Note that an antisymmetric 4 × 4 matrix has 4 × 3/2 = 6 independent components,which agrees with the 6 transformations of the Lorentz group: 3 rotations and 3 boosts.It’s going to be useful to introduce a basis of these six 4 × 4 anti-symmetric matrices.We could call them (M A ) µν , with A = 1, . . .,6. But in fact it’s better for us (althoughinitially a little confusing) to replace the single index A with a pair of antisymmetricindices [ρσ], where ρ, σ = 0, . . .,3, so we call our matrices (M ρσ ) µ ν. The antisymmetryon the ρ and σ indices means that, for example, M 01 = −M 10 , etc, so that ρ and σagain label six different matrices. Of course, the matrices are also antisymmetric onthe µν indices because they are, after all, antisymmetric matrices. With this notationin place, we can write a basis of six 4 × 4 antisymmetric matrices as(M ρσ ) µν = η ρµ η σν − η σµ η ρν (4.7)where the indices µ and ν are those of the 4 × 4 matrix, while ρ and σ denote whichbasis element we’re dealing with. If we use these matrices for anything practical (forexample, if we want to multiply them together, or act on some field) we will typicallyneed to lower one index, so we have(M ρσ ) µ ν = ηρµ δ σ ν − ησµ δ ρ ν (4.8)Since we lowered the index with the Minkowski metric, we pick up various minus signswhich means that when written in this form, the matrices are no longer necessarilyantisymmetric. Two examples of these basis matrices are,(0 1 0 0)(M 01 ) µ ν = 1 0 0 00 0 0 00 0 0 0(0 0 0 0)and (M 12 ) µ ν = 0 0 −1 00 1 0 00 0 0 0(4.9)The first, M 01 , generates boosts in the x 1 direction. It is real and symmetric. Thesecond, M 12 , generates rotations in the (x 1 , x 2 )-plane. It is real and antisymmetric.We can now write any ω µ ν as a linear combination of the M ρσ ,ω µ ν = 1 2 Ω ρσ (M ρσ ) µ ν(4.10)where Ω ρσ are just six numbers (again antisymmetric in the indices) that tell us whatLorentz transformation we’re doing. The six basis matrices M ρσ are called the generatorsof the Lorentz transformations. The generators obey the Lorentz Lie algebrarelations,[M ρσ , M τν ] = η στ M ρν − η ρτ M σν + η ρν M στ − η σν M ρτ (4.11)– 82 –
where we have suppressed the matrix indices. A finite Lorentz transformation can thenbe expressed as the exponentialΛ = exp ( 12 Ω ρσM ρσ) (4.12)Let me stress again what each of these objects are: the M ρσ are six 4 × 4 basiselements of the Lorentz group; the Ω ρσ are six numbers telling us what kind of Lorentztransformation we’re doing (for example, they say things like rotate by θ = π/7 aboutthe x 3 -direction and run at speed v = 0.2 in the x 1 direction).4.1 The Spinor RepresentationWe’re interested in finding other matrices which satisfy the Lorentz algebra commutationrelations (4.11). We will construct the spinor representation. To do this, we startby defining something which, at first sight, has nothing to do with the Lorentz group.It is the Clifford algebra,{γ µ , γ ν } ≡ γ µ γ ν + γ ν γ µ = 2η µν 1 (4.13)where γ µ , with µ = 0, 1, 2, 3, are a set of four matrices and the 1 on the right-hand sidedenotes the unit matrix. This means that we must find four matrices such thatandγ µ γ ν = −γ ν γ µ when µ ≠ ν (4.14)(γ 0 ) 2 = 1 , (γ i ) 2 = −1 i = 1, 2, 3 (4.15)It’s not hard to convince yourself that there are no representations of the Cliffordalgebra using 2 × 2 or 3 × 3 matrices. The simplest representation of the Cliffordalgebra is in terms of 4 × 4 matrices. There are many such examples of 4 × 4 matriceswhich obey (4.13). For example, we may take( ) ( )0 10 σiγ 0 = , γ i =(4.16)1 0−σ i 0where each element is itself a 2 × 2 matrix, with the σ i the Pauli matrices( ) ( ) ( )0 10 −i1 0σ 1 = , σ 2 = , σ 3 =1 0i 00 −1(4.17)which themselves satisfy {σ i , σ j } = 2δ ij .– 83 –
- Page 5 and 6:
4.7.2 Some Useful Formulae: Inner a
- Page 7 and 8:
0. Introduction“There are no real
- Page 9 and 10:
At distances shorter than this, the
- Page 11 and 12:
which allows us to express all dime
- Page 13 and 14:
1. Classical Field TheoryIn this fi
- Page 15 and 16:
and the potential energy of the fie
- Page 17 and 18:
1.2 Lorentz InvarianceThe laws of N
- Page 19 and 20:
Example 3: Maxwell’s EquationsUnd
- Page 21 and 22:
(where the sign in the field transf
- Page 23 and 24:
from which we see thatδφ = −ω
- Page 25 and 26:
Another Cute TrickThere is a quick
- Page 27 and 28:
2. Free Fields“The career of a yo
- Page 29 and 30:
Substituting into the above express
- Page 31 and 32:
Hmmmm. We’ve found a delta-functi
- Page 33 and 34:
2.3.2 The Casimir Effect“I mentio
- Page 35 and 36:
This is still infinite in the limit
- Page 37 and 38: This space is known as a Fock space
- Page 39 and 40: From this result we can figure out
- Page 41 and 42: 2.6 The Heisenberg PictureAlthough
- Page 43 and 44: But what about arbitrary spacetime
- Page 45 and 46: where T stands for time ordering, p
- Page 47 and 48: Im(p 0)Im(p 0)−E p+ E pRe(p 0)−
- Page 49 and 50: We may expand ψ(⃗x) as a Fourier
- Page 51 and 52: which confirms (2.120). So we learn
- Page 53 and 54: 3. Interacting FieldsThe free field
- Page 55 and 56: means that for experiments at small
- Page 57 and 58: The interaction picture is a hybrid
- Page 59 and 60: Actually these last two terms doubl
- Page 61 and 62: • Obviously we can’t cope with
- Page 63 and 64: where the ± signs on φ ± make li
- Page 65 and 66: Now, using Wick’s theorem we see
- Page 67 and 68: solid lines to denote its charge; w
- Page 69 and 70: p 1p 1/p 1p 1/p 2p 2/p 2p 2/Figure
- Page 71 and 72: Notice that the momentum dependence
- Page 73 and 74: 3.5.2 The Yukawa PotentialSo far we
- Page 75 and 76: where we’ve introduced the dimens
- Page 77 and 78: • We do not consider diagrams wit
- Page 79 and 80: is the probability for the transiti
- Page 81 and 82: meaning that the flux is given in t
- Page 83 and 84: But the last term vanishes. This fo
- Page 85 and 86: the S-matrix elements, where we wer
- Page 87: 4. The Dirac Equation“A great dea
- Page 91 and 92: 4.1.1 SpinorsThe S µν are 4 × 4
- Page 93 and 94: which can be anti-hermitian if all
- Page 95 and 96: so the requirement (4.44) becomes
- Page 97 and 98: 4.4 Chiral SpinorsWhen we’ve need
- Page 99 and 100: 4.4.2 γ 5The Lorentz group matrice
- Page 101 and 102: 4.4.4 Chiral InteractionsLet’s no
- Page 103 and 104: where we’ve made use of the prope
- Page 105 and 106: which, after direct substitution, t
- Page 107 and 108: for any 2-component spinor ξ which
- Page 109 and 110: HelicityThe helicity operator is th
- Page 111 and 112: Proof:2∑s=1u s (⃗p) ū s (⃗p)
- Page 113 and 114: Claim: The field commutation relati
- Page 115 and 116: The δ (3) term is familiar and eas
- Page 117 and 118: Let’s pause our discussion to mak
- Page 119 and 120: In what follows we will often drop
- Page 121 and 122: 5.6 Yukawa TheoryThe interaction be
- Page 123 and 124: where we’ve put the φ propagator
- Page 125 and 126: The denominators in each term are d
- Page 127 and 128: Finally, we can also compute the sc
- Page 129 and 130: We could again try to take the non-
- Page 131 and 132: These remain true even in the prese
- Page 133 and 134: line can be reached by a gauge tran
- Page 135 and 136: which means that ⃗ ξ is perpendi
- Page 137 and 138: 6.2.2 Lorentz GaugeWe could try to
- Page 139 and 140:
The minus signs here are odd to say
- Page 141 and 142:
a 1,2 †⃗p, while |φ〉 contain
- Page 143 and 144:
where we’ve introduced a coupling
- Page 145 and 146:
Let’s now consider a complex scal
- Page 147 and 148:
6.4.1 Naive Feynman RulesWe want to
- Page 149 and 150:
which is the claimed result. You ca
- Page 151 and 152:
Electron ScatteringElectron scatter
- Page 153 and 154:
momenta, we have the amplitudes giv
- Page 155:
6.7 AfterwordIn this course, we hav