MA 105 : Calculus

Division 1, Lecture 12

Prof. Sudhir R. Ghorpade

IIT Bombay

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Recap of the previous lecture

Curves given by polar equations. Area of related planar

regions
Volume of a solid. Slice Method, Examples
Solids of Revolution
Washer Method
Shell Method
Examples
Parametrized curves

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Parametrized Curve

A parametrized curve or a path C in R2 is given by

(x(t), y (t)), where x, y : [α, β] → R are continuous functions.
Here [α, β] is called the parameter interval.
We wish to define the ‘length’ of C .
Basic assumption: The (Euclidean) length of a line segment
joining points (x1 , y1 ) and (x2 , y2 ) in R2 is equal to
p
(x2 − x1 )2 + (y2 − y1 )2 .

We shall assume that C is smooth, that is, the functions x, y

are continuously differentiable on [α, β]. This means that
x, y are differentiable on [α, β], and their derivatives x 0 , y 0 are
continuous on [α, β].

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Arc Length of a Smooth Curve
Partition [α, β] into α
= t0 < t1 < · · · < tn = β.
Let Pi := x(ti ), y (ti ) for i = 1, . . . , n, and draw the line
segments joining P0 to P1 , P1 to P2 , . . ., Pn−1 to Pn .
The sum of the lengths of these line segments is
X n q
(x(ti ) − x(ti−1 ))2 + (y (ti ) − y (ti−1 ))2
i=1
Xn q
= (x 0 (si ))2 + (y 0 (ui ))2 (ti − ti−1 ) ,
i=1

for some si , ui ∈ (ti−1 , ti ) for i = 1, . . . , n by the MVT.

We define the arc length of C by
Z βp
` (C ) := x 0 (t)2 + y 0 (t)2 dt.
α

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Special Cases
Special cases:
(i) Let a curve C be given by y = f (x), x ∈ [a, b].
Here α := a, β := b, x(t) := t and y (t) := f (t) for t ∈ [a, b].
Suppose f is continuously differentiable on [a, b]. Then
Z b p
` (C ) := 1 + f 0 (x)2 dx.
a

(ii) Let a curve C be given by x = g (y ), y ∈ [c, d].

Here α := c, β := d, x(t) := g (t) and y (t) := t for t ∈ [c, d].
Suppose g is continuously differentiable on [c, d]. Then
Z dp
` (C ) := g 0 (y )2 + 1 dy .
c

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Arc Length in Polar coordinates
Let C be given by a polar equation r = p(θ), θ ∈ [α, β]. As a
parametrized curve, C is given by (x(θ), y (θ)), where

x(θ) := p(θ) cos θ and y (θ) := p(θ) sin θ, θ ∈ [α, β].

Suppose the function p is continuously differentiable on [α, β].

p
For θ ∈ [α, β], we note that x 0 (θ)2 + y 0 (θ)2 is equal to
q
(p 0 (θ) cos θ − p(θ) sin θ)2 + (p 0 (θ) sin θ + p(θ) cos θ)2
p
= p(θ)2 + p 0 (θ)2 .

Hence Z β p
` (C ) := p(θ)2 + p 0 (θ)2 dθ.
α

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Examples
(i) Let C be given by y = x 2 , x ∈ [0, 1]. Then

1 2√
Z 1p Z
`(C ) = 2
1 + (2x) dx = 1 + u 2 du
0 2 0
1√ 1 √

= 5 + ln 2 + 5 .
2 4
√
(Use Integration by Parts. Also, if√f (u) := ln(u + 1 + u 2 ) for
u ∈ R, then note that f 0 (u) = 1/ 1 + u 2 for u ∈ R, and so
Rx √ √ √
1 + u 2 du = 12 x 1 + x 2 + ln x + 1 + x 2 for x ∈ R.)



0

(ii) Let C be given by x = (2y 6 + 1)/8y 2 , y ∈ [1, 2]. Then

Z 2  1/2 Z 2

3 1 2 1  123
1+ y − 3 dy = y 3 + 3 dy = .
1 4y 1 4y 32

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

(iii) Let a > 0 and ϕ ∈ [0, π]. Let C denote the arc of a circle
of radius a given by x(θ) := a cos θ, y (θ) := a sin θ for
θ ∈ [0, ϕ]. Then C is given by the polar equation r = p(θ),
where p(θ) = a for θ ∈ [0, φ], and so
Z ϕ√
` (C ) = a2 + 02 dθ = a ϕ.
0
Rπ
Hence the length of a circle of radius a is −π
a dθ = 2πa.
(iv) Let C be given by r = 1 + cos θ for θ ∈ [0, π]. Then
Z πp
` (C ) = (1 + cos θ)2 + (− sin θ)2 dθ
0
Z πp Z π
θ
= 2(1 + cos θ) dθ = 2 cos dθ = 4.
0 0 2

(Note: cos(θ/2) ≥ 0 for θ ∈ [0, π].)

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
Curves in R3
Suppose C is a smooth parametrized curve in R3 given by
(x(t), y (t), z(t)), where x, y , z : [α, β] → R are continuously
differentiable functions on [α, β].
In analogy with the definition of the arc length of a curve in
R2 , we define the arc length of C by
Z βp
` (C ) := x 0 (t)2 + y 0 (t)2 + z 0 (t)2 dt.
α

Example
Let C denote a helix in R3 given by
x(t) := a cos t, y (t) := a sin t, z(t) := b t, t ∈ [α, β], where
a, b, ∈ R, a > 0 and b 6= 0. Then
Z βp √
` (C ) = (−a sin t)2 + (a cos t)2 + b 2 dt = (β−α) a2 + b 2 .
α

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Surface of Revolution
A surface of revolution is generated when a curve C in R2 is
revolved about a line L in R2 .
First suppose the curve C is a slanted line segment P1 P2 of
length λ2 , and C does not cross L. Let d1 and d2 denote the
distances of P1 and P2 from L with d1 ≤ d2 . Then the surface
of revolution is a frustum F of a right circular cone with base
radii d1 and d2 , and slant height λ2 . We find its surface area.
P2
b
λ2
P1 d2
b

d1 L
ℓ1

ℓ2

Figure: Frustum of a right circular cone

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Consider a cone with base radius d and slant height `. If we
slit open this cone, we obtain a sector of a disk of radius `.

ℓ
ℓ
ϕ ℓϕ = 2πd

b
d

Figure: Right circular cone and sector of a disk

Since `ϕ = 2πd, the surface area of the cone is equal to

1 2 1 2πd
` ϕ = `2 = π`d.
2 2 `
Hence the surface area of the frustrum F of the cone is
π`2 d2 − π`1 d1 = π(d1 + d2 )(`2 − `1 ) = π(d1 + d2 )λ2 .
Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
Now suppose C is parametrized by (x(t), y (t)), t ∈ [α, β].
Partition [α, β] into α
= t0 < t1 < · · · < tn = β.
Let Pi := x(ti ), y (ti ) for i = 0, 1, . . . , n, and draw the
line segments P0 P1 , P1 P2 , . . ., Pn−1 Pn .
Let d0 , d1 , d2 , . . . , dn be the distances of P0 , P1 , P2 , . . . , Pn
from the line L. Let λ1 , λ2 , . . . , λn be the lengths of the line
segments P0 P1 , P1 P2 , . . . , Pn−1 Pn . Suppose they don’t cross L.
b
Pn
b
Pi−1 Pib
P0
b
P1 L
b

b b

b
b

Fix i ∈ {1, . . . , n}. When the line segment Pi−1 Pi is revolved

about the line L, it generates a frustum Fi (of a right circular
cone) whose surface area is π(di−1 + di )λi .
Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
Let ρ(t) denote the distance of the point (x(t), y (t)) on the
curve C from the line L. Then di = ρ(ti ) for i = 0, 1, . . . , n.
Thus the sum of the surface areas of the frustrums F1 , . . . ,Fn is
n
X

π ρ(ti−1 ) + ρ(ti ) λi ,
i=1

If the functions x 0 and y 0 are continuously differentiable on

[α, β], then the length λi of the line segment Pi−1 Pi is given by
q
λi = (x(ti ) − x(ti−1 ))2 + (y (ti ) − y (ti−1 ))2
p
= x 0 (si )2 + y 0 (ui )2 (ti − ti−1 )

for some si , ui ∈ (ti−1 , ti ) for i = 1, . . . , n (by the MVT).

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Area of Surface of Revolution
Let C be a smooth curve parametrized by
(x(t), y (t)), t ∈ [α, β]. Suppose the curve C does not cross
the line L given by ax + by + c = 0. We define the area of
the surface S generated by revolving C about the line L by
Z β p
Area (S) := 2π ρ(t) x 0 (t)2 + y 0 (t)2 dt,
α

where ρ(t) is the distance of (x(t), y (t)) from the line L,

√
that is, ρ(t) := |ax(t) + by (t) + c|/ a2 + b 2 for t ∈ [a, b].
Note: Since the curve C does not cross the line L, the curve C
lies entirely on one of the sides of the line L, that is,
either ax(t) + by (t) + c ≥ 0 for all t ∈ [α, β],
or ax(t) + by (t) + c ≤ 0 for all t ∈ [α, β].

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

Special Cases:
(i) Let the line L be the x-axis, and let the curve C be given
by y = f (x) for x ∈ [a, b], where f is continuously
differentiable. If f ≥ 0 on [a, b] or f ≤ 0 on [a, b], then
Z b p
Area(S) = 2π |f (x)| 1 + f 0 (x)2 dx.
a

(ii) Let the line L be the y -axis, and let the curve C be given
by x = g (y ) for y ∈ [c, d], where g is continuously
differentiable. If g ≥ 0 on [c, d] or g ≤ 0 on [c, d], then
Z d p
Area(S) = 2π |g (y )| 1 + g 0 (y )2 dy .
c

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12

(iii) Let the line L be given by θ = γ, where γ ∈ (−π, π], and
let the curve C be given by r = p(θ) for θ ∈ [α, β], where
p is continuously differentiable on [α, β]. Suppose C does
not cross L. Now the curve C is also given by
(p(θ) cos θ, p(θ) sin θ) for θ ∈ [α, β].
Also, ρ(θ) = p(θ)| sin(θ − γ)| for θ ∈ [α, β].

(p(θ) cos(θ), p(θ) sin(θ))

b
C
θ)

p(θ)| sin(θ − γ)|

p(

L
θ γ
b

Figure: Distance of a point on a polar curve from a ray.

Z β p
Thus Area(S) = 2π p(θ)| sin(θ−γ)| p(θ)2 + p 0 (θ)2 dθ.
α
Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
Examples:
(i) Let S denote the surface generated by revolving the curve
y = (x 3 /3) + (1/4x), x ∈ [1, 3], about the line y = −1. Then
Z 3 p
Area(S) = 2π (y + 1) 1 + (y 0 )2 dx
1
3 s 2
x3
Z  
1 1
= 2π + +1 1 + x − 2 dx
2
1 3 4x 4x
3
x3
Z   
1 1
= 2π + +1 x 2 + 2 dx
1 3 4x 4x
= 1823π/18.

(iii) Let 0 < b < a and let C denote the circle given by
(a + b cos t, b sin t), t ∈ [−π, π]. Let S denote the surface
generated by revolving the curve C about the y -axis. Then
a + b cos t > 0 for all t ∈ [−π, π], and so
Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
 Z π p
Area (S) = 2π (a + b cos t) (−b sin t)2 + (b cos t)2 dt
−π
Z π
= 2πb (a + b cos t)dt
−π
= 4π 2 ab.

Note: S is in fact the surface of a torus in R3 .

(iii) Let a > 0, and S denote the surface generated by
revolving the semicircle p(θ) = a, θ ∈ [0, π], about the x-axis.
Then
Z π √
Area(S) = 2π a sin θ a2 + 02 dθ = 4πa2 .
0

Note: S is in fact the sphere of radius a in R3 .

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
Euclidean Spaces
Let m ∈ N, and consider the m dimensional Euclidean space

Rm := {(x1 , . . . , xm ) : xj ∈ R for j = 1, . . . , m}.

If m = 1, then an element of R1 := R is called a scalar and if

m > 1, then an element of Rm is called a vector.
We shall consider sequences in Rm , and functions from subsets
of Rm to R, and we shall study their properties.
We shall set up the basic structure in Rm , and then restrict
ourselves to m := 2, 3.
For x := (x1 , . . . , xm ), y := (y1 , . . . , ym ) ∈ Rm , and a ∈ R, let

x + y := (x1 + y1 , . . . , xm + ym ) ∈ Rm ,
a x := (a x1 , . . . , a xm ) ∈ Rm ,

called the sum of x and y, and the scalar multiple of a and x .

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
Norm of an element of Rm
Let x := (x1 , . . . , xm ) ∈ Rm . We define the norm of x by
kxx k := (x12 + · · · + xm2 )1/2 .
For m = 1, the norm of x ∈ R is its absolute value |x|. Note:
max{|x1 |, . . . , |xm |} ≤ kxx k ≤ |x1 | + · · · + |xm |.
Let 0 := (0, . . . , 0). Then kxx k = 0 ⇐⇒ x = 0 .
For x := (x1 , . . . , xm ) ∈ Rm and y := (y1 , . . . , ym ) ∈ Rm , let
x · y := x1 y1 + · · · + xm ym ∈ R,
called the dot product or the scalar product of x and y .
Theorem (Cauchy-Schwarz Inequality and Triangle Inequality)
For x , y ∈ Rm , |xx · y | ≤ kxx kkyy k and kxx + y k ≤ kxx k + kyy k.

Proof: If kxx k = 0 or kyy k = 0, then x = 0 or y = 0 , so done.

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
Suppose kxx k =
6 0 and kyy k =
6 0. Then

|xj | |yj | 1  |xj |2 |yj |2 

≤ + for j = 1, . . . , m,
kxx k kyy k 2 kxx k2 kyy k2

since ab ≤ (a2 + b 2 )/2 for all a, b ∈ R. Hence

m
X kxx kkyy k
|xx · y | ≤ |xj | |yj | ≤ (1 + 1) = kxx kkyy k.
j=1
2

for all x , y ∈ Rm . As a consequence,

m
X
2
kxx + y k = (xj + yj )2 = kxx k2 + kyy k2 + 2 x · y
j=1

≤ kxx k2 + kyy k2 + 2 kxx kkyy k = (kxx k + kyy k)2 .

Thus kxx + y k ≤ kxx k + kyy k for all x , y ∈ Rm .

Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12
Let x , y ∈ Rm . We say that x and y are orthogonal or
perpendicular (to each other) if
x · y = 0.

If x 6= 0 and y 6= 0 , then the angle between them is the unique

θ ∈ [0, π] such that cos θ = x · y /kxx k kyy k. (Cosine Rule)
Let m ∈ N and x 0 ∈ Rm . For x ∈ Rm , kxx − x 0 k is called the
distance of x from x 0 . For r > 0, the subset
B(xx 0 , r ) := {xx ∈ Rm : kxx − x 0 k < r }
of Rm is called the neighbourhood of x 0 of radius r .
If m := 1, then B(xx 0 , r ) is the interval (x0 − r , x0 + r ) in R.
If m := 2 and x 0 := (x0 , y0 ), then B(xx 0 , r ) is the disk
{(x, y ) ∈ R2 : (x − x0 )2 + (y − y0 )2 < r 2 } in R2 .
If m := 3 and x 0 := (x0 , y0 , z0 ), then B(xx 0 , r ) is the ball
{(x, y , z) ∈ R3 : (x − x0 )2 + (y − y0 )2 + (z − z0 )2 < r 2 } in R3 .
Prof. Sudhir R. Ghorpade, IIT Bombay MA 105 Calculus: Division 1, Lecture 12