Ma432 Classical Field Theory

Notes by Chris Blair

These notes cover a lot of the 2008-2009 Ma432 Classical Field Theory course given by Dr Nigel
Buttimore (replaced by Ma3431 Classical Field Theory and Ma3432 Classical Electrodynamics,
the former corresponding to at least the first four sections of these notes). The emphasis is
mostly on the Lagrangian formulation of classical electrodynamics and the solution of Maxwell’s
equations by Green’s function methods. They are probably slightly suspect, particularly with
regard to indices and brackets (and no doubt contain other more unsettling errors).
I am told that Dr Buttimore has changed his units from those in these notes, so use at your
own discretion.

Contents
1 Simple field theory 3
1.1 Introduction to field theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Field theory as a continuum limit . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Euler-Lagrange equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 Special relativity 6
2.1 Rapidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 Tensor notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3 Covariant field theory 10

3.1 Relativistic free particle action . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2 Relativistic interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.3 Electromagnetic field tensor . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.4 Maxwell’s equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.5 Four-current and charge conservation . . . . . . . . . . . . . . . . . . . . . . 15

4 Noether’s theorem 17
4.1 Derivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.3 Stress-energy tensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.3.1 Stress-energy tensor for electromagnetic field . . . . . . . . . . . . . . 20

5 Solving Maxwell’s equations 22

5.1 Time-independent solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
5.1.1 Green’s function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
5.1.2 Magnetostatic and electrostatic potentials . . . . . . . . . . . . . . . 23
5.2 Time-dependent solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

1
 5.2.1 Green’s function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5.2.2 Lienard-Wiechart potentials . . . . . . . . . . . . . . . . . . . . . . . 26
5.2.3 Electromagnetic fields from Lienard-Wiechart potentials: method one 28
5.2.4 Electromagnetic fields from Lienard-Wiechart potentials: method two 29
5.2.5 Properties of the electromagnetic fields due to a moving charge . . . . 30
5.2.6 Local form of the electromagnetic fields due to a moving charge . . . 31

6 Power radiated by accelerating charge 33

7 Bibliography 34
Notes for Classical Field Theory Section 1: Simple field theory

1 Simple field theory

1.1 Introduction to field theory

You are probably already familiar with the notion of electric and magnetic fields. Loosely
speaking, a field in a physics is a physical quantity defined at every point of space and
time, which can be valued as a single number (scalar field), a vector (vector field, such as
electromagnetism and gravity), or as a tensor.
We will restrict ourselves to the study of the electric and magnetic fields, and do so in a
unified manner consistent with special relativity. The methods we use are based in the La-
grangian approach to classical mechanics. Recall that in classical mechanics the Lagrangian
L was defined as L = T − V where T and V are the kinetic and potential energies R of the
system in question. The action of the system was defined to be the quantity S = Ldt, and
the equations of motion of the system were found from the principle of least action, which
states that the true time evolution of the system is such that the action is an extremum.
The equations of motionR (known as the Euler-Lagrange equations) were thus derived from
the condition δS = δ Ldt = 0.
In studying fields which take on different valuesRat different space points it is convenient
to express the Lagrangian itself as an Rintegral, L = d3 x L, where L is called the Lagrangian
density. The full action is then S = dtd3 x L. Note that when we approach this from the
special relativistic point of view the separate time and space components will be unified into
a single package.

1.2 Field theory as a continuum limit

To begin, let us show how a simple field theory may be derived by taking the contin-
uum limit of a system of N particles on a spring with spring constant k. Let the particles
have equilibrium positions a, 2a, . . . N a and denote the deviation of the ith particle from its
equilibrium by φi . The force on the ith particle is
( +k(φ − φ )
2 1 i=1
Fi = −k(φi − φi−1 ) + k(φi+1 − φi ) 1 < i < N
−k(φN − φN −1 ) i=N
The Lagrangian is
N N
X 1 X 1
L=T −V = mφ̇2i − k(φi+1 − φi )2
i=1
2 i=1
2
and the equations of motion are
mφ̈i = −k(φi − φi−1 ) + k(φi−1 − φi ) 1<i<N
We now take the limit a → 0 while keeping (N − 1)a fixed by letting N → ∞. If we write
x = ai as the position of the ith particle then we can regard φi ≡ φ(x = ai, t), and using the

3
Notes for Classical Field Theory Section 1: Simple field theory

equations of motion in the form

m 1h i
φ̈i = ka 2 (φi−1 − φi ) − (φi − φi−1 )
a a
we can apply the definition of the derivative

∂φ φ ([i + 1]a) − φ(ai)

= lim
∂x a→0 a
i

twice to obtain the equations of motion in the limit a → 0:

∂ 2φ ∂ 2φ
µ =κ 2
∂t ∂x
where κ = lima→0 ka and µ = m a
is the mass density which we keep fixed. We see that our
simple field obeys the wave equation.
If we define
N
X 1 1k
L= aLi Li = µφ̇2i − (φi+1 − φi )2
i=1
2 2a
then in the limit we obtain the Lagrangian density
 2  2
1 ∂φ 1 ∂φ
L= µ − κ
2 ∂t 2 ∂x
R
such that L = dxL.

1.3 Euler-Lagrange equations

A more general Lagrangian density would be of the form L(∂t φi , ∂x φi , φi , t, x). We can
use Hamilton’s Principle of Least Action to find the general form of the equations of motion.
Let us consider the simplest case where the field is one-dimensional and the Lagrangian
density is invariant under time and space translation. Then we have that
Z
δL dx dt = 0

and in full Z  
∂L ∂L ∂L
δ(∂t φ) + δ(∂x φ) + δφ dx dt = 0
∂(∂t φ) ∂(∂x φ) ∂φ
Noting that δ(∂t φ) = ∂t (δφ) and δ(∂t φ) = ∂t (δφ) we rewrite this as
Z      
d ∂L ∂L d ∂L ∂L ∂L
δφ − ∂t δφ + δφ − ∂x δφ + δφ dx dt = 0
dt ∂(∂t φ) ∂(∂t φ) dx ∂(∂x φ) ∂(∂x φ) ∂φ

4
Notes for Classical Field Theory Section 1: Simple field theory

We integrate out the total time and space derivatives and use the fact that the δφ term must
vanish at the endpoints to then obtain
Z  
∂L ∂L ∂L
δS = ∂t + ∂x − δφ dx dt = 0
∂(∂t φ) ∂(∂x φ) ∂φ

hence we obtain the Euler-Lagrange equations for this field:

∂L ∂L ∂L
∂t + ∂x − =0
∂(∂t φ) ∂(∂x φ) ∂φ

For a vector field just replace φ by Ai .

5
Notes for Classical Field Theory Section 2: Special relativity

2 Special relativity
We will now introduce the machinery that allows us to express field theory in a manner
consistent with the theory of special relativity. In particular, we seek to formulate the theory
of fields in a manner that is Lorentz covariant - that is, related from one frame to another
via Lorentz transformations. Note that we do not introduce special relativity systematically
but assume some prior knowledge of the subject. For completeness we note that the two
postulates of special relativity are that the laws of physics take the same form in all inertial
(non-accelerating) reference frames, and that the speed of light c in vacuum is an absolute
constant regardless of frame.

2.1 Rapidity
The basic Lorentz transformations in 1 + 1 dimensions are
 vx  1
t0 = γ t − 2 x0 = γ(x − vt) where γ = q
c 1− v2
c2

for a frame S 0 moving with velocity v with respect to the frame S. We have that
s
1 + vc
ct0 − x0 = (ct − x)
1 − vc

We define the rapidity ζ(v) as  v 

1 1+ c
ζ(v) = ln v
2 1− c
so that
ct0 − x0 = eζ (ct − x)
Note that rapidities add. We can then show that

v = tanh ζ γ = cosh ζ

which allows us to write the Lorentz transformations as

ct0 = ct cosh ζ − x sinh ζ x0 = x cosh ζ − ct sinh ζ

In the full 1 + 3 dimensions we can write this transformation in matrix form as

 
cosh ζ − sinh ζ 0 0
− sinh ζ cosh ζ 0 0
Λ=  0

0 1 0
0 0 0 1

6
Notes for Classical Field Theory Section 2: Special relativity

called a boost in the x-direction. Note that the most general proper Lorentz transformation
can be written as a product of a 3-rotation to align the new x-axis with the direction of
motion, a boost along the new x-direction with velocity v and a second 3-space rotation.

2.2 Tensor notation

A basic invariant in special relativity is the interval ds separating two (infinitesimally
close) events in four-dimensional space-time: ds2 = c2 dt2 − dx21 − dx22 − dx23 . From this we
get the metric tensor:
g µν = diag (1, −1, −1, −1) = gµν
which is used to raise and lower indices as follows:

xµ = g µα xα xµ = gµα xα

Note that we sum over repeated indices. Upper indices are said to be contravariant, and
lower indices are said to be covariant. Note that (in this metric) raising a time-index has no
effect, x0 = x0 , while raising a space-index changes the sign, xi = −xi . Note also that indices
with Greek letters can take any value in {0, 1, 2, 3} while indices with Roman letters refer to
spatial indices, {1, 2, 3}. Thus in our notation we have xµ = (ct, ~x) and xi = ~x.
A Lorentz transformation relates events x0 in the frame S 0 to events x in the frame S,
and is written as
x0 µ = Λµβ xβ
The metric tensor is invariant under Lorentz transformations

gαβ = gµν Λµα Λνβ

A four-dimensional vector, or four-vector, is written as Aµ and transforms like the coordinates

xµ :
A0µ = Λµν Aν
while a second rank tensor T µν transforms like the product of (components of) two four-
vectors:
T 0µν = Λµα Λνβ T αβ
and similarly for tensors of higher rank.
We form the four-dimensional Kronecker delta by lowering the index of g µν :

gλν g νµ = g µλ = δ µλ = diag (1, 1, 1, 1)

We also have the four-dimensional Levi-Civita symbol, αβγδ which is +1 for even permu-
tations of αβγδ and −1 for odd permutations, with 0123 = 1. Note that for the covariant
form, αβγδ , we have 0123 = −1. Both the Levi-Civita symbol and the Kronecker delta are
invariant under Lorentz transformations.

7
Notes for Classical Field Theory Section 2: Special relativity

We can form a scalar invariant under Lorentz transformations (a Lorentz scalar) by con-
tracting two four vectors
aµ bµ = a0µ b0µ
The invariant time element dτ is given by

c2 dτ 2 = dxµ dxµ

It is related to the usual time element by

dt d d
dτ = ⇒ =γ
γ dτ dt
µ µ
This allows us to define vectors of four-velocity V µ = dx dτ
and four-momentum pµ = m dx dτ
where m is the rest mass of the particle. The zero component of the four-momentum is related
to the energy E = γmc2 by cp0 = E, so we can write the four-momentum as pµ = Ec , p~ .

Note that the contraction of the four-momentum with itself is pµ pµ = m2 c2 .

If this (or indeed the scalar formed by contracting any four-vector with itself) is equal
to zero we say that the four-vector is light-like, if it is greater than zero we say that it is
space-like, if it is less than zero it is timelike.
This can be related to the idea of light-cones:
ct v=c

v<c

Future
v>c

Past

This picture can be understood as follows: events that occur inside the lightcone are
timelike; it is possible to find a Lorentz transformation such that any two events occur at the
same point in space, but at different times. Similarly, events that occur outside the lightcone
are spacelike in that is possible to find a Lorentz transformation to a frame such that any
two events occur at the same point in time, but are separated in space.

8
Notes for Classical Field Theory Section 2: Special relativity

The boundary of the light-cone is given by a line corresponding to the motion of a particle
of velocity c. The motion of a particle with velocity less than c lies within its light-cone.
Events that occur within the past light-cone of a particle can affect the particle in the
present, while events that occur outside it cannot.
Finally we must consider calculus in space-time. We will be integrating over the four-
dimensional volume element

d4 x = dx0 dx1 dx2 dx3 = c dt dx1 dx2 dx3

which is an invariant. Derivatives are denoted by

∂ ∂
∂µ ≡ ∂µ ≡
∂xµ ∂xµ

Note that differentiating with respect to a lower index gives an upper index, while differen-
tiating with respect to an upper index gives a lower index, so for instance

∂xµ ∂xµ ∂xρ ∂(∂α xβ )

= δµν = g µρ = g µρ δρν = g µν = δ µα δ νβ
∂xν ∂xν ∂xν ∂(∂µ xν )

9
Notes for Classical Field Theory Section 3: Covariant field theory

3 Covariant field theory

We now seek to formulate field theories in a special relativistic context. We will be seeking
actions Z Z Z
3
S = L d x dt = L dt = Lγ dτ

which are relativistically invariant. Recall that the first postulate of special relativity is that
the laws of physics are the same in all (inertial) reference frames - these laws take the form
of the equations of motion derived from the condition δS = 0, hence S must be Lorentz
invariant. As dτ is an invariant scalar we see that we must have Lγ also a Lorentz scalar.
The simplest way to achieve this is to contract the available four-vectors.
Note that the Euler-Lagrange equations of motion for a field Aµ are
 
∂L ∂L
∂µ − =0
∂(∂µ Aν ) ∂Aν

3.1 Relativistic free particle action

For a free particle, the only scalar which preserves translational invariance is pµ pµ = (mc)2 ,
suggesting a Lagrangian of the form L = Cpµ pµ = Cm2 c2 , where C is some constant. Let us
look at the non-relativistic limit of this Lagrangian. We have
Z Z
S = L dt = Lγ dτ

We want the non-relativistic limit of L to agree with the Lagrangian for a non-relativistic
free particle, L = 12 m~v 2 . Consider the Taylor expansion of mγ −1 :

~v 2 m ~v 2
   
−1 1
mγ = m 1 − 2 = m − 2
+O 4
c 2 c c
1
⇒ −mc2 γ −1 = −mc2 + m~v 2
2
2
As the constant term −mc is unimportant, we see that we can take our Lagrangian to be
mc2
L=−
γ
hence we have action Z Z
dt
2
S = −mc = −mc2 dτ
γ
or Z
1
S=− pµ pµ dτ
m

10
Notes for Classical Field Theory Section 3: Covariant field theory

3.2 Relativistic interactions

We now let our particle be acted on by some field with potential Aµ (xν ). Possible scalars
include Aµ pµ and Aµ Aµ . Again we want to choose the Lagrangian so that the non-relativistic
limit gives us the interaction Lagrangian for a particle in the presence of an electric field (in
the non-relativistic limit the particle’s velocity goes to zero, and so it will not interact with
the magnetic field). This is L = −eΦ. This suggests we look at eAµ pµ = ep0 A0 − e~v · A.~ Now,
E 0 µ
p0 = c = mγc, and we identify A with Φ. Then in the limit ~v → 0 we have epµ A = mcγeΦ.
Thus we take our interaction term to be
e
Lint = − Aµ pµ
mcγ
The total action is now
Z  Z 
1 e  e 
S=− pµ + Aµ pµ dτ = − pµ + Aµ dxµ
m c c
µ
using pµ = m dxdτ
. The dynamics of the system can then be found by varying this action.
First, let us note that
dxµ µ
δ(pµ pµ ) = δpµ pµ + pµ δpµ = pµ δpµ + pµ δpµ = 2pµ δpµ = 2m δp
dτ
but also
δ(pµ pµ ) = δ(m2 c2 ) = 0
and hence if m 6= 0 we have
dxµ µ
δp = 0 ⇒ dxµ δpµ = 0
dτ
Now let us compute the variation:
Z 
e 
δS = −δ pµ + Aµ dxµ
c
Z  Z 
e  µ e 
=− pµ + Aµ δ(dx ) − δpµ + δAµ dxµ
c c
Z  Z
e  e ∂A µ
=− pµ + Aµ d(δxµ ) − δxν dxµ
c c ∂xν
where we have used the above result to eliminate the δpµ term. We now use
Z  Z   Z 
e  e e 
pµ + Aµ d(δx ) = d pµ + Aµ (δx ) − d pµ + Aµ δxµ
µ µ
c c c
Z Z 
d  e  e 
= pµ + Aµ (δxµ ) dτ − d pµ + Aµ δxµ
dτ c c

11
Notes for Classical Field Theory Section 3: Covariant field theory

and the fact that the variation δxµ vanishes at the end points to obtain
Z  Z
h e  µ iτ2 e 
µ e ∂Aµ ν µ
δS = − pµ + Aµ δx + dpµ + dAµ δx − δx dx
c τ1 c c ∂xν
| {z }
=0
Z   Z
e ∂Aµ ν µ e ∂Aµ ν µ
= dpµ + ν
dx δx − δx dx
c ∂x c ∂xν
e ∂Aµ dxν µ ∂Aµ dxµ ν
Z    
dpµ
= dτ + δx − δx dτ
dτ c ∂xν dτ ∂xν dτ

Switching the dummy variables µ and ν in the second Aµ term we have:

e ∂Aµ dxν ∂Aν dxν

Z   
µ dpµ
δS = dτ δx + − =0
dτ c ∂xν dτ ∂xµ dτ

Hence we find equations of motion

dxν
 
dpµ e ∂Aν ∂Aµ
= µ
−
dτ c ∂x ∂xν dτ

3.3 Electromagnetic field tensor

The electric and magnetic fields can be expressed in terms of the 4-potential Aµ as

~ = − 1 ∂t A
E ~ − ∇Φ
~ (A0 ≡ Φ)
c
~ =∇
B ~ ×A
~

Note in passing that E~ has odd parity (i.e. transforms as E

~ 7→ −E ~ under space reversal,
~r 7→ −~r) and is even under time reversal, while B~ has even parity and is odd under time
reversal.
We now introduce the electromagnetic field tensor

Fµν = ∂µ Aν − ∂ν Aµ

where we identify
F i0 = E i F ij = −εijk B k
as we have
E i = −∂0 Ai − ∂i A0 = −∂ 0 Ai + ∂ i A0 = F i0
i j
−εijk B k = −εijk εklm ∂l Am = −(δli δm
j
− δm δl )∂l Am = −∂i Aj + ∂j Ai = ∂ i Aj − ∂ j Ai = F ij

12
Notes for Classical Field Theory Section 3: Covariant field theory

Explicitly, the contravariant and covariant forms of the tensor are:

   
0 −E 1 −E 2 −E 3 0 E1 E2 E3
1 3 2 
 E 0 −B B   −E 1 0 −B 3 B 2 
F µν = 
E 2 B 3 F µν =  
0 −B 1  −E 2 B3 0 −B 1 
3 2 1
E −B B 0 −E 3 −B 2 B 1 0

The equations of motion derived in the last section can then be written
dpµ e dxν
= Fµν
dτ c dτ
To write these in three-dimensional form we first set µ = 0 and sum over ν (noting that
d
dτ
= γ dtd )
dp0 e dxi e dE ~ · ~v
= F0i = F i0 v i ⇒ = eE
dt c dt c dt
using p0 = Ec and the fact that raising a spatial index changes the sign, as well as the
antisymmetry of Fµν . For µ = i

dpi e dx0 e dxj dpi e

= Fi0 + Fij ⇒− = −eF i0 − εijk v k B k
dt c dt c dt dt c
d~p ~ + e ~v × B
~
⇒ = eE
dt c
Note that Fµν is invariant under an important class of transformations known as gauge
transformations. A gauge transformation of the electromagnetic four-potential is a trans-
formation of the form Aµ 7→ Aµ + ∂µ φ for some scalar field φ, and under this we have

Fµν 7→ ∂µ Aν − ∂ν Aµ + ∂µ ∂ν φ − ∂ν ∂µ φ = ∂µ Aν − ∂ν Aµ = Fµν

The invariance of the electromagnetic field tensor and hence the observable fields allows us to
simplify problems by choosing a particular gauge, i.e. a particular choice of the Aµ satisfying
certain conditions. For example we will later explicitly solve Maxwell’s equations (introduced
in the next section) in Lorenz gauge: ∂µ Aµ = 0.

3.4 Maxwell’s equations

In the moonlight opposite me were three young women, ladies
by their dress and manner. I thought at the time that I must be
dreaming when I saw them, they threw no shadow on the floor.
Bram Stoker, Dracula

The simplest choice of a Lagrangian density for the electromagnetic field tensor is L =
CFµν F µν where C is some constant. We will now find the equations of motion satisfied by

13
Notes for Classical Field Theory Section 3: Covariant field theory

the field using the Euler-Lagrange equations. We have

∂(Fαβ F αβ ) ∂Fαβ
= 2F αβ
∂(∂µ Aν ) ∂(∂µ Aν )
 
αβ ∂(∂α Aβ ) ∂(∂β Aα )
= 2F −
∂(∂µ Aν ) ∂(∂µ Aν )
= 2F αβ δαµ δβν − δβµ δαν

= 2F µν − 2F νµ
= 4F µν
∂L
and as ∂Aν
= 0 we have the equation of motion for a free field

∂µ F µν = 0

~ (see the next section) then we

If we introduce a source of charge and 3-current J µ = (cρ, J)
have an interaction term
1
Lint = − Aµ J µ
c
and we obtain
1 ν
∂µ F µν = − J
4Cc
1
and choosing C = − 16π (i.e. Gaussian cgs units) gives

4π ν
∂µ F µν = J
c
In three dimensions this becomes
~ ·E
∇ ~ = 4πρ

and
1∂ ~ ~ ~ = 4π J~
− E+∇×B
c ∂t c
These are two of Maxwell’s equations. The other two may be expressed using the dual
tensor F̃ µν defined by
1
F̃ µν = µνρσ Fσρ = µνρσ ∂ρ Aσ
2
The components of the dual tensor may be seen to be

F̃ i0 = B i F̃ ij = ijk E k
~ with B
in words, replace E ~ and B
~ with −E.
~ Now consider

∂µ F̃ µν = µνρσ ∂µ ∂ρ Aσ

14
Notes for Classical Field Theory Section 3: Covariant field theory

This is the contraction of an antisymmetric tensor with a symmetric tensor, and so is equal
to zero. Hence we have the equation

∂µ F̃ µν = 0

By comparing with the above Maxwell’s equations in the case that J µ = 0 and interchang-
ing the electric and magnetic fields in the manner mentioned above, we find the other two
Maxwell’s equations:
~ ·B
∇ ~ =0
and
1∂ ~ ~ ~ =0
B+∇×E
c ∂t
Note that the equation ∂µ F̃ µν = 0 can also be written in the form

∂λ Fµν + ∂µ Fνλ + ∂ν Fλµ = 0

The various contractions arising from the field tensor and its dual are
 
Fµν F µν = 2 B~ ·B~ −E ~ ·E~ Fµν F̃ µν = −4E~ ·B~ ~ ·E
F̃µν F̃ µν = 2(E ~ −B
~ · B)
~

3.5 Four-current and charge conservation

The four-current density is given by
dxµ 3
J µ (t, ~x) = e δ (~x − ~xe (t))
dt
where ~xe (t) is the path of a particle of charge e generating the field Aµ . The charge q and
current e~v are then given by
Z Z
q = ρ(t, ~x)d x3 ~ ~x)d3 x
e~v = J(t,

Consider the gauge transformation Aµ → A0µ = Aµ + ∂µ φ. This gives an interaction action

Z
0 1
S = − 2 (J µ Aµ + J µ ∂µ φ) d4 x
c
We vary this with respect to φ:
Z Z
0 1 1
δS = − 2 δ(J ∂µ φ)d x = − 2 (−∂µ J µ δφ + ∂µ [J µ δφ]) d4 x
µ 4
c c
The second term vanishes on the boundary, and we are left with
Z
0 1
δS = 2 ∂µ J µ δφd4 x = 0 ⇒ ∂µ J µ = 0
c

15
Notes for Classical Field Theory Section 3: Covariant field theory

so charge is conserved. This equation can also be written

∂ρ ~ ~
+∇·J =0
∂t
Let us integrate this over a surface Σ,
Z Z
∂ρ 3 ~ · Jd
~ 3x = 0
d x+ ∇
Σ ∂t Σ
Z Z
∂
⇒ 3
ρd x + J~ · dAd
~ 3x = 0
∂t Σ ∂Σ
or Z
∂q
+ J~ · dA ~=0
∂t ∂Σ
We see that invariance under gauge transformations leads to charge conservation. In fact
there is a close relationship between certain types of transformational invariance and conser-
vation laws, which we treat in detail in the next section.

16
Notes for Classical Field Theory Section 4: Noether’s theorem

4 Noether’s theorem
Noether’s theorem states for every continuous symmetry there is a conserved quantity.

4.1 Derivation
Let us suppose we have a Lagrangian density L(φ, ∂µ φ, ε) invariant under a transformation

φ → φ∗ = φ + δφ xµ → x∗µ = xµ + ∆xµ

parametrised by a small quantity ε such that

φ∗ (ε = 0) = φ x∗µ (ε = 0) = xµ

Now the change in xµ means that the volume we integrate over will change; that is, we have
Z Z
∗ ∗µ 4 ∗
cδS = L(φ , x )d x − L(φ, xµ )d4 x
Σ0 Σ
Z Z
∗ µ µ 4
⇒ cδS = [L(φ , x ) − L(φ, x )]d x + L∆xµ dΣµ
Σ ∂Σ
µ
where dΣµ is a surface element (in the xµ -direction) and ∆x dΣµ can be thought of as giving
the change in the boundary ∂Σ caused by the transformation of the coordinates (for more
details see Classical Mechanics by Goldstein, 3rd edition, page 592). We have also switched
dummy variables in the first integral from x∗µ to xµ .
Now the first integrand is just the variation of L with respect to φ, that is
Z   Z "     #
∂L ∂L ∂L ∂L ∂L
δφ + δ(∂µ φ) d4 x = ∂µ δφ + − ∂µ δφ d4 x
∂φ ∂(∂µ φ) ∂(∂µ φ) ∂φ ∂(∂µ φ)
| {z }
=0 (E-L)

while the second can be rewritten using the divergence theorem

Z Z
µ
L∆x dΣµ = ∂µ (L∆xµ ) d4 x
∂Σ Σ

hence Z  
∂L
cδS = ∂µ δφ + L∆x d4 x = 0
µ
Σ ∂(∂µ φ)
Now, consider

δφ = φ∗ (xρ ) − φ(xρ )
= φ∗ (x∗ρ ) − φ(xρ ) − [φ∗ (x∗ρ ) − φ∗ (xρ )]

17
Notes for Classical Field Theory Section 4: Noether’s theorem

We now Taylor expand φ∗ (x∗ρ ) with respect to xρ , obtaining

φ∗ (x∗ρ ) − φ∗ (xρ ) ≈ φ∗ (xρ ) + ∂ν φ∗ (xρ )∆xν − φ∗ (xρ ) = ∂ν φ∗ (xρ )∆xν

or, using φ∗ = φ + δφ

φ∗ (x∗ρ ) − φ∗ (xρ ) ≈ ∂ν φ(xρ )∆xν + ∂ν δφ∆xν

and the term on the right is of second order and so can be neglected.
Similarly we expand φ∗ (xρ ) with respect to ε to find

∂φ∗ ∂φ∗ ∂φ∗

φ∗ (x∗ρ ) − φ(xρ ) ≈ φ∗ (x∗ρ ) +ε − φ(xρ ) = φ(xρ ) + ε − φ(xρ ) = ε
ε=0 ∂ε ∂ε ∂ε
ε=0 ε=0 ε=0
∗µ
and as we also have ∆xµ = x∗µ − xµ ≈ ∂x∂ε |ε=0 ,
" #
∂L ∂φ∗ ∂x∗ν
Z  
∂L µ
cδS = ∂µ ε − ∂ν φ − δν L ε d4 x = 0
Σ ∂(∂µ φ) ∂ ∂(∂µ φ) ∂
=0 =0

and so we can conclude that the current

∂L ∂φ∗
  ∗
µ ∂L ν µν ∂xν
J = − ∂ φ−g L
∂(∂µ φ) ∂ ∂(∂µ φ) ∂
=0 =0

is conserved. In the case of a vector field Aλ this becomes

∂L ∂A∗λ
  ∗
µ ∂L ν λ µν ∂xν
J = λ
− λ
∂ A −g L
∂(∂µ A ) ∂ ∂(∂µ A ) ∂
=0 =0

Note that this derivation assumes ε has no indices (i.e. is a scalar). To be more precise we
should take into account the possibility that ε may be a vector or even matrix quantity, and
write it as εα where α stands for any possible index. In this case we should go back to the
second from last line of the proof and extract the conserved quantity
!
∗
  ∗
∂L ∂φ ∂L ∂x ν
J µα εα = − ∂ ν φ − g µν L εα
∂(∂µ φ) ∂εα ∂(∂µ φ) ∂α
εα =0 εα =0

where there is now one conserved current J µα for each εα (see example iv) below).

4.2 Examples
Let us assume that the field does not change, i.e. φ∗ (x) = φ(x).

18
Notes for Classical Field Theory Section 4: Noether’s theorem

i) Let L be invariant under time translation, x∗0 = x0 + ε, x∗i = xi , then

∂L 0
J0 = ∂ φ−L≡H
∂(∂0 φ)

is conserved (this is the Hamiltonian for the field).

ii) Suppose L invariant under x∗µ = xµ + ε, then

∂L ν
∂ φ − g µν L
∂(∂µ φ)

is conserved (this is the energy-momentum density for the field).

iii) Suppose L is symmetric under rotations about the x3 axis, that is,

x∗1 = x1 cos ε + x2 sin ε x∗2 = −x1 sin ε + x2 cos ε x∗3 = x3 x∗0 = x0

then the angular momentum density of the field is conserved,

∂L
(x1 ∂2 φ − x2 ∂1 φ)
∂(∂3 φ)

iv) Suppose that L is completely rotationally invariant in the space dimensions. An infinites-
imal rotation can be written as
xi 7→ xi + εij xj
where εij = −εji is a three by three skew-symmetric real matrix, and so an element of so(3)
(i.e. a generator of rotations). We then have that
 
∗
∂L l ∂x
−J ijk εjk =  ∂ φ − g il L l  εjk
∂(∂i φ) ∂jk
εjk =0

is conserved. Now,
∂x∗l ∂ εln xn
= = δlj δnk xn − δnj δlk xn = δlj xk − δlk xj
∂jk ∂jk

giving  
ijk ∂L  j k

ij ik
−J εjk = ∂ φ xk − ∂ φ xj − g Lxk + g Lxj εjk
∂(∂i φ)
so we can pick out our conserved currents
∂L  j 
−J ijk = ∂ φ xk − ∂ k φ xj − g ij Lxk + g ik Lxj
∂(∂i φ)

19
Notes for Classical Field Theory Section 4: Noether’s theorem

For example, consider rotations about the x1 axis. Then we have the following conserved
currents:
∂L  2 
−J 123 = ∂ φ x3 − ∂ 3 φ x2
∂(∂1 φ)
∂L  1 
−J 112 = ∂ φ x2 − ∂ 2 φ x1 − Lx2
∂(∂1 φ)
113 ∂L  1 3

−J = ∂ φ x3 − ∂ φ x1 − Lx3
∂(∂1 φ)
J 132 = −J 123 J 121 = −J 112 J 131 = −J 113 J 111 = J 112 = J 133 = 0
Note that J 132 corresponds to the angular momentum density about the x1 axis.

4.3 Stress-energy tensor

If we assume that our system is invariant under the transformation xν → xν + εν then we
have that the tensor
∂L
T µν = ∂ ν Aλ − g µν L
∂(∂µ Aλ )
is conserved (note we have changed the λ indices). This tensor is known as the stress-energy
tensor.

4.3.1 Stress-energy tensor for electromagnetic field

1
For a free electromagnetic field, L = − 16π Fµν F µν . The stress energy tensor is given by

1 ∂(Fρσ F ρσ ) ν 1 µν
T µν = − ∂ Aλ + g Fρσ F ρσ
16π ∂(∂µ Aλ ) 16π
1 µλ ν 1 µν
⇒ T µν = − F ∂ Aλ + g Fρσ F ρσ
4π 16π
Although conserved, T µν is not gauge invariant (as Aλ appears explicitly) or symmetric.
We can form a tensor with nicer properties as follows: first, write F µλ ∂ ν Aλ = F µλ ∂ ν Aλ =
g µρ Fρλ ∂ ν Aλ giving
1 1 µν
T µν = − g µρ Fρλ ∂ ν Aλ + g Fρσ F ρσ
4π 16π
and then substitute in
∂ ν Aλ = F νλ + ∂ λ Aν
to obtain
1 µρ 1 µρ 1 µν
T µν = − g Fρλ F νλ − g Fρλ ∂ λ Aν + g Fρσ F ρσ
4π 4π 16π
and now define
1 µρ
Θµν = T µν + g Fρλ ∂ λ Aν
4π

20
Notes for Classical Field Theory Section 4: Noether’s theorem

1 µρ 1 µν
⇒ Θµν = g Fρλ F λν + g Fρσ F ρσ
4π 16π
(note interchange of µ and ν to remove the leading minus sign). This tensor is conserved as
1 µρ 1
the difference between it and the stress-energy tensor is 4π g Fρλ ∂ λ Aν = 4π F λµ ∂λ Aν and

∂µ (F λµ ∂λ Aν ) = (∂µ F λµ )∂λ Aν + F λµ ∂µ ∂λ Aν = 0

where the first term vanishes due to the equations of motion and the second term vanishes
as it is the contraction of a symmetric and antisymmetric tensor.
The tensor Θµν is gauge invariant, symmetric, traceless (Θµµ = Θµν gµν = 0) and can be
used to define the angular momentum density M µνσ = Θµν xσ − Θµσ xν .
But my very feelings changed to repulsion and terror when I
saw the whole man slowly emerge from the window and begin
to crawl down the castle wall over the dreadful abyss, face down
with his cloak spreading out around him like great wings.
Bram Stoker, Dracula

21
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

5 Solving Maxwell’s equations

We now wish to solve the equation ∂µ F µν = 4π c
J ν for a given J ν . This equation can be
written
4π ν
∂µ ∂ µ Aν − ∂µ ∂ ν Aµ = J
c
We impose the Lorenz gauge, ∂µ Aµ = 0 so that this becomes
4π ν
∂µ ∂ µ Aν = J
c
This equation will be solved using Green’s function methods. Recall that given some dif-
ferential
R 0equation0 Df (x) = g(x) the RGreen’s function G solves R DG(x) = δ(x); so that
f = dx G(x − x )g(x ) as then Df = dx DG(x − x )g(x ) = dx δ(x − x0 )g(x0 ) = g(x).
0 0 0 0 0

We will apply Fourier transform methods to obtain the Green’s functions we need. The
four-dimensional Fourier transform and its inverse are
Z Z
µ 1 4 −ikν xν ˜ µ ˜ µ 1 ν
f (x ) = 2
d ke f (k ) f (k ) = 2
d4 x eikν x f (xµ )
(2π) (2π)
The Fourier representation of the delta-function is
Z ∞
1
δ(x) = dk e−ikx
2π −∞

5.1 Time-independent solutions

First let us obtain solutions with no time dependence. This means we will be solving the
equation
~ 2 Aν = − 4π J ν
∇
c

5.1.1 Green’s function

We need to find a Green’s function satisfying ∇ ~ 2 G(~x) = δ 3 (~x). In terms of Fourier

transforms, this is
Z Z Z
~ 1 −i~k·~ ~ 1 3 ~ 2 −i~k·~ ~ 1 ~
∇ 2
3/2
3
d ke x
G̃(k) = − 3/2
d kk e x
G̃(k) = 3
d3 k e−ik·~x
(2π) (2π) (2π)
1 1
⇒ G̃(~k) = −
(2π)3/2 ~k 2
giving Z Z ∞ Z
1 1 ~ ~ ~2
G(~x) = − d k e−ik·~x =
3
dα d3 k e−ik·~x−αk
(2π)3 ~k 2 0

22
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations
R∞
as 0
dα e−αx = α1 . Now the k integral can be expressed as a Gaussian:
Z Z “ ”2  π 3/2 ~x2
~ x2
−i~k·~
x−α~k2 −α ~k+ 2α
ix
− ~4α
3
d ke 3
= d ke e = e− 4α
α
so we now have  3/2
Z ∞
1 1 x2
~
G(~x) = − π 3/2 dα e− 4α
(2π)3 0 α
−1/2 −3/2
Let u = α , then −2du = α dα and we have
Z 0
r
1 2
− ~x4 u2 1 1 4π
G(~x) = 3/2 du e = − 3/2
4π ∞ 4π 2 ~x2
using the fact that the Gaussian integral is symmetric, and hence
1
G(~x) = −
4π|~x|

5.1.2 Magnetostatic and electrostatic potentials

For an electrostatic system, we have ∇ ~ 2 A0 = ∇~ 2 Φ = −4πρ(~x) = − 4π P e δ 3 (~x − ~xe ),

c e
where the sum ranges over the different charges in the system, with ~xe signifying the position
of the charge e. Using the Green’s function above we see that
~0 3 0
x − ~xe ) X
Z Z P
3 0 ρ(x ) 3 0 e e δ (~ e
Φ(~x) = d x 0
= d x 0
=
|~x − ~x | |~x − ~x | e
|~x − ~xe |

giving
~ = −∇Φ
~ =
X ~x − ~xe
E e
e
|~x − ~xe |3
Similarly the vector potential is given by
Z ~ 0)
~ x) = 1
A(~ d 3 x0
J(x
c |~x − ~x0 |

5.2 Time-dependent solutions

We will now seek solutions of the full equation
4π ν
∂µ ∂ µ Aν = J
c

23
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

5.2.1 Green’s function

The Green’s function we seek is a solution D(z ρ ) of ∂µ ∂ µ D(z ρ ) = δ 4 (z ρ ) where z ρ =

xρ − x0 ρ . Appealing to Fourier transforms again, we have
Z Z Z
1 −ikν z ν 1 µ −ikν z ν 1 ν
∂µ ∂ µ
2
4
d k D̃(k)e =− 2
4 µ
d k kµ k D̃(k )e = 4
d4 k e−ikν z
(2π) (2π) (2π)
from which
1 1 1 1
D̃(k µ ) = − 2 µ
=− 2
4π kµ k 4π k0 − k 2
where k = |~k|. We now must solve
Z Z Z ∞
1 1 −ikν z ν 1 i~k·~ 1 0
µ
D(z ) = − 4
4
dk 2 2
e =− 4
3
d ke z
dk0 2 2
e−ik0 z
(2π) k0 − k (2π) −∞ k0 − k
Note the sign change in the first exponential when converting to three-dimensions. Now,
to evaluate the k0 integral we use contour integration, treating k0 as a complex number,
~ 0
k0 = Re k0 + iIm k0 . This gives eik·~z = eIm k0 z e−iRe k0 . So that the integral converges we must
choose Im k0 < 0 for z 0 > 0. This condition is imposed as z 0 = x0 − x00 = c(t − t0 ), and so
positive z 0 ensures that contributions to the Green’s function and hence to the potential Aµ
only come from events that occur at times t0 < t, i.e. events in the past. Thus causality is
ensured.
Now, the poles of the integrand are ±k and so lie on the real axis. To avoid them, we
displace our contour by an infinitesimal amount i so that it lies just in the upper-half plane
(formally we should let  → 0 at the end). Our contour of integration Γ then looks like:
Im k0

Re k0
−k k

R
Γ

We than have 0 0 0
R
e−ik0 z e−ik0 z e−ik0 z
I Z Z
dk0 2 = dk0 2 +
Γ k0 − k 2 −R k0 − k 2 semi-circle k02 − k 2
−iϕ
Now on the semicircle we can write k0 = Re = R cos ϕ − iR sin ϕ, so that the integral
becomes an integral over ϕ from 0 to π. So we have
0 Z π
e−ik0 z
Z
0 0 1
2 2
= −i dϕ Reiϕ e−Rz sin ϕ−iRz cos ϕ 2 −2iϕ
semi-circle k0 − k 0 R e − k2

24
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

and sin ϕ ≥ 0 for this range of ϕ, hence the integrand goes to zero as R → ∞, as required.
The value of the k0 integral we are interested in will hence be given by the −2πi times the
sum of the residues inside the contour. The residues are:
0 0
e−ik0 z e−ikz
lim (k0 − k) =
k0 →k (k0 − k)(k0 + k) 2k
0 0
e−ik0 z eikz
lim (k0 + k) =−
k0 →−k (k0 − k)(k0 + k) 2k
hence Z ∞
1 −ik0 z 0 πi  ikz0 −ikz 0

dk0 2 2
e = e − e Θ(z 0 )
−∞ k0 − k k
where the Heaviside function (
01 z0 > 0
Θ(z ) =
0 z0 < 0
is added as the residues only contribute for positive z 0 . So we now have
Θ(z 0 )
Z
z1
 0 
µ 3 i~k·~ ikz −ikz 0
D(z ) = πi d k e e − e
(2π)4 k
∞  Z 2π Z π
Θ(z 0 )
Z
21

ikz 0 −ikz 0
= i dk k e −e dφ dθ sin θ eikz cos θ
16π 3 0 k 0 0

upon switching to polar coordinates and choosing the coordinate frame such that the k 3 axis
makes an angle of θ with ~z, and letting z = |~z|. Integrating over the angles, we get
Z ∞   eikz e−ikz 
µ Θ(z 0 )  0
ikz −ikz 0
D(z ) = i dk k e −e −
8π 2 0 ikz ikz
0 Z ∞
Θ(z ) 
ik(z 0 +z) −ik(z 0 +z) ik(z 0 −z) −ik(z 0 −z)

=− 2 dk e +e −e −e
8π z 0
but if we let k 7→ −k
Z ∞ Z −∞ Z 0
−ik(z 0 +z) ik(z 0 +z) 0 +z)
dk e = d(−k)e = dk eik(z
0 0 −∞

hence
∞
Θ(z 0 )  Θ(z 0 )
Z  0 0
µ
dk eik(z+z ) − eik(z −z) = δ(z 0 − z) − δ(z 0 + z)


D(z ) = − 2
8π z −∞ 4πz
remembering the integral representation of the delta function. Owing to the Heaviside func-
tion only the first delta function will contribute, so
Θ(z 0 ) 0
Dret (z µ ) = δ(z − z)
4πz

25
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

The subscript signifies that this is the retarded Green’s function (that is, the Green’s function
resulting from the contribution of events in the past).
We can put the Green’s function in covariant form using
δ(z0 − z) δ(z0 + z)
δ(zµ z µ ) = δ([z0 − z][z0 + z]) = +
|z0 + z| |z0 − z|
δ(a) δ(b)
as δ(ab) = |b|
+ |a|
. Hence

δ(z0 − z) δ(z0 − z)
Θ(z 0 )δ(zµ z µ ) = Θ(z 0 ) = Θ(z 0 )
|z0 + z| |2z|
and so
Θ(z 0 )
Dret (z µ ) =
δ(zµ z µ )
2π
Recalling that z µ = xµ − x0µ we can state our final results for the Green’s function as
Θ(x0 − x00 )
Dret (xµ − x0µ ) = δ(x0 − x00 − |~x − x~0 |)
4πz
and in covariant form,
Θ(x0 − x00 )
Dret (xµ − x0µ ) = δ([xµ − x0µ ][xµ − x0µ ])
2π
Note that if we had taken z 0 < 0 and closed our contour in the upper half-plane (with poles
displaced upwards) we would have obtained the advanced Green’s function
Θ(x00 − x0 )
Dadv (xµ − x0µ ) = δ(x0 − x00 + |~x − x~0 |)
4πz

5.2.2 Lienard-Wiechart potentials

Of night and light and the half-light.

W.B. Yeats, “He Wishes For The Cloths Of Heaven”
4π ν
We can now work out the potentials Aµ that solve the Maxwell equation ∂µ ∂ µ Aν = c
J .
They are given by Z
4π
µ σ
A (x ) = d4 x0 Dret (x − x0 )J µ (x0σ )
c
where
dx0σ
J µ (x0σ ) = e 0 δ 3 (~x0 − ~x0e (t))
dt
or in covariant form
dx0σ 4 0σ
Z
µ 0σ
J (x ) = ec dτ δ (x − x0σ e (τ ))
dτ

26
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

as integrating over τ and using that

X δ(τ − τi )
δ(f (τ )) =
|f 0 (τi )|
i:f (τi )=0

we have −1
dx0σ dx00

0σ 0
µ
J (x ) = ec 3
δ (~x − ~x0e (t))
dτ τ0 dτ τ0
00 00 0σ 0σ
where x00 − x0e (τ 0 ) = 0. Now, dxdτ |τ 0 = dxdt dτ
dt dt
|τ 0 = c dτ |τ 0 and dxdτ |τ 0 = dxdt dτ
dt
|τ 0 so this reduces
to the local form.
So substituting this in, we have
Θ(x0 − x00 ) dx0σ 4 0σ
Z
4π
µ σ
A (x ) = dτ d4 x0 δ([xσ − x0σ ][xσ − x0σ ])ec δ (x − x0σ e (τ ))
c 2π dτ
Z
= 2e dτ Θ(x0 − x0e )δ([xσ − xeσ (τ )][xσ − xσe (τ )])V µ (τ )
µ
where we have written V µ ≡ dx dτ
e
. Now, we need the roots of the argument of the delta
function:
[xσ − xeσ (τ )][xσ − xσe (τ )] = 0 ⇒ (x0 − xe0 (τ ))2 − |~x − ~xe (τ )|2 = 0
There are two possibilities:
x0 − xe0 (τ ) = ±|~x − ~xe (τ )|
The Heaviside function constrains us to choose the positive option. We see that the unique
root of [xσ − xeσ (τ )][xσ − xσe (τ )] = 0 which ensures causality is given by the retarded time
τ0 :
x0 − xe0 (τ0 ) = |~x − ~xe (τ0 )|
Physically, the retarded time gives the unique time at which the charged particle intersects
the past light-cone of the observation point. Now, as
d d
[xσ − xeσ (τ )][xσ − xσe (τ )] = −2[xσ − xeσ (τ )] xσe (τ )
dτ dτ
We then have that
δ(τ − τ0 )
Θ(x0 − x0e )δ([xσ − xeσ (τ )][xσ − xσe (τ )]) =
2[xσ − xeσ (τ0 )]V σ (τ0 )
so, writing Rσ = xσ − xσe (τ0 ), we have

V µ (τ )
Aµ (xσ ) =
Rσ (τ )V σ (τ )
τ0

Note that the retarded time is in this notation defined by Rσ (τ0 )Rσ (τ0 ) = 0 and that then
~ = |~x − ~xµe (τ0 )|.
R0 = R = |R|

27
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

5.2.3 Electromagnetic fields from Lienard-Wiechart potentials: method one

We wish to evaluate the electromagnetic fields F µν = ∂ µ Aν − ∂ ν Aµ arising from the

motion of a charged particle. Consider the integral expression for the potential
Z
A (x ) = 2e dτ Θ(x0 − x0e )δ([xσ − xeσ (τ )]2 )V ν (τ )
ν σ

We will differentiate this with respect to xµ . First, note that

∂µ Θ(x0 − x0e ) = δ(x0 − x0e )
and this will give a term δ(−|~x − ~xe (τ )]2 ) which only contributes for ~x = ~xe (τ ) and can be
neglected. Thus we have
Z
∂ A = 2e dτ Θ(x0 − x0e )∂ µ δ([xσ − xeσ (τ )]2 )V ν (τ )
µ ν

Let us now write

∂ µ δ([xσ − xeσ (τ )]2 ) = ∂ µ δ(Rσ (τ )Rσ (τ ))
∂
= δ(Rσ (τ )Rσ (τ ))∂ µ (Rσ Rσ )
∂Rσ Rσ
d ∂τ
= δ(Rσ (τ )Rσ (τ )) ∂ µ (Rσ Rσ )
dτ ∂Rσ Rσ
and as Rσ = xσ − xσe (τ ) we have
∂ µ (Rσ Rσ ) = 2Rµ
and
d dτ 1
(Rσ Rσ ) = −2Rσ V σ ⇒ σ
=−
dτ dRσ R 2Rσ V σ
so (using ρ as our dummy variable in the denominator as σ is used in the numerator)
Rµ d
∂ µ δ([xσ − xeσ (τ )]2 ) = − δ(Rσ (τ )Rσ (τ ))
Rρ V ρ dτ
Thus
Rµ V ν d
Z
µ ν
∂ A = −2e dτ Θ(x0 − x0e ) δ(Rσ (τ )Rσ (τ ))
Rρ V ρ dτ
and we can integrate this by parts to obtain
d Rµ V ν
Z  
µ ν 0 0 σ
∂ A = 2e dτ Θ(x − xe )δ(Rσ (τ )R (τ ))
dτ Rρ V ρ
Recalling from the derivation of the potentials that
δ(τ − τ0 )
Θ(x0 − x0e )δ(Rσ (τ )Rσ (τ )) =
2Rσ (τ0 )V σ (τ0 )

28
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

we find
Rµ V ν
 
µ νe d
∂ A =
Rσ V σ dτ Rρ V ρ
with this evaluated at the retarded time. Carrying out the differentiation, with a dot denoting
a derivative with respect to proper time,
!
µ ν µ ν µ ν 
e −V V + R V̇ R V 
∂ µ Aν = − −Vλ V λ + Rλ V̇ λ
Rσ V σ Rρ V ρ (Rρ V ρ )2

and hence
eVρ V ρ  µ ν  e   eRρ V̇ ρ  µ ν 
F µν = R V − R ν µ
V + R µ ν
V̇ − R ν µ
V̇ − R V − R ν µ
V
(Rσ V σ )3 (Rσ V σ )2 (Rσ V σ )3
Note that we can write F µν as a sum of two parts, one a velocity field containing terms
depending on V µ and the other an acceleration or radiative field containing derivatives of the
velocity, V̇ µ :
µν µν
F µν = Fvel + Frad
where
µν eVρ V ρ  µ ν ν µ

Fvel = R V − R V
(Rσ V σ )3
and
µν e 
µ ν ν µ
 eRρ V̇ ρ  µ ν ν µ

Frad = R V̇ − R V̇ − R V − R V
(Rσ V σ )2 (Rσ V σ )3

5.2.4 Electromagnetic fields from Lienard-Wiechart potentials: method two

Consider the covariant form of the equation defining the retarded time:

Rσ (τ0 )Rσ (τ0 ) = 0

This equation defines τ0 as a function of the observation point xµ . To find how τ0 changes
with xµ we differentiate

∂µ (Rσ (τ0 )Rσ (τ0 )) = 0 ⇒ 2Rσ ∂µ Rσ = 0

Now,
d σ
∂µ Rσ = ∂µ xσ − ∂µ xe (τ ) = δµσ − x (τ )∂ µ τ = δµσ − V µ ∂µ τ
dτ e
so at the retarded time
Rσ (δµσ − V µ ∂µ τ0 ) = 0
and hence
Rµ µ Rµ
∂µ τ 0 = ⇒ ∂ τ0 =
Rσ V σ Rσ V σ

29
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

with the expression on the right evaluated at the retarded time.

We will use this to now evaluate the electromagnetic field tensor resulting from the four-
potential
eV ν
Aν (xρ ) = σ
V Rσ
τ0

In what follows we will denote differentiation with respect to proper time by a dot. Note
that our expressions must all be evaluated at the retarded time - so in effect we will work
out ∂ µ Aν treating Rσ and Vσ as functions of τ with the awareness that in reality everything
we do is evaluated at τ0 . This allows us to substitute the above expression for ∂ µ τ0 for ∂ µ τ
as we derive. So we have
∂ µV ν eV ν  µ ρ 
∂ µ Aν = e − σ (∂ V )R ρ + V ρ µ
(∂ Rρ )
Vσ Rσ (V Rσ )2
V̇ ν ∂ µ τ eV ν  ρ µ ρ µ µ

=e − σ V̇ R ρ ∂ τ + V (δ ρ − V ρ ∂ τ )
Vσ Rσ (V Rσ )2
!
V̇ ν Rµ eV ν V̇ ρ Rρ Rµ V ρ
V ρ R µ
=e − +Vµ−
(Vσ Rσ )2 (V σ Rσ )2 V σ Rσ Rσ V σ
Rµ V ν V ρ Vρ eV µ V ν eRµ V̇ ν eRµ V ν V̇ ρ Rρ
=e − + −
(V σ Rσ )3 (V σ Rσ )2 (V σ Rσ )2 (V σ Rσ )3

Hence we again find we can write F µν = ∂ µ Aν − ∂ ν Aµ as a sum of two parts, one involving
terms containing the four-velocity of the charge V µ and the other involving terms involving
the acceleration V̇ µ , that is,
µν µν
F µν = Fvel + Frad
where
µν ec2  µ ν ν µ

Fvel = R V − R V
(V σ Rσ )3
and
µν e 
µ ν ν µ
 eV̇ ρ Rρ  µ ν ν µ

Frad = R V̇ − R V̇ − R V − R V
(V σ Rσ )2 (V σ Rσ )3
where we have used that V ρ Vρ = γ 2 c2 − γ 2 v 2 = γ 2 c2 (1 − v 2 /c2 ) = c2 .

5.2.5 Properties of the electromagnetic fields due to a moving charge

We can immediately work out some important properties of the velocity and radiative
fields. Consider
µν ec2
Rµ Fvel =− σ Rν V µ 6= 0
(V Rσ )3

30
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

where we have used that Rµ Rµ = 0 at the retarded time (remember that all our expressions
for the fields are evaluated at τ0 ), and

µν eRµ Rν V̇ µ eV̇ ρ Rρ Rµ Rν V µ
Rµ Frad =− + =0
(V σ Rσ )2 (V σ Rσ )3

upon making a single fraction and switching some of the dummy variables to agree. Setting
µν
ν in Rµ Facc to zero and j successively we find
µ0
Rµ Frad i0
= 0 ⇒ Ri Frad = 0 ⇒ Ri Erad
i ~ rad = 0
= 0 ⇒ ~n · E

µj 0j ij Ri k ~ = −~n × B~
Rµ Frad = R0 Frad + Ri Frad = −RE j + Ri εijk B k = 0 ⇒ E j = −εjik B ⇒E
R
where ~n is a unit direction in the direction of R.~ Note that this implies ~n, E
~ rad and B
~ rad are
mutually orthogonal and have the same magnitude (in Gaussian/Heaviside-Lorentz units).
µν µν
We can similarly consider the dual tensor, F̃ αβ = αβµν Fµν . For both Fvel and Frad we see
every term in Rα F̃ αβ = αβµν Rα Fµν will contain αβµν Rα Rµ or αβµν Rα Rν and hence contract
to zero. Thus,
Rµ F̃ µν = 0
and hence
~ =0
~n · B ~ = ~n × E
B ~

5.2.6 Local form of the electromagnetic fields due to a moving charge

µν µν
To transform our covariant expressions for Fvel and Frad to our local frame we recall that
~ ~
V = γ(c, ~v ) = γ(c, cβ), where ~v = cB. We need to work out V̇ µ = dτd V µ . Now, dτd = γ dtd , so
µ

we have
d d 1 β~ · α
~ 4~
γc = cγ q = cγ q 3 = cγ β · α
~
dτ dt
1 − β~ 2 1−β ~ 2

d ~
where α
~= dt
β, and
 
d d ~ d d ~

~ 3 ~

γcβ = cγ γβ = cγ β γ + γ β = cγ βγ (β · α α = cγ 4 (β~ · α
~ ) + γ~ ~ )β~ + cγ 2 α
~
dτ dt dt dt

We also have that V σ Rσ = Rγc − γ V~ · R ~ Hence we have

~ = γcR(1 − ~n · β).

i i0 ec2  i 0 0 i

Evel = Fvel = R V − R V
(V σ Rσ )3
ec2  
= γcRi − Rγcβ i
~ 3
γ 3 c3 R3 (1 − ~n · β)

31
Notes for Classical Field Theory Section 5: Solving Maxwell’s equations

so
~
e(~n − β)
~ vel =
E
~ 3
γ 2 R2 (1 − ~n · β) t0

~ vel = ~n × E
B ~ vel
Turning to the radiative fields, we have

i i0 e 
i 0 0 i
 eV̇ ρ Rρ  i 0 0 i

Erad = Frad = R V̇ − R V̇ − σ R V −R V
(V σ Rσ )2 (V Rσ )3
~ = R~n,
so, writing R
~ 3
γ 3 c3 R3 (1 − ~n · β)   h i
~ rad = γcR 1 − ~n · β~ γ 4 cR~nβ~ · α
E ~ − R cγ 4 (β~ · α ~ )β~ + cγ 2 α~
e   
− γ 4 cRβ~ · α ~ − γ 4 cR(β~ · α~ )~n · β~ − γ 2 cR~n · α ~ cγR~n − cγRβ~
"
  α
~

5 2 2
=γ c R 1 − ~n · β~ β~ · α ~ −
~ (~n − β)
γ2
  #
~
n · α
~ 
− β~ · α ~ −
~ (1 − ~n · β) ~n − β~
γ2
"
1 ~ + (~n − β)

~ β~ · α ~
= γ 5 c2 R 2 2 α ~ (1 − ~n · β) ~ (1 − ~n · β)
γ
#
1
− β~ · α ~ (1 − ~n · β)~ + ~n · α
~
γ2
h i
3 2 2 ~
= γ c R (~n − β)~n · α ~ −α ~
~ (1 − ~n · β)

hence
~ −α ~
~ rad = e ~n · α
E
~ (~n − β) ~ (1 − ~n · β)
cR ~ 3
(1 − ~n · β) t0

~ rad = ~n × E
B ~ rad
We can use the vector identity ~a × (~b ×~c) = ~b(~a ·~c) −~c(~a · ~b) with ~a = ~n, ~b = ~n − β~ and ~c = α
~
to see that
~ ×α
~n × ([~n − β] ~ ) = ~n · α
~ (~n − β)~ −α ~
~ (1 − ~n · β)
so we can write the radiative electric field as
~
~ rad = e ~n × ([~n − β] × α
E
~)
Rc (1 − ~n · β)~ 3
t0

32
Notes for Classical Field Theory Section 6: Power radiated by accelerating charge

6 Power radiated by accelerating charge

The essential idea is that the energy flux per unit area in the direction ~n is given by ~ · ~n
S
where S~= cE ~ ×B ~ rad is the Poynting vector. We have that E~ rad is perpendicular to ~ rad
B
4π rad
and ~n is perpendicular to both with B ~ rad = ~n × E~ rad , so
~ = c |E
|~n · S| ~ rad |2
4π
Now the differential power radiated into a solid angle element dΩ in the direction ~n is
~
dP = R2 |~n · S|dΩ
This expression is in terms of the time t at the observation point; it is often more convenient
to work with the time t0 in the charge’s own frames. We can write the energy radiated
between times t1 and t2 as
Z t2 Z t02
E= ~ 2
|~n · S|dt dΩR = ~ dt dt0 dΩR2
|~n · S|
t1 t01 dt0
so we see that we should define
~ dt 2 c ~ dt
dP (t0 ) = R2 |~n · S|dΩ = R |Erad |2
dΩ
dt0 4π dt0
To evaluate the derivative, we use that
ct − ct0 = R
from the definition of the retarded time. As R = |R| ~ = |~x − ~xe (t0 ) we find
dt ~ · ~v
R dt
0
− 1 = − ⇒ 0 = 1 − ~n · β~
dt Rc dt
and hence
dP (t0 ) c ~ 2 ~
= R2 |E rad | (1 − ~
n · β)
dΩ 4π
and using the expression for E ~ rad evaluated in the preceding section we have
2
~
dP (t0 ) e2 ~n × ([~n − β] × α~)
=
dΩ 4πc ~ 5
(1 − ~n · β)
with this expression evaluated at the retarded time.
An important consequence of the electromagnetic radiation of an accelerating charged
particle is that classically an electron turning in circular motion will lose energy due to
radiation, leading to a decay of its orbit.
The remainder of this section of the course is not covered by these notes.
Because I do not hope to turn again
Because I do not hope
Because I do not hope to turn
T.S. Eliot, “Ash Wednesday”

33
Notes for Classical Field Theory Section 7: Bibliography

7 Bibliography
• Obviously most of the material above was taken from my notes from Dr Buttimore’s lec-
tures and shamelessly LATEXed out under my own name. However I can claim full credit
for any mistakes that have appeared, and would appreciate any corrections/suggestions
to cblair[at]maths.tcd.ie.

• The covariant derivation of the Lienard-Wiechart potentials was borrowed from Elec-
trodynamics by Fulvio Melia and Classical Electrodynamics by Jackson (note that the
first method for obtaining the electromagnetic fields from the potentials is taken from
Jackson while the second was contributed to the 432 course by former students).

• The Classical Theory of Fields by Landau and Lifshitz.

• There are some good exam-oriented notes by Eoin Curran at http://peelmeagrape.

net/eoin/notes/fields.pdf.

• The quotations throughout these notes were all mentioned in lectures by Dr Butti-
more; their precise relevance is left as an exercise for the reader. The text of Drac-
ula by Bram Stoker is available online at http://classiclit.about.com/library/
bl-etexts/bstoker/bl-bsto-drac-1.htm. The poems by Yeats and Eliot can pre-
sumably also be found online, or in any major collection of their work.

34