1 Classical Mechanics
1 Classical Mechanics
1.1 Mechanics
What is mechanics? Mechanics concerns with motion of objects, which includes velocity and acceleration,
and the cause of motion which includes force and energy. As the object moves, it does not change their
properties. This assumption is called rigid body and idealized particles. Mass and moment of inertia are the
underlying properties that lead to dierent motions for dierent objects.
1.2 Classical vs Modern Mechanics
Modern in Physics means 20th century. It involves Quantum Mechanics and Relativity. Classical Mechanics
is also termed prequantum mechanics. This includes special relativity and E&M.
Between the 17th and the 19th century, there is the Newtonian Mechanics, which is a human-scale approxi-
mation.
1.3 Newtonian Mechanics
In principle, Newtons Equation of Motion predicted the motion of any objects from the force. But there
are still many problems unsolved. For example, the 3-body problem is impossible to solve. That is, forces
may not even been known when there is more than 2 objects attracting each other by gravity. This leads to
the quest for more powerful mathematics.
1.4 Generalizing Equation of Motion
Newtons Equation deals with the objects position. The goal is to solve for x(t), y(t), and z(t) precisely.
So, if there are N objects, the goal would be to solve for 3N coordinates. Once those coordinate is know for
all time, acceleration, velocity can all be found.
But there are innite other ways to describe motion. For example, the motion of a pendulum could be better
described by the length of the pendulum and the angle of vibration. Because of the physical constraint of
the system, some coordinates could be constrained and thus not all the 3N coordinates of N objects are free.
The question is, how we can derive equation of motion for generalized coordinates.This was done by Lagrange.
1.5 Lagrangian formulation
Newtons Equation is all about force, which is the cause of motion. Start with
F =
F(x, t) for all particles,
we could solve for the 3N functions for the 3N coordinates with the initial state of each particle (that is, the
initial position and initial velocities).
Lagrangian formulation forgets about the force and introduce something else:
Lagrangian: L = L(q, q, t), where q is the generalized coordinate and q is the time derivative.
Lagranges Equation:
d
dt
L
q
L
q
= 0
Everything about this system is embodied in a scalar function L that saties the Lagranges Equation. It
does not depend on a particular coordinate system. The coordinate system can be anything and could be
switched easily.
In words, the partial derivative of the Lagrangian with respect to the coordinate q is equal to the time
derivative of the partial derivative of the lagrangian with respect to the time derivative of the coordinate.
1
1.6 Hamiltons Principle
2
1
Ldt = 0
Hamiltons Principle derives Lagranges Equation from a simple rule:
The time integral of L is stationary for the path taken by an actual physical system.
This statement seems rather weird because it relate path, time and Lagrangian in a way that doesnt appeal
the human intuition.
Newtons Laws were found by induction. That is from conclusion of experimental results. If one day,
it is violated, the Newtonian Mechanics is no longer valid. Deriving the laws from a principle would mean
knowing the origin. Indeed, the laws are eventually connected to Feynmans path integral.
1.7 Hamiltonian Formulation
q =
H
p
, p =
H
q
where (p, q) are canonical variables and H is a function called Hamiltonian.
Hamiltonian Formulation is more exible than Lagrangian formulation. The canonical variables are similar
to position and momentum. (Arent p=mv?) Treating positionand momentum as independent allows wider
range of variable transformations. This also has eerie similarity to what QM does with the uncertainty
principle.
1.8 Success out of failure
The quest for tools that solve the 3-body problem failed. But the byproduct (Lagrangians and Hamiltonians)
turned out to be the cornerstones of Quantum Mechanics. The development of QM was guided by analogies to
Lagrangian and Hamiltonian formulations. Indeed, many pioneers of QM grew up with Classical Mechanics.
One could say that Classical Mechanics is the missing link between Newton and Schrodinger.
2
2 Elementary Principles
2.1 Single Particle
Particle is an object with insignicant size. Examples are electron in a cathode ray tube, baseball thrown by
a pitcher, the Earth orbiting the Sun. A particle has mass m, and location r. Velocity is the time derivative
of position and linear momentum is the product between mass and velocity. The Newtons Second Law
relate the time derivative of momentum to a quantity called Force.
2.2 Inertial System
The origin O of r is somewhat arbitrary. A choice of the origin is a choice of the reference frame. Inertial
system is a reference frame in which F = p holds. More precisely, Newtons second law states that:
there exist reference frames in which the time derivative of the linear momentum
equals to the force.
Indeed, there are innitely many of them.
Consider two inertial systems A and B. A particle is at r
A
in A and r
B
in B. The origin of A is at r
B
r
A
in B. According to both inertial system, force is the same. This enforce the condition that the accelation
of the particle according to both object must be the same and thus, the dierence in velocity is a constant.
This lead to the conclusion that:
Two inertial systems are moving relative to each other at a constant velocity.
Equivalence of such systems was pointed out by Galileo, and hence the name Galilean System.
2.3 Angular Momentum
L = r p
Angular momentum of a particle is dened as the cross product between the position vector r of the trajectory
and the linear momentum. Moment of force (torque) is dened as
N = r
F. Using the two denitions and
the Newtons second law, we could deduce that
N =
L
as follows,
L =
r p
=
r p +r
p
But p is parallel to v =
r,
r p = 0
L = r
p
=
N
Like momentum, the denition of angular momentum depends on the denition of the origin. The equation
relating torque and angular momentum are the same for dierent reference frame. However, both the quan-
tity for torque and angular momentum are dierent in dierent reference frame.
3
2.4 Momentum Conservation
From the two conservation theorems, it follows that:
If the total force is zero, linear momentum does not change with time.
If the total torque is zero, angular momentum does not change with time.
These two conservation laws are independent can could not be derived from the other.
2.5 Work by External Force
When a particle moves from point 1 to 2 under force F, work done by the force F is dened by:
W
12
=
2
1
F ds
Alternatively, we could dene the kinetic energy as T
mv
2
2
and then derive W
12
= T
2
T
1
. Thus, in this
case, without any potential energy, work done is equal to the change in the kinetic energy.
2.6 Conservative Force
If W
12
is the same for any possible path from 1 to 2, the force
F is conservative. Equivalently, the contour
integral force with respect to position is 0:
F ds
2.7 Potential Energy
If a force is conservative, it could be expressed in terms of the gradient of the potential energy. Work is then
expressed by the dierence in the potential energy, which is also the dierence between the kinetic energy.
Thus,
If the force is conservative, the total energy T+V is conserved.
That is the energy conservation Theorem. Indeed, energy could be converted to some other form apart from
kinetic and potential (e.g. heat and light or other EM radiation).
2.8 System of Particles
For systems that involve more than 1 particles, we can just index the forces, torques, linear and angular
momentums. For forces and torques, it could be external and internal. Internal forces are from other particles
in the system. Thus, the force on particle i is,
F
i
=
F
ji
+
F
(e)
i
where
F
ji
is the force from particle j and
F
(e)
i
is the force from outside.
Adding up all the forces on all particles,
F
i
=
i,j
i=j
F
ji
+
F
(e)
i
=
i
p
i
The sum of internal forces vanishes if
F
ij
+
F
ji
= 0. According to the weak law of action and reaction,
Forces two particle exert on each other are equal and opposite.
4
compare with the strong law of action and reaction,
Forces two particle exert on each other are equal, opposite and along the line joining
the particles.
If the strong law is true, the weak law is true. If the weak law is true, the strong law may not be true.
2.9 Sum of Particles
How could we sum up the eect of a system of particles and predict the motion of the system as a whole? If
the relative position of the particles stay unchanged, we could describe the motion just as a single particle
by applying the conservation of energy and momentum. Assuming mass is constant with respect to time,
F
i
=
F
(e)
i
=
i
p
i
=
d
2
dt
2
i
m
i
r
i
We could dene the center of mass as,
R =
i
m
i
r
i
M
And thus preserving the Newtons Second law that
M
R =
F
(e)
i
F
(e)
2.10 Total Linear Momentum
The sum of the linear momenta is
P =
i
p
i
=
i
m
i
r
i
= M
R
Taking the derivative,
P = M
R = sum
i
F
(e)
i
F
(e)
Conservation of total linear momentum follows,
If the total external force is zero, the total linear momentum P is conserved, assum-
ing the weak law of action and reaction.
2.11 Total Angular Momentum
The sum of the angular momentum is,
L =
L
i
=
r
i
p
i
And similarly by assuming the weak law of action and reaction,
L =
i
N
(e)
i
= N
(e)
Conservation of total angular momentum follows,
If the total external torque N
(e)
is zero, the total angular momentum L is conserved.
5
2.12 Laws of Action and Reaction
Most forces we know obey strong law of action and reaction. For example, gravitational force, electrostatic
force. However, there are rare exceptions. For example, the Lorenz force felt by moving charges (the magnetic
force). Conservation of linear and angular momenta fails.
If we take into account the EM eld, and think that particles exchange forces with the eld, the eld itself
has linear and angular momenta and thus the conservation laws are restored for the eld as long as the eld
is constant. If the eld changes, the conservation laws would not apply.
2.13 Total Angular Momentum
Dene particle is position from the center of mass as
r
i
= r
i
R
and the velocities as
v
i
= r
, v =
R
And the total angular momentum
L =
i
r
i
p
i
=
i
(r
i
+R) m
i
(v
i
+v)
Since
i
r
i
= 0 and so
i
v
i
= 0
L = RMv
. .. .
L at center of mass
+
i
r
i
m
i
v
i
. .. .
L around center of mass
2.14 Kinetic Energy
The work done by force is
W
12
=
2
1
F
i
ds
i
where Positions 1 and 2 are congurations (i.e. sets of positions)
Similarly,
W
12
= T
2
T
1
where T =
i
1
2
m
i
v
2
i
T can be split into two pieces just like the angular momentum,
T =
i
1
2
m
i
(v +v
i
)(v +v
2
i
)
=
i
1
2
m
i
(v
2
+ v
2
i
+ 2vv
i
)
=
i
1
2
m
i
(v
2
+ v
2
i
)
=
1
2
Mv
2
. .. .
Motion at center of mass
+
i
1
2
m
i
v
2
i
. .. .
Motion around center of mass
6
2.15 Potential Energy
Assume that the external force is conservative. The total work done by the external force is negative of the
sum of potential dierence by the external force.
Assume also conservative internal forces. By the strong law of action and reaction, the potential energy
between 2 particles i and j is a function of the distance between the two particles. It turns out that the sum
of the work done is negative half the sum of the potential dierence. This is because potential dierence is
additive for every distinct pair of particles.
Hence, if all forces are conservative, one can dene total potential energy:
V =
i
V
i
+
1
2
i,j
i=j
V
ij
. .. .
internal potential energy
The internal potential energy of a rigid body is constant because the relative conguration is xed. The
total energy T + V is conserved.
2.16 Constraints
Equation of motion m
i
r = F
i
= F
e
i
+
j
F
i
j assumes the particle can move anywhere in space.
This is not generally true. In fact never true Free space is an idealization. For example, roller coaster is
constrained on a rail; billiard balls is constrained on a pool table. How can we accommodate constraints in
the equation of motion? The answer is: it depends on the type of constraint.
2.17 Holonomic Constraints
Constraints may be expressed by
f(r
1
, r
2
, r
3
, . . . , t) = 0
Particle on the x-y plane has the constraint that z = 0. Rigid body has the constraint that (r
i
r
j
)
2
C
2
ij
= 0.
All other cases are called nonholonomic. For example z 0 and f( r) = 0
2.18 Independent Variables
A holonomic constraint reduces the number of independent variables by 1. For example, if x + y + z = 0,
then we are only able to freely choose 2 variables and the last is constrained and depends on the value chosen
for other variables. The constrain can also be stated in a dierent coordinate system by parameterization.
For example, the constraint that a particle is on a sphere x
2
+ y
2
+ z
2
= c
2
in Cartesian coordinate system
can be stated as = c in spherical coordinate system. The corresponding independent variables are (, ).
This freedom of choice of set of variables is the concept of generalized coordinates.
2.19 Generalized Coordinates
N particles have 3N degrees of freedom. Introducing k holonomic constrains reduces it to 3N k. Thus,
using the generalized coordinates q
1
, q
2
, . . . , q
3Nk
r
i
= r
i
(q
1
, q
2
, . . . , t)
For example, for the constraint that a particle lies on the sphere with radius 1 centered at the origin,
r
i
= r
i
(, )
7
Transformation from Cartesian coordinate system is as follows:
x = sincos
y = sinsin
z = cos
How can we transform the equation of motion to the generalized coordinates? That requires Lagranges
Equations.
Constraint is an idealized classical concept. Because of uncertainty, nothing is perfectly constrainted in
Quantum Mechanics.
2.20 Constraints and Force
A holonomic constraint is an innitely strong force, or equivalently, an innitely high potential wall. Reality
is always smoother. For example, electron of a hydrogen atom feels strong radial (binding) force, which it
can move freely around the nucleus. This binding force makes the r direction special (constrainted) while
the circular direction is still free (unconstrained).
2.21 Force and Symmetry
Without forces, all coordinate systems are equal and x-y-z system is the simplest. Forces break the symmetry
and thus some coordinate system works better than others. Generalized coordinates oer natural way of
handling systems with such forces. Constraints are extreme cases where the force is innitely strong.
2.22 Lagranges Equations
d
dt
L
q
L
q
= 0 where L(q, q, t) T V
The Lagrangian embodies everything about a constrained system and expresses it in terms of the generalized
coordinates q
j
, their time-derivatives, and time t.
Example 1 Consider a particle moving on the x-axis
x = x(t), y = 0, z = 0
The kinetic and potential energies:
T =
m
2
x
2
, V = V (x) L =
m
2
x
2
V (x)
Lagranges Eqn
m
2
x
2
V (x)
This is equivalent to Newtons equation
F
x
=
V
x
8
2.23 Virtual Displacement
Consider a system with constraints.
r
1
= r
1
(q
1
, q
2
, . . . , q
n
, t)
r
1
= r
1
(q
1
, q
2
, . . . , q
n
, t)
.
.
.
r
N
= r
N
(q
1
, q
2
, . . . , q
n
, t)
Moving all particles slight by r
i
r
i
= r
i
+ r
i
r
i
=
j
r
i
q
j
q
j
2.24 DAlemberts Principle
From Newtons Equation of Motion, F
i
= p
i
. Part of the force is applied and part of it is due to constraints:
F
i
= F
(a)
i
....
applied
+ f
i
....
constraint
Applied force is
F
(a)
t
= F
(a)
t
(r
1
, r
2
, . . . , r
i
, . . . , r
N
, t)
Constrainted force f
i
usually does no work because the constraint force constrains the motion along the
direction so that the movement is perpendicular to the force f
i
r
i
= 0. The exception is: frictional force.
F
i
p
i
= 0
F
(a)
i
+f
i
p
i
= 0
Multiply by r
i
i
(F
(a)
i
p
i
)r
i
= 0 ,which is called DAlemberts Principle
dropping the exponent (a) and use F
i
for the applied force instead,
i
(F
i
p
i
)r
i
= 0
i
F
i
r
i
=
i
p
i
r
i
i
F
i
j
r
i
q
j
q
j
=
i
p
i
j
r
i
q
j
q
j
i
F
i
r
i
q
j
q
j
=
i,j
p
i
r
i
q
j
q
j
j
Q
j
q
j
=
i,j
mr
i
r
i
q
j
q
j
, where Q
j
i
F
i
r
i
q
j
9
On the R.H.S.,
r
i
r
i
q
j
=
d
dt
q
j
v
2
i
2
q
j
v
2
i
2
mr
i
r
i
q
j
=
d
dt
T
q
j
T
q
j
Combining it with the L.H.S.
d
dt
T
q
j
T
q
j
Q
j
q
j
= 0
Since q
j
are independent,
d
dt
T
q
j
T
q
j
= Q
j
Assume forces are conservative,
Q
j
i
F
i
r
i
q
j
=
i
V
r
i
q
j
=
V
q
i
Substituting the expression for Q
j
,
d
dt
T
q
j
(T V )
q
j
= 0
Assume V is independent of q
j
,
d
dt
(T V )
q
j
(T V )
q
j
= 0
d
dt
L
q
j
L
q
j
= 0 where L = T(q
j
, q
j
, t) V (q
j
, t)
3 Lagranges Equation
In deriving the Lagranges Equation, we made the following assumptions:
Holonomic Constraints r
i
= r
i
(q
1
, q
2
, . . . , q
n
, t)
Constraint forces do no work f
i
r
i
= 0 and forget frictions
Applied forces are conservative F
i
=
i
V
Potential V does not depend on q
j
V
qj
= 0
Example 2 Time-Dependent
Transformation functions may depend on t. That is, the generalized coordinate moves.
Consider a particle with mass m on a rail connected by a spring with constant K and natural length l moving
with angular velocity . Let the extended length of the spring be r. Thus,
10
Trans. fnt:
x = (l + r) cos t
y = (l + r) sin t
Kinetic Energy
m
2
[ r
2
+ (l + r)
2
2
]
Potential Energy
K
2
r
2
Lagrangian
m
2
[ r
2
+ (l + r)
2
2
]
K
2
r
2
Ls Eqn m r m
2
(l + r) + Kr = 0
m r + (K m
2
)
r
m
2
l
Km
2
= 0
The Lagranges equation lead to the following interpretations:
Cases description of the system
K > m
2
harmonic oscillator with =
Km
2
m
center of oscillation is shifted by
m
2
l
Km
2
K < m
2 K
2
r
2
K = m
2
moves away exponentially
3.1 Arbitrarity of Lagrangian
Lagrangian is not unique for a given system. For example, if a Lagrangian L describes a system, then:
L
= L +
dF(q, t)
dt
also works for the system
Lemma 1
d
dt
dF
dt
dF
dt
= 0
Proof 1
dF
dt
=
F
q
q +
F
t
L.H.S. =
d
dt
dF
dt
dF
dt
=
d
dt
F
q
q +
F
t
dF
dt
=
d
dt
F
q
dF
dt
=
d
dt
F
q
dF
dt
= 0
= R.H.S. (proved)
11