Todo list

spin and parity of odd-A nuclei . . . . . . . . . . . . . . . . . . . . . . . . . 25

mention more sophisticated techniques for nuclear structure . . . . . . . . . 25

Derive Breit-Wigner shape . . . . . . . . . . . . . . . . . . . . . . . . . . .

FT 28
A
DR

1
 DR
A FT

2
The Oxford Nuclear and Particle Physics Basics

A.J. Barr
A FT

December 7, 2016
DR
 DR
A FT

ii
Preface

This book has grown out of a third year undergraduate lecture course taught at
the University of Oxford. That course is designed to introduce all physics students
to the main concepts of both nuclear and particle physics, without recourse to
the mathematical framework developed at the advanced undergraduate or graduate
level. This book aims to provide a corresponding introductory text. It is intended to
be accessible, and to be useful to all physics undergraduates whether experimentally
or theoretically inclined.

Subatomic physics has a particularly close interplay between experimental results

and the theories which describe them. The book uses the examples of nuclear and
particle physics to show both how models – of various levels of sophistication – are
developed to describe physical phenomena, and how those models can themselves
FT

be tested in future experiments. Examples include the semi-empirical mass formula

for nuclei, the quark model of hadrons, the Standard Model of particle physics.
A
DR

Non-relativisitic quantum scattering theory is developed via the Lippmann-Schwinger

equation, from which the propagator and interaction terms arise naturally. Feynman
diagrams for relativistic scattering are introduced by analogy, and used in tree-level
calculations of example experimental observables.

Applications of nuclear physics are dealt with in a dedicated chapter, which outlines
the important reactions both for power generation in earth, and for stars like the
sun. The physics of the accelerators and detectors – which form the toolset of the
experimental physicist – are also given a dedicated chapter. We do not shy away
from areas of active research, such as in neutrino or Higgs boson physics, where
recent progress have being made.

Each chapter ends with with a summary of the main key results. Additional infor-
mation, and a list of futher reading can be found after each chapter summary. At
the end of the book we provide problems for each chapter. They are of varying
difficulty, and are intended both to test the concepts developed in that chapter, and
to provoke further thought and discussion.

By its nature, subatomic physics deals with particles which are very small and
very fast-moving, so we have to deal both with quantum mechanics and special
relativity. The book assumes its readers already to have some familiarity with
special relativity at the level of performing Lorentz transformations and calculating
Lorentz invariants. In terms of quantum mechanics, the book expects the reader to
have met the bra-ket notation of Dirac. They should also have encountered angular

iii
momentum in quantum mechanics, perturbation theory, and the Fermi Golden Rule
for calculating transition rates. However no group theory is assumed, nor is it
expected that students will have yet delved into the deeper waters of quantum field
theory.

A FT
DR

iv
Contents

1 Introduction 1

2 Nuclear physics 3
2.1 Structure of matter and energy scales . . . . . . . . . . . . . . . . 3
2.2 Binding Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2.1 Semi-empirical mass formula . . . . . . . . . . . . . . . . . 4
2.3 Spin and parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4 Decays and reactions . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4.1 Alpha Decays . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4.2 Beta decays . . . . . . . . . . . . . . . . . . . . . . . . . . 14
FT

2.4.3 Gamma decays . . . . . . . . . . . . . . . . . . . . . . . . 19

2.5 Shell Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
A
DR

2.5.1 Solution of the infinite spherical well . . . . . . . . . . . . . 21

2.5.2 Spin orbit interaction . . . . . . . . . . . . . . . . . . . . . 24
2.6 Nuclear Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.6.1 Cross sections . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.6.2 Resonances and the Breit-Wigner formula . . . . . . . . . . 26
2.6.3 Nuclear scattering and form factors . . . . . . . . . . . . . . 29
2.7 Key points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.A Natural units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.B Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.B.1 Decays and the Fermi Golden Rule . . . . . . . . . . . . . . 34
2.B.2 Density of states . . . . . . . . . . . . . . . . . . . . . . . . 34
2.B.3 Fermi G.R. example . . . . . . . . . . . . . . . . . . . . . . 35
2.B.4 Lifetimes and decays . . . . . . . . . . . . . . . . . . . . . 35
2.B.5 The flux factor . . . . . . . . . . . . . . . . . . . . . . . . 36
2.B.6 Luminosity . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

3 Nuclear applications 39

4 Modern nuclear 41

5 Hadrons 43

v
CONTENTS CONTENTS

6 Scattering theory 45

7 Feynman diagrams 47

8 The Standard Model 49

9 Accelerators and detectors 51

A Examples 53
A.1 Radioactivity and nuclear stability . . . . . . . . . . . . . . . . . . 54
A.2 Quarks and scattering . . . . . . . . . . . . . . . . . . . . . . . . . 66
A.3 Relativistic scattering . . . . . . . . . . . . . . . . . . . . . . . . . 84
A.4 The Standard Model . . . . . . . . . . . . . . . . . . . . . . . . . 91
A.5 Energy production, stars . . . . . . . . . . . . . . . . . . . . . . . . 99

A FT
DR

vi
Chapter 1

Introduction

FT
A
DR

1
 CHAPTER 1. INTRODUCTION

FT
A
DR

2
Chapter 2

Nuclear physics and decays

2.1 Structure of matter and energy scales

Subatomic physics deals with objects of the size of the atomic nucleus and smaller.
We cannot see subatomic particles directly, but we may obtain knowledge of their
structures by observing the effect of projectiles that are scattered from them. When
a particle of any kind is scattered from a target, the typical resolution achieved is
FT

order the de Broglie wavelength,

h
A

λ= (2.1)
DR

p
where h is Planck’s constant, and p is the momentum of the projectile. Therefore,
if we wish to resolve small distances, smaller than the size of the atom, we will
need to do so with high-momentum projectiles. Smaller objects also tend to have
larger binding energies holding them together, so require larger energies to excite
their internal components. Some typical sizes of objects are given below, together
with the momementum of the projectile required to resolve their size, and typical
binding energies in electron-volt (eV) units.
keV 103 eV
MeV 106 eV
Object Size p = λh Binding energy GeV 109 eV
TeV 1012 eV
Atom 10−10 m 10 keV/c ∼ eV
Nucleus ∼ 10−15 m 1 GeV/c ∼ MeV
Quark < 10−19 m > TeV/c > TeV

We can see that small objects also tend to have high binding energies, and hence
projectiles of large energy will be required in order to excite them or break them up.
The momenta are indicated in units of eV/c where c is the speed of light. These
units make it easy to compare the momentum of the projetile to its corresponding
energy E = pc for the case of a massless projectile such as a photon. The most
convenient unit for describing the size of nuclei is the femtometer 10−15 m.1 No
1 The unit of 10−15 m or femtometer is sometimes called the ‘fermi’ reflecting the many seminal

contributions of the Italian physicist Enrico Fermi to the field of nuclear physics.

3
 2.2. BINDING ENERGY CHAPTER 2. NUCLEAR PHYSICS

sub-structure has yet been found for quarks even when using very high energy (TeV)
projectile.

2.2 The Nuclear Periodic Table and Binding Energy

Nuclei are found to be made out of two constituents: protons and neutrons. We
label nuclei by their atomic number Z which is the number of protons they contain,
by their neutron number N , and by their mass number A = Z + N .

The symbol used to identify a nucleus is

A
Z XN

where X is the name of the chemical element. For example the Carbon-14 nucleus,
which contains 8 neutrons and 6 protons is denoted 146 C8 . Since the element’s
name specifies the number of electrons, and hence the atomic number Z, and since
A = N + Z, we can fully specify the nucleus by just the symbol for the chemical
and the mass number,
A
X e.g. 14 C.
Isotopes Same Z
Isotones Same N Most nuclei are spherical in shape. The nuclear radius r can be measured in scat-
tering experiments, and follows the general rule
FT

Isobars Same A

Notation for related nuclei

r = r0 A1/3
A

(2.2)
DR

where the constant r0 is the characteristic nuclear size and is about 1.2 × 10−15 m.
The fact that r is proportional to A1/3 indicates that the volume of the nucleus
V ∝ r3 is proportional to the mass number A. Each proton or neutron is therefore
making an equal contribution to the overall nuclear volume.

2.2.1 Binding energy and the semi-empirical mass formula

The mass m(A, Z) of the nucleus containing Z protons and A − Z neutrons should
be given by the mass of its consituents, less the mass associated with the binding
energy. The mass-energy is therefore

m(A, Z)c2 = Zmp c2 + (A − Z)mn c2 − B(A, Z), (2.3)

where mp ≈ 938.3 MeV/c2 and mn ≈ 939.6 MeV are the masses of the proton and
neutron respectively. In nuclear physics it is convenient to measure energies in units
of MeV and masses in units of MeV/c2 . Using these units makes it easy for us to
convert from mass to mass-energy and vice versa. By assuming such units, we can
omit the factors of c2 in what follows. 2

We can build up a functional form for the binding energy B(A, Z) by considering
the forces between the nuclear constituents. To find the full quantum mechanical
2 For more on ‘natural units’ see appendix 2.A.

4
CHAPTER 2. NUCLEAR PHYSICS 2.2. BINDING ENERGY

Atomic or nuclear masses?

In nature almost all nuclei exist within atoms, so measuring atomic masses
is generally easier than measuring nuclear masses. The mass of the nu-
cleus is, to an excellent approximation, equal to the mass of the atom after
subtraction of the mass Zme of the Z atomic electrons. There is a small
reduction of the atomic mass from the binding energy Batomic of the elec-
trons to the nucleus. This atomic binding energy is typical of order keV
or smaller, about a thousand times smaller than nuclear binding energies
B(A, Z), and so the atomic binding energy can safely be ignored in most
contexts, so

matom = mnucleus + Zme − Batomic

≈ mnucleus + Zme .

The mass of the electron, is about half an MeV, of the same order as the
nuclear minding energies, so certainly can’t be ignored. In examples it’s easy
to forget to check whether atomic or nuclear masses are being considered
– so take care, particularly in β decay calculations where electrons (and
positrons) are created and annihilated.

ground state for all of the protons and neutrons would be a very difficult problem.
However we can understand a great deal about nuclear behaviour by building up a
model of the mass which encapsulates its key features. This we will do over the rest
FT

of the section, building up towards the semi-empirical mass formula of equation

(2.5). The ‘semi-empirical’ means that the model is built partly partly by demanding
A

agreement with data, and partly from our understanding of the underlying theory.
DR

Firstly, we will need an attractive force in order to hold the nucleus together against
the mutual electrostatic replusion of its constituent protons. That force must be
very strong, since the Coulomb electrostatic repulsion between a pair of protons,
each of charge e and separated by distance d ≈ 1 fm, is

e2
F =
4π0 d2
≈ 230 N

which is macroscopic – comparable to the weight of a child.

What form should that nucleon-nucleon attractive force take? We can get clues
about the force by looking at the binding energy per nucleon B/A is shown for some
common nuclei, shown in Figure 2.1. For nuclei this binding energy is typically of
order 8 MeV per nucleon. It can be seen that the most stable nuclei are found
for values of A around 60. Different behaviours can be seen in different regions.
There is a broad flattish plateau for the central region 30 < A < 200 for which
B/A ≈ 8 MeV. For A below about 30 the binding energy per nucleon is smaller
than the plateau value and is spiky. There is a systematic drop in B/A for large A,
particularly for A > 200.

To obtain a value of B/A that is rather flat, we cannot permit strong attractions
between each of the constituent nucleons and every one of the others. If every
nucleon felt an attraction to each of the others, then the binding energy would be

5
2.2. BINDING ENERGY CHAPTER 2. NUCLEAR PHYSICS

Figure 2.1: Binding energy per nucleon (B/A) as a function of A for some common
nuclei. Data taken from [1]. Plot from [source].

expected to grow as approximately B ∝ A(A − 1) ∼ A2 , and hence B/A would be

approximately proportional to A. Such a linear growth in B/A is ruled out by the
data (Figure 2.1).
FT

To obtain the flat B/A found in nature we must assume that the strongly attractive
force acts only between nearest neighbour nucleons. In this way, each nucleon
A

binds to the same number of nearest neighbours, independently of the size of the
DR

nucleus, and hence the binding energy per nucleon is the same regardless of the
nuclear size,
B ≈ αA

where α is a constant with units of energy. The use of nearest-neighbour interactions

indicates that the force must either be short-range, or screened from long-range
interactions by the effects of the nucleons in between.

In modelling a nearest-neighbour force we ought to make a correction for the fact

that those nucleons on the surface have fewer neighbours. To correct for the reduced
number of binding opportunities on the surface we reduce the binding energy by an
amount proportional to the surface area, leading to the corrected formula
2
B ≈ αA − βA 3 . (2.4)

The new contribution is negative since it reduces the binding energy. We have made
use of the observation (2.2) that since the volume of the nucleus scales as r3 ∝ A,
the surface area scales as r2 ∝ A2/3 .

These two terms (2.4) in this first approximation to the binding energy are known
as the volume term and the surface term respectively. Together they form what is
known as the liquid drop model, since a similar result would be found for a drop of
fluid with nearest neighbour interactions and a surface tension parameterised by β.
The liquid drop model is consistent with the observation that each nucleon requires
the same volume of space, in agreement with equation (2.2).

6
CHAPTER 2. NUCLEAR PHYSICS 2.2. BINDING ENERGY

So far, so good. However there is nothing in this liquid drop model to prevent the
growth of arbitrarily large nuclei. Such large nuclei are not observed in nature, so
we must be missing something. The obvious candidate is the Coulomb repulsion,
which interacts over long distances, and so will tend to push larger nuclei apart.
This electrostatic repulsion between protons will reduce the binding energy by an
amount proportional to Z(Z − 1) ≈ Z 2 because every proton feels the repulsion
from all of the other protons (not just nearest neighbours). The binding energy will E
be reduced by the electrostatic binding energy which can be parameterised by ΕF
2
Z
 1 .
A3 P N

Here  is a another constant with dimensions of energy, which we will calculate a

value for in the examples. The Coulomb repulsion energy is inversely proportional to
1

gZ HEL gN HEL
the radius of the nucleus, and hence to A 3 , since the potential energy of a uniform
2
sphere of charge Q is proportional to Q /r.
The density of states g(E) for
protons and neutrons as a
Two further terms are required to give a good match between our model and the function of energy E.
data. Both of them are quantum mechanical in origin.

Firstly there is an asymmetry term. The origin of this term is as follows. Since
protons are identical fermions, the Pauli exclusion principal states that no two of 1
them may exist in the same state. Nor may any neutron occupy the same state as p(Ei ) =
e(Ei −µ)/kB T + 1
any other neutron. However it is possible for a proton and a neutron to exist in the The Fermi-Dirac function gives
same state since the two particles are not identical. The allowed states are therefore the probability p(Ei ) of filling a
distinct, and are separately filled for the protons compared to the neutrons.
FT

state with energy Ei for a

system at temperature T and
with chemical potential µ. kB
A

We can work out the size of the asymmetry effect by calculating the number of is the Boltzmann constant.
DR

states available. Neutrons and protons are both fermions, and so obey Fermi-Dirac
statistics. The temperatures we are interested in are small compared to the chemical
potential (kB T  µ). Under these circumstances the Fermi-Dirac distribution tends
towards a step function — all levels are filled up to some energy level, known as the
Fermi Energy F , with all states with energy above F left vacant. E
ΕF
At large mass number A the Coloumb repulsion term would tend to favour larger
N and smaller Z, since neutrons do not suffer from the Coulomb repulsion as
P N
protons do. However this energic advantage of neutrons over protons will be partially
cancelled out by the fact that the additional neutrons must (on average) be placed
in higher energy levels than additional protons, since all of the lower-energy neutron
states will already be filled. gZ HEL gN HEL
The density of states g(E) for
If we approximate the states available as being a continuum, which is not a bad protons and neutrons as a
approximation provided that both Z and N are larger than about 10, then the function of energy E.
1
density of available states is found to be proportional to E 2 . In the examples we
show that this leads to an energy equation of the form
(N − Z)2
γ .
A
This asymmetry term reduces the binding energy, doing so most when the differ-
ence between the number of protons and of neutrons is largest.

Finally there is a pairing term which accounts for the observation that nuclei with
either even numbers of protons (Z even) or with even numbers of neutrons (N

7
 2.2. BINDING ENERGY CHAPTER 2. NUCLEAR PHYSICS

Z Z
4 5 6
100 100
7.5 7.5
7.75
80 80
8 7
8
60 60
8.25
8.5
8.5
40 8.5 40 3
8
7.75
20 7.5 20
8.5
8.25
8 7.25
7.57
77 N N
50 100 150 50 100 150

(a) (a)

Figure 2.2: Diagram showing binding energies as a function of proton and neutron
number for (a) data [1] and (b) the Semi-Empirical Mass Formula.

even) tend to be more stable than those with odd numbers. The pairing term is
zero for odd-A nuclei. Even-A nuclei have two possibilities. If both Z and N are
even then the nucleus is more tightly bound and have an extra binding contribution,
so B is increased by δ. If both Z and N are odd then the nucleus is less tightly
bound and so B is decreased by δ.

Putting all five terms together we obtain a formula for the binding energy,
FT

2 (A − 2Z)2 Z2
B(A, Z) = αA − βA 3 − γ −  1 + δ(A, Z),
A

A A3
DR

having eliminated N in favour of A. Substituting this into the formula defining the
binding energy (2.3) we obtain the semi-empirical mass formula (SEMF)

2 (N − Z)2 Z2
M (A, Z) = Zmp + (A − Z)mn − αA + βA 3 + γ +  1 − δ(N, Z).
A A3
(2.5)

Other than for A < 30, where our approximations are less valid, the SEMF gives a
rather good description of the binding energies of the observed nuclei (Figure 2.2).
In particular the SEMF correctly predicts the shape of the curved valley of stability
in the Z, N plane within which the stable nuclei are found. The relative numbers
of protons and neutrons along this valley reflects a trade-off between the Coulomb
Sketch of the shape of the
valley of stability. and asymmetry terms. At low A the asymmetry term favours N = Z. At larger A
the Coulomb term starts to compete with the asymmetry term, reducing the ratio
of protons to neutrons.

The SEMF confirms the observation that the most energetically favourable nuclei
are found close to 56 Fe. The energetically favourable location of iron goes some
way to explaining why this it one of the most common elements inside the Earth.

It is energetically favourable for nuclei far from that valley to migrate towards the
valley by nuclear decay. The ways in which nuclei can migrate around this space
are explored in the following section.

8
CHAPTER 2. NUCLEAR PHYSICS 2.3. SPIN AND PARITY

Volume Surface Asymmetry Coulomb Pairing

α β γ  δ
√
15.835 18.33 23.2 0.71 11.2/ A

Figure 2.3: Typical values of the SEMF parameters (in MeV). From Bowler.

2.3 Spin and parity

2.4 Decays and reactions

A table of the nuclides can be found in Figure 2.4. The stable long-lived nuclides
lie along the valley of stability where the binding energy per nucleon is largest. The
valley lies along N ≈ Z for light nuclei but has N > Z for heavier nuclei. Nuclei far
from that valley, and very heavy nuclei, tend to be unstable against nuclear decay.

While unstable nuclei will decay spontaneously, other reactions can be initiated by
firing projectiles at a nucleus. Reactions are said to be elastic if the final state
contains the same set of particles e.g. the elastic scattering of a photon from a
nucleus, via an excited intermediate state:

X + γ −→ X ∗ −→ X + γ.
FT

Reactions are inelastic if there is a change in particle content during the reaction
A

e.g. radiative capture of a neutron

DR

A
X + n −→ A+1 X∗ −→ A+1 X + γ.

For all nuclear decays and reactions we define the Q value to be amount of energy
‘released’ by the decay,
X X
Q= Mi − Mf
Z
The first sum is over the masses of the initial particles in the decay (including Β+, EC
their binding energies), while the second sum is over the masses of the final-state
particles (including their binding energies). A positive Q value shows that a reaction
Α
is energetically favourable. Β-

2.4.1 Alpha Decays N

The changes in (Z, A) induced
by various decays.
Alpha decays occur when (usually heavy) nuclei eject an ‘α particle’, that is a helium
nucleus containing two protons and two neutrons. The decay is

A
ZX −→ A−4 4
Z−2 Y + 2 He.

The change of mass number of the heavy nucleus is ∆A = −4 and the change in
its atomic number is −2.

9
 DR
A FT
2.4. DECAYS AND REACTIONS

10
Figure 2.4: Table of the nuclides as a function of the number of neutrons (N , on the x-axis) and the number of protons (Z on the y-axis).
Darker colours represent longer-lived nuclides, which can be found in the ‘valley of stability’. The ‘Magic Numbers’ indicate particularly stable
CHAPTER 2. NUCLEAR PHYSICS

nuclei and are described in the shell model (see appendix 2.5). From [Link]
CHAPTER 2. NUCLEAR PHYSICS 2.4. DECAYS AND REACTIONS

Figure 2.5: Decay modes of the nuclei. From [1].

FT

The process of decay of heavy nuclei is often via sequential chains involving both
alpha and beta decays. Since ∆A = −4 for α decays and ∆A = 0 for β and γ
A

decays we see that for any nucleus starting with mass number A (and if it only
DR

decays via alpha, beta or gamma decay processes) all other nuclei in that chain
must have some other set of mass numbers A0 = A − 4m, where m is an integer
indicating the number of alpha decays that have occured.

The helium nuclei emitted by alpha decay are highly ionising, and so travel only a
small distance through matter – they can be stopped, for example, by a piece of
card.

Rate calculation for α decays

We can model the α decay as a process in which ‘proto α particles’ are pre-formed
inside the nucleus. Each is assumed to have a large number of collisions with the
edge of the nucleus, but a small probability on each collision of tunnelling through
the Coulomb barrier and escaping.

If the Q value of the decay is positive, then the decay is energetically favourable,
but it may still be suppressed by a large tunnelling factor. Let us try to model
the probability of tunnelling through the barrier. We will assume that the large
exponential in the quantum tunnelling factor will dominate the calculation of the
rate of decay, so we will neglect differences in the probability of formation of the
proto-alpha particle, and its rates of hitting the barrier.

11
 95 Americium (Am)

94 Plutonium (Pu) alpha decay

93 Neptunium (Np) 93 Neptunium (Np)
DR
beta decay
92 Uranium (U) 92 Uranium (U) A
91 Protactinium (Pa) 91 Protactinium (Pa) stable isotope
FT
90 Thorium (Th) 90 Thorium (Th) thorium chain
89 Actinium (Ac) 89 Actinium (Ac) radium chain
88 Radium (Ra) 88 Radium (Ra) actinium chain
2.4. DECAYS AND REACTIONS

87 Francium (Fr) 87 Francium (Fr) neptunium chain

86 Radon (Rn) 86 Radon (Rn)

12
85 Astatine (At) 85 Astatine (At)

84 Polonium (Po) 84 Polonium (Po)

83 Bismuth (Bi) 83 Bismuth (Bi)

82 Lead (Pb) 82 Lead (Pb)

81 Thallium (Tl) 81 Thallium (Tl)

Element
241 240 239 238 237 236 235 234 233 232 231 230 229 228 227 226 225 224 223 222 221 220 219 218 217 216 215 214 213 212 211 210 209 208 207 206 205
Atomic mass

Figure 2.6: There are therefore four non-overlapping decay chains for the heavy elements, corresponding to A = n, A = n+1, A = n+2,
and A = n + 3 respectively, where n is an integer.
CHAPTER 2. NUCLEAR PHYSICS
CHAPTER 2. NUCLEAR PHYSICS 2.4. DECAYS AND REACTIONS

The time independent Schrödinger equation defines the energy eigenstate |Ψi,
 2 
p
E|Ψi = + V |Ψi.
2m
where E is the energy of the alpha particle, p is the momentum operator, m is its
mass, and V is the potential in which it moves.

For simplicity, we will ignore the spherical geometry and treat the problem as one-
dimensional in the radial direction r. In doing so we neglect a term l(l + 1)/(2mr2 )
in the effective potential, where l is the angular momentum of the outgoing alpha
particle. This term will in fact turn out to be an important correction to the potential
if conservation of angular momentum requires an emission with l > 0.

Returning to the 1-dimensional case, we assume that the heavy daughter nucleus
has negligible recoil energy, so that the outgoing alpha particle has kinetic energy
Q. The Schrödinger equation then becomes

1 ∂2
 
Qhr|Ψi = − + V (r) hr|Ψi, (2.6)
2m ∂r2
h
where we use natural units such that ~ = 2π = c = 1 (see appendix 2.A). In the
Dirac notation hr|Ψi represents the wave function — that is the amplitude to find
the alpha particle located between r and r + dr, and is often written Ψ(r). Without
losing any generality we can write the wave function as the exponential of some
other function η(r),
FT

hr|Ψi = exp[η(r)]. (2.7)

A

After inserting (2.7) into (2.6) and dividing by exp(η) we find

DR

1  00
η + (η 0 )2 + V (r),

Q=−
2m
where the primes indicate derivatives by r. Placing the nucleus at the origin, as-
suming the nucleus to be much heaver than the alpha particle, and working in its
rest frame, we can model the potential V (r) felt by any α particle by the function
(
const r < Ra
V (r) = zZαEM
r r > Ra

where inside the nucleus V is large and negative, and outside the nucleus it is given
by the Coulomb potential and hence characterised by the charges z and Z of the
α-particle and the daughter nucleus respectively. The constant αEM in the Coulomb
potential is the dimensionless electromagnetic fine structure constant

e2 1
αEM = ≈ .
4π0 ~c 137

Within the barrier the potential is smoothly varying, so η should be a smoothly

varying function of (r). We then expect η 00  (η 0 )2 , and we can safely neglect the
η 00 term compared to the (η 0 )2 . 3
3 This is known as the WKBJ approximation, named after Wentzel, Kramers, Brillouin and

Jeffreys, is a good approximation if many wave-lengths (or in the classically forbidden regions, as
here, many factors of 1/e) of the wave-function occur before the potential changes significantly.

13
 2.4. DECAYS AND REACTIONS CHAPTER 2. NUCLEAR PHYSICS

The tunnelling probability can be found from the ratio of the mod-squared ampli-
tudes:
|hRb |Ψi|2
P = = e−2G .
|hRa |Ψi|2
where the ‘Gamow factor’ G (> 0) is given by

√ Z Rb p
−G = η(Rb ) − η(Ra ) = − 2m dr V (r) − Q.
Ra

Choosing the minus sign before the radical ensures that we select the exponentially
The exponentially rising falling solution. The inner limit of the integration is the radius of the nucleus
solution would be of interest
1
only if the alpha particle were Ra ≈ r0 A 3 ,
tunnelling into the nucleus.
This alternative solution is the
relevant one in fusion reactions, and the outer limit
zZαEM
for example. Rb = ,
Q
the radius for which Q > V , i.e. where the α particle enters the classically allowed
region.

The Gamow factor integral will be dominated by that part where the potential V (r)
is much larger than the released energy Q, so to a good approximation one can
neglect Q in the integral. The rate dependence on Q is therefore dominated by the
(a)
10
9
Po
Pb
dependence of the upper integration limit Rb on Q, leading to the Geiger-Nutall
10
6
equation
FT

Rn
T1/2 [sec]

Z
log Γ = −c1 √ + c2 ,
3
10
Ra (2.8)
Q
A

0
10
Yb
DR

for the dependence of the alpha-decay rate on the charge Z of the daughter nucleus
186
-3 Po Hf
10 W
Hg Pt Os
and the energy release, where c1 and c2 are constants.
T 1/2 (exp)/T1/2 (GN)

-6
10 186
Po 186-208
10 g.s.->g.s. decays Po
190
188 Po
Po
1

(b)
0,1
0,34 0,36 0,38 0,40 0,42 0,44 0,46 2.4.2 Beta decays, electron capture
Qa-1/2

The logarithms of experimental There are three related nuclear decay processes which are all mediated by the weak
partial α-decay half-lives for the
even-even Yb-Ra nuclei with nuclear interaction. Neutron-rich isotopes can decay through the emission of an
neutron number N < 126 as a electron e and an anti-neutrino ν̄e in the beta decay a process:
−1/2
function of Qα (in
MeV −1/2 ). The straight lines A
ZX
A
−→ Z+1 Y + e− + ν̄e . (2.9)
show the description of the
Geiger-Nuttall law with A and
B values fitted for each isotopic
The effect is to increase the atomic number by one, but to leave the mass number
chain. The sub-figure shows the unchanged. At the level of the individual nucleons the reaction is
deviation of the experimental
α-decay half-lives from those n −→ p + e− + ν̄e . (2.10)
predicted by the GN law for the
light Po isotopes. From [5].
The emitted electron can be observed and its energy measured. The associated
anti-neutrino has a very small interaction probability, and so is expected to escape
unobserved. Long before neutrinos were observed, Wolfgang Pauli realised that an
additional, invisible, massless particle was required in order to conserve energy and
momentum in the decay (2.9). His arguments ran as follows. The emitted electrons
are observed to have a variety of different kinetic energies, up to Q. Meanwhile
the mass difference between the parent and daughter nucleus is fixed to a single

14
CHAPTER 2. NUCLEAR PHYSICS 2.4. DECAYS AND REACTIONS

value. The energy given to the recoiling daughter nucleus is small and is fixed by
momentum conservation, so it can’t be responsible for the deficit observed when
the electron has energy less than Q. Energy conservation is then only possible if
the total energy Q can be shared between the electron and some other unobserved
partice – the (anti-)neutrino.

Pauli also argued that without the neutrino the reaction (2.10) would violate angular
momentum conservation. Adding the angular momenta of just the two observed
final state spin-half particles – the electron and the proton – according to the rules
of quantum mechanical angular momentum addition we would find
1 1
⊕ = 0 or 1,
2 2
Neither of the possibilities of total angular momentum s = 0 or s = 1 match the When adding angular momenta
spin of the initial neutron, which has s = 12 . However by adding a third spin-half j1 and j2 in quantum
particle to the final state – the s = 12 anti-neutrino – we can reconstruct a state mechanics the permitted values
of the total angular momentum
which has total angular momentum equal to that of the proton (s = 12 ) since range from |j1 − j2 | to j1 + j2
in integer steps.
1 1 1 1 3
⊕ ⊕ = or .
2 2 2 2 2

Isotopes which have a surplus of protons can proceed via one of two processes. The
first is the emission of a positively charge anti-electron. This is known as positron
emission or β + decay
A A +
Z X −→ Z−1 Z + e + νe (2.11)
FT

The positron is the anti-particle of the electron. It has the same mass as the
A

electron, but positive charge.

DR

EOOE

The second method of decay of proton-rich nuclei is by the nucleus removing one
of the atomic electrons, the electron capture process:
Z
A
ZX + e− −→ Z−1
A
Z + νe (2.12)

These two processes ((2.11) and (2.12)) result in the same change to the nucleus,
and so compete with one another to reduce the Z number of proton-rich nuclei.
OO
When considering whether electron capture or β + decay will dominate we note that
EE

• The Q value for positron emission is 2 × me c2 smaller than that for the Z
corresponding electron capture.
Mass as a function of Z for
• Electron capture relies on there being a substantial overlap of an electron
nuclides of the same A, for
wave-function with the nucleus. odd-A nuclei (above) and
even-A nuclei (below). The
even-A case has two curves
When viewed at the level of the nuclear consitutents, all three of the interactions separated by 2δ.
above — β decay (2.9), β + decay (2.11) and electron capture (2.12) — involve the
interaction of four particles: a proton, a neutron, an (anti-)electron and an (anti-)
neutrino.

n −→ p + e− + ν̄e (2.13)
p −→ n + e+ + νe , (2.14)
−
p+e −→ n + νe . (2.15)

15
2.4. DECAYS AND REACTIONS CHAPTER 2. NUCLEAR PHYSICS

We note that all of the reactions (2.13)–(2.15) are assumed to be occurring inside
the complex environment of the nucleus. Of these three reactions, only neutron
decay (2.13) can occur in isolation, since it is the only one with Q > 0, (the neutron
being about 1.3 MeV/c2 heavier than the proton). The other two reactions (2.14)–
(2.15) occur only within a nucleus, when the energy released from the rearrangement
of the nuclear constituents is sufficient to compensate for the endothermic nature
of the reaction at the level of the individual nucleon.

We note that all of three transitions — β − , β + , and e− capture — leave the mass
number A unchanged. The parent and daughter nuclei are isobars. The decay
processes (2.13)–(2.15) allow transitions between isobars, and mean that for odd-A
nuclei for any value of A there is usually only one stable isobar, that for which the
mass of the system is minimum.

Even-A nuclei may have one stable isobar, but can also have two or very occasionally
three. Multiple stable states are possible for A even because the binding curves for
(even-Z, even-N ) and (odd-Z odd-N ) are separated by 2δ, where δ is the pairing
energy in the SEMF. For an even-even nucleus, since all of the reactions (2.13)–
(2.15) change both |Z| and |N | by one they result in an odd-odd nucleus, and so a
transition to the higher of the two curves. None of the reactions permit a change
in Z of two units, and the probability of two such reactions happening at once is
extremely small, so an even-even nucleus with some Z can be stable provided that
both the neighbouring nuclei with Z + 1 and Z − 1 have larger mass.
FT

Fermi theory of beta decays

A
DR

To understand the lifetimes of the nuclei, we wish to calculate the expected rates
for β ± decays. We follow the method and approximations of Enrico Fermi.

If we put the initial state particles on the left hand side of the diagram, and the final
state particles on the right hand side, then we obtain the following three diagrams
for the reactions (2.13)–(2.15).

e− νe
e νe
n p
ν̄e e+
p n
p n

We shall assume that each of these four-particle interactions will happen at a single
point in space. The amplitude for each reaction is given by the same constant —
a four-body coupling constant which tells us the amplitude for each interaction at
that point in space. For each of the three diagrams that coupling is the Fermi
constant,
GF ≈ 1.17 × 10−5 GeV−2

which has units of inverse energy squared.

16
CHAPTER 2. NUCLEAR PHYSICS 2.4. DECAYS AND REACTIONS

In using a single, constant factor, we implicitly make the simplifying assumption

that the four-body interaction does not depend on the spins of the incoming or
outgoing particles. To simplfy calculations we will also follow Fermi in assuming
that the wave-functions of the electron and the anti-neutrino can be represented by
plane waves. This ignores the effect of the Coulomb attraction between the electron
and the nucleus, and is a good approximation provided that the electron energy is
sufficiently high.

We recall that in quantum mechanics, the rate of some a transition from an initial
state to a final state characterised by a continuum of energy levels is given by the
Fermi Golden Rule4
2π dN
Γ= |Af i |2 (2.16)
~ dEf
Here the transition rate Γ is given in terms of the amplitude Af i connecting the
dN
initial and the final states, and the density dEf
of states at the final energy. It is
dN
the Af i and dEf that we shall have to calculate.

To be concrete, let us consider the beta decay reaction (2.13). We denote the initial
nuclear wave-function by hx|Ψi i, the final nuclear wave-function by hx|Ψf i. The
electron and anti-neutrino wave-functions are approximated as plane waves

hx|φe i ≡ φe = exp (i pe · x) (2.17)

hx|φν i ≡ φν = exp (i pν · x). (2.18)
FT

We can now write down the initial state |Ψi i, which is just that of the parent nucleus
A

|Ψi i = |ψi i,
DR

and final state |Ψf i, which is the product of the daughter nucleus state |ψf i, the
electron state |φe i and the anti-neutrino state |φν̄e i

|Ψf i = |ψf i × |φe i × |φν̄e i.

The matrix element Af i controls the transition from the initial to the final state

Af i = hΨf |A|Ψi i

It can be obtained by working in the position representation and recognising that the
amplitude GF associated with the point-like interaction (2.13) should be integrated
over the volume of the nucleus,
Z
Af i = d3 x GF φ∗e φ∗ν ψf∗ ψi .

The φ and ψ terms are the position representations (wave functions) of the four
particles, and in the final state are found in complex conjugate form. The inte-
gral sums over the amplitudes for the point-like reaction to occur anywhere in the
nucleus, since the reaction could have occurred anywhere within.

To perform the integral we first Taylor expand the exponentials in the plane wave
functions (2.17)–(2.18) for the electron and the neutrino. The expansion is useful
4 For a refresher, see appendix 2.B.1

17
2.4. DECAYS AND REACTIONS CHAPTER 2. NUCLEAR PHYSICS

because the exponents p · x are small.5 The product of φ∗e and φ∗ν can therefore be
written
e−i(pe +pν )·x ≈ 1 − i(pe + pν ) · x + . . . (2.19)
Provided that the first term in this expression does not vanish when performing the
integral, it can be expected to dominate, and the whole integral can be approximated
by
Z
Af i = GF d3 x ψf∗ ψi

≡ GF Mnucl

where in the lower line Mnucl denotes the overlap integral between the neutron in
the parent nucleus and the proton in the daughter nucleus. The size of the quantity
Mnucl depends on the participating nuclei, and is known as the nuclear matrix
element.

In some particularly simple cases Mnucl can be calculated analytically. In particular,

if the initial-state neutron, and the final-state proton happen to inhabit the same
state within a nucleus, the overlap integral is maximal, i.e. for those nuclei

|Mnucl | = 1.

An example of a maximum overlap integral is found for the simplest case of the
isolated neutron decay. 6
FT

To complete the job of calculating Γ we need to find the density of states factor
dN
dEf . The density of states for the outgoing electron can be calculated from the
A
DR

density of states inside a box7 ,

d3 p
dN = .
(2π)3

Assuming spherical symmetry the angular integrals yield 4π so

4π p2 dp
dN = .
(2π)3

A similar result holds for the neutrino. The states allowed by the daughter nucleus
are fixed by total momentum conservation, so provide no further contribution to
the density of states. The recoil energy of the heavy daughter nucleus is negligible,
so conservation of energy gives

Ee + Eν = Q,

where Ee is the kinetic energy of the electron. Hence the rate of decays that yield
electrons with momenta between pe and pe + dpe is

(Q − Ee )2 2
dΓ(pe ) = G2F |Mnucl |2 pe dpe . (2.20)
2π 3
5 The size of x is of order the typical nuclear size, i.e. ∼ 10 fm, which in natural units is

10 fm/(197 MeV fm) ∼ 10−1 MeV−1 . The typical momenta of the out-going particles are of order
MeV, so the dot products in the exponents are of order 10−1 .
6 Such decays are called ‘super-allowed’.
7 See appendix 2.B.2 for the source of this term.

18
CHAPTER 2. NUCLEAR PHYSICS 2.5. SHELL MODEL

In the relativistic limit where Ee  me we can perform the integral and obtain the
simple result
Γβ ∝ Q5 .

i.e. the rate depends on the fifth power of the available energy.

‘Forbidden’ decays (We have assumed that the first term in (2.19) will
dominate, but it can vanish due to selection rules. For example, if the
nuclear matrix element has odd parity the first term vanishes, since then
we are integrating the product of an odd and an even function. In that case,
the next term in the series is required, and the reaction rate is suppressed.
Such decays are said to be ‘first forbidden’. In general the larger the change
in angular momentum required in the nuclear transition, the further along
the series one will need to go to find a non-zero term, and the slower will
be the decay.

From the differential rate (2.20), we see that the observed electrons will have energy
Ee in the range
0 < Ee < Q.
The upper bound is when all the available energy is carried by the electron, while
the lower one is when all that energy is carried by the unobserved neutrino. The
kinetic energy of the heavy recoiling daughter nucleus is assumed to be negligible. Spectrum of electron kinetic
energy expected for beta
decays.
FT

2.4.3 Gamma decays

A
DR

Gamma decays are electromagnetic transitions, and are found when excited nuclear
states relax to their ground states.

Similarly to the beta decay case, one can work out the rate using the Fermi golden
rule. If one represents the initial nuclear wave-function by Ψa and the final nuclear
wave-function by Ψb , then the appropriate matrix element is found to be
Z
hΨf |M |Ψi i = d3 x Ψ∗b (A · Ĵ) e−ik·x Ψa

where A represents the electromagnetic 4-potential and Ĵ = q P̂/m is the electric 4-

current operator. The electromagnetic selection rules and transitions are analogous
to those of atomic physics.

MORE

2.5 Shell Model

The SEMF (recall §2.2.1) provides a reasonable description of the binding energies
of the nuclei for A > 30. However it assumes that the density of states is a
continuous function, whereas in a quantum mechanical system the energy states
will really take discrete levels. We can expect the continuum approximation to

19
2.5. SHELL MODEL CHAPTER 2. NUCLEAR PHYSICS

HB - SEMFLA
9

0.35
8
BA
7
0.30

0.25
5

0.20 4

N
0 5 10 15 20 25 30
0.15

0.10

0.05

0.00
40 60 80 100 120 140 N

-0.05

HB - SEMFLA
(a)
10

9
0.20
FT

8
BA
7
A
DR

6
0.15

0.10 3
Z
0 5 10 15 20 25 30

0.05

0.00
40 60 80 100 Z

(b)

Figure 2.7: Difference between the measured binding energy (per nucleon) and
the SEMF prediction. (a) The x-axis shows the number of neutrons in the nu-
cleus; curves show isotopes (same Z). (b) The x-axis shows the number of
protons in the nucleus; curves show isotones (same N ). In both cases the inset
shows the binding energy per nucleon for the low-A nuclei. The magic numbers
{2, 8, 20, 28, 50, 82, 126} are marked with dashed lines.

20
CHAPTER 2. NUCLEAR PHYSICS 2.5. SHELL MODEL

We will later find that Beta decays are mediated at very small length
scales (∼ 10−18 m) by charged spin-1 force-carrying particles known as
W ± bosons.
e− νe
−
e
e+
d W− u W+ W+
ν̄e νe

u d u d
The Feynman diagrams above show the W bosons responsible for β −
±

decay, β + decay, and electron capture. For projectiles with wavelength

λ  λc,W , where
~
λc,W =
mW c
is the W boson Compton wavelength, or equivalently for projectiles with
momentum p  mW , the small-distance behaviour of the interaction is
not apparent. We do not resolve the W boson and instead we get what
appears to be a single four-body interaction.

break down, for example, when the number of nucleons is small, or when there are
large gaps between energy levels.

Differences from the SEMF at small A (e.g. the tightly bound isotopes 42 He and
16
8 O) are already obvious in Figure 2.1. The discrepancies are shown in more detail
FT

in Figure 2.7 which plots the measured binding energy (per nucleon) and the SEMF
A

prediction. We can see that there are regions of particularly high stability — that
DR

is with anomalously large B/A — clearly visible. The particular stable nuclei are
found around particular special values of N or Z:

{2, 8, 20, 28, 50, 82, 126}.

These are known as the magic numbers.

There are other interesting observations about these ‘magic number’ nuclei. Nuclei
with a magic number of protons tend to have a large number of stable isotopes.
Similarly those with a magic number of neutrons tend to have a large number of
stable istones. Magic nuclei also tend to be more abundant than their neighbours
in the periodic table of the nuclei.

In fact these special values are similar to the filled shells of electrons in atomic
physics. We shall find that they correspond to configurations of nuclear shells that
are exactly filled with either protons or neutrons.

In the next section, we consider more precisely how those discrete energy levels can
change the binding energy.

2.5.1 Solution of the infinite spherical well

As a first approximation, and one which we can solve exactly, we can take a non-
relativistic quantum mechanical model of a nucleus, treating the nucleons indepen-

21
 2.5. SHELL MODEL CHAPTER 2. NUCLEAR PHYSICS

dently. We model the potential as being spherical, and centered on the origin, with
the potential
(
V0 r ≤ R0
V (r) = . (2.21)
∞ r > R0

That is the potential takes a constant value everywhere inside the nucleus, and
is infinite outside. Now we know this does not correspond exactly to the nuclear
potential, which really has a finite potential barrier, and smooth edges rather than
infinite cliff edges, but solving it will teach us something interesting, so let’s give it
a go.

We want to solve the time independent Schrödinger equation

H|ψi = Eψ

in the position basis, to find the energy levels E corresponding to solutions φ(~x) =
h~x|φi. The Hamiltonian for the system of a single particle in a potential V (~x) is

−~2 2
H= ∇ + V (~x).
2m
Our nuclear potential V (r), given by (2.21), means that problem has spherical
symmetry, so let’s try to solve in spherical polar coordinates (r, θ, φ). The solutions
ψ(~x) can be separated into the product
FT

ψ(~x) = R(r) Y (θ, ϕ)

A

of a radial function R(r) which depends only on the radial distance, and a second
DR

function Y (θ, ϕ) which depends only on the polar and azimuthal angles θ and ϕ
1.0

0.8
respectively.
j0 (x)
0.6
j1 (x)
0.4 j2 (x)
j3 (x)
The angular part is solved in a standard way by the spherical harmonics Yl,m (θ, ϕ).
0.2
j4 (x)
They are labelled by two integer quantum numbers,l and m, where l ≥ 0 and
2 4 6 8 10

-0.2 −l ≤ m ≤ l.8
Spherical Bessel functions of
the first kind jl (u) The radial equation is

~2 d2 ~2 l(l + 1)
− (rR(r)) + R(r) + V (r)R(r) = ER(r).
2mr2 dr2 2mr2
It can be simpified using the substitution

2(E − V0 )m
k2 = .
~2
The solutions for the constant potential V0 found inside our nucleus are the linear
combinations
R(r) = al jl (kr) + bl yl (kr),
where the al and bl are constants, and the functions jl (u) and yl (u) known as
spherical Bessel functions of the first kind and of the second kind respectively. The
first few of these spherical Bessel functions are tabulated below.
8 See e.g. Binney and Skinner “The physics of Quantum Mechanics”.

22
CHAPTER 2. NUCLEAR PHYSICS 2.5. SHELL MODEL

l jl (u) yl (u)
sin u
0 u − cosu u
sin u cos u
1 u2 − u − u2 − sinu u
cos u

(3−u ) sin u 3 cos u

2
(u −3) cos u 3 sin u
2
2 u3 − u2 u3 − u2

Table 2.1: The first few Spherical Bessel functions.

The yl (u) are not finite at u = 0 so can’t be part of our wave function. 9 That
leaves us with the jl (u). For our potential, with its infinite step outside the nucleus,
the wave function must vanish at r = r0 . Thus, our radial solutions should be of
the form R(r) = al jl (kr) where k is chosen such that jl (kr0 ) = 0. The jl are
oscillatory, so many such solutions exist.

The solutions are therefore characterised by three numbers: the integers l and m
associated with the spherical harmonics, and a further integer n which identifies
which root of the Bessel function is used in the radial equation. It’s worth taking
note that with this infinite spherical well potential (2.21), large l solutions are
possible even when n is small. This stands in contrast to the solutions of the
Coulomb-like potential V ∝ r−1 common in atomic physics where l ≤ n.

Let’s first take the example for no angular momentum, l = 0. The roots y0 (u) = 0
occur where kr is an integer multiple of π, thought not at kr = 0 since sin u → u as
u → 0. kr = π, 2π, . . . and so on. To satisfy that boundary condition we must have
FT

that r0 k = nπ with n a positive integer. The En,l,m with no angular momentum

(l = 0) are therefore
A

n 2 ~2 π 2
En,0,0 = V0 + .
DR

2mr02

The solutions with larger l can also be determined from the appropriate zeros of
the jl , but they dont have closed-form solutions. In general it will be the case that

~2 kn,l
2
En,l,m = V0 + ,
2m
where kn,l is the value of k which provides the nth zero of jl (kr0 ). Part of the
spectrum is shown in the margin, using the traditional spectroscopic notation where
states with l = 1, 2, 3 and 4 are labeled s,p,d and f respectively.

How many states can we expect at each energy level? The nucleons are spin-half
fermions, so there is a spin degeneracy of 2s + 1 = 2 for each spatial state. For each Low energy part of the
spectrum of the infinite
value of n, l there are 2l + 1 spatial states. So for a particular En,l the multiplicity spherical potential well. The
of states is 2(2l + 1). We would therefore expect the magic numbers of full shells levels are labeled with their
to start with a two for a filled 1s level. Adding the six 1p states give the next spectroscopic labels above, and
magic number of eight. Adding the two 2s and the ten 1d states at the same time, by their degeneracy to the side.
which we might attempt since they have fairly similar energies. would give us a
further twelve states bringing the total to 20, the next magic number. So we have
a magic number sequence of (2, 8, 20,. . . ). These first few magic numbers are well
predicted by this simple model.
9 More precisely we require the probability density |ψ|2 ∝ r 2 y (kr)2 to be finite. For the y this
l l
can be true only for y0 (kr), but that function suffers from a kink in |φ|2 at the origin so it can’t
fit our needs either.

23
2.5. SHELL MODEL CHAPTER 2. NUCLEAR PHYSICS

A FT
DR

Figure 2.8: The plots show ψn,l,m for an infinte spherical well. The slices are in the
x, z plane at ϕ = 0. The quantum number n increases down the column of plots,
while and l increasing across rows. In all cases m = 0.

2.5.2 Spin orbit interaction

The predictions from our simple model of the nucleus would not keep working.
Adding the 1f level, which has l = 3 would give us a further 14 states, bringing the
total to 34. However the next magic number is only 28, so we wanted to add only
8 more states, rather than the 14 predicted by the model.

The naive model has too many degenerate states because it missed the spin-orbit
factor in the Hamiltonian
HS-O = VS−O (r) L · S.
This term accounts for the interaction between the spin of the nucleons and the
field in which they are moving. A similar term arises in atomic physics from the
coupling between the electron magnetic moment and the magnetic field resulting
from its movement around the nucleus.

We can use the standard result for the quantum mechanical angular momentum
operator J that
J2 = L2 + S2 + 2L · S

24
CHAPTER 2. NUCLEAR PHYSICS 2.6. NUCLEAR SCATTERING

Figure 2.9: Shell model prediction of the magic numbers.

to determine the average value of L · S. For an eigenstate of the J2 , L2 and S2

operators,
1

hL · Si = j(j + 1) − l(l + 1) − s(s + 1) .
2
Each nucleon has a fixed value of s = 12 , so the possible values of j are l ± 21 . We
then obtain (
1
l for j = l + 12
hL · Si = 2 1
− 2 (l + 1) for j = l − 12 .

The j = l ± 21 states therefore receive corrections to their energies which for positive
VSO increase the energy of j = l + 12 and decrease the energy of the j = l − 12
states. The required sign of VSO does indeed turn out to be positive.10

Returning to the magic numbers it can be seen that the 1f states which have l = 3
will split into into eight states with j = 7/2 and six with j = 5/2. It is the additional
eight lower-energy 1f7/2 states which are required to make the next magic number.
The remaining magic numbers can be obtained in a similar manner, as can be seen
in Figure 2.9.
spin and parity of odd-A
FT

nuclei
A

mention more sophis-

DR

ticated techniques for

nuclear structure

2.6 Nuclear Scattering

The structure of the nucleus can be probed by scattering projectiles from it. Those
projectiles might be protons, electrons, muons, or indeed other nuclei.

2.6.1 Cross sections

Many experiments take the form of scattering a beam of projectiles into a material.
Inside the material there are many scattering centres (e.g. nuclei) of a similar type.

Provided that the material is sufficiently thin that the flux is approximately constant
within it, the rate of any reaction Wi will be proportional to the flux of incoming
projectiles J (number per unit time) the number density of scattering centres n in
the material (i.e. the number per unit volume), and the width δx of the material.
We can write this as
Wi = σi n J δx. (2.22)
10 That this is the opposite sign from that obtained from the Coulomb potential in atomic physics

can be attributed to the different functional form of the potential V (r) in the two cases.

25
2.6. NUCLEAR SCATTERING CHAPTER 2. NUCLEAR PHYSICS

The constant of proportionality σi has dimensions of area. It is known as the cross

section for process i and is defined by

Wi
σi = (2.23)
n J δx
We can get some feeling for why this is a useful quantity if we rewrite (2.22) as
σ
Wi = (n A δx) J
| {z } |{z}A
Nscat
Pscatt

where A is the total area of the material. Here Nscat is the total number of scattering
centres in the target material. The cross section can be interpreted as the effective
area presented to the beam per scattering centre for which a particular reaction
can be expected to occur. For example if all particles coming within distance d of
a nucleus were scattered then the scattering cross section would be pid2 , the area
of a disk of radius d, and a fraction πd2 /A of the projectiles will be scattered.

When many different scattering processes are available the total rate of loss of beam
is given by W = ΣWi , and the corresponding total cross section is therefore
X
σ= σi .
i

We could choose to quote cross sections in units of e.g. fm2 or in natural units of
GeV−2 , however the most common unit used in nuclear and particle physics is the
FT

so-called barn (b) where

A

1 barn = 10−28 m2
DR

We can convert the barn to natural units of MeV−2 using the ~c conversion constant
as follows
1 barn = 10−28 m2
= 100 fm2 /(197 MeV fm)2
= 0.00257 MeV−2 .

dσ
The differential cross section dΩ is the cross section per unit solid angle of scattered
particle. It is defined to be the rate of scattering per target per unit incoming flux
density per unit solid angle (dΩ) of deflected particle.

2.6.2 Resonances and the Breit-Wigner formula

Sometimes the projectile can be absorbed by the nucleus to form a compound

state, then later reemitted. The time-energy uncertainty relationship of quantum
mechanics tells us that if a state has only a finite lifetime (of order ∆t), then it has
an uncertaintity on its energy ∆E given by
∆E∆t ∼ ~.

26
CHAPTER 2. NUCLEAR PHYSICS 2.6. NUCLEAR SCATTERING

As well as scattering experiments, the size of various nuclei can be deter-

mined by other methods, including:
• Shifts in the energy-levels of atomic electrons from the change of
their Coulomb potential caused by the finite size of the nucleus.
• Muonic equivalents of the above. Muons are about 200 times heavier
than electrons, so their “Bohr raduis” is about 200 times smaller. One
observes the series of x-rays from the atomic (muonic) transitions

The decay width can be generalised to a particle which has many different decay
modes. The rate of decay into mode i is given Γi . The total rate of decay is
given by the sum over all possible decay modes
X
Γ= Γi .
i=1...n

The fraction of particles that decay into final state i, is known as the branching
ratio
Γi
B= .
Γ
The quantity Γi is known as the partial width to final state i, whereas the sum
of all partial widths is known as the total width.

Using the fact that the decay rate Γ = 1/τ , and using natural units to set ~ = 1,
we find that
FT

∆E ∼ Γ. (2.24)
In these units, the uncertainty in the rest-energy of a particle is equal to the rate
A

of its decay. This means that if we take a set of identical unstable particles, and
DR

measure the mass of each, we will expect to get a range of values with width of
order Γ.

Long-lived intermediate states have small Γ and hence well-defined energies. We

tend to think of these reasonably long-lived intermediate states as ‘particles’. The
neutron is an example of an unstable state that lives long enough for the word
‘particle’ to be meaningfully applied to it.

Short-lived intermediate states have large widths and less well defined energies.
When the intermediate state is so short-lived that its width Γ is similar to its mass,
then the decay is so rapid that it is no longer useful to think of it as a particle — it’s
really some transition through which the state happens to be momentarily passing.

We can develop these ideas more quantitatively by considering the general process
A + B −→ O −→ C + D. (2.25)
The initial particles A and B collide to form an unstable intermediate O, which then
decays to the final state C and D (which may or may not have the same particle
content as the initial state). An example of a familiar process is the absorption and
then emission of a photon by an atom, with an intermediate excited atomic state,
A + γ −→ A∗ −→ A + γ.
Alternatively the reaction could represent an inelastic nuclear interaction, for exam-
ple the nuclear absorption of a neutron to form a heavier isotope followed by its

27
 2.6. NUCLEAR SCATTERING CHAPTER 2. NUCLEAR PHYSICS

de-excitation
25 ∗
Mg + n −→ 26 Mg −→ 26 Mg + γ.
Other reactions can create and annihilate other types of particle.

The reaction will proceed most rapidly when the energies of the incoming particles
are correctly tuned to the mass of the intermediate. The reaction rate will be fastest
when the total energy (as measured in the centre-of-mass frame) of A + B is equal
to the rest-mass energy E0 of the intermediate state. The energy need only match
E0 to within the uncertainty Γ in the energy of the intermediate.

Under the condition that the transition from A + B to C + D proceeds exclusively

via the intermediate state ‘0’, of mass m0 and that the width of the intermediate
is not too large (Γ  m0 ), the probability for the scattering process, as a function
of total energy E takes the familiar Lorentzian shape
1
p(E) ∝ . (2.26)
(E − E0 )2 + Γ2 /4
This is the same peaked shape as seen in resonances in situations involving oscil-
lators, and so the excited intermediate state is often called a resonance, and the
process is known as resonant scattering.

Derive Breit-Wigner
shape Taking into account density of states and flux factors, and the possibilities of decay
FT

into multiple different final states, the overall cross-section for the process (2.25)
is given by the Breit-Wigner formula (a full derrivation of which can be found in
A

Appendix ??.
DR

π Γi Γf
σi→0→f = g 2 . (2.27)
k (E − E0 )2 + Γ2 /4
Since excited states are very common, this is an important result not just in nuclear
Σ and particle physics, but also in any process where excitations are found. The terms
in this equation are as follows:

G • Γi is the partial width of the resonance to decay to the initial state A + B

• Γf is the partial width of the resonance to decay to the final state C + D

E0 E • Γ
Pis the full width of the resonance at half-maximum (and equal to the sum
j Γj over all possible decay modes)
The Breit-Wigner line shape.
• E is the centre-of-mass energy of the system
• E0 is the characteristic rest mass energy of the resonance
• k is the wave-number of the incoming projectile in the centre-of-mass frame
which is equal to its momentum in natural units.
(2J+1)
• g = (2s1 +2)(2s 2 +1)
is the probability that the initial particles have the correct
angular momentum to form the resonance. J is the spin of the resonance, and
s1 and s2 are the spins of the two projectiles. This factor averages over the
(2s + 1) spin states of each of the initial state particles (which are assumed
to be unpolarised), and sums over the (2J + 1) spin states of the resonant
intermediate state.

28
CHAPTER 2. NUCLEAR PHYSICS 2.6. NUCLEAR SCATTERING

The cross-section is non-zero at any energy, but has a sharp peak at energies E close
to the rest-mass-energy E0 of the intermediate particle. Longer lived intermediate
particles have smaller Γ and hence sharper peaks.

Resonant scattering experiments can tell us about the excited states of nuclei,
and hence provide further information about nuclear structure and interactions.
All sorts of particles which are too short-lived to travel macroscopic distances can
nevertheless be created as intermediate states and studied from the properties of
their Breit-Wigner peaks.

2.6.3 Nuclear scattering and form factors

When a target only slightly perturbs the wave-function of the projectile, the resulting
scattering behaives rather like optical diffraction.

In optics, the properties of a microscopic aperture can be understood from the

pattern obtained when light, of wavelength similar to the size of the aperture, is
diffracted by that apperture. Far from the aperture, the optical pattern observed is
the two dimensional Fourier transformation of the aperture function. This is true
even if the optical aperture is too small to observe directly.

Now consider scattering a wave from a three-dimensional projectile. Again, the

observed diffraction pattern comes from a Fourier transform of the object, but now
FT

the aperture function is replaced with the potential V (x0 ).

A

In the Born approximation, which is valid for weak potentials, the amplitude f (∆k)
DR

for scattering a projectile such that its change in momentum is ∆k, is proportional
to the 3D Fourier transform of the scattering potential V ,
Z
0
f (∆k) = A d3 x0 V (x0 ) e−i∆k·x (2.28)

where A is a normalising constant. The probability to scatter into some small angle
dΩ is then proportional to |f (∆k)|2 .

Let us consider the scattering of a projectile of charge z from a nucleus with charge
Z and spherically symmetric local charge density ρ(r), centred at the origin. The
potential at some point x0 is given by summing over the Coulomb potentials from
Diffractive scattering from a slit
distributed charges at all other locations x00 , and from an object.
ze2 ρ(x00 )
Z
0
V (x ) = d3 x00 0
4π0 |x − x00 |
ρ(x00 )
Z
= zα d3 x00 0
|x − x00 |

where in the second step we again use the relation (valid in natural units) that the
e2
electromagnetic fine structure constant α = 4π 0
.

Substituting this form of the potential into the Born relation (2.28) we find

ρ(x00 )
Z Z
0
f (∆k) = zαA d3 x0 d3 x00 e−i(∆k)·x 0 .
|x − x00 |

29
 2.6. NUCLEAR SCATTERING CHAPTER 2. NUCLEAR PHYSICS

We may simplify this expression by defining a new variable X = x0 − x00 . This

change of variables allows us to factorize the two integrals, giving the result

e−i∆k·X
 Z  Z 
1 3 00 00 −i∆k·x00 3
f (∆k) = d x ρ(x )e × ZzαA d X . (2.29)
Z |X|
| {z } | {z }
Form Factor Rutherford

The scattering amplitude from a distributed charge is therefore equal to the product
of two terms. The second term can be recognised as the Rutherford scattering
amplitude – the amplitude that would be obtained from scattering from a point
charge density ρ(x) = Zδ(x). The second term therefore tells us nothing about
the internal structure of the nucleus. All of the interesting information about the
nuclear structure is encapsulated in the first term,

1
Z
00
Fnucl (∆k) = d3 x00 ρ(x00 ) e−i∆k·x
Z

which is known as the nuclear form factor. The form factor is the three-dimensional
Fourier transform of the normalised charge density ρ(x)/Z. All of the interesting
information about the size and structure of the nucleus is found in Fnucl (∆k).
We will find interesting scattering — that is interesting ‘diffraction patterns’ — if
the exponent is of order unity. For this to be true the de Broglie wave-length of
the projectile must be of the same order as the nuclear size, as was noted in the
introduction to this chapter.
FT

The mod-squared of the optical amplitude gives the intensity. Similarly it is |F |2

A

which is important when considering the flux of scattered projectile particles. The
DR

rate at which particles are scattered into unit solid angle is given by
 
dN 2 dN
= |Fnucl (|∆k|)| . (2.30)
dΩ dΩ Rutherford

Sketch of a nuclear form factor This equation is more often written in terms of the differential cross section for
diffraction pattern. scattering
 
dσ dσ
= |Fnucl (|∆k|)|2 .
dΩ dΩ Rutherford

By examining the form factor for particles scattered with various changes in mo-
mentum |∆k| we can infer information about ρ(x) and hence about the size and
shape of the nuclear potential V (x).

30
CHAPTER 2. NUCLEAR PHYSICS 2.7. KEY POINTS

2.7 Key points

• In natural units (appendix 2.A), ~ = c = 1 and

[Mass] = [Energy] = [Momentum] = [Time]−1 = [Distance]−1

A useful conversion constant is

~c ≈ 197 MeV fm

• The nuclear mass is well described by the semi-empirical mass formula

2 (A − 2Z)2 Z2
M (A, Z) = Zmp + (A − Z)mn − αA + βA 3 + γ +  1 − δ(A, Z).
A A3

• The binding energy leads to a valley of stability in the (A, Z) plane where
the stable nuclei lie

• In a reaction or decay, the Q-value is the energy released in a decay

X X
Q= Mi − Mf

If Q > 0 the reaction is exothermic – it gives out energy, whereas if Q < 0

FT

the reaction is endothermic, and energy must be supplied for it to proceed.

A

• Alpha decay rates are dominated by quantum tunnelling through the Coulomb
DR

barrier.

• Beta decay rates and electron capture are governed by the Fermi coupling
constant
GF ≈ 1.17 × 10−5 GeV−2

• The cross section is defined by:

Wi
σi = (2.31)
n J δx

The differential cross section is the cross section per unit solid angle

dσi
dΩ

• Cross sections for sub-atomic physics are often expressed in the unit of barns.

1 barn = 10−28 m2

• The Breit Wigner formula for resonant scattering is

π Γi Γf
σi→0→f = .
k (E − E0 )2 + Γ2 /4
2

31
2.7. KEY POINTS CHAPTER 2. NUCLEAR PHYSICS

• In elastic nuclear scattering the form factor

 
ρ(x)
Z
F (|∆k|) = d3 x e−i∆k·x ,
Z

is the 3D Fourier transform of the normalised charge density, and is related

to the Rutherford scattering differential cross section by
 
dσ dσ
= |Fnucl (|∆k|)|2 .
dΩ dΩ Rutherford

A FT
DR

32
Appendix

Appendix 2.A Natural units

In the S.I. system of units, times are measured in seconds and distances in meters.
In those units the speed of light takes the value close to 3 × 108 ms−1 .

We could instead have chosen to use unit of time such that c = 1. For example
we could have used units in which time is measured in seconds and distance in
light-seconds. In those units the speed of light is one (one light-second per second).
Using units in which c = 1 allows us to leave c out of our equations (provided
we are careful to remember the units we are working in). Such units are useful in
FT

relativistic systems, since now the relativistic energy-momentum-mass relations are

simplified to
A
DR

E = γm
p = γmv
E 2 − p2 = m2 .

So for a relativistic system setting c = 1 means that energy, mass and momentum
all have the same dimensions.

Since we are interested in quantum systems, we can go further and look for units in
which ~ is also equal to one. In such units the energy E of a photon will be equal
to its angular frequency ω
E = ~ω = ω.
Setting ~ = 1 therefore means that the units of energy are the same as the units of
inverse time. Units with ~ = 1 imply that time (and via c = 1 distance too) must
have the same dimensions as inverse energy, E −1 .

So in our system natural units with ~ = c = 1 we have have that all of the following
dimensions are the same:

[Mass] = [Energy] = [Momentum] = [Time]−1 = [Distance]−1

We are still free to choose a convenient unit for all of these quantities. In subatomic
physics it is common to use units of energy (or inverse energy). The nuclear energy
levels have typical energies of the order of 106 electron-volts, so we shall measure

33
2.B. TOOLS CHAPTER 2. NUCLEAR PHYSICS

energies, momenta and masses in MeV, and lengths and times in MeV−1 . At the
end of a calculation we might wish to recover, for example, a “real” length from
one measured in our MeV−1 units. To do so we can make use of the conversion
factor
~c ≈ 197 MeV fm

which tells us that one of our MeV−1 length units corresponds to 197 fm where
1 fm = 10−15 m.

Appendix 2.B Tools for cross-section calculations

2.B.1 Decays and the Fermi Golden Rule

In subatomic physics we are interested in the decays of unstable particles, such

as radioactive nuclei, muons from the atmosphere, or Higgs bosons. Using time-
dependent perturbation theory in quantum mechanics it is possible to show that
that the transition rate of an unstable state into a continuum of other states is
given by the Fermi Golden Rule:

2π dN
Γ= |Vf i |2 , (2.32)
~ dEf
FT

where
A
DR

• Γ is the rate of the decay

• Vf i = hf |V |ii is the matrix element of the Hamiltonian coupling the initial

and the final states, obtained by acting on the initial state with the perturbing
Hamiltonian, then finding the inner product of that pertubed state with the
final state.
dN
• dEf is the density of final states per unit energy, which tells us how many
states are accessible from the continuum.

A derivation of this result can be found in all good quantum mechanics textbooks.

2.B.2 Density of states

The density of states for a single particle within a cubic box with sides length a can
be calculated as follows. The plane wave solution is of form

hx|Ψi ∝ exp (ik · x) .

If we require periodic boundary conditions, with period a equal to the side of the
box, then the values of the wavenumber kx are constrained to kx = 2πn/a for

34
CHAPTER 2. NUCLEAR PHYSICS 2.B. TOOLS

integer n. Similar conditions hold for ky and kz . The number of momentum states
within some range of momentum d3 p = d3 k (for ~ = 1) is therefore given by
d3 p
dN = V
(2π~)3
where V = a3 is the volume of the box.

2.B.3 Fermi G.R. example

Consider the isotropic decay of a neutral spin-0 particle A into two massless daugh-
ters, B and C
A −→ B + C.
The Fermi G.R. gives the decay rate (in natural units) of A as
dN
Γ = 2π |Vf i |2 .
dEf
The density of final states can be found from the allowed momenta pB of particle
B.
d3 pB
dN = V
(2π)3
When pB is fixed there is no further freedom for pC since the sum of the momenta
of the two final state particles is fixed by total momentum conservation. This
FT

constraint means that for the two body final state there is no additional term in the
density of states for pC .11 Since all decay angles are equally probable, the integrals
A
DR

over the angles contribute 4π, leading to

4π p2B dpB
Γ = 2π |Vf i |2 V.
(2π)3 dEf
dpB
The relativistic decay products each have momentum |pB | = Ef /2 so dE f
= 12 .
Normalising to one unstable particle in our unit volume gives V = 1, and results in
a decay rate
1
Γ = |Vf i |2 p2B
2π
1
= |Vf i |2 m2A .
8π

2.B.4 Lifetimes and decays

The number of particles remaining at time t is governed by the decay law12

dN
= −ΓN,
dt
3
11 For d p
a three-body final state there would be terms in dN of the form (2π)3 for two of the three
particles, the third again being fixed by momentum conservation.
12 The decay law was discovered experimentally by Frederick Soddy (1877-1956). Soddy, who

had been a scholar at Merton, was also first person to understand that radioactivity led to the
transmutation of the elements — in effect making him the first true alchemist.

35
2.B. TOOLS CHAPTER 2. NUCLEAR PHYSICS

where the constant Γ is the decay rate per nucleus. The equation is easily integrated
to give
N (t) = N0 exp (−Γt) .
We can calculate the particles’ average proper lifetime τ , using the probability that
they decay between time t and t + δt
1 dN
p(t) δt = − δt = Γ exp (−Γt) δt.
N0 dt
The mean lifetime is then

τ = hti
R∞
t p(t) dt
= R0∞
0
p(t) dt
1
=
Γ

The decay law can be justified from precise experimental verification. In essence
it represents a statement that the decay rate is independant of the history of the
nucleus, its method of preparation and its environment. These are often excellent
approximations, provided that the nucleus lives long enough that has mass m  Γ
where Γ is its decay width, and provided it it not bombarded with particularly
disruptive projectiles, such as high-energy strongly interacting particles.
A FT

2.B.5 The flux factor

DR

When calculating a cross section σ from a rate Γ, we need to take into account
that for scattering from a single fixed target
W
σ=
J
where J is the flux density of incoming particles. The flux density is itself given by

J = np v

where np is the number density of projectiles and v is their speed. If we normalise to

one incoming particle per unit volume, then np = 1 and the cross section is simply
related to the rate by
W
σ=
v

2.B.6 Luminosity

In a collider — a machine which collides opposing beams of particles — the rate

of any particular reaction will be proportional to the cross section for that reaction
and on various other parameters which depend on the machine set-up. Those
parameters will include the number of particles in each collding bunch, their spatial
distributions, and their frequency of bunch crossings.

36
CHAPTER 2. NUCLEAR PHYSICS 2.B. TOOLS

We can define a parameter called the luminosity L which encapsulates all the
relevant machine parameters. It is related to the rate W and the cross section σ by
W
L= .
σ
For any collider the luminosity tells us the instantaneous rate of reaction for any
cross section. The product of the time-integrated luminosity and the cross section
tell us the expected count of the events of that type
Z
Nevents, i = σi L dt .

For a machine colliding trains of counter-rotating bunches containing N1 and N2

particles respectively at a bunch-crossing rate f , we can show that the luminosity
is
N1 N2 f
L= ,
A
where A is the cross-sectional area of each bunch (perpendicular to the beam di-
rection).

We have assumed above that the distributions of particles within each bunch is
uniform. If that is not the case (e.g. in most real experiments the beams have
approximately Gaussian profiles) then we will have to calculate the effective overlap
area A of the bunches by performing an appropriate integral.
A FT
DR

Further Reading
• “An Introduction to Nuclear Physics”, W. N. Cottingham and D. A. Green-
wood, 2001 for the basics
• “Nuclear Physics”, M.G. Bowler, Pergamon press, 1973

• “Introductory Nuclear Physics”, P.E. Hodgeson, E. Gadioli and E. Gadioli

Erba, OUP, 2003
• The BNL table of the nuclides provides good reference data
[Link]

Bolwer and Hodgeson et. al. are good books which go well beyond the material in
this introduction.

37
2.B. TOOLS CHAPTER 2. NUCLEAR PHYSICS

A FT
DR

38
Chapter 3

Applications of Nuclear Physics

FT
A
DR

39
 CHAPTER 3. NUCLEAR APPLICATIONS

FT
A
DR

40
Chapter 4

Modern Nuclear Physics

FT
A
DR

41
 CHAPTER 4. MODERN NUCLEAR

FT
A
DR

42
Chapter 5

Hadrons

FT
A
DR

43
 CHAPTER 5. HADRONS

FT
A
DR

44
Chapter 6

Non-relativistic scattering theory

FT
A
DR

45
 CHAPTER 6. SCATTERING THEORY

FT
A
DR

46
Chapter 7

Feynman diagrams

FT
A
DR

47
 CHAPTER 7. FEYNMAN DIAGRAMS

FT
A
DR

48
Chapter 8

The Standard Model

FT
A
DR

49
 CHAPTER 8. THE STANDARD MODEL

FT
A
DR

50
Chapter 9

Accelerators and detectors

FT
A
DR

51
Examples

FT
A
DR

52
Appendix A

Examples

FT
A
DR

53
Examples

A.1 Radioactivity and nuclear stability

Solutions

Three vectors are written in bold e.g. x.

Four vectors are written sans-serif, e.g. P.
The metric is diag(1, −1, −1, −1) so that with X · X = c2 t2 − x · x,
time-like intervals have positive signs, and propagating particles satisfy
P · P = +m2 .

1.1. a) What assumptions underlie the radioactivity law

dN
= −ΓN ?
dt
b) If some number N0 of nuclei are present at time t = 0 how many are present at
some later time t?

c) Calculate the mean life τ of the species.

d) Relate the half-life t 21 to τ .

e) How do things change when the particle moves relativistically?

FT

a) Assumptions: rate of loss depends only on number of number of nuclei, not on

A

environment. Approximately valid because the typical energy scales in the environment
DR

are usually much too small to induce transitions.

b) Integrating the differential equation

dN 0
Z N Z t
0
= − Γ dt0
N0 N 0

we find the familiar N = N0 e−Γt .

c) The mean life is

R R ∞ −Γt
t P (t) dt te dt 1
τ = hti = R = R0 ∞ −Γt =
P (t) dt 0
e dt Γ

since the probability of decaying between times t and t + dt is P (t) = Γe−Γt .

d) Taking logs:
−t/τ = ln n/n0
1
and setting n/n0 = 2
, giving
t 1 = τ ln 2
2

e) Particles which are travelling relativistically are time dilated by factor γ

 
x
n(x) = n0 exp − .
βγcτ

54
Examples

1.2. A sample consists originally of nucleus A only, but subsequently decays

according to
A B Γ Γ
A −−→ B −−→ C.

Write down differential expressions for dA dB dC

dt , dt and dt . Solve for, and then sketch,
the fractions of, A(t), B(t) and C(t). At what time is the decay rate of B maxi-
mum?

The differential equations are

d
A = −ΓA A
dt
d
B = ΓA A − ΓB B
dt
d
C = ΓB B.
dt
The first is not coupled:
A = A0 e−ΓA t .
Trying solution B = αeΓA t + βeΓB t with boundary condition α + β = 0 and we find
α = ΓAB0−Γ
ΓA
A
, and β = −α, so

A0 ΓA  −ΓA t 
B= e − e−ΓB t .
ΓB − ΓA
We can get C by direct integration of B, with B.C. B(0) = 0
FT

Z
A0 h    i
C = ΓB B dt = ΓA e−ΓB t − 1 − ΓB e−ΓA t − 1 .
A

ΓB − ΓA
DR

Decays of B are maximum when B is maximum. We can calculate Ḃ using the

expression for B above, which is zero when
∗ ∗
0 = ΓA e−ΓA t − ΓB e−ΓB t .

So t∗ = (ΓA − ΓB )−1 ln(ΓA /ΓB ).

Check for different {ΓA , ΓB }:

:3, > :3, > : 3 , 5>

1 1 1 2 1
4 3
NN0 NN0 NN0
1.0 1.0 1.0

0.8 0.8 0.8

0.6 0.6 0.6

0.4 0.4 0.4

0.2 0.2 0.2

t t t
1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

:1, > :1, > 81, 5<

1 2
4 3
NN0
NN0 NN0
1.0
1.0 1.0

0.8
0.8 0.8

0.6
0.6 0.6

0.4
0.4 0.4

0.2
0.2 0.2

t
t t 1 2 3 4 5
1 2 3 4 5 1 2 3 4 5

:4, > :4, > 84, 5<

1 2
4 3
NN0
NN0 NN0
1.0
1.0 1.0

0.8
0.8 0.8

0.6
0.6 0.6

0.4
0.4 0.4

0.2
0.2 0.2

t
t t 1 2 3 4 5
1 2 3 4 5 1 2 3 4 5

55
Examples

1.3. Consider Compton scattering γ + e− → γ + e− of a photon of energy Eγ

(∼ MeV) from a stationary electron.

a) Show that the energy of the scattered photon is

me Eγ
Eγ0 = ,
me + Eγ (1 − cos θ)
where θ is the angle through which the photon is scattered.

b) If an incoming photon is scattered through an angle ∼ 180◦ at the surface of

a material, how much of the original photon’s energy would you expect not to be
deposited in the material?
a) Letting the photon momentum be P and the electron momentum be Q, with primes
after the scatter. Then energy-momentum conservation is given by

P + Q = P0 + Q0

To eliminate the components of Q0 ,

2
Q02 = P + Q − P0 .

The Q and Q terms cancel on each side, and the P2 and P02 terms are zero. The
2 02

cross terms must give

0 = P · Q − P0 · Q − P · P0 .
Without loss of generality the four-vectors can be expressed as
FT

E0
     
E me
 E   0  0
 E cos θ 
P0 = 
A

P=  0 
 Q=  0 

 E 0 sin θ  .

DR

0 0 0

So we have that
0 = Eme − E 0 me − eE 0 (1 − cos θ)
which can be rearranged into the form given.

b) You don’t see the the energy from the scattered photon, so you fail to see mE/(m+
2E) of the original energy.

1.4. Briefly explain the origin of each of the terms in the semi-empirical mass
formula (SEMF)
2 (N − Z)2 Z2
M (N, Z) = Zmp + N mn − αA + βA 3 + γ +  1 + δ(N, Z)
A A3
and obtain a value for .

Show that we can include the gravitational interaction between the nucleons by
adding a term to the SEMF of the form
−ζA5/3
and find the value of ζ.

Use this modified SEMF to obtain a lower bound on the mass of a gravitationally-
bound ‘nucleus’ consisting only of neutrons (a neutron star).

56
Examples

Obtain value for  by building uniform sphere of charge from spherical shells, out to
the nuclear radius of 1
R = r0 A 3
with r0 ≈ 1.25 fm. The shell of width dr at radius r makes a contribution to the energy
of
1
dE = Q(r)dQ
4π0 r
where Q(r) = Q0 (r/R)3 and

4πr2 dr r2 dr
dQ = Q0 4 = 3Q 3
3
πR 3 R

so that the total electrostatic energy is given by

Z R 4
3Q2 r 3 Q2
E= dr =
4π0 0 R6 5 4π0 R
Comparing with the SEMF we see that

3 e2 1 3 1 197 MeV fm
ε= = ≈ 0.69 MeV.
5 4π0 r0 5 137 1.25 fm
which compares well with the value of 0.71 MeV obtained by fitting the data.

To include the contribution from gravity we need to add an attractive gravitational term
Q2
calculated in much the same way as the Coulomb term, except with 4π 0
replaced by
−GN M 2 , so assuming mn = mp = mN , taking M ≈ mN A and R = r0 A1/3 we have
FT

3 GN m2N
ζ= ≈ 5.6 × 10−37 MeV.
5 r0
A
DR

For a neutron star with Z = 0 and A = N we then have binding energy

B = αN − βN 2/3 − γN + ζN 5/3 .

For large N we can ignore the surface term which has coefficient of the same order as
α and γ but is ∝ N 2/3 . We have zero binding energy when B = 0 i.e. for

(γ − α)N = ζN 5/3 .

Since γ − α ≈ (23.2 − 15.8) = 7.4 MeV we obtain a lower bound on the size of a
neutron star
 3/2  3/2
γ−α 7.4
N> = = 4.8 × 1055 ,
ζ 5.6 × 10−37

meaning that the mass must be larger than 8.0 × 1028 kg which is 0.06 Msun . Observed
neutron stars typically contain 1.3 to 2 solar masses.

1.5. An analysis of a chart showing all stable (t 12 > 109 years) nuclei shows that
there are 177 even-even, 121 even-odd and 8 odd-odd stable nuclei and, for each
A, only one, two or three stable isobars. Explain these observations qualitatively
using the SEMF. Energetically 106 106
48 Cd could decay to 46 Pd with an energy release
106
of greater than 2 MeV. Why does 48 Cd occur naturally?
e-e energetically most favourable, o-o least favourable

Only 2-3 stable nuclei due to asymmetry and Coulomb terms. For const A energy is
parabolic in Z, so isobars beta ± decay towards bottom of parabola.

57
Examples

For 106
48 Cd to decay requires ∆Z = 2, which must proceed via either (a) two beta
decays the first of which is e-e to o-o and is energetically forbidden or (b) a double
beta decay which has a very low rate.

OO

EE
Z=48
Z=46

1
1.6. The radius r of a nucleus with mass number A is given by r = r0 A 3 with
r0 = 1.2 fm. What does this tell us about the nuclear force?

a) Use the Fermi gas model (assuming N ≈ Z) to show that the energy F of the
Fermi level is given by
  23
~2 9π
F = .
2mr02 8

16
b) Estimate the total kinetic energy of the nucleons in an O nucleus.

c) For a nucleus with neutron number N and proton number Z the asymmetry term
in the semi-empirical mass formula is
FT

γ(N − Z)2
A

.
DR

Assuming that (N − Z)  A use the Fermi gas model to justify this form and to
estimate the value of γ. Comment on the value obtained.

The proportionality r ∝ A1/3 tells us that the every nucleon gives the same contribution
to the volume (‘close-packed’). If the nuclear force extended beyond nearest neighbours
then the nucleons in large-A nuclei could stack up in the higher energy levels in a smaller
volume. So ‘close packing’ is evidence in favour of nearest-neighbour forces.

a) In terms of the Fermi momentum pX the total number of states (including factor
of 2 for spin) is
4π p3X
X =2× V.
3 (2π~)3
Eliminating V = 4π
3
Ar03 the Fermi momentum for either species is
 1
~ 9πX 3
pX = .
r0 4A

Assuming X/A ≈ 1
2
and using F = p2N /2m we get the result given.

dN 1
b) The Fermi energy is 33.5 MeV. The density of states dE
is ∝ E 2 , so
R F dN R F 3
E dE dE E 2 dE 3
hEi = R F dN = R 1 = F .
dE
dE F
E dE
2 5

58
Examples

c) The total energy is

E = N hEN i + ZhEZ i
3
= (N p2N + Zp2Z )
10m
5 5
3 N3 +Z3
= F 2 .
5 (A/2) 3

Assume Z ≈ N . Let (Z − N )/A = δ so that N = A/2(1 − δ) and Z = A/2(1 + δ).

Expanding for δ  1, (1 ± x)P ≈ 1 ± P x + P (P − 1)x2 /2!. The quadratic term (with
P = 35 ) must provide the asymmetry term

3 A 10 2 1 1 (N − Z)2
Easym = F δ = F Aδ 2 = F
5 2 9 3 3 A
so the asymmetry term prefactor γ = 13 F ≈ 11.2 MeV. This is smaller than the typical
value of about 23 MeV because we have not taken into account the change in the depth
of the potential.

1.7. Alpha decay rates are determined by the probability of tunnelling through
the Coulomb barrier. Draw a diagram of the potential energy V (r) as a function
of the distance r between the daughter nucleus and the α particle, and of the wave
function hr|ψi.

The decay rate can be expressed as Γ = f P , where f is the frequency of attempts

FT

by the alpha particle to escape and P = exp (−2G) is the probability for the alpha
particle to escape on any given attempt.
A
DR

By using a one-dimensional Hamiltonian

p2r
H= + V,
2m
∂
with a 1D momentum operator pr = −i~ ∂r , and by representing the wave function
by
hr|Ψi ≡ Ψ(r) = exp[η(r)],
for some function η(r) with r = x show that
√
2m b p
Z
G= V (r) − Q dr.
~ a

Since the maximum potential barrier height is usually much larger Q, we can ap-
proximate the integrand by neglecting Q inside the integrand. Show √ that this
approximation leads to a transition rate λ of the form ln λ = C − D/ Q.

[Hint: can you convince yourself that η 00  (η 0 )2 ?]

Substituting Ψ = exp[η(r)] into Schrödinger and dividing by exp(η) we find that the
η(r) must satisfy
~2  00
η + (η 0 )2 + V (r).

Q=−
2m
We can assume η 00  (η 0 )2 if we are in the semi-classical regime where the poten-
tial is “smooth”, meaning that there are many e-folds of the wave function before

59
Examples

the potential changes significantly. This should be true except close to the nuclear
boundary.

If you need to convince yourself of this, consider that for the very low tunnelling probabilities we
are dealing with there must be very many e-foldings of the wave-function within the classically
forbidden region. We can then see that the potential is approximately constant over each
e-folding, so that a e−kr form (with k approximately constant over an individual e-folding) is
a decent approximation.

The potential is (
−V0 r < ra
V (r) = zZα
r
r > ra .
We can estimate the tunnelling probability (neglecting the ∇ in the flux) by simply find-
ing the amplitude for the particle to make its way from the nucleus into the classically
allowed region. This amplitude is

|Ψ(rb )|
= eηb −ηa = e−G
|Ψ(ra )|

where the exponent is given by

r Z rb
2m p
−G = ηb − ηa = V (r) − Q dr
~2 ra

The lower bound is the nuclear radius

1
ra ≈ r0 A 3
FT

and the upper bound is the edge of the classically allowed region
A

Zzα
DR

rb = .
Q
Neglecting Q will be reasonable for any reasonably long-lived nucleus, since if Q ∼
V (ra ) then tunnelling will occur rapidly, and the nucleus will decay very quickly. With
this approximation r r
2m Zzα/Q Zzα
Z
−G = dr
~2 ra r
which yields rr 
2mZzα Zzα
−G = 2 − r a .
~2 Q
√
The rate λ ∝ exp −2G so log λ ∝ C + D/ Q as required.

1.8. Complete the integral from the previous question without making the approx-
imation that Q is small. You might wish to consider the substitution r = rb cos2 θ
helps. Show that s
π 2mc2
G= Zzα F(ra /rb )
2 Q
where the dimensionless function
2  −1 √ p 
F(r) = cos r − r(1 − r)
π
lies in the range between 0 and 1, and for small Q approaches 1.

Compare this solution to that from the previous question.

60
Examples

We can rewrite r rb
Z r
2mQ rb
G= − 1 dr
~2 ra r
The integral can be solved by substituting r = rb cos2 θ giving
r Z θa
2mQ
G = r b 2 sin2 θ dθ
~2 0
r
2mQ
= rb [θa − sin θa cos θa ]
~2

which can be rewritten in the form given, with F taking the form below.

1.0

0.8

0.6

0.4

0.2

0.2 0.4 0.6 0.8 1.0

FT

1.9. a) What are the basic assumptions of the Fermi theory of beta decay?
A

b) The Fermi theory predicts that in a beta decay the rate of electrons emitted with
DR

momentum between p and p + dp is given by

dΓ 2π 2 nuc 2 1
= G |Mf i | (E − Q)2 p2 ,
dpe ~ 4π 4 ~6 c3
where E is the energy of the electron, and Q is the energy released in the reaction.
Justify the form of this result.

c) Show that for Q  me c2 the total rate is proportional to Q5

d) What spin states are allowed for the combined system of the electron + neutrino?

e) ‡ Why are transitions between initial and final nuclei with angular momenta
differing by more than ~ suppressed?

[Hint: what form does the Schrödinger equation take for angular momentum quantum
number l 6= 0?]
a) Point-like interaction, matrix element constant, heavy nucleus, plane-wave solution
for electron and neutrino.

b) 2π
~
and |Mf i |2 come from Fermi’s Golden Rule. The other terms come from the
density of states
4πq 2 dq 4πp2 dp
dNν = 3
dNe =
(2π~) (2π~)3
A heavy recoil nucleus allows any values for p and q without gaining energy itself.
Energy conservation gives Ee = Q − Eν so the neutrino momentum is q = (E − Q)/c.

61
Examples

For a fixed value of dpe we have that dEf = dEν so the density of states

dN dN (E − Q)2 p2 dp
= =
dEf dEν 4π 4 ~6 c3
giving the desired result.

c) For E  me , we can approximate p by E and dp by dE giving

Z Q
A 5
Γ=A (E − Q)2 E 2 dE = Q .
0 30

d) The electron and neutrino are each spin-half so can be in a combined S = 0 (Fermi)
or S = 1 (Gamow-Teller) state.

e) The matrix element is

Mf i = hΨf Ψν Ψe |O|Ψi i
The leptonic wave functions hr|Ψe i, hr|Ψν i are plane waves (only approximately true
for the e− ) which we can expand as

hr|Ψi ∝ exp (ik · r) = 1 + ik · r + . . .

since for typical energies k · r  1.

‘Allowed’ transitions have non-zero matrix elements for the first term in this expansion,
which is only non-zero for no lepton angular momentum L = 0. The decay is ‘first
forbidden’ if the nuclear matrix element is such that the the first term in the expansion
FT

is zero and the second ∝ k · r is non-zero (i.e. if we require lepton L = 1). And so on
for second, third, . . . forbidden transitions. The decay is ‘super-allowed’ when nuclear
A
DR

matrix elements also have maximal overlap integrals.

1.10. A high energy photon can create an electron-positron pair within the
material. When a positron comes to rest it will annihilate against an electron from
the material

e+ + e− → γ + γ.

What will be the energy of these secondary photons?

62
Examples

The figure shows the energy spectrum from the gamma decay of 24 Na as measured
in a small Ge(Li) detector. Suggest the origins of the peaks A, B, C and the edge
D. For such a detector describe the stages by which gamma ray energy is converted
into a measurable voltage pulse.
Each photon has 511 keV of energy. In the plot, peak A comes from complete absorption
of a photon in the material. Comment that peak width comes mostly from resolution
of detector.
A FT

All other features come from loss of some fraction of energy from the absorption
DR

material. D is the Compton edge from reverse scattering of the photon, near the
boundary of the material. Edge occurs at position corresponding to θ = π.

Some of the other features result from pair creation followed by annihilation to two
511 keV photons. If one photon escapes then we get peak B. If both escape or the e+
escapes we get the small peak next to C.

24
The large peak at C itself seems to be a second gamma emission line of Na [see
spectrum]

244 240
1.11. The figure shows the α-decay scheme of 96 Cm and 94 Pu.

63
Examples

Show — either by (a) using a suitable approximation of F calculated in a previous

question or (b) by redoing the corresponding integral neglecting Q — that we would
expect the rates to satisfy and equation of the form
BZ
log Γ = A − √ .
Q

The Q value for the ground state to ground state transition is 5.902 MeV and for
this transition A = 132.8 and B = 3.97( MeV)1/2 when Γ is in s−1 . The branching
ratio for this transition is given in the figure. Calculate the mean life of 244 Cm.

Estimate the transition rate from the ground state of 244 Cm to the 6+ level of
240
Pu using the same A and B and compare to the branching ratio given in the
figure.

Suggest a reason for any discrepancy.

For these decays rb = Zzα/Q ≈ 47 fm which is much larger
√ nuclear radius ra ≈
1.25A1/3 = 8 fm, so we can assume F ≈ 1, leaving G ∝ Z/ Q. (Alternatively redo
the integral from the previous question omitting Q.

For the values given, the decay constant for the 0+ state is
 
3.97 × 94
Γ0+ = exp 132.8 − s−1
5.902

So
FT

τ0+ = 1/Γ+ 9
0 = 1.09 × 10 s = 34.6 years.
A

However this only accounts for 76.7% of the total rate so the lifetime is
DR

τ = 0.767 × 34.6 years = 26.5 years

The same values applied to the 6+ state give

τ6+ ≈ 5.8 × 1010 s

which is ≈ 53×τ0+ . We’d then expect a branching ratio of 76.7%/53 = 1.4%; however
we observe only 0.0036%.

When there is angular momentum we must modify the potential in the 1d Schrödringer
equation
Zzα Zzα ~2
→ + l(l + 1)
r r 2mr2
meaning that the effective potential barrier is larger when l 6= 0. This supresses larger
δJ transitions beyond what would be expected from the reduced Q value alone.

1.12. Discuss the evidence for shell structure in the atomic nucleus. What is
meant by the ‘magic numbers’, and what values do they take?

Consider a model for the nucleus with a three-dimensional harmonic potential,

V (x) = 21 κ2 x2 .
By solving the single-particle Schrödinger equation in Cartesian coordinates, or oth-
erwise, show that the degeneracy of the nth energy level is (n+1)(n+2). Calculate
the degeneracies of the first five energy levels and compare this model with the magic
numbers. How could the model be improved?

64
Examples

1.13. Deduce from the shell model the spins and parities of the ground states of
the following nuclei, stating any assumptions you make: 73 Li, 178 O, 20 27 14
10 Ne, 13 Al, 7 N,
39 41
19 K, 21 Sc.

Evidence: abundances; deviation of B/A from SEMF; spins (as below)

Magnetic moments measured from NMR match expectation from unpaired nucleons.
Calculate by using Hamiltonian
Hµ = −µ · B
with µ = µL + µS with j = l ± 12 . [See e.g. Cottingham & Greenwood §5.6.]

From something like the Saxon-Woods potential (rounded off square well), with spin-
orbit interation Vls (r) ∝ L · S, the energy levels are ordered such that

E(l + 12 ) < E(l − 21 )

Order of shells is:

1s 1 1p 3 1p 1 1d 5 2s 1 1d 3 1f 7 2p 3 1f 5 2p 1 1g 9
2
|{z} | 2{z 2} | 2 {z2 2 2
} |{z} | 2 2
{z 2 2
}
2 +6=8 +12=20 +8=28 +22=50

The next magic numbers are 82 and 126 (see e.g. Cottingham & Greenwood).

Spin – parities:
FT

3−
A

7
3 Li p in 1p 3 2
2
DR

17 5+
8O n in 1d 5 2
2
20 +
10 Ne even-even 0
27 5+
13 Al p in 1d 5 2
2
14 +
7N n and p in 1p 1 0 or 1+
2
39 3+
19 K p in 1d 3 2
2
41 7−
21 Sc p in 1f 7 2
2

65
Examples

A.2 Quarks and scattering

Solutions

2.1. What is meant by the ‘cross section’ and the ‘differential cross section’ ?

Consider classical Rutherford scattering of a particle with mass m and initial speed
v0 from a potential
α
V (r) =
r

a) Show from geometry that the change in momentum is given by

|∆p| = 2p sin(Θ/2).

b) Considering the symmetry of the problem, show that

dθ
bv0 = r2
dt
where b is the impact parameter, r is the location of the particle from the origin
and θ is the angle ∠(r, r∗ ) where r∗ is the point of closest approach.

c) Starting from Newton’s second law show that

FT

 
2α Θ
A

|∆p| = cos .
v0 b 2
DR

d) Show that the scattering angle Θ is given by

α
tan(Θ/2) = (A.1)
2bT
where b is the impact parameter (the closest distance of the projectile to the nucleus
p2
if it were to be undeflected) and T = 2m is the initial kinetic energy.

e) Calculate the Rutherford scattering cross section σ for scattering of projectiles

by angles greater than Θmin .

f) Show that the differential cross section

dσ 1  α 2 1
= 4
dΩ 16 T sin (Θ/2)

where T is the kinetic energy of the particle.

Why is it not possible to calculate the total cross section for this reaction?
The cross section for an interaction i is given by:
Wi
σi =
J
where Wi is the rate the interaction i and J is the incoming flux density of projectiles.

66
Examples

The differential cross section for the same reaction

dσi 1 dWi
=
dΩ J dΩ
where the second term is the rate at which scattered particles impinge on some solid
angle dΩ.

a) Conservation of energy gives p = p0 , after which |∆p| = 2|p| sin(Θ/2) by geometry.

b) Conservation of angular momentum gives mbv0 = j = mr2 dθ

dt
.

c) Newton II gives Z
α
|∆p| = dt cos θ
.
r2
We can use the angular momentum relationship to rephrase the integral as
Z π−Θ
α 2 2
|∆p| = dθ cos θ
bv0 −π +Θ
2 2
 
2α Θ
= cos
v0 b 2

d) Equating the two values obtained for |∆p| gives

 
Θ α
tan = .
2 2bT
FT

e) Θ decreases monotonically with b so σ(Θ > Θ0 ) = σ(b < b0 ) = πb20 with

A

 
α Θ
DR

b0 = cot .
2T 2

f) dΩ = sin Θ dΘ dφ and from geometry dσ = b dφ db, so

dσ b db
= −
dΩ sin Θ dΘ
1  α 2 1
=
16 T sin4 (Θ/2)
db
where in the last step we substitute − dΘ = α
4T
cosec2 (Θ/2) and 1
sin Θ
=
1/(2 sin(Θ/2) cos(Θ/2).

The total cross section diverges as b → ∞, or equivalently as Θ → 0.

3+
2.2. The J P = 2 decuplet contains the following baryons:

∆− ∆0 ∆+ ∆++
Σ∗− Σ∗0 Σ∗+
Ξ∗− Ξ∗0
Ω−

What is the quark content of each of these baryons?

67
Examples

The constituent quarks have no relative orbital angular momentum. How does the
Ω− baryon state behave under exchange of any pair of quarks? Explain how this is
achieved in terms of the space, spin, flavour and colour parts of the state vector.

1+
Account for the absence of sss, ddd and uuu states in the J P = 2 octet.

What is meant by quark confinement?

Quarks only found in hadrons (qqq, baryons or q q̄ mesons) which are colour singlet
states. Existence of sub-structure is inferred from (e.g.)

• structures in hadron multiplets

• high-energy probes scattering off point-like objects (deep inelastic scattering)
• energy levels in quarkonium
• ratio of cross sections of e+ e− to hadrons compared to µ+ µ−.

The isospin mutiplets have quark content

∆ = (ddd, ddu, duu, uuu)

Σ∗ = (dds, uds, uus)
Ξ∗ = (dss, uss)
Ω = (sss)
FT

The Ω state vector can be written as a direct product

A

|Ωi = |Ψspace i|Ψspin i|Ψflavour i|Ψcolour i.

DR

The flavour, spin and space parts of the |Ωi are symmetric under exchange of any
two quark labels. |Ψispace is symmetric because all are in spherically symmetric L =
0 states. The possible values of |Ψispin with different mS can found by successive
application of S− = S1− + S2− + S3− to | ↑1 ↑2 ↑3 i, so all possible values of |Ψispin are
symmetric under exchange of any labels. The colour part is the totally antisymmetric
combination  
|r1 i |g1 i |b1 i
|ψc i =  |r2 i |g2 i |b2 i 
|r3 i |g3 i |b3 i
so the combination satisfies the required antisymmetry under exchange of labels of any
pair of identical fermions.
+
For J P = 21 we still require antisymmetry under exchange of any two quarks. Since
again the product |Ψspace i|Ψflavour i|Ψcolour i is S × S × A/S under any exchange, we
would require the spin part to be symmetric under any exchange, which isn’t possible
for total spin J = 12 .

2.3. Write down the valence quark content for each of the different particles in
the reactions below and check that the conservation laws of electric charge, flavour,
strangeness and baryon number are satisfied throught.

(1) π− + p → K0 + Λ
(2) K− + p → K 0 + Ξ0
(3) Ξ− + p → Λ+Λ
(4) K− + p → K + + K 0 + Ω−

68
Examples

Draw a quark flow diagram for the last reaction.

By the magic of the quark model, if we check conservation of flavour, then conservation
of electric charge, strangeness and baryon number follow automatically.

(1) π− + p → K0 + Λ
dū + uud ds̄ + uds
(2) K− + p → K 0 + Ξ0
sū + udd ds̄ + uss
(3) Ξ− + p → Λ+Λ
dss + uud uds + uds
(4) K− + p → K + + K 0 + Ω−
sū + uud us̄ + ds̄ + sss

FT

2.4. Consider the decay of the ρ0 meson (J P = 1− ) in the following decay modes:
A
DR

a) ρ0 → π 0 + γ

b) ρ0 → π + + π −

c) ρ0 → π 0 + π 0

For case (b) and (c), draw a diagram to show the quark flow.

Consider the symmetry of the wave-function required for π 0 + π 0 and explain why
this decay mode is forbidden.

From consideration of the relative strength of the different fundamental forces,

determine which of the other two decay modes will dominate.

Pions are spin-0 so to conserve angular momentum π 0 π 0 would have to be in a L = 1

state, which has negative parity. However π 0 π 0 is a state of identical bosons so must be
symmetric under exchange of labels. But in the CM frame the operation of exchange
of labels is equivalent to the parity operation. It is not possible to write down an L = 1
state which is symmetric under interchange of the identical pions.

The strong decay ρ0 → π + π − dominates. The electromagnetic decay contains of factor

of α relative to the strong decay. In fact the branching ratio is 6 × 10−4 , meaning that
the form factor must also be important.

69
Examples

2.5. A wave function is modelled as the sum of the incoming plane wave and an
outgoing (scattered) spherical wave,

eikr
 
(+) ik·x 0
hx|Ψ i=A e + f (k , k) .
r

Calculate the flux associated with the plane wave and the spherical wave separately.
Hence show that the cross section into solid angle dΩ is

dσ 2
= |f (k0 , k)| .
dΩ

justifying any assumptions you make.

[Hint: Remember that the flux is given by ~

2mi
(ψ ∗ ∇ψ − ψ∇ψ ∗ ).]

The cross section is defined by

W = σnJ.
For an outgoing spherical wave, the rate of particles making their way into solid angle
dΩ is
WdΩ = Jo r2 dΩ
where Jo is the outgoing flux at radius r.
FT

We ignore the interference terms since while the incoming wave is approximately a plane
wave close to the scattering center, but can be assumed to be negligible sufficiently far
A

from the beam.

DR

For the incoming wave eik·x the flux is

~k
Ji = |A|2 .
m

∂
For the outgoing flux, ∇r = ∂r
. The derivatives which hit the 1/r terms cancel, so

~k
Jo = |A|2 |f |2 .
mr2
The contribution to the cross section from a small part of solid angle dΩ is the ratio
of the outgoing rate of particles scattered into dΩ to the incident flux. The area over
which the outgoing flux should be integrated to get the rate is r2 dΩ, so

Jo (r2 dΩ)
dσ = = |f |2 dΩ .
Ji

2.6. In the Born approximation, the scattering amplitude is given by

1
f (1) (k0 , k) = − (2m)(2π)3 hk0 |V |ki
4π

Explain the terms in this equation, and state the conditions for which it is valid.

70
Examples

b) The Yukawa potential is given by

g 2 e−µr
V (r) = .
4π r
Show that for this potential
g2 1 e−µr
Z
hk0 |V |ki = d3 x e−i∆k·x
4π (2π)3 r
where ∆k = k0 − k.

c) Hence show that

g2 1
|hk0 |V |ki| =
(2π) q + µ2
3 2

where q = |∆k|.

d) Show that within the Born approximation

dσ g4 4m2
= (A.2)
dΩ (4π)2 [2k 2 (1 − cos Θ) + µ2 ]2
where Θ is the scattering angle.

e) Find the total cross section for the case when µ 6= 0.

f) When µ → 0, V (r) ∝ 1/r. Compare the differential cross section (A.2) to the
FT

classical Rutherford scattering cross section

A

dσ 1  α 2 1
DR

= .
dΩ 16 T sin4 (Θ/2)
What must be the relationship between g and α for Born approximation to reproduce
the classical Rutherford formula for electron–proton scattering?
a) In the Born approximation the interaction is modelled as a single interaction of the
incoming wave with the potential. This is valid provided that the potential is sufficiently
weak.
k is the incoming wave-vector. k0 is a wave-vector with the same magnitude as k
pointing towards the detector. m is the mass of the projectile (or the reduced mass for
a two-body system). V is the scattering potential. b) We need the amplitude for the
states Z Z
hk0 |V |ki = d3 x d3 x0 hk0 |x0 ihx0 |V |xihx|ki

The conventionally-normalised wave-functions for the states with definite momentum

are
3
hx|ki = (2π)− 2 e+ik·x
0
3 ·x0
hk0 |x0 i = (2π)− 2 e−ik .

and the potential is local so that

hx0 |V |xi = δ 3 (x − x0 )V (x0 ).

Using the delta function to do the x0 integral

g2 1 e−µr
Z
hk0 |V |ki = 3
d3 x e−i∆k·x
4π (2π) r

71
Examples

where ∆k = k0 − k.

c) In spherical polars, with θ relative to the q direction the integral is given by

2π 1 ∞
r2 −iqr cos θ −µr
Z Z Z
I = dφ d(cos θ) dr e e
0 −1 0 r
Z ∞
1 h −iqr cos θ −µr i+1
= 2π dr re e
0 −iq cos θ=−1
Z ∞
2π  
= dr e−iqr−µr − e+iqr−µr
−iq 0
∞
2π e−iqr−µr eiqr−µr

= −
−iq −iq − µ iq − µ r=0
 
2π −1 −1
= −
−iq −iq − µ iq − µ
2π (iq − µ) − (−iq − µ)
=
+iq (−iq − µ)(iq − µ)
 
2π 2iq
=
iq q 2 + µ2
4π
=
µ2 + q 2

Hence
g2 1
hk0 |V |ki = .
(2π)3 µ2 + ∆k2
FT

d) The scattering amplitude is given by

A

1
f (1) (k0 , k) (2m)(2π)3 hk0 |V |ki
DR

= −
4π
g2 2m
= − .
4π µ2 + ∆k2

The differential cross section is then

2
g 2 /(4π)

dσ
= |f (1) (k0 , k)|2 = (2m)2 .
dΩ µ2 + ∆k2

From geometry
∆k2 = (2k sin(Θ/2))2

dσ
R
e) The cross section is σ = dΩ dΩ
.
2 π
g2
 Z
sin Θ
σ = (2m)2 2π dΘ
2 .
4π 0 µ2 + 4k2 sin2 (Θ/2)

The integral can be written

Z π
2 sin(Θ/2) cos(Θ/2)
I = dΘ  2
0 µ2 + 4k2 sin2 (Θ/2)
π
1/(2k)2

=
µ2 + 4k2 sin2 (Θ/2) 0
2
= .
4k2 µ2 + µ4

72
Examples

Putting this together we get

m2 g 4
σ= .
π(4k2 µ2 + µ4 )

f) For massless intermediate particles (such as photons) the total cross section σ → ∞.
The differential cross section
2
g 2 /(4π)

dσ
= (2m)2
dΩ 4k2 sin2 (Θ/2)
 2 2
1 g 1
=
16T 2 4π sin4 (Θ/2)

Comparing this to the classical cross section for electrostatic scattering we find that

e2 g2
αe = = EM
4π0 4π
meaning that gEM is a dimensionless measure of the electric charge.

2.7. Consider the scattering of an electron from a nucleus R with extended spherical
charge density N (|x00 |) which is normalised such that d3 x00 N (|x00 |) = 1. The
potential at any point is then
N (|x00 |)
Z
V (x ) = zZα d3 x00 0
0
,
|x − x00 |
FT

where Z and z are the charges of the nucleus and the projectile respectively.
A
DR

a) By expanding in the position basis, and defining a new variable X = x0 − x00 ,

show that the Born approximation to the scattering amplitude can be expressed in
the form
f (k0 , k) = f (k0 , k)point × Fnucl (∆k)
where f (k0 , k)point is the scattering amplitude from a point charge,the nuclear form
factor Z
0
Fnucl (∆k) = d3 x0 e−i∆k·x N (|x0 |)

is the 3D Fourier transform of the charge density distribution, and ∆k is the change
in momentum of the projectile.

Hence show that  

dσ dσ
= |Fnucl (∆k)|2 .
dΩ dΩ Rutherford

b) Consider the example form factors in Figure A.1 (i) and (ii). How would these
form factors scale along the ∆k-axis if the radius r of the corresponding sphere of
charge was doubled? By relating the momentum transfer to the scattering angle,
use the data to estimate the size of the silver nucleus. Compare to the expectation
1
for an incompressible nucleus, r = r0 A 3 with r0 = 1.25 fm.

c) How might one accelerate protons to kinetic energy of 17 MeV, and subsequently
detect the scattered protons experimentally?

73
Examples

NHraL
1.0

0.8

0.6

0.4
FT

0.2
A
DR

0.5 1.0 1.5 2.0

ra

ÈFHÈDkÈaLÈ
(i)
2

10

0.1

0.001

0 2 4 6 8 10
ÈDkÈa
(ii)

Figure A.1: (LHS) Scattering cross section for K.E. = 17 MeV protons normalized
to the Rutherford scattering cross section (from [2]).
(RHS) (i) Normalised charge density distribution ρ(r/a) ∝ [1+exp((r/a−1)/δ)]−1
with R = 1, δ = 0.2 (This is the so-called Saxon-Woods form [6]) (ii) The cor-
responding nuclear form factor |F (|∆k|a)|2 . The constant a is the approximate
radius of the nucleus.

74
Examples

a) Expanding in the position basis,

Z Z
1 0 0 Zα 0
hk0 |V |ki = d 3 0
x d3 x00 e−ik ·x 0 N (x00 )e+ik·x .
(2π)3 |x − x00 |

Defining a new variable X = x0 − x00 and defining ∆k = k0 − k we have that

e−i∆k·X
Z  Z 
00 Zα
hk0 |V |ki = d3 x00 N (x00 )e−i∆k·x × d3
X
(2π)3 |X|
= Fnucl (∆k) × hk0 |V |kiRuth .

The differential cross section ratio is then

|hk0 |V |ki|2
   
dσ . dσ
= = |Fnucl (|∆k|)|2 .
dΩ dΩ Ruth |hk0 |V |kiRuth |2

b) The form factor calculation depends on k · r, so if one doubles the radius of the
sphere, the form factor distribution is squeezed into a factor of two smaller in k.
p √
The momentum of the projectile is p = 2mp Ep = 2 × 938 × 17 ≈ 179 MeV/c.
From the plot the short scale structure shows minima for Ag about every 40◦ between
which the momentum transfer |∆k| = 2p sin Θ/2 ≈ 122 MeV/c. From the comparison
plots, |F (|∆k|)|2 has minima around |∆k|r0 ≈ 4 so

4~c 4 × 197 MeV fm

r0 ≈ = = 6.5 fm.
|∆k|c 122 MeV

This rough estimate compares with our expectation 1.25 fm × (108)1/3 ≈ 5.95 fm.
A FT

c) A 17 MeV proton beam can be generated from a cyclotron, or by a linac. The

DR

scattered protons could be detected with (for example) a crystal scintillators connected
to photomultiplier tubes.

2.8. The cross section for the production of γ-rays by neutrons incident on a
certain nucleus (N, Z) is dominated by a resonance and given by the Breit-Wigner
formula,
π Γn Γγ
σ(n, γ) = 2 . (A.3)
k (E − E0 )2 + Γ2 /4
a) Define the symbols in this formula and explain the physical principles that underlie
it, and the conditions under which it applies.

b) What are the mass and lifetime of the resonant state?

c) All spin effects have been ignored in (A.3). How would the formula differ if spins
are included?

d) On the same plot draw how the inelastic (n, γ) and elastic (n, n) cross sections
would behave close to the resonance, when Γγ = 4Γn , labeling important quantities
including the peak cross section values.

[In (d) you may assume that resonant scattering dominates both cross-sections, and that
decays other than to n and γ are negligible.]

75
Examples

a) σ is the cross section, W/j. k is the wave-number.

P E is the total energy of the
system. E0 is the resonance energy. 1/τ = Γ = Γi = FWHM provided Γ  E0 .
Applies for scattering dominated by transitions through a long-lived intermediate state
A → B ∗ → C so that the width is small compared to the mass and also small compared
to the energy spacing between resonances.

b) mR = E02 /c2 , τ = ~/Γ

c) When spins are included one must include for a + b → R → b + c a factor

2JR + 1
g=
(2Ja + 1)(2Jb + 1)
which gives the probability that a + b collide in the correct spin state to form the
resonance.

d) Relative heights should be in correct proportion. Labels should include E0 , FWHM=

Γ. The maximal cross sections are
π
σmax (n, n) = 4g(Γn /Γ)2
k2
with Γn /Γ = 1/5. The (n, γ) cross section is 4 times larger.

2.9. The cross section for the reaction π − p → π 0 n shows a prominent peak
when measured as a function of the π − energy. The peak corresponds to the ∆
resonance which has a mass of 1232 MeV, with Γ = 120 MeV. The partial widths for
FT

the incoming and outgoing states are Γi = 40 MeV, and Γf = 80 MeV respectively
for this reaction.
A
DR

At what pion beam energy will the cross section be maximal for a stationary proton?

Describe and explain the similarities and differences you would expect between the
cross section for π − p → π 0 n and the one for π − p → π − p, for centre-of-mass
energies not far from 1.2 GeV. Giving values for the variables in the Breit-Wigner
formula where possible. Use quark-flow diagrams to explain what is happening.

By considering the quark content of the intermediate states, discuss whether you
would expect similar peaks in the cross sections for the reactions (a) K − p →
products and (b) K + p → products.
Using the partial widths for the ∆0 resonance from above, the reaction π − p → π − p
has the same FWHM (since it has the same Γ), same E0 and half the cross section as
π − p → π 0 n.

At resonance the CM energy must be such that

m2∆ = (Pπ + Pp )2 = m2π + m2p + 2Eπ mp

So
m2∆ − m2π − m2p
Eπ = = 329.7 MeV.
2mp
Thus the kinetic energy of the pion must be 190.1 MeV.

K + +p has quark content (us̄)+(uud) which fits no baryon. K − +p has quark content
(sū) + (uud) = (uds) for which one would expect to find Λ0 and Σ0 resonances.

76
Examples

2.10. Consider the Hamiltonian H = H0 + V , where H0 is the free-particle Hamil-

tonian, and V is some localised potential. Let the ket |φi represent an eigenstate
of H0 , and the ket |ψi represent an eigenstate of H which shares the same energy
eigenvalue as the |φi in the limit  → 0.

Show that the Lippmann-Schwinger equation

1
|ψ (±) i = |φi + V |ψ (±) i.
E − H0 ± i

is consistent with the states defined above by multiplying it by the operator (E −

H0 ± i) and taking the limit  → 0.

After multiplying and  → 0 we find (E − H0 − V )|ψi = (E − H0 )|φi, both of which

are null kets by definition.

2.11. Show that the Green’s function

~2 D 1 E
G± (x, x0 ) ≡ x x0
2m E − H0 ± i

can be written
∞
i eiq∆ − e−iq∆
Z
G± (x, x0 ) = dq q (A.4)
4π 2 ∆ 0 q 2 − k 2 ∓ iδ
~2 k 2
FT

where  and δ are infinitesimals, E = 2m and ∆ = |x − x0 |.

A

[Hint: start by inserting identity operators d3 p0 |p0 ihp0 | and d3 p00 |p00 ihp00 | on each side
R R
DR

of the operator, and changing Ĥ0 →p̂2 /2m.]

After inserting the completeness relations we have to calculate the amplitude

D 1 E δ 3 (p0 − p00 )
p0 p00 =
E− (p̂2 /2m) ± i E − (p00 2 /2m) ± i

The δ 3 allows us to do one of the momentum integrals, leaving

0
eiq·(x−x )
Z
1
G± = d3 q
(2π)3 k − q 2 ± i
2

~2 k 2
where E = 2m
and p0 = ~q. If we write this out in polar coordinates
∞ 2π 1
eiq∆ cos θ
Z Z Z
1
dq dφ d(cos θ)q 2 ,
(2π)3 0 0 −1 k2 − q 2 ± i

then doing angular integrals we get the desired result.

2.12. If you have done the course on functions of a complex variable, finish the
integral in (A.4) using appropriate contour integrals to obtain

1 e±ik∆
G± (x, x0 ) = − .
4π ∆

77
Examples

The integral from 0 to ∞ is half the integral along the whole real axis
Z ∞
i eiq∆ − e−iq∆
G± (x, x0 ) = 2
dq q 2 .
8π ∆ −∞ q − k2 ∓ i

Since  is a vanishingly small number we can rewrite the denominator

(q 2 − k2 ∓ i) = (q 2 − [k 1 ± i/k2 ]2 ) ≈ (q 2 − [k ± i0 ]2 ).

p

The poles in the complex q-plane are therefore at

q= ±(k + i0 ) (G+ )

q= ±(k − i0 ) (G− ).

Do G+ first. Closing the eiq∆ integral in the upper half-plane we pick up a pole for
ik∆
G+ at q = k + i0 with residue ke2k . The e−iq∆ integral when closed in the lower
half-plane gives an equal contribution. Cauchy then tells us that
 
i k ik∆
G+ (x, x0 ) = 2 2 (2πi) e
8π ∆ 2k
1 eik∆
= − .
4π ∆
We can get G− by making the replacement k 7→ −k.

2.13. The pions can be represented in an isospin triplet (I = 1) while the nucleons
form an isospin doublet (I = 12 ),
FT

π+
 
A

 
 π0  p
and
DR

n
π−

while the ∆ series of resonances have I = 32 .

By assuming that the isospin operators I, I3 , I± obey the same alegbra as the
quantum mechanical angular momentum operators J, Jz , J± , explain why the
ratio of Γi /Γf ≈ 12 was found in a previous question.

[Hint: you will need the Clebsch-Gordon coefficients for hj1 j2 m1 m2 |j1 j2 JM i for J = 32 ,
j1 = 1, j2 = 21 , M = − 21 .]

Isospin values for the dramatis personae are

Particle π− p
π0 n ∆0

1 1 1
, − 12 3
, − 12


(I, I3 ) (1, −1) ,
2 2
(1, 0) 2 2

By analogy with the angular momentum case we shall need the C.G. coefficients for

h j1 = 1, j2 = 21 , m1 , m2 | J = 32 , M = − 12 , j1 , j2 i
q
Where for (π − , p) ⇒ (m1 = −1, m2 = 21 ) the C.G. coefficient is 1
3
while for
q
(π 0 , n) ⇒ (m1 = 0, m2 = − 12 ) it is 2
3
. The matrix element for the decay contains

78
Examples

these coefficients for each decay type, so the rate contains the mod-squared of the
above, hence q 2
1
Γ(∆0 → p π − ) 3 1
= q 2 = .
Γ(∆0 → n π 0 ) 2 2
3

2.14. A proton is travelling through a material and scattering the electrons in the
material.

a) Express the scattering angle in terms of the impact parameter b, the reduced
mass µ, the relative speed v, and the scattering angle in the ZMF. Hence show that
the momentum transfer is
2µv
q=√ .
1 + z2

where z = bµv 2 /α.

b) Write down the energy given to an electron for a collision for a given impact
parameter b. Integrate this up with area element 2πb db to show that the average
energy lost by the projectile per distance travelled is

zmax
4πne α2
 
dE z dz
Z
− = ,
FT

dx me v 2 zmin 1 + z2
A

where ne is the number density of electrons.

DR

[Hint: recycle results from the Rutherford scattering question.]

a) From the question on Rutherford scattering, for a two-body system in the ZMF
frame, we now have v as the relative velocity and µ ≈ me replacing m. Thus we
change p to me v and T to 21 me v 2 . The angular relationships are then

α 1
tan(Θ∗ /2) = =
bµv 2 z
and
q
sin(Θ∗ /2) = .
2µv
where µ ≈ me is the reduced mass. Eliminate Θ∗ using cot2 +1 = cosec2 ,
 2
2µv
z2 + 1 = ,
q

leaving
2µv
q= √ .
1 + z2

b) The energy loss by the proton per collision in the lab frame is the energy gained by
the electron which is
q2
Ee = .
2me

79
Examples

The average energy loss in a material with electron number density ne is then
 
dE
− = ne hσ ∆Ei
dx
Z bmax
= ne db (2πb)Ee (b)
bmin
Z bmax
b db
= 4πne me v 2
bmin 1 + z2
zmax
4πne α2
Z
z dz
= .
me v 2 zmin 1 + z2

It’s easy to do the integral. The limits can be found from the points where Tmin is of
the order of the ionisation energy (zmax ); and at Tmax when relativistic effects become
important (zmin ).

2.15. a) A projectile travels through a medium of thickness x with n targets per

unit volume. Show that the fraction absorbed or deflected by the medium is
Pabsorb (x) = 1 − e−nσx ,
where σ is the absorption cross section.

b) Estimate, stating any assumptions you make, the thickness of lead that would
be required to have a 50% chance of stopping a 2.3 MeV neutrino coming from a
solar nuclear fusion reaction.
FT

The cross section for the scattering of a neutrino from a stationary target is approx-
A

imately
DR

4π p2CM dpCM
σtot = 2πG2F
(2π)3 dECM
where ECM is the centre-of-mass energy of the system, and pCM is the momentum
of the neutrino in the centre-of-mass frame.

c) Justify the form of this expression.

d) Explain how Figure A.2 supports a model in which the proton contains point-like
constituents.

[The density of lead is about 11.3 g cm−3 . Some data for the cross section of neutrinos ν
scattering from nucleons – meaning protons or neutrons – are shown in Figure A.2.]
a) The flux j(x) will vary with distance since particles are absorbed. For a thin slice of
material with area A at distance x and thickness δx the number of targets is Aδx and
the fraction of the flux absorbed is
δj Wabsorb n(A δx) σj
− = = = nσ δx.
j Warrive jA
Integrating, j(x) = j(0)e−nσx . σT /Eν ≈ 0.7 × 10−38 cm2 / GeV, so at 2.3 MeV the
cross section is about 1.6 × 10−41 cm2 .

b) We require j/j0 = 12 , so σnx = ln 2. Let’s assume the scattering of each nucleon

is incoherent. The number density of nucleons in lead is
11.3 g cm−3
n = ρ/mN = = 6.7 × 1024 cm−3 .
1.67 × 10−24 g

80
Examples

0 10 20 30 50 100 150 200 250 300 350

1.0 1.0

0.8 0.8

]VeG/ mc

0.6 0.6
2
83Ð
01[
nE/Ts

0.4 0.4

0.2 _ 0.2
[1] NuTeV [5] CDHSW [9] GGM-PS n [13] CRS

[2] CCFR (96) [6] GGM-SPS [10] IHEP-JINR [14] ANL

[3] CCFR (90) [7] BEBC WBB [11] IHEP-ITEP [15] BNL-7ft

[4] CCFRR [8] GGM-PS n [12] SKAT [16] CHARM

0.0 0.0
0 10 20 30 50 100 150 200 250 300 350

En [GeV]

Figure A.2: σtot /Eν for neutrino-nucleon νN (and anti-neutrino–nucleon ν̄N )

interactions as a function of neutrino energy. From [4].
FT

The distance x is then

A

ln 2 0.69
= 6.3 × 1015 cm,
DR

=
nσ 6.7 1024 × 1.6 × 10−41
which is about 2.5 light-days.

We have extrapolated over many orders of magnitude. In fact the cross section falls
rather faster than ∝ Eν at ∼ MeV energy levels due to nuclear shell effects inhibiting
scattering.

c) Leaving aside the relativistic complications: 2π from FGR. Fermi matrix element
GF is squared in rate. Flux factor is unity for particles travelling close to c. Density of
d3 p 4πp2 dp
states for two-body final state is (2π) 3 ≈ (2π)3 dE , measured in CM frame. Subsituting

numbers into the formula gives a prediction of 3.9 × 10−39 cm2 / GeV, which is of the
right order, even based on this simplified calculation.

d) Leaving constants aside, we consider Pν = (Eν , Eν , 0, 0) and Pq = x(mp , 0, 0, 0),

where x is the fraction of the momentum of the proton carried by the quark. For
Eν  mp the centre-of-mass energy squared is
s = m2p + 2x mp Eν ≈ 2xmp Eν
and momentum of the neutrino in the centre-of-mass is pCM ≈ ECM /2. We then have
that σ ∝ E, as found in the figure.

The absence of any other structure (the trivial form factor F (q 2 ) ≈ 1) across the
approximate energy range 3 GeV2 < s < 300 GeV2 shows that the targets are point-
~c
like on length-scales of order √s
. The absence of structure is observed in the length
range 10−17 m→10−16 m. If the quarks are not point-like then their typical sizes must
therefore be smaller than 10−17 m.

81
Examples

In fact scattering of quarks from proton-proton interactions at the LHC further con-
strains any quark substructure to scales smaller than ~c/TeV ∼ 10−19 m.

2.16. The figure shows the fraction N/N0 transmitted when protons of kinetic
energy E = 140 MeV impinge as a collimated beam on sheets of copper of various
thicknesses x. By considering the 2-body kinematics of proton–electron collisions
and proton–nucleus collisions, account for the attenuation for values of x between 0
and 20 mm, and for the sudden change in behaviour around the value of x marked
R(E).

A FT
DR

Results similar to the figure were obtained for protons of E = 100 MeV, except that
in this case a value of R(E) = 14 mm was obtained. Offer a brief explanation for
the change in R(E). What is the relative size of the nuclear scattering cross section
σNucl in copper compared to the geometric cross section?

[The density of copper is 8.9 g cm−3 , and it has relative atomic mass 63.5. You may
assume that the nuclear radius is given by r = r0 A1/3 with r0 = 1.25 fm ]

Scattering from electrons will lead to gradual energy loss along path
 
dE
− ≈ 40 MeV/10 mm.
dx

When all the energy is used up the flux stops. Some smearing at the edge since energy
loss is a stochastic process.

Scattering from nuclei will lead to large change in momentum of the proton but with low
probability (low x part of graph). From the graph the gradient at low x is δN/N0 ≈ 0.05
in 10 mm so
−δn 1
σ =
n nl
−δn mCu
=
n ρσ
63.5 × 1.66 × 10−27 kg
= 0.05 ×
(8.9 g cm−3 )(10 mm)
= 0.6 barn.

82
Examples

The geometric cross section is

2
σgeom = πr02 A2/3 = π(1.25 fm2 )(64) 3 = 0.79 barn.

A FT
DR

83
Examples

A.3 Relativistic scattering

Solutions

3.1. The Klein-Gordon equation,

 2 
∂ 2 2
− ∇ + m ϕ(r, t) = 0. (A.5)
∂t2
is the relativistic wave equation for spin-0 particles.

a) Show that
e−µr
ϕ=
r
is a valid static-field solution.

b) Show that another possible solution to the Klein-Gordon equation is:

φ(X) = A exp [iP · X]
Where P and X are the momentum and position four-vectors respectively. What
restrictions does (A.5) place on the components of P?

What are the physical interpretation of these solutions?

1 ∂
r2 ∂r
∂


The radial part of the Laplacian in spherical polars is r 2 ∂r
so
FT

e−µr e−µr
  
2 1 ∂ 2
A

∇ ϕ = r −µ − 2
r2 ∂r r r
DR

1 ∂ 
(−µr − 1)e−µr

=
r2 ∂r
1 
−µe−µr − µ(−µr − 1)e−µr

= 2
r
= µ2 e−µr = µ2 ϕ.

The equation shows a possible scalar field configuration for particles of mass µ with
a static source at the origin. For example the field of virtual pions around a nucleus.
These are virtual particles which only travel a short distance. Though we have not
calculated it here, these virtual particles do not satisfy the usual energy–momentum
invariant: E 2 − |p|2 6= µ2 for these guys.

Expanding out the dot product

φ(X) = Aei(Et−p·x)
2 2 2 2
d
so dt 2 φ = −E φ, and similarly ∇ φ = −|p| φ. Therefore the K-G equation is satisfied
2 2 2
provided E − |p| = m . This is the relativistic propagation of a field containing real
(i.e. non-virtual) particles of mass m, satisfying the usual energy–momentum invariant.

3.2. Draw all the tree-level electromagnetic Feynman diagram(s) for the following
processes:

a) e− + e+ −→ e− + e+

84
Examples

Two diagrams: one for each of t-channel and s-channel photon

b) e− + e− −→ e− + e−

2 diagrams both with t-channel photon exchange, one with outgoing legs swapped.

c) e− + e− −→ e− + e− + µ+ + µ−

Electrons exchange a photon. Muon pair created from a virtual photon emitted from
any of the charged legs (so 4 of them). Diagrams also exist where electrons are swapped.

d) γ −→ e+ e− in the presence of matter

2 diagrams. Virtual photon from Z attached to either e+ leg or e− leg.

e) γ + γ −→ γ + γ

Requires an electron loop. Three different topologies:

A FT
DR

Two different directions for the electron to flow around each of those loops. Other
particles may also enter in the loop. . .

3.3. Why does the ratio

σ(e+ + e− → µ+ + µ− )
σ(e+ + e− → τ + + τ − )

tend to unity at high energies? Would you expect the same to be true for

σ(e+ + e− → µ+ + µ− )
?
σ(e+ + e− → e+ + e− )

Well above threshold the density-of-states factors become the same. The µ and τ
production can progress through the s-channel photon channel so have very similar
cross sections. The electron scattering cross section also has a t-channel diagram so
has a different cross section.

85
Examples

-2
10

-3
10

-4
10

-5
σ [mb]

10

-6
10

-7
10

-8
10
2
1 10 10

3
10

2
10

R
10
FT

1
A

-1
DR

10
2
1 √ 10 10
s [GeV]

Figure A.1: The cross section σ(e+ e− → hadrons) and the ratio of cross sections
+ −
e → hadrons) √
R = σ(e
σ(e+ e− → µ+ µ− ) as a function of the center of mass energy s.

3.4. Draw leading order electromagnetic Feynman diagrams for the processes

e+ + e− → µ+ + µ− and e+ + e− → q + q̄.

How do the vertex and propagator factors compare?

Figure A.1 shows the ratio of the cross sections for the process of electron–positron
annihilation to hadrons, and√the corresponding cross section to the muon–antimuon
final state as a function of s, the centre-of-mass energy.

Considering the number of quarks that can be created at particular centre-of-mass

√ values of R would you expect for centre-of-mass energy in the range
energy, what
2 GeV < s < 20 GeV? How do your predictions match the data? How do these
measurements support the existence of quark colour?

What is causing the sharp peaks in σ and R at centre-of-mass energy of ≈ 3 GeV,

10 GeV, and 100 GeV?

86
Examples

From the Feynman diagrams, the ratio is expected to be

X 2
Nc qi

where there are three colours and the sum is over the charges of the kinematically
accessible quark states.

Below 3 GeV only u, d, s are accessible and so

 2
12 12

2
R=3 + + =2
3 3 3

When cc̄ becomes accessible above about 3 GeV we find

 2
12 12 22

2 10
R=3 + + + =
3 3 3 3 3
2 √
and when a further 3 × 13 is added for s > 10 GeV we get R = 11
3
, all of which
predictions (green lines on the plot) match the data rather well.

The peaks are for cc̄, bb̄ bound state resonances at 3 GeV and 10 GeV, and for Z 0 at
92 GeV. The lower mass peaks come from the ω and φ mesons.

3.5. At the HERA collider 27 GeV positrons collided with 920 GeV protons. Why
can these collisions can be considered to be due to positrons scattering off the
quarks in the protons?
FT

For these collisions draw one example of a Feynman diagram for each of the cases
A

of weak charged-current, weak neutral-current and electromagnetic interaction.

DR

Calculate the center-of-mass energy of the quark–positron system assuming that

the 4-momentum of the quark Pq can be represented as a fixed fraction f of the
proton 4-momentum Pp , in the approximation where both particles are massless.

What is the highest-mass particle that can be produced in such a collision in the
approximation that a quark carries about 31 of the proton momentum?

We can represent the quark 4-momentum by

Pq = f Pp = f Ep (1, 0, 0, 1)

and the positron 4-momentum by

Pe = Ee (1, 0, 0, −1).

The CM energy is then:

2
ECM = (Pq + Pe )2 = P2q + P2e + 2Pq · Pe ≈ 0 + 0 + 4f Ep Ee ,

so p
ECM = 2 f Ep Ee .
1
Assuming that the quark carries 3
of the momentum of the proton the CM energy
q
of the positron–quark system is so ECM = 2 13 Ep Ee = 182 GeV which would be
sufficient to produce top quarks (via charge current interactions with virtual b quarks
in the proton sea).

87
Examples

How does the propagator for the weak charged current and electromagnetic inter-
actions vary with 4-momentum transfer P2 ? Hence explain the fact that at low
values of the momentum transfer it is found that the ratio of weak interactions to
electromagnetic interactions is very small whereas at very high values it is found
that the ratio is of the order of unity.
The positrons behave as if they scatter off individual quarks provided their de Broglie
wavelength is much smaller than the size of the proton. Here we have 27 GeV positrons
with resolution of order ~c/p ≈ 197 MeV fm/27GeV which is a factor of about 140
smaller than the radius of the proton. One ought to do the calculation in the ZMF,
but it is clear that the probe is going to be much smaller than the proton.
The propagators are
1 1
W : 2 γ: 2
PW − m2W Pγ
As with many texts we write the negative of the square of the momentum transfer
−P2 = Q2 .

At low energy where the momentum transfer −P2 = Q2  m2W the W propagator is
suppressed relative to the electromagnetic one by a factor ∼ Q2 /m2W . At large energies
when mW  Q2 the term m2W can be neglected and the relative cross sections are
determined by relative coupling constants which are of order unity.

3.6. The J/Ψ has mass 3097 MeV, width 87 keV and equal branching ratios of 6%
to e+ + e− and µ+ + µ− final states. What would you expect for these branching
ratios if the J/Ψ decayed only electromagnetically? What does this tell you about
the “strength” of the strong interaction in this decay? For comparison, the Ψ00 has
FT

mass 3770 MeV, width 24 MeV, but branching ratio to e+ + e− of 10−5 .

A

Draw diagrams for the decays D0 → K − + π + and D0 → K − + e+ + νe . Disre-

DR

garding the differences in the 2-body and 3-body density of states factors, what do
you expect for the relative rates of these decays?
If J/Ψ decayed electromagnetically only then the expected final states, and their relative
amplitudes and density-of-states factors are

uū dd¯ ss̄ e + e− µ+ µ−

2 2 2 1 2 1 2




Q 3 3 3
1 1
DoS 3 3 3 1 1
4 1 1
Ratio 3 3 3
1 1

The factors of three in the quark density of states come from colour factors.

1
The sum of the above is 4, so the expected BR to ee (or µµ) would be 4
= 25% for
pure electromagnetic decays.

If the strong force was unsuppressed, we would expect a branching ratio to leptons of
order
|gEM |4
|gS |4
from the vertex factors, which is of order 10−4 . This is close to the value observed for
the Ψ00 (3770).

However the observed branching ratio is 6%, which is smaller than the value of 25%
that we would expect in the absence of strong interactions, but larger than the value

88
Examples

of ∼ 10−4 we would expect if strong interactions are turned on at their full strength.
So we conclude that the strong interaction is present, but suppressed (known as ‘OZI’
suppression).

In the decay of the D0 a colour factor should lead to enhancement of a factor of NC

(three) in the density of states for the pion case. However since the quark and anti-
quark are in a colour-singlet object in the final state, it’s not a good approximation to
say that the quarks are free particles which can have any√colour. The amplitude to
find each colour-anticolour pair inside the pion will be
1/ 3, since the colour part of
the meson state vector is 1/sqrt3 |rr̄i + |gḡi + |bb̄i . Thus the net effect of colour is
√
to add a factor of NC from the d.o.s and a factor of 1/ 3 to the amplitude, hence to
multiply the rate by
√ 2
NC × 1/ NC = 1.
A FT

2
There is also a factor of Vud ≈ (0.972 ) from the CKM element in the decay to the
DR

pion, but this is fairly close to unity.

Looking at the data, the PDG gives

B(D0 →K − + π + ) = 3.88%
B(D0 →K − + e+ + νe ) = 3.55%

which is close to 1:1 as predicted.

3.7. By conserving momentum at each vertex in the centre-of-mass frame (or

otherwise) determine whether the propagator momentum is space-like (P2 < m2 )
or time-like (P2 > m2 ) for (i) e− + µ− → e− + µ− and for (ii) e+ e− → µ+ + µ− .
(i) For the elastic t-channel scattering in the CM frame no energy is exchanged, but
momentum is exchanged so |pγ | > 0, but Eγ = 0 so the propagator is space-like.
(ii) For the s-channel process in the CM frame pγ = 0, but Eγ > 0 so the propagator
is time-like.

3.8. Draw Feynman diagrams showing a significant decay mode of each of the
following particles:
a) π 0 meson
b) π + meson
c) µ−
d) τ − to a final state containing hadrons
e) K 0

89
Examples

f) top quark

Suggest:
a) π 0 → γ + γ
b) π + → µ+ + νµ
c) µ− → e− + ν̄e + νµ
d) τ − → π − + ντ
e) K 0 → π + + π −
f) t → b + W +

FT
A
DR

90
Examples

A.4 The Standard Model

Solutions

4.1. a) Draw Feynman diagrams for the the production of W ± bosons being
produced in a pp̄ collider. If the W + boson is close to its Breit-Wigner peak, what
possible decays may it have? (Which final states are kinematically accessible?)

b) What fraction of W + decays would you expect to produce positrons?

c) Suggest why the W was discovered in the leptonic rather than hadronic decay
channels.

d) How could the outgoing (anti-)electron momentum be determined? How might

the components of the neutrino momentum perpendicular to the beam be deter-
mined?
a) Production is dominantly via valence quarks

ud¯ → W + and dū → W −

b) Kinematically accessible decays for W + are:

Final state vertex factor colour factor

e+ ν̄e 1 1
FT

µ+ ν̄µ 1 1
τ + ν̄τ 1 1
A

cs̄ cos θc 3
DR

cd¯ sin θc 3
ud¯ cos θc 3
us̄ sin θc 3

where θc is the Cabibbo mixing angle. The fraction of decays to electrons (ignoring
leptonic tau decays) is then
1 1
=
1 + 1 + 1 + 3 cos2 θc + 3 sin2 θc + 3 cos2 θc + 3 sin2 θc 9

Some may realise that there is in principal also a decay mode to cb̄ but this is very
suppressed since Vcb is about 0.042 ∼ 10−3 .

c) The hadronic decays compete with scattering via the strong interaction which will
dominate.

d) The electron momentum can be determined from e.g. its bending in a magnetic
field
p⊥ = 0.3BR.
for p⊥ in GeV, B in Tesla and R in m.

The transverse components of the neutrino momenta can be determined from momen-
tum conservation in the directions perpendicular to the beam. The same cannot be
done for the longitudinal component since the longitudinal momenta of the partons
within the protons in the initial state are not known.

91
Examples

4.2. Write down Feynman diagrams for the decays of the muon and the tau
lepton. Are hadronic decays possible? By considering the propagator factor in each
case explain why one might expect on dimensional grounds that lifetimes should be
in the ratio
5
Γ(τ − → e− + ν + ν̄)

mτ
= .
Γ(µ− → e− + ν + ν̄) mµ

Using the following data

mτ = 1777.0 MeV ττ = 2.91 × 10−13 s

mµ = 105.66 MeV τµ = 2.197 × 10−6 s
BR(τ − → e− + ν + ν̄) = 17.8%

test this prediction.

As well as the two leptonic modes, the τ decays to a variety of hadronic modes e.g.

τ − → π − ντ
τ − → π − π 0 ντ
τ − → π − π + π − ντ

The propagator factor in each case is

1
FT

.
P2 − m2W
A

Since EW  mW the P2 term can be ignored so the partial rate must be proportional
DR

to m−4
W . By dimensional analysis, and since the mass of the decaying particle is the
only relevant scale in the problem, we again find Sargent’s rule

m5f
Γ∝
m4W

Our theoretical value for the τ lepton lifetime is

 m 5
µ
ττtheory = 0.178 τµ = 2.91 × 10−13 s
mτ
precisely matching the experimental value.

4.3. Which of the Standard Model fermions couple to the Z 0 boson? To which
final states may a Z 0 boson decay?

Explain why for the Z 0 the sum of the partial widths to the observed states (e+ e− ,
µ+ µ− , τ + τ − , hadrons) does not equal the FWHM of the Breit-Wigner.

By referring to the properties of the Breit-Wigner forumla, suggest how the LEP
e+ e− collider operating at centre-of-mass energies in the range 80 GeV to 100 GeV
could have inferred that there are three neutrino species with mν < mZ /2, even
though the detectors were unable to detect those neutrinos.

92
Examples

The Z 0 couples to all of the SM fermions (though the couplings are not equal).

The first four terms represent the visible final states, so

Γ = Γvis + Γν ν̄

If Γνi can be independently inferred (e.g. from the rate of neutral current neutrino
scattering) then the number of light neutrinos can be determined from
Γ − Γvis
nν =
Γνi

X
Γ = Γe+ e− + Γµ+ µ− + Γτ + τ − + Γqq̄ + Γνi ν̄i
i
σhad [nb]

2ν

3ν
30 ALEPH
DELPHI
4ν
L3
OPAL
20
average measurements,
error bars increased
by factor 10

10
A FT

0
86 88 90 92 94
DR

Ecm [GeV]
LEP’s Z 0 lineshape as measured and as predicted for different number
of light neutrino species. From [3].

4.4. Consider a model with two neutrino mass eigenstates ν2 and ν3 with masses
m2 and m3 and energies E2 and E3 , mixed so that

|νµ i = |ν2 i cos θ + |ν3 i sin θ

|ντ i = −|ν2 i sin θ + |ν3 i cos θ.

Consider a beam of neutrinos created from from π − → µ− ν decays. Show that the
observed flux of muon neutrinos observed at a distance L from such a source is
   
2 2 E3 − E2 L
J(L) = J(L = 0) × 1 − sin (2θ) sin .
2~ c

If m2 and m3 are much less than the neutrino momentum, |p|, show that
  
2 2
L
|νµ (L)| ≈ |νµ (0)| × 1 − sin2 (2θ) sin2 A m22 − m23 .
|p|

What is the first length L∗ at which the νµ detection rate is at a minimum?

If a range of neutrino energies are present what will be the ratio of the rate (per
neutrino) of ν + n → µ− + p for L  L∗ and L  L∗ .

93
Examples

What would be the corresponding ratio for neutral-current scattering?

Solar neutrinos emitted in p–p fusion have been detected via the processes

νe + d → p + p + e− and,
νx + d → p + n + νx .

Suggest why the charged-current reaction showed only a third of the neutrino flux
of the neutral-current reaction.

See notes for oscillation formula.

For a wide range of energies present and when L  L∗ , the term involving sin2 (. . . L)
will average to 21 , in the two-body case, and we will get a survival probability of
1 − 12 sin2 2θ.

For three-state maximal mixing we will similarly expect to find 13 of the neutrinos in
each flavour eigenstate. The neutral current measures all neutrino species, whereas the
charged current process (as described) is sensitive only to νe .

4.5. Draw all leading Feynman diagrams for the following processes:

a) νµ + n → p + µ−
FT

t-channel W exchange
A
DR

b) νµ + e− → νµ + e−

t-channel Z 0 exchange

c) ν̄e + e− → ν̄e + e−

Z 0 in t-channel; also W in s-channel.

d) ν̄e + p → e+ + n

t-channel W exchange

4.6. Write brief notes on:

a) The evidence that there are three and only three families of quarks and leptons.
b) The Cabibbo angle and quark mixing.
c) The evidence for confinement of quarks in hadrons.

94
Examples

4.7. How does helicity of a state change on application of the parity operator
P
(which reverses the coordinate axes: x 7→ −x)?

If |Φi is an eigenstate of the parity operator, what can be said about the parity of
the state (1 + a S · p)|Φi?

60
Co nuclei (J P = 5+ ) are polarised by immersing them at low temperature in
a magnetic field. When these nuclei β decay to 60 Ni (4+ ) more electrons are
emitted opposite to the aligning B field than along it1 . Explain carefully why this
demonstrates parity violation in the weak interaction.

Justify the direction of the parity-violating effect.

The helicity observable is

S·p
h=
|p|

Under the parity operator, the momentum and spin observables

P P
p 7→ −p S 7→ +S

so we must have that

P
h 7→ −h.
The eigenvalues of the helicity operator change must sign under P whereas those of
the unit operator are unchanged. The operators do not commute [P, 1 + aS · p] 6= 0.
so the state
FT

(1 + a S · p)|Φi
A

cannot be an eigenstate of parity for general a. It must be a mixed parity state, which
DR

we can write as
|ψi = |ψ+ i + |ψ− i.
The effect of parity violation will show up in the interference between the |ψ+ i and the
|ψ− i.

60
Co → 60 Ni + e− + ν̄e ,
For the 60 Co decays P leaves B unchanged, but reverses pe . The fact that the electron
fluxes
Je (pe · B) 6= Je (−pe · B)
show that the final state has mixed parity, whereas the initial 60 Co nuclei were in a
definite (+ve) parity states. So clearly the parity operator does not communte with
the evolution operator (and hence neither with the Hamiltonian operator).

If we put the magnetic field up the page

B⇑ µ⇑ JCo ⇑

when conserving momentum and spin we will favour the final state with the electron
moving down the page

ν̄ ↑⇑ e− ↑⇓
more than .
e− ↓⇑ ν ↓⇓

1 Reported in [7].

95
Examples

4.8. Neutrino and anti-neutrino states have only ever been observed with the
following eigenvalues of the helicity operator respectively:
ν : − 12 ~ ν̄ : + 12 ~

What values of the projection operators

 
1 p
P± = 1±σ·
2 |p|
must be present in weak processes for (anti-)neutrinos reactions?

What is the implication for parity in the weak interaction?

Let |ν± i be the neutrino eigenstates of helicity with eigenvalues ± 12 .
ν ν̄
⇐ ⇒
−→ −→
For the neutrino interactions we need P− operators

P− |ν+ i = 0 P− |ν− i = 1

whereas for anti-neutrinos we need P+

P+ |ν̄+ i = 1 P+ |ν̄+ i = 0

[The operators for the weak interaction are of course really 21 (1 ± γ 5 ), but first we’d
FT

need to solve Dirac. . . so this is the best we can do for now.]

A

The complete non-existence (or possibly non-interaction) of positive helicity neutrinos

DR

and negative helicity anti-neutrinos means that the weak interaction violates parity
maximally.

4.9. Write down Feynman diagrams and for the processes π + → e+ νe and
+ +
π → µ νµ . By considering the helicities of the final state particles, suggest why
the π + (J P = 0− ) decays dominantly to µ+ νµ .
The neutrino is -ve helicity, so considering conservation of linear and angular momentum
in the ZMF, the charged lepton must be -ve helicity √ too. The amplitude to find a
right-chiral spin- 21 object in a -ve helicity state is 1 − β (again, need Dirac). So the
amplitude is smaller for the (more relativistic) electron, and for the (less relativistic)
muon.

4.10. The Large Hadron Collider has been designed to accelerate counter-rotating
beams of protons to energies of 7 TeV, and to collide those beams at a small number
of interaction points.

a) Use dimensional analysis to estimate the smallest length scale which this machine
could be used to resolve. How does this compare to the size of e.g. atoms, nuclei
and protons?

b) The LHC beam pipe is evacuated to reduce loss of beam from collisions with gas
moleceules. If less than 5% of the beam protons are to be lost from collisions with
gas nuclei over a ten hour run, estimate the maximum permissible number density
of H gas atoms in the beam pipe.

96
Examples

c) The machine collides counter-rotating bunches of protons, each bunch having

circular cross section with radius 17 µm (in the direction perpendicular to travel).
How many protons are required in a bunch to have an average of ten interactions
per bunch crossing?

d) If such bunches collide every 25 ns, what is the luminosity of the machine?
(Express your answer in units of cm−2 s−1 .)

e) If the cross section for producing a Higgs Boson is 50 pb, how many will be made
each second?

f) What is the kinetic energy of each bunch of protons in the LHC?

[Some data for proton-proton cross sections can be found in Figure A.1.]
Cross section (mb)

2
10

total
⇓

pp
FT

elastic
A

10
DR

Plab GeV/c
-1 2 3 4 5 6 7 8
10 1 10 10 10 10 10 10 10 10

√s GeV
2 3 4
1.9 2 10 10 10 10

Figure A.1: Proton–proton scattering cross section data as a function of laboratory

momentum (upper scale) and of the centre-of-mass energy (lower scale). From [4].
Cross section (mb)

2
10
a) ~c/E = 197 MeV fm/14 TeV so we can resolve ≈ 0.14 × 10−19 m — i.e. about
⇓ the proton.
80 000 times smaller than total

b) We want the fraction lost to be small so 0.05 ≈ nσtot ctbeam . From the plot,
σtot (plab = 7 TeV) ≈ 50pmb
−p
= 50 × 10−31 m2 , so the number density of stray H atoms
should be
0.05 elastic 0.05
10 nH < = ≈ 1015 m−3 .
σtot ctbeam (50 10−31 m2 )(3 108 m s−2 )(10 60 60 s)
Plab GeV/c
c) At-1CM energy of 14 TeV the2 total pp3 cross section
4 is about
5 1006 mb. The7 number 8
of 10 1
interactions 10
per bunch 10
crossing 10
is hN 102 10 10
tot i = Np σtot /A so we require
10 10

A π(17 × 10−6 m)2

Np2 = hNtot i = 10 ≈ 9 × 1020 .
σtot 100 × 10−31 m2

97
Examples

So the protons/bunch should be Np ≈ 3 × 1010 .

d) The rate, W = σL. For σtot = 100 mb, we have specified that

Wtot = 10/25 ns = 4 × 108 s−1

therefore
L = Wtot /σtot = 4 × 1033 cm−2 s−1 .

e) WHiggs = σHiggs L which with σHiggs = 50 pb gives

WHiggs = (50 × 10−12 × 10−24 )(4 × 1033 ) = 0.2 s−1

f) Ebunch = N Ep = (1010 )(7 × 1012 × 1.6 10−19 ) ≈ 11 kJ (macroscopic).

A FT
DR

98
Examples

A.5 Energy production, stars

Solutions

5.1. a) Why does 235 U fission with thermal neutrons whereas 238
U requires
neutrons with energies of order MeV?

b) The fission of 235 U by thermal neutrons is asymmetric, the most probable mass
numbers of fission fragments being 93 and 140. Use the semi-empirical mass
formula to estimate the energy released in fission of 235
92 U and hence the mass of
235
92 U consumed each second in a 1 GW reactor.

c) In almost all uranium ores, the proportion of 235 U to 238 U is 0.0072. However,
in certain samples from Oklo in the Gabon the proportion is 0.0044. Assuming that
a natural fission reactor operated in the Gabon 2 × 109 years ago, estimate the total
energy released from 1 kg of the then naturally occurring uranium. How might the
hypothesis that 235 U was depleted by fission be tested?

[t1/2 (238 U) = 4.5 × 109 years, t1/2 (235 U) = 7.0 × 108 years.]
a) The capture of a neutron changes 235 U from O-E to E-E, so releases the energy from
the pairing term. This energy is sufficient to overcome the energy barrier to fission.

b) We are interested in the process

n + 235 140
X + 93 Y + 3n
FT

92 U −→

The three neutrons are needed for conservation of baryon number. We need to work
A

out the atomic numbers of the two isotopes X and Y . The quick and dirty way to do
DR

this is to assume about the same Z/N ratio as in the parent Uranium isotope, which
would give Z(A = 140) ≈ 55 and Z(A = 93) ≈ 37. The better way is to look at
the binding energy EB as a function of Z for constant A and to find out the point
where dE dZ
B
= 0. (In fact to more precisely find the most stable isobar one should also
account for the mass difference between neutrons and protons and the mass of any
beta electron.) For odd-A nuclei (which have no pairing term) the value of Z at the
minimum
4s + (mn − mp )A
Zmin = 2 .
2(4s + dA 3 )
For A = 140 this yields Z = 56 and for A = 93 we get Z = 40, so we were reasonably
close before. (Consulting a table of nuclides shows the stable isobars are 140 58 Ce and
93
41 Nb respectively.) The energy released is

Efission = M (235 140 93

92 U) − M ( 58 Ce) − M (41 Nb) − 2M (n)

This is very close to 200 MeV. We shall assume that all of the energy, including that of
the beta decays is included in the power output by the reactor. (In reality some energy
will be lost through neutrinos, and some daughters may have long lifetimes.) Then for
power P the rate of reactions W is
P 1GW
W = = = 3.2 × 1019 s−1
Efission 200 MeV
and the rate of consumption of Uranium mass is M (U ) × W which is about 12 mg/s.

c) For most natural uranium

235
U(t)
238 U(t)
= 0.0072

99
Examples

whereas at Oklo
235
U(t)
238 U(t)
= 0.004
Oklo

Calculating backwards,
235
U(t) = 235 U(0) exp(−Γ235 t)
238
U(t) = 238 U(0) exp(−Γ238 t)

so
235 235
U(0) U(t)
238 U(0)
= 238 U(t)
exp[(Γ235 − Γ238 )t]

thus that part of the ratio which was consumed is

(0.0072 − 0.004) exp[(Γ235 − Γ238 )t]

h   i
= 0.0032 exp ln(2) t−1 −1
1 ,235 − t 1 ,238 t
2 2

= 0.017

From 1 kg of uranium is 17 g used. The energy released is

17 g
· 6 × 1023 mol−1 · 170 MeV ≈ 1 TJ
235 g mol−1
Look for fission fragments in rock to test the hypothesis.
FT

5.2. Write down the semi empirical mass formula. Which terms are responsible for
A

the existence of a viable chain reaction of thermal-neutron-induced uranium fission?

DR

What distinguishes the isotopes of uranium that support such a reaction?

In the construction of a nuclear fission reactor an important role is often played

by water, heavy water or graphite. Describe this role and explain why are these
materials are suitable.

Why is the fissile material not completely mixed up with the moderator?
235
Fission barrier from surface term. Energy release from capture of n on U sufficient
to overcome barrier (pairing term).

Water as coolant and as moderator. Heavy water or graphite as moderator.

The moderator is separated from the uranium to cool neutrons below the temperature of
the 238 U(n, γ) capture reaction before they come back into contact with the uranium.
These capture reactions would otherwise reduce the number of neutrons below that
needed to maintain a reaction.

5.3. A neutron produced in a fission reaction is emitted with considerable energy.

Discuss how the design of the reactor determines the competition between i) neutron
absorption by sharp resonances; ii) neutron decay; iii) neutron energy transfer to
the reactor media (and thence to turbines); iv) neutron absorption by resonances
with high branching ratios to further fission. Most of these processes happen very
fast indeed. How is it possible to control the reactor flux with a response time of
seconds to minutes, or even longer?

100
Examples

First parts: See textbooks (e.g. Krane Sec 13.5).

Last part: per fission about 2.5 prompt neutrons per fission. Control relies on ≈ 0.018
delayed neutrons per fission with time constants of order seconds (time constant for
reactor as a whole is larger, of order minutes). Delay is due to neutron emission after
beta decay.

5.4. a) Draw a diagram showing how the fusion process

p + p → d + e+ + ν
is related to the Fermi theory of beta decay.

b) Find the approximate height B of the Coulomb barrier for pp fusion. To ap-
proximately what temperature would one need to heat hydrogen for pp fusion to
overcome the Coulomb barrier?

c) If plasma at this temperature is to be magnetically confined what will be a typical

Larmor radius (gyro-radius) for the duterium ions in the magnetic field?

d) Given the following reactions and energy release (in MeV)

d + d → 3 He + n Q = 3.27,
d+d→t+p Q = 4.04,
d + t → 4 He + n Q = 17.6,
FT

suggest two reasons why the artificial fusion reactors depend largely on the d + t
A

reaction.
DR

e) Tritium has a half-life of about 12 years, and must be generated through reactions
with both 6 Li and 7 Li. Write down the form of these reactions, and explain why
the 7 Li reaction is helpful even though it is endothermic.

[d means 2 H and t means 3 H. You may assume the magnetic field strength is 13.5 Tesla,
which is what has been proposed for the ITER Tokamak.]
a) p → n + e+ + ν can be described by the same four-fermion matrix element assumed
in Fermi theory of beta decay.

b) The barrier height is

1 q1 q2 α
B= =
4π0 r1 + r2 r1 + r2
1 197 MeV fm
Taking r1 = r2 ≈ 1 fm we find that B = 137 2 fm
≈ 700 keV. For a fast rate we
would need to heat each such that 2 kB T ≈ B meaning that T ≈ 5.4 × 109 K. To
3

confine material at this temperature one either uses magnetic confinement (e.g. JET)
or inertial confinement (e.g. NIF).

c) The Larmor radius is √

mv⊥ 2mE
r= ≈
qB qB
For B = 13.5 T, m2 H ≈ 2 GeV, E ≈ 700 keV,
√
1 2 · 2 · 7 × 10−4
r[m] =
0.3 13.5

101
Examples

so is about 1.3 cm.

d) The larger Q value makes dt ignition easier – the Lawson criterion is

12kB T
nτ >
hvσiQ

d + t also has a lower Coulomb barrier than p + p. The nuclear sizes are larger, and
the nucleons can arrange themselves such that the protons are as far apart as possible.
The Boltzmann-averaged cross section hvσi is therefore larger.

e) The reactions are

n + 6 Li → α + t Q = 4.8 MeV
and
n + 7 Li → α + t + n Q = −2.47 MeV
The latter is important since we need to produce > 1 tritium for each neutron, because
of neutron and tritium losses and decays.

5.5. Show that for a star in a state of hydrostatic equilibrium (with pressure
balancing gravity), the pressure gradient is given by
dP Gm
= −ρ 2
dr r

where m is the mass contained within the sphere of radius r. Hence show that the
pressure at the center of a star satisfies
FT

Z M Z M
Gm dm Gm dm
A

Pc = 4
> ,
4πr 4πR4
DR

0 0
where R is the radius of the star.

Estimate the pressure and temperature at the the center of the sun.
For an element at radius r, with volume dr × dS and hence mass dm = ρdr dS the
gravitational and pressure forces must balance
Gm dm
0 = − + [P (r) − P (r + dr)]dS
r2
Gm dm dP dm
= − −
r2 dr ρ
giving the required result, which can also be expressed as
dP Gm
=−
dm 4πr4
The central pressure follows from integrating the above. The pressure the pressure at
the center of the sun must then satisfy
GM 2
PC > = 4.5 × 1013 Pa = 4.5 × 108 atm.
8πR4
since G = 6.67 N m2 kg−2 , M = 2 × 1030 kg, R = 7 × 108 m. The actual pressure is
about a factor of 10 larger.

The temperature can be found from the gas law P V = nkB T so that
mH P
T =
kB ρ

102
Examples

Taking the average density of the sun from a spherical approximation,

2 × 1030 kg
ρ= 4 = 1.8 × 103 kg m−3 .
3
π(7× 108 m)3

For mH = 1.67 × 10−27 kg and kB = 1.38 × 10−23 J K−1 we find T > 6 × 106 K.
Models suggest a core temperature of about 1.6 × 107 K This is in the right ball-park
for fusion, though clearly not in the MeV ∼ 1010 K regime at which the pp Coulomb
barrier becomes unimportant. Nevertheless tunnelling together with the tail of the
Maxwell-Boltzmann can do the job.

[For more on stellar structure see e.g. Prailnik or Clayton]

5.6. The rate of thermonuclear fusion reactions is approximately proportional to

mv 2
   
2πZ1 Z2 αc
exp − exp − .
v 2kT

Sketch the form of this curve and explain the origin of these two terms.

Find the value of v at which the rate is maximal. At what temperature should one
run a Tokamak?
The first term is the tunnelling probability, the as same we calculated for α decay in
the limit where Q = mv 2 was small. [We’ve simplified things by only considering
one-particle dymanics.] The second term is the Boltzman factor.
A FT

R
DR

1.000
0.500

0.100
0.050

0.010
0.005

v
0.0 0.5 1.0 1.5 2.0 2.5 3.0

The maximum occurs when

mv 2
 
d 2πZ1 Z2 αc 2πZ1 Z2 αc mv
0= − − = −
dv v 2kB T v2 kB T
at  1
2πZ1 Z2 αkB T 3
v/c = .
mc2

5.7. a) Assuming that the energy for the sun’s luminosity is provided by the
conversion of 4H → 4 He, and that the neutrinos carry off only about 3 percent of
the energy liberated how many neutrinos are liberated each second from the sun?

103
Examples

b) What neutrino flux would you expect to find at the Earth?

c) By what sequence of reactions do the above conversions dominantly proceed?

d) Why might the alternative rare process

p + e− + p → d + νe

be of interest when studying solar neutrinos from the earth?

[4M (1 H) − M (4 He) = 26.73 MeV. The earth is on average about 1.50 × 1011 m from the
sun, and is subject to a radiation flux of about 1.3 kW m−2 .]

2
The solar power P is 4πrearth J where J is the energy flux at the earth.

P = 4π · (1.5 × 1011 m)2 · (1.3 × 103 W m−2 ) = 3.7 × 1026 W.

In principle we now need to multiply this by 1/(1−fν ) to work out what the total power
(since above we have calculated the visible power output excluding the neutrinos). It
is clear from the question that the fraction of the power found in neutrinos fν is small,
so this correction can be neglected at the level of precision we are using.

There are two neutrinos liberated in the process

FT

4 1 H → 4 He + 2 e+ + 2 νe + 26.73 MeV.
A
DR

The neutrino production rate is then

(3.7 × 1026 W)
Γν = 2 × = 1.7 × 1038 s−1
26.73 MeV × (1.6 × 10−19 × 106 J/MeV)

The solar neutrino flux at earth is then

Γν
Fν = 2
= 6.1 × 1014 m−2 s−1 .
4πrearth

The p − p chain relies on initial production (via weak interaction) of deuteron

p + p → d + e+ + ν

then further proton absorption

d + p → 3 He + γ,

then fusion of two Helium nuclei and emission of a pair of protons. . .

3
He + 3 He → 4 He + 2p.

The more direct alternatives e.g.

p + p → 2 He

are not useful since 2 He is unstable against disintegration.

The pep process is interesting since the emitted neutrinos are monoenergetic.

104
Examples

Solar neutrino predictions from J.N. Bahcall and A.M. Serenelli New
Solar Opacities, Abundances, Helioseismology, and Neutrino Fluxes
ApJ, 621, L85 (2005).

5.8. Give brief accounts of the methods of synthesis of:

12
a) C
FT

28
b) Si
A
DR

56
c) Fe
238
d) U
a) C generated via Helium burning at T ∼ 108 K (cf pp at ∼ 4 × 106 K and CNO
12

at ∼ 1.5 × 107 K). Though unstable with lifetime 2.6 × 10−16 s, 8 Be exists at small
concentration through the reaction
4
He + 4 He ↔ 8 Be.
Collision with another α particle leads to
8
Be → 12 C,
a process which is enhanced by a 12
C∗ resonance at ∼ 7.6 MeV.

28
b) Si generated during carbon and oxygen burning at T ∼ 109 K
12
C + α → 16 0
16
O + 16 0 → 28 Si + 4 He.

c) The Coulomb barrier is too large for

28
Si + 28 Si → 56 Fe.
The reaction proceeds at ∼ 3×109 K via the ‘equilibrium’ process. Photons of ∼ 1 MeV
knock nucleons out of the nuclei creating a soup of nuclei, nuclepons, some alphas and
photons. The nucleons are slowly gathered up by nuclei until the isotopes in the peak
of the B/A curve are obtained.

105
Examples

d) Elements heavier than the peak value of B/A (> 56 Fe) require energy to be created,
and are formed during gravitational collapse of the stellar core when under highly non-
equilibrium conditions by the capture of liberated neutrons in conjunction with beta
decays.

5.9. The two figures show properties of the ‘valley of stability’ of nuclei in the
N -Z plane and the binding energy per nucleon versus mass number A, for nuclei
with lifetimes greater than 108 years.

A FT
DR

Using the data in the figures, estimate the energy released in the thermal-neutron
induced fission of 235 U, given that the daughter nuclei tend to cluster asymmetrically
around A = 140 and 94. Where are the daughter nuclei in relation to the valley
of stability and what happens to them subsequently? Compare this with the 238 U
decay chain, which comprises eight α decays and six β decays to 206 Pb with a total
release of 48.6 MeV and a lifetime of 2 × 1017 s.

There is a flow of heat from the Earth’s interior amounting to a total of the order
of 35 TW. Much of this may be accounted for by decay of radioactive elements.
Using the model above for a typical fission, estimate the rate of fissions needed to
produce such a heat flow and the associated flux of neutrinos.

In a certain model of the Earth it is postulated that there is a self-sustaining fission

reactor at the Earth’s centre, fuelled by 235 U, contributing as much as 5 TW to the
overall heat flow. If the ‘geo-neutrino’ flux could be measured, how might the ‘core
reactor’ model be tested?

106
Examples

235
U has B/A ≈ 7.6 MeV so total binding energy of 235 × 7.6 = 1786 MeV. The
fission fragments have binding energies of about 94 × 8.65 + 140 × 8.4 = 1989 MeV,
so the energy release is 203 MeV.

The daughter nuclei are below the valley of stability (N too large), so will emit neutrons
and beta decay until it reaches the valley.

n → pe− ν̄e

Drawing a straight line on the Z vs. N plot, the A ≈ 94 daughter is about 4 beta
decays from the valley. The A ≈ 140 daughter is also about 8 beta decays from the
valley.

The total neutrino rate required from the 238 decay chain is

35 × 1012 W
W =6× = 2.7 × 1025 s−1
(48.6 × 106 eV)(1.6 × 10−19 J eV−1 )

The ratios of neutrinos to energy is

nν 8
=
E 235,fission 203 MeV
nν 6
=
E 238,decay 48.6 MeV
so the proposed a self-sustaining 235 reactor would decrease the neutrino flux to
6 × 30 TW 8 × 5 TW
FT

+
48.6 MeV 203 MeV
A

= (2.3 + 0.1) × 1025 s−1

DR

5.10. The CNO cycle proceeds in the following steps

Q / MeV rate lifetime

12 13
C+p → N+γ 1.944 r12
13 13
N → C + e+ + ν 2.221 τN
13 14
C+p → N+γ 7.550 r13
14 15
N+p → O+γ 7.293 r14
15 15
O → N + e+ + ν 2.761 τO
15 12
N+p → C + 4 He 4.965 ‘fast’

How much energy is released per cycle? Estimate the fraction of that energy in
neutrinos.

The beta decay time constants are of the order of minutes. The shortest proton
capture time is for 15 N which is of the order of years, whereas the other capture
timescales are significantly longer. Write the coupled linear differential equations of
(12 C, 13 C, 14 N) in the form
d
U = MU
dt
where M depends on the rx but not the τx .

107
Examples

Show that this set of coupled differential equations admits a solution

X
U(t) = ai eλi t ui
i=1,3

with
1
λ1 = 0 λ2,3 =
(−Σ ± ∆)
2
and find Σ and ∆. Why must the elements of u2,3 sum to zero?

Express the relative equilibrium abundances of 12 C, 13 C and 14 N in terms of the

rx , and show that the equilibrium fractions of the beta decaying isotopes satisfy
equations of the form
B = τB rA A.
The net effect of one cycle is
12
C + 4p → 12 C + 4 He + 2e+ + 2νe + 26.7 MeV
Cross-check: Qp−p = 4M (H) − M (4 He) − 26.73 MeV.

Annihilation of the positrons releases a further 2 MeV. [Is this already include in Q
values? Cross-check suggests yes. If so my neutrino energy values below are too
large.] We might expect the to neutrinos carry off about half the energy from each
beta decay, so about 2.5 MeV. So about 10% of the CNO energy would escape as
neutrinos. [Clayton suggests a little less]

The ‘fast’ reactions (including the two β decays) rapidly equilibrate, after which the
FT

other three species evolve according to

 12     12 
−r12
A

C 0 r14 C
d  13  
DR

  13 
 C  =  r12 −r13 0  C .
dt 14 14
N 0 r13 −r14 N

If we try a solution
U = ui eλi t
then
λi ui = Mui .
Label eigenvalues λi and eigenvectors ui . Since the differential equation is linear and
homogeneous, a linear superposition of solutions of the form ui eλi t will also be a
solution. The ai are determined from the initial abundances.

The eigenvalues follow directly from the secular equation

0 = (−r12 − λ)[(−r13 − λ)(−r14 − λ)] − 0 + r14 [(−r12 )(−r13 )]
= −λ3 − λ2 (r12 + r13 + r14 ) − λ(r12 r13 + r12 r14 + r13 r14 )
So are 0, 12 (−Σ ± ∆) with


Σ = r12 + r13 + r14 ,

 12
∆ = Σ2 − 4 (r12 r13 + r12 r14 + r13 r14 )


The equilibrium abundances are those for the eigenvector associated with the null
eigenvalue, which is  
1/r12
u1 =  1/r13 
 

1/r14

108
Examples

The eigenvectors u2,3 give the directions in which the system evolves towards equilib-
rium. Each has components which sum to zero as required by conservation of baryon
number.

From generalisation of the null eigenvector above (check), or directly from the equilib-
rium condition
d 13
1 13

N = r12 12 C −


0= N
dt τN
we find
13
N = τN r12 12 C
15 15
with a similar result for O (and indeed for N).

A FT
DR

109
Examples

FT
A
DR

110
Bibliography

[1] G. Audi, A. H. Wapstra, and C. Thibault. The 2003 atomic mass evaluation:
(ii). tables, graphs and references. Nuclear Physics A, 729(1):337 – 676, 2003.
The 2003 NUBASE and Atomic Mass Evaluations.
[2] A. E. Glassgold. Nuclear density distributions from proton scattering. Rev. Mod.
Phys., 30(2):419–423, Apr 1958.
[3] LEP. Precision electroweak measurements on the Z resonance. [Link].,
427:257–454, 2006.
[4] PDG. Review of particle physics. Phys. Lett., B667:1, 2008.
[5] C. Qi, A. N. Andreyev, M. Huyse, R. J. Liotta, P. Van Duppen, and R. Wyss.
FT

On the validity of the GeigerNuttall alpha-decay law and its microscopic basis.
Phys. Lett., B734:203–206, 2014.
A
DR

[6] Roger D. Woods and David S. Saxon. Diffuse surface optical model for nucleon-
nuclei scattering. Phys. Rev., 95(2):577–578, Jul 1954.
[7] C. S. Wu et al. Experimental test of parity conservation in beta decay. Phys.
Rev., 105:1413–1415, Feb 1957.

111