Mark Hindmarsh
University of Sussex
m.b.hindmarsh@sussex.ac.uk
http://www.pact.cpes.susx.ac.uk/users/markh/Teaching/
Syllabus
1
• Gauge field theory (Ryder 4.4, Cheng & Li Ch 8). Internal symme-
tries; Gauge symmetry 1: Abelian; The electromagnetic field; Gauge
symmetry 2: non-Abelian; Non-Abelian gauge field theory; Sponta-
neous symmetry breaking 1: Abelian; Spontaneous symmetry breaking
2: SU(2).
• Quantum gauge theory (Cheng & Li Ch 9, Ryder Ch7). Path-integral
quantisation; Fade’ev-Popov procedure, ghosts; Feynman rules in co-
variant gauge.
• Electroweak theory (Cheng & Li Ch 11). 1-family SU(2)×U(1) La-
grangian; Higgs mechanism; Mass spectrum; Family replication.
• QCD (Cheng & Li Ch 10). QCD Lagrangian and symmetries; Asymp-
totic freedom; Anomalies
Teaching methods
There will be 3 lectures a week throughout the term. Problem sheets will
be given every two weeks.
Assessment
Problem sheets will count for 100% of the total mark for the course, with a
25% weighting for a take-home exam at the end of the course.
Reading list
2
http://www.slac.stanford.edu/~mpeskin/QFT.html
The other books are complementary in some way. Kaku is very complete
but rather rushed. Itzykson & Zuber is compendious but lacks a coherent
development of the subject.
Prerequisites
Relativistic Quantum Fields 1, Further Quantum Mechanics.
Course Lecturer
3
Contents
1 Fermions 6
1.1 Plane wave solutions . . . . . . . . . . . . . . . . . . . . . . . 7
1.2 The interpretation of negative energy states . . . . . . . . . . 9
1.3 Quantising the spinor field . . . . . . . . . . . . . . . . . . . . 10
1.4 Conserved charge . . . . . . . . . . . . . . . . . . . . . . . . . 13
4 Electroweak theory 29
5 QCD 29
6 Renormalisation 29
A Pauli matrices 29
B Dirac matrices 30
B.1 Standard representation of Dirac matrices . . . . . . . . . . . 30
D Feynman rules 31
4
I Non-Abelian gauge field 37
L Problem Sheets 44
L.1 Problem Sheet 1. . . . . . . . . . . . . . . . . . . . . . . . . . 44
L.2 Problem Sheet 2. . . . . . . . . . . . . . . . . . . . . . . . . . 46
L.3 Problem Sheet 3. . . . . . . . . . . . . . . . . . . . . . . . . . 49
L.4 Problem Sheet 4. . . . . . . . . . . . . . . . . . . . . . . . . . 52
L.5 Problem Sheet 5. . . . . . . . . . . . . . . . . . . . . . . . . . 54
5
1 Fermions
Dirac knew of the problem with the probability interpretation of the Klein-
Gordon equation, and traced it to the fact that it was second order in time
derivatives. He therefore set out to find a relativistic wave equation with
only one time derivative. The requirement that the form of the equation be
unchanged under Lorentz transformations, which mix up ∂/∂t and ∇, means
that the equation must be first order in spatial derivatives as well. Hence
Dirac proposed a relativistic free particle wave equation
∂
ψ = −iα·∇ψ + βmψ.
i (1)
∂t
There clearly has to be something rather special about the objects α, β and
ψ in order that Lorentz covariance be preserved. In fact, α and β are 4×4
Hermitean matrices, and ψ is an object with four components called a spinor.
Now, we want a wave equation with plane wave solutions which satisfy
the relativistic energy-momentum relation E 2 = p2 + m2 . To see how this
property emerges from the Dirac equation, we act on both sides of (1) with
(i∂t − iα · ∇), to obtain
! ! !
∂ ∂ ∂
i − iα · ∇ i + iα · ∇ ψ = m i − iα · ∇ βψ. (2)
∂t ∂t ∂t
At this point, it is more convenient to switch to index notation for the 3-
vectors α and ∇, noting that α · ∇ = αi ∂i . We expand out the brackets on
the left hand side of (2), and add and subtract −iβαi ∂i on the right hand
side. The result is
∂2
! " ! #
i j ∂ i i i
− 2 + α α ∂i ∂j ψ = m i − iβα ∂i ψ + iβα ∂i ψ + iα β∂i ψ . (3)
∂t ∂t
It is now useful to define the anticommutator, represented by curly brackets:
{A, B} = AB + BA. (4)
(Sometimes you will also see the anticommutator written as [A, B]+ ). We
rewrite the left hand side as (−∂t2 + 21 αi αj ∂i ∂j + 12 αj αi ∂j ∂i )ψ, which we are
entitled to do by renaming the indices on α and ∂, to obtain
∂2
!
1
− 2 + {αi , αj }∂i ∂j ψ = m2 β 2 ψ + i{β, αi }∂i ψ. (5)
∂t 2
In order to reproduce the relativistic relation between energy and momentum,
we must end up with an equation like the Klein-Gordon equation. If αi and
β satisfy
{αi , αj } = δ ij , {β, αi } = 0, β 2 = 1, (6)
6
where 1 is the 4×4 identity matrix, ψ satisfies the equation
∂2
!
− 2 + ∇2 ψ = m2 ψ. (7)
∂t
γ 0 = β, γ i = βαi . (8)
The conditions (6) on αi and β are neatly unified into the matrix equation
{γ µ , γ ν } = 2η µν 1. (10)
(6 p − m)u = 0. (14)
6 a 6 b = a · b 1, (15)
7
to show that
(6 p + m)(6 p − m)u = (p2 − m2 )u = 0. (16)
This brings us back to the relation E 2 = p2 + m2 , although by a rather more
rapid route. It is conventional to choose p0 ≡ E always to be positive, in
which case we must also include solutions proportional to e+ip·x separately.
We write them
ψ(x) = veip·x , (17)
which satisfy
(6 p + m)v = 0. (18)
Solutions with time dependence e−iEt (eiEt ) are called positive (negative)
energy solutions.
We can use the first equality in Eq. (16) to show that the following spinors
are solutions to the Dirac equations for u and v:
! !
χ± 0
u± (p) = N (6 p + m) v± (p) = −N (6 p − m)
0 χ±
There are a couple of remarks that are worth making about the solutions (21)
and (22). Firstly, when the 3-momentum is small, that is when |p| m,
8
two of the components are much smaller than the other two. For the positive
energy solutions, it is the lower two components that are negligible, while the
opposite is true for the negative energy solutions. Secondly, the appearance
of two degrees of freedom for each set of solutions (that is, χ± ) demands some
explanation. One can verify that χ± are eigenstates of the Pauli matrix σ 3 ,
with eigenvalues ±1:
σ 3 χ± = ±χ± . (23)
This degeneracy is actually a result of the fact that the particles described
by the Dirac equation (such as the electron) have an extra property, spin,
which is a type of angular momentum intrinsic to the particle itself, unlike
ordinary angular momentum which particles gain by virtue of rotation. Spin
is quantised in half-integer units, with the spin of the electron being s = 21 .
As with other types of quantised angular momentum, the value of the spin
projected along any axis (such as the z axis) takes the values s, s − 1, . . . , −s.
Thus Dirac-type particles have only two spin states, which we conventionally
call “up” and “down”.
With the given normalisation convention, the spinors have the properties
9
a photon with energy E1 + E2 . There indeed seems to be nothing to stop the
energy of the electron from tumbling down towards minus infinity, releasing
an infinite amount of energy in the process. This is clearly nonsense. Dirac’s
way of circumventing this problem was to suppose that all the negative en-
ergy states are filled already, and the Pauli exclusion principle prevents a
positive energy electron from radiating a photon and occupying a filled nega-
tive energy state. One can still imagine exciting one of these negative energy
electrons into a postive energy state, whereupon it becomes a real electron,
leaving behind a hole. A hole is an absence of a negative electric charge,
which has positive charge. Thus Dirac realised that his theory predicted
the existence of a positively charged spin 12 particle with exactly the same
mass as the electron. Initially, Dirac identified this particle with the proton,
hoping that somehow Coulomb interactions would provide a mass difference,
but after others (including Weyl, Oppenheimer, and Tamm) showed that this
couldn’t be right, he eventually had to abandon this conservative position
and predict a new particle now know as the positron. Its discovery in 1932
by Carl Anderson in cosmic rays settled the issue.
However, this picture of a negative energy “sea” of electrons leaves much
to be desired. Bosons, particles with integer spin such as the photon (spin
1) and the pion (spin 0) do not obey the Pauli Exclusion Principle and
yet they still have negative energy states. Thus there can be no Dirac sea
picture and we are left where we were before. It is only if we abandon simple
wave mechanics and turn to quantum field theory that we find a satisfactory
resolution of this problem.
10
from which we derive the Dirac equation i 6 ∂ψ − mψ = 0. Similarly, we can
vary with respect to ψ and obtain
Z
δS = d4 x (iψ̄ 6 ∂δψ − mψ̄δψ). (31)
where the last term is an integral over the space-time surface at spatial
infinity (|x| → ∞) with end-caps at |t| → ∞. As usual, we suppose that the
variations die away at infinity so that we can drop the surface term, so we
recover the equation for the adjoint spinor i∂µ ψ̄γ µ + mψ̄ = 0.
The momentum conjugate to ψ is given by
∂L
π= = iψ̄γ 0 , (33)
∂ ψ̇
which shows that in the standard reresentation ψ and iψ † are canonically
conjugate variables. Thus we can find the Hamiltonian density:
11
We can now compute the Hamiltonian, which is the spatial integral of the
Hamiltonian density H, or
Z
H= d3 x (−iψ̄γ i ∂i ψ + mψ̄ψ) (38)
This means that the fermionic creation and annihilation operators must an-
ticommute, and we quantise the spinor field by anticommutation relations
There are four kinds of 1-particle states in this theory, created by the four
operators c†A (p) and d†B (p). We write them
As you can probably guess from the way the states are built up for the
complex scalar field, c†A (p) creates a particle with momentum p and spin
state A, while d†A (p) creates an antiparticle.
Let us examine the 2-particle state
12
If we interchange the particles, we must interchange the creation operators,
which gives the state a relative minus sign:
Thus we see something that has to be taken for granted in ordinary quan-
tum mechanics: the states of the spinor or Dirac field are antisymmetric
under particle interchange. We call these states fermionic, and the particles
fermions. Furthermore, suppose we take both momenta and both spin states
equal to p and A respectively: we then find
Hence two fermions can never be in the same state. This is precisely the
Pauli exclusion principle, which is fundamental to the understanding of the
physics of the atom. On the other hands, bosons, like the scalar particle we
considered previously, have commuting field operators, and so any number
of them can be in the same state.
Armed with the anticommutation relations for the creation and annihila-
tion operators, and the spinor completeness relations
uA A
vaA v̄bA = (γ · p)ab − mδab ,
X X
a ūb = (γ · p)ab + mδab , (53)
A A
the fields can be shown to satisfy the equal time anticommutation relations
where the factor of h̄ has been revived, and labels a and b have been put in
to keep track of the rows and columns of the spinors ψ and ψ̄.
13
By substituting the plane wave expansion for the Dirac field operator, one
can show that
XZ d3 p 1 † †
Q= c (p)c A (p) − d (p)dA (p) . (58)
A (2π)3 2E A A
Applying this operator to to the single particle states, one can show that
Hence the states created by c†A (p) have equal and opposite charge to those
created by d†A (p), which fits in with their interpretation as particles and
antiparticles.
One can also show that the existence of this conserved charge is a conse-
quence of a symmetry of the Lagrangian, as we should expect from Noether’s
theorem. This symmetry is
14
In this section we will study internal symmetries, which transform fields
amongst themselves:
δφa = Tab φb , δxµ = 0 (64)
The non-infinitesimal matrix Tab is called the generator the symmetry.
With this transformation it is straightforward to show that the following
non-infinitesimal 4-vector is conserved:
∂L
jµ = Tab φb . (65)
∂(∂µ φa )
We can extend these considerations to N complex scalar fields. For this we
need to recall that when differentiating with respect to a complex variable
z, one treats z and z̄ as independent. Given this fact, it follows that the
infinitesimal conserved current which follows from a symmetry of the action
S[φa , φ̄a ] is
∂L ∂L
fµ = δφa (x) + δ φ̄a (x)
∂(∂µ φa ) ∂(∂µ φ̄a )
!
∂L ∂L
− ∂ν φa (x) + ∂ν φ̄a (x) − Lδνµ δxν . (66)
∂(∂µ φa ) ∂(∂µ φ̄a )
Internal symmetry transformations for complex fields take the form
†
δφa = Tab φb , δ φ̄a = Tab φ̄b , δxµ = 0. (67)
Example 1. Consider a theory with two real fields φ1 and φ2 , and an action
1
Z
S= d4 x ∂µ φa ∂ µ φa − V (φ) , (68)
2
q
~ The
where φ = φ21 + φ22 is the magnitude of the 2-dimensional field vector φ.
action is clearly symmetric under rotations in the 2-dimensional “internal”
space of the field,
φ~→φ ~ 0 = Ωφ,
~ (69)
where Ω is a 2×2 orthogonal matrix
!
cos θ sin θ
Ω= . (70)
− sin θ cos θ
15
Using the antisymmetric symbol ab , defined by 12 = −21 = 1 and 11 =
22 = 0, we can rewrite the infinitesimal change in the field in component
form,
δφa = θab φb . (72)
In order to evaluate j µ we require
∂L
= ∂ µ φa . (73)
∂(∂µ φa )
Note that the set of 2×2 orthogonal matrices form a group under matrix
multiplication. We have been considering those matrices which are contin-
uously connected to the identity, and therefore have unit determinant. The
group of 2×2 orthogonal matrices with unit determinant is called SO(2).
where |φ| is the modulus of φ. This is clearly invariant under the transforma-
tion φ → φ0 = e−iθ φ, i.e. under multiplication by complex numbers of unit
modulus. If θ is infinitesimal, we have
is conserved. Note that the set of numbers of unit modulus form a group
under ordinary complex multiplication,
√ which is called U(1). Note also that
if one writes φ = (φ1 +iφ2 )/ 2, with φ1 and φ2 real, one can recover both the
action and the conserved current of the SO(2) scalar field theory. This is just
a consequence of the fact that the groups U(1) and SO(2) are isomorphic.
16
Example 3. Consider now the spinor field, whose action is
Z
S= d4 x ψ̄(i 6 ∂ − m)ψ. (79)
This also enjoys the symmetry ψ → ψ 0 = e−iθ ψ, but has an extra chiral
5
symmetry ψ → ψ 0 = e−iθ5 γ ψ, where γ 5 is defined in Appendix B. Note
5
that the adjoint spinor transforms as ψ̄ → ψ̄ 0 = ψ̄e−iθ5 γ , and so the mass
term ψ̄ψ is not invariant under this transformation. Similar considerations
to the previous examples show that the conserved current associated with
this symmetry is
j5µ = ψ̄γ µ γ 5 ψ. (84)
17
The simple spinor action is no longer invariant under this space-time depen-
dent transformation, as one can easily show:
Z
S → S0 = d4 x ψ̄eiα(x) (i 6 ∂ − m)e−iα(x) ψ,
Z
= S− d4 x ψ̄γ µ ψ∂µ α. (86)
18
2.3 Maxwell’s equations in covariant form
This section recaps some important results, and introduces the formulation
of the theory in an explicitly Lorentz covariant manner.
Firstly, we recall Maxwell’s equations in free space, writing them down
in natural units, in which the permittivity and permeability of free space µ0
and 0 are both unity:
Homogeneous Inhomogeneous
∇·B = 0 ∇·E = ρ (92)
∂ ∂
B+∇∧E=0 − E+∇∧B=j
∂t ∂t
The homogeneous and inhomogeneous (i.e. having a source term on the right
hand side) equations have differing status. The homogeneous equations imply
the existence of potentials φ and A from which the physically measurable
quantities E and B can be calculated. These potentials are specified only
up to a gauge transformation, φ → φ − Λ̇ and A → A + ∇Λ, where Λ is an
arbitrary function of space and time. The inhomogeneous equations imply
that the source terms must obey a current conservation equation, ρ̇+∇·j = 0.
To summarise:
Homogeneous Inhomogeneous
potentials φ, A current conservation
B = ∇∧A
(93)
E = −Ȧ − ∇φ
ρ̇ + ∇·j = 0
φ → φ − Λ̇
A → A + ∇Λ
All these quantities can be assembled into explicitly Lorentz covariant ob-
jects. The gauge potentials belong together in a 4-vector potential Aµ =
(φ, A), while the charge density ρ and the current density j can be put to-
gether into a 4-vector current density j µ = (ρ, j). Recall that putting quanti-
ties together into 4-vectors is not just a matter of notation: it means that the
quantities transform just like the space-time coordinates xµ under a Lorentz
transformation.
The electric and magnetic fields E and B also belong together in a Lorentz
covariant object, as they are mixed up by Lorentz transformations (an ob-
server moving through a magnetic field also sees an electric field). However,
this object cannot be a 4-vector as there are a total of 6 components of the
electric and magnetic fields, when they are taken together. In fact, the object
is an antisymmetric tensor, the field strength tensor F µν , whose entries are
19
as follows:
0 −E 1 −E 2 −E 3
E1 0 −B 3 B 2
F µν = . (94)
2 3 1
E B 0 −B
E 3 −B 2 B 1 0
Note that E 1 represents the first component of the electric field vector, which
in Cartesian coordinates is the x component. The reverse relations may be
written
1
E i = F i0 , B i = − ijk F jk , (95)
2
which introduces the Levi-Civita symbol ijk . The Levi-Civita symbol is
defined by
+1 if i 6= j 6= k cyclic,
ijk = −1 if i 6= j 6= k anticyclic, (96)
0 otherwise.
The field strength tensor has a neat expression in terms of the gauge potential
4-vector Aµ :
F µν = ∂ µ Aν − ∂ ν Aµ . (97)
Let us check this explicitly for the components corresponding to the electric
field:
F i0 = ∂ i A0 − ∂ 0 Ai = ∂ i A0 − Ȧi = −∂i φ − Ȧi = E i , (98)
where we have used the expression for the electric field in terms of the gauge
potentials in equation (93).
When expressed in terms of 4-vectors and tensors, electromagnetism looks
very simple and beautiful. For example, the covariant expression of the
current conservation equation is simply
∂µ j µ = 0. (99)
20
We can use the four-dimensional Levi-Civita tensor to re-express the ho-
mogeneous equations more compactly. This tensor has four indices, and is
defined by
if µ 6= ν 6= ρ 6= σ, symmetric,
+1
µνρσ = −1 if µ 6= ν 6= ρ 6= σ, antisymmetric, (101)
0 otherwise.
µνρσ ∂ ρ F µν = 0. (103)
δAν (y)
= δνµ δ 4 (y − x). (105)
δAµ (x)
21
2.5 Gauges
The physical quantities, the electric and magnetic fields, contained in the
field strength tensor Fµν , are invariant under a gauge transformation
The last step follows because we can take the partial derivatives in any order.
There are two related problems with the theory as formulated above.
The solution to these problems begins with fixing the gauge: imposing a
constraint on the field to remove the freedom to make a gauge transformation.
For each constraint, the field equations may be solved.
There are in principle an infinite number of constraints one could apply,
but in practice three turn out to be convenient.
1. Temporal gauge A0 = 0.
3. Lorentz gauge ∂ · A = 0.
Only the last is obviously Lorentz invariant, and so is most useful for
relativistic field theory. The other two gauges are useful in other circum-
stances: for example, temporal gauge is the easiest to use for solving the
field equations numerically on a lattice, while Coulomb gauge is useful for
non-relativistic applications in atomic physics.
22
2.6 Coulomb gauge solutions
This is also called radiation gauge, and is defined by
∇·A = 0, (110)
which implies that the first of the inhomogeneous equations in (92) becomes
∇2 φ = −ρ. (111)
2
− ∇2 A = 0. (115)
∂t
One of the important discoveries of the last century was that this equation has
plane wave solutions, which carry energy and momentum: electromagnetic
waves. A plane wave solution with a particular wavenumber k may be written
23
which implies
k·a = 0. (118)
Hence the gauge potential A is orthogonal to the wave vector k. This means
that both the electric and magnetic fields are also orthogonal: we say that
the waves are transverse.
∂µ Aµ = 0, (119)
and is manifestly Lorentz invariant. In this gauge the equation of motion for
the gauge field is simplified, for
∂µ F µν = ∂µ ∂ µ Aν − ∂µ ∂ ν Aµ = ∂ 2 Aν = j ν . (120)
However, an irritating feature of this gauge is that it does not quite specify
Aµ fully. One can still make a gauge transformation Aµ → Aµ − ∂ µ Λ which
satisfies the Lorentz gauge condition (119), as long as the function Λ satisfies
∂ 2 Λ = 0. Such functions are called harmonic. In classical field theory this is
not too much of a problem, but in setting up the quantum theory of gauge
fields care must be taken.
In free space, the field equation is (120)
∂2
!
2
∂ A ≡µ
− ∇2 Aµ = 0. (121)
∂t2
A particular solution is
Aµ (t, x) = aµ e−ik·x , (122)
where k 0 = ωk = |k|, and aµ is a constant 4-vector.
The Lorentz gauge condition (119) gives ∂ · A = −ik · ae−ik·x = 0, which
implies
k · a = 0. (123)
Once again the field is orthogonal to the wavenumber k, but this time in the
4-vector sense.
The general solution may be constructed from a superposition of plane
wave solutions
Z
d¯3 k µ
Aµ (x) = a (k)e−ik·x + a∗ µ (k)eik·x , (124)
2ωk
24
where the amplitudes aµ (k) must all satisfy k · a(k) = 0.
It would be more convenient to be able to choose the amplitudes of the
components of the gauge field independently, and to this end we introduce a
set of four basis 4-vectors, or polarisation vectors, A µ (k), with A = 0, 1, 2, 3,
for each wavevector k. They must satisfy a completeness condition in order
for them to be considered as basis vectors, and it is also convenient to make
them orthonormal, in a 4-vector sense.
A good choice is
0 = (1; 0),
1 = (0, n̂ × k)/|n̂ × k|,
2 = (0, k × (n̂ × k))/|k × (n̂ × k)|,
3 = (0; k̂),
(127)
25
By multiplying with 3µ we see that this is equivalent to shifting a3 (k),
3
a3 (k) → a0 (k) = a3 (k) − i|k|λ(k). (132)
Thus although the Lorentz gauge does not entirely remove the freedom to
make gauge transformations, it does confine the ambiguity in the solution to
one and only one of the sets of constants in the solution. There are only two
physical amplitudes to choose, a1 and a2 .
Let us end this section by calculating the Hamiltonian in the Lorentz
gauge.
Z
HLg = d3 x πµ Ȧµ − LLg
1Z 3
= − d x(Ȧµ Ȧµ + ∇Aµ ·∇Aµ ). (133)
2
We already notice from this expression that the Hamiltonian is not positive
definite, as the timelike component of the gauge field A0 appears with a
negative sign. It turns out that we have already cured this problem in the
classical theory. To see this, let us substitute the plane wave expansion of
the field operator, obtaining
Z
d¯3 k
HLg = − ωk ηAB a∗A (k)aB (k). (134)
2ωk
We saw earlier that the Lorentz gauge condition Eq. 119 implies a0 = a3 ,
and so we find that
X Z d¯3 k
HLg = ωk a∗A (k)aA (k), (135)
A=1,2 2ωk
which is clearly positive definite, and depends only on the physical tranversely
polarised components.
26
the canonical momentum vanishes. This turns out to be cured by fixing the
gauge, as we can verify in the Lorentz gauge:
LLg
πµ = = −Ȧµ . (136)
∂ Ȧµ
Hence the equal time canonical commutation relations can be guessed to be
those for four independent real fields
By substituting the plane wave expansion for the field operator Aµ (x), one
can find (after some algebra) that
Hence we have four pairs of ladder operators, one pair for each polarisation.
We define the vacuum state |0i by
and we can construct excited states by acting with the raising operators
a∗A (k). For example,
|k, Ai = a∗A (k)|0i. (142)
How do we impose the gauge constraint in the quantum theory? Firstly,
it is clear that demanding ∂ · A = 0 is too strong a condition on the operator
Aµ , as it is inconsistent with the equal time canonical commutation relations
Eq. 137. A weaker condition is to ask that matrix elements of the gauge
condition vanish. This we cannot do for every possible state, but we only
accept as physical states for which this is true: i.e.
for any two physical states |φi and |φ0 i. Any equivalent condition is to
demand that the positive frequency part of the gauge condition gives zero
when acting on a physical state:
where Z
d¯3 k
∂ · A(+) = − ik · ae−ik·x . (145)
2ωk
27
Hence physical states must satisfy
X Z d¯3 k
hφ0 |:HLg :|φi = ωk hφ0 |a∗A (k)aA (k)|φi. (148)
A=1,2 2ωk
28
However, proving this relation will have to wait until we have studied the
path integral quantisation of the gauge field. Meanwhile, we will jump the
gun and write down the Feynman rules for the electromagnetic field
k, r
Ingoing µ (k)
k,
Outgoing r ∗µ (k)
µ νr Z
iηµν
Internal r d¯4 k 2
k + i
(σ 1 )2 = (σ 2 )2 = (σ 3 )2 = 1. (155)
This set of relations can be neatly expressed using the 3 dimensional Levi-
Civita symbol εijk :
[σ i , σ j ] = 2iεijk σ k , (158)
29
and with the anticommutation relations
{σ i , σ j } = 2δ ij . (159)
One may thus write the product of any two Pauli matrices as
1 1
σ i σ j = {σ i , σ j } + [σ i , σ j ] = δ ij + iεijk σ k . (160)
2 2
The Pauli matrices form a representation of the angular momentum al-
gebra with (spin) angular momentum 12 .
B Dirac matrices
Dirac matrices (in four space-time dimensions) are 4×4 matrices defined by
the anitcommutation relations
{γ µ , γ ν } = 2η µν 14 , (161)
where η µν is the Minkowski metric, and 14 is the 4×4 identity matrix (which
is often dropped and left implicit in the equation).
Properties of the Dirac γ-matrices include:
(i). (γ µ )† = γ 0 γ µ γ 0 .
30
C Identities for Dirac matrices
(i). tr(γ µ γ ν ) = 4η µν
(v). γ µ γµ = 41
(vi). γ µ 6 aγµ = −2 6 a
(viii). γ µ 6 a 6 b 6 cγµ = −2 6 c 6 b 6 a
D Feynman rules
The rules which one derives from the LSZ reduction formula are for calcu-
lating S-matrix elements Sαβ = hα|Ŝ|βi. In momentum space one finds, for
a real scalar field, that the following rules are useful:
Z
i
Internal line (propagator) d¯4 k
k2 − m2 + i
k2 k3
k1 k4
31
state. Furthermore, there are also uninteresting contributions from graphs
with no vertices.
It is therefore convenient to define matrix elements Mαβ through
We can derived slightly simpler Feynman rules for these matrix elements,
in which the propagators lose their associated integrations and the vertices
their δ-functions.
Consider a diagram with E external lines, V vertices, and I internal lines.
For the simple φ4 theory we know that external lines end on a vertex, internal
lines have vertices at each end, and vertices must be attached to four lines.
Hence
E + 2I = 4V.
There are I integrations and V δ-functions, and we know that there is always
one δ-function left over. Hence we can trivially perform V − 1 integrations,
leaving
L=I −V +1
integrations to perform.
We call the number of free integrations the number of loops, as for simple
graphs it does indeed correspond to the number of closed loops in the graph.
The following sections give the Feynman rules for working out the matrix
elements Mαβ for several theories of interest.
32
E Real scalar field
1 1 1
Z
4
S= dx ∂µ φ∂ µ φ − m2 φ2 − φ4
2 2 4!
k
→
Ingoing line 1
k
→
Outgoing line 1
→
k i
Internal line (propagator)
k2 − m2 + i
Vertex −iλ
33
k
→
Ingoing particle line 1
→
k
k
→
Outgoing particle line 1
→
k
→
k i
Internal line (propagator)
k2 − m2 + i
Vertex −iλ
34
G U(1) gauge field
!
Z
1 1
S= d x − Fµν F µν − (∂ · A)2
4
4 2ξ
k,
→
Ingoing line µ
k,
→
Outgoing line ∗µ
µ ν
→
k ηµν + (ξ −
Internal gauge line (propagator) i
k2
35
k, A
→
Ingoing particle line uA (p)
→
k, A
k, A
→
Outgoing particle line ūA (p)
→
k, A
→
p i
Internal line (propagator)
6 p − m + i
k1
k2
36
I Non-Abelian gauge field
!
Z
4 1 a aµν 1
S= d x − Fµν F − (∂ · Aa )(∂ · Aa ) − c̄a ∂ · (Dc)a
4 2ξ
37
k, , a
→
Ingoing line
k, , a
→
Outgoing line
µ, a ν, b
→
k ηµν
Internal gauge line (propagator) iδab
a b
→
k
Internal ghost line (propagator)
k2 , ν, b
k1 , µ, a
k3 , ρ, c ig[(
3-point gauge vertex (
ν, b ρ, c
−ig 2 [feab
4-point gauge vertex +feac
µ, a σ, d +fead f
38
µ, a
k, c
• Closed ghost loops are associated with a factor (−1).
39
J Complex scalar field with non-abelian gauge
symmetry
Z
S= d4 x (Dµ Φ)† (Dµ Φ) − m2 Φ† Φ − Vint (Φ, Φ† ) .
40
k
→
Ingoing particle line 1
→
k
k
→
Outgoing particle line 1
→
k
→
k iδmn
Internal line (propagator)
k2 − m2 + i
k1 , m
µ, a
k2 , n
a
3-point gauge-scalar vertex igTmn (k1 + k2 )µ
ν, b n
41
K Spinor field with non-abelian gauge sym-
metry
Z
S= d4 x Ψ̄(i 6 D − m)Ψ.
42
k, A, m
→
Ingoing particle line umA (p)
→
k, A, m
k, A, m
→
Outgoing particle line ūmA (p)
→
k, A, m
m n
→
p iδmn
Internal line (propagator)
6 p − m + i
k1 , m
µ, a
k2 , n
a
3-point vertex igTmn γµ
43
L Problem Sheets
L.1 Problem Sheet 1
1. (a) If we write
ψ(x) = ue−ip·x + veip·x ,
1
with p0 = E ≡ |(p2 + m2 ) 2 |, show that the the following expres-
sions for u and v give solutions to the Dirac equation (i 6 ∂ −m)ψ =
0:
! !
− 21 χ± − 12 0
u± (p) = (E+m) (6 p+m) v± (p) = −(E+m) (6 p−m)
0 χ±
! !
1 0
where χ+ = , and χ− = .
0 1
(b) Show also that
u†A uB = 2EδAB , vA† vB = 2EδAB
where the indices A and B range over the spin states + and −,
and that
ūA uB = 2mδAB , v̄A vB = −2mδAB ,
where ūA = u†A γ 0 , and v̄A = vA† γ 0 are adjoint spinors.
2. The Lagrangian density for a spinor field may be written
L = ψ̄(i 6 ∂ − m)ψ.
(a) Given that the general expression for the canonical energy-momentum
tensor is
∂L ∂L
θµ ν = ∂ν ψ + ∂µ ψ̄ − Lδνµ ,
∂(∂µ ψ) ∂(∂µ ψ̄)
evaluate the momentum density θi 0 and the energy density θ0 0 ,
showing that the latter is equal to the Hamiltonian density H.
(b) Given the plane wave expansion of the spinor field
X Z d¯3 p
ψ(x) = cA (p)uA (p)e−ip·x + d†A (p)vA (p)eip·x ,
A=± 2E
show that
X Z d¯3 p
H= E c†A (p)cA (p) − dA (p)d†A (p) .
A=± 2E
44
3. We define the matrix γ 5 = iγ 0 γ 1 γ 2 γ 3 .
(γ 5 )† = γ 5 , (γ 5 )2 = 1, {γ 5 , γ µ } = 0,
j µ = ψ̄γ µ ψ.
45
4. (a) Consider a function of a pair complex Grassmann variables f (θ, θ̄).
46
(a) As shown in the lectures, there is a U(1) symmetry leading to a
conserved current j µ = iφ̄(∂ µ φ) − i(∂ µ φ̄)φ. Show that
Z
d¯3 k ∗
:Q: = (a ak − b∗k bk ).
2ωk k
(b) Show that [:Q:, ak ] = −ak , [:Q:, a∗k ] = a∗k , [:Q:, bk ] = bk , and
[:Q:, b∗k ] = −b∗k ,
(c) Hence or otherwise show that the states a∗ (k)|0i and b∗ (k)|0i
(where |0i is the vacuum state) are eigenstates of the charge op-
erator, and give their charges.
(d) Show that [:Q:, φ(x)] = −φ(x), and prove that of Q|0i = 0 then
the vacuum expectation value of the field operator, h0|φ(x)|0i,
must vanish.
47
2. Consider a massless spinor with action
Z
S= d4 x ψ̄i 6 ∂ψ,
where Dµ1 = ∂µ − ie1 A1µ and Dµ2 = ∂µ − ie2 A2µ , giving the coupling
constants e1 and e2 in terms of eA and eB , and the vector fields
A1µ and A2µ in terms of Vµ and Aµ .
δFρσ (x) ∂ 4 ∂
= ρ
δ (x − y)δσµ − σ δ 4 (x − y)δρµ ,
δAµ (y) ∂x ∂x
48
4. Consider an action for an SU(2) doublet spinor coupled to an SU(2)
gauge field
1 a aµν
Z
4
S= d x − Fµν F + Ψ̄i 6 DΨ − mΨ̄Ψ .
4
where Dµ = ∂µ − igAaµ τ a , Fµν
a
= ∂µ Aaν − ∂ν Aaµ + gabc Abµ Acν , and τ a =
a a
σ /2, where σ are the Pauli matrices.
a
(b) Show that F̃µν = 12 µνρσ F aρσ satisfies the identity
∂ ν F̃µν
a
+ gabc Abν F̃µν
c
≡ 0.
You may find the identity abc cde = δad δbe − δae δbd useful.
(c) (Optional) Defining a 4-vector “current” Kµ = Aaν F̃µν
a
, show that
1
∂ µ Kµ = F aµν F̃µν
a
2
.
49
(c) The Feynman propagator for the massive vector field is defined in
the usual way by
Give an argument which shows that the propagator must have the
form Z
Dµν (x) = δ̄ 4 k [A(k 2 )ηµν + B(k 2 )kµ kν ]e−ik·x
Show that the spinor field has a mass gv, and interacts with the
fields h and a through the terms −ghψ̄ψ and −igaψ̄γ 5 ψ
50
(b) If the potential has the form V (Φ) = −µ2 Φ† Φ + λ(Φ† Φ)2 , show
that the Lagrangian has the U(1) global symmetry Φ → Φe−iβ .
Suppose that we wish to gauge this symmetry: how should we
amend the Lagragian? (You should display the transformation
properties of any extra fields you introduce).
(c) Suppose now that we have two SU(2) scalar doublets Φ1 and Φ2 ,
transforming under the combined SU(2)×U(1) gauge symmetry
as
Φ1 → Φ01 = U Φ1 e−iβ , Φ1 → Φ02 = U Φ2 eiβ .
Verify that the following potential is gauge invariant:
V (Φ1 , Φ2 ) =
−µ21 Φ†1 Φ1 − µ22 Φ†2 Φ2 + λ1 (Φ†1 Φ1 )2 + λ2 (Φ†2 Φ2 )2 + λ3 (Φ†1 Φ2 )(Φ†2 Φ1 ),
−1
δf
δ[f (φ) − ω] = δ[φ − φ̃] det .
δφ
(b) The measure for the functional integration over a complex scalar
field is Dφ̄Dφ = x dφ̄(x)dφ(x). Introducing “polar” fields ρ(x)
Q
Show that R
d4 x ln ρ(x)
Dφ̄Dφ = DρDθe .
51
L.4 Problem Sheet 4
1. The axial gauge is defined by
Ga (A) = tµ Aaµ = 0,
δGa (A(x))
Mab (x, y) = ,
δαb (y)
and hence write down the complete action for a non-Abelian gauge
field Aaµ in the axial gauge, include gauge-fixing and ghost terms.
(b) Show that the momentum space propagator in the axial gauge is
tµ k ν + k µ tν (ξk 2 − t2 ) µ ν
" #
µν δab
Dab (k) = 2 −η µν + + k k .
k + i k·t (t · k)2
µ ν
52
where
Z
[q µ (q + k)ν + (q + k)µ q ν − η µν (q · (q + k) − m2 )]
iΠµν (k) = −4e2 d¯4 q .
[(q + k)2 − m2 + i][q 2 − m2 + i]
γ µ γµ = 41, γ µ γ ν γµ = −2γ ν .
Show that applying the Feynman rules (in Feynman gauge) to this
diagram results in the expression
i i
iΣ(p)
6 p − m + i 6 p − m + i
where
Z
4m − 2 6 q
iΣ(p) = −e2 d¯4 q .
[(p − q)2 + i](q 2 − m2 + i)
53
µ, a ν, b
(a) Find the expression for the contribution of this diagram to the
polarisation tensor Πµν
ab resulting from the application of the Feyn-
man rules (in Feynman gauge) to this diagram.
(b) In a non-Abelian gauge theory, with no other fields, what other
diagrams contribute to Πµν
ab ?
54
(d) Define m2 (φ) = m2 + 21 gφ2 . Show that, using an upper cut-off
Λ m(φ) on the momentum integration,
Λ2 2 1 1
4 2 2
V1 (φ) = N 2
m (φ) + N 2
m (φ) ln m (φ)/Λ − .
16π 16π 2
55
2. (a) Write down the covariant derivative terms for the fermion fields
in the Glashow-Weinberg-Salam (GWS) model of electroweak in-
teractions, in terms of their weak eigenstates, SU(2)×U(1) gauge
fields Aaµ and A0µ , and gauge couplings g and g 0 . Show that the
interaction terms can be written
Ls,i = g(Wµ J µ + Wµ† J †µ + Zµ JZµ ) + eAµ Jem
µ
,
√
where Wµ = (A1µ − iA2µ )/ 2, Zµ = cos θW A3µ − sin θW A0 µ, Aµ =
sin θW A3µ + cos θW A0µ , and θW is the weak mixing angle. Be sure
to give full expressions for the currents J µ , JZµ , and Jem
µ
.
(b) Show that JZµ and Jem µ
maintain their form in the mass eigenstate
basis, i.e. that (unlike the charged current) these currents do not
couple fermions with different flavour indices.
(c) Let Φ be the scalar doublet of the GWS model, with Φ† Φ = (v +
h(x))2 /2 in the symmetry-broken phase, where h is the Higgs field.
Show that the couplings of the Higgs to fermion mass eigenstates
are proportional to the masses of the fermions.
(d) Show also that there are Higgs-W interaction terms of the forms
hW W and hhW W , giving the coupling constants.
3. The bare interaction Lagrangian of QED is Li = e0 ψ̄0 γµ ψ0 Aµ0 .
(a) Show that if we demand that the action be dimensionless in d
dimensions, the mass dimension of the bare coupling constant e0
is /2, where = 4 − d.
(b) Show that the photon self energy of Problem 4.2(b) becomes
iΠµν (k) =
Z 1 Z
(1 − d2 )η µν l2 − 2x(1 − x)pµ pν + η µν (m20 + x(1 − x)k 2 )
−4e20 dx d¯d l ,
0 (l2 + α)2
where α = m20 −x(1−x)k 2 . Note that in d dimensions, ηµν η µν = d.
(c) Perform the integrals to show that iΠµν (k) = (η µν k 2 −k µ k ν )iΠ(k 2 ),
where
e20 Z 1 2
2
Π(k ) = − dxx(1 − x) − ln α − γ .
2π 0
(d) Given that the bare photon propagator can be written
−iηµν
iDµν (k) = ,
k 2 (1
− Π(k 2 ))
56
show how the divergence in Π(k 2 ) can be dealt with using the
photon wavefunction renormalisation constant Z3 .
(e) Introducing an arbitrary renormalisation parameter µ to define a
dimensionless renormalised coupling e = Z3 e0 µ−/2 , find the QED
β-function,
de
β=µ
dµ
in the limit → 0. What is peculiar about the relation between
the renormalised and bare couplings?
57