EP2102 Classical Dynamics

Aditya Mohapatra EP21B002

Contents
1 Newtonian Mechanics 2

2 Calculus of Variation 2
2.1 Snell’s Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2 Problem of calculus of variation . . . . . . . . . . . . . . . . . . . 3
2.2.1 Parametrasition: . . . . . . . . . . . . . . . . . . . . . . . 4
2.2.2 Brachistochronic Problem . . . . . . . . . . . . . . . . . . 5
2.2.3 Geodesic . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2.4 The δ Notation . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3 Hamiltonian Principle and Lagrange Equation 8

3.1 Hamiltonian Principle . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2 Generalized Coordinates . . . . . . . . . . . . . . . . . . . . . . . 8
3.3 Finding Equation of motion . . . . . . . . . . . . . . . . . . . . . 9

4 Essence of Lagrangian 11
4.1 Canonical Momentum . . . . . . . . . . . . . . . . . . . . . . . . 11
4.2 Homogeneity of time : energy conservation . . . . . . . . . . . . 11
4.3 Homogenity of space : Momentum consevation . . . . . . . . . . 12
4.4 Isotropy of Space : Conservation of Angular momentum . . . . . 12
4.5 Mechanical Similarities . . . . . . . . . . . . . . . . . . . . . . . . 14
4.6 The Virial Theorem . . . . . . . . . . . . . . . . . . . . . . . . . 15

5 Integration of Equations of motion 17

5.1 Motion in One dimension . . . . . . . . . . . . . . . . . . . . . . 17
5.2 Central Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.2.1 Turing points in central force . . . . . . . . . . . . . . . . 18
5.2.2 Effective potential . . . . . . . . . . . . . . . . . . . . . . 18
5.3 Kepler’s Problem (1/r potential) . . . . . . . . . . . . . . . . . . 19
5.4 The Laplace-Runge-Lenz(LRL) Vector . . . . . . . . . . . . . . . 21
5.4.1 Stable circular orbits: . . . . . . . . . . . . . . . . . . . . 22

6 Scattering 22
6.1 Scattering Cross-section . . . . . . . . . . . . . . . . . . . . . . . 23

1
1 Newtonian Mechanics
d⃗
p d2 r
F⃗ = m =m
dt dt

Conservation laws:
Total force = 0 , =⇒ p⃗˙ = 0 =⇒ p⃗ is conserved.

2 Calculus of Variation
2.1 Snell’s Law

Let speed be v1 , v2 in n1 , n2

2
 h1 = xcot(θ)

h2 = (c − x)cot(θ)

h21 +x2 h22 +(c−x) 2

tD1 +D2 = v1 + v2

Fermat’s principle of least time says time has to be minimum.

dtD1 +D2
=0
dx

=⇒ v2 sin(θ1 ) = v1 sin(θ2 )
So Nature follows that path which minimizes the action. In the case of snell’s
law time is equivalent to action.

We use calculus for variation to extremize the action.

2.2 Problem of calculus of variation

We have a function f (y(x), y ′ (x), x)dx defined on y = y(x) between x1 and x2 .
R x2
y(x) for which J = x1
f (y(x), y′ (x), x)dx is an extremum.

3
 dy
Here y ′ (x) = dx ,
x is an independent variable

2.2.1 Parametrasition:
We put the above problem to the apparatus of differential calculus which we
know.
The value of J is stationary relative to any neighbouring path to y(x) .We label
all the those neighbouring paths as y(x, a) where a is infinitesimal parameter.
And we donate y(x, 0) as the correct path. If we have any arbitrary function
η(x) such that η(x1 ) = η(x2 ) = 0. Then the equation which gives the set of all
paths is:

y(a, x) = y(0, x) + aη(x)

For such family of curves J is also a function of a.

Z x2
J(a) = f (y(a, x), y ′ (a, x), x)dx
x1

Since J is stationary for y(x, 0) relative to other members of that family.

d 
=⇒ J(a) =0

da a=0

Finding Solution of variational calculus problem:

R x2∂f ∂y
d
da J(a) = x1 ∂y ∂a + ∂f ∂ ẏ
∂ ẏ ∂a
x
∂f ∂y  2
R x2 ∂f ∂y
Rx d ∂f ∂y


= x1 ∂y ∂a + ∂ ẏ ∂a  − x12 dx ∂ ẏ ∂a dx
x1

Since all paths pass through x1 and x2 , so y ′ (a) = 0 at x1 , x2 .

x2
=⇒ ∂f ∂y 
∂ ẏ ∂a  =0
x1
R x2
=⇒ d
da J(a) = x1
( ∂f
∂y −
d ∂f ∂y
dx ∂ ẏ ) ∂a dx =0
∂y
∂a = η(x) which is arbitrary function. So other term inside the integral has to
be 0 for all x ∈ (x1 , x2 ).
∂f d ∂f
=⇒ − =0
∂y dx ∂ ẏ

4
 ∂f d ∂f
=⇒ =
∂y dx ∂ ẏ

This Equation is called Euler’s Equiation

2.2.2 Brachistochronic Problem

Find the shape of the curve down which a bead sliding from rest and
accelerated by gravity will slip (without friction) from one point to
another in the least time.

T + U = constant
T = kinetic energy , U = Potential energy
velocity
√ at a given y (initial velocity =0):
v = 2gy

Time require from A to B is :

B
p
dx2 + dy 2
Z
ds
tAB = = √
A v 2gy
x2
1  1 + ẏ 2  12
Z
= √
x1 =0 2g x

∂f
Since f doesn’t depend on y ; =⇒ ∂y =0
∂f d ∂f
∂y = dx ∂ ẏ

d ∂f
=⇒ dx ∂ ẏ =0
∂f 1
=⇒ ∂ ẏ = constant = 2a

x = a(1 − cos(θ))

dx = a sin(θ)dθ

5
 R
y= a(1 − cos(θ))dθ

By putting inituial condition

x = a(1 − cos(θ))
y = a(θ − sin θ)

2.2.3 Geodesic
Line that represents the shortest path between any two points when
the path is restricted to a particular surface.

Let us consider sphere

considering the geodesic constrained to follow the surface of a sphere of radius R.

the element of path length

q on the surface of the sphere is given in spherical
2 2
coordinates as ds = R dθ + (sin θdϕ) .
Therefore the distance s between two points A and B is:
s 
Z B  2
dθ
s=R  + sin2 θ dϕ
A dϕ

The function f for ensuring that s be an extremum value uses

p
f = θ′2 + sin2 θ
Now solving this using varitaional calculus gives the trajectory.

2.2.4 The δ Notation

∂J
R x2 ∂f d ∂f
∂a = x1 ( ∂y − dx ∂ ẏ )η(x)dx
R x2
∂J
∂a da = x1
( ∂f
∂y −
d ∂f
dx ∂ ẏ )η(x)dadx

∂J
δJ = da
∂a
∂y
δy = da
∂a

δy represents difference of varied path from the correct y(0, x) at point x.

(It also corresponds virtual displacement, Read up about it.)

Similarly δJ is the variation from correct J at x.

Since J is stationary for correct path δJ = 0

6
2.3 Questions
Find the shortest path between the p (x, y, z) points (0, 1, 0) and (0,
1,0) on the conical surface z = 1 − x2 + y 2 What is the length of the
path? Note: this is the shortest mountain path around a volcano.

Solution:

In cylindrical coordinate the equation of surface is:

z =1−r

=⇒ dz = dr
p
ds = dr2 + dz 2 + r2 dϕ2
R Rp
s= ds = dr2 + dz 2 + r2 dϕ2
Rp
= dr2 + dr2 + r2 dϕ2
 
R q 2
= 2 + r2 dϕ
dr 2 dr
Z
J= f (ϕ, ϕ̇, r)

q 
2
Take f = 2 + r2 dϕ
dr 2

∂f
=0
∂ϕ
Since
d ∂f ∂f
= =0
dr ∂ ϕ̇ ∂ϕ

∂f
=⇒ = constant
∂ ϕ̇
r2 2ϕ̇
=⇒ q  =a
2+ r2 ϕ̇2

By solving this we can get condition.

7
3 Hamiltonian Principle and Lagrange Equation
3.1 Hamiltonian Principle
Of all the possible paths along which a dynamical system may move from
one point to another within a specified time interval (consistent with any con-
straints), the actual path followed is that which minimizes the time integral of
the difference between the kinetic and potential energies.

In terms of variational calculus:

If T = kinetic Energy
V = Potential Energy
2
Zt
δ (T − V )dt = 0
t1

Consider first a single particle, moving in a conservative force field. For such
a particle, the kinetic energy T will just be a function of the velocity(T = T (ẋ))
of the particle, and the potential energy will just be a function of the position
of the particle(V = V (x)). We define the Lagrangian is thus also a function of
the position and the velocity of the particle.

We will see the transformation from Euler equation to Lagranage:

x ←→ t

yi (x) ←→ xi (t)
yi (ẋ) ←→ xi (ṫ)
f (y, ẏ, x) ←→ L(x, ẋ)

And the Euler- Lagrange Equation is :

d ∂L ∂L
= =0
dt ∂ ẋ ∂x

3.2 Generalized Coordinates

• If we try to describe a system of n particles, we need in general 3n coor-
dinates to specify the position of its components.
• If external constraints are imposed on the system, the number of degrees
of freedom may be less. If there are m constraints applied, the number of
degrees of freedom will be 3n - m.

8
 • The coordinates do not need to be the coordinates of a coordinate system,
but can be any set of quantities that completely specifies the state of the
system. The state of the system is thus full specified by a point in the
configuration space (which is a 3n - m dimensional space).
• The generalized coordinates of a system are written as q1 , q2 , q3 ,
...... The generalized coordinates are of course related to the physical
coordinates of the particles being described:
The Euler-Lagrange Equation in generalized coordinates is:
d ∂L ∂L
− =0
dt ∂ q̇i ∂qi

3.3 Finding Equation of motion

Rt2
q = q(t) is the coordinate for which S = Ldt is minimum. // For variation
t1
δq in path ,

q = q + δq

δq(t1 ) = 0 and δq(t2 ) = 0

Zt2
δS = δ L(q, q̇, t)dt = 0
t1

Zt2  
∂L ∂L
δS = δq + δ q̇ dt
∂q ∂ q̇
t1

Zt2  
∂L ∂L d
= δq + δq dt
∂q ∂ q̇ dt
t1

Zt2  
∂L t2 ∂L d ∂L
= δq  + − δqdt = 0
∂ q̇ t1 ∂q dt ∂ q̇
t1

∂L d ∂L
− =0
∂q dt ∂ q̇

9
L = T- V
n
1 2
P
L= 2 mi q̇i − V (q1 , q2 , . . . , qn )]
i=1
∂L
∂ q̇i = mi q̇i

∂L ∂V
∂q = ∂q

∂V
=⇒ mi q̈i = − =F (Newton’s Law)
∂q

Body of mass m , position r = q1 , θ = q2

 
T =m
2 ṙ 2
+ r 2 2
θ̇

V = V (r)

∂L ∂L ∂V
∂ ṙ = mṙ and ∂r = mrθ̇2 − ∂r

∂V
=⇒ mr̈ = mrθ̇2 −
∂r

∂L ∂L
∂ θ̇
= mr2 θ̇ and ∂θ =0

10
4 Essence of Lagrangian
4.1 Canonical Momentum
Generalized momentum for jth generalized coordinates is:
∂L
pj =
∂ q̇j

If the potential is velocity dependent then canonical momentum is not nec-

essarily mechanical momentum.

4.2 Homogeneity of time : energy conservation

Time is homogeneous means one instant of time (or duration) is identical to any
other instant (or duration) of time.

Which means Lagrangian doesn’t explicitly depend on time; i.e. L(q, q̇, t) =
L(q, q̇)
∂L
=⇒ =0
∂t

we know:

∂L ∂L dq ∂L dq̇
= + (1)
∂t ∂q dt ∂ q̇ dt
And
∂L d ∂L
= (2)
∂q dt ∂ q̇

Using (2) in (1):

 
∂L d ∂L dq ∂L dq̇
∂t = dt ∂ q̇ dt + ∂ q̇ dt

 
d
=⇒ 0 = dt q̇ ∂L
∂ q̇

=⇒ q̇ ∂L
∂ q̇ = Constant

11
 q̇i ∂∂L
P
q̇i = 2T
i

2T − L = T + V = Energy = Constant

4.3 Homogenity of space : Momentum consevation

The Lagrangian doesn’t change changing position of origin in space.

L is constant for r −→ r + δr
 
d ∂L
δL = dt
δq + ∂L
∂ q̇
∂ q̇
δ q̇
 
d ∂L
=⇒ 0 = dt ∂ q̇
δq + ∂L d
∂ q̇ dt
δq

 
d ∂L d
=⇒ 0 = dt ∂ q̇
δq As dt
δq =0

 
d ∂L
=⇒ 0 = dt ∂ q̇
= ṗ

=⇒ p = constant

4.4 Isotropy of Space : Conservation of Angular momen-

tum
L is independent of orientation of coordinate system.

12
When the coordinate system is rotated by angle ϕ , we can write position and
velocity as :
⃗ × ⃗r
δ⃗r = δ ϕ
˙ = δϕ
δ⃗r ˙
⃗ ×⃗r

Also
X  ∂L ∂L ˙

δL = · δ⃗ri + · δ⃗ri =0
∂⃗ri ˙i
∂⃗r
i

X 
=⇒ ˙ i · δ⃗ri + p⃗i · δ⃗r
⃗p ˙i =0
i

X 
=⇒ ⃗ × ⃗ri + p⃗i · δ ϕ
p⃗i · δ ϕ ⃗ × ⃗ṙi = 0
i

⃗· d X
δϕ ⃗ri × p⃗i = 0
dt i

13
 ⃗ = ⃗r × p⃗ = Constant
This shows; Angular Momentum = L

4.5 Mechanical Similarities

Potential Energy is a homogeneous function of coordinates.

=⇒ V (η⃗r1 , . . . , η⃗rn ) = η k V (⃗r1 , . . . , ⃗rn )

Lets do the following transformations:

⃗rα −→ η⃗rα

t −→ βt

Then the following gets transformed by:

⃗vα −→ βη ⃗vα
η2
KE = β 2 KE

P E = ηk P E
k
If we take β = η 1− 2 then the L gets scaled by η k and eq of motion doesn’t
change.

• PE is homogeneous function of degree k , the equation of motion permits

a series of geometrically similar path and time
Let the transformations:(l, t) −→ (l′ , t′ )

t′ l′
t =β , l =η

 ′ 1− k2
t′ l
=
t l
k
(Put k = -1 , i.e V = r , you get t2 ∝ r3 , that is keplers law.)
′
(Put k = 1 , i.e V = kr2 , you get tt = 1 , which shows time period of harmonic
oscillator is independent of amplitude)

14
  ′  k2  ′ k
v′ l E′ l
= , =
v l E l

4.6 The Virial Theorem

rα is position of α th particle.

◦ Motion is bounded.

◦ V is homogeneous function of ⃗r.

Let each particle experience a force Fα

=⇒ ⃗p˙ α = F⃗α
P
Let G = α p⃗α · ⃗rα

=⇒ dG
=
P ˙ α +⃗p
p⃗α ·⃗r ˙ α · ⃗rα
dt
α
dG
2T + F⃗α · ⃗rα
P
=⇒ dt =
α

Rk
⟨ dG 1 dG
= 2 ⟨T ⟩ + ⟨ F⃗α · ⃗rα ⟩
P
dt ⟩ = k dt
0 α

1 X⃗
⟨T ⟩ = ⟨ Fα · ⃗rα ⟩
2 α

We know F⃗α = ∇Vα

And V (r) = crn+1

=⇒ ⃗r · ∇V (r) = (n + 1)V (r)

n+1
⟨T ⟩ = ⟨V ⟩
2

15
If degree of homogeneity is k :
2E kE
⟨V ⟩ = , ⟨T ⟩ =
k+2 k+2

16
5 Integration of Equations of motion
5.1 Motion in One dimension
It has 1 degree of freedom.

The generalized coordinate is q.velocity is q̇.

If Cartesian q = x.

The Lagrangian , L = 12 mẋ2 − U (x)

The first integral of motion;

1 2
E= mq̇ + U (q)
2
In Cartesian;
q
2(E−U (x)) dx 2(E−U (x))
ẋ2 = m =⇒ dt = m
r Z
m dx
=⇒ t = p + constant.
2 E − U (x)
• Kinetic Energy is positive.

• Motion can only take place only id U (x) < E

• The points for which U (x) = E are called turning points where velocity
=0.
• Region bounded by two turning points is called finite.

A finite motion in 1-D is called oscillatory motion with period ;

√ Zx2
dx
T = 2m p
E − U (x)
x1

5.2 Central Force

Two bodies move under the influence of a mutual central force.

The central force problem becomes simplified by separating bodies to center

of mass and particles revolving around it.

m1 , m2 having position r1 , r2

Take R = rcm , r = r1 − r2

17
 |˙ 2 − U (r)
L = 12 M ⃗r|

=⇒ L = 21 M (ṙ2 + r2 θ̇2 ) − U (r)

=⇒ M (r̈ − rθ̇2 ) = − ∂U
∂r = F (r)

and
=⇒ M r2 θ̇ =constant= l

dA 1 l
= r2 θ̇ = = constant
dt 2 2M

5.2.1 Turing points in central force

ṙ = 0 , the system may or may not be in rest.

l2
E − U (r) − 2µr 2 =0

Let rmax and rmin satisfy these equations.

The motion is bounded and closed in a path.(because velocity non zero at turn-
ing point)

There are only two types of central field in which all finite motions
takes place in a closed orbit , PE of those fields vary as: 1r and r2 .

5.2.2 Effective potential

We prefer to split energy energy into radial kinetic energy and effective potential
energy to make problem analogous to 1-D motion.

l2
Uef f = − U (r)
2mr2

Solving equation:

s
l2
 
dr 2
= E − U (r) −
dt M mr2

Where E is the total Energy , l is the angular momentum.

18
For finding r ans function of theta:

∂2 1 M r2
 
1
2
+ = − 2 F (r) (3)
∂θ r r l

Where F (r) = − ∂U
∂r

5.3 Kepler’s Problem (1/r potential)

l2 k
Uef f = −
2mr2 r

19
In the above figure for different value of E we can predict the trajectory

• E= 0 : Parabolic

• E3 < E < 0: Elliptical

2
• E = E3 : Circular , Uef f −min = − M2lk2

• E < E3 : No motion
Solving for r and θ using (3) :

α
= 1 + ε cos θ (4)
r

where ; r
l2 2El2
α= ε= 1+
Mk M k2

Eccentricity Energy Trajectory

ε>1 E>0 Hyperbola
ε=1 E=0 Parabola
1>ε>0 0 > E > Uef f min Elipse
ε=0 E = Uef f min Circle

For planetary motion the orbits are elliptical, with major axis 2a and minor axis
2b.

α α
rmin = 1+ε
; rmax = 1−ε

Period of planetary motion:

4π 2 M 3
T2 = a
k

20
5.4 The Laplace-Runge-Lenz(LRL) Vector
For any central force , newtons second law motion is :

˙ = F (r) ⃗r
⃗p
r

˙ with angular momentum ⃗l:

The cross product of⃗p

˙ × ⃗l
⃗p
˙
= F (r) ⃗rr × m(⃗r ×⃗r)
mF (r)
= r (⃗ r(⃗r · ⃗ṙ) − (r2⃗ṙ))

Using ⃗r · ⃗ṙ = 12 dt
d
= rṙ and F (r) = − rk2 and with a
bit of manipulation:

 
d ˙  d ⃗r
⃗p × ⃗l = −mF (r)r2
dt dt r

 
d ˙ ⃗
 d mk⃗r
=⇒ ⃗p × l =
dt dt r

˙ × ⃗l − mk⃗r = constant
⃗ =⃗p
=⇒ A
r
⃗ is LRL vector.
A

⃗ · ⃗l = 0
A

Now we can derive orbit equation in another way using LRL vector:

p × ⃗l) − mkr
⃗ · ⃗r = ⃗r · (⃗
A

21
 ⃗ · ⃗r = ⃗l · (⃗r × p⃗) − mkr
=⇒ A

⃗ · ⃗r = ⃗l · ⃗l − mkr
=⇒ A

=⇒ Ar cos θ = l2 − mkr

 
l mk A
= 2 1+ cos θ A = mke
r l mk

LRL vetror is part of 7 constants of motion.(Energy , 3 momentums , 3 LRL)

5.4.1 Stable circular orbits:

∂Uef f ∂ 2 Uef f
=0 >0
∂r ∂r2

F ′ (ρ) 3
+ >0
F (ρ) ρ

6 Scattering
sin θ
tan ψ =
cos θ + m 1
m2

ψ is in lab frame.

θ in COM frame.

22
6.1 Scattering Cross-section

Angular momentum ; p
l = m1 u1 b = b 2m1 T0
Kinetic Energy;
1
T0 = m1 u21
2

But in experiment we can’t control a single atomic particle , so we do with

a beam of particle.

Let the beam intensity = I= number of particles crossing a unit area nor-
mal to the beam in unit time.

number of particles scattered into a solid angle dΩ

σ(Ω)dΩ = , dΩ = 2π sin θdθ
Incident intensity

23
In the above figure case;

I2πbdb = Iσ(θ)2π sin θ|dθ|

 
b  db 
=⇒ σ(θ) =  
sin θ  dθ 

Due to symmetry of central force ; α = β.

24