PHYS3936

ADVANCED STREAM

PARTICLE PHYSICS

COURSE NOTES

2021

Last revised: February 27, 2021

1
Kevin Varvell and Bruce Yabsley

1
School of Physics, The University of Sydney, NSW 2006, Australia
Contents

Introduction 2
Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Suggested Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Assumed Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Basic Building Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
The Atom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
The Electron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
The Nucleus and the Proton . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
The Neutron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
The Photon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
The Positron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
The Neutrino and Anti-neutrino . . . . . . . . . . . . . . . . . . . . . . . . . 7
The Muon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
The Particle Explosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Electron-Proton and Deep Inelastic Scattering . . . . . . . . . . . . . . . . . . . . 8
Fundamental Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
An Aside on Units for Particle Masses . . . . . . . . . . . . . . . . . . . . . . . . 10

1 The Particle Zoo 11

1.1 Fundamental Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.1.1 Leptons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

i
 1.1.2 Quarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.1.3 Gauge Bosons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.1.4 Higgs Boson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.2 Composite Particles: Hadrons . . . . . . . . . . . . . . . . . . . . . . . . . . 13
1.2.1 Mesons (q q̄) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2.2 Baryons (qqq/q̄ q̄ q̄) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2.3 Excited States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.2.4 Hadron Spin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.2.5 Exotic Hadrons? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2 Angular Momentum and Particle Physics 20

2.1 Angular Momentum in Classical Mechanics . . . . . . . . . . . . . . . . . . . 20
2.2 Angular Momentum in Quantum Mechanics . . . . . . . . . . . . . . . . . . 21
2.3 Spin in Quantum Mechanics . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.3.1 What is spinning? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4 Fermions and Bosons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Combining Angular Momentum Vectors . . . . . . . . . . . . . . . . . . . . 24
2.6 Total Angular Momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.7 The spin of a proton? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.7.1 Proton as a single unresolved object . . . . . . . . . . . . . . . . . . . 25
2.7.2 Proton as a composite object . . . . . . . . . . . . . . . . . . . . . . 25
2.8 Spin is not conserved! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3 Special Relativity and Particle Physics 27

3.1 Non-Relativistic Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.2 Relativistic Kinematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.3 Total Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.4 4-Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.5 Relativistic Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

ii
 3.6 Energy, Momentum and Mass . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.7 Colliding Beams of Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.7.1 Why Colliding Beams? . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.8 Reconstructing Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.8.1 Example: Reconstructing a decay . . . . . . . . . . . . . . . . . . . . 34

4 Interactions 35
4.1 Decay and Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.2 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.3 Feynman Diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.3.1 The Vertex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.3.2 A Scattering Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.4 Example: Electromagnetic Interactions . . . . . . . . . . . . . . . . . . . . . 38
4.4.1 Electron-Electron scattering . . . . . . . . . . . . . . . . . . . . . . . 39
4.4.2 Electron-Positron Scattering . . . . . . . . . . . . . . . . . . . . . . . 39
4.4.3 Virtual Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.4.4 Compton Scattering and Electron-Positron Annihilation . . . . . . . 41
4.5 Relating Scattering and Decay . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.6 Higher Order Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

5 Rates of Interactions 45
5.1 The Rates of Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5.1.1 Fermi’s Golden Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5.1.2 The Matrix Element . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
5.1.3 The Density of States . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.2 Calculation of Decay Rates - details non-assessable . . . . . . . . . . . . . . 47
5.2.1 Non-relativistic formulation . . . . . . . . . . . . . . . . . . . . . . . 48
5.2.2 Normalisation of the wavefunction revisited . . . . . . . . . . . . . . 50
5.2.3 Relativistic formulation . . . . . . . . . . . . . . . . . . . . . . . . . . 51

iii
 5.2.4 Expression for the two-body decay rate . . . . . . . . . . . . . . . . . 51
5.3 Branching Fractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

6 The Electromagnetic and Strong Interactions 53

6.1 Coupling Constants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
6.2 The Standard Model Vertices . . . . . . . . . . . . . . . . . . . . . . . . . . 54
6.3 The Electromagnetic Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
6.3.1 Examples of EM Interactions . . . . . . . . . . . . . . . . . . . . . . 56
6.3.2 Suppression of Higher Order Processes . . . . . . . . . . . . . . . . . 57
6.4 The Strong Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
6.4.1 The Yukawa Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . 58
6.4.2 What holds the proton and neutron together? . . . . . . . . . . . . . 58
6.4.3 Properties of the Strong Interaction . . . . . . . . . . . . . . . . . . . 59
6.4.4 The Strong versus EM Interaction . . . . . . . . . . . . . . . . . . . . 60
6.4.5 Examples of Strong Interactions . . . . . . . . . . . . . . . . . . . . . 61
6.4.6 But what about Yukawa? . . . . . . . . . . . . . . . . . . . . . . . . 61

7 The Weak Force 63

7.1 Beta Decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
7.2 The Weak Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
7.2.1 Properties of the Weak Interaction . . . . . . . . . . . . . . . . . . . 64
7.2.2 The Lepton-W ± Vertices . . . . . . . . . . . . . . . . . . . . . . . . . 64
7.2.3 Quark-Lepton Symmetry . . . . . . . . . . . . . . . . . . . . . . . . . 65
7.2.4 β Decay Revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
7.2.5 Weak Coupling Constants . . . . . . . . . . . . . . . . . . . . . . . . 66
7.2.6 The Cabibbo Angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
7.2.7 The Quark and Lepton-Z 0 Vertices . . . . . . . . . . . . . . . . . . . 69
7.2.8 A Further Example of Weak Decay . . . . . . . . . . . . . . . . . . . 70
7.3 General Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

iv
 7.4 Electroweak Unification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
7.5 A Brief Aside - Gravity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

8 The Higgs Boson 73

8.1 Electroweak Unification and the Higgs Field . . . . . . . . . . . . . . . . . . 73
8.2 Spontaneous Symmetry Breaking (SSB) . . . . . . . . . . . . . . . . . . . . 74
8.2.1 SSB in the Standard Model . . . . . . . . . . . . . . . . . . . . . . . 75
8.3 Higgs Couplings to other particles . . . . . . . . . . . . . . . . . . . . . . . . 75
8.4 How the Higgs boson is made at the LHC . . . . . . . . . . . . . . . . . . . 77
8.5 How do we detect the Higgs boson once produced? . . . . . . . . . . . . . . 78
8.6 The Discovery of the Higgs Boson . . . . . . . . . . . . . . . . . . . . . . . . 79
8.6.1 Higgs decaying to two photons . . . . . . . . . . . . . . . . . . . . . . 79
8.6.2 Higgs Decaying to two Z bosons . . . . . . . . . . . . . . . . . . . . . 80
8.6.3 Is this a Standard Model Higgs? . . . . . . . . . . . . . . . . . . . . . 80
8.7 The Higgs beyond the LHC? . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

9 Detecting Particles 84
9.1 The LHC and the ATLAS Experiment . . . . . . . . . . . . . . . . . . . . . 85
9.2 Some Elements of Particle Detectors . . . . . . . . . . . . . . . . . . . . . . 86
9.3 Reconstructing Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
9.4 Hands-on exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

10 Symmetry 89
10.1 Symmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
10.2 Noether’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
10.3 Conservation Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
10.3.1 Energy and Momentum . . . . . . . . . . . . . . . . . . . . . . . . . 91
10.3.2 Exercises on Energy and Momentum Conservation . . . . . . . . . . . 91
10.3.3 Angular Momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

v
 10.3.4 Exercises on Angular Momentum Conservation . . . . . . . . . . . . 92

11 More on Symmetry 94
11.1 More Conservation Laws . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
11.1.1 Electric Charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
11.1.2 Baryon Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
11.1.3 Lepton Number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
11.1.4 Quark Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

12 Discrete Symmetries 99
12.1 Charge Conjugation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
12.1.1 Decay of the neutral pion . . . . . . . . . . . . . . . . . . . . . . . . 100
12.2 Parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
12.2.1 Intrinsic Parity of Fermions . . . . . . . . . . . . . . . . . . . . . . . 101
12.2.2 Meson and Baryon Parity . . . . . . . . . . . . . . . . . . . . . . . . 101
12.3 Time Reversal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
12.4 Are these discrete symmetries universal? . . . . . . . . . . . . . . . . . . . . 103
12.4.1 The Fall of Parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

13 The Quark Model 105

13.1 Strong Isopin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
13.2 Strong Hypercharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
13.3 Hadrons and the Quark Hypothesis . . . . . . . . . . . . . . . . . . . . . . . 107

14 More on the Quark Model 111

14.1 Symmetry of the Baryon States . . . . . . . . . . . . . . . . . . . . . . . . . 111
14.1.1 A Problem with this Symmetry . . . . . . . . . . . . . . . . . . . . . 112
14.2 Applications of Strong Isospin . . . . . . . . . . . . . . . . . . . . . . . . . . 113

15 QCD 115

vi
 15.1 Colour and The Strong Force . . . . . . . . . . . . . . . . . . . . . . . . . . 115
15.1.1 The ∆++ State Revisited . . . . . . . . . . . . . . . . . . . . . . . . . 116
15.2 Colour and Gluons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
15.3 How many Gluons? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

16 More on QCD 120

16.1 Evidence for Colour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
16.2 The Coupling Strength of the Strong Interaction . . . . . . . . . . . . . . . . 122
16.3 Quark and Gluon Jets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
16.3.1 Separating Quarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
16.3.2 Jets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
16.4 Quark-Gluon Plasma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
16.5 What really is a Proton? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
16.6 A Brief Aside: Group Theory . . . . . . . . . . . . . . . . . . . . . . . . . . 128
16.6.1 Gluons and Group Theory . . . . . . . . . . . . . . . . . . . . . . . . 128
16.6.2 The Quark Model and Group Theory . . . . . . . . . . . . . . . . . . 129
16.6.3 The Weak Force and Group Theory . . . . . . . . . . . . . . . . . . . 129

17 Particle Mixing, CP Symmetry and CP Violation 130

17.1 Symmetry Violation: Communicating with Aliens . . . . . . . . . . . . . . . 130
17.2 CP Symmetry? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
17.3 K 0 Mesons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
17.3.1 KS Regeneration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
17.4 CP Violation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
17.5 Talking to aliens revisited . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
17.6 CP Violation – a summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

18 Neutrinos 138
18.1 The Definition of Helicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

vii
 18.2 Photons and Helicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
18.3 Neutrinos and Helicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
18.4 C, P , CP and Neutrinos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
18.5 Detecting Neutrinos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
18.5.1 Example of Neutrino Detection - Super-Kamiokande . . . . . . . . . 142
18.6 Sources of Neutrinos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
18.6.1 Accelerators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
18.6.2 The Sun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
18.6.3 The Earth’s Atmosphere . . . . . . . . . . . . . . . . . . . . . . . . . 145
18.6.4 Nuclear Reactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
18.7 The Solar Neutrino Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
18.8 Neutrino Mixing (or oscillations) . . . . . . . . . . . . . . . . . . . . . . . . 147
18.9 Atmospheric Neutrino Mixing . . . . . . . . . . . . . . . . . . . . . . . . . . 149
18.10Solar Neutrino Oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

19 Beyond the Standard Model 153

19.1 Lots of Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
19.2 The Precision of Standard Model Predictions . . . . . . . . . . . . . . . . . . 154
19.2.1 The g-factor anomaly for the electron . . . . . . . . . . . . . . . . . . 155
19.2.2 The g-factor anomaly for the muon . . . . . . . . . . . . . . . . . . . 155
19.3 Beyond The Standard Model: Unification . . . . . . . . . . . . . . . . . . . . 156
19.4 Dark Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
19.4.1 Mass/Energy Content of Universe . . . . . . . . . . . . . . . . . . . . 157
19.4.2 Evidence for Dark Matter . . . . . . . . . . . . . . . . . . . . . . . . 157
19.5 What is the Dark Matter? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
19.6 Unification and the Standard Model . . . . . . . . . . . . . . . . . . . . . . . 159
19.7 Supersymmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
19.8 Unification, Supersymmetry, and Dark Matter . . . . . . . . . . . . . . . . . 161
19.9 Detecting WIMPs Terrestrially . . . . . . . . . . . . . . . . . . . . . . . . . 161

viii
19.9.1 Example: The CDMS experiment . . . . . . . . . . . . . . . . . . . . 162
19.9.2 A dark matter experiment in Australia . . . . . . . . . . . . . . . . . 162
19.9.3 Example: Detecting WIMPs with Accelerators . . . . . . . . . . . . . 162

1
Introduction

Outline

One way to state the goal of Particle Physics, also known as High Energy Physics, is to say
that it is to uncover the basic constituents of matter, examining their fundamental properties
and interactions, including their role in the early Universe.
The goal of these lectures therefore is to cover the basic constituents of matter, such as
quarks and leptons, examining their fundamental properties and interactions. Some brief
discussion of limitations and extensions to the currently accepted Standard Model of Particle
Physics will also be given.

Suggested Reading

We do not have a set text for this module. Whilst there are many, many textbooks on
particle physics, finding one which is at the right level and from which a suitable length
course can be derived is not easy. Rather the module is defined by the material in these
notes and in the lecture slides, and to complement this we provide a list of resources which
may be of use. This is by no means an exhaustive list.

• The Standard Model in a Nutshell, by Dave Goldberg, Princeton University Press,

2017. A recent book, which we have not yet formed a strong opinion on, but looks
good at a first look.

• Modern Particle Physics, by Mark Thomson, Cambridge University Press, 2013. A

very good and fairly recent book, though overall pitched at a level higher than in this
course.

• Particle Physics, by Duncan Carlsmith, Pearson, 2013. Another fairly recent book at
a roughly similar level to Thomson.

• Introduction to Elementary Particles, 2nd, Revised Edition, by David Griffiths, Wiley-

VCH, 2008. Griffiths has written textbooks on several topics in physics, and many find
his style very accessible.

2
 • Particle Physics, 3rd Edition, by B. Martin and G. Shaw, Wiley, 2008. A text perhaps
closer to the level of this course than the previous, newer texts.

• Introduction to High Energy Physics, 4th Edition, by D. H. Perkins, Cambridge Uni-

versity Press, 2000. Now getting somewhat dated, but a text at roughly the level of
this course.

• University Physics, 13th Edition, by H. D. Young, R. A. Freedman and A. L. Ford,

Pearson, 2011. The latter sections of the general undergraduate text used here at the
University of Sydney includes covers some aspects of high energy physics.

Assumed Knowledge

The current underlying theory for high energy particle physics, the Standard Model, is math-
ematically precisely defined in the language of Quantum Field Theory (QFT). We will not
deal with QFT in this module - an introductory course is offered in Honours, but never-
theless, as indicated in Figure 1, QFT and therefore particle physics deals with relativistic
situations, and is a microscopic theory of nature, dealing with individual particles and their
interactions where the uncertainty principle comes into play. It is therefore underpinned
both by special relativity and quantum mechanics.

Figure 1: The road from classical physics to quantum field theory. Figure taken from lectures
by M. Thomson, Cambridge Part III.

The following will therefore for the most part be assumed knowledge for this module:

• A basic understanding of special relativity, as for example obtained from the module
in PHYS2913/2013.

• The third year Quantum Physics module which is being completed in parallel with this
module. In particular, several concepts from quantum physics are required: bosons and
fermions, de Broglie wavelength, and basic notions of spin and angular momentum.

3
Having said that these topics are assumed knowledge, some aspects will be reviewed in a
particle physics context.

Basic Building Blocks

The material in the remainder of this introduction will not be covered directly in lectures.
It serves as a type of historical survey of the subject, and some students will have a degree
of familiarity with the material from high school physics or earlier undergraduate courses.
It is recommended to read this material prior to the first lecture.

The Atom

The idea that matter may be broken down into indivisible building blocks is of course not
new, and typically is traced back to the Ancient Greeks, such as Leucippus and Democritus
∼430BC.

The Electron

The concept of a light (compared to an atom), negatively charged particle being the con-
stituent of “cathode rays” is generally attributed to J.J. Thomson around 1896, who per-
formed experiments with these rays. Thomson received the 1906 Nobel Prize.
The charge of the electron was precisely determined by R.A. Millikan in his famous oil-drop
experiment. He received the Nobel Prize in 1923.

The Nucleus and the Proton

It was Ernest Rutherford and co-workers who demonstrated experimentally that the atom
contained a small positively charged nucleus, rather than having positive charge more dif-
fusely spread throughout as had been previously postulated by Thomson. The experiment
scattered alpha particles from gold atoms, with the alpha particles suffering unexpectedly
large deflections, as illustrated in Figure 2.
It was also Rutherford who coined the term proton, as the particle comprising a hydrogen
nucleus, when his experiments indicated that such an object could be extracted from the
nuclei of heavier atoms and thus could be considered as a building block of atomic nuclei.

4
Figure 2: The Rutherford experiment, where the scattering of alpha particles was consistent
with (b) and not (a). Figure from Young and Freedman.

The Neutron

A suitable model for the atom at the time seemed to be a postively-charged nucleus of
protons and a surrounding cloud of electrons. Rutherford nevertheless postulated the idea
of a neutron to explain the disparity between the atomic number and the atomic mass of
atoms. The neutron as a proton and an orbiting electron was one hypothesis.
Further alpha particle scattering experiments showed that in certain cases a penetrating
form of radiation was emitted. James Chadwick perfomed experiments which showed that
this was electromagnetic in nature but was consistent with a neutral particle of around the
mass of the proton.
The defining set of reactions was the following:
4
2
He + 94 Be → 12
6
C + 10 n

1
0
n + 10
5
B → 7
3
Li + 42 He

in which the neutron emitted in the first reaction was subsequently able to induce the second.
Chadwick received the 1935 Nobel Prize for his work.

The Photon

The photon as the quantum of electromagnetic radiation was postulated by Einstein in one
of his famous 1905 papers, in this case explaining the photoelectric effect. Einstein received
the 1921 Nobel Prize for this.
With the photon on board, a rather complete picture of fundamental particles able to explain
the basic properties of atoms known at the time seemed to be in hand – a positively-charged

5
nucleus consisting of protons and neutrons surrounded by electrons, with the electrons bound
by the electromagnetic force which is mediated by photons. This does beg the question as
to what holds the positively-charged nucleus together in such a small volume, given that the
protons should repel each other.

The Positron

Quantum mechanics was further developed in the 1920s, and in particular Paul Dirac at-
tempted to find an equation to describe the electron which was consistent with special
relativity. He succeeded spectacularly, receiving the 1933 Nobel Prize, but in the process
of doing so not only managed to explain the two “spin” states of the electron, but also
found that his equation had twice as many solutions as he expected, and that in particular
the extra solutions seemed to describe particles in negative energy states. His attempts to
interpret what was going on led to the revolutionary concept first of a sea of negative energy
states and “holes”, as illustrated in Figure 3, and subsequently to the more modern way
of thinking in terms of antiparticles, with the holes interpretated as real, positive-energy
particles known as positrons.

Figure 3: Dirac’s positive and negative energy states (a); the idea of pair production (b);
and electron-positron annihilation (c). Figure from Young and Freedman.

If positrons were real they should be observable, and indeed C.D. Anderson detected them
when examining cosmic rays with a cloud chamber in 1932, receiving the 1936 Nobel Prize.
The cloud chamber image of a positron as observed by Anderson is shown in Figure 4. In
this figure, the positron can be seen bending in the magnetic field in which the chamber is
embedded. The curvature is less below the lead absorber dividing the top and bottom parts
of the chamber. From this it is known that the particle entered from the bottom, and then
knowing this, the charge can be inferred by the direction of bending.

6
Figure 4: The positron discovered in Wilson’s cloud chamber. Image from Physical Review
43 (1933) 491.

The Neutrino and Anti-neutrino

Nuclear beta decay was known about at around the same time as Dirac’s work, and studies
of it produced an anomaly. It was assumed that a nucleus transformed itself into another
nucleus with the emission of an electron. If this was the case, however, the electrons from
this process should all have the same energy. In fact, they produced a spectrum of energies
up to a given maximum energy, in apparent violation of the law of conservation of energy.
Wolfgang Pauli postulated the existence of the neutrino in 1930 as a way of explaining
the anomaly and rescuing the law of conservation of energy. The neutrino and electron
would share the energy between them. In 1933 Fermi then produced a theory of beta
decay building on Dirac’s work, and subsequent calculations by Bethe and Peierls of the
probabibilty of neutrinos scattering off other matter and thus being observed gave such low
values that it was assumed that neutrinos would never be observed. This assumption proved
to be incorrect, though it took over two decades before neutrinos were observed by Reines
and Cowan in 1956 by placing a suitable detector close to a nuclear reactor. Reines received
the Nobel Prize in 1995 - Cowan had died in the intervening years and the prize cannot be
awarded posthumously.
The particle emitted with an electron in nuclear beta decay is actually an antineutrino. There
are some forms of beta decay in which a positron is emitted, in which case the accompanying
particle is indeed a neutrino.

The Muon

In 1937 a particle with mass approximately 200 times that of the electron was discovered in
cosmic rays. As we will see later, at around this time a particle of approximately this mass
had been postulated as a carrier of the strong force between protons and neutrons (nucleons)
in a nucleus, and at first it was thought that this was the particle.

7
Over the next decade, studies of the properties of the new particle showed that it could
not be the strong-force carrying particle. Rather it was realised this particle was like a big
brother of the electron and positron, and it was named the muon (µ± ).
Why the electron and positron should have a higher mass replica was not clear, leading I.I.
Rabi to ask “Who ordered that?”.

The Particle Explosion

Figure 5: Bubble chamber image of particle interactions.

https://hst-archive.web.cern.ch/archiv/HST2005/bubble_chambers/BCwebsite/
gallery/gal3_neutralkaon1.htm

The 1940s and 1950s saw an explosion in the number of new particles discovered, examples
being the charged pions π ± (1947); charged kaons K± (1949); neutral pion π 0 (1950); Λ0 , K0
(1951); and ∆0 (1952).
Much of this rapid rise in discovery rates can be traced to the invention of the bubble chamber
in 1952 by Donald Glaser, for which he received the 1960 Nobel Prize. An example bubble
chamber image is shown in Figure 5.
An explanation for why all of these new particle should exist was lacking at the time, and
was disturbing for the scientific community. It lead to comments such as this from Enrico
Fermi: “If I could remember the names of all these particles, I’d be a botanist”.

Electron-Proton and Deep Inelastic Scattering

In 1953 Robert Hofstadter, using a particle accelerator, scattered electrons off protons as

in Figure 6, and revealed that the proton consisted of an extended charge distribution. He

8
 Figure 6: The scattering of electrons from protons.

received the 1961 Nobel Prize. The same proved to be the case for the neutron.
This then opened up the question as to whether the proton and neutron are really funda-
mental particles, or perhaps could be composites of other more basic particles. With the
explosion of the number of types of particles also needing an explanation, in 1964 Murray
Gell-Mann and George Zweig independently proposed that the majority of particles are com-
posite, made from a small number of fundamental building blocks. These building blocks we
now know as quarks, and Gell-Mann received the 1969 Nobel Prize for this work.
For a time it was not accepted that quarks were real particles rather than a mathematical
device for classifying particle states. However, in 1967, scattering of electrons from protons
at higher energies, so-called deep inelastic scattering in which the proton is broken up, see
Figure 7, revealed that the proton appeared to consist of point-like constituents, named
partons at the time by Richard Feynman but subsequently coming to be identified with
quarks. J. Friedman, H. Kendall and R. Taylor received the Nobel Prize in 1990 for these
experiments.

Figure 7: Deep inelastic scattering of electrons from a proton.

9
Fundamental Particles

Looking ahead, further studies led to the picture in which the fundamental building blocks
of the so-called Standard Model of Particle Physics are those shown in Figure 8.

Figure 8: The currently accepted fundamental building blocks. Taken from http://www.
particleadventure.org/the-role-of-the-higgs-boson.html

Drawing out some of the properties of these particles and the relationships between them
will form a significant part of the remainder of this module.

An Aside on Units for Particle Masses

In particle physics, particle masses are generally expressed in units of GeV/c2 .

By way of example, in SI units the proton has a mass of 1.67 × 10−27 kg. Using E = mc2 ,
this corresponds to an energy of E = 1.50 × 10−10 J. As 1 eV = 1.602 × 10−19 J, the mass of
the proton corresponds to an energy E = 0.938 GeV, and so we arrive at

mproton = 0.938 GeV/c2

Expressing masses (and momenta in fact) in terms of energy equivalent units is a convenience
used universally in particle physics. There is seldom a need to substitute explicitly for the
speed of light c.

10
Chapter 1

The Particle Zoo

1.1 Fundamental Particles

The fundamental particles were shown pictorially in Figure 8. They can basically be broken
down into the quarks and the leptons, which form the building blocks of matter, the force
carrying particles, which are termed gauge bosons, and the Higgs boson, which plays a special
role which we will return to later in the module. Note that each of the quarks and leptons
mentioned in the figure also has a corresponding antiparticle.
We will consider some properties of the leptons, quarks and bosons in turn.

1.1.1 Leptons

Leptons are spin 21 ℏ fermions. To our present level of experimental probing, they are point-
like particles at least down to dimensions of (< 10−18 m), and they are treated mathemat-
ically as point-like particles in the Standard Model. The defining characteristic of leptons
is that they are blind to the strong force. Leptons can be observed as free particles. Some
properties of leptons are given in Table 1.1.

Particle Lifetime (s) Mass (GeV/c2 )

e+ , e − stable 5.11 × 10−4
−
+
µ ,µ 2.2 × 10−6 0.106
−
+
τ ,τ 3.0 × 10−11 1.784
νe , ν̄e stable < 2.0 × 10−6
νµ , ν̄µ stable < 1.9 × 10−4
ντ , ν̄τ stable < 1.8 × 10−2

Table 1.1: Some properties of the leptons.

11
1.1.2 Quarks

Quarks are spin 12 ℏ fermions. Like the leptons, to our present level of experimental probing,
they are point-like particles at least down to dimensions of (< 10−18 m), and they are
also treated mathematically as point-like particles in the Standard Model. The defining
characteristic of quarks is that they do feel the strong force. They are fractionally-charged
particles and come in three colours, where colour is a type of distinct charge to electric
charge, related to the strong force. Quarks have never been observed as free particles. Some
properties of quarks are given in Table 1.2. The masses are approximate for the reason that
they cannot be definitively measured since quarks have never been observed as observed as
free particles and are not expected to be under normal conditions.

Particle Charge (e) Mass (GeV/c2 )

d − 13 ∼ 6 × 10−4
u + 23 ∼ 3 × 10−4
s − 13 ∼ 0.1
c + 32 ∼ 1.3
b − 13 ∼ 4.2
t + 32 ∼ 174

Table 1.2: Some properties of the quarks.

Anti-quarks have opposite charge to the quarks – for example d̄ is + 31 e.

1.1.3 Gauge Bosons

These propagate the fundamental forces, which in the Standard Model refers to the electro-
magnetic (EM), strong and weak forces. Gravity is not described by the Standard Model,
for the reason that the Standard Model is a quantum field theory, and gravity cannot cur-
rently be described by a viable quantum theory. Its hypothesised force-carrying boson, the
graviton, is included in Table 1.3 below for completeness.
An obvious question which might be raised upon inspecting Table 1.3 is why are some of the
gauge bosons stable and massless, and why are some short-lived and massive? This will be

Particle Lifetime (s) Mass (GeV/c2 ) Spin Force

γ (photon) stable massless 1ℏ EM
g (gluon) stable massless 1ℏ Strong
Z0 2.7 × 10−25 91.19 1ℏ Weak
W± 3.1 × 10−25 80.42 1ℏ Weak
graviton stable massless 2ℏ Gravity

Table 1.3: Some properties of the force-carrying particles.

12
taken up later.

1.1.4 Higgs Boson

The Higgs boson was for a long time the “missing piece” of the Standard Model, a hypo-
thetical heavy, spinless, neutral boson which was believed to be responsible for providing a
mechanism via which other fundamental particles acquire mass. It was discovered at CERN
in 2012 by the ATLAS and CMS experiments at the Large Hadron Collider, and subsequent
studies so far seem to confirm that it has the properties that were predicted for it. It is
the second heaviest known fundamental particle (after the top quark), with a mass of 125
GeV/c2 .

1.2 Composite Particles: Hadrons

Many of the plethora of particles discovered in the particle discovery days of the 1950s and
1960s are composite particles made of quarks and antiquarks, which collectively are referred
to as hadrons.
Hadrons are characterised by a quantum number known as baryon number, defined as

1
B = (nq − nq̄ )
3

where nq specifies the number of quarks and nq̄ the number of antiquarks of which the
particle is composed.
The different classifications of hadrons depending on their quark content are shown in Ta-
ble 1.4.

Quarks B Name
qqq +1 Baryon
q̄ q̄ q̄ −1 Anti-Baryon
q q̄ 0 Meson
q q̄q q̄ 0 Tetraquark?
qqq q̄q/q̄ q̄ q̄ q̄q +1/−1 Pentaquark?

Table 1.4: Classification of hadrons by quark content.

B can only take the values −1, 0 or +1, as a consequence of the allowed combinations of
quarks and antiquarks, and the assignments of baryon number +1/3 to quarks and −1/3 to
antiquarks.

13
The question marks against the combinations in the last two rows of the table indicate that
these are hypothetical combinations of quarks and antiquarks for which there is not firm
experimental evidence. We will say a little more about this below.

1.2.1 Mesons (q q̄)

Mesons are hadrons which have integral spin, and are therefore by definition bosons. They
have baryon number equal to zero, and feel all forces. Some examples of mesons, along with
some of their properties, are given in Table 1.5.

Particle Quarks Mass (GeV/c2 ) Lifetime (s)

π+, π− ¯ ūd
ud, 0.139 2.6 × 10−8
K +, K − us̄, ūs 0.494 1.3 × 10−8
J/Ψ cc̄ 3.097 7.7 × 10−21
Υ bb̄ 9.460 1.3 × 10−20

Table 1.5: Some examples of mesons, with associated properties.

In terms of quark content, in a quantum mechanical sense some mesons are quantum super-
positions of quark-antiquark states. The neutral pion, for example, can be represented as
the superposition of uū and dd̄ given below, which has the consequence that it can interact
with other particles either as though it has quark content uū or quark content dd̄.

1
|π 0 ⟩ = √ (|uū⟩ − |dd⟩)
¯
2

The omega meson, on the other hand, is the following superposition:

1
|ω 0 ⟩ = √ (|uū⟩ + |dd⟩)
¯
2

1.2.2 Baryons (qqq/q̄ q̄ q̄)

Baryons are hadrons which have half-integral spin, and are therefore by definition fermions.
They have baryon number ±1, and feel all forces. Some examples of baryons, along with
some of their properties, are given in Table 1.6.

1.2.3 Excited States

There are only so many qqq/q̄ q̄ q̄ and q q̄ configurations possible. Quarks in a particle occupy
energy levels, in an analogous sense to electrons in an atom. This means that particles

14
 Particle Quarks Mass (GeV/c2 ) Lifetime (s)
p uud 0.938 stable
n ddu 0.940 920
Λ uds 1.116 2.6 × 10−10
∆ ++
uuu 1.232 6.0 × 10−24
Ξo uss 1.315 2.9 × 10−10
−
Ω sss 1.672 8.2 × 10−11

Table 1.6: Some examples of baryons, with associated properties.

can exist in excited states. In particle physics these states are generally viewed as distinct
particles, labelled by their mass. Denoting a nucleon (proton or neutron) by n, then examples
of excited nucleons are listed in Table 1.7. Each looks like a more massive version of the
ground state particle, and in general the excited states are highly unstable.

n n(1650) n(1710) n(2000) n(2190)

n(1440) n(1675) n(1720) n(2080) n(2200)
n(1520) n(1680) n(1900) n(2090) n(2250)
n(1535) n(1700) n(1990) n(2100) ...

Table 1.7: Excited nucleons. http://pdg.lbl.gov/. Masses in brackets are in MeV/c2

1.2.4 Hadron Spin

Is it just the mixture of quark types that defines a particle? Clearly that is not the case, as
illustrated in Table 1.8.

Quarks Particle Particle

uud p(938) ∆+ (1232)
udd n(939) ∆o (1232)

Table 1.8: Examples of distinct baryons sharing the same quark content.

Particles can have the same mix of quarks, but quite different masses. This is due to the
relative spin orientations of quarks and antiquarks, which are themselves spin 1/2 objects.
Hadrons can be found to form families linked by their spin but differentiated by other
quantum numbers. Examples of these families can be seen in Figure 1.1, and we will revisit
this when discussing the quark model in more detail.

15
Figure 1.1: Example baryon octet and decuplet of particles. The numbers in blue give
the approximate masses in MeV/c2 . The axes (Iz , Y ) are quantum numbers that will be
explained when treating the quark model.

1.2.5 Exotic Hadrons?

The majority of identified hadrons have been attributed to be qqq, q̄ q̄ q̄ or q q̄ states. As was
indicated in Table 1.4, there is nothing precluding more exotic combinations.
The Belle experiment in Japan, for example, has observed a particle known as the Z(4430)
which could be a candidate for a tetraquark state. How it is produced and decays is shown
in Figure 1.2.

Figure 1.2: Possible mechanism for production of a tetraquark state. Image: KEK.

Until very recently, the weight of evidence has been against pentaquarks having been ob-
served in experiments, although a number of claims had been made. However, in August
2015 the LHCb experiment at CERN’s Large Hadron Collider (LHC) announced compelling
evidence for a pentaquark state seen in the decay products of a particle known as a Λ0b ,
Λ0b → Pc+ + K − , where the pentaquark Pc+ decays to a bound state of a charm and anti-
charm quark (known as a J/ψ) and a proton. It can be seen as a bump in the graph in
Figure 1.3. The press release image from CERN is shown in Figure 1.4.

16
It is even possible to envisage bound states containing gluons. Examples are given in the
last row in Figure 1.5, which shows all of the potential quark/antiquark/gluon combination
states commonly considered.
A more detailed understanding of these strongly bound states comes from that part of the
Standard Model known as Quantum Chromodynamics, or QCD. We will touch on some of
this later in the module.

17
Figure 1.3: LHCb data revealing the production of a pentaquark state. https://
cerncourier.com/a/lhcb-reports-observation-of-pentaquarks/

Figure 1.4: Possible mechanism for production of a pentaquark state. Image: https://www.
arabnews.com/science-technology/news/776441

18
Figure 1.5: Normal and exotic composite particles. Image: https://physicsworld.com/
a/evidence-grows-for-tetraquarks/

19
Chapter 2

Angular Momentum and Particle

Physics

In this chapter we briefly review angular momentum in classical and quantum physics, and
discuss its relevance in particle physics, particularly in terms of the intrinsic angular mo-
mentum that a particle possesses, which is denoted as its spin.

2.1 Angular Momentum in Classical Mechanics

In classical mechanics, the angular momentum is given by the cross product of the position
vector of an object (with respect to some chosen origin) and its momentum vector.
L=r×p (2.1)
This is illustrated in Figure 2.1 for the case of an object undergoing circular motion about
some axis.

Figure 2.1: The angular momentum for an object undergoing circular motion.

Classically, the angular momentum vector L can be of arbitrary length. Also, all three
components of L can be determined at one time. In the absence of a net external torque,
the angular momentum of a system is a fixed vector - “angular momentum is conserved”.

20
If we consider a charged particle undergoing circular motion, for example an electron in a
classical orbit around a nucleus as in Figure 2.2, then we can think of the charge forming a
current I. If the area of the loop is A we can define a magnetic moment which has magnitude

|µ| = IA (2.2)

Figure 2.2: An electron in a classical orbit around a nucleus.Figure from Young and Freed-
man.

It should be noted that orbital angular momentum is a property of an extended system (of
more than one object), in this example the nucleus and electron.

2.2 Angular Momentum in Quantum Mechanics

In quantum mechanics the angular momentum for a particle in an orbit cannot take arbitrary
values, it is quantised. The “length” of the orbital angular momentum vector is given by
√
|L| = ℓ(ℓ + 1)ℏ (2.3)

More precisely we say that the operator L̂2 takes values ℓ(ℓ+1)ℏ where ℓ is a positive integer.
At the same time, the measurement of the projection of the orbital angular momentum onto
a chosenaxis will only take one of the 2ℓ + 1 values Lz ϵ [−ℓℏ, ℓℏ] where ℓ is a non-negative
integer. This is illustrated for the case where ℓ = 2 in Figure 2.3.
It is a feature of quantum mechanics that one cannot make measurements of the eigenvalues
of the operators L̂x , L̂y and L̂z simultaneously, only of, say, L̂2 and L̂z .

2.3 Spin in Quantum Mechanics

In quantum mechanics, spin is the intrinsic angular momentum acribed to a single particle.
The concept of spin arose as a consequence of Dirac’s seeking a relativistic version of the
Schrŏdinger equation to describe the electron. The Dirac equation has four solutions, which

21
Figure 2.3: Geometrical illustration of orbital angular momentum quantisation for the case
ℓ = 2. Figure from Young and Freedman.

can be interpreted as two particle solutions, spin up and spin down, and two antiparticle
solutions, spin up and spin down.
An electron or positron can only have a total spin angular momentum of (1/2)ℏ, with
√ ( ) √
1 1 3
|S| = +1 ℏ= ℏ (2.4)
2 2 2

and can be in two possible spin states (“up” or “down”) corresponding to eigenvalues
1
Sz = ± ℏ (2.5)
2
which give the projections of the spin onto the z-axis. This is illustrated in Figure 2.4.

2.3.1 What is spinning?

It is tempting to adopt the classical picture of spin as illustrated in Figure 2.5, but this is
incorrect. The electron is pointlike as far as we can resolve experimentally, and is strictly
pointlike in a mathematical sense in current accepted theories.
If an electron was a tiny rotating sphere, it could not rotate fast enough to provide the
amount of angular momentum represented by the spin. Rather, spin is to be interpreted as
an intrinsic, vector quantity associated with a particle, analogous to the way that electric
charge is an intrinsic scalar quantity associated with a particle.

22
 Figure 2.4: Spin half projections.

Figure 2.5: How not to interpret spin.

2.4 Fermions and Bosons

All particles that we will meet can be characterised by their spin, which constitutes one of
the quantum numbers that they possess. As should be familiar from quantum mechanics,
particles can be classified into two groupings based on their spins, fermions and bosons.
Fermions carry half integral spin S = 1/2, 3/2 and so on in units of ℏ. Examples are electrons,
protons and quarks. Fermions obey the Pauli-exclusion principle and are characterised by
antisymmetric wavefunctions.
Bosons carry integer spin S = 0ℏ, 1ℏ, 2ℏ and up. Examples are photons (also known as γs),
pions, kaons and the Higgs boson. Bosons can exist in the same quantum state as other
bosons in multi-particle systems. They are characterised by symmetric wavefunctions.

23
Figure 2.6: Combining two spin 1/2 vectors (blue) to form a spin 1 vector (black) in the
three possible spin projections.

2.5 Combining Angular Momentum Vectors

Angular momentum vectors can be combined using rules derived from quantum mechanics
and the theory of groups to form systems with different total angular momentum.
An example of combining two spin (1/2ℏ) vectors to form a total angular momentum 1ℏ
state (a triplet state) is shown in Figure 2.6.

Figure 2.7: The triplet (1ℏ) and singlet (0ℏ) states that can be formed by combining a
pair of spin (1/2)ℏ states. From Eisberg and Resnick, Quantum Physics of Atoms,
Molecules, Solids, Nuclei and Particles, 2nd ed.

In fact it is also possible for two spin 1/2 vectors to form a state with zero total angular
momentum (a so-called singlet state). All four possibilities are illustrated in Figure 2.7.

24
2.6 Total Angular Momentum

If a system is composed of a number of particles, labelled 1, 2, 3 . . . and there is orbital

angular momentum associated with each pair of particles (1 ↔ 2) (1 ↔ 3) etc, then the
total angular momentum is given by

J = L12 + L13 + · · · + S1 + S2 + S3 + . . . (2.6)

i.e. orbital and spin angular momenta can be combined.

2.7 The spin of a proton?

The proton is a “spin 1/2” object, but it can be viewed in two ways knowing that it is not
a fundamental particle but rather a composite object containing quarks and gluons.

2.7.1 Proton as a single unresolved object

If we view the proton as it was historically viewed for a long time, up to the 1960s or
thereabouts, as a single object with no resolvable substructure, then we would just describe
it as having an intrinsic spin of S = 1/2ℏ, as is the case for the electron. There can be no
orbital angular momentum ascribed to an object with no constituent parts. Classically one
would say that for there to be orbital angular momentum in a system, there needs to be at
least two subsystems in motion with respect to each other.

2.7.2 Proton as a composite object

If we know on the other hand that the proton is composed of more fundamental objects, the
quarks and gluons, then we have to think in terms of the intrinsic spins of these particles, as
well as taking into account any orbital angular momentum between them. The total angular
momentum would be constructed using an equation of the form of equation 2.6. We would
then associate this total angular momentum J of the composite object with its spin. This
is why you may see both the symbols S and J used for spin in the literature.

2.8 Spin is not conserved!

In a classical physical system that experiences no external torques, total angular momentum
is conserved, Indeed we will discuss this later. In a quantum system of particles which

25
is isolated from its surroundings, this is also true, total angular momentum is conserved.
However, spin is not! It is not the (vector) sum of intrinsic spins that is conserved, but
the vector sum of intrinsic spins of the constituents of the system plus the orbital angular
momenta between them i.e. the total angular momentum. Remember that J = L + S.

26
Chapter 3

Special Relativity and Particle

Physics

Particle Physics must deal, in most cases, with the situation in which particles are travelling
at or close to the speed of light, and therefore special relativity is relevant. In this chapter
we will discuss aspects of special relativity that are of most relevance to particle physics,
since these may not necessarily have been emphasised in an introductory course on special
relativity such as the one offered in second year physics.

3.1 Non-Relativistic Kinematics

Firstly let us recall some basic notions and formulae from non-relativistic kinematics, as met
in Newton’s laws of motion.
In Newtonian mechanics, the magnitude of the momentum of an object of mass m travelling
with a speed v is given by
p = mv (3.1)
The kinetic energy of the object is given by
1 p2
T = mv 2 = (3.2)
2 2m
This second relationship is illustrated in Figure 3.1. It follows that velocity, and hence
momentum, are not bounded in classical physics.

3.2 Relativistic Kinematics

In high energy physics, particles are usually travelling at relativistic velocities and so any
description of them should be consistent with special relativity.

27
Figure 3.1: The relationship between kinetic energy and momentum in non-relativistic me-
chanics.

In particular, in any inertial frame of reference there is a limiting velocity given by the speed
of light (c). The relationship between kinetic energy and velocity is illustrated in Figure 3.2,
comparing the relativistic and non-relativistic cases.

Figure 3.2: The relationship between kinetic energy and velocity in non-relativistic and
relativistic mechanics.

Looking again at the relationship between kinetic energy and momentum, and comparing
now the relativistic and non-relativistic cases, the situation is as in Figure 3.3. In particular,
at high momentum, the figure shows that the kinetic energy T becomes proportional to the
momentum p.

28
Figure 3.3: The relationship between kinetic energy and momentum in non-relativistic and
relativistic mechanics.

3.3 Total Energy

The total energy E of a particle has two components, the kinetic energy T and the rest mass
energy E = mo c2 , expressed as
E = T + mo c 2 (3.3)
Expressed in terms of momentum instead of kinetic energy, the total momentum is given by
the expression √
E = c p2 + (mo c)2 (3.4)
This relationship is shown graphically in Figure 3.4, for the case of two finite masses m1 and
m2 , for which m1 > m2 > 0. The so-called lightcone gives the the relationship for a massless
particle.

Figure 3.4: The relationship between total energy and momentum in relativistic mechanics.

29
3.4 4-Vectors

In special relativity, the time and space coordinates of a particle are “mixed” (unlike in New-
tonian theory where time is absolutely defined). We can define a 4-vector which encapsulates
the spatial and time components of, say, a particle in the following way:

X = (x0 , x1 , x2 , x3 ) = (ct, x, y, z) (3.5)

The length of a 4-vector is defined as the square root of the dot product, the dot product
being given by (note the minus signs, which reflect the fact that time and space, while linked,
are not treated exactly in the same way in special relativity)

X.X = c2 t2 − x2 − y 2 − z 2 (3.6)

This quantity is an invariant, or scalar, in special relativity.

4-vectors are not confined to expressing the time and spatial components of a particle. In
particular, by way of another example, the energy and momentum (3-)vector of a particle
form a 4-vector
P = (E/c, p) = (E/c, px , py , pz ) (3.7)
The dot product of any two 4-vectors is an invariant. In the case of the energy-momentum
4-vector
P.P = (E/c)(E/c) − p.p = (E/c)2 − p2 = m2o c2 (3.8)
where p = |p| and mo is the rest mass of the particle. While the observed energy and
momentum of a particle is dependent on the frame of reference in which the observing is
done, the rest mass of a particle is a scalar quantity independent of frame of reference,
defined as it is in terms of the length of a 4-vector.

3.5 Relativistic Transformations

For an observer in a frame of reference S ′ moving the z axis with velocity βc with respect
to the frame S, (−1 ≤ β ≤ 1), the energy and momentum in S and S ′ are related through
the Lorentz transformations
E′ = γ(E − βcpz ) E = γ(E ′ + βcp′ z )
p′ x = px px = p′ x
(3.9)
p′ y = py py = p′ y
p′ z = γ(pz − βE/c) pz = γ(pz ′ + βE ′ /c)

where √
1
γ= (3.10)
1 − β2
and β = v/c is the fractional velocity of the moving frame with respect to the speed of light.

30
3.6 Energy, Momentum and Mass

The formula resulting from the invariant length of the energy-momentum 4-vector is
√
E 2 = p2 c2 + m2o c4 E = c p2 + m2o c2 (3.11)

which is the generalization of Einstein’s famous E = mc2 (E = mo c2 in our notation) to

the case of a moving particle. Perhaps more accurately, Einstein’s famous formula should
be viewed as a special case of the more general formula above relating energy, momentum
and rest mass. It can be seen that a particle possesses energy (a contribution to the total
energy) both from possessing a rest mass, and from the fact that it is moving (has non-zero
momentum). In the case of a massless particle (mo = 0) only the second of these contributes,
and the formula reduces to E = pc which you may have met before in the context of a photon.
The following are very useful relationships between a particle’s energy, magnitude of mo-
mentum and mass in a frame which has a velocity βc.

E = γmo c2 p = γβmo c p = βE/c (3.12)

Note that the first two of these formulae only apply if |β| < 1, i.e. mo > 0.
Remember, in high energy physics, when we say mass we mean “rest mass” which is an
invariant quantity. The older way of discussing mass, which one still finds in some textbooks,
was to say that the mass of an object increases the faster it goes (is observed to be going).
The “mass” that is referred to in this case is not the rest mass mo but the quantity γmo ,
which indeed increases with increasing γ and therefore increasing |β|. Clearly the rest mass
mo is more fundamental, the other “mass” just being a reflection of the frame of reference
that the observer happens to be in.

3.7 Colliding Beams of Particles

Figure 3.5: Colliding beams

31
Consider two beams of particles which are made to collide in a small volume of space as in
Figure 3.5. If the beams are composed in one case of particles and in the other their cor-
responding anti-particles (for example protons and anti-protons or electrons and positrons)
then the total energy and momentum at collision is

E ′ = 2mo γc2 p′ = 0 (3.13)

where mo is the rest mass of either particle or antiparticle, which will be the same.
For the CERN LHC (Large Hadron Collider), currently running, beams of 6.5 TeV (6500
GeV) protons collide head on with a second beam of protons of the same energy. In time
these energies will be increased to 7 TeV (7000 GeV).
We can use
Ebeam = γmp c2 (3.14)
from the previous section to calculate that

v = 0.9999999910c ! (3.15)

3.7.1 Why Colliding Beams?

Consider the situation in Figure 3.6, which compares the centre of mass or CMS system (more
accurately referred to as the centre of momentum system for colliding beams of protons with
the laboratory system in which one beam is moving and the other beam is stationary. The
latter system is historically referred to as the laboratory system because all experiment in
HEP used to be performed in this system, where the stationary second beam was actually a
fixed target (usually a block of material).

Figure 3.6: Centre-of-momentum versus laboratory frame

In the center of mass frame we have

E ′ = 2mp γcms c2 p′ = 0 (3.16)

32
How much energy would we need in the laboratory frame, with a single beam of protons
hitting a fixed target of protons?1 We can transform the energy using the Lorentz transfor-
mations and we find
E = γcms (E ′ + βcms cp′ )
Eo + mp c2 = γcms (2mp γcms c2 )
Eo = 2mp γcms 2
c2 − mp c2 (3.17)
2 2
Eo = 2(mpmγcms
pc
2
c )
− mp c 2
Again considering the Large Hadron Collider, to achieve the same energies for a fixed target,
the laboratory-frame beam therefore needs to be of energy
[ ]
2(7000)2
Eo = 0.938
− 0.938 GeV
= 1.04 × 108 GeV (3.18)
∼ 1 × 1017 eV
This therefore would be the energy necessary for a fixed target machine to achieve the same
thresholds - much, much greater energies than in the colliding beam case, and not achievable
technologically currently or at any time in the foreseeable future, if ever. Note, however,
that fixed target machines are capable of much higher luminosity (i.e. number of collisions
per unit time) - an important feature when searching for rare processes. Also note that
in the above formula we expressed the proton mass as 0.938 GeV/c2 rather than explicitly
substituting for c.

3.8 Reconstructing Events

Consider a particle decaying into two other particles, represented in cartoon form in Fig-
ure 3.7.

Figure 3.7: Two-body decay

Particle A may live for a very short amount of time before decaying, and so in many cases
we will only observe particles B and C in our detectors. An example of such a decay is

∆++ → p + π +

where the ∆++ particle has a mean lifetime of approximately 10−24 s.

1
In practice it would have to be a gas or liquid hydrogen target if it was just protons, but a target of any
material would be a good approximation since the mass of a neutron is similar to the mass of a proton.

33
We can still infer the existence of particle A using special relativity and the fact (which we
will discuss in more detail later) that 4-momentum is conserved in all processes. Since then
(for the 4-vectors of these particles)

PA = PB + PC (3.19)

if we can measure PB and PC we can calculate PA . Remembering that

PA2 = EA2 /c2 − pA .pA ≡ m2a c2 (3.20)

then we can calculate the mass of the decaying particle. In a practical experiment we may
also have to take take into account the presence of background, that is that not all observed
protons and pions will necessarily have come from the decay of interest.
Clearly this approach can be generalized to the decay a single particle to N particles where
N > 2.

3.8.1 Example: Reconstructing a decay

Imagine we undertake a particle physics experiment, and are able to measure the energy and
momenta of a proton (p) and pion (π + ). The 4-momentum of each is (E/c, px , py , pz )

p = (1.76, 1.48, −0.17, 0.0)

(3.21)
π + = (0.24, 0.10, 0.17, 0.0)

in units of GeV/c. Summing these two 4-vectors gives

p + π + = (2.00, 1.58, 0.0, 0.0) (3.22)

This must be the 4-momenum of the particle which decayed to give the p and π + and so the
rest mass is √
2.002 − 1.582 = 1.23 GeV/c2 (3.23)
The mass of the decaying particle was therefore 1.23 GeV/c2 , corresponding to the rest mass
of a ∆++ particle.

34
Chapter 4

Interactions

What do we mean when we talk about a particle? The properties of a particle tell us how
it interacts. In a classical sense, an electron’s mass tells us how it reacts to the force of
gravity. The charge of an electron tells us how it will interact with an EM field. When we
treat particles quantum mechanically, we shall see how properties such as electric charge
are examples of quantum labels that control which interactions are allowed and also help
determine the interaction rate.
In general we can divide particle interactions into two general classes, based on the number
of particles in the initial state – decay and scattering.

4.1 Decay and Scattering

A decay is when a single particle spontaneously becomes a collection of particles – “one

particle in, two or more particles out”. Some examples of particle decays are given in
Table 4.1.

n → p + e− + ν̄e (β − Decay)
p → n + e+ + νe (β + Decay)
∆o → n + πo
µ+ → e+ + ν̄µ + νe
Σ+ → p + πo
Σ+ → n + π+

Table 4.1: Some examples of particle decays.

Scattering describes the interaction between groups of particles – “two particles in, two or
more particles out”. Some examples of scattering processes are given in Table 4.2.
A general transition from some initial state at an earlier time to a final state at some later

35
 e+ + e− → γ+γ (Annihilation)
p+p → p+p (Elastic Scattering)
p+p → p + p + πo
νµ + e− → µ− + νe
K+ + p → K o + ∆++

Table 4.2: Some examples of particle decays.

time may be a combination (sequence) of scattering followed by decays, for example

K+ + p → Ko + ∆++

- p + π+

4.2 Forces

In classical physics, forces are represented by interactions with fields (e.g. the EM field).
The field in classical EM is the product of a potential and is produced by the net effect
of a distribution of electric charge. An introduced test charge responds to the field and
experiences a force – effectively through action at a distance. In quantum field theory, at a
fundamental level, forces are seen as the exchange of particles, the gauge bosons,and there
is no action at a distance.
A classical analogy, not to be taken too literally, of a force arising due to the exchange of an
object is illustrated in Figure 4.1.

Figure 4.1: Classical analogy for a force as an exchange: (a) repulsive; (b) attractive.

36
4.3 Feynman Diagrams

All particle interactions can be represented by Feynman diagrams. At one level a Feynman
diagram can be thought of as a way of visualising a particle interaction in space-time. Since
space-time is four-dimensional, and Feynman diagrams are drawn in two dimensions, only
the time axis and one “representative” spatial axis are drawn. The convention that we will
follow is that the time axis is drawn horizontally and the spatial axis is drawn vertically.
Beware that some textbooks use the opposite convention, with the time axis flowing vertically
up the page.

4.3.1 The Vertex

The basic building blocks of Feynman diagrams are vertices. An example of a vertex is
shown in Figure 4.2.

Figure 4.2: A Feynman diagram representing a vertex.

Fermions are represented as solid lines with an arrow, and bosons are represented as wavy
lines (photons, W and Z bosons), helicies or pig tails (gluons) or dashed lines (Higgs bosons).
Note that particles (fermions) are represented as travelling forward in time, whereas anti-
particles (anti-fermions) are represented as travelling backwards in time!

4.3.2 A Scattering Diagram

A simple scattering process can be represented by a Feynman diagram, as in Figure 4.3, by

constructing it from two vertices with one leg of each vertex connected together.
In Figure 4.3, the gauge boson appears to leap across space at a particular point in time. This
should not be taken too literally. Consider Figure 4.4 for a scattering process a + b → c + d.
There are actually two distinct ways this could occur – two time orderings. In the first,
particle a emits gauge boson X which is absorbed by particle b. In the second, particle b

37
 Figure 4.3: A Feynman diagram representing a scattering process.

emits X which is absorbed by particle a. The Feynman diagram at the left is a stylised
picture encompassing both possibilities for getting from initial state a + b to final state c + d.

Figure 4.4: A Feynman diagram represents the two possible time-orderings for a scattering
a + b → c + d. From Thomson, Modern Particle Physics.

In actuality Feynman diagrams are more than mere cartoons, even though in this course
we will mostly use them as such. They in fact visually represent terms in a mathematical
expansion of the quantum mechanical amplitude for a process, which combined with a set
of rules (the so-called Feynman rules) allow calculations of the amplitude to be made in an
orderly manner and in principle at least to any desired level of precision.
The basketball analogy introduced earlier is simplistic. The gauge boson carries information
between the particles, energy and momentum, and quantum numbers, for example.

4.4 Example: Electromagnetic Interactions

Examples of electromagnetic processes will be used to illustrate drawing Feynman diagrams

for scattering processes, and to discuss the concept of virtual particles.

38
4.4.1 Electron-Electron scattering

An example Feynman diagram for two electrons interacting via the electromagnetic (EM)
force is given in Figure 4.5. Note that electrons don’t randomly emit photons, only during
an interaction (by definition).

Figure 4.5: An example Feynman diagram for two electrons scattering via the EM force.

4.4.2 Electron-Positron Scattering

In Figure 4.6 are shown two Feynman diagrams for electron-positron scattering. Notice how
these are simply rotations of one another.

Figure 4.6: Feynman diagrams for electron-positron scattering.

An example of what this type of scattering looks like in real life is given in Figure 4.7, which
is taken from the Belle experiment at KEK in Japan. In the right-hand picture, an 8 GeV
e+ enters from the left and collides with a 3.5 GeV e+ entering from the right.

39
Figure 4.7: An example of electron-positron scattering in the Belle detector at KEK in
Japan. Left is a view along the direction of the entering electron-positron beams, and right
is a side-on view.

4.4.3 Virtual Particles

The total energy and momentum of the electrons before and after the interaction are the
same. What about during the interaction? If we imagine just the top half of a scattering
diagram such as the left-hand one in Figure 4.6, such a diagram could represent the emission
of a real photon by an electron.

e− e−

Figure 4.8: Feynman diagram for emission of a photon

But can such a process occur? Consider the photon emission in the electron rest frame, as
depicted in Figure 4.9. Comparing the energy and momentum before and after the emission:

me c2
Before Etot = √ P⃗tot = 0
After Etot = (me c2 )2 + (|⃗pe |c)2 + |⃗pγ |c P⃗tot = 0 (|⃗pe | = |⃗pγ |)

The expression for Etot after emission must be greater than before, since |⃗pe | and |⃗pγ | are both
non-negative and in fact must be positive since the photon cannot be at rest. Clearly energy
and momentum cannot both be conserved. Hence an isolated electron cannot spontaneously
emit a photon. What then if instead the electron was, as suggested above, the top electron
in a scattering diagram, and the photon gets absorbed by the bottom electron? How do we
get around this problem of energy conservation? One way to think about this is that we can
use Heisenberg’s Uncertainty Principle to “borrow” the energy.

40
Figure 4.9: Emission of a photon in the CMS frame. Top: before emission; bottom: after
emission. The electron is in red, and the photon blue.

The Uncertainty Principle stated in terms of energy and time says that

∆E∆t ∼ ℏ

If the exchange particle has a mass m = ∆E/c2 , it can travel R ∼ c∆t and

ℏ
R∼
mc
R can be considered the effective range of the force.
The exchanged particle is undetectable and is said to be virtual (as opposed to real). It is
characterised by not satisfying the standard special relativity relation between mass, energy
and magnitude of momentum, i.e. if one worked out the 4-momentum of the particle from
the known 4-momenta of the particles it couples to at a vertex, assuming 4-momentum
conservation, and then uses the special relativistic relation

E 2 = |⃗p|2 c2 + m2 c4

to solve for the mass of the particle, if would not give a result equal to the rest mass of a
real particle of that type.
Another way to think about this is that to detect a virtual particle and determine its mass
would require a measurement to be made, so something would have to interact with it and
we would no longer be dealing with the same process and Feynman diagram. Or, to ask the
question “was the virtual particle really there” is not a question that quantum mechanics
allows us to ask.

4.4.4 Compton Scattering and Electron-Positron Annihilation

Feynman diagrams for two other electromagnetic scattering processes, Compton scattering
and electron-positron annihilation, are shown in Figure 4.10.
Note that these configurations have a virtual electron/positron.

41
Figure 4.10: Feynman diagrams for Compton scattering (left) and electron-positron annihi-
lation (right).

4.5 Relating Scattering and Decay

As well as rotations of the complete Feynman diagrams, you can also flip paths of an in-
dividual particle on a Feynman diagram to get a related process. This can be seen in the
example in Figure 4.11. In the figure, ere A - D represent some general particles, which are
not necessarily identical.

A C A C

_
B

B D
D

A+B →C +D A→B+C +D

Scattering Decay

Figure 4.11: Example of how a scattering process can be related to a decay process through
flipping one of the incoming particles to an outgoing particle.

Any legitimate particle interaction must be representable by a Feynman diagram. Note

however that the ability to draw a Feynman diagram for a process does not ensure that it
will occur in nature. Further quantum rules (conservation laws) can still prevent such an
interaction occurring. We will return to this point later.

42
4.6 Higher Order Contributions

Using Feynman diagrams we have depicted processes such as electron-electron scattering as

the exchange of a single photon. But can there be more complicated paths via which to get
from a given initial state to a given final state?
Consider a more general representation of e− e− scattering such as that represented in Fig-
ure 4.12.

e− e−

e− e−

Figure 4.12: A general representation of Feynman diagrams for electron-electron scattering.

What can be in the blob? Some examples are shown in Figure 4.13. the top Feynman
diagram is the one that we have already met, known as the lowest-order diagram describing
the scattering process. The lower two diagrams represent the exchange of two and three
virtual photons between the two electrons.

e− e−

photon

e− e−
e− e− e− e−

e− e− e− e−

Figure 4.13: The lowest order diagram for electron-electron scattering (top) and two exam-
ples of higher-order diagrams (bottom).

43
In principle there are an infinite number of Feynman Diagrams for any process. This is
illustrated in Figure 4.14, which depicts the lowest-order diagram as well as some other of
the higher-order diagrams which will contribute to the overall calculation of the scattering
probability.

Figure 4.14: Examples of higher order diagrams which contribute terms in a perturbation
expansion for a scattering process.

Each of these individual Feynman diagrams in fact represents a term in a perturbation

expansion which can in principle be used to calculate the interaction rate to any desired
accuracy. The more such diagrams are taken into account the more accurate will be the final
result. We will come back to this point below. In general all of the particles which represent
“internal lines” in these Feynman diagrams will be virtual, as was the case for the single
photon in the lowest-order diagram.

44
Chapter 5

Rates of Interactions

5.1 The Rates of Interactions

In this section an outline of the method for calculating the probability of an interaction
occurring is given.

5.1.1 Fermi’s Golden Rule

The transition rate for an interaction process (scatter or decay) is expressed in terms of
Fermi’s Golden Rule.
2π
Γf i = |Tf i |2 ρ(Ef ) (5.1)
ℏ
Here Γf i gives the number of transitions per unit time from some initial state |i⟩ to some final
state |f ⟩. Tf i is known as the transition matrix element, or alternatively as the quantum
mechanical amplitude, and contains the basic physics describing the process. ρ(Ef ) is the
density of final states and takes care of the kinematics of the interaction and ensures for
example that energy and momentum are conserved.

5.1.2 The Matrix Element

In general the matrix element Tf i is a complex number and can be expressed as follows using
bra and ket notation:

Tf i = ⟨f |Ĥ|i⟩ (5.2)

where Ĥ is an operator which effects the transition.

45
What if there is more than one path to the final state?

If there is more than one way to get from the initial state to the final state, that is, if one
can draw more than one distinct Feynman diagram to represent the transition, then each
Feynman diagram contributes to the amplitude. This allows for quantum interference to
take place.
If Ĥ1 is an operator corresponding to the first diagram, and Ĥ2 to the second, and so on,
then

Tf i = ⟨f |Ĥ1 |i⟩ + ⟨f |Ĥ2 |i⟩ + . . . (5.3)

The probability of interaction is proportional to |Tf i |2 . This is an example of the well-known

principle from quantum mechanics that when there are multiple ways to get from an initial
to final state, then the amplitudes must be added first and then the result squared in order
to calculate the probability. This is a coherent sum of amplitudes.

What if there are different final states?

Figure 5.1 depicts an example of the same initial state having the possibility to produce two
distinct final states.

e+ µ+ e+ τ+

γ γ

e− µ− e− τ−

Figure 5.1: Electron-positron annihilation to a muon pair and a tau pair.

.
We have previously seen the example of the electron and positron scattering or annihilating
to produce an electron and positron in the final state. There are also many other possible
final states.
If |f1 ⟩ and |f2 ⟩ are different final states, then we can think of a superposition of these as
containing the possibilities:

|f ⟩ = a1 |f1 ⟩ + a2 |f2 ⟩ (5.4)

The probability that we obtain final state |f1 ⟩ is given by

|⟨f1 |Ĥ1 |i⟩|2 = a21 (5.5)

46
and the probability that we obtain the final state |f2 ⟩ is given by
|⟨f2 |Ĥ2 |i⟩|2 = a22 (5.6)
The coefficients are normalized so that
a21 + a22 = 1
In this case the rule is that to get the probability of interaction the individual amplitudes
for each final state must be squared and then the results added. This is an incoherent sum
since the different final states are independent of each other, and no interference takes place.
Clearly the above can be generalised to multiple final-state possibilities, rather than just
two.

5.1.3 The Density of States

The density of states is related to conservation of momentum, and can most easily be un-
derstood by way of an example. The mean lifetime of a free neutron is very long compared
to the lifetimes of most weakly decaying particles, being around 15 minutes. The sum of
the masses of the daughter particles, the proton, electron and electron antineutrino, is very
close to the mass of the parent neutrino. This restricts the number of 3-momentum states
available for the daughter particles to occupy - we say there is little “phase space” available,
and this in turn means that the decay occurs slowly, leading to a long lifetime.
For those interested, the density of states is discussed in more detail in the following, non-
assessable, section. Do in any case, however, look at the section after next, Section 5.3,
which explains what a branching fraction is.

5.2 Calculation of Decay Rates - details non-assessable

To outline the steps involved in calculating the rate of a transition using Fermi’s Golden
Rule, we will consider a simpler case than that of a scattering process. The argument here
closely follows the one given in Thomson (2013). The simplest process that we can have is
the decay of a particle into two final-state particles. The Feynman diagram is just a single
vertex, and we can denote the process
i→1+2 (5.7)
In the centre-of-mass system of the decaying particle, the final state particles will emerge
back-to-back, as illustrated in Figure 5.2.
To calculate a decay rate we need to know the matrix element and the density of states.
The matrix element can be expressed as in Equation 5.2. Note that this is actually the
lowest-order version, which corresponds to what is termed the Born approximation. In

47
 Figure 5.2: A two-body decay as viewed in the centre-of-mass system

terms of perturbation theory, when the more complicated higher-order terms are taken into
account, we would actually express the matrix element in the form
∑ ⟨f |Ĥ|j⟩⟨j|Ĥ|i⟩
Tf i = ⟨f |Ĥ|i⟩ + + ··· (5.8)
j̸=i
Ei − Ej

In our example, using the Born approximation, we would write

Tf i = ⟨ψ1 ψ2 |Ĥ|ψi ⟩ (5.9)
The density of states can be written
 
 dn 
ρ(Ef ) =   (5.10)
dE Ef

To evaluate this we need to think about the normalisation of the wave functions, and this
presents a challenge if we want the resulting calculation to be consistent with special rela-
tivity. Firstly we will consider the non-relativistic case.

5.2.1 Non-relativistic formulation

The particle wave-functions can be represented as plane waves, e.g.

ψi = N e− ℏ p.x
i
(5.11)
where N is a normalisation factor and p and x are the momentum and position 4-vectors of
the particle. We have
p.x = Et − p.r = Et − ℏk.r (5.12)
where k is the wave vector of the particle and p = ℏk. The conventional way to normalise
the wavefunction is to demand one particle in a cube of side a as in Figure 5.3, with the
wave function vanishing at the boundaries.
This means that ∫
ψ ∗ ψ dV = 1 = N 2 a3 (5.13)

48
 Figure 5.3: Particle confined in a box of side a. Image from Thomson (2013).

and so N 2 = 1/a3 .
In momentum space the momentum is then quantised
2πnx 2πny 2πnz
px = py = pz = (5.14)
a a a
and a single state in momentum space occupies volume
( )3
2π (2π)3
= (5.15)
a V

If there is one particle per unit volume, then the number of states in an element d3 p =
dpx dpy dpz is
d3 p 1 d3 p
dn = × = (5.16)
(2π)3 V (2π)3
V
The density of states is
   
 dn   dn d|p| 
 
ρ(Ef ) =   =    (5.17)
dE Ef d|p| dE Ef

Integrating over a shell d3 p = 4πp2 dp in momentum space then gives

4πp2
ρ(Ef ) = × βc (5.18)
(2π)3

where we have used p = βcE.

Can also write the density of states as an integral, using the delta function.
  ∫
 dn  dn
ρ(Ef ) =   = δ(E − Ei )dE (5.19)
dE Ef dE

since energy is conserved, Ei = Ef . This means that Fermi’s Golden Rule takes the form
∫
2π
Γf i = |Tf i |2 δ(E − Ei )dn (5.20)
ℏ

49
with the integral being over all allowed states, regardless of energy.
For a two-body decay, only one particle in the final state needs to be considered since
momentum is conserved and this fixes the momentum of the second particle once that of the
first is known. This can be expressed using a delta function.
∫
2π d3 p1
Γf i = |Tf i |2 δ(Ei − E1 − E2 ) (5.21)
ℏ (2π)3
In fact if we want to include both particles 1 and 2 symmetrically we can use a second delta
function
∫
(2π)4 d3 p1 d3 p2
Γf i = |Tf i |2 δ(Ei − E1 − E2 )δ 3 (pi − p1 − p2 ) (5.22)
ℏ (2π)3 (2π)3

5.2.2 Normalisation of the wavefunction revisited

There is an issue with normalising the wavefunction such that

∫
ψ ∗ ψ dV = 1 (5.23)

which is the following. If we change our Lorentz frame of reference, what happens to this
expression? The situation is illustrated in Figure 5.4 The volume element dV shrinks by a

Figure 5.4: Particle confined in a box of side a, under a Lorentz transformation. Image from
Thomson (2013).

factor 1/γ, which means that Equation 5.23 is not describing a Lorentz invariant situation
since it implies that the particle density increases by a factor γ = E/m as viewed in the new
frame. This leads to the conclusion that the appropriate normalisation for the wavefunction
is one in which we demand that there is a certain amount of energy per unit volume rather
than number of particles. What is conventionally chosen is
∫
ψ ′∗ ψ ′ dV = 2E (5.24)

where the primes on the wavefunction indicates that it is different from the previous ψ. In
fact
ψ ′ = (2E)1/2 ψ (5.25)

50
Note that the factor of 2 is merely convention. The important point is that normalising
to something proportional to energy makes the normalisation condition explicitly Lorentz
invariant. Here there is an amount of energy 2E per unit volume.

5.2.3 Relativistic formulation

With this normalisation condition, the expression for the matrix element in our example
would normally re-expressed as follows:

Mf i = ⟨ψ1′ ψ2′ |Ĥ|ψi′ ⟩ = (2Ei .2E1 .2E2 )1/2 Tf i (5.26)

which then leads to an expression for the transition rate as follows:

∫
(2π)4 d3 p1 d3 p2
Γf i = |Mf i |2 δ(Ei − E1 − E2 )δ 3 (pi − p1 − p2 ) (5.27)
2Ei ℏ (2π)3 2E1 (2π)3 2E2

The factors of the form d3 p/((2π)3 2E) are know as the Lorentz Invariant Phase Space (LIPS)
for each particle and the expression for the decay rate has been rewritten in a Lorentz
invariant form. We can see that the transition rate is inversely proportional to the energy
Ei = γmi of the decaying particle, which is what we would expect from time dilation.

5.2.4 Expression for the two-body decay rate

To calculate the transition rate for the two body decay, the simplest frame of reference to
use is the rest frame of the decaying particle i.e. setting Ei = mi c2 and pi = 0. If one works
through the maths (if you are interesred, many textbooks do this, Thomson is an example),
then the expression obtained is
∫
|p∗ |
Γf i = |Mf i |2 dΩ (5.28)
32π 2 ℏmi c
where dΩ = sinθdθdϕ is an infinitesimal solid angle and p∗ is the momentum of either final
state particle. One can see that if the matrix element can be obtained from the underlying
physics (in the jargon, from the Lagrangian density governing the interaction), then the
decay rate or lifetime of the particle (τ = ℏ/Γ) is readily obtained.
The above discussion could be generalised to decays to N particles rather than two, and a
similar line of reasoning can be followed for scattering rather than decay, in which case the
relevant quantity obtained is the scattering cross section.

51
5.3 Branching Fractions

Note that the terms branching fraction and branching ratio tend to be used interchangably
in particle physics.
Consider a situation where a particle can decay in two ways to distinct final states.

A → B+C
A → D+E

Here there is no interference since the final states are distinguishable, so to get the total
decay rate the two separate decay rates are simply added incoherently.
If the decays are effected by operators Ô1 and Ô2 then

Rate(A → anything) = Rate(A → B + C) + Rate(A → D + E) (5.29)

2 2
∝ | ⟨BC|Ô1 |A⟩ | + | ⟨DE|Ô2 |A⟩ | (5.30)

The branching fraction (also called branching ratio) for A → f is defined to be

Rate(A → f )
Br(A → f ) = (5.31)
Rate(A → anything)

This concept is clearly extendable to situations where there are more than two decay modes.

52
Chapter 6

The Electromagnetic and Strong

Interactions

In this chapter we will discuss some of the properties of the electromagnetic and strong
interactions, but first we will introduce the concept of a coupling constant, and look at the
basic vertices of the Standard Model coupling particles to gauge bosons.

6.1 Coupling Constants

A number of properties of an interaction determine its rate.

g
A

C
Figure 6.1: A basic interaction vertex with coupling constant.

A coupling constant g is associated with each interaction vertex and contributes to the
matrix element Mf i for a process. For example, the probability for a decay

A→B+C

53
as depicted in the Feynman diagram in Figure 6.1, has the property
Prob(A → B + C) ∝ |Mf i |2 ∝ g 2
The coupling constant g depends upon the nature of the interaction, i.e. in the case of the
Standard Model whether it is EM, strong, or weak.
Consider a scattering process, such as that depicted in Figure 6.2.

Figure 6.2: A scattering Feynman Diagram constructed from two vertex diagrams.

As indicated previously, the Feynman diagram for a simple scattering process is formed from
the combination of vertices, two in this case. Again, there are specific details associated with
each interaction, but simplistically, given a coupling constant g associated with each vertex,
then for two vertices we have
Interaction rate ∝ g 2 g 2 = g 4

6.2 The Standard Model Vertices

Which basic vertices are allowed in the Standard Model? This is determined by the structure
of the underlying mathematics of the Standard Model as written down in Quantum Field
Theory. Translated into Feynman diagrams, for couplings of gauge bosons to fermions, this
amounts to vertices of the type displayed in Figure 6.3.
Each vertex has the characteristic of the relevant gauge boson for the type of interaction,
coupled to a pair of fermions. There are additional allowed vertices to this set. In the first
diagram the electron could be replaced by a µ− or τ − , or by any of the quarks. Similarly in
the third diagram, the electron could be replaced by a µ− or τ − , with the type of neutrino
changed to match, and in the fourth diagram, the type of both neutrinos could be changed
to νµ or ντ . In both the third and fourth diagrams, the two leptons could be replaced with
two quarks. Finally, in any of the diagrams, the particles shown could be replaced by their
antiparticles, reversing the directions of the arrows (or equivalently, the incoming fermion
could be rotated to be outgoing, changing it to its antiparticle, and vice versa).

54
Figure 6.3: Examples of Standard Model vertices. From Thomson, Modern Particle Physics.
See text for discussion of additional allowed vertices.

Each vertex also has a coupling constant characteristic of the type of interaction, which is
related to the strength of the force. If we change the particles in a diagram as discussed in
the previous paragraph, the coupling constant at the vertex may change, depending on the
quantum numbers of the new pair of particles. For example the coupling constant in the first
diagram is in fact proportional to the charge of the participating fermions, and so if quarks
were substituted the electric charge of the quarks would constitute the coupling constant.
The quantities α in the figure are related to the strength of the interaction, and will be
discussed further in the next two chapters.

6.3 The Electromagnetic Force

The electromagnetic force, or interaction, occurs between charged particles and is mediated
by the photon. The basic Feynman diagram for scattering is given in Figure 6.4, where A
and B are charged quarks or leptons. If either of the particles was an antiparticle then the
arrows on the appropriate leg of the diagram would be reversed.

Properties of the EM Interaction

Some of the characteristic properties of the EM interaction are listed below.

• The basic coupling constant which is associated with a vertex coupling a particle of
unit electric charge to a photon is e, the charge of a positron or proton. The size of
the coupling constant g = e is such that a related quantity, the fine structure constant
α has the value
e2 1
α= ≃
4πϵ0 ℏc 137

55
 Figure 6.4: The basic Feynman diagram for scattering via the EM force.

• Processes which occur via the EM interaction tend to happen on a medium (in particle
physics terms) time scale of order 10−16 − 10−20 s.

• The gauge boson mediating the interaction is the photon, which has m = 0. The
massless nature of a photon is related to the range R of the EM interaction being
large, effectively R = ∞, as embodied classically in the Coulomb force law.

• We will discuss conservation laws later, but for the time being note that most conser-
vation laws, with the exception of Strong Isospin, are obeyed by the EM interaction.

6.3.1 Examples of EM Interactions

π0 → γ + γ µ− + p → µ− + p

Figure 6.5: Two examples of EM interactions. Left is π 0 decay, right is muon-proton scat-
tering.

Figure 6.5 shows two examples of processes which are mediated by the EM interaction, one
¯
decay and one scattering. The π 0 can be considered a bound state of either a uū or dd.
In the Feynman diagram shown, the u an ū quarks annihilate each other to produce a pair
of photons. In the muon-proton scattering diagram, the muon scatters from one of the u
quarks in the proton, via the exchange of a virtual photon.

56
6.3.2 Suppression of Higher Order Processes

Figure 6.6: The basic Feynman diagram for scattering via the EM force.

Given the size of the fine structure constant α is of order 0.01 (1/137), it is clear that
Feynman diagrams with more powers of α associated with them (i.e. more vertices) will
tend to be suppressed when calculating the rate for a process. For example, in Figure 6.6
the processes where an electron and positron annihilate to produce two photons is compared
to the process where three photons are produced.
The ratio of the rates for the two processes is therefore as follows

Rate(e− + e+ → 3γ)
∼ O(g 2 )
Rate(e− + e+ → 2γ)

which indicates that the rate for the three photon case will be heavily suppressed compared
to that for the two photon case. Because we are considering EM, the coupling constant g is
represented by the electric charge of the particles coupling to the photon at the vertex, so
g 2 = e2 .
The same sort of argument would apply when comparing lower order (less vertices) diagrams
to higher order (more vertices) diagrams in EM for the same initial and final state. In EM,
rate calculations can be generally be carried out to a good approximation by just considering
the lowest order diagrams.

6.4 The Strong Force

A nucleus is a bound state of positively charged protons and (usually) neutral neutrons.
What then holds the nuclei of atoms together within a small volume, given that the EM
repulsion between the protons should cause the nucleus to fly apart? Clearly there must be

57
a strong attractive force, stronger at these distances than the EM force, acting between the
nucleons to hold them together.
Whatever the nature of this force, it must therefore have two properties:

• Strong: to overcome the protons EM repulsion

• Short range: because it effectively only operates within the nucleus

6.4.1 The Yukawa Hypothesis

Figure 6.7: The strong force viewed as the exchange of a pion.

In 1937, Yukawa presented a model for the strong force, with the force being mediated by
particle exchange.
Given the short range (R ∼ 10−15 m) range of the force, he concluded that the exchanged
particle must be massive. In particular, he estimated its mass m to be as follows, in what is
an argument based on the uncertainty principle:
ℏ
m∼ ∼ 0.2 GeV/c2
Rc
Yukawa had effectively predicted the existence of the π as the exchange particle, as illustrated
in Figure 6.7.
To be more specific, to accomodate all transitions between a proton and a neutron, three
charge states for the pion would be required, as illusrated in Figure 6.8.

6.4.2 What holds the proton and neutron together?

Since we know that the fundamental building blocks of matter are the quarks, something
which was unknown to Yukawa, a more fundamental force is required to hold the individual
protons and neutrons (nucleons) together.

58
 Figure 6.8: The three diagrams for pion exchange in the Yukawa hypothesis.

The modern view is that the strong force, or interaction, actually occurs between quarks
and is mediated by gluons. The basic Feynman diagram for scattering is given in Figure 6.9,
where A and B are quarks.

6.4.3 Properties of the Strong Interaction

Some of the characteristic properties of the strong interaction are listed below.

• The basic coupling constant which is associated with a vertex coupling a quark or
antiquark to a gluon is gS , and is associated with a type of charge known as colour
charge. The size of the coupling constant is such that the strong interaction equivalent
of the fine structure constant of EM, αS has the property
gS2
αS = ≃1
4πϵ0 ℏc

• Processes which occur via the strong interaction tend to happen on a short time scale
of order 10−22 − 10−24 s.
• The gauge boson mediating the interaction is the gluon, which has m = 0. The massless
nature of a gluon should therefore imply that the range R of the strong interaction
should be large, effectively R = ∞, which is not what we observe in Nature. Something
else is clearly going on which we need to get some insight into.

59
 Figure 6.9: The basic Feynman diagram for scattering via the strong force.

• For the time being note that all conservation laws that we will discuss later are obeyed
by the strong interaction.

6.4.4 The Strong versus EM Interaction

There are some very obvious differences between the properties of the strong and the EM
interaction.

• The strong interaction occurs very fast by comparison with the EM, i.e. is overwhelm-
ingly the more probable if a process could occur via either interaction.

• Leptons do not interact with gluons.

• The strong interaction is confined to short distances despite being mediated by a

massless particle.

We will deal with these differences when we discuss Quantum Chromodynamics later, but
basically the gluons possess properties which distinguish the behaviour of the strong inter-
action.
To interact with gluons, quarks carry an additional quantum number called colour. Strong
interactions change the colour of quarks, as indicated in the Feynman diagram of Figure 6.10.
An important feature of gluons which is hinted at in the figure is that gluons themselves
carry colour charge, unlike the situation in EM where photons do not carry electric charge.
We will return to this point.

60
Figure 6.10: The basic Feynman diagram for scattering via the strong force, incorporating
colour.

6.4.5 Examples of Strong Interactions

We give below two examples of Feynman diagrams for strong interaction processes, one of
a decay and one of a scatter. In the first example, illustrated in Figure!6.11, a ∆++ particle
decays to a proton and a pion.

d¯

g
u d
u u
u u

¯ + p(duu)
∆++(uuu) → π +(ud)

Figure 6.11: The decay via the strong interaction of a ∆++ particle to a proton and pion.

In the second example, illustrated in Figure 6.12, a proton scatters from an antiproton to
produce a neutron and an antineutron.

6.4.6 But what about Yukawa?

As we said above, the gluon is massless and the strong force ought to have infinite range.
But the Yukawa hypothesis requires the force between nucleons to mediated by a massive
particle and have short range.

61
 u u
d d
u d
g
g
ū d¯
ū ū
d¯ d¯

¯ → n(udd) + n̄(ūd¯d)
p(uud) + p̄(ūūd) ¯

Figure 6.12: Proton-antiproton scattering to a neutron-antineutron.

Figure 6.13: The inter-nucleon force viewed as a kind of van der Waals force.

One way to view this is that the basic force is between quarks, and that the inter-nucleon
force is a residual of the inter-quark force, in analogy with the van der Waals force between
molecules in EM. This is illustrated in Figure 6.13. The result conspires to then look like
π exchange. We still have the question as to why the basic inter-quark force is short-range,
which we will look at again when discussing QCD.

62
Chapter 7

The Weak Force

In many ways the weak interaction is the most complicated and the most interesting of the
three interactions of the Standard Model. In this chapter we will cover some of its basic
properties, before returning to some of its more complicated aspects later in the module.

7.1 Beta Decay

Radioactive beta decay is the only well-known “everyday” example of the weak interaction,
and at first sight it does not really look like a force in the same sense as the EM and strong
forces do. It is not binding particles together, for example, as in the case of atoms and nuclei.

Figure 7.1: Examples of β − decay (top) and beta+ decay (bottom).

Figure 7.1 gives examples of the two forms of nuclear beta decay, which result in either an
electron or a positron being emitted, along with an antineutrino or neutrino and a recoiling
nucleus. The underlying transitions at the nucleon level are the following:
n → p + e− + ν̄e p → n + e+ + νe
which look more like particle decays analogous to those for the strong and EM interactions.

63
7.2 The Weak Force

β decay is just one example of the weak interaction. As with the other forces we have looked
at, the weak interaction is mediated by particles. Unlike the other forces, however, there are
three particles involved , W + , W − & Z 0 . The basic weak scattering Feynman diagram is
shown in Figure 7.2. A and B are quarks or leptons.

Figure 7.2: The basic Feynman diagram for scattering via the weak force.

7.2.1 Properties of the Weak Interaction

Some of the characteristic properties of the weak interaction are listed below.

• The basic coupling constant which is associated with a vertex coupling a weakly inter-
acting particle to a W ± or Z, gW , is such that the related dimensionless quantity, αW
has the value
2
gW 1
αW = ≃
4πϵ0 ℏc 30
Intrinsically this makes αW larger than α of EM, but in fact the large masses of the W
and Z bosons contribute to making the weak force weaker than EM at low energies.

• Processes which occur via the weak interaction tend to happen on a medium to long
time scale of order 10−6 − 10−13 s.

• The gauge bosons mediating the interaction are the W ± or Z, which are massive.
Hence the weak interaction is short range.

• As we will find later, the weak interaction violates a number of conservation laws.

7.2.2 The Lepton-W ± Vertices

The basic lepton vertex with the W + is given in Figure 7.3.

64
 W+

νℓ

ℓ−

Figure 7.3: The basic vertex coupling a charged lepton and neutrino to a W + .

Here l is e, µ or τ , and the particle type and neutrino type must match. This is related to
conservation of lepton number, which we will discuss later. To get the corresponding vertex
for coupling to a W − , the neutrino must changed to an antineutrino and the charge of the
charged lepton must be made postive (reversing the arrows on the two particles). The above
diagram can also be transformed with rotations and flips to represent different processes.

7.2.3 Quark-Lepton Symmetry

The coupling constant, gw , for W ± and each lepton family is the same. Let’s assume that
in the weak interaction there is quark-lepton symmetry.
( ) ( ) ( ) ( )
νe u νµ c
− , − ,
e d µ s
The above assumes only two generations of quarks and leptons, which at the time these ideas
were first developed were the only two known.
If this symmetry holds, we can draw quark equivalents of the lepton verticies by making the
replacements
( ) ( )
νe → u νµ → c
,
e− → d µ− → s

With this transformation we get the vertex transformation depicted in Figure 7.4.
There is a similar transformation for the second generation. Note that the weak interaction
can change the flavour of quarks.

7.2.4 β Decay Revisited

With this symmetry, we can now get further insight into β decay. At the quark level
it involves the coupling of a d and u quark at one vertex and an electron and electron

65
Figure 7.4: Transformation of the basic charged weak vertex under quark-lepton symmetry.

antineutrino at the other, as illustrated in Figure 7.5.

e−

W− ν̄e

d u
d d
u u

n(udd) → p(uud) + e− + ν̄e

Figure 7.5: Quark-level Feynman diagram for β − decay.

7.2.5 Weak Coupling Constants

Consider the weak scattering process illustrated in Figure 7.6. If the quark-lepton symmetry
just discussed was perfect, then we would expect

gw = gud = gcs

and the only quark flavour changes would be

u ←→ d c ←→ s

Note that energy considerations favour heavy to light quark transitions.

66
Figure 7.6: Feynman diagram for charged weak scattering of an electron from a down quark.

µ− µ−
d W− s W−

ū ū
ν̄µ ν̄µ
π −(dū) → µ− + ν̄µ K −(sū) → µ− + ν̄µ

Figure 7.7: Feynman diagram for the decay of a charged pion (left) and charged kaon (right).

Consider however the two decays illustrated in Figure 7.7, both of which are observed in
Nature.
If the weak quark-lepton symmetry is perfect, then the weak decay of the K − would be
forbidden, since there is no non-zero coupling constant gus and no possibility of s ↔ u
transitions. The decay is, however, observed.

7.2.6 The Cabibbo Angle

To try to explain this situation, Cabibbo in the 1960s suggested that the symmetry is actually
between
( ) ( ) ( ) ( )
νe u νµ c
, ,
e− d′ µ− s′
where d′ and s′ are quantum superpositions of the d and s quarks

|d′ ⟩ = |d⟩ cos θc + |s⟩ sin θc

|s′ ⟩ = −|d⟩ sin θc + |s⟩ cos θc

and where θc is known as the Cabibbo angle.

67
A Reminder Regarding State Vectors in Quantum Mechanics

If |ψ⟩1 and |ψ2 ⟩ are basis states of a two-state quantum system, then a general superposition
state is
|ψ⟩ = a1 |ψ1 ⟩ + a2 |ψ2 ⟩.
Given this, the probability that an operation returns an eigenvalue of |ψ1 ⟩ is a21 , and the prob-
ability that an operation returns an eigenvalue of |ψ2 ⟩ is a22 . The coefficients are normalized
so that
a21 + a22 = 1

In terms of coupling constants, we can consider the coupling of the W to the pairs of quarks
in the manner suggested in Figure 7.8.

Figure 7.8: Decomposition of the charged weak coupling at a vertex.

The angle θc has been measured to be ∼ 12.7 degrees. Since the contribution of a vertex to
the rate calculation for a process goes as g 2 where g is the coupling constant at the vertex,
we have
Rate(u → s) g2 1
∼ us = tan2 θc ∼
Rate(u → d) 2
gud 20
and the u → s transition is said to be Cabibbo suppressed.
Another way of expressing the superpositions is in matrix form
( ′ ) ( )( )
|d ⟩ cos θc sin θc |d⟩
=
|s′ ⟩ − sin θc cos θc |s⟩
from which it is clear that the Cabibbo angle is related to a rotation between two different
bases. The unprimed basis corresponds to the mass eigenstates for the quarks whilst the
primed basis corresponds to the weak eigenstates.
The above formalism can be generalised to the third generation of quarks, in which case the
2 × 2 rotation matrix is replaced by a 3 × 3 matrix known as the Cabibbo-Kobyashi-Maskawa
(CKM) Matrix.

68
7.2.7 The Quark and Lepton-Z 0 Vertices

As well as by the charged W ± boson, the weak force is also mediated by the neutral Z 0 .
The basic vertices for this are given in Figure 7.9.

Figure 7.9: The basic vertices coupling leptons and quarks to a Z 0 .

Unlike the case for coupling to the W ± , the coupling to the Z 0 is found experimentally to
not mix the quark states. For example, we find the situation in Figure 7.10 for the decays
of charged kaons. Whilst the coupling to a charged W can change the flavour of quark as in
the left diagram, the coupling to a Z 0 cannot as in the right diagram.

Allowed Forbidden

Figure 7.10: Allowed and forbidden decays of charged kaons.

In the jargon of particle physics, we say that there are no Flavour-Changing Neutral Currents
(FCNC). To be more precise, the statement that the process in our example here, K + →
π 0 + νµ + ν̄µ is forbidden should really be that the process cannot occur via the Feynman
diagram shown, which couples a Z 0 to two differing flavours of quarks at a single vertex. In
practice, it may be possible to draw a higher-order Feynman diagram to represent the process,
involving a larger number of vertices which are allowed in the SM. In this case, because it
is a higher-order process the matrix element will be much smaller, and the process will be
very rare and very difficult to observe experimentally.

69
7.2.8 A Further Example of Weak Decay

Figure 7.11 shows a further example of a weak decay, in this case the decay of a Λ baryon
to a proton and pion.

d
ū
d
ū
W− d
W−
s
s u u
u u
d d d
u
Λ(uds) → p(uud) + π −(dū)
Λ(uds) → p(uud) + π −(dū) u

Figure 7.11: Two examples of Feynman diagrams for Λ decay.

Two examples of Feynman diagrams are shown in the figure, indicating again that there is
often more than one path to the same outcome. All such configurations need to be considered
when calculating the rates for processes such as this.

7.3 General Interactions

In general, interactions can be a combination of interactions of different type. Take the

example in Figure 7.12, which shows a Feynman diagram for the decay of a charged kaon
to three charged pions. The rate-limiting step is the weak part involving the W + , but in
order to generate the third pion, a quark must radiate a gluon which produces an additional
quark-antiquark pair, a step which happens very fast.
When drawing such a Feynman diagram, convention has it is that it is permissible to not
explicitly include the gluon, leaving it as assumed. In principle of course a quark could also
radiate a photon which produces the quark-antiquark pair, but such a path would be far less
likely than the one shown.

70
 u
d¯

u
s̄ u
W+ g d¯

d
ū
¯ + π +(ud)
K +(us̄) → π +(ud) ¯ + π −(dū)

Figure 7.12: An example of a decay which requires both a weak and a strong component to
proceed.

7.4 Electroweak Unification

There is a lot of similarity between the neutral weak (Z 0 ) interaction and the EM interaction,
as seen in Figure 7.13. We shall see later that this is related to the unification of the weak
and EM interactions. This means that the weak and EM interactions are in fact two aspects
of the same thing.

7.5 A Brief Aside - Gravity

Gravity does not form a part of the Standard Model. It is accurately described by General
Relativity, which is a classical theory based on the properties of curved spacetime. The
Standard Model on the other hand is a quantum (field) theory based on flat spacetime.
Despite much effort on the part of theorists, gravity does not fit into quantum theory and
this remains one of the great unsolved problems in fundamental physics.
From what we do know about the characteristics that a quantum theory of gravity would
possess, we can estimate the gravitational equivalent of the fine structure constant to be,
considering the basic “charge” for gravity to be the electron mass:

Gm2e
αG = ≃ 10−45
ℏc
Compared to the other forces, gravity is therefore extremely weak, and at all accessible
energy scales available to experimental particle physics, it can be completely neglected.

71
Figure 7.13: The neutral weak and EM interaction.

72
Chapter 8

The Higgs Boson

The Higgs boson is fundamental to the Standard Model of particle physics. First postulated
in 1964, it took nearly fifty years, until 2012, for its discovery at CERN’s Large Hadron
Collider (LHC) to be announced.
The search for the Higgs boson, and now further study of its properties, forms one of the
major goals of the LHC program. As explained below, the Higgs boson was the missing
piece of the Standard Model, related to the origin of mass for fundamental particles. Obvious
questions that one might ask about the Standard Model are why do the fundamental particles
have mass in the first place? Why are these masses different for different particles, as
illustrated in Figure 8.1? Why are the photon and gluons massless, and the W and Z
bosons massive?
The Higgs mechanism gives insight into some of these questions but not all. It is also
true that for several reasons the Standard Model, complete with Higgs, is incomplete, and
believed to be a “low-energy” approximation to a deeper theory.

8.1 Electroweak Unification and the Higgs Field

In the late 1960s, Glashow, Weinberg and Salam unified the electromagnetic and weak in-
teractions, continuing a quest towards a simplified picture of the forces of Nature which
was begun much earlier by Maxwell, who unified electricity and magnetism in the 1800s.
They drew on work from a few years earlier by Higgs and others, which had showed how
fundamental particles might acquire a mass through interaction with a new field permeating
the Universe, now known as the Higgs field. In the absence of the Higgs field, fundamental
symmetry principles dictate that all of the particles in the Standard Model should be mass-
less, clearly in contradiction with observation, and our very existence. Associated with the
Higgs field would be its quantum, a new particle called the Higgs Boson.

73
Figure 8.1: The masses of the particles in the Standard Model. Image: https://
scienceblogs.com/startswithabang/2010/11/04/the-horror-of-the-higgs.

8.2 Spontaneous Symmetry Breaking (SSB)

It is possible for the underlying equations describing a physical system to have a high degree
of symmetry but for the “ground state” or state of lowest energy realised by the system to
be less symmetric. An everyday example is illustrated by the pencil in Figure 8.2. In its
metastable state standing vertically, the pencil is in a higher-energy state and the system has
rotational symmetry about the vertical axis through the pencil. If the pencil is disturbed in
some way, then it falls and the system transitions to its ground state of lowest energy with
the pencil lying on the table. Because the pencil has to fall in one unique direction out of
all possible, the rotational symmetry is now broken by this ground state. The term for this
is spontaneous symmetry breaking.
There are many systems in Nature where at high energies they may be more symmetric than
at low energies, and indeed the underlying mathematical description of the particles and
forces describing the Universe seems to be one of them. In such cases “phase transitions” may
occur from the high energy to low energy state. An example you will probably be familiar
with is the behaviour of ferromagnetic materials above and below the Curie temperature.

74
Figure 8.2: Spontaneous symmetry breaking. The pencil standing on end has perfect rota-
tional symmetry. When it falls to its lowest energy state, a preferred direction is selected,
breaking the symmetry. Image: http://www.nobelprize.org/.

8.2.1 SSB in the Standard Model

The Higgs mechanism uses spontaneous symmetry breaking to explain how a massless par-
ticle may acquire a mass in the physical realisation of a theory without violating the re-
quirement of an underlying symmetry known as gauge symmetry. In particular, Glashow,
Weinberg and Salam’s theory was able to explain how the photon of electromagnetism can
be massless but the W and Z bosons of the weak force can be massive, through the existence
of a Higgs field and spontaneous symmetry breaking.
Without the Higgs field, the spin 1 photon, W + , W − and Z are strictly massless, each with
two spin states only (why?). This would in fact have been the case in the early Universe
at high temperature and energy. With the Higgs field, the W + , W − and Z acquire a
mass, which implies that they now each have three spin states, while the photon remains
massless. This is the case today in the Universe, at much lower temperature and energy.
The Universe presumably underwent an electroweak phase transition as it expanded and
cooled. In the technical jargon of quantum field theory, the Higgs field is said to have
acquired a non-zero vacuum expectation value at the electroweak phase transition. Otherwise
massless particles interacting with the physical vacuum containing the Higgs field with non-
zero vacuum expectation value behave as if they have a mass, travelling at less than the
speed of light because of the presence of the medium.

8.3 Higgs Couplings to other particles

A characteristic of the Higgs field in the Standard Model is that the Higgs boson couples most
strongly to the heaviest particles, with a coupling proportional to the mass of the particle
concerned. It should be noted that this is essentially “put in by hand” - the Standard
Model does not provide an explanation as to why the Higgs coupling is of this form, which
is another way of saying that the Standard Model cannot predict the values of the masses
of the fermions and bosons it describes, other than predicting that the photon and gluon
should be massless. Clearly it is not a complete description of Nature.

75
Figure 8.3 shows Feynman diagrams for vertices coupling the Higgs boson to a fermion-
antifermion pair, a pair of charged W bosons, and a pair of neutral Z bosons. For the
left-hand diagram, the largest coupling will be of the Higgs boson to a top-antitop pair,
since the top quark is by far the most massive fermion in the Standard Model. The Z and
W are the next most massive particles with the exception, as we now know, of the Higgs
boson itself.

Figure 8.3: Higgs boson coupling to a fermion-antifermion pair (left), charged W boson pair
(centre) and neutral Z boson pair (right). Image adapted from Thomson (2013).

Because the coupling of the Higgs boson to other particles is proportional to their mass, it
does not couple directly to photons or gluons. It can couple to photons and gluons through
what are called virtual loops, with examples given in Figure 8.4. Other quarks could feature
in the fermion loops in the left two diagrams in the Figure, but the greatest contribution
will come from diagrams featuring the top quark because it has the strongest coupling to
the Higgs boson.

Figure 8.4: Higgs boson coupling to massless gluons (left) and photons (centre and right)
through virtual loops. Image from Thomson (2013).

Like the masses of the other fundamental particles in the Standard Model, the mass of
the Higgs itself is not predicted, and is therefore a parameter which must be determined
experimentally. Prior to its discovery, however, there were hints as to what range its mass
was likely to lie in, based on the previous searches for it at earlier accelerators than the
LHC, and also on studies of the properties of the particles which were already known. For
example, it was known that the mass had to be greater than 115 GeV/c2 , since if was less it
should have been seen in previous experiments, which had the energies to probe mass ranges
up to that value. It was also expected that the mass should be less than around 1 TeV/c2

76
(1000 GeV/c2 ), since otherwise the Standard Model without something playing the role of
the Higgs field would be inconsistent, being unable to explain correctly certain scattering
probabilities of known particles.
Not knowing the mass of the Higgs boson a priori made it incredibly difficult to design
an accelerator and detectors to discover it. The ATLAS and CMS detectors at the LHC
were designed to be sensitive to Higgs boson masses anywhere within the above range.
However the ease with which the Higgs boson could be disentangled from the overwhelming
background processes was very dependent on the exact value of the mass. As the properties
of particles such as the top quark and the W and Z bosons became clearer, it began to
appear that the Higgs boson mass was more likely to lie in the lower part of the expected
mass range, a conclusion which was able to be drawn because the observed properties of the
particles coupling to the Higgs, particularly the heaviest ones which couple most strongly,
are sensitive to what the Higgs mass is and therefore can be used to constrain it.

8.4 How the Higgs boson is made at the LHC

Two ways in which a real Higgs boson can be produced at the LHC are shown in Figure 8.5.
At high energies the objects colliding at the LHC are not so much the protons themselves as
their individual constituents, the quark, antiquarks and gluons of which they are composed.

Figure 8.5: Examples of how the Higgs boson can be produced at the LHC. Image from
Thomson (2013).

The left-hand process is known as gluon-gluon fusion, where quarks in each proton radiate
virtual gluons which fuse through a virtual top-quark loop to form a Higgs boson. The right-
hand process is known as vector boson fusion, in which each proton radiates a virtual W
boson and the two W bosons fuse directly to produce a Higgs boson. Two other processes
not illustrated here, known as top fusion and associated production (or Higgs-strahlung)
respectively also contribute significantly to Higgs boson production at LHC energies.

77
8.5 How do we detect the Higgs boson once produced?

Once produced, the Higgs boson is expected to decay with a mean lifetime of ∼ 10−22 s. Even
travelling at the speed of light it would not on average travel more than a few nuclear radii
before decaying, making it impossible to directly observe. As we have previously discussed,
we can nevertheless detect particles with such short lifetimes by detecting the particles into
which they decay, and then using special relativity and the conservation of 4-momentum to
reconstruct them. See section 3.8 for a example of how this is done.
The vertices which were depicted in Figures 8.3 and 8.4 serve as examples of processes
representing Higgs boson decays. Which decay process is likely to be the most probable
depends on the mass of the Higgs, as shown in Figure 8.6. At low masses, by far the most
probable decay is H → bb̄, while at high masses the most probable decays are H → W + W −
followed by H → ZZ. In all cases shown except for H → γγ, the products of the Higgs
boson decay are themselves not directly observable, and must be identified by what they
decay into, or by the jets of particles that they produce through hadronisation if they are
quarks or gluons.

Figure 8.6: The variation in the predicted branching fraction to different decay products
of the Higgs boson as a function of its mass. Image taken from J. Ellis, Higgs Physics,
arXiV:1312.5672[hep-ph].

Some decay modes are easier to detect than others, due to their being less background
processes which can mimic them. Despite having a low branching fraction, the decay H → γγ
was considered the “golden channel” for searching for Higgs boson decay because the signal
is clean (two high energy photons) and the backgrounds moderate. Another clean channel is
the decay to two Z bosons, which in turn decay each to a pair of charged leptons, particularly
e+ e− or µ+ µ− . Generally decay channels involving quarks and gluons will be very difficult to
detect, due to the very large backgrounds from all of the other quarks and gluons produced
in the rather messy proton-proton collisions.

78
8.6 The Discovery of the Higgs Boson

As previously mentioned the Higgs boson was discovered by the ATLAS and CMS experi-
ments at the LHC in 2012. The LHC commenced serious data-taking in late 2010, and the
data from 2011 and the first half of 2012 was used in the discovery. It was decays in the two
golden channels referred to above which constituted the first experimental evidence for the
Higgs boson.

8.6.1 Higgs decaying to two photons

In Figure 8.7 is shown the ATLAS data from the search for the Higgs boson decaying to two
photons, in this case including all of the data taken in the 2011 and 2012 running periods
at centre-of-mass collision energies of 7 TeV and 8 TeV respectively. The horizontal axis is
the invariant mass of pairs of photons. The data is represented by the black filled circles
with uncertainty bars, and an enhancement above the continuous background distribution
represented by the dashed red line is seen around 125 GeV/c2 . The full red line is a fit to
the data using a background and the prediction for a Higgs boson of mass 126.8 GeV/c2 .
The lower half of the plot shows the same data with the background substracted away.

Figure 8.7: Evidence for the Higgs boson decaying to a pair of photons from
the ATLAS experiment. https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/
HIGG-2013-02/.

The CMS experiment produced a very similar result to ATLAS at the same time, which of
course gave greater weight to the hypothesis that a new particle was indeed being produced.

79
8.6.2 Higgs Decaying to two Z bosons

In Figure 8.8 is shown the ATLAS data from the search for the Higgs boson decaying to two
Z bosons, again including all of the data taken in the 2011 and 2012 running periods. The
horizontal axis is in this case the invariant mass of four leptons, either 2e+ 2e −, 2µ+ 2µ− , or
e+ e− µ+ µ− . The data is again represented by the black filled circles with uncertainty bars,
and the purple and light blue regions added together represent what would be expected to be
observed in the absence of a Higgs boson. An enhancement is again seen around 125 GeV/c2 ,
only a handful of events in this case but significantly above the background prediction. The
red region is the prediction for a Higgs boson of mass 124.3 GeV/c2 .

Figure 8.8: Evidence for the Higgs boson decaying to a pair of Z bosons from
the ATLAS experiment. https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/
HIGG-2013-02/.

Again the CMS experiment produced a very similar result. One can see in these distributions
that the signal peaks representing the Higgs boson candidates are quite broad, a few GeV
in width. The expected decay width of the Higgs boson is much less, around 4 MeV, which
means that the width of the peaks here is the result of finite detector resolution, not the
intrinsic width of the Higgs boson. When all of the evidence now available is considered, the
best current estimate of the Higgs boson mass is 125.09 ± 0.24 GeV/c2 , taking the current
Particle Data Group listings.

8.6.3 Is this a Standard Model Higgs?

Subsequent studies by ATLAS and CMS have demonstrated tha the spin and parity J P of
the new boson is consistent with being 0+ , as expected for the Standard Model Higgs boson.
Note that we will discuss parity when we get to discrete symmetries later in the course. The

80
other major test is whether it decays to all final states expected, and in the proportions and
with overall rate consistent with the Standard Model predictions. At the time of writing
the Higgs boson has been determined to decay to the five final states depicted in Figure 8.9,
which shows only the ATLAS data. Decays both to final state bosons and fermions have
now been observed. The quantity µ represents the ratio of the measured rate of decay to any
given final state to that predicted by the Standard Model, known as the signal strength. For
a Standard Model Higgs boson, µ should therefore have the value 1 for all decay channels. At
right in the figure, the vertical black line in all cases represents the value of µ obtained, with
the green band showing the one-standard deviation uncertainty on the value of µ. Within
the still quite large uncertainties, everything to date appears consistent with Standard Model
expectations.

Figure 8.9: The ATLAS and CMS measurements of the signal strength for Higgs boson
production in the decay modes in which it has been observed to date. https://atlas.web.
cern.ch/Atlas/GROUPS/PHYSICS/CONFNOTES/ATLAS-CONF-2015-044/fig_12.png.

8.7 The Higgs beyond the LHC?

What is required to further study the properties of the newly discovered Higgs boson is more
data. Compared to the first run of the LHC from 2010-2012, when the maximum collision
energy reached was 8 TeV, the machine is currently running at a much higher centre-of-
mass collision energy of 13 TeV, which is approaching its design specification of 14 TeV. At
these higher energies, the probability of producing Higgs bosons goes up significantly, and
along with improvements to the intensity of the proton beams, the number of Higgs bosons
available for study should increase at a good rate.

81
Precision studies of Higgs boson properties could be done at a future e+ e− collider. The
advantage of an electron-positron collider is the backgrounds from strongly interacting par-
ticles, the quarks and gluons, is much lower, since there are none of these in the colliding
particles. Once the mass of the Higgs boson is known, as it now is, the collider specifications
can be chosen to maximise the production rate of Higgs bosons.
Figure 8.10 shows a schematic of the proposed Internaltional Linear Collider (ILC), which
may one day be built, most likely in Japan, to study the Higgs boson in detail.

Figure 8.10: The layout of the proposed International Linear Collider (ILC). Graphic cour-
tesy of ILC.

The centre-of-mass energy of the collisions at the ILC is likely to be 500 GeV.It is not
practical to build the machine with a collision energy corresponding to the rest mass energy
of the Higgs boson, since the small coupling of the Higgs boson to electron and positron
(which are extremely light) is so small that the rate of Higgs boson production would be too
low. Instead, processes such as those depicted in Figures 8.11 and 8.12 will be used, where
the electron and positron produce virtual particles, in this case W and Z bosons, which
couple more readily to the Higgs boson because of their much greater mass.
Whether it is the ILC or some other machine which is built next, there is a lot still to
be learned about the Higgs boson, either at that machine or at the current Large Hadron
Collider.

82
Figure 8.11: Production of a Higgs boson from electron-positron annihilation in association
with a Z boson.

Figure 8.12: Production of a Higgs boson from electron-positron annihilation via fusion of
two W bosons.

83
Chapter 9

Detecting Particles

The material in this chapter provides a very brief introduction to the Large Hadron Collider,
the ATLAS experiment and the detection of particles.

Figure 9.1: Schematic layout of the LHC.

84
9.1 The LHC and the ATLAS Experiment

A schematic of the Large Hadron Collider (LHC) at CERN is given in Figure 9.1. The
large ring is 27km in circumference, located an average of around 100m underground on the
Swiss-French border.

Figure 9.2: The ATLAS detector. Note the scale, given by the human figures in the
schematic. The detector is around 44m in length and 25m in height. Image: CERN.

The LHC was finally successfully commissioned in 2010, and has been systematically ramping
up its collision energy. It collides two beams of high energy protons head-on in the centre of
two very large, general purpose detectors known as CMS and ATLAS.
The University of Sydney is a member of the ATLAS Collaboration. A schematic of the
ATLAS detector is shown in Figure 9.2.
In its first running period from 2010-2012, the LHC reached just over half design energy,
with beams of 4.0 TeV protons. Currently the LHC is operating with 6.5 TeV proton beams
and aims to increase that to 7.0 TeV proton beams, the design specifications, in the next
two or three years.

85
 Figure 9.3: Schematic of interactions of different particle types in a particle detector.

9.2 Some Elements of Particle Detectors

Modern particle detectors are extremely complex pieces of equipment, consisting of a number
of sub-detector systems designed with the following features:

Tracking To determine momentum of all the charged particles emerging from the in-
teraction point. This requires good spatial resolution, low Z (atomic number) material
and a magnetic field.

Vertexing To determine secondary decay vertices of short-lived states if possible. This

requires extremely good spatial resolution, and low Z material.

Calorimetry To determine the energy of electromagnetic showers (from electrons and

photons) and of neutral hadrons. This requires high Z material to contain the shower
and a means of detecting the energy deposited.

Particle Identification To separate e.g. charged pions from muons and electrons.

Fast electronics and computing To record signals from the detector in a high event
rate environment. This encompasses triggering, to only record interactions of interest.

Solid Angle Coverage To minimise regions where particles can escape the detector.

Figure 9.3 gives a schematic of the different elements of a typical particle detector, showing
different types of particles and their characteristic penetration. In Figure 9.4 a more realistic
schematic, specifically showing a segment of the ATLAS detector.

86
Figure 9.4: A more realistic depiction of the interaction of different particle types in a wedge
of the ATLAS detector, viewed along the beam direction.

9.3 Reconstructing Events

Here we review material that we have met previously in Section 3.8, on how to reconstruct a
short-lived parent particle decaying to daughter particles which can be detected in a detec-
tor. Recall that we can use conservation of energy and momentum, and special relativistic
concepts to do this.
Suppose we have the decay A → B + C. Since, for the 4-vectors of these particles
PA = PB + PC
when we measure PB and PC we can calculate PA . Remembering that
PA2 = EA2 /c2 − p⃗A .p⃗A ≡ m2a c2
we see that we can calculate the mass of the decaying particle under the hypothesis that the
above decay actually occurred. Clearly this can be generalized to N particles.
By way of an addition example to the one we saw previously, the table below gives the
momentum and energy components of the pions from the decay of a KS meson (the mea-
surement uncertainties on these quantities have been omitted). One can use the principle just
discussed, and the information in the table to show that the decaying particle is consistent
with being a neutral kaon.

Particle px c (GeV) py c (GeV) pz c (GeV) E (GeV)

π− 2.80879 -0.51130 0.45066 2.89367
+
π 0.76380 0.04410 0.04419 0.77895

87
Conservation of energy and momentum means that the 4-momentum of the kaon before decay
must equal the sum of the 4-momenta of the two pions after the decay. So by adding the two
4-momenta
√ of the pions, and finding the length of the resultant 4-vector from which it follows
E 2 − p2 c2 = mc2 , we can check whether m corresponds to the mass of a known particle,
in this case the K 0 . Using the data given, the mass obtained is 0.511 GeV/c2 (check this for
yourself), reasonably close to the neutral K mass of around 0.498 GeV/c2 . To really check for
compatibility we would need the uncertainties on the energy and momentum measurements.

9.4 Hands-on exercises

There are a set of on-line exercises which you are invited to try, which will reinforce some of
this material. In them, you can classify a handful of real collision “events” in the ATLAS
detector using the following website and associated programs:
https://atlas.physicsmasterclasses.org/en/index.htm
and in particular the “Z-Path” link:
https://atlas.physicsmasterclasses.org/en/zpath.htm
where you could try for example following the link “Practice!” link on the right-hand side
of the ”Identifying Particles” or Identifying Events” pages.
Further, it is possible to explore how the ATLAS detector identifies different types of particles
and classifies collision “events”. In particular in these exercises you can look at decays of Z
bosons and Higgs bosons, decays such as, for example

Z 0 → µ+ + µ−
H → γ+γ

and similar. The Higgs (and Z 0 ) are very short-lived and can be identified from their decay
products as outlined in the previous chapter. For an individual decay we can never be sure
that we have a Higgs as opposed to something else giving the same “signature”, for example
two unrelated photons from other processes which just happen to give an invariant mass
close to the Higgs rest mass. By collecting enough examples, however, a peak at the Higgs
mass would become visible above the “background”. This is how the Higgs was discovered.

88
Chapter 10

Symmetry

10.1 Symmetry

What do we mean by symmetry? A symmetry occurs when we apply a transformation to

all particles of a system and the laws of physics remain unchanged. This means that there
is no experiment you can do to tell if the transformation has been made.

Figure 10.1: Mirror-flipped cars

Consider the car in the left-hand image in Figure 10.1. If we mirror-flip the car (and the
driver) we expect the performance to be the same. Note, that in this macroscopic example,
the car symmetry is actually not perfect.
There are of course, many more examples that you can think of.

• If we translate all of the components of the car 1 metre to the left, the car is unchanged.

• If we rotate all of the components of the car by 45◦ , the car is unchanged.

• If we wait a day (translate in time), the car is unchanged.

89
That’s all quite obvious. But, symmetries have a deeper meaning in physics.

10.2 Noether’s Theorem

A deep principle in physics is due to Emmy Noether (1882-1936).

For every fundamental symmetry there is a conserved quantity

As an example, consider a translation of a system

x → x′ = x + δx

The Hamiltonian for a free particle is

1 ∂2
H=−
2m ∂x2
and so

H(x′ ) = H(x + δx) = H(x)

and the Hamiltonian is invariant under translation. We can define a translation operator

D̂ψ(x) = ψ(x + δx)

where ψ(x) is an arbitrary wavefunction. We can expand ψ(x + δx) = ψ(x) + δx ∂ψ(x)
∂x
and
see that
i ∂
D̂ = 1 + δx p̂ p̂ = −iℏ
ℏ ∂x
Define a new wavefunction ψn (x) = H(x) ψ(x) and apply D̂

D̂ψn (x) = ψn (x + δx) = H(x + δx)ψ(x + δx) = H(x)ψ(x + δx) = H(x)D̂ψ(x)

[ ]
D̂H(x) − H(x)D̂ ψ(x) = 0

From quantum mechanics, anything that commutes with the Hamiltonian is a constant of
i
motion. Substituting D̂ = 1 + δx p̂ into the commutation relation we get
ℏ
[p̂H(x) − H(x)p̂] ψ(x) = 0

and momentum is a constant of motion.

Other symmetries lead to other conservation laws, as we will explore further below.

90
10.3 Conservation Laws

That the quantities listed in Table 10.1 are conserved is familiar to you from classical physics.
The last column of the table indicates which symmetry relates to the conserved quantity
through Noether’s Theorem. It can also be seen that all of the three interactions of the
Standard Model respect these symmetries.
Quantity Strong EM Weak Symmetry
Momentum Yes Yes Yes Spatial Translation
Energy Yes Yes Yes Time Translation
Angular Momentum Yes Yes Yes Spatial Rotation

Table 10.1: Some conserved quantities, and their origin in symmetry considerations.

We will now discuss these conservation laws in turn.

10.3.1 Energy and Momentum

These are amongst the most fundamental conservation laws. Even if you can draw a le-
gitimate Feynman diagram, if you cannot conserve energy and momentum an interaction
cannot proceed.
This can be easily seem in the context of decays. The sum of the rest masses of the daughter
particles from a decay cannot be greater than the mass of the parent particle.
Remember from special relativity (refer back to Chapter 3) that energy and momentum form
a 4-vector.
P = (E/c, p) = (E/c, px , py , pz )
The dot product of two 4-vectors is an invariant, that is, has the same value in all Lorentz
frames of reference.
P.P = (E/c)(E/c) − p.p = (E/c)2 − p2 = m2 c2
where p = |p| and m is the rest mass of the particle.

10.3.2 Exercises on Energy and Momentum Conservation

You might like to consider the following exercises related to energy and momentum conser-
vation and particle physics processes.

1. Is this decay possible on energy and momentum grounds?

B0 → π+ + π−
(10.1)
Mass (GeV/c2 ) 5.2796 → 0.1396 + 0.1396

91
 2. Is this decay possible on energy and momentum grounds?
Σ+ → Λ0 + µ+ + νµ
(10.2)
Mass (GeV/c ) 1.1894 → 1.1156 + 0.1057 + 0.0000
2

3. Is this scattering process possible on energy and momentum grounds?

π+ + π− → p + p
(10.3)
Mass (GeV/c ) 0.1396 + 0.1396 → 0.9383 + 0.9383
2

You might like to ask yourself the question – is mass conserved in an interaction? (The
answer is no – why?)

10.3.3 Angular Momentum

Angular momentum is conserved in all interactions. Remember, there are two components to
total angular momentum, spin angular momentum and orbital angular momentum. Because
individual components of angular momentum are quantized, the total angular momentum is
quantized. One can review these concepts. by referring back to Chapter 2.
Note that most particles in the lowest energy state have orbital angular momentym L = 0.
As an example of thinking about the angular momentum aspect of interactions, consider the
following decay, where the spin of the particles participating is indicated.
Ω− → K − + K̄ 0
Spin (ℏ) 32 → 0 + 0
This decay cannot occur. The total angular momentum from the right-hand side will be
J = L + S1 + S2
where in this case S1 is the spin vector of the charged kaon, S2 is the spin vector of the
neutral kaon, and L is the orbital angular momentum between them. There is no way that
we can add two integer spin vectors to produce a total angular momentum which is half-
integral, since the orbital angular momentum must be integral. This is also clear from just
considering the z-components, which have to add arithmetically.

10.3.4 Exercises on Angular Momentum Conservation

You might like to consider the following exercises related to angular momentum conservation
and particle physics processes.

1. Is this decay possible on angular momentum grounds?

J/ψ → π + + π − + π 0
(10.4)
Spin (ℏ) 1 → 0 + 0 + 0

92
 2. Is this scattering process possible on angular momentum grounds?

π + + p → ∆++ + π 0
(10.5)
Spin (ℏ) 0 + 12 → 3
2
+ 0

You might also like to ask yourself the question – is spin conserved in an interaction? (Again
the answer is no – why?)

93
Chapter 11

More on Symmetry

11.1 More Conservation Laws

We will examine a further conservation law with which you will be familiar already, electric
charge, and several less familiar conservation laws, which are summarised in Table 11.1.
Unlike the cases in the previous Chapter, however, these latter conservation laws are not
necessarily respected by all Standard Model forces.

Quantity Strong EM Weak

Electric Charge Yes Yes Yes
Baryon Number Yes Yes Yes(?)
Lepton Number - Yes Yes(?)

Table 11.1: Conservation laws of electric charge, baryon number and lepton number.

Lepton number is only ascribed to particles which do not feel the strong force. The question
marks against the baryon and lepton number entries in the table for the weak force reflect the
fact that the situation with regard to these symmetries is unclear when potential Beyond-
Standard-Model (BSM) scenarios are considered. We will return to this towards the end of
the module.
We will also define “quark numbers” (related to quark flavour), as described in Table 11.2.
Here we see that the weak interaction does not conserve these quantities. This should be
clear from the observation that the weak force, through coupling to the charged W bosons,
can change the flavour of a quark, whereas this is not the case for the strong and EM
interactions.
The quotes on “Upness”, “Downness”, “Charmness”, “Bottomness” and “Topness” indicate
that these terms are not in common use in the field. Later, when we consider the quark
model, we will see that there is in fact a concept known as strong isospin which is related to
what we refer to here as “upness” and “downness”.

94
 Quantity Strong EM Weak
“Upness” Yes Yes No
“Downness” Yes Yes No
Strangeness Yes Yes No
“Charmness” Yes Yes No
“Bottomness” Yes Yes No
“Topness” Yes Yes No

Table 11.2: Conservation laws related to the flavour of quarks.

11.1.1 Electric Charge

As with energy and momentum, electric charge is conserved in all interactions. Consider the
case of the decays of charged pions. The decay channels below are classified as to whether
they conserve electric charge or not, and are hence allowed or forbidden (on the criterion of
electric charge conservation only, there may of course be other considerations which forbid
a process even if electric charge is conserved).

π + → µ+ + νµ π + → µ− + ν̄µ
π − → µ− + ν̄µ π − → µ+ + νµ

Allowed F orbidden

Electric charge can only be created or destroyed in equal positive and negative pairs in any
process.

11.1.2 Baryon Number

We met baryon number earlier in the module. The baryon number of a particle made from
quarks is

1
B = (nq − nq̄ )
3
and we have already noted that B can only take the values −1, 0 or +1. The common quark
and antiquark combinations and their baryon numbers, along with examples, are given in
Table 11.3.
Baryon number is conserved in (almost) all interactions. Here are examples of processes
which conserve baryon number and do not conserve baryon number. The first is allowed and
the second forbidden.
p + n → p + n + p + p̄
B = 1 + 1 → 1 + 1 + 1 + −1
Allowed

95
 Quarks B Example Name
qqq +1 p (uud), n (udd) Baryon
q̄ q̄ q̄ −1 ¯ n̄ (ūd¯d)
p̄ (ūūd), ¯ Anti-Baryon
+ ¯ −
q q̄ 0 π (ud), K (sū) (Anti-)Meson

Table 11.3: Common quark/antiquark combinations and their baryon number, with exam-
ples.

p + n → p + µ+ + µ−
B= 1 + 1 → 1 + 0 + 0
F orbidden

Note that there is no equivalent concept of meson number, and no associated law of con-
servation of meson number. This is illustrated by the existence of the following process for
charged pion decay:

π − → µ− + ν̄µ

11.1.3 Lepton Number

There are three distinct lepton numbers, associated with the three flavours, or generations,
of lepton.

Lℓ ≡ N (ℓ− ) − N (ℓ+ ) + N (νℓ ) − N (ν̄ℓ )

where ℓ = e, µ or τ .
Each of these, Le , Lµ and Lτ , are (almost) conserved. The reason for the “almost” is that
this is not the case for the phenomenon of neutrino oscillations, which will form the topic
of a later lecture. In that case, however, total lepton number L ≡ Le + Lµ + Lτ is still
conserved.
Here are examples of an allowed and forbidden decay mode of the tau lepton, analysed in
terms of lepton numbers.

τ− → e− + ν̄e + ντ
Le = 0 → 1 + −1 + 0
Lµ = 0 → 0 + 0 + 0
Lτ = 1 → 0 + 0 + 1
Allowed

96
 τ− → e− + ν̄e + νµ
Le = 0 → 1 + −1 + 0
Lµ = 0 → 0 + 0 + 1
Lτ = 1 → 0 + 0 + 0
F orbidden
Finally here is an example of an allowed scattering process where hadrons are also involved.
Note that the convention is that anything that is not a lepton is assigned Lℓ = 0.
ν̄µ + p → µ+ + n
Lµ = −1 + 0 → −1 + 0

11.1.4 Quark Numbers

There are four quark numbers, defined as follows:

S ≡ −[N (s) − N (s̄)] C ≡ [N (c) − N (c̄)]
B̃ ≡ −[N (b) − N (b̄)] T ≡ [N (t) − N (t̄)]
where N (q) designates the number of quarks of a particular flavour q, with an obvious
meaning for N (q̄). The convention that explains the source of the minus signs when treating
the strange and bottom quarks is that the definitions for positively charged quark flavours
carry a plus sign and for negatively charged quarks carry a minus sign. The tilde over the B
in the definition of the bottom quark number is simply to distinguish the symbol from that
for baryon number, which we have previously denoted B.
It would also be possible, as you might imagine, to define quark numbers related to the up
and down quarks. The reason this is not generally done is a historical convention, since the
up and down quarks are the subject of a related concept known as strong isospin, which we
will discuss in the following section.
Some examples of mesons and their corresponding quark numbers are given in Table 11.4.

P article Q S C B̃
π− (dū) −1 0 0 0
K− (sū) −1 −1 0 0
D− (dc̄) −1 0 −1 0
B− (bū) −1 0 0 −1

Table 11.4: Some mesons and their corresponding quark numbers.

As the strong and EM interactions create or destroy like quark/antiquark pairs, these quark
numbers are conserved in these interactions. Here is an example:
π − (dū) + p (uud) → K o (ds̄) + Λ (uds)
S= 0 + 0 → 1 + −1

97
As the weak interaction produces unlike quark/antiquark pairs, it does not have to conserve
these quark numbers, as illustrated in the decay of the Λ meson.

Λ (uds) → π − (dū) + p (uud)

S= −1 → 0 + 0

98
Chapter 12

Discrete Symmetries

Symmetries such as spatial translation, time translation and rotation are examples of con-
tinuous symmetries. In this chapter we will consider three discrete symmetries – charge
conjugation, parity and time reversal.

12.1 Charge Conjugation

Charge conjugation involves replacing all positive charges with negative charges and vice
versa. For an electron, for example,
e− → e+
The operator also inverts other quantum numbers, such as lepton number, baryon number
and strangeness, but leaves mass, energy, momentum and spin unchanged. It thus changes
particle to antiparticle.
Charge conjugation also reverses the direction of electric and magnetic fields, which makes
sense when one considers that classically a static distribution of electric charge serves as the
the source of an electric field and a current as the source of a magnetic field.
The electromagnetic interaction is invariant under charge conjugation. We would expect
that if we consider motion of a charged particle in an electromagnetic field, then an equally
valid physical system would be one in which the same type particle but of opposite electric
charge moves in an electromagnetic field which has had the directions of the electric and
magnetic field components reversed. As another example, in terms of atomic energy levels,
we would expect that the spectrum of anti-hydrogen ought to be the same as the spectrum
of hydrogen.
Since charge conjugation flips electromagnetic fields and the quantum of the electromagnetic
field is the photon, we would expect the state representing a photon to flip. Representing
the charge conjugation operator by ĉ, we would write this as
Ĉ|ψγ ⟩ = −1|ψγ ⟩

99
which is in the form of an eigenvalue equation. More generally, for some state |ψ⟩, we would
write

Ĉ|ψ⟩ = εC |ψ⟩

where |ψ⟩ is an eigenfunction of ĉ and εC the corresponding eigenvalue. If a system exhibits

charge conjugation symmetry and a particle is represented by |ψ⟩, then εC is the value of a
quantum number characterising that particle.

12.1.1 Decay of the neutral pion

We can use this concept of symmetry of the electromagnetic interaction under charge con-
jugation to understand the way in which a neutral pion decays. The decay

π0 → γ + γ

is the normal way in which the π 0 is observed to decay. In terms of charge conjugation,
considering eigenvalues, we write

εC (π 0 ) = εC (γ)εC (γ) = (−1)2 = +1

Th decay

π0 → γ + γ + γ

is not observed in nature and we can see that it is forbidden since

εC (3γ) = εC (γ)εC (γ)εC (γ) = (−1)3 = −1

and charge conjugation would be violated if this decay occurred.

The strong interactions are also invariant under charge conjugation.
Note that the charge conjugation quantum number is a multiplicative quantum number,
which is characteristic of discrete symmetry operations.

12.2 Parity

The parity transformations involves a simple spatial operation:

 
x → −x
 y → −y 
z → −z

with the time component left unchanged. This can be seen is a reflection of a physical system
through the origin of the coordinate system. It can also be viewed as a mirror reflection in

100
a plane which passes through the origin, followed by a rotation of 180◦ about an axis which
contains the origin and is perpendicular to the mirror plane.
We can define an operator, P̂ , to make this transformation

P̂ |ψ⟩ = p|ψ⟩

where |ψ⟩ is an eigenfunction of the parity operator and p the corresponding eigenvalue. If
|ψ⟩ represents a particle then p is a quantum number characteristic of that particle which
we call its parity.
If the parity operator is applied twice we see expect to get back to the state we started with,
so

P̂ P̂ |ψ⟩ = P̂ p|ψ⟩ = p2 |ψ⟩ ≡ |ψ⟩

and the eigenvalues are p = ±1, assuming that p is real.

Note that parity transformations also reverse the direction of momentum but leave spin, or
more generally angular momentum, unchanged.

12.2.1 Intrinsic Parity of Fermions

Quantum mechanics, or more correctly quantum field theory, tells us that fermions can
only be created in particle- antiparticle pairs. A consequence of this is that the parity of
a fermion-antifermion pair is a well- defined quantity, but the intrinsic parity of a fermion
is arbitrary since it is unmeasurable. Hence if we want to define the intrinsic parity of a
fermion we employ a convention:

quark, lepton parity = +1

Once we have this convention, then what can be measured is the relative parity between
fermion and antifermion, and this is found to be negative. It then follows that:

antiquark, antilepton parity = −1

To see this formally one needs the Dirac equation, which we will not study in these lectures.

12.2.2 Meson and Baryon Parity

Hadrons, i.e. mesons and baryons, are composed of quarks and antiquarks, and so the
parities of the composite particle are obtained by combining the parities of the components.
Taking into account the above convention, for a meson it follows that:

pmeson = (−1)L pq1 pq̄2

101
where L is the orbital angular momentum between quark q1 and antiquark q̄2 , and pq1 and
pq̄2 are their intrinsic parities. The factor (−1)L arises from the spherical harmonic functions
which form part of the wavefunction for the meson. You may have encountered this factor
when solving Schrödinger’s equation for a hydrogen atom.
In the case of a baryon, the analogous expression is
pbaryon = (−1)L12 (−1)L3 pq1 pq2 pq3
The situation with orbital angular momentum is more complicated in this situation with
three quark constituents. there will be orbital angular momentum between each pair of
quarks, and all of these must be added vectorially to give the total orbital angular momen-
tum, following the quantum mechanical rules for adding momenta. In the expression above,
L12 represents the orbital angular momentum between q1 and q2 , and then considering q1 q2
to be a single system, L3 represents the orbital angular momentum between this system and
q3 . Clearly the decomposition could be done in a different order, although the end result
would be the same. pq1 , pq2 and pq3 are the intrinsic parities of the three quarks.
For the lowest mass mesons and baryons, it is generally the case that the quarks will be
in a configuration of minimum orbital angular momentum, L = 0, and from this the above
expressions imply

meson, antimeson −1, −1

baryon, antibaryon +1, −1

As with charge conjugation, parity must be conserved in strong and electromagnetic inter-
actions.

12.3 Time Reversal

Time reversal amounts to the following transformation:

( )
t → −t
with the spatial components left unchanged.
The operation of time reversal reverses the sign of momentum and angular momentum vec-
tors, even though it leaves the positions of particles unchanged. A movie showing the evolu-
tion of the physical system in time would appear to an external observer to run backwards
in the time reversed case.
Basically this means we can run any particle interaction backwards. If time reversal sym-
metry applies, the backwards reaction must also represent a viable situation which could be
observed in Nature. Hence, for example, the two processes
A+B →C +D
C +D →A+B

102
should have the same rate.
Time reversal, like charge conjugation and parity, is a symmetry of the strong and electro-
magnetic interactions. interactions.

12.4 Are these discrete symmetries universal?

It was once thought that the symmetries of charge conjugation, parity and time reversal
were absolute. The symmetries implied the existence of conserved quantities as required by
Noether’s Theorem, and applying them would not alter the way we see the universe behave.
As we’ve stated, the strong and electromagnetic interactions do appear to completely obey
these symmetries. What about the weak interaction?

12.4.1 The Fall of Parity

Parity conservation was indeed considered obvious and universal for a long time. In 1956
however, Lee and Yang, who received the Nobel Prize in 1957, questioned this for the weak
interaction, realising that there had been scant if any experimental tests. They suggested
a test involving radioactive beta decay, which occurs via the weak interaction, with the
experiment then being carried out by C.S. Wu and her team.
Wu looked at the β-decay of 60 Co, with the situation represented schematically in Figure 12.1.

Figure 12.1: β-decay of radioactive 60 Co in the normal world and in the mirror world. Image
from https://www.nist.gov/pml/fall-parity/parity-whats-not-conserved.

A sample of cobalt nuclei are all aligned such that their spin vectors are pointing in the same

103
direction, say up in the normal-world image in Figure 12.1. This is represented by the nuclei
spinning clockwise with respect to an axis pointing vertically upwards. In the mirror the
spin vectors would point downwards, as the nuclei appear to spin anticlockwise with respect
to the same vertical axis, which doesn’t change in the mirror.
The direction of emission of the electrons (β-rays) is measured. Parity conservation would
result in isotropic emission, since if there was a prefered direction of emission with respect
to the axis of spin of the nuclei, then that direction would be opposite in the mirror world.
It would then be possible to tell from the result of the experiment whether one was in the
normal world or the mirror world, which would imply that parity is violated, since the laws
of physics would differ between the normal and mirror world.
Figure 12.1 suggests an asymmetric situation in which more electrons are emitted in the
hemisphere opposite to the spin direction than in the one in which it lies. This is indeed
what Wu’s team observed when they supercooled 60 Co nuclei and placed them in a magnetic
field to align their spins - the experiment showed non-isotropic β emission, demonstrating
that the weak interaction violates parity. The opposite was seen with β + emission (58 Co) in
subsequent experiments, with positrons emitted preferentially in the direction aligned with
the spin vector.
Note that as presented above the Wu experiment actually shows that mirror symmetry is
violated by the weak interaction. The parity transformation involves an extra rotation of
180◦ about the axis perpendicular to the mirror. However, since we already know that all
interactions exhibit symmetry under rotations, a violation of mirror symmetry also implies
a violation of parity, since the final rotation cannot “undo” the violation.
The Wu result rocked the foundations of physics to the core, as it was a completely unex-
pected result. We will gain a little more insight into the nature of parity violation when we
discuss the properties of neutrinos in a later chapter, where we will also see that the weak
interaction violates charge conjugation symmetry.

104
Chapter 13

The Quark Model

In this chapter we examine the quark model, which was originally developed in the 1960s to
explain the proliferation of hadronic states which were turning up in scattering experiments.
Firstly we look at two quantities conserved in the strong interaction which were found useful
in classifying the various hadrons known at the time.

13.1 Strong Isopin

Strong Isospin is a set of quantum numbers that is conserved in the strong interaction. In an
analogous way to spin, particles in a strong isospin family are seen as different projections
of the strong isospin vector. For example, the proton and neutron represent a strong isospin
family:
( )
1 p Iz = + 12
I=2
n Iz = − 12
Notice the change of 1 between the two Iz values.
Here are some other examples of strong isospin families:
( ) ( + ) ( ) ( 0 ) ( )
1 p K K̄ 0 Ξ Iz = + 12
I=2
n K0 K− Ξ− Iz = − 21

   +   
π+ Σ Iz = +1
I=1  π 0   Σ0   Iz = 0 
π− Σ− Iz = −1

   
∆++ Iz = + 23
 ∆+   Iz = + 1 
(Λo ) I, Iz = 0    2  I= 3
 ∆o   Iz = − 1  2
2
∆− Iz = − 23

105
In strong interactions, isospin is vectorially conserved. the mathematical rules for combining
strong isospin vectors are identical to those for combining angular momenta, even though
the physical meaning of strong isospin is completely unrelated to that of angular momentum.
For example, we can visualise the two strong isospin states of the nucleon as in Figure 13.1.
STRONG ISOSPIN SPACE
z

+1/2 Proton I=1/2, Iz = +1/2

−1/2 Neutron I=1/2, Iz = −1/2

Figure 13.1: The nucleon strong isospin doublet represented as vectors in strong isospin
space.

Thus the total I vectors of the initial state in a strong interaction must be able to add in
such a way as to match the sum of the total I vectors of the final state, and the projections
Iz of the initial state must sum arithmetically to the same value as those of the final state.
As an example, consider the following potential decay of a Λ baryon.
Λ0 → p + π −
I = 0 → 12 + 1
Iz = 0 → 12 − 1
F orbidden
Strong isospin considerations tell us that this decay cannot occur via the strong interaction.
If we vectorially add an object of half-integral strong isospin (the proton) to an object of
integral strong isospin (the pion) then the only possible results can be half integral strong
isospin, in this example I = 1/2 or I = 3/2. This is incompatible with the strong isospin of
the Λ0 being I = 0, hence the decay is forbidden. Note that since the weak interaction does
not respect strong isospin conservation, the decay can still occur via the weak interaction,
which in fact it does.

13.2 Strong Hypercharge

Strong isospin as a way to classify hadrons, or strongly interacting particles. Another quan-
tum number which is useful for the same classification purpose is strong hypercharge, defined

106
as

Y =B+S

where B is the baryon number of the state in question and S is the strangeness. The concept
of strong hypercharge was introduced at the time when only objects containing u, d and s
quarks were known, even though it was not then known that the objects contained quarks.
The strong hypercharge, strong isospin and electric charge of a state are related by

Q 1 1
= Iz + Y = Iz + (B + S)
e 2 2

where the electric charge Q is being expressed in units of e, the charge of a proton or positron.
This relationship can be generalised to more quark flavours, but for now we will just stick
with three.
As an aside, note that it is possible instead of using the strong hypercharge and third
component of strong isospin, Y and Iz to classify states, one could instead use the quark
numbers “upness” and “downness” introduced earlier.
1 1
U = Iz + Y + B D = Iz − Y − B
2 2
and write the previous relationship as
Q 1
= (B + U + D + S)
e 2
In fact
U +D
Iz =
2
This makes it clear that all flavours of quarks are being treated on an equal footing. It is for
historical reasons that strong isospin is and strong hyperchage are generally still used, even
though with the benefit of hindsight, we now know about up, down and strange quarks, as
discussed below.

13.3 Hadrons and the Quark Hypothesis

Figure 13.2 shows that if a two dimensional plot of strong hypercharge versus third com-
ponent of strong isospin is made, the lowest mass baryon states form families, known as
multiplets. Similar types of plots could be made for the mesons known at the time. What is
more, the masses of states which form horizontal families of strong isospin in these plots are

107
Figure 13.2: The J = 12 and positive parity baryon octet (left) and the J = 32 and positive
parity baryon decuplet (right). The masses in blue refer to the approximate masses of the
particles in that row and are in units of MeV/c2 .

all very similar. In the early 1960s these facts were known, but an explanation as to why
this was so was actively being sought.
The breakthrough in understanding of why there are so many different hadron states and why
they form the multiplets of strong isospin and strong hypercharge that they do came in 1964,
when Gell-Mann and Zweig independently hypothesised that hadrons could be described in
terms of mathematical constructs which subsequently came to be known as quarks (Gell-
Mann’s term, Zweig had called them aces). Under the quark hypothesis, baryons consist
of three quarks (qqq), antibaryons consist of three antiquarks (q̄ q̄ q̄), and mesons consist of
a quark-antiquark pair (q q̄). The quarks came in three flavours (u, d, s). The need for
heavier quarks was not known at that time since heavier mesons and baryons had not been
discovered. The first charmed mesons were discovered in 1974.
Under this scheme, the up and down quarks (u, d) form a strong isospin doublet (I = 12 ) with
S = 0. The strange quark (s) forms a strong isospin singlet with (I = 0) and with S = −1.
Antiquarks have opposite third component of strong isospin and strong hypercharge quantum
numbers to their corresponding quarks, with the situation summarised in Table 13.1.

F lavour B J I Iz S Y Q/e
u, ū ( 13 , − 13 ) 1
2
1
2
(+ 2 , − 21 )
1
0 (+ 31 , − 13 ) (+ 23 , − 23 )
d, d¯ ( 31 , − 13 ) 1
2
1
2
(− 12 , + 21 ) 0 (+ 31 , − 13 ) (− 13 , + 13 )
s, s̄ ( 31 , − 13 ) 1
2
0 0 (−1, +1) (− 23 , + 32 ) (− 13 , + 13 )

Table 13.1: The quantum numbers of up, down and strange quarks.

The up, down and strange quarks and antiquarks form triplets when plotted on the same
type of Y versus I3 diagram as we have seen for the hadrons of which they are composed.
This is illustrated in Figure 13.3.

108
 Y Y

1 1
d u s Ι=0
Ι=1/2
-1 1
I3 -1 1
I3
s Ι=0 Ι=1/2
-1
u -1
d

Figure 13.3: The quark (left) and antiquark (right) triplets.

With 3 quark flavours, there are 27 possible qqq permutations

uuu, uud, udu ... sss

and 9 possible q q̄ permutations

¯ us̄ ... ss̄
uū, ud,

which can be formed.

It turns out that these are split up into different multiplets depending on the overall sym-
metry of the hadron state under permutations of the constituent quarks and antiquarks. For
the baryon states, there is one decuplet (10 states) of hadrons with spin J = 3/2 i.e. all
three quark spins aligned and positive parity; two octets (each 8 states) with J = 1/2, one
of positive parity and one negative; and a singlet state (1 state) with J = 1/2 and negative
parity. The first two of these, containing the lowest mass baryons in which the quarks have
no orbital angular momentum between them, were shown in Figure 13.2.
For the mesons, nonets (9 state multiplets) are formed. When there is no orbital angular
momentum between the quark and antiquark, then either the spins of the two particles will
be opposite, giving a spin 0 meson, or they will be aligned, giving a spin 1 meson. These
two nonets are shown in Figure 13.4.
One reason that the quark model was so compelling was that not all of the particles in the
multiplets were known at the time that quarks were postulated. In particular, within the
baryon decuplet, shown in Figure 13.5, the Ω− state containing three strange quarks had not
been observed, and Gell-Mann was able to predict its existence and its approximate mass.
The Ω− was subsequently discovered, vindicating the model.

109
Figure 13.4: The spin 0 (left) and spin 1 (right) lowest-mass meson nonets, with quark
content of each particle shown.

Figure 13.5: The J = 32 baryon decuplet, with quark content indicated. See Figure 13.2 for
the masses of the baryons.

110
Chapter 14

More on the Quark Model

14.1 Symmetry of the Baryon States

We can look in a little more detail how to combine the three quark flavours in a baryon to
construct a flavour state. If we label the combination of three quark flavours as q1 q2 q3 and
treat the position of a given quark as significant in this product of quark flavour states, then
first we can note that the baryon states |uuu⟩, |ddd⟩ and |sss⟩ are symmetric (S) under the
exchange of any two of the three quarks. This is not the case for states where two of the
quarks only are the same, e.g. |uud⟩, |udu⟩ and |duu⟩. We find that we can construct a
symmetric (S) state from these
1
|ψ⟩ = √ (|uud⟩ + |udu⟩ + |duu⟩)
3
i.e. we get the same state back if we swap the quarks in positions 1 and 2 (1 ↔ 2) in all
three parts of the sum, or the quarks in positions 2 and 3 or the quarks in positions 1 and
3. Alternatively, we can construct a state with mixed (M) symmetry
1
|ψ⟩ = √ (|udu⟩ − |duu⟩)
2
where we find the swap 1 ↔ 2 results in ψ → −ψ but 2 ↔ 3 results in ψ ↛= ψ, i.e. we do
not get the a multiple of the same state in the second case.
The uds combinations can form a symmetric (S) state
1
|ψ⟩ = √ (|uds⟩ + |dsu⟩ + |sud⟩ + |usd⟩ + |sdu⟩ + |dus⟩)
6
or they can combine to give an anti-symmetric (A) state
1
|ψ⟩ = √ (|uds⟩ + |dsu⟩ + |sud⟩ − |usd⟩ − |sdu⟩ − |dus⟩)
6

111
 Quarks Symmetry Q/e S I Iz Y
3
uuu S 2 0 2
+ 32 1
3
uud SM M 1 0 2
+ 12 1
udd SM M 0 0 3
2
− 12 1
ddd S −1 0 3
2
− 32 1
uus SM M 1 −1 1 +1 0
uds SM M M M A 0 −1 1 0 0
dds SM M −1 −1 1 −1 0
uss SM M 0 −2 12 + 12 −1
dss SM M −1 −2 12 − 12 −1
sss S −1 −3 0 0 −2

Table 14.1: Symmetry properties under permutation, and quantum numbers of 3-quark
flavour states.

or one of four mixed states. All of the possibilities are summarised in Table 14.1.
Notice from the table that there are 10 symmetric (S) states out of the 27 that it is possible
to construct. These are the the members of the J = 32 decuplet of Figure 13.5. While other
states are mixed they can be made symmetric also with appropriate choice of the spins of
the quarks. For example, although the proton would have mixed symmetry on the basis of
the table and flavour alone, we can constuct a state for a proton with spin ↑, for example,
as
|ψ⟩ = √1 [
18
2|u ↑ d ↓ u ↑⟩ + 2|d ↓ u ↑ u ↑⟩ + 2|u ↑ u ↑ d ↓⟩
−|u ↓ d ↑ u ↑⟩ − |d ↑ u ↑ u ↓⟩ − |u ↑ u ↓ d ↑⟩
−|u ↑ d ↑ u ↓⟩ − |d ↑ u ↓ u ↑⟩ − |u ↓ u ↑ d ↑⟩]
where the arrow following any quark flavour indicates its spin orientation. To convince
yourself that this wavefunction is indeed symmetric, try interchanging any two quarks. This
type of procedure can be followed for any of the 27 states, and all can be represented by a
symmetric state. The outcome is the multiplets (one decuplet, two octets and one singlet)
which were mentioned in the previous section.

14.1.1 A Problem with this Symmetry

Because baryons are fermions, there is a problem with the quark model as just presented
when the symmetry of the states that we have discussed is taken into account. In quantum
mechanics, fermions should be represented by antisymmetric states.
Consider the ∆++ particle which is at the top right-hand corner of the baryon decuplet in
Figure 13.5. The quark content is uuu, which means that the flavour part of this state

112
is completely symmetric with respect to the exchange of any two of the up quarks. The
quark spins are all up, ↑↑↑, so again the state is completely symmetric when quark spins
are considered. Finally the quarks are in the lowest energy state corresponding to orbital
angular momntum L = 0, so again the state is completely symmetric when the spatial part
is considered. In terms of the wavefunction of the state, we can write

ψ∆++ = ψspace ψf lavour ψspin

with all three components symmetric. This situation violates the Pauli exclusion prin-
ciple. Technically speaking, the state is symmetric, whereas fermions need to be in an
anti-symmetric state.
Somehow there must be a way to make the state anti-symmetric overall. This will require
the introduction of a new quantum number, colour, which we will meet in the next chapter,
on Quantum Chromodynamics (QCD).

14.2 Applications of Strong Isospin

Strong isospin considerations can be used to make predictions about the relative rates for
processes. For this the rules for coupling angular momentum states in quantum mechanics
can be used, as embodied in tables of Clebsch-Gordan coefficients. A copy of these Clebsch-
Gordan tables can be found in the Resources section of the module in BlackBoard.
As an example, when charged pions scatter off nucleons at appropriate energies, short-
lived baryons known as ∆ baryons can be produced. Compare the production rates for the
following two cases (all other considerations being equal):

π + + p → ∆++
π − + p → ∆o

Since pions have I = 1 and nucleons have I = 12 we can calculate the rates from the 1 × 1
2
Clebsch-Gordan coefficient table given in Figure 14.1.
Figure 14.1 can be interpreted as follows. For the first process, in terms of Iz , π + and p are
+1 and + 12 respectively. The circled pink boxes in the upper (part of) the figure correspond
to these Iz values. These can only combine to give (I, Iz ) of 32 , + 32 . This can be seen by
reading horizontally across from the pink boxes to the sole yellow one, and then reading up
to the blue ones, which give the (I, Iz ) values of the combination, which correspond to those
of the ∆++ .
For the second process, in terms of Iz , π − and p( are −1 1
( 1 + 21.) In this case, there are
) and
two possible (I, Iz ) combinations that can result, 2 , − 2 or 2 , − 2 , as can be seen by the
3 1

presence of two yellow boxes when reading across.

Using the table and the contents of the yellow boxes, we can construct the general wave-
function for these two processes. The rule is that the coefficient in the wavefunction for the

113
 Figure 14.1: Using the 1 × 1/2 Clebsch-Gordan table to combine pion and proton states.

combined result corresponding to each possible (I, Iz ) valueis given by the square root of the
absolute value of the content of the relevant yellow box. If the box contains a minus sign,
that is written in front of the coefficient.
For our example we get
( )
π + + p ≡ √32 , + 32
( ) √2 (1 1)
π− + p ≡ 1 3
3 2
, − 1
2
− 3 2, −2

In quantum mechanics, given a wavefunction which is the superposition of different eigenfunc-

tions, the probability of obtaining a particular eigenvalue when performing a measurement
is given by the value of the coefficient in from of that eigenfunction squared. In our example
the probabilities associated with the possible final (I, Iz ) outcomes is therefore
( )
π + + p ( 32 , + 32 , 100%) ( )
π − + p 32 , − 12 , 33% , 12 , − 21 , 67%

Recall that in terms of (I, Iz )

( )
∆++ ≡ ( 23 , + 32 )
∆o ≡ 32 , − 12
and, as strong isospin is conserved in the strong interaction
π + + p → ∆++ 100% of the time
π − + p → ∆o 33% of the time

noting that we do not produce a ∆0 when π − + p is in the ( 21 , − 12 ) state since the strong
isospin quantum numbers are not those of that particle.
Remember again that tables of coefficients such as these are the same ones that tell you how
to combine quantized angular momentum (e.g. electron orbital and spin AM). They can
be used to combine strong isospin states because the same basic quantum mechanical rules
apply.

114
Chapter 15

QCD

15.1 Colour and The Strong Force

In the preceding chapter, the issue of the symmetry of the ∆++ baryon state was raised. To
recap, the wavefunction can be written as a product of a spatial part, a flavour part and a
spin part, as follows:

∆++ = ψspace ψf lavour ψspin

Considering these in turn, it can be seen that ψspace is symmetric since there is no orbital
angular momentum between the quarks, L = 0. ψf lavour is symmetric since the particle is
composed of three identical upquarks, uuu. Finally, ψspin is symmetric since to produce a
spin 3/2 particle, it must be possible for all three up quark spins to be aligned. It follows
therefore that the total wavefunction is symmetric.
Remembering that baryons are fermions with half-integer spin of 21 , 32 . . . , they must obey
Pauli’s exclusion principle and therefore must have an anti-symmetric wavefunction. Hence
we have a contradiction.
The solution to this problem is to introduce a new quantum number known as colour to
differentiate between the quarks. Colour, or colour charge as it is also referred to, comes in
three varieties for quarks and a corresponding three varieties for antiquarks, as follows:

Quarks r = red g = green b = blue

Anti-Quarks r̄ = anti-red ḡ = anti-green b̄ = anti-blue

The way to view colour charge and colour anti-charge is through an analogy with the situation
for electric charge and the electromagnetic interaction. Rather than think of a single type
of electric charge which can be positive and negative, one could think of positive charge as
“charge” and negative charge as “anti-charge”. Hence if the charge of a proton is denoted
e, then the charge on an antiproton could be denoted ē (= −e). One difference between

115
electric charge and colour charge is that electric charge comes in a single variety whereas
colour charge comes in three varieties.
Hadrons are colour singlets in as much as they do not possess quantum numbers correspond-
ing to net colour. In baryons (qqq/q̄ q̄ q̄), the colours of the three quarks add to ‘colourless’.
In mesons (q q̄) the quarks come in colour/anti-colour pairs in such a way as to make the
meson ‘colourless’.
From the analogy with real colour it should be obvious where the terminology comes from,
although of course the two concepts are totally unrelated. Whilst the way we have just
introduced colour may seen adhoc, it has a firm mathematical footing in the mathematics
of the underlying field theory describing the strong interaction, Quantum Chromodynamics
or QCD.
At the level at which we are discussing colour here, the significance of having three varieties
of colour which combine in ways to make the composite objects of which they are composed,
the hadrons, colourless, builds in the experimental observation that we do not see composite
objects in Nature of the type qq, qq q̄, q q̄ q̄ etc. There is no way to make these objects have no
net colour, whereas there is for the known combinations qqq q̄ q̄ q̄, q q̄, and for the hypothesised
tetraquark and pentaquark states, q q̄q q̄ and qqqq q̄.

15.1.1 The ∆++ State Revisited

Taking into account the colour of quarks, the ∆++ wavefunction can now be written in the
following form

ψ∆++ = ψspace ψf lavour ψspin ψcolour

To make the whole state antisymmetric, the colour part ψcolour can be chosen antisymmetric.
An appropriate choice is
1
ψcolour = √ (rgb + gbr + brg − rbg − grb − bgr)
6
where the in each term the colour of the first, second and third up quarks has been specified.
The total ∆++ state becomes
|ψ∆++ ⟩ = √1 [
6
|ur ↑ ug ↑ ub ↑⟩ + |ug ↑ ub ↑ ur ↑⟩ + |ub ↑ ur ↑ ug ↑⟩
−|ur ↑ ub ↑ ug ↑⟩ − |ug ↑ ur ↑ ub ↑⟩ − |ub ↑ ug ↑ ur ↑⟩ ]

where now the colour of each up quark is specified by the subscript and its spin is written
directly after. In this way the symmetry property of the total state is consistent with what
quantum mechanics requires for a fermion.

116
15.2 Colour and Gluons

Colour is the charge seen by the strong force, and the basic scattering process for the strong
force is the exchange of a gluon between quarks (and/or antiquarks) as illustrated in Fig-
ure 15.1.

Figure 15.1: The basic scattering process of the strong interaction.

A major difference between the strong and electromagnetic interactions is that unlike the
photon of electromagnetism, which has zero electric charge itself, gluons carry net colour
and change a quark’s colour during the strong interaction. More precisely, gluons carry one
unit of colour charge and one unit of colour anticharge. Since colour must be conserved at
each vertex, one way of viewing this is to explicitly show the flow of colour as in the example
in Figure 15.2.

Figure 15.2: Example quark-gluon vertex including colour.

Gluon interactions inside particles continuously change the individual colours of the quarks
and antiquarks that they interact with, whilst the net colour of the hadron remains “white”.
An implication of the gluons carrying net colour themselves is that they self-interact, mak-
ing the strong force complicated than the electromagnetic one. In Figure 15.3, the top
two diagrams show three and four gluon coupling vertices, which have no analogue for the
electromagnetic interaction. The bottom diagram in the figure shows that a virtual gluon
can split temporarily into a quark-antiquark pair. This diagram does have an analogue in
electromagnetism.
Qualitatively at least, the fact that gluons self-interact give us an explanation as to why the

117
 Figure 15.3: Gluon self-interaction diagrams.

properties of the strong interaction, particularly with regard to range and to confinement
for quarks, are so different from those for electromagnetism.

15.3 How many Gluons?

In some way we can think of gluons as colour-anticolour combinations. The fundamental

colour and anticolour triplets can be visualised as in Figure 15.4.

Figure 15.4: The fundamental colour (left) and anti-colour (right) triplets.

Forming all coloured quark-antiquark pairs we get nine combinations.

118
 Figure 15.5: The gluon octet.

Eight of the combinations form an octet, as shown in Figure 15.5. These are the gluon states.

|rb⟩, |rg⟩, |bg⟩, |br⟩, |gr⟩, |gb⟩ Edges

1 1 ( )
√ (|rr⟩ − |gg⟩) , √ |rr⟩ + |gg⟩ − 2|bb⟩ Centre
2 6

The leftover combination is a colour singlet

1 ( )
√ |rr⟩ + |gg⟩ + |bb⟩
3
Nature does not appear to use the colour singlet state. This state would be the QCD
analogue of the photon in electromagnetism (which is an electric charge singlet, another way
of saying that it carries no electric charge). If it existed it would generate a long-range force
between colour charges in the same way as the photon generates a long-range force between
electric charges, in contradiction which what we observe.

119
Chapter 16

More on QCD

16.1 Evidence for Colour

It is reasonable to ask whether colour is just a convenient way of solving problems such as
the symmetry of baryon wavefunctions, or does it have direct observable consequences? The
following experimental observation is an example of one which gives us direct evidence for
the existence of colour.
Consider the relative interaction rates for the scattering of electrons and positrons to either
a quark-antiquark pair, which ultimately produced hadrons in the final state, or to a pair of
muons, as illustrated in Figure 16.1, where the lowest-order Feynman diagrams are drawn.
The index i in the left-hand diagram represents any flavour of quark for which there is enough
energy in the collision to produce a quark-antiquark pair of that flavour. In particular we
will look at the ratio of the rates for these processes
σ(e+ e− → hadrons)
R=
σ(e+ e− → µ+ µ− )

The interactions are electromagnetic, with the respective vertices on the left-hand side of
each diagram being identical. The vertices on the right-hand side each have a coupling
constant associated with them which is proportional to the magnitude of the electric charge
of the fermions connected to the vertex. If the electric charge is measured in units of e, the
charge on a proton or positron, we can write the following for the coupling constants g and
rates for the two interactions:

gi ∝ Qi e (Quarks) gµ ∝ e (M uons)
∑ ∑
RQ ∝ gi2 ∝ Q2i e2 Rµ ∝ gµ2 ∝ e2

Note that for scattering to quarks, each flavour of quark that it is possible to produce a
quark-antiquark pair for contributes a distinct final state. Therefore to get the total rate to

120
 e+ q̄ e+ µ+

Qi i = u, d, ...
γ γ

e− q e− µ−

Figure 16.1: Electron-positron scattering to a quark-antiquark pair (left) and to an muon

pair (right).

hadrons we must calculate the rate for each separately and then sum the rates. If you are
unsure of why you should review the material in Chapter 5. We therefore get the following
prediction for the ratio of rates R:
∑ 2 2 ∑
RQ Qi e
R= = 2
= Q2i
Rµ e

What therefore do we expect with and without colour? Suppose we consider just the three
lightest quark flavours (uds). In the case of no colour, we have
∑ ( )2 ( )2 ( )2
2 1 1 2
R= 2
Qi = + − + − =
3 3 3 3
With colour, each quark flavour comes in three distinct colours, which translates to three
times as many possible final states. We therefore have
∑ ∑ 2
Qi 2
R= colour
2
=3× =2
e 3
Repeating this now for the cases where four or five quark flavours can participate, we get
the comparison given in Table 16.1.
The actual value of R as a function of energy in the e+ e− interaction obtained experimentally
is shown in Figure 16.2.
There is a lot of structure in this graph, which arises from the fact that at particular energies
corresponding to the rest mass energy of particles a resonance structure due to the production
of that particle is observed. Averaging these out and just looking at the trends evident in
R, given in pink, we can see steps in R which come into effect when the energy threshold
required to produce a pair of the next heaviest quarks is crossed. As indicated in the Figure,
if there was no colour quantum number then for any chosen energy R would be expected to
be a factor of three lower. The observed values of R are clearly consistent with there being
colour and inconsistent with no colour.

121
 Number of quarks R
Colour No Colour
2
3 uds 2 3
10 10
4 udsc 3 9
11 11
5 udscb 3 9

Table 16.1: Predictions for the ratio rates of quark-pair to muon-pair production in electron-
positron scattering, for different numbers of participating quark flavours.

Figure 16.2: The ratio of the rate of electron-proton scattering to quark-antiquark

√ pairs to
that for muon pairs, as a function of the centre-of-mass energy of the collision, s. Image:
Particle Data Group.

16.2 The Coupling Strength of the Strong Interaction

We are familiar with the electromagnetic and gravitational forces falling off with distance.
Classically both follow an inverse-square law. The strong force clearly behaves somewhat
differently. It effectively has a finite range, with the range being of order of the typical size
of a nucleus. It also appears to confine quarks and gluons inside of nucleons.
In fact the strength of the strong interaction, which can be expressed in terms of αS , the
strong interaction equivalent of the fine structure constant which we first met in Chapters
6 and 7, is found to be quite a strong function of the energy scale of the strong interaction
taking place. Figure 16.3 shows measurements of αS over a range of energy scales and made
with a variety of processes. The most striking thing to notice from this figure is that as the
energy scale Q tends towards infinity, αS becomes small. Conversely, as Q becomes small,

122
Figure 16.3: The variation of the strong interaction strength αS with energy scale Q. From
https://arxiv.org/pdf/hep-ex/0606035.pdf.

αS becomes large.
What does this mean? Remembering that there is an inverse relationship between the energy
or momentum of an scattering process and the time scale or distance scale which it probes
(this follows from the Heisenberg Uncertainty Principle), the behaviour of αS suggests that
the force between quarks is small when they are close together and large when they are far
apart. This phenomenon is given the technical name asymptotic freedom.
When the formal theory of QCD is considered, it is found that the strong coupling constant
varies with energy in the following way
12π
αs (Q2 ) =
(33 − 2Nf ) ln (Q2 /Λ2 )

where Q is the energy scale of the interaction Nf is the number of relevant quark flavours,
and Λ ≈ 0.2 GeV is a reference energy scale which has to be determined from experiment. It
can be seen therefore that once Λ is specified, the above equation for αS allows us to relate
the strength of the strong interaction at any other energy to that at the reference scale.

16.3 Quark and Gluon Jets

16.3.1 Separating Quarks

Given the behaviour of the strong coupling strength discussed above, it follows that it takes
an increasing amount of energy to separate quarks to greater differences. This is illustrated

123
Figure 16.4: The shape of the QCD potential energy function with separation between colour
charges.

in Figure 16.4, which compares the shape of the classical electromagnetic potential as a
function of separation, which leads to the Coulomb force law, to an analogous potential
for QCD which captures the properties seen for quark interactions. In the latter case the
potential is confining, and at larger distances the force between quarks, which is given by
the derivative of the potential, becomes constant rather than falling away.
One could qualitatively think about this in terms of lines of force, as shown schematically in
Figure 16.5. In the case of QCD, the colour lines of force are confined to the region between
the colour charges, reflecting the short range of the force, rather than giving the familiar
Coulomb pattern of lines of force between electric charges.
Figure 16.6 qualitatively shows what happens when quarks are separated. It becomes en-
ergetically favourable to create new quarks rather than extend the lines of force to larger
distances. This process can continue with the newly created particles, resulting in jets of
hadrons.

16.3.2 Jets

The formation of QCD jets is something that is confirmed by experiment. In an e+ e−

collider, for example, it is common to observe interactions in which two jets of particles are

124
Figure 16.5: Schematic comparison of the lines of force between two charges for the electro-
magnetic and the strong interactions.

Figure 16.6: Schematic view of QCD field lines between two colour charges as they separate.

produced, through the process which was illustrated in the left-hand side of Figure 16.1,
e+ + e− → q + q̄, with one jet from the quark and one from the antiquark. In the same way
that one cannot have an isolated quark, one cannot have an isolated gluon. If for example a
quark radiates a gluon, as in Figure 16.7, then the gluon will also ultimately produce a jet
of hadrons. The emission of a gluon can therefore split a single quark jet into multiple jets.
It was the observation of so-called 3-jet events in the late 1970s at DESY in Hamburg which
provided direct evidence for the existence of gluons. Figure 16.8 shows an example of a 3-jet
event from the ALEPH detector at the CERN Large Electron Positron (LEP) collider, which
ran in the 1990s.

125
 Figure 16.7: Radiation of a gluon by a quark producing a gluon-initiated jet.

Figure 16.8: A 3-jet event observed in the ALEPH detector at the CERN Large Electron
Positron (LEP) collider.

16.4 Quark-Gluon Plasma

A consequence of asymptotic freedom is that at very high energies, or equivalently very

short distances, because the coupling strength of the strong interaction decreases, quarks and
gluons behave essentially as free objects and can become unbound. This would presumably
have been the case in the very early Universe when energy densities were very high. This
state of matter is known as a quark-gluon plasma (QGP).
Experimentally, heavy-ion collisions are studied in an attempt to create conditions associated
with a quark-gluon plasma. Schematically this is illustrated in Figure 16.9.
Heavy-ions colliding are able to produce a higher energy density than is possible with proton-
proton collisions, for a fleeting moment. This fragment of quark-gluon plasma rapidly cools
and condenses into hadrons, and from the properties of the hadrons observed it is possible
to study whether the QGP was formed. In the early Universe, as the Universe expanded and
cooled, a phase transition would have occured, from a QGP state into the phase observed
today where quarks and gluons are bound into hadrons.

126
 Figure 16.9: Schematic of a heavy-ion collision producing a quark-gluon plasma.

16.5 What really is a Proton?

The simplistic view of a proton as three quarks hanging out, two up and one down, each
possessing a spin and a colour, is illustrated in Figure 16.10.

Figure 16.10: A simplified view of a proton.

This really is a simplistic view, since once one takes into account that quarks can radiate
gluons, which can in turn form quark-antiquark pairs which can themselves annihilate to
gluons, it becomes clear that as well as the three valence quarks that we associate with
the proton, there are at any given time gluons and quark-antiquark pairs present. A more
realistic view of a proton might be more like that shown in Figure 16.11.
The quarks and antiquarks which exist as virtual pairs within the proton are known as sea
quarks and antiquarks. Of course one cannot tell for any individual quark whether it is a
valence quark or a sea quark. The idea of valence quarks is that they represent the net
quantum numbers of the proton, since those of the sea cancel out.
Protons and other hadrons are therefore dynamic places. One observational consequence of
this is that quarks and antiquarks only carry ∼ 50% of the total momentum of a proton, the
rest being carried by gluons. Another is that a significant fraction of the proton’s quarks are
s/s̄ quarks, illustrated schematically in Figure 16.12. Scattering experiments using electrons
or neutrinos as probes of the proton or neutron allow us to quantitatively explore these
issues.

127
 Figure 16.11: A more complicated view of a proton.

Figure 16.12: A view of the proton showing the formation of a virtual strange-antistrange
pair. http://www.sciencemag.org/cgi/content/full/290/5499/2083.

16.6 A Brief Aside: Group Theory

This is non-assessable material, for your interest.

For those of you who have studied continuous groups, in particular Lie groups, certain of
these groups are fundamental to the Standard Model.

16.6.1 Gluons and Group Theory

We will return briefly to the colour quantum numbers possessed by quarks and gluons. In
group theory language, the triplet of colours and triplet of anticolours form fundamental 3
and 3 representations of SU (3), and we combine them

3⊗3=1⊕8

The gauge bosons in a theory always come from the adjoint representation of the symmetry
group, in this case the 8 representation. If the colour singlet gluon existed in Nature, the
natural symmetry group would be U (3) not SU (3), with U (3) requiring 9 gauge bosons
rather than 8.
The conservation of colour by the strong interaction is in fact a result of an exact symmetry
of the equations governing the Standard Model under transformations of the SU (3) group.

128
This follows from Noether’s Theorem, which we discussed earlier.

16.6.2 The Quark Model and Group Theory

SU (3) in fact gets used elsewhere in particle physics, and we were implicitly using it in a
different context when in Chapter 13 we were discussing the multiplets of hadrons which
arise in the quark model. If we assume just three flavours of quarks, u, d and s, which were
the first uncovered historically, then each quark or hadron multiplet which we constructed
in the plane of strong hypercharge Y versus the third component of strong isospin I3 (or Iz
as it is sometimes labelled) are in fact representations of SU (3), this time of flavour rather
than colour. So this is a different application of the same underlying mathematics of groups
and their representations.
In the case of flavour SU (3), unlike colour SU (3), we say that the underlying symmetry is
broken, or approximate, or inexact. This is clear because it is possible experimentally to tell
the different hadrons within a multiplet apart, through properties such as electric charge, or
strangeness, or mass.
If we consider any row of a flavour SU (3) multiplet, then what we have are multiplets of
strong isospin. These strong isospin multiplets are in fact representations of the group SU (2),
which is a sub-group of the group SU (3). It logically follows that the SU (2) symmetry is only
approximate, since the different members of a strong isospin multiplet are distinguishable by
their electric charge. We can say that the electromagnetic interaction breaks the symmetry.
The u, d and s quarks form the fundamental representation 3 of flavour SU (3) and the
corresponding antiquarks the 3 representation. If we introduce extra flavours of quarks, e.g.
c, then we can generalise the situation to include hadrons containing charm quarks as well,
and the symmetry is SU (4) in that case. given the large disparity in mass between the c
quark and the lighter quarks, SU (4) is very broken, but it still serves to label the allowable
hadrons which can be constructed from four quark flavours.

16.6.3 The Weak Force and Group Theory

The SU (2) group also plays another role in particle physics, when it comes to the weak
interaction. Here, for example, the u and d quarks, or the electron neutrino and electron,
forms doublets (multiplets) of SU (2), where the two members of the doublet are distinguished
by possessing different values of a quantum number known as the third component of weak
isospin. Exactly how the symmetry of all this plays out is rather complicated, and involves
so-called electroweak theory (the coupling together of the weak and electromagnetic forces)
and the Higgs boson. The details are beyond the scope of these lectures.

129
Chapter 17

Particle Mixing, CP Symmetry and

CP Violation

17.1 Symmetry Violation: Communicating with Aliens

Figure 17.1: A Zygon from the BBC series Dr Who. Image from http://en.wikipedia.
org/.

We have seen with parity and the weak interaction our first example of a symmetry violation
associated with the fundamental interactions of the Standard Model. At this point let us
pose a somewhat fanciful question which we will take up below.
Suppose we receive a signal from an alien civilisation, perhaps such as that depicted in
Figure 17.1, and they would like to come and meet us. How could we be sure they are not
made of anti-matter? If they were then a first encounter such as that of Figure 17.2 would
have disastrous consequences.
Could we ask the aliens to undertake some physics experiments and send us the results,
from which we could answer the question of whether they are made of matter or antimatter?

130
If our contact with the aliens was neutral and unpolarized, we could not communicate the
charge of the electron to them, or our definition of right and left. What to do?

Figure 17.2: First contact. Image: Universal Pictures.

17.2 CP Symmetry?

Returning to the discussion of discrete symmetries in Chapter!12, recall that parity violation
by the weak interactions came as a fundamental shock to physicists. For reasons which
will become clearer when we discuss neutrinos later, Wu’s experiment also violates charge
conjugation symmetry. Applying both charge conjugation and a parity transformation at
the same time, however, seems to result in a physically acceptable process, which would
be observable in our universe. At least this is what was believed in the aftermath of Wu’s
experiment – while the weak force individually does not conserve C and P , it was thought
that the combined property (CP ) was conserved.

17.3 K 0 Mesons

The neutral kaons, K 0 and K̄ 0 are strange particles produced in strong interactions, for
example through these processes:
π− + p → Λ + K 0
π + + p → K + + K̄ 0 + p
They are a particle/antiparticle pair with a mass of around 0.498 GeV/c2 and quark content
K 0 (ds̄) ¯
K̄ 0 (ds)

These particles can mix via the weak force as illustrated by the Feynman diagrams in Fig-
ure 17.3.
K 0 ↔ K̄ 0
This means that a particle which is initially produced as a K 0 can behave as a K̄ 0 at some
later time having propagated some distance, and vice versa.
Mixing is allowed between these particles because we can see that the following conditions
are met:

131
 s̄ q̄ d¯

W W

d q s

s̄ W d¯

q q̄

d W s

Figure 17.3: Mixing of K 0 and K̄ 0 via the weak force.

• Electric charge is conserved (both have Q = 0).

• Baryon number is conserved (both have B = 0).

• The weak interaction allows for a change of strangeness, since it does not conserve
stangeness (|∆S| = 2 in this case.)

If we make a pure beam of K 0 particles, it soon becomes a mix of K 0 and K̄ 0 , as illustrated

in Figure 17.4.
The particles in the beam can decay into pions, with the two decay modes characterised by
different lifetimes, τS and τL .

K → 2π K → 3π

τS = 9 × 10−11 s τL = 5 × 10−8 s

S stands for short (lifetime) and L for long. Initially there are both 2π and 3π decays present
in the beam. But the 2π decays end rapidly, leaving only the 3π decays. Does this reflect
different behaviour of the K 0 and K̄ 0 ? i.e.
K 0 → 2π
K̄ 0 → 3π

132
Figure 17.4: The intensity of K 0 and K̄ 0 as a function of time in a beam which is initially
all K 0 . τS is the mean lifetime of the state KS , defined in the text.

Whilst this might sound plausible it turns out not to be the case. Figure 17.5 shows that
both K 0 and K̄ 0 can decay into two pions. As an exercise you might like to convince yourself
that both can also decay into three pions.
How then can we explain the difference in lifetimes? Or, stated another way, why doesn’t
everything quickly decay into 2π? The relative amounts of the decay mode in the beam at
any given time (2π or 3π) in fact depends on the relative amounts of K 0 and K̄ 0 present.
This suggests the beam of particles consists of a quantum superposition of K 0 and K̄ 0 , in
the following sense. We can define two superpositions
1 ( )
|KS ⟩ = √ |K 0 ⟩ + |K̄ 0 ⟩
2
1 ( 0 )
|KL ⟩ = √ |K ⟩ − |K̄ 0 ⟩
2
If CP is not violated, what are the CP eigenvalues of KS and KL ? Applying both charge
conjugation and parity operations,

Ĉ P̂ |K 0 ⟩ = |K̄ 0 ⟩
Ĉ P̂ |K̄ 0 ⟩ = |K 0 ⟩

This can be seen by first noting that applying P̂ only to either state returns an eigenvalue of
-1, since we saw in the previous chapter that mesons and antimesons have negative parity.

P̂ |K 0 ⟩ = −|K 0 ⟩
P̂ |K̄ 0 ⟩ = −|K̄ 0 ⟩

133
 ū

s̄
d
W

d d¯

ū

d¯
d
W

s d¯

Figure 17.5: Feynman diagrams for the decays K 0 → 2π (top) and K̄ 0 → 3π (bottom).

Since Ĉ changes a particle to its antiparticle then we cannot write an eigenvalue equation in
this case. However Ĉ will introduce some phase factor eiζ multiplying the final state when
applied to |K 0 ⟩ and e−iζ when applied to |K̄ 0 ⟩ (since that gets us back to |K 0 ⟩), i.e.

Ĉ|K 0 ⟩ = e+iζ |K̄ 0 ⟩

Ĉ|K̄ 0 ⟩ = e−iζ |K 0 ⟩

These phases are unobservable, meaning that we are free to choose ζ as we wish. The
common convention is to choose ζ = π, from which Ĉ also introduces a factor of -1 in both
cases and therefore the above result for Ĉ P̂ follows.
Now applying this to the superpositions

Ĉ P̂ |KS ⟩ = +|KS ⟩ (CP = +1)

Ĉ P̂ |KL ⟩ = −|KL ⟩ (CP = −1)

What CP eigenvalues do the 2π and 3π decays have? The 2π state is CP = +1, which
again seems reasonable from Figure 17.6. While the argument is a little more complex and

134
Figure 17.6: A neutral kaon decaying to two charged pions in its rest frame, showing the
effects of applying first a charge conjugation and then a parity transformation. The original
system is regained.

we will omit it, the 3π state is CP = −1. So we can identify

KS → 2π (CP = +1)
KL → 3π (CP = −1)

These two states KL and KS act like they are particles that decay via the weak interaction.
The way to think about this is that the states K 0 and K̄ 0 are eigenstates of the strong
interaction hamiltonian, they are the states which are produced in a strong interaction and
they would be stable states if the weak interaction did not exist. The states KS and KL are
eigenstates of the full hamiltonian including the weak interaction and so they are the states
which are relevant for the decays via the weak interaction. The two sets of states are related
to each other via a change of basis.

17.3.1 KS Regeneration

After many KS lifetimes, the beam will be purely KL , with only 3π decays being observed.
The beam can then interact with matter, with differing reactions for K 0 and K̄ 0 possible.
The target is composed entirely of particles, so in one case we have particles interacting
with particles and in the other antiparticles interacting with particles, so it is reasonable to
expect different scattering reactions to be possible. Because of this difference, the target can
regenerate the KS → 2π component of the beam, so that after the target, 2π decays will
again be observed.

17.4 CP Violation

In 1964, Christenson, Cronin, Fitch and Turlay studied KL decays by placing a detector a
sufficient distance from the source of a beam of neutral kaons that all of the decays to two

135
pions (from KS decay) should have died out, leaving only decays to three pions. To their
surprise, approximately one in a thousand of the decays were observed to be to two pions,
indicating that KL did indeed decay into 2π, albeit rarely. This indicates violation of CP
symmetry, since we have

KL (CP = −1) → 2π (CP = +1)

which should be forbidden if CP is perfectly conserved. A corollary of this is that we must

also have

KS (CP = +1) → 3π (CP = −1)

since all of the kaons in the beam eventually decay to something.

This implies the astonishing result that the weak interaction imprints a fundamental asym-
metry on nature, contrary to the belief which was held following the work of Lee, Yang,
Wu et al. that CP symmetry would be respected by the weak interaction even if C and P
separately were not. As it turns out, this CP violation appears to be related to the reason
there is any matter in the universe at all!

17.5 Talking to aliens revisited

We can now return to the question posed at the end of Chapter 12 as to whether it is possible
to ask an alien cilvilisation which wishes to make physical contact with us to undertake some
physics experiments and send us the results, from which we could answer the question of
whether they are made of matter or antimatter?
It turns out that while KL decays into 3π as we have indicated, the dominant decay mode
(with a branching fraction of 66%) is as follows:

KL → π ± + l∓ + ν̄l (νl )

where l is a lepton from the first two generations, e or µ. You might like as an aside to ask
yourself why the lepton cannot be a τ . While KL is electrically neutral, and has no spin and
would be expected to be spherically symmetric, it is found experimentally that
Rate(KL → e+ + νe + π − )
= 1.00648 ± 0.00035
Rate(KL → e− + ν̄e + π + )
i.e. there is a significant asymmetry between the rate of decay to positrons as opposed
to electrons, with the positrons being produced slightly more frequently. This is also a
consequence of CP violation. So, all we need to do is ask the aliens to set up the appropriate
K 0 /K̄ 0 experiment, and tell them that the dominant charged lepton emitted is what we call
positive, i.e. it has the same charge as the charged particles which the nuclei in our bodies
contain. If they report the opposite observation, that the charge of the dominant charged
lepton in their experiment has the opposite charge to the charged particles in the nuclei in

136
their bodies, then we know that it would be very unwise, on these grounds if not other, to
make physical contact with them!
You might also as an exercise like to think about how we would agree with the aliens as to
what we each mean by right and left.

17.6 CP Violation – a summary

Individually parity and charge conjugation symmetry are both strongly violated by the weak
interaction. Combined, however, the weak interaction still mildly violates the combined CP
symmetry.
It is nevertheless thought that C, P and T are absolutely conserved when applied together,
a transformation called CP T . The argument as to why this is is a very general one from
Quantum Field Theory, which we won’t give here but which you can find in textbooks on the
subject. It is interesting to note that as a consequence of this, CP violation reveals that the
weak interaction must have some time asymmetry, i.e. time reversal symmetry is violated!
The situation is summarised in Table 17.1.
Quantity Strong EM Weak Symmetry
Parity (P̂ ) Yes Yes No Reflection in Space
Charge Conjugation (Ĉ) Yes Yes No Particle-Antiparticle Inversion
Time Reversal (T̂ ) Yes Yes Almost Reverse direction of time
Ĉ P̂ Yes Yes Almost Both Ĉ & P̂
Ĉ P̂ T̂ Yes Yes Yes As name suggests

Table 17.1: Discrete symmmetries and their relationship to the interactions of the Standard
Model.

137
Chapter 18

Neutrinos

18.1 The Definition of Helicity

The helicity of a particle is a measure of the projection of the spin vector of a particle onto
its direction of motion. If we denote the helicity quantum number h, spin vector p and
momentum vector p, then
S·p
h=
|p|
Helicty is a signed quantity, and as illustrated in Figure 18.1, a particle which has nega-
tive helicity, with spin vector and momentum vector pointing into opposite hemispheres, is
generally refered to as being left-handed (LH). A particle for which the two vectors point
into the same hemisphere is refered to as right-handed (RH). Here the hemispheres are ob-
viously defined as being with respect to plane perpendicular to the direction of motion of
the particle.

Figure 18.1: Illustrating negative helicity (left-handed) particles and positive helic-
ity (right-handed) particles. http://hyperphysics.phyastr.gsu.edu/hbase/particles/
neutrino3.html

Helicity is quantized, since spin is, and remembering that S = |Sz |max , it will take values −S
to +S. Note, however, that helicity is not in general a Lorentz-invariant quantity. Applying

138
a Lorentz transformation to a particle will change the energy and the momentum vector
of the particle but not the spin vector. If the boost vector of the transformation is large
enough in magnitude, then the helicity of a particle of non-zero mass can change sign. Stated
another way, we can find a frame of reference in which we have overtaken the particle and in
which it therefore appears to us to be travelling in the opposite direction. This means that
helicity is not in general a good quantum number to label massive particles with.

18.2 Photons and Helicity

A significant caveat in the last statement made was that it applied to particles of non-zero
mass. Photons are bosons with a spin of 1ℏ, and they are massless, which means that they
have a velocity equal to the speed of light, c, in all Lorentz frames of reference. It is not
possible for an observer to change her/his frame of reference in order to overtake a photon
and observe it to be travelling in the opposite direction, since this would require travelling
at greater than the speed of light, in violation of the principles of special relativity.
As a result of this, real, massless photons possess only two possible helicities, rather than
the three that one might otherwise expect for a particle of spin 1. Their spin vector (or
more correctly the projection of the spin vector) will either be measured to be aligned with
the direction of motion (right-handed photon) or anti-aligned with the direction of motion
(left-handed photon).
Note as an aside that even the previous statement has a caveat, as we refered to real photons.
A virtual photon will not generally have zero mass, and so when calculating quantum me-
chanical amplitudes involving virtual photons, we must take into account the three helicity
states that such photons can possess.
In electromagnetism, we are aware that electromagnetic radiation can be polarized, which
in terms of photon helicity can be taken to mean that in general we are dealing with a
quantum superposition of left-handed and right-handed photon states. The electromagnetic
interaction does not have a bias towards either helicity state, so generally, photons can be
produced equally well in LH and RH states.

18.3 Neutrinos and Helicity

In 1958 Goldhaber et al scattered electrons from heavy nuclei and examined the spin of the
emitted photons. The reaction that they were examining was the following:

e− +152 Eu(J = 0) →152 Sm(J = 0) + νe + γ

This is a weak interaction since a neutrino is produced in the final state. The photon is
emitted subsequent to the weak process taking place, when the samarium nucleus, which is

139
produced in an excited state, relaxes to its ground state. In this experiment, using conser-
vation of angular momentum and measuring the spin properties of the photons, the helicity
of the neutrinos could be deduced. Amazingly, from this and other experiments involving
antineutrinos, it was realised that all neutrinos are left-handed and all anti-neutrinos are
right-handed. Neutrinos were believed to be massless at the time, like photons, but these
experiments showed that in terms of helicity neutrinos were fundamentally unlike photons,
and that this strict handedness for neutrinos and antineutrinos represents a a fundamental
asymmetry of the universe.
It would seem that neutrinos provide a way of allowing us to communicate right and left to
an alien civilisation. Closer to home, can we now gain a deeper understanding of the Wu
experiment, which showed that parity is violated by the weak interaction?
In the Wu experiment, which uses β-decay, not only is an electron emitted but also an
accommpanying electron antineutrino. The situation is illustrated in Figure 18.2. Remem-
bering that angular momentum is conserved, it can be seen that the cobalt nucleus has total
angular momentum J = 5 and the daughter nickel nucleus J = 4. Electrons and antineu-
trinos have J = 1/2. If the spin vectors of the two nuclei are upwards as in the figure,
and if the antineutrino must be right-handed, then the prefered direction of emission of the
antineutrino must be upwards, with the emitted electron downwards and in a left-handed
helicity state, in order for total angular momentum to be conserved. If the antineutrino was
found to be emitted strictly downwards then it would have to be left-handed, which is not
allowed. Thus it is the strict handed-ness of antineutrinos which produces the asymmetry
in emission directions of the electrons in β-decay. In β + -decay an electron neutrino and
positron is emitted, and an asymmetry also occurs due to the left-handed nature of the
neutrino.

Figure 18.2: The decay of a cobalt-60 nucleus in the Wu experiment, showing the particles
and their spin vectors before and after the decay.

It should be noted that conservation of angular momentum does not require the antineutrino
to be emitted strictly upwards in every decay. If the situation is analysed carefully using
quantum mechanical principles, it is found that the electrons have an angular distribution

140
with the intensity of the emitted electrons a function of the angle and of their velocity. If θ
is the angle of the electron with respect to the vertical in Figure 18.2 and v is the magnitude
of the electron’s velocity, then the intensity I is given by
1 v
I(θ, v) = (1 − cos θ)
2 c
Typically v ∼ c owing to the lightness of electrons. For all practical purposes the electron
and antineutrino are emitted back-to-back, since the cobalt and nickel nuclei are essentially
at rest.

18.4 C, P , CP and Neutrinos

We have seen in the previous section that massless neutrinos are left-handed. Denote this as
νL , Charge conjugation changes particle to anti-particle but leaves helicity unchanged, since
it affects neither the spin nor momentum vector.

νL → ν̄L

ν̄L however does not exist in Nature.

The parity transformation changes helicity, since the momentum vector reverses and spin
does not, but leaves particle type unchanged.

νL → νR

Again νR does not exist in Nature.

Applying both charge conjugation and parity transformations, however, results in

C P
νL → ν̄L → ν̄R

and ν̄R does exist in Nature. This implies that as far as neutrinos themselves are concerned,
the weak interaction exhibits CP symmetry.

18.5 Detecting Neutrinos

While we are discussing neutrinos, let us look at their detection. Why aren’t the neutrinos
emitted in, for example, β-decay detected? The answer is that neutrinos interact very weakly.
They are the only fermions which interact solely via the weak interaction.
Neutrinos can be detected through scattering from nuclei, e.g.

νe + n → p + e− νµ + n → p + µ−

141
where the electron or muon and the recoiling proton are detectable. Even though this
demonstrates that neutrinos can be detected, the mean free path, dp , for neutrinos in lead
is around (the exact value is a function of neutrino energy)

dp ∼ 1.5 × 1016 m

To put this distance into context, note that the distance to the nearest star to the Sun is
4.1 × 1016 m.

18.5.1 Example of Neutrino Detection - Super-Kamiokande

Given such a small probability of interaction for an individual neutrino, in order to observe
neutrino interactions a large flux of them is required to start with. As well as having a large
flux of neutrinos, detection is also aided by having a large target mass.
Figure 18.3 shows the Super-Kamiokande detector in Japan.

Figure 18.3: The Super-Kamiokande detector.Image: Kamioka Observatory, ICRR (Institute

for Cosmic Ray Research), the University of Tokyo

It consists of 50 × 103 tonnes of pure water with some 11200 photo-multiplier tubes sur-
rounding it. It primarily detects neutrinos coming from the Sun, or produced in cosmic ray
interactions in the Earth’s atmosphere. Despite its size, and relatively large fluxes of neu-
trinos from these sources, many billions per day, the detection rate is only a few neutrinos
per day.
The Super-Kamiokande detector is what is known as a water-Cherenkov detector. It relies
on the principle that a charged particle, in this case an electron or muon produced when the
neutrino scatters, if travelling faster than the speed of light in a medium emits Cherenkov
radiation. This radiation consists of photons - you may be familiar with the blue light

142
which is visible in the water-filled core of some nuclear reactors, which is due to Cherenkov
radiation. The light is emitted in a cone of angle
1
cos θ =
nβ
where n is the refractive index of the medium, water in this case, and β is the velocity of
the particle as a fraction of the speed of light. Figure 18.4 illustrates the situation.

Figure 18.4: Cherenkov radiation, showing how a wavefront is produced at a particular angle
forming a cone around the direction of travel of the charged particle. Image: Wikipedia -
Arpad Horvath.

In the case of Super-Kamiokande, the photons forming the cone are detected by the pho-
tomultiplier tubes and a characteristic ring pattern is seen when the intensity of light over
all phototubes is mapped out onto a two-dimensional image of the detector (the detector is
essentially cylindrical - imagine cutting a cylinder along its axis at a certain point on the
surface and then unfolding it to form a plane).
It is possible to distinguish electron-type from muon-type neutrinos in a detector such as
this, as can be seen in Figure 18.5. The massive µ− produces a clean neutrino signal, while
the much lighter e− crashes around (scatters more) in the water and and causes a cascade,
leading to a messy signal. The pattern also allows the direction of the charged particle, and
hence the neutrino, to be ascertained to some extent.

18.6 Sources of Neutrinos

There have basically been four sources of neutrinos which have been used for studying the
properties of this particle. They are accelerators, the Sun, the Earth’s atmosphere and
nuclear reactors. A new field, neutrino astronomy, is also beginning to emerge now. The
first four of these sources will be discussed in this section.

143
Figure 18.5: The Cherenkov signal for an electron neutrino interaction in Super-Kamiokande
(left) and a muon neutrino interaction (right) in Super-Kamiokande.

18.6.1 Accelerators

Neutrinos from accelerators consist mainly of νµ from π, K decay, for example through the
decays

π + → µ+ + νµ
K + → µ+ + νµ

A highly schematic diagram of how the neutrinos are produced is shown in Figure 18.6. The
parent pions and kaons are themselves usually produced by bombarding a target material
such as copper or tungsten with an energetic proton beam extracted from an accelerator.

Figure 18.6: Production of neutrinos using an accelerator.

The energies of neutrinos produced in accelerators are generally in the range of a few hundred
MeV up to a few hundred GeV. Because the neutrinos are produced in a two-body decay and
the parent particle is moving, it is not possible to get mono-energetic beams of neutrinos.
This is actually true of all sources of neutrinos that we will discuss.

144
In the future it is conceivable that we will see the development of “neutrino factories” using
muons stored in a circular accelerator ring which decay via, for example,

µ+ → e+ + νe + ν̄µ

In this way very intense beams of neutrinos could be produced. They will still not be mono-
energetic. If the muons are of sufficiently high energy it would be feasible to store them long
enough to make these factories work.

18.6.2 The Sun

The Sun is a middle-aged star, converting hydrogen into helium. The principal cycle is
p+p → d + e+ + νe
d+p → 3 He + γ
3
He + 3 He → 4 He + p + p
Overall the Sun’s burning cycle, whilst complicated, essentially results in

2e− + 4p → 4 He + 2νe + 26.7 MeV

About 3% of the energy emitted by the Sun is in the form of neutrinos. The neutrinos travel
through the Sun and out in a short time, whilst for comparison the photons take 105 to 106
yrs to do the same. The νe flux at the Earth’s surface should be ∼ 6 × 1010 /s/cm2 , based
on what is known about the Sun through the Standard Solar Model. The photon flux is
∼ 4×1021 /m2 /s, but, unlike the neutrino flux goes away at night for an earth-based detector.
Figure 18.7 shows the energy spectrum of neutrinos from the Sun. The spectrum is complex,
because a number of different reactions contribute. The grey banding on this figure distin-
guishes energy ranges which are relevant to some of the detectors which have been used to
study solar neutrinos.

18.6.3 The Earth’s Atmosphere

The source of the neutrinos produced in the Earth’s atmosphere is cosmic ray interactions,
which are mostly initiated by very fast protons or helium nuclei. The decay scheme via
which the neutrinos are born is depicted in Figure 18.8. The decays of negative hadrons also
contribute. Because of their penetrating power, the Earth is rather transparent to neutrinos,
so at any point on the surface neutrinos produced in the atmosphere are arriving from all
directions, including below.
The energies of the neutrinos are generally in the range of about 100 MeV to 2 GeV. At the
Earth’s surface, neutrinos would be expected to be observed in approximately the following
ratio

N (νµ + ν µ )/N (νe + ν e ) ∼ 2

145
Figure 18.7: The Sun’s neutrino energy spectrum Figure from McKeown, R.D. et al.
Phys.Rept. 394 (2004) 315-356. arXiv:hep-ph/0402025.

18.6.4 Nuclear Reactors

Reactors are an abundant source of ν e . It was at a reactor that neutrinos were first observed.
The ν e are a by-product of fission reactions, mainly from isotopes such as 235 U, 238 U, 239 Pu
and 241 Pu. The fission products undergo β-decay

(Z, A) → (Z + 1, A) + e− + ν e

The ν e produced are of low energy, up to about 8 MeV. Fluxes are of the order of 7×1020 ν e /s
from a typical power reactor of about 4 GW.

18.7 The Solar Neutrino Problem

Physicists have developed detailed models of the centre of the Sun. These predict a particular
flux of neutrinos. It has been found over a range of experiments and detection techniques
that only a fraction of this flux of electron neutrinos is observed at the Earth. This is
illustrated in Figure 18.9, and naturally raises the question as to whether the solar models
are wrong.
All of the reactions used to measure the fluxes of neutrinos involve electrons – the prediction
is for the number of νe reaching Earth. Given the low interaction cross-section for neutrinos,
it is very unlikely that they interacted with something on the way andwere scattered or
absorbed.

146
 Figure 18.8: Neutrino production in the earth’s atmosphere.

18.8 Neutrino Mixing (or oscillations)

In the Standard Model neutrinos are massless, but if they are massive, they can mix. Taking
a simplified example in which there are just two neutrino species rather than three, we can
write
( ) ( )( )
|νe ⟩ cos θ sin θ |ν1 ⟩
=
|νµ ⟩ − sin θ cos θ |ν2 ⟩

The states on the left-hand side are weak eigenstates; these states form a basis with which to
discuss the weak interactions of neutrinos, and coupling to the W/Z bosons. The right-hand
side states are mass eigenstates. These states are used when a measurement of mass is made
(i.e. they are states with a definite mass). Equivalently, these mass or energy eigenstates
form a relevant basis for discussing the propagation of neutrinos in vacuum. The above
matrix equation shows that the weak states are quantum superpositions of the mass states,
and vice versa.
There is an analogy here to the situation with quark mixing, which in that case is framed
in terms of the Cabibbo angle and more generally the CKM matrix.
Since the two mass eigenstates have a definite mass, for a fixed amount of momentum they
will have different energies and velocities.
√ m2i c4
Ei = (pc)2 + (mi c2 )2 ∼ pc + mi ≪ Ei
2pc
and
pc2
vi =
Ei

147
Figure 18.9: Comparison of the measured fluxes of neutrinos from the Sun in three different
types of experiments, compared to the predictions of the Standard Solar Model.

For two neutrinos with different masses, the wavefunctions will beat between the two states,
as illustrated in Figure 18.10. Remember that an amplitude in quantum mechanics is related
to probability. Here the probability of the neutrino being detected as a e, µ or τ depends
upon time.

Figure 18.10: Illustrating beating between two neutrino mass states of differing mass.

If the angle θ introduced above is non-zero we can have non-trivial mixing of the states, and
we find after following through the calculation that
( )
L
P (νe → νµ ) = sin (2θ) sin
2 2
1.27 (m22 − m21 )
E

where L is the distance travelled (in km) & E is the energy of the neutrino (in GeV). The
“mass difference squared” ∆m2 = m22 − m21 is in units of eV2 .

148
If we start with a pure source of νe (for example from the Sun), as they propagate towards
the Earth there is an oscillation
νe ↔ νµ
Since the above formula for the probability shows that the mixing depends on the mass
difference squared between the neutrinos and the mixing angle, if either of these is zero,
then there is no mixing and no oscillations.

18.9 Atmospheric Neutrino Mixing

The SuperKamiokande detector in Japan is able to detect neutrinos produced in the atmo-
sphere via, for example, the interaction
νℓ + nucleus → ℓ− + anything
and labels them as muon-like (ℓ = µ) or electron-like (ℓ = e) depending upon their Cherenkov
signature, as we saw in the previous chapter. Figure 18.11 gives a nice description of the
principles of atmospheric neutrino experiments.
What was originally measured by SuperKamiokande and reported around 1998 was the
following ratio
(Nµ /Ne )observed
R= = 0.65 ± 0.05
(Nµ /Ne )expected
which clearly indicated that there appeared to be too few muon neutrinos being detected
compared to expectation. Furthermore, because the neutrinos arrive at the detector from all
directions, because interactions occur in the atmosphere all around the Earth and the Earth
itself does little to impede the resulting neutrinos, the experiment could look at their data
as a function of the ratio of distance travelled to neutrino energy. Neutrinos which travel
to the detector from the atmosphere immediately above would typically travel some 13km,
whereas those that have travelled through the Earth from the atmosphere on the other side
will typically have travelled some 13000km. The data is shown in Figure 18.12, and confirm
that the observed deficit is in the muon neutrinos and is greater the further that they have
travelled.
Upon detailed analysis, the favoured interpretation of the SuperKamiokande data was that
νµ → ντ oscillations were occurring with sin2 2θ ∼ 1 and ∆m2 ∼ 2−3×10−3 eV 2 . The dashed
lines in Figure 18.12 show the expectation for νµ - ντ oscillations with ∆m2 = 2.2 × 10−3
eV2 and sin2 2θ = 1, which describes the data extremely well.

18.10 Solar Neutrino Oscillations

Neutrino oscillations also provide a solution to the solar neutrino problem. While originally
created as νe , on their journey to the Earth, the neutrinos mix into a combination of νe , νµ

149
 Figure 18.11: Atmospheric neutrino experiment principles.

and ντ . The detectors which were observing a deficit in the number of neutrinos arriving
at the Earth, including SuperKamiokande which also detects solar neutrinos, were doing so
because they can only detect νe neutrinos if source is solar. This is because of the energy
spectrum of the solar neutrinos. Since they are detected through scattering to produce a
charged lepton, solar neutrinos to not have sufficient energy to create a muon or a tau lepton,
which means that νµ or ντ cannot be detected in this case.
The issue was finally resolved by the Sudbury Neutrino Observatory (SNO) in Canada with
the detector shown in Figure 18.13.
SNO can detect all three flavours of neutrinos by using heavy water rather than normal
water, and “neutral current” interactions involving virtual Z bosons rather than virtual W
bosons. The relevant interactions are
νℓ + d → νℓ + p + n
νℓ + e− → νℓ + e−

where ℓ = e, µ and the electron is an atomic electron in the second case. Since the neutrino

150
Figure 18.12: SuperKamiokande data on atmospheric neutrinos. See text for an explanation
of the dashed lines. From Phys. Rev. Lett. 81 (1998) 1562.

remains a neutrino in this case there is no threshold effect due to having to produce the
heavier charged lepton in the final state. SNO can in addition observe the “charged current”
interaction

νe + d → e− + p + p

analogous to the charged current interactions which the other solar neutrino experiments
detect.
The SNO results, which are included in Figure 18.14, show that the total number of ν arriving
is as expected from the solar models, while confirming that the number of νe detected is low
as in the other experiments.
The subject of neutrino oscillations and neutrino mass is a huge one and the field very active
these days, and represents a potential portal for exploring physics beyond the Standard
Model. We have only scratched the surface in what we have discussed here.

151
Figure 18.13: The SNO detector. Image from A. McDonaldNobel Lecture: The Sudbury
Neutrino Observatory: Observation of Flavor Change for Solar Neutrinos. Nobelprize.org.
Nobel Media AB 2015. Web. 7 Dec 2015.

Figure 18.14: Solar neutrino results including those from SNO.

152
Chapter 19

Beyond the Standard Model

The Standard Model of particle physics encapsulates our present understanding the fun-
damental fermions (quarks and leptons) and their interactions via the exchange of gauge
bosons. It provides a description of the electromagnetic and weak forces, unified into the
electroweak force, and of the strong force using QCD. Gravity is not included in the Standard
Model.
The Standard Model of particle physics is extremely successful in its ability to incorporate
the unified electro-weak force and the strong force through QCD. But it cannot be the
whole story. For example it contains many free parameters (see below), and only massless
neutrinos, at least as originally formulated. The strong and electroweak interactions are not
unified, and it does not contain gravity.
The quest to unify the electroweak and strong interactions has lead to the idea of Grand
Unified Theories (GUTs) and the idea of gravity being included also to the term “Theory
of Everything” (ToE). It is thought that extensions to the Standard Model may also solve
problems in other areas of physics, such as providing an explanation for dark matter in
astrophysics.

19.1 Lots of Parameters

One dissatisfying aspect of the Standard Model is the number of free parameters that it
contains. By free parameters we mean quantities which we have to measure the values of,
but for which the theory itself makes no prediction of what these values should be. Examples
that we have met in these lectures are:

• the masses of the six quarks, and the three charged leptons

• the mass of the Higgs boson

• the coupling strengths of the electromagnetic, weak and strong interactions

153
 • the Cabibbo angle

If all of the free parameters are tabulated, including the (tiny) masses of the neutrinos, then
the number is greater than twenty. Ideally, a deeper theory than the Standard Model would
be able to predict the values of these parameters, or at least predict relationships between
them and reduce the number of independent one. Perhaps a Theory of Everything might
ultimately contain, say, one free parameter, the size of a universal coupling strength.

19.2 The Precision of Standard Model Predictions

An example of the extremely precise nature of some Standard Model predictions is the g-
factor anomaly for leptons. The lepton g-factor gℓ relates the magnetic moment ⃗µ of a lepton
to its spin ⃗s, through the relation
eℏ
⃗µℓ = gℓ ⃗s
2mc
In non-relativistic quantum mechanics, as described by the Schrödinger equation, we have

gℓ = 1

A relativistic quantum-mechanical treatment, as delivered by Dirac and his Dirac equation,

gives

gℓ = 2

When the full quantum field theory treatment is given, which includes higher order correc-
tions in the calculation as illustrated in Figure 19.1, then one obtains

gℓ = 2 (1 + aℓ )

which the fractional difference of the value for gℓ from 2 is denoted aℓ and is related to the
idea of an anomalous g-factor.

Figure 19.1: Higher order corrections contribute to the calculation of g-factors.

154
19.2.1 The g-factor anomaly for the electron

The g-factor anomaly for the electron is one of the most precisely known quantities in physics.
The measured and predicted values for ae are as follows:

ae = (1159652180.73 ± 0.28) × 10−12 experimental

ae = (1159652181.78 ± 0.77) × 10−12 theoretical

Experiment and theory agrees to around 10 significant figures, a major vindication of the
theory of Quantum Electrodynamics (QED) and perturbative methods for calculating with
the theory. By way of contrast, it can be noted that the Newtonian gravitational constant
G is known to a relative uncertainty of ∼ 10−4 .

19.2.2 The g-factor anomaly for the muon

When the same comparison is made for the g-factor anomaly for the muon, aµ is measured
and predicted as:

aµ = (1165920.9 ± 0.6) × 10−9 experimental

aµ = (1165918.0 ± 0.5) × 10−9 theoretical

In this case the values, whilst still extremely precisely determined, actually disagree by
around 3.0 standard deviations. This is one of the rare occasions where the Standard Model
prediction for a physical quantity may be in disagreement with experiment.
Could this be a hint of physics beyond the Standard Model? If there are new particles
associated with a deeper theory for which the Standard Model is an approximation, then
these new particles can contribute to the calculation of predictions, for example by appearing
in virtual loops, as illustrated in Figure 19.2. In this example the new particles are associated
with Supersymmetric theories, to be touched upon below. When there effects are included
it may lead to better agreement between prediction and measurement.

Figure 19.2: New particles from beyond-Standard-Model theories may contribute to the
calculation of quantities such as the g-factor for the muon.

155
19.3 Beyond The Standard Model: Unification

As we alluded to in an earlier chapter, in terms of coupling strength, at high energy the

strong force weakens. In fact it turns out that at high energies the electromagnetic and weak
forces strengthen. This is illustrated schematically in Figure 19.3.

Figure 19.3: Schematic of the unification of the coupling strengths for the electromagnetic,
weak and strong interactions at high energy. Image: CERN Courier

This leads to the idea is that at somevery high energy (∼ 1015 − 1016 GeV) all of these forces
could be observed to have the same strength. From the energy regime where we can directly
study the change in coupling strength with energy, there is a large extrapolation required
over many orders of magnitude to predict whether the coupling strengths do in fact meet
at a single high energy scale. With just the Standard Model governing the running of the
coupling strengths, the extrapolation does not quite give converge to a single point in the
way it does in Figure 19.3. Rather than abandon the idea of unification which has served
physics for so long, most physicists take this failure of the extrapolation of the coupling
strengths to a single point as a hint that physics beyond the Standard Model is needed.

19.4 Dark Matter

As an example of a phenomenon which clearly seems to indicate the need for physics beyond
the Standard Model, we will consider dark matter.

156
19.4.1 Mass/Energy Content of Universe

Figure 19.4 shows our current understanding of the Universe’s energy budget, based on
observation.

Figure 19.4: The energy budget of the Universe. http://map.gsfc.nasa.gov/universe/

The amount of energy that is baryonic in nature, that is, bound up in the atoms that form
the visible matter in the Universe, is in fact less than 5% of the total energy budget of the
Universe. At least five times as much energy is in the form of an as yet unidentified dark
matter, which is non-baryonic but does not seem to correspond to any known particle in the
Standard Model. Even more mysterious is the fact that over 70% of the energy budget seems
to be in the form of Dark Energy, required to explain the accelerating rate of expansion of
the Universe discovered around 15 years ago by studying Type Ia supernovae.

19.4.2 Evidence for Dark Matter

The original evidence for dark matter came from measuring the rotation curves of galaxies,
which indicated that much of the matter in the Universe is “invisible”. Figure 19.5 shows
curves for a typical galaxy. Curve A is the predicted distribution of velocities of stars as
a function of distance from the galactic centre, based on Newton’s Law of Gravitation,
assuming that all of the mass of the galaxy exists in the form of stars and other luminous
matter. Curve B is what is actually observed.
The observed rotation curves can only be explained if a large amount of matter in the
galaxy is non-baryonic very weakly interacting with normal matter, its presence only inferred
through its gravitational effects. The distribution of the dark matter would be something
like that shown schematically in Figure 19.6, forming a spherical distribution extending
significantly beyond the visible dimensions of the galaxy.

157
Figure 19.5: Typical rotation curves for galaxies. A is the expected distribution in the
absence of dark matter and B is the observed distribution.

Figure 19.6: Schematic for the distribution of visible matter (white) and dark matter (grey)
in a typical spiral galaxy. http://www.universeadventure.org/

19.5 What is the Dark Matter?

One obvious candidate for the dark matter is that it is ordinary matter that is very difficult
to observe, because it is in the form of dust, extinct stars, dead planets and the like. This
can be referred to as cold matter, the cold indicating that the matter is non-relativistic.
Neutron stars, black holes, white dwarfs and the like also fall into this category of cold
matter, generally going by the name of MACHOs, or Massive Compact Halo Objects. The
idea that this ordinary matter forms the dark matter is generally disfavoured by gravitational
lensing experiments - there simply do not seem to be enough baryons.
Another candidate for dark matter is neutrinos, which would constitute a form of hot dark
matter with the hot referring to the fact that neutrinos are relativistic. Neutrinos are the
second most abundant Standard Model species in the Universe. However, this idea is also
disfavoured. Neutrinos are too light to produce sufficient total mass, and too relativistic to
be confined to galactic halos.

158
 Volume 260, number 3,4 PHYSICS LETTERS B
The most promising candidates for dark matter are WIMPs, or Weakly Interacting Massive
Particles. These hypothetical particles are characterised by exactly the properties 2. Definition needed
of the coupling constants
:i ~ ~ World average 87
by dark matter - weakly enough
50 interacting
"~- C~?'(~) to explain
(U ~t the lack
M eR D~6 1116~) of observation, and massive AmMdi

enough to explain the gravitational effects that they produce. This impliesIn athemass unified
in SU
the( 2 ) L @ U ( 1 ) theory,
2 ing well known relations hold between t
range 100-200 GeV/c , electric charge neutral, and stable, that is, left over from Big Bang.
30 below. constants and the gauge boson masses:
We will say more on WIMPs ~o'" ld ~ ld ~ t
20 45 ~ " q] e=v/4na=gsin 0w = g ' cos 0w,
. . . .

o M w = ~vg" ,

19.6 Unification and the Standard Model Mz = ~ v ~ 2+g2,

103 10 5 10 7 109 ld 1 ld 3 ld 5 ld 7
/~ [GeV] from which it follows that

e2 g,2 1 M2w
'~- 50
sin20w- g2 - g,2 + g2 - M 2 •

Here g and g' are the coupling constants o

S U ( 2 ) L and U ( 1 ), respectively, c~ is the
30 I ~'i ~ ' ~ ' d 5 ld4 ld5 10'6
L ~ L ..... I~ ~ ~ ~ 1~ ~~ ture constant, 0,~ is the electroweak mixin
20 45 ~ v is the v a c u u m expectation value o f the
If the model contains Higgs representation
I0 40
doublets, the theory has an additional de
10 3 105 10 7 109 ld ~ ld 3 ld 5 ld 7
dom, usually p a r a m e t r i z e d by the p-para
/4 [GeVl In the SM based on the group SU ( 3 ) c ®
U ( 1 ) the usual definitions o f the couplin
Fig. 1. (a) First order evolution of the three coupling constants
are
Figure 19.7: Extrapolation
in theofminimal
the coupling strengths
standard model for the
(world average Standard
values in 1987 Model forces to high
energy. From: Ugo Amaldi,from ref,
Wim[ 11 ).de
TheBoer,
small figure
andis aHermann
blow-up of theFürstenau,
crossing area). “Comparison
5 ' / 2 / 4 ~of- - grand
50~/3C0S20M~ ,
(b) As above but using Mz and oq(Mz) from DELPH 1data. The
unified theories with electroweak
three coupling and strong
constants coupling
disagree constants
with a single measuredOL2at
unification point = g 2LEP”, Phys. ,
/ 4 n = a/sin~-O~s
Lett. B 260, 447–455 (1991).
by more than 7 standard deviations.
a3=g2/4g,
unified theories, was only two s t a n d a r d deviations in where gs is the SU ( 3 )c coupling constant
1987. 5
Above we stated that at high energy the strong force weakens, while the inelectromagnetic 5 the definition o f a~ has been inclu
In this p a p e r we extend that analysis with recent p r o p e r n o r m a l i z a t i o n at the unification p
and weak forces strengthen. This gives rise to the idea of unification of the forces – that at
more precise LEP data. We do this along lines similar The coupling constants, if defined as e
very high energy all of these
to the forces have the
ones recently a d osame strength.
p t e d by Ellis et al. [2 ] and ues including loop corrections in the g
Langacker [3]. We use published d a t a from the propagators, become energy d e p e n d
Figure 19.7 illustrates the result of extrapolation of the Standard Model coupling strengths
D E L P H I Collaboration, o f which we are members, n i n g " ) . A running coupling constant r
to high energies. This extrapolation
but similar results doesn’t
couldquite work from
be derived in thetheStandard
data o f Model. Over twenty
specification of a renormalization schem
years ago it was already other
clear LEP
thatexperiments.
the couplings just missed each other extrapolated will use the tousual highm o d i f i e d m i n i m a l
energies, and that the difference
The paperwas has statistically
been organized significant,
as follows: that
In sec-is, unlikely to go away
scheme ( M S ) [5]. The energy depende
with more data. tions 2 and 3 the coupling constants are defined and pletely d e t e r m i n e d by the particle conte
their new d e t e r m i n a t i o n s are described. In section 4 couplings inside the loop diagrams of th
the evolutions o f the coupling constants to high ener- sons, as expressed by the renormaliz
gies in the m i n i m a l s t a n d a r d model ( S M ) and in the equations, The first order r e n o r m a l i z
19.7 Supersymmetry
minimal supersymmetric standard model (SUSY) equations are
are compared. A s u m m a r y is given in section 5.
0
~t ~ - c q ( # ) = bic~2(#) + .... i = 1 , 2 , 3
Supersymmetry (SUSY) is a beyond-Standard Model symmetry which has several attractive ogt
features. It proposes that Standard Model fermions have boson superpartners
whereof~tspin
is the 0, and at which
energy the coupli
that Standard model bosons have fermion superpartners of spin 1/2. The naming scheme
448
for the superpartner particles is somewhat whimsical, as can be seen below.

159
 quark (1/2) ↔ squark (0) W (1) ↔ W ino (1/2)
electron (1/2) ↔ selectron (0) Z (1) ↔ Zino (1/2)
muon (1/2) ↔ smuon (0) P hoton (1) ↔ P hotino (1/2)
tauon (1/2) ↔ stauon (0) Gluon (1) ↔ Gluino (1/2)

Figure 19.8 gives the Standard Model particles and their SUSY counterparts. This figure is
actually a simplification when it comes to the Higgs particles. In the simplest SUSY model,
there are in fact 5 Higgs bosons, three neutral and two charged, of which one has the prop-
erties of the single Higgs boson of the Standard Model.

Figure 19.8: The Standard Model particles (left) and their superpartners (right).

The new particles can affect calculations of the rates for processes and other quantities, such
as the g-factor anomaly aµ for the muon, through additional loop diagrams which contribute
to the quantum mechanical amplitudes. An example of new diagrams is given in Figure 19.9.

Figure 19.9: SUSY diagrams contributing to the g-factor anomaly for muons.

160
19.8 Unification, Supersymmetry, and Dark Matter

Supersymmetric particles change the evolution of the couplings for the Standard Model
interactions, as shown in Figure 19.10. Now it is possible to have a scenarion in which the
curves meet at ∼ 1016 GeV.
Volume 260, n u m b e r 3,4 PHYSICS LETTERS B 16

60 SUSY 2 n d order ld6 ld7

-- without smeoring

£ ~10 + .... .... +.M ..

--., ___ ° :::
++°i
40

+ 103

20

0,1 0.105 0.11 0.115

10 3 10 5 10 7 10 9 ld I 1(~ 3 ]d 5 !017

# [OeV]
(.D .

Figure 19.10: Extrapolation of the coupling strengths for the Standard Model forces to
high energy, including SUSY effects. From: Ugo Amaldi, Wim de Boer, and Hermann
Fürstenau, “Comparison of grand unified theories with electroweak and strong coupling
5 5F J
constants measured at LEP”, Phys. Lett. B 260, 447–455 (1991).?
12 i 4+ i! o+ o~o+ o,1 o,,+

Additionally, it is expected that the lightest supersymmetric particle willFig.

O+o2 io3
knowing the masses of the SUSY particles
'° +
'° + l
'° 5
it is not clear exactly
:),\/,!, be3.stable.
which particle
The Msus¥Without

the errors inthis

(a) and Mou T (b) energy scales are
function o f ~ 3(Mz). The uncertainties in MOUT and Ms
c q ( Mwill
z ) andbe,c%(Mz) are small. The ful
Msus,r [OeV] Mou+ [GeV]
but quite likely it will be a mixture of the photino, zino etc, known generically as a Neutralino. sumes that all SUSY particles have the mass of the SU
Fig. 2. (a) Second order evolution of the three coupling con- The dashed, dotted and dash-dotted lines indicate the
Neutralinos should interact via
stants theminimal
in the weakSUSY interaction.
model. MsusYThis then
has been fittedmake
by re- the the
neutralino
SUSY particle a spectrum
naturalis smeared over the range
candidate for dark matter (i.e.crossing
quiring a WIMP).
of the couplings in a single point. The two lower in the figure.
plots show t h e z z distribution for the SUSY scale MsusY (b) and
for the unification scale MGo x (c) taking into account their fluence on M~vT, as shown by the dashed and
correlation.
lines in figs. 3a and 3b.
19.9 DetectingTheWIMPs Terrestrially
widths of the Z 2 distributions are dominated
The values of MGVT and MsvsY are correla
taking this correlation into account, one finds
by the error of c~3(Mz). We have repeated the fits for
The solar system orbits different values of
the galactic c% (Mz)
centre at and the results
a speed are shown
of approximately 250 km s−1 . GeV
M s l J s y = 10 3"0+10
The,
in figs. 3a and 3b. One observes that−1 Msusv is a steep
Earth orbits the Sun atfunction
a speedof of around 30 km s . The WIMPs, ~ifGU which T = only GeV,
interact
1016"0+03
cq: for c%(Mz) between 0.10 and 0.12,
weakly with other matter, Msvsvdo varies
not movebetweenwith30 the
TeVvisible
and 10matter.
GeV. TheEffectively, therefore,
o~6~jT =25.7_+ 1.7. the
Earth moves through the 68% WIMPs,
CL rangeor of from the obtained
c~3 values, viewpoint of a stationary Earth there is a
by averaging
Because of the threshold behaviour, the mas
flux of neutralinos. The DELPHI
estimated results
flux(see
of eq. (26)), is also
neutralinos ∼ 109 /m2 /s.
is indicated.
Until now the assumption was made that the slopes heavy gauge bosons (Mx) is typically 0.3 MGU
If the proton decay is dominated by X-bos
In order to directly detectchange
the WIMPs,
from SM the valuesmethod
to SUSY employed would
values exactly at involve elastic scattering
Msvsv. This abrupt change is unphysical, not only change, the proton lifetime for Mx = 3 ! i 0 ~5 G
of the WIMPs from nuclei. Seasonal variations in a detected signal would help distinguish
be estimated as
because the particles are virtual, but also because dif-
the signal from the background, which mostly originates from radioactivity.
ferent SUSY particles are likely to have different
masses. To model the actual behavior we have 1 M4 1033.2+1.2 yr,
Z'P r°t°n ~ c~tjv2 M~ -
smeared this change over 1-3 orders of magnitude
symmetrically around Msusv by taking the average of where Mp is the proton mass. However, the es
the SM and SUSY slopes in this interval. This smear- of eq. (31) is not unique because in many
ing lowers the fitted value of Msusv and has little in- models faster decay processes can contribute

452
161
19.9.1 Example: The CDMS experiment

An example of a direct detection experiment is the CDMS (Cryogenic Dark Matter Search)
experiment in the USA, deep underground at the Soudan mine in Minnesota.
The principle of CDMS is to detect the energy imparted to a nucleus by a WIMP, which is
typically around a few keV. Germanium and silicon crystal detectors are held at cryogenic
temperatures of around 50 mK. What is actually detected are the vibrations of the crystal
lattice, which correspond to phonons. To reduce backgrounds the experiment is placed far
underground.

Figure 19.11: The principle of WIMP detection with the CDMS experiment. http://cdms.
berkeley.edu/.

19.9.2 A dark matter experiment in Australia

An underground laboratory for dark matter research is now being developed in Australia, in
a former gold mine in Stawell, Victoria. For a popular article on this, see:
http://www.scienceinpublic.com.au/aip/aip-congress-stawell
See also https://www.darkmatter.org.au/ for some additional information.

19.9.3 Example: Detecting WIMPs with Accelerators

Another approach to detecting WIMPs is to produce them directly. The LHC experiments
are searching for neutralinos produced in proton-proton collisions at high energy.

162
Figure 19.12: A schematic of a collision producing a WIMP in the ATLAS detector at CERN,
showing the subsequent decay chain. http://www.pd.infn.it/~dorigo/met_atlas.jpg

Figure 19.12 shows a schematic of the production of a SUSY particle, in this case a gluino,
and the subsequent decay chain, which ultimately results in a neutralino (WIMP candidate).
In the Figure, χ̃01 is the lightest neutralino, which is stable. χ̃02 is the next lightest neutralino,
g̃ is a gluino and ℓ̃R is a slepton. Since the sparticles in a collision such as this would have to
be produced in pairs, a second decay chain in which a second neutralino is produced would
be expected also.
In Figure 19.13 a simulation of a collision producing such a pair of neutralinos in the ATLAS
detector is shown. The two lightest neutralinos can be seen exiting the detector at top left
and bottom left, whilst most of the visible activity in the detector is on the other side of the
detector, as it would need to be to balance momentum transverse to the beam axis, which
is into the figure. The dashed arrow labelled MET gives the direction of the Missing Energy
Transverse vector, which results from the addition of the transverse momentum vectors of
the two neutralinos.

Figure 19.13: A simulated event in the ATLAS detector in which a pair of neutralinos is
produced. http://www.pd.infn.it/~dorigo/met_atlas.jpg

163
At this point in time there is no evidence for the production of supersymmetric particles in
the LHC experiments, despite an intensive search using the data collected in 2011 and 2012
at 7 TeV and 8 TeV centre-of-mass collision energy (so-called Run 1 of the LHC), and from
2015 to 2018 at 13 TeV (Run 2). The data is still being analysed, and more sophisticated
methods of analysing it continue to be developed. When the LHC resumes operations in
2021, the energy is likely to push up to 14 TeV, which is the design collision energy, and the
intensity of the beams will be increased. From the mid-2020s, the “high-luminosity LHC”
(HL-LHC) will operate, collecting a much larger set of data than the current one. So the
search for a deeper theory is not over yet.

164
Errata and additions

Date Comment
None currently.

165