PHYH-DSE-5: Classical Dynamics (Sem-5)

Sem-V (Honours), (July-Dec-2019)

Somenath Jalal
Department of Physics, Hooghly Women’s College

Last Updated on January 30, 2020

by somenath.jalal@gmail.com
2
 CONTENTS

1 Classical Mechanics of Point Particles 5

1.1 Review of Newtonian Mechanics . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Motion of a charge particle in external electric and magnetic fields . . . . . 5
1.3 Magnetic field-gyroradius and gyrofrequency . . . . . . . . . . . . . . . . . 6
1.4 Motion in crossed electric and magnetic fields. . . . . . . . . . . . . . . . . 6
1.5 Generalized coordinates and velocities . . . . . . . . . . . . . . . . . . . . . 6
1.6 Derivation of Euler-Lagrange Equation . . . . . . . . . . . . . . . . . . . . 6
1.7 Hamilton’s principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.8 Lagrangian and the Euler-Lagrange equations . . . . . . . . . . . . . . . . 10
1.9 1D Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.10 Hamilton’s equations of motion and its Applications: . . . . . . . . . . . . 11

2 Small Amplitude Oscillations 13

2.1 Potential Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Small Oscillation about Potential minima . . . . . . . . . . . . . . . . . . . 15

3 Special Theory of Relativity 23

3.1 Postulates of Special Theory of Relativity . . . . . . . . . . . . . . . . . . 23
3.2 Lorentz Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.3 Minkowski space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.4 The invariant interval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.5 Light cone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.6 World lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.7 Space-time diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.8 Time-dilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.9 Length contraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.10 Twin paradox . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.11 Four-vectors: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.12 Space-like, time-like and light-like. . . . . . . . . . . . . . . . . . . . . . . . 24
3.13 Four-velocity and acceleration. . . . . . . . . . . . . . . . . . . . . . . . . . 25

3
CONTENTS

3.14 Metric and alternating tensors. . . . . . . . . . . . . . . . . . . . . . . . . 25

3.15 Four-momentum and energy-momentum relation . . . . . . . . . . . . . . . 25
3.16 Doppler effect from a four-vector perspective . . . . . . . . . . . . . . . . . 26
3.17 Concept of four-force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.18 Conservation of four-momentum . . . . . . . . . . . . . . . . . . . . . . . . 26
3.19 Relativistic kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.20 Application to two-body decay of an unstable particle. . . . . . . . . . . . 26

4 Fluid Dynamics 29
4.1 Density ρ and pressure P in a fluid . . . . . . . . . . . . . . . . . . . . . . 29
4.2 Continuity equation and mass conservation . . . . . . . . . . . . . . . . . . 29
4.3 Stream-lined motion, laminar flow, Poiseuille’s equation for flow of a liquid
through a pipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.4 Navier-Stokes equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.5 Qualitative description of turbulence . . . . . . . . . . . . . . . . . . . . . 33
4.6 Reynolds number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

4
 CHAPTER 1
CLASSICAL MECHANICS OF POINT
PARTICLES

1.1 Review of Newtonian Mechanics

If total force is zero on a system of particles, then linear momentum is conserved.

p~˙ = 0 (1.1)

If torque ~τ is zero, then angular momentum of a particle is conserved.

~˙ = 0
L (1.2)

1.2 Motion of a charge particle in external electric

and magnetic fields
Let us consider a charged particle of mass m and charge q is entering a region with velocity
~ = E ĵ) and magnetic field
~v = ẋî + ẏ ĵ. In that region there are both electric field (E
~ = B k̂. The choice of fields are such that we derive all the details of motion of the
B
charge particle without loss of generality of choice of arbitrary fields.
The Lorentz force is given by

F~ = mẍî + mÿ ĵ = qE ĵ + q~v × B

~ (1.3)

** Incomplete
We shall derive the same equation of motion using Lagrangian for the problem.

5
Chapter 1. Classical Mechanics of Point Particles

1.3 Magnetic field-gyroradius and gyrofrequency

1.4 Motion in crossed electric and magnetic fields.
1.5 Generalized coordinates and velocities
What are Constraints of motion?
It is a condition that a body obeys during its motion. A constraint limits the motion of
a system.
Example: An ant walking on a football is constraint to move on the spherical surface
of the football.

Generalized co-ordinates
Minimum number of independent co-ordinates needed to specify the motion of a rigid body
are called generalized co-ordinates. These co-ordinates can be distance, angle, energy,
angular momentum, Fourier coefficients etc. The number of generalized co-ordinates are
the number of degrees of freedom.
Properties:
(i) Generalized co-ordinates decide the configuration of the system.
(ii) Generalized co-ordinates can be varied arbitrarily and independently of each other,
without violating the constraints on the system.
(iii) There is no uniqueness in the choice of generalized co-ordinates. A better choice
of Generalized co-ordinates will simplify the mathematical rigor of the problem.
In general
~ri = ~ri (q1 , q2 , · · · , q3N , t) (1.4)
where qi s are generalized co-ordinates and t is the time co-ordinate for the position
variable of the ith particle inside a system of N number of particles. And no. of degrees
of freedom are 3N .

Generalized velocity
Time derivative of generalized co-ordinate is named as generalized velocity.

1.6 Derivation of Euler-Lagrange Equation

A virtual or infinitesimal displacement of a system refers to change in the configuration of
the system as a result of any arbitrary small change of the co-ordinate δ~ri , consistent with
the forces and constraints imposed on the system at the given instant of time t. Suppose
the system is in equilibrium and thus total force on each particle vanishes. F~itot = 0. Thus
X
F~itot · δ~ri = 0 (1.5)
i

6
 1.6. Derivation of Euler-Lagrange Equation

We decompose F~itot into applied forces and forces of constraints. Thus

F~itot = F~i + f~i (1.6)

We restrict ourselves to systems for which the net virtual work of the forces of constraints
is zero. Thus
X
f~i · δ~ri = 0 (1.7)
i

The equation of motion is given as

F~itot = p~˙i (1.8)
So we get
X
F~itot · δ~ri = 0 (1.9)
i
X  X
F~i + f~i · δ~ri = p~˙i · δ~ri = 0 (1.10)
i i
X X
F~i · δ~ri = p~˙i · δ~ri = 0 (1.11)
i i
X 
˙p~i − F~ · δ~ri = 0 (1.12)
i
i

This above equation is called D’Alembert’s Principle.

Let us now write the transformation equation as
~ri = ~ri (q1 , q2 , · · · , qn , t) (1.13)
assuming n = 3N − k number of independent coordinates. (this is position of ith particle
and there are N number of particles subject to k no of constraints relations. So there are
3N-k number of degrees of freedom.)

d~ri X ∂~ri ∂~ri

~vi = = q˙k + (1.14)
dt k
∂qk dt

From the above equation we get some important relations. One is derivative of particle
velocity with respect to generalized co-ordinate qj
   
∂~vi ∂ d~ri d ∂~ri
= = (1.15)
∂qj ∂qj dt dt ∂qj
∂~vi X ∂ 2~ri ∂ 2~ri
= q˙k + (1.16)
∂qj k
∂q j ∂qk dt∂q j

Other is derivative of particle velocity with respect to generalized velocity q̇j

∂~vi ∂~ri
= (1.17)
∂ q̇j ∂qj

7
Chapter 1. Classical Mechanics of Point Particles

The virtual displacement δ~ri can be written as

X ∂~ri
δ~ri = δqj (1.18)
j
∂q j

In terms of the generalized coordinates the virtual work of applied force F~i becomes
!
X X X ∂~ri X X ∂~
ri
X
F~i · δ~ri = F~i · δqj = F~i · δqj = Qj δqj (1.19)
i i j
∂qj j i
∂qj j

Now the work done from the rate of change of momentum is written as
X X X X ∂~ri X ∂~ri
p~˙i · δ~ri = mi~r¨i · δ~ri = mi~r¨i · δqj = mi~r¨i · δqj (1.20)
i i i j
∂q j i,j
∂q j

X d  ∂~ri
 
d ∂~ri

= ˙
mi~ri · ˙
− mi~ri · δqj (1.21)
i,j
dt ∂qj dt ∂qj

Using Eq. (1.15) and Eq. (1.17) we get

X X d  ∂~vi

∂~vi

˙p~i · δ~ri = mi~vi · − mi~vi · δqj (1.22)
i i,j
dt ∂ q̇j ∂qj

Define total kinetic energy

X1
T = mi vi2 (1.23)
i
2
∂T X ∂~vi
= mi~vi · (1.24)
∂ q̇j i
∂ q̇j
∂T X ∂~vi
= mi~vi · (1.25)
∂qj i
∂qj

Thus from Eq. (1.22)

X X  d  ∂T  ∂T 
p~˙i · δ~ri = − δqj (1.26)
i j
dt ∂ q̇ j ∂q j

Now combining Eq. (1.12, 1.19 and 1.26) we get

X  d  ∂T  ∂T 
− − Qj δqj = 0 (1.27)
j
dt ∂ q̇ j ∂q j

Points to ponder: The constraints are holonomic and the generalized co-ordinates are
independent of each other, and the virtual displacements δqk are linearly independent.

8
 1.7. Hamilton’s principle

P
Thus Individual coefficints of δqj are set to zero. So next we drop j part.!! (Attend
class to understand this:)
∂V
Qj = − (1.28)
∂qj

The potential V is independent of generalized velocities.

   
d ∂(T − V ) ∂(T − V )
− =0 (1.29)
dt ∂ q̇j ∂qj

We define a new function , called the Lagrangian L as

L=T −V (1.30)

and finally get

 
d ∂L ∂L
− =0 (1.31)
dt ∂ q̇j ∂qj

The above Eq. (1.31) is called Euler-Lagrange equation or simply Lagrange equation.

1. Find the Euler Lagrange equation for velocity dependent potential. (i.e., for dissi-
pation case)

1.7 Hamilton’s principle

Total energy is conserved:
On conservation of energy: Consider L = L(qj , q̇j )
 
∂L d ∂L
− =0 (1.32)
∂ q̇j dt ∂qj
dL ∂L ∂qj ∂L ∂ q̇j
= + (1.33)
dt ∂qj ∂t ∂ q̇j ∂t
   
dL d ∂L ∂L ∂ q̇j d ∂L d
= q̇j + = q̇j = (pj q̇j ) (1.34)
dt dt ∂ q̇j ∂ q̇j ∂t dt ∂ q̇j dt
d
(pj q̇j − L) = 0 (1.35)
dt
Thus we get
n
X
pi q̇i − L = constant (1.36)
i=1

Let us consider

9
Chapter 1. Classical Mechanics of Point Particles

X1
T = mq̇i2 (1.37)
i
2
∂T
= mq̇j = pj (1.38)
∂ q̇j
X X
pj q̇j = mq̇j2 = 2T (1.39)
j j

We define the quantity

n
X
pi q̇i − L = constant = H = 2T − L = T + V (1.40)
i=1

this is total energy of a system.

1.8 Lagrangian and the Euler-Lagrange equations

L = T (q˙j ) − V (qj ) (1.41)

1 1
= mq˙j 2 − kqj2 (1.42)
  2 2
d ∂L ∂L
− =0 (1.43)
dt ∂ q̇j ∂qj

Cyclic coordinate
Lagrangian is function of generalised coordinates, generalised velocities and time. Thus

L = L(q1 , q2 , · · · , qn , q̇1 , q̇2 , · · · , q̇n , t) (1.44)

Now suppose for some system we encounter qjth coordinate is absent from the Lagrangian.

L = L(q1 , q2 , · · · , qj−1 , qj+1 , · · · , qn , q̇1 , q̇2 , · · · , q̇n , t) (1.45)

Thus from Euler-Lagrangian equation we have

 
d ∂L ∂L
= =0 (1.46)
dt ∂ q̇j ∂qj
∂L
= pj (let0 s say) constant (1.47)
∂ q̇j

1.9 1D Examples
one-dimensional examples of the Euler-Lagrange equations- one-dimensional Simple Har-
monic Oscillations and falling body in uniform gravity; applications to simple systems
such as coupled oscillators Canonical momenta Hamiltonian.

10
 1.10. Hamilton’s equations of motion and its Applications:

1.10 Hamilton’s equations of motion and its Appli-

cations:
Let us write a function f which is function of two variables x and y.

f = f (x, y) (1.48)
∂f ∂f
df = dx + dy (1.49)
∂x ∂y
= udx + vdy (1.50)
∂f ∂f
u= , v= (1.51)
∂x ∂y
We intend to change the variable of the function from (x, y) to (u, y). Thus the functional
form changes from f (x, y) to g(u, y). Let us consider

g = ux − f (1.52)
dg = udx + xdu − df (1.53)
= udx + xdu − udx − vdy (1.54)
dg = xdu − vdy (1.55)
∂g ∂g
dg = du + dy (1.56)
∂u ∂y
From the above relations we get
∂g ∂g
x= , v=− (1.57)
∂u ∂y
These relations are called Legendre transformation equation.
let us replace

f =L (1.58)
x = q̇ (1.59)
y=q (1.60)
u=p (1.61)
v = ṗ (1.62)
g=H (1.63)

Thus we get equivalent equation known as Hamilton’s equation of motion as

∂H ∂H
q̇ = , ṗ = − (1.64)
∂p ∂q

Hamiltonian for a harmonic oscillator, solution of Hamilton’s equation for Simple

Harmonic Oscillations; particle in a central force field- conservation of angular momentum
and energy.

11
Chapter 1. Classical Mechanics of Point Particles

12
 CHAPTER 2
SMALL AMPLITUDE OSCILLATIONS

2.1 Potential Energy

• If the system is conservative, Potential Energy is function of generalised coordinates
only. V = V (q1 , q2 , q3 , · · · )

• The system is in equilibrium when the generalised force vanishes

1 d2 V
   
1 dV
V (x) = V (x0 ) + (x − x0 ) + (x − x0 )2
1! dx x0 2! dx2 x0
1 d3 V
 
+ (x − x0 )3 + · · · (2.1)
3! dx3 x0

Let us understand how to get Taylor expansion of a function for two variable function.
Consider the function
∆t ∂F (∆t)2 ∂ 2 F
F (t) = F (t0 + ∆t) = F (t0 ) + + + ··· (2.2)
1! ∂t 2! ∂t2
Let us consider the function F (t) = f (x0 + t∆x, y0 + t∆y). We re-define x(t) = x0 + t∆x
and y(t) = y0 + t∆y. Thus we can do Taylor expansion of the function F (t) with respect

13
Chapter 2. Small Amplitude Oscillations

to t0 = 0 i.e.; for F (0) = f (x0 , y0 ) and at t = 1, F (1) = f (x0 + ∆x, y0 + ∆y). Thus
∆t = 1. We calculate
∂F ∂f ∂x ∂f ∂y ∂f ∂f
= + = ∆x + ∆y (2.3)
∂t ∂x ∂t ∂y ∂t ∂x ∂y
Similarly we have

∂ 2F
 
∂ ∂f ∂f
= ∆x + ∆y (2.4)
∂t2 ∂t ∂x ∂y
   
∂ ∂f ∂f ∂x ∂ ∂f ∂f ∂y
= ∆x + ∆y + ∆x + ∆y (2.5)
∂x ∂x ∂y ∂t ∂y ∂x ∂y ∂t
2 2 2 2
∂ f ∂ f ∂ f ∂ f
= (∆x)(∆x) + (∆y)(∆x) + (∆x)(∆y) + (∆y)(∆y) (2.6)
∂x∂x ∂x∂y ∂y∂x ∂y∂y
Now collecting all the terms we get the two variables Taylor expansion as

∆t ∂F (∆t)2 ∂ 2 F
F (t) = F (t0 + ∆t) = F (t0 ) + + + ··· (2.7)
1! ∂t 2! ∂t2
1 ∂F 1 ∂ 2F
F (x0 + ∆x, y0 + ∆y) = F (0) + + + ··· (2.8)
1! ∂t 2! ∂t2 
∂f ∂f
F (x0 + ∆x, y0 + ∆y) = f (x0 , y0 ) + ∆x + ∆y +
∂x ∂y
 2
∂ 2f ∂ 2f

1 ∂ f 2 2
+ (∆x) + 2 (∆y)(∆x) + (∆y) + · · · (2.9)
2! ∂x∂x ∂x∂y ∂y∂y

In general for small displacement (linear term!) qi = qi0 + ηi

d2 V
 
1 XX
V (q1 + η1 , q2 + η2 , q3 + η3 , · · · ) = V (q10 , q20 , q30 · · · ) + η1 ηj + ···
(2.10)
2! i j dqi ∂qj qi0 ,qj0

Understand how the second order term is coming.

The odd terms vanishes since  
dV
=0
dx x0
Minima of potential energy and points of stable equilibrium, expansion of the potential
energy around a minimum, small amplitude oscillations about the minimum,
We will be interested into Taylor expansion of potential V which is a function of
variables q1 , q2 , q3 · · · etc. The expansion is around the point q10 , q20 , · · · . When the
potential is single variables we have

(x − x0 )2 ∂ 2 V
   
∂V
V (x) = V (x0 ) + (x − x0 ) + + ··· (2.11)
∂x x=x0 2! ∂x2 x=x0

How the above expression works when V = V (q1 , q2 , q3 , · · · )

14
 2.2. Small Oscillation about Potential minima

2.2 Small Oscillation about Potential minima

normal modes of oscillations.
We shall study two similar problem here. One is frequency of oscillation for a tri-
atomic molecule like CO2 .

k k

m m
M

Figure 2.1: CO2 molecule as a 3 body and 2 spring system.

The Lagrangian is given by

   
1 2 1 2 1 2 1 2 1 2
L=T −V = mẋ1 + M ẋ2 + mẋ3 − k(x2 − x1 ) + k(x3 − x2 ) (2.12)
2 2 2 2 2

The Euler Lagrange equation of motion are written as

 
d ∂L ∂L
− =0 (2.13)
dt ∂ ẋ1 ∂x1
 
d ∂L ∂L
− =0 (2.14)
dt ∂ ẋ2 ∂x2
 
d ∂L ∂L
− =0 (2.15)
dt ∂ ẋ3 ∂31

And we get
k k
ẍ1 + x1 − x2 = 0 (2.16)
m m
2k k k
ẍ2 + x2 − x1 − x3 = 0 (2.17)
M M M
k k
ẍ3 + x3 − x2 = 0 (2.18)
m m
q
k
We note here the natural frequency of a simple spring mass system is ω0 = m .
Students are requested to derive the above three equations explicitly, step-by-step form
the Lagrangian and
Let us consider the solutions are of the form

x1 = A1 eiωt (2.19)
x2 = A2 eiωt (2.20)
x3 = A3 eiωt (2.21)

15
Chapter 2. Small Amplitude Oscillations

ω is the unknown frequency with which the system may oscillate and for the problem at
hand, we are here to find the possible values of ω. And of course the unknown amplitudes
A1 , A2 and A3 are to be determined. (or their relations among themselves.)
Let us substitute the assumed solutions in Equations (2.16, 2.17, 2.18) and get

k k
( − ω 2 )A1 − A2 + 0.A3 = 0 (2.22)
m m
k 2k 2 k
− A1 + ( − ω )A2 − A3 = 0 (2.23)
M M M
k k
0.A1 − A2 + ( − ω 2 )A3 = 0 (2.24)
m m
In matrix form we have

k k
  
(m − ω2) −m 0 A1
 −k 2k
(M − ω2) − Mk   A2  = 0 (2.25)
M
k k 2
0 −m (m − ω ) A3
We should find out the determinant of the above 3 by 3 matrix and find the value of
ω. In other words we have to find the eigenvalues and eigenvectors of the 3 by 3 matrix.
 k k

m
−m 0
 −k 2k
− Mk 
M M
k k
0 −m m
q q
k 2k k
The eigenvalues are ω1 = 0, ω2 = m and ω3 = M +m . Let us calculate the eigenvec-
tors or eigen functions corresponding to three eigenvalues.
 
1
1
1. For ω1 = 0 we have A1 = A2 = A3 . So the eigen function is Ψ1 = √3  1 
1
Corresponds to no oscillation, i.e., translation of the system as a whole.
q
k
2. For ω2 = m , we solve for A1 , A2 , A3 , and we get A1 = −A3 and A2 = 0. We may
 
1
write the eigen function as Ψ2 = √12  0  , this corresponds to oscillation of the
−1
mass of Oxygen atoms in opposite direction. Called symmetric mode of vibration.
q
2k k
3. For ω3 = M +m , we have to put this value in Eq. (2.43) and get A2 = − 2m M
A1
 
A1
2m
and A3 = A1 . Thus the eigen function is Ψ3 =  − M A1  or the normalised form
  A1
1
of the eigen vector is Ψ3 = q 1 m2  − 2m M
 . This is called antisymmetric mode
2+4 2
M 1

16
 2.2. Small Oscillation about Potential minima

of vibration and here the oxygen atoms vibrate in phase and the carbon atom moves
opposite to the direction of vibration.

See the figure to understand the three modes of vibration calculated above.

O C O

O C O

O C O

Figure 2.2: Normal modes and normal mode frequencies of CO2 molecule. Treated clas-
sically.

17
Chapter 2. Small Amplitude Oscillations

N atoms connected with (N-1) strings

The other problem is to find out normal mode frequencies of a system of N identical masses
connected in a linear fashion to (N -1) - identical springs. Without loss of generality, we
have attached the first and last particle with the wall with strings. This is to maintain
the boudary conditions.

k k k

m m m
m

Figure 2.3: A chain with N number of molecules connected with N − 1 number of springs.

The kinetic energy

1
T = m ẋ21 + ẋ22 + ẋ23 + ẋ24 + ẋ25 · · · + ẋ2N −1 + ẋ2N


(2.26)
2
and the potential energy is
1 
V = k ∗ x21 + (x2 − x1 )2 + (x3 − x2 )2 + (x4 − x3 )2 + · · · + (xN − xN −1 )2 + ∗ x2N (2.27)

2
We can write the Euler Lagrange’s equation of motion as
 
d ∂L ∂L
= (2.28)
dt ∂ ẋ1 ∂x1
 
d ∂L ∂L
= (2.29)
dt ∂ ẋ2 ∂x2
··· = ···
··· = ···
 · · · = · · ·
d ∂L ∂L
= (2.30)
dt ∂ ẋN ∂xN
Where L = T − V . The equation of motion or Euler-Lagrange equation of motion are
writtesn as

mẍ1 + ∗ 2kx1 − kx2 =0 (2.31)

mẍ2 + 2kx2 − kx1 − kx3 =0 (2.32)
mẍ3 + 2kx3 − kx2 − kx4 =0 (2.33)
mẍp + 2kxp − kxp−1 − kxp+1 =0 (2.34)
mẍN + ∗ 2kxN − kxN −1 =0 (2.35)

18
 2.2. Small Oscillation about Potential minima

k
Let us devide each equation by m and define ω02 = m . Here ω0 is the natural frquency of
a single spring-mass system. To solve the above set of coupled differential equation we do
the following. First we consider the solutions are of the form

x1 = A1 eiωt (2.36)
x2 = A2 eiωt (2.37)
··· ···
xp = Ap eiωt (2.38)
··· ···
xN = AN eiωt (2.39)
(2.40)

this is because, all the particles oscillates with same frequency ω but with different am-
plitutes. And there could be lots of ways the system can oscillate. Actually N different
ways with maximum N different frequencies. Now the way of oscillation is denoted by
the eigenvector of the matrix. And the frequency of oscillation is given by the unknown
ω which we are interested into, in the following exercise. The coefficient of the solutions
Ap and the ω are the unknown quantities. Solving the set of N second order differential
equations means we get eigen frequencies ω1 , ω2 , · · · , ωN . And each eigenvalue or each
frequency ω1 will give us a set of (A1 , A2 , · · · , AN ). This set of A0 s are the eigenvector or
”way of oscillation”. We substitute the assumed solutions into the differential equations
and get

( 2k − ω2) k
  
m
−m 0 0 0 0 A1
k 2k 2 k

 −m (m − ω ) −m 0 0 0 
 A2 

k 2k 2 k

 0 −m (m − ω ) −m 0 0 
 ··· 
=0
k

 0 0 −m ( 2k
m
− ω 2
) − k
m
0 
 Ap 

k
 0 0 0 −m ( 2k
m
− ω2) −mk  ··· 
k 2k
0 0 0 0 −m ( m − ω2) AN

Now we do the following. As we understand that the there is no particle at the edges,
x0 = 0 and xN +1 = 0. So the index p runs from 1 to N. Let us look at the Eq. (2.34)

mẍp + 2kxp − kxp−1 − kxp+1 = 0

and substitute xp = Ap eiωt . We get

(−ω 2 + 2ω02 )Ap − ω02 Ap−1 − ω02 Ap+1 eiωt = 0

(2.41)
Ap+1 + Ap−1 2ω 2 − ω 2
= 0 2 (2.42)
Ap ω0

Let us consider the amplitude of pth particle is

Ap = B sin pθ. (2.43)

19
Chapter 2. Small Amplitude Oscillations

where B is constant. As p goes from 1, 2, · · · , N ; we must have this information in θ. this

is the time to write and remind the complete solution as

xp = B sin (pθ) eiωt

As we have boundary conditions xN +1 = 0, we get

AN +1 = 0 (2.44)
sin(N + 1)θ = 0 = sin(mπ) (2.45)
mπ
θ= ; m = 1, 2, · · · , N (2.46)
N +1

here θ has N different values, which will give N different values of Ap . Let us find
relations between the coefficients Ap , Ap−1 and Ap+1 from Eq. (2.42). This will give us
the eigenvalues or the eigen frequency ω.

Ap+1 + Ap−1 2ω02 − ω 2

= (2.47)
Ap ω02
sin (p + 1)θ + sin(p − 1)θ 2ω 2 − ω 2
= 0 2 (2.48)
sin(pθ) ω0
ω2
2 cos θ = 2 − 2 (2.49)
ω0
θ mπ
ω = 2ω0 sin = 2ω0 sin (2.50)
2 2(N + 1)
mπ
ωm = 2ω0 sin (2.51)
2(N + 1)

Once we get the eigenvalue ωm , we can derive  the eigen

 functions from Eq.(2.43).
A1
 A2 
 
 ··· 
Each value of m, we get the eigen vector as Ψm =    for p = 1, 2, 3, · · · , N from
 A p


 ··· 
AN
Ap = B sin(pθ). We can in principle re-derive the known solutions of 3 atoms sytem (C02 )
and check the results.

Second Method
A good reference on this topic: Greiner: System of Particles and Hamiltonian Dynamics,
Chapter-7: Vibrations of coupled mass points, Page: 83.

20
 2.2. Small Oscillation about Potential minima

To find the eigen frequencies we solve the following matrix given as

 2k k
( m − ω2)

−m 0 0 0 0
k 2k 2 k

 −m (m − ω ) −m 0 0 0 

k 2k 2 k
 0 − m
( m
− ω ) − m
0 0 
DN =  k 2k 2 k
 (2.52)

 0 0 −m (m − ω ) −m 0 

k
 0 0 0 −m ( 2k
m
− ω 2
) − k
m

k
0 0 0 0 −m ( 2k
m
− ω2)

The eigenfrequencies are obtain as solution of the equation DN = 0. Expanding DN with

respect to first row, we get

DN = (2ω02 − ω 2 )DN −1 − ω04 DN −2 (2.53)

* Let us consider the solution of the determinant equation as

DN = ω02N sin ((N + 1)Θ) (2.54)

Substituting this equation in Eq (2.53), we get

ω = 2ω0 sin(Θ/2) (2.55)

For any nontrivial soltion of the system of equations , we must have DN = 0, i.e., DN
must vanish for all N , we have

sin ((N + 1)Θ) = 0; (2.56)

pπ
Θ= ; here p =1,2,3, · · · N integer (2.57)
N +1
The eigen frequencies are
 
pπ
ωp = 2ω0 sin (2.58)
2(N + 1)

Once we get the eigen values, we can derive the eigen functions from the matrix Eq. (2.52)
after taking care of the boundary conditions properly. We can in principle re-derive the
known solutions of 3 atoms sytem (C02 ) and check the results.

21
Chapter 2. Small Amplitude Oscillations

22
 CHAPTER 3
SPECIAL THEORY OF RELATIVITY

3.1 Postulates of Special Theory of Relativity

3.2 Lorentz Transformations
One can write the Lorentz transformation as
x̄µ = Λµν xν (3.1)
Here Λµν is called the Lotentz transformation matrix.

3.3 Minkowski space

In Minkowski space-time notation, both space and time are treated in same dimension
namely the space dimention. Time is multiplied with velocity of light c to get the dimen-
sion as meter. The Lorentz transformation is written as
x̄0 = γ(x0 − βx1 ) (3.2)
x̄1 = γ(x1 − βx0 ) (3.3)
x̄2 = x2 (3.4)
x̄3 = x3 (3.5)
here we define the symbols as
v
β= (3.6)
c
1
γ=q (3.7)
v2
1+ c2

v is the velocity of intertial frame O0 with respect to O. along x axis (say). x1 = x,

x = y, x3 = z and x0 = ct
2

23
Chapter 3. Special Theory of Relativity

3.4 The invariant interval

The interval is defined as

I = (∆x)µ (∆x)µ = −(∆x0 )2 + (∆x1 )2 + (∆x2 )2 + (∆x3 )2 (3.8)

3.5 Light cone

3.6 World lines

3.7 Space-time diagrams

3.8 Time-dilation

3.9 Length contraction

3.10 Twin paradox

3.11 Four-vectors:

3.12 Space-like, time-like and light-like.

Refrerring to Eq (3.8),

1. For I < 0, the interval is time like, as at the same place two events are seperated
by time or temporal seperation only.

2. For I > 0, the events are happening at same time, but at different space points.

3. For I = 0, the interval is light like.

24
 3.13. Four-velocity and acceleration.

3.13 Four-velocity and acceleration.

Keeping in mind that the frame velocity of O0 is v whereas the particle velocity is u. τ is
proper time where as t is relativistic time.
dxµ
ηµ = (3.9)
r dτ
u2
dτ = 1 − 2 dt (3.10)
c
dx0 c
η0 = =q (3.11)
dτ 1− u
2
c2
0 0 1
η̄ = γ(η − βη ) (3.12)
η̄ 1 = γ(η 1 − βη 0 ) (3.13)
η̄ 2 = η 2 (3.14)
η̄ 3 = η 3 (3.15)
When we represent acceleration as four-acceleration vector we define
dη µ d2 xµ
αµ = = (3.16)
dτ dτ 2

3.14 Metric and alternating tensors.

3.15 Four-momentum and energy-momentum relation
The momentum is defined as
p~ = m~η (3.17)
where p is for momentum, m is mass and η is velocity. In tensor notation, (in four vector
formalism) we have
pν = m η ν (3.18)
All the four components of the momentum can be writen as
c
p0 = mη 0 = m0 q (3.19)
2
1 − uc2
p~ = m ~u (3.20)
m0
m= q (3.21)
2
1 − vc2
pµ pµ = (~p · p~) − (p0 )2 = −m20 c2 (3.22)
E
p0 = (3.23)
c
In is a good exercise to ckeck the famous equation
E 2 = p2 c2 + m20 c4 (3.24)

25
Chapter 3. Special Theory of Relativity

3.16 Doppler effect from a four-vector perspective

3.17 Concept of four-force
Newton’s first law for motion is valid in Special theory of relativity as
d~p
F~ = (3.25)
dt
But Newton’s

3.18 Conservation of four-momentum

3.19 Relativistic kinematics
Total energy is related to kinetic energy and rest mass energy as

E = mc2 = K + m0 c2 (3.26)

Momentum is writtes as
m0
p = mu = q u (3.27)
u2
1− c2

3.20 Application to two-body decay of an unstable

particle.
Let us assume an ustable Pion at rest decays into muon and neutrino (massless). The
problem is to find the energy and velocity of the outgoing muon in terms of rest mass of
piopn and muon. For this we have two conserved quantuty. Momentum and total energy.

pπ = 0 (3.28)
pµ + pν = 0 (3.29)

Thus muon and neutrino fly in opposete direction with equal momentum as the pion
was at rest before blast.
From the energy consideration we have rest mass energy of the pion befor decay is
equal to the total energy of muon and neutrino after decay.

mπ c2 = Eµ + Eν (3.30)
Eµ = Kµ + mµ c2 (3.31)
Eµ2 = p2µ c2 + m2µ c4 (3.32)
E ν = K ν = pν c = pµ c (3.33)

26
 3.20. Application to two-body decay of an unstable particle.

mπ and mmu are the rest mass of pion and muon respectively. Kµ and Kν are the kinetic
energy of muon and neutrino respectively. As neutrino has zero rest mass, its energy is
totally the kinetic energy.
Thus we get
q
2 2 2
Eµ = mπ c − Eν = mπ c − pµ c = mπ c − Eµ2 − m2µ c4 (3.34)
q
Eµ2 − m2µ c4 = mπ c2 − Eµ (3.35)

Squaring and solving for Eµ we get

c2 (m2µ + m2π )
Eµ = (3.36)
2mπ
And also the velocity of the muon is got from the relations below:
 
1 mµ  2
KE(µ) = Eµ − mµ c2 =  q u (3.37)
2 1− u
2
c2

27
Chapter 3. Special Theory of Relativity

28
 CHAPTER 4
FLUID DYNAMICS

4.1 Density ρ and pressure P in a fluid

Attend class:

4.2 Continuity equation and mass conservation

Figure 4.1: Fluid entering and coming out from a cubical box of length ∆x width ∆y and
height ∆z

small force acting on the x = 0 plane, is

Fx (at x = 0) = P ∆y∆z (4.1)

29
Chapter 4. Fluid Dynamics

Force acting on x = ∆x plane is

∂P
Fx (at x = ∆x) = (P + ∆x)∆y∆z (4.2)
∂x
Net force is
∂P ∂P
Fx = ∆x∆y∆z = ∆V (4.3)
∂x ∂x
∂P
∆max = ∆V (4.4)
∂x
1 ∂P
ax = ; ρ = density (4.5)
ρ ∂x
In vector form the acceleration looks like
1~
~a = ∇P (4.6)
ρ
We should introduce inertial or effective force per unit mass as f~ in the form
d~v 1~
f~ = = ~a − ∇P (4.7)
dt ρ

4.3 Stream-lined motion, laminar flow, Poiseuille’s

equation for flow of a liquid through a pipe
When a liquid flows through a tube, motions are generally two types. Turbulent motion
and steady or streamline motion. If the pressure difference across two ends of the tube
is small, then the motion is stream line. If the pressure difference is large, the motion
is turbulent. But how small be the pressure difference, that is one thing to think about.
Nonetheless we consider the simplest case of regular motion where the motion is flowing
uniformly, named streamline motion and there is no source or sink inside the tube. That
means equation of continuity is valid in the region and if one think about the water flow
as a field vector, the the curl of the vector is zero. etc..

Poiseuille’s Equation for Flow of a Liquid through a Capillary

Tube.
Method 1
The viscous force acting on the inner wall of the cylindrical shell (of radius r) is
dv
F1 = 2πrlη (4.8)
dr
the force (higher in value) acting on the outer surface of cylindrical shell (of radius r + dr)
is
 
d dv
F2 = F1 + 2πrlη dr (4.9)
dr dr

30
4.3. Stream-lined motion, laminar flow, Poiseuille’s equation for flow of a liquid through
a pipe
d
2πrlη dv


Thus the net force is dr dr
dr. This net force is balanced by the force arinsing
due to pressure difference between the two ends of the cylindrical tube. At the left the
force is
0
F1 = P × 2πrdr (4.10)

and at the right end of the cylinder (where pressure is low as the liquid is flowing from
left to right) the force is

0 dP
F2 = (P + l)2πrdr (4.11)
dx
Equating the net force due to pressure difference and net force due to viscous drag
 
d dv dP
2πrlη dr = l × 2πrdr (4.12)
dr dr dx

We set dP
dx
= − Pl as the pressure gradient is uniform. Ans get the differential equation of
the form
d dv P
(r ) = −r (4.13)
dr dr ηl
We integrate this above equation and get

dv P r2
r =− +A (4.14)
dr 2ηl
dv Pr A
=− + (4.15)
dr 2ηl r
Again we integrate and get

P r2
v(r) = − + A log r + B (4.16)
4ηl
Now the problem at hand has definite solution if we impose boundary conditions of your
choice. As for example r = 0, v(r) = vmax and at r = R, v = 0. We set A = 0 for the
first condition, otherwise velocity becomes infinite log(0) = -ve infinite. etc..

Method 2
Let us consider a tube of cylindrical shape, with uniform cross section with radius R
and length l. Consider concentric cylinders of radius r, whole length is also l. We are
studying fluid motion under the influence of viscosity. Thus the viscous drag is considered.
This viscous force is such that the velocity of fluid touching the wall of the tube is zero.
Whereas the velocity of the fluid at the center of the cylinder is maximum. The force due
to viscosity is proportional to the surface area A(= 2πrl) as it acts like a frictional force on
the immediate concentric cylinders. The viscous force is proportional to velocity gradient

31
Chapter 4. Fluid Dynamics

Figure 4.2: Viscous force on laminar flow

dv


dr
. η being the proportionality constant called viscous coefficient. The equation is given
by
dv dv
F = ηA = η2πrl (4.17)
dr dr
The flow of liquid is happening due to pressure difference at two ends of the tube. If
pressure difference is P , we have
dF = −P 2πrdr (4.18)
Integrating both side we have
Z r=r
F = −2πP rdr = −P π r2 (4.19)
r=0

Thus
dv
η2πrl = −P π r2 (4.20)
dr
P πr2 Pr
dv = − dr = − dr (4.21)
2πrlη 2lη
Integrating we get the velocity for a fluid on the surface of cylindrical tube of radius r.
We keep in mind r = R, the velocity is zero. And at r = r, we have velocity v = v(r)
which is function of radius r of the tube. This decides the integration limits.
Z r
Pr P
v = v(r) = − dr = (R2 − r2 ) (4.22)
r=R 2lη 4lη
In Poiseuille’s method we collect the water flowing through a capillary tube and take
data for collected fluid volume per second. When the fluid is flowing with velocity v, the
volume of liquid collected per second is given as (v 2 π r dr). Integrating we get the total
volume of the fluid collected
Z R
Q P Pπ 4
= 2 π v r dr = 2 π (R2 − r2 ) r dr = R (4.23)
t r=0 4lη 8lη
Thus working formula for coefficient of viscosity of a fluid η is given as
Pπ
η= R4 (4.24)
8l(Q/t)

32
 4.4. Navier-Stokes equation

4.4 Navier-Stokes equation

4.5 Qualitative description of turbulence
4.6 Reynolds number

33