Lecture Notes

PHYSICS 101

S. Murugesh

IIST-Trivandrum

Last update: August 12, 2013

Contents

1 Preliminaries: Vectors and Kinematics 2

1.1 Vector Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.1 Completeness and Linear Independence . . . . . . . . . . . . . . . . . . . . . 3
1.1.2 The Inner/Scalar/Dot Product . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.1.3 The Vector/Cross Product . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2 Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.1 Position, Velocity and Acceleration . . . . . . . . . . . . . . . . . . . . . . . 8
1.2.2 Circular Motion: Polar Coordinates . . . . . . . . . . . . . . . . . . . . . . . 8

2 Newtonian Mechanics 10
2.1 Laws of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Linear Momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Chapter 1

Preliminaries: Vectors and Kinematics

We shall start with a brief review of some basic math ideas we will frequently invoke, without dis-
cussing them in detail (See Topics in Algebra by I. N. Herstein for a complete rehash). Particularly,

1.1 Vector Spaces

For the time being it suffices to know that a vector space is a Set of elements, say S, with a property
called addition, denoted +, defined on the set S. We say it is called addition, because we are free
to define the addition, but as long as it is.

commutative : ~s1 + ~s2 = ~s2 + ~s1 (1.1)

associative : ~s1 + (~s2 + ~s3 ) = (~s1 + ~s2 ) + ~s3 (1.2)
linear : a(~s1 + ~s2 ) = a~s1 + a~s2 , (1.3)

where a is a scalar We shall denote such a vector space by (S, +), or simply S. The term defined
on implies that
+ : S × S → S. (1.4)
i .e., ∀s1 , s2 ∈ S ⇒ s1 + s2 ∈ S. (1.5)
Further, the set must compulsarily have a element called zero (denoted 0), such that

∀s1 ∈ S ⇒ s1 + 0 = s1 . (1.6)

Elements of a vector space are then called vectors1 .

Examples:

1. Let S be R1 , the real line. Evidently, S is a vector space with + being the usual arithmetic
addition.

2. The set of all positive and negative integers, including 0, is a vector space under usual addition.

3. The set of all n’th order real polynomials forms a vector space under addition.
1
All along these notes vectors will be denoted by bold faces
 4. The set of all possible complex periodic piece-wise smooth functions (with period, say, L)
forms vector space.

5. The set S of all N × M real matrices forms a vector space, where + is the usual matrix
addition. The most common being the Euclidean n-dimensional space, which is the set of all
column (or row) matrices of size n.

6. The set Q of all rational numbers forms a vector space under addition.

1.1.1 Completeness and Linear Independence

Understood that the operation addition just maps two elements (vectors) in S to a third element
in S, a subset C ⊂ S, is said to be complete in S if any vector in S can be written as a linear
combination of vectors in C. I.e., a subset C = {ci }, i = 1, 2, ....n, is said to be complete if

∀s ∈ S ∃ {ai }, i = 1, 2, ...n : ai ci = s. (1.7)

ai are referred to as scalars and belong in some field2,3 . Such a subset C is said to span the vector
space S. Note that vectors are denoted by bold faces, while scalars are not (Also implied in Eq.
(1.7) is one of the axioms of vector spaces - multiplying a scalar by a vector gives another vector!).
Examples:

1. The set of functions defined as C = {x0 , x1 , x2 , ....xn } forms a complete set for the vector
space S of all possible polynomials upto order n.

2. The infinite set of functions defined by C = {exp(inπx/L)}, n = 0, 1, 2, ...∞, is a complete set

in the vector space of all complex periodic piece-wise smooth functions of period L.

3. The vector space S of all 2 × 2 real matrices is spanned by the subset

       
1 0 0 1 0 0 0 0
C={ , , , }.
0 0 0 0 1 0 0 1

C is thus complete in S.

For a given vector space S, if n is the number of elements in a complete set C in S, then the minimum
value n can take, i.e., the number of elements in the smallest complete set, is called the dimension
of the vector space S. In fact, if C in example 1 above spans the vector space of all polynomials of
order n, then so does C ′ = {x0 , x1 , x2 , ...xn , x1 + x2 }. We say the set C ′ is over-complete. Further,
notice that the complete set need not be unique. For, again in example 1 above, the vector space
of polynomials can also be spanned by the complete set C ′′ = {x0 + x1 , x0 − x1 , x2 , ...xn }. The
complete subsets C and C ′ give two different basis for the vector space S. Choosing one over the
other amounts to merely a change of basis vectors.
2
The nature of {ai } tells the nature of the vector space. Thus, if {ai } can only be real, then the vector space is
a real vector space. Further discussion will take us into the subject of Fields, which we shall deliberately avoid. For
a further reading, see [1]
3
We have adopted (here, and henceforth) the Einstein summation P convention, wherein summation is assumed
when repeated indices occur, unless stated otherwise. Thus, ai ci ≡ i ai ci .
 A set of vectors {si }, i = 1, 2, ..m is said to be linearly independent if

ai si = 0 ⇒ ai = 0 f or all i, (1.8)

where ai are all scalars.

The following statements must be apparent:

1. The maximum number of linearly independent vectors possible in a n-dimensional vector space
is n.

2. Any set of n linearly independent vectors in a n-dimensional vector space forms a complete
set in that vector space, and thus forms a basis. Conversely,

3. If a set of n vectors is complete, in a n-dimensional space, then the vectors are linearly
independent.

1.1.2 The Inner/Scalar/Dot Product

Just as the ‘addition’ used in the previous section was a map of any two given elements of the set
to a third element of the set S, the inner product, indicated by a ’·’, is a map of any two given
elements of the set to a element of the underlying field F (remember the elements of the field are
scalars and can be real or complex depending on the nature of the vector space).

· : S × S → F. (1.9)
s1 , s2 ∈ S ⇒ ·(s1 , s2 ) ≡ s1 · s2 ∈ F. (1.10)
A vector space endowed with a valid inner product is called a inner product space. Euclidean inner
product, specifically called the dot product, is the most common. Defined in Rn , this is:

∀u = {u1 , u2 , ...un }, v = {v1 , v2 , ...vn } ∈ Rn ⇒ u · v = u1 v1 + u2 v2 + ...un vn ∈ F. (1.11)

No matter what the specific definition of the scalar product is, it should obey certain tenets:

1. the · product treats both vectors alike, i.e.,

u · v = v · u. (1.12)

2. for any two scalars a and b , and vectors u, v and w,

(au + bv) · w = au · w + bv · w. (1.13)

3. the · product is positive definite

u·u≥0 (1.14)
u·u=0 ⇐⇒ u = 0, (1.15)

Notice that the vector 0 is different from the number ‘0’.

 4. the norm, or length, of a vector is defined as
√
||u|| = + u · u. (1.16)

5. The inner product of any vector with the vector 0 is always ‘0’
0 · u = 0, ∀u.

Two vectors are said to be orthogonal iff their · product vanishes. A vector u is normal if u·u = 1.
A set of vectors ui , i = 1, 2, ..m are orthonormal if
ui · uj = δij , (1.17)
for all i and j 4 . The following statements should be obvious:
1. Orthogonal vectors are also linearly independent. The converse, however, need not be true.
2. In a n-dimensional vector space the maximal number of orthonormal vectors possible is n.
3. Given any n linearly independent vectors in n-dimensional space, a suitable · product can
always be defined such that the n given vectors are orthonormal.
Let {ai }, i = 1, 2, .., n form a basis for a n-dimensional vector space S. Then any vector u in
S can be written as a unique linear combination of the basis vectors: u = ui ai . In particular,
4. if the basis set {ai } is orthonormal (ai · aj = δij for all i, j), then ui = u · ai , i = 1, 2, .., n are
the components of u in that basis.
Example: The Euclidean 3-Space/R3
The best way to construct the usual 3-D vector space (or any vector space for that matter) would
be to first start by defining a complete set of orthonormal basis vectors - in this case {î, ĵ, k̂}. By
calling them orthonormal, we have already assumed an inner product, namely î · ĵ = ĵ · k̂ = k̂ · î = 0
and î · î = ĵ · ĵ = k̂ · k̂ = 1. The Euclidean 3-space is then just the set of all possible linear
combinations of the three basis vectors of the type (say, v =) aî + √ bĵ + ck̂ where a, b, c ∈ R. With
this definition of orthonormality, the length of the vector v = v = a2 + b2 + c2 . The dot product
of two vectors v1 =) a1 î + b1 ĵ + c1 k̂, v2 =) a2 î + b2 ĵ + c2 k̂ is u · v = a1 a2 + b1 b2 + c1 c2 .

Exercise Convince yourself that this definition of the inner product is the same as the usual
definition: u · v = uv cos θ.

Exercise Construct, or define, Euclidean 3-space by any other means than that suggested above.

Exercise Suggest an alternate valid inner product in 3-D, than the regular Euclidean · product.

Notice that if the components of vectors u, v in Eq. (1.11) are represented as as n × 1 column
matrices u and v, their dot product can also be written as
u · v = uT v, (1.18)
4
The Kronecker delta, δij , here is just the the identity matrix. Thus, its ij’th element δij = 1 if i = j, and 0
otherwise.
where uT is the transpose matrix of u of order n × 1, and the · product becomes just the usual
matrix multiplication. Some crucial differences have to remembered:

1. u and uT do not belong to the same vector space.

2. One has to be careful that the transposed matrix goes in the left, for

uT v 6= vuT . (1.19)

1.1.3 The Vector/Cross Product

Vector spaces do not have a inner product inherent in their construction - all they need are a set of
elements and a valid addition. The inner product structure is usually added on, and, given a vector
space, it may be possible to define more than one valid inner product. A cross/vector product is
another such structure that may be possibly defined on a vector space.

× : S × S → S. (1.20)

s1 , s2 ∈ S ⇒ ×(s1 , s2 ) ≡ s1 × s2 ∈ S, (1.21)
with the following properties.

1. Antisymmetry:
s1 × s2 = −s2 × s1 . (1.22)

2. Cross product of parallel vectors vanish:

s1 × s2 = 0 ⇒ either, s1 = 0, s2 = 0, ors1 = as2 (1.23)

where a is some number.

3. Distributive:
s1 × (as2 + bs3 ) = as1 × s2 + bs1 × s3 . (1.24)

4. Jacobi identity:
s1 × (s2 × s3 ) + s2 × (s3 × s1 ) + s3 × (s1 × s2 ) = 0. (1.25)

The Euclidean vector product is defined using the basis vectors as

î × ĵ = k̂, ĵ × k̂ = î, (1.26)

Using the relations in Eq. (1.28) in the Jacobi identity, Eq. (1.25), we deduce that k̂ × î is
parallel to ĵ (verify!!) or
k̂ × î = aĵ (1.27)
where a is some scalar. The scalar a is conventionally taken to be 1, giving us the complete relations

î × ĵ = k̂, ĵ × k̂ = î, k̂ × î = ĵ. (1.28)

The vector product is then extensible to the entire vector space using linearity, Eq. (1.24).
 Problem 1.1: Let ai , i = 1, 2, 3 be three arbitrary non-coplanar vectors. Find vectors bi , i =
1, 2, 3 such that bi · aj = δij . Sketch three vectors ai from the origin, and then sketch the corre-
sponding bi vectors, also from the same origin.

Problem 1.2: If a and b are two Euclidean vectors, specify their properties in each of the
following cases:
a)a + b = c and a + b = c.
b)a + b = a − b.
c)|a + b| = |a − b|.
d)a + b = c and a2 + b2 = c2 .
e)|a + b| = |a| = |b|.

The Levi-Civita tensor, product, and some notation

We introduced the Kronecker delta in Eq. (1.17). Another similar useful object is the Levi-Civita
tensor, whose elements are denoted as ǫijk , where each of the three indices i, j, k take any of the
values 1, 2, 3, making it 27 in total. The tensor is antisymmetric with respect to interchange of any
two indices, and ǫ123 = 1. Thus, ǫ123 = ǫ231 = ǫ312 = 1, ǫ213 = ǫ132 = ǫ312 = −1, and all other
elements vanish.
The Levi-Civita tensor comes in handy in writing the vector product. The component form of
the equation A = B × C can be written as Ai = ǫijk Bj Ck (Remember the summation convention.
There is summation assumed over repeated indices j and k.).

Exercise Verify the product relation for Levi-Civita tensor:

ǫijk ǫilm = δjl δkm − δjm δkl . (1.29)

Note: The index i is repeated, and hence summed over, but not j and k.

Problem 1.3: Show that A · (B × C) = ǫijk Ai Bj Ck = (A × B) · C.

Problem 1.4: Use the Levi-Civita matrix to a) show that A×(B × C) 6= (A × B)×C), and
that b) A×(B × C) = B · (A × C) − C · (A × B).

Problem 1.5: Show that A · B × C is just the area of the parallelepiped formed by A, B and
C as edges.

1.2 Kinematics
Mechanics deals with motion - causes of motion, Dynamics, and its geometrical description, Kine-
matics5 . A vectorial representation is indispensable in describing the geometry of motion of an
object, or collection of objects. For all our purposes in this course we treat the 3-dimensional space
we live in to be real and Euclidean (although it should be cautioned that our present state of knowl-
edge tells us that this is only a reasonable approximation). Which implies that I can represent the
5
Of course there is also the case of Statics - condition of rest, about which we shall not make much fuzz here.
position vector r of an object by three spacial co-ordinates {x, y, z}, which are just the distances
I need to travel along three linearly independent and mutually orthogonal directions to reach the
objects from where I am. How I choose these directions is entirely upto me, but I just need to pick
the three directions in a way that they are perpendicular to each other. Further, in dealing with
a set of such vectors, the Euclidean rules for scalar and vector products applies. The unit vectors
along the three directions shall be denoted by x̂, ŷ and ẑ.

1.2.1 Position, Velocity and Acceleration

A fundamental assumption in Newtonian mechanics is that space and time are independent and
absolute entities. By which we mean that change in time is always uniform, or to put it more
practically, time intervals are same for all observers, no matter when or where. A similar assumption
holds for space too, that the distance between any two points in space is same for all observers,
and at all times6 . In describing the motion of an object, we take position vector to vary with time,
r(t) = {x(t), y(t), z(t)}. The rate of chage of the position vector gives its velocity
dr dx dy dz
v= ≡ ṙ = { , , }. (1.30)
dt dt dt dt
Notice that velocity is a vector, although not in the same vector space as position. The rate of
change of velocity gives its accelaration,
dv
a= ≡ v̇. (1.31)
dt

1.2.2 Circular Motion: Polar Coordinates

Velocity and acceleration tell us about a vectorial change in time, that of the position and velocity
vectors, respectively. A vector can change either on account of a change in its magnitude or direction.
In our discussion on Euclidean space so far we have chosen the Cartesian basis, x̂, ŷ and ẑ that
neither change in space or in time. However, it is not the useful basis for all circumstances. Here,
we shall discuss one simpler case of circular motion in a plane (say the x − y plane). Suppose the
positon vector of the object at some point in time is given by r = {x, y}. It is convenient to use a
polar coordinate system defined by
x = r cos θ, y = r sin θ, (1.32)
and hence
r = r cos θx̂ + r sin θθ̂. (1.33)
p
Geometrically, we realize that r is just the radial distance x2 + y 2 and θ is the angle the position
vector r makes with the x axis. Analogous to the unit vectors x̂ and ŷ we intend to define the unit
vectors r̂ and θ̂. The unit vector x̂ may be defined as the change in position vector r with respect
to x, if y is held constant (and vice versa for ŷ):
dr dr
x̂ = , ŷ = . (1.34)
dx y dy x
6
Another thought we now know is just an approximation. The impatient reader may refer an introductory book
on relativity such as [2]
 dr 1
r̂ = =p (xx̂ + yŷ) = cos θx̂ + sin θŷ, (1.35)
dr θ x2 + y 2
1 dr 1
θ̂ = =p (−yx̂ + xŷ) = − sin θx̂ + cos θŷ. (1.36)
|dr/dθ| dθ r x2 + y 2
Notice that the directions of r̂ and θ̂ vary with position, but are orothogonal to each other ev-
erywhere. In Eq. (1.36), θ̂ is made normal through division by the modulus |dr/dθ|. Before we
proceed further to write the velocity and acceleration in r̂ and θ basis, it is useful to first find the
time derivatives of the two basis vectors. By direct differentiation of Eqs. (1.35) and (1.36), we find

dr̂ dθ̂
= θ̇θ̂, = −θ̇r̂. (1.37)
dt dt

That the change in θ~ is along the r̂ direction, and vice-versa should not be surprising. On the plane,
given that r̂ · θ̂ = 0, it immediately follows that

dr̂ dθ̂
· θ̂ = −r̂ · . (1.38)
dt dt
We can now find the velocity and acceleration in polar coordiates:
dr
v= = vr r̂ + vθ θ̂ = ṙr̂ + rθ̇θ̂, (1.39)
dt

a = ar r̂ + aθ θ̂ = v̇ = (r̈ − rθ̇2 )r̂ + (rθ̈ + 2ṙθ̇)θ̂. (1.40)

Caution must be exercised while integrating acceleration in polar coordinates to obtain the velocity
vector, for, unlike the Cartesian case, ar 6= v̇r , and aθ 6= v̇θ - the catch being that the unit vectors
themselves change in time on account of their dependence on (x, y). The component of acceleration
along the radial direction has two contributions - radial (r̈) and centripetal (−rθ̇2 ) accelerations,
while along the angular direction it composes of angular (rθ̈) and coriolis (ṙθ̇) accelerations. The
negative sign in the centripetal acceleration indicates that its direction is inward towards the center.
Chapter 2

Newtonian Mechanics

2.1 Laws of Motion

We introduced velocity and acceleration, obtained as first and second time derivatives of the position
vector. One can go on with this exercise and find further derivatives, but they would be just a matter
of detail. The importance of acceleration is embodied in the
First law of Motion: Every body continues in its state of rest, or uniform, motion unless or
untill it is acted upon by a net force.
Thus acceleration is an indication, and a measure, of the net force acting on a body. More strictly,
the acceleration of a body is determined by the net force acting on the body, and vice-versa - a
concept referred to as determinism. By detailing acceleration and force, we have in fact jumped
the gun - not having mentioned anything about the state of the observer. Would any two observers
seeing the same moving object always note the same acceleration? Say, Calvin on a train and
Hobbes on the platform! They certainly would, if the train were at halt, or even if it were moving
at constant speed (at least for the typical speeds trains are known to reach!). This is easy to show
if we relate the coordinates of the two observers. Let r = {x, y, z} be the coordinates of the moving
object as seen by Hobbes from the platform, and r′ = {x′ , y ′ , z ′ } be that noted by Calvin. We
shall take the train to be moving with a constant velocity v along the x direction (i. e., v = vx̂).
Then, the relation between the coordinates of the moving object assigned by the two is given by
the Galilean transformation relations:
x′ (t) = x(t) − vt, y ′ (t) = y(t), z ′ (t) = z(t). (2.1)
Remember, v is a constant - the train is not accelarating. Also we have taken the times noted by
either observer to be the same t, on account of our assumption that time is absolute.
Now, let us find the velocity and acceleration of the object as seen by Calvin and Hobbes. To
Hobbes these would be
v = ṙ = {ẋ, ẏ, ż} (2.2)
a = v̇ = {ẍ, ÿ, z̈}. (2.3)
Whereas, for Calvin these would be
v′ = ṙ′ = {ẋ′ , ẏ ′ , z˙′ } = {ẋ − v, ẏ, ż} = v − vx̂ (2.4)
a′ = v̇′ = {ẍ′ , ÿ ′ , z¨′ } = {ẍ, ÿ, z̈} = a, (2.5)
since v is a constant. So, even though the velocities of the object as observed by Calvin and Hobbes
are different, they do agree on the accelaration, on account of Hobbes being at rest and Calvin in
uniform relative motion with respect to Hobbes. Lesson: Measurement of acceleration better be
done from a reference frame at rest, or from any other reference frame in uniform relative motion
with respect to one at rest - what we shall call as inertial reference frames. Experimentally, the
inertial mass is independent of the velocity or accelration of the object, atleast for the kind of
velocities we shall deal with here. So our lesson learnt extends to forces too. Forces should be
measured from inertial reference frames. Unless otherwise stated, any reference frame we shall use
henceforth will be assumed inertial.

Exercise: You are stuck inside a sound proof lab without windows, say, inside a train. How
will you know it is accelerating? Can you figure out its velocity, if it is in a uniform speed? (This
is easy!)

The importance of inertial frames - the gist of the first law, lies in the answer to this simple exercise
above. The state of rest, or uniform motion, is purely relative, whereas acceleration isn’t.

Exercise A critical assumption in the Calvin and Hobbes observation was that the platform
itself is at rest. What if it isn’t? Well, in reality it is not. The Earth, afterall, is moving around in
the Solar system, accelerating, and so does the platform. How do we account for Earth’s acceleration
in their observation?
Armed with the first law we can perform a few experiments to arrive at a better understanding
of Force. For instance, we may tie a rope to a cubic block of wood on ice, with some heavy object
hung from the other end of the rope (see figure), run over a pulley. With negligible friction between
the wooden block and ice, its acceleration can be measured with reasonable accuracy. If another
identical block of wood, showing the same acceleration, is attached the the first block, we may
notice that the acceleration is now reduced to half. We conclude that the force depends directly
on the acceleration, and the inherent tendency to resist the force is called the inertial mass of the
object - more the inertial mass, less the acceleration for a given force. We may also check that
forces behave as vectors, and obey the vector addition rules, by tying ropes to pull the block in
different directions and measuring resultant acceleration. Thus we state the
Second Law of Motion: F = ma,
a quantitative expression for the net force F acting on a body of inertial mass m, and moving with
an acceleration a.
The third law of motion: Every action has an equal and opposite reaction,
is an assertion that forces come in pairs - if A exerts a force on B, then so should B on A, which is
equal and opposite in direction:
A consequence of the third law is that the net force in a closed system must be zero. The
third law is also observed in two forms, the strong form - when, along with being equal and
opposite, the force acts along the line joining A − B. Such as, when two objects collide, two charges
attracting/repelling, or the gravitational attraction between two massive objects. There are cases
when the force between A and B, although equal and opposite, doesn’t act along the line A − B
(for instance, a set of charges in motion - the magnetic Lorentz force). Such forces are said to obey
the weak form of the action and reaction law. The two forms of the third law have their own
consequences, which will be discussed in the subsequent sections.
 To begin with applications of Newton’s laws to mechanical systems, we shall for simplicity
consider here systems that consist of rigid bodies that do not show rotational motion. This allows
us to treat every massive object as a point particle. Given the initial condition(s) for such a system,
and all the forces acting on it, our endevour is to describe its future. To obtain the equations
governing the motion of a given mechanical system,

1. we first dissect the system into subsystems - typically every massive object in the system is
considered a ssubsystem,

2. identify an appropriate coordinate frame for the entire system, and assign coordinates for each
of the subsystems,

3. list out all the forces acting on each of the sybsystems,

4. apply the condition that the forces between any two subsystems is equal in magnitude and
opposite in direction,

5. the net force on any subsystem must be equal to its mass times acceleration.

An illustrative sketch of the above five points will give us a free body diagram of the system. The
forces between these subsystems could be i) contact forces - such as the impact of a collision, or
friction between surfaces of two bodies, ii) forces acted upon through a spring, or an inextensible
rope or iii) action at a distance forces such as gravitational attraction.

2.2 Linear Momentum

If the position vector of a particle of mass m is given by r, then the second law of motion translates
to
F = mr̈, (2.6)
where F is the net force on the particle.
Bibliography

[1] I. N. Herstein. Topics in Algebra, 2nd Edition. Wiley, 1975.

[2] B. Schumacher. Introduction to Special Relativity. Overseas Press, New Delhi, 2007.