Professional Documents
Culture Documents
Lecture Notes On Particle Physics PH3520 - 2012/2013: Glen D. Cowan Updated by V. Boisvert Physics Department
Lecture Notes On Particle Physics PH3520 - 2012/2013: Glen D. Cowan Updated by V. Boisvert Physics Department
Preface
The main goal of PH3520 is to become acquainted with the basic ingredients and ideas of elementary particle physics. Well see that all matter can be described as consisting of a relatively small number of dierent types of particles, which interact in very specic ways. This description is now well enough established that its called the Standard Model. Well spend most of the course talking about the Standard Model and how it was developed. This will lead into discussions of theory and experiment and the interplay between the two. We begin in Chapters 1 and 2 with an overview of the Standard Model and its theoretical framework. Chapters 3 and 4 then provide a basic description of experimental techniques, including an overview of particle accelerators and detectors. In Chapters 5 through 8, we take a step back and go through the discoveries that led to the various components of our current picture. We will describe these developments using the language of todays Standard Model, which was in most cases not possible at the time of the discoveries themselves. In Chapters 9 and 10 we continue our description of the electroweak and strong interaction with emphasis on recent experiments that have provided verication of the Standard Model to a high level of precision. Chapter 11 describes one of the predicted and possibly very recently observed pieces of the Standard Model, the Higgs boson. The course notes for PH3520 are intended to contain essentially all of the material from the lectures. For further information, the book best suited for this course is Particle Physics by Martin and Shaw [1]. The books by Hughes [2] and Perkins [3] are also good but are somewhat more advanced. The history of particle physics from the standpoint of experiment is covered in the book by Cahn and Goldhaber [4]. It would be appreciated if corrections, suggestions and comments on these notes could be communicated to [email protected]. VB September, 2012
ii
Contents
Preface 1 Overview of particle physics and basic concepts 1.1 1.2 1.3 1.4 1.5 1.6 Particles of Nature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The fundamental particles of the Standard Model . . . . . . . . . . . . . . . . . . Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Antiparticles, baryon number, lepton number . . . . . . . . . . . . . . . . . . . . Open questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Some basic concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6.1 1.6.2 1.6.3 Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Special relativity and invariant mass . . . . . . . . . . . . . . . . . . . . . Observables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i 1 1 1 3 4 5 7 7 7 11 15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 16 17 17 19 20 21 22 23 25
2 Theoretical framework 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 Quantum eld theory
Feynman diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electromagnetic interactions, coupling strength . . . . . . . . . . . . . . . . . . . Weak interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strong interactions, hadrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Triple gauge boson couplings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Virtual particles, propagators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Space and time location of vertices . . . . . . . . . . . . . . . . . . . . . . . . . . Perturbation theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
iii
iv 3 Particle accelerators 3.1 3.2 3.3 3.4 3.5 3.6 Why high energies? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Accelerator principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Linear accelerators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Circular accelerators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fixed target vs. colliding beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . Some existing and planned accelerators . . . . . . . . . . . . . . . . . . . . . . . 27 27 28 28 30 33 35 37 37 38 40 42 44 45 47 50 51 53 53 53 54 58 60 65 65 68 72 76 79 79 79 80
4 Particle detectors 4.1 4.2 The particles we can detect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ionization energy loss, tracking detectors . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 4.2.2 4.2.3 4.3 4.4 4.5 4.6 Cloud chambers and bubble chambers . . . . . . . . . . . . . . . . . . . . Multiwire proportional chambers . . . . . . . . . . . . . . . . . . . . . . . Solid state detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Scintillators and photomultipliers . . . . . . . . . . . . . . . . . . . . . . . . . . . Electromagnetic showers, calorimeters . . . . . . . . . . . . . . . . . . . . . . . . Hadron calorimeters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Multicomponent detector systems . . . . . . . . . . . . . . . . . . . . . . . . . . .
5 Leptons 5.1 5.2 5.3 5.4 5.5 The electron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The positron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The muon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The neutrino . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The tau lepton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6 Hadrons 6.1 6.2 6.3 6.4 Nuclear forces, the pion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strange particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hadron resonances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The quark model of hadrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7 Inside hadrons 7.1 Elastic electronproton scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.1 7.1.2 The basic set-up, dierential cross sections . . . . . . . . . . . . . . . . . Predictions for d /d . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Lecture Notes on Particle Physics 7.1.3 7.2 Comparison with experiment the size of the proton . . . . . . . . . . . .
v 83 85 88 89 93 93 94 96
Deep inelastic electronnucleon scattering . . . . . . . . . . . . . . . . . . . . . . 7.2.1 7.2.2 Structure functions and scaling . . . . . . . . . . . . . . . . . . . . . . . . The parton model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8 Heavy quarks 8.1 8.2 8.3 8.4 8.5 8.6 8.7 The Cabibbo angle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The GIM hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The charmed quark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The bottom quark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 The CKM matrix and CP violation . . . . . . . . . . . . . . . . . . . . . . . . . 103 Lifetimes of charmed and bottom hadrons . . . . . . . . . . . . . . . . . . . . . . 106 The top quark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 111
9 The electroweak standard model 9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8
Theoretical need for the intermediate vector boson . . . . . . . . . . . . . . . . . 111 Gauge symmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 Non-conservation of parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Electroweak unication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Discovery of weak neutral currents . . . . . . . . . . . . . . . . . . . . . . . . . . 119 The discovery of the W and Z bosons . . . . . . . . . . . . . . . . . . . . . . . . 121
Z physics: SLC and LEP I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 e+ e ! W+ W : the LEP II programme . . . . . . . . . . . . . . . . . . . . . . . 128 133
10 Quantum chromodynamics
10.1 Evidence for colour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 10.2 The structure of QCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 10.3 The strong coupling constant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 10.4 Jets of hadrons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 10.5 Hadronization and the string model . . . . . . . . . . . . . . . . . . . . . . . . . 141 10.6 Gluon emission and event-shape variables . . . . . . . . . . . . . . . . . . . . . . 143 10.7 Measuring s using (e+ e ! hadrons) . . . . . . . . . . . . . . . . . . . . . . . 146
149
11.1 The Higgs eld . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 11.2 The Higgs mechanism and Margaret Thatcher . . . . . . . . . . . . . . . . . . . . 150 11.3 The theoretical need for the Higgs . . . . . . . . . . . . . . . . . . . . . . . . . . 151 11.4 Properties of the Higgs boson . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 11.5 Relating gauge boson masses and coupling strengths . . . . . . . . . . . . . . . . 153 11.6 Evidence for the Higgs from higher order eects . . . . . . . . . . . . . . . . . . . 154 11.7 Searches for the Higgs boson in e+ e collisions . . . . . . . . . . . . . . . . . . . . 155 11.8 Searching for the Higgs in protonproton collisions . . . . . . . . . . . . . . . . . 157 11.9 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 Bibliography 163
Chapter 1
1.1
Particles of Nature
According to our current understanding of Nature, all matter is made up of a small number of dierent types of particles. Remarkably, this basic idea repeats itself on dierent distance or energy scales. To describe processes where the energies involved are typically around 1 eV, the fundamental building blocks are atoms, of which there are about 100 dierent kinds. A typical atomic distance scale is around 10 10 m. If we collide particles together with higher energies, we nd out that atoms are not elementary but rather are made up of more fundamental objects, namely, negatively charged electrons surrounding a small positively charged nucleus. In processes involving energies around 1 MeV or more, one nds that the nucleus itself is not elementary but is made of neutrons and protons, with diameters of somewhat less than 10 15 m. In reactions at even higher energies we nd out that protons and neutrons are not elementary but are made out of more fundamental objects called quarks and gluons (collectively called partons). In our current picture, called the Standard Model, we have a relatively small number of fundamental particles whose interactions are governed by a set of abstract mathematical laws. It appears we can describe all processes that take place in the universe by a set of physical laws plus initial conditions. Our knowledge of the rules is not complete, however, and it is not clear, for example, whether the separation between physical laws and initial conditions will be part of a more nal theory. The fact that the Standard Model works as well as it does is nevertheless remarkable, and were therefore condent that were on the right track.
1.2
In the Standard Model, the fundamental particles of matter are spin-1/2 fermions, which interact by emitting and absorbing spin-1 particles called gauge bosons. The fermions come in two types:
quarks and leptons. There are six types or avours of quarks: up, down, strange, charm, bottom and top, usually referred to by their rst letters: u, d, s, c, b, t. The u, c and t quarks have an electric charge Q = 2/3, while the d, s and b quarks have Q = 1/3. Here we are measuring the charge Q in units of the protons charge, i.e., Q = q/e , (1.1)
where q is the usual electric charge and e is that of the proton. The quarks, their charges, and approximate masses (strictly speaking, their rest mass energies mc2 ) are listed in Table 1.1. Table 1.1: Quarks and their properties [6]. name (symbol) up (u) down (d) strange (s) charm (c) bottom (b) top (t) charge (Q) 2/3 1/3 1/3 2/3 1/3 2/3 mass (c2 ) 3 MeV 6 MeV 100 MeV 1.2 GeV 4.2 GeV 175 GeV
Free quarks have never been observed directly. They are always found bound together in groups of three, called baryons, or as a quark q and an antiquark q, called a meson. Examples of baryons are the proton, composed of uud, and the neutron, which is udd. Examples of mesons are pions, which can be charged ( + = ud, = du) or neutral ( 0 = mixture of uu and dd, K0 = ds). Mesons and baryons are collectively called hadrons. There are also six leptons, three with charge 1 and three which are neutral, called neutrinos. These are listed in Table 1.2. The simplest version of the Standard Model treats the neutrinos as having zero rest mass. Therefore Table 1.2 only lists for them experimental upper limits [6]. In fact, indirect evidence for very small neutrino masses has recently been discovered in a phenomenon called neutrino oscillations. For these lectures we will usually treat the neutrinos as being massless, as this will be an excellent approximation for the reactions considered. Table 1.2: Leptons and their properties. name (symbol) electron (e ) muon ( ) tau ( ) electron neutrino (e ) muon neutrino ( ) tau neutrino ( ) charge (Q) 1 1 1 0 0 0 mass (c2 ) 0.511 MeV 105.7 MeV 1.777 GeV < 2 eV < 0.19 MeV < 18.2 MeV
The fermions interact by exchanging spin-1 gauge bosons. The type of gauge boson exchanged determines the type of interaction: strong, electromagnetic or weak. In principle, the gravitational interaction should also be describable in terms of the exchange of a gauge
boson, the graviton, but gravity is extremely weak compared to the other interactions and we will neglect it in these lectures. At present there does not exist an accepted theory describing the gravitational interactions of particles at the quantum mechanical level, although a theoretical framework called string theory may eventually provide this. The electromagnetic interaction is mediated by the exchange of a virtual photon (denoted by ), and the weak interaction proceeds by exchange of a virtual Z (electrically neutral), or a W (charge +1 or 1). Both quarks and leptons participate in electromagnetic and weak interactions. In addition, quarks participate in the strong interactions, mediated by exchange of the gluon, whereas leptons do not. Some properties of the gauge bosons are listed in Table 1.3. Table 1.3: Gauge bosons and their properties. particle photon ( ) W Z gluon (g) charge (Q) 0 1 0 0 mass (c2 ) 0 80.4 GeV 91.2 GeV 0 interaction electromagnetic weak weak strong
The particles , W and Z can be seen as mediating a unied electroweak interaction. Those involving the W and Z are weak in the sense that the masses of these particles lead to a suppression of the interaction probability when the energy of the reaction is much less than the bosons mass. In addition to the fundamental fermions and gauge bosons, the simplest version of the Standard Model predicts the existence of a neutral spin-0 particle known as the Higgs boson. This particle is related to the mechanism by which all particles of the Standard Model acquire mass. The mass of the Higgs boson itself is not predicted by the Standard Model. In July 2012 ATLAS and CMS reported that they had observed a new particle consistent with the Standard Model Higgs boson. We will discuss this in Chapter 11.
1.3
Families
From Tables 1.1 and 1.2 we see that there are three quarks with charge 2/3, three with 1/3, three leptons with charge 1 and three types of neutrinos. We can group these particles into families, as shown in Table 1.4. Table 1.4: The three families of quarks and leptons. charge (Q) 2/3 1/3 0 1 1 u d e e family 2 c s 3 t b
The three families are given in order of increasing mass. The families also group the particles according to the reactions in which they participate with the highest probability: u and d quarks interact in a sense as a pair, as does c with s, and t with b. Electron neutrinos are produced in association with electrons, muon neutrinos with muons, tau neutrinos with taus.
1.4
For every particle there exists an antiparticle with the same mass but opposite charge. These are denoted either with a bar or with a superscript that indicates the opposite charge, i.e., the antiquarks are u, d, s, c, b, t , and the antileptons are e+ , + , + , e , , . The interactions of quarks and leptons are such that certain quantities or quantum numbers are conserved. The rst example you are no doubt familiar with is the electric charge. In addition, we dene the baryon number B to be 1/3 for the quarks, 1/3 for antiquarks, and zero for all other fundamental particles. The baryon number of a system of particles is simply the sum of the individual baryon numbers. So, for example, the baryon number of a proton (uud) or neutron (udd) is 1. In all reactions that have been observed, the total baryon number is conserved, i.e., the baryon number of the initial state particles equals that of the nal state. There are indirect hints that baryon number may not be perfectly conserved, and we will return to this point in Section 8.5. In a similar way, one assigns a lepton number L of 1 to the leptons, 1 to the antileptons, and 0 to all the other particles. The lepton number is also conserved in all reactions that have been observed, but the indirect hints mentioned above about baryon number violation hold also for lepton number. In addition to the total lepton number L, we can dene a separate lepton number for each family: Le = 1 for e and e , L = 1 for and , L = 1 for and , with 1 for the corresponding antiparticles, and otherwise they are zero. For any system of particles we have L = Le + L + L . (1.2)
In the simplest version of the Standard Model with massless neutrinos, these quantum numbers are conserved separately. The recent observation of neutrino oscillations, however, has indicated that this is not the case in Nature. For the reactions we will talk about in this course, however, Le , L and L will all be separately conserved. So, for example, in the beta decay of a neutron, n ! p e e , (1.3)
the initial state has B = 1, L = 0, and the nal state has the same. Note that Le , L and L are also all conserved. The reaction p ! 0 e+ , (1.4)
however, violates both baryon and lepton number conservation and has never been observed, despite extensive attempts to see it experimentally.
1.5
Open questions
To date there exists no signicant discrepancy between experimental observations and the predictions of the Standard Model. Nevertheless, it is widely believed that the Standard Model cannot be Natures nal answer to the structure of matter. Below we list some open questions and unsolved problems. Well add more as we go along. For a readable discussion of how close we might be to solving these problems, see the book by Weinberg [7]. The model itself contains around two dozen free parameters, whose values must be obtained by comparing experimentally measured quantities with the corresponding predictions and adjusting the parameters to give the best agreement. It is hard to believe that Nature at her most fundamental level would be described by so many arbitrary parameters. It is widely believed that the Standard Model represents a low energy eective theory, similar to the way that Newtons theory of gravity represents an eective approximation to General Relativity in the limit of weak, slowly varying gravitational elds. This more general particle physics theory, if and when we nd it, should explain why the parameters (at least some of them) have the values that they do. The masses of particles are poorly understood. They are all free parameters, and no one has yet succeeded in nding any understandable pattern in their values. In order for particles to have mass in the Standard Model, the Higgs boson should exist, and although a particle consistent with the Higgs has been observed by ATLAS and CMS as reported in July 2012, it has not yet been established if it is the Standard Model Higgs, or one of several Higgs particles. Gravity is not yet included. A nal Theory of Everything should incorporate all of the four known interactions: gravitational, electromagnetic, weak and strong. Already, electromagnetic and weak interactions are understood to be dierent aspects of a unied electroweak interaction. No one has succeeded, however, in describing electromagnetism and gravity as dierent manifestations of the same thing. This is a di cult task, especially when one considers that gravity is so weak. For example, the ratio of the forces of electrical repulsion to gravitational attraction between two electrons is more than 1042 . A promising candidate for such a unied description of all four interactions is called superstring theory, where the various elementary particles are described as dierent vibrational states of small loops of string. The loops have extremely small sizes, on the order of 10 33 cm, so in all experiments that we can carry out today they appear as pointlike particles. Currently we do not know whether superstring theory is correct or even which of this class of theories should be taken to describe our universe. An introduction to the basic ideas of superstrings can be found in the book by Greene [8].
Lecture Notes on Particle Physics Why are there three families of particles? Almost all phenomena of our usual world can be described with particles from the 1st family: nuclei are made of neutrons and protons, which are made of u and d quarks; the nuclei are surrounded by electrons; the nuclear reactions in stars involve electron neutrinos. Why Nature has given us three families of particles, the heavier two of which are almost completely decoupled from everyday life, is not understood. Why do the equations describing elementary particles possess the mathematical symmetries they do? For example, many processes we see are symmetric with respect to right and left (i.e., a mirror-image experiment would run the same way as the original) and with respect to matter and antimatter, but in the weak interaction these symmetries are violated. It is not clear why Nature should be this way. It is possible that the asymmetry between particles and antiparticles may be related to the presence of the 2nd and 3rd families, and this may have played an important role at times just after the Big Bang, when the temperature of the universe, i.e. the average energies of the particles, was much higher. This in turn could explain why the universe appears to be made almost exclusively of matter, rather than a mixture of matter and antimatter. Are our current elementary particles truly elementary or do they possess some sort of substructure, which we could resolve in higher energy interactions? One possibility for this is the superstring theory mentioned above. Other theories have been proposed in which quarks and leptons are composed of particles called preons. No experimental evidence for this type of compositeness has been seen. Based on the measurements of experiments like WMAP [9], we can estimate that the energy density of the Universe is approximately: 4% normal matter, 21% dark matter and 75% dark energy. The description of dark matter and dark energy is beyond the scope of this course, but we have good indirect experimental evidence for the existence of dark matter. We now need a direct observation of dark matter and a theoretical framework as to how this new particle ts into the current Standard Model.
1.6
In this section we introduce some basic tools and concepts that will be needed in the lectures to follow. We recall the system of units used in particle physics and review some important aspects of Special Relativity. Finally well discuss what it is that we usually measure in particle physics experiments, namely, cross sections and decay rates.
1.6.1
Units
The system of units used in particle physics is summarised in Table 1.5. Making the indicated replacements amounts to working in units where h = c = 1. That is, whenever we write p for momentum, this really stands for the usual momentum (having units GeV/c) times c, so that the units of our particle physics momentum are GeV. Table 1.5: Units used in particle physics. quantity energy E momentum p mass m distance x time t units GeV GeV GeV GeV 1 GeV 1 replacement pc ! p mc2 ! m x/( c) ! x h t/ ! t h
For converting between the two systems, its su cient to know c and c: h
To convert from particle physics units back to normal units, simply insert the factors of c and c necessary to obtain the desired dimension. For example, cross sections in standard units h having dimensions of length squared will have units GeV 2 in the particle physics system. To get back to normal units, i.e. fm2 , you have to multiply by ( c)2 . h In addition we will usually give the electric charge of a particle as a multiple of the protons charge e. In particle physics units, e is a dimensionless number with a value of 0.303. Its easier, however, to remember the ne structure constant , given by = e2 1 . 4 137 (1.5)
1.6.2
For almost all of the particle reactions we will look at, the speeds involved will be signicant fractions of the speed of light, and therefore we will need to use Special Relativity. The most
important relations we will need are the relativistic equations relating a particles energy E, momentum vector p = (px , py , pz ) and rest mass m (in the following, the term mass will always mean rest mass):
E = p =
m, m,
E 2 = p 2 + m2 .
p
2 , and p = |p| is the magnitude of the Here = v/c gives the speed of the particle, = 1/ 1 momentum vector. Note that in these relations we have set c = 1 as discussed in Section 1.6.1. A handy consequence of (1.6) and (1.7) is the relation between the energy, momentum and of a particle: = p/E.
Now consider the case where a particle of mass M , energy E and momentum vector P decays into n daughter particles with masses mi , energies Ei and momentum vectors pi , with i = 1, . . . , n. From the relations above we know that for the original particle, E2 = P 2 + M 2 . Furthermore, from conservation of energy and momentum we have
n X i=1 n X i=1
(1.9)
E = P =
Ei , pi .
(1.10) (1.11)
Ei
!2
n X i=1
pi
!2
(1.12)
Equation (1.12) denes the invariant mass (squared) of a system of particles, regardless of whether they originated as the decay products of a single more massive particle.1 Notice that the second term on the right-hand-side of (1.12) says to construct the vector sum of the momenta and then to square it, i.e. to take the scalar product of it with itself. Recall that the energy and momentum of a particle together form a four-vector, which we can write as p = (E, p) , (1.13)
1 We often speak of invariant mass but then write down the equation for the mass squared, since its a bit easier to work with. You must determine from context which is actually meant.
where p = (px , py , pz ) is the usual momentum vector and the index takes on the values 0, 1, 2, 3. Recall also that the square of a four-vector means the energy component squared minus the sum of the squares of the momentum components. That is, the invariant mass squared of a particle (or system of particles) is simply the square (in the four-vector sense) of the total four-momentum. In particle physics there is a widespread but unfortunate notational convention that you must be aware of. If we dene p to be a four-vector, then sometimes we write p2 to mean E 2 |p|2 , i.e. the four-momentum squared. At other times we might dene p to be a usual three-vector, and we write p2 to mean |p|2 . So if you see something like p2 written in an equation, make sure you understand which possibility is meant. An important property of the invariant mass is that, as the name implies, it is a Lorentz invariant. That is, if we were to observe the system of particles in a dierent reference frame moving at a speed releative to the original one, then the energies and momenta would be dierent, but the invariant mass would be the same. We can see this by using a Lorentz transformation to obtain the energy E and momentum P in the boosted reference frame,
E0 =
0 Pk =
Pk E + Pk (1.14)
0 P? = P? .
Here Pk and P? are the components of P parallel and perpendicular to the boost and the quantities in the boosted frame are written with primes. The invariant mass squared in the 0 boosted frame M 2 is therefore
02
= E
02
Pk2
2 Pk
2 P?
= ( E = E2 = M2 ,
Pk )2
2 P?
E + Pk )2
2 P?
(1.15)
i.e., it is the same as in the original reference frame. Suppose we consider a system of n particles for which the invariant mass squared is given by equation (1.12). If we go to a reference frame in which the total momentum is zero, then taking the square root of both sides of (1.12) gives M=
n X i=1
Ei = Ecm .
(1.16)
That is, the invariant mass is the total energy in the rest frame of the system of particles. This is usually called the centre-of-mass energy, written as Ecm .
10
One use of the invariant mass is that it allows us to tell whether a particular set of particles originated as the decay products of a more massive particle, since then the invariant mass will be equal to the rest mass of the original particle. Suppose we consider a high energy reaction such as proton-proton collisions. These collisions lead to events with large numbers of particles being produced. The Higgs particle could also be produced, and if this happens, then it should decay a small fraction of the time into two photons. In a high energy collision, however, many other photons could be produced in addition to those from the Higgs. This is illustrated schematically in Fig. 1.1. In principle we can measure the energy and direction of all of the photons in such an event.
Figure 1.1: Proton-proton collisions at high energy result in the production of many particles. How do we know if any of the pairs of photons came from a Higgs particle? We simply compute the invariant mass squared for all pairs of photons in the event. Using equation (1.12) for two photons i and j gives
m2 ij
= (Ei + Ej )2
2 2 = Ei + 2Ei Ej + Ej
= 2Ei Ej (1
cos ij ) ,
where we used the fact that for photons E = p (i.e. they have zero rest mass). By computing this for all possible pairings of the photons and entering them into a histogram, we might get after including many events something like Fig. 1.2. The clustering of entries around a single invariant mass corresponds to those photon pairs which did in fact come from Higgs decays. The rest of the entries are from pairs of photons that did not originate from the same particle, and hence their invariant mass is simply some random value. The plot is from a simulation of Higgs boson production in proton-proton collisions with a hypothetical Higgs mass of 130 GeV, and the peak around 130 GeV is clearly visible. So the invariant mass allows us to infer the existence of a new particle without seeing it directly. For this reason, invariant mass has played a role in many important particle discoveries, and we will use it repeatedly in these lectures.
11
8000
Number of events
7000
6000
5000
4000
110
120
130
140
(GeV)
Figure 1.2: A histogram of the invariant mass of pairs of photons showing evidence for a particle with a mass of 130 GeV (from [5]).
1.6.3
Observables
There are basically two types of quantities that get measured in particle physics experiments: cross sections and decay rates (or functions thereof). These quantities are the interface between experiment and theory. Basically they represent probabilities for a specied reaction to occur, and hence they can be predicted by a quantum mechanical theory like the Standard Model. One measures these probabilities by in eect giving the reaction the opportunity to take place many times and determining the fraction of times it actually does. In this section we describe two important types of reactions: particle scattering and particle decays. Cross sections An important type of particle physics experiment is called scattering, where two particles are collided together and one measures what comes out. In elastic scattering, the outgoing particles are the same as the incoming, e.g. e+ e ! e+ e . In inelastic scattering, new particles are produced, and the incoming ones may be destroyed, e.g. e+ e ! + . One of the initial state particles may be at rest (a xed-target experiment), or two beams of particles may be collided together. In any case, we want to measure how often a certain nal state is produced, i.e. the number of events. Consider a beam of particles of cross sectional area A, hitting a broad target of
12
thickness d. The beam ux J is the number of particles in the beam per unit area per unit time, so the number of incident particles per unit time is given by JA. The number of events N will be proportional to the number of times we let the particles collide, i.e., it is proportional to the number of incoming particles per unit time, JA, times the number of target particles they can hit. This is proportional to the thickness d times the number of target particles per unit volume n. The product of all of these factors is called the luminosity L, L = JAdn , (1.18)
which has units of cm 2 s 1 . The luminosity can depend on the time t, since, for example, the ux might vary in time. The total number of events produced is thus proportional to the integral of the luminosity over time. The proportionality constant is called the cross section, N=
Z
L dt .
(1.19)
The cross section is a measure of the probability for a given nal state to be produced. It depends only on the nature of the particle interaction, and is independent of the details of the experimental set-up, namely, beam ux, target density and target thickness. So theorists can compute cross sections without knowing anything about how intense the beams are in a given accelerator. The experimentalist then combines this with the integrated luminosity to get the prediction for the expected number of events of a certain type. In normal units the cross section has the dimension of area, which in particle physics units corresponds to GeV 2 . Usually one converts the prediction back to normal units before comparing with experiment. The unit of area used for cross sections is the barn, dened as 10 24 cm2 . Most cross sections of interest are signicantly smaller than 1 barn, and one often sees, for example, nanobarns (1 nb = 10 33 cm2 ), picobarns (1 pb = 10 36 cm2 ) and femtobarns (1 fb = 10 39 cm2 ). The formula for the luminosity is a bit more complicated when the target is itself another beam of particles, e.g. in an e+ e collider. In any case, equation (1.19) always gives the relation between the number of events, integrated luminosity and cross section. For example, at the PEP-II e+ e storage ring at the Stanford Linear Accelerator Center (SLAC) in California, the average luminosity is around L = 3 1033 cm 2 s 1 . One years nominal running gives around 107 s of useful time, i.e., given the practical constraints of running an accelerator and particle detector, one can take data a third of the time. The cross section for 0 the reaction e+ e ! B0 B is about 0.6 nb (the B0 is a meson consisting of db), so in one year, the number of events of this type will be
N (e+ e ! B0 B ) =
Lt
2
107 s
10
cm2 1 nb
33
(1.20)
13
Often the particles produced in high energy reactions are unstable. One of the quantities that we can both predict and measure is the mean lifetime, i.e., the average proper decay time. Remember that if the particle is moving, then because of time dilation its proper decay time is given by the observed decay time divided by . One of the more general predictions of quantum mechanics is that the proper decay times of unstable particles will follow an exponential probability distribution. That is, the probability of observing a proper decay time t in the interval [t, t + dt] is p(t) dt = 1 e
t/
dt .
(1.21)
The constant of proportionality 1/ comes from the requirement that the integral of p(t) from 0 to 1 be equal to unity. The parameter is in fact the mean lifetime, as can be seen by computing the expectation value of t:
Z 1
0 0
hti = =
t p(t) dt t e
t/
Z 1 1
dt (1.22)
= .
As an example, muons can be trapped in a detector and the time that it takes for them to decay can be measured. The time for each decay can be entered into a histogram as shown in Fig. 1.3. The curve shown along with the histogram has the exponential form of equation (1.21). The value of the mean lifetime has been adjusted (or tted) to give the best agreement between the curve and the histogram, resulting in = 2.2 s. The total decay rate is dened by = 1 , (1.23)
and it therefore carries the same information as the mean lifetime. If a particle X can decay into a number of dierent states i, then the partial decay rate i is dened as the total decay rate times the probability for the decay to result in i, which is also called the branching ratio B for the state i,
i
B(X ! i) .
(1.24)
The branching ratio is often expressed as a percent, and the sum of all of the branching ratios for a particle must be unity. Equivalently, the sum of the partial decay rates is equal to the total decay rate,
X
i i
(1.25)
14
Figure 1.3: A histogram of decay times for muons For the K0 meson, for example, the main decay modes and branching ratios are shown in S Table 1.6 [6]. Table 1.6: Principal decay modes and branching ratios of the K0 meson. S mode + 00 + branching ratio 68.43% 31.39% 0.0018
For some types of particles, the decay time is too short to be experimentally resolved. We can still measure the mean lifetime, however, by exploiting the uncertainty relation between mass (i.e. energy) and time, m 1 . (1.26)
(Recall we are working in units where = 1.) If a type of particle decays with a mean lifetime h , then its mass must be uncertain by an amount on the order of 1/ . Therefore if we look at the invariant mass distribution of the decay products of the particle, the peak will have a certain width. Some of this will stem from imperfect measurement resolution (this is in fact the case in Fig. 1.2), but for su ciently small lifetimes the so-called natural width will dominate. Thus by measuring the width of a peak in an invariant mass distributions one can indirectly determine the mean lifetime of a particle.