y
3
2. You have assumed that the derivative of f(x ) = sinx is f'(x) =
cosx. Based on this assumption, evaluate f'(2) . Then write the X
equation of the line through the point (2 ,f ( 2)) that has slope
f'(2). Plot that line as y 2 on your grapher and show it on the
graph in Problem 1. What relationship does the line have to
the sine graph? How does this relationship confirm that the Figure 3-6c
derivative of the sine function really is the cosine function?
g(X) ' .
3. Figure 3-6d shows g(x) = sin 3x.. Make a conjecture about what 3
an equation for g '(x) might be. Enter g(x) as y 1 on your grapher.
Then plot Y 2 = your presumed derivative and y 3 = the numerical
derivative of g . Does the result verify or refute your conjectured
derivative? Sketch the correct g' graph. If your conjecture was
wrong, write the correct equation for g '(x).
4. Figure 3-6e shows h(x) = sinx. 2 . Make a conjecture about what Figure 3-6d
an equation for h '(x) might be. Then verify (or refute!) your
h(X)
conjecture by appropriately graphing on your grapher. Sketch
the correct h' graph. If your conjecture was wrong, see if you
can figure out a correct equation for h '(x).
5. You have assumed that the derivative of a power formula works
for any constant exponent. Use this fact and the patterns you
have observed in Problems 1-4 to make a conjecture about the
derivative function for t(x ) = sinx 0 7 Verify your conjecture by Figure 3-6e
plotting graphs as you did in Problems 3 and 4.
6. Suppose that f(x.) = sin [g(x) J for some differentiable function g. What special name
is given to function f in this case? What special name is given to function g? What
special name is given to the sine function in this case? Write a statement telling how
you could find an equation for f'(x.).
7. For each function below, tell which is the inside function and which is the outside.
a. f(x) = sin3x b. h(x) = sin 3x
c. g(x) = sinx. 3 d. r(x ) = z cosx
1
e. q(x.) = -- f. L(x) = log (secx)
tanx.
8. Write a paragraph in your journal, telling what you learned as a result of doing this
prob lem set that you did not know before.
106 Chapte
r 3: Derivatives,Antiderivatives,
andIndefinite Integrals
3-7 Derivatives of Composite Functions-
The Chain Rule
A function with an equation such as
f(x) = sin (x2 )
is called a composite function. It is composed of two other functions. The function
x 2 in parentheses is called the inside function. The sine function outside the
parentheses is called the outside function. It operates on the answer to the first
function. Your purpose in this section is to learn how to find the derivative of a
composite function.
OBJECTIVE Given th e equation for a composite function, writ e the equation for its derivative function.
Suppose you set out to differentiate
2
f(x) = sin(x ).
Your first reaction might be: This is just the derivative of the sine function, so the
answer is cos(x 2 ). As you saw in Problem Set 3-6, the answer is really
f'(x ) = cos (x2 ) 2x.
f(x)
The anticipated result, cos (x 2 ), is multiplied by the derivative of the inside function,
2x. Figure 3-7a shows the graphs of f, f', and the numerical derivative of f. The
algebraic and numerical derivative (dashed) are the same. The property that tells
I I I I I I
you how to differentiate a composite function is called the chain rule.
I
I I I I I I
\ I
\I
I
I
I I I
I
I I
I
To see why the chain rule works, realize that there are really three functions
involved in f(x) = sin (x2 ): the inside function, x 2 ; the outside function, sine; and
Figure 3-7a the composite function, sin (x2 ). Each one has its own derivative. For simplicity, let
u = x 2 and let y = sin u. Graphs of the three functions are shown in Figure 3-7b.
y = sin u
,,
Slope= du / dx
= 2x ~
, Slope= dy/ du
=C OS U
I
I ,,
I X u
Inside function Outside function Composite function
Figure 3-7b
Observe that y depends on u and that u depends on x. The derivative of u with
respect to x, du / dx, is 2x. It tells the slope of the xu-graph (left). Similarly, the
derivative of y with respect to u, cos u, tells the slope of the uy-graph . You seek
the slope of the xy-graph, dy / dx (right). It is the limit of 6.y / 6.x . (The words with
respect to u, for instance, mean that u is the independent variable.)
Section3-7: Derivatives
of Composite
Functions-The
ChainRule 107
Although derivatives are not fractions, you can use th e definition of derivative to
write them that way.
dy = lim t.y dy = lim t.y du = lim t.u
dx 6x - O t.x du .lu -0 t.u dx 6x - o t.x
To find the fraction for dy / dx in terms of the fractions for dy / du and du / dx, use
some algebra. Multiplying by a clever form of 1, namely t.u / t.u, gives th e following .
dy = lim (t.y . t.u)
dx 6x-o t.x t.u
dy = lim ( t.y . t.u) Reason ?
dx 6x-o t.u t.x
dy = Jim t. y Jim t.u Limit of a produ ct of two functi ons.
dx .lx - o t.u 6x-o t.x
If u is a continuous function of x, then t.x - 0 implies t.u - 0. By the limit of a
composite function property (Section 2-3) you can replace t.x - 0 with t.u - 0.
dy = lim t.y lim t.u
dx .1u- o t.u ox - O t.x
The two limits now fit the definition of derivative. Therefore,
dy dy du
dx = du dx
The name cha in rule is us ed because when a composite function has severa l insid e
functions, you get a who le "chain" of derivatives multiplied together. Note that since
t.u is a change in the dependent variable, it might go to zero somewhere before t.x
gets to zero . There is a way to get past this diffi cult y by using the extended mean
value theorem, which you may learn lat er in your mathematical career.
Property:TheChainRule
dy/dxForm:
If y is a differentiable function of u and u is a differentiable function of x,
then the derivative of y with respect to x is given by
dy dy du
dx = du dx
f(x) form:If f'(x) = g(h(x)), then f '(x) = g'(h(x)) h'(x)
Outside
function, form:To differentiate a composite function, differentiate
insidefunction
the outside function with respect to the inside function, then multiply by the derivative of
the inside function with respect to x.
It is safest to remember the cha in rule as a procedure, in its outside function, inside
function form. This form indicates something to do. It is also easy to remember the
dy / dx form. Just think of "canceling" the du's, and you are left ,vith dy / dx.
With the chain rule at hand, you can accomplish this sect ion's objective .
108 Chapter
3: Derivatives,
Antiderivatives,andIndefinite Integrals
Example 1 If f(x) = (x 3 + 7) 2 , find f'(x).
Solution f(x ) = (x 3 + 7) 2 Writ e th e gi\'e n func tion's equati on.
3 1 2
f'(x ) = 2(x + 7) 3x Apply the chain rule.
\ \ Derivative or th e ins ide fu nction
Deriva tive of the out sid e fwic tion
2
f'(x ) = 6x (x + 7) 3
Simplify.
Example 2 If y = cos 4 x, find dy / dx.
Solution y = cos 4 x = (cosx) 4
Write the fourth pow er as an out side fu nc tion.
3
:: = -!(cosx) (- sinx ) Differ enti ate th e out side and inside fu nctio ns.
= - 4 cos 3xsin X Simplif y.
Problem Set 3-7
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Q1. Write one condition for f to be continuous at x = c.
Q2. Write another condition for f to be continuous at x = c.
QJ. Write the third condition for f to be continuous at x = c.
Q4. Is the signum function, f (x) = sgn x, continuous at x = O?
05. Find dy / dx if y = 20x 4 15 .
Q6. Find f(x) if f'(x) = 3ox- 4 .
Ql. Function fin Problem Q6 is called a(n) - ?- .
QB. Sketch the graph of y = sinx.
Q9. Sketch the graph of y = cos x .
QIO. Sketch the graph of y = - sinx.
l . State the chain rule in each form.
a. Use dy / dx terminology.
b. Use f'(x ) terminology.
c. State verbally, using the words inside function and outside function.
2. Given f(x ) = (x 2 - 1) 3 :
a. Differentiate using the chain rule.
b. Expand the power, then differentiate term by term.
c. Show that the answers to a and b are equival ent.
Section3-7:Derivativesof Composite
Functions-The
ChainRule 109
bf:'
C
For Problems 3-22, find an equation for the derivative function . You may check your
answer by comparing its graph with the numerical derivative graph.
3. f(x) = cos 3x 4. f(x) = sin 5x 5. g (x) = cos (x3 )
6. h(x ) = sin(x 5) 7. y = (cos x) 3 8. f(x) = (sinx ) 5
9. y = sin 6 x 10. f (x) = cos 7 x 11. y = -6s in 3x
d
12. f(x) = 4cos( - 5x) 13. dx (cos 4 7x) 14. :x (sin 9 13x)
15. f(x) = 24 sin 5134x 16. f (x) = -100sin 61S(-9x) 17. f (x) = (5x + 3) 7
18. f(x) = (x 2 + 8)9 19. y = (4x 3 - 7)- 6 20. y = (x 2 + 3x - 7)- 5
21. y = [cos (x 2 + 3) J100 22. y = [cos (Sx + 3)4 ]5
23. Graphical Ver ificat ion Probl em : For f(x ) = 5 cos 0.2x, plot the graph
of function f. Where x = 3, plot a line on the graph with slope equal ----
to f'(3) . Show that the line really is tangent to the graph.
24. Beanstalk Prob lem: Jack's beanstalk grows in spurts (Figure 3-7c).
Its height, y feet, above the ground at time t hours since he planted
it is given by
y = 7 sin rr t + 12 t 1 2 .
Write an equation for dy / dt. Plot the graph of y and th e velocity
graph on the same screen. Use a window from about t = 0 tot = 10.
Do there appear to b e times when the beanstalk is shrinking? Justif y Figur e 3-7c
your answer .
25. Ba ll oon Vol ume Prob lem : A spherical balloon is being inflated with air (see Figure
3-7d). The volume of the sphere depends on the radius, and the radius depends on
time. Thus, the volume is a composite function of tim e.
a. The volume of a spher e is V = (4 rr / 3) r 3 , where r is the radius in
cm. Find an equation for dV / dr. What are the units of dV / dr?
b . At time t = 0, the radius is 10 cm. If r increases at 6 cm / min,
writer as a function oft.
c. Find an equation for dr / dt. Surprising ?! What are the units
ofdr / dt?
d. By appropriate use of th e chain ru le, find dV / dt when t = 5 min.
Based on the units of dV / dr and dr / dt, explain why th e units
of dV / dt are cm 3 / min .
e. Find dV / d t direct ly by substituting r from 2 Sb into the equation
for V Show that you get the same answer you did in 25d for Figure 3-7d
dV / dt when t = 5.
26. L\u and L\x Prob lem: The derivation of th e chain rule states that if u is a continu ous
function of x, then L\x - 0 impli es L\u - 0.
a. Sketch a graph showing that this may not be true if u has a step discontinuit y.
b . Sketch a graph showing why this is true if u is continuous.
110 Chap
ter 3: Deriva
tives,Antiderivatives
, andIndefiniteIntegrals
3-8 Proof and Application of Sine and
Cosine Derivatives
In Section 3-6 you discovered that the derivative of the sine function is the cosine
function, and th e derivative of the cosine function is the opposite of the sine
function. In this section you will prove algebraically and geometrically that these
prop erties are true . You will also apply the derivatives of these sinusoidal functions
to some problems from the real world.
OBJECTIVE Be able to derive algebraically the formulas for the derivatives of sin x and cos x. Find
rates of change of sinusoidal functions in real-world problems .
Background:
TheLimitof (sinx}/x
The function
(sin x)
y =--
x
y
takes on the indeterminate form 0/ 0 as x approaches zero . If you use a friendly
window that includes x = 0 (Figure 3-8a) to graph the function, you will see that
the fraction gets very close to 1 as x gets close to zero. By usin g the appropriate
X
geometry, it is possibl e to pro ve that 1 is the limit of y as x approaches zero.
-1 l
Figure 3-8b shows a unit circle with an arc of length x cut off by an angle of x
radians . The line segment tangent to the arc has length tan x (hence the term
Figure 3-8a tangent) . The half-chord perpendicular to the u-axis has length sin x. The distance
along the u-axis from the origin to this chord is cos x.
V Consider the areas of three regions shown in Figure 3-8b.
Unit circle
Large triangle : Area= (l)( tanx ) Area = <ha e)(altitude) .
sin x
Sector inscribed in the large triangl e: Area = (x) A fraction of a unit circle, (x / 2rr ). (rr . 12 )
Small triangle inscribed in the sector: Area= (sinx)(cosx)
tanx
u
1 From the areas of these figures, you can write the following three-member inequality,
COS X
then use the appropriate algebra.
Figure 3-8b
(sinx)(cosx) < (x) < O)(tanx)
x tanx
cosx < -- < -- Divide all thre e members of the inequality by sin x, then multiply by 2.
sinx sinx
1 sinx sinx
-->-->-- The order reverses because all three members of the inequality
cosx x tanx are th e same sign.
sinx
secx > -- > cosx Becaus e 1/ (cosx ) = secx, and (sin x) / (tanx) = cosx .
X
Section
3-8:ProofandApplication
of SineandCosine
Derivatives 11 1
~C
y Figure 3-8c shows how the graphs of the three members of this inequality would
I
,secant appear on your grapher. Sec x and cos x both approach 1 as x approaches zero . So
'' ...... 1 , ~ ' sinx (sin x) / x is also "squeezed" to 1. Be careful not to read too much into this answer.
X (Sin 0)/ 0 is undefined, but has 1 for a limit as the argument, x, approaches zero.
cosin~ ',, x
The indeterminate form 0 / 0 can approach numbers other than 1.
-1
Property:Limitof (sinx}/x
Figure 3-Bc
. sinx
1Im --
x-0 X
=1
The function (sinx ) / x approaches 1 because it stays between two other functions,
eac h of whose limits equals l. This fact is an example of the squeeze theorem. The
theorem follows dir ectly from the definition of limit. Figure 3-8d shows functions
f, g, and h. If g(x) and h (x) can be kept within E units of L just by keeping x close
enough, but not equal to, c, and if f(x) is always between g(x) and h(x), then f(x)
can also be kept vvithin E units of L. Hence, Lis also the limit of f (x ).
X
Property: TheSqueezeTheorem
Figure 3-Bd
lf: 1. g(x) :5 h(x) for all x in a neighborhood of c, where x * c,
2. limx -c g(x) = Iimx -c h(x) = L, and
3. f is a function for which g(x) 5 f(x) 5 h(x) for all x in a neighborhood of c;
then limx -c f(x) = L.
No te: A neighborho od of a numb er c is an open int erval containing c . Because the
inter val is open, c cannot b e at an endpoint. Thus, there are numbers in the interval
on both sides of c.
Derivative
of theSineFunction
Let f (x) = sinx . You suspect that f'(x) = cosx. By the definition of derivative,
'( ) = r sin(x + h) - sinx
f X l!:J h .
Algebra with fractions is usually easier if th e numerator and the denominator have
one term each, perhaps consisting of a product of several factors . The following
property from tri gonometr y can be used to transform the numerator above from a
sum to a product .
sin A - sinB = 2 cos (A+ B)sin (A - B) A sum and a product prop erty
fr om trigonometry .
. , . 2cos[(x+ h )+x]s in [(x+ h )-x]
.. f (x) = hm h Use (x + h) for A and x for B.
h-0
f'(x) = lim
2 cos ( x +
h
i) sin i
h- 0
2 sin~
f'(x) = lim cos ( x + ~ ) lim - h 2 Limit of a product property.
h- 0 h- 0
112 Chapter
3: Derivatives,
Antiderivatives,
andIndefiniteIntegrals
, 2 sin!:!
f (X) = (COS X) lim - h 2 Tak e limit of first factor, assuming cosine is continuo us.
h-0
sin !:1.
f' (x ) = (COS X) lim
h -0
---,.,1--Express th e seco nd factor as sin (argu ment ), whose limit is 1.
argum ent
2
f'(x) = (cos x) (1) = cos x, which you discovered graphically in Section 3-6.
Derivative
of theCosineFunction
Using the cofunction properties from trigonometry , this new problem becomes an
old problem.
y = cos x = sin ( f - x) =} y' = cos ( f - x) ( -1) = - cos ( f - x) = - sin x
So the property you discovered graphically in Section 3-6 is correct.
Sinusoidal
Equations
fromReal-World
Information
The graph of a sine or a cosin e function is called a sinusoid. Such functions occur
in the real world, particularl y with periodic motion. To write an equation for a given
sinusoid, you must first recall
( (x) One cycle
the relationships between the - - - - , - - - - - - - - - - - - - , - - - - Upper bound
equation and the graph . Figure
3-8e shows a sinusoid that is
moved away from the origin. A
.-+-...._...._..-/-
_.... ~~~soida l
cycle of the graph goes from one - ~- - Lower bound
point to the point where the graph :~ Phase
: displa cement
: 1
Vertical
first starts repeating itself. The ' displacement
period is the number of x-units
j---Period---+- j i X
taken to complete one cycle (for
instance, the x-distance between
two high points) . Figure 3-8e
The phase displacement is the x-coordinate of the beginning of the "first" cycle,
where the argument of th e sine or the cosine equals zero. The horizontal axis
running along the middl e of the graph is called the sinusoida l axis . The amplitude
is the y-distancc between the sinusoidal axis and a high point or a low point. The
vertical displacement is the y-distance from the x-axis to the sinusoidal axis .
The general equation of a sinusoid is
f(x ) = C + A cosB(x - D ) ,
((x)
where A, B, C, and D stand for constants and the function can be either cosine or
sine. To see the effects of these constants on the graph, plot a sinusoid such as
y = 5 + 3 cos 2 (x - 1)
X
1 X= 4.14 )'=8 on your grapher (Figure 3-8f). The 5 indicates how far the graph is shifted up wards
and is thus equal to the vertical displacement (count up 5 spaces from the x-axis).
Figure 3-8f The 3 indicates how far the graph is dilated in the y-direction and is thus equal to
the amplitude (count 3 spaces from the sinusoida l axis).
Section
3-8: Proof andApplicationof SineandCosine
Derivatives 113
t hv
~
C
The argument of the cosine, 2 (x - 1), is zero at x = 1. So the "first" cycle starts
th ere , meaning the phase displacement is 1. To find the p eriod, recall that y = cos x
has a period of 2 rr, the number of radians in a circle. The 2 in 2 (x - 1) makes the
argument change twice as fast. So the period is only 1/ 2 of 2rr, or simply rr . Thus,
th e next high point comes at x = 1 + rr ::::::
4.14.
From the information above you can reach the following conclusions .
Properties:Graphol a Sinusoid
If y = C + A cos B ( x - D), then the following are true.
1. The sinusoidal axis is along the line y = C.
2. The amplitude equals IAI.
3. The period equals ~;.
4. The phase disp lacement equals D.
Example 1 A mass is bouncin g up and down on a spring hanging from the ceiling (Figure 3-8g).
Its distance, y feet, from the ceiling varies sinusoida lly with tim e t seconds, making
a complete cycle every 1.6 seconds. At t = 0.4, y reaches its greatest value, 8 ft. The
smallest y gets is 2 ft .
a. Write an equation for yin terms oft.
b. Write an equation for the derivative, y ' .
c. How fast is the mass moving for these values oft ?
i. t =1 ii. t = 1.5 iii. t = 2.7
d. At t = 2.7 sec, is the mass moving up or down ? Justify your answer .
e. What is th e fastest the mass moves? Where is the mass when it is moving this
fast?
Figure 3-Bg
Solution First sketch the graph. Show the high point at t = 0.4. The next high point will
be at t = 0.4 plus the period, or 2.0. Halfway betw een the two high points, at
y t = 1.2 will be a low point. From thes e points you can sketch a reasonable graph,
::I\/1:
as shown in Figure 3-8h.
a. C = (2 + 8) = 5 Sinu so idal ax.is is halfway between upper and lower bounds.
A= 8 -5=3 Amplitude is from sinu soidal ax.is to upp er bound.
2rr
0.4 2.0
B= - = l.25rr There are 1.25 rr cycles in 2rr units of x because th e period is 1.6.
1.6
D = 0.4 A high point occurs at x = 0.4.
Figure 3-Bh
equation is y = 5 + 3 cos l.25rr(t - 0.4).
b. y' = -3s inl.25rr(t - 0.4) l.25rr Derivative of cos is - sin. Use th e chain rule!
y ' = -3 .75rr sin l.2 5rr(t - 0.4)
11 4 Chapter
3: Derivatives,
Antiderivatives,
andIndefinite
Integrals
c. Plot the graphs of y and y ' on the same screen (Figure 3-8i), then trace the
y ' graph to find the values .
i. t = l: y' = - 8.3304 . .. "' - 8.3 ft / sec
ii. t = 1.5: y ' = 10.884 .. . "' 10.9 ft / sec
iii. t = 2.7 : y ' = -4.508 ... "' - 4.5 ft / sec
d. At t = 2.7, the mass is going up; y ' is negative, meaning y (the distance
between the mass and the ceiling) is getting smaller.
I /
I
\ I ,,
I/
e. The fastest the mass moves is 3.75rr, or about 11.8 ft / sec, which equals
y2 = -4 .508383
the amplitude of y ' . Tracing shows a high point of y ' at t = 1.6 (dotted line
in Figure 3-8i). At this time the mass is halfway between its high and low
Figure 3-8i points.
Problem Set 3-8
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Differentiate: f (x) = x 9
Q2. Find dy I dx : y = 3 cos x .
Q3. Find y ' : y= (5x 6 + 11) 2.4 _
Q4. Find s'(x): s (x) = 2- 13 . (Be careful!)
. d . (x + 5) (x - 7)
Q5. Fln 1lmx- 7
x- 7
Q6. Find limx-o 2x.
Ql. Is f(x) --
= 1/ (x - 3) comtnuous at x = 4?
QB.If f'(x) = 2x sinx 2 , find f(x).
-:-'
Figure 3-8j
Q9. Fill in the blank: cos2 x + sin 2 x = -?-.
Q10. Sketch the graph of the derivative of the function shown in Figure 3-8j.
1. Ferris Wheel Problem: When you ride a Ferris whee l, your
distance, y(t) feet, from the ground varies sinusoidally with
time t seconds since the wheel started rotating. Suppose that
the Ferris wheel ha s a diameter of 40 ft and that its axle is 25 ft
above the ground (see Figure 3-8k). Three seconds after it starts,
your seat is at its high point. The whee l makes 3 rev /min .
a. Sketch the graph of function y. From the sketch, figure out
the particular equat ion for y(t).
b . Write an equat ion for y '(t).
y( t)
c. When t = 15, is y(t ) increasing or decreasing? How fast?
Where is the seat when y(t ) is changing the fastest? !
d. What is the fastest y(t ) changes?
Figure 3-8k
Section
3-8:ProofandApplication
of SineandCos
ineDerivati
ves 115
t /2v
~
C
2. Pendulum Probl em: A pendulum hung from the ceiling makes a
comp lete back-and-forth swing each 6 sec (see Figure 3-81). As the
pendulum swings, its distance, d cm, from one wall of the room
depends on th e number of seconds, t, since it was set in motion. At
t = 1.3 sec, dis at its maximum of 110 cm from the wall. The lower
bound of dis 50 cm. Assume that d is a sinusoidal function oft.
Wall
a. Write an equation expressing d as a function oft.
b. Write an equation for the derivative function.
c. How fast is the pendulum moving when t = 5? When t = 11? How
do you explain the relationship b etween these two answers? - -------- ----
d. When t = 20, is the p endulum moving toward or away from the
wall? Explain .
Figure 3-81
e. What is the fastest th e p endulum swings7 Where is the p endulum
when it is svvinging its fastest?
f. What is the first positi ve value of t at which the pendulum is
swinging O cm / sec7 Where is th e p endulum at this time?
3. Playground Probl em: Opportunity Park, a playground in Midland, Texas, has a sinu-
soidal ramp for children to walk up (see photo). A bri ck curb starts 0.75 ft above the
ground and slopes up to 3.25 ft above the gro und , -!4 ft away. The concrete walkway
varies from 0.7 5 ft b elow this curb to 0.25 ft below it , as shown in Figure 3-8m. Let x
be the number of feet from the b eginning of the ramp.
a. Write an equation for f(x), the di sta nce from
the ground to the top of th e bri ck curb.
b . Write an equation for g(x), the distanc e
from the ground to the top of the concrete
walkway. Take into account th e slope of the
ramp .
c. Write an equation for g '(x). What is the slope
of the ramp at x = 9? What is th e slope at
x = 15? What are the units of th e slope? At
these places, would a child walking up the
ramp be going upward, downward, or on th e
level? Explain.
d . As a child walks up the ramp, what are the steepest up and steepest down slopes?
Concrete wa lkway Brick curb
3 1
X
8 16 2-1 32 40 -1-1
Figure 3-Bm
116 Chapter
3: Derivatives,
Antiderivatives,
andIndefinite
Integrals
4. Daylight Problem : The length of daylight (sunrise to sunset) is, approximately, a
sinusoidal function of the day of the year. The long est day is around June 21, the
172nd day of the year. In San Antonio, Texas, the long est da y has 14 hr, 3 min of
daylight. The shortest day has 10 hr, 15 min of daylight.
a. Write an equation expressing the length of daylight, in minutes per day, for San
Antonio as a function of day of the year. How much daylight is ther e on August 7?
b. Write an equation for the derivative of the function in 4a . At what rate is the length
of daylight changing on August 7? What are the units of this rate of change?
c. What is the greatest rate of change of daylight length ? On what two da ys of the
year is the rate equal to this number?
5. Pendulum Experiment : Swing a weight on a string from the ceiling of your classroom.
Make appropriate measurements to find an equation for distance from the wall as a
function of time since you let the pendulum go. Based on your equation, calculate the
fastest the pendulum moves as it swings .
6. Day light Researc h Project: Obtain a chart of sunrise and sunset times for various
days of the year in your locality . Use this chart to derive a sinusoidal equation for
the number of minutes of daylight in a day as a function of th e day of the year. See
Problem 4 for ideas. Draw a graph of predicted length of daylight versus day . Use your
grapher to get the plotting data. On the same axes, plot the actual length of daylight
derived from your chart. Discuss how well the sinusoidal model fits the real data . What
is the greatest change in length of daylight from the table? How does this number
compare with the derivative of the sinusoidal function at that time of year?
7. Squeeze Theorem, Numerically : Define functions f, g, and h as shown.
f (x ) = -Z x2 + 8x - 2
2
g(x) = Zx + 2
h (x) = 4x
a. Plot the three functions on your grapher. What is the limit of each function as x
approaches 1?
b. Show that the squeeze theorem applies to the three functions. Which function is
the upper bound, which is the lower bound, and which is in between?
c. Make a table of values of f(x), g(x), and h(x) for each 0.01 unit of x from 0.95 to
1.05. Use a time-efficient method, such as the table or trace feature of your grapher.
d. From your table, determine a value of o that can be used to keep both the upper
and lower bound functions within 0.1 unit of the limit when x is within o units of 1.
e. Based on your table, explain how the conclusion of the squeeze theorem is true.
8. Lim it of (sinx ) / x, Numerically: The adjacent table shows values of (sinx) / x as x gets
closer to zero. Don't forget to use radian mode!
a. Use the table or trace feature to verify that these numb ers
X (s inx) /x
are correct.
0.5 0.9588510 ...
b. Make a tabl e of values for each 0.01 unit of x from 0.05 to
0.4 0.9735458 .. .
0.01. Do the values seem to be getting closer to l?
0.3 0.9850673 . ..
c. Use smaller increments for x to find the value of x at which 0.2 0.9933466 ...
your grapher rounds the answer to exactly 1. 0.1 0.9983341. ..
Sec
tion 3-8:Proof andAppl
icationof SineandCosi
ne Derivatives 117
t hv
~
C
d. The National Bureau of Standards lists the value of
sin 0.001 to 23 plac es as
sin0.001 = 0.00099 99998 33333 34 166 667.
(The spaces are left for ease of reading.) How close to 1 is (sin0.001 )/ 0.001? How
does this answer compare with what your calculator gives you?
e. Just for fun, see if you can explain why sin 0.001 has so man y repeated digits .
9. Limit of (sin x) / x Probl em : Without lookin g at th e text, see if you can pro ve that 1 is the
limit of (sinx )/ x as x approaches zero. If you get stuck, look at th e text only enou gh
to get going .
10. Derivative of the Sine Function : Without looking at the text, see if you can prove directly
from the definition of derivative that cos x is the derivative of sin x. If you get stuck,
look at the text only enough to get started again.
11. Derivative of th e Cosine Function : By a clever application of the chain rule and the
appropriate trigonom etr y properties, prove that -s in x is th e derivative of cos x.
12. Squeeze Theorem Problem:
a. State the squeeze theorem.
b. Prove dir ectly from the definition of limit that th e squeeze th eorem is really tru e.
c. Sketch a graph that shows you und erstand what th e squeeze th eorem really means.
13. Group Discuss ion Problem: Figure 3-8n shows th e gra ph of
1
y = 2 + (x - 1) sin ( -. - - ). y
X - 1
The function has a dis continuity at x = 1 b ecaus e of division
by zero .
a. What does the limit of y seem to be as x approaches 1?
b . Plot the grap h using a friendly window that ha s x = 1 as a
grid point and has bounds approximately as shown in Figur e
3-8n. Then zoom in on the point (1, 2) several tim es. Sketch
X
the result.
c. Find two linear functions that are upp er and lower bounds
for the graph, and us e the number in 13a for th eir limits as x Figure 3-Sn
approach es l. Plot these on your grapher, then sketch them.
d. Prove that your answer in 13a is correct. The squeeze theorem
will help!
e. Why do you suppose th e two linear functions are said to form an envelope for th e
y-graph ?
f. If x is far away from 1, y seems to approach th e constant 3. Based on what you
ha ve learne d in this section, see if you can explain why this is true.
14. Journal Problem: Updat e your journal with thin gs you have learne d since the last entr y.
You should includ e such things as those list ed here.
The one most important thing you hav e learned since your last journal entr y
What you now bett er und erstand about formulas for derivatives of functions
Properties you have learne d and what they're nam ed
Any prop ert ies or techniques about which you are st ill un sur e
118 Chapter
3: Derivatives
, Antiderivatives,and Indefinite Integrals
3-9 Antiderivatives and Indefinite Integrals
If an object is falling freely under the action of gravity, it speeds up. From physics
you may have learned that the velocity is given approximately by
v (t) = - 9.8t,
where v(t) is in meters per second and tis in seconds. The negative sign indicates
that as time increases, the distance the object is above the ground decreases.
Velocity is the instantaneous rate of change of position. That is, the velocity is the
derivative of position with respect to time. If y is the number of meters the object
is above the ground, then
dy
dt = -9 .8t.
From this differential equation it is possible to find an equation for y as a function
of t. You have to go backward from the process of taking the derivative and ask
yourself: "What could I differentiate to get -9.8t for the answer?" If you differentiate
y = t 2 , you get dy / dt = 2t. The variab le is correct, but its coefficient is not - 9.8. So
you ask, "What could I multiply 2 by to get - 9.8?" The answer is -9 .8/ 2, or -4.9 .
The function is thus
y = - 4.9t 2 .
y Interestingly, there are other functions you could differentiate to get - 9.8t, such as
y = - 4.9t 2 + 3.7,
y = - 4.9t 2 - 1776,
y = - 4.9t 2 + rr, and
X
y = - 4.9t 2 + C,
where C stands for a constant. Each function is an antiderivative of - 9.8t. An
antiderivative is also called an indefinite integral for reasons you will learn in
Chapter 5. The equation
Some antiderivatives of y = - 4.9t 2 + C
dy/ dt = -9 .Bt
is called the general equation for the antiderivative . Each of the other equations in
Figure 3-9a which a particular value appears for the constant C is called a particular equation.
Figure 3-9a shows four particular equations in which C = 40, 20, 0, and -20 .
Definition:Antiderivative
The function f is an antiderivative of function g if and only if f'(x) = g(x) for all values of
x in their domains.
OBJECTIVE Given the equation for the derivative of a function, find an equation for the function (that
is, find an antiderivative).
Sect
ion3-9:Antiderivatives
andIndefinite
Integrals 119
~
C
Example 1 If f'(x ) = x 7 , find the general equation for the antiderivative f(x).
Solution Your thought process should go like this.
The exponent is 7, so the function differentiated must have had an exponent
of 8.
The derivative of x 8 is 8x 7 , which is eight times too big.
Therefore, the function differentiated must have been (1/ 8)x 8 .
The function differentiated cou ld also have had some constant added to it.
That constant does not show up in the derivative because the derivative of a
constant is zero.
Thus, the general equation for f(x) would be
f(x) = x8 + C.
The particular equation of an antiderivative can be found if you know one point on
the graph . The coordinates of this point are called an initial condition . The initial
condition gives you the information you ne ed to find the particular value of the
constant C.
Example 2 If f'(x) = sin2x, find the particular equat ion for the antiderivative, f(x), using the
initial condition f(TT /2) = 9. Plot the antiderivative on your grapher . Show that it
contains the point (TT/ 2, 9).
Solution Your thought process should go like this.
The function differentiated must have been cosine .
But the derivative of cosine is the opposite of sine.
The argument of the cosine must have been 2x because the argument does
not change when you differentiate.
However, the derivative of - cos 2x is 2 sin 2x, which is twice what is needed .
Thus, the general equation must be
f(x) = - cos 2x + C.
To find the particular equation, substitute TT/ 2 for x an d 9 for f(x).
9 = - cos (2TT/'2.) + C
9=-(-l)+C
8.5 = C
7t 21t X
f(x) = - cos2x + 8.5
X = 1.570796 y =9 Figure 39 b shows the graph of this solution. If you use a friendly window with
a multiple of TT as the scale factor, you can trace to TT / 2 exactly and find that y
Figure 3-9b actually does equal 9.
Example 3 Galileo drops a cannonball from a leaning tower. Two secon ds later the cannonball
is at y = 35 m above the ground. Its velocity is given by dy / dt = - 9.8t . Find th e
particular equation for the antiderivative, expressing y as a function oft. Based on
the equation, how tall is the leaning tower 7 When does the cannon ball reach the
ground?
'
120 Chapter
3: Derivatives,
Antiderivatives,
andIndefiniteIntegrals
dy
Solution dt = - 9.8t
.. y = - 4.9t 2 + C Sec the beg innin g of th.is sect io n.
2
35 = - 4.9(2 ) +C Sub st itu te th e inHia l co nd ition y ; 35 wh en t ; 2.
54.6 = C
the particular equation is y = - 4.9t 2 + 54.6.
The cannonball was dropped when t = 0. At that tim e, y = 54.6, so the tower is
54.6 m tall.
The cannonball reaches the ground when y = 0.
112
0 = -4.9t 2
+ 54.6 ~ t = ( 54.6)
_ = 3.338 ... , about 3.3 sec
49
Problem Set 3-9
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten probl ems in less than five minut es.
QI. What theorem has a hypothesis that f(x ) is between g(x ) and h(x ) for all x close to c?
Q2. What is th e limit of (sin x ) / x as x approa ches zero ?
Q3. What is th e derivative of y = cos x?
Q4. If y ' = cos x, what function could y equal ?
QS. If g(x) = 2x, can g '(x) be found using th e power rul e for derivativ es?
Q6. Simplify: x 12 /x 4
Ql. Simplify: (x 12 ) 4
QB. Write log 32 in terms of log 2.
Q9. Sketch the graph of y = lx l.
QIO. What is the outside function in y = sin 3 x?
For Problems 1-16, find a function whose derivati ve is given. That is, write th e general
equation for the antiderivative.
l. f '(x ) = 7x 6 2. f'(x ) = 10x 9
3. f' (x ) = x 5 4. f' (x ) = x 4
5. f '( x ) = x - 9 6. f' (x ) = x- 1066
7. f '(x ) = cosx 8. f' (x ) = sinx
9. f '( x ) = 36x 712 10. f' (x ) = 77x 4 13
11. f '( x ) = sin 5x 12. f' (x ) = cos 4x
Sec
tion 3-9:Antiderivat
ivesand IndefiniteIntegrals 121
,~. 13. f'(x) = (4x + 5)7 (Be clever!)
t"
14. f'(x) = (8x + 3) 5
~ 7t
(Be clever!!)
i rt
-
15. f'(x)=x 2
+ 6x - 5 16. f '(x) = x 2 - lOx + 7
For Problems 17-22, find the particular function f(x) that has the given function f'(x) for
its derivative and contains the given point.
17. f'(x ) = x 4 and f(l) = 10 18. f'(x) = x 7 andf( - 1) = 100
19. f'(x ) = cosx and f(rr / 2) = 5 20. f'(x) = sinx and f(rr) = 8
21. f '(x ) = x2 - 8x + 3 and f( - 2) = 13 22. f'(x) = x 2 + 12x - 7 and f(3) = -10
23. Displacement Problem: Ann Archer shoots an arrow into the air. Let d(t) be its
displacement above the ground at time t seconds after she shoots it. From phys ics
she knows that the velocity is given by
d '( t) = 70 - 9.8t.
a. Write the general equation for d ( t).
b . Write the particular equation for d(t), using the fact that Ann is standing on a
platform that puts the bow 6 m above the ground when she shoots the arrow .
c. How far is the arrow above the ground when t = 5? When t = 6? When t = 9? How
do you explain the relationship among the three answers 7
d . When is the arrow at its highest? How high is it at that time?
24 . Acce lerat ion Problem: Calvin is swinging on the playground. His velocity, v(t) feet
per second, gets lar ger and smaller as he swings. His gir lfri end Phoebe tells him that
his acceleration, v '(t ) , is given by
v '(t) = 18 sin 3t,
where t is th e number of seconds since he reached a low point on a backswing.
a. Write the general equation for v(t).
b. Write the particular equation for v(t) if Calvin's velocity is - 6 ft / sec when t = 0.
How do you interpret this negative velocity?
c. Make a table of values of v(t) for each 0.2 sec from t = 0 through t = 2. Round to
the nearest 0.1 ft / sec.
d. What is the maximum velocity Calvin reaches? What is his acceleration at this
time7 Surprising?!
25 . Derivative and Antiderivative Problem : Let g '(x) = 0.6x .
a. Find the general equation for the antiderivative, g(x).
b . Find the particular equation for g(x) in each case .
i. g(O) = 0 ii. g(O) = 3 iii. g(O) = 5
c. Plot the graph of g '(x) and the three graphs for g (x ) on the same screen, then
sketch the results. Why are the three graphs of g(x ) called a family of functions?
122 Chapter
3: Derivatives,
Antiderivatives,
andIndefinite Integrals
3-10 Chapter Review and Test
In this chapter you have learned some algebraic techniques for differentiating
functions. You can differentiate powers, sums, linear combinations, sine and cosine,
and composite functions . You have learned formally about velocity and acceleration .
You have also been exposed to the fourth concept of calculus, the indefinite integral,
which is just an antiderivative.
The advantage of the algebraic techniques over the numerical and graphical ones is
the ability to get the exact values of a derivative fairly quickly. The disadvantage of
the algebraic techniques is that the way you calculate a derivative has nothing to do
with what derivative means. You must always bear in mind that a der ivative is an
instantaneous rate of change.
The Review Problems below are numbered according to the sections of this chapter .
The Concepts Problems allow you to apply your know ledge to new situations . The
Chapter Test is more like a normal classroom test your instructor might give you .
Review Problems
RO. Update your journal with things you've learned since your last entry . You should
include such things as those listed here .
The one most important thing you have learned in studying Chapter 3
Which boxes you have been working on in the "define, understand, do, apply" table
(See page 71.)
Key words, such as chain rule, difference quotient, sinusoid
Ways in which you have used graphs, tables, algebra, and writing to understand
concepts
Any ideas about calculus that you're still unclear about
Rl. Let f(x) = x 3 .
a. Calculate the numerical derivative off at x = 2.
b . Write an equation for m(x), the slope of the line connecting the points (x, x 3 )
and (2, 8). What form does m(2) take? What seems to be the limit of m(x) as x
approaches 2?
c. Use algebra to simp lify the expression for m(x) . You might recall how to factor
a difference of two cubes, a 3 - b 3 = (a - b)(a 2 +ab+ b 2 ). Then take the limit of
m(x) as x approaches 2.
d. Show that the result of Rlc gives, exactly, the value of the derivative in Rla .
R2. a. Write the definition of derivative (derivative at x = c form).
b. If f(x) = 0.4x 2 - x + 5, find f'(3) directly from the definition of derivative.
C. Plot the difference quotient in a neighborhood of 3. Sketch the result.
d. Plot on the same screen the graph of function f in R2b and a line through the
point (3, f(3)) with slope f'(3). Sketch the result .
e. What relationship does the line in R2d have to the graph?
f. Does function f seem to have the property of local linearity at x = 3?
How do you know?
Section3-10: Cha
pter ReviewandTest 123
R3. a. Plot y 1 = x 4 - 4x 3 - 7x 2 + 34x - 24. Use a window that goes from at least x = -4
to x = 5 and has a fairly large y-range. Sketch the result.
b. Plot y 2 = the numerica l derivative of Y1..Sketch the result.
c. What feature does the y 1 graph have where the derivative graph eq uals zero?
d. Leaky Tire Prob lem: The air pressure in a leaky tire decreases accord ing to th e
equation p(t) = 35(0 .9) 1 , where p(t) is pressure in po unds per square inch an d t
is time in hours since it was last inflated. Plot p(t) and its numer ical derivat ive on
your grapher. Then sketch the result. At approximately what rate is the pressure
changing when t = 3? When t = 67 When the tire was just filled? What are the
units of pressure change? How do you explain the sign of the pressure change? Is
the rate of change of pressure getting closer to zero as time goes on?
R4. a. Write the definition of derivative (the h or 6.x form).
b. What word means "find the derivat ive" 7
c. Write the property for the derivative of a power function .
d. Prove the property of the derivative of a constant times a func tion.
e. Prove the property of the derivative of a sum of two funct ions.
dy d
f. How do you pronounce dx 7 How do you pronounce dx (y)7 y
..j
What do these symbols mean?
g. Find an equation for the derivative function .
i. f(x) = 7x 9 15
x2
ii. g(x) = 7x - 4 - - x + 7
6
iii. h (x) = 7 3
h. Compare the exact value of f'(32) in R4g with the numer ical
derivative at that point.
i. Sketch the derivative function's graph for the function Figure 3- 1Oa
shown in Figure 3-lOa.
RS. a. If x is the displacement of a moving object from a fixed plane as a function of
time, t, write the appropriate calculus symbol for the velocity, v. Write two forms
of the symbol for the acceleration , a, one in terms of v and the oth er in terms of x.
b. Spaceship Problem: A spaceship is approaching Mars. It fires its retrorocket s,
causing it to slow down, stop, rise up again, then come back down. Its displace m ent ,
y kilometers, from the surface is found to be
y = - O.Olt 3 + 0.9t 2 - 25t + 250,
where t is time in seconds since the retrorocket was fired .
i. Write equations for the velocity and acceleration of the spaceship.
ii. Find the acceleration at time t = 15 sec. At that time is the spaceship spee ding
up or slowing clown? How do you tell?
iii. Find by direct calculation the values oft at which the spaceship is stopped .
iv. When does the spaceship touch the surface of Mars? What is its velocity at that
time? Explain what you think will happen to the spaceship at that time.
R6. a. Plot and sketch y 1 = cos x and y 2 = the numerica l derivative of y 1 .
b. Explain how the graphs show that cos'x = - sinx.
c. Verify numerically that sin' 1 = cos l.
124 Chapter 3: Derivatives,Antiderivatives, andInde
finite Integral
s
d. What name is given to a function such as f(x) = cos (x 2 ) for which a function such
as cosine is performed on an inside function, x 2 ? What does f '(x) equal ?
R7. a. State the chain rule in each form.
i. dy/dx
ii. f(x)
iii. Using outside function and inside function
b. Prove the chain rule. What assumption must be made about the quantity t..u so
that the proof is valid?
c. Differentiate f(x) = (x2 - 4) 3 two ways and show that the two answers are
equivalent.
i. Use the chain rule.
ii. Expand the binomial power first.
d. Differentiate.
i. f(x) = cos (x3 ) ii. g(x) = sin 5x
iii. h(x) = cos 6 x iv. k(x ) = sin3
e. Shark Problem: The weight of a great white
shark, W pounds, is given approximately
by W = 0.6x 3 , where x is the length of the
shark in feet. Suppose that a baby shark is
growing at the rate of dx / dt = 0.4 ft/ day.
How heavy is the shark when it is 2 ft
long? 10 ft long7 At what rat e is it gaining
weight (pounds per day) when it is these
lengths? Tell how the chain rul e allows you
to calculate these answers.
. sinx
RB. a. Evaluate: ltm --
x-o X
b. Make a table of values of (sin x) /x for each 0.01 unit of x from - 0.0 5 to 0.05. Show
from the table that (sinx) /x really does approach the limit in R8a as x approaches
zero.
c. State the squeeze theorem. Between what two quantiti es can (sinx) /x be
"squeezed" to prove that the limit is the number you wrote in R8a?
d. Prove directly from the definition of derivative that sin' x = cos x.
e. Given that sin ' x = cos x, use appropriate trigonometric properties to prove that
cos ' x = -sinx.
f. Clock Problem: A clock with a sweep second hand is fastened to the wall. As the
sweep hand turns, the distance from its tip to the floor varies sinusoidally with
time . The center of the clock is 180 cm from the floor, and the sweep hand is 20
cm long from the center to its tip . Write an equation for the distance from tip to
floor as a function of the number of seconds since the hand was at the 12. Write
an equation for the instantaneous rate of change of this distance with respect to
time. How fast is the distanc e changing when the sweep hand points at the 2? At
the 3? At the 7? How do you explain the signs you get for these rates ?
Section
3-10: Chapter
Review
andTest 125
t At.v
~
C
R9. a. If f'(x) = 36x5, find the general equation for f(x).
b. If dy / dx = sin0.2x, and y = 3 when x = 0, find the particular equation for y .
c. Distance Problem: Iona Carr gets into her automobile, which is 100 ft from her
house . She accelerates it in such a way that its velocity, v(t) feet per second, is
v(t) = 6t 112 _
Find the particular equation for y(t), her displacement from the house t seconds
after she started accelerating . How far from the house is she 1 min after she
starts? How fast is she going at that time 7
d. What is the difference between an antiderivative and an indefinite integral ?
Concepts Problems
Cl. Introduction to the Derivativ e of a Product: Let f (x ) = x 7 and g (x ) = x 9 . Let
h(x) = f(x) g(x).
a. Write an equation for h (x ) in as simpl e a form as possible.
b . Find h'(x) .
c. Find f'(x ) and g'(x ) . Does h'(x ) = f'(x) g'(x)?
d. Show that for these particular functions, h'(x ) = f'(x ) g(x ) + f(x ) g '(x ) .
. Let f (x ) = x - .sin 2x .
. Functwn:
C2 . Grap h of an I nterestmg
smx
a. What form does f(O) take? What nam e is given to a form like this? Explain why f
is discontinuous at x = 0.
b. The discontinuity inf at x = 0 is removable . What number should f (O) be defined
to equal so that th e function is continuous at x = O?
c. Make a conjecture about whether or not the function as defined in C2b is
differentiable at x = 0. If you think it is differentiabl e, make a conjecture about
what f'(O) equals. If you think it is not differentiable, tell why not .
d. Use the definition of derivative (derivative at x = c form) to establish wheth er or
not your conjecture in C2c is true.
Chapter Test
Tl. Write the formal definition of limit.
T2. Write the definition of derivative (either form).
T3. Prove directly from the definition of derivative that the derivative of f (x ) = 3x4
is 12x 3 .
T4. Is the function t (x ) = cos 3x incr easing or decreasing when x = 5? At what rate?
TS. Sketch the graph of a function for which f'(S ) = 2. Sketch a line tangent to the graph
at that point . What would happen to the graph and th e tangent line if you were to
zoom in on the point where x = 5? What is the name of the property that expresses
this relationship between the graph and the tangent line7
126 Chapter
3: Derivatives,Antiderivatives,andIndefinite Integra
ls
T6. Amos Take must evaluate f'(5) where f(x) = 7x. He substitutes 5 for x, getting
f (5) = 35. Then he differentiates the 35 and gets f'(5 ) = 0 (which equals the score
his instructor gives him for the problem!) . What mistake did Amos make? What is
the correct answer?
T7. Plot the graph of f(x) = (sinx) /x. Then sketch the resu lt. Show in particular what
happens to f(x) as x approaches zero. In the proof of the limit of this function as
x approaches zero, the function is bounded below by y = cos x and bounded above
by y = sec x . Name and state (without proof) the theorem used in this proof.
For Problems T8-Tl2, find an equation for the derivative function .
TS. f(x) = (7x + 3) 15 T9. g(x ) = cos (x 5 ) y
. 4 ..
TlO. :x (sin 5x) Tll . y = 6Qx2 13 - X + 25
Tl2 . f(x) = cos (sin 5 7x) X
Tl3 . Estimate the value of the derivative, y', at x = l for the graph
shown in Figure 3-1Ob.
Tl4. Find velocity and acceleration equations.
y = 3+ 5[ 16
Figure 3-1Ob
Tl 5. Find an equation for the antiderivative: f'(x) = 72x 51.;.
Tl6. Find an equation for f(x) if f'(x) = 5sinx and f(O) = 13.
Tl 7. Carbon Diox ide Prob lem: The concentration of carbon dioxide in the atmosphere
fluctuates slightly. The concentration is at a minimum in the summer when tree
leaves absorb the carbon dioxide and at a maximum in the winter when trees are
bare. The fluctuation is only about 2 ppm (parts per million) with an average
concentration of 300 ppm. Assuming that the concentration varies sinusoidally with
time, t days since the first of the year, the concentration, c(t ) , is
2rr
c(t ) = 300 + 2 cos t,
365
whose graph is shown in Figure 3-1Oc.
ppm
a. Write an equation for c'(t) .
b. The greatest rate of increase occurs at t = 273, which is
S = Summer
at the autumnal equinox. What is this rate in parts per W = Winter
million per day? 100
d
c. The earth's atmosphere contains a total of about 6 x 10 15
tons (t) of air. Show that the amount of carbon dioxide in w s w s w
the atmosphere is increasing by about 2390 tons / sec at Figure 3-1Oc
t = 273. Surprising, isn't it!?
Section
3-10: Chapter
Review
andTest 127
CHAPTER
4 Products, Quotients,
and Parametric Functions
Unlike the pendulum in a grandfather clock, this Foucault pendulum
is free to swing in any direction. The path of the pendulum is as
constant as possible with respect to space, and thus seems to move
with respect to the earth as the earth rotates underneath it. Parametric
functions are used as mathematical models of the path of an object
whose x- and y-coordinates both vary with time.
129
1~1 Mathematical Overview
You have already learned a way to differentiate power functions .
In Chapter 4 you will learn algebraic methods for differentiating
products and quotients. You will also apply these methods to
parametric functions in which both x and y depend on some third
variable such as time . You will gain this knowledge in four ways.
Graphically The logo at the top of each even- y( t)
numbered page of this chapter shows
the path followed by a pendulum Slop e= dy
_ dx
that is allowed to swing in both --
the x- and y-directions.
Numerically t X y
0 7 4 x( t)
1 5.620 ... 5.501 ...
2 1.030 .. . 4.176 .. .
3 2.039 ... 3.053 ...
dy dy/ dt . .
Algebraically dx = dx/ dt , the parametric cham rule.
Verbally I had my first major surprise in calculus! The derivative of a product
of two functions is not the product of the two derivatives. The same
thing goes for quotients. I also proved that the derivative property
for power functions works when the exponent is negative and not
an integer.
130
4-1 Combinations of Two Functions
You have learn ed to differentiat e algebra ically a funct ion that is a linear combination
of two oth er functions, su ch as f (x) = 3 cos x + 2 sin x. In this section you will explor e
derivat ives of a product , a quotient , an d a composition of two other functio ns. For
exam pl e,
= 3 COSX X = 3 COS t
p (x ) = (3 cosx )(2 sinx ), q (x ) . , and
2 smx y = 2 sint
OBJECTIVE Explore on your ovm or with your study group the derivatives of functions formed by
multiplying, dividing, and composing rwo functions.
Exploratory Problem Set 4 1
Let f (x ) = 3 cos x and let g (x ) = 2 sin x, as shown in Figures 4-la and 4-1b.
1. Find equation s for f ' (x) an d g' (x ).
2. D er iva tiveofaProd ucto fTwoF unct ions:P l otp (x ) = f (x ) g (x) =
6 f(x ) = 3cosx
(3 cosx)(2 sin x) . Use a frie ndl y window th at is about th e size
shown in Figur es 4-la and 4-lb. Sketch the res ul t. Find p'( 2)
num erically . How do you te ll whether p (x ) is increasi ng or
decreasi ng at x = 2? Does p '(2) equal f '(2) g'(2)?
3. D eri vativ e of a Quoti ent of Two Functions: Plot q (x ) = f(x )/ g(x) =
(3 cos x)/ (2 sin x). Sketch the resu lt. What faini liar funct ion is q?
Find q'(2) num erically. Is q (x ) increas ing or decreasing at x = 2? Figure 4-1a
Does q'( 2) equ al f' (2)/g'( 2)?
4. Parame tri c Fun cti on: Let x = 3 cost an d y = 2 sin t . If you draw g(x) = 2sinx
6
an xy -grap h for var iou s values of t, the result is called a
p arametr ic funct ion . The ellipt ical path traced on t he floor by
X
th e pen dulum pictur ed on page 129 is an examp le of su ch a .. ~.....,. ~ ,~ .......... ........ ~ .,,...... ..... ..1....
10
function . Put your grap her in paramet ri c mo de. Enter x = 3 cos t
and y = 2 sin t. Use a window with x - and y-ranges like tho se
shown in Figure 4-la. Use a window of O ::; t ::; 2rr , with a t-st ep
of 0. 1. Plot an d sketch the grap h. Does the grap h rea lly seem to
Fig ure 4-1b
b e an ellipse?
5. Der ivat ive of a Para metri c Fun ct ion: Trace the grap h in Probl em 4 to th e point where
t = 2. Find L'.x and L'.y as t goes from 1.9 to 2.1 an d u se the res ul ts to find a symmetr ic
difference qu otient that approximates dy /dx at t = 2. Give evidence to show that dy /dx
at thi s point could equa l (dy /dt )/( dx /d t ) given that the derivat ives are tak en at t = 2.
6. Conjec tur es? Ju st for fun, see if you can calculate the correct valu es of p '( 2) an d q'(2)
from Probl em s 2 and 3, using th e values of f' (2) , g '( 2), f (2), and g(2 ).
Section 4-1: Comb
inations of Two Fun
ctions 131
)
8 .
)
4-2 Derivative of a Product of Two Functions
The derivative of a sum of two funct ions is equal to the sum of the two derivat ives.
As you learned in Exploratory Problem Set 4-1, there is no such distrib u tive property
for the derivative of a product of two func tions. Here is an example.
Derivative of the prod uct f (x) = (x5 )(x 8 ) Product of the der ivatives
f (x) = x 13 (5x 4 )(8x 7
)
f'(x) = 13x 12 Not the same. = 40x 11
In this sect ion you will learn how to differe nt iate a product of two functions without
having to mu ltip ly first.
OBJECTIVE Given a fun ction that is a product of two other fun ction s, find, in one st ep, an equ ation
for the derivati ve function.
What follows involves some fairly comp licate d algebra. Therefore, it helps to
streamli n e the symbols somewhat. Instead of writing f(x) = g(x) h (x), writ e
y = UV,
where u and v stand for differentiab le funct ions of x. The id ea will be to find dy /dx
in terms of du /dx and dv /dx.
u V y
y+t,.y =
(u+t,.u)(v+t,.v)
u +t,.u
I
x
'I
x +t,.x
X
t
X
x
[
x+t,.x
..X
u = one factor v = other factor Y= UV
Figure 4-2a
The graphs of two funct ions u and v and their product y = uv are shown in Figure
4-2a. By the definition of derivative,
dy = lim (y + 6y) - Y
dx -'>x- o 6X
Becau se f(x ) = g(x) h (x), f(x + 6x) is equal to g(x + 6x) h (x + 6x). Thu s
(y + 6y ) in the equation above can be replace d by ( u + 6 u)(v + 6v ), as incticated
on the right side in Figure 4-2a, and y can be rep laced by uv . Therefore,
dy = lim ( u + 6 u)( v + 6v ) - uv
dx -'>x- 0 6X
132 Chapter 4: Products, Quoti
ents, andParametri
c Func
tions
dy = lim uv + 6uv + u6v + 6u6v - uv
6u v means (6u ) v .
dx t.x-0 6x
. 6uv + u6v + 6u6v
= 11m --------
t.x - o 6x
= 1,Lill (6uv
--
u6v + --
+ -- 6u6v)
t,x - 0 6x 6x 6x
= hm. (6u 6v
- v+u -+ 6u -6v) .
t.x -O 6x 6x 6x
Using the limit of a sum and the limit of a product properties, and using the fact
that the limits of 6u /6x and 6v /6x are du /dx and dv /dx, respectively, gives
dy = du V + u dv + Odv Because u is a continuous function, t.u - Oas
dx dx dx dx t.x - 0.
du dv
= - v+u- Derivative of a produ ct formula.
dx dx
Short form, where u' and v ' are derivatives ~th
y' = u'v + uv ' respect to x .
The formula is best rememb ered as a procedure, as shown in the box.
Property:Derivativeof a Productof TwoFunctions
If y = uv, where u and v are differentiable functions of x, then y ' = u'v + uv '.
Wor ds: Derivative of first times second, plus first times derivative of second.
With this pattern you can accomplish the objective of differ entiating a product in
one step.
Example 1 If y = x.Jcos 6x, find dy /dx.
Solution !~ = 4x 3 cos 6x + x.J( - sin 6x) 6
= 4x 3 cos 6x - 6x 4 sin 6x
As you write the derivative, say to yourself, "Derivative of first times second, plus
first times derivative of second." Don't forget the chain rule'
Example 2 If y = (3x - 8)7 (4x + 9)5, find y ' and simplify.
Solution y' = 7(3x - 8) 6 (3)(4x + 9) 5 + (3x - 8)7(5)(4x + 9).J(4)
The product rule and the chain rule have been used. Each term has common
binomial factors. Simplification includes factoring out these common factors.
y' = (3x - 8) 6 (4x + 9) 4 [21(4x + 9) + 20(3x - 8)]
= (3x - 8) 6 (4x + 9).J(]44x + 29)
Note that the factored form in Example 2 is considered simpler because it is easier
to find values of x that make the derivative equal zero.
Section4-2:Derivative
of a Product
ofTwoFunctions 133
)
19 .
)
Problem Set 4-2
DoTheseQuickly
The following prob lems are intended to refresh your skills. You should be able to do all
ten prob lems in less than five minutes .
01. Differentiate: y = x 314
02. Find y ' : y = l 7x.
03. Find dy / dx : y = (Sx - 7) - 6 .
04. Find ddx(sin2x).
as. Differentiate: = cos 3 t
V
06. Differentiate: L = m 2 + Sm+ 11
Ql. If dy /dx = cos x 3 3x 2 , find y .
QB. In Figure 4-2b, if x = - 2, y ' ""'-?-
Figure 4-2b
Q9. Sketch the graph y = cos x.
010. If u = v 2 /6, where u is in feet and vis in seconds, how fast is u changing when v = 12?
For Problems 1-22, differentiate and simplify. You may check your answer by comparing
its graph with the numerica l derivative graph .
1. f (x) = x 3 cosx 2. f (x ) = x 4 sinx
3. Find g '( x) : g(x) = xL5 sin 7x. 4. Find h' (x ) : h(x ) = x - 6 3 cos lOx.
5. Find!: : y = x 7 ( 2x + 5) 10 . 6. Find !y : y = x 8 (3x + 7)9 .
X
7. Find z ' : z = x4 cos 5 3x. 8. Find v ' : v = x 5 sin 3 6x.
9. Find dx[ (2
d _4x - 3 ~ 'I
) ~5-:.J. 10. Find :x (p ): p = ( 7x - 4)9 cos 2x.
11. y = (6x + 11)4 (5x - 9) 7 12. y = ( 7x - 3)9 (6x - 1)5
13. p = (x2 _ l )1o(x2 + l) 1s 14. P (x ) = (x 3 + 6)4 (x 3 + 4)6
15. a(t) = 4sin3tcos5t 16. v = 7cos2ts in6t
17. y = 10 cos 8 5x sin 5 8x 18. y = 7 sin 3 4x cos 4 3x
19. z = x 3 ( 5x - 2)4 sin6x (Be clever!) 20. u = 3x 5 (x 2 - 4 ) cos lOx (Be clever!!)
21. y = cos (3 sinx) 22. y = sin(Scosx )
23. Product of Three Functions Prob lem: Prove that if y = uvw, where u, v, and w are
differentiable functions of x , then y ' = u' vw + uv ' w + uvw ' .
24. Product of n Functions Conjecture Prob lem: lf y = u 1 u2 u3 . .. Un , wher e u 1 , , Un are
differentiable funct ions of x, make a conjecture about what an equation for y ' would be.
134 Chapter
4 Products,
Quotients,andParametric Function
s
For Problems 25- 28, differentiat e and simplif y.
25. z = x 5 cos6 x sin 7x 26. y = 4x 6 sin 3 x cos 5x
27. y = x 4 (2x - 3) 5 sinxcos 2x 28. u = x 5 (3x - 1)2cos 2x sin 3x
29. Odd Function and Even Function Derivativ e Probl em: A function is called an odd
function if it has the property f( - x ) = - f (x ) . Similarly, f is called an eve n function
if f (- x ) = f (x ) . For instanc e, sin e is odd because sin(-x ) = - sinx, and cosine is even
because cos ( - x ) = cos x . Use th e chain rule appropriately to prove that the derivative
of an odd function is an even function and that the deri vative of an even function is
an odd function .
30. Double A rgum ent Properti es Prob lem: Let f (x ) = 2 sinx cos x and let g (x ) = sin 2x. Find
f'(x ) and g '(x) . Use the appropri ate trigonom etric prop erties to show that f' (x) an d
g'( x ) ar e equival ent. Then show that f (x ) and g (x ) ar e also equiva lent . Do the same
for th e functions f (x ) = cos 2 x - sin 2 x and g (x ) = cos 2x.
31. Der ivativ e of a Power Induction Prob lem: Prove by mathematical induction that for any
positi ve integer n, if f (x ) = x n, th en f '(x) = nx n- 1 .
32. Derivativ e Two Ways Probl em: The function y = (x + 3) 8 (x - 4)8 can be diff erentiated
by two methods: First, consider the function as a product of two composite functions;
second, multipl y and diff erentiat e the result , y = (x 2 - x - 12) 8 . Show that both
methods give th e same result for th e derivative.
33. Confirmation of th e Produ ct Prop erty: Let f (x ) = x 3 sinx . (See
Figure 4-2c.)
a. Sketch this graph, th en draw what you think th e deri vative
X
graph, f', would look like . Show, especiall y, plac es where f' (x )
would equal zero.
b . Write an equation for f'( x ). Plot f and f' on your graph er . How
do es the graph you pr edicted compare vvith the actual graph ?
c. Plot th e num erical derivati ve on your graph er . How do es this Figure 4-2c
graph confirm that your equ ation for f' (x ) is correct ?
34. Repeated Roots Problem: In this problem you will sketch th e graph of the function
f (x ) = (5x - 7)4 (2x + 3) 5 .
a. Plot on your graph er th e graph off. Use an x-windo w from -2 to 2 and a y- window
that is tall enough to fit th e graph . Sketch the result.
b. Find f '(x ). Simplify . For exampl e, facto r out any common factors .
c. Find all thre e values of x for which f '( x ) = 0. The factor ed form of f' (x ) from 34b
should be conveni ent for this purpos e.
d. Find f(x) for each of th e valu es of x in 34c. Show th ese thr ee points on your graph.
e. The graph of f should have a horizontal tangent line at each of th e points in 34d .
State wheth er the following stat ement is true or false: The graph ha s a high point
or a low point where f' (x) = 0.
Section 4-2:Derivative
ofa Product
ofTwoFunctions 13 5
35. Pole Dance Prob lem: In a variation of a Filipino pole dance, two pairs of bamboo poles
are moved back and forth at floor level. The dancer steps into and out of the region
between the poles (Figure 4-2d), trying to avoid being pinched between them as they
come together. The area, A= LW, varies with time, t, where distances are in feet and
time is in seconds.
r
Poles
Area
varies
t
w
'
--L-
Figure 4-2d
a. Write dA /dt in terms of L, W, dL Jdt, and dW /dt .
b. Suppose that W = 2 + 2 cost and that L = 3 + 2 sin 2t . At what rate is the area
changing when t = 4 sec? When t = 5 sec? At these times is the area increasing or
decreasing?
4-3 Derivative of a Quotient of Two Functions
In this section you will learn how to find the derivative of a quotient of two functions .
For example,
f(x ) = Bx
sin5x
- 3
OBJECTIVE Given a function whose equation contains a quotient of two other functions, find an
equation for the derivative function in one step and simplify the answer.
136 Chapter
4: Products,
Quotients,
andParametric
Functions
Suppose that f(x) = g(x) /h (x) . Using y, u, and v for th e three function values lets
you write the equation in a simp le form.
u
y = v
By the definition of derivativ e,
u + t.u u)
-dy = 11m -t, y = 11m v+t.v
-----
- v .
dx t.x - 0 t.x t.x - 0 ( t.x
Multiplying the numerator and th e denomin ato r by (v + t.v )(v) eliminates the
complex fraction.
dy = lim ( (u + t.u)v - u (v + t.v))
dx ~x -o t.x (v + t.v) (v)
Further algebra allows you to take th e limit.
. (uv + t.uv - uv - ut.v)
-dy - 11m
dx - t.x - o t.x (v + t.v)(v)
= lim ( 1 . t.uv - ut. v) Simplify , then asso ciate !'!.x with
t.x - o (v + t.v)(v) t.x the num erator.
1
= lim [
t.x - o (v+t.v)(v)
( t.u . v - u . t, v)
t.x t.x
J Distribut e !'!.x, then associate it
with !'!.u and !'!.v.
Limits of products, quotient, and
= l_ . ( du . v _ u . dv) sum; definition of derivative ;
v2 dx dx !'!.v - 0 as !'!.x - 0.
u' v - uv ' Where u' and v ' are derivatives
y ' = - 1 ( UV' - UV ') =
v2 v2 with respect to x.
Property: Derivative of a Quotientof Two Functions
If y = !!., where u and v are differentiable functions , and v =1=0, then
V
u' v - uv '
y '=-- -
v2
Words: Derivativ e of top times bottom, minus top times derivative of bottom, all divided
by bottom squared.
Note that the num era tor has th e same pattern as th e derivative of a product, nam ely
u' v + uv ' , except that a subtraction sign (- ) is us ed instead of an addition sign ( +).
Example 1 D1'ff erent1ate:
. f(x ) = Bx
sin_5x
3
Solution f'(x) = cos 5x (5)( 8x - 3) - sin 5x (8)
(Bx - 3) 2
f'(x) = 5 cos 5x(8x - 3) - 8 sin 5x
(Bx - 3) 2
As you diff erentiat e, say to your self, "Derivative of top times bottom, minus top
times deri vat ive of bottom, all divided by bottom squared ." You must also, of course,
apply the chain rul e when necessary .
Section4-3:Derivative
of a Quotientof TwoFunctions 137
(5x - 2) 7
Example 2 Differentiate : y = ---
( 4x 3 + 9)
Solution , 7(5x - 2) 6 (5) (4x + 9) 3 - (5x - 2) 7 3(4x + 9) 2 (4 )
y = (4x+9) 6
(5x - 2) 6 (4x + 9)2[35(4x + 9) - 12 (5x - 2)]
Factor the num erator.
(4x +9) 6
(5x - 2) 6 (80x + 339)
Cancel commo n factors.
(4x + 9) 4
Example 3 Differentiate: :x( ~ )
7 3
Solution Although the quoti ent rul e can b e used here, it is simpler to tra nsform to a
power.
d ( 5 )
dx 7x 3 = dx
d (5
?x
-3) = -7x15 -4
Problem Set 4-3
DoTheseQuickly
The following problem s are int ended to refresh your skills. You shou ld b e able to do all
ten problems in less than five minutes.
QI. Find :x
(x 1066 ). Y .
.., 4 .... ,...
Q2. Antidifferentiate: f'(x) = 60 x 4
Q3. Find y': y = x 3 sin x. .::::::T_r
Q4. Find !: y = cos(x 7) . X
QS. Differentiate: f(x) = (3 5)(28 )
Q6. Find a'(t): a(t) = sin 9 t.
:f!E .... ' 4
Ql. Write th e definition of derivative. 1+!H~
..
QB. Write the ph ysica l meaning of derivative.
Figure 4-3a
Q9. Factor: (x - 3)5 + (x - 3)4 (2x)
Q10. Sketch th e graph of th e derivative of the function grap hed in Figure 4-3a.
For Problems 1-26, differentiat e and sim plif y. You ma y check your an swer by comparing
its graph with the numerical derivative graph .
x3 x4
1. f(x) = - . - 2. f(x) = -
smx cosx
3
3_ g(x) = COS X 4. h (x) = sin sx
xs x3
sin lOx cos 12x
5 Y = cos20x 6. y = ---
sin 18x
138 Chapter
4: Products,Quotient
s, and Parametric
Functions
I 'f _ 3x - 7 , .f _ lOx + 9
7 y 1 y - 6x + 5 8. y 1 y - 5x - 3
. dz . (Bx + 1) 6
9. Fmd dx 1f z = ( 5x _ 2 ) 9 . 10. Find dA if A = (4x - 1)47 .
dx (7x + 2)
2 2
. d , .f _ 4x + Bx + 1
11. f' d P' 1.f p = 5x 2 - lOx + 3 12. Fm r 1 r - 4x 2 - Bx + 3 .
m 3x + 6x - 8
13. Find :x (60x - 413) . 14. Find dd (24x - 713).
X
12 51
15 . r(x) = 3 (Be clever!) 16. t(x) = 17 (Be clever!)
X X
14 20
17. v(x) = --- 18. a(x) = - .- 2-
cos 0.5x sm x
1 1
19. e(x) = - 20. s(x) = 2
X X
0 1
21. Find W' (x) if W(x) = (x 3 : 22. Find T'(x) if T(x) = - - --
cosx sinx
23. T(x) = sinx 24 . C(x) = cosx
cosx smx
1 1
25. Find C' if C = -.- - . 26. Find S' if S = - -.
smx cosx
27 . Black Hole Prob lem: Ann Astronaut's spaceship gets trapped in the gravitational field
of a black hole! Her velocity, v(t) miles per hour, is given by:
_ 1000
( )
vt-3 - t'
where t is time in hours .
a. How fast is she going when t = 1? When t = 2? When t = 3?
b. Recall that acceleration is the instantaneous rate of change of velocity . Write an
equation for the acceleration function, a(t) .
c. What is her acceleration when t = 1? When t = 2? When t = 3? What are the units
of acceleration in this problem?
d. Using the same screen, plot graphs of velocity and of acceleration as functions of
time. Sketch the results.
e. Ann is in danger if the acceleration exceeds 500 (mi/hr)/hr . For what range of times
is Ann's acceleration below the danger point?
28 . Catch-Up Rate Problem: Willie Ketchup is out for his morning walk . He sees Betty Wont
walking ahead of him and decides to catch up to her. He quickly figures that she is
walking at a rate of 5 ft / sec. He lets x ft / sec stand for the rate he will go .
a. Explain why Willie catches up at a rate of (x - 5) ft / sec.
b. Betty is 300 ft ahead of Willie when he first starts catching up. Recall that distance
= rate x time and write an equation for t(x), the number of seconds it will take him
to catch up to her.
c. How long will it take Willie to catch up if he walks 6 ft/sec? 8 ft / sec? 10 ft / sec?
5 ft / sec? 4 ft / sec 7 5.1 ft / sec? What is a reasonable domain for x?
Section
43: Derivative
of a Quotient
of TwoFunctions 139
d. At what instantaneous rate does Willie's catch-up time change with respect to speed
if he is going 6 ft / sec? What are the units of this rate?
e. Explain why t (x) has no value for the derivative at x = 5.
29. Confirmation of Quotient Formula Problem: Find f'(x ) for the function
f(x) = 3x + 7 .
2x + 5
Use the answer to evaluate f' (4). Use x = 4.1, 4.01, and 4.001 to show that the difference
quotient [f(x) - f(4)] / (x - 4) gets closer and closer to f '( 4 ) .
30. Derivative Graph and Table Problem:
a. For the function f in Figure 4-3b, sketch what you think the
f(x)
graph of the derivative, f', looks like.
b. The equation for f(x ) is
x2 8 -
f(x) = -- . X
x- 3
Make a table of values of f(x) and f'(x) for each 0.01 unit of x
from 2.95 to 3.05.
c. Plot on your grapher the graph of f' . In what ways does your Figure 4-3b
sketch in 30a differ from the actual derivative graph?
d. Based on your graphs and table, describe the way f (x) changes as x approaches 3.
e. Find the value of x between 1 and 3 where f(x) stops increasing and starts
decreasing.
f. Find the value of x between 3 and 5 where f(x) stops decreasing and starts
increasing .
g. What are the domain and the range of f7 What are the domain and the range of f'7
31. Proof of the Power Rule for Negative Exponents: The proof you have used for the power
rule for derivatives assumes that the exponent is a positive integer. You have seen by
example that the rule also works for some powers with negative exponents. Suppose
that y = x- 5 . Use the quotient rule of this section and write y as
1
y = xs
to prove that y ' = 5x- 6 . Prove that, in general, if y = xn, where n is a negative constant,
then y' = nxn- t. To do this, it helps to write y as
1
y = XP'
where pis a positive number equal to the opposite of n .
* 32. Figures 4-3c and 4-3d show the graphs of y = sec x and y = tanx, respectively. Using a
photocopy of each graph, sketch what you think the derivative graph looks like. Check
your answers by grapher, using the numerical derivative feature.
*This problem prepares you for the next section .
14 0 Chapter
4: Products,
Quotients,
andParametric
Funct
ions
y
y ; I:
: ' .
' :
: I -~ I ;. .. --~-.. iI ; ...
:
: I
"j "
.}I i,
:
I
t
: I :
i" , ..... ,. ":,
' : 1
.... r-
:
: ;
I :
;,
.....
:, !1'..
I
I
.i
I
--: t ~
I
.. 1 : ~ ... :. : ....
-~ '
,.
I :
I
: I :
..:.(
I I
.. -~.. I ..
I
-~:. I I ,
I
I: X
: I I: X
I I
I 1 I I . I I
,
I ... ; . 1 I ~-
::
:..L.l.i/T\ .L
..
I :
. I ! I ;
...;- ,- . ~-- ---~--- i':"...
I
'I . I . : f
...f..~. !...f./:i
...... \ f\ ,,
: I
""(: :....
Figure 4-3c Figure 4-3d
33. Journal Prob lem: Update your journal with what you've learned since the last entry.
Include such things as those listed here.
The one most important thing you have learned since your last journal entry
The primary difference between how to differentiate a sum of two functions and
how to differentiate a product or a quotient of two functions
The meaning of a parametric function
What you better understand about the meaning of derivative
Any technique or idea about derivatives that is still unclear in your mind
4-4 Derivatives of the Other Trigonometric
Functions
Recall from trigonometry that the tangent, cotangent, secant, and cosecant functions
can be written in terms of sine and cosine.
sinx cosx 1 1
tanx = -- , cotx = -. - , secx = - -, cscx = - .-
cos x smx cos x smx
Each of the above is a quo tient. Now that you can differentiate quotients, you can
differentiate the other four trigonometric functions.
OBJECTIVE Given a function whose equation contains any of the six trigonometric functions, find the
equation for the derivative function in one step.
Derivative
of Tangent
andCotangent
Functions
sinx
y = tanx = --
cosx
, (cosx)(casx ) - (sinx )(- sinx ) cos 2 x + sin 2 x 1
:. Y = cos 2 x cos 2 x cos 2 x
= sec2 x Derivative of tang ent.
Section4-4:Derivatives
of theOtherTrigonometric
Funct
ions 141
In Problem 3 7 of Problem Set 4-4, you vv:illshow that if y = cot x, then y ' = - csc 2x.
Derivative
of Secantand(osecantFunctions
1
y = secx = --
cosx
(O)(cosx) - (1)( - sinx) sin x 1 sinx
.. y ' = --------- = -- = -- . --
cos2x cos 2x cos x cos x
=, secx tan x Der ivative of secant.
In Problem 38 of Problem Set 4-4, you will show that if y = csc x, then
y ' = -csc x cot x.
Properties: Derivativesof the Six Trigonometric
Functions
sin' x = cosx cos' x = - sinx
tan' x = sec 2 x cot'x = - csc 2 x
sec'x = sec x tanx csc' x = -csc xcotx
Note: x must be in radians since tl:tis was assumed for the sine derivative .
Memory Aids :
a. The derivatives of the "co-" functions have a negative sign (- ) .
b. To find the derivatives of the "co-" functions in the right column, replace each function
in th e left column with its cofunction (e.g., sec with csc).
Example 1 Differentiate: y = 3 tan 5 7x
Solution y' = 3(5tan 4 7x)(sec 2 7x)(7)
= 105 tan 4 7x sec 27x
Note that this example involves two applications of th e chain rule . The outermost
function is the fifth pow er function. This fact is eas ier to see if you write the original
function as
y = 3(tan 7x) 5 .
The next function in is tan . Its derivative is sec 2 . The innermost function is 7x; its
derivat ive is 7. You should begin to see the "chain" of derivatives that gives the
chain rule its nam e.
y ' = 3(5 tan 4 7x) . (sec 27x) (7)
Here are th e three "links" in th e chain.
Problem Set 4-4 gives you practic e differentiating all six trigonometric functions.
142 Chapter
4: Product
s, Quotients,
andParame
tric Function
s
Problem Set 4.4
DoThese Quickly
The following problems are intended to refresh your skills. You should be
able to do all ten problems in less than five minutes.
QI. (sinx) / (tanx) = -?- y
Q2. 1/ (secx ) = - ?-
Q3. cos 2 3 + sin 2 3 = - ?- I
I
:n:
I
Q4. Differentiate: f(x ) = x sinx I
' X
QS. Differentiate : g (x ) = x/ (cosx) I 7l I
Q6. Differentiate: h(x) = (3x) - 517 I I
I I
I I
Ql. Find dy /dx: y = (cosx) - 3 I I
I I
I I
. Find limx_2 (x2
QB - 7x + 10)/ (x - 2). I
I
I
I
Q9, lim t,p- o(t.j /6.p ) = - ?-
Figure 4-40
. Sketch the graph of the derivative for the function graphed in Figure 4-4a .
QJO
For Problems 1- 36, differentiate and do obvious simplification. You may check each
answer by comparing its graph with the numerical derivative graph.
1. f(x) = tan 5x 2. f(x) = sec 3x
3. y = sec 7x 4. z = tan9x
5. Find g ' (x): g(x) = cot llx . 6. Find h' (x): h(x) = csc lOx.
7. r (x ) = csc 20x 8. p (x ) = cot3lx
9. Find :x (y): y = tan 5 4x . 10. Find :x (y): y = tan 7 9x.
d -
11. Find :x (3 cot 6 8x). 12. Find d X (5 sec 1 9x .) .
13. Find y ' : y = 8 sec 514 4x. 14. Find y ' : y = 88csc 11' 8 2x.
15. Find v' : v = csc (x - 7 ) . 16. Find u' : u = cot (3x- 5 ) .
17. Find!~: p = secxtanx. 18. Fm d d:
dm m = cscxcotx.
X -
19. y = x- 3 cotx 20. y = x 512secx
21. y = secx cscx 22. y = tanx cotx
tanx cotx
23. y = sinx 24. y = cosx
5x7 4 csc lOx
25. Y = cot 14x 26. y = x~O
27. w = tan(sin 3x) 28. t = sec(cos4x)
29. S (x ) = sec 2 x - tan 2 x 30. m (x) = cot 2 x - csc 2 x
Section
4-4:Derivatives
of theOtherTrigonometric
Functions 143
i-A 4MM1Ai21#5&4ii ti
31. A(x) = sinx 2 32. f(x) = cos x 3
33. F(x) = sin 2 x 34. g(x) = cos 3 x
35. C(x) = sin (sinx) 36. h(x) = cos (cosx)
37. Derivative of Cotangent Problem: Derive the formu la for y ' if y = cotx. You may write
cotx either as (cosx) / (sinx ) or as 1/ (tanx) .
38. Derivative of Cosecant Problem: Derive the formula for y ' if y = csc x by first trans-
forming csc x into a function you already know how to differentiate.
39. Confirmat ion of Tangent Derivative Formula: Figure 4-4b shows the graph of f(x) =
tanx as it might appear on your grapher.
a. Sketch the graph . Without using the grapher, sketch what you
think the derivative graph, f', would look like. 'I
I
b. Find an equation for f'(x). Plot the f and f' graphs on the same IX
screen. Did the f' you predicted look like the actual one? If
not, write down where your thinking went astray and what you
learned from this problem.
c. Calculate the symmetric difference quotient for f'(l ), using
Figure 4-4b
6.x = 0.01. How close is the answer to the actual value of f'(l )?
40. Confirmation of Secant Derivative Formula: Figure 4-4c shows an f(x) .
accurate graph of f(x) = sec x in the closed interval [- rr/2, rr/2].
I
a. Write the formula for f'(x ). Use it to find the value of f(l ) . + : ..
I
, .:.: .....:..
:
I I :
b. Photocopy or trace the graph onto your own graph paper. At I
I
I
the point (1, sec 1), draw a line with slope f'(l) . I
I
r . . ....~-.
c. How does your drawing confirm that the derivative formula I
I
for secant gives the correct answer at x = 1? .!I , ,
I
ct. Plot on the same screen the graphs off and f'. Copy the f ' I
I
I
graph onto the paper you used in 40b . I
X
e. What is happening to the f graph at values of x where the f ' : -1
graph is negative ?
Figure 4-4c
41. Light on the Monument Problem: Suppose you stand 10 ft away
from the base of the Washington Monument and shine a flashlight
at it (Figure 4-4d). Let x be the number of radians the light beam
makes with a horizontal line . Let y ft be the vertical distance from
the flashlight to the spot of light on the wall.
a. Show that y = 10 tanx. 53 5 ft
b . As you rotate the flashlight upward, at what rate is y increasing
with respect to x when x = 1? What are the units of this rate?
What is this rate in feet per degree?
c. At what rate does y change when your light points at the top of
y
the vertical monument wall, where y = 535 Jt ?
-1 10'1-
Figure 4-4d
144 Chapter
4: Prod
ucts,Quotients, and ParametricFunction
s
42. Point of Light Problem: As the beacon light at an airport rotates, it casts a point of light
onto objects it encounters. Suppose that a building is located northeast of the beacon
and that the building's north-south wall is 500 ft east of the beacon (Figure 4-4e). Let
x be the angle (in radians) the light beam makes with the east-west line. Let y be the
number of feet north of this line where the point of light is shining when the angle
is X.
a. Show that y = 500 tanx .
b . Let t be the number of seconds since the North Window
light was pointed due east. Find dy /dt. (Be-
+
cause x depends on t, x is an inside function.
By the chain rule, the answer will contain
dx /dt.) Rot a tion
c. Suppose that the beacon is rotating counter- y
clockwise at 0.3 rad / sec. What does dx /dt
equal? How fast is the point of light moving _l
along the wall when it passes the window Beacon
light
located at y = 300 ft?
Figure 4-4e
43. Antiderivative Problem: For each function below, write an equation for the antideriva-
tive. Remember "+C"!
a. y ' = cosx b. y ' = sinZx
c. y ' = sec 2 3x d. y ' = csc 2 4x
e. y' = 5 sec x tan x
44. Journal Problem: Update your journal with what you've learned since your last entry.
Include such things as those listed below.
The one most important thing you have learned since your last journal entry
The extension you have made in the derivative of a power formula
How to differentiate all six trigonometric functions, using what you have recently
learned
Anything you need to ask questions about during next class period
4-5 Derivatives of Inverse Trigonometric Functions
You have learned how to differentiate the six trigonometric functions. In this section
you will explore the inverses of these functions.
OBJECTIVE Differentiate each of the six inverse trigonometric functions.
Section4-5:Derivatives
of Inverse
Trigonometric
Functions 145
Background
ItemNumber1:TheInverseof a Function
The left-hand grap h in Figure 4-Sa shows how the population of a certain city
may grow as a function of time. If you are inter ested in findin g the time at which
th e population reaches a certain value, it may be mor e convenient to reverse the
variables and write time as a function of population. The relation you get by
interchanging th e two variables is called the inverse of the original function . The
graph of the inverse is shown on th e right-hand graph in Figure 4-Sa.
Population Time
/
/
Inverse /
relation /
,,
/
/
// y = x ,,'
/
/
, ,,
, ,,
, , 'or igina l
,,' fun ctio n
Time / Population
/
Function grap h Inverse relation grap h
Figure 4-50
For a linear function such as y = 2x + 6, interchanging the variables gives x = 2y + 6
for the invers e relation . Solving this equation for y in terms of x gives y = 0.Sx - 3.
The symbol f - 1 , pronounced "f inverse, " is used for the invers e function off. For
this exampl e,
if f(x) = 2x + 6, th en r- (x)
1
= o.sx - 3.
If f - 1 turns out to be a function (no two values of f - 1 (x) for th e same value of x),
th en the original function f is said to be invert ible . Realiz e that the - 1 exponent
does not m ean th e reciprocal of f(x). The inverse of a function undoes what the
r-
fun ction did to x . That is, 1 (f(x)) = x. For instanc e, th e square root function is the
r-
invers e of the squaring function. If f(x) = x.2 , then 1 (x) = # , and ../x2= x . Notice
in Figure 4-Sa that if the same scales are us ed for th e two axes, th en the graphs off
and f - l are mirror images With respect tO th e 45 line y = X .
Definition,Symbol,andProperty: Inverseof a Function
Definition: If y = f (x ), then the inverse of function f has the equ ation x = f(y ).
Symbol: If X = f (y), then y = r- 1 (x) .
Definition: If r-1 is a function, then f is said to be inver tible.
Property: If f is invertible, then f - 1 (f (x)) = X and f(f - 1 (x)) = x.
146 Chapter
4: Products,
Quotients
, andParametric
Funct
ions
Background
ItemNumber
2: TheInverseTrigonometric
Functions
The inverses of the trigonometric functions follow from the definition . For instance,
y = tanx Function .
x = tany Inver se function.
In the equation x = tany, y is called the inverse tangent of x, abbreviated
y = tan - 1 x
The symbol arctan x (for "the arc whose tangent is x") is sometimes used to help
you distinguish tan - 1 x from 1/tanx.
A clever way to plot the inverse of any function is to use the parametric mode on
your grapher. For the tangent function you would use the following .
X1 = t
Y1= tant Tan gent fun ction.
x2 = tant
Y2 = t Inverse tan gent relation.
Figure 4-Sb shows the two graphs on separate screens.
y = tanx y = tan - 1x = arctanx
Figure 4-5b
Unfortunately, the inverse tangent relation is not a function. There are many values
r: , of y for the same value of x . To make a function that is the inverse of tan x, it is
: , ~ci pal
- - - - - - rr/ 2 - ~,' -- _b_!:an
_qi customary to restrict the range to - rr / 2 < y < rr /2, or, briefly, (- rr /2, rr /2). This
-_y X
restriction includes only the branch of the graph that is nearest the origin. Figure
.-,,I'
I,...-.~_,_.,
_ .,,, / L -- ,r/ 2 - - - - - -
_ -::""'.
-:-:? 4-Sc shows this principal branch. The result is called the inve rse tangent function
tan- 1 ,/ :
/ I (as opposed to the inverse tangent relation).
,," tan 1
Figure 4-Sc also shows that the function and inverse function graphs are symmetrical
Figure 4-5c with respect to the line y = x. Everything that is an x-feature on the tangent graph
is a y-feature on the inverse tangent graph .
Sect
ion4-5: Derivativesof Inverse
Trigonomet
ric Functions 147
t~ -
~
The other five inverse trigonometric functions are defined the same way. For each,
a principal branch is a function. The principal branch is near the origin, continuous
if possible, and positive if there is a choice between two branches. The graphs are
shown in Figure 4-Sd.
y = tan - 1x
--- .- ... .
--
- ----------
X x X
-1
- - - - - - - rt/ 2
Sine inverse Tangent inverse Secant inverse
y = csc 1x
------ - -- ---------- ..- .. .
1t /2
X X X
-1 1
~ l
-rt/ 2
Cosine inverse Cotangent inverse Cosecant inverse
Figure 4-5d
The definitions and the ranges of the inverse trigonometric functions are summa-
rized in the following table.
Definitions:InverseTrigonometric
Functions
(PrincipalBranches}
y = sin - 1x if and only if sin y = x and y E [ - f, f J
y = cos - 1x if and only if cosy = x and y E [O,rr]
y = tan - lx if and only if tany = x and y E (-f, f)
y = cor- 1x if and only if coty = x and y E (0, rr)
y = sec 1x if and only if secy = x and y E [ 0, rr], but x * i
y = csc - 1 x if and only if cscy = x and y E [-f, f ], but x * 0.
Note: The names arcsin, arccos, arctan, arccot, arcsec, and arccsc can be used to help
distinguish, for instance, tan - 1 x from 1/tanx.
148 Chapter
4: Products,
Quotients,
andParametric
Functions
Algebraic
Derivative
of theInverseTangent
Function
Example 1 shows how to differentiate th e inverse tang ent function. The definition
of invers e function lets you turn this new problem into the old probl em of
differentiating the tang ent .
Example 1 Differentiate y = tan - 1 x .
Use the definition of tan 1 to \\Tite the eq uation in
Solution y = tan - 1 x => tany = x. terms of tangent.
The derivative of tan is scc 2 . Beca use y depends on x,
sec 2 y y' = 1 it is an inside f'll.llction. They ' is the deri\'ative of
V
this inside fun ction (from the chain rul e).
' 1 Use algebr a to so lve for y'.
y = sec 2 y
= cos 2 y
u
To find y ' in terms of x, consider that y is an angl e whose tangent and cosine
are being found. Draw a right triangle with angle y in standard position (Figure
Figure 4-5e
4-Se). By trigonometry ,
opposite leg
tan y = -"-"-----=
adjacent leg
Because tany = x, which equals x/1, put x on the opposite leg and 1 on
the adjacent leg. The hypotenuse is thus Jx2 + 1. Because cosine equals
(adjacent) / (hypotenuse), you can write
y' = 1 = _ l_ _
(Jxz+T
)2 x2 + 1
In Example 1, both the left and right sides of th e equation tan y = x are functions of
x. The technique of differ entiating both sides of such an equation with resp ect to x
is called implicit differentiation. You will study this technique more extensively in
Section 4-8.
Derivative
of theInverseSecantFunction
The der ivative of the invers e secant is trick y. Example 2 shows what happens.
Example 2 Differentiate: y = sec 1x.
Solution y = sec - 1 x => sec y = x Transform the new problem into an old pr ob lem .
secytany y ' = 1 Remember th e chain rule' That's where y' com es from.
' 1 Use algebr a to so lve for y ' .
Y = secytany
Consider y to be an angle. Draw a triangle with angle yin th e standard position
(Figure 4-Sf). Because secant equals (hypot enuse) / (adjacent) and secy = x = x/1,
write x on the hypotenus e and write 1 on th e adjacent leg. Because th e range of
the inverse secant function is Quadrants I and II, y will terminate in the second
quadrant if xis negative . In this case, x must be drawn in the negative direction,
Fig ure 4-5f as is done with negativ e radii in polar coordinates . By the Pythagorean theorem,
the third side squared is x 2 - l. This side is below the horizontal axis, so the
Section4-5: Derivativesof Inverse
Trigonometric Functions 149
)
8 .
)
radical for the third side is negative if x is negative. Because tangent equals
(opposite) / (adjacent), tany = -~ in this case . Because you started with
secy = x, the denominator of the derivative will be
x~ ifx > 0, or - x~ ifx < 0.
To avoid two different representations, use the notation Ix I and write the
answer in this way.
, 1
y =l xl~
The derivatives of the six inverse trigonometric functions are shown below . In the
following problem set you will derive the other four properties.
Properties: Derivativesof the Six InverseTrigonometric
Functions
1 d (cos - 1x) = - --- 1
-d (sin - 1x) = --- -
dx JI-=x2 dx JI=x2
d 1 1 d - 1 l
dx (tan - x) = 1 + x2 dx(cot x) =- 1 +x 2
~ (sec- 1x) = 1
~(csc 1x) =-
1
dx lxl~ dx lx l~
Note: Your grapher must be in the radian mode.
Memory Aid: The derivative of each "co-" inverse function is the opposite of the derivative
of the corresponding inverse function because each co-inverse function is decreasing at
x = 0 or 1 (Figure 4-Sd).
Example 3 Differentiate: y = cos - 1 5x7
V
Solution y = cos - 1 5x7 ~ cosy= 5x7
- siny y ' = 35x 6
35x 6 35x6
y' = - -- =- -;==== (Figure 4-5g) u
siny J l - 25x 14
Sx 7
Figure 4-5g
150 Chapter4: Products,
Quotients,
andParametr
ic Functions
Problem Set 4-5
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes .
QJ. sin'x = -?-
Q2. cos'x = -?- f(x)
7
Q3. tan 'x = - ?- 6 Quarter circle
Q4. cot'x = -?- 5
\ Line
QS. sec'x = -?- ~
Q6. csc'x = -?-
Refer to Figure 4-5h for Problems Q7-Ql0 .
3
2
1
~
Ql. f '(l) = -?- X
1234567
QB. f'(3) = - ?-
Q9. f'(4) = -?-
Figure 4-5h
Q10. f'(6) = -?-
For Problems 1-4, duplicate on your grapher the graphs in Figure 4-5d. For Problems 3
and 4, recall that csc y = 1/ sin y and that cot y = 1/tan y.
2. y = sin - 1 x
3. y = csc 1 x 4. y = coc 1 x
5. Explain why the principal branch of the inverse cotangent function goes from O to rr
rather than from - rr/2 to rr/2.
6. Explain why the principal branch of the inverse secant tunction cannot be continuous.
7. Evaluate sin (sin - 10.3 ).
8. Evaluate cos - 1 (cos 0.8).
For Problems 9-12, derive the formula shown .
9. ~(sin - 1 x) =
dx
h 1- x2
d
10. -,-- (cos - 1x) = -
dx
-===
1
.Jf=xZ
d 1 d 1
11. -d (csc- 1x) = - I~ 12. dx (cot - 1x) = - + x2
X , 1
X X -1 1
For Problems 13-24, find the derivative algebraically .
13. y = sin - 1 4x 14. y = cos - 110x
15. y = cot - 1x0 5 16. Y = tan - 1xo.s
17. y -- sec 1~
3 18. y,..=csc 1 fo
19. y = cos- 1 5x2 20. f (x) = tan - 1 x 3
('
21. g(x) = (sin - 1x) 2 22. u = (sec 1x) 2
Section4-5:Derivatives
of Inverse
Trigonometric
Functions 151
23. v = x sin - 1 x + (1 - x 2 ) 112 (Surprise!) 24. I(x) = cot - 1 (cotx) (Surprise!)
25. Radar Problem: An officer in a patrol car sitting 100 ft from the highway observes a
truck approaching (Figure 4-5i).
a. At a particular instant, t seconds, the truck
is x ft down the highway. The line of sight
to the truck makes an angle of 0 radians to
a perpendicular to the highway . Explain why
e = tan - 1 (x/ 100) .
b. Find d0 /dx. Use the chain rule to write an
equation for d0 /dt .
c. When the truck is at x = 500 ft, the angle is
observed to be changing at a rate d0 /dt = - 0.04 Figure 4-5i
rad / sec. How fast is the truck going? How many
miles per hour is this?
26. Exit Sign Problem : The base of a 20-ft tall exit
sign is 30 ft above the driver's eye level (Figure
4-5j). When cars are far away, the sign is hard
to read because of the distance . When they are
very close, the sign is hard to read because the
driver has to look up at a steep angle. The sign
is easiest to read when the distance x is such
that the angle 0 at the driver's eye is as large as
possible .
a. Write 0 as the difference of two inverse
cotangents. Figure 4-5j
b . Write an equation for d0 /dx .
c. The sign will be easiest to read at the value of x where 0 stops increasing and starts
decreasing . This happ ens when d0 /dx = 0. Find this value of x .
d. Confirm that your answer in 26c is correct by plotting 0 as a function of x and thus
showing that the graph really does hav e a high point at that value.
27. Numerical Answer Check Problem: For f(x) = cos - 1x, make a table of values that show
f'(x) numerically and f' (x) by the formula. Start at x = - 0.8 and go to x = 0.8, with
t,.x = 0.2 . Show that the formula and the numerical derivative give the same answers
for each value of x.
28. Graphical Analysis Prob lem: Figure 4-5k shows the graph of y = sec~' x
y = sec 1 x. . ! . j . -~~ 4
a. Calculate the derivative at x = 2. Based on what the graph ~ -;
shows, why is the answer reasonable?
-----, - --.~iz _,- '__-~
b. What does y equal when xis 2? What does (d/dy) (secy)
X
equal for this value of y ?
4
c. In what way is the derivative of the inverse secant function
related to the derivative of the secant function? Figure 4-5k
29. General Derivative of the Inverse of a Function : In this problem you will derive a genera l
formula for the der ivative of the inverse of a function .
152 Chapter
4: Products,
Quotients,
andParametricFunctions
1
a. Let y = sin - 1 x . Show that ddy = - - .
x cosy
b. By directly substituting sin - 1x for yin 29a, you get ddy = (1 )" Show that
X COS Sill - 1X
the formula of this section and this problem's formula for dy /dx give the same
value when x = 0.6 .
c. Show that the following property is true for the derivative of the inverse of a
function.
Property: Derivativeof the Inverseof a Function
f r-1 d -I 1
I y = (x), then dx (f (x)) = f'(f - I(x)).
d. Suppose that f(x) = x 3 + x . Leth be the inverse function off. Find x if f(x) = 10.
Use the result and the property given above to calculate h '(lO).
30. Quick! Which of the inverse trigonometric derivatives is preceded by a negative ( - )
sign?
4-6 Differentiability and Continuity
It is time to pause in the study of derivatives and take care of some unfinished
business. If a function f has a value for f'(c), then f is said to be differen tiable at
x = c. If f is differentiable at every value of x in an interval, then f is said to be
differentiable on that interval.
Definitions
ot o point:Function f is differentiable at x = c if and only if
Differentiability f'(c) exists.
(That is, f'(c) is a rea l number.)
Differentiability
ononinterval:Function f is differentiable on an interval if and only if it is
differentiable for every x value in the interval.
Function f is differentiable if and only if it is differentiab le at every value
Differentiability:
of x in its domain.
In Section 2-4, you learned that a function f is continuous at x = c if limx-c f(x) =
f(c). A function can be continuous at x = c without being differentiable at that
point. But a function that is differentiable at x = c is automatically continuous at
that point. Figure 4-6a illustrates the two cases.
Section
46:Differentiability
andContinuity 153
f( x )
Not "smooth ":
L = {(c) ------ "Cusp "
X X
C
Continuous, but not differentiable Differentiable, thus continuous
Figure 4-6a
OBJECTIVE Prove that a differentiable function is continuous, and use this property to prove that
certain functions are continuous.
To prove that a function f is continuous at x = c, you must show that limx-c f (x) =
f(c ). One form of the definition of derivative contains all th ese ingredients.
f'(c) = lim f (x ) - f (c)
x-c X - C
The trick is to perform some mathematically correct operations that lead from the
hypoth esis to the conclusion . In this case it is easi er to start somewhere "in the
middle" and pick up th e hypothesis along the way. Here goes!
Property Prove that if f is differentiabl e at x = c, then f is continuous at x = c.
Proof You must prove that limx-c f (x) = f(c) .
lim [f(x) - f(c ) ] Start wit h something th at cont ain s limi t, f (x), and f (c).
x-c
_ x-c
- 1.Im [f----
(x) - f(c) ( X - C )] Multiply by- - .
x-c X - C x -c
= lim f(x) - f(c) lim (x - c) Lim.it of a produ ct.
x-c X - C x-c
= f' (c) 0 Definition of derivative and limit of a linear function .
=0 Becau se f is diff erenti able at x = c,
f' (c ) is a real numb er;
(numb er)-0 = 0 .
.. lim [f(x) - f(c)] = 0 Transit ive prop erty .
x-c
But lim [f (x) - f(c ) J = [lim f (x) J - [lim f(c) J = [lim f(x)] - f (c).
x- c x- c x-c x-c
[lim f (x) ] - f (c) = 0, also. Trans itive property again.
X- C
.. lim f(x) = f(c)
x-c
.. f is continuous at x = c, Q.E.D. Definition of continuit y.
The secret to this proof is to multipl y by 1 in the form of (x - c )/( x - c). This
transformation causes the difference quotient [f(x) - f (c) ]/( x - c ) to appear inside
154 Chapter
4: Product
s, Quotients,
andParametr
ic Functio
ns
the limit sign. The rest of the proof involves algebra and limit properties and the
definitions of derivative and continuity.
Property: DifferentiabilityImpliesContinuity
lf function f is differentiable at x = c, then f is continuous at x = c.
Contrapositive If function f is not continuous at x = c, then f is not
of the Property:
differentiable at x = c.
(The converse and the inverse of this property are false.)
This property and its contrapositive provide a simple way to prove that a function
is continuous or not differentiable, respecti vely.
Example 1 Prove that f(x ) = x 2 - 7x + 13 is continuous at x = 4.
Solution f' (x ) = 2x - 7
f'(4 ) = 2(4) - 7 = 1, which is a real number .
.'. f is differentiable at x = 4.
.'. f is continuous at x = 4, Q.E.D. Differentiabilit y impli es continuit y.
Note that you could prove that f is continuous by applying the limit theorems. The
technique in Example 1 provides a faster way if you can find the derivative easily.
2
Example 2 Is the function g (x) = (x - ) (; + 3 ) differ entiable at x = 2? Justify your answer.
x-
Solution The function g has a (removable) discontinuity at x = 2.
.'. g is not differentiable at x = 2. Contrap ositive of diff erentiability
impli es continuity.
The most significant thing for you to understand is the distinction between the
concepts of differentiability and continuity. To help you acquire this understanding,
it helps to look at graphs of functions and state which, if either, of the two properties
applies .
Example 3 State whether or not the following functions are differ entiable or continuous at
X = C.
h (x)
X
C C
a. b. C. d.
Figure 4-6b
Sec
tion4-6:Differentiabi
lity andContinuity 155
t~~-
Q
Solutions a. The function is continuous but not differentiable . At the cusp, the rate of
change approaches a different number as x - c from the left side than it
does as x - c from the right.
b. The function is neither continuous nor differentiable . There is no limit for
g (x ) as x - c.
c. The function is neither continuous nor differentiable . Although the
graph appears "smooth" as x goes through c, the difference quotient
[h(x) - h(c)] / (x - c) approaches +oo (positive infinity) as x approaches c
from the left side, and approaches -oo as x approaches c from the right.
d. The function is continuous and differ entiable. The discontinuity elsewhere
has no effect on the behavior of the function at x = c.
Example 4 Find values of the constants a and b that make function f diff erentiabl e at x = 2.
Check your answer by graphing .
ax 3 , if x .s; 2
f (x ) = { b(x - 3) 2 + 10, if x > 2
Solution Let Yi = ax 3 and let Y2 = b(x - 3) 2 + 10.
For f to be differentiable at x = 2, it must also be continuous at that point.
Thus, at x = 2, Yi must equal Y2 and y; must equal y 2.
y ; = 3ax 2 and y 2 = 2b(x - 3)
.. 8a = b + l0 Y1 = y2 atx = 2.
12a = - 2b =} b = - 6a y; = Y2at x = 2.
f(x)
8a = - 6a + 10 By sub stinition .
10
a = ?.
7
5 30
b = - 6. 7 = -7
y
Figure 4-6c shows the graph off with thes e values of a and b. The broken end
portions of the graph show Yi and Y2 outside their respective domains . As you
Figure 4-6c can see, the graphs have the same slope at x = 2, and they are continuous. This
means that f is differentiable at x = 2.
156 Chapter
4: Products,
Quotients,
andParametric
Funct
ions
Problem Set 4-6
DoTheseQuickly
Th e follo wing prob lems are intended to refr esh your skills . You sh ou ld be able to do all
ten problems in less th an five minut es.
'' 4 .....
y ,. ..
01. Write th e definition of continuit y.
.... ;... :::::::::::::
Q2. Write th e definition of derivati ve.
-~----.;. i+
03. Find y : y ' = 12x- 3
X
Q4. Find cos ' x. . . . 6
QS. Find dy / dx: y = tanx .
+---
... .-:"f +
i"+
-~-- .!. .;. --~ --~-
06. Find : x sec- x.1 ..;......... :....( .. .'.-;..,;----:.-
'
:---~----
(i----1
} 1---j-
Ql. If f (x) = x4, find f'(2 ) .
QB.Find dy /dx: y = (x 3 + 1 ) 5 . Fig ure 4-6d
Q9. Estimat e th e definit e integral from -2 to 2 of the function in Figur e 4-6d.
Q10. Sketch the derivat ive graph for the function in Figure 4-6d .
For Prob lems 1- 12, state whether the function is con tinu ou s, differe n tiable, both, or
n eith er at x = c.
2. 3. 4.
~
l. y y
X X X
'
C C C
5. 6. 7. 8.
/
Y. C C
X
C
I
I
X
9. 10. y 11. 12. y
X C __..
X
Sec
tion4-6: Differen
tiability andCont
inuity 157
18
I
Q I
-
.U)
x(t )
I
For Problems 13-20:
a. Sketch the graph of a function that has the indicated features .
b. Write the equation for a function that has these features.
13. Is differentiable and continuous at the point (3, 5)
14. Is differentiable and continuous at the point (- 2, 4)
15. Has a finite limit as x approaches 6, but is not continuous at that point because f(6)
is undefined
16. Has a finite limit as x approaches 1, has a value for f(l), but still is not continuous at
that point
17. Has a value for f( - 5) but has no limit as x approaches -5
18. Has a cusp at the point (- 1, 3)
19. Is continuous at the point (4, 7) but is not differentiable at that point
20. Is differentiable at the point (3, 8) but is not continuous at that point
For Problems 21-30, sketch the graph. State whether the function is differentiable,
continuous, neither, or both at the indicated valu e of x = c.
21. f(x) =l x-3 1,c=3 22. f(x) = 4 + lxl, c = 2
23. f(x) = sinx, c = 1 24. f(c) = tanx, c = rr/2
2 if X $ 3 2 if X ~ 1
25. f(x) = { x - 4x + 8, 26. f(x) = { x - 6x + 8,
11 - X, if X >3 7 - 4x, if X < 1
C = 3 C = 1
if X < 1 if X $ 1
27. f(x)={x 2 -x+l, 28. f(x ) = { x2 - x + 1,
X + 1, if X ~ 1 2 - X, if X > 1
C =1 C = 1
x3 - 8 Ix - 21
29. f (X) = -- , C =2 30. f(x) = x 2 + 2x + 4 - -- ,c=2
x-2 x- 2
For Problems 31-34, find values of the constants a and b that make the function
differentiable at the point where the rule for the function changes. Check by grapher.
2
x3 , ifx < l {-(x-3) + 7, ifx ~ 2
3 1. f(x) = a(x - 2) 2 + b, if x ~ 1 32 f (x) = ax 3 + b, if x < 2
{
2 a/x, if x ::; 1
33 f(x) = {ax2 + 10, if x < 2 34 f (x) = 12 - bx 2, if x > 1
x - 6x + b, if x ~ 2 {
35. Railroad Curve Problem: Curves on a railroad track are made in the shape of sections
of cubic parabolas rather than arcs of circles . This ensures that locomotives are
gradually eased into curved paths and are thus prevented from derailing. Suppose a
track goes along the negative x-axis and starts curving at x = 0. By the time the curve
reaches x = 100 yd, the track lies at a 45 angle to its original direction (Figure 4-6e).
158 Chapter
4: Products,
Quotients,
andParametr
ic Funct
ions
a. The equation for the path of the track is
y
o,
if X < 0 Straight
if O :::;x ::; 100
y = ax 3,
{ X + b, if X > 100 \
Find the values of a and b that make the path differentiable Tr ack
at X = 100. X
b. Show that at x = 0 (where the train will first enter the curve)
the rate of change of slope is zero. Also show that thereafter
the slope increases uniformly with x. Figure 4-6e
36. Bicycle Frame Design Problem: Figure 4-6f shows a side view of a
bicycle frame's front fork, holding the front wheel. To make the
bike track properly, the fork curves forward at the bottom where
the wheel bolts on. Assume that the fork is bent in the shape of
a cubic parabola, y = ax 3 + bx. What should the constants a and X
b be so that the curve joins smoothly to the straight part of the
fork at the point (10 cm, 20 cm) with slope equal to 5?
2 lx - 21 .
37. Let f(x) = X - if X * 2
~
{
4, if X = 2
Find an equation for f'(x). Show that function f is not differen-
tiable at x = 2, even though the left and right limits of f'(x) are Figure 4-6f
equal as x approaches 2.
38. Baseball Line Drive Problem: Milt Farney pitches his famous fastball. At time t = 0.5
sec after Milt releases the ball, Joe Jamoke hits a line drive to center field. The distance,
d(t ) feet, of the ball from home plate is given by the two-rule function shown below .
60.5 (~ -~ - t), if t ::;0.5
d(t) = . +t
1( 150 2 - t1), if t ~ 0.5 ~n
Figure 4-6g shows the graph of function d. 200
a. Find an equation for d'(t). Be careful about the inequality
signs at t = 0.5.
b. Prove quickly that d is continuous at t = l. 100
c. Find the limit of d ' (t ) as t - 0.5 - and as t - 0.5 +. State the
real-world meanings of these two numbers .
d. Explain why dis continuous, but not differentiable, at t = 0.5. 0.5 2
e. What is the significance of the number 60.5 in the first rule
for d(t)? Figure 4-69
39. Continuity Proof Problem: Use the fact that differentiability implies continuity to
prove that the following kinds of functions are continuous.
a. Linear function, y = mx + b
b. Quadratic function, y = ax 2 + bx + c
c. Reciprocal function, y = 1/x, provided x *0
Section4-6:Differentiabi
lity andContinuity 159
d. Identity function, y = x
e. Constant function, y = k
40 . D ifferentiability Im plies Continuity Proof: Prove that if f is differentiable at x = c, then
it is cont inuou s at that point. Try to do the proof without looking at the proof in the
text. Consult the text only if you get stuck.
4-7 Derivative of a Parametric Function
Figure 4-7a shows ho w a pendulum
hung from the ceiling of a room
might move if it were to swing in both
the x- and y-directions. It is possible
to calculate its velocity in both the
x- and y-direct ions, and along its
curved path. These rates help you
determine facts about the path of a
moving object. In this sec tion you
will use parametric functions to
make these determinations.
/ Path x
Figure 4-7a
OBJECTIVE Given equations for x and yin terms oft, find dx /dt, dy /dt, and dy /dx.
As the pendulum in Figure 4-7a swings , it goes back and forth sinusoidally in both
the x- and y-direc tions . By using the methods you learned in Section 3-8, you can
find equations for these sinusoids. Suppose that the equations for a particular
pendulum are
X = 50 COS l.2t
y = 30 sin l.2t,
where x and y are in centimeters and t is time in seconds . The variable t on which
x and y both depend is called a parameter . (The wor d par ameter means "parallel
measure .") The two equations, one for x an d one for y, are called parametric
equations.
You can use the parametric mode on your grapher to plot the xy-graph of th e
pendulum's path . The result is an ellips e, as shown in Figur e 4-7b . The ellips e goes
from - 50 to 50 cm in the x-directi on an d from -30 to 30 cm in th ey -direction . The
Figure 4-7b points 50 and 30 appear in th e param etric equation s as the amplitudes of the two
sinusoids. (If the pendulum were at rest, it would han g over the origin .)
160 Chapter
4: Products
, Quotients
, andParametric
Functions
The rates of change of x and y with respect to t can be found by differentiating.
!; = - 60 sin l.2t and !~ = 36 cos l.2t
y Evaluating these derivatives at a certain value of t, say t = 1, shows that the
_ "C':7
_s___
to..:pe = -0.2 33
30 pendulum is moving at about -55.9 cm/ sec in the x-direction and at about
13.0 cm/ sec in they-direction. If you divide dy/dt by dx/dt, the result is the slope
X
50
of the ellipse, dy /dx, at the given point.
dy = 13.044 .. . = - 0.233 ...
dx -55.922 ...
Figure 4-7c A line with this slope at the point where t = 1 is tangent to the graph (Figure 4-7c).
The property illustrated by this example is called the parametric chain rule.
Property: TheParametricChainRule
If x and y are differentiab le functions oft, then the slope of the xy -graph is
dy dy /dt
dx = dx /dt
. x = 3cos2rrt
Example 1 G1ven: .
y = 5 sm rrt,
a. Plot the xy-graph. Use a t-range that generates at least one complete cycle of x
and y . At-step of 0.05 is reasonable . Sketch the result.
b. Describe the behavior of the xy-graph as t increases.
c. Find an equation for dy/dx in terms oft .
d. Calculate dy /dx when t = 0.15. Show how the answer corresponds to the graph.
e. Show that dy/dx is indeterminate when t = 0.5. Find the approximate limit of
dy /dx as t approaches 0.5. How does the answer relate to the graph?
f. Make a conjecture about what geometrical figure the graph represents. Then
confirm your conjecture by eliminating the parameter t and analyzing the
resu lting Cartesian equation .
g. How do the range and the domain of the parametric function relate to the range
and the domain of the Cartesian equation in part f?
Solution a. Figure 4-7d shows the graph as it might appear on your grapher. The period
for xis 2rr/2rr = l; for y it is 2rr/rr = 2. Thus a minimal range is O s t s 2.
b. If you watch the graph being generated, you will see that the points start at
(3, 0), go upward to the left, stop, retrace the path through the point (3, 0),
X
then go downward to the left, eventually corning back to the point (3, 0).
c. dx 6 . 2
dt = - TT sm rrt and !~ = 5rr cos rrt
dy 5rr cos rrt 5 COS TTt
dx - 6rr sin 2rrt -6 sin 2rr t
Figure 4-7d
Section 4-7: Derivative of a Para
metric Function 16 1
)
8 .
)
d. t =O .1 5 ~
dy = ----
5cos0 .15rr = --4.455- ...- =- 0917
. . ..
dx -6 sin0 .3rr - 4.854 .. .
As shown in Figure 4-7d, a tang ent line to th e gra ph at the point where
t = 0.15 has slope of about - 1, which corresponds to the exact value of
- 0.917 ....
e. t = 0 .5 d = 5 cos 0.5rr
-dy
~ . . h 1s
= -O, wlnc . m. d etermmate
. . A grap h of d y /d x
x - 6 sm rr 0
versus t (Figure 4-7e) shows a removable discontinuity at t = 0.5.
dy/ dx dy /dx
0.498 - 0.4166749
0.5
0.499 - 0.4166687
-1 0.500 (no value )
0.501 - 0.4166687
0.50 2 -0.4 16674 9
Figure 4-7e
To find the limit of d y /dx more pr ecisely, either zoom in on the discontinuit y
or us e the tab le feature . The limit appears to be - 0.4166666 . . . , which equals
- 5/12.
f. The graph appears to be a parabola . Eliminating the param eter t involves
solving one equation for tin terms of x (or y) and substituting the result into
the other equation. S0metin1es there are shortcuts that will let you do th.is
mor e easily, as shown below.
x = 3 cos 2rrt and y = 5 sin rrt The given parametric equations.
Th e doubl e argument pr operty gets cos Zrrl in
x = 3(1 - 2sin 2 rrt) ter m s of sin rrt, which appea rs in the or igin al
par am etr ic equ ation for y.
But sin rrt = (y/5) . From the original param etri c equations.
Subs titutin g y/5 for sin rrt eliminate s th e
.. X = 3( 1 - 2(y/5) 2 )
para m eter t.
6
X = -- y2 + 3 By algeb ra.
25
As conjectured, th.is is the equation of a parabola opening in the negative
x-direction .
g. The Cartesian equation has domain x ~ 3, which is unbounded in the negative
x-direction. The parametric graph stops at x = -3 .
162 Chapter
4: Products,
Quotients,
andParametric
Functio
ns
Problem Set 4-7
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Differentiate: y = 0.2x 1215
Q2. F"m d dy_ _ X - 3
dx Y - x - 1 y
Q3. Find f'(x) : f(x) = x cos x .
Q4. Find y ' : y = sin (x5 ) .
d X8
QS. Find d(y): y =5 .
X X X
Q6. D"ff .
1 erennate: y = - -
sin 8
cos 3
Ql. Find 0': 0 = cos - 1x.
QB. For Figure 4- 7f, sketch the graph of y'.
Q9. If v'(S) = - 3, what can you conclude about v(t) at t = S?
Q10. If u' (7) = 4, what can you conclude about u (x) at x = 8? Figure 4-7f
1. Parabola Problem: A parametric function has the following equations.
X= 2+ t
y = 3 - t2
a. Make a table of values of x and y for each integer value oft from - 3 through 3.
b. Plot the graph of this function on graph paper, using the points in la.
c. Find dy /dx when t = 1. Show that the line through the point (x, y) from la, with
slope dy /dx, is tangent to the graph at that point.
d. Eliminate the parameter t and show that the resulting Cartesian equation is that of
a parabola .
e. Find dy /dx by direct differentiation of the equation in ld. Show that the value
of dy /dx calculated this way is equal to the value you found in le by using the
parametric chain rule.
2. Semicubical Parabola Problem: A parametric function has the following equations.
X = t2
y = t3
a. Make a table of values of x and y for each integer value oft from - 3 through 3.
b. Plot the graph of this function on graph paper, using the coordinate pairs found
in 2a.
c. Find dy /dx when t = 1. Show that the line through the point (x, y) from 2a, with
slope dy /dx, is tangent to the graph at that point.
d. Eliminate the parameter t. Find y in terms of x. From the result, state why this
graph is called a semicu bical parabola.
Section
4-7:Derivativeof a Parametric
Function 163
t~ .
DD
e. Find dy /dx by direct differentiation of the equation in 2d. Show that the value of
dy /dx calculated in this way is equal to the value you found in 2c by using th e
parametric chain rul e.
3. Ellipse Problem: The ellipse in Figure 4-7g has the par ametr ic
equations y
X = 3 COS t
y = S sin t.
a. Plot the graph on your grapher . Sketch the result or photocopy
. .
the text graph . X
b . Find an equation for dy /dx. ..,i ... ,
c. Evaluate the point (x, y) when t = rr/4 , and find dy /dx when
t = rr/4. Plot a line on your graph at this point (x, y) that has
slope dy /dx. Is the line tangent to the graph? ! l
' . '
d. Determine whether the following statement is true or fals e:
When t = rr/4, the point (x, y) is on a lin e throu gh th e ori gin Figure 4-7g
that makes a 45-degree angl e with the x- and y -axes .
e. Use your equation for dy /dx from 3b to find all th e points
where the tangent lin e is vertical or hori zo ntal. Show these
.
_,
. .
:- 1 -:---~-. ~. : -
Y.
.;... ;... ;. . ; ' ...' -~;..
...; .:. <8 ..., .., ; : :; ~ ::
points on your graph. . ' ' ' ''
f. Eliminate the paramet er t and thus confirm that your graph
actually is an ellipse. This elimination ma y be don e by-cleverl y
applying the Pythagorean propert y for sine and cosine . __._
~ ::-1--'_,__i....J...-'-+-+-'_,__+-+-9
"'!......._.
., . ,. X
8..:.
4. Astroid Probl em: The star-shaped curve in Figure 4-7h is called ... ~-+
an astroid . Its parametri c equations are
X = 8 COS3 t
y = 8sin 3 t .
a. Plot the graph on your gra pher. Sketch the result or photocop y Figure 4-7h
th e text graph.
b . Find an equation for dy /dx .
c. Evaluate the point (x, y) when t = 1, and find dy /dx when t = l. Plot a line on your
graph at this point (x, y) that has slope dy /dx . Is th e line tangent to th e graph?
d . At each cusp, dy /dx has the indeterminat e form 0/ 0. Explain the diff eren ce in
b ehavior at th e cusp at th e point (8, 0) and at the cusp at the point (0, 8) .
e. Eliminate th e paramet er t. This transformation ma y be done by solving th e two
equations for th e squares of cos t and sin t in terms of x and y, th en using th e
Pythagorean prop er ty for sin e and cosine .
S. Circle Problem : A param etric function has the follm ving equations .
X = 6 + 5 COS t
y = 3 + 5 sin t
a. Plot the graph of this function. Sketch the result.
b . Find an equation for dy /d x in ter ms oft.
164 Chapter
4: Product
s, Quot
ients, andParametric
Functions
c. Find a value oft that makes dy /dx equal zero . Find a value oft that makes dy /dx
infinite . Show a point on the graph for which dy /dx is infinite . What is true about
dx /dt and about dy /dt at a point wher e d y/ dx is infinite?
d. Eliminate the parameter t. This may be done by expressing the squares of cosine
and sine in terms of x and y, then by applying the Pythagorean property for sine
and cosine.
e. From the equation in Sd, you should be able to tell that the graph is a circle . How
can you determine th e center and the radius of the circle just by looking at the
original equations?
6. Line Segment Problem: Plot the graph of th e parametric function
x = cos 2 t
y = sin 2 t.
Show that dy /dx is constant . How does this fact corr espond to what you observe
about the graph? Confirm your observation by eliminating the parameter to get an
xy-equation. Describe the difference in domain and rang e between th e parametric
function and the xy-equation .
7. Deltoid Problem: The graph of the parametric function y
X = 2 COS t + COS 2t 3- ...
y = 2 sin t - sin 2 t
is shown in Figure 4- 7i. It is called a deltoid, for reasons that
should be obvious. X
a. Confirm by grapher that the equations shown above give
the deltoid in Figure 4-7i.
b. Find an equation for dy /dx in terms oft.
c. Show that at two of the cusps, the tangent line is neith er r ...,
horizontal nor vertical, yet the derivative dy /dx fails to
exist. Find the limit of dy /dx as t approaches the value at Figure 4-7i
the cusp in Quadrant II.
8. Witch of Agnesi Problem: The Witch of Agnesi, y
named for Italian mathematician Maria Gaetana ..... , ........ .... 10
Agnesi (1718 - 1799), has the equations
x = 2atant
y = 2acos 2 t,
where a stands for a constant. -10 -5 0 5 10 1.5
a. Figure 4-7j shows a curve for which a= 3.
-5
Plot the graph. Sketch the result or photo-
copy the text graph.
.. -
~
b. Find dy /dx in terms oft.
c. Eliminate the parameter to get an equation Figure 4-7j
for y in terms of x .
d. Differentiate the equation in 8c to get an equation for dy /dx in terms of x.
Sect
ion4-7:Derivativeof a Parametric
Function 165
e. Show that both equations for dy/dx give the same answer at t = rr/4, and that a line
through the point where t = rr/4 with this value of dy /dx as its slope is tangent to
the curve .
9. Invo lute Problem: A string is wrapped around a circle 1 in.
in radius. As the string is unwound, its end traces a path Involute y
called the involute of a circle (Figure 4-7k). The parametric
equations of this involute are
x = cos t + t sin t
y = sin t - t cos t,
where t is the number of radians from the positive x-axis to X
the radius drawn to the point of tangency of the string.
a. Use your grapher to confirm that these parametric equa-
tions give the graph shown in Figure 4-7k.
b . Find dy/dx in terms oft. Simplify as much as possible.
c. Show that the value you get for dy /dx at t = TT is consistent Figure 4-7k
with the graph.
10. Clock Prob lem: A clock sits on a shelf close to a wall (Figure
4-71). As the second hand turns, its distances x cm and y
cm from the wall and the shelf, respectively, depend on the Wall
number of seconds, t, since the second hand was pointing
straight up.
a. Write parametric equations for x and yin terms oft.
b. At what rates are x and y changing when t = 5 sec? T
c. What is the slope of the circular path traced by the second
hand when t = 5? 25 cm -1 Shelf
d. Confirm that the path really is a circle by finding an
xy-equation. Figu re 4-7 1
11. Pendulum Project: Suspend a small mass from the ceiling on the end of a nylon cord .
Place meter sticks on the floor, crossing them at the point below which the mass hangs
at rest, as shown in Figure 4-7m. Determine the period of the pendulum by measuring
)'
Meter sticks
Figure 4-7m
166 Chapter
4: Products,
Quotients,
andParametr
ic Functions
the time for 10 swings. Then start the pendulum in an elliptical path by pulling it 30
cm in the x-direction and pushing it sideways just hard enough for it to cross the
y-axis at 20 cm. Write parametric equations for the path this pendulum traces on the
floor. Place a coin on the floor at the point at which you predict the pendulum will be
at time t = 5 sec. (Lay the coin on top of a ruler tilted at an angle corresponding to the
slope of the path at that time.) Then set the pendulum in motion again. How close do
the predicted point and slope come to those you observe by experiment ?
12. Spring Problem: Figure 4-7n shows a "spring" drawn
by computer graphics. Figure out equations for a y
parametric function that will generate this graph. 5
How did you verify that your equations are correct?
Use your equations to find values of x and y at which
the graph has interesting features, such as horizontal
or vertical tangents and places where the graph seems
to cross itself.
13. Lissajous Curves: Graphs of parametric functions X
with the equations 10
X = nt
COS Figure 4-7n
y = sint,
where n is (usually) an integer, are called Lissajous curves,
or sometimes Bowditch curves. In this problem you will
investigate some of these curves.
a. Figure 4-70 shows the Lissajous curve with the parametric
equations
y
X = COS 3t
y = sint.
Use your grapher to confirm that these equations generate
this graph. X
b. Plot the Lissajous curve with the equations
x = cos4t
y = sint.
Sketch the resulting curve.
c. In what fundamental way do the curves differ for n = 3 Figure 4-70
(an odd number) and for n = 4 (an even number)?
d. Sketch what you think these curves would look like.
i. x = cos St ii. x = cos6t
y = sin t y = sin t
e. Plot on your grapher the two curves in 13d. Did your predicted graphs match the
actual ones in number of vertices and in behavior at the extreme points?
f. What two familiar curves are special cases of Lissajous curves when n = 1 and
n = 2?
Section4-7:Derivative
of a Parametric
Function 167
)
S .
)
4-8 Graphs and Derivatives of Implicit Relations
If y equals som e fun ction of x, su ch as y = x 2 + sin x, th en th ere is said to be an
expli cit relation ship between x and y . The word expli cit com es from th e sam e root
as the wor d ex plain. If x and y ap pear in an equ ation such as
x 2 + y 2 = 25,
then there is an implicit relation between x and y becaus e it is only "impli ed" that
y is a fun ction of x. In thi s section you will see how to diff erenti ate such an impli cit
relation without first solving for y in terms of x . As a result, you will be abl e to
pro ve th at th e power rul e for deriva tives works when th e expon ent is a rational
numb er, not ju st an intege r. In Section 4-5, you u sed implicit differ entiation to find
derivatives of th e invers e tri gonometri c fun ctions.
OBJECTIVE Given the equation for an implicit relation, find the derivative of y with respect to x, and
show by graph that the answer is reasonable.
Example 1 Consid er th e impli cit relat ion x 2 + y 2 = 25 plotte d in Figure 4-8a.
. --~.. - 5~
Figu re 4-Ba
a. Tell why th e graph is a circle.
b. Differe ntiat e impli citly to find d y /d x.
c. Calcul ate the two values of y when x = 3.
d. Thr ough the point determin ed in p art c with th e lower valu e of y , draw a line
with slop e dy /d x. How is thi s line related to th e graph ?
Solutions a. The graph is a circle by the Pythagorea n theorem. Because x 2 + y 2 = 2 5 = 52 ,
all point s on the gra ph are five unit s from the origin, implyin g th at th e graph
is a circl e.
168 Chapter
4: Product
s, Quotients
, andParametric
Function
s
b. For simplicity, use y ' for dy /dx.
2x + 2yy ' = 0 The y ' com es from the chain ru le.
2yy ' = -2x
-x
y' = -
y
C. X = 3 : 9 + y 2 = 25
y 2 = 16 ~ y = 4
3
d. Atthe point (3, -4), slope= y' = - - = 0. 75 . The line goes through the point
4
(3, -4) and is tangent to the graph (Figure 4-8b) .
.. ~ .\
-5 5 9
:,J' !
y
4';,..'
. j ..
r
-:- .. ,....
1/
.!.......
.,-l /~:.
Figure 4-8b
Example 2 For the implicit relation y 4 + x 3 y 5 - 2x 7 = 13, find dy /dx.
Solution For simplicity, use y ' for dy /dx. Be on the lookout for the derivative of a product
property. Of course, you must also obey the chain rule wherever y appears as
an inside function.
4y 3 y ' + 3x 2 y 5 + x 3 5y 4 y ' - 14x 6 = 0
Note that y ' shows up as a result of the chain rule and is only to the first power,
so you can use relatively easy algebra to isolate y ' .
4y 3 y ' + x 3 5y 4 y ' = - 3x 2 y 5 + 14x 6
(4y+ 5x 3 y 4 )y ' = - 3x 2 y 5 + 14x 6
3
, -3x 2 y 5 + 14x 6
y = 4y 3 + 5x 3y4
The answer to Example 2 expresses y ' in terms of both x and y, which is okay
because the original relation had x and y together. Given a point (x, y) on the graph,
the answer above could be used to find y'. In practice, it is usually harder to find a
point on the graph than it is to do the calculus' Thus you will work with problems
like that in Example 2 mainly for practice in differentiating.
Section
4-8:Graphs
andDerivatives
of ImplicitRelations 169
s0(t)
x(t )
Technique:ImplicitDifferentiation
To find dy /dx for a relation whose equation is written implicitly:
1. Differentiate both sides of the equation with respect to x. Observe the chain rule by
multiplying by dy/dx each time you differentiate an e>-'Pressioncontaining y .
2. Isolate dy /dx by getting all of the dy /dx terms onto one side of the equation, and all
other terms onto the other side . Then factor, if necessary, and divide both sides by the
coefficient of dy /dx.
Example 3 If y = x 7f3, prove that the power rule for deri vatives gives th e right answer for y'.
Solution y3 = x 7 Cube both sides of the given equ atio n.
Use the power rule for integer expon ent s to
3y 2y' = 7x 6 diff ere nt iate impli citly with respect to x.
7x 6
y' = 3y 2
7 x6 7 x6 7
y' = --- = --- = - x4/3
3 (x i/3)2 3 xl4/3 3
This is the answer you would get by direct application of th e power rule using
fractional exponents, Q.E.D.
In the following probl em set you will pro ve, in general, the prop ert y proved in
Example 3. You will also us e impli cit diff eren tiation to verify the results you
obtain ed by diff erentiating parametric fun ctions from the preceding section .
Problem Set 4-8
DoTheseQuickly
The following probl ems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
y
OJ. If f (x) = cosx, then f(rr) = -? -
02. Differentiate: y = x200 1 !:
2 ... '.-
-:.
03. lim x-s(x - Sx)/(x - 5) = -?-
04. Differentiate: f(u) = cot u X
QS. A definite int egral is a-?- of x and y.
06. y = tan - 1 3x => y ' = - ?-
Ql. If dy /dx = 3x 2, what fun ction could y equal ?
QB. A derivative is an-?-
Figure 4 -Bc
Q9. Sketch the derivative of the function graphed in Figure 4-8c.
010. If the position of a moving object is given by x(t) = t sin t feet, what is th e velocity of the
object when t = 3 sec7
170 Chapter
4: Products,
Quotients
, andParametric
Functions
For Problems 1-20, differentiate implicitly to find y ' in terms of x and y.
1. x 3 + 7y 4 = 13 2. 3x 5 - y 4 = 22
3. x6y 9 + 3x - y 3 = 104 4. 4x + 8x 2 y 6 + y 4 = 21 3
5. x + xy + y = sin 2x 6. COS (xy) = X - 2y
7. X0.5 _ yD.5 = 13 8. x1.2 + y1.2 = 64
9. 4y 2 + 9x 2 = 36 10. 25y 2 - 16x 2 = 400
11. (x 3y4) 5 = X - y 12. (xy) 6 = X + y
13. cos 2 x + sin 2 y = 1 14. sec 2 y - tan 2 x = 1
15. tan (xy) = xy 16. cos (xy) = xy
17. siny = x 18. cosy= x
19. cscy = x 20. coty = x
21. By implicit differentiation, derive the formula for y ' if y = cos - 1 x. Express the answer
explicitly in terms of x .
22. By implicit differentiation, derive the formula for y ' if y = tan - 1 x. Express the answer
explicitly in terms of x.
23. If y = x 1115 , prove that the derivative of a power formula for powers with integer
exponents gives the correct answer for y '.
24. Derivative of a Rational Power: Suppose that'y = x", where n = a /b for integers a
and b. Write the equation y = x 01b in the form y b = x 0 . Then use the derivative of a
power formula for integer exponents to prov e that the formula also works for rational
constant exponents.
25. Circle Problem: Consider the circle x 2 + y 2 = 100 (Figure
4-8d) .
a. Show that the point (- 6, 8) is on the graph.
b. Evaluate dy /dx at the point (- 6,8) . Explain why your
answer is reasonable.
c. Show that the parametric equations
X = 10 COS t
y = 10 sin t
give the same value for dy /dx at (x, y) = (- 6, 8).
26. Hyperbola Problem: Consider the hyperbola x 2 - y 2 = 36
(Figure 4-8e).
Figure 4-Bd
a. Show that the point (10, - 8) is on the graph .
b. Evaluate dy /dx at the point (10, - 8). Explain why your
answer is reasonable .
c. Show that the parametric equations
t +-+-+-J-':
10
y
__, +-+,.. -+-+'. 1
x = 6 sect . ~ . . ... ~- 1 ~... ~
'
y = 6 tan t ..l
give the same value for dy /dx at x = 10.
.j.;
,. :
j
27. Cubic Circle Problem: Figure 4-8f shows the cubic circle
x 3 + y 3 = 64.
a. Find dy /dx at the points where x = 0, x = 2, and x = 4.
Show that your answers are consistent with the graph.
b. Find dy /dx at the point where y = x.
c. Find the limit of dy /dx as x approaches infinity. Figure 4-8e
d. Why do you suppose this graph is called a cubic circle?
28. Ovals of Cassini Project: Figure 4-8g shows the
ovals of Cassini,
[ (x - 6) 2 + y 2 ][(x + 6) 2 + y 2 ] = 1200.
In 1680, Italian astronomer Giovanni Domenico
Cassini (1625-1712) used th ese figures for
relative motions of the earth and the sun.
a. Find th e two values of dy /dx when x = 8.
Show that your answers ar e reasonable. "'!-!-,
J
I
b. Find the four x-intercepts . What does dy /dx
seem to be at these points ? Confirm your Figure 4-8f
conclusion for the largest intercept.
c. Starting with the original equation above, use
the quadratic formula to find an equation for
y 2 explicitly in terms of x. Use this equation
to duplicate Figure 4-8g.
d. Replace the number 1200 in the original
equation with the number 1400, then plot
the graph. In what way or ways is this graph
different from that shown in Figure 4-8g 7
e. Show that for any point on the graph, the
product of its distances from the points Figure 4-8g
(6, 0) and (- 6, 0) is constant.
4-9 Chapter Review and Test
In this chapter you have learned more algebraic techniques for differentiating
functions. You can now differentiate products, quotients, parametric functions, and
implicit relations. You can differentiate implicit relations, and all six trigonometric
functions. The properties are sometimes surprising; for instance, for a product,
(uv)' = u' v + uv ' , not u' v ' . Derivatives of parametric functions and of implicit
relations allow you to analyze geometric figures that are much more comp licated
than those you've dealt with so far.
The Review Problems below are numbered according to the sections of this chapter.
The Concepts Problems allow you to apply your knowledge to new situations . The
Chapter Test is more like a typical classroom test your instructor might give you.
Review Problems
RO. Update your journal with what you've learned since the last entry. Include such things
as those listed here.
The one most important thing you have learned by studying Chapter 4
Which boxes you have been working on in the "define, understand, do, apply" table
Key terms, such as parametric function, implicit differentiation, and differentiab ili ty
Surprising properties, such as the derivative of a product and the parametric chain
rule
Ways in which you have used graphs, tables, algebra, and writing to understand
concepts
Any ideas about calculus that you're still unclear about
Rl. Suppose that x = g(t) = t 3 and that y = h(t) = cost.
a. If f(t) = g(t) h(t), show by counterexample that f ' (t) does not equal g ' (t) h' (t).
b. If f(t) = g(t) / h(t), show by counterexample that f'(t) does not equal g'(t)/h'(t) .
c. Show by example that dy/dx equals (dy/dt)/(dx/dt).
R2. a. State the property of the derivative of a product .
b. Prove the property of the derivative of a product, using the definition of derivative.
c. Differentiate and simplify .
i. f(x) = x 7 cos 3x ii . g(x) = (sinx)(sin2x)
iii. h(x) = (3x - +
7) 5 (5x 2) 3 iv. s(x) = (5 3 ) (x B)
d. Differentiate f(x) = (3x + 8) ( 4x + 7) in two ways.
i. As a product of two functions
ii. By mu ltiplying the binomials, then differentiating
Show that your answers are equivalent.
R3. a. State the property of the derivative of a quotient.
b. Prove the property of the derivative of a quotient, using the definition of derivative .
c. Differentiate and simplify .
. f( ) = sin 1Ox .. ( ) _ (2x + 3) 9 iii. h(x) = (100x 3 - 1) - 5
I. X x5 ll. g X - (9x - 5)~
d. Differentiate y = l /x 10 as a quotient and as a power with a negative exponent.
Show that both answers are equivalent.
e. Find t'(x) if t(x) = sinx. Use the result to find t'(l).
cosx
f. Plot the difference quotient [t(x) - t(l)] / (x - 1), using a window centered at
x = 1, with t.x = 0.001. (Enter t(l) as ( sin 1)/ ( cos 1) to avoid rounding errors .)
Section 49:ChapterRev
iewandTest 173
t~ -
~
Sketch the result. Show that the difference quotient approaches t '(l) by tracing
the graph and making a table of values for several x-values on either side of l.
R4. a. Differentiate.
i. y = tan 7x ii. y = cot(x 4 ) iii. y = 3 sec x iv. y = cscx
b . Derive the formula cot 'x = - csc 2 x.
c. Plot the graph of y = tan x. Make a connection between the slope of the graph and
the fact that the derivative is equal to the square of a function.
d. Suppose that f(t) = 7 sect. How fast is f(t) changing when t = 1? When t = 1.57
When t = l. 5 77 How do you explain the dramatic increase in the rate of change of
f(t) even thought doesn't change very much?
RS. a. Differentiate .
iii. c (x ) = (cos - 1 x) 2
b . Plot on your grapher the graph of y = sin - 1 x. Use parametric mode. Evaluate the
derivative at x = 0, x = 1, and x = 2. How do your results agree with the graph?
R6. a. State the relationship between differentiability and continuity.
b. Sketch a graph for each of the following .
i. Function f is neither differentiable nor continuous at x = c.
ii. Function f is continuous but not differentiable at x = c.
iii. Function f is differentiable but not continuous at x = c.
iv. Function f is differentiable and continuous at x = c.
x 2 +1, ifx < l
c. Let f(x ) = { - x 2 + 4x - 1, if x ~ 1
i. Sketch the graph.
ii. Show that f is continuous at x = l.
iii . Is f differentiable at x = 17 Justify your answer.
1
d Let (x) = {sin - x, ifO :;; x :s; l
g x 2 + ax + b, 1f x < 0
Find the values of the constants a and b that make g differentiable at x = 0.
Confirm your answer by graphing.
R7. a. Figure 4-9a shows the spiral with the parametric
equations
X = (t/TT ) COS t X
y = (t/ rr ) sin t.
Write dy /dx in terms oft . The graph appears to pass
vertically through the point (6, 0). Does the graph
contain this point? If not, why not? If so, is it really Figure 4-9a
vertical at that point? Justify your answer .
17 4 Chapter
4: Products,
Quotients,
andParametric
Funct
ions
b. Ferris Wheel Problem: Figure 4-9b shows a Ferris wheel of
Rotation
diameter 40 ft. Its axle is 25 ft above the ground. (The same
Ferris wheel appeared in Problem 1 of Problem Set 3-8.)
The seat shown is y(t) feet from the ground and x(t) feet \;\
out from the axle. The seat first reaches a high point when
t = 3 sec. The wheel makes 3 rev / min. Given that x and y 1-x(t)-
vary sinusoidally with time, t, as the Ferris wheel rotates
clockwise, write parametric equations for x and y in terms
oft. When t = 0, is the seat moving up or down ? How fast? ,. y( t)
How can you determine these facts? When t = 0, is the seat t
moving to the left or to the right? How fast? How can you
determine this ? What is the first positive value oft for which
!
dy /dx is infinite? Figure 4-9b
RB. a. Find y':y = (12x 113 + 7) 6 .
b. Find dy /dx: y 3 sin (xy) = x 4 5 .
c. Cissoid of Diocles Problem: The cissoid of Diodes in
Figure 4-9c has the equation 4y 2 - xy 2 = x3 (The
word cissoid comes from Greek and means "ivylike.")
In Curves and Their Prop er ties (National Council
of Teachers of Mathematics, 1974) Robert C. Yates
reports that the Greek mathematician Diodes (ca.
250-100 B.C.) used cissoids for finding cub e roots.
i. Find dy /dx when x = 2. Show that your answers
are reasonable .
ii. Find dy /dx at the point (0, 0). Interpret your
answer.
iii . Find the vertical asymptote . Figure 4-9c
Concepts Problems
Cl. Historical Problem: Newton's Method: Suppose you are
trying to find a zero (an x-intercept) of function f, and you
can't solve th e equation f(x) = 0 by using algebra. English
mathematician Isaac Newton (1642-1727) is cre dited with
finding a way to solve such equations approximately by f(x)
using derivatives. This process is called Newton's method.
Figure 4-9d shows the graph of function f for which a zero
is to be found . Pick a convenient value x = x 0 close to the
zero. Calculate Yo = f(x 0 ). The tangent line to the graph of
fat that point will have slope m = f'(x 0 ). lf you extend this
X
tangent line to the x-axis, it should cross at a place x 1 that
is closer to the desired zero than your original choice, x 0 .
By repeating this process, you can find the zero to as man y Figure 4-9d
decimal places as you like!
a. Let m be the slope of the tangent line at the point (x 0 , y0 ) . Write an equation for
the tangent line .
b. The point (x1, 0) is on the tangent line. Substitute these coordi nates for x and y
in the equation from Cla . Use the result to show that x 1 = x 0 - (y 0/m ).
Sect
ion 4-9:Chapter Review
andTest 175
t~"-j
-
~
c. If a tangent line is drawn at (x1,Y1), it will cross the x-axis at Xz. Explain why
f (x 1)
Xz = X1 - {'(Xi).
d. Write a program to use the formula from Cle to calculate values of x iteratively .
The equation for f(x) can be stored in they = menu . You can use the numerical
derivative feature to calculate the value of f'(x). The input shou ld be the value of
x 0 . The program should allow you to press ENT ER, then read the next value of x .
Test the program on f(x) = x 2 - 9x + 14. Start with x 0 = 1, then again with x 0 = 6.
e. Use the program to find the three zeros of g(x) = x 3 - 9x 2 + 5x + 10. Compare
your answers with those you get by using your grapher's built-in solve feature .
f. The equation of the graph in Figure 4-9d is f (x ) = secx - 1.1. Use ewton's
method to estimate the zero shown in the figure. If you start with x 0 = 1, how
many iterations, n, does it take to make Xn + L indistinguishable from Xn on the
calculator?
Chapter Test
Tl. Is the function c(x) = cot 3x increasing or decreasing when x = 5? At what rate?
. . secx - sec 2
T2. If f (x ) = secx, find f'(2 ). Show that the difference quotient x _ approaches
2
f '( 2) by making a table of values from x = 1.997 through x = 2.003 with an x-step of
0.001.
T3. Sketch the graph of a function that has all these features.
The value f( - 2) = 7.
The function f(x) is increasing slowly at x = - 2.
The value f(l) is positive, but f(x) is decreasing fast at x = 1.
The function f(x) is continuous at x = 2 but is not differentiable at that point .
T4. State (without proof) the derivative of a quotient property.
T5. State (without proof) the parametric chain rule.
T6. Use the most time-efficient method to prove that the general linear function,
f(x) = mx + b, is continuous for every valu e of x = c.
For problems T7-Tl4 , find f '( x) or dy /dx . Simplify.
T7. f(x) = sec 5x TS. y = tan 7/3x
d
T9. dX (3 csc llx) no . f(x) = (2x - 5) 6 (5x - 1) 2
Tll. f (x ) = cot 3x Tl2. Sx3y 7 = y L6
cos 2 3x
Tl3 . x = sec 2t Tl4. y = 4 sin - 1 (5x3)
y = tan2t 3
Tl 5. Rotated Ellipse Prob lem: Figure 4-9e shows the graph of
9x 2 - 20xy + 25y 2 - 16x + lOy - 50 = 0.
176 Chapter
4: Products,
Quotients,
andParametric
Functions
The figure is an ellipse that is rotated with respect to the
x- and y-axes . Evaluate dy /dx at both points where x = - 2.
y
Show that each answer is reasonable. : l
-+------,----,----,....: s+-- ~___,.-,--+-
7
Tl6. Let y = x f3 . Transform this equation into an implicit rela- ! .. -~
tion involving only integer exponents. Then differentiate
implicitly to find y ' . Show that your answer agrees with the
answer found directly, using the formula for the derivative
of a power on the original equation.
x 3 + 1, if X :$ 1
Tl7. L et
f(X ) = { a(x - 2) 2 + b, if x > 1
What values of the constants a and b make function f
differentiable at x = 17 Show that the limits of f(x) as x
approaches 1 from the right and from the left can equal Figure 4-9e
each other but that f is still not differentiable at x = 1.
Sectio
n 4-9:Chapter
Review
andTest 177
CHAPTER
5 Definite and
Indefinite Integrals
Cable cars are used for steep streets in San Francisco. For each
100 feet the car goes horizontally, it might rise as much as 20 feet
vertically. The slope of the street is a derivative, dy/ dx, a single
quantity. The differentials dy and dx can be defined separately,
in such a way that their ratio is the slope. These differentials play
a crucial role in the definition, computation, and application of
definite integrals.
179
~I Mathematical Overview
In your study of calculus so far you have learned that a definite
integral is a product of x and y, where y varies with x. In Chapter 5
you will learn the formal definition of definite integral, that an
indefinite integral is an antiderivative, and that the two are related
by the fundamental theorem of calculus . You will gain this knowledge
in four ways.
Graphically The logo at the top of each even- f (x)
numbered page of this chapter shows
how a definite integral can be analyzed
by slicing a region into vertical strips .
Numerically x f(x) integral
1.0 5 0.0
1.5 8 2.5
2.0 10 7.0 X
a b
Algebraically r
1
COS X dx = sin 4 - sin 1, the fundamental theorem.
Verbally Finally I found out why definite and indefinite integrals share a name.
Indefinite integrals can be used to calculate definite integrals exactly.
I also learned a way to analyze many different problems involving
the product of x and y by slicing the region under the graph into
vertical strips.
180
5-1 A Definite Integral Problem
So far you have learned three of the four concepts of calculus. In this chapter you
will learn about the fourth concept - indefinite integral - and why its name is so
similar to that of definite integral. In this section you will refresh your memory
about definite integrals .
OBJECTIVE Work the prob lems in this section, on your own or with your stu dy group, as an assignment
after your last test on Chapter 4.
Exploratory Problem Set 5-1
Oil Well Problem: An oil well that is 1000 ft deep is to be extended
to a depth of 4000 ft. The drilling contractor estimates that the cost
in dollars per foot, c(x), for doing the drilling is c(x) $/ ft
c(x) = 20 l.0003x,
80
where x is the number of feet below the surface at which the drill is
operating. Figure 5-la shows an accurate graph of function c. 60
l. How much does it cost per foot to drill at 1000 feet? At 4000 ft? 40
2. The actual cost of extending the well is the definite integral of 20
c(x) from x = 1000 to x = 4000. Estimate this integral by using
X ft
the trapezoidal rule with n = 6 increments .
1000 4000
3. The average cost per foot is equal to the total cost from Problem 2
divided by the number of feet that were drilled (3000). Calculate Figure 5-1 a
the average cost per foot.
4. Estimate the definite integral again, this time assuming that each of the six strips is
a rectangle instead of a trapezoid . Let the altitudes of the rectangles be the costs per
foot at the midpoints of the six strips (that is, c(l 2 50), c(l 750), c(22 50), and so on).
How close is your answer to the answer you got by using the trapezoidal rule?
5. Is the average cost per foot for extending the well from 1000 ft to 4000 ft equal to,
more than, or less than the average of the two values in Problem 17
6. What is the approximate instantaneous rate of change of the cost per foot at x = 1000?
At x = 4000? What are the units of th.is rate 7
7. The answer to Problem 6 is a rate of change of a rate of change. What mathematical
word identifies such a rate7 What other physical quantity is a rate of change of a rate
of change?
Sec
tion5-1: A Definite Integral Prob
lem 181
5-2 Review of Antiderivatives
Recall from Section 3-9 that function f is said to be an antiderivative of another
function g if and only if g (x ) is the derivati ve of f (x ) . For instan ce, if
g( x ) = 5x 4 and f (x ) = x5,
th en f is an antid erivativ e of g because f '( x ) and g (x ) both equal 5x4. Note that
x 5 + 7, x 5 - 13 .2, and x 5 + rr ar e also antid erivativ es of g (x ) be caus e th e derivativ e
of a constant equals zero . In gen eral, if f (x ) is an antiderivative of g (x ) , then so
is f (x ) + C, wh er e C is an arbitrary constant. For r easons you ,,viJ.lsoon learn, the
constant C is called the constant of integration. The equation f (x ) = x 5 + C is called
the general equation for the antiderivative. An equation such as f (x ) = x 5 - 1001,
wher e a p articular value has been substituted for the constant of integration, is
called a particular equation for the antiderivative.
In this section you will refresh your memory about ho w to find an antiderivative .
OBJECTIVE Given the equation for the derivative of a function, write the general equation for the
antiderivative function.
Example 1 If f ' (x ) = x 7, find the general equation for f (x ) .
Solution f (x ) = x 8 +C Because f' (x ) = x7.
Problem Set 5-2
DoTheseQuickly
The following problems ar e intended to refr esh your skills . You should b e able to do all
ten probl ems in less than five minut es .
y
01. Find dy /dx: x 3 y 2 = 5.
Q2. Find dy / dx : x = cost and y = st - 1.7.
Q3. Differentiat e: f (x) = 3 7
04. Find f '(x): f (x) = sin - 1 x.
QS. Find y ' : y = secx.
06. Find th e se cond derivati ve: y = x 6 . X
Ql. Find the velocity: x ( t ) = cos 3t. --~-
t
QB. Find the acceleration: x(t ) = cos 3t. ' '
~ .. ;-:
Q9. log 5x + log 3x = lo g - ?-
010. Sketch th e graph of y' for y = lx l + 2, shown in Figure 5-2a. Figure 5-2a
182 Chapter
5: Definite
andIndefinite
Integrals
Write the general equation for the antiderivative function .
l. f'(x) = 7x 6 2. f'(x) = 10x 9
3. f '(x) = x - 9 4. f'(x) = x- 1066
5. f '(x) = cosx 6. f'(x) = sinx
7. f '(x) = csc 2 x 8. f '(x) = secxtanx
9. f'(x) = -csc x cot x 10. f'(x) = sec 2 x
11. f'(x) = sin 5x 12. f'(x) = cos 4x
13. f'(x) = sec 2 8x 14. f '(x) = csc 2x cot 2x
15. f'(x) = (4x + 5) 7 16. f'(x) = (8x + 3) 5
17. Family of Functions Problem: Figure 5-2b shows the graphs of
three functions.
5
y = X + 0.3 X
5
y = X + 0. 7
y = X5 + 1.1
a. Each of the functions is an antiderivative of the same function.
Write an equation for that function. Figure 5-2b
b. What word from geometry (beginning with the letter c) describes the three graphs?
In what way could it be said that the graphs are the same distance apart from one
another? Why is it not strictly correct to say that the graphs are "parallel" to on e
another?
c. Suppose that f'(x) = 3x 2 - 2. Write the general equation for the antiderivative f(x ).
Then write the equations for three particular solutions, the first containing the
point (1, 0), the second containing the point (1, 1), and the third containing the
point (1 , 2).
ct. Plot the graphs of the thre e particular solutions in 17c. Sketch the results.
e. Write a paragraph in your journal that desc ribes how the constant of int egration, C,
aff ects th e graph of the antiderivative. Explain why an antiderivative can be called
a family of functions .
5-3 Linear Approximations and Differentials
You have learned how to find an equation for an antiderivative by asking "What did
they differentiate to get this function?" but nowhere have you seen a symbol for an
antiderivative.
Sec
tion5-3:LinearApproximations
and Differentials 183
In this section you will learn how to find a linear
function that goes throug h a fixed value on the
graph of a given function and that approximate ly
fits the given function when x is close to that fixed
valu e. If you pick a point on the linear function
gra ph, the changes in x and y from the fixed
point are called differentials of x and y, written
dx and dy (Figure 5-3a). Differentials will appear
in the ant id erivative sym bol, as you will see in
X
Section 5-4.
Figure 5-3a
OBJECTIVE Given the equation for a function f and a fLxedpoint on its graph, find an equation for
the linear function that best fits the given function. Use the equation to find approximate
values for f (x) and values for the differentials dx and dy.
For the linear function y = L(x ) = mx + b to fit a given function f close to x = c, two
criteria should be met .
The fun ction valu es should be equal at x = c. That is, I(c) = f(c).
The slopes should be equal at x = c. That is, I '(c) = f'(c).
Example 1 shows you a way to meet these criteria and find how good the fit is.
Example 1 For f(x) = 15 - x 3, find an equat ion of the linear function that best fits f at the
fixed po int (2, 7). Sketch the graphs off and of the linear function . Find th e error in
approximat ing f(x) with the linear function for values of x close to 2.
Solution f(2) = 15 - 23 = 7, soy = 7 when x = 2.
f'(x) = -3x 2 ~ f'(2) = - 12, so m = - 12 .
.. the line has the equation y - 7 = - 12(x - 2) ~ y = 7 - 12(x - 2).
Figure 5-3b shows the graphs of f and of the linear function. Using your
grapher's table feature gives the following values of f(x) and y, and the error,
f(x) - y.
yor f(x\ X f (x) y error, f (x ) - y
1 '\ Y 1.7 10.087 10.6 - 0.513
'" 1.8 9.168 9.4 - 0.232
1.9 8.141 8.2 - 0.059
X
2.0 7 7 0
2.1 5.739 5.8 - 0.061
2.2 4.352 4.6 - 0 .248
Figure 5-3b 2.3 2.833 3.4 -0.567
The linear function fits f(x) perfectly at x = 2, and y "" f(x) when x is close to
2. The error, f (x) - y, gets larger in absolute valu e as x gets farther from 2.
184 Chapter
5: Definite
andIndefinite
Integrals
Parts of the lin ear equation y = 7 - 12 (x - 2) in Example 1 are significant.
7 is f(2), the value of the original function at the fixed point.
- 12 is f '(2), the value of the derivative there.
(x - 2) is the differential dx, the x-displacement from the fixed point (2, 7).
-12 (x - 2) is the differential dy, they-displacement from the point (2, 7) for the
linear function.
Property: Linearizationof a Function
The linear function that best fits function f for values of x close to x = c is
y = f(c) + f'(c)(x - c) or, equivalently, y = f(c) + f'(c) dx.
Example 2 If f(x) = sin x, write an equation for the linear function that best fits at x = l. Use the
linear function to approximate f(l.02). What is the error in the approximation? What
are the values of dx, dy, 6.x, and t.y? Sketch a graph that shows these quantities
and the error.
Solution f(l) = sinl = 0.84147 .. .
f'(x) = COS X => f'(l) = COS 1 = 0.540 30 ...
:.y = (sin 1) + (cos 1) (x - 1) y = f (c) + f'(c)(x - c)
y = 0.84147 ... + 0.54030 ... (x - 1).
If x = 1.02, then
y = 0.84147 ... + 0.54030 ... (1.02 - 1) = 0.8 52277030 ...
f(l.02) = sinl.02 = 0.852108021 .. .
Small error.
:. error = sin 1.02 - 0.852277030 ... = - 0.000169008 ...
1.02 is close to l.
dx = l.02 - 1 = 0.02
dy = (cos 1) (0.02) = 0.01080604 ...
0.X = dx = 0.02
t.y = sin 1.02 - sin 1 = 0.01063 703 ... ::::::
dy.
Figure 5-3c illustrates the diff eren- yand f (x)
tials dx and dy, and shows their Erro r
relationship to t.x and 6.y. The error
is also equal to t.y - dy.
""'xf. _,
/
/
sin 1.02
dy
r i j
sin 1 -----;;:,; -----1-
1 X
~_....__l _l_.___, .02
Figure 5-3c
Section
5-3: LinearApproximations
andDifferentials 185
In Example 2, dy = (cos 1) (x - 1). In general, dy is given by
dy = f '(x) dx.
This fact can be us ed as a definition for the differential dy.
Definition:Differentials
The differentials dx and dy are defined as follows.
dx = t:.x
dy = f'(x) dx
Thus dy -c-dx is equal to f '( x). Note that dy is not usually equal to t:,y.
An important aspect of the definition of differ entials is that it allows for dy and dx
in th e symbol for derivative,
dy
dx'
to be treated as separate quantiti es.
Examp le 3 shows you how to find the antiderivative when the differential of a
function is given.
Example 3 Given dy = (3 x + 7) 5 dx, find an equation for th e antiderivative, y.
Solution dy = (3x + 7) 5 Divide b oth sides of th e given equ ation by dx.
dx
Thought process :
It looks like someo ne diff ere ntia ted (3x + 7) 6
But th e diff erential of (3x + 7) 6 is
6(3x + 7) 5 3 dx, or 18(3x + 7) 5 dx.
So th e function mu st be on.ly l / 18 as big
as (3x + 7) 6 , and th e answe r is:
y = fs(3 x + 7) 6
+C Why is +C needed?
The problems in th e following problem set are design ed to help you become
comfor tabl e with linear approximations for functions and with th e differentials dy
and dx.
Problem Set 5.3
DoTheseQuickly
The following problems ar e intended to r efr esh your skills . You should be able to do all
ten problems in less than five minutes .
Ql. Sketch a graph that illu strates the meaning of definite int egral.
Q2. Write the physical meaning of derivative.
Q3. Differentiat e: f (x ) = 3x - 5
186 Chapter
5: DefiniteandIndefin
ite Integrals
Q4. Find the antiderivative: y ' = cos x.
QS. If y = tan t meters, how fast is y changing when t = rr/3 sec?
Q6. Find limt>x- o[f(x + .6x) - f(x)] /.6x if f(x) = secx.
Ql. Find limx- o secx.
QB
. What is the limit of a constant?
Q9. What is the derivative of a constant?
QlO. If limx -c g(x) = g(c), then g is - ?- at x = c.
1. For f(x) = 0.2x4, find an equation of the linear function that best fits f at x = 3. What
is the error in this linear approximation of f(x) if x = 3.1? If x = 3.001? If x = 2.999?
2. For g(x) = secx, find an equation of the linear function that best fits at x = rr/3.
What is the error in this linear approximation of g(x) if dx = 0.04? If dx = - 0.04? If
dx = 0.001?
3. Local Linearity Problem I: In Problem Set 3-2, you learned that
/
the function f(x) = x 2 has the property of local linearity at any /
point, such as x = 1. Figure 5-3d shows the graph off and the
linear function that fits best at x = 1. Write an equation for this
X
linear function. Plot f and the linear function on the same screen,
/
then zoom in repeatedly on the point (1, 1). Tell why the words /
local lin ear ity describe the relationship between the linear graph
and the graph of f when x = 1. Figure 5-3d
4. Local Linearity Problem II: Figure 5-3e shows the graph of
f(x) = x 2 - O.l(x - 1) 113.
Explore the graph for x close to 1. Does the function have local
linearity at x = 1? Is f differentiable at x = 17 If f is differentiable X
at x = c, is it locally linear at that point? Is the converse of this 1
statement true or false? Explain.
5. Steepness of a Hill Problem: On roads in hilly areas, you Figure 5-3e
sometimes see signs like this.
Steep hill 20% grade is shown.
20% grade
The grade of a hill is the slope (rise/ run) written as a
percentage, or, equivalently, as the number of feet the hill
rises per hundred feet horizontally. Figure 5-3f shows the
latter meaning of grade .
Figure 5-31
a. Let x be the grade of a hill. Explain why the angle, 0
degrees, that a hill makes with the horizontal is given
by
0 = isotan - 1~
TT 100'
Sect
ion 5-3: Linear
Approximations
andDifferentials 187
b. Find an equation for de in terms of x and dx. Then find
de in terms of dx for grades of x = 0%, 10%, and 20%.
c. You can estimate e at x = 20% simply by multiplying
de at x = 0 by 20 . How much error is there in the value
of e found by using this method rather than by using
the exact formula that involves the inverse tangent
function?
d. A rule of thumb you can use to estimate the number
of degrees a hill makes with the horizontal is to divide
the grade by 2. Where in your work for Sc did you
divide by approximately 2? When you use this method
to determine the number of degrees for grades of 20%
and 100%, how much error is there in the number?
6. Table of Differentials Problem: Figure 5-3g shows both the graph y
of y = x 3 and a tangent line drawn at x = l . Plot this diagram
on your grapher . Use TABLEor TRACE to make a table of values
of t,x, t,y, and dy and of the difference between t,y and dy for
values of x on both sides of 1. Pick some values of x that are very
close to 1 and some that are fairly far away. How well does dy
approximate tly for x close to 1 and for x farther away from 1?
For Problems 7-24, find an equation for the differential dy . Figure 5-3g
7. y = 7x 3 8. Y = - 4x 11
9. y = (x.j + 1) 7 10. y = (5 - 8x).j
2
11. y = 3x 2 + Sx - 9 12. y = X +X +9
13. y = - sx - 1.7 14. y = 15x 113
15. y = sin3x 16. y = cos4x
17. y = tan 3 x 18. y = sec 3 x
19. y = 4x cosx 20. y = 3xsinx
2 x3 X
X X
21. y =---+ 2 22. y =---+ 6
2 4 3 5
23. y = cos (sec x) 24. y = sin(cscx)
For Problems 25-38, find an equation for the antiderivative y.
25 . dy = 20x 3 dx 26. dy = 36x 4 dx
27. dy = sin4xdx 28. dy = cos 0.2x dx
29. dy = (O.Sx - 1)6 dx 30. dy = (4x + 3) - 6 dx
31. dy = sec 2 x dx 32. dy = cscxcotxdx
33. dy = 5 dx 34. dy = - 7dx
188 Chapter5: Definite and Indefinite Integrals
35 . dy = (6x 2 + lOx - 4) dx 36 . dy = (10x 2 - 3x + 7) dx
3 7. dy = sin 5 x cos x dx (Be clever!) 38. dy = sec 7 x tanx dx (Be very clever!)
For Problems 39 and 40, do the following.
a. Find dy in terms of dx.
b . Find dy for the given values of x and dx.
c. Find lly for the given values of x and dx.
d. Show that dy is close to lly.
39. y = (3x + 4)2(2x - 5)3, x = 1, dx = - 0.04
40. y = sin 5x, x = rr/3, dx = 0.06
5-4 Formal Definition of Antiderivative and
Indefinite Integral
You have learned that an antiderivative is a function whose derivative is given. As
you will learn in Section 5-8, the antiderivative of a function provides an algebraic
way to calculate exact definite integrals . For this reason an antiderivative is also
called an indefinite integral . The word indefinite is used because there is always a
"+C" whose value is not determined until an initial condition is specified .
Relationship:AntiderivativeandIndefiniteIntegral
Indefinite integral is another name for antiderivative.
ln this section you will learn the symbol that is used for an indefinite integral or an
antiderivative.
OBJECTIVE Become familiar with the symbo l used for an indefinite integral or an antiderivative by
using this symbol to evaluate indefinite int egrals.
In Section 5-3, you worked problems such as "If dy = x 5 dx, find y." Indefinite
integration, which you now know is the same thing as antidifferentiation, can be
considered to be the operation performed on a differential to get the expression for
the original function. The integral sign used for this operation is a stretched-out S
like that shown here.
f
Sec
tion 5-4: Forma
l Definition
of Antider
ivativeandIndef
inite Integral 189
As you will see in Section 5-5, the S shape comes from "sum ." To indicate that you
want to find the indefinite integra l of xs dx, write
f XS dx .
The whole expression, f xs dx, is the integral. The function x 5 "inside" the integral
sign is called the integrand. These words are similar, for example, to the words
radical, used for ft , and radicand, used for the number 7 inside. Note that although
the dx must appear in the integral, only the function xs is called the integrand.
Writing the answer to an indefinite integral is called eva luating it or doin g the
inte gration . Having seen the techniques of the last two sections, you should
recognize that
f xs dx = x + C.
6
From the discussion above, the formal definition can be understood .
Definition:IndefiniteIntegral
g(x) = f f(x) dx if and only if g'(x) = f(x).
That is, an indefinite integra l of f(x) dx is a function whose derivative is f(x). The function
f (x) inside the integral sign is called the integrand.
Words: The express ion f f(x) dx is pro nounced "The integra l of f (x) with respec t to x. "
Notes: An indefinite integral is the same as an antideriva tive. The symbol f is an operator,
like "cos" or the minus sign, that acts on f(x) dx .
Another symbol for g'(x) = f(x) is :x Jf(x) dx = f(x).
Integralof a ConstantTimesa Function
andof a Sumof SeveralFunctions
To develop a systematic way of integrating, it helps to know some properties of
in definite integrals, like those shown below .
f Scos x dx and f (x + sec x - x) dx
5 2
=Sfcosxdx = f xs dx + f sec x dx - f x dx
2
= 5 sinx + C = l6 x 6 + tan x - l2 x 2 + C
In the first case, 5, a constant, can be multiplied by the answer to f cos x dx. In the
second case, the integral of each term can be evalu ated separately and the answers
can be added toge ther. Becau se of the "only if" part of the definition of indefinite
integra l, all you have to do to prove these facts is differentiate the answers . The
following is a proof of the integral of a constant times a function property . You will
prove the integral of a sum property in Problem 45 of Problem Set 5-4.
190 Chapter
5: Definite andIndefinite Integrals
Property If k stands for a constant, then Jk f(x) dx = k Jf(x) dx .
Proof Let g(x) = k Jf(x ) dx .
Derivativ e of a co nsta nt time s a
Then g ' (x) = k :x (Jf(x) dx) function.
Definition of ind efinit e inte gral
= kf(x) (the "only if" part).
.. g(x) = Jkf(x) dx
Definition of indefinite int egral
(the "if" part).
Transitive property of
.'. J k f(x) dx = k J f(x ) dx, Q.E.D. equality.
TwoPropertiesof IndefiniteIntegrals
Integral
of a Constant
Timesa Function:
If f is a function that can be integrated and k is a
constant, then
f k f(x) dx = k f f(x) dx .
Words: "You can pull a constant multiplier out through the integral sign."
Integral : If f and g are functions that can be integrated, then
of a Sumof TwoFunctions
f (f(x) f f
+ g(x)) dx = f(x) dx + g(x) dx.
Words : "Integration distributes over addition."
Be sure you don't read too much into these properties! You can't pull a variable
through the integral sign. For instance,
Jx cos x dx does not equal x Jcos x dx.
The integral on the right is x sinx + C. Its differential is (sin x + x cos x) dx, not
x cos x dx. The integral f x cos x dx is the integral of a produ ct of two func-
tions . Recall that the derivative of a product does not equal the product of the
derivatives.
The"dx"inanIndefinite
Integral
There is a relationship betw een the differential, dx, at the end of th e integral and
the argument of the function in the integrand. For example,
f cosxdx = sinx+C.
Sectio
n 5-4: FormalDefinitionof Antiderivat
iveandIndefinite
Integra
l 191
It does not matter what lett er you use for the variable.
f cos r dr = sin r + C
f cost dt = sin t + C
f cos u du = sin u + C
The words "with respect to" id entify the variable in the differential that follows the
integrand function . Note that in each case, dx, dr, dt, or du is the differential of the
argument of the cosine. This observation provides you with a way to integrate some
compos it e functions .
Example 1 Evaluate: f 5 cos (Sx + 3) dx
Solution f cos(Sx + 3)(5dx) Commu te th e 5 on th e left, and associate it with the dx.
= sin (Sx + 3) + C The differenti al of the insid e function, Sx + 3, is 5 dx.
Example 2 Evaluate: f (7x + 4) dx 9
Solution This is the integral of the ninth power function . If th e dx were the differential
of the inside function, 7x + 4, then the int egral would have the form
f Un du .
Transform the integral to make the dx the differential of (7 x + 4). When you do
so, the work looks like this.
f (7x + 4) dx 9
= f (7x + 4) 0 7 dx)
9
Multiply by 1, using th e form (1/7)(7).
= f ~( 7x + 4) ( 7dx) 9 Assoc iat e 7 dx, the diff eren tial of th e in side fun ction.
Commute 1/ 7.
= ~ f (7x + 4) ( 7 dx )
9 Integra l of a constant times a function . (Pull the 1/ 7 out
through the integral sign .)
= .!.
7
...!...(7x+4)
10
10 +c
Integra l, f u" du .
=fri( 7x + 4) 10 + c
Once you understand the process you can leave out some of the steps, thus
shortening your work. In Problem Set 5-4, you will practice evaluating indefinite
integrals.
192 Chapter
5: DefiniteandIndefiniteIntegrals
Problem Set 5-4
DoThe
se Quickly
The following problems are int end ed to refresh your skills. You should be able to do all
ten problems in less than five minutes .
QI. Find the antiderivative: 3x 2 dx.
Q2. Find the indefinite integral: x 5 dx .
Q3. Find the deri vative: y = x 3 .
04. Differentiate: y = (1/6)x6
QS. Integrate: cos x dx
Q6. Differentiate : y = cos x
QJ. The "if" part of a theor em is called the - ?- .
QB. The "then" part of a th eorem is called the - ?- .
Q9. If h(x) = r ' (x), then r is a(n) -?- or a(n) -?- of h .
QIO. lim j _ 0 (sinJ )/ (j) = - ?- .
For Problems 1-42, evaluat e the indefinite integral.
1. J 6x 5 dx 2. J Sx4 dx
3. Jx 10 dx 4. Jx 20dx
5. J4x - 6 dx 6. J 9x - 7 dx
7. J 102t 41 dt 8. J 72r - u dr
9. f 30p - 2/5 dp 10. J 56v- 317dv
- 11. f cosx dx 12. f sin x dx
13. Jsin 3m dm 14. Jcos Su du
15. J4 cos 7x dx 16. J20sin 9x dx
17. J(4v + 9) dv 2 18. J(3 p + 17 ) 5 dp
19. J(8 - 5x) dx 3 20. J(20 - x ) 4 dx
21. f (6 + 7b ) - 4
db 22. J(10 + 13t) - 6 dt
23. J(sinx )6 cosx dx 24. J(cos x)8 sin x dx
25. Jcos 4 e sin e de 26. J sin 5 0co s 0d0
27. Jsin 3 rrx cos rrx dx 28. J cos 8 rrx sin rrx dx
29. f (x 2 + 3x - 5) dx 30. J(x 2 - 4x + 1) dx
Sect
ion5-4: Formal
Definition
of Antiderivotive
andIndefinite
Integral 19 3
~a b
31. f (r - 2 + r 2 ) dr 32. f (u 3 - u- 3 ) du
2
33. f(x + 5) 3 dx (Beware!) 34. f (x 3 - 6) 2 dx (Beware!)
35. f sec 2 x dx 36. f csc 2 x dx
37. f csc 3x cot 3x dx 38. f sec 5x tan 5x dx
39. f tan 7 x sec 2 x dx 40 . f cot 8 x csc 2 x dx
41. f csc 9 x cot x dx (Be clever!) 42 . f sec7 xtanxdx (Be clever!)
43 . Distance from Velocity Problem: As you drive
along the highway, you step hard on the
accelera tor to pass a truck (Figure 5-4a).
Assume that your velocity, v(t) feet per (~} =o
second , is given by -- -------- ---
v(t) = 40 + 5./f ,
l[=:] :---------~c::Jl I
,------H ow far 7--- - -
Start to Fini sh
where t is the number of seconds since you pass pa ssin g
started accelerating . Find an equa tion for
D(t), your displacement from the starting Figure 5-40
point, that is, from D (O) = 0. How far do you
go in the 10 sec it takes to pass the truck?
44. Definite Int egra l Surpr ise! Figure 5-4b shows the region that f(x)
represents the definite integral of f(x) = 0.3x 2 + 1 from x = 1 to 5
X = 4.
a. Evaluate the integral by usi ng the tra pezoidal rule with
n = 100 increments.
b. Let g(x ) = J f(x) dx. Integrate to find an equa tion for g(x) .
c. Evaluate the quantity g(4) - g(l). What is int eresting abou t Figure 5-4b
your answer?
45 . Integral of a Sum Property: Prove that if f and g are functions that can be integrated,
then J (f(x) + g(x)) dx = J f(x) dx + J g(x) dx.
46. Integral Table Problem: Calvin finds the form ula
J x cos x dx = x sin x + cos x + C in a table of integrals. v(t)
Phoebe says, "That's right!" How can Phoebe be sure 30
the for mul a is right?
'.'l3_.?L ________
_
*47. Introduction to Riemann Sums: Suppose the velocity 20
of a moving object is given by '.'R~l_______
2
v (t) = t + 10,
10
where v(t) is in feet per minute and tis in minutes.
Figure 5-4c shows th e region tha t represents the
integral of v(t) from t = 1 tot = 4. Thus the area of
2 3 4
th e re gion equals the distance trave led by the object.
Figure 5-4c
*This problem pr epares you for the next sect ion.
194 Chapter
5: DefiniteandIndefin
ite Integrals
In this problem you will find the integral, approximately, by dividing the region into
rectangles instead of into trapezoids . The width of each rectang le will still be M, and each
rectangle's length will be v(t) found at the t-value in the middle of the strip. The sum of
the areas of these rectangles is called a Riemann sum.
a. Use a Riemann sum with n = 3 strips, as shown in v(t)
Figure 5-4c, to find an approximation for the definite 30
integral of v(t) = t 2 + 10 from t = 1 to t = 4.
b. Use a Riemann sum with n = 6 strips, as shown in
20
Figure 5-4d, to find another approximation for the
definite integral. The altitudes of the rectangles will be
values of v(t) fort at the midpoints of the intervals, 10
namely, t = 1.25, t = 1.75, t = 2.25, t = 2.75, t = 3.25,
and t = 3. 75. The widths of the strips are, of course,
tit = 0.5.
2 3 4
c. Explain why the Riemann sum you used in 4 7b should
be a better approximation for the int egral than that Figure 5-4d
used in 47a.
d. Make a conjecture about the exact value of the definite integral.
e. Check the conjecture you made in 47d by evaluating the integral using the
trapezoidal rule with n = 100 increments.
f. How far did the object travel betw een t = 1 min and t = 4 min ? What was its average
velocity for that time interval?
48 . Journal Problem: Update your journal with what you've learned since your last entry.
Include such things as those listed here.
The one most important thing you've learned since your last journal entry
The difference, as you understand it so far, between definite integral and indefinite
integral
The difference between a differential and a derivative
Any insight you may have gained about why two different concepts are both named
integra l, and how you gained that insight
Algebraic techniques you have learned for finding equations for indefinite integrals
Questions you plan to ask during the next class period
5-5 Riemann Sums and the Definition of
Definite Integral
Recall that a definite integral is used for the product of f(x) and x , such as
(rate)(time). Thus the iutegral is equal to the area of the region under the graph of
f, as shown in Figure 5-5a. The trapezoidal rule lets you find definite integrals by
slicing the region under a graph into strips, then by approximating the area of each
strip with the area of a trapezoid.
Sec
tion 5-5: Riemann SumsandtheDefinition of Defini
te Integral 195
Another way to estimate a definite integral is f (x)
to slice the region into strips and approximate
the strips with rectangles instead of with trape-
zoids. Figure 5-Sb shows the region in Figure
5-Sa sliced into n strips of variable width t-.x1 ,
t..x 2, C-.x3 , ... (n = 6, in this case). These strips
are said to partition the interval [a, b] into n
subintervals, or increments. Values of x (called
c 1, c 2, c 3 , ... ) are picked so that one value is
in each (closed) subinterval. These x-values are X
called sample points. a b
At each of the sample points, the corresponding
Figure 5-5a
function values f(c1), f(c 2), f(c 3), ... are the
altitudes of the rectangles. The area of any one
rectangle is thus
Ar ecr = f(ck) t..xk,
where k is 1, 2, 3, ... , up to n. The integral is
approximately equal to the sum of the areas of
the rectangles.
n
I
Area ::::: f( cd t..xk
X
k=l
atc ct c3
t b
c4 c5 c6
1 2
This sum is called a Riemann sum (pronounced Sampl e point s
"ree-mahn," with the accent on the first A Riemann sum
syllable) after G. F. Bernhard Reimann (1826-
1866). Figure 5-5b
Definition:RiemannSum
A Riemann sum, Rn, for function f on the interval [a, b] is a sum of the form
n
Rn = L f(cd l-.Xk,
k=l
where the interval [a, b] is partitioned into n subintervals of widths t:.xk, and the numbers
{ckl are sample points, one in each subinterval.
If each sample point is picked so that f(c) is the lowest in its respective subinterval,
then each rectangle has an area that is less than the area of the strip . In this case,
the Riemann sum is called a lower sum. Similarly, an upper sum is a Riemann sum
with each sample point taken where f(c) is the highest in its respective subinterval.
A midpoint sum is formed by choosing each sample point at the midpoint of the
respective subinterval. An upper sum is an upper bound for the area of the region,
and a lower sum is a lower bound. A lower Riemann sum, an upper Riemann sum,
and a midpoint Riemann sum are shown in Figure 5-Sc. Equal values of t.x have
been used.
196 Chapter
5: DefiniteandIndefiniteIntegrals
1 It
f'(x) f'(x) f'(x)
X X X
a b a b a b
Samp le polnts Sample poin ts Samp le point s
Ln: lower Riemann sum Un: upper Riemann sum Mn: midpoint Riemann sum
(inscribed rectang les) (circumscribed rectangles) (intermediate rectangles)
Figure 5-5c
The symbols Ln, Un, Mn, and R 11 are us ed for lower, upp er, midpoint, and genera l
Riemann sums, respec tively. Suppose that the limits of L,, and Un are equ al to each
oth er as the larg est value of Lix approaches zero . If this is the case, function f is said
to be integrabl e on the interval [a, b]. The common limit of the upper and lower
sum s is de fined to be the definite int egral off on the int erval [a, b] . The integral
sign (a stretc hed S for "sum") with th e a and b of [a, b] attached to it is used for a
definite integra l, like this .
J,rab The definit e int egra l sign.
Any Riemann sum , Rn, for a given part ition of [a, b] is bounded by the upp er and
lower sum .
Thu s if func tion f is integrab le on [a, b], then any Riemann sum will also have
the definite integra l as its limit . This resul t is a dir ect application of th e sque eze
th eorem .
Definitions:DefiniteIntegralandIntegrability
If the lower and upp er sum s for function f on the interval [a , bl have a common limit as
6.x approach es 0, then f is said to be integrable on [a, b]. This common limit is defined
to be the definite integral of f from x = a to x = b. The numb ers a and b are called lower
and upp er limits of integration , or bounds of integration. In symbols,
f.ab f (x) dx = lim Ln = lim Un.
nx-0 nx- 0
Not e: The differential symbol d x is used in definite int egrals instea d of 6.x .
Section5-5: RiemannSum
s andtheDefinition of Definite Integral 197
I t I;:\
-Hf+
Property: Limitof a RiemannSum
If f is integrable on [a, b], and if Rn is any Riemann sum for f on [a, b], then
f
a
b f(x) dx = lim Rn.
Ll.x-0
OBJECTIVEUse Riemann sums to find approximate values of definite integrals.
Example 1 For the integral r (1 / x) dx, do the following.
a. Find U6 for the integral.
b . Find 1 6 for the integral.
c. Find M6 for th e integral. Show that M6 is betwe en 1 6 and U6 .
d. Find T6 , the integral found by application of the trape zoidal rul e using six
increments . Show that T6 is also between 1 6 and U6 . Is T6 higher or lower than
the actual integral? Explain.
Solutions a. Sketch the graph and the six subintervals of width ~x = 0.5 (Figure 5-5d).
Becaus e f (x) is decreasing, the left ends of th e subintervals will give the
highest values of f(x). Pick the sample points at the left ends.
f(x) X = C 1/x (1/x)(t.x)
1.0 1 0.5
1.5 0.6666666 .. . 0.33 33333 . . .
2.0 0.5 0.25
2.5 0.4 0.2
3.0 0.3333 333 . .. 0.166 6666 ...
3.5 0.2857142 ... 0.1428571 ...
U5 = 1.5928571 ...
X
2 3 4
Upper Riemann sum
Figure 5-5d
b. For the lower sum, the sample points should be taken at the right ends of
the subintervals (Figure 5-5e).
198 Chapter
5: Definite
andIndefinite
Integrals
f(x) X=C 1/x (1/x)(t.x)
l 1.5 0.6666666 . .. 0.3333333 ...
2.0 0.5 0.25
2.5 0.4 0.2
3.0 0.3333333 ... 0.1666666 .. .
3.5 0.2857142 ... 0.1428571 . . .
4.0 0.25 0.125
L5 = 1.2178 571 ...
2 3 4
Lower Riemann sum
Figure 5-5e
c. For the midpoint sum, use sample points c = 1.25, c = 1.75, c = 2.25, and so
on (Figure 5-5f).
f (x) X= C 1/x
1.25 0.8
1.75 0.5714285 .. .
2.25 0.4444444 ...
2.75 0.3636363 .. .
3.25 0.3076923 . ..
3.75 0.2666666 ...
Sum= 2.7538683 . . .
M5 = 0.5(2.7538683 ... ) = 1.3769341 . . .
X
2 3 4
Midpoint Riemann sum
Figure 5-5f
The tabl e above shows a shortcut that allows you to take advantage of the
fact that all the 6x's are equal. See if you can figure out why it works! The
midpoint sum is betw een the lower and upp er sums,
1.2178571 . .. < 1.3769341. .. < 1.5928 571 . ...
d. By using your trap ezo idal rul e program, you should find that T6 =
1.40 5357 .... This value is also between the lower and upper sums.
1.21785 71 ... < 1.405 357 ... < 1.5928571. ..
The trapezoidal sum T6 is higher than the actual value of the integral. Figure
5-5g shows that the trapezoid is circumscribed around th e outside of the
strip, thus the trape zoid has a great er area. This happens because the graph
is "concave upward," a property you will explor e in Chapter 8. The sum of
Sec
tion 5-5: Riem
annSumsandthe Defini
tionof DefiniteIntegral 199
I t /ffi\ I
*+
the areas of the trapezoids will therefore be greater than th e area of the
region that represents the integral.
f (x)
Trap ez oid
Figure 5-5g
Problem Set 5-5
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
QI. Differentiate: y = xsinx
Q2. Integrate: f sec 2 x dx
Q3. Differentiate: f(x) = tanx
Q4. Integrate: f x3 dx
QS. Differentiate : z = cos 7x
Q6. Integrate: f sin u du
Ql. Find lim x_5 (x 2 - 2x - 15)/ (x - S).
QB. Sketch a function graph with a cusp at the point (4, 7).
Q9. Write the converse of this statement: If a = 2 and b = 3, then a + b = S.
Q10. The statement in Problem Q9 is true. Is the converse true?
For Problems 1-6, calculate approximately the given definite integral by using a Riemann
sum with n increments . Pick the sample points at the midpoints of the subintervals.
1. r
1 x 2 dx n = 6 2. J6x dx
2
3
n=8
3. r
- 1
3x dx n = 8 4. J2zxdx
- 1
n=6
s. f sinxdx, n= S 6.
1
J0 cosxdx, n= S
For Problems 7 and 8, calculat e the Riemann sums U4 , 1 4 , and M 4 , and the sum found by
applying the trapezoidal rule, T4 . Show that M4 and T4 are between the upper and lower
sums.
7.
1.2
1 tanxdx
0.4
8. r (10/x ) dx
200 Chapter
5: DefiniteandIndefinite
Integrals
9. Sample Point Problem: Figure 5-5h shows the graph of the h(x)
sinusoid
h(x) = 3 + 2 sinx.
a. At what values of x should the sample points be taken to get
an upper sum for the integral of h on (0, 6) with n = 6?
b. Where should the sample points be taken to get a lower X
Riemann sum for the integral7 1 2 3 4 5 6
c. Calculate these upper and lower sums.
Figure 5-5h
10. Program for Riemann Sums Problem: Write a program to compute Riemann sums for a
given function . If you write the program on your grapher, you can store the integrand
as y 1 , just as you did for the trapezoidal rule program. You should be able to input a
and b, the lower and upper bounds of integration; n, the number of increments to use;
and p, the percentage of the way through each subinterval to take the sample point.
For instance , for a midpoint sum, p would equal 50. Test the program by using it for
Problem 7 of this problem set. If your program gives you the correct answers for U4 ,
L4, and M4, you may assume that it is working properly .
11. Limit of Riemann Sums Problem: In Problem 1 of this problem set, you evaluated
r x 2 dx
by using midpoint sums with n = 6. In this problem you will explore what happens to
approximate values of this integral as n gets larger.
a. Use your programs to show that L 10 = 18. 795, U10 = 23.295, M1o = 20.9775, and
T 10 = 21.045 .
b. Calculate L 100and L500 . What limit does L,, seem to be approaching as n increases?
c. Calculate U100 and U500 . Does U11 seem to be approaching the same limit as L,,?
What words describe a function f(x) on the interval (1, 4] if L,, and U,, have the
same limit as n approaches infinity (and thus t:,.x approaches zero)?
d. See if you can figure out why the trapezoidal sums are always slightly greater than
your conjectured value for the exact integral and why the midpoint sums are always
less than your conjectured value.
12. Exact Integral of the Square Function by Brute Force Project: In this problem you will
find, exactly, the integral J] x 2 dx by actually calculating the limit of the upper sums.
a. Find, approximately, the value of the integral by calculating the upper and lower
Riemann sums with 100 increments, U1ooand L 100. Make a conjecture about the
exact value.
b. If you partition (0, 3] into n subintervals, then each one will be t:,.x = 3 / n units wide.
The partition points will be at x equals
0, 1 , 2 , 3 , .. . , n -
Which of these partition points would you pick to find the upper sum?
Section
5-5: Riemann
SumsandtheDefinitionof DefiniteIntegral 20 1
~a b
c. They -values will be the values of f(x) at these samp le points, f<x)
namely, the squares of these numbers . For instance, at the 10
end of the fifth increment, x = 5(3/n) and f(x) = [5(3 /n)]2
(Figure 5-5i). Write a formula for Un, an upper sum with n
increments.
5 Region
d. The formula you wrote in 12c can be rearranged by factoring
out common factors, leaving only the squares of the counting
y .....
numbers inside parentheses. From algebra, this sum is
X
1 2 + 22 + 32 + + n 2 = (n + 1)(2n + 1).
t 3
Use this information to write a closed formula (no ellipsis: Samp le polnt, x = 5(3/ n)
... ) for Un. Confirm that this formula gives the right answer Figure 5-5i
for U100 -
e. Use the formula to predict U1000 . Does Un seem to be approaching the limit you
conjectured in 12a?
f. Find algebraically the limit of Un from 12e as n approaches infinity. This limit is the
definite integral of x 2 from x = 0 to x = 3 and equals the exact area and integral!
si
13. Exact Integra l of the Cube Function Project: Find the exact value of x 3 dx . Use as a
guide the technique you used in Problem 12 of this problem set. Recall that the sum
of the cubes of the counting numbers is given by
n
L k 3 = 13 + 23 + 33 + 4 3 + ..
k= l
. + n 3 = [ i (n + 1) r.
?
5-6 The Mean Value Theorem and Rolle's Theorem
Suppose that you go 500 ft in 10
sec as you slow down on a freeway
exit. Your average velocity for the
10-sec time interval is 50 ft/sec .
It seems reasonable to conclude
that sometime in that interval,
your instantaneous velocity was
also equal to 50 ft/ sec (Figure
5-6a). In this section you will learn
the mean value theorem, which
states conditions under which this Figure 5-6a
conclusion is true. You will also
learn Rolle's theorem and use it as a lemma to prove the mean value theorem . In
Sections 5-7 and 5-8, you will see how the mean value theorem leads to an algebraic
method for finding exact definite integrals .
20 2 Chap
ter 5: Definite andIndefinite Integrals
OBJECTIVE Learn the mean value theorem and Rolle's theorem and learn how to find the point in an
interval at which the instantaneous rate of change of the function equals the average rate
of change.
Let f(x) be your displacement as you exit the freeway. As shown in Figure 5-6b, your
average velocity from time x = a to x = b is displacem ent divided by time.
. f(b ) - f(a )
Average velonty = b
- a
The average velocity is the slope of the secant
line connect ing two points on the graph. As you f(x)
Slope=
can see from the figure, there is a time x = c average
between the points a and b at which the tangent
line parallels the secant line . At this point the
f(a) ---
' ''
,
''
I velocity
instantaneous velocity, f'(c), equals the average '" '
' "
velocity . ''
'
The mean value theorem gives two sufficient
conditions for there to be an instantaneous rate f(b)
of change equal to the average rate from x = a X
and x = b. First, the function is differentiabl e a C b
for all values of x between a and b. Second, th e
function is continuous at x = a and x = b, even Figure 5-6b
if it is not differentiable at these points. The
function in Figure 5-6b meets these two conditions. Note that f is not differentiable
at x = a because the tangent line would be vertical at that point.
Property: TheMeanValueTheorem
If 1. f is differentiable for all values of x in the open interval (a, b), and
2. f is continuous at x = a and at x = b,
then there is at least one number x = c in (a, b) such that
f '( c) = f(b~ =:(a).
The two conditions in the "if" part are the hypotheses of the mean value theorem.
The "then" part is the conclusion. The left-hand and center graphs of Figure 5-6c
show why the conclusion might not be tru e if the hypotheses are not met by a
function. For the function on the left, there is a point between a and b at which the
function is not differentiable. You cannot draw a uniqu e tangent line at the cusp,
and ther e is no other place at which th e tan gent line parallels the secant. In the
center graph, the function is differentiabl e for all values of x between a and b, but
the function is not continuous at x = a. At no place is th ere a tangent line parallel
to the secant. The third function is continuous but not diff erentiable at x = a. There
is a tangent line parallel to the secant.
Section
5-6:TheMeanValueTheorem
andRolle's
Theorem 203
f(x) f(x)
f(x) No uniqu e tangent
, at the cusp No tangent
par aUel to seca nt Cusp is okay
ar H n~~?jn ~-~
,,."' : / I
/ '
B
A ;
X X X
a C b a b a C b
Not differentiable Not continuous Not differentiable at one
at an x between a and b at one of the endpoints endpoint,but continuous there
Figure 5-6c
Roi/e'sTheorem
The proof of the mean value theorem uses as a lemma another theorem from
f(x) the analysis of real numbers, Rolle's theorem. Named after seventeenth-century
f'( c)=O French mathematician Michel Rolle (pronounced "roll"), the theorem states that
between two consecutive zeros of a function, there is a place where the derivative
is zero (Figure 5-6d). Sufficient conditions for th.is conclusion to be true are the
same as the hypotheses of the mean value theor em, with the additional hypothesis
f(a) = f(b) = 0.
X
C
Consecutive ze ros of f (x)
Property: Roi/e's Theorem
Rolle's theorem If 1. f is differentiable for all values of x in the open interval (a, b), and
2. f is continuous at x = a and at x = b, and
Figure 5-6d
3. f(a) = f(b) = 0,
then there is at least one number x = c in (a, b ) such that f' (c) = 0.
Geometrical
Proofof Roi/e'sTheorem
Suppose that f(x) is positive somewhere between x = a and x = b. As Figure 5-6e
shows, f(x) must reach a maximum for some value of x (x = c) between a and b.
The fact that there is such a number relies on the completeness axiom (Section 2-6)
and on the fact that f is continuous (because it is differentiable) . If x is less than
c, then t.x is negative. If x is greater than c, then t.x is positive. Because f(c) is
the maximum value of f(x), t.y will always be negative (or possibly zero). Thus the
difference quotient,
t.y f(x) - f(c)
t.x X - C
is positive (or zero) when x is less than c and negative (or ze ro) when x is greater
than c. The limit of this differ enc e quotient is, of course, f'(c). Thus
0 s f '(c) s 0.
By the squeeze theorem, f'(c) = 0.
20 4 Chapter
5: DefiniteandIndefinite Integra
ls
f(x) Maximum
t!.x is nega tive. \ t!.x is positive .
t!.y is negative. i t!.y is negative .
f(c)
C b
Figure 5-6e
A similar case can be made if f(x) is negative for some x between a and b, or if f(x)
is always zero. In conclu sion, if the three hypotheses are true, then there is always
a number x = c between a and b such that f '(c) = 0, Q.E.D.
Algebraic
Proofof theMeanValueTheorem
The mean value theorem can be proved using Rolle's theorem as a lemma. The
left-hand graph in Figure 5-6f shows function f, the secant line through points A
and B on the graph off, and the tangent line you hope to prove is parallel to the
secant line .
y Tang ent y y Tangent
f(b) f(b) f (b)
f
((a) f(a) ((a)
g(x) ' '
:: ,~-
'
I
1-,,', ::
/
,
'
' h
'
I
' I h(C) ' '
X 1/ ' l X
X
, ''
a b a X b a C b',
I
Secant line through A and B Define functions g and h c is at maximum of h(x) .
Figure 5-6f
Let g be the linear function whose graph is the secant line through points A and B.
Let h be the function defined by
h(x) = f(x) - g(x).
As shown in the center graph of Figure 5-6f, the values of h(x) are the vertical
distances between the f graph and the g graph . Because the f and g graphs coincide
at both x = a and x = b, it follows that h(a) = h(b) = 0. The graph of his shown in
the right-hand graph of Figure 5-6f.
Section5-6: TheMeanValueTheorem
andRolle
's Theorem 205
Function h is continuo u s and differentiable at the same places as function f. This
is true because g is a linear functien, which is continuous and differentiable every-
where, and because a difference such as h (x) between continuous, differentiable
functions is also continuous and differ entiabl e.
Therefore h satisfies the three hypotheses of Rolle's theorem. Function h is
differentiable on (a , b), continuous at x = a and x = b, and has h(a) = h(b) = 0.
By the conclusion of Rolle 's theorem, th ere is a number x = c in (a, b) such that
h' (c) = 0. All that remains to be done is the algebra.
h '( c) =0
However, h '( x) = f'(x) - g ' (x), which imp lies that h '( c) = f'(c) - g'(c) = 0.
Therefore f'(c) = g '(c).
However, g is a linear function. Thus g '( x) is everywhere equal to the slope
of the graph. This slope is [f(b) - f (a )] / (b - a). Hence
f'(c) = f(bi =:(a)' Q.E.D.
Be careful! Do not read more into the mean value theorem and Rolle's theorem than
they say . The hypotheses of these theor ems are sufficient conditions to imply the
conclusions. They are not necessary conditions. Figure 5-6g shows two functions
that satisfy the conclusions of the mean value theorem even though the hypotheses
are not true.
f(x)
,
,,
X
-1 8
2
g(x) = {x. -Ix.I, if x tc 0
f(x) = 2x 213 - x.2/ 16 on [-1, 8] l, ifx=O
Not differentiable at x = 0 ot continuous at x = 0
Figure 5-6g
Example 1 Given f(x) = x.113, plot the graph . Explain why f satisfies the hypotheses of the mean
value theorem on [O, 8]. Find a value of x = c in the op en interval (0, 8) at which the
conclusion of the theorem is true, and show on your graph that the tangent really
is parallel to the secant .
Solutions Figure 5-6h suggests that f is differentiable everywhere except, perhaps, at
x = 0. Because
f'(x) = x-213,
f'(O) would be 0- 213 = 1/0 213 = 1/0, which is infinite. But f is differentiable on
the open interval (0, 8). Function f is continuous at x = 0 and x = 8 because the
. 206 Chapter
5: Definite and Indefinite Integrals
limits of f(x ) as x - 0 and as x - 8 are O and 2, th e values of f(O) and f(8),
respectively.
Thus f satisfi es the hypoth eses of th e m ean value th eorem on [O,8). The slope
X of the secant line (Figure 5-6h) is
8 3 113- 0 1/3 1
m sec = 8- 0 4.
Figure 5-6h Setting f'(c) = 1/4 gives
l..c 213 = .!. ~ c 213 = i4
or
3 4
3 12
.. c = = 1.5396 . . .
Only the positive value of c is in (0, 8). Thus the answer is c = 1.5396 . ...
X
8 To plot th e tangent line, find the particular equation of the line with slope 1/ 4
X= l.5396 y=l.1547 through the point (1.5396 .. . , f (l.5396 ... ) ).
Figure 5-6i f(l.5 396 . . .) = (1. 5396 ... ) 113 = 1.1547 . . .
.. y - 1.1547 . .. = 0.25(x - 1.5396 ... ) ~ y = 0.2 5x + 0.7698 ...
Enter this equation as y 2 into your grapher . Figure 5-6i shows that the line reall y
is tangent to th e graph and that it is also parallel to the secant line.
Example 2 Given f (x) = x sinx, find the first interval of nonnegative x-values on which the hy-
potheses of Rolle's th eorem ar e tru e. Then find the point x = c in the corresponding
open interval at which th e conclusion of Rolle's theorem is true. Illustrat e with a
graph.
Solutions As Figure 5-6j shows, f (x) = 0 at x = 0 and at x = TT because sinx is zero at
those points . To establish differentiabilit y,
f'(x) = sinx + XCOS X,
which exists for all x. Thus f is diff erentiable on (0, TT). The function is
continuous at O and TT because it is differentiable there . So the hypotheses are
met for th e interval [O, TT].
f'(c) = 0 = sine + ccosc = 0
.' . C = 2.02 875 Use your grap her's solve feature.
Figure 5-6j
Note that f'(O) is also zero , but c * 0 because O is not in the open int erval
(0, TT).
Sec
tion5-6:TheMeanValueTheorem
andRolle's Theo
rem 20 7
Problem Set 5-6
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minut es.
QI. Integrate: f (x + 2) dx
Q2. Integrate: f 1Ot dt
Q3. Integrat e: f csc 2 x dx
Q4. Differentiate: g (x ) = csc x
Q5. Differentiate: p (x ) = sin 5 x
Q6. Sketch a graph that shows f12 x 2 dx.
Ql. Sketch a graph with a removable discontinuity at the point (3, 5).
QB. Sketch a graph of a function that is continuous at the point (2, 1) but not differentiable at
that point.
Q9. Find limx-o 1/x.
Q10. If f(x) = 2x + 6, th en r- (x ) =
1 _7_
1. State the mean value theorem. If you have to refer to th e definition in this book to
find out what it says, th en practic e stating it until you can do so vvithout looking .
2. State Rolle's th eorem . If you have to refer to its definition in this book to find out
what it says, th en practice stating it until you can do so without looking.
For Problems 3-6, plot the graph. Find a point x = c in the given interval at which the
mean value theorem conclusion is true. Plot the secant line and the tangent line, showing
that the y are reall y parallel. Sketch the resulting graphs.
3. g (x ) = ~. [l, 4 ] 4. f(x ) = x 4 , [-1, 2]
5. c(x) = 2 + cosx, [ 0, f] 6. h (x ) = 5 - ft , [1, 9]
For Probl ems 7-10, plot th e graph. Find an interval on which the hypotheses of Rolle's
theorem ar e met. Then find a point x = c in that interval at which the conclusion of Rolle's
theorem is true. Plot a hori zontal line through the point (c, f ( c) ) and show that the line
really is tangent to the graph. Sketch the result.
7. f(x) = xcosx (use [o.f ]) 8. f(x) = x2 sinx
9. f (x ) = (6x - x 2 ) 112 10. f (x ) = x 4/3 - 4x 1/3
11. Compound Int erest Problem: Suppos e you invest $ 1000 in a retirement account. The
account pa ys int erest continuously at a rate that mak es the annual perc entage rate
(APR)equal 9%. Thus the number of dollars, d(t), in your account at time t years is
given by
d(t) = 1000 (1.09 1 ).
a. When you retire 50 years from now, how much money will be in the account?
Surprising?!
b. At what average rate does your money incr ease?
208 Chapter
5: DefiniteandIndefiniteIntegrals
c. Differ entiate numericall y to calculate the instantaneous rate at which your money
is increasing now, at t = 0, and when you retire, at t = 50. Is the average of these
two rates equal to th e average rat e you found in llb ?
d. Solve numeri cally to find th e tim e at which th e inst antan eous rat e equal s the
average rate . Is this tim e half way between now and the time you retir e?
12. Baseball Line Drive Prob lem (Second Inning ): In Probl em 38 of
Probl em Set 4-6, the displac ement, d(t), of a baseball from hom e
plate was given to be d(t)
d(t) =
60 .5 (
0
5
- t), t
.5 + t
if :$ o.5
150
1
150 ( 2 - +), if t ~ 0. 5
where tis the number of seconds since the ball was pitch ed (Figure
50
5-6k). Discuss th e way in which the mean valu e theorem appli es to
fun ction d . For instance , do the hypotheses apply on the int erval
[O, l]? [O, 2]? [0.5, 2]? Is th e conclusion tru e anywh ere in (0, 1)?
0.5 1 1.5 2 2.5
(0, 2)? (0.5, 2)? How do your answers illustrate the fact that the
hypoth eses of th e m ean value theor em are sufficient conditions, Figure 5-6k
not nec essary ones?
13. Sketch a graph that clear ly shows you und ersta nd both th e hypotheses and the
conclusion of Rolle's th eore m.
For Problems 14- 16, sketch a gra ph that shows why the conclusion of Rolle's th eorem
might not be true if f m eets all the hypoth eses of the theorem on the int erva l [a, b]
except the following .
14. The function f is discontinuou s at x = b.
15. The function f is continuous, but not diff erentia bl e, at x = din (a, b ).
16. The value of f (a ) is not equal to zero.
17. Sketch a graph that shows that a function may satisfy th e conclusion of Rolle's
theorem even though the function do es not me et all of th e hypoth eses.
18. What was Rolle's first name ? When did he live? In what countr y?
For Probl ems 19-28, plot (if neces sary) and sketch the graph. State which of the hypotheses
of Rolle's theorem are not m et on the given interval. Then state whet her or not the
conclusion of Rolle's theor em is tru e on the corresp ondin g open interval.
19. f (x) = x 2 - 4x on [O, 1) 20. f (x) = x 2 - 6x + 5 on [1, 2]
21. f (x ) = x 2 - 4x on [O, 2] 22. f (x) =x 2
-6x + 5 on[l,4)
23. f(x ) = x 2 - 4x on [O, 3] 24. f (x) = Ix - 21- 1 on [l, 3]
1
25. f(x ) = - on [O, 5] 26. f(x) = x - [x] on [1, 2]
X
( [x] is the greates t int eger less than
or equal to x.)
x3 - 6x 2 + llx - 6
27. f(x ) = 1 - (x - 3)213on [2, 4] 28. f(x) = on [l, 3)
x-2
Section5-6: TheMeanValueTheorem
andRolle
's Theo
rem 209
. x3 - 7x 2 + l 3x - 6
29. Given g(x) = x _ , explain why the hypotheses of the mean value
2
theorem are not met on any interval containing x = 2. Is the conclusion true if the
theorem is applied on [l, 3]? On [1, 5]? Justify your answers . A graph may help.
30. Given h(x) = x 213, explain why the hypotheses of the mean value theorem are met on
[O, 8] but are not met on [ -1, 8]. Is the conclusion of the mean value theorem true for
any x = c in (- 1, 8)? Justify your answer. A graph may help.
31. Suppose that f(x) = Ix - 31+ 2x.
a. Use the definition of absolute value to write two equations for f(x), one that works
if x :=:-:3 and one that works if x < 3.
b. Sketch the graph of f.
c. Is f continuous at x = 3? Justify your answer.
d. Is f differentiable at x = 3? Justify your answer.
e. Which hypothesis of the mean value theorem is not met on the interval [l, 6]?
f. Show that there is no point x = c in (1, 6) at which the conclusion of the mean value
theorem is true.
ft
g. Is f integrable on [O, 5]? If not, why not ? If so, evaluat e f (x ) dx geom etrically.
32. New Jersey Turnpike Problem: When you enter the New Jersey Turnpike you receive a
card that indicates your entrance point and the time at which you entered . When you
exit, therefore, it can be determin ed how far you went and how long it took, and thus
what your average speed was .
a. Let f(t) be the number of miles you traveled int hours . What assumptions must
you make about f so that it satisfies the hypotheses of the mean value theorem on
the interval from t = a, when you entered the turnpike, to t = b, when you exited?
b . Suppose that your average speed is 60 mi/ hr. lf f meets the hypotheses of the mean
value theorem, prove that your speed was exactly 60 mi/ hr at some time between
t = a and t = b.
33. Roi/e's Theorem Proof Jllustrated by Graph and by Table: The proof of Rolle's theorem
shows that at a high point, f(c), for the open interval (a, b), th e difference quotient
f (x ) - f (c )
x- c
is always positive (or zero) when x < c and always negative when x > c. In this problem
you will show graphically and numerically that this fact is tru e for a fairly complicated
function.
a. Figure 5-61 shows the graph of
f(x) = 25 - (x - 5)2 + 4 cos (2rr(x - 5)) .
Plot the graph as y 1 . Use a friendly window that includes
X = 5, with 6.X = 0.1.
b. Explain why the high point of the graph is at x = 5. What does
f(5) equal? Figure 5-61
c. Find f'(x) . What does f '(5) equal?
2 10 Chapter
5: DefiniteandIndefin
ite Integrals
d. Let Y2 be the difference quotient
. Yi - f(5)
Y2= x - 5.
Plot on your grapher the graph of y 2 . Sketch the result.
e. Make a table of values of the difference quotient for each 0.5 units of x from x = 2
to X = 8.
f. Read the proof of Rolle's theorem, which appears in this section. Explain how the
work you've done in this problem relates to this proof. Tell which hypothesis of
Rolle's theorem has not been mentioned so far in this problem. Is this hypothesis
true for function f? Can the conclusion of Rolle's theorem be true for a function if
the hypotheses aren't? Explain.
34. Mean Value Theorem Proof fllustrated by Graph and by Table: In the proof of the mean
value theorem, a linear function, g, and a difference function, h, were created . Rolle's
theorem was then applied to function h. In this problem you will derive equations for
g(x) and h(x), and you'll illustrate the proof by graph and by table of values.
a. Figure 5-6m shows the graph of
f(x) = 1 + X + COS TTX f(x )
and a chord drawn between the endpoints of the graph for the
interval [2, 4.5]. Plot the graph as y 1 Use a friendly window with 4
D..xno larger than 0.05.
b. Figure out an equation for the linear function g(x) and plot it 2 4.5
as Y2-
c. Use a time-efficient way to plot the function h(x) mentioned in Figure 5-6m
the proof of the mean value theorem.
d. Find a point x = c in (2, 4.5) at which the conclusion of the mean value theorem is
true .
e. Make a table of ten values of h(x) for values of x around c, spaced 0.05 units apart.
Show that h(c) is an upper (or lower) bound for the values of h(x) in the table.
f. Read the proof of the mean value theorem, which appears in this section. Show
how functions g and h in this problem are used with Rolle's theorem to prove the
mean value theorem .
g. Show that function f meets the hypotheses of the mean value theorem on [2, 4.5].
35. Corollary of the Mean Value Theorem: A corollary of a theorem is another theorem
that follows easily from the first one. Explain why the corollary of the mean value
theorem shown in the box below is true.
Property: Corollaryof the MeanValueTheorem
If f is differentiable on the closed interva l [a, b],
then there is at least one number x = c in (a, b) such that
f'(c) = f(b~ =:(a).
Section
5-6: TheMeanValueTheorem
andRolle'sTheorem 211
36. Converse of a Theorem: It is easy to show that if two differentiable
functions differ by a constant, then their derivatives are equal for
all values of x in the domain . For instance, if
f(x) = sinx and g(x ) = 2 + sinx,
then f'(x) = g'(x) for all x (Figure 5-6n). The converse of a theorem
is the statement that comes from interchanging the hypothesis and
the conclusion, as shown in the box below . figure 5-6n
Property: Converseof Equal-Derivative
Theorem
Theorem:
If f(x) = g(x) + D for all x in the domain, where Dis a consta n t,
then f'(x) = g'(x) for all x in the domain .
Converse
of theTheorem:
If f'(x) = g'(x) for all x in the domain,
then f(x) = g(x) + D for all x in the domain, where D is a constant.
As you realize, the converse of a theorem is not necessarily true . For instance, the
converse of the mean value theorem is false. However, the above converse is true and
can be proved by contradiction with the help of the mean value theorem .
a. Suppose that there are two values of x such that f (a) = g (a ) + D 1 and f(b) =
g(b) + D 2 , where D 1 * D 2 . Let h(x) = f(x) - g(x). Explain why the mean value
theorem applies to hon the interval [a, b].
b. The mean value theorem lets you conclude that there is a number x = c in (a, b )
such that
h' (c) = h(b~- h(a).
-a
Show that h'(c) also equals (D 2 - D 1 )/(b - a).
c. Show that h' (x) = 0 for all x in (a, b), and thus h'(c) = 0. Use the result to show that
D 1 and D 2 are equal, which proves the converse by contradicting the assumption
made in 36a .
3 7. Antiderivative of Zero : Prove as a corollary of the property in Problem 36 that if
f'(x ) = 0 for all values of x, then f(x) is a constant function .
38. Let f(x) = (cosx + sinx) 2 . Let g(x) = sin2x . On the same screen, plot graphs off and
g. Sketch the result. Make a table of values of the two functions for convenient values
of x, say 0, 1, 2, . . .. What seems to be true about values of f(x) and g(x) at the same
values of x? Prove algebraically that f'(x) = g '( x) for all values of x.
39. Maximum and Minimum Values of Continuous Functions: When you study the analysis
of real numbers you will show that if a function f is continuous on a closed interval
[a, b], then f(x) has a maximum value and a minimum value at values of x in
that interval. Explain why a function that meets the hypotheses of Rolle's theorem
automatically meets these requirements .
40. Intermediate Value Theorem versus Mean Value Theorem : The words intermediate and
mean both connote the concept of betweenness. Both the intermediate value theorem
and the mean value theorem assert the existence of a number x = c that is between
two numbers a and b . However, the hypotheses and conclusions of the theorems are
212 Chapter
5: DefiniteandIndefinite
Integra
ls
quite different . Write a paragraph or two describing how the two theorems differ and
how they are alike . Graphs will help.
41. Journa l Problem: Update your journal with what you've learned since the last entry .
Include su ch things as those listed here .
The one most important thing you have learned since your last journal entry
The fac t that you can now fill in one more square in your chart of calculus concepts
The two new theorems you have just learned and how th ey are related to each other
The difference between the mean value theorem and the intermediate value theorem
The differe nce between definite integral and indefinite integral
The different kinds of Riemann sums
Any techniq u e or idea about which you plan to ask at th e next class period
5-7 Some Very Special Riemann Sums
In Sections 5-5 and 5-6, you learned that a definite integral is the limit of Riemann
sums and that there is a mean value theorem, which equates the slope of a tangent
line to the slope of a secant line . In this section you will explore Riemann sums in
which the sample points are picked very cleverly based on the mean value theorem.
The conclus ions you reach will pave the way to evaluating exact definite integrals,
without using Riemann sums or the trapezoidal rule .
OBJECTIVE Calculate Riemann sums for given sets of sample points and reach some conclusions
about how the sample points were chosen.
Problem Set 5-7
Note: Your instructor might have you work these problems in groups so that you can
con sul t your classmates about what you are discovering.
For Probl ems 1-7, let f(x) = x 1f2 (Figure 5-7a).
1. Find an estimate for the integral f(x)
2
I = J:1
xl
12
dx
by us in g the trapezoida l rule with n = 3 subintervals . Write
down all th e decima l places your calculator will give you. Does
this value overest imate or underestimate the actual integral ?
Explain.
X
2. Fin d a mi dp oint Riemann sum for integral I in Problem 1. Use 2 3 4
n = 3 incre m ents. Show that this sum is not equal to the value
you found b y using the trapezoidal rule . Does the midpoint sum Figure 5-7a
overestimate or underestimate the actual integral? Explain.
Sec
tion 5-7: Some Very Specia
l Riemann
Sums 2 13
3. Find a Riemann sum for integral I by using th e subintervals in Problem 1, but use the
following sample points. (Let k stand for the subinterval numb er .)
k X=C
1 1.48 584256
2 2.491610 26
3 3.49402722
How does this sum compare with your answers to Problems 1 and 2?
4. Find a Riemann sum for I by using six subint ervals of equal width and the following
sampl e points.
k X = C
1 1.2458 0513
2 1.74701361
3 2.247680 40
4 2.748 10345
5 3.2-1839587
6 3.74861006
How does the int egral approximated by this Riemann sum compare with other values
in this problem set7
5. Let g(x) = (2/3)x3f2. Find the point in th e open interval (1, 1.5) at which th e conclusion
of the mean value theorem is tru e for function g. Where have you seen this number in
this problem set 7
6. How is function g related to function f 7
7. Make a conjecture about th e exact value of th e int egra l in Probl em 1.
For Problems 8-12, let f (x ) = x 3 (Figure 5-7b). ( (x)
8
8. Find an approximate value of the integral 7
6
I = Jo2
f(x) dx 5
by using a midpoint Riemann sum with n = 100 subintervals. Do 4
3
not round off. Does this sum underestimate or overestimate the
2
int egral ?
X
9. Find another approximat e value for integral I usin g the trape-
zoidal rule with n = 100 . Do not round off. Based on your answer
to Problem 8, mak e a conjecture about what integer th e exac t Figure 5-7b
value of integral I is.
10. Let g(x) = f f (x ) dx . Assuming that the constant of integration is C = 0, find g(O),
g(0.5), g (l ), g(l.5), and g(2). Use these values to find th e slopes of the secant lines
connecting (0, g(O)) to (0.5, g(0 .5)) , and so forth .
2 14 Chapter
5: DefiniteandIndefinite
Integrals
11. The mean value theorem applies to function g on each interval (0, 0.5), (0.5, 1), (1, 1.5),
and (1.5, 2). Find the values of x = c 1 , c2 , c3 , and c~ in each interval for which g ' (c)
equals the slope of the respective secant line in Problem 10. Do not round off.
12. Find the Riemann sum for the integral in Problem 8, using as sample points the values
of c you found in Problem 10. Keep all the decimal places . How does this sum compare
with the exact (integer) answer you conjectured in Problem 9?
13. Make a conjecture about how you could calculate the exact value of the integral by
using a Riemann sum with just one increment and a judiciously-chosen sample point.
Then test your conjecture to see whether or not it works .
5-8 The Fundamental Theorem of Calculus
f (x)
f
Riemann sums tend to get closer to the actual
value of the corresponding integral as the
number of intervals increases. The value of a
Riemann sum also depends on just where you
pick the sample points in each interval. For
instance, a midpoint sum should be closer to
the actual value of the integral than a sum for
X
which sample points are taken at one end of the
a b
interval.
Figure 5-8a
What if you could pick the sample points in such
a way that the Riemann sum is independent of g(x) g
the number of intervals? The Riemann sum
Mean valu e theorem
would be the same no matter how many or conclusion is true here .
how few intervals you used . In Section 5-7, you
discovered a way to do this. In this section you
will use what you discovered to put together
the fundamental theorem of calculus . This
theorem lets you evaluate definite integrals
exactly, by algebra, using indefinite integrals.
X
Figure 5-8a shows a function f for which it is b = xn
desired to find the definite integral f(x)
r f(x) dx .
Use mean value theor em
c-value s for sampl e points.
f
The top graph in Figure 5-8b is function g, an
indefinite integral off. That is, g(x) = Jf(x ) dx.
By the definition of indefinite integral, g '( x ) =
f(x). Because g is differentiable, the mean
value theorem applies to it on [a, b] or on any X
subinterval of [a, b].
Figure 5-8b
Section5-8:TheFundamental
Theorem
ofCalculus 215
Divide the interval [a, b] into n subintervals of equal width 6.x . Let C t , c2 , c3 , ... , Cn
be the points in the first, second, third, .. . , nth subinterval at which the conclusion
of the mean value theorem is true for function g on that subinterval. Thus
'( ) g(xi) - g(a)
g Ci = 6.x
'( ) g(x z) - g(xi)
g Cz = 6.x
'( ) g(x 3) - g(x z)
g C3 = t:,.x
g '( Cn ) = -----
g(b) - g(Xn - 1)
- .
6.x
Now, use the points c1, c2, c3, ... , Cn as sample points for a Riemann sum of the
original definite integral, as shown in the bottom graph of Figure 5-8b. That is,
Rn= f(c1) 6.x + f(c z) 6.x + f(c 3) 6.x + + f(cn)6.x.
However, f(x) = g'(x) for any value of x. Therefore,
Rn = g ' (c 1 ) 6.x + g'(c z) 6.x + g '( c3) 6.x + + g '( cn) 6.x.
Replacing the g ' (c 1) with g (xi) - g (a ) , and so forth, canc eling th e 6.x's, and
6.x
arranging in column form gives
R,, = g(x 1) - g(a)
+ g(x z) - g(x1)
+ g(x 3) - g(x z)
+ g(b) - g(Xn - 1)
Rn= g(b) - g(a) .
All the middle terms "telescope," leaving only - g (a ) from the first term and g(b)
from the last . The result is independent of the number of increments. The quantity
g(b) - g(a) is, in fact, the exact value of the definite integral. This is what the
fundamental theorem of calculus says. The definite integral can be calculated by
evaluating the antiderivative at the upper limit of integration and then subtracting
from it the value of the antiderivative at the lower limit of integration . The formal
statement of this theorem and its proof follow.
Property: TheFundamental
Theoremof Calculus
If f is an integrab le function, and if g(x) = Jf(x) dx,
then J% f(x) dx = g(b) - g(a).
Proof:
Partition the interval [a, b] into n subintervals of equal width 6.x. Let Ln and
U,, be lower and upper sums for I:
f (x ) dx. Let Rn be the Riemann sum equal to
21 6 Chapter
5: Definite
andIndefinite
Integrals
!MA&
g(b) - g(a),as derived above. Because any Riemann sum is between the upper
and lower sum, you can write
1 11 :s; R 11 :s; U11
By the definition of integrability, the limits of Ln and Un as 6X approaches zero
are equal to each other . By the squeeze theorem, therefore, the limit of Rn as
6X approaches zero is also equal to this common limit. The common limit is
defined to be the definite integral. Therefore
b f(x)
J, dx = lim Rn= lim [g(b) - g(a)] = g(b) - g(a), Q.E.D.
a 4x-O 4x-O
The last equality is true because the quantity g(b) - g(a) is a constant with
respect to 6x.
OBJECTIVE Learn what the fundam ent al theorem of calculus says. Show that it produces reasonabl e
answers by comparing them with approximations obtained by using th e familiar techniques
of Riemann summing and the trap ez oidal rule.
You-perhaps with your study group members - will accomplish this objective by
working the prob lems in the following problem set. In Problem 10, you will record
in your journal what you have learned . In Section 5-9, you will see examp les of how
to use the fundamental theorem, and you'll practice evaluating definite integrals.
Problem Set 5-8
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes .
QI. r(x) = f m(x) dx if and only if - ?-
Q2. Write the definition of derivative.
Q3. How fast is f(x) = x2 changing when x = 3?
Q4. Find dy : y = sec x.
QS. Find y' : y = (x 2 + 3)4.
Q6. Find dz /du: z = sin u.
Ql. Find f'(x) : f (x) = 73 .
QB. Solve for the minimum positive value of x: cos x = 0.5.
Q9. cos2 x + sin 2 x = -?-
QIO. Sketch a graph that shows the conclusion of the mean value theorem.
1. For the integral I = J; 1ox- 15 dx, do the following :
a. Find the exact value of I by using the fundamental theorem. What happens to "+C"
from the indefinite integral?
b. Sketch a graph that shows an upper sum with n = 5 increments for this integral.
Section 5-8: TheFundamenta
l Theorem
ofCalculus 21 7
c. Find U5 , L5 , and th e averag e of these two sums. Does the average overestimate or
underestimate th e integral ? Explain.
d. Find midpoint Riemann sums M 10 , M 100 , and Mwoo for I. Do the Riemann sums
seem to be convergin g to th e exact value of /7
5
2. Let I= f~ sinx dx . Find th e exact value of I by using th e fundamental theorem .
Show that midpoint Riemann sums approach this value as the number of increm ents
approaches infinity . Sketch an appropriate graph to show why midpoint Riemann sums
overestimate th e value of the definit e integral.
3. State the fundamental th eorem of calculus .
4. Prove that if f is an integrabl e function, then for any partition of the inter val [a, b ]
I:
into n subintervals of equal vvidth th ere is a Riemann sum for f(x) dx whose value
is independent of th e numb er n.
5. Prove the fundamental theorem of calculus by usin g the result of Probl em 4 as a
lemma .
6. You have proved that it is possible to pick a Riemann sum whose value is ind ependent
of the number of increments. Suppose that the int erval [a, b] is taken as a whole .
That is, suppose th ere is just one "subinterval." Where should th e sampl e point for
this interva l be picked so that th e corresponding Riemann "sum" is exactly equal to
the definite integral from a to b? How does your an swer relate to the fundam ental
theorem ?
7. Fr eeway Exit Prob lem: In the design of free way exit ramps it is important to allow
enough room for car s to slow down before th ey ent er th e frontage road . Suppos e that
the velocity is given by v (t) = 100 - 20(t + 1) 112, wher e v(t ) is in feet per second and
t is the number of seconds since you start ed slowing down. Write a definite inte gral
equal to the number of feet the car goes from t = 0 to t = 8. Evaluate the int egral
exactly by using the fund am ental th eorem.
h (x)
8. The Fundamenta l Theor em A noth er Way: Let h be th e squar e
root function h (x) = x 112. Let P be th e region und er th e graph of
h from x = 4 to x = 9 (Figur e 5-8c). 3
a. Evaluate the Riemann sum R 10 for P, pickin g sample points 2
at the midpoints of th e subint ervals. Don't round th e answ er .
b. Let u be a valu e of x in th e interval [4, 9]. Let A( u) be the X
ar ea of the portion of th e region from x = 4 to x = u (Figure 4 9
5-8d). Let t.u be a small change in u. The area of the strip
Figure 5-Bc
from x = u to x = u + t.u equals A(u + t.u ) - A( u) . Explain
why this area is between h (u ) t.u and h (u + t.u )t. u. Writ e h (x)
the result as a thr ee-memb er inequali ty. Strip
c. Use the inequalit y you found in 8b to prov e that dA /du = 3
h (u ). This equation is called a differential equation.
2
d. Multiply both sid es of th e differential equ ation given in 8c by
du . Then take th e ind efinite int egral of both sides . Find the
X
constant of integration by obser ving that A(4) must equal
4 u u +t. u
zero .
Figure 5-Sd
218 Chapter
5: Definiteand Indefin
ite Integrals
e. Find the area of region P by evaluating A(9) . Explain why your answer to 8a is
consistent with your answer to this problem.
9. Riemann Sum Sketching Problem: Sketch appropriate rectangles to show the following.
a. An upper sum with four increments, where the sample points are taken at the left
endpoint of each subinterval
b. A lower sum with three increments, where the sample points are taken at the left
endpoint of each subinterval
c. A Riemann sum with five increments, where the sample points are taken at the
midpoint of each subinterval
d. A Riemann sum with four increments, where no sample point is taken at the middle
or at either end of a subinterval
e. A Riemann sum with three increments, where two different sample points are at
the same x-value
f. A subinterval for an upper sum in which the sample point must be somewhere
between the two endpoints
10. Journal Problem: Update your journal with what you've learned as a result of doing this
problem set. Include such things as those listed here.
A statement of the fundamental theorem of calculus
How sample points can be chosen so that a Riemann sum is independent of the
number of increments, and how this fact leads to the fundamental theorem
Evidence to show that Riemann sums really do get close to the value of a definite
integral found by the fundamental theorem and as n approaches infinity
What you now better understand about the meaning of Riemann sum
Anything about the fundamental theorem that you're still unclear about
5-9 Definite Integral Properties and Practice
You now know precise definitions of the four concepts in calculus, and you know
what you should be able to do with each concept.
Define it. Understand it. Do it. Apply it.
Limit
Derivative
Definite integral
Indefinite integral
In this section you will work on the "Do it" box for definite integrals. Because you
will be using the fundamental theorem, you will also be working on the "Do it" box
for indefinite integrals.
Section
59:Definite
IntegralProperties
andPractice 2 19
OBJECTIVE Be able to evaluate quickly a definite integral, in an acceptable format, using the funda-
mental theorem of calculus.
To evaluate a definite int egral such as
4
Jx
I
2
dx '
you could start by writing an indefinite integral, g(x) = f x 2 dx, integrating, then
finding g(4) - g(l). Here is a compact format that is customarily used.
r 2
x dx
4
= lx
3
3 Il Pronoun ced "(l /3)x 3 evaluated from x = I to x =4."
= !3 . 43 - ! . 13
3
= 21
In the first step you find the indefinite integral . The vertical bar at the right reminds
you that the upp er and lower limits of integration are to be substituted into the
expression at its left.
As mentioned in Section 5-5, the values 1 and 4 are called limits of integration.
This terminology is unfortunat e because these values have nothing to do with
the concept of limit as you have defined it. The term lower and upper bounds of
integra tion wou ld be more suitable. However, the word limit is firmly entrenched in
mathematical literature, so you should get used to the ambiguity and interpret the
wor d in its proper context .
The following are some properties of definite integrals that are useful both for
evaluating definite integrals an d for understanding what they mean.
Integral
witha NegativeIntegrand
Suppose you must evaluate I:(x 2
- 5x + 2) dx. The result is
4
(x3 - ~ x 2 + 2x) I i
= ( 364 - 40 + 8 ) - ( 1 5
3- 2 + 2)
= -10 .5.
The answer should be surprising . How could an area be negative? Figure 5-9a reveals
the reason why. The region lies below the x-axis. The Riemann sum has the form
"'i.J(x) 6.x.
Figure 5-9a Each value of f(x) is negative, and each 6.x is positive. Thus each term in the sum is
negativ e, and as a result the integral is negative.
220 Chapter
5: DefiniteandIndefinite
Integrals
Integral
froma Higher
Number
to a LowerNumber
Suppose an integral has a lower limit of integration that is greater than the upper
limit, such as
0
f(x) =COSX r
Jrr/3
cosxdx.
Evaluating the integral gives
0
sinx Irr/3
negative.
= sinO - sin f
= 0 - J3
2
Figure 5-9b
= -0.86 602 ...
The integral is negative, yet the integrand is positive, as shown in Figure 5-9b. The
reason why is th e same as that shown in the previous property but mor e subtle. In
this case each 6.x is negativ e. Because 6.x = (b - a)/n, 6.x will be negative whenever
bis less than a. Thus each term in the sum
IJ (x ) 6.x
will be negative, making the integral itself negative. Combining the observations
above leads you to conclude that if both f(x ) and 6.x are negative, the integral is
positi ve. For instanc e,
r - x dx = -x 2
I: -
= + = 4.
Sumof Integrals
withtheSameIntegrand
Suppos e you int egrat e a function such as x 2 + 1 from x = l to x = 4, then int egrate
the same function from x = 4 to x = 5. Figure 5-9c shows the two regions whose
areas equal the two integrals. The sum of the two areas equals the area of th e region
from x = l all th e way to x = 5. This fact suggests that the two integrals should add
f (x) = x 2+ 1 up to the integral of x 2 + 1 from x = l to x = 5. This turns out to be true, as you can
see below.
I (x
1
4 2
+ l ) dx = (.!.x
3
3
.= ~
+ x) 1-1
l 3
+ 4 - .!.
3
- 1 = 24
~3 -
5
5
X l4
\ x 2 + 1) dx = ( .!.x
3
3
+ x) 1
-I
= 125
3
+5- 4 = 21 .!.
3
1 4 5
-
0
2
x 3 + x) 15l =
+ l) dx = ( .!. 125
+ 5 - .!.- 1 = 45.!. which equals 24 + 21.!.
Figure 5-9c Jl (x 3 3 3 3' 3
In general, I! f (x) dx = J; f (x ) dx + S:f (x ) dx.
Integrals
betweenSymmetric
Limits
ea f(x) dx is called an integral between symmetric limits . If f happens to be either
an odd function (such as sinx or x 5) or an even function (such as cos x or x 6 ), then
the int egral has properties that make it easier to evaluate .
Section5-9:DefiniteIntegralProperties
andPractice 221
d i\
*+
f(x) f(x)
Odd function, f(-x) = -f(x) Even function, f(-x) = f(x)
e.g ., f(x) = sin x e.g., f(x) = cosx
Figure 5-9d
The left-hand graph in Figure 5-9d shows an odd function, where f( - x) = - f(x) .
The area of the region from x = - a to x = 0 equals the area from x = 0 to x = a,
but the signs of the integrals will be opposite. Thus the integral equals zero! For
instance,
f- 2 x 3dx = .!.x412
2
4 -2
= .!.(2)4 -
4
.!.(-2)4 = 0.
4
The right-hand graph in Figure 5-9d shows an even function, where f( - x) = f (x).
The areas of the regions from x = - a to O and from x = 0 to a are again the same.
This time the integrals have the same sign . Thus you can int egrate from zero to a,
then double the result. For instance,
3
3
J-. 3 x dx = x 5 I = l:, (3) 5 l:, (-3)
4 5
-3
- = 48.6 - (- 48.6) = 97 .2, and
r3 x 4 dx = 2. lx
2 Jo 5 13
:, o = 2. 1(3)
,
5 - 5
:, 0) = 2(48 .6) = 97.2
2 . 1(
Integral
of a Sum,andIntegral
of a Constant
Timesa Function
In Section 5-4, you learned that the indefinite integral of a sum of two functions
equals the sum of the integrals and that the integral of a constant times a function is
the constant times the integral of the function. By the fundamental theorem, these
properties apply to definite integrals, too.
r (f(x) + g(x)) dx = r f(x ) dx + r g (x ) dx, and
r kf(x ) dx = k r f (x) dx .
Figures 5-9e and 5-9f show geometrically what these two properties say. In Figure
5-9e the regions representing the two integrals are shaded differently. The second
region sits down on top of the first without a change in area. Thus the integral of
the sum of the two functions is represented by the sum of the two areas. In Figure
5-9f, the region representing th e integral of f is magnified by a factor of k in the
vertical direction. The region representing the integral of kf thus has k times the
area of the region representing f.
222 Chapter
5: Definite
andIndefin
ite Integral
s
y
kf(x)
f!kf(x)dx
,, ,, f(x)
X
a a b
Figure 5-9e Figure 5-9f
UpperBounds
forIntegrals
y Suppose that the graph of one function is always below the graph of another. Figure
5-9g shows functions for which
f(x) < g(x) for all x in [a, b].
The area of the region corresponding to the integral off is smaller than that for g.
Thus
Figure 5-9g
J:f(x) dx < rg(x) dx.
The prop erties shown above are summarized in the following table.
Propertiesof DefiniteIntegrals
1. PositiveandNegative
Integrands:
The integral I: f(x) dx is
positive if f(x) is positive for all values of x in [a, b] and
negative if f (x) is negative for all values of x in [a, b],
provided a < b.
2. Reversal
of Limits
of Integration I:
: f (x) dx = - I: f(x) dx
3. Sumof Integrals I: f(x)
withtheSameIntegrand: dx = Ii f(x) dx + It f(x) dx
4. Integrals betweenSymmetric Limits: If f is an odd function, then t'a f(x) dx = 0.
If f is an even function, then t'a f(x) dx = 2 It f(x) dx.
S. Integral of a Sumandof a Constant Times a Function:
I: (f(x) + g(x)) dx = I: f (x) dx + I: g(x) dx
I: kf(x) dx = k f% f(x) dx
6. Upper
Bounds
forIntegrals:
If f(x) < g(x) for all x in [a, b], then f: f(x) dx < I: g(x) dx.
Section5-9: Definite
IntegralProperties
andPractice 223
~a b
Problem Set 5.9
DoTheseQuickly
The following problems are intend ed to refresh your skills . You should be able to do all
ten problems in less than five minutes.
Ql. Evaluate: f x 5 dx
Q2. Evaluate: f (3x + 7) 5 dx
Q3. Evaluate: f x - 4 dx
Q4. Evaluate: f sin 5 x cos x dx
as.Evaluate : f cos 5x dx
Q6. Evaluate: f (cos 2 x + sin 2 x ) dx
Ql. Evaluate: f sec2 x dx
QB.Find y ' : y = x 2 secx .
Q9. I:
f(x) dx is a(n) - ?- int egral.
Q10. f f (x) dx is a(n) - ?- int egral.
For Problems 1- 26, evaluat e the int egral exactl y by using the fundamental theorem. You
may check your answ er b y Riemann sum or by trapezoidal rule.
1. r x 2 dx 2. r x 3 dx
3. r 2 (1 + 3x) 2 dx 4. t (5x - 2) 2 dx
5. f 60x
8
1
213dx
6. r 24x 312dx
7. ts dx 5 8. rso dx
Jzo
9. r2(x 2+ 3x + 7) dx 10. r3(x 2+ 4x + 10 ) dx
11. rl v'4x + 5dx 12. f 3
J 2x + l0dx
13. rTT
/3
4 sinx dx 14. frr
12
- rr / 2
6cosxdx
rrr /3
15. J, (sec 2 x + cosx) dx 16. Jo (secxtanx + sinx)dx
rr/6
1.4 0. 7
17.
f1.
2
(Sx + 1) 6 dx 18.
1 (4x - 1) dx
0 .4
5
19. Ji2sin3 xcosxdx 20. f (1 + cosx) sinxdx
3
-3
4
0.2 ro.4
21. 1 cos 3x dx
0.1
22. Jo sin 2x dx
5
23 .
f (x 6x + 4 sinx + 2) dx
-5
7 - 3 (Be clever!)
24 . fi (cosx + 10x tanx ) dx 3
- (Be very clever !!)
25. fi x - dx (Beware')
2
26. fzft dx (Beware!)
224 Chapter
5: DefiniteandIndefinite
Integrals
For the functions in Problems 27-30, state whether or not the definite integral equals the
area of the region bounded by the graph and th e x-axis . A graph may help.
27. r (x 2 - lOx + 16) dx 28 . r cosxdx
29. J:sin %x dx 30. f(x) = f
8
1
(: 2 -) dx
For Problems 31-36, suppose that
r f(x ) dx = 7, J: g(x) dx = 12, and ( g(x) dx = 13.
Evaluate the given integral or state that it cannot be evaluated from the given information.
31. r f(x) dx 32 . r 4f(x) dx
33. r g(x ) dx 34 . r f(x ) dx
35. r (f(x) + g(x)) dx 36. r (f(x) + g(x)) dx
37. The property of the upper bound of an integral states :
If f(x) < g (x) for all x in [a, b], then f(x) dx < J: J:
g(x) dx . Write the converse of this
statement, then show by example that the converse is false.
38 . "Plus C" Problem: When you write an indefinite integral, you always write "+C, " but
you don't do this with a definite integral . Evaluate the integral
r 2
x dx
by using the indefinite integral f x 2 dx = x3 + C. Then explain what happens to C and
why on e does not write + C for a definite integral.
5-10 A Way to Apply Definite Integrals
You have learned that a definite integral has a physical meaning as the product of
x and y, and has a geometric meaning as the area of a region under a graph. In this
section you will put these two ideas together to arrive at a systematic method for
applying definite integrals .
OBJECTIVE Given a probl em in which a quantity y varies with x, learn a systematic way to write
a definite integral for the product of y and x, and evaluate the integral by using the
fundamental theorem.
Section
5-10:AWayto Apply DefiniteIntegrals 225
Suppose you are driving 60 ft/ sec (about 40 mi/ hr). You speed up to pass a truck.
Assume that t seconds after you start accelerating, your velocity is given by
V = 60 + 6t 112
The "big" problem of finding the distance you have gone from t = 0 sec tot = 10 sec
can be solved by dividing it into little problems. Divide the time interval [O, 10] into
subintervals, each of width dt units. Figure 5-1Oa shows a representative subinterval
and a narrow strip of width dt. Pick a sample point (t, v) on the graph within the
subinterval. If dt is small, the velocity throughout the subinterval will be almost the
same as it is at the sample point . Let y be the number of feet your car has gone . The
30
dr
distance dy it goes in the time interval dt is approximately
10
dy = vdt
dy = (60 + 6t 112) dt.
Figure 5- 1Oa
The total distance is approximately equal to the sum of the distances dy . Thus
Y "" I dy = I (6o+ 6t' 12) dt.
This sum is a Riemann sum. Its limit as dt approaches zero is a definite integral. So
th e distanc e traveled between t = 0 and t = 10 is
( JO The definite inte gral is the exact
Y = Jo (60 + 6t 112)dt distanc e.
10
= (60t + 4t 312) [ = 600 + 4(10 312) - 0- O
0
= 726.49 . .. :::::726 ft.
The thought proc ess in the problem above can be applied to any problem invo lving
a definite integral. The process is summarized below.
Technique:
Applicationol a DefiniteIntegral
To find the product of P = y times x, where y = f(x) and x goes from a to b, do the
following.
a. Divide the interval [a, b] into subintervals. Draw a strip of width dx, correspon din g to
one representative subin terval.
b. Pick a sample point (x, y) on the graph in that subinterval.
c. Write dp as a function of x and dx, using the fact that y is essen tially constant
throughout the subinterval.
d. Add up the dP's and take the limit; that is, integrate .
226 Chapter
5: Definite
andIndefinite
Integrals
Hiiiii A ffl
Problem Set 5-10
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
Ql. Evaluate: f 72xu dx
Q2. Evaluate: ft 72xu dx
... ... ~.
~ ~-. -:
f(x)
Q3. Find y' : y = cos - 1x.
Q4. Find f'(x) : f(x) = x3 sinx. ~ . ~ ...
Q5. Sketch f'(x) for Figure 5-lOb. X
Q6. Is g(x) = 2 + Ix - 11continuous at x = 1?
Ql. Is f(x) = sinx increasing or decreasing at x = 7?
QB, What hypothesis of Rolle's theorem is not a hypothesis of Figure 5-1Ob
the mean value theorem?
Q9. How fast are you going at time t = 9 sec if your displacement is d(t) = lOOtL5 ft?
Q10. In Problem Q9, what is your acceleration at time t = 9 sec?
1. Displacement Problem: Suppose you take a long trip. You are
V
impatient to arrive at your destination, so you gradually let
your velocity increase according to the equation 100
V = 55 + 12t 0 6 ,
where vis in miles per hour and tis in hours (Figure 5-lOc).
Your displacement equals (velocity)(time), but velocity varies . so
a. On a copy of the graph, draw a narrow vertical strip of
width dt. Show a sample point, (t, v), on the graph within
the strip .
b. The distance you go in time dt is approximately equal t= a t= b
to the value of v at the sample point times dt. Write an
equation for dy, the number of miles you go in dt hours. Figure 5-1Oc
c. Write a Riemann sum for your approximate displac ement, y, between t = a and
t = b.
d. Write an expression for the exact displacement you get by adding up all the dy's
and taking the limit as dt approaches zero .
e. How far did you travel in the first hour (t = 0 to t = 1)? How far did you travel in
the second hour? Show that the property of the sum of two integrals with the
same integrand applies to your answers.
f. How fast were you going at the end of 2 hr?
g. Just for fun, see if you can figure out how long it would take to make the entire
300-mi trip, assuming you don't stop for food, fuel, speeding tickets, and so on.
Section
5-10: A Wayto ApplyDefiniteIntegral
s 227
2. Area Prob lem: In this problem you will find the area of a region whose length varies.
a. Sketch the graph of the parabolic region under the graph of y = 6x - x 2 and betwee n
the two x-intercepts. Show a narrow vertical strip of the region of width dx . Show a
sample point (x, y) on the graph within the strip.
b. The area of the strip, dA, is approximately equal to its length, y, at the samp le
point times its width, dx. Write an equation for dA in terms of x and dx.
c. Write a Riemann sum for the total area of the region.
d. Find the definite integral of dA from 2b. Explain why this integral is the exact area.
e. Show that the area of this parabolic region is two-thirds of the area of the
circumscribed rectangle .
3. Work Prob lem : The amount of work needed to stretch a spring
equals the force exerted on the spring times the displacement of F
the end of the spring where the force is being exerted. According 6
to Hooke's law, this force is proportional to the displacement.
Suppose that for a particular spring, the force is given by
F = 0 .6x,
where F is in pounds and x is in inches. Figure 5-lOd shows the
graph of F versus x.
a. On a copy of the graph, draw a narrow vertical strip of width
dx in the region under the graph . Pick a sample point (x, F) Figure 5-1Od
on the graph within the strip.
b. The work, dW, done in stretching the spring by dx is approximately equa l to the
force at the sample point times the displacement dx . Write dW in terms of x
and dx .
c. The definite integral can be interpreted as : "Add up the dW's and take the limit as
dx goes to zero." Use a definite integral to find the exact number of inch-pounds of
work done in stretching the spring from x = 0 to x = 9.
d. Is the force at x = 9 twice that at x = 4.5? Is the work done in stretching the spring
from O in. to 9 in . twice that done in stretching it from O in. to 4.5 in .?
4. Degree-Days Prob lem: A quantity used to measure the expense
T
of air-conditioning a building is the degree-day. If the outside
temperature is 20 degrees above that inside for two days,
then there have been 40 degree-days. Usually the temperature
difference varies. Suppose the temperature difference, T, is
T = 20 - 12cos 2rr(x - 0.1),
where x is time in days (Figure 5-1Oe). X
0.5
a. On a copy of the graph, sketch a narrow vertical strip of width
dx days . Show a sample point (x, T) on the graph within the Figure 5-1Oe
strip.
b. The number of degree-days, dD, during the time interval dx is approximately the
value of Tat the samp le point times the length of time, dx. Write dD in terms of x
and dx .
228 Chapter5: Definite andIndefiniteIntegrals
c. To find the total number of degree-days, D, from x = 0 to x = 0.5, add up all of the
dD's and fin d the limit. That is, find the definite integral from x = 0 to x = 0.5.
d. The answer to 4c is the number of degree-days from midnight to noon. Find the
number of degree-days from noon to the next midnight. What is the total number
of degree-days from one midnight till the next?
5. Heat Prob lem: The amount of heat needed to warm a substance equals the heat
capacity of the substance times the number of degrees the substance is warmed . From
experimental data it is found that the heat capacity for 1 lb mol (a pound mole equals
about 18 lb) of steam at normal atmospheric pressure is approximately
C = - 0.016T 3
+ 0.678T 2 + 7.45T + 796, for Tin the interval [O,30],
where Tis in hundreds of degrees Fahrenheit and C is in Btu per 100 degrees. (A Btu,
or British thermal unit, is the amount of heat needed to warm one pound of water by
one Fahrenheit degree.)
a. Plot the graph of C and sketch the result. Show a narrow vertical strip of width dT
for the region under the graph. Show a sample point on the graph within the strip.
b . The heat , dH , needed to warm the steam by dT degrees is approximately equal to
the heat capacity at the sample point times dT . Write an equation for dH in terms
of T and dT.
c. Write an integral equal to the amount of heat needed to warm one pound mole
of steam from 1000 degrees to 3000 degrees . Evaluate the integral by using the
fundamental theorem.
d. The amount of heat needed to warm 2000 mol of steam is 2000C. How much heat
would it take to warm this amount from 1000 degrees to 3000 degrees? What
property of definite integrals lets you easily find this answer without doing any
more calculus?
e. The actual heat capacity of steam rises at a slower and slower rate beyond 3000
degrees (T = 30). Does this mathematical model give reasonable answers beyond
the given domain?
6. Total Cost Prob lem: A mining company wants to dig a silver mine horizontally into the
side of a mountain. The price per meter for digging the mine increases as the mine
shaft gets longer because it is more expensive to bring out the dirt and the rock and to
shore up the tunnel. Suppose the price, P, in dollars per meter for digging at a point
x meters from the entrance is given by
P = 100 + 0.06x 2 .
a. On a graph of P, draw a narrow strip of width dx . Show
a sample point (x, P) on the graph within the strip.
b. Write an equation for dC, the number of dollars it costs
to dig dx meters. Write an equation for C, the total cost
of digging from x = 0 to x = b.
c. How much would it cost to dig from Om to 100 m?
From 100 m to 200 m? From Om to 200 m? Show that
the property of a sum of two integrals with the same
integrand applies to the three answers.
ion 5l 0: A Wayto Apply Definite Integrals
Sect 229
7. Golf Course Prob lem: Adolph Ball Company is constructing a golf course in Scorpion
Gulch. The plot for one of the putting greens is shown in Figure 5-lOf. To estimate
the area of the green, Mr. Ball draws parallel lines 10 (scale) ft apart and measures
their lengths (feet). From the information given in the table, figure out about how
many square feet of grass sod Mr. Ball must purchase to cover the green. Describe how
you find the answer. Why can't the fundamental theorem of calculus be used to find
this area?
X width y
20 0 80
30 38
40 50
50 62
60 60
70 55
80 51 10 X
90 30 100
100 3
Figure 5-1 Of
5-11 Numerical Integration by Simpson's Rule
and a Grapher
Before you learned the fundamental theorem of calculus, you found approximate
values of definite int egra ls by using the trapezoidal rule or Riemann sums. In
both methods the actual graph is rep laced by segments of other function graphs,
constant functions for Riemann sums and linear functions for the trapezoidal rule
(Figure 5-1 la). The areas of the rectangles or the trapezoids can be found by using
only the y-values at various points in the domain and 6.x.
f(x) . f(x)
Consta nt fun ctio n graph
Linear function graph
a b a b
Riemann sum Trapezoidal rul e
Graph is approximated by Graph is approximated by
zero-degre e functions. first-degree functions.
Figure 5-11 a
230 Chapter
5: Definite
andIndefinite
Integrals
In this section you will learn a similar numerical technique in which the graph of a
function is approximated by parabolas. This technique is called Simpson's rule. Like
the trapezoidal rule, it is useful if the function you want to integrate is specified only
by a table of data . You will also learn how to use your grapher's built-in numerical
integration feature for functions specified by equation.
OBJECTIVE Use Simpson's rule or your grapher's built-in integrate feature to approximate a given
definite integral.
by Simpson's
Integrals Rule
Figure 5-llb shows the function from Figure 5-lla, with parabolas approximating
th e graph. As you can see, there is less error than there would be either for Riemann
sums or for th e trapezoidal rule.
((x) ( (x)
Quadrati c function graph
X
a -h h
Simpson's rul e
Graph is approximated by
second-degree functions.
Figure5-11b Figure 5-11c
It is possible to find the area of a parabolic region in terms of three y-values, one
at the left, one at the middle, and one at the right of the region . Figure 5-llc shows
the region under the parabola
2
y=ax + bx + c,
from x = -h to x = h. The area of the region is
A = fh
- /1
(ax 2 + bx + c ) dx
= t ah 3 + 2ch
Factor (cleverly!). The area is in
= (h ) (2ah 2 + 6c). terms of h.
Set this result aside for a moment while you substitute -h, 0, and h for x in
y = ax 2 + bx + c.
Yo = ah 2 - bh + c
Yi = +C
2
Y2 = ah + bh + c
Section
5-11: Numerical bySimpson's
Integration Ruleanda Gropher 231
Adding the first and third equations gives
Yo+ Y2= 2ah 2 + 2c,
which is interesting because the quantity (2ah 2 + 6c) appears in the equation for A
by integration. Substituting (y 0 + y2) for the quantity (2ah 2 + 2c) and substituting
4Y1 for the remaining four e's gives
A = h(Yo + Y2 + 4Y1)
Area in terms of three y -values.
So the area of a parabolic region may be found by adding the first y-value, four
times the middle value, and the last value, then by multiplying by one third of the
spacing between x-values. Using this property, you can approximate by parabolas
the area of a region under a graph, using only the y-values at regularly-spaced
points. Suppose that integral I is
I= r f(x) dx.
Let x 0 , x 1, x 2 , ... ,Xn be values of x spaced ~x units apart, where x 0 = a and Xn = b .
Let y 0 , y1, y2, ... , Yn be the corresponding values of f(x) (Figure 5-1 ld). Because
it takes three points to determine a parabola, group the strips in pairs and draw
parabolic arcs, as shown in the figure. Integral I is equal to the area of the region
under the graph, and this area is approximately equal to the sum of the areas of the
parabolic regions.
f(x) Function
.. . ~
Y,,-< Y11_ 1 Y11
X
Figure 5-l l d
By using the areas of the parabolic regions in terms of y, you can write
I "' h(Yo + 4Y1 + Y2) + h(y 2 + 4y3 + y4)
+ h(y4 + 4Ys + Y6) + + h (Yn- 2 + 4Yn- l + Yn) .
Factoring out (1 /3 ) (h) from each term, combining like terms inside the parentheses,
and replacing h with ~x gives the following formula, called Simpson's rule.
232 Chapter
5: Definite
andIndef
initeIntegra
ls
Property: Simpson'sRule
If th e int erval [a , b ] is divided into an even numb er, n , of subint ervals of equ al width ~x ,
then th e integra l of f(x) from x = a to x = b is approximat ely equ al to
(t. x)(yo + 4Y1 + 2y2 + 4y3 + 2y4 + + 2Yn-2 + 4Yn- l + Yn) .
Simpson's rule is applied by multiplying each y-value by the appropriate weighting
factor, adding them up, then multiplying the result by one third of the spacing
between x-values .
Example 1 The given data express y as a function of x . Use Simpson's rule with as many
increments as possible to evaluate
ro
30
ydx .
X y
30 83
35 79
40 74
45 68
50 61
55 49
60 37
65 31
70 33
Solution Assume the function has a smooth graph that can be approximated by parabolic
segments . For a small number of data points, such as those in this example,
write down the weighting factor by each point. Then multiply the factor by the
y-value and add the results . The following table shows a reasonable way to
present your solution .
X y factor y x factor
30 83 1 83
35 79 4 316
40 74 2 148
45 68 4 272
50 61 2 122
55 49 4 196
60 37 2 7-!
65 31 4 124
70 33 1 _TI
Sum 1368
Integral "" ~(1368) = 2280.
n bySimpson's
Section5-11: NumericalIntegratio Rule anda Grapher 233
You need only write out the x-values and the appropriate weighting factors.
Accumulate the rest on your calculator. For larger numbers of data points, write a
program that multiplies, accumulates, and does the final calculation.
Example 2 The data shown express y as a function of x. Use Simpson's rule with as many
increments as possible to evaluate
r f (x ) dx.
X y
1 192
3 249
5 398
7 544
Solution The problem cannot be worked by Simpson's rule - there must be an even
number of increments. Because there are four data points, there are only three
increments between them. The integral can be approximated by the trapezoidal
rule .
by a Grapher
Integrals
Your grapher has a built-in feature that will allow it to evaluate approximately
an indefinite integral. Internally, the method used may resemble Simpson's rule.
Example 3 shows two ways to evaluate an integral with a typical grapher. You may
need to consult your instruction manual to find out how to evaluate on your grapher.
~--sinx dx
Example 3 Evaluate: l0.3 X
Solution The numerical integration feature on a typical grapher might look like this.
fnlnt( (sin x) /x, x, 0.3, 4)
The integrand appears first, followed by a comma. The name of the variable of
integration appears next; x, in this case. The third and fourth numbers are the
lower and upper limits of integration, respectively. Thes e four quantities are in
the same order you would use in pronouncing 'The integral of (sin x) /x , with
(sinx)/ x respect to x, from 0.3 to 4."
1
Upon executing this command, the answer is 1.459699 ....
4 X
Some graphers have a numerical integrator that can be used directly from the
1 graphing screen . The equation is stored on the y-menu. Then the graph is
ff(x) dx = 1.4596991 plotted, using a friendly window containing the two limits of integration. Upon
accessing the integral command on a menu such as CALCULATE,the grapher
Figure 5-11 e allows you to trace to the lower limit, press ENTER,trace to the upper limit,
and press ENTER.
The graph er will display the answer and may shade the region on the graph, as
shown in Figure 5-lle . The answer, shown on the screen, is 1.4596991....
234 Chapter
5: Definite
andIndefinite
Integrals
Example 3 illustrates an instance in which the fundamental theorem cannot be
used. There is no elementary transc endental function equal to the indefinite integral
Jsi: x dx. Later in your math emat ical career , you will learn that this is an example
of th e sine-integral function. The next exampl e shows you how to graph a function
defined by a definite integral.
x 1
Example 4 Let f(x) = I -t dt.
l
a. Explain why 1/t is not the derivative of a pow er fun ction, and thus why you
cannot yet us e the fundamental theorem to evaluate th e integral.
b. Use your grapher's numerical integration feature to plot the graph of f from
x = 0.1 to about x = 10.
c. Make a table of values of f(x) for x = l, 2, 3, . .. , 6.
d. Why is f (x) negative for values of x less than 1?
e. See if you can find any relationship among f (2) ,f (3) , and f (6) .
Solutions a. 1/t equals t - 1 . If this were the deri vative of a pow er, the antiderivative's
exponent would be zero. Its coefficient would thus be 1/ 0, which is infinite .
Because the antiderivative of 1/t is not known yet, the fundamental theorem
cannot be used.
b. On th e y = menu, en ter the equation. The instru ction on a typical grapher
would be
Yi = fnlnt(t - 1 , t, 1, x) .
The graph is shown in Figure 5-11 f.
c. Use your gra ph er 's table or trace feature to get the table shown.
d. The values of f(x) are nega tive for x < l because th e upper limit of integration
is less than the lower limit. This makes all the values of dt negativ e while the
values of t - 1 are positiv e. Thus each t - 1 dt is negative .
e. You should discover from the table that
f (6) = f(2) + f (3) .
X f (x)
1 0
2 0.693147 ...
3 1.098612 ...
4 1.386294 ...
5 1.609437 ...
6 1. 791759 ...
Figure 5-1 l f
Sec IntegrationbySimpson's
tionS11: Numerical Ruleanda Graph
er 235
Problem Set 5-1 1
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
Q 1. Evaluate: J(x 2 + l) dx
Q2. Evaluate: f~3 (x 2 + l) dx
Q3. Evaluate : f sec 2 x dx
d
Q4. Evaluate: dx (sec 2 x)
QS. Sketch a velocity versus time graph for a moving object.
Q6. Show a strip of width dt on the graph you sketched in Q5.
Ql. Show a sample point (t, v) on the graph within the strip in Q6.
QB. Write the displacement of the object in Q5 for the time interval dt.
Q9. Write a Riemann sum for the approximate displacement of the object from t = a to t = b.
Q10. Write an expression for the exact displacement of the object from t = a to t = b .
l. Velocity Problem: People who sail ships at sea use dead reckoning time (min) speed (kn)
to calculate the distance a ship has gone. (The term dead reckoning 0 33
comes from "ded-reckoning," which is short for "deduced reck- 2 25
oning.") Suppose that a ship is maneuvering by changing speed 4 27
rapidly. The table shows its speeds at 2-min intervals . (A knot- 6 13
abbreviated kn-is a nautical mile per hour. A nautical mile is about 8 21
2000 yd.) 10 5
a. Use Simpson's rule to find the distance traveled in the 12-min 12 9
time interval.
b. Find the distance again by using the trapezoidal rule.
c. Which answer should be closer to the actual distance? Explain.
2. Spleen Mass Problem: Figure 5-1lg is a CAT scan that shows a
cross section of an eleven-year-old girl's body, looking up from
below. The spleen is located at the right of the cross section (the
girl's left). Doctors want to know whether or not the spleen has
a mass that is within the normal range, 150 g to 200 g. Using
the information contained in this and other CAT scans taken
at 0.8-cm intervals up and down the girl's body, the doctors
measure the spleen's cross-sectional areas.
D(cm) A (cm 2 ) D (cm) A (cm 2 )
0 6.8 4.8 38.4 Figure 5-1 lg
0.8 6.8 5.6 33.9
1.6 20 .1 6.4 15.8
2.4 25.3 7.2 6.1
3.2 29.5 8.0 2.3
4.0 34.6
236 Chapter
5: Definite
andIndefinite
Integrals
a. The volume of an object equals its cross-sectional area times its thickness. If the
cross section's area varies, the volume is the integral of the area with respect to the
thickness. Use Simpson's rule to estimate the volume of the spleen .
b . The density of the spleen is about the same as that of water, namely, 1.0 g/ cm 3 Is
the girl's spleen within the normal mass range ?
3. Tensile Strength Test Problem: The tensile
strength of a metal bar is measured by the
amount of force needed to str etch the bar until --1 ,- (str
Displa ce ment
e tch)
it breaks (Figure 5-1 lh). At first the force varies
Pull
directly with the distance stretched. After a
certain point, the bar begins to deform (to yield)
and the force goes down. The amount of work
done in breaking the bar is equal to the force
times displacement (the distance stretched). 63~ Brea ks!
- Pul
Suppose the following data is measured.
Figure 5-1 l h
inches pounds inches pounds
0.00 0 0.30 290
0.05 120 0.35 280
0.10 240 0.40 270
0.1 5 360 0.4 5 270
0.20 370 0.50 190 (breaks)
0.25 330
a. Draw a scatter plot of force versus displacement.
b. Work equals force times displacement. If the force varies, the work must be found
by integration. Use Simpson's rule to find the inch-pounds of work done in breaking
the bar.
4. Heat Capacity Problem: The amount of heat needed to warm a mole of gas by 1 degree
is called the molal heat capacity of the gas. The amount of heat added to a mole of
gas in warming it by a certain number of degrees is, then, equal to the heat capacity
times the number of degrees. Most gases have heat capacities that change as they are
heated . For example, steam at the normal boiling point of water has a heat capacity
just over 8 (Btu/ lb mol)/ F (The abbreviation Btu stands for British thermal unit,
0
which is the amount of heat needed to warm one pound of water by one Fahrenheit
deg.) However, if the steam is at 5000 F, its heat capacity is over 13 (Btu/ lb mol)/ F.
Therefore calculus must be used to find the amount of heat added! Figure 5-1 li (next
page) shows graphs-from Hougen and Watson's Chemical Process Principles Charts
(John Wiley, 1946)- of heat capacity versus temperature. Use the chart to find the
heat capacity of steam (H2 0) for each 500 degrees from 500F to 4500 F. Then use
Simpson's rule to find the number of Btu needed to warm a pound mole of steam from
500F to 4500 F.
5. Sine-Integral Function Problem: The sine-integral function, Six, is defined by
.
SIX= l x -sintd
- t.
0 t
a. Use your grapher's integrate feature to plot the graph of Six from about x = - 20
to x = 20 and a y window of - 2 to 2. Sketch the result.
b . (Sin x)/ (x) takes on the indeterminate form 0/ 0 at x = 0. What limit does (sinx) / (x)
seem to approach as x approaches zero?
Section
5-11:Numerical bySimpson's
Integration Ruleanda Gropher 237
,_,_ t( ~ .L!
ti,{_U
't. ,__
J.,-.,. i-trtt:tttt_ l:i:,I _ljj_
--~I-+
--so, _g_
j..J u IJ~H-11
tt::r_ - .-Ht:J_J'+I
_-ti_02\-t =;;: i+H:!_
1
rn:t-:;:r+B
-;-t-1
14
-, . ~ - -. j:JJ: Jf.fft - ti ffi_ . --1:rt
50 2 i j li ~
tllt:i:..t+i -+ .
:1 1
-. -}ai=~wr:t.!i..'
-T t - +
13
- ..1.
-I : ; f . .r -I{ T .
+ .. l : _; 1.-1-
,-; . I J
12
TRUE MOLAL HEAT CAPACITIES
11
OF GASES ~T CONSTANT
- ~r , ft
14- ' '
--~ t~
-- -t-'+. PRESSURE
= u3*- ' -~. I -
' . ~ . -rrt +1-fflli'
, - tJ+Jlffi-j
I . +,..
l:J:J+ll+-+flt+JCH=Ji+m:l-11=ITT=ltt+t
- i=i 1-1
.,.+Ti
' ":hw., -c -~
~ ! ffib .
-1
10 J:' . p: . ~il '' : . ' "',.. ++:tL: --
~r-, 1 ~:-
- ~
j
' .- -
--
.0 9
i v I
f . - -, i t+ +
-I ~ 1<
+ ._ -I
:,
t-
a, e
-
,;L'
1 -1
-I
t -- -- ,
T -
-t
AIR t.. ,
I I j !,-.,
N2 ttJ_
-1-
:rH:i.t a+
,
', r
-i
. _
'
_~
-
.1 _
+ -:j: '
.:I
r-.
I H+'+++H-H
I- , . ~ _,_::tttttt-tffi-tH.+cH-H
rn:W:J:-11-7_,rtl::_r-
7
-H+~+H-H+i-++++l
'-H+I
l:t .- . :i:r-
- J + -r--t '
. -W.-4:Il+-1-1-
1-
+'lH'-H+++H-
I l-J--.
! -
,.I- -+
t: !:!J,-= - 1- wt
l_ -
tlt:t 1
IL -l:l-r
::i:
~=-
ti-~- : d
L
f :tr':{ tl1:i- --~ :j:
..Jiti
$ T_c,. , :t
L
. 4~. f.. :H:j:+ltl+J t:irr:i:r-4+1.f1. Li
ii
1000
' 1500T . i-
J J}l
2000
-}j i
U :
2500 3000
d: ~llL1- ..i.crntt
3500
H-l+i+l+I=
4000 4500
r :i d.
tW : .J:ffl
5000
Temperature, F
Figure 5- 1 l i
c. The National Bureau of Standards Handbook of Mathematical Functions lists Si 0.6
as 0.5881288096 . How close does your grapher's value come to this number?
d. Does Six seem to approach a limit as x approaches infinity ? If so, what number do
you conjecture this limit is? If not, explain how you know why not.
e. Compare the graphs of Six and f(x) = (sinx) / (x). How do you know from the
graphs that one is the derivative of the other? y
6. Error Function Problem: If you measure statistics on a larg e 0.5
population, such as test scores or peoples' heights, the
numbers are often normally distributed. That is, most of
th e data points are close to th e mean and fewer are farther
away. In statistics courses, you will learn that for a normal
(or Gaussian) distribution, th e relative frequency with which 2
-2 -1
a particular data point occurs is given by -x X
1 2
Figure 5-1 l j
y = foe - I '
238 Chapter
5: Definite
andIndefinite
Integrals
where t is the number of standard deviations the data point is from the mean and e is
the base of natural logarithms (approximately 2.718, as you will learn in Chapter 6). The
area under the graph from t = -x tot =xis equal to the fraction of the population within
x standard deviations of the mean. This area is called the error function of x (erf x).
erf x = }rr( e- 12
dt
a. What property of definite integrals explains the fact that the coefficient in the
integral is 2/.jrr but that the coefficient in the equation for y is only 1/.jrr?
b. Plot the graph of erf x for x = 0 to x = 2. Use your grapher's built-in function ex.
Sketch the result.
c. If x is enough standard deviations from the mean, virtually all the data should be
within that number of standard deviations. How do the values of erf x confirm this
fact?
d. The Nationa l Bureau of Standards Handbook of Mathematical Functions lists erf 0. 5
as 0.5204998778. How close does your grapher's value come to this number?
2
e. Compare the graphs of erf x and f(x) = 2/.jrre - x How do you know from the
graphs that one is the derivative of the other?
For Problems 7-10, evaluate the integral approximately by using your grapher's built-in
integrate feature.
7.
1.4
l cosx dx
0.3
8. r (x 2 - 3x + 5) dx
9. J 2" dx
0
3
10. l
0. 1
l. 4
tanxdx
11. Answer Check Problem: Evaluate the integral in Problem 7 exactly by using the
fundamental theorem. Find a decimal approximation for your answer. How closely
does your grapher's numerical answer match the exact answer?
12. Simpson's Rule from Equation Problem: Use Simpson's rule with n = 4 increments to
estimate the integral given in Problem 9. You should be able to figure out how to do this
without having seen an example! Estimate the integral again by using the trapezoidal
rule. Assuming that your grapher's answer is the actual answer, does Simpson's rule
really seem to give a more accurate approximation than the trapezoidal rule? Justify
your answer.
13. Program for Simpson's Rule from Data: Write a program to evaluate integrals by using
Simpson's rule. Data for they-values may be entered into a table such as L 1 before the
program is run. Input should include the number of increments, n, (one less than the
number of data points) and the value of ~x. An easy way to calculate the terms is to
start a counter k at 2, then have the program evaluate L1 (k - 1) + 411 (k) + I 1 (k + 1).
Then increment k by 2 and repeat the calculation, each time adding the result to the
sum of the previous calculations. When k exceeds n, the program should leave the loop
and evaluate the integral by multiplying by ~x and dividing by 3. The output should
be this value of the integral. Test your program by using the data given in Problem 1.
14. Program for Simp son's Rule from Equation: Write another program to evaluate integrals
by using Simpson's rule. This time, they-values should be calculated by equation rather
than from a table of data. The equation can be stored in the y-menu, say at y 1 The input
should be the upper and lower limits of integration and the number of increments, n.
The program should then calculate ~x as (b - a) /n. The computations can be done by
IntegrationbySimpson's
Section5 11: Numerical Ruleanda Grapher 239
finding Y1(x - 6-x) + 4Y1(x) + Yi(x + 6-x). Again, the output should be the approximate
value of the integral. Test your program by using n = 100 increments to evaluate the
integral given in Problem 5. The answer should be 0.6899295234.
15. Relative Accuracy Problem : In th.is problem you will investigate the relative accuracy
of integrals found by Simpson's rule, the trapezoidal rule, and your grapher's built-in
integrate feature.
a. Evaluate forr sinx dx in four ways.
i. By Simpson's rule, with n = 50 increments
ii. By the trapezoidal rule, with n = 50 increments
iii . By the built-in integrate feature
iv. Exactly, using the fundamental theorem
b. Which of the first three techniques in 15a give you an answer closest to the actual
value in part iv of l 5a? By what percentage does each answer differ from the actual
answer?
5-12 Chapter Review and Test
In previous chapters, you learned about definite integrals and antiderivatives. In this
chapter, you learned about the symbols and formal definitions for these concepts.
The mean value theorem and Rolle's theorem led to the fundamental theorem of
calculus, which allows you to find definite integrals exactly, by algebra, rather than
just numerically or graphically . This connection between the antiderivative and the
definite integral explains why the word integral is used also for the antiderivative .
You learned a method for applying definite integrals by slicing a region under a
graph, picking a sample point, forming a Riemann sum, then integrating. Finally you
learned two more ways of evaluating definite integrals numerically : Simpson's ru le
and your grapher's built -in integrate feature .
The Review Problems below are numbered according to the sections of this chapter.
The Concepts Problems allow you to apply your knowledge to new situations. The
Chapter Test is more like a typical classroom test your instructor might give you .
Review Problems
RO. Update your journal with what you've learned since the last entry . Include such things
as those listed here .
The one most important thing you have learned in your study of Chapter 5
Which boxes you have been working on in the "define, understand, do, apply" table
The difference between definite integral and indefinite integral
The fundamental theorem of calculus, how it is proved, and what it is useful for
Application of definite integrals to real-world problems involving a product of
variables
Simpson's rule and your grapher's numerical integration feature
Any ideas about calculus you're still unclear about
24 0 Chapter
5: DefiniteandIndefinite Integrals
Rl. Heat Capacity Problem: The amount of heat it takes to heat a
pound of a substance from one temperature to another equals
the heat capacity of the substance (British thermal units per
degree) times the number of degrees by which the substanc e is
heated. Most substances have a heat capacity that depends on o.s
temperature. For instance, the heat capacity of steam at high
temperatures is given approximately by
C(T) = - 0.4 + 0.3 logT,
where C(T) is in Btu/ deg, Tis temperature in Fahrenheit, and
log is the common base-10 logarithm. A graph of this function
1000 5000
is shown in Figure 5-12a.
a. Explain why the number of Btu needed to heat steam is Figure 5- l 2a
equal to a definite integral.
b. Find the approximate number of Btu needed to heat a pound of steam from
l000 F to S000F. Show how you arrive at your answer.
c. Find the average heat capacity for the interval 1000 deg to 5000 deg. That is, find
the number of Btu/deg you could multiply by (5000 - 1000 ) to get the number of
Btu you found in Rlb .
R2. Find the general equation for f(x), the antiderivative function.
a. f '(x) = 4x 317 b. f'(x) = 10cos2x c. f'(x ) = (12x + s) - 3
R3. a. For f(x) = sin rrx, find an equation of the linear function that best fits f at x = l.
What is the error in approximating f (x) by th.is linear function if x = 1.1? If
X = l.001?
b. Find dy.
xs x-3
i. y = csc 5 2x ii.y =s-T iii. y = ( 7 - 3x) 4
c. Find the general equation for the antiderivative, y.
i. dy = secxtanxdx ii. dy = (3x + 7) 5 dx iii . dy = 5 dx
d. For y = (2x + 5) 112 , do the following.
i. Find dy in terms of dx.
ii. Find dy if x = 10 and dx = 0.3.
iii. Find 6.y if x = 10 and dx = 0.3.
iv. Show that dy is approximately equal to 6.y.
R4. a. Write the definition of indefinite integral.
b. Evaluate the indefinite integral.
i. f l 2x 213 dx ii. f sin 6 x cos x dx iii. f (x 2 - Bx + 3) dx
RS. a. Write the definition of integrability.
b. Write the definition of definite integral.
c. Use the following to evaluate approximately J;
2 sec x dx .
4
i. The upper Riemann sum with six increments
ii. The lower Riemann sum with six increments
iii. The midpoint Riemann sum with six increments
iv. The trapezoidal rule with six increments
Section5-12: Chapter
Review
andTest 241
d. Draw a diagram that shows the meaning of each of the sums in RSc.
e. Let I= It
xdx.
i. Divide the interval [O, SJ into n subintervals, each of equal width 5/n. Write a
formula for the upper sum of I in terms of n.
ii. Transform the formula in RSe.i to eliminate the ellipsis ( . . .). Recall from
algebra that (1 + 2 + 3 + + n) = (0.Sn)(n + 1).
iii. Evaluate I exactly by taking the limit of the formula given in RSe.ii as n
approaches infinity (and thus ~x approaches zero).
iv. Use appropriate geometry to show that your answer to RSe.iii is correct.
R6. a. What is the difference between the hypotheses of a theorem and the conclusion?
b. A long pendulum swings slowly back and forth. Its displacement, d(t) meters,
from one wall of the museum in which it hangs is
d(t) = 20 + 3 sin ~t,
where t is time in seconds. What is its average velocity from t = 0 sec to t = 2
sec? At what time in the interval (0, 2) does the instanta neous velocity equal this
average velocity?
c. Find an interval [a, b] on which the hypotheses of Rolle's theorem are satisfied
for the function g(x) = x 413 - 4x 113. Then find the value of x in (a, b) at which
the conclusion of the theorem is true . Show that g is not differentiable at one
of the two endpoints of [a, b], and explain why this fact is consistent with the
hypotheses of the theorem.
d. Sketch a graph that shows why the conclusion of the mean value theorem might
not be true for [a, b] if the function is not differentiable at some value of x in
(a, b) .
e. Sketch a graph that shows why the conclusion of the mean value theorem might be
true in (a, b) even though the function is discontinuous at one of the endpoints
of [a, b] .
f. In the proof of the mean value theorem for f(x) on [a, b], tvvonew functions were
defined, g and h . Tell how they were defined and how, as a consequence, you can
use Rolle's theorem as a lemma to prove the mean value theorem.
g. Let f(x) = 3 + Scos 81rx. Plot the graph on your grapher. Quickly find all the points
in the interval (0, 1) at which the conclusion of Rolle's theorem is true .
h. What can you conclude about the values of r(x) and s(x) if r ' (x) = s' (x) for all
values of x in an interval?
I:
R7. Find a Riemann sum, R 3 , for x1.s dx by choosing the sample points in a special way:
Let g(x) = f x L 5 dx, the indefinite integral. Let c1 , c2 , and c3 be the values of x in the
three subintervals, (1, 2), (2, 3), and (3, 4), at which the conclusion of the mean value
theorem is true for the integral, function g. Use these numbers as sample points for
R 3 for the given integral. If the sample points are picked in this way, what can you
conclude about the answer you get for the Riemann sum7
RB. a. Evaluate f! 1 (10 - x2 ) dx exactly by using the fundamental theorem of calculus .
b. Check your answer to R8a by using the trapezoidal rule with a reasonable number
of increments . Show that the answer is close to the exact answer.
242 Chapter
5: DefiniteandIndefinite Integrals
c. Find the midpoint Riemann sums M10 , M100 , and M1000 for the integral in R8a.
Show that the greater the number of terms in the Riemann sum, the closer the
sum gets to the actual value of the integral.
R9. a. Evaluate the following integrals by using the fundamental theorem of calculus.
Check your answers by using a Riemann sum or the trapezoidal rule.
i. f{ x - 2 dx ii. J; (x 2 + 3) 5 (xdx) iii. forr(sinx - 5) dx
b . Sketch a graph of R9a.iii to show why you get a negative answer.
c. Quick! Evaluate t ~(4 sinx
0 + 6x 7 - 8x 3 + 4) dx.
d . Sketch a graph to illustrate the property I: f(x) dx = J; f(x) dx + f1f(x) dx.
RlO. Disp lacement Prob lem : Suppose that as a rocket rises from its launching pad, its
upward velocity, v feet per second, is given as a function of time, t seconds, by
V = 150t 0 5 .
a. Sketch the graph of v versus t. Show a narrow strip of width dt for the region
under the graph . Show a sample point (t, v) on the graph within the strip .
b . The disp lacement, dy, of the rocket in this time interval is approximately equal to
the velocity at the sample point times the time, dt . Write dy in terms oft and dt .
C. Write a Riemann sum for the vertical displacement of the rocket in the time
interval [O,9].
d. Find the limit of the Riemann sum given in RlOc as dt approaches zero . What
mathematical quantity is this7
e. Show that the displacement of the rocket for the time interval [O,9] is the sum of
the displacements for the intervals [O,4] and [4, 9].
Rll . a. Sketch a graph to illustrate the idea behind Simpson's rule and why you need to
use an even number of increments.
b . Sketch a graph to show why Simpson's rule is expected to give a more accurate
value of an integral than the trapezoidal rule .
C. The table below shows the velocity, v(t) meters per minute, of a moving object.
Write an integral to represent the displacement of the object from its position
at time t = 3 min . Use Simpson's rule appropriately to calculate the distance the
object travels between t = 3 min and t = 5 min.
v(t) v ( t)
3.0 29 4.2 28
3.2 41 4.4 20
3.4 50 4.6 11
3.6 51 4.8 25
3.8 44 5.0 39
4.0 33
0
d. You can't yet evaluate f} log x dx using the fundamental theorem because you
haven't learned an algebraic way to find the indefinite integral. Use your grapher's
built-in integrate feature to find the value of this integral. Sketch the graph of the
integrand and use the graph to explain why your answer is reasonable .
Section512: Chapter Rev
iewandTest 243
~
Conce pts Problems
Cl. In this problem you will investigate rb- - 1- dx for various values of b.
Jo X 2 + 1
a. Evaluate the integral approximately, using midpoint Riemann sums with 100
terms for b = l, b = 2, b = 3, b = 4, and b = 5.
b. Evaluate the integral given in Cla for b = 100 by using the midpoint Riemann sum
with n = 1000 terms .
c. Plot the graph of the value of the integral as a function of b. Use the same scale
on both axes . What value does the integral seem to be approaching as b gets very
large?
d. On another piece of graph paper, plot y = tan x. Use the same scales on this graph
as you used in Cle .
e. Turn over the graph paper you used for Cld and look at the graph by holding the
paper up to a light . Rotate the paper if necessary . What do you n otice?
C2. Mean Value Theorem for Quadratic Func ti ons: Let f be the general qu adratic function ,
f(x) = ax 2 +bx+ c. Given any interval [d, e], show that the point x =kin (d, e)
at which the conclusion of the mean value theorem is true is the midpoint of that
interval .
C3. Sum of the Squares Problem : In Problem 12 of Problem Set 5-5, you found that the
ft
exact value of x 2 dx can be found by using the sum of the squares of the who le
numbers. You also learned that the indefinite integral f x 2 dx is a cubic function of
x . In this problem you will see that the sum of the squares of the whole numbers is
also a cubic function. Let S(n) be the sum of the squares of the integers from zero
through n . That is,
S(n) = 0 2 + 12 + 2 2 + 3 2 + + n 2 .
a. Find S(O), S(l), S(2), and S(3).
b . Assume that S(n) is a cubic function of n . That is,
S(n) = an 3 + bn 2 + en+ d,
where a, b, c, and dare constants. Find the particular equation for S(n) by finding
these constants.
c. The constants in the equation turn out to be fractions . Factor out the appropriate
fraction, leaving a polynomial with integer coefficients inside th e parentheses .
Then factor the polynomial. You can check a table of mathematical functions,
such as CRC Tables, to see if you are correct.
d . Use the equation you found in C3c to predict S(4) and S(5). Show that your
answers are correct by actually adding the squares of the integers.
e. Predict the value of S(lOOO).
f. Use mathematical induction to prove that the equation given in C3c works for any
positive integer n.
C4. Sum of the Cubes Prob lem: Let S ( n) = 0 3 + 13 + 23 + 33 + + n 3 be the sum of the
cubes of the integers from zero through n. Find S(O), S(l), S(2), S(3), and S(4). You
should find that all these numbers are perfect squares! Then derive a closed formula
(no ellipsis) for S(n) in terms of n, as you did in Problem C3. You may assume that
244 Chapter
5: Defin
ite andIndefinite Integrals
the formula is quartic: S(n) = an 4 + bn 3 + cn 2 + dn + e. Find values of the constants
a, b, c, d, and e, then use mathematical induction to prove that your formula is right.
C5. Radio Wave Integral Problem: AM radio signals are transmitted by sending a high-
frequency wave whose amplitude varies in the pattern of the sound being carried.
(The amplitude is modulated, hence the abbreviation AM.) Figure 5-l 2b shows
f(x) = 4 sinx sin lOx,
where f(x) is the strength of the signal at any instant, x, in time. The sound
represented by y = 4 sinx, with a frequency of 1 cycle per 2rr x-units, is being
"carried" by the signal y = sin lOx, with a frequency of 10 cycles per 2rr x-units.
((x)
5
-5
Figure 5-l 2b
a. Evaluate approximately the integral
r 4 sinx sin lOx dx.
Show that the integral is very close to zero .
b. You do not yet know how to use the fundamental theorem to evaluate the integral
because it is the integral of a product. However, the integrand can be changed to
a sum by using the trigonometric property
2 sinA sinB = -cos (A+ B) + cos (A - B).
Make the transformation, then show that the integral is exactly equal to zero by
using the fundamental theorem to evaluate it.
c. The integral given in C5a is represented by the shaded region in Figure 5-l 2b.
What is the geometrical interpretation of the fact that this integral equals zero?
d. Show that the integral
f
0
rr 4 sinx sin nx dx
is equal to zero for any integer n > l.
e. Record in your journal the fact that this kind of integral is related to Fourier
series, which you will study later in your mathematical career.
Section
5-12: Chapter
Review
andTest 245
C6. Riemann Sums with Unequal Increments: Figure f (x)
5-12c shows the graph of f(x ) = 1.2". The interval
[1, 9] is partitioned into subintervals of unequal
width. The largest value of 6.x is called the norm
of the partition, written IIP II. An upper and lower
sum, Un and Ln, are shown.
a. Why is the difference Un - L 11 no greater than
IIPll(l.2 9 -1.2 1 )?
X
b. Suppose that the number of subintervals
is allowed to approach infinity in such a 1--1 9
\ Norm= largest t;x
way that the limit of IIP II is zero . Use the
observation you made in C6a to conclude that Figure 5-l 2c
f is integrable on [l, 9].
c. Prove that g(x) = l /x is integrable on [l, 4].
d. Could the reasoning of this problem be used to prove that '1(x) = sin xis integrable
on the interval [O, 3]? State why or why not .
Chapter Test
Tl. Write the definition of indefinite integral.
T2. Write the definition of definit e integral.
T3. State the fundamental theorem of calculus.
Ii
T4. Given 3x dx, do the following.
a. Find the upper and lower sums U4 and L4 for this integral and find their average.
How does this average compare with the integral by trapezoidal rule with n = 4?
Does this average overestimate or und eres timate the actual value of the integral?
Explain.
b. Find the midpoint Riemann sum M4 for this int egral. Explain why this Riemann
sum underestimates the actual value of the integral.
c. Evaluate the integral approximately by using Simpson's rule with n = 4 increments.
d. Evaluate the integral by using your grapher's numerical int egration feature.
Evaluate th e indefinite integrals.
17
TS. f<3x - ll) dx T6. f (cos 2x) 4 sin 2x dx
For Problems T7- TlO, evaluate the integral exactly by using the fundamental theorem.
You may check your answers by using a numerical method if you have time .
4
T7. rrr /
Jrr/6
cos 2x dx TS. J_\sec 2
xdx
2 8
T9. [ (12x 3 + 10x 2 ) dx Tl 0. Ji 7x - 2f3 dx
Tll. Sketch a graph that clearly shows that you know the hypotheses and the conclusion
of Rolle's theor em.
Tl2. Let f(x) = 0.3x 2 and let g (x) = f f (x ) dx.
246 Chapter
5: DefiniteandIndefini
te Integrals
a. Integrate to find an equation for g(x ) . Set the constant of integration, C, equal to
zero.
b. On graph paper, plot graphs off and gin the domain [O, 4]. Plot them on different
sets of axes, but use the same scales for both gra phs.
c. Draw the secant line on the g graph from (1, g(l)) to (4, g(4)). Find its slope.
Then find the point x = c in (1, 4) at which the conclusion of the mean value
theorem is true for function g . Draw a line through the point (c, g(c)) parallel to
the secant line and thus show that it is tangent to the graph of g at that point.
d. Using the value of c you calculated in Tl2c, find the point (c, f(c)) on the graph
of f. Draw a rectangle whose base is the segment from x = 1 to x = 4 and whose
altitude is f(c ). How does the graph show that the area of this rectangle is equal
to the value of the definite integral of f(x) dx from x = 1 to x = 4?
Section
5-12: Chapter
Review
andTest 247
..
CHAPTER
6 The Calculus of Exponential
and Logarithmic Functions
.. s
... ,
,.,.:;
..
Rabbits introduced into Australia in the middle 1800s had no
natural enemies. Their population grew unchecked. Such unrestricted
population growth can be modeled by exponential functions, in which
the exponent is a variable. Exponential functions and their inverses,
the logarithmic functions, do not behave the same way as the power
functions you have analyzed in calculus so far.
249
~I Mathematical Overview
Exponential functions, where the variable is an exponent, model
population growth. In Chapter 6 you will learn the calculus of these
functions by exploring the integral of the reciprocal function, which
turns out to be a logarithmic function. You will learn the calculus of
these functions in four ways.
Graphically The logo at the top of each even-
numbered page of this chapter shows
the natural logarithm function and
its inverse, the base e exponential
function.
ln
Numerically X exp x In x X
1 2.718 ... 0
2 7.389 ... 0.693 ...
3 20.085 ... 1.098 ...
x 1
Algebraically ln x = J,
1
t dt, the definition of natural logarithm.
Verbally I didn't realize that it was possible to define a function to be the definite
integral of another function. The advantage of doing this is that you
know immediately the derivative of that new function. I was surprised
to find out that the integral of 1/ t from t = 1 to t = x turns out to be a
logarithm function. I shall have to remember that derivatives and
integrals of exponential function s are not the same as for powers.
250
6-1 Integral of the Reciprocal Function: A Population
Growth Problem
You have learned how to differentiate power functions with constant exponents,
such as
f(x) = x5.
You have also shown by example that functions with variable exponents, such as
f(x) = 5"
cannot be differentiated by using the same rule as used for power functions. A
frontal assault on finding the derivatives of such exponential functions is rather
difficult. The method of choice, then, is to approach the exponential function
through its inverse, the logarithmic function. In this section you will develop some
background needed for this approach.
OBJECTIVE Work the problems in this section, on your own or with your study group, as an assignment
after your last test on Chapter 5.
Exploratory Problem Set 6-1
Population Problem: A population (such as people, wildlife, or bacteria) tends to grow at
a rate proportional to the number of individuals present. For example, the more people
there are in a community, the more babies are born per year. Assume that a small town has
a population of P = 1000 people now (time t = 0 yr) and that the population is growing
at an instantaneous rate of 5% per year . In the following questions you will predict the
population after various numbers of years.
1. Explain why the differential equation dP /dt = 0.05P expresses the information above.
2. What are the units of dP /dt?
3. The differential equation in Problem 1 can be transformed by separating the variables .
1
p dP = 0.05 dt
If N is the value of P when t = 10 yr, then integrating both sides gives
1/ P
(N _pldP = (!O0.05 dt
11000 Jo
Evaluate the right-hand integral. Explain why the indefinite integral _
0 001
J (1 / P) dP, which equals J p - 1 dP, cannot be evaluat ed by using
the normal method for power functions.
p
4. Figure 6-la shows the integrand 1/P. The integral on the left 1000 N
in Problem 3 equals the area of the region under this graph.
Evaluate this integral numerically for N = 1500 and N = 2000. Figure 6-1a
5. Find, numerically, the value of N that makes the two integrals in Problem 3 equal to
each other. Write a paragraph that describes how you accomplished this computation.
6. Predict the population of the town when t = 20 yr.
7. Is the P-versus-t graph linear or nonlinear? Justify your answer.
6l : Integralof the Reciprocal
Section Function:
A Population
GrowthProblem 251
6-2 Antiderivative of the Reciprocal Function
In Section 6-1, you saw that the integral of the reciprocal function can appear in
problems of population growth. Although f (1 / x) dx can be written as the integral
of a power, f x- 1 dx, you cannot use the pattern "Increase the exponent by 1, divide
by the new exponent, and add C." This pattern would result in division by zero .
fx- 1
dx = 6
x
0
+C
In Section 5-11, you were introduced to the sine-integral function and the error
function, which were defined by definite integrals. In this section you will investigate
another such function,
x 1
g(x) =
I t I
- dt
and you will see how this problem leads to the antiderivative of 1/ x.
OBJECTIVE Imestigate the function g(x) = g (1 / t) dt and reach some conclusions about its values,
graph , and derivative.
You may work in groups, with guidance from your instructor as necessary.
Problem Set 6-2
l. Let g(x) = J;(l / t) dt . Figure 6-2a shows that g (x) is the area of a region under the
graph of y = 1 / t. Explain why this area is a function of x.
y = 1/ f
2. Use a suitable numerical method to make a table of values of
g(x) for x = 8, 7, 6, 5, 4, 3, 2, 1, and 0.5. Keep six decimal places.
Ar ea = g(x)
3. Explain why g(x) is negative if xis less than l. A graph may help.
4. Figure 6-2b shows the graph of function g. Use your grapher's
numerical integration feature to plot this graph. To avoid f= l f=X
division-by-zero errors, set the minimum value of x at 0.1 rather
than at 0. Use your grapher's trace feature to confirm that the Figure 6-20
value for g(5) you found in Problem 2 is correct.
5. Show that the following are true for the values you calculated in Problem 2.
a. g(6) = g (3) + g(2 )
b. g(4) = g(8) - g(2)
C. g(8) = 3 g(2)
d. g( ) = - g(2)
252 Chapter
6: TheCalculus
of Exponential
andLogarithmic
Functions
g(x).
6. Recall the properties of logarithms.
logbl = 0
logb (xy) = log bx + log by
logb ( f) = logbx - logb y
log b (xY) = y log bx
logb( ~) = - log bx Figure 6-2b
Show that the values given in Problems 2 and 5 are examples of
these properties.
7. Plot on the same screen you used in Problem 4 the graph of y = log x (that is, the
base -10 log). Sketch the two graphs. Describe the similarities and the differences
between them.
8. Plot on the same screen you used in Problem 4 the graph of the natural logarithm
function. This function will appear on your grapher as ln x . The l stands for logarit hm,
and the n stands for natural . Describe the results . Surprising?!
9. Conjecture: g(x) = ln x . Test this conjecture by finding g ' (3) numerically. How does
your result confirm the conjecture?
10. By the definition of indefinite integral, g(x ) = f f(x) dx if and only if g' (x) = f(x) . Which
part of the definition, "if" or "only if," explains why
J ~ dx = In X + C?
11. If g(x), as defined in Problem 1, is really a logarithm function, then it has a base, b,
such that g(b) = 1. Solve numerically to find b. Where have you seen this number
before?
12. In the population problem of Problem Set 6-1, you evaluated
IN _!_dP.
p
1000
Use the result of Problem 10 in this problem set to evaluate the integral by using the
fundamental theorem of calculus . Show that the answers you find for N = 1500 and
N = 2000 agree with the answers you found in Problem 4 of Problem Set 6-1.
6-3 Natural Logarithms, and Another Form of the
Fundamental Theorem
The function g from Problem Set 6-2 is called the natural logarithm funct ion . As
you have shown, this function has the properties of logarithms. It arose "naturally"
out of a population problem. The abbreviation for this function is ln, pronounced
"el en." The letter l stands for the word logarithm, and the letter n stands for th e
word natural.
Secti
on6-3: Natural
Logarithms,
andAnother
FormoftheFundamen
tal Theorem 25 3
exp I
t
'fl _
y = 1/ t
Definition:TheNaturalLogarithm
Function
ln x = Jx,!_dt
1 t '
where x is a positive numbe r.
Area = lnx
The geometric meaning of the natural logarithm is illustrated in Figure 6-3a. The
natural logarithm of x equals the area of the region under the graph of y = l / t
X
from t = l to t = x. In this section you will learn to do calculus with the ln function .
Specifically, you will show that the derivative of ln xis 1/ x. Along the way you will
Fig ure 6-30 learn another form of the fundamental theorem of calculus. This new form will
enable you to find the derivative of a function defined as a definite integral.
OBJECTIVE Given a fun ction involving ln, find its derivative function; and
given an int egral of the form f (l / u ) du, do the int egra tion.
Background:
Functions
Defined
asDefinite
Integrals
Suppose a definite integral has a variable for its upper limit of integration, for
instance
r"sin t dt .
Ji
fu'
1 t= l l =X ~
Evaluating the integral gives
- cost I:= -cos x + cos 1
= - COS X + 0.5403 ....
Figure 6-3b
The answer is an expression involving the upper limit of integration, x . Thus the
integral is a function of x. Figure 6-3b shows what is happening. The integral equals
the area of the region under the graph of y = sin t from t = l to t = x. Clearly (as
mathematicians like to say!) the area is a function of the value you pick for x. Let
g (x) stand for this function.
The interesting thing is what results when you find the derivative of g.
g(x) = J:'sintdt = -cosx+0 .5403 . ..
.. g' (x) = sinx
The answer is the integrand, evaluated at the upper limit of integration, x . Here's
why this result happens. Because g(x) equals the area of the region, g ' (x ) is the
rate of change of this area. Its value, sinx, is equal to the altitude of the region at
the boundary where the change is taking place. Figure 6-3c shows this situation for
three values of x .
254 Chapter
6:TheCalculus
of Exponential
andLogarithmic
Functions
f (t) = sint f(t) = sint f(t) = sint
sinx is sma ll
sinx is negativ e.
Area is incr eas ing s lowly.
"Area" is decreas ing.
Figure 6-3c
The taller the region is, the faster its area increases as x changes . Think of painting
horizontally with a brush whose width is the altitude of the region at x. The wider
the paint brush, the faster the region gets painted for each inch the brush moves .
If sinx is negative, the "area" decreases because the integrand is negative. This
property is called the second form of the fundamental theorem of calculus, and it
is stated below.
Property: Fundamental
Theoremol Calculus
:
Derivativeol an IntegralForm
If g(x) = J;f(t) dt, where a is a constant, then g'(x) = f(x).
This form of the fundamental theorem can be proved algebraically, using as a lemma
the form of the theorem you already know.
Proof:
Let g(x) = f (t) dt. I:
Leth be an antiderivative off . That is, let h(x) = f f(x) dx .
.. g(x) = h(x) - h (a ) Fundamenta l th eorem (first form).
:. g '( x) = h '( x) - 0 Derivative of a constant is zero .
:. g'(x) = f(x), Q.E.D. Definition of ind efinite int egra l.
The derivative of the natural logarithm function follows directly from the
second form of the fundamental theorem.
x 1
In x = - dt
f
I
(where xis a positive numb er)
t
1n ' X = -1
x'
Property:Derivativeol Inx
d 1
- (lnx) = -
dx X
To find derivatives involving ln, all you have to remember is that the derivative of
In x is 1 /x. You must also observe such rules as the chain rule and the product,
power, and quotient rules.
Section
6-3: Natural
Logarithms,
andAnother
FormoftheFundamental
Theorem 255
Qfil
J; /xp
,
' I
....
In
Example 1 If y = ln (7x 5 ), find dy / dx.
Solution y= ln (7x 5 )
dy = ~(35x.4) Find the derivativ e of In (argument), then use the chain rule.
dx 7x ':)
5
X
Simplify.
In Section 6-4, you will find out why the answer above is so simple.
Example 2 If y = csc (ln x.), find dy / dx.
Solution y = csc (ln x)
ddy = - csc (ln x) cot (ln x) .!_
X X
Observ e that Example 2 is a straight application of the derivati ve of cosecant, a
method you learned earlier, followed by the application of the chain rule on the
inside function, ln x.
Example 3 If f(x ) = x.3 ln x, find an equation for f'(x) . Then show graphically and numerically
that your answer is correct.
Solution f(x) = x 3 ln x
2
f'(x) =3 x2 lnx +x 3 .!_=x (3 lnx + l). Equation for derivative.
I
X
I
y I
I
I
I Graphical Check: First plot the following.
y2 and y3 /
I
I
I
X
Y1 = x 3 ln x
',,
.........
_,,., 1 Y2 = numerical derivati ve of y 1
X = 0.5 y= -0.2698604 Y3 = x.2 ( 3 ln X + 1)
Only two graphs appear on the screen, as shown in Figure 6-3d. Tracing the y 2
and y 3 graphs shows that their values are almost identical.
Figure 6-3d
Numerical Check: Use your grapher's table feature to generate values of these
three functions.
X YI Y2 Y3
0.5 - 0.0866 . .. - 0.2698 ... - 0.2698 . ..
1.0 0 1 1.0000 . ..
1.5 1.3684 ... 4.986 9 ... 4.9869 ...
2.0 5.5451 ... 12.3177 ... 12.3177 ...
2.5 14.3170 ... 23.4304 ... 23.4304 ...
The numerical and algebrai c derivativ es give essentially the same values.
Integral
of theReciprocal
Function
Because ln'x = 1/x, the integral J(l /x) dx equa ls ln x + C. But this works only for
positive values of the variable. The function ln x is undefined for negative values of
256 Chapter
6: TheCalcu
lus of Exponential
andLogarithmic
Functions
x becaus e finding ln x would require integrating across a discontinuity in the graph
(Figure 6-3e). However, it is possible to find J(l / x) dx if xis a negative number.
y= 1/ t
t =x
In of a negative
number is
und efined.
Figure 6-3e
Let x = -u. If xis negative, then u is a positive number. The differential, dx, is
dx = d(-u) = - du.
f .!_dx = f J_(
X -U
- du) Substitute for x and dx.
=ft du. By algebra.
The integral ha s the sa me form as f (1/ x) dx, and u is a
= ln u + C. positive number.
= ln (-x) + C. Revers e substitution .
Putting this result together with the original result gives
.!_d , _ {ln x + C if x is positive
fx x - ln (-x) + C if xis negative
The two answers can be combin ed with the aid of the abso lute value function, as
follows.
Integralol the Reciprocal
Function
f ;du = lnlul + C
The variable u, rather than x, has been used above to indicate that the integral of
the reciprocal of a function, u, can be found in this way as long as the rest of the
integrand is du, the differential of the denominator.
sec 2 5x
Example 4 Integrate: f ---
tan Sx
dx
Solution sec 2 5x dx
f tan5x
Write the fraction as a reciprocal. Multi ply the other
= .!_ f - 1
-
5 tan Sx
5 sec 5xdx 2 factor by 5 to make it equal to the diff erential of the
denominator.
=
1
5 ln ltan5x l+C Int egrate the recipro cal function.
Section
6-3:Natural
Logarithms,
andAnother
FormoftheFundamental
Theorem 257
exp
t ;.21
J~
l1f'1 I
Example 5 shows a case in which the absolute value comes into action. The argument
is oft en negative when the integral has negative limits of int egration .
Example 5 Use the fundamental th eorem to evaluate exactly . Check numer ically.
-s x 2 dx
J -4 1+ x3
-5 x 2 dx
Solution
J -4 1 +x3
= -l -- lJ-5 3x 2 dx
Make the num erator equal
to the differ enti al of th e
3 -4 1 + x3 denominator.
= ln 11 + x I 1--s~
3 Integrate the recipro cal
function.
= .!.
3
ln I - 124 1- .!.
3
ln I - 6 31 = .!.
3
ln 124 - .!.
3
ln 6 3 Exact answer .
Decimal valu e of exact
= 0.225715613 ... answer.
Using your grapher's num erical integration feature gives 0.225715613 . . . , or an
answer very close to this, which checks! .
Note that the answer is a positive number. Figure 6-3f reminds you that a definite
integral is a limit of a Riemann sum. The terms have the form f (x) dx. The integrand
function is negative, and so are th e dx's because -5 is less than -4. If both f(x) and
dx ar e negative, th e terms in the Riemann sum are positive .
f(x)
!J.x is negative .
--- -------- f(c) is negative.
Figure 6-31
Example 6 shows how to differentiate an integra l when the upper limit of integration
is a function of x rather than x itself.
Example 6 Find f '( x ) : f (x) = Jfx sect dt.
Solution Let g(x) = Jsecxdx .
Then g'(x) = secx. Definition of indefinite integral (ant iderivative) .
.. f (x ) = g (5x) - g(2) Fundamental theorem , g(b) - g(a) form .
f'(x) = g '(5 x) 5 - 0 Chain rul e and derivative of a constant.
= 5 sec 5x
258 Chapter
6: TheCalcu
lusof Exponen
tial andLogarithmic
Functions
Once you see the pattern that appears in Example 6, you can write down the answer
quickly, in one step . The sec 5x is the integrand, evaluated at the upper limit of
integration. This is the result you would expect from the fundamental theorem in
its derivative of the integral form . The 5 is the derivative of 5x, which is the inside
function.
Problem Set 6-3
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes .
Ql. Integrate: f x- 0-3 dx
Q2. Integrate: It x 2 dx
Q3. Differentiate: f(x) = cos 2 x
Q4. lim x- 100COSX = COS 100, SO COS is - ?- at X = 100 .
QS. If f(x) = x 3, then f'(2) = 12. Thus f is-? - at x = 2.
Q6. Find y ' : y = sin - 1x.
Ql. Find y ' : y = csc x.
QB. "if(x)tix is called a(n) - ?- .
Q9. f f(x) dx is called a(n) - ?- .
Q10. log3 + log4 = log - ?- .
0. Look A head Problem: Look at th e derivatives and the integrals in Problem Set 6-9.
Make a list, by problem number, of those you presently know how to do.
For Problems 1-26, find the derivative .
l. y = ln 7x 2. y = ln 4x
3. f(x ) = ln x 5 4. f (x ) = ln x 3
5. h(x) = 61n x - 2 6. g (x) = 13 ln x- 5
7. r(t) = ln 3t + ln 4t + ln St 8. v (z) = ln 6z + ln 7z + ln 8z
9. y = (ln 6x) (ln 4x) 10. z = (ln 2x) (ln 9x)
ln llx ln 9x
11 . Y = ln 3x 12. Y = In 6x
13. p = (sin x) (ln x) 14. m = (cos x) (ln x)
15. y = cos (ln x) 16. y = sin (ln x)
17. y = ln (cos x) (Surprise ?!) 18. y = ln (sinx) (Surprise?!!)
19. T(x) = tan (ln x) 20. S (x) = sec (ln x)
21. y = (3x + s) - 1 22. Y = (x3 - 2) - 1
Sect
ion 6-3:Natu
ral Logarithms, andAnotherForm of theFundamental
Theorem 2 59
ffe
I
n
23. y = x 4 ln 3x 24. y = x 7 ln 5x
25. y = ln (1 / x) 26. y = ln (1/ x) 4
For Problems 27-46, integrate.
27. J 7 / x dx 28. J 5/ x dx
29. J ..!..dx
3x 30. J J_
Bx dx
x2
31. J x3 + 5 dx 32. I --x x 5 4 dx
6 -
5 3
33. J x dx6 34. J x dx 4
9- x 10 - x
J sec xtan 2
36. J sec xdx
x dx
35.
1 + secx 1 + tanx
37. J cosxdx 38. J sinx dx
smx cosx
39. r
0.5
(1 / w)dw 40 . ro
01(1/ v) dv
_
41. C\ (1 / x) dx 42. C\ o ; x) dx
43.
r
4
xl /2 dx
1 + x3/2
44 . J8
x- 113 dx
1 2 + xZ/3
45. I (ln x) 5 -dx
X
(Be clever!) 46 . J ln x X
dx (Be very clever!)
For Problems 47-54, find the derivative .
47. f(x) = J;cos 3t dt 48. f(x) = r (t 2 + lOt - 17) dt
49 . :x (J;tan 3tdt) 50. :x (.C2 dt) 1
51. f(x ) = f 3t dt 2
52. g(x) = l cosx
0
./f dt
53. h (x) = Jt-s.Jf+t2 dt 54. p(x) = Lx3(t 4
+ 1) 1 dt
55. Evaluate J; (5 / x) dx by using the fundamenta l theorem in its g(b ) - g(a) form. Then
verify your answer numerically. Ind icate which numerical method you used.
56. Look Ahead Prob lem Foll ow Up: In Problem O you were asked to look at Prob lem Set
6-9 and ind icate which problems you knew how to do. Go back and make another list
of the problems in Problem Set 6-9 that you know how to do now but that you didn't
know how to do before you worked on Problems 1-54 in this problem set.
57. Population Prob lem: In the popu lation problem of Problem Set 6-1, you evaluated
l_ dP
IN
1000 p '
where P stood for population as a function of time, t. Use what you have learned in
this section to evaluate this integral by using the fundamental theorem of calculus,
260 Chapter6:TheCalculus
of Exponential
andLogarith
mic Function
s
getting an answer in terms of N. Use the result to solve numerically for N, the number
of people when t = 10 yr, in the equation - .. You exert
N l rl O fp ound s
I1000
p dP = Jo 0.05 dt.
58. Tire Pump Work Problem: Figure 6-3g shows a bicycle
tire pump. To compress the inside air, you exert a force Outside air exerts
of F lb on the movable piston by pushing the pump 30 pound s
handle. The outside air exerts a force of 30 lb, so the
total force on the piston is F + 30. By Boyle's law, this
total force varies inversely with h, the distance between
the top of the pump base and the bottom of th e movable
piston. Consequently, the general equation is
k
F + 30 = h'
Figure 6-39
where k is a constant of proportionality.
a. Assume that the inside air is not compressed when h = 20 in., so that F = 0 when
h = 20. Find the proportionality constant, k, and write the particular equation
expressing F as a function of h .
b. Draw a sketch that shows the region under the graph of F between h = 10 and
h = 20.
c. The amount of work done in compressing the air is defined to be the product of
the force exerted on the piston and the distance the piston moves. Explain why this
work can be found by using a definite integral.
d. Calculate the work done by compressing the air from h = 20 to h = 10. What is
the mathematical reason why your answer is negative ? (If you study physics or
engineering, you will learn a scientific reason why th e answer is negative.)
e. The units of work in this problem are inch-pounds (in-lb). Why is this name
appropriate?
59. Radio Dial Derivative Problem: Figure 6-3h shows a typical AMradio dial. As you can
see, the distances between numbers decrease as the frequency increases. If you study
AM Radio x10 kHz
1-------- 10cm
------- -i
Figure 6-3h
the theory behind the tuning of radios, you will learn that the distance from the left
end of the dial to a particular frequency varies logarithmically with the frequency.
That is,
d(f)=a+blnf ,
Section6-3:Natu
ralLogari
thms,and Another
FormoftheFundamental
Theorem 261
exp
/ In
,' l
where d(f) is the number of centimeters from the number 53 to the frequency
number f on the dial, and where a and b stand for constants.
a. Figure out the constants a and b to find the particular equation for this logarithmic
function.
b. Use the equation you found in 59a to make a table of values of d(f) for each of the
frequencies shown in Figure 6-3h . Then measure the distances with a ruler to the
nearest 0.1 cm. If your calculated and measured answers do not agree, go back and
fix your errors.
c. Write an equation for d '(f). Put a new column in the tab le you made in 59b that
shows the instantaneous rates of change of distance with respect to frequency.
d. The numbers on the dial in Figure 6-3h are given in tens of kilohertz . (One hertz
equal one cycle per second .) What are the units of d'(f)?
e. Do the values of d '(f) increase or decrease as f increases? Explain how this fact is
consistent with the way the numbers are spaced on the dial.
''60. Propert ies ofln Prob lem: In this problem you will explore some properties of ln that
you will prove in Section 6-4.
a. Evaluate In 2, In 3, and ln 6 by calculator. How are your results related to one
another?
b. Make a conjecture : ln (ab ) =-?- . Test your conjecture with two other values of a
and b.
c. Evaluate ln (10 / 2), ln 10, and ln 2. What do you observe?
d. Make a conjecture: ln (a / b) = -?-. Test your conjecture with two other values of a
and b.
e. Evaluate ln (2JO) and ln 2. How are the resu lts related to each other?
f. Make a conjecture: ln (ab) = -?-. Test your conjecture on two other values of
a and b .
g. Find ln 5 and log 5, where log is the base-10 logarithm . Assuming that In x = k log x,
find the value of k .
h . Test the assumption you made in 60g by using a different value of x. In the equation
ln k = klogx, does k seem to be a constant or is it definitely not a constant?
61. Journal Problem: Update your journal with what you've learned since the last entry .
Include such things as those listed here.
The one most important thing you have learned since your last journal entry .
The difference between the graph of y = ln x and the graph of y = 1 / t, from which
ln x is defined .
The second form of the fundamental theorem of calcu lus, as the derivative of a
definite integral. You might give an example such as g(x) = f( sin t dt, where you
actually do the integrating, then show that g'(x ) = sinx.
The algebraic proof of the fundamental theorem in its second form.
The geometric interpretation of the fundamental theorem, second form, as the rate
at which the area of a region changes.
Evidence (numerical, graphical, and algebraic) you have encountered so far to indi -
cate that ln really is a logarithm.
*This problem prepares you for Section 6-4.
262 Chapter
6: TheCalcu
lusof Exponent
ial andLogarithmic
Functions
6-4 In x Really Is a Logarithmic Function
In Section 6-3, you defined th e function
x l
ln X = I -t dt '
l
where xis a positive number, and called it the natural logarithm of x. In this section
you will see why you are justified in calling this function a logarithm.
y = lnx
4 -I
3 3
2
X X
5678910
-1 -1
-2 -2
-3 -3
-4 -4
y= lnx
Figure 6-4a
Figure 6-4a shows the graphs of y = 1n x and y = log 2x. The graphs look
remarkably similar. In each, 1 is the x-intercept and the y-axis is an asymp-
tote. Each is increasing and concave downward . It is the similarity that suggests
ln might actually be a log. If you can show that ln also has the properties of logs,
you will be justified in calling it a log.
OBJECTIVE Prove that In has the prop er ties of logarithm s.
Background
ItemI: Properties
of Logarithms
Logarithm is just a fancy name for exponent. The name was picked some 300
years ago because converting numbers to powers of 10 gave a logical way to do
arithmetic before calculators were invented . For instance, a product of several fac-
tors could be computed quickly by adding their logs (that is, exponents) columnwise
in one step instead of by multiplying each pair of factors one at a time.
The three major properties of logarithms come from the corresponding properties
of exponents .
n 6-4:In x Real
Sectio ly Is a Logarithmic Function 26 3
Propertiesof Exponents Propertiesof Logarithms
1. Product of powers 1. Log of a product
b e . bd = b c+d l ogb (cd ) = logbc + logbd
2. Quotient of powers 2. Log of a quotient
b e / bd = b e- d logb (c I d) = logbc - logbd
3. Power of a power 3. Log of a power
(b e )d = b ed logb (er) = r l ogbc
For the log prope rt ies, b > 0, b * l, c > 0, d > 0, and r is any real number.
The properties are proved by transforming logarithms to exponential form using
the following definition.
AlgebraicDefinitionof Logarithm
0
a = logbc if and only if b = c
Words: A logarithm is an exponent.
Here is the thought process in transforming y = log2x to exponential form .
log 2x is pronounced "log to th e base 2 of x," so 2 is the base . Write down 2.
A logarithm is an exponent. Because y equals a log, y must be the expon ent .
Write 2Y.
The only other number in y = log 2x is x, so x must be the "answer ." Write 2Y = x .
Background
ItemII:TheUniqueness
Theorem
forDerivatives
In Problem 36 of Problem Set 5-6, you may y Th e fun ction s
have seen that if f'(x) = g ' (x ) for all x, then diff er by a
f(x) and g (x) differ at most by a constant. constant.
Figure 6-4b shows two such functions. If f(x)
and g(x) also have a point in common, then
they are actually just one ("unique") function.
This property is called the uniqueness
theorem for deriva tives. Intuition tells
you that if functions start at the same
point and change the same way, the y are X
the same function. The property can be
proved algebraically by using th e mean value
theorem as a lemma .
Figure 6-4b
26 4 Chapter
6: TheCalculus
of Exponential
andLogarithmic
Functions
Property: TheUniquenessTheoremfor Derivatives
If: 1. f '(x) = g '(x) for all values of x in the domain, and
2. f(a) = g(a) for one value, x = a, in the domain, then f(x) = g(x) for all values of x
in the domain.
Words : If two functions have the same derivative everywhere and they also have a point in
common, then they are the same function.
Proof (by contradiction): Assume that the conclusion is false. Then there is a number
x = b in the domain for which f(b) * g(b) (Figure 6-4c, left side). Let h be the
difference function, h (x) = f(x) - g(x ) (Figure 6-4c, right side) .
f {x) or g(x) h (x) = f(x) -g(x)
f(b) - - - - - - - - - - - - -f
'
g(~ - - -----------
f(a) = g(a) - - - -,
h (a) = O
a a C b
Figure 6-4c
Because f(a) = g(a) and f (b) * g(b), it follows that h (a) = 0 and h (b) * 0.
Thus th e secant line through th e points (a, h (a)) and (b, h (b )) will have a slope not
equal to zero.
Because both f and g are given to be differ entiable for all x in the domain, h is also
differentiabl e, and thus the mean value th eorem applies to function h on [a, b ].
Therefore there is a numb er x = c in (a, b) such that h '( c) equals the slop e of the
secant line. Thus h '(c) * 0.
But h' (x) = f'(x) - g ' (x) . Because f'(x) is given to b e equal to g ' (x) for all x in the
domain, h '( x) = 0 for all x. Thus h '(c) does equal zero.
This is a contradi ction. Therefore the assumption is false, and f (x) does equal g(x)
for all x in the domain, Q.E.D.
Section6-4: Inx ReallyIsa Logarithmic
Function 265
Logarithm
Properties
of In
Four properties of logarithms for natural logs are given in the box below .
Property:LogarithmPropertiesof In
(a and b are positive, and r is any real number.)
Product: ln (ab) = ln a+ ln b
Quotient: ln (a/b) = ln a - In b
Power: ln(a') = rln a
Intercept : ln 1 = 0 because ff (1/t) dt = 0
The uniqueness theorem can be used to show that ln actually does have these
properties. The proof for a product is shown below. You will prove the other
properties in Problem Set 6-4.
Proofof theIn of a Product
Property
Prove that ln (ab) = ln a + ln b for all a > 0 and b > 0.
Proof: Let b stand for a positive constant, and replace a with the variable x.
Let f(x) = ln (xb), and let g(x) = ln x + In b.
1 1 1 1
Then f' (x ) = - b b = - for all x > 0, and g' (x) = - + 0 = - for all x > 0.
X X X X
Substituting 1 for x gives
f(l)= ln (lb) = In band g(l ) = ln 1 + ln b = 0 + ln b = In b.
Thus f'(x) = g'(x) for all x > 0 and f(l) = g(l). Thus, by the uniqueness theorem
for derivatives, f(x) = g(x) for all x > 0.
That is, In (xb) = ln x + ln b for any positive number x and any positive number b.
Replacing x with a gives
ln (ab) = ln a + ln b for all a > 0 and all b > 0, Q.E.D.
Summary
Because the function In has the three major properties of a logarithm, and because
its graph looks like that of a logarithm function - including an x-intercept of 1 and
the y-axis as an asymptote-you are justified in calling ln a logarithm. In Problem
Set 6-4, you will prove the rest of these properties, and you'll find the number e that
is the base of the natural logarithm function.
266 Chapter
6:TheCalculus
ofExponentia
l andLogarithmic
Functio
ns
Problem Set 6-4
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Differentiate: y = tan - 1x y
Q2. Integrate: f (4x + 1) 5
dx
Q3. Find limx-o(sinx) /( x).
Q4. Find limx-3 7. X
QS. log 12 - log48 = log-?-
Q6. log 7 + log 5 = log-?-
Ql. 3 log 2 = log-?-
QB. Sketch the graph of f; y dx for the function
shown in Figure 6-4d.
Figure 6-4d
Q9. Write one hypothesis of the mean value theorem.
QlO. Write the other hypothesis of the mean value th eorem .
For Problems 1-6, show that the properties of ln actually work by evaluating both sides
of the given equation.
l. ln 24 = ln 6 + ln 4 2. ln 3 5 = ln 5 + ln 7
3. ln /
2
66 = ln 2001 - ln 667 4. ln 1 1
~~ = ln 1001 - ln 77
3 4
5. ln(1776 ) = 3 1n 1776 6. ln(1066 ) = 41n 1066
7. Prove the uniqu eness theorem for derivatives. Try doing so without looking at the
proof given in this text. If you get stuck, look at the text proof just long enough to
get going again.
8. Prove by counterexample that ln (a+ b ) does not equal ln a + ln b.
9. Prove that ln (a / b ) = ln a - ln b for all a > 0 and b > 0.
10. Prove that ln (ab) = b ln a for all a > 0 and for all b.
11. Prove that ln (a/b) = ln a - ln b again, using the property given in Problem 10 as a
lemma.
y
12. Write the definition of ln .
13. Base of Natural Logarithms Problem : Because you have demon-
strated that ln is a logarithm, it must have some number for its
base. The question is, what number. In this problem you will
investigate what that numb er could be. Let y = ln x = logbx,
a. Explain why x = bY.
b. Based on your answer to 13a, what would x equal if y = l? Figure 6-4e
c. Figure 6-4e shows y = ln x with a line drawn across at y = l. Use your grapher's
trace or solve feature to find, approximately, what x equals when y = l.
d. Have you seen the numb er you found in 13c before ? If so, where? What letter is
commonly used for this number 7
6-4:In x ReallyIsa Logarithmic
Section Function 267
,t~n
~ 6-5 Derivatives of Exponential Functions -
Logarithmic Differentiation
So far, the only way you have been able to find derivatives of exponential functions
such as
f(x ) = 2x
is by using numeric techniques. Now it is time to learn an algebraic technique
that will let you find the derivative exactly. This technique is called logarithmic
differenti ation, and it uses the properti es of ln to transform an equation to one
that you can differentiate implicitl y. The techniqu e also provides an algebraic way
to find derivatives of products and quotients mor e easily .
OBJECTIVE Differentiat e algebraically a function who se equation has a variable exponent. Use the
prop erties of ln to differen tiate produ cts , powers, and quotient s.
To find f'(x) for the function above, start by taking the ln of both sides of the
equation. This is th e easiest thing you could ever do-yo u just write "ln" in front of
each side!
ln f(x) = ln 2x
The expression on the right is the ln of a power, so you can write
ln f (x ) = x(ln 2) .
Now, differentiate both sides implicitly with respect to x. Observe that 1n 2 is a
constant.
f (~/ '( x ) = l(ln 2)
The f'(x) on the left side comes from the chain rule. Finally, use algebra to isolate
f' (x ).
f'(x) = f(x) (ln 2) = 2x ln 2
The answer is simply the original function multiplied by a constant, ln 2 in this case.
The same proce dure can be used if the equation has more complicated variable
exponents.
Example 1 Find f '(x) if f(x ) = (x 3 + 4)Cosx.
Solution f(x ) = (x3 + 4)cosx
ln f (x) = ln (x 3 + 4)C0 sx Take ln of both sides.
ln f(x ) = cosx[ln (x 3 + 4)] ln of a power.
-f(1 ) f'(x) = -s inx[ln (x 3 + 4)] + cosx [ 3x
34
2
J
Differentiate implicitly on the left.
X X + Derivative of a product on the right.
f'(x) = (x 3 + 4)cosx ( -sinx [ln (x 3 + 4)] + cos x [ ):
2
4
J)
268 Chapter
6:TheCalculus
of Exponential
andLogarithmic
Functions
Logarithmic
Differentiation
forProducts
andQuotients
The properties of logarithms allowed you to turn a power into a product, which you
know how to differentiate. The other properties let you transform products and
quotients to sums and differences, which are even easier to differentiate. Example
2 shows you how this is done.
5
3 7
Example 2 If f(x) = ( ~ \ ) , find f'(x).
Slil X
5
Solution ln f (x ) = ln ( 3x + 7 ) Take In of both sides.
Slil 4 X
ln f(x) = 5 ln (3x + 7) - ln (sin4x) Log of a quoti ent and a power.
1 5 1 Differentiate implicitly (use the
- - f'(x) = _ _. 3 - - - cos4x 4
f(x) 3x + 7 sin4x chain rul e).
5 1
f'(x) = f(x) (- - 3 - -.- - cos4x 4)
3x+7 sm4x
----
sin4x 3x+ 7
5
(3x+ 7) ( -- 15 -4cot4x )
Problem Set 6-5
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Differentiate : y = ln 7x 3
Q2. Integrate: J(5x) - 3 dx
Q3. Differentiate : y = cos - 1 x
Q4. Integrate: f (- x )6 dx
QS. Differentiate: y = sec x
Q6. Integrate: f - sinx dx
Ql. Differentiate: y = tan x cot x
QB. ln 12 + ln 3 = ln -?-
Q9. 3 ln 2 = ln - ?-
Q 10. 0/0 is called a(n) - ?- form.
0. Look Ahead Problem: Look at the derivatives and the integrals in Problem Set 6-9. Make
a list, by problem number, of those you presently know how to do .
For Probl ems 1-18, find an equation for the derivative of the given function.
l. f(x) = o.42x 2. f(x) = 10 - o.2x
3. g(X) = 4(7X) 4. h(X) = 1000(1.Q3X)
5. c(x) = x 5 3x 6. m(x) = 5x x 7
7. y = (cosx) 07x 8. y = (tanx) 4x
9. y = (csc 5x) 2x 10. y = (cos2x) 3x
6-5:Derivatives
Section of Exponential
Functions-Logarithmic
Oifferentiatlon 269
l ex~
' '~ J
l
11. f( t) = t secr 12. r(u) = uln u
13. V = (x 4 - l )X 14. 2 = (sin t) cscr
15. y = 2x ln x 16. y = X(3X2)
17. y = 5(3x-4)X 18. y = 8(4x - 5)X
For Problems 19-24, find an equation for the derivative function by applying the following
methods: (a) Differentiate directly, using the chain rule and so on. (b) First apply the
logarithm properties of ln, then differentiate. Show that both answers are equivalent.
19. y = ln 3x 7 20. y = ln 10x 8
21. y = ln[(3x + 4 )(2 x - 9)] 22. y=ln[(4x - 7) (x + 10 )]
6x - 5
23 y = ln 7x
5x + 2
- 8 24 . y = In x +
3 1
25 . For y = ln 3x 7 (Problem 19), show that your answer for y ' gives the right value when
x = 2 by evaluating the differ en ce quotient t:,.y / t:,.x for t:,.x = 0.001.
26. For y = ln 10x 8 (Problem 20), show that your answer for y ' gives the right value when
x = 2 by evaluating the difference quotient t:,.y / t:,.x for t:,.x = 0.001.
For Probl~ms 27-30, find an equation for the derivative function by logarithmic differen-
tiation.
27. y = (5x + 11)7 (7x - 3) 5 28. y = (4x + 3) 8 (8x - 9) 4
29. y = (3 - 4x) 5 (7 + 5x) 4 30. y = (10 + 3x) 10 (4 - 5x) 3
(4x + 1) CSC X x 7 cos x
31. y = 5
32. y = 5x + 6
sin x
9 5
33 . Integral Review Problem: Evaluate r
Js 3 -
-
X
dx by using the fundamental theorem.
34. Look Ahead Problem Follow-Up : In Problem 0, you were asked to look at Problem Set
6-9 and indicate which problems you kn ew how to do. Go back and make another list
of the problems in Problem Set 6-9 that you know how to do now but that you didn't
know how to do before you worked on Problems 1-30 in this problem set.
35. Continued Exponentiation Problem :
a. Let f (x) = xx.Find an equation for f'(x ) . Simplify as much as possible .
b. Let g(x) = xxx_Find an equation for g '( x ) . Simplify as much as possible. Observe
that the x" in the exponent is the innermost function .
36 . Derivative with Variable Bas e and Exponent Generalization Problem :
Let f(x) = x 3 , g(x ) = 3x, and h(x ) = x".
a. Differentiate function f logarithmically , as you ha ve for variable exponent
functions in this section. Show that the answ er reduces to the familiar form nxn - 1
from the derivative of the power function.
"b . Write an equation for g '( x) . In what ways is your equation similar to the equation
for f'(x). In what ways is it different?
c. Write an equation for h ' (x) . Show that your answer can be written as a sum of two
terms, one of which is similar to the derivative off (constant exponent), the other
of which is similar to the derivativ e of g (constant base) .
2 70 Chapter
6: TheCalculus
of Exponential
andLogarithmic
Functions
3 7. Compound Interest Prob lem: In a real -world situation
that is driven by internal forces, the variables are FASTER !
often related by an exponential function. For instance,
the more money there is in a savings account, the
faster the amount in the account grows (Figure 6-Sa).
If $1000 is invested with interest compounded
continuously, and the interest is enough to make
the annual percentage rate (APR)equal 6%, then the
amount of money, m(t) dollars, in the account at
time t years after it is invested is
1
m(t) = 1000(1.06)
a. Find an equation for the derivative, m'(t) . At what
rate is the amount growing at the instant t = 0 yr?
At t = 5 yr? At t = 10 yr? What are the units of
these rates? The more there is, the faster it grows.
b. Find the amount of money in the account at t = 0,
t = 5 yr, and t = 10 yr. Does the rate of increase Figure 6-50
seem to be getting larger as the amount increases?
c. Show that the rate of increase of money is directly proportional to the amount
present. One way to do so is to show that m ' (t) / m(t) is constant.
d. Show that you earn exact ly $60 the first year. Then explain why the rate of increase
at time t = 0 is less than $60/ yr.
38. Door-Closer Problem: In Section 1-1, you were introduced to a problem in which a door
was pushed open . As the automatic door-closer slowed the door down, the numb er of
degrees, d(t), the door was open after t seconds was given to be
d(t ) = 200t 2- 1 ,0 :s; t :s; 7.
Use what you have learned about derivatives of exponential functions to analyze th e
motion of the door . For instance, how fast is it opening at t = 1 and at t = 2? At what
time is it open the widest, and what is the derivative at that time? A graph might help .
,,39_ Limit of an Interesting Expression: Figure 6-Sb shows f(n) = (1 + 1/ n )".
Wh ere?
As the exponent n gets large, the base gets closer and closer to l. And
1 raised to a large power is still 1. But a number greater than 1 raised
to a large power is very large. Investigate what happens to f(n ) as n
gets very large. Go to at least n = 1, 000, 000 . Where have you seen
this number before?
X= 7 S y= 2.546499
40 . Journal Problem : Update your journal with what you've learned
since the last entry . Include such things as those listed here. Figure 6-5b
The one most important thing you have learned since your last journal entr y
The way to differentiate algebraically an exponential function
How you know that 1n really is a logarithmic function
What you now better understand about logarithms
Any technique or idea you plan to ask about at the next class period
* This problem sets the stage for Section 6-6.
6-6 The Number e, and the Derivative of Base b
Logarithm Functions
You defined ln x to be a function whose derivative is 1/ x. You then found that the
resulting function has the properties of logarithms . In this section you will attack
the problem from the other end, starting with a base b logarithm and u sing the
definition of derivative. This task, though difficult to do, will lead you to the number
that is the base of the natural logarithm .
OBJECTIVE Find out algebraic ally wha t numb er is the base of th e 1n function.
Differentiate algebraically a logarit hm fun ction with any permis sibl e numb er as its b ase .
ItemI: Limitof (1 + n)11n asn Approaches
Background Zero
The expression (1 + n) l f n assumes interesting values if n is close to zero . The base,
?,,-Limit = e = 2.718 28. _
3 (1 + n), is close to 1 and the exponent, 1/ n, is very large. As Figure 6-6a shows, the
values approach an asymptote a bit above 2.7. If you trace the graph, here are some
y = (l+l / x)' values.
X = 1/ n 1/ n (1 + n )l fn answer
5 10 1.110 = 2.593742 ...
100 1.01100 = 2.704813 ...
Figure 6-60 10,000 l.0001 I 0,000 = 2.718145 .. .
The value of (1 + n) I f n is caught between two opposing properties.
( 1)any exponent = 1
(greater than l)l arge positive exponent = large number.
The result is a stand off. The limit, named e, is 2.7182818284 . . . . With 1/ n =
10,000,000, you should get 2.718281816 .. . , which is close toe . Prob lem 43 in
Problem Set 6-5 introduced you toe . It is a naturally-occurring constant, like rr . The
letter e is used because it is the base of the natural exponential function, as you will
see in Section 6-7.
Definitionof e
e = lim(l + n ) 11n = lim(l + 1 / n)n
n-0 n- oo
e = 2.7182818284 . . . (a nonrepeating decima l)
') "7') rL __ ...__ /.TL_r_ L .. l. . __ rr ..___ __ ..: _ 1 __ _1 , __ ,,1 r
In spite of the fact that 1828 repeats once in the decimal part of e, e is a nonter-
minating, nonrepeating decimal that cannot be expressed exactly by using only the
operations of algebra. Like rr, e is a transcendental number . The proof of this fact
appears in abstract algebra courses .
Background
ItemII:Limit-Function
Interchange
forContinuous
Functions
If you take the limit of a continuous function that has another function inside, such
as
lim sin (tanx ),
x-c
it is possible to interchange the limit and the outside function,
sin (lim
x-c
tan x).
To see why continuity is sufficient for this interchange, consider a simpler case. The
definition of continuity states that if g is continuous at x = c, then
limg(x ) = g(c ).
x.-c
But c is the limit of x as x approaches c. Replacing c with the limit gives
VElg
(x)= g(VElx)'
which shows that the limit and the outside function have been interchanged.
Property:Limit-Function Interchangefor Continuous Functions
For the function f(x) = g(h(x) ), if h(x) has a limit, L, as x approaches c and if g is
cont inu ous at L, then ~Elg(
h(x))= g (FElh (x)) .
Derivative
of theBaseb Logarithm
Function
fromtheDefinition
of Derivative
Suppose that f(x) = logbx, where b is a positive constant not equal to 1. The graph
is shown in Figure 6-6b. By the definition of derivative,
'( ) = r logb (x + h) - logbx
f X ,;!_lJ h .
f (x)
log(x + h)
log x
Figure 6-6b
Section e, andthe Derivative
6-6:TheNumber ofBaseb Logarithm
Functions 273
exp
l .....
' In
* I
The following sequence of steps shows how this difference quotient can be
transformed into the form ( 1 + n) 1/ n, whose limit is e as n approaches zero .
logb (X + h)
f'(x) = lim x Log of a quoti ent prop ert y, applied in reverse.
h -0 h
Algebra , used to get (1 + something) into the
= \iEJ [1ogb ( 1 + ~) J argu ment of th e log.
= lim
h -0
[.!.
X
~log
h
b (1 + .!:1.)J
X
Multiply by x/xan d rearran ge.
Limit of a con stan t times a function. 1/ x is
= .!.
X
lim [~log b (1 + .!:1.)J
h- 0 h X ind epend ent of h.
= -1 lim [ logb ( 1 + -h)x/h] Log of a pow er property, applied in revers e.
X h -0 X
= -1 logb [ lim ( 1 + -h)x/h]
Interchan ge log and limit , asswning log is
X h -0 X continuous.
The exp ress ion in parentheses ha s the form
f'(x) = _!_ logbe (1 + n) 1' ", whose limit is e as n - 0.
X
Finding th e limit in th e last step depends on the fact that h - 0 implies h/ x - 0,
because x is restricted away from zero by the definition of logarithm .
Now, suppose you choose e as the base of the logarithm function . The derivative is
f (x) = logex => f'(x) = _!_ logee = _!_ 1= _!_ logbb = 1 for an y permi ss ible base b.
X X X
The derivativ e is simpler because it involves no logarithms. But, mor e importantly,
it is identical to the derivative of ln x, namely, 1/ x. Because ln 1 = logel = 0, the two
functions meet the hypotheses of the uniqueness theorem for derivatives. Therefore
ln and loge are th e same function!
Property:Equivalence
of NaturalLogsandBasee Logs
ln x = logex for all x > 0.
Algebraic
Differentiation
Technique
forBaseb Logarithm
Functions
It would be possible for you to memorize th e formula above for the derivative
of base b logs. However, be cause calculators include an ln key, and because the
derivative of 1n is so simple, it is bett er to first transform th e base b log into a
natural log. To do so, recall from algebra th e change-of-base property.
Property:Change-of-Base
Propertyfor Logarithms
logax .
logbx =- b m general, and
1oga
logex ln x 1 . .
logbx = log, b = ln b = In b In x m particular.
274 Chapter
6: TheCalcu
lusof Exponentia
l andLogarithm
ic Functions
In Problem 21 of Problem Set 6-6, you will prove this property. Examples 1 and 2
demonstrate how to differentiate algebraically some base b log functions.
Example 1 Find an equation for f '(x) if f(x) = log10 x. Check the formula by evaluating f'(2)
and showing that the line at the point (2,log 10 2), with slope f'(2), is tangent to the
graph.
1
Solution f(x) = log 10 x = ln ln x Use the change -of-base property.
10
f'(x) = ln\o ~ = xl: 10 Derivativ e of a constant times a function.
f (x) The line through the point (2, log 10 2), which equals (2, 0.3010 ... ), has slope
1
f'( 2 ) = 2.302~85 . . . . ""0 217
Thus the line's equation is
y = 0.30102999 y - 0.3010 "" 0.217(x - 2), or y"" 0.217x - 0.13 3.
Figure 6-6c shows the log graph and the line . The line really is tangent to the
Figure 6-6c graph, Q.E.D.
Example 2 If f(x) = log 4 3x, find f'(x) algebraically and find an approximation for f'(S). Show
that your answer is reasonable by plotting f on your grapher and by showing that
f'(S) has a sign that agrees with the slope of the graph .
Solution ln 3x 1
f(x) = log-13x = ln = ln ln 3x
4 4
f(x) f'(x) = -
1- (_!_ 3)
ln 4 3x
1
X
5 xln 4
By calculator, f '(S) = 0.144269 ....
x=S y=l.9534453
To plot the graph, enter the transformed equation involving ln , shown above.
As Figure 6-6d shows, f(x) is increasing slowly at x = 5, which agrees with the
Figure 6-6d small positive value of the derivative .
Sec e, andtheDeriva
tion 66: The Number tiveof Baseb Logar
ithmFunctions 275
[iJ
J ,' In
,' [
Problem Set 6-6
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
01. Differentiate : f(x) = cos 2 x
y
02. Differentiate: g (x ) = cos x 2
03. Differentiate : h (x ) = cos 2x .. ;
---;-.-;
1 ....... ;.
Q4. Differentiate: c (x ) = cos 2 5
X
05. Differentiate: L (x ) = ln x
06. Differentiat e: M (x ) = ln x 5
.
.
T,
l
Ql. Differentiate: N (x) = ln 5 x
.......' ......... ........ ' ..
QB. Differentiate : O(x ) = tan - 1x .' ..:
'
:'
Q9. Sketch the graph of y ' for the function shown in Figur e 6-6e.
Figure 6-6e
Q10. State the conclusion of the uniqueness theorem for derivatives.
For Problems 1-4, find an equation for the derivativ e of the given function, and show
numerically or graphically that the equation gives a reasonable value for the
derivative at the given valu e of x.
l. f (x ) = log 3x, x = 5 2. f (x ) = log 7 x, x = 3
3. f(x ) = log o5X, x=9 4. f(x ) = logos x, x =4
For Problems 5-12, find an equation for the derivativ e of the given function.
5. f (x ) = 13 loge x 6. f (x ) = 5 loge x
7. g (x ) = 8log e(x 5 ) 8. h (x ) = 10log e(x 04 )
9. T (x ) = log 5 (sinx ) 10 . R (x ) = log4(Secx )
11. p (x ) = (loge x ) (log 5 x ) 12. q (x ) = log 9 x
lOg3 X
For Problems 13 and 14, find the derivative at the given value of x, then show that your
answer is correct by finding th e numerical derivative.
13. Y = log10X, X =4 14. y = logo.2X, x = 5
For Problems 15 and 16, plot th e graph of the given function, then find the derivative at
th e given value of x = c . Plot a line through th e point (c ,f (c)) , with slope f ' (c ) . Explain
why your results are reasonable.
15. f (x ) = logo.9x, c =2 16. f (x ) = log14 x, c = 3
17. Derivat ive of Logarithm Proof: Deriv e the formula for f' (x) if f(x) = logb x, starting
with the definition of derivative. Try to do this without looking at the proof in this
text . If you need to look, do so just long enough to get moving again.
18. Equivalence of Natural Logs and Base e Logs : Use th e uniqueness th eorem for
deri vatives to prove that loge x = ln x for all x > 0.
2 76 Chapter
6: TheCalculus
of Exponential
andLogarithmic
Functions
19. Lava Flow Problem: Velocities are measured in miles per hour. When a velocity is
low, people sometimes prefer to think of how many hours it takes to go a mile. Lava
flowing down the side of a volcano flows more slowly as it cools. Assume that the
distance, y miles, from the crater to the tip of the flowing lava is given by
y = 7 (2 - 0.9x),
where x is the number of hours since the lava started flowing (Figure 6-6f).
Figure 6-6f Geologist collecting data about lava flow
a. Find an equation for dy / dx. Use the equation to find out how fast the lava tip is
moving when x = 0, 1, 5, and 10 hr. Is the lava speeding up or slowing down as
time passes?
b. Transform the equation y = 7 (2 - 0.9x) (from 19a) so that xis in terms of y .
Take the log of both sides, using some appropriate base for the logs.
c. Differentiate the equation given in 19b with respect to y to find an equa tion for
dx/dy. Calculate dx /dy when y = 10 mi. What are the units of dx /dy?
d. Calculate dx /dy for the value of y when x = 10 hr.
e. Naive reasoning suggests that dy/dx and dx/dy are reciprocals of each other .
Based on your answers above, in what way is this reasoning true and in what way
is it not true?
20. Compound Int erest Problem: If interest on a savings account is compounded contin-
uousl y, and the interest is enough to make the annual percentage rate APR equal
6%, then the amount of money, M, after t years is given by the exponential function
M = 1000 x l.06r. This equation can be solved fort in terms of M.
t = log1 0G(i:0 0
)
a. Show how the transformations are done to get tin terms of M.
b. Write an equation for dt/dM.
c. Evaluate dt /dM when M = 1000. What are the units of dt /dM7 What real-world
quantity does dt / dM represent?
d. Does dt/dM increase or decrease as M increases? How do you interpret the
real-world meaning of your answer?
Sec e, andtheDerivative
tion6-6:TheNumber of Baseb Logarithm
Functions 277
21. Proof of the Change-of-Base Property: The following is a proof of the change-of-
base property for logarithms . For each step in the proof, write a reason to justify that
step.
log b X
Prove that log ax = - - - .
1ogb a
Proof:
Let y = log ax.
: . aY = x a. _ __ ___________ _
: . log b (aY) = log bx b. ___ __ ____ _____ _
: . ylog b a = log b x c. ______________ _
. y = --
logb X d. _____ _______ _
log b a
. logb X
.. logax = - -- , Q.E.D. e. _ __________ _ __ _
1ogb a
22. The Two Forms of the Definition of e: Let f(n) = (1 + n ) 11n and let g(n) = (1 + 1/ n)n
as in the definition of e. Show that the limit of g(n ) as n approaches infinity is
equal to the limit of f (n ) as n approaches zero and thus that the two forms of the
definition are equivalent.
23. Definition of e Journal Problem: Write in your journal what you understand about the
number e. Include such things as those listed here .
The definition given in this text, and why e = limx-oo( 1 + 1 / x )" is equivalent to this
definition.
The graph of Y1 = (1 + 1/ x)", using windows of [O, 10] and [O, 100,000] for x and
[O, 3] for y. Include a discussion about how you can tell that, even though
the graph looks like a horizontal straight line for th e larger window, it is actually
increasing slightly. Explain why your grapher gives no value of y if x is zero.
What is meant by th e fact that e is a transcendental number.
24. Population Problem Revisited: In Section 6-1, and at several intervening times, you
have worked with the population problem in which
N 1 r10
I
1000
p dP = Jo 0.05 dt,
where P = N is the number of people at time t = 10 yr. Integrating gives
In IN I - ln 1000 = 0.5 .
Use the fact that In x = Iogex to solve this equation explicitl y for N. Show that th e
answer is approximately 1649 people, which you found numerically in Problem 57 of
Problem Set 6-3.
25. Limit and Function Int erchange Journal Problem: Write an entry in your journal about
the property that lets you interchange the function name and the limit sign. Include
the following .
What the property says
How the definition of continuity makes the property work
An example applied to a function you know is continuous
How the property is used in finding the derivati ve of base b logarithms algebraically
278 Chapter
6: TheCalculus
of Exponential
andLogarithmic
Functions
6-7 The Natural Exponential Function:
The Inverse of In
In Section 6-6, you learned that the natural logarithm function, ln, has the number e
as its base. From Section 4-5, recall that the inverse of a function is the relation you
get when you interchange the two variables. Thus, the inverse of y = ln x = loge x is
X = loge y.
By the algebraic definition of logarithm (Section 6-4),
This function is called the natural exponential function because the constant e
arose "naturally" in differentiating logarithmic functions. In this section you will
learn to differentiate and integrate quickly the natural exponential function.
OBJECTIVE Given a function whose equation involves a variable power of e, find equations for its
derivative and its integral functions.
ln Section 6-5, you learned how to use logarithmic differentiation to find the
equation for the derivative of a function with a variable exponent. The derivative of
f(x) = ex is as follows .
f(x) = ex
ln f(x) = ln ex = x In e = x I = x =;, In f(x) = x Tak e ln of both side s and simplify.
Differ entiat e implicitly. Observe
. f(\) f'(x) = 1 the chain rul e.
f'(x) = f(x) Algebra.
f'(x) = ex Substitute f (x) = e".
The natural exponential function has a remarkable property: It is its own derivative!!
Figure 6-7a shows that the slope of the graph at any point is equal to they-value at
that point.
y=ex
y = e" j
y = 7.389 .. . ----- , Slope =
l 7.389 ...
Area =
ex+ C
y = 2.718 ...
X X
1 2 3 1 X
Figure 6-7a Figure 6-7b
Sec
tion6-7: The Natural Exponential Func
tion: TheInverse
of In 279
ii
Because ex is the derivative of eX, you can find an equation for the integral of ex dx.
I ex dx =e x+ C
The natural exponential function is practically indestructible! Differentiate it or
integrate it- the answer comes out the same! Figure 6-7b shows that the area under
the graph from a constant lower limit to the variable upper limit x is
Area = ex + C.
The number e, although it is a "messy decimal," is the base of choice in calculus for
logarithmic and exponential functions because the derivatives and the integrals are
so simple . The calculus of the natural exponential function is summarized below.
Properties:Calculusof the NaturalExponentialFunction
Derivative: If f(x) = eX, then f'(x) = ex.
Integral : f ex dx = ex + C
Example 1 If f(x) = 7e 5x , find an equation for f'(x ). Use the equation to find the value of f'(0 .3).
Use numerical differentiation to show that your answer is correct.
Solution f(x) = 7esx
The Sx in th e expon ent is th e inside function for
f'(x) = (7)(e 5x)(5) th e expon ential fun ction .
= 35e 5x
f'(0 .3) = 35e sco.3> = 156.8591 . . .
Finding the numerical derivative of 7e 5x at x = 0.3 gives an answer close to
156.8591 . . .. Your grapher may show something like 156.8597 .. . , depending
on the tolerance to which it is set.
Example 2 Find an equation for the indefinite integral Je0-2x dx.
Solution The key is recognizing that in f ex dx, the variable could be any letter.
J e 1 dt = e1 + C, J eu du = eu + C, etc.
The only thing that is important is for the d . .. to be the differential of the inside
function, namely, the exponent. In the given integral, the differential of 0.2x is
0.2 dx. The dx is already there, but not the 0.2, so the first step is to multiply
and divide by 0.2.
I eD.2xdx
= o\I e D.2x(0.2 dx) Make th e dx th e diff erential of th e insid e fun ction .
= -e1 o.2x + c f e" du = e" + C
0.2
= 5eo.2x+ C
280 Chapter
6:TheCalcu
lusof Exponential
andLogarithmic
Functions
The symbol exp (x), or just exp x, is often used in place of ex. Using this symbol has
several advantages.
It uses the more familiar f(x) terminology.
It helps you realize that the exponent of an exponential is really an inside function .
It makes exponentials easier to read if the exponent is a complicated expression,
such as
e sin x /( 3x+7) = exp ( sinx ) .
3x +7
It gives the function a name, exp, which can be used as ln, cos, and others are
used .
It appears in some computer languages.
Inverse
Properties
ofInandexp
The fact that ln and exp are inverses of each other can sometimes be used to simplify
an expression before you differentiate or integrate. For instance,
ln ex = x In e = x 1 = X.
This result is an example of the more general relationship between a function and
its inverse, which you saw in Section 4-5. Specifically if f is an invertible function,
then
f(f - 1(X) ) = X and f - 1 (f (X)) = X.
Property:InverseRelationship
BetweenInandexp Functions
exp(ln x) = x and ln(exp x) = x, or
eln x = x and 1n eX= X
Example 3 Differentiate: y = cos (ln e7x )
Solution y = cos (ln e 7x) => y = cos 7x Function of an inver se fun ction prop erty .
.. y ' = - 7sin7x
Section
6-7:TheNatural
Exponential
Function:
TheInverse
of In 28 1
p
I.....
I
In
Problem Set 6-7
DoTheseQuickly
The following problems are int ended to refresh your skills . You should be able to do all
ten probl ems in less than five minut es .
Ql. Differentiat e: y = In x
Q2. Differentiat e: y = logex y
i........ # , ..
Q3. Evaluat e: ln 1
.. 1
Q4. Evaluat e: log3l . X
QS. Evaluate: ln e5
Q6. Int egrat e: f sec 2x dx
---~-~
---,. ..
Ql. (d / dx )( secx ) = -?-
QB. (d/dx)(sec 1
x) = - ?- Figure 6-7c
Q9. Sketch th e graph of the derivativ e for the function shown in Figure 6-7c.
QlO. True or false: Continui ty implies differ entiability.
0. Look Ahead Problem : Look at th e derivatives and integrals in Problem Set 6-9. Make a
list, by problem number, of thos e you pres entl y know ho w to do.
For Problems 1- 36, find an equation for th e derivati ve function. Simplify your answer.
1. y = e4x 2. y = e9x
3. Y = 17e- SX 4. y = 667e - 3x
5. f (x) = e- x 6. f (x) = 3e - x
7. h (x) = x 3eX 8. g(x) = x - 6ex
9. r(t ) = e 1 sint 10. s(t) = e 1 tan t
11. u = 3exe - x 12. V = e- 4xe4x
13. y = e 2uln 3u 14. y = e- Su ln 4U
expx In x
15. y = ln X 16. y = expx
17 . y = 4e secx 18. y = 7ecosx
19. f (x) = csc ex 20. f (x) = cot ex
21. y = 3 ln e2X 22. y = 4 ln eSx
23 . Y = (ln e3x) (ln e4x) 24. y = (ln e- 2x) (ln eSx)
25. g(x) = 4eln 3x 26. h (x) = 6eln 7x
27. y = 2001(e 3x)S 28. Y = 1001 (e 4X)l
29. y = eX + e-X 30. y = eX - e- X
31. u = (5 + e2r)7 32. V =( 3+e - 1 ) 5
282 Chapter
6: TheCalculus
of Exponential
andLogarithmic
Functions
33. y = exp(5x 3) 34. y = 8 exp(x 5 )
35. y = (sin3)(ln 5)(e 2 ) 36. y = (tan4)(e 6 )(ln 2)
For Problems 3 7 and 38, find f'(x) algebraically for the given value of x, then confirm your
answer numerically.
37. f(x) = e0.4x, x = 2 38. f(x) = e- 2x, x = 0.6
For Problems 39 and 40, find f '(x) algebraically for the given value of x. Then confirm
your answer by graphing a line through the given point with the appropriate slope.
39. f(x ) = 5xex, X =-l 40 . f(x) = 6x 2 e-x, X = 2
For Problems 41-54, evaluate the indefinite integral.
41. Jesxdx 42. J e7x dx
43 . J 6expxdx 44. Jexp(0.2x) dx
45 . J 3e- 2x dx 46. J - 4e- Gxdx
47. I esinxcosX dx 48. J etanx sec 2x dx
49. I e3lnxdx 50. J 60e 10 Sx dx
51. J (1 + e2X)50e2XdX 52. J (1 _ e4x)l00e4X
dX
53. J(3 + e ) 2 dx 54. J(2 + e") 3 dx
For Problems 55-58, evaluate the definite integral by using the fundamental theorem.
Show by numerical integration that your answer is correct.
55. r e0.4xdx 56. r e0 -2x dx
57. Jo2
(e" - e-x ) dx 58. ( (ex + e-x) dx
59. Rabbit Populat ion Prob lem: When rabbits were introduced to Australia in the middle
of the nineteenth century, they had no natural enemies . As a result, th eir population
grew exponentially with time . The photograph at the beginning of this chapter shows
what happened. The general equation of the exponential function for the number of
rabbits, R(t), is
R(t) = aekr_
a. Suppose there were 60,000 rabbits in 1865,
when t = 0, and that the population had grown
to 2,400,000 by 1867. Substitute these values
of t and R ( t) to get two equations involving
the constants a and k . Use these equations to
find values of a and k, then write the particular
equation expressing R ( t) as a function of t.
b. How many rabbits does your model predict
there would have been in 1870?
Sec
tion6-7: The Natural Exponent
ial Function:
TheInverse
of In 283
c. According to your model, when was the first pair of rabbits introduced into
Australia?
d. See George Laycock's The Alien Animals (Ballantine Books, 1966) for the eventual
outcome of the rabbit problem.
60. Depreciation Problem: The value of a major purchase, such as a house, depreciates
(decreases) each year because the purchase gets older . Assume that the value of
Richard Holmes's house is given by
v(t) = 85,000 e - o.osr,
where v(t) is the number of dollars the house is worth at t years after it was built.
a. How much was it worth when it was built?
b. How much did it depreciate during its eleventh year (from t = 10 to t = 11)?
c. What is the instantaneous rate of change (in dollars per year) of the value at
t = 1O?Why is this answer different from the answer you found in 60b?
d. When will the value have dropped to $30,000?
61. An Exponential Function is Not a Power Function! Figure out a way to show that the
formula for the derivative of a power function (that is, variable base and constant
exponent) does not work for an exponential function (that is, constant base and
variable exponent).
62. Proof of the Function of an Inverse Function Property: Prove the function of an inverse
function property, mentioned in this section and in Section 4-5. The author's student
Leighton Ku developed an easy proof that uses the definition of inverse function,
namely,
y = f - 1(X) if and only if X = f(y) .
*63. Z ero.;z
1 . era
P ro blem: Let f()x = lnx+sin(x-1
_ ex - I )
1
a. Show that f(x) takes on the indeterminate form 0/ 0 as x approaches 1.
b. Plot the graph off. Use a friendly window that includes 1.
c. What limit does f(x) seem to be approaching as x approaches 1?
d. Find the derivative of the numerator and the derivative of the denominator.
Evaluate each derivative at x = 1. What does the ratio of the two derivatives equal?
Surprising?!
64. Journal Problem: Update your journal with what you've learned since the last entry.
Include such things as those listed here.
The one most important thing you have learned since your last journal entry
The reasons for using e as a base for logs and exponential functions
The algebraic techniques for finding derivatives and integrals of the natural
exponential function and for finding the derivatives of any log function
Some instances in which exponential or log functions are used as mathematical
models
Anything about logs or exponentials that you plan to ask about at the next class
meeting
*This problem prepares you for Section 6-8.
284 Chapter
6: TheCalculus
ofExponential
andLogarithmic
Functions
6-8 Limits of Indeterminate Forms: !'Hospital's Rule
When you use one of the formulas to find a derivative, you are really using a shortcut
f(x) to find the limit of the indeterminate form 0/0. These formulas can be used to help
you evaluate other expressions of the form 0/0 . Figure 6-8a shows the graph of
4
= x ~ 2x
2
f(x) - 3.
nx
X lf you try to evaluate f(l ) , you get
1
Figure 6-80
f(l) = 1 ~:; 3 = .
The graph suggests that the limit of f(x) is 4 as x approaches 1.
The technique you will use to evaluate such limits is called l'Hospital's rule, named
after G. F. A. de l'Hospital (1661-1704), although it was probably known earlier
by th e Bernoulli brothers. This French name is pronounced "lo-pee-tal ' ." In older
writing, it is sometimes spelled l'H6pital, with a circumflex placed over the letter
o. You will also learn how to use !'Hospital's rule to evaluate other indeterminate
forms such as 00/ 00, 1 o0 , 00, and oo - oo.
00
,
OBJECTIVE Given an expr es sion with an ind eterminate form, find its limit using !'Hospital' s rul e.
L'Hospital's rul e is easy to use but tricky to derive. Therefore you will start out
by seeing how it works, then get some insight into why it works . The procedure
for finding th e limit of a fraction that has the form 0/0 or oo / oo is to take the
derivatives of the numerator and the denominator, then find the limit. For instance,
if g (x) / h (x) = (x 2 + 2x - 3) / (ln x) (given above),
Jim g(x) = lim g'(x) = lim 2x + 2 = .= 4
x - 1 h (x) x - l h'(x) x- 1 l/x 1 '
which agrees with Figure 6-8a. The following is a formal statement of l'Hospital's
rule.
Property:/'Hospital'sRule
If f(x) = gh ((x))
X
an d ifl im g(x)
x-c
= limh (x) = 0,
x-c
then li mf(x)
x-c
= lim
x-c
h :((x)), provide d the latter limit exists .
g
X
Corollaries of the rule lead to the same concl u sion if x - oo or if bo th g(x) an d h (x)
approach infinity.
Here's why l'Hospital's rule works. Because g(x) and h (x) both approach zero as x
approaches c, g(c) and h (c) either equal zero or can be defined to equal zero by
removing a removable discontinuity. The fraction for f (x) can be transformed to a
ratio of differ ence quotients by subtracting g(c) and h(c ), which both equal zero,
and by multiplying by clever forms of 1.
Section6-8: Limits of Indeterminate
Forms
: l' Hospitol's
Rule 28 5
g(x) - g(c)
g(x) g(x) - g(c) x - c
f(x) = h (x) = h (x) - h (c) = h (x) - h (c)
x - c
. g(x) - g(c)
11m =-------''----
1. f( ) x- c X- C g '( c) Limit of a quotient.
}~ x = lim h (x) - h (c) h '(c) Definition of derivative.
x- c X - C
If th e derivativ es off and g are also continuous at x = c, you can write
g'(c) = g'(limx ) = limg '(x) and h ' (c) = h '( limx ) = limh'(x ).
x-c x- c x-c x-c
Therefore
limf (x) = l~mx-c g:(x) . g '( x)
x-c hm x-c h (x) = hm
x-c -h ' (X )' Q.E.D.,
where the last step is justified by the limit of a quotient property used "backwards."
A formal proof of l'Hospital' s rule must avoid the difficulty that [h (x) - h(c )] might
be zero somewhere other than at x = c . This proof and the proofs of the corollaries
are not shown here because they would distract you from what you are learning.
A geometric derivation of l'Hospital's rule is presented in Problem 34 of Problem
Set 6-8 .
Example 1 shows you a reasonable format to use when you apply !'Hospital's rule.
The function is that given at the be ginning of this section.
Example 1 Find L = lim xz + 2x - 3
x- 1 ln x
Solution . x 2 + 2x - 3 0 L'Hospital's rul e applies becaus e the limit has the
11m ---- - -
form 0/ 0.
x- 1 ln X 0
= lim 2x + 2 - i L'Hospital's rule is no lon ger needed because the
x-l l /x 1 limit is no longer ind eterminate.
=4
Example 2 shows how to use l'Hospital's rule for an indeterminate form other
than 0/ 0.
Example 2 Evaluate lim x 2 e - x.
x - oo
Solution As x goes to infinity, x2 gets infinitely lar ge and e - x goes to zero . A graph of
y = x 2 e - x suggests that the expression goes to zero as x becom es infinite (Figure
6-Sb). To show this by l'Hospital's rule, first transform the expression into a
fraction.
y L'Hospital's rul e does n't app ly yet.
x- oo
x2 oo L'Hospital's rul e does apply now. Find th e
= lim-- - derivative of th e num erator and the denominator.
x- oo e x 00
2x oo
= lim - - - L'Hospital' s rul e applies again.
X x -oo ex 00
= lim 3.. - 3_ L'Hosp ital' s rul e is no lon ger neede d.
x- oo ex 00
Figure 6-8b =0 (finite)/ (infinite) - 0
286 Chapter
6:TheCalculus
of Exponential
andLogarithmic
Functions
Indeterminate
Exponential
Forms
If you raise a number greater than 1 to a large power, the resu lt is very large . A
positi ve numb er les s than 1 rais ed to a large power is close to zero. For instance,
lim l.OF = oo and lim 0 .99 x = 0.
x-oo x-oo
If an expression approaches 1 the answ er is indeterminate . You saw such a case
00
,
with th e definition of e, which is the limit of (1 + 1 / x) x as x approaches infinit y.
Expressions that take on th e form 00
and o0 are also indeterminate. Example 3
shows you how to evaluate an indeterminate form with a variable base and exponent.
Like logarithmic diff erentiation shown earlier in this chapter, the secret is to take
the log of th e expression. Then you can transform the r esult to a fraction and apply
l'Hospital's rule.
Evaluate limx 1i( l - xl.
Example 3 x-1
Solution The function f (x ) = x 11(l -xJ takes on th e indeterminate form 1 at x = l. The 00
gra ph of f (Figure 6-8c) shows a removable discontinuit y at x = 1 and shows
that the limit of f(x ) as x approaches 1 is a number les s than 0.5. The limit can
be found by using !'Hospita l's ru le after taking the log.
Let L = limx 110 - xl .
f(x )
x-1
0.5 Then ln L = In [limx 11(l - xl ] = lim[ln x 11(l -xl ] Reverse ln and lim .
x- 1 x- 1
. [ -- 1
= 11m ]n x J = 11m
. --]n X
- -0 L'Hospital's rul e appli es now.
1 X x-1 1- X x- I 1- X 0
X=l )'= Find th e derivative of the num era tor
= lim l / x = - 1
x-1 - 1 and th e d eno min ato r.
Figure 6-Bc
: . L = e- 1 = 0.367879 . .. In L = - l => L = e- 1
The answ er to Example 2 agrees with the graph in Figur e 6-8c.
Problem Set 6-8
DoTheseQuickly
The following problems are intend ed to refres h your ski lls. You should be able to do all
ten problems in less than five minutes .
Ql. e ~ -?- (as a decimal)
1
Q2. e- = -?- (without negative exponents)
Q3. ln e = - ?-
Q4. ln (exp x) = -7-
QS. e 111x
= -?-
Q6. If logb x = In x, then b = -?- .
Ql. logb x =-?- (in terms of the function ln)
QB. If f (x) = ex, then f'(x) = -?- .
Q9. f e-x dx = -?-
Q10. If f (x) = Jt' x sintdt, then f'(x ) = -?- .
10
Section
6-8:Limti s of Indeterminate
Forms:
!'Hospital's
Rule 287
For Problems 1 and 2, estimat e graphi cally the limit of f (x ) as x approaches zero. Sketch
the graph. Then confirm your es tim ate by using !'Hospit al's rul e.
. 2 sin 5x . 4 tan 3x
1. 11m --- 2 . 11m - --
x- o 3x x-o 5x
For Problems 3-30, find the indicat ed limit. Use l'Hospital's rule if necessary.
tanx sinx
3. 11m - - 4. 11m - -
x- 0 X x-0 X
1 - cosx x2
5. lim 6. lim
x-0 xz x-0 cos 3x - 1
sin x 1 - COSX
7. 11m -- 2 8. lim
x- o x x-0 X + X 2
3x
9. l 1m --
ln X
10. lim ~ 2
x- o 1/ x x-0 x
1 ex - e In x - x + 1
11. 1m -- 12. lim 2
x- 1 5lnx x- 1 x - 2x + 1
13. 11m--
3x + 5
- r
14. 1m-tanx-
x-2 COSX x-2 X - 2
ex x3
15. lim - 16. lim -
x-oo xz x-oo ex
3x + 17 . 2 - 7x
17. 11m 18. 11m --
x-oo4x - 11 x- 3 + 5x
00
x 3 - 5x 2 + 13x - 21 3x 5 + 2
19. 11m 3 20. 11m _
x- 4x + 9x 2 - llx - 17
00 x- 7x >- 8
00
21. lim xx 22. lim (sin x) sinx
x- o+ x- o+
23. lim (sin x ) tan x 24. lim x l /(x- 1)
x- rr;2- x- 1 ~
25 . lim(l + ax) 11x (where a = positive constant) 26. lim(l + ax )lfx (where a = constant)
X--+oo x-0
27. lim x3/(lnx) 28. lim (7x)S/(ln x)
x- o+ x- o+
29. lim(!
x- 0 X
1
- - - )
ex - 1 x- o x sinx
(1
30. 11m - - -- 1 )
31. Infinity Minus Infinity Problem: Let f (x ) = sec 2 x - tan 2 x. Because both sec (rr / 2)
and tan (rr / 2) ar e infinit e, f (x ) tak es on th e indet erminat e form oo - oo as x approaches
1. Naive thinking might lead you to susp ect that oo - oo is zero becaus e the difference
b etween two equal number s is zero. But oo is not a numb er. Plot th e graph off . Sketch
the result, showing what happ ens at x = 1, 3, 5, ... . Explain th e graph based on what
you recall from tri gonom etr y.
32. /'Hospital's Surprise Problem! Try to evaluate lim sec x by using l'Hospital's rule.
x- rr/2 tanx
W)lat happens ? Find the limit by using som e oth er m ethod.
33. Zero to the Zero Problem: Often , th e indet erminat e form o0 equals 1. For instance, th e
expression (sin x) sinx approa che s 1 as x approa ches 0. But a fun ction of the form
f (x ) = xk/(lnx )
288 Chapter
6: TheCalculus
of Exponential
andLogar
ithmic Functions
(where k stands for a constant) that goes to o0 does not, in general, approach 1. Apply
!'Hospital's rul e appropriately to ascertain the limit of f (x) as x approaches zero. On
your grapher, investigate the graph of f(x ) . Explain your results.
34. L'Hospital's Rule, Geometrically: In this problem you will
y
investigate
2
f (x) = g(x) = 0.3x - 2.7
0.2x 2 - 2x + 4.2'
h(x)
which approaches 0/ 0 as x approaches 3. You will see !'Hospital's
rule geometrically .
a. Confirm that g(3) = h(3) = 0.
X
b . Figure 6-8d shows the graphs of g and h, along with the
tangent lines at x = 3. Find equations of the tangent lines.
Leave the answers in terms of (x - 3). I
I
c. Because g and h are differentiable at x = 3, they have local I
I
I
linearity in a neighborhood of x = 3. Thus the ratio g(x) I h(x ) I
is approximately equal to the ratio of th e two linear functions Figure 6-Bd
you found in 34b. Show that this ratio is equal to g'(3)/ h'(3).
d. Explain the connection between the conclusion you made in 34c and !'Hospital's
rule. Explain why the conclusion might not be true if either g(3) or h (3) were not
equal to zero.
e. Plot the graph off. Sketch the result, showing its behavior at x = 3.
35. Continuous Compounding of Interest Problem: Suppose that $1000 is earning interest
at 6% per year, compounded annually (once a year). At the end of the first year, it will
earn (0.06)(1000), or $60, so there will be $1060 in the account. This number can be
easily found by multiplying the original $1000 by 1.06, which is (1 + interest rate). At
the end of each subsequent year, the amount in the account at the beginning of that
year is multiplied again by 1.06, giving the following.
year total at end of year
0 1000
1 1000(1.06)
2 1000(1.06) 2
3 1000(1.06) 3
t 1000 ( 1.06)t
a. If the interest is compounded semiannually (twice a year), the account gets half the
interest rate for twice as many time periods. If m(t) is the number of dollars in the
account after t years, explain why
2
m(t) T
= 1000 ( 1 + 0 06) t
b. Write an equation for m(t ) if the interest is compounded n times a year. Then find
the limit of this equation as n approaches infinity to find m(t) if the interest is
compounded continuously. Treat t as a constant, because it is n that is varying as
you find the limit.
Section
6-8:Limitsof Indeterminate
Forms:
!'Hospital's
Rule 289
c. How much more money will you have with continuous compounding rather than
with annual compounding after 5 yr? After 20 yr? After 50 yr?
d. Quick! Write an equation for m(t) if the interest is 7% per year compounded
continuously .
36. Order of Magnitude of a Function : Let L be th e limit of f(x) / g(x) as x approaches
infinity. If L is infinite, then f is said to be of a higher order of magnitude than g . If
L = 0, then f is said to be of a lower order of magnitude than g. If Lis a finite nonzero
number, then f and g are said to hav e the same order of magnitude.
a. Rank each kind of function according to its order of magnitude.
i. Power function, f(x) = xn, where n is a positive constant
ii. Logarithmic function, g(x) = ln x
iii. Exponential function, h (x) = e"
b . Quick! Without using l'Hospital's rule, evaluate the following limits .
. 11m -ln 3x
1. - ii. Jim -xioo iii. lim e 0.3x 1v.11mJx V. lim --
ex
x- oo 5 x x-oo e 0 .01 x x -oo 1001 n x x-oo x x-oo e0, 2x
3 7. Journal Problem: In your journal, write something about various indeterminate forms .
Include examples of functions that approach the following forms.
Kindsof Indeterminate
Form
Limits that take the following forms can equal different numbers at different times and thus
cannot be found just by looking at the form:
0 00
0' 00
oo-0, 00-00, 1 00
,00, and o0 .
Try to pick examples for which the answer is not obvious. For instance, pick a function
that goes to 0/ 0 but for which the limit does not equal l. Show how other indeterminate
forms can be algebraically transformed to 0/0 or to oo / oo so that !'Hospital's rule can
be used.
6-9 Derivative and Integral Practice for
Transcendental Functions
The properties you have learned in this and previous chapters allow you to
differentiate and integrate algebraically almost all of the elementary transcendental
functions. These functions include trigonometric, inverse trigonometric, logarithmic,
and inverse logarithmic (exponential) functions . In this section you will learn how
to integrate exponential functions with any base (not just e) and how to integrate
the remaining four trigonometric functions (sec, csc, tan, and cot) . In Chapter 9, you
will learn integration by parts, which will let you integrate logarithmic and inverse
trigonometric functions algebraically.
29 0 Chapter
6: TheCalculus
of Exponential
andLogarithmic
Functions
OBJECTIVEBe able to differentiat e and integrat e algebrai cally fun ction s involving logs and exponen -
tials quickly and correct ly so that you ecm concen trate on the applications in the following
chapt ers .
Integrals
of Baseb Exponential
Functions
In Section 6-7, you learned an algebraic method for integrating exponential functions
with e as the base. Exponentials with other bases can be integrated by transforming
them to base e first. Suppose you must integrate
Jsxdx .
Any p ositive number, such as 8, can be written as a power of e.
Let 8 = ek, where k stands for a positive constant.
Then ln 8 = ln ek Take In of bo th sid es.
= k(lne) Reason?
=k Reason?
.. 8 = eln B. Sub stin1te In 8 for k.
This equ ation is an example of th e genera l relationship between a func tion an d its
inverse ,
f(f - 1(X)) = X and f - 1(f (x ))= X.
The relationship leads to a definition of exponentials with bases other thane .
bX= (elnb)X Repla ce b with e1n b_
= exlnb Multiply th e exponent s.
Definition:Exponentialwith Baseb
Using this definition, the integra l above can be transform ed and integrated as
follows.
J sxdx = J e x !n B dx Definition of 8' .
= - - Jexln (ln 8dx)
1
ln 8
8
Transform to the differ ential of th e in side fun ction .
= --1 ex1ns + c Je" du = e" + C
ln 8
= 1:88 x + C Sub stitut e back 8 = e 1n 8.
Once you see the pattern, you can integrate or differentiate exponentials quickly .
Section 6-9: Derivative and IntegralPracticefor Transcendental
Functions 291
exp
Il, '.'.!
' I_
Properties:DerivativeandIntegralof an Exponential
Function
Derivative: :x(bx) = bx In b Multiply bx by In b .
Integral : f bx dx = bx l: b + C Divide bx by In b.
Integrals
of tan,cot,sec,andcsc
The tangent function can be int egra ted by first transforming it to sine and cosine,
using the quotient properties from trigonometry.
I tanxdx = J sinxdx
cosx
= -J- 1
cosx
-(- sinxdx )
Tran s form to the integra l of th e recipro cal
function .
= - ln Icos x I + C
= + ln 1-1-1
COSX
+C Inn = - ln (l / n )
= In lsecx l + C
The cotangent function is int egra ted the same way. The integral of secant and
cosecant are trickier! A clever transformation is required to turn the integrand into
the reciprocal function. The key to the transformation is that the derivative of sec x
is sec x tan x and the derivativ e of tan x is sec 2 x. Here's how it works .
secx + tanx
I I
sec x dx = sec x -----
secx + tanx
dx Multipl y by a "clever" form of l.
1
=J (sec 2 x + sec x tanx) dx Write as th e reciprocal function.
secx + tanx
(sec 2 x + secx tanx) dx is th e
= ln [sec x + tanx l + C diff erential of th e denominator.
The formulas for f cot x dx and for f csc x dx are derived similarly . These integrals
are listed, along with sine and cosine, in the box belo w.
Properties:Integralsof the Six Trigonometric
Functions
f sinx dx = - cosx + C
Jcosxdx = sinx + C
f tanx dx = - ln Icos x i+ C = ln lsecx l + C
Jcotxdx = ln lsinxl + C = - ln Iesex i+ C
Jsecxdx = ln lsecx + tanx l + C
Jcscxdx = -ln lcscx + cot x i+ C = In lcscx - cot x i+ C
The following boxes contain properties of logs and exponentials for doing calculus
algebraically .
292 Chapter
6: TheCalculus
of Exponential
andLogarithmic
Functions
Properties:NaturalLogsandExponentials
Definition
of theNatural
Logarithm
Function:
1n x = f".:!.
l t
dt (where xis a positive number)
Calculus
of theNatural
Logarithm
Function:
d 1
- On x) =-
dx X
f ln x dx (to be introduced in Chapter 9)
Integral
of theReciprocal
Function
(fromthedefinition):
f ~du = ln lul + C
Logarithm
Properties
of In:
Product: ln (ab) = ln a + ln b
Quotient : ln (a/ b)= ln a - ln b
Power: ln (ab) = b ln a
Intercept : ln 1 =O
Calculus
of theNatural
Exponential
Function:
d
-(ex)= ex
dx
Inverse
Properties
of LogandExponential
Functions
:
Function
Notation
forExponential
Functions:
exp(x) = ex
Definition
of e:
e = lim
n-0
(l + n ) 11" = lim (l + 1 / n)n
n-oo
e = 2. 7182818284 . .. (a transcendental number, a nonrepeating decimal)
Sec
tion6-9:Derivative
andIntegralPractice
for Transcendental
Functions 293
Properties:Baseb Logsand Exponentials
of NaturalLogsandBasee Logs:
Equivalence
ln x = logex for all x > 0
of Baseb Logs:
Calculus
d 1 1
- (logbx ) =- -
dx ln b x
f log bx dx (to b e introdu ced in Chapter 9)
of Baseb Exponential
Calculus Functions:
!!._(bx) = (ln b )bx
dx
f b'' d X = _ l_ bx + C
In b
Change
of BaseProperty
forLogarithms:
logax = logb x = ln x
logba ln a
Problem Set 6-9
For Problems 1-5 6, find y '. Work all the problems in th e order they appear , rather than
just th e odds or just the evens. (This is wh y th ey are numb er ed down the p age instead of
across.)
1. y = ln (3x + 4) 13. y = eSlnx
2. y = In (3x 5 ) 14. y = e COSX
3. y = In (e 3x) 15. Y = COS (eX)
4. y = ln (sin4x ) 16. y = (cos 3 x)(e 3x)
5. y = ln (cos 5x) 17. y = exp (x 5 )
6. y = In (e 5 ) 18. y = exp (ex)
7. y = ln (cos (tanx)) 19. sin y = ex
8. y = ln .Jx 2 - 2x + 3 20. y =ex ln x
X1
9. y = cos (ln x) 21. y =
l t
dt f-
10. y = sinx ln x 22. tany = ex
11. y = el x 23. y = ln (elnx)
12. y = ex3 24 . y = 2"
294 Chapter
6: TheCalc
uluso~ I andLogarithmic
Funct
ions
log3X
25. y = exln2 41. Y = log3e
log 10x
26. y = e2lnX 42. y=--
log10e
27. y = x 2 43. y = (logsx) (ln 8)
28 . y = exln 2. 44 . y = (log4x) 10
29 . y = x x 45 . Y = lOg5X7
30 . y = xlnx - x 46. y = tane
31. y = ex (x - 1) 47. y = esinx
32. y =( ex+ e- x) 48 . y=lncscx
33 . Y = (ex - e- x) 49. Y = 3s
ex
34. y = 1 + ex 50 . y = ln (cos 2x + sin 2x)
35. Y = 5x 51. y = sinx
36. y = log sx 52 . y = sin - 1x
37. y = x - 7 Iog 2x 53. y = cscx
38. y = 2- xcosx 54 . y = tan - 1 x
39 . y = e- 2x1n 5x 55. y = tanx
7x
40 . y = ln 7 . 56. y = cotx
For Problems 57-80, integrate . Work all the problems in the order they appear, rather
than just the odds or just the evens .
,/ 57. I e4xdx 67. J 2xdx
58. J e 4 dx 68. J (x - 02 + 3x) dx
v 59. I x3ex4 dx / 69 . J ~dx
60. I cos X . esinx dx 70. f 4x dx
/ 61. J (lnxx)s dx / 71. J (ln x) 9 ~ dx
62. J 5x dx 72. J cosxdx
63 . I exlnSdx v 73. J elnx dx
64. J (ex + e- x) dx 74 . J ln (e 3x) dx
65 . rl
1
-dt
t
./ 75. f Odx
66 . J e- x dx 76. J cos x sec x dx
') Q C::
77. f sec 2x dx 79. Jcot4xdx
78. f tan 3x dx 80. f csc 5x dx
For Problems 81-90, find the limit of the given expression.
81. lim 1 - cosx 86. lim (1 + 0.0 3 ) x
x- 0 X x -oo X
82. l X 87. xlim(l + 0.03x) l /x
}!] 1 - COSX -oo
X 2x
83. lim 88. lim -
x- rr/2 1 - COS X x -oo x2
84. lIm X 89. lim(0 .5x) 31(2-x)
1 + COS X
x - rr x-2
85. lim
X-> Q
Sx - sin 5x
x3
1
90. lim ( - 3 -
x- 0 e x - 1
_
3x
l_)
6-10 Chapter Review and Test
In this chapter you have studied the calculus of logarithmic and exponential
functions. It all started with a real-world population problem in which dP /dt was
directly proportional to p. In solving this differential equation for P, you encountered
f p - 1 dP, which you could integrate numerically but not algebraically .
The problem was solved by definition. The function ln x was defined as a definite
integral whose derivativ e is 1/ x. This function was found to hav e th e properties
of logarithms, giving credence to the term natural logarithm . By starting at the
other end, you found that the derivativ e of the function y = loge x is also 1 / x
and thus concluded that ln x = loge x. This fact allowed you to differentiate the
inverse function, y = ex. As a result, you became able to differentiate and integrate
exponential functions such as y = 2x, another problem you had been able to solve
numerically but not algebraically.
The Review Problems below are numbered according to the sections of this chapter .
The Concepts Problems allow you to apply your knowledg e to new situations . The
Chapter Test is like a typical classroom test your instructor might give you.
Review Problems
RO. Update your journal with what you've learned since the last entr y. Include such things
as those listed here .
The one most important thing you have learned in studying Chapter 6
Which boxes you have been working on in the "define, understand, do, apply" table
Your ability to do calculus algebraically, not just numericall y, on logs and
exponentials
rl - - ,L_ - I . TL . r -I I r ,..
The new techniques and properties you have learned
Any ideas about logs and exponentials you must ask about before the test on
Chapter 6
Rl. a. If money in a savings account earns interest compounded continuously, the rate
of change of the amount of money in the account is directly proportional to the
amount of money there . Suppose that for a particular account, dM / dt = 0.06M,
where M is the number of dollars and t is the number of years the money has
been in the account . Separate the variables so that M is on one side and t is on
the other. If $100 is in the account at time t = 0, show that when t = 5 yr,
5
Ix
JOO
M- 1 dM = r
Jo
0.06 dt = 0.3,
where x is the number of dollars in the account after five years.
b. Use your grapher's numerical integration and solve features to find, approximately,
the value of x for which the left-hand integral equals 0.3.
c. To the nearest cent, how much interest will the account have earned when
t = 5 yr?
R2. a. Explain why f x - 1 dx cannot be evaluated by using the power rule for integrals.
b . LetL(x) = ft (1 /t) dt. Evaluate L(2), L(3), L(4), L(B), and L(12) by using numerical
integration. Show that these values are equal to the natural logarithms ln 2, ln 3,
ln 4, ln 8, and ln 12 you can find on your grapher.
c. Show that the values of function L in R2b have the properties of logarithms,
namely, L(ab) = L(a) + L(b), L(a / b) = L(a) - L(b), and L(ab) = bln a.
R3. a. Differentiate .
i. y = (ln 5x) 3 ii . f(x) = ln x 9 iii. y = csc (ln x)
x2
iv. g(x) = J1 csc t dt
b . Integrate.
.f
1.
sec x tan x d X
----
.. J-310 d
11. - X
secx -2 X
c. Memory Retention Problem: Paula Tishan prides herself on Y Number of names
being able to remember names. She knows that the number 50 she remembers
of names she can remember at an event is a logarithmic
function of the number of people she meets there, and she
figures that her particular equation is
y = 1 - 101 ln 101 + 1011n (100 + x),
umb er of
where y is the number of names she remembers when she peopl e she meets
X
meets x people. The graph is shown in Figure 6-1 Oa.
50
i. How many names can she remember if she meets 100
people? Just 1 person? What percentage of the people Figure 6-1Oa
she meets do these two numbers represent?
ii. At what rate does she remember names if she has met 100 people? Just 1
person?
iii. What is the greatest number of names she is likely to be able to remember
without forgetting any? What assumptions do you make in coming up with
your answer 7
Section
6-10:Chapter
Review
andTest 297
V,~n
ri
R4. a. State the definition of ln.
b . State the algebraic definition of logarithm.
C. State the uniqueness theorem for derivatives.
d. State the property of the In of a power.
e. State and prove the property of the ln of a quotient.
RS. a. Write the equation of the derivative function.
i. y = 1oox ij. f(x) = 3.7 10-2x iii. r(t) = t tanr
b. Differentiate logarithmjcally: y = (5x - 7) 3 (3x
+ 1) 5 Concentration in parts
per million
C. Vitamin C Problem : When you tak e vitamm C, its concen-
tration, C(t) parts per million, in your bloodstream rises
rapidly, then drops off gradually. Assume that if you take a
500 mg tablet, the concentration is given by
C(t) = 200t x 0.61 ,
where t is time in hours since you took the tablet. Figure Figure 6-1Ob
6-1Ob shows the graph of C.
i. Approximately what is the highest concentration, and when does it occur?
ii. How fast is the concentration changing when t = 1? When t = 5? How do you
interpret the signs of these rates?
iii. For how lon g a period of tim e ,vill the conc entration of th e vitamin C be above
50 ppm 7
iv. If you take the vitamin C with a cola drink, the vitamin C decomposes more
rapidly. Assume that the bas e in the equation changes from 0.6 to 0.3. What
effect will this change have on the highest concentration and when it occurs?
What effect will this change have on the leng th of time the concentration is
above 50 ppm ?
R6. a. Write the definition of e.
b . Write an equation relating base e logs and natural logs.
c. Write an equation expressing log bx in terms of In.
d. Differentiate.
ii. f(x) = logz(COSX) iii. y = log 5 9x
e. Check your journal for Section 6-6. What is one important thing you wrote?
R7. a. Sketch the graph.
i . y = ex ii. y = exp ( - x) iii. y = In x
b. Differentiate.
i. f (x) = xt. 4 exp(Sx) ii. g(x) = sin e - Zx iii. y = eln X
C.Integrate .
i. f1oe - ZXdX ii. fe COSX
SinXdX iii. fzeX p (- 0.lX ) dX
d. Tell why it is convenient in calculus to use the (untidy!) number e as a base .
e. Radioactiv e Decay Probl em: Strontium 90 is a radioactive isotope formed when
uranium fissions . The percentag e of th e isotop e remaining after t years is given by
p(t ) = 1ooe- o.02s1.
i. How much of the original strontium 90 is left after 5 yr?
ii. At what rate is the percentage of strontium 90 changing after O yr? After 5 yr?
298 Cha
pter 6: TheCalcu
lusof Exponent
ial andLogarithm
ic Functions
iii. What is the half -life of strontium 90 (the time required for it to decay to 50%)?
iv. How long would it take for only 0.001% of the strontium 90 to remain?
f. Chemotherapy Prob lem : When a patient receives chemoth erapy, the concentration,
C(t) parts per million, of chemical in the blood decreases exponentially with time.
Assume that
C(t) = 1soe-01 61,
where t is the number of days since the treatment (Figure 6-1 Oc).
C(t)
150
Expo sur e=
E{x) ppm-da ys
X 20
Figure 6-1Oc
i. The amount of exposure to the treatment, E(x), after t = x days may be
expressed as the product of the concentration and the number of days . Explain
why a definite integral must be used to calculate the amount of exposure.
ii. Write an equation for E(x) . How much exposure has the patient received in S
days? In 10 days? Does there seem to be a limit to the amount of exposure as
x becomes very large? If so, what is the limit? If not, why not ?
iii. Quick! Write an equation for E' (x). At what rate is E(x) changing when x = S?
When x = 107
RS. Evaluate the limits.
. 2x 2 - 3 x2 - cosx + 1
a. 1Im--- b. lim C. limx 3e - x
x -oo 7- Sx 2 x-0 ex - X - 1 x-oo
d. lim xt an (rr x / 2) e. lim3x 4 f. lim (tan 2 x - sec 2 x)
x- 1 x-2 X- TT/ 2
g. Write as many indeterminate forms as you can think of.
R9. a. Differentiate .
i. y = ln (sin 4 7x) iii . Y = COS(2X) iv. y = log 3 (x 4 )
b. Integrate.
i. Je - l.7x dx ii. Jzsccx(secx tanx dx)
iii. J(S + sinx) - 1 cosxdx . f 2 dz
IV. I
5 I
c. Find the following limits.
. l" tan3x
I.}~~
.. 1Im
ll.
x-oo
(1- -3)x
X
Section6-10: Cha
pter ReviewandTest 299
,
I
n
Concepts Problems
Cl. Derivation of the Memory Equation: In Problem R3c, the number of names, y,
remembered as a function of peop le met, x, was said to be
y = l - 1011n 101 + 1011n (100 + x).
Suppose that, in general,
y=a + bln(x+c).
State why the following conditions are reasonable for y and y'. Calculate the constants
a, b, and c so that these conditions will be met .
y = l when x = l
y' = l when x = l
y = 80 when x = 100
C2. Integra l ofln Prob lem : In Chapter 9 you will learn how to antidifferentiate y = In x . In
this problem you are to try to discover what this antiderivative equals by examining
graphs and a table of values. Figure 6- lOd shows
Yi = ln X
Y2 = Jln x dx, with C = 0.
The table shows values of x, Yi, and Yz- From the tables and the graph, see if you can
figure out an equation for y 2 Write a description of methods you tried and whether
or not these methods helped you get the answer.
X ln X JIn x dx
0.5 - 0.6931427 - 0.846 5736 y Integral ,'
1.0 0 -1 graph ,'
2.0 0.6931427 - 0.61370 56
3.0 1.098612 3 0.2958 369
4.0 1.3862944 1.5451774
5.0 1.609437 9 3.0471896
6.0 1.7917595 4.7505 568
10.0 2.3025851 13.025851 Figure 6-1Od
C3. Continued Exponentiation Function Problem: Let g(x) = x', a function for which both
the base and the exponent are variable. Find g '(x ) . Then suppose that the number of
x's in the exponent is also variable. Specifically, define the continued exponentiation
function, cont(x), as follows:
x
cont(x) = x-'' ,
where there is a total of x x's in the exponent . For instance,
cont(3) = 3333 = 3327 = 31525597 484981 1
which has over 3.6 billion digits . Figure out how to define cont(x) for noninteger
values of x . Try to do so in such a way that the resulting function is well defined,
300 Chapter
6: TheCalculus
of Exponentia
l andLogarithmic Functions
continuous, and differentiable for all positive values of x, including the integer values
Of X. y =lnx
C4. Every Real Number Is the In of Some Positive Number : Figure
6-lOe shows the graph of f (x) = In x . The graph seems to be
increasing, but slowly. In this problem you will prove that there is
no horizontal asymptote and that the range of the ln function is X
all real numbers.
a. Prove by contradiction that In is unbounded above. That is,
suppose that there is a positive number M such that ln x :::;M
for all values of x > 0. Then pick a clever value of x and show
that you get an answer greater than M for ln x. Figure 6-1Oe
b. Prove (quickly!) that ln is unbounded below.
c. Prove (quickly!) that ln is continuous for all positive values of x.
d. Prove that for any two numbers a and b, if k is between In a and ln b, then there
is a number c between a and b such that Inc= k. You should find that sketching
a graph and using the intermediate value theorem are helpful.
e. Use the connections among C5a-d to prove that any real number, k, is the natural
log of some positive number. That is, the range of 1n is {real numbers}.
f. Use the fact that In and exp are inverses of each other to show that the domain of
exp is the set of all real numbers and that its range is the set of positive numbers.
cs. Derivative of an Integral with Both Variable Upper and Lower Limits: You have learned
how to differentiate an integral such as g(x) = J;sin t dt between a fixed lower limit
and a variable upper limit. In this problem you will see what happens if both limits
of integration are variable.
a. Findg '(x) ifg (x) = fx~ sintdt .
b. Find g ' (x) if g(x) = J:
;nxsintdt.
c. Write a generalization: If g (x ) = S:c~/f(t) dt, then g ' (x) = -?-.Include this gener-
alization in your journal.
d(cabin)
C6. What d oes f b.
ca m
equa 17.
Chapter Test
Tl. Write the definition of natural logarithm.
T2. Evaluate In 1.8 approximately by using a midpoint Riemann sum with n = 4 in-
crem ents. Show that your answer is close to the value of ln 1.8 you would get by
calculator .
T3. If g(x) = J; sintdt, write an equation for g'(x). What theorem tells you how to get
this answer quickly?
T4. The uniqueness theorem for derivatives states that if f(a) = g(a) for some number
x = a, and if f'(x) = g'(x) for all values of x, then f(x) = g(x) for all values of x . In
the proof of the theorem, you assume that there is a number x = b for which f (b)
does not equal g(b). Show how this assumption leads to a contradiction of the mean
value theorem.
Section
6-10:Chapt
er Review
andTest 301
TS. Use the uniqueness theorem to prove that ln x = loge x for all positive numbers x.
T6. Let f(x) = ln (x 3 ex) . Find f' (x) in the following two ways. Show that the two answers
are equivalent.
a. Without first simplifying the equation for f(x )
b. First simplifying the equation for f(x) by using ln properties
For Problems T7-Tll, find an equation for the derivative. Simplify.
T7. y = e2x ln x 3 T8. v = ln (cos lOx)
T9. f(x) = (logz 4x)7 TlO. t(x) = ln (cos 2 x + sin 2 x)
[n X
Tll . p(x ) = I1 e 1sint dt
For Problems Tl2-Tl5, evaluate the integra l.
Tl2 . Je 5 X dx Tl3. J(ln x )6 dt
Tl4 . Jsec Sx dx Tl 5. Jo25x dx (algebraically)
. d . 5 - 3x Tl 7. Find lim (tanx) cotx_
Tl6. Fm 1im -1-4 -.
x -oo n X x - rr 12-
Tl8. Force and Work Problem: If you pull a box across the floor,
F(x)
you must exert a certain force. The amount of force needed
may increase with distance if the bottom of the box becomes
damaged as it moves. Assume that the force need ed to move a
particular box is given by
F(x) = 60e 0 1X,
where F(x) is the number of pounds that must be exerted when
5 ~
X
the box has moved x feet (Figure 6-lOf). Answer the following Figure 6-1Of
questions.
a. At what rate is th e force changing when x = S? When x = 10?
b. Recall that work (foot-pounds) equals force times distance moved. Explain why a
definite integral is used to find the work done moving the box to x = 5 from x = 0.
c. Write an integral for the work done in moving the box from x = 0 to x = 5. Evaluate
the integral by using the appropriate form of the fundamental theorem to get
a mathematical-world answer (exact). Then write a real-world answer, rounded
appropriately.
6-11 Cumulative Review: Chapters 1-6
The following problems constitute a "semester exam" that will allow you to
demonstrate your mastery of the concepts as you hav e studied them so far. Another
cumulative review appears at the end of Chapter 7.
302 Chap
ter 6:TheCalculus
of Exponential
andLogarithmic
Functions
Problem Set 6-1 1
1. The derivative of a function at a point is its instantaneous rate of change at that point.
For the function
f(x) = zx,
show that you can find a derivative numerically by calculating f'(3), using a symmetric
difference quotient with 6x = 0.1.
2. A definite integral is a product of x and y, where y is allowed
to vary with x. Show that you can calculat e a definite integral
graphical ly by estimating the integral of g (x ) , shown in Figure g(x)
6-lla, from x = 10 to x = 50. s
3. Derivatives and definite integrals are defined precisely by 6 ...
using the conce pt of limit. Write the definition of limit .
4. Intuitiv ely, a limit is a y-value that f(x) stays close to when x
is close to a given numb er c. Show that you und ersta nd the
symbol s for, and the meaning of, limit by sketching the graph
X
of one function for which both of the following are true.
10 50
Jim f(x) = 4 and lim f(x) = - oo
x-3 - x-3 + Figure 6-11 a
5. Limits are the basis for the formal definition of derivative. Write this definition .
6. Show that you can operate with the definition of derivative by usin g it to show that if
f(x) = x 3 , then f'(x) = 3x 2 .
7. Properties such as that in Problem 6 allow you to calculate derivatives algebraically,
thus gett ing exact answers. Find f'(5) for the function in Probl em 6. Then find
symme tri c difference quotients for f'(5) by using 6x = 0.01 and 6x = 0.001. Show
that the difference quotients really get closer to the exact answer as 6x decreases.
8. When composite functions are involved, you must remember
th e chain rule. Find the exact value of f'(7) for
f(x) = ~ 3x - 5. f (x)
9. Derivatives can be interpreted grap hically. Show that you
understand this graphica l int erpretation by constructing an
ap propriate line on a copy of the graph in Figure 6-1 lb for
the function in Problem 8.
10. Definite integrals can be calculated numerically by using
Riemann sums. Show that you understand what a Riemann
sum is by finding an upper sum, using n = 6 subintervals for
4 ?
J x- dx .
I Figure 6-1 1b
11. The definition of definite integral involves the limit of a Riemann sum. For the integral
in Problem 10, find midpoint Riemann sums with n = 10 and n = 100 increments.
What limit do these sums seem to be ap pro aching?
Section6-11:Cumula
tiveReview:
Chapter
s 1--6 303
12. Indefinite integrals are antiderivatives. Evaluate the following.
a. J cos 5 x sinx dx b. J ~ dx c. J tanx dx
d. Jsecxdx e. J (3x - 5) 1 i 2 dx
13. The fundamental theorem of calculus gives an algebraic way to calculat e definite
integrals exact ly, using indefinite integrals . Use the fundamental theorem to evaluat e
r x 2 dx
from Problem 10. Show that your answer is the number you conjectur ed in Problem
11 for the limit of the Riemann sum s.
14. The fundamental theorem is proved by usin g the mean value theorem as a lemma .
State the mean value theorem. Draw a graph that clearly shows you und erstand its
conclusion.
15. M,c:hof calculu s involves learnin g how to do algebraicall y the things you have learned
how to do graphically or numerically. Use implicit differentiation to find dy /dx if
y = x 917,
thereby showing how the power rule for the derivative of functions with integer
exponents is exten ded to functions with noninteger exponents.
16 . Explain why th e power rule for derivatives never gives x- 1 as the answer to a
differentiation problem.
17. The fundamental theorem in its other form lets you take the derivative of a function
defined by a definit e integral. Find f '(x) if
tan x
f(x) = J 1
cos 3t dt.
18. Show how the fund amenta l theorem in its second form lets you write a function who se
derivative is x - 1 .
19. The function you shou ld have written in Problem 18 is the natural logarithm function.
Use the uniqueness theorem for derivatives to show that this function has the
property of the log of a power. That is, show that
ln xa = aln x
for any constant a and for all values of x > 0.
20. Using the parametric chain rule, you can find dy /dx for functions such as
X = 5 COS t
y = 3 sint.
Write a formula for dy /dx in terms oft .
304 Chapter
6: TheCalcu
lus of Exponentia
l andLogarithmic
Func
tions
21. The ellipse in Figure 6-1 lc has the parametric equations given
in Problem 20. Find dy /dx if t = 2. Show graphically that your
answer is reasonable. :lUL+
:
,nUiU
: . :
..
. : : ;
;---;--;
22. Derivatives can be applied to real-world problems. Suppose '
: : : : . : X
that a car's position is !- ..~-...,
~
y = tan - 1 t,
where y is in feet and t is in seconds. The velocity is the
instantaneous rate of change of position, and the acceleration
is the instantaneous rate of change of velocity. Find an Figure 6-1 l c
equation for the velocity and an equation for the acceleration,
both as functions of time.
23. Derivatives can also be applied to problems from the mathematical world. For instance,
derivatives can be used to calculate limits by using l'Hospital's rule . Find
e 3x - 1
lim---
x-o sin 5x
24. In the differentiation of the base b logarithm function, the limit
L = lim(l + n) 11n
n- 0
appears. By appropriate use of !'Hospital's rule, show that this limit equals e, the base
of natural logarithms.
25 . Simpson's rule can be used to find definite integrals numerically if the integrand is
specified only by a table of data. Use Simpson's rule to find the integral of f(x) from
X = 2 to X = 5.
X f(x )
2.0 100
2.5 150
3.0 170
3.5 185
4.0 190
4.5 220
5.0 300
26. For the solid cone in Figure 6-11 d, cross sections perpendicu-
lar to the x-axis, x units from the vertex, are circles of radius
y = (r / h)x, where r is the radius of the cone's base and h is
its altitude (both constants). The volume of an object equals X
(cross-sectional area)(height) . Definite integrals provide a way
in which to evaluate (dependent variable)(independent vari-
able). Use this information to derive the geometric formula
for the volume of a cone,
V = rrrh.2
Figure 6-1 l d
27. It is important for you to be able to write about mathematics. Has writing in your
journal helped you better understand calculus? If so, give an example. If not, why not?
Section6-11: Cumulative
Review
: Chapters
1--6 305
CHAPTER
7 The Calculus of
Growth and Decay
Bristlecone pine trees in California's White Mountains are still alive
after thousands of years. Their ages can be found by counting the
growth rings and by measuring carbon 14, absorbed when the tree
grew. The rate of decay of carbon 14 is proportional to the amount
remaining . Integrating the differential equation expressing this fact
shows that the amount remaining is an exponential function of time.
307
~I Mathematical Overview
If you know the rate at which a population grows, you can use
antiderivatives to find the population as a function of time. In
Chapter 7 you will learn ways to solve differential equations for
population growth and other related real-world phenomena. You
will solve these differential equations in four ways.
Graphically The logo at the top of each even- y
numbered page of this chapter
shows three particular solutions
of the same differential equation.
The graph here also shows the
slope field for this differential
equation.
Numerically X. Y1 Y2 Y3 X
0 0.5 1.0 1.5
1 0.64 ... 1.28 ... 1.92 ...
2 0.82 ... 2.64 ... 2.47 ...
3 1.05 ... 2.11 ... 3.17 ...
Algebraically :. = 0.2 Sy~ y = Ce 0 -25 x, a differential equation.
Verbally I learned that the constant of integration is of vital importance in the
solution of differential equations. Different values of C give different
particular solutions. So I must always keep in mind, "Remember +CJ"
308
7-1 Direct Proportion Property of
Exponential Functions
In Chapters 1-6, you learned the heart of calculus. You now know precise definitions
and techniques for calculating limits, derivatives, indefinite integrals, and definite
integrals. In this chapter you will solve differential equations, which express the
rate at which a function grows. The function can represent population, money in a
bank, water in a tub, radioactive atoms, or other quantities. A slope field, shown in
the graph on this chapter's facing page, will let you solve complicated differential
equations graphically. Euler's method provides you with a way to solve them
numerically. Antiderivatives let you solve them algebraically. The experience you
gain in this chapter will equip you to make intelligent application of these calculus
concepts when they arise in your study of such fields as biology, economics, physics,
chemistry, engineering, medicine, history, and law.
OBJECTIVE Discover, on your own or with your study group, a property of exponential functions by
working a real-world problem.
Exploratory Problem Set 7-1
1. Suppose the number of dollars, D(t), in a savings account after t years is
D(t) = 500(1.06t).
D(t)
Calculate the number of dollars at t = 0 yr, t = 10 yr, and t = 20 yr.
2. For Problem 1, calculate D '(O), D'(lO), and D '(20) . What are the
units of D '(t)? Does the rate increase, decrease, or stay the same
as the amount in the account increases?
3. For the account described in Problem 1, let R(t) be the instanta- 10
neous rate of change of money in dollars per year per dollar in
the account. Calculate R(O), R(lO), and R(20) . Figure 7-1a
4. The values of R(t) you found in Problem 3, when multiplied by 100, are percentage
interest rates for the savings account. Does the percentage interest rate go up, go
down, or stay the same as the amount of money in the account increases?
5. Recall that if y is directly proportional to x, then y = kx, where k stands for a constant
(called the constant of proportionality). Show that the following property is true.
DirectProportion
Propertyof Exponential
Functions
If f is an exponential function, f(x) = a bx, where a and b are positive constants, then f'(x)
is directly proportional to f(x).
6. Just for fun, see if you can prove the converse of the property given in Problem 5. That
is, prove that if f'(x) is directly proportional to f(x), then f is an exponential function
of X.
Section7-1:DirectProportion
Property
of Exponential
Functions 309
7-2 Exponential Growth and Decay
At the beginning of Chapte r 6, you encountered a popul ation growt h problem in
which the rate of chan ge of the population, dP/dt, is dir ectly proportional to that
population. Recall from Section 3-9 that an equation such as dP/dt = kP is called a
differential equation . Findin g an equat ion for P as a function of t is called solving
the differential equation. In this section you will learn an efficient pro cedure for
solving this sort of differential equat ion.
OBJECTIVEGivena real-world situation in which the rate of change of y with respect to x is directly
proportional to y, write and solve a differential equation and use the resulting solution as
a mathematical model to make predictions and interpretations of that real-world situation.
Example 1 Population Problem: The popul ation of the little town of Scorpion Gulch is now
1000 people. The population is presently growing at abou t 5% per year . Write
a differential equation that expresses this fact . Solve it to find an equation that
expresses population as a function of time.
Solution Let P be the number of people t years after the present . The diff erential
equation is
dP The grow th rate is dP / dt. Five percent of the population
dt = 0.05P. is 0.05 times the population.
pdP = 0.05dt
Use algebra to separate the variab les on oppos ite s ides
of the differentia l eq uatio n.
f d: = f 0.05dt Integrate both sides of the differentia l equat ion.
ln IP I = 0.05t + C Do the integrating.
Exponentiate both sides of the int egrate d equation.
IP I = eo.os, . ec Exponent ial of an In on the left, produ ct of powers with
equa l bases on the right.
p = Cteo.osr ec is a positive constan t. Let C1 = ec . This is the
genera l solution.
1000 = C1 ew.os)(o) = C1 Sub stitut e O fort and 1000 for P.
:. P = 1000 eo.os, Substitute 1000 for C 1 This gives the particular
soluti on .
The differential equation above was solved by separating the variables. The
general solution represents a family of functions (Figure 7-2a), each with a
different consta nt of integration. The population of 1000 at t = 0 is called an initial
condition, or sometimes a boundary condition . The solution of a diff ere ntial that
m eets a given initial condition is called a particular solution. Figure 7-2a shows th e
particular solut ion from Example l. It also shows two other particular solutions,
with C1 = 500 and C1 = 1500.
310 Chapter
7: TheCalculus
of Growth
andDecay
P(t) A family of
functions
Initial
population
= 1500
\
1500
500
10 20
Figure 7-2a
Example 2 involves a differential equation for something other than population.
The air pressure in a car or bicycle tire can be measured in pounds per square inch
(psi). If the pressure in each of the four tires of a 3000-lb vehicle is 30 psi, the tires
would flatten enough on the bottoms to have "footprints" totaling 100 in 2 . Lower
tire pressure would cause the tires to deform more so that pressure times area
would still equal 3000 .
Example 2 Punctured Tire Problem: You run over a nail. As the air leaks out of your tire, the rate
of change of air pressure inside the tire is directly proportional to that pressure .
a. Write a differential equation that states this fact. Evaluate the proportionality
constant if the pressure was 35 psi and decreasing at 0.28 psi/min at time zero .
b. Solve the differential equation subject to the initial condition implied in step a.
c. Sketch the graph of the function. Show its behavior a long time after the tire is
punctured .
d. What will the pressure be at 10 min after the tire was punctured?
e. The car is safe to drive as long as the tire pressure is 12 psi or greater. For how
long after the puncture will the car be safe to drive?
Solutions a. Let p = no. of psi pressure.
Let t = no. of minutes since the puncture .
dp _ k Rat e of chang e of pr es sur e is dir ectly prop ortion al to
dt - p pr essur e.
- 0. 28 = k( S) Substitute for dp / dt and p. Beca use pi s decreasin g,
3 dp / dt is negativ e.
- 0.008 = k
... dt
dp
= -0 .008p
Section7-2: Exponential
Growth
andDecay 3 11
b. dp = - 0.008 dt Separat e th e variab les.
p
f dpp = f - 0.008 dt Integrate both sides .
Use C1 here in anti cipatio n of usin g th e simpl er
ln IPI = - 0.008t + C1 symbol C lat er on.
elnlpl = e(-0.008t+Ci)
1
P1 = ec1e- o.oosr
p = Ce - 0.00Bt Replace e C1 with C. Replace IP I with p because
p pressur e is a positiv e quantity.
35 35 = ce - 0008<0l ~ 35 = C Subs ti tu te th e initi al condition .
... p = 35e -0.008t Write the particul ar so lution .
c. The graph is shown in Figure 7-2b. The t-axis is a horizontal asymptote. You
should be able to sketch the graph of a decreasing exponential function such
as this one without havin g to plot it on your grapher first.
Figure 7-2b d. p = 35e - OOOS(
lO) = 32.30907 ...
The pressure will be about 32.3 psi.
e. 12 = 3 5e- o.oosr SubstitLJte 12 for p .
g = e- o.oosr
35
ln( H) = - 0.008t
1n(Hl
t = - 0.008 = 133.805 .. .
The car will be safe to drive for about 134 min, or a bit less than two and a
quarter hours.
In Problem Set 7-2, you will work more problems in which differential equations
lead to expon ential functions.
Problem Set 7-2
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Q1. Sketch : y= ex
Q2. Sketch: y = e-x
Q3. Sketch: y = ln x
Q4. Sketch : y = x2
QS. Sketch: y = x3
Q6. Sketch: y = 1 /x
Ql. Sketch: y =x
QB. Sketch : y =3
Q9. Sketch: x =4
QIO. Sketch : y =3- x
312 Chapter
7: TheCalcul
us ofGrowth
andDecay
l. Bacteria Problem : Bacteria in a lab culture (Figure 7-2c) grow in
such a way that the instantaneous rate of change of bacteria is
directly proportional to the number of bacteria present.
a. Write a differential equation that expresses the relationship.
Separate the variables and integrate the equation, solving for
the number of bacteria as a function of time.
b. Suppose that initially there are 5 million bacteria. Three
hours later, the number has grown to 7 million. Write the
particular equation that expresses the number of millions of Figure 7-2c
bacteria as a function of the number of hours.
c. Sketch the graph of bacteria versus time.
d. What will the bacteria population be one full day after the first measurement ?
e. When will the population reach 1 billion (1000 million)?
2. Nitrogen 17 Problem: When a water-cooled nuclear power
plant is operating, oxygen in the water is transmuted to
nitrogen 17. After the reactor is shut down, the radiation
from this nitrogen 17 decreases in such a way that the rate
of change in the radiation level is directly proportional to
the radiation level.
a. Write a differential equation that expresses the rate of
change of the radiation level in terms of the radiation
level. Solve the equation to find an equation that
expresses the radiation level in terms of time.
b. Suppose that when the reactor is first shut down, the
radiation level is 3 x 10 17 units . After 60 sec the level
has dropped to 5.6 x 10 13 units. Write the particular Figure 7-2d
equation.
c. Sketch the graph of radiation level versus time.
d. It is safe to enter the reactor compartment when the radiation level has dropped
to 7 x 10- 3 units. Will it be safe to enter the reactor compartment 5 min after the
reactor has been shut down? Justify your answer .
3. Chemical Reaction Problem: Calculus foeride (a rare substance) reacts in such a way
that th e rate of change in the amount of foeride left unreacted is directly proportional
to that amount.
a. Write a differential equation that expresses this relationship. Integrate it to find an
equation that expresses amount in terms of time . Use the initial conditions that the
amount is 50 mg when t = 0 min, and 30 mg when t = 20 min.
b. Sketch the graph of amount versus time.
c. How much foeride remains an hour after the reaction starts?
d. When will the amount of foeride equal 0.007 mg?
Sect
ion7-2: Exponential
Growth
andDecay 313
4. Car Trade -in Problem : Major purchases, like cars
and houses, depreciate in value. That is, as time
goes on, their value goes down . A reasonable
mathematical model for the value of an object
that depreciates assumes that the instantaneous
rate of change of the object's value is directly
proportional to the value .
a. Write a differential equation that says that
the rate of change of a car's trade -in value is
directly proportional to that trade -in value .
Integrate the equation and express the trade-in
value as a function of time.
b. Suppose you own a car whose trade-in value
is presently $4200. Three months ago its
trade-in value was $4 700. Find the particular equation that expresses the trade-in
value as a function of time since the car was worth $4200.
c. Plot the graph of trade-in value versus time. Sketch the result .
d. What will the trade-in value be a year after the time the car was worth $4 700?
e. You plan to get rid of the car when its trade-in value drops to $1200. When will
this be?
f. At the time your car was worth $4 700, it was 31 mo old. What was its trade-in value
when it was new?
g. The purchase price of the car when it was new was $16,000 . How do you explain
the difference between this number and your answer to 4f?
5. Biological Half-Life Problem: You accidentally inhale some mildly
poisonous fumes (Figure 7-2e). Twenty hours later you still feel a
bit woozy, so you go to a doctor. From blood samples, she measures
a poison concentration of 0.003 72 mg/ml and tells you to come
back in 8 hr. On the second visit, she measures a concentration of
0.00219 mg/ ml .
Let t be the number of hours that have elapsed since you first
visited the doctor and let C be the concentration of poison in your
blood (in milligrams per milliliter) . From biology, you realize that
the instantaneous rate of change of C with respect to t is directly
proportional to C.
Figure 7-2e
a. Write a differential equation that relates these two variables .
b . Solve the differential equation subject to the initial conditions specified. Express C
as a function oft.
c. The doctor says you might have had serious body damage if the poison concentra-
tion had ever been as high as 0.015 mg/ ml . Based on your mathematical model,
was the concentration ever that high? Justify your answer.
d. Plot the graph of this function. Sketch the results.
e. The biological half-life of a poison is the length of time it takes for the concentration
to drop to half of its present value . Find the biological half -life of this poison .
6. Carbon 14 Dating Problem : Carbon 14 is an isotope of carbon that is formed when
radiation from the sun strikes ordinary carbon dioxide in the atmosphere . Thus plants
3 14 Chapter
7: TheCalculus
ofGrowth
andDecay
such as trees, which get their carbon dioxide from the atmosphere, contain small
amounts of carbon 14. Once a particular part of a plant has been formed, no more
new carbon 14 is taken in by that part. The carbon 14 in that part decays slowly,
transmuting into nitrogen 14. Let P be the percentage of carbon 14 that remains in a
part of a tree that grew t years ago.
a. The instantaneous rate of change of P with respect to t is directly proportional to
P. Use this fact to write a differential equation that relates these two variables.
b. Solve the differential equation for Pin terms oft. Use the fact that the half-life of
carbon 14 is 5750 yr. That is, if P = 100 when t = 0, then P = SOwhen t = 5750.
c. The oldest living trees in the world are the bristlecone pines in the White Mountains
of California. Four thousand growth rings have been counted in the trunk of one of
these trees, meaning that the innermost ring grew 4000 years ago . What percentage
of the original carbon 14 would you expect to find remaining in this innermost ring?
d. A piece of wood claimed to have come from Noah's Ark is
found to have 48.37% of the carbon 14 remaining. It has
been suggested that the Great Flood occurred in 4004 B.C.
Is the wood old enough to have come from Noah's Ark?
Explain.
e. Plot the graph of P versus t for times from O yr through
at least 20,000 yr. Use your grapher's trace function to
demonstrate that your answers to 6c-e are correct by
showing that they lie on this graph. Sketch the results.
f. See Colin Renfrew's article "Carbon 14 and the Prehistory
of Europe," Scientific American, (October, 1971) for some
surprising results of slight inaccuracies in the carbon 14
dating technique!
7. Compound Interest Problem I: Banks compound interest on
savings continuously, meaning that the instant the interest is
earned, it starts earning more interest. Thus the instantaneous
rate at which the amount in your account changes is directly
proportional to the amount in the account. As a result, the
amount of money, M, increases at a rate proportional to the
amount of money in the account (Figure 7-2f),
FASTER!
dM = kM
dt '
where M is in dollars, t is in years, and k is a proportionality
constant.
Based on what you have learned thus far in calculus, figure
out how M varies with t . To find the value of k for a particular
savings account, realize that if $100 is invested at an interest
rate of 7% per year, then Mis increasing at a rate of $7 per
year at the instant M = 100. Once you have found a function
that expresses M in terms of t, investigate the effects of
leaving various amounts for various times at various interest The more there is, the faster
rates. For instance, which option gives you more money in the it grows.
long run: investing twice the amount of money, leaving the
money twice as long, or finding an interest rate twice as high? Figure 7-2f
Sec
tion 7-2:Exponential
Growth
andDecay 315
8. Compound Interest Problem II: If interest in a savings account is compounded at
discrete int ervals rather than continuous ly, then the amount of money, M, in the
account is
k )nt
M = Mo ( 1 +
11 ,
where M is the number of dollars at time t years after the investment was made, Mo
is the numb er of dollars invested when t = 0, k is the interest rate as a decimal, and
n is the numb er of times per year the interest is compo und ed. The ideas behind this
equation are shown in Problem 35 of Problem Set 6-8. Compare the amount , M, you
would have after a specified time if the money were compounded yearly, quarterly
(four times a year), monthly, and daily. Compare these amounts with that which you
would get if the interest were compo und ed contin u ously, as in Problem 7. Reach some
conclu sion s about the relative effec ts of higher interest rate versus more frequent
compounding of interest . See if you can show that the function from Problem 7 for
continuou s compounding is a logical consequence of taking the limit of the compoun d
interest formula in this problem as n approaches infinity.
9. Generalization Problem: In this sect ion you have worked problems in which the rate
of change of y is directly proportional to y . Solving this differential equation has
always led to an exponentia l function . Good mathematicians are quick to spot possible
generalizat ions that will shorten the problem-solving process. Prove the following
theorem .
Theorem:Converseof the DirectProportionPropertyof Exponential
Functions
If!:= ky, where k stands for a constant, th en y = Cekx.
7-3 Other Differential Equations for
Real-World Applications
In Section 7-2, you worked real-world problems in which dy /dx is directly propor -
tional toy . In this section you will work real-world problems in which the derivative
has a more comp licated property than being directly proportional to y . Some of the
resu lting functions will be exponential and others will not.
OBJECTIVE Given the relationship between a function and its rate of change, Tite a differential
equation, solve it to find an equation for the function, and use the function as a
mathematical model.
316 Chapter
7: The Calculus
ofGrowth
andDecay
Example 1 Tin Can Leakage Problem: Suppose you fill a tall (topless) tin can with water , then
pun ch a hole near the bottom with an ice pick (Figure 7-3a). The water leaks quickly
at first, then more slowly as the depth of the water decrea ses . In engineering or
physics, you will learn that the rate at which water leaks out is directly proportional
to th e square root of its depth . Suppose that at time t = 0 min, the depth is 12 cm
and dy/ dt is - 3 cm/ min .
a. Write a diff erential equation stating that th e instantan eous rate of change of
y with respect to tis dir ectly proportional to the squar e root of y. Find the
proportionality constant.
b. Solve the differential equation to find y as a function of t. Use the given infor -
mation to find the particular solution. What kind of a function is this?
c. Plot th e graph of y as a function of t. Sketch the graph. Consider the domain of t
in which the function gives reasonable answers.
Figure 7-3a d. Solve algebraically for the time at which the can becomes empty. Compare your
answer with the time it would take at the initia l rate of - 3 cm/ min.
k is the proportionality constant.
Solutions a. dy = k yl / 2 The 1/ 2 power is equivalent to the
dt squar e root.
At t = 0, y = 12 and
dy
dt = -3. !~ is negative because y is
decreasing as l increases .
.'. - 3 = k (12 112 ) ~ k = -3(12 - 1 12 )
Separa te the varia bles . It is simp ler
b. y - 112 dy = k dt
to write k inst ead of -3(12 - 1' 2 ).
f y-l f2d y = f kdt
2y 112 = kt + C
112 Substitute the initial condition
2(12 ) = k 0+C = C y = 12 when t = 0.
Do the algebr a before substituting
y =( kt + C) 2 = (k 2 t 2 + 2kCt + C2 ) fork and C.
y
y = ftt 2 - 3 t + 12
Use your pencil an d pap er to see
how to get this!
This is a quadratic function.
c. Figure 7-3b shows th e graph of this function. At time t = 8, the can becomes
empty. Beyond that time th e mod el would indicate that the can is filling back
up . Befor e t = 0, th e can was not draining. Thus the domain in which the
Figure 7-3b mathematical model gives reasonabl e answers is O ::; t :s; 8. The part of the
grap h beyond t = 8 is dotted to show th e quadratic natur e of the function .
d. 0 = l16 t 2 - 3t + 12
3 ~9 - 4 (ft) ( 12)
t = 3 =8
2( 16)
The can takes 8 min to drain, which is just twice as long as it would take at the
original rat e of - 3 cm/ min.
Sec
tion7-3:OtherDifferential
Equations
for Real-World
Applicatio
ns 317
Example 2 shows what can happen if a population is growing at a constant rate
because of one influence and decaying at another rate because of a second influence.
The "population" is water in a lake behind a dam.
Example 2 Dam Leakage Problem: A new dam is constructed across Scorpion Gulch (Figure
7-3c). Engineers want to predict the amount of water in the lake behind the dam as
a function of time. At time t = 0 days the water starts flowing in at a fixed rate F
ft 3/ hr. Unfortunately, as the water level rises, some leaks out. The leakage rate, L,
is directly proportional to the amount of water, W ft 3 , present in the lake. Thus the
instantaneous rate of change of W is equal to F - L.
Wft 3 in lake
Dam
Figure 7-3c
a. What does L equal in terms of W? Write a differential equation that expresses
dW/dt in terms of F, W, and t.
b. Solve for W in terms of t, using the initial condition W = 0 when t = 0.
c. Water is known to be flowing in at F = 5000 ft 3/ hr. Based on geological consid-
erations, the proportionality constant in the leakage equation is assumed to be
0.04 / hr. Write the equation for W, substituting these quantities .
d. Predict the amount of water after 10 hr, 20 hr, and 30 hr. After these numbers of
hours, how much water has flowed in and how much has leaked out7
e. When will the lake have 100,000 ft 3 of water?
f. Find the limit of W as t approaches infinity . State the real-world meaning of this
number.
g. Draw the graph of W versus t. Clearly show the asymptote.
Solutions a. In a problem this complicated, it helps to start by writing the definitions of
the letters being used and whether they stand for variables or constants.
W = no . of ft 3 of water in the lake (dependent variable)
t = no . of hours since water started flowing (independent variable)
F = no . of ft 3/ hr the water flows in (a constant)
L = no. of ft 3/hr the water leaks out (a variable)
Then you proceed by putting together the information asked for in the
problem.
Meanin g of "dir ectly prop orti onal." k s tand s for
L = kW th e constant of proportionality .
dW/dt =F - L dW /d t is th e instant aneou s rat e of change of W .
=F -
.'. dW /dt kW
3 18 Chapter
7: TheCalculus
ofGrowth
andDecoy
b. Separating the variables appears to be tricky. Recall that F and k are constants.
dW
F _ kW = dt Multiply by dt. Divide by F - kW.
dW 1 Integrate both sid es.
IF- kW= dt
The differential of the denominator, d(F - kW), equals -kdW . You can make
the numerator equal to - kdW by multiplying and dividing by -k .
1 I - kdW 1
-k F - kW= dt
1
- k In IF - kW 1= t + C Integral of the reciprocal function.
ln(F - kW) = -kt - kC F > kW because water comes in faster than it goes out.
eln(F-kW) = e-kt-kC Exponentiate both sides .
F - kW = e - kte - kc Do suitable algebra.
W = .!(F - e - kt e - kc)
k
W = _!(F - C1e-k 1 ) Use C1 for e- kc (general soluti on).
k
Substituting the initial condition W = 0 when t = 0,
0= k1 (F - 0
C1 e ) .
0 = i (F - C1), which implies that C 1 = F.
.. W = i(F - Fe - kt)
W = ~(l - e- kc)
k
c. Substituting 5000 for F and 0.04 for k gives
W = 12 5000 ( 1 - e- 0 -041 ) . This is the particular solution.
d. Try using your grapher's trace or table feature to find values of W. Round
to some reasonable value, such as to the nearest cubic foot. The values of F
are calculated by multiplying 5000 by t. The values of leakage are found by
subtraction.
w F L
10 41,210 50,000 8,790
20 68 ,834 100,000 31,166
30 87,351 150,000 72,649
e. Substituting 100,000 for W and using the appropriate algebra gives the
following.
100000 = 125000(1 - e- 0 -041 )
0.8 = 1 - e- o.o4r~ e- o.o..ir = 0.2
- 0.04t = ln0.2 Tak e ln of both side s.
t = 40.2359 ...
Thus it will take a bit more than 40 hr for the lake to fill up to 100,000 ft 3
Section
7-3: OtherDifferential
Equations
for Real
-WorldApplications 319
f. lim W = lim 125000(1 - e- 0041 )
t -oo t - oo
= 125000(1 - 0)
= 125000
Wis 125,000 ft 3 at most.
Note that e-o.o4 r = 1/ e 004 1 (which has the form 1 / oo as t approaches infinity)
thus approaches zero.
Figure 7-3d g. The graph is shown in Figure 7-3d.
Example 3 The lake in Example 2 starts filling with water. The actual amount of water at time
t = 10 hr is exactly 40,000 ft 3 . The flow rate is still 5000 ft 3/h.r, as predicted. Use
th.is information to find a more precise value of the leakage constant k.
Solution Substituting F = 5000 and the ordered pair (t, W) = (10, 40000) gives
40000 = 5okoo o - e- 1ok).
Th.is equation cannot be solved anal ytically for k because k appears both
algebraically (by division) and transc end entall y (as an exponent) . Fortunately,
your grapher will allow you to evaluate k as precisel y as you like. You might
first divide both sides by 5000 to make th e numbers more manageable.
8= f (1 - e - lOk)
Then us e your grapher's solve or inters ect feature to find the value of k . The
result is
0.1
k ""0.046421 ....
The value of k can also be found by plotting and tracing (Figure 7-3e). It is close
Figure 7-3e to the 0.04 assumed in Example 2.
Problem Set 7-3
DoThese
Quickly
The following problems are intended to refresh your skills . You sho uld be able to do all
ten problems in less than five minutes.
Ql. If dy /dx = ky, then y = -?-
Q2. If dy /dx = kx, then y = -?-
Q3, If dy /dx = k, then y = -?-
Q4, If dy /dx sinx, then y = -?-
=
1
Q5, If y = sin - x, th en dy /dx = -?-
Q6, ln (escosx) = -?-
Ql. eln tanx = -?-
QB. Sketch a y-graph (Figure 7-3f) if y(l) = 0. Figure 7-3f
Q9. What do es it mean for f to be integrabl e on [a, b]?
QlO. If J: v(t) dt = 17 and J: v(t) dt = 33, then JJv(t) dt = -?-
320 Chapter
7: TheCalculus
ofGrowth
andDecay
1. Sweepstakes Problem I: You win a well-known national sweep-
stakes1 Your award is an income of $100 a day for the rest of
your life! You decide to put the money into a fireproof filing
cabinet (Figure 7-3g) and let it accumulate there. But temptation
sets in, and you start spending the money at S dollars per day .
a. Let M be the number of dollars you have in the filing cabinet
Out at
and let t be the number of days you have been receiving S $/ day
the money . Assuming that the rates are continuous, write a
diff erential equation that expresses dM /d t in terms of S.
b. Your spending rate, S, is directly proportional to the amount
of money, M . Write an equation that expresses this fact, then Figure 7-3g
substitute the result into the differential equation.
c. Separate the variables and integrate the differential equation in 1b to get an
equation for Min terms oft. As the initial condition , realize that M = 0 when t = 0.
d. Suppose that each day you spend 2% of the money kept in the filing cabine t. That
is, the proportionality constant in the equation for S is 0.02. Substitute this value
into the equa tion you found in le to get M explicitly in terms oft .
e. Plot the graph of M versus t. Sketch the result.
f. After 30 days, 60 days, and 90 days, how much money will you have in the filing
cabinet? How much has come in? How much has been spent?
g. After a year, how much money is in the filing cabinet? At what rate is the amount
increasing at this time ?
h. What is the limit of M as t approaches infinity?
2. Sweepstakes Problem II: You win a well-known national sweepstakes!
Your award is an income of $100 a day for the rest of your life! You
Interest in
put th e money into a savings account at a bank (Figure 7-3h), where it at kM $/ day
earns interest at a rate directly proportional to the amount, M, which
is in the account. Assuming that the $100 rate is continuous, dM /dt
equals 100 + kM, where k is a proportionality constant. Solve this
differential equation subject to the initial condition that there was no
money in the account at t = 0 days. Find the proportionality constant
if the interest rate is 0.02% (not 2%!)per day, or roughly 7% per year.
Transform the solution so that Mis in terms oft . Use the result to
explore the way M varies with t. A graph might help . Consider such
information as how much of Mand of dM /dt comes from the $100
per day and how much comes from interest after various numbers of Fig ure 7-3 h
days. What is the limit of M as t approaches infinity?
3. Electrical Circuit Problem: When you turn on the switch in an electric
circuit (Figure 7-3i), a constant voltage (electrical "pressure"), E,
is applied instantaneously to the circuit. This voltage causes an
electrical current to begin to flow through the circuit. The current
is I = 0 amp (ampere) when the switch is turned on at time t = 0
sec. The part of this voltage that goes into overcoming the electrical - Power source
resistance of the circuit is directly proportional to the current, I .
The proportionality constant, R, is called the resistance of the
circuit. The rest of the voltage is used to get the current moving
through the circuit in the first place and varies directly with the
Figure 7-3i
Section 7-3: Other Different
ial Equationsfor Real
-WorldApplicatio
ns 32 1
ttfM5/MMtttttt 1ffllMt&M
4 &9#
instantaneous rate of change of the current with respect to time. The constant for this
proportionality, L, is called the inductance of the circuit.
a. Write a differential equat ion stating that E is the sum of the resistive voltag e and
the inductive voltage.
b. Solve this differential equation subject to the initial condition that I = 0 when t = 0.
Write the resulting equation with I as a function oft.
c. Suppose that the circuit has a resistance of 10 ohms and an indu ctance of 20
H (henries) . If the circuit is connected to a normal llO-v (volt) outlet, write the
particular equation and plot the graph. Sketch the result. Show any asymptotes .
d. Predict the current for the following times.
i. 1 sec after the switch is turned on
ii. 10 sec after the switch is turned on
iii. At a steady state, after many seconds
e. At what time, t, will the current reach 95% of its steady-state
value? Heat lost
Heat in at
R cal/s ec to room
4 . Newton's Law of Cooling Problem: When you turn on an
electric heater, such as a "burner" on a stove (Figure 7-3j),
its temperature increases rapidly at first, then more slowly,
and finally approaches a constant high temperature. As the
burner warms up, heat supplied by the electricity goes to
two places.
i. Storage in the heater materials, thus warming the
heater Figure 7-3j
ii. Losses to the room
Assume that heat is being supplied at a constant rate, R. The rate at which heat is
stored is directly proportional to the rate of change of temperature. Let T be the
number of degrees above room temperature. Let t be time in seconds. Then the storage
rate is C(dT /dt). The proportionality constant, C (calories per degree), is called the
heat capacity of the heater materials. According to Newton's law of cooling, the
rate at which heat is lost to the room is directly proportional to T. The (positive)
proportionalit y constant, h, is called the heat transfer coefficient.
a. The rate at which heat is supplied to the heater is equal to the sum of the storage
rate and the loss rate. Write a differential equation that expresses this fact.
b. Separate the variables and integrate the differential equation. Explain why the
absolute value sign is not necessary in this case. Transform the answer so that
temperature, T, is in terms of time, t . Use the initial condition that T = 0 when
t = 0.
c. Suppose that heat is supplied at a rate, R = 50 cal/sec. Assume that the heat capacity
is C = 2 cal/ degC, and that the h eat transfer coefficient is h = 0.04 (cal/sec) / degC.
Substitute these values to get Tin terms of t alone.
d. Plot the graph of T versus t. Sketch the result.
e. Predict Tat times of 10, 20 , 50, 100, and 200 sec after the heater was turned on .
f. Find the limit of T as t approaches infinity. This is called the steady-state tempera-
ture.
g. How long does it take the heater to reach 99% of its stea dy-state temperature?
322 Chapter
7: TheCalculus
of Growth
andDecoy
5. Hot Tub Problem: Figure 7-3k shows a cylindrical hot tub that
is 8 ft in diameter and 4 ft deep. At time t = 0 min, the drain
is opened and water flows out. The rate at whi ch it flows is
proportional to the square root of the depth, y feet. Because
the tub has vertical sides, th e rate is also proportional to the
square root of the volume, V cubic feet, of water left .
--
I
,...
-:::!.I \
..
~
~
8 ft
.......... ..-::. -t
.... ... . t
y4ft
I
~Li
Drain
a. Write a differential equation for the rate at which water
flows from the tub. That is, write an equation for dV /dt in
terms of V. Figure 7-3k
b. Separate the variables and integrate the differential equation you wrote in Sa.
Transform the result so that V is expressed explicitly in terms of t. Tell how V
varies with t.
c. Suppose that the tub initially (that is, when t = 0) contains 196 ft 3 of water and
that when the drain is first opened the water flows out at 28 ft 3/ min (that is, when
t = 0, dV /dt = -28) . Find the particular solution of the differential equation that
fits these initial conditions.
d. Naive thinking suggests that the tub would be empty after 7 min since it contained
196 ft 3 and the water flowed out at 28 ft 3/ min. Show that this conclusion is false.
Justify your answer.
e. Does this mathematical model predict a time when the tub is completely empty, or
does the volume, V, approach zero asymptotically? If there is a time, tell what time.
f. Draw a graph of V versus t in a suitable domain. / Burette
g. See Problem C4 in Section 7-7 to see what would happen if a hose were left
running into the hot tub while it was draining .
6. Burette Experiment: In this problem you will simulate the Hot Tub Problem in
this problem set and the Tin Can Leakage Problem in Example 1. Obtain a burette
y
(see Figure 7-31)from a chemistry lab. Fill the burette with water, then open the
stopcock so that water runs out fairly slowly. Record the level of water in the
burette at various times as it drains. Plot volume versus time on graph paper.
Does the volume seem to vary quadraticall y with time, as it did in Example 1?
Find the best -fitting quadratic function for the data. Discuss the implications of
the fact that the volume read on the burette equals zero before the depth, y, of
the water equals zero. Figure 7-31
7. Differential Equation Generalization Problem: The solutions of dy /dx = kyn are
functions with different behaviors, depending on the value of the (constant) exponent
n. If n = 1, then y varies exponentially with x. If n = 0.5, as in the Hot Tub Problem,
then y varies quadratically with x. In this problem you will explore the graphs of
various solutions of this equation.
a. Write the solution of the equation for n = 1. Let k = 1 and let the constant of
integration C = - 3. Graph the solution and sketch the graph.
b. Solve the equation for n = 0.5. Let k = 1 and C = - 3, as in 7a. Graph the solution.
c. Show that if n = - 1, then y is a square root function of x, and if n = - 2, then y is
a cube root function of x. Plot both graphs, using k = 1 and C = - 3, as in 7a.
d. Show that if n > 1, then there is a vertical asymptote at x = - C / k. Plot two graphs
that show the difference in behavior for n = 2 and for n = 3. Use k = 1 and C = - 3,
as in 7a.
Section
7-3:OtherDifferential
Equations
for Real-World
Applications 323
e. What kind of function is y when n = O? Graph this function,
using k = 1 and C = -3 , as in 7a-d . B. ......... .
::
---:11-r-:-
I-:-1:
8. Advertising Project: A soft-drink manufacturing com-
pany introduces a new product . The company's sales-
people want to predict the number of bottles per day
they will sell as a function of the number of days since
.
i,{ ;;:i:1:<
.'llr,:1,~J;j r
) ..:,---;
1.
..-;----;----;
:
c
the product was introduced. One of the parameters
will be the amount per day spent on advertising. Here
are some assumptions the salespeople make about the Figure 7-3m
sales.
The dependent variable is B bottles per day; the inde-
pendent variable is t days (Figure 7-3m).
They will spend a fixed amount, M dollars per day, on Spend
advertising . M $/da,rAdvertise .
Part of M, an amount proportional to B, maintains
present sales (Figure 7-3n).
The rate of change of B, dB/dt, is directly proportional
to the rest of M.
Advertising costs need to be $80 per day to maintain
sales of 1000 bottles per day .
Due to advance publicity, dB/dt will be 500 bottles per
day when t = 0, independent of M . Figure 7-3n
Use what you have learned in this section to find an equation for B as a function oft .
Then show the effect of spending various amounts, M, on advertising. Calculations
and graphs would be convincing. You might include such information as whether sales
will continue to go up without bound or will eventually level off. You could also make
an impression on management by assuming a certain price per bottle and by indicating
how long it will take before the product starts making a profit.
9. Water Heater Project : Suppose you have been hired by a water-heater manufacturer to
determine som e characteristics of a new line of water heaters (Figure 7-30). Specifically,
they want to know how long it will take to warm up a tank of cold water to various
temperatures, and how long it takes from the time the thermostat turns off the heat to
the time the thermostat turns on the heat again. Here are some things you learn from
the engineering and design departments.
Heat will be supplied at a constant rate of 1200 Btu (British
thermal units) per minute.
Heat will be lost to the surroundings at a rate, L, propor-
tional to the difference between the heater temperature and Heat lost to
~ surroundings
the room temperature . That is, Water at at L = Btu/ min
Tde g
L = h(T - 70),
-- Heat in at
where L is loss rate in Btu/ min, T is the water temperature, h or zero
Btu/ min
70 deg is the room temperature, and h is a proportionality
constant called the heat transfer coefficient.
the water will warm at a rate, dT /dt, proportional to (1200 - L). Figure 7-30
324 Chapter7: TheCalculus
of Growth
andDecay
the water would warm up at 3 deg/ min if there were zero losses to the surroundings.
in 10 min the heater will warm water to 96 deg from the room temperature of 70 deg.
Use this information to derive an equation that expresses temperature, T, in terms of
the number of minutes, t, since the heater was turned on. Use the equation to find
the information the manufacturer is seeking (see the beginning of this problem). For
instance, you might investigate how long it would take to warm water to 140 deg, to
160 deg, or to 180 deg. You can find out how long it takes, when the heat is off, for
the water to cool from 160 deg to, say, 15 5 deg (when the heat turns on again). You
can impress your boss by pointing out any inadequacies in the proposed design of the
heater and by suggesting which of the parameters might be changed to improve the
design .
10. Vapor Pressure Project: The vapor pressure, P, of
a liquid or a solid (Figure 7-3p) increases as the
temperature increases. The rate of change of the Vapor , with
vapor pr essur e
vapor pressure, dP/dT, is directly proportional to P of Pmm Hg
and inversely proportional to the square of the Kelvin
temperature, T. In physical chemistry you will learn Liquid
that this relationship is called the Clausius-Clapeyron
equation.
a. Write a differential equation that expresses dP/dT
in terms of P and T. lntegrate the equation, then Figure 7-3p
solve for P in terms of T.
b. The table shows the vapor pressure (millimeters of mercury, or mm Hg) of
naphthalene (moth balls, C10 H8 ) from an old edition of Lange's Handbook of
Chemistry. Use the data for 293 degK (20 deg C) and 343 degK (70 degC) to find
the two constants in the equation you wrote in 10a. You may solve the system
of simultaneous equations either in their logarithmic form or in their exponential
form, whichever is more convenient . Don't be afraid of large numbers! And don't
round them off!!
degC degK mm Hg degC de g K mm Hg
10 283 0.0 21 70 34 3 3.95
20 293 0.0 54 80 353 7.4 (melting point)
30 303 0.133 90 363 12.6
40 313 0.320 100 373 18.5
50 323 0.815 110 383 27.3
60 333 1.83 200 47 3 496.5
c. How well does your function fit the actual data? Does the same equation fit well
above the melting point? If so, give information to support your conclusion. If not,
find an equation that fits better above the melting point . Do any other types of
functions available on your grapher seem to fit the data better than the function
from the Clausius-Clapeyron equation?
d. Predict the boiling point of naphthalene, which is the temperature at which the
vapor pressure equals atmospheric pressure, or 760 mm Hg.
e. What extensions can you think of for this project?
7-3:OtherDifferential
Section Equations
for Real-World
Applications 325
I~
7-4 Graphical Solution of Differential Equations
by Using Slope Fields
The function y = e03 x is a particular solution of the differential equation dy /dx =
0.3y . The left-hand graph in Figure 7-4a shows this solution, with a tangent line
through th e point (2, e0 6 ). In the right-hand graph of this figure, the curve and most
of the tangent line have been delete d, leaving only a short segment of the tangent,
centered at the point (2, e0 6 ). This segment could have been drawn without ever
having solved the differential equation. Its slope is 0.3e 0 6 a:::0.55, the number you
get by substituting e 0 6 for y in the original differ en tial equation.
y y
4 4
2 2
y= e0.3x Tangent segme nt at x = 2
Figure 7-4a
The left-hand graph in Figure 7-4b shows what results if you draw a short segment
of slop e 0.3y at every grid point (point with integer coordinates) on the plane . The
result is called a slope field or sometimes a direction field. The right -hand graph in
Figure 7-4b shows the solution y = e03 x , from Figure 7-4a, drawn on the slope field .
The line segments show the direction the graph takes. As a result, you can draw the
graph of another particular solution just by picking a starting point and going to
the left and to the right "parallel" to the line segments . The dotted curves on the
right-han d graph of Figure 7-4b show three such graphs.
; ; f{':~t/:
y
/ / / / / /
/ / / / / / .. ' '
/ / / / / / / ::::,.......... / /
-- ---- 2
X
2
-- X
Slope field y= e0 -3 x and other
particular solutions
Figure 7-4b
Slope field s are tedious to draw by hand and are best done by grapher. Once you get
a slope field, however, it allows you to graph any particular solution without ever
326 Chapter
7: TheCalculus
of Growth
andDecoy
solving the differential equation . As you can appreciate from difficulti es you may
have had integrating the different ial equations in Section 7-3, such an approximate
solu tion met hod is welcome ' In this section you will sketch the graphs by han d.
However, you will compare some of th e graphica l solutions with exact, algebraic
soluti ons . In Section 7-5, you will learn a numerica l method for plotting such
approximate graphs on the grapher itself.
OBJECTIVE Given a slop e field for a differenti al equation, graph an approximate particular solution
b y hand and, if possibl e, confirm th e solution algebraicall y.
Example 1 Figure 7-4c shows the slop e field for the differentia l equation
dy
- ---
0.36x
dx y
.,,.,,,..,,..,,.,... ____
____
.,,.,,,..,,.,,..,,,..,,.
////;.,;,
___,,,,,,,
______ ,,
, .................
,...
////;;,.,,. ____ ,,,,,,,,
//////;;, .... -,,,,,,\\\
II/Ill///, ...,,,\\\\\\\x
I I I II I I I t-1-,1-1-+++-t-+-+-+-+-
I I I I I\\''' ~///PIii/i
\\\,,,,,...,_ ____
,,,,,,,, .,,.,.,,,.;//////
,,,,,
,, ______
........
....,,,
.,,..,,..,,/////
_______ .,,..,,..,,..,,.,,//
,....,,..,,..,,..,,.,,.,,
Figure 7-4c
a. From the differentia l equation, find the slope at the points (5, 2) and (-8, 9).
Mark th ese points on the figure . Tell why the calcul ated slopes are reason able.
b. Start at the point (0, 6) and draw a grap h that represents the p ar ticular solution
of the differential equation that contains that point . Go both to the right and
to the left. Where does the curve seem to go after it touches the x-axis? What
geomet ric figur e does the graph seem to be?
c. Start at the p oint (5, 2), from pa rt a, and draw another particular solu tion of the
differentia l equation . How is this solution related to tha t in part b?
d. Solve the differential equation algebraically. Find the particular solution that
contains the point (0, 6). Verify that th e grap h rea lly is the figure ind icated in
part b.
. dy 0.36(5)
Solutions a. At the pomt (5, 2), dx = - = - 0.9 .
2
. dy 0.36( - 8)
At the pomt (- 8, 9), dx = - = 0.32.
9
Section
7-4: Graphical
Solut s by UsingSlope Fields
ion of DifferentialEquation 327
The circled points in Figure 7-4d show slopes of about - 1 an d 0.3, which
agree with th e calculations.
b . Figure 7-4d shows th e gra ph . Start at the
boxed point (0, 6). Where the graph goes --0------- ~---------
---------- ----------
between grid points, make its slop e an
average of the slopes shown . Don't tr y ,,,.,,,..,,,..,~..----
to head for the grid points themselves!
~~~~;;:--:-:-...--
I'll// ,,,,,,..._
I I I I. I I//,...
The graph may not pass through these
i. ,,,\' ..,,, ........,;// ..;,,,;
points . The graph appears to b e an ellipse. \ ~ \ \ ' ~.~ ' .... - - .,.. ..,,.; / / I 1:"1
''\ ;'I\'' ...............
,':"'- - -:,,,., ,;:., /// ,.; /
The dotted line shows the same ellipti cal
patt ern below the x-axis. ~~~
:~::-:-~:~: -~:~.:-::~-~-~~;
c. The solution that contains the point (5, 2) is
the inner ellipse in Figure 7-4d. It is similar
(has the same proportion s) to the ellipse Figure 7-4d
describ ed in part b .
d. y dy = - 0.36x dx => f y d y = - 0.36 f x dx => 0.5y 2 = - 0.18x 2 + C
Substituting the point (0, 6) gives 0.5(36 ) = 0 + C => C = 18.
:. 0.5y 2 = - 0.18x 2 + 18 => 9x 2 + 25y 2 = 900
This is the equation of an ellipse cent ere d at the origin, as shown in part b .
Figure 7-4e shows slope fields for three simple differential equations. For each slope
field or its differential equation, three particular solutions are shown, along with
the corresponding initial conditions. Note that th e graph follows the pattern but
usually goes between lattice points rath er than through th em.
// ///
/ / / // / /
dy - 1
dx - ~~ = X ~ = COSX
(-1, 1.1), (1, 1.8), (0.5, -1.5) (1, 1.7), (2, 0.6) , (-2 , -1) (0, 2), (1, 0.5), (-2, -2.5)
Figure 7-4e
328 Chapter
7: TheCalculus
ofGrowth
andDecay
Problem Set 7 - 4
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes .
QJ. Differentiate: y = x5
Q2. Differentiate: y = 5x 25
Q3. Integrate: f x 7 dx
Q4. Integrate: f 7x dx
10
QS. Differentiate: xy =3
Q6. For Figure 7-4f, f! f(x) dx = -?-. X
Ql. Sketch the graph : y = x 2 2 7
QB. Sketch the graph: y = 2x Figure 7-4f
Q9. If g(x) = f f(x) dx, then f; f(x) dx = -?-.
. Find the slope of the line perpendicular toy= x 2 at the point (3, 9).
QJO
):_ ____.,..,,,,...,,,,,.
l. Figure 7-4g shows the slope field for the differential equation
,,
___
__
........
__ ____
.........................
____
, ........
.,,..,.,,,.,,,...,,.,.
.,,,..,...,,..,,,..,,,.,,,
dy X __
,,,,...,...,, _ ___.,...,,,,...,,,,..,,//
__
dx 2y ,,,,,,,..., __ __
,,,,,,..., ....
__..,.,...,,..,,.,,,,,,//
.,...,,,..,,,,,////
Use a photocopy of this figure to answer the following
,,,,,,,,--
,,,,,,,,, _ _,.,.,,;/////
_ ,.////Ill/
I I I \ \ \ \ \ ,, _./Ill/ff/I x
questions .
a. Show that you understand the meaning of slope field by ////////--- ,,,,,,,,
I I I I I I I I/.,.
,,,,,,,,
........
__
, \ \ \ \ \ I I I I
calculating dy/dx at the points (3, 5) and (- 5, 1) and by
1/II///,',
////;,,..,, __
__
_
__
__ ,,,,,,,,
...,,,,,,,,
showing that the results agree with the figure.
////.,,.,.,...,,.
//;.,,.,,,.,.,,. __ _ ___
____ ____
.,,,,..,,,..,,.,,,..., ,,
, ...., ........
,,, ....,,
b. Sketch the graph of the particular solution of the differen-
tial equation that contains the point (1, 2). Draw on both
sides of the y-axis. What geometric figure does the graph Figure 7-4g
seem to be?
c. Sketch the graph of the particular solution that contains the point (5, 1). Draw on
both sides of the x-axis.
d. Solve the differential equation algebraically. Find the particular solution that
contains the point (5, 1). How well does your graphical solution from lb agree with
the algebraic solution?
2. On dot paper, like that shown in Figure 7-4h, draw the slope
field for the differential equation Y.
dy = _ X
dx 2y
x
Then solve the differential equation algebraically. Find the
particular solution that contains the point (5, 1). Plot this
solution on the figure . What geometric figure is the graph?
3. a. On a photocopy of the slope field shown in Figure 7-4i,
sketch two particular solutions: one that contains the Figure 7-4h
point (3, 2), and one that contains the point (1, - 2).
Section7-4: Graphical
Solution
of Differential byUsingSlopeFields
Equations 329
!~
b. In Quadrant I, the slope is always negative and gets steeper y
as x or y increases. The slope at the point (1, 1) is about '' I I I J /
J I/ '' '' \
\
\ \ \ I
-0 .2. Make a conjecture about a differential equation that J /.,, ....'
could generate this slope field . Give evidence to support
I
I
J J I
"/,,,,.,,,,,,,
/ /.,,,
- __ '' ' ' ' \'
....' ,,,,
X
your conjecture . ,,,,
..... .....&....... .....J,--1......&......&.
.......,_ _..,,,
.....&......&......&......&......&......&......&.
/ //I
......
..../
4. Dependence on Initial Conditions Prob lem: Figure 7-4j shows \
'' \
' ' .... / J J J I
''' .... .,,
\ \ \ / J J I I I
the slope field for I \ \ \ \ / I ' I I I I
'
dy
\ \ \ \
'' / J I I I I I
dx = O.lx + 0.2y . Figure 7-4i
a. On a photocopy of this figure, draw the particular solution
/Ill/I/Ill ~ l/1111111
that contains th e point (0, 2). Show the graph on both sides /I/Ill/Ill /IIIIIIIII
-'///II/Ill IIIIIIIIII
of the y -axis . __
.,,.,,,,,,,,.,,,,,/////
.,,.,,,,,,,,,,,.,,,// IIIIIIIIII
b. Show that the particular solution containing the point (0,
____
_____ .,,.,,,,,,,.,,,,,//Ill/I/Ill
II/Ill/Ill
.,,..,,,.,,,,,,,.,,
/Ill/I/Ill
-5 ) exhibits a different behavior from that in 4a. ///Ill/Ill
,,,...,, _____ ,,,,,,,.,,.,,///1/I
.,,..,,..,.,,.,,,,,,.,,/// x
c. The solution to 4a curves upward, and that to 4b curves
downward . It seems reasonable that somewher e between
,,,,,,,
'l ' l
,,,,,,,,,
I I I I I I
_ ______ ;~
..... ~~~-=~
........
I I
~ ~~
.,,..,,.,,,.,,
,,,,,,,,,, ...., ....______ .,,.
these two solutions there is one that has a straight-line ,,,,,,,,,, ,,,, _____ _
graph . Draw this solution . Where do es th e graph cross the ''''''''''
\\\\\\\\\\ ,,,,,,, ___
,,,,,, _
....__
y-axis? ''''''''''J''''''''''
111\\\\\\\J''''''''''
\\\\\\\\\\ J \\\\\\\'''
I I I I I I I \ \ \ '/" \ \ \ \ \ \ \ \ \ \
Figure 7-4j
5. On a photocopy of the slop e field shown in Figure 7-4k, draw the particular solutions
that contain the points (0, 1), (3, 4), (0, - 4), and (-8 , - 2).
6. On a photocopy of the slope field shown in Figure 7-41, draw the particular solutions
that contain the points (0, 4), (0, 8), (0, 10), and (0, 15). How does the change from
(0, 8) to (0, 10) for an initial condition affect the graph?
y ,.,,_,,,,,,_..,,, ~ , ..../,,/_,,,
II/JI/Ill/.,,,,,.,,.,,..,,. ____ ,
II/Ill////,,,.,,,,,,,,,,,,,.,,..,,. __ -;r,,/_,,,,,, /,,/_,,,,,,_
_____.,,,..,,..,,.,,,,,
////,'///// ////////// .,,,,-/ ....,.,,_., /,,.,,,,-/ .... ,
,,, .....____.,,.,,, /II/Ill/II .,,....,.,,,,, .../-, ...,_.,,,,.,,,,
,_,_,_,_,,_
/
,,,,,,
,,,,,,..., ___ _ /Ill/Ill/I
__ _ /II/Ill/II
- ,-/-,-/-,
,,.,,.,,,,,/-, / ..., ....,,,,,,,,,,
,,,,,,..., ___ -'/I/Ill/II ,,/ _,,,,.,, ..., /,,,.,./_,...,,,
,,,,,, ....___ .;,;///Ill/I _ .,,,,,,
.,,.,_, , .,./_.,,
- /_,,, ,,,,....,_/,,
,,.,....,,,,/_,,
.... ,_
___ .,.,,.,,,,,,,,///11
............................. __ .,,,,.,,.,,.,,.,,,,,,,,x .,,_,, _.,,.,,_,, ,,,,_,,_,,,,_,
.J. -L-L...L-L-L...L-L....L...L ....L-J.....L...,1.....L....L....L....L-J.-J.._
---,.S,,,,, _,, ....../ ......,,
,, ...._..,,,,_,,, //,,, ........//-
/////;..---- ,,,,,,,
___ ,,,,,,,,,, ............ , ...._..,,/.,._,,,
;r/_,
;'// ......,,
... ,, ... _.,,,,
.... _.,,
/////,,,,.,,.
,,,,,,,,,, _ ,,//_,_,,, ....,,//_,_,,,, _
,,,.,,.,,.,____..., ,,,,,,,,,,
ll//;;;---
///_,_,,,,
,,,,__,,,,, ... ,,,,/,,,,_,,,,
1///.;,---
,,,,,,,,,, ... ... ,,'///_, __,,
,,,,,,,,,, _,,,,,,, ........ _.,,,,,,///,,,-
,,,,,,,
,,,, ....,, ........
....__
_ ___ ,,,,,,/////
''''''''''}''''''''''
''''''''''J''''' ...___
''''''''''T'''' ____ .,.,_ -I.....L....L....1
....L....1....L....1--1....1 .:.:.:.:
..::..::~
5
~~~-'
Figure 7-4k
Figure 7-4 1
7. Rabbit Population Overcrowding Prob lem: 1n the population problems of Section
7-2, the rate of change of population is proportional to the population. 1n th e real
world, overcrowdin g limits the size of the population . One mathematical model, the
330 Chapter 7: The Calculus
of Growth andDecoy
logistic equation, says that dP/dt is proportional to the product of the population
and a constant minus the population. Suppose that rabbits are introduced to a small
uninhabited island in the Pacific. Naturalists find that the differential equation for
population growth is
dP
dt = 0.038P(l0.5 - P),
~
where P is in hundreds of rabbits and tis in months. Figure 7-4m shows the slope
field.
a. Suppose that 200 rabbits arrive at time t = 0. On a
l
t' It I I I It, It I I I I I I I
photocopy of Figure 7-4m, graph the particular solution. I I I I 11 I I I I
11 I I I I I I I I
I
I
I
I
I
I
I
I
I
I
I I I I I
I I I I I
I I I I I I I I I I I I I I I I I I I I
b . Draw another particular solution if the 200 rabbits had I I I I I I I\ I I I I I I I I\ I I\
\\\\\\\\\\\\\\\\\\\\
been introduced at time t = 4. What would be the differ- \\ ''' ''''' ' '''' '''''
ences and the similarities in the population growth? ''''' ' ' ' ''''''''''''
,,,,,,,,,,,,,,,,,,,,
c. Draw a third particular solution if 1800 rabbits had been
;;;;;;;;;;;;;;;;;;;;
introduced at time t = 0. With this initial condition, what is ////////////////////
////////////////////
the major difference in population growth? What similarity ////////////////////
////////////////////
does this scenario have to those in 7a and b7 ////////////////////
;;;;;;;;;;;;;;;;;;;;
,,,,,,,,,,,,,,,_,_,, t
d. Think of a real-world reason to explain the horizon- ~~~ ~ ~~~~1a~~~~~~~~~~
tal asymptote each graph approaches. Where does this
asymptote appear in the differential equation? Figure 7-4m
8. Terminal Velocity Problem: A sky diver jumps from an
airplane. During the free-fall stage, her speed increases at
the acceleration of gravity, about 32.16 (ft/ sec)/ sec. But wind
resistance causes a force that reduces the acceleration. The
t~
~, \\\' \\\\\\\\\\\\\\\
,,,,,,,,,,,,,
resistance force is proportional to the square of the velocity. \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \
Assume that the constant of proportionality is 0.0015, so that k,,,,,,,,,,,,,,,,,,,,
f ''' ''' ''''''''' ''' ''
dv
dt = 32.16 - 0.0015v 2 ,
where v is in feet per second and t is in seconds . The slope
!--------------------
--------------------
;;;;;;;;;;;;;;;;;;;;
( ////////////////////
[ ;//Ill/I
lQQr, I I I I I I I I I I
II Ill/I/Ill
I I I I I I I I I I
I
r. 11111111111111111111
field for this differential equation is shown in Figure 7-4n. r. 1 I I I I I I I I I I I I I I I I I I I
r.
t
I I I I I I I I I I
I I I I I I I I I I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
a. What does the slope appear to be at the point (5, 120)? t I I I I I I I I I I I I I I I I I I I I
t I I I I I I I I I I I I I I I I I I I I
What does it actually equal? Explain any discrepancy t I I I I I I I I I I I I f I I I I I I I
tt , , , , , , , , , , , , , , , , , , , , t
I I I I I I I I I I I I f f f f ff f f
between your two answers.
, 1 1 1 1 1 1 1 1 '1b'1 1 1 1 1 1 1 1 . ..
b. The diver starts at time t = 0 with zero initial velocity.
On a photocopy of Figure 7-4n, ske tch her velocity as a Figure 7-4n
function of time.
c. The velocity approaches an asymptote . What does this terminal velocity appear
to equa l? About how long does it take her to fall until she is essentially at this
velocity?
d. A second diver started 5 sec later with zero initial velocity. Sketch the velocity-time
graph. What similariti es doe s this graph have to the graph you sketched in Sb?
e. Suppose that the plane is going down steeply as a third diver jumps, giving him
an initial downward velocit y of 180 ft/sec. Sketch this diver's velocity-time graph.
How is it different from the graphs you sketched in 8b and d?
Section
7-4:Graphical
Solution
of Differential byUsingSlopeFields
Equations 331
f. The mathematical models for free fall in this problem and for population in
Problem 7 have some similarities. Write a paragraph that discusses the similarities
and the differences. Do you find it remarkable that two different phenomena have
simi lar mathematica l models?
9. Escape Velocity Prob lem : If a spaceship has a high enough initial velocity , it will escape
the earth's gravity and be free to go elsewhere . Otherwise it will stop and fall back to
earth.
a. By Newton's third law of motion, the force, F, on the
spaceship equals its mass, m, times the acceleration, a.
By his law of gravitation, Fis also equal to mg /r 2 , where
g is the gravitational constant and r is the distance from
the center of the earth to the spaceship . Give reasons
for each step in the following transformations.
mg
ma=- r2
dv g
dt r2
dv dr g
dr dt = r 2
dv g
- V =-
dr r2
dv = ...fl_
dr r 2v
b . If r is in earth-radii (1 earth-radius= 6380 km) and vis in
kilometers per second, then
,,------------------
V
,,------------------
dv - 62.44 ,,------------------
,, __________________
,,------------------
dr
The sign is negative becaus e gravity acts opposite to the
,,------------------
,,------------------
direction of motion . Figure 7-40 shows the slope field
,,------------------
,,------------------
10
,,,-----------------
,,,-----------------
for this differential equation . Confirm that the differen- ,,,-----------------
,,,-----------------
tial equat ion produces the slopes shown at the points ,,,-----------------
,,,,----------------
1 ,,, _______________ _
(r, v) = (5, 2), (1, 10), and (10, 4). I,,,,,, ____________ _
I\\\,,,, ___________ _
c. If the spaceship starts at earth's surface (r = 1) with an I I\\\,,,,, _________ _
I I I 1\\\,,,,,,,------ r
initial velocity of v = 10 km/ sec, it will not escape earth's
10
gravity. On a photocopy of Figure 7-40, sketch this partic-
ular solut ion . About how far from earth's surface does th e Figure 7-40
ship stop and start falling back?
d. Show on your photocopied figure that if the spaceship starts from the earth's
surface with an initial velocity of 12 km / sec, it v.1:ill escape the earth's gravity . About
how fast will it be going when it is far from earth ?
e. If the spaceship starts from earth's surface with an initial velocity of 18 km/ sec,
what will its velocity approach far from earth? Does it lose as much speed starting
at 18 km/sec as it does starting at 12 km / sec? How do you explain this observation?
f. Show that the spaceship will escape from earth's gravity if it starts with an initial
velocity of 10 km/ sec from a space platform in orbit that is 1 earth-radius above
the earth's surface (that is, r = 2).
332 Chapter
7: TheCalculus
ofGrowth
andDecay
10. Slope Fields on the Grapher: Generate on your grapher the slope
~,y \\\\\\\\\\\\\\
, ' ' ' ' ' ' ' ' ' ' ' ' '
field for ..............................................................
dy ,.,,,..,,,.,,,.,,,..,,,.,.,,,.,.,,,..,,,.,
dx = 0.Sy(l - 0.15y ), ,s, ,, ... .,. .,. ,,. .,. ,, ,, , ,,. .,. ., .,
t..,,,,,,,.,,,,,,,,,,,,;;;;
as shown in Figure 7-4p. If your grapher does not have a built-in t.;;.,.,,,,,;.,;.,;,.,
t.,,,.,.,.,.,,,,,,,.,,,,,,.,.,.,,.,
program to generate slope fields, obtain one or write one. Save ~ .,..,. .,. .,..........., , , ,, .,. .,. .,..~
~
this program to use in Section 7-5.
Fig ure 7-4p
7-5 Numerical Solution of Differential Equations
by Using Euler's Method
In Section 7-4, you sketched approximate solutions of differential equations, using
their slope fields . In this section you will learn a numerical method for calculating
approximate y-values for a particu lar solution, and you'll plot the points either
by hand or on the grapher. This method is called Euler's method, after Swiss
mathematician Leonhard Euler (1707-1783). (Euler is pronounced "oi' -ler.")
OBJECTIVE Given a differential equation and its slope field, calculate points on the graph iterati vely
by starting at one point and finding the next point by following the slope for a given
x-distance .
Euler's method for solving a differential equation numerically is based on the fact
that a differentiable function has local linearity at any given point. For instance, if
dy Jdx = cos xy, you can calculate the slope at any point (x, y) and follow the linear
function to ano~her point 6.x units away. If 6.x is small, the new point will be close
to the actual point on the graph . Figure 7-Sa illustrates the procedure.
Y Start here. Find the slop e.
!
.;~-~~/ A
/
Follow the slope to
get here .
Repea t.
Go back this
way, too.
KV '-;,._
---,
"'-V":
1
~ X
Figure 7-50
Section
7-5:Numerical byUsingEuler'sMethod
Solutionof DifferentialEquations 333
Start at a given point (x, y) on the graph .
Calculate the slope at this point by using the differential equation .
For a given value of 6.x, calculate the value of dy, specifically (dy /dx) (6.x), which
is the change in y along the linear graph .
Add 6.x and dy to the previous values of x and y to get a new point, (x, y) .
Repeat the process, starting at the new point (x, y).
As the figure shows, you can go in both directions from the starting point, as long
as it makes sense in the context of the problem to do so.
In the following problem set you will work on your own or with your study group to
learn the details of how to use Euler's method.
Problem Set 7-5
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
QJ. If dy /dx is directly proportional toy, then dy /dx = -?- .
Q2. If dy /dx = 3y, then the general solution for y is -7- .
QJ. If dy /dx = 0.1 xy, what is the slope of the slope-field line at the point (6, 8)?
Q4. If y = Ce0 2x and y = 100 when x = 0, then C = -?-. y
QS. f dv / (1 - v) = _7_
Q6. (d/dx) (sec x) = -?-
l
Ql. Find f'(x) if f(x) = Ji'(3t + 5)4 dt. X
QB
. Sketch the graph of y ' for Figure 7-5b.
Q9. Differentiate implicitly: x 3 y 5 = x + y.
QJO. If limx- 4 f(x) = f(4), then f is-?- at x = 4. Figure 7-5b
l. How Euler's Method Works: Figure 7-5c shows the slope field for the differential
equation
dy X
dx 2y
a. Start at the point (0, 3), calculate the slope at that point, then assume that the graph
is linear between x = 0 and x = 0.5. What will the value of y be at x = 0.5?
b. Find the slope at the point (0.5, y), in Problem la, and use it to find an approximate
value of y at x = 1, assuming that the graph is linear between x = 0.5 and x = l.
c. Repeat the computations you did for la and b to make a table of values, as shown
on the next page.
d. On a photocopy of Figure 7-5c, plot the y-values you calculated in le by using
Euler's method. For which values of x do the Euler's method y-values seem to
follow the slope field? For which values of x is this numerical solution clearly
incorrect?
334 Chapter
7:TheCalculus
ofGrowth
andDecay
X y slope dy
0 3 0 0
y
0.5 3 - 0.083 3 ... - 0.0416 .. . 4 .. - - -~
2.9583 . ..
.................' ' \- '\
1
1.5
2 .... '
....;._ ~ - \ \ \
2.5
3
'- .. , \ \ \ . \ \
x
3.5
4 ~ -- / / l I
4.5 .-, . .... /
/ /
5
5.5
6 -4 _,. / / / /
6.5
7 Figure 7-5c
2. Numerical Program for Euler's Method: Obtain or write a program for computing
y-values by Euler's method. The program should allow you to enter the differential
equation, say as y 1 , in terms of both x and y. Then you should be able to input the
initial condition (x, y) and the value of .6.x, such as the point (0, 3) and the number 0.5
in Problem 1, respectively . Test your program by u sing it to calculate the values in the
table given in Problem le .
3. Accuracy of Euler's Method: In this problem you will use your program from Problem 2
to solve the differential equation
dy X
dx - 2y'
from Problem 1, with different values of .6.x, and you'll compare your results with the
exact solution.
a. The curve in Quadrant IV of Figure 7-5c shows the exact particular solution of the
differential equation that contains the point (0, - 4). Solve the differential equation
algebraically. Use your result to calculate y when x = 5 and show that this point is
on the graph .
b . Use your program from Problem 2 to calculate values of y by Euler's method,
starting at the point (0, - 4), using .6.x = 0.5. Record in a table the y-values for
x = 0, 1, 2, 3, 4, 5, 6, and 7.
c. On a photocopy of Figure 7-5c, plot the values from the table you made in 3b. Write
some comments on how well Euler's method fits the exact solution.
d. Repeat 3b and c, using .6.x = 0.1. Again, record only the calculated y-values for
x = 0, 1, 2, 3, 4, 5, 6, and 7. Comment on the relative accuracy of Euler's method as
smaller values of .6.x are used .
e. Figure out a time-efficient way to calculate y when x = 5, using Euler's method with
.6.x = 0.01. How closely does this value match the exact value you calculated in 3a?
4. Graphical Program for Eu ler's Method: Obtain or write a grapher program for plotting
a particular solution of a differential equation by Euler's method. You can adapt
your program from Problem 2 if you like. The differential equation can be entered as
Sect
ion75: NumericalSolutionof Differential byUsingEuler'sMethod
Equations 335
y 1 . The input should include the initial point and the value of t!,.x. As each point is
calculated the grapher should draw a line segment to it from the previous point. The
program should allow the graph to be plotted to the right or to the left of the initial
point, depending on the sign of t!,.x.The program should work in conjunction with the
slope-field program of Section 7-4 so that the solution can be superimposed on the
corresponding slope field. Test the program by using the information in Problem 3c,
and show that the graph resembles that in Figure 7-5c.
5. Figure 7-5d shows the slope field for the differential equation I
I
I
I
I
I
I
I
I I /
I I /
Y.
' \
... \
I
\
I
\
I
\
I
I
I
I
...'....'\
I I I I I /,. \ \ \
dy I I I I / ; .,,.
..... \
\ \ \ \
dx = -0.2xy. I//;.,,..,,. - - ...',, 'x'
a. Use your grapher programs to plot the slope field and the
,,,, ........._ - ,.,. .,,. / / / I
particular solution that contains the point (3, 2). Sketch the
\
\
\
\ \\ ' ''"' ...
\
,,,. ;
,. /
/
I
/
I
I
I
I
I
I
I
\ I \ \
\
\ ...\ / I I I I I I
solution on a photocopy of Figure 7-5d. I I I \ \ \' / I I I I I I
b. Plot the particular solution that contains the point (1, - 2).
Sketch the solution on the photocopy . Figure 7-5d
6. Figure 7-5e shows the slope field for the differential equation y,1111111,,,,.,.,,..,.
I I I I I I / / / / .,. .,,. .,,. .,,.
dy I I I / / / / ; ;
__
.,,. ,
......,,,,. _
.., -
dx = - 0.l X + 0.2 y .
/////,;'.,,..,,.,.,.
/ / /
5";;,.,,. ., ; ......
.,,.,.,..,,.-- - - - ...,-
______
a. Use your grapher programs to plot the slope field and the ., ......- - - -
; .,,.______
.,,..,,. - ....,,,
....' .....
............
particular solution that contains the point (0, 2). Sketch the , ______ ,,,,,,,
solution on a photocopy of Figure 7-5e.
b. Plot the particular solution that contains the point (0, 4). Sketch
the solution on the photocopy. Figure 7-5e
c. The solution you plotted in 6a curves downward, and that in 6b curves upward. It
seems reasonable that somewhere between these two solutions there is one that
has a straight-line graph. By experimenting on your grapher, find this particular
solution . Record the initial point you used and sketch the solution on th e photocopy.
7. U.S. Population Project: The following table shows the U.S. population (in millions) from
1940 through 1990. In this problem you will use this data to make a mathematical
model for predicting the population in future years and for seeing how far back the
model fits for previous years .
a. For the years 1950, 1960, 1970, and 1980, find symmetric difference quotients,
t!,.P/M, where Pis population in millions and tis time in years since 1940. (Why
can't you do this for 1940 and 19907)
year population
1940 131.7
1950 151.4
1960 179. 3
1970 20 3.2
1980 226.5
1990 248.7
b. For each year given in 7a, find t!,.P/1!,.t
as a fraction of P. That is, find (t!,.P/t!,.t)/P.
c. It is reasonab le to assume that the rate of growth of a population in a fixed region
such as the United States (as a fraction of the size of that popu lation) is some
336 Chapter
7: TheCalculus
ofGrowth
and Decay
function of the population. For instance, when the population gets too large, its
growth rate slows because of overcrowding. Find the function (linear, logarithmic,
exponential, or power) that best fits the values of (6.P/M) /P as a function of P.
Justify your answer.
d. Assume that dP/dt obeys the same equation as 6.P/M. Write a differential equation
based on your answer to 7c. Transform the equation so that dP/dt is by itself on one
side. The result is called the logistic equation. The word logistics usually refers to the
art of planning and coordinating the details of an operation, such as a military campaign,
but originates from the Greek word logistikos, meaning "reckoning" or "reason."
e. Plot a slope field. Use -50 .:,;t :,; 100 and O .:,;P :,; 500. Print p
the slope field or sketch it on a photocopy of Figure 7-5f.
500
f. Make a table of population predicted by Euler's method
for each 10 yr from t = - 50 through t = 100. Use as an
initial condition the population in 1940 (t = 0). Use steps
of M = 1 yr. Plot the points on the graph you drew in 7e
and connect them with a smooth curve.
. t
g. According to this mathematical model, what will be the
100
ultimate population of the United States? How does this
number appear in the differential equation and on the figure 7-51
slope field ?
h. Plot the populations for the six given years on the graph you drew in 7e. Does the
population really seem to follow the solution by Euler's method?
i. Write a paragraph describing how well the predicted populations in 7 f agree with
the actual populations from 1940 through 1990.
j. Consult some reference material to find the results of censuses dating back through
1900. How well do your predicted values compare with the actual ones? How can
you explain any large discrepancies between predicted and actual values?
k. Suppose that in the year 2010, 200 million people immigrate to the United States.
Predict the population for the next 40 yr. What does the logistic-equation model
say about the population growth under this condition ?
8. Algebraic Solution of the Logistic Equation: It is possible to solve the logistic differential
equation like that in Problem 7 algebraically. Suppose that
dy
dx = 3y(10 - y).
Separating the variables and integrating gives
f y(lOl - y) dy =3 f dx.
In this problem you will learn how to do the integration on the left side. Then you will
apply what you have learned so that you can algebraically solve the logistic equation
in Problem 7.
a. The fraction in the integral on the left can be split into partial fractions like this:
----
1 A
= - + --
B
y(lO - y) y 10 - y'
where A and B stand for constants. Using suitable algebra, find the values of A
andB.
Section
7-5:Numerical
Solution
of Differential byUsingEuler'sMethod
Equations 337
b. Int egrate the differential equation . Show that the integrated equation can be
transform ed into
10
y = 1 + ke - 30x '
where k is a constant related to the constant of integration.
c. Solve the logistic equation from Problem 7d algebraically . Transform the answer so
that population is in terms of time. Use the initial condition that P = 131. 7 in 1940
(when t = 0) to evaluat e k.
d. Use the algebraic solution you found in Sc to predict the population in 1950, 1960,
1970, 1980, and 1990 . How well do the approximate solutions found by Euler's
method in Problem 7 f compare with these exact solutions? How well do the exact
solutions compare with the actual population in these years? Write a paragraph
that describes your observations about how well diff erent mathematical models
agree with each other and about how well they fit data from the real world .
9. Journal Problem: Update your journal with what you've learned since the last entry .
Include such things as those listed here.
The one most important thing you have learned since your last journal entry
How slope fields and numerical methods can be used to solve differential equations
without finding an algebraic solution
How much faith you would put into a computer -generated prediction of the U.S.
population for the year 2050
What you now better understand about differential equations
Any technique or idea about differential equations that you're still unclear about
7-6 Predator-Prey Population Problems
If a population, such as anin1als or people, has plenty of food and plenty of room,
the population tends to grow exponentially with time . You have seen that such
exponentia l growth is a result of the fact that the rate of growth is proportional to
the size of the population. The more people there are, the more babies are born per
year. In Section 7-5, you used the logistic equation to model population growth in
an environment where overcrowding limits the ultimate size of the population.
In this section you will explore the effects on the populations of two species in
an environment where one of the species eats the other as its food supply . If the
predator population is small, the prey population grows . Then, as a result of the
in creased food supply, th e predator population starts to grow and can eventually
become so large that it reduces the prey population. Then the predator population
will decrease because there is less food, and so it goes' The models you vvillus e for
foxes and rabbits in the following problem set also apply to such populations as
bats and mosquitoes, cattle and grass, and sharks and other fish .
338 Chapter
7:TheCalcu
lusofGrowth
andDecay
OBJECTIV
E Use slope fields to solve problems of popu lation growth in an environm ent where one
popul ation relies on another population for its food supply.
Problem Set 7-6
DoTheseQuickly
The following problems are intended to refr esh your skills . You should be able to do all
ten problems in less than five minutes .
Q1. J:f (x ) dx = lim I f(c)!::,,,x is a (brief) statem ent of th e-?- .
s: n-oo
Q2. f(x) dx = g(b ) - g(a ) is a (brief) stat ement of the -? - .
Q3. Jf(x)dx = g(x) if and only if f(x) = g ' (x) is a statement of the - ?- .
Q4. " . . . then there is a point cin (a, b) such that f(c) = k" is the conclusion of - ?- .
QS. " . .. then there is a point c in (a, b) such that f '(c ) = O" is the conclusion of- ?- .
Q6. " . .. then there is a point c in (a, b) such that f'(c) = f(b~ - f(a ) " is the conclusion of
- a
-?-
Ql. " . . . then f '(x) = g '(h(x)) h'(x)" is the conclusion of -7- .
QB. f (x ) = cosx + C is the - ?- solution of a differ ential equation.
Q9. f(x) = cos x + 5 is a(n) -7- solution of a differential equation.
Q10. f(O) = 6 is a(n) _ 7_ condition for th e differential equation in Q9.
Ona Nyland moves to an uninhabited island . Being lonely for compan y, she imports som e
pet rabbits. The rabbits multiply and become a nuisance! So she imports some foxes to
control the rabbit population . In Problems 1-1 5, you will investigate the populations of
foxes and rabbits, based upon assumptions about the way the animals grow and interact
with one another.
1. Let R be th_enumber of hundreds of rabbits and let F b e the number of foxes at
any given time, t. If there were no foxes, the rabbit population would grow at a rate
proportional to the population . That is, dR /dt would equal k 1R, where k 1 is a positive
constant . Show that the rabbit population would grow exponentially with time under
this condition.
2. If there were no rabbits for the foxes to eat, the fox population would decrease at a
rate proportional to the popu lation. That is, dF /dt would equal - k 2 F, where k 2 is a
positive constant . Show that th e fox population would decrease exponentially under
this condition.
3. Assume that foxes eat rabbits at a rate proportional to the number of encounters
between foxes and rabbits. This rate is proportional to the product of the number
of rabbits and foxes . (If there ar e twice as man y rabbits, th ere are twice as many
encounters and vice versa.) Thus the rabbit population decreases at a rate k 3 RF in
addition to increasing at k 1 R. The fox population increas es at a rate k 4 RF as well as
decreasing at k 2 F. Write diff erential equations for dR /dt and for dF /dt under these
conditions.
Section7-6: Predator-Pr
eyPopula
tionProblems 339
4. Use the chain rule to write a differential equation for dF/dR.
What happens to t?
FIll/,... ________ ,,,,,,,
5. Assume that the four constants in the differential equation Ill/,... _______ ,,,,,,,,
are such that ll//;---------------
III// __________ ,,,,,
dF -F + 0.025RF 50 , , , //----- ...,, ....,,,,,,
1111;-----,,,,,,,,,,
dR R - 0.04RF . 1111/-------,,,,,,,,
fl(/// _____ ,,,,,,,,,
I I 111/,...-,,,,\\\\\\\\
If R = 70 and F = 15, calculate dF/dR. f f f I f f f - f ff I f f I i f f f f
6. Figure 7-6a shows the slope field for this differential equation. ,,,,,-------------
ff\\\,-----/////////
........
,,,-----------------
Show the initial condition given in Problem 5. Then show the ,~------------------ R
relative populations of rabbits and foxes as time progresses.
so 100
How do you tell from the differential equations you wrote for Figure 7-60
Problem 3 whether to start going to the right or to the left?
7. How would you describe the behavior of the rabbit and fox populations?
8. Is there a fixed point at which both the rabbit and the fox populations do not change?
Explain .
9. The logistic equation of Section 7-5 shows that, because of
F
,,,.,,.,..._______________
overcrowding, the rate of change of population is decreased ll//;---------------
111,,... ______________ _
by an amount proportional to the square of the population .
50
11///---------------
I//;,--------------
Assume that fII I,,,.,,.
I//,,,. ______________
_______ ,,,,,,, _
dR 2
dt =R - 0.04RF - O.OlR .
I I//;------,,,,,,,,,
I I 11/-------,,,,,,,,
I If/// _____ ,,,,,,,,,
Calculate dF/dR (not dR /dt!) at R = 70 and F = 15 under this ',,,,,,_,,,,,,,,_,
I I 11,--
\ ,,,, ____
I I .... ,,,,,,,,,
.,..,,,,,,,, __ ...
........
condition . ,,, _________ ,.,../1,,,- R
~~~~~~~~~~~~~~~~~~~
10. The slope field in Figure 7-6b is for dF/dR, which you calcu-
so 100
lated in Problem 9. Use the initial condition in Problem 9 to Figure 7-6b
sketch the predicted populations.
11. How does the graph in Problem 10 differ from that in Problem 6? How does overcrowd-
ing by rabbits affect the ultimate rabbit population7 The ultimate fox population?
12. Ona seeks to reduce the rabbit population by allowing hunters F////, __________ , __, ....
to come to the island. She allows the hunters to take 1000
////.,..
////,,------------ ___________,, __ ........
~ ,,,,
//// ________,, ____
....
_______________ __
rabbits per unit of time, so dR /dt is decreased by an additional ////.,.. ____________ ,,_
10. Calculate dF/dR at the point (R,F) = (70, 15) under these
////.,..
////.,..
__________ ,, ___
______________ _
////.,.. _______________
conditions.
13. The slope field in Figure 7-6c is for the differential equation II///,--------------
111, .... ---------------
/ll//
/lll/1/--,,,, .... -------------- ........................ ....
dF - F + 0.025RF /1,,, _____ .,..,,,,,----R
- ~~~~~~~~~~~~~~~~~~~ ...-
dR R - 0.04RF - O.OlR 2 10.
- so 100
Trace the predicted populations under these conditions, Figure 7-6c
starting at the point (70, 15).
14. Describe what happens to the populations of rabbits and foxes under these conditions.
15. Worried about the fate of the foxes in Problem 13, Ona imports 15 more of them.
Starting at the point (70, 30), trace the populations. According to this mathematical
model, what is the effect of importing more foxes? Surprising?!
340 Chapter
7:TheCalculus
ofGrowthand Decay
7-7 Chapter Review and Test
In this chapter you have seen that by knowing the rate at which a population changes,
you can write an equation for the derivative of the population. This differential
equation can be solved numerically by Euler's method, graphically by slope field, or
exactly by algebraic integration.
The Review Problems below are numbered according to the sections of this chapter.
The Concepts Problems allow you to apply your knowledge to new situations. The
Chapter Test is more like a typical classroom test your instructor might give you.
Review Problems
RO. Update your journal with what you've learned since the last entry. Include such things
as those listed here.
The one most important thing you have learned in studying Chapter 7
Which boxes you have been working on in the "define, understand, do, apply" table
The proportion property of exponential functions and their derivatives, and this
property's converse
The fact that a function equation can be found from the rate of change of the
function
How differential equations can be solved graphically and numerically
Any ideas about calculus that you're still unclear about
Rl. Punctured Tire Problem: You run over a nail! The pressure, P(t) pounds per square
inch (psi), of the air remaining in your tire is given by
1
P(t) = 35(0.98 ),
where tis the number of seconds since the tire was punctured. Calculate P(O),P(lO),
and P(20). Show by example that although P'(t) decreases as t increases, the ratio
P'(t) /P(t) stays constant. Prove in general that P'(t) /P(t) is constant.
R2. Ramjet Problem: A ramjet (Figure 7-7a) is a relatively
simple jet engine. The faster the plane goes, the more air Air is rammed in. Jet out
is "rammed" into the engine, and thus the more power the
engine generates . Assume that the rate at which the plane's
speed changes is directly proportional to the speed.
a. Write a differential equation that expresses the assump- Figure 7-7a
tion above.
b. Solve the differential equation . Show the integration step and describe what
happens to the absolute value sign.
C.Evaluate the constants in the equation if the plane is going 400 mi/ hr at time t = 0
sec and 500 mi/hr at time t = 40 sec.
d. When will the plane reach the speed of sound, 750 mi/ hr?
R3. a. Find the general solution of the differential equation dy /dx = 6y 112 .
b. Find the particular solution of the equation in R3a that contains the point (3, 25).
C. Plot the graph of the particular solution you found in R3b. Sketch the result.
Section
7-7:Chapter
Review
andTest 34 1
d. Find dy/dx for this differential equation when x = 2. Show on your graph that
your answer is reasonable.
e. Memory Retention Problem : Paula Tickle starts her campaign for election to the
senate. She meets people at a rate of about 100 per day, and she tries to remember
as many names as possible . She finds that after seven full days, she remembers
names of 600 of the 700 people she met. Assume that the rate of change of the
number of names she remembers, dN /dt, equals 100 minus an amount that is
directly proportional to N.
i. Write a differential equation that expresses the assumption above, and solve
the equation subject to the initial condition that she knew no names when
t = 0.
ii. How many names should Miss Tickle remember after 30 days?
iii. Does your mathematical model predict that her brain will "saturate" after a
long time, or does it predict that she can remember unlimited numbers of
names?
iv. After how many days of campaigning will Paula be able
to remember the names of only 30 of the people she y
meets that day7 -,//////////////////
--;/////////////////
---/////////////////
R4. Figure 7-7b shows the slope field for ___
----////////////////
, ////////////////
,------/////////////
dy = - 20 + 0.05y .
dx xy ,-------------------
,-------------------
a. Calculate the slope at the points (2, 5) and (10, 16). Show
,-------------------
,-------------------
10 , , - - - - - - .- - - - ,,,,.
.,,..,,..,,..,,..,,,.
,,,..,,.
that these slopes agree with the graph. ,,------------------
,,------------------
,,,,
,,,-----------------
,,,,, _______________ _
,,,,,,,______________
____________ __
b. On a photocopy of Figure 7-7b, draw the particular ,,,,,,,,,, _________ _
solutions that contain (1, 8) and (1, 12). Describe the
I\\\\\\\\\''''''''''
major difference in the behavior of the two graphs. I 111 I 111\\\\\\\\\\\\ X
10
c. Does the particular solution that contains (1, 10) behave
like that containing (1, 12) or that containing (1, 8)? Figure 7-7b
Justify your answer.
RS. a. For the differential equation given in Problem R4, use Euler's method to calculate
values of y for the particular solution that contains (1, 9). Use t:,.x = 1. Where does
the graph seem to cross the x-axis? On a photocopy of Figure 7-7b, plot the points
you calculated.
b . Use Euler's method, as you did in RSa, but with an increment of t:,.x = 0.1. Record
they-value for each integer value of x that shows in Figure 7-7b. Plot these points
on the photocopy you used in RSa.
c. Write a few sentences commenting on the accuracy of Euler's method far away
from the initial point when you use a relatively large value of t:,.x.
d. At what value of x would the graph described in RSc cross the x-axis?
R6. Predator-Prey Problem: Space explorers visiting a planet in a nearby star system
discover a population of 600 humanlike beings called Xaltos living by preying on a
herd of 7000 creatures that bear a remarkable resemblance to yaks. They figure the
differential equation that relates the two populations is
dy - 0.S(x - 6)
dx (y - 7)
342 Chapter
7:TheCalc
ulus of Growth andDecoy
where xis the number of hundreds of Xaltos and y is the number of thousands of
yaks. Figure 7-7c shows the slope field for this differential equation .
a. The numerator of the fraction in the differential equation
is dy / dt and the denominator is dx/ dt. Explain why the
two populations presentl y seem to be in equilibrium with
each other.
y
----------
,,
-----------
______ ......,,,,,,,,,
........, ....,,,,,
....,,,,,,,,
,....,, ______ ,,,,,,,,,,
b. Suppose that 300 more Xaltos move into the community.
Starting at the point (9, 7), draw the particular solution of
the differential equation on a photocopy of Figure 7-7c.
............,----,,,,,,,''
//.,, ___ ,,,,,,,
,,,.,.,.._,,,,,,,,,,
t t t t - t t t t t t t t t t t t
,,,, ... --l'lllllllltlt
\ \ \ \'
...
\ \
5 ,,,, ___ .,,....,,,,,,,,,,,
Explain why the graph goes clockwise from this initial ,,..., _____
,,,..., ___
.,,..,,,,,............,...,,,,,
,,,,,//I'll I I I I
point. Describe what happens to the two populations as
,--------,,,
,-------,,, ...,...,...,,,
............
............
...,...,...,.,,x
time goes on.
5
c. Suppose that instead of 300, 1300 more Xaltos move
Figure 7-7c
into the community. Draw the particular solution, using
this initial condition. What dir e circumstance befalls the
populations under this condition ? Surprising?!
d. What if only 900 more Xaltos move in instead of the 1300 described in R6c? Would
the same fate befall the populations? Justify your answer.
Concepts Problems
Cl. Differential Equations Leading to Polynomial Functions: You have shown that if
dy / dx is directly proportional to y, then y is an exponential function of x. In this
problem you will investigate similar differential equations that lead to other kinds of
functions.
a. If dy/ dx is directly proportional to y 112 , show that y is a quadratic function of x.
b. Make a conjecture about what differential equation would make ya cubic function
of x.
c. Verify or refute your conjecture by solving the differential equation. If your
conjecture was wrong, make other conjectures until you find the one that is right.
d. Once you succeed with Cle, you should be able to write a differential equa tion
whose solution is any specified degree . Demonstrate that you have seen the pattern
by writing and solving a differential equation whose solution is an eighth-degree
function.
,,c2. Film Festival Problem: In this chapter you have assumed a certain behavior for the
derivative of a function. Then you have integrated to find an equation for the function.
In this problem you will reverse the procedure. You will use measured values of a
function, then find the derivative to make use of the mathematical model.
In order to make money for trips to contests, the math club at Wyden High plans
to rent some video cassettes and present an all-night Halloween film festival in the
school gym. The club members want to predict how much money they could make
from such a project and to set the admission price so that they make the greatest
amount of money.
'' Adapted from data by Land y Godbold, as cited by Dan Teague.
Section
7-7: Chapter
Review
andTest 343
The club conducts a survey of the entire student body, concluding with the question
"What is the most you would pay to attend the festival?" Here are the results.
maximum dollars number of people
2.00 100
2.50 40
3.00 60
4.00 120
4.50 20
5.50 40
6.00 80
a. Make a chart that shows the total number of people likely to attend as a function
of the admission price.
b. Plot the data you charted in C2a. What kind of function might be a reasonable
mathematical model for people in terms of dollars? Fit an equation of this kind to
the data.
c. The amount of money club members expect to make is the product of price
and number of people. Write an equation that expresses amount of money as a
function of price.
d. What price should club members charge to make the greatest amount of money?
Justify your answer .
e. Why would club members expect to make less money if they charged more than
the price you determined in C2d? Why would they expect to make less money if
they charged less than the price in C2d?
C3. Gomper tz Growt h Curve Prob lem: Another function with a g( t)
sigmoid (S-shaped) graph sometimes used for population
growth is the Gompertz function, whose general equation is
g(t) = ae-ce- kl ,
where g(t) is the population at time t, and a, c, and k
are positive constants . The graphs of these functions
look somewhat like Figure 7-7d. In this problem you will
-5 5 10
investigate effects of the constants, maximum growth rates,
and limiting population values . f igure 7-7d
a. Let a= 10, c = 0.8, and k = 0.5 so that the equation is
g(t) = 1oe- o.se-0.5
t
Plot the graph of this particular Gompertz function . Confirm that it looks like the
grap h in Figure 7-7d. What does the limit of g(t) appear to be as t approaches
infinity? Confirm your answer by taking the limit in the equation. If g(t) represents
population, what is the significance of this limit in the real world?
b. Find the equation of the particular Gompertz function that fits the United States
population figures for 1960, 1970, and 1980 .
344 Chap
ter 7: TheCalculus
of Growth
andDecay
year population (millions)
1960 179
1970 203
1980 226
You may let time be zero in 1970. To get the exponential constants down to where
you can deal with them, you may take the ln of both sides of the equation twice.
By clever use of algebra, you can get two equations involving only the constant a.
Then you can use your grapher to calculate the value of a. Plot the graph of the
function. At what value does the population seem to level off?
c. Suppose that the 1980 data point had been 227 million instead of 226 million.
How would this change affect the predicted ultimate population of the United
States? Does the Gompertz equation seem to be fairly sensitive to slight changes
in initial conditions?
C4. Hot Tub Problem, Continued: In Problem 5 of Problem Set 7-3, you wrote a differential
equation for the volume of water remaining in a hot tub as it drained. That equation
is
dV = - 2v1 12
dt '
where V is the volume of water that remains t minutes after the drain is opened. By
solving the differential equation, you found that the 196 ft 3 of water initially in the
tub drained in 14 min. Suppose that while the drain is open, water flows in at the rate
F ft 3/min . Explore the effect of such an inflow on the remaining amount as a function
of time.
Chapter Test
Tl. Phoebe's Space Leak Problem: Phoebe is returning to earth in her spaceship when she
detects an oxygen tank leak. She knows that the rate of change of pressure is directly
proportional to the pressure of the remaining oxygen .
a. Write a differential equation that expresses this fact and solve it subject to the
initial cond ition that pressure is 3000 psi (pounds per square inch) at time t = 0
when Phoebe discovers the leak.
b. Five hours after she discovers the leak, the pressure has dropped to 2300 psi. At
that time, Phoebe is still 20 hr away from earth . Will she make it home before the
pressure drops to 800 psi? Justify your answer.
T2. Swimming Pool Chlorination Problem: Suppose that a pool is filled with chlorine-free
water. The chlorinator is turned on, dissolving chlorine in the pool at a rate of 30 g/ hr.
But chlorine also escapes to the atmosphere at a rate proportional to the amount
dissolved in the water. For this particular pool, the escape rate is 13 g/hr when the
amount dissolved is 100 g.
a. Write a differential equation that expresses this information and solve it to express
number of grams of chlorine in the pool as a function of the number of hours the
chlorinator has been running . Be clever to find an initial condition!
b. How long will it take for the chlorine content to build up to the desired 200 g?
Section7-7:Chapter
Review
andTest 345
T3. The slope field in Figure 7-7e is for the differential equation
;;;; ______y----------
ddy = - 0.36~. //;;
////
____________ ,,,,
___________ ,,,,,
X y
a. On a photocopy of Figure 7-7e, sketch the particular
//////;;-- ___,,,,,,,
Ill/ff///- ,,\\\\\\\\x
I ll ll H++++-1--t--t--t-t- -t-t -t-t _.
solution that contains the point (0, 4). \\\\\\\\'' -//I Ill/I
b. Use Euler's method for the particular solution you ,,,,,,______
\\\\\\,,,_
,,,, ____ --~////Ill
-----;////
------////
sketched in T3a to find y at x = 6. Use t.x = 0.1.
c. Separate the variables and solve the differential equa- ---------- ----------
tion algebraically. Transform the solution so that y is Figure 7-7e
expressed explicitly in terms of x.
d. Evaluate y when x = 6 for the algebraic solution you found in T3c. How close does
the solution by Euler's method you found in T3b come to this exact solution? How
close did the graphical solution you found in T3a come to the exact solution at
X = 6?
T4. Write a paragraph telling the most important thing you learned as a result of studying
this chapter.
7-8 Cumulative Review: Chapters 1-7
In your study of calculus so far, you have learned that calculus involves four major
concepts, studied by four techniques. You should be able to do four major things
with the concepts.
concepts techniques be able to do
Limits Graphical Define them.
Derivatives Numerical Understand them.
Indefinite integrals Algebraic Do them.
Definite integrals Verbal Apply them.
Two of these concepts, derivatives and definite integrals, are used to work problems
involving, respectively, the rate of change of a function, or the product of x
and y for a function in which y depends on x. Both derivatives and definite
integrals are founded on the concept of limit . Indefinite integrals, which are simply
antiderivatives, provide an amazing link between derivatives and definite integrals
via the fundamental theorem of calculus .
The following problems constitute a "semester exam" in which you are to demon-
strate your mastery of these concepts as you have studied them so far .
346 Chapter
7:TheCalc
ulusof Grow
th andDecay
Problem Set 7-8
Rocket Problems: Ella Vader (Darth's daughter) is driving in her rocket ship. At time t = 0
min she fires her rocket engine. The ship speeds up for a while, then slows down as the
planet Alderaan's gravity takes its effect. The graph of her velocity, v(t) miles per minute,
is shown in Figure 7-8a. In Problems 1-16, you will analyze Ella's motion .
1. On a sketch of Figure 7-8a, draw a narrow vertical strip of
width dt. Show a sample point (t, v ( t)) on the graph within
the strip. What physical quantity does v(t) dt represent? v(~ ....~ ..) ....;.....
+....;)~ '....) ..
2. If you take the sum I: v(t) dt from t = 0 tot = 8, what calculus
concept equals the limit of this sum as dt approaches zero?
3. Ella figures that her velocity is given by -t )(---~-1
r:( ..(
v(t) = t3 - 21t 2 + lOOt + 80. 100 .. ;. t >++
+'+++
Use the fundamental theorem of calculus to find the distance
she goes from t = 0 to t = 8. : : ! i ! i i : i i'
4. Calculate midpoint Riemann sums with n = 100 and n = 1000 5 10
increments. How do the results confirm that the fundamental
theorem gives the correct answer for the integral even though Figure 7-Ba
it has nothing to do either with Riemann sums or with limits ?
5. On a sketch of Figure 7-8a, draw a representation of an upper sum with n = 8
increments.
6. Explain why, for an integrable function, any Riemann sum is squeezed to the same
limit as the upper and lower sums as the widths of the increments approach zero.
7. Write the definition of definite integral. Write a statement of the fundamental theorem
of calculus . Be sure to tell which is which.
8. Calculate the integral in Problem 3 numerically by using your grapher's integrate
feature. Calculate the integral again graphically by counting squares. Compare the
answers with the exact value.
9. Use symmetric difference quotients with M = 0.1 min and M = 0.01 min to estimate
the rate of change of Ella's velocity when t = 4 min.
10. Write the definition of derivative.
11. For most kinds of functions there is a way to find the derivativ e algebraically . Use the
appropriate method to find the exact rate of change of Ella's velocity when t = 4.
12. At t = 4, was Ella speeding up or slowing down ? Justify your answer.
13. On a photocopy of Figure 7-8a, draw a line at the point (4, v(4)), having slope v'(4).
Clearly show how you construct the line. How is the line relat ed to the graph?
14. What is the physical name of the instantaneous rate of change of velocity?
15. Ella's maximum velocity seems to occur at t = 3. Use derivatives appropriately to find
out whether the maximum occurs when t is exactly 3 sec.
16. Find an equation for v"(t), the second derivati ve of v(t) with respect tot.
Section7-8:Cumulative
Review:
Chapters
1-7 34 7
Compound Interest Problems: When money is left in a savings account for which the interest
is compounded continuously, th e instantaneous rat e at which the money increases, dm /dt,
is directly proportional to m, the amount in the account at that instant.
17. Write a differential equation that expresses the prop er ty above .
18. Show the steps in solving the differential equation in Problem 17 form as a function
oft.
19. In one word, how does m vary with t ?
20 . The solution in Problem 18 is called th e-? - solution of the differential equation.
What word goes in th e blank ?
21. Find the particular solution in Problem 18 if m is $10,00 0 at t = 0 and $10,900
at t = 1.
22. In Problem 21, th e amount of mon ey in th e account grew by $900 in one year. True
or false : The amount of money will grow b y $9,000 in 10 yr. Justify your answer.
Discrete Data Problems: The techniques of calculus were invented for dealing with
continuous functions . The concepts can also be applied to functions specified by a table
of data . The following table gives values of y for various values of x.
X y
30 74
32 77
34 83
36 88
38 90
40 91
42 89
23. Use Simpson's rul e to estimat e I: y dx.
24 . Estimate dy /dx if x = 36 . Show how you get your answer.
Mean Value Theor em Problems: The proof of the fundamental theor em is based on the
mean value theorem. This theorem is a corollary of Rolle' s th eor em.
25 . State Rolle's theorem .
26 . Sketch a graph that illustrat es the conclusion of the m ean
value theorem .
Graphing Problems: Calculus is useful for analyzing the behavior
of graphs of functions .
27. Figure 7-8b shows th e gra ph of function f. On a photocop y
of this figur e, sketch th e deri va tive graph, f '.
28 . The function
f(x) = 2" - ~
X -1
has a discontinuity at x = 1. Sketch the graph. What kind of Figure7-Bb
discontinuity is it7
348 Chapter
7: TheCalculus
ofGrowth
andDecay
29. The function g(x) = x 113(x - 1) has g(O) = 0. Show that g'(O) is undefined. Show what
the graph of g looks like in a neighbor hood of x = 0. You may use your grapher's
cube root .
D ifferential Equation Prob lems: Figure 7-8c shows the slope field for the differential
equation
r _________________
_
ddy = 0.25~.
J--------------------
X y
30. On a photocopy of the graph, sketch the particular solutions --------------------
----------------////
-------------///////
that contain the points (0, 3) and (10, 4). ---------///////1111
------/////llllllllf
--///I I I I I I I It I It t t I X
31. The two solutions in Problem 30 share a common asymptote. _...,,......
--- \\\\ \\\
,,,,,,,,,,,,,,, \\ \\ \ \ \ \
Sketch the asymptote. State an initial condition that would ______
_________ ,,,,,,,,\\\\\\
,,,,,,,,,,,
___________
_____________ ,,,,,,,,,
,,,,,,,
give the asymptote as the graph of the solution. ________________ ,,,,
32. Solve the differential equation by separating the variables
and integrating . Find the equation of the particular solution Figure 7-8c
that contains the point (10, 4).
33. Use the function in Problem 32 to calculate the exact value of y when x = 10.5.
34. Demonstrate that you understand the idea behind Euler's method by calculating the
first point to the right of the point (10, 4) in Problem 32, with 6.x = 0.5 . How does
this value compare with the exact value in Problem 33?
A lgebraic Techniques Problems : You have learned algebraic techniques for differentiating,
antidifferentiating, and calculating limits .
35. Find :x (sin - 1x 3 ).
36. Find :: if x = ln(cos t) and y =sect.
37. Find J 4 ~x3x .
38. Find h'(x) if h(x) = 5x_
39 _ Find lim sin 5x + cos
x- 0 X
;x- 5x - 1
40. Plot the graph of the fraction given in Problem 39. Sketch the resu lt. Show how the
graph confirms your answer to Problem 39.
Journa l Prob lems: You have kept a calculus journal in which you record what you've
learned and what you're still unsure about .
41. Write what you think is the one most important thing you have learned so far as a
result of taking calculus.
42 . Write one thing in calculus about which you are still unsure.
Section7-8: Cumulative
Review:
Chapters
1-7 349
, ..
CHAPTER
8 The Calculus of Plane
and Solid Figures
A cable hanging under its own weight forms a curve called a catenary .
A cable supporting a uniform horizontal load, such as the cables in
the Golden Gate Bridge, forms a parabola. By slicing such graphs into
short segments the differential of arc length can be found. Integrating
this differential allows computation of the exact length of hanging
cables and chains, important information for construction of bridges .
351
--..... ,. ,i,-,;,..,, -,J,
Mathematical Overview
In Chapter 8 you will learn how definite integrals let you find exact
area, volume, and length by slicing an object into small pieces, then
adding and taking the limit . You will also use derivatives to find
where geometric figures have maxima, minima, and other interesting
features. You will explore the geometrical figures in four ways.
Graphically The logo at the top of each even-
numbered page of this chapter
shows an object for which you
can find length, area, volume,
and points of inflection .
Numerically X f'(x) f(x)
1.8 0.72 13.931
1.9 0.33 13.984
2.0 0 14 f-Max.
2.1 -0.27 13.987
2.2 -0.48 13.949
b
Algebraically V= Ja (x ~ -
TT xf) dy, volume by slicing into washers .
Verbally I think the most important thing I learned is that when you find
area, volume, length, and so forth, you use the same technique.
Draw a picture showing a representative slice of the object, pick
a sample point within the slice, find the differential of the quantity
I'm trying to find, then add up the differentials and take the limit,
which means integrate.
352
8-1 Cubic Functions and Their Derivatives
Recall that the graph of a quadratic function, f(x) = ax 2 + bx + c, is always a
parabola . The graph of a cubic function, f(x) = ax 3 + bx 2 +ex + d, is called a cubic
parabola. To begin your application of calculus to geometric figures, you will learn
about the second derivative, which tells the rate at which the (first) derivative
changes . From the second derivative you can learn something about the curvature
of a graph and whether the graph curves upward or downward .
OBJECTIVE Work alone or with your study group to explore the graphs of various cubic functio ns and
to make connections between the function 's graph and its derivatives.
Figure 8-1a shows the graphs of three cubic parabolas. They have different shapes
depending on the relative sizes of the coefficients a, b, and c. (The constant d affects
only the vertical placement of the graph, not its shape.) Sometimes they have two
distinct vertices, sometimes none at all. In Exploratory Problem Set 8-1, you will
accomplish the objective of this section.
Exploratory Problem Set 8 1
l. In Figure 8-1a,
f(x) = x 3 - 6x 2 + 9x + 3,
g(x) = x 3 - 6x 2 + 15x - 9, and
h(x) = x 3 - 6x 2 + 12x - 3.
y
For each function, find an equation for the derivative . Plot the
function and its derivative on the same screen. Then list as f
many connections as you can find between the function graph 10
and the derivative graph . Sketches will help .
2. What connection can you see between the graph of the
derivative of a function and whether or not the function has X
two distinct vertex points (high or low points) ?
Concave side
3. The sec ond derivativ e of a function is the derivative of the is downward.
(first) derivative . For instance, f " (x) (pronounced "f double
prime of x") is equal to 6x - 12. Find equations for the second
derivatives g"(x) and h"(x). What do you notice7 Figure 8-1a
4. Figure 8-la illustrates what it means for a curve to be concave upward at a given point,
or concave downwar d. What connection do you notice between the second derivative
and the direction of the concave side of the graph?
5. A graph has a point of inflect ion where it changes from concave downward to concave
upward . Tell two ways you could locate a point of inflection using derivatives .
Sec
tion8-1: Cub
ic Functions
andThei
r Derivatives 35 3
8-2 Critical Points and Points of Inflection
If a movin g object comes to a stop, several thin gs could happ en. It could remain
stopp ed, star t off again in th e sam e dir ection, or start off again in some diff erent
dir ection. When a car stop s or reverses dir ection, its velocit y goes through zero
(hop efully!). When a baseball is hit by a bat, its velocity chan ges abruptly and is
und efined at th e instant of conta ct. Figur e 8-2a shows how displa cem ent, d, and
velocity, v (derivative), could vary with time, x .
;:c>
d or v d or v d or v
Crit ical point
d
i Critical poi nt
~ : d
:.\
o---- v
V
X
~ i c~
X
C X
V
C
V
Car pau ses, th en starts Car sto ps and backs up . Baseball is hit. Derivative
again . Derivat ive is zero Derivative is zero and is un defined and chang es
but does not change sign . does chan ge sign . Local sign. Local minimum
"Platea u " point for d. max imum point for d. point for d.
Figu re 8-2a
A point where th e derivative is zero or undefin ed is called a critical point . The
word comes from "crisi s." (When one reac hes a crisis, thin gs stop and can go in
diff erent direc tion s.) "Critical point" is sometim es us ed for th e point on th e x-axis
and som etim es for the point on th e graph its elf. You must decide which is meant
from th e cont ext.
They -value at a critical point can b e a local maximum or a loca l minimum (Figure
8-2a, cente r and right). The word "local" is u sed to indi cat e th at f (c) is the maximum
or minimum of f (x) when xi s kept in a neighborhood (locality) of c. The global
maximum and global minimum ar e the largest an d small est of th e local maxima
and mini ma , res pectively. (Maxima and minim a are th e plur al form s.) A critical point
with zero derivative but no m aximum or minimum (Figure 8-2a, left) is called a
plateau point .
There are connec tion s between th e derivative of a function and the behavior of
it s gra ph at a critical point . For in stan ce, if th e derivative chan ges from positive
to negative (Figure 8-2a, cent er), th ere is a maximum point in th e fun ction graph .
As you saw in Section 8-1, th e seco nd derivative of a fun ction tells which way the
concave side of th e graph points . A point of inflection occur s where th e con cavity
chan ges direc tion .
OBJECTIVE From information about the first and second der ivatives of a function, tell whether th e
y-value is a local maximum or minimum at a critical point, tell whet her th e graph h as a
point of inflection , and use this informa tion to sketch the grap h or find the equ ation of
th e function .
354 Chapter
8: TheCalculus of PlaneandSolidFigures
Example 1 For the function graphed in Figure 8-2b, sketch a number-line graph for f ' and a
number-line graph for f " showing the sign of each derivative in a neighborhood
of the critical point at x = 2. Indicate on the number lines whether f (2) is a local
maximum or a local minimum, or whether the graph has a point of inflection at
X = 2.
f (x)
Figure 8-2b
Solutions Sketch a number line for f' and another one for f " . Each one needs three
regions: one for x, one for the derivative, and one for f (x ). Figure 8-2c shows a
convenient way to sketch them.
f (x) / /
f'(x) + +
X 2
f (x) '-.__/ p.i. ~
f "(x) +
X 2
Figure 8-2c
The graph is vertical at x = 2, so f'(2) is infinite. Insert the symbol oo in the f'(x)
region above x = 2, and sketch a vertical arrow above it in the f(x) region .
The graph off slopes up on both sides of x = 2. Since the derivative is positive
when the function is increasing, put a plus sign in the f '(x) region on both sides
of x = 2. Show upward sloping arrows in the f(x) region above the plus signs.
Since there is not a maximum or minimum value of f (x ) at x = 2, you don't need
to write any words in that region.
The graph is concave up for x < 2 and concave down for x > 2. Since a positive
second derivative indicates concave up and vice versa, put a plus sign in the
f"(x) region to the left of x = 2 and a minus sign to the right. Draw arcs in
the f(x) region to indicate the direction of concavity of the f graph. Since the
concavity changes (from up to down) at x = 2, the graph has a point of inflection
there. Write "p.i." in the f (x ) region above x = 2.
Section
8-2: Critical PointsandPointsofInflection 35 5
A NoteonConcavity
andCurvature
The word concave comes from the Latin cavus, meaning "hollow." So do "cave" and
"cavity." If the second derivative is positive, the first derivative is increasing . Figure
8-2d shows why the concave side is upward in this case and vice versa. As shown
in Figure 8-2e, the larger the absolute value of f " (x), the more sharply the graph
curves. However, as you will learn in Section 10-7, the curvature also depends on
the slope of the graph . For a given value off " (x), the steeper the slope, the less the
curvature.
r f(x)
Derivative is
decr easin g.
""f" (l)= 0.4
Small curvatur e
X
f " (l) = -2
Derivativ e is / Large curvatur e
increasin g.
X
Figure 8-2d Figure 8-2e
In Example 2, you will reverse the procedure of Example 1 and construct the graph
of a function from the number lines for its first and second derivatives .
Example 2 Figure 8-2f shows number-line graphs for the first and second derivatives of a
continuous function f. Use this information to sketch the graph off if f(4 ) = 0. The
abbreviation "e.p." signifies an endpoint of the domain . Describe the behavior of the
function at critical points .
f'(x) e.p . + 0
. I
X 2 3 4
f"(x) e.p. + = + 0
. I
X 2 3 4
Figure 8-2f
Solution Sketch the number lines . Add arrows and arcs in the f(x) region to show the
slope and concavity in the intervals between critical points (Figure 8-2g). Add
words to describe what features the graph will have at the critical points of
f and f'. Sketch a continuous function (no asymptotes) having the prescribed
features, crossing the x-axis at x = 4 (Figure 8-2h). The graph you draw could be
somewhat different, but it must have the features shown on the number lines
in Figure 8-2g.
356 Chapter
8: The Calculus
of Plane andSolidFigures
f( x)
f'(x)
min.
e.p. +
/
max.
t \._ -
plateau
0
\._
f(x)
X 1 2 3 4
no p.i. p.i.
f(x) '-._/ '-._/ ~ X
("(x) e.p. + = + 0
X 1 2 3 4
Figure 8-2g Figure 8-2h
In Example 3, you are given both the equation for the function an d an accurat e
grap h. You will be asked to find crit ical features algebraically, some of which ma y
be hard to see.
Example 3 Figure 8-2i shows the graph of f (x) = x 4 13 + 4x 113 .
a. Sketch numb er lines for f' and f " showing features that app ear clearly on the
graph.
b. Find equations for f'(x) and f "(x) . Show algebraically that the critical points you
drew in part a are correct . Fix any errors .
c. Write x - and y - coor dinat es of all maxima, minima, and points of inflection.
f(x)
15
Figure 8-2i
f(x)
f'(x)
\._ -/
min.
0 +
t /
+
X -1 0 1
Solutions a. Figur e 8-2j show s the two number
lines. f'(x) is zero at x = - 1, and infi-
nit e at x = 0. The graph is concave up f(x) '-._/ p.i. ~
for x < 0, and appears to be concave ("(x) +
down for x > 0. X 0
Figure 8-2j
Section
8-2:CriticalPoints
andPoints
ofInflection 357
b. f'(x) = 5xl f3 + f x - 2/3 = 5x - 2/3 (x + 1) Factor out the power of x "'~th the
smaller exponen t.
f " (x) = x -213 _ ~x - 5/3 = x - 513(x _ 2)
Critical points occur where either f'(x) = 0 or f '( x) is undefined.
f'(x) =0 = f x - 213(x + 1) = 0
A product is ze ro if and on ly if one of its
x - 213= 0 or x + 1 = 0
factors is ze ro.
x- 2 13 = 1 / x 2 /3 , which canno t equal zero .
.'. X =- 1 So the other factor mu st be zero .
f'(x) is undefined x=0 0- 213 = 1 / 0 2/3 = l / 0, which is infini te .
.'. critical points occur at x = 0 and x = - 1, as observed in part a.
Inflection points occur where f' has critical point s; that is, f "(x) is zero or
und efined .
f"(x) = 0 =x - 513(x - 2) =0
x - 513= 0 or x - 2 = 0
.'. X =2 Why?
f " (x) is undefined x =0 Why?
In order for ther e to be a point of inflection, f " (x) must change sign.
At x = 2, the factor (x - 2) in f "( x ) changes sign . Any power of a positive number is
At x = 0, the factor (4 / 9)x - 5i 3 changes sign. positive. If xis negative, th en x- 513
Inflect ion points are at x = O and x = 2. is negativ e. The cube root of a
negativ e numb er is nega tive, the
fifth power of that answer is also
negative, and th e recipro cal
(negative expo nent ) of that
negative answer is still negative.
Since the point at x = 2 did not show up in the original number line for f " in
part a, add this feature to your sketch, as shown in Figure 8-2k.
f(x) '-.._./ p .i. ~ p.i. '-.._./
f"(x) + 0 +
X 0 2
Figure 8-2k
c. To find the y-coordinates of the maxima and minima, substitute the x values
from part b into the f(x) equation.
f (-1) = (- 1) 4 13 +4( - 1) 113 = 1 - 4 = - 3
f (O) = 0
113
f (2) = (2) 4 13 + 4(2) = 7.559 ...
The local and global minima of f(x) are both -3 at x = -1.
Points of inflection are at (0, 0) and at (2, 7.559 ... ).
Th ere are no local or global maxima. f (x) approa ches infini ty as x approach es oo.
358 Chapter
8: TheCalculus
of PlaneandSolidFigures
A NoteonPowerswithNegative
Bases
Noninteger powers with negative bases are awkward . For instance, ( -1) 113 has a
r eal value, - 1; ( - 1) 116 has no r eal values; and ( - 1) rr has infinitely many distinct,
comple x valu es . As a result, calculators usuall y requir e you to enter noninteger
powers such as x 413 in the form (x 113 )-l or (x 4 ) 113 , ,vith an exponent having 1 as its
numerator and an integer denominator . That way th e calculator can tell what kind
of number th e answer will be, r eal or imaginary .
A NoteonInfiniteSlope
andInfiniteCurvature
If f(x) is defined but f'(x) is infinite, such as at x = 0 in Example 3, then the graph
will have a vertical tangent at that point . If f (x ) and f' (x ) are defined but f "(x) is
infinite, then the graph will have infinite curvature at that point. Infinite curvature
may be hard to imagine, as you will see in Problem 40 of Problem Set 8-2.
A NoteonUndefined
versus
Infinite
There are several reasons a function or its derivative ma y be undefined . Figure 8-21
shows five possibilities .
f (l ) is undefin ed because it is infinite . That is , the limit of f(x ) = oo as x
approach es L (Do not say, "f(l) = oo." This is bad form.)
f(2 ) is undefined but is not infinite . Ther e is a finit e limit for f(x) as x ap-
proaches 2.
f '( 2) is und efin ed b ecaus e f( 2) is und efined.
X f'( 3) is und efined because it is infinite. That is, th e limit of f' (x) = oo as x
2 3 4 5
approaches 3.
Figure 8-21 f' (4 ) is und efined but is not infinite . Th e left and right limits of f '(x ) ar e both real
number s, but th ey are not equal to each oth er .
Sometimes you will find criti cal points from just an equation for a function .
Example 4 shows how to do this graphicall y and num ericall y and how to confirm
th e results algebraically.
Example 4 Let f (x ) = - x 3 + 4x 2 + 5x + 20, with domain x E [- 2.5, 5] .
a. Plot the graph. Estimate the x- and y -coordinates of all local maxima or minima
and of all points of inflection. Tell the global maximum and minimum.
b . Write equations for f '(x) and f "(x ). Use them to find , either numerically or
algebraically, the precis e values of the x-coordinates in part a.
c. Tell why ther e are no other critical points or points of inflection.
Solutions a. Figur e 8-2m shows the graph in the given domain . By tracing, you find the
following :
Local minima of 20 at th e endpoint x = 5, and about 18.625 at x "" - 0.5.
Global minimum is about 18.62 5.
Local maxima of 48.12 5 at th e endpoint x = -2 .5, and about 44.192 at
X "" 3.2.
Global maximum is about 48.1 25 .
Point of inflection is at approximat ely (1.3, 31).
Section
8-2: CriticalPointsandPointsof Inflection 3 59
b. f'(x ) = - 3x 2 + Bx + 5
f "(x) = - 6x + 8
max. y max.
The graph off ' is shown on th e same screen as f in Figure 8-2m. To locate
the critical points precisely, eith er use your graphers solve feature to find
numerically where f'(x) = 0, or use the quadrati c formula. Thus,
-2 .5 ,, " ' 5 - 8 .J64 - 4( -3 )(5)
/
'
f' '- X = 2( - 3)
" x = - 0.5225 .. . or 3.1892 .. . ,
Figu re 8-2 m both of which confirm the estimates in part a.
To find the point of inflection precisely, set f "(x) = 0 and solve. Thus,
- 6X +8 =0 <:} X = !,
which confirms the estimate of x "" 1.3 in part a.
c. Since f'(x) is quadratic, there can be at most two zeros, both of which were
found in part b . Since f "(x ) is linear, th ere is exactly one zero, which was
found in part b. Therefor e, there are no more critical points or points of
inflection.
Figure 8-2n and the accompanying boxes pr esent the definitions and properti es of
this section.
f(x) n1ax., e.p.,
p.i.,
cu s p no max.
cusp
(no p.i. )
p .i., p.i.
!
p.i.,
cusp, p lateau i min .
min .
1
e.p.
min .
up down up do wn up up up down X
Concavity
Figure 8-2n
360 Chapter
8: TheCalculus
of Plane and Solid Figures
Definitions:CriticalPointsandRelatedFeatures
A critical point on a graph occurs at x = c if f(c) is defined, and if and only if f'(c)
either is zero or is undefined .
f (c) is a local maximum (or relativ e maximum) of f(x ) if and only if f(c) ?: f(x) for all
x in a neighborhood of c (that is, in an open interval containing c).
f(c) is a local minimum (or relative minimum) of f(x) if and only if f(c) ::; f(x) for all x
in a neighborhood of c.
f(c ) is the global maximum (or absolute maximum) of f(x ) if and only if f(c) ?: f(x) for
all x in the domain off.
f (c) is the global minimum (or absolute minimum) of f(x ) if and only if f(c) ::; f(x) for
all x in the domain of f.
The value of f"(c) is called the concavity of the graph off at x = c.
The point (c,f(c)) is a point of inflection if and only if f " (x) changes sign at x = c . (Old
spelling: inflexion, meaning "not bent.")
The point (c,f(c)) is a cusp if and only if f ' is discontinuous at x = c.
The point (c,f(c)) is a plateau point if and only if f '( c ) = 0, but f ' (x) does not change
sign at x = c.
Properties: Maximum
, Minimum
, andPointof Inflection
If f'(x) goes from positive to negative at x = c and f is continuous at x = c, then f(c) is
a local maximum.
If f'(x) goes from negative to positive at x = c, and f is continuous at x = c, then f(c)
is a local minimum.
If f"(c ) is positive, then the graph off is concave upwards at x = c.
If f " (c) is negative, then the graph off is concave downwards at x = c.
If f " (x) changes sign at x = c and f is continuous at x = c, then (c, f(c)) is a point of
inflection (by definition).
A maximum or minimum point (but not a point of inflection) can occur at an endpoint
of the domain of a function.
SymbolsforDerivatives
The following table lists different symbols for first and second derivatives . Each
one is convenient for a specific purpose. Make sure you ar e comfortable with any of
them.
function first second
value derivative derivative
f (x) f'(x) f " (x)
y y' y"
dy d2y
y
dx dx 2
Section
8-2:CriticalPointsandPointsof Inflection 361
ft:i,X
?
00
The symbol d 2 y/dx 2 is read "the second derivative of y with respect to x," or more
briefly, "d squared y , dx squared ." The symbol comes from thinking of d/dx as an
operator that acts on dy /dx and then doing "algebra" on d/dx(dy /dx) .
2
- d ( -dy) - - d ( - d (y) ) - ( - d ) (y) -- d 2 (y)- -d 2y2
dx dx dx dx dx dx 2 dx
Problem Set 8-2
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes .
a,. Sketch: y = x2
Q2. Sketch : y = x3
Q3. Sketch: y = cos x
Q4. Sketch: y = sin - 1 x
QS. Sketch: y = e-"'
Q6. Sketch: y = ln x
Ql. Sketch : y = tan x
QB. Sketch : y = x
Q9. Sketch : y = 1/ x
Q10. Sketch: x = 2
For Prob lems 1-12, sketch number-line graphs for f' and for f " showi ng what happens to
the value and to the sign of each derivative in a neighborhood of x = 2.
l. 2. 3.
f (x) f(x) f (x)
2
X
1
u 2
X
2
X
4. 5. 6.
f (x) f(x) f (x)
l
A X
1
V X
1
X
2 2 2
362 Chapter
8: TheCalcu
lusof Planeand SolidFigure
s
7. 8. 9.
f (x) f(x) f(x)
V 2
X
1
2
X
1
2
X
10. 11. 12.
f(x} f (x) f(x)
1 1
X X X
2 2
For Problems 13-18, mark the numb er lines with inform ation about the behavior of th e
graph . Sketch a graph of a continuou s fun ction f consistent with the information abou t
th e derivatives .
13. 14.
f'(x) + 0 0 + 0 f'(x) - 0 + 0 + 0 -
I I I I
-2 -3 -1 3
X
I 3 X
f"(x) 0
I
+ 0
I
rcx) + 0 - 0
I I
+ 0
I
-1 -1
X
I 2 X -2 2
15. 16.
f'(x) + 0 f'(x) 0 + = + 0
I I I I
X -2 X 2 3 4
f"(x) + + 0 f"(x) +
I I
X -2 X 3
17. 18.
f'(x) e.p. + o- zero- o e.p. f'(x) e.p. + 0 0 + e.p.
I I I I I I I
X -1 3 5 X 2 6 7
f"(x ) e.p. o- zero- o e.p. ('(x) e.p. o- zero- o + e.p.
I I I I I I I I
X -1 1 3 5 X 1 3 5 7
Section8-2: CriticalPoints andPoints of Inflection 363
19. Let f(x ) = 6x 5 - 10x3 (Figure 8-20). f(x )
a. Use derivativ es to find th e x-coordinat es for all critical
points of f and f'.
b. Explain why ther e are criti cal point s in part a that do not
show up on this graph.
-1
c. Explain why there is no maximum or minimum point at
x = 0, even though f '(O) equals zero . -5
Figure 8-20
f (x)
20. Let f (x) = O.lx4 - 3.2x + 7 (Figure 8-2p).
a. Use derivati ves to find the x-coordinates of all criti cal
points of f and f '.
b. Explain why th ere is no point of inflection at x = 0, even
though f "(O) equals zero .
c. Und er what conditions for f'(x ) and f "(x) can a graph be
"straight" without b eing horizontal ? I
1
I
2
I
3
I.
4
X
Figure 8-2p
21. Let f(x ) = xe -x (Figure 8-2q).
a. Use derivatives to find th e x-coordinat es of all critical
points of f and f' . X
b . How can you tell that ther e is a point of inflection even
though it does not show up on th e graph ?
c. Does the graph cross the x-axis at any point oth er th an
(0, 0)? Justif y your answer.
Figure 8-2q
f (x)
22. Let f (x ) = x 2 lnx (Figure 8-2r).
a. Use derivativ es to find th e x-coordinates of all critical
points of f and f' .
b. Show that the limit of f (x) is zero as x approaches zero
from the right , but not from th e left. L'Hospital's rule will
help. X
c. Are there any criti cal points th at do not show up on the
graph? If so, wher e, and what kind 7 If not, expl ain how
you know.
Figure 8-2r
364 Chapter
8: TheCalculus
of PlaneandSolid Figures
23. Let f(x) = x 513 + Sx213 (Figure 8-2s). f(x)
a. Use derivatives to find the x-coordinates of all critical
points off and f'.
b. Explain why there is a tangent line at the cusp, even though
f '(x) is undefined there.
c. Is there a point of inflection at the cusp? Is there a point X
of inflection anywhere else?
-5
Figure 8-2s
f(x)
24. Let f(x) = x 1 2 - 3x0 2 (Figure 8-2t). 10
a. Use derivatives to find the x-coordinates of all critical
points of f and f'. 5
b. The tangent is vertical at x = 0. How do you know that
there aren't several different values of y at x = O? X
-5
c. Is the graph straight or curved when xis less than -2? If
it is curved, which way is the concave side directed?
-5
Figure 8-2t
xz
25. Let f(x) = - 2- - (Figure 8-2u). f(x)
X - 1
a. Use derivatives to find the x-coordinates of all critical
points of f and f '.
b. Why are the discontinuities at x = 1 not removable?
c. As x gets very large, does the graph cross the x-axis,
_J IL X
approach it without crossing, or level off somewhere
above the x-axis? Justify your answer.
Figu re 8-2u
f(x)
x-1 5
26. Let f(x) = x + xz _ , (Figure 8-2v).
1
a. Use derivatives to find the x-coordinates of all critical
points of f and f' .
X
b. Explain why the discontinuity at x = 1 is removable, but -5
the discontinuity at x = - 1 is not removable .
c. By appropriate algebra, find an equation for the diagonal
asymptote the graph approaches as lxl becomes infinitely
large. /\ -5
Figure 8-2v
Section8-2:Critical
PointsandPointsof Inflection 365
For Problems 27-32,
a. Plot the graph. Estimate the x- and y-coordinates of all local maxima or minima and
all points of inflection . Tell the global ma'<imum and minimum.
b. Write equations for f'(x) and f"(x) . Use them to find numerically or algebraically the
precise values of the x-coordinates in part a.
c. Explain why there are no other critical points or points of inflection.
27. f(x) = - x 3 + 5x2 - 6x + 7 28. f(x) = x3 - 7x 2 + 9x + 10
29. f(x) = O.lx 3 + l.5x 2 + 7.6x - 3 30. f(x) = - x 3 + 9x 2 - 28x + 20
31. f(x) = 3x 4 + 8x 3 - 6x 2 - 24x + 37, for x E [- 3, 2]
32. f(x) = (x-1) 5 +4, forx E [-1,3]
33. Point of I nflect ion of a Cubic Funct ion: The general equation of a quadra tic function is
y = ax 2 + bx+ c, where a /. 0.
You recall from algebra that the "middle" of a quadratic function graph (that is,
the vertex) is at x = - b/(2a). The "middle" of a cubic function graph is at its point
of inflection . Prove that if f(x) = ax 3 + bx 2 +ex + d, where a /. 0, then the point of
inflection is located at x = - b/(3a).
34. Maximum and Minimum Points of a Cubic Function: The maximum and minimum
points of a cubic function are located symmetrically on either side of the point of
inflection. Prove that this is true in general for the cubic function f(x) = ax 3 + bx 2 +
ex + d. In terms of the coefficients a, b, c, and d, how far on either side of the point
of inflection do the maximum and minimum points occur?
35. Equation from Critical Points: Find the particular equation of the cubic function with
a local maximum at (5, 10) and its point of inflection at (3, 2). Use your grapher to
confirm your equation.
36. Can You Generate This Figure? Figure 8-2w shows the graph of a
piecewise-defined function in four different parts of its domain,
as follows:
b
(1, 2]: f(x) =a+ --
x - l
[2, 3] : f(x) = C +~ 2
[3,4] : f(x) = c+~
X
[4,5] : f(x) = mx + d
1 2 3 4
If f(3) = 1, f is continuous for all x in (1, 5], and is differentiable
at x = 2, find the values of the constants in the equations. Confirm Figure 8-2w
that your answers are correct by plotting the graph .
366 Chapter
8: TheCalcu
lusof PlaneandSolid Figures
Jll
37. Concav ity Concept Prob lem: Figure 8-2x shows the graph of
f(x) = x 3 . Tangent lines are drawn where x = - 0.8, - 0.5, 0.5 ,
and 0.8.
a. Calculate the slope for each given tangent point .
X
b. What is happening to the slope as x increases from - 0.8
to - 0.5? As x increases from 0.5 to 0.8? How do the values
of the seco nd derivative confirm th ese findings?
c. On which side of the tangent line do es th e graph of a
function lie if the graph is concave up at th e point of
tangency ?
Figu re 8-2x
38. Naive Graphing Prob lem: Ima Evian plots the graph of y = x 3 ,
using x = - 1, 0, and 1 (Figure 8-2y). From these three points she f(x) ".
concludes that the graph is a straight line. Explain to Ima how she 1 ,,.
could use derivativ es to avoid making this false conclusion . ""
39. Connections Between a Zero First Derivative and the Graph: If
-1 "
1"
" X
f' (c) = 0, the only thing you know for sure about the graph off is "
that it is horizontal at x = c . At least five things may be true about "
~"
" -1
the graph of f as x increases through c.
The graph stops increasing and starts decreasing (Figure 8-2z).
The graph stops decreasing and starts increasing (Figure 8-2aa).
The graph stops increasing but starts increasing again (Figure 8-2bb). Figu re 8-2y
The gra ph stops decreasing but starts decreasing again (Figure 8-2cc).
The graph is locally constant at x = c (Figure 8-2dd).
f(x)
X X
X
C
Figure 8-2z Figu re 8-2aa Figure 8-2bb
f(x)
(Locall y
constant)
X
X
C
Figure 8-2cc Figure 8-2dd
a. If you find that the first derivative of a function is zero at x = c, why can't you
conclude that there is a maximum point in the graph of f(x) at x = c 7
Section8-2: Critical PointsandPointsof Inflect
ion 367
b. Write the particular equation for a function that has a maximum at (1, 4), as in
Figure 8-2z .
c. Write the particular equation for a function that has a minimum at (1, 2), as in
Figure 8-2aa .
d. Write the particular equation for a function that has a horizontal tangent at (1, 3),
as in Figure 8-2bb or 8-2cc, but has no maximum or minimum point there.
e. Write the particular equation for a function that is locally constant in a neighbor-
hood of th e point (1, 2), as in Figure 8-2dd .
40. Infinite Curvature Problem: Show that the graph of
f(x) = lO(x - 1)4 13 +2
is defined and differentiable at x = l, but that the second derivative is infinite there .
Explore the behavior of f (x) close to x = l by zooming in on that point on the graph
or by constructing a table of values . Describe what you discover.
41. Histor ical Problem-The Second Derivative Test: In the days before graphers, one way
to find critical points was to use algebra to solve the equation f '( x) = 0. Once you
found a critical point, you had to find out whether the point was a local maximum, a
local minimum, or neither . This could be done by examining the sign of the second
derivative at each critical point . In this problem you will see how this second derivative
test was done. Suppose that
f(x) = x x
5
-
4
+ !x + 2x
3 2
- 3x + 7.
a. Find an equation for f'(x) . You should find that the resulting polynomial can be
factored into linear factors.
b. For what values of xis f'(x) equal to zero? What does f "(x) equal at each of these
values?
c. Based on your answers to 41b, which way is the graph concave at each critical
point? Without actually plotting the graph, how can you tell from the concavity
whether the graph has a local maximum or a local minimum at these critical
points? Confirm your answers by plotting the graph of f.
d. The following box gives a statement of the second derivative tes t. For each of the
thre e parts, make a sketch showing why that part is true.
Procedure
: TheSecondDerivativeTest for MaximaandMinima
If f'(c) = 0 and f "(c) is positive, then there is a local minimum at x = c.
If f'(c) = 0 and f "( c) is negative, th en there is a local maximum at x = c.
If f '(c) = 0 and f "(c) is zero, then no conclusion can be made about local maxima or minima
at X = C.
36 8 Chapter
8: TheCalcu
lusof Plan
e andSolid Figur
es
42 . Exponentia l and Poly nom ial Look-A like Prob lem: The f(x)
graphs of the functions 2
f(x) = e006 x and g(x) = 1 + 0.06x + 0.0018x 2 + 0.000036x 3
X
are shown in Figure 8-2ee. The graphs look remarkab ly -10 -5 5 10
alike! Are they really the same function? Examine the
g(x)
function values and the derivative values. For example,
do both f and g have the same derivative at x = O? Do
they have the same derivative at other places ? Show that
function g has an x-intercept but function f does not. X
- 10 -5 5 10
43 . A Pat hologica l Function: Consider the function defined by
1 Figure 8-2ee
f( X ) _
-
{ (x - 1) 3 sin -X - -l + 2, if X 'f l,
2, if X = l.
a. Plot the graph of the trigonometric branch of f. Use a friendly window of about
(0, 2] for x and a y-window of (1.99, 2.01].
b . As defined, is f continuous at x = l? Justify your answer.
c. Use the definition of derivative to prove that f'(l) = 0.
d. Zoom in on the point (1, 2). Does the graph appear to be locally linear at (1, 2) if you
use f( 1) = 2? How do you reconcile your answer with the fact that sin[ 1 / (x - l)]
makes an infinite number of cycles between 1 and any value of x close to 1? Why
do you suppose that f is said to be a "pathological" function?
e. In the reading material for this section it was stated that at least five things could
be true about f if f' (c) = 0. Tell a sixth thing that could be true .
44. Journa l Prob lem: Update your journal with things you've learned since the last entry.
You should include such things as those listed here.
The one most important thing you have learned since the last journal entry
The relationships between the signs of the first and second derivatives and the
behavior of the graph of the function itself.
The way the first and second derivative can be used to locate the maxima, minima,
and points of inflection algebraically
How your understanding of derivat ive has improved
Any techniques or ideas about the behavior of graphs that are still unclear to you
8-3 Maxima and Minima in Plane and Solid Figures
In Section 8-2, you found maximum and minimum values of a function where the
equation was already given. The problems of this section require you first to find
an equation for the area, volume, or perimeter of a geometric figure, then to use
the now-familiar techniques to find extreme valu es. As a res ul t you will be able to
investigate such things as how the canning industry saves money by packaging the
maximum volume of product with the minimum amount of metal.
Sec
tion8-3: Maxima
andMinimain Plane andSolid Figures 369
~?
)__fil
OBJECTIVE Given a plane or solid figure, find the maximum or minimum perimeter, area, or volume .
Example 1 Barb Dwyer must build a rectangular corral along the river bank . Three sides of the
corral will be fenced with barbed wire. The river forms the fourth side of the corral
(Figure 8-3a). The total length of fencing available is 1000 ft . What is the maximum
area the corral could have? How should the fence be built to enclose this maximum
area? Justif y your answers.
y
y Corral y
Figure 8-3a
Solution The first thing to realize is that the area is to be maximized. So you need an
equation for area as a function of one or more variables. Letting A stand for
area and x and y stand for the len gth of fence parallel to and perpendicular to
the river, respectively, you can write
A= xy.
Next you must find A in terms of one variable. Since there is a total of 1000 ft
of fencing, you can write an equation relating x and y.
x + 2y = 1000 = x = 1000 - 2y, where y E [O, 500] Why this domain for y?
..A= (1000 - 2y) (y) = lOOOy - 2y 2
Figure 8-3b shows the graph of A versus y . It is a parabola opening downward,
with its maximum point halfwa y between the two y-intercepts . Since these
intercepts are y = 0 and y = 500, and these points are the ends of the domain,
the maximum point is at
y = 250.
A
200,000
Confirming this fact by derivatives,
A ' = 1000 - 4y
A ' = 0 if and only if
.Y
1000 - 4y = 0 ~ y = 250. 500
.. X = 1000 - 2(250 ) = 500
:. A= (500 ) (250) = 125, 000 Figure 8-3b
Barb should make the corral 250 ft perpendicular to the river, and 500 ft parallel
to the river. The maximum area the corral can have is 125,000 square feet .
370 Chapter
8: TheCalculus
of Plane
andSolidFigures
Example 2 The part of the parabola y = 4 - x 2 from x = 0 to x = 2 is rotated about the y-axis
to form a surface. A cone is inscribed in th e resulting paraboloid with its vertex
at the origin and its bas e touching th e parabola (Figure 8-3c). At what radius and
altitud e does the maximum volum e occur ? What is this maximum volume? Justify
your answer.
r-igure 8-3c
Solution Pick sample point (x , y) on th e parabola in the first quadr ant where the con e
touche s it. Since it is volum e you are to maximi ze , find an equation for volume
Vi n term s of x and y.
Pick sampl e point (x , y) . Use V = ()(bas e
V -- !3 rrx 2 y
area)(height) to get \/ in term s of x and y.
V = rrx 2 (4 - x 2 ) ,x E [O, 2] Get V in term s of one variable, and specify a domain .
= %(4x 2 - x4 ) Sums are easier to differentiate than prod ucts.
V' = %(Bx - 4x 3 )
V' = 4;(x)(2- x2 ) Produ cts are easier to equate to zero than su ms.
X
V' = 0 <>x = 0 or 2 - x 2 = 0 <>x = 0 or x = ../2 The - /2 is out of the dom ain .
Figure 8-3d shows that th e maximum comes at x "' 1.4. At the criti cal point
Figure 8-3d x = 0, th e volume is a minimum . At the oth er endpoint, x = 2, the volume is
also minimum.
The maximum volume is at x = J2. The volume at this point is
V = %(2)(4 - 2)
= 4rr /3 = 4.18879 ....
There are some key steps in Examples 1 and 2 that will lead you to success in
max-min probl ems. Thes e steps are listed in the following box.
Sectio
n 83: MaximaandMinimain PlaneandSolid Figures 371
Technique:
Analysis of Maximum-Minimum
Problems
1. Make a sketch if one isn't already drawn.
2. Write an equation for the variable you are tr ying to maximize or minimize.
3. Get the equation in terms of one variable and specify a domain .
4. Find an approximate maximum or minimum by grapher.
5. Find the exact maximum or minimum by seeing where the derivative is zero or infinite .
Check any endpoints of the domain.
6. Answer the question by writing what was asked for in the problem statement .
Problem Set 8-3
DoTheseQuickly
The following problems are intend ed to refresh your skills. You should be able to do all
ten problems in less than five minutes.
r
QJ. Differentiate: y = (3x + s) - 1
Q2. Integrat e: J(x + 6) - 1 dx
Q3. Differentiate: y = x - 2 13
1
Q4. Integrate: f x - 213 dx X
Q5. Integrate: f x - 2 dx
Q6. Integrate : J x 0 dx
Ql. (d / dx )(c os - 1 x) =-?-
QB. fcotxdx =-?-
Q9. Sketch the graph of y = x 113 . Figure 8-3e
QJO.Sketch the graph of y " for the cubic function in Figure 8-3e.
1. D iv ided Stock Pen Prob lem: A rectangular stock pen (Figure
8-3f) is to be built using a tota l of 600 ft of fencing. Part of
this fencing will be used to build a fence across the middle
of the rectangle (see diagram) . Find the length and width of
the rectangle that give the maximum total area. Justify your Figure 8-3f
answer.
2. Motel Prob lem: A six-room mote l is to be built
with the floor plan shown in Figure 8-3g. Each
room is to have 350 ft 2 of floor space.
a. What dimensions should the rooms be in
order to have the minimum total length
of walls to build ? Justify your answer. Figure 8-39
b. How would your answer to 2a change if th e motel had ten rooms? Just three rooms?
372 "Chapter
8: TheCalculus
ofPlaneandSolidFigures
3. Two-Field Problem: Ella Mental has 600 ft of fencing to enclose
two fields . One is to be a rectangle twice as long as it is wide,
and the other is to be a square (Figure 8-3h). The square
field must contain at least 100 ft 2 . The rectangular one must
contain at least 800 ft 2 .
a. If x is the width of the rectangular field, what is the domain
of x?
b . Plot the graph of the total area contained in the two fields
as a function of x.
c. What is the greatest area that can be contained in the two Figure 8-3h
fields ? Justify your answer.
4. Two-Corral Problem: Suppose that you work on Big Bill
Spender's Ranch . Big Bill tells you to build a circular
fence around the lake and to use the remainder of your
1000 yd of fencing to build a square corral (Figure 8-3i).
To keep the fence out of the water, the diameter of the
circular enclosure must be at least 50 yd.
a. If you must use all of 1000 yd of fencing, how would
you build the fences so as to enclose the minimum Figure 8-3i
total area? Justify your answer.
b. What would you tell Big Bill if he asked you to build
the fences so as to enclose the maximum total area?
5. Open Box I: A rectangular box with a square base and no top
(Figure 8-3j) is to be made of a total of 120 cm 2 of cardboard.
a. Find the dimensions of the box of maximum volume .
b. Make a conjecture about the depth of th e maximum-volume
box in relation to the base length if the box has a fixed surface
area .
6. Open Box II (Project): For this project you are Figure 8-3j
to investigate the volume of the box formed
by cutting out squares from the four corners
of a 20 by 12 unit piece of graph paper (Figure
8-3k) and folding up the edges to form a box
without a top. / Cu:t o~t ~ b.y ~ squa~es: fr~m; th~ c~r~er~. ;
a. Each group should pick a different value -r
r-1-,r-r:-11--:-r-y-
of x, such as 1, 2, 3, 4, ... , cut out the : : : : : : : : : : : : : :
squares from the graph paper, then fold t
----1--t--r-tt-itt-rr-1
----i----
and tape it to form a box. What is the
largest possible value of x?
b. Calculate the volume of each box. Which
integer value of x gives the largest vol-
J
: /t
j
T
1111:i!!!J:
-t
11
! :::
...{....~ ttt
/;t1 ....
---i---
---~-r-r-t1-1-1-r--r/11-r--r1
i....
ume?
c. Find the value of x that gives the maxi-
mum volume. What is this volume ? Figure 8-3k
d. Construct a box of maximum volume.
Sec
tion 8-3:Maxima
andMinimain PlaneandSolidFigures 373
f!J;CI
100 1
7. Open Box III: A glass fish tank is to be constructed to hold 72 ft 3 of water. Its base and
sides are to be rectangular. The top, of course, is to be open. It is to be constructed so
that its width will be 5 ft but the length and depth are variable. Building the tank costs
$10 per square foot for the base and $5 per square foot for the sides . What is the cost
of the least expensive tank? Justify your answer.
8. Open Box IV (Project): The diagram shows an open-top box with
rectangular base x by y units and rectangular sides. The depth
of the box is z units (Figure 8-31).
a. Hold the depth constant. Show that the maximum volume of z
the box for a given amount of material occurs when x = y.
b. Set y = x, but let both vary as the depth varies. Find the
values of x and z that give the maximum volume for a given Figure 8-31
amount of material.
c. In what ratio are the values of x and z for the maximum-volume box in 8b? Is the
maximum-volume box tall and skinny or short and fat? Based on geometry, why is
your answer reasonable?
9. Shortest-Distance Problem: What point on the graph of y = ex is closest to the origin
(Figure 8-3m)? Justify your answer.
10. Track and Field Problem: A track 400 min periphery is to be laid out on the practice
field (Figure 8-3n) . Each semicircular end must have a radius of at least 20 m, and each
straight section must be at least 100 m. How should the track be laid out so that it
encompasses the least area? Justify your answer.
X
- ---x---~
Figure 8-3m Figure 8-3n
11. Ladder Problem: A ladder is to reach over a fence 8 ft high to
a wall that is 1 ft behind the fence (Figure 8-30). What is the
length of the shortest ladder that can be used7 Justify your
answer.
Figure 8-30
374 Chapter
8: TheCalculus
of Plane
andSolidFigures
12. Ladder in the Half Problem: A nonfolding ladder is to be taken around a corner where
two hallways intersect at right angles (Figure 8-3p). One hall is 7 ft wide, and the other
is 5 ft wide. What is the maximum length the ladder can be so that it will pass around
such a corner, given that the ladder must be carried parallel to the floor?
13. Rotated Rectangle Problem: A rectangle of perimeter 1200 mm is rotated in space using
one of its legs as the axis (Figure 8-3q). The volume enclosed by the resulting cylinder
depends on the proportions of the rectangle . Find the dimensions of the rectangle that
maximize the cylinder's volume.
7 ft---+-
-Rectan gle
t
5 ft
i
Figure 8-3p Figure 8-3q
14. Rotated Rectang le Generalization Prob lem: The rectangle of maximum area for a given
perimeter P is a square. Does rotating a square about one of its sides (as in Problem
13) produce the maximum-volume cylinder? If so, prove it. If not, what proportions do
produce the maximum-volume cylinder?
15. Tin Can Problem: A popular size of tin can with "normal"
proportions has a diameter of 7.3 cm and an altitude of 10.6 cm
(Figure 8-3r). f-- 7.3
cm ---j
a. What is its volume?
cc::>:>
b. The volume is to be kept the same, but the proportions are
to be changed. Write an equation expressing total surface of
the can (lateral surface plus two ends) as a function of radius
T
10.6 cm
and altitude. Transform the equation so that the volume is in
terms of radius alone.
c. Find the radius and altitude of the can such that it has
_j
minimum surface area . Is the can tall and skinny or short and
fat? What is the ratio of diameter to altitude? Justify your Figure 8-3r
answers.
d. Does the normal can use close to the minimum amount of metal? What percent
of the metal in the normal can could be saved by using cans with the minimum
dimensions?
e. If the United States uses 20 million of these cans a day and the metal in a normal
can is worth $0.06, how much money could be saved a year by using minimum-area
cans?
Section
8-3:Maxima andMinimain PlaneandSolidFigures 375
16. Tin Can Generalization Project: The tin can of minimum cost in Problem 15 is not
necessarily the one with minimum surface area. In this problem you will investigate
the effects of wasted metal in the manufacturing process and of overlapping metal in
the seams.
a. Assume that the metal for th e ends of the can in Problem 15 costs k times as
much per square centimeter as the metal for the cylindrical walls. Find the value
of k that makes the minimum-cost can have the proportions of the actual can . Is
it reasonable in the real world for the ends to cost this much more (or less) per
square centimeter than the walls? Explain.
b. Assume that the ends of the tin can in Problem 15 are cut from squares and that
the rest of the metal from the squares is wasted . What value of kin 16a would make
the actual can have minimum cost under this assumption? Is the can that uses the
minimum amount of metal under this assumption closer to the proportions of the
actual can or farther away?
c. Assume that the ends of the can are made from metal disks that overhang by 0.6
cm all the way around. They are made this way in order to have enough overlap to
fabricate the joints at the top and bottom of the can . Assume also that there must
be an extra 0.5 cm of metal in the circumference of the can in order to have an
overlap for the vertical seam . How do these assumptions affect the dimensions of
the minimum-area can in Problem 15?
17. Cup Prob lem: Suppose you have been hired by the Yankee Cup Company. They
currently make a cylindrical paper cup of diameter 5 cm and altitude 7 cm. Your job
is to see whether paper can be saved by making cups that hold the same amount of
liquid but have different proportions.
a. Find the dimensions of a same-volume cup that uses a minimum amount of paper .
b. What is the ratio of diameter to altitude for the minimum-area cup?
c. Paper costs $2.00 per square meter. Yankee makes 300 million of this kind of cup
per year. Write a proposal to your boss telling her whether or not you think it would
be worthwhile to change the dimensions of Yankee's cup to those of the minimum
cup . Be sure to show that the area of the proposed cup really is a minimum.
d. Show that in general if a cup of given volume V has minimum total surface area,
then the radius is equal to the altitude.
18. Duct Problem: A duct made of sheet metal
connects one rectangular opening in an air-
conditioning system to another rectangular
opening (Figure 8-3s). The rectangle on the
X
left is at x = 0 in . and the one on the right is
at x = 100 in. Cross sections perpendicular
to the x-axis are rectangles of width z =
30 + 0 .2x and y = 40 - 0.2x. Figure 8-3s
a. Find the areas of the two end rectangles .
b . What is the cross -sectional area of the duct at x = 80?
c. What x-values give the maximum and the minimum cross-sectional areas?
376 Chapter
8:TheCalculus
of Plane
andSolidFigures
19. Rectangle in Sinusoid Problem: A rectangle is inscribed y
in the region bounded by one arch of the graph of
y = cos x and the x-axis (Figure 8-3t). What value of x
gives the maximum area? What is the maximum area ?
Figure 8-31
20. Building Problem : Bill Ding plans to build a new hardware store. He buys a rectangular
lot that is 50 ft by 200 ft, the 50-ft dimension being along the street . The store is to
have 4000 ft 2 of floor space. Construction costs $100 per linear foot for the part of the
store along the street and only $80 per linear foot for the parts along the sides and
back. To what dimensions should Bill build the store to minimize construction costs?
Justify your answer .
y
21. Triangle under Cotangent Problem: A right triangle has a vertex at
the origin and a leg along the x-axis . Its other vertex touches the
graph of y = cot x, as shown in Figure 8-3u .
a. As the right angle approaches the origin, the altitude of the
triangle approaches infinity, and the base length approaches
zero. Find the limit of the area as the right angle approaches the
origin.
X
b. What is the maximum area the triangle can have if the domain
is the half-open interval (0, rr / 2]? Justify your answer.
Figure 8-3u
22. Triangle under Exponential Curve Problem: A right triangle
has one leg on the x-axis . The vertex at the right end of
that leg is at the point (3, 0). The other vertex touches the
y
graph of y = ex. The entire triangle is to lie in the first
9
quadrant. Find the maximum area of this triangle. Justify
your answer.
23. Rectangle in Parabola Problem: A rectangle is inscribed
in the region bounded by the x-axis and the parabola
y = 9 - x 2 (Figure 8-3v).
a. Find the length and width of the rectangle of greatest
area. Justify your answer.
X
b. Find the length and width of the rectangle of greatest
perimeter. Justify your answer.
c. Does the rectangle of greatest area have the greatest Figure 8-3v
perimeter?
Section
8-3:Maxima
andMinimain PlaneandSolid Figures 377
24. Cylinder in Paraboloid Problem: The parabola y = 9 - x 2 is y
rotated about the y-axis to form a paraboloid . A cylinder is
coaxially inscribed in the paraboloid (Figure 8-3w).
a. Find the radius and altitude of the cylinder of maximum
volume. Justify your answer.
b. Find the radius and altitude of the cylinder of maximum
lateral area.
c. Find the radius and altitude of the cylinder of maximum
total area (including the ends) . Justify your answer. X
d. Does the maximum-volume cylinder have the same di-
mensions as either of the maximum-area cylinders in 24b
Figure 8-3w
or c?
e. Does rotating the maximum area rectangle in Problem 23a
produce the maximum volume cylinder in this problem?
f. If the cylinder of maximum volume is inscribed in the paraboloid formed by
rotating the parabola y = a 2 - x 2 about the y-axis, does the ratio (cylinder ra-
dius) :(paraboloid radius) depend in any way on how long the paraboloid is? (That
is, does it depend on the value of the constant a?) Justify your answer.
25. Cylinder in Sphere Problem: A cylinder is to be inscribed
in a sphere of radius 10 cm (Figure 8-3x). The bottom and
top bases of the cylinder are to touch the surface of the
sphere. The volume of the cylinder will depend on whether
it is tall and skinny or short and fat.
a. Let (x, y) be the coordinates of a point on the circle, as
shown. Write an equation for the volume of the cylinder
in terms of x and y.
b . What radius and altitude of the cylinder will give it the
maximum volume? What is this volume? Justify your
answer.
c. How are the radius and altitude of the maximum-volume
cylinder related to each other? How is the maximum Figure 8-3x
cylinder volume related to the volume of the sphere?
26. Conical Nose Cone Problem: In the design of a missile nose
cone, it is important to minimize the surface area exposed to
the atmosphere. For aerodynamic reasons, it is important that
7 y
the cone be long and slender . Suppose that a right circular
nose cone is to contain a volume of 5rr ft 3 Find the radius
and height of the nose cone of minimum lateral surface area,
subject to the restriction that the height must be at least twice
the radius of the base . (The differentiation may be easier if
you minimize the square of the area.)
X
27. Cylinder in Cone Problem : A cone of altitude 7 cm and base
5
radius 5 cm has a cylinder inscribed in it, with the base of the
cylinder contained in the base of the cone (Figure 8-3y).
a. Find the radius of the cylinder of maximum lateral area Figure 8-3y
(sides only).
378 Chapter
8: TheCalcu
lusof PlaneandSolid Figures
b. Find the radius of the cylinder of maximum total area (including the top and bottom
bases). Justify your answer.
28. Genera l Cyli nder in Cone Problem: A given cone has a cylinder inscr ibed in it, with its
base contained in the base of the cone.
a. How should the radius of the cone and cylinder be related for the lateral surface of
the cylinder to have a maximum area? Justify your answer.
b. Find the radius of the cylinder that gives the maximum total area.
c. If the cone is short and fat, the maximum total area occurs y
where the altitude of the cylinder drops to zero and all
the material is used in the top and bottom bases. How
8m
must the radius and altitude of the cone be related for this X
phenomenon to happen?
29. Elli ptica l Nose Cone Prob lem: The nose of a new cargo plane is
to be a half-ellipsoid of diameter 8 m and length 9 m (Figure
8-3z). The nose swings open so that a cylindrical cargo container
can be placed inside. What is the greatest volume this container
could hold? What are the radius and altitude of this largest Figure 8-3z
container? Justify your answer.
30. Submarine Pressure Hull Project: According to a new design,
y
the forward end of a submarine hull is to be the shape of Cylinder
a paraboloid 16 m long and 8 min diameter (Figure 8-3aa).
Since this is a doubly curved surface, it is hard to bend thick X
steel plates into this shape. So the paraboloid is to be made
of relatively thin steel, and the pressure hull built inside as a
cylinder (a singly curved surface). A frustum of a cone is also a
singly curved surface, which would be about as easy to make and
which might be able to contain more volume (Figure 8-3bb). How Figure 8-300
much more volume could be contained in the maximum-volume y
Frus tum of con e
frustum than in the maximum-volume cylinder? Some things
you will need to find are the equation of this particular parabola
and the equation for the volume of a frustum of a cone .
31. Local Maximum Property Problem: The definition of local
maximum is as follows : f(c) is a local maximum off on the
interval (a, b) if and only if f(c) 2'..f(x) for all values of x in
(a, b). Figure 8-3cc illustrates this definition. Figure 8-3bb
a. Prove that if f(c) is a local maximum off on (a, b) and
f is differentiable at x = c in (a, b), then f'(c) = 0. To do
this, you can consider the sign of the difference quotient
when x is to the left and to the right of c, then take left
and right limits.
b . Explain why the property in 31a would be false without
the hypothesis that f is differentiable at x = c.
c. Explain why the converse of the property in 31 a is false. X
a X C b
Figure 8-3cc
Section
83:Maximo
andMinimain PlaneandSolid Figures 379
32. Corral with Short Wall Project: Millie Watt needs to use an electric fence to build a
rectangular corral along a wall. Part or all of the wall forms all or part of one side
of the corral. The total length of fence (excluding the wall) is to be 1000 ft. Find the
maximum area that can be enclosed if the following is true.
a. The wall is 600 ft long (Figure 8-3dd, left).
b. The wall is 400 ft long (Figure 8-3dd, center) .
c. The wall is 200 ft long (Figure 8-3dd, right).
W:./.1
,M N tt.4.t--t#.ttN-t:#l t#t--#/.tijf.Wl.-./:i!#t
IJ
i
Corral ~
~l l Corral
~=~3.~~
~
~
~#ttl.'~ttl
Figure 8-3dd
33. Journal Problem: Update your journal with things you've learned since the last entry .
You should include such things as those listed here.
The one most important thing you have learned since the last journal entry
The critical features you can tell about the graph of a function from its two deriva-
tives
The way you solve a real-world problem involving a maximum or minimum value
Any techniques or ideas about extreme-value problems that are still unclear to you.
8-4 Area of a Plane Region
In Section 5-10, you learned the following procedure for setting up a definite integral.
Slice the region under a graph into narrow vertical strips of width dx.
Show a representative strip and a sample point (x, y) on the graph within the
strip.
Write the differential y dx for the product of x and y corresponding to that strip.
- --- dx X Add up the y dx's and take the limit (that is, integrate).
4
Figure 8-4a illustrates the process. In this section you will apply this technique to
Figure 8-4a find the areas of regions bounded by two curves or by curves for which slicing is
best done horizontally.
OBJECTIVE Given a plane region bound ed by the graphs of one or more relations, write a definite
integral representing the area of the region, and evaluate the integral numerically or
algebraically (using the fundamental theorem of calculus) to find the area.
380 Chapter
8: TheCalculus
of PlaneandSolidFigure
s
Example 1 Find algebraically (by the fundamental theor em of calculus) th e ar ea of the "triangu-
lar" region in Quadrant I th at is bounded by th e graphs of y = 4 - x 2 and y = 4x - x 2
and the y-axis.
y
Solution Ske tch th e graph as shown in Figure 8-4b. 4
Be careful to select th e prop er "triangular"
region, in this case th e one bound ed by the
y-axi s. Then dra w a repr esent ative slice.
Choos e sampl e points, one on each graph,
for th e sam e valu e of x . Findin g d A involves
geometr y and algebra .
d A = (Y1 - Yz) d x
= (4 - x 2 - 4x + x 2 ) d x
=(4 - 4x ) dx
Figure 8-4b
Be sur e you subtra ct (larger y-value) - (small er y-valu e) so that th e height of
th e rectangle will be positive. Before you can integrate , you must find the limit s
of integration . The first strip is at th e y-axis , so the lower limit of integration is
x = 0. The last strip is at th e point where th e two graph s cross . If you cannot
tell this point by insp ection (x = 1, in thi s case), you can set the two y-values
equ al to each oth er and solve for x . The answer is th e upp er limit of int egration .
4 - x 2 = 4x - x 2 => 4 = 4x => x = 1
A"" ,Ld A = ,L (4-4 x ) d x Total area is app roxim ately the sum of th e dA's .
A = J (4 - 4x ) dx
0
1
A definite integra l is a limit of a Riemann sum.
= 4x - 2x 2 j ~ Use the fundam ental theo rem.
=4-2 - 0+ 0 =2
Sometim es verti cal slicing mak es th e problem hard to analyze . Figure 8-4c shows
th e region bound ed by th e gra ph s of x = 3 + 2y - y 2 and x + y = -1.
Awkward to slice verti cally Appropriate to slice horizontall y
Figure 8-4c
Section84:Areaofa PlaneRegion 381
f!i...
?
The length of the vertical strip is given by different rules for different parts of the
domain. If x is negative the length of the strip is given by
curve - line.
For positive values of x, the length is given by
curve - curve.
This difficulty is circumvented by slicing hori zontally, as shown in Example 2.
Example 2 For the region in Figure 8-4c bounded by th e graphs of x = 3 + 2y - y 2 and
x + y = - 1, write an int egral for the ar ea and evaluate it num ericall y. Evaluate the
integral exac tly by th e fundamental theor em and compare your answers .
Solution See the right-hand graph in Figure 8-4c. Pick sample points (x 1, y) and (x2, y) .
Then
dA = (curve - line)d y = (x1 - x2) dy
= [3 + 2y - y 2 - (- y - 1)] dy = (4 + 3y - y 2) dy.
The graphs int ersect where the two x-values are equal.
3 + 2y - y 2 = -y - 1
O= y2 - 3y - 4 ~ (y + l )(y - 4) = 0 ~ y = - 1 or y = 4
The total area is found by adding the dA's and taking the limit (that is ,
integrating) .
A = f 1
(4 + 3y - y 2 ) dy Write an integral equal to the area.
= 20 .83333 ... Use num erical integration.
The answer can also be found using the fundamental theor em .
4
A = 4 y + i y 2 - .!.y 3 I
2 3 - I
= 16 + 24 - Q.:-
3
(- 4) - i2 + ::..!.
3
= 20 ~, which equals 20.8333 ... .
382 Chapter 8: The Calculusof PlaneandSolidFigures
Problem Set 8-4
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
QJ, Integrate : Jx - 2 dx y
Q2. Integrate: J u - 1 du
Q3. Integrate: J un du
Q4. Integrate: J(lnx )7(dx / x )
QS. Integrate: Jln 3 dx
Q6, Integrate: J4x dx
Ql. Integrate: Jsec x tanx dx
QB.Sketch the graph of y ' : y = Ix - 21 + 3. Figure 8-4d
Q9. In Figure 8-4d, y = 4 - x 2 . Write the area of the rectangle in terms of x and y at the sample
point .
Q10. Find the maximum area of the rectangle in Figure 8-4d.
For Problems 1-14, sketch the region, write an integral for the area, and calculate the area
either approximately by numerical int egration, or exactly by the fundamental theorem.
1. Bounded by the graph of y = - x 2 + 6x - 5 and the x-axis
2. Bounded by the graph of y = x 2 - x - 6 and the x-axis
3. Bounded by the graph of x = (y - l)(y - 4) and th e y-axis
4. Bounded by the graph of x = 5 + 4y - y 2 and the y-axis
5. Bounded by the graphs of y = x 2 - 2x - 2 and y = x + 2
6. Bounded by the graphs of y = - 2x + 7 and y = x 2 - 4x - 1
7. Bounded by the graphs of y = 0.5x 2 + 2x and y = - x 2 + 2x + 6
8. Bounded by the graphs of y = 0.2x 2 + 3 and y = x 2 - 4x + 3
9. Bounded by the graphs of y = 2e 02 x and y = cos x, between x = O and x = 5
10. Bounded by the graphs of y = sec2 x and y = e 2", in Quadrant I, for x ::; 1
11. Bounded by the graphs of y = x + 3 and x = - y 2 + 6y - 7
12. Bounded by the graphs of y = - 2x + 11 and x = 0.25y 2 - 0.5y - 0.75
13. Bounded by the graphs of y = x 3 - 4x and y = 3x 2 - 4x - 4 (The grapher will help
you find the region .)
14. Bounded by the graphs of y = x 213 and y = (x + 1) 112 + 1 (The grapher will help you
find the region and locate the intersection points.)
Section8-4:Areaof a PlaneRegion 383
:,n
l(j
;4'1.
15. Wanda Wye needs to find dA for the region bounded by y = x 2 and
y = x (Figure 8-4e). She wants to know whether to use
(line - curve) or (curve - line).
Explain to Wanda how she can always choose the correct way. Also,
tell her what her answer will be if she chooses the wrong one .
16. Peter Doubt must find the area of the region shown in Figure 8-4f.
a. Explain to Peter why hori zontal slicing is not appropriate for
this problem .
b. Peter is worried because some of the region is below the x- Figure 8-4e
axis . Explain why, with proper slicing, each value of dA will be
positive and that he thus need not worry.
17. Parabolic Region Problem: Prove that the ar ea of a parabolic
region as shown in Figure 8-4g is always 2/ 3 of the area of the
circumscribed rectangle . Show the significance of what you have
proved by using the result to find quickly the area of the parabolic
region bounded by the graph of y = 67 - 0.6x 2 and the horizontal
line y = 7.
18. Sinusoida l Region Problem: Show that the area under one arch of
the sinusoid y = sin x (Figure 8-4h) is a rational number by finding Figure 8-41
this number. Demonstrate the significance of this fact by finding
y
quickly the area under one arch of y = 7 cos 5x.
19. Ellipse Area Problem: Figure 8-4i shows the ellipse
9x 2 + 25y 2 = 225 .
Write an integral equal to the area of the region . Evaluate the integral
numerically. This answer can also be found without calculus, using
the x-radius and they-radius of the ellipse . Make a conjecture about
a formula for the area of an ellips e. How does your conjectured
formula compare with that for the area of a circle?
Figure 8-4g
20. Area of a Region Parametrically Problem : The ellipse in Figure 8-4i
has parametric equations
y
X = 5 COS t
3
y = 3 sint .
The area of a representative vertical
strip is dA = 2y dx . Find the area by
--~~~-~~~-,-- 5
X Figure 8-4 h
expressing bothy and dx in terms of t .
You must also find out what the limits
of integration are for t , corresponding
to the x-limits in Problem 19. Figure 8-4i
2 l. Golf Course Problem: Figure 8-4j shows a small puttin g green to be
constructed on a new golf course . The dimensions x and y are in
yards. Find, approximately, the area of the putting green . Tell how
you arrived at your answer. Figure 8-4 j
384 Chapter8: TheCalculus
ofPlaneandSolid Figures
22. Area Check Prob lem: Find the area of the region bounded by the graph of y = x 2 and
the line y = x + 6 exact ly, using the fundamental theorem. Then calculate midpoint
Riemann sums with n = 10, n = 100, and n = 1000 increments. Show that the Riemann
sums are approaching the exact answer as the number of increments becomes large .
23. Curve Sketchi ng Review Prob lem: Given t (x ) = x + sinx, plot the graph oft using an
x-window of ab out O ::; x ::; 4rr. Sketch the result. Then calculate the x-coordinates of
all points at which the tangent to the graph is horizontal. Does t '(x) change signs at
any of these points? Show any such points on your graph .
24 . Max imum-Minimum Review Problem : A rectang le is to be inscribed in the ellipse of
Problem 19. The sides of the rectangle are to be parallel to the x-, and y-axes, with the
corners touching the ellipse . Find the maximum area such a rectangle could have .
8-5 Volume of a Solid by Plane Slicing
In Section 8-4, you refreshed your memory about setting up definite integrals by
slicing a region into narrow strips . In this section you will extend the technique to
find volumes of solids that have curved boundaries. The idea is to slice the solid into
thin, flat slabs, find the volume of a representative slab by multiplying its area by
its thickness, then add the volumes and take the limit as the thickness approaches
zero.
OBJECTIVE Given a solid whose cross-sectional area varies along its length, find its volume by
slicing into slabs and performing the appropriate calculus, and show that your answer is
reasonable.
Figure 8-Sa shows a rectangular solid. Its volume is (length)(width)(height). The area
of a cross section is (width)(height) . So volume of this solid is
Volume = (area)(length).
If a solid has some other shape, its volume is still (area)(length). But th e area may
vary for different cross sections . Example 1 shows you how to find the volume .
Figure 8-5a
Example 1 (Rotate a region around the x-axis.) The region un der the graph of y = x 113 from
x = 0 to x = 8 is rotated about the x-axis to form a solid . Write an integral equal to
the volume of the solid and evaluate it to find the volume . Assume that x and y are
in centimeters . Show that your answer is reasonable.
of a SolidbyPlaneSlicing
Section 8-5: Volume 385
Solution First, draw a picture of the region (Figure 8-Sb, left). Next, slice the region into
strips as thou gh you planned to find the area . The idea is to slice perpendicular
to the axis around which the region will rotate . Show a representative strip as
in Figure 8-Sb. Pick a sample point (x, y) that is on the graph of y = x 113 .
y y
Sampl e point Disk of
volum e
(x,y) i dV
----dx
X X
8 8
Draw th e region . Rotat e th e region and the strip to
Slice a strip perpendicular to the axis of rotation . form a solid. The rotating stri p forms
Pick a sample point (x, y) . a flat disk of volum e dV.
Figure 8-5b
Then draw what the solid looks like as the region rotates around the x-axis
(Figure 8-Sb, right). As the region turns, the rotating strip generates a disk. This
disk can be thought of as a short, fat cylinder whose "altitude" is dx and whose
cross section at any valu e of x is close to what it is at th e sampl e point (x, y).
The volume of th e disk, dV, is
dV = (cross-sectional area at the samp le point)(altitude, dx ) .
Since the radius of the disk is y, its volume will be
dV = rry 2 dx = rr (x 113 )2 dx = rrx 213 dx.
The solid itself can be considered to be made up of a stack of these disks . So
the approximate volume of the solid is equal to the sum of the disk volumes .
V "" L dV = L rrx 213
dx
The exact volume is the limit of this Riemann sum - that is, the definite integral.
Definite int egral equal to th e
V = fosx 2/ 3 dx volum e.
8 Evalu ate using the fundam ental
= rr. 15 x 5 13 1 0 = J(3
:,
2)rr - ~(O)rr
)
= 19.2rr theorem.
Write a real-world ans wer , suitabl y
= 60 .318 .. . "" 60.3cm 3 round ed, with uni ts.
Checks:
Volume of circumscribed cylinder would be rr(2 2 )(8) = 32rr > 19.2 rr . ./
Volume of ins cribed cone would be ( )rr (22 ) (8) = 10.66 ... rr < 19.2rr . ./
Numerical integration: V "" 19.20000037rr "" 19.2rr ./
Example 2 (Rotate a region around th e y-axis .) The region in Quadrant I bounded by the
parabola y = 4 - x 2 is rotat ed about the y-axis to form a solid paraboloid . Find
the volume of th e paraboloid if x and y are in inches. Show that your answ er is
reasonable.
386 Chapter
8: The Calculus
of PlaneandSolidFigures
Solution First sketch the region as shown in th e left-hand diagram of Figure 8-5c. Slice
the region into strips perp endicular to the axis of rotation (the y-axis, this time).
Show one representative strip with a sample point (x, y) on the graph. Then
sketch the solid that would result if this region is rotated about the y-axis (right
diagram) . As in Example 1, the strip generates a flat disk as it rotates . Let dV be
the volume of th e disk and proceed as follows .
y y
X X
Draw the region . Rotat e the region and th e strip
Slice a strip perp endicular to form a solid. The rotating
to the axis of rotation . strip forms a flat disk.
Figure 8-5c
Volum e = (area)(length ); the radius is x
dV = TTX2 d y thi s tim e.
= TT(4 - y)dy Get dV in term s of one variabl e.
4 4 Add volum es of slices and tak e th e
:. V r rr(4 - y) dy = rr (4y -
= Jo -
Y2 ) [
o limit (int egrate).
= rr (16 - 8) - rr (O - 0) = 8rr = 25.132 .. . :::,25.1 in 3
Checks : Volume of circumscribed cylinder would be rr(2 2 )(4) = 16rr > 8rr . ./
Volume of inscribed cone would be (1 / 3)rr(2 2 ) (4) = 5.33 ... rr < 8rr . ./
Numerical integration: V = Brr = Brr ./
Example 3 (Rotate a region bounded by two curves.) Let R be the region that is bounded by the
graphs of y 1 = 6e- 0 2x and y2 = -/x, and by the vertical lines x = 1 and x = 4. Find
the volume of the solid generated wh en R is rotated about the x-axis . Assume that
x and y are in feet. Show that your answer is reasonable .
Solution Sketch region R and slice it into strips perpendicular to the axis of rotation
(x-axis). Show a representative strip with two sample points, one on each graph
(Figure 8-5d, left). As R rotates, the strip will trace out a disk with a hole in the
middle (a washer, for those of you who are familiar with nuts and bolts) . The
volume dV of the washer will be the volume of the outer disk minus the volume
of the inner disk.
dV = rry f dx - TT Yi dx Volume of out er disk minus volum e of inner disk.
dV = rr(y f - Yi) dx Star t here if you 're bra ve enough!
dV = rr (36e - 04 x - x ) dx Substitut e for y 1 and Y2 and do th e squ aring .
on 8-5: Volumeof a Solid byPlaneSlicing
Secti 387
.'.V = TT r
(36e -0.4x - x ) dx Add th e volumes of the strips and find the limit.
Use num erical integration sin ce an "exact" answer
= 34.65811 .. . TT was not called for.
The rr could ha ve been includ ed in the num erical
108.9 ft 3
= 108.881. .. :::::: int egration .
X X
Draw the region. Slice a strip Rotat e the region and the strip
perpendicular to the axis of to form a solid. The rotating
rotation. Pick two sample strip forms a flat washer
points , (x, y1) and (x , Yz). of volume dV.
Figure 8-5d
Checks: Volume of outer cylinder would be TT ( 6e- 02 )2 (4 - 1) = 72.39 ... TT . ./
Volume of inner cylinder would be TT ( 1) 2 ( 4 - 1) = 3TT . ./
.'. volume of solid is bounded above by 69.39 ... TT 2: 34.658 . .. TT. ./
In Examples 1-3, the figure has been generated by rotating a region. Such a figure
is given the (obvious!) name solid of revo lution. As the strip rotated, it generated
a disk or washer. The same disk or washer would be generated if the solid were
generated first, then sliced with planes perpendicular to the axis of rotation. As a
result, the technique you have been using is called finding v olumes by plane slices .
Once you realize this fact, you can find volumes of solids that are not generated
by rotation. All you have to do is find the cross-sectional area in terms of the
displacement perpendicular to th e cross section . Example 4 shows how.
Example 4 (Plane slices of a noncircu lar solid) A 2 in. by 2 in. by 4 in . wooden block is carved
into the shape shown in Figure 8-5e. The graph of y = J;r--=x is drawn on the back
face of the block. Then wood is shaved off the front and top faces in such a way
that the remaining solid has square cross sectio ns perpendicular to the x-axis. Find
the volume of the solid . Show that your answer is reasonable.
Solution The right-hand diagram in Figure 8-5e shows a slab formed by slicing the
solid with planes perpendicular to the x-axis . Each such slab has a length
388 Chap
ter 8: TheCalcu
lusof PlaneandSolid Figures
2
...........
17 X
4
4 Slab of
volum e dV
Block Carve it int o a solid
with curved sides. Slice the solid into
slabs of volume dV.
Figure 8-5e
approximately equal to th e y-value at the sample point shown. Thus the cross-
sectional area of the slice is about equal to y 2 since the cross section is a
squar e. Using the fact that volume equals cross-sectional area tim es length (or
thickness, in this case), you can write th e following.
Do the geometry and
dV = y 2 dx = (4 - x) dx algebra .
2 3
. .V= fo4(4 - x) dx = 4x-x 1:= 16 - 8 - 0 + 0=8in Do the calculus .
Checks:
Volume of block is (2)(2)(4) = 16 > 8 . ./
Volume of inscribed pyramid would be (1 / 3)(2) (2) (4) = 5.33 ... < 8 . ./
Numerica l integration: V = 8 = 8 .I
All of the examples above involve th e same basic reasoning, as summarized here.
Volumeof a Solidby PlaneSlicing
Technique:
Cut th e solid into flat slices , formed either by strips in a rotated region or by planes
passed throu gh the solid. Get disks, washers, or slabs whose volumes can be found in
terms of the solid's cross section at sample point(s) (x, y).
Do geometry to get dV in terms of the sample point(s).
Do algebra to get dV in terms of one variab le.
Do calculus to ad d up all the dV's and tak e th e limit (that is, integrate).
Check your answer to mak e sure it's reasonabl e.
Problem Set 8-5
DoThese
Quickly
The following problems are intended to refresh your skills. You sho uld be able to do all
ten problems in less than five minutes.
Ql. Integrate: J(x 2 + x + 1) dx
Q2. Integrate: fx 314
dx
Q3. Differentiate: y = x 2 ! 3
Section of a SolidbyPlaneSlicing
8-5: Volume 389
Q4. Integrate: f e- dx
3x
QS. Integrate : f csc x cot x dx
Q6. Differentiate: y = ln 5x
Ql. Write the definition of definite integral.
QB. The conclusion, "... then there is a value of x = c in (a, b) such that f'(c)
= [f(b) - f(a)] / (b - a)" is for the - ?- theorem.
Q9. Sketch the graph of a function that is continuous at x = 4, but not differentiable there .
Q10. Sketch the graph of y = 2-x.
y
1. Parabo loid Problem: The region in the first quadrant under the
graph of y = 9 - x 2 is rotated about the y-axis to form a solid
paraboloid (Figure 8-Sf). Assume that x and y are in feet.
a. Find the exact volume of the paraboloid using the fundamental
theorem of calculus.
b. Find the approximate volume by integrating numerically.
X
c. Show that your answers are reasonable by doing a quick
geometrical check.
Figure 8-51
2. Cone Problem: A solid cone is formed by rotating about
y
the y-axis the first-quadrant triangle bounded by the line
y = 10 - 2x and the two axes (Figure 8-Sg). Assume that x
and y are in centimeters .
a. Find the exact volume of the cone using the fundamental
theorem of calculus.
b. Show that the volume in 2a agrees with the formula from
geometry.
c. How does the volume of the cone compare to the volume X
of the cylinder that can be circumscribed about it?
3. Let R be the region under the graph of y = 4x - x 2 from x = 1
to x = 4. Find the exact volume of the solid generated by Figure 8-Sg
rotating R about the x-axis.
4. Let R be the region under the graph of y = x.1 5 from x = 1 to x = 9. Find the exact
volume of the solid generated by rotating R about the x-axis.
5. Let R be the region in Quadrant I bounded by the graphs of y = ln x and y = 1. Find
the exact volume of the solid generated by rotating R about the y-axis.
6. Let R be the region bounded by the y-axis, the lines y = 1 and y = 8, and the curve
y = x 314 . Find the exact volume of the solid generated by rotating R about the y-axis .
390 Chapter 8: The Calculusof Plane and Solid Figur
es
JI 04 tix ,xr I , 9%tiS5
7. Washer Sli ces Problem: Figure 8-Sh shows the solid formed
)'
by rotating around the y-axis the region that is bounded by
the graphs of y = x~ and y = Bx. Find the exact volume of the 16
solid using the fundamental theorem of calculus. Assume that
x and y are in inches. Show that your answer is reasonable by I'
a quick geometrical check and by numerical integration. '
:'
'
8. Exponential Horn Problem: A horn for a public address /'.
system is to be made with the inside cross sections increasing
exponentially with distance from the speaker. The horn will
...
have the shape of the solid formed when the region bounded
by y = e0 Ax and y = x + 1 from x = 0 to x = 3 is rotated about
the x-axis (Figure 8-Si). Find the volume of the material used
to make this speaker. Assume that x and y are in feet. Show
that the exact answer by the fundamental theorem agrees
with the answer obtained by numerical integration.
9. The region between x = 0, x = 8 and bounded by the graphs
of y = x 113 and y = lOe- 0 i x is rotated around the x-axis to
X
form a solid. Find its exact volume.
2
10. The region bounded by the graphs of y = 4 - x and y = 4 - x 2
is rotated around the y-axis to form a solid. Find its exact Figure 8-5h
volume.
11. Paraboloid Volum e Formula Problem: Prove that the volume
of a paraboloid is always one-half the volume of the circum-
scribed cylinder. To do this, realize that any paraboloid is
congruent to a paraboloid generated by rotating a parabola
of general equation y = ax 2 about the y-axis .
12. Riemann Sum Limit Problem: The region in Quadrant I X
bounded by y = 0.3x1.s, x = 4, and the x-axis is rotated about
the x-axis to form a solid.
a. Find the volume of the solid by performing the appropriate
calculus.
b. Find three midpoint Riemann sums, M10 , M100 , and M100 0 ,
for the volume of the solid. Show that these sums are get-
ting closer to the exact value as the number of increments
increases.
Figure 8-5i
13. Different Axis Problem I: The region in Quadrant I under the
graph of y = 4 - x 2 is rotated around the line x = 3 to form a
solid (Figure 8-Sj). Sketch the washer formed as the horizontal 4
slice shown rotates. What are the inner and outer radii of the
washer? Find the volume of the solid.
14. Different Axis Problem II: The region in Quadrant I under the
(0, y)
-
(x, y )
Inner :
radiu ~
'
graph of y = 4 - x 2 shown in Figure 8-Sj is rotated around
the line y = - 5 to form another solid. Find its volume.
.....
...,.~
:...-:.::::~
..:t.......... X
z-'4 6
Figure 8-5j
Section of a SolidbyPlane
8-5:Volume Slicing 391
15. New Integra l Prob lem I: The region under the graph of y = sinx from x = 0 to x = 1.2
is rotated about the x-axis to form a solid.
a. Write an integral for the volume of this solid.
b. Evaluate the integral approximately by numerical integration.
c. Transform the integrand using a clever application of the double-argument
properties from trigonometry so that it is linear in sine or cosine of a multiple of
x. Then find the exact volume by evaluating the integral using the fundamental
theorem. Show that the answer you got in 15b is close to this exact value.
16. New Integral Prob lem II: The region bounded by the graph of y = tanx, the line y = 1,
and the y-axis is rotated about the x-axis to form a solid.
a. Write an integral for the volume of this solid.
b. Evaluate the integral approximately by numerical integration .
c. See whether you can get the indefinite integral and thus find the exact value of the
volume .
17. Pyramid Prob lem : A pyramid has a square base 8 cm by 8 cm and altitude 15 cm (Figure
8-5k). Each cross section perpendicular to the y-axis is a square. Find the volume of the
pyramid. Show that it is equal to one-third the volume of the circumscribed rectangular
solid with the same base and altitude .
18. Horn Prob lem: Figure 8-51 shows a horn-shaped solid formed in such a way that a plane
perpendicular to the x-axis cuts a circular cross section . Each circle has its center on
the graph of y = 0.2x 2 and a radius ending on the graph of y = 0.16x 2 + l. Find the
volume of the solid if x and y are in centimeters.
y
Figure 8-5k Figure 8-51
19. Triangu lar Cross-Section Problem : Region R under the graph y
of y = x06 from x = 0 to x = 4 forms the back side of a solid
(Figure 8-5m). Cross sections of the solid perpendicular to
the x-axis are isosceles right triangles with their right angle
on the x-axis .
X
a. Find the volume of the solid.
b . Quick! Tell what the volume of the solid would be if the
cross sections were squares instead of right triangles . Figure 8-5m
392 Chapte
r 8: TheCalc
ulusof PlaneandSolidFigures
kffiLl t!M
20. Wedge Problem: Figure 8-Sn shows a cylindrical log of radius 6 in . A wedge is cut from
the log by sawing halfway through it perpendicular to its central axis, then sawing
diagonally from a point 3 in. above the first cut. Your job is to find the volume of the
wedge .
(x,y)
-
(x,z)
Cut a wedge . Slide it out. Slice the wedge.
Figure 8-5n
a. Write the equation of the line that runs up the top surface of the wedge (in the
xy-plane).
b. Write the equation of the circle in the xz-plane that forms the boundary for the
bottom surface of the wedge.
c. Slice the wedge into slabs using planes perpendicular to the x-axis. Find an equation
for the volume dV of a repr esentative slab. Use th e result to find th e volume of the
wedge.
21. Generalized Wedge Problem : Find a formula for the volume of a wedge of altitude h cut
to the central axis of a log of radius r, as in Figure 8-Sn, where one cut is perpendicular
to the axis of the log.
22. Cone Volume Formula Proof Problem: Prove that the volume
of a right circular cone of base radius r and altitude his given y
by
V = rrr2 h.
You may find it helpful to draw the cone in a Cartesian
coordinate system . If you put either the vertex of the cone or
the center of its bas e at the origin, a slanted element of the
cone will be a line segment whose equation you can find in
terms of x and y .
23. Sphere Problem: Figure 8-50 shows a sphere of radius 10 cm.
a. Find the volume of the spher e by calculus.
b. Show that the answer you got in 23a agrees with the Figure 8-50
formula that you learned in geometry.
ion8-5:Volume of a Solid byPlaneSlicing
Sect 393
;;;;
24. General Volume of a Sphere Problem: By calculus, derive the formula
V = 13 rrr 3
for the volume of a sphere of radius r.
25. Volume of an Ellip soid Problem: Figure 8-5p shows the
ellipsoid
Find the volume of the ellipsoid in terms of a, b, and c, the a
radii along the three axes. To get dV, show that every cross
section perpendicular to the x-axis is an ellipse similar in
proportions to the ellipse in the yz -plan e. Use the fact that
the area of an ellipse is rruv, where u and v are the radii Figure 8-5p
alon g the major and minor axes .
26. Highway Cut Problem: Figure 8-5q shows a cut through a hill that is to be made for a
new highway. Each vertical cross section of the cut perpendicular to the roadway is an
isosceles trapezoid whose sides make angles of 52 with the horizontal. The roadway
at the bottom of the trapezoid is 50 yd wide . The cut is 600 yd long from its beginning
to its end .
y
X
~, ~ X y X y
0 0 330 47
30 5 360 49
End of cut
60 9 390 51
t Beginning of cut 90 15 420 46
Highway cut through hill 120 28 450 39
150 37 480 37
180 42 510 31
Isosceles trapezoid 210 42 540 22
240 48 570 14
Depth
270 53 600 0
300 52
50 yards
Vertical cross section
Figure 8-5q
You are to find how many cubic yards of earth must be removed in order to construct
the cut. To estimate this volum e, survey crews have measured the depth, y, of the cut
at various distances, x, from the beginning. The table above shows these depths. Write
an integral involving y and dx that represents the volume of earth to be removed . Then
evaluate the integral by a suitable numerical technique. If earth removal costs $12 .00
per cubic yard, about how much will it cost to make the cut?
27. Submarine Problem: The countries of Parra and Noya are spying on each other. The
Noyaks find that the Parrians are designing a new submarine, the Black November.
394 Chapter
8: TheCalculus
of Plane
andSolidFigures
The cross sections perpendicular to the horizontal axis of the sub will be circles with
centers on that axis . The only quantitative piece of information the Noyaks have found
is that the radius y at any distance x from the bow of the sub is given by
y = 2x 0 5 - 0.02x i.5 ,
where x and y are in meters. Your mission is to find out as much about the submarine
as possible from this information .
a. Plot the graph of the radius as a function of distance from the bow . Sketch the
result .
b . The sub will end where the equation indicates that the radius becomes negative .
How long will the Black November be? How does this compare to the length of a
football field?
c. What is the beam (the maximum diameter) of the sub? How far from the bow is this
maximum diameter?
d. Fast submarines have a length -to -beam ratio of 7 or more. Do you expect the Black
November to be fast or slow?
e. What will be the volume of the submarine?
f. The displacement of a ship is the number of tons the ship weighs . [The term
displacement is used because a (floating) ship weighing T tons will displace T tons
of water .] Given that a cubic meter of seawater is about 1042 kg, how many metric
tons will the Black November displace? (A metric ton is 1000 kg.)
y
*28. Preview Prob lem: Figure 8-Sr shows the solid formed by rotating a
region in Quadrant I about the y -axis. The region has been sliced
into strips parallel to the axis of rotation, rather than perpendicular
to it. Sketch the geometrical figure formed by the strip shown in
the figure. What name could be given to this figure? See whether
you can find an expression for dV, the volume of the figure formed
by rotating the strip, in terms of the sample point (x., y).
Figure 8-5r
*This problem prepares you for Section 8-6.
Section8-5: Volumeofa Solid byPlane Slicing 395
8-6 Volume of a Solid of Revolution by
Cylindrical Shells
Figure 8-6a shows the region under the graph of y = 4x - x 2 from x = 0 to x = 3.
Suppose that this region is to be rotated about the y -axis to form a solid . Slicing the
region perpendicular to the y-axis as you did in Section 8-5 wou ld be awkwar d. As
shown on the left in Figure 8-6a, the lengths of the strips are not given by a single
rule for all values of y. As a res ult, you have to consider two parts of the solid , one
from the x-axis up to the cusp and the other from the cusp up to the vertex of th e
parabola .
y ~-~- Length =
cur ve- line
(always)
X X
Awkward to slice perpendicular Appropriate to slice para llel
to the y-axis to the y-axis
Figure 8-6a
If you slice the region into strips parall el to the y-axis, the length of the stri p is
always equal to (curve - line). As shown in Figur e 8-6b, these paralle l strips will
generate cylindrical shells as the region rotates.
dx
X .................................. .........................................
,,._ _ _____ 2nx - ------
Draw the solid. Draw the shell Roll th e shell out flat into a rectangu lar solid.
The rotatin g strip forms with sample point Length = circumference of shell.
a cylindrical shell . (x,y). Width= altitude of shel l. Thickness = dx
Figure 8-6b
OBJECTIVE Find the volum e of a solid of revolution by slicing it into cylindri cal shells.
396 Chapter
8: TheCalculus
of PlaneandSolidFigures
The shells in Figure 8-6b are like tin cans without ends . Since a shell is thin, its
volume dV can be found by cutting down its side and rolling it out flat (Figure
8-6b, right). The resulting rectangular solid will have the following approximate
dimensions.
Length: Circumference of the shell at the sample point (2rrx, in this case)
Width: Altitude of the shell at the sample point (y, in this case)
Thickness: Width of the strip (dx, in this case)
Consequently, the volume of the shell is given by the following property .
Property:Differentialof Volumefor Cylindrical
Shells
dV = (circumference) (altitude) (thickness)
The volume of the solid will be approximately equal to the sum of the volumes of
the shells (Figure 8-6c). The exact volume will be the limit of this sum-that is, the
definite integral. The innermost shell is at x = 0, and the outermost is at x = 3. Thus
the limits of integration will be from O to 3. (The part of the solid from x = -3 to
x = 0 is just the image of the region being rotated, not the region itself.) Example 1
shows the details of calculating the volume of this solid.
Inner shell
t
Outer shell
is at x = 0 is at x = 3
Figure 8-6c
Example 1 The region under the graph of y = 4x - x 2 from x = 0 to x = 3 is rotated about the
y-axis to form a solid. Find the volume of the solid by slicing into cylindrical shells.
Use the fundamental theorem to obtain the exact answer. Show that your answer is
reasonable .
Solution The volume of a representative cylindrical shell is
Recall the rolled-au t sh ell in Figure
dV = (circumference) (altitude) (thickness)
8-6b .
= (2rrx) (y)(dx).
Substituting 4x - x 2 for y gives
dV = 2rrx(4x - x 2 ) dx = 2rr(4x 2
- x 3 ) dx.
Section
8-6:Volume
of a Solid of Revo
lutionbyCylindricalShe
lls 39 7
The volume is found by adding all the dV's and taking th e limit.
V = J 2rr(4x
3
0
2
- x 3 ) dx
3
) I
3 4
= 2rr (ix
3
!x
4
-
o
= 2rr (36 - --
0 + 0)
= 31.Srr = 98.96 ....
Checks:
Volume of circumscribed cylinder is rr (3 2 ) (4) = 36rr > 31.Srr. v"
Numerical inte gration: Integral = 31.Srr v"
In case you are wondering whether th e distortion of th e shell as you roll it out flat
causes the final answer, 31.Srr, to be inaccurat e, the answer is, No. As t,.x approaches
zero, so do the inaccuracies in the shell approximation. In Section 11-7, you will
learn that if the approximate value of dV diff ers from the exact value by nothing
more than infinit esimals of higher order, for instan ce (dx )(dy ), then the integral will
give the exact volume.
Cylindrical shells can be used to find volumes when these conditions are encoun-
tered.
The rotation is not around the y-axis.
The axis of rotation is not a bound of integration.
Both ends of the shell's altitude are variable.
Example 2 shows how this can be don e.
Example 2 Let R be th e region bound ed below by the graph of y = x 112 , abo ve by th e graph of
y = 2, and on th e left by the graph of y = x. Find the volume of the solid generated
when R is rotated about the line y = - l . Assume that x and y are in feet. Show that
your answer is reasonable.
Solution First draw the region, as in th e lef t-hand diagram of Figure 8-6d . Find the points
of intersection of the graphs. Slice parallel to th e axis of rotation. Mark the two
resulting sample points as (x1, y) and (x 2 , y) . Then rotate the region about the
line y = - l. As shown in the center diagram of Figure 8-6d, it helps to draw
only th e back half of th e solid. Otherwise the diagram becomes so cluttered it
is hard to tell which lines are which. Roll out the shell and find dV.
dV = (circumference) (altitude) (thickness)
Altitud e is always larger value minu s
smaller value. Radius is th e difference
between the y-values, y - (- 1) = y + 1.
For th e curve y = x 112 , solve to get x 1 = y 2 .
For the line y = x, "solve" to get X2 = y.
2
:. d V = 2rr (y + l )(y y) d y
r
-
2
:. V = rr 2(y + 1) (y - y) dy Inn ermost shell is at y = l ; out erm ost shell is at y = 2.
= 4.5rr Integrate num erica lly or algebraically.
= 14.1371 . . . "" 14.lft 3
398 Chapter
8: TheCalculus
of PlaneandSolidFigures
X X
4
2rr(y + 1)
-1
Draw the region. Slice Rotate about the x-axis. Show only Roll out the (whole)
parallel to axis of rotation. the back half of the solid . The rotating shell into a flat
Show two sample points. strip generates the cylindrical she ll. rectangular solid.
Figure 8-6d
Check:
Outer cylinder - inner cylinder = rr (32 ) (3) - rr(2 2 ) (3) = 15rr > 4.5rr ./
Problem Set 8-6
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Sketch the graph : y = x 2
Q2. Sketch the graph: y = -x 2
Q3. Sketch the graph: y = x- 2
y
Q4. Sketch the graph : y = zx
QS. Sketch the graph: y = 2-x
Q6. Sketch the graph: y = 2x X
Ql. Sketch the graph : y = In x 2
QB. Sketch the graph of a continuous function whos e derivative -1
is shown in Figure 8-6e.
Q9. y = x 3 - 3x has a local minimum at x = - 7 - .
Q10. f sec 2 xdx = -?-. Figure 8-6e
Section
8-6:Volume
of a Solidof Revolution
byCylindrical
Shells 399
l. Figure 8-6f shows the solid formed by rotating about the y
y-axis the region in Quadrant I under the graph of y = 4 - x 2 .
a. Find the volume dV of a cylindrical shell. Transform dV
so that it is in terms of one variable.
b. Find the exact volume of the solid by using the fundamen-
tal theorem.
c. Find the volume again by plane slicing . Use planes perpen-
dicular to the y-axis to form slabs of thickness dy . Show X
that you get the same answer as in 1b.
2. Figure 8-6g shows the solid formed by rotating around the
x-axis the region under the graph of y = x 213 from x = 0 to
Figure 8-6f
X = 8.
y
a. What is the altitude of the cylindrical shell in terms of the
sample point (x, y)?
b. Find the volume dV of a cylindrical shell. Transform dV
so that it is in terms of one variable.
c. Find the exact volume of the solid using the fundamental X
theorem .
d. Find the volume again, by plane slicing . Is the answer the
same as in 2c?
Figure 8-6g
For Problems 3-18, find the volume of the solid by slicing into cylindrical shells . You may
use numerical integration . Use familiar geometric relationships to show geometrically
that your answer is reasonable.
3. Rotate around the y-axis the region under the graph of y = -x 2 + 4x + 3 from x = l to
X = 4.
4. Rotate around the y-axis the region under the graph of y = x 2 - Bx + l 7 from x = 2 to
X = 5.
5. Rotate about the x-axis the region bounded by the y-axis and the graph of
X = -y 2 + 6y - 5.
6. Rotate about the x-axis the region bounded by the y-axis and the graph of
x = y2 - lOy + 24. y
8 t--1r--t---
7. Rotate around the y-axis the region above the graph of y = x3 that
is bounded by the lines x = l and y = 8 (Figure 8-6h).
Figure 8-6h
400 Chapter
8: TheCalcu
lusof Plane
andSolid Figures
8. Rotate around the y-axis the region in Quadrant I above the graph of y = 1/ x that is
bounded by the lines y = 4 and x = 3.
9. Rotate around the x-axis the region in Quadrant I above the graph of y = 1/ x 2 that is
bounded by the lines x = 5 and y = 4 (Figure 8-6i).
10. Rotate around the x-axis the region in Quadrant I below the graph of y = x 213, above
the line y = 1 and bounded by the line x = 8.
11. Rotate around the y-axis the region bounded by the graph of y = x 2 - 6x + 7 and the
line x - y = - 1 (Figure 8-6j).
4 f--+------+-
X
X
Figure 8-6i Figure 8-6j
12. Rotate around the x-axis the region in Quadrant I bounded by the
y
graph of y = x 113 and the line y = O.Sx - 2.
13. Rotate around the line x = 5 the region under the graph of y = x 312
from x = 1 to x = 4 (Figure 8-6k).
14. Rotate around the line x = 3 the region under the graph of y = x - 2
from x = 1 to x = 2.
15. Rotate around the line x = 4 the region bounded by the graph of X
y = x 4 and the line y = Sx + 6. 4 5
16. Rotate around the line x = -1 the region bounded by the graph of Figure 8-6k
y = -Jx.and the lines x + y = 6 and x = 1.
17. Rotate around the line x = - 2 the region bounded by the graphs of
y = - x 2 + 4x + 1 and y = 1.4x (Figure 8-61).You will need to find y
one of the intersections numerically.
18. Rotate about the line y = - 1 the region in Figure 8-61 from Prob-
lem 17. Tell why it would not be appropriate to find the volume of
this figure by cylindrical shells.
-2
Figure 8-61
Sect
ion8-6:Volume
of a Solidof Revolution
by Cylindrical
Shells 401
For Problems 19 and 20, find the volume of the solid by slicing into plane slabs, thus
verifying the answer obtained by cylindrical shells.
19. Use the solid given in Problem 7. 20. Use the solid given in Problem 8.
21. Limit of Riemann Sum Problem: The region under the graph of y = x 113 from x = 0 to
x = 8 is rotated around the x-axis to form a solid . Find the volume exactly by slicing
into cylindrical shells and using the fundamental theorem. Then find three midpoint
Riemann sums for the integral, using n = 8, n = 100, and n = 1000 increments. Show
that the Riemann sums approach the exact answer as n increases.
y
22. Unknown Integral Problem: Figure 8-6m shows the region
under y = sin x from x = 0 to x = 2, rotated about the y-axis
to form a solid.
a. Write an integral for the volume of this solid using
2
cylmdrical shells. Evaluate the integral numerically.
b. Explain why the integral cannot be evaluated by the
fundamental theorem using the techniques you have Figure 8-6m
learned so far.
23. Parametric Curve Problem: Figure 8-6n shows
the ellipse with parametric equations
x = Scos t y
3
y = 3sin t.
a. Slice the region horizontally, then rotate it
about the x-axis to form an ellipsoid. Find X
the volume of the ellipsoid by first writing 5 17
dV in terms of the parameter t . '
b. Slice the region vertically, then rotate it I
'
about the x-axis to form the same ellipsoid. I
Show that you get the same volume.
c. Find the volume of the solid generated by Figure 8-6n
rotating the ellipse around the line x = 7.
24. Journal Problem: Update your journal with things you've learned since the last entry.
You should include such things as those listed here.
The one most important thing you have learned since the last journal entry
The basic concept from geometry that is used to find volumes by calculus
The similarities of slicing into disks, washers, and other plane slices
The difference between plane slicing and cylindrical shells
Any techniques or ideas about finding volumes that are still unclear to you .
40 2 Chapter
8: TheCalcu
lus of Plane andSolid Figures
8-7 Length of a Plane Curve- Arc Length
At the beginning of this chapter you were introduced to th e geometry of plane and
solid figures, such as the bell-shaped solid shown in Figure 8-7a. You can now find
the volume of a solid by plane slices or cylindrical shells and the area of a plane
region . In the next two sections you will find the length of a curved line and the area
of a curved surfac e in space.
Figure 8-7a
OBJECTIVE Give the equaUon for a plane curve, find its length approximately by calculating and
summing the lengths of the chords, or exactly by calculus.
Example 1 Find approximately the length of th e parabola y = x 2 from x = - 1 to x = 2 (Figure
8-7b, left).
f(x) f(x)
X
-1 2 -1 2
Figure 8-7b
Solution The right-hand diagram in Figure 8-7b shows three chords drawn to the graph .
The sum of the lengths of the chords is approximately the length of the graph .
By th e Pythagorean theorem,
L "" -/2.+ -/2.+ M = 5.99 0704 . . . "" 5.99 units. Expl ai n why.
In general, the length of any one chord will b e
6.L = ~ t,,x 2 + t,,y z_ ,J6x 2 + 6y 2 m eans .J(6x) 2 + (6y) 2.
Section
8-7: Length
of a PlaneCurve---Arc
Length 403
Using smaller values of L'>x(that is, a greater number n of chords), the following is
true.
n = 30: L "' 6.1241726 9 .. . By th e pro gram of Problem 33 in Problem Set 8-7.
n = 100 : L "' 6.12558677 . . .
n = 1000: L "' 6.12572522 .. . The valu es are appro achin g a limit!
The limit of the sums of the chord lengths equals the exact length of the curve.
Property:Lengthof a PlaneCurve(ArcLength)
A curve between two points in the xy-plane has length
L = lim
t.x - 0,t.y-O
I ~tix 2 + tiy 2,
provided that this limit exists .
The limit of a chord length sum can be found exactly by transforming it to a Riemann
sum,
Lg (c )L'>x.
The first thing to do is make the factor L'>xappear. Although L'>x2 is not a factor of
both terms in the expression L'>x2 + L'>y2, it can still be factored out.
2
2 2
,'.',.x + ,'.',.y = [ 1 + ~: : J,'.',.x2 = [ 1 + ( ~:) ] L'>x2
So a chord length sum can be written
The remaining radicand contains L'>y/ L'>x,which you should recognize as the
difference quotient that approaches f '( x ) as L'>xapproaches zero. In fact, if f is a
differentiable function, the mean value theorem tells you that there is a number
x = c within the interval where f ' (c) is exactly equal to L'>y/ L'>x(Figure 8-7c). So a
chord length sum can be written
I ~ l + f'(c ) L'>x,
2
where the sample points, x = c, are chosen at a point in each subinterval where the
conclusion of the mean value theorem is true. The length L of the curve is thus the
X
limit of a Riemann sum and hence a definite integral.
C
Figure 8-7c L = I:~l + f'(x) 2 dx The exac t length of th e cur ve!
The quantity dL = ~ l + f' (x ) 2 dx, the differential of curve length (often called "arc
length"), can be written in a form that is easier to remember and use. Recalling
that f' (x ) = dy / dx, and that the differentials dy and dx can be written as separate
quantities, you can write
404 Chapter
8: TheCalculus
of PlaneandSolidFigures
FF3NfG$ iWWPffirtifS#54ihWi3?fiBH
!i@ 55J!5l!FI WM
dL = J 1 + ( ~)
2
dx, or, more simply,
dL = ~dx 2 + dy 2. Differe ntial of arc length .
It is easy to remember dL in the last form above because it looks like the Pythagorean
theorem. There are also some algebraic advantages of this form, as you will see in
later examples .
Example 2 Write an integral to find the exact length of the curve in Example 1 and evaluate it.
Solution y = x 2 ~ dy = 2x dx
dL = ~ dx 2 + (2x dx )2 = .J1 + 4x 2 dx
.'. L = f 2
- J
.J1 + 4x 2 dx
When you study trigonometric substitution in Chapter 9, you will be able
to evaluate integrals like this using the fundamental theorem . Numerical
integration gives
6.1257266 . .. .
Equations for more complex curves can be written in parametric form. Example 3
shows how this can be done.
Example 3 The ellipse in Figure 8-7d has parametric equations
X = 6 + 5 COS t
y
y = 4 + 3 sint .
Write an integral equal to the length of the graph and evaluate it numerically. Check
your answer for reasonability.
Figure 8-7d
Solution x=6 +5c ost ~ dx = -5sintdt
y = 4 + 3 sin t ~ dy = 3 cost dt
dL = ~ dx 2 + dy 2
= ~(- 5sintdt)2 + (3costdt) 2
= .J25 sin 2 t + 9 cos 2 t dt Explain th e "dt."
2 rr
L = J0 .J25sin t + 9cos 2 tdt
2
,:::;25.52699 ...
Check: A circle of radius 4 has length 2rr 4 = 25.13274 . .. . ./
The indefinite integral in Example 3 cannot be evaluated using any of the elementary
functions. It is called an elliptic integral, which you may study in later courses.
Sect
ion8-7: Length
of a PlaneCurveArc
Length 40 5
Occasionally a curve length problem will involve an integral that can be evaluated
by the fundamental theorem . Example 4 shows one instanc e.
Example 4 Plot the graph of y = x 312. Find exactly the length from x = 0 to x = 9.
Solution The graph is shown in Figure 8-7e. It starts at the origin and rises gently to the
point (9, 18).
dy = x 112 dx
dL = ~dx 2 + dy 2
= ~ dx 2 + (x l/ 2 dx) 2
= Jf+x dx
r r
18 y
112
L = Jf+xdx = (1 + x) dx
9
=.?.(l + x) 3/21 =.?.(103i2)_ .?.(l 312)
3 O 3 3
X
= 00 3/2 - 1) Exact length .
9
= 20.41518 .. . Appro xim ation for exac t length .
Figure 8-7e
Problem Set 8-7
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten prob lems in less than five minutes .
QJ. Sketch: y = x 2
Q2. Show the region under y = x 2 from x = 1 to x = 4. y
Q3. Write an integral for the area of the region in Problem Q2.
Q4. Do the integration in Problem Q3.
QS. Evaluate the integral in Problem Q4.
Q6. Sketch the solid generated by rotating the region in Problem Q2
about the y-axis.
Ql. Write an integral for the volume of the solid in Problem Q6. :x
a C b
QB. Do the integration indicated in Problem Q7.
Q9. Evaluate the integral in Problem Q8. Figure 8-7f
Q10. Figure 8-7f illustrates the -? - theorem .
For Problems 1-4,
a. Sketch the graph for x in the given interval.
b. Find its approximate length using five chords with equal values of 6.x.
c. Find its length more precisely using a definite integral evaluated numerically .
406 Chapter
8: TheCalculus
of Plane
andSolidFigures
MM
*"*'e 6@
1. y = eX XE [0, 2] 2. y = 2x X E [0, 3 ]
3. y = tan x x E [0, 1. 5] 4. y = secx x E [O, 1.5]
For Problems 5-16,
a. Plot the graph for x in the given inter val. Sketch the result.
b. Find the approximate length using a definite integral evaluated numerically.
C. Show that your answer is reasonable.
5. y = x 2 - 5x + 3 X E [1, 6] 6. y = 4x - x 2 X E [0, 4]
7. y = 16 - x 4 X E [- 1, 2] 8. y = x 3 - 9x 2 + 5x + 50 X E [- 1, 9]
2
9. y = (lnx) X E [0 .1, e] 10. y = x sinx X E [0, 4rr]
11. y = tanx X E [0, 1.5] 12. y = secx X E [0, 1.5]
13. Astroid: t E [0, 2rr] 14. Cardioid: t E [0, 2rr]
x = 5 cos 3 t x = 5(2cost - cos 2t)
y = 5sin 3 t y = 5 (2 sin t - sin 2 t)
15. Epicycloid: t E [0,2rr] 16. Involute of a circle: t E [0, 4rr]
x = 5 cost - cos St X = COS t + t Sin t
y = 5 sin t - sin St y = sin t - t cos t
For Problems 17-20,
a. Plot the graph for x in the given int erval. Sketch th e result.
b . Find the exact length using a definit e integral evaluated by the fundamental theorem.
c. Show that your answer is reasonable. (For Problem 19, find a common denominator
under the radical sign .)
3
17. y = 4x 312 X E [0, 4] 18. y = ux + x1 XE [l, 2]
19. y = 3x 213 + 5 X E [l, 8] 20. (xz + 2)3/2 X E [0, 3]
21. Golden Gate Bridge Problem: The photograph
shows the Golden Gate Bridge across San Fran-
cisco Bay in California. The center span of th e
bridge is about 4200 ft long. The susp ension ca-
bles hang in parabolic arcs from towers about 750
ft above the water's surface. These cables come
as close as 220 ft to the water at the center of the
span. Use this information to write an equation
of the particular quadratic function expressing
the distance of the cables from the water as a
function of the horizontal displac em ent from
center span. Use the equation to calculate the
length of the parabolic cable.
Section
8-7:Lengthof a PlaneCurve-
--ArcLength 407
ffli.?
Lfil>
22. Chain Prob lem: When a chain hangs under its own weight, its shape is a catenary
(which comes from the Latin word for "chain") . Figure 8-7g shows a catenary with its
vertex on the y-axis. Its equation is
y = 0.2(e x + e- x ),
where x and y are in feet. Find the length of this chain from x = - 4 to x = 4. How does
this length compare with that of a parabola,
y = ax 2 + c,
which has the same vertex and endpoints?
Parking
lot
Seating area
Pla)~J1g field
Figure 8-7g Figure 8-7h
23. Stadium Prob lem: Figure 8-7h shows the seating area for a sports stadium. The ellipses
have the following parametric equations .
Outer Ellipse: Inner Ellipse:
X = 120 COS t x = lOOcos t
y = lOOsint y = 50sint
Both x and y are in meters. Find the lengths of the boundaries between the outer
ellipse and the parking lot, and between the inner ellipse and the playing field.
24. Parabola Surprise Problem! A parabola has parametric equations
x = 8 cos 2t
y = 5 sint.
Find the length from t = 0 tot = 2rr . Why does the answer seem unreasonabl y high ?
25. Imp licit Relation Prob lem I: Use the fundamental theorem to find exactly the length of
the graph of 9x 2 = 4y 3 between the points (0, 0) and (213 , 3). Consider y to be the
independent variable.
26. Imp li cit Relation Problem II: Use the fundamental theorem to find exactly the length
of the sernicubical parabola x 2 = y 3 between th e points ( - 1, 1) and (8, 4). Consider
y to be the independent variable. You will have to break the graph into two branches
(sketch a graph).
408 Chapter
8: TheCalculus
of PlaneandSolid Figures
27. Spiral Problem: Figure 8-7i shows the spiral whose parametric
equations are
x = tcost X
y = tsint.
What range oft generates the part of the spiral shown in the figure?
Find the length of the spiral by the fundamental theorem if you
can, or by numerical methods. Figure 8-7i
28. Length of a Circle Problem: Write parametric equations for a circle of radius r centered
at the origin . Then use appropriate algebra, trigonometry, and calculus to prove the
familiar circumference formula C = 2rrr.
29. Sinusoid Length Investigation Problem: Write an integral for the length of one cycle of
the sinusoid of constant amplitude A,
y = A sinx .
Find lengths of the sinusoid for various values of A. From the results of your work, try
to reach a conclusion about how the length varies with A. For instance, does doubling
A double the length?
30. Ellipse Length Investigation Problem: Write an integral for the length of the ellipse
X = COS t
y = A sint.
Find lengths of the ellipse for various values of A From the results of your work, try
to reach a conclusion about how the length varies with A. For instance, does doubling
A double the length ?
31. Fatal Error Problem: Mae Dan error wants to find the length of y = (x - 2 )- 1 from x = 1
to x = 3. She partitions [1, 3] into five equal subintervals, and gets 18.2774 ... for the
length. Explain to Mae why she has made a fatal error in her approach to the problem.
32. Mistake Problem: Amos Take finds the length of the curve y = sin 2rrx from x = 0
to x = 10 by dividing the interval [O, 10] into 5 subintervals of equal length. He gets
an answer of exactly 10. Feeling he may have made a mistake, he tries again with 20
subintervals, and gets the same answer, 10. Show Amos that he did make a mistake .
Show him how he can get a quite accurate answer using only 5 subintervals.
33. Program for Arc Length by Brute Force: Write a program to calculate the approximate
length of a curve by summing the lengths of the chords. The equation for the function
can be stored in the y = menu . The program should allow you to input the lower
and upper bounds of the domain and the number of increments to be used. The
output should be the approximate length of the curve . To make the program more
entertaining to run, you might have it display the increment number and the current
sum of the lengths at each pass through the loop . You may assume that your program
is working correctly if it gives 6.12417269 ... for the length of y = x 2 from x = -1 to
x = 2 (Example 1) with n = 30 increments.
Section8-7: Lengthof a PlaneCurve--Arc
Length 409
8-8 Area of a Surface of Revolution
Suppose that the graph of a function y = f (x) is rotated around the x-axis . The
result will be a surface in space (Figure 8-8a, left). You are to find the area of the
surfac e.
y
y
X X
Rotat e a curved graph, Rotat e (straight) chords ,
get a doub ly curved surface . get singly curved surfaces.
Figure 8-80
A graph cur ved in one direction that rotates in another direction forms a doubly
curve d surface . Like a map of the Earth, a doubly curved surface cannot be flattened
out. But if you draw chords on the graph as you did for finding arc length, the
rotating chords generate singly curved frustums of cones, as shown in the right-
hand diagram in Figure 8-8a. The frustums can be flattened (Figure 8-8b), allowing
you to find their areas by geometry.
y
Frustum is Roll it
singly cured out flat.
~
X .
Rotat ed chords form One of th e A sin gly cur ved surface
frustums of cones. frustums. can be roll ed out flat.
Figure 8-8b
OBJECTIVE Find the area of a surface of revolution by slicing the surface into frustums of cones.
4 10 Chapter
8: TheCalculus
of Plane
andSolidFigures
The surface area of a cone is S = rr RL, where R is the base radius, and L is the slant
height (Figure 8-8c). The area of a frustum is the area of the big cone minus the area
of the small one; that is,
S = rrRL - rrrl,
where R and L are for the large cone and r and l are for th e small one. By clever
algebra (which you will be asked to do in Problem 26), this equation can be
transformed to
S = 2rr ( R; r)(L -1).
The quantity (R + r) / 2 is th e average of the two radii. The quantity (L - l) is the
slant height of th e frustum. It is the same as DI in the arc length problems of Section
Figure 8-8c 8-7. So the differential of surface area, dS, is
dS = 2rr (average radius)(slant height) = 2rr (average radius) dL.
Note that 2rr(average radius) equals the distance travel ed by the midpoint of the
chord as the chord rotat es around.
Property:Area of a Surfaceof Revolution
If y is a differentiable function of x, then the area of the surface formed by rotating the
graph of the function around an axis is
S = J:(circumference) dL = 2rr J:(radius) dL,
where dL = ~ dx 2 + dy 2 and a and b are the x- or y-coordinates of the endpoints of the
graph .
The radius must be found from information about the surface .
Example 1 The grc1ph of y = sin x from x = 1 to x = 3 is rotated around various axes to form
surfaces. Find th e area of the surface if the graph is rotat ed about
a. the y-axis
b . the line y = 2
Show that your answers are reasonable.
Solution a. Figure 8-8d shows the surface for the graph rotated around the y-axis.
dy = cosxdx
.Jdx 2 + cos 2 x dx 2 = .J1 + cos 2 x dx
Radius = x, so
dS = 2rrx .Jl + cos 2 x dx .
.'. S = 2rr r x.Jl + COS 2 X dx .
By numerical integration,
S ~ 9.5111282 .. . rr = 29.88009 .. . .
Section
8-8:Areaofa Surface
of Revolution 411
7
/i.
.
As a check on this answer, cons ider the area of a flat washer of radii 1 and 3
(Figure 8-Se). Its area is
rr (3 2 - 12 ) = 25 .132 .. . .
So the 29.88 . .. answer is reasonable .
Figure 8-8d Figu re 8-8e
b . Figure 8-Sf shows the surface for rotation around y = 2. Note that dL is the
same as in part a. Only the radius is different.
radius = 2 - y = 2 - sinx
So the surface area is
S = 2rr r (2 - Sinx) ,/ 1 + COS2 X dx .
By numerical integration,
S :::::5.836945 ... rr = 18.337304 . . . .
As a reasonability check, the answer shou ld be a bit more than a cylinder of
altitude 2 and radius 1 (Figure 8-8g). That area is
2rr(1 2 )(2) = 12.566 . . . ,
which is in the ballpark.
1-- 2- 1
X X
3 3
Figure 8-8f Figure 8-89
412 Chapter
8:TheCalculus
ofPlaneandSolid Figures
Problem Set 8-8
DoTheseQuickly
The following problems are intended to refresh your skills . You shoul d be able to do all
ten problems in less than five minutes.
QJ. If y = x 3 , then dI (arc length) = -?- .
Q2. If y = tanx, then dI = -?-.
Q3. Jsin 5 x cos x dx = -?- .
Q4. Ji5x 3 dx = -?-.
QS. If y = xeX, then y ' = -?-.
Q6. The maximum of y = x 2 Bx + 14 on the interval (1, 6] is-?- .
-
Ql. If lim Un = limin for function f, then f is -?- .
QB.Write the definition of derivative.
Q9. Write the physical meaning of derivative.
Q10. Jsec2xdx = -?-.
l. Paraboloid Problem: A paraboloid is formed by rotating around the y-axis the graph
of y = 0.5x2 from x = 0 to x = 3.
a. Write an integral for the area of the paraboloid . Evaluate it numerically .
b. Show that your answer is reasonable by comparing it with suitable geometric
figures .
c. The indefinite integral in la is relatively easy to evaluate. Do so, and thus find the
exact area. Show that your answer in la is close to the exact answer .
2. Rotated Sinusoid Prob lem : One arch of the graph of y = sinx is rotated around the
x-axis to form a football-shaped surface.
a. Sketch the surface .
b. Write an integral equal to the area of the surface. Evaluate it numerically.
c. Show that your answer is reasonable by comparing it with suita ble geometric
figures .
3. In-Curved Surface, Prob lem I: The graph of y = lnx from x = 1 to x = 3 is rotated
around the x-axis to form a surface. Find the area of the surface.
4. In-Curved Surface, Problem II: The graph of y = lnx from x = 1 to x = 3 is rotated
around the y-axis to form a surface. Find the area of the surface .
5. Reciprocal Curved Surface Problem I: The graph of y = 1/x from x = 0.5 to x = 2 is
rotated around the y-axis to form a surface . Find its area .
6. Reciprocal Curved Su rface Problem II: The graph of y = 1/x from x = 0.5 to x = 2
is rotated around the x-axis to form a surface. Find its area. How does this answer
compare with that in Problem 5?
7. Cubic Parabo loid Problem I: The cubic paraboloid y = x 3 from x = 0 to x = 2 is rotated
around the y-axis to form a cuplike surface . Find the area of the surface.
Sec
tion 8-8: Areaof a Surface of Revolution 413
8. Cubic Paraboloid Problem II: The part of the cubic parabola
3
y = -x + 5x 2 - Bx + 6
in Quadrant I is rotated about the y-axis to form a surface. Find the area of the
surface.
For Problems 9-16, write an integral equal to the area of the surface . Evaluate it exactly,
using the fundamental theorem. Find a decimal approximation for the exact area.
9. y = Jx, from x = 0 to x = 1, around the x-axis
10. y = x3 , from x = 1 to x = 2, around the x-axis
5 y
x4 x-2
11. Y = - +- from x = 1 to x = 2 around the x-axis
8 4 ' '
12. y = x2, from x = 0 to x = 2, around the y-axis
13. y = (x2 + 2) 3 12 , from x = 0 to x = 3, around the y-axis
X
14. y = 2x 113 , from x = 1 to x = 8, around the y-axis 0 l 2 3 4 5
x3 1
15. y = + x, from x = 1 to x = 3, around the line y = -1
3 4
x3 1
16. Y = + x, from x = 1 to x = 3, around the line x = 4
3 4
17. Sphere Zone Problem: The circle with equation x 2 + y 2 = 2 5 Figure 8-Bh
is rotated around the x-axis to form a sphere (Figure 8-8h).
a. Slice the sphere ,vith planes perpendicular to the
x-axis . Write the differential of surface area, dS, in
terms of x. 5 y
b. Find the area of the zone between
i. x = 0 and x=1
Radius gets
ii. x = 1 and x= 2 shorter .
iii. x = 2 and x= 3
iv. x = 3 and x =4 X
V . x = 4 and x= 5
c. As you progress from the center of a sphere toward
a pole, you would expect the areas of zones of equal
altitude to decrease because their radii are decreasing,
but also to increase because their arc lengths are ------
-1
increasing (Figure 8-8i). From the results of 17b, which
of these two opposing features seems to predominate Figure 8-Bi
in the case of a sphere?
18. Sphere Total Area Formu la Prob lem: Prove that the surface area of a sphere of radius r
is given by S = 4rrr 2 .
19. Sphere Volume and Surfac e Problem: The volume of a sphere may be found by slicing
it into spherical shells (Figure 8-8j). If the shell is thin, its volume is approximately
equal to its surface area times its thickness. The approximation becomes exact as the
thickness of the shell approaches zero. Use the area formula in Problem 18 to derive
the volume formula for a sphere,
3
V = 1rrr .
4 14 Chapte
r 8: TheCalculus
of PlaneandSolidFigu
res
20. Sphere Rate of Change of Volume Problem: Prove that the
instantaneous rate of change of the volume of a sphere with
respect to its radius is equal to the sphere's surface area.
21. Paraboloid Surface Area Problem: Figure 8-8k shows the
paraboloid formed by rotating around the y-axis the graph
of a parabola y = ax 2 . Derive a formula for the surface area of
a paraboloid in terms of its base radius r and the constant a in
the equation .
22. Zone of a Paraboloid Problem: Zones of equal altitude on a sphere
have equal areas (Problem 17). Is this property also true for a
paraboloid (Figure 8-8k)? If so, support your conclusion with
appropriate evidence. If not, does the area increase or decrease Figure 8-8j
as you move away from the vertex?
y
23. Ellipsoid Problem: The ellipse, with x-radius 5 and y-radius 3 and
parametric equations
X = 5 COS t
y = 3 sint,
is rotated around the x-axis to form an ellipsoid (a football-shaped
surface). Write an integral for the surface area of the ellipsoid and
evaluate it numerically. Show why the Cartesian equation (x / 5) 2 +
(y / 3) 2 = 1 for the same ellipsoid would be difficult to use because of X
what happens to dL at the end of the ellipsoid, at x = 5.
24. Cooling Tower Problem: Cooling towers for some power plants are Figure 8-8k
made in the shape of hyperboloids of one sheet (see photograph) . This
shape is picked because it can be made using all straight reinforcing rods. A framework
is made, then concrete is applied to form a relatively thin shell that is quite strong, yet
has no structure inside to get in the way. In this problem you will find the area of such
a cooling tower.
Figure 8-81
Section
8-8: Areaofa Surface
of Revolution 415
f!J!..
?
LID
The cooling tower depicted in Figure 8-81 is formed by rotating about the y-axis the
hyperbola with the parametric equations
x = 3 5 sect
y = 100 + 80tant,
where x and y are in feet.
a. The hyperbola starts at y = 0. What is the radius of the hyperboloid at its bottom?
b. The hyperbola stops where t = 0.5 . What is the radius at the top of the hyperboloid?
How tall is the cooling tower 7
c. What is the radius of the cooling tower at its narrowest? How high up does this
narrowest point come ?
d. Find the surfac e area of the hyperboloid.
e. The walls of the cooling tower are to be 4 in. thick.
Approximately how many cubic yards of
concrete will be needed to build the tower?
25. Lateral Area of a Cone Problem: Figure 8-Bm shows a con e of radius Rand slant height
L. The con e is a singly curved surface, so it can be cut and rolled out into a plane
surface that is a sector of a circle. Show that the area of the lateral surface of a cone is
S = rrRL.
Figure 8-8m
Figure 8-8n
26. Latera l Area of a Frustum Problem: Figure 8-8n shows that a frustum of a cone is
a difference between two similar cones, one of radius and slant height R and L; the
other, r and I. By the properties of similar triangles,
R r
L I.
Use this fact and clever algebra to transform the area of the frustum so that it is in
terms of the average radius and the frustum slan t height,
- +-
S = rrRL - rrrl = 2rr ( R r)(L - 1).
2
416 Chapter
8: TheCalculus
of PlaneandSolidFigure
s
8-9 Lengths and Areas for Polar Coordinates
You have seen how to find lengths of curves specified by parametric equations and
by regular Cartesian equations. In this section you will find lengths and areas when
the curve is specified by polar coordinates. In polar coordinates, the position of a
point is given by the displacement from the origin (the pole) and the angle with the
positive x-axis (the polar axis).
Suppose that an object is located at point (x, y) in the Cartesian plane (Figure 8-9a).
Let r (for "radius") be the directed distance from the pole to the object. Let 0 be the
directed angle from the polar axis to a ray from the pole through the point (x, y).
Then the ordered pair (r, 0) contains polar coordinates of (x, y). Note that in (r, 0)
the variabl e r is the dependent variable, not the independent one. Figure 8-9a also
shows how (r, 0) can be plotted if r is negative . In this case, 0 is an angle to the ray
opposite the ray through (x, y).
y
(x,y)
( r, 0)
r >O
0 X
Polar axis
(x,y)
( r, 0)
Figure 8-9a
Suppose that the polar coordinates of a moving object are given by
r = 5 + 4 cos 0.
By picking values of 0, you or your grapher can calculate and plot the corresponding
values of r. The polar graph in this case (Figure 8-9b) is a lima(on , a French word
for "snail." (The cedilla under the c makes its pronunciation "s.")
Figure 8-9b
Section
8-9:Leng
thsandAreasforPolarCoordinates 417
OBJECTIVE Given the equation of a polar function, find the area of a region bounded by the graph and
the length of the graph .
Samp le Area
point
(r, 0 ) Figure 8-9c shows a wedge-shaped region between a polar curve and the pole, swept
out as e increases by a small amount d0. The point (r, 0) on the graph can be
us ed as a samp le point for a Riemann sum . The area of the region is approximate ly
the area of a circular sector of radius r and central angle d0. The sector ha s area
d0 / (2TT) of a whol e circle . Let dA be the sector's area .
.. dA = TTr ZTT
de = !r
2 2 d0
2
The area of an entire region swept out as e increases from a to b is found by
Figure 8-9c
summin g the areas of the sectors between a and b and taking th e limit as d0
approaches zero (that is, integrating).
Area of a Regionin Polar Coordinates
The area A of the region swept out between the graph of r = f(0) and the pole as 0
increases from a to b is given by
A = lim ' l.r 260 = J,bl.r 2d0 .
l'.0-0 L 2 a 2
Example 1 Find the area of the region enclosed by the limac;:onr = 5 + 4 cos e.
Solution The graph is shown in Figure 8-9b. You can plot it with your grap her in polar
mode . If you start at e = 0, the graph makes a complete cycle and closes at
0 = ZTT.
2
dA =( S + 4cos0) d0 Find th e area of a se ctor, d A .
The entire limac;:onis generated as e increases from Oto ZTT.
2rr Add th e sector areas and take th e limit (that is,
.A =
0J (5 + 4cos0) 2 d0
= 103.672 . .. square units
int egrat e).
By num erical inte grati on.
As a rough check, the lima c;:onis somewhat larger than a circle of diameter 10
units . The area of the circle is 25TT= 78.5 .... So 103.6 ... is reasonable for the
limac;:on. (Since the limac;:onis symmetrical , you could integrate from O to TTand
double the answer .)
The integral of a power property cannot be us ed in Example 1 because de cannot be
made the differential of the inside function . In Section 9-5, you will learn a technique
for evaluatin g this integral algebraically using the fundamental theorem. You will
find that the exact answer is 33TT.If you divide the unrounded num erical answer,
103.672 ... , by TTyou should get 33.
418 Chapter
8: TheCalculus
ofPlaneandSolidFigur
es
Example 2 shows you how to find the area swept out as a polar curve is generated
if r becomes negative somewhere, or if part of the region overlaps another part.
Example 2 Figure 8-9d shows th e lima\on r = l + 3 sin e.
4
Figure 8-9d
a. Find the area of the region inside the inner loop .
b . Find the area of the region between the outer loop and the inner loop.
Solution a. Figure 8-9d shows a wedge-shaped slice of the region and a sample point
(r, 0) on the inner loop of the graph.
dA = r2 d0 = O+ 3 sin0) 2
By plotting and tracing, you will find that the inner loop correspon ds to
value s of 0 roughly between TTand 2TT.You will also find that r is negative,
as shown in Figure 8-9d. However, dA will be positive for positive values of
d0 because the r is squar ed. To find the limits of integration precisel y, set r
equal to zero .
1 + 3 sin 0 = 0 =>sin 0 = -
=>e = -0.3398 ... + 2TTn or TT- (- 0.3398 ... ) + 2TTn
By usin g n = l in the first equation and n = 0 in the second equation, you
can find values of 0 in the desired range.
0 = 3.4814 . .. or 5.9433 ...
For convenience, store these values as a and b in your grapher.
(b 1
A= Ja 2 (1 + 3sin0)2d0 "' 2.527636... By num ericalintegration.
As a rough check, the inner loop has an area slightl y smaller than that of a
circle of diam eter 2. See Figure 8-9d. That area is TT 12 = 3.141 . . ..
b . The outer loop begins an d en ds where the inner loop ends and begins.
The appropriate values of the limit s of integr ation as found in part a are
a = - 0.3398 . .. and b = 3.4814 ... . Store the se values in your grapher and
repeat the numerical integration.
A = I:O + 3 sin0) 2 d0 "' 14.751123 .. .
Section
8-9: Lengths
andAreasforPolarCoordinates 419
It is important for you to realize that this is the area of the entire outer
loop , between the graph and the pole, swept out as 0 increases from
- 0.33 ... to 3.48 .... The area of the region between the two loops is the
difference between this and the area of the inner loop.
A "" 14.751123 ... - 2.527636 ... = 12.223487 .. .
ArcLength
Figure 8-9e shows a part of a polar curve traced out as 0 increases by d 0. An arc of
a circle dr awn at (r, 0) would have length r d0 since 0 is measured in radians . The
length dL is close to that of the hypotenuse of a right triangle of legs r d0 and dr.
By the Pythagorean theorem,
dL = ~ dr 2 + (r d0) 2.
Factoring out d0 2 and taking its square root,
dL = (:;f + r 2 d0 .
Figure 8-9e
The length of the entire path traced as 0 increases from a to b is found by surnrning
the dL's and finding the limit (that is, integrating).
Lengthof a Curvein PolarCoordinates
The length L traced out along the polar curve r = f(0) as 0 increases from a to bis given
by
L = lirn = 2,dL =
.10---,0
J.b
a
V dr 2 + (r d0)2 = J.b
a
V'(dr/ d0)2 + r 2 d0.
Example 3 Find the length of the limac,:on r = 1 + 3 sin 0 in Example 2 (Figure 8-9d).
dr
Solution de = 3cos0
.-. dL = ~-(3_c
_o_s_0_)2_+_(1_+_3_s_in
_0_)_
2 de
420 Chapter
8: TheCalculus
of PlaneandSolidFigures
By plotting the graph and tracing, you will find that the graph starts repeating
itself after e has increased by 2rr radians. So convenient limits of integration
are Oto 2rr. A smaller interval would not generate the entire graph. A larger
interval would count parts of the graph more than once.
L= J:'r~( 3cos0 )2 + (1 + 3sin0) 2 d0
L a:: 19.3 768 ... units By num erical int egration.
The inner and outer loops of the lima~on are close to circles of diamet ers 2 and
4, respectively. The circumferences of these circles add up to 2rr 1 + 2rr 2 =
18.84 .... So 19.3 ... is reasonable for the length of th e bma~on.
Problem Set 8-9
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Differentiate: f(x ) = 5x 3 - 7x 2 + 4x - 11
Q2. Differentiate: g (x) = (4x - 9) 3
Q3. Differentiate: h(x ) = sin 3 x
Q4. Differentiate: u(x) = sec 3x
QS. Differentiate: v(x) = e-x
Q6. Differentiate: r(x) = 1/ x
QJ. Integrate : J(1 / x) dx
QB. Integrate: Jx dx
Q9. Integrate: J 3 dx 10 I
Q10. Integrate: J dx
1. Figure 8-9f shows the polar graph of the circle r = 10 sine, with 5
diameter 10.
a. Find the area of the region swept out between the graph and the
pole in one revolution as e increases from Oto 2rr.
5
b . Why is the answer in la twice the area of the region insid e the
circle? Why don't you get a negative value for the area integral
as e increases from rr to 2rr , even though r is negative for these Figure 8-9f
values of 0?
2. Find the length of the circle r = 10 sinein Figure 8-9f that is traced out as e makes one
revolution, increasing from Oto 2rr radians. Why is the answer twice the circumference
of the circle? Why do you suppose the polar coordinate length formula is phrased
dynamically, in terms of the length "traced out," rather than statically in terms of the
length "of" the curve?
Section8-9:Lengths
andAreasforPolarCoord
inates 421
For Problems 3-10,
a. Plot the graph, thus confirming the one shown here.
b. Find the area of the region enclosed by the graph.
c. Find the length of the graph.
3. The lima~on r = 4 + 3 sin 0 4. The lima~on r = S - 3 cos 0
S. The curve r = 7 + 3 cos 2 0 6. The four-leaved rose r = 8 cos 20
7. The cardioid r = S + S cos 0 8. The ellipse r = ~O
3- cos
9. The three-leaved rose r = sin 3 0 (Be careful of the range of 0-values!)
10. The cissoid of Diocles r = 4 sec 0 - 4 cos 0, and the lines 0 = -1 and 0 = 1
422 Chapter
8: TheCalculus
of PlaneandSolidFigures
11. Figure 8-9g shows the lemniscate of Bernoulli with polar
equation
5
r = -V49cos 20 .
What range of values of e causes the right-hand loop to b e
generated? What is the total area of the region enclosed by
7
both loops?
12. The polar graph of r = csc e + 4 is a conchoid of Nicomedes.
The graph is unbounded but contains a closed loop. Find Figure 8-9g
the area of the region inside the loop.
13. Figure 8-9h shows these polar graphs.
Cardioid: r = 4 + 4 cos 0
Circle: r = 10 cos 0 .
Where do the graphs intersect ? What is the area of the
region outside the cardioid and inside the circle ?
14. Find the area of the region that is inside the circle r = 5
and outside the cardioid r = 5 - 5 cos e.
15. Figure 8-9i shows the Archemedian spiral r = 0.50. Figure 8-9h
a. Find the length of the part of the spiral shown.
b. Find the area of the region in Quadrant I that lies
between the outermost branch of the spiral and the
next branch in toward the pole.
16. For the lirnai;:on r = 4 + 6 cos e,find the area of the region
that lies between the inner and outer loops.
17. Column Scroll Problem: The spiral design at the top of Ionic
columns in ancient Greek archit ecture is an example of a
lituus (pronounced "lit' -you-us"). Plot the lituus
r = se - 112
Figure 8-9i
traced out as e increases from Oto 6rr .
a. Find the length from e = rr / 2 to e = 6rr.
b. Let e = l radian . Sketch an arc of a circle centered at the
pole, from the polar axis to the point (r, 1) on the lituus.
Find the area of the sector of the circle corresponding
to this arc. Repeat the calculation for e = 2 and e = 3.
What seems to be true about these areas?
Section
89:Lengths
andAreasforPolarCoordinates 423
18. Line Problem: Show that the graph of r = sec e is a line. Find the length of the segment
from e = 0 to e = 1.5 using the calculus of polar coordinates . Then confirm that your
answer is correct by appropriate geometry .
19. LP Record Project: In this project you are to calculate the length of the groove on an
old 33 rpm record. Obtain such a record . Figure out a way to measure the number of
grooves per centimeter in the radial direction. Then figure out a polar equation for the
spiral formed by th e grooves. By integrating, calculate the length of the groove from
the outer one to the inner one. Perform a quick calcu lation to show that your answer
is reasonable .
20. Kepler's Law Project : Figure 8-9j shows the path of
a spaceship in an elliptical orbit. The earth is at the
pole (one focus of the ellipse) . The polar equation
of the ellipse is
100 50 0= 0.8
r=----
3 - 2 cos e'
where e is in radians and r is in thousands of
miles. In this problem you will investigate the
speed of the spaceship at various places.
0 =0
a. Find the area of the elliptical sector from e = 0
100
to 0 = 0.2. Shaded regions
have equal areas.
b . German astonomer Johannes Kepler (1571-
1630) observed that an object in orbit sweeps
out sectors of equal area in equa l times. This
fact is known as Kepler's second law of plane-
tary motion. (His first law states that the orbit Figure 8-9j
is an ellipse where the object being orbited is
located at one focus.) If th e sector in Figure 8-9j
starting at e = 0.8 has area equal to the one in
20a, at what value of 0 does the sector end?
c. Kepler's third law states that the period of an orbiting object is related to its
distance from the object being orbited. If a is half the major axis of the orbit, then
the period P is
P=ka1. 5
Find the value of the constant k using data for the moon. The moon is about
a = 240,000 mi from earth and has a period of about 655 hr (27.3 d 24 hr / d).
d. What is the period of the spaceship in Figure 8-9j?
e. How many hours does it take the spaceship to travel from 0 = 0 to e = 0.2? How
man y hours does it take to travel from e = 0.8 to the value of 0 in 20b 7
f. How many miles does the spaceship travel on its elliptical path between e = 0 and
e = 0.2? How many miles does it travel between e = 0.8 and the value of e in 20b?
g. Find the average speed (distance / time) of the spaceship for each of the two arcs in
20f .
h. See whether you can explain physically why a spaceship would move faster when it
is closer to the earth than it does when it is farther away.
424 Chapter
8: TheCalculus
of PlaneandSolidFigures
21. The Derivative dy / dx for Polar Coordinates Problem:
Figure 8-9k shows the polar graph of the spiral
r=0
... --
~ -~-"
.;...- . ..
........
; ....... :-.;..
.. !..-~.f
superimposed on a Cartesian xy-plane . A tangent line is ---~-.
plotted at the point (r, 7).
a. Estimate the slope of the tangent line.
b. Polar and Cartesian coordinates are related by the
parametric equations ; ,_ '.
. ~" -=--"t:
.. :, :
X = r COS 0 -~-;
:
--:-- ....:.
'. ..
, .:; :..
y = r sine. I '\. -:----~---~--
By appropriate use of the parametric chain rule, find
an equation for dy / dx and use it to show algebraically Figure 8-9k
that the slope you found in 21a is correct.
22. Project - The Angle Between the Radius and the Tangent y
Line: A remarkably simple relationship exists between
dr / d0 in polar coordinates and the angle 1./J(Greek letter
psi, pronounced the same as "sigh") measured counter- I
I
clockwise from the radius to the tangent line (Figure 8-91). I
In this project you will derive and apply this relationship .
a. Explain why tan e = y /x.
b. Let cf>(Greek letter phi, "fee" or "fye") be the angle from
the positive horizontal direction to the tangent line
(Figure 8-91). The slope of the tangent line is dy / dx. X
Explain why I
I
I
dy / d0
tancp = dx / d0 Figure 8-91
c. You recall from trigonometry that tan (A - B) = tan A -AtanBB. Use this property,
1 + tan tan
the results of 22a and b, and appropriate algebra to show that
x(dy / d0) - y(dx / d0 )
tan 1./J= --'-----'-- - -
x (dx / d 0) + y(dy / d0)
d. Use the fact that x = r cos 0 and y = r sin 0 to show that the numerator in 22c
equals r 2 .
e. Use the fact that r 2 = x 2 + y 2 to show that r :; equals the denominator in 22c, and
thus the following property holds.
Property:TheAnglePsiin PolarCoordinates
If (/I is the angle measured counterclockwise from the radius to the tangent line of a polar graph,
then
r r
tan(/} = dr/d0 = ?
Section
8-9: Lengths
andAreasforPolarCoordinates 425
f. Figure 8-9m shows the cardioid ,'---
r = a - a cos e,
where a stands for a nonzero constant. Prove that for
)::;'
this cardioid the angle l/J is always equal to one-half of '
-+-~' -----~ ~ -1....,~
, Za
e. You will find that the half-argument properties for
tangent, which you may recall from trigonometry, are
helpful.
g. Figure 8-9n shows a cross section through a chambered
nautilus shell . The spiral has the property that the angle l/J
is a constant. Use this fact and the property in 22e to find Figure 8-9 m
the general equation of this equiangular spiral . Choose two
values of e on the photograph and measure the corresponding
values of r. Use these values as initial conditions to find the particular equation
for the outer spiral. Confirm that your equation is correct by plotting it, tracing
to another value of e,and showing that the point is actually on the photographed
spiral. Use the constants in your equation to calculate the value of l/J. Measure a
photocopy of the shell to show that your calculated value of l/J is correct.
Equiangular spiral. t/1 is constant.
Figure 8-9n
8-10 Chapter Review and Test
In this chapter you have seen a major application of derivatives and integrals
to geometrical problems. The rate of change of the area or volume of a figure
describes how fast these quantities change as a given dimension changes . Maximum
or minimum areas or volumes occur where the rate of change equals zero. Areas,
volumes, and curved lengths can be calculated by slicing a figure into small pieces,
adding the pieces, and taking the limit. The resulting limits of Riemann sums are
4 26 Chapter
8: TheCalculus
of Plane andSolidFigures
equal to definite integrals. You evaluated th ese integrals numerically or by the
fundamenta l th eorem if you were able to find the indefinite integral.
The Review Problems are numbered according to the sections of this chapter. The
Concepts Problems allow you to apply your knowledg e to new situations. The
Chapter Test is typical of a classroom test .
Review Problems
RO. Update your journal with things you've learned since the last entry . You should include
such things as those listed here.
The one most important thing you have learned in studying Chapter 8
How the volume, length, and surface area of a geometric figure are found
How volume, length, and surface area are found in polar coordinates or with para-
metric functions
Which boxes you have been working on in the "define, understand, do, apply" table
Rl. Three cubic functions have equations
f(x) = x3 - 9x 2 + 30x - 10,
g (x ) = x 3 - 9x 2 + 27x - 10, and
h(x) = x 3 - 9x 2 + 24x - 10.
a. Plot the graphs on the same screen . Sketch the results .
b. Write equations for the first and second derivative of each function.
c. Which function has two distinct values of x at which the first
derivative is zero? What are these values of x7 What features occur f'(x)
at these points?
d. Which function has a horizontal tangent line somewhere but no
local maximum or minimum points?
e. Each function has a point of inflection . Show that the second
derivative is zero at each of the points of inflection .
R2. a. For the function in Figure 8-1Oa, sketch a number-line graph
X
for f' and for f " showing the sign of each derivative in a 2
neighborhood of the critical point at x = 2. Indicate on the
number lines whether there is a local maximum, a local Figure 8-1Oa
minimum, or a point of inflection at x = 2.
b. Sketch the graph of a function whose f'(x) 0 + 0 +
derivatives have the features given in X 1 --2 '
5
Figure 8-lOb.
X
((x )
I._
-2
+ 0
3
- 0 +
'
5
Figure 8-1Ob
Sectio
n 8-10: Chapter Rev
iewandTest 427
c. Figure 8-1 Oc shows the graph of f(x)
f(x) = x 2!3 - x.
i. Write equations for f'(x ) and f " (x) .
ii. The graph appears to slope downward for all x . Does 10
f(x) have any local maxima or minima? If so, where? If
not, explain how you can tell .
iii . The graph appears to be concave downward for all x .
Are there any points of inflection? If so, where? If not,
explain how you can tell.
iv. Write the global maximum and minimum values of f(x)
for x in the closed interva l [O, 5].
I Figure 8-1Oc
d. For f(x) = x 2e-x, find the maxima, minima, and points of inflection and sketch the
graph .
R3. a. Storage Battery Prob lem: A normal automobile battery has
six cells divided by walls (Figure 8-1Od). For a particular
battery, each cell must have an area of 10 in 2 , looking down
from the top . What dimensions of the battery will give the
minimum total wall length (including outsides)7 A typical
battery is 9 in . by 6.7 in. (that is, 1.5-in. cell width) . Does
minimum wall length seem to be a consideration in battery Figure 8-1Od
design?
b . Cylinder in Cubic Paraboloid Prob lem: A rectangle is inscribed in the region in
Quadrant I under the cubic parabola y = 8 - x 3 . Two sides of the rectangle are on
the x- and y-axes, and the opposite corner of the rectangle touches the graph. The
figure is rotated about the y-axis . The curve generates a cubic paraboloid, and the
rectang le generates a cylinder. What rectangle dimensions give the largest-volume
cylinder?
R4. a. Find the area of the region bounded by the graph of y = lnx, the y-axis, and
the lines y = 1 and y = 2. Use the fundamental theorem. Verify your answer
numerically.
b . Find the area of the region bounded by the graphs of y = x 113 and y = x / 3 - 2/ 3.
c. Mystery Prob lem: Mr. Rhee must find the area of the region bounded by the graphs
of y = x 3 and y = x. He finds, correctly, that the graphs intersect at x = -1 and
x = 1. So he integrates
1
J (x 3 - x) dx = 1x 4 -
- 1 4
1x 2
2
J 1
- 1
= .!.
4
- .!. - .!. + .!. = 0.
2 4 2
(Surprise!)
Explain to Mr. Rhee what went wrong .
RS. a. The region under the graph of y = e0 -2x from x = 0 to x = 4 is rotated around the
x-axis to form a solid . Find its volume .
b. The region in Quadrant I bounded by the graphs of y = x 0 25 and y = x rotates
around the y-axis to form a solid . Find its volume.
4 28 Chapter
8:TheCalculus
ofPlaneandSolidFigu
res
c. Oblique Cone Problem: Figure 8-lOe shows an oblique circu lar
cone with base at y = 0. Each cross section perpendicular to 6 y
the y-axis is a circle, with diameter extending from the graph
of y = x + 2 to y = 3x - 6. Find the volume of the cone. Is its
volume larger or smaller than that of a right circular cone of
the same altitude, 6, and base radius, 2?
R6. a. The region in Quadrant I bounded by the graphs of y = x 113 X
and y = x 2 is rotated about the x-axis to form a solid . Find the
volume of the solid by cylindrical shells.
b. Find the volume of the solid in R6a by plane slices . Show that Figure 8-1Oe
the answer is the same.
c. Various Axes Probl em: The region bounded by the parabola y = x 2 and the line
y = 4 is rotat ed to form various solids. Find the volume of the solid for the
following axes of rotation:
i. y-axis ii. x-axis
iii. Line y = 5 iv. Line x = 3
R7. a. Write an integral equal to the length of the parabola y = x 2 between x = - 1 and
x = 2. Evaluate th e integral numerically .
b. Find exactl y the length of the graph of y = x 312 from x = 0 to x = 9 using the
fundamental theorem. Find a decimal approximation for the answer. Check your
answ er by employing suitable geometry.
c. Find the length of the following spiral from t = 0 to t = 4.
X = t COS TTt
y = t sin rrt
RB. a. Find, exactly by the fundamental theorem, the area of the surface formed by
rotating around the y-axis the graph of y = x 113 from x = 0 to x = 8. Find a
decimal approximation for the answer . Check your answer by employing suitab le
geometr y.
b. The graph of y = tanx from x = 0 to x = 1 is rotated around
the line y = - 1 to form a surface . Write an integral for the
area of the surface . Evaluate it numericall y.
5
c. The spiral in Probl em R7c is rotated about the y-axis to
form a "sea shell ." Find its surface area .
R9. Figure 8-lOf shows the spiral
5
r =0
from 0 = 0 to 0 = Srr / 2.
a. Find the length of this part of the spiral.
b . Find the area of the re gion in Quadrant I that is outside the Figure 8-1Of
first cycle and inside the second cycle of the spiral.
Section
8-10: Chapter
Review
andTest 429
Concepts Problems
Cl. Oil Viscosity Problem: The viscosity (resistance to flow) of normal motor oil decreases
as the oil warms up. "All-weather" motor oils retain about the same viscosity
throughout the range of operating temperatures. Suppose that the viscosity of
lOW-40 oil is given by
= 130 - 12T + 1ST 2 - 4T 3 , for O ~ T ~ 3,
where (Greek letter mu, pronounced "mew" or "moo") is the viscosity in centipoise
and Tis the temperature in hundreds of degrees.
a. At what temperature in this domain will the maximum viscosity occur?
b. What is th e minimum viscosity in this domain 7 Justify your answer .
C2. "Straight Point" Problem: Show that the graph of f(x) = (x - 1) 4 + x has a zero second
derivative at x = 1 but does not have a point of inflection there. Sketch what the
graph will look like in the vicinity of x = l. Describe what is true about the graph at
X = l.
C3. In.finite Derivative Problem : The functions f(x) = x 213 and g (x ) = x - 213 both have
infinite first derivatives at x = 0, but the behavior of each is quite different there .
Sketch a graph showing the difference .
C4. Chapter Logo Problem: The logo on each even-numbered
page of this chapter shows a solid formed by rotating about y
the line x = 4 the part of the graph
y = 3 + 5 [ 0.5 + 0.5 cos ( f (x - 5)) r
from x = 5 to x = 7.5. A cylindrical hole 1 unit in radius is
coaxial with the solid. Figure 8-1Og shows the coordinate 3
system in which this diagram was drawn.
a. Find the length of the segment of graph that was rotated .
X
b. Find the x-coordinate of the point of inflection. 4
c. Find the area of the doubly curved surface of the solid .
Figure 8-1Og
d. Find the volume of the solid.
CS. Area by Planimeter Project: You have learned how to calcu-
late the area of a region algebraically using the fundamental
theorem and also numerically . There is a mechanical device
called a planimeter (compensating polar planimeter) that
finds the area geometrically from a drawing of the region.
In this problem you are to borrow a planimeter and use it
to find the area of Texas from a map.
a. Photocopy the map of Texas in Figure 8-1Oh. Be sure
that the scale is photocopied, too, because some copy
machines shrink the picture .
b. Set up the planimeter with the tracer point at a conve- 0 100 200 300 400 500 miles
nient starting point on the map . Set the dial to zero. 0 200 400 600 800 kilom eter s
Then trace around the boundary until you return to the
Figure 8-1Oh
starting point.
43 0 Chapter
8: TheCalculus
of PlaneandSolidFigures
MAU ;q;
c. Read the final setting on the dial. The planimeter may have a vernier scale for
-
reading tenths of a unit.
d. Measure the scale on the map to find out how many
miles correspond to 1 cm. Be clever! Then find out
how many square miles correspond to 1 cm 2 . Finally,
calculate the area of Texas to as many significant
digits as the data justify.
e. Find out from the planimeter's instruction manual
the theoretical basis on which the instrument works .
Write a paragraph or two describing what you learned.
f. Check an almanac or encyclopedia to see how accurate
your measurement is. Figure 8-1Oi
C6. Ho le in th e Cylinder Project: A cylinder of uranium 10 cm in
diameter has a ho le 6 cm in diameter drilled through it. (Figure
8-1 Oi)The axis of the ho le intersects the axis of the cylinder at
right angles. Find the volume of the uranium drilled out . Find
the value of the uranium drilled out . You may assume that
uranium is worth $200 a gram. Be resourceful to find out the
density of uranium.
C7. Three -Ho le Project : A cube 2 cm on each edge has three
mutually perpendicular holes drilled through its faces (Figure
8-1 Oj). Each hole has a diameter of 2 cm, so it comes right to
the cube faces which it parallels . Find the volume of the solid
remaining after the three holes are drilled. Figure 8-1Oj
Chapter Test
Tl. Figure 8-1 Ok shows the graph of f(x)
f(x) = x 3 - 7.8x 2 + 20.25x - 13. 10
Ascertain whether the graph has a relative maximum and a relative
minimum, a horizontal tangent at the point of inflection, or just a X
2 3 4
point of inflection with no horizontal tangent. Justify your answer.
Figure 8-1Ok
T2. Figure 8-101 shows critical values of f'(x) and f'(x) + 0 + =
I II
f" (x) for a continuous function f, as well as X 1 2 4
the signs of the derivatives in the intervals
between these points. Sketch a possible graph f"(x) 0 +=+ 0 - +
L II
off using the initial condition that f(O) = 3. X l 2 3 4
Figure 8-101
Section 8-10: Chapt
er Reviewand Test 431
Problems T3-T8 are concerned with the region R shown in Figure 8-lOm . Region R is
bounded by the graph of y = x 3 from x = 0 to x = 2, the y-axis, and the line y = 8.
T3. Write a definite integral to find the length of the graph of y = x 3
from x = 0 to x = 2. Evaluate the integral to find the length .
T4. Write a definite integral equal to the area of the surface
generated by rotating the segment of graph in Problem T3
about the y-axis. Evaluate the integral.
TS. A rectangular region is inscribed in region R as shown in Figure
8-lOm. As R rotates about the y-axis, the rectangular region
generates a cylinder. Find exactly the maximum volume the
cylinder could have . Justify your answer.
X
T6. Using slices of R perpendicular to the y-axis, write an integral
equal to the volume of the solid formed by rotating R around the
y-axis . Evaluate the integral algebraically using the fundamental Figure 8-1Om
theorem.
T7. Using slices of R parallel to the y-axis, write another integral equal to the volume in
Problem T6. Show that the volume determined this way is exactly the same as that in
Problem T6.
TB. Suppose that a cylinder is circumscribed about the solid in Problem T6. What fraction
of the volume of this cylinder is the volume of the solid ?
T9. For the ellipse with param etric functions
X = 5 COS t
y = 2 sin t
a. Plot the graph and sketch it.
b. Find the length of the ellipse.
c. Show that the volum e of the ellipsoid formed by rotating the ellipse about the
x-axis is
V = ~rr (x-radius) (y-radius) 2 .
For Problems no and Tll, use the spiral r = se0 10 shown in Figure 8-lOn. 30
no. Find the length of the part of the spiral shown .
Tl l. Find the area of the region in Quadrant I that is outside the
second revolution of th e spiral and inside the third revolution. 40
Figure 8-1On
432 Chapter
8:TheCalculus
ofPlaneandSolidFigures
CHAPTER
9 Algebraic Calculus Techniques
for the Elementary Functions
The Gateway Arch in St. Louis is built in the form of a catenary, the
same shape a chain forms when it hangs under its own weight. In
this shape the stresses act along the length of the arch and cause no
bending. The equation of a catenary involves the hyperbolic functions,
which have properties similar to the circular functions of trigonometry.
433
-----........--....'"' "''
Mathematical Overview
In Chapter 9 you will learn ways to integrate each of the elementary
functions and their inverses. These are
algebraic
trigonometric
logarithmic
hyperbolic
You will do this in four ways .
Graphically The logo at the top of each even -
numbered page of this chapter
shows the graphical meaning of
the integration by parts formula,
used to integrate products.
Numerically X
1 0
ln X r
0
ln t dt
u
2 0.693 ... 0.386 .. .
3 1.098 ... 1.295 ...
4 1.386 .. . 2.545 ...
5 1.609 ... 4.047 ...
Algebraically f u dv = uv - f v du, the integration by parts formula.
Verbally The most fu.ndamental method of integration seems to be integration
by parts. With it I can integrate products of functions. I can also use it
to find algebraic integrals for the inverse trigonometric functions .
4 34
9-1 Introduction to the Integral of a Product of
Two Functions
Suppose you are to find the volume of the solid formed by rotating the region
under y = cosx around the y-axis (Figure 9-la). The value of dV, the differential of
volume, is
dV = 2rrx y dx = 2rrxcosxdx.
Thus the volume is
rrr/ 2
V = 2rr Jo xcosxdx.
The integrand, x cos x, involves a product of two functions. So far you have been
able to evaluate such integrals only by approximate numerical methods because
you usually could not find the antiderivative of a product.
)'
X
1t
Figure 9-1 a
In this chapter you will learn algebraic techniques for integrating the so-called
elementary functions. These are the algebraic functions, involving no operations
other than addition, subtraction, multiplication, division, and roots (rational
exponential powers); and the elementary transcendental functions, which are the
trigonometric and inverse trigonometric functions, logarithmic and exponential
functions, and hyperbolic functions. The hyperbolic functions are defined in terms
of exponential functions but have properties similar to those of the trigonometric
functions.
Some of the techniques you will learn were essential before the advent of the
computer made numerical integration easily available . They are now interesting
mostly for historical reasons and because they give you insight into how to approach
a problem . Your instructor will guide you to the ones that are important for your
course. You will see how algebraically "brave" one had to be to learn calculus in the
days "BC" ("before calculators"). But for each technique you learn you will get the
thrill of knowing, "I can do it on my own, without a calculator!"
OBJECTIVEOn your own or with your study group, find the indefinite integral Jx cos x dx, and use
the result to find exactly the volume of the solid in Figure 9-1a using the fundamental
theorem .
Section9-1: Introductio
n to theIntegralofa Product
ofTwoFunctions 435
Exploratory Problem Set 9-1
1. Find the volume of the solid in Figure 9-la approximately by numerical integration.
2. Let f (x ) = x sinx. Use the derivative of a product formula to find an equation for f '(x) .
You should find that x cos x, th e integrand in Problem 1, is one of the terms.
3. Multiply both sides of the equation for f'(x) in Problem 2 by dx. Then int egrate both
sides. (That's easy 1 You just write an integral sign(" f ")in front of each term!)
4. The integral f x cos x dx should be one term in the equation of Problem 3. Use suitable
algebra to isolate this integral. Then do the inte grating on the other side of th e
equation. Recall what f f'(x) dx equals!
5. Use the result of Problem 4 to find th e exact volume of the solid in Figure 9-la.
6. Find a decimal approximation for the exact volume in Problem 5. How close did the
approximation in Problem 1 com e to this exact volum e?
7. The technique of this exerc is e is called int egra tion by parts. Why do you suppose
this name is used ? How do you suppose the function f(x) = x sinx was chos en in
Problem 2?
9-2 Integration by Parts- A Way to Integrate
Products
Figure 9-2a shows th e solid generated by rotating about th e y-axis the region under
th e graph of y = cos x. The volume of this solid is given by
rr / 2
V = 2rr f
O
x cos x dx.
In Section 9-1, you saw that this produ ct of functions can be integrated algebraical ly.
ln this section you will learn why th e integration by part s technique works.
rr
2
Figure 9-20
43 6 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
OBJECTIVE Given an integral involving a product, evaluate it algebraically using integration by parts.
To see how to integrate a product of two functions, it helps to start with the formula
for the differential of a product. If y = uv, where u and v are differentiable functions
of x, then
dy = du v + u dv
Commuting the du and v in the right-hand side of the equation and integrating both
sides gives
Jdy = Jv du + J u dv
Either integral on the right-hand side can be isolated and written in terms of the
remaining quantities. For instance,
Ju dv = Jdy - Jv du .
Since f dy = y (ignoring, for the time being, "+C") and y = uv, you can write
J udv = uv - J vdu.
For f x cos x dx, let u = x and let dv = cos x dx. Then the following holds.
du= dx and v = Jcosxdx = sinx + C Differentiate u and int egra te dv.
Substi tut e for u, v, and du in th e
.. Jxcosxdx = x(sinx + C) - J(sinx + C) dx equation f u dv = uv - f v du.
A new constant of integrat ion
= xsinx + Cx + cosx-Cx + C1 comes from th e second int egral.
The old constan t of int egra tion
= xsinx + cosx + Ci cance ls out!
The integral of a product of two functions can always be written as
J(one function)(differential of another function) .
Associating the integrand into two factors leads to the name "integration by parts."
It succeeds if the new integral, f v du, is simpler to integrate than the original one,
f udv .
Integrationby Parts
Technique:
A way to integrate a product is to write it in the form
f (one function)(differential of another function).
If u and v are differentiable fun ctions of x, then
f u dv = UV - f V du.
It's a good id ea to memorize the integration by parts formula as you will use it
often. Example 1 shows a way to use this formula.
Section9-2: Integration
byParts-A Wayto Integrate
Product
s 437
Example 1 Do the integrating: f Sxe 3x dx
Solution f Sxe 3
x dx Let u = Sx, dv = e3x dx .
Write u and dv to the right, out of th e
way, as shown.
.. du = S dx, v = f e3 dx
x
Differenti ate u and int egrate dv to find
du and v .
= e3x+ C.
= Sx(e3x + C) - J(e + C)S dx 3x Use th e int egra tion by par ts formul a,
f udv = u v - f vdu.
= xe x + SCx - f e 3 dx - f SC dx
3 x
= xe3 x + SCx - ~e3 x - SCx + C1
= :'>.
3
e3x+ c 1
xe 3x _ :'>.
9
Again, the original constant C in integrating f dv conveniently "drops out." The C1
com es from th e last integral. In Problem 46 of Problem Set 9-3, you ,vi.llprove that
this is always the case. So you don't need to worry about putting in the + C until the
last integral disappears. Example 2 shows that you may have to integrate by parts
mor e than onc e.
Example 2 Do th e integrating: f x 2 cos 4x dx
Solution Jx 2
cos4x dx u = x2 dv = cos4xdx
du = 2xdx v = .!.
4
sin4x
= x 2 . sin4x - J(sin4x )( 2xdx ) Use f udv = uv - f vdu.
Note that th e int egral still
involves a produ ct of two
fun ctions.
= x2 sin4x - f (2x)( sin4xdx )
Ass ociate th e dx with the sin e
factor, as it was originall y.
u = 2x dv =sin4xdx
du = 2dx v = _.l...
16
cos4x
= x 2 sin4x - [- xcos4x - f (-fc; cos4x )( 2) dx ] Use int egration by part s again .
2
= x sin4x + xcos4x - iz sin4x + C
Integration by parts is successful in this example because at both steps th e second
integral, f v du, is less complex than f u dv at the start of the step . You could also
have chosen these terms.
u = cos4x
du= - 4sin4xdx
:i
However, the new integral, - f x 3 sin 4x dx, would have been more complicated
than the original one . The following conclusions will help you decide how to split
up an integral of a product into appropriate parts.
438 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
Technique:
Choosingthe Parts in Integrationby Parts
To evaluate I u dv, the following criteria should be met.
Primary
criterion:
dv must be something you can integrate.
Secondary
criterion:
u should , if possible, be something that gets simpler (or at least not
much more complicated) when it is clifferentiated.
Problem Set 9-2
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
QJ. Differentiate: y = xtanx
Q2. Integrate: Jx 10 dx !/
Q3. Sketch: y = e -x
Q4. Integrate: Jcos 3x dx
y
,--r:1
QS. Differentiate: y = cos 5x sin 5x !H ' X
Q6. Sketch: y = 2/ x C a b
Ql. r(x) = Jt(x)dxifandonlyif-?-. Figure 9-2b
QB.Definition: f'(x) = -?-.
Q9. If region R (Figure 9-2b) is rotated around the line x = c, the volume of the solid is- ?-.
Q10. If f(6.2) = 13, f(6.5) = 19, and f(6.8) = 24, then f'(6.5) "' -?-.
For Problems 1-10 integrate by parts.
l. J xsinxdx 2. J x cos 3xdx
3. J xe 4x dx 4. J 6x e- 3x dx
5. J (x + 4)e - sxdx 6. J (x + 7)e 2x dx
7. J x 3 lnxdx 8. J x 5 ln3xdx
9. J x 2 ex dx 10. J x 2 sinxdx
11. Integral of the Natural Logarithm Problem: The integral fln x dx can be evaluated by
parts, but you must be clever to make the "obvious" choice of parts! Find Jlnx dx.
Section9-2: Integration
byPorts-AWoyto Integrate
Products 439
9-3 Rapid Repeated Integration by Parts
U~iply dv
There is a pattern that helps you remember the integration by parts formula. Write u
du ~V and dv on one line . Below them write du and v. The patt ern is, "Multiply diagonally
Integrate down, then integrate across the bottom." The arrows in Figure 9-3a remind you of this
Figure 9-3a
pattern . The plus and minus signs on the arrows say to add uv and subtract Jv du.
This pattern is particularly useful when you must integrate by parts several times.
OBJECTIVE Use the pattern of Figure 9-3a to simplify repeated integration by parts.
Here's the way Example 2 of Section 9-2 was done.
f x cos4xdx
2
u = x2 dv = cos4xdx
du = 2xdx v=.!.sin4x
4
=x 2
sin4x - J(sin4x ) (2xdx)
2
= x sin4x - J(2x )( sin4xdx ) u = 2x dv = sin4x dx
du = 2 dx v = _ ...!...
16
cos4x
2
= x sin4x - [ - xcos4x - f (~ cos4x) (2) dx]
2
= x sin 4x + x cos 4x - -=lzsin 4x + C
Note that the function to be differentiated appears in the left-hand column and the
function to be integrated appears in the right-hand one. For instance, 2x appears as
part of du in the first step and again as u in the next step. If you head the left column
"u" and the right column "dv," and leave out dx and other redundant information,
the work can be shortened as follows.
u dv
f x cos 4x dx
2
x2 + cos4x
2x ~l sin4x
= x 2 sin4x + xcos 4x - -=lzsin4x + C ~ 4 1
2~ - 16 cos4x
_--............
1 . 4
0- - 64 Sln X
The second diagonal arrow has a minus sign because the minus sign from the first
Jv du carries over. The third diagonal arrow has a plus sign since (- ) (-) from the
step before gives a plus. If you put in the third step (fourth line), the last integral is
f Odx, which equals C, the constant of integration.
RapidRepeatedIntegrationby Parts
Technique:
Choose parts u and dv. Differentiate the u's and integrate the dv's.
Multiply down each diagonal.
Integrate once across the bottom .
Use alternating signs shown on the arrows.
If you get O in the u-column, the one integral will be JO dx, which equals C.
Examples 1-4 show some special cases and how to handle them .
440 Chapter
9:Algebraic
Calculus
Techniques
fortheElementary
Functions
MaketheOriginal
Integral
Reappear
When you do repeated integration by parts,
the original int egral may reappear. u dv
Example 1 Do the integrating: Je 6xcos 4x dx e
6
x ~ cos4x
6
6e x ~ sin4x
Solution f e xcos 4x dx
6
36e6 x .........-
_ ..!_cos4x
16
= !e
4
6 xsin4x + i.. e 6xcos4x - ~
16 16
Je6xcos4x dx Why + and - , not - and +7
The integral on the right is the same as the original one on the left but has a
different coefficient. Adding ~ Je6xcos4x dx to both sides of the equation gives
.?l
16
f
e 6xcos 4x dx = !e
4
6xsin4x + i.. e 6xcos 4x + C
16
The "+ C" must be displayed on the right-hand side of the equation because no
indefinite integral remains th ere . Multiplying both sides by ~ and simplifying
gives
fe 6x cos 4x dx = ..!..
13
e6x sin 4x + l...
26
e6x cos 4x + C1
In case you have doubts about the validity of what has been done, the answer can
be checked by differentiation.
e6xsin4x + l...e
Y = ..!..
13 26
6xcos4x + C1
6xsin4x + .i..e 6xcos4x + 1- e6xcos4x - i.. e 6 xsin4x
Y , = i..e
13 13 13 13
y ' = e6x cos 4x Th e ori ginal int egrand .
UseTrigonometric
Properties
to MaketheOriginal
Integral
Reappear
In Example 1, you found that the original integral reappeared on the right-hand side
of the equation . Sometim es you can use properties of the trigonometric functions
to mak e this happen. Example 2 shows how this can be done. ote the clever choice
of u and dv!
Example 2 Do the integrating: Jsin 2 x dx u dv
sinx ~ sinx
Solution
COS X --=-- -COS X
f
= - sinx cos x + cos 2 x dx
= -sinxcosx + J(l - sin 2 x) dx By th e Pytha gor ean prop erti es .
= -sinxcosx + J ldx - Jsin 2xdx J sin 2 x dx r eapp ear s!
.'. 2 J sin 2 x dx = -sinx cosx + x + C Do th e indi ca ted algebra.
Jsin 2 xdx = -sinxcosx + x + Ci
Section9-3:RapidRepeated byParts
Integration 441
It would have been possible to make f sin 2 x dx reappear by carrying the integration
by parts one more step. Something unfortunate happens in this case, as shown here .
u dv
Jsin 2
x dx sinx ~ sinx
= -s inxcosx + cosxsinx + Jsin 2 xdx COS X ~ -COS X
-s inx -- -sinx
This time the original integral appears on the right-hand side but with 1 as its
coefficient. When you subtract f sin 2 x dx from both sides and simplify, you wind up
with
0 = 0,
which is true, but not very helpful! Stopping one step earlier as in Example 2 avoids
this difficulty.
Reassociate
Factors
BetweenSteps
The table form of repeated int egrat ion by parts relies on the fact that the u and
dv stay separate at each step. Sometimes it is necessary to reassociate some of the
factors of u with dv, or vice versa, before taking the next step . The table format can
still be used. Example 3 shows how.
Example 3 Do the integrating: f x 3 ex2 dx
2
Solution The quantity x e dx can be integrat ed like this:
J x ex2 dx = J ex2 (2x dx) = ex2 + C
2
The technique is to mak e xe dx show up in the dv column .
u dv
Choose the parts this way.
Draw a dashed line.
1 x2
2 ~ 2 xe
Assoc iat e the x with th e other factor.
0 + 1 x2
- 4e
The dashed line across the two columns indicates that the factors above the line
have been reassociated to the form shown below the line. The arrows indicate which
of the terms are actually multiplied together to give the answer.
Integral
of theNaturalLogarithm
Function
In Probl em 11 of Problem Set 9-2, you were asked to integrate lnx. To do this, you
must be clever in selecting the "parts ." Example 4 shows you how.
442 Chapter
9: Algebraic
Calculus
Techn
iquesfar theElementary
Funct
ions
Example 4 Do the integrating : Jln x dx
Solution f lnxdx u dv
= xlnx - f ~(x) dx lnx~ 1
dv can be integrated!
1/X ---=-
X
= xlnx - x + C.
Property:Integralof the NaturalLogarithmFunction
JInxdx = xlnx - x + C
In Problem Set 9-3, you will practice rapid repeated integration by parts. You will
justify th e fact that the "+C" can be left out in the integration of dv . You will also
practice integrating other familiar functions .
Problem Set 9-3
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
10 in less than 5 minutes.
QI. Differentiate: f(t) = te 1
f
Q2. Integrate : r 5 dr y
a
Q3. Differe n tiate: g(m) = m sin 2m
f
Q4. Integrate: sec 2x dx
QS.Integrate : f (x 3 + 11) 5 (x 2 dx)
Q6. Integrate : f (x 3 + 11) dx '
a
'
b
X
Ql. Find Jim (sin x)
X---+0 X '
Figure 9-3b
QB.Find Jim
x-.O
(sinx)
2x .
Q9. If region R (Figure 9-3b) is rotated about the x-axis, the volume of the solid is -?-.
QIO.Sketch the parabo loid formed by rotating y = x 2 from x = 0 to 2 around the y-axis.
For Problems 1-20, do th e integrating.
l. f x3e2xdx 2. f x e-x dx
5
3. f x sinx dx
4
4. f x cosxdx
2
5. f x cos 2x dx
5 6. f x sin Sx dx
3
7. f exsinx dx 8. f eXcosxdx
9. f e xcos Sx dx
3
10. f e Xsin2x dx
4
Section9-3: RapidRepeated byParts
Integration 44 3
!~ .'.
11. Jx 7 ln3xdx 12. Jx 5 ln6xdx
13. Jx 4 In 7 dx 14. Je 7xcos 5 dx
15. f sin xcosxdx
5 16. f x (3 - x 2 2 13
) dx
17 . Jx 3 (x + 5) 112 dx 18. fx ~
2
dx
19. Jlnx 5 dx 20. J eln7x dx
For Problems 21-32 , do the integrat ing . You may reassociate factors between st eps or us e
the trigonometric functions or logarithm properties in the integration by parts.
21. f x 5ex2dx 22. Jx 5 ex3 dx
23. f x(lnx) 3 dx 24. f x3(lnx) 2
dx
25. Jx3 (x2 + 1)4 dx 26. Jx 3 ~ dx
27. Jcos 2 x dx 28. J sin 2 0.4x dx
29. Jsec 3x dx 30. J sec 2 x tanx dx (Be clever')
31. Jlog3x dx 32. f log 10 x dx
For Problems 31-38, write the antiderivative .
33. J sinxdx 34. Jcosx dx
35. f cscxdx 36. f secxdx
37. Jtanx dx 38. Jcotx dx
39. Wanda Y. Knott evaluates f x 2 cos x dx, letting u = x 2 and dv = cos x dx. She gets
x2 sinx - f 2xsinxdx .
For the second integral, sh e lets u = sinx and dv = 2x dx. Show Wanda why her
second choice of u and dv is inappropriate.
40. Amos Take evaluates f x 2 cos x dx by parts, letting u = cos x and dv = x 2 dx. Show
Amos that although his choice for dv can be integrated, it is a mistake to choose the
parts as he did.
41. If you evaluate f exsinx dx, the original integral reappears after two integrations by
parts . It will also reappear after four integrations by parts. Show why it would be
unproductive to evaluate the integral this way.
42 . The integral f cos 2 x dx can be integrated by clever use of trigonometric properties,
as well as by parts. Substitute (1 + cos 2x) for cos 2 x and integrate. Compare this
answer with that which you obtain using integration by parts, and show that the two
answers are equivalent .
444 Chapte
r 9: Algebraic
Calculus
Techniques
for the Elemen
tary Functio
ns
43. Area Problem: Sketch the graph of y = xe - x from x = 0 to x = 3.
Where does the function have its high point in the interval (0, 3]?
Find algebraically, with the fundamental theorem, the area of
the region under the graph from x = 0 to x = 3.
44. Unbounded Region Area Prob lem : Figure 9-3c shows the region
under the graph of y = 12x 2 e-" from x = 0 to x = b. Find an
equation for this area in terms of b. Then find the limit of the
area as b approaches infinity. Does the area approach a finite
number, or does it increase without bound as b increases? Justify Figure 9-3c
your answer.
45 . Volum e Problem: The region under the graph of y = lnx from x = l to x = 5 is rotated
around the x-axis to form a solid. Find its volume exactly, using the fundamental
theorem.
46. Proof Problem: In setting up an integration by parts problem, V
fu dv
you select dv to equal something useful, and integrate to find
v . Prove that whatever number you pick for the constant of
integration at this point, it will cancel out later in the integration
by parts process.
47 . Areas and Integration by Parts: Figure 9-3d shows the graph of C
function v plotted against function u. As u goes from a to b, v
u
goes from c to d. Show that the integration by parts formula can a b
be interpreted in terms of areas on this diagram. (This diagram
is the same as the logo at the top of each even page in this Figu re 9-3d
chapter.)
48. Integral of In Generalization Problem: Derive a formula for f ln ax dx, where a stands
for a non-zero constant.
*49. Introduction to Reduction Formulas Problem: For f sin 7 x dx, integrate once by parts. Use
the Pythagorean properties in an appropriate manner to write the remaining integral
as two integrals, one involving sin 5 x and the other involving sin 7 x. Then use algebra
to combine the two integrals involving sin 7 x, and thus express f sin 7 x dx in terms
off sin 5 xdx. Use the resulting pattern repeatedly to finish evaluating Jsin 7 xdx. The
pattern leads to a reduction formula, as you will learn in the next section .
50. Journal Problem: Update your journal with techniques and concep ts you have learned
since the last entry . In particular, tell about integration by parts, the kind of integral it
is used for, and the rapid way in which it can be accomplished.
*This problem prepares you for the next section.
Section
9-3:RapidRepeated byParts
Integration 445
9-4 Reduction Formulas and Computer Software
In this section you will develop formulas that allow you to integrate algebraically
powers of the trigonometric functions. Repeated integration by parts can be used
for integrals such as
f sin x dx.
6
The choice for dv is sin x dx because it can be integrated and the result is no
more complicated . But the other part, u = sin 5 x, gets more complex when you
differentiate it.
u dv
sin 5 x ~ sinx
Ssin4 x cosx ---=- -cosx
So you just put the integral back
together and hope for the best!
f sin xdx
6
= - sin 5 xcosx + 5 Jsin 4 xcos 2 xdx
= - sin 5 xcosx + SJ sin 4 x(l - sin 2 x) dx Use th e Pythagor ean prop erty to
4
re turn to sines.
= - sin 5xcosx + SJ sin xdx - SJ sin 6 xdx
The original integral appears on the right-hand side with a coefficient of - 5. Adding
SJ sin 6 x dx to the first and last members of the above equation (and using the
transitive property to ignore the twe, members in the middle) gives
f
6 sin 6 x dx = -sin 5 x cos x + SJ sin 4 x dx.
Dividing by 6 gives
f sin xdx 6
= - sin 5 xcosx + J sin 4 xdx .
The integral f sin 6 x dx has been replaced by an expression in terms off sin 4 x dx.
The new integral has the same form as the original but is "reduced" in complexity .
By repeating the above integration with n instead of 6 as the exponent, you can
find an equation expressing this integral with any nonzero exponent in terms of an
integral with that exponent reduced by 2.
Jsinnxdx = -~sinn - 1xcosx + n ~ 1 Jsinn- 2 xdx
An equation such as this one is called a reduction formula. It can be used again on
the new integral. For the above example, the work would look like this.
Jsin 6 x dx
= -i sin x cos x + Jsin x dx
5 4
" + (- sin 3 xcosx + J sin 2 xdx)
" - i...
24
sin 3 x cos x + ~8 f sin x dx
2
" - " + (- sin 1 xcosx + J sinxdx)
" - " - ~ sinx cos x + ~ J dx. Th e las t int egr al equ als x + C.
= _ !sin
6
5xcosx - i...sin 3xcosx - i...sinxcosx + i...x + C
24 16 16
An swer.
446 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
You can use ditto marks, as shown, to avoid having to rewrite the first terms so
many times .
Definition:ReductionFormula
A red uction formula is an equation express ing an integ ral of a particu lar kind in terms of
an integral of exactly the same kin d, but of reduced complexity.
OBJECTIVE Given an integral involving powers of fun ctions, derive a reduction formul a, and given a
r eduction formul a, us e it to evaluate an indefinite integral either by pencil and paper or
by compu ter sof tware.
The following formulas for integration of the six trigonometric functions (recall
from Section 6-9) are repeated here to refresh your memory.
Properties:Integralsof the Trigonometric
Functions
f sinxdx = - cosx + C f cosxdx = sinx + C
f tanxdx = lnlsecxl + C f cotxdx = - In fcscxl + C
= - ln Icos x I + C = In lsinx l + C
f secxdx = ln Jsecx + tanx l + C f cscxdx = - In fcscx + cot x i+ C
= - In fsecx - tan xi+ C =lnlcscx-cotxl +C
The work at the beginning of this section shows that a reduction formula can be
derived by choosing a particular case, integrating, and looking for a pattern in the an-
swer . However, to be perfectly sure the result is correct, start with a general integral
of the tr:pe in question, using a letter such as n to stand for the (constant) exponent.
Example 1 shows you how the derivation can be done for the integral of seenx dx .
Example 1 a. Derive a reduction formula for f secnx dx that can be used if n ~ 2.
b. Use the reduction formula to evaluate f sec5 x dx.
c. Check your answer to part b by computer software .
Solutions a. The best choice for dv is sec2 xdx, because it can be integrated easily and its
integral, tanx, is simpler than the u dv
original integrand . The work looks
like this . secn-2 x + sec 2 x
f secnxdx (n - 2)secn - 3 secx tanx tanx ~
= secn- 2 xtanx - ( n - 2) Jsecn- 2 x tan 2 xdx
= secn- 2 x tanx - ( n - 2) Jsecn- 2 x(sec 2 x - 1) dx
= secn- 2 x tanx - (n - 2) J(secnx - secn- 2 x) dx
= secn- 2 x tanx - (n - 2) Jsecnx dx + ( n - 2) Jsecn- 2 x dx
Section9-4: Redu
ction Formula
s andComputer Software 447
The desired integral now appears in the last member of the equation, with
- (n - 2) as its coefficient. Adding ( n - 2) f secnx dx to the first and last
members (and eliminating the middle members by transitivity) gives
(n - 1) J secnxdx = secn- 2 xtanx + (n - 2) J secn- 2 xdx.
Dividing both members by (n - 1) produces the desired reduction formula .
1 2
seenx dx = - - sec"- 2 x tan x + n - Jsecn- 2 x dx
I n- 1 n- 1
b. Jsec5x dx
= sec 3 xtanx + J sec 3 xdx
= sec 3 xtanx + (secxtanx + Jsecxdx)
= sec 3 xtanx + secxtanx + Jsecxdx
= sec 3 x tan x + sec x tan x + ln Isec x + tan x I + C
c. Symbol-manipulating software may present the answer in a slightly different
form. For instance, one symbol-manipulating calculator gives
4
3(cos(x)) ln (lcos(x )I) - 3(cos(x)) 4 - ln( lsin(x) - 11) + sin(x)(3(cos(x) ) 2 + 2) + C
-
8(cos (x)) 4
By converting sinx / cosx to tanx, and cosx to 1/ secx, you get an answer
equivalent to that in part b.
Sometimes a reduction formula can be derived without integrating by parts as
Example 2 shows .
Example 2 Derive a reduction formula for f cotnx dx that can be used if n 2'.. 2.
Solution Jcotnx dx
= Jcotn- 2 x (cot 2 x dx)
Use the Pythagore an prop erti es to transform to
= Jcotn- 2 x (csc 2 x - 1) dx cose cant.
= J cotn-zx csc 2 x dx - J cotn-zx dx
1
= -- - cotn- Jx - Jcotn- 2 x dx Integral of a pow er function.
n- 1
The last expression has the integral of cotangent with an exponent 2 less than
the original integral and is thus a reduction formula .
Example 3 Use the reduction formula in Example 2 to evaluate f cot 5 x dx.
Solution Jcot 5 x dx
= -cot x -
4
Jcot xdx
3
= - cot 4 x- (- cot 2 x - Jcotxdx)
= -cot x + cot x + In Isin x i+ C
4 2
You will derive the following reduction formulas in Problem Set 9-4.
448 Chapter
9: Algeb
raicCalculus
Techniques
for the Elementary
Functions
~ tl:i ;;ft& itM
ReductionFormulas
: Integralsof Powers of Trigonometric
Functions
(For reference only . Do not try to memorize these!)
f sinnxdx = -!isinn- 1xcosx + n ~ 1 f sin n- xdx, for n ~ 2 2
f cos x dx = icosn - lx sinx + n ~ 1 f cos x dx, for n ~ 2
11 11
-
2
tannxdx = - - tann - lx -f tann- xdx, for n ~ 2
1 2
.r n- 1
f cotnxdx = --n - -l cotn- lx -f cot" - xdx, for n ~ 2
1 2
2
+ n - J.sec 11- 2 x dx, for n ~ 2
f secnxdx = _l_l
n-
sec 11
-
2 x tanx
n- 1
2
11 1
f csc xd x = - -n - -l csc 11 2
- xcotx + n -
n- 1
Jcscn- 2 xdx, for n ~ 2
Don't try to memorize the reduction formulas! You are almos t boun d to mak e a
mistake, and there is nothing quite as useless as a wrong form ul a. Fortunate ly you
will not have to make a career out of evaluating reduction formu las . Your purpose
here is to see how the computer comes up with an answer as in Example 1 when it
integrates a power of a trigonometric function.
Problem Set 9-4
DoTheseQuickly
The following prob lems are intended to refresh your skills. You should be able to do all
ten prob lems in less than five minutes.
Ql. In integration by parts, f udv = -7-.
Q2. Sketch: y = 3 cos x
Q3. Sketch: y = cos 2x
Q4. Differentiate : y = x ln Sx
QS. Integrate: f sin x cos x dx
5
Q6. Integrate: f dx/x
Ql. Integrate: Ji3e2x dx
QB. Sketch the graph of a function that is continuous at (3, 1) but not different iable there .
Q9. Find: (d / dx)(tan - 1x)
QlO. Integrate: f sec x dx
For Problems 1-6, take the first step in integration by parts or use appropriate trigono m etry
to write the given integral in terms of an integral with a reduced power of the same
function.
l. f sin x dx
9
2. f cos 10
x dx
Section9-4: Redu
ctionFormulasand Computer
Software 449
3. Jcot 12
x dx 4. J tan 20 x dx
5. J sec 13 x dx 6. Jcsc 100
xdx
For Problems 7-12, derive the reduction formula in the table on page 449 directly, using
n as the exponent, rather than using a particular constant as in Problems 1-6.
7. J cosnx dx, n =1:-0 8. J sinnx dx, n=1:-0
9. J tannx dx, n =1:-0 10. J cotnx dx, n =1:-0
11. J cscnx dx, n =1:-0 12. J secnx dx, n =1:-
0
For Problems 13-18, integrat e by one of these methods .
a. Use pencil and paper, and the appropriate reduction formula.
b. Use computer software.
13. J sin 5x dx 14. J cos 5 x dx
15. J cot 6 x dx 16. J tan 7 x dx
17. J sec 4 x dx 18. J csc 4 x dx
X
19. Cosine Area Problem: Figure 9-4a shows the graphs of
y = cosx,
y = cos 3 X, and
y = cos 5 x. Figure 9-4a
a. Which graph goes with which function?
b. By numerical integration find the approximate area of the region under each graph
from x = - TT / 2 to TT / 2.
c. Find exactly each area in 19b by the fundamental theorem. Use the reduction
formulas.
d. Based on the graphs, explain why the areas you calculated are reasonable .
e. Plot the graph of y = cos 100 x. Sketch the result.
f. As the exponent n gets larger, the graph of y = cosnx gets "narrower ." Does the
limit of the area as n approaches infinity seem to be zero? Explain .
*20. Integral ~f cos 5 x Another Way: The int egral Jcos 5 x dx can be written
f cos x dx = J(cos x) cos x dx.
5 4
The factor cos 4 x can be converted to powers of sine by appropriate use of the
Pythagorean properti es from trigonometry. The result will be three integrals that you
will be able to evaluate by the fundamental theorem. Find the area of the region under
the graph of y = cos 5 x from x = - TT / 2 to TT / 2 using this technique for integration .
Compare your answer with that in Problem 19c.
*This probl em prepar es you for th e next section .
450 Chapter
9: Algebra
ic Calcu
lusTechniques
for the Elementary
Functions
21. Integral of Secant Cubed Problem: The integral J sec 3x dx seems to appear often, as you
will see in the next few sections. Use an appropriate technique to do the integration.
Then see whether you can figure out a way to remember the answer.
22. Reduction Formula for f sinnax dx: Derive a reduction formula for
f sinnaxdx, n 2. 2, a *- 0.
Then use the reduction formula to evaluate f sin 5 3x dx.
1
23. Prove that f sin 3 ax dx = - a (cos ax)(sin 2 ax + 2) + C (a * 0) .
3
1
24. Prove that f cos 3 axdx = a (sinax)(cos 2 ax + 2) + C (a * 0).
3
9-5 Integrating Special Powers of Trigonometric
Functions
In Section 9-4, you found algebraically the indefinite integrals of any positive integer
power of any trigonometric function. In this section you will see how to integrate
some special powers of these functions without having to resort to reduction
formulas .
OBJECTIVE Be able to integrate odd powers of sine or cosine, even powers of secant or cosecant, and
squares of sine or cosine without having to use the reduction formulas.
OddPowers
of SineandCosine
Example 1 Do the integrating: f sin 7 x dx
Solution The key to the technique is associating one sinx factor with dx, then transform-
ing the remaining (even) number of sines into cosines using the Pythagorean
properties.
Jsin 7 x dx
= Jsin 6 x(sinx dx )
= f (sin x) (sinx dx )
2 3
= J( 1 - cos 2 x )3 (sinxdx )
= J( 1 - 3cos 2 x + 3cos 4 x - cos 6 x)(sinxdx)
= J sinxdx - 3 J cos 2 xsinxdx + 3 J cos 4 xsinxdx - J cos 6 xsinxdx
= - cosx + cos 3x - cos 5x + i cos 7 x + C
Sec
tion 9-5: Integrating
Special
Powers
of Trigonometric
Functions 45 1
Each integral in the next-to -last line of Example 1 has the form of the integral
of a power, fun du . So you wind up integrating power functions rather than
trigonometric functions. This technique will work for odd powers of sine or cosine
because associating one of the factors with dx leaves an even number of sines or
cosines to be transformed into the cofunction .
Squares
of SineandCosine
In Section 8-9, you found the area inside the lima~on r = 5 + 4 cos e (Figure 9-Sa) is
2rr
J
A= 0 (5 + 4cos0 )2 d0.
Figure 9-50
Expanding the binomial power gives
A= J: rr(25 + 40cos e + 16cos 20)d0.
The last term in the integral has cos 20. The integrals f sin 2x dx and f cos 2x dx occur
frequently enough to make it worthwhile to learn an algebraic shortcut. The double
argument property for cosine is
cos 2x = cos 2 x - sin 2 x.
The right-hand side can be written either entirely in terms of cosine or entirely in
terms of sine .
cos 2x = 2 cos 2 x - 1
cos 2x = 1 - 2 sin 2 x
Performing algebra on these two equations gives the following.
Property:DoubleArgumentPropertiesfor Sineand Cosine,Transformed
cos2 x = (l+cos2x)
sin2 x = o - cos2x)
452 Chapter
9: Algebraic
Calculus
Techn
iquesfar the Elementary
Functions
These two equations allow f sin 2 x dx and f cos 2x dx to be transformed so that the
integrand is linear in cos 2x.
f cos xdx2
= fc1+cos2x)dx= x+si n2x+C
f sin xdx2
= J(1 - cos2x)dx = x-sin2x +C
Example 2 Do the integrating: f sin 2 8x dx
Solution Jsin 2 8x dx
f
= (1 - cos 16x) dx
= x - ~ sin l 6x + C.
EvenPowersof SecantandCosecant
The technique for integrating odd powers of sine and cosine can be adapted to even
pow ers of secant and cosecant. Example 3 shows how.
Example 3 Do the integrating: f sec 8 Sx dx
Solution Jsec 8 Sx dx
= f sec Sx(sec Sxdx )
6 2
= f (sec Sx) (sec Sxdx)
2 3 2
= J(tan 2 Sx + 1) 3 (sec 2 Sxdx)
= J(tan 6 Sx + 3 tan 4 Sx + 3tan 2 Sx + l)(sec 2
Sxdx)
= J tan 6
Sx sec Sx dx + 3 J tan Sx sec Sx dx + 3 J tan 2 Sx sec 2 Sx dx + J sec 2 Sx dx
2 4 2
7 5 3
= 1-
35 tan Sx + .l..
25 tan Sx + l5 tan Sx + l5 tanSx + C
The advantage of the techniques in Examples 2 and 3 is that you don't have to
remember the reduction formulas. The disadvantages are that you must remember
certain trigonometric properties, the binomial formula for expanding powers of
binomials, and just which powers of which functions can be integrated this way.
Problem Set 9-5 gives you some opportunities for practice. You will also find the
exact area of the limac;:onin Figure 9-Sa.
Problem Set 9.5
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Find f'(l) : f(x ) = x 3 - 7x
Q2. Find g'(2): g(x) = lnx
Q3. Find h'(3) : h (x) = (2x - 7)6
Q4. Find t'(4): t(x ) = sin fi x
Section9-5:Integrating
SpecialPowers
of Trigonometric
Functions 453
QS. Find p'(5): p(x) = xe x f'(x)
: -{ .... I
Q6. Solve: x 113 =8
Ql. Sketch the graph: (x/ 3) 2 + (y / 5)2 = 1
f
QB. u dv = uv - f v du is the _ 7_ formula. :;c-:
- 0--'---<1
, X
r--+~
Q9. For Figure 9-5b, sketch the (continuous) antiderivative, - 1 . __
1/ 3: . 5~ 8
f(x), that contains point (2, 1).
f f
Q10. [f(x)]n dx = g(x) + [f (x) Jn-t dx is called a - ?- formula.
Figure 9-Sb
For Problems 1-30, do the integration.
1. f sin x dx
5 2. f cos x dx
7
3. f cos 9x dx
7
4. f sin l0xdx
3
5. Jsin 3x cos 3x dx
4
6. Jcos 7x sin 7x dx
8
7. Jcos 8x sin 8x dx
6 3
8. Jsin 2x cos 2x dx
4 3
9. f sin x cos x dx
5 2
10. f cos x sin x dx
3 2
11. f cos x dx
2 12. f sin x dx
2
13. f sin 5x dx
2
14. f cos 6xdx
2
15. Jsec xdx
4
16. Jcsc xdx6
17. f csc 6x dx
8
18. f sec 4
100x dx
19. f tan x sec x dx
10 2
20. Jcot x csc x dx
8 2
21. f sec x tan x dx
10
22. f csc x cot x dx
8
23. f sec 20 dx
10
24. f csc 12 dx
8
25. f (cos x - sin x) dx
2 2
26. J(cos x + sin x) dx
2 2
27. f (sinx )- dx 2 28. f (cos 3x)- 2 dx
29. Jsec xdx
3 30. f csc x dx
3
31. Area Prob lem I: Figure 9-5c shows the re gion bound ed by
X
the graph of
y = cos 5x sin 3x
fromx = 0 to x = 2rr .
a. Using integration by parts, find the ind efinite integral Figure 9-Sc
f cos 5x sin 3x dx .
b. Show that the region has just as much area below the x-axis as it has abov e.
4 54 Chapter
9: Algebraic
Calculus
Techniques
for the Elementary
Functions
32. Area Problem II : Let f(x) = sin 3 x.
a. Plot the graph from x = 0 to x = rr. Sketch the result.
b . Find the area of the region under the graph off from x = 0 to x = rr exactly, using
the fundamental theorem and the integration techniques of this section.
c. Verify your answer to 32b by integrating numerically.
d. Quick! Find the integral of f(x) from x = - rr to x = rr. State what property allows
you to answer this question so quickly.
y
3 3. Volume Problem I: One arch of the graph of y = sin x is rotated
about the x-axis to form a football-shaped solid (Figure 9-5d).
Find its volume. Use the fundamental theorem.
34. Volume Problem II: The region under the graph of y = sec 2 x
X
from x = 0 to x = 1 is rotated about the line y = - 3 to form a
solid.
a. Find the exact volume of the solid.
b. Find the exact volume of the solid if the region is rotated
Figure 9-5d
about the line x = - 3.
35. Limaron Area Problem: Figure 9-5e shows the limac;:on
r = 5 + 4cos e,
which appears at the beginning of this section. By numerical
integration the area of the region from e = 0 to e = rr / 4 is
29.101205 ... square units. Find this area exactly, using the
fundamental theorem. Write the answer using radicals and 0=
rr , if necessary. Show that this exact answer, when evaluated,
5
gives the same answer as obtained numerically.
36. Cardioid Area Problem : Figure 9-5f shows the general cardioid
r = a(l + cos 0),
where a is a constant. A circle of radius a is inscribed in
the cardioid . By the fundamental theorem, find the exact Figure 9-5e
area inside the cardioid in terms of a. Explain how this area
corresponds to the area of the circle.
3 7. Journal Problem: Update your journal with things you've learned
since the last entry. You should include such things as those
listed here.
The one most important thing you have learned since the
last journal entry
The basis behind integration by parts, and what kind of
function can be integrated that way
The reason one might want to go to the trouble of using the
fundamental theorem to integrate, rather that simply using
a Riemann sum or the trapezoidal rule
Any techniques or ideas about derivatives that are still un- Figure 9-5f
clear to you
Section
9-5: Integrating
Special
Powers
of Trigonometric
Functions 455
il 9-6 Integration by Trigonometric Substitution
Recall from geometry that the area of a circle is A = rr r 2 , where r is the radius. This
formula can be derived by calculus if you know how to integrate certain square root
functions.
The circle in Figure 9-6a has equation
x 2 + y 2 = 25 .
Draw a vertical strip. Pick a sample point in the strip on the upper half of the circle.
The area of any one strip is 2y dx . The upper half of the circle has equation
y =) 25 - x 2 .
Figure 9-6a
So the area of the entire circle is
A =2 f 5
) 25 - x 2 dx .
The indefinite integral f .J25 - x 2 dx cannot be done as a power because dx is not
V the differential of the inside function, 2 5 - x 2 . But there is an algebraic way to do
the integration.
Here's your suggested thought process:
5
X
Hmm ... This looks Pythagorean. Like somebody was trying to find the third side
of a right triangle.
Draw a right triangle as in Figure 9-6b , placing the ang le 0 in standard position
u
in a uv-coordinate system .
The radicand is 2 5 - x 2 . So the hypotenuse must be 5 and one leg must be x.
Figure 9-6b Put x on the vertical leg (to avoid a minus sign in dx later) and .J25 - x 2 on the
other leg .
..i = sin0 ~ x = 5sin0 ~ dx = 5cos0d0 Use trigonom etric ratio s .
.J25 - x 2
and 5 = cos e ~ ,J2 5 - x 2 = 5 cos e Use trigonometric ratio s .
.'.f -l25 - x dx 2
= f (5cos0)(5cos0d0 ) Do th e trigonom etric substitution.
= 25 f cos 0 d0 2
= - fo + cos20)d0 Double- argum ent prop erti es.
= 9.2 0 + 9.
4
sin 2 0 + C Do th e inte grating .
= 9.2 0 + 9.2 sin 0 cos 0 + C Double -argum ent propertie s.
= 25 sin _1 ~ + 25 . ~ . .J25 - x + C
2
Do the reverse substitution. See
2 5 2 5 5 triangl e in Figure 9-6b.
25 . _ x x-J
= - sm 1 - + - 25 - x 2 + C Answer.
2 5 2
456 Chapter
9: Algebraic
Calculus
Technique
s for theElementary
Functions
Multiplying by 2 from th e origina l integral and evaluatin g from x = - 5 to x = 5 gives
A =2 f 5
.J25 - x 2 dx = 25 sin - 1 ~ + x -J25 - x 2 I~5
= 25sin - 1 1 + 5.J25- 25 - 25sin - 1 (- 1) - 5.J25 - 25
IT -IT
= 25 + 0 - 25 = 25IT,
2 2
which agrees vvith the answer from the area formula, A = rr 52 .
The substitution above rationalizes the integrand by taking advantage of th e
Pythagorean properties of trigonometric functions. The technique works for square
roots of quadrati cs. Your success in u sing this trigonometric substitution depends
on drawing the triangle, then deciding what to call the legs and hypotenuse . After
the subs tituti on has been made, you must also be able to integrate th e trigonometric
functions that appear .
OBJECTIVE Be able to rationaliz e an integrand containing the square root of a quadratic binomal by
using trigonometric substitution, then do the int egrati on.
In the preceding work the radical contained (constant) 2 - x 2 The situation is
different if the radical contains x 2 -(constant}2 . Example 1 shows you what happens.
Example 1 .
Do the mtegratmg: . I -Jxdx 2
_
9
Solution Draw the triangle . This time x is the hypotenuse and 3 is a leg. Putting 3 on th e
horizontal leg allows you to use secant instead of cosecant (Figure 9-6c), thu s
avoidin g a minus sign when you find dx.
1= sec 0 => x = 3 sec 0 =>
-Jx 2 - 9
dx = 3 sec 0 tan 0 d 0, and = tan 0 => .Jx 2 - 9 = 3 tan 0
3
dx
I -Jx 2 - 9
= J 3 sec 0 tan 0 d 0 Do th e substi tution.
V 3tan0
= Jsec 0 d0
= ln Isec 0 + tan 0 1+ C
= ln ~
3+ +C
X 1 Do th e reverse substitution.
I
3
u
3 = ln Ix+ .Jx2 - 9 1- ln 3 + C Log of a quoti ent.
Figure 9-6c = ln Ix + .Jx2 - 9 1+ C1 In 3 is constant.
Section
9-6:Integration
byTrigonometric
Substitution 457
Example 2 shows what to do if the sign in the quadratic is a plus instead of a minus.
It also shows that trigonometric substitution can be used even if there is no radical
and if the constant is not a square.
Example 2
. . :J
Do t h e mtegratmg dx
X2 + 37
Solution Draw a triangle (Figure 9-6d). Since the sign between terms is a plus sign, both
X and m are legs. By putting X on the vertical leg, you can use tan e instead of
cote. The hypotenuse is .Jx2 + 37.
~ = tan0 =>X = V37 tan0 =>
2
V dx = msec 0 d 0, and
37
.Jx~ = sece => .Jx2 + 37 = m sece
. f x dx+ 37 = f J3737sec
.. 2
0d0
sec e
2
2
X
e u = 1
J37 J d0 1
= me+c
Figure 9-6d
1
= J37tan
_1 X
J37 +C Since x/ v'37= tan 0, 0 = tan - 1 (x/ v'37)
.
DefiniteIntegralsby Trigonometric
Substitution
The reverse substitution in the preceding examples is done to revert to x, as in the
original integral. For a definte integral, you can avoid the reverse substitution if you
change the limits of integration to e instead of x . Example 3 shows you this can be
done with the integral for the area of a circle at the beginning of this section.
Example 3 Find the area of the zone of the circle
x 2 + y 2 = 25
between x = - 2 and x = 3 (Figure 9-6e).
-2 5
Figure 9-6e
458 Chapter
9: Algebraic
Calculus
Techniques
for the Elementary
Functions
Solution This problem is the same as at the beginning of this section except for the limits
of integration. The area is
A =2 J -./25 - x 2 dx .
3
-2
Let x = 5 sin 0. Then dx = 5 cos 0 d0, .J25 - x 2 = 5 cos e, and e = sin - 1 ~ -
If x = 3, then e = sin - 1 i = sin - 1 0.6. If x = - 2, then 0 = sin - 1 -/ = sin - 1 (-0.4).
sin - 10.6
:.A=2 f,sm-. , ( - 0 .4)
(5cos0)(5cos0d0) Substitut e 0 limit s.
sin- 1 0.6
= 2 5 0 + - sin 2 eIsin - 1 (-0 .4)
See th e probl em at the beginning of th.is sec tion .
= 2 5 sin - 1 0.6 + -sin (2 sin - 1 0.6) - 2 5 sin - 1 ( - 0.4) - -sin[2sin - 1 ( -0.4)]
= 47 .5406002 ...
By numerical integration, A ::::::
47.5406 . . ., which is close to 47.5406002 .. ..
In Problem 3 5 of Problem Set 9-6, you will see what happens in trigonometric
substitution if x is negative, and thus is not in Quadrant I. e
Problem Set 9-6
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
QJ. Integrate: Jcos 3x dx
Q2. Integrate: J sin 4x dx
Vlne
locity
Q3. Integrate: Jtan 5x dx
''
Q4. Integrate: J cot 6x dx ' '
'
'
QS. Integrate: Jsec 7x dx I , I '
I ,
' '
Q6. Find: (d / dx)(tan5x)
a c d e b
Ql. Find y ' : y = sin 4x
QB.For Figure 9-6f, the maximum acceleration on [a, b] is at t = -?-. Figure 9-6f
Q9. State the fundamental theorem of calculus.
QJO.Write the definition of indefinite integral.
For Problems 1-16, do the integration .
l. J .J49 - x 2 dx 2. f .J lOO - x 2dx
3. f .Jx 2 + 16dx 4. f .JBl + x 2 dx
5. J .J9x 2 - 1 dx 6. f .J l6x ldx2 -
7. f .J l 7dx- x 2
8. f dx
.J l3 - x 2
Section tionbyTrigonometric
9-6:Integra Substitution 459
~rr
k
9. J R+T1 dx 10. J 1 dx
.Jx 2 - 121
11. f x .Jx2 2 - 9dx 12. fx 2
.J9 - x 2 dx
13. f (1 - x2)3/2 dx 14. f (x 81) 312 dx
2
-
15. f 81dx+ x2 16. f 25xdx2 + 1
Some integrals that can be done by trigonometric substitution can also be done other
ways.
a. For Problems 17 and 18, evaluate the integral by trigonometric substitution .
b. Evaluate the integral again as a power. Show that the two answers are equivalent .
J
17. I .Jxxdx+ 25
2
18
xdx
.Jx 2 - 49
Integrals such as in Problems 19 and 20 can be evaluated by trigonometric substitution .
For instance, in Problem 19, the hypotenuse would be 3 and one leg would be (x - 5). Do
the integration .
19
f ,./9 - dx(x - 5)2
20
J
,./36 - (x + 2)2
dx
Integrals such as in Problems 21 and 22 can be transformed into ones like Problems 19
and 20 by first completing the square . In Problem 21, x2 + 8x - 20 = (x 2 + Bx + 16) - 36,
which equals (x + 4) 2 - 36. Do the integration .
21 J dx 22 J dx
.Jx 2 + Bx - 20 ) x2 - 14x + 50
For Problems 23 and 24, evaluate the integral exactly using the fundamental theorem .
Compare the answer vvith the one you get by numerical integration.
23. J ) 100 - x
8
- 3
2 dx 24. J v'x
- l
4
2 + 25 dx
25. Arc Length of a Parabola Problem : Use the fundamental theorem to find exactly
the length of a parabola y = 3x2 from x = 0 to x = 5. Find a decimal approximation
for the answer. Compare the decimal approximation with the answer by numerical
integration .
26. Area of an Ell ipse Problem:
a. Use the fundam ental theorem to find exactly the ar ea
of the region bounded by the ellipse 9x 2 + 25y 2 = 225
between x = - 3 and x = 4 (Figure 9-6g). Compare a
decimal approximation of this answer with the answer
by numerical integration.
b . Find the exact area of the entire ellipse. How is this area
related to the 5 and 3, which are the x- and y-radii of the Figure 9-69
ellipse?
2 7. Circle Area Formu la Prob lem: Derive by calculus the area formula, A = rrr 2, for the
circle
x 2 + y 2 = r 2.
460 Chapter
9:Algebraic
Calculus
Techniques
fortheElementary
Functions
28. Ellipse Area Formula Problem: Derive by calculus the area
formula for an ellipse. You may start with the general equation
for an ellipse with serniaxes a and b, br(~)2+(tJ2= 1
(;f +(tf =l
(Figure 9-6h). Show that the formula reduces to the area of a
circle if a = b.
29. Ellipsoid Problem: The ellipse in Problem 28 is rotated about the
y-axis to form an ellipsoid . Find the volume inside the ellipsoid
in terms of the constants a and b. What difference would there
CE> Figure 9-6h
X
be in the answer if the ellipse had been rotated about the x-axis? y
4
30. Hyperbola Area Problem: Use the fundamental theorem to find
exactly the area of the region bounded above and below by the
hyperbola
X
x2 - y z = 9
3
from x = 3 to x = S (Figure 9-6i). Find an approximation for this
answer, and compare it with the answer obtained by integrating
numerically.
31. Hyperboloid Problem: The region in Problem 30 is rotated about
the y-axis to form a hollow solid . (The inside surface of the solid is Figure 9-6i
a hyperboloid of one sheet.) Find the volume of the solid.
32. Average Radius Problem: The average radius of a solid of rotation may be defined to
be the distance x for which
Volume = 2rrx. A,
where a is the area of the region being rotated. Find the average radius of the
hyperboloidal solid in Problem 31. Is the average radius more than, less than or exactly
halfway through the region in the x direction as you progress outward from the y-axis?
33. Area of an Ellipse, Parametrically: The ellipse in Figure 9-6h has parametric equations
X= a COS t
y = b sint.
Find the area of the ellipse directly from the parametric equations.
Show that the answer is the same as in Problem 28. How does the
integration technique used in this case compare with the trig-
onometric substitution method used in Problem 28?
34. Length of a Spiral in Polar Coordinates: In Problem 1 S of Problem
Set 8-9, you found the length of the spiral with polar equation
r= o.se.
The spiral is shown in Figure 9-6j. ow that you know how to
integrate by trigonometric substitution, you can find the exact
length . Find the length of the part of the spiral shown using the
fundamental theorem. Compare this answer with the value you Figure 9-6j
get by numerical integration.
Section9-6:Integration
byTrigonometric
Substitution 4 61
V
35. Trigonometric Substitution for Negative Values ofx: In trigonometric substitutions you
let x / a equal sine, tan e, or sec e. If x is negative, then e is not in Quadrant I. Thus
8=sin - 1 ~. 8 = tan - 1 ~. or 0 = sec 1
~.
a a a
If you restrict e to the other quadrant in the range of the inverse trigonometric
function, you find the same indefinite integral as if you naively assumed that e is
always in Quadrant I. Show that this is the case for each of these three trigQr_iometric
substitutions.
9-7 Integration of Rational Functions by
Partial Fractions
In unrestrained population growth the rate of change of the population is propor-
tional to the number of people. This happens because the more people there are,
the more babies are born each year. In restrained population growth there is a max-
imum population a region can sustain. In this case, the rate of population growth
is also proportional to how close the number of people is to that maximum. For
instance, if the region can sustain 10.5 million people, then a differential equation
for population growth could be
dP
dt = 0.038P(l0.5 - P),
where P is population in millions, and 0.038 is the proportionality constant. You
may already have seen this differential equation in connection with the logistic
equation in Section 7-4. Separating the variables and integrating gives
f P(lO .~ _ P) dP = 0.038 f dt.
The integral on the left-hand side contains a rational algebraic function of P. That
is, the integrand can be written as (polynomial) / (polynomial). In this section you
will learn an algebraic method to find the antiderivative on the left-hand side of
the equation . The method involves breaking the rational expression into a sum of
relatively simple partial fractions, each of which is easy to integrate . In theory, at
least, many ratios of polynomials that have a numerator of degree lower than the
denominator degree can be vvritten as
Polynomial constant constant constant
--- + - - - + ... + --- .
Polynomial linear linear linear
OBJECTIVE Find the integral of a rat ional algebraic funct ion by first reso lving the integr and into
partial fractions .
462 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
Example 1 (Heaviside Method) Evaluate the indefinite integral: J X 2 4+x t
X
41
- 10
dx .
Solution The first step is factoring the denominator. The rational expression thus
becomes
4x + 41
(x+5)(x-2)"
Your thought process goes something like this: "Hmrnm ... It looks like someone
has been adding fractions, where (x + 5)(x - 2) is the common denominator!"
So you write
4x + 41 A
----- = -- + -- B .
(x+5)(x-2) x+5 x- 2
A clever way to isolate the constant A is to multiply both sides of the equation
by (x + 5).
4x + 41 A B
(x +5)( )( )=(x + 5) -- +(x +5 ) --
x +5 x - 2 x +5 x- 2
4x+41 = A+(x + 5)__!__
x -2 x -2
Substituting - 5 for x in the transformed equation gives
4( -5) + 41 =A+ (-5 + 5)-B - =A + 0.
-5-2 -5-2
:. - 3 = A
Similarly, multiplying both sides by (x - 2) isolates the constant B.
4x + 41 A B
(x - 2 ) (x + 5)(x - 2) = (x - 2 ) x + 5 + (x - 2 ) x - 2
4x + 41 A
--- = (x- 2)-- +B
x+5 x +5
Substituting 2 for x eliminates the A term and gives
4 ( 2 ) + 4 l =( 2 - 2)~ + B=O + B.
2+5 2+5
:. 7 = B
Substituting -3 for A and 7 for B, and putting the integra l back together gives
I x 4x+ 3x+ 41- 10 dx = J (--=l_
2
+ _7_) dx
x +5 x - 2
= -3 ln Ix + 51 + 71n Ix - 21 + C.
Transforming into partial fractions as shown in Example 1 is called the Heaviside
method after Oliver Heaviside (1850-1925). The method can be shortened enough
to be done in one step in your head! Here's how.
Section 9-7: Integration by Partia
of Rational Functions l Fractions 463
Example 2 (Heavisid e Shortcut ) Inte grate by res olving into partial fractions: J (x - x5~ x2 - 1
) dx
Solution Thou ght pro cess :
Write th e int egral and th e denominators of the parti al fractions .
f
2
_ x_-- - dx =
(x - 5)( x - 1)
f (- +-
x- 5 x- 1
) dx
Tell yourself, "If xis 5, th en (x - 5) equals zero." Cover up th e ( x - 5) with
your finger, and substitut e 5 into what is left .
f _8_x,--(-!-
_-1) dx ' hm . 5 - 2
Do ant
5 1 4
3
enc: _ = .
The an swer, 3 / 4, is th e num erator for (x - 5). To find the numerator for
(x - 1), repeat th e pro cess , but cover up th e (x - 1) and substitute 1, the
numb er that mak es (x - 1) zero .
f (x- ~)-82 dx Do arithm etic:
1
1- 5
-
2
= !.
4
Finger
Fill in th e 3 / 4 and 1 / 4 wher e th ey belong. The entir e pro ces s is just on e st ep,
like this.
f (x - x5)(x
- 2
- 1)
dx -- f (-x --5 + --x - 1 )d X
= ln Ix - 5 I + ln Ix - 11+ C
In Probl em Set 9-7, you will pr actice int egratin g by par tial fraction s. You will also
find out what to do if the denomin ator has
Unfac torabl e quadratic factors
Rep eated linear factor s
Problem Set 9.7
DoTheseQuickly
The following probl ems ar e int end ed to refr es h your skills. You should be able to do all
ten probl ems in less than five minut es .
Q1. Factor: x 2 - 25
Q2. Multiply: (x - 3) (x + 5)
Q3. Factor: x 2 - 4x - 12
Q4. Multipl y: (x + 7) 2
QS. Factor : x 2 + 8x + 16
Q6. Multiply: (x - 8) (x + 8)
,..
Ql. If f (x) = lnx, th en r- (x) =
1 - ?- ,
46 4 Chapter
9: Algebraic
Calculu
s Techn
iquesfor theElementary
Functions
QB. For the data in the tabl e, use Simpson's rule to find J; g (x) dx .
X g(x)
1 10
3 15
5 16
7 14
9 13
Q9. Show th at x 2 + 50x + 1000 cannot be facto red usin g real numbers only.
Q10. Show that x 2 + 36 cannot b e factored using real numb ers only .
For Probl ems 1-10, in tegrate by first reso lving the integrand into partial fra ctions .
1. J llx -15 d 2. J 7x+25 d
x2 - 3x + 2 x x2 - 7x - 8 x
3.
f (5x- ll ) dx
4.
f (3x - 12) dx
x2 - 2x - 8 x2 - 5x - 50
5. J 21 dx 6. f lOxdx
x 2 + 7x + 10 x2 - 9x - 36
7.
f 9x 2 - 25x - 50
dx 8. f 2
7x + 22x - 54 d
(x + 1) (x - 7) (x + 2) (x - 2)(x+4)(x-1) x
J 4x 2 + l 5x - 1 d 10. J - 3x 2 + 22x - 3 1 d
9.
x3 + 2x 2 - 5x - 6 x x3 - 8x 2 + l 9x - 12 x
Improper
Algebraic
Fractions
If the num era tor is of higher degree than th e denominator , long division will reduce th e
integrand to a polynomial plus a "prop er" fraction . For instanc e,
x3 - 9x 2 + 24x - 17 x- 2
----- 2 - -= x - 3+--- 2 - .
x - 6x + 5 x -6x + 5
For Problems 11 and 12, perform the integration b y first dividin g.
l l. J 3x 3 + 2x 2 - l 2x + 9 dx 12_ J x3 - 2
7x + 5x + 40 dx
x- 1 x2 - 2x - 8
Unfactorable
Quadratics
Heaviside's method does not work for integrals such as
7x 2 - 4x d
f (x 2 + l )(x - 2) x
9-7: Integration of Rational Functions by Partial Fractions
Section 465
'W zmrz:EiltifDi a mm@ mfffllrrmmzz r
that have an unfactorable quadratic in the denominator (unless you are willing to use
imaginary numbers!). However, the quadratic term can have a linear numerator. In this
case you can write
7x 2 - 4x Ax + B C
------=---+--
(x2+ 1)(x-2) x2 + 1 x- 2
(Ax+ B)(x - 2) + C(x 2 + 1)
(x2 + l)(x - 2)
Ax 2 2Ax + Bx - 2B + Cx 2 + C
-
(x 2 + l)(x - 2)
So Ax 2 + Cx 2 = 7x 2, - 2Ax + Bx = - 4x, and - 2B + C = 0. Solving the system
A+ + C= 7
- 2A + B =-4
-2B + C = 0
gives A = 3, B = 2, and C = 4. Therefore,
7x 2 - 4x 3x + 2 4
--- - --=--+-- 2
(x2 + 1)(x - 2) x + 1 x - 2
For Problems 13 and 14, integrate by first resolving into partial fractions.
4x 2 + 6x + 11 J ----4x 2 - l 5x ---dx
1
13
f -- 2 --
(x +l)(x + 4)
- dx 14.
x 3 -5x 2 + 3x + l
RepeatedLinearFactors
If a power of a linear factor appears in the denominator, the fraction could have come
from adding partial fractions with that power or any lower power. For instance, the
integral
2
f (xx + 4 4x)(x +_ 18l) 2 dx can be wntten
- . f [x A+ + x B_ + (xC +_ Dxl) 2 ] dx.
4 1
However, the numerator of the original fraction has only three coefficients, and the
right-hand side of the equation has four unknown constants. So one of the constants is
arbitrary and can take on any value you decide. The smart move is to let D = 0 so that
there will be three partial fractions that are as easy to integrate as possible.
For Problems 15 and 16, integrate by resolving the integrand into partial fractions.
2 2
15 _ J 4x + 18x + 62 dx 16 _ J 3x3 - 53x2 + 245 dx
(x + 5)(x + 1) x - 14x + 49x
"Old Problem" New Problems: Sometimes a problem that seems to fit the pattern of a new
problem actually reduces to an old problem. For Problems 17 and 18, evaluat e the integral
with this id ea in mind.
dx 1
17.
f - 3 -----
x - 6x + 12x - 8
2
- 18.
J -- -- ----
x + 4x + 6x 2 + 4x + 1
4 3
dx
19. Rumor Problem: There are 1000 students attending Lowe High. One day 10 students
arrive at school bearing the rumor that final exams will be canceled! On the average,
each student talks to other students at a rate of 2 students / hour, passing on the
rumor to students, some of whom have already heard it and some of whom have
466 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
not. Thus the rate at which students hear the rumor for the first time is 2 times the
number who have already heard it, times the fraction of the students who have not
yet heard it. If y is the number of students who have heard the rumor at time t hours
since school started, then
dy _ 1000 - y
2
dt - y 1000
a. Solve this differential equation algebraically,
subject to the initial condition that y = 10
when school started at t = 0. ////////////////
b. How many students had heard the rumor I I I I I I I I I I I I I I I I
I I I I I I I I I I I I I I I I
after the first hour 7 At lunch time (t = 4)? At
IIIIIIIIIIIIIIII
the end of the school day (t = 8)? IIIII/IIIII/IIII
c. How many students had heard the rumor IIII/IIIIIIIIIII
at the time it was spreading the fastest? At I I I I I I I I I I I I I I I I
what time was this? I I I I I I I I I I I I I I I I
//////////////// t
d. Figure 9-7a shows the slope field for the
4 8
differential equation . Plot the solution in
19a on a photocopy of the figure. How does Figure 9-7a
the curve relate to the slope field?
20. Epidemic Problem: A new dis ease arrives at the town of Scorpion Gulch. When a person
with the disease contacts a person who has not yet had the disease, the uninfected
person may or may not catch the disease. The disease is not fatal, but it persists with
the infected person forever after . Of the N people who live there, let P be the number
of them who have the disease after t days .
a. Suppose that each person contacts an average of 3 people per day. In terms of P
and N, how many contacts will there be between infected and uninfected people
per day?
b. Suppose that the probability of passing on the disease to an uninfected person is
only 10% at each contact. Explain why
dP =0 3PN - P
dt . N .
c. Solve the differential equation in 20b for P in terms of t. Use the initial condition
that P = Po at time t = 0.
d. If 1000 people live in Scorpion Gulch and 10 people are infected
y
at time t = 0, how many will be infected after 1 week?
e. For the conditions in 20d, how long will it be until 99% of the
population is infect ed?
21. Area Problem: Find an equation for the area of the region under
the graph of
j 2 b
X
25
Y = x 2 + 3x - 4
from x = 2 to x = b, where b is a constant greater than 2 (Figure
9-7b). Let b = 7 and check your answer by numerical integration .
(l
Does the area approach a finite limit as b approaches infinity? Figure 9-7b
Justify your answer .
Section
9-7:Integration
of Rational by PartialFractions
Functions 467
22. Volume Problem: The region in Probl em 21 is rotated about the y-axis to form a solid.
Find an equation in term s of th e constant b for the volume of the solid. Check your
answer by numeri cal integration using b = 7. Does the volume approach a finit e limit
as b approaches infinit y? Justify your ans wer.
23. Equivalent A nswers Problem: Evaluat e this int egral thre e ways.
3 dx
fx X -
2 - 6x + 8
a. By first resolving into partial fractions.
b. By completing the square, followed by tri gonom etric substitution.
c. Directly, as th e int egra l of the reciprocal fun ction .
d. Show that th e thre e answers are equivalent.
24. Logisti c Curve Probl em, A lg ebraically : For um estrained population growth, the rate of
change of population is directly proportional to the population. That is, dp / dt = kp,
where p is popul ation, t is time, and k is a constant . One assumption for restrain ed
growth is that there is a certain maximum size m for th e population, and th e rate goes
to zero as the popul ation approaches that size. Use this information to ans wer the
following questions.
a. Show that the differential equation dp / dt = kp(m - p ) has the properti es men-
tioned .
b. At what value of p is th e growt h rate the greates t?
c. Separate the variabl es, then solve the equ ation by inte gratin g. If you have worked
correctly, th e int egra l on one sid e of the equation can be evaluate d by partial
fractions .
d. Transform your answer so that p is explicitly in terms of t . Show that it can be
expressed in th e form of the logistic equation ,
1+b
P = Po l + be - kt '
where p 0 is th e population at tim e t = 0 and b is a constant.
e. Census figur es for the United States are as follows.
1960: 179. 3 million
1970 : 203.2 million
1980 : 226 .5 million
Let t be th e number of years that ha ve elaps ed since
1960. Use these as initial conditions to evalu ate p 0 , b,
and k. Write th e particular solution.
f. Predict th e out com e of the 1990 census . How clos e
does your answer come to the actual 1990 population,
248.7 million ?
g. Based on the logist ic model what will be the ultimate U.S. population?
h. Is this mathem atical model very sensitive to the initial conditions ? For inst ance,
suppose that the 1970 popu lation had really be en 204 .2 milli on instead of 203.2
million . How much would this affect the pr edicted ultimat e popu lation?
468 Chapter
9: Algebra
ic Calcu
lusTechniques
fortheElementary
Functions
9-8 Integrals of the Inverse Trigonometric Functions
At the beginning of this chapter it was mentioned that there are thr ee types of
functions that , along with their inv er ses, come under th e category elementary
transcendental functions. They ar e as follows:
Trigonometric Inverse trigonometric
Logarithmic Inverse logarithmic (exponential)
Hyperbolic Inverse hyperbolic
The name transcendental implies that the operations needed to calculate values
of the functions "transcend," or go beyond, th e operations of algebra (addition,
subtraction, multiplication, division, and root extraction). You have already learned
algebraic calculus techniques for the first four types of functions, except for
integrating the inverse trigonometric functions . In this section you will learn how
to integrat e the inverse trigonom et ric functions algebraically. In Section 9-9, you
will explore the hyperbolic functions, which are related both to exponential and
trigonometric functions.
OBJECTIVEBe able to integrate (antidifferentiate) each of the six inverse trigonometr ic functio ns.
Background:
Definition
andDerivatives
of theInverseTrigonometric
Functions
In Section 4-5, you learned definitions of the in verse trigonom etric functions, and
how to find algebraic formul as for their d er ivati ves . These definitions and derivative
formulas are r epeat ed her e for easy ref erenc e.
Definitions:InverseTrigonometric
Functions
(PrincipalBranches}
y = sin- 1 x if and only if siny = x and Y E [-'f , 'f ]
1
y = cos- x if and only if cosy = x and y E [0, rr]
y = tan - 1 x if and only if tarry = x and y E (- 'f, 'f )
y = cor- 1 x if and only if coty = x and y E (0, IT)
1
y = sec x if and only if secy = x and y E [O,rr], buty *'f
y = csc- 1 x if and only if csc y = x and y E [-'f, 'f ], buty*O
Note: The names arcsin, arccos, arctan, arccot, arcsec, and arccsc can be used to help
distinguish, for instance , tan - 1 x from 1/ tanx .
Sectio
n 9-8: Integralsof the InverseTrigonometr
ic Functions 469
Properties:Derivativesof the Six InverseTrigonometric
Functions
d 1
-d ( sin-
. 1 x ) = --- 1 1
dx (cos- x) = - J I - x 2
dx v1T'=x2
1
d
dx (tan - 1 x) = _l _+_x_2 .!!_(coc 1 x) = -- 1-
dx 1 + x2
1 .!!_(csc- 1 x) = - 1
.!!_(sec
1 x) = -~ ==
dx lxlJx'T=I dx lx1Jx 2 - 1
Note: Your grapher must be in the radian mode.
Memory Aid: The derivative of each "co-" inverse function is the opposite of the derivative
of the corresponding inverse function because each co-inverse function is decreasing as x
starts increasing from zero.
Integrals
The te chniqu e for integrating invers e trigonometric fun ctions is (surprisingly!)
integration by parts. Example 1 shows you how this is don e.
Example 1 Integrate : Jtan - 1 x dx
Solution
dv
xdx
1
= x tan - x -
f --x 2 + 1 ~ 1
1
2 X
= x tan - 1x - ln jx2 + 11+ C x +1
The last integral can be transformed to the integral of the reciprocal function .
The absolute value in the argument of ln is optional since x 2 + 1 is always
positive.
In Problem Set 9-8, you will derive algebraic formulas for the integrals of the other
five inverse trigonometric functions. These formulas are listed in the following
box. They were more important in the first 300 years of calculus, before such
technology as your grapher made numerical integration quick and easily accessible.
Like climbing Mount Everest, they are int eres ting more from th e standpoint that
they can be done, rather than becaus e th ey are of great practical use .
Properties:AlgebraicIntegralsof the InverseTrigonometric
Functions
Jsin- 1x dx = x sin- 1x + ~ + C
Jcos- 1 x dx = x cos- 1 x - -J1 - x 2 + C
Jtan - 1xdx = xtan - 1 x - m lx2 + 11+ C = xtan - 1 x - ln R+l +C
Jcor- 1x dx = x cot - 1 x + ln lx2 + 1 I+ C = x cor- 1 x + ln J x 2 + 1 + C
Jsec- 1xdx = xsec 1
x- sgnxlnlx +-Jx2 - 11+ C
Jcsc- 1xdx = xcsc 1x + sgnxln Ix+ -Jx2 - 11+ C
4 70 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
Problem Set 9-8
DoTheseQuickly
The followin g problems are int ended to refresh your skills. You should be able to do all
ten prob lems in less than five minutes.
QI. To integrate a product, use-? - .
Q2. To integrate a rational function, use -?-.
Q3. To integrate f .Jx2+I dx, use th e-?- trigonometric substitution.
Q4. To integrate f R--=-I dx, use th e-?- trigonometric substitution.
QS. To integrate f .Jl - x dx, use th e-?-
2 trigonom etric substitution.
Q6. Integrat e: f (x + l )7(x dx )
2
Ql. If f(x) = 3 + Ix - 5[, then the maximum of f (x) on [l, 6) is -?- .
QB. If f(x) = 3 + [x - 51, then the minimum of f(x) on [l, 6) is -? - .
Q9. If f(x) = 3 + [x - 5[, then f'(5) is-?-.
. If h(x ) = x 3 + x, then the graph of h has a point of inflection at x = -? - .
QJO
For Problems 1-6, find the indefinite integral. Check your answer against thos e in the
preceding box .
1. Jtan - 1 xdx 2. Jcot - 1 xdx 3. Jcos - ix dx
4. Jsin - 1 x dx 5. Jse c 1
x dx 6. Jcs c 1
x dx
7. Answer Verification Problem: Evaluate th e int egral f{ tan - 1 x dx algebraically . Find a
decimal approximation for the answer. Th en evaluate the int egral numeri cally. How
close does the numerical answer come to the exact, algebrai c answer?
8. Simpson's Rule Review Problem: Plot the graph of y = se c 1 x
from x = l to x = 3. You may do this in parametric mode, with
x = l /c ost and y = t . Sketch the graph . Use Simpson's rule with Jr
2 i-------,
n = 10 incr ements to find a numerical approximation for the area
of the region under this graph from x = l to x = 3. Then find th e
exact area by integrating algebraically . How close does the answer
using Simpson's rule come to the exact answ er?
9. Area Probl em: Figure 9-8a shows the region above the graph of
y = sin - 1 x, below y = rr / 2, and to the right of the y-axis. Find the
area of this region by vertical slices . Find th e ar ea again, this time X
using hori zontal slices. Show that the two answers are equivalent .
Figure 9-8a
10. Volume Probl em: Figure 9-8b shows the region under th e
graph of y = tan - 1 x from x = Oto x = l, rotated about the y= tan- 1 x
y-axis to form a solid . Find the exact volume of th e solid Jr
4
using the fundamental theorem . Show that your answer is ..
reasonable by numerical integration and by comparing it
with a suitable geometric figure . , ....... :..-::.:.... ~---
.. .. -~.
,
X
Figure 9-8b
Sec
tion 9-8:Integra
ls of the Inverse
Trigonomet
ric Functions 471
9-9 Calculus of the Hyperbolic and Inverse
Hyperbolic Functions
Figure 9-9a shows what a chain might look like if suspended between two nails
driven into the frame of a chalkboard. Although the shape resembles a parabola,
it is actually a catenary from the Latin catena, meaning "chain ." Its graph is the
hype rbolic cos ine function,
y = a + b cosh ex
(pronounced "kosh," with a short "o"). As shown in Figure 9-9b, a parabola is more
sharply curved at the vertex than a catenar y is .
Figure 9-90 Figure 9-9b
Cosh and the related hyperbolic sine and tangent are important enough to be
on most graphers. In this section you will see that the hyp erbolic functions are
related both to the natural exponential function and to the circular sine and cosine
functions. You will see why a chain hangs in the shape of a hyperbolic cosine and
why the functions are called hyp erbolic.
Definitions
of the SixHyperbolic
Functions
"Out of a clear, blue sky," define two functions, u and v :
and
The u graph is identical to y = coshx, as you could see by plotting y = u and
y = coshx on the same screen (Figure 9-9c, left) . The v graph is identical to
y = sinhx (pronounced "sinch of x," although the letter "c" does not appear in
writing) . The right -hand diagram in Figure 9-9c shows y = sinhx and y = v. The
dashed graphs are y = o.se x and y = O.se-x.
y =u y= v
= coshx_.,.- = sinhx
...
:: ..-:::. X
.. .. -~ -
Figure 9-9c
472 Chapter
9: Algebraic
Calculus
Techniques
fortheElementary
Functions
The reason these functions are called "hyperbolic" shows up if you eliminate x and
get an equation with u and v alone. Squaring u and v, then subtracting gives the
following.
u2 = (e2x + 2 + e- 2x) Why just 2 for the middl e term ?
y 2 = (e2X_ 2 + e- 2X)
u2 - v2 = 1 A unit equilateral hyperbola in the uv-coordinat e system.
This is the equation of a hyperbola in a uv-coordinate system . Function u is the
horizontal coordinate of a point on the hyperbola, and function v is the vertical
coordinate. These coordinates have the same relationship to the unit equilateral
hyperbola u 2 - v 2 = 1 as the "circular" functions cosine and sine have to the unit
circle, u 2 + v 2 = 1 (Figure 9-9d).
/
V / V
/
/
//
1
( u, v)
/
(Negative /
/
branch is /
/ : v = sinhx
extraneous.) /
/
/
/
/
/ u u
<- -- 1
''
''
'
''
',, u2 - v 2 = 1
'
''
''
Hyperbolic functions Circular functions
Figure 9-9d
The other four hyperbolic functions are defined in the same way as the other four
circular (trigonometric) functions, namely, as reciprocals and quotients of the first
two. The symbols come from adding an "h" to the corresponding circular function
symbol. There are no widely accepted pronunciations for symbols such as tanh x
and cschx other than "hyperbolic tangent of x," and so on .
Definitions:TheHyperbolic
Functions
sinhx = (ex - e-x)
sinhx coshx
tanhx = - - cothx = --
coshx sinhx
1 1
sechx = -coshx
- cschx = - -
sinhx
From the fact that u 2 - v 2 = 1 and the above definitions, the Pythagorean properties
follow.
Section
9-9:Calculus
of the Hyperbolic
andInverse
Hyperbolic
Functions 4 73
au ..;;;;;
PythagoreanPropertiesof the HyperbolicFunctions
cosh2 x - sinh2x = 1
1 - tanll x = sech2 x
coth 2 x - l = csch2 x
OBJECTIVE Be able to ilifferentiate and integrate any of the six hyperbolic functions and their in,erses,
and use the hyperbolic cosine function as a mathematical model.
Derivatives
of theHyperbolic
Functions
The derivatives of cosh and sinh hav e an interesting relationship .
and .!:!_sinhx = .!.(e"
2
+ e-") = coshx
dx
This "cyclical" property of the derivatives is similar to that of the circular sine and
cosine.
sin ' x = cos x, and cos' x = - sinx.
The derivatives of the other four hyperbolic functions can be found by first
transforming to sinhx and coshx . Example 1 shows you ho w.
Example 1 Find the derivative: y = coth x
Solution By definition of coth,
coshx
Y = sinhx
Applying the derivative of a quotient property gives the following.
(sinhx)(sinhx) - (coshx)(coshx)
y'
sinh 2 x
sinh 2 x - cosh 2 x
-1
By the Pythagorean prop erties .
sinh 2 x
y' = - csch 2 x
Th e derivatives of all the hyperbolic functions are summariz ed in the following box.
In Problems 3 7 and 38 of Problem Set 9-9, you will be asked to derive some of thes e.
Properties:Derivativesof HyperbolicFunctions
sinh'x = coshx cosh'x = sinhx
tanh' x = sech 2 x coth' x = - csch2 x
sech'x = -sec hx tanhx csch'x = - cschx cothx
474 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
Integrals
of Hyperbolic
Functions
The integrals of cash and sinh come directly from the derivative formulas. The
integrals of tanh and coth are found in much the same way you integrated tan and
cot . Example 2 shows how coth is integrated. The integrals of sech and csch require
clever substitutions you have not yet learned. 1n Problems Cl and C2 of Problem Set
9-13, you will see how to do this.
Example 2 Integrate: f coth x dx
Solution The technique is to use the definition of coth. The resulting quotient has a
numerator that is the derivative of the denominator. You wind up integrating
the reciprocal function.
f cothxdx
= f coshx dx
sinhx
= f Sl
1
.nh coshx dx
X
= ln lsinhx l + C
The integrals of the six hyperbolic functions are listed in the following box. In
Problem Set 9-9, you will be asked to derive some of these.
Properties:Integralsof the HyperbolicFunctions
Jsinhxdx = coshx + C f coshxdx = sinhx + C
J tanh xdx = ln(coshx) + C Jcoth xdx = ln lsinhxl + C
Jsechxdx = sin - 1 (tanh x) + C Jcschxdx = ln I tanh (x / 2) I + C
InverseHyperbolic
Functions
Recall that the inverse of a function is the relation you obtain by interchanging the
two variables . For instance, if y = sinhx, then the inverse relation has the equation
X = sinhy.
The dependent variable y is called the inverse hype rbolic sine of x. The symbol
often used is similar to that for inverse circular functions,
y = sinh - 1 x.
This is read, "Inverse hyperbolic sine of x," or simply "sinch inverse of x ." The term
y = argsinhx is also used, in honor of the fact that y is the "argument whose sinh
is x." This symbol parallels y = arcsinx for the inverse circular functions.
Section
9-9: Calculusof theHyperbo
lic andInverseHyperbo
lic Functions 475
The graphs of y = sinh x and y = sinh - t x are shown in Figure 9-9e. As is true vvith
any function and its inverse, the two graphs are reflections of each other in the line
y = X.
y ,'y = sinh x
,,
/ .,,..1.1 )' = X
,,
/
'
Figure 9-9e
Derivatives
of theInverseHyperbolic
Functions
To differentiate an inverse hyperbolic function, just use the definition to transform
back to th e natural function. Then differentiate implicitly.
a Example 3 Find the derivative: y = cosh - 1 x, for x > 1.
Solution y = cosh - lx
coshy = x
Sln11y -dy =l dy / dx comes fr om the chain rul e.
dx
dy 1
Divide each memb er by sinh y .
dx sinhy
dy 1
cosh 2 y - sinh 2 y = 1, so sinh 2 y = cosh 2 y - 1.
dx _/cosh 2 y - 1
dy
dx
1
Jx"2-=-I
cos h y = x
Differentiation formulas for the other five inverse hyperbolic functions can be
derived in a similar way, as you will do in Problem 39 of Problem Set 9-9. The six
derivatives are summarized in the following box.
Properties:InverseHyperbolicFunctionDerivatives
!!_ (sinh - 1 x) = __ i-_ d 1
-d (cosh - 1 x) = -J 2 , x > 1
dx R+1" X X - 1
1 1
d ( nh - 1 )
dx ta x =1- xz ' Ix I < 1 d (cot h - 1 x ) = - --
-d , Ix I > 1
X 1 - X2
d 1 d ( h-1 ) 1 *Q
dx (sech - 1 x) = - x~ x in (O, 1) dx csc x = - lx l../l"+x2'
X
476 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
Integrals
of InverseHyperbolic
Functions
The integrals of the inverse hyperbolic functions can be found by straightforward
integration by parts. Example 4 shows you how.
Example 4 Integrate: Jsinh - 1 xdx
Solution Jsinh - 1xdx u dv
1
= xsinh - x - J kxdx ~l
X +1
----=--X
= x sinh - 1 x - J(x2 + 1) - 112 (2x dx)
f
= xsinh - 1 x - (x 2 + 1) 112 + C Can you give a reason?
= x sinh - 1 x - (x2 + 1) 112 + C
The following box lists the indefinite integrals of the inverse hyperbolic functions.
You should understand that these properties exist, and should know how to derive
them if you are called upon to do so. However, unless you plan to make a career out
of integrating inverse hyperbolic functions, there is no need to memorize them!
Properties:InverseHyperbolicFunction
Integrals
f sinh- x dx = x sinh - x - (x + 1) + C
1 1 2 1 12
f cosh- xdx = xcosh - x- (x + 1) + C, x > 1
1 1 2 112
f tanh - xdx = xtanh - x+ln ll-x 1+C, lxl <l
1 1 2
f coth - xdx = xcoth - x+ln ll-x 1+C, lxl > 1
1 1 2
f sech- x dx = x sech- x + sin- x + C, x in (0, 1)
1 1 1
f csch- xdx=xcsch - x+ sgnxsinh - x+C, X*O
1 1 1
Hyperbolic
Cosine
as a Mathematical
Model
A chain hangs in the shape of the catenary
y = kcoshfx + C,
where k and C stand for constants. In Problem 25 of Problem Set 9-9, you will learn
why this is true. Example 5 shows you how to derive the particular equation.
Example 5 A chain hangs from above a chalkboard (Figure 9-9f). Its ends are at (90, 120), and
its vertex is at (0, 20), where x and y are in centimeters along and above the chalk
tray, respectively.
a. Find the particular equation of the catenary .
Section
99:Calcu
lusof theHyperbol
ic andInverse
Hyperbolic
Functions 477
: eeteeeerttt @
b. How high is the chain above the chalk tra y when x = 50?
c. At what values of x is th e chain 110 cm above the chalk tray ?
10
Figure 9-9f
Solutions a. y = k cosh fx + C Write th e general equation of th e
catenary.
20 = k cosh f (0) + C = k + C Sub stitut e (0, 20); cosh O = 1.
120 = kcosh t (90) + C Sub stitut e (90, 120).
Eliminate C by subtr actin g
100 = k cosh ~ - k equation s.
0 = kcosh ~ - k - 100 ~ k = 51.780122 ... Use your gra ph er's solve fea tur e.
20 = 51. 78 .. . + C ~ C = -3 1.78 ...
. . y = 51.78 ... cosh 5 1.;sx - 3 1.78 .. .
0
b . y = 51.78 ... cosh 51\ 8 __ - 31.7 8 . .. = 46 .07 55 ...
The chain is about 46.l cm above the chalk tray when x = 50.
c. 110 = 51.78 ... cosh 5
1.; x -
8
3 1.78
cosh 5 1.; x = 2.7381 .. .
8
;
51 8
x = c osh - 1 2.7381 . . . = 1.66526 .. .
X = 86.2278 ...
The chain is about 110 cm above the chalk tray when x"" 86.2 cm
Problem Set 9.9
DoTheseQuickly
The following problems are intended to refr esh your skills. You should be able to do all
ten problems in less than five minutes .
QI. What trigonom etric substitution should b e us ed for f (x 2 + 5) 312 dx ?
Q2. Integrat e: Jxe x dx
Q3. Integrat e: Jsec 2 3x dx
478 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
Q4. Integrate : x- 112 dx f
QS. Integrate: fx- 1
dx
Q6. f sin nxdx = ;sin n- 1 xcosx + n f
~ 1 sin n- 2 xdx is calleda (n ) -?-.
Ql. True or false: sec' x = In Isec x + tan x I + C
QB
. Write the formula for dL, the differential of arc length.
Q9. Nam e the appropriate int egration method for J(x + 5) / [ (x - 3) (x + 2)] dx.
. If f (x )
QJO = sin - 1x, then f' (x ) = - ?- .
1. Hyperbolic Function Graphing Problem: Sketch the graphs of each of the six hyperbolic
functions. You may plot these on your grapher first to see what they look like. The
graphs of coth, sech, and csch ma y be plott ed by taking advantage of their definitions
to writ e the equations for thes e functions in terms of functions that appear on your
grapher .
2. Invers e Hyperbolic Function Graphing Problem: Sketch the graphs of the inverses of
th e six hyperbolic functions. This is most easily done in parametric mode on your
grapher, letting x = cosh t and y = t, for example. For each on e that is not a function ,
dark en what you think would be the principal branch Uust one value of y for each
valu e of x), and write an inequality that restricts the range to specify this branch.
For Problems 3-22, do th e integration or differentiation.
3. f (x ) = tanh 3 x 4. f(x ) = 5 sech 3x
5. f cosh 5
x sinhx dx 6. f (sinhx )- coshxdx
3
7. g (x) = cschx sinx 8. g(x) = tanx tanh x
9. f se ch 4x d x
2
10. f sech 7x tanh 7x dx
11. h (x) = x 2 coth x 12. h(x) = x 2 5 csch4x
13. r tanh xdx 14. r
-4
sinhxdx
sinh 5x cosh6x
15. q (x ) = 16. r (x ) =
ln3x cos 3x
17. f0
1
x sinhx dx 18. r x 2 coshxdx
19. y = 3 sinh - 1 4x 20. y = 5tanh - 1 (x 3 )
21. f tanh - 1
5x dx 22 . f 4 cosh - 6x dx
1
Hyperbolic Substitution Prob lems: For Problems 23 and 24, integrate by hyperbolic
substitution, using the fact that cosh 2 t = sinh 2 t + 1, and sinh 2 t = cosh 2 t - l.
23. f Jx 2 + 9dx 24. f Jx 2 - 25 dx
Section9-9:Colculus
of the Hyperbo
lic andInverseHyperbolic
Functions 479
25. Hanging Chain or Cable Problem : If a chain (or a flexible cable
that doesn't stretch) is hung between two supports, it takes
the shape of a hyperbolic cosine curve,
y
h w
v = - cosh - x + C
/ w h '
where x and y are horizontal and vertical distances to a point
V
on the chain, h is the horizontal tensile force exerted on the
chain, and w is the weight of the chain per unit length. In this
problem you will show why this is true. Figure 9-9g shows
the graph of the chain in an xy-coordinate system. Any point
on the chain experiences horizontal and vertical forces of
X
h and v, respectively . Force h is constant, and depends on
0 X
how tightly the chain is pulled at its ends . Force v varies and
equals the weight of the part of the chain below point (x, y). Figure 9-99
The resultant tension vector points along the graph.
a. Explain why the slope, y ' = dy / dx, of the graph at point (x, y) is equal to v I h.
b. The weight of the chain from O to x equals the length, s, times the weight per unit
length, w . Explain why this equation is true.
I w
y = - s
h
c. If you differentiate both sides of the equation in 23b, you get the differential
equation
w
d(y') = h ds
Using what you know about arc length, show that this differential equation can be
written
112
d(y') = : (1 + (y ' )2) dx
d. Separating the variables in the equation in 23c and integrating gives
f [1 + (y') 2] - 112 d(y ' ) = f : dx.
Perform a hyperbolic substitution on the left, letting y' = sinh t, and integrate to
find
sinl 1 - 1y , = w x + C .
11
e. Use the fact that y ' = 0 at the vertex, x = 0, to evaluate the constant of integration, C.
f. Based on the above work, show that
.nhw
I
y = Sl h X.
48 0 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Funct
ions
g. From 25f, show that the equation of the hanging chain is as shown in the following
box.
Property:Equationof HangingChainor Cable
h w
y = w cash h x + C
where w is the weight of chain per unit length,
h is the horizontal tensile force on chain (in units consistent with w),
xis the distance from the axis of symmetry to point (x, y) on the chain,
y is the vertical distance from the x-axis to point (x, y), and
C is a vertical distance determined by the position of the chain.
26. Can You Duplicate This Graph? Figure 9-9h shows th e graph of a
hyperbolic cosine function with general equation
1
y = k cosh k x + C.
a. Find the particular equation. Check your equation by showing
that its graph agrees with points in the figure.
b. Calculate y if x is 20.
c. Calculate x if y is 4. Show that your answer is consistent with Figure 9-9h
the graph .
d. Find the slope of the graph if xis 3. Show that a line of this slope through the point
on the graph where x = 3 is tangent to the graph .
e. Find the area of the region und er the graph from x = - l to x = 3.
f. Find the length of the graph from x = - l to x = 3.
27. Power Line Problem: An electrical power line is to be sus-
l- 300ft-J
p ended between pylons 300 ft apart (Figure 9-9i). The cable
weighs 0.8 lb/ ft, and will be connected to the pylons 110 ft Figure 9-9i
above ground .
a. It is planned to use a horizontal tensil e
forc e of h = 400 lb to hold the cable. Write
the particular equation of the resulting
catenary . Use the equation to calculate
how close to the ground the cable will
come.
b . How long will the cable in 27a be ? How
much will it weigh?
c. For the cable in 27a, where will the
maximum total tensile force be, at the
middle or at the ends ? What will this
maximum tension be equal to ?
d. The power company decides that the cable must come no clos er than 100 ft from
the ground. How high a horizontal force would be needed to achiev e this clearance
if the cables are still connected on 110 ft pylons?
Sect
ion9-9: Calculus
of theHyperbolic
andInverse
Hyperbolic
Functions 4 81
kl' 28. Hangi ng Chain Experiment: Obtain a length of chain; about 6-10 feet is reasonable .
Hang the chain between two convenient supports such as nails driven into the upper
frame of a chalkboard. Let the chain hang down fairly far, as shown in Figure 9-9f (page
4 78). Set up a coordinate system with the y-axis halfway between the two supports . The
x-axis can be some convenient horizontal line like the chalk tray . In this experiment
you are to derive an equation for the vertical distance from the x-axis to the chain and
to check it by actual measurement.
a. Measure the x- and y-coordinates of the two supports and the vertex. Use centime-
ters.
b . Find the particular equation of the particular catenary that fits the three data
points.
c. Make a table of values of y versus x for each 10 cm, going both ways from x = 0
and ending at the two supports .
d. Mark the chain links at the supports then take down the chain. Plot the points you
calculated in 28c on the chalkboard . Find some way to make sure the y-distances
are truly vertical. Then hang back the chain to see how well it fits the catenary .
e. Find the particular equation of the parabola (quadratic function) that fits the three
measured data points. Make a table of y-values as in 28c and plot these values on
the chalkboard . How does the parabola differ from the catenary?
f. Use your equation from 28b to calculate the length of the chain between the two
supports. Then stretch out the chain on the floor and measure it. How close does
the calculated value come to the measured value? y
29. Bowl Problem: The graph of y = sinhx from x = 0 to x = 1 is
rotated about the y-axis to form a bowl (Figure 9-9j). Assume
that x and y are in feet.
a. Find the surface area of the bowl.
b. The bowl is to be silver-plated inside and out. The cost of
plating is $ 5 7 per square foot. How much will plating cost?
X
c. How much liquid could be held inside the bowl if it were
filled to within a half-inch of the top?
30. Gateway Arch Problem: The Gateway to the West Arch in St. Figure 9-9j
Louis is built in the shape of an inverted catenary (Figure
9-9k). This shape was used since compression forces act
tangentially to the structure, as do the tensile forces on a
chain, thus avoiding bending of the stainless steel from which
the arch is constructed . The outside of the arch is 630 ft wide
at the base and 630 ft high. The inside of the arch is 520 ft
wide and 612 ft high.
a. Find particular equations of the inner and outer catenaries.
b. Verify that your equations are correct by plotting them on
your grapher.
c. The stress created by wind blowing against the arch
depends on the area of the region between the two graphs.
Find this area. Figure 9-9k
482 Chapter
9: Algebraic
Calculus
Techniques
for theEleme
ntaryFunctions
d. A spider starts at the point where the left end of the outside of the arch meets the
ground and crawls all the way up, then down the other side to the ground. As she
crawls she leaves one strand of web. How long is the strand she leaves?
e. How steeply must the spider climb when she first starts up?
f. A. V. Ator wants to fly a plane underneath the arch. The plane has a (total) vvingspan
of 120 ft. Below what height, y, can A. V. fly to ensure that each wing misses the
inside of the arch by at least 50 ft horizontally?
31. Derivative Verification Problem : For H(x) = cschx,
a. Find H '( 1) exactly, using the differentiation formula.
b. Find H'(l) approximately, using the symmetric difference quotient with ~x = 0.01.
By what percentage does the approximate answer differ from the exact answer?
2
32. Integral Verification Problem: Evaluate J1 sechx dx by the fundamental theorem,
using the antiderivative formula for the hyperbolic secant. Then evaluate the integral
approximately using numerical integration. How does the numerical answer compare
to the exact answer?
33. Integration by Parts Problem: Evaluate f exsinh 2x dx by parts. Then integrate again by
first using the definition of sinh to transform to exponential form. Show that the two
answers are equivalent. Which technique is easier?
34. Integration Surprise Problem! Try to integrate by parts.
Jexsinhxdx
What causes integration by parts to fail? Recalling the definition of sinh, figure another
way to do the integrations, and do it.
35. Derivations of the Pythagorean Properties of Hyperbolic Functions:
a. Starting with the definition of coshx and sinhx, prove that cosh 2 x - sinh 2 x = 1.
b. Divide both members of the equation in 37a by cosh 2 x, and thus derive the property
1 - tanh 2 x = sech 2 x.
2
c. Derive the property coth x - 1 = csch 2 x.
36. Double-Argument Properties of Hyperbolic Functions:
a. Explain why sinh 2x = (e 2x - e- 2x ).
b . Derive the double-argument property sinh 2x = 2 sinh x cosh x.
c. Derive the double-argument property cosh2x = cosh 2 x + sinh 2 x.
d. Derive the other form of the double-argument property, cosh 2x = 1 + 2 sinh 2 x.
e. Derive the property sinh 2 x = (cosh 2x - 1).
37. Hyperbolic Radian Problem: In trigonometry you learn that the argument, x radians, in
the circular functions sinx or cos x equals an arc length on the unit circle (Figure 9-91,
left). In this problem you will show that the same is not true for x "hyperbolic" radians
in sinhx and coshx. You will show that the arguments both in circular and hyperbolic
radians equal the area of a sector of the circular or hyperbolic region shown in Figure
9-91.
a. Show that the unit circle u2 + v 2 = 1 between u = cos 2 and u = 1 is 2 units long,
but the hyperbola u2 - v 2 = 1 between u = 1 and u = cosh 2 is greater than 2 units
long.
Section
9-9:Calculus
of theHyperbolic
andInverse
Hyperbolic
Functions 483
1 V V
u 2 -v 2 = 1
(coshx, sinhx)
x = area of
sector x = area of
sector
Figure 9-91
b . Show th at 2 is th e area of th e hyp erboli c sector in Figur e 9-9 1 with the point
(u,v ) = (cosh 2,sinh 2) as its upp er bound ary .
c. Show that xis the area of th e circul ar sector with th e point ( u, v ) = (cos x, sin x) as
it s upper boundar y.
d. Show in general that xi s the area of the hyp erboli c sector with th e point ( u , v ) =
(coshx, sinhx ) as its upp er bound ary.
38. A lgebrai c Derivativ es of the Oth er Five Inve rse Hyper bolic Fun ctions: Derive th e
differ ential formulas given in thi s section for th e followirlg expr ession s.
a. :x (sinh - 1 x)
b. :x(tanh - x)1
c. :x(coth - x)1
d
d. dx (sech- 1x )
d
e. dx (csch- 1x)
9-10 Improper Integrals
Suppo se that you are driving alon g th e highway at 80 ft/ sec (about 55 mi / hr ). At
tim e t = 0 sec you take your foo t off the accelerator and let th e car start slowing
down . Assum e that your velocit y is given by
v (t) = 8Qe- Olt ,
wher e v (t) is in feet per second . According to this math emati cal mod el the velocity
approach es zero as tim e increases but is never equ al to zero . So you are never quit e
stopp ed. Would th e dis tance you go appro ach a limitin g valu e, or would it incre ase
without bound ? In this section you will learn ho w to an swer such ques tion s by
evaluatin g improp er int eg rals.
4 84 Chapter
9: Algebra
ic Calculus
Technique
s fortheElementary
Functions
OBJECTIVE Given an improper int egral, tell wheth er or n ot it converges (that is, app ro aches a finite
numb er as a limit). If it does , find the numb er to which it converges .
Figure 9-1Oa shows the velocity function mentioned above . The distance the car
goes between t = 0 and t = b is equal to the area of the region under the graph.
Thus
Distance = r 80e -o.1tdt
b
= - 8ooe - 0.ll I0
= - 800e - 0 2b + 800 .
If b = 10 sec, the distanc e is 505.6 .. . ft. As b approaches infinity, the -800e - 02 b
term approaches zero . Thus the distance approaches 800 ft. The mathemat ical
model te lls you that the car never passes a point 800 ft from where you started
slowing. The integral
0
fo80e - O.ltdt
is called an improper integral because one of its limits of integration is not finite .
The integral converges to 800 because the integral from O to b approaches 800 as
b approaches infinity . Suppose the velocity function had been v(t ) = 320(t + 4) - 1 .
The graph (Figure 9-lOb) looks almost the same. The velocity still approaches zero
as time increases. The distance would be
Distance= r 320(t + 4) - 1 dt
b
= 320lnlt + 4 110
= 3201n lb + 4 1- 320ln4.
v(t ) v(t)
80 80 -
Area = distanc e travel ed Area = distance traveled
IO b 20 IO b 20
Figure 9-1Oa Figure 9-1Ob
As b approaches infinity, so does ln Ib + 41. The integral diverges . Unlike the first
mathematical model, this one tells you the car would go arbitraril y far from the
starting point if you waited long enough!
A definite integral is improper if the following hold.
The upper or lower limit of integration is infinite .
The integrand is discontinuous for at least one value of x at or between the limits
of integration.
Section
9-10: Improper
Integrals 485
Definition:ImproperIntegrals
00
f f(x) dx = lim (b f(x) dx
Ja b-ooJa
t oof(x) dx = a~~ oo J:
f(x) dx
f,abf(x) lb
dx = lim k f(x) dx
k-c + '
+ lim
k-c-
f.kf(x)
a
dx, f is discontinuous at x = c in [a, b]
An improper integral converges to a certain number if each applicable limit shown above
is finite. Otherwise, the integral diverges.
Note that an improper integral with an infinite limit of integration always diverges
if the integrand has a limit other than zero as the variable of integration approaches
infinity.
Example 1 For the improper integral J0
00
x 2 e - x dx,
a. Graph the integrand and tell whether or not the integral might converge .
b. If the integral might converge, find out whether or not it does, and if so, to what
limit it converges.
Solution a. Figure 9-1 Oc shows that the integral might converge because the integrand
seems to approach zero as x gets very large .
b. Replace oo with b and let b approach infinity.
roox 2 e - x dx = lim rb x 2 e - x dx
Jo b -oo Jo
X Integrating by parts twice gives
b
lim (- x 2e- x - 2xe - x - 2e -") I = lim( - b 2 e - b - 2be - b - 2e - b + 0 + 0 + 2) .
b -oo Cl b -oo
Figure 9-1Oc 2
The limit of b e - b can be found by two applications of !'Hospital's rule.
b2 (X)
limb 2 e - b = lim - - - Writ e th e expr ess ion as a quotient .
b -"' b -"" eb oo
2b oo Use !'Hospit al's rul e. (Tak e th e derivative of
= lim - - - numer ator and denominator. )
b-co eb (X)
= lim3_ - 3_ Use !'Hospit al's rule again.
b- coeb (X)
=0 Limit of th e form (finite/ infinite) is ze ro.
Similarly, the second and third terms in the limit each go to zero . Therefore,
lim( - b 2 e - b - 2be - b - 2e - b + 0 + 0 + 2) = 2.
b-co
Example 2 For the improper integral J0
00
x0 2 dx,
a. Graph the integrand and tell whether or not the integral might converge.
b. If the integral might converge, find out whether or not it does, and if so, to what
limit it converges.
486 Chapter
9: Algebraic
Calculus
Techniques
for the Elementary
Functions
Solution a. The graph in Figure 9-1Od indicates that the integral does not converge. Since
the limit of the integrand, x 0 2 , is not O as x approaches infinity, the integral
diverges .
Figure 9-1Od
b. Nothing remains to be done because the integral diverges.
Example 3 For the improper integral f 2
-2 X
.!.dx,
a. Graph the integrand and tell whether or not the integral might converge .
b. If the integral might converge, find out whether or not it does, and if so, to what
limit it converges.
Solution a. Figure 9-lOe shows the graph of the function. Since there is a discontinuity
at x = 0, you must evaluate two integrals, one from -2 to b and the other
from a to 2. You write
2
f-22 .!.
x
dx = lim
b - o-
f b .!.dx + lim J, .!.dx.
-2 x a-o + a x
Figure 9-1Oe
At first glance, you might think that the integral converges to zero . If a and
b are the same distance from the origin, then there is just as much negative
"area" below the x-axis as there is area above. In order for the integral
to converge, however , both of the integrals shown above must converge.
Checking the first one,
lim
b-o-
fb
-2
.!.
x
d x = lim ln Ix I I b = lim (ln Ib I -
b-o - -2 b- o-
ln I- 2 I) .
Since ln O is infinite, this integral diverges. Thus, the original integra l also
diverges .
b. There is nothing to be done for part b because the integral diverges .
Section9l 0: ImproperIntegrals 48 7
t1
Problem Set 9-1 0
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Q 1. Sketch: y = e-x
Q2. What function has a graph like Figure 9-1 Of?
Q3. What function has a graph like Figure 9- lOg?
Q4. What function has a graph like Figure 9- lOh?
Q5. What function has a graph like Figure 9- lOi?
Q6. f coshx dx = -?-
Ql. y = coshx =>y ' = -?-
QB. y = cosx =>y ' = -?-
Q9. f cosxdx = -?-
Q10. What is the maximum value of y if y = - x 2 + lOx + 7?
I
X ,x X X
Figure 9-1Of Figure 9-1Og Figure 9-1Oh Figure 9-1 Oi
For Problems 1-20,
a. Tell from the graph of the integrand whether or not the integral might converge.
b. If the integral might converge, find out whether or not it does, and if so, the limit
to which it converges.
1. r""_!_dx
h x2
2. r""_!_dx
h x4 3. f
l
00
!:.
X
dx 4. r1 }:_
Jo X
dx
5. f"x~.2 dx 6. f
1
00 _2_
xt. 2
dx 7. rl _ l_ dx
Jo xo.2
8. rl _2_
Jo xi.
2 dx
9.
oo
l 1+
0
dx
X
2 10. l
0
oo dx
1+X
11.
1 dx
1
o xlnx 12. r x(l!xx )2
13. f2""e-0.4xdx 14. fa""eo.0 2x dx 15. el Jx dx 16. r (x - 3) - 2 13 dx
17. J xe -x dx
00
0
18. fo3(x-1) - 2 dx 19. J cosxdx
0
00
20. Jo"
" sinx dx
488 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
21. Divergence by Oscillation Problem: The improper integrals in Problems 19 and 20 are
said to diverge by oscillation. Explain why these words make sense. A graph may help.
22. p-Integra/ Problem : An integral of the form y y
Ip = f
00
1
~
XP
dx,
where p stands for a constant, is called a p-
integral. For some values of the exponent p,
the integral converges and for others it doesn't. X X
Figure 9-lOj shows an example for which the
two graphs look practically identical, but only
one of the integrals converges. In this problem f-- 1
- dx
I XI.OJ f- X0.99
I
1
dx
Converges Diverges
your objective is to find the values of p for
which the p-integral converges and those for Figure 9-1Oj
which it diverges.
a. Show that Ip converges if p = 1.001, but not if p = 0.999.
b. Does Ip converge if p = 1? Justify your answer.
c. Complete the statement, "Ip converges if p -?-, and diverges if p -?-."
23. Volume of an Unbounded Solid Problem: Figure 9-lOk shows the region under the graph
of y = 1 / x from x = 1 to x = b.
y = 1/ x y y
Figure 9-1Ok
a. Does the region's area approach a finite limit as b approaches infinity 7 Explain.
b. The region is rotated about the x-axis to form a solid. Does the volume of the solid
approach a finite limit as b - oo?If so, what is the limit? If not, explain why not .
c. The region is rotated about the y-axis to form a different solid. Does this volume
approach a finite limit as b - oo?If so, what is the limit? If not, explain why not.
d. True or false: "If a region has infinite area, then the solid formed by rotating that
region about an axis has infinite volume."
Section9-10:Improper
Integrals 489
24. Infinite Paint Bucket Problem: The graph of y = - 1/ x from y
x = 0 to x = 1 is rotated about the y-axis to form an infinitely
deep paint bucket (Figure 9-101). Explain why a vertical cross X
section along the y-axis will have an infinite area and thus
why the surface area of the bucket itself is infinite . Then show
that the bucket could be completely filled with a finite volume
of paint, thus coating the infinite surface area . Surprising7? 1
25. The Gamma Function and Factorial Function : In this problem
you will explore
f (x) = Jo"
" txe - t dt,
where x is a constant with respect to the integration . Figure
9-lOm shows the integrand for x = 1, x = 2, and x = 3.
a. Find f(l), f(2), and f(3) by evaluating the improper Figure 9-101
integral. Along the way you will have to show, for instance,
that
lim b 3 e - b = 0. y
b -oo
b. From the pattern you see in the answers to 25a, make a
conjecture about what f(4), f(5), and f(6) are equal to.
c. Integrate by parts once and thus show that
f(x) =x f(x - 1) .
Use the answer to confirm your conjecture in part (b).
d. The result of the above work forms a basis for the
definition of the factorial function. Explain why this
definition is consistent with the definition Figure 9-1Om
x! = (x)(x - l)(x - 2) ... (2)(1).
Definition:TheFactorialFunction
(andthe GammaFunction}
The factorial function.
The gamma function. (f is the uppercase
[(x) = (x - 1)!
Greek letter gamma.)
e. Confirm that the integral for 3! approaches 6 by integrating numerically from t = 0
to t = b for some fairly large value of b. How large a value of b makes the integral
come within 0.000,001 of 6?
f. The improper integral can be used to define factorials for noninteger values of x .
Write an integral equal to 0.5!. Evaluate it numerically, using the value of b from
25e . How can you tell from the graphs in Figure 9-lOm that your answer will be
closer than 0.000,001 to the correct answer? How does your answer compare with
the value in the National Bureau of Standards Handbook of Mathematical Functions,
namely, 0.51 = 0.8862269255?
g. Quick! Without further integration, calculate 1.5!, 2.5!, and 3.5!. (Remember 25c.)
h. Show that O!= 1, as you probably learned in algebra.
4 90 Chapter
9: Algebraic
Calculus
Techniques
far theElementary
Functions
i. Show that (-1 )!, (- 2)!, (- 3)!, .. . , are infinite but (-0.5)!, (- 1.5)!, and (- 2.5)! are
finite .
j. Show that the value of 0.5! in 25f can be expressed rather simply in terms of rr.
26. Spaceship Work Problem: A 1000-lb spaceship is to be sent to a distant location . The
work required to get the spaceship away from the earth's gravity equals the force times
the distance the spaceship is moved . But the force F, which is 1000 lb at the earth's
surface, decreases with the square of the distance from the earth's center,
F = 1000
r2 '
where r is the number of earth-radii. Since there is always some force no matter how
far you travel from earth, additional work is always being done. Does the amount of
work increase without bound as r goes to infinity? Show how you arrive at your answer.
2 7. Piecewise Continuity Problem: Figure 9- lOn shows the graph
)I
of
y = 2x _ Ix - 21
x- 2
Suppose that you are to evaluate
3
fI
(2x _ lx - 21) dx.
X- 2
Although the integrand is discontinuous on the closed interval
(1, 3], there is only a step discontinuity at x = 2. The integrand
X
is continuous everywhere else in (1, 3]. Such a function is said
2 3
to be piecewise-continuous on the given int erval. In this
problem you will show that a piecewise-continuous function Figure 9-1On
is integrable on the given interval.
Definition:PiecewiseContinuity
Function f is piecewise-continuous on the interva l [a, b] if and only if there is a finite number of
values of x in [a, b] at which f(x) is discontinuous, the discontinuities are either removable or
step discontinuities, and f is continuous elsewhere on [a, b].
a. Write the integral above as the sum of two integrals, one from x = 1 to x = 2 and
the other from x = 2 to x = 3.
b. Both integrals in 27a are improper. Write each one using the correct limit terminol-
ogy.
c. Show that both integra ls in 27b converge. Observe that the expression Ix - 21/ (x - 2)
equals one constant to the left of x = 2, and a different constant to the right. Find
the value to which the original integral converges.
d. Explain why the following property is true.
Property:Integrabilityol Piecewise-Continuous
Functions
If function f is piecewise-continuous on the interval [a, b ], then f is integrable on [a, b ].
Sec
tion9-10: Improper
Integrals 491
e. True or false: "A function is integrable on th e interval [a, b] if and only if it is
continuous on [a, b]." Justif y your answer.
28. Journal Problem: Update your journal with things you've learned since the last entry.
You should include such things as thos e list ed her e.
The one most important thing you hav e learned since the last journal entr y
The big integration techniqu e that allows you to integrate a product of two functions
Oth er integration techniques that involve substitutions and algebraic transforma-
tions
Hyperboli c fun ctions
Improp er int egrals
Why th e fundamental theor em, not numeri cal integration, is neede d for improper
integrals
Any techniques or ideas about the calculus of transc endental functions that are still
uncl ear to you
9-11 Miscellaneous Integrals and Derivatives
By the tim e you finish this section you will have seen all of the classical algebraic
techniqu es for performin g calculus. These techniques were the only way calculus
could be don e before the advent of the computer made numerical methods easily
impl ementabl e.
OBJECTIVE Algebraically integrate or differentia te expressions containing the elementary functions.
Techniques
: Differentiation
Sum: (u + v )' = u' + v'
Product: (uv) ' = u ' v + uv ' Or, use logar ithmi c dif feren tiation technique .
Quotient: (u/v)' = (u'v - uv ')/v 2 Or, use logarithmic differentiation technique.
Composite: (f(u ))' = f'(u)u' (chain rule)
Implicit: f(y) = g(x) => f'(y)y' = g'(x) (Use the cha in rule, where y is th e insid e fun ction .)
Power function: (x")' = nx 11- i
Exponential function : ( nx)' = (n ") Inn (Use the logarithmic differentiation technique.)
Logarithmic function: (log bx) ' = (1/x)(l/ lnb )
Logarithmic diff erentiation technique :
y = f(x) => ln y = lnf (x) => (l/y)y' = [lnf (x)]' => y ' = y[Inf(x)]'
Trigonometric function: sin 'x = cosx , cos' x = - sinx
Inverse trigonometric fun ction: Differentiate implicitly .
Hyperbolic function: sinh 'x = coshx, cosh'x = sinhx
Inverse hyperbolic function: Differentiat e implicitly .
492 Chapter
9: Algeb
raicCalculus
Techniques
for the Elemen
tary Functions
Techniques:IndefiniteIntegration
Known derivative: f f '(x) dx = f(x ) + C
Sum:f (u+v)dx = fudx + fvdx
Product: f u dv = uv - f v du Int egrat e by parts .
Reciprocal Function: J u- du = ln lul + C
1
Power Function : f u" du = un+i/ (n + 1) + C, n * - 1 ("u-substitution" method).
Power of a Function: ff" (x) dx Use a reduction formula.
Square root of a quadratic: Integrat e by trigonometric substitution; complete the square
first, if necessary.
Rational algebraic function: Convert to a sum by long division and by resolving into
partial fractions.
Inverse function [exponential (logarithmic), trigonometric, or hyperbolic]: Integrate by
parts.
Problem Set 9-1 1
For Problems 1- 100, differentiate the given function, or evaluate the given integral.
l. y = sec 3x tan 3x 2. y = sinh 5x tanh 5x
3. f x cosh4x dx 4. f X COSX dx
5. f(x) = (3x + 5) - 1 6. f(x) = (5 - 2x) - 1
7. f (3x + 5) - 1 dx 8. J (5 - 2x) - 1 dx
9. t (x) = tan 5 4x 10. h(x) = sech 3 7x
11. J sin 2 x dx 12. f cos 2
xdx
6x - 11 5x + 9
13. y = 14. y = X - 4
x +2
15. f 6xx +- 211 dx 16. f 5xx-4+ 9 dx
17. f(t) = Jl+t2 18. g(t) = Jt2-=-I
19. f Jl+t2 dt 20. f Jt2-=-Idt
21. y = x 3 eX 22 . Y = x 4e - x
23. f x ex dx
3 24. f x e - x dx
4
25. f(x ) = sin - 1 x 26 . g(x) = tan - 1 x
27. f sin - xdx
1
28 . f tan - xdx 1
29 . J x 2 + 4x
1 d
- 5 x
30. J x2 - 1 d
6x - 7 x
31. J 1 d 32 . J 1 d
J x 2 + 4x - 5 x J x2 - 6x - 7 x
Sec
tion911: Misce
llaneous
Integrals
andDerivatives 493
~
33. f(x) = tanh x 34. f(x) = coth x
35. f tanh xdx 36. f coth x dx
37 . y = e2xcos 3x 38. y = e- 3xcos4x
39. f e 2xcos 3x dx 40. Je- "cos4xdx
3
41. g(x) = x 3 ln5x 42 . h(x ) = x 2 ln8x
43. f x 3 ln 5x dx 44. f x 2 ln8x dx
X X
45. 46 .
y= (x+2)( x + 3)(x+4) y= (x- l )(x-2)(x-3)
47. f X dx 48. f X dx
(x + 2)(x + 3)(x + 4) (x - l )(x - 2)(x - 3)
49. y = cos 3xs inx 50. y = sin 5 x cos x
51. f cos 3 x sinx dx 52. f sin 5 xcosxdx
53. f cos 3 x dx 54. Jsin x dx 5
55. f cos x dx
4
56. Jsin 6 x dx
57. g(x) = (x 4 + 3) 3 58. f(x) = (x 3 - 1)4
59. f(x 4
+3) 3 dx 60. J(x 3
- 1)"' dx
61. f (x"' + 3) 3 x 3 dx 62. f (x 3 - l )"'x2 dx
63 . f(x 4
+3) dx 64. f (x 3 - 1) dx
65. f (x) = r (t 4 + 3)3dt 66. h (x) = r (t 3 - 1) 4 dt
67. r xex dx 68. Jo2xe -x dx
69. r(x) = xex 70. s(x) = xe -x
ln x + 2 (lnx ) 3 + 4
71. q(x) = 72. r(x ) =
X X
73. f lnxx+ 2 dx 74. f (lnx~ 3 + 4 dx
3
75 . f(x ) = e2 76. f (x) = e"
2
77. f xe" dx 78. f x2exl dx
80. f x -"e".! dx
2
79. f x 3 eX dx
In Prob lems 81-100, a, b, c, d, and n stand for constants .
81. f e0 "cos bx dx 82. Jea"sinbxdx
83. f sin 2 cx dx 84. Jcos cx dx2
494 Chapter
9: Algebraic
Calculus
Techniques
for theEleme
ntary Functions
85. f(x) = ax + b 86 . f (x) = (ax + b )n
ex + d
87. Jax + b dx 88 . J(ax + b)"dx
ex + d
89 . f (x 2 + a 2) - l 12x dx 90 . f (a 2 - x 2)- li2xdx
91. f (x 2 + a 2)- 112 dx 92. f (a 2 - x2)- 112dx
93 . f(x) = x 2 sinax 94. f(x) = x 2 cos ax
95 . f x sinaxdx
2 96 . f x2cos axdx
97. f sinhax dx 98 . f coshaxdx
99. f cos - axdx
1 100. J sin - 1 ax dx
Historical
Topic
#I-RationalizingAlgebraic
Substitutions
Before calculators and computers wer e readily available to perform numerical integration,
it was important to be able to find algebraic formulas for as man y integrals as possible.
Much time was spent by users of math ematics searching for clever int egration techniques,
and by students of mathematics in learning thes e techniqu es. Two such techniques are
shown here and in Problem 107. An integral such as
I = f 1 +1 VXdx
X
that has a radical in the denominator can be transformed to a rational integrand by
substituting a variable either for the radical or for the entire denominator. Here's how
you would go about it . Let u = 1 + 1/x. Then (u - 1) 3 = x, from which dx = 3( u - 1) 2 du .
Substituting u for the denominator and 3(u - 1) 2 du for dx gives
1=3 f (u -u 1)
2
f
du = 3 (u - 2 + ,:;)du = 2 u 2 - 6u + 3ln lu l+ C
1 3
= ~ ( 1 + 1/x)2 - 6 ( 1 + 1/x) + 3 ln I 1 + vx I + c.
For Problems 101-106, evaluate the integrals by algebraic substitution.
1 1
101. J r;; dx 102 . J r;; dx
l +v x 1 -v x
103. 104. 1 d
J Jx+vx x
105. 106. 1 d
J .jex-=-I x
Historical
Topic
#2-RationalFunctions
of sinx andcosx by u = tan(x/2)
107. From trigonometr y, you recall the double-argum ent properti es for cosin e and sin e,
cos2t = 2cos 2 t-1, and sin2t = 2sintcost.
a. Explain how thes e properti es justify th e following equations .
cos x = 2 cos 2 i - 1, and sin x = 2 sin 1cos 1
Sectio
n 9-11: Miscel
laneousIntegrals and Deriva
tives 495
b. Show that the equations in 107a can be transformed to
1 - tan 2 '.! . 2 tan i
cosx = l ~ , and smx = ? x
+ tan 2
2 1 + tan- 2
c. Let u = tan (x/2). Show that x = 2 tan - 1 u, and thus that the following propert ies
are true.
Property:u = tan (x/2) Substitution
1 - u2 . 2u 2
cosx = + u2 , SlUX = -- and dx = -- du
1 1 + u2 ' 1 + u2
d. Let u = tan (x/2). Use the resu lt s of 107c to show that this integral reduces to
I du:
1 d
J 1 + COSX X
e. Do the integration in 107d and then do the reverse substitutio n to show that the
integral equals tan (x/2) + C.
108. A noth er Ind efin ite Int egra l of Secant:
a. Transform the integral
f secxdx,
using th e substitut ion u = tan (x/ 2) from Problem 107 to get
J 1 ! u2 du.
b. Perform the int egration in 108a. Show that the result is
ln tan ! I + C.
111 +- tan
2
c. Recall from trigonometry the following .
. tan (A
1.
tanA + tanB
+ B) = ------
1 - tan A tanB
ii . tanf = 1
Use this information to show that
Jsecxdx = ln ltan(:f + i )I + C.
d. Evaluate f~sec x dx two ways: (i) using the result of 108c and (ii) using the more
familiar int egra l form ul a. Show that th e answers are equivalent.
For Problems 109-111, use the substitution u = tan (x/2) to evalu ate the integral.
1 1
109. J dx 110. J . dx
1 - cosx 1 + smx
111.
I ---1 -cosx
cosx
d X
496 Chapter
9: Algebraic
Calculus
Techniques
for theElementary
Functions
9-12 Integrals in Journal
In this chapter you have learned algebraic techniques by which the elementary tran-
scendental functions can be integrated and differentiated. The integration tech-
niques include:
Recognition of the integrand as the derivative of a familiar function
Integral of the power function, fun du , n * - 1
Integral of the reciprocal function, f u- 1 du
Integration by parts
Reduction formulas
Trigonometric substitution
Partial fractions
Other substitutions
These techniques let you find the equation of a function whose derivative is given.
They also allow you to use the fundamental theorem to find the exact value of a
definite integral.
Although differentiating a function is relatively easy, the reverse process, int egrating ,
can be like unscrambling eggs 1 Integrals that look almost the same, such as
y f (x 2
+ 1) 10 xdx and f (x 2
+ 1) 10 dx,
2
may require completely different techniques. An integrand, such as e -x in
X
f e - x2 dx,
2
may be an elementary function but not the derivative of any other elementary
Figure 9-12a function . [This integral gives the area under the bell curve in statistics (Figure
9-12a).]
In this section you will record a short table of integrals in your journal . Constructin g
the table will bring together the various techniques of integration. The en d product
will give you a reference guide that you can us e to recall vario u s integrals and how
you derived them. As a result, you will better be able to use publications such as
CRC Tables, and to understand the output from symbol-manipulating comp ut ers.
Problem Set 9-1 2
1. Table of Integrals Problem: Record a short table of integrals in your journal. The table
should be arranged by the nature of th e integrand, rather than by the technique us ed .
The table shou ld include integrals of the algebraic functions and each one of the
elementary transcendental functions :
Power and reciprocal
Exponential and Logarithmic
Circular and reverse cir cular
Hyperbolic and inverse hyperbolic
You should include examples of other frequently occurring forms such as rat ion al
function, square root of a quadratic, and power of a trigonometric function (especia lly
f sin 2x dx, f sec 3 x dx, and so on). For each entry, you should state or show how the
formula is derived .
Section9-12: Integral
s inJournal 497
9-13 Chapter Review and Test
In this chapter you have learned to do algebraically the calculus of elementary
transcendental functions -exp onential and logarithmic, circular (trigonometric),
hyperbolic, and their inverses. The Review Problems are numbered according to the
sections of this chapter. The Concepts Problems allow you to apply your knowledge
to new situations. The Chapter Test is typical of a classroom test.
Review Problems
RO. Update your journal with things you've learned since th e last entry. You should
include such things as those listed here .
The one most important thing you have learned in studying Chapter 9
Which boxes you have been working on in the "define, understand, do, apply" table.
Any techniques or ideas about calculus that are still unclear to you
Rl. Let f(x) = x cos x . Find f'(x), observing the derivative of a product property . From
the results, figure out an equation for the indefirute integral Jx sinx dx. Check your
work by using the equation to evaluate the definite integral
f xsinxdx
and comparing it with the approximate answer you get by numerical integration.
R2. Integrate: J Sx sin 2x dx
R3. a. Integrate: Jx 3 cos 2x dx
b. Integrate : Je4 x sin 3x dx
C. Integrate : Jx(lnx) 2 dx
d. The region under the graph of y = x In x from x = 1 to x = 2 is rotated about the
y-axis to form a solid. Find the volume of the solid.
R4. a. Integrate by parts once to express Jcos 30 x dx in terms of an integral of a reduced
power of cos x .
b. Use the appropriate reduction formula to evaluate Jsec 6 x dx.
c. Derive the reduction formula for Jtan nx dx.
RS. a. Integrate without reduction formula : Jcos 5 x dx
b. Integrate without reduction formula: Jsec 6 x dx
C. Integrate without reduction formula : Jsin 2 7x dx
d. Integrate without reduction formula: Jsec 3 x dx
e. Integrate without reduction formula: Jtan 9 32 dx
f. Find exactly the area of the region inside th e limac;on with polar equation
r = 9 + 8 sin 0 from 0 = 0 to 0 = TT / 4.
R6. a. Integrate : J -Jx2 - 49 dx
b . Integrate: J -Jx2 - lOx + 34 dx
c. Integrate: J -Jl - 0.25x 2 dx
498 Chapter
9: Algebraic
Calculus
Techn
iquesfor theElementary
Functions
d. Find exactly, using the fundamental theorem, th e area of y
the zone of a circle of radius 5 between th e lines 3 unit s
and 4 units from the center (Figure 9-13a).
R7. Integrate:
(6x + l)dx X
a. J x2 - 3x - 4
5x2 - 2lx - 2
b f (x - l)(x + 2)(x -3)
dx
c. f 5x + 3x + 45 dx
2
x 3 + 9x
d. J 5x2 + 27x + 32 dx Figure 9-13a
x(x + 4) 2
e. Differential Equation Problem: Figure 9-l 3b I I I I I I I I YI I I I I I I I
shows the slope field for the differ ential //////// ////////
equation
dy
dx = 0.l (y-3)(y - 8).
'''''''' ''''''''
Solve this differential equation subject to ''''' '' , s ''' '' '''
the initial condition that y = 7 when x = 0.
On a photocopy of the slope field, plot the //////// ////////
graph of your solution, thus showing that I I I I I I I I I I I I I I I I
X
it is reasonable.
5
R8. a. Sketch the graph : y = cos - 1 x
Figure 9-l 3b
b. Differentiate: f(x) = sec 1 3x
c. Integrate: Jtan - 1 5x dx
d. Find the area of the region in Quadrant I bounded by the graph of y = cos - 1 x.
R9. a. Sketch the graph: f (x ) = sinb x
b . Sketch the graph: g (x ) = cosh - 1 x
c. Differentiate: h(x) = x 2 sechx
d. Differentiate: f(x) = sinh - 1 5x
e. Integrate: J tanh 3x dx
f. Integrate: J cosh - 1 ?x dx
g. Using the definitions of coshx and sinhx, prov e that cosh 2 x - sinh 2 x = l.
h. Find a particular equation of the catenary with vertex at (0, 5) and point (3, 7). Use
the equation to predict the value of y if x = 10. Find the values of x if y = 20.
0
RlO. a. Evaluate: ( (x - 2 )- L2 dx
O
b. Evaluate: J / 2 tanx
TT
dx
C. Evaluate : f 1
-1
x- 213 dx
d. Evaluate: r( ft - I:=~I) dx
e. For what values of p does th e p-integral ft' x - Pdx converge?
Section 9-13: Chapter
Review
andTest 499
IA M 1Hidihlil : &&41 5WMiiri4iWii#&f t Eij
Rll. a. Differentiate : f (x) = x sin - 1 x
b . Int egrate: f x sin - 1 x dx
C. Differ entiate : tanh(ex)
d. Integrate: J(x 3 - x) - 1 dx
e. Differentiate: f (x) = (1 - x 2 ) 1 i 2
f. Integrate: J(l - x 2 ) 1 i 2 dx
g. Differentiate: g(x) = (lnx )2
h. Integrate: f x lnx dx
Rl2 . Explain why f (9 - x 2 ) - 112 dx has an inverse sine in the answer but f (9 - x2) - 112 xdx
does not.
Concepts Problems
Cl. Integral of sechx Problem: Derive th e formula
f sechx dx = sin - 1
(tanh x) + C.
The integrand can be transformed to a square root involving tanh x by use of the
Pythagorean propert y relating sechx and tanh x . Then a very clever trigonometric
substitution can be used to rationalize the radical. Confirm that th e formula works by
evaluating the integral on the int erval [O, 1], then checking by num erical integration .
C2. Int egra l of cschx Problem: Derive the formula
f cschx dx = ln I tanh 11+ C.
The integrand can be transformed to functions of tanh (x/2) by first observing that
sinh 2A = 2 sinhA coshA,
from which
1
cschx = 2 Sl.nh x COS h x .
2 2
Clever algebra, followed by an application of the Pythagorean properties, produces
the desir ed result. Then substitute u for tanh (x/ 2) . You will have to be clever again
to figure out what to substitute for dx in terms of du. The result ing integral is
remarkably simple! Confirm that the formula works by evaluatin g the integral on the
interval [l, 2], then checking by numerical integration .
C3. Another Integral of csc x: Derive the formula
Jcsc x dx = ln Itan i I + C.
Confirm that the formula works by evaluating the integral on the int erval [0.5, 1],
then checking by numerical integration.
500 Chapter
9: Algebraic
Calculus
Techniques
for the Elementary
Functions
C4. Another Definition of TT Problem : Figure 9-13c shows the
region under the graph of y = (x 2 + 1) - 1 , extending to
infinity in both directions. Show that the area of this
infinitely long region is exactly equal to TT. This fact is
remarkable because the integrand has nothing to do with
circles, yet the answer is the most fundamental number
concerned with circles! Figure 9-13c
CS. Upper Bound Prob lem: Figure 9-l 3d shows the graphs of f(x) = ln x and g (x) = tan - 1 x.
As x gets large, both graphs are increasing, but concave down. The inverse tangent
graph approaches TT / 2. Prove that the graph of f(x) = lnx is unbounded above. You
can do this by assuming that it is bounded above by some number M, then finding a
contradiction by finding a value of x in terms of M for which lnx > M.
f(x) = lnx g(x ) = ar c tan x
X X
Figure 9-13d
Chapter Test
[Author's Note: This test is longer than a normal 60-minute classroom test.]
For Problems Tl-T6, evaluate the indefinite integral.
Tl . fsin 5 xcosxdx T2. Jx 3 sinh6xdx
T3. Jcos - 1 x dx T4 . f sec 3x dx
TS. f e 2xcos 5x dx T6. f ln3x dx
For Problems T7 and TB, differentiate .
T7 . f(x) = sech 3 (e 5x ) TB. g(x) = sin - 1 x
T9 . Given f(x) = tanh - 1 x, find a formula for f'(x) in terms of x by appropriate implicit
differentiation. Demonstrate that the formula is correct by approximating f '(0.6)
using numerical differentiation.
TlO. Find the particular equation of the form y = kcosh(l / k)x + C for the catenary
containing vertex (0, 1) and point (5, 3).
X - 3
Tl 1. Integrate J x 2 _ x + dx three ways:
6 5
a. By trigonometric substitution, after completing the square .
b . By partial fractions.
c. As the integral of the reciprocal function.
d. Show that all three answers are equivalent .
Section9-13:Chapter
Review
andTest 501
Tl 2. Evaluate f cos 2 x dx by appropriate use of the double-argument properties.
T13. Evaluate f cos 5 x dx
two ways:
a. By transforming four of the cosines to sines, and integrating as powers of sine.
b. By using the reduction formula,
Jcos"xdx 1
= ncos 11
-
1 n- lJ cosn-
xsinx + -n - 2
xdx .
c. Show that the two answers are equivalent.
0
Tl 4. Evaluat e the improper int egral : Joxe - 0 1" dx
502 Chapter
9: Algebraic
Calculus
Technique
s for theElementary
Functions
CHAPTER
10 The Calculus of Motion-
Averages, Extremes, and Vectors
The distance a spaceship has gone equals velocity multiplied by time.
But the velocity varies. Displacement is the integral of velocity, and
velocity is the integral of acceleration. By measuring acceleration of
the spaceship at frequent time intervals, the displacement can be
calculated by numerical calculus methods.
503
Mathematical Overview
Chapter 10 extends your study of objects in motion. You will
distinguish between such things as
distance vs. displacement
acceleration vs. velocity
maximum vs. minimum
linear vs. plane motion
You will do this in four ways.
Graphically The logo at the top of each even- y
numbered page of this chapter shows
the position vector, velocity vector,
and acceleration vector for an object
moving in a curved, plane path .
X
Numerically time accel. vel.
0 1.3 20.0
2 1.7 23.0
4 2.2 26.9
6 2.1 31.2
8 1.5 34.8
Algebraically Distance= tI
a
v(t) Idt. Displacement= t
a
v(t) dt.
Verbally Now I know the precise definition of the average value of a function.
It is the integral of the function between two limits divided by the
difference between those limits.
504
10-1 Introduction to Distance and Displacement for
Motion along a Line
You have learned that velocity is the rate of change of position with respect to
time. In this chapter you will concentrate on the distinction between distance (how
far) and displacement (how far, and in what direction). You will sharpen your
understanding of the difference between speed (how fast) and velocity (how fast
and which direction). Once you have made these distinctions for motion in one
dimension (along a line), you will use vectors to analyze motion in two dimensions
(in a plane). Along the way you will find maximum, minimum, and average values of
velocity and position functions.
OBJECTIVE Given an equation for the velocity of a moving object, find the distance traveled and the
displacement from the starting point for a specified time interval.
Suppose that you drive 100 mi and then return 70 of those miles (Figure 10-1a).
Although you have gone a distance of 170 mi, your displacement, which is measured
from the starting point, is only 30 mi. Problem Set 10-1 will clarify the distinction
between these two quantities. You may work on your own or with your study group.
Start Turn
Go 100 mi. Displa cement = + 100
'
Return 70 miles .
Displa cement = - 70
' Net : Finish
displa cement
=30 mi
Figure l 0-1 a
Exploratory Problem Set 1 0-1
Calvin's Swimming Problem: Calvin enters an endurance swimming contest. The objective
is to swim upstream in the river for a period of 10 min. The river flows at 30 ft/min.
Calvin jumps in and starts swimming upstream at 100 ft/min. Phoebe ascertains that as
he tires his speed through the water decreases exponentially with time according to the
equation Ve = 100(0.8) 1 . Thus Calvin's net velocity (Figure 10-lb) is only
V = 100(0 .8) 1 - 30.
l. At what time will Calvin's velocity become negative? V
2. How far will Calvin go upstream-that is, while his velocity is
positive? How far will he go back downstream (while his velocity is
negative) until t = 10? What is the total distance he will have gone 10
in the 10 min? l
3. What will Calvin's displacement from the starting point be at the
end of the 10 min? Will he be upstream or downstream of his Figure l 0-1 b
starting point?
Section10-1:Introduction
to Distance
andDisplacement
forMotion
alonga Line 505
4. Write a definite integral that can be used to find the displacement after 10 min in one
computation. Check it by doing the integration and comparing with Problem 3.
5. Write one definite integral that represents the total distance Calvin goes in the 10 min.
A clever application of absolute value will help. Integrate numerically.
10-2 Distance, Displacement, and Acceleration for
Linear Motion
Most real objects such as cars and birds travel in two or three dimensions . In
Section 10-7, you will see how to use vectors to analyze such motion . For the time
being, consider only objects moving in a straight line. In Section 10-1 , you saw the
distinction between the distance a moving object travels and its displacement from
the starting point. If the velocity is positive, the displacement is positive. If the
velocity is negative, then the displac ement is negative. In the latter case the distance
traveled is the opposite of the displacement. The two ideas can be combined with
the aid of the absolute value function, as follows .
Property:Distanceand Displacement
Displacement = J:(velocity) dt
Distance = J:lvelocityl dt
OBJECTIVE Given velocity or acceleration as a function of time for an object in linear motion, find the
displacement at a given time and the distance traveled in a given time interval.
Example 1 A moving object has velocity v(t) = t 2 - 7t + 10 ft/sec in the time interval [l, 4].
a. Find the time subintervals in which the velocity is positive and those in which it
is negative .
b . Find the distance the object travels in each of these subintervals.
c. Use a single integral to find the displacement in the time interval [l, 4] and use
a single integral to find the distance traveled in this interval.
d . Show how the answers to part c can be found from the answers to parts a and b .
Solutions a. The graph of v versus t (Figure 10-2a) shows that the velocity changes from
positive to negative at t = 2. This fact can be confirmed algebraically by
solving
t 2 - 7t + 10 = 0 ~ t = 2 or t = 5.
506 Chapter
10:TheCalcu
lusofMotion-Averages,
Extremes,
andVectors
, v(t ) Positi ve velocity: [ 1, 2)
10 Pos itive displa ce ment Negati ve velocity: (2,4 ]
\
\
\
I
I
I
b. For [1, 2],
I
2 2 5 Num erica lly, or b y
Displacement = J (t1
- 7t + 10) dt = 1.8333 . .. = 1 6 .
fundam ent al th eor em .
.-. distan ce = 1 i6 ft
Figure l 0-2a
For [2, 4],
4
Displa cement = J
2
(t 2 - 7t + 10) dt = -3 .33 33 . . . = -3 3I .
.-. distan ce = 3 ft
C. For[l,4],
4
Displa cement = J
1
(t 2 - 7t + 10) dt = - 1.5
Distanc e= r lt 2 - 7t + 101dt
Use th e ab so lut e va lue fun ct ion o n
your grap h er.
Use num erica l in tegrat ion (see
= 5.1666 .. . = 5ft no te.)
d. For displa cement, 1 + ( - 3 ) = - 1.5 (checks).
\ Absolut e value of v(t)
For distanc e, 1 + 3 = 5 (checks).
10 Not e: If you want to evaluat e th e int egral in part c algebraicall y u sing th e fundamental
\
\ Both are p os itive. ,' theor em, divid e th e inter val [1, 4] as in part b. The int egrand has a cusp at t = 2
I
I (Figure 10-2b) at which the absolut e-value fun ction chang es from + (t 2 - 7t + 10) to
I
I ',,,,' t - (t 2 - 7t + 10). (Recall from algebr a that th e absolut e valu e of a negative number
2 4 is th e opposit e of that numb er.) The final ans wer is still th e sum,
1+ (+3 l = 5 ft .
Figure l 0-2b
Figure 10-2b also shows th at th e int egrand for distance is always positiv e (or zero),
m eanin g that th e distanc e trav eled is po sitive. This is tru e even though the object
in Exampl e 1 is displa ced a negative amount from its startin g point.
Exampl e 2 A car accelerat es for 24 sec. Its acceleration in (mi/ hr)/ sec is m easur ed each 3 sec
and is list ed in th e tab le.
time acceleration
(sec) [(mi/ hr )/ sec]
0 1.3
3 1.7
6 2.2
9 2.1
12 1.5
15 0.3
18 - 0.4
21 - 1.1
24 - 1.4
a. Plot th e graph of acceler ation ver sus tim e.
Sect
ion10-2: Distance,
Displacement
, andAcce
lerationfor Linear
Motion 50 7
b . At time t = 0, the car was going 20 mi/ hr. Predict its velocity at each 3 sec instant
from O through 24.
c. Plot the graph of velocity versus time.
d. Approximate ly how far did the car go in this 24-sec interval?
Solutions a. The graph is shown in Figure 10-2c.
b. Since acceleration is the derivative of velocity with respect to time, the
2
a
,,
,,- ...
velocity is the integral of acceleration .
Ja dt
,
," '
"
\
\
\
v =
You can estimate the average acceleration for each 3-sec interval. For
\
\ interval 1,
3 6 9 12 15 '{ 8 21 24
' t
Average accel eration = 0.3 + 1.7) = 1.50.
-1
'' The change in velocity is thus (1.50 )( 3) = 4.50. Since the initial velocity was
\..
-.. given to be 20, th e velocity at the end of the interval is about 20 + 4.50 =
24.50.
Figure 10-2c
The process is equivalent to integrating by the trapezoidal rule . The calcula-
tions can be done by extending the given table using a computer spreadsheet
or at least a spreadsheet format.
average
tim e acceleration acceleration 6v velocity
(sec) [(mi/ hr)/ secl [(mi/ hr)/ sec] (mi / hr ) (mi/hr)
. ' ---
V
' ... 0 1.3 20 (Given)
40 ,, '
3 1.7 1.50 4.50 24.5
,,
30 , ," 6 2.2 1.95 5.85 30.35
, ,; 9 2.1 2.15 6.45 36.8
20 12 1.5 1.80 5.40 42.2
10
15 0.3 0.90 2.70 44 .9
18 - 0.4 - 0 .05 - 0.1 5 44 .75
t
21 - 1.1 - 0.75 - 2.2 5 42.5
3 6 9 12 15 18 21 24
24 - 1.4 -1.25 -3.7 5 38.75
Figure 10-2d
C. The graph is shown in Figure 10-2d.
d. Finding the displacement at the end of each 3-sec interval can be done the
same way you found the velocity. Find an average velocity for the interval,
multiply by M, and add the result to the displacement at the beginning of
the interval. The displacement at t = 0 is zero because you are finding the
displacement from the starting point. The only catch is that you must use
3/ 3600 hr for t.t since vis in miles per hour. The calculations can be done
by extending the tab le from part b, putting appropriate instruct ions into the
spreadsheet.
508 Chapter
10:TheCalculus
ofMotion-Averages,
Extremes,
andVectors
average average
time acceleration acceleration t.v velocity velocity displacement
(sec) [(mi/ hr)/ sec] [(mi/hr)/ sec] (mi/hr) (mi/ hr) (mi/ hr) (mi)
0 1.3 20 0.000
3 1.7 1.50 4.50 24.5 22.25 0.018 ...
6 2.2 1.95 5.85 30.35 27.425 0.041 . ..
9 2.1 2.15 6.45 36.8 33.5 75 0.069 . ..
12 1.5 1.80 5.40 42 .2 39.5 0.102 . ..
15 0.3 0.90 2.70 44.9 43 .55 0.138 ...
18 - 0.4 - 0.05 -0.1 5 44.7 5 44.825 0.175 ...
21 - 1.1 - 0.75 - 2.25 42 .5 43.625 0.212 ...
24 -1.4 - 1.25 -3 .75 38.75 40.625 0.246 .. .
The last column shows the total displacement from t = 0 to the end of the
time interval . For the 24 sec the total displacement was about 0.246 mi, or
about 1300 ft.
Problem Set 1 0-2
Quickly
DoThese
The following problems are intended to refresh your skills . You shou ld be able to do all
ten problems in less than five minutes .
QI. You traveled 30 mi/ hr for 4 hr . How far did you go?
Q2. You traveled 75 mi in 3 hr. How fast did you go?
Q3. You traveled 50 mi at 40 mi/ hr. How long did it take ?
Q4. Differentiate : f(x) = lnx
QS. Integrate: JIn x dx
Q6. Differentiate: f(t) = tan t
Ql. Differentiate: g(t) = tanh t
QB.Integrate : f x 2 dx
Q9. Integrate : J 2x dx
QJO.Differentiate : h(x) = 2"
For Problems 1-4, an object moving in a straight line has velocity v(t) in the given time
interval.
a. Find the time subintervals in which the velocity is positive, and those in which it is
negative.
b . Find the distance the object travels in each of these subintervals.
c. Use a single integral to find the displacement in the given time interval and use a single
integral to find the distance traveled in this interval.
d. Show how the answers to part c can be found from the answers to parts a and b.
e. Find the acceleration of the object at the midpoint of the time interval.
Sec
tion102: Distance,Displacemen
t, andAccelerat
ionfor Linear
Motion 509
f~. ,
!~ 1
1. v(t) = t 2 - lOt + 16 ft/ sec from t = 0 tot = 6 sec
2. v(t) = tan0.2t cm/sec from t = 10 sec tot = 20 sec
3. v(t) = sec fi t - 2 km/ hr, from t = 1 sec tot = 11 hr
4. v(t) = t 3 - 5t 2 + 8t - 6 mi/ min, from t = 0 to t = 5 min
For Problems 5-8, first find an equation for the velocity of a moving object from the
equation for acceleration. Recall that acceleration is the derivative of velocity. Then find
the disp lacement and distance traveled by th e moving object in the given time interval.
5. a(t) = t 112 (ft/ sec)/ sec, v(O ) = - 18 ft/ sec, from t = 0 tot = 16 sec
6. a(t) = t- 1 (cm/sec )/ sec, v(l) = 0 cm/ sec, from t = 0.4 sec tot = 1.6 sec
7. a(t) = 6sint (km/ hr)/ hr, v(O) = -9 km / h, from t = 0 tot = rr hr
8. a (t ) = sinht (mi/ hr)/ hr, v (O) = -2 mi/ hr, from t = 0 tot= 5 hr
9. Meg's Velocity Prob lem: Meg floorboards her car, giving it a velocity of v = t 112 - 2
ft/ sec at time t seconds after she started accelerating.
a. Find th e time(s) at which v = 0.
b. Find her net disp lacement for the time interval (1, 9].
c. Find the total distance she travels for the time interval (1, 9].
10. Periodic Motion Problem: The velocity of a moving object is given by v = sin 2t cm/sec.
a. Find the distance the object trav els from the tin1e it starts (t = 0) to the first time
it stops.
b. Find the displacement of the object from its starting point and the total distance
it travels from t = 0 tot= 4. 5rr. Be clever!
11. Car on the Hill Prob lem: Faye Ling's car runs out of gas as she is drivin g up a long
hill . She lets the car roll without putting on th e brakes. As it slows down, stops, and
starts rolling backwards, its velocity up th e hill is given by
V = 60 - Zt,
where v is in feet per second and t is th e numb er of seconds since the car ran out of
gas .
a. What is the car's net displacement betwe en t = 10 and t = 40?
b. What is the total distance the car rolls b etween t = 10 and t = 40?
12. Rocket Prob lem : If a rock et is fired straight up from earth, it experiences acceleration
from two sources :
Upward, au in (meters per second) per second, du e to the rocket engine
Downward, a ctin (meters per second) per second, due to gravity
The net acceleration is a = au + ac1 . However, th e up war d acceleration has a dis-
continuity at the time the rock et engine stops. Suppos e that the accelerations are
given by
au = {40 cos 0.015t, 0 ~ t ~ 100
0, t > 100
ac1= -9 .8 for all t .
510 Chapter
l 0: TheCalculus
ofMotion
- Averages,
Extremes,
andVectors
a. Plot graphs of a and v versus t for th e first 300 sec.
b. At what value of t does a become negative? At what value of t does v become
negative?
c. Find the displacement of the rocket at time t = 300 and the distance the rocket
traveled between t = 0 and t = 300. What does the relationship between these two
numbers tell you about what is happening in the real world at t = 100?
d. How fast, and in what dir ection is the rocket traveling at t = 300?
13. Subway Problem: A train accelerates as it leaves one subway time, t acce leration, a
station, then decelerates as it comes into the next one. Three (sec) [(mi/hr)/sec]
calculus students take an acceleromete r aboard the train. They 0 1.2
measure the accelerations, a, given in the table in (miles per hour) 5 4.7
per second, at the given values of t seconds. 10 2.9
a. Calculate the velocity and displacement at the end of each time 15 0.6
interval. Assume that th e velocity was zero at time zero. You 20 0
may use a computer spreadsheet. 25 0
b. Show that the train has stopped at t = 60. 30 0
35 0
C. How can the velocity be zero at t = 0 but the acceleration be
40 - 0.4
positiv e at that time?
45 - 1.4
d. How can acceleration be zero from t = 20 to t = 35, while the 50 - 3.8
velocity is not zero? 55 -3 .2
e. How far is it between the two stations? 60 0
14. Spaceship Problem: A spaceship is to be sent into orbit around the earth. You must
find out whether the proposed design of the last stage booster rocket will get the
spaceship going fast enough and far enough so that it can orbit. Based on the way the
fuel burns, the acceleration of the spaceship is predicted to be as in the table on the
next page, where time is in seconds and acceleration is in (miles per hour) per second.
a. Initiall y the spaceship is 400 mi from the launch pad, going 6000 mi/ hr. Calculate
the velocity and acceleration at the end of each time interval. You may use a
spreadsheet.
Sectionl 0-2: Distance,Displacement,
andAcce
lerationforLinear
Motion 511
4 rr w
@,W'ifiift@1111,e,a1111.;wG:M:J
,Qs
time acceleration
(sec) [(mi/ hr)/ sec]
0 3
10 14
20 30
30 36
40 43
so 42
60 64
70 78
80 89
90 6
100 0 (Rocket burns out)
b. Consulting the specifications, you find that when the last stage finishes firing (100
sec, in this case) the spaceship must
Be at least 1000 mi away from the launch pad
Be moving at least 17,500 mi/hr
Based on your work, conclude whether each of these specifications will be
Definitely met
Definitely not met
Too close to say without more information
15. Physics Formula Problem: Elementary physics courses usually deal only with motion
under a constant acceleration, such as motion under the influence of gravity. Under
this condition, certain formulas relate acceleration, velocity, and displacement. These
formulas are easily derived by calculus. Let a be the acceleration (a constant). Let v 0
be the initial velocity (when time t = 0). Let s 0 be the initial displacement (again, when
time t = 0). Derive the following formulas.
a. v = v 0 + at
b. s = v 0 t + at 2 + so
16. Elevator Project: When a normal elevator starts going up, you feel a jerk until it gets
up to speed . This is because the acceleration changes almost instantly from zero to
some positive value . In this problem you will explor e another way for an elevator's
acceleration to be arranged such that the jerk is minimized.
a. Suppose an elevator starts from rest (v = 0) at the bottom floor (s = 0 ft) and
is given a constant acceleration of 2 ft/ sec 2 for 6 sec. Thereafter , the elevator
rises with constant velocity. Find the velocity and displacement of the elevator as
functions of time. Sketch three graphs: acceleration, velocity, displacement, for
times t = 0 to t = 10 sec .
b. How does the acceleration graph show that passengers on the elevator get a jerk at
t = 0 and another jerk at t = 6?
c. If the acceleration increases gradually to a maximum value, then decreases gradua lly
to zero, the jerks will be eliminated. Show that if the acceleration is given by
a= 2 - 2cos(rr / 3)t
for the first 6 sec, then the elevator's acceleration has this property.
d. Using the acceleration function in 16c, find the velocity as a function of time.
512 Chapter
10:TheCalculus
ofMotion-Averages,
Extremes,
andVectors
e. Sketch the graph of velocity as a function of time, if the velocity remains at its value
at t = 6 seconds as the elevator goes up. How does this velocity graph differ from
that in 16a?
f. How far does the elevator go while it is getting up to top speed 7
g. The elevator is to be slowed down the same way it was speeded up, over a 6 second
time period . If the elevator is to go all the way to the 50th story, 600 ft above the
bottom floor, where should it start slowing down?
h. How long does it take for the elevator to make the complete trip?
i. If the elevator were to go up just one floor (12 ft), would the new acceleration and
deceleration functions in 16c and f still provide a smooth ride7 If so, how do you
tell? If not, what functions could be used to smooth out the ride?
10-3 Average Value Problems in Motion and
Elsewhere
v(t)
40
Suppose that the velocity of a moving object is given by
v(t ) = 12t - t 2 ,
20 where t is in seconds and v(t) is in feet per second . What would be meant by
I
I
I the average velocity in a time interval such as from t = 2 to t = 11? The equation
I
I
,,,, "distance= (rate)(time)" is the basis for the answer. Dividing both sides by time gives
I
I
I ' \
, \(
rate = (distance)/(time) . This concept is extended to velocity by using displacement
--+ _._ .........
_,_,.___._
.......__.......__
.................
..
2 11 \ instead of distance.
. displacement
Figure l 0-3a average velocity = -~- . - - -
tune
Figure 10-3a shows the graph of v(t) . The total displacement in feet, s, is the integral
of the velocity from 2 to 11.
(11 2
S=h (l2t-t )dt
= 6t2 - ! t3 J ll
3 2
v(t) Area of rectangl e = 261
40
= area und er graph
The time to travel the 261 ft is (11 - 2), or 9 sec. So the average velocity is
261
Vav = g = 29.
20
Plotting the average velocity on the v(t) graph (Figure 10-3b) reveals several
conclusions.
I \
I The area of the rectangle with altitude equal to Vav and base 9 is also equal to the
-~~~~_,_~-
-.-
\
I \
I ,t total displacement . So the area of the rectangle equals the area under the v (t)
2 11 \
graph.
Figure l 0-3b Because the area of the rectangle equals the area of the region, there is just as
much of the region above the Vav line as there is "empty" space in the rectangle
Sectio
n l 0-3:Average
ValueProblemsinMotion
andElsewhe
re 513
v(t) below the line. If you could "pour" the region above the line into the spaces below
40 Area here
the line, it would just fit (Figure 10-3c).
The average velocity Va v is not equal to the average of the initial and final ve-
locities. From the equation, v(2) = 20 and v(ll) = 11. The average, 29 (which
equals displacement/time), is higher than either one and is thus not equal to their
20 ,,
/: average.
I
I
I '
'
,,,, Another object, starting at the same time and place, but moving with a constant
I
I
' \
velocity equal to the average velocity, would finish at the same time and place as
I ' ,t the first object.
2 11 \
The facts that average velocity equa ls (disp lacement) / (time), and that displacement
Figure l 0-3c
is found by integrating the velocity lead to the following general definition of
average velocity .
Definition:Average Velocity
If v(t) is the velocity of a moving object as a function of time then the average velocity
from time t = a to t = b is
Va v =
J:v(t) dt
b_ a ,
. total displacement
That 1s,average velooty = .
tota1 nme
The reasoning used to define average velocity can be extended to define average
value for any function . The following is the definition of average value of a function.
Definition:Average Valueof a Function
If function f is integrable on [a, b], then the average value of y = f(x ) on the interval x = a
to x =bis
Yav =
J:b _ a
f(x) dx
OBJECTIVE Calculate the average value for a function given its equation.
Problem Set 1 0-3
DoTheseQuickly
The foll0vving problems are intended to refresh your skills. You shou ld be able to do all
ten problems in less than five minutes .
Q1. What is your average speed if you go 40 mi in 0.8 hr?
Q2. How far do you go in 3 min if your average speed is 600 mi/ hr?
Q3. How long does it take to go 10 mi at an average speed of 30 mi/hr?
514 Chapter
10: TheCalculus
ofMotion
- Averages,
Extremes,andVectors
Q4. The first positive value of x at which y = cos x has a local maximum is -?- .
QS. y = ex has a local maximum at what value of x?
Q6. y = x 2 - 3x + 11 has a local minimum at x = -?- .
Ql. For f(x) = (x - 5) 2 , the global maximum for x E [l, 3] is -?-.
QB.If f(x) = x 08 ,
then f '(O) = -?- .
Q9. Name the theorem that states that under suitab le conditions, a function's graph has a
tangent line parallel to a given secant line .
Q 10. The graph of ( 1) + ( f
2
)2= 1 is a(n) - 7 -.
For Problems 1- 6,
a. Find the average value of the function on the given interval.
b . Sketch a graph showing the geometrical interpretation of the average value .
1. f(x) = x 3 - x + 5, x E [1, 5] 2. f(x)=x 112 -x+7,xE[l,9 ]
3. g(x) = 3 sin0 .2x, x E [1, 7] 4. h(x) = tanx, x E [0.5, 1.5]
5. v(t) = fl , t E [l, 9] 6. v(t) = 100(1 - e - t), t E [O,3]
For Problems 7-10, find a formula in terms of k for the average value of the given function
on the interval [O,k], where k is a positive constant (a stands for a constant).
7. f(x) = ax 2 8. f(x) = ax 3
9. f(x) = aex 10. f(x) = tanx, k <
11. Average Velocity from Acceleration Problem : Suppose you are driving 60 ft / sec (about
40 mi/ hr) behind a truck. When you get th e opportunity to pass, you step on the
accelerator, giving the car an acceleration a = 6 / fl , where a is in (feet per second)
per second and t is in seconds . How fast are you going 2 5 sec later when
you have passed the truck? How far did you travel in that time? What
was your average velocity for the 25-sec interval? Speed
50
12. Ida's Speeding Ticket Problem: Ida Livermore is rushing to take
pizzas to her customers when she is stopped for speeding . Her ticket
states that she was clocked at speeds up to 50 mi/hr during a 4-min 30 ---------------
per iod, and that her fine will be $140 ($7 for each mile per hour over
the 30-mi/ hr speed limit) . Ida is good at calculu s. She figures that
her speed was a quadratic function of time over the 4-min period, '' Time
being 30 mi/ hr at the beginning and end, and peaking at 50 mi/ hr
1 2 3 4
(Figure 10-3d). She argues she should be charged only for her average
speed above 30 mi/ hr. How much less will her fine be if she wins her Figure l 0-3d
appeal?
13. Average Velocity for Constant Acceleration Prob lem: Prove that if an object moves
with constant acceleration a, such as it does in ideal free fall, then its average velocity
over a time interval is the average of the velocities at th e beginning and end of the
interval . (This result leads to one of the physics formulas you may have learned,
and which may have led you to a false conclusion about average velocity when the
acceleration is not constant.)
Sectionl 03: Average Value Prob
lemsinMotion and Elsewhere 515
14. Average Velocity for Other Accelerations Problem: Show by counterexample that if the
acceleration of an object varies over a time interval, then the average velocity over that
interval might not equal the average of the velocities at the beginning and end of the
interval.
15. Average Voltage Problem: For the normal alternating current supplied
to houses, the voltage varies sinusoidally with time (Figure 10-3e), v = A Isin 120ntl
A
making 60 complete cycles each second . Thus 110
v = A sin 120rrt, I \
T'2o\
where v is in volts, t is in seconds, and A is the maximum voltage \
\ I
I
' ....,,,.'
during a cycle. The nominal "llO volts" is the average of the absolute
value of the voltage. Use the fundamental theorem to find the average Figure l 0-3e
value of y = 1Asinl20rrt l from t = 0 tot= 1/ 60. Use the result to
calculate the maximum voltage A if the average is 110 v. Show how this
number can be found more easily using y = sinx and an appropriate y=sinx
interval of integration.
16. Root Mean Square Deviation Problem : To measure how hilly a land-
X
scape is or how rough a machined surface is, people ask the question,
"On average, how far do points on the surface deviate from the mean
level?" If you simply average the deviations, you will get zero . As
shown in Figure 10-3f, there is just as much area above the mean as
there is below. One way used to overcome this difficulty is Figure l 0-31
Square the deviations .
Find the average of the squares.
Take the square root of the average to get an answer with the same dimensions as
the original deviations.
The result is called the root mean square deviation . For instance, the roughness of a
machined surface might be reported as, "0.1 microinches, rms," where rms stands for
root mean square, and a microinch is one-millionth of an inch.
a. Suppose that the deviations from average are sinusoidal, as in Figure 10-3f. That
is, d = k sin x, where d is deviation, x is displacement along the surface, and k is a
constant amplitude . Find the average of d 2 for one complete cycle. Use the result
to calculate the rms deviation .
b. Plot the graph of y = sin 2 x. Sketch the result. Show that the resulting graph is itself
a sinusoid and find its equation .
c. Show that the answer to 16a can be determined graphically from 16b, without
having to use calculus .
d. Suppose a surface is "lumpy," as in Figure 10-3g, and has the shape y =l sinx l
of the graph of y = lsinxl. Find the average value of y, Yav Then
find therms deviation, using the fact that
Deviation= y - Yav X
7t
Based on your answer, would this surface be rougher or smoother
than a sinusoidal surface with the same maximum distance Figure l 0-3g
between high points and low points as in Figure 10-3g7
516 Chapte
r 10:TheCalculusofMotion- Averages,
Extremes,and Vec
tors
10-4 Related Rates
Thus far in this chapter you have considered only motion along a line. Now you
will investigate situations in which more than one distance is varying. You will find
relationships between the rates at which the distances or other variables change.
OBJECTIVE Given a situation in which several quantities vary, predict the rate at which one of them is
changing when you know other related rates.
Example 1 An airplane is flying 600 mi/hr on a horizontal path that will take it directly over an
observer . The airplane is 7 mi high (Figure 10-4a).
a. Write an equation for the rate of change of the line-of-sight distance, z, between
the observer and the airplane in terms of the horizontal displacement, x, from
observer to airplane.
b. Plot dz / d t as a function of x. Sketch the result.
c. How fast is z changing when xis 10 mi? When xis - 5 mi (that is, 5 mi beyond
the observer) ? Interpret the answers.
d. Interpret the graph in part b when x = 0 and when lx l is very large.
Observer
\
Figure l 0-40
Solutions a. You know the rate of change of x with respect to time t, and you want the
rate of change of z. So the key to the problem is establishing a relationship
between z and x. Here's how.
dx
Know: dt = -600 Why is it negative?
dz
Want: dt
Find a relation ship bet, veen x and z
z 2 = x 2 + 72
(Pythagorean theorem).
2zdz = 2xdx Differentiate implicitly with res pect to t.
dt dt Rememb er the chain rule!
dz = ~ dx = _ 600 ~ = - 600x By algebra and the Pythagorean theorem.
dt z dt z -Jx 2 + 49
Sectionl 0-4:Related
Rates 517
b. Figure 10-4b shows the graph.
c. When x = 10, dz / dt = -491.539 ... , meaning that the distance is decreasing
at about 492 mi / hr as the airp lane approaches. When x = - 5, dz / dt =
348.74 ... , meaning that the distance is incr easing at about 349 mi / hr as the
airplane flies away.
d. At x = 0, the airplane is directly overhead . Although it is still moving at 600
Figure l 0-4b mi/hr, the distanc e between the airplane and the observer is not changing
at that instant. Thus, dz / dt = 0, as shown by the graph . For large values of
lx l the rate approaches 600 mi/hr, the speed of the airplane. This happens
because z and x are very nearly equa l when the plane is far away.
Example 2 Suppose you are drinking root beer from a conical paper cup . The cup has a diameter
of 8 cm and a depth of 10 cm. As you suck on the straw, root beer leaves the cup at
the rate of 7 cm 3 /sec. At what rate is the level of the liquid in the cup chang ing
a. When the liquid is 6 cm deep?
b. At the instant when the last drop leaves the cup?
Solutions a. The secret to getting started is drawing an appropriate diagram, then
identifying the known rate and the wanted rate. Figure 10-4c shows a cross
y section through the cup. Since the level of the liquid varies, you should label
8 cm ---1 the depth with a variable, even though you are looking for the rate when the
depth is 6.
Let y = no. of cm deep.
Let x = no. of cm radius of liquid surface.
dV dy
10 cm
Know: dt = - 7cm 3 / sec Want: dt
V = rrx 2y Volume of a con e.
By the properties of similar triangles, x can be expressed in terms of y, thus
reducing the problem to just the two variabl es whose rates you know or
want.
Figure l 0-4c X y
4 = lQ =>X = 0.4y
.
.. V = 1 rr(0.4y) 2 (y) = 4 rry 3
3 75
dV 4 2 dy
dt = 25rry dt
dy 25 dV
--- 2 Get a general formula befor e you substitut e particular valu es .
dt 4rry dt
dy 25 -175
dt rr. 62 ( - 7) = rr = - 0.3868 .. . = - 0.39cm / sec
4 144
b. At the instant the last drop leaves, y = 0. Substituting y = 0 into the general
formula of part a leads to division by zero. So the liquid level is changing
infinit ely fast!!
518 Chapter
l 0: TheCalculus
ofMotion-Averages
, Extremes,
andVectors
Problem Set 1 0-4
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes .
Q1. The number of feet from the starting point is called - ?- .
Q2. The total feet traveled is called - ?- .
Q3. f lvl dt is us ed to find the -?-.
Q4. f v dt is used to find the - ?-.
QS. If velocity and acceleration are both negativ e, is the object speeding up or slowing down?
Q6. If dy / dx is negative, then y is getting -?- .
Ql. f xe" dx = - ?- .
QB.If y = xcosx, then y' = - ?- .
Q9. If y = lnx, then dy / dt = -?- .
Q 10. The only point of inflection in the graph of y = xex is at x = - ?- .
1. Bacter ia Spreading Prob lem: Bacteria are growing in a circular colony one bacterium
thick. The bacteria are growing at a constant rate, thus making the area of the colony
increase at a constant rate of 12 mm 2 / hr. Find an equat ion expressing the rate of
change of area as a function of the radius, r, of the colony. Plot dr / dt as a function of
r . How fast is r changing when it equals 3 mm? Describe the way dr / dt changes with
the radius of the circle.
2. Ba lloon Prob lem: Phil LaRupp blows up a spherical balloon . He reca lls that the volume
is (4 / 3) rr r 3 . Find dV / dt as a function of r and dr / d t . In order for the radius to increase
at 2 cm/ sec, how fast must Phil blow air into the balloon when r = 3? When r = 6? Plot
the graph of dV / dt under these conditions . Sketch the result and interpret the graph.
3. Ellipse Prob lem: You recall that the area of an ellipse is A = rrab,
where a and b are the lengths of the semiaxes (Figure 10-4d). b
Suppose that an ellipse is changing size but always keeps the a = 2b
same proportions, a = 2b . At what rate is the length of the major
axis changing when b = 12 cm and the area is decreasing at 144
cm 2 / sec?
Figure l 0-4d
4. Bathtub Problem: A tub has cross sections the shape of the
quartic parabola y = x 4 , where both x and y are in feet. The
tub is 5 ft long, has vert ical ends (which make it uncomfortable
for bathing, but convenient for calculus!), and sits as flat as
possible on the floor (Figure 10-4e). When you pull the plug,
water flows out at a rate proportional to the depth of the
water remaining, dV / dt = ky.
a. Find dV, the differentia l of volume for a horizontal slice,
in terms of x and dy . Then get an equation for dV / dt
involving dy / d t. In terms of the constant k, figure out how
fast the depth is changing when the water is 6 in. deep. Figure l 0-4e
Section 10-4: Related Rates 519
b. Would the rate be faster, slower, or th e same if you are sitting in the tub when the
water is 6 in. deep ?
Second
5. Base Runner Problem: Milt Farney hits a line drive to center field. Milt
As he rounds second base, he heads directly for third, running 90 ft 20 'ft/sec
at 20 ft/sec (Figure 10-4f). Write an equation expressing the rate '
of change of his distance from hom e plate as a function of his Third First
displacement from third base. Plot the graph in a suitable domain.
How fast is this distan ce changing when he is halfway to third? At 90 ft 90 ft
third? Is the latter answer reasonable? Explain.
6. Tugboat Problem: A tugboat moves a ship up to the dock by pushing Home
its stern at a rate of 3 m/ sec (Figure 10-4g). The ship is 200 m long. Figure l 0-41
Its bow remains in contact with the dock and its stern remains in
contact with the pi er. At what rate is the bow moving along the
dock when the stern is 120 m from the dock ? Plot the graph of this
rate as a function of the distance between the stern and the dock.
7. Rectangle Problem I: The length of a rectangle is increasing at 3
ft/ min, and the width is decreasing at 2 ft/ min . When th e length is
50 ft and the width is 20 ft, is the area of the rectangle increasing
or decreasing ? At what rate ?
8. Rectangle Problem II: A rectangle of len gth L and width W has a
constant area of 1200 in 2 . The length changes at a rate of dL / dt Figure l 0-49
inches per minute .
a. Find dW / dt in terms of Wand dL / dt.
b . At a particular instant the length is increasing at 6 in./min and the width is
decr easing at 2 in./ min. Find the dimensions of the rectangle at this instant.
c. At the instant in 8b, is the length of th e diagonal of the 0 1 m/ sec
rectangle increasing or decreasing? At what rate ? - o.3 m/s ec
9. Luke and Leia's Trash Compactor Problem : Luke and Leia are
trapped inside a trash compactor on the Death Star (Figure 10-4h). End
The side walls are moving apart at 0.1 m/ sec, but the end walls Side
are moving together at 0.3 m/ sec . The volume of liquid inside the Figure l 0-4h
compactor is 20 cubic meters, a constant.
a. Write an equation expressing the rate of change of depth of the liquid in terms of
the width and length of the region inside the compactor.
b. When the side walls are 5 m lon g and the end walls are 2 m long, is the depth of
liquid incr easing or decreasing? At what rate?
10. Darth Vader's Problem : Darth Vader's spaceship is approaching the origin along the
positive y-axis at 50 km/ sec. Meanwhile, his daughter Ella's spaceship is moving away
from the origin along the positive x-axis at 80 km / sec . When Darth is at y = 1200 km
and Ella is at x = 500 km, is the distance between them increasing or decreasing? At
what rate?
520 Chapter
l 0: TheCalcu
lusofMotion-Averages
, Extremes,andVectors
11. Point on a Parabo la Prob lem: A point moves along the parabola
y = x 2 (Figure 10-4i) in such a way that the rat e of change of x with
respect to tim e t is directly proportional to th e value of x . That is,
dx / dt = kx, where k stands for a constant .
a. Find a general equation for dy I dt as a function of x.
b . Find the constant of proportionality, k, if dx / dt = 10 when
X = 5.
c. Find the value of dy I dt when x = 7. X
dx/dt = kx
12. Point on a Tangent Graph Prob lem: A particle moves along the
graph of y = tanx . Its velocity in the x-dir ection is 5 units per Figure l 0-4i
minute. When x = 1, what is its velocity in the y-direction?
13. Barn Ladder Probl em: A ladder to the loft in a barn is arranged so
that it can be pushed up against the wall when it is not in use . The
top of the ladder slides in a track on the wall, and the bottom is
free to roll across the floor on wheels (Figure 10-4j). To make the
ladder easier to move, a counterweight is attached to the top of the
ladder by a rop e over a pulle y. As the ladder goes away from the
Weight
wall, the counterweight goes up and vice versa.
a. The ladder is 20 ft long . Write an equation expressing th e
i
velocity the counterweight moves as a function of the distance Wall
the bottom of the ladder is from the wall and the velocity the Floor
bottom of the ladder moves away from the wall.
b. Find the velocity of the counterweight when the bottom is 4 ft Figure l 0-4j
from the wall, and is being pushed toward the wall at 3 ft/ sec.
c. If the ladder is allowed to drop all the way to the floor with the bottom moving at 2
ft/ sec and the top remains in contact with the wall, how fast is the counterweight
moving when the top just hits the floor? Surprising?!
14. Kinetic Energy Problem: The kinetic energy of a moving object equals half the product
of the mass and the square of the velocity.
K = lmV 2
2
As a spaceship is being fired into orbit, all thr ee of thes e quantities vary. Suppose
that the kinetic energy of a particular spaceship is increasing at a constant rate of
100,000 units per second and that mass is decreasing at 20 kg/sec because fuel is
being consumed . At what rate is the velocity changing when the
spaceship's mass is 5000 kg and it is traveling at 10 km / sec?
15. Conical Water Tank Prob lem: The water tank shown in Figure 10-4k
has diameter 6 m and depth 5 m.
a. If the water is 3 m deep, and is rising at 5 m/ hr, at what rate is
th e volume changing?
b. If the water is being pumped out at 2 m 3/ hr, at what rate is the
depth changing
i. When the water is 4 m deep? Figure l 0-4k
ii. When the last drop is pumped out?
c. If water flows out under the action of gravity, the rate of change
of volume is directly proportional to the square root of the water's
Section
l 0-4: Related Rates 52 1
depth. Suppose that the volume Vis decreasing at 0.5 m 3 / hr when
the depth is 4 m.
i. Find dV / dt in terms of the depth of the water.
ii. Find dV Id t when the water is 0.64 m deep.
iii. Find the rate of change of depth when the water is 0.64 m deep.
16. Conical Tank Generalization Problem: Suppose that a conical tank as
in Figure 10-4k has a radius of a units and a depth of b units. Water Figure 10-41
enters or leaves the tank at dV / dt cubic units per unit time, where
dV / dt is not necessarily a constant. Write an equation expressing the
rate of change of depth of the water in the tank as a function of depth
and dV / dt.
17. Cone of Light Problem : A spotlight shines on the wall, forming a cone
of light in the air (Figure 10-41). The light is being moved closer to
the wall, making the cone's altitude decrease at 6 ft/ min . At the same
time, the light is being refocused, making the radius increase at 7
ft/min. At the instant when the altitude is 3 ft and the radius is 8 ft,
Figure 10-4m
is the volume of the cone increasing or decreasing? How fast?
18. Slag Heap Problem: Slag left over from the manu-
facture of cement is being piled outside the cement
plant. The resulting slag heap is a cone whose el-
ements make an angle of 40 with the horizontal
(Figure 10-4m). Environmentalists measure the
circumference of the heap one day, finding it to be
3000 ft, and increasing at 7 ft/ day. About how fast
is the cement plant generating slag7
19. Cone in Hemisphere Problem: A right circular cone
is inscribed in a hemisphere (Figure 10-4n). The
figure is expanding in such a way that the combined
surface area of the hemisphere and its base is increasing at a constant
rate of 18 in 2 / sec. At what rate is the volume of the cone changing
when the radius of the common base is 4 in.?
20. Planetary Motion Problem: On October 10, 1988, Mars was at its
closest position to earth. It then receded at an increasing rate
Figure 10-4n
(Figure 10-40). In this problem you will analyze the rate at which
the distance between the two planets changes. Assume that the
orbits of the two planets are both circular and are both in the
ecliptic plane. The radius of earth's orbit is 93 million mi and
the radius of Mars' orbit is 141 million mi. Assume also that the
speed each planet moves along its path is constant. (This would
be exactly true if the orbits were circular.) Mars orbits the sun
once each 687 earth-days. The earth, of course, orbits once each
365 earth-days . Answer the following questions.
Figure 10-40
522 Chapter
l 0: TheCalculus
ofMotion-Averages,
Extremes,
andVectors
a. What are the angular velocities of earth and Mars about the sun in radians per day?
What is their relative angular velocity? That is, what is d0 / dt?
b. What is the period of the planets' relative motion? On what day and date were the
two planets next at their closest?
c. Write an equation expressing the distance D between the planets as a function of e.
d. At what rate is D changing today? Transform your answer to miles per hour.
e. Will D be changing its fastest when the planets are 90 apart? If so, prove it. If not,
find the angle e at which D is changing fastest. Transform the answer to degrees.
f. Plot the graph of D versus time for at least one period of the planets' relative
motion. Is the graph a sinusoid?
21. Speeding Piston Project: The engine in a car is going 3000 revolutions
per minute (rpm). As the crankshaft turns in the xy-plane, the piston y
goes up and down inside the cylinder (Figure 10-4p). The radius of the
crankshaft is 6 cm. The connecting rod is 20 cm long and fastens to a 18cm
point 8 cm below the top of the piston. Let y be the distance from the
top of the piston to the center of the crankshaft. Let B be an angle in
standard position measured from the positive x-axis and increasing
as the crankshaft rotates counterclockwise.
a. Write an equat ion expressing yin terms of e.You should find that
the law of cosines is helpful.
b. Find an equation for the rate of change of y with respect to t in
terms of e and d0 / dt.
c. Find an equation for the acceleration of the point on top of the
piston. Remember that de/ dt is a constant since the engine is
going at a constant 3000 rpm.
d. Between what two values of e is the top of the piston going down Figure 10-4p
with an acceleration greater than that of gravity (980 cm/ sec 2 )?
10-5 Minimal Path Problems
Suppose you are swimming in the ocean. When you finish, you could swim straight
to the place on the shoreline where you left your towel, or you could swim straight
to the shoreline, then walk to the towel (Figure 10-Sa). Swimming straight to the
y~ _ Ath is bes t
7
1~ 7...._
I ,
f__
...._
Ocean
T ",. ._...._...._
1 ',, Shoreline ...._...._
- - ....,._ - ~-;.:';.:"-:...-It::"'"..:'"!,.'*"'..:'----;.:-_ ...
Towel
Beach
Figure 10-5a
Section
l 0-5:Minimal
PathProblems 523
towel minimizes the distance you must go. Heading for the closest point on the
shoreline minimizes your time in the water. But your total time might be minimized
if you head for a point between the towel and the closest point because you can
walk faster than you can swim.
A third alternative is to swim to some point between the towel and the closest point
on the shoreline, shortening the walk while increasing the swim by just a little. Your
objective in this section will be to analyz e such problems to find the minimal path,
which takes the least total time.
OBJECTIVE Given a situation in which something goes from one place to another through two different
media at different rates, find the path that minimizes a total time or cost.
Example 1 Suppose you are 200 yd from the beach and want to get to the place where you
left your towel in the minimum possible time . You can walk on the beach at llO
yd/ min but can swim only 70 yd/ min. Let x be the distance from the foot of your
perpendicular to the beach to the point where you will make landfall (Figure 10-Sb).
What value of x minimizes your total time if your towel is
a. 600 yd from the foot of the perpendicular?
b. 100 yd from the foot of the perpendicular ?
You Swim distance
,--- : ~
I
200 ycl :
,
:
!
Landfall point
Walk clttance
Towel
- X - 1---600 - X---+-
I 600 yd----+-
Figure l 0-5b
Solutions a. Distance to swim is -Jx2 + 200 2.
Distance to walk is (600 - x), where x 5 600 .
.. total time will be
Minimum
atx = l65 t = t5-Jx2 + 200 2 + 1;0 (600 - x).
10 /...
To find the minimum total time, find the distance, x, at which t stops
decreasing and starts increasing. This can be done graphically by plotting
X t as a function of x and tracing (Figure 10-Sc). The minimum is somewhere
600 between x = 160 and 170.
Figure l 0-5c The exact value can be found algebraically by finding the value of x that
makes the derivative of t equal to zero.
dt = _!_(x2 + 200 2) - lf 2(2x) - _J_
dx 14 0 ll O
= ..':...
70
(x 2 + 200 2)- 112 - _!_
110
524 Chapter
l 0: TheCalculus
ofMotion-Averages,
Extremes,
andVectors
dt =O if and only if
dx
3__
70
(x2 + 2002)- 112= __!_
110
Take th e recipro cal of both sides, th en
70(x 2 + 200 2) 112= llOx multipl y by x.
4900(x 2 + 200 2 ) = 12100x 2
4900(200 2) = 7200x 2
+ 164.9 ... is in the desired rang e, thus
X = 164.991 5 . .. confir min g the gra phi cal solution .
Head for a point about 165 yd from the foot of the perpendicular .
b . If the towel is only 100 yd from the foot of the perpendicular, the domain of
xis [O, 100], and the equation fort is
Minimum at
10 end point ! t = t5J x2 + 200 2 + iio(100 - x).
/ X
The graph of this equation is shown in Figure 10-5d. Surprisingly, the
minimum is at the same value, x "" 165, as in part a. The 600 or 100
100 disappears when you differentiate, so the zero of the derivative is not
affected by this number. However, the 165 is out of the domain. It is beyond
Figure l 0-5d the towel! By the graph in Figure 10-5d, you can see that the minimum time
comes at the endpoint, x = 100.
Head straight for the towel.
These problems are sometimes called "drowning swimmer" problems, because they
ar e often phrased in such a way that a person on the beach must rescue a drowning
swimmer and wants to take the path that reaches the swimmer in a minimum length
of tim e. (The swimmer might drown if one were to take all that time to do th e
calculations')
In Problem Set 10-5, you will work more minimum path problems . Then you will
show that there is a remarkably simple way to find the minimum path if you will
do calculus algebraically on the general case instead of "brute-force" plotting on a
particular case.
Problem Set 10-5
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
QI. Solve: Jx = 9
Q2. Differentiate: y = ) 100 - x 2
Q3. lntegrate: Jx,,/100 - x 2 dx
Q4. Differentiate: y = sin - 1 3x
QS. Inte grate: Jxe 2" dx
Q6. Differentiate : y = tanh x
Ql. Going 60 cm at 40 cm/ hr takes -?- hr .
Section l 0-5: Min
imalPathProblems 525
For Problems Q8-Ql0, use th e velocity-time function in Figure 10-Se:
QB
. At what time(s) is the moving object at rest ?
Q9. At what time(s) does the moving object change directions?
Q10. At what time(s) does the acceleration function
have a local maximum ?
Figure l 0-5e
1. Swim and Run Problem: In a swim-and-run biathlon, Ann ---100m---.
Athlete must get to a point on the other side of a 50-m-wide
river, 100 m downstream from her starting point (Figure
10-Sf). Ann can swim 2 m/ sec and run 5 m/ sec. Toward what
point on the opposite side of the river should Ann swim in Run at 5 m/sec
order to minimize her total time?
Figure l 0-5 f
2. Scuba Diver Problem: A scuba diver heads for
a point on the bottom that is 30 m below the
surface and 100 m horizontally from the point
where she entered the water. She can move
13 m/ min on the surface but only 12 m/ min
as she is descending. How far from her entry
point should she start descending to reach her
destination in minimum time?
3. Pipeline Problem: Earl Wells owns an oil lease. A
new well 300 m from the road is to be connected
to storage tanks 1000 m down the road from the
well (Figure 10-Sg). Building pipeline across the
field costs $ 50 per meter, while building it along
the road costs only $40 per meter. How should
the pipeline be laid out in order to minimize its
total cost7
4. Elevated Walkway Problem: A walkway is to be built from the corner of one building to
the corner of another building across the street and 400 ft down the block. It is 120 ft
across the street (Figure 10-Sh). Engineering studies show that the walkway will weigh
3000 lb/ft where it parallels the street, and 4000 lb/ ft where it crosses the street. How
should the walkway be laid out in order to minimize its total weight?
r,___
Buildin g
1000 m --- 1
On road: $40 / m
120 ft
tJ -
300 Pipelin e
Tank
battery
-Through field: $50/ m
Well
Figure l 0-59 Figure l 0-5h
526 Chapter
10: TheCalculus
of Motion
- Averages,
Extremes,
andVectors
5. Minimum Path Discovery Problem: In this problem you will
explore a relationship between the minimum path and the
two speeds or costs per meter.
a. In the Swim and Run Problem, let e be the (acute) angle
Run: Sm/sec
between the slant path and a perpendicular to the river
(Figure 10-5i, top diagram). Show that the sine of this
angle equals the ratio of the two speeds. That is, show that
sine= 2/ 5. On road: $40/m
Through field:
b. In the Pipeline Problem, let e be the acute angle between
the slant path and a perpendicular to the road (Figure
$50/ m e,
...
10-5i, bottom diagram). Show that the sine of this angle .
equals the ratio of the two costs per meter. That is, show Figure 10-5i
that sine= 40 / 50.
6. Minimum Path Generalization Problem: A swimmer is at a
distance of p ft from the beach. His towel is at the water's
edge, k ft along the beach (Figure 10-5j). He swims at an angle
,
Swimmer
r: ~''
Swim at rate s
e to a line perpendicular to the beach and makes landfall at 0 : Wal at rate w
a point x ft from the point on the beach that was originally
closest to him. He can swim s ft/min and walk w ft/min, Towel
where s < w. Prove that his total time is a minimum if the
sine of the angle the slant path makes with the perpendicular
equals the ratio of the two speeds. That is, prove the following Figure 10-5j
property.
Property:Minimum Path
For the path shown in Figure 10-Sj, if sin 0 = s/w, then the total time taken is a minimum.
7. Scuba Diver Problem Revisited: Work Problem 2 again, using the minimum path
property of Problem 6. Which way is it easier to work the problem? Tell one reason
why mathematicians find general solutions to typical problems.
8. Elevated Walkway Problem Revisited: Work Problem 4 again, using the minimum path
property of Problem 6. Since weight equals distance multiplied by cost per foot (rather
than distance divided by velocity), you will have to adapt the property appropriately.
Which is the easier way to work the problem? Tell one reason why mathematicians
find general solutions to typical problems.
9. Pipeline Problem, Near Miss: Suppose you present your boss with a solution to the
Pipeline Problem (Problem 3), but in order to save some trees, the value of x must be a
few meters away from the optimum value. Will this fact make much difference in the
total cost of the pipeline? Tell how you reach your conclusion.
Section10-5: Min
imalPathProblems 527
10. Calvin and Phoebe's Commuting Problem: Calvin lives at the j-500 ft--+-j
corner of Alamo and Heights Streets (Figure 10-5k). Phoebe Phoebe's D 14-x -1
lives on High Street, 500 ft from its intersection with Heights High Street
Street. Since they have been going steady, Calvin finds that he
is spending a lot of of time walking between the two houses. Heights
Street
He seeks to minimize the time by cutting across the field to a
point that is x ft from the Heights-High intersection. He finds
that he can walk 5 ft/sec along the streets but only 3 ft/sec 1200 ft
across the rough, grassy field. What value of x minimizes
Calvin's time getting to Phoebe's?
11. Robinson Crusoe Problem: Robinson Crusoe is shipwrecked
on a desert island. He builds a hut 70 yd from the shore . His Alamo Street
wrecked ship is 120 yd from the shore, 300 yd down from the Calvin's D
hut (Figure 10-51).Crusoe makes many trips between the hut
and the ship and wants to minimize the time each trip takes. Figure 10-5k
He can walk 130 yd/min and pole his raft 50 yd/ min. Where
on the shoreline should he moor his raft so that the trips can
be made in minimum time?
12. Robinson Crusoe Follow-Up Problem: Let 0 1 and 02 be the T:
120 yd;
Ocean
angles between the two paths and a line perpendicular to the
beach (Figure 10-5m) in the Robinson Crusoe Problem. Use
the answer to the problem to calculate the measures of these
angles . Then show that the ratio of the sines of the angles
equals the ratio of the two speeds . That is, show ,,---
... 300 yd---
sin 0 1 50 Figure 10-51
sin0 2 130
13. Robinson Crusoe Generalizat ion Problem: Figure 10-5n shows the general case of the
Robinson Crusoe Problem, where the wreck, A, and the hut, B, are a and b units from
the shore, respectively, and k units apart parallel to the shore. The velocities through
the water and on land are v 1 and v2 , respectively. Prove that the minimum time from
wreck to hut is where
sin0i
----
V1
k B
Figure 10-5m Figure l 0-5n
528 Chapter
l 0: TheCalculus
ofMotion-Averages,
Extremes,
andVectors
14. Snell' s Law of Refra ction Probl em: About 350 years
ago the Dutch physicist Willebrod Snellius ob- Incident light ray
served that when light pass es from one substan ce
into another, such as from air to wat er, th e rays
bend at the int erfa ce (Figure 10-50). He found
that the angles 0 1 and 02 th at th e incomin g and Substance 1 Interface
outgoing rays make with a p erp endi cular to the Substan ce 2
interface obey th e following rul e:
Sin 0 1 = V1
sin 02 v2 Refracted light ray
where v 1 and v2 are th e sp eeds of light in th e two Figure l 0-50
substan ces.
a. In what ways doe s thi s real-world situation corr espond to the Robin son Crusoe
Probl ems?
b. What can you conclud e about th e time light takes to get from one point to anoth er
when it pass es throu gh diff erent substanc es?
c. When you are above th e surfa ce of the water in a swimmin g pool, objects on the
bottom always app ear closer to the sur face th an th ey actually ar e. Explain how thi s
can be true, and what this fact tells you about th e relative speed s of light in air and
in wat er.
15. Journal Problem: Updat e your journal with thin gs you ha ve learned since th e last entr y.
You should includ e su ch thin gs as tho se listed here.
The on e most important thing you hav e learn ed since th e last journal en try
The big diff erenc e between displa cement and distan ce
The meaning of average valu e of a function
How one solves a typical relat ed rat es probl em
The main id ea behind minimum path probl ems
Any techniques or ideas about obj ects in mo tion th at are still uncl ear
10-6 Maximum and Minimum Problems in Motion
and Elsewhere
In Section 10-5, you found minimum path s, whereby one could tra vel between
points in th e least time or build stru ctur es between point s for th e least cost. In this
section you will find maximum and minimum valu es in oth er ph enom ena.
OBJECTIVE Given a situation in the real or mathematical world where a function is to be ma'<imized or
minimized, write an equation for the function and find the maximum or minimum values.
The techniqu e is similar to that in th e pr evious section and in th e an alysis of plan e
and solid figur es from Section 8-3. Therefor e, no spe cific exampl es are given in this
section.
l 0-6:MaximumandMinimumProblem
Section s in MotionandElsewhere 529
Problem Set 1 0-6
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Q1. If f(x) = x sinx, find f '(x) .
Q2. If g(x) = xlnx, find g"(x).
Q3. If h(x) = xex, find Jh(x) dx.
Q4. The minimum path problems of Section 10-5 are equivalent to what law from physics?
QS. ln(exp x) = -?-
Q6. Sketch a paraboloid.
Ql. Sketchy = x 113 .
QB. s:
lvelocityld(time) = total distance or net displacement?
Q9. Who is credited with inventing calculus?
Q10. Find the sum: 1 + 2 + 3 + + 98 + 99 + 100
1. Rocket Problem: Jeff is out Sunday driving in his spaceship. As he approaches Mars, he
fires his retro rockets. Starting 30 seconds later, his distance from Mars is given by
1
D = t+-
t'
where D is in thousands of miles and t is in minutes. Plot the graphs of D and D '
versus t. Sketch them on your paper. What are his maximum and minimum distances
from Mars in the tin1e interval [0.5, 3]? Justify your answers.
2. Truck Problem : Les Moore owns a truck. His driver, Ouida Liver, regularly makes the
100-mile trip between Tedium and Ennui. Since Les must pay Ouida $10 per hour to
drive the truck, it is to his advantage for her to make the trip as quickly as possible.
However, the cost of fuel varies directly with the square of the speed, and is $.10 per
mile at a speed of 30 mph. The truck can go as fast as 85 mph. Plot cost and derivative
of cost versus speed. Sketch the graphs. What speed gives the minimum total cost for
the 100-mile trip? (Ignore the cost of tickets, and so on, for going over the speed limit.)
Justify your answer.
3. Number Problem I: Find the number which exceeds its square by the greatest amount.
That is, find x if x - x 2 is to be as large as possibl e.
4. Number Problem II: Find the number greater than or equal to 2 which exceeds its
square by the greatest amount.
5. Fran's Optimal Study Time Problem: Fran Tick forgot to study for her calculus test
until late Sunday night. She knows she will score zero if she doesn't study at all, and
that her potential score will be
S = lOOt
t+1
if she studies for t hours. She also realizes that the longer she studies, the more
fatigued she will become. So her actual score will be less than the potential score. Her
530 Chapter
l 0: TheCalculus
of Motion-Averages,
Extremes,
andVectors
"fatigue factor" is
F = _ 9_
t+9
This is the number she must multiply by the potential score to find her actual gra de, G.
That is, G = S F.
a. Sketch the graphs of S, F, and G versus time, t.
b . What is the optimum number of hours for Fran to study? That is, how long should
she study to maximize her estimated grade, G7
c. How many points less than the optimum will Fran make if she studies
i. 1 hour more than the optimum?
ii. 1 hour less than the optimum?
6. Motor Oil Viscosity Problem : The viscosity (resistance to flow) of normal motor
oil decreases as temperature goes up. All-temperature oils have roughly the same
viscosity throughout their range of operating temperatures. Suppose that lOW-30
motor oil has viscosity
= 130 - 12T + 15T 2 - 4T 3 ,
where (Greek letter "mu") is the viscosity in centipoise, T is the temperature in
hundreds of degrees, and the equation applies for temperatures from 0 through
300 C.
a. Find the temperature in the domain at which the maximum viscosity occurs.
b . Find the minimum viscosity in the domain.
c. Suppose that the oil is being heated in such a way that T = ./f, where t is time in
minutes . At what rate is the viscosity changing when the temperature is 100(?
7. Cylinder-in-the -Cone Problem I: A right circular cone has an altitude of 6 inches and
a base of radius 10 inches . A right circular cylinder is inscribed in the cone, coaxial
with it.
a. Plot the graphs of volume and total surface area of the cylinder as a function of its
radius. Sketch the graphs .
b . What are the radius and altitude of the cylinder which give it the maximum volume?
The maximum total surface area? Do the two maxima occur at the same radius?
8. Cylinder -in -the-Cone Problem II: A cone of radius 6 in. and altitude 18 in. has a cylinder
inscribed in it . The cylinder's altitude starts at O in . and increases at 2 in./min.
a. When the altitude of the cylinder is 12 in., will its volume be
increasing or decreasing? At what rate?
b . What will be the maximum volume of the cylinder in the time
interval [O,9]7 Justify your answer . y = x4+ 5
c. What will be the maximum volume of the cylinder in the time
interval [4, 6]7 5
9. Quartic Parabola Tank Problem : A water storage tank has the shape
of the surface formed by rotating about the y-axis the graph of X
y = x 4 + 5, where x and y are in meters (Figure 10-6a). At what rate
is the depth of the water changing when the water is 3m deep in Figure 10-60
the tank and is being drained at 0.7m 3 per minute?
tion l 0-6: MaximumandMinimumProblemsin MotionandElsewhere
Sec 531
10. Cylinder in Paraboloid Problem: A paraboloid is formed by rotating
the graph of y = 4 - x 2 around the y-axis. A cylinder is inscribed in
the paraboloid (Figure 10-6b).
a. If the radius of the cylinder is 1.5 units, and is increasing at
0.3 units per second, is the volume of the cylinder increasing or
decreasing? At what rate ?
b. What is the maximum volume the cylinder could have? Justify your
answer.
11. Pig Sale Problem: Ann Aggie's pig weighs 1000 lb, and is gaining 15 X
lb/ day. She could sell it for $900 at today's price of 90 cents per 2
pound, or wait till it gains some more weight and hope to get more
than $900. Unfortunately, the price per pound is dropping at 1 cent Figure l 0-6b
per pound each da y. So she must decide when is the best time to sell.
a. Write functions for weight and for price per pound in terms of the number of days
after today. Then write a function for the total amount Ann will get for the pig.
b. Find the time when the derivative of total amount will b e zero. Convince Ann that
at this time the total amount is a maximum, not a minimum.
c. If Ann sells at the time in llb, how much will she get for the pig?
12. Bridge Problem: Suppose that you work for a construction company that has a contract
to build a new bridge across Scorpion Gulch, downstream from the present bridge
(Figure 10-6c). You collect information from various sources in order to decide just
where to build the bridge. From the surveyors you find that the width of the gulch is
Width= 10(x 2 - Bx + 22) ft,
where x is the number of miles downstream from the old bridge.
The depth of the water is
Depth = 20x + 10ft.
a. The Structural Department specifies that the bridge can be
Figure 10-6c
built in water as much as 130 ft deep, but that it can be no
more than 310 ft long. The City Traffic Department specifies that the bridge must
be at least 1 mi downstream from the present bridge. What is the domain of x?
b. What are the shortest and longest lengths the bridge could be?
c. The cost of building the bridge is proportional to the product of the length of the
bridge and the depth of the water. Where should the bridge be built to minimize
the cost? Justify your answer.
d. Is the shortest bridge also the cheapest bridge? Explain.
10-7 Vector Functions for Motion in a Plane
Until now you have considered velocity and acceleration of an object moving back
and forth in a line. The displacement, x, from some fixed point depends on time, t.
532 Chapter
10:TheCalculus
ofMotion- Averages,
Extremes,
andVectors
#I
Y Curved path In this section you will consider objects moving along a path in a plane. Both x and y
will depend on t. For such motion the velocity and acceleration may act at an angle
to each other . Therefore, it is convenient to use vectors to represent the objects'
Acceleration position, velocity, and acceleration.
Figure 10-7a shows the path a moving object might take in a plane. Its ve locity
vec tor points in the direction of motion and has magnitude (length) equal to the
Position speed of the object . Its acceleration vector, acting at an angle to the path, changes
X
both the object's speed and direction, thus pulling it into a curved path . The position
of the object can be represented by a pos ition vect or, which tells the direction and
Figure l 0-7a distance from the origin to the object at any given time, t .
Parametric equations can be used to write the position vector as a function of time.
The resulting vector fun ction can be differentiated to find velocity and acceleration
vectors .
OBJECTIVEGiven the equation of a vector fW1ction for the position of a moving objec t, find the
first and second derivatives of position with respect to time, and interpr et the way these
vectors influence the motion of the object.
Background:
Vectors
A vec tor quantity is a quantity that has both magnitude and direction. Quantities
that have only magnitude are called scalar quantities . Volume, mass, time, distance
(not displacement!), and money are scalar quantities . They can be represented by
points on a "scale" such as a number line. (The word "scalar" comes from the Latin
scalaris, meaning "like a ladder.")
Vector quantities can be represented mathematically by direct ed line segments,
called simply vectors . An overhead arrow such as v denotes a variable used for a
Tail (beginning)
vector (Figure 10-7b). The length of the line segment represents the magnitude of
Figure l 0-7b the vector quantity, also called its absolute value or its norm. The direction the
segment points represents the dir ection in which the vector quantity acts .
Two vectors are equal if and only if they have the same magnitude and the same
direction (Figure 10-7c, left) . You are free to move vectors around from place to
place, as long as you keep them pointing the same way and don't change their
lengths . The opposi te of a vector, written -v , is a vector of the same magnitude as
v but pointin g in the opposite direction (Figure 10-7c, middle). The zero vect or is a
vector of magnitude zero. It can be pointing in any direction!
II
Equal vectors Opposite vectors
Figure l 0-7c
Zero vector
Section l 0-7: Vector
Functions
forMotion
ina Plane 533
,Qs
Vectors are added by moving one of them so that its tail is at the head of the other
(Figure 10-7d). The sum is the vector from the beginning of the first vector to the
end of the second. Vectors are subtracted by adding the opposite of the second
vector to the first one.
Vector addition Vector subtraction
Figure l 0-7d
Vectors are most easily worked with in a coordinate system (Figure 10-7e). Any
vector v is a sum of a vector in the x-direction and one in the y -direction (called
components of the vector). If 7 and ] are unit vectors in the x- and y-directions,
respectively, then
V = Xi+y] .
The dot product (inner product or scalar product) of two vectors is the number
you get by placing the vectors tail-to-tail (Figure 10-7f), then multiplying their
magnitudes and the cosine of the angle between the vectors; that is,
v1 v2 = lv1 Ilv2Icos e.
In Problem 14 of Problem Set 10-7, you will show that the dot product is also equal
to
V1 V2 = X1X2 + Y1Y2,
where v1 = ><1 7 + Y1J and v 2 = ><27 + Y2J. The dot product is useful for finding the
length of a "shadow" that one vector would cast on another. This scalar projection,
P, of v 1 on v 2 is given by
p = lv1 Icos e,
as shown in Figure 10-7g. Multiplying by a clever form of 1 makes a dot product
appear on the right-hand side of th e equation .
p = lv1llv2lcose = v1 v2
lv2I lv2I
The vector projection of v1 on v 2 is a vector in the direction of v 2, with magnitude P.
Derivatives
al a Position
VectorFunction-Velocity
andAcceleration
The derivative of a scalar-valued function is the limit of 6y I t:.x as t:.x approaches
zero. Similarly the derivative of a vector function, r,is
dr = lim t:.r.
dt L'>C- 0 6t
534 Chapter
l 0: TheCalculus
ofMotion-Averages,
Extremes,
andVectors
fflti t
y
4
3 _,
v,
2 _,
v I ..., v1
' yj lI Vector
I
I
..., I
I ~ prnjec<io,
J xi I X
_,
-1
..., 1 2 4 v2
I 3
-1 I Scalar projection I
v;
Figure 10-7e Figure 10-71 Figure 10-7g
Figure 10-7h shows vector r to the position of a moving object at time t. Vector 6.r
y is the vector from this position to a new position at time t + 6.t. Vector 6.r can be
resolved into horizontal and vertical components, 6.x 1 and 6.y]. So,
dr = lim 6.xi + 6.y]
dt M-0 6.t
. (6.x--, 6.y-,) 6t di str ibut es, and can be associate d with
= 11m - 1+ - J
M - 0 6.t 6.t b.x and 6y.
r 6.x--,
1
r 6.y-, The limi t of a sum property applies to
= "'~~ M + "'~~ M J vector s.
dx--, dy.., Definition of derivative. (The unit vecto rs
Figure 10-7h = dt I + dt ). are con stant.)
Thus, if the components of a position vector ar e specified by functions oft, you can
find dr / dt simply by differentiating each function and multiplying the answers by
7 and]. As you will see in Example 1, the resulting vector is tangent to the path of
the moving object. By the Pythagorean th eorem, its length is
.Jdx 2 + dy 2 dL
dt dt.
But dL / dt is the speed of the object along its curved path. Thus dr / dt is the velocity
vector for the moving object. The acceleration vector is found by differentiating
the velocity vector 's components, the same way th e velocity is found from the
displacement.
Properties:VelocityandAccelerationVectors
If r(t ) = x(t) 7 + y(t)] is the position vector for a moving object,
then v(t) = r ' (t) = x'(t) 7 + y'(t)] is the velocity vector,
and a(t) = v '( t) = r " (t) = x"(t)i + y"(t)] is the acceleration vector.
The speed of the moving object equals lv(t) I = ) (dx / dt) 2 + (dy / dt) 2 .
Section10-7: Vecto
r Functions
forMotion
ina Plane 535
Example 1 Given the vector equation
r(t) = (5 sint)7 + (5 cos 2 t)]
for the position, r(t), of a moving object, where distances are in feet and time tis
in seconds,
a. Plot the path of the object. Sketch the result. Then show the position vectors
r(O), r(0.5), r(l) , r(l.5), and r(2). Interpret the location of r(2) .
b. Calculate the difference vector t,.r = r(l) - r(0.5) by subtracting the respective
components. Sketch t,.f with its tail at the head of r(0 .5). Where is the head of
t,.r?
c. The average ve locity vector of the object for interval [0.5, t] is the difference
quotient
r(t) - r (0.5)
Vav = t - 0.5
Find the average velocities for [0.5, 1] and [0.5, 0.6]. Sketch these vectors starting
at the head of the position vector r(0.5).
d. Find the (instantaneous) velocity vector, v(0 .5). Plot it starting at the end of
r(0.5), thus showing that it is tangent to the path . How do the average velocity
vectors in part c relate to the v(0 .5)?
e. Find the speed of the object at time t = 0.5.
Solutions a. Plot with your grapher in parametric mode
x = 5 sint
y = 5 cos2 t .
lf you use the grid on option it will be easier to sketch the graph on dot paper
(Figure 10-7i). The position vectors are as follows.
r(O) = 07 +]
r(0.5) = 2.39 .. . 7 + 3.85 .. . ]
rO) = 4.20 .. . 7 + 1.45 .. . ]
r(l.5) = 4.98 .. . 7 + 0.02 .. . ]
r(2) = 4.54 ... 7 + o.86 .. . ]
The heads of the vectors lie on the path. (That's why they are called position
vectors!) They progress clockwise from the vertical. At time t = 2 the object
has started back in the counterclockwise direction, as you can see by tracing.
b. t,.r = r(l) - r(0.5)
= (5sinl - 5sin0.5)7 + (5cos 2 1 - 5cos 2 0.5)]
= 1.81. .. 7 - 2.39 . .. ]
Figure 10-7j shows this difference vector . If you start it at the head of r(0 .5),
then t,.r ends at the head of r(l), as you can see by counting spaces on the
dot paper.
536 Chapter
l 0: The Calcu
lusofMotion-Averages,Extremes,
andVectors
ff&m#E*
l Wi
y y
5
- :-2 .391]
i"<zi
r (l.5) X X
5
Figure 10-7i Figure 10-7j
C. t = 1: Viv= r(l ) - r (0. 5)
' 1 - 0.5
1.81. .. 7- 2.39 . . . ]
0.5
= 3.62 ... 7- 4.78 ... J
_ . _ _ r (0.6) - r (0.5)
t - 0 .6 . V av- 0.6 - 0.5
0.4 26 .. . 7 - 0.444 .. . ]
0.1
= 4.26 ... 7- 4.44 ... ]
Figure 10-7k show s these average vectors plotted on the diagram in Figure
10-7j. You plot them by counting spaces on th e dot paper.
4.26 . . . i
.j:
...''
. ' .
Figure 10-7k
d. v(t ) = r '(t ) = (5c ost )7 + (-lOcostsint)]
.. v (0. 5) = 4.38 . . . 7 - 4.20 ... J
Sectionl 0-7:VectorFunctions
for Motionin a Plane 537
Figure 10-71 shows that vis tangent to the path at the position r(0.5) . The
average velocity vectors approach v as a limit as t:.t approaches zero.
-,
4.38 .. i
. - _.- - - - - - - - - - - - ._- ..,.
''
'
Figure l 0-71
e. Speed = lv(0.5) 1
---- - ----
= ~ (4.38 ... )2 + (- 4.20 ... )2
= 6.079 .. .
Thus the object is traveling at about 6.1 ft/sec.
Example 2 shows how to find the acceleration vector and how it helps interpret the
motion .
Example 2 Given the vector equation
r(t) = (Ssint)i + (5cos 2 t)]
for the position of a moving object, as in Example 1, do the following :
a. Write vector equations for the velocity vector, v(t), and the acceleration vector,
a(t ) .
b . On a graph of the path of the object, sketch the position vector r(4) at time t = 4.
From the head of r(4), sketch the vectors v(4) and a(4).
c. How fast is the object going at time t = 4?
d. Compute Zit(4), the tangential component of the acceleration (parallel to the
path) .
e. Is the object speeding up or slowing down at t = 4? How are you able to tell? At
what rate is it speeding up or slowing down?
f. Compute Zin(4), the normal component of acceleration (perpendicular to the
path) . Toward which side of the path does Zin(4) point? What effect does this
component have on the motion of the object?
538 Chap
ter 10: The Calc
ulus of Motion-Ave rages,Extremes,
andVec
tors
Solutions a. The velocity is found as in Example 1. The acceleration is found by differen-
tiating the velocity vector.
v(t ) = r '(t ) = (5cost)7 + (- lOco s tsint )]
From trigonometry,
= (5 cost )7 + (-5 sin2t)] 2 sint cos t = sin 2t.
a(t) = v '(t ) = (-5sint )7 + (- 10 cos2t )]
b. Substituting 4 for t gives
r(4l = -3 .78 ... 7 + 2.13 . . . ]
v (4l = -3 .26 . . . 7 - 4.94 ... ]
a (4) = 3.78 ... 7 + 1.45 .. . ]
Figure 10-7m shows v (4) and ci(4) drawn on a graph plotted as in Example 1.
The vectors are drawn by counting spac es on th e dot paper. Note that
although v (4) is tangent to th e path, th e acceleration is at an angle to
the path.
a(4>
Figure l 0-7m
c. Speed = lv(4 ) I
= ~-(5_ c_o_s_4_)_2 _+_(___
5_s_in_8_)_
2
= 5.928 . ..
Thus th e object is going about 5.93 ft/ sec.
d. The tangential acceleration, a 1 (4), parallel to the path, is th e vector projec-
tion of a(4) on v(4 ) (Figure 10-7n). First find th e scalar projection, P.
v(4l
p = a (4 ) . v (4) (-5 sin4) (5 cos 4 ) + (- 10 cos 8) (-5 sin8 ) = -3 2998
Figure l 0-7n lv (4) I 5.928... .
Section
10-7:Vector
Functions
forMotion
ina Plane 539
To find a vector in the direction of v(4) with length 3.2998 . . . , pointed the
other way, multiply the scalar projection, -3 .2998 .. . by a unit vector in the
direction of v(4) .
_ v(4 )
at( 4 ) = P lv(4) I
Zit(4) = (-3.2998 . . .) (5cos4~'.9 ; i ~.~sin8)]
Zi1(4) = 1.818 ... 7 + 2.753 .. . ]
e. Since the tangential acceleration points in the direction opposite to v(4), the
object is slowing down when t = 4. Both the obtuse angle between v(4) and
a(4) (Figure 10-7n) and the negative value of P, the scalar projection of a(4)
on v(4), reveal this fact. The rate at which the object is slowing equals the
magnitude of Zit(4), namely, IP I, or 3.2998 .. . . So the object is slowing down
at about 3.3 (ft/sec) / sec when t = 4.
f. As shown in Figure 10-7n, the normal acceleration, Zin(4), perpendicular to
the path, equals the vector that, when added to a 1 (4 ) , gives a (4). Therefore,
Zin(4) = a(4) - Zit (4) = (3.78 ... 7 + 1.45 . . . ]) - (l.818 ... 7 + 2.753 . . . ] )
= 1.96 ... 7- 1.29 . . . ].
The normal component of acceleration will always point toward the concave
side of the graph, as shown in Figure 10-7n. It is responsible for pulling the
object out of a straight-line path and into a curved path .
From Example 2, you can reach the following conclusions about acceleration vectors.
Property:Componentsof the AccelerationVector
The accelera tion point s toward the concave side of the p ath of the object.
The scalar proj ection, P, of a on v tells the magnitu de of the tangential component of
acceleration, Zit,which acts along the path.
IPI is equal to th e rate of change of speed of the object.
If P > 0, or the angle between Ziand v is acute, the object is speeding up.
If P < 0, or th e angle be tween a and v is obtuse, the object is slowing down.
The normal component of acceleratio n, Zin,changes the direct ion of motion.
Since a = a1 + a11, the normal component can b e foun d by Zin= a - Zit.
In Exampl e 3, you will find the distance traveled by an object moving in a curved
path.
Example 3 For the object in Example 1, r(t) = (5 sin t) 7 + (5 cos 2 t)] . Find the distance traveled
by the object in the time interval [O,4].
Solution Recognize that a vector equation is the same as two parametric equations . The
distance the object travels thus equa ls the arc length of a parametric curve
(Section 8-7).
540 Chapter
10: TheCalculus
ofMotion-Averages,
Extremes,
andVectors
dL = ~x'(t) 2 + y'(t) 2 dt = ~(5 cos t) 2 + (-10 cost sin t)2 dt
L= f ~( 25 + 100sin 2t)cos 2tdt
0
4
= 19.7245 . . . "" 19.72 ft By num erica l integration
Check: A circle of radius 5 approximates the curve. A 4-radia n arc of th e circle
has length (4 / 2rr) (2rr 5) = 20, which compares well with 19.72.
Problem Set 1 0-7
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes .
Ql. Integrate: f x sinx
dx
2 3
Q2. Differentiate: x e x
Q3. Integrate : f 2x dx
52001
Q4. Evaluate :
51998
QS. Find f x 3 dx if the integral equals 11 when x = 2.
Q6. Find dy / dx if x = e 3t and y = tan6t .
Ql. The function in Problem Q6 is called a(n) -?- function .
QB.Integrate: flnx dx
Q9. In polar coordinates, the graph of r = 0 is a(n) -?- .
y
Q10. Simplify: e 21nx
l. Parabolic Path Problem I: An object moves along
4
the parabolic path (Figure 10-70)
r(t) = (10sin0.6t)7 + (4cosl.2t)],
X
where distance is in feet and time is in seconds. - 10 . 10
a. Find equations for v(t) and a(t).
b . Calculate ?(0. 5), v(0.5), and a(0.5). On
a photocopy of Figure 10-70, plot ? as a
position vector, and plot v and a with their
tails at the head of?. Explain why the three
vectors are reasonable . Figure 10-70
c. Based on the graphs of the vectors in lb, does the object seem to be speed ing up
or slowing down at time t = 0.5 sec? How do you tell?
d. Verify your answer to le by finding the tangential and normal components of
a(0.5). Sketch these components on the diagram in lb starting at the tail of a.
e. At what rate is the object speeding up or slowing down at t = 0.5?
f. Calculate ?(7), v(7), and a(7) . Sketch these vectors on the diagram of lb . At tim e
t = 7, do es the object seem to be speeding up or slowing down?
Section 10-7:Vec
torFunctionsforMotion in a Plan
e 541
g. Show that at time t = 0 the acceleration
vector is perpendicular to the path. How do y
you interpret this fact in terms of motion of
the object at t = O?
2. Parabolic Path Problem II: An object moves
along th e parabolic path (Figure 10-?p),
r(t) = (8cos0.St)7 + (6sin0.4t)],
where distan ce is in feet and time is in seconds.
a. Find equations for v(t) and a(t).
b. Calculate r(l), v(l), and a(l). On a photo-
copy of Figure 10-?p, plot r as a position
vector, and plot v and a with their tails at
the head of r. Explain why the three vectors Figure 10-7p
are reasonable .
c. Based on the graphs of the vectors in 2b, does the object seem to be speeding up
or slowing down at time t = 1 sec? How do you tell?
d. Verify your answer to 2c by finding the tangential and normal components of a(1).
Sketch these components on the diagram in 2b starting at the tail of a.
e. At what rate is the object speeding up or slavving down at t = l?
f. Calculate r(l0.5), v(l0.5), and a(l0.5). Sketch these vectors on the diagram of 2b.
At time t = 10.5, does the object seem to be speeding up or slowing down?
g. What is the first positive value of t at which the object is stopped? What is the
acceleration vector at that time? Plot this vector on the diagram in 2b. Surprising?
3. Elliptical Path Problem: An object moves along
y
the elliptic al path (Figure 10-7q),
r(t) = (10 cos it) 7 + (6 sin %tJ],
where t is in seconds and distances are in feet.
a. On your grapher, plot the path followed by
the heads of the velocity vectors if their tails
are placed at the heads of the respective
position vectors . You can use parametric
mode to plot r + v. Sketch the result on a
photocopy of Figure 10-7q.
b. Sketch the velocity vectors for each integer ~6 ~ .
value oft from O through 12. You may find
their beginnings by tracing the r vector and Figure 10-7 q
their ends by tracing r + v.
c. Prove that the heads of the velocity vectors in 3b lie along an ellips e.
d. On your grapher, plot the path followed by the heads of the acceleration vectors
when their tails are placed at the heads of the res pective position vectors. Sketch
the result on your diagram from 3a.
e. Sketch the acceleration vectors as you did for the velocity vectors in 3b . What seems
to be true about the direction of each of these acceleration vectors?
542 Chapter
l 0: TheCalculus
of Motion-Averages,
Extremes,andVectors
4. Spiral Path Problem: An object moves on the
spiral path (Figure 10-7r), y
r(t) = (0.5tcost)i + (0.5tsint)],
S
where distance is in miles and time is in hours.
a. Find vector equations for v(t) and a(t).
b. Find the position, velocity, and acceleration X
at t = 8.5 and at t = 12. On a photocopy - 10 . . 10
of Figure 10-7r, plot r(8 .5) and T-(12),
thus showing that these vectors really do
terminate on the path. -5
c. Plot v(8.5) and a(8 .5) starting at the head
of r(8.5) . Do the same for v(12) and a(12).
Explain why these velocity vectors have the figure 10-7r
proper relationships to the path.
d. At t = 8.5 and at t = 12, is the object speeding up or slowing down? How do you
tell?
e. Find the tangential and normal components of the acceleration vector at t = 12.
Show these components on the diagram from 4b .
f. How fast is the object going at t = 12 hr ? At what rate is the speed changing then?
g. On your grapher plot the path followed by the heads of the acceleration vectors
when they are placed with their tails at the heads of the respective position vectors.
What geometrical figure does the path appear to be? Prove algebraically that your
conjecture is correct.
5. Parabo lic Path Problem III: An object moves along the parabola y = x2 . At various
times, t, the object is at various points (x, y), where x and y are in centimeters and t
is in seconds.
a. Write the position vector r(x) as a function of x alone (and the two unit vectors 7
and], of course!). Then find the velocity vector v(x) as a function of x and dx / dt.
b. Assume that the object moves in such a way that x decreases at a constant rate of
3 cm/ sec. Find v(2) . How fast is the object moving when x = 27
c. Sketch the graph of the parabola and draw r(2) and v(2) at the point (2, 4). Explain
why the graph of v(2) is reasonable.
d. Find the acceleration vector, a(x), and evaluate a(2) . Sketch a(2) on your graph .
e. Find the tangential and normal components of acceleration at x = 2. Show these
components on your graph . Based on the graphs, why are your answers reasonable?
f. When x = 2, is the object speeding up or slowing down? Justify your answer .
g. The object changes its motion and goes in such a way that its speed along its curved
path is 5 units per minute. Write an expression in terms of x for dL, the differential
of arc length along the curve . Find dx / dt when x = 2.
Sectionl 0-7: Vector
Functions
forMotionina Plane 543
6. Velocity Vector Limit Problem: An object moves along one
petal of the four-leafed rose shown in Figure 10-7s. y
r= (12sintcos0 .5t)7 + (12sintsin0.5t)]
On a photocopy of this diagram, plot r (1). From the end of
r (l),
plot the average velocity difference quotient vectors, 10
_( ) r(t) - r(l)
qt= ----
t- l
fort = 2, t = 1.5, and t = l.l. Then plot the velocity vector,
v(l ) . How does the velocity vector relate to the path of the
object and to the average velocity vectors ?
7. Find the distance traveled by the object in Problem 1 from X
time t = 0 to t = 2. 10
8. Find the distance traveled by the object in Problem 3 in one Figure 10-7s
complete cycle.
9. Baseball Problem: Sol Teen pitches a baseball. As it leaves his hand, it moves horizon-
tally toward home plate at 130 ft/ sec (about 90 mi/ hr), so its velocity vector at time
t = O sec is
v(O) = - 1307 + o].
The minus sign is used since the distance from the plate is decreasing. As the ball
moves, it drops vertically due to the acceleration of gravity. The vertical acceleration
is 32 (ft/ sec)/ sec. Assuming that there is no loss of speed due to air resistance, its
acceleration vector is
a(t) = Oi- 32].
a. Write an equation for v(t), the velocity as a function of time.
b. When Sol releases the ball at time t = 0 sec, it is at y = 8 ft above the playing field
and x = 60.5 ft from the plate. Write the position vector, r(t) = x(t)i + y(t)].
c. Sol's sister, Phyllis, stands at the plate ready to hit the ball . How long does it take
the ball to reach the plat e, at x = O? As it passes over the plate, will it be in Phyllis's
strike zone, between y = 1.5 ft and y = 4.5 ft above the plate? Show how you reach
your conclusion .
d. Sol pitches another time. Phyllis hits the ball, making it leave a point 3 ft above the
plate at an angle of 15 to the horizontal, going at a speed of 200 ft/ sec. What are
the initial horizontal and vertical velocities? Assuming that the horizontal velocity
stays constant and the vertical velocity is affected by gravity as above, write the
position vector r(t) as a function of the number of seconds since she hit the ball .
e. Will Phyllis make a home run with the hit in 9d? The fence for which it is heading
is 400 ft from the plate, and 10 ft high . You may do this by plotting the position
graph in parametric mode and seeing where the ball is when x = 400.
544 Chapter
l 0: TheCalculus
of Motion
- Averages,
Extremes,
andVecto
rs
10. Sinusoidal Path Problem: An object moves along the graph of y
y = sinx (Figure 10-7t), where x and y are in meters and t is
minutes. Its x-acceleration is 3 m/ min 2 , a constant. At time t = 0,
X
the object is at the point (0, 0) and has velocity equal to the zero
vector.
a. Find r(t), the position vector as a function of time.
b. How fast is the object moving when it is at the point (6, sin 6)?
Figure l 0-71
11. Figure Skating Problem: One figure roller skaters do in com-
petition has a large loop and a small loop (Figure 10-7u). y
Specifications by the Roller Skating Rink Operations Association
of America require the outer loop to be 240 cm from the origin
where the loops cross and the inner loop to be 60 cm from
X
the origin. The figure closely resembles a limaron with polar
equation
d =a+ b cost,
where dis the directed distance from the origin at angle t radians
and a and b are constants. Figure l 0-7u
a. At t = 0, d = 240. At t = rr, d = - 60 so that the point is
on the positive x-axis . Find the particular equation of this
limac;:on.
b. Suppose that Annie Lips skates with an angular velocity of
1 rad / sec . So t is also her time in seconds, and her position
vector is
r(t) = (dcost)i + (dsint)],
where d is given by the equation in l la. Find Annie's velocity
vector at time t = l. How fast is she going at that time ?
C. Find Annie's acceleration vector at t = l. Find the tangential
and normal components of the acceleration. Is she speeding
up or slowing down at this time? At what rate?
12. River Bend Problem: A river meanders slowly across the plains
(Figure 10-7v). A log floating on the river has position vector
r(t) = (0.St + sint)7 + (4cos0.St)],
where distances are in miles and t is in hours.
a. When t = 14, what are the log's velocity and acceleration
vectors? How fast is it going? What are its tangential and
normal acceleration vectors? Is it speeding up or slowing 4 y
down? At what rate ?
b. How far does the log move along its curved path from t = 0
tot = 14? What is its average speed for this time interval? X
Figure l 0-7v
Section l 0-7: Vector
Functions
forMotion
ina Plane 545
13. Roller Coaster Problem: Assume that a roller coaster track is a prolate cycloid (Figure
10-7w) and that the position of a car on the track (in feet) at time t seconds is
r (t) = (5t - 12 sin t) 7 + ( 15 + 12 cos t)].
a. Write the velocity and acceleration vectors as functions oft.
b. Find the velocity and acceleration vectors at the point shown, where t = 2.5 sec. Plot
these vectors on a photocopy of Figure 10-7w, starting at the point on the graph .
r:-
Figure l 0-7w
c. Find the tangential and normal components of the acceleration vector at t = 2.5
sec.
d . Analyze the motion of the roller coaster at t = 2.5 sec. For instance, how can you
tell that the velocity vector is reasonable? Is the normal component of acceleration
reasonable? How fast is the roller coaster going? Is it speeding up or slowing down?
e. Show that the acceleration vector is straight down when the roller coaster is at a
high point, and straight up when it is at a low point.
f. How long is the track from one high point to the next?
14. Dot Product Problem: The dot product of two vectors is defined to be
v1 V2 = lvi llv 2l cose .
If v 1 = X i 7 + Yi) and v 2 = X 2 7 + Y 2], show as in Figure 10-7x that
V1 V2 = X1X2 + Y1Y2,
Assume the distributive property, a (b + c) = a b +a c. You will need to figure
out what 7 7,] ],and 7 ] equal, based on the angles between these unit vectors.
Figure l 0-7x
546 Chapter
l 0: TheCalculus
ofMotion
- Averages,
Extremes,
andVectors
15. Three-Dimensional Vector Problem: A three-dimensional z
vector (Figure 10-?y) can be resolved into three mutually
perpendicular components. If 7, ] , and k are unit vectors in
the x-, y-, and z-directions, respectively, then position vector
r from the origin to the point (x, y, z) can be written
' -->
r =xi + y] + zk. : zk
y
Suppose that a moving object's position is given by the vector
function 1,.:,
' J' :
r(t) = (10sin0 .8t )7 + (10cos0 .6t)] + (6t 0 5 )k.. ~/--c.- - - - - - j
Find the velocity and acceleration vectors at time t = 1. At /X YJ
that time, is the object speeding up or slowing down? Justify Figure 10-7y
your answer.
16. Curvature Project: Figure 10-7z shows an object moving with y
...,
velocity v along a path. The velocity makes an angle cp (Greek letter V
"phi") with the positive x-axis. The curvature of the path is defined
to be the rate of change of cp with respect to distance, s, along the
path. That is, K (lower case Greek letter "kappa") is given by
K = dcp/ ds.
a. Explain why dcp/ ds = (dcp/ dt)(dt / ds). X
b. Explain why tancp = dy / dx, which equals (dy / dt) / (dx / dt ).
c. Let x' and x" be the first and second derivatives of x with respect Figure 10-7z
to t, and similarly for y. Show that the following is true.
Technique:
Calculationof Curvature
d<f>= x' y " - x"y '
ds lvi3
where the derivatives are taken with respect to t.
d. Figure 10-7aa shows the ellipse
X = 5 COS t
y = 3 sin t.
Show that the maximum curvature is at each end of the major
axis.
e. Show that the curvature of a circle,
y
X = r COS t 3
y = r sint
X
is constant. 5
f. Show that the curvature of this line is zero.
x=5cos 2 t
y = 3 sin 2 t Figure 10-7 aa
Section10-7:Vector
Functions
forMotion
ina Plane 547
g. The radius of curvature is defined to be the reciprocal of the curvature. Find the
radius of curvature of the ellipse in 15d at the right-most vertex, (5, 0).
h. On your graph er, plot the ellips e in Figure 10-7aa . Then plot a circle tangent to the
ellipse at (5, 0), on the concave side of the ellipse, with radius equal to the radius
of curvat ur e. Sketch the result. This circle is called the osculating circle ("kissing"
circle). Appropriately, it is the circle that best fits the curve at the point of tangency .
10-8 Chapter Review and Test
In this chapter you have studied applications of calculus to motion. You distin-
guished between distance traveled by a moving object and displacement from its
starting point. You made precise the concept of average velocity and extended it
to other average values. ext, you learned how to calculate the rate of change of a
variable quantity from a related rate . You extended your study of maximum and
minimum values to problems involving motion, and then to other similar problems.
Finally, you applied the concepts to objects moving in a plane, using vectors as a
tool.
The Review Problems are numbered according to the sections of this chapter. The
Concepts Problems allow you to apply your knowledge to new situations . The
Chapter Test is typical of a classroom test.
Review Problems
RO. Updat e your journal with the things you have learned since the last entry . You should
include such things as those listed here.
The one most important thing you have learned in studying Chapter 10
Which boxes you have been working on in the "define, understand, do, apply" table.
The distinction among displacement, velocity, and acceleration
How to find average rates
How to us e rates of change to find extreme values of functions
How to find a rate that is related to another rate
How to analyze motion of objects moving in two dimensions
Any techniques or ideas about calculus that are still unclear
Rl. Popeye and Olive Problem : Olive Oyl is on a conveyor belt moving 3 ft/sec toward
the sawmil l. At time t = 0, Popeye rescues her and starts running the other direction
along the conveyor belt. His velocity with respect to th e ground, v ft/ sec, is given by
V = .If - 3.
When does Popeye's velocity become positive? How far have he and Olive moved
toward the sawmill at this time? What is their net displacement from the starting
point at t = 25? What total distance did they go from t = 0 tot = 25?
548 Chapter
l 0: TheCalculus
ofMotion-Averages,
Extremes,
andVectors
'Mt$ [i #@$
R2. a. The velocity of a moving object is given by v(t) = 21 - 8 cm/ min.
i. Graph velocity versus time. Sketch the result.
ii. Find the net displacement between t = 1 and t = 4.
iii. Find the total distance traveled between t = 1 and t = 4.
b. Acceleration Data Problem: An object initially going 30 ft/ sec has the following
accelerations in (feet per second) per second measured at 5-sec intervals .
time acceleration
0 2
5 8
10 1
15 0
20 - 10
25 -2 0
Find the estimated velocities at the ends of the time intervals . For each entry in
the table, tell whether the object was speeding up, slowing down, or neither at
that instant.
R3. a. Average Velocity Problem: An object moves with velocity v(t) = sin (rrt / 6). Find
the average velocity on the time interval:
i. [0, 3]
ii. [3, 9]
iii . [O, 12]
b. Average Value Problem: For the function f(x) = 6x 2 - x 3 ,
i. Find the average value of f(x) on the interval between the two x-intercepts.
ii. Sketch a graph showing the geometrical significance of this average value .
iii. Show that the average value is not equal to th e average of the two values of
f(x) at the end points of the interval.
R4. Rover's Tablecloth Problem: Rover grabs the tablecloth and
starts backing away at 20 cm/ sec. A glass near the other
end of the tablecloth (Figure 10-8a) moves toward the edge
and finally falls off. The table is 80 cm high, Rover's mouth
is 10 cm above the floor, and 200 cm of tablecloth separate
Rover's mouth from the glass. At the instant the glass
reached the table's edge, was it going faster or slower than Figure l 0-Ba
Rovers 20 cm/ sec? By how much ?
RS. a. Campus Cut-Across Problem : Juana Cross makes daily Math
Grass buildin g
trips from the Math Building to the English Building. She Start
has three possible routes (Figure 10-8b): Stop
Along sidewalk all the way Sidewalk
Straight across the grass ,___ -
700 ft I
Angle across to the other sidewalk English building
She figures her speed is 6.2 ft/ sec on the sidewalk and
5.7 ft/ sec across the grass. Which route takes the least Figure l 0-Bb
time? Explain.
Section10-8:Chapter
Review
andTest 549
b. Resort Island Causeway Problem : Moe Tell ovms a resort on
the beach. He purchases an island 6 km offshore, 10 km down Land - 6 km ----l
the beach (Figure 10-8c). So that his guests may drive to the
island, he plans to build a causeway from the island to the
beach, connecting to a road along the beach to the hotel. The
road will cost $5 thousand per kilometer, and the bridge will
cost $13 thousand per kilometer . What is the minimum cost 10km
for the road and causeway system? How much money is saved
by using the optimum path over what it would cost to build a
causeway from the hotel straight to the island?
R6. a. An object's acceleration is given by a(t)
[O, 10]. Find the following:
= 6t - t 2 in the interval
1 Resort
i. The maximum and minimum accelerations fort in (0, 10] Figure l 0-8c
ii. The maximum and minimum velocities fort in [O, 10],
assuming v(O) = 0
iii. The maximum and minimum displacements from the starting point for t in
[O, 10]
b. Inflation Problem: Saul T. O'Tile lives in a third-world country where the inflation
rate is very high. Saul saves at a rate of 50 pillars (the currency in his country) a
day. But the value of money is decreasing exponentially with time in such a way
that at the end of 200 days a pillar will purchase only half of what it would at the
beginning.
i. Write an equation for the purchasing power of the money
Saul has saved as a function of the number of days since
he started saving. y
ii. On what day will Saul's accumulated savings have the
maximum total purchasing power? Justify your answer. 5
R7. a. Make a sketch showing how the velocity and acceleration
vectors are related to each other and to the curved path x
of an object moving in a plane if
i. The object is speeding up
ii. The object is slowing down .
b. An object moves along the hyperbola shown in Figure
10-8d . The position vector at any time t minutes is given
by
r = (5 cosh t) 7 + (3 sinh t)]. Figure l 0-8d
i. Find the position, velocity, and acceleration vectors for the object at time t = l.
ii. Draw these vectors at appropriate places on a photocopy of the object's path.
iii. At time t = 1, how fast is the object moving? Is it speeding up or slowing
down? At what rate ?
iv. How far does the object move between t = 0 and t = 1?
v. Show that if the tails of the velocity vectors are placed at the respective points
on the path, then their heads lie on one of the asymptotes of the hyperbola.
550 Chapter
l 0: TheCalculus
ofMotion
- Averages,
Extremes,
andVec
tors
Concepts Problems
Cl. One-Problem Test on Linear Motion and Other Concepts: A particle moves up and
down the y-axis with velocity v feet per second given by
V = t3 - 7t 2 + 15t - 9
during the time interval [O,4]. At time t = 0, its position is y = 4.
a. Sketch the velocity-time graph.
b. At what time(s) is the particle stopped7
c. At what time is the velocity the maximum? The minimum? Justify your answ ers .
d. At what time(s) does the velocity-time graph have a point of inflection?
e. What is happening to the particle at the point(s) of inflection?
f. Find the position, y, as a function of time.
g. Sketch the position-time graph.
h. At what time is y the maximum? The minimum? Justify your answers.
i. At what time(s) does the position-time graph have a point of inflection?
j. What is happening to the particle at the point(s) of inflection?
k. Is y ever negative? Explain.
l. What is the net displacement of the particle from t = 0 tot= 4 7
m. How far does the particle travel from t = 0 to t = 4?
n. What is the average velocity from t = 0 to t = 4?
o. What is the average speed from t = 0 to t = 4?
C2. New York to Los Angeles Problem : What is the shortest time in which a person could
possibly get from New York to Los Angeles? If you ignore such things as getting to
and from airports, the kind of vehicle to be used, and so forth, the problem reduces
to how much stress the human body can take from acceleration and deceleration
("g forces"). Recall that an acceleration of lg is the same as the acceleration due to
gravity.
a. From what you have heard in the media or elsewhere, about how many g can the
human body withstand?
b. About how far is it from New York to Los Angeles7
c. You must be stopped both at the beginning of the trip and at the end .
What, then, is the minimum length of tim e a human being could
take to get from New York to Los Angeles?
C3. Spider and Clock Problem: A spider sitting on a clock face attaches
one end of its web at the "12," 25 cm from the center of the clock.
As the second hand passes by, she jumps onto it and starts crawling
toward the center at a rate of 0. 7 cm/ sec (Figure 10-8e). As the clock
turns, the spider spins more web. The length of this web depends
on the number of seconds the spider has been crawling and can be
calculated using the law of cosines. Find the rate of change of this Figure l 0-Be
length at the instant the spider has been crawling for 10 sec.
Section10-8:Chapter
ReviewandTest 551
C4. Submerg ing Cone Prob lem: A cone of base radius 5 cm and altitude 12 cm is
being lowered at 2 cm/min, vertex down, into a cylinder of radius 7 cm that
has water 15 cm deep in it (Figure 10-8f). As the cone dips into the water, the
~
water level in the cylinder rises. Find the rate at which the level is rising when
the vertex of the cone is
a. 10 cm from the bottom of the cylinder
--
~-
\J ~.,.
b. 1 cm from the bottom of the cylinder
CS. The Horse Race Theorem: Sir Vey and Sir Mount run a horse race. They start at
time t = a at the same point. At the end of the race, time t = b, they are tied. Let
f(t) be Sir Vey's distance from the start, and g(t) be Sir Mount's distance from
the start. Assuming that f and g are differentiable, prove that there was a time Figure l 0-8f
t = c between a and b at which each knight was going exactly the same speed .
C6. Hem ispherica l Railroad Prob lem: A mountain has the shape of a
perfect hemisphere with a base radius of 1000 ft (unlikely in the Top
real wor ld, but it makes an interesting problem!). A railroad track
is to be built to the top of the mountain. Since the train can't go
straight up, the track must spiral around the mountain (Figure ~ ~ ....
10-8g). The steeper the track spirals, the shorter it will be, but the
slower the train will go. Suppose that the velocity of the train is
given by
v = 30 - 60sin0,
where vis in feet per second and e is the (constant) angle the track Figure l 0-8g
makes with the horizontal. If the track is built in the optimum
way, what is the minimum length of time the train could take to
get to the top?
Chapter Test
Tl. Truck Passing Problem : You accelerate your car to pass a truck, giving it a velocity
V = .Jt+ 60,
where v is velocity in feet per second and t is time in seconds since you started to
pass. Find out how far you go in the 2 5 sec it takes you to get around the truck.
T2. Power L ine Prob lem: Ima Hunter builds a camp house in the country that she wants
to supply with electricity. The house is 3 mi from the road. The electrical contractor
tells her it will cost $2520 ($360 a mile) to run the power line 7 mi along the highway
to the point nearest the camp house and $2400 ($800 a mile) more to run it the 3
mi from the highway to the camp house. You believe that Ima could save money by
making the line cut off from the highway before the 7-mi point and angle across to
the house. How should the power line be run to minimize its total cost7 How much
could Ima save over the $4920 the contractor proposes?
T3. For the function f(x) = x 3 - 4x + 5, find the maximum, the minimum, and the average
value of the function on the interval [1, 3]. Sketch a graph showing the geometrical
significance of the average value .
T4. An object moving along a line has velocity v(t) = 10(0.5 - 2- 1 ) ft/ sec.
552 Chapter
l 0: TheCalculusofMotion-Averages,
Extremes,andVectors
a. Find the distance it travels and its net displacement from the starting point for
the time interval (0, 2].
b. Find its acceleration at time t = 0.
c. At time t = 0, was the particle speeding up or slowing down? Justify your answer.
TS. An object is moving at 50 cm/sec at time t = 0. It has
accelerations of 4, 6, 10, and 13 (cm/sec)/sec at times
t = 0, 7, 14, and 21 sec, respectively. Approximately what
was the object's average velocity for the 21-sec time . ... I . , .- ..
interval? About how far did the object go7
Figure 10-8h shows the path of an object moving with vector
equation
r(t) = (10cos0.4t)i + (10sin0.6t)]
where distance is in miles and time is in hours.
T6. Find an equation for v(t).
X
T7. Find an equation for a(t). 5 10
T8. Find r(2) . Make a photocopy of the graph in Figure 10-8h Figure l 0-Sh
and draw r(2) on it.
T9. Find v(2). On the photocopy, plot this vector starting at the object's position when
t = 2. How is this vector related to the path of the object?
TlO. Find a(2) . On the photocopy, plot this vector starting at the object's position when
t = 2.
Tll. Sketch the components of a(2), one of them directed tangentially to the path, and
the other normal to it.
Tl 2. Based on the components of a(2), would you expect that the object is slowing down
or speeding up when t = 2? How can you tell?
Tl3 . At what rate is the object speeding up or slowing down when t = 2?
Tl4. Explain why the normal component of a(2) is pointing toward the concave side of
the path.
Tl 5. Find the distance the object travels between t = 0 and t = 2.
Section l 0-8:Chapter
Review
andTest 553
CHAPTER
11 The Calculus of
Variable-Factor Products
I
II
-
I-beams used in construction must be stiff so that they do not bend
too much. The stiffness depends on the shape of the beam's cross-
section. Stiffness is measured by the second moment of area of the
cross-section, which is defined to be area times the square of the dis-
tance from the centroid of the cross-section. Since different parts of
the cross-section are at different distances from the centroid, definite
integrals are used to compute the stiffness of a given type of beam.
555
lbl Mathematical Overview
A definite integral gives a way to find the product of x and y, where y
varies. In Chapter 11 you will appl y integrals to
work = force x displacement
force = pressure x area
mass = density x volume
moment = displacement x quantity
You will do the applications in four ways.
Graphically The logo at the top of each even- Force
numbered page of this chapter
reminds you that work equals
force times displacement.
Numerically force disp. work Displacement
50 10 0 a b
53 12 103
58 14 214
70 16 342
90 18 502
Algebraically My = t
a
x dA, the definition of moment of area.
Verbally We calculated the balance point of a piece of cardboard by finding
its centroid. Then we showed that we were right by cutting out the
cardboard. It actually did balance on a pencil point placed at the
calculated centroid!
556
11-1 Review of Work-Force Times Displacement
In previous chapters you have run across the fact that the work done in moving an
object from one place to another equals the force with which it is pushed or pulled
Force, 11 lb ]Fl times the displacement through which it moves. For instance, if you pu sh a chair 7
hr1 feet across the floor with a force of 11 pounds (Figure 11-la) you do 77 foot -pound s
fN I~ Displa~e;~
r' of work.
In Prob lem Set 11-1, you will refresh your memory about how to comp ut e the work
; ~
done if the force is variable. As you study this chapter you will see how the thought
process you use for this one application can be used for many others . You will learn
Figure 11-1a such things as how to find the ba lance point of a solid object, and how to calcu late
volumes and masses of objects quickly, without actually doing any integration.
OBJECTIVE By your self or with your study group, find the work done in moving a chair across the
floor if the force you exert on it varies as you pu sh.
Exploratory Problem Set 1 1 -1
Chair Wor k Problem: Suppose that you push a chair across the floor with a force
F = 20xe - o.sx,
where F is the force in pounds and x is the distance in feet the chair has
moved since you started pushing. F
l. Figure 11-1b shows the graph of F . On a photocopy of this graph , draw a
narrow vertical strip of width 6.x = 0.2 centered at x = 4. Approxima tely
what is the force at any value of x in this strip? Approximate ly how
much work is done in moving the chair a distance 6.x at x = 4?
2. Write an equation for dW, the work done in moving the chair a distance
of dx feet when the chair is at point x, where the force is given by the Figure 11-1b
equation above .
3. Add up all the dW's from x = 0 to x = 7. That is, find the definite integral of dW.
4. How much work was done in moving the chair from x = 0 to x = 7?
5. If you continue to push the chair with the force shown and it continues to move, what
limit would the amount of work approach as x approaches infinity?
11-2 Work Done by a Variable Force
In ordinary English the word work is used in different contexts an d with different
meanings . For instance, you may feel that you did a lot of work on your calculu s
assignment last night. Physically, however, the word work has a precise definition .
Section 11-1: Reviewof Work-F orceTimesDisplacement 557
It is the product of the force applied to an object and the displacement the object
moves as a r esult of that force . For instance, if you push a chair 7 ft across the
floor by exerting a force of 11 lb, you have done 77 ft-lb of work, as you saw in
Section 11-1.
Definition:Work
If an object moves a certain displacement as a result of being acted upon by a certain
force, then the amount of work done is given by
Work = (force)(displacement).
In most real-world situations, the force does not remain constant as the object
moves. By now you should realize that finding the work under these conditions is a
job for definite integration!
OBJECTIVE Given a situation in which a varying force acts on an object, or where different parts of
the object move through different displacements, be able to calculate the amount of work
done.
There are two ways to analyze a work problem .
1. Move the whole object a small part of the displacement .
2. Move a small part of the object the whole displacement.
The following two examples show how this analysis can be done.
Example 1 Move the whole object through a small displacement. A ship is at anchor in 80 ft
of water. The anchor weighs 5000 lb, and the chain weighs 20 lb/ ft (Figure l l-2a).
The anchor is to be pulled up as the ship gets under way.
a. How much force must be exerted to lift th e anchor as it comes aboard the ship?
Write an equation expressing force in terms of the displacement, y, of the anchor
r
80-y from the bottom of the ocean .
l so ft
b. How much work must be done to raise the anchor the 80 ft from the bottom to
the point where it comes aboard the ship?
Anchor
T
y
!
Figure 11-20
Solutions a. When the anchor is at the bottom and neglecting buoyancy, you must pull
with enough force to lift the 5000-lb anchor and the 80 ft of chain. Letting F
stand for force,
F = (20)(80) + 5000 = 6600 lb.
558 Chapter
11: TheCalculus
ofVariable-Factor
Products
i Mft&ifiti
"-W#ti
J iii &
At the ship, the only force is that needed to lift the 5000-lb anchor .
F = 5000 lb
In betw een , the force varies linearl y with the length of the chain. If y is the
displacement from the bottom to the anchor, then this length is equal to
(80 - y). Therefore,
F = 20(80 - y) + 5000
= 6600 - 20y.
b . If the anchor is rais ed a small displ acement, dy, the force would be essentially
constant, the same as at some sampl e point in that particular subinterva l. So
the work, dW, don e in lifting the anchor this sma ll displacement would be
dW = Fdy
= (6600 - 20y) dy.
The total amount of work, W, can be fou nd by adding up the dW's, then
taking the limit as dy goes to zero . You should recognize by no w that this
pro cess is definit e integration.
rso
W = Jo (6600 - 20y) dy
80
= 660 0y - 1Oy2 I O You could inte grate numerica lly.
= 464,000 ft-lb
Example 2 Move a small part of the object the whole displacement. A conical tank has a top
diam eter of 10 ft and an altitude of 15 ft (Figure ll -2b). It is filled to the top with
liquid of density k lb/ft 3 . A pump takes suction from the bottom of the tank and
pumps the liquid up to a level 8 ft above the top of the tank. Find the total amount
of work don e.
t
l--10
ft -I 8 ft
i
(x,y)
1
15 ft
1 X
Pump
Figure l l -2b
Solutions The first thing to realize is that the amount of work done liftin g any small
volume of liquid is ind ependent of the path the liquid takes. It depends only
on how far the volume is displaced upward from where it starts to where it
finishes. Liquid at a sample point, (x, y) travels down through th e tank, through
Sec
tion 11-2:WorkDone bya Variable
Force 559
the pump, and up to the discharge 8 ft above the top of the tank (and thus 23 ft
above the bottom of the tank, where y = 0). So its net displacement is equal to
(23 - y) .
If you "slice" the liquid horizontally, liquid at each point in the slice will be
displaced essentially the same distance as that at the sample point. The force,
dF, needed to lift the water the displacement (23 - y) is equa l to the weight of
the slice, namely, k dV . Letting W stand for amount of work, the work done in
lifting one slice is
dW = (kdV)(23 - y)
2
= k TTx (23 - y ) dy . Subst itute rr x 2 d y for dV and commute .
The element of the cone where the sample point (x, y) is located is a line
segment thro ugh the origin, containing the point (5, 15). So its slope is 15 / 5 = 3,
and its equation is
Y = 3x ' or x = 2'..
3.
2
dW = kTT ( i) (23 - y) dy Substitut e y / 3 for x.
k TT
= 9(23y 2 - y 3) dy
k rr 1s Add up th e dW's and take the limit (that is,
w = 9 Io (23y 2 - y 3) dy integrat e).
= kTT ( 23 y 3 - .!.y~)I LS You could int egrat e num erically.
9 3 4 o
5375
= krr = 4614.214 .. . k
4
If the liquid were water, with density k = 62.4 lb/ ft 3 , the total work would be
about 287,927 ft-lb .
Note: Work is an equivalent physical quantity to energy . For instance, the work done
compressing a spring is stored in that spring as energy. Foot-pounds of work can
be converted direc tly to joules or calories by the appropriate conversion factors .
Although first moment of force, called torque, also has the units (distance)(force),
torque is not the same physical quantity as work. For this reason, torq ue is usually
called "pound-feet" rather than foot -pounds (see Section 11-4).
560 Chap
ter 11:TheCalculus
ofVariable-Factor
Products
1# F
Problem Set 1 1 2
DoTheseQuickly
The following problems are intended to refresh your skills. You should b e
able to do all ten problems in less than five minutes.
Q1. What is the area under one arch of y = sinx?
Q2. What is the area under y = 4 - x 2 from x = - 2 to x = 2?
Q3. Find a velocity equation if the acceleration is a = tan t.
Q4. Find the acceleration equation if the velocity is v = ln t?
QS. Name the theorem that allows definite integrals to be calculated
by antiderivatives .
Q6. 2.f(x) dx is a - ?- sum .
Ql. What is the average valu e of y = sinx for one comp lete cycle?
QB.Name the technique for integrating f ex cos x dx.
Q9. Name the technique for finding dy / dx if x 3 y 5 = x sin 2 y.
Q10. Name the quick method for resolving an expression into partial
fractions.
l. Leaking Bucket Problem: Miss Hapse pulls a bucket of water up from the Figure l l -2c
bottom of the well (Figure l l-2c). When she starts pulling, she exerts
a force of 20 lb. But by the time she gets it to the top, 50 ft up, enough
water has leaked out so that she pulls with only 12 lb. Assume that
bo th the rate she pulls the bucket and the rate the water leaks are
t-- lOtons
up here
con stant, so the force she exerts decreases linearly with displac eme nt
from the bottom. How much work did Miss Hapse do in pulling the
buck et out of the well?
2. Spaceship Problem: A spaceship on the launch pad weighs 30 tons
(Figure l l-2d). By the time it reaches an altitude of 70 mi, it weighs
only 10 tons because 20 tons of fuel have been used. Assume that
the weight of the spaceship decreases lin ear ly with displacement
above the earth.
a. How many mile-tons of work were done in lifting the spaceship
to an altitude of 70 mi? . 30 tons
. / down here
b. The rocket engines exert a constant thrust (that is, force) of 90
tons. How much work was done by the engines in lifting the --------
spaceship to 70 mi? What do you think happens to the excess Figure l l -2d
energy from part a?
Origina l length
3. Spring Problem: It takes work to compress a spring (work = force x ofspring -
displacement). However, the amount of force exerted while compress-
ing the spring varies, and is directly porportional to the displacement,
s, the spring has been compressed (Figure ll-2e). This property is
known as Hooke's Law. Let k be the proportionality constant. Find the ---
--
Force
/:_ ( Displacem ent
work required to compress a spring from s = 0 to s = 10. s
Figure l l -2e
Section112: WorkDonebya Variable
Force 561
4. Table Moving Problem: You push a table across the floor. At first,
you push hard to get it moving, then ease off as it starts to move . y
The force drops to zero at a displacement of 4 ft, where the table Pump to here.
stops. Assume that the force, F lb, is given by T10 ft
F = -x 3
+ 6x 2 - 12x + 16,
--------J_
where x is the number of ft the table has been displaced. y= x 2
a. Sketch the graph of F and show that it really does have these
properties.
b. How much work is done pushing the table the 4ft ?
X
5. Conical Reservoir Problem: A conical reservoir 30 ft in diameter 4
and 10 ft deep is filled to the top with water of density 62.4 lb/ ft 3 .
Find the work done in pumping all of this water to a level of 7 ft Figure l l -2f
above the top of the reservoir.
6. Paraboloidal Tank Problem: A tank is made in the shape of the
paraboloid formed by rotating about the y-axis the graph of y = x 2
from x = 0 to x = 4 (Figure ll-2f). The tank is filled with benzene,
an organic liquid whose density if 54.8 lb/ ft 3 Find the work done
in pumping a full tank of benzene to a level of 10 ft above the top
of the tank .
7. Spherical Water Tower Problem: A spherical water tower 40 ft in
diameter has its center 120 ft abov e the ground (Figure ll-2g). A
pump at ground level fills the tank with water of density 62.4 lb/ft 3 .
a. How much work is done in filling the tank half full?
b . Quick! How much work is done in filling the tank completely?
(Be careful!)
Figure l l -2g
8. Flooded Ship Problem: A compartment in a ship is flooded to
a depth of 16 ft with sea water of density 67 lb/ ft 3 (Figure ll- y
2h). The vertical bulkheads at both ends of the compartment
have the shape of the region above the graph of
y = 0.0002x 4 ,
where x and y are in ft . The compartment is 15 ft long . How
much work must the bilge pumps do to pump all of the water
X
over the side of the ship, 30 ft above the bottom?
9. Carnot Cycle Problem: An automobile engine works by burning
gasoline in its cylinders. Assume that a cylinder in a particular Figure l l-2h
engine has diameter 2 in. (Figure l l -2i). When the spark plug fires,
the pressure inside the cylinder is 1000 psi (pounds per square
inch), and the volume is at its minimum, 1 in 3. As the piston goes
out, the hot gases expand adiabatically (that is, without losing heat
to the surroundings). The pressure drops according to the equation
P = k1v -1 .4,
where p is pressure, V is volume, and k 1 is a proportionality
Figure l l-2i
constant. When the piston is farthest out, and the volume is 10 in 3 ,
562 Chapter
11:TheCalculus
ofVariable-
FactorProducts
the exhaust valve opens and the pressure drops to atmospheric pressure, 15 psi.
As the piston comes back, the cool gases are compressed and the cycle is
repeated. For compression, p = k2 v - 1.4,where k2 is a differ ent proportionality
constant .
a. Find the work done on the piston by the expanding hot gas .
b . Find the work done by the piston as it compresses the cool gas.
c. Find the net amount of work done. This is the amount of work that is available for
moving the car .
d. How is "Carnot" pronounced? Who was Carnot?
11-3 Mass of a Variable-Density Object
The density of an object is defined to be the mass per unit volume. For instance,
water has a density of 1 g/c m 3 ; iron, 7.86 g/ cm3; and uranium, 18.5 g/cm 3 . Since
density is calculated by dividing the mass of an object by its volume, the mass is
Plunger equa l to density times volume.
Mass = (density)(volume)
High The density of a real object may vary from point to point within the object . For
density example, the dens ity of the materials composing the earth varies from about 12
g/cm 3 at the center of the earth to about 4 g/cm 3 at the surface.
Lower
density As another example, uranium oxide pellets, used as fuel in nuclear reactors, are
made by compacting uranium oxide powder with a press (Figure l l-3a). The powder
Figure l l-3a closer to the plunger in the press compacts to a higher density than that farther
away.
In this section you will explore ways of predicting the mass of an object if you know
how its density behaves at various places.
OBJECTIVE Given a function specifying the density of an object at various places within that object ,
calculat e th e total mass of the object.
Example 1 shows how definite integration can be used to calculate the total mass
of a hypothetical object where the density varies using techniques you know. Your
purpose is to apply these techniques to real-world problems.
Example 1 A solid is formed by rotating about the x-axis the region under the graph of y = x 113
from x = 0 to x = 8. Find the mass of the solid if the densit y, p (Greek letter "rho")
a. Varies axially (in the direction of the axis of rotation), being directly proportional
to the square of the distance from the yz-plane .
b . Varies radially (in the direction of the radius), being directly proportional to the
distance from the x -axis.
Section 11-3: Massofa Variable
-DensityObject 563
Solutions a. Figure l l-3b shows the solid . A vertical slice of the rotated region generates
a disk parallel to the yz -plane. So each point in the disk has essentially the
same density as at the sample point (x , y ) . Letting m stand for mass and V
stand for volume, the mass of a representative slice is as follows .
dm = pdV
p is dlr ectly proportiona l to the squar e of
p = kx 2 x, th e dis tance from th e y z -plan e.
By geometry, volum e= (cross -sectional
dV = rr y 2 dx area) (length ).
dm = kx 2 rry 2 dx
Sub stitu te for p and for dV .
= krrx 2 (x 113) 2 dx = krrx 813 dx
8 13 Add the dm 's and find the limit. That is,
.. m = Jo8krrx dx integrate.
= l...krrxll
11
/3 80
1
= 6 144 krr (Find th e actual mass by substituting for k
11 and doin g th e arithm etic.)
b. Since the density varies radially (with the distance from the x-axis), slicing the
rotated region horizontally will produc e a constant density within the slice .
Rotating slices parallel to the axis of rotation produces cylindrical shells, as
shown in Figure ll-3c . Again, picking a sample point on the curve, the mass
of a representative shell is as follows.
dm = pdV
= ky 2rry(8 - x ) dy p = ky and dV = 2rry(8 - x) dy .
= 2k rr y 2 (8 - y 3) dy = 2k rr (8y 2 - y 5
) dy
Add the dm's and find the limit.
m = 2k rr Jo2(8y 2 - y 5 ) dy Why are th e bounds O to 2?
2
= 2krr oy 3 - Y6) I 0
= 2krr ( ~3 - B3 - 0 + 0) = :
3
kTT
y y
dy
X X
8 8
Figure l l-3b Figure l l -3c
Note: Part b of the example shows the real reason for slicing objects into cylindrical
shells. If the density varies radially, it will be constant (essentially) at all points in
the shell. Slicing into plane slices would give a slice in which the density varies.
564 Chapter
11: TheCalculus
ofVariab
le-Factor
Product
s
15
Problem Set 11-3
DoTheseQuick
ly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Q1. What is the volume of a cone inscribed in a 6-cm 3 cylinder?
Q2. What is the volume of a paraboloid inscribed in a 6-cm 3 cylinder?
Q3. y varies linearly with x . y is 12 if x = 0 and 20 if x = 2. Find y if x = 3.
Q4. Sketch the graph: y = sin x.
QS. Sketch the graph: y = lnx.
06. Sketch the graph: y = 2x.
Ql. Sketch the graph: y = x 2 .
QB. Density = (-?-) / (- ?-)
Q9. Work = (- ?- )(- ?-)
Q10. If y = sin - 1 x, then y ' = -?-.
1. The region bounded by the graph of y = In x, the x-axis, and the line x = 3 is rotated
about the line x = 0 to form a solid. Find the mass of the solid if
a. The density varies inversely with the distance from the axis of rotation.
b. The density varies linearly with y, being 5 when y = 0, and 7 when y = 1.
2. The region bounded by the graph of y = sinx and the x-axis, between x = 0 and x = TT,
is rotated about the y-axis to form a solid. The density of the solid varies directly with
the distance from the axis of rotation. Find the mass of the solid.
3. The region under the graph of y = 9 - x 2 is rotated about the y-axis to form a solid.
Find the mass of the solid if
a. The density is a constant, k.
b. The density is constant in any thin horizontal slice but varies directly with the
square of y in the y-direction.
c. The density does not vary in the y-direction but is directly proportional to the
quantity (1 + x), where xis the distance between the sample point and the axis of
rotation .
d. Which of the solids in 3a, b, and c has the greatest mass ? Assume that the constant
k is the same in all three parts .
4. The region bounded by the graphs of y = .jx and y = O.Sx is rotated about the x-axis
to form a solid. Find the mass of the solid if
a. The density varies axially, being directly proportional to the distance between the
sample point and the yz-plane.
b. The density varies radially, being directly proportional to the square of the distance
between the sample point and the axis of rotation .
Section 113: Massofa Variable
-Density
Object 565
50 oz/ in3 80 oz/ in 3
5. Two Cone Probl em: Two con es have the same size, base radius
3 in . and altitude 6 in. (Figure ll-3d). Both have th e same \ \
weight densiti es at their two en ds, 50 oz/ in 3 and 80 oz / in 3 ,
but one has the higher density at the bas e and the other has
the higher density at the vertex. In both, th e dens ity varies
linearly with the distanc e from th e plan e of the bas e.
a. Without performing any calculations, pr edict which cone
has the higher mass . Explain your reasoning .
I
80 oz/ in 3
I
50 oz/ in3
b. Confirm (or refute!) your prediction by calculatin g the Figure l l-3d
mass of each cone .
6. Two Cylinder Problem: Two cylinders hav e th e same shap e, ro,;;~Oo,;,o' 50 oz/ in3
3-in. ra dius and 6-in. altitude (Figure l l- 3e). One has density
50 oz/ in 3 along the axis, and 80 oz/ in 3 at the walls. The other
has densit y 80 oz/ in 3 along the axis, and 50 oz/ in 3 at the
l---- 80 oz/ in3
walls. In both cylinders the density varies linearly with the
distanc e from th e axis.
a. Without performi ng any calcu lations, predict which cylin-
der has the higher mass. Explain .
b. Confirm (or refute!) your pr ediction by calculating the Figure l l -3e
mass of each cylinder.
y
7. The region bound ed by the graphs of y = 4 - 2x 2 , y = 3 - x 2 , and
x = 0 (Figure l l-3f) is rotated about th e x-axis to form a solid . Both
x and y are in cent im eters. The density var ies directly as the square
of the distance from the yz-p lan e (that is, the "bas e" of the solid) .
Find the mass of the solid.
8. The region in Probl em 7 (Figur e ll-3f) is rotated about the y-axis
to form a differ ent solid. The densit y decreases expon entiall y with X
distance from the y-axis, according to th e eq uation p = e-x. Find
z
the mass of the solid .
Figure l l-3 f
9. Ura nium Fuel Pellet Problem : Uranium oxide is us ed as a fuel in nuclear
pow er plants th at generate electricity . Powder ed uranium oxide is compressed
into pellets as shown in Figur e l l -3a. The compressing mak es th e pellets slight ly
denser at the top than they are at the bottom. Suppose that th e cylindrical pellets have
a bas e diam eter of 1 cm and an altitude of 2 cm. Assume that th e density is constant
in the radial direction, but varies with y, the distan ce from th e bottom of the pellet (as
shown in th e table), being 9 g/cm 3 at the bottom and 10 g/ cm 3 at the top. Predict the
mas s of the pellet, taking into considerat ion the variabl e density.
y(cm) density (g/ cm 3)
2.0 10.0
1.6 9.9
1.2 9.8
0.8 9.6
0.4 9.4
0 9.0
566 Chapter
11:TheCalculus
ofVariable-Fac
tor Products
ween e M#M ttMit
10. The "triangular" region in the first quadrant bounded by the graphs of y = 4 - x 2 ,
y = 4x - x 2 , and x = 0, is rotated about various axes to form various solids.
a. Find the mass of the solid formed by rotating the region about the y-axis if the
density varies directly with x, the distance from the y-axis.
b . Find the mass of the solid formed by rotating the region about the x-axis if the
density varies directly with x, the distance from the yz-plane.
c. Find the mass of the solid in 10a if the density varies directl y with y, the distance
from the xz-plane, instead of with x.
11. Find the mass of a sphere of radius r if
a. The density varies directly with the distance from a plane through the center .
b. The density varies directly with the distance from one of its diameters.
c. The density varies directly with the distance from the center . (Use spherical shells.)
12. Mass of the Earth Prob lem : The density of the earth is about 12 g/ cm 3 at its center, and
about 4 g/ cm 3 at its surface. Assume that the density varies linearly with the distance
from the center . Find the mass of the earth.
Useful information:
1 mi= 5280 ft; 1 ft= 12 in; 2.54 cm= 1 in.
Radius of the earth is about 3960 mi.
Slice into spherical shells.
13. The region under the graph of y = ex from x = 0 to x = rr / 2 is rotated about the y-axis
to form a solid. The density is given by p = cos x. Find the mass of the solid.
14. Buckminster's Elliptica l Dome Problem : Architect Buckminster
Fuller (1895-1983) once proposed that a dome should be built
over Manhattan Island and an air-conditioning system be built
to regulate the temperature, remove air pollution, and so forth.
Your job is to find out how much air would be inside such
a dome . Figure l l-3g shows what that dome might look like.
Assume that the dome is a half-ellipsoid 8 mi long, 2 mi wide, Figure l l -3g
and 1/ 2 mi high .
a. The three-dimensional equation for an ellipsoid is
where a, b, and care the x-, y-, and z -radii, respectively. Show that any cross section
of the ellipsoid parallel to the xy-plane is an ellipse.
b . The weight density of air at sea level (z = 0) is about 0.08 lb/ ft 3 . But it decreases
with altitude according to the equation p = 0.08e - 0 22 , where z is in mi. Find the
mass of air inside the dome. How many tons is this?
c. Suppose that you assume the density of the air is constant throughout the dome,
0.08 lb/ ft 3 How many mor e lb of air would you have assumed are in the dome than
are actually there ? Surprising?!
d. The volume of a (whole) ellipsoid is V = (4/3 )rrabc. See if you can derive this
formula by integrating the dV from this problem.
Sectio
n 11-3:Massof a Variable-Density
Object 56 7
11-4 Moments, Centroids, Center of Mass, and the
Theorem of Pappus
If you hold a meter stick at one end and hang a weight on it, the force caused
by the weight twists the met er stick downwards (Figure ll-4a). The farther from
your hand you hang the weight, the more th e twisting. Doubling the displacement
doubles the twisting for a given weight. And doublin g the weight hung at the
same displac ement also doubles the t,visting. The amount of twisting is given the
name torque, pronounced "tork." The torque with resp ect to an axis through your
hand is defined to be the amount of force created by the weight, multiplied by the
displacement from your hand perpendicular to the direction of the force.
Torque = (force)(displacement)
I-- Greater _
1
I
Moment
J 1/ dis<:::efo
J )
Greater
("twis t") mom ent
Figure l l -4a
Torque is just one exampl e of a more general conc ept called moment . The word
com es from the Latin mov ere, "to move," and momentum, "moving pow er." The
mom ent of a physical quantit y equals the magnitude of that quantity, multipli ed by
some power of the displacem en t from a reference point, line, or plane to the place
where the quantity is located.
Definition: nth Moment
nth moment of quantity = (magnitude of quantity){displacement)n
The torque produced by the weight in Figure l l -4a is thus the first moment of force
with respect to an axis through your hand. You can find any order moment of any
quantity with respect to a point, line, or plane. Some moments have interesting
meanings in the real world . Others are of interest just in the mathematical world . In
this section you will explore first and second moments of mass, volume, arc length,
and area . You will use th e results to calculate the center of mass of a solid, th e point
where all of the mass could be concentrated to produc e the same first moment, and
the related centroid, which is a center of volum e, length, or area. Calculus is used if
parts of the object are at diff erent displacem ents .
OBJECTIV
E Given the description of a solid or a plane region, find its first or second moment of
volume, area, or mass with respect to a point, line, or plane, and its center of mass,
volume, or area.
568 Chapter
11: TheCalculus
ofVariable-Factor
Products
Example 1 A solid paraboloid has the shape of the solid formed by rotating about the y-axis
the region in Quadrant I above the graph of y = x2 and below the line y = 9 (Figure
ll-4b) .
)' = x 2
a. Find the volume of the solid if the dimensions are in centimeters.
Centroid, b . Find Mxz, the first moment of volume with respect to the xz-plane.
(x,y,zJ
c. Find y ("y bar"), they-coordinate of the centroid, at which y-volume= Mxz from
X step b.
z
d. At the centroid (x , y, z) , x-volume= Myz, and z -volume= Mxy, where the moments
Figure 1 l-4b are with respect to the yz- and xy-planes, respectively. Explain why both x and z
are zero for this solid.
Solutions a. Let V stand for volume. Slicing perpendicular to the y-axis gives
dV = TTX2 dy = TTYdy .
V = Io9TTydy
= :!:l:y2 19
2 0
= 40.STT = 127.234 ... cm 3
b. Let Mxz stand for the moment ,,vi.threspect to the xz-plane . Slicing perpen-
dicular to the y-axis as in part a makes the slice parallel to the xz-plane, so
every point in the slice has approximately the same displacement from th e
xz-plane as the sample point (x, y).
dMxz = ydV Moment = (dis placemcnt)(vo Jume)
2
= y TTYdy = TTY dy Substitute for dV
r9 TTY2 dy
Mxz = Jo
Add up the dM's and take the limit
(integrat e)
= f y 3 1:= 243 TT= 763.407 ... cm~ The units are (cm){cm3 ).
C. y 40.STT= 243TT (y)(volum e) = moment
_ 243TT
y = _ TT = 6 cm The units arc (cm4 )/ (cm 3 ).
40 5
Note that the centroid is two-thirds of the way up from the vertex to the
base . This fact is to be expected since the solid is v.rider near the base .
d. The displacement is a directed distanc e. Since the solid is symmetrical v.rith
respect to the xy- and yz-planes, th ere is just as much negative moment
on one side of the plane as there is positive moment on the other side.
The moments v.rith respect to these planes are thus zero. So the x- and
z-coordinates of the centroid are also zero, and the centroid lies on the
y -axis.
The word centroid means "center-like." The centroid of an object is a geometrical
center.
Section 11-4:Mome
nts, Centroids,Cen
ter of Mass,andtheTheorem
of Pappus 569
Definition:Centroid
The centr oid of a solid is the point (x, y, z) at which
xV = Myz, yV = Mxz, and zV = Mxy,
where Vis volume, and Myz , Mxz, and Mxy are the first moments of volume of the solid
with respect to the yz-, xz-, and xy-planes, respective ly.
Centroids for plane regions and curves (center of area and center of arc length) are
similarly defined . For instance, a region of area A in the xy-plane has a centroid at a
displacement from the y-axis given by xA = M y .A curve of length Lin the xy-plane
has a centroid at a displacement from the x-axis given by yL = Mx.
If a solid has a uniform density, its center of mass, or balance point, will b e at the
centroid . If the density of the solid is not uniform, the center of mass can b e at a
place other than the centroid. Example 2 shows how to calculate the center of mass.
Example 2 Suppose that the solid in Example 1 has a density that is constant radially but varies
axially, being equal to y 112 g/ cm 3
a. Find the mass of the solid.
b. Find th e first moment of mass with respect to the xz-plane .
c. Find the y-coordinate of the center of mass, the point (x, y, z ) for which
(y)(mass) = moment of mass with respect to the xz-plane . Show that the answer
is reasonable .
Cent er Centroid ,
of mass (x,y, z)
Figure l l -4c
Solutions a. Figure ll -4c shows the solid, sliced horizontally as in Example l. Since the
density varies only axially and all points in the slice are about the same
displacement from the xz -plane, the mass, dm, of any slice is the density at
the sample point times the volume of the slice .
dm =p dV =yL rry dy =rry dy
12 312 From Example 1, dV = rry dy.
m =J: rr y dy =~rry
312
I:=97.2 rr ~ 305.4 g
512 Units ar e (g/ cm 3)(cm 3).
b . Let Mxz stand for th e first moment of mass with respect to the xz-p lan e. All
points in a slice are about the same displacement from this plane. So the
570 Chapter
11:TheCalculus
ofVariab
le-FactorProducts
moment of the slice, dM xz, equals the displacement from the xz-plane to the
sample point, times the mass of the slice.
dM xz = Y dm = yTTy 3 12 dy = rry 512 dy
r9 - 2 2 2 9 43 74 TT
Mxz = Jo TTY)/ dy = TTy71 0 = -- -:::::: 1963.0 cm-g
1 Unit s are
(cm)(g).
7 7
c. The y-coordinate of the center of mass, y, is found using the fact that
(displacement from center of mass)(total mass) = moment .
Y m = Mx,
- = 4374 TT/ ? = 5 17 =6428 cm Unit s are [ (cm ) (g)] /( g ).
Y 97 .2TT
The answer is reasonable because it is a little larger than 6 cm, the y-
coordinate of the centroid. Since the density is greater near the base than
at the top of the solid, the center of mass is closer to the base than is the
centroid.
Example 3 shows how to extend the concept of moment and centroid to a plane
region.
Example 3 Let R be the region in Quadrant I bounded by the graph of y = 3 cos x and the two
coordinate axes, where x and y are in inches.
a. Find the first moment of area of R with respect to the y-axis.
b. Find the first moment of area of R with respect to the x-axis .
c. Find the centroid of R, the point (x, y) at which (x)(area) = first moment with
respect to the y-axis, and (y) (area) = first moment with respect to the x-axis.
Solutions a. Figure l l-4d shows the region R. Slicing parallel to the y-axis gives strips in
y which all points are about the same displacement from the y-axis. Use A for
3 area and My for moment with respect to the y-axis. The moment, dM y, of a
y=3cosx
strip of area dA is
dMy = xdA = x(3 cosx) dx
- - _x_ - - (x , y) = 3xcosxdx.
rrr/ 2
My = Jo 3xcosxdx
X
TT/2
3 Why are the
= 3xsinx + 3cosx IO = ; + 0 - 0- 3 = l.71238 ... in 3 units in 37
Figure 11-4d
b. To find Mx, the moment with respect to the x-axis, slice horizontally (Figure
l l -4e). That way, each point in the strip will be about the same displacement
from the x-axis.
dM x = ydA = y(xdy)
You could put cos - t (y / 3) in place of x and integrate from O to 3. If you are
integrating algebraically, it is easier to substitute 3 cos x for y and - 3 sin x dx
X
for dy. Since xis TT/ 2 when y is 0, and O when y is 3, the limits of integration
are TT/ 2 to 0.
Figure 11-4e
Section 11-4: Moments,
Centroids,
Centerof Mass,andtheTheorem
of Pappus 571
0
Mx = Jrrf / 2 (3 cosx)(x)( -3 sinxdx)
0 0
= -9 Jrrf / 2 xcosxsinxdx = _22 Jrrf / 2 xsin2xdx
= - 22 (-! 2 xcos2x + .!.
.j
sin2x) 1/ 2
TT
= -( 0 + 0 - ~ + 0) = 9
; = 3.534291. . . "" 3.53 in 3
y c. To find th e centroid, you need to find the area of the region first . You can
3 either integrate 3 cos x from O to rr / 2 or recall that the area under a half-arch
of y = cosx equals 1 so that th e total area is 3.
xA = My
- - - - y = l.J 78 . . _ 3rr / 2-3 .
X = = 0.570796 ... :::< 0.57 m.
y = 3cosx
3
X
yA=Mx
Figure l l -4f
9 18
y = ~ = 1.178097 . . . "" 1.18 in.
Figure l l -4f shows the location of the centroid, point (0. 57 ... , 1.178 ... ).
The photograph shows th e region in Example 3 drawn on cardboard . The cutout will
balance on a pencil point placed at th e centroid!
The last examp le for this section shows you th e meaning of second moment.
Example 4 For the region R in Example 3,
a. Find the second moment of ar ea of the region with respect to the y-axis .
b. Find x, the x-coordinate of the center of second moment, which is the displace-
ment from the y-axis for which
(Displacement )2(area) = second moment.
Solutions a. Slice parall el to the y-axis as previously shown in Figure l l-4d so that points
in a strip will be about the same displacement from that axis.
Second moment = (area) (squ are of
dMy = x 2 dA = x 2 (3 cosx dx ) displ acemen t)
frr / 2 2 .J
My = Jo 3x cos x dx = 1.40 2203 . .. in Why are the unit s in -l?
572 Chapter
11:TheCalculus
ofVariable
-FactorProducts
b. To find the center of second moment, you must us e the fact that displacement
squared times area equals second moment.
(x )2A = My(x) 2 (3) = 1.402203 .. .
x= 4 2 3
~ 0.68 in. Why do th e units come
~l. 0 ~0 ... = 0.683667 . ..
out inches?
The "stiffness"of a beam is related to th e second moment of area of a region that
has the cross-sectional shape of the beam. The resistance of a uniform flat plate
to being rotated about an axis is related to the second moment of area about that
axis. The higher the second moment, the more difficult it is to start the region
rotating, or stop it once it gets started. For this reason the second moment of area
is sometimes called the moment of inertia of the region. The displacement to the
center of second moment from the axis of reference is called the radius of gyration .
The same terms apply to second moments of volume or mass of a solid figure.
Problem Set 1 1 4
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
Ql. Evaluate - x2 if xis 5.
Q2. Evaluate x2 if x is -11.
Q3. Write sin 2x in terms of functions of x.
Q4. Write cos 2 x in terms of cos 2x.
QS. Resolve into partial fractions: 1/ [ (x - 2) (x - 5)]
Q6. Integrate: f dx / [(x - 2)(x - 5)]
Ql. Differentiate: y = 1/ [(x - 2)(x - 5)]
QB.What hypothesis of Rolle's theorem is not a hypothesis of the mean value theorem?
Q9. Write the definition of indefinite integral.
Q10. Is f(x) = x 3 - 7x increasing or decreasing at x = 2?
1. Paraboloid Problem: The region in Quadrant I under the graph of y = 9 - x 2 is rotated
about the y-axis to form a solid. Assume that dimensions are in centimeters.
a. Find the volume of the solid.
b. Find its first moment of volume with respect to the xz-plane. y
c. Find its centroid. 5
2. Ellipsoid Problem: A half-ellipsoid is formed by rotating about the
X
x-axis the region in the first quadrant bounded by (Figure l l -4g)
12
(t2 ) + ( f) = 1 (dimensions in feet).
2 2
: ;'
:/
a. Confirm by appropriate integration that the volume is ! rr (12) (52 ) .
b. Find the first moment of volume with respect to the yz-plane.
Figure l l -4g
c. Find the centroid.
Section11-4:Moments,
Centroids,
Centerof Moss,andtheTheorem
of Poppus 573
3. Paraboloid Mass Problem: The paraboloid in Problem 1 has density (g/ cm 3 ) that is
constant in the radial direction, but directly proportional to the cube root of y in the
axial direction.
a. Find the mass of the solid .
b. Find the first moment of mass of the solid with respect to the xz-p lane.
c. Find the center of mass.
d. True or false: "The center of mass of a solid is always at the centroid."
4. Ellipsoid Mass Problem: The half-ellipsoid in Problem 2 has a weight density (lb/ ft 3 )
that is constant in the radial direction, but varies directly with x in the axial direction.
a. Find the weight of the solid.
b. Find the first moment of weight of the solid with respect to the y
yz -plane. y = ex
c. Find the center of weight (which is the same as the center of mass
in this case).
d. True or false : "The center of mass of a solid is always at the centro id ."
5. Exponen tial Region and Solid Problem: Let R be the region under the
X
graph of y = ex from x = 0 to x = 2 (Figure l l -4h).
0 2
a. Find the x-coor dinat e of the centroid of R .
b. R is rotated about the x-axis to form a solid . Find the x-coordinate Figure 11-4h
of the centroid.
c. True or false: "The centroid of a region and the centroid of th e
y
solid formed by rotating that region have the same x-coordinate ."
6. Secant Curve Region Problem: Let R be the region under the graph
of y = sec x from x = 0 to x = TT/3 (Figure ll-4i).
a. Find the x-coordinate of the centroid of R .
b. R is rotated about the x-axis to form a solid . Find the x-coordinate
of the centroid .
X
c. True or false: "The centroid of a region and the centroid of the 0 rr/ 3
solid formed by rotating that region ha ve the same x-coordinate."
Figure 11-4i
7. Centroid of a Triangle Experiment: Prove that the centroid of a
triangle is located one-third of th e way up from the base (Figure
l l-4j) . Then draw a triangle on cardboa rd , draw lines one-third of
the way from each base, and cut out the triangle. If your work is '
done accurately, you should find that the lines intersect at one point !h
'
and that you can balanc e the triangle on th e point of a pencil placed ----1------1;; ---
at that point.
b
8. Centroid Cut -Out Exper im ent: Let R be the region under the graph
of y = x 213 from x = 0 to x = 2, where x and y are in centimeters. Figure 11-4j
a. Find the area of R.
b. Find the first moment of R with respect to the x-axis .
c. Find the first moment of R with respect to the y-axis .
d. Find the coordinates of the centroid of R.
574 Chapte
r 11: TheCalculus
ofVariab
le-FactorProd
ucts
e. Plot an accurate graph of R on an inde x card or graph paper. Mark
the centroid. Then cut out th e region and try to balance the region
on a pencil point at the centroid.
9. Second Moment of Area Problem: Let R be the region under the graph
of y = sinx from x = 0 to x = TT (Figure ll-4k), where x and y are in
centimeters. Figure l l-4k
a. Show that the centroid of R is at x = TT / 2.
b. Find the second moment of area of the region
with respect to the y-axis . H
c. Find the radius of gyration (displacement to H -- - -- -- --
center of second moment) of R with respect to
the y-axis .
10. Second Moments for Plane Regions Problem: Find B B
the second moment of area for the following
shapes (Figure 11-41):
a. A rectangle of base Band height H, with respect
to an axis along the base. Ti
H
J___
b. A rectangle of base B and height H, with respect
to an axis through the centroid, parallel to the
base.
c. A triangle of base B and altitude H, with respect B
to an axis along the base.
Figure l 1-41
d. A triangle of base Band altitude H, with respect
to an axis parallel to the base, one-third of the
way to the opposite vertex (that is, through the
centroid).
11. Second Moments for Solid Figures : Find the second
moment of volume and radius of gyration for these H
solids (Figure 11-4m):
a. A cylinder of radius R and altitude H, with
respect to its axis.
b. A cone of base radius R and altitude H, with
respect to its axis .
Figure l l-4m
c. A sphere ofradius R, with respect to a diameter. I, Rotat e
'
I Rotate
12. Rotation of Solids Problem: The amount of ~ -P>
resistance an object has to starting or stopping
rotation is measured by its second moment of
mass with respect to the axis of rotation . If the
density is constant, the second moment of mass I
..... 1'
-~~-~
equals the density times the second moment
of volume. Suppose that 1000 cm 3 of clay is
made into a sphere, and another 1000 cm 3 is
made into a cylinder whose diameter equals its Equal
volume s
altitude (Figure 11-4n). Which one has higher
resistance to rotating? Justify your answer.
Figure l l -4n
Section11-4:Moments,
Centroids,
Centerof Mass,andtheTheorem
of Pappus 575
13. Beam Moment Problem: The stiffness of a beam
is directly proportional to the second moment of
area of the beam's cross section with respect to an
axis through the centroid of the cross section. In
this problem you will investigate the stiffness of 12
beams with the same cross-sectional area but with
different shapes.
2 12
a. Prove that if a rectangle has base B and altitude
H, then the second moment of area vvith respect Turned on edge Lying flat
to an axis through the centroid and parallel to
the base is equal to BH 3 / 12.
b. Find the stiffness of a 2" by 12" beam (Figure
11-40) if it is
i. turned on edge.
ii. lying flat.
Use k for the proportionality constant . Based 4 4
on the results, tell why boards used for floor
joists in houses are turned on edge rather than
All parts 4 in . wide rx 8" web
being laid flat. Figure 11-40
c. An I-beam is made with the same cross-sectional
area as the 2" by 12" in 13b. Find its stiffness if
i. all three parts (two flanges and one web) are 4 in . wide.
ii . the two flanges are 2" by 4" , but the web is l " by 8".
d. Does increasing an I-beam's depth without changing the
cross -sectional area make much change in the beam's
stiffness? What physical limitations keep people from
making a beam very tall and thus very stiff?
14. Introduction to the Theorems of Pappus: The region R under the
graph of y = x 3 from x = 0 to x = 2 is rotated about the y-axis
to form a solid (Figure l l-4p).
a. Find the area of R.
b. Find the volume of the solid using vertical slices of R.
c. Find the first moment of area of R with respect to the y-axis. y
What do you notice about the integral?
d. Find the x-coordinate of the centroid of R .
e. A theorem of Pappus states that the volume of a solid of
revolution equals the area of the region being rotated times
the distance the centroid of the region travels. Show that this y =x3
problem confirms the theorem.
,(x , y)
I
I R
Figure 1 l -4p
576 Chapter
11:TheCalculus
ofVariable-Factor
Products
15. Theorem of Pappus Problem: Pappus was a Greek mathematician who lived in Alexan-
dria in the fourth century A.D. One of his theorems states the following.
Theorem:
TheTheorem
of Pappusfor Volumes
The volume, V, of a solid of revolution is given by
V = 2rrRA,
where A is the area of the region being rotated, R is the displacement from the axis of rotation
to the centroid of the region, and the region is not on both sides of the axis of rotation . The
quantity 2rrR is thus the distance the centroid travels as the region rotates.
In Problem 14, you saw an example of this theorem. In this problem you will use the
theorem, once forward and once backward.
a. Toroid Prob lem: A toroid (Figure l l -4q) is
formed by rotating a circle of radius r about
an axis R units from the center of the circle,
where r s R. Find the volume of the toroid.
b. Centroid of a Semicircle: A semicircle of
radius r is rotated about its diameter to
form a sphere (Figure ll-4r). You know
formulas for the area of a semicircle and for
the volume of a sphere. Use these facts to
find the displacement from the center of a
semicircle to its centroid.
R-
Figure l l -4q Figure l l -4r
16. Theorem of Pappus Proof: Prove the theorem of Pappus for volumes.
Sectio
n 11-4: Moments,
Centroids,Centerof Mass,andtheTheorem
of Pappus 577
11-5 Force Exerted by a Variable Pressure -
Center of Pressure
Pressure exerted by a fluid such as air or water is defined to be the force exerted
by the fluid per unit area. For instance, water in your home's pipes is usually at a
pressure of 40 to 100 psi (lb/in 2 ) . The air in a scuba diving tank is compressed to
about 3000 psi. This means that each square inch of the tank's wall is pushed with
a force of 3000 lb . As a result of the definition of pressure,
Force = (pressure)(area).
In many real-world situations, the pressure acting on a surface is different at various
places on the surface. For instance, the pressure acting on an airplane's wings is
usually higher near the middle of the wing than it is at either the leading or trailing
edge. So the total force, which holds up the plane, must be found by integrat ing
rather than just by multiplying.
In this section you will learn how to calculate the force exerted by a variab le
pressure. You will also calculate the momen t of force, and use it to find the ce nter
of pressure, where the entire force could be concentrated to produce the same
moment.
OBJECTIVEGiven a region and a function specifying the pressure acting on the region, be able to
calculate the total force, the moment of force, and the center of press ure.
Example 1 Weir Problem: A weir (a small dam) is to be built across a stream. When finished, the
weir will have a vertical parabolic cross section the shape of the region above the
graph of y = x 2 and below the line y = 4, where x and y are in feet (Figure 11-Sa).
The pressure at any sample point below the surface of the water will be directly
proportional to the displacement from the surface to that point. The proportionality
constant is the density of water, 62.4 lb/ ft 3 .
Figure l l -5a
a. Predict the total force acting on the weir when the water is all the way to the top .
b. Find the first moment of force with respect to the x-axis .
578 Chapter
11:TheCalculus
ofVariab
le-Fac
tor Products
c. Find the "center of pressure," the point on the face of the weir where the entire
force could be applied to produce the same moment with respect to the x-axis.
d. The weir face is wider at the top, so you would expect the center of pressure to
be more than halfway up. But the pressure is greater toward the bottom, so you
would also expect th e center of pressure to be less than halfway up . Based on
your answer to part c, which of these two competing effects predominates?
Solutions a. Slice the region parallel to the x-axis so that the pressure will be essentially
constant at any point in the slice. Pick a sample point (x, y) in the slice and
on the graph. Let F stand for force, p for pressure, and A for area. By the
definition of pressure,
dF = pdA.
Since pressure varies directly with depth, (4 - y ) , the pressure at the sample
point will be
p = k(4 - y),
where k = 62.4 lb/ ft 3, which will be substituted at the end to get a numerical
answer. Since dA = 2x dy = 2y 112 dy, you can write
dF = 2k(4 - y)y 112dy.
To find the total force acting on the weir's face, you add up all the dF's
and take the limit as dy goes to zero. This process is, of course, definite
integration.
F = Jo42k (4 - y )y 112 dy
4
f = 2k ( li y 3!2 _ i y 512) I
3 :, O
256k (256 )( 62.4)
15 15
= 1064 .96, or about 1065 lb
b. The first moment of force with resp ect to the x-axis is defined to be the
product of the force and the displacement from the x-axis to the point where
the force is acting. Fortunately, all points in the horizontal slice (Figure l l-5a)
are essentially the same displacement from the x-axis as the sample point
(x, y ) . Using M for moment, the moment acting on the slice is
dM = ydF = y pdA,
where y and p are both measured at the sample point.
dM = y 2k (4 - y)y 112 dy Substitute for p and d.A
= 2 k (4y 3/2 - y S/2) dy
.-. M = 2k Io4(4y 3/2 - y S/2) dy Add th e d.Ms and take th e limit (that is, inte grat e).
4
= 2k (~:, y 5/2 _ .?.
7
y 512) I0
1024k (1024) (62.4)
= 1825.645 ... , or about 1826 lb-ft.
35 35
Section11-5: ForceExerted
by a Variable
Pressure
---Center
of Pressure 579
c. By the definition of center of pressure, its vertical coordinate, y, is the
number for which
yF = M .
. -= 1024k / 35 = g r 1~ f
.. y 2 s6k I 15 7 ' 0
7 t
By symmetry, the x-coordinate, x, of the center of pressure is zero, so the
center of pressure is at (0, 1 ~).
d. The center of pressure is less than halfway up, meaning that the increasing
pressure at greater depths predominates over the decreasing area.
As you learned in Section 11-4, the first moment of force with respect to an axis is
called torque. In part b of Example 1 above, the 1826 lb-ft is the amount of torque
exerted by the water on the dam face . It measures the amount of twisting the force
does on the dam face, tending to make it rotate about the x-axis and fall over.
Problem Set 1 1 -5
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
Ql. The geometrical center of a solid is called its - ?-.
Q2. The point where all the mass of a solid could be concentrated is called its - ?- .
Q3. The displacement from an axis to the center of second moment with respect to that axis
is called the - ?- .
Q4. The process of adding up parts of a physical quantity then taking the limit as the size of
the parts goes to zero is called -? - .
QS. The process of finding the antiderivative is called - ?- .
Q6. Density equals-?- divided by - ?- .
Ql. Simplify: (x 113 )(x 116 )
QB. Integrate: f sec x dx
Q9. Differentiate: y = tan - 1 x
Q10. (O.S)(ex + e-x ) is defined to be-?-. y
1. Trough Problem: A trough has a vertical end in the shape of the
region above the graph of y = 2x4 and below the line y = 2, where
x and y are in feet (Figure 11-Sb). The trough is filled with liquid of
density k lb/ft 3 .
X
a. Find the force acting on the end of the trough. Recall that force
= (pressure)(area), and the pressure at a point varies directly
with the displacement from the surface of the liquid to that Figure l l -5b
point .
b. The force in la causes a moment with respect to the x-axis . However, different
points on the end of the trough are at different displacements from the x-axis . Find
the moment of force with respect to the x-axis .
580 Chapter
11: TheCalcu
lusofVariable-
FactorProduc
ts
ffl 4 4
c. Find the center of pressure, where the entire force could be concentrated to produce
the same moment with respect to the x-axis.
2. Dam Problem: At its narrowest point, Scorpion Gulch is 20 ft wide and 100 ft deep. A
dam at this point has its vertical face in the shape of the region bounded by the graphs
of y = x 2 and y = I 00, where x and y are in ft. Answer the following questions.
a. Confirm that the dam is 20 ft wide at the top.
b. Find the area of the vertical dam face.
c. When the gulch is filled with water to the top of the dam, the greatest force will be
exerted. It is important to know whether this force will be large enough to rip the
dam from its foundations. Find this total force.
d. The dam could also fail by being pushed over. The first moment of force with
respect to the x-axis is the quantity that the dam must withstand to prevent this.
Find the first moment of force with respect to the x-axis.
e. How far above the bottom of Scorpion Gulch could the entire force be concentrated
to produce the same moment as in 2d?
3. Ship's Bulkhead Problem: A bulkhead on a ship is a vertical
wall that separates two compartments. Bulkheads are often y
designed so that they will withstand the water pressure if the
compartment on one side of it gets flooded. Suppose that a
bulkhead is to go all the way across a ship (Figure 11-Sc). The
bulkhead is to be 40 ft wide at the top and 32 ft from bottom to
top. The cross section of the ship where the bulkhead will go is
in the shape of the quartic ellipse
(to)+ ( Y ;232)
4 4 =i
where y is the vertical displacement (in ft) from the bottom of
the ship to the sample point and x is the horizontal displacement Figure l l -5c
(also in ft) from the center line of the ship to the sample point.
a. Find the area of the bulkhead.
b . Find the force that the bulkhead must be designed to withstand from water pressure
if the compartment on one side is filled to 32 ft with seawater of density 67 lb/ft 3 .
c. The bulkhead must also be designed to withstand the torque caused by this force.
Find the torque with respect to the x-axis, taking into account that different parts
of the force act at different displacements from the axis.
d. Find the center of pressure, at which the total force could be concentrated to
produce the same torque .
e. Find the centroid of the bulkhead. Is the center of pressure located at the centroid?
f. When the ship is floating, the water outside is expected to come up toy = 16 ft.
Assuming that the ship has a uniform cross section, the center of buoyancy of the
ship is located at the centroid of the part of the bulkhead that lies below the water
line. Find the center of buoyancy. (The center of mass of the ship must be below the
center of buoyancy, or the ship will capsize. Ships often carry ballast, consisting
of rock, metal scrap, lead, and so on, at the bottom for the specific purpose of
lowering the center of mass.)
Section 11-5: ForceExerted
bya Variable
Pressure
---Center
of Pressure 581
;;;;:;
4. Oil Truck Problem: An oil truck has a tank the shape of an elliptical
cylinder (Figure 11-Sd). The tank is 12 ft wide and 6 ft high. Your job
is to analyze the forces acting on the elliptical end of the tank.
a. Write an equation for the ellipse using axes with origin at the
ellipse's center.
b. Suppose that the tank is half full. Find the force acting on the
ellipse. Recall that pressure at a sample point is directly propor-
tional to the point's displacement below the liquid's surface. The
proportionality constant is the density of oil, 50 lb/ ft 3 . (If your Figure l l-5d
answer comes out negative, see if you can find your mistake!)
5. Airplane Wing Problem I: An airplane wing has the shape of the region bounded by the
graph of y = 60 cos -fux and the x-axis, where x and y are in ft .
a. Find the area of the wing.
b. Assume that the pressure pushing up on the wing when the plane is in flight is
constant in the y-d irection, but is directly proportional to the quantity ( 10 - Ix I) in
the x-direction. Find the total force acting on the wing .
c. In order for the plane to fly, each wing must lift 96 tons. What must the propor-
tionality constant equal?
y
6. A irplan e Wing Problem II: Suppose you have been hired by Fly-By- ight
Aircraft Corporation. You are to analyze the forces that will act on
the wings of a new plane. From the Design Department you find that,
looking from the top, the wing's shape (Figure ll-5 e) is the region
bounded by the graph of y = 100 - x 2 and the x-axis, where x and y
are in feet. The x-axis runs along the line where the wing joins the y = 100 - x 2
fuselage.
a. The Research Department finds that in normal flight the pressure
pushing up on the wings varies linearly with x. The pressure is 90
lb/ ft 2 at the y-axis, and 20 lb/ ft 2 at x = 10. Find the total force
acting on the wing.
b. Find the first moment of force with respect to the y-axis. X
c. For proper balance you need to know where in the x-direction the
force could be concentrated to produce the same first moment of Figure l l-5e
force. Find the x-coordinate of the center of pressure .
d. During a certain banking maneuver, the pressure pattern changes. It becomes
directly proportional to y, and does not vary in the x-direction. The pressure is
60 lb/ ft 2 at y = 50. Find the total force acting on the wing.
e. Find the first moment of force in 6d with respect to the x-axis.
f. The first moment of force in 6e measures the amount of n,visting needed to make
the plane bank. How far out on the wing could th e total force be concentrated and
produce the same first moment with respect to the x-axis?
582 Chapter
11:TheCalculus
ofVariable-Factor
Products
;re 1
7. Doub le-Integration Airplane Wing Problem: The place where an
airplane's wing joins the fuselage must be designed to withstand
the bending torque caused by air pressure acting on the wing.
r(x , t) (x,y)
The stiffness of this joint is measured by the second moment
of area of the joint with respect to a horizontal axis. Suppose
that a new design of plane is to have a cross section as shown in
Figure 11-Sf. The equation of the curve is
~ -==f 1~ "-/
'll""
dx ::::-,
4:
y = 0.25(x - 4) - (x - 4) 113 .
Figure l l -5f
Your mission is to calculate the second moment of area of this
region with respect to the x-axis.
a. Slice the region vertically into strips of width dx. Slice each strip horizontally,
forming rectangles of dimensions dx by dt. Find the moment of a strip by finding
the moment of a rectangle, then integrating from t = 0 to t = y . Recall that x and
dx will be constants with respect to this integration.
b . Find the total second moment of the region by adding up the moments of the
vertical strips and taking the limit (that is, by integrating with respect to x).
8. Double Integration Variable Pressure Problem: A variable pressure
acts on the region bounded by the curve y = e-x and the lines y = l
and x = ln 5. Find the force acting on this region if
a. The pressure is constant in the y-direction but varies directly
with the square of x.
b. The pressure is constant in the x-direction but varies inversely
with y .
X
c. The pressure varies both directly with the square of x and
inversely with y . To solve this problem, you must slice one way,
then slice the slice the other way. One integration gives dF . A Figure l l-5g
second integration gives f. See Figure 11-Sg for suggestions.
9. Problems 7 and 8 involve double integrals. Why do you suppose this name is used?
10. Floodgate Problem: A vertical floodgate at the bottom of a dam has
the shape of the region bounded by the graph of y = 5 tan 2 (rr / 8)x y
and the line y = 5, where x and y are in feet (Figure 11-Sh). The lake
behind the dam is filled with water to the level where y = 20 ft.
a. Find the area of the floodgate .
b . Find the total force acting on the floodgate.
c. The force acting on one side of the floodgate makes it difficult
to open. The force with which the equipment must pull up on
the gate equals the coefficient of friction between the gate and
the dam multiplied by the force of the water acting on the gate.
Experience shows that this force is about 10,000 lb. What does Figure l l -5h
the coefficient of friction equal?
Sec
tion 11-5: ForceExertedbya VariablePressure--Center
of Pressure 583
11-6 Other Variable-Factor Products
The area of a rectangular region equals its length times its width. For the region in
Quadrant I bounded by the graph of y = 4 - x 2 (Figure l l -6a), the length and width
vary. If you slice the region into horizontal strips, you are taking a small amount
of length, dy, in which the width is essentially constant. If you slice into vertical
strips, you are taking a small amount of width, dx, in which the length is essentially
constant. In the former case, dA = x dy. In the latter case, dA = y dx . Integrating
either one gives the exact area by adding the areas of the strips and taking the limit.
y y
4 4
Length
is
constant
X X
2 2
Figure l l -6a
Most applications of definite integration involve similar reasoning . A product has
one or both factors that vary. You take a small amount of one quantity (slice) in
which the other quantity is essentially constant, evaluate the "constant" quantity
at a sample point in the strip, and find the product. The trick is in deciding which
factor to slice . For example, volume = (cross-sectional area)(altitude). For plane
slices (Figure ll-6b left), you take a small amount of altitude, dy, in which the cross-
sectional area, rrx 2 , is constant. For cylindrical shells you take a small amount of
cross-sectional area, 2rrx dx, in which the altitude, y, is constant (Figure l l-6b, right).
y y Altitude is
Cross-sectional
area is "constant" "constant"
1----1r--- '\ (x, y) \
I
I
I
........ .........
X
____
. . .... .J1.....
____ _
I
Plane slices Cylindrical shells
Figure l l -6b
584 Chapter
11: TheCalcu
lusofVariable-Factor
Produc
ts
Slicing may be easier one way than another . Work = (force)(displacement) or
W = (F)(D). If the whole object is moved the same amount, you should take small
11 displacements, dD, in which the force is essentially constant, so dW = F dD. But
different parts of a fluid may move different amounts. So it is preferable to take
small amounts of force (that is, weight) for which the displacement is constant. In
this case, dW = D dF (Figure l l-6c).
In this section you will work problems in which one factor of a product varies.
You will be expected to read and interpret the definition of each physical quantity,
then translate it into the appropriate mathematics. It is the ability to see the
underlying similarities in seemingly different phenomena that will enable you to
make intelligent applications of mathematics.
Figure l l -6c
OBJECTIVE Given a real-world situation in which a product has a factor that varies, calculate the value
of th e pro duct.
SomeVariable-Factor
Products
Distance = (rate)(time)
Area = (width)(length)
Volume = (cross-sectional area)(height)
Work = (force)(displacement)
Force = (pressure)(area)
Mass = (density)(volume)
Moment = (displacement)n(mass, volume, area, force, and so on)
Worth of a region = (price)(area)
Problem Set 11-6
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
Q1. f xioo dx = _7_
Q2. J~1 tanxdx
= -?-
Q3. f lnxdx
= -? -
Q4. Write 2 sin x cos x in terms of a trigonometric function of 2x.
QS. Work =- ?-
Q6. If y = tan - 13x, then y ' = - ?- .
Ql. y = x 3 + 6x 2 has a point of inflection at x = -?- .
d
QB
. dx (sec zx) = _ 7_
Q9. If y = cos 3x, then d 2 y / dx 2 = - ?- .
Q10. What figure is the graph of 3x 2 - 7y 2 = 39?
Section 11-6:OtherVariable-Factor
Products 58 5
1. Heat Capacity Problem: The number of calories (heat, as energy) required to warm
a substance from temperature T 1 to temperature T2 equals the "heat capacity" of
the substance (calories per degree) times the change in temperature (T 2 - Ti), where
T is in degrees. Unfortunately, most substances have heat capacities that vary with
temperature . Assume that calculus foeride (a rare, tough substance!) has a heat capacity
given by
112
C = 10 + 0.3T ,
where C is in calories per degree and T is in degrees. How many calories would be
needed to warm a gram of calculus foeride from 100 to 900 ?
2. Phoebe's Speeding Problem: Phoebe is caught speeding. The fine is $3.00 per minute
for each mile per hour above the 5 5 mph speed limit. Since she was clocked at speeds
up to 64 mph during a 6-minute period, the judge fines her
(3.00)(time)(mph over 55) = (3.00)(6)(64 - 55) = $162.
Phoebe is good at calculus. She argues that her speed varied over the 6 minutes. It was
55 mph at t = 0 and t = 6, and was 64 mph only at t = 3. She figures that her speed,
v, was
V = 5 5 + 6t - t 2 .
a. Show that this equation gives the correct speeds at the times 0, 3, and 6 minutes.
b . What should Phoebe propose to the judge as a more reasonable fine?
3. Tunnel Problem : The amount of money it takes to dig a tunnel equals the length of
the tunnel times the cost per unit length. However, the cost per unit length increases
as the tunnel gets longer because of the expense of carrying workers and tools in, and
carrying dirt and rock out. Assume that the price per foot varies quadratically with the
number of feet, x, from the beginning of the tunnel.
a. Find the particular equation for the price per foot if the following prices are known:
x price
0 $500
100 $820
200 Sl180
b. Find the cost per foot for digging at a point 700 feet from the beginning of the
tunnel.
c. Find the total number of dollars for digging a tunnel 1000 feet long if the workers
start at one end and dig through to the other end.
d. How much money could be saved by starting the 1000-foot tunnel from both ends
and meeting in the middle?
4. Water Pipe Problem: The flow rate of water through a pipe (cubic inches per second)
equals the velocity of the water (in inches per second) times the cross-sectional area
of the pipe .
a. Show that velocity times cross-sectional area gives the right units for flow rate.
586 Chapter
11:TheCalculus
ofVariable-Factor
Products
b. In real pipes, the flow rate varies at different points across
the pipe, being a maximum at the center and dropping to zero
at the pipe walls (Figure l l-6d). Assume that the velocit y, v, Low flow
through any cross-section of a 4-inch diameter pipe is given rat e near --+,i ll \
wall s
by
High flow
2 rat e near
V = 4 - X ,
cent er
where x is the number of inches from the center of the pipe,
and v is in inches per second. Show that the velocity really is
a maximum at the center and zero at the pipe walls. Figure 11-6d
c. What is the flow rate in cubic inches per second?
d. How many gallons per minute (gpm) are flowing through the pipe? (There are 231
cubic inches in a gallon.)
e. How many gpm would be flowing through the pipe if all the water were moving at
the maximum velocity?
f. As far as the mathematics is concerned, this problem is identica l to some other
kind of problem you have worked. Which kind?
5. Wire-Pulling Problem: Yank Hardy tries to pull down a tree. He attaches one end of a
long wire to the tree, and the other end to the bumper of his Jeep. As he drives the Jeep
slowly away, the wire stretches tighter and tighter, and finally breaks . At first the force
increases linearly with x, the number of inches the wire stretches . At x = 2 inches the
wire "yields" (that is, begins to break). The table shows the forces to the point where
the wire breaks at x = 5.
x (inches) force (pounds)
0 0
0.5 150
1 300
1.5 450
2 600 (before yielding)
2 450 (after yielding)
2.5 470
3 440
3.5 420
4 410
4.5 390
5 330 (breaks)
a. Plot the graph of force versus x.
b. Describe the behavior of the force function at x = 2.
c. Find the work done in stretching the wire from x = 0 to x = 2.
d. Find the work done in stretching the wire from x = 2 to x = 5.
e. Find the total work done in breaking the wire.
f. Is it possible for a discontinuous function to be integrable7 Explain.
Section11-6:OtherVariable-Factor
Products 587
6. Variable Attract ion Problem: A solid paraboloid is formed by
rotating about the y-axis the region in Quadrant I bounded-by the
graph of y = 4 - x 2 (Figure l l -6e), where x and y are in centimeters .
The solid has a uniform density of k g/cm 3 .
a. Find the mass of the solid.
b. The solid is attracted by a force that is inversely proportional to
the square root of the distance from the base of the solid, and
dir ectly proportional to the mass. That is,
Force= (constant)(mass)(y - 112 ). X
But different parts of the solid are at different distances from
the base . By appropriate slicing, find the total force exerted on Figure 1 l -6e
this solid.
7. Moment of Inertia Problem: The second moment of mass of an object with
respect to an axis is defined to be the mass times the square of the distance between the
object and the axis. It is sometimes called the moment of inertia because it measures
how difficult it is to start or stop rotating the solid around the axis . Find the second
moment of mass of the solid in Problem 6 with respect to the y-axis.
8. Degree-Days Problem: Engineers who design heating and air conditioning systems
use a quantit y called degree-days to measure how much above or below normal the
weather has been. For example, if the temperature is 10 degrees above normal for 2
days, then the weather has been ( + 10) (2) = + 20 degree-days. If it is 30 degrees below
normal for half a day, the weather has been (- 30)(1 / 2) = - 15 degree-days. However,
the temperature varies throughout the day, so degree-days should be calculated by
calculus rather than by arithmetic. Suppose that one morning the temperature starts
out at normal, and 6 hours later has risen 20 above normal, and that any time D days
after the morning reading, the temperature, T, is
T = 20 sin2rrD.
a. Show that this equation gives the right values of T for times D = 0 and D = 1 / 4.
b. Find the number of degree-days between D = 0 and D = 1 / 4.
9. Rocket Car Problem: Iona Carr is building a rocket powered car that she plans to use in
racing. In this problem you are to help Iona figure out the car's speed, and the distance
it ,vill travel in a given time. The car, with Iona in it and a full load of fuel, will have a
mass of 2000 kilograms. When the engine is running it will develop 7000 N (newtons)
of thrust, which means the car will be pushed with a cons tant force of 7000 N. You
recall from physics that force = (mass)(acceleration), and 1 N is 1 kg-m/ sec 2 . However,
the car uses fuel at a rate of S kg/sec, so its mass is decr easing.
a. Write an equation expressing mass as a function of time.
b. Write an equation expressing acceleration as a function of time.
c. The answer to part b is a differential equation because acceleration is the derivative
of velocity. Solve the differential equation for velocity as a function of time if
v(O) = 0.
d. Twenty seconds after the car starts, how fast will it be going? How far will it have
gone?
588 Chapter
11:TheCalculus
ofVariable-Factor
Products
10. Field Worth Problem : Ann Aggie has a tract of land the shape of the y
"triangular" region in Quadrant I bounded by the y-axis and by the 4
graphs of y = 4 - x 2 and y = 4x - x2 (Figure l l -6f) where x and y
are in kilometers. The land's value per square kilometer is directly
proportional to its distance from the railroad tracks (along the y-axis), Tracks
being $200 thousand per square kilometer at the point farthest from
the tracks. To the nearest thousand dollars, what is the total worth
\
of the land? How much less is this than it would be if the entire tract
were worth $200,000 per square kilometer ?
11. Sinusoidal Land Tract Problem: A tract of land has the shape of the
region in Quadrant I under the graph of y = cos x (Figure l l -6g). Find
X
the total worth of the land if the worth per square unit is
a. constant in the y-direction, but directly proportional to x in the
x-direction. Figure l l-6f
b. constant in the x-direction, but directly proportional to y in the y
y-direction.
12. Painted Wall Problem: Calvin has a contract to paint the wall of a new
auditorium. The wall is the shape of the region under the graph of
y = 9 - x 2 , where x and y are in meters, and the x-axis runs along the
X
ground. Since it is harder to paint higher up he charges a price per
square meter that is directly proportional to the square of the distance
above the ground. At a point 2 m above the ground, he charges $12 Figure l l-6g
per square meter. What total amount will Calvin charge for the job?
13. City Land Value Problem: Suppose that you have been hired by the Tax Assessor's
Office in the town of Scorpion Gulch. You are to calculate the total worth of all land
within the city limits. The town is circular, with a radius of 3 kilometers. You find
that land is worth $10 million per square kilometer right at the center of town, and
$1 million per square kilometer at the edge of town. Find the total worth of the land
assuming that the price per square kilometer
a. varies linearly with the distance from the center of town.
b. varies exponentially with the distance from the center of town.
c. is given by the table of values below.
d. This problem is mathematically equivalent to another type of problem you have
worked. What kind of problem?
e. What real-world reason(s) can you think of to explain the pattern of the data in 13c?
distance (km) million $/km 2 distance (km) million $/ km 2
0 10 1.8 8
0.3 12 2.1 5
0.6 15 2.4 3
0.9 14 2.7 2
1.2 13 3 1
1.5 10
Section11-6:OtherVariable-Factor
Products 589
th~
14. D ivi ng Board Problem: Calvin sits on the end of a diving board p
(Figure l l -6h). His anatomy exerts a pressure on the diving board Variable
that is constant in the z-direction (perpendicular to the page), and
is equal to
Diving board
p = 150[(x - 8) 112 - 0.5(x - 8)]
X =8 X ,:,10
(in pounds per square foot) in the x-direction . The diving board is
2 feet wide in the z-direction . Figure l l -6h
a. To the nearest pound, how much does Calvin weigh?
b. What is the average pressure Calvin exerts in the interval between x = 8 and x = 107
c. To the nearest pound -foot, find Calvin's first moment of force with respect to the
yz-plane .
d. Calvin wishes to exert the same first moment by standing on tip-toes at some point
near the end of the board. Where should he stand?
15. Skewness Prob lem : Figure ll-6i shows the graphs of y
2
f(x)=9-x and g(x) =- x 3 - x 2 + 3x + 9.
The region under each graph has the same area, as you will show
in 15b, below. But the region under the g graph is "skewed" to the
right. In this problem you will calculate the "skewness," which is
used in statistics to measure how "unbalanced" a region is.
a. Show that the only x-intercepts of both functions f and g are
-3 and 3 (Figure ll -6i). X
-3 -2 -1 2 3
b. Show that the regions under the two graphs have equal area .
How do the properties of definite integrals betwe en symmetrical Figure l l -6i
limits help you calculate the area easily7
c. At what value of x in the interval [ -3, 3] do the maxima of the f and g graphs
occur?
d. Find the x-coordinate of the centroid of the region under the g graph. Recall that
this is the number x such that (x)(area) = first moment of area of the region with
respect to the y-axis .
e. True or false: "The centroid of the region is on a vertica l line through the maximum
on the g graph ."
f. True or false: "There is just as much area to the left of the centroid as there is to
the right."
g. The skewness of a region is defined to be the third moment of area of the region
with respect to a vertical line through the centroid. Calculate the skewness of the
region under the graph of function g.
h. Show that the skewness of the region under the parabola, function f, is equa l to
zero. Why is the word skewness appropriate in this case 7
i. Sketch the graph of a region whose skewness has the opposite sign from that un der
the graph of function g .
590 Chapte
r 11:TheCalculus
ofVariable-Factor Products
ff At f t t1ifM\M: Mii&H
"
16. Moment of Arc Length Problem : You have found moments of area,
y
mass, and volume. It is also possible to find moments of length.
4
Fi.gure l l-6j shows the arc of the parabola y = x 2 from x = 0 to
x = 2. The moment, dM y, of the arc dL with respect to the y -axis is
the length, dL, times its distance, x, from the y-axis. That is,
dMy = xdL.
a. Find dM y explicitly in terms of x and dx.
b. Find the total moment of the parabolic arc with respect to the
y-axis.
c. Find the length of the parabolic arc. X
2
d. Find the x-coordinate of the centroid of the parabolic arc . This
point would be the center of mass of a thin, uniform wire bent Figure 1 l -6i
in the shape of the arc.
e. Find the surface area of the paraboloid formed by rotating the arc about the y-axis .
f. What interesting thing do you notice about the integral in 16e?
17. Another Theorem of Pappus Problem: In Section 11-4 you learned the theorem of
Pappus for volumes. The theorem states the volume of a solid of revolution equals
the area of the region being rotated times the distance traveled by the centroid of that
region . There is a similar theorem for surface area.
Theorem:TheTheoremof Pappusfor Surfaces
The area, S, of a surface of revolution is given by
S = 2rr RL,
where L is the length of the curve being rotated, and R is the displacement from the axis of
rotation to the centroid of the curve, and the curve is not on both sides of the axis of rotation .
The quantity 2rrR is thus the distance the centroid travels as the curve rotates.
Demonstrate that this theorem is true for the paraboloid in l 6e.
18. Application of Pappus' Other Theorem: A toroidal surface (like an
inner tube) is formed by rotating a circle of radius r about an axis R-
(Figure l l -6k) R units from the center of the circle. It is hard to find
the surface area of the toroid directly by integration, but easy by
using the theorem of Pappus for surfaces. Find a formula for the
area of a toroidal surface in terms of rand R.
Figure 1 l-6k
Section11-6: OtherVariable-Factor
Products 59 1
11-7 Chapter Review and Test
In this chapter you have applied definite integration to prob lems involving a product
of two quantities , where one of the quantities is a variable. By now you should be
able to take any such situation, familiar or unfamiliar, and perform the appropriate
mathematics. The ability to see similarities among seemingly dissimilar phenomena
is the key to intelligent application of mathematics .
The Review Problems are numbered according to the sections of this chapter. The
Concepts Problems allow you to apply your knowledge to new situations. The
Chapter Test is more like a classroom test.
Review Problems
RO. Update your journal with what you've learned since the last entry. You should include
such things as those listed here .
The one most important thing you have learned in studying Chapter 11
Which boxes you have been working on in the "define, understand, do, apply" table
Physical quantities that can be calculated as products, such as work, moment, and
mass
Centroid, center of mass, center of gravity, center of pressure, and so on
Any techniques or ideas about calculus that are still unclear
Rl. Work Prob lem: In Miss Calculate's first-grade classroom, Manuel
Dexterity drags Bob Tail across the floor. Manuel pulls hard at
first, then eases off. The force he exerts (Figure 11-?a) is given by
F = 30e - 0 2" ,
where F is in pounds and x is Bob's displacement in feet from 10
the starting point. The work Manuel does is the force times the
disp lacement. Find the number of foot-pounds of work he does in Figure l 1-7a
dragging Bob from x = 0 to x = 10 ft.
R2. a. Magnet Problem: A magnet repels another magnet with a force
inversely proportional to the square of their distance apart.
That is, F = k / x 2 , where Fis in pounds and xis in inches . Find
the work done in moving the magnets from 3 in . apart to 1 in.
apart. -- -,, --
II
---:~--
-1 10 in .
b. Conical Cup Problem: Phil LaRupp puts a 10 in . long straw into II
7 in.
a conical cup filled to the top with root beer (Figure 11-7b).
The cup has a top diameter of 6 in ., and an altitude of 7 in .
The root beer has a density of 0.036 lb/ in 3 . How much work j
will Phil do in raising all the liquid in the cup to the top of the
straw? Figure l l -7b
R3. Variab le Density Problem: The region in Quadrant I bounded by the graph of y = 8 - x 3
is rotated about the y-axis to form a solid. Find the mass of the solid if the density
a. is constant in the radial direction but equal to ky in the y-direction, where k is a
proportionality constant.
b . is constant in the axial direction but equals e" in the radial direction .
592 Chapter
11: TheCalculus
ofVariable-
Factor Produ
cts
srn RI 'If
R4. a. Triangle Centroid Problem: Figure ll-7c shows a triangle of y
base b and altitude h. Write an equation for the width of the
triangle in terms of y, the distance from the base to a sample
point . Find the first moment of area of the region inside the
triangle with respect to its base. Show that the centroid is
one-third of the distance from the base to the opposite vertex.
b. Second Moment of Volume Problem: The region under the
graph of y = ex from x = 0 to x = l is rotated about the y-axis X
to form a solid. Find the second moment of volume of the solid b
with respect to the y -axis.
Figure l l-7c
RS. Wind Force Prob lem: A well-known tower has the shape of a
slender pyramid (Figure 11-7d). The base of the pyramid, at
ground level, is a square of length 150 ft. The building is 400 ft
tall. When the wind blows, the pressure acting on the triangular
face of the building is grea ter at the top than at the bottom
because the wind speed increases with altitude. Assume that the
pressure due to the wind is given by
p = 200(1 - e - OOlY),
where p is in pounds per square foot, and y is the height in feet
above the ground. Assume also that the face of the building is a
vertical triang le of base 150 ft and altitude 400 ft. Calculate the
total force of the wind acting on that face .
R6. Oil Well Prob lem: Suppose you work for a company that plans to
drill a well to a depth of 50,000 ft, farther than anyone has ever
drilled before. Your job is to estimate the cost of drilling. From
historical records you find that it costs about $30 per foot to drill Figure l 1-7d
at the surface, and about $50 per foot at a depth of 10,000 ft.
a. Assume that the number of dollars per foot varies exponentially with depth . Write
the particular equation expressing cost per foot in terms of depth .
b . The total cost of the well will be the cost per foot times the number of feet. Since
the cost per foot varies, you realize that this is a job for calculus! Your boss needs
the estimated cost of the well. What are you going to tell him?
Concepts Problems
Cl. Cubic Parabo la Region Problem: The following problems concern a region in the
xy-plane of an xyz-coordinate system . The region is bounded by the graphs of
y = x3, y = 8, and x = 0 (Figure ll-7e).
a. Find the area of the region .
b. Find the first moment of area of the region with respect to
i. The x-axis.
ti . The y-axis.
c. Find the centroid of the region.
Section 11-7:Chapter
Review
andTest 593
d. Find the volume of the solid generated by rotating the region +Y__ _, Y = B
i. About the x-axis .
ii. About the y-a)ds.
iii. About the line x = 3.
e. Show that for each of the solids in Cld, the volume is equal to the
area of the region times the distance traveled by the centroid of the X=O
region as it rotates to form the solid.
f. Find the first moment of volume with respect to the xz-plane for
the solid in C1d, part ii.
g. Find the centroid of the solid in Cld, part ii.
h . Does the y-coordinate of the centroid of the solid in Cld, part ii X
equal the y-coordinate of the centroid of the region?
i. Find the mass of the solid in Cld, part ii if the density varies directly Fig ure l 1-7 e
with the square of the distance from the y-axis.
j . Find the second moment of mass of the solid in Cld, part ii with respect to the
y-axis.
k. Find the force acting perpendicular to the region due to a pressure equal to (3 - x)
in the x-direction (and constant in the y-direction).
1. Find the work done by moving the region from z = 1 to z = 3. Assume that the
force in Clk acts on the region when it is at z = 1, and that the force varies
inversely with the square of z as the region moves in the z-direction, and acts in
the positive z-direction.
m. Suppose that the object in Cld, part ii is made of a substance with a heat
capacity of 0.3 cal/g/ C and a density of 5.8 g/ cm3, and that its dimensions are in
centimeters. Suppose also that the object is warmed from 0C to a temperature
that is constant in the radial direction but is given by T = 10 - y in the a)dal
direction . Find the amount of heat needed to cause this temperature change .
C2. Moment vs. Volume Problem: Show that finding the first moment of the area of a
region with respect to the y-axis is mathematically equivalent to finding the volume
by cylindrical shells for the solid formed by rotating that region about the y-axis,
provided that the region is not on both sides of the axis of rotation .
C3. Parabo loid Moment Conjecture Prob lem: A solid paraboloid is
forme d by rotating about the y -axis the region in Quadrant I
under the graph of y = 9 - x 2 . Show that the first moment of
volume of the solid with respect to the plane of its base equals y
the second moment of volume of the solid with respect to its
axis . Does this property hold in general for any solid paraboloid? y + e.y
y
Justify your answer.
C4. Infinitesimals of Higher Order: X
a. Figure ll-7f shows a lower fuemann sum for y = mx, where a x x+dx b
m * 0. The area of each strip is dA :::::y dx, the area of the
rectangle. The length of each piece of graph is dL :::::dx . Both It works for area,
appro)dmations become exact as t,..x approaches zero. Show but not for length.
that on integrating from a to b, dA :::::y dx gives the exact area Figure l l -7f
of the region, but dL :::::dx does not give the exact length .
594 Chapter
11:TheCalculus
ofVar
iable-FactorProducts
b. Figure ll-7g shows the cone formed by rotating about the
x-axis the graph of y = mx(m =1=0) from x = 0 to x = h. Plane
y + t.y
sections cut the cone into frustums of volume dV "' rry 2 dx. y
Each frustum has area dS "' 2rry dx. Both approximations
become exact as t::,.xapproaches zero . Show that on integrating
from O to h, dV "' rry 2 dx gives the exact volume of the cone,
but dS "' 2rry dx does not give the exact area.
c. Find the exact area of a strip in Figure ll -7f and the exact
volume of a frustum in Figure l l-7g. [The volume of a frustum
It works for volume ,
of altitude dx is V = (rr / 3) (R 2 + Rr + r 2 ) (dx ), where R is the but not for surface area .
larger radius and r is the smaller radius of the frustum.)
Figure l 1-7 g
d. A quantity such as 2rry dx that approach es zero as t::,.xapproaches
zero is called a first-order infinit esimal. A quantity such as 0.5!::,.ydx
that is the product of two or more first-order infinitesimals is called a higher-order
infinites imal. Show that the approximations dA "' y dx and dV "' rr y 2 dx differ
from the exact values in C4c only by infinitesimals of higher order .
e. You recall that the differential of arc length is dL = .j dx 2 + dy 2 . The approximation
dL "' dx = -/dx2leaves out the first-order infinitesimal .j dy 2 . Make a conjecture
about how accurate a differential of a quantity must be so that it yields the exact
value when it is integrated.
f. The second-order infinitesimal O.St::,.ydx appears in the exact value of the area
of the strip in Figure 1f-7f in C4d. The limit of the Riemann sum of such a
higher-order infinitesimal equals zero. Give a reason for each step in this example
of that statement.
I 0.St::,.ydx = 0.5!::,.yI dx
= 0.5!::,.y(b - a)
.. limIO .St::,.ydx= O.S(O)(b-a )=0
L'>x- 0
Property:Infinitesimals
ol HigherOrder
If dQ "" t.Q leaves out only infinitesimals of higher order , then
J:dQ is exactly equal to Q.
Chapter Test
Tl. A packing case is dragged across the floor from x = 0 to x = 10 ft . As it moves it
becomes damaged, causing the force needed to move it to increase as shown in the
table. Find the exponential function that best fits these data. Use the function to find
the total amount of work done .
X forc e (lb)
0 30
2 34
4 38
6 43
8 49
10 55
Section 11-7: Chapter
Review
andTest 59 5
t~ vork
_.:re
T2. For th e region und er th e graph of y = ex from x = 0 to x = 2, where x and y are in
inches,
a. Find the first moment of area with respect to th e y-axis. Tell the units in your
answ er.
b. Find the second mom ent of area with respect to th e y-axis. Tell the units in your
answer .
c. Find th e x-coordinate of the centroid of th e region . Tell the units in your answer.
T3. A solid is formed by rotating about the x-axis the region und er the graph of y = x 1 12
from x = 0 to x = 16 where x and y are in centimeters. Its density is constant axially
and equal to 3y radially (in grams per cubic centimeter). Find its mass .
T4. A trough 8 ft deep and 4 ft across has an end the shape of the
region above the graph of y = Ix 3 I (Figure 11-7h).
a. Find the tot al force acting on the end of the trough when it is
full of water of densit y 62.4 lb/ ft 3 .
b . Find the cent er of pressur e, where the force could be concen -
trat ed to produ ce the same mom ent. X
TS. Theat er in the Round Problem: A round th eater is to be built with
the stag e at the center . The seats will be in circular rows starting Figure l l -7h
at r = 30 ft from th e center and ending at r = b ft (Figure ll-7i) .
a. The value of th e seating area for any given performance is
given by v = 150 /r (in dollars per squar e foot). Find the total
value of the seating area in terms of th e outer radius, b.
b. How bi g must th e th eater be in order to receive $60,000 for a
performance ?
Figure l l-7i
596 Chapter
11: TheCalc
ulusofVariable
-Factor Products
CHAPTER
12 The Calculus of Functions
Defined by Power Series
rescr1pt1oas
If you take regular doses of a medication, the amount in your system
is the sum of what remains from the series of the doses you have
taken. The limit of that series as the number of doses becomes large
is important to know for determining long-term effects of the
medication. Such limits can be found by calculus techniques.
597
1 ftl Mathematical Overview
How does a calculator find sines and logs when all it can do is add
and multiply? In Chapter 12 you will see that these transcendental
functions can be calculated as accurately as you please using "infinite
polynomials" called power series. You will study these series in
four ways .
Graphically The logo at the top of each even- y
Power ;
numbered page of this chapter
shows that the first few terms of
a power series can fit the sine
function close to x = 0.
"7
' ,,. X
Numerically sin 0.6 = 0.56462473 ...
n series for sin 0.6
0 0.6
1 0.564
2 0.564648
3 0.564642445 .. .
4 0.564642473 .. .
1 1
Algebraically sin x = x - 3! x 3 + 51x 5 - .. . , the Maclaurin series for sine.
Verbally Perhaps the most surprising thing we learned about power series is
that sometimes they converge as more and more terms are added,
and sometimes they don't . For each power series there is an interval
ofx-values for which the series converges.
598
12-1 Introduction to Power Series
f( x)
Suppose that f(x) = 6/ (1 - x ) (Figure 12-la). If you divide 1 - x into 6, you get a
"polynomial" that continues forever!
P(x) = 6 + 6x + 6x 2 + 6x 3 + 6x 4 + 6x 5 + ...
X
The result is called a power series. The word series indicat es that th ere is an infinite
number of terms. The word power tells us that each term contains a pow er of x.
Figure 12-1a
OBJECTIVE Find values of P(x) for a given power series, and compare the results with the correspond-
ing values of the function from which the power series originates.
The following problem set is designed to let you work toward this objective either
on your own or with your study group following your test on Chapter 11.
Exploratory Problem Set 1 2-1
Let f (x ) = 6/ (1 - x ) and let P(x) = 6 + 6x + 6x 2 + 6x 3 + 6x 4 + 6x 5 + .
l. On the same screen, plot f(x) and the polynomial function P5 (x) (the six terms of P (x )
through 6x 5 ). Sketch the results. Use a friendly x-window from about [ -2 , 2], and a
y-window of [- 100, 100).
2. On the same screen as Problem 1, plot the polynomial function P6 (x) . From th e graphs
of P5 and P6 , determine the x-values for which P (x ) is close to f (x ) and th e x-values
for which the graph of P bears little or no resemblance to the graph of f.
3. Show that P6 (0.5) is closer to f (0.5) than P5 (0.5 ) is, but that P6 (2) is not clos er to f (2)
than P5 (2) is.
4. If the limit of the sum of a series as you add more and more terms is equal to the
corresponding value of f(x), then the series is said to converge to f (x ). Otherwise,
the series is said to diverge (or to converge to some other number) . Make a conjecture
about the interval of values of x for which the series P (x ) = 6 + 6x + 6x 2 + 6x 3 + 6x4 +
6x 5 + ... converges to f (x) = 6/ (1 - x) .
5. Write the values of P0 (1), P 1 (1), ... , P4(1) . Write the values of P0 (- 1), Pi(- 1), .. . , P4(- 1).
Tell why each series diverges. Does this result affect your conjecture in Problem 4?
6. By how much do P5 (0.5) and P5 (- 0.5) differ from the respe ctive valu es of f (0.5) and
f (- 0.5)? How do these differences compare with 6x 6 , the first term of th e series that
is left out of the sum ?
7. Each term of P(x) after the first is equal to the preceding term multipli ed by the sam e
number, x, in this case . What name is given to this type of series? What nam e is given
to the multiplier x?
Section12-1: Introduction
to PowerSeries 599
12-2 Geometric Sequences and Series as
Mathematical Models
In Problem Set 12-1, you saw how the rational function f(x) = 6/ (1 - x) could
be expanded as a series, 6 + 6x + 6x 2 + 6x 3 + . Any term after the first can be
generated by multiplying the preceding term by the same number, x in this case .
Thus the series fits the definition of geometric series you may have learned in
algebra . In this section you will reverse th e process, and see how a geometric series
can be represented, at least in some instances, as a rational function. By so doing
you will be able to analyze some functions in the real world in which the function
values change discretely (by jumps) rather than continuously .
OBJECTIVE Given a fun ction defined by a geometr ic series, tell wheth er or not the series can be writt en
as a rational algebraic function, and if so, Tit e an equation for that tar get function.
Background
A geometr ic series is defined to be a series t1 + t2 + t 3 + t4 + for which there is
a constant, r, called the common ratio such that t11 = rt,,_1 for any integer n > l.
The numbers t 1 , t 2 , t3 , . .. are called terms of the series, henc e the letter "t." The
variable n is called the term index. If n starts at 1, the term index is the same as the
term number. The nth partial sum, S11, of a geometric seri es is the indicated sum
of the first n terms. By clever algebra it is possibl e to derive a closed formula (no
ellipsis, ... ) for S 11 as a function of the first term, t 1 , and the common ratio, r .
S,, = ti + t2 + t3 + t4 + + t n- 1 + t ,,
Each term is r times the
pr eceding term.
Multipl y both sides of
th e equation by r.
Subtra ct the third equation
from th e second on e. The
middle terms "telescope."
1 - r 11
S,, = t1 - --
1- r
A Convergent
Geometric
Series-DrugDosage
Suppose that a person takes 500 mg of vitamin C every 8 hours . Assume that at the
end of any 8-hr period, only 60% of the vitamin C present at the beginning of the
period remains in the person's system . At th e end of the first 8-hr period, only 300
mg of the original 500 remains. After the second dos e, the amount jumps to 800 mg .
Immediately after the nth dose, the amount of vitamin C remaining in the system is
the sum of the 500 mg from the last dose and the remains of each previous dose .
S11 = 500 + 300 + 180 + 108 + 180 is 60% of 300; 108 is 60% of 180, and so forth .
2
= 500 + 500(0.6) + 500(0 .6 ) + 500(0.6 3 ) + + 500(0.6 11- 1
)
600 Chapter
12:TheCalculus Defined by Power Series
of Functions
This series is geometric, with first term 500 and common ratio 0.6. After 10 doses,
1 - 0.6 10
S10 = 500 l _ _ = 1242.441 ...
06
The person would have only about 1240 mg of vitamin C in his or her system, in
spite of having taken a total of 5000 mg.
If the person continues taking the 500 mg doses for a long time, does the amount in
the body become arbitrarily high? To find out, it is instructive to look at the partial
sums graphically and numerically . As shown in Figure l 2-2a, the amount of vitamin
C rapidly levels off toward 1250. As shown in the accompanying table, the partial
sums stay the same for more and more decimal places . For both of these reasons
the sequence of partial sums converges to 1250 . As you saw in Section 12-1, if the
sequence of partial sums converges, then the series converges also .
Sn
-----1250 ...... .....
.-; - ,...
~.-.~ -
!
n
25
26
Sn
1249.996446
1249.997867
.. .
...
27 1249.998 720 ...
. 28 1249.999232 .. .
n 29 1249.999539 ...
10 30 1249.999723 ...
31 1249.999834 ...
Figure 12-2a 32 1249 .999900 ...
From the formula for Sn, you can tell algebraica lly that the series converges to 1250.
Take the limit of Sn as n approaches infinity .
6
lim Sn = lim 500 l - ~ : = 500 ____l_O Because 0.6" approaches zero.
n -oo n -oo 1- . 1 - .6
= 1250
So th e amount of vitamin C in the person's system never exceeds 1250 mg, no
matter how long the dosage is continued.
A geometric series will converge if the common ratio r is a proper fraction (that is,
Ir I < 1). This is because the term rn in the formula for S,, approaches zero as a limit
as n approaches infinity . The number S to which the series converges is given by
1 - rn 1
S = lim t1 -- = ti -- .
n -oo 1- r 1- r
A Divergent
Geometric
Sequence-Compound
Interest
In Chapter 7, you saw that an exponential function can be used as a mathematical
model for invested money if the interest is compounded continuous ly. A sequence is
more appropriate if the interest is compounded at discrete intervals such as once a
day, once a month, or once a quarter. Suppose you invest $500 in a savings account
that pays 6% per year interest compounded quarterly. For the first 3 months you
have only the initial $500 . Then $7.50 is added to the account, 1.5% of the $500 (a
fourth of 6%),and you have $507.50 for the next 3 months. At the end of the second
quarter the account increases by $7.6125, which is 1.5% of the $507.50 .
Section 12-2: Geometr
ic Sequences
andSeriesas MathematicalMode
ls 601
To find a pattern in the amounts, observe how the $5 07 .50 can b e calculat ed.
500 + 500(0 .015) = 500(1 + 0.015) = 500 (1.01 5) After the first quarter.
Repeating the com putation for the second quarter without simp lifyin g 500(1.0 15 )
gives
500(1.015) + 500( 1.01 5) (0 .01 5)
= 500(1.015)(1 + 0.015) = 500(1.015)2 . Second quart er.
The amounts are terms in a geometric sequence with first term 500 and a common
ratio 1.015.
500, 500(1.015), 500( 1.01 5) 2 , 500(1.015) 3 , 500( 1.01 5)\ .. . Dollarsint heaccoun t.
0 1 2 3 4 Quarters
The expone nt of 1.015 equals the number of quarters. After 10 years the amount
would be
t 40 = 500( 1.015 )40 = 907.0092 .. . :::::$907 .01 Why 40, w hy not 10?)
The amo unt of int erest earned in the 10 years would be $407.01 , th e differ ence
between the $907 .01 an d th e initial in ves tm ent of $500 . In this instanc e it is more
convenient to start n at O instead of 1 so that th e term index will equal the number
of qu arters elapsed.
Note that the se qu ence diverges because the common ratio is greater than l.
Thi s fact can be useful in the real wor ld . For instanc e, it alerts bank ers to the
consequences of money left in dormant accounts. If th e $500 had been in veste d by
George Washin gton the year h e died, 1799, his heirs could claim $74,439 ,56 1.08
in 1999!
WordsAssociated
withSequences
andSeries
The follovving bo xes summa ri ze pertinent definitions and prop er ties conc ernin g
sequences and series.
Definitions:Vocabulary
Associatedwith SequencesandSeries
A sequence is an infinite ordered set of nwnb ers .
Exampl e: 2, 3, 5, 7, 11, 13, 17, 19 , 23 , . .. The sequence of prim es.
A series is th e indicated swn of the terms of a sequence.
Exampl e: 2 + 3 + 5 + 7 + 11 + 13 + 17 + 19 + 23 + The series of primes.
The terms of a sequence or series, t 1, t2, t3, ... , tn, . .. , are the numbers that appear in the
sequence or series.
The term index is the variable integer subscript n used to calculate the term value. If n
starts at l, the term index is also equal to the term number.
A partial sum of a series is the sum of a finite number of terms in the series.
Example: S4 = 2 + 3 + 5 + 7 = 1 7 Fourth partial sum of the series of primes.
602 Chapter
12: TheCalculus
of Funct byPowerSeries
ionsDefined
Definition
s
The sequence of partial sums for a series is the sequence whose terms are the first partial
sum, the second partial sum, the third partial sum, and so on.
Example: For the series of primes, above, the sequence of partial sums is
S1,S2,S3,S4,Ss,S5,S7,Ss, .. . = 2, 5, 10, 17, 28,41, 58, 77, ....
A sequence converges to L if its nth term approaches a finite limit L as n approaches
infinity. If the sequence does not converge, it is said to diverge.
Examp 1es.. l, 1 2l , 1 43 , 187 , 1 TI;,
15
... converges to 2 .
2, 3, 5, 7, 11, 13, ... (the sequence of primes) diverges.
1, 0, 1, 0, 1, 0, l, . . . diverges by oscillation.
A ser ies converges if and only if its sequence of partial sums converges.
Example: 1 + + + + t5 + converges since its sequenc e of partial sums is
f,
1, 1 , 1 , 1 1 ~ .... , which converges to 2.
A geo metric series is a series for which each term after the first term is given by
t,, = r tn- l for some constant r called the common ratio.
Propertiesof GeometricSeries
The nth partial sum of a geometric series is given by
1 - rn
s,,= ti .-- .
1- r
A geometric series converges if Ir I < 1.
The number to which a convergent geometric series converges is S = t 1 - --1 .r .
1
Problem Set 1 2-2
DoThese
Quickly
The following problems are intended to refresh your skills . You should be able to work
all ten problems in less than five minutes .
Q1. Definition : Lis the limit of f (x) as x approaches infinity if and only if -?-.
02. "If g(x) = Jf(x) dx, then I:f(x) dx = g(b) - g(a)" is a stat ement of - ?-.
Q3. (d / dx) J; f(t) dt = f (x ) is a stat ement of - ?- .
04. ... such that f ' (c ) = [f (b ) - f (a ) ] I (b - a ) is part of the conclusion of -?-.
QS. Instantaneous rate of chang e of a function is the physical meaning of -?-.
Q6. (d / dx)(xcosx) = - ?-
Ql. Jxcosxdx = -?-
QB. The differential of area in polar coordinates is d A = - ?- .
Section12-2:Geometri
c Sequences
andSeries as Mathematical
Models 603
Q9. What function has the graph shown in Figure 12-2b 7
QlO. What function has the graph shown in Figure 12-2c?
g(x)
Figure l 2-2b Figure l 2-2c
1. Write the first few terms of the geometric series with first term 200 and common ratio
- 0.6. Write the corresponding partial sums. Plot the graph of the partial sums and
sketch the result. To what number does the series converge? Show this number on
the graph. What is the first value of the term number, n, for which the partial sum is
within 0.0001 of the limit? What can you say about the proximity of S,, to the limit for
greater values of n?
2. Write the first few terms of the geometric series with first term 30 and common ratio
1.1. Write the corresponding partial sums. Give numerical and graphical evidence
that the series diverges. For instance, what does S100 equal? What meaning does the
algebraic formula S = t 1 I ( 1 - r) have for this series?
3. Drug Dosage Problem: Suppose you have been
hired by a pharmaceutical company. One of
your projects is to determine the recommended
dose rates for a new nasal allergy spray. For
the drug to be effective but safe, the amount in
the person's body should be between 50 and 80
g (micrograms). Research has shown that each
6 hours the amount of drug remaining in the
body is 80% of what it was at the beginning of
the 6-hr period . Each puff of the spray delivers
7 g of the drug.
a. Suppose that a person uses one puff of the spray each 6 hours. Write a geometric
series for the amount of drug remaining in the person's system after n puffs . How
soon would the amount in the person's system exceed 50 g?
Would the amount ever exceed 80 g?
b . Describe the effects of using two, three, or four puffs each 6 hours.
For instance, how soon will be amount exceed 50 g? Will it ever
exceed 80 g?
c. Using the spray each 6 hours is inconvenient. Could several puffs of
the spray be taken twice a day or just once a day and have the levels
stay between 50 and 80 g? Give evidence to support your answer.
4. Inscribed Squares Problem: Figure 12-2d shows an outer square of ----4 cm---+-
side 4 cm. The midpoints of the sides of the square are the vertices
for an inscribed square. More squares are inscribed using the same Figure l 2-2d
pattern, forever!
60 4 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
5t
a. Show that the perimeters of the squares form a geometric sequence.
b . The total perimeter of all the squares is a geometric series . Find the tenth partial
sum of the series.
c. Does the series for the total perimeter converge, or does the total perimeter
approach infinity?
d. The sum of the areas of the squares is also a geometric series. Does this series
converge, or does the sum of the areas become infinite?
5. Compound Interest Problem: Meg A. Buck invests a million dollars in a certificate of
deposit (CD) that earns 9% interest a year, compounded once a month.
a. Write a geometric sequence for the amounts the CD is worth after 0, 1, 2, and 3 mo.
b. How much will the CD be worth at the end of the first year? How much interest will
have been earned?
c. How do you explain the fact that after 12 mo the term index is 12, but there are 13
terms in the sequence?
d. The annual percentage rate (APR) an investment earns is the amount of interest
for 1 yr expressed as a percentage of the worth of the investment at the beginning
of the year. What is the APR for Meg's CD?
e. When will Meg's CD be worth 2 million dollars?
6. Regular Deposits Problem: Ernest Lee invests $100 a month in an indivi dual ret ir ement
arrangement (IRA). The interest rate is 10.8% per year, compoun ded monthly . After
0 months, Ernest has only the first $100 in the IRA. After 1 month, he has $200, plus
interest on the first $100 . After n months, he has $100 invested the last month, plus
$100 with 1 month's interest for the preceding month, plus $100 with 2 months'
interest, and so forth.
a. Write the amount Ernest has at the end of 5 months as a partial sum of a geometric
series.
b. The 5 (months) in 6a is the term index. How many terms are in the partial sum?
Why is the number of terms not equal to the term index?
c. How much will Ernest have after 10 years? How much of this is principal and how
much is interest?
7. Bounc ing Ball Problem: A superball is catapulted from floor Heig ht (ft)
10
level. It rises to 10 ft above the floor then starts back down.
On the next bounce it rises to 9 ft above the floor. On each
subsequent bounce it rises to 90% of the maximum height on
the previous bounce (Figure 12-2e).
a. The ball travels 20 ft vertically before the first bounce.
Write th e first few terms of the sequence of distances the
ball travels between bounces.
b . Calculate the fourth partial sum of the series of distances
corresponding to the sequence in 7a. Time (sec)
c. To what number does the series in 7b converge? What
does this fact imply for the total distance the ball travels Figure 12-2e
as it comes to rest?
Section122: GeometricSequences
and Seriesas Mathemat
icalModels 605
d. From physics you learn that the distanc e an object drops from rest under the
influence of gravity is d = (l / 2)gt 2 . If dis in feet and tis in seconds, then g"" 32.2
ft/ sec 2 . The time taken for an up-and-down cycle is twice the time to fall from
a high point. How long does it take the ball to make th e 20-ft first up-and-down
cycle? How lon g does it take to make th e 18-ft second cycle?
e. According to this math ematic al model the ball makes an infinite number of bounc es
before it comes to rest. Does this infinite number of bounces take an infinite length
of tim e, or does the model predict that the ball eventually comes to rest ? Explain.
8. Snowflak e Curve Problem: A figure called the snowflake curve is generated as
shown in Figure 12-2f. An equilateral triangle has the one-third points marked
on each side. For the first iteration the middle one-third of each side is erased 9 cm 9 cm
and tvvo line segments equal to the length of th e erased segment are adde d to
form sides of sma ller equ ilatera l triangles. In the second iteration the process is
repeated. Each old segment is replaced with four new segments, each of which is
one-thir d as lon g as the segment they replace. The snowflake curve is the figur e 9c m
that results from takin g the limit as th e numb er of it erat ions approaches infinity.
This limit was first considere d by Helge von Koch in 1904.
a. Suppose that th e pre-image (the original triangle) has sides 9 cm long . Write
the total length of the curve at the first, second, and third iterations . How
can the length at one particular it era tion be generate d from the length at the
previous iteration?
b. Does the sequence of total lengt hs in 8a converge or diverge? What does the
answer tell you about the total length of the snowflake curve ? Surprising?
c. The area enclosed by each iteration is a partial sum of a geometric series.
Writ e th e first few terms of this series. Does th e series converge or diverge?
If it converges, tell the limit to which it converg es. If it diverges, tell how you
know it diverges .
d. The snowflake curve is a classic exampl e of a fractal curve. It is so "fractured"
that it is more than one-dimensional but less than two-dimensional. Its dimen-
sion is a "fraction" equa l to about 1.26. The curve is continuous everywher e
but differentiable nowhere. For mor e about such curves , see, for example, Figure l 2-2f
Benoit Mend elbro t's book, The Fractal Geometry of Nature, published by W.H.
Freema n and Company in 1983.
*9. Derivatives of a Geometric Series: In this problem you will consider the derivatives of
the geome tri c series from Problem Set 12-1 and see how they relate to the rational
algebr aic function from which the series was derived.
Let P(x) = 6 + 6x + 6x 2 + 6x 3 + 6x4 + 6x 5 + .
Let f(x) = 6/ (1 - x) .
Assume that the derivatives of a ser ies can be found by differentiating each term.
Write series for P' (x), P" (x), and P"' (x) . (P"' (x) is the third derivative, the deriva-
tive of P" (x ) .] Show that P' (O),P"(O), and P "' (O) equal the corresponding values
of f'(O),f " (O), and f "' (O). How do you suppose p (nl (O) relates to f (nl(O), the nth
derivative ?
*This problem prepares you for the next sect ion.
606 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
12-3 Power Series for an Exponential Function
In Section 12-1, you saw that the function
100 Y
6
f (x) = 1 - X
,' -1 could be written as a power series,
I
,'P5 P(x) = 6 + 6x + 6x 2 + 6x 3 + 6x 4 + ... .
I
I
- 100
Figure 12-3a shows that for values of x between - 1 and 1, the more terms of the
Figure 12-3a series used, the closer the P graph is to the f graph .
Values of f(x) = 6/ (1 - x) can be calculated directly. Values of the elementary
transcendental functions cannot be calculated directly using only a finite number of
operations of algebra, namely, +, -, x , and+. Fortunately, it is possible to express
many of these functions as power series. Partial sums of these series can be used to
calculate sin x, ln x, ex, and so forth to as many decimal places as you need. In this
section you will derive a power series for f(x) = 5e 2x .
OBJECTIVE Givena particular exponential function, derive a power series that fits the function for
values of x close to zero.
In the following problem set you will accomplish this objective.
Problem Set 1 2-3
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
10 in less than 5 minutes.
Ql. Write 1/ 3 as a repeating decimal.
Q2. Write 4 / 9 as a repeating decimal.
Q3. Write 0.6666 . .. (repeating) as a ratio of two integers.
Q4. Write 0.4 + 0.04 + 0.004 + 0.0004 + as a ratio of two integers.
QS. Write the next term of the arithmetic series 1 + 5 + 9 + .
Q6. Write the next term of the geometric series 1 + 5 + 2 5 + .
Ql. First moment of mass = times
- 7- -7- .
QB. Center of volume is called - ?- .
Q9. Do the integra tion : Jx - 1 dx
Q10. Differentiate: y = sec 2x
1. Let f (x) = 5e 2 x . Find f' (x), f " (x), f' " (x), and f <4 l (x), the first, second, third, and fourth
derivatives of f(x).
2. Function f is locally linear at x = 0. Let P 1 be the linear function P 1 (x) = c 0 + c 1 x that
best fits f (x) at x = 0. (co and c1 are used for the constants rather than b and m so that
Sect
ion 12-3:PowerSeriesfor an Exponen
tial Function 607
you will see a pattern later.) Find the values of c0 and c 1 that make P 1 (0) = f (0) and
P;(O) = f'(O) .
3. Function f is also locally quadratic at x = 0. That is, there is a quadratic function
P2 (x) = c0 + c1x + c2x2 that best fits f(x) at x = 0. Find the values of the constants
co, c1, and Cz that make P2(0) = f(O), P; (O) = f ' (O), and P;' (O) = f " (O). How do the
values of c0 and c 1 compare to the corresponding values for the linear function P 1?
4. Function f is also locally cubic and locally quartic at x = 0. Find equations for the
cubic and quartic functions,
P3(X) = Co+ C1X+ CzX2 + C3X3 and
P.J(x) = Co+ C1X+ CzX2 + C3X3 + C.JX-l,
that best fit f(x) at x = 0. For these equations, P3 (0 ) and P.J(O) must equal f(O), and
the first three derivatives (for P3 ) or the first four derivatives (for P4 ) must equal the
corresponding derivatives off at x = 0. How do the coefficients c0 , c 1 , and c2 compare
with those for the quadratic and linear functions in Problems 2 and 3?
5. Plot graphs off, P3, and P.Jon the same screen. Use a friendly window of about [ - 2, 2]
for x and a window of [ - 20, 100] for y. Sketch the results.
6. For what range of values of x is the graph of P.Jindistinguishable from the graph off
on your grapher?
7. Show that the value of P.J( 1) is closer to the actual value off ( 1) than the value of P3( 1)
is.
8. In Problem 4, you should have found that for the fourth derivative Pj .Jl (0) to equal
f l4 l (0), the value of c.Jwas given by
80
24C4= 80 =>C.J= .
24
Write the 80 as the product of 5 and a power of 2. Write the 24 as a factorial. What
pattern do you notice ?
9. Show that c3, c2, c 1 , and even c0 follow the pattern in Problem 8.
10. Let P(x) be the power series P(x) = c0 + c 1x + c2x 2 + c 3 x 3 + c.Jx-1+ c5 x 5 + .Note
that P3 (x) and P 4 (x) are partial sums of this series. Make a conjecture about the values
of c5 and c6 such that P 5 (x) and P6 (x) best fit f(x) at x = 0.
11. In previous courses you probably learned how to express series in~ (sigma) notation.
For instance,
~
Ln
-
1
- xn = l + .!.x + .!.x 2 + .!.x 3 + .
+l 2 3 -1
11=0
The expression 1/ (n + 1) xn is evaluated for each integer value of n starting at 0, and
going to infinity, and the terms are added. Use what you have learned in this problem
set to 1,,vritein~ notation the series for P(x ) that best fits f (x ) = 5e2".
608 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
12-4 Power Series for Other Elementary Functions
In Sections 12-1 and 12-3, you saw that two quite diff eren t functions could be
represented in similar form as pow er series, at least for values of x close to zero .
6
Rational function: f (x) = - - Series : P(x) = 6 + 6x + 6x 2 + 6x 3 + 6x 4 + 6x 5 +
1-x
Exponential function: f (x) = Se2" Series: 5 + lOx + 10x 2 + ~ x 3 + x 4 + x 5 + f
The two series have th e same form. The only difference is the values of the
coefficients, all 6's for one series and 5, 10, 10, 20/ 3, ... for the other.
The process of finding the right coefficients for a particular fun ction is called
expanding the function as a power series. If a function and th e series match each
other at x = 0, the function is said to be expanded about x = 0.
OBJECTIVE Given an elementary function, find the first few terms of the power series that best fits the
function, find a pattern that allows you to write more terms of the series, write the series
in sigma notation, and plot the graph to see how well the series fits the function.
A power series can be written in the following general form.
Definition:PowerSeries
A power series for f (x) expanded about x = 0 can be written
where co, c1, c2, ... stand for constant coefficients.
Informal ly: A power series is a "polynomial" with an infinite number of terms.
Example 1 By equating derivatives, show that the following are the first three nonzero terms
of the power series for f (x) = sinx expanded about x = 0.
x - l.31x 3 + l.51x 5
Solution
In order for P(x) to fit f (x ) = sinx at x = 0, the function value and each
derivativ e off (x) must equal the corresponding function value and deri vative
of P (x) at x = 0. Assume that the series can be differ entiated termwis e.
P(x) =Co+ C1X+ + C3X3 + C4X-l+ C5X5 + C5X6 +
C2 X 2 =:> P(O) = Co
P' (x) = C1+ 2C 2 X + 3C3X2 + 4C4X3 + Scsx 4 + 6C5X5 + . . . =:> P'(O) = C1
P" (x) = 2c2 + 6C3X + l 2C4X2 + 20c sx 3 + 30C5X4 + .. . =:> P" (0) = 2c2
P"' (x) = 6c3 + 24c4X + 60c 5x 2 + 120c 6 x 3 + =:> P "' (0) = 6c 3 = 3 1c3
p <4l (x) = 24c 4 + l 20c 5x + 360c 6 x 2 + =:> p (4 ) (0) = 24C4 = 4!C4
p <5l (x) = l 20c 5 + 720c 5x + =:> p <5l (0) = l 20c 5 = 5!c5
Section 12-4: PowerSeriesforOtherElemen
taryFunctions 609
[Recall that 3! (three factorial) is the product of the first three counting numbers,
1 2 3 .) For the function, the derivatives off (x ) are as follows.
f(x) = sinx ~ f(O) = 0
f'(x) = cosx ~ f'(O) = 1
f" (x) = - sinx ~ f " (0) = 0
f "' (x) = - cosx ~ f "'( O) = - 1
f< l (x) = sinx
4 ~ f '~l (0) = 0
f CS) (X) = COS X ~ f CS)(0) =1
Equating the function values and corresponding derivatives off and P gives
Co= 0
C1 =1
2C 2 =0 ~ C2 =0
3 !C3 = -1 ~ C3 = - t
4 !C4= 0 ~ C4 = 0
S!cs = 1 ~ Cs = -t
Thus th e sum of the first three nonzero terms is
P (x) = X -
l
3! X
3 I -
+ 5! X ' , Q.E.D.
Once you have found derivatives for the series, you can remember the pattern and
use it when you are called upon to expand other functions as series by equating
derivatives .
Property:Derivativesof a Power Series
If P (x) =co+ C1X + QX 2 + C3X 3 + C4 X 4 + C5X5 + + CnXn + ,
then P (O) = co, P' (0) = c1, P" (0) = 2!c2, . .. , p <nl (0) = n!cn, .. . .
All you hav e to do to expand a function f as a power series about x = 0 is find the
values off (0), f' (0), f " (0), .. . , f <nl (0), ... , se t them equal to the above values of
P (O) and derivatives, and solve for th e values of c.
Example 2 For the series sinx = x - l31x 3 + .l51x 5 - ...
a. Demonstrate that you understand the pattern in the series by ,,vriting the next
thr ee terms.
b. Write the series using sigma notation.
Solutions a. The exponents of x are th e odd integers. Each coefficient is the reciproca l of
the factorial of the exponent. The signs alternate, with the first term being
positive. Thus,
sinx = x - .l3! x 3 + .lx
5!
5 - .l71 x 7 + .l
9!
x9 -
ll !
+ .. .
___!_xn
610 Chap
ter 12: TheCalculus
of Functions byPowerSeries
Defined
b. The series can be written in sigma notation this way:
I,( - l)n 1 x2n+J
n=O (2n + l)!
This symbol is pronounced, "the sum from n = 0 to infinity of
(- l)nx 2n+l / (2n+l)!." It means to let n =0,1,2,3, . . . and add up the
resulting terms . The secret to finding a formula for tn (the term with index
n) is writing the values of n under the terms . In this case it is helpful to start
n at O rather than l.
x- .!.x
31
3 + .!.x
51
5 - .!.x 7
71
+ . ..
Write th e values of th e term ind ex, 11, und er th e
0 1 2 3 resp ective term s.
By comparing the values of n with numbers in the terms, you can see that the
exponent and denominator in each term are one more than twice the value
of n. The factor (-l)n makes the signs alternate . The index of summation, n,
could start at 0, 1, or whatever number you feel is appropriate . For instance,
the answer above could be written
I
n=J
(- l)n +l l
(2n - 1)!
x 2 n-l Altern ative form .
Example 3 a. Plot the sixth partial sum of the power series for sinx expanded about x = 0.
b. Find, approximately, the interval of x-values for which the sixth partial sum is
within 0.0001 unit of the value of the sine function.
c. Find a wider interval for which the ninth partial sum is within 0.0001 unit of
sinx .
Solutions a. The sixth partial sum is S5 (x) since the index of summation starts at n = 0.
S 5 (x) = x - .!.x
3!
3 + .!.x
51
s- .!.x
7!
7 + .!.x
91
9 - ...Lx
11 1
11
A time-efficient way to enter the partial sum uses the formula for the nth
term that was found in Example 2 and the grapher's sequence commands.
For a typical grapher you can enter the formula this way:
y 1 = SUM SEQUENCE(( - l) " n / (2n + l )! * x" (2n + 1), n, 0, 5, 1).
/x The sequence command tells the grapher to generate a set of numbers using
the formula inside the parentheses . The n after the comma tells the grapher
that n is the index of summation . The last three numbers tell the grapher to
start n at 0, end it at 5, and increase it by steps of 1 each time (thus giving
six terms) . The sum command tells the grapher to add up the terms of the
Figure l 2-4a sequence it has calculated . The grapher performs this computation for each
value of x in the window you specify . Figure 12-4a shows the result.
Section12-4: PowerSeriesforOtherElementary
Functions 6 11
ttt X sinx - sum
-4 - 0.0100020 . . .
-3 - 0.0002454 .. .
-2 - 0 .00000129 . ..
- 1 - 0.000000000159 . ..
0 0
1 0.000000000159 . ..
2 0.00000129 ...
3 0.0002454 ...
4 0.0100020 . ..
b . A table of values of the differences between sinx and the series quickly
shows the interval of x-values. Enter (sinx - y 1 ) as y2 . From the table you can
see that if xis between - 2 and 2, the absolute value of the difference is less
than 0.0001. By exploring the interval between 2 and 3 with anothe r tabl e,
stepping x by 0.1, you can find that the series is within 0.0001 uni t of sin x
for -2 .7 < X < 2.7.
The answer can also be found using the solve feature of your grapher. If
you have entered sinx as y 1 and the partial sum as y 2 , then set (y 1 - y 2 ) -
0.0001 equal to 0. The result is x ~ 2. 7986 . ... By symmetry, the interva l is
-2.7986 ... < X < 2.7986 ....
c. The ninth partial sum is S8 (x) . If you are using the sequence commands, you
can change the 5 in part (a) to 8. The grapher will then calculate 9 terms of
the seri es, starting at n = 0, for each value of x . The resulting graph is as in
Figure 12-4b. Numerically, as in part (b) you can find that the ninth part ial
sum is within 0.0001 of sinx if
Figure l 2-4b - 4.8974 . . . < X < 4.8974 . ..
Power series relate to functions the way decimals relate to irrational numbers. For
instance,
TT = 3.141592653 ... and ,) 7359 = 85.7846140 . ...
The human mind can grasp the size of the approximation 85.78 ... more easily than
it can grasp the exact value, ) 7359 . Similarly, a computer can calculate values of
sinx from a power series more easily than it can calculate them direct ly fro m a
definition of sine.
The more decimal places you use for a number such as TT, the more accurate
the approximation is. In many cases, the more terms you use for the partia l su m
of a series, the better the partial sum fits the function values . As was shown in
Figure 12-4b, the ninth partial sum of the series above seems to coincide with
sinx for - 7 < x < 7. The sixth partial sum in Figure 12-4a coincides only for about
-4 < X < 4 .
612 Cha
pter 12: TheCalculus ctions Defined by Power Series
of Fun
Problem Set 1 2-4
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes .
Q 1. Sketch the graph of y = sin x.
Q2. Sketch the graph of y = cos x.
Q3. Sketch the graph of y = ex.
Q4. Sketch the graph of y = lnx .
QS. Sketch the graph of y = cosh x.
Q6. Sketch the graph of y = tan - 1x.
Ql. In the expression 3x 5 , the number 5 is called the -? - .
. In the expression 3x 5 , the number 3 is called the -? - .
QB
Q9. The expression x 5 is called a - ?- .
Q10. The area of the region between the graph of y = 9 - x 2 and the x-axis equals- ?- .
1. Exponential Function Series Problem:
a. Show by equating derivatives that the pow er series expansion for ex about x = 0 is
P(x ) = 1 + x + l.x
21
2 + l.x 3 + . . .
3! .
b. Write the next two terms of the series.
c. Write the series using sigma notation.
d. Plot the fourth partial sum of the series . On the same screen plot y = ex. Use a
window of [ - 3, 3] for x and [ - 2, 10] for y . Sketch the result .
e. For what interval of x-values are the two graphs indistinguishable from each other ?
f. For what interval of x-values is the fourth partial sum within 0.0001 unit of e x7
g. For what wider interval is the ninth partial sum of the series within 0.0001 of ex?
2. Cosine Function Series Problem:
a. Show by equating derivatives that the power series expansion for cos x about x = 0 is
P(x) =1- l.x
21
2 + l.4 x 4
1
- l.x
61
6 + l.x
81
8 - .. .
b . Write the next three terms of the series .
c. Write the series using sigma notation. Start the index of summation at n = 0.
d. Figure l 2-4c shows the graph of the fifth partial sum, S4(x ) . Plot
this graph on your grapher. Then plot y = cos x on the same
y
screen. Sketch both graphs.
e. Plot th e graph of the eighth partial sum, S7(x ). For what inter val
of x-values is th e S7(X) graph indistinguishable from that of
X
y = cos x? Sketch th e result.
f. For what interval of x is the eighth partial sum within 0.0001
unit of cos x ?
g. Give a reason why the series for P(x) agrees with the prop erties Figure 12-4c
of the cosine function .
Sect
ion 12-4:Power
Series
for OtherElementary
Functions 613
3. Sine Series Problem: Let P(x) = x - ix 3 + i x 5 - fi x 7 + , which is the sine series.
a. Show that S3 (0.6), the fourth partial sum, is approximately equal to sin 0.6.
b. The tail of a power series is the series of terms left after a given partial sum. Write
the value of sin 0.6 to as many places as your calculator gives. Use this number
to evaluate, approximately, the tail of the P (0.6) series for S1(0.6), S2(0.6), and
S3 (0.6). Show that in each case, the value of th e tail of the series is less in magnitude
than the absolute value of the fi_rst term of the tail.
c. Assuming that your observation in 3b about the tail of the seri es is correct for all
values of n, figure out how many terms of th e series for P (0.6 ) it would take to get
a partial sum that estimates sin 0.6 correct to at least 20 decimal places.
4. Hyperbolic Sine and Cosine Series Problem: Let P (x) = I. ( 1
) x 211
n =O 2 n + l 1.
1
+
a. Write the first four terms of the series (through n = 3).
b. Show by equating derivatives that the series P (x) represents sinhx.
c. Show that S3 (0.6), the fourth partial sum of the series, is approximately equal to
sinh 0.6.
d. For what interval of x-values is S3 (x) within 0.0001 unit of sinh x?
e. Assume that the derivative of the series equals the sum of the derivatives of the
terms. Differentiate each term of the series for P (x) to get a series for P ' (x).
f. Since the derivative of sinhx is coshx, the series you found in 4e should be the
series for coshx. Demonstrate that this is tru e by showing that the value of P' (0.6)
is approximately equal to cosh 0.6. Use the fourth partial sum of the derivative
series.
g. Integrate the series for P(x) term-by-term to get a power series for JP(x) dx. Show
that the result is the series for coshx in 4f if the integration constant is picked
appropriately.
5. Natural Log Series Problem: Let P(x) = (x - l ) - (x - 1) 2 + (x - 1)3 - (x -1 )4 + .. .
This series is the power series for lnx expanded about x = 1.
a. By equating derivatives, show that P(x) and lnx have the sam e function value at
x = l, and the same first, second, and third derivative values at x = 1.
b. Write the next two terms of the series .
c. Write the series using sigma notation. Start the index of summation at n = 1.
d. Figure 12-4d shows the graph of the fourth partial sum of the y
In~ ..........
series . The graph fits y = lnx reasonably well when xis clos e to 1
1. Plot y = ln x and S10 (x), the tenth partial sum of the series, X
on the same screen. Sketch.
e. By appropriate use of the TRACE or TABLE feature, compare
S10 (1.2), S10 (1.95), and S10 (3) with the values of ln 1.2, ln 1.95,
and ln 3. For what interval of x-values does the tenth partial sum
of the series seem to fit the ln function? Is this interval much Figure l 2-4d
larger than that for th e fourth partial sum in Figure 12-4d ?
6. Convergence and Divergence Problem: From Problem Set 12-1 you recall that a series
converges to f(x) for a particular value of x if the partial sums of P(x ) approach the
value off (x) as the number of terms in the partial sum approaches infinity . In Problem
6 14 Chapter
12:TheCalculus DefinedbyPowerSeries
of Functions
5 the series P(x) converges to lnx when x = 1.2 and x = 1.95. But th e series diverges
for x = 3. In this problem you will see reasons why this is true .
a. Make a table of values for the first few terms of P (3) . What is happening to the
absolute value of the terms?
b. By appropriate use of !'Hospital's rule, show that the absolute value of the nth term
for P(3) approaches infinity as n approaches infinity. Explain how this fact tells
you that the series for P(3) cannot possibly converge .
c. Make a table of values for the first few terms of P(l.2) . Show that these terms
approach zero for a limit as n approaches infinity, and thus the series could
converge .
d. In Problem 3 it was observed that the value of the tail of a series which remained
after the nth partial sum was smaller in absolut e valu e than the absolute value of
the first term of the tail . Is this observation true for P(l.2)? Justify your answer.
7. Invers e Tangent Series Problem:
a. Let f (x) = tan - 1 x. Let P(x) be the power series
P(x) = f
n =O
(-l)n _ l_ x 2n+l
2n + 1
Write out the first few t erms of the series.
b . Plot the graph of f and the graphs of the sixth and seventh partial sums on the
same scree n. For what values of x do the partial sums represent the graph off quite
well? For what values of x do the partial sum graphs bear little or no resemblance
to the graph of f?
12-5 Taylor and Maclaurin Series, and Operations on
These Series
In Problem 5 of Problem Set 12-4, it was stated that the expansion of lnx about
X = l is
ln x = (x - 1) - (x -1) 2 + (x - 1) 3 - (x - 1)4 + ....
The coefficients can be found by equating derivatives at x = 1 as you will see in
Example 1 of this section . In general, if function f is expanded about x = a, then the
coefficients are
, f " (a) f "' (a)
Co = f (a), C1 = f (a), C2 = _ 2_!_ ' C3 = -3- ! - , .. . .
The powers are (x - a)n rather than xn becaus e (x - a ) equals zero when x = a.
The result is called the Taylor series expansion of {(x) about x = a after the
British mathematician Brook Taylor (168 5-1731). If a = 0, the Taylor series is called
the Maclaurin series expansion of f(x) after the Scottish mathematician Colin
Maclaurin (1698-1746), although neither Taylor nor Maclaurin was the first to
publish this kind of series .
Sect
ion12-5:TaylorandMaclaurin
Series
, andOperations
onThese
Series 615
ii Ii
TaylorSeries andMaclaurinSeries
If f is a function with differentiabl e derivatives, then f (x) can be written as a Taylor seri es
expans ion about x = 1 as follows:
f " (a ) f '" (a ) . r (n) (a)
f (x ) =f (a) + f'(a)(x - a) + -- (x - a) 2 + --( x - a )3 + + - - (x - a)n + . ..
21 3! n!
If a = 0, the series is called a Maclaurin series.
OBJECTIVEGiven the Taylor series for eX, sin x, cos x, sinh x, cosh x, ln x, 1/ ( 1 - x), and tan - 1 x,
perform operations on these series to derive power series for related functions.
To accomplish the objective efficiently, it is a good idea to memorize the following
eight series. These are the series you derived in Sections 12-1 through 12 -4.
EightWe/I-Known Power Series
"" 1
ex= 1 + X + -x1 2 + -x1 3 + -x1 4 + ... = L -xn
2! 3! 4! n=O n!
sinx = x-..!.x 3+ ..!.xs_ l.x7+
3! S! ?!
= f n =O
( - l)n
(2n
1
+ 1)!
x 2n+J
1
cos x = 1 _ l.x 2 + ..!.x4
2! 4!
- ..!.x 6
6!
+ = f n =O
(- 1)"- - x 2 n
(2 n)!
.
sinh X = X + -31 x 3 + -1 x 5 + -71 x 7 + .. . = L"" ---x 1 2
n+ l
! S! ' n =O (2n + 1)!
cosh x = 1+ 1
-x 2
2!
+ 1
-x 4
4!
+ 1
-x 6
6!
+ = L"" --x1
n =O (2 n)!
2n
ln x = (x - 1) - (x - 1) 2 + (x - 1) 3 - (x - 1) 4 + = f (- l)n + ~(x -
n= I
1 1 )n
1 ""
-- = 1 + x + x 2 + x3 + x 4 + = L xn (A geometric series .)
1- X n =O
1
tan - 1x = x _ .!.x3 + .!.x5 - .!.x7 + =
3 5 7 n =O
f (-1)"-- -x 2n + 1
2n + 1
Example 1 Show by equating derivativ es that the Taylor series for ln x expa nded about x = 1 is
lnx = (x - 1) - (x - 1) 2 + (x - 1) 3 - (x - 1) 4 + ... .
Solution So that P ( 1) will equal c0 , P' (1) will equal c1 , and so forth, the series is written
in powers of (x - 1) in stead of in pow ers of x as b efore.
P(x) = Co+ C1(x - 1) + C2(X - 1) 2 + C3(X - 1) 3 + C4(X - 1) 4 + . .. .
616 Chapter
12:TheCalculus
of Functions
Defined
byPower
Series
[If you were to use P(x) = c0 + c 1x + c2x 2 + ,then P(l) would equal
c0 + c1 + c2 + instead of just c0 .) Equating derivatives gives
f(x) = lnx ~ f(l) = 0 ~ Co = 0
f'(x) = 1/x = x- 1 ~ f ' (1) = 1 ~ C1= 1
f"(x) = - x - 2 ~ f"(l) =- 1 ~ 2!c2= -l ~ C2 = - 2I
I
f "' (x) = + 2x- 3 ~ f' " (l) = 2 ~ 3!c3 = 2 ~ C3 = 3
f (4l(x) = - 6x-4 ~ f (4) (1) = - 6 ~ 4 1c4 = - 6 ~ C4 = _l 4
[( 5 > (x) = +24x - 5 ~ f (5 ) (1) = 24 ~ 5!c5 = 24 ~ c-:, -- l5
.-. lnx = (x - 1) - (x - 1) 2 + (x - 1)3 - (x - 1)4 + , Q.E.D
Example 2 Expand f (x) = sinx as a Taylor series about x = TT / 3.
Solution f(x) = sinx ~ f(?f) = sin?f = "}
f'(x) = cosx ~ f'(?f) = cos 1=
f " (x ) = - smx
~ f " (3 = - sm 3 = -
TT ) TT
2
J3
f" ' (x) = - cosx ~ f' " (?f) = - cos ?f = - ,and so forth .
2] [(x - TT/ 3)2]
.-. sinx = -)3/ 2 + (1 / 2)(x - TT/ 3) - [ 1;;
- [ ;n
1
[(x - TT/ 3) 3] + [ ~; 2] [(x - TT/ 3) 4 ] + .
Example 2 shows that once you have found the derivatives for the function you are
expanding, you can just substitute them directly into the formula for the Taylor
series. It isn't necessary to equate derivatives.
Figure 12-5a Figure 12-Sa shows that the fifth partia l sum in Example 2 fits sinx well if xis in a
neighborhood of TT / 3.
It is usually easier to derive a series by starting with one of the known series than it
is to calculate deriva tives off (x) . Examples 3-6 show you ways this can be done .
Example 3 Write the first few terms of the Maclaurin series for sin(3x) 2.
Solution Take the known series for sinx, and replace the x with (3x) 2. The rest is algebra .
sin(3x) 2 = (3x) 2 - t, [(3x) 2 ]3 + t [(3x )2]5 - t [(3x) 2]7 +
= 32x2 - ~3! x 6 + ~51 x10 - ~71 xl-1+ ...
1
Example4 Write the first few terms of the Maclaurin series for g(x) = - - 3 .
l +x
Solution This series can be derived by performing long division, or by sub stituting (- x3 )
for x in the geometric series from 1 / (1 - x) .
g(x) = 1 + (-x 3) + (-x 3)2 + (- x3)3 + (-x 3)4 + ...
= 1 - x3 + x 6 - x g + x1 2 - .. .
Section12-5: TaylorandMaclaurin
Series,andOperatio
ns onThese Series 617
I~
Example 5 By app ropriate operations show that the Maclaurin series for tan - 1 x is
tan - 1x = x - x 3 + x s - ~x 7 + . ..
Solution If you start the process of equating derivatives, the result is
f(x) = tan - 1 x
f'(x) =_ l_ .
1 + x2
The expression for the first derivative can be expan ded as a Maclaurin seri es by
substituting - x 2 for x in the geometr ic series from 1 / (1 - x) , as in Example 4.
f' (x) = 1 - x2 + x 4 - x6 + . . .
The series for tan - 1 x can be found by int egrating, assuming that an infinit e
series can be integrated termwise.
f
tan- 1 x = (l - x 2 + x 4 - x 6 + ) dx
= x _ lx
3
3 + l x s _ l x 7 + . .. + C
S 7
Since tan - 1 0 = 0, the constant of integration C is also zero. Thus
tan - 1x = x- x 3 + x s - ix + ,Q.E.D.
7
Example 6 Write a power series for f (x) = J; t cos t s dt. Evalu ate th e sixth partial sum at x = 0 .8.
Solution The technique is to write a series for the integrand, then int egra te term by term.
As in Example 5, assume that the su m of an infinit e number of terms can be
int egrated termwise, which is true in this case, bu t not always . First, replac e x
with t s in the Maclaurin series for cosine .
COS
2! + l.4!t 20 - l.61t 30 + . ..
t s = 1 - .l.tlO
Then multiply each term by t, and integrate.
cos t s = t - l.t 11 21 31
t 21 + l.t
41 - l.61t + ...
= Jo
(" ( t - l.t 11 21 - .l t 31 + ) dt
f(x) 2! + l.t 4! 51
l 2 I 12 l
f (x) = ;/ - iz.Trt + 224!t - JITi 22 1
t 32 + ... Ixo
f (X) = 2l X 2 - iz.21X
1 J2
+ 221,4 1 X 22 - JITi
1
X32 +
To find S6 (0 .8), substitute and do the arithmetic . For decimals longer than your
calculator can handl e, add and subtract column-wise, as shown here .
S5(0.8) = 0.8 2 - 12\ 0.81 2 + 22\ 0 .822 - 32\10 .832 + 42\1 0 .842 - s2\010 .8 s2
0.3200000000000 .. .
- 0 .0028633115306 ...
+ 0.0000139748061 .. .
- 0 .0000000 343872 . . . Add th e digit s column -wise. For insta nce, for the
13th decimal place, add O - 6 + 1 - 2 + 2 - 0
+ 0 .0000000000502 .. .
to obta in - 5. Write down 5 and carry - 1.
- 0.0000000000000484 .. .
0.31715062893851 ...
618 Chapter
12: TheCalcul
us of Functions d by Power Series
Define
The answer could, of course, be found numerically . But th e series takes onl y six
terms to give 13-pla ce accuracy. As you will see in the following section, it is possible
to det ermine th e accuracy of an integral if you use a series rather than a Riema nn
sum or other num erical methods .
Problem Set 1 2 5
DoTheseQuickly
The following problems are int ended to refresh your skills. You should be able to do all
10 in less than 5 minutes .
QJ. Evaluate: 4!
Q2. Evaluate: 3!
Q3. Evaluate: 4!/ 4
Q4. What does n equal if 4!/4 = n!?
QS. If m!/ m = n!, then n = -?- .
Q6. 0' = m!/ m. What do es m equal ?
Ql. Why do es O!equal 1?
QB . Why is ( - 1) ! infinite ?
Q9. Differentiat e: f (x) = -Jx 2 - 7
. Integrate:
QJO f sinhx dx
For Problems 1-8, write from memory the power series.
l. f(u ) = e" 2. f (u) = ln u
3. f (u ) = sin u 4. f ( u) = cos u
5. f (u ) = coshu 6. f (u ) = sinh u
7. f(u ) = (1 - u)- 1 8. f (u) = tan - 1 u
For Probl ems 9-24, deri ve a pow er series for the given function. Write enough terms of
the series to show the pattern.
9. xsinx 10. x sinhx
11. coshx 3
12. cosx 2
.,
13. lnx 2 14. e-x-
15. ( e - 12 dt 16. J;sin t 3
dt
17. J: t 2
sin t
5
dt 18. r In t 3 dt
X l
19. Jc - sinht 2 dt 20. fox cos t 0 5 dt
0 t
1 9
21. 4 22.
x +l x2 +3
l 9
23. l -tX
0 4 +l
dt 24. l
0
X
-
t2 + 3
dt
Section12-5:Taylor
andMaclaur
in Series,andOperation
s onTheseSerie
s 619
-
For Problems 25-30, expand the function as a Taylor series about the given value of x.
Write enough terms to reveal clearly that you have seen the pattern.
25. f (x.) = sinx., about x = TT / 4 26. f(x.) = cosx, about x = TT / 4
27. f(x.) = lnx., about x = 1 28. f (x) = logx, about x = 10
29. f(x) = (x - 5) 713 , about x = 4 30. f(x) = (x + 6) 42 , about x = -5.
31. Find the Maclaurin series for cos 3x by equating derivatives . Compare the answer,
and the ease of getting the answer, with the series you obtain by substituting 3x for
x. in the cosine series.
32. Find the Maclaurin series for ln(l + x.) by equating derivatives. Compare the answer ,
and the ease of finding the answer, with the series you obtain by substituting (1 + x.)
for x in the Taylor series for lnx., expanded about x = 1.
33. Accuracy for ln x. Series Value: Estimate ln 1.5 using S4 (1.5), fourth partial sum of the
Taylor series . How close is your answer to the real answer? How does the error in
the series value compare with the first term of the tail of the series, t5 , which is the
first term left out in the partial sum7
34. Accuracy Interval for In x Series: Find the interval of values of x for which the fourth
part ial sum of the Taylor series for ln x gives values that are within 0.001 unit of
ln x.
35. Inverse Tangent Series and an Approximation for TT: You recall that tan(TT/ 4) = 1.
Thus tan - 1 1 = TT / 4. In this problem you will use the inverse tangent series to estimate
TT.
a. Write the first few terms of the Maclaurin series for tan - 1 1. Then use the
appropriate features of your grapher to find the 10th partial sum of this series .
Multiply by 4 to find an approximate value of TT. How close does this approximation
come to TT?
b . Find another approximation for TT using the 50th partial sum of the series in 35a.
Is this approximation much better than the one using the 10th partial sum 7
c. By appropriate trigonometry, show that
tan - 1 1 = tan - 1 + tan - 1 .
Use the result to write TT / 4 as a sum of the Maclaurin series. Estimate the value
of TT by finding the tenth partial sums of the two series. Comment on how much
better this method is for estimating TT than the methods of 35a and 35b.
36. Tangent Ser ies Problem : You recall that tanx = (sinx) / (cosx). Long divide the
Maclaurin series for sin x by that for cos x to get a power series for tan x . Use enough
terms of both sine and cosine series to find four terms of the tangent series . Show by
calculator that the fourth partial sum for tan 0.2 is close to tan 0.2.
37. Taylor Series Proof Problem: Prove by mathematical induction that for all positive
integers n, the nth derivative of the general Taylor series is equal to f (nJ(a).
38. Historical Prob lem: What were Taylor's and Maclaurin's first names7 When did they
live in relation to Newton and Leibniz, who invented calculus?
620 Chapter
12:The CalculusofFunctions DefinedbyPower
Series
* 39. Ratio of Terms Prob lem: A Taylor series usually gives better and better approximations
for values of a function the more terms you us e. For some series this is tru e only for
certain valu es of x. For instance, the series for th e natural logarithm ,
lnx = I( - l)n +t.!.(x - l)n,
11= 1 n
converges to lnx only for O < x :,; 2. If xis outside this interval of convergence, the
series does not converge to a real number. It diverg es and thu s cannot r epresent
lnx. In this problem you will investigate the ratio of a term in this series to the term
b efore it, and try to discover a way to find from this ratio whether or not the series
converges.
a. As shown above, formula for tn, the nth term in the series for lnx, is
tn = (-l)n +1.!_(X - l)n .
n
Let rn be th e ratio ltn+i / tnl- Find a formula for rn in terms of x and n.
b. Calculate r 10 for x = 1.2, x = 1.95, and x = 3.
c. Let r b e th e limit of r,, as n approaches infinity . Find an equation for r in terms of x.
d. Evaluat e r for x = 1.2, x = 1.95, and x = 3.
e. Make a conjecture : "The series converges to lnx whenever the value of x makes r
- ?- , and diverges whenever the value of x makes r - ?- ."
f. If your conjecture is correct, you can use it to show that the series converges if
x is in the int erval O < x < 2. Check your conjec ture b y showing that it gives this
interval.
40. Journal Problem: Updat e your journal with what you 've learn ed since th e last entry.
Includ e such things as those listed here.
The one mo st important thing you've learn ed since the last journal entry.
The difference between a sequence and a series.
Th e distin ction betwee n term index and term number.
The definition of geometric series.
The meaning of power series, and for what purpose the y ma y be useful.
What it mean s for a series to converge and to diverge.
Anything about series that is still unclear .
*T his problem pr epares you for the next section.
Section12-5: TaylorandMacla
urin Series,andOpera
tionsonTheseSeries 621
12-6 Interval of Convergence for a Series-
The Ratio Technique
A series converges to a certain number if the limit of the nth partial sum is that
numb er as n approaches infinity . Power series often converge if x is within 1 unit of
a, th e constant about which the series is expanded. For instanc e the ser ies
1nx = ex - 1) - ex- 1) 2 + ex - 1) 3 - ex - 1)4 + ... = f. c- l)n +1~cx -
n= l
l)n
converges when x = 1.6. The quantity (x - l ) equals 0.6, and the powers 0 .6n
approach zero rapidly as n gets larg e. But if x = 4, the quantity (x - 1) is 3, and
the powers 3n become infinitely large as n approaches infinity . You can see what
happ ens from a table of values .
n nth term, x = 1.6 nth term, x = 4
1 0.6 3
2 - 0.18 -4.5
3 0.072 9
4 - 0.0324 - 20.25
5 0.015552 48.6
6 - 0.007776 - 121.5
7 0.00399908 ... 312.428 ...
20 - 0.00000182 ... - 174339220.05
Figure 12-6a shows what happens to th e partial sums of the natural logarithm
series. The left-hand graph shows that the partial sums for x = 1.6 converge rapidly
to a number around 0.5 as n approaches infinity . The right-hand graph shows that
th e partial sums for x = 4 diverge .
Sn
20
, Converges 10
0.5 - - - - - _..~_.::..=
,.-
.r-:...
4-......-.
-----
- Diverges
n
2 3 4 5
- ....
-10
n ~
5 10 -20
Figure 12-60
Surprisingly, the series for sinx,
00
. = x - -1 x3 + -1 xs - -1 x 7 + ... =
smx L (-l)n ___ 1 x2n+1
3! 5! 71 n =O (2n + 1)! '
converges no matter how large x is ! At x = 10, for instanc e, the power 102n+i
is very large . But the denominator, (2n + 1)!, is much larger. If n = 20, then
(2n + 1)! = 41! = 3.3 .. . x 1049 , which is 300 million times as big as 1041 . In this
section you will develop a method called the ratio techni que (sometimes called the
622 Chapter
12: TheCalculus
of Functions byPowerSeries
Defined
ratio test) for finding precisely the interval of convergence - that is, the interval of
x-values for which a power ser ies converges.
OBJECTIVE Givena power series in x, use the rati.o technique to find the interval of convergence.
The ratio technique is based on bounding the given series with a convergent
geometric series. To see how the technique works, consider the series for lnx when
X = 1.6.
lnl.6 = (1.6 - 1) - (1.6 - 1) 2 + (1.6 - 1) 3 - (1.6 -1 ) 4 +
= 0.6 - (0.6) 2 + (0.6) 3 - (0.6) 4 + ...
If you take the ratios of the abso lut e values of adjacent terms, Itn+ 1 / tn I, you get th e
sequence
(0.6), i (0.6), (0.6), ~ (0.6), ~ (0.6), ; (0.6 ), i (0.6), ~ (0 .6), .. .
= 0.3, 0.4, 0.45, 0.48, 0.5, 0. 514 . .. , 0.525, 0.5333 . .. ,
A given term in the ln 1.6 series is formed by multiplying the preceding term by the
approp riat e one of these ratios. So each term is less than 0.6 times the preceding
term.
Pick a geometric series with common ratio between 0.6 and 1, say r = 0.7, and
with the first term equal to a term in the tail, say lt4I = 0.0324. Since lr l < 1, the
geometric series converges to
0 324
S = = 0.108.
1- 0.7
As shown in Figure 12-6b, the geometric series is an upper bound for the absol ut e
values of th e terms in the tail of the ln 1.6 series. Starting the geometric series at
t 20 gives an upp er bound of It20 I/0.3 , which equals 0.00000609 .... Since the terms
of the ln 1.6 series alternate in sign and approach zero as n approaches infinity,
a lower bound for the tail is -0 .00000 609 .... Since the bounds for the tail can be
made as close to zero as you like, the series converges .
'94 = 14
I\
I\
\ .. 95 = 0 .794
\'
I '
1t51= 0.48lt) ~ \ ,9G= 0.?9 5
\ ' ,97 = 0.796
ltGI = 0.Slt; le..
'
1t7 1= 0.514 ... 1t6 1.._ ..._
.,-._n
....__
10
Figure l 2-6b
In general, a po wer series will converge if the ratio of the absolute values of adjacent
terms can be kept less than some number R, an d R is less than 1. In that case
Section12-6:Interval
of Convergence
fora Series-TheRatioTechnique 623
you can always find a geometric series with common ratio between R and 1 that
converges and is an upper bound for the tail of the series.
One way to show that there is such a number R is to take the limit, L, of the ratios
of adjacent terms . As shown in Figure 12-6c, if L < 1, then you can pick an epsilon
small enough so that R = L + E is also less than 1. Then any geometric series with
common ratio r between R and 1, and with a suitable first term, will be an upper
bound for the tail of the given series .
Ratio
- - - - - - - - - - -- - - - - - - - - - - - - - -
. . .-.- .
l
- - - - _. - -. - -
R
L +E ;- -
L
:
: From her e on
: the ratio s are
: less than L + E.
'
5 n=N 10 n
Figure l 2-6c
This fact can be used as a relatively simple way to find the interval of values of x
for which a series converges . Example 1 shows you how this technique is used with
the series for ln x.
Example 1 Find the interval of convergence for ln x = I. (-1)
11= 1
n +1 l. (x -
n
1) n.
Solution L r l tn+ll r 1(- l) n+2(x-1)" +1 n I
= ,,1~ r"; = ,,1:...~ (n + 1) (- 1) 11+ 1 (x - 1) 11
= lim I (x - l)(n ) I What happ ens to ( - 1) 11+ 1 and ( - 1 ) 11+ 2 7
n-oo n+1
= Ix - 11lim ~ n- Since Ix - 11is ind epend ent of n.
n-00 n+1
= lx - l ly By !'Hospit al's rul e, first embedding t 11 in a
continu ous fw1ction.
= Ix -1 1
So the series will converge if
[x - 1[ < 1 => - 1 < x - 1 < 1 => 0 < x < 2. (Open) inter val of conv erg enc e.
Here is a formal statement of th e ratio technique used in Example 1.
Technique:
TheRatio Technique
for Convergence
of Series
For the series f
n=l
tn, if L = lim
n-oo
Itntn+l I,then:
i. The series converges if L < l.
ii . The series diverges if L > l .
iii. The series may either converge or diverge if L = l.
624 Chapter
12:TheCalcu tionsDefinedbyPower Series
lusof Func
The int erval of convergence in Exampl e 1, 0 < x < 2, goes 1 unit eith er side of
x = 1, the value of x abo ut which the ser ies is expanded . Th e half-width of the
interval of convergence is called th e radius of convergence. The word radius is
used because if x is allowed to be a complex number, the series converges for all x
inside a circle of that radius (Figure 12-6d).
Imaginari es
........ ... \
Reals
-1
1 /2
-i ..\ ~~;;:: ..converges
for x in here.
Figure l 2-6d
oo n
Example 2 For the series n~i ~(x - S)n,
a. Write out the first few terms.
b. Find the interval of convergence.
c. Find the radius of convergence .
Solutions a. (x - 5) + ~(x - 5) 2 + ,b (x - 5) 3 + fi (x - 5) 4 +
b. L = lim
11- 00
I t11+1
tn
I
. l( n +l)(x-S)n +I
= 11-00
hm 3n + l
= hm -
n -oo
. In+
n
-1 -1 (x - 5) I
3
1
= Ix - 5 I lim [ n + I
n- oo 3n
= lx - 51
The series will converge if L < l.
Ix - SI< 1 =>Ix - 51 < 3 =>- 3 < x - 5 < 3 =>2 < x < 8
c. The radius of convergence is the distance from the midpoint of the int erva l
of convergence to one of its en dpoint s.
Radius of convergence is 3.
If x equals the number at an endpoint of the interval of converge nce, the limit of the
ratio of terms equals 1. The series may or may not converge in this case. In Section
12-7, you will learn tests for convergence that can be used when the rati o technique
doesn't work. Examp le 3 shows you that the radi u s of convergence of a series is
zero if the limit of the ratio of terms is infinite.
Section
126: IntervalofConvergence
fora Series-TheRatioTechnique 625
Example 3 For the series I"" -----i(x
n' - 3)n,
n= I n
a. Write out the first few terms .
b. Show that though the first few terms decrease in value, the radius of convergence
is zero.
c. For what one value of x does the series converge?
Solutions a. (x - 3) + (x - 3) 2 +f? (X- 3) 3 +fz (x - 3) 4 +
b . Note that the factorials simplify nicely when you divide adjacent terms .
4
. I -tn +l I - 11m
L - I1m . I -------
(n + l)!(x - 3)n+l --- n I
- n-oo t,, - n-oo (n + 1) 4 n!(x - 3)n
= lim I (n + 1) (n!) . ( - n- )\x - 3) I
n-oo n! n+l
4
= lx - 3llim
n-oo
l (n + 1) ( -n +
n- l ) 1
Since n/ (n + 1) goes to 1 as n approaches infinity, its fourth power also goes
to 1. Thus the quantity inside the abso lute value sign approaches the other
(n + 1), and L is infinite for all values of x not equal to 3. The radius of
convergence is thus equal to zero, Q.E.D.
c. If x = 3, the series becomes O + 0 + 0 + , which converges to zero . So 3 is
the only value of x for which the series converges.
From Example 3 there follow two special cases .
SpecialCases:Zero andInfiniteRadiusof Convergence
For a power ser ies in (x - c)n with radius of convergence r:
If r = 0, the series converges only at x = c.
If r is infinite, the series converges for all values of x.
Problem Set 1 2-6
DoTheseQuickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
Q1. x - x 3 / 3' + x 5 / 5' - x 7 / 7' + = -? -
Q2. x + x 3 / 3! + x 5 / S!+ x 7 / 7! + =- 7-
Q3. l - x + x 2 / 2!-x / 31 +x 4 / 4! - = -?-
3
Q4. l + x + x 2 / 2' + x 3 / 3! + x4 / 4! + = -?-
Q5. 1 + x + x 2 + x 3 + x4 + x 5 + ... = -?-
Q6. Integrate : f cos 2x dx
626 Chapter
12: TheCalculus Defined byPower Series
of Functions
Ql. Differentiate: f (x) = tan 3x
QB. Find the limit as x approaches zero of f(x) = cos4x.
Q9. Find the limit as x approaches zero of g(x) = (1 + x)' 1x .
QIO. (Force)(displacement) = _7_
For Problems 1-6,
a. Write out the first few terms.
b . Find the interval of convergence.
c. Find the radius of convergence.
; n n
00 xn
1. L, -x 2
n = l 4n n 2n
n~l
00
(2x+3)n (5x - 7)n
00
3. I --- 4 . I ---
n= 1 n n= 1 2n
5.
oo
I
n3
-( x - 8)n 6.
n' + 2)n
I __:_(X
oo
4
n=I n! n= l n
For Problems 7-12, show that these familiar series for the transcendental functions
converge for all real values of x.
7. sinx = x - l.
3!
x 3 + l.x
51
5 - l.x
71
7 + ...
2!
x 2 + l.x
8 cosx = 1 - ..!.. 4!
4 - l.x
6!
6 + ...
9 . sinhx = x + l.x
3!
3 + l.x 5 + l.x 7 + ...
51 7! 10. coshx = 1 + t,x 2 + ,x4 + i x 6 + . ..
11 ex = 1 + x + l.x
21
2 + l.x 3 + l.x 4 +
3! 4! :
12. e-x = 1 - x + t,x 2 - t, x 3 + ,x 4 -
13. Show that the series O!+ l!x + 2!x2 + 3!x3 + converges only for the trivial case,
X = 0.
14. Mae Danerror writes out the first few terms of the series
oo I
I l;()n = 1 + O.Olx + 0.0002x 2 + 0.000006x 3 + .. ..
n=O
She figures that since the coefficients are getting small so fast, the series is bound to
converge, at least if she picks a value of x such as 0.7, which is less than 1. Show Mae
that she is wrong, and that the series converges only for the trivial case, x = 0.
15. Amos Take evaluates the Maclaurin series for cosh 10 and gets
cosh 10 = to( 2
~)! 102 n = 1 + 50 + 416 .666 .. . + 1388.888 ... + ... .
He figures that since the terms are increasing so fast, the series could not possibly
converge. Show Amos his mistake by showing him that the series does actually
converge, even though the terms increase for a while.
16. For the Taylor series for ln 0.1 expanded about x = l, construct a table of values
showing the term index, n; the term value, tn; and the absolute value of the ratio
of terms, Itn+i / tn I. Make a conjecture about what number the ratio seems to be
approaching as n approaches infinity. By taking the limit of the ratio, show that your
conjecture is correct, or change the conjecture.
Section12-6: Interval ofConvergence
fora Series-The RatioTechnique 627
Iii!
17. Inverse Tangent Series Problem: The series
x3 xs x?
P(x) = x - 3 + 5 - 7 + . ..
converges to tan - 1 x for certain values of x.
a. Find the open interval of convergence of the series.
b. On the same screen, plot the graphs of tan - 1 x and the fourth and fifth partial sums
of the series. How do the graphs confirm what you found algebraically in 17a.
c. Evaluate the fourth partial sum of the series for x = 0.1.
d. Find the value of the tail of the series after the fourth partial sum by comparing
your answer to 17c with the value of tan - 1 0.1 you obtain with your calculator.
e. Show that the remainder of the series in 17d is less in magnitude than the absolute
value of the first term of the tail of the series. y
18. Volume Problem: Figure 12-6e shows the solid generated
by rotating about the y-axis the region under the graph of
y = x 2sin 2x from x = 0 to x = 1.5.
a. Find the volume of this solid. Do the integrating numer-
ically, and write all the decimal places your grapher will
give you .
b. Find the volume exactly by integrating by parts then Figure 12-6e
using the fundamental theorem.
c. Evaluate your answer to 18b. Write all the decimal places your calculator will give
you. How does this exact answer compare with the numerical answer you got in
18a.
d. Write the integrand for the indefinite integral in 18a as a Maclaurin series. Then
do the integration.
e. Show that x = 1.5 is in the interval of convergence for the integrated series in 18d.
f. Estimate the volume of the solid by evaluating the first five non-zero terms of the
series in 18d. How does this estimate compare with the exact answer in 18b?
g. The integrated series in 18d is an alternating series whose terms decrease in value
and approach zero as n approaches infinity. Thus the remainder of the series after
a given partial sum is no larger in magnitude than the absolute value of the first
term of the tail following that partial sum. How many terms of the series would
you need to use in order to estimate the volume correct to 10 decimal places 7
19. The Error Function: Figure 12-6f shows the normal distribu-
tion curve, sometimes used to "curve" grades. Its equation y
is Area= erfx
y =- 2 e - t 2
fo
The area of the region under the graph from zero to infinity
-2 -1 2
turns out to be l. So the area from t = 0 to t = x is the X
fraction of a normally distributed population that is within
x units of zero . The resulting integral is called the error Figure l 2-6f
function of x, written erfx.
erfx = }rrfoxe- 12
dt
628 Chapter
12: TheCalculus
of Functions byPower
Defined Series
,.
The fundamental theorem cannot be used to evaluate erfx since e- 12 is not the
derivative of an elementary function. But a power series can be used. Let
f (x) = r e-
12
dt.
a. Find the Maclaurin series for f (x). Write enough terms to show clearly the pattern .
b . On the same screen, plot the sixth partial sum of the series for f (x) and the value
of f(x) by numerical integration. Use an x-window of about - 5 :,; x :,; 5. For what
interval of x-values does the partial sums graph fit the numerical integration
graph reasonably well?
c. Does the series in 19a converge for all values of x7 Justify your answer.
d. Does erf x really seem to approach 1 as x approaches infinity? How do you tell?
20. The Sine-Integral Function: The function
f(x) = rxsint dt
Jo t
is called the sine-integral function of x, abbreviated Six (no pun intended!) Since the
antiderivative of (sin t) / tis not an elementary transcendental function, values of Six
cannot be found directly using the fundamental theorem. Power series give a way to
do this.
a. Write a power series for the integrand by a time-efficient method . Integrate the
series to find a power series for Si x.
b. Is the radius of convergence for the Si x series the same as that for the integrand
series?
c. Find the third partial sum of the series for Si 0.6. How does this value compare
with the value you get by numerical integration?
d. Plot the graph of Si x by numerical integration. On the same screen, plot the graph
of the tenth partial sum of the series for Six. Use an x-window of about - 12 to 12.
For what interval do the partial sums seem to fit the numerical integration values
reasonably well?
nth root of tn
21. The Root Technique: A series of positive terms can be shown
to converge if the nth root of the nth term approaches a
1 ---------------
constant less than 1 for its limit as n approaches infinity.
. . .
Figure 12-6g shows such a series. L - - - - -- - - - - - - -~ - - - - ' ' - - - - - - -.- .
'
lim '!/f;,where L < 1.
Let L = n-oo
a. Show that for any number E > 0 there is a number k such n
that if n > k then ';ff,; < L + E. 5 k 10
b. Show that E can be made small enough so that L + E is Figure 12-69
also less than 1.
c. Show that for all integers n > k, tn < (L + E)n - k
d. Show that the tail of the series after t 11 is bounded above by a convergent geometric
series.
e. Explain how the above reasoning verifies that the series converges.
Section12-6: IntervalofConvergence
fora Series-TheRatioTechnique 629
The result of Problem 21 is called the root tec hniqu e, or roo t tes t, stated as follows :
Technique
: TheRoot Technique
for Convergenceof Series
00
For the series I tn, if L = n-limoo '1ftnlthen:
n= l
i. The series is absolutely convergent if L < 1,
ii. The series is absolutely divergent if L > 1,
iii . The series may either converge or diverge if L = 1.
22. A Special Limit Problem: To use the root technique it helps to know the limit of the
nth root of n.
Let L = n-oo
lim ~
Prove that L = l. (Try taking lnL, finding its limit, then raising e to that power to get L)
Property:Limitof the nth Root of n
Jim
n-oo
'Zin= 1
23. Use the root technique to show that the open interval of convergence of the Taylor
series for lnx is O < x < 2.
24. Use the root technique to show that I ~=J ~xn converges for all values of x.
n
25. Use the root technique to show that .Z::
~=1 nnxn converges only for x = 0.
26. "Which One Wins?" Problem: In Problems 13 and 24, you showed that I ~=l n'xn
converges for no values of x except x = 0, and that I ~=l ~xn converges for all values
' n
of x . For what values of x does I ~=l n~ xn converge?
n
12-7 Convergence of Series at the Ends of the
Convergence Interval
In Section 12-6, you learned the ratio technique for finding the interval of x-values
for which a power series converges . Since the limit of the ratios of terms values is
1 or - 1 at the endpoints of the interval, other techniques are needed to test for
convergence there.
OBJECTIVE Given a series of constants for which the ratio technique is inconclusive, prove either that
the series converges or that it diverges .
63 0 Chapter
12:TheCalculus DefinedbyPower Series
of Functions
To accomplish this objective, it is helpful for you to consolidate your knowledge
about the tail and the remainder of a series.
Definitions:Tailand Remainderof a Series
The tail of a series is the indicated sum of the terms remaining in the series beyond the
end of a particular partial sum.
Example:
2 + 3 + 5 + 7 1+ 11 + 13 + 17 + 19 + 23 + The series of primes.
4th partial sum I --Tail -- -
The remainder of a series, Rn, is the value of the tail after partial sum Sn, provided the tail
converges.
Examples:
For 1 + + + + T6+ , R5 = T6because S5 = 1 f
, the series converges to 2,
and 2 - 1 f= -fs.
For the series of primes, R4 is infinite because the series diverges.
See the box in Section 12-2 for other vocabulary relating to series.
Convergence
of Sequences
There is one major propert y of sequences that leads to several methods of testing
for convergence. You might think at first that a sequence converges if there is an
upper bound for the terms of the sequence. Not true! The sequence
2, 3, 2, 3, 2, 3, 2, 3, . ..
is bounded above by 3, and does not converge . It diverges by oscillation. However,
a sequence such as
1 -2 -3 -4 -5
-
2 ' 3 ' 4 ' 5 ' 5
does converge since the terms are strictly
increasing as well as being bounded above Upp er bound
1 ---- - -----
(Figure 12-7a). The number 1 is an upper bound
for the terms because the numerators are always
less than the denominators. Term tn = n / (n + 1)
can be made arbitrarily close to 1 by picking a
0. 5
large enough value of n. Beyond that value of
n, the terms are even closer to 1 since they are
strictly increasing. Thus 1 is the limit of tn as n
approaches infinity. n
5 10
In Problem 23 of Problem Set 12-7, you will
prove that a sequence converges if its terms are Figure 12-70
bounded above and are strictly increasing.
Section12-7: Convergence
of Seriesat the Endsofthe Convergence
Interval 631
Property: Convergence
of Sequences
If a sequence {t 1, t2 , t3 , . .. , tn , . . . } is increasing and bounded above, then the sequence
converges.
Somep -SeriesConverge,
andOthersDon't
-b
The ser ies 1 + + + ~ + is called a p-series because each denominator is a
power of the term number. In this case , p = 3. The tail of the series following the
terms shown is
1 1 1
-+-+-+
53 63 73 .
f(x) or tn
Figure 12-7b shows how the terms of a series
can be bounded above by an improper integral.
The sequence {tn} is said to be embedded in
the continuous function f (x) = 1/ x 3 becaus e f(x) =-
X
f (x) = tn whenever x = n . Drawing 1 unit to the .01
left of the graph at each point gives inscribed
rectangles, each of whose area equals tn. So the
integral of 1 /x 3 from 4 to infinity,
00
1 x3 dx
4
__!__
is an upper bound for the tail of the series
starting at n = 5. The tail is a lower Riemann Figure l 2-7b
sum for the integral. Integrating gives
i::
n =S n 3
__!__< r
4
=
__!___dx
x3
lim ( - !x - 2
b -co 2
lb)= lim ( - !b2 -
4 b -co
2
+ ! . 4- 2 )
2
=I...
32
The sequenc e of partia l sums in the tail is incr easing because each term is positive
and is bounded above by 1/32 . Thus the series converges because th e tail converges.
In addition, you know that 1/ 32 is an upper bound for R 4 , the remaind er after partial
sum S4.
The p-series
1 1 1 1
-+
10.6 -+-+
2~6 3~6 -40~ +
diverges, even though the terms,
1 + 0.65975 3 . .. + 0.517281 . .. + 0.435275 ... + ,
are decreasing and approach zero for a limit. Figure 12-7c shows the tail after
four terms embedded in f (x) = 1/ x06. Drawing rectangles 1 unit to the right of the
graph at each point makes the tail an upp er Riemann sum for a divergent improper
integral.
I
00
1
3n > f co 1
06
.
dx = hm 2.5x I
0.4
( b)
n =S 5 X b- co 5
= lim (2.5 b 04 - 4.759 . .. ) = oo
b -co
632 Chapter 12: TheCalcu
lusofFunctionsDefinedbyPowerSeries
w
Since the tail is bounded below by infinity, the f(x) or tn
tail is also infinite. Thus the series diverges. 0. 5 1
f (x)= -X o.,c
The p-series will converge if p > 1 because
the exponent - p + 1 in b - P+ l will still be
negative after the integration. The series will
diverge if p < 1 because the exponent - p + 1
in b - p+ l will be positive after integration.
ts t
If p = 1, the series becomes a divergent 6
t7 ts tg .. .
harmonic series, as shown below.
n
56 78910
Figure l 2-7c
Property:Convergenceol a p-Series
The p-series
n=l
I nP
1
= fe + tr;+ f,, + ;f,,+
converges if p > 1, and diverges if p ~ 1.
TheHarmonic
SeriesDiverges
The series
1 +!+!+!+!+!+
2 3 4 5 6
is called a harmonic series. Its terms are
the reciprocals of the terms in an arithmetic f (x) or tn
series. The partial sums, 0.2
5 2 I 2 17
1 , 1 2I 1 6 12 60' .. ,
are increasing. But they are not bounded
above. As shown in Figure 12-7d, the tail after
five terms can be embedded in f(x) = 1/ x,
0.1 -..
~
,.
and the series written as an upper Riemann tg . . .
sum.
1 1
I- >
00
n )5 X
00
1 - dx=lim(lnb
n-oo
- 1n6) = 00
6 7 8 9 10
n
n=6
Since the tail is bounded below by infinity, the Figure l 2-7 d
tail is also infinite. Thus the series diverges.
TheAlternating
Harmonic
SeriesConverges
The series
is called an alternating harmonic series. The partial sums,
1 5 7 47 37
l, 2' 6' 12' 60' 60' ,
Section
12-7:Convergence
of Series
at theEndsof theConvergence
Interval 633
get bigger and smaller. But as shown in Figure
12-7e, they seem to be converging toward
some number between 1/2 and 1.
Associating pairs of terms gives _
? ..,._________ _________________
) + (- 1) + (-)+ (~ - )+
(1- ...
0 .5
= !2 +l...+
12
l...
30
+ l...+
56
.. ,
Since each term of the series of differences
is positive, the sequence of even-numbered
n
partial sums is strictly increasing . That is,
5 10
S2, S4, S 5 , Ss, ... = 0.5, 0.58333 ... ,
Figure 12-7e
0.61666 ... , 0.634523 .. . ,
Associating another way gives
1 - (!2 - l)
3
- (!4 - !)
5
- ( !6 - !)
7
- (!8 - !9 ) + = 1 - !6 - _l_ -
20
1... - 1...
42 72
-
Since each term after the first is subtracted from 1, the odd-numbered partial sums
are bounded above by 1.
S1 , S 3 , S 5 , S 7 , S 9 , .. . = 1, 0.83333 .. . , 0.78333 ... , 0.75952 ... , 0.74563 ... , ...
Thus the sequence of even-numbered partial sums converges because it is increasing
and bounded above. The odd-numbered partial sums approach the same limit since
the terms approach zero. By computer, the limit is 0.69314 7. .. , which happens to
equal ln2.
In general, a strictly alternating series will converge if the absolute values of its
terms decrease uniformly toward zero for a limit.
Property:AlternatingSeries ConvergenceTest
If a series t 1 - t2 + t3 - t 4 + has the following properties,
the terms in the tail are strictly alternating in sign,
the terms in the tail are strictly decreasing in abso lute value, and
the limit of tn is zero as n approaches infinity,
then the series converges.
Corollary: The remainder is bounded by the first term of the tail, IRn I < Itn+1 I.
A formal proof of the theorem would duplicate in general the reasoning in the
preceding example.
AbsoluteConvergence
of a Series
If the signs of the terms are not strictly alternating, or if all the terms are positive,
it is more difficult to prove convergence. What is normally done is to consider all
terms to be positive (which makes the sequence of partial sums increasing), then
634 Chapter
12: TheCalculus
of Functions byPowerSeries
Defined
look for an upper bound . If the series of abso lute values of terms converges, the
series is said to be absolut ely convergent.
Definition:Absoluteand ConditionalConvergence
A series t 1 + t2 + t3 + t4 + is said to be absolutely convergent if the series of absolute
values of the terms,
l~ l+l ~ l+l ~ l+l ~ I+
converges.
If a series converges but is not absolutely convergent, then it is said to be conditionally
convergent.
Absolute convergence may be considered to be a "worst case ." It is most difficult
to make a series converge if all terms have the same sign. If a series is absolutely
convergent, it converges even though some of its signs may be negative and some
positive.
There are several ways to establish upper bounds for a series of positive terms.
These methods, or tests for convergence, are summarized in the following box. You
will see why some of these tests work in Problem Set 12-7.
Properties:Tests for Convergence
of a Seriesof Constants
Increasing Series Test
Converges if the tail is bounded above by a convergent integral or series.
Diverges if the tail is bounded below by a divergent integral or seri.es.
Alternating Series Test
Converges if these three conditions apply:
1. The terms in the tail are strictly alternating in sign.
2. The absolute values of the terms in the tail are strictly decreasing.
3. The limit of tn is O as n approaches infinity.
Upper bound for the remainder: ltail l < lfirst term of tail l.
nth Term Test Diverges if n-
limootn * 0.
Geometric Series Test
Converges if Icommon ratio ! < 1.
Diverges if Icommon ratio I 2: 1.
p-Series Test
Converges if p > 1.
Diverges if p !, l.
Harmonic Series Test
Diverges if all signs are the same.
Section
12-7:Convergence
of Series
at theEndsof theConvergence
Interval 635
Armed with these properties, you can now determine the convergence of power
series at the endpoints of the interval of convergence, where the ratio technique is
inconclusive.
Example 1 The open interval of convergence for the Taylor series for lnx expanded about x = l
is O < x < 2. Determine whether or not the series converges at x = 0 and at x = 2.
Solution The series for lnx is lnx = (x - 1) - (x - 1) 2 + (x - 1) 3 - (x - 1) 4 + .. .
At x = 0, the series becomes - 1 - - - - , which diverges because it
is the opposite of the divergent harmonic series . At x = 2, the series becomes
1 - + - + , which converges because it meets the hypotheses of the
alternating series test.
.-. the comp lete interval of convergence is O < x 5 2.
"' 2n(x-l) 11
Example 2 Find the complete interval of convergence of the power series I ( )
n =O 1n n + 2
Solution By the ratio technique,
. 12n+l(x - l)n +1 ln (n + 2) I
L = hm00
11- ln(n + 3) 2n(x - l)n
hm
. I ln(n
= 2 lx - l l 11-00 (
+ 2) I oo
) - -oo !'Hos pital 's rul e appli es .
1n n + 3
. In+
= 2 1x - l l hm --
n-oo
3 I-
n+2
-o
0
!'Hos pital's rul e appli es again .
= 2 Ix - 11 1 = 2 Ix - 1 1
L < l <=> 21x - 11 < 1 <=> < X <
At x = , the series is
1 1 1 1
ln 2 - ln 3 + ln 4 - ln 5 + '
which converges because it meets the hypotheses of the alternating series test .
At x = , the series is
1 1 1 1
ln 2 + ln 3 + ln 4 + ln 5 + '
for which the terms starting at 1 / ln 3 are larger than + + + , a divergent
harmonic series .
.'. the complete interval of convergence is 5 x < ,
Sometimes it is necessary for you to tell whether or not a given series of constants
converges, even if the series does not come from the endpoint of a Taylor series.
Examples 3-5 show you how to determine whether such series converge.
Example 3 Determine whether or not the series converges. I (1+ ~)n
n= l
636 Chapter
12: TheCalculus ns DefinedbyPowerSeries
of Functio
Solution
Example 4
The series begins 2 + 1.25 + 1.1111 ... + 1.0625 +
.-. the series diverges because tn approaches 1, not 0, as n approaches
~~-
Determine whether or not the series converges. I""
n=O n 2
n
+7
.
Solution . b.egms 0 + l + TT
Th e senes 2 + 3 + 4 + 5 + 6 + , whihc 1s
" equa 1 to
8 16 23 32 43
0 + 0.125 + 0 .1818 . .. + 0.1875 + 0.17391 .. . + 0.15625 + 0.13953 ... + .
The series might converge because the terms are decreasing after a while and
approach 0. (The n2 in the denominator dominates the n in the numerator, as
you could tell by !'Hospital's rule) . Since it's reasonably easy to integrate the
given function, you can compare the series with an improper integral. The lower
limit of integration can be any nonnegative number, since only the tail of the
series is in question.
r""_ x_ dx = lim rb_x_ dx
Jo x 2 + 7 b -oo Jo x 2 + 7
b
= blim
-oo
ln lx2 + 7110
= lim
b -oo
In
lb 2
+ 71- In7 = oo
.-. the series diverges because the tail could be bounded below by a divergent
in~~ill.
Example 5 Determine whether or not the series converges. f -n 1-+ 7
n=o
-
2
Solution 'b
Th e senes " 11 + + ii1 + l + I + .
egms 7 8 16 23
The series might converge because the terms decrease and approach O as a limit
(tn approaches the form 1 / oo). Each term after the first one in the given series is
smilller than the corresponding term of a convergent p-series .
1 I 1
p-series: I; =1 : 2
1 +4+9+l(3+
Value of n: 1 2 3 4
Given series:
, oo 1 1 1 I
= 7 + g+u+15+23 +
L.n =O ~7
I l
n +
.. the series converges since the partial sums are increasing and bounded above
by the limit of a convergent p-series.
Problem Set 12-7
DoTheseQuickly
The following problems are intended to refresh your skills. You should be able to do all
ten problems in less than five minutes.
QJ. 7 + 14 + 28 + 56 + are terms of a - ?- series.
Q2. The next term in the series in Ql is found from the preceding term by-?-.
Section 127:Convergence
of Seriesot the EndsoftheConvergence
Interval 637
~
Q3. The number 2 for th e series in Ql is called the -? - of the series.
Q4. ex- 1) - ex - 1) 2 + ex - 1) 3 - is the Taylor series expansion for- ?-.
QS. The first three terms in the Maclaurin series expansion of cos 2x are-?-.
Q6. The coefficient of x 6 in th e Maclaurin series expansion off ex) = sinx 2 is -?-.
Ql. If the interval of convergence is 3 < x < 8, then the radius of convergence is -?- .
QB. The open interval of convergence for nfo
is-?-.
ex- n4)n
oo
3
Q9. The volume of the solid formed by rotating the region under the graph of y = 9 - x 2 about
th e y-axis is -? - .
Q10. If ret)= ee21)7 + esin3t)], then vet)= - ?- .
00 6
l. Vocabulary Problem I: For the series I -
n=l n!
a. Write th e first five terms of the series.
b . Calculate the first five terms of the sequence of partial sums.
c. Write the first three terms of the tail of the series after the fifth partial sum.
d. Are th e terms of the series increasing or decreas ing?
e. Are th e partial sums of the series increasing or decr easi ng?
2. Vocabulary Prob lem II: Explain why a series of positive terms always has an increasing
sequence of partial sums, even if the term values themselves are decreasing .
00 6
3. Vocabu lary Problem III: For the series I
e- l)n +J -
n=I n!
a. Write the first five terms of the series .
b. Calculate the first five terms of the sequence of partial sums.
c. Write the first three terms of the tail of the series after the fifth partial sum .
d. What is th e effect of the e- 1) n + 1 factor in the formula for the series?
e. Show that the sequence of even-numbered partial sums is increasing.
f. Show that th e sequence of partial sums is bounded above by the first term of the
series.
4. Vocabulary Problem N: The series 6 - 3 + 1 - 0.25 + 0.05 - from Problem 3, above,
converges to 3.792723 . If the series 6 + 3 + 1 + 0.25 + 0.05 + also converges,
then the former series is said to be -?- convergent. If the series 6 + 3 + 1 + 0.25 +
0.05 + does not converge, then the series 6 - 3 + 1 - 0 .25 + 0.05 - is said to
be- ?- convergent .
For Problems 5 through 14, find the interval of convergence for the given power series,
including convergence or divergence at th e endpoints of the interval.
00
oo 5"xn
5. L nex-3)n 6. I -
n=l n=l n2
00 xn 00
e- l)nex-6)n
7. I- 8. I
n2n
n=l n n=4
e-l)n +lex + 5)2n
00
ex+ l)n
00
9. I 10. I
n=l 2n n=l n2
638 Chap
ter 12:The Calcul
us of Functions Defined byPowerSeries
11. I ln(n + 1) xn
12. I
00
5(x - 3)n
n =O n+1 n= I
oo 4n 00 1
13. I - 14. I -
n =O xn n = l xn
15 . Upper Bound by Convergent Improper Integral : Given
00 1
S= I 2 .
n =I n f (x) or tn
a. Write the first five terms of the series, and calculate the .04
fifth partial sum, S 5 .
\. f (x) = 1,
''.... x-
b. The tail of the series after S5 is
00
Rs = L -n1
n =G 2
= -
1
36 49
1 1
+ - + - + ...
64
.02
Figure 12-?f shows that the tail is a lower sum for the
improper integral
n
Joo_!_dx
5 x2 s 6 7 s 9 10
Why is the lower limit of summation 6, but the lower Figure 12-71
limit of integration is only 5?
c. By evaluating the improper integral, find an upper bound for the tail of the series.
d. Explain how the above reasoning allows you to conclude that the original series
converges.
e. Estimate th e number to which the series converges by summing the first 1000
terms.
f. Find an upp er bound for the remaind er of th e series in part (e).
g. True or false: "The remainder of the series is close to the value of the first term of
the tail."
h. Complete: "The value of the series using 1000 terms has no more than a-?- % error."
i. How many terms would it take to guarantee that th e value of the series is correct
to the sixth decimal place (error less than 0.000000 5)?
16. Divergence by Comparison with Divergent Improper Integral: Consider the seri es I l..
n
n =l
a. Find S5 , the fifth partial sum of the series.
b . The tail after Ss is f (x) or tn
00 0.2 .... 1
1 1 1 1 \ t X)= x
Rs = L- =- +- +- + ...
n 6 7 8
~
n=6 I~
t
The improper integral f;' dx is a lower bound for this tail ,, --
'.
(Figure 12-7g). Evaluate the integral. How do es the result 0.1
tell you that the series diverges ?
c. Why would it b e futile to try to evaluate this series by
computer ?
d . True or false : "If the terms of a series approach zero for a n
limit as n aproaches infinity, then the series converges." 6 7 8 9 10
Figure 12-79
Section12-7: Convergence
of Seriesat theEndsoftheCon
vergence Interval 639
17. Integral Test Problem : Use the results of Problems 15 and 16 to explain th e following:
Technique:TheIntegralTest for Convergenceof Series
GivenS = f f (n) and I =
n=O
r
a
f (x) dx where f(x) decreases monotonically to 0.
S converges if I converges, and S diverges if J diverges.
18. Follow-Up Problem: In Problem 15, you conclud ed that
1... + 1... + ..!..+ ...
36 49 64
f (x) or tn
converges because it is a lower sum for an improper integral . You ... 1
\ f(x) = :;;z
could also have made the series an upper sum for the integral
rb_l_dx
Ja x2 .02
as shown in Figure 12-7h. Explain why you learn nothing about
the convergence of this series by making it an upper sum .
19. The Factorial Reciprocal Series Converges: Suppose that 1 is
substituted for x in the Maclaurin series for e". The result is the
series of reciprocals of factorials . n
00 6 7 8 9 10
1 1 1 1 1 1 1 1
I -=-+-+-+-+-+-+-+
n=O n 1 0 1 11 21 31 4! 51 6! Figure l 2-7h
a. Explain why it would be difficult to find an upp er bound for the tail of the series
using an improper integral.
b . Comparing th e given series with the geometric series with first term 1 and ratio
shows
Term index, n: 0 l 2 3 4 5 6
Given:
l l l l l l
Geometric: 1 +2+4 + g+15+ 3"z+ 6-1+ ...
1--TaH- - -
From term t4 on, the geometric series terms are upper bounds for the terms of the
given series. To what number does the tail of the geometric series converge?
c. Find an upper bound for the remainder of the given series after S3 , the fourth
partial sum .
d. Find an upper bound for th e entire given series.
e. Explain how the reasoning in 19a-d allows you to conclude that the given series
converges.
20. Compar ison Test for the Exponential Function Series: The following ar e the first seven
terms of the Maclaurin series for e0 6 .
n : 0 1 2 3 4 5 6
e 06 = 1 + 0.6 + 0.18 + 0.036 + 0.0054 + 0.000648 + 0.0000 648 +
--- 5th partial sum ---1 - - Tail ---
a. Show how the seventh term, t6 , is calculated.
640 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
b . Show that the fifth partial sum, S4 , differs from the actual value of e 06 by more
than t5 , the first term of the tail, but not by much more.
c. Show that each term of the geometric series
0.000648 + 0.0000648 +
with common ratio 0.1 is an upper bound for the corresponding term in the tail.
d . To what value does the geometric series in 19c converge?
e. Based on your answer to 19d, what number is an upper bound for the sum of the
tail of the series? What number is an upper bound for the entire series? Show that
the latter number is just above e 0 6 .
21. A lternating Series Remainders Property Problem: The following are the first four terms
of the Maclaurin series for sin0.6:
n: 0 1 2 3
sin0.6 = 0.6 - 0.036 + 0.000648 - 0.000005554285714 +
a. Show how t3 , the fourth term, is calculated.
b. Calculate S 1 and S2 , the second and third partial sums, respectively.
c. Calculate R1 and R2, the remainders after S1and S2 (the values of the tail), by finding
the difference between the partial sum and the value of sin 0.6 by calculator. Show
that in both cases , the magnitude of the remainder is less than the absolute value
of the first term of the tail.
d . Use the appropriate property to prove that the series for sin 0.6 converges .
22. Infinit e Overhang Problem: Figure 12-7i shows a pile of blocks. The top block sits
so that its center of mass is exactly on the edge of the second one down. The
third block is placed under the center of mass of the first two. The fourth is
placed under the center of mass of the first three, and so on.
1--L-j
a. Show that the overhangs of the blocks are terms in the harmoni c
sequence L, L, L, L,... where Lis the length of each block. To
find the centroid of a particular pile, find the sum of the moments
of the blocks by summing each one's moment with respect to the
y-axis, then dividing by the number of blocks. You must, of course,
find the centroid of one pile before you can find the centroid of Figure 12-7i
th e next .
b . What depth of pile is the first to have its top block projecting entirely beyond its
bottom block?
c. Explain why, in theory at least, it would be possible to mak e a pil e of blocks with
any desired overhang, using nothing but gravity to hold the pile together .
d. If you pil e up a normal 52-card deck the way the blocks are piled in this problem,
by how many card- lengths would the top card be offset from the bottom card?
Section
12-7: Convergence
ofSeriesat theEndsoftheConvergence
Interval 641
23. Convergence of Sequences Proof: Figure 12-?j shows a sequence t,,
_________Upp er Bound ________
that is increasing and bounded above. Prove that the sequence
.............
..........
..
converges . You use the least upp er bound postulate (which
states that any set of real numbers that is bounded above has
a least upper bound) to establish the existence of a least upper
bound, L, then prove that L is the limit of the sequence. n
1 2 3 4 5 6 7 8 9 10
24. Sequences vs. Series Prob lem: Explain why the sequence 1.9, 1.99,
1.999, 1.9999, ... converges, but the series 1.9 + 1.99 + 1.999 + Figure l 2-7 j
1.9999 + does not converge.
For Problems 25-30, determine whether the series converges or diverges . Justify your
answer.
00 3 co 3n
25. I - 26. I 4n
n=O 4n n=O
00 1 00 1
27. I --- 28. I -
n=o (2n + l)! n=o ( - 3 ) n
oo n3 00
29. I --
n =2 n~ - 1
30. I sinn
n=O
31. Journal Prob lem: Update your journal with things you have learned since the last entry.
Include such things as those listed here .
The one most important thing you have learned since your last journal entry .
The way you find an interval of convergence for a power series.
The ways you can determine convergence at the endpoints of the interval of conver-
gence.
What you understand better, now, about the meaning of convergence.
Any techniques or ideas about series that are still unclear.
12-8 Error Analysis for Series
From time to time in this chapter you have estimated the remainder of a series,
the value of the tail of the series after a certain number of terms . This remainder
represents the error in the value of a function that you get by using a partial sum of
the series. For certain alternating series you found that the entire tail is bounded by
the first term of the tail. In this section you will learn about the Lagrange form for
the remainder of a Taylor series, an expression similar to the first term of the tail.
Joseph Louis Lagrange (1736-181 3) applied mathematics in many areas, including
the motion of planets, and helped set up the French metric system.
642 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
OBJECTIVE Given a series, determine the number of terms needed to obtain an approximation for the
limit to which the series converges correct to a specified accuracy.
The general term of the Taylor seri es expansion of f(x) about x = a is
f (n ) (a)
tn = --- (x - a)n .
n!
The first term of the tail of a Taylor series is tn+l For any value of x in the interval
of conver gence, it turns out that there is a value of c between a and x for which
R n (x ) is given by
f'n + l ) (c) n+l
R 11(x) = (n + l)! (x - a ) .
The only difference between this remainder and the first term of the tail is that
f <n+l l (a) is replaced by f '11+ 1l (c) . Usually this derivative is awkward to calculate .
Often, though, an upper bound, M, can be found for it. If this is the case, an upper
bound can be found for the remainder. Finding the value of n that gives a remainder
sufficiently small allows you to determine how many terms to use to get the desired
accuracy in the partial sum representing f (x ) .
Property: LagrangeFormof the Remainder
of a TaylorSeries
If f (x ) is expanded as a Taylor series about x = a and x is a number in the interval of
convergence, then there is a number c between a and x such that the remainder Rn after
the partial sum Sn is given by
R = f (n + l ) ( c ) (x - a )n+l .
n (n + l )!
If Mis the maximum value of f (n + l l (x) on the interval between a and x, then
IRnl .$ M I n+ 1
(n + l)! x - al .
The property is an extension of the mean value theorem, which concludes that there
is a number c between a and x for which
f'(c) = f(x) - f(a). f(x)
x - a
Multiplying by (x - a) gives f (x)
f (x ) - f(a) = f'(c)(x - a ). T
Error =
As shown in Figure 12-8a, f (x) - f (a ) is the error f(x) - f(a)
in using f(a) as an approximation for f (x).
Solving this equation for f(x ) gives f (a) l
X
f(x ) = f(a) + f'(c)(x - a).
a C X
Figure 12-Ba
Section
12-8: ErrorAnalysis
for Series 643
In th.is form you can see that f(a) is the first term ( n = 0) of the Taylor series
expansion of f(x) about x = a, and f'(c) (x - a) is the Lagrange form of the remainder
after n = 0. You will be asked to supply the algebraic details of the derivation in
Problem 21 of Prob lem Set 12-8 .
Example 1 a. Estimate e 2 using the 11th partial sum (n = 10) of the Maclaurin series fore '.
b. Use the Lagrange form of the remainder to estimate the accuracy of using this
partial sum .
c. How does this estimate of the remainder (part b) compare with the value
calculated by subtracting S10 from the value of e 2 on your calculator?
JO 1
Solutions a. S10 = I - 2n = 7.3889470 . . .
n =O n!
b. All derivatives of ex are equal to ex. Since you are estimating e 2 from scratch,
you should not assume that e is known to be 2.718 .... However, you know
that e < 3. So a value of Mis 3 2' and thus r <n+l) (x) < 9 for all X between 0
and 2.
IR10I < l~! (2 - 0) 11 = 0.0004617 .. .
Thus S to may be off by as much as 5 in the fourth decimal place and thus
should match e 2 to 1 in the third decimal place.
C. e 2 = 7.38905609. . . By calculator.
S1 0 = 7.38899470 . .. From p art a.
The difference is 0.00006138 . . . , which is significantly less than the upper
bound 0.0004617 . .. by Lagrange's form. Note that although the difference
has zeros in the first four decimal places, the partial sum and the more
precise value of e 2 still differ by 1 in the third decimal place.
Example 2 How many terms of the Maclaurin series for sinh x are needed to estimate sinh 4
correct to five decimal places? Confirm your answer by subtracting the partial sum
from sinh4 using your calculator.
Solution The general term of the series is tn = ( n ~ l)!x 2n+I, where n is the term index.
2
All derivatives of sinhx are either coshx or sinhx. Both functions are increasing
on the interval (0, 4], with coshx > sinhx. Thus the derivatives are all bounded
by cosh4. Because you are trying to estimate sinh4, you should not assume
that you know cosh 4 exactly. However, 2 < e < 3, so cosh4 < 0.5(3 4 + 2- 4 ) =
40.53125 < 41, which means that the absolute values of the derivatives are
bounded by M = 41.
JRnl = 41 42n+3 Why 2n + 3?
(2n + 3)'
To get five-place accuracy, IRn I should have zeros in the first five decimal places
and no more than 5 in the sixth place. That is, IRnl < 0.000005 . Using the table
feature,
n 41 / (2n + 3) ! 4 2 n+3
7 0.001980 319 .. .
8 0.0000 92646 . . .
9 0.00000 3529 . . .
64 4 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
The 0.000003529 ... for n = 9 is the firs t value less than 0.000005 . There for e,
you should use at least 10 terms (because the term index, x, starts at 0).
sinh4 = 27.28991719 .. . By ca lculator.
S9 = 27.28991711 . . .
Difference = 0.000000088 . .. , which is conside rably less than 0.000005 .
Example 3 For the Taylor series for lnx expanded about x = 1, how many ter ms would be
needed in the partial sum to compute ln 1.4 to five decimal places 7
Solution lnl.4 = (1.4 - 1) - (1.4 -1) 2 + (1.4 - 1) 3 - . ..
= 0.4 - 0.08 + 0.021333 . .. -
Since this series meets the requirements of the alternating series test , JR,,I <
It,,+1I.
1
Make - - (0.4"+1) < 0.000005 .
n +l
n < 9.731 ... < 10 Solve num erically for 11.
Beca use th e term ind ex start s at 1, it is
Use 10 terms . equal to th e te rm numb er.
As a check, S 10 = 0.3364694 45 ... , and ln 1.4 - S 10 = 0.00000279 ... , whi ch
checks.
Example 4 For the Maclaurin series for e 2 in Example 1, find approximate ly the valu e of c for
which the Lagrange form is equal to the remainder R1o-
Solution R = f ( 11) ( c ) (2 - 0) 11 = ~ 2 11 = 0.00006138 . ..
C
From Exam ple 2.
10 11! 11!
. C = 0.00006138 . .. . 11! =
.. e 211 1. 1965 . ..
c = ln 1.1965 ... = 0.1794 . .. , which is between O and 2.
Example 5 Estimate the remainder of the p-series I
n= l
\
n1 2
after 20 terms.
Solution Since this is a series of constan t s not
necessarily related to a Taylor ser ies, the y or tn
Lagrange form of the remainder does not y = (x - 1r 102
\ is an uppe r bo und .
apply . Instead, compare the tail of the y = x-1. 02 ..
series with an improper integra l that is is a lowe r \
an upp er bound. bound. \ ....
Figure 12-8b shows the first few terms of
the tail starting at term t 2 1. The area of Rzo
begins
each rectangle in the histogram equals at t2 1. /LI
th e resp ective term value, and thus th e t22
t23 tz4
sum of their areas equals the value of n or x
the tail, R20 . The area under the graph of 21 22 23 24 25
y = x- 1.o2 is a lower bound for R20 since
the rectang les form an upper Riemann Fig ure l 2-8b
sum. If the graph is shifted one space to
Section 128: ErrorAnalysis for Series 645
the right, the equation is y = (x - 1) - 1.20 , and the area under the shifted graph
is an upper bound for R 20 . The areas of each region can be expressed as an
improper integral whose value is an upper or lower bound.
R 20 =
00
n =2l
1
I --02
n 1.
> r
x - l.02 dx = lim (-Sox - 0 02 1t1)
21 b -oo
= lim (- SOb- 002 + 50. 21- 00 2 )
b -oo
= 0 + 47 .04631 .. . = 47.04631 . . .
Similarly,R 20 < J (x -
00
2 1.
1) - 10 2 d x = lim[ -S O(b - 1) - 0 -02 + 50. 20- 002 ] = 47 .09224 ... .
b -oo
So R 20 is between 47.046 ... and 47.092 .... Since S20 is only 3.50977 . .. , the
series is nowher e close to converging after just 20 terms.
Problem Set 1 2-8
DoThese
Quickly
The following problems are intended to refresh your skills . You should be able to do all
ten problems in less than five minutes.
QI. The radius of convergence of a power series can be found using the -7- technique .
Q2. A geometric series converges if and only if -?- .
Q3. The Maclaurin series for cos x converges for what values of x?
Q4. The Taylor series for lnx, expanded about x = 1, has what radius of convergence?
QS. Write the first four non-z ero terms of the Maclaurin series for tan - 1 x.
Q6. Sketch a partial sum of the Maclaurin series for sine compar ed to the actual sine graph .
Ql. Do th e integratin g: f sec x dx
QB.Differ entiat e: y = tanx
Q9. Do the inte grating: f sec 2x dx
QlO. v\Thichtwo people are credited with having invente d the calculus ?
For Problems 1 through 4,
a. Find the indicated partial sum.
b. Use the Lagrang e form of the remainder to estimate the number of decimal places to
which the partial sum in part (a) is accurate.
c. Confirm your answer to part (b) by subtracting th e partial sum from the calculator
value.
l. cosh4 using the sixth partial sum (n = 5) of th e Maclaurin series
2. sinh 5 usin g th e tenth partial sum (n = 9) of the Maclaurin series
3. e 3 usin g the 15th partial sum of the Maclaurin seri es
4. ln 0. 7 using 8 terms of th e Taylor series expansion about x = 1
For Problems 5 throu gh 8, use the Lagrange form of th e remainder to find th e numb er of
terms need ed in the par tial sum to estimate the function value to the specified accuracy.
5. sinh 2 to 6 decimal places using the Maclaurin series
6. cosh 3 to 8 decimal pla ces using the Maclaurin series
646 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
7. ln 0.6 to 7 decimal places using the Taylor series expansion about x = 1
8. e 10 to 5 decimal plac es usin g the Maclaurin series
For Problems 9 and 10 calculate the value of c in the appropriate int erval for which the
Lagrange form of the remaind er is equal to the remainder calculate d by subtracting the
partial sum from the function value by calculator.
9. cosh 2 using 5 terms (S4)
10. e5 using 20 terms (S 19 )
For Problems 11 and 12 show that th e hypoth eses of the alternating series test apply to
the function, then find the number of terms needed in the partial sum to get the specified
accurac y.
11. cos 2.4 to 6 decimal plac es using the Maclaurin series
12. e- 2 to 7 decimal pla ces using the Maclaurin series
13. p-Ser ies Probl em I: For the p-seri es I; =1 (1/ n 3 ) find an upper bound for the tail of the
series after 10 terms. How does this estimat e of the error introduced by stopping at
S1o compare with the value of tu? How many terms would be needed to ensure that
the partial sum is correct to at leas t five decimal pl aces?
14. p-Series Prob lem II: The series
1 1 1
l +-+ - + -+
-12 ,./3 v'4
is a p-series. Explain why th e method of Problem 13 would not b e appropriate for
estimating the remainder of thi s series.
00
15. p-Series Problem III: Amos Take wants to calculate the limit to which I (1 / n 1.os)
n= l
converges. With his grapher he calcu lat es S99 = 4.69030101 .... Show Amos that
although he has used man y terms of th e series, his answer is nowhere close to the
value to which th e seri es converges.
16. Ratio Technique and p-Series Problem : The ratio technique can be used to determine
whether or not a series of constants converges. If the limit of the ratio of adjacent
terms is less than 1 in absolute value, then the series converges . Show that the ratio
technique neve r gives conclusive results for a p-series.
17. Geometric Series as an Upp er Bound Problem : In Exampl e 1 of this section you saw
that the error in calculating e2 using the eleventh partial sum (S 10 ) of the Maclaurin
series could be estimated using the Lagrange form of the remainder. The error can
also be estimated by boundin g the tail of the series with a convergent geometric
series that has first term equal to tu (the first term of the tail) and common ratio
equal to t 12 / tu . Which method gives a bett er estimate of the error, the geometric
series or the Lagrang e remainder?
18. Values of ex from Values of e- x Problem: The value of e2 can be calculated by first
finding the value of e- 2 , th en taking the reciprocal. After the first few terms th e series
for e- 2 meets the hypothes es for the alt ernating ser ies test. Thus the error for any
partial sum is bounded by th e first term of the tail of the series after that partial sum.
Estimate the error in the estimate of e- 2 using the eleventh partia l sum (S 10 ). Then
estimate e2 by calculating l / S 10 . Is the error in the answer any smal ler than the error
in using S10 directl y for e 2 as in Exampl e 1 of this section?
Section 12-8:ErrorAnalys
is for Series 647
19. Sin x for Any Argument Using a Value of x in [O, TT/4] y
Problem: As you have seen, the Maclaurin series for sin x
converges more slowly the farther x is away from 0. Suppose
that you wanted to compute sin 250.
a. The values of sin x repeat themselves with a period of
2TT.Find a number bin [O,2TT] for which sinb = sin 250.
b. Each value of sinx for x in [O,2TT]is equal to a value of
sine for some number c in [ - TT/2, TT/ 2] (Figure 12-8b). Figure l 2-8c
Find the value of c for which sin c = sin 2 50.
c. Each value of sin x for x in [ -TT/2, TT/2] is equa l to sin d or cos d for some
number din [O, TT/4] (Figure 12-8c). Find the value of d for which sin 250 = sind
or cos d. Demonstrate that your value of d gives the correct answer for sin 2 50.
d. Show that the technique of this problem can be used to calculate values of sinx
correct to at least 10 decimal places using just six terms of the appropriate
Maclaurin series. How many terms are needed to calculate sin 250 to 10 places
directly from the series?
e. Project: Write a program to calculate sinx correct to 10 decimal places by
Maclaurin series using the technique of this problem. Programs similar to this are
used internall y by calculators to evaluate sines and cosines efficiently.
20. The National Bureau of Standards Handbook of Mathematical Functions lists the value
of sin 1 to 23 decimal places as 0.84147 09848 07896 50665 250. (The spaces are
used in lieu of commas for ease of reading.) How many terms of the Maclaurin series
for sinx would have to be used to get this accuracy? How many terms would it take
if the technique of Problem 19 were used7
21. Derivation of th e Lagrange Form of the Remainder: Earlier in this section you saw
that the conclusion of the mean value theorem leads to a spec ial case of the Lagrange
form of the remainder. If function f has derivatives of all orders, as do exponential,
trigonometric, hyperbolic, and many other functions, then the mean value theorem
applies to each derivative.
a. Show that applying the mean value theorem to f'(x) on the interval [a, x] gives
f'(x ) = f'(a) + f " (c) (x - a)
for some number c between x and a.
b. Assume that a and c are constants and x is the variable. Integrate both sides of
the differential equation in part (a) with respect to x. Use the point (a, f (a ) ) as
the initial condition. Show that the answer can be transformed to
f(x) = f(a) + f'(a)(x - a) + f " (c)(x - a) 2
c. You should recognize that the first two terms of the right-hand side of the
equation in part (b) are terms in the Taylor series expansion of f(x) about x = a,
and the third term is the Lagrange form of the remainder. By applying the mean
value theorem to f " on the interval [a, x] and integrating twice show that there is
a number c in (a, x) for which
f(x) = f(a) + f'(a)(x - a) + tff " (a)(x - a) 2 + t f' " (c)(x - a) 3
d. Without actually doing the algebra, name the mathematical technique that could
be used to prove for any integ er n > 0 that there is a number c in the interval
648 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
(a, x) for which the Lagrange form of the remainder is exactly equal to the error
in usin~ the partial sum S,,(x) of the Taylor series as an approximation for f(x).
22. A Pathological Funct ion: Figure 12-8d shows the function {(x)
f(x) = {e-x-z, if x * 0
0, If X = 0
Function f has derivatives of all orders at x = 0, and each derivative X
equals zero there.
a. By equating derivatives , show that the Maclaurin series for f(x) Figure l 2-Bd
would be O + Ox + Ox2 + Ox3 + ... .
b. Show that the Maclaurin series converges for all values of x, but that it does not
converge to f(x) except at x = 0.
c. Substitute - x- 2 for x in the Maclaurin series for ex. Write the first four terms of
the power series and simplify.
d. The resulting power series is called a Laurent series, the name applied to a power
series in which some powers can have negative exponents . By finding a partial
sum of the series, make a conjecture about whether or not the Laurent series
evaluated at x = 2 converges to f(2).
23. The Maclaurin Series for ex Converges to eX: If you worked Problem 22 you found
that a Maclaurin series can converge, but not to the target function. Use the Lagrange
form of the remainder to show that the Maclaurin series for eX does converge to ex
for all valu es of x by showing that the remainder of the series approaches zero as n
approaches infinity.
12-9 Chapter Review and Test
In this chapter you have seen how a function can be expanded as a power series.
There are two advantages to doing this. First, it allows you to calcu late values of
transcendental functions by doing only the four operations of arithmetic . Calculato rs
and computers use series internally to calculate values of sines, logs, and so forth.
Second, it allows you to determine how accurate a numerical integral is . If xis within
the interval of convergence, you can get any desired accuracy by using enough terms
of the series.
The Review Problems are numbered according to the sections of this chapter. The
Concepts Problems allow you to apply your knowledge to new situations. The
Chapter Test is more like a typical classroom test your instructor might give you.
Review Problems
RO. Update your journal with the things you have learned since the last entry. You should
include such things as those listed here.
The one most important thing you have learned in studying Chapter 12.
Which boxes you have been working on in the "define, understand, do, apply" table.
How to write from memory some special, well-known series.
Section 129: Chapter
Review
andTest 649
How you find a power series for a given function, either by equating derivatives or
by operating on another known series.
What it means for a sequence or series to converge .
How you determine the accuracy of a function value found by series .
Any ideas about series that are still unclear.
Rl. Let f(x) = t:'.x and let P11 (x) = 9 + 9x + 9x 2 + 9x 3 + 9x 4 + 9x 5 + + 9x". On the
same screen, plot f(x), P5 (x), and P6 (x) . Sketch the results . For what values of x does
P11 (x) seem to be close to f(x), and for what values of x does the graph of P bear little
resemblance to the graph off? Show that P6 (0.4) is closer to f(0.4) than P5 (0.4) is.
Show that P5(0), P5' (0), and Pt (0) equal f ' (0), f " (0), and f " ' (0), respectively . What
kind of series is P11(x) a subseries of?
R2. a. Biceps Problem: You start an exercise program to increase
the size of your biceps. The first day you find that the
circumference of each bicep increases by 3 mm . You
assume that the amount of increase on each subsequent
day will be only 90% of the amount of increase the day
before. By how much do you predict your biceps will
have increased after 10 days? At the limit, what will be
the total increase in each bicep?
b. Present Value Problem: You win $10 million in the state
lottery' However, you will receive only $0.5 million
now, and $0.5 million a year for the next 19 years (20
payments) . How much money must the state invest
now so that they will have $0.5 million to pay you at
the beginning of the 20th year? Assume that the state
can get 10% per year interest, compounded annually,
on their investment. How much money, total, must the
state invest now to make all 19 future payments? This
amount is called the present va lue of your remaining
$9.5 million.
R3. Let P(x) be the power series c0 + c1 x + c2 x 2 + c3 x 3 + c4 x 4 + . Let f (x) = 7e 3x. By
equating derivatives find the values of c0 , c1 , c2 , and c3 that make P(O), P' (0), P" (0),
and P" ' (O) equal f(O), f'(O), f"(O), and f" ' (O), respectively.
R4. For a-c, show that the fourth partial sum, S3 (0.12) is close to f (0.12).
a. ex = 1 + x + ..!__x
21
2 + ..!__x
3!
3 + ..!__x
41
4 +
00 1
b. cosx = I (-ll"--x 2
"
n =O (2n)!
C. sinhx = x + ..!__x
3!
3 + ..!__x
5
s + .!..x
1 7!
7 + ...
d. Show that the 20th partial sum of I ( - 1)
n= I
+1
11
..!.(x
n
- 1)" gives values close to lnx
if x = 1.7, but not if x = 2.3.
RS. a. What is the difference between a Maclaurin series and a Taylor series?
b. Write the Maclaurin series for ln(x + 1) by performing appropriate operations on
the Taylor series for ln x.
C. Integrate the series in RSb to find a Maclaurin series for f In (x + 1) dx.
650 Chapter
12:TheCalculus DefinedbyPower
of Functions Series
ct. Show that the series in R5c is equivalent to the one you would obtain by finding
the antiderivative of ln (x + 1) and writin g that as a Maclaurin series .
e. Write the first few terms of a power seri es for J;
t cos t 2 dt.
1
f. Write tan - x as the definite integral of an appropriat e function from O to x .
Write the integrand as a Maclaurin series. Then write the first few terms of the
corresponding Maclaurin series for tan - tx.
g. Suppose that f is a function whose deri vatives of all orders are defined for all real
values of x. If f (3) = 5, f'(3) = 7, f " (3) = -6, and f "'( 3) = 0.9, write the first
four terms of th e Taylor series for f (x) expan ded about x = 3.
00
R6. a. Write out th e first few terms of I (-3) - " (x - 5) 11
n= l
b . Find the (open) interval convergence and radius of convergence of the series
in R6a .
c. Show that the Maclaurin series for coshx converges for all value s of x.
ct. Write the fifth five terms of the Maclaurin series for e 1 2 . Then calculate the error
in using the fifth partial sum to approximat e e 12 by subtracting the partial sum
from e1.2 by calculator. How does the error compar e to the valu e of the first term
in the tail of th e series after th e fifth partial sum ?
e. On th e same screen plot the graphs of ln x, the Taylor series about x = 1 for ln x
using 10 terms, and the same series usin g 11 terms. Sketch the gra ph s. Then write
a paragraph stating ho w the graphs relat e to the interval of conve rgence .
R7. a. Find the tenth partial sum of the geometric series with t 1 = 1000 and common
ratio 0.8.
b . By how much do es the tenth partial sum in R7a differ from the limit to which the
series converges?
C. Th e rest of the seri es in R7a following th e tenth partial sum is called the - ?- of
the series.
ct. The valu e of the rest of the series, R7b, is calle d the - ?- of the series.
e. Use an appropriate improper integral to find an upp er bound for th e remainder
of the p-series
1+.!.+..L+...!...+
8 27 64
.. ,
after the tenth term .
f. Explain how your result in R7e can be used to prov e that the p -series
1 + .!.+...!...+...!...+
8 27 64
.. ,
converges .
g. Show that there is a convergent geometric ser ies that is an upper bound for the
tail of the series 2/ 1! + 4 / 2! + 8/3! + 16 / 4! + 32 / 5! + after a suit able number
of terms .
h. Show that th e alternating harmonic series
1 - ! + ! _ .!.+.!. _ !+ .. ,
2 3 4 5 G
converges.
i. Find an upper bound for the remainder of the ser ies in R7h after 10,000 te~ms .
Section
12-9:Chapter
Review
andTest 651
j . Find the complete interval of convergence, including the endpoints .
00
1on(x - 3)" .. oo (- l) n(x + l) n
i. I 2 11. I - ----
n= l n n =l n 2 11
k. Tell whether or not the following series of constants conv erge. Justify your answer.
. ~ 10
00
1 L.. - ii. I ( n -3 + 5- 1)
n =o n! n= I
RB. a. Use the Lagrange form of the remainder to estimate the error in using the fourth
partial sum of the Maclaurin series to estimate cosh 2.
b. It is desired to calculate e 3 using enough terms of the Maclaurin series to get a 20
decimal place accuracy. Use the Lagrange form of the remainder to calculate the
number of terms that should be used.
c. The Maclaurin series for coshx converges for all values of x . Use the Lagrange
form of the remainder to show that the value the series converges to when x = 4
really is equal to cosh 4.
d. Calculate, approximately, the number c in the interval (0, 0.6) for which the
Lagrange form is equal to the remainder of the Maclaurin series for sinh 0.6 after
the fourth partial sum (n = 3).
e. Use the fact that the Taylor series for lnx is alternating if xis between 1 and 2 to
find the number of terms of the series needed to compute In 1.3 to 20 decimal
places.
f. Use an improper integral to find an upper bound on th e remainder of th e p-series
1 00
I -p4
n= ]
after the 50th partial sum (n = 50).
Concepts Problems
Cl. Series with Imaginary Numbers Prob lem: If you substitute ix for x in the Maclaurin
series for cosine and sine, you get some startling results! In this prob lem you will
see how these results lead to another similarity between trigonometric functions and
hyperbolic functions.
a. Substitute ix (where i is J=T) for x in the Maclaurin series for cosine. When you
simplify, you should find that cos ix is real-valued and equals cosh x.
b. Substitute ix for x in the Maclaurin series for sine. You should find that each term
has i as a factor. Factor out the i to show that sin ix = i sinh x .
c. Substitute ix for x in the Maclaurin series for ex. Use the result and the answers
to Cla and b to show that the following formula is true.
ei x = cosx + i sinx
d. Show that e irr = - 1. This one short formula combin es four of the most mysterious
numbers of mathematics!
C2. Pract ical Calcu lation of Pi Problem: In Problem 35 of Problem Set 12-5 you computed
TT by using the Maclaurin series for tan - 1 x to comput e TT/ 4, which equals tan - 1 1. You
652 Chapter
12:TheCalculus
ofFunctions
Defined
byPowerSeries
made use of the composite argument property from trigonometry,
tanA + tanB
ran (A + B) =------ ,
1 - tan A tanB
to show that
tan- 1 .!. + tan - 1 .!. = !!.
2 3 ~ '
thus obtaining more accuracy with fewer terms. Show that the double series
4 tan - 1 .!.
5
- tan - 1 _!._
239
also converges to rr / 4. How many terms of this series would be needed to get
rr correct to the numb er of decimal places reported in William Shaaf's booklet
Computation of Pi (Yale University Press, 1967), namely,
rr = 3.14159 26535 89793 23846 26433 83279 50288 41971 69399 37510 ....
C3. Series Solution of a Differ ential Equation: In future courses on differential equations
you will learn to solve a differential equation directly in terms of a power series. In
this problem you will get a preview of the technique used. Consider the second-order
differential equation
y" + 9xy = 0
with initial conditions y ' = 7 and y = 5 when x = 0.
a. Assume that there is a power series equal to y. That is,
17= 0
Assuming that the series can be differentiated termwise, write equations for y'
and y".
b. Use the two initial conditions in the appropriate places to evaluate c0 and Ct.
c. Substituting the series for y and y " into the original differential equation and
combining terms with equal powers of x gives
2c2 + 6c3x + 12c~x2 + 20c x 3 + 30c x 6 + 5 6
+9 x (co + + C2X + C3 X + C4 X + C5X 5 + C5X 6 + ... ) = 0
C1 X
2 3 4
2c2 + (6c3 + 9ca)x + (12c4 + 9ci)x 2 + (20c 5 + 9c2)x 3 + (30c 6 + 9c3)x4 + = 0
Since the right-hand side of the equation is zero, each coefficient on the left-hand
side must equal zero. Use this fact to calculate the values of c2 through c6 .
d. Use the terms of the series through the sixth power to compute y when x = 0.3 .
e. Just for fun, see if you can tell whether or not the series you found in C3d
converges when x = 0.3.
Chapter Test
00
Tl. You know that the p-series I 2 / n 3 converges. Suppose you evaluate the series
17= 1
by adding up the first 1000 terms . The tail of the series, starting at term number
1001, represents the error involved if you stop at 1000 terms. By comparing with a
convergent improper integral, find an upper bound for the tail of the series.
Section12-9:Chapter
Review
andTest 653
T2. Based on your answer to Problem Tl, to how many decimal places could you guarantee
the answer to be accurate if you stop after 1000 terms?
T3. By equating derivatives, show that the Taylor series for lnx expanded about x = 1 is
lnx = (x - 1) - (x - 1) 2 + (x - 1) 3 - (x - 1)4 + . .. .
T4. Let f (x) = f~';r dt. Write a power series for f
1 3 (x).
TS. Find the interval of convergence of the series in Problem T4.
T6. Find an approximation for f (0.6) using 20 terms of the series from Problem T4.
T7. Find an approximation for f (0.6) in Problem T4 by numerical integration.
TS. Find f (0.6) in Problem T4 exactly, using the fundamental theorem. To how many
decimal places are the answers to Problems T6 and T7 correct?
T9. Demonstrate that the error in the value off (0.6) by series in Problem T6 is less than
the first term in the tail of the series .
Tl 0. Write a power series for coshx . Express the answer in sigma notation .
Tl 1. Use the Lagrange form of the remainder to find the number of terms of the series in
Problem Tl O needed to estimate cosh 3 correct to 10 decimal places.
Tl 2. Write a power series for sin (x 2 ). Write the answer in sigma nota tion.
T13. Show that the geometric series 1000 + 999 + converges, but the geometric series
0.0001 + 0.0002 + does not converge .
Tl 4. Find the open interval of convergence and the radius of convergence:
f (2x - s)n
n= l 3n
TlS. Does the series in Tl4 converge or diverge at the endpoints of the interva l of
convergence? Justify your answer.
12-10 Cumulative Reviews
In this section are several cumulative reviews that may be considere d to b e
"rehearsals" for your final exam. Each review touches on most of the concepts and
techniques of calculus, particularly those of the second half of the book.
Problem Set 1 2-1 0
Cumulative Review No. 1- The Dam Problem
Suppose you are hired as a mathematician by Albee Dam Construction Company, which
has been award ed a contract to build a dam across Scorpion Gulch. The following
questions pertain to your part in this project . Before Mr. Albee will allow you to work on
his dams, he must be sure that you know some of the fundamental definitions, theorems ,
and techniques of mathematics . These are contained in Problems 1 thro ugh 5.
654 Chapter s of FunctionsDefined by Power Series
12: The Calculu
1. There are four major concepts of calculus. Name
those concepts, and state their definitions .
2. Define each of the following:
a. Continuity of a function at a point
b. Continuity of a function on an interval
c. Convergence of a sequence
d. Convergence of a series
e. Natural logarithm
f. The exponential ax where a > 0
3. Mr. Albee is satisfied with your knowledge of
definitions, and proceeds to quiz you on your
knowledge of various properties. State the follow-
ing:
a. The mean value theorem b. The intermediate value theorem
c. The squeeze theorem d. The uniqueness theorem for derivatives
e. The limit of a product property f. The integration by parts formula
g. The fundamental theorem of calculus h . The Lagrange form of the remainder
i. The chain rule for parametric functions j. The polar differential of arc length
4. To make sure you know enough algebraic techniques, Mr. Albee asks you to find the
following limits, derivatives, and integrals.
a. f'(x), if f(x) = J; J l + sech t dt b. f'(x), if f (x) = ax
c. f'(x), if f(x) = x a d. f'(x), if f(x) = x x
e. J e 6xcos 3x dx f. J cosh 5 xsinhxdx
g. J sec3 xdx h. J(sin5x) - 1 cos5xdx
. cos 7x - l
i. 11m---- j. lim(l - x) 3fx
x- o 13x 2 x- 0
5. Mr. Albee wants to be sure you know graphical and numerical y
methods. 10 \ \ \
\ \ \ \ \ \
\ \ \
",
\ " "'- '-
..............
a. Figure 12-lOa shows the slope field for
\ \ \ \
\'-"'-'''
""' ,__
__-
_
dy
dx = 0.2x - 0.3y + 0.3
\
\'-"'-'' _____,,,.
On a photocopy of the figure, sketch the particular solution "'''
,,-----////
____ ,,,,/
containing (1, 8).
,---//////
b . Use Euler's method with 2\x = 0.5 to estimate the value of - - _. ,,,. / / / / I I
y for the solution in Sa if x = 9. How does your answer -/////////
X
compare with your graphical solution?
10
Figure 12-1Oo
Section12-10
: Cumulat
iveReviews 655
You pass your preliminary tests and start to )I
work on the dam project. At the dam site, Line z= (()')
Scorpion Gulch has a parabolic cross-section
whose shape is the graph of y = O.lx 2 , where
x and y are in yards . The back face of the dam
(where the water will be) is vertical, and lies
in the xy-plane (Figure 12-lOb). The front face Horizonta l
slopes in such a way that the thickness is z = 30 cross -section
yards at the bottom of the gulch (where y = 0),
and z = 10 yards at the top of the dam. The
dam is to be 40 yards high .
6. Your first project is to analyze the forces that
the water will exert on the vertical back face of
the dam when the lake is full (i.e., 40 yards deep) .
Assume that the water density is k pounds per Figure 12-1Ob
cubic yard .
a. Write an equation for the pressure in terms of y.
b . Find the area of the dam's back face.
c. Find the force exerted by the water on the back face of the dam.
d. Find the first moment of this force with respect to the x-axis.
e. Find the "center of pressure." This is the point on the back face at which the entire
force could be concentrated to produce the same first moment with respect to the
x-axis .
7. Your next project is to determine some of the physical characteristics of the dam itself.
a. Find a linear equation expressing the thickness of the dam, z, in terms of the
altitude, y.
b . At what y-value will the dam's horizontal cross-sectional area be a maxirnum7 a
minimum?
c. A cement mixer truck holds 5 cubic yards of concrete. How many truckloads of
concrete should you order when it is time to pour the dam?
d. Find the length of the joint between the dam's back face and the sides of the gulch.
8. The dam is finished. A speedboat on the lake behind the dam moves with vector
equation
r= (100cos0.03t)7 + (50sin0.03t)]
where distances are in feet. How fast is the boat going when t = 50 seconds?
9. The waves from the boat displace the water's surface according to the equation
z = (sint) / t. The average displacement over the time interval [O, t] involves the sine
integral function,
.
Sit= lrsinud
-- u
0 U
Write Si t as a power series. Use the ratio technique to determine the interval of
convergence. Estimate the error in calculating Si 0.6 using just the first three non-zero
terms of the series. Calculate Si 0.6 by numerical integration on your grapher .
656 Chapter
12:TheCalculus
of Functions byPowerSeries
Defined
10. The drain in the dam has a cross-section the shape of the polar curve r = 5 + 4 cos 0
where r is in feet. Find the area of the drain's cross-section .
11. At time t = 0 hours, the drain is opened . Initially water flows out at 5 million gallons
per hour. But the rate is directly proportional to the amount of water remaining.
There were 300 million gallons of water behind the dam at t = 0. Predict the amount
remaining at t = 10.
Cumulative Review No. 2- The Ship Problem
After graduation you apply for work at Sinkin Ship Construction Company. Mr. Sinkin
gives you the following preliminary test to see how much calculus you know .
l. Define derivative . 2. Define definite integral.
3. State the mean value theorem. 4. Find f' (x) if f(x) = J; g(t) dt.
5. Integrate : f tanh 5 x sech 2 x dx 6. Integrate: f x sinh 2x dx
7. Integrate: I (x +3x )(x
+ 14
_ 8. Wnte
. t h e Mac laurm
. senes
. f or J --sinh x d x.
3 2 ) dx X
9. Find the (open) interval of convergence for this series:
f n(x - S)n
n= l 311
10. Evaluate the improper integral f; x- 0 998 dx .
11. Find the average value of y = x 2 on the interval [3, 9].
12. Given f(x) = x 2 . If 8 = 0.01, is this small enough to keep f(x) within 0.08 unit of f(4)
when xis within 8 units of 4? Justify your answer .
Mr. Sinkin is satisfied with your work on these questions and assigns you to the design
team for a new ship. The hull of this ship is shown in Figure 12-1 Oc.
13. The volume of the ship equals the cross-sectional area times the length . At the
forward end of the ship, the cross-sectional area varies with x, the distance from the
bow . Write an integral for the volume of the part of the ship from x = 2 to x = 10.
Then evaluate the integral approximately by Simpson's rule, given the cross-sectional
areas shown in the table. Dimensions are in feet and square feet.
- Forward Aft - X Area
Bulkhead
tl(F'a o='"
2 153
4 217
6 285
8 319
10 343
Keel Propeller
Figure l 2-1Oc
14. The propeller will have four blades the shape of the four-leaved rose r = 4 sin 20,
where r is in feet . Find the area of one blade.
Section12l 0: Cumulative
Reviews 65 7
15. At the stern of the ship, the deck has a shape similar to the region bounded by the
ellipse
(i /+( t /= 1
between x = 1 and x = 5. Find the area of this region.
Your next project is to analyze a vertical bulkhead (a wall) that goes across the ship. The
bulkhead has the shape of the region that lies above the graph of y = 0.0016x 4 and below
the line y = 16.
16. Find the area of the bulkhead .
17. The welders who will install the bulkhead need to know the length of the graph of
y = 0.0016x 4 that forms the edge of the bulkhead. Find this length.
18. The bulkhead must be strong enough to withstand the force of the water if the
compartment on one side of the bulkhead becomes flooded. You recall that force
equals pressure times area, and that the pressure at any point below the water's
surface is proportional to that point's distance from the surface. The proportionality
constant is 62.4 pounds per cubic foot, the density of water. Find to the nearest 100
pounds the force exerted by the water.
A vertical keel is to extend below the bottom of the ship. When turned upside down, the
keel is similar in shape to the region under the graph of y = (ln x) / x, for x ?. 1.
19. Find the limit of y as x approaches infinity .
20. Find the x-coordinate of the maximum of the function. Justify your answer.
21. Find the x-coordinate(s) of all points of inflection of the graph.
22. Sketch the graph, consistent with your answers above.
The radar equipment needs values of natural logarithms to 20 decimal places.
23. Show that the Taylor series for ln x expanded about x = 1 converges for O < x :s:2.
24. How many terms of the series would be needed to calculate ln 1.4 to 20 decimal
places?
Your last project is analysis of the motion of the ship .
V
25. In linear motion the velocity of the ship is given by a
differential equation whose slope field is shown in Figure
12-lOd . Describe the way the velocity changes if the ship
40
starts from v = 0 ft / sec at t = 0 min . How would the velocity
differ if somehow the ship were given an initial velocity of 30
50 ft / sec?
20 / / / / / / / / / /
26. In a sharp turn, the position vector of the ship is given by
//////////
r = (lnt)i + (sin2t)]. Find the acceleration vector. 10 //////////
I I I I I I I I I I t
2 3 4
Figure 12-1Od
6 58 Chapter
12:TheCalcu
lusof Functions byPowerSeries
Defined
Cumulative Review No. 3-Routine Problems
Calculus involves four fundamental concepts. In the first four problems you are to
demonstrate that you understand these concepts .
1. Demonstrate that you understand the definition of limit by sketching the graph of a
function that has a limit Las x approaches c, and showing an epsilon neighborhood
of L and a corresponding delta neighborhood of c that is clearly smaller than is
necessary.
2. Write the formal definition of derivative. Then tell a graphical meaning of the
derivative, and a physical meaning of the derivative.
3. Write the definition of Jf(x) dx. 4. Write the definition of I: f(t) dt.
The definition of limit is not used to find limits . So other techniques are developed .
5. There is a technique for finding "interesting" limits of the form 0 / 0 or 00/ 00. Name
this technique, and use it to evaluate
lim X COSX.
x-0 1 - e 5x
The definition of derivative is awkward to use. So you develop shortcuts.
6. Find y ' if y = tan (sin 5x) . Name the property that allows you to differentiate such
"composite functions."
7. Products and powers can be differentiated logarithmically . Find y ' if
y = (5x - 3)(2x + 7) 4 (x - 9).
8. Formulas for derivatives of inverse cicrular functions are found by implicit differen-
tiation. Do this for the inverse tangent function.
In Prob lems 9 through 12 you will demonstrate your knowledge of certain basic techniques
and when to use them. Integrate .
9. J sin 7 xcosxdx 10. f .Jx 2 + 9dx
ll. J 3x - ll dx 12. J sin - 1 x dx
x 2 + 2x - 3
Definite integrals are hard to evaluate using the definition . Fortunately, there is a theo rem
relating definite integrals to indefinite ones .
13. Name and state the theorem that relates definite and indefinite integrals .
14. The mean value theorem plays a key role in the proof of the theorem in Problem 13.
Demonstrate that you understand what the mean value theorem says by sketching
an appropriate graph.
The techniques of calculus can be applied to problems in the real and mathematical
worlds.
15. If f(x ) = J;h(t ) dt, find f'(x).
16. Given: f(x ) = xe - x for x 2. 0. Find the x-coordinates of all points of inflection .
17. Find the length of the graph of y = sin x from x = 0 to x = 2.
Section 12-10: CumulativeRev
iews 659
18. The integral Ji6 x - 3 14 dx is improper.
a. Show that the integral converges.
b. Use the result to find the average value of y = x - 314 from x = 0 to x = 16.
19. Find the area of the region inside the circle with polar equation r = 10 cos e from
e e
= o.sto = 1.
20 . A particle travels in the xy-plane in such a way that its position vector is
r = r 2 7 + 3r- 1].
At time t = l, what is its velocity vector7 How fast is it going? Is the particle's distance
from the origin increasing or decreasing7 At what rate7
21. The region in Quadrant I bounded by the coordinate axes and the graph of y = cos x
is rotated about the y-axis to form a solid . Find its volume.
22. A rectangle in the first quadrant has one corner at the origin, and the diagonally
opposite corner on the line y = - 1. Sx + 6.
a. Show that the rectangle has maximum area when x = 2.
b. If the rectangle and line are rotated about the y-axis to form a cylinder inscribed in
a cone, show that a different value of x produces the maximum-volume cylinder.
23 . The crew's compartment in a space ship has an irregular shape due to all the equip-
ment in it. Cross sections at various distances from the bottom of the compartment
have areas as follows (feet and square feet):
distanc e area
3 51
5 37
7 41
9 63
11 59
Use Simpson's rule to estimate the volume of the crew's compartment between 3 and
11 ft .
24. The error function, used in "curving" grades, is defined by
erf x = 2rr 112 ( e- 12 dt .
a. Write the first few terms of the Maclaurin series for the integral in erf x.
b. Use the ratio technique to prove that the series for erf x converges for all values
Of X.
660 Chapter
12:TheCalculus
af Functio byPowerSeries
ns Defined
FINAL EXAMINATION
This section contains an examination that may be considered to be a "dress
rehearsal" for the exam your instructor will give you. We suggest that you make the
rehearsal as realistic as possible by putting yourse lf under simulated test conditions
and giving yourself a two-hour time limit. No answers are provided in the back of
the book.
A GuidedTourThrough
Calculus
Calculus involves two basic concepts:
a. Instantaneous rates of change, and
b . Products in which one factor's value depends on the other factor.
Both of these are based on the underlying concept of limit. They are linked together
by the fundamental theorem, which allows you to calculate limits of Riemann
sums by using antiderivatives. On this test it will be your objective to answer the
questions in a way that demonstrates that you understand these concepts and their
relationships.
1. The first problem you encountered was finding the instantaneous rate of change
of a function at a given point. Find, approximately, the derivative of f(x) = sinx
at x = 1 by finding how much sin 1 differs from sin 1.1, sin 1.01, and sin 1.001,
and doing the appropriate division .
2. Later, you found techniques for calculating derivatives exactly. Find f'(l) if
f(x) = sinx . Show that the three approximate values of f'(l) you calculated
in Problem 1 are converging toward the value of f ' (l) you calculated in this
problem .
3. The intuitive idea of "instantaneous rate of change" is made precise by a formal
definition of derivative. Write both forms of this definition, one for f'(x), and
one for f'(c), where c is a particular value of x.
4. The definition of derivative in Problem 3 involves the concept of limit . The limit
of f(x) as x approaches c is the number f(x) can be kept close to, just by keeping
x close enough to c. Suppose that f(x) =e x.What does the limit of f(x) equal as
x approaches 2? How close would you have to keep x to 2 in order for f(x) to be
within 0.1 unit of this limit?
5. The intuitive idea of "closeness" is made precise in the formal definition of limit.
Write this definition.
6. The quantities epsilon and delta from the definition of limit appear in Problem
4, above. Which one is epsilon, and which is delta?
7. Your intuitive introduction to variable-factor products came from the distance
= rate x time equation . Draw an appropriate graph which shows that if the rate
is constant, the distance can be represented as the area of a rectangle.
8. If the rate varies, the distance can still be represented as the area of the region
under a graph . Suppose that the rates shown in the table have been measured
at the given times. Plot the graph of rate vs. time . Find the distance traveled
between 1.0 and 2.8 by counting squares.
FINAL EXAMIN A TION 66 1
~
Time Rate
1.0 7
1.3 9
1.6 13
1.9 12
2.2 10
2.5 8
2.8 5
9. More recently you have learned ways to calculate definite integrals such as in
Problem 8, without having to draw the graph and count. Find the distance using
Simpson's rule . Show that it is approximately the same as the distance you
found by counting squares.
10. In Problems 8 and 9 you knew no equation for rate in terms of time. If you
do know such an equation, there are other ways to calculate the distance.
Suppose that v(t) = te - 1 Find the distance traveled between t = 0 and t = 2 by
calculating a Riemann sum with n = 5 increments, taking sample points at the
midpoint of each subinterval.
11. You now know the fundamental theorem of calculus, which allows you to
calculate limits of Riemann sums exactly, using antiderivatives. Find, exactly,
the distance traveled in Problem 10. By what percentage does the Riemann sum
in Problem 10 differ from the exact value?
12. State the fundamental theorem of calculus .
13. The proof of the fundamental theorem relies on the mean value theorem. Show
that the mean value theorem does apply to f(x) = x 213 on the inverval [O, l], in
spite of the fact that f is not differentiable at x = 0. Calculate the point x = c
in the interval (0, 1) at which the conc lusion of the mean value theorem is true.
Then plot the graph accurately, and show that the line through (c, f(c)) with
slope f ' (c) really does satisfy the requirements of the mean value theorem.
14. In order to use the fundamental theorem, you must be reasonably good at
finding indefinite integrals. Write an integral that can be evaluated by the given
technique, and do the integration.
a. Partial fractions.
b. Trigonometric substitution.
15. Sometimes integration by parts results in the same integral appearing on both
sides of the equals sign. Show how this situation can be handled by doing the
integration for f sec 3 x dx.
16. In addition to their role in evalua ting definite integrals, indefinite integrals
have applicat ions in their own right. For instance, if you know how a function
changes, you can find an equation for the function itself. Suppose that the
instantaneous rate of change of y with respect to x is directly proportional to
y . Write the appropriate differential equation and solve it to express yin terms
Of X.
662 Chapter
12: TheCalculus
of Functions byPowerSeries
Defined
17. Definite integrals arise from variable-factor products . The volume of a solid is
equal to its cross-sectional area times its height . Sketch the solid formed by
rotating about the y -axis the r egion in Quadrant I under the graph of y = 4 - x 2
Then slice the height in such a way that the cross -sectional area is (essentially)
constant at any point in the slice.
18. Sketch the solid in Problem 17 again, and draw a slice using cylindrical shells .
Show that this method slices the cross-sectional area so that the height is
(essentially) constant at any point in the slice.
19 . You recall that the moment of a quantity is the magnitude of that quantity times
a distance from a reference point, line, or plane. Find the first moment of area
of the region in Problem 17 with respect to the y-axis. Use the result to find the
x-coordinate of the centroid of the region .
20. Once you understand the concept of variable-factor products, you can analyze
any such problem, even when the quantities are unfamiliar . For instance,
the number of calories needed to warm a substance from temperature T1 to
temperature T2 equals the "heat capacity" of the substance (calories per degree)
times the change in temperature, T2 - T1 . Unfortunately, most real substances
have heat capacities that vary with temperature . Assume that calculus foeride
(a rare, tough substance!) has a heat capacity given by
C = 10 + 0 .3T 112 ,
where C is in calories per degree and T is in degrees celsius. How many calories
would be needed to warm the calculus foeride from 100 to 900 F?
21. Definite integrals can be applied to the mathematical world. For instance,
a function might be defined as a definite integral. As you may recall, the
sine-integral function, Six, is defined as
.
SlX =
l x --sinud U
0 U
a. Write an equation for Si'x.
b. Expand the integrand as a Maclaurin series, then integrate to get a series for
Six .
c. Evaluate Si 0.7 approximately, using the first two nonzero terms of the series .
d. Find an upper bound for the tail of the series that is left after the first
two terms. Based on this result, tell to how many decimal places you can
guarantee that your answer in 21c is correct .
e. Prove that the series for Si x converges for all values of x.
22. Parametric equations can be used to apply calculus to vector functions. Suppose
that an object is moving in such a way that its position vector, r,
is given by
r= (t 3
)7 + (t 2
)].
Plot accurately the path of the object from t = 0 to t = l. Show the location of
the object when t = 0.5. Calculate the velocity and acceleration vectors when
t = 0.5 . Plot these two vectors with their tails at the object . Is the object speeding
up or slowing down? Explain .
FINAL EXAMIN AT ION 66 3
14-
23. Calculus can be applied to figures in polar coordinates. Figure FE-1 shows the
polar graph of
r = cos e
from 0 = 0 to 0 = IT / 2. Slicing the region as shown gives a wedge of angle d0.
Any point on the graph that is with.in the angle d0 has (essentially) the same
radius as at the sample point. The area of the wedge is approximately equa l to
the area of a sector of a circle of radius r. The sector, of course, is d 0 / ( 2IT ) of
the area of a circle of radius r . Use th.is information to find dA, the area of the
wedge, in terms of e. Then find the area from e = 0 to 0 = IT / 6 by adding the
dA's and taking the limit as d0 approaches O (i.e., definite integrating).
Figure FEl
664 Chapter
12: TheCalculus
of Functions byPower
Defined Series
Appendix A: Summary of Properties
of Trigonometric Functions
1. Reciprocal tan (A _ B) = tan A - tan B
1 + tan A tan B
1
cot x = -t or tan x cot x = 1 tan (A + B) = tan A + tan B
anx
1 - tan A tan B
sec x = - 1- or cos x sec x = 1 7. Double-Argument
COSX
sin 2x = 2 sin x cos x
csc X = -.- 1 or Sill
. X csc X = 1
Sill X cos 2x = cos 2x - sin 2x = 1 - 2 sin 2x
2. Quotient = 2 cos 2 x - 1
tan 2x = 2 tan x2
tan x = sin x = sec x 1 - tan x
COSX CSCX
cos 2 x = ( 1 + cos 2x)
cot X = C?S X = csc X
Sill x sec x
sin 2 x = ( 1 - cos 2x)
3. Pythagorean
cos 2x + sin 2x = 1 8. Ha lf-Argument
-----
1 + tan 2x = sec 2 x sinx = )( 1 - cos x)
cot 2x + 1 = csc 2 x
cosx= ( 1+cosx )
4. Odd-Even 1 - COSX
tanx = 1 + COSX
sin (- x) = - sin x (odd)
COS(- X) = COSX (even) Sin X 1 - COSX
1 + COSX Sin X
tan (- x) = - tan x (odd)
cot (-x) = -c ot x (odd) 9. Sum and Product
sec (- x) = sec x (even) 2 cos A cos B = cos (A + B) + cos (A - B)
CSC(-X) = - CSCX (odd) 2 sin A sin B = -cos (A + B) + cos (A - B)
2 sin A cos B = sin (A + B) + sin (A - B)
5. Cofunction 2 cos A sin B = sin (A + B) - sin (A - B)
COS(90 - 0) = Sin 0; COS( ; - X) = sin X y
COSX + COS = 2 COS(X + y)COS(X - y)
cot (90 - 0) = tan 0; cot ( I - x)= tan x cos x - cos y = - 2 sin (x + y)sin (x - y)
csc (90 - 0) = sec 0; csc ( I - x)= sec x sin x + sin y = 2 sin (x + y)cos (x - y)
6. Composite-Argument sin x - sin y = 2 cos (x + y)sin (x - y)
cos (A - B) = cos A cos B + sin A sin B
10. Linear Combination of Sine and Cosine
cos (A + B) = cos A cos B - sin A sin B
A cos x + B sin x = C cos (x - D); where
sin (A - B) = sin A cos B - cos A sin B
--- A B
C=YA 2 + B 2 ,cosD=c,andsinD = C
sin (A + B) = sin A cos B + cos A sin B
665
Appendix B: Answers to Selected Problems
CHAPTER 1 15. a. Quadratic (or polynomial) .
b. j(3) = 30
Problem Set 1-1 c. Increasing at about 11.0. (2.99 to 3.01)
The Concept of Instantaneous Rote
17. a. Exponential. b. f (2) = 9
l. Pendulum Prob lem c. Increasing at about 9.89. (l.99 to 2.01)
a. 95 cm, Q.E.D.
19. a. Rational algebraic. b. j(4) = - 1
b. t = 5.1, rate "" 26.34 cm / sec c. Decreasing at about 1.00. (3.99 to 4.01)
t = 5.01, rate "" 27.12 cm / sec
t = 5.001, rate "" 27 .20 cm / sec 21. a. Linear (or polynomial) .
So the instantaneous rate of change at t = 5 is b. j (5) = - 8
about 27.20 or 27 .21 cm / sec . c. Decreasing at 3. (Exact)
c. Instantaneous rate invo lves division by zero. 23. a. Circular (or trigonometric).
d. Fort = 1.5 to 1.501, rate "" - 31.42 cm/sec. The b . f (2) = 0.90929 .. .
pendulum is approach ing the wall. Since rate of
c. Decreasing at about 0.416. (1.99 to 2.01).
change is negative the distance is decreasing .
e. The instantaneous rate of change is the li m it of 25. Accurate Graph of a Cubic Function Prob lem
the average rates as the time interval approaches
l(x)
zero . It is called the derivative . 5
f. Before t = 0 the pendulum was not yet moving .
For large values of t the pendulum's motion will X
5
die out due to friction.
Problem Set 1-2
Rotesof Change by Equation,Graph, or Tobie
a. j(x) is increasing fast when the graph slopes
1. a. Increasing slowly. steeply upward.
b . Increasing fast. b. j(x ) is decreasing for about - 3 < x < 6.
3. a. Decreasin g fast. c. False. j(x) is positiv e, but decreasing between
b . Decreasing slowly. x = -3 and x = 6.
d. j(x) is decreasing fastest at about x = 1 or 2.
5. a. Increasing fast.
b. Increasing slowly. 27. See definition in text.
c. Decreasing slowly. Problem Set 1-3
d. Increasing fast. One Kindof Integralof a Function
7. a. Increasing slowly .
b. Increasing slowly. l. a. Approximately 3. a. Approximately
30.8 2.0
c. Increasing slowly.
b. Approximately b. Approximate ly
9. a. Increasing fast. 41.8 1.0
b. Neither increasing nor decreasing . h(x)
. 7 /(x)
c. Increasing fast.
d. Increasing slowly .
11. a. x = 1.5 : rate "" 6.0 cm / min. Increas ing.
b. x = 3.0: rate "" 6.8 cm / min . Increasing .
c. x = 4.0: rate "" - 0.3 cm/min . Decreasing . X
X
13. Rolli ng Tire Problem -1 0 5 6
a. i. - 1.0 in. / sec . ii. 0.0 in. / sec. iii. 1.15 in ./ sec.
b . 1.7 sec, since y = 0 at that time. 5. Distance "" 680 feet. 7. Derivative "" 3.42 .. .
667
9. Sports Car Problem wind up being inscribed inside the region the rule
a. Graph. und eres timates the integral. Thus for Problem 9
the rule underestimat es, and for Problem 10 it
v(I)
overestimates .
100
13. Football Problem
Integral "" 2 2 7.8 in 3
The integral will have units (in 2 ) (in. ) = in 3 . So the
int egral represents the volum e of the football.
15. Derivative from Graph Problem
a. Derivative "" -5.6 .
b. Derivative "' 5.6
8.7 10
Note the different scales on the two axes.
b. Range is O ~ y < 100. 17. Meaning of Limits Problem
c. x "" 8.7 sec. a. j(5) would involve division by ze ro.
d. Distance "" 300 ft.
b. j(x ) = 2(x + 5), x * 5.
e. Rate "" 6.2 (ft / sec) / sec f(4.9) = 19.8, j(5 .l ) = 20.2
f. Acceleration . c. Both are close to 20.
11. About 7.1 cm d. f(4 .99) = 19.98, j(5.01 ) = 20.02
13. See text meaning of definite integral. Both are close to 20.
e. "If xis within 0.01 unit of 5, then j(x) is within
Problem Set 1-4 0.0 2 unit of L."
DefiniteIntegralsby Trapezoids,from Equationsand Data
Problem Set 1-5
1. Spaceship Problem limitof a Function
a. Graph.
1. Has a limit, 3.
3. Has a limit, 3.
5. Has no limit.
7. Has a limit, 7.
9. Has no limit.
30
11. Definition of Lim it, Problem 1:
b. Distance = rate times time. Since the vertical a. See text definition of limit.
axis represents distance and th e horizontal axis b. j (4) = 3(4) - 7 = 5, Q.E.D.
represents time, their product (i.e., the area)
c. Keep x within 0.2 unit of 4.
represents distance.
d. Graph.
c. Graph. Distance "" 281,000 ft.
d . Definite integral. l(x)
e. Yes, it will be going fast enough . 5.6
s- -~--
v(30) = 27,919 .04 ... > 27,000. 4.4
3. Aircraft Carrier Landing Problem
Distance "" 396 feet. No danger of running off!
5. Integral "" 70.8 X
7. Programs will vary, depending on calculator.
9. a. 30.8125 c. 41.766095 e. e.g., Let x = 4.1. Then f(4.l) = 5.3, which is
b. 41.7095 within 0.6 unit of 5.
11. Trapezoida l Rule Error Problem f. 0 = E/3 = (0.00012) / 3 = 0.00004
If the trapezoids wind up being circumscribed
outside the region bounded by the graph, the trape- 13. Definition of Limit, Problem 3:
zoids enclose too much area and the trapezoidal = (x 2 -6x+13)(x - 2)
rule overestimates the integral. If the trape zoids a. f( X ) X - 2
668 Appendix
B:Answers
toSelected
Problems
!( 2) = (5)(0) = Q Q.E D b . From 4 to 4.1, average rate :::::-29.3 ft / sec
(0) O' ..
From 3.9 to 4, average rate :::::- 40.1 ft / sec
The limit is 5. Substitute 2 for x into the simpli- Instantaneous rate :::::-3 4.7 ft / sec
fied expression, j(x) = x 2 - 6x + 13, x * 2 Going down . Distance from water is decreasing.
b. j(l.995) = 5.010025, j(l.996) = 5.008016; c. Instantaneous rate :::::70.8
!(2.005) = 4.990025, !(2.006) = 4.988025.
d. Going up at about 70.8 ft / sec.
If xis within 0.004 unit of 2, th en .... You must
use the value of x that is closer to 2. e. Derivative .
C. f (x) = 4.99: X = 2.00501256 .. . R2. a. Instantaneous rate of change of a function .
f(x) = 5.01: X = 1.99501243 . . .
1.9950124 3 .. . < X < 2.00501256 .. . b. x = -4: Decreasing fast
x = 1: Increasing slowly
d. Graph (exaggerated to show concept) x = 3: Increasing fast
x = 5: Neither increasing nor decreasing
c. j(2 ) = 25. Derivative :::::40.2.
d . t = 2: 3.25 m / sec
t = 18: 8.75 m / sec
delta? __; t = 24: 11.5 m / sec
_.__
1
delta? 1-.....----
I
Speed is constant, 7 m / sec, from t = 6 to 16 sec.
I
At t = 24, Mary is giving her final spurt toward
1.995012. .. 2 2.00501...
the finish line.
e. Delta = 0.00498756 ... R3. a. Graph .
f. L = 5, c = 2, epsilon = 0.01,
delta = 0.00498756 ... v( t)
10
15. One-Sided Limit Problem
a. The quantity Ix - 2 1/ (x - 2) changes from - 1 to
1 at x = 2, thus causing the jump.
b. From the left, the limit is 3.
c. From th e right, the limit is 5.
d. Pick any value of epsilon less than 1. No matter
what numb er you claim to be the limit, there are
2 10
values of x on one side of 2 or th e oth er which
mak e f (x) come out more than epsilon units
away from that number. b. Distance :::::58.1 ft.
17. Derivative:::::- 0.1111 using r( -3 .01) and r(-2 .99) c. Definit e integral
19. Don 't Believe Everything You See Probl em! R4. a. Graph is correct.
"Look, Ima. The two curves are so close together that
the grapher can't tell the difference. But you should b. Integral :::::15.0
reali ze that by using TRACE on the two graphs, you C. X f (X)
will see a slight difference in the y-values for the 1.0 5.3
two graphs at the same x-value. Or you could use a 1.5 5.575
wider window, thus showing the curv e of the sine 2.0 5.6
graph away from the straight of th e linear graph." 2.5 5.375
3.0 4.9
Problem Set 1-6 3.5 4.175
CalculusJournal 4.0 3.2
1. Answers will vary . Integral :::::14.9375
RS. a. See definition of limit in text.
Problem Set 1-7
b. Limit = - 10
Chapter Review and Test
c. Delta = 0.00498756 .. .
Review Problems d. e.g., Let x = 3.004.
Rl. Bungee Jumping Problem !(3.004) = - 10.007984, which is within 0.01
a. t =4 :d :::::15.4ft. unitof - 10.
Problem
Set1-7:Chapter
Review
andTest 669
e. Graph. (example) h(x )
- 1
Problem Set 2-2
Graphical and Algebraic Approach to the Definition of Limit
R6. The journal gives you a chance to summarize things
you have learned and condense the class notes you 1. limx-3 / (x) = 5. For E = 0.5, 8 ;::;0 .2 or 0.3.
have taken . Since it also gives you a chance to write 3. limx-6 / (x) = 4 . For = 0.7, 8 ;::;0 .5 or 0.6 .
E
down unanswered questions, you do not have to (The right side is the more restrictive.)
know the answers to everything you write.
5. limx-s f(x) = 2. For E = 0.3, 8 ;::;0.5 or 0.6 .
(The right side is more restrictive.)
7. a. Graph. (Same as Problem 1.)
CHAPTER 2 b. limx-3 J (x) = 5.
c. Graph is symmetrical about x = 3.
Exploratory Problem Set 2 - 1 Max. 8 = 3 - (3 + sin - 1 (-0.25)) = 0 .25268 . . . .
Numerical Approach to the Definitionof Limit d. Max. 8 = sin - 1 (E/ 2),
which is positive for any positive value of E.
1. a. Grapher graph agrees with text graph .
9. a. Graph. (Same as Problem 3.)
b. X f(x) b. lim x-G f (x ) = 4 .
1.9995 2.920629947 .. . c. The right side is more restrictive.
1.9996 2.9263193 70 .. . Max . 8 = (7 - (2 .3 / 3) 3 ) - 6 = 0.5493 . . . .
1.9997 2.9 33 056704 .. .
d. Max. 8 = 1 - ((3 - E) / 3) 3,
1.9998 2.941519645 .. .
which is positive for all positive values of E.
1.9999 2.95358 4 111 . ..
2 3 11. a. Graph. (Same as Problem 5 .)
2.0001 3.04641 5888 .. . b . lim x-5 J (x ) = 2.
2.0002 3.058480354 .. . c. The right side is more restrictive.
2.000 3 3.06694 3295 . . . Max . 8 = (5 + Jo.3) - 5 = 0.54772 ....
2.0004 3.073680629 . . . d. Max. 6 = .ft, which is positive for all E > 0.
2.0005 3.079370052 .. .
13. Limit s Applied to Derivat ives Prob lem
C. 1.9997 S X S 2.0003.
3t 2 - 48
d. 1.9999 S X S 2.0001. a. m ( t) =
t- 4
e. Yes. To make 3 - 0.01 < f (x) < 3 + 0.01, keep x b. Graph. Removable discontinuity at x = 4.
in the interva l 2 - 0.01 3 < x < 2 + 0 .01 3 . That is,
1.999999 < X < 2.000001 (but X * 2). d(t )
30
f. L = 3, c = 2, epsilon = 0.07, and delta = 0.0003.
g. Keep x in the interval 2 - epsilon 3 < x <
2 + epsilon 3.
3. Graphs . No limit. Graph cycles infinitely as it ap-
proaches x = 3.
c. Limit = 24 ft / sec.
h(x)
d. m ( t) = 3t + 12, if t * 4.
Keep t within 0.04 unit of 4.
Problem Set 2-3
The LimitTheorems
X
1. Limit of a Function Plus a Function Problem
Graph.
6 70 Appendix
B:Answers
toSelec
tedProb
lems
/
7. a. Graph .
y g+h
/
/ l(x)
'
\
\
/
/
10 /
' /
/
'.<'. h X
!!-//
X
limx- 2f(x) = 10, limx - 2B(X) = 4, and
limx- 2 h(x) = 6.
.'. limx-2 f(x) = limx -2 g(x) + limx - 2 h(x), b. limx - 4 f (x) = 5 .
Q.E.D. c. Proo{."
Example: Use each 0.01 from x = 1.97 to 2.03 . limf (x ) = lim (3x - 7)
X f(x) x- 4 x-4
1.97 9.9722 .. . = lim 3x + lim( -7)
1.98 9.9810 .. . x-4 x-4
1.99 9.9902 . . . Limit of a sum.
2.00 10
2.01 10.0102 . . . =3 limx -7
x-4
2.02 10.0209 . . .
2.03 10.0322 . . . Limit of a constant times a func-
All of these f (x) values are close to 10. tion , and limit of a constant.
3. Lim it of a Constant Prob lem = 3 4- 7 = 5,Q.E.D .. .. Limi.t of x.
Graph. 9. a. Graph .
l(x)
~ Limit=7
7
X
3 - 13
The limit is 7 since f (x) is always close to 7, no
matter what value x takes on. (It shouldn't bother b . limx-3f(x) = -13
you that f (x) = 7 for x =t= 3, if you think of the c. Proo{."
definition of limit for a while .) limf (x) = lim(x 2 - 9x + 5)
X 3 X-3
5. Lim it of a Product Prob lem
1.5 2 = 2.25 and tan 1.5 = 14.101419 .. . = limx 2 + lim (-9x) + lim5
x-3 x-3 x- 3
.'.f(l.5) = (2.25)(14.10 .. .) = 31.72819 ...
Limit of a sum .
l(x) = limx limx + (-9) limx + 5
30 x- 3 x-3 x- 3
Limit of a product, constant times
function, and constant.
X = 3 3 + (-9)(3) + 5 = -1 3, Q.E.D.
1.5
Limit of X.
By tracing on the graph or by table close to x = 1.5, 11. a. Graph.
with an increment of 0.000001,
r(x)
j(l.499980) = 31.71835 . . .
f(l.500020) = 31.73803 .. . -2 X
Since both of these are within 0.01 of the limit, and
the next increment of 0.000001 gives values further
away, let o = 0.000020 . -8
If x = 1.5 + o then x 2 is within 0.0000600004 of
1.5 2 and tanx is vvithin 0.0039981 ... of tan 1.5.
Pro
blemSet 2-3:TheLimit
Theorems 671
b. r(x ) = x - 6, x * - 2 17. a. Graph.
limx-- 2 r(x) = -8.
/(x )
c. Proof: X
limx-- 2 r (x ) = limx-- z(X - 6 )
Since x * - 2.
= limx-- 2 X + limx-- 2(- 6 )
Limit of a sum.
= - 2- 6 = - 8, Q.E.D.
Limit of X.
b. j(x) = x 3 - 9x 2 + 3x + 5, x *2
13. a. Graph. lim x - 2f(x) = - 17
c. Proof:
41 l(x ) .J
limx- 2 f (x )
= limx- 2(x 3 - 9x 2 + 3x + 5)
Since x * 2.
10
X
= limx- 2 x 3 + limx-2 (-9 x 2)
5
+ limx -2 3x + limx-2 5
Limit of a sum .
= limx- 2 X limx- 2 X limx- 2 X
b. f (x) = x 2 + 2x + 6, x *5 - 9 limx -2 X limx-2 X
limx- s f(x) = 41. + 3 limx- 2 X + 5
c. Proof: Limit of a product and a constant.
= 2 2 2 + (- 9) ( 2 2 ) + 3 2 + 5
limx-s f(x)
= limx- s (x 2 + 2x + 6 )
= - 17, Q.E.D.
Limit of x.
Since x * 5.
= limx- 5 x 2 + limx- s (2x ) + limx- 5 6 19. Limit of a Composite Function Problem
Limit of a sum. a. Graph. y is close to 1 when x is close to 0.
= l.iCT
x - 5 X limx-5 X + 2 limx-5 X + 6
Y= cos X
Limit of a product, limit of a
/ Close to 1
constant 1
= 5 5 + 2 5 + 6 = 41, Q.E.D. X
Limit of x. 0
15. a. Graph.
b. Graph. y is close to tan 1 = 1.557 .. . when xis
close to 1.
X Y = tan x
2
tan 1
b. f (x ) = x 2 - 5x + 3, x * -1 X
limx-- 1f (x ) = 9.
c. Proof:
limx- - 1f (x ) c. Graph. y stays close to tan 1 = 1.557 ... when x
= limx-- d x2 - 5x + 3) is close to 0.
Since x * - 1. y = tan (cos x )
= limx- - 1 x 2 + limx- - 1(- 5x ) + limx-- 1 3 2 / Close to tan 1
Limit of a sum. ~
= limx-- 1 X lin1x- - l X
+(-5) limx-- 1X +3 X
Limit of a product and a constant.
= (- 1)( - 1) + (- 5)( - 1) + 3 = 9, Q.E.D.
Limit of X.
672 Appendix
B:Answers
toSelected
Problem
d. limx- o cosx = 1. limu- 1 tanu = 1.55 .... 11. a. Has no right and left limits .
limx-o tan (cosx ) = 1.55 .... b. Has no limit.
21. Exact Derivative Problem c. Discontinuous. No limit. No value for j(-2).
a. Derivativ e :::::,
12.61
13. a. Has right and left limits.
x3 - 8 (x - 2) (x 2 + 2x + 4)
b x -2- - -------- x-2 = x 2 + 2x + 4, b. Has a limit.
provided x * 2. c. Cont inu ous. (Note that th e x-value, 5, is not at
This expression approaches 12. th e discontinuity.)
Proof: 15. a. Has right and left limits.
limx- z(x 3 - 8 / x - 2) = limx-z (x 2 + 2x + 4) b. Has a limit.
Since x * 2. c. Discontinuous. j( l ) * limit.
= limx- 2 x 2 + limx- 2 2x + limx- 2 4
Limit of a sum . 17. a. Has right and left limits.
= limx-2 X limx-2 X + 2 limx- 2 X + 4
b. Has no limit.
Limit of a product and limit of a
constant. c. Discontinuous. No limit.
= 2 2 + 2 2 + 4 = 12, Q.E.D
Limit of x. 19. a. Has right and left limit s.
b. Has a limit.
c. Graph . The line appears to be tangent to the
graph off at (2, 8). c. Discontinuous. No j(c).
21. Graph. (example) 23. Graph. (examp le)
f(x) f (x)
12
X X
11.8 235 ... (n = 100)
23. Integral :::::, 3 5
25. Journal Problem (Not selected.)
Problem Set 2-4
25. Graph . (example) 27. Graph. (example)
Continuity
f(x ) l (x )
)
1. a. Has right and left limits.
b. Has no limit.
c. Discontinuous. Has no limit.
3. a. Has right and left limits .
b . Has a limit. X
X
-3
c. Continuous . 2
5. a. Has right and left limits.
b. Has a limit.
29. Graph. (example) 31. Graph. (example)
c. Continuous.
f(x ) l(x)
7. a. Has right and left limits.
b . Has a limit.
6
c. Continuous . (Note that the x-va lu e, -3 , is not at
the asymptote .)
9. a. Has right and left limits .
/' 3
X X
b. Has a limit . 2 5
-2
c. Discontinuous . No valu e for j(-6).
Prob
lemSet2-4: Continuity 6 73
33. Graph . (example) 35. Graph. (example) 51. Graph. Discontinuous . There is no value of h (2) .
/ (x ) l(x)
h(x)
... 8
X
1(4) X
-4
37. Graph . (examp le) 39. Graph. (example) 53. a. Graph .
Not continuous
/(x) l(x) (No valu e for j (2))
6
d(x)
X
X 3 .
41. Graph . (example)
l(x)
b . limx-2 - d(x) = 3,
limx-2 + d (x) = 3,
Limit = 3.
55. a. Graph.
X ot continuou s
-9
m(x)
43 . Discontinuous at x = -3.
45 . Discontinuous at x = rr / 2 + rrn, where n stands for
an integer.
47. Graph. Discontinuous. limx-z j (x) = 2 and X
j(2) = 3 .
l(x)
b. limx-2 - m (x ) = 9,
limx-2 + m (x ) = 7,
No limit.
57. a. Graph. Not continuous
49. Graph . Discontinuous becaus e s(x) has no limit as
X - 2- . X
(No real function values to the left of x = 2)
s(x)
b. limx-2 - T (x ) : non e.
X
2
limx-2 + T (x ) = 3,
No limit.
67 4 Appendix
B:Answers
toSelected Problems
59. k = 1.3. Graph. Problem Set 2-5
Limits Involving Infinity
f(x )
1. Graph. (example) 3. Graph. (example)
f(x )
__ _ __ Z f(x )
X
----- -5 ______ _
61. k = 2.5. Graph. 63. k = - 1/ 2. Graph.
u(x ) 5. a. Graph.
X
;\__
f(x)
3
--------- ----
X
X
65. k = -1 or 2. Graphs .
f(x)
k= -1
b. lim x-3 + J(x) = oo , limx-3- J(x) = - oo,
limx-3 J(x), non e, limx-oo J(x ) = 2,
X limx- - ooJ(x) = 2
c.6=1 /9 8
X d. D = 100 3
7. a. Graph.
r(x)
3
67. Two Constants Problem
a. b - 1 = a.
b . a = -1 ~ b = 0. Graph . Continuous at x = 1.
X
f(x) f(x)
X X
b. lim x-oo r(x) =2
c. D = 1000
e.g., a= 1, b = 2 d. The line y = 2 is not an asymptote since the
a=-1 , b=O
graph crosses this line an infinit e numb er of
times as x increases.
c. e.g., a= 1 ~ b = 2. Graph, above . e. limx-o r(x) = 3. By zooming in on th e graph
Continuous at x = 1. around x = 0 you can see that r(x) is very clos e
to 3 wh en x is close to, but not equal to, zero .
69 . Continuity of Polynomial Function s 9. The limit is infinit e. Y is unbounded as x approaches
For any value of c, P( c) is determined by addition infinity. If ther e were a numb er E such that log x < E
and multipli cation. Since the set of real numbers is for all x > 0, th en you could let x = 10 2 , so that
closed under multiplication and addition, P (c) will log x = log 10 2 = 2, which is great er than E which
be a unique, real number for any real value x = c . was assumed to be an upper bound.
P (c) is the limit of P (x ) as x approaches c by the
properties of th e limit of a product of functions (for 11. Limits Ap pli ed to Integral s Probl em
powers of x), limit of a constan t times a function a. The definit e integral is the product of the
(for multiplication by th e coefficients), and limit independ ent and depend ent variables. Since
of a sum (for the individual terms) . Therefore, P is distanc e = (rat e)( tim e) , the integral represents
contin uous for all values of x. distan ce in this case.
Problem
Set2-5:Limits
Involving
Infinity 675
b. Tg = 17.8060005 .. . 9. Foot Race Problem
T4s = 17.9819616 .. . Let f (t) = Jesse 's speed - Kay's speed.
Tgo = 17.9935649 . . . j(l) = 20 - 15 = 5, which is positive.
T4so = 17.9994175 .. . j(3 ) = 17 - 19 = - 2, which is negative .
c. The exact answer is 18. It is a limit since the sums Since the speeds are assumed to be continuous,
can be made as close to it as you like, just by f is also continuous, and th e intermediate value
making the number of trapezoids large enough theorem app lies.
(and thus their widths close to zero). The sums Thus, there is a value of t between 1 and 3 for which
are smaller than the integral since each trapezoid j ( t ) = 0, meaning that Jesse and Kay are going
is inscribed under the graph, and thus leaves out exact ly the same speed at that time.
a part of its respective strip of the region. The existence of the time tells you neither what that
d. Tri is within exactly 0.01 unit of 18 when it equals time is, nor what the speed is. An existence theorem
17.99, which occurs between n = 45 and n = 90. such as the intermediate value theorem does not
By experimentation, tell these things.
T 57 = 17.9900158 ...
11. Cosine Function Problem
T55 = 17.9897900 .. .
The intermediate value theorem cannot be used on
Therefore the approximation is within 0.01 units
the cosine function until it has been proved that
of 18 for any value of n ::c:.67 .
cos is continuous. Once it has been proved (Chap-
13. Search lig ht Problem ter 3), you can conclude that there is a number
Length= 100 sec x = 100 / cos x c = cos - 1 0.6 = 0.92729521 .. . .
x must be within 0.100167 ... unit of rr / 2.
13 . The Extreme Value Theorem
Problem Set 2-6 This means that a function graph has a high point
The IntermediateValue Theoremand Its Consequences and a low point on any interval in which the function
is continuous. Graph .
l. f is continuous because it is a polynomial function.
j(l) = 18, j(4 ) = 3. l(x)
Since 8 is between 18 and 3, there is a number x = c
between 1 and 4 for which f(c)
Graph. c = 1.4349 ...
18
f(x)
= 8, Q.E.D.
N ,
I
I
I
I
..
, X
8
j If the function is not continuous, there might be
3 X
a point missing where the maximum or minimum
1 '
C 4
would have be en. Graph .
3. Converse of the Intermediate Value Theorem ? f(x)
a. For 1 ~ y < 2 or for 5 < y ~ 8, the conclusion
would be true. But for 2 ~ y ~ 5 it would be false
since there are no values of x in [l, SJwhich give
these value s for f (x).
b . The conclusion of the theorem is true since every a b
number y in [4, 6) is a value of g (x) for some
valu e of x in [l, 5].
5. Let j(x) = x 2 . Another possibility would be a graph with a vertical
Since f is a polynomial function , it is continuous asymptote somewhere between a and b.
and thus the intermediate value theorem applies .
Since J(l) = 1 and !(2) = 4, there is a number c be-
Problem Set 2-7
tween 1 and 2 such that J(c) = 3. By the definition
of square root, c = ./3, Q.E.D. Chapter Review and Test
7. The intermediate value theorem is called an exis- RO. (Not selected.)
tence theorem since it tells you that a number such
as ./3 exists. It does not tell you how to calculate Rl. a. See text definition of limit.
that number. b. Graph. f (2) = 0 115 + 3 = 3, Q.E.D.
676 Appendix
B:Answers
toSelected
Problems
f(x ) g(x) = x 2 - lOx + 2, x * 3
limx-3 B(X) = - 19, whi ch agrees with the graph.
3
c. The part of the definition is, ". . . , but not equal
to c, .... " Since the definition restricts x away
from 3, the quantity (x - 3) in the denominator
X
cannot equal zero. So you are allowed to cancel.
d ( )= (x +3)( x - 2)
. m X X - 4
X j (x ) . (6 + 3) (6 - 2)
1.9995 2.78132 .. . lrmx- 6 m(x) = ( _ ) = 18.
6 4
1.9996 2.79087 .. . Proof:
1.9997 2.80256 . . . limm (x ) = limx- 6(X + 3)( x - 2)
1.9998 2.81794 .. . x- 6 limx-6 (X - 4 )
1.9999 2.84151. . .
2 3 Limit of a quotient.
2.0001 3.15848 .. . limx- 6(X + 3) limx - 6(X - 2)
2.0002 3.18205 .. . limx- 6(X - 4 )
2.000 3 3.19743 .. .
2.0004 3.20912 .. . Limit of a product.
2.0005 3.21867 . . . 9 .4
Keep x within 0.0003 unit of 2. 2
c. Graph . Graph has a cusp and a minimum point. Limit of a linear function (3 times).
There is a limit.
= 18, Q.E.D.
g(x ) For limit as x approaches 4, the property cannot
be used since the denominator goes to zero . The
3
property specifically excludes the denominator
function having zero as its limit.
X
e. Chuck's Rock Problem
2
From 5 to 5 .1 sec, average velocity = - 15. 5
m/ sec.
35t - 5t 2 - 50
av. vel. = - --- - = - 5(t - 2), fort* 5.
R2. a. Maximum value of 8 is 0.0003 2. t- 5
instantaneous velocity = limit = - 5 ( 5 - 2) = - 15
b . Graph is symmetrical about x = 2.
m/ sec.
Let j (x ) = 3 + E.
Rate is ne gativ e, so distanc e above starting point
(X-2 ) 115 +3= 3 + E
is getting smaller, which means the rock is going
(x-2) = E5
down .
Let 8 = E5 , which is positive for all 8 > 0.
Instantaneous velocity is a derivative .
c. limx - 2f (x ) = 3. Maximum 8 is 0.6 or 0.7.
R4. a. See text definition of continuity at a point.
d. Left sid e of x = 2 is more restrictive.
Maximum value of 8 is 0.64. b. i. Graph . (example) ii. Graph. (example)
e. Let f (x ) = 3 - E. y y
2+Jx=l =3- E
X=(l - E)2 + 1 /
Let 8 = 2 - (( 1 - E)2 + 1) = 1 - (1 - E)2 which is
positive for all positive E < 1.
X X
R3. a. See the limit property statements in th e text.
b. Graph .
iii. Graph. (example) iv. Graph. (example)
g (x ) y y
-1 9 .
3
X
L.
3 4
X
Prob
lemSet 2-7: Chapter Rev
iewandTest 677
v. Graph. (example) vi. Graph. (example) d. n trapezoidal rul e
y y 50 467.9074 .. .
100 467.9669 ...
1(6)
200 467 .9882 ...
400 467 .9958 . . .
The limit of these sums seems to be 468.
X X D = 223.
6
R6. a. See text stat ement of int erm ediat e value theorem .
vii. Graph. (example) Basis is th e completeness axiom.
y See text statement of extreme value theorem .
5 ~ The wor d is corollary .
b. f(3) = 8, f(4) = -4 .
X
So f(x) = 0 for some x between 3 and 4 by the
int erm ediat e value theorem .
The prop er ty is continuity.
The valu e of x is approximately 3.7553.
c. i. Graph. c. Graph .
t (x )
f(x)
2 X
X --4
ii. The left limit is 4 and the right limit is 2. f( - 6) = 1 and f(-2) = 5
So f is discontinuous at x = 2, Q.E.D. Pick y = 3, and ther e is no value of x. Thi s fact
iii. k = 12. do es not con tradi ct the interm ediate value th e-
RS. a. See th e text definition of infinit e limit. orem. Function f does not me et the continuity
b. Graph. (example) hypoth esis of the th eorem .
/ (X) I
I
CHAPTER 3
I
3 --,
I
Exploratory Problem Set 3-1
X
Graphical Interpretation of Derivative
2'I
I
1. Graph is correct.
c. f(x) = 6 - 2-x . 3. m(x) = x - 3, provid ed x * 5.
limx-oo f (x) = 6 since 2- x approaches zero. 5. m(S) has the form 0/ 0, whi ch is an indeterminate
Graph . form. It is und efined becaus e of division by zero.
t(xl I
6 = l =--- - 7. The lin e is tangent to th e gra ph .
0.1 unit 9. (Not selected.)
Problem Set 3-2
Difference Quotients, and One Definition of Derivative
X
1. See text definition of derivativ e.
D
3. a. f' (3) = 3.6
Let D = 9.9653 . ... Keep x > D. b. Graph of the diff eren ce quotient , m(x).
67 8 Appendix
B:Answers
toSelecte
d Problems
m(x)
C. X j(x)
2.997 667.66 ...
2.998 1001
2.999 2001
3.000 error
X
3.001 -1999
3
3.002 - 999
3.00 3 - 665 .66 ...
c. and d. Graph of the function and tangent line . The difference quotients are all large positive
numbers on the left side of 3. On the right side
/(x )
they are large negative numbers. For a derivative
to exist the difference quotient must approach
the same number as x gets closer to 3.
19. Tangent Lines as Lim its of Secant Lines
X
a. The tangent line on the graph has a slope of -1.
j' (x ) =- 1, Q.E.D.
5. f' ( - 2) =1 7. j'(l) =-4 b. Graph. As the x-distance between the point and
3 decreases, the secant lines (solid) approach the
9. j'(3) = - 0.7 11. j'(-1) =0 tangent line (dotted).
13. The derivative of a linear function equals the slope.
The tangent line coincides with the graph . /(x )
15 . a. First, find f' (1); then plot a stra ight line through
point ( l ,f ( 1) ) using f' (1) as the slope.
b. Near the point (1,f (l ) ), which is (1, 1), the
tangent line and the curve appear coincidental.
c. The curve appears to get closer and closer to the
line and finally touch it at the point ( l ,f ( 1) ) .
d. Near point ( 1, 1) the curve looks linear.
e. If a graph has local linearity, the graph near that
point looks like the tangent line ; therefore the
derivative at that point could be said to equal the
slope of the graph at that point.
17 . a. Graph .
f(x )
c. The same thing happens with secant lines from
7 the left of x = 3. See graph, above.
5 d. Graph . g(x) = 4 - 61cos %x i
b. Difference quotient is
x-5
m(x )=--
x-3
Graph.
m(x )
,3
I
I
I
I e. A derivative is a limit. Since the left and right
limits are unequal, there is no derivative at x = 3.
Problem
Set3-2: Difference
Quotients,
andOne Definit
ionofDerivative 679
- 61cos !!xi
6 5. Sinu soid Prob lem #1
f. m (x) = . By table,
x - 3 a. Graph .
X m(x)
y
2.9 3.1401 .. .
2.99 3.1415 .. . 4
3 error
3.01 -3. 1415 .. .
, - ' r
3.1 -3 .1401 .. . / \ ,/ ' X
Conjecture: The numbers are rr and - rr .
Problem Set 3-3 b. Amplitude = 1, period = 2rr = 6.28 3 ...
DerivativeFunctionsNumericallyand Graphically c. The f' graph ha s amplitude 1 and period 2rr.
1. Cubic Function Problem # 1 d. Graph.
a. Graph . y
40 y
// -2 ' / 5 ' /
' - "', , and g' -...,
/
/
/
/
I
I
I
The graphs of f and 9 are the same shape,
spaced 1 unit apart vert ically. The graphs off'
and g ' are iden tical! Thi s is to b e expected since
b. f' (x) is positive for -2 < x < 2.
the shapes of th e f and g gra phs are th e same.
The graph off is increasing for these x-values .
c. j(x) is decreasing for x satisfying lxl > 2. 7. (Not selected)
f' (x) < 0 for these values of x. 9. Tolera nce Problem (Epsilon and Delta)
d. Where th e f' graph crosses the x-axis, the f a. Area is wit hin 0.2401 in 2 of the nomin al.
graph has a high point or a low point. b. Let x be the numb er of inch es.
e. See th e graph in part (a). Keep x within 0.000 8 in. of 12 in.
f. Conjecture: f' is quadrati c. c. The 0.02 in part (b) corres pond s to E an d the
3. Quartic Funct ion Probl em #1 0.0008 correspon ds to fJ.
a. Graph. 11. Difference Quot ient Acc ura cy Problem
a. j'( l ) = 2.
200 y b . Forward: 2.31
h
I
I Backward : 1. 71
I
I Symmetric: 2.01
I
/ h' The symmetr ic differ ence quoti ent is closer to
/
I
X
the actual b ecause it is the average of the other
/-2 2.5
two, and the other two spa n the actua l derivative.
I
c. f ' (O) = - 1.
d . Forward: - 0.99
b. The h ' graph looks like a cubic function graph. Backward: - 0.99
Conjecture : 7th degree function has a sixth Symmetric: - 0.99.
degree function for its derivative . All thr ee diff ere nce quotients are equ al since
c. By plotting the graph u sing a friendly window, f(x) cha nges just as much from - 0.1 to Oas it
then tracing, the zeros of h ' are - 2, 1, 2. 5. do es from Oto 0.1.
d. If h' (x) = 0, the h graph ha s a high point or a low e. The journ al entry shou ld note that in gene ral the
point. This is reasonabl e becaus e if h ' (x) = 0, symmetr ic difference quoti ent is more accura te
the rate of change of h (x) is zero, which would than eith er the forward or back war d one, except
happ en when the graph stops going up and starts when the function is increasing or decreas ing at
going down, or vice versa. th e same rate on both sid es of x = c.
e. See graph in part (a). 13. Journal Problem ( ot selec ted)
680 Appendi
x B:Answers
toSelectedProblem
Problem Set 3 -4
Derivative of the Power Function, and Another Definition of
Derivative
\
1. f' (x) = 20x 3 '
3. dv/dt = -o .ss1t- 84
5. M'(x) =0
b. The graph off' is shown dotted in part (a).
7. dy / dx = 0.6x - 8 c. There appear to be only two graphs because the
exact an d the numerical derivative graphs almost
9. _Ec._(13
- x) = -1
d.x coincide.
11. dy / dx = 2.3xl.3 - iox - 3 - 100 d . f(3) = -6.2
f' (3) = 3.8 (by formula)
13. dv /dx = 18x - 24 f' (3) ::,:;3.8000004 (depending on grapher)
15. j'(x) = 24x 2 + 120x + 150 The two values off' ( 3) are almost identical!
29 . j(x) = x 112+ 2x - 13
17.P'(x) = X-l
j'(x) = x- 112 + 2j'(4) = ~
19. j(x) = 7x.J Increasing by 9/4 y-uni t s per x-unit at x = 4.
f '(x) m h -0 -c
= 11 I I ) -
x+ 1 -rx
h
4 31. Decreasing by 1.5 y-units per x-unit at x = 9.
33. Graph. High and low points of the f graph are at
= lim,, _ 0 (28x 3 + 42x 2 h + 28xh 2 + 7h 3 ) = 28x 3 . the x-intercepts of the f' grap h.
By formula, f' (x) = 7 4x 3 = 28x 3 , which checks.
21. v'(t)
= lim [10(t +h)2 - s(t+h)+7J- [1012-s 1+7J
1,- 0 h
2-s 11= lim(20t + 10h - 5)
= lim zoth+1011
h- 0 h h- 0
= 20t - 5.
By formula, v ' (t) = 10 2t - 5 = 20t - 5, which 35. Formula Proof Problem No. 1
checks. If j(x) = k g(x), thenj'(x) = k g '( x).
Proof:
2 3. Misconception Problem
j'(x) = lim11--o f' (x + h,~-f(x)
Mae should realize that you differentiate functions,
not values of functions. If you substitute a value for _ . kg(x + h )- k g(x)
- 1lilll,-0 11
x into j(x) = x4, you get j(3) = 34 = 81, which is
= limh -0 k g(x+ h,~-g(x)
a new function, g(x) = 81. It is the derivative of g
that equals zero. -- k 11m
. h- 0
g(x+ h )-g(x)
h
Moral: Differentiate before you substitute for x. = k g' (x), Q.E.D.
25. Graph . Dashed line is the derivative. 37. Derivative of a Power Formula
If f(x) = xn , thenf'(x) = nxn - l_
Proof:
f' (x) = lim 11-o (x+h/;'- x"
. x" +nx 11- 1 11+zn(n- l )x11- 211
2+ + h" - x"
\
\
= lim11- o h
= lim1,-o( nx 11- 1 + n(n - l)xn - 2h + + h 11- 1)
\
\
X = nx" - 1 + 0 + 0 + + 0
9 = nxn - L, which is from the second term in the
binomial expansion of (x + h)n, Q.E.D.
39. Introduction to Antiderivatives
a. j(x) = x 3 - 5x 2 + 5x
b. g(x) = j (x) + 13 is also an answer to part (a)
27. Numerical vs. Exact Derivative Problem since it has the same derivative as j(x). The
a. Graph. derivative of a constant is zero.
Problem
Set3-4:Derivative
ofthePower
Function,
andAnother
Definition
ofDerivative 681
c. The nam e antiderivative is picked because it is b. At t = 1 the football is going up at 8 .2 m/ sec.
an inverse operation of taking the derivative. At t = 3 the football is going down at 11.4 m/ sec.
The ball is going up when the derivative is pos-
Problem Set 3-5 itive and coming down when the derivative is
Displacement,Velocity,and Acceleration negative .
The ball is going up when the graph slopes up
1. v = 20t 3 7.2t'- 4 + 7, a = 60t 2 l0.08t 0.4
- - and corning down when the graph slopes down .
3. x = - + 13t
t3 35t + 27. Graph.
2 - c. d (4 ) = -2 1.2, which suggests that the ball is go-
The object starts out at x = 2 7 ft when l = 0 sec. It ing davvn at 21.2 m/ sec . However, d (4 ) = - 6.4,
moves to the left to x = 0.16 ft when t = 1.7 sec. It which reveals that the ball has gone under-
ground. The function gives meaningful answers
turns there and goes to the right to x = 76 ft when
in th e real world only if the domain of t is
t = 7 sec. It turns there and speeds up, going to restricted to values that make d ( t) non-negative.
the left for all higher values of t.
y
11. Average Rate vs. Instantaneous Rate Problem
Turns at I = 7, X = 76.
The average rate is defined to be the change in the
dependent variable divided by the change in the
-
Turns at I = 1.7, X= 0. 16.
indep end ent variable (such as total distance divided
Starts at I = o, x= 27 by total time). Thus, the difference quotient is an
X average rat e.
10 The instantaneous rate is the limit of this aver-
age rate as the chang e in the independent variab le
approach es zero.
5. a. v = -3 t 2 + 26 t - 35, a = -6 t + 26
b. xis decreasing at 12 ft/ sec at t = 1. Problem Set 3-6
c. The object is slowing down at 20 (ft/ sec)/ sec Introductionto Sine, Cosine, and Composite Functions
since the velocity and acceleration are in oppo-
1. Graph.
site directions when t = 1.
y
d. At t = 7, x has a relative maximum sinc e v (7) = 0
at that point and is positiv e just befor e t = 7 and
negative just aft er.
e. No, x is never negative for t in [O,9). It starts out
at 27 ft, decreases to just above O around t = 1.7
sec, and does not become negative until some
time between t = 9.6 and 9.7 sec. . . . ' . . . . . . . . . . . . .. . . . .. . . . . . . . .. . . . .
. . . . . . . . ..
. . . ..
..... . .... .... .........................
.
7. Car Problem
a. Graph .
3. Conjecture: g' (x) = 3 cos 3x
300 y
Graph confirms conjecture .
X
d'
10
10
b. Velocity is positive for O ~ t < 15.
Calvin is going up the hill for the first 15 sec. 5. Conjecture: t '( x ) = 0.7x - 0 -3 cosx 0 -7
c. At 15 seconds his car stopped. Graph confirms conjecture!
d(15) = 324, so distanc e is 324 feet.
y
d. He'll be back at the bottom when t = 33 sec.
e. Car runs out of gas 99 ft from the bottom.
X
9. Velocity from Displacement Problem
10
a. d ' (l) = 18 - 9.8 = 8.2
d '( 3) = 18 - 9.8 3 = - 11.4
d ' is called velocity in physics.
682 Appendix
B:Answers
toSelected
Problems
7. a. Inside: 3x. Outside: sine. e. V = 4f (6t + 10 )3
c. Inside: cube. Outside: sine. ..~'; = 24rr(6t + 10) 2
e. Inside: tangent. Outside: reciprocal. When t = 5, !';
= 38400rr, which checks.
Problem Set 3-7
Problem Set 3-8
Derivativesof Composite Functions-The Chain Rule
Proofand Application of Sine and Cosine Derivatives
1. a. Let y = J(u), u = g(x).
1. Ferris Wheel Problem
_ c!.1!.
<!:2'_ . du
dx - du dx a. Graph . y(t) = 25 + 20cos fo(t - 3)
b. y' = j'(g(x)) g'(x)
c. To differentiate a composite function, differen-
tiate the outside function with respect to the
inside function, then multiply by the derivative
of the inside function with respect to x.
_ Increas ing
3. f ' (x) = -3 sin 3x 5 X
3 15 23
5. g '( x) = -3 x2 sin(x 3)
7. y' = -3 cos 2 x sinx
5
b. y' (t) = -2 rr sin -fu(t - 3)
9. y' = 6sin XCOSX
c. y'(15) = 3.69316 ...
11. y' = -18cos3x y(t) is increasing at about 3.7 ft/ sec.
13. /x(cos 4 7x) = -2 8 cos 3 7x sin 7x The seat is at y ( t) = 2 5 ft above the ground then .
15. j '( x) = 160sin 213
4xcos4x d. The fastest y (t) changes is 2rr, or 6.28 ... ft/ sec.
3. Playground Problem
17. j'(x) = 35(5x + 3) 6
a. Curb has slope (3.25 - 0.75 )/ 44 = 2.5 /44.
19. y ' = - 72x 2 (4x 3 - 7) - 7
.. equation is J( x) = 0.7 5 + (2.5/ 44)x
21. y' = - 200x cos 99 (x 2 + 3) sin(x 2 + 3) b. Sinusoid has period 8 ft, so B = 2rr / 8 = rr / 4.
23. Graphical Verification Problem Amplitude = 0.5(0.75 - 0.25 ) = 0.25 ft.
f' (x) = - sin 0.2x Low end of ramp is a low point on the sinusoid .
f' (3) = - sin 0.6 = -0.5646 ... and f (3) = 4.126 ... .. sinusoidal axis is at y = 0.25 when x = 0, and
Line has equation y = - 0.5646 ... x + 5.820 ... goes up with slope 2.5/ 44.
Graph. Line is tangent to graph. Sinusoid is at a low point when x = 0. So phase
displacement is O if the cosin e is subtracted .
.-. equation is
g(x) = 0.25 + ~-J
x - 0.25 cos :f(x )
(There are other correct forms.)
rr . rr ( )
C. g , ( x ) = TI
2.5
+ 16 sm -:r x
3
X
g' (9) = ~-J + ~ sin :f (9) = 0.1956 .. . ft / ft.
Going up at about 0.2 vertical ft per hori zontal ft
g '( lS) =~-J + ~ sin:f(lS) = - 0.0820 ... ft / ft .
Going down at about 0.08 vert ical ft per horizon-
tal ft.
25. Ball oon Volume Problem
Negative derivative implies g (x) is getting
a. !;,= 4rrr 2
(cm3 / cm), or cm 2 . smaller, and thus child is going down, and vice
b. r = 6t + 10 versa.
d. By tracing th e g ' graph, maximum value of g '(x)
c. ~; = 6 cm/ min
is 0 .253 1 ... ft/ft (about 14.2 up ).
dV dV dr
d. dt = dr dt Minimum is -0.1395 ... ft/ ft (about 7.9 down).
dV
When t = 5, = 6400rr .
dr 5. Pendulum Experiment (Not selected)
..!'; = 6400rr 6 = 38400rr cm 3/ min .
Since dV / dr has units cm 2 , and dr / dt has units 7. Squeeze Theorem, Numerically
cm/ min, dV / dt has units cm 2 ~:, a. Graph. j(x) = -2x 2 + Bx - 2, g(x) = 2x 2 + 2,
which becomes cm 3 / min, Q.E.D. h(x) = 4x. Limits are each equal to 4.
Problem
Set3-8:Proof
andApplication
ofSineandCosine
Derivat
ives 683
y 25 . Derivative and Antiderivative Problem:
2
a. g (x) = 0.3x +C
b. i. g(x ) = 0.3x 2
4
ti. g(x) = 0.3x 2 + 3
X
tii. g(x) = 0.3x 2 + 5
c. Graph. All the graphs are obtained by shifting
the solution function vertically through some
b. Each function is continuous since it is a polyno- displacement C. So all the graphs are "related"
mial function. and thus can be called a family.
j(x) s h(x) s g(x).
c. x J(x) h(x) g (x ) g (x )
0.95 3.795 3.8 3.805
0.96 3.8368 3.84 3.8432
0.97 3.87 82 3.88 3.8818
0.98 3.9192 3.92 3.9208
0.99 3.9598 3.96 3.9602
1.00 4 4 4 -
--g ' grap h
1.01 4 .0398 4.04 4.0402
1.02 4 .0792 4.08 4 .0808
1.03 4.1182 4.12 4 .1218 Problem Set 3-10
1.04 4.1568 4.16 4 .1632 Chapter Review and Test
1.05 4.195 4.2 4.205
d. From the table, /5 = 0.01 or 0.02 will work, but Review Problems
0.03 is too lar ge . RO. (Not selected)
e. All the values of h(x ) are between the corre- Rl. a. f' (2) "" 12
sponding values of j (x ) and g(x), and the three
functions all approach 4 as a limit. b. m (x) = -~:}
9. Limit of Sine Problem (See text proof.) m(2) takes the indeterminate form 0/ 0.
By tracing the graph or constructing a table, th e
11. Derivative of the Cosine Function (See text proof.)
limit of m(x) seems to be 12 as x approaches 2.
13. Group Discussion Problem (Not selected)
c. m(x) = x 2 + 2x + 4.
limx- 2 m(x) = 12
Problem Set 3-9
Antiderivativesand IndefiniteIntegrals d. The answer to part (c) is exactly 12.
The answer to part (a) is approximately 12.
1. j(x) =x7 +C = ix 6 + C
3. J (x ) R2. a. j'( c ) = lim f( x)- f(c)
x~c x- c
5. j(x) = - x - 8 + C 7. J(x) = sinx + C
b. j(x) = 0.4x x +5 2 -
9. j(x) = 8x 9 12 + C 11. j(x) = -0.2 cos Sx + C
2
f' (x) = lim O..Jx -_x + 5- 5.6
13. f(x) = tz (4x + 5) 8 + C x-3 x-3
15. J(x) = x + 3x
3 2
- 5x + C = Jim (x - 3)( 0.4x +0.2 )
x- 3 x-3
5
17. f(x) = x + 9.8 = lim(0.4x + 0.2) = 1.4
x -3
19. J(x) = sinx + 4
21. j(x) = x 3 - 4x 2 + 3x +
1 3
~
c. m(x) = 0.4
x;=;-O.G
. Graph .
23. Displacement Problem m (x )
a. d(t) = 70t - 4 .9t 2 + C 2
b. d(t) = 70t - 4.9t 2 + 6
c. d(5) = 233.5 meters
d(6) = 249.6 meters X
d(9) = 239.1 meters 3
These three numbers show that d(t) has a high
point somewhere between 5 and 9.
d. Arrow was highest, 256m, at about t = 7.1 sec . d. Line: y = l.4x + 1.4. Graph .
684 Appendix
B:Answers
toSelected
Problems
y h. j'(32 ) = 201.6 exactly .
Numer ical derivative is equa l to or very close to
201.6.
i. Graph. (Dotted line.)
4 y
X
e. The line is tangent to the graph.
f. Yes, f does have local linearity at x = 3. Zooming
in on the point (3, 5.6) shows that the graph looks
more and more like the line.
R3. a. Graph .
RS. a. v = !I or x ' (t).
y 2
a= du
dt
orv '( t ) ' a= dt2
d x orx " (t)
50
b. Spaceship Probl em
I '
I ' i. v = -0.03t 2 + l. 8t - 25
/ '
I , X
a = - 0.06t + l.8
I
ii. a(lS) = 0.9 (km/sec) / sec
I
I
v(lS ) = - 4.75 km/ sec
The spaceship is slowing down at t = 15 since
b. Y2 = num erical derivative . See graph in part (a). the velocity and acceleration hav e opposite
c. The Y r graph has a high point or a low point at signs.
each x-value wher e the Y2 graph is zero. iii. v = -0.0 3t 2 + l.8t - 25 = 0
d. Leaky Tir e Problem By SOLVE or quadratic for mul a,
Graph. t = 21.83 5 .. . or t = 38.164 .. . .
The spaceshi p is stopped at ab ou t 21.8 and
y 38.2 secon ds.
iv. y = 0 whe n t = 50 . v(50) = - 10.
20
Since the spaces hip is going 10 km/ sec when
it reaches the surface, it is a cras h landing!
R6. a. Graph .
- - -p, - - - - - -- --- -
y
derivative cosine
Decreasing at about 2.69 psi/hr when t = 3. /
I \
Decreasing at about 1.96 psi/hr when t = 6. \ J(
Decreasing at about 3.69 psi/hr when t = 0. \
\
The units are psi/hr.
The sign of the pr essur e change is negative since
the pressure is decreasing.
Yes, the rate of pr essure change is getting closer b. The graph of th e derivative is the same as th e
to zero . sine graph, but inverted in they-direction. Thus,
R4. a. See text definition of derivative. (cosx )' = - sinx is confirme d.
b. Differentiate. C. COS 1 = 0.5403023058 . . .
c. If y = xn, then y ' = nxn - l . Num erical derivative ::::::
0.5403022158 . ..
The two ar e very close!
d. See solut ion to Problem Set 3-4, Problem 35.
d. Composite function.
e. See the proof in Section 3-4. f' (x) = - 2x sin (x 2 )
f. ~ is pronounced, "Dee y, dee x." r;/2:__ r;/2:_ du
R7 a. 1.
/x (y ) is pronounced , "Dee, dee x, of y ." d x - du dx
ii. j (x) = g(h(x)) ~ j'(x) = g ' (h(x) ) h ' (x)
Both mean the derivative of y with respect to x.
g. i. f' (x)
= x OB iii. The derivative of a composite fun ction is th e
derivative of th e outside function with respect
ii. g'(x ) = -2 8x - 5 - x - 1 to the insid e function, times the deri vativ e of
iii. h' (x) =0 th e inside function with respect to x .
Problem
Set3-l 0:Chapter
Review
andTest 685
b . See the derivation in the text. This derivation R9. a. f (x ) = 6x 6 + C
constitutes a proof. b . y = - 5 cos0.2x + 8
.6.u must be non-zero throughout the interval. c. Distance Problem
c. i. f (x ) = (x 2 - 4) 3 y (t ) = 4t 3 12 + 100
f'(x ) = 3 (x 2 - 4 ) 2 . 2x = 6x (x 2 - 4) 2 y(60 ) = 1959.03 .. .
ii . j(x ) = x 6 - 12x 4 + 48x 2 - 64 V (60 ) = 46.475 . . .
f' (x) = 6x 5 - 48x 3 + 96x She is about 1959 feet away from the house,
Expanding the answer to part (i) gives goin g about 46.5 ft/ sec 1 minute after she starts .
f' (x) = 6x 5 - 48x 3 + 96x , which checks. d. An ant iderivative and an indefinit e int egral ar e
d. j'(x) = - 3x 2 sinx 3 the sam e thing.
g ' (x) = 5cos5x
h ' (x) = -6sinxcos 5 x CHAPTER 4
k ' (x ) = 0
e. Shark Problem Exploratory Problem Set 4-1
dd
i= ~~ . ~~ = l.8x 2
0.4 = 0.72x 2 Combin a tions of Two Functions
If X = 2, W = 0.6 2 3 = 4.8 lb 1. f' (x ) = - 3 sinx, g ' (x ) = 2 cos x
dW / dt = 0.72(2 2 ) = 2.88 3. Derivative of a Quotient of Two Functions
Shark is gaining about 2.88 lb/ day Graph.
If X = 10, W = 0.6 10 3 = 600 lb
dW / dt = 0.72 (102 ) = 72
Shark is gaining about 72 lb/ day
The chain rul e is used to get d W / dt from d W / dx
X
by multiplying th e latter by dx / dt.
10
. sinx
R8 . a. 1lIIl x-o )( =l.
b. X (sinx) / x
- 0.05 0.999583 38541 . . .
- 0.04 0.99973 33 5466 .. . q is the cotangent funct ion.
- 0.03 0.9 998 5000674 . . . q ' (2) = - 1.8141 ...
- 0.02 0.99993333466 .. . q (x ) is decreasing at x = 2.
- 0.01 0.99998333 341 .. . j' (2) / g '(2 ) = 3.277 5 .. . * j' (2)/ g'(2).
0.00 (No value) 5. Derivativ e of a Parametri c Function
0.01 0.99998333341 . . . Graph. x = 3 cos t, y = 2 sin t
0.02 0.9999 3333466 . . .
0.03 0.99985000674 .. . y
0.04 0.9997 3335466 . . . I = 2 here
0.0 5 0.999583 38541 .. .
X
The values approach 1 as x approaches 0.
c. See text statement of squee ze theorem .
Squeeze (sinx) / x betw een cosx and sec x.
d. See proof in text (Section 3-8).
e. cos x = sin(rr / 2 - x) .6.x = -0 .54466 .. . , .6.y = - 0.16618 ...
cos ' x = cos (rr / 2 - x ) (- 1) = - sinx, Q.E.D. d y / dx "' .6.y / .6.x = 0.3051 .. .
f. Clock Problem dy / dt 2 cos 2 .
At t = 2, -d / d = . = 0.3051 .. . , which
d(t) = 180 + 20cos :fut
X t - 3 Sill 2
d '( t) -- - 237T sin!I..t agrees with the difference quotient .
30 Also, from th e graph in Problem 4, the slope of a line
At 2, t = 10. d '( lO) "'- 1.81 cm/ sec. tangent to the graph at the point where t = 2 wou ld
At 3, t = 15. d '( 15) "' - 2.09 cm/ sec. be posit ive, and less than 1, which agrees ,vith th e
At 7, t = 35. d' (35 ) "' 1.05 cm/ sec . value of 0.3051 . .. for the difference quotient.
At th e 2 and 3, th e tip is goin g down, so the
distance from the floor is decreasing, which is Problem Set 4-2
implied by the negative derivatives. Deriva tive of a Product of Two Functio ns
At the 7, the tip is going up, as implied by the
positive derivative . 1. f' (x ) = 3x 2 cos x - x 3 sinx
686 Appendi
x B:Answers
toSelected
Problems
3. g ' (x) = l.5x 0 5 sin 7x + 7x1.s cos 7x 31. Derivative of a Power by Indu ction Problem
Prove that if f 11 (x) = xn, th en f;, (x ) = nxn - l for
5. dy / dx = x 6 (2x + 5) 9 (34x + 35)
all integers ;?. 1.
7. z ' = 4x 3 cos 5 3x - l 5x 4 cos 4 3x sin 3x
Proof: (by induction on n)
9. (d / dx)[(4x - 3)6 sin5x] If n = l, then Ji (x) = x 1 , which implies that
= 24(4x - 3) 5 sin Sx + 5 (4x - 3) 6 cos Sx J; (x ) = 1 = 1x 0 , which anchors the induc-
11. y' = (6x + 11) 3 (5x - 9) 6 (330x + 169) tion. Assume that for some integer n = k > 1,
fk (x ) = kx k-1.
13. P' = l0x(x 2 - 1) 9 (x 2 + 1) 14 (5x 2 - 1) For n = k + l.fk +l (x) = x k+t = (xk ) (x).
15. a'(t) = 12cos3tcos5t - 20sin3tsin5t By the derivative of a product property, fk +J(x) =
(xk )' (x) + (xk)(x)' = (xk) '( x) + xk.
17. y' = - 400 cos 7 Sx sin Sx sin 5 8x Substituting for (xk) ' from the induction hypoth-
+ 400 cos 8 Sx sin 4 8x cos 8x esis, fk +1 (x ) = (kx k- 1 )( x ) + xk = kxk + xk =
(k + l )x k = ( k + l )x (k+i>-1, completing the induc-
19. z' = 3x 2(5x - 2) 4 sin6x + 20x 3 (5x - 2) 3 sin6x
+6x 3 (5x - 2)4 cos 6x tion.
.-J ;,(x ) = nxn - 1 for all int egers ;?. 1, Q.E.D.
21. y ' = - 3 sin(3 sinx ) cosx
33. Confirmation of the Product Property
23. Product of Three Functions Problem
a. Graph . f (x) = x 3 sin x, with the graph off ' .
If y = u v w, where u, v, and w are differentiable
f' should have been sketch ed.
functions of x , then y' = u'vw + uv ' w + iww ' .
Proof: f(x )
I
y = uvw = (uv)w I
I 5
I
.-.y' = (uv)'w + (uv)w ' = (u ' v + uv ' )w I
I I
I
I I
X
+(uv)w '
I / I
= u ' vw + uv ' w + uvw ' ,Q .E.D. I
I
/ I
I f "I
25 . z ' = 5x 4 cos 6
x sin 7x - 6x 5 cos 5 x sinx sin 7x \
,, I
+ 7x 5 cos 6 x cos 7x
27. y ' = 4x 3 (2x - 3) 5 sinx cos 2x b. f' (x ) = 3x 2 sinx + x 3 cos x
+ 10x 4 (2x - 3) 4 sinxcos2x Graph in part (a) is correct.
+x 4 (2x - 3) 5 cosx cos 2x
c. The numerical derivative graph duplicates the
- 2x 4 (2x - 3) 5 sinx sin 2x
algebraic derivative graph, as in part (a), thus
29. Odd and Even Function Derivative Problem showing that the algebraic derivative is right.
Prove that the derivative of an odd function is an
even function, and the derivative of an even function 35. Pole Danc e Problem
is an odd function . a. A = LW
Proof: dA = dL . W + L . d W
For any function, the chain rule gives dt dt dt
b. At t = 4, dA / dt = 7.132 .. . , so A is increasing.
At t = 5, dA / dt = - 4.949 .. . , so A is decreasing .
d~f( - x) = f'( - x) (- 1) = - j'(-x).
Problem Set 4-3
For an odd function, Derivative o f a Ouotienf of Two Functio ns
2 3
_!!__f(- x ) = _!!__(-j(x) ) = -j'(x). 1. j' (x ) = 3x smx - x cosx
dx dx 2
sm x
.-. - j ' (-x) = -j'(x ) or f' (- x ) = j' (x ) ,
, ( ) = - 3 x cos 2 x sm
.
x - 5 cos 3 x,
3. g X x 6
and the derivative is an even function.
For an even function, 10 cos lOx cos 20x + 20 sin 20x sin lOx
5. y ' = ---------------
cos2 20x
d d
dxf(-x) = dxj(x) = j'(x). , 57
7
Y = (6x + 5) 2
.'. - f' ( - x) = f' (x) or f' (-x) = - f' (x),
dz= (Bx + 1)5 (120x + 141)
and the derivative is an odd function , Q.E.D. 9
dx (5x - 2) 10
Problem
Set4-3:Derivative
ofa Quotient
ofTwo
Functions 687
, = 60x 2 - 98x + 62 Proof
ll. P
(3x 2 + 6x - 8) 2
Let n = - p, where p is a positive integ er.
13. d~ (6ox - 413 ) = -8ox - 713 1
..y = x -P= -
xP
15. r'(x) = -36x - 4
. , O . x P - l px r - 1 since pis a positive
.. y =
l7. v'(x) = 7sin0.5x x 2P integer.
cos 2 0.5x
Px r - 1 p - l -2 p - p- 1
- 1 = -~ = -px = - px .
19. e' (x) = - = -x - 2
X2
Replacing - p with n gives
21. W ' (x) = 150x 2 (x 3 - 1)4
1
y' = nx 11
- , Q.E .D.
23. T' (x) = sec 2 x ("T" is for "tangent function.")
33 . Not selected.
25. C'(x) = - cscxcotx ("C" is for "coseca nt func-
tion.") Problem Set 4-4
27. Black Hol e Prob lem Derivativesof the Other TrigonometricFunctions
a. v(l) = 500 mph l. f' (x) = 5 sec 2 5x
v(2) = 1000 mph
v(3) = 1000 / 0. No value for v(3). 3. y' = 7 sec 7x tan 7x
1000 5. g ' (x ) = -11 csc 2 llx
b. a(t) = (3 - t) 2
7. r ' (x) = - 20 csc 20x cot 20x
c. a(l) = 250 mph / hr 9. (d / dx )(y) = 20tan.J4xsec 2 4x
a(2) = 1000 mph / hr
a(3) = 1000 / 0. No value for a(3). 11. (d / dx) (3 cot 6 8x) = -144 cot 5 8x csc 2 8x
d. Graph. 13. y ' = 40 sec 514 4x tan4 x
15. v ' = 7x - 8 csc(x - 7 ) cot(x - 7 )
I
aorv
I
I
17. dp / dx = secx tan 2 x + sec 3 x
I
a,
I
19. y ' = -3 x - .Jcotx - x - 3 csc 2 x
2000 I
i
,
I V
21. y ' = sec 2 x - csc 2 x
23. y' = secx tanx
3 35x 6 cot 14x + 70x 7 csc 2 14x
25 . y' = -----~2------
cot 14x
e. Range is O :,:;t < 1.58 5 .. .. 27. w ' = 3sec 2(sin3x) cos3x
29. Confirmation of Quotient Formula Problem 29. S'(x) = 0
31. A'(x) = 2xcosx 2
3x + 7 , 1
j(x) = 2x + 5 ~ f (x) = (2x + 5) 2 3 3. F' (x) = 2 sin x cos x
j'(4 ) = 0.005917159 ... 35. C'(x) = cos (sinx) cosx
37. Derivative of the Cotangent Problem
For 4.1,f' (4) "" 0.005827505 . . . COSX
y = cotx = - .- ~
smx
For 4.01,f ' (4) "" 0.005908070 .. .
- sin x sin x - cos x cos x
For 4.001, f' (4) "" 0.005916249 .. . y ' = ------,c-------
sin 2 x
f' (4) (exact) = 0.005917159 .. . -1
= -- 2 = -c sc 2 x or:
sin x
Difference quotients are approaching f' (4).
1
31. Derivative Proof for Negative Integer Exponents y = -- = (tanx) - 1 ~
tanx
If y = x", where n is a negative integer, then
y' = nxn - 1_ y' = -l (tanx )- 2 sec 2 x = - csc 2 x
688 Appendix
B:Answers
toSelectedProblems
39. Confirmation of Tangent Derivative Formu la: 1
11. y = csc x =>cscy = x => - cscycoty y ' =>
a. See graph in part (b). 1 1
y ' = --- -- -== = lf X > 0
b . J(x) = tanx =>f' (x) = sec 2 x. Graph. cscycoty x v'x 2 -l
If x < 0, then y is in QIV. So both csc y and cot y
Predicted graph should be close to actual one. are negative, and thus their product is positive .
I
. ' 1 Q
I
y
,. ,,
I '
I
.y = - lx lv'x 2 - 1' .E.D.
I
[Since csc y = (hypotenuse) / (opposite leg), put
tanl.01 - tan0 .99 = x on the hypotenuse, 1 on the opposite leg.
C. 2(0.01) 3 .426464 l 6 ...
Adjacent leg = v'x 2 - 1, and csc y = x and
tan ' 1 = sec 2 l = (1 / cos 1) 2 = 3.42551882 . .. cot y = (adjacent) / (opposite).]
Difference quotient is within 0.001 of actual. , 4 4
13
41. Light on the Monument Problem Y = cosy = v'l - l 6x 2
0 5 0.5x - 0 5
a. y / 10 = tanx =>y = lOtanx, Q.E.D. 15 y ' = __0 5x
--2-
b. At x = 1, y is increasing at 34 .2551 ... , or about . csc y l +x
34.3 ft / radian, which is 0.5978 .. . ft / degree . l7 '= 1 3
c. At y = 535, y is increasing at about 28,632.5 Y 3secytany 1x1Jx2- 9
ft / radian .
, lOx
1 19 . y = JI - 25x 4
43 . a. y=sinx +C b. y = - Z cos 2x +C
1 1 21. g'(x) = 2 sin - 1 x ~
c. y = tan 3x + C d. y = - cot 4x + C vl - X"2
3 4 1
23 . v ' = sin - x. The surprise is that you now have seen
e. y = 5 secx + C
a formula for the antiderivative of the inverse sine.
Problem Set 4-5 25. Radar Problem
Derivatives of Inverse TrigonometricFunctions a. tane = x / 100, so 0 = tan - 1 (x / 100), Q.E.D.
b de = 100 de 100 dx
ForProblems I through4, see Figure 4-5d. dx 10000 + x 2 ' dt 10000 + x 2 dt
1. Graph, y = cos - 1 x 3. Graph, y = csc 1x c. Truck is going 104 ft / sec "" 71 mph .
5. The principal branch of the inverse cotangent func- 27. Numerical Answer Check Problem
tion goes from O to rr so that the function will be numerical algebraic
continuous. X derivative derivative
7. sin(sin - 1 0.3) = 0.3 - 0.8 - 1.666671 .. . -1.666666 . . .
1
- 0.6 - 1.250000 . . . -1.25
9. y = sin - x =>siny = x =>cosy y ' = 1 => - 0.4 - 1.091089 ... -1.091089 ...
y , = -- 1 ~ 1 ,
Q.E.D. - 0.2 - 1.020620 ... - 1.020620 ...
cosy vl - X" 2 0 - 1.000000 ... -1
0.2 - 1.020620 ... - 1.020620 ...
0.4 - 1.091089 ... -1.091089 ...
X 0.6 -1.250000 ... -1.25
0.8 -1.666671 ... -1.666666 ...
y
''The precise value for the numerical derivative will
depend on the tolerance to which the grapher is set.
[Since sin y = (oppos ite leg)/ (hypotenuse), put x on 29. Genera l Derivative of the Inverse of a Function
the opposite leg and 1 on the hypotenuse. a. y = sin - 1 x => siny = x =>cosy y' = l =>
Adjacent leg = v'l - x 2 , and
cosy = (adjacent) / (hypotenuse).] y ' = -- 1 , Q.E.D.
cosy
Probl
emSet 4-5: DerivativesofInverse
Trigonome
tricFunctions 689
b I= 1 1 = 1.25 19. a. Graph. (example) b. (Equations will vary.)
Y cos(sin - 1 x) cos(sin- 1 0.6)
1 1 1 - QED
1. 2 ),
I l(x )
y = J 1 - x2 J 1 - 0.6 2 = 0.8 =
C. y = 1 - 1(X) ~ l(y) = X ~ f' (y) . d~ (y) = 1 ~
X
d 1 d - 1 1
dx (y) = j'(y) ~ dx (j (x)) = j'(j - l(x))'
Q.E.D.
d. l(x) = x 3 + x = 10
(x - 2)(x 2 + 2x + 5) = 0
x = 2 (only) . h(lO) = 2 .
2 1. Graph. Continuous
Since h(x) = 1- 1 (x) and f' (x) = 3x 2 + 1,
1 1 1 f(x )
h'(lO)
j'(h(lO)) 1'(2) 3-2 2 +1
1/13
Problem Set 4-6
X
Differentiabilityand Continuity 3
1. Continuous. 7. Both.
3. Neither. 9 . Neither.
5. Neither. 11. Continuous. 23. Graph. Both
13. a. Graph. (example) b. (Equations will vary.) f(x )
f(x )
25 . Graph. Neither
15. a. Graph. (example) b. (Equations will vary.)
f(x)
/
X X
6 3
17. a. Graph (example) b. (Equations will vary .) 27. Graph. Neither
f(x ) l(x )
/ -5
X
/
X
690 Append
ix B:Answers
toSelected
Prob
lems
29 . Graph. Neither The same thing happens from the right . As the
l(x)
following graph shows , the secant lines become
vertical as x approaches 2 from either sid e.
12
l(x) Secant
slope
becomes
infinite.
X 4
31. a= - 1.5 , b = 2.5 . Graph . X
f(x)
Thus f is not differentiabl e at x = 2, even though
X the right and left limits off' (x) are equal to each
other. The function must be continuous if it is to
have a chance of being diff erentiable.
39. Continuity Proof Problem
33. a= -0.5, b = 16. Graph . a. y = mx + b =:. y ' = m, which is independent of
x.
l(x) .. linear functions are differentiable for all x .
. linear functions are continuous for all x.
10
b. y = ax 2 + bx + c =:. y ' = 2ax + b, which exists
for all x by the closure axioms .
X
. . quadratic functions are differentiable for all x.
.. quadratic functions are continuous for all x .
c. y = 1/ x = x - 1 =:. y ' = - x - 2 , which exists for
35. Railr oad Curve Prob lem all x * 0 by closure and multiplicativ e inverse
axioms .
a. a= 1/ 30000, b = - 200 / 3
.. the reciprocal function is differentiable for all
Check: Graph shows that y is differentiable at
x = 100 with these values of a and b. X * 0.
. the reciprocal function is continuous for all
b. Rate of change of slope is (y ' ) ' , abbreviated y " . X c/= 0.
y" = 0 if x < 0 and y" = 6ax if x > 0. d. y = x =:. y ' = 1, which is independent of x.
.-. the ideritity function is differentiable for all x .
Both of these qu antities approach zero as x .'. the identity function is corrtinuous for all x.
approaches zero. Since y " = 6ax for x > 0, e. y = k =:> y ' = 0, which is independent of X.
the slope increases uniformly with x for positive constant functions are differ entiable for all x.
values of x, Q.E.D . constant functions are continuous for all x .
2x, if X < 2
37. j'(x)= 2x, ifx > 2
{
un defined, if x = 2 Problem Set 4-7
Taking the left and right limits gives Derivative of a Parametric Function
limx -2- f'(x) = 2 2 = 4 1. Parabo la Problem
limx--2+ f' (x) = 2 2 = 4 a. t X y
Using the definit ion of derivative, taking the limit -3 - 1 -6
from the left, -2 0 -1
- 1 1 2
0 2 3
f , (x) = lim x 2 + 1 - 4 1
x-2 - x-2 - O' 1 3 2
2 4 - 1
which is infinite. 3 5 -6
ProblemSet4-7: Derivativeof a Param
etric Fun
ction 691
b . Graph. 5. Circle Problem
a. Graph.
dy = -2t b. dy / dx = - cott
c. dx c. dy / dx = 0 if t = 0.5TT,l.5TT,2.5TT,.. .
If t = 1, dy / dx = -2 , and (x , y) = (3, 2). dy I dx is infinite if t = 0, TT, 2TT,. ..
Line through (3, 2) with slope -2 is tangent to At a point where dy / dx is infinit e, dx / dt
the graph. See part (b) must be 0. If dy / dt * 0, dy / dx is infinite. If
d. X = 2 + t ~ t = X - 2 ~ y = 3 - (X - 2) 2 dy I dt = 0, dy I dx is indeterminate, and could
This is the Cartesian equation of a parabola be infinite.
because only one of the variables is squared. x - 6 y-3
d. - - = cost and - - = sin t
e. By direct differentiation, dy / dx = - 2(x - 2). At 5 5
(x,y) = (3,2), dy / dx = -2(3 - 2) = -2 , which
agrees with part (c).
(x;6) y; 2+ ( 3 ) 2 = cos2 t + sin 2 t
dy / dx = - 2(x-2) = -2 (2 + t - 2) = -2 t, which
agrees with part (a). (x ;Gr + (y ; 3r =l
3. Ellipse Problem This is a standard form of the equation of a circle
a. Graph . x = 3 cost, y = 5 sin t centered at (6, 3) with radius 5.
e. The 6 and 3 added in the original equations are
the x- and y-coordinates of the center, respec-
tively.
The coefficients, 5, for cosine and sine in the
X
original equations are the x- and y-radii, respec -
tively. Since the x- and y-radii are equal, the
graph is a circle.
7. Deltoid Probl em
a. Grapher confirms figure in text .
b dy = 5 cost _
dx -3 sint b. dy = cost - cos 2t
C. If t = TT /4, (x, y) = (2.121 ... , 3.535 .. . ),
dx - sin t - sin 2 t
dy / dx = -5 / 3. c. Cusps occur where both dx / dt and dy / dt = 0.
Graph, part (a). The line is tangent to th e graph. Graphical solution shows that this occurs at
t = 0, t = 2TT/ 3, t = 4TT/3 , t = 2TT,....
d. False. The line from (0, 0) to (2.1 . . . , 3.5 ... ) does
not make an angle of 45 with the x-axis. [This
shows that the t in parametric functions is not dxl dt or dy/dt
the same as the e in polar coordinates.] _ dxl dt
I ' '\
e. Tangent line is horizontal if dy / dx = 0. I
.-.cost = 0 and sin t * 0.
This happens at t = TT / 2, 3TT/2 , . ... I
I I
Points are (0, 5), (0, - 5). dy/dt \, 1
Tangent line is vertical if dy / dx is infinit e .
.. sin t = 0 and cost * 0.
This happens at t = 0, TT,2TT,.... At t = 0, 2TT,... , the tangent appears to be hori-
Points are (3, 0), (-3, 0). See graph in part (a). zontal. At t = 2TT/3 , 4TT /3 , ... , there appears to
f. (x / 3) 2 +( y / 5) 2 = l,whichisastandardformof b e a tangent line but not horizontal.
the equation of an ellipse centered at the origin, A numerical solution shows the following values
with x-radius 3 and y-radius 5. as t approaches 2TT/ 3.
692 Appendix
B:Answers
toSelected
Problems
t dy / dx
2rr / 3 - 0.1 -1.547849 .. .
2rr / 3 - 0.01 -1.712222 . . .
2rr / 3 - 0.001 -1.730052 .. . X
2rr / 3 Indeterminate
2rr / 3 + 0.001 - 1.734052 .. .
2rr / 3 + 0.01 - 1.752225 .. .
2rr / 3 + 0.l - 1.951213 . . .
ii. Graph. (x = cos 6t, y = sin t)
dy / dx seems to be approaching about - 1. 732
as t approaches 2rr / 3.
The exact answer is -vG , which you can find with
l'Hospital's rule when you study Section 6-8.
X
9. Involute Problem /,
a. x = cos t + t sin t
y = sin t - t cos t
Grapher confirms figure in text.
e. See grapher graphs in part (d).
b. dy = tsint =tant f. Graph, n = 1. (x = cost , y = sin t)
dx t cost
c. At t = rr, dy I dx = tan rr = 0.
The string will be pointing straight up from the
x-axis. The diagram shows that the tangent to
X
the graph is horizontal at this point.
Graph, n = 2. (x = cos2t,y = sint)
X
1 y
11. Pendulum Project
Not selected
13. Lissajous Curves
If n = 1 the graph is a circle.
a. Grapher confirms figure in text. If n = 2 the graph is a parabola .
b. Graph. (x = cos 4t, y = sin t)
Problem Set 4-8
1 y Graphs and Derivativesof ImplicitRelations
3x 2 I 2x 5y 9 +1
X
1. y ' = - 28y 3 3. y = y 2 - 3x6y B
I = 2 COS 2x - 1 - y
5. y X +1
7_ y' = y o.s ; x o.s 9. y ' = - :~
c. If n is an even number, the graph comes to 1 - 15xl 4y 20 , cosx sinx
end points and retraces its path, making two 11 y' = -----=-- 13
. 1 + 2ox1 sy1 9 Y = cosysiny
complete cycles as t goes from Oto 2rr.
If n is an odd number, the graph does not come 15. y' = - y / x 17. y '= secy
to end points. It makes one complete cycle as t 19. y' = - siny tany
goes from Oto 2rr . / 1 1
d. i. Graph . (x = cos5t,y = sint) 2 1. Y = - siny
Problem
Set4-8:Grap
hs andDerivatives
ofImplicit
Relations 693
23. y = x 1115 ~ y 5 = x 11 ~ 5y.J y ' = llx 10 ~ Problem Set 4-9
, llx 10 llx 10 llx 10 11 Chapter Review and Test
y = 5y4 5(xll /5).J = 5x.J.J/s = Sx6 i5,
Review Problems
which is the answer obtained using the derivative RO. Not selected
of a power formula, Q.E.D.
Rl. a. = g(t ) = t 3 ~ g ' (t) = 3t 2
X
25. Circle Problem y = h(t) = cost ~ h '( t) = - sint
a. At (- 6,8), (-6) 2 + 8 2 = 100, which shows that If j(t) = g(t) h(t) = t 3 cost, then, for example ,
(-6, 8) is on the graph, Q.E.D. f' (1) = 0.7794 . . . by numerica l different iation.
g '( l ) h '( l ) = 3( 12 ). (-sinl) = -2.5244 . ...
b. dy / dx = - x /y. At (-6, 8), dy / dx = 0.75.
A line at (- 6, 8) with slope 0.75 is tangent to the :.j' (t) * g ' (t) h ' (t), Q.E.D.
graph, shmving that th e answer is reasonable. b. If j(t) = g(t ) / h(t) = t 3 / cost, then, for examp le,
f' (1) = 8.4 349 . .. by numerical differentiation.
g '( l) / h '( l) = 3(1 2)/(-sinl) = 3.5651 .. ..
:.j'(t) * g '( t) / h '( t), Q.E.D.
c. y = cost.
x = t 3 ~ t = x 113 ~ y = cos (x 113)
10
dy = - sin(xl /3) . 1:,x - 2/3
dx 3
dy . 1
At x = 1, dx = - sml
3 = - 0.280490 ...
dy cost If x = 1, then t = 1 113 = 1.
C -=---
. dx sin t dy / dt = - sint = - sinl = _
0 280490
dx / dt 3t 2 3 '
At x = -6, t = cos - 1(-0.6) which equals dy / dx , Q.E.D.
sin(cos - 1(- 0.6)) = 0.8
R2 . a. If y = uv, then y ' = u'v + uv' .
. dy - 0.6 . . . b. See proof of product formula in text.
dx = -c5:'B = 0.75, which agrees with part c. i. f' (x) = 7x 6 cos 3x - 3x 7 sin 3x
(b),Q.E.D. ii. g ' (x ) = cos x sin 2x + 2 sinx cos 2x
27. Cubic Circle Problem iii. h'(x) = 15(3x - 7) 4 (5x + 2) 2(8x - 5)
a. dy / dx = - x 2 /y 2 iv. s' (x) = 1000x 7 (Be careful!)
x=O:dy / dx = O d. j(x) = (3x + 8)(4x + 7)
The tangent is horizontal (see graph below). i. j'(x) = 3(4x + 7) + (3x + 8)(4) = 24x + 53
X = 2: dy / dx = - Q.2732 .. ..
ii. j(x) = 12x 2 + 53x + 56
The tangent lin e has a small negativ e slope, which
agre es with the graph.
f' (x) = 24x + 53, which checks.
x = 4: dy / dx is infinite . u'v - uv'
The tangent line is vertical. R3. a. Ify = u /v ,theny '=
V2
rn Y b . See proof of quotient formula in text.
~ C.l.
. f' ( ) _ lOx cos lOx - 5 sin lOx
X-
X
6
r-:-:-...~ ii .
8
' (x) = 18 (2x + 3) (5~ - 11)
10 g (9x - 5) 0
I\
iii. h(x) = - 1500x 2(10Qx 3 - 1) - 6
~ d . y = 1/ xlO
As a quotient:
b. y = x: dy / dx = - 1 Ox 10 - l-10x 9 -10
y' = ------ = - = -10x - 11
C. y = (64-x3)1 /3 x 20 xll
As x becomes infinite, (64 - x 3 ) 113 gets clos er to As a power:
(-x 3 ) 1i3 , which equals -x . The graph has a di- y = x - 10
agonal asymptote at y = -x , and dy / dx - - 1. y ' = - 1ox - u, which checks.
d. The name comes from analogy with th e equation c. t ' (x) = sec 2 x
of a circle, such as x 2 + y 2 = 64 . t '( l) = 3.4255 ...
69 4 Appendix B: Answe
rsto Selected Pro
blems
f. m(x) = t(x) - t(l) tanx - tan 1 t-window : [ - rr / 2, rr / 2]
x- 1 x-1
Graph.
y
rt/2
m (x )
3.42 ... X
X m(x) 1
y ' (O) = Jf=a2 1, which agrees with the
0.997 3.40959 . . .
0.998 3.4 1488 .. . graph.
0.99 9 3.420 19 . . .
1 no value y ' ( 1) = .Jf=TZ
l = l , w hi ch IS. 1"nfi mt. e.
1.001 3.43086 .. .
0
1.00 2 3.43622 . . . The graph becom es vertical as x approaches
1.00 3 3.44160 .. . 1 from the negat ive side. y' (2) is und efined
Th e values get closer to 3 .4 2 5 5 . . . as x ap - b ecaus e y(2) is not a real numb er .
proaches 1 from either side, Q.E.D.
R6. a. Differ entiability impli es continuity.
R4. a. i. y' = 7 sec 2 7x
b. i. Graph. (example) ii. Graph . (exampl e)
ii. y ' = -4x 3 csc 2 (x 4 )
iii . y ' = 3secxta nx / (x ) l(x)
iv. y' = -cscx cotx
b . See deri vatio n in text for tan ' x = sec 2 x.
c. Graph . The graph is always slopin g up ward, X X
which is connected to the fact that tan ' x equals C C
the square of a function, and is thu s always
positiv e.
iii. Graph. iv. Graph . (example)
y l(x) l(x)
No such
I
function.
I X
X
rt
d. f' ( t ) = 7 sec ttan t C. i. Graph .
j'(l) = 20 .17 ...
j ' (l. 5) = 1395 .44 ...
j'(l.57) = 110 38634. 0 ...
There is an asymptote in the secant graph at
t = rr / 2 = 1.57079 . . .. Ast gets closer to this
valu e, secant changes very rapidly! X
RS. a. 1.. y ' = 1 +39x 2
ii. f is cont inuous at x = 1 becau se right and
ii. ddx(secI x) =- ---=l==
left limit s both equal 2, which equal j(l).
lx 1Jx -l 2
... '( ) - _2cos - 1 x iii. f is differentiable. Left and right limits of
ill. C X - JT=x2 f' (x) are both equal to 2, and f is continuous
b . Graph, y = sin - 1 x , plott ed as at X = 2.
x = sint d. a= 1, b = 0
y =t Graph, diff erentiabl e and continuous at x = 0.
Prob
lemSet4-9:Chapter
Review
andTest 695
g(x)
CHAPTER 5
X
Exploratory Problem Set 5 - 1
A Definite Integral Problem
Oil Well Probl em
1. c(lOOO) ""$ 27.00 / ft
dy sint + tcost c(4000) ""$66 .39 / ft
R7. a - = ---- -
. dx cos t - t sin t 3. Average cost "" $43.86/ft
Where the graph crosses the positive x-axis, 5. Average cost using Problem 4 ""43 .74
t = 2TT,4TT,6TT,. . . Average of c( lOOO) and c(4000) is 46.69.
If t = 6TT, x = 6 and y = 0. The average of c(lOOO) and c(4 000 ) is significantly
..(6,0) ison the graph. high er than the actual average figur ed either way.
If t = 6TT, then dy / dx = 6TT.
So the graph is not vertical where it crosses the 7. The mathematical word for such a rate is th e second
x-axis. It has a slop e of 6TT = 18.84 .. . . derivativ e. The physic al quantity is acceleration .
b. Ferris Wheel Problem Problem Set 5-2
X = 20 Sin~ (t - 3) Review of Antiderivotives
TT 1
y=25+20cos (t -3) 1. j(x) =x7+C 3. j(x) =- 8 +C
10
TT
8x -
dx / dt = 2TTcos (t - 3) 5. j(x) = sinx + C 7. j(x) = -co tx + C
10
TT
dy / dt = - 2TTsin (t - 3) 9. j(x) = cscx + C
10
1
11. j(x) = -
When t = 0, dy / dt = 5.0832 .. . .
The Ferris wheel is going up at about 5 .1 ft / sec.
5 cos5x +C
When t = 0, dx / dt = 3.6931 . ... 1
13. f(x) =
The Ferris wheel is going right at abou t 3.?ft / sec. 8 tan Bx + C
dy dy / dt 1
15. f(x) = (4x + 5) 8 + C
dx dx / dt 32
dy / dx is first infinite at t = 8 sec. 17. a. Each is an ant iderivative of y' = 5x 4 .
RB. a. y' = 24x - 2 i3( 12x 1 13 + 7) 5 b. The word is congruent.
For any on e value of x, each pair of points on
, dy 4.5x 35 - y 4 cos(xy) y = x 5 + 0.3 andy 2 = x 5 + 0.7, for example, is
b . Y = dx = 3y 2 sin(xy) + xy 3 cos(xy) the same vertical distanc e apart, 0.4 . However,
c. Cissoid of Diod es Problem the graphs are not real ly parallel because the
. dy 3x 2 + y 2 perpendicular distance from one to ano ther is
I. -=--~-
dx By- 2xy not constant.
At (2, 2), dy / dx = 2. At (2, - 2), dy / dx = - 2. c. f(x) = x 3 - 2x + C
Lines at these point s with these slopes are For (1,0), C = 1
tangent to the grap h (see diagram). f( x ) = x 3 - 2x + 1 cont ains (1, 0 )
For(l ,l ),C=2
5 y f(x) = x 3 - 2x + 2 contains (1, 1)
For (1, 2), 2 = 13 - 2 1 + C ~ C = 3
j(x) = x 3 - 2x + 3 contains (1, 2)
X d. Graph.
,2
ii. At (0, 0), dy / dx has th e ind eterminate form X
0 / 0, which is consistent with the cusp.
iii. Asymptote is at x = 4.
696 Appendix
B:Answers
toSelected
Problem
e. The constant C affects the vertical position of
21. dy = (x - ) dx
th e graph without affecting eit her it s shape or
its horizont al position. The antid erivative is a 23. dy = - sin(secx) secx tan xdx
"family" of functions since there is mor e than
4
ju st one function, but th ey look so much th e 25. y = 5x + C
sam e. 1
27. y =-
4 cos4x+C
Problem Set 5-3
2 -
29. y = - 1) 1 + C
7 (0.5x
Linear Ap proxi mations and Differentia ls
1. y = 21.6x - 48.6 31. y = tan x + C
x = 3.1: Error= 0.11042
x = 3.001: Error= 0.0000108 . . . 33. y = 5x + C
x = 2.999 : Error= 0.0000107 .. . 35. y = 2x 3 + 5x 2 - 4x + C
3. Local Linearity Problem #1
y = 2x -1 37. y = sin 6 x + C
Graph shows zoom by factor of 10.
39. a. dy = 6(3x + 4)(2x - 5) 2 (5x - 1) dx
b . dy = - 60.48
c. 6y = -60.0218 .. .
d. - 60.48 is close to - 60.0218 ....
Problem Set 5 -4
graph tangent line
1 Formal Definition of Indefinite Integra l
: l,. -
1
Local linear ity des cribes the prop erty of the function 1. x 6 + C 3. - x 11 + C
11
because if you keep x close to 1 (in th e "locality" of
4 -
1) the curved grap h of the function looks like the 5. -- x - )+ C 7. 20t 5 1 + C
straight graph of the tang ent lin e. 5
5. Steepness of a Hill Problem 9. 50p 3/s + C 11. sinx + C
a. Let A b e th e number of radians in e degr ees . 1
13. - cos3m+C
3
15. ~ sin 7x +C
By trigonometry, tan A= ~ ~A = tan - 1 ~
1 0 1 0 1 1
(4v + 9) 3 + C
4
Since 1 radian is 180 / rr degrees, 17. 19. -- (8 - 5x) +C
12 20
180 _1 X
1
e = -;:-tan
100
, Q.E.D.
21. - (6 + 7b) - 3 + C
21
23. ~ sin 7 x +C
1.8 / rr
b. de = 1 + (x / 100) 2 dx 1
25. - -=-
::i
cos 5 e + C 27. _!_sin
4rr
4
rrx + C
X = 0: de = 0.5729 ... dx
1
X = 10: de= 0.5672 . . . dx 29 . -31 x 3 + -23 x 2 -::ix+C
-
31. - r - 1 + - r 3 + C
X = 20 : de= 0.5509 ... dx 3
c. The erro r is 0.1492 . .. which is about 1.3%. 0
,
1
33. 7
+ 3x 5 + 25x 3 + 125x + C
d. 0.5729 ... is approximately 0.5. So multiplying by 7x
it is approximately equivalent to dividing by 2.
35. tanx + C
For a 20% gra de this estimate gives 10, com-
pared to the actual angle of 11.309 . . . an error 0 1
,
37. -- csc3x + C
of about 11.6%. 3
For a 100% gra de this estimate gives 50, com - 1
par ed to th e actual angle of 45 , an error of about 39. tan 8 x + C
8
11.1%.
1
7. dy = 2lx 2 dx 9. dy = 28x 3(x 4 + 1) 6 dx 4 1. - - csc 9 x + C
9
11. dy = (6x + 5) dx 13. dy = 8.5x - 2 7 dx 43. Distanc e from Velocity Problem
2 xsec 2 xdx 10 312
15. dy = 3cos3xdx 17. dy = 3tan D(t) = 40 t + t
3
19. dy = (4 cosx - 4x sinx) dx D (lO) = 505.4092 ... "" 505 feet
Pro
blemSet5-4:Formal
Definition
ofIndefinite
Integral 697
45 . Integral of a Sum Properly d. The trapezoids are circumscribed around the re-
Prove that if f and g are functions that can be gion under the graph and thus contain more area
integrated, then f (j(x) + g(x)) dx = f j(x) dx + (see left diagram). For rectangles, the "triangular"
f g(x) dx. part of the region that is left out has more area
Proof than the "triangular" part that is included since
Let h(x) = f f(x) dx + f g(x) dx. the "triangles" hav e equal bases but unequal
By the derivative of a sum propert y, al ti tu des (see right diagram).
d d
h '(x)=- d ff(x)dx +- d fg (x)dx y y
X X Trapezo id Rectangle
By the definition of indefinite integral applied twice includes leaves out
to the right side of the equation,
h' (x) = j(x) + g(x).
By the definition of indefinite integral applied in the
other direction, X X
h(x ) = f (j(x) + g(x))dx.
By the transitive property, then, f (j(x) + g (x) )dx = 12. Exact Integral of Square Function by Brute Force
f j(x)dx + f g (x) dx, Q.E.D.
47. Introduction to Riemann Sums
a. ft x 2
dx
U100 = 9.13 545; L1 00 = 8.86545
a. Integral "' 50.75 Conjecture: Integral equals 9 exactly.
b. Integral "' 50.9375 b. The sample points will be at the right of each
c. As shown in Figures 5-4a and 5-4b, the Riemann interva l, 1 3/ n, 2 3/n, 3 3/n, ... , n 3/ n.
sum with 6 increments has smaller regions in- c. U,, = (3/n)( l 3/ n) 2 + (3 / n)(2 3/ n) 2 +
cluded above the graph and smaller regions (3/n.)(3 3/n) 2 + + (3/n)( n 3/ n) 2
excluded below the graph. So the Riemann sum d. U,, = (3/ n ) 3 (1 2 + 2 2 + 3 2 + . . . + n 2 )
should be closer to the integral. = (3 / n) 3 (n / 6)(n + 1)(2n + 1)
d . Conjecture : Exact value is 51. = (4.5/ n 2 )(n + 1)(2 n + 1)
e. By trapezoidal rule with n = 100, U 1oo = (4.4 / 100 2 ) (101)(201) = 9.13454, which
integral "' 51.00045, which agrees ,vith th e is correct.
conjecture. e. Using the formula, U 1ooo = 9.013504 . . . , which
f. The object went 51 ft. does seem to be approaching 9.
Average velocity= 17 ft / min. h . Un = 4 .5 . n + 1 . 2n + 1
n n
Problem Set 5-5 = 4.5(1 + l / n)(2 + 1/ n )
RiemannSums, and the Definitionof DefiniteIntegral As n approaches infinity, 1 /n approaches zero.
.'.U,, approaches 4.5(1 + 0)(2 + 0) = 9
1. R6 = 20 .9375
13. Not selected.
3. R s = 23 .97054 ...
5. Rs = 0.9 58045.,. Problem Set 5-6
7. L4 = 0.7 3879 . . . , U4 = 1.16866 .. . TheMean Value Theoremand Roi/e'sTheorem
M4 = 0.92270 .. . , T4 = 0.95373 .. . 1. See statement of mean value theorem in text.
.'.M4 and T4 are between L 4 and U 4 , Q.E.D.
3. Graph, g(x) = 6/x; [1,4]
9. Samp le Point Problem
a. Take sample points at x = 1, rr / 2, 2, 3, 4, and 6. g(x)
b. Take sample points at x = 0, 1, 3, 4, 3rr / 2, and 5. 6
c. U6 = 21.71134 ... , L5 = 14.53372 ...
11. Limit of Riemann Sums Problem 2
a. The program should give the values listed in text. X
C 4
b. Liao= 20.77545, Lsoo = 20.955018.
Ln seems to be approaching 21.
C = 2
c. U100 = 21.22545, Usoo = 21.045018.
Tangent at x = 2 paral lels the secant line.
U,, also seems to be approaching 21.
f is integrable on [ 1, 4] if L,.1 and U,, have the
same limit as n approaches infinity. 5. Graph, c( x ) = 2 + cosx; [ 0, iJ
698 Appendix
8:Answers
toSelected
Prob
lems
f(x )
c (x)
X
a d b
C rr/2
17. See Figure 5-6g.
C = 0.69010 ...
Tangent at x = 0.690 . .. parallels the secant line. 19. J(l) = - 3 * 0 .
Conclus ion is not true.
7. Graph, J(x) = xcosx j' (2) = 0, but 2 is not in th e interval (0, 1).
Graph.
f(x )
f(x )
X 0
C 1
-3
f' (x) = cosx - x sinx 21. J(2 ) = - 4 0. *
:.f is differentiable for all x. Conclusion is not true. f' (2) = 0, but 2 is not in the
f (O) = j(TT / 2) = 0 open interval (0, 2).
.-. hypQtheses are met on [O, rr / 2]. Graph .
C = 0.86033 ... f(x )
Horizontal line at x = 0.86033 . . . is tangent.
X
9. Graph, J(x) = (6x - x 2 ) 1 12 0 2
f(x )
23. f( 3) = -3 * 0.
Conclusion is true.
X
f' (2) = 0 and 2 is in the interval (0, 3) .
0 C 6
Graph.
f(x )
J'(x) = (1 / 2)(6x - x 2 ) - 112 (6 - 2x)
.J is differentiable on (0, 6). 0 3
"
f is continuous at x = 0 and x = 6.
f (O) = f (6) = O; interval is [O, 6]; c = 3
Horizontal line at x = 3 is tangent. -3
11. Compound Interest Problem
a. $74,357.52 Surprising! 25. J(O) does not exist.
b. Average rate ~ $1,467.15 per year Conclusion is not true.
c. d ' (0) ~ $86.18 per year f' (x) never equals 0.
d ' (50) ~ $6,407.96 per year Graph.
The average of these is $3,247.07 per year, which
f(x )
does not equal the average in part (b).
d. t ~ 32.893 . .. years
This time is not halfway between O and 50.
13. See Figure 5-6d. X
5
15. Graph. (example)
Problem
Set5-6:TheMean
Value
Theorem
andRolle's
Theorem 699
27. f is not differentiable at x = 3. Conclusion is not 33. Rolle' s Theorem Proof Illu strat ed by Graph and
tru e. Table
f ' (x) never equa ls 0. Graph . a. Grapher graph agrees with Figure 5-6k.
b. At x = 5 the cosine is at a high point and the
f(x )
parabola y = 25 - (x - 5) 2 is also at a high point.
j(5) = 29
c. j'(x) = -2x 2 + lOx - 8rrsin(2rr(x - 5)) ;
j'(5) = 0
X d. Graph, diff ere nce quotient Y 2 = m(x).
2 4
29. g is dis conlinuous at x = 2. Thus th e hypotheses of
the mean valu e th eorem are not met. The conclusion
is not true for [ l, 3] b ecaus e the tangent line would
ha ve to contain (2, g(2) ), as shown in th e left graph .
The conclusion is tru e for ( 1, 5) since the slope of
the secant line is 1, and g ' (x) = 1 at x = 3, which is
in the int erval (1, 5). See the right graph .
g (x ) g (x )
e. X m(x) X m(x)
2.0 3 5.5 - 16.5
X 2.5 5.7 6.0 - 1
3.0 2 6.5 - 6.833 . . .
3.5 6.8333 . . . 7.0 -2
-l.O 1 7.5 - 5.7
4 .5 16.5 8.0 -3
5.0 no value
3x - 3, if X ~ 3
31. a. j(x) ={ f. As shown in parts (d) and (e), the diff erence
X + 3' if X < 3
quotient is positive when x is less than 5 and
b. Graph .
negative wh en x is grea ter than 5. In the proof
f(x ) of Rolle's th eorem the left limit of the differen ce
quoti ent was shown to be positive or zero and
th e right limit was shovvn to be negative or zero .
The unm ent ion ed hypoth esis is differentiability
X
on the int erval (a , b). Function f is differentiab le
on any interval containin g x = 5. Since ther e is
a value off' ( 5), both the left and right limits
of the difference quoti en t must be equal. This
c. f is continuous at x = 3. The right and left limits nwnb er can only b e zero, which esta blish es th e
both equal 6.
conclusion of the th eorem. The conclusion of
d. f is not differentiable at x = 3. The left limit of Rolle's th eorem can be true even if th e hypothe -
f' (x) is 1 and right limit is 3. ses aren't met. For inst anc e, j(x) = 2 + cos x
e. f is not diff erentiabl e at x = 3, which is in ( 1, 6). has zero derivativ es every rr units of x, although
f. The secant line has slop e 11 / 5. The tang ent line j(x) is never equal to zero.
has slope either 1 or 3, and thus n ever 11/ 5. 35. Corollary of th e Mean Value Theorem
g. f is int egrabl e on [1, 6]. The int egra l equals 41.5, The hypothe ses of th e mean value theorem stat e
the sum of th e areas of the two trapezoids shown that f should be differentiable in th e open int erval
in the diagram below. (a, b ) , and continuous at x = a and x = b. If f
is differentiable in th e closed inter val [a, b], then
15 / (x )
it is automatically continuous at x = a and x = b
b ecaus e diff ere ntiabilit y impli es con tinuity.
37. A ntid erivativ e of Ze ro
By th e definition of antiderivative (indefinite inte -
X gral) , g(x) = f Odx if and only if g' (x) = 0. Any
other function f for which f' (x) = 0 diff ers from
700 Appendix
B:Answers
toSelected
Problem
s
g(x) by a constant. Thus the antiderivative of O is a
constant function, Q.E.D.
39. Max imum and Minimum Values of Continuou s
Functions The hypoth eses of Rolle's theorem say
that f is differentiable on the open interval (a, b ) . X
Since diff erentiability implies continuity, f is also 4 9
continuous on (a, b). Combining this fact with the
c. Pick sample point s at left ends of subinterva ls
hypoth es is of con tinui ty at a and at b allows you
for U5 and at right ends for L 5 .
to conclude that the function is continuous on the
Us = 3.80673199 ... , L 5 = 2.92710236 .. .
closed int erva l [ a , b] .
Average = 3.36691717 ...
41. Not selected . Average overestimates the integral, 3.33333 .. . .
This fact is cons istent with the fact that the graph
Problem Set 5-7 is concave up, and thus area above each lower
rec tangle is les s than half the differen ce b etween
Some VerySpecial RiemannSums
each upp er rectangle and lower rectangl e.
l. T3 = 4.646264 3 ... . d. Use sampl e points at the midpoints.
Since the curv e is concave downward, it lies above M w = 3.329 112 29 . . .
each trapezoid, and thus encloses mor e ar ea than MIOo= 3.3332 9093 .. .
the trapezoids. So the trapezoidal rule und eresti- M,ooo= 3.333332 90 .. .
mat es the actual area . See graph. Sums are converg ing toward 10/3 .
Trapezoid 3. See text statement of the fundamental theorem .
area
< graph 5. See the text proof of the fundamental theorem.
area.
7. Freeway Exit Prob lem
Distanc e= Jt(l oo- 20(t + 1) 1 12 ) dt = 453 feet
8. The Fundamenta l Theorem A noth er Way
3. R 3 = 4 .66666667 (Remarkable!)
a. M10 = 12.66753 . . .
Answer is between the answ er to Problem 1 (which is
a low er bound) and the answer to Problem 2 (which b. h (u )6u and h (u + 6u )6u are terms in a lower
is an upp er bound) . and an upper sum, res p ectively, because h (x ) is
increasing.
5. C "=' 1.24580513 .h (u)6u < A(u + 6 u) - A(u) < h (u + 6u)6u
This is th e sample point u sed in Problem 4. A(u + 6u ) - A(u) h( A )
C. h ( U ) <-------< U + u.U
7. Conjecture: Exact area = 4 ~ 6u
But th e limits of h (u) and h(u + 6u ) both
9. T10 0 = 4 .0004 equal h (u) since hi s continuous and h (u ) is
Conjecture : Integral = 4, exactly. indep end ent of 6u . Therefore, by th e squeeze
theorem,
11. C1 = 0.31498 026 . . .
C2 = 0.77680912 . . . lim A(u + 6u ) - A(u) _ h( )
cw- o 6u - u.
C3 = 1.266449 25 .. .
C-1 = 1. 76182468 .. . But the limit on th e lef t is defined to be dA / dil .
Each value of c is within the respective int erva l. :.dA / du = h(u ) , Q.E.D.
13. Conjecture: Area = g ( 2 ) - g (O) d. dA = h (u)du
C = 1.25992 ...
A(u) = (2 / 3) u 3 12 - 16/ 3.
j(c) = 2 (exactly) e. A(9) = 12 ~ which agrees with M10 = 12.667 ... .
.-.R1 = (2) (2) = 4, th e exact answer.
9. a. Graph. (example) b. Graph. (example)
Problem Set 5-8
/(x)
The FundamentalTheoremof Calculus
1. a. I = ft
1ox-1.sdx
= (- 10/ 0.5)(9 - 05 ) - (- 10 / 0.5)(4 - 0 5 )
= - 20 / 3 + 20 / 2 = 10/3 = 3.3333 3 . . .
X
The + C and -C add up to zero. X
b . Graph, upp er sum, n = 5.
Problem
Set5-8:TheFundamental
Theorem
ofCalculus 701
c. Graph. (example) d. Graph. (exampl e) the g grap h is grea ter than the area und er th e f
grap h, but th e g graph touches or crosses the f
/(x )
graph somewher e in [a, b]. One count erexampl e is
j(x) = 1.5 andg (x) = 2 + cosx on [0,2rr]. Graph .
3 y
g
X X
e. Graph . (example) f. Graph. (example) 0 2n
/(x) /(x )
Problem Set 5 10
A Way to Apply Definite Integrals
X X
1. Displacement Probl em
a. Graph, showing a strip and a sample point ( t, v ).
Problem Set 5 -9 100
Definite Integral Properties and Practice
1. 21 3. 125 50
5. 1116 7. 30 di
2
9. 103 1 I= a I= b
13. 8
b. dy = V d t = (55 + 12t 0 6 ) dt
15. (7/6)-/3 - 1/ 2 = 1.52072 ... C. R = I dy = I(55 + 12t 0 6 ) dt
29 d. Displacement = J: (55 + 12t 06 ) dt
17. 363 8835
e. 1st hour: Disp. = If (55 + 12t 06 ) dt = 62.5 mi
2
4 4
19. (sin 2 - sin l ) = 0.045566 ... 2nd hour: Disp. = J1 (55 + 12t 06 ) dt "' 70.2 mi.
lst2hr.:Disp. = f5(55+ 12t 06 )dt = 132.7 35 ... ,
21. i(sin 0.6- sin0. 3) = 0.0897074 ... which equal s the sum of the two integrals above.
f. v (2) "' 73.2 mph.
23. 20 g. Approx. 4.1 34 hour s.
25. no value
3. Work Problem
27. Integral = -( area) 29. Int egral * area a. Graph, F = 0.6x, showin g strip and sample point.
/(x )
/(x)
6 X 6 F
(x , y)
7 X
-5
0
dx - I
'x
0
31. I: j(x) dx = - 7 b. dW = Fdx = 0.6xdx
c. W = 24.3 inch pounds
33. f~g(x) dx = 25
d. F (4.5) = 2.7 pounds an d F(9) = 5.4 pounds.
35. f~j(x) dx + f~g (x ) dx cannot be determined. So the force at x = 9 is twice the force at x = 4.5.
37. Converse: Work done in stretc hin g from O to 4 .5 inch is
"If I:
j (x) dx <
all x in [a, b]."
I:
g(x) dx, then j(x ) < g(x) for 6.075 inch pounds . So the work don e str etchin g
from O to 9 inches is four times the work done
The converse can be shown to be false by any coun- stre tching from O to 4.5 inches, not twice the
terexampl e in which the area of the region und er work.
70 2 Appendix
B:Answers
toSelected
Problems
5. Heat Problem 5. Sine-Integral Function Problem
3 2
a. Graph, C = - 0.016T + 0.678T + 7.45T + 796, a. Graph, y = Six
showing strip and sample point (T, C).
Six
2
C
-20 20
dT
T -2
10 30
b. (sinx) / (x) approaches 1 as x approaches 0.
b . dH = (-0 .016T 3 + 0.678T 2 + 7.45T + 796) dT
c. Answers will vary depending on the grapher. The
c. H = I;i( -0 .016T 3 + 0.678T 2 + 7.45T + 796) dT TI-82 gives Si0.6 = 0.58812881 using TRACEor
= 21,576 Btu 0.588128809608 using TABLE,both of which are
d. 43,152,000 Btu correct to as many decimal places as NBS.
Integral of a constant times a function . d. By TABLE,Six seems to be oscillating between
e. The mathematical model does not give reason- about 1.53 and 1.61 when xis between 20 and
able answers beyond 300Q (T = 30). As stated 30. The limit is somewhere between these two
in the text, the actual heat capacity rises slowly. numbers, say about 1.57. The actual limit is TT/2 ,
The model indicates that it decreases rapidly. See which equa ls 1.570796 ....
graph .
e. Graph, j(x) = (sinx) / (x), superimposed on
C
y = Six. The f graph is positive, and greatest
when xis between - TT and TT, which agrees with
1000 the large positive slope of the Six graph in this
End of . region. Each place the Six graph has a high or low
domain:
point the j(x) graph has a zero, corresponding
T to the zero slope of the Six graph .
10 30
Si X
2
7. Golf Course Problem
Using trapezoids, area ""3475 ft 2 . X
The fundamental theorem cannot be used since - 20 20
the function is specified only by data, not by an
equation whose antiderivative can be found.
Problem Se t 5 - 11
4
Numerical Integrationby Simpson's Ruleand Grapher 7. f~.3 COS X dx "= 0.689929523 . . .
1. Velocity Prob lem 9. f52x dx "" 10.0988965 ...
a. Distance "" 3.444 . . . "" 3.4 nautical miles
11. Answer Check Problem
b. T6 = 3.7333 ... "" 3.7 nautical miles sin 1.4 - sin0.3 = 0.689929523 ... ,
c. The answer by Simpson's rule should be closer which agrees with Problem 7.
since the graph is represented by curved seg-
ments instead of straight ones. 13. Program for Simpson's Rule from Data
Not selected.
3. Tensile Strength Test
a. Graph, scatter plot of the data. 15. Relat ive Accuracy Problem
Pounds a. I = forrsinx dx
.. i. Sso = 2.00000017 .. .
300 .... ii. Tso = 1.99934198 .. .
I =2
iii. INTEGRATE:
Inches iv. I= -c osx l0 = -(-1) + 1 = 2
0.5 b. The built-in integrate function is closest, followed
b. Work "" 132.8333 . . . "" 132 .8 inch pounds by Simpson's rule, then trapezoidal rule.
ProblemSet5-11: Numer
icalIntegration
bySimpson's
Ru
leandGrapher 703
Problem Set 5 - 12 Graphs of M6 and T6 , respectively .
Chapter Review and Test Y
Review Problems
RO. Not selec ted
Rl. Heat Capacity Problem
a. Hea t needed is (Btu/degree)(degrees), which is a
product of independent and dep endent variabl es.
Since (Btu/ degree) varies, a definit e integral must
WW )
0.2
X
1.4
e. i. Un = (1 5/ n )(S / n) + (2 5/ n)(S / n)
+ (3 5/n)(S / n) + + (n 5/ n)(S / n)
b e used . = (25/ n 2 )( 1 + 2 + 3 + + n)
b. C = - 0.4 + 0. 3 log T ii. Un = (25/n 2 )(0 .5n)(n + 1) = 12.5(1 + 1/ n)
By trapezoidal rule with n = 100, th e amount of iii . lim n-co Un = 12. 5 b ecaus e 1 / n - 0 as n - oo.
h eat added from T = 1000 to T = 5000 is about iv. Int egral = area of triangle with bas e 5 and
2527 Btu . (Exact answer: 2527.3016 ... ) altitude 5, which is (1 / 2)(5)(5) = 12.5.
c. Average :::::0.632 Btu / degree R6. a. The hypoth esis is the "if" part of a theorem and
R2. a. j(x) = 2.8x 10 17 + C the conclusion is th e "then" part. (Hypo- means
b . j(x) = 5 sin2x + C under, and -thesis m eans th eme .)
1 b. Average velocity = 1.5 m / sec
c. J(x) = - (12x + s) - 2 + c
24 Instant aneo us velocity= d '( t) = 0.75 rr cos ~t
R3. a. Linear function is y = - rrx + rr Instantan eous velocity equals average velocity at
At x = 1.1, Error= - 0.005142 .. . time t = 1.12132 . .. :::::1.12 sec
At x = 1.001, Error= - 0.000000005 ... c. Int erval is [O, 4], c = 1.
b. i. dy = - 10 csc 5 2x cot 2x dx At x = 0, g ' (0) takes the form 1/ 0 which is
ii . dy = (x 4 + x - 4 ) dx infinit e.
iii. dy = - 12 (7 - 3x) 3 dx Thus, g is not differ en tiabl e at x = 0. However,
c. i. y = sec x + C the function need not be diff erentiabl e at th e end
points of the interval , just at int erior point s.
ii. y = 118(3x + 7)6 + C d. See Figure 5-6c, left diagram.
iii . y = Sx + C e. See Figure 5-6c, right diagram.
d. i. dy = (2x + s) - 1 12 dx f. See text derivation of Rolle's theorem.
ii . dy = 0.06 g. Graph . Points are 1/8 , 1/4, 3/8 , 1/ 2, 5/ 8, 3/4,
iii . ~ y = 0.059644 ... 7/ 8.
iv. 0.06 is close to 0.059644 . . ..
R4. a. See te xt definition of indefinite integral.
b . i. 7.2x 5 13 + C
ii. i 7
sin x + C
1 3 - 4x 2 + 3x + C
iii.
3x
RS. a. See text definition of int egra bilit y. h . If r ' (x) = s' (x) for all x in an interva l, then
b . See text definition of definite integral. r (x) = s(x) + C for som e constant C.
C. i. 1/5 = 2.845332 .. . R7. C1 = l.513915927 .. .
ii. L5 = 1.872703 .. . C2 = 2.508338988 . . .
C3 = 3.505954424 .. .
iii . M5 = 2.209073 .. .
R 3 = (1.5 31. .. )1. 5 + (2.508 ... )1.5 + (3.505 . . . )1.5 =
iv. T5 = 2.359108 .. .
12.4, which is the exact value of the int egra l.
d . Graphs of U6 and L6 respectively .
5
Sr u r J
RB. a. Integral = 92 / 3 = 30.666 ...
b. T100 = 30.6656, which is clos e to 92 / 3.
c. M10 = 30.72
~~
M10 0 = 30.6672
M1 000 = 30.666672
These Riemann sums are approaching 92/ 3.
704 Appendix
B:Answe
rsto Selected
Prob
lems
R9. a. i. 4/ 5 CHAPTER 6
ii. (1 / 12)( 19) 6 - (1 / 12)(12) 6 = 3671658.08 .. .
Problem Set 6 - 1
iii. 2 - 5rr
Integralof the Reciprocal Function
, a Population Growth
b. Graph, y = sinx - 5.
Integral is negative since each y-value in the Problem
Riemann sum is negative. 1. The annual rate of population growth, dP I dt, is 5%
y (0.05) times the current population P.
X r10
3. JO 0.05 dt = 0.5
Using the power function method on Jp - l dP gives
(1 / 0)P 0 , which involves division by zero .
5. From the answer to Problem 4 you can tell that
the value of N which makes the integral equal 0.5
c. Integral = 80 is somewhere b etween 1500 and 2000 . Setting the
numerical int egral between t = 1000 and t = N
d . Graph . Total area = sum of two areas. equal to 0.5 and doing a numerical solution for N
gives
N "" 1648.72 ... , or about 1649 people.
7. The P versus t graph is non-linear. From t = 0 to
t = 10 the population increased by 649 people .
X
From t = 10 to t = 20 it increased by 1069 people.
a C b
Problem Set 6-2
RlO . Displacement Prob lem Antiderivotiveof the ReciprocalFunction
a. Graph, showing strip and sample point (t, v) .
1. Each value of x uniqu ely determines the area ; as x
1v
(r increases, so does the area. Note that the area does
not depend at all on t.
P6
b . dy = V dt = 150t OSdt
3. Graph, showing that each M is negat ive and each
value of 1 / t is positive, so that the Riemann sums
are negative, and thus the integral is negative.
1/ 1
C. R = I
150t 0 5 dt
d. limn-ooR = f5
l 50t 0.s dt = 2 700 ft., a definite
integral.
e. The 2700 in part (d) is the displacement, 2700 ft. ~ Negative
For [0,4], y = 800.
For [4, 9], y = 1900. 5. a. g(3) + g(2) = 1.098617 . .. + 0.693142 . . .
Thus 2700 = 800 + 1900, Q.E.D. 1.791759 . . . , which equals g(6).
Rll . a. See text graph, Figure 5-lld. b. g(8) - g (2) = 2.079441 . . . - 0.693147 . ..
Each parabola requires three points, thus taking 1.386294 . .. , which equals g(4) .
two subintervals. Therefore, Simpson's Rule uses c. 3 g(2) = 3 0.693147 ... = 2.079441 . . . , which
an even number of increments. equals g(8).
b. See graphs in Figures 5-lla and 5-llb . d. - g(2 ) = - 0.69314 . . . = g(0 .5) = g(l/2).
c. Displacement = 67.6 m 7. Graphs, y = g(x) and y = logx.
d. Integral "" 6.0913
Graph. By counting squares, integral "" 6.
v(I)
1 X
c- I
/1 10
Prob
lemSet 6-2:An
tiderivot
iveof theReciprocal
Function 705
Both graphs have asymptotes at x = 0, x-intercepts b. f dcm d ' part (c)
of 1, slopes tending to zero as x approaches infinity. 53 0 0.1707
Both are increasing and concave down . 60 1.1227 0.1508
70 2.5197 0.1 292
9. g ' (3) ~ g (4 ) - g (2) ~ 1.386 29 - 0.69 314
80 3.7265 0.11 3 1
2 2
~ 0.3466 100 5.7461 0.090 5
By numeri cal differentiation, 120 7.3962 0.07 54
(d / dx )( lnx ) ~ 0.33333 , 140 8.7914 0.0646
which is close to 0.3466. 160 10.0 0.0 565
Th e measured distances are close to the calcu-
11. Solving num erically for g (b ) = l gives lated distances.
b ~ 2.71828 .. .
c. d ' (j ) = b If = 9.0507 / f. See table in part (b).
d. d ' (j) is in cm / 10 kH z.
Problem Set 6-3
e. d ' (j ) decreases as f gets larger; this is consis-
Natural Logarithms
, and Another Formof the Fundamental
tent with the spaces between the numbers getting
Theorem small er as f increases.
l. y ' = l / x 3. j' (x ) = 5/ x 61. Not selected .
5. h '( x ) = - 12 / x 7. r '( t) = 3/ t
Problem Set 6 -4
ln 24x 2
9. y ' = (1 / x )(ln4x + ln6x) or : - - - In x Reallyis a LogarithmicFunction
x
1. ln6 + ln4 = 1.79175 . .. + l.38629 ...
, _ ln 3x - ln llx . ln (3/ 11)
11 y - x (ln 3x )2 or. x (ln 3x ) 2 = 3.1780 5 ...
ln24 = 3.17805 . .. (checks)
13. p ' = (cosx ) (lnx ) + (sinx) (l / x )
3. ln 2001 - In 667 = 7.60140 ... - 6.50279 . ..
15. y ' = - sin(lnx ) (1/ x ) = 1.09861 ...
l 7. y ' = - tanx (Surprise!) ln (2001 / 667 ) = ln 3 = 1.09861 . .. (checks)
19. T '( x ) = sec 2 (lnx ) . (1/ x ) 5. 3 ln 1776 = 3( 7.48211 ... ) = 22.44635 .. .
ln (1776 3 ) = ln 5601816 576 = 22.44635 ... (checks)
21. y ' = -3(3 x + 5) - 2 23 . y ' = 4x 3 ln 3x + x 3 7. See text proof of the uniqueness theorem .
25. y' = -l / x 27. 71n lx l + C 9. Prove that ln (a / b ) = lna - lnb for all a > 0, b > 0.
1 1
29. ln lx l + C 31. ln lx 3 +5l+ C Proof:
3 3 Let J (x ) = ln (x / b ), g (x ) = lnx - lnb for x, b > 0.
1 Then j' (x ) = (b / x)(l / b ) = l / x, and g'(x) =
33. - x 6 l+ C
6 ln l9 - 35. ln 11 + sec x i+ C
(1 / x ) - 0 = 1/ x .
37. 1n I sin x i+ C 39. ln8 = 2.079441 . .. ..j' (x ) = g ' (x ) for all x > 0.
J (b ) = ln (b / b ) = lnl = 0.
41. ln 30 = 3.401197 .. . g (b ) = lnb - lnb = 0 .
2 ..J ( b ) = g (b ).
43 . 0n28 - ln9 ) = 0.756653 . ..
3 ..f (x ) = g (x) for all x > 0 by the uniq u eness
theorem.
1
45. (ln x) 6 + C 47. f ' (x) = cos 3x .. ln (x / b ) = lnx - lnb for all x > 0.
6 :. ln (a / b ) = lna-lnb for all a > 0 and b > 0, Q.E.D.
2
49. tan 3 X 51. f '(x) = 2x 3x
11. Prove that ln (a / b ) = lna - lnb for all a > 0, b > 0.
53. h '( x ) = 3.}l + (3x - 5)2
Proof:
55. Fundamental Th.eorem: 5 ln 3 - 5 ln 1 = 5.493061 .. . ln (a / b ) = ln (a b- 1 ) = lna + lnb - 1
Midpoint Riemann sum: M t0o = 5.492 987 ... = ln a + ( - 1) In b = ln a - ln b.
Trape zoid al rul e: T 100 = 5.-193209 . . . . ln (a / b ) = lna - lnb,Q.E.D.
Numerical integration: 5.493 06 1 .. .
13. Base of Natura l Logarithms Problem
57. Population Probl em a. y = log b x and x = bY are equivalent by the
Ito
00 (1 I P ) dP = lnN - ln 1000 . definition of logarithm.
N"" 1649 peopl e. b. x would equal b 1 , which equals b .
59. Radio Dial Derivative Problem c. When y = l, x = b ~ 2.7183.
a. a= -3 5.934084 ... , b = 9.0 507041 ... d . This number is called "e."
706 Append
ixB:Answers
toSelec
ted Pr?blems
Problem Set 6-5 m '( t) _ 1000(1.06)t(lnl.06) - l
c. m(t) - 1000(1.06 ) 1 - nl. 06
Derivativesof ExponentialFunctions-LogarithmicDifferentiation
..m ' (t) / m(t) = In 1.06, a constant.
l. f' (x) = o.4 2x 2 ln 0.4
d. m (l ) = 1060.00. So you earn $60.00.
3. g ' (X) = 4(7 X) ]n 7 The rat e starts out at only 58.27 $/year, butt
5. c'(x) = x 5 3x(5/x + ln 3) incr ease d enough by year's end to make the tc
for th e year equal $60.00 .
7. y ' = (cosx )0 -7x(0 .7ln (cosx) - 0.7x tanx )
2x [2In(csc5x)-10xcot5x]
39 . Limit of an Interesting Express ion:
9. y' = (csc5x) Using TRACEor TABLE ,
11. j'(t) = t sect(secttantlnt + (1/t) sect) n (l + l / n ) 11
4 10 2.59374246 01 . . .
13. v ' = (x 4 - l )x [1n(x-! - 1) +
4
x
(x-l - 1)
J 100 2.7048138 294 .. .
1000 2.7169239322 .. .
15. y' = 2x1nx[ln2 + 1/(x lnx )] 10000 2.71 81459268 .. .
17. y' = 5(3x - 4 )X[ln (3x - 4 ) + 3x/(3 x - 4)] 100000 2.7182682371 .. .
1000000 2. 718 28 0469 3 .. .
19. a. y ' = 7/ x
b. y' = 7 / x (checks)
The valu es are increasing toward e, the base ol
natural logarithms.
, 12x - 19
2 1. a. Y = (3x + 4)(2x - 9)
Problem Set 6-6
b. , = _3_ + _2_ = 12x-19
y 3x + 4 2x - 9 (3x + 4)(2x - 9)
The Number e and the Derivativeof Bose b Logarithms
54 1. f'(x) = l /( xln3), j'(5) = 0.182047 ...
23 . a. '= - Graph, showing tan gent with small positive slop
Y (5x + 2)(7x - 8)
b , = _5_ __ 7_ = - 54
/(x)
Y 5x +2 7x-8 (5x+2)( 7x - 8)
25. From Problem 19, y' = 7 /x
At X = 2, y ' = 7 / 2 = 3.5. X
ln 3(2 .001 ) 7 - ln 3(2) 7
O.OOl = 3.499 12 . .. , close to 3.5.
27 I= ( 35 +~ )
y Y 5x + 11 7x - 3
3. f'(x) = 1/( xln0.6 ), j'(9 ) = -0.2 17 512 ...
29
Y
I =
Y
(---=IQ_
3 - 4x
+~
7 + 5x
) Graph, showing tan gent with sma ll negative slo1
31. y ' = y[4 /( 4x + 1) - 6co tx]
l(x)
33 . -5ln3 = -5.493 061. . .
X
35 . Continued Exponentiation Problem 9
a. j'(x) = xx(lnx + 1)
b. g ' (x) = x xx xx[(lnx) 2 + lnx + 1/ x]
37 . Compound Int erest Problem
a. m ' ( t ) = 1000 (1.06 )t (ln 1.06)
m '( O) = 58 .27 $/year 5. f'(x) = 13/x
m '(5) = 77.98 $/ye ar
m '( lO ) = 104.35 $/year. 7. g '( x ) =4 0 / x
b. m (O) = $1000 .00 9. T'(x) = (cotx) /( ln 5)
m (5) = $1338.23
m (lO ) = $1790.8 5 11. p '( x) = (2 ln x) / (xln5)
The rat es are increasi ng. $338 .23 is earned be- 13. y' = 1/(4 ln l0) = 0.108573620 . . .
tween O and 5 yea rs, $452 .62 is earned between
Num erical deri va tive is 0.1085736 .. . (checks)
5 and 10 years, which agrees with the increasing
derivatives shown in part (a). 15. j'(x) = 1/( xln0 .9);j'(2) = -4 .7456 ... ;graph :
Problem
Set6-6:TheNumber
eandtheDerivative
ofBose
b Logarithms
f (x)
11. u ' = 0
13. y ' = 2e 2" ln 3u + e 2u (1/ u )
X
, = exp x 1nx - exp x ( 1 / x )
15 y (ln x )2
Tangent!
17. y ' = 4esccx sec x tan x
Not selected. 19. j' (x) = - csce x co t ex ex
Hour s/Mil e Lava Flow Prob lem 21. y ' = 6
a. y = 7 (2 - 0 .9x ) 23. y ' = 24x
dy / dX = 7( - Q.9X)(ln Q.9)
d y / dX = 0.73 7523 .. . (Q.9X) 25. g ' (x ) = 12
x = 0: dy / dx = 0.73 7 . . . mph 27. y ' = 3001 5e 15x
x = 1: dy / dx = 0.66 3 . .. mph
x = 5: dy / dx = 0.435 .. . mph 29. y ' = ex - e- x
x = 10: dy / dx = 0.257 . . . mph 31. u ' = 14e 2l(5 + e 2t)6
Lava is slowing down. (Phew!)
b. Y / 7 = 2 - 0 .9x 33. y ' = 15x 2 exp (5x 3 )
0.9x=2-y /7 35. y ' = 0
xln0 .9 = ln (2 - y / 7)
X = (1/l n0 .9)[ ln (2 - y / 7)]
37. j( x ) = e OAx => j'(x) = 0 .4e 0Ax
dx 9.49 122 1. . . Algebrai cally: f' (2) = 0.4e 0 8 = 0.8902 16 . . .
C. - = -- -- - Numeri cally: f' (2) "' 0 .8902 16 . . . (ch ecks)
dy 14 - y
y = 10: d x / dy = 2.372 . . . hour s / mil e 39 . f' (x) = se x (1 + x )
d. If x = 10, th en dx / d y = 3.88865 1. . . J'( - 1) = 0, J (- 1) = - 1.839 .. .
Graph : Line at x = - 1 is tangent to th e gr aph .
e. 3.8 8 .. . is th e rec iproca l of 0.2 5 7 ... , th e valu e of
dy / dx when x = 10, not when y = I O. t(x)
Proof of the Change of Base Pr operty
a. Definiti on of logari thm (algebr aic). X
b . Take logb of both sid es . -1
c. Log of a p ower pro perty.
d. Divid e by logb a . Tangent !
e. Substitution .
Not selected . 1
41. - esx + C 4 3. 6ex p x + C
Limit and Function Int er chan ge Journ al Problem 5
A typical journal entr y should include th e fact th at 47. esinx + C
4 5. _ie -2x+ c
if th e out sid e fun ction is continuou s and th e in- 2
sid e fun ction h as a limit as x approa ches c, then 1 4 1
limf (g (x )) = f( lim g (x)). A simple exampl e is 49 . - x + C 51. - (1 + e2x) s1 + c
4 102
lim x-c f (x) = J( lim x-c x ) = j (c ) , which is th e
definition of continui ty. Th e pro p erty is u se d to re- 1
53. 9x + 6e x + - e 2x + C
ver se the log and th e limit in findin g th e derivative 2
of th e logarithm fun ction algebr aically. 55. a. 2.5(e 08 - e 04 ) = 1.834 290 ...
blem Set 6-7 b. Num erically, int egral "' 1.834290 ... (ch ecks)
Natural ExponentialFunction, the Inverseof In 57. a. e 2 + e- 2 - 2 = 5.52439 1 .. .
y ' = 4e 4X b . Num er ically, int egral "' 5.52439 1 .. . (ch ecks)
y ' = - 85e - Sx 59. Rabbit Popul ati on Pr oblem
a. R (t) = 60000 el. 844 .t
j'( x ) = - e-x
b . R (5) "' 607 million rabbit s.
h '(x) =x 2ex(3 +x ) c. t = -5.589 .. . . So th e first pair of rabbit s was
r' (t ) = et sin t + e 1 cost introdu ced about 5.6 year s earli er, or in 1859.
Appendi
x B: Answe
rstoSelectedProb
lems
61. An Exponential Function Is Not a Power Function' f(x)
Counter example: Let f (x ) = ex.
Using th e derivati ve of a power formula, f' (x) - 1 ::;
would equal x ex- 1 . Then f' (0) would equal X
O e 0 = 0. But th e graph of j (x ) = ex cross es -1 3 5
the y -axis with a slope of e 0 = 1. So the derivativ e
of a pow er formula produ ces a wrong answer.
63. Zero/ Zero Problem
. 1n l + sinO 0 33 . Zero to the Zero Problem
a. lrmx-J j (x ) - - - , Q.E.D. L = lim x-o + xk f (lnx ) - 00
1 -e 0 0
b . Graph. lnL = limx-o +[k / (ln x) ln x] = limx-o+ k = k.
:.L = ek
f(x) Graph. Th e gr aph turn s out to b e a h ori zontal line
y = ek defined for x > 0.
X
y
Y= e '
-2 /
X
c. j(x ) appears to approach - 2.
d. d~ (lnx + sin (x - 1)) = 1/x + cos (x - 1);
By the definition of a power,
At x = 1, 1/ 1 + cos (l - 1) = l + cosO = 2. x k/(lnx) = (x k) I / lnx = (ek lnx) 1/ lnx = ek
d~ (1 - ex- 1) = - ex- 1; -e 1- 1 =-ea= -1 35. Continuou s Compounding of Int erest Problem
Ratio = 2 / (- 1) = - 2, which equals the apparent a. For yearly compoundin g, m(t ) = 1000 (1 + 0.06 )t
limit! For semi-annual compounding, m ( t ) = 1000 (1 +
0.0 6/ 2) 2 1 b ecau se th ere ar e two compoundin g
Problem Set 6-8 p eriod s p er year, each of which gets half th e
int eres t rat e.
Limitsof IndeterminateForms- l'Hospital'sRule
b. m(t ) = 1000 (1 + 0.0 6/ n )n'
1. Limit = 10/3 . Graph . limn-oo m (t) = 1000 e 0 -06 t
When int erest is compounded continuously,
y m ( t ) = 1000 e 006 t.
c. t m ( t ), annual m ( t ), cont. diff erence
5 1,338.23 1,349.86 11.63
20 3,207.14 3,320.12 112.98
X
50 18,4 20.15 20,085.54 1,665 .38
d. m (t) = 1000e 0071 .
3 7. Not selected.
3. Limit = l 5. Limit = 1/ 2 Problem Set 6 -9
7. Limit = oo 9. Limit = 0 Derivativeand Integral Practice for TranscendentalFunctions
11. Limit = e/5 13. Limit = - 26.4 3297 .. .
1. y' = 3/(3 x + 4 ) 3. y' = 3
15. Limit = 00 17. Limit = 3/ 4
19. Limit = 1/ 4 21. Limit = 1 5. y' = -5 tan x 7. y' = - tan (tanx) sec 2 x
23. Limit = 1 25. Limit = 1 9. y' = -( 1/x) sin (ln x) 11. y' = 7e7x
27. Limit = e 3 = 20.08 5 ... 29. Limit = 1/ 2 13. y' = 5x 4 15. y' = -ex sin ex
3 1. Infin ity Minu s Infinity Problem
IT IT
) = sec 2 x - tan 2
j(x
2 2x
Graph. Wher e secant and tang ent are defined, the 19 ,=~ = ex
Pythagorean properti es tell that j (x ) = 1. .y cosy J 1- e2x
Problem
Set69:Derivative
andIntegral
Practice
forTranscendental
Functions 70 9
21. y ' = 1/ x 23. y ' = 1/x b. L(2) = 0.693 .. . , which equals ln2.
L(3) = 1.098 . . . , which equals ln 3.
25. y ' = 2xln2 27. y ' = 2x
L(4) = 1.386 ... , which equals ln4.
29. y ' = x x (lnx + 1) 31. y ' = xe x L (8) = 2.079 ... , which equals ln8.
L(12) = 2.484 ... , which equals 1n 12.
33. y ' = (ex+ e- x ) 35. y ' = 5x 1ns c. L(3 4) = 2.484 . .. = 1.098 ... + 1.386 .. .
= L(3) + L(4)
x -8
37. y' = ln (- 7lnx + 1) L(12 / 3) = 1.386 ... = 2.484 ... - 1.098 .. .
2 = L(12 ) - L(3)
L(2 3 ) = L(8 ) = 2.079 ... = 3 0.693 . . . = 3L(2)
39. y ' = e- 2x (- 2ln5x + 1/ x)
R3. a. i. y ' = (3/ x )(lnSx ) 2
ii. j' (x ) = 9 / x
41. y ' = 1/ x 43. y ' = 1/ x
iii. y ' = - csc(lnx) cot(lnx) (1 / x)
' 7 47. y' = esinx cosx
45 Y = xlnS iv. g ' (x ) = 2x csc x 2
b. i. ln I sec x I + C
49. y ' = 0 51. y ' = cosx
ii. 10(ln3 - ln2) = 4 .054651. . .
53. y ' = -cscxcotx 55. y' = sec 2 x
iii. ln lx 3 -4 l+ C
5 7. .!.e.ix +C 1 x" + C
59 . 4e
4 c. Memory Retention Problem
1 6
5x i. y (lOO) :::::70 names; 70% remembered
61. -(lnx) +C 63. - +C y ( 1) = 1 name; 100% remembered .
6 ln S
2x .. , 101
65. lnx 67. ln2 + C ll. y = 100 + X
1 y'(IOO ) = 0.505 names / person
69.3 ln lx l+ C 71. - (lnx) to + C y '( l ) = 1 name / person
10
iii. Paula has probably not forgotten any names as
73 . .!.x 2
2 +C 75. C long as x - y < 0.5. After meeting 11 people
she remembers about 10.53 ... :::::11 names,
1 but after meeting 12 people she remembers
77.
2 ln I sec 2x + tan2x l + C about 11.44 ... :::::11 names.
R4. a. See text definition of ln.
79. ln I sin4x l+ C
b. See text definition of logarithm.
c. See text statement of the uniqueness theorem.
81. Limit= 0 83. Limit = rr / 2
d. See text statement of In of a power.
85. Limit= l!S = 20.8333 ... e. See text for In of a quotient property, and see the
solution for Problems 9 and 11, Problem Set 6-4,
for proof of the quotient property.
87. Limit = eO = 1
RS. a. i. y ' = 10ox ln 100
89 . Limit= e- 3 12 = 0.22313 ... ii. j'(x) = (0 .74lnl0)(I0 2x)
iii. r ' ( t) = t tan t [ sec 2 t ln t + (tan t) I ( t)]
Problem Set 6 - 10
Chapter Review and Test b. y , = (Sx - 7) 3 (3x + 1) 5 (
Sx15_ + x15+ )
7 3 1
Rev iew Problems c. Vitamin C Problem
RO. Not selected. i. From Figure 6-lOb th e maximum concentra-
tion is about 150 ppm at about 2 hours.
Rl. a. dM / dt = 0.06M ~ M - 1 dM = 0.06dt
ii . C(t) is incr easing at about 58.7 ppm/hr when
1
Ii
:. fi~oM - dM = 0.06 dt, Q.E.D. t = 1, and decreasing at about 24.2 ppm / hr
b. X :::::134.9858 ... when t = S. The concentration is increas-
c. The interest would be $34 .99. ing if C' ( t ) is positive and decreasing if it is
R2. a. Integrating x - 1 by the power rule resu lts in negative.
x - 1+ 1 iii. C(t ) = SO for t :::::0.2899 ... and t :::::
division by zero: -- + C. 6.3245 .... So C(t ) > SO for about 6 hours.
- 1+ 1
710 Appendix
B:Answers
toSelected
Problems
iv. Graph, C1 ( t) = 200t 0.3t iv. t = 460.5170 . . . , or about 461 years .
f. Chemotherapy Problem
C (t)
i. The exposure is the product of C ( t) and t,
where C(t) varies . Thus a definite integral
must be used.
ii. E(x) = 937 .5(-e - 0 - 16x + 1)
(5) = 516.25 ... ppm days
(10) = 748.22 . . . ppm days
As x grows very large, E (x) seems to approach
From the graph, the maximum is about 60 ppm 937.5.
around t = 1. iii. E'(x) = 150e - 0 16x = C(x)
C(t) = 50 fort "" 0.409 ... and t = 1.473 .... '(5) = 67.39 ... ppm (or ppm days) per day
C(t) > 50 for about 1.06 hours. '(10) = 30 .28 .. . ppm
In conclusion, the concentration peaks sooner
at a low er concentration, and stays above 50 RS. a. Limit = - 2/ 5
ppm for a much shorter time. b. Limit = 3
c. Limit = 0
R6. a. e = limn -o( l + n ) 1 tn = limn -oo(l + l / n)n
d. Limit = e- Z/rr = 0.529077 ...
b. log e x = lnx
lnx e. Limit = 48 (Don't be fooled!)
C. ]og b X = In b f.Limit =- 1
. , 1 g. Examples of indeterminate forms:
d 1. y = xln4 oI o, ooI oo, o oo, o0 , 1 ooO, oo - oo 00
,
ii. f' (x) = - t:; R9. a. i. y ' = 28 cot 7x
ii. y ' = x - 4 e2x(2 x - 3)
iii. y' = log 5 9 iii. y = cos(2x) => y' = - sin (2x) 2x In 2
e. ot selected. . , 4
iv. y = xln3
R7. a. i. y = ex
ii. y = exp(-x) b . i. (-1 / 1.7)e-l.?x + C
iii. y = lnx ii. (1/ln2)2 5 ecx + C
iii. ln(5 + sinx) + C (No absolute value is needed.)
iv. ln 5 (by definition!)
C. i. Limit = oo
ii. Limit = e- 3 = 0.049787 ...
Problem Set 6-11
b. i. j' (x) = x 0 ..i exp(5x) (1.4 + 5x ) CumulativeReview, Chapters I Through6
ii. g ' (x) = -2e - 2x cos e- 2x
1. j'(3) "" 5.549618 ...
iii . y ' = l
C. i. -5e - 2x+ c 2. f(gg (x) dx "" 200
ii. -ecosx +c 3. L = lim x -c j(x ) if and only if for any E > 0 there is
iii . -lOexp (- 0.2 ) + lOexp(0.2)
= 4.0267200 .. . a 6 > 0 such that if x is within 6 units of c but not
d. The numb er e is us ed as the base in calculus equal to c, then j(x) is within E units of L.
since the algebraic formulas for the derivatives 4. Graph. (example)
are simpler.
f(x)
Exponentials: d(ex) / dx = ex; d(a x) = a x Ina
Logs: d(lnx) / dx = l / x; d(logbx) / dx X
(1 / x) / lnb. 31
e. Radioactive Decay Problem
i. p(5) = 88.2496 ... , so about 88%remains.
ii. p' (0) = -2.5. Decreasing at about 2.5%/year . ,( ) _ . j(x + h ) - j(x)
p'(5) = -2.5e- 0 125 = -2.2062 ... 5. f x - 11mh-o h or
Decreasing at about 2.2% per year. '(C ) _- 1.lffi x-c j(x) - j(c )
iii. t = 27.7258 ... , or about 28 years. ! -----
X -C
Problem
Set6-11:Cumulative
Review
, Chapters
l Through
6 711
6. J(x) = x3 15. y = x 917
'( ) -lim (x+ h )3 -x 3 Either:
! X - h- 0 h y 7=x9
. x 3 + 3x 2h + 3xh 2 + h 3 - x 3 7y 6 y ' = 9x 8
= limh - 0 h I' = 9xB 9xB = ~ x B-5.J/7
J 7y6 7(x9 f 7)6 7
= limh - o(3x 2 + 3x h + h 2) = 3x 2, Q.E.D.
= ~x 2 17 = ~x 9 i 7- 1 as from the derivative of a
7. j'(5) = 75 7 7 '
6x = 0 .01: f'(5) "'75 .0001 power formula.
6x = 0.001: j'(5)"' 75.000001 Or:
The symmetric differences are getting clos er to 75 9 9
lny =
as 6x gets clos er to zero, Q.E.D. 7 lnx => ( 1/ y)y ' =
7 (1 / x) =>
9 9 9 9 -
8. J'(7) = 0.375 = 3/ 8 y ' = - (l / x)-y= - (1 / x)x 9/7 = - x 2/7 = - x 9/1-1
7 7 7 7 '
9. Graph . Line with slope of 3/8 is tangent to th e gra ph as from the derivative of a power formula.
at X = 7.
16. If x - 1 were the derivative of a power, then the
l(x) __
power would have to be x 0 . But x 0 = 1, and so its
derivative eq uals 0, not x - 1 Thus x - 1 is not th e
derivative of a power, Q.E.D.
17. f' (x ) = cos(3 tan x) sec 2 x
X
5 18. f(x) = f;"(1 f t ) d t =>j'(x ) = 1/ x, Q.E.D.
19. Prove lnx a = a ln x for any constant a and all
10. Optional graph showing upp er sum, U6 = 24.875.
X > 0.
Proof:
Let J(x) = lnxa and g(x) = alnx.
10 1 1 a
Then f'(x) = - ax a- , a - - , an d
xa X X
X g'(x) =a..!.=~
4
X X
:.f' (x) = g' (x) for all x > 0.
11 . M10 = 20.9775, M 1oo = 20.999775 j( l ) = ln (la) = lnl = 0, andg(l) = a ln l = 0.
Sums seem to b e approaching 21. .'.j(l) = g(l).
: .j(x ) = g(x ) for all x > 0, an d thus lnx a = alnx
12. a. - 1 cos 6 x + C for all X ~ 0, Q.E.D.
6
b . ln lx l + C 20 dy = 3 cost
c. - ln Icos x I + C dx - 5 sin t
d . 1nlsecx + tan x i + C 21. At t = 2, (x, y) = (- 2.08 ... ,2 .72 ... ), and
e. i(3x - 5) 312 +C dy I dx = 0.2745 . ..
Graph, showing that a lin e of slope 0.27 ... at point
13. Integral = 21 , as conjectured in Problem 10. (- 2.08 ... , 2.72 ... ) is tangent to the curve .
14. Graph, example, showing tangent lin e parall el to y
secant line at x = c.
f(x)
X
a Cb
Statement: 22. V = (1 + t2) - l
If f is differentiable on (a, b ) and continuou s at 2t
x = a and x = b, then th ere is a numb er x = c in a=-(l+t 2)2
(a, b ) suchthatf' (x) = J(b~ = ~(a)_ 23 . Limit= 3/ 5 /
7 12 Appendix
B:Answers
toSelected
Proble
ms
24. L = limn - o(l + n) t i n - 1" a. no. of millions of bacteria; t = no . of hours.
dB/ dt = kB=} f dB/ B = f kdt
lnL = lim n- o ( ln(l + n))- oo 0
B = C1ek1
= limn -0 ln (l + n) - Q b. B = 5e <l/3l0n <7/ Sllt = 5(7 /5)' 13= 5eo.1121s7... ,
n 0 c. Graph .
= lim,, _ o 1/( 1 + n) = 1 B
1
:. L = e 1 = e, Q.E.D.
2
25. Int egra l :,:,556
3
2. 5
26. Area of cross -section = rr y
Since th e end of the radius is on a lin e through the
origin with slope r / h, y = (r / h )x .
.".Area = 7T [ (~) X r
Graph, showing Area as a function of x, with strip
= rr;;/ x 2.
d . About 74 million .
e. t = 47.24001 . ..
Abou t 47 hours since start.
of width dx and sample point (x, y) on the graph 3. Chemica l React ion Problem
within the strip. a. F = no. of mg; t = no. of minut es.
dF / dt = kF
Area F = 50(0.6)1 120 = 5oe - o.02ss41...t
b . Graph.
F
50
dx X
0 h
dV = (Area) dx
. .1, h rr r 2
.. V = J0 (Area) dx = fo x 2 dx
---;;;F
= rrhr22 . ! X 31h -
-
! 1T rh 22 (h 3- O3) -- ! rr r 2h, Q.E.D. C. 10.8 mg
3 O
3 3 d . t = 347.4323 .. . About 5 hr 47 min .
27. Not selected. 5. Biological Half-Life Prob lem
a. dC / dt = kC
b. C = 0 .00 3 ne - 0.066
2277...t
CHAPTER 7
c. Either: C = 0.015 ~ t = -21.05 ... , which is
before th e poison was inhaled,
Exploratory Problem Set 7-1
Or: t = -20 ~ C = 0 .0139 . .. , which is less th an
DirectProportionPropertyof ExponentialFunctions 0.015 .
1. t D(t ) .". concentration never was that high.
0 500 d. Grap h .
10 895.42
20 1603.57 0.015 C
...... 0.0139
3. t R(t )
0 0.0582689081 . ..
10 0.0582689081 ...
20 0.0582689081 . ..
- 20 20
5. j(x) = a bx~ j'(x) = a (lnb) bx
= (lnb)(a b x)= (lnb) j(x) e. t = 10.4661 .. . :,:,10 .5 hours .
So j'(x) is dir ectly proportional to j(x), Q.E.D.
7. Compound Interest Problem No. 1
Problem Set 7 -2 dM / dt = kM ~ M = C ekt, where C is the initial
investment. :.M var ies expo nentially with t.
Exponential Growth and Decoy
Let i = int erest rate as a decimal.
l. Bacteria Problem dM / dt =Ck. ekt
Problem
Set7-2:Expo
nentia
l Grow
thandDecoy 713
At t = 0, dM/dt = Ci d. i. I= 4. 32 81 ... ;:s 4.33 amps
:.Ci = Ck e0 ~ i = k ii. I = 10.9258 ... ;:s 10.9 3 amps
Examples : iii. I= 11 amps
$1000 at 7% for 5 years: $1419.07
e. t = -2 ln0 .05 ;:s 6 seconds
$1000 at 7% for 10 years : $2013.75
$1000 at 14% for 5 years: S2013.75 5. Hot Tub Problem
$1000 at 14% for 10 years: $4055.20 dV - kv1 12
a. dt -
Leaving the money twice as long has the same effect
as doubling the interest rate. Doubling the amount b. f v - 112 dv = k f dt ~
invested doubles the money at any particular time.
2V 1i 2 = kt + c ~v= (kt;cr
9. Generalization Problem
d y I dx = k y ~ f d y I y = f k dx ~ In Iy I = kx + C 1 V varies quadratically with t.
lyl = ekx +C1 ~ y = Cek x , Q.E.D. c. Initial conditions t = 0; V = 196; dV / dt = -2 8:
195112 = k . 0 + C ~ C = 28
Problem Set 7 -3 2
Other DifferentialEquationsfor Real World Applications and: - 28 = k 196 1 12 ~ k = -2
1. Sweepstakes Problem No. 1 .-.V = ( - 2 \+
28 ) 2 ~V= (t - 14 ) 2
a. dM / dt = 100 - S
d. False. Since dV / dt = 2t - 28, the water flows
b. S = kM ~ dM / dt = 100 - kM (which is positive) out at 28 only when t = 0. For instance, at t = 5,
dM dV / dt = - 18, which means water flows out at
C. f 100 - kM = f dt
only 18 ft 3 / sec. So it takes longer than 7 minutes
M = 100 (1 - e - kt) to empty the tub.
k e. O = (t - 14 ) 2 ~ the tub is empty at
d. M = 5000(1 - e- 0 021 ) t = 14 minut es.
e. Graph . f. Graph .
M V
5000 ..... . . .. .. _
100
30 60 90 14
f. t = 30: $2255 .94 ($3000 in, $74 4 .06 spent) g. See Problem Set 7-7, #C4.
t = 60: $3494.03 ($6000 in, $2505.97 spent) 7. Differential Equation Generali zation Problem
t = 90: $4173.51 ($9000 in, $4826.49 spent) a. n = 1, k = 1, C = -3 : y = 0.04978 ... ex. Graph.
g. t = 365: (366 could be used.) : M = 4996.6 22 ...
;:s $4996.62 in the account. y
dM / dt = 0.06755 ...
M is increasing at about $0.07 per day.
X
h. limt -oo M = 5000
3. Electrical Circuit Problem
a. E=Rl+L(dl / dt)
b l=E_(l -e - U</L)t)
. R 1
b. n = 0.5, k = 1, C = -3 : y = - 3) 2 . Graph .
c. I = 11 ( 1 - e- 0 51 ) Graph . 4 (x
y
11 ..
5 10 3
714 Appendix
B:Answers
toSelected
Prob
lems
Note: x ~ 3 because the symbol y 0 5 stands for a y
positive number. 2
c. n = - l, k = l, C = -3 : y = J 2x - 6. Graph .
X
5 y
9. Water Heater Project
T "' 70 + 102.26 ... (1 - e- o.oz933 t )
Time data for various temperatures can be found by
grapher , or by substituting for T and solving for t .
n = - 2, k = l, C = -3 : y = 1/3x - 9. Graph. T
Never reaches180
170
5 y
39 72 130
T t
140 39 min.
155 61 min .
d. For n > l, !~
-( n - 1)
= kyn ~ f y - n dy = k f dx 160 72 min .
170 130 min.
~ - Y = kx + C since n > l, so 180 ever!
n - l
The limit of T as t increases is 70 + 102 .26 .. . (1 + 0)
-1
which equals 172.26 . Thus, the temperature never
y = "-1/(n - 1) (kx + C)'
reaches 180 .
which has a vertical asymptote at x = - C / k since When the heater turns off the differential equation
the denominator = 0 for this point. becomes
Note that the radical will involve a sign when dT
the root index is even (i.e., when n is odd). dt = - kh(T - 70) ~
For n = 2, k = l, C = -3: y = - (x - 3) - 1 . Graph . T = 70 + C 2e - kht
Using T = 160 at time t = 0 when the heater turns
y off,
2 T = 7o + 90 e- o.oz933,.,t
To find the time taken to drop to 155, substitute:
X
15 5 = 70 + 90 e-0 ,02933...t
Solving numerically or algebraically gives t = l.9 .. ..
Thus it takes only 2 minutes for the temperature to
drop 5 ! By contrast, from the above table it takes
11 minutes (t = 61 to t = 72 in the table above) to
- 1 warm back up from 155 to 160 .
For n = 3, k = l, C = - 3: y = .J2x _ 6 . Graph. The design of the heater is inadequate because it
takes much longer to warm up by a certain amount
than it does to cool back down again . Near 172
y
I
a slight increase in the th ermostat setting for the
I
I heat er makes a great increase in the time taken to
--- - !.. reach that setting . For instance, it takes an hour (72
minutes to 130 minutes) to warm the 10 degrees
from 160 to 170 . These inadequacies could be
corrected most easily by adding more insulation.
The resulting decrease in h would make the heater
cool mor e slowly, heat up faster, and reach the 180
e. n = 0, k = l, C = -3: y = x - 3, which is a linear degrees it currently cannot. Decreasing h would
function, Q.E.D. Graph. also reduce the power consumption.
Problem
Set7-3:Other
Differential
Equations
forReal
World
Applications 715
Problem Set 7-4 also true for this differential equat ion. In quad-
GraphicalSolutionof DifferentialEquationsUsingSlope Fields rants I and III the slopes are all negative, and in
quadrants II and IV they are all positive.
1. a. At (3, 5), dy / dx = 3/ 10 = 0.3. (Note: The algebra ic solution is y = ce - o.ix 2 )
At (- 5, 1), dy / dx = - 5/ 2 = - 2.5.
On the graph, the line at (3, 5) slopes upw ard 5. Graph. (differential equation is
with a slope less than 1. At ( - 5, 1) the line slopes dy / dx = (x / 5) sin(rry / 8 ) + y / 10.)
downward with a slope much steeper than -1.
Y/ /.,,,,,.. ___ __ ....,_
b. Graph . The figure looks like one branch of a //////////
!///////// / / / / / - --.....
=- -
hyperbola opening in the y-direction. (The lower .::;' ___ ____,,,.,,,..,,,,.
////////// / / / / / ~ // / /
branch shown on the graph is also part of the , , , -.. - - - - .- ,,,.
/
,,, / /
/ / /
/ '1/ I
/, / .i I
I I
I
I
I
I
solution, but you are not expected to find this \ ~ ~ '' ' ' - - - _.5 {J "4)/ / I 1/ I I I I
\ \ "".' , ' -.. - - - Y '/ ._ / I I I I I
graphica lly.)
,,,,
\ \ \, ' ~ ....
.........
- - - ,..,.,./
...
/ /4 /
///////
/ / / / /
~ ............ ,,"
-...;:,, ......,-..---- : / 11~2!:: == (In)~ -.. ...... ' ' ""' \ \ \
~., ' ........... - - - - /1//// ,,...-:::- =s::. ~~ : : ~ : :
' ~-> ' ' ........... - - -s 1/ / / ~,.... ,,...
..,.. - - - .._ ::::::~
"\ \ \ \ \ \ \
\ '~ >-' / / ,,,..,,..
-/::..
-.:: ___- .._....._....._,,
- - -- .......
' "' \ ' ' ' \ \ \
,,,,,,,,,, ''"
\ \ \ \ ,.~(-,,,l-._,,
\ :s~' . ' ' ' '"''''''
\ \ \ , - ~ '' a,.. - -
I I I I\ \\\\\\\''" '' '''
'' '..............
,' .......
'~'::.:::=.
'' ~
\\\\\\\\\'- "''....._ _____ ,,..
I I I I I i I I/,,..
I I I I <"/ // . /,,,: .<_. --
! I / / , 1/ / - -~->' ~ \ \
, ,,,,
; ;_. ;;..; :::: = --- - --.:
- '-~ ,' ., '
. .......
7. Rabbit Population Problem
.<,-:,/ / /,.....- --- - - - ---
.......... :'..
( ,' / - - - - - - - - - - - - , .;
/ ,....
................
...........
~
a. Graph. Initial condition (0, 2)
R I I I I I I I I I I I I I I I I I I
I O, 1 ~ ) I I I I I I I I I I I I I I I I I
\ I I \ \ I \ I \ \ \ \ I I I I I I I
\ \ \ \ \ \ \\I I \ I I \ I \ I \
c. Graph, above . The figure looks like the right-hand ~ \ \ \ \ \ \
branch of a hyperbola opening in the x-direction, ' ' ,,,,,,,,,,,,,,,,,
\ \ \ \ \ \ \
(The left-hand branch is also part of the solution,
but you are not expected to find this graphically.)
d. x 2 - 2y 2 = 23. /
:::::::-::,,,..:;~-:
//////.1////
,, : ~:: ~ ////////
/
This is the particular equa tion of a hyperbola / ////h/// b //////////
/ / / / h / / / 1 // / / / / / / / / / /
opening in the x-direction , which confirms the
observations in part (c). / ; ;,; ; ; ;,,r;;;;;;;;;;;;
/ h// / f~ /////////////
_ (Q,
///
?) _ j, 4y 21. ___
//////
_____
/// //////
____ (
3. a. Graph,
; I ; I I / Y, \ \ I I I
I I I \ \ I I I
'
I / I,
I -...
-- -"
I I I r,' / \ I I
(\ 2), b. Graph, above. Initial condition (4, 2)
I I I I / '- I
I / -~ -._ The graph is identical to that in part (a) shifted
I
--...'"->,- - ''' X
over 4 months. This behavior is to be expected
- since dP / dt depends only on P, not on t , and
'''
..- ~r.,. ~ / / / I
..::
,~_ -i,}2/ I both initial conditions have the same value of P .
\ \
I
\
' -._ ,., I I I
I
I
I
I
I
I
I I \
I
I
\
' \ '- / I I I I I I
c. Graph, above. Initial condition (0, 18)
I I I I I I I The population is decreasing to the same asymp -
\ I \
' / I
tote, P = 10.5, as in parts (a) and (b).
d. The asymptote at P = 10.5 indicates that the
dy island can sustain only 1050 rabbits. If the pop-
b. dx = -0 .2xy
ulation is lower than that, it increas es. If the
Evidence: At ( 1, 1) the slope was given to be population is higher than that, it decreases. The
-0.2, which is true for this differential equation. number 10.5 is a value of P that makes dP / dt
As x or y increase from this point, the slope equa l zero. Note that there is another asymptote
gets steeper in the negative direction, which is at P = 0, which also makes dP / dt equal zero .
716 Appendix
B:Answers
toSelected
Prob
lems
9. Escape Velocity Problem b. At (0.5, 3), dy / dx = -0.0833 ....
mg At X = 1, y ::::2.9583 ....
a. ma = - 2 by hypothesis
r c. x y slope dy
dv fl..... di VIcte b ym; a = -dv 1 2.9583... - 0.1690 .. . - 0.0845 .. .
dt r2 dt 1.5 2.87 38... - 0.2609 . . . - 0.1304 .. .
dv dr g 2 2.7433... - 0.3645 .. . - 0.1822 ...
dr dt = r 2 chain rul e
2.5 2.5610... - 0.4880 .. . - 0.2440 ...
dv g dr
dr . V = y 2 V =- (r = distance) 3 2.3170 ... - 0.5463 . . . - 0.3236 ...
dt 3.5 1.9933 . .. - 0.8779 . . . - 0.4385 .. .
dv g
divide by v . 4 1.5543 .. . - 1.2866 .. . - 0.6433 .. .
dr r 2v 4.5 0.9110. . . - 2.4696 .. . -1.2348 .. .
dv 5 - 0.3237... 7.7213 . . . 3.8606 .. .
b. dr (5, 2) = - 1.2488
5.5 3.5368 . .. - 0.7775 . . . - 0.3887 ...
dv 6 3.1481... - 0.9529 . . . -0.4764 ...
dr (1, 10) = - 6.244
6.5 2.6716 .. . -1.2164 . . . -0 .6082 .. .
dv 7 2.0634.. . -1.6962 .. . -0.8481 .. .
dr (10,4 ) = -0 .1561
These slopes agree with those shown . d. Graph.
c. Graph. Initial condition (r, v) = (1, 10) y - - - -
Spaceship stops abou t 4 earth radii, or about
25,000 km above the surface.
3
,
I \( , __ ______________ _
X
~~ =~==
~ ~- - -=-==
- =
- -==-== - -=-==
- = - =
- -
ll-~t., ---=-=-=-=: =-=-=-=-=-=: = -- -- ,,,,.,_::,,'
-- :/- -0--/
~(Uii:: : : : : : : : : : : : : : :
~,,-- -- - ,,.,... ,.,.
,to.). ------ -------- - ,,,,.
. /
,~(<. - - - - - - - - - - - - - -
~\~?S~iitiJll~~
i I '-' - - - - (iQ":"41
- - - - - -
The values of y follow the slope field through
x = 4 .5. At x = 5, y becomes negative . As a
result, dy I dx is positive and the curve jumps
up in the positive direction. Thus, for x > 5 the
I I : ~~t~====~
\ ~,,,,,,------ -=====
\ -,= y-va lues by Euler's method are clearly wrong.
I I I I /////// _____ _ _ Note that the values follow another curve that
has initial condition (5.5, 3.5368 . .. )
d. Graph above . Initial condition (r, v) = (1, 12) 3. Accuracy of Euler's Method
The graph levels off between 4 and 5 km / sec. 1
a. y 2 = - x 2 + 16 =} y = - v'l6 - 0.5x 2
e. Graph, above . Initial condition (r, v) = (1, 18) 2
The negative root is picked because y is negative
The graph levels off at v :::: 14 km / sec. Here
whenx = 0.
the spaceship loses about 4 km / sec of velocity,
For x = 5, y = -v'3.5 = -1.8708 ....
whereas it loses 7 or 8 km/sec when starting at
The point (5, - 1.87 . .. ) is on the graph (see
12 km / sec. Both cases lose the same amount of
Problem 1, part (d)).
kinetic energy, which is proportional to v 2 (the
change in v 2 is the same in both cases). b. 6.x = 0.5 c. Graph, 6.x = 0.5.
f. Graph, above . Initial condition (r, v) = (2, 10)
X y 4 y
The graph levels off at about 6 km / sec, so the
spaceship does escape. Alternatively note that
0
1
-4
-3 .9687 .. .
- - ' ' ' ' '
the solution through (2, 10 ) lies above the solu-
2 -3 .8097 ...
- ' ' ' \ \
tion through (1, 12).
3 -3 .5085 .. . ...... ' \ \ \ \ \
X
4 -3 .0292 ... 5
Problem Set 7 -5 5 - 2.2823 .. . ,,,,. / I I I I
NumericalSolutionof DifferentialEquationsby Euler'sMethod 6 - 0.9420 .. . - ~ / / I '
7 - 1.8486 .. .
1. How Euler's Method Works
a. At (0, 3), dy / dx = 0. At x = 0.5, y ::::3.
ProblemSet75:Numerical
Solution
ofDifferential
Equations
byEuler's Me
thod 717
Euler's method gives a good approximation close 1 dP
d. Pdt "' 0.02802596 ... - 0.0000792747 .. . P
to the starting point, but gets progressively
worse farther away from the starting point. The dP
~ dt = P (0.02802596 ... - 0.0000792747 ... P)
answers are meaningless beyond the x-intercept,
where the slope is infinite. e. Graph.
d. 6.x = 0.1 Graph, 6.x = 0.1
X y 4 y
- .... .... .... ....
0 -4 .... ....
1 - 3.9434 . . . ' '
2 - 3.7555 ...
- .... ....
' ' \ ~
3
4
- 3.4155 ...
- 2.871 5 ...
....
' \ \ I
5
I
X
5 - 1.9696 ... / / I I I I
6 0.7600 .. . - / / / I I
~ o- ___ o ____ 50 ____ 100
7 2.1652 ... / / /
f. year t Euler actual" Euler'"'
The approximation is better if 6.x is smaller. 1890 -50 44.6 .. . 62.9 46.1 ...
The values still diverge slightly from the ac- 1900 -40 56.9 .. . 76.0 58 .3 . ..
tual values as x gets farther from the starting 1910 -30 71.7 ... 92.0 73.9 . ..
point. For instance, if x = 5, y "' -1. 9696 . .. 1920 -20 89.2 ... 105 .7 90.1 . . .
by Euler's method compared to the exact value, 1930 - 10 109.3 ... 122.8 109.8 .. .
- 1.8708 ... , from part 3(a). The method still fails 1940 0 131.7 131.7 131.7
beyond the x-intercept. 1950 10 155.4 . .. 151.4 155 .0 .. .
e. Using 6.x = 0.01, y "' -1.8812 ... when x = 5, 1960 20 180.1 . . . 179.3 179 .2 . . .
which is quite close to -J3.S = - 1.8708 . ... 1970 30 204.7 ... 203.2 203.5 ...
dy 1980 40 228.2 . . . 226.5 226 .9 . ..
5. a-b. Graph, dx = - 0.2xy 1990 50 249.4 ... 248.7 248.8 ...
2000 60 269.3 ...
I I I I I I / )'.,_ \ \ \ I I I
2010 70 286.1 ...
I I I I I I /
'...."\ \ \ \ \ 2020 80 300 .2 ...
- --
I I I I I ~ / I
I I I I ........' (~~ I 2030 90 311.8 . ..
I / / /
/
- ........ '\ \ X
2040 100 321.1 ...
\ '\ ',- .............
....- - ..--_,,/ /
-
/ / I *Data from The World Book Encyclopedia
I
I
I
I
\
\
\
I
'\
\
'' '-.... --i<1)
/ l I
I I
I
I
**Note that although linear regression gives the
I I \ \ \ .... I I I I
/
"best" fit for (6.P/ 6.t )/ P vs. P, actually plotting
I I I I \
\ ' /
the graph shows that the data point for 1960 is
considerably out of line.
7. U.S. Population Project
a-b. year p 6.P I M (t:,P I 6.t) I P (a P/a t)/ P
950
1940 131.7 0.015 '~
, 1960
1950 151.4 2.38 0.01571 . .. 0.014 ',
0.013 ',
1960 179.3 2.59 0.01444 ...
0.012 '
1970 203.2 2.36 0.01161. .. 1970'..
1980 226.5 2.275 0.01004 . . .
0.011 ' P
981),
150 200
1990 248.7
c. Using linear regression on the values of
(6.P I M ) I P without round-off gives Using the two end points, 1950 and 1980, gives
1 6.P (6.P / M )/ P = 0.002716 ... - 0.00007558 . . . P
p t:,t "' 0.02802596 ... - 0.0000792747 .. . P Using this equation gives populations much
The correlation coefficient is r = - 0.98535 .. . closer to the actual for the given years, as shown
For the other types of regression: in the right-most column above . This is, of course,
r = - 0.978 . . . for logarithmic no guarantee that the lat er model fits any better
r = - 0.981 ... for exponential in the future than the former one.
r = -0.981 ... for power g. The population growth rate is zero if dP / dt = 0.
Thus a linear function fits best because r is Let P(0 .02802596 ... - 0.0000792747 . . . P) = 0.
closest to -1. P = 0 or
718 Appendix
B:Answers
toSelectedProblems
RI //_...,..._....
I///.------ _ _ _ _ ___- ------- ....-_
P = (0.02802596 ... )/ (0.0000792747 . .. ) = _ __ __ __
353.5 ... 'I///..----
I/_.,..,... ____ -- _------
_____ - ----..... .._
..._.._.._
Predicted ultimate population "" 353.5 million . 50 1 //..--- - - - - - ----- - --
Differential equation: P = 353.5 . .. makes /I//...-------------.-.-...--...
I I///.----- ... -. ....................... .._.._..._
dP / dt = 0. I I// ,.,, ..- - - -. -.. .......................... ......
!// h /------..- -~ _____ ...
Graph: P = 353.5 ... is a horizontal asymptote . I / /, / / - - - - - - - ._ - (Z.0_3Q) __ _
h. Graph, part (f). Data do follow th e solution .
i. Example answer: "The predicted populations
agree fairly well with the data for the six given
years. The fit is exact for 1940 since this point
was used as an initial condition. For the other five
years the predicted populations are a bit higher 15. Graph in Probl em 13, starting at (70, 30) . With this
than the actual population." many foxes and hunters chasing rabbits, the rab-
j. Actual data are underlined in the table in part (f). bits become extinct. At this point the foxes have
k. The predicted population for 2000 from part been reduced to just 5. After the rabbits become
(f) is 269.3 . . . million . Using 469.3 million as extinct, the foxes decrease exponentially with time,
an initial condition in 2000 gives the following eventually becoming extinct themselves.
predictions:
year t Euler
2000 60 469 .3 Problem Set 7-7
2010 70 433.9 .. . Chapter Review and Test
2020 80 410.6 . . .
Review Problems
2030 90 394.8 .. .
RO. Journal entri es will vary.
2040 100 383 .7 .. .
The logistic model predicts that the population Rl. Pu nctured Tire Prob lem
,vi.ll drop, approaching the ultimate value of P(t) = 35(0.98 t)
353.5 million from above. This behavior shows P ' (t) = 35 (0.98 1 ) ln0.98
up in the slop e field of part (c), since the slopes
t P ( t) P' (t) P ' (t ) I P(t)
are negative for populations above 353 .5.
0 35 - 0.7070 .. . - 0.2020 .. .
9. Not selected. 10 28. 59 7 .. . - 0.5777 . . . - 0.2020 . . .
20 23.366. . . - 0.4720 .. . -0.2020 . . .
Problem Set 7-6
P ' (t) = 35 (0.98t ) ln0 .98 =ln0 = - 0 2020
Predator-Prey
Population Problems P(t) 35 (0.98t) .98 . . .. '
dR dR which is a cons tant, Q.E.D.
1. dt = k 1R ~ R = k 1dt ~ ln IRI = k 1 t + C
~ IRI = eCek1t ~ R = C1ek1t R2. Ramjet Probl em
R is increasing since k1 > 0.
a. V = speed in mph, t = time in seconds .
dR dV = kV
3. dt = k1R - k3RF
dt
dF
dt = -k 2F + k4RF b. f
dV
v
= kf dt
5. dF I dR = 0.4017 . .. ln IVI = kt + C ~ IVI = ekt +c = ec . ekt ~
V = C1ekt
7. The populations vary periodically, and the graph is Mathematicall y, C 1 can be positi ve or negative,
cyclical. The foxes reach their maximum 1/4 cycle so the absolute value sign is not needed for V.
after the rabbits . In the real world V is positive, which also makes
9. dF / dR = - 0.5357 ... the absolut e value sign unnec essar y.
c. V = 40Qe0.005578...t
11. The populations now spiral to a fixed point. The
rabbit population stabilizes at the same value as in d. t = 112.68 . . . "" 113 seconds
Problem 8, R = 40 (4000 rabbits), which is surpris -
ing. The stable fox population decreases from 2 5 to R3. a. y = (3x + C ) 2
15. b. y = (3x - 4 ) 2
13. Graph. The slope at (70, 15) is about -0.4. c. Graph .
Pro
bel mSet7-7:ChapterReview
andTes
t 719
X y( 6x = 1) y(6x = 0.1 )
1 9 9
2 7.227 . .. 7.707 ...
3 6.205 . . . 6.949 ...
4 5.441. .. 6.413 ...
5 4 .794 . . . 5.999 . . .
6 4.200 . . . 5.662 ...
7 3.616 . .. 5.377 . . .
8 3.007 . . . 5. 130 ...
9 2.326 .. . 4.910 ...
10 1.488 ... 4 .712 ...
d. At x = 2, y' = 12 and y = 4. 11 0.2185 ... 4.529 . . .
A line through (2, 4) with slope 12 is tangent to 12 - 8.091 . .. 4.359 . ..
the graph, showing that 12 is reasonable . 13 4.199 ...
14 4.045 .. .
e. Memory Retention Problem 3.896 . . .
15
= 100 - kN
i. dN I dt 16 3.750 ...
N = 2210 .6 ... (1 - e - o.o-1s236t) 17 3.604 . . .
ii . About 1642 names. 18 3.457 .. .
19 3.306 .. .
iii . Brain saturates at about 2211 names. 20 3.150 . . .
iv. t "" 27 days. 21 2.986 .. .
22 2.811 ...
R4. a. At (2, 5), dy / dx = -1.75.
23 2.621. ..
At (10, 16), dy / dx = 0.675
24 2.410 ...
The slopes at (2, 5) and (10, 16 ) agree with
these numb ers as shown on the graph in part
28 .9 0.1344 . . .
(b).
29 -0.3 796 . . .
b. Graph, initial conditions (1, 8) and (1, 12) . Graph.
~ /////////////////
I JL .,...,,,,~,, -'/// ///// / /////,,,.,.,,. ,/'////////
- .,. / / /
_,,,,,,,,,,,,,,,,,,,,,,/,,6.
/ / / / f"
/
/ /
/ __ _,,,,,,,,.,,,,,,.,,,,,,,,,,,,,,,,,.
////////////////
"/////
......
- / / / / / ,,..
09 ;.,"!fa)/ / / / - .,....,....,.,..,.,..,.,..,.,.,,,,,,,,,,,.,,,.,,,.,.,.,,,,,,,,,,,,.,,,
/ / / / / / / / V / / / /
, _____ _,.,,,.,.,..,,,,,...,,,,,,,,,,.,....,,,..,,,,,,,...,,,
(t ========: : : : : : : :
,,--- --------- ---
15 0~ ' ............
- -............
\...,:,,, - _________
- - - - - - - - - -_
l2'\5~ '\ , , -...., ............
......-.. _____ _
''' " "''''''''-
\ "''''"''''''''
''''''''' X' 5
' '''''''' f
Ill/////////
/ ~ II I/Ill/I////
For 6x = 1, the graph crosses the x-axis at
about x = 11, and for 6x = 0.1, the graph
The solution containing (1 , 8) crosses the crosses the x-axis at about x = 28.9.
x-axis near x = 7, converges asymptotically
b. Table in part (a), 6x = 0.1
to the y-axis as x approaches zero, and is
Graph in part (a) shows different pattern.
symmetric across the x-axis. The solution
con taining ( 1, 12) goes to infinit y as x goes to c. The accuracy far away from the initial con-
infinity. dition is very sensitive to th e size of the
increment. For instanc e, in part (a) the first
c. Graph, initial condition (1, 10). step took the graph so far down that it crosses
The so lution containing (1, 10 ) behaves more the x-axis before running off the edge of the
like the one containing (1 , 12), although a grid. The greater accuracy with 6x = 0.1
slight discr epancy in plotting may make it shows that th e graph actually does not cross
seem to go th e other way. the x-axis before x = 20.
RS. a. Table, initial condition (1, 9), 6x = 1 d. Graph crosses the x-axis close to x = 28 .9.
720 Appendix
B:Answers
to Selec
ted Prob
lems
R6. Predator-Prey Problem purely algebraic methods seems to give the correct
dy - 0.5(x - 6) value of the limit of the Riemann sum.
a -= --- --
. dx (y - 7)
5. Graph, showing the upper sum Us.
dy = 0 when x = 6 and dx = 0 when y = 7.
So the stab le point is (6 , 7), corresponding to
the present population of 600 Xalto s natives and
7000 yaks.
b. Graph, initial condition (9 , 7).
/
/ ,,...__________
Jl-- ---------,,,,,,,,, ,,,,,,,,
/
/
,,,,,,...
,,,,,,...
_______
_______ ,,,,,,,,,\
,,,,,,,,,, 8
6 . Any Riemann sum is bounded by the corresponding
lower and upper sums. That is,
I
\5 Ln :s R,, :s U11
\. ,,,-----/////
By the definition of integrability, the limits of Ln
and U11 are equal to each other, and to the definite
integral. By the squeeze theorem, then, the limit of
Rn is also equal to the definite integral.
Suddenly there are too many predators for the 7. Definition: J:
J(x) dx = lim .c,x-o Ln = lim.c,x- o U,,
number of pre y, so the yak population declines. provided that the two limits are equal.
Since y is decreasing from (9, 7), the graph Fundamental theorem: If J is integrable on
follows a clockwise path. [a, b], and g (x) = f J(x) dx, thens: J(x) dx =
c. Graph, initial condition (19, 7), in part (b). g (b ) -g(a) .
The graph crosses the x-axis at x "' 14.4, indicat- 8. Numerically, the integral equals 1280.
ing that the yaks are hunted to extinct ion . (The By counting, there are approximately 52 squares.
Xaltos would then starve or become vegetarian!) Thus the integral "' 52(25)(1) = 1300 .
d. Graph, initial con dition (15, 7}, in part (b).
9. M = 0.1: v '( 4) "'- 19.9 (mi / min) / min
The graph never crosses the x-axis, but crosses M = 0.01: v '( 4) "' -19.9999 (mi / min) / min
the y-axis at y "' 2.3, indicating that the yak
population becomes so sparse that the predators , . J(x) - J(c)
10. J (c) = llm x-c ---- or
become extinct. (The yak population would then X -C
exp lod e!) '( ) - li f(x + 6.x ) - J(x)
! X - m .c,x-o 6.x
Problem Set 7 -8 11. v ' (t) = 3t 2 - 42t + 100 ~ v '( 4) = -2 0
CumulativeReview, Chapters I through7 12. Slowing down. v ' (4 ) < 0 ~ velocity is decreasing.
Rocket Problem 13. Graph, showing line of slope -2 0 through (4,208) .
1. Graph, showing strip and sample point. The line is tangent to the graph .
v ( t) dt represents the distance traveled in time dt. v(I)
v(I)
200
(I, v(I)) :- 100
100
0 5 10
2. Definite integral
3. 1280 mi
14. Acceleration
4. = 1280.0384
M100
M1000 = 1280.000384
15. v '( t) = 0 =
t = 10.958 . . . or 3.041 . ..
The Riemann sums seem to be approaching 1280 So the maximum is not at exactly t = 3.
as n increases. Thus, the 1280 that was found by 16. v " (t ) = 6t - 42
Problem
Set7-8:Cumulative
Review,
Chapters
1through
7 721
Compound Interest Prob lem 1
29. g '( x ) = x - 2 /3(4x - 1)
3
dm = km g' (0) is undefin ed becau se 0- 2 13 takes on the form
17. dt
1/ 0 213 or 1/ 0. Graph.
dm
18 . f -m = k f dt ~ ln 1m l = kt + C ~
g (x)
1m l = ekt+c ~ m = C1ekt
19 . Exponentially
20 . General X
-1
21. m = 10,000e 1n (L0 9 H = 10,000 (l.0 9)t
22 . False. Th e rate of increase changes as the amount
in the account increas es . At t = 10,
m = 10,000 (1.09 ) 10 "" 23,673.64. Diff er ential Equation Probl ems
The amount of mon ey would grow by $13,673.64, 30 . Graph, initial conditions: (0, 3) and (10, 4).
not just $9, 000 .
Discrete Data Prob lem
23 . Integral "" 1022
24. By symmetric difference quotient, y' "" 1.75 .
I I I I\\
Mean Value Theor em Prob lem -- ___ ..,__..._,.._,,,,,,
_______ ,, '
\ \ \ \ \
\
,,, \\
25 . See text statement of Rolle's th eore m.
26. Graph . (example)
3 1. Graph, Probl em 30. Any initial condition such as
(2 , 1) for which y = 0.5x gives th e asymptote.
32. x 2 - 4y 2 = 36 or y = 0.5 J x 2 - 36
33. X = 10.5: y = 4.30842 ...
dy 10
34 . At (10, 4), dx = 0.25 = 0.625.
4
Using ~ x = 0.5, y (l0. 5) "" 4 + (0.625)(0.5)
= 4. 3 12 5, which is clos e to th e exact value of
Graphing Problems 4.30842 ....
27. Graph. A lgebraic Techniqu es Problems
d . - 1 3) - 3x 2
35 . -d ( Sill X - ~
X v l - xv6
36 . dy / dx =- sect =-y
1
3 7. Int egral = - 3x l + C
3 ln 14 -
38 . h ' (x) = 5x ln 5
39 . Limit = -4 .5
40. Graph, removable disco ntinuity at (0, - 4.5 ).
5 y
X
28 . Graph . Step discontinuit y at x = 1
-1
,''') X
41. Journal entri es will vary.
4 2. Journal entries will vary.
722 Appendix
B: Answe
rstoSelected Problems
CHAPTER 8 Problem Set 8 -2
CriticalPointsand Pointsof Inflection
Exploratory Problem Set 8-1
Cubic Functionsand TheirDerivatives l. max.
l. j(x) = x 3 - 6x 2 + 9x + 3 9
2
f' (x) = 3x 2 - 12x + 9 Graph.
no p.i.
y ,,.,1 ~
I
/1 I /"(~
X 3. plateau
4 f(x)
/ '(x) + 0
X 2
p.i.
g(x) = x 3 - 6x 2 + 15x - 9
g' (x) = 3x 2 - 12x + 15 Graph. ?
I I 5. min.
I I
I l(x)
,.
I f '(x ) undef.
X
X
3 4 no p.i.
f(x)
f "(x) 0 undef. 0
X 2
h (x) = x 3 - 6x 2 + 12x - 3
h ' (x) = 3x 2 - 12x + 12 Graph. 7. no max./min.
l(x)
/ '
/'(x ) +
X 2
I
\ y
I
I I I no p.i .
I
I
,
I + ? +
X 2
2 3 4
9. no max./min.
l(x)
/ '(x )
X 2
Positive derivative ~ increasin g function .
Negative derivativ e ~ decreasing function.
p.i.
Zero derivati ve ~ function could be at a high point
'-.__/ ~
or a low point, but not always. +
d
3. g"(x) = dx (3x 2 - 12x + 15) = 6x - 12
d 11. no max./min.
h "(x) = dx (3x 2 - 12x + 12) = 6x - 12 f(x)
/ '(x ) ' undef. '
All the second derivativ es are the same! X
5. Inflection points occur where the first derivativ e no p.i.
graph reach es a minimum. l(x) '-.__/
Inflection points occur where the second derivative / "(x ) undef. +
graph crosses x-axis. X
Prob
lemSet82: CriticalPoints andPoi
ntsofInflec
tion 723
l(x )
13. Number lines and sample graph .
max. min. max.
f(x)
/ ' ~ /+ '
f' X + 0 0 0 X
X -2 3 -1 3 5
p.i. p.i. 19. a. x = - 1, 0, or 1 (critical points for j (x))
f(x) x = 0, -../l.72 (critical points for f' (x ))
.,..--..._ '-._/ .,..--..._
f" X 0 + 0 b. The graph begins after the !-critical point at
-1
x = - 1; th e f' -critical point at x = --../l.72 is
X 2
shown, but hard to see.
c. f' (x) is n egative for both x < 0 and x > 0.
l(x)
21. a. x = 1 (critical point for J(x ))
x = 2 (critical point for f' (x ))
b. Since J(x) approaches its hori zonta l asymptot e
(y = 0) from above, the graph mus t be concave
X up for larg e x; but the graph is concave down
-2 - 1 2 3 near x = 1, an d the graph is smooth; somewhere
the concavit y must change from down to up.
c. No. e- x * 0 for all x, so x e-x = 0 x = 0. =
15. Number lines and sample graph.
23. a. x = - 2, and x = 0 (critical points for j( x))
x = 1 (critical point for f' (x); f' (0) is undefined,
max. plateau
so f' has no critical point at x = 0)
f(x)
/+ ~ ~ b. The y-axis (x = 0) is a tang ent line since the
f' X 0 slope approaches infinity from both sides .
X -2
c. No inflection point at x = 0 since concavity is
down for both sides, but there is an inflectio n
no p.i. p.i. point at x = 1.
f(x) '-._/ '-._/ .,..--..._ 25 . a. x = 0 (critical po int for f( x )), (f( l) are und e-
f'1 X + + 0 fined so f has no critical point at x = 1.)
X -2 No critical points for f' (x) .
b . limx--1 f (x) and limx- 1J(x ) are undefined.
/(x) c. Levels off to y = 1 for large x.
2 7. a. Max. (2 .5, 7.6) Min. (0.8, 4.9 ) p.i. (1.7, 6.3)
No global max. or min.
-2
17. Number lines and sample graph. X
min. max. min.
f(x) = -3 x 2 + lOx - 6
/ ~ b. j'(x)
e.p. o~- zero-~o - e.p.
f' X
X - 1
+
3 5
f '( x) = 0 = x =
1
3 (5 ./7) 2.548 .. . or
0.784 ...
no p.i.
j"(x) = - 6x + 10; j"(x ) = 0 = 5
x = - =
3
f(x) .,..--..._ .,..--..._ 1.666 . ..
f" X e.p. 0- zero- 0 e.p. c. Critical and inflec tion points only occur where f,
X -1 3 5 f', or f " are undefined (no such points exist) or
are zero (all such points are found above).
724 Appendi
x B:Answers
toSelected
Prob
lems
29. a. No local max. or min. (Plateau at (-5, -16)?) l(x) (5, 10)
p.i. (- 5, - 16). No global max. or min .
-5
37. Concavity Concept Problem
-20
a. f' (-0.8) = 1.92
J'( - 0.5) = 0.75
f'(0 .5) = 0.75
b. j' (x) = 0.3x 2 + 3x + 7.6; f' (x) * 0 for all x, f' (0.8) = 1.92
confirming that there are no local max. or min., b . The slope seems to be decreasing from - 0.8 to
and refuting the apparent plateau at x = - 5. - 0.5; f " (x ) = 6x < 0 on - 0.8 5 x 5 -0.5, which
j"(x) = 0.6x + 3; j"(x) = 0 = x = - 5 confirms that the slope decreases.
c. Critical and inflection points only occur where f, The slope seems to be increasing from 0.5 to
j', or f " are undefined (no such points exist) or 0.8; f" (x) = 6x > 0 on 0.5 5 x 5 0.8, which
are zero (all such points are found above). confirms that the slope increases.
c. The curve lies above the tangent line.
31. a. Max. (-3,82), (- 1,50), (2,77)
Min. (- 2, 45), (1, 18) 39 . Connections Between a Zero First Der ivative and the
p.i. (- 1.5,45 .7), (0.2, 32.0) Graph
a. The graph may have a minimum or plateau there.
l(x) b. Example: J(x) = - (x - 1) 2 + 4 = - x 2 + 2x + 3
80 c. Example: j(x) = (x - 1)2 + 2 = x 2 - 2x + 3
d . Example: j(x) = (x -1 )3 + 3 = x 3 - 3x 2 + 3x + 2
e. Example: f(x) = 2 for 1 5 x 5 4
X
41. Historical Problem-The Second Derivative Test
-3 2 a. f' (x) = x 4 - 4x 3 + 2x 2 + 4x - 3
= (x + l)(x - 1) 2 (x - 3)
b. f " (x) = 4x 3 - 12x 2 + 4x + 4
b . j'(x) = 12x 3 + 24x 2 - 12x - 24 j' (x)=Oatx =- 1,1,3
= 12(x + 2)(x - l)(x + 1) j"(-1 ) = - 16; j"(l) = O; j"(3) = 16
j'(X) = 0 = X = -2, - 1,1 c. x = - 1: concave down, local maximum
f' (x ) is undefined= x = - 3, 2 x = 1: not concave, can't tell max. / min.
j"(x ) = 36x 2 + 48x - 12 = 12(3x 2 + 4x - 1); x = 3: concave up, local minimum
1
j"(x ) = 0 = x = - (2 -../7 ) = 0.21. . . or Graph, confirming these resu lts.
3
-1.54 ... l(x)
f" (x) is undefined = x = - 3, 2 max.
c. Critical and inflection points only occur where f,
f', or f " are undefined (only at endpoints) or are
zero (all such points are found above) . X
- 1 3
33. Point of Inflection of a Cubic Function
J(x) = ax 3 + bx 2 +ex + d; f' (x ) = 3ax 2 + 2bx + c;
j"(x) = 6ax + 2b ~ j"(x) = 0 at x = - b /( 3a). d . Graphs. The third graph shows that if both
Since the equation for f " (x) is a line with nonzero f' (x) and f " (c) are 0, there could be either a
slope, f"(x) must change sign at x = - b /( 3a ), maximum, a minimum, or a plateau.
so the concavity changes and - b / ( 3a ) is indeed a
point of inflection . l(x)
max.
35. Equat ion from Critica l Points
1 9 15 5 min. X
j(x) = -2x 3 + 2x2 - 2x - 2 X
Graph, with max . (5, 10) and p.i. (3, 2) C
Prob
lemSet8-2:CriticalPoints
andPoints ofInflection 725
43. A Pathological Funct ion 17. Cup Problem
Not selected . a. r = 1/43. 75 = 3.5236 ... , h = 1/43. 75 = r
Minimal cup has r ""3.52 cm, h "" 3.52c m
Problem Set 8-3
b. Ratio is d : h = 2r : h = 2 : 1
Maximo and Minima in Plane and Solid Figures
c. Saving "" $754, 000 p er year. Proposals vary .
1. Divided Stock Pen Problem:
Make the total width 150 ft and length 100 ft. (Note: 19. Rectangl e in Sinusoid Problem
The maximum area was not asked for.) Maximum area "" 1.1 2219 . .. at x "" 0.86033 ....
3. Two Field Problem 21. Triangle Und er Cotangent Problem
a. Domain: 20 :s;x :s;93.333 ... a. Limit = 1/ 2
b. A(x) = 22500 - 450x + 4.25x 2 Graph . b. Maximal area does not exist, but the area ap-
proaches 1/ 2 as a limit as x approaches 0.
A (x )
20,000 23. Rectangle in Parabola Problem
~ a. Max. rectangle has width = 2 .)3, length = 6.
b. Max. rectangle ha s width = 2, length = 8.
X
c. No. The maximum area rectangle is 2.)3 by 6.
20 93.3 The maximum perimeter rectangle is 2 by 8.
25. Cylind er in Spher e Probl em
c. Greates t area = 17522.222 ... "" 17,522 ft 2 a. V(x) = 2rr x 2.J l00 - x 2
5. Open Box No. 1
a. Max. at x = J4o = 6.324 ... , z = J4o / 2 b. Max. cylind er has radius =
10
;6= 8.1649 ... ,
3.162 ... Make the box 6.32 cm square by 3.16 . 20 .)3
cm deep. altitude = - - = 11.5470 . .. ,
3
b. Conjecture: An open box with squ are base and 4000rr .)3
fixed surface area A will have maximal volum e and volwn e = = 2418.39 ...
9
wh en the base len gth is twice the height, which c. Altitude = Radius ../2, Ve = V5 I .)3 .
occurs when x = .JA / 3.
7. Open Box No . 3
27. Cylind er in Cone Probl em
Minimum cost is $523.47 . a. Max. lat era l area at radiu s x = 2.5 cm .
9. Short est Distance Problem b. Max. total area is with the de generate cylind er
Closest point to the origin is consistin g only of the top and bottom , ra dius 5
(x,y) "" (- 0.4263,0.6529 ). and al titud e 0.
11. Ladder Problem 29 . Elliptica l Nose Cone Prob lem
Shortest ladder has lengt h 5Js"" 11.18 ft. Max. volum e is 32 rr .)3 = 174.1 ... m 3 at X = J'[7 =
5.196 ... m, and y = .) 32 / 3 = 3.265 . .. m .
13. Rotated Rectangle Problem
Max. volum e with rectangle 400 mm wide (radius), 31. Not selected.
200 mm high.
33. Not selected .
15. Tin Can Problem
a. Let r = radius; h = height.
Problem Set 8-4
V = 141.2185rr = 443.65 ... cm 3
Area of o Plane Region
b. A= 2rr (141.2185r - 1 + r 2)
c. Min. at r = 1/70.60925 = 4.1332 ... , Plan of attack for area probl ems:
h = 2 1/70.60925 = 8.266 ...
Radiu s "" 4.1 cm, altitude "" 8.3 cm Do geometry to get dA in terms of sam pl e point
Since altitude = 2 x radius, altitude = diameter. (x,y).
So minimal can is n eith er tall and skinny nor Do algebra to get dA in terms of one variab le.
short and fat.
Do calculus to add up the dA's and tak e the limit (i.e.,
d. Norm ally-proport ion ed can is taller an d thinn er
int egra te).
than minimal can.
1.465 ... "" 1.5% of metal wou ld be saved.
1. Graph. A = J15 (- x 2 + 6x - 5) dx = 10 2
e. Saving is about $6.4 million! 3
726 Append
ixB:Answers
toSelec
tedPro
blems
y
X X
2
4 2 1
3. Graph . A=f (-y + 5y - 4)dy = 4
1
2 13. Graph . A= f! 1 (x 3 - 3x 2 + 4) dx = 6~
X
-2
5. Graph. A = f_4 1 (- x 2 + 3x + 4) dx = 20 5
6 15. Wanda: You can always tell the right way because
the altitude of the strip should be positive. This will
happen if you take (larger value) - (smaller value).
In this case, if you slice vertically it's line minus
curve (see graph).
X
y
7. Graph . A= f! 2 (- l.5x 2 + 6)dx = 16
For curve minus line, you'd get the opposite of the
right answer. Note that if you slice horizontally it
would be curve minus line.
X
17. Parabolic Region Prob lem
(Proof not selected.)
Area of region= ~ (20)(60) = 800
9. Graph. A= ft(2e 0 2x - cosx) dx = 18.1417 .. .
19. Ellipse Area Problem
y A= es
" 2
5) 225 - 9x
2 dx :::o 47.123889 ...
Conjecture: A = nab, where a = semimajor axis
and b = semiminor axis .
This is consistent with the area of a circle with
X
radius r, in which a = b = r and the area = rrr 2 .
21. Golf Course Problem
Approximately 37.l yd 2 , counting squares.
1 23 . Curve Sketching Review Problem
11. Graph. A= fi4 (- y 2 + Sy - 4 ) dy = 4
2 Graph.
Prob
lemSet8-4:Area
ofa Plane
Reg
ion 727
t(x ) 21. Genera li zed Wedge Problem
2h rr ~~~ 2
V = - Jo x .Jr 2 2- x dx = - r 2h
10
r 3
23. Sphere Probl em
4
X a. V = TT(l000 ) cm 3
rr 2rr 3n 4rr
3
b . Formula: V = 4 TTr 3 = 4 TT10 3 = 4 TT(1000) cm 3 ,
t ' (x ) = 0 =
x = TT+ 2TTn = ... , TT,3TT,5TT,...
Q.E.D.
3 3 3
t ' (x) do es not change signs . (plateau points)
25. Volum e of an Elli psoid Probl em
Problem Set 8-5 4
\I = - TTabc
Volumeof o Solid by PlaneSlicing 3
Note that the volum e formu la for a sphere is a spe-
Plan of attack for volum e probl ems: cial case of the volum e formula for an ellipsoid in
which a = b = c = r, the radius of the sp h ere .
Do geometry to get dV in terms of sample point
(x ,y ). 27 . Sub marine Problem
Do algebra to get dV in terms of one var iabl e. a. Graph .
Do calculus to add up the dV's and take the limit (i.e.,
integrate ). y
1. Paraboloid Problem
10
X
a. V = 40.5TT = 127.2345 . .. V
100
b . V "" 127 .2345 .. . , whic h checks.
c. Ins crib ed cone has volum e 27 TT.Circumscribed
cylind er has volum e 81 TT.
3. V = 30.6TT = 96 .132 .. .
b. 100m lon g, or about as long as a football field.
TT 2
5. V = (e - 1) = 10.0359 . .. c. Maximize D(x) = 2(2x 0 5 - o.02x1. 5 )
2
D'( x ) = 2(x - 05 - 0.03x 0 5 ) = 2x - 05 (1 - 0.03x)
7. Washer Slices Problem
64 D ' (x) = 0 <=:> X = _l_
V = TT= 67 .0206 ... 0.03
3 D(O ) = D (lOO) = 0,
Circums cribed cylind er minus th e cone has 80 v'3
volum e = TTr2 h - TTr 2 h / 3 = 128TT/ 3 D ( l00 / 3) = - - = 15. 396 0 .. .
9
= 134.04 12 ... > V . .D(l 00 / 3) is a maximum, which is confirmed by
Num er ical int egra tion gives V "" 67 .0 206 .. . . th e graph in part (a).
9. V = TT(-50 oe - 1.6 + 480.8 ) = 1193.3394 . . . Maximum diameter is (80v'3) / 9 "" 15.40m, 33m
11. Not selected. from bow.
13. Different Ax is Probl em N o. 1 d. Sub will be fairl y slow.
Inner radius is 3 - x, outer radius is 3. 1
e. V = 3333 m3
V = 24TT= 75.3982 . . . 3 TT"" 10472
15 . New Int egra l Problem No . 1 f. Disp lacement "" 10,912 metric tons
a., b . V = M TTsin
2 2
x dx"" 1.354448 .. .
Problem Set 8-6
TT 12 TT .
c. V = J
2 0
(1 - cos2x) dx = 0.6TT-
4 sm2.4 = Volumeof o Solid of Revolutionby CylindricalShells
1.354448 ... , whic h agrees. 1. a. d \! = 2TTX y dx = 2TT(4x - x 3 ) dx
17. Pyramid Prob lem b . 0 = 4 - x 2 = (2 - x) (2 + x) at x = 2.
V = 32 0 cm 2 , whi ch is 1/3 of 960. \I= 8TT = 25 .1327 .. .
19. Tr ian g ular Cross-Section Probl em c. V = 8TT = 25.1327 ... , which is the same answer.
1 3. V "" 268.606 1 .. . (exactly 85.5 TT)
a. V = .4 4 2 2 = 4.79821 . . .
4 Circumscrib ed hollow cylinder has volume 329 .8 ... ,
b. Volume would doubl e, to = 9.5964 .. . a reasonab le upper bound for calcu lated volume .
728 Appendix
B:Ans
werstoSelected
Problems
5. V "' 201.06 19 .. . (exactly 64 IT) 3. a. Graph.
Circumscribed hollow cylinder has volume 301.5 ... ,
y
a reasonable upper bound.
7. V "' 36.4424 ... (exactly l l.6IT)
10
Circumscr ibed hollo w cylinder has volum e 65.9 .. . ,
a reasonable upp er bound for calculated volum e.
X
9. V "' 217 .82 54 . . . (exactly 69.336 IT) 0 1.5
Circumscribed cylinder has volume 226.1 ... , which
is a reasonable upper bound for calculat ed volume.
b . L "' 14.4 394 .. .
5
11. V "' 458.1489 ... (exactly 145 IT) c. L "' 14.4488 . .. num erically.
6
Circumscribed hollow cylinder has volume 769.6 ... , 5. a. Graph.
a reasonable upp er bound for calculat ed volume.
3
13. V "' 161.5676 .. . (exactly 51 IT)
7
Circumscribed hollow cylinder has volum e 3 76.9 . . . ,
a reasonable upp er bound for calcu lat ed volume.
X
15. V "' 390.1858 ... (exactly 124.2IT) -1
Circumscr ibed hollow cylinder has volume
1055.5 .. . , a reasonable upper bound .
17. V "' 163.8592 ... b. L "' 15.8617 ...
Circumscribed hollow cylinder has volum e 3 16.1 . . . , c. Low point is (2 .5, -3.25) . Chords from (1, -1) to
a reasonable upp er bound for calcul ate d volum e. (2.5, -3.25) and from (2.5 , -3.25) to (6, 9) have
length 15 .4 ... , a reaso nable lower bound for L.
19. V = 36.4424 ... (exactly ll.6IT), which agrees.
7. a. Graph, y = 16 - x 4 , x in [- 1, 2]
21. V = 19.2IT = 60 .3185789 ...
Rs= 19.3662109 .. . IT = 60.8407460 ...
R 100 = 19.2010666 .. . IT = 60.3219299 .. .
R 1000 = 19.2 00010 6 . .. IT= 60.3186124 .. .
Rn is approaching l 9.2IT as n increases .
23 . Parametric Curv e Problem
X
a. dV = 180IT cos 2 t sin t dt -1 2
V = 60 IT = 188.4955 .. .
b . V "' 188.4955 ... (exactly 60 IT), which agrees
with the volum e found in part (a). b. L "' 18.2470 .. . (L "' 256.6585 .. . for [O,4] .]
c. V "' 2072 .6169 ... (exactly 210 IT2 ) c. Chord s from (- 1, 15) to (0, 16) and (0, 16 ) to
( 2, O) have lengt h 17.5 ... , a reasonable lower
Problem Set 8-7 bound .
Lengthof a Plane Curve-Arc Length 9. a. Graph.
1. a. Graph.
y
1 \ / X
e
X
0 2
b. L "' 7.6043 ...
c. Chords from x = 0.1 to x = 1, and x = 1 to
b. L "' 6. 7848 .. . x = e, have length 7.3658 .. . , a reasonable lower
c. L "' 6. 7886 ... numerically bound .
Set8-7:Length
Problem ofa Plane
Curve-Arc
Length 729
11. a. Graph .
10
5
0 1.5
b. L = 8J8 - 5,/5 = 11.4470 . ..
c. Distance between endpoints is JTio =
b. L "" 14.4488 . . .
11.4017 . . . , which is a reasonabl e lower bound .
c. Distance between the endpoints is 14.1809 ... ,
which is a reasonable lower bound for L. 2 1. Golden Gat e Bridge Problem
3
Construct an x-axis at water level and y-axis
13. a. Graph, x = 5 cos 3 t , y = 5 sin t, tin [O, 2rr]. through the verte x of the parabola.
5 y Equation of parabola is y = ~~ x 2 + 220.
44 00
L ""43 72.0861 .. . ""43 72 feet
X
Th e an swer is reasonable since the 4,200 feet be-
5 tween support s is a lower bound for L.
23. Stadium Probl em
Out er ellips e: L ""6 92. 5791 . .. "" 692.6 m
Inner ellips e: L ""484.4 224 ... ""484.4 m
b. L = 30 (exactly!) 25. Impli cit Relation Probl em No . 1
2
c. Circle of radius 5 has circumfer enc e 31.415 2 . .. , L = 4 = 4.6666 ...
which is close to the calculated value of L.
3
27. Spiral Problem
15. a. Graph . Spiral is generated as t goes from O to 7rr.
y L = _!_
TT
r7rr
Jo
.Ji+t2 dt = 77.6508 . . .
29 . Sinusoid Length Investigation Problem
X
A L
0 6.28 3 185 . .. (=2rr )
1 7.640 39 5 .. .
2 10.5407 34 .. .
3 13.9 74417 . . .
b. L = 40 (exactly')
c. Max./ min. values of x, y are 3 J3. Circle of Doubling A doubles the amplitude of the sinusoid .
radius 3./3 has circumfer ence 32.6483 ... , which However, it less than doubles the length of the
is clos e. sinusoid, for much th e same reason that doubling
on e leg of a right triangle does not double the
17. a. Graph. hypot enuse . In th e limit as A approach es infinity,
doublin g A approaches doubling the length.
y
30 3 1. Fatal Error Probl em
Th e function y = (x - 2i- 1 has a vertical asymptote
at x = 2, which is in th e int erval [l, 3]. So the length
is infinite. Mae's partition of the int erval skips over
X
the discontinuity. Graph.
y
25
1
b. L = (145 3 12 - 1) = 32. 3153 .. .
54
c. The chord connecting the endpoints has length
32.2490 ... , a reasonable lower bound for L. X
19. a. Graph.
730 Appendix
B: Answers
toSelected
Prob
lems
33. Not selected . b. The area of the circle is TT 52 = 25 TT.
The calculated area is twice this because the
Problem Set 8-8 circle is traced out twice as e increases from O to
Area of a Surface of Revolution 2TT.Although r is negative for TT < e < 2TT, dA
is positive because r is squared.
l. Par aboloid Prob lem
3 3. a. Calculator graph confirms that the text figure is
a. S = fo 2TTx J l + x 2 dx "" 64.1361 . ..
traced out once as e increases from Oto 2TT.
b. Th e inscr ibed cone has lateral surface area
50.9722 . .. , a reasonab le lower bou n d for S. b . A ""6 4.4026 .. . (exactly 20.5TT)
c. L "" 28.8141 .. .
c. S = ~ TT(lOM - 1) = 64 .1361 . . . , which agrees .
5. a. Calculator graph confirms that the text figure is
3. Ln-Curved Surface, Problem I traced out once as e increases from Oto 2TT.
S "" 9.0242 .. . b . A "" 168.0752 ... (exactly 53.5 TT)
5. Recipr ocal Curved Surface Problem I c. L "" 51.4511 . . . .
S"" 15.5181 . ..
7. a. Calculator graph confirms that the text figure is
7. Cubic Parabo loid Problem I traced out once as e increases from Oto 2TT.
S "" 77.3245 . . . b. A"" 117.8097 ... (exactly 37.5TT)
9. S -- 4TT(1.25 3/ 2 - 0.125 ) -- 5.3304 . .. c. L = 40 (exac tly)
3
155
9. a. Graph makes one comp lete cycle as e increases
11. S =4 TT= 14.4685 . . . from O to TT.
256
13. S = 49.5 TT= 155.5088 .. .
5
15. S = 101 TT = 318.1735 ...
18
17. Sphere Zone Problem
Not selected .
19. Spherical Volume and Surface Prob lem
Pick a sample poin t in the sp herical she ll at radi u s
r from th e center. Surface area at the samp le point b. A "" 0.7853 .. . (exactly 0.25TT)
is 4TTr 2 . Volume of shell is approximate ly (surface
area)( thickness). c. L "" 6.6824 .. .
dV = 4TTr 2 dr 11. Right-hand loop: -TT/4 ~ 0 ~ TT/ 4.
V = I: 2
4TTr dr = ~ TTr
3
1:= ~ TTR3, Q.E.D.
Area of both loops is 49 /2 = 69.2964 . . .
13. Int ersections: e = cos - 1 (2/3) = 0.8410 . . . + 2TTn
21. Para boloid Surface A r ea Prob lem Region ou tside the cardioid and inside the circle is
TT
S = - 2 [(1 + 4a2r2)3 i2 - l] generated as e goes from - 0.841 ... to 0.841 ...
6a A"" 18.886 3 ... (exact ly 26cos - 1 (2/3) - (4/3))5)
23. Elli psoid Problem
S= J;
6TTsin t .J~(
--5-si_n_t ~)2~+~( 3~c_o_s_t~)
2 dt
15. a. L "" 89 .8589 ...
13 3
"" 165 .7930 . .. b. A = TT = 25.1925 .. .
16
25. La tera l Area of a Cone Prob lem
17. Colu mn Scroll Problem
From th e figure in the text, a circle of radius L has
area TTL 2 and circumfere n ce 2TTL. Th e circumfer- a. Graph, r = 50 - 112 , from 0 = 0 to 0 = 6TT.
ence of the cone 's b ase is 2TTR , whi ch is eq ual to
the arc length of the sector of th e circle of radi us L.
Thus th e sector is (2TTR ) I (2TTL) = R / L of the circle,
and h as sur face area
S = TTL 2 (R / L) = TTRL, Q.E.D.
Problem Set 8-9
Lengths and Areas for Polar Coordinates
1. a. A"" 157.0796 . . . (exact ly 50TT) L "" 31.0872 ... from e = TT/2 to 6TT
Problem
Set8-9: Length
s andAr
eas forPolar
Coordinate
s 731
b. Graph, showing sectors of central angles 1, 2, p.i.
and 3 radians. ,,----....
+ undef.
2'
b. Graph, example .
f(x )
A(l) = A(2) = A(3) = 12.5 X
-2 3
A(e) = 12.5, which is independent of the value
of e.
19. LP Record Project c. j (x ) = x 213 - x
Not selected. i. j' (x ) = ~ x - t / 3 - 1, j" (x ) = -92x - 4 /3
21. The Derivativ e dy / dx for Polar Graphs ii. Zooming in shows that there is a local mini-
a. Count 5 spaces to the right and about 7.5 spaces mum cusp at (0, 0 ) and a local maximum with
down from the given point. Slope ""- 1.5. zero derivative at x "" 0.3. Graph.
b. r = e
f(x)
X = 0cose => dx = de cose - esinede 0.5
y = e sine => dy = de . sine + e cos ede
dy dy / d0 sine + ecose
dx = dx I de = cos e - e sin e X
At e = 7, dy / dx = -1.54338 ... , thus confirm- 0
ing the answer found graphically .
Problem Set 8 - 10 j'(x) =Oat x = (2/3 ) 3 = 8/ 27, and j' (x )
Chapter Review and Test is undefined at x = 0, thus locating precisely
the min. and max. found by graph . Since there
Review Problems are no other critical values of x, there are no
RO. Journal entries will vary. other max. or min. points.
Rl. a. Graphs. iii. f" (x) is undefined at x = 0, and f" (x ) < 0
everywhere else; f " never changes sign, so
there are no inflection points.
iv. J(O ) = 0, f (8 / 27 ) = 4 / 27,
f(5 ) = - 2.0759 .. . .
Global max. at (8/2 7, 4 / 27).
X
Global min . at (5, - 2.0759 .. .).
d. Graph. Local min. at x = 0, local max. at x = 2.
Points of inflection at x = 2 .J2""3.4 and 0.6.
b. j'(x) = 3x 2 - 18x + 30; j"(x) = 6x - 18 f(x)
g ' (x) = 3x 2 - 18x + 27; g " (x) = 6x - 18
h ' (x ) = 3x 2 - 18x + 24; h"(x) = 6x - 18
c. h has h ' (x ) = Oat x = 2 and x = 4.
Local max. at x = 2, local min . at x = --L
d. g has a horiz. tan . at x = 3, but no max . or min. 2
e. Each point of inflection appears at x = 3, where
second derivative 6x - 18 equals zero. R3. a. Storage Battery Problem
R2. a. Number-line graphs. Optin1albatteryhasc ells P 0/12 = 2.415 2 ... cm
wide and 10) 12 / 70 = 4.1-103 . .. cm long, giving
no max. a battery of overall dimensions about 14.5" by
f(x) or min.
4.1", which is longer and narrower than the typ-
f '(x ) undef.
ical batt ery, 9" by 6.7". Minimal wall length does
X
not seem to be a major consideration.
732 Appendix
B:Answers
toSelected
Problems
b. Cylinder In Cubic Paraboloid Problem CHAPTER 9
Max. rectangle has x = Vl6 / 5 = 1.47 36 ... ,
y = 8 - 16/5 = 4.8. Exploratory Problem Set 9- 1
Introductionlo the Integralof a Productof Two Functions
R4. a. A = e 2 - e = 4 .6707 ...
b. A = 6.75 1. V ::::3.586 4 ...
c. The graphs cross at x = O; for -1 < x < 0, 3. ff'
(x) dx = f x cos x dx + f sin x dx
x 3 > x, but x 3 < x for O < x < l, so the region 5. V = rr 2 - 2rr
to the right cancels the one to the left! Mr. Rhee
should use: 7. The method involves working separately with the
A= J~ 3 3 different "parts" of the integrand . The function
1 (x -x) dx + M(x - x )dx
x sinx was chosen because one of the terms in its
derivative is x cos x, which is the original integrand .
See Section 9-2.
Problem Set 9-2
Integrationby Ports-A Woy to IntegrateProducts
1. - x cosx + sinx + C
1 1
3. - xe4 x - - e4x + C
4 16
21 - 1
RS. a. V = 2.Srr(e 1 6 - 1) = 31.04 70 ... 5. --e - ,x - - xe - 5x + C
25 5
2
b. V = rr = 0.6981. ..
9 7. -1 x 4 1nx- - 1 x 4 +C
4 16
c. Oblique Cone Problem 9. x 2ex - 2xe x + 2ex + C
V "" 25.1327 ... (exactly 8rr)
The right circular cone also has volume 8rr. 11. xlnx-x +C
R6. a. V "" 1.2566 . .. (exactly 0.4rr) Problem Set 9-3
b. V = 0.4rr, which is the same . Rapid Repeated Integrationby Ports
c. Various Axes Prob lem
1. -1 x3e2x _ -3 x 2e2x + -3 xe 2x _ -e2x
3 .+C
i. V"" 25.1327 ... (exactly 8rr) 2 4 4 8
ii. V "" 160 .8495 ... (exactly 5 l.2 rr ) 3. - x 4 cos x + 4x 3 sinx + 12x 2 cos x - 24x sinx
- 24cosx + C
iii. V:::: 174.2536 ... (exactly SSft rr)
1
iv. V "" 201.0619 .. . (exactly 64rr) 5. x 5 sin 2x + x 4 cos 2x - ~x 3 sin 2x- ; x 2 cos 2x
15 . 15
R7. a. L = J~1 .JI+ (2x) 2 dx "" 6.1257 . .. + xsm2x + cos 2x + C
4 8
2 .
b. L = _ (21.25 312 -1) = 28.7281 .. . Distance be-
6 75 7. - ex cosx + ex sinx + C
tween the endpoints is ) 9 2 + 27 2 = 28.4604 ... ,
so the answer is reasonable . 5 3 3 3
9. e x sin Sx + e x cos Sx + C
C. L ::::25 .7255 ... 34 34
11. I_x 8 ln 3x - ...!._
x8 + C
RS. a. S = ;; (145312 - 1) = 203 .0436 . . . 8 64
The disk of radius 8 has area 64rr = 20 1.0619 .. . , ln 7
13. - -x 5 + C (ln 7 is a constant!)
which is close . 5
b. dL = ~ dx 2 + dy 2 = ) 1 + sec 4 x dx
15. sin 6 x +C
f5
S = 2rr (tanx + 1) ) 1 + sec 4 x dx "" 20.4199 ...
C. S ::::272 .0945 . . . 17. ~x 3(x + 5)3/2 - ix 2(x + 5)5/2 + 16 x(x + 5)7/2
3 5 35
R9. a. L "" 32.4706 . . .
- ~(x + 5) 9 12 + C
_ 7.75 _ 315
b. A - - - rr 3 - 38 .7578 ...
6 19. Sxlnx - Sx + C
ProblemSet 9-3:Rapid Repea
tedIntegration
byParts 733
24 . 2 48
- - Sill X COS X - - COS X +C
105 105
1 2 3 2 3 3
23. x (lnx ) 3 - x (1nx ) 2 + x 2 Jnx - x 2 + C Problem Set 9-4
2 4 4 8
ReductionFormulasand ComputerSoftware
25. :ox 2(x 2 + l )s - 610(x 2 + 1)6 + C
1. Jsin 9 x dx = -i sin 8 x cos x + if sin 7 x dx
27. cosxsinx + x + C 1
3. f cot 12 x dx =- - cotll x - f cot 10 x dx
1 1 11
29.
2 secx tanx +
2 1111sec x + tan x i+ C
J sec 13 x dx 1 11
12J sec
11 11
5. =
1 12sec x tanx + x dx
31. x log 3 x - ln x + C 7. ot selected .
3
33. - cosx + C 9. Not selected .
35. - In lcscx + cot x i+ C 11. Not selected .
37. - ln lcosx l + C 13 . - 51 Sm. 4 X cos X -
4 .
15 sm 2 X cos X -
8
15cos X +
C
39. For th e first int egral , Wanda integrated cos x and dif-
ferentiated x 2 , but in the second integral she plans 1 1
15. - - cot 5 x + - cot 3 x - cot x - x + C
to differ entiate f cos x dx and integrate 2x, effec- 5 3
tively canceling out what she did in the first part. 1 2 x tanx + 2 tanx + C
She will get f x 2 cos x dx = x 2 sin x - x 2 sinx + 17.
3 sec 3
f x 2 cos x dx, which is tru e, but not very us eful!
19. Cosine Area Problem
41. After two integrations by parts,
a. y = cos x is on top; y = cos 3 x is in the middl e;
f ex sinx dx = - ex cosx + eX sinx - f ex sinx dx, y = cos s x is on the bottom .
but after two more integrations,
f ex sinx d x = -ex cosx + ex sinx + ex cosx b. For y = cosx, area "" 2.0000 .. .
-eX sinx + f ex sinx dx. For y = cos 3 x, area "" 1.3333 .. .
Two integrations produced th e original integral with For y = cos 5 x, area "" 1.06666 . . .
the opposit e sign (which is useful), and two more in- c. Ai = 2, A3 = 4 / 3, A s= 16 / 15
tegrations reversed the sign again to give the origina l d. Based on the graphs , the area under cos x should
int egral with the same sign (which is not useful). be greater than that under cos 3 x, which in turn
43. A rea Probl em is greater than the area under cos 5 x. This is ex-
A = - 4e - 3 + 1 = 0.8008 ...
actly what happ ens with the calculated answers:
A1 > A3 > As
4 5. Volume Prob lem e. Graph.
V = 5rr(ln5 ) 2 - 10rrln5 + Brr = 15.2589 ...
y
47 . A reas and Int egration by Part s Y=C OS 1oo X
For int egration by part s, Ju dv = uv - fv du.
Applying limits of integration gives
Jcd U d V = UV Iuu =b
=a - Jab V d U X
f! u dv = (bd - a c) - s: v du
-0. 5rr 0.5rr
The quantity (bd - ac ) is the ar ea of the "L-shaped"
region, which is th e area of the larger rectangle mi-
nus the ar ea of th e smaller one . Thus, the integral of f. Yes, Jim,,_"" C'~;2 cos" x dx = 0.
u dv equals the area of the L-shaped region minus 21. Integral of Secant Cubed Problem
the area repres ent ed by the integral of v du . 1 1
J sec 3 x dx =
2 sec x tan x +
2 1n I sec x + tan x I + C
49. Introduction to Reduction Formulas Problem
Note that the answer is half the derivative of secant
f sin 7 x dx = - sin 6 x cos x + 6 f sin 5 x cos 2 x dx
= - sin 6 x cos x + 6 f sin s x dx - 6 f sin 7 x dx plus half the integral of secant.
J sin 7 x dx = - .!.
sin 6 x cos x + ~ f sin s x dx 23. f sin 3 ax dx = - ~ sin 2 ax cos ax
7 7 3
7 d l . 5 6, .j 2 .
3 J smaxdx
f Sill X X =- l Sill X COS X -
35
Sill X COS X +
734 Appendix
B: Answers
toSelec
tedProblems
(From Problem 22) 31. Area Problem No. 1
1 . 2 5 3
= - a sm 2 ax cos ax - a cos ax + C a. sin 5x sin 3x + cos 5x cos 3x + C
3 3 16 16
1
= - a cos ax(sin 2
ax + 2) + C, Q.E.D. b . JJrrcos 5x sin 3x dx = 0.
3 Since the integral finds the area above minus area
1 2 below, this calculation shows the two areas are
Or: d~ ( - a (cos ax) (sin ax + 2))
3 equal.
1 2
= - a (-a sin ax)(sin ax+ 2) 33. Volume Problem No . 1 V = rr 2 / 2
3
35. L ima <;on Area Problem
- ~ (cos ax)(2a sin ax cos ax) 25 .
3 A = TT+ 2 + 10 v'2 + rr = 29 .1012 ... , which
3 8
= (sin ax + 2 sin ax - 2 sin ax cos 2 ax) agrees with the numerical answer.
= (sin 3 ax + 2 sinax(l - cos 2 ax)) 3 7. Not selected.
Problem Set 9-6
= (sin 3 ax+ 2 sinax(sin 2
ax) )= sin 3 ax.
Integrationby TrigonometricSubstitution
..Jsin 3 ax dx = - ~ (cos ax) (sin 2 ax + 2) + C,
~
3 9
1. ~ sin - 1 + x ) 49 - x 2 + C
QED.
Problem Set 9-5 3. x .Jx 2 +16+8ln l .Jx 2 +16+x l +C
IntegratingSpecial Powersof TrigFunctions
2 1
5. x .J9x 2 - 1- ln I 3x + ) 9x 2 - 1I + C
1. - cos X + -3 cos 3 X - - COS S X + C
5 ,
7. sm -
. l X
v'17+C
3 .!. sin 9x - .!. sin 3 9x + J__ sin 5 9x - J__ sin 7 9x + C
. 9 9 15 63 9. In .Jx 2 + 1 + x
J J +C
5. \ sin s 3x + C 11 . .!.x 3 .Jx 2 - 9 - ~x .Jx 2 - 9
1
4 8
l - 1 81
7. - cos 1 Sx + cos 9 Sx + C -
8
In Ix + Jx 2 - 9 +CJ
56 72
1 2 1 1
9. --cos 3 x+-cossx--cos 7 x+C 13. x(l - x 2 ) 3 ' 2 + sin - x + x .J l - x 2 + C
3 5 7
1 X + 1 Sill
11. Z . 2X + C 15. tan - 1 ~ + C
4
1 1 . 0x + C 17. a . .Jx 2 + 25 + C
13 . -x- -sml xdx 1
2 20 b. f --;===c==== - f(x 2 + 25) - 112 (2x dx)
1 Jx + 2'.:> 2
2
15. tan 3 x + tanx + C = Jx 2 + 25 + C, which agrees with part (a).
3
Moral: Always check for an easy way to integrate
17. -
1
42
cot 7 6x - :ocot 5
6x - cot 3 6x - cot 6x + C before trying a more powerful technique!
x - 5
1 11
19. sin - 1 - - + C
19. - tan x +C 3
11
21. In x + 4 + .Jx 2 + Sx - 20 + C
J J
21. 1 sec 10 x + C
10 23. 50 sin - t 0.8 + 25 sin(2 sin - 1 0.8)
23. (sec 10 20)x + C - 50 sin - r (- 0.3) - 25 sin(2 sin - 1 (- 0.3))
= 99.9084 ...
25. sin2x + C Numerical integration: 99.9084 ... , which checks .
25. Arc Length of a Parabola Problem
27. -cotx+C 5 1
1 1 L = 2J901+ 12 1n IJ90T+ 30 I = 75.3828 . ..
29.
2 secxtanx +
2 ln I secx + tanx l + C Numerical integration: L = 75.3828 . . . (check.)
Prob
lemSet 9-6:Integrat
ionbyTrigonomet
ricSubstitut
ion 735
27. Circle A r ea Formu la Probl em
Not select ed. ~0;;;;- ; ///;;; ; ;
Ill// !Ill/II/I/
I I I I I I I I I I I I I I I I
29 . Ellip soid Probl em I I I I I I I I I I I I I I I
I I I I I I I I I I I I I I I I
Rotatin g about th e y-axis, V = ~ rr a 2 b . I I I I. I I I I I I I I I I I I
I I I I I I I I I I I I I I I I
///,//II/I/Ill/I/
Rot atin g ab out th e x -axis , V = ~ rr ab 2 . / .....::/ /2/ / / /4 / / / /6/ / / /t
3 1. Hy perboloid Pro blem
2 6
V = ~ rr = 268. 0825 ... 21. A r ea Problem
b- 1
33 . A r ea of an Ell ipse, Param etri cally A = 5 ln b + + 5 In 6
4
A = rrab , as in Probl em 28.
6
With this m ethod, you get f sin 2 t d t , just like with A(7) 5 ln
=
11 + 5 1116 = 5.928 1 ...
tri g substitution , but h ere you get th e inte gral
directl y, not indi r ectly. lim b-ooA( b ) = lim b-oo 5 ln t + 5 ln6
3 5. Trig Substitution for Negativ e Valu es of x (!'Hos pital' s rul e)
For th e sin e and tang ent substitution, the range = 5 1116= 8.9587 . .. ,
of the inverse sine and inverse tang ent mak e th e so th e ar ea do es approach a finite limit.
correspondin g radical positiv e. For th e secant sub-
stitution th e situation is mor e complicated but still 23. Equival ent A nswers Pr obl em
gives an ans wer of th e sam e algebra ic for m as if x 1 1
a.
had b een onl y po sitive . 2 ln Ix - 2 1+
2 ln Ix - 4 1+ C
b. In Jx 2 - 6x + 8 + C
Problem Set 9 -7 1
c. ln lx 2 - 6x + 8 1+ C
Integrationof RationalFunctionsby PartialFractions 2
1. 4 ln lx - 11+ 7 ln lx - 2 1+ C d. Each can b e tra n sfo rm ed to ln Jx 2 - 6x + 8 + C.
7 3
3.
2 ln lx + 2 1+
2 ln lx - 4 l+ C
Problem Set 9 -8
5. - 7 ln lx + 51+ 7 ln lx + 2 1+ C Integralsof the InverseTrigFunctions
7. 2 ln Ix + 11+ 3 ln Ix - 71+ 4 ln Ix + 2 1+ C 1. Not selected .
9. - ln Ix + 3 1+ 2 ln Ix + 11+ 3 ln Ix - 2 1+ C 3. Not selected.
3 5 2
11. x +
2x -7x + 2 ln lx-l l+ C 5. Not selected .
1 2
7. A nswer Verification Probl em
+ 11+ 2 tan - 1 x + 3 ln lx +4 1+ C
13 .
2 ln lx -)
4 tan 4 -
rr
- 1 ln 17 _- 3.44 78 .. . (checks)
15. ln lx + 51+ 3 ln lx + 11+ 2(x + 1) - 1 + C
4 2 2
9. A rea Probl em
17. - 1 (x - 2) - 2 + C Both m ethod s give A = 1.
2
19. Rumor Probl em
1000 Probl e m Set 9-9
a. Y = 1 + 99e - 2 t Calculusof the Hyperbolicand InverseHyperbolicFunctions
b . y( l ) = 69.453 1 . . . :::::69 stud ent s.
y(4) = 967 .856 7 ... :::::968 stud en ts. 1. Hy per boli c Functi on Gra phin g Probl em
y(8) = 999 .9888 . . . :::::1000 stud ents - everyon e
I y
knows by th e end of th e day' I y
\
1 cash, ,
\ I ' I
\
c. y " = (lOOOy ' - 2yy ') is a ma x. wh en ,'
500 f
X
y = 500.
This occurs wh en t = ln 99 = 2.29 75 .. . hour s. sinh
'I
coth\
d . Graph , which follo ws th e slop e field pat tern .
736 Append
ix B:Ans
werstoSelectedProb
lems
y 35. Derivations of the Pythagorean Properties of Hyper-
bolic Functions
X
Not selected .
' 37. Hyperbolic Radian Problem
'' I a. On the circle,
I
csc h,
rl du _ 1
~ = - COS
I
L= Jcos2 1+ 2 = 2
vl - u2
On the hyperbola, L = cosh 2 - 1 = 2. 762 ... . So
3. 3 tanh 2 x sech 2 x
the length of the curve is greater than 2, Q.E.D.
1
5. cosh 6 x + C b. The area of the triangle that circumscribes the
6 sector is 0. 5 ( 2 sinh 2 cash 2) = sinh 2 cosh 2. The
7. - cschx cothx sinx + cschx cos x area of the region between the upper and lower
1 branches of the hyperbola from u = 1 to cash 2
9. +C
4 tanh4x is
11. 2x coth x - x 2 csch 2 x fJ
A= 2 sinh 2 t dt "' 11.644958 ...
Thus the area of the sector is cash 2 sinh 2 -
13. ln(cosh3) - ln (cosh 1) = 1.875547 . . .
11.6449 58 ... = 2, Q.E.D.
5 cosh 5x 1n 3x - x - 1 sinh 5x c. By definition of the circular functions x is the
15
(ln 3x) 2 length of the arc from (1, 0) to (cos x sinx) . So
17 . cosh 1 - sinh 1 = e- 1 = 0.36787 . . . the total arc has length 2x. The circumference of
a unit circle is 2TT,and its area is TT. Thus
19 12 2x
J l6x 2 + 1 Asector= TTTT = X, Q.E.D.
2
1
21. x tanh - 1 5x + 1n 11 - (5x) 2 I + C d. Area of circumscribing triangle = cosh x sinh x.
10 Area between branches= coshx sinhx - x.
Area of sector =
23. 0.5x J x 2 + 9 + 4.5 sinh - 1 ~ + C
coshx sinhx - (coshx sinhx - x)
25. Hanging Chain or Cable Problem Area of sector= x, Q.E.D.
Not selected.
Problem Set 9-10
27. Power Line Problem ImproperIntegrals
1
a. y = 500cosh x + 110- 500cosh0 .3
500 1. a. Graph, y = 1 / x 2 . Might converge. Integrand
y(O) = 87.3307 .. . "' 87.3 ft approaches zero as x approaches infinity.
b. L = 1000 sinh 0.3 = 304.5202 ... "' 304.5 ft
weight= 243.6162 ... "' 243 .6 lb. y
c. T (l50 ) = 400cosh0.3 = 418.1354 ... "' 418 .1 lb.
d. h = 901.3301 ... "'9 01.31b
29. Bowl Prob lem
a. S "' 5.07327 ... "' 5.07 ft 2 X
b. Cost "' $289.18
c. V "' 1.2 531 7 . . . "' 1. 2 53 ft 3
1
31. Derivative Verification Problem b. Integral converges to
2.
a. H'(l) = - cschlcothl = -1.1172855 .. . 3. a. Graph, y = 1/x . Might converge . Integrand
b. H '( l) "'- 1.11738505 ... approaches zero as x approach es infinity.
The answers differ by 0.0000995 ... , which is
about 0.0089% of the actual answer.
33. Integrat ion by Parts Pr oblem
2 1 .
By parts: ex cosh2x - +C
3 3ex smh2x
. 1 3 1
+
By transf. to exponential form:
6e x 2 e- x + C
X
Transforming to exponential form is easier !
Problem
Set9l 0:Improper
Integra
ls 737
b. Integral diverges. 13. a. Graph, y = e - OAx. Might converge. Integrand
5. a. Graph, y = 1 / x 0 2 . Might converge. Integrand approaches zero as x approaches infinity.
approaches zero as x approaches infinity.
y
I 2
b. Integral converges to 2.5e- 0 8 = 1.1233 ...
b. Integral diverges. 15. a. Graph, y = Jx . Does not converge . Inte grand is
7. a. Graph, y = 1 / x 0 2 .
Might converge. Integrand undefined for x < 0.
becomes infinite only as x approaches zero. y
-2 0 2
I X
b. (Not applicable)
b. Integral converges to 1.2 5. 17. a. Graph, y = xe - x . Might converge . Integrand
9. a. Graph, y = 1 / ( 1 + x 2 ).
Might converge . Integrand seems to approach zero as x approaches infini ty.
approaches zero as x approaches infinity. y
0
b. Integral converges to 1.
b. Integra l converges to TT / 2.
19. a. Graph , y = cos x. Diverges. Int egran d does not
11. a. Graph, y = 1/ (x lnx ). Might converge. Integrand approach zero as x approaches infinity.
becomes infinit e only as x approaches O or 1.
y
X
b. (Not applicable.)
b. Split the integral into two pieces. 21. Divergence By Oscillation Problem
I~1 / (x ln x) dx = As b - oo, It
cos x dx oscillates between -1 and 1
It 1/ (xl nx) dx + I! 1/ (x ln x ) dx and never approaches a limit. Similarly sinx dxIt
= lim a- o+ I~ 1/(xlnx) dx oscillates between O and 2.
+ limb- 1-
= oo+oo
I: 1/(xlnx) dx
23. Volume of an Unbounded Solid Problem
For the integral to converge, both limits must a. A - oo. The improper integral diverges.
exist. Since neither exists, the integral diverges. b. V - TT
738 Appendix
B:Answers
toSelected
Problems
C. V - oo corresponding to each continuous piece may be
d. False. The volume could approach a constant as added up separately.
in part (b) or become infinite as in part (c). e. False. Some discontinuous functions (notably,
25. Gamma Function and Factorial Function piecewise continuous functions) are integrable.
a. j(l) = l , j(2) = 2, f(3) = 6
Problem Set 9 -11
b . Conjecture:
j(4) = 4f(3) = 24 = 41 MiscellaneousIntegrals and Derivatives
j(S) = Sj(4) = 120 = 5! 1. y ' = 3 sec 3x tan 2 3x + 3 scc 3 3x
j (6) = 6f (5) = 720 = 51
c. j(x) = xf(x - 1), Q.E.D. 1
3. xsinh4x- cosh4x + C
d. Part (a) shows that j (l) = 1 = l!. 16
Part (c) shows that j (n ) = nj (n - 1) 5. f'(x) = - 3(3x + sJ - 2
= n(n - l)j(n - 2)
1
= n(n - l)(n - 2) . . . (2)(1) = n!, Q.E.D . 7. ln1(3x + 5)1 + C
3
e. ft t 3 e- t dt < 0.000, 001 for b ~ 24.
f. 0.5! "" 0.886227 311 .. . using b = 24. 9. t ' (x) = 20 tan 4 4x sec 2 4x
From the graphs, t 0 5 e- t < t 3 e- 1 for x ~ 24.
11. 1 X -
1 Sll1
. X COS X +C
The error in 0.5! by stopping at b = 24 is the area 2 2
under the "tail" of the graph from b = 24.
13. y' = 23 / (x + 2) 2
Error = J;'.;t 0 5 e- 1 dt < J2: t 3 e- 1 dt < 0.000, 001.
The difference between the tabulated value of 15. 6x-23lnlx + 2l+C
0.5! and the value calculated here is
0.8862269255
-0.886227311 .. .
17. j'(t) = k1+ t
= -0.000000386,
which is less in absolute value that 0.000,001. 19. tv'l+t2
+ ln I v'f+t2 + t I + C
Note, however, that the difference is negative,
21. y' = 3x 2ex +x 3 ex = x 2ex (3 +x)
because the calculated value is larger than the
tabulated value. This observation means that 23. x 3 ex - 3x 2ex + 6xex - 6ex + C
either the tabulated value is incorrect or there
25. j'(x) = (1 - x 2 ) - 112
is more inaccuracy in the numerical integration
algorithm than there is in the error caused by 27. x sin - 1 x + J I - x 2 + C
dropping the tail of the integral. 1 1
g. 1.51 = 1.3293 .. . 29. -
6 ln Ix + 5I +
6 1nIx - 11+ C
2.5 1= 3.3233 .. .
3.5! = 11.6317 .. . 31. tn lx+2+ .Jx 2 +4x - 5 1 +C
h. 0! = J0 t 0 e- t dt = 1, Q.E.D.
00
33. f' (x) = sech 2 x
i. (- 1)! = 01; 0, which is infinite. So (- 2)! and (-3)1,
which equal (- 1 )1/ (- 1) and (- 2)1/ (- 2), are also 35. In I coshx l + C
infinite. However, 37. y ' = e 2x (2cos 3x - 3sin3x)
(-0.5)! = 0.5 1/ (0.5 ) = 1.77245 ...
(-1.5)! = (- 0.5)! / (- 0.5) = - 3.54490 .. . 39. l_e 2x sin 3x + ~ e 2x cos 3x + C
(-2.5)! = (- l.5)!/(-1.5) = 2.36327 ... , 13 13
all of which are finite. 41. g ' (x) = (3x 2 )ln5x+x 3 (5/5x) = x 2 (3ln5x + l)
j . 0.5! = f = 0.886226925 ... (checks)
43 . !x 4
ln5x - J__x 4 + C
27. Piecewise Continuity Problem 4 16
3 2 3
a. fi ydx = f1 (2x + l ) dx + J2 (2x- l)dx 45. y ' = y(x - 1 - (x + 2) - 1 - (x + 3) - 1 - (x + 4) - 1 )
b. = limb-2- Jt(2 x + 1) dx + lima - 2~ J;(2 x - 1) dx 47. -lnlx + 21 +3ln lx + 3 l- 2ln lx + 4 1+ C
c. Integral converges to 6/ ln 2 = 8.6561 .... 49 . y' = - 3 cos 2 x sin 2 x + cos 4 x
d. The integral is defined by dividing the inter-
val into Riemann partitions and adding up th e 51. - -1 cos 4 x + C
subintervals . But the Riemann partitions may be 4
chosen so that the discontinuities are at end- . C
points of subintervals. Then the subintervals 53 . Sll1 X - 3l.3sm X + or 31 cos 2 X .
Sll1 X + 32 Sll1
.
X +
C
ProblemSet911: Misce
llaneous
Integrals
andDeriva
tives 739
55 . 1 cos 3 x smx
.
+ 3 cosx smx
. + 3x + C 101. 2 ( 1 + .fie ) - 2 ln Il + .fie I + C
4 8 8 or2 fi -2ln l l +fil + C1
57. g ' (x) = 12x 3 (x 4 + 3) 2
103. ~ (1 + 1/x)3 - 6 (1 + 1/x)2 + 12 (1 + 1/x)
3 + x 9 + 27 x s + 27x + C
59 . __!_x1 - 4 ln (1 + 1/x) + C or
13 5
61. -(x
1 4
+ 3) 4 + C
1( 1/x)3 - 2 ( ~/x/ + 4 1/x - 4 ln 11 + 1/xI + C1
16
105. ln ( fex+T - 1) - 1n ( fex+T + 1) + C
1
63. x 5 + 3x + C 107. Ratio nal Function ofsinx and cosx by
5
u = tan (x / 2)
65 . j'(x) = (x 4 + 3) 3 a. cosx = 2 cos 2(x / 2) - 1 and
67. e 2 = 7.3890 ... sinx = 2 sin(x / 2) cos(x / 2) .
69. r ' (x) = xe x + ex 1 - tan 2(x / 2)
b . cos x = 1 + tan 2(x / 2 ) ' Q.E.D.
7 1. q'(x) = - l - lnx 2tan (x / 2 )
X2 sinx = ( ) , Q.E.D.
1 + tan 2 x 12
73.
1
+ 2) 2 + C 2du 1- u2 . 2u
2 (lnx c. dx = --
1 + u 2'
cosx = --
1+ u2
and smx = --
1+ u2
75 . f' = 2xe x2 1
(x) d. f dx
1 + cosx
1 x2 C
77 . 2e 1 2du
+
=f 1- u2 1+u2
l+- -
79 . 1 x 2 ex2 - 1 ex2 + C 1 +u2
2 2 2du
= f (1 + u 2) + (1-u 2) = f du, Q.E.D.
b
81 . a 2 + ax . b a ax b C
b 2 e sm x + a 2 + b 2 e cos x + ,
e. tan (x / 2) + C
(for a, b not b ot h 0)
x + C, (for a = b = 0) 109. - cot(x / 2) + C
1 111. - cot(x / 2) - x + C
83 . x - e sin2ex + C, (fore * O); C, (fore = 0)
4
Problem Set 9- 12
, ad - be IntegralsinJournal
85. f (x ) = (ex+ d) 2 , (fore , d not both 0)
(undefined for e = d = 0) 1. Not se lected.
ax be - ad
87. - +
e e2
ln lex+ di + C, (fore * 0) Problem Set 9-13
Chapter Reviewand Test
a b
2dx 2 + dx + C, (fore= 0, d * 0, undef., e = d = 0)
Review Problems
89. Jx 2 + a 2 + C RO. Journal entr ies will vary .
9 1. lnl v x 2 +a 2 +xl+C
Rl. f'(x) = cos x - x sinx
f X sin X dx = sin X - X cos X + C
93. f' (x) = 2x sin ax + ax 2 cos ax f;x sin x dx = sin 4 - 4 cos 4 - sin 1 + cos 1
1 2 . 2 = 1.5566 ... , whic h checks .
95. -- x 2 cos ax + 2 xsmax + 3 cos ax + C,
a a a
(for a* 0) R2. -i xcos2x + sin2x +C
C, (for a= 0)
1 R3. a. x 3 sin2x+ ~x 2
cos2x - ~xsin 2x
97. - coshax + C, (for a* 0)
a 3
f sin hax dx = C, (for a = 0) - - cos2x + C
8
4 4 3 4
99 . x cos - 1 ax - ~ v11 - (ax) 2 + C, (for a* O) b. e x sin 3x -
25
e x cos 3x + C
a 25
TT 1 1 1
c - x 2(lnx ) 2 - -x 2 lnx + - x 2 + C
2 x+C,(fora=O) . 2 2 4
740 Appendix
B:Answers
toSelected
Problems
16 14 IT y
d. V = rr ln 2 - rr = 6.7268 ...
3 9
1 29
R4. a. Jcos 30 xdx = cos 29 x sin x + Jcos 28 xdx IT
2
30 30
1 4 8 X
b. sec 4 x tan x + sec 2 x tan x + tan x + C
5 15 15
1
c. f tan" xdx = - - tan 11- 1 x - f tan 11
-
2xdx,
n- 1 b. f' (xl = IX Iv 9X
1z
Q.E.D - 1
1
c. xtan - 1 5x - In ll + 25x 2 l +C
. 2. 3 l. s C 10
RS. a. Slll X -
3 Slll X + S Slll X +
d. A = 1
1 5 2 R9. a. Graph, J(x ) = sinhx
tan 3 x + tanx + C
b.
5 tan x +
3 b. Graph, g(x) = cosh - 1 x
c. 1 x - 1 sml4x
. +C y y
2 28
1 1
d.
2 secxtanx +
2 In I secx + tanx l + C X
X
e. (tan 9 32)x + C
113
f. A= - - rr + 64 - 36)2 = 57.46 33 . . .
8
9 c. h ' (x ) = - x 2 sechx tanhx + 2x sechx
R6. a. x v'x 2 - 49 - ~ in [x + v'x 2 - 49[ + C
d. f' (x ) = J 25: 2 + 1
b . .!.(x - 5)v'x 2 - lOx + 34 1.
2 e.
3 ln I cash 3x I + C
+ ~ In [ v'x 2 - lOx + 34 + x - s[+ c
f. x cosh - 1 7x - .!.J 49x 2 - 1+ C
7
c . sin - 1 ~
2 + .!.x
2 J l - O.25x + C
2
g. cosh 2 x - sinh 2 x = 1, Q.E.D.
t
d . A = 25 (sin - 1 0.8 - sin - 1 0.6) = 7.0948 ... h. y = 2.5269 ... cash _ ... + 5
2 5269
R7. a. ln lx + l l +5ln lx - 4 l+ C - 2.5269 . ..
y(lO) = 68.5961 ...
b. 3 lnlx-l l +4ln lx + 2 1- 2ln lx -3l+ C y = 20 =:>X = 6.6 324 ...
X
c. 5 ln lx l + tan - 1 +C RlO. a. Converges to 5.
3 b. Integral diverges.
1
d. In lx 2 (x + 4) 3 1+ ~- +C c. Converges to 6.
x +4 10
5 d. Converges to = 3.3 33 .. .
e. y = 3 + 1 + 0.25e o.sx
3
e. ft' x - P dx converges if p > 1, diverges if p :s;;l.
Graph, showing that solution fits slope field.
Rll. a. f' (x ) = sin - 1 x + ~
1- x
I/II Ill YI!//////
b .!.x 2 sin - x - .!. sin - 1 x + .!.x v'l - x 2 + C
1
. 2 4 4
2
c. ex sech ex
1 1
d. - In Ix I +
2 1n Ix - 11+
2 In Ix + 11+ C
I I I I I 'I I I I I I I / e. j' (x ) = - x(l - x 2 ) - 112
f . .!.
2
sin - 1 x + .!.x
2
Jl - x2 + C
1
RB. a. Graph, y = cos - 1 x g. g' (x ) = lnx -
X
Problem
Set9-13:Chapter
Review
andTest 741
h . !x 2 lnx - !x 2 +C
2 4 Dist . = 179 : ft.
5
Rl2. For f (9-x 2 ) - 1 i 2 x dx, the x dx can be transformed 7. v(t) = -6cost - 3
to the differential of the inside function by multi- - 9.42 km (Exact: -3 rr km)
Oispl. = -9.42-17 ... ::::::
plying by a constant. Dist. = 13.5 338 ... "' 13.53 km (Exact: 6)3 + rr)
- 1 f(9 - x 2 )- 112 (-2 xdx) = - (9 - x 2 ) 1i 2 + C,
2
9. Meg's Velocity Problem
a. 4 sec
and thus has no inverse sine. For J(9 - x 2 ) - 1i 2 dx,
transforming the dx to the differential of the in- b. Displ. = 1 ft
side function, -2x dx, requires multiplyirlg by a
variable. Since the integral of a product does not c. Dist. = 4 ft
equal the product of the two integrals, you can't 11. Car on the Hill Problem
divide on the outside of the integral by -2x. So a a. Displ. = 300 ft
more sophisticated technique must be used, in th.is
b. Dist . = 500 ft
case, trig substitution. As a result, an inverse sine
appears in the answer. 13. Subway Problem
f (9 - x 2 ) 115 dx = sin - ~ + C1 a. t end aa vg V end V avg Send
3 sec mph / sec mph mph mi
0 0 0
CHAPTER 10 5 2.95 14.75 7.375 0.0102 ...
10 3.8 33 .75 24.25 0.0439 .. .
Exploratory Problem Set 10-1 15 1. 75 42.5 38.125 0.0968 ...
Introductionto Distanceand Displacementfor Motion Along a 20 0.3 44 43.25 0.1569 ...
Line 25 0 44 44 0.2180 ...
30 0 44 44 0.2791 ...
1. 5.3955 ... "" 5.40 minutes. 35 0 44 44 0.3402 ...
3. 100.0231 .. . "" 100.0 ft upstream 40 -0 .2 43 43.5 0.4006 ...
4 5 -0.9 38.5 40.75 0.4572 ...
5. f~0 llOOetlno.s- 301 dt = 203.6452 .. . "" 203.6 ft 50 - 2.6 25.5 32 0.5017 . ..
55 - 3.5 8 16.75 0.525
Problem Set 10-2
60 - 1.6 0 4 0.5305 . ..
Distance,Displacement, and Acceleration
b. V end = 0 at t = 60 => the train is at rest.
1. a. Positive on [O, 2); negative on (2, 6]. c. The train is just starting at t = O; its acceleration
2 2 must be greater than zero to get it moving, even
b. [O, 2): 14 ft; (2, 6]: 26 ft though it is stopped at t = 0. Acceleration and ve-
3 3
locity are different quantities; the velocity can be
c. Displ. = - 12 ft; Dist.= 41 ft zero, but changing, which means the acceleration
is non-zero.
d. Displ. = 14~ + (-260 = - 12 ft d. Zero acceleration means the velocity is constant,
but not necessarily zero.
Dist. = 14 t + 26t = 41 ft e. 0.5305 ... "" 0.53 mi
e. - 4 (ft/sec)/sec 15. Physics Formula Problem
3. a. Positive on (8, 11]; negative on [l, 8). dv
a. a = dt => v = fa dt = at + C;
b. [l, 8): 4.9420 ... "" 4.94 km; (8, 11]: 4.7569 ... ""
4.76 km v = v 0 when t=0 => C = 0 => v = Vo + at
ds
c. Displ. = -0.1850 ... ::::::- 0.19 km; Dist. = b. V = dt => s = f V dt = f (Vo + at) dt
9.6990 . .. "" 9.70 km
1 2
d. Displ. = -4 .9420 ... + 4. 7569 ... = - 0.1850 . .. "" = v ot + +C
- 0.19 km
2 at
1
Dist. = - (-4.9420 ... ) + 4. 7569 ... s = so when t = 0 => C = so => s = v ot + - at 2 + s0
2
= 9.6990 ... "" 9. 70 km
1T
e. 0.1851 ... "' 0.19 (km / h) / h (exactly J2) Problem Set 10-3
24
Average ValueProblemsin Motion and Elsewhere
5. v(t) = 2 t 3/ 2 - 18; Displ. = - 14 14 ft;
3 15 l. a. Yau= -ll
742 Appendix
B:Answers
toSelected
Problems
b. Graph. The rectangle has the same area as the The average of Vo and v 1 is
shaded region. 1 1 1
(vo + v1) = (v o+ v o+ a (t1 - to))= Vo+ aU1 - t 0 )
2 2 2
l(x)
..vau = the average of Vo and v 1 , Q.E.D.
15. Average Voltage Problem
2A
Average = - ; A = 55TT= 172. 78 ... V.
TT
X Problem Set 10-4
Related Rotes
1. Bacteria Spreading Problem
3. a. Ya u = 2.0252 ...
dr 6 2
b. Graph. The rectangle has the same area as the -d = - = - = 0.6366 ... mm / hr
t TTr TT
shaded region. when r = 3 mm.
g(x)
drldt
3
Y_= 2.0252 ...
1
5. a. Yau = 2 dr mverse
ly wit h t h e ra di us .
6 dt vanes
b. Graph. The rectangle has the sam e area as the
shaded region. 3. Ellipse Problem
Major axis is decreasing at 12 / TTcm / sec.
v(t )
5. Base Runner Problem
Let y = Milt's distance from home plate.
Let x = Milt's displacement from third base.
dy _ - 20x
dt - -Jx Z+ 90 2
dyl dt
10
1 ?
7. Y au = ak "
3 X
9 . Ya u = 'i<a(ek - 1)
11. Average Velocity from Acceleration Problem
v(25) = 120 ft / sec; Displ. = 2 500 ft;
Va v = 100 ft / sec At x = 45, it "" - 8.9 ft / sec (exact: - 4 .Js )
13. Average Velocity for Constant Acceleration Problem
Consider an object with constant acceleration a, for At x = 0, it = 0 ft / sec, which is reasonable since
a time interval [to, t 1]. Milt is moving perpendicular to his line from home
v(t) = fadt = at+C plate.
At t = to, v (t) = Vo ~ Vo = at o + C ~ C = v o - at o
7. Rectangle Problem I
..v(t) = at+ Vo - ato = vo + a(t - to)
Area is decreasing at 40 ft 2 / min.
fi01 (vo + a(t - to)) dt
Vav = 9. Luke and Leia's Trash Compactor Problem
t1 - to
1 1 a. Let L = length, W = width, H = depth (meters)
= - - (voti + - a(t1 - to) 2 - Voto - .!.a(to - to) 2) dH 20 20
t1 - to 2 2 dt = - LW 2 (0.1) - L2W (- 0.3)
1
= vo + a(t1 - to) b. Depth is increasing at 0.02 m / sec
2
Problem
Setl 0-4:Related
Rates 743
11. Point on a Parabola Problem 9. Pipelin e Probl em, Near M iss
dy = 2kx 2 A graph or a table of times for path s clos e to the
a. dt optimum shows that a near miss will hav e virtua lly
b. k = 2 no effect on the minimal cost.
C. 196 11. Robinson Crusoe Probl em
13. Barn Ladder Problem 47.8809 ... "' 47 .9 yd. from the line perp en dicular
a. Let x = distance from bottom of ladd er to wall. to the ship.
Let y = distance from top of ladder to floor. 13. Not selected.
Let v = velocity of the weight.
X dx 15. Journal entries will vary.
V = -;====cc==-
-)400 - x z dt
Problem Set 10-6
b. v = - 0.61 23 . .. ft / sec (exact: -fi; / 4 )
Maximum and MinimumProblemsin Motion and Elsewhere
c. v is infinit e.
15. Conical Water Tank Problem 1. Rocket Probl em
Minimum is D (l ) = 2, or 2,000 mi.
a. 16.2rr = 50.8938 ... "' 50.9 m 3 / hr
-5 0 Maximum is D (3) = 3, or about 3333 mi.
b. i. - = - 0.110 5 ... "' -0. 11 m / hr
1 44' 1T
ii. - 00 3. N umb er Problem I
C. i. !~ = - 0.25./h
X = 0.5.
5. Fran's Opt im al Study Time Problem
ii. - 0.2 m 3 / hr a. Graph, showing maximum of G at t = 3 hours.
iii. - 0.4 317 ... "' -0.43 m / hr
S, G, or 20 F
17. Cone of Light Problem s
Decreasing at 16rr "' 50.3 ft 3 / min.
50 1.1'-' ---....:_ G
19. Cone in Hemisphere Problem
Increasing at 12 in 3 / sec. Ftim es 20
------- -- I
21. Speeding Piston Project [Not selected]
10
Problem Set 10-5
Minimal PothProblems b. Fran should study for 3 hours .
c. i. G(4) is about 1 point less .
1. Swim and Run Problem ii. G(2) is about 1 or 2 points less.
Swim toward a point about 21.8 m downstr eam.
3. Pipelin e Problem 7. Cylind er-in-th e-Cone Probl em I
Go 600 m along road, then start crossing field 400 m a. Graphs of V and A .
up the road from th e well.
VorA
5. Minimal Path Discovery Probl em 600 /
,
/
~ ,/
a. For minimal path, x = 100 / .J21.
,'
.'.sin8 = .J ; = 0.4 = 2/ 5,Q.E.D. V
50 + X 2
b. For minim al path, x = 400. X
10
sine= .J ~ + xz = 0.8 = 40 /5 0, Q.E.D.
300
7. Scuba Diver Problem Revisited b. Maximum V: r = 6~ in., h = 2 in.
12
sine=
X
13
= 30tan (sin - l ~n
Swim 100 - 72 = 28 m, then dive.
= 72
Maximum A: r = IO in., h = 0 in. (degenera te)
The maximum volum e and maximum area do not
occur at the sam e radius.
9. Quartic Parabola Tank Problem
The algebraic solution is easier than befor e because - 0.1286 . .. "' - 0.129 m / min.
no algebrai c calculus needs to be don e.
Mathematicians find general solutions to gain in- 11. Pig Sale Probl em
sight, and to find patterns and methods to allow a. w = 1000 + 15t (lbs ); p = 0.90 - O.Olt ($/ lb)
easier solution of similar problems. A= 900 + 3.5 t - 0.15t 2 ($)
744 Appendix
B:Answers
toSelected
Problem
b. Maximum A at t = 11 !, not a minimum, since
y.
dA
dt goes from pos1t1ve
. . to negative
. t h ere .
c. About $920.42
Problem Set 10-7
VectorFunctionsof Motion in a Plane
1. Parabo li c Path Problem I
a. ii(t) = (6cos0.6t)i+ (-4.8sinl.2tJI
a(t) = (-3.6sin0.6t)i+ (-5. 76cosl.2t
b . i(0.5) = 2.9552 ... { + 3.3013 .. .J
ii (0.5) = 5. 7320 ... {- 2. 7102 . . .J
JI
b . Graph, part (a), showing vectors ii .
C. for i + V, show that ( ; ) \
0
which is the equation of an ellipse .
= 1+ ( (tr i r,
a(0.5) = -1.0638 . . . i-4 .7539 . . ./ d . Graph, part (a).
e. The direction of each acceleration vector is the
opposite of the corresponding position vector,
y
and thus directed toward the origin.
5. Parabolic Path Problem m
-( ) ,- 2 ,- _( ) dx ,- dx ,-
a. r X = Xt + X J; V X = dt L+ 2xdt J
X
b. v(2J = -3i:- 12I
Speed= lv(2) 1 = .JI53 ::e 12.4 cm / sec.
c. Graph, showing i(2) and v(2).
This is reasonable because v(2) points along the
curve to th e left, indicating that x is decr eas ing .
y
a
These vectors mak e sense becaus e the head of i 20 i a,
is on the graph, ii is tangent to the graph, and a
points to the concave sid e of the graph.
c. The object is sp eeding up. The angl e between a
and ii is acute .
X
d. a(0 .5) v(0.5) = 6.7863 . . . , so the angle is acute.
ii1(0.5) = 0.9676 .. . i- 0.4575 .. .J
iin(0 .5) = -2.0314 .. . i-4 .2964 .. ./
Graph, part (b).
e. Object is speeding up at 1.070 3 . . . d . a(x) = 18/. a(2) = 18/. Graph, part (c).
::e 1.0 7 (ft / sec) / sec.
e. a 1(2) = 4.2352 ... i + 16.9411 . . .J
f. i(7) = - 8.71 57 . . . i - 2.0771. . .J an(2) = - 4.2352 . .. i + 1.0588 . . .J
v(7 J = -2.9415 ... i - 4.1020 .. .J a 1 (2) is parallel to the curve. iin(2) is normal to
a(7l = 3.1376 ... { + 2.9911 .. .J the curve and points inward to the concave side.
Graph, part (b). f. When x = 2 the object is slowing down. This is
Object is slowing down . The angle between a and true because th e angle between a(2) and v(2 )
..,;is obtuse. is obtuse, as shown by the graph and by the
g. a(O) v(O) = 0, so a(O) and v(O) are perp en- fact that the dot product is negativ e. Also, iir(2)
dicular . This means the object is neither slowing points in the opposite direction of v(2).
down nor speeding up at t = 0. g. dL = J 1 + 4x 2 dx
dx 5
3. Elli ptical Path Problem
dt = .Ju = 1.2126 ... . ::e 1.21 cm / sec.
a. Graph, showing path of i + ii . 7. 12.0858 . . . ft.
Prob
lemSet l 0-7: Vec
tor FunctionsofMo
tionina Plane 745
9. Baseball Problem 15. Three-Dim ensiona l Vector Problem
a. v(t) = - 130{ - 32t/ v(l) = (8cos0.8)i + (- 6sin0.6 )/ + 3k
a(l) = (- 6.-!sin0 .8){ + (-3 .6cos0 .6)/ - 1.5k
b. r(t) = (- 130t + 60.5){ + (- 16t 2 + 8)/
d (l) . v (l ) = -20.0230 .. . , so the object is
c. The ball passes 4 .5346 ... ft over the plate, which
slowing .
is slightly above the strike zone .
d. At t = 0, dx / dt = 200cosl5 , Problem Set 10-8
dy / dt = 200sinl5 . Chapter Review and Test
As in part (a), v (t) = C1{ + (- 32t + C2)/
= (200cos 15)[ + (- 32t + 200sin 15)/ Review Problems
f(t) = (200tcosl5 )i+ (- 16t 2 + 200tsin 15 + RO. Journa l
3)/ Rl. Popeye and Olive Prob lem
e. Phyllis mak es the home run since the ball is Velocity becom es positive at t = 9 seconds. They
about 41.6 ft. above the wall when x = -!00. have moved 9 ft closer to the sawmill.
11. Figure Skating Problem From t = o tot= 25, displ. = 8 ft, dist.= 26 ft.
a. d = 90 + 150cost
R2. a. i. Graph .
b. 11(1) = -212.1270 .. . [ - 13.79-!8 ... J
Speed = 212 .5750 ... "" 212.6 cm / sec v(t)
C. a (l ) = 76.2168 ... {- 348.5216 .. .I
dr(l) = 53.3266 ... { + 3.4678 .. J 5
dn (l) = 22.8 902 ... { - 351.9894 .. J
Anni e is slowing down at 53.4 392 ... "" 53.4 3
cm / sec. The angle between the acceleration and
the velocity vectors is obtuse, as revealed by the
negative dot product.
ii . Displ. "" - 3.8 cm (Exactly 14/ 1n 2 - 24)
13. Roller Coaster Problem
ill. Dist. "" 10.9 cm (Exactly 2 / 1n2 + 8)
a. v ( t) = (5- 12 cost )i + (- 12sint )J
a(t ) = (12sint)i + (- 12 cost)J
b. Acceleration Data Prob lem
tend a a a,g Vend
b. v(2.5) = l-!.6137 ... { - 7.1816 .. J 2 speeding up
0 30
d (2 .5) = 7.1816 ... { + 9.6137 .. J 5 8 5 55 speeding up
Graph, showing v (2.5) and d(2.5 ). 10 1 4 .5 77.5 speeding u p
y 15 0 0.5 80 neither
20 - 10 -5 55 slowing down
25 -2 0 - 15 - 20 slowing do,vn
Note that the object is speeding up, slowing down
or neither, exactly when a cnd > 0, a end < 0, or
a cnct= 0, respective ly, in the origina l table .
R3. a. Average Velocity Prob lem
X i. Vavg = 2/ rr = 0.6366 .. .
ii. Vavg = 0
iii. Vavg= 0
C. d1(2.5) = l.9791 ... i - 0.9726 .. J b. Averag e Value Problem
d 11(2. 5) = 5.202-! .. . { + 10.586 3 .. J i. Average on [O, 6] is 18.
d. v (2.5 ) is reasonable since its graph points along ii . Graph. The rectangle has the same area as the
the path in the direction of motion . d (2.5) is shaded region .
reasonable since it points toward the concave
side of the path . The roller coaster is traveling
at lv( 2.5) 1 = 16.2803 ... ft / sec. Its speed is
increasing at 2.2 0 5 2 . . . ft / sec 2 , since the scalar
projection of d (2 .5) on ii (2.5) = 2.2052 . ...
e. d(O + 2rrn) = o{ - 12I, pointing straight down .
d(rr + 2rrn) = Oi + 12/, pointing straight up . X
6
f. 78 .7078 .. . "" 78.7 ft.
746 Appendix
B:Answers
toSelected Prob
lems
iii. The average of the two values of j(x) at the
endpoints is zero, not 18.
R4. Rover's Tab lecloth Problem
The glass moves at the same speed as th e tablecloth,
or about 18.7 cm/ sec, which is about 1.3 cm/sec
slower than Rover.
RS. a. Campus Cut-Across Prob lem
T(O) = 145.1612 .. .
T(467.3544 ... ) = 126.7077 ...
T(700) = 127 .7212 . . . iii. Speed = 7.4804 .. . "" 7.48 units /min.
Heading for a point about 467 ft from th e in- a(l). v (l) = 34s inhlcoshl = 61.6566 ...
tersection gives the minimum time, although it The object is speeding up at 8.2423 ... ""
takes only a second longer to head straight for 8.24 units / min 2 .
the English building . iv. 4.5841 .. . "" 4.58 units
b. Resort Island Causeway Prob lem v. r(t) + v (t) = (5 cash t + 5 sinh t){
The minimum cost is $122,000 by going 7.5 + (3 sinh t + 3 cash t)J
km along the beach, then cutting across to the Note that the y-coordinate is 0.6 times the
island. This path saves about $29,600 over the x-coordina te, so the head lies on y = 0.6x,
path straight to the island. one asymptote of the hyperbola .
R6. a. i. Max. acceleration = 9 at t = 3.
Min . acceleration= -4 0 at t = 10. CHAPTER 11
ii. Max. velocity = 36 at t = 6.
Exploratory Problem Set 1 1- 1
Min. velocity = - 33 at t = 10. Review of Work-Force TimesDisplacement
iii. Max. disp lacement = 182 .!.at t = 9. 1. Graph, showing strip and sample point (x, F) .
4
F
Min. displacement = 0 at t = 0.
b. I nf/.ation Problem (x , F)
i. Let t = number of days Saul has been saving.
Let V(t) = real value (in constant day zero i:!.X= 0.2
pillars) of money in account after t days . X
4
V(t) = 50t(O.sooost)
ii. Saul's greatest purchasing power will be af- F = 10.8268 .. . "" 10.83 lb in the strip.
ter about 289 days because V' (t) goes from W = 2.1653 ... "" 2.17 ft-lb
positive to negative at t = 288 . 5390 . . . .
3. Integral= 69.1298 ...
R7. a. i. and ii. Graphs. 5. W = 80 ft -lb
Problem Set 1 1-2
Work Done by a VariableForce
a 1. Leaking Bucket Problem : 800 ft-lb
3. Spring Problem : 50k
5. Conical Reservo ir Problem
1396752.0937 ... "" 1.4 million ft-lb
7. Spherica l Water Tower Problem
b. Lr(ll =7 .7154 ... { +3 .5256 . . ./ a. 117621229 .... "" 117.6millionft-lb
v(ll = 5.8760 ... { + 4.6292 .. ./ b. 250925288.4 ... "" 250.9 million ft-lb
a(l) = 7.7154 . . .F+3.s2s6 . .. J 9. Carnot Cycle Problem
ii. Graph, showing r(l), v(l), and a(l). a. 1504.7320 ... "" 1504 .7 in-lb
Probl
emSet 11-2: Wo
rk Done bya Vari
able Force 747
b. -566.9574 ... So about 567 in-lb of work is don e d. False. The centroid is at (0, 3, 0) but the cent er of
in compressing the gases . mass is at (0, 3.6 , 0) .
C. 937 .7746 ... :::::
937.8 in -lb . 5. Exponentia l Region and Solid Prob lem
d. Not selected. a. x = 1.3130 .. .
b . x = 1.5 373 .. .
Problem Set 11-3
c. False. For th e so lid, x is farther from the yz-
Mass of o VariableDensityObject plane .
1. a. 8.1419 ... k 7. Centroid of a Triangle Exper im ent
b. 108.1103 ... Construct axes with the origin at a vertex and th e
x-axis along the b ase, b.
3. a. 40.5 TTk
Slice the trian gle parallel to the x-axis .
b. 546. 75 TTk
C. 105 .3TTk
Width of a st rip is b-*Y
d. The solid in part (b) has the largest mass.
5. Two Cone Prob lem
dA = ( b - * y) dy
a. Prediction: Cone on the left wit h higher density 2
dM x = ydA = (by-t y ) dy
at bas e ha s greate r mass , because higher density
is in the larg er part of th e cone .
Mx -- Ioh ( by b
- hy 2) dy -- 1 by 2- bhy 31h
b. For th e left-hand cone, m = 1305 TToz. For th e 2 3 0
right-hand con e, m = 1035TT oz. = !bh 2
:. the cone on the left has the higher mass, as 6
predicted in part (a). !bh 2
2
7. Yi = 4 - 2x an d Y 2 = 3 -
x-axis
x 2,
rotate d abo ut the y A = Mx => y = t-- = h, Q.E.D.
2bh
Graphs int ersect at ( 1, 2) in Quadrant I.
Slice perpendicular to th e axis of rotation, generat- 9. Second Mo m ent of A r ea Problem
ing plane washers. a. Slice the regio n parallel to the y-axis so th at eac h
Pick samp le points (x, Yi) and (x , Y 2 ) . point in a str ip will be abo u t x units from the
p = kx 2 , dV = TT(Yf - y } ) dx y-axis, where x is at the samp le point (x ,y ).
dm = p dV = TTkx 2 (7 - 10x 2 + 3x 4 ) dx dA = ydx = sinx dx
I
m = Io dm =
16
TTk = 2.3935 ... k A = IZ'sinx dx = 2 (exactly)
21 (Thi s may b e "well known" by now.)
9. Ura nium Fuel Pellet Problem: m "' 15.14 g dM y = xdA = xsinxdx
My= IZ'xsinxdx = 3.1415 .. . = TT(exactly)
1 1
11. a. = k TT
x A= My=> x =
2 , Q.E.D.
m
2TTr
b . m = _!_ 2 4 (or ju st note the symmetry)
4 TT r k
b. M 2y = 5.8696 ...
c. m = TTkr 4 c. x = 1.71 3 1 ...
13. m = 8.6261 .. . 11. Second Moments for Solid Figur es
Problem Set 11-4 a. M = TTHR4 , r = ~ R
Moments, Centroids, Center of Mass, and the Theorem of 1
Pappus b. Mh = 10 TTHR4 , r = ..Jo.3R
1. Paraboloid Problem C. M = ~15 TTR5 , r = v'0"4R
u.~
a. V = 40 .5TT
13. Beam Moment Prob lem
b . Mxz = 121.5TT
c. Centroi d is at (0, 3, 0). a. Set up axes with x-axis through the centroid .
dM2 = y dA = y 2 B dy
3. Paraboloid Mass Problem 0.5// l
a. m = 170.1375 . . . k M2 = B I~l~1
-1Y 2 dy = !y 3 = -B H3
3 I- 0. SH 12 '
b. Mxz = 612.4952 ... k Q.E.D.
c. Center of mass is at (0, 3.6, 0). b. i. Stiffness = 288k
748 Appendi
x B: Ans
werstoSelected
Problems
ii. Stiffness = 8k Problem Set 11-6
Board up on edge is 36 times stiffer. Other Variable-Factor
Products
c. i. Stiffness = 160k
1. Heat Capacity Problem: H = 13,200 calories
ii. Stiffness = 448k (2.8 times stiffer!)
d. Incr easing the depth does seem to increase stiff- 3. Tunnel Problem
ness greatly, but making the beam very tall would a. P(x) = 0.00Zx 2 + 3x + 500
also make the web very thin , perhaps too thin to b. P(700) = $3580/ft.
withstand much forc e. c. Cost is about $2,666,667.
15. Theorem of Pappus Problem d. Cost is about $1,416,667
a. Toroid Problem: V = 2rr 2 r 2 R Saving is about $1,250 ,000!
b. Centroid of a Semicircle: r= ~ r 5. Wire-Pulling Problem
3 a. Graph, connected scatter plot.
Problem Set 11-5 F
600
ForceExertedby a VariablePressure-Center of Pressure
1. Trough Problem
a. F = 2.8444 ... k
b. Mx = 2.1880 ... k X
c. Center of pressure is at ( 0, ~~) .
b. F has a step discontinuity at x = 2.
3. Ship's Bu lkh ead Problem 2 .
C. W = 600 in-lb d. W :::::1266 m-lb
a. A :::::1186.6077 ... :::::1186.6 ft 2 3
C.
Mx
~ (;,-:
d.:,(:
=M2
- s,r::
b. F :::::1199294.1645 ... :::::1.199 million lb
dMx""' 13992028.2564 ...
e. Total work ""'1866f in-lb
f. Yes, a piecewise-continuous function such as this
one can be integrable. See Problem 27 in Problem
Set 9-10 (Improper Int egrals).
7. Moment of Inertia Probl em
:::::13.992 million lb-ft M 2y = 33 .5103 ... k g-cm 2
d. y ""'11.6668 .. . ft. x = 0 by symmetry. 9. Rocket Car Problem
Center of pressur e is at about (0, 11.67) ft. a. m = 2000 - St
e. y""' 16.9150 ... ft. x = 0 b y symmetry. b . a= 1400(400 - t ) - 1
Centroid is at about (0, 16.92 ) ft.
400
Centroid is diff ere nt from center of pressure. c. v (t) = 1400 In _ ti
1400
f. y ""'8.6566 ... ft. x = 0 by symmetry. d. v (20) = 71.8106 ... ""'71.81 m / sec
Center of buoyancy is at about (0, 8.66) ft s = 711.9673 ... ""'712.0 m
5. Airplane Wing Problem I
11. Sinusoidal Land Tract Problem
a. A= 76 3.94 37 ... :::::763.9 ft 2
a. W = 0. 5707 ... k b. W = 0.3926 ... k
b . F = 4863.4168 . .. k
13. City Land Value Problem:
c. Make k 2:. 0.0197 . .. tons / ft 2
a. W = 113.0973 ... ""'113.1 million dollars.
7. Double Int egrat ion Airplane Wing Problem
b. W = 71.4328 ... ""'71.4 million dollars.
a. dM 2x = 1 (0.25(x - 4) - (x - 4) 1/ 3 ) 3 dx c. W = 163.9911 ... ""'164.0 million dollars.
3
d. This problem is equivalent to volum e by cylin-
8 drical shells, where the value of the land per
b. M 2x = 0.5333 ... ( exactly )
15 square unit takes the place of the altitude of the
9. The integrals in Problems 7 and 8 can b e written in cylinder. It is also equivalent to the water flow in
the form I::! s::cdf(x, t) dt dx Problem 4 of this problem set.
Since two integrals appear, the result is called a 15. Skew ness Problem
double integral. (Hiding inside each integral is a a. j(x) = 9 - x 2 = (3 - x) (3 + x) = O only at
second integral!) X = 3 .
Problem
Set11-6:Other
Variable-Factor
Products 749
1 R3. Variab le Densily Prob lem
g(x) = -- x 3 - x 2 + 3x + 9
3 a. m = 57.6rrk
1
= - (x - 3)(x + 3) 2 = 0 only at x = 3 b. m = 64 rr
3
b. AJ = fl 3 (9 - x 2 ) dx = 36 R4. a. Triang le Centro id Prob lem
1 , 1 _ 1
Ag = f~3 (- x3 - x 2 + 3x + 9) dx = 36
Mx =
6
b. Second Mo ment of Volume Prob lem
bh- , A =
2bh ~ y =
3
h, Q.E.D.
To simplify algebraic integration you could use
M 2y = 3.5401 . ..
AJ = 2J5(9-x 2 )dx
f5
Ag = 2 (9 - x 2 ) dx, where the odd terms RS. Wind Force Prob lem
F = 3736263.2708 ... "" 3.736 million lb
integrate to zero between symmetrical limits.
Thus the two integrals are identica l. R6. Oil Well Prob lem
c. The high point off comes at x = 0. S )x /10000
The high point of g comes at x = 1. a. r (x) = 30 (
3
d. x = 0.6 (Or r(X) = 3Qe- ln0.6X/10Q
QQ = 3Qe0.00005108256 .X)
e. False. For the symmetrical region under the f b. C = 6965243.17 . . . "" 6.965 million dollars
graph, the centroid is on the line throug h the
high point . But for the asymmetrical region un-
der the g graph the high point is at x = 1 and CHAPTER 12
the centroid is at x = 0.6.
f. False. Exploratory Problem Set 12-1
Area to left = 17.1072 Introductionto Power Series
Area to right = 18.8928 1. Graphs, j(x) and P5 (x).
g. S = -17 .7737 ...
h. By symmetry, the centroid of the area under f is 100 Y
on the y-axis, so x = 0. Then
dS = x 3 dA = x 3 (9 - x 2 ) dx
X
S = f~3 x 3 (9 - x 2 ) dx = 0 (odd function inte-
I
/.,.- - 1
grated between symmetrical limits) I
I
i P5
The "skewness" being zero reflects the symmetry I
I -100
of this region . It is not skewed at all. I
. 1 3
1. For example, graph g( - x ) = x - x 2 - 3x + 9.
3 3. Ps (0. 5) = 11.8125, P5(0.5) = 11.90625 , j(0 .5) =
12 .
.'.P6 (0.5) is closer to f(0.5) than P 5 (0.5) is.
New
graph Ps (2) = 378 , P5(2) = 762, j(2) = -6.
.'.P5(2) is not closer to j(2) thanPs(2) is.
5. Po(l) = 6 Po(-1) = 6
I
I Pi(l) = 12 Pi(-1) = 0
\ X
P2(l ) = 18 P2(- l) = 6
-3 3
P3 (l ) = 24 P3( - l) = 0
P~(l) = 30 P~( - 1) = 6
17. Another Theorem of Pappus Prob lem: Not selected . For x = 1, the sums just keep getting bigger and
bigger as more terms are adde d. For x = -1, th e
Problem Set 1 1-7 sums oscillate between O an d 6. In neit h er case does
Chapter Review and Test the series converge . If the answer to Prob lem 4 h ad
included x = 1 or x = - 1, the conjecture woul d
Review Problems
have to be modified.
RO. Journa l entries will vary.
7. Geometric series . x is the common ratio .
Rl. Work Problem
W = 129.6997 ... "" 129 .7 ft-lb
Problem Set 12-2
2
R2. a. Magnet Problem: W = - k ft-lb Geometric Sequences and Series as MathematicalModels
3
b . Conica l Cup Prob lem 1. Series: 200 - 120 + 72 - 43.2 + 25.92 - 15.552 +
W = 11.2814 .. . "" 11.28 in-lb Sums: 200, 80, 152, 108.8, 134. 72,119 .168, . . .
750 App
endix B: Answe
rstoSelec
ted Probl
ems
Graph, showing convergence to 125. S. Compound Interest Problem
a. months dollars
Sn 0 1,000,000.00
1 1,007,500.00
12_5_ - - - .- - - . - .. -.- --- -
2 1,015,056.25
100 3 1,022,669.17
b. Worth is $1,093,806.90; Interest is $93 ,806. 90
n c. The first deposit is made at time t = 0, the sec-
10
ond at time t = 1, and so forth, so that at time
t = 12, the term index is 13.
S11 will be within 0.0001 unit of 125 for all values of d. 9.3806 . .. % APR
n ~ 28. e. After 93 months.
3. Drug Dosage Problem 7. Bouncing Ball Problem
a. Series : I; =l 7(0.8n - l) = 7 + 5.6 + 4.48 + 3.584 + a. Sequence: 20, 18, 16.2, 14.58, 13.122, . ..
b. S4 = 20 + 18 + 16.2 + 14.58 = 68.78 ft
Sums: 7, 12.6, 17.08, 20.664, 23.5312, ... c. S = 200. Ball travels 200 ft before stopping.
1
S = 7. - - = 35 d. 20-foot cycle, t = 2.2291 ... sec .
1- 0.8
The amount approaches 3Sg as a limit, and thus 18-foot second cycle, t = 2.1147 ... sec.
never reaches SO or 80 g. e. The model predicts that the ball comes to rest
b. 2 puffs : The amount will first exceed SO g after after about 43.4 sec.
the 6th dose, and stay above SO g after the 11th 9. Derivatives of a Geometric Series
dose. It will never reach 80 g. P '( O) = 6 and j'(O) = 6,
3 puffs: The amount will first exceed SOg after P "( O) = 12 and j"(O) = 12,
the 3rd dose and stay above SO g after the 5th P"' (0) = 36 and f"' (0) = 36
dose. Conj ecture: p (n) (0) = f' 11) (0) for all values of n .
The amount will first exceed 80 g after the 7th
Problem Set 12-3
dose and stay above 80 g after the 14th dose.
4 puffs: The amo unt will first exceed SOg after Power Series for an Exponential Function
the 2nd dose and stay above SO g after the 3rd 1. j(x) = Se2x
dose . f' (x) = 1Qe2x
The amount will first exceed 80g after the 4th f" (x) = 20e 2x
dose and stay above 80g after the 6th dos e. f"' (x) = 40e 2x
c. Tw ice a day: r = 0.8 2 = 0.64, p = no. of puffs f'4 ) (x) = 8Qe2x
Amount just after the nth dose:
3. Co = 5, Ct = 10, C2 = 10.
For p = 4 puffs, S = 77.7777 .. ..
For p = S puffs, S > 80, whic h is uns afe. S. Graphs off, P3, and ?4.
Amount just before the nth dose:
For p = 4, S = 49.7777 ... , which is just barely 100 y I Ip, ,'
I ,
below th e minimum effect ive amount. I ,'
1/ I P,
Graph for p = 4. v/
/
80 ......
Amt. after dose n
50
. . .... .
,
Amt. before dose n
7. P3(l ) = 31.6666666 .. .
P4(l) = 35.00000 00 .. .
n j(l) = Se2 = 36.9452804 ...
10
.'.P4 (1) is closer to j(l) than P3(1), Q.E.D.
_ 40 _ 5 2 3 _ 20 _ Ll:_
From the grap h you can see that if the maximum 9. C3 - 6 - 3! ' C2 - 2 - 2! '
amo unt s after dose n are kept below th e allow-
abl e 80g, then the minimum amounts befor e 10 5 . 21 5 . 2
dos e n are significantly below the minimum c1 = - = -- c = 5 = --(O! = 1)
effec tive amount of SO g for 3 or 4 days . The
1 l! ' 0 O!
situ ation wou ld be worse for daily doses. 11. P(x) -- "' 00
.L,.,n =O
~ X 11
n.I
Problem
Set12-3:Power
Series
foranExponential
Function 751
Problem Set 12-4 y
PowerSeries for Other ElementaryFunctions In x
l . Exponential Function Ser ies Problem X
a. j(x) = ex j(O) = P(O) = 1 Co= 1
j'(x) = ex J'(O) = P'(O) = 1 C1 = 1 S ,o
1
j"(x) = eX j"(O) = P"(O) =1 2!C2 = 1 C2 = -
' 21
1 e. S10(1.2) = 0.182321555 . . .
f "' (x) = ex f'" (0) = P'" (0) = 1 3!C3 = 1 C3 = -
' 31 lnl.2 = 0.182321556 .. .
Sio(l.95) = 0.640144911 .. .
... P(x) -- 1 +x + 1 x 2 +. 1 x 3 + , Q.E.D. ln 1.95 = 0.667829372 .. .
21 31 S10(3) = - 64.8253968 . . .
ln 3 = 1.0986122 . . .
b. Next two terms .... . + 4!x
1 4
+ 1 x s + ...
51 S1o(x) fits lnx in about 0.1 < x < 2.
1 S lO ( 1. 2) and ln 1. 2 agree through the 8th decimal
L
00
c -x n place . The values of S 1o( l.95) and lnl.95 agree
n!
n =O
only to 1 decimal place. The values of S 10(3) and
d . Graph, y S3(x) (fourth partial sum) and 1n 3 bear no res emblance to each other .
y = ex.
7. Inverse Tangent Series Problem
1 1 1
a P(x) = x - - x 3 + - x s - -x 7 + ...
f
y
ex / . 3 5 7
I
f s,
/
I b. Graph.
I
y : s,
X
I l(x)
3 I
/
/ X
e. Interval is about -1 < x < 1. I
I
Ss
I
f. Interval is -0 .2237 ... < x < 0.2188 .. . I
g. Interval is -1.5142 ... < x < 1.4648 . . .
Both partial sums fit the graph off very well for
3. Sine Series Problem about - 0.9 < x < 0.9 . For x > land x < - 1 the
a. S3(0.6) = 0.564642445 .. . partial sums bear no resemblance to the graph
sin 0.6 = 0.564642473 . . . off .
.".S3(0.6) = sin 0.6, Q.E.D.
b. sin 0.6 - S1(0.6) = 0.0006424733 . . . Problem Set 12-5
t2 = 0.000648 Taylorand MaclaurinSeries, and Operations on these Series
sin 0.6 - S2(0.6) = -0 .00000552660 .. .
t3 = -0 .00000555428 .. . 1. 1 + u + ~u 2 + !u 3 + !u 4 +!us+ ...
21 3! 4! 5!
sin 0 .6 - S3(0.6) = 0.0000000276807 .. .
t4 = 0.0000000277714 .. . 3 u - !u 3 + !u s - !u 7 + !u 9 - J:_u 11 + ..
. 3! 5! 7! 9! 11!
In each case the tail is less in magnitude than the
absolute value of the first term of the tail, Q.E.D. 5. 1 + !u 2 + ~u 4 + !u 6 +!us+ _l_u10 + .. .
Use at least 9 terms . 21 4! 6! 8! 10!
7. (1 - u) - 1 = 1 + u + u 2 + u 3 + u 4 + u s+
5. Natural Logarithm Series Problem
a. P(l) = 0 = j(l) 9. X 2 - -1 X 4 + -1 X 6 -
1
-X 8 +
1 10
- X -
P'(l) = 1 = j'(l) 31 5! 7! 9!
P"(l) = - 1 = j"(l) 1 1 1
P'"(l) = 2
j'"(2), Q.E.D.
=
11. 1 + - x6 + -x12 +-xis+ -1 x 24+ .. .
2! 4! 6! 81
1 (X - 1) s - Ei
b .. . . S 1 (X - l )6 +
13. (x 2 - 1) - (x 2 -1) 2 + (x 2 - 1) 3 - ...
1
c. P(x) = L (-l)n
00
+l. -(x - l)n 2
(or 2lnx = 2(x - 1) - (x - 1) 2 + - (x - 1) 3 - . )
n =l n 3
d. Graph, S10(x) and lnx.
752 Appendix
B: Answers
toSelected
Prob
lems
15 x - .!.x 3 + .!. __!_x
5 - .!. __!__x
7 + .!. __!__x
9 - tangent ser ies is - 1 ::;;x ::;;1. In general, power
. 3 5 2! 7 3! 9 4! ser ies converge slowly at the endpoints of th e
convergence interval.
17 .!.xs _ _!_ . _!_xis+_!_ . __!_x
2s _ _!_ . __!_x
3s + .. .
. 8 18 3! 28 5! 38 7! 3 7. Not selecte d.
1 1 1 1 1 1 1 39. Ratio of Terms Problem
19 2x 2 + 6 3!x 6 + 10 Six
IO 14
+ 14 7!x + n
a. r,, = -- Ix - 1 I
21. l -x 4 + x B- x12+x l6_ ... n +l
2
23 x - .!.xs + .!.x9 - J_x 13 + J_x 17 - ... b. rlO = ii for x = 1.2
. 5 9 13 17
95
2 5 ./2 + ./2
. 2 2
(x-~)- ./2 (x-~)
4 2 2! 4
2
r10 = f or x = 1.95
11
20
ii
- ./2
2 3!
(x-~) ./2 (x-~)
4
3 +
2 4! 4
4
rm =
c. r = Ix - 11
for x = 3
+ ./2
2 5!
(x-~)s 4
d. r = 0.2 for x = 1.2
r = 0.95 for x = 1.95
r = 2 for x = 3
27. (x - 1) - (x - 1) 2 + (x - 1) 3 - (x - 1)4 + . . . e. The series converges to ln x whenever the value
of x makes r < 1, an d diverges whenever the
7 7.4 741 value of x mak es r > 1.
29. - 1 + (x - 4 ) - (x - 4) 2 + (x - 4) 3-
3 3221 3331 f. r = Ix - 11 < 1 ~ - 1 < (x - 1) < 1 ~ 0 < x < 2.
7 4 1 (-2) 4
344! (x - 4) +
Problem Set 12-6
741( -2) (-5) s Interval of Con vergence for a Series-The Ratio Technique
3s5 1 (x - 4) - ...
9 81 729 6 1 2 2 3 3 4 4 5 ,5
31 Both give cos 3x = 1 - - x 2 + -x 4 - -x + . . . 1. a. 4x + 16x + 64x + 256x + 1024x +
. 2! 4! 6!
Substitution gives the answer much more easily. b. Open int erval of convergence is ( - 4, 4).
c. Radius of convergence = 4.
33. Accuracy for ln x Series Value
S4 (1.5) = 0.40104166 . . .; lnl.5 = 0.40546510 . .. (2x + 3) 2 (2x + 3) 3 (2x + 3)4
3. a. (2X + 3 ) + 2 + 3 + 4 +...
error = 0.00442344 ...
1 - b. (-2 , - 1)
fifth term = (1.5 - l P = .00625
5 1
Error is sma ller in absolute value than t 5 . C. 2
3 5. Inv erse Tangent Ser ies and an Approximation for TT 2
a. 4S9 (1) = 3.04 183961 . . . 5. a. (x-8)+!(x- 8) 2 + Z(x - 8) 3 + ~!(x - 8) 4 + ..
TT = 3.141 59265 ... b. Series converges for all values of x .
The error is about 3%. c. Radius of conve rgen ce is infinite.
b . 4S4g(l ) = 3.12159465 . . .
7. L = x 2 O < 1 for all x. The series converges for all
TT = 3.14159265 ...
The error is about 0.6%.
x.
4 ( l) 2n+l 9. L = x 2 0 < 1 for all x. The series converges for all
C. 4Sg = I
9
n =O
.
(-1) 11 --
2n + 1 2
- x.
+ (-l)"- 4-
( .!.) 2n+l
X
11. e -
_ , 00
Ln =O
_!_ n
n .1
X
~l
n =O 2n + 1 3
= 3.141 59257 .. . L = lim n- ooI (:: +~)I ;,; I = lxl limn - oon
TT = 3.1415 9265 . ..
= lxl 0
The answer differs from TT by only 1 in the 7th : .L < 1 for all x and th e series converges for all x.
decimal plac e. The improv ement in accuracy is
accounted for by the fact that the inverse tan- 13. L = lx l oo = oo for all x * O; L = 0 at x = 0.
gent series converges much more rapidly for .-. the series converges only for x = 0.
x = 1/ 2 and x = 1/3 than it does for x = 1. 211
In Problem 17 of Problem Set 12-6 you will see 15. cash 10 = I: =o (2~ )! 10
that the int erval of converge nce for the inverse L = 10 2 O = O < 1 ~ series converges.
Set12-6:
Problem Interval
ofConvergence
fora Series-The
Ratio
Techn
ique 753
17. Inverse Tangent Series Problem 1 1 1
c. 120 + 840 + 6720 + . ..
a. Open interval of convergence is ( -1 , 1) .
b. Graphs fit very well for - 1 < x < 1. Partial sums d. Terms are decreasing.
diverge from tan - 1 x for x outside this interva l. e. Partial sums are increasing.
I
y
3. Vocabu lary Problem III
I
a. 6 - 3 + 1 - ! + ~ -
I $ 3
I
I
. 4 20
3 4
I
b. 6,3,4,34,35'"""
tan- 1 /
I s,
s,, I
I 1 1 1
I
I
c. - 120 + 840 - 6720 + ...
c. S3(0.l) = 0.09966865238095 ... d. ( - 1)" + 1 makes the signs of the terms alternate .
d. tail= 0.00000000011020 ... e. Sz = 3, s~= 3.75, S6 = 3.791666 ... , .. .
e. First term of tail = 0.0000000001111 l ... , which The even partial sums are increasing . For the two
is larger than the tail. terms added to get the next even partia l sum, the
positive term is larger in absolute value th an the
19. The Error Function negative term .
1 1 1
a. J(x) = x- !x 3 + - -x 5- - -x 7 +- -x 9 - f. S1 = 6, S3 = 4, Ss = 3.8, ...
3 5 2! 7 3! 9 4! The odd partial sums are decreasing . For the two
-- 1 x 11 + ... terms added to get the next odd par tial sum, the
11 5! positive term is smaller in absolute value than
b. Graph. Approximately - 1.5 < x < 1.5. the negative term . Since the odd partia l sums are
y
all greater than the even partial sums, the entire
f(x )
sequence is bounded above by the greatest odd
partial sum, S1 = 6.
X
5. Interval of convergence is (2, 4).
s, 7. Interval of convergence is [ - 1, 1).
9. Interval of convergence is [ -6, -4].
c. L = x 2 0 < 1 for all x. 11 . Interval of convergence is [ - 1, 1).
d. Erf x does seem to be approaching 1 as x 13. Intervals of convergence are (-oo, -4) and (4, oo ) .
increases, as shown by the following table gener-
ated by numerical integration. 15. Upper Bound by Convergent Improper Integra l
x erf x 1 1 1 1 5269
a. 1 + 4 + 9 + 16 + 25 + ; Ss 3600
1 0.8427007929 .. .
1.46361111 . . .
2 0.9953222650 .. .
3 0.9999779095 .. . b. The sixth term of the series has a rectangle
4 0.9999999845 .. . between x = 5 and x = 6 (the heigh t of the
5 0.9999999999 .. . rectangles being used is the valu e of the func tion
at the rightmost edge.) So the integral must be
21. The Root Technique from 5 to oo to include this rectangle .
Not selected.
23. Open interval of convergence is (0, 2). c. Tail is bounded above by !5 .
25. L = 0 if x = 0, and is infinite if x * 0. d. The sequence of partial sums is increasing (since
.. the series converges only if x = 0. the terms of the series are positive) and bounded
above . Thus the sequence of partial sums must
Problem Set 12-7 converge. The series is a finite sum added to the
Convergence of Series al the Ends of the Convergence Interval tail, so the series converges.
e. S1000 = 1.643934 .. .
1. Vocabulary Problem I
1 1
f. Upper bound for tail is 0.001.
a. 6 + 3 + 1 + - + - + . (Lower bound for tail is 1/1001 = 0.0009990 .... )
4 20
1 3 g. False.
b. 6,9,10,104,1010'""" h . No more than a 0.0608 % error.
754 Appendix
B:Answ
ers toSelected
Probl
ems
i. It would take at least 2 million terms! 27 . Converges by comparison with:
Geometric series with to= 1 and r = 1/6.
17. Integral Test Problem
Factorial reciprocal series in Problem 19.
Assume f (x) is positive and decreasing for all x
beyond x = D. 29. Diverges. Use th e int egra l test or compare with a
If J converges, then the tail of th e series can be harmonic series .
bounded above by th e number to which the in-
31. Not selected.
tegral converges . Because the terms are positive
the partial sums are increasing. Thus th e sequence Problem Set 12-8
of partial sums converges b ecaus e it is increasing
ErrorAnalysisfor Series
and bounded above. Since the tail of the series
converges, so does th e series. 1. a. S 5 (4) = 27.2699118 ...
If J diverges, then the tail of the series can be b. S 5 (4) is within 2 of cosh4 in the units digit.
bounded below by a divergent improper int egral.
c. cosh4 - S5 (4) = 0.0383 ... , which is well within
Thus the tail is infinite, which implies that the series
the upper bound found by Lagrange's form.
diverges .
3. a. S14(3) = 20 .0855234 ...
19. The Factoria l Reciprocal Series Converges
b . S 14(3) is within 3 units of e 3 in the 4th decimal
a. It is difficult to find the appropriate function
. 1 place.
f(x ) to mtegrate, such that j(n) = - at the c. e 3 - S 1-1(3) = 0.00001346 . .. , which is within the
n!
integer points. So the integral test is impractical upper bound found by Lagrange's form.
for this series. 5. Use at least 7 terms (n = 6).
1
b. The geometric series converges to . 7. Use at least 32 terms .
8
9. c = cosh - 1 1.0 309 ... = 0.2482 ... , which is between
c. An upper bound for the tail of the series is . 0 and 2.
d. An upper bound for the entire series is 2 !!. 11. cos 2.4 = 1 - 2.88 + 1.3824 - 0.2654208
+0.0273004 ... -
e. The sequence of partial sums is increasing (since The terms are strictly alternating. They are decr eas -
the terms of the series are positiv e) and bounded ing in absolute value after t 1, and they approach
above by 2. 7916 .. . . Thus the sequence of partial zero for a limit as n - oo.
sums converges, so the series converges. Therefo r e th e hypotheses of the alternating series
21. Alternating Series Remaind ers Property Problem test apply. Use 8 terms (n = 7).
1 13. p-Series Problem I
a. t 3 = - 0.6 7 = -0 .00000555428571 ...
71 Use 317 terms.
1
b . S 1 (0.6 ) = 0.6 - 0.6 3 = 0.564 15. p-Series Problem III
31 R 99 is bounded above by 15.8945 ....
S2(0.6) = 0.6 - io.6 3 + io .6 5 = 0.564648 R 99 is also bounded belo w by 15.8865 .. ..
So, Amos, the 4.69030101 ... you calculated for S99
C.R1 = sin0 .6 - S1(0.6) = 0.0006424 ... is not close to the actual limit to which the series
R2 = sin 0.6 - S2(0.6) = -0 .000005 5266 . .. converges.
It 2 I = 0.000648
..IR1 I < lt 2I 17. Geometric Series as an Upper Bound Problem
1t 31 = 0.0000055542 . . . By Lagrange form, IR10 I < 0.0004617 ... .
. IR2I < lt 3I By geometric ser ies IR 1ol < 0.00006156 ... .
d. The terms are strictly alternating in sign. The geometric series gives a better est imate of the
The terms are strictly decreasing in absolute remainder than does the Lagrange form.
value. 19. Sin x for A ny Argument Using a Value of x in
The terms approach ze ro for a limit as n - oo. [0,rr / 4]
Thus the series converges by the alternating a. b = 4.9557730 ... radians.
series test.
b . c = - 1.32741228 . . . radians.
23 . Convergence of Sequences Proof c. d = 0.243384039 . .. radians.
Not selected . d. IR5 (x)I < lt5(rr / 4) 1 = 3.8980 ... x 10- 13, which
25. Converges because it is a geometric series with is small enou gh to guarantee that sinx will be
common ratio 1/4, which is less than 1 in absolute correct to 1O decimal places.
value . Direct calculation would take about 349 terms
Problem
Set12-8: Error
Analysisfor Series 755
e. Not select ed. RS. a. A Maclaur in series is a Taylor ser ies expanded
21. Derivation of the Lagrange Form of the Remainder about x = 0.
Not selec ted. b. ln (x + 1) = x - 1 x 2 + 1 x 3 - 1 x 4 +
2 3 4
23. The Maclaurin Series for ex Converges to ex 1 1 1
Not selected . c. fln (x+l)dx = - x 2 --- x 3 +-- x4
- +C
32 2 4.3
d. f ln (x + 1) dx = (x + 1) ln(x + 1) - (x + 1) + C1
Problem Set 12-9 =x ln (x + l )+ ln(x+l)-x+C (C=C 1 -1 )
Chapter Review and Test = !:.x2 - _ l _ x 3 + _ l _ x4 - . . . + C,
2 32 4.3
Review Problems which is the same as the series in part (c).
RO. Journal
9
fX
e. 10 tcost 2 dt- _ .lx,2 - _ .l 21
_ 6 _ l_
x + . x 10
2 6 10 41
Rl. j(x) = -- and P (x) = 9 + 9x + 9x 2 + 9x 3 + 1
1-x --- x 14 + ...
Graph, j(x), P5 (x), and P6 (x), showing that P(x) 14 6!
is close to j(x) for x between about - 0.7 and 0.6,
and bears little resemblance to j(x) beyond 1. f. tan - lx = ft 1 : t 2 dt = ft(l-t2 + (t 2)2
- (t 2 )3 + (t 2 ) 4 - ) dt ( lt l 5; 1)
I y ,, 1 1 - 1 - 1
I P6 It = x - -x 3 + -x " - -x 1 + -x 9
I 3 5 7 9
\ 20
g. j(x) = S+7(x-3)-3(x-3 ) 2 + 0.15(x-3) 3+
'
,,
X
I
I P,
I
R6. a. L = (-3) - n (x - 5) 11
I n=l
= _!:.(x - 5) + !:.(x - 5) 2 - J_(x - 5) 3 +
3 9 27
Ps (0.4) = 14.93856 b. Open int erva l of convergence is (2, 8).
P5(0.4) = 14.975424 Radius of convergence = 3.
j(0.4) = 15
c. L = x 2 O < 1 for all x .
. .P6 (0.4) is closer to j(0.4) than P 5 (0.4) is, Q.E.D .
Series conve rg es for all x, Q.E.D.
Ps (O) = 9 = j(O)
1
A((O) = 9 = j'(O) d. e l.2 = l + l.2 + l (1.2) 2 + (1.2) 3 + l (1.2) 4 +
P5' (0) = 18 = f" (0) 21 31 41
Ps"(0) = 54 = f" ' (0) Error= e1.2 - S4 (1.2) = 0.02571692 ...
First term of the tail is t 5 = 0.020736
P11 (x) is a subset of a geometric ser ies.
The error is greater than t 5 , but not much greater.
R2. a. Biceps Problem e. Graphs.
About 19.5 mm increase in 10 days .
About 30 mm increase eventually. y '
/ S 11
b. Present Valu e Problem / In
They must invest $8 1,754 .00 now in order to
make th e last payment. 2 I
They mu st invest $4,182,460.05 now to make all \ S ,o
I
19 payments. I
1
R3. co = 7, c1 = 2 1, c2 = 31.S, c3 = 31.5
R4 . a. e012 = l.127496851 . . . The open int erva l of convergence is (0, 2) . Both
S3(0.12) = 1.1 27488 , which is close to e 0 12. partial sums fit ln well within this interva l. Above
x = 2 the partial sums diverge rapidly to oo .
b. cos 0.12 = 0.9928086358538 ...
Below x = 0 the partial sums give answers, but
S (0.12) = 0.9928086358528, which is close .
ther e are no real values for ln x.
c. sinh(0.12) = 0.1202882074 311 .. .
S3(0 .12) = 0.1 202882074310 ... , R7. a. S1o = 4463 .129099 (exactly)
which is close. b. S - S 10 = 536 .870912, which differs from the
limi t by about 10.7%.
d. lnl. 7 = 0.5 30628251 .. .
S20 (1.7) = 0.530612301 ... , which is close. c. "Tail"
ln 2.3 = 0.83290912 .. . d. "Remainder"
S20(2.3) = -4.42067 878 ... , which is not close. e. The remainder is bounded above by 0.005 .
756 Appendi
x B:Answe
rs toSelected
Probl
ems
f. The series converges because the sequence of b. Continuity on an interval : See Section 2-4.
partial sums is increasing and bounded above. c. Convergence of a sequence: A sequence con-
g. 2/ 1! + 4 / 2! + 8 / 3! + 16 / 4! + 32 / 5! + verges if and only if limn- ootn exis ts.
= 2 + 2 + 1.3333 ... + 0.6666 .. . + 0.2666 ... + d. Convergence of a series: A series converges if and
= I ~=12n ; n! only if the sequence of partial sums converges .
The terms are decreasing starting at t 2 .
e. Natural logarithm: See Section 6-3.
R 1 is bounded by the geometric series with first
term 2 and common ratio 1.3333 ... /2 = 2/3. f. Exponential: ax = e x lna .
Since Icommon ratio I is less than 1, the geometric 3. a. Mean valu e theorem: See Section 5-6.
series converges (to 2/ (1 - 2 / 3) = 6). b. Int ermediate value theorem: See Section 2-6.
Thus the tail after the first partial sum is bounded
above by a convergent geometric series, Q.E.D. c. Squeeze theorem: See Section 3-8.
h. Use the A lternating Series test. d. Uniqu eness theorem for derivatives: See Section
The terms alternate in sign, decrease in absolute 6-4.
value, and approach zero for a limit as n - oo. e. Limit of a product property: See Section 2-3.
i. Upper bound is 1/ 10001. f. Integration by parts formula: See Section 9-2.
j. i. Interval of convergence is (2.9, 3.1]. g. Fundamental theorem of calculus: See Section
ii. Interval of convergence is ( - 3, 1]. 5-8.
k. i. The tail after So is bounded above by the h. Lagrange form of the remainder: See Section
convergent geometric series with first term 12-8.
10 and common ratio 0.5. Thus the series i. Parametric chain rule: See Section 4-7.
converges. j. Polar differential of arc length : See Section 8-9.
(Other justifications are possible.)
ii. Diverges because tn approaches 0.2, not 0, as 4. a. f' (x) = ,J 1 + sechx
n - oo. b. f' (x) = a x Ina
R8. a. Error is less than 0.03.
c. j'(x) = axa - l
b . Use at least 34 terms (n = 33). d. f' (X) = X x 1nX + X x
2 6
c. Using the Lagrange form of the remainder, the e. Je 6x cos 3x dx = \ e 6x sin 3x + e x cos 3x + C
value of cosh 4 is given exactly by 1 15
1 cosh 6 x
cosh4 = ,to(
k
2 n)!
211
4 + Rk(4), where
f. Jcosh 5 x sinhx dx = +C
1
Jsec 3 x dx =
R (4) = f~-
(2k+2)(c )
-- . 42k+2
g.
2 sec x tanx
k (2k+2 )! 1
and c is between O and 4. +
2 ln I secx + tanx l + C
1Rk(4) 1 5; M l4 12k+2
(2k + 2)1 h. J(sin 5x) - 1 cos 5x dx = ln I sin 5xl +C
lim k-oo IRd4) I = O
Since the remainder approaches zero as n ap- . limi __ 49
1. t - 26
proaches infinity, cosh 4 is given exactly by
cosh4 = J 1
0
(
2
211
~) 1 4 , Q.E.D.
j. limit= e- 3 = 0.04978 ...
5. a. Graph.
d. c = cosh - 1.00328 ... = 0.0809 ... , which is in Y\ \ \ \ \ \ \ "\ "- "-
the interval (0, 0.6). \ \ \ \ \ \ "'-"- "----._
e. Use at least 35 terms. \ \ \
\\\"'-"-' "" "_______ -
----- -_
f. An upper bound is 2.6666 ... x 10 6 .
\ " "" -------- -
Problem Set 12-10: Cumulative Reviews
" ------ - - -
" " " "- - - - ?' / /
/ / / /
CumulativeReview N umber I-The Dam Problem ------//////
- ---///////
1. Limit: See Sections 1-5, 2-2, 2-5, and 2-7. -/////////
Derivative: See Sections 3-2 and 3-4.
Indefinite integral: See Section 3-9.
Definite integral: See Section 5-4. b. If x = 9, y :::::5.413 ... , which agrees with the
2. a. Continuity at a point: See Section 2-4. graph.
Problem
Set12-10:
Cumulative
Reviews:
Cumulative
Review
Number
1-TheDam
Problem 757
6. a. p = k (40-y) 18. F = 113595 .73 ... :::::113,600 lb
b. A = 1066. 6 ... yd 2 19. limit = 0
c. F = 1706 6.6 ... k lb
20 . There is a maximum at x = e since y' goes from
d. M = 292,571.4 ... k lb-yd positiv e to n ega tive th ere .
e. Cent er of pressur e is at ( 0, 17 ~). 21 . There is a point of inflec tion at x :::::4.48 ft since y"
chan ges sign th ere.
7. a. z = 30 - 0.5y
22 . Graph .
b. Max. at y = 20; min . at y = 0.
c. 3840 truckloads .
d. L = 92.9356 ... :::::92.9 yd
8. Speed= 2.9943 ... :::::2.99 ft / sec X
'
40
9. s1 t -- t - 1
~t
3
+
1
~t
1 7 s
- 7 . 71t +
L = t 2 0 < 1 for all values of t, and the series
converges for all values of t . 23 . ln x=(x- 1)-!(x- 1) 2 +( x - 1) 3 -
Third partial sum is S2 (0.6) = 0.5881296 L= lx-1 1
Answer is correct to 1 in th e sixth decimal place. L < l =O < x<2
Si0.6 :::::0.588128809 ... At x = Oht e senes . 1s
. - 1 - -1 - -1 - -1 - whihc
2 3 4 '
10. A= 103.672 5 ... :::::103.7 ft 2 (Exactly 33rr) is a diverg ent harmonic series.
11. At t = 10, V = 253.9445 . . . :::::253 .9 million gal. At x = 2 th e senes . 1 - -1 + -1 - -1 + whihc
. 1s
' 2 3 4 '
Cumulative Review Number 2-The Ship Problem converges since it mee ts the three hypotheses of
the alternating series test.
1. Derivativ e: See Sections 3-2 and 3-4
.-. interva l of converg enc e is O < x ::; 2, Q.E.D.
2. Definite int egral: See Section 5-4.
24. Use 46 terms.
3. Mean value theorem: See Section 5-6.
25 . If th e velocity is O ft/ sec at time t = 0, the ship
4. j'(x) = g(x) speeds up , ap proachin g approximately 34 ft/sec
1 as ymptoti cally as t incr eases .
6
x +C
5.
6 tanh If th e velocity is 50 ft/sec at tim e t = 0, the ship
slows down, again approaching 34 ft /sec asymptot-
6. x cosh 2x - sinh 2x +C ically as t increases.
26. a= (-l / t 2 ){ + (- 4sin2t)f
7. - ln lx+3 1+ 4ln lx-2 1+ C
Cumulative Review Number 3- Routine Problems
1 3 1 5 1 7
8. x + ~x +~ x + 7 . 7 !x + + C
1. Graph . o is clearl y smaller than necessary .
9. Open interval of convergence is 2 < x < 8. l(x)
10. 500 L+e
ll.y =39 L ------
L- e
12. j(4) = 16
X
j(3.99) = 15.9201, which is within 0.08 unit of 16.
c-6 c c+ 6
j(4.01) = 16.0801, which is not within 0.08 unit of
16.
Thus, o = 0.01 is not small enou gh to keep j(x) 2. See Sections 3-2 and 3-4 for definitions of deriva tive.
within 0.08 unit of 4 . Graphi cal meaning: Slope of tangent line .
Physical m eanin g: Inst antan eous rate of change.
13. V = ff0 A dx :::::2140 ft 3
14. A = 6.2831 :::::6.28 ft 2 (Exactly 2rr )
3. g(x ) = f j(x) dx if and only if g' (x) = j(x) .
15. A = 17.6021. . . :::::17.6 squar e units
4. I:J (t ) dt = limM- o L,, = lim M- o U,,, where L,, and
U,, are lower and upper Riemann sums, respectively,
16. A= 256 ft 2 provid ed th e two limits are equal.
17. L = 42 .5483 .. . :::::42.55 ft. 5. !'Hospital' s rul e. Limit= - 0.2
758 Appe
ndix B:Answe
rstoSelect
ed Problem
s
6. y = tan (sin 5x) 18. a. Int egral converges to 8.
y ' = sec 2 (sin 5x) 5cos5x. Chain rul e. b. Average valu e = 0.5.
7 y ' =y ( -5- +-8- + _ l_ ) 19. A "" 13.3478 ...
5x - 3 2x + 7 x - 9
20. ii (1)=2{ - 3f
8 ' = _ l_ Speed= Ju= 3.6055 ...
y l +x2 Distance from origin is decreasing at 2.2135 .. ..
9. sin 8 x +C 21. V "" 3.5864 .. . (Exactly 2rr (rr / 2 - 1))
1 9
22. a. A' = 0 =
x = 2.
10. x .Jx 2 +9+ In1.Jx 2 +9+x l +C A(O) = 0, A(4) = 0, A(2) > 0.
2 2 Thus maximum area is at x = 2, Q.E.D.
11. 5 ln lx+31 - 2 ln lx-l l +C
1
b. Maximum volum e is at x = 2 .
12. xsin - x+ .J l - x 2 + C
13. Fundamental theore m of calculus. 23. V "" 394 ft 3
See Section 5-8 for statement.
= It e- t dt
2
14. See Figure 5-6a. 24. a. j(x)
15. j'(x) = h (x) =X -
1 3 + -- 1 X 5 - -- 1 X 7 + --X1
- X
9
- ..
3 5 2! 7 3! 9 4!
16. Only point of inflect ion is at x = 2.
b. L = x 2 0 < l for all values of x, and thus the
17. L "" 2.3516 .. . series converges for all valu es of x , Q.E.D.
Set12-10
Problem : Cumulati
veReviews:
Cumulative
Review
Number
3-RautineProbl
ems 759
Glossary
The following are descriptions of the major terms used in calculus,
along with references to page numbers where formal definitions and
statements of the terms can b e found. For references to the many
other significant terms used in this text, please see the index.
Acceler ation (pp . 99, 305): The instantaneous rate Continuity (p. 54): A function is continuous at x = c
of change of velocity . if and only if f(c) is the limit of f(x) as x approaches c.
Antideriv ative (pp. 119, 182, 189): g(x) is an Converge nce of a seri es (pp. 631, 635): A series con-
anti derivative of f(x) if and only if g'(x) = f(x). An verges to a particular value if the limit of the partial
antider ivative is the same as an indefinite integral. sums of the series equals that value as the number of
terms approaches infinit y.
Av erage value of a function (p. 514): The integral of
f(x) from x = a to x = b, divided by the quantity (b - a). Critical point (p. 3 54): A point on a graph where the
derivative is either zero or undefined. Maximum and
Calculus (p. 215): A word meaning "calculation,"
minimum values of functions may occur at critical
coming from the same root word as "calcium,"
points.
chosen because calculations centuries ago were done
using pebbles (calcium carbonate). The word is part Cusp (p. 54): A point on the graph at which the func-
of the title of an appendix to Isaac Newton's tion is continuous, but the derivative is discontinuous.
Princ ipia, entitled The Calculus of Infin itesimals,
which means calculating with quantities that Cylindrical shells (p. 396): A technique for slicing a
solid of revolution into thin shells so that each point
approac h zero as x approaches a particu lar value .
in the shell is virtually the same distance from the
Centroid (p. 570): The centroid of an object is its axis of rotation as the sample point is. Cylindrical
geometric center, found by dividing the first moment shells are used for setting up integrals for calculating
of area or volume with respect to an axis by the area the volume, mass, moment, etc., of a solid object.
or volume . If the object has uniform density, the
centroid and center of mass are at the same point, Definite integral (pp. 16, 19 7): Physical meaning: The
and the obj ect will balance at that point. product (dependent variable)(change in independent
variab le) for a function where the dependent variable
Chain rule (pp. 107, 161): The method for finding may take on different values as the independent
the derivative of a composite function, namely, the variable changes throughout an interval. Geometrical
derivat ive of the outside function with respect to the meaning: The area of the region under the graph of
inside function, multiplied by the derivat ive of the f(x) from the x-value at the beginning of an interval
inside function with respect to x. to the x-value at the end of that interval.
Concave (p. 356): Literally, "hollowed out," but used Deriva tive (pp. 10, 80, 81, 91): Physical meaning:
to refer to the side of the graph of a function or The derivative of a function fat x = c is the
relation that looks hollowed out (contrasted to the instantaneous rate of change of f(x) with respect to
other side, which is convex) . x at x = c. Geometrica l meaning : The slope of the line
tangent to the graph at x = c. (p. 81)
Concept s of calculu s (p. xiii, 33): There are four
major concepts of calculus: limit, derivative, definite Differe nce quotien t (p. 80): The ratio (change in
integral, and indefinite integral (antiderivative). f(x))/ (change in x). The limit of a difference quotient
as the change in x approaches zero is the derivative.
Con stant of inte gration (p. 182): Two antiderivatives
of the same function differ by at most a constant . Differentia bility (p. 153): The property possessed by
The constant term of an antiderivative equation is a function at x = c if f'(c) exists. Function fis
calle d the constant of integration . differentiable on an interval if and only if f'(x) exists
for all values of x in that interval.
761
Differential (p 186): If y = f(x), then the differential Implicit relation (p. 168): A relationship between
dx is the same quantity as 6.x, a change in x; and the two variables where operations may be performed
differential dy is equal to f'(x)dx. Thus, the quotient on the dependent variable as well as the independent
dy-;- dx is equal to the derivative, f'(x). The differen- one. For x 2 + y 2 = 25, there is an implied relationship
tial dy is also the change in y along a tangent to the between x and y.
graph, rather than along the graph itself.
Improper integral (p. 486): A definite integra l in
Differential calculus: An obsolete term for calculus which either one or both limits of integration is
of derivatives only. infinite, or the integrand is undefined for some value
of x between the limits of integration, inclusive .
Differential equation (pp . 119, 310): A differential
equation is an equation that contains the derivative Indefinite integral (pp. 119, 190): g(x) = f f(x)dx if
of a function. A solution of a differential equation is and only if g'(x) = f(x) . An indefinite integral is the
a function whose derivative appears in the differen- same as an antiderivative .
tial equation.
Indeterminate form (pp . 26, 290): A form such as
Differentiation (p. 92): The process of finding the 0/ 0, 0, 00/00 , etc., that an expression may take as x
derivative of a function . approaches a certain value, and for which there may
be a finite limit.
Displacement (p. 99): The directed distance an
object is from a given reference point at a given time. Infinitesima l (p. 595): A quantity that approaches
zero as 6.x approaches zero, such as dy, dA, dV, etc.
e (p. 272) : A naturally occurring constant equal to
2.71828 ... used as the base for the natural logarithm Initial condi tion (pp . 120, 310): A given value of x
and natural exponential function to make the and f(x) used to find the constant of integration.
calculus of these functions simpler.
Integrability (p. 197): The property possessed by
Euler's method (p. 333): A numerical method for a function if the definite integral exists on a given
solving a given differential equation by assuming the interval.
graph follows tangent segments for short distances
from point to point. Integral calculus: An obsolete term for the calculus
of integrals only.
Explicit relation (p. 168): A function for which f(x) is
given in terms of x and constants only. For instance, Integration : The process of finding either the defi-
f(x) = 5x 2 gives f(x) explicitly in terms of x.
nite integral or the indefinite integral of a function.
Function (p. 4) A relationship between two variable Integration by parts (p. 437): An algebraic method
quantities for which there is exactly one value of the for finding the antiderivative of a product of two
dependent variable for each value of the independent functions .
variable in the domain . Intermediate value theorem (p. 6 7): A property of
Fundamental theorem of calculus (pp. 215, 304): continuous functions which states that for any given
The theorem that tells how to calculate exact values number y between f(a) and f(b), there is a number
x = c between a and b for which f(c) = y.
of definite integrals by using indefinite integrals . In
its alternate form, the theorem tells how to find the Interval of converge nce (p. 621): The interval of
derivative of a definite integral between a fixed lower values of x for which a given power series converges.
limit of integration and a variable upper limit of
integration . (Sometimes called the fundamental Lagrange form of the remainder of a Taylor series
theorem of integral calculus.) (p. 643): A way to find an upper bound on the error
introduced by using only a finite number of terms
Grapher: A graphing calculator or computer used to of a Taylor series to approximate the value of a
generate graphs of given functions. function. The remainder is bounded by a multiple
Hyperbolic functions (p. 4 73): Functions with of the first term of the tail of the series after a given
properties similar to the trigonometric (circular) partial sum.
functions, but defined by points on a unit equilateral !'Hospital's rule (p. 285): A property for finding
hyperbola rather than by points on a unit circle. limits of the form 0/ 0 or 00/ 00 by taking the deriv -
Implicit differentiation (p. 149): The process of ative of the numerator and the denominator . (Some-
differentiating without first getting the dependent times spelled l'H6pital's rule.)
variable explicitly in terms of the independent
variable.
762 Glossary
Limit (pp. 4, 10, 27, 40, 61, 74): The limit of f(x) as x Objectives of calculus (p. 71): There are four major
approaches c is the one number you can keep f(x) as things you should be able to do with each of the four
close as you like to, just by keeping x close enough to major concepts of calculus: define it, understand it,
c, but not equal to c. (pp. 27, 40, 74) do it, and apply it.
The limit of f(x) as x approaches infinity is the
number you can keep f(x) as close as you like to, just Parameter (p. 160): The independent variable in a
by keeping x far enough away from zero. (p. 61) parametric function.
The limit of f(x) as x approaches c is infinite if and Parametric function (p. 160): A parametric function
only if f(x) can be kept as far as you like from zero is a function where two variables each depend on a
just by keeping x close enough to c, but not equal to third variable. For example, the x- and y-coordinates
c. (p. 61) of a moving object might both depend on time.
The limit of f(x) as x approaches infinity is infinite
if and only if you can keep f(x) as far as you like from Partial sum of a series (p. 602): The nth partial sum
zero just by keeping x far enough away from zero. of a series is the sum of the first n terms of the series .
(p. 62) Point of inflection (pp. 354-361) : A point where a
Limits are used in the formal definitions of deriv- graph changes from concave up to concave down, or
ative and definite integral. vice versa. Points of inflection occur where the
Linearization of a function (p. 185): The linear func- second derivative of a function has a critical point
tion that best fits function (for values of x close to (i.e., is either zero or undefined).
x = c is y = {( c) + {'( c)(x - c), or, equivalently, y = f( c) Power series (p. 609): A series (with an infinite
+ f'(c)dx. number of terms) in which each term contains a
Local linearity (p. 82): A function is locally linear at power of the independent variable.
x = c if the graph of the function looks more and Radius of convergence (p. 625) : The distance from
more like the tangent line to the graph as one zooms the midpoint of the interval of convergence to one of
in on the point (c, f(c)). its ends.
Logarithmic differentiation (p. 268): An implicit Ratio technique (p. 622) : A technique for determin-
differentiation process where the natural log of a ing the interval of convergence for a power series by
function is taken first, usually so that variables can finding the values of x for which the absolute value
be gotten out of exponents. of the ratio of adjacent terms can be kept less than l.
Logistic equation (p. 331): A differential equation (Sometimes called the ratio test.)
(or its solution) for population growth that takes into Reduction formula (p. 44 7): A formula whereby a
account the assumption that a population will even- complicated antiderivative can be expressed in terms
tually level off at a maximum sustainable value. of a simpler antiderivative of the same form.
Maclaurin series (p. 615 ): A Taylor series expanded Removable discontinuity (p. 26): If a function is
about x = 0. Sine, cosine, exponential, and hyperbolic discontinuous at x = c, but may be made continuous
functions can be calculated using only the operations there by a suitable definition of f(c), then the discon-
of arithmetic by first expanding the function as a tinuity is removable . For instance, f(x) = (x 2 - 25)/
Maclaurin series . (x - 5) is discontinuous at x = 5 because of division by
Mean value theorem (p. 202): The property that zero, but the discontinuity can be removed by
expresses sufficient conditions for a function graph defining ((5) = 10.
to have a tangent line parallel to a given secant line Riemann sum (pp. 196, 198): A sum of the form
at a value of x = c betvveen the endpoints of the I f(x)dx where each term of the sum represents the
secant line. area of a rectangle of altitude f(x) and base dx. A
Moment (p. 568): The product of a quantity such as Riemann sum gives an approximate value for a
force or mass and the power of a distance from a definite integral. The limit of a Riemann sum as dx
point, line, or plane at which that quantity is located. approaches zero is the basis for the formal definition
of definite integral.
Natural exponential function (p. 279): Exponential
function with base e. Rolle's theorem (p. 204): The property that expresses
sufficient conditions for a function graph to have a
Natural logarithm function (p. 254): The classic horizontal tangent for some value of x = c between
example of a function defined as a definite integral two zeros of the function .
between a fixed lower limit and a variable upper limit.
Glossary 763
Sample point (p. 196): A point x in a subinterval for Theorems of Pappus (pp . 5 77, 591): For volume,
which a term of a Riemann sum, f(x)dx, is found; or Volum e = (area of rotat ed region)(distance traveled
the corresponding point (x, f(x)) on the gra ph of f by centroid) . For surfaces, Area = (length of rotated
its elf. arc)(distance traveled by centroid) .
Separating the variables (pp. 251, 310): The most Trapezoi dal rule (p. 19): A numerical way of approx-
elementary technique for transforming a differential imating a definite integral by slicing the region under
equation so that it can be solved. a graph into trapezoids and adding the areas of the
trap ez oids . The technique is similar to Simpson's
Simpson's rule (pp. 233, 305): A numerical way of
rule, but the graph is replaced by parts of linear
approximating a definit e integral by replacing the functions rather than by parts of quadratic functions.
graph of the integrand with segments of parabolas,
then summing the areas of the regions und er the Trigonometric substitution (p. 456) : An algebraic
parabolic segments. The technique is similar to the method for finding antiderivatives where the
trapezoidal rule, except that the grap h is replaced by integrand involves quadratics or square roots of
segments of quadratic functions instead of by quadrati cs.
segments of linear functions.
Uniqueness theorem for deriva tives (p. 264): The
Slope field (pp. 309, 326): A graphical represe ntation property that states that if two funct ions have
of the slope specified by a differential equatio n at id en tical deri vatives everywhere in an interval, and
each grid point in a coordinate system. A slope field, hav e at least one point in common, then they are the
which can be generated by grapher, allows graphical same ("unique") function .
solutions of diff erentia l equations .
Vector (p. 533): A quantit y that has both magnitude
Speed (p. 99): The absolute value of velocity. and direction. Position, velocity, and acceleration
vectors are us ed to analyze motion in two or three
Squeeze theorem (p. 112): If f(x) is always between dimensions.
the values of two other functions, and the two other
functions approach a common limit as x approaches Velocity (p. 98): The instantaneous rate of change
c, then f(x) also approaches that limit. of displacement.
Step discontinuity (p. 53): If f(x) approaches differ- Vertical asymptote (p. 53): A vertical line x = c that
ent numbers from th e right and from the left as x the gra ph of a function does not cross because the
approaches c, then there is a step discontinuity at limit of f(x) as x approaches c is infinite .
X = C.
Washers (p. 387): A technique for slicing a solid of
Taylor series (p. 615): A power series representing a revolution into thin slices so that each point in the
function as non-negative integer pow ers of (x - a). washer is virtually the same distance from a plane
The coefficients of the terms are such that each perpendicular to the axis of rotation as is the sample
order derivative of the series equals the correspond- point. Washers are used to set up integrals for calcu-
ing order derivative of the function at the point lating the volume, mass, moment, etc., of a solid
where x= a. object.
Techniques of calculus: There are four major kinds
of technique used in calculus : algebrai c, numeric,
graphical, and verbal.
764 Glossary
Index of Problem Titles
Acceleration Data Problem, 549 Beam Moment Probl em, 5 76 Confirmation of Quotient Formula
Acceleration Problem, 122 Beanstalk Problem, 110 Prob lem, 140
Accurate Graph of a Cubic Function Biceps Problem, 650 Conical Cup Probl em, 592
Problem , 13 Bicycle Frame Des ign Problem, 159 Conical ose Cone Problem, 378
Accurate Graph of a Rational Function Biological Half-Life Problem, 3 14 Conical Reservo ir Problem, 562
Problem , 13 Black Hole Problem, 139 Conical Tank Genera liza tion Problem,
Aircraft Carrier Landin g Problem, 22 Board Price Problem, 5 522
Airplane Wing Problem I, 582 Bouncing Ball Problem, 605 Conical Water Tank Problem, 521
Airplane Wing Problem ll, 582 Bowl Probl em, 482 Cont inued Exponen tiat ion Function
Algebraic Technique Probl em, 349 Bridge Problem, 532 Probl em, 300
Alternating Series Remaind ers Prop ert y Buckminster's Elliptical Dome Problem, Cont inu ed Exponentiation Problem, 270
Problem, 64 1 567 Cont inui ty Proof Problem, 159
Another Definition of rr Problem , 501 Building Problem, 3 77 Continuous Compoundin g of Int ere st
Another Theorem of Pappu s Probl em, Bungee Probl em, 33 Problem, 289
59 1 Con vergence and Divergence Problem,
Answer Check Probl em, 239 614
Answer Verification Probl em, 468 Calvin and Phoebe's Accelerat ion Cooling Tower Problem, 415
Arc Length of a Parabola Problem, 460 Problem, 76 Cosine Area Problem, 45 0
Area Check Problem, 385 Calvin and Phoebe' s Commutin g Cosine Fun ction Problem, 70
Area of an Ellipse Probl em, 460 Probl em, 528 Cosine Function Series Problem, 613
Area of a Region Param etr icall y Calvin's Swimmin g Problem, 505 Cubic Circle Problem, 171
Probl em, 384 Campus Cut-Across Probl em, 549 Cubic Function Problem I, 87
Area Problem, 228 , 445, 467, 468 Carbon 14 Dating Problem, 3 14 Cubi c Function Problem II, 87
Area Probl em I, 454 Carbon Dioxide Problem, 127 Cubic Parabola Region Problem, 593
Area Probl em II, 455 Card ioid Area Probl em, 455 Cubic Paraboloid Problem I, 413
Astroid Problem, 164 Carnot Cycle Problem, 562 Cubi c Paraboloid Problem II, 4 14
Average Radius Problem, 461 Car on the Hill Problem, 510 Cup Problem, 375
Average Value Problem , 549 Car Problem , 103 Curve Sketching Review Problem, 385
Average Velocity for Constant Car Trade-In Problem, 314 Cylind er in Cone Problem, 378
Acceleration Problem, 515 Catch-Up Rate Probl em, 139 Cylind er in Cubic Paraboloid Problem ,
Average Velocity for Other Chain Problem, 408 428
Accelera tions Probl em, 516 Chair Work Problem, 5 5 7 Cylinder in Paraboloid Problem , 378,
Average Velocity from Acceleratio n Chapt er Logo Probl em, 430 532
Problem, 515 Check th e Answer by th e Table Cylind er in Sphere Prob lem, 3 78
Average Velocity Problem, 549 Problem, 50 Cylind er-in -th e-Cone Probl em 1, 53 1
Average Versus Instantaneo us Velocity Chemical Reaction Problem, 313 Cylind er-in-th e-Cone Problem 11, 531
Problem, 104 Chemoth erapy Problem, 299
Average Voltage Probl em, 516 Chuck's Rock Problem, 72
Circle Area Formula Problem, 460 Dam Leakage Probl em, 318
Circle Problem, 164,171 Dam Probl em, 581, 654
Bacteria Problem , 313 Cissoid of Diodes Problem, 175 Darth Vader's Problem, 520
Bacteria Spreading Probl em, 519 City Land Value Probl em, 589 Daylight Prob lem, 117
Balloon Problem, 519 Clock Probl em , 125, 166 Definition of e Journal Probl em , 278
Balloon Volume Probl em, 110 Column Scroll Problem, 423 Definition of Limit Probl em, 13
Barn Ladd er Problem , 521 Compound Inter est Problem, 208, 270, Degree-Days Probl em, 228, 588
Baseball Line Drive Problem, 159 277, 348, 605 ouand oxProblem, 110
Baseball Line Drive Problem (Second Compound Inter est Problem I, 315 Deltoid Probl em , 165
Inning) , 209 Compound Interest Problem II, 316 Dependence on Initial Conditions
Baseball Problem, 544 Concavi ty Concept Problem, 367 Problem, 330
Base of Natural Logarithms Probl em, Cone in Hemisph ere Problem, 522 Depreciation Problem, 284
267 Cone of Light Problem , 522 Derivative and Antiderivative Problem,
Base Runn er Problem, 520 Cone Probl em , 390 122
Bathtub Problem, 519 Cone Volum e Formula Proof Problem, Derivative dy/ dx fo r Polar Coor dinat es
393 Problem, 425
765
Derivative from Grap h Problem, 24 Exact Int egra l Conjecture Problem, 24 Hyperbolic Function Graphing Problem,
Derivative Graph and Table Problem, Exact Value of a Derivative Problem, 35 479
140 Exit Sign Problem, 152 Hyperbolic Radian Problem, 483
Derivative of a Power Induction Exponential and Polynomial Look-Alike Hyperbolic Sine and Cosine Problem,
Problem, 135 Problem, 369 614
Derivative of Cosecant Problem, 144 Exponential Function Problem, 70, 88 Hyperbolic Substitution Problem, 479
Derivative of Cotangent Problem, 144 Exponentia l Funct ion Series Problem, Hyperbo loid Problem, 461
Derivatives and Continuity Problem, 73 613
Derivative Two Ways Problem, 135 Exponent ial Horn Problem, 391
Derivative Verificat ion Problem, 483 Exponential Region and Solid Problem, Ida's Speeding Ticket Problem, 515
Derivative with Variable Base and 574 Implicit Relation Problem I, 408
Exponent Generalization Problem, Implicit Relation Problem II, 408
270 Infinite Curvature Problem, 368
Difference Quotient Accuracy Problem, Fata l Error Problem, 409 Infinite Derivative Problem, 430
89 Ferris Wheel Problem, 115, 175 Infinite Overhang Problem, 641
Different Axis Problem I, 391 Field Worth Prob lem, 589 Infinite Paint Bucket Problem, 490
Different Axis Problem 11,391 Figure Skating Problem, 545 lnf'inity Minus Infini ty Problem, 288
Differential Equation General izat ion Film Festival Problem, 343 Inflation Problem, 550
Problem, 323 Flat Tire Problem, 12 Inscribed Squares Problem, 604
Differential Equation Problem, 349 Flooded Ship Problem, 562 Integra l of csch x Problem, 500
Discrete Data Problem, 348 Floodgate Problem, 583 Integral as a Limit Problem, 24
Discussion Problem: Meaning of Limit, Follow-Up Problem, 640 Integral of In Generalization Problem,
13 Football Problem, 2-4 445
Displacement from Velocity Problem, Foot Race Problem, 70 Integral of In Problem, 300
104 Force and Work Problem, 302 Integral of Secant Cubed Problem, 451
Displacement Problem, 122, 227, 243 Fran's Optimal Study Problem, 530 Integral of sech x Problem, 500
Distance from Velocity Problem, 194 Integra l of the Natural Logarithm
Distance Problem, 126 Problem, 439
Divergence by Oscillation Problem, 489 Gateway Arch Problem, 482 Integral Review Problem, 270
Divided Stock Pen Problem, 3 72 General Cylinder in Cone Problem, 3 79 Integral Tab le Problem, 194
Diving Board Problem, 590 Generalization Problem, 316 Int egral Test Problem, 640
Door-Closer Problem, 2 70 Generalized Wedge Problem, 393 Integral Verification Problem, 483
Dot Product Problem, 546 General Volume of a Sphere Problem, Int egrat ion by Parts Problem, 483
Double Argument Properties Problem, 394 Integration Surprise Problem!, 483
135 Geometric Ser ies as an Upper Bound Introduction to Reduction Formulas
Double Int egration Airplane Wing Problem, 647 Problem, 445
Problem, 583 Glacier Problem, 75 Inverse Hyperbolic Funct ion Graphing
Double Inte gration Variable Pressure Golden Gate Bridge Problem, 407 Problem, 4 79
Problem, 583 Golf Course Problem, 230, 384 Inverse Tangent Series Problem, 628
Drug Dosage Problem, 604 Gompertz Growth Curve Problem, 344 Involu te Problem, 166
Duct Problem, 3 75 Graph ical Ana lysis Problem, 152
Graph ical Verification Problem, 110
Graphing Problem, 348 Journal Problem, 90, 118, 141, 195, 213,
Electrical Circuit Problem, 321 Group Discussion Prob lem, 118 261,270,284,339,349,369,379,
Elevated Walkway Problem, 526 402,445,455,492,529,621,642
Elevated Walkway Problem Revisited,
527 Hanging Cha in or Cable Problem, 480
Ellip se Area Formula Problem, 461 Heat Capacity Problem, 237,241,586 Kinetic Energy Problem, 521
Ellip se Area Problem, 384 Heat Problem, 229
Ellipse Length Investigat ion Problem, Hemispherical Railroad Problem, 552
409 Higher Math Prob lem, 96 Ladder in the Hall Problem, 375
Ellipse Problem, 164, 519 Highway Cut Problem, 39-4 Ladder Problem, 374
Ellipsoid Mass Problem, 574 Historical Problem, 620 Latera l Area of a Cone Problem, 416
Ellipsoid Problem, 415, 461, 573 Historical Problem: Newton's Method, Latera l Area of a Frustum Problem, 416
Elliptical Nose Cone Problem, 379 175 Lava Flow Problem, 277
Elliptical Path Problem, 542 Historical Problem - The Second Leaking Bucket Problem, 561
Elliptical Table Problem, 23 Derivative Test, 368 Leaky Tire Problem, 124
Epidemic Problem, 467 Horn Problem, 392 Length of a Circle Problem, 408
Equation from Graph Problem, 74 Hot Tub Problem, 323 !'Hospital's Surprise Prob lem!, 288
Equivalent Answers Problem, 468 Hot Tub Problem, Cont inu ed, 345 Light on the Monument Problem, 144
Error Function Problem, 238 How the Grapher Works Problem, 88 Limacon Area Problem, 455
Escape Velocity Problem, 332 Hyperbola Area Problem, 461 Limit and Function Interchange Journal
Exact Derivative Problem, 51 Hyperbola Prob lem, 171 Problem, 278
766 Inde
x of Problem
Titles
Limit of (sin x)/x Probl em, 118 Newton 's Law of Cooling Problem, 322 Product of Three Functions Problem,
Limit of a Composit e Fun ction Prob lem, New York to Los Angeles Problem, 551 134
50 Nitrogen 17 Problem, 313 Program for Riemann Sums Problem,
Limit of a Const ant Probl em, 49 umber Problem I, 530 20 1
Limit of a Constant Times a Function Number Problem ll, 530 Program for Trapezoidal Rule Problem,
Problem, 49 Num erical Answer Check Problem, 152 23
Limit of a Fun ction Plu s a Function Numer ical Derivative Error Problem!, 90 Proof Problem, 445
Problem, 49 Numerical Versus Exact Derivative Properties of ln Problem, 261
Limit of a Product Probl em, 50 Problem, 97 p-Series Problem I, 647
Limit of a Quotien t Prob lem, 50 p-Series Problem II, 647
Limit of Riemann Sum Problem, 201, p-Series Problem lll, 647
402 Oblique Cone Problem, 429 Punctured Tire Problem, 311,341
Limit of x Problem, 50 Odd Function and Even Function Pyramid Problem, 392
Limits Appli ed to Derivatives Problem, Derivative Function, 135
45 Oil Truck Problem, 582
Limits Applied to Int egrals Prob lem, 65 Oil Viscosity Prob lem, 429 Quartic Function Probl em I, 88
Line Problem, 424 Oil Well Problem, 181, 593 Quartic Function Prob lem II, 88
Line Segment Problem, 165 "Old Problem" New Problems, 466 Quarti c Parabola Tank Problem, 531
Ln-Curved Surface, Problem I, 413
Ln-Curved Surface, Probl em II, 413
Local Linearity Problem, 82 Paint ed Wall Prob lem, 589 Rabbit Population Overcrowding
Local Linearity Probl em I, 187 Parabola Problem, 163 Problem, 330
Local Linearity Problem II, 187 Parabo la Surprise Prob lem !, 408 Rabbit Population Problem, 283
Local Maximum Prop erty Problem , 3 79 Parabolic Path Problem I, 541 Radar Problem, 15 2
Local Nonlin ear ity Probl em, 83 Parabolic Path Problem II, 542 Radio Dial Derivative Probl em, 261
Logistic Cur ve Problem, Algebraical l y, Parabolic Path Probl em III, 543 Radio Wave Integral Problem, 245
468 Parabo lic Region Problem, 384 Railroad Curve Problem, 158
Look Ahead Problem, 259, 269, 282 Paraboloida l Tank Problem, 562 Ramjet Problem, 341
Look Ahead Problem Follow Up, 259, Parabo loid Mass Problem, 574 Ratio of Terms Problem, 621
270 Paraboloid Moment Conjecture Ratio Techniqu e and p-Series Problem,
Luke and Leia's Trash Compactor Prob lem, 594 647
Problem, 520 Paraboloid Problem, 390, 413, 573 Reciprocal Curved Surface Problem I,
Paraboloid Surface Area Problem, 415 413
Paraboloid Volume Formula Prob lem, Reciprocal Curved Surface Problem II,
Magnet Problem, 592 391 413
Mass of the Earth Problem, 567 Parametric Curve Problem, 402 Rectangle in Parabola Problem, 377
Mathematical Indu ction Problem - Pendulum Problem, 5, 116 Rectangl e in Sinusoid Problem, 377
The Limit of a Power, 51 Periodic Motion Problem, 510 Rectangle Problem I, 520
Maximum-Minimum Review Problem, Phoeb e's Space Leak Prob lem, 345 Rectangle Problem II, 520
385 Phoebe's Speeding Problem, 586 Regular Deposits Problem, 605
Meaning of Limits Problem, 24 Physic s Formula Problem, 512 Reimann Sum Limit Problem, 391
Mean Value Theor em Problem, 348 Piecewise Continuity Problem, 491 Relative Accuracy Problem, 240
Meg's Velocity Problem, 510 Pig Sale Probl em, 532 Repeated Roots Probl em, 135
Memory Retention Problem, 297, 342 p-Integral Prob lem , 489 Resort Jsland Causeway Problem , 550
Minimum Path Discovery Problem, 527 Pipeline Prob lem, 526 River Bend Problem, 545
Minimum Path Genera lizat ion Problem, Pipeline Prob lem, Near Miss, 527 Robinson Crusoe Follow-Up Problem,
527 Pizza Delivery Prob lem, 51 528
Misconception Problem, 96 Planeta ry Motion Problem, 522 Robinson Crusoe Generalization
Mistake Problem, 409 Playground Problem, 116 Problem, 528
Moment of Arc Length Prob lem, 591 "Plus C" Problem, 225 Robinson Crusoe Problem, 528
Moment of Inert ia Problem, 588 Point on a Parabola Problem, 521 Rocket Car Problem, 588
Moment vs. Moment Problem, 594 Point on a Tangent Graph Problem, 521 Rocket Problem, 347, 510, 530
Motel Problem, 372 Pole Dance Problem, 136 Roller Coaster Problem, 546
Motor Oil Viscosity Problem, 53 1 Popeye and Olive Problem, 548 Rolling Tire Problem, 12
Mystery Problem, 428 Population Problem, 251,259,310 Root Mean Square Deviation Problem,
Population Problem Revisited, 278 516
Postage Stamp Probl em, 70 Rotated Rectangle Generalization
Naive Graphin g Problem, 367 Power Line Problem, 481, 552 Problem, 375
Natural Log Series Probl em, 614 Practical Calculation of Pi Problem, 652 Rotated Rectangle Problem, 3 75
Negative Velocity Problem, 18 Predator-Pr ey Problem, 342 Rotated Sinusoid Prob lem, 413
New Int egra l Problem I, 392 Present Value Problem, 650 Rotation of Solids Problem, 575
New Int egral Problem II, 392 Preview Problem, 395 Rover's Tablecloth Problem, 549
New Jersey Turnpike Problem, 210 Product of n Functions Conjecture Rumor Problem, 466
Prob lem, 134
Inde
x of Prob
lemTitles 767
Sample Point Problem, 20 1 Sports Car Problem, 17 Tugboat Problem, 520
Scuba Diver Problem, 526 Spring Problem, 167, 561 Tunn el Problem, 586
Scuba Diver Problem Revisite d, 527 Squeeze Theorem Introdu ction Two Cone Problem, 566
Searchlight Probl em, 66 Problem, 73 Two Constants Problem, 59
Secant Curve Region Problem , 574 Squeeze Theorem Problem, 118 Two-Corra l Problem, 3 73
Second Moment of Area Probl em, 575 Stadium Problem, 408 Two Cylind er Problem, 566
Second Moment of Volume Problem, Steep ness of a Hill Problem, 187 Two-Field Problem , 3 73
593 Storage Battery Problem, 428
Second Moments for Plane Regions "Straight Point " Problem, 430
Problem, 5 75 Subm arine Problem, 394 Unbound ed Region Area Problem, 445
Semicubical Parabola Problem, 163 Subm erging Cone Problem, 552 Unknown Integral Probl em, 40 2
Series with lmaginar y Numb ers Subway Problem, 511 Uppe r Bound Problem, 501
Problem, 652 Sum of the Cub es Problem, 244 Uranium Fuel Pellet Problem, 566
Shark Probl em, 125 Sum of the Squares Problem, 244
Ship's Bulkhead Problem, 581 Surpri se Funct ion Problem!, 59
Ship Problem, 657 Sweeps tak e Probl em I, 32 1 Valu es of ex from Values of e - x
Shortest-Distance Probl em, 374 Sweepstake Problem II, 321 Problem, 64 7
Simpson's Rule from Equation Problem, Sweetheart Problem, 69 Variable Attraction Problem, 587
239 Swim and Run Problem , 526 Variable Density Problem, 592
Simp son's Rule Review Problem, 468 Swimming Pool Chlorinat ion Problem, Velocity from Displa ceme nt Probl em ,
Sine-Integra l Function Problem, 237 345 104
Sine Series Problem, 614 Symmetr ic Difference Quotient Velocity Problem, 236
Sinusoidal Land Tract Prob lem , 589 Problem, 89 Velocity Vector Limit Problem, 544
Sinusoidal Path Problem, 545 Vocabulary Problem 1, 638
Sinusoidal Region Problem, 384 Vocabulary Problem II, 638
Sinusoid Length Investi gation Problem, Table Moving Problem, 562 Vocabulary Problem III, 638
409 Table of Int egrals Problem, 497 Vocabulary Problem rv, 638
Sinusoid Problem I, 88 Tab les of Differentials Problem, 188 Volum e of an Ellipsoid Problem, 394
Sinusoid Problem II, 88 Tangent to a Graph Problem, 35 Volume Probl em, 445, 468, 47 1, 62 8
Sin x for Any Argument Using a Value Tan gent Series Probl em, 620 Volum e Problem I, 455
of x in [O, rr/4 ] Problem, 648 Taylor Series Proof Probl em, 620 Volume Problem II, 455
Skewness Problem, 590 Tensile Strength Test Problem, 23 7 Volum e of an Unbounded Solid
Sky Diver's Acceleratio n Problem, 103 Terminal Velocity Problem, 331 Problem, 489
Slag Heap Problem, 522 Theater in th e Round Problem, 596
Slide Problem, 18 Theor em of Pappus Problem, 5 77
Snell's Law of Refract ion Problem, 5 29 Thr ee-Dim ensiona l Vector Problem, 54 7 Walking Problem, 21
Snowflake Curve Problem, 606 Tin Can Leakage Probl em, 317 Washer Slices Probl em, 391
Spaceship Problem, 21, 79,124,511, Tin Can Problem, 375 Water over th e Dam Problem, 22
561 Tire Pump Work Problem, 261 Water Pipe Probl em, 586
Spaceship Work Problem, 491 Tolera nce Problem (Epsilon and Delta), Wedge Problem, 393
Sphere Probl em, 393 89 Weir Problem, 578
Sphere Rate of Change of Volum e Toro id Problem, 577 "Which One Wins!" Problem , 630
Problem, 415 Total Cost Problem, 229 Wind Force Problem, 593
Sphere Total Area Formula Problem, Track and Field Problem, 374 Wire-Pulling Problem, 587
414 Trapezoidal Rule Error Problem, 23 Witch of Agn esi Probl em, 165
Sphere Volume and Surface Problem, Triangle Centroid Probl em, 593 Work Problem, 66, 228, 592
414 Triang le und er Cota ngent Problem, 377
Sph ere Zone Problem, 414 Triangle und er Expon ential Curve
Spherical Water Tower Problem, 562 Problem, 3 77 Zero Times lnfinity Problem, 66
Spider and Clock Problem, 551 Triangular Cross-Section Problem, 392 Zero to the Zero Problem, 288
Spiral Path Problem, 543 Trough Problem, 580 Zero/Zero Problem, 284
Spiral Problem, 408 Tru ck Passing Problem, 552 Zone of a Paraboloid Problem, 415
Spleen Mass Probl em, 236 Truck Problem, 530
768 Indexof Problem
Titles
General Index
Absolute convergence, 635 Calculus Convergence, 599, 603
Absolute value, 533 fundamen ta! theorem of, 215-217, absolute, 635
Absolute value function, 6 304 of alternating harmonic series,
Acceleration, 98-102, 305 second form of, 255 633-634
for linear motion, 506-509 of hyperbolic and inverse hyperbolic conditional, 63 5
normal component of, 538 functions, 472-478 interval of, 621
tangential component of, 538 of natural exponential functions, 280 for series, 622-626
Acceleration vector, 533, 535 Calculus journal, 31-32 of p-series, 632-633
components of, 540 Cardioid, 423 radius of, 625-626
Algebraic derivative, 93 Catenary, 408, 433, 472 infinite, 626
Algebraic fraction, improper, 465 Center of mass, 568-573 of series, ratio technique for, 624
Algebraic function, 435 Center of pressure, 578-580 tests for series of constants, 635
rational, 462 Centroid, 568-573 Convergence of sequences, 631-632
Alternating harmonic series, definition of, 5 70 Convergence of series
convergence of, 633-634 Chain rule, 107-109, 303 absolute, 634-637
Alternating series convergence test, 634 parametric, 161, 304 at ends of convergence interval,
Alternating series test, 635 Change, rate of, by equation, graph, or 630-637
Amplitude, 113 table, 6-10 integral test for, 640
Anesi, Maria Gaetana, 165 Change of base property, 2 78, 294 root technique for, 630
Angle psi, in polar coordinates, 425 Circle Convergent geometric series, 600-601
Annual percentage rate, 605 involute of, 166 Converse
Antiderivative, 33, 98, 119-121, 182, osculating, 548 of direct proportion property of
304 Circular function, 6 exponential functions, 316
definition of, 119 Cissoid of ciocles, 422 of theorem, 212
formal, 189-192 Common ratio, 600, 603 Corollary, 211
general equation for, 182 Completeness axiom, 67, 204 Cosecant function
particular equation for, 182 Composite function, 105-106, 303 derivative of, 142
of reciprocal function, 252-253 derivative of, 107-109 even powers of, 453
of zero, 212 limit of, 48 integrals of, 292
Approximation, linear, 183-186 Compound interest, 315 Cosine function, 105
Archemedian spiral, 423 Computer software, 446-449 derivative of, 105
Arclength,403-406,420-421 Concave downward, 353 double argument properties for, 452
Arctan x, 147 Concave upward, 353 hyberbolic, 4 72
Area Concavity, 356-358 proof and application of, 111-115
of ellipse, 461 Conchoid of Nicomedes, 423 square of, 452-453
of plane region, 380-382 Conclusion, 203 Cotangent function
for polar coordinates, 417-421 Conditional convergence, 635 derivative of, 141-142
of surface of revolution, 410-412 Cones, frustums of, 410-412 integrals of, 292
Astroid, 164 Constant, 90 Critical point, 354-362
Asymptote, vertical, 53 Constant function Cubic function, 353
Average radius, 461 derivative of, 93 point of inflection of, 366
Average velocity, 536 limit of, 48 Cubic parabola, 353
Axial variation, 563 Constant of integration, 182 Curvature, 356-358
Axioms, completeness, 67, 204 Continued exponentiation function, 300 calculation of, 547
Continuity, 52-56, 153-156 infinite, 359
definition of, 54 radius of, 548
Backward difference quotient, 84 piecewise, definition of, 491 Curve, length in polar coordinates, 420
formula for, 89 Continuous function Cusp,361
Base e, 272 limit-function interchange for, 273 definition of, 54
Biological half-life, 314 maximum and minimum values of, Cycle, 113
Boundary condition, 310 212 Cycloid, prolate, 546
Bowditch curves, 167 Continuous graph, 67 Cylindrical shells, 396-399
Contrapositive, 15 5 differential of volume for, 397
769
Definite integral, 181 Differential equation, 119, 218, 251, integra ls of, 291
definition of, 195-200, 303 310 properties of, 294
functions defined as, 254-256 leading to polynomial function, 343 derivative and integral of, 268-269,
properties of, 219-224 for real-world applications, 316-320 292
by trigonometric substitution, solution of, 310 direct proport ion proper ty of, 309
458-459 graphical by using slope fields, 326 converse of, 316
de !'Hospital, G. F. A., 285 numerical using Euler's method, natural, 279-281
Deltoid, 165 333-334 calculus of, 280
Density, 563-564 series, 653 power series for, 607-608
Dependent variable, 4 Differential of volume, for cylindrical properties of, 293
Derivative(s) shells, 397 Exponent ial function series,
of a function, 9- 10, 303 Differentiation, 92, 492 comparison test for, 640
algebraic, 93 implicit, 149, 168, 170, 304 Exponential growth and decay, 310-312
of a product, 126 logarithmic, 268-269 Exponentiation function , con tinued ,
of base b logarithmic function, properties of, 93 300
272-275 Diocles, 175 Exponents, 263
of composite function, 107-109 Direction field, 326 negat ive, proof of power ru le for, 140
definition of, 80-81, 91, 303 Discontinuity power formula for, 97
of dependent variable with respect to removable, 26 properties of, 264
independent variable, 4 step, 27, 30, 53, 62 Extreme value theorem, 67, 70
of exponential function, 268-269, Discrete point, 67
292 Displacement, 98-102
geometrical interpretation of, 81 force times, 5 5 7 Factorial function, definition of, 490
of geometric series, 606 for linear motion, 506-509 Factoria l reciprocal series, convergence
graphical interpretation of, 79 for motion along a line, 505 of, 640
of hyperbolic function, 474 phase, 113 Family,offunctions, 183,310
of implicit relation, 168-170 vert ical, 113 First moment, 568
of integral, with variable upper and Distance First-order infinitesimal, 595
lower limits, 301 for linear mot ion, 506-509 Foot-pounds, 55 7
of integral form, 2 5 5 for motion along a line, 505 Force
of inverse hyperbolic function, 4 76 related rates, 517-518 exerted by variable pressure,
of inverse of a function, 153 Divergence, 599, 603 578-580
of inverse trigonometric function, of harmonic series, 633 moment of, 5 78
145-150,470 Divergent geometric sequence, 601-602 times displacement, 557
Of Jn X, 255 Doubly curved, 410-412 variable, work done by, 557-560
nth, 606 Forward difference quotient, 84
of parametric function, 131, 160-162 formula for, 89
of position vector function, 534-541 e, 272-275 Foucault pendulum, 129
of power function, 90-95 definHion of, 2 72, 2 78 Fourier series, 245
of power series, 610 Elementary function, 435 Fractal curve, 606
of product of two functions, 132-133 power series for, 609-612 Fraction
of quotient of two functions, Elementary transcendental function, algebraic, improper, 465
136-138 469 partial, 337, 462
of rational power, 171 Ellipse, area of, 461 integration of rational functions
second, 100 Elliptic integral, 405 by,462-464
of a function, 353 Equal-derivative theorem, converse of, Frustums of cones, 410-412
of sine and cosine functions, 105 212 Fuller, Buckminster, 567
proof and application of, 111-115 Equation Function(s), 4. See also specif ic type
symbols for, 361-362 general, 6 average value of, 514
third, 606 particular, 6 constant times
of trigonometric function, 141-142 rate of change by, 6-10 derivative of, 93
uniqueness theorem for, 264-265, Equiangular spiral, 426 integral of, 190-191, 222
304 Error analysis, for series, 642-646 limit of, 48
Derivative function, 84-87 Error function, 628 definite integra l of, 13-16
Difference quotient, 80-81 Error function of x, 239 derivative, 84-87, 303
backward, 84 Euler, Leonhard, 333 second, 353
formulas for, 89 Euler's method, 309, 333-336 even, 135, 222
forward, 84 Even function, 135, 222 family of, 183, 310
symmetric, 84, 303 Existence theorem, 69 inside, 105, 107
Differentiability, 153-156 Explicit relation, 168 differential of, 280
Differential, 94, 183-186 Exponential forms, indeterminate, 287 inverse of, 146
definition of, 186 Exponential function, 6 derivative of, 153
base b limit of, 25-27
linearization of, 185
770 General Index
Function(s) (continued) definitions of, 472-473 of natural logarithm functions,
odd,135,222 derivatives of, 4 74 442-443
order of magnitude of, 290 integrals of, 475 with negative integrand, 220
outside, 105, 107 inverse, 475-478 of powers of trigonometric functions,
polynomial, 90 derivatives of, 476 449
product of two integrals of, 4 77 of product of two functions, 435
derivativ e of, 132-133 Pythagorean properties of, 4 74 of reciprocal functions, 251, 256-259
integral of, 435 Hypotheses, of mean value theorem, by Simpson's rule, 231-234
limit of, 46, 48 203 of sum of functions, 190-191, 222
quotient of two sum with same integrand, 221
derivative of, 136-138 between symmetric limits, 221-222
limit of, 4 7, 48 Ident ity function, limit of, 48 of trigonometric functions, 292, 447
step, 70 Image theorem, 67, 70 upper bounds for, 223
sum of, integral of, 190-191 Implicit differentiation, 149, 168, 170, Integral between symmetric limits,
sum of two 304 221-222
derivative of, 93 Implicit relation, graphs and derivatives Integral form, derivative of, 255
integral of, 222 of, 168-170 Integral sign, 189
limit of, 48 Improper integrals, 484-488 Integral test, for convergence of series,
types of, 6 definition of, 486 640
Fundamental th eorem of calculus, Increasing series test, 63 5 Integrand, 190
215-217,304 Increments , 196 negative, definite integral with, 220
second form of, 255 Indefinit e integral, 33, 119-121, 304 sum of integrals with same, 221
definition of, formal, 189-192 Integration
Indefinite integration, 189, 493 bounds of, 197
Galileo, 120 Ind ependent variable, 4 constant of, 182
Gamma function, definition of, 490 Ind eterminate forms, 26, 46 indefinite, 189, 493
Gaussian distribution, 238 kinds of, 290 limits of, 197, 220
General equation , 6, 119 limits of, 285-287 numerical by Simpson's rule and
for antiderivative, 182 Inductan ce, 322 grapher, 230-235
General solution, 310 Inertia, moment of, 573, 588 by parts, 436-439
Geometric sequence, divergent, Infinite curvature, 359 rapid repeated, 440-443
601-602 Infinite limits, 53, 60-64 by reduction formula, 446
Geometric series, 600-603 definition of, 61 by trigonometric substitution,
convergent, 600 - 601 as x approaches infinity, 62 456-459
derivatives of, 606 Infinit esimals, 594-595 Intermediate value theorem, 67-68
properties of, 603 Infinite slope, 359 converse of, 68
Global maximum, 354,361 Infinity, reciprocals of, 63 versus mean value theorem, 212-213
Global minimum, 354, 361 Initial condition, 120, 310 Interval
Gompertz function, 344-345 Inside function, 105, 107 continuity on, 54
Graph differential of, 280 differentiability on and, 153
continuous, 67 Instantan eous rate, 3-4 Interval of convergence, 621
of implicit relation, 168-170 Integrability, definition of, 197 for a series, 622-626
rate of change by, 6-10 Integral(s), 190 Inverse, ofln, 279-281
of sinusoid, 114 application of, 225-226 Inverse function, 146
Grapher of base b exponentia l function, derivative of, 153
numerical integration by, 230-235 291 Inverse function property, proof of, 284
slope fields on, 333 of constant times a function, Inverse hyperbolic functions, 475-478
Gyration, radius of, 573 190-191, 222 derivatives of, 476
definite. See Definite integral integrals of, 477
derivative of, with variable upp er and Inverse hyperbolic sine, 475
Hanging chain/cable, equation of, 481 lower limits, 301 Inverse tangent function, 147
Harmonic series elliptic, 40 5 Inverse trigonometric functions
alternating, convergence of, 633-634 of exponential function, 292 derivatives of, 145-150, 470
divergence of, 633 by grapher, 234-235 integrals of, 469-4 70
Harmonic series test, 635 from higher number to lower Invertible, 146
Heat capacity, 23 7, 241, 586 number, 221 Involute of circle, 166
Heaviside, Oliver, 463 of hyperbolic functions, 4 7 5
Heaviside method, 463-464 improper, 484-488
Higher-order infinitesimal, 595 definition of, 486 Journal, 31-32
Higher order of magnitude, 290 indefinite. SeeIndefinite integral
Hooke's law, 228 of inverse hyperbolic functions, 477
Hyperbolic cosine, 4 77-4 78 of inverse trigonometric functions, Kepler's second law, 424
Hyperbolic function, 4 72-4 78 469-470 Kepler's third law, 424
General
Index 771
Lagrange, Joseph Louis, 642 Logarithmic differentiation, 268-269 Normal component of acceleration, 540
Lagrange form, 642 Logarithmic function, 263-266 Normal distribution curve, 628
of remainder of Taylor series, 643 base b, derivative of, 272-275 nth derivative, 606
Larent series, 649 Logistic equation, 331, 337 nth partial sum, 600
Laycock, George, 284 algebraic solution of, 33 7-338 nth term, nth root of, 629-630
Least upper bound postulate, 642 Lower bound, 196 nth term test, 635
Leibniz, Gottfried, 94 Lower order of magnitude, 290
Lernniscate of Bernoulli, 432 Lower sum, 196
Length of a plane curve, 403-406, Odd function, 135, 222
One-sided limit, 54
420-421
Maclaurin, Colin, 615 symbols for, 55
!'Hospital's rule, 285-287, 305
Maclaurin series, 615-619 Operator, 94, 190
Limacon, 41 7
Mass Order of magnitude, of a function, 290
Limit, 4, 10
center of, 568-573 Oscillation, 603
absolute value definition of, 74
of a function, 25-27 of variable-density object, 563-564 Osculating circle, 548
Maxima Outside function, 105, 107
definition of, 303
in plane and solid figures, 369-372 Ovals of Cassinis, 172
formal, 27
second derivative test for, 368
graphical and algebraic approaches
Maximum problems, in motion, 529
to, 40-43 Pappus, 577
Mean value theorem, 202-207, 304
numerical approach to, 39 Parabola, 231
algebraic proof of, 205-207
equal right and left, 5 5 cubic, 353
corollary of, 211
infinite, 53, 60-64 semicubical, 163
proof illustrated by graph and table,
definition of, 61 Parameter, 160
211
as x approaches infinity, 62 Parametric chain rule, 161, 304
for quadratic functions, 244
of integration , 197, 220 Parametric equations, 160
versus intermediate value theorem
of interminate forms, 285-287 212-213 ' Parametric function, 131
one-sided, 54 derivative of, 160-162
Mendelbrot, Benoit, 606
symbols for, 55 Partial fraction, 33 7, 462
Midpoint sum, 196
of product or sum of two functions, integration of rational functions by,
Minima
46-47 462-464
in plane and solid figures, 369-3 72
properties of, 48 Partial sum, 602
second derivative test for, 368
properties involving infinity, 62-63 Particular equation, 6, 119
Minimal path problems, 523-525
of quotient of two functions, 4 7 for antiderivative, 182
Minimum path, 527
of Riemann sum, 198 Particular solution, 310
Minimum problems, in motion, 529
as x approaches infinity, 61 Partition, 196
Moment, 568-573
Limit theorems, 45-49 norm of, 246
of force, 5 78
Linear approximation, 183-186 Parts, integration by, 436-443
of inertia, 5 73, 588
Linear combination, 90 Pendulum, 116,117,160, 166-167
Motion
Linear factor, repeated, 466 in a plane, vector functions for Foucault, 129
Linear function, 6 532-533 , Period, 113
Linearity, local, 82 Phase displacement, 113
average value problems in, 513-514
Linearization, of function, 185 linear, distance, displacement, and Piecewise-continuous function
Linear motion, distance, displacement, acceleration for, 506-509 integrability of, 491 '
and acceleration for, 506-509 maximum and minimum problems in Piecewise continuity, definition of, 491
Lissagous curves, 167 529 ' p-integral, 489
Lituus, 423 p-series, 635
1n Plane slicing, volume by, 385-389
inverse of, 279-281 Natural exponential function, 279-281 Plane curve, length of, 403-406
logarithm properties of, 266 calculus of, 280 Plane figure, maxima and minima in
Local linearity, 82 Natural logarithm, 253-259, 304 369-372 '
Locally cubic, 608 base of, 305 Plane region, area of, 380-382
Locally quadratic, 608 properties of, 293 Plane slicing, 385-395
Locally quartic, 608 Natural logarithm function Planetary motion
Local maximum, 354, 361 definition of, 2 54 Kepler's second law of, 424
Local minimum, 354, 361 integral of, 442-443 Kepler's third law of, 424
Logarithm(s) Negative exponents, power rule for, Plateau point, 354
algebraic definition of, 264 proof of, 140 Point
base b, properties of, 294 Neighborhood, 112 continuity at, 54
change-of-base property for, 274 Newton, Isaac, 175 critical, 354-362
natural, 253-259, 304 Newton's method, 175-176 differentiability at, 153
base of, 305 Norm, 533 discrete, 6 7
equivalence with base e logs, 274 of partition, 246 plateau, 354
properties of, 293 Normal distribution, 238 sample, 196
properties of, 263-264
772 Gene
ral Index
Point of inflection, 353, 354-362 Ratio, common, 600, 603 error analysis for, 642-646
of cubic function, 366 Rational algebraic function, 6, 462 geometric, 600
Polar axis, 417 Rational function, integration by partial Maclaurin, 616
Polar coordinates fractions, 462-464 tail of, 631
angle psi in, 425 Rational power, derivative of, 171 Taylor, 616
lengths and areas for, 417-421 Ratio technique, 622-626 Series of constants, tests for
spiral in, length of, 461 for convergence of series, 624 convergence of, 63 5
Pole, 417 Reciprocal function Shaaf, William, 653
Polynomial function, 6, 90 antiderivative of, 252-253 Sigma notation, 611
differential equation leading to, 343 integral of, 251, 256-259 Signum function, 60
Position vector, 533, 535 Reciprocals, of zero and infinity, 63 Simpson's rule, 305
Position vector function, derivatives of, Reduction formula, 446-449 numerical integration by, 230-235
534-541 definition of, 447 Sine function, 105
Power formula, for exponents, 97 Related rates, 517-523 derivative of, 105
Power function, 6 Remainder of a series, 631 double argument properties for, 452
derivative of, 90-95 Removable discontinuity, 26 proof and application of, 111- 115
Power rule, for negative exponents, Resistance, electrical, 3 21 square of, 452-453
proof of, 140 Riemann sum, 194-200, 213-215, 303 Sine-integral function, 235, 629
Powers, with negative bases, 359 definition of, 196 Singly-curved surface, 410
Power series, 599, 616 limit of, 198 Sinusoid, 88, 105, 113
definition of, 609 RolJe, Michel, 204 graph of, 114
derivatives of, 610 Rolle's theorem, 204-207 Sinusoidal axis, 113
for elementary functions, 609-612 geometrical proof of, 204-205 Skewness, 590
expanding the function as, 609 proof illustrated by graph and table, Slope, 179
for exponential functions, 607-608 210-211 infinite, 359
Predator-prey, 338-339 Root mean square deviation, 516 of tangent line, 81
Present value, 650 Root technique, 629-630 Slope field, 309, 326
Pressure, center of, 578-580 for convergence of series, 630 solution of differential equations
Principal branch, 147 and, 326-328
Product Snellius, WilJebrod, 529
derivative of, 126 Same order of magnitude, 290 Snowflake curve, 606
logarithmic differentiation for, 269 Sample points, 196 Solid, volume of
of two functions, derivative of, Sandwich theorem, see squeeze by cylindrical shelJs, 396-399
132-133 theorem by plane slicing, 385-389
Projection, vector and scalar, 539 Scalar projection, 539 Solid figure, maxima and minima in,
Prolate cycloid, 546 Scalar quantities, 533 369-372
p-series, convergence of, 632-633 Secant derivative formula, confirmation Solid of revolution, 388
p-series test, 63 5 of, 144 Speed,99, 535
Pythagorean theorem, 168, 403, 535 Secant function Spherical shelJs, 414
derivative of, 142 Spiral, length in polar coordinates, 461
even powers of, 453 Squeeze theorem, 112, 11 7
Quadratic function, 6, 317 integral of, 292 Standard deviation, 239
mean value theorem for, 244 inverse, derivative of, 149-150 Step discontinuity, 27, 30, 53, 62
unfactorable, 465-466 Secant line, tangent line as limit of, 84 Step function, 70
Quotient Second derivative, 100 Subintervals, 196
difference. See Difference quotient Second derivative test, 368 Sufficient conditions, 203
logarithmic differentiation for, 269 Second moment, 5 72 Surface of revolution, area of, 410-412
of two functions, derivative of, Semicubical parabola, 163 Surfaces, theorem of Pappus for, 591
136-138 Separation of variables, 310 Symmetric difference quotient, 84, 303
Sequence, 602 formula for, 89
convergence of, 631-632
Radial variation, 563 Sequence of partial sums, 602
Radical, 190 Series, 602 Table, rate of change by, 6-10
Radicand, 190 convergence of Tail of a series, 631
Radius absolute, 634-637 Tangent, 81
average, 461 at ends of convergence interval, Tangent derivative formula,
of curvature, 548 630-637 confirmation of, 144
of gyration, 573 integral test for, 640 Tangent function
Radius of convergence, 625-626 interval of, 622-626 derivative of, 141-142
infinite, 626 ratio technique for, 624 integral of, 292
Rate of change, by equation, graph, or remainder of, 631 inverse, 14 7
table, 6-10 root technique for, 630 algebraic derivative of, 149
Rates, related, 517-523
General
Index 773
Tangent line proof illustrated by graph and indep end ent , 4
as limit of secant fines , 84 tab le, 210-211 separating, 25 1, 310
slope of, 81 squeeze, 112, 117 Vector fun ction, 533
Tangent ial component of acceleratio n, uniqueness, for derivatives, Vector proj ection, 539
540 264-265 Vector quantity, 533
Taylor, Brook, 615 w1iqueness for derivatives, 304 Vectors, 533-541
Taylor series, remainder of, Lagrange Third derivative, 606 Velocity, 98-102, 305
form of, 643 Three-leaved rose, 422 average, 513-5 14, 536
Taylor series expans ion, 615-619 Tolerance, 85 terminal, 103, 33 1
Tens ile stre ngth, 237 Torq ue, 568 Velocity vector, 533, 535
Termi nal velocity, 103, 331 Trapezo id al rule, 19-20 Vertica l asymptote, 53
Term index, 600, 602 Trapezoids, definit e int egrals by, Vertical displacement, 113
Terms, 600, 602 18-20 Volume
Theorem of Pappus, 568-573 Trigonometric functions, 6 by cylindrical shells, 396-402
for surfaces, 591 derivatives of, 141-142 by plane slicing, 385-389
for volum es, 5 77 integrals of, 292 , 447 th eorem of Pappus for, 577
Theorems inverse von Koch, Helge, 606
converse of, 212 derivatives of, 145-150, 470
existence, 69 int egra ls of, 469-470
extreme valu e, 67, 70 powers of Witch of Agnesi, 165
fundamenta l of calculu s, 215-217, int egra ls of, 449 Work, 557
304 integrating, 451-453 definition of, 558
second form of, 255 Trigonometric sub stitution done by variab le force, 557-560
image, 67, 70 definite inte gra ls by, 458-459
intermediate valu e, 67-68 integration by, 456-459
converse of, 68 for negative values of x, 462 Yates, Robert C., 175
intermediate valu e versus mean
value, 212-213
limit, 45-49 Uniqu eness theor em, for derivatives, Zero
mean value, 202-207 264-265 , 304 antiderivative of, 212
algebraic proof of, 205-207 Upper bound, 196 infinit e radius of convergenc e
corollary of, 211 Upp er sum, 196 and, 626
proof illustrated by graph and reciprocals of, 63
tab le, 211 Zero vector, 533
Pythagorean, 168,403,535 Variable-factor products, 584-585
Rolle's, 204-207 Variables
geometrica l proof of, 204-205 dependen t, 4
774 Genera
l Index
Photo Credits
Chapter 1
1: Peter Yates/Saba. 17: Peter Yates/ Saba. 22: courtesy Paul Foerster.
Chapter 2
37: Mary E. Messenger / Photo Network. 51: Mary E. Messenger / Photo Network.
76: Biological Photo Service/ Terraphotographics.
Chapter 3
77: Chip Maury/ AP/ Wide World Photos. 103: Chip Maury/ AP/Wide World Photos .
116: courtesy Paul Foerster . 125: Eric Sander / Gamma Liaison.
Chapter 4
129: Griffith Observatory / Anthony Cook. 136: Dance Collection, NY Public
Library for the Performing Arts, Astor, Lenox and Tilden Foundations.
166: Griffith Observatory / Anthony Cook.
Chapter 5
179: Paul Fusco/ Magnum Photos. 188: Paul Fusco/ Magnum Photos.
229: Brophy Collection, courtesy the Bisbee Mining and Historical Museum
236: courtesy James R. Stewart, M.D.
Chapt er 6
249: UPI/ Bettmann . 2 77: Brad Lewis/ Gamma Liaison. 283: UPI/Bettmann .
299: Geoff Tompkinson/Science Photo Library/ Photo Researchers.
Chapt er 7
307: John Gerlach / Tony Stone Images. 314: Chester Higgins Jr ./ Photo Researchers.
315: John Gerlach/ Tony Stone Images . 332: NASA.
Chapter 8
351: Lou Jacobs, Jr ./ Gamma Liaison. 407: Lou Jacobs, Jr./ Gamma Liaison.
415: David Woodfall/ Tony Stone Images. 423: Kathleen Campbe ll/ Gamma Liaison.
426: Comstock.
Chapt er 9
433: Comstock / Adam Tanner. 446: courtesy Soft Warehou se. 468: UPI/Bettmann.
481: Mike Valeri/ AP/ Wide World Photos.
Chapter 10
503: NASA. 511: Jean-Marc Gibou x/ Gamma Liaison. 512: NASA. 522: Graham
Finlayson / Tony Stone Images. 526: Comstock . 545: courtes y USAC/ RS.
Chapt er 11
555: Garry Conner / PhotoEdit. 572: both, courtesy Paul Foerster.
576: Garry Conner / PhotoEdit. 577: David Lissy/ The Picture Cube.
Chapt er 12
597: Will & Deni McIntyre/Tony Stone Images. 604: Will & Deni McIntyre/ Tony
Stone Images. 650: Bob Daemmrich / The Image Works. 655: U.S. Department of
the Interior / Bureau of Reclamation.
775
You might also like
- AHA ACLS Precourse Self Assessment Answers 202392% (24)AHA ACLS Precourse Self Assessment Answers 202332 pages
- The Manual of Free Energy Devices and Systems (1991)93% (208)The Manual of Free Energy Devices and Systems (1991)128 pages
- Seventy-Eight Degrees of Wisdom - A Book of Tarot, Revised - Rachel Pollack98% (41)Seventy-Eight Degrees of Wisdom - A Book of Tarot, Revised - Rachel Pollack370 pages
- Joe Tippens' Protocol and Scientific Backing PDF86% (7)Joe Tippens' Protocol and Scientific Backing PDF2 pages
- Bessel Van Der Kolk - The Body Keeps The Score - Brain, Mind, and Body in The Healing of Trauma-Penguin (2014)100% (12)Bessel Van Der Kolk - The Body Keeps The Score - Brain, Mind, and Body in The Healing of Trauma-Penguin (2014)490 pages
- Kiss Your Dentist Goodbye A Do It Yourself Mouth Care System For Healthy - Clean Gums and Teeth by El100% (14)Kiss Your Dentist Goodbye A Do It Yourself Mouth Care System For Healthy - Clean Gums and Teeth by El210 pages
- 50 Challenging Calculus Problems (Fully Solved) - Chris McMullen100% (13)50 Challenging Calculus Problems (Fully Solved) - Chris McMullen236 pages
- Adult ADHD Test 18 Question ADHD-ASRS-v1-1 Adhd-Npf - Com Quality Approved100% (14)Adult ADHD Test 18 Question ADHD-ASRS-v1-1 Adhd-Npf - Com Quality Approved3 pages
- Fundamentals of Differential Equations (Nagle, Saff, Snider)100% (11)Fundamentals of Differential Equations (Nagle, Saff, Snider)739 pages
- Borax - The Inexpensive Detox, Arthritis, Osteoporosis and Mycoplasma Cure90% (10)Borax - The Inexpensive Detox, Arthritis, Osteoporosis and Mycoplasma Cure10 pages
- Mark Sloan, Ray Peat, Raymond Peat - The Ultimate Guide to Methylene Blue_ Remarkable Hope for Depression, COVID, AIDS & Other Viruses, Alzheimer’s, Autism, Cancer, Heart Disease, ... Targeting Mitoch100% (15)Mark Sloan, Ray Peat, Raymond Peat - The Ultimate Guide to Methylene Blue_ Remarkable Hope for Depression, COVID, AIDS & Other Viruses, Alzheimer’s, Autism, Cancer, Heart Disease, ... Targeting Mitoch171 pages
- Mathematics For Engineers I Basic Calculus100% (12)Mathematics For Engineers I Basic Calculus403 pages
- Cognitive Therapy Therapy Techniques For Retraining Your Brain - Jason M. Satterfield97% (71)Cognitive Therapy Therapy Techniques For Retraining Your Brain - Jason M. Satterfield222 pages
- Circumcision Photo Guide: What To Expect After Surgery: Patient and Family Education17% (12)Circumcision Photo Guide: What To Expect After Surgery: Patient and Family Education2 pages
- Calculus Concepts and Applications Foerster100% (7)Calculus Concepts and Applications Foerster731 pages
- W. Michael Kelley - The Humongous Book of Trigonometry Problems-ALPHA (2012) PDF100% (5)W. Michael Kelley - The Humongous Book of Trigonometry Problems-ALPHA (2012) PDF455 pages
- Project Looking Glass: QAnon Post 3094 Ir0nbelly93% (28)Project Looking Glass: QAnon Post 3094 Ir0nbelly89 pages
- Calculus - Concepts and Applications - Foerster83% (58)Calculus - Concepts and Applications - Foerster731 pages
- Advanced Functions and Introductory Calculus100% (18)Advanced Functions and Introductory Calculus498 pages
- Introduction To Linear Algebra For Science and Engineering 1st Ed89% (57)Introduction To Linear Algebra For Science and Engineering 1st Ed550 pages
- Mehdi Rahmani-Andebili - Precalculus - Practice Problems, Methods, and Solutions (2021)100% (1)Mehdi Rahmani-Andebili - Precalculus - Practice Problems, Methods, and Solutions (2021)119 pages
- (Industrial and Applied Mathematics) Martin Brokate, Pammy Manchanda, Abul Hasan Siddiqi - Calculus For Scientists and Engineers (2019, Springer)100% (11)(Industrial and Applied Mathematics) Martin Brokate, Pammy Manchanda, Abul Hasan Siddiqi - Calculus For Scientists and Engineers (2019, Springer)655 pages
- Calculus Made Easy: Being a Very-Simplest Introduction to Those Beautiful Methods of Reckoning Which are Generally Called by the Terrifying Names of the Differential Calculus and the Integral CalculusFrom EverandCalculus Made Easy: Being a Very-Simplest Introduction to Those Beautiful Methods of Reckoning Which are Generally Called by the Terrifying Names of the Differential Calculus and the Integral Calculus4.5/5 (2)
- Conceptual Calculus: A Grand Review Guide for the Ap® Calculus ExamFrom EverandConceptual Calculus: A Grand Review Guide for the Ap® Calculus ExamNo ratings yet
- Calculus graphical numerical algebraic Fourth Edition. Edition Franklin D. Demana - Download the ebook today and experience the full contentNo ratings yetCalculus graphical numerical algebraic Fourth Edition. Edition Franklin D. Demana - Download the ebook today and experience the full content55 pages
- What Parents and Teachers Should Know About Adhd: Center For Children and FamiliesNo ratings yetWhat Parents and Teachers Should Know About Adhd: Center For Children and Families2 pages
- Private Tutoring Service - Jakarta in Home Tutoring PDFNo ratings yetPrivate Tutoring Service - Jakarta in Home Tutoring PDF8 pages
- Project Report: Functions of Human Resource ManagementNo ratings yetProject Report: Functions of Human Resource Management15 pages
- Study Skills of Normal-Achieving and Academically-Struggling College StudentsNo ratings yetStudy Skills of Normal-Achieving and Academically-Struggling College Students16 pages
- Assignments - Interpret and Plan Assignments How To - Interpret An Assignment QuestionNo ratings yetAssignments - Interpret and Plan Assignments How To - Interpret An Assignment Question2 pages
- MASTER TEACHERS GREEN TEMPLATE Results Based Performance Management SystemNo ratings yetMASTER TEACHERS GREEN TEMPLATE Results Based Performance Management System44 pages
- Computer Training Curriculum For Older Adults FINAL-esNo ratings yetComputer Training Curriculum For Older Adults FINAL-es274 pages
- Peer Tutoring - Exploring The Effects On Learning Grade 9 MathematicsNo ratings yetPeer Tutoring - Exploring The Effects On Learning Grade 9 Mathematics14 pages
- Teaching Number Concepts: Staff DevelopmentNo ratings yetTeaching Number Concepts: Staff Development4 pages
