Fundamentals of Nuclear Physics

Noboru
Takigawa · Kouhei Washiyama
Fundamentals
of Nuclear
Physics
Fundamentals of Nuclear Physics
Noboru Takigawa Kouhei Washiyama
•
Fundamentals of Nuclear
Physics
123
Noboru Takigawa Kouhei Washiyama
Department of Physics Center for Computational Sciences
Graduate School of Science University of Tsukuba
Tohoku University Tsukuba, Ibaraki
Sendai Japan
Japan
ISBN 978-4-431-55377-9 ISBN 978-4-431-55378-6 (eBook)

DOI 10.1007/978-4-431-55378-6
Library of Congress Control Number: 2016958483
Translation from the Japanese language edition: Genshikaku Butsurigaku by Noboru Takigawa, ©
Asakura Publishing Company, Ltd. 2013. All rights reserved.
© Springer Japan 2017
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer Japan KK
The registered company address is: Chiyoda First Bldg. East, 3-8-1 Nishi-Kanda, Chiyoda-ku,
Tokyo 101-0065, Japan
To Noriko, Akiko, Tsuyoshi
and to our parents.
Preface
This book provides an introduction to nuclear physics. Research in nuclear physics

covers a wide variety of subjects, and one can list many key words: nuclear
structure and reactions of stable and unstable nuclei, fission and decay of a nucleus,
extreme states such as the limits of existence and high-spin states, properties at high
temperature and high density, hypernuclei, neutron stars, and nucleosynthesis,
among others. All of these are the subjects of nuclear physics. In addition to these
rather static properties, nuclear reactions such as heavy-ion collisions introduce new
aspects of research, i.e., dynamical properties of nuclei or reaction mechanisms,
such as heavy-ion fusion reactions, dissipation phenomena and liquid–gas phase
transition. Many of these phenomena can be understood from the point of view that
a nucleus is a quantum many-body system of nucleons stabilized by nuclear force.
On the other hand, phenomena at higher energies, driven by, e.g., high-energy
heavy-ion collisions, require a different approach: the approach based on the
quantum chromodynamics (QCD). The study of quark–gluon plasma and of the
QCD phase diagram is representative and forms a large stream of current nuclear
physics.
In this book, we largely restrict our subjects and describe basic features of
nuclear structure and of nuclear decays. b-decay and most excitation motions are
left to other books. Also, nuclear reactions and current subjects such as physics of
unstable nuclei, hypernuclei, and nuclear physics based on QCD are untouched
except for occasional very brief references. Even with these limitations, we could
only briefly mention recent developments. However, we have tried to convey part
of them through columns on the QCD phase diagram of nuclear matter, superheavy
elements, superdeformed states, and overview of the synthesis of elements. We
hope that together with the main text they help readers to grasp our current
knowledge of the nucleus and some recent research trends in nuclear physics. By
restricting the subjects, our aim was to contain many experimental data of basic
nuclear properties or suitable illustrations and explain the main structural features of
nuclei in some detail. This will be useful because most of the phenomena listed in
the first paragraph, but omitted from the book, are intimately related to those
basic properties. We also have attempted to explain how the nuclear model has
vii
viii Preface
developed from the original phenomenological level of a shell model to the modern
understanding based on a many-body theory such as the Hartree–Fock calculations.
We also have described nuclear force, the basis of nuclear physics, in some detail.
We sometimes introduce semi-classical approximation to the original quantum
mechanical formalism. We hope that it helps the readers to intuitively grasp the
underlying physics of complicated nuclear phenomena.
In writing the book our intention was to create not only a good introduction to
nuclear physics, but also a good reference book for physicists to learn the appli-
cation of quantum mechanics and mathematics. Toward that aim, in addition to
describing the basic phenomena of a nucleus, we attempted to convey how the basic
subjects of modern physics such as quantum mechanics, statistical physics, math-
ematics for physics, e.g., complex integrals, are used to describe or interpret various
phenomena of nucleus. We also considered the standard level of knowledge of
junior and senior students, and gave a detailed description to enable them to derive
each equation. Finally, we added the appendix to prove a number of important
formulae in the main text and to show some fundamental formulae.
The contents of the book are based on the lectures that one of the authors, N.T.,
has delivered at Tohoku University, Sendai, Japan, for a long time to junior and
senior students in the undergraduate physics course and also to beginning graduate
students. We have included as sidebars some additional material that was presented
in the class in order to keep the atmosphere of the lecture. Many textbooks and
original papers and figures therein have helped in preparing the lectures and this
book. Several of the figures are taken from them. Using this opportunity, we wish to
thank the authors. The papers cited at various places are not at all complete.
Moreover, it does not mean that they are necessarily the representative papers on
each citation. Nevertheless, we hope that they can help readers to do further study.
This book is an English translation of the Japanese edition, which one of the
authors, N.T., published in 2013. The appendix contains the errata to the original
Japanese edition.
We would like to thank Akif Baha Balantekin for many useful comments for the
writing of this English edition. We thank D.M. Brink, A.B. Balantekin, N. Rowley,
F. Michel, S.Y. Lee, P. Fröbrich, S. Landowne, K. McVoy, W.A. Friedman,
G.F. Bertsch, A. Brown, H. Weidenmüller, H. Friedrich, H. Esbensen,
M.S. Hussein, L.F. Canto, C. Bertulani, P.R.S. Gomes, D. Hinde, M. Dasgupta,
G. Pollarolo, A. Bonasera, M. Di Toro, C. Spitaleri, C. Rolfs, I. Thompson,
S. Ayik, K. Hara, Y. Abe, H. Sagawa, A. Iwamoto, T. Tazawa, M. Ohta, J. Kasagi,
and many other colleagues and friends around the world; and K. Hagino and
A. Ono in the Nuclear Theory Group of Tohoku University, Sendai, for useful
discussions. We are grateful to students of Tohoku University, especially foreign
students, for stimulating us to write this English book. We would like to thank
F. Minato, S. Yusa, S. Iwasaki and his wife, H. Tamura, and H. Koura for preparing
figures, and K. Morita, T. Hirano, Y. Aritomo, T. Kajino, and Y. Motizuki in
preparing the columns; and P. Möller, T. Wada, K. Matsuyanagi, T. Tamae, and
K. Kato for kindly reading sections of the Japanese edition and making many
helpful suggestions. We also thank all the mentors and collaborators who provided
Preface ix
us opportunities to work in their institutions. We are grateful to H. Niko and

R. Takizawa for their help as the editors of this English version. Above all
N.T. would like to thank his wife, Noriko, and his family for their support over the
years.
Sendai, Japan Noboru Takigawa

September 2016 Kouhei Washiyama
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 The Constituents and Basic Structure of Atomic Nuclei . . . . . . . . . 1
1.2 Properties of Particles Relevant to Nuclear Physics . . . . . . . . . . . . 2
1.3 The Role of Various Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Useful Physical Quantities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5 Species of Nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6 Column: QCD Phase Diagram of Nuclear Matter . . . . . . . . . . . . . . 11
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2 Bulk Properties of Nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1 Nuclear Sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1.1 Rutherford Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1.2 Electron Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.3 Mass Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.2 Number Density and Fermi Momentum of Nucleons . . . . . . . . . . . 28
2.2.1 Number Density of Nucleons . . . . . . . . . . . . . . . . . . . . . . . 28
2.2.2 Fermi Momentum: Fermi-Gas Model, Thomas–Fermi
Approximation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.3 Nuclear Masses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.3.1 The Binding Energies: Experimental Data and
Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.3.2 The Semi-empirical Mass Formula (The Weizsäcker–
Bethe Mass Formula)—The Liquid-Drop Model . . . . . . . . . 42
2.3.3 Applications of the Mass Formula (1): The Stability Line,
the Heisenberg Valley . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.3.4 Applications of the Mass Formula (2): Stability
with Respect to Fission . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.3.5 Application to Nuclear Power Generation . . . . . . . . . . . . . . 57
2.3.6 Fission Isomers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
xi
xii Contents
3 The Nuclear Force and Two-Body Systems . . . . . . . . . . . . . . . . . . . . . 65

3.1 The Fundamentals of Nuclear Force . . . . . . . . . . . . . . . . . . . . . . . . 65
3.1.1 The Range of Forces—A Simple Estimate
by the Uncertainty Principle . . . . . . . . . . . . . . . . . . . . . . . . 65
3.1.2 The Radial Dependence . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.1.3 The State Dependence of Nuclear Force . . . . . . . . . . . . . . . 67
3.2 The General Structure of Nuclear Force . . . . . . . . . . . . . . . . . . . . . 71
3.2.1 Static Potentials (Velocity-Independent Potentials) . . . . . . . 71
3.2.2 Velocity-Dependent Potentials. . . . . . . . . . . . . . . . . . . . . . . 73
3.3 The Properties of Deuteron and the Nuclear Force . . . . . . . . . . . . . 74
3.3.1 The Effect of Tensor Force: The Wave Function
in the Spin–Isospin Space . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.3.2 The Radial Wave Function: Estimate of the Magnitude
of the Force Between Proton and Neutron . . . . . . . . . . . . . 77
3.4 Nucleon–Nucleon Scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
3.4.1 Low-Energy Scattering: Effective Range Theory . . . . . . . . . 78
3.4.2 High-Energy Scattering: Exchange Force . . . . . . . . . . . . . . 81
3.4.3 High-Energy Scattering: Repulsive Core . . . . . . . . . . . . . . . 83
3.4.4 Spin Polarization Experiments . . . . . . . . . . . . . . . . . . . . . . . 85
3.5 Microscopic Considerations: Meson Theory, QCD . . . . . . . . . . . . . 85
3.6 Phenomenological Potential with High Accuracy: Realistic
Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.6.1 Hamada–Johnston Potential . . . . . . . . . . . . . . . . . . . . . . . . . 89
3.6.2 Reid Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.7 Summary of the Nuclear Force in the Free Space . . . . . . . . . . . . . 91
3.8 Effective Interaction Inside Nucleus . . . . . . . . . . . . . . . . . . . . . . . . 92
3.8.1 G-Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.8.2 Phenomenological Effective Interaction . . . . . . . . . . . . . . . . 94
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
4 Interaction with Electromagnetic Field:
Electromagnetic Moments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 97
4.1 Hamiltonian of the Electromagnetic Interaction
and Electromagnetic Multipole Moments . . . . . . . . . . . . . . . . .... 97
4.1.1 Operators for the Dipole and Quadrupole Moments . . .... 98
4.1.2 Various Corrections. . . . . . . . . . . . . . . . . . . . . . . . . . . .... 100
4.1.3 Measurement of the Magnetic Moment: Hyperfine
Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.2 Electromagnetic Multipole Operators . . . . . . . . . . . . . . . . . . . . . . . 102
4.3 Properties of the Electromagnetic Multipole Operators . . . . . . . . . . 103
4.3.1 Parity, Tensor Property and Selection Rule . . . . . . . . . . . . . 103
4.3.2 Definition of the Electromagnetic Moments . . . . . . . . . . . . 104
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Contents xiii
5 Shell Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

5.1 Magic Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5.2 Explanation of the Magic Numbers by Mean-Field Theory . . . . . . 109
5.2.1 The Mean Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.2.2 Energy Levels for the Infinite Square-Well Potential . . . . . 110
5.2.3 The Harmonic Oscillator Model . . . . . . . . . . . . . . . . . . . . . 111
5.2.4 The Magic Numbers in the Static Potential Due
to Short Range Force . . . . . . . . . . . . . . . . . . . . . . . . . .... 112
5.2.5 Spin–Orbit Interaction . . . . . . . . . . . . . . . . . . . . . . . . . .... 114
5.3 The Spin and Parity of the Ground and Low-Lying States
of Doubly-Magic 1 Nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . .... 117
5.4 The Magnetic Dipole Moment in the Ground State of Odd
Nuclei: Single Particle Model . . . . . . . . . . . . . . . . . . . . . . . . . .... 119
5.4.1 The Schmidt Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 120
5.4.2 Configuration Mixing and Core Polarization . . . . . . . . .... 121
5.4.3 [Addendum] The Anomalous Magnetic Moments
of Nucleons in the Quark Model . . . . . . . . . . . . . . . . . .... 123
5.5 Mass Number Dependence of the Level Spacing hx . . . . . . . .... 124
5.6 The Magnitude and Origin of Spin–Orbit Interaction . . . . . . . .... 125
5.7 Difference Between the Potentials for Protons
and for Neutrons: Lane Potential . . . . . . . . . . . . . . . . . . . . . . .... 125
5.8 The Spin and Parity of Low-Lying States
of Doubly-Magic 2 Nuclei and the Pairing Correlation . . . . .... 127
5.8.1 The Spin and Parity of the Ground and Low Excited
States of 210 82 Pb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 127
5.8.2 The Effect of d-Type Residual Interaction: Pairing
Correlation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 128
5.9 Column: Superheavy Elements (SHE) . . . . . . . . . . . . . . . . . . .... 130
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 133
6 Microscopic Mean-Field Theory (Hartree–Fock Theory) . . . . . . . . . . 135
6.1 Hartree–Fock Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
6.1.1 Equivalent Local Potential, Effective Mass . . . . . . . . . . . . . 136
6.1.2 Nuclear Matter and Local Density Approximation . . . . . . . 137
6.1.3 Saturation Property in the Well-Behaved Potential,
Constraint to the Exchange Property . . . . . . . . . . . . . . . . . . 140
6.2 Skyrme Hartree–Fock Calculations for Finite Nuclei . . . . . . . . . . . 141
6.2.1 Skyrme Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
6.2.2 Skyrme Hartree–Fock Equation . . . . . . . . . . . . . . . . . . . . . . 143
6.2.3 Energy Density and Determination of Parameters . . . . . . . . 144
6.2.4 Comparison with the Experimental Data . . . . . . . . . . . . . . . 147
6.2.5 The Equation of State, Saturation, Spinodal Line,
Surface Thickness . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 148
6.2.6 Beyond the Hartree–Fock Calculations:
Nucleon–Vibration Coupling; x-Mass . . . . . . . . . . . . .... 154
xiv Contents
6.3 Relativistic Mean-Field Theory (rxq Model) . . . . . . . . . . . . . . . . . 155

6.3.1 Lagrangian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
6.3.2 Field Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
6.3.3 The Mean-Field Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
6.3.4 Prologue to How to Solve the Mean-Field Equations . . . . . 158
6.3.5 Non-relativistic Approximation and the Spin–Orbit
Coupling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
6.3.6 Parameter Sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
6.4 Pairing Correlation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
6.4.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
6.4.2 Multipole Expansion of the Pairing Correlation,
Monopole Pairing Correlation Model and Quasi-Spin
Formalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
6.4.3 BCS Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
6.4.4 The Magnitude of the Gap Parameter . . . . . . . . . . . . . . . . . 168
6.4.5 The Coherence Length . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
7 The Shapes of Nuclei. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 171
7.1 The Observables Relevant to the Nuclear Shape: Multipole
Moments and the Excitation Spectrum . . . . . . . . . . . . . . . . . . .... 171
7.2 Deformation Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 174
7.3 The Deformed Shell Model . . . . . . . . . . . . . . . . . . . . . . . . . . .... 177
7.4 Nucleon Energy Levels in a Deformed One-Body Field:
Nilsson Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 180
7.5 The Spin and Parity of the Ground State of Deformed
Odd Nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 182
7.6 Theoretical Prediction of Nuclear Shape . . . . . . . . . . . . . . . . . .... 182
7.6.1 The Strutinsky Method: Macroscopic–Microscopic
Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
7.6.2 Constrained Hartree–Fock Calculations . . . . . . . . . . . . . . . . 184
7.7 Column: Superdeformed States . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
8 Nuclear Decay and Radioactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
8.1 Alpha Decay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
8.1.1 Decay Width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
8.1.2 The Geiger–Nuttal Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
8.2 Fission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
8.3 Electromagnetic Transitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
8.3.1 Multipole Transition, Reduced Transition Probability . . . . . 203
8.3.2 General Consideration of the Selection Rule
and the Magnitude . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 205
8.3.3 Single-Particle Model Estimate: Weisskopf Units
and Experimental Values. . . . . . . . . . . . . . . . . . . . . . . .... 206
Contents xv
8.3.4 Connection Between Electromagnetic Transitions

and the Shapes and Collective Motions of Nuclei . . . . . . . . 209
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
9 Synthesis of Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
9.1 The Astrophysical S-Factor and Gamow Factor . . . . . . . . . . . . . . . 215
9.2 Gamow Peak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
9.3 Neutron Capture Cross Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
9.4 Synthesis of Heavy Elements: S-Process and R-Process . . . . . . . . . 219
9.5 Column: Overview of the Synthesis of Elements . . . . . . . . . . . . . . 220
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Appendix A: Important Formulae and Their Derivation . . . . . . . . . . . . . 223
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
Chapter 1
Introduction
Abstract It would be useful to have an overview of some fundamental aspects of

nuclei before discussing each subject in detail. In this connection, we briefly describe
in this chapter the constituents and basic structure of atomic nuclei, properties of
particles which are closely related to nuclear physics, the role of the four fundamental
forces in nature in nuclear physics, nuclear species, the abundance of elements and
the phase diagram of nuclear matter.
1.1 The Constituents and Basic Structure of Atomic Nuclei
The atomic nuclei are self-bound many-body systems of protons (p) and neutrons (n)
by strong interaction. Although other baryons such as Δ(1232) are also contained,
their amounts are small. For example, the percentage of Δ(1232)Δ(1232) contained
in the lightest nucleus deuteron (d) is about 1%. π -mesons in virtual states mediate the
interaction between the constituent particles and affect the electromagnetic properties
of protons and neutrons. Furthermore, each proton and neutron is also a composite
particle consisting of three quarks. The other hadrons also consist of quarks. One can
therefore take also the view that atomic nuclei are many-body systems of quarks.
The picture of nuclei and of nuclear phenomena, hence the appropriate way to
describe them, depend on the object and method of observation and the related
energy scale, and lead to various models for nuclei. This book restricts to low-
energy phenomena and discusses the nuclear structure and properties primarily from
the point of view that nuclei are many-body systems of protons and neutrons. The
governing law is quantum mechanics. This contrasts to quantum chromodynamics
for high-energy phenomena. Among various quantum many-body systems, nuclei
have characteristics that the number of constituents is small and also that the leading
forces are strong interactions.
Each nucleus is represented, for example, as 168 O. O is the symbol of the chemical
element. It represents oxygen in this example, hence the number of protons is 8.
This number is called the atomic number, and is given at the left lower side. It is
often omitted, because it has a one to one correspondence to the symbol of element.
The number at the left upper side is called mass number and is given by the sum
© Springer Japan 2017 1

N. Takigawa and K. Washiyama, Fundamentals of Nuclear Physics,
DOI 10.1007/978-4-431-55378-6_1
2 1 Introduction
Fig. 1.1 Conceptual

illustration of nuclear
structure: example of 4521 Sc
of the atomic number and the neutron number. They are denoted by A, Z and N ,
respectively, and A = Z + N .
Figure 1.1 is a conceptual illustration of nuclear structure exemplified by 45 21 Sc.
The enclosing circle has been drawn to indicate the finiteness of the nuclear size. In
reality, it is absent, because nuclei are not given by any external boundary conditions,
but are self-bound systems. The arrows indicate that protons p and neutrons n inside a
nucleus are not fixed at lattice points like atoms in solids, but are moving around with
finite velocities. We learn later that nuclei behave like either liquid or gas depending
on the observables or phenomena we are interested in.
1.2 Properties of Particles Relevant to Nuclear Physics
Table 1.1 gives the properties of particles which are closely related to this book. As
the table shows, proton and neutron resemble each other in many properties such as
the mass and the spin except for electric properties, and are jointly called nucleons.
In order to distinguish them, one introduces the concept of isospin space related to
charge, and considers proton and neutron to be two different states in the isospin
space. The operators and states in the isospin space obey the same law as that of
angular momentum, and are called isospin operators and isospin states, respectively.
Nucleon has two states in the isospin space. Similarly to the spin operators for
electron ŝ, one therefore introduces the isospin operators t̂ by

1 01 1 0 −i 1 1 0
tˆx = ˆ
, ty = ˆ
, tz = , (1.1)
2 10 2 i 0 2 0 −1
and, analogously to the Pauli spin operators σ̂ , τ̂ = 2t̂ by

01 0 −i 1 0
τ̂x = , τ̂ y = , τ̂z = , (1.2)
10 i 0 0 −1
1.2 Properties of Particles Relevant to Nuclear Physics 3
Table 1.1 Properties of particles relevant to nuclear physics. I : isospin, S: strangeness, Rc : radius
of charge distribution, μ: magnetic dipole moment, m in the magneton e/2mc is m e for electron,
m μ for μ particle, m p for proton and neutron, and m p for Λ and Σ particles. The number represents
the mean value if error is not given. The lifetime of proton depends on methods. The mass of ν is
from the decay of tritium. The lifetime of ν is from nuclear reactor with m νe in units of eV. The
quark structure for ρ ±,0 is the same as that for π ±,0 . Taken from [1]
Name I I3 Jπ S mc2 (MeV) Rc μ τ (mean life) (s) Quark
e
(fm) ( 2mc ) model
1+
p 1
2 − 21 2 0 938.3 0.88 ± 2.79 >1.9 × 1029 y uud
0.01
1+
n 1
2
1
2 2 0 939.6 0.34a −1.91 885.7 ± 0.8 udd
γ 1− <6 × 10−23 stable

W± 1 80.4 × 103 3.1 × 10−25
Z0 1 91.2 × 103 2.7 × 10−25
νe 1
2 <2 × 10−6 b >300m νe
e− 1
2 0.511 1.00 >4.6 × 1026 y
μ− 1
2 105.7 1.00 2.2 × 10−6
π+ 1 1 0− 0 139.6 2.6 × 10−8 ud̄
π− 1 −1 0− 0 139.6 2.6 × 10−8 dū
π0 1 0 0− 0 135.0 0.84 × 10−16 √1 (uū−dd̄)
2
ρ ±,0 1 1− 775.5 4.5 × 10−24 (ud̄, dū)

ω 0 1− 782.7 7.9 × 10−23 c
K+ 1
2
1
2 0− 1 493.7 1.24 × 10−8 us̄
K− 1
2 − 21 0− −1 493.7 1.24 × 10−8 sū
K0 1
2 − 21 0− 1 497.6 d s̄
K̄0 1
2
1
2 0− −1 497.6 s d̄
1+
Λ 0 0 2 −1 1115.7 −0.61 2.63 × 10−10 uds
1+
Σ+ 1 1 2 −1 1189.4 2.46 0.80 × 10−10 uus
1+
Σ0 1 0 2 −1 1192.6 (7.4 ± 0.7) × 10−20 uds
1+
Σ− 1 −1 2 −1 1197.4 1.5 × 10−10 dds
1+
Ξ0 1
2
1
2 2 −2 1314.8 2.9 × 10−10 uss
1+
Ξ− 1
2 − 21 2 −2 1321.3 1.6 × 10−10 dss
3+
Δ 3
2 2 0 ∼ 1232 ∼ 6 × 10−24 d
a The mean square charge radius of neutron is rn2 = −0.1161 ± 0.0022 fm2
b μ < 0.54 × 10−10 μ
ν B
c c (uū + dd̄)+c ss̄
1 2
d Δ++ =uuu, Δ+ =uud, Δ0 =udd, Δ− =ddd
4 1 Introduction
2
and considers proton and neutron to be simultaneous eigenstates of t̂ and tˆz in such
a way that1
1 1 1 1 1 0

|n = =
, |p = − = . (1.3)
2 2 0 2 2 1
As Table 1.1 shows, the isospin is one of the important quantum numbers to specify
the property of each particle. Its magnitude is assigned to be I when there exist 2I + 1
particles which have common properties for all aspects such as the mass and spin
but the electric charge. For example, the isospin of π -mesons is 1, since there exist
three particles which differ only in the electric charge. In nuclei, the isospin quantum
number of each state and the symmetry concerning the isospin play important roles
reflecting the symmetry properties of nuclear force in the isospin space.
If a particle is a structureless fermion, one can deduce from the Dirac equation
that its magnetic dipole moment, which is often simply called magnetic moment,
is given by μ = e/2mc, where m is the mass of the particle. In fact, the magnetic
moment of an electron is 1 in units of the Bohr magneton μB = e/2m e c.2 However,
Table 1.1 shows that the magnetic moment of a proton significantly deviates from the
nuclear magneton μ N = e/2m p c. Also, the magnetic moment of a neutron is not
zero, but is nearly comparable in magnitude and opposite in sign to that of a proton.
They are called anomalous magnetic moments and imply that both the proton and
the neutron are composite particles with intrinsic structure.3
Exercise 1.1 Derive the approximate equation for two large components ϕ in the
four-component spinor ψ starting from the Dirac equation in the presence of elec-
tromagnetic fields and assuming that the velocity v of the particle is much smaller
than the speed of light in vacuum c, i.e., v c, and show that the term which
describes the interaction with the magnetic field in the effective Hamiltonian is given
by H = − 2mce
σ · B, where B is the magnetic field. One can thus prove that the
magnetic moment of a Dirac particle is given by e/2mc.
The fact that both proton and neutron are not point particles, but have intrinsic
structure, can be seen also directly from the data of charge distribution. The radius
of the charge distribution of a proton is about 1 fm.4
1 There exists an alternative definition, where proton and neutron are inverted such that |p
= | 21 21 ,
|n = | 21 − 21 . Since N ≥ Z for most stable nuclei, we adopt the Definition (1.3) in this book.
2 Precisely speaking, the experimental value of the magnetic moment of an electron is larger than the
prediction of the Dirac theory by about 0.1%, and can be explained by quantum electrodynamics.
3 It is Otto Stern who experimentally determined the magnetic moment of a proton for the first
time. There remains an episode that Pauli visited Stern while he was conducting the experiment
and denied the significance of the experiment based on the Dirac theory. Despite the criticism,
Stern continued his experiment, and discovered the anomalous magnetic moment of a proton, and
consequently was awarded the Nobel Prize in Physics in 1943.
4 Recent experiments of the scattering of high-energy electrons, and also of polarized electrons, are
shedding new lights on the intrinsic structure of nucleons. For example, it is getting uncovered that
the electric and magnetic charge distributions are different.
1.2 Properties of Particles Relevant to Nuclear Physics 5
Fig. 1.2 A physical nucleon

dressed with a virtual
π -meson cloud
The anomalous magnetic moments of nucleons can be understood by either the

meson theory or the quark model. The latter will be explained in Sect. 5.4.3. Here,
we learn the former.
Let us assume that the proton observed in experiments is a superposition of a
bare proton as a structureless Dirac particle and a bare neutron as a Dirac particle
surrounded by a virtual π + -meson (the left part of Fig. 1.2),

|p ↑) = 1 − Cp |p ↑ + Cp |(n × π + ) ↑ . (1.4)
Since the π -meson is a pseudoscalar particle, the orbit of the π -meson around the
neutron is p-orbit. Corresponding to Eq. (1.4), the magnetic moment of a proton will
be given by
e
μp = (1 − Cp )μ N + Cp . (1.5)
2m π c
The first and the second terms on the right-hand side represent the contribution of
the interaction of the electromagnetic field with a bare proton and with π -meson,
respectively. Because of the difference between the masses of a nucleon and a π -
meson, one can explain the anomalous magnetic moment of a proton by assuming
the admixture of the (neutron×π + -meson)-component in a proton to be about 30%.
Similarly, if one assumes that a neutron is not a genuine Dirac particle, but contains
the component, where π − -meson is moving around a proton as a Dirac particle, by
Cn in proportion (the right part of Fig. 1.2), one obtains

e
μn = Cn μ N − (1.6)
2m π c
for the magnetic moment of a neutron. Assuming Cp = Cn , one obtains
μp + μn = μ N . (1.7)
This agrees well with the experimental data (μp + μn )exp ∼ 0.88μ N .
6 1 Introduction
1.3 The Role of Various Forces
It is known that four types of forces exist in nature. In this section, we briefly survey
the role of four forces in nuclear physics.
The strong interaction is principally responsible for the stability and the struc-
ture of nuclei. The electromagnetic interaction provides a powerful probe of nuclear
structure through the electron scattering from nuclei as well as electromagnetic tran-
sitions and moments thanks to its well understood nature and the weakness of the
force. Furthermore, it governs the lifetime of excited states of nuclei through the
electromagnetic transition by γ -ray emission. The weak interaction governs the sta-
bility of nuclei through β ± -decay. The representative example is the beta decay of
neutron (Fig. 1.3), which is given by
n → p + e− + ν̄e . (1.8)
As shown in Table 1.1, the mean life of the neutron in free space is τ ∼ 14.8 min,
and the corresponding half-life is T1/2 ∼ 10.2 min. The weak interaction plays an
important role also in the synthesis of elements beyond Fe.
Exercise 1.2 Explain the reason why the third particle besides the proton and the
electron is needed in the final state of the β-decay of the neutron. Also, discuss the
properties of that particle.
All of the strong, the electromagnetic and the weak interactions are relevant to
the decay of nuclei. Though there are a large variety of decays, the time scale of the
lifetime associated with them is of the order of 10−21 s, 1 ps = 10−12 s, and 1 min,
respectively, reflecting the difference among their strengths.
The gravitational force plays almost no role in the structure of nuclei. However,
it plays a crucial role in the synthesis of elements. Also, as we learn later, although
there exists no stable nucleus of dineutrons, there exist neutron stars because of the
gravitational force.
Fig. 1.3 The β-decay of a

neutron
1.4 Useful Physical Quantities 7
1.4 Useful Physical Quantities
It is sometimes worth making order-of-magnitude estimates of various physical quan-

tities. In that connection. it is useful to remember the following approximate values
related to the fundamental physical constants c (the speed of light in vacuum), (the
Planck constant divided by 2π ), e (electron charge magnitude), kB (the Boltzmann
constant) as
c = 2.99792458 × 108 m/s ≈ 3.00 × 108 m/s, (1.9)

c = 197.326968 MeV fm ≈ 200 MeV fm, (1.10)
e2 1 1
= ≈ (fine structure constant), (1.11)
c 137.035999074 137
1
k B T = 0.02482 eV ≈ eV at T = 288 K. (1.12)
40
The quantity e2 /c is called the fine structure constant. Equation (1.11) holds when
the proportional coefficient in the static electric force between two particles with
electric charge q1 , q2 is determined such that the force is given by F(r ) = q1 q2 /r 2
when the two particles are apart from each other by the distance r . Equation (1.12)
represents the kinetic energy of thermal neutrons. It is useful to convert the temper-
ature given in units of Kelvin to the corresponding energy in units of MeV.
Exercise 1.3 The range of force is given by the Compton wave length /mc of the
corresponding gauge particle. Estimate the range of the strong interaction and of the
weak interaction.
Exercise 1.4 As Fig. 1.4 shows, the force between two protons is dominated by a
repulsive Coulomb interaction at large distances, and turns attractive in the region
inside their touching radius due to the nuclear force, i.e., due to the strong interac-
tion. Estimate the height of the Coulomb barrier VCB = V (r B ) by assuming that the
touching radius is given by r = r B ∼ 2 × Rp ∼ 2 fm, where Rp is the radius of the
proton.
Exercise 1.5 The temperature of the Sun at the core is about 16 million K. Estimate
the collision energy E at the core of the Sun.
Fig. 1.4 Illustration of the V(r)

collision between two
protons in the Sun VCB Newtonian mechanics
E
Quantum tunneling
rB r
8 1 Introduction
Nuclear reactions in the Sun occur very slowly, because they take place by quan-
tum tunneling as indicated by Exercises 1.4 and 1.5. In reality, there exists another
important hindrance factor. As we learn later, there exists no stable bound state in the
diproton system. The only stable dinucleon system is deuteron consisting of one pro-
ton and one neutron. In order for the fusion of two protons to take place, the inverse
reaction of Eq. (1.8), where a proton is converted into a neutron by weak interaction,
must therefore be involved. Because of the superposition of the quantum tunneling
and weak interaction, the nuclear reaction between two protons is doubly hindered.
Consequently, the Sun burns very slowly. It has burnt already for 4.6 billion years
and is expected to continue to shine for another almost same period.
1.5 Species of Nuclei
The display of nuclei on the two-dimensional plane, where one axis, say the abscissa,
represents the neutron number and the other, say the ordinate, the proton number, is
called Nuclear chart or Segré chart. There are 256 stable nuclei if one includes those
nuclei which have long lifetimes of the order comparable to the lifetime of the Sun
such as U. They lie in the vicinity of the diagonal line of the nuclear chart for the
reason we learn later. The reason why there exist more stable nuclei than the number
of stable elements about 92 in nature is because there exist about three stable nuclei
for each element on average. For example, there exist two stable nuclei, called proton
p and deuteron d, for the element hydrogen.
Incidentally, the nuclei which have the same number of protons (i.e., the same
atomic number), but differ in the neutron number, hence in the mass number as well,
are called isotopes to each other, and those whose neutron numbers are the same,
but the proton numbers are different, the isotones, and those which have same mass
numbers the isobars, respectively.
The study of unstable nuclei with short lifetimes is currently one of the hot subjects
of nuclear physics. If one includes unstable nuclei whose lifetime is longer than 1 µs,
about 7000 nuclei are theoretically predicted to exist, among which about 3000 nuclei
have already been discovered experimentally. Also, a group of nuclei stabilized by
shell effects (see Chap. 5) are predicted to exist in the region far beyond U. They
are called superheavy nuclei or superheavy elements, on which extensive studies are
going on both experimentally and theoretically.
1.5 Species of Nuclei 9
Fig. 1.5 Nuclear chart.

Made from the 2010 version
of Japan Atomic Energy
Agency (JAEA)
Fig. 1.6 Three-dimensional

nuclear chart. Made by the
experimental group of
nuclear physics in the
graduate school of science of
Tohoku University
This book deals with nuclei made from the first-generation quarks, i.e., u and d
quarks, from the point of view of the quark model. In recent years, however, nuclei
containing Λ or Σ or Ξ particles, which include one or two second-generation s
quark (s) as constituents, are also under extensive studies. These novel nuclei are
called hypernuclei. A big progress of their study is expected to be stimulated by
the operation of, e.g., J-PARC. Figure 1.5 shows the ordinary nuclear chart, while
Fig. 1.6 a multi-layers nuclear chart, where the ordinate represents the strangeness
number. The numbers appearing in Fig. 1.5 are the magic numbers, which we learn
in Chap. 5.
To end this chapter, we mention the abundance of elements. Figure 1.7 shows the
relative abundance of elements5 in the conventional unit, i.e., by taking the abundance
5 Seealso [4], where the relative abundance of even–even nuclear species with A ≥ 50 in the solar
system is given.
10 1 Introduction
Fig. 1.7 Schematic curve of atomic abundances as a function of atomic weight based on the data
of Suess and Urey [2]. Taken from [3]
of Si to be 106 , i.e., H(Si) = 106 . The hydrogen dominates by far, and the general
trend is that the abundance sharply decreases with increasing mass number up to
A ∼ 100, then decreases much more slowly. In addition, it is noticeable that there
appears a sharp peak of Fe group and that the abundances of nuclei with particular
neutron numbers are large. Also, one notices that each of the latter abundances
has twin peaks. These features originate from the stability of nuclei, and the magic
numbers, and the way of nucleosynthesis. We will gradually learn them in this book.
1.6 Column: QCD Phase Diagram of Nuclear Matter 11
1.6 Column: QCD Phase Diagram of Nuclear Matter

Water changes its state from ice (solid phase) to water (liquid phase), then to vapour (gas
phase) when the temperature and the pressure vary. Similarly, nuclear matter changes its state
or phase with temperature and density.
Figure 1.8 conceptually represents the phase diagram of nuclear matter from the point of
view of QCD by taking the temperature and the chemical potential, which corresponds to the
baryon density, for the ordinate and the abscissa, respectively. The region marked as nuclear
matter is the state of nucleus treated in detail in this book.
As the figure shows, it is conjectured that a phase of quark–gluon plasma is achieved
irrespective of the density if the temperature becomes high. Related experimental as well as
theoretical studies are extensively performed including the high-energy heavy-ion collisions
(Au + Au) using the RHIC (Relativistic Heavy Ion Collider) at the Brookhaven National Lab-
oratory (BNL) in USA, and the Pb + Pb Collision experiments by the Large Hadron Collider
(LHC) at the European Organization for Nuclear Research (CERN). The stars and the arrows
in the diagram indicate the regions expected to be achieved by collision experiments and the
expected paths of time evolution, respectively. There still remain, however, many uncertainties,
including the phase diagram itself, and studies from various aspects are under way.
Fig. 1.8 QCD phase diagram. Taken from [5], courtesy of Brookhaven National Laboratory
12 1 Introduction
References
1. W.-M. Yao et al. (Particle Data Group), J. Phys. G 33, 1 (2006)

2. H.E. Suess, H.C. Urey, Rev. Mod. Phys. 28, 53 (1956)
3. E.M. Burbidge, G.R. Burbidge, W.A. Fowler, F. Hoyle, Rev. Mod. Phys. 29, 547 (1957)
4. Aage Bohr, Ben R. Mottelson, Nuclear Structure, vol. I (Benjamin, New York, 1969)
5. The Nuclear Science Advisory Committee, The Frontiers of Nuclear Science – A Long Range
Plan (2007). http://science.energy.gov/np/nsac/
Chapter 2
Bulk Properties of Nuclei
Abstract Their sizes and masses are the most fundamental properties of nuclei. They
have simple mass number dependences which suggest that the nucleus behaves like a
liquid and lead to the liquid-drop model for the nucleus. In this chapter we learn these
bulk properties of nuclei and their applications to discussing nuclear stability, muon-
catalysed fusion and the structure of heavy stars. As an example of the applications
we discuss somewhat in detail the basic features of fission and nuclear reactors. We
also mention deviations from what are expected from the liquid-drop model which
suggest the pairing correlation and shell effects. We also discuss the velocity and the
density distributions of nucleons inside a nucleus.
2.1 Nuclear Sizes
2.1.1 Rutherford Scattering
At the beginning of the twentieth century when quantum mechanics was born, various
models were proposed for the structure of atoms such as the plum pudding or raisin
bread model of J.J. Thompson, which assumes that the plus charge distributes over
whole atom together with electrons, and the Saturn model by Hantaro Nagaoka.
Rutherford led these debates to conclusion through the study of scattering of alpha
particles on atom. He proposed the existence of a central part of the atom, i.e., the
core part, which bears all the positive charge that cancels out the total negative charge
of electrons and also carries the dominant part of the mass of the atom. Rutherford
named this core part nucleus, and gave the limiting value to its size.1 At that time,
Rutherford had been engaged in the detailed studies of the properties of alpha particles
emitted from radioactive materials, and knew that the alpha particle is the ionized
He. What Rutherford remarked in the experimental results of Marsden is that alpha
1 Itwas 1911 when Rutherford submitted his article on the atomic model to a science journal.
The idea and the formula of Rutherford were derived by stimulation of experimental results of his
coworker Marsden who had been engaged in the study of scattering of alpha particles emitted from
natural radioactive elements on matter. Furthermore, they have been confirmed experimentally to
be correct by his collaborators Geiger and Marsden.
DOI 10.1007/978-4-431-55378-6_2
14 2 Bulk Properties of Nuclei
Fig. 2.1 Connection

between the classical orbits
of the Rutherford scattering
and differential cross section
particles passing through matter sometimes make a large angle scattering, although
most of them go nearly straight. This experimental feature cannot be explained by
the Thompson model which assumes that positive charges distribute over the whole
atom.2
It is established today that alpha particle is the nucleus of He. Here, we learn how
the Rutherford model for atom was born, and how the information on the nuclear
size is obtained through the scattering experiments of alpha particles.
Let us now consider the scattering of two structureless charged particles by
Coulomb interaction, which is called Coulomb scattering or Rutherford scattering.
Here, the two charged particles represent the alpha particle and the nucleus of the tar-
get atom. Since the mass of electrons is extremely small, one can ignore the scattering
by electrons. The correct expression of the differential cross section for the Coulomb
scattering can be obtained by classical mechanics, although, correctly speaking, one
should use quantum mechanics. One important thing in this connection is that one to
one correspondence holds between the impact parameter b and the scattering angle θ
as illustrated in Fig. 2.1. The alpha particles which pass the area 2π bdb of the impact
parameter between b and b + db in Fig. 2.1 are scattered to the region of solid angle
dΩ = 2π sin θ dθ around the scattering angle θ . On the other hand, the differential
cross section is defined as the number of particles scattered to the region of solid
angle dΩ when there exists one incident particle per unit time and unit area. Hence
by definition
dσ 2π b
= . (2.1)
dθ |dθ/db|
Using the relation

θ
b = a cot (2.2)
2
with
Z 1 Z 2 e2
a= (2.3)
μv2
2 The history of the progress and discoveries of modern physics in the period from late in the
nineteenth century to the beginning of the twentieth century is vividly described in the book by E.
Segré [1].
2.1 Nuclear Sizes 15
which holds between the impact parameter b and the scattering angle θ in the case
of Coulomb scattering, we obtain
dσ dσR 1 dσ a2 1
= ≡ = , (2.4)
dΩ dΩ 2π sin θ dθ 4 sin (θ/2)
4
where the index R in σR means the Rutherford scattering. In Eq. (2.3), μ is the reduced
mass, and v is the speed of the relative motion in the asymptotic region, i.e., at the
beginning of scattering. Note that Eq. (2.4) exactly agrees with the formula of the
differential cross section dσR /dΩ obtained quantum mechanically for the Rutherford
scattering. The characteristics of the Coulomb scattering given by Eq. (2.4) are that
the forward scattering is strong, but also that backward scattering takes place with a
certain probability as well. These match the experimental results of Marsden.
The ground state of the natural radioactive nucleus 21084 Po decays with the half-life
of 138.4 days by emitting an alpha particle (210
84 Po → 82 Pb + α). The kinetic energy
206
of the alpha particle is about 5.3 MeV corresponding to the Q-value of the decay
5.4 MeV. The differential cross section for the scattering, where this α particle is used
to bombard the Au target of atomic number 79, agrees with that for the Rutherford
scattering right up to the backward angle θ = π . This suggests that the sum of the
radii of Au and α (R(Au) + R(α)) is smaller than the distance of closest approach
d(θ = π ) for the scattering with the impact parameter b = 0 leading to the backward
scattering θ = 180◦ in the case of Rutherford scattering. Since the distance of closest
approach d and the scattering angle θ or the impact parameter b is related by

θ
d = a 1 + csc = a + a 2 + b2 (2.5)
2
for the Rutherford scattering, the above mentioned experimental results give the upper
boundary of the radius as R(Au) + R(α) < 4.3 × 10−12 cm. This upper boundary is
much smaller than the radii of atoms, which are of the order of 10−8 cm. Rutherford
was thus guided to his atomic model.
Figure 2.2 shows the ratio of the experimental differential cross section for the
scattering of α particles of 48.2 MeV by Pb target to that for the corresponding Ruther-
ford scattering dσR /dΩ as a function of the scattering angle. The experimental cross
section gets rapidly smaller than that of the Rutherford scattering beyond θ = 30◦ .
This can be understood as the consequence of that the distance of closest approach
is small for the scattering corresponding to large angle scattering as Eq. (2.5) shows,
and hence the overlap between α particle and Pb becomes large, and consequently,
those phenomena such as inelastic scattering which are excluded in the Rutherford
scattering take place. Figure 2.3 shows classical trajectories obtained by fixing Pb
at the origin and by making the incident energy of α particles to 48.2 MeV so as
to match with Fig. 2.2. The circle shows the region corresponding to the sum of the
radii of Pb and α particle, i.e., about 9.1 fm. The figure confirms that the distance of
closest approach for the trajectories corresponding to large angle scattering becomes
indeed smaller than the sum of the radii of Pb and α particle.
Fig. 2.2 Differential cross

section of the elastic
scattering of α particles by
Pb. Taken from [2]
Fig. 2.3 Trajectories of

Rutherford scattering
Exercise 2.1 Estimate the sum of the radii of α and Pb from the result of Fig. 2.2
based on the idea mentioned above.
The ratio of the differential cross section shown in Fig. 2.2 resembles the Fresnel
diffraction of classical optics. The Fresnel diffraction occurs in the case where the
source of light is located in the vicinity of the object that causes diffraction, e.g., an
absorber. The scattering of α particles by a nucleus with large atomic number behaves
like a Fresnel scattering, because the large Coulomb repulsion strongly bends the
trajectory and works to make the source of light effectively locate in the vicinity
of the scatterer. It is also because the partial waves corresponding to small impact
parameters whose distance of closest approach is small are removed from the elastic
scattering due to, e.g., inelastic scattering. Concerning the elastic scattering, this
plays effectively the same role as that of an absorber.
This picture holds when the Coulomb interaction dominates the scattering process.
The strength of the Coulomb interaction increases in proportion to the product of
the charges of the projectile and target nuclei. On the other hand, roughly speaking,
the strength of the nuclear interaction increases in proportion to the reduced mass
A1 A2 /(A1 + A2 ). Hence the refraction effect due to the nuclear interaction becomes
non-negligible for the scattering by a target nucleus with small atomic number. In
fact, a similar differential cross section appears not by diffraction effect, but by
refraction effect. The interpretation and analysis then get complicated. Accordingly,
one can safely estimate the nuclear size based on the consideration of the Coulomb
trajectory together with the strong absorption due to inelastic scattering as described
in the present section when the atomic number of the target nucleus is large.
2.1.2 Electron Scattering
The scattering of α particles by a nucleus had led to the Rutherford atomic model,
and provided a way to estimate the nuclear size. It thus played an important historical
role. However, as stated at the end of the last section, it has a limitation regarding
applicability. In contrast, the scattering of electrons by a nucleus, which we learn in
this section, is a powerful method to study the nuclear size, more exactly, the dis-
tribution of protons inside a nucleus, because only well understood electromagnetic
force is involved.3,4
2.1.2.1 The de Broglie Wavelength of Electron
In the experiments of electron scattering, electrons are injected on the target nucleus
after they are accelerated by, e.g., a linear accelerator. In order to deduce the infor-
mation on the nuclear size from electron scattering, the de Broglie wavelength of the
electron must be of the same order of magnitude as that of the nuclear size or smaller.
In this connection, let us first study the relation between the de Broglie wavelength
of electron and the kinetic energy of electron supplied by the accelerator.
The de Broglie wavelength of electron λe is given by
h
λe = (2.6)
pe
in terms of the momentum of the electron pe and the Planck constant h. On the other
hand, it holds that

E total = m 2e c4 + pe2 c2 = m e c2 + E kin = m e c2 + E e (2.7)
3 R.Hofstadter was awarded the Nobel Prize in Physics 1961 for the study of high energy electron
scattering with linear accelerator and the discovery of the structure of nucleon. He performed also
systematic studies of nuclei by electron scattering.
4 The electron scattering is a powerful method to learn the structure of nucleons mentioned in
Chap. 1, and also to study nuclear excitations such as giant resonances and hypernuclei as well.
Table 2.1 The de Broglie wavelength of electron for some representative acceleration energies
Acceleration energy E e (MeV) 100 200 300 1000 4000
de Broglie wavelength λe (fm) 12.4 6.2 4.1 1.2 0.31
by using the relation of relativity between the momentum and the energy of an
electron, since the total energy of electron is given by adding the energy E e supplied
by the accelerator, i.e., the kinetic energy E kin , to the rest energy. Hence,
pe2 = 2m e E e + E e2 /c2 . (2.8)
By inserting Eq. (2.8) into (2.6), we obtain
λe c c 200
= ≈ ≈ fm . (2.9)
2π E e (1 + 2m e c /E e )
2 1/2 Ee E e /MeV
We ignored the rest energy of electron m e c2 in the third and fourth terms of Eq. (2.9)
by assuming that it is much smaller than the acceleration energy E e . We have given not
the wavelength itself, but the quantity which is obtained by dividing the wavelength
by 2π to facilitate the estimate of the order of magnitude.
Table 2.1 gives the de Broglie wavelength of electron estimated by Eq. (2.9) for
some representative acceleration energies.
2.1.2.2 Form Factor
As Table 2.1 shows, one has to inject electrons on a nucleus after having accelerated
them to much higher energy than the rest energy of electron m e c2 ≈ 0.51 MeV in
order to learn the nuclear size, which is of the order of fm. Hence one needs to use the
Dirac equation, to which a relativistic Fermi particle obeys, in order to theoretically
derive the proper expression of the differential cross section [3].
Exercise 2.2 Evaluate the ratio v/c of the velocity of the electron v to the speed of
light in the vacuum c when the acceleration energy of the electron E e is 100 MeV.
However, here let us simplify the problem in the following way, and learn how the
information on nuclei can be obtained from the analyses of the electron scattering:
1. Treat the scattering of electrons by the electromagnetic field made by a nucleus,
instead of considering the scattering of electrons by the nucleons inside the
nucleus.
2. Consider only the Coulomb force (electric force) and ignore the magnetic force.
3. Use non-relativistic Schrödinger equation.
We express the Coulomb potential as V (r). The scattering amplitude to the angle θ
is then given by

1 2μ
f (1) (θ ) = − e−iq·r V (r)dr (2.10)
4π 2
following the first order Born approximation, which is valid because of the high
energy scattering, and also because the involved electric force is weak compared to
the kinetic energy. The μ is the reduced mass. It can be identified with the mass of
electron m e to a high degree of accuracy. The q is the momentum transfer divided
by and is given by
q = kf − ki , (2.11)
q = |q| = 2k sin(θ/2) . (2.12)
where ki is the wave-vector of the incident electron, kf is the wave-vector of the

electron scattered to the direction of angle θ and k is the wave number of the electron
corresponding to the incident energy.
Exercise 2.3 Derive Eqs. (2.10)–(2.12).
We remark that the potential V (r) obeys the following Poisson equation,
ΔV = 4π Z e2 ρC (r) , (2.13)
in order to relate the scattering amplitude to the distribution of protons inside a

nucleus. The Z is the atomic number of the nucleus to be studied, and ρC (r) is
the charge density at the position r measured from the center of the nucleus.5 It is
normalized as
ρC (r)dr = 1 . (2.14)
By repeating the integration by parts twice in Eq. (2.10), and by using Eq. (2.13), we
obtain,
Z e2 1
f (1) (θ ) = F(q) , (2.15)
2μv sin (θ/2)
2 2
where F(q) is defined as

F(q) ≡ e−iq·r ρC (r)dr . (2.16)
The differential cross section is therefore given by
dσ (1) dσR
= | f (1) (θ )|2 = |F(q)|2 . (2.17)
dΩ dΩ
5 The distribution of protons ρ can be derived from ρC by taking into account the intrinsic structure
p
of proton.
Correctly, the expression of the differential cross section is obtained as
dσ (1) dσM
= |F(q)|2 , (2.18)
dΩ dΩ
2
dσM Z e2 v2
= 1 − 2 sin (θ/2)
2
dΩ 2E e sin2 (θ/2) c
2 2
Z e cos(θ/2)
≈ , (2.19)
2E e sin2 (θ/2)
by replacing the differential cross section of the Rutherford scattering dσR /dΩ by
that for the Mott scattering dσM /dΩ which takes into account relativistic effects
for electrons. The dσM /dΩ gives the differential cross section of the scattering of
electrons by the Coulomb force made by a point charge. Equation (2.18) shows that
the information on the density distribution of protons inside a nucleus can be obtained
through the ratio of the experimental differential cross section to that for the Mott
scattering. The F(q) defined by Eq. (2.16) is the factor which represents the effects
of the finiteness of the nuclear size and is called the form factor.
Especially, if the nucleus is spherical, i.e., if the charge distribution is spherical,
the form factor is given by
∞ ∞
sin(qr ) 2
F(q) = 4π ρC (r ) j0 (qr )r 2 dr = 4π ρC (r ) r dr , (2.20)
0 0 qr
where j0 (x) is the spherical Bessel function of the first kind.
Exercise 2.4 Prove Eq. (2.20) in the following two methods:

1. Perform directly the integration over the angular part of r (dΩr ) in Eq. (2.16).
2. Expand at first the e−iq·r in terms of Legendre functions, and then use the ortho-
normal property of the spherical harmonics.
2.1.2.3 Density Distribution
(1) Estimate of the Nuclear Radius from Diffraction Pattern As Table 2.1 implies,
the experiments of electron scattering are performed with high energy in order to
study nuclear size and the density distribution of protons inside a nucleus. The cor-
responding differential cross section is expected to show a diffraction pattern similar
to that of the Fraunhofer diffraction in optics. Figure 2.4 shows the differential cross
section for the elastic scattering of electrons from Au target at 153 and 183 MeV. The
monotonically decreasing solid line is the Mott scattering cross section for 183 MeV.
The figure shows that the observed cross section is smaller than that for the Mott scat-
tering, and that it has indeed the diffraction pattern, i.e., oscillation, of the Fraunhofer
type.
Fig. 2.4 The elastic

electron-scattering
differential cross section
from Au at energies of 153
and 183 MeV. Taken from [4]
Let us assume that the charge distributes uniformly inside a nucleus with a spher-
ical shape of radius R in order to see how the nuclear size is estimated from the
diffraction pattern. The resulting form factor reads
3 3
F(q) = j1 (q R) = (q R)−2 [sin(q R) − (q R) cos(q R)] . (2.21)
qR qR
We then find that the zeros of j1 (x)/x correspond to the angles where the differen-
tial cross section shows local minima. We denote the magnitude of the transferred
momenta corresponding to those scattering angles θ1 , θ2 , . . . by q1 , q2 , . . .. The first
zero of j1 (x)/x is x1 ≈ 4.49, and the interval between the successive zeros Δx is
about π afterwards. One can therefore estimate the radius either by R ∼ x1 /q1 or by
R ∼ π/(q2 − q1 ) using the q1 , q2 , . . . evaluated from θ1 , θ2 , . . ., which are extracted
from the experimental data.
Exercise 2.5 Estimate the radius of the nucleus of Au from the experimental data
shown in Fig. 2.4.6
6 The dips in diffraction pattern are buried by the distortion effects due to Coulomb force in the case
of target nuclei with a large mass number such as Au. The diffraction pattern appears more clearly
for the target nuclei with small mass number.
(2) The Woods–Saxon Representation of the Density Distribution Accurate infor-

mation on the density distribution7 of nucleons inside a nucleus can be obtained
by making the Fourier transformation of the form factor given by the experiments
of electron scattering. However, one usually takes the method to first postulate a
plausible functional form, then determine the parameters therein to reproduce the
experimental data. In that case, a widely adopted choice is to assume the following
functional form called the Woods–Saxon type,8
ρ0
ρ(r ) = , (2.22)
1 + e(r −R)/a
where R is the parameter to represent the radius. a is the parameter to represent the
thickness of the surface area and is called the surface diffuseness parameter. The
density falls from 90 to 10% of the central density over the region of thickness of
4.4 times a around R. The ρ0 is the central density, and is given as a function of R
and a through the normalization condition,

2
4π 3 2 a
ρdr ∼ ρ0 R 1+π = A. (2.23)
3 R
The two solid curves in Fig. 2.4 which reproduce fairly well the experimental
differential cross section have been calculated with the best fit parameters for the data
at 183 MeV by assuming the Woods–Saxon representation for the charge distribution.
The following values are obtained from the analyses of experimental data for a
large number of stable target nuclei,
R ∼ (1.1–1.2)A1/3 fm , a ∼ 0.6 fm , ρ0 ∼ (0.14–0.17) fm−3 , (2.24)
as parameters in the Woods–Saxon parametrization.9 The fact that the radius is

proportional to the 1/3 power of the mass number, i.e., the number of nucleons
composing the nucleus, and that the density is independent of the mass number
7 Instable nuclei, the protons and neutrons distribute inside the nucleus almost in the same way.
Here, we therefore treat the density distribution of protons and of nucleons as the same except for the
absolute value. In these days, extensive studies are performed on the nuclei far from the β-stability
line, which are called unstable nuclei. It is then getting known that some of them have very different
distributions for protons and for neutrons. For example, the region where there exists only neutrons
largely extends over the surface region in some nuclei such as 11 Li in the vicinity of the neutron
drip line. Such layer is called the neutron halo. Recently, it is reported from the inelastic scattering
of polarized protons (see [5]) and also from the elastic scattering of polarized electrons that even
a typical stable nucleus 208 Pb has a larger radius of the neutron distribution than that of the proton
distribution by 0.15–0.33 fm. Relatedly, the study of the existence of the region consisting of only
neutrons, which is called neutron skin, is an active area of research.
8 There exist deviations from the Woods–Saxon type for individual nucleus. They are explained by
shell model.
9 The value ρ ∼ 0.17 fm −3 is widely accepted as the density of nuclear matter (see [6] for the
0
argument about the detailed mass number dependence of the central density of nuclei with large
mass number).
suggest that the nucleus has a property as a liquid.10 The fact that the density is
independent of the mass number is called the saturation property of nuclear density.
Incidentally, the surface area dominates in light nuclei. The Gaussian type is
therefore often used, instead of the Woods–Saxon type, as a more realistic functional
form.
2.1.2.4 The Root-Mean-Square Radius
The radius of a nucleus is the average of the radii of the region of spatial occupation of
the nucleons composing the nucleus. Hence the concept of root-mean-square radius
is often used in discussing nuclear size. We first define the mean-square radius by
∞ ∞
2
r ≡ r 2 ρ(r )4πr 2 dr ρ(r )4πr 2 dr . (2.25)
0 0
The root-mean-square
radius Rr.m.s. is then defined by taking the square root as
Rr.m.s. ≡ r 2 .
We assumed a sharp density distribution without a surface diffuseness, i.e., a step
function, in a simple analysis of the diffraction pattern of the electron scattering,
e.g., as shown in Eq. (2.21). We name the resulting radius the equivalent radius or
the effective sharp radius and designate it as Req . By definition, Rr.m.s. and Req are
related to each other as
5 2 5
Req = r = Rr.m.s. . (2.26)
3 3
Since Req is proportional to the 1/3 power of the mass number A, let us represent it
as
Req = r0 A1/3 . (2.27)
The proportional coefficient r0 , called radius parameter, is then independent of

individual nucleus, and takes experimentally the value11 of
r0 ∼ 1.1–1.2 fm . (2.28)
10 Although the nucleus appears like a drop of liquid, it is not a classical liquid, but a drop of quantum
liquid, since commutation relations govern the nucleus. In the classical liquid, the momentum space
is not deformed even if it is spatially deformed. On the other hand, in the case of nuclei, a spatial
deformation leads to the deformation in the momentum space through the uncertainty relation (see
the footnote concerning quantum liquid in Sect. 8.3.4).
11 Equations (2.24), (2.27) and (2.28) hold for stable nuclei. It has been found that the radii of
unstable nuclei, especially of nuclei in the vicinity of neutron drip line, significantly deviate from
these equations. For example, in the case of Li isotopes, the radius of 11 Li which has a neutron halo
as mentioned before and thus called a halo nucleus or a neutron halo nucleus is much larger than
what we expect from Eq. (2.27), although the radii of stable isotopes 6,7 Li and of the isotopes 8,9 Li
which lie close to the stability line vary with A almost following Eq. (2.27).
Fig. 2.5 The form factor of the electron scattering by proton. Taken from [7]
If we assume the density distribution of the Woods–Saxon type given by Eq. (2.22),
Eq. (2.26) is replaced by
n
r ρ(r )dr 3 n(n + 5) 2 a
2
r ≡
n
= R 1+
n
π + ··· . (2.29)
ρ(r )dr n+3 6 R
From Eqs. (2.16) and (2.25), we obtain
1 2 2
F(q) = 1 − r q + · · · ,
3!
1
|F(q)|2 = 1 − r 2 q 2 + · · · , (2.30)
3
for small values of the momentum transfer. These formulae show that the nuclear
size can be estimated by studying the form factor for the forward scattering or the
behaviour of the form factor in the region of small momentum transfer. Figure 2.5
shows a similar analysis for the radius of proton. It is derived from the analy-
sis of this experimental data that the root-mean-square radius of proton Rr.m.s. is
0.7–0.8 fm.12,13
12 The measurement with high accuracy in the extreme forward region, where q 2 is extremely small,
is a difficult task and is a challenging subject in these days. In practice, one estimates the radius by
analysing the data in the region where the measurement with high accuracy is doable, by assuming,
e.g., the exponential or the Gaussian density distribution as shown in Fig. 2.5. Also, it is known that
it is necessary to take into account the magnetic effects for the case of nucleons.
13 As Fig. 2.5 shows, the experimental data for the region of small momentum transfer can be
reproduced equally well by assuming the charge distribution in a proton to be either exponential or
Gaussian type. Historically, the charge distribution of exponential type has been considered to be a
good approximation, and correspondingly, the form factor of dipole type has been used. However, it
is getting known from recent studies that the experimental data significantly deviate from the form
factor of the dipole type, and it is argued that the charge distribution inside a proton is closer to a
Gaussian type.
Exercise 2.6 The radius of the charge distribution of a nucleus can be estimated
also from the energy spectrum of the X-ray of the corresponding muonic-atom, or
the energy difference between the corresponding energy levels of mirror nuclei.14
Study the principle of each method.
2.1.3 Mass Distribution
All the methods mentioned above are the method to study the distribution of protons,
in other words, the charge distribution in nuclei, and give no information on the
distribution of neutrons. One has to use other methods in order to study the distribution
of nucleons including neutrons, which is often called the mass distribution, inside
a nucleus. The scattering of protons or neutrons by a nucleus, where the strong
interaction is involved, and nucleus-nucleus collisions are the examples. Similarly
to the case of electron scattering, high incident energies are appropriate. Here we
consider two methods.
2.1.3.1 High-Energy Neutron Scattering: Fraunhofer Scattering
Let us first consider the high-energy neutron scattering. Since the interaction between
the neutron and the target nucleus is strong, neutrons are absorbed or the target nucleus
is excited when the neutrons reach the area of the target nucleus. The associated cross
section, i.e., the sum of the cross sections of inelastic scattering and of absorption, is
intuitively expected to be given by σ (inel) ∼ π R 2 by using the nuclear radius R. On
the other hand, in the scattering of relatively high energy by a short range interaction,
the elastic scattering corresponds to the shadow scattering in quantum mechanics and
the magnitude of its cross section is also expected to be the same, i.e., σ (el) ∼ π R 2 .
In consequence, the total cross section is given by σ (total) ∼ 2π R 2 . One can therefore
obtain the information on nuclear size by measuring the cross section of high-energy
neutron scattering.
In order to make the arguments more precise, let us consider the partial wave
analysis of neutron scattering by a nucleus, and postulate that the S-matrix of the
elastic scattering of each partial wave is given by

(el) 0 for ≤ k R
S = (2.31)
1 for > k R ,
14 The pair nuclei such as 15 Nand 15

7 8 8 O7 which have interchanged numbers of protons and neutrons
are called mirror nuclei to each other. Their energy levels are nearly equal, but shifted.
by emphasizing the above mentioned absorption effects and the characteristic of

high-energy scattering that the phase shift is small. By inserting the extreme postulate
Eq. (2.31) called the sharp-cut-off-model into the partial wave expansion formulae,
Eqs. (A.6), (A.8) and (A.10), for the total cross sections of elastic scattering σ (el) and
of inelastic scattering σ (inel) and for the total cross section σ (total) = σ (el) + σ (inel) , or
into their integral representations transformed to the form of Poisson sum formula
(see Eq. (A.23)), we can confirm that σ (el) ∼ π R 2 , σ (inel) ∼ π R 2 , σ (total) ∼ 2π R 2 .
Experimentally, the total scattering cross section σ (total) can be determined rel-
atively easily by the attenuation method. In this method, the information on the
cross section is obtained by comparing the counts, which are observed with detec-
tors placed in the forward direction, in the measurements with and without scatterer.
If we represent the former and the latter as I and I0 , respectively, then it holds that
I
= exp (−σ (total) nt) , (2.32)
I0
where n is the number density of the target nucleus inside the scatterer, and t the
thickness of the scatterer.
Incidentally, the scattering where the S-matrix is given by Eq. (2.31) corresponds
to the Fraunhofer diffraction in classical optics. A semi-classical scattering theory
leads to the following concise analytic expression for the associated differential cross
section of the elastic scattering,
dσ 1
= R2 |J1 (k Rθ )|2 . (2.33)
dΩ θ sin θ
The property of the Bessel function J1 (x) suggests that the elastic scattering has large
intensities at forward angles. This is nothing but the shadow scattering mentioned at
the beginning of this section.
Another important thing is that an oscillatory diffraction pattern is observed. The
interval between two successive angles, where the cross section takes local maxima,
is given by Δθ D ∼ π/k R from the property of the Bessel function. Hence one can
deduce a value for the nuclear size from the angular spacing of the diffraction pattern.
The validity of the Fraunhofer diffraction method described in this section can be
assessed by comparing thus obtained nuclear radius with the nuclear radius obtained
from the measurement of the total cross section.
One can also use a high-energy proton–nucleus scattering to study the mass distri-
bution. However, the simple model given by Eq. (2.31) cannot be applied in that case
because of the long range Coulomb interaction. In fact, the cross section diverges in
the Coulomb scattering. A special consideration is therefore required to handle the
scattering by Coulomb interaction.
2.1.3.2 Analysis of High-Energy Nucleon–Nucleus

and Nucleus–Nucleus Reaction Cross Sections
by Glauber Theory
The high-energy potential scattering can be well described by the eikonal approxi-
mation, where the scattering phase shift is determined based on the assumption that
the reaction takes place along a straight line trajectory. If one applies this idea to
the nucleon–nucleus and nucleus–nucleus scatterings, and takes the folding model,
where the potential in the eikonal approximation is approximated by folding interac-
tions between either incident nucleon or incident nucleus and constituent nucleons
in the target nucleus, then one obtains the following expressions
∞
σ R = 2π bdb [1 − T (b)] , (2.34)
0
T (b) = exp(−σ̄ N N Ov (b)) , (2.35)
∞
Ov (b) = dz drρT (r)ρ P (r − R) , (2.36)
−∞
which give the total reaction cross section σ R in terms of the total reaction cross
section of the nucleon–nucleon scattering (see Appendix A.4.2). The σ̄ N N in the
transparency function T (b) is the average of the total cross section in the proton–
neutron and proton–proton scatterings. If one denotes the collision energy in the
(N N ) (nn)
nucleon–nucleon scattering by E lab , that in the nucleus–nucleus scattering by E lab ,
and the mass number of the projectile nucleus by A P , then one should use the energy
given by the relationship
(N N ) (nn)
E lab = E lab /A P , (2.37)
(the reasoning is given in Appendix A.4.2). The Ov (b) is called the overlap function.
It is given by integrating the overlap of the densities of the incident nucleus (or
incident nucleon) and of the target nucleus for the impact parameter b over the
collision direction (named z-direction) which is perpendicular to b. The R = (b, z)
is the relative coordinate between the projectile and target nuclei. Since the overlap
integral clearly depends on the nuclear size and the distribution of nucleons inside
nuclei, one can extract information on those quantities from the analysis of the
reaction cross section based on Eq. (2.34).
Equations (2.34)–(2.37) are called the optical limit formulae of the Glauber theory,
whose derivation is given in Appendix A.4.2. It is necessary to extend them either by
taking into account higher order terms in the so-called cumulant expansion method
or by using the diffractive eikonal approximation which goes beyond the optical limit
approximation in order to correctly estimate the size and the reaction cross section
of the nuclei in the vicinity of the drip line, whose studies are recently extensively
going on, by analyzing their scattering data in terms of the Glauber theory (see, e.g.,
[8–10]).
2.2 Number Density and Fermi Momentum of Nucleons
2.2.1 Number Density of Nucleons
Let us denote the number density of nucleons inside a nucleus by ρ. Since ρ is defined
as the mass number A divided by the volume V , we have
A 3
ρ≡ = (2.38)
V 4πr03
from Eq. (2.27) for spherical nuclei. As Eq. (2.28) shows, the radius parameter r0 is
almost constant independently of the mass number. Equation (2.38) therefore shows
that the number density of nucleons inside a nucleus is almost constant independently
of individual nucleus. This property is called the density saturation, and is one of the
experimental evidences of the liquid-like behaviour of nuclei leading to the liquid-
drop model of nucleus.
2.2.2 Fermi Momentum: Fermi-Gas Model, Thomas–Fermi

Approximation
One of the simple pictures for the nucleus as a many-body system of nucleons, which
are fermions, is the so-called Fermi-gas model or Thomas–Fermi approximation. In
that method, each nucleon is considered to be moving independently from each other
in a mean field of volume V . However, the Pauli exclusion principle that each quan-
tum state can be occupied by at most only one fermion is respected. Consequently,
the nucleons become to distribute from the state of minimum momentum to that of
the highest momentum called the Fermi momentum. Since there exists one quantum
state for each volume (2π )3 in the phase space, it holds
kF
V
A= p 2 d pdΩ p × 2 × 2 (2.39)
(2π )3 0
if we represent the Fermi momentum as pF = kF . The last factor 2 × 2 is the sta-
tistical weight in the spin and isospin spaces. Hence we obtain
1/3 1/3
3π 2 9π 1
kF = ρ 1/3 = (2.40)
2 8 r0
by using Eq. (2.27). Similarly to the density, the Fermi momentum also does not
depend on the mass number. The Fermi wave number is proportional to the 1/3
power of the density as one can expect from the consideration of their dimensions.
2.2 Number Density and Fermi Momentum of Nucleons 29
Table 2.2 Nucleon number density, average nucleon distance, Fermi wave number, velocity, Fermi
energy
r0 (fm) ρ(fm−3 ) a (fm) kF (fm−1 ) vF /c E F (MeV)
1.10 0.179 1.77 1.38 0.29 39.5
1.15 0.157 1.85 1.32 0.28 36.1
1.20 0.138 1.94 1.27 0.27 33.1
Table 2.2 shows the number density of nucleons, average distance between nucle-
ons a, Fermi wave number kF , the speed of nucleon at the Fermi surface in units of
the speed of light in vacuum vF /c and the Fermi energy E F for three representative
values of r0 . The table shows that the speed of nucleons in the nucleus is at most 30%
of the speed of light in vacuum. It is the reason why we adopt the non-relativistic
theory to describe nuclei in most part of this book.15
Exercise 2.7 Show that the mass density of nuclei ρmass is about 2.9 × 1014 g/cm3
by assuming that ρ = 0.17 fm−3 .
Sidebar 1: Why don’t stars collapse? It looks puzzling that stars do not collapse
despite the fact that the attractive gravitational forces are acting between the con-
stituent particles. This is because the collapse is prevented by the degeneracy pressure
of electrons and of other constituent particles in the star. As stated in the main text, all
the fermions cannot occupy the smallest momentum state due to the Pauli exclusion
principle. Instead, they distribute up to the Fermi surface with high momentum. The
resulting pressure of the quantum mechanical origin, i.e., due to the Pauli exclusion
principle, is the degeneracy pressure. The Fermi momentum increases and hence the
degeneracy pressure increases if the volume of a star decreases by starting collapse. It
is why stars do not collapse so easily. On the contrary, if the density of fermions such
as electrons inside a star decreases for some reason, then the star starts to collapse
because of the break-down of the balance between the attraction by the gravitational
force and the degeneracy pressure. For example, a supernova explosion is initiated
by the electron capture as we will describe in Sect. 2.3.
Exercise 2.8 Show that the degeneracy pressure of electrons is given by
2/3
3 1
P= h 2 ρ 5/3 (2.41)
π 20m e
with the density of electrons ρ and the mass of electron m e by using the fact that the
pressure P is given by P = −∂ E/∂ V when the temperature is zero, and by identi-
fying the energy E with the total kinetic energy of electrons. Note that particles with
15 One can take the opposite point of view that 30% is a large number. The relativistic descriptions
of nuclear structure and nuclear reactions are also actively going on based on that point of view.
The relativistic treatment has the advantage that the spin of nucleons is naturally introduced. In this
book, we only touch the general framework of the relativistic mean-field theory in Sect. 6.3.
smaller mass work more effectively, i.e., give larger contribution, to the degeneracy
pressure if the number densities of different particles are the same as Eq. (2.41)
shows.
Sidebar 2: Why don’t atoms collapse? A similar question arises for the size of
atoms. Let us consider the neutral hydrogen atom as the example. The attractive
Coulomb force acts between the proton and the electron in the hydrogen atom. The
corresponding potential energy is given by V (r ) = VC (r ) = −e2 /r when they are
separated by the distance r . It therefore looks energetically most stable at r = 0, i.e.,
when the proton and the electron stick to each other. Nonetheless, the size of the
Hydrogen atom is known to be the Bohr radius given by aB = 0.5 × 10−8 cm, which
is larger than the radius of proton Rp ∼ 0.8 × 10−13 cm by 5 orders of magnitude.
This is also a consequence of quantum mechanics. The world of quantum mechanics,
i.e., the microscopic world, is governed by the uncertainty principle Δxα Δpα .
On the other hand, the Hamiltonian describing the motion of the electron around
p2
the proton is given by H = 2μ + VC (r ). Therefore, the uncertainty relation leads to
1 2
2 2
approximately E 2μ ( r ) − er for the corresponding energy. The kinetic energy
increases when r decreases and cancels the gain of the Coulomb energy, and even
leads to a positive total energy. The radius of the Hydrogen atom, i.e., the Bohr radius
aB , is obtained as the value of r which minimizes the energy E.
Sidebar 3: Atoms should have collapsed if the uncertainty principle or the com-
mutation relation were absent Since the uncertainty principle is derived from the
commutation relation [x̂α , p̂β ] = iδαβ , where Ô is the operator corresponding to an
observable O, atoms would become point particles if the commutation relation did
not govern the nature.
Sidebar 4: Application to Screening Effect As we learnt in Sect. 1.4 concerning
the nuclear reactions in the Sun, many nuclear reactions in stars occur by quantum
tunneling through the potential barrier called the Coulomb barrier made by the long-
range Coulomb interaction. An interesting phenomenon in this connection is the
screening effect caused by the surrounding charged particles which are in the plasma
state in hot stars. It leads to an enhancement of the tunneling effect. Recently, the
screening effect is argued also in connection with laboratory experiments, where
low-energy nuclear reactions are caused in matter, e.g., in host metals [11]. It is
reported that the reaction rate is enhanced and the screening effect is considered to
be one of the origins of the enhancement. The interest is whether the enhancement
of the reaction rate by the screening effect due to the presence of the host material is
large and can explain the observed enhancement [12].
Let us denote the effective Coulomb potential, i.e., the screened Coulomb poten-
tial, around a nucleus with electric charge Z e by ϕ(r ). If we represent the change of
the density of a charged particle ζ with the electric charge Z ζ e in the surrounding
induced by this nucleus as δρζ (r), then ϕ(r) obeys the Poisson equation
2.2 Number Density and Fermi Momentum of Nucleons 31
⎡ ⎤

Δϕ(r) = −4π ⎣ Z eδ(r) + Z ζ eδρζ (r)⎦ . (2.42)
ζ
The nuclear reactions in matter at laboratories are performed at the room tem-
perature. Hence, we can use the Thomas–Fermi approximation for the density of
electrons of the host material, and obtain
Ze
ϕ(r) = exp(−qTF r ) (2.43)
r
for the solution of Eq. (2.42) within the framework of the linear approximation,
where the inverse of the screening length qTF is given by

6πρ0
qTF = e2 . (2.44)
εF(0)
The ρ0 and εF(0) are the density of free electrons in the metal and the Fermi energy,
respectively, before they are polarized.
On the other hand, for nuclear reactions in hot stars, the densities of the charged
particles composing the plasma are well represented by the Boltzmann distribution.
Consequently, the screened potential is given by the following well-known Debye–
Hückel formula,
Ze
ϕ(r) = exp(−κDH r ) , (2.45)
r
κDH = 4πβe2 Z ζ2 ρζ(0) , (2.46)
ζ
where β = 1/kT , where k and T are the Boltzmann constant and the temperature
of the star, respectively.
Exercise 2.9 Derive Eqs. (2.44) and (2.46).
2.3 Nuclear Masses
We denote the mass of the nucleus with the mass number A and the atomic number
Z by M(A, Z ). The nuclear mass is one of the basic quantities of each nucleus. At
the same time, it is a fundamental quantity which governs various nuclear reactions.
Fig. 2.6 The binding energy

per nucleon as a function of
the mass number A [13]
2.3.1 The Binding Energies: Experimental Data

and Characteristics
One usually uses the binding energy B(A, Z ) for the mass M(A, Z ) in various
discussions. The B(A, Z ) is defined as the energy gain obtained by forming a nucleus
instead of being completely apart as individual nucleons. It is therefore given by16,17
B(A, Z ) = Z Mp c2 + (A − Z )Mn c2 − M(A, Z )c2 . (2.48)
The filled circles in Fig. 2.6 show the experimental value of the binding energy
per nucleon B/A for the most stable even–even nucleus for each mass number, and
the open circles that for the most stable nucleus in the case of odd mass number.18
The inset shows the details for nuclei with small mass number, i.e., for light nuclei.
The binding energy has the following characteristics including those shown in these
figures.
(1) The binding energy rapidly increases with the mass number in the region of
small mass number (see Table 2.3).19
16 The separation energy is also an important related physical quantity. It is the energy needed to
remove a particle such as a nucleon or an alpha particle from the nucleus. For example, the separation
energy of neutron is given by Sn (N , Z ) ≡ B(A, Z ) − B(A − 1, Z ) in terms of the binding energies
of the relevant nuclei.
17 The binding energy of a nucleus is usually calculated from the mass of the corresponding neutral
atom Matom (A, Z ) as
B(A, Z ) = Z MH c2 + (A − Z )Mn c2 − Matom (A, Z )c2 , (2.47)
where MH is the mass of the Hydrogen atom.

18 Exactly speaking, they have been obtained by using Eq. (2.47).
19 We name deuteron and triton and denote by d and t in the case of nuclei, while name deuterium
and tritium and denote by D and T in the case of neutral atoms.

2.3 Nuclear Masses 33
Table 2.3 Binding energies of some light nuclei

Nuclide d (21 H) t (31 H) 3 He
2 α (42 He)
Binding energy (MeV) 2.225 8.482 7.718 28.296
Table 2.4 Correlation between the even–odd property of the proton and neutron numbers and the
stability of nuclei. The δ represents a systematic variation which appears in the binding energy B
in connection with the even–odd property of the proton and neutron numbers. The Δ is called the
odd–even mass parameter
A Z N Stability Pairing
correlation
energy δ (MeV)
√
Even Even Even Most stable Δ ≈ 12/ A
Odd Even Odd Less stable 0
Odd Odd Even Less stable 0
√
Even Odd Odd Least stable Δ ≈ −12/ A
(2) The B/A shows a sharp maximum at each light α-nucleus.20,21

(3) The binding energy per nucleon B/A is always about 8 MeV except for light
nuclei (saturation property of the binding energy).
(4) Roughly speaking, the B/A steadily increases with the mass number until the
region of Fe whose atomic number is 26,22 then gradually decreases with A.
(5) The B/A shows local maxima when the number of either protons or neutrons
takes special numbers, e.g., Z = 50, in the region of large mass number.
(6) There exist systematic dependences on the even–odd property of the number of
protons and that of neutrons as shown in Table 2.4.23,24
Let us introduce here a few topics associated with the characteristics of the binding
energy mentioned from (1) through (6).
Topic 1: The Energy Generation in the Main Sequence Stars The characteristic
(1) mentioned above is the reason why the energy is generated by nuclear fusion in
stars in the main sequence. The binding energies of light nuclei including the values
20 Those nuclei such as 84 Be, 12 16

6 C, 8 O whose atomic number and the mass number are the same
multiples of those of α particle are called α-nuclei.
21 The binding energy of 8 Be is less than twice that of α by 0.1 MeV, so that 8 Be is unstable and
4 4
exists only transiently.
22 The nucleus which has the largest value of B/A is 62 Ni (see [14]).
28
23 Only 5 nuclei, i.e., d, 6 Li, 10 B, 14 N, 180 Ta (isomer), are stable among the so-called odd–odd
3 5 7 73
nuclei, which have odd numbers for both the proton and the neutron numbers. Incidentally, it is
209 Bi which has the largest atomic number among stable nuclei, although exactly speaking it decays
83
with the half-life of 1.9 × 1019 years. There exist no stable isotopes for Tc and Pm, whose atomic
numbers are 43 and 61, respectively.
24 It is related to the pairing correlation between nucleons, which will be discussed later in detail.
listed in Table 2.3 are the important quantities to determine the energies released
by those nuclear reactions and also the energies of the emitted particles such as
neutrinos.
The nuclear reactions which power the main sequence stars can be classified into
(a) pp-chain (proton–proton chain) reactions, which consist of ppI-chain, ppII-chain
and ppIII-chain reactions, and (b) CNO cycle. Their reactions and the associated
energy releases are as follows:
• ppI chain
1 p + p → 2 H + e+ + νe + 0.42 MeV (νpp )
2 e+ + e− → 2γ + 1.02 MeV
3 2 H + p → 3 He + γ + 5.49 MeV
4 3 He + 3 He → 4 He + 2p + 12.86 MeV
• ppII chain
5 3 He + 4 He→ 7 Be + γ
−
6 Be + e → 7 Li + νe (νBe )
7
7 7 Li + p → 2 4 He
• ppIII chain
8 7 Be + p → 8B + γ
9 8 B → 8 Be + e+ + νe (νB )
10 8 Be → 2 4 He
• CNO cycle
11 12 C + p → 13 N + γ + 1.94 MeV
12 N → 13 C + e+ + νe + 1.20 MeV
13
13 13
C + p → 14 N + γ + 7.55 MeV
14 14
N + p→ 15 O + γ + 7.29 MeV
15 15
O → 15 N + e+ + νe + 1.74 MeV
16 15
N + p → 12 C + 4 He + 4.97 MeV
17 e + e− → 2γ + 1.02 MeV
+
We introduced the notations νpp , νBe , νB to distinguish neutrinos of different origins.

By adding the nuclear reactions in 1 – 4 with appropriate weights, one can see
that the energy of 26.72 MeV in total is released by converting four protons into one
4
He nucleus in the ppI chain,
4p + 2e− → 4 He + 2νe + 26.72 MeV. (2.49)
Also for the CNO cycle, we obtain the same net reaction formula as Eq. (2.49) by
adding 11 – 16 and subtracting twice of 17 on each side. The central temperature
of the Sun is about 16 million K, and the standard solar model predicts that about
98.5% of the solar energy is produced by nuclear reactions in the pp chain, while the
remaining about 1.5% by the CNO cycle.
Fig. 2.7 Conceptual representation of the pp chain and CNO cycle. Taken from [15]
Figure 2.7 represents conceptually the pp chain and the CNO cycle, and Fig. 2.8 the
spectra of solar neutrinos emitted by various reactions. Note that two-body reactions
lead to line spectra, while three-body reactions continuous spectra. Experimentally,
the high-energy neutrino νB is detected either by water Cherenkov detectors or by
using the reaction of 37 Cl,
νe + 37 Cl → 37 Ar + e− . (2.50)
The low-energy neutrino νpp is studied, on the other hand, through the capture reac-
tions by Ga and deuterium.
Fig. 2.8 The solar neutrino

spectra. Taken from [16].
The solid lines are the
spectra for the pp chain and
the dashed lines those for the
CNO cycle
The solar neutrino problem25 was a serious problem, which lasted from mid-
1960s to the beginning of the twenty-first century. It is now attributed to the neutrino
oscillation.
Topic 2: The Energy Problem: Muon-Catalysed Fusion (μCF) The binding ener-
gies of light nuclei listed in Table 2.3 are important quantities in connection with the
so-called energy problem as well. For example, if the nuclear reactions p + d → 3 He,
d + t → 4 He + n take place, the energy of 5.4 and 17.5 MeV is released per reaction for
the two cases, respectively. However, like the p + p → d + e+ + νe reaction in the Sun
mentioned in Sect. 1.4, the high Coulomb barrier between the projectile and target
nuclei has to be overcome in order for these reactions to occur. The so-called high-
temperature fusion reactions attempt to overcome the Coulomb barrier by heating
the nuclei to high temperature. In this case, the nuclei of high temperature need to be
confined in a small space. To that end, various methods such as the magnetic confine-
ment, where the hot fusion fuel is confined by magnetic fields in a small space in the
form of plasma, and the inertial confinement have been developed. However, none
of the methods have yet reached the practical usage for the power production. The
muon-catalysed fusion is an alternative method, where one tries to induce fusion
reactions at low temperature by lowering the Coulomb barrier with the screening
effect by μ− particle.
Figure 2.9 shows the muon-catalysed pd fusion which has been confirmed exper-
imentally for the first time as the evidence of the μCF.26 Figure 2.10 conceptually
shows the muon-catalysed dt fusion which is currently thought to be most promis-
25 It is the problem that the observed number of solar neutrinos is only about 1/2–1/3 of the
theoretical prediction.
26 In 1956, Alvarez et al. discovered a trajectory of muon which stops in the liquid hydrogen
contaminated with deuterium. The flight length in matter is determined by the stopping power. The
length 1.7 cm of the trajectory observed by Alvarez et al. exactly matched the flight length of muon
with energy of 5.4 MeV which is released in the p + d → 3 He reaction.
Fig. 2.9 The first

experimental evidence of the
μCF. Taken from [17]
Fig. 2.10 The μCF cycle.

dtμ → 4 He +n + μ + 17.5
MeV reaction. Taken from
[18]
ing from the point of view of the practical usage for the power production. In the
latter, the liquid target of the mixture of about 50% each of deuterium and tritium
is injected by μ− particle produced by nuclear reactions. Thus produced dμ and tμ
atoms collide to form a dtμ molecule. Eventually the nuclear reaction
d+t →α+n (2.51)
takes place. The μ− particle which stimulated tunneling effect through screening
the Coulomb barrier finishes the role of a catalyst by forming a muonic atom by
sticking to the α particle produced in the nuclear reaction with a certain probability.
However, with a much higher probability it leaves the nucleus by using a part of
the energy released in the reaction, and repeats the cycle as a catalyst. Whether the
muon-catalysed fusion is useful for the practical usage for the energy generation or
not depends on the number of cycle per muon, whose mean life is 2.2 µs, and the
reactivation probability by collision of the muon lost by forming a muonic atom
such as αμ. The dtμ reaction is more promising than the pdμ reaction, because the
dtμ molecule is produced resonantly by the so-called Vesman mechanism, and also
because the sticking probability is smaller (see [18] for details).
Finally, we wish to refer to the nuclear reaction in matter [11] as a closely related
research. It is the study, for example, to see whether the rates of the d + d → t + p,
3
He + n reactions are significantly enhanced over those in the free space if a metal
target such as Pd and PdO is continuously bombarded with deuterons.
Topic 3: The Alpha Particle Model and Cluster Structures Symmetry plays impor-
tant roles in physics. The characteristic (2) of the binding energy mentioned in
Sect. 2.3.1 originates from the property of nuclear force that favours states with good
spatial symmetry. Historically, based on this property, the model called the α-particle
model has been proposed which views the light α-nuclei as aggregates of α parti-
cles and describes their ground states and low-lying excited states by special spatial
configurations or relative motions of the constituent α particles or by the rotations
of the whole nucleus. In this model, for example, the ground state of 12 C is thought
to be an equilateral triangle configuration of three α particles, while the ground state
of 16 O a regular tetrahedron of four α particles. In these days, the standard model
for nuclear structure is the shell model which views the nucleus as an ensemble of
nucleons which are moving independently from each other in a common mean field
of a single center in the zeroth order approximation (see Chap. 5). On the contrary,
the α-particle model is a molecular point of view or a molecular structure model
which assumes the existence of multi-centers or of density localization. This idea
has been extended to variational calculations and currently to the cluster model (the
study of cluster structure) by using the mean-field theory or the molecular dynam-
ics calculations, and intensive studies are going on from that points of view. The
molecular structures are expected to appear at the energies near the threshold energy
for each molecular structure, and are characterized by having a large decay width to
each corresponding decay channel as an indicator. Figure 2.11 shows27 the excitation
energy of various molecular structures in light nuclei expected from this idea. It is
called the Ikeda diagram (see [20, 21] for details of the study of cluster structure).
Figure 2.12 shows the energy-level diagram of 12 C in the region of low excitation
energies [22] in order to help a deeper understanding. The second 0+ state denoted
by 02+ at the excitation energy 7.65 MeV, which is close to the threshold energy for
the α-decay 7.37 MeV, is a state for which the 3α-particle model works well.28 The
ground state locates somewhat far from the threshold energy for the α-decay. Conse-
quently, the density localization becomes less significant. Nevertheless, the picture
that the state consists of 3α-clusters which form an equilateral triangle configura-
tion is still valid to some extent. One can think that it is an experimental evidence
that there appears a 3− state at a low excitation energy, i.e., at 9.64 MeV, which is
expected from the D3h symmetry of the point group associated with the equilateral
triangle configuration. The existence of a rotational band, 0g.s. , 21+ (4.44 MeV), 41+
(14.08 MeV ≈ 4.44 MeV × 3.33) also supports this point of view. However, if there
27 The threshold energies for α-decay, and those for emitting several α particles, and also the thresh-
old energy of the decay of 24 Mg into two 12 C nuclei, taken from [19].
28 The synthesis of 12 C in stars proceeds as 8 Be + α → 12 C(0+ ), 12 C(0+ ) → 12 C(0+ ) In that sense,
2 2 g.s.
the 02+ state of 12 C plays a crucial role in the synthesis of heavy elements, and is named Hoyle-state
after F. Hoyle who predicted the existence of this state.
Fig. 2.11 The Ikeda

diagram: Threshold energies
corresponding to various
cluster structures [19]
Fig. 2.12 The low-lying + 15.110

1
energy levels of 12 C 4
+ 14.083
– 13.352
(2 ) +
1– 12.710
2 11.828
(2+) – 11.160
1 10.844
–
3 9.641
+
0 7.6542
+
2 4.4389
0+ 0 (MeV)
12
C
appear states only up to 4+ member, the rotational property of the spectrum can
be explained also by a usual shell model without assuming the density localization
(Elliott’s SU (3) model). A more accurate description of the ground state and the
ground-state rotational band is provided by a hybrid model, which hybridizes the
α-cluster model and the shell model in order to take into account the broken spatial
symmetry due to the spin–orbit interaction [23].29 Incidentally, the cluster structure
29 Inthe case of 12 C, the spin–orbit interaction reduces the extent of deformation and increases
the excitation energy of the 21+ state. From the point of view of shell model, a force with smaller
spin–orbit splitting leads to a high degeneracy and to a larger deformation due to the Jahn–Teller
effect, while a force with a strong spin–orbit force leads to a large splitting between the p3/2 and
Fig. 2.13 The onion layer

structure of a presupernova
star
and the deformation of nuclei, which is described in detail in Chap. 7 can be thought
to be a kind of Jahn–Teller effect.30
Exercise 2.10 Let us assume that three α particles form an equilateral triangle, i.e.,
the D3h state of the point group. Show that in this case the component of the angular
momentum along the axis perpendicular to the plane made by the three α particles,
which is called the K quantum number, takes only multiples of 3.
Topic 4: The Onion Layer Structure of a Massive Star and the Iron Core In
massive stars whose mass M exceeds 12 times the solar mass, i.e., M > 12M ,
there exists a core consisting of Fe and Ni at the center and successive layers of grad-
ually smaller atomic number distribute towards the surface (onion layer structure).
Figure 2.13 shows the inner structure of a presupernova star. The core part consists
of Fe and Ni, because the binding energy per particle becomes the largest in the Fe
region as remarked in Sect. 2.3.1 as the characteristic (4) of the binding energy per
nucleon B/A.31,32 For the same reason, nuclei with small atomic number and mass
(Footnote 29 continued)
p1/2 states. Consequently, the degeneracy disappears and the spherical shape is predicted to appear.
Inversely, if one considers from the point of view of the cluster model, too large deformation or
clusterization is predicted and consequently the excitation energy of the 21+ state becomes too low
compared with the experimental data if one assumes the cluster state which extends the pure state
of the maximum spatial symmetry of the shell model denoted by (0s)4 (0 p)8 [444]11 L configuration
by introducing a finite distance between three alpha clusters, but ignores the spin–orbit interaction,
and the experimental data can be well reproduced by admixing states with lower spatial symmetry
by the spin–orbit interaction.
30 In the case when the electronic states degenerate in energy in a polyatomic molecule where the
atoms have a highly symmetric geometrical configuration such as a regular polyhedron, it becomes
energetically more stable if the degeneracy is removed by assuming a configuration of atoms with
lower symmetry. This is known as the Jahn–Teller effect. Similar situations occur in many physical
systems including metal microclusters.
31 A statement often appears in the literature that the central part of massive stars consists of 56 Fe
26
due to the fact that the nucleus which has the largest binding energy per nucleon is 56 26 Fe. However,
in reality, the nucleus which has the maximum binding energy per nucleon is 62 28 Ni. It is understood
that the reason why the major component of the core is 56 26 Fe despite this fact is because 62
28 Ni is
56
hard to be synthesized by nuclear reactions in stars, while 26 Fe can be easily synthesized as the last
nucleus in the Si-burning reaction.
32 The temperature of the central part has to be sufficiently high in order for the core of Fe and Ni
to be formed in the central part of stars. In the stars whose mass is not so large, the central part
number proceed towards Fe and Ni by fusion reactions. On the contrary, nuclei with
large atomic number and mass number proceed to Fe and Ni by fission.
Topic 5: The Supernova Explosion It is known that there exist two types of super-
nova explosion, i.e., that initiated by carbon detonation and by gravitational col-
lapse.33 In the latter case, and in massive stars whose core part is made of Fe and
Ni, the gravitational collapse is triggered when Fe in the core captures surrounding
electrons by electron capture reaction,
56
26 Fe + e− → 56
25 Mn + νe , (2.52)
and consequently when the degeneracy pressure of electrons decreases. Also, if the
central temperature exceeds 5 × 109 K, an endothermic photo-disintegration reaction
of Fe,
26 Fe + γ → 13 He + 4n − 125 MeV,
56 4
(2.53)
takes place in addition, and at higher temperatures, the photo-disintegration reaction

of 4 He,
4
He + γ → 2n + 2p − 28.3 MeV, (2.54)
occurs as well. The internal energy which should have increased by contraction does
not lead to the increase of the pressure by being used in these endothermic reactions.
Thus, the contraction of the star cannot be prevented. Consequently, the gravitational
collapse of the star proceeds and eventually is bounced back by the hard core in the
nuclear force (see Sect. 3.7) when the density increases so high as nucleons start
to overlap. The collapse thus changes into explosion and neutrinos and many other
particles are emitted into space. This is a scenario of the supernova explosion. The
supernova explosion is an interesting subject to be studied also in connection with
the synthesis of elements, e.g., as a site to synthesize heavy elements such as U by
rapid neutron capture reactions called r-process.
Exercise 2.11 Show that the thermal pressure in high temperature ideal gas is given
by
PT = ρT (2.55)
in terms of the temperature T and the density ρ. (Hint: Assume that the momentum
distribution of the gas is given by the Boltzmann distribution. See Sect. 6.2.5.)
therefore consists of nuclei with smaller mass number than that of Fe, though they have an onion
layer structure. For example, the central part of stars whose mass lies between 8M < M < 12M
consists of O, Ne and Mg.
33 There is an alternative way to classify the supernova explosion into the type II and I, respectively,
depending on whether the spectrum contains the hydrogen line or not.

2.3.2 The Semi-empirical Mass Formula (The

Weizsäcker–Bethe Mass Formula)—The Liquid-Drop
Model
The bulk behaviour of the binding energy can be well represented by a simple formula
1 (N − Z )2 Z2
B = bvol A − bsurf A2/3 − bsym − bC 1/3 + δ(A) , (2.56)
2 A A
which takes into account major physical effects. Equation (2.56) is called the
Weizsäcker–Bethe mass formula.34 The first and the second terms on the right-hand
side are called the volume and the surface35 terms, respectively, because the nuclear
radius is proportional to A1/3 . The property that the bulk binding energy is the vol-
ume term which is proportional to the number of constituent particles comes from
the fact that the nuclear force which stabilizes a nucleus as an isolated bound sys-
tem is of short range as we learn in the next chapter. The existence of the surface
term will be easily understood from the point of view of the liquid-drop model for
nuclei. The third term is called the symmetry energy term and tends to locate sta-
ble nuclei along the diagonal line in the nuclear chart. We will learn its origin later
somewhat in detail. The fourth term represents the effect of Coulomb energy. As we
have already remarked in Sect. 2.3.1 the last term originates from the pairing correla-
tion between nucleons and represents the fact that the binding energy systematically
changes according to the even–odd property of the atomic number and of the neutron
number. It will be described in detail in Sect. 5.8.
The values of the parameters in the mass formula somewhat vary depending
on the experimental data to be reproduced. The solid line in Fig. 2.6 represents the
theoretical values given by the mass formula when the parameters bvol , bsurf , bsym , bC
are determined by the least-square fit to the binding energies of even–even nuclei
for which N , Z > 7 and the experimental error of the binding energy B is √ smaller
than 200 keV [25]. We fixed the δ for the pairing correlation to be δ = 12/ A MeV.
The dotted line represents the similar fit to odd nuclei. In both cases, the fitting was
performed to the most stable isobar for each mass number A, i.e., the experimental
data shown in the figure. We observe that the mass formula well reproduces the bulk
behaviour of the binding energy. Thus obtained values of the parameters are listed
in Table 2.5.
Comment: The Reason Why Stable Nuclei Locate along the Diagonal Line of
the Nuclear Chart: The Symmetry Energy and Its Origin As stated above, the
symmetry energy locates stable nuclei nearly along the diagonal line in the nuclear
chart. Let us consider the contribution of the kinetic energy to the binding energy in
34 At present, there exist several refined mass formulae [24] reflecting various studies such as those
of unstable nuclei, superheavy elements and nucleosynthesis of elements through r-process. In those
refined mass formulae, for example, the effects of surface diffuseness and of the symmetry energy
correction to the volume as well as the surface terms are considered.
35 If we define the nuclear surface tension σ by σ ≡ b 2/3 /4π(r A1/3 )2 , then σ ∼ 1 MeV/fm2 .
surf A 0
Table 2.5 The values of parameters in the mass formula used to obtain the solid and dotted lines
in Fig. 2.6. All the values are in units of MeV
bvol bsurf bsym bC δ
√
Even–even (solid line) 16.2 19.0 47.0 0.755 12/ A
Odd (dotted line) 16.3 19.2 45.4 0.770 0
order to understand the origin of the symmetry energy. If we use the Thomas–Fermi
approximation, we have
(n) (p) 3 (n) 3 (p) 3 1 (N − Z )2

E kin = E kin + E kin = εF N + εF Z ≈ εF A + εF . (2.57)
5 5 5 3 A
By comparing the second term on the right-hand side of Eq. (2.57) with the symmetry
energy term in the mass formula, we obtain
(kin) 2
bsym = εF ∼ 25 MeV. (2.58)
3
The comparison of this value with the value of the parameter of the symmetry energy
bsym given in Table 2.5 shows that about half of the symmetry energy arises from
the kinetic energy due to the Pauli exclusion principle (Pauli effect). The other half
originates from the fact that the nuclear force in the isospin singlet term, i.e., that
between n and p in the isospin singlet configuration, is stronger than that in the
isospin triplet terms, i.e., that in the isospin triplet configurations of the pp, np and
nn systems as we learn in the next chapter.
2.3.3 Applications of the Mass Formula (1): The Stability

Line, the Heisenberg Valley
The mass formula is useful, for example, to predict the masses of unknown nuclei
and also to discuss the bulk aspects concerning the stability of nuclei such as the
stability with respect to fission. Here and in the next subsection, let us study some
of the applications of the mass formula.
2.3.3.1 The Formula for the Stability Line
In the region of small atomic number, the stable nuclei locate along the diagonal
line or in the vicinity in the nuclear chart. This is due to the symmetry energy term
in the mass formula. However, the contribution of the Coulomb energy becomes
non-negligible as the atomic number gets large. Consequently, the location of stable
nuclei gradually deviates from the diagonal line and the neutron number of stable
nuclei becomes about 1.6 times the proton number (N ∼ 1.6Z ) in the region of large
atomic number such as U. This trend can be derived from the mass formula. If we
denote the atomic number of the most stable nucleus for a given mass number A0 by
Z βs (A0 ), then ∂∂Z B(A, Z )| A=A0 = 0 at Z = Z βs (A0 ). Hence we obtain
A/2 A/2
Z βs (A) = ≈ (2.59)
1+ bC
2bsym
A2/3 1 + 0.00803A2/3
from the mass formula. We used the values listed in Table 2.5 in obtaining the third
term. If we denote the value of Z in the case when we let N = 1.6Z by Z̃ , for example,
Z βs = 61, Z̃ = 58 for A = 150, and Z βs = 81, Z̃ = 80 for A = 208, indicating that
the values of Z βs and Z̃ are very close to each other. In fact, 150 208
62 Sm and 82 Pb are
36
stable nuclei. In this way, the ratio between the proton number and the neutron
number for heavy nuclei is determined by the balance between the Coulomb energy
and the symmetry energy.
2.3.3.2 The Heisenberg Valley
Figure 2.14 is a schematic three-dimensional representation of the nuclear mass

M as a function of the neutron number N and the proton number Z . There exist
three planes corresponding to the even–even, even–odd or odd–even, and odd–odd
nuclei due to the pairing term δ in the mass formula. The line along a cross section
represents the variation of the masses for a given isobars. It behaves like a quadratic
function because of the symmetry energy. This property is shown in Fig. 2.15 for the
nuclei with odd-mass number, and in Fig. 2.16 for even-A nuclei. As these figures
Fig. 2.14 The Z , N

dependence of the nuclear
mass: The Heisenberg valley
36150 Ce and 208 Hg are unstable nuclei. The relationship

58 80 N ∼ 1.6Z should be taken only as a rough
relationship that holds in the region of large mass number. To be accurate, it is safer to use Eq. (2.59).
Fig. 2.15 The Z , N

masses of the isobars with an
odd A
Fig. 2.16 The Z , N

masses of the isobars with an
even A
show, there exist one stable nucleus in the case of odd A and 1–3 stable nuclei in the
case of even A which are stable with respect to β-decay or electron capture.
These stable nuclei form a kind of valley in Fig. 2.14 as the mass number is varied.
This valley is called the Heisenberg valley. Its projection on the N Z -plane follows
almost along Eq. (2.59). The synthesis of light elements up to Fe by nuclear fusion
and the synthesis of heavy elements up to Bi by the slow neutron capture reaction,
called s-process, proceed along the Heisenberg valley.
Incidentally, the conditions for the β − -decay, β + -decay and electron capture to
occur are given by the following equations using the mass M(A, Z ) of the neutral
atom of atomic number Z and mass number A, the mass of electron m e and the
binding energy Δe of the electron to be captured in the electron capture in the atom
before it is captured,
M(A, Z ) > M(A, Z + 1) for β − decay, (2.60)

+
M(A, Z ) > M(A, Z − 1) + 2m e for β decay, (2.61)
M(A, Z ) > M(A, Z − 1) + Δe /c 2
for electron capture. (2.62)
Here, we ignored the change of the binding energy of electron associated with the
change of atom.
Exercise 2.12 Prove Eqs. (2.60)–(2.62).

Fig. 2.17 The mass

distribution of the fission
fragments in the
thermal-neutron-induced
fission of 235
92 U. Taken from
[26]
2.3.4 Applications of the Mass Formula (2): Stability

with Respect to Fission
Fission or nuclear fission is a process where a nucleus with a large atomic number
such as U, Th and Pu disintegrates into two fragments, called fission fragments, with
nearly equal mass numbers. As an example, Fig. 2.17 shows the mass distribution
of the fission fragments in the thermal-neutron-induced fission of 235 U.37 The fis-
sion was discovered by Hahn and Strassmann in 1938 [28].38 Bohr and Wheeler
gave a theoretical interpretation soon after the discovery of fission. The following
discussions follow their paper [29].39
Figure 2.18 schematically represents a process, where a nucleus viewed as a liquid-
drop disintegrates into two fission fragments of equal mass, i.e., of a symmetric
fission.40 The configuration d in the figure is called scission point.
37 In the spontaneous fission or in the induced fission by a small energy, hence with low excita-
tion energy, fission with a striking mass-asymmetry occurs as shown in Fig. 2.17 because of the
shell effect which we study in Chap. 5. Also, a few neutrons are emitted simultaneously in each
spontaneous fission and in each thermal neutron-induced fission. For example, the average num-
bers of neutrons emitted in thermal neutron-induced fission of 235 U and 239 94 Pu are 2.4 and 2.9,
respectively [27].
38 Hahn and Strassmann, who were studying the decay of U by irradiation with neutrons, discovered
fission by identifying the fission fragments, which had been believed to be 88 Ra, 89 Ac, 90 Th, as
56 Ba, 57 La, 58 Ce. Lise Meitner made also a significant contribution to the series of studies as a
collaborator of Hahn.
39 The basic data and theoretical descriptions of fission are given in detail in [27].
40 As we learn in Chap. 7 many nuclei which undergo spontaneous fission such as actinides to which
U belongs are deformed even in their ground states due to the shell effect, i.e., a quantum effect,
and the long axis is about 30% longer than the short axis. Also, as shown in Fig. 2.17, most of the
fission is an asymmetric fission, where the masses of the two fission fragments are significantly
different. Following the original picture of Bohr–Wheeler which is based on the liquid-drop model,
Fig. 2.18 A schematic

representation of a
symmetric fission
2.3.4.1 The Q-Value of a Spontaneous Fission
Let us first evaluate the Q-value of a spontaneous fission by assuming that a nucleus
with a large mass number disintegrates into two fission fragments with equal mass.41
We have
Q ≡ 2B(A/2, Z /2) − B(A, Z ) (2.63)

2
Z bsurf (21/3 − 1)
= A2/3 bC (1 − 2−2/3 ) − (2.64)
A bC (1 − 2−2/3 )
≈ A2/3 bC (1 − 2−2/3 )(Z 2 /A − 17.7) (2.65)
from the mass formula. Hence, if Z 2 /A > 17.7, i.e., if Z > 35 when we assume A ∼
2Z , the nucleus eventually undergoes fission even if we leave it alone. This is called
spontaneous fission.42 For example, if one assumes the fission 236 92 U→ 2 × 46 Pd,
118
43
then the Q-value is about 182 MeV.
Fig. 2.19 An illustration of

the fission barrier by the
liquid-drop model
we develop in this book our discussions by assuming that the ground state of nuclei prior to fission,
which is a metastable state, is spherical. Incidentally, the shell effect was discovered and the shell
model has been established much later.
41 The heat associated with a nuclear reaction or with an atomic or a molecular reaction is called the
Q-value of the reaction. It becomes positive or negative for exothermic or endothermic reactions,
respectively. It corresponds to the energy released by fission in the case of a nuclear fission.
42 Fission is stimulated if the nucleus is irradiated by a photon, or neutron or by other particles. Such
reaction is called induced fission and is distinguished from the spontaneous fission. It is represented
as, for example, (γ ,f) and (n,f) for photofission and neutron-induced fission, respectively.
43 About 200 MeV is a measure of the Q-value for the fission of actinides such as U.
Fig. 2.20 Illustration of the

fission barrier by the
liquid-drop model with shell
effect [27]
The fact that the Q-value is positive does not necessarily mean that fission takes
place instantaneously. Figure 2.19 schematically shows the potential barrier govern-
ing fission. How to take fission coordinates is itself an important question. Here, we
take the nuclear deformation as the fission coordinate corresponding to the concep-
tual picture for fission shown in Fig. 2.18. This potential barrier has to be overcome
by a quantum tunneling in order for fission to occur.44 Consequently, the lifetime
sf
for fission is extremely long. For example, the half-life is T1/2 ∼ 1017 years in the
case of U. On the other hand, heavy nuclei decay also by emitting α particle. As
238
we learn in Sect. 8.1, the α-decay also occurs by a quantum tunneling. In that case,
the abscissa of the potential barrier which corresponds to Fig. 2.19 is the relative
coordinate between the α particle and the remaining nucleus in the α-decay, called
α
the daughter nucleus.45 In the case of 238 U, the half-life of α-decay is T1/2 ∼ 109
years. As is clear in the WKB formula for a quantum tunneling Eqs. (8.6), (8.7) or a
revised formula in the uniform approximation Eq. (2.77), the tunneling probability
dramatically decreases following the exponential law as the mass parameter μ of the
decaying particle or of the decaying degree of freedom gets large. The μ is nearly
equal to the mass of the α particle in the case of α-decay, while the reduced mass of
the fission fragments gives a rough measure of μ in the case of fission if the relative
coordinate between fission fragments is used as the fission coordinate. Obviously, the
44 In reality, fission is a quantum tunneling in a multidimensional space with many degrees of

freedom.
45 In the case of α-decay, the potential barrier to be transmitted by a quantum tunneling is located
outside the sum of the radii of the α particle and the daughter nucleus. In other words, if we consider
the α-decay as a kind of fission process, then the potential barrier appears outside the scission point.
That is the reason why the coordinate for the α-decay can be taken to be the coordinate of the
relative motion.
latter is much larger. In addition, roughly speaking, the barrier height is proportional
to the product of the charges of the fission fragments. One can therefore expect that
the barrier height to be transmitted is much higher for fission. It is why the half-life
for fission is much longer than that for α-decay. Consequently, the decay of heavy
elements occurs mainly by α-decay.
Figure 2.20 shows schematic potential surfaces for fission by Nix. The solid and
the dotted lines are the potential surfaces with and without shell correction, which
will be discussed later, respectively. It should be remarked that the shell correction
introduces a dip in the potential surface at the barrier region.
2.3.4.2 Potential Barrier with Respect to Symmetric Spontaneous

Fission and the Fissility Parameter
(1) Fission Coordinate: Deformation Parameter Let us next study how the poten-
tial surface with respect to fission shown conceptually in Fig. 2.19 depends on the
atomic number and the mass number. Since the isotopes of 235,238 U exist in nature
with finite but long lifetimes, one can expect that the potential surface for fission
has a potential barrier as depicted in Fig. 2.19 up to at least Z = 92. On the other
hand, there exist no stable nuclei with an extremely large atomic number. One can
therefore also expect that the potential barrier ultimately disappears for large enough
atomic number. How large is the atomic number at the border?
As stated before, it is not clear at all how to choose the coordinates or parameters
for fission. It will be natural to use the distance between two fission fragments as the
fission coordinate once they are more apart than the scission point in the late stage
of fission. However, the deformation parameters introduced in the following will be
more natural to describe the initial stage of fission.
Let us consider fission as a process, where the deformation of a nucleus grows
with axial symmetry. The radius R(θ ) of an axial-symmetrically deformed nucleus
can be expanded as

R(θ ) = R0 1 + α0 + αλ Pλ (cos θ ) (2.66)
λ=2,4,...
by using the Legendre functions Pλ (cos θ ), which form a complete set. Here, αλ
are called the deformation parameters. Among them, α2 , α4 are named quadrupole
and hexadecapole deformation parameters, respectively.46 α0 is given as a func-
tion of αλ =0 from the condition of volume conservation by considering nucleus
as an incompressible liquid. Let us truncate the expansion in Eq. (2.66) by the
λ = 2 term in order to understand the physical meaning of the deformation para-
meter. If we let the radius for θ = 0 and θ = π/2 to be a and b, respectively,
46 We do not consider λ = 1, because it describes the translational motion of the nucleus as a

√
whole. There holds the relationship αλ = (2λ + 1)/4π α20 between the αλ in this section and the
deformation parameters αλμ , which will be introduced in Chap. 7.
Fig. 2.21 Schematic picture for quadrupole, octupole and hexadecapole deformations
√
we obtain a ≡ R(0) = R0 (1 + α2 ), b ≡ R(π/2) = R0 (1 − α2 /2) ∼ R0 / 1 + α2
by using P2 (cos θ ) = (3 cos2 θ − 1)/2. One can see that one obtains a nuclear shape
which is elongated (shrunk) along the symmetry axis for positive (negative) α2 . Also,
as the expressions for a and b show, the volume is conserved for small deformation
in accord with the property of a nucleus as an incompressible liquid. In Eq. (2.66),
we assumed the symmetric fission, i.e., the reflection symmetry with respect to the
plane which equally divides the symmetry axis. Correspondingly, the deformation
parameters of odd order such as the octupole deformation λ = 3 have been omitted.
In Fig. 2.21, we show the nuclear shape corresponding to λ = 2, 3 and 4.
Exercise 2.13 Confirm that √the relation shown in Fig. 2.21 between the nuclear
shape and the sign of α4 = 9/4π α40 in the presence of hexadecapole deforma-
tion holds by using the explicit functional form of the Legendre polynomial P4 (z),
P4 (z) = (35z 4 − 30z 2 + 3)/8.
(2) Stability with Respect to Fission: Fissility Parameter Both the surface energy
E surf and the Coulomb energy E C change when a nucleus changes its shape. Hence
it is necessary to study how they depend on the deformation parameters in order to
evaluate the potential barrier with respect to fission. The following formulae can be
obtained from the mass formula:

2 5 (n − 1)(n + 2)
E surf = bsurf A2/3
1 + α22 + α32 + αn2 , (2.67)
5 7 n≥4
2(2n + 1)

Z 2
1 2 10 2 5(n − 1) 2
E C = bC 1/3 1 − α2 − α3 − α . (2.68)
A 5 49 n≥4
(2n + 1)2 n
If we consider within the framework of the quadrupole deformation and keep

terms only up to the second order of the deformation parameter, the change of the
energy by deformation is given by
Fig. 2.22 A schematic illustration of the multi-dimensional potential surface for fission and the
potential barrier. After [29]
2
1 Z
ΔE = bC α2 A
2 2/3
(1 − χ ) , (2.69)
5 A cr
2
Z bsurf
=2× ∼ 50.3 , (2.70)
A cr bC
2 2
Z Z
χ≡ . (2.71)
A A cr
We see that there appears a potential barrier for the nuclei with χ < 1 and they
become metastable with respect to fission. On the contrary, nuclei with a large atomic
number for which χ > 1 have no potential barrier and immediately disintegrate by
fission.47 The χ is called the fissility parameter.
(3) The Height of the Fission Barrier In reality, the spontaneous fission is a quantum
tunneling in a multidimensional space specified either by various deformation para-
meters or by appropriately chosen coordinates. Figure 2.22, which has been quoted
from a paper by Bohr and Wheeler, schematically illustrates fission. The lines in the
figure on the left side represent equipotential surfaces. In a simplified picture, one
considers that a spherical metastable state decays along a path going through the
saddle point. Let us define the height of the fission barrier as the height of the saddle
point measured from the potential minimum corresponding to the metastable state
for fission, and denote it by E f . The value of E f can be approximately written down
as a function of χ if the fissility parameter χ is small or close to 1. An important
point in doing so is that the position of the saddle point moves from a configuration
47 The critical nucleus is (Acr , Z cr ) ∼ (340, 131) if we use N ∼ 1.6Z . However, in more refined
estimates which go beyond the quadrupole deformation we used here the potential barrier by the
liquid-drop model almost disappears when the atomic number exceeds the critical value Z cr ∼ 100
[30].
with small deformation near a in Fig. 2.18 to a configuration near the scission point d
as the fissility parameter gets smaller starting from 1. If we take a simplification that
the scission point coincides with the saddle point when χ = 0, E f can be estimated
as
2/3 2 1/3
A Z A
E f = 2bsurf − bsurf A2/3 + 2bC
2 2 2
2 1/3
5 Z A
+ bC 2 − bC Z 2 /A1/3 . (2.72)
3 2 2
By taking the ratio to the surface energy, we have
Ef
≡ f (χ )
E surf

5 1
= (21/3 − 1) + 21/3 + − 2 χ
3 2 × 22/3
≈ 0.260 − 0.215χ . (2.73)
On the contrary, the saddle point appears in the region of small deformation when χ
is close to 1. Hence it makes sense to express the energy surface in the expansion with
respect to deformation parameters. It is, of course, needed to include terms beyond
the second order in order to determine the position and the height of the saddle point.
By keeping up to the fourth order with respect to α2 and second order concerning
α4 , we have

2 2 116 3 101 4 2 2
ΔE surf+C = bsurf A 2/3
α + α + α + α α4 + α42
5 2 105 2 35 2 35 2

Z2 1 2 64 3 58 4 8 5
− bC α2 + α2 + α2 + α22 α4 + α42 . (2.74)
A1/3 5 105 35 35 27
If we determine α4 for a given α2 by a variational principle, we obtain
243 2
α4 = − α , (2.75)
595 2
since there exists a coupling term between α2 and α4 . In consequence, when χ is
close to 1, the height of the potential barrier is given by
Ef 98 11368
≡ f (χ ) = (1 − χ )3 − (1 − χ )4 + · · · , (2.76)
E surf 135 34425
in units of the surface energy.

Fig. 2.23 The fissility parameter dependence of the height of the fission barrier plotted as the ratio
to the surface energy. After [29]
Figure 2.23 shows the behaviour of f (χ ) after the paper by Bohr and Wheeler
[29]. A line which interpolates Eqs. (2.73) and (2.76) is shown as f (χ ). The f ∗ (χ )
represents f (χ ) given by Eq. (2.73). The value of E f which is expected when the
surface energy for the shaded region around χ ∼ 0.75 is set to 530 MeV is shown
in the inset on the right corner, since the surface energy does not so change in that
region.48
The experimental value of E f can be deduced, for example, from the excitation
function of (n,f), (t,pf), (t,df), (d,pf), (γ ,f) reactions which are induced by bombard-
ing a target nucleus with either light nuclei or photons.49 As an example, Fig. 2.24
shows the excitation functions of the fission probabilities for the 239 U(n,f) and for
the 239 U(d,pf) reactions. They are denoted by the solid line and the dots, respectively.
Both fission excitation functions exhibit a characteristic threshold rise followed by a
flat plateau. A reasonable estimate of the barrier height can be obtained by identifying
it to the energy at which the fission probability becomes 1/2 of its plateau value. It
is because the tunneling probability or the barrier transmission probability becomes
1/2 when the decay energy E coincides with the height of the fission barrier, as can
be seen from the following formula for the barrier transmission probability in the
uniform approximation:
1
PUA = r> √ , (2.77)
1 + exp(2 r< dr 2μ(V (r ) − E)/)
which is valid from below to above the barrier as long as the fission barrier is a
quadratic function or can be well approximated by a quadratic function. In Eq. (2.77),
48 Bohr and Wheeler estimated the value of E f for the other nuclei by assuming (Z 2 /A)cr to be
47.8 from the experimental data that E f is about 6 MeV for 239 U.
49 The scattering cross section plotted as a function of the collision energy is called the excitation
function.
Fig. 2.24 The excitation

functions of the
induced-fission probabilities
of 239 U [27]. The solid line
is for (n,f) and the dots are
for (d,pf). The abscissa is the
excitation energy
Fig. 2.25 Fission barriers. 0.06

The data are taken from the Lu even-even
compilations in [27] 0.05 Ta odd-even
Os Ir odd-odd
0.04 Tl
Bi
Ef / Esurf
Hg
0.03 Po At
0.02
Liquid drop model Ra
Pa Np Am Bk
0.01
Th U Pu Cm Cf
0
0.6 0.65 0.7 0.75
χ = Z2/49A
r< and r> are the classical turning points on both sides of the potential barrier:
V (r< ) = V (r> ) = E, and μ is the reduced mass.50 The two turning points coincide
when the decay energy E matches the height of the fission barrier, so that the resulting
barrier transmission probability becomes 1/2.
Figure 2.25 shows the experimental values of E f (MeV) in units of the surface
energy, which has been assumed to be given by E surf = 17A2/3 MeV. The solid line
has been calculated by keeping only the first term on the right-hand side of Eq. (2.76).
exp.
Also, we presented in Table 2.6 the experimental values of E f together with the
BW exp.
predictions by Bohr and Wheeler E f and the binding energy of a neutron Bn in
the compound nucleus.
What is important in connection with the reactor, which we learn in Sect. 2.3.5, is
the sign of E f − Bn . If the sign is negative, fission is easily induced even if the energy
of the incident neutron is 0. On the other hand, if the sign is positive, an appropriate
50 In classical mechanics, a particle is reflected at the position where the total energy of the particle
becomes identical with the potential energy. Such positions are therefore called classical turning
points or simply turning points.
Table 2.6 The threshold energy for fission E f and the binding energy of neutron Bn , in units of
MeV. The former and the latter are taken from [27, 31], respectively
exp. exp. exp. exp.
Nucleus Ef Bn E f − Bn E BW
f
(compound
nucleus)
233 Th 6.4 4.8 1.6 6.9
232 Pa 6.3 5.6 0.7 5.5
235 U 5.8 5.3 0.5 5.0
236 U 5.9 6.5 −0.6 5.25
239 U 6.2 4.8 1.4 5.95
240 Pu 5.9 6.5 −0.6
amount of energy has to be supplied in order to induce fission with a large probability.
As we learn later, it is the reason why fission can be induced for 235 U by a thermal
neutron with an extremely small energy, while it is necessary to inject a neutron with
the energy of the order of MeV in order to induce fission of 238 U. The difference
between the two arises from the difference between neutron binding energies in the
corresponding compound nuclei, 236 U and 239 U. Because of the pairing correlation
appearing in the mass formula, the Bn for the former is larger by Δ than the value
estimated in the simple liquid-drop model without the correction term δ. On the
other hand, it gets smaller by Δ in the latter. In fact, the difference between the two
Bn (236 U) − Bn (239 U) = 1.7 MeV is close to 2Δ ≈ 2 × √12 238
∼ 1.6 MeV.51
(4) Spontaneous Fission Half-Lives Figure 2.26 shows the spontaneous fission half-
lives t1/2 for several actinides (U, Pu, Cm, Cf, Fm) as a function of the fissility
parameter χ (filled circles are for even–even nuclei, open circles are for even–odd
nuclei. Exactly speaking, the abscissa is slightly different from the χ defined in this
book, because a symmetry energy correction is added to the surface energy term in
the mass formula). Note that the experimental data lie by and large along a straight
line in the semi-logarithmic graph. The detailed structures come from the deviation of
the ground state masses from the prediction of the liquid-drop model because of the
shell effect. Indeed, if we denote the deviation of the ground state mass from that of
the smooth liquid-drop model by δm (MeV), then the values of log10 t1/2 (yr) + 5δm
beautifully align as a function of χ [32] as shown in Fig. 2.27.
The feature shown in Figs. 2.26 and 2.27 can be understood as follows from the
point of view of quantum tunneling. Let us denote the curvature of the potential
surface for fission in the vicinity of the potential pocket which yields the metastable
state by ω0 ≡ ( ∂∂rV2 /μ f )1/2 , where r is an appropriately chosen fission coordinate
2
and μ f is the corresponding mass parameter. Let us then approximate the potential
surface in the vicinity of the potential barrier by a quadratic function with curvature
Ωfb . Furthermore, assuming that fission occurs with energy much lower than the
51 The reactor must have become fairly different from the present one if there existed no pairing
correlation between nucleons.

Fig. 2.26 Spontaneous

fission half-lives of
even–even (filled circles) and
even–odd (open circles)
nuclides as a function of the
fissility parameter [27]
Fig. 2.27 Spontaneous

fission half-lives corrected
with shell correction [27]
height of the fission barrier, let us represent the tunneling probability in the stan-
dard WKB formula by ignoring 1 in the denominator of the formula in the uniform
approximation (2.77). The decay width for fission is then given by
ω0
Γf = 5 × Pfis (2.78)
2π
ω0 2π
∼ 5× exp − (E f − E) . (2.79)
2π Ωfb
In Eq. (2.78), following Bohr and Wheeler, we introduced the statistical factor 5
to represent the degeneracy of the quadrupole oscillation which induces fission.
Although E f is given by Eq. (2.76), it will be well represented as E f = a − bχ in
the linear approximation, since Figs. 2.26 and 2.27 are intended for a narrow region
of χ . Finally, we obtain
2π
log10 t1/2 ∼ A − B χ (2.80)
Ωfb
using two constants A and B. One can thus understand why a linear relationship
holds between the fission half-life and the fissility parameter in the semi-logarithmic
plot.
Exercise 2.14 Estimate the curvature of the fission barrier Ωfb for actinides from
Fig. 2.26 or 2.27.52
2.3.5 Application to Nuclear Power Generation
The nuclear power generation in operation today exploits the energy release, called
nuclear energy, due to fission. In this subsection, we learn the principles of reactor
and several key words.
2.3.5.1 Nuclear Reactions in Reactor, Principles of Reactor, Chain

Reaction, Neutron Moderator, Neutron Absorber
Figure 2.28 illustrates the nuclear reactions induced by injecting a neutron into a
mass of uranium which is a mixture of 235 U and 238 U. 235 U disintegrates into two
large nuclei with about half the atomic and mass numbers of 235 U (see Fig. 2.17),
and two to three neutrons on average.
The phenomenon shown in Fig. 2.28 can be understood as follows. As already
stated concerning the Q-value for fission, the nuclear energy of around 200 MeV is
Fig. 2.28 Illustration of

nuclear reactions in reactor
(chain reaction)
52 The information on both the potential surface and the effective mass is needed in order to theo-
retically evaluate the fission width. Nowadays, the potential surface can be obtained with relatively
high accuracy using, for example, the macroscopic–microscopic method, which refines the liquid-
drop model by adding the shell correction and will be described in Sect. 7.6.1. On the contrary, it
is difficult to estimate the mass parameter with high reliability. One can hope to obtain important
phenomenological information on the effective mass for fission by using the results of this exercise.
released per fission. Though most of the released energy is used as the kinetic energy
of the fission fragments, a part of it becomes their excitation energies. Roughly
speaking, the fission fragment loses its excitation energy by emitting one neutron
if the excitation energy amounts to 8 MeV, which is an average binding energy of
nucleon. The resulting nucleus is usually a neutron-rich unstable nucleus, so that it
continues to lose its energy by repeating β-decay. Since the β-decay proceeds in
obedience to selection rules [33], the resulting daughter nucleus is not necessarily in
the ground state, but sometimes is above the threshold energy for neutron emission.
In that case, the state decays by emitting a neutron. Since the β-decay usually takes
place in the time scale of the order of seconds to minutes, those neutrons emitted
after the β-decay are emitted with a time delay compared with the neutrons emitted
immediately after fission. Those neutrons are therefore called delayed neutrons.
In comparison, the neutrons emitted immediately after fission are called prompt
neutrons. In consequence, two large fragments and two to three neutrons are emitted
by the decay of U.53 The average number of neutrons emitted in the thermal-neutron-
induced fission of 235 U is 2.47 as stated before, although it fluctuates due to the
existence of various decay processes as shown in Fig. 2.17.
The emitted neutrons are absorbed by the surrounding 235 U, and successively
induce fission leading to a chain reaction.54 It is the principle of reactor to utilize
the nuclear energy by appropriately removing the emitted neutrons with an absorber
called control rod in order to keep the number of thermal neutrons in the space
constant, and thus by safely sustaining the chain reaction. A minimum condition is
needed for the amount of U fuel in order to sustain the chain reaction, which is called
the critical mass.
2.3.5.2 The Neutron Absorption Cross Section, Enriched Uranium,

and Neutron Moderator
The cross section of the neutron-induced fission of U strongly depends on the isotopes
and the energy of the incident neutron. Figure 2.29 shows the neutron-induced fission
cross section of 235 U and 238 U (cited from [33], see also [34]). The figure shows that
238
U does not cause fission unless the energy of neutron becomes as high as 1 MeV.
This is the reason why no fission of 238 U is seen in Fig. 2.28. On the other hand,
the energy of thermal neutron 1/40 eV is sufficient to induce fission of 235 U. The
difference between the threshold energies for the induced fission of 235 U and 238 U
originates from the pairing correlation between nucleons as stated before.
53 Most of the neutrons are the prompt neutrons which are emitted at times shorter than 4 × 10−14 s.
Though the ratio of the delayed neutron is small, it plays an important role in controlling nuclear
reactors [6, 33, 34].
54 It is Fermi and his collaborators who succeeded in the chain reaction of the induced fission of U
by a slow neutron for the first time, and the experiment was performed at the University of Chicago
on December 2, 1942.
Fig. 2.29 Neutron induced

fission cross section [33]
Because of this difference, the enriched uranium, which has the composition of
235
U with a larger percentage than that of the naturally occurring uranium 0.720%,
is used as the fuel for reactors.55
Figure 2.29 shows that the induced fission cross section of 235 U decreases nearly
proportionally to the inverse of the velocity of the incident neutron v as the energy
of neutron increases (1/v-law). We learn the reason in Sect. 9.3. On the other hand,
the energy spectrum of the neutrons emitted in the thermal neutron induced fission
of 235 U can be well represented as
√ √
N (E) ∝ E exp(−E/E 0 ) ∼ E exp(−E/1.29 MeV) (2.81)
using the laboratory energy E, and distributes with its maximum at near E = E 0 /2 ∼
0.65 MeV. Figure 2.30 shows the laboratory neutron kinetic energy spectrum for
thermal neutron-induced fission of 235 U. The average energy of the emitted neutrons
is Ē = 1.5E 0 ∼ 2 MeV.56 Although these fast neutrons can induce fission of 238 U,
the cross section is much smaller than that of the thermal neutron-induced fission
of 235 U as Fig. 2.29 shows. Moreover, the induced fission cross section of 235 U by
high-energy neutrons is small, so that the cross section becomes effectively small.
The neutron moderator is therefore introduced in reactors so as to reduce the kinetic
energy of neutrons to that of the thermal neutrons.57
Exercise 2.15 Summarize why enriched uranium is used for nuclear reactors.
Sidebar: Natural Nuclear Fission Reactor A number of uranium ores, whose

isotope ratio of 235 U is significantly smaller than the standard value 0.720% in the
55 Incommercial reactors, the enrichment of 235 U is typically about 3–4%.

56 The center of mass energy spectrum of neutrons emitted from a nucleus in the excited state with
temperature T is expected to be given by εn exp(−εn /T ), and is well described by the theory called
the evaporation theory. The dashed line in Fig. 2.30 represents the energy spectrum expected from
the evaporation theory transformed to that in the laboratory frame by taking into account the finite
velocity of the fission fragment.
57 The reactor which uses high energy neutrons, instead of neutrons which have been cooled down
to thermal neutrons due to the elastic scattering in the neutron moderator, is called fast reactor or
fast neutron reactor.
Fig. 2.30 Kinetic energy spectrum of neutrons (Laboratory system) [27]
naturally occurring uranium, have been discovered in the uranium mines at Oklo in
Gabon (Gabonese Republic), Africa. This suggests that there has been once natural
nuclear fission reactors in the uranium deposits at Oklo.
2.3.5.3 Nuclear Waste Problem: Transmutation
Many of the fission products are radioactive nuclei with finite lifetimes as written
in Fig. 2.28. In particular, the relatively long half-lives of 137 90
55 Cs and 38 Sr are 30.14
and 28.78 years, respectively, which are of the same order as the human life span. In
addition, transuranium elements such as 237 93 Np, whose half-life exceeds 200 million
years, are produced. In the nuclear power industry this is a serious nuisance. These
radioactivities cause a biological hazard in discarded nuclear waste for a long time.
It is of critical importance to handle these long-lived nuclear wastes, and the studies
of nuclear transmutation are going on.
2.3.6 Fission Isomers
At the end of this section, let us learn about fission isomers. Figure 2.31 shows the
excitation function of the neutron-induced fission of 230 Th, i.e., 230
90 Th(n,f). Unlike
Fig. 2.24, there appears a clear resonance structure with the width of 40 keV at the
incident energy of neutron of about 720 keV. Figure 2.32 compares the excitation
functions of the 230 230
90 Th(n,f) and 90 Th(d,pf) reactions, which have been obtained by
Fig. 2.31 The excitation

function of the neutron
induced fission of 230 Th,
showing a resonance. Taken
from [35]
Fig. 2.32 Detailed structure

of the excitation function of
two induced fissions of
230 Th. Taken from [36]
later experiments with high resolution. The abscissa of Fig. 2.32 is the excitation
energy of 231
90 Th.
The excitation function of fission will be a monotonic function if the potential
surface for fission has only a single barrier as predicted by the simple liquid-drop
model and as has been shown in Fig. 2.19. In reality, the potential surface has multiple-
barriers depending on the nucleus due to the shell correction energy representing the
shell effect. The situation is shown in Fig. 2.33. It shows how the understanding for
the structure of fission barrier has evolved with time.58
The structure seen in Fig. 2.31 was originally interpreted as the meta-stable β-
vibrational level produced in the second potential minimum. However, using the fine
58 One of the powerful methods to obtain the potential surface with multiple-potential barriers is the
macroscopic–microscopic method which modifies the liquid-drop model by adding shell correction
(see Sect. 7.6.1). In the fission of 231 Th, the first fission barrier is lowered by axial asymmetry. The
axial symmetry is recovered in the second fission barrier. However, the barrier height gets lower by
considering mass asymmetry, and furthermore there appears the third potential minimum.
Fig. 2.33 Time evolution of the theory for fission barrier, taken from [37]
structures seen in Fig. 2.32 as the experimental evidence, these structures are now
understood to be associated with the rotational motion involving parity doublet, i.e.,
the positive and negative parity rotational bands, which have almost the same moment
of inertia, yielded in the third potential minimum associated with the mass asymmet-
ric fission [36, 38] (see the right-bottom curves in Fig. 2.33). The position, width and
spin–parity of each resonance state shown in the figure have been determined by the
high-resolution measurement of the excitation function using, for example, the time
of flight (TOF), the measurement of the angular distribution of fission fragments, the
simultaneous measurements of (n,f) and (d,pf) reactions, and the theoretical analyses
by the parity-doublet rotational model. The ratio of the length of the long axis to that
of the short axis obtained from the level splitting is 3:1, and matches the deformation
at the third potential minimum.59
Figure 2.34 shows a few energy levels near the ground state of 238 U and metastable
states in the second potential well together with their spin and parity. The latter states
are called fission isomers.
The fission isomers are a kind of either superdeformed states, which we learn
in Sect. 7.6.2 and whose ratio of the length of the long axis to that of short axis is
about 1.6–1.8, or hyperdeformed states which have even larger ratio of the lengths
of two axes.60 As we learn in Chap. 7, a characteristic of deformed states is that there
59 The ratio of the length of the long axis to that of the short axis is 2:1 for the second potential
minimum.
60 The ratio of the length of the long axis to that of the short axis is about 1.3 for usual deformed
nuclei such as the ground state of U. In comparison, the moment of inertia of fission isomers is 2.2
Fig. 2.34 The structure of

238 U. Fission isomers. Taken
from [39]
Fig. 2.35 The structure of

240 Pu. Fission isomers.
Taken from [40]
appear a characteristic group of states called rotational bands. Figure 2.35 compares
the rotational bands built on the ground state and on a fission isomer by taking the
case of 240 Pu as an example. The large difference of the deformation between the
two bands can be deduced from the difference of the corresponding level spacing.
times that of the ground state for 238 U as shown in Fig. 2.34. 8 Be, which has the dumb-bell structure
of two touching α particles, has also 2:1 for the ratio of the length of the long axis to that of the
short axis, and hence is one of the hyperdeformed states.
References
1. E. Segré, From X-Rays to Quarks: Modern Physicists and Their Discoveries (W. H. Freeman
and Company, San Francisco, 1980)
2. M. Nogami, Nuclear Physics, Japanese edn. (Shoukabou, Tokyo, 1973)
3. T. de Forest Jr., J.D. Walecka, Adv. Phys. 15, 1 (1966)
4. B. Hahn, D.G. Ravenhall, R. Hofstadter, Phys. Rev. 101, 1131 (1956)
5. A. Tamii et al., Phys. Rev. Lett. 107, 062502 (2011)
6. W.N. Cottingham, D.A. Greenwood, An Introduction to Nuclear Physics, 2nd edn. (Cambridge
University Press, Cambridge, 2001)
7. R. Hofstadter, F. Bumiller, M.R. Yearian, Rev. Mod. Phys. 30, 482 (1958)
8. N. Takigawa, M. Ueda, M. Kuratani, H. Sagawa, Phys. Lett. B 288, 244 (1992)
9. J.S. Al-Khalili, J.A. Tostevin, Phys. Rev. Lett. 76, 3903 (1996)
10. B. Abu-Ibrahim, Y. Suzuki, Phys. Rev. C 61, 051601 (2000)
11. J. Kasagi, H. Yuki, T. Baba, T. Noda, T. Ohtsuki, A.G. Lipson, J. Phys. Soc. Jpn. 71, 2881
(2002)
12. Y. Kato, N. Takigawa, Phys. Rev. C 76, 014615 (2007). And references therein
13. G. Audi, A.H. Wapstra, C. Thibault, Nucl. Phys. A 729, 337 (2003)
14. R. Shurtleff, E. Derringh, Am. J. Phys. 57, 552 (1989)
15. T. Arima, Master thesis, Tohoku University
17. L. Alvarez et al., Phys. Rev. 105, 1127 (1957)
18. S.E. Jones, Nature 321, 127 (1986)
19. K. Ikeda, N. Takigawa, H. Horiuchi, Prog. Theor. Phys. Suppl. E68, 464 (1968)
20. K. Ikeda, T. Marumori, R. Tamagaki, H. Tanaka, J. Hiura, H. Horiuchi, Y. Suzuki, F. Nemoto,
H. Bando, Y. Abe, N. Takigawa, M. Kamimura, K. Takada, Y. Akaishi, S. Nagata, Prog. Theor.
Phys. Suppl. 52 (1972); K. Ikeda, R. Tamagaki, S. Saito, H. Horiuchi, A. Tohsaki-Suzuki, M.
Kamimura, Prog. Theor. Phys. Suppl. 62 (1977)
21. H. Horiuchi, K. Ikeda, K. Kato, Prog. Theor. Phys. Suppl. 192, 1 (2012)
22. C.M. Lederer, V. Shirley, Table of Isotopes, 7th edn. (Wiley, New York, 1978)
23. N. Takigawa, A. Arima, Nucl. Phys. A 168, 593 (1971)
24. P. Möller, J.R. Nix, W.D. Myers, W.J. Swiatecki, At Data Nucl. Tables 59, 185 (1995); P.
Möller, J.R. Nix, Nucl. Phys. A 536, 20 (1992); H. Koura, M. Uno, T. Tachibana, M. Yamada,
Nucl. Phys. A 674, 47 (2000); H. Koura, T. Tachibana, M. Uno, M. Yamada. Prog. Theor. Phys.
113, 305 (2005)
25. S. Yusa, private communication
26. K.F. Flynn, L.E. Glendenin, ANL-7749 (1970); cited in [27]
27. R. Vandenbosch, J.R. Huizenga, Nuclear Fission (Academic Press, New York, 1973)
28. O. Hahn, F. Strassmann, Naturwiss. 27, 11 (1939)
29. N. Bohr, J.A. Wheeler, Phys. Rev. 56, 426 (1939)
30. F. Ivanyuk, K. Pomorski, Phys. Rev. C 79, 054327 (2009); Int. J. Mod. Phys. E 19, 514 (2010)
31. A.H. Wapstra, G. Audi, Nucl. Phys. A 432, 1 (1985)
32. W.J. Swiatecki, Phys. Rev. 100, 937 (1955)
33. K. Yagi, Nuclear Physics, Japanese edn. (Asakura, Tokyo, 1971)
34. W.S.C. Williams, Nuclear and Particle Physics (Clarendon press, Oxford, 1991)
35. G.D. James, J.E. Lynn, L.G. Earwaker, Nucl. Phys. A 189, 225 (1972)
36. J. Blons, Nucl. Phys. A 502, 121c (1989)
37. P. Möller, A.J. Sierk, T. Ichikawa, A. Iwamoto, R. Bengtsson, H. Uhrenholt, S. Åberg, Phys.
Rev. C 79, 064304 (2009)
38. P. Möller, J.R. Nix, in Proceedings of the Third IAEA Symposium on the Physics and Chemistry
of Fission, Rochester, 1973, Vol. I (International Atomic Energy Agency, Vienna, 1974), p. 103
39. R. Vandenbosch, Nucl. Phys. A 502, 1c (1989)
40. V. Metag, D. Habs, H.J. Specht, Phys. Rep. 65, 1 (1980); The original paper is H.J. Specht, J.
Weber, E. Konecny, D. Heunemann. Phys. Lett. B 41, 43 (1972)
Chapter 3
The Nuclear Force and Two-Body Systems
Abstract The strong interaction between nucleons which is called the nuclear force
plays the central role in nuclear physics. Compared with the Coulomb interaction
between two charged particles, which plays the exclusive role in condensed matter
physics, the nuclear force is fairly complicated reflecting the fact that nucleons are
composite particles and have spin and isospin degrees of freedom. There are several
different approaches to uncover nuclear force. The one is to consider the most general
possibility under the constraints from the consideration of symmetry or transforma-
tion invariance. The other is a phenomenological approach. The meson theory is also
an attractive historical approach, which is often combined with phenomenological
approaches. Recently, approaches based on QCD are also introduced. A difficulty
of nuclear force is that the effective force inside nucleus is fairly different from the
bare interaction between two nucleons in free space. In this chapter we overview the
basic properties and the current understanding of the nuclear force.
3.1 The Fundamentals of Nuclear Force
3.1.1 The Range of Forces—A Simple Estimate

by the Uncertainty Principle
The four forces in nature, i.e., the gravitational force, the electromagnetic force, the
weak interaction and the strong interaction, can be thought to originate from exchang-
ing gauge particles, i.e., graviton, photon, weak bosons (W± and Z0 particles) and
mesons (or gluons at the more fundamental level of the quantum chromodynamics
(QCD)), which mediate each force, between two particles. Following this idea, one
can relate the range of each force to the mass of the corresponding gauge particles
using the uncertainty principle.
If we denote the mass of a gauge particle by m, then the uncertainty of the energy
associated with the propagation of the gauge particle is ΔE ∼ mc2 . Hence, the
corresponding uncertainty of time is Δt ∼ /ΔE ∼ /mc2 . Assuming that the gauge
particle moves with the light velocity during that time interval, the range of the force
d can be estimated to be
DOI 10.1007/978-4-431-55378-6_3
66 3 The Nuclear Force and Two-Body Systems

d ∼ c × Δt ∼ , (3.1)
mc
and coincides with the Compton wavelength of the gauge particle. As is well known,
the electromagnetic force reaches infinite distance, because the mass of photon is
zero. On the other hand, the range of the weak interaction is very short, since the
masses of W± and Z0 particles are large. If we assume that the strong interaction is
mediated by exchanging mesons as we learn later in the meson theory, the range of
the force which originates from the exchange of the lightest meson, i.e., π -mesons,
can be estimated to be about 1.4 fm by using that the rest energy of the π -mesons is
about 140 MeV. Incidentally, although the graviton has not yet been discovered, its
mass is expected to be zero, because the gravitational force has an infinite range like
the electromagnetic force.
3.1.2 The Radial Dependence
Though we made a rough estimate of the range of forces using the uncertainty
principle in the previous section, the strength of the nuclear force acting between
two particles varies as a function of the distance between them similarly to that the
strength of the Coulomb interaction gets weaker in inverse proportion to the square
of the distance r between the two charged particles as their distance increases.
Corresponding to the fact that the strength of the Coulomb force is proportional
to 1/r 2 , the Coulomb potential is given by VC (r ) = Q 1 Q 2 e2 /r , where Q 1 e and Q 2 e
are the electric charges of the two charged particles. This formula can be derived by
thinking that one of the charged particles produces an electric field through interaction
with the electromagnetic field, and then exerts a force on the other charged particle
via the electric field. The equation which determines the electric field in this case is
known as the D’Alembert equation.
The same idea can be applied to the strong interaction. If we denote the field of
the strong interaction by φ(r, t), then the equation called Klein–Gordon equation,
which is a generalization of the D’Alembert equation for the electric force, is given
by
1 ∂2 mc 2
∇2 − 2 2 − φ(r, t) = −4πρ(r, t) , (3.2)
c ∂t
where m is the mass of the gauge particle. It is the mass of a meson in the meson theory
for the nuclear force. There appears on the left-hand side the Compton wavelength
of the gauge particle because of the finiteness of the mass of the gauge particle. The
ρ(r, t) appearing on the right-hand side represents the density at the spatial position
r at time t of the charged particles which create the electric force in the D’Alembert
equation for the Coulomb potential. It is the density of hadrons such as nucleons,
which are the sources of mesons or gluons, in the case of the strong interaction.
In the static limit to ignore the time variation, the solution of Eq. (3.2) is given by
3.1 The Fundamentals of Nuclear Force 67

e−mc|r−r |/
φ(r, t) ≈ ρ(r , t) dr . (3.3)
|r − r |
Exercise 3.1 Derive Eq. (3.3) by the following two methods 1 and 2.
1. The method of Fourier transformation
a. Show that the Fourier transformation of φ(r) is given by
4π
φ̃(k) = ρ̃(k, t) . (3.4)
k2 + (mc/)2
b. Derive Eq. (3.3) by making the inverse Fourier transformation of φ̃(k).

2. The Green function method
Define the Green function in the static case by
mc 2
∇ −
2
G N (r − r ) = −4π δ(r − r ) , (3.5)

and then show that it becomes

e−mc|r−r |/
G N (r − r ) = . (3.6)
|r − r |
Exercise 3.2 Prove the following equation.
1
∇2 = −4π δ(r − r ) . (3.7)
|r − r |
The nuclear potential φ N (r ) can be obtained by letting ρ(r ) = ρ0 δ(r ) in the

integral on the right-hand side of Eq. (3.3) in the limit of ignoring the size of the
particle such as a nucleon which is the source of the force. Here, we took the position
of the hadron which is the source of the force as the origin of the spatial coordinate. We
can see that the distance dependence changes from the 1/r law for the electric force
to e−μr /r with μ = mc/. This functional form including the exponential damping
factor is called the Yukawa type, and the strength of the nuclear force becomes
e−1 ≈ 1/2.7 in the distance of 1/μ. This distance corresponds to the range of the
force which we obtained in the previous section and coincides with the Compton
wavelength of the gauge particle.
3.1.3 The State Dependence of Nuclear Force
In the following, we discuss the interaction between two nucleons, called nuclear
force, among the strong interactions. Since each nucleon has the spin degree of
freedom, and also the degree of freedom in the isospin space, which has been touched
in Sect. 1.2, in addition to that in the usual space, the nuclear force is much more
complex than the Coulomb force, and causes a great variety of phenomena. In this
section, we learn the classification of the states of two-nucleon systems, the exchange
operators and the projection operators in preparation for studying more details about
the nuclear force.
3.1.3.1 The Classification of the States of Two-Nucleon Systems
The wave function of two-nucleon systems ψ can be expressed as the product of the
wave functions in the usual space, the spin space and the isospin space as
ψ(x1 , x2 ) = ϕ(r)ζ (s1 , s2 )η(t1 , t2 ) (3.8)
if there is no interaction between the coordinates in the usual and spin spaces such
as the spin–orbit interaction which we will learn in Sect. 3.2.2. The s and t are the
z-components of the coordinates in the spin and isospin spaces, the lower indices
1 and 2 refer to the first and the second nucleons, r ≡ r1 − r2 is the coordinate for the
relative motion, r1 and r2 the spatial coordinates of the first and the second nucleons,
respectively. Also, x is used to represent the coordinates in the usual, spin and isospin
spaces altogether simultaneously, i.e., x = (r, s, t).
If one uses the simultaneous eigenstates of the square Sˆ2 of the total spin
Ŝ ≡ ŝ1 + ŝ2 and its z-component Ŝz , |S, Sz , then the wave function in the spin space
ζ is one of the following states,
|1, 1S = α1 α2 , (3.9)

1
|1, 0S = √ [α1 β2 + β1 α2 ] , (3.10)
2
|1, −1S = β1 β2 , (3.11)
1
|0, 0S = √ [α1 β2 − β1 α2 ] . (3.12)
2
We ignored the coordinates on the left-hand side, but used the index S to explicitly
indicate that it is the wave function in the spin space. On the right-hand side, we
expressed the coordinates s1 and s2 simply with the lower index 1 and 2. The α
and β represent the spin-up and spin-down states, | 21 21 and | 21 − 21 , respectively.
The states with S = 1 and S = 0 are symmetric and antisymmetric concerning the
two nucleons, respectively, and does not change and changes the sign when the two
nucleons are interchanged. The former and the latter are called the spin-triplet and
the spin-singlet states, respectively.
Similarly, the wave function in the isospin space η can be expressed using the fol-
lowing simultaneous eigenstates of the square of the total isospin and the magnitude
of its z-component,
|1, 1I = n1 n2 , (3.13)

1
|1, 0I = √ [n1 p2 + p1 n2 ] , (3.14)
2
|1, −1I = p1 p2 , (3.15)
1
|0, 0I = √ [n1 p2 − p1 n2 ] , (3.16)
2
and each state can be classified into either isospin-triplet or isospin-singlet states.
The lower index I means isospin space. As defined in Eq. (1.3), n and p represent
neutron (isospin-up) and proton (isospin-down) states, respectively.
On the other hand, the wave function for the relative motion can also be classified
into either symmetric or antisymmetric states with respect to the exchange of two
nucleons. Since the exchange of two particles corresponds to the parity transforma-
tion (or the parity inversion) of the coordinate of the relative motion r, the former
state is called E-state in the sense of even, while the latter O-state in the sense of
odd state. For example, if the wave function ϕ(r) is given by ϕ(r) = R (r )Ym (θ, φ)
by using the spherical harmonics Ym , then the states with an even number for the
orbital angular momentum such as the s-wave are E-states, while those with odd-
values such as the p-wave are O-states.
Let us now represent the state for two-nucleon systems as 2T +1 2S+1 E or 2T +1 2S+1 O
when the total spin and the total isospin are S and T . The symbols E and O represent
the symmetry property, i.e., the even–odd property, of the wave function of the
relative motion. Since nucleons are fermions, the wave function of two-nucleon
systems has to change sign for the simultaneous coordinate interchange in the whole
space. Namely, ψ(x2 , x1 ) = −ψ(x1 , x2 ). Hence only the following four states, 13 E,
31
E, 11 O, 33 O, are allowed for two-nucleon systems. For example, 13 E is called singlet
triplet even state. A characteristic of the nuclear force is that it significantly differs
in these four states.
3.1.3.2 The Exchange Operators
(1) The Spin-Exchange Operator Various exchanges take place when two nucleons
interact. For example, when a proton in the spin-up state collides with another proton
in the spin-down state, the orientation of the spin is interchanged between them with a
certain probability (spin exchange). This happens because the nuclear force contains
terms which are proportional to the inner product σ̂ 1 · σ̂ 2 of the spins of the two
nucleons. We can rewrite the inner product as
1
σ̂ 1 · σ̂ 2 = (σ̂1+ σ̂2− + σ̂1− σ̂2+ ) + σ̂1z σ̂2z (3.17)
2
by using the raising and lowering operators of the spin. The spin exchange processes
are caused by the first and the second terms on the right-hand side.
It is convenient to introduce the following operator in order to describe the spin

exchange effect,
1
Pσ = (1 + σ̂ 1 · σ̂ 2 ) . (3.18)
2
One can easily show that Pσ |1MS = |1MS (M = 1, 0, −1) and Pσ |00S =
−|00S by operating Pσ on the spin-triplet and spin-singlet states. Since the
spin-triplet and spin-singlet states are symmetric and antisymmetric regarding the
exchange of the spins of two nucleons, this means that Pσ has the same function
as the exchange of the spins of the two nucleons. For this reason, Pσ is called the
spin-exchange operator, alternatively the Bartlett operator.
(2) The Charge-Exchange Operator The exchange of isospin can also occur in two-
nucleon systems. For example, a proton and a neutron are interchanged when they
collide (charge exchange). In order to describe this phenomenon, it is convenient to
introduce the following charge-exchange operator, which is also called the isospin-
exchange operator or the Heisenberg operator, by using the isospin operator τ of
each nucleon,
1
Pτ = (1 + τ̂ 1 · τ̂ 2 ) . (3.19)
2
(3) The space-Exchange Operator (Majorana Operator) One can also consider
the operator which interchanges the coordinates of two nucleons in the usual space.
It is denoted by PM ,
PM ϕ(r) = ϕ(−r) . (3.20)
Since the wave function for two-nucleon systems has to be antisymmetric in whole
space,
PM Pτ Pσ ψ = −ψ . (3.21)
Hence,
PM Pτ Pσ = −1 . (3.22)
One can therefore write

PM = −Pτ Pσ . (3.23)
3.1.3.3 The Projection Operators
As we see later, the nuclear force strongly depends on the states of two nucleons
through the σ̂ 1 · σ̂ 2 and τ̂ 1 · τ̂ 2 terms in the nuclear force.
One often uses the projection operators defined by
1 1
Πst ≡ (3 + σ̂ 1 · σ̂ 2 ) , Πss ≡ (1 − σ̂ 1 · σ̂ 2 ) , (3.24)
4 4
1 1
Πit ≡ (3 + τ̂ 1 · τ̂ 2 ) , Πis ≡ (1 − τ̂ 1 · τ̂ 2 ) , (3.25)
4 4
in order to explicitly show the state dependence of the nuclear force. The indices
st, ss, it and is represent to project on the spin-triplet states, spin-singlet state,
isospin-triplet states and isospin-singlet state, respectively.
Exercise 3.3 Show that Πst satisfies the following conditions which are required
for the projection operator on the spin-triplet states.
2 = Π , Π Π = 0 , Π |1M
Πst st st ss st S = |1MS , Πst |00S = 0 . (3.26)
3.2 The General Structure of Nuclear Force
One of the methods to construct nuclear force or nuclear potential is to take into
account various conservation laws suggested by fundamental experiments and to
consider the most general structure of the nuclear potential under those constraints.
Each conservation law is related to the invariance or symmetry under an associated
transformation. For nuclear force, one can first list up that
• the potential is a scalar quantity,
• it is invariant under space reflection, i.e., Parity invariant,
• it is time reversal invariant
as the basic properties. Furthermore, the proton–proton and proton–neutron scatter-
ings at low energies suggest that the nuclear force is the same for the three states of
isospin triplets to a high accuracy. This is called the charge independence and implies
that the nuclear force is a function of the inner product of the isospins of two nucleons
τ̂ 1 · τ̂ 2 . On the other hand, the way of thinking that the force between two neutrons
and that between two protons are the same is called the charge symmetry. The charge
symmetry and the charge independence are suggested also from the facts that the
energy levels of mirror nuclei, e.g., 157 N8 and 158 O7 , very much resemble except for
the energy shift due to the Coulomb energy, and that there appear corresponding
energy levels among isobars such as 148 O6 , 147 N7 and 146 C8 .1
In this section, we give the general form of the nuclear potential based on the
considerations of the invariance or the symmetry.
3.2.1 Static Potentials (Velocity-Independent Potentials)
Let us first consider the potentials, called static potentials, which do not depend on
the velocity of the relative motion between the nucleons.
1 A group of energy levels which have the same J π but differ in the z-component of the isospin form
an isobaric multiplet. For example, there exist many isobaric triplets for 14 14 14
8 O6 , 7 N7 , and 6 C8 .
3.2.1.1 Central Potentials
The static potentials can be classified into the central potentials, which depend only
on the distance r between the nucleons, and the noncentral potentials, which depend
on the angle of the relative coordinate r. The most general central static potential
can be written as
VC = V0 (r ) + (σ 1 · σ 2 )Vσ (r ) + (τ 1 · τ 2 )Vτ (r ) + (σ 1 · σ 2 )(τ 1 · τ 2 )Vσ τ (r )

(3.27)
by considering the symmetry and invariance conditions mentioned at the beginning of

this section. The strength and the functional form of V0 (r ), Vσ (r ), Vτ (r ) and Vσ τ (r )
cannot be determined by the consideration of symmetry alone. Equation (3.27) can
be rewritten as
VC = VW (r ) + VM (r )PM + VB (r )Pσ + VH (r )Pτ , (3.28)
where the indices W, M, B and H stand for Wigner, Majorana, Bartlett, and
Heisenberg, respectively, or
VC = Vst (r )Π̃st + Vts (r )Π̃ts + Vtt (r )Π̃tt + Vss (r )Π̃ss (3.29)
by using the exchange and projection operators. Here, Π̃st is the projection oper-
ator on the isospin-singlet, spin-triplet states and is given by Π̃st ≡ Πis × Πst .
Π̃ts , Π̃tt , Π̃ss are defined in a similar way. The radial functions in the different rep-
resentations are transformed as
VW = V0 − Vσ − Vτ + Vσ τ ,
VM = −4Vσ τ ,
VB = 2Vσ − 2Vσ τ ,
VH = 2Vτ − 2Vσ τ , (3.30)
and
Vts = VW + VM − VB + VH = V0 − 3Vσ + Vτ − 3Vσ τ ,

Vst = VW + VM + VB − VH = V0 + Vσ − 3Vτ − 3Vσ τ ,
Vss = VW − VM − VB − VH = V0 − 3Vσ − 3Vτ + 9Vσ τ ,
Vtt = VW − VM + VB + VH = V0 + Vσ + Vτ + Vσ τ . (3.31)
Exercise 3.4 Show the relationships (3.30) and (3.31).

3.2 The General Structure of Nuclear Force 73
3.2.1.2 Noncentral Potentials
One can make an operator which satisfies the invariant conditions for nuclear force
in the following way,
3(r · σ 1 )(r · σ 2 )
S12 = − σ1 · σ2 , (3.32)
r2
by using the relative coordinate r between the nucleons and the spin of each nucleon.
Equation (3.32) can be represented in the form,
S12 = (24π )1/2 [[σ1(1) × σ2(1) ](λ=2) × Y2 (r̂)](0)

0 , (3.33)
by using the notation of the tensor product. The S12 is therefore called the tensor
operator. Using the total spin of two-nucleon system,
1
S= (s1 + s2 ) , (3.34)
2
S12 can be rewritten also as

(Ŝ · r)2 2
S12 =2 3 2
− Ŝ . (3.35)
r
Equation (3.35) shows that S12 has a finite value only when it is operated on the
spin-triplet states. Hence one can represent the noncentral potential as

VT = VTeven (r )Π (T = 0, L even ) + VTodd (r )Π (T = 1, L odd ) S12 , (3.36)
where Π (T = 0, L even ) and Π (T = 1, L odd ) are the projection operators on the

isospin 0, even parity states and on the isospin 1, odd parity states, respectively.
This force is called the tensor force. As the form of Eq. (3.32) shows, the tensor
force resembles the force acting between two magnetic dipoles. We learn later that
the tensor force is responsible for binding the deuteron among various two-nucleon
systems, and plays a vital role in the existence of nuclei, e.g., in locating the stable
nuclei along the diagonal line in the nuclear chart.
Exercise 3.5 Show Eqs. (3.33) and (3.35).
3.2.2 Velocity-Dependent Potentials
A representative of the velocity-dependent potentials is the spin–orbit interaction,

whose corresponding potential is given by
even
VLS = VLS (r )Π (T = 0, L even ) + VLS

odd
(r )Π (T = 1, L odd ) L̂ · Ŝ . (3.37)
Here, L̂ is the angular momentum operator of the relative motion, and is given by
L = r × p = r × (p1 − p2 ) . (3.38)
As we learn later, the spin–orbit interaction plays an important role in nuclear physics,
e.g., in governing the magic numbers.
3.3 The Properties of Deuteron and the Nuclear Force
3.3.1 The Effect of Tensor Force: The Wave Function

in the Spin–Isospin Space
Of two-nucleon systems, only deuteron is stable. Neither proton–proton nor neutron–

neutron bound state exists. In that sense, the deuteron is the simplest stable nucleus in
nature. The detailed analyses of the experiments concerning deuteron are therefore
important in obtaining various information on the nuclear force together with the
analyses of the nucleon–nucleon scatterings, which we learn in the next section.
We show in Table 3.1 the properties of the deuteron in comparison with those of
the proton and the neutron.
As we learn in detail in Sect. 3.6 (for example, see Fig. 3.5), the central force is
attractive in distant regions for all two-nucleon systems. Moreover, it is stronger in
the spin-singlet state, to which proton–proton and neutron–neutron systems belong,
than in the spin-triplet states of deuteron. Nevertheless, only deuteron exists as a
stable state in nature. This is because even an attractive force cannot yield a stable
Table 3.1 Properties of the proton, the neutron and the deuteron. u: Unified atomic mass unit;
Q: Electric quadrupole moment. Taken from [1]. Values of the magnetic and electric quadrupole
moments for deuteron are taken from [2, 3], respectively
Proton Neutron Deuteron
Electric charge e 0 e
Mass (u) 1.00727646688(13) 1.0086654(4) 2.01410219(11)
Mass (MeV) 938.272029(80) 939.550(5) 1875.61282(16)
Binding energy (MeV) 2.22452(20)
Spin 1/2 1/2 1
Mean life >1031 to 1033 years 887.5 ± 0.8 s Stable
Magnetic moment (μ N ) 2.792847351(28) −1.9130427(5) 0.8574382308(72)
Q (e fm2 ) 0 0 0.285783(30)
3.3 The Properties of Deuteron and the Nuclear Force 75
Fig. 3.1 Model experiment

of the tensor force. Toys
courtesy of Gyo Takeda
state unless it is strong enough in the three dimensional world governed by quantum
mechanics, and also because an extra attractive force leading to a stronger attraction
as a whole is added to deuteron due to the tensor force, since the spin of the deuteron
is 1. In order to demonstrate the mechanism why only deuteron is stable among two-
nucleon systems, Fig. 3.1 shows an assembly of toys of magnetic dipoles floating on
the surface of oil. The directions of the magnetic dipoles shown by the arrows align
due to the interaction between the magnetic dipoles when two toys approach close
enough. Deuteron becomes stable for a similar reason. Since the nuclear force is of
short range, the relative motion between the two nucleons is desirable to be in the
s-wave which has a finite probability at zero distance in order to be stabilized by
the nuclear force. On the other hand, the wave function of two-nucleon systems has
to be antisymmetric in the usual, spin and isospin spaces as a whole with respect to
the exchange of two nucleons. Consequently, only deuteron which is in the isospin-
singlet state (spin-triplet states) and can supplement the insufficiency of the central
force by the tensor force becomes to exist in nature as a stable system.
Here, let us examine what information on the effects and magnitude of the ten-
sor force can be obtained from the magnetic and electric quadrupole moments of
deuteron, although the binding energy is an important quantity in determining the
nuclear force [4].
The operator of the magnetic moment of deuteron is given by (see Sect. 4.1.1),
μ̂d = μp σ̂ p + μn σ̂ n + L̂p , (3.39)
where Lp is the orbital angular momentum of proton and is given by Lp = L/2 by

using the angular momentum of the relative motion between proton and neutron L.
All the magnetic moments are given in units of the nuclear magnetic moment μ N .
The μˆd can be rewritten as

1
μ̂d = μ p + μn Ĵ − μ p + μn − L̂ (3.40)
2
by using the total angular momentum of deuteron J ≡ L + 21 (σ p + σ n ) and L.2 As

we learn in Sect. 4.3.2, the magnetic moment is defined as the expectation value of
the z-component of μ̂ in the case when the z-component of the angular momentum
M takes its maximum value M = J ,
μ = J M = J |μ̂z |J M = J . (3.41)
Since the operator of the magnetic moment is a tensor of rank 1, the value of the
magnetic moment becomes 0 and disagrees with the experimental data if one assumes
that the spin of deuteron is 0. For finite J , the projection theorem (see Appendix A.6.4)
gives,
1
μ= J M = J |μ̂ · Ĵ|J M = J , (3.42)
J +1
The spin of deuteron is 1 as shown in Table 3.1. This is due to the existence of the
tensor force, which mixes the L = 2 state (D-state) to the L = 0 state (S-state) for
the relative motion as Eq. (3.33) implies. If we denote the amplitude of the mixing
by α D , then the wave function of deuteron can be formally expressed as
|ψ D = |13 S1 + α D |13 D1 , (3.43)
where the lower index 1 shows that the magnitude of the total angular momentum J
is 1. Inserting Eq. (3.43) into the state |J M = J in Eq. (3.42), we obtain

3 1
μ = μn + μp − μn + μp − PD , (3.44)
2 2
where PD = |α D |2 represents the probability of D-state in deuteron.

The fact that the magnetic moment of deuteron is nearly equal to the sum of the
magnetic moment of the proton and that of the neutron in Table 3.1 supports that the
spin of deuteron is 1, that is, the spins of the proton and the neutron in the deuteron are
aligned. Moreover, the difference between them suggests that the deuteron contains
the D-state by about 4 %.
The admixture of the D-state means that the deuteron is deformed by that amount.
We learn in Chap. 7 that the quadrupole moment of each nucleus gives a measure
of the deformation of the nucleus. The admixture of the D-state suggested by the
magnetic moment is consistent with the fact that the quadrupole moment is finite in
Table 3.1.
Exercise 3.6 Derive Eq. (3.44).
2 We used that σ p − σ n becomes 0 when it is operated on the |S = 1, Sz = 1 state.

3.3 The Properties of Deuteron and the Nuclear Force 77
3.3.2 The Radial Wave Function: Estimate of the Magnitude

of the Force Between Proton and Neutron
Let us next examine the radial wave function between proton and neutron. For sim-
plicity, we assume that the orbital angular momentum is zero. If we ignore the differ-
ence between the neutron and proton masses and represent the radial wave function
as R0 (r ) = u(r )/r , then the Schrödinger equation for u(r ) is given by
d2 u MN
+ 2 [E − V (r )] u(r ) = 0 , (3.45)
dr 2
where M N is the nucleon mass. If we further assume that the potential is a square
well potential of depth V0 and range a, then the equation can be analytically solved
and its solution is given by

A sin K r for r < a
u(r ) = (3.46)
Be−γ r for r ≥ a ,
√
M N (V0 − W )
K = , (3.47)
√
MN W
γ = , (3.48)

where W ∼ 2.22 MeV is the binding energy. From the continuity condition at r = a,
we obtain
K cot K a = −γ . (3.49)
Exercise 3.7 The assessment of the depth of the potential.

1. If we assume V0 W , then K a ≈ π/2 from Eq. (3.49). Using this, derive
π 2 2
V0 a 2 ≈ . (3.50)
4M N
2. Show that V0 ∼ 50 MeV by assuming that the range of the force nearly equals
the Compton wavelength of pions: a ≈ /m π c ∼ 1.45 fm.
The wave function
u = C e−γ r − e−αr (3.51)
is also often used for the wave function given by Eq. (3.46), and is called the wave
function of Hulthén-type [5]. The normalization constant is given by

3γ
C ≈ 2γ 1+ (3.52)
2α
if one assumes α γ . The wave function of the Hulthén-type has a desirable prop-
erty that it satisfies the boundary condition that it is zero at the origin.
3.4 Nucleon–Nucleon Scattering
The neutron–proton and proton–proton scattering data provide important information

on the nuclear force. The experimental data are given as the differential cross sections
and scattering cross sections. More detailed information on the spin dependence can
be obtained if one uses polarized projectile or polarized target [6]. Here, we discuss
the properties of nuclear force obtained from the scattering data of unpolarized beam
and target nucleus.
3.4.1 Low-Energy Scattering: Effective Range Theory
Let us consider the scattering of a neutron by a proton in order to avoid special

considerations needed to handle the long range Coulomb interaction. When both
neutron and proton are not polarized and their spins randomly orient all the directions
with equal probability, the state of two nucleons takes the spin-singlet state and the
spin-triplet states with the statistical weight 1:3. The experimentally obtained cross
section is therefore given by
1 (0)
σ = σ + 3σ (1) (3.53)
4
if we denote the cross sections in the spin-singlet state and in the spin-triplet states
by σ (0) and σ (1) , respectively.
Exercise 3.8 Show Eq. (3.53).
Exercise 3.9 Explain experimental methods to determine the cross section such as
the direct measurement and the measurement of the absorption cross section.
Let us study how the information on the nuclear force can be obtained from the
analysis of the scattering cross section. In general, the differential cross section is
given by
dσ
= | f (θ )|2 (3.54)
dΩ
by using the scattering amplitude f (θ ), where θ is the scattering angle. As we explain
in Appendix A.1, one of the standard methods to analyze the experimental data of
scattering is to make the partial wave expansion of the wave function and perform
the phase shift analysis. In this method, the scattering amplitude f (θ ) is given by
3.4 Nucleon–Nucleon Scattering 79
∞
1
f (θ ) = (2 + 1)(e2iδ − 1)P (cos θ ) (3.55)
2ik =0
by using the phase shift δ for each partial wave and the Legendre function P (cos θ ).
k is the incident wave number.
In the case of scattering by a short range force, only the s-wave contributes to
the scattering if the incident energy of the scattering is sufficiently low. Namely, the
phase shift can be set equal to zero to a good approximation for the partial waves
≥ 1. This can be understood by noting that the angular momentum is given as the
product of the incident momentum and the impact parameter. If the incident energy is
low, the impact parameter becomes large for the partial waves ≥ 1, hence becomes
outside of the range of the nuclear force. More important thing is that the s-wave
phase shift can be represented as
1 1
k cot δ = − + k 2 re (3.56)
a 2
by using two parameters irrespective of the details of the force (effective range theory
[7]). In other words, one can determine, at best, only two parameters from the data
of low-energy scattering. The parameters a and re are called the scattering length
and the effective range, respectively.
Exercise 3.10 Discuss the range of the incident energy for which only the s-wave
contributes to the scattering by assuming that the range of nuclear force is about 2 fm.
We obtain
4πa 2
σ = (3.57)
(1 − re ak 2 /2)2 + (ak)2
by inserting the result of Eq. (3.56) into the relationship between the cross section
and the phase shift. We can obtain the information on the scattering length and the
effective range by experimentally studying in detail the change of the cross section
at low energies as a function of the incident energy. An alternative way to determine
the scattering length with high precision is to use the cross sections of the scattering
of a thermal neutron by a proton, and by a parahydrogen molecule. Since the energy
of the thermal neutron3 is as low as 1/40 eV, the cross section is given by

σthermal-neutron = π 3(a (1) )2 + (a (0) )2 (3.58)
by taking the limit of k → 0. On the other hand, the cross section of the scattering
of the thermal neutron by a parahydrogen molecule4 is given by [8, 9]
σparahydrogen = π(3a (1) + a (0) )2 . (3.59)
3 Neutronswhich are in thermal equilibrium after multiple scattering inside matters of the normal
temperature.
4 Hydrogen molecule consisting of two protons, where two proton spins aligned antiparallel.
Table 3.2 Nuclear force parameters determined from low-energy scattering and the properties of
deuteron, taken from [11]
Scattering system (state a (fm) re (fm)
indices)
Expt. Argonne v18 Expt. Argonne v18
p + p (T = 1, S = 0) −7.8063 ± 0.0026 −7.8064 2.794 ± 0.014 2.788
n + p (T = 1, S = 0) −23.749 ± 0.008 −23.732 2.81 ± 0.05 2.697
n + n (T = 1, S = 0) −18.5 ± 0.4 −18.487 2.80 ± 0.11 2.840
n + p (T = 0, S = 1) 5.424 ± 0.003 5.419 1.760 ± 0.005 1.753
One can determine a (1) and a (0) by using the observed values of these cross sections.
Exercise 3.11 Prove Eq. (3.59).
The following relation between the binding energy of deuteron, scattering length
and the effective range,
1 1
≈ α − re(1) α 2 , (3.60)
a (1) 2
can be used in order to further determine the effective range for the spin-triplet states.
Here, α = [(M N /2 )ε D ]1/2 , ε D being the binding energy of deuteron.
Table 3.2 shows the scattering length and the effective range obtained from the
analyses of the low-energy scattering and deuteron [10–14]. It can be shown on the
way of derivation of the effective range theory, i.e., of Eq. (3.56) that there exist
bound states if the sign of the scattering length a is positive, while there exists no
bound state if it is negative. Hence the results shown in Table 3.2 show that only
the isospin 0, spin-triplet states, i.e., deuteron, is stable, i.e., a bound state, of two-
nucleon systems. In this way, the nuclear force has a strong state dependence. Also, it
should be remarked that the scattering length is nearly equal in all the proton–proton,
neutron–neutron and neutron–proton scatterings which belong to the isospin-triplet
states (T = 1) (charge independence of the nuclear force).
The effective range theory tells that only two parameters on the nuclear force can
be determined from low-energy scattering data. However, it is convenient to have an
explicit expression of the functional form and the strength of the potential in order
to perform various calculations. Here we mention that there exist also such attempts.
Exercise 3.12 Assume that the nucleon–nucleon potential is given by the sum of
the central (VC ) and the tensor (VT ) forces of the Yukawa-type, which have equal
weight of Wigner and Majorana terms,
V (r ) = VC (r ) + VT (r ) , (3.61)
e−μr e−r/aC
VC (r ) = V0 (w + m PM ) = V0 (w + m PM ) , (3.62)
μr r/aC
e−r/aT
VT (r ) = V0T S12 (w + m PM ) , (3.63)
r/aT
w = 1 − m = 0.5 . (3.64)
Show that the parameters should be taken as
V0 = −48.1 MeV, V0T = 23.1 MeV, aC = 1.17 fm, aT = 1.74 fm, (3.65)
in order to reproduce the experimental values of the effective range and the scattering
length in the spin-singlet and spin-triplet states.
3.4.2 High-Energy Scattering: Exchange Force
Equation (3.30) shows that the spin and isospin dependence of the nuclear force leads
to exchange forces. The existence of those exchange forces manifests itself in the
differential cross section of nucleon–nucleon scattering at high energies. Figure 3.2
shows the angular distribution of neutrons scattered by protons in the neutron energy
range 85–105 MeV. It is remarkable that a large amount of scattering takes place to
backward angles as well as to forward angles.
Fig. 3.2 The differential cross section of neutron–proton scattering, taken from [15]
Let us now assume that the nuclear force contains only the Wigner force. Since one
can use the Born approximation for high-energy scattering, the scattering amplitude
is given by
MN ∞
f W (θ ) = − 2 VW (r )r sin qr dr , (3.66)
q 0
with
q = 2k sin(θ/2) , (3.67)
to the first order. M N is the nucleon mass and q the wave number corresponding
to the momentum transfer. If we assume the Gauss-type for the form factor of the
nuclear force,
r2
VW (r ) = −VW 0 exp − 2 , (3.68)
RW
then we have √
MN π 3 (q RW )2
f W (θ ) = VW0 R exp − . (3.69)
2 q 4 W 4
Equation (3.69) shows that the cross section should monotonically decrease towards
backward angles if the nuclear force consists of the Wigner force alone. This con-
tradicts the experimental data shown in Fig. 3.2.
In order to resolve this contradiction, let us assume that the nuclear force contains
a Majorana term
VM (r )PM , (3.70)
and examine its effect in the first-order Born approximation. Since the scattering
amplitude is linearly proportional to the nuclear force in the first-order Born approx-
imation, it gets an additional term,
∞
MN
f M (θ ) = − VM (r )r sin q r dr (3.71)
2 q 0
with
q = ki + k f , (3.72)
q = 2ki sin[(π − θ )/2] , (3.73)
due to the Majorana force. If we assume the Gaussian-type for VM (r ) in the same
way as that for the Wigner force, the differential cross section increases with the
scattering angle. One can thus explain the large cross section at backward angles
appearing in the experimental data.5
5 The Serber force, which is also called the Serber exchange force, is one of such forces. It consists
of the Wigner and the Majorana terms of the same sign with equal weight and is assumed to be
proportional to 1 + PM (see Table 3.4).
3.4.3 High-Energy Scattering: Repulsive Core
Another characteristic of the nuclear force is that it becomes strong repulsion in

the close distance where two nucleons strongly overlap. This repulsive force in the
short distance is often called the repulsive core. As can be seen from the phase shift
analysis, the sign of the potential and the sign of the phase shift δ have a one to one
correspondence if the sign of V (r ) is fixed independently of the distance r , i.e., δ is
positive for attractive potentials and negative for repulsive potentials.
Figure 3.3 shows the behaviour of the phase shift for several partial waves as a
function of the incident energy. Now, let us focus on the 31 S0 channel (T = 1, S = 0,
L = 0 channel). The figure indicates that the potential for this channel behaves effec-
tively as an attractive potential in the scattering at low energies, while as a repulsive
potential in the scattering at high energies. If we consider the uncertainty relation
between the distance and the momentum, this implies that the nuclear potential
behaves as a repulsive potential at short distances, while as an attractive potential at
long distances. Figure 3.4 represents this property in a simple way by combining a
square potential barrier and a square well potential.
In the following, we show that this simple model can well explain the energy
dependence of the phase shift, and that it gives an estimate of the extension of the
repulsive core. We first note that the phase shift for the partial wave of the angular
momentum is given by
∞
Mk
δ (k) = − V (r ) j2 (kr )r 2 dr . (3.74)
2 0
This equation confirms the following relation between the attractive or repulsive
property of the potential and the sign of the phase shift
T=1 T=0
60
1 3
S0 S1
3 1
P0 P1
40 3
P1 3
Phase shift (degree)
D1
3
P2
20
-20
0 100 200 300 0 100 200 300 400

E lab (MeV) E lab (MeV)
Fig. 3.3 The phase shifts for nucleon–nucleon scattering in several channels. Made from the analy-
sis in [16]
Fig. 3.4 A simple model for V(r)

the nuclear force (repulsive V1
core plus short range
attraction)
c a
0
r
b
–V0
δ > 0 for V (r ) < 0 , (3.75)

δ < 0 for V (r ) > 0 , (3.76)
which we mentioned before for the case when the sign of the potential is fixed.
The s-wave phase shift is given by

M sin(2kc) sin(2ka)
δ0 = − (cV1 − bV0 ) − c(V1 + V0 ) + aV0 , (3.77)
22 k 2kc 2ka
when the potential is given by Fig. 3.4. Here b is defined as
b =a−c . (3.78)
Let us assume now that

b
V1 > V0 and a c, (3.79)
c
and study the sign of the phase shift in the scattering at low and high energies. In the
scattering at low energies, where the conditions 2kc 1, 2ka 1 hold, we obtain
MaV0
δ0 (k) ≈ >0, (3.80)
22 k
by ignoring the last term of Eq. (3.77) and approximating as sin(2kc)/2kc ≈ 1.

Equation (3.80) shows that the phase shift is positive. On the other hand, in the
scattering at high energies, where the conditions 2kc 1, 2ka 1 hold, we obtain
M
δ0 (k) ≈ − (cV1 − bV0 ) < 0 , (3.81)
22 k
by ignoring the second and third terms in Eq. (3.77). Thus the phase shift becomes
negative. The radius of the repulsive core c is given by 2k̄c ≈ 1 if we denote the wave
cr
number which corresponds to the critical energy E c.m. where the phase shift becomes
zero by k̄. E c.m. ∼ 125 MeV from Fig. 3.3, so that one can estimate c ≈ 0.3 fm, though
cr
more accurate studies give c = 0.4–0.5 fm.

3.4.4 Spin Polarization Experiments
Since 1990s, experiments with polarized incident nucleons and/or target nucleus
became popular. Those experiments are very useful to obtain information on the
spin-dependent forces.
Suppose that the force between two nucleons is given by
V = f (r ) (V0 + Vσ σ 1 · σ 2 ) . (3.82)
If we define the spin raising and lowering operators by
σ± ≡ σx ± iσ y (3.83)
as is often done in the calculations concerning angular momentum, then the inner
product of the spins can be rewritten as
1
σ1 · σ2 = (σ1+ σ2− + σ1− σ2+ ) + σ10 σ20 . (3.84)
2
The spin-flip, i.e., the exchange of direction of spins between the incident and target
nucleons, takes place through the first and the second terms on the right-hand side
of Eq. (3.84). Hence one can expect that detailed information on the spin-dependent
terms in the nuclear force can be obtained from the experiments related to the spin
polarization.
Exercise 3.13 Study methods to polarize the incident particle or the target nucleus.
Exercise 3.14 Consider the proton–neutron scattering by the following potential,
V = (1 + κ σ̂ 1 · σ̂ 2 ) f (r ) , (3.85)
and assume that the proton and the neutron are in the spin-up and spin-down states,
respectively, at the beginning of the scattering. Obtain the probability of exchanging
the direction of spins between the proton and neutron at the end of the scattering
as a function of the strength of the spin-dependent term κ in the first-order Born
approximation.
3.5 Microscopic Considerations: Meson Theory, QCD
In the meson theory, the force between two nucleons is assumed to work through
the exchange of various mesons such as π -mesons, ω-mesons and ρ-mesons. In this
section, we discuss the properties of the nuclear force from this point of view.
Let us first derive the force which arises from the exchange of π -mesons. The first
step is to properly postulate a Lagrangian density. Since the π -meson field obeys the
Klein–Gordon equation, the corresponding Lagrangian is given by
1 1
Lpion = ∂μ Π a ∂ μ Π a − m 2π Π a Π a ,
2 2

1 ∂ 2Π a ∂ 2Π a ∂ 2Π a ∂ 2Π a 1
= − − − − m 2π Π a Π a . (3.86)
2 ∂t 2 ∂x2 ∂ y2 ∂z 2 2
Here, we used the natural unit ( = c = 1). Π is the π -meson field and the upper
index a is the index in the isospin space. The a = 1, 2, 3 correspond to the x, y, z-
components, respectively, and Π 1 , Π 2 , Π 3 are related to the fields of three charge
states π + , π 0 , π − of π -mesons as
√ √
π + = (Π 1 + iΠ 2 )/ 2, π − = (π + )∗ = (Π 1 − iΠ 2 )/ 2, π 0 = Π 3 . (3.87)
On the other hand, the Lagrangian density for the interaction between nucleons and
pion fields are given by
Lint
NR
= gπ ϕ † σα τ a ϕ∇α Π a , (3.88)
in the non-relativistic approximation by considering that it is a scalar in the combined

spaces of the spin and usual space coordinates, it is scalar in the total isospin space
including both nucleons and pions, and that π -mesons are pseudoscalar particles.
ϕ is a two-dimensional spinor representing the nucleon field, σ and τ are the spin
and isospin operators of nucleons, respectively. Corresponding to Eq. (3.88), the
interaction Hamiltonian is given by

H =− gπ ϕ † σα τ a ϕ∇α Π a dr . (3.89)
It is necessary to determine the meson fields which mediate the force in order to
determine the interaction between two nucleons. They are determined by the Euler–
Lagrange equation,
∂ ∂L ∂L
− =0, (3.90)
∂ x μ ∂ (∂qi /∂ x μ ) ∂qi
where qi represents the meson field Π and the nucleon field ϕ in a unified way. We
obtain the equation,

(∇ 2 − m 2π )Π a (x) = gπ ∇α ϕ † σα τ a ϕ , (3.91)
for the π -meson field by applying the Euler–Lagrange equation to q = Π a when

the Lagrangian density is given by Eqs. (3.86) and (3.88). We introduced the static
approximation and ignored the term of time derivative of the π -mesons. Note that
this equation has the same form as Eq. (3.2). The solution can be found as

gπ e−m π |r−r | †
Π a (r) = − dr
∇α ϕ (r )σα τ a ϕ(r ) (3.92)
4π |r − r |
3.5 Microscopic Considerations: Meson Theory, QCD 87
in the same way as that used to derive Eq. (3.3). Inserting this result into Eq. (3.89),
we have

gπ2 e−m π |r−r | †
H= drdr ϕ † (r)σα τ a ϕ(r)∇α ∇ [ϕ (r )σβ τ a ϕ(r )] . (3.93)
4π |r − r | β
The interaction potential can be obtained as
δ2 H
V (r1 , σ 1 , τ 1 , r2 , σ 2 , τ 2 ) =
δρ(r1 )δρ(r2 )
g2 e−m π |r1 −r2 |
= − π (τ 1 · τ 2 )(σ 1 · ∇ 1 )(σ 2 · ∇ 2 ) (3.94)
4π |r1 − r2 |
through the functional derivative with respect to the density. In deriving Eq. (3.94),
we treated the nucleon field in the mean field approximation by identifying ϕ † (r)ϕ(r)
δρ(r)
with the mean nucleon density ρ(r) of c-number, and used the rule δρ(r 1)
= δ(r − r1 )
for the functional derivative. By performing the differentiation, Eq. (3.94) can be
rewritten as
V OPEP (r = r1 − r2 , σ1 , τ1 , σ2 , τ2 )

g2 e−μr 3 3
= π m π c2 (τ1 · τ2 ) (σ1 · σ2 ) + 1 + + S12
3c μr μr (μr )2
4π g 2
− 3 π m π c2 (τ1 · τ2 )(σ1 · σ2 )δ(r) , (3.95)
μ 3c
where μ = m π c/. The index OPEP of V OPEP means one-pion exchange potential.
Note that there appear spin- and isospin-dependent terms as well as a term which
corresponds to the tensor force.
Exercise 3.15 In the relativistic theory, the Lagrangian density for the interaction
between the nucleon field and the π -meson field is given by
Lint
R
= −gπ ψ̄γ5 γμ τ a ∂ μ Π a ψ . (3.96)
Show that Eq. (3.88) can be obtained from Eq. (3.96) by introducing non-relativistic
approximations.
Equation (3.95) shows that the range of the nuclear force due to the exchange
of one pion is given by the Compton wavelength of π -meson and that the radial
dependence, i.e., the form factor, is given by the Yukawa function e−μr /μr . These
results agree with the conclusions obtained in Sects. 3.1.1 and 3.1.2 based on the
uncertainty principle and by making reference to the electromagnetic field. In general,
the range of the force due to the exchange of a meson of mass m is of the order
of the corresponding Compton wavelength /mc. Hence the contribution of the
exchange of heavier mesons such as ρ-meson, whose mass (more precisely the rest
energy) is 775.5 MeV, ω-meson, whose mass is 782.7 MeV, and σ -particles, whose
exchange corresponds to the exchange of two π -mesons, becomes non-negligible as
the distance between two nucleons decreases.
The meson theory has difficulty in discussing the nuclear force in the short range
region where the repulsive core appears. Another approach, called the quark clus-
ter model, has been developed instead. In this approach, the nucleon is thought to
consist of three quarks and the properties of nuclear force are studied microscopi-
cally by taking into account the effects of antisymmetrization among quarks [17].
We wish to mention that the study of nuclear force based on the chiral perturbation
theory (χ PT) as an approximate theory (effective field theory) for QCD [18], and
more recently, the study of nuclear force using the Monte Carlo calculation of QCD
defined on the lattice (Lattice QCD calculation) [19] are also going on as alternative
approaches.
3.6 Phenomenological Potential with High Accuracy:

Realistic Potential
In parallel with the theoretical derivation of the nuclear force from microscopic
points of view, attempts to determine semi-phenomenological nuclear forces which
carry quantitative accuracy as well as high applicability have also been developed
by combining the phenomenology and meson theory. Historically, the Hamada–
Johnston potential [20] and the Reid hard-core and soft-core potentials [21],6 and
later, as refined potentials, the Bonn potential [12], the Argonne potential [11], the
Paris potential [13], and the Nijmegen potential [14] are known as the representatives
of such semi-phenomenological potentials.
These potentials have common features that they have a repulsive core and are
described by OPEP in the distant region and contain many parameters. The parame-
ters are determined so as to reproduce the experimental data of the nucleon–nucleon
scattering at laboratory energies below 350 MeV, and the properties of deuteron. In
that sense, they are often called realistic potentials. Each potential differs, e.g., in the
strength of the tensor force, the treatment of the medium and short distances, non-
locality, and off-shell effects. Table 3.3 summarizes the key issues of determination
as well as the properties of these realistic phenomenological potentials. As Table 3.3
shows, the refined realistic potentials, i.e., Paris, Argonne v18 , Bonn and Nijmegen
potentials, take into account the charge independence breaking (CIB)7 or the charge
6 We name the potentials whose height of the repulsive core is infinite and finite the hard- and the
soft-core potentials, respectively.
7 One of the main reasons of the CIB is the difference of the masses of three π -mesons.
3.6 Phenomenological Potential with High Accuracy: Realistic Potential 89
Table 3.3 Properties of several realistic phenomenological potentials: Core: property of the repul-
sive core. Inputs: information used to determine the potential parameters; ere: expansion coefficients
in the effective range theory, d: properties of deuteron, t: properties of triton, pp: proton–proton
scattering data, np: neutron–proton scattering data, nn: neutron–neutron scattering data. E max : Max-
imum energy of the nucleon–nucleon scattering used to determine the potential, in units of MeV.
Components: C: central force. T: tensor force. LS: spin–orbit force. LL: L2 term. (LS)2 : Second-
order spin–orbit force. Characteristics: p 2 -dep. C: square-of-momentum dependent central force,
CIB: charge independence breaking, CD: charge dependence, EM: include detailed electromagnetic
potential
Name Core Inputs E max Components Characteristics
Hamada–Johnston Hard ere, d, pp, np 315 C, T, LL, LS, (LS)2
Reid hare-core Hard ere, d, pp, np 350 C, T, LS
Reid soft-core Softa ere, d, pp, np 350 C, T, LS
Paris (Soft) d, ere, NN 330 C, T, LS, (LS)2 p 2 -dep. C
Argonne v18 Softb ere, d, pp, np, nn 350 C, T, LL, LS, (LS)2 CIB, EM, local
Bonn (Soft) d, t, pp, np 350 CD, nonlocal
Nijmegen Soft pp, np 350 C, T, LS, (LS)2 CD, local
a Yukawa type
b Woods–Saxon type
dependence (CD) of the nuclear force8,9 and are used in accurate calculations of
few-body systems and in the study of nuclear structure from microscopic point of
view.
Here, we explain the Hamada–Johnston and Reid potentials which have relatively
concise expressions and are convenient to see the bulk properties of the nuclear force.
3.6.1 Hamada–Johnston Potential
Figure 3.5 shows the Hamada–Johnston potential. The Reid hard core potential
resembles the Hamada–Johnston potential qualitatively.
The Hamada–Johnston potential postulates a repulsive core of infinite height for
the short distance inside of r = rC = 0.49 fm, and assumes that the potential is given
by
V = VC (μr ) + VT (μr )S12 + VL S (μr )L · S + VL L (μr )L 12 (3.97)
in the region outside of the core. Here, the operator of the spin-orbit interaction of
the second order L 12 is defined as
8 The details of the treatment and the accuracy depend on the potential. For example, the Nijmegen
potential takes into account only a part of the charge symmetry breaking (CSB), while the Argonne
potential v18 handles the CSB phenomenologically.
9 The accurate treatment of the CSB of the nuclear force is important in explaining the mass difference
between triton and 3 He, and more in general, the Nolen–Schiffer (NS) anomaly [22] concerning
the energy difference between the corresponding energy levels of mirror nuclei.
singlet triplet even triplet odd

100
VC, even VC VC
VLL, even VT VT
50 VC, odd VLS VLS
VLL, odd VLL VLL
V (MeV)
-50
-100
0 1 2 0 1 2 0 1 2
r (fm) r (fm) r (fm)
Fig. 3.5 Hamada–Johnston potential [20]. The figure has been made based on Eq. (3.99) through
(3.106) and by using the values of parameters given in [23]
1
L 12 = (σ 1 · σ 2 )L2 − [(σ 1 · L)(σ 2 · L) + (σ 2 · L)(σ 1 · L)]
2
= (δ L J + σ 1 · σ 2 )L2 − (L · S)2 , (3.98)
and the radial functions are assumed to be given by

VC = V0 (τ 1 · τ 2 )(σ 1 · σ 2 )Y (x) 1 + aC Y (x) + bC Y 2 (x) , (3.99)

VT = V0 (τ 1 · τ 2 )(σ 1 · σ 2 )Z (x) 1 + aT Y (x) + bT Y 2 (x) , (3.100)

VL S = g L S V0 Y (x) [1 + b L S Y (x)] ,
2
(3.101)
Z (x)
VL L = g L L V0 2 1 + a L L Y (x) + b L L Y 2 (x) , (3.102)

x
with
gπ2
V0 =m π c2 = 3.65 MeV, (3.103)
3c
mπ c
x = μr = r, (3.104)

1
Y (x) = e−x , (3.105)
x

3 3
Z (x) = 1 + + 2 Y (x) . (3.106)
x x
3.6 Phenomenological Potential with High Accuracy: Realistic Potential 91
3.6.2 Reid Potential
In the Reid hard-core potential, the outside region of the repulsive core is assumed
to be given by
V = VC (μr ) + VT (μr )S12 + VL S (μr )L · S , (3.107)
and the radial dependence is parametrized as

∞
e−nx
VC (x) = an ,
n=1
x
∞

b1 1 1 1 −x k 1 −kx e−nx
VT (x) = + + 2 e − + 2 e + bn ,
x 3 x x x x n=2
x
∞
e−nx
VL S (x) = cn . (3.108)
n=1
x
The position of the hard core rC depends on the total isospin of two nucleons, and
is taken as rC = 0.42 fm for the 1 S state and 0.43 fm for the other states in the case
when T = 1, and rC = 0.55 fm for the T = 0 state. The Reid soft-core potential
represents the effect of repulsive core by using a Yukawa function of short-range
repulsive type.
The parameters aC , . . . , b L L in the Hamada–Johnston potential and an , bn , cn in
the Reid potential are determined phenomenologically through the analyses of the
experimental data.
3.7 Summary of the Nuclear Force in the Free Space
Let us here summarize the nuclear force in the free space, i.e., the nuclear force
obtained from the analyses of the nucleon–nucleon scattering and of deuteron.
Figure 3.6 shows how various nuclear forces such as the central and tensor forces
vary with the distance r between two nucleons. They are shown separately for each
state of the two-nucleon systems such as 31 E and 11 O.10 The figure shows also the
mesons which contribute to the nuclear force in each spatial region.
The main characteristics of the nuclear force appearing in the figure can be sum-
marized as follows.
1. The nuclear force is of short range in contrast with the gravitational force and the
electromagnetic force. Note that the unit of the abscissa is 10−15 m.
10 The even–odd property of the spatial motion and the spin multiplicity are shown in the figure.
The triplet and singlet in 3 E and 1 E represent the multiplicity in the spin space.
Fig. 3.6 Summary of the nuclear potential, taken from [24]
2. The long distance region is governed by the one-pion exchange potential (OPEP),
and is attractive.
3. It turns to a strong repulsion in the short distance, where two nucleons overlap.
4. It strongly depends on the state of two-nucleon systems through, e.g., the tensor
force. For example, of the two-nucleon systems, it is only deuteron, whose total
isospin is 0 and total spin is 1, for which the attraction is strong enough to yield
a bound state.
3.8 Effective Interaction Inside Nucleus
3.8.1 G-Matrix
Further considerations are necessary in order to use for the studies of nuclear structure
and nuclear reactions even if the bare nuclear force, i.e., the nuclear force acting
between two nucleons in isolation, i.e., in the free space, is known. Although there
exist some variations depending on each potential, as we learnt in the previous section,
the short range part of the nuclear force is represented either by a repulsive core of
infinite strength or a strong repulsive core even if it is not infinite. If we use the
3.8 Effective Interaction Inside Nucleus 93
Fig. 3.7 Conceptual representation of the G-matrix. The wavy line and the dotted line represent G
and the bare potential V , respectively
bare nuclear force with a hard core as it is, e.g., in the mean-field theory, which is a
standard theory for studying nuclear structure, we get into trouble of the divergence
of the matrix elements of the interaction.
One of the methods to circumvent the difficulty associated with the fact that the
nuclear force is strong, e.g., the problem of hard core, is the G-matrix theory, in other
words the Brueckner theory [25]. In this method, one uses not the bare nuclear force,
but the effective nuclear force called G-matrix, which takes into account the effects
of multiple scattering and Pauli principle, as the interaction between nucleons in
performing the theoretical calculations of nuclear structure and nuclear reactions.11
Figure 3.7 conceptually represents the G-matrix. In the case of the free space, the
effective interaction which takes into account the multiple scattering is called the
T -matrix, and is given by
1
T =V +V T , (3.109)
E − K 1 − K 2 + iε
in the operator representation when the total energy of the two-nucleon system is E.
By repeating the insertion of the whole right-hand side to the T on the right-hand
side of Eq. (3.109), one will be able to understand that T is the effective interaction
which takes into account multiple scattering. K 1 , K 2 are the kinetic energy operators
of the nucleons 1 and 2 in the intermediate states.
The G-matrix can be thought to be the extension of the T -matrix to the nuclear
force inside nucleus. The important changes from the case of free space are to take
into account the Pauli exclusion principle so as to exclude the states which are
already occupied by the other nucleons from the intermediate states of nucleons
in the multiple scattering, and to use the energy in the medium in calculating the
propagator. Consequently, the G-matrix is given by the following equation,
Q
G =V +V G, (3.110)
E− H1(0) − H2(0)
where Q is the operator to restrict the intermediate states by following the Pauli
exclusion principle. H1(0) and H2(0) are the unperturbed Hamiltonian, and correspond,
e.g., to the Hamiltonian of the mean field in the shell model. Equation (3.110) is called
the Bethe–Goldstone equation.
11 The G-matrix theory is not the standard Rayleigh–Schrödinger perturbation theory, but a pertur-
bation theory which takes into account special higher-order terms, i.e., the ladder scattering.
We give here the matrix representations corresponding to Eqs. (3.109) and (3.110)
in order to make the meaning of the word matrix clear. In the case of T -matrix, it is
given by
TkE1 k2 ,k1 k2 = Vk1 k2 ,k1 k2

1
+ Vk1 k2 ,p1 p2 TE (3.111)
p1 p2
E− (p21 /2M) − (p22 /2M) + iε p1 p2 ,k1 k2
by using the wave number vector specifying the plane wave as the state index. k1 , k2
and k1 , k2 are the wave number vectors of the incoming and outgoing nucleons,
respectively. In the case of the G-matrix, using, e.g., the state index of the shell
model as the matrix index, we have
1
E
G cd,ab = Vcd,ab + Vcd,mn GE . (3.112)
mn>ε F
E − εm − εn mn,ab
Leaving the details of the G-matrix theory to other books [25], here we remark
that the effective interaction inside nucleus becomes density dependent through the
Fermi energy εF which appears on the right-hand side of Eq. (3.112) in order to
restrict the intermediate states (see Sect. 2.2 as for the relation between the Fermi
energy and the density).
3.8.2 Phenomenological Effective Interaction
As we have hitherto studied, it is not easy to determine the nuclear force exactly,
and to perform the study of nuclear structure and nuclear reactions from the point
of view of many-body problems based on the nuclear force determined from more
fundamental points of view such as the meson theory and QCD. The studies using
handy phenomenological interactions have therefore also been developed in parallel
with the studies based on the practical phenomenological nuclear force with high
accuracy described in Sect. 3.6 and those based on the G-matrix theory. Those phe-
nomenological potentials consist of several effective terms such as a strong exchange
term (Majorana term) to guarantee the saturation property of density, and/or the term
which takes the density dependence into account referring to the G-matrix theory.
Here, we mention some examples of those phenomenological effective interactions.12
12 As is clear from the projection operator method [26] by Feshbach. The effective interaction used
in a theoretical calculation depends on the size of the model space which is explicitly taken into
account in that calculation. The unitary model operator approach (UMOA) is one of such theoretical
approaches [27].
3.8 Effective Interaction Inside Nucleus 95
Table 3.4 Exchange property of the nuclear force. Each column gives the coefficient of each
exchange term
Name of force Wigner Majorana Bartlett Heisenberg
Serbera 0.5 0.5 0.0 0.0
Rosenfeld −0.13 0.93 0.46 0.26
a It well explains the large backward cross sections in the neutron–proton scattering
3.8.2.1 Examples of the Radial Form Factor and Exchange Property
The radial form factor of the nuclear force is often assumed to be either the Yukawa-
type or the Gauss-type as described below, or a sum of those terms with different
range parameters,
e−μr
V (r ) = −V0 (Yukawa potential), (3.113)
μr
V (r ) = −V0 e−r /r02
2
(Gauss potential). (3.114)
As for the exchange property, we show the cases of Serber and Rosenfeld forces in
Table 3.4 as representative examples.
3.8.2.2 Effective Interactions Used for Variational Calculations

of Light Nuclei
Here, we mention the Volkov interaction [28], which is often used for the study of
cluster structure of light nuclei by variational methods, as an example of the effective
interaction,13

V (r ) = −V1 e−(r/r1 ) + V2 e−(r/r2 ) (1 − m + m PM ) ,
2 2
(3.115)
where V1 = V2 = 60 MeV, r1 = 1.80 fm, r2 = 1.01 fm. The Majorana exchange term
is used to guarantee the saturation of nuclei in the Volkov force. Regarding the cluster
structure of light nuclei, larger values of m predict more distinct cluster structure.
13 The Brink–Boeker force [29], the Hasegawa–Nagata force [30], and the Minnesota force [31]
have also been often used. These forces are common in assuming the radial factor of Gaussian type
so that the calculations of the matrix elements using the wave functions of the harmonic oscillator
model can be easily performed. The Hasegawa–Nagata force has a characteristic that it contains the
spin–orbit force and the tensor force. Except for the Minnesota force, they determine the parameters
so as to reproduce the radii and the relative binding energies of some light nuclei. The Minnesota
force determines the parameters so as to reproduce the scattering data. The characteristics of various
effective interactions and the way to calculate the matrix elements of the Hamiltonian when the
wave function is given by a Slater determinant are described in detail in [32].
3.8.2.3 Effective Interactions Used for Hartree–Fock Calculations
Besides those forces, there exist the Skyrme force and the Gogny force as the effective
interactions which are often used for the non-relativistic Hartree–Fock calculations.
We describe the former in detail in Chap. 6.
References

2. P.J. Mohr, B.N. Taylor, D.B. Newell, Rev. Mod. Phys. 84, 1527 (2010)
3. M. Pavanello, W.-C. Tung, L. Adamowicz, Phys. Rev. A 81, 042526 (2012)
4. G.E. Brown, A.D. Jackson, The Nucleon-Nucleon Interaction (North-Holland, Amsterdam,
1976)
5. L. Hulthén, K.V. Laurikainen, Rev. Mod. Phys. 23, 1 (1951)
6. K. Kubo, K. Katori, Spin and Polarization, Japanese edn. (Baifukan, Tokyo, 1994)
7. L.I. Schiff, Quantum Mechanics (McGraw-Hill Kogakusha, Tokyo, 1968)
8. M. Nogami, Nuclear Physics, Japanese edn. (Shoukabou, Tokyo, 1973)
9. N.A. Jelly, Fundamentals of Nuclear Physics (Cambridge University Press, Cambridge, 1990)
10. R. Machleidt, Nucl. Phys. A 689, 11c (2001); L. Jäde, H.V. von Geramb, Phys. Rev. C 57, 496
(1998); K. Amos et al. Adv. Nucl. Phys. 25, 276 (2002)
11. R.B. Wiringa, V.G.J. Stoks, R. Schiavilla, Phys. Rev. C 51, 38 (1995); R.B. Wiringa, R.A.
Smith, T.L. Ainsworth. Phys. Rev. C 29, 1207 (1984)
12. R. Machleidt, K. Holinde, Ch. Elster, Phys. Rep. 149, 1 (1987); J. Haidenbauer, K. Holinde,
Phys. Rev. C 40, 2465 (1989); R. Machleidt. Adv. Nucl. Phys. 19, 189 (1989)
13. M. Lacombe, B. Loiseau, J.M. Richard, R. Vinh Mau, J. Côté, P. Pirès, R. de Tourreil, 21, 861
(1980)
14. M.M. Nagels, T.A. Rijken, J.J. de Swart, Phys. Rev. D 17, 768 (1978); P.M.M. Maessen,
T.A. Rijken, J.J. de Swart, Phys. Rev. C 40, 2226 (1989); V.G.J. Stoks, R.A.M. Klomp, C.P.F.
Terheggen, J.J. de Swart. Phys. Rev. C 49, 2950 (1994)
15. R.H. Stahl, N.F. Ramsey, Phys. Rev. 96, 1310 (1954)
16. R.A. Arndt, M.H. MacGregor, Phys. Rev. 141, 873 (1966)
17. K. Yazaki, Prog. Theor. Phys. Suppl. 91, 146 (1987)
18. R. Machleidt, D.R. Entem, J. Phys. Conf. Ser. 20, 77 (2005)
19. N. Ishii, S. Aoki, T. Hatsuda, Phys. Rev. Lett. 99, 022001 (2007); S. Aoki, T. Hatsuda, N. Ishii,
Comput. Sci. Disc. 1, 015009 (2008); Prog. Theor. Phys. 123, 89 (2010)
20. T. Hamada, I.D. Johnston, Nucl. Phys. A 34, 382 (1962)
21. P.V. Reid, Ann. Phys. 50, 411 (1968)
22. J.A. Nolen, J.P. Schiffer, Ann. Rev. Nucl. Sci. 19, 471 (1969)
24. R. Tamagaki, Frontiers of Physics, vol. 15 (Kyouritsu, Tokyo, 1986). Japanese edition
25. G.E. Brown, Unified Theory of Nuclear Models and Forces (North-Holland, Amsterdam, 1967)
26. G.R. Satchler, Direct Nuclear Reactions (Clarendon Press, Oxford, 1983)
27. S. Fujii, R. Okamoto, K. Suzuki, Phys. Rev. Lett. 113, 182501 (2009); K. Suzuki, R. Okamoto,
BUTSURI 42, 263 (1987); K. Suzuki, S.Y. Lee, Prog. Theor. Phys. 64, 2091 (1980)
28. A.B. Volkov, Phys. Lett. 12, 118 (1964); Nucl. Phys. 74, 33 (1965)
29. D.M. Brink, E. Boeker, Nucl. Phys. 91, 1 (1967)
30. A. Hasegawa, S. Nagata, Prog. Theor. Phys. 45, 786 (1971)
31. D.R. Thompson, Y.C. Tang, Phys. Rev. 159, 806 (1967)
32. D.M. Brink, in Proceedings of the International School of Physics “Enrico Fermi”, Course
XXXVI, Varenna, 1965, ed. by C. Bloch (Academic Press, New York, 1966), p.247
Chapter 4
Interaction with Electromagnetic Field:
Electromagnetic Moments
Abstract As mentioned in Chap. 1, by measuring the magnetic moment of proton,

Stern showed that the proton is not an ideal point particle which can be described
by the Dirac equation. Also, we learnt in Chap. 3 that the magnetic moment of
deuteron tells that deuteron is in the spin-triplet and isospin-singlet states, and that
one can get information on the magnitude of the tensor force through the admixture
of the D-state in deuteron. In this way, the electromagnetic properties of nuclei
provide valuable information on the structure of nucleons and nuclei. Also, the
interaction of nuclei with the radiation field leads to the emission and absorption
of a γ -ray, governs the lifetime of each energy level, and provides information on
the nuclear structure as well as nuclear collective motions through the strength of
the electromagnetic transitions between nuclear levels. Furthermore, the measure-
ment of the angular correlation between the cascade γ -rays enables us to determine
the spin of each energy level. The electromagnetic transitions play an important
role also in the synthesis of elements through, e.g., the radiative neutron capture
(see Cottingham and Greenwood, An Introduction to Nuclear Physics, 2nd edn.
(Cambridge University Press, Cambridge, 2001); Thompson and Nunes, Nuclear
Reactions for Astrophysics: Principles, Calculation and Applications of Low-Energy
Reactions (Cambridge University Press, Cambridge, 2009) [1, 2]). In this chapter
we learn the electromagnetic moments such as the magnetic dipole moment and the
quadrupole moment which provides important information on the nuclear shape. The
electromagnetic transitions by emitting γ -rays will be discussed in Sect. 8.3 after we
learn about nuclear structure in Chaps. 5 and 7.
4.1 Hamiltonian of the Electromagnetic Interaction

and Electromagnetic Multipole Moments
The Hamiltonian of the interaction between a nucleus and the electromagnetic field
is given by

1
Hem = ρC (r)ϕ(r)dr − j(r) · A(r)dr , (4.1)
c

DOI 10.1007/978-4-431-55378-6_4
98 4 Interaction with Electromagnetic Field: …
where ρC (r) and j(r) are the electric charge density and the electric current density
of the nucleus, and ϕ(r) and A(r) are the scalar and vector potentials to represent
the electromagnetic field. We obtain

1 ∂ E j
Hem = Qϕ(0) − P · E(0) − μ · H(0) − Qi j + ··· , (4.2)
6 ij ∂ xi r=0

Q = ρC (r)dr , (4.3)

P = rρC (r)dr , (4.4)

1
μ= r × j(r)dr , (4.5)
2c

Q i j = ρC (r)(3xi x j − δi j r 2 )dr , (4.6)
E = −∇ϕ , (4.7)
H =∇×A, (4.8)
by expanding ϕ(r) and A(r) around the center of the nucleus. The indices i, j run
from one to three using the notation r = (x1 , x2 , x3 ). δi j is the Kronecker delta. We
assumed that the magnetic field is constant over the nucleus, and used A = 21 H × r,
which is valid in that case. Q is the total charge and P, μ, Q i j are called electric dipole
moment, magnetic dipole moment and electric quadrupole moment, respectively.
Exercise 4.1 Show that A = 21 H × r satisfies H = ∇ × A if the magnetic field is

uniform.
4.1.1 Operators for the Dipole and Quadrupole Moments
The electric charge density ρC and the electric current density j in Eqs. (4.4)–(4.6)
must be considered as the operators ρ̂C and ĵ in order to compare the experimental
values of these multipole moments and the related transition probabilities with their
theoretical values. As we describe in Chaps. 5 and 7 and in Sect. 8.3, there exist
various ways such as the microscopic methods, e.g., the shell model, which uses
the nucleonic degrees of freedom, and the collective model which uses the surface
deformation parameters as coordinates to describe nuclear structure and nuclear
reactions. Since ρ̂C and ĵ are needed to be expressed in terms of relevant coordinates,
their explicit expressions change depending on the model or the method we take
to describe nuclear structure or reactions. Here, we explicitly write down ρ̂C and ĵ
by assuming that we adopt a microscopic model which uses the nucleonic degrees
of freedom. If we treat each nucleon as a point particle, and represent the spin and
isospin of the k-th nucleon as ŝk and t̂k , then they are given by
4.1 Hamiltonian of the Electromagnetic Interaction … 99
1
ρ̂(r) = e − tˆz (k) δ(r̂ − rk ) , (4.9)
k
2
1
1
ĵ(r) = e − tˆz (k) v̂k δ(r̂ − rk ) + δ(r̂ − rk )v̂k
2 k
2
e
+ gs (k)∇ × ŝk δ(r̂ − rk ) , (4.10)
2M N k
where
i
1
v̂k = Ĥ , r̂k ∼ p̂k , (4.11)
m
1
gs (k) ≡ [gn (k) + gp (k)] + tˆz (k)[gn (k) − gp (k)] . (4.12)
2
gs is called the spin g-factor. We use the values of the anomalous magnetic moments
gn = 2μn /μ N = −3.826 and gp = 2μp /μ N = 5.586 discussed in Chap. 1 for gn
and gp . The first and second terms on the right-hand side of Eq. (4.10) are called
the convection current and the magnetization current, respectively. The former is
the usual electric current due to the orbital motion of protons, while the latter the
electric current associated with the magnetic moments of protons and neutrons.
Equation (4.10) can be derived from the Dirac equation, which the Fermi particle
should obey, by making non-relativistic approximation.
Exercise
0 4.2 In the Dirac theory, the current density is given by j k = cψ † α k ψ with
k
σ
α ≡ σ k 0 . On the other hand, the two-dimensional spinor χ representing the
k
third and fourth components of ψ, i.e., the small components if the velocity is small,
can be approximated as χ ≈ (σ · p/2mc)ϕ by using the first and second components
ϕ if the velocity of the particle is sufficiently smaller than the light velocity. Noting
these properties, derive Eq. (4.10).
By inserting Eqs. (4.9) and (4.10) into Eqs. (4.3)–(4.6), we obtain
1
Q̂ = e − tˆz (k) , (4.13)
k
2
1
P̂ = e − tˆz (k) r̂k , (4.14)
k
2

1
μ̂ = r × ĵc (r) + ĵmag (r) dr ≡ μ̂convection + μ̂mag , (4.15)
2c
1
μ̂convection = μN − tz (k) ˆ k ,
ˆ (4.16)
k
2

μ̂mag = μN gs (k)ŝk , (4.17)
k
1

Q̂ i j = e − tˆz (k) 3x̂i (k)x̂ j (k) − δi j r̂ 2 (k) . (4.18)
k
2
4.1.2 Various Corrections
To make discussions with high accuracy, one must take into account various cor-
rections such as (1) the effect of the finite size of proton, (2) that the velocity
v̂k ≡ i [ Ĥ , r̂k ] deviates from p̂k /M N in the presence of velocity dependent forces
such as the spin–orbit interaction, (3) the effect of meson exchange current, i.e.,
that the electromagnetic field operates on the mesons exchanged between the nucle-
ons, (4) the quenching effect, i.e., that the anomalies of the magnetic moment of
nucleons get reduced inside nucleus because of the Pauli exclusion principle, and (5)
relativistic effects.
4.1.3 Measurement of the Magnetic Moment: Hyperfine

Structure
The detailed study of the energy levels of atoms shows that the level structure is more
complex than what is expected from the motion of the electron alone. An additional
structure called the hyperfine structure appears. Consider a nucleus with an odd mass
number. The spin of the nucleus is a half integer. We represent the corresponding
nuclear spin operator as Î. The magnetic moment operator of the nucleus μ̂nucleus
is then given by μ̂nucleus = gμ N Î by using the g-factor. The energy levels of the
whole atom including both electrons and the nucleus split into close-lying energy
levels due to the influence of the magnetic field made by the electron on this nuclear
magnetic moment. This is why the hyperfine structure arises in the level structure.
As an example, Fig. 4.1 shows the low-lying energy levels of a neutral hydrogen
atom [3].
As one can expect from the above arguments, the study of hyperfine structure is a
powerful method to determine the magnitude of the magnetic moment of the nucleus.1
In the following, let us study somewhat in detail the principle of the method.
If we denote the magnetic field made by electrons by Ĥe , then the interaction
Hamiltonian between the electrons and the nucleus is given by
Ĥhyperfine = −μ̂nucleus · Ĥe . (4.19)
1 Stern developed a tenuous molecular beam method in order to realize the interaction-less situation
which is often postulated in idealized theoretical arguments. He discovered the anomalous magnetic
moment of proton by the experiment which sends the tenuous stream of molecules through an inho-
mogeneous magnetic field similarly to the Stern–Gerlach experiment which showed experimentally
the quantization of spatial orientation of angular momentum.
4.1 Hamiltonian of the Electromagnetic Interaction … 101
Fig. 4.1 Low-lying energy levels of atomic hydrogen. Taken from [3]
To be specific, let us consider either the hydrogen atom or an alkali atom, where
the orbital angular momentum of the outer electron is 0 in the ground state. In that
case, Ĥe is given by [4]
16π
Ĥe = − μB |ψ(0)|2 ŝ , (4.20)
3
if we denote the Bohr magneton by μB , the value of the wave function of the outer
electron at the origin by ψ(0), and the spin operator of the outer electron by ŝ. Hence,
we have
16π
Ĥhyperfine = Ĥspin–spin = gμ N μB |ψ(0)|2 (ŝ · Î) . (4.21)
3
Equation (4.21) shows that each energy level splits into two energy levels by the
presence of the inner product (ŝ · Î). In order to emphasize this feature, the index spin–
spin has been attached to the Hamiltonian operator in the second term of Eq. (4.21)
in the sense of spin–spin interaction.
Equation (4.21) can be used to deduce the g-factor from the measurement of
the hyperfine splitting. Thus the magnetic moment can be determined. Note that the
probability for electrons to exist at the origin is finite for s-orbits. In general, the
operator of the magnetic interaction of the electrons with the nucleus is of the form
of Ĥ J I = a(Ĵ · Î) if the total spin of the electrons is denoted by Ĵ. Hence each energy
level with given J splits into either 2I + 1 levels if I ≤ J or 2J + 1 levels if I ≥ J .
One can therefore obtain information on J and I by counting the number of energy
levels appearing in the hyperfine structure.2
2 The energy spacing between successive energy levels labeled with the value of the total spin
F̂ = Jˆ + Iˆ in the hyperfine structure deviates from the rule expected from the (Ĵ · Î) factor for the
magnetic hyperfine splitting considered in this section due to the quadrupole hyperfine splitting
if the quadrupole moment of the nucleus is finite. In other words, deviations of the level interval
from the interval rule for the magnetic hyperfine structure provide important information on the
quadrupole moment or the shape of the nucleus.
Exercise 4.3 Show that the strength of the magnetic field made by the electron at the
position of nucleus is 105 –106 Gauss depending on the species of atom by obtaining
the magnitude of |ψ(0)|2 as a function of the atomic number Z and the principal
quantum number n by using the wave function of the hydrogen atom.
Exercise 4.4 Besides the study of hyperfine structure, there exist various methods
to measure the magnetic moment such as the method of nuclear magnetic resonance,
molecular beam method, Rabi’s molecular beam resonance method (Rabi’s refocus
method) and the perturbed angular correlation method (PAC). Study the principle
and features of each method.
4.2 Electromagnetic Multipole Operators
It is convenient to begin with the generalized definition in order to discuss in a

unified way not only the low-order moments introduced in Sect. 4.1.1, but also all
the electromagnetic properties of nuclei including higher-order multipole moments
as well as the emission and absorption of photons.
Let us first consider the static electric potential φ(r) made by a nucleus in the far
distance from the nucleus. We denote the proton density inside the nucleus by ρp (r).
Using the formula of the multipole expansion of 1/|r − r | for the case of r > r , we
have

1
φ(r) = e ρp (r ) dr
|r − r |

4π e 1 1 ∗
= dr ρp (r )(r ) Y m (Ωr ) Y m
(Ωr ) . (4.22)
r m 2 + 1 r
If we define the μ component of the electric multipole operator by

Q̂ λμ = ρp (r)r λ Yλμ (Ωr )dr , (4.23)
then, the static electric potential is given by
Q̂ m Y ∗ (Ωr )
m
φ(r) = 4π e . (4.24)
m
2 + 1 r +1
Similarly we define the μ component of the magnetic multipole operator by

M̂λμ ≡ μ(r) · ∇(r λ Yλμ (Ωr ))dr (4.25)

1
= [r × j(r)] · ∇r λ Yλμ (Ωr ) dr . (4.26)
c(λ + 1)
4.2 Electromagnetic Multipole Operators 103
We used A · (B × C) = B · (C × A) = C · (A × B) and the relationship j(r) =

c∇ × μ(r) in order to transform from Eq. (4.25) to (4.26).
By inserting the results of Eqs. (4.9) and (4.10) into Eqs. (4.23) and (4.26), we
obtain the expressions for the electromagnetic multipole operators in terms of nucle-
onic coordinates as
1
Q̂ λμ = e − tˆz (k) rkλ Yλμ (θk , ϕk ) , (4.27)
k
2
2

M̂λμ = μ N gs (k)ŝk + ˆ
g (k)k · ∇r λ Yλμ (θ, ϕ) r=rk . (4.28)
k
λ+1
4.3 Properties of the Electromagnetic Multipole Operators
4.3.1 Parity, Tensor Property and Selection Rule
4.3.1.1 Parity
If we denote the operator of the Parity transformation, which inverts all spatial coor-
dinates, by P̂, and represent the coordinate using the polar coordinates (r, θ, ϕ), the
definition of P̂,
P̂(r, θ, ϕ)P̂ −1 ≡ (r, π − θ, ϕ + π ) , (4.29)
leads to
P̂Yλμ (r, θ, ϕ)P̂ −1 = Yλμ (r, π − θ, ϕ + π ) = (−1)λ Yλμ (r, θ, ϕ) . (4.30)
Equations (4.27) and (4.28) therefore give the following properties of the electric
and magnetic multipole operators concerning the parity transformation,
P̂ Q̂ λμ P̂ −1 = (−1)λ Q̂ λμ , (4.31)
−1 λ+1
P̂ M̂λμ P̂ = (−1) M̂λμ . (4.32)
Note that both the spin and the angular momentum operators are invariant under the
parity transformation.
4.3.1.2 Tensor Property
Furthermore, it can be derived from Eqs. (4.27) and (4.28) that both Q̂ λμ and M̂λμ
are the μ component of a rank-λ tensor. The former can be easily proved by noting
that it is directly proportional to the spherical harmonics Yλμ and that Yλμ is the μ
component of a rank-λ tensor. The latter looks complicated at first glance because
of the existence of the inner product of vector operators in front of Yλμ . However,
the tensor property does not change and is identical with that of Yλμ , since the inner
product of two vectors is a scalar, i.e., a rank-0 tensor. Details of tensor algebra are
given in Appendix A.6.
4.3.1.3 Selection Rule
It is necessary to calculate the expectation values of multipole operators in order to

compare the experimental values of the multipole moments with theoretical values.
If we represent the nuclear state as |ψ I M , where I and M are the magnitude of the
angular momentum and its z-component in the space-fixed coordinate system, then
the expectation values of multipole operators are defined by
Q λμ =
ψ I M | Q̂ λμ |ψ I M , (4.33)
Mλμ =
ψ I M | M̂λμ |ψ I M . (4.34)
The following selection rules can be obtained from the parity and tensor properties
of the multipole operators mentioned above,
Q λμ = 0 (for odd λ), (4.35)

Mλμ = 0 (for even λ), (4.36)
Q λμ , Mλμ = 0 (only when μ = 0 and 0 ≤ λ ≤ 2I ). (4.37)
Equations (4.35) and (4.36) hold when one constructs the wave function |ψ I M by
considering only the strong interaction, because |ψ I M then has a definite parity.
4.3.2 Definition of the Electromagnetic Moments
4.3.2.1 Wigner–Eckart Theorem
If we represent the μ component of a rank-λ tensor as T̂λμ , the following theorem

called the Wigner–Eckart theorem holds,3

β I2 M2 |T̂λμ |α I1 M1 = (2I2 + 1)−1/2
I1 λM1 μ|I2 M2
β I2 T̂λ α I1 . (4.38)
Equation (4.38) implies that it is sufficient to know the matrix element for a particular
z-component in order to fully determine the matrix elements of a tensor operator. The
3 There exists an alternative definition which omits the factor (2I2 + 1)−1/2 on the right-hand side.
4.3 Properties of the Electromagnetic Multipole Operators 105
matrix elements for the other z-components are automatically determined according
to the Clebsch–Gordan coefficients.
4.3.2.2 Magnetic Dipole Moment
Since the operator of the magnetic dipole moment is a tensor of rank-1, we define
the magnetic dipole moment which should be compared with the experimental
data by
μ ≡
I I |μ̂z |I I (4.39)

4π
=
I I | M̂10 |I I , (4.40)
3
following the Wigner–Eckart theorem.4
4.3.2.3 The Electric Quadrupole Moment
Similarly we define the electric quadrupole moment by

16π
Q≡
I I | Q̂ 20 |I I . (4.41)
5
The tensor component indices of both the wave functions and the operator in
Eq. (4.41) refer to the space-fixed coordinate system, i.e., the laboratory system.
On the contrary, in the case of a statically deformed nucleus, one defines the intrinsic
electric quadrupole moment by

16π
Q0 ≡
Q̂ 20 (4.42)
5
= ρ p (r )(3z 2 − r 2 )dr , (4.43)
by taking the axes of the coordinate system along the directions of the principal axes,
i.e., by taking the body-fixed coordinate frame (or the intrinsic coordinate system).
For example, if the nucleus is a spheroid, Q 0 is given by
2
Q0 = Z e(a 2 − b2 ) (4.44)
5
2 1
= Z e R0 3α2 1 + α2
2
(4.45)
5 4
4 Note that the z-component of a vector corresponds to the zeroth component of a tensor of rank-1,
i.e., Vz = T̂10 (see Appendix A.6.2).

3 16π
∼ Z e R02 α20 , (4.46)
4π 5
by using the length of the semi-axis in the direction of the symmetry axis a and that
in the perpendicular direction b, and the deformation α2 and the radius parameters
R0 introduced in Eq. (2.66). These equations show that one can learn the shape of the
nucleus, more precisely the shape of the distribution of protons inside the nucleus,
if information on the intrinsic electric quadrupole moment can be obtained. Q 0 > 0
and Q 0 < 0 mean the shapes prolonged and shrunk in the direction of the symmetry
axis and are called prolate and oblate shapes, respectively. In Eq. (4.46), we used
the deformation parameter α20 , which we will introduce in Chap. 7, and ignored the
second order term with respect to α20 .
What one measures experimentally is not the value of Q 0 , but that of Q. If a
nucleus is axial-symmetrically deformed, the component of the angular momentum
projected on the symmetry axis, K , becomes a good quantum number in addition to
the magnitude of the angular momentum I and its component along the space-fixed
z-axis M. In this case, for even–even nuclei or for odd nuclei, where all nucleons
rotate simultaneously5 and I = K = 1/2 or K ≥ 3/2, one can prove that Q and Q 0
are related as
3K 2 − I (I + 1)
Q = Q0 . (4.47)
(2I + 3)(I + 1)
Especially, if I = K as in the case of the ground state,
I 2I − 1
Q= Q0 . (4.48)
I + 1 2I + 3
Equation (4.48) shows that Q = 0 irrespective of Q 0 if I = 0 or 1/2. This result

can be expected from the coupling rule of angular momenta or of tensors and the
orthogonality of states with different values of angular momentum. The derivation
of Eq. (4.47) is given in Appendix A.7.
References
1. W.N. Cottingham, D.A. Greenwood, An Introduction to Nuclear Physics, 2nd edn. (Cambridge
University Press, Cambridge, 2001)
2. I.J. Thompson, F.M. Nunes, Nuclear Reactions for Astrophysics: Principles, Calculation and
Applications of Low-Energy Reactions (Cambridge University Press, Cambridge, 2009)
3. J.D. Bjorken, S.D. Drell, Relativistic Quantum Mechanics (McGraw-Hill, New York, 1964)
4. L.D. Landau, E.M. Lifshitz, Quantum Mechanics–Non-Relativistic Theory, 2nd edn. (Pergamon
Press, Oxford, 1965), p. 465
5 Thereexist two possibilities for odd nuclei. In one case, the core part made from even–even
nucleus makes a collective rotation, while the last odd nucleon moves almost independently from
this motion. In the other case, both the core and the last nucleon rotate together as a whole. The
former and the latter are called weak coupling and strong coupling states, respectively.
Chapter 5
Shell Structure
Abstract Although the bulk properties of nuclei such as the mass and size, and
also fission and the compound nucleus reactions suggest that the nucleus behaves
like a liquid, there exist characteristic properties which cannot be understood from
such a point of view. The existence of magic numbers in various phenomena is the
most evident example, suggesting the shell structure. The magic numbers appear
in various systems in nature. The existence of the noble gases in the periodic table
of the elements is the most popular example. Contrary to those magic numbers for
atoms, which are associated with the long range Coulomb interaction, the magic
numbers for nuclei which originate from the short range force differ in numbers.
Also, the spin–orbit interaction plays a crucial role in the magic numbers for nuclei.
In this chapter we discuss how the magic numbers arise, and discuss the spin and
parity properties of the nuclei in the vicinity of magic numbers. Although the shell
model which is based on the mean-field theory succeeds, it is suggested that it is
important to take into account the effects of the residual interaction, i.e., the pairing
correlation in order to explain the details of the nuclear phenomena. As an example
of the current topics, we also discuss the present status of the research of superheavy
elements which are stabilized by shell effects.
5.1 Magic Numbers
The saturation properties of density and binding energy per nucleon suggest the
validity of the liquid-drop model for nuclei. The mass formula is also based on the
picture of a nucleus as a liquid drop, and various basic properties of fission can be
explained based on the liquid-drop model. Also, the compound nucleus reaction, one
of the representative nuclear reactions, strongly supports the validity of the liquid-
drop model for nuclei.
The liquid-drop model is the picture that physical quantities vary continuously
or monotonically as functions of the mass and atomic numbers and that there exist

DOI 10.1007/978-4-431-55378-6_5
108 5 Shell Structure
neither special mass number nor special proton and neutron numbers. However, there
exist many experimental data which contradict this picture. We mention here some
of the examples.
Figure 1.7 shows that the nuclei whose neutron number N is 50, 82, or 126 exist
significantly more than the surrounding nuclei. Figure 2.6 and more detailed analysis
of the binding energy show that the nuclei whose proton number and/or neutron
number are one of 20, 28, 50, 82, 126 have larger binding energy per nucleon than
the other nuclei whose binding energy can be well represented by the mass formula
and are more stable than them. Also, as stated concerning Fig. 2.17, most of the
spontaneous fission and the induced fission at low energies are asymmetric fission,
where the mass number and the atomic number of the fission products which have
large yield are related to the special numbers mentioned above. As we learn in Chap. 7,
the shape of nucleus is also intimately related to the mass and atomic numbers, and
the shape of the nucleus whose proton number and neutron number are one of the
special numbers mentioned above is spherical, i.e., the absolute value of their intrinsic
quadrupole moment is either 0 or small. On the other hand, 42 He and 168 O, whose proton
and neutron numbers are both 2 or 8, respectively, are much more stable than the
surrounding nuclei.
In this way, 2, 8, 20, 28, 50, 82, 126 are special numbers for nuclei and are called
magic numbers.1
Sidebar: The Connection Between the Magic Numbers and Life The atomic
numbers 2, 10, 18, 36, 54, 86 of the noble gases in the periodic table of the elements
can also be thought to be the magic numbers of atoms. The respiration of organism
is guaranteed by the rich abundance of oxygen thanks to the fact that the magic
number for nuclei is not 10, but 8. As we learn in the next section, the change of
the magic number from 10 to 8 is caused because of the short range character of
the nuclear force. If the nuclear force were not of short range, organism could not
sustain respiration, i.e., could not obtain energy by reacting oxygen with glucose,
and should have got into serious trouble or had to rely upon some different processes
to obtain energy.
1 The magic numbers can be thought to be the proton and/or neutron numbers for which the nucleus
becomes spherical, or those at which the binding energy per nucleon, i.e., the separation energy of
a nucleon, becomes particularly large compared with that in the surrounding nuclei, or the proton
and neutron numbers at which the nuclear radius suddenly changes as a function of the proton and
neutron numbers. If we consider in this way, the magic number is determined by the combination of
the proton and neutron numbers, and the magic numbers for nuclei which are unstable with respect
to β-decay such as neutron-rich unstable nuclei or neutron-deficient nuclei, so-called proton-rich
nuclei, can be different from the known magic numbers for stable nuclei. The change of magic
numbers or the appearance of new magic numbers in the region of unstable nuclei is one of the
central research subjects in current nuclear physics.
5.2 Explanation of the Magic Numbers by Mean-Field Theory 109
5.2 Explanation of the Magic Numbers by Mean-Field

Theory
5.2.1 The Mean Field
The magic numbers for atoms can be explained based on the picture that electrons are
moving independently in the Coulomb field VC made by the nucleus. In this picture,
the central roles are played by the facts that the Coulomb field is of long ranged and
the functional form is given by the inverse of the distance from the center r and the
Pauli exclusion principle.
Let us derive the magic numbers for nuclei by similarly assuming that nucleons
are moving independently from each other inside the nucleus, but obeying the Pauli
exclusion principle. However, there exists no central particle which governs the force
like the nucleus for electrons inside an atom. Instead, we assume that each nucleon
is moving in the mean field made by the nucleons themselves. In this case, the field
of force at the position r will be given as

A
V (r) = v N N (r − ri ) = v N N (r − r )ρ(r )dr , (5.1)
i=1
by adding the forces exerted on the nucleon at r from the other nucleons. v N N (r − r )
is the nuclear potential working between two nucleons at r and r , and ρ(r) is the
number density of nucleons at the position r. We used that ρ(r) is given by

A
ρ(r) = δ(r − ri ) , (5.2)
i=1
in transforming to the last term of Eq. (5.1).

If we assume
v N N (r) = −V0 δ(r) (V0 > 0), (5.3)
by emphasizing the short-range character of the nuclear force, we obtain
VN (r) = −V0 ρ(r) (5.4)
for the nuclear potential. Equation (5.4) suggests that the nuclear part of the mean
potential which nucleons feel inside the nucleus has the radial dependence similar
to that of the number density of nucleons inside the nucleus.
For this reason, the nuclear potential for nucleons inside a nucleus is often assumed
to be
V0
VN (r ) = − , (5.5)
1 + exp[(r − R)/a]
after Eq. (2.22), and is called the Woods–Saxon potential. Here, we assumed a spher-
ical nucleus.
The value of V0 can be estimated to be 40–50 MeV from the facts that the Fermi
energy is 33–40 MeV and the binding energy per nucleon is about 8 MeV.
5.2.2 Energy Levels for the Infinite Square-Well Potential
In the following, let us argue by ignoring the Coulomb force, although one has to add
the contribution of the Coulomb force for protons. One needs to perform numerical
calculations in order to determine the energy levels in the Woods–Saxon potential.
Let us therefore replace the potential by a square-well potential with an infinitely
high wall at r = R in order to gain a qualitative understanding,

− V0 for 0 ≤ r < R
V (r ) = (5.6)
+ ∞ for r ≥ R .
The wave function in the region of the potential well is then given by
ψ(r) = A j (K r )Ym (θ, ϕ) , (5.7)

2M N
K = (E + V0 ) , (5.8)
2
j (K R) = 0 . (5.9)
If we denote the nth zero of the spherical Bessel function j (x) by xn , then we obtain
the following formula to give the energy levels
2
E n = −V0 + x2 . (5.10)
2M N R 2 n
Table 5.1 shows the values of some of the zeros xn , and numbers them in order of
magnitude.
Table 5.1 The zeros xn of the spherical Bessel function j (x)
n j0 (x) j1 (x) j2 (x) j3 (x) j4 (x) j5 (x) j6 (x)
1 1 3.14 2 4.49 3 5.76 5 6.99 7 8.18 9 9.36
12 10.51
2 4 6.28 6 7.73 8 9.10
11 10.42 11.70 12.97 14.21
3
10 9.42
13 10.90 12.32 13.70 15.04 16.35 17.65
4 12.57 14.07 15.51 16.92 18.30 19.65 20.98
5.2.3 The Harmonic Oscillator Model
The harmonic oscillator model, where the mean field is assumed to be given by the
following formula, is often used as a more convenient method,
1
V (r ) = M N ω2 r 2 . (5.11)
2
In this case, the energy eigenvalues and the quantum numbers are given by

3
E n x ,n y ,n z = ω N + , (5.12)
2
N = nx + n y + nz , (5.13)
n x , n y , n z = 0, 1, 2, . . . , (5.14)
in the Cartesian coordinates, i.e., in the Descartes coordinates, and by

3
E nm = ω N + , (5.15)
2
N = 2(n − 1) + , principal quantum number, (5.16)
n = 1, 2, 3, . . . , radial quantum number, (5.17)
= 0, 1, . . . , azimuthal quantum number (orbital angular momentum),
(5.18)
m = −, − + 1, . . . , − 1, , magnetic quantum number, (5.19)
in the spherical polar coordinates. n is the quantum number or index which shows
the numerical order of the states with the same angular momentum when they are
counted from the lowest energy state. nr ≡ n − 1 = 0, 1, 2, . . . is also often used
for n. n r is the number of nodes of the radial wave function at positions other than
the origin of the coordinate system, i.e., within the potential well.
The energy levels and the wave functions in the case of Cartesian coordinates can
be easily obtained by the series expansion method or by the method using the cre-
ation and annihilation operators as is explained in the standard textbook of quantum
mechanics, and the wave function is given by

1
ψn x (x) = Nn x Hn x (αx) exp − α 2 x 2 , (5.20)
2

α = M N ω/ , (5.21)
by using the Hermite polynomials Hn .2 Nn x = ( √π2nαx (n x )! )1/2 is the normalization

constant.
One can easily guess which kinds of angular momentum states are included in
each energy level when the polar coordinates are used by considering the parity
and the degeneracy of each energy level, which can be easily determined in the
Cartesian coordinates, and the fact that the representations in terms of the Cartesian
coordinates and the polar coordinates can be transformed to each other using a unitary
transformation.
Exercise 5.1 Let us write the wave function in the polar coordinates as
u nr (r )
ψnr m (r) ≡ Rnr (r )Ylm (r̂) = Ylm (r̂) . (5.22)
r
Show that the normalized radial wave function u nr is then given by
√
1/2
2α Γ (n r + + 23 ) +1 − 21 x 2 3 2
u nr (r ) = x e F −n r ; + ; x (5.23)
Γ ( + 23 ) nr ! 2

1/2
√ nr ! (+ 1 )
x +1 e− 2 x L nr 2 (x 2 ) ,
1 2
= 2α (5.24)
Γ (n r + + 2 )
3
where x = αr . F is the confluent hypergeometric function,

nr
3 k k nr (2 + 1)!!
F −n r ; + ; z = (−1) 2 zk , (5.25)
2 k=0
k (2 + 2k + 1)!!
(+ 21 )
and L nr is the generalized Laguerre polynomial.
5.2.4 The Magic Numbers in the Static Potential

Due to Short Range Force
The first and second columns from the left in Fig. 5.1 give the single-particle energy
levels together with the quantum numbers based on Eqs. (5.15) and (5.10), respec-
tively. The magic numbers appear when one energy level is fully occupied and the
energy gap to the next energy level is large. If we denote the number of states in the
√
2 The parameter b which is defined by b ≡ 1/α = /M N ω is also often used and is called the
oscillator parameter. It gives a measure of the strength of the potential which confines nucleons and
has the dimension of length. It gives the extension of the 0s-state and the measure of the size of
clusters such as the α-cluster.
Fig. 5.1 Nucleon energy levels for several spherical mean fields. H.O. stands for the harmonic
oscillator model, ISQ stands for the infinite square-well potential, WS stands for the Woods–Saxon
potential. The last column represents the energy levels when the spin–orbit interaction is taken into
account
energy level with the principal quantum number N by N (N ) in order to predict the
magic numbers based on the harmonic oscillator model, then
N (N ) = (N + 1)(N + 2) (5.26)
regardless of the even-odd property of N . We show in Table 5.2 the N (N ) and the
total number of states whose principal quantum number is smaller than or equal to
N , Ntot (N ) ≡ NN =0 N (N ) .
Table 5.2 The degeneracy of energy levels and the magic numbers for the harmonic oscillator
model
N 0 1 2 3 4 5 6 7
N (N ) 2 6 12 20 30 42 56 72
Ntot (N ) 2 8 20 40 70 112 168 240
Table 5.2 shows that the magic numbers 2, 8 and 20 can be explained by the
harmonic oscillator model. Figure 5.1 shows that the infinite square-well potential
also explains the small magic numbers 2, 8 and 20. In addition, in the square-well
potential, the degeneracy concerning the angular momentum is resolved and the state
with larger angular momentum becomes energetically lower. This is because the
state with larger orbital angular momentum among the states with the same principal
quantum number has larger probability in the surface region, and hence more strongly
feels the difference, i.e., a negative perturbation, between the square-well potential
and the harmonic oscillator potential.
The third column in Fig. 5.1 shows the energy levels for the Woods–Saxon poten-
tial. The structure of the energy levels in this case resembles that for the square-well
potential.
N
Exercise 5.2 Prove Eq. (5.26) by using that N (N ) is given by =0,2,... (2+1)×2
N
when N is an even number and by =1,3,... (2 + 1) × 2 when N is an odd number
because of the degeneracies concerning the magnetic quantum number and the spin.
5.2.5 Spin–Orbit Interaction
Suppose that the spin–orbit interaction is acting on each nucleon in addition to

the static central force in order to explain the magic numbers larger than 20. The
Schrödinger equation to determine the energy levels of nucleons is then given by

2
− Δ + V (r ) + vLS (ˆ · ŝ) ψ(r) = E ψ(r) . (5.27)
2M N
We assume here that vLS = ξLS (constant) for simplicity.

The modified energy levels for each N can be obtained by diagonalizing the
Hamiltonian with the degenerate states according to the perturbation theory for
degenerate case. Here, however, we use a more convenient method. Since the opera-
ˆ ŝ does not commute with any components of the orbital angular momentum and
tor ·
spin operator, their magnetic quantum numbers cannot be good quantum numbers
separately. On the other hand, since
2
2

ˆ · ŝ, ˆ = ˆ · ŝ, ŝ2 = ˆ · ŝ, ĵ = ˆ · ŝ, ĵα = 0 , (5.28)
Fig. 5.2 The ground and first excited states of 5 He and 5 Li [1]. The numbers on the left of each
first excited state are the excitation energies in MeV. Also shown are the threshold energies for the
neutron and proton emissions. As seen in the figure, all the levels are above the threshold energy for
nucleon emission and are resonance states with large widths. The threshold energies for the neutron
and proton decays are after [2]
one can choose the magnitude of the orbital angular momentum, the magnitude of
spin, the magnitude of the total angular momentum ĵ ≡ ˆ + ŝ and its z-component,
, s = 21 , j, m, as good quantum numbers to specify states. Consequently, the energy
level splits into two as

E n − ξ L S 21 ( + 1) (for j = − 21 )
Enj = (5.29)
E n + ξ L S 21 (for j = + 21 ) ,
according to the magnitude of the total angular momentum. We used

1 2 2
ˆ · ŝ|s jm = (ĵ − ˆ − ŝ2 )|s jm
2
1 1 1
= j ( j + 1) − ( + 1) − + 1 |s jm (5.30)
2 2 2
to derive Eq. (5.29).

We show in Fig. 5.2 the ground and first excited states of 5 He and 5 Li in order
to learn the strength and the sign of the spin–orbit interaction. The figure suggests
that the sign of ξ L S is negative,3 hence the energy of the j = + 1/2 state shifts
downward, while that of j = − 1/2 state upward.4
3 For electrons in atoms, the spin–orbit interaction is repulsive. Hence the energy of the j = − 1/2
state gets lowered as shown in Fig. 4.1.
4 It is difficult to determine the resonance parameters such as the level position for a resonance state
with a large width. In fact, [3] gives values which are very different from those in Fig. 5.2 for the
energy splitting between the p1/2 and p3/2 states. For example, the excitation energy of the p1/2
state of 5 He is 4.6MeV. Reference [1] also gives two different parameter sets depending on the
method of analysis. The resonance position and the width given in Fig. 5.2 and mentioned also in
the following exercise have been obtained from the pole position of the S-matrix on the complex
energy plane (extended R-matrix method; after [4]). On the other hand, the conventional method,
i.e., standard R-matrix method, determines the parameters from the poles of the cross section by
restricting the energy to be real as experiments are performed. The former has the advantage that
the results do not depend on the choices of the boundary condition and the channel radius which
appear in the R-matrix theory. See [5, 6] for details of the R-matrix theory.
Exercise 5.3 Estimate the magnitude of ξLS for protons and neutrons from Fig. 5.2.5
The width of the level is Γ = 648 keV for the p3/2 state and Γ = 5.57 MeV for the
p1/2 state for 5 He, while Γ = 1.23 MeV for p3/2 state and Γ = 6.60 MeV for p1/2
state for 5 Li [1]. Estimate the half-life and the mean life of the ground and excited
states of 5 He and 5 Li from these values.
The last column of Fig. 5.1 gives the nucleon energy levels when the effect of the
spin–orbit interaction is added based on these considerations. Each energy level is
designated by the set of the radial quantum number n, the orbital angular momentum
and the total angular momentum j. The figure shows that the magic numbers 28, 50,
82, 126 arise due to the spin–orbit interaction.6 Note that there are 2 j + 1 degenerate
states with different magnetic quantum numbers in a single energy level with the
angular momentum j.
Note that the magic numbers 50, 82 and 126 are caused by the penetration of the
j = + 1/2 levels with a large orbital angular momentum into the energy region
of the energy levels with the principal quantum number N smaller by one due to the
spin–orbit interaction and the surface effect, i.e., the effect that the realistic potential
is deeper in the surface region than that in the harmonic oscillator model. These states
are called intruder states. The 1g9/2 , 1h 11/2 and 1i 13/2 levels play the principal role
as the intruder states for the magic numbers 50, 82 and 126, respectively.
Exercise 5.4 Answer the following questions concerning the transformation from
the | s = 21 m m s basis to the |s jm j basis.
1. Show that the total number of states is (2 + 1) × 2 in both bases.
2. The states are given by |Ym α and |Ym β in the former basis, where α and
β represent the spin-up and spin-down states, respectively. Write the general
expression of the Unitary transformation which represents the |s j = +1/2 m j
and |s j = − 1/2 m j states in the |sm m s basis. Also, write explicitly the
| f 7/2 m = 7/2 and | f 5/2 m = 5/2 in terms of Y3m , α, β.
Sidebar: Japanese Contribution to the Development of Shell model—Pioneers
of History: Tadayoshi Hikosaka and Takahiko Yamanouchi The independent
particle model, i.e., the shell model, thus succeeds in explaining the existence and
the actual numbers of magic numbers through the shell layer structure of the energy
levels of nucleons, and the Pauli exclusion principle. Historically, pioneering studies
of the independent particle model (shell model) had been developed in Japan too by
Tadayoshi Hikosaka, Tohoku University, and Takahiko Yamanouchi, University of
Tokyo. However, their pioneering works had been forgotten before their theory is
accepted as the correct effective model for many nuclear properties.
5 It
is necessary to handle both the ground and excited states as resonance states in order to achieve
a quantitatively accurate estimate. Here, however, use the harmonic oscillator model and ignore the
radial dependence of the spin–orbit force in order to obtain a rough estimate of ξLS .
6 The tensor force also strongly affects the magic numbers and the shape of nucleus, and is one of
the active research subjects in recent years in connection with, e.g., the structure of unstable nuclei
such as the change of the magic number. For example, 40 Ca is spherical, but 32 Mg is deformed
among the N = 20 isotones (see [7]).
There remain historical episodes that both Hikosaka and Yamanouchi received
strong objections when they discussed with Niels Bohr based on the success of
the liquid-drop model to explain fission and the compound nucleus reactions. On
the contrary, the existence of magic numbers was a strong experimental evidence in
the cases of Mayer and Jensen who established the shell model later. There remains
also an episode that Jensen got the inspiration of the spin–orbit interaction while he
was attending a dance party. Incidentally, Hikosaka is the same person as who has
proposed the heterogeneous reactor.
Exercise 5.5 The puzzle of the magic numbers: Comparison with the cases for atoms
and microclusters. Discuss the magic numbers for many-body systems of fermions
when the interactions among the constituent particles can be approximated by the
mean fields listed below in the zeroth order approximation.
1. An attractive potential proportional to 1/r when the distance from the center is
denoted by r (periodic table of the elements).
2. Spherical Woods–Saxon potential (magic numbers for Alkali metals [8]).
3. Three-dimensional harmonic oscillator plus attractive spin–orbit interaction
(magic numbers for nuclei).
4. Three-dimensional harmonic oscillator plus repulsive spin–orbit interaction.
5. Two-dimensional harmonic oscillator.
5.3 The Spin and Parity of the Ground and Low-Lying

States of Doubly-Magic ±1 Nuclei
The nucleus in which both proton and neutron numbers are magic numbers such as
16 208
8 O8 and 82 Pb126 is called doubly-magic nucleus or doubly-closed shell nucleus.
One can predict the spin and parity of the ground state of the nuclei whose proton
or neutron number is smaller or larger than those of a doubly-magic nucleus by one
by using the shell model. The spin and parity of low-lying states of those nuclei can
also be predicted to some extent.
Consider 20982 Pb for example. For the ground state, the single-particle energy levels
are successively filled with nucleons from the lowest energy level according to the
Pauli exclusion principle. Figure 5.1 suggests that all the 82 protons are just filled in
the states from the n j = 1s1/2 level to the 3s1/2 level. The 126 neutrons occupy all
the states from the 1s1/2 to the 2 f 5/2 levels. Let us call this part, which corresponds
to doubly-magic numbers, the core part. One then fills the last 127th neutron, called
the valence neutron, in the next energy level 2g9/2 . Since there is only one way to fill
all the 2 j + 1 degenerate states for a given n j with the same number of particles,
Jtot =0 because of 2Jtot + 1 = 1 if we denote the total spin of these 2 j + 1 nucleons
by Jtot . Also, the total parity of these nucleons is plus irrespective of the even–odd
property of the orbital angular momentum , since 2 j + 1 is always an even number.
Consequently, the total spin and parity of the nucleons forming the core are 0+ . As
Fig. 5.3 The ground and

low-lying excited states of
209 Pb and 209 Bi (the energies
are in MeV) [3]
the result, the spin and parity of the whole nucleus is determined by those of the
valence neutron. Hence the spin and parity of the ground state of 209 82 Pb are expected
to be 9/2+ .
The left side of Fig. 5.3 shows the experimental energy-level diagram for 209 82 Pb
(Data from [3]). The spin and parity of the ground state agree with the prediction of
the shell model described above. One can keep the core part as it is and change the
level to fill the valence neutron from the 2g9/2 state to another state successively from
the low energy level in order to predict the spin and parity of the other states. As an
alternative, one can consider the state, where the core is polarized and the nucleon
in the energy level n h h jh in the core, where the index h stands for hole, is promoted
to the n p pj p level outside the core, where the index p stands for particle. If the
original valence nucleon is in the n j state, the corresponding wave function can be
[ j ]
written, for example, as (n h h jh )−1 ⊗ [(n p pj p ) ⊗ (n j )][J ] ; j m .7 The state
of this kind of configuration is called two-particle one-hole (2p1h) state. Although
there are many possibilities depending on the way to make a hole, the choice of the
particle state to promote the original core nucleon, and the way to couple the angular
momenta, the state, where the core neutron is promoted to the same level as that of
the valence neutron and where the total angular momentum of two valence neutrons
7 We used the convention to omit the core part in writing the configuration, and to make the core
part as the reference configuration.
5.3 The Spin and Parity of the Ground and Low-Lying States … 119
J couples to zero, appears at low energy because of the pairing interaction between
the valence nucleons. In this case, the spin and parity of the whole nucleus become
identical with those of the hole state, i.e., jh and (−1)h .
An interesting another possibility is that a series of characteristic states appear in
low energy region by the process that the core part makes a low energy collective
excitation and weakly couples with the valence nucleon. An example can be seen in
209
Bi shown on the right side of Fig. 5.3 [3]. The ground state and the first through
third excited states can be understood as the single-particle states or 2p1h state in the
shell model. The septet states above them, 3/2+ , 5/2+ , 7/2+ , 9/2+ , 11/2+ , 13/2+ ,
and 15/2+ , are the states, where the core of 209 Bi makes a vibrational excitation of
the octupole type (see Chap. 7) and weakly couples with the valence proton in the
h 9/2 orbit, and can be expressed as |[3− ⊗ h 9/2 ] J ; J + [9] (see also [10]).
Exercise 5.6 1. Predict the configurations for the low-lying excited states of 209
82 Pb.
2. Predict the configurations as well as the spin and parity of the ground state and
low-lying excited states of 17 O, 15 O and 209
83 Bi, and compare the predicted spin
and parity with those in the experimental data in [3].
Incidentally, in the case of odd–odd nuclei in which both proton and neutron
numbers are odd numbers, the parity and the spin of the ground state and low-lying
excited states are expected to be given by (−1)p +n and one of the values between
| jp − jn | and jp + jn , where p , jp , n . jn are the orbital and total angular momenta of
the orbits of the last proton and neutron. For example, in the case of 147 N, p = n = 1,
jp = jn = 1/2 and the experimental data for the spin and parity of the ground state
are 1+ .
5.4 The Magnetic Dipole Moment in the Ground State

of Odd Nuclei: Single Particle Model
The idea which attributes the properties of not only the doubly-magic ±1 nuclei,
but all nuclei with an odd mass number, i.e., the even–odd or odd–even nuclei, or
simply the odd nuclei, in general entirely to the last proton or neutron whose total
number is an odd number is called the single particle model. The single particle
model often well predicts the spin and parity of the ground state and some states in
the vicinity of the ground state of odd nuclei. This happens because nucleons move
around pairwise by forming a pair of total angular momentum and parity 0+ due
to the pairing interaction (see Sect. 5.8), and consequently the total spin and parity
of the nucleus is determined by the last unpaired nucleon or hole regardless of the
actual number of valence nucleons outside the core or the number of holes inside the
core.
Exercise 5.7 Predict the configurations and the spin and parity of the ground and
some low-lying excited states of 45
21 Sc and compare the predicted spin and parity with
the experimental data [3].
Here, let us learn that the single particle model provides a rough estimate of
the magnetic dipole moment of odd nuclei, and further study how the discrepancy
between the experimental data and the prediction of the single particle model is
explained by the concept of core polarization or the configuration mixing.
5.4.1 The Schmidt Lines
The magnetic moment is defined by μ ≡ j j|μ̂z | j j in the single particle model

following Eq. (4.39), where the operator of the magnetic moment μ̂ is given by
ˆ ,
μ̂ = μ N (gs ŝ + g ) (5.31)
according to Eqs. (4.15)–(4.17).

Here let us derive the explicit expression of μ by using the projection theorem for
the expectation value of vector operator (see Appendix A.6.4). From Eq. (A.106),
1
μ= j j| ĵ Z | j j j j|(μ̂ · ĵ)| j j (5.32)
j ( j + 1)
1
= j j|(μ̂ · ĵ)| j j . (5.33)
j +1
The expectation values of the operators ŝ · ĵ and ˆ · ĵ , which are needed to evaluate
j j|(μ̂ · ĵ)| j j , can be easily obtained in the same way as that was used to obtain the
expectation value of the operator ˆ · ŝ in Sect. 5.2.5. Finally, μ is given by

μ N g j − 21 + 21 gs for j = + 1/2
μ= (5.34)
μ N g j + 23 − 21 gs j
j+1
for j = − 1/2 .
Thus one can expect that the magnetic moments are given by the solid lines in
Fig. 5.4 depending on whether j = −1/2 or j = +1/2 by remarking that g = 1.0,
gs = g p ≈ 5.58 in the case of odd-Z nuclei and g = 0.0, gs = gn ≈ −3.82 in
the case of odd-N nuclei. These lines are referred to as the Schmidt lines. The
experimental values lie between the two Schmidt lines except for a small number of
nuclei such as 3 H, 3 He, 13 C and 189
76 Os .
Exercise 5.8 Derive Eq. (5.34) for the Schmidt lines by using that the wave function
of the | j j state is given by

Y |α for j = + 1/2
| j j = √ (5.35)
√ 1
2(+1/2)
−Y−1 |α + 2 Y |β for j = − 1/2 .
5.4 The Magnetic Dipole Moment in the Ground State of Odd Nuclei … 121
Fig. 5.4 The magnetic moment for odd nuclei. a Odd-Z nuclei. b Odd-N nuclei. After [11]
5.4.2 Configuration Mixing and Core Polarization
Figure 5.4 shows the success of the shell model in the sense that it well explains the
gross behaviour of the magnetic moment, while it also shows the limitation of the
single particle model in reproducing the magnitude. A large part of the deviation of
the experimental values from the corresponding Schmidt values can be explained
by considering the configuration mixing due to the residual interaction which is not
incorporated in the mean field [12]. Let us denote the residual interaction by V , and
write the wave function which takes into account the effect of the residual interaction
as
[J ] [ j]

| j j = α0 | j j + α ph j ⊗ jh−1 ⊗ j p ; jj + ··· (5.36)
ph

n
= α0 | j j + αi |Ci . (5.37)
i=1
by assuming that V is a two-body interaction. | j j = | jm j = j is the zeroth order

wave function in the shell model (single particle model). Figure 5.5 illustrates the
Fig. 5.5 Configuration

mixing by the core
polarization
state given by Eq. (5.36). In Eq. (5.37), we used an abbreviated notation |Ci for the
additional terms due to the configuration mixing.
The effect of configuration mixing is often called the effect of core polarization. In
fact, for the nuclei in the vicinity of doubly-magic nuclei, the correction terms mixed
by the configuration mixing or a part of them are caused by a process where the core
part is excited, i.e., polarized. However, it will be more appropriate to understand
that the effect of configuration mixing is a wider concept than that of simple-minded
core polarization. In Eq. (5.36), we used the indices p and h suggesting the particle
or hole states as indices of quantum numbers in order to facilitate imagining the
physical process.
The magnetic moment is given by

n
μ = |α0 |2 j j|μ̂ Z | j j + 2Re α0∗ αi j j|μ̂ Z |Ci + O(αi αi ) , (5.38)
i=1
when there exists configuration mixing, or core polarization. If we 20assume, n =

20, αi = 0.1(i = 1 − 20) and α02 = 0.80, i.e., α0 = 0.89, then α0 i=1 αi = 1.78.
Hence a large correction can arise if there are many different ways to polarize the
core and if they contribute cooperatively even if the mixing amplitude of each core
polarization is small. In fact, Arima and Horie have succeeded in reproducing the
experimental values of the magnetic moment by taking into account the effect of
core polarization [12].
Various types of configuration mixing can be considered depending on the residual
interaction. For example, if V = (σ 1 · σ 2 )Vσ (r ) or V = (σ 1 · σ 2 )(τ 1 · τ 2 )Vσ τ (r ),
then J in Eq. (5.36) becomes 1, and one can expect that the effect of core polarization
becomes large for the nuclei in which the j = + 1/2 level among the group of
levels split by the spin–orbit interaction from a degenerate energy level in the static
mean field is occupied and the counter j = − 1/2 level is empty.
So far, we used the values of free nucleons for the g-factor of nucleons. As
mentioned in Sect. 4.1.2, those values can change inside nuclei e.g., by a quenching
effect. More generally, the effects of meson and the excited state of nucleon Δ33
also contribute to the deviation from the Schmidt values. Especially, the magnetic
moments of 3 H and 3 He, which lie outside the Schmidt lines, are explained by the
meson effect. It is suggested from the detailed studies on the interaction between
pions and nucleons and from the electromagnetic form factors of these particles that
the anomalous magnetic moment of nucleons is weakened inside the nucleus by
about 10 %. In this connection, we wish to mention that the study of the I π = 11−
isomer state of 210 Po shows that the g -factor of proton is larger by about 10 % as an
example of the meson effect [13].
5.4 The Magnetic Dipole Moment in the Ground State of Odd Nuclei … 123
5.4.3 [Addendum] The Anomalous Magnetic Moments

of Nucleons in the Quark Model
In Chap. 1, we attributed the anomalous magnetic moments of nucleons to the effects

of meson cloud. Alternatively, here we show that the anomalous magnetic moments
of nucleons can be naturally derived from the point of view of the quark model.
In the quark model, proton is made up of two u quarks and one d quark. It is
considered that all the baryons existing in nature are in the color-singlet state, which
is antisymmetric concerning the color degree of freedom, and are spatially symmetric
in low-energy states. Hence the wave function for proton in the combined spaces of
the spin and flavor degrees of freedom must be symmetric in total, and is given by
[14]

1
|p ↑ = uud(↑↓↑ + ↓↑↑ −2 ↑↑↓) + udu(↑↑↓ + ↓↑↑ −2 ↑↓↑)
18

+ duu(↑↓↑ + ↑↑↓ −2 ↓↑↑) . (5.39)
Similarly, the wave function for |n ↑ state is given by interchanging u and d quarks.
On the other hand, the operator of the magnetic moment of the ith quark is given
by
(q) (q) e (q) (q)
μ̂i = Q i (q)
σ̂ i ≡ μi σ̂ i (5.40)
2m i c
(q) (q)
if we assume that quarks are structureless Dirac particles, where Q i e and m i are
the charge and the mass of the ith quark. The magnetic dipole moment of proton is
given by
3
(q)
μp = p ↑ |μ̂i z |p ↑ , (5.41)
i=1
following the definition (4.39). Using Eq. (5.39) in Eq. (5.41), we obtain
1
μp = (4μu − μd ) . (5.42)
3
Similarly, we obtain 1
μn = (4μd − μu ) (5.43)
3
for the magnetic dipole moment of neutron. If we identify the quarks with constituent
quarks and assume that m u = m d , then it follows
μn 2
=− . (5.44)
μp 3
This value well agrees with the experimental value μn /μp ≈ −0.685.
Exercise 5.9 Estimate the value of μp by assuming m u c2 = m d c2 ≈ 300 MeV, and

compare the estimated value with the experimental data.
5.5 Mass Number Dependence of the Level Spacing ω
The spacing between two successive energy levels, i.e., ω in the harmonic oscillator
model, plays the central role in the shell model. Here, we estimate the value of ω.
We learnt in Sect. 2.1 that the mean-square radius for stable nuclei is given by
3 2 3
r 2 = R = r02 A2/3 (5.45)
5 eq 5
by using a radius parameter r0 ∼ 1.2 fm and the mass number A.
We have
1 2
A
r 2 ≡ r (5.46)
A k=1 k

1
Nmax
3
= N+ (N + 1)(N + 2) × 2 (5.47)
A N =0 M N ω 2
1 1
≈ (Nmax + 2)4 (5.48)
A MN ω 2
from the definition of the mean-square radius. We used the following virial theorem,

1 1 1 3
M N ω r N = V N = E N = ω N +
2 2
, (5.49)
2 2 2 2
to transform from Eq. (5.46) to (5.47).

On the other hand, we have

Nmax
2
A= (N + 1)(N + 2) × 2 ≈ (Nmax + 2)3 . (5.50)
N =0
3
We assumed Nmax + 2 1 in deriving Eqs. (5.48) and (5.50). Equations (5.48) and
(5.50) lead to

1 3 4/3
r 2 = A1/3 . (5.51)
2 2 MN ω
5.5 Mass Number Dependence of the Level Spacing ω 125
Thus we obtain
4/3
1 3 (c)2 5 −2 −1/3
ω = r A ∼ 41.1A−1/3 MeV (5.52)
2 2 M N c2 3 0
from Eq. (5.45).
5.6 The Magnitude and Origin of Spin–Orbit Interaction
It will be worth commenting on the magnitude and origin of the spin–orbit interaction
[15]. For nuclei with large mass numbers, the spin–orbit interaction is suggested to
be
VLS ∼ −20ˆ · ŝ A−2/3 MeV (5.53)
from the experimental data for the energy difference between the j = + 1/2
and j = − 1/2 states split by the spin–orbit interaction. Since ∼ k F R =
(9π/8)1/3 A1/3 for nucleons in the vicinity of the Fermi surface, VLS becomes non-
negligible compared with ω given by Eq. (5.52), and plays an important role in
determining magic numbers.
The spin–orbit interaction used in Eq. (5.27) is the one-body spin–orbit interaction.
It originates partly from the two-body spin–orbit interaction, which we studied in
Sect. 3.2.2, by averaging over the coordinates of one of the nucleons. The tensor
force is another source of the one-body spin–orbit interaction. In the relativistic
mean-field theory which we learn in Sect. 6.3, the spin–orbit interaction originates
from the coupling of nucleons to the σ -, ρ-, ω-mesons, and depends on the isospin
of nucleons due to the contribution of the ρ-meson.
5.7 Difference Between the Potentials for Protons

and for Neutrons: Lane Potential
The potential given by Eq. (5.1) corresponds to the Hartree potential which we learn
in Sect. 6.1. More precisely, the strength of the potential V0 depends on the isospin
of nucleons reflecting, e.g., the exchange effects, and is replaced by
1 N−Z
V = V0 − tˆz V1 , (5.54)
2 A
if we take the definition that tˆz |n = 21 |n , and tˆz |p = − 21 |p . Hence the potential for
nucleons depends on the isospin of nucleons. The isospin dependent term is called
the Lane potential. One can estimate the strength of V1 as

(pot) (kin)
V1 ∼ 4bsym = 4 bsym − bsym ≈ 100 MeV
from the magnitude of the contribution of the potential energy to the symmetry energy
term in the mass formula.
Exercise 5.10 Show that

(pot) 1 (N − Z )2
E sym ≈ V1 (5.55)
8 A
if we express the magnitude of the potential energy given by the Lane potential as
(pot) (pot)
E sym , hence that V1 ∼ 4bsym .
The Lane potential can be derived in the following way. Suppose that the interac-
tion between two nucleons is given by
V (x1 , x2 ) = −V0 (w + m PM )v(r1 − r2 ) = −V0 (w + m PM )v(r12 ) (5.56)

1
= −V0 w − m (1 + τ 1 · τ 2 )(1 + σ 1 · σ 2 ) v(r12 ) , (5.57)
4
where PM is the Majorana operator, and w and m are the strength parameters of the
Wigner and Majorana forces, respectively. By ignoring the correlation between the
nucleons, and adding on average with respect to the second nucleon, we obtain

m m 1
V (r1 ) = −V0 w − 1− (t̂ · T̂) V (r1 ) , (5.58)
4 w− m
A
4
V (r1 ) ≡ ρ(r2 )v(r12 )dr2 , (5.59)
for the nucleon potential. T̂ represents the total isospin operator of nucleons excluding
the nucleon under consideration. The isospin dependent term is nothing but the Lane
potential. Let us consider two cases in order to estimate the magnitude. One of
the cases is the Serber exchange potential, where m = w = 1/2. In this case,
w − m/4 = 3/8, m/(w − m/4) = 4/3. The other case is the Volkov force, which
is often used for the study of cluster structure of light nuclei. In this case, m = 0.6,
w = 1 − m = 0.4. Hence w − m/4 = 0.25, m/(w − m/4) = 2.4. Since V0 and V1
in Eq. (5.54) are about 40 and 100 MeV, respectively, the Volkov force well agrees
with the phenomenological strength of the Lane potential.
Physically, the Lane potential can be considered to reflect the fact that the force
between proton and neutron is stronger than the forces between protons and between
neutrons. Since the difference is intimately related to the effect of the tensor force,
one can suppose that the Volkov interaction is a force which effectively incorporates
the effect of the tensor force.
5.8 The Spin and Parity of Low-Lying States of Doubly-Magic … 127
5.8 The Spin and Parity of Low-Lying States

of Doubly-Magic ±2 Nuclei and the Pairing Correlation
5.8.1 The Spin and Parity of the Ground and Low Excited
States of 210
82 Pb
The spin and parity of the ground and low excited states of magic number ±2 nuclei
provide a good example of showing the important role of the pairing correlation in
nuclei.
Consider 210
82 Pb as an example. Figure 5.1 suggests that the configuration for low-
lying states including the ground state is (2g9/2 )2n . The power 2 means to fill two
nucleons.8 Hence, from the law of addition of angular momentum, we expect that
there appear 0+ , 1+ , . . . , 7+ , and 8+ states in low-energy region.
Figure 5.6 shows the energy-level diagrams for 210 210
82 Pb128 and 84 Po126 in low-
+ + + + +
energy region [3]. There appear indeed the 0 , 2 , 4 , 6 , 8 states in low-energy
region in the figure for 21082 Pb as expected.
One can prove that the odd-spin states do not appear because of the Pauli exclusion
principle as follows. Let us use the total spin j to represent the quantum number of
the single-particle level. The state in which two neutrons or two protons fill the jα
and jβ states is then given by

|( jα jβ )J M a = N |( jα jβ )J M − P12 |( jα jβ )J M (5.60)
by making antisymmetrization in order to satisfy the Pauli exclusion principle. N is

the normalization constant. The lower index a denotes antisymmetrization and P12
is the operator to interchange the particles 1 and 2. We used that the wave function
in the isospin space is symmetric concerning the exchange of two nucleons, since
we are considering the case where the two nucleons are the same kind of nucleons.
One can rewrite the second term of Eq. (5.60) as

P12 |( jα jβ )J M = P12 | jα jβ m α m β jα jβ m α m β |J M
m α ,m β

= | jβ jα m β m α jα jβ m α m β |J M
m α ,m β

= (−1) jα + jβ −J | jβ jα m β m α jβ jα m β m α |J M
m α ,m β
= (−1) jα + jβ −J |( jβ jα )J M . (5.61)
We used the following property of the Clebsch–Gordan coefficient,
8 We used the lower index n in order to indicate that it is the energy level for neutrons.
Fig. 5.6 The energy-level

diagrams for 210 Pb and
210 Po (the energy is in MeV)
jα jβ m α m β |J M = (−1) jα + jβ −J jβ jα m β m α |J M , (5.62)
in transforming from the second to the third lines in Eq. (5.61). Finally, we obtain

|( jα jβ )J M a = N |( jα jβ )J M − (−1) jα + jβ −J |( jβ jα )J M . (5.63)
Equation (5.63) shows that only even numbers are allowed for the total spin J when
one fills two identical nucleons in the same jα = jβ = j orbit, because 2 j is an odd
number.
5.8.2 The Effect of δ-Type Residual Interaction: Pairing

Correlation
The shell model which assumes that nucleons move independently from each other
in a mean field under the constraint of Pauli exclusion principle well explains the
basic features of nuclei such as the magic numbers. However, it has to be refined by
taking into account the effects of residual interaction in order to explain the details
appearing in the experimental data.
We rewrite the Schrödinger equation
⎡ ⎤
2
⎣− ∇i2 + vi j ⎦ Ψ (x1 , . . . , x A ) = EΨ (x1 , . . . , x A ) (5.64)
i
2M N i> j
5.8 The Spin and Parity of Low-Lying States of Doubly-Magic … 129
for an A-nucleon system, where the nucleons are interacting with each other through
the two-body interaction v, into
⎧ ⎫
⎨ 2 ⎬
− ∇ 2 + U (xi ) + Vi j Ψ (x1 , . . . , x A ) = EΨ (x1 , . . . , x A )
⎩ 2M N i ⎭
i i> j
(5.65)
by introducing a mean field U (x). V is the residual interaction if we consider the
ĥ (0) ≡ −2 ∇ 2 /2M N + U as the unperturbed single-particle Hamiltonian of the shell
model which uses, e.g., the harmonic oscillator potential for U .
We learnt in Sect. 5.4.2 that the residual interaction plays an important role in the
magnetic moment. The degenerate 0+ , 2+ , . . . , 8+ levels in 210 82 Pb are also resolved
by the residual interaction.
Let us assume that the residual interaction between two nucleons is given by
V (r1 − r2 ) = −V0 δ(r1 − r2 ) . (5.66)
Using the perturbation theory, one can then show (see Appendix A.11) that the
energy of two-nucleon system, where each of the single-particle levels (n α α jα ) and
(n β β jβ ) is occupied by one nucleon and their angular momenta couple to the total
angular momentum J , changes by the amount of
V0
ΔE J = − 1 + (−1)α +β −J F(n α α , n β β )A( jα jβ ; J ) , (5.67)
2
where
∞
1 2
F(n α α , n β β ) = u (r )u 2n β β (r )dr , (5.68)
0 r 2 n α α
" #2
1 1 (2 jα + 1)(2 jβ + 1) 1 1
A( jα jβ ; J ) = jα jβ − J 0 ,
1 + δn α n β δα β δ jα jβ 4π 2J + 1 2 2
(5.69)
where u n (r ) is the radial wave function defined according to Eq. (5.22). Equa-
tions (5.67)–(5.69) show that the energy shift by the residual interaction depends
on the total spin J only through the factor A via the Clebsch–Gordan coefficient and
2J + 1. Table 5.3 shows the factor A as a function of J for the case when we fill
two identical nucleons in the same jα = jβ = 9/2 orbit. The table suggests that
the energy split among the 0+ , 2+ , . . . , 8+ states shown in Fig. 5.6 has been induced
by the short-range residual interaction of δ-function type. The J = 0 state appears
as the lowest energy level, because in that state two nucleons move in the same
orbit with opposite directions, so that the spatial overlap is large. Consequently, the
correlation energy of attractive δ-function type is maximized. The spatial overlap of
two nucleons gets smaller with increasing J . Accordingly the energy shift from the
energy in the independent particle model becomes smaller. Note that the energy shift
Table 5.3 The angular momentum dependence of the energy shift factor A for the case of jα =
jβ = 9/2
J 0 2 4 6 8
8π A 10 80/33 (2.42) 180/143 (1.26) 320/429 (0.75) 980/2439 (0.40)
for the J = 0 state is especially large, and that the interval between the successive
energy levels gets narrower with increasing J . The J dependence shown in Table 5.3
can be qualitatively understood from the following asymptotic formula which holds
when both j and J are large [16],
#2 $
"
1 1 2 1 4 j ( j + 1) (J + 21 )2 (J + 21 )2
j j − J0 ∼ −
2 2 π j + 21 (2 j + 1)2 J (J + 1) (2 j + 1)2
$ 2
2 J
∼ 1− . (5.70)
πj 2j
In this way, in the ground state of nuclei, two identical nucleons occupy two orbits,
which are time-reversed orbits to each other, as a pair of the total angular momentum
J = 0. This is called the pairing correlation in nuclei and corresponds to the Cooper
pair and superfluidity in condensed matter physics.9
5.9 Column: Superheavy Elements (SHE)

As stated in Sect. 2.3.4 concerning the stability with respect to fission, the liquid-drop model
suggests that the fission barrier disappears when the atomic number exceeds about 100, because
the Coulomb force which prefers to fission then overcomes the surface tension which resists
fission. Hence it predicts that there exist no heavy nuclei beyond Z ∼ 100. However, theoretical
calculations predict that there appear fission barriers due to shell effect and that the atomic
number Z = 114 and the neutron number N = 18410 are the magic numbers next to the
known magic numbers Z = 82 and N = 126 and that there exists a region of relatively stable
nuclei around the set of these numbers in the nuclear chart. They are called superheavy nuclei
and the corresponding elements are called superheavy elements (SHE). Extensive activities
to examine such theoretical prediction and to explore the limit of existence of nuclei in the
region of large mass number are continued.
Figures 5.7 and 5.8 show reports of the Joint Institute for Nuclear Research (JINR) in
Dubna, Russia and the RIKEN at Wako, Japan, which are currently the world centers of
experimental activities in this subject. In Fig. 5.7, the superheavy nuclei synthesized by the
so-called hot fusion which uses actinides such as Cf and Cm as the target and 48 Ca as the
projectile are shown by circles and the superheavy nuclei synthesized by the cold fusion which
9 One can rigorously discuss the level structure in the presence of the pairing correlation in an
algebraic way by using the quasi-spin formalism if one simplifies that only the pair whose total
angular momentum J is zero is affected by the attractive residual interaction [17].
10 These numbers vary somewhat depending on the theoretical models.
5.9 Column: Superheavy Elements (SHE) 131
Fig. 5.7 Superheavy elements and the shell correction energy (in MeV), taken from [18]
bombards Pb or Bi target with, e.g., Ni by crosses.11 The numbers attached to the contour
lines are the magnitudes of the shell correction energy.
As the names cold fusion and hot fusion suggest, superheavy elements are synthesized by
making two heavy nuclei with atomic numbers Z 1 and Z 2 collide to fuse to form a nucleus with
a larger atomic number Z 1 + Z 2 . A feature of heavy-ion fusion reactions used to synthesize
superheavy elements with large atomic number is that the fission barrier locates inside the
fusion barrier for the entrance channel. Therefore, it is not guaranteed that a compound nucleus
with a certain lifetime is formed even if the colliding nuclei succeed to approach inside
the fusion barrier for the entrance channel. Thus the formation cross section of superheavy
elements gets very small.
Together with the experimental studies, many theoretical studies have been undertaken to
elucidate the reaction mechanism and to suggest suitable combinations of the projectile and
target nuclei as well as experimental conditions for the synthesis of superheavy nuclei using
transport theories such as the stochastic differential equations and various other methods [19].
Superheavy elements have short lifetimes, and the cross section to synthesize them is
so small as a few events in the experiments of a few months. Hence it is a hard task to
experimentally confirm that the superheavy nuclei have been synthesized. The elements whose
atomic numbers are 107–112 have been synthesized at the GSI Helmholtz Centre for Heavy
Ion Research in Darmstadt, Germany, by the cold fusion reactions using Bi and Pb as the
target nuclei. The electromagnetic separator called SHIP (Separator for Heavy Ion reaction
Products) was used in a series of experiments, and the identification of the nuclear species of the
synthesized superheavy nuclei has been done by confirming that the synthesized superheavy
nuclei reach to a known nucleus after repeating α decays. As Fig. 5.8 shows, the experiments
at RIKEN in 2004 and 2005 also used the fact that the α-decay chains from the superheavy
nucleus have been observed and that the lifetime of the daughter nucleus 262 105 Db which caused
fission at the last stage does not contradict the known lifetime of the spontaneous fission
as the evidence of the synthesis of element 113. The synthesis of the isotope 278 113 of the
element 113 was confirmed by the experiment in August, 2012, where six successive α-chains
11 The reactions which have a large reaction Q-value and hence synthesize a superheavy nucleus
as the evaporation residue after emitting about three neutrons by the evaporation process are called
hot fusion, while the reactions which have a small Q-value and emit one neutron in synthesizing
a superheavy nucleus is called cold fusion. Since the average binding energy of nucleon is about
8 MeV, one neutron is emitted for each 8 MeV of the excitation energy.
Fig. 5.8 The observed decay chains which show the synthesis of the element 278 113 by the
209 Bi + 70 Zn → 278 113 + n reactions [20]
101 Md were observed [20]. Of the observed six α-decays the last two, i.e., the
leading to 254
α-decays of Db and Lr, agreed with the known α-decays. Thus, the synthesis of the element
113 became unquestionable. The right to name the element Z = 113 was then given to
Morita et al. at RIKEN by IUPAP/IUPAC in December, 2015. The name was announced
as Nh (Nihonium) on June 8, 2016. By comparison, for the experiments at JINR using the
hot fusion it has been an obstacle to confirmation of the synthesis of superheavy nuclei that
the decay chains from the synthesized nuclei do not connect to the known region. However,
those elements have been finally recognized by IUPAP/IUPAC, since the synthesis of the
112th element (283 Cn) by hot fusion has been reproduced at GSI, and in addition the number
of events of the synthesis of 114th and 116th elements has been increased by the double-
check experiments by the collaboration between JINR and Lawrence Livermore National
Laboratory, USA and it was accepted that their decay can be reasonably interpreted as the
decay through 283 Cn. The Z = 114 and Z = 116 elements have been named Fl (Flerovium)
and Lv (Livermorium), respectively, in May, 2012. Furthermore, the Z = 115, 117 and 118
elements, which were synthesized by hot fusion, were announced as Mc (Moscovium), Ts
(Tennessine), Og (Oganesson), respectively, on June 8, 2016.
References 133
References
1. D.R. Tilley et al., Nucl. Phys. A 708, 3 (2002)

2. National Nuclear Data Center. http://www.nndc.bnl.gov/nudat2/
4. A. Csótó, G.M. Hale, Phys. Rev. C 55, 536 (1997)
5. M.A. Preston, Physics of the Nucleus (Addison-Wesley, London, 1962)
6. A.M. Lane, R.G. Thomas, Rev. Mod. Phys. 30, 257 (1958)
7. T. Otsuka, T. Suzuki, R. Fujimoto, H. Grawe, Y. Akaishi, Phys. Rev. Lett. 95, 232502 (2005)
8. W.D. Knight et al., Phys. Rev. Lett. 52, 2141 (1984)
9. I. Hamamoto, Nucl. Phys. A 126, 545 (1969); 141, 1 (1970); 155, 362 (1970)
10. A. Bohr, B.R. Mottelson, Nuclear Structure, vol. II (Benjamin, New York, 1975)
11. N.J. Stone, At. Data Nucl. Data Tables 90, 75 (2005)
12. A. Arima, H. Horie, Prog. Theor. Phys. 12, 623 (1954)
13. T. Yamazaki, T. Nomura, S. Nagamiya, T. Katou, Phys. Rev. Lett. 25, 547 (1970)
14. F. Halzen, A.D. Martin, QUARKS and LEPTONS (Wiley, New York, 1984)
15. A. Bohr, B.R. Mottelson, Nuclear Structure, vol. I (Benjamin, New York, 1969)
16. A. Molinari, M.B. Johnson, H.A. Bethe, W.M. Alberico, Nucl. Phys. A 239, 45 (1975)
18. Yu. Oganessian, J. Phys. G 34, R165 (2007)
19. Here, we quote some of the related theoretical papers. (a) Papers concerning stochastic dif-
ferential equations: Y. Aritomo, T. Wada, M. Ohta, Y. Abe, Phys. Rev. C 59, 796 (1999); C.
Shen, G. Kosenko, Y. Abe, Phys. Rev. C 66, 061602 (2002); Y. Abe, D. Boilley, G. Kosenko,
C. Shen, Acta Phys. Pol. B 34, 2091 (2003); Y. Aritomo, Phys. Rev. C 80, 064604 (2009); V.
Zagrebaev, W. Greiner, Phys. Rev. C 78, 034610 (2008); (b) Paper on the quantum diffusion
theory: N. Takigawa, S. Ayik, K. Washiyama, S. Kimura, Phys. Rev. C 69, 054605 (2004); (c)
Paper using coupled-channels calculations: N. Rowley, N. Grar, K. Hagino, Phys. Lett. B 632,
243 (2006)
20. K. Morita et al., J. Phys. Soc. Jpn. 81, 103201 (2012)
Chapter 6
Microscopic Mean-Field Theory
(Hartree–Fock Theory)
Abstract In the last chapter we discussed the shell structure of nuclei by assuming
a simplified mean field such as the Wood–Saxon potential or the harmonic oscillator
potential, which appears realistic. However, properly speaking, it is necessary to self-
consistently determine the mean field and the whole nucleus as well as the individual
states of nucleons inside the nucleus, because the mean field itself is determined
based on the interaction between constituent nucleons. In this chapter we describe
the basic ingredients of the Hartree–Fock theory, which is a representative of such
self-consistent approaches, and of some related microscopic theories. It will not
be an overstatement to say that the self-consistent calculations from microscopic
points of view are at the center of current theoretical studies of nuclear structure.
An important aspect in performing microscopic calculations such as the Hartree–
Fock calculations is that the force between the constituent particles, i.e., the nuclear
force, is not necessarily well established or involves some features which make the
theoretical treatment difficult in contract to the systems which are objects of, e.g.,
condensed matter physics, where the interaction between the constituent particles
is the well-known Coulomb interaction. The studies based on microscopic theories
utilizes interdependence of theory and experiments. One analyzes experimental data
with theories using the new nuclear force which looks more realistic. One then
improves the nuclear force so as to incorporate the results of the analysis, then
reconfirms thus obtained new theoretical predictions with experimental data.
6.1 Hartree–Fock Equation
In the Hartree–Fock theory, we approximate the solution of Eq. (5.64) by a single

Slater determinant such that

ψ1 (x1 ) ψ1 (x2 ) · · · ψ1 (x A )

1 ψ2 (x1 ) ψ2 (x2 ) · · · ψ2 (x A )
Ψ (x1 , . . . , x A ) = √ . .. .. .. , (6.1)
A! .. . . .
ψ A (x1 ) ψ A (x2 ) · · · ψ A (x A )

DOI 10.1007/978-4-431-55378-6_6
136 6 Microscopic Mean-Field Theory (Hartree–Fock Theory)
and determine the single-particle wave function {ψ} by the variational principle.
From the condition of the variational principle
δ
(Ψ |H |Ψ − EΨ |Ψ ) = 0 , (6.2)
δψi∗ (x j )
we obtain

2
− Δ + ΓH (r) ψk (r) + ΓEX (r, r )ψk (r )dr = εk ψk (r) , (6.3)
2M N
A

ΓH (r) = v(r − r ) |ψ j (r )| dr = v(r − r )ρ(r )dr ,
2
(6.4)
j=1

A
ΓEX (r, r ) = −v(r − r ) ψ ∗j (r )ψ j (r) = −v(r − r )ρ(r, r ) , (6.5)
j=1

A
ρ̂ ≡ | j j| . (6.6)
j=1
We assumed a central potential of the Wigner-type for the nucleon–nucleon interac-

tion v. Equation (6.3) is called the Hartree–Fock equation, and ΓH and ΓEX are the
Hartree and Fock or exchange potentials, respectively. Note that the Hartree potential
matches the potential assumed in Eq. (5.1) and that the Fock potential is non-local
if the range of the nuclear force is finite.
Equation (6.3) which determines single-particle wave functions and the corre-
sponding single-particle energies has to be solved self-consistently, since the Hartree
and Fock potentials depend on the density of the nucleus at each position. The his-
torically standard method for that aim is to expand the single-particle wave functions
{ψ} in terms of the wave functions of the harmonic oscillator model which we learnt
in the previous section and determine the expansion coefficients by the iteration
method. Recently, however, it is getting popular also to use an alternative method
which directly solves the Hartree–Fock equation by numerical integration without
introducing the expansion basis by dividing the whole space into small meshes. Since
the accuracy of the expansion method depends on the size of the expansion basis, the
direct integration method is useful, especially for the currently active area of study
of unstable nuclei where the binding energy of valence nucleons is small.
6.1.1 Equivalent Local Potential, Effective Mass
As we learnt in the previous chapter, the shell model usually assumes a local mean
field. One can convert the integro-differential equation due to non-local potential
into an equivalent local equation if certain conditions are fulfilled. One can then
6.1 Hartree–Fock Equation 137
understand the characteristic effects caused by the non-local potential in different

physical terms.
We show in Appendix A.5 that the effect of non-locality can be represented by
the energy or momentum dependence of the effective local potential if the change
of the potential within a wave length is small and hence if the wave function can
be well approximated in the WKB approximation, i.e., by an extension of the plane
wave using the local wave number. Furthermore, if the non-locality is moderate, the
non-local effect can be represented in terms of a position-dependent effective mass.
6.1.2 Nuclear Matter and Local Density Approximation
6.1.2.1 Local Density Approximation
As Eq. (6.5) shows, the Fock potential depends on the non-diagonal matrix elements
of the density operator. Its primary effect can be understood by considering the
nucleus locally as a nuclear matter1 and by approximating the wave functions of
nucleons with the plane waves distributed up to the Fermi surface corresponding to the
density ρ((r + r )/2) at each position. This is called the local density approximation.
Following this idea, ρ(r, r ) is given by
2 kF (R)
1 V
ρLDA (r, r ) ≡ √ e−ik·r eik·r dk × ×2×2 (6.7)
V 0 (2π )3
1
= 3ρ(R) j1 (kF (R)s) , (6.8)
kF (R)s
where s ≡ r − r , R = (r + r )/2.
Exercise 6.1 Using Eqs. (6.3) and (6.8), show that the contribution of the Fock term
to the Coulomb energy is given by

2
1 r1 |ρ̂ (p) |r2 1 27 Z 2 e2 1
ECoul
ex
= − e2 dr1 dr2 × = − , (6.9)
2 |r1 − r2 | 2 16 R 3 (kF(p) )2
when the density is uniform. The last factor 1/2 in the second term is the factor
which takes into account the orthogonality in the spin space when two protons are
exchanged.
1 Such a medium made of nucleons which has uniform density and an infinite spatial extension is
called nuclear matter.
The Coulomb energy is thus given by

2/3
3 e2 Z 2 3
ECoul ≈ 1−5 (6.10)
5 R 16π Z
including the Fock term for nuclei with a large mass number in the local density
approximation.
6.1.2.2 Effective Mass in the Local Density Approximation
Let us study how the effective mass depends on the parameters of the nuclear force
and the nuclear density by using the local density approximation.2
We begin by noting that the potential energy U in the Hartree–Fock approximation
is given in general by
1
U= αβ|v|αβa , (6.11)
2 α,β=occupied
where α and β denote the total quantum numbers in the usual, the spin and the isospin
spaces, and the ket state |αβa represents the antisymmetrized state (|αβ − |βα).
The sum is carried over all states below the Fermi surface. We assume that the nuclear
force is the central force of the Yukawa type given by Eq. (3.62),
e−μr
v(r ) = V0 (w + m PM ) = v0 (r )(1 − m + m PM ) . (6.12)
μr
If we express v as Eq. (3.29) by introducing the projection operators, then the potential
in each projected state is given by
e−μr e−μr
Vtt (r ) = Vss (r ) = V0 (1 − 2m) , Vst (r ) = Vts (r ) = V0 . (6.13)
μr μr
Let us consider especially the infinite nuclear matter consisting of equal numbers
of protons and neutrons, where both spin-up and down states are occupied. In that
case, the total potential energy can be expressed as

1 Ω 2
U= d 3
k d 3
k (9Vtt(0) +3Vts(0) +3Vst(0) +Vss(0) )kk |v̂0 |kk
2 (2π )3 k<kF k <kF

− (9Vtt(0) − 3Vts(0) − 3Vst(0) + Vss(0) )kk |v̂0 |k k (6.14)
by taking into account the statistical weight due to the degree of degeneracy and the
symmetry property of the spatial wave function for each state of two-nucleon system.
2 Similar consideration can be found in [1, 2].

We expressed the interaction in each projected state as the product of a common factor
v0 (r ) = V0 e−μr /μr and a state dependent coefficient Vtt(0) , Vts(0) , Vst(0) and Vss(0) , which
are 1 − 2m, 1, 1 and 1 − 2m, respectively. Equation (6.14) shows that the weights
of the direct and exchange terms are given by
9Vtt(0) + 3Vts(0) + 3Vst(0) + Vss(0) = 16 − 20m ,

9Vtt(0) − 3Vts(0) − 3Vst(0) + Vss(0) = 4 − 20m . (6.15)
The following expressions can be obtained for the local and non-local terms in
Eq. (6.3) by performing the functional differentiation of U with respect to ψk∗ (r).

ΓH (r) = (4 − 5m) v0 (|r − r |)ρ (r )dr , (6.16)
ΓEX (r, r ) = −(1 − 5m)v0 (|r − r |)ρ (r, r ) , (6.17)

kF
kF
ρ̂ ≡ |k k | = |ψk ψk | . (6.18)
k =0 k =0
We divided each term by 4 by specifying spin and isospin in taking the functional
differentiation.
Let us now calculate the Fock term in the Hartree–Fock equation by using the

explicit form of the nuclear force v0 (r − r ) = V0 e−μ|r−r | /μ|r − r | and the result
of Eq. (6.8). In that procedure, we use the plane wave for the wave function accord-
ing to the nuclear matter approximation. The Fock term can then be expressed as
ΓEXELP
(r)ψk (r), where the equivalent local potential (ELP) for the Fock term ΓEXELP
(r)
is given by
−μs
1 e j1 (kF (r)s) ik·s
ΓEX
ELP
(r) = − (1 − 5m)V0 3ρ(r) e ds (6.19)
4 μs kF (r)s
∞
1
= −(1 − 5m)V0 3πρ(r) e−μs j1 (kF (r)s)ds
μkF (r) 0

1
−πρ(r) k + ···
2
(6.20)
μ(μ2 + kF2 )2
by noting that ρ (r, r ) = ρ(r, r )/4. This leads to the following expression for the
effective mass,

1 1 M N c2 1
= 1 + 2π(1 − 5m)V0 ρ(r) . (6.21)
M∗ MN (c)2 μ(μ2 + kF2 )2
Equation (6.21) shows that the contribution of the Fock term to the effective mass
is zero, i.e., M ∗ = M N , in the limit when the range of the nuclear force is zero, i.e.,
when μ → ∞, and gets larger with increasing range. The same equation as Eq. (6.21)
can be obtained also from the evaluation of the energy of nucleons with momentum
k. If we take kF = 1.27 fm−1 , ρ = 0.138 fm−3 (Table 2.2), m = 0.5 (Serber force),
V0 = −48.1 MeV, a = 1/μ = 1.17 fm (see Eq. (3.64)), then M N∗ = 0.76M N . This
well agrees with the results of the standard Hartree–Fock calculations such as the
Skyrme Hartree–Fock calculations, which we learn in the next section.
Incidentally, the Hartree potential given by Eq. (6.16) becomes
3
2 kF
ΓH (r) = V0 (4 − 5m) (6.22)
3π μ
in the nuclear matter approximation.
6.1.3 Saturation Property in the Well-Behaved Potential,

Constraint to the Exchange Property
A characteristic of the potential given by Eq. (6.12) in comparison with potentials

with repulsive core which have been introduced in Chap. 3 is that one can perform
Hartree–Fock calculations as it is, since the matrix elements of the nuclear force
do not diverge.3 We comment here on how the saturation property, i.e., the fact
that nuclei do not collapse or that nuclei exist with a finite density, is guaranteed
within the framework of the Hartree–Fock theory for such a well-behaved potential.
Similarly to Sect. 6.1.2.2 we consider the infinite nuclear matter consisting of the
same numbers of neutrons and protons, where both spin-up and down states are
occupied, and assume that the radial dependence of the nuclear force is given by
v0 (r ) = V0 e−μr /μr similarly to Eq. (6.12). Then, the condition
4
16 − 20m < 0, i.e., m > , (6.23)
5
has to be satisfied in order for nuclei to exist with a finite density as we can see from
Eq. (6.22).
The condition given by Eq. (6.23) contradicts the requirement of two-nucleon
scattering, which suggests m ∼ 1/2 (for example, Serber force). In this way, it is dif-
ficult to satisfy the requirement of the saturation property with well-behaved poten-
tials such as that given by Eq. (6.12) solely by the exchange property of the nuclear
force. Hence it is necessary to attribute the saturation property to some effective
3 Such potentials are often called well-behaved potentials.

many-body interaction, which takes into account the medium effects. The Skyrme
interaction, which we describe in the following, is one example of such interactions.4
6.2 Skyrme Hartree–Fock Calculations for Finite Nuclei
6.2.1 Skyrme Force
It is not so easy to systematically study nuclear structure from the point of view
of a quantum many-body problem based on the Brueckner theory (see Sect. 3.8).
On the other hand, the Hartree–Fock calculations using a phenomenological nuclear
force began in early 1970 s and have been very popular up to now. Especially, the
Hartree–Fock calculations using the simple force invented by Skyrme are known
under the name of the Skyrme Hartree–Fock calculations [3] and have become a
standard calculation for nuclear structure.
Skyrme invented the extreme nuclear force of δ-function type to emphasize the
short-range property of the nuclear force. In addition, he introduced a three-body
force to guarantee the saturation property,

V = V (i, j) + V (i, j, k) . (6.24)
i< j i< j<k
(A) Two-Body Force The force currently used as the Skyrme force, which is an
extension of the force introduced originally by Skyrme himself, reads
V (1, 2) = t0 (1 + x0 Pσ )δ(r1 − r2 )
1
+ t1 (1 + x1 Pσ ) δ(r1 − r2 )k2R + k2L δ(r1 − r2 )

2
+ t2 (1 + x2 Pσ )k L · δ(r1 − r2 )k R
+ iW0 (σ 1 + σ 2 ) · k L × δ(r1 − r2 )k R , (6.25)
where Pσ is the spin exchange operator defined by Eq. (3.18), and k R and k L are
defined by
1
kR ≡ (∇ 1 − ∇ 2 ) , (6.26)
2i
1
k L ≡ − (∇ 1 − ∇ 2 ) , (6.27)
2i
and operate on the right and the left, respectively.
4 The study of equation of state using the simple perfect rigid sphere model shows that the saturation
property is guaranteed by the repulsive core in the high density region higher than about six times
the standard density ρ0 ∼ 0.14 fm−3 if one assumes the radius of the repulsive core to be 0.4 fm.
(B) Three-body force The three-body force5 which was originally introduced by
Skyrme reads
V (1, 2, 3) = t3 δ(r1 − r2 )δ(r2 − r3 ) . (6.28)
However,
currently, it is more popular to introduce an additional two-body term
V
i< j DD (i, j) using the density-dependent two-body force

1 r1 + r2
VDD (1, 2) = t3 (1 + x3 Pσ )ρ α δ(r1 − r2 ) , (6.29)
6 2

instead of the term of the three-body force i< j<k V (i, j, k). This model is designed
to guarantee the saturation and control the nuclear incompressibility through the value
of α, which is often assumed to be either 1/3 or 1/6. The latter gives, of course, a
smaller incompressibility.
The W0 term in Eq. (6.25) represents the two-body spin–orbit interaction. The t0
term corresponds to the limit of zero-range nuclear force, and is the attractive term
which is indispensable for the nucleus to exist as a self-bound system. The t1 and
t2 terms have been introduced to take into account the effects of the finite range
property of the nuclear force. In order to understand this, let us consider a two-body
central force with a finite range,6,7
v(r ) = V0 e−(r/a) .
2
(6.30)
In this case, the matrix elements using the eigenstates of the momentum operator are
given by
1 √ 3 2
k|v̂|k = (a π) V0 e−a (k−k ) /4
2
(6.31)
Ω
1 √ 1
= (a π)3 V0 1 − a 2 (k − k )2 + · · · . (6.32)
Ω 4

Equation (6.31) can be derived by inserting the completeness relation dr|rr| = 1
on both sides of the operator v̂ and using r|k = √1Ω eik·r and also that the matrix
element of a local potential v̂ in the Dirac notation is given by
5 This term is an effective interaction, which represents the many-body effects and the effect of
repulsive core appearing in the G-matrix theory and has been introduced to avoid the collapse of
nuclei, and should be distinguished from the genuine three-body force appearing in the discussion
of nuclear force based on the meson theory as the force involving three nucleons [1]. The magnitude
of the latter has been discussed through detailed calculations of the binding energy and the form
factor of light nuclei such as the triton and 3 He using, e.g., the Faddeev equation.
6 Instead of the Yukawa type, we assumed the Gauss type as a two-body central force to facilitate
calculations.
7 Here, we eliminate the center-of-mass motion and formulate only for the relative motion, and treat
it as a one-body problem.
6.2 Skyrme Hartree–Fock Calculations for Finite Nuclei 143
r|v̂|r = δ(r − r )v(r) . (6.33)
Since Eq. (6.32) is an expansion with respect to the range of the nuclear force
a, let us keep up to the first term representing the finite range effect, i.e., up to the
second order with respect to a in Eq. (6.32), and transform back to the coordinate
representation. We then have

√ 3
r|v̂|r = (a π ) V0 δ(r)δ(r )

1
+ a 2 ∇r2 δ(r)δ(r ) + δ(r)∇r2 δ(r ) + 2∇r δ(r) · ∇r δ(r ) , (6.34)

4
Ω

which can be obtained by inserting the completeness relation (2π)3
dk|kk| = 1
on both sides of v̂. Correspondingly, one obtains

drdr F(r)r|v̂|r G(r )

√
= (a π )3 V0 drdr δ(r)F(r)δ(r )G(r )

1
+ a2 drdr δ(r) (∇r2 F(r))δ(r )G(r ) + F(r)δ(r )(∇r2 G(r ))

4

1 2
+ a drdr δ(r)∇r F(r)δ(r ) · ∇r G(r ) (6.35)
2
for functions F(r) and G(r ). This means that the v(r) approximated up to the second
order of the range of the force is given by
2 2
√ 3 1 2 1 1
v(r) ≈ (a π ) V0 δ(r) − a − ∇ δ(r) + δ(r) ∇
4 i L i R

1 1 1
+ a 2 − ∇ δ(r) ∇ . (6.36)
2 i L i R
By comparing Eq. (6.25) with√Eq. (6.36), one can expect that t0 < 0, t1 > 0, and
t2 < 0. Also, the coefficient (a π )3 in Eq. (6.36) can be understood by considering
2 √
one of the representations of the δ-function, δ(x) = lima→0 e−(x/a) / πa .
6.2.2 Skyrme Hartree–Fock Equation
In the Skyrme Hartree–Fock calculations, thanks to the feature of the interaction that
it is of the δ-function type one can explicitly write down the Hartree–Fock equation
in terms of the parameters of the nuclear force and the nucleon density ρ [3]. We give
here the Hartree–Fock equation for the N = Z system in the case when the Coulomb
force is ignored and when we assume the original interaction by Skyrme, i.e., when
we set x1 = x2 = 0 for the two-body force, and assume the effective three-body force
given by Eq. (6.28),

2
−∇ · ∇ + U (r) − iW(r) · (∇ × σ ) ψi = εi ψi , (6.37)
2M N∗ (r)
where the effective mass M N∗ (r), scalar U (r) and spin–orbit W(r) potentials are
given as
2 2 1
∗ = + (3t1 + 5t2 )ρ(r) , (6.38)
2M N (r) 2M N 16
3 3 1 1 3
U (r) = t0 ρ + t3 ρ 2 + (3t1 + 5t2 )τ + (5t2 − 9t1 )∇ 2 ρ − W0 ∇ · J ,
4 16 16 32 4
(6.39)
3 1
W(r) = W0 ∇ρ + (t1 − t2 )J , (6.40)
4 16
with
3 2
τ (r) = |∇ψi (r, s, q)|2 ∼ k ρ, (6.41)
q=p,n i,s
5 F

J(r) = −i ψi∗ (r, s, q) ∇ψi (r, s , q) × s|σ |s . (6.42)

q=p,n i,s,s
The τ and J are the kinetic energy and the spin–orbit densities, respectively. There
appears the effective mass M N∗ (r ) in the Hartree–Fock equation reflecting the
momentum-dependent terms in the Skyrme force (see Eq. (6.25)). In accord with
the result of Eq. (6.21), the effective mass stays the same as the bare nucleon mass
in the limit of zero-range force, i.e., in the limit of t1 = 0, t2 = 0, and gets smaller
with increasing range of the nuclear force.
6.2.3 Energy Density and Determination of Parameters
In the case of the Skyrme force we can express the interaction energy as a functional
of the nucleon density. On the other hand, in the Thomas–Fermi approximation, the
kinetic energy, e.g., for protons, is given by
(p)
pF (r ) (p)
(p) 1 p2 3 (kF (r ))2
E kin = dpdr × 2 = ρp (r)dr . (6.43)
(2π )3 2M N 5 2M N
Consequently, if we represent the total energy as

E = E Skyrme = H (r)dr (6.44)
using the energy density H , then H can be expressed as a functional of the proton
and neutron densities. Especially, if we consider the case of N = Z and ignore the
Coulomb force, then H is given by
2 3 1 1
H = τ + t0 ρ 2 + t3 ρ 3 + (3t1 + 5t2 )ρτ
2M N 8 16 16
1 3 1
+ (9t1 − 5t2 )(∇ρ)2 − W0 ρ∇ · J + (t1 − t2 )J2 , (6.45)
64 4 32
corresponding to Eqs. (6.37)–(6.42). Let us next move to the problem of determining
the values of parameters such as t0 and t1 . To that end, let us consider the case of
N = Z and the nuclear matter, where the surface effect can be ignored. In this case,
the derivative terms in Eq. (6.45) can be ignored, and the total energy per nucleon is
given by
E HV 3 3 1 3
= = TF + t0 ρ + t3 ρ 2 + (3t1 + 5t2 )ρkF2 , (6.46)
A ρV 5 8 16 80
where TF = 2 kF2 /2M N is the Fermi energy. Equation (6.46) leads to

∂2 E 6 9 15 3
K ≡ kF2 = TF + t0 ρ + t3 ρ 2 + (3t1 + 5t2 )ρkF2 (6.47)
∂kF2 A 5 4 8 4
for the incompressibility of the nuclear matter.8 On the other hand, the pressure is
given by
∂ ∂ E
P=− E = ρ2 (6.48)
∂V ∂ρ A
by definition when the temperature is 0.

Exercise 6.2 Express the pressure as a function of the nucleon density ρ and the
parameters of the nuclear force.
It is known from the mass formula that the binding energy per nucleon is about
15 MeV for the nuclear matter. On the other hand, the incompressibility K is
suggested to be about 200 MeV9 from the excitation energy [4] of the isoscalar
8 Alternatively, the nuclear incompressibility is defined also by k ≡ −V ∂∂V P = ρ ∂∂ρP . The two
0)
nuclear incompressibilities are related as K = 9k(ρ ρ0 if we denote the equilibrium density, which
gives P(ρ0 ) = 0, by ρ0 .
9 It is, however, not so easy to determine the incompressibility of nuclear matter to a good accuracy
from the data of actual nuclei with finite sizes.

monopole vibration (breathing mode of vibration; the excitation motion of the

nucleus, where the protons and neutrons make in phase collective motion, and the
nucleus repeats expansion and contraction of the volume without changing the shape)
and the study of astrophysical phenomena such as neutron stars and supernova explo-
sions. Also, the pressure has to be 0 at the standard density, i.e., at the saturation
density ρ = ρ0 ∼ 0.17 fm−3 , since nuclei are isolated systems. The parameters of
the Skyrme force are determined so as to satisfy these conditions and to reproduce
the radii and the binding energies of some of the representative nuclei such as 16 O
and 208 Pb, and several different parameter sets have been proposed depending on the
experimental data, which have been taken into account. Table 6.1 lists some exam-
ples.
SIII [5] is one of the frequently used parameter sets, where the values of parameters
have been determined so as to reproduce the experimental data of the binding energy
and the radius of spherical nuclei 16 O, 40 Ca, 48 Ca, 56 Ni, 90 Zr, 140 Ce, and 208 Pb.
SkM∗ [6] determines the parameters to reproduce the fission barrier of 240 Pu as well.
Alternatively, SLy4 [7] determines the parameters so as to be suitable for the study of
neutron-rich nuclei and the neutron matter by paying attention to the isospin degree
of freedom. Table 6.1 shows also the properties of the nuclear matter calculated
with each parameter set. Since the nuclear incompressibility K is suggested to be
about 200 MeV from the excitation energy of the compressional vibration, i.e., the
monopole vibration, the size of the neutron star and also supernova explosion as stated
Table 6.1 The values of parameters of the Skyrme force and the properties of the nuclear matter
obtained by using them. ρ0 is the baryon number density, E/A is the binding energy per nucleon,
K is nuclear incompressibility and m ∗N is the effective mass of nucleon in the central region
SIII SkM∗ SLy4
t0 (MeV fm3 ) −1128.75 −2645.00 −2488.91
t1 (MeV fm5 ) 395.0 410.00 486.82
t2 (MeV fm5 ) −95.0 −135.00 −546.39
t3 (MeV fm3+α ) 14000.0 15595.00 13777.0
x0 0.45 0.09 0.834
x1 0.00 0.00 −0.344
x2 0.00 0.00 −1.000
x3 1.00 0.00 1.354
α 1 1/6 1/6
W0 (MeV fm5 ) 120.0 130.0 123.0
Properties of the nuclear matter
ρ0 (fm−3 ) 0.145 0.1603 0.160
E/A (MeV) −15.87 −15.78 −15.969
m ∗N /M N 0.76 0.79 0.70
K (MeV) 356 216.7 229.9
before, SkM∗ and SLy4 which give soft equation of state (EOS) look reasonable,
while SIII gives too hard EOS.10
6.2.4 Comparison with the Experimental Data
The Hartree–Fock calculations assuming the Skyrme force well reproduce the exper-
imental data of fundamental observables such as the nuclear radii and the charge
distributions. As an example of the comparison with the experimental data, Fig. 6.1
compares the energy levels for nucleons obtained by the Hartree–Fock calculations
using SkM∗ and SLy4 forces with the experimental data. The figure shows that the
Hartree–Fock calculations with the Skyrme force well reproduce the features of the
experimental data for the single-particle levels.
However, if one examines closely, it is noticed that the Hartree–Fock calculations
give too large level spacing, i.e., smaller level density, compared with the experimen-
tal data. That trend is more significant for the SLy4 force than for the SkM∗ force,
especially, for neutrons. The two forces give similar results for protons. In order to
see the origin of this problem and its feature, we compare in Fig. 6.2 the effective
mass and potentials of nucleons obtained for each calculation. The figure shows that
Fig. 6.1 Nucleon energy levels near the Fermi surface. Comparison of the experimental data with
the Hartree–Fock calculations using the SkM∗ and SLy4 forces
10 It is known that the fragmentation reactions observed in medium energy nucleus–nucleus colli-
sions also reflect the hard or soft property of the EOS [8].
Fig. 6.2 The effective mass and the potential for nucleons in 208 Pb for SkM∗ (left) and for SLy4
(right)
the effective mass in the Hartree–Fock calculations is significantly smaller than the
bare mass. Moreover, it shows that the effective mass for neutrons is considerably
smaller for SLy4 force than that for SkM∗ force. On the other hand, the effective mass
for protons is nearly the same for two forces. Since the level spacing in the square
well potential increases with decreasing mass (see Eq. (5.10)), the above mentioned
problem of the Hartree–Fock calculations concerning the nucleon level density could
be sought one of its origins in the incorrect treatment of the effective mass, i.e., in
that they ignore the e-mass, which we discuss later.
6.2.5 The Equation of State, Saturation, Spinodal Line,

Surface Thickness
6.2.5.1 The Equation of State and Saturation
The equation of state is one of the fundamental properties of a physical system and
is often expressed by representing the pressure as a function of density. One can
analytically write down the equation of state if one assumes the Skyrme force. The
equation of state plays an important role in various problems such as the nuclear
saturation, compressional vibrational excitation, nuclear fragmentation reactions11
and astrophysical phenomena such as neutron stars and supernova explosions. Here,
we discuss the equation of state for the nuclear matter with N = Z based on the
Skyrme force. We ignore the Coulomb force.
11 Nuclear reactions observed in the nucleus-nucleus collisions at medium energies of several tens
MeV per nucleon, where the colliding nuclei dissociate into nucleons and fragments with various
small mass numbers.
The pressure at finite temperature is given by the derivative of the Helmholtz free
energy F = E − T S:
∂
P=− F . (6.49)
∂V T
Since the internal energy E is given by the sum of the interaction energy E int and
the kinetic energy E kin , we divide the pressure into the part originating from the
interaction energy and that from the kinetic energy
P = PV + PT . (6.50)
(1) Pressure due to the Interaction Energy: PV We estimate the pressure originating
from the interaction by replacing the total energy E with the interaction energy E int
in the formula (6.48) at zero temperature. Corresponding to Eq. (6.46), let us assume
that the interaction energy per nucleon is given by
E int
= −C1 ρ + C2 ρ 1+α . (6.51)
A
We have changed the power in the second term on the right-hand side of Eq. (6.51) by
considering the general case when the nuclear force is given by Eq. (6.29) instead of
Eq. (6.28), although Eq. (6.46) corresponds to the case when α = 1.12 The pressure
due to the interaction is then given by
PV = −C1 ρ 2 + (1 + α)C2 ρ 2+α . (6.52)
(2) Pressure due to the Kinetic Energy: PT In order to obtain the pressure due to
the kinetic energy, we first calculate the kinetic energy per unit volume based on

E kin 4 2 k 2 1
= d3 k , (6.53)
V (2π )3 2k2
2M N exp[( 2M − λ)/T ] + 1
N
then obtain PT by inserting E kin in the place of E in Bernoulli’s formula [9, 10],
2
PV = E, (6.54)
3
which holds for both classical and quantum mechanical ideal gases irrespective of
the statistical property. λ in Eq. (6.53) is the Fermi level or Fermi potential.
(2a) The Degenerate Limit: the Case of T = 0 When T = 0, one can easily show
that
12 The prescription which takes into account the finite range property of the nuclear force through the
t1 and t2 terms following Skyrme would need to be reconsidered when we use α smaller than 2/3 in
discussing the saturation, since the last term on the right-hand side of Eq. (6.46) is proportional to
ρ 5/3 .
2 2 2 kF2
PT = ρεF = ρ ∝ ρ 5/3 , (6.55)
5 5 2M N
where εF is the Fermi level or Fermi energy at T = 0.

(2b) The Case of Non-degenerate System at High Temperature: At high temperatures,
one can approximate the Fermi distribution by the Boltzmann distribution. We then
have
∞
E kin 4 2 2 k 2
4 −( 2M N −λ)/T
∼ 4π dkk e
V (2π )3 2M N 0
4 1 1 λ/T 3 √
∼ 4π e π (2M N T )5/2 , (6.56)
(2π ) 3 2M N 3 8
∞
N 4 2 2
−( k −λ)/T
∼ 4π dkk 2 e 2M N
V (2π ) 3
0
4 1 λ/T 1 √
∼ 4π e π (2M N T )3/2 , (6.57)
(2π )3 3 4
and obtain
E kin 3
= ρT . (6.58)
V 2
Using Bernoulli’s formula, the pressure due to the kinetic energy is finally given by
PT = ρT , (6.59)
which is well known for the ideal gas.

(2c) The Case of Fermi Gas at Low Temperature: The kinetic energy at temperature
T is given by
1 π2
E kin (T ) = E kin (T = 0) + AT 2 (6.60)
4 εF
when the temperature T satisfies the condition T λ. The pressure due to the kinetic
energy is therefore given by

2 5 2 T 2
P = ρεF 1 + π (6.61)
5 12 εF
by using Eq. (6.54).

Let us guess the behaviour of the equation of state based on these results. We first
consider the case when the temperature is 0. Equations (6.52) and (6.55) suggest that
the effect of kinetic energy, i.e., the degeneracy pressure, dominates in the region
of low density and the pressure is positive and increases with density. Eventually,
the pressure due to attractive two-body interaction becomes to overcome and the
Fig. 6.3 The conceptual

illustration (solid lines) of
the equation of state of
nuclear matter (isothermal
lines). The dotted and the
dashed lines are the
isothermal and isentropic
spinodal lines, which connect
the points where ( ∂∂ρP )|T = 0
and ( ∂∂ρP )| S = 0, respectively
(taken from [13])
pressure decreases with density and turns to be negative. For even higher density,
the pressure turns to increase with density because of the three-body force or the
density-dependent force, i.e., the second term on the right-hand side of Eq. (6.52),
which have been introduced to guarantee the saturation, and becomes 0 at the normal
density ρ0 ∼ 0.17 fm−3 , for which the nucleus exists as a stable isolated system, and
then continues to increase. However, it is necessary to refine the theory by taking
into account the effect of repulsive core and various effects, which are ignored in
Eq. (6.55), in order to discuss the region of very high density.13 As Eq. (6.61) shows,
the pressure increases with increasing temperature and eventually converges to the
equation of state for the classical ideal gas given by Eq. (6.59).
Figure 6.3 shows the equation of state drawn based on these considerations.
Roughly speaking, the pressure behaves like a cubic function of the density at low
temperatures, and becomes a monotonically increasing function of density at tem-
peratures higher than the critical temperature TC .
Finally, we express the equation of state in the form of the energy density H (ρ).
To that end, we remark that the incompressibility defined by Eq. (6.47) can be
expressed also as
2
2 ∂ E
K = 9ρ , (6.62)
∂ρ 2 A ρ=ρ0
13 The equation of state at high density is one of the crucial issues in the problems of the neutron stars
and the supernova explosions. However, it is not precisely known yet. Concerning neutron stars, it
has been reported that the pressure gets lower if one takes into account baryons other than protons
and neutrons, that is, hyperons. This is natural, because for a given baryon density the degeneracy
pressure must decrease if more different kinds of baryon exist. A problem is that the softening
of the equation of state by the presence of hyperons contradicts the experimental observation of
massive neutron stars with the mass as large as about two times the solar mass. This problem
is often referred to as the hyperon puzzle or the hyperon crisis. Various possibilities such as the
recovery of the stiffness of the equation of state at the high density side due to the repulsive effect
of a three-body force working universally among all baryons [11] are now under investigation to
resolve this puzzle and to establish a theory which is compatible with the data of both neutron stars
and hypernuclei [12]. In this connection, we note that the density at deep inside of neutron stars is
thought to be significantly higher than the normal nuclear density, although the actual value is not
known and varies over a wide range between 4 and 8 times the normal nuclear density depending
on the theoretical models.
where ρ0 is the density of nucleus in the ground state. Equation (6.62) shows that
the incompressibility corresponds to the curvature in the case when we expand the
energy per nucleon as a function of the density ρ in the vicinity of ρ0 . Hence the
H (ρ) can be expressed as

E K ρ 2 1
H (ρ) = ρ = ρ 1− + ε0 + B(∇ρ)2 , (6.63)
A 18 ρ0 2
with
1
B= [9t1 − (5 + 4x2 )t2 ] + δ BW , (6.64)
32
1 2 1
δ BW = , (6.65)
18 2M N ρ
where ε0 is the binding energy per nucleon ∼ −15 MeV. We used the extended
Thomas–Fermi approximation,
3
τ (ρ) = τTF (ρ) + τ2 (ρ) = (3π 2 )2/3 ρ 5/3 + τ2 (ρ) , (6.66)
5
1 (∇ρ)2 1
τ2 (ρ) = + Δρ , (6.67)
36 ρ 3
for the kinetic energy, since we have taken into account the (∇ρ)2 term, which is
related to the surface energy, in the interaction energy. The first term on the right-
hand side of Eq. (6.67) is called the Weizsäcker correction. The last term δ BW in
Eq. (6.64) is the term representing the Weizsäcker correction.
6.2.5.2 Liquid–Gas Phase Transition, Spinodal Line
The behaviour shown in Fig. 6.3 resembles the van der Waals equation of state.
Like van der Waals, the problem of liquid–gas phase transition in nuclei, which
are self-bound systems with finite number of constituents interacting through strong
interactions, has been discussed based on this equation of state in connection with
the fragmentation reactions in medium energy nucleus–nucleus collisions [14].
The nucleus once becomes a state of high temperature and high density after the
collision at high energy, then cools down nearly isentropically with expansion, and
eventually is observed as fragments of various masses including nuclei with relatively
large mass number. In this procedure, the region of negative nuclear incompressibility
plays an important role as the region of mechanical instability. This can be understood
in the following way.
Let us consider the Helmholtz free energy density F ≡ FA ρ corresponding to
the Hamiltonian density H ≡ EA ρ introduced by Eq. (6.44). By taking the second
Fig. 6.4 The unstable region

(the oblique lines: the
boundary is called the
spinodal line) and the
compressed region (shaded
region) which is expected to
lead to the fragmentation
reaction in heavy-ion
collisions. The dotted line is
an example of the trajectory
in heavy-ion collisions. The
figure is after D.K. Scott
derivative of F with respect to the density, we have

∂ 2F ∂ F ∂2 F
=2 +ρ 2 . (6.68)
∂ρ 2 ∂ρ A ∂ρ A
On the other hand, from the definition of pressure (6.49), we have

∂P ∂ F ∂2 F
= 2ρ +ρ 2
. (6.69)
∂ρ ∂ρ A ∂ρ 2 A
Equations (6.68) and (6.69) lead to the relationship,
∂ 2F 1 ∂P
= . (6.70)
∂ρ 2 ρ ∂ρ
This relationship shows that the negative gradient of the pressure, i.e., negative incom-
pressibility, is equivalent to that the second derivative of the free energy density with
respect to density is negative.
∂P ∂ 2F
<0 ↔ <0. (6.71)
∂ρ ∂ρ 2
The condition ∂∂ρF2 < 0 means that the system is unstable with respect to the density
2
fluctuation, i.e., the mechanical instability. It is because the change of the free energy
density due to a local fluctuation of the density is given by
1
ΔF = [F (ρ + Δρ) + F (ρ − Δρ)] − F (ρ)
2
1 ∂ 2F
= (Δρ)2 , (6.72)
2 ∂ρ 2
and it becomes ΔF < 0 if ∂∂ρF2 < 0. The former, i.e., ΔF < 0, means that the free
2
energy gets lower in the state with density fluctuation.

Figure 6.4 shows the mechanically unstable region, whose boundary is called the
spinodal line, and the region which is compressed immediately after the collision
and is expected to be required in order to reach instability region after cooling in the
nucleus–nucleus collision at medium energies.
6.2.5.3 Surface Diffuseness
The surface property of nuclei is one of the intriguing physical quantities concern-
ing nuclear structure as well as nuclear reactions. Equation (6.63) suggests that the
parameter B given by Eq. (6.64) is related to the surface thickness. In fact, using
Eq. (6.63) one can show [15] that the density distribution of nuclei is given by

r−R
ρ = ρ0 tanh2 , (6.73)
2b
B
b2 = 9 ρ0 . (6.74)
K
If we approximate the functional form of Eq. (6.73) by the standard Woods–Saxon
type, then the surface diffuseness parameter a is given by a ≈ 2b/3. Equations (6.74)
and (6.64) suggest that the surface thickness of nuclei is governed by the nuclear
incompressibility and the Weizsäcker term in the asymptotic surface region, where
the density is extremely small, or by the range of nuclear force through the parameter
9t1 − (5 + 4x2 )t2 in more inside region.
6.2.6 Beyond the Hartree–Fock Calculations:

Nucleon–Vibration Coupling; ω-Mass
We remarked in Sect. 6.2.4 that the single-particle level density obtained by the
Hartree–Fock calculations is too small compared with the experimental data, and
suggested the possibility that one of the reasons of this failure is related to the
effective mass.
The effective mass hitherto discussed within the framework of the Hartree–Fock
theory arises from the antisymmetrization of nucleons. In revised theories which go
beyond the Hartree–Fock approximation, the single-particle potential has an energy
dependent correction term in addition to the potential UHF in the Hartree–Fock theory
due to, e.g., the coupling of nucleons to the surface vibration, which is one of the
nuclear collective excitations. In consequence, the energy of the single-particle state
is given by
p2
e= + UHF (r, p(r )) + UPVC (r, e) . (6.75)
2m
The lower index PVC in the third term on the right-hand side means particle–
vibration coupling. We denoted the bare nucleon mass M N by m. Also, we rep-
resented the Hartree–Fock potential by an equivalent momentum dependent local
potential in the spirit of the WKB approximation, although it is in general a non-
local potential.
By taking the derivative of both sides of Eq. (6.75) with respect to p, and using
the definition of the effective mass m1∗ ≡ 1p ddep , we obtain
m∗ mk mω
= , (6.76)
m m m

mk m ∂UHF −1
= 1+ , (6.77)
m p ∂p
mω ∂UPVC
=1− , (6.78)
m ∂e
where m k stands for k-mass and m ω for ω-mass or e-mass.
The ω-mass m ω is larger than the bare mass m, and is concentrated in the surface
region reflecting the coupling between nucleons and the surface vibrational motion
[16]. Thanks to the ω-mass, the final effective mass becomes larger than the k-
mass introduced in the Hartree–Fock approximation. Hence the agreement between
the theoretical value and the experimental data concerning the single-particle level
density is improved by taking into account the effect of particle–vibration coupling.
Incidentally, this is related to the fact that the parameter a,14 which is called the
level density parameter, takes the value A/8 MeV−1 in the region of low excitation
energy in contradiction to the expectation. The ω-mass approaches 1 with increas-
ing excitation energy. Consequently, the level density parameter a is expected to
approach the expected value A/15 MeV−1 as the excitation energy increases. It has
been confirmed that it is the case indeed from the study of, e.g., the excitation energy
dependence of the evaporation spectrum of α particles from excited nuclei [18].
6.3 Relativistic Mean-Field Theory (σ ωρ Model)
Theoretical studies which use the relativistic theory and introduce mesons to mediate
the nuclear force have also been extensively developed to determine nuclear proper-
ties such as the binding energy and the shape of nuclei within the framework of the
mean-field theory in parallel with the non-relativistic Hartree–Fock theory with the
Skyrme force. The relativistic approach has the advantage that the spin–orbit force is
14 If one considers the nucleus as a Fermi gas, then the level density ρ
L exponentially increases with
the excitation energy E and can be roughly given by ρ L (E) ∼ exp(a E)1/2 (see [17] for details).
naturally introduced.15,16 In this section, we learn the σ ωρ model, which is a widely

used relativistic mean-field theory17 [20, 21].
6.3.1 Lagrangian
In the σ ωρ model, the Lagrangian density is assumed to be given by18

L = ψ̄ iγμ ∂ μ − M N − gσ σ − gω γμ ωμ − gρ γμ τa ρ aμ ψ
1 1 1 1 a aμν 1 2 a aμ
+ ∂μ σ ∂ μ σ − U (σ ) − Ωμν Ω μν + m 2ω ωμ ωμ − Rμν R + m ρ ρμ ρ
2 4 2 4 2
1 μν 1 − τ3 μ
− Fμν F − eψ̄γμ ψA (6.79)
4 2
in terms of the field operators of nucleons ψ and mesons σ, ω, ρ.19 U (σ ) is the

potential energy term of the σ meson. We assume that it is given by
1 2 2 1 1
U (σ ) = m σ + g2 σ 3 + g3 σ 4 . (6.80)
2 σ 3 4
The second and third terms on the right-hand side of Eq. (6.80), which are often
called non-linear terms, are the self-interaction terms of the σ mesons and have been
introduced in order to reproduce the experimental data of the binding energy per
particle and Fermi momentum of the nuclear matter as well as the surface energy
and surface thickness of nuclei.20
15 Another attractive point of the relativistic treatment is that one can reproduce the saturation
property of nuclei within the framework of the mean-field theory. Although it is different from the
relativistic mean-field theory which we describe in this section, in the case of the Brueckner–Hartree–
Fock calculations using the nuclear force obtained from the experimental data of the nucleon–
nucleon scattering, the relativistic treatment can simultaneously reproduce the Fermi momentum
and the binding energy per nucleon of the nuclear matter. On the other hand, if one uses non-
relativistic theories, there exists a correlation between the Fermi momentum and the binding energy
per nucleon, called the Coester line, and one cannot simultaneously reproduce both of them in the
way to match with the mass formula irrespective of the nuclear force.
16 The relativistic approach reproduces the experimental data of the polarization phenomena in the
proton–nucleus scattering much better than the non-relativistic approach.

17 The π mesons are not introduced, because one tries to establish the theory under the mean-field
approximation as will be described later. Recently, the theory which explicitly includes the π mesons
is also being developed (see, e.g., [19]).
18 In this section, we use the natural units, where one sets = 1, c = 1.
19 μ and ν denote the four-dimensional time–space components (see Appendix A.10), while a the
component in the isospin space.

20 As Eq. (6.74) shows, the surface thickness of nuclei is intimately related to the nuclear incompress-
ibility. The incompressibility K takes a large value of about 500 MeV if the non-linear terms are not
introduced, and the surface properties of nuclei are not well reproduced. The value K ∼ 500 MeV
is too large also in comparison with about 210 MeV, which is suggested from the experimental value
of the excitation energy of the breathing mode.
6.3 Relativistic Mean-Field Theory (σ ωρ Model) 157
The field tensors are given by
Ω μν = ∂ μ ων − ∂ ν ωμ , (6.81)
μ aν ν aμ
Raμν
= ∂ ρ − ∂ ρ − gρ ε ρ ρ
abc bμ cν
, (6.82)
F = ∂ μ Aν − ∂ ν Aμ ,
μν
(6.83)
where εabc is the Levi-Civita symbol.
6.3.2 Field Equations
Following the Euler–Lagrange equation,

∂ ∂L ∂L
− =0, (6.84)
∂xμ ∂(∂qi /∂ x μ ) ∂qi
where qi is a field operator, the equations for the field operators are obtained as

μ 1 − τ3
γ −i∂μ + gω ωμ + gρ τa ρμ + e
a
Aμ + (M N + gσ σ ) ψ = 0 , (6.85)
2
dU (σ )
∂ ν ∂ν σ + = −gσ ρ S , (6.86)
ν dσ

∂ ∂ν + m 2ω ωμ = gω j μ , (6.87)
ν
∂ ∂ν + m 2ρ ρ aμ = gρ j aμ , (6.88)
ν μ
∂ ∂ν A = ejpμ . (6.89)
Equation (6.85) is the Dirac equation, Eqs. (6.86)–(6.88) are the Klein-Gordon equa-
tions and Eq. (6.89) is the d’Alembert equation. The source terms for each meson
and electromagnetic field are given by

A
ρ S (x) = ψ̄i (x)ψi (x) , (6.90)
i=1

A
j μ (x) = ψ̄i (x)γ μ ψi (x) , (6.91)
i=1

A
j aμ (x) = ψ̄i (x)γ μ τa ψi (x) , (6.92)
i=1

A
1 − τ3
jpμ (x) = ψ̄i (x)γ μ ψi (x) . (6.93)
i=1
2
6.3.3 The Mean-Field Theory
Equations (6.85)–(6.89) are the equations for the field operators, so that it is not easy
to solve them as they are. One therefore usually introduces the approximation that
replaces the operators by their expectation values, which are c-numbers,
σ → σ = σ (r) (6.94)
μ μ μ
ω → ω = ω (r). (6.95)
The resulting theory is called the mean-field theory.

Consequently, we obtain

α · (−i∇ − V(r)) + β M ∗ (r) + V (r) ψi (r) = εi ψi (r) , (6.96)

∗
M (r) = M N + gσ σ (r) = M N + S(r) , (6.97)
1 − τ3 0
V (r) = gω ω0 (r) + gρ τa ρ a0 (r) + e A (r) , (6.98)
2
1 − τ3
V(r) = gω ω(r) + gρ τa ρ a (r) + e A(r) , (6.99)
2
for the nucleon field, and
(−Δ + m 2σ )σ (r) = −gσ ρ S (r) − g2 σ 2 (r) − g3 σ 3 (r) , (6.100)

(−Δ + m 2ω )ωμ (r) = gω j μ (r) , (6.101)
(−Δ + m 2ρ )ρ aμ (r) = gρ j aμ (r) , (6.102)
− ΔAμ (r) = ej pμ (r) , (6.103)
for the meson and electromagnetic fields, as the equations for the stationary mean
fields. See Appendix A.10 for the notations and the variables that appear in the above
equations.
Equations (6.96)–(6.99) show that a scalar and a vector fields act on nucleons due
to the interaction with the meson and the electromagnetic fields, and that the effect
of the σ meson appears as a space dependent effective mass.
The relativistic mean-field theory usually assumes the time-reversal symmetry
and is applied to even–even nuclei. In this case, the total current density becomes
zero, so that the spatial components of the vector potential becomes zero (V(r) = 0)
as seen from Eqs. (6.101)–(6.103).
6.3.4 Prologue to How to Solve the Mean-Field Equations
Here, we briefly explain the practical method to solve the equations of the relativistic
mean-field theory. For spherical nuclei [22, 23], we express the wave function of a
nucleon in the ith state as

j
G i (r )

i Y jm (θ, φ)
ψi (r, s, t) = j
Fi (r )
r χi (t) , (6.104)

r
(σ · r̂)Y jm (θ, φ)
by noting that the good quantum number is not (, s = 21 , m , m s ), but (, s, j, m j =

m) due to the spin–orbit interaction, which we will describe in Sect. 6.3.5. The Y jm
are the spherical spinors representing the angular momentum states, and are defined
by
1 1

Y jm = m m s | jmYm | m s . (6.105)
m m
2 2
s
The χ (t) are the wave functions in the isospin space.

By inserting Eq. (6.104) into Eq. (6.96), we obtain the following coupled differ-
j j
ential equations for the radial wave functions G i (r ) and Fi (r ),

j ∂ κi j j
εi G i (r ) = − + Fi (r ) + [M N + S(r ) + V (r )]G i (r ) , (6.106)
∂r r

j ∂ κi j j
εi Fi (r ) = + + G i (r ) − [M N + S(r ) − V (r )]Fi (r ) , (6.107)
∂r r
where κ is defined by

− ( j + 1/2) for j = + 21
κ= (6.108)
+ ( j + 1/2) for j = − 21 .
On the other hand, the meson and the electromagnetic fields obey the radial
Laplace equations, which are given by

∂2 2
− 2 − + m 2Φ Φ = SΦ (r ) , (6.109)
∂r r
where
⎧
⎪
⎪ −gσ ρ S (r ) − g2 σ 2 (r ) − g3 σ 3 (r ) for the σ field,
⎪
⎨g ρ (r )
ω B for the ω field,
SΦ (r ) = (6.110)
⎪
⎪ gρ ρ3 (r ) for the ρ field,
⎪
⎩
eρC (r ) for the Coulomb field.
⎧ A
⎪
⎪ 4πr 2 ρ S (r ) = i=1 (|G i (r )|2 − |Fi (r )|2 ) ,
⎪
⎨4πr 2 ρ (r ) = A (|G (r )|2 + |F (r )|2 ) ,
B i i
Ni=1 Z
⎪
⎪ 4πr 2
ρ (r ) = (|G (r )| 2
+ |Fn (r )| ) −
2
p=1 (|G p (r )| + |F p (r )| ) ,
2 2
⎪
⎩
3
n=1 n
4πr 2 ρC (r ) = Zp=1 (|G p (r )|2 + |F p (r )|2 ) .
(6.111)
The m Φ is taken to be the mass of the corresponding meson if Φ is either σ or

ω or ρ mesons, and zero if Φ is photon. One then solves these coupled equations
self-consistently by an iteration method with either the expansion method using
the harmonic oscillator wave functions as the expansion basis or with the direct
integration method by discretizing the space into small lattices.
We describe actual applications in Sect. 7.6.2 by taking the study of nuclear shape
as an example after we learn pairing correlations.
6.3.5 Non-relativistic Approximation and the Spin–Orbit

Coupling
We learnt in Sect. 5.2.5 that the spin–orbit coupling plays a crucial role in determining
the magic numbers and the stability of nuclei. Here, we introduce the non-relativistic
approximation by using the Tani–Foldy–Wouthuysen transformation (TFW transfor-
mation), and obtain the information on the one-body spin–orbit coupling from the
relativistic mean-field theory.
The TFW transformation gradually removes the odd power terms of α which
admix the small and large components from the equation by repeating the unitary
transformation which makes the small components, i.e., the third and fourth com-
ponents, contained in the relativistic wave function small, and thus increases the
accuracy of the approximation in the power series of the inverse of the mass [22, 24].
By performing the TFW transformation on Eq. (6.96), we obtain the following
Hamiltonian for the spin–orbit coupling,

1 1 1 ∂V 1 ∂S
Ĥs = −β ·s (6.112)
2 M N2 r ∂r r ∂r

1 1 1 ∂
∼ (V (r ) − S(r )) · s . (6.113)
2 M N2 r ∂r
Equation (6.113) is the equation for the large components, i.e., the first and second
components, of the Dirac spinor.21, 22
21 Note that the central force is governed by V + S in contrast to the · s force (see Eq. (6.106)).
22 The case of electrons in atoms corresponds to the case when there exists only the V term. Thanks
to the S term, the sign of the spin–orbit potential for nucleons inside a nucleus becomes opposite
to that for the electrons in atoms [25].
If we express the V (r ) − S(r ) in terms of the meson fields by using Eqs. (6.97)
and (6.98), and determine each meson field by approximating the left-hand side of
Eq. (6.109) with the mass term, then we have
1 2 1 1
V (r ) − S(r ) ∼ g ρ S + 2 gω2 ρ B + 2 gρ2 τ3 ρ3 (6.114)
m 2σ σ mω mρ
1 1 1
≈ 2 gσ2 ρ B + 2 gω2 ρ B + 2 gρ2 (ρ E − ρ N E ) , (6.115)
mσ mω mρ
where ρ B is the baryon density (nucleon density), ρ E and ρ N E are the densities of
nucleons which have the equal or non-equal isospin to that of the nucleon under con-
sideration, respectively. In transforming from Eqs. (6.114) to (6.115), we remarked
Eq. (6.111) and assumed that the main components of the Dirac spinor are much
larger than the small components (|G| |F|).
Since the nucleon density decreases with r , the results shown in Eqs. (6.113)
and (6.115) match the attractive nature of the spin–orbit coupling, i.e., that vLS is
negative as we learnt phenomenologically in Chap. 5. Also, they show that the spin–
orbit coupling is strong in the surface region where the density variation is large.
Furthermore, Eq. (6.115) suggests that the spin–orbit coupling depends on the isospin
due to the spin–orbit coupling stemming from the coupling to ρ mesons, hence that
it is necessary to revise the isospin-independent spin-dependent term (W0 term)
assumed in the original Skyrme force. This effect is important, e.g., in understanding
the isotope shift of nuclear radius over a wide range across a magic number.23
The spin–orbit coupling is expected to have different effects in stable nuclei and
in unstable nuclei in the vicinity of the limit of existence through the difference of
the surface properties and the isospin dependence. The exploration of such effects
will be an interesting subject.
6.3.6 Parameter Sets
Table 6.2 shows several parameter sets which have been used for actual calculations.
The common notation NL stands for non-linear.
23 Sharma [26] succeeded in explaining the kink phenomenon which the radii of Pb isotopes show
when they cross the neutron magic number N = 126 by replacing the W0 term in Eq. (6.25) with
iW0 (1 + xw Pτ )(σ 1 + σ 2 ) · k L × δ(r1 − r2 )k R .
Table 6.2 Input parameter sets for the relativistic mean field calculations and the nuclear matter
properties obtained by them
NL1 NL2 NL-SH NL3
m N (MeV) 938.000 938.0 939.0 939.0
m σ (MeV) 492.250 504.89 526.059 508.194
m ω (MeV) 795.360 780.0 783.00 782.501
m ρ (MeV) 763.000 763.0 763.00 763.000
gσ 10.138 9.111 10.444 10.217
gω 13.285 11.493 12.945 12.868
gρ 4.976 5.507 4.383 4.474
g2 (fm−1 ) −12.172 −2.304 −6.9099 −10.431
g3 −36.265 13.783 −15.8377 −28.885
Nuclear matter properties
ρ0 (fm−3 ) 0.1542 0.146 0.146 0.148
E/A (MeV) −16.43 −17.016 −16.328 −16.299
m ∗N /M N 0.571 0.670 0.60 0.60
K (MeV) 212 399.2 354.95 271.76
6.4 Pairing Correlation
6.4.1 Overview
We learnt in Sect. 2.3.1 that the binding energy per nucleon for even–even nuclei is
systematically larger than that for odd nuclei and that for odd–odd nuclei. Also, by
using 21082 Pb as an example, we have shown in Sect. 5.8 that the level structure, i.e.,
the spin, parity and the level spacing, of the ground state and low-lying excited states
of doubly-magic ±2 nuclei can be explained by taking into account the short-range
residual interaction, which is ignored in the mean field, i.e., in the zeroth order shell
model.
In general, detailed examination of the role of residual interactions which are not
included in the mean field is necessary to understand the nuclear properties in detail.
Of the residual interactions the especially important is the pairing correlation. The
ground state and low-lying excited states of nuclei become superfluid due to the
pairing interaction. As stated in the beginning of this section, the pairing correlation
strongly affects the binding energy of nuclei as well as the low-lying level structure,
i.e., the spin, parity and the distribution of the excitation energies of low-lying states.
Furthermore, as we will learn later, it plays an important role in determining the
shapes of nuclei. Also, the pairing correlation has a strong influence on various
phenomena and quantities such as the moment of inertia of the rotational motion of
deformed nuclei and the cross section of two-nucleon transfer reactions.
6.4 Pairing Correlation 163
The pairing interaction among electrons, which makes metals superfluid, is

induced through the electron–phonon coupling [27]. In contrast, the pairing correla-
tion in nuclei originates partly from the direct attractive interaction between nucleons
and partly from the mechanism that the pair of nucleons, which occupy time-reversal
states to each other in the vicinity of Fermi surface, form a Cooper pair via surface
vibrations (see Ref. [28] for details).
In solid state physics, the origin of superfluidity depends on the system. For
example, it is known that many metals and alloys become superfluid by the 1 S pair,
while heavy electron systems and liquid 3 He become superfluid by 3 P pair [27]. The
mechanism to become superfluid depends on the system for the nucleus as well. In
this section, by having the ground and low-lying excited states of stable nuclei in
mind, we discuss the role and the theoretical treatment of the main pairing correlation
due to the pair of nucleons in the isospin-triplet and spin-singlet state, i.e., in the 1 S
state.24
6.4.2 Multipole Expansion of the Pairing Correlation,

Monopole Pairing Correlation Model and Quasi-Spin
Formalism
Let us first consider the behaviour of the residual interaction and its effect in the case
when a single-particle level with total angular momentum j is occupied by nucleons.
Here, we use the second quantization representation. The residual interaction V is
in general represented by two creation and two annihilation operators, because it is
a two-body operator. Since Hamiltonian is a tensor of rank-0, we can represent V as

V = EJ C+ (J M)C− (J M) , (6.116)
J =even M
1
C+ (J M) ≡ √ j jm M − m|J Ma †jm a †j M−m [1 + (−1) J ] , (6.117)
2 m>0
C− (J M) = C+† (J M) , (6.118)
by separately coupling two creation and two annihilation operators into a reduced
tensor.
The J = 0 term in Eq. (6.116) is called the monopole pairing, and the other terms
multipole pairing, i.e., the quadrupole pairing etc. As Table 5.3 shows, the monopole
pairing is much stronger than multipole pairing when the residual interaction is given
by the δ function. Hence one often approximates V by keeping only the monopole
24 In neutron stars, it is considered that neutrons in the outer region with low density called crust
are in the 1 S0 superfluid state, the neutrons in more inner region with high density are in the 3 P2
superfluid state, and protons are in the 1 S0 superfluid state. Also, these superfluidities are considered
to play important roles in the cooling and the phenomena called glitches of pulsars. See [29].
term. This is called the monopole pairing model. In that case, the residual interaction
is given by25
V = −G S+ S− , (6.119)
1
G = −2E 0 , (6.120)
2j + 1

S+ = (−1) j−m a †jm a †j−m , (6.121)
m>0

S− = S+† = (−) j−m a j−m a jm . (6.122)
m>0
Using the so-called quasi-spin formalism [1], one can analytically show that the
degeneracy of the energy levels of the N -body system | j N is then resolved and the
resulting levels can be classified according to the quantum number named seniority
s, which represents the number of unpaired nucleons, i.e., the number of nucleons
whose pairwise angular momentum is not coupled to zero.
Especially, in the case when N = 2, i.e., when the number of valence nucleons is
2, the seniority s is either s = 0 or s = 2, and all the energies for the J = 2, 4, 6, . . .
states corresponding to s = 2 are degenerate and only the energy of the J = 0 state
for s = 0 gets lower. This result well represents the main feature of the level structure
of 210
82 Pb which we learnt in Sect. 5.8 (Table 5.3) as an example of the doubly-magic
±2 nuclei by emphasizing the characteristic property of the pairing correlation with
the assumption that there exists only the monopole pairing correlation.
6.4.3 BCS Theory
The quasi-spin formalism can be applied not only to the case when the pairing
interaction is weak, so that the single level approximation is applicable like the
case which we described in the previous section, but to other cases as well. For
example, if there exist many close lying levels, then one will be able to apply the
quasi-spin formalism by considering those levels as approximately degenerate levels,
and by considering them as a single energy level with a large effective angular
momentum. Recently, there are attempts of the quasi-spin formalism which does
not introduce the degeneracy approximation. However, more general and standard
treatments which handle the energy distribution of plural single-particle energy levels
as it is are the Bardeen–Cooper–Schrieffer (BCS) or the Hartree–Fock–Bogoliubov
(HFB) theories. In this section, we learn the BCS theory.
25 Since j jm √
− m|00 = (−) j−m / 2 j + 1, S+ and S− are the creation and annihilation operators,
respectively, of the pair of nucleons whose total angular momentum couples to 0.
6.4.3.1 The Time-Reversed State Representation of the Monopole

Pairing Correlation Model
Before we describe the BCS theory, let us reconsider the meaning of the monopole
pairing correlation model given by Eqs. (6.119)–(6.122) from the point of view of
the correlation between the time-reversed states.
We fix the phase such that the time-reversed state is given by
|α J M = Tˆ |α J M = (−1) J −M |α J − M , (6.123)
when we represent the single-particle state as |α J M by using the magnitude of

the angular momentum J , the magnitude of its z-component M and the additional
quantum number α. The Tˆ is the time-reversal operator, and is given by
Tˆ = −iσ y K̂ (6.124)
with the operator K̂ , which implies to take complex conjugate of all c-numbers.26
Exercise 6.3 Following the time-reversal operator defined by Eq. (6.124), show that
the time-reversed state is given by Eq. (6.123) if we define the single-particle state
as 1

|njm = Rn (r ) m m s jm i Ym |m s . (6.125)
2
Hint: Use the following property of the Clebsch–Gordan coefficients:
j1 j2 m 1 m 2 |J M = (−1) j1 + j2 −J j1 j2 − m 1 − m 2 |J − M .
Note that the phase i in Eq. (6.125) has been introduced to guarantee the phase in
Eq. (6.123).
Using Eqs. (6.123), (6.119) can be rewritten as

V = −G a †jm a †jm a jm a jm . (6.126)
m>0,m >0
Equation (6.126) shows that the monopole pairing model is the model which assumes
that the residual interaction works only between the states which are time reversal to
each other. This corresponds to that the Cooper pair of electrons in metals is formed
for the pair of electrons which are in the time-reversed states (k ↑, −k ↓) to each
other. Remark that the phase which appears in the quasi-spin to define the monopole
pair and the phase which appears as the result of time-reversal operation are both
(−1) j−m and identical with each other.
26 For Eq. (6.123) and correspondingly also for Eq. (6.124), we determined the phase in the same way
as that in Ref. [30]. In Ref. [17], the phase of Eq. (6.123) is taken to be (−1) J +M . Correspondingly,
the time-reversal operator is taken to be Tˆ = iσ y K̂ .
6.4.3.2 HF+BCS Theory
One of the standard methods to theoretically explore the structure of nuclei is first
to determine single-particle levels in the mean field by using, for example, either
the non-relativistic mean-field calculations such as the Skyrme Hartree–Fock calcu-
lations or the relativistic mean-field theory, then to take into account the effects of
the residual interaction by the BCS theory. This method is often called the Hartree–
Fock+BCS theory.
Here, we proceed by having a non-relativistic theory in mind. We denote the
single-particle state determined by the Hartree–Fock equation by |k. As we discuss
in Chap. 7, most nuclei except for those whose proton or neutron numbers are either
one of the magic numbers or in their vicinity are deformed. The njm which we
used to write Eq. (6.126) are therefore not in general good quantum numbers for
the eigenstates of the Hartree–Fock equation. Instead, as we learn in Sect. 7.4, |k is
given by a superposition of the states which have njm as quantum numbers,

|k = Dαk |α . (6.127)
α={njm}
However, one can define the time-reversed state by |k̄ ≡ Tˆ |k in the deformed case
as well. In many cases, the study is performed by assuming an axially-symmetric
deformation. In that case, the magnitude of the projection of the total angular momen-
tum along the symmetry axis m = Ω is a good quantum number, and the magnitude
of the projection of the total angular momentum of the time-reversed state along the
symmetry axis is −m = −Ω.
In the BCS theory, one assumes that the total Hamiltonian is given by

Ĥ = εk (ak† ak + ak̄† ak̄ ) − G ak† ak̄† ak̄ ak , (6.128)
k>0 k,k >0
by extending Eq. (6.126). The summation is taken over the states whose component
of the angular momentum along the symmetry axis is positive.
Equation (6.128) implies that nucleons pairwise occupy the time-reversed states
to each other. Hence we assume that the total wave function is given by

|BCS = (u k + vk ak† ak̄† )|0 , (6.129)
k>0
u 2k + vk2 = 1 . (6.130)
The vk2 represents the probability that the pair of nucleons occupy the k, k̄ levels,
and Eq. (6.130) is the normalization condition. The |0 is the vacuum state. The
wave function assumed by Eq. (6.129) is in general a mixture of different number
of nucleons N . Therefore, introducing the Lagrange multiplier λ in order to fix
the particle number on average, and by performing the variational calculation to

minimize Ĥ = Ĥ − λ N̂ , we obtain

u 2k 1 εk − λ
= 1± , (6.131)
vk2 2 (εk − λ)2 + Δ2

Δ≡G u k vk . (6.132)
k>0
The λ is called either the chemical potential or the Fermi energy, and is determined
by the condition,
BCS| N̂ |BCS = 2 vk2 = N . (6.133)
k>0
Since the nucleus is a many-body system with a small number of constituents, the
fluctuation of the particle number is a serious problem. Various methods such as the
projection operator method which projects on the state with fixed number of particles
or the double constraint method which gives constraint on both the mean value and
the fluctuation of the particle number have been invented27 [31–33].
Equations (6.131) and (6.132) lead to
G Δ
Δ= . (6.134)
2 k>0 (εk − λ)2 + Δ2
Equation (6.134) is called the gap equation, and determines the gap parameter Δ.
If the pairing correlation is sufficiently strong, or if there are plenty of levels in the
vicinity of the Fermi surface, there exists a solution of finite value of Δ besides
the trivial solution Δ = 0. The former is the superfluid state. Figure 6.5 shows the
occupation probability of each single-particle level vk2 as a function of the single-
particle energy εk for the case when Δ is finite. The figure shows that the occupation
probability of each level deviates from 1 or 0, and hence the distribution near the
Fermi surface becomes blurred due to the pairing correlation.28
27 The quasi-spin method is advantageous in the sense that it has no fluctuation of the particle number,
although it is difficult to be applied to general cases because of the restriction on the degeneracy of
the energy levels.
28 Equation (6.128) corresponds to the monopole pairing correlation approximation described in the
previous section. In more advanced treatments the strength of the pairing correlation G depends on
the states and is generalized to G kk . As the result, the gap parameter becomes state dependent Δk ,
and the gap equation is generalized to [34]
1 G kk Δk
Δk = . (6.135)
2 (εk − λ)2 + Δ2k
k >0
.
Fig. 6.5 The occupation 2

probability of each
vk
1.0
single-particle state in the ~0.85
BCS theory
0.5
~0.15
ε
0.0
λ–Δ λ λ+Δ
6.4.4 The Magnitude of the Gap Parameter
One of the standard ways to determine the magnitude of the gap parameter Δ is to
estimate it phenomenologically from the binding energies of nuclei by remarking that
the binding energies systematically differ for odd, odd–odd and even–even nuclei
due to the pairing correlation (see Sect. 2.3.1). Practically, we use, for example,
1
Δn = [B(A − 2, Z ) − 3B(A − 1, Z ) + 3B(A, Z ) − B(A + 1, Z )] , (6.136)
4
1
Δp = [B(A − 2, Z − 2)−3B(A − 1, Z − 1)+3B(A, Z )−B(A + 1, Z + 1)] ,
4
(6.137)
for even–even nuclei. Figure 6.6 shows thus obtained gap parameter for neutrons Δn
as a function of the mass number.29 The Δp behaves in a similar way. As is written
in the figure, the average behaviour of Δ can be well represented by
12
Δ ∼ √ MeV (6.138)
A
for both neutrons and protons.30

For nuclei in the vicinity of doubly-magic ±2 nuclei, the magnitude of the gap
parameter Δ can be estimated also from the level spacing of low-lying states.
For example, in the case of 210 82 Pb, the level structure shown in Fig. 5.6 gives
2Δ ∼ 1.3 MeV, yielding Δ ∼ 0.7 MeV. This value well matches the empirical for-
mula (6.138).
One can also theoretically determine Δ so as to be consistent with Hartree–
Fock calculations by solving the gap equation (6.134) once the strength of the pair-
ing correlation G is known. According to Nilsson and Prior [38], G p ≈ 17/A MeV
for protons and G n ≈ 25/A MeV for neutrons. Also, Nilsson and Ragnarsson [39]
29 The masses of nuclei have been taken from [35].

30 There exist also empirical formulae which take into account the isospin dependence [36].
6
5
Δn (MeV)
−1/2
4 Δ = 12A
3
2
1
0
20 40 60 80 100 120 140 160 180 200 220 240
A
Fig. 6.6 The experimental values of Δn . After [37]
suggest that G p ≈ 26/A MeV, G n ≈ 20.5/A MeV for the nuclei in the rare-earth
region. One can find some estimates also in [28].31
We discuss in Sect. 7.6.2 the correlation between the shape of nuclei and the
nuclear superfluidity by first estimating G globally, and then by determining Δ as a
function of nuclear deformation by solving the gap equation.
6.4.5 The Coherence Length
Here, we remark the spatial extension of the Cooper pair.

The coherence length or the correlation length ξ , which gives the measure of the
spatial extension of the pairing correlation, is given by [27, 28]
vF
ξ= . (6.139)
2Δ
Exercise 6.4 Derive Eq. (6.139) based on the idea of the uncertainty relation.
Since vF is about 30% of the speed of light, Eq. (6.138) shows that the coherence
length of the pairing correlation in stable nuclei is longer than the nuclear size, i.e.,
the diameter 2R ∼ 2 × 1.2 A1/3 fm. For example, for 210 Pb, 2R ∼ 14.3 fm, while
ξ ∼ 35.7 fm.
As Eq. (2.40) shows, the Fermi momentum gets smaller when the nuclear density
becomes smaller. From Eq. (6.139) one can therefore expect that the coherence length
becomes small in the region where the nuclear density is small. As a related topic, it
is being argued recently in connection with unstable nuclei that the spatial property
of the Cooper pair changes from that in the case of normal density when the nuclear
density gets smaller than the nuclear saturation density ρ0 , and that there appears a
spatially localized two-body correlation [41].
31 One needs to determine the magnitude of G and the range of states k over which the sum is taken
on the right-hand side of Eq. (6.134) in a consistent way [40].
References
2. K. Sugimoto, M. Muraoka, Nuclear Physics, Japanese edn. (Kyouritsu, Tokyo, 1998)
3. D. Vautherin, D.M. Brink, Phys. Rev. C 5, 626 (1972)
4. J.P. Blaizot, Phys. Rep. 64, 171 (1980)
5. M. Beiner, H. Flocard, N. Van Giai, P. Quentin. Nucl. Phys. A 238, 29 (1975)
6. J. Bartel, P. Quentin, M. Brack, C. Guet, H.-B. Hakansson, Nucl. Phys. A 386, 79 (1982)
7. E. Chabanat, P. Bonche, P. Haensel, J. Meyer, R. Schaeffer, Nucl. Phys. A 627, 710 (1997);
635, 231 (1998)
8. A. Ono, H. Horiuchi, Prog. Part. Nucl. Phys. 53, 501 (2004)
9. H. Ichimura, Statistical Mechanics, Japanese edn. (Shoukabou, 1992)
10. R. Kubo, H. Ichimura, T. Usui, N. Hashitsume, Statistical Mechanics–An Advanced Course
with Problems and Solutions (Elsevier, Amsterdam, 1990)
11. S. Nishizaki, Y. Yamamoto, T. Takatsuka, Prog. Theor. Phys. 105, 607 (2001); 108, 703 (2002);
Y. Yamamoto, T. Furumoto, N. Yasutake, Th. A. Rijken. Phys. Rev. C 90, 045805 (2014)
12. T. Takatsuka, Genshikaku Kenkyu 57, Suppl. 3, 270 (2013); K. Masuda, T. Hatsuda, T. Takat-
suka, Eur. Phys. J. A 52, 65 (2016); I. Vidana. J. Phys. Conf. Ser. 668, 012031 (2016)
13. P.J. Siemens, A.S. Jensen, Elements of Nuclei: Many-Body Physics with the Strong Interaction
(Addison and Wesley, California, 1987)
14. T. Furuta, A. Ono, Phys. Rev. C 74, 014612 (2006); 79, 014608 (2009)
15. T.H.R. Skyrme, Philos. Mag. 1, 1043 (1956); Proc. Phys. Soc. (Lond.) A 70, 433 (1957); Nucl.
Phys. 9, 615 (1958)
16. J.P. Jeukenne, A. Lejeune, C. Mahaux, Phys. Rep. 25, 83 (1976); C. Mahaux, P. F. Bortignon,
R.A. Broglia, C.H. Dasso. Phys. Rep. 120, 1 (1985)
18. S. Shlomo, J.B. Natowitz, Phys. Lett. B 252, 187 (1990)
19. H. Toki, S. Sugimoto, K. Ikeda, Prog. Theor. Phys. 108, 903 (2002)
20. B.D. Serot, J.D. Walecka, Adv. Nucl. Phys. 16, 1 (1986)
21. P. Ring, Y. K. Gambhir, G. A. Lalazissis, Comp. Phys. Commun. 105, 77 (1997); Y.K. Gambhir,
P. Ring, A. Thimet, Ann. Phys. (N.Y.) 198, 132 (1990)
22. W. Greiner, Relativistic Quantum Mechanics (Springer, Berlin, 2000)
23. J. Meng, Nucl. Phys. A 635, 3 (1998)
24. J.D. Bjorken, S.D. Drell, Relativistic Quantum Mechanics (McGraw-Hill, New York, 1964)
25. H. Kurasawa, Butsuri 49, 628 (1994); J.V. Noble. Nucl. Phys. A 329, 354 (1979)
26. M.M. Sharma, G. Lalazissis, J. König, P. Ring, Phys. Rev. Lett. 74, 3744 (1995)
27. T. Tsuneto, Superconductivity and Superfluidity (Iwanami Modern Physics 17, Japanese edi-
tion) (Iwanami, Tokyo, 1993)
28. D.M. Brink, R.A. Broglia, Nuclear Superfluidity-Pairing in Finite Systems (Cambridge Uni-
versity Press, Cambridge, 2005)
29. R. Tamagaki, High density nuclear matter (Frontiers of Physics 15, Japanese edition)
(Kyouritsu, 1986); Prog. Theor. Phys. 44, 905 (1970)
30. W. Greiner, J.A. Maruhn, Nuclear Models (Springer, Berlin, 1995)
31. K. Dietrich, H.J. Mang, J.H. Pradal, Phys. Rev. 135, B22 (1964)
32. H.J. Lipkin, Ann. Phys. (N.Y.) 9, 272 (1960); 12, 452 (1961)
33. Y. Nogami, Phys. Rev. 134, B313 (1964)
34. P. Ring, P. Schuck, The Nuclear Many-Body Problem (Springer, Berlin, 1980)
35. G. Audi, A.H. Wapstra, C. Thibault, Nucl. Phys. A 729, 337 (2003)
36. D.G. Madland, J.R. Nix, Nucl. Phys. A 476, 1 (1988)
37. H. Koura, private communication
38. S.G. Nilsson, O. Prior, Mat. Fys. Medd. Dan. Vid. Selsk. 32, 16 (1961)
39. S.G. Nilsson, I. Ragnarsson, Shapes and Shells in Nuclear Structure (Cambridge University
Press, Cambridge, 1995)
40. A. Bulgac, Y. Yu, Phys. Rev. Lett. 88, 042504 (2002)
41. M. Matsuo, Phys. Rev. C 73, 044309 (2006)
Chapter 7
The Shapes of Nuclei
Abstract The nucleus behaves like a liquid drop as the mass formula and the
saturation of density imply. However, a significant difference from the classical liquid
drop, which is always spherical in order to make the energy by surface tension min-
imum, is that many nuclei are deformed except for those near the closed shells. The
shape of nuclei is one of the central research subjects of nuclear structure together
with the size of nuclei. The shape of nuclei is intimately related to the collective
excitations of nuclei, and also strongly affects nuclear reactions including heavy-
ion fusion reactions (see Balantekin and Takigawa, Rev. Mod. Phys. 70, 77 (1998);
Dasgupta et al., Annu. Rev. Nucl. Part. Sci. 48, 401 (1998); Hagino et al., Comput.
Phys. Commun. 123, 143 (1999); Esbensen, Nucl. Phys. A 352, 147 (1981); Hagino
and Takigawa, Butsuri 57, 588 (2002); Hagino and Takigawa, Prog. Theor. Phys. 128,
1001 (2012) [1–6]). In this chapter we describe several basic concepts concerning
the shape of nuclei.
7.1 The Observables Relevant to the Nuclear Shape:

Multipole Moments and the Excitation Spectrum
As pointed out in Chap. 4, the magnitude of the electric quadrupole moment provides
direct information on the nuclear shape, more precisely, on whether the distribution
of protons inside the nucleus is spherical or deformed. We learn in Sect. 8.3 that
the electromagnetic transitions give related valuable information on the shape of
the nucleus. Figures 7.1a and 7.1b show the energy levels of 166 167
68 Er and 68 Er in the
vicinity of the ground state taken from [7] (the excitation energies are in keV).
The level structure of 166 68 Er resembles that for
210
Pb and 210 Po shown in Fig. 5.6
π + + + + +
in the sense that the I = 0 , 2 , 4 , 6 , 8 levels appear in low energy region in
this order. However, it significantly differs in that the level spacing increases with
increasing angular momentum I . The level spacing well agrees with the spectrum
E I = 2 I (I + 1)/2I , I being the moment of inertia, of the rotational motion,
which is a characteristic collective motion of statically deformed nuclei. The level
structure of 167
68 Er is similar. Figures 7.1a and 7.1b therefore suggest that these nuclei
are deformed nuclei.

DOI 10.1007/978-4-431-55378-6_7
172 7 The Shapes of Nuclei
(a) (b) (c) + 1468

(0) 24+
+ 1429
1400 1415
+ 1313
+ 2+
1200 6 1216.0 0 1223
+ 1075.3
5
1000 + 956.2
4+
8+ 911.2
E [keV]
800 3+ 859.4
2 785.9
+ 617.4
600 + 2
6 545.5
9/2–– 15/2+ 432.4 442.0
400 7/2– 7/2– 412.3 430.0
+ 5/2– 13/2+ 293.7 346.6
4 265.0 5/2– 3/2– 264.9 281.6
200 1/2 11/2+ 178.0 207.8
+
2
+
80.6 9/2++ 79.3
0 0 0 7/2 0 0+ 0
π 166 167 112
J 68Er E [keV] 68 Er 48Cd
Fig. 7.1 The level structure of 166 167 112

68 Er (a), 68 Er (b) and 48 Cd (c)
On the other hand, Fig. 7.1c shows the energy-level diagram for 112 Cd in the
vicinity of the ground state. It is noticed that the first excited state is I π = 2+ state, and
that there appear almost degenerate second excited group of states with I π = 02+ , 22+ ,
and 41+ at the excitation energy of about twice that of the first excited state. This
implies that 112 Cd is a spherical nucleus and can be easily excited to the vibrational
excitation of the quadrupole type. The 02+ , 22+ , and 41+ levels can be considered to be
the two phonon states.1
In this way, the level structure of nuclei also provides important information on
the shape of nuclei.
Figure 7.2 shows the change of the energy-level diagrams for Sm isotopes in the
vicinity of the ground state as the neutron number increases. The figure suggests
+
6
+
4
2000 +
+
6 +
4
6 +
+
2+ + 4+ +
+
4+
0 2 2 +
10
1500 + 3+ + 3+
2+ 4 + 2+
4+ 0+ 2+ 4+ 4+
E [keV]
6+ +
4+ 0 3+ 0
2 8+ + 2+
0 + 2 0+ +
1000 2 4 0+
+ + 8
+ 4+ 2+
2 0 6+
2
+ 0 +
500 6
+ +
2 4 +
+
4
2+ +
+ + + + 2+
0 0 0 0 0 0 0
π144 146 148 150 152 154
J 62 Sm Sm Sm Sm Sm Sm
Fig. 7.2 The isotopic variation of the low-lying level structure of Sm isotopes
1 Recently, it is argued that the 02+ state has a different property from that of the two-phonon state.
7.1 The Observables Relevant to the Nuclear Shape: Multipole Moments … 173
100 3.0 ≤ E4/E2

2.7 ≤ E4/E2 < 3.0
2.5 ≤ E4/E2 < 2.7
80
E4/E2 < 2.5
Proton number
60
40
20
0
0 20 40 60 80 100 120 140 160
Neutron number
Fig. 7.3 The region of nuclei whose ground state is deformed
that the Sm isotopes change their shape from the spherical shape to the deformed
shape as the neutron number increases from the magic number N = 82 (the phase
transition of the nuclear shape, i.e., the shape transition).
Figure 7.3 shows the region in the nuclear chart where nuclei are deformed in their
ground state. Those nuclei have been selected as the nuclei, which have either the
ground-state rotational band or whose E 4 /E 2 ratio is large, where E 2 is the excitation
energy of the first 2+ state and E 4 that of the first 4+ state except for some shape
coexisting nuclei.2 They are sorted according to the value of E 4 /E 2 .
Typical deformed nuclei are located in the regions of rare earth such as Er isotopes
and actinides such as U isotopes. In addition, 8 Be, 12 C, 20 Ne, and 24 Mg are also
deformed in their ground state. Even if the ground state is spherical, some nuclei have
deformed states in the region of relatively low excitation energy. Also, there exist
nuclei which have states with different deformations in the same energy region. They
are called the shape coexistence. 16 O and some of the Ge isotopes are representative
examples of these nuclei.3 Ge isotopes are also known to make shape transition in
the ground state from the spherical shape to a deformed shape as the neutron number
increases.
2 The excitation energy of each level has been taken either from the Table of Isotopes [7] or from
http://www.nndc.bnl.gov/chart/.
3 Light Hg isotopes are known to be the nuclei with shape coexistence [8]. They have a rotational
band of prolate shape in the region of a low excitation energy in addition to the ground-state rotational
band of oblate shape. The excitation energy of the 4+ member of the prolate band becomes lower
than that of the 4+ member of the oblate ground-state band for the isotopes whose mass number
is 186 or smaller. Hence we used the excitation energy of the second 4+ state for E 4 for the
182,184,186 Hg isotopes in making Fig. 7.3. We have omitted 180 Hg from the figure, since the level
structure is presently unclear.

7.2 Deformation Parameters
Various deformation parameters are conventionally used to represent nuclear defor-

mation. Here, we list some of them and discuss their properties and connections
among them. We introduced the deformation parameters αλ by Eq. (2.66) in order to
describe fission. The parameter δ which is defined by
3 R32 − R⊥2
δ≡ (7.1)
2 R32 + 2R⊥2
is also often used when the deformation is axially symmetric by assigning the sym-
metry axis to the third axis, which we sometimes call the z-axis in the following. The
R3 and R⊥ represent the equivalent radii in the direction of the symmetry axis, and
in the perpendicular direction, respectively.
For general deformations, one expands the radius as
⎡ ⎤
∞
λ
R = R(θ, ϕ) = R0 ⎣1 + α00 + ∗
αλμ Yλμ (θ, ϕ)⎦ (7.2)
λ=2 μ=−λ
by extending Eq. (2.66). The α00 is not an independent parameter in the case of
incompressible deformations, but is given by
∞ λ
1
α00 = − |αλμ |2 (7.3)
4π λ=2 μ=−λ
by using the other deformation parameters due to the condition of the volume con-
servation.
One can derive the following properties for the deformation parameters:
∗
αλ−μ = (−1)μ αλμ , (7.4)
P̂αλμ P̂ −1 = (−1)λ αλμ , (7.5)
(αλμ )ncs = R̂(ω)αλμ R̂ −1
(ω) = Rμ(λ) μ (ω)(αλμ )ocs , (7.6)
by using the condition that the nuclear radius R is real and invariant under the
parity and the rotational transformations and also from the properties of the spherical
harmonics. P̂ and R̂ are the parity inversion and the rotation operators, respectively.
When we take the point of view to rotate the system, i.e., the nucleus, instead of the
axes, R̂ is given by
ˆ ˆ ˆ
R̂(ω) = e−iφ Jz e−iθ Jy e−iψ Jz , (7.7)
7.2 Deformation Parameters 175
Fig. 7.4 The quadrupole

vibration
in terms of the Euler angles ω = (φ, θ, ψ). The Rμ(λ) μ (ω) is the matrix which repre-
sents the rotation, and is defined by [9, 10]
Rμ(λ) μ (ω) ≡ λμ |R̂(ω)|λμ = Dμλ μ∗ (ω) . (7.8)
The indices ncs and ocs in Eq. (7.6) represent the components after and before
the rotational transformation, respectively. Equation (7.6) shows that {αλμ } forms a
tensor of rank-λ and that αλμ is its μ component (see Sect. A.6).
Exercise 7.1 Show Eqs. (7.4)–(7.6).
We have shown in Fig. 2.21 the shape of deformation for λ = 2, 3 and 4 in con-
nection with fission.4 The surface vibrations of spherical nuclei can be described by
using αλμ as dynamical variables (geometrical collective model).5 Those vibrations
correspond to the oscillations of the deviation of the shape from the spherical shape
shown in Fig. 2.21 as functions of time. The vibrations for λ = 2 and 3 are called the
quadrupole and octupole vibrations, respectively. Figure. 7.4 conceptually shows the
quadrupole vibration as an example.
For statically deformed nuclei, it is more convenient to use the Euler angles, which
show the direction of principal axes, instead of the αλμ as a part of the variables in
order to describe collective motions. As an example, let us consider the case of
the quadrupole deformation. There are originally 2(2λ + 1) variables, since αλμ are
complex numbers. Among them 2λ + 1 variables are independent, since there exist
4 We assumed the prolate type, whose shape resembles that of the ball of the American football, for
the quadrupole deformation λ = 2. Note that the change of the shape depends on the sign of α40
when the hexadecapole deformation α40 is added. It becomes the barrel type which has expanded
peaks and belly compared to the shape in the case when there exists only the quadrupole deformation
if α40 > 0, while the peanuts shape when α40 < 0.
5 There exist alternative ways to describe nuclear energy level structure and excitations. One of the
powerful theoretical frameworks is the interacting boson model (IBM) [11] which is an algebraic
method using the group theory.
2λ + 1 conditions because of Eq. (7.4). Hence if λ = 2, there are two independent

variables besides three Euler angles. On the other hand, since {αλμ } is a tensor of
rank λ, the variables α2μ in the space-fixed coordinate system are transformed to new
variables a2ν in the nucleus fixed coordinate system, i.e., the body-fixed coordinate
system, where the coordinate axes are taken along the principal axes of the nucleus,
by a rotation of the coordinate system:

α2μ = a2ν Dμν
2
(ω) . (7.9)
ν
By taking the Euler angles in the direction of the principal axes, we have
a21 = a2−1 = 0, a22 = a2−2 . (7.10)
The Euler angles φ, θ, ψ and the real variables a22 , a20 become new variables.
Conventionally, however, the parameters β and γ which are defined by
1
a20 = β cos γ , a22 = √ β sin γ (7.11)
2
are usually used for a22 and a20 . The nuclear radius in the direction of θb , ϕb measured
in the body-fixed system is given by

5 √
R(θb , ϕb ) = R0 1 + β cos γ (3 cos θb − 1) + 3 sin γ sin θb cos 2ϕb

2 2
.
16π
(7.12)
β is the parameter which gives the magnitude of the quadrupole deformation, and is
related to α2μ as
|α2μ |2 = a20
2
+ 2a22
2
= β2 . (7.13)
μ
γ is the parameter for discussing the axial symmetry. Equation (7.12) shows that
1. γ = 0◦ , 120◦ , 240◦ represent the prolate type of axially symmetric deformation
where the 3, 1 and 2 axes are the symmetry axes, respectively.
2. γ = 180◦ , 300◦ , 60◦ represent the oblate type of axially symmetric deformation
where the 3, 1 and 2 axes are the symmetry axes, respectively.
3. The γ which is not a multiple of 60◦ represents a triaxial deformation.
4. One can represent all the shapes by using only the region β ≥ 0, 0◦ ≤ γ ≤ 60◦
in the case of quadrupole deformation. The other regions of β, γ can be reached
by suitable transformations.
Figure. 7.5 shows the connection between the deformation parameters β, γ and the
nuclear shape.
The following relations hold among different deformation parameters,
7.2 Deformation Parameters 177
Fig. 7.5 The relation between the deformation parameters β, γ and the nuclear shape

5 2
α2 = α20 ∼ δ , α20 = β ∼ 1.06δ , δ ∼ 0.95α20 = 0.95β , (7.14)
4π 3
when the deformation is small.
7.3 The Deformed Shell Model
Let us study the distribution of energy levels for nucleons inside a deformed nucleus.
One can then understand the origin of nuclear deformation.
As Eq. (5.4) shows, in the approximation to set the range of the nuclear force to zero
the single-particle potential for nucleons is proportional to the density of nucleons.
This implies that the mean field is also deformed if the nucleus is deformed. This is
called the self-consistency condition.
Suppose an axially symmetric deformation and assume that the single-particle
field is given by the deformed harmonic oscillator field as
MN 2 2
V = ω⊥ (x + y 2 ) + ωz2 z 2 . (7.15)
2
Both oscillator parameters ω⊥ and ωz must be functions of the magnitude of deforma-
tion δ because of the self-consistency condition. In order to determine their functional
forms, we remark that the shape of the equipotential surfaces of the potential rep-
resented by Eq. (7.15) is given by the ratio of the distances in the directions of the
three axes ω1⊥ : ω1⊥ : ω1z . Therefore, if we denote the radius along the α (α = x, y, z)
axis by aα , the mean radius by R0 and the common oscillator parameter by ω0 , then
it is required that

R0 2
ωz (δ) = ω0 (δ) ∼ 1 − δ ω0 (δ) , (7.16)
az 3

R0 1
ω⊥ (δ) = ω0 (δ) ∼ 1 + δ ω0 (δ) , (7.17)
a⊥ 3
in order for the mean field to be consistent with the density distribution6 [12]. If we
denote the mean oscillator parameter by ω̄,7 then we obtain

1
ω0 (δ) ∼ ω̄ 1 + δ 2 , (7.18)
9
from the incompressibility condition ω⊥2

ωz = ω̄3 .
The potential given by Eq. (7.15) can be expressed as
1
V = M N ω02 (δ)r 2 − M N ω02 (δ)r 2 βY20 (θ, φ) (7.19)
2
by using the deformation parameter.
Since it is the problem of the three-dimensional harmonic oscillator, the energy
levels can be easily obtained by using either the Cartesian or cylindrical coordinates,
and are given by
ε(n x , n y , n z ) = ε(n z , n ρ , k ) (7.20)

1
= ω⊥ (n x + n y + 1) + ωz n z + (7.21)
2

1
= ω⊥ (2n ρ + |k | + 1) + ωz n z + (7.22)
2

3 1
∼ ω0 (δ) N + + δ(N − 3n z ) , (7.23)
2 3
where N = n x + n y + n z = n z + 2n ρ + |k | and k is the quantum number which

represents the magnitude of the projection component of the orbital angular momen-
tum along the symmetry axis.
As remarked concerning Eqs. (7.16) and (7.17), strictly speaking, the δ in
Eq. (7.23) differs from the δ in Eq. (7.1), and while the latter is defined by
3 Rz2 − R⊥2
3 ω⊥2
(δ) − ωz2 (δ)
δ≡ = , (7.24)
2 Rz2 + 2R⊥2 2 ω⊥ (δ) + 2ωz2 (δ)
2
6 The transformation to the last term in Eqs. (7.16) and (7.17) is correct only to the first order of the
deformation δ. Strictly speaking, it is therefore different from the δ defined by Eq. (7.1).
7 The rough estimate of ω̄ is given by Eq. (5.52).
7.3 The Deformed Shell Model 179
the former is defined by

ω⊥ (δ) − ωz (δ)
δ≡3 . (7.25)
ωz (δ) + 2ω⊥ (δ)
Hence some literatures, e.g., [13], distinguish it by denoting with a different notation
such as δosc .
Figure 7.6 shows how the three levels (six levels if we include the spin), which are
degenerate in the spherical shape, split with the deformation by taking the case of N =
1 levels as the example. Among them the n z = 0 level is two-fold degenerate (four-
fold degenerate if one includes the spin). For positive δ, i.e., for prolate deformation
which is stretched in the direction of the symmetry axis, the state which has the
finite quantum number in that direction, i.e., the N = 1, n z = 1 state, in other words,
the state with a smaller |k | (the k = 0 state in the present case), has lower energy,
while for negative δ, i.e., for oblate deformation which is shrunk in the direction of
the symmetry axis, the state which has the finite quantum number in the direction
perpendicular to the symmetry axis, i.e., the N = 1, n z = 0 state, in other words, the
state with a large |k | value (the k = ±1 state in the present case) has lower energy.
This can be easily expected from the uncertainty principle, which suggests that the
kinetic energy can be smaller if the movable space gets larger.
Note also that the change of the energy due to deformation is larger on the prolate
side if we compare the lower level on the prolate and oblate sides. This can be
understood from Eq. (7.23) which shows that the variation of the energy is given by
2δ/3 for the lower n z = 1 level on the prolate side, while it is given by δ/3 for the
n ⊥ = 1(n z = 0) level which is the lower level on the oblate side.
Figure 7.6 suggests that the deformation of nucleus occurs in close connection with
the shell structure and the Pauli exclusion principle. As an example, let us consider
8
4 Be from the point of view of the shell model. The first two protons (and neutrons)
occupy the lowest 1s state (N = 0 state). Figure 7.6 shows that it is energetically more
favorable to deform the nucleus to the prolate shape and to fill the remaining two
protons (and neutrons) in the n z = 1 level. For this reason, 8 Be becomes a deformed
nucleus of prolate type.8 As the number of nucleons increases, it becomes necessary
to fill higher levels with nucleons, and the sign of deformation changes. Roughly
Fig. 7.6 The deformation ε(δ)

dependence of the
single-particle energy levels
N=1, nz=0 N=1, nz=1
δ
-0.3 -0.2 -0.1 0 0.1 0.2 0.3
8 This is a kind of Jahn–Teller effect. There exists an alternative view for 8 Be in terms of the cluster
model or the α-particle model, where 8 Be is thought to be a dumb-bell consisting of two alpha
speaking, it can be expected that the nuclei which belong to the beginning of a major
shell have prolate deformation, while those towards the end of the shell oblate. For
example, in the case of N = 1 shell, i.e., the 1 p shell, 8 Be, which belongs to the
first half, has a prolate deformation, while 12 C, which belongs to the last half, has an
oblate shape. However, in reality most of the deformed nuclei are prolate, and those
which are oblate in their ground state are limited to 12 C and small number of nuclei
such as some of the Hg isotopes, which we mentioned in connection with Fig. 7.3
and will discuss again in Sect. 7.6.2.
7.4 Nucleon Energy Levels in a Deformed One-Body Field:

Nilsson Levels
The spin–orbit coupling is indispensable in order to discuss real nuclei. Also, the
radial dependence of the harmonic oscillator type is unrealistic. Hence we assume
that the Hamiltonian for nucleons is given by
2 MN 2 2 MN 2 2
ĥ = − Δ+ ω⊥ (x + y 2 ) + ω z + C · s + D2 (7.26)
2M N 2 2 z
by extending the deformed harmonic oscillator model introduced in the previous

section. The 2 term has been introduced in order to effectively improve the harmonic
oscillator model so as to yield the level structure for the realistic radial dependence
of the Woods–Saxon type, and is designed in such a way that the states with larger
orbital angular momentum feel stronger attraction (see Fig. 5.1). The single-particle
levels obtained by Eq. (7.26) are called Nilsson levels or Nilsson diagram or Nilsson
scheme, since this model was proposed by Nilsson [14] for the first time.
The Nilsson levels and the associated wave functions can be obtained by diago-
nalizing the Hamiltonian given by Eq. (7.26) by using a suitable complete set as the
expansion basis. For example, using the wave functions of the deformed shell model
considered in the previous section |α = |N n z k Ω as the basis, we expand the wave
function as
|φi=N n z k Ω = Ciα |α , (7.27)
α
and determine the expansion coefficients Ciα and the corresponding single-particle
energy εi by diagonalizing ĥ. Ω = k + m s is the magnitude of the projection com-
ponent of the total angular momentum along the symmetry axis, where m s is the
magnitude of the projection component of the spin along the symmetry axis.
Figure 7.7 shows an example of the Nilsson levels. The followings are some
remarks concerning the Nilsson levels.
particles. However, the overlap between the wave functions of the shell model and of the cluster
model is fairly large.
7.4 Nucleon Energy Levels in a Deformed One-Body Field: Nilsson Levels 181
Fig. 7.7 Neutron energy levels for prolate deformation for 82 < N < 126, taken from [15]
1. The 2 j + 1 levels, which are degenerate for the spherical shape, split into (2 j +
1)/2 levels by deformation.
2. Each split energy level is twofold degenerate together with the time reversed
state.
3. As the right-hand side of Eq. (7.27) shows, different N , n z , k values are mixed in
a single energy level. Nevertheless, the set of quantum numbers {N , n z , k , Ω},
which is suitable in the region of large deformation and hence is sometimes called
the asymptotic quantum numbers, is conventionally used as the label for each
Nilsson level. However, the parity, which is determined by N is a good quantum
number, and Ω is also a good quantum number as long as the deformation is
axially symmetric. Note that Ω is not the component of the angular momentum
along a space-fixed axis, but the projection of the angular momentum along an
intrinsic, i.e., body-fixed, axis.
4. Roughly speaking, for prolate deformation (δ > 0), the states with small Ω gain
negative energy by deformation and the amount of the gain is larger for smaller
Ω. The shift of energy by deformation gradually changes with Ω, and turns to be
positive for states with large values of Ω. The trend of the effect of deformation
is opposite for oblate deformation. However, the effect of deformation is more
complex for oblate deformation.
7.5 The Spin and Parity of the Ground State of Deformed

Odd Nuclei
One can speculate the spin and parity of the ground state of deformed odd nuclei by
referring to the Nilsson diagram.
Consider 16768 Er as an example. From the analysis of the inelastic scattering of α
particles, i.e., (α, α ) scattering, it is known that the quadrupole deformation of 166 68 Er
is β2 = 0.276 [16]. If the quadrupole moment of 167 68 Er is assumed to be nearly equal,
then the spin of the ground state of 167 68 Er is predicted to be 7/2 from the Ω value of
the 99th energy level for neutrons in Fig. 7.7. Also, the parity is expected to be plus,
since this level results from the split of the i 13/2 level. These predictions agree with
the experimental results shown in Fig. 7.1b.
7.6 Theoretical Prediction of Nuclear Shape
There are various theoretical methods to confirm nuclear shapes determined or pre-
dicted by experimental data or to predict unknown nuclear shapes. Here, we explain
two standard methods which determine the nuclear shape based on the calculation
of binding energies.
7.6.1 The Strutinsky Method: Macroscopic–Microscopic

Method
One of the methods is the method which reconciles the liquid-drop model and the
shell model. It is called either the macroscopic–microscopic method or the Strutinsky
shell correction method after the name of the inventor [17]. This method gives also the
binding energy of nuclei and the potential surface for fission with high accuracy [18].
Leaving the detailed explanation of the method to other books [19], here we
mention the basic concept of this method.
As we see in Fig. 2.6, the binding energy consists of the part E LDM which smoothly
depends on the mass number and can be well explained by the liquid-drop model,
7.6 Theoretical Prediction of Nuclear Shape 183
and the part which deviates from it and oscillates, i.e., detailed structure;
E tot = E LDM + E osc . (7.28)
The oscillating part arises because the distribution of single-particle energy lev-
els deviates from the average level density and is non-uniform reflecting the shell
structure.
In the Strutinsky method, the part which depends smoothly on the mass number is
estimated macroscopically based on the liquid-drop model, while the oscillating part
E osc as the difference between the energy microscopically calculated by the shell
model and the mean energy Ẽ sh calculated by introducing the average level density
E osc = E sh − Ẽ sh . (7.29)
If we denote the accurate or shell model level density by g(ε), the average level
density by g̃(ε), the Fermi energy by λ, and the Fermi energy when the average level
density is used by λ̃, then

A
g(ε) = δ(ε − εi ) , (7.30)
i=1

A λ
E sh = εi = εg(ε)dε , (7.31)
i=1 −∞
λ̃
Ẽ sh = ε g̃(ε)dε , (7.32)
−∞
λ λ̃
A= g(ε)dε = g̃(ε)dε , (7.33)
−∞ −∞
where we denote the mass number by A.

The folding method,
∞
1 ε − ε
g̃(ε) = g(ε ) f dε , (7.34)
γ −∞ γ
where f is a Gaussian, whose width γ is of the order of ω0 , is a candidate of the

method to introduce the average level density. However, one must be sure that the
result does not change when the averaging procedure is repeated [19].
Incidentally, Eq. (7.29) is revised as
E osc = E sh − Ẽ sh + E pair − Ẽ pair (7.35)
if the pairing energy is taken into account in the BCS method.

A feature of this method is that the parameter dependence is smaller than the
microscopic method, which we will discuss in the next section, because the parame-
ters are phenomenologically determined based on, e.g., the mass formula. Hence, it
has a high predictive power. As stated at the beginning of this section, it therefore
provides a powerful way to calculate the binding energy and the shape of the ground
state as well as the energy surface for fission. The same idea has been applied also
to the theoretical study of the structure of metal clusters.
7.6.2 Constrained Hartree–Fock Calculations
The other method is the constrained mean-field calculations with the linear or the
quadratic constraints. If one is interested in the quadrupole deformation, the method
of linear constraint performs variational calculations for the effective Hamiltonian
given by
Ĥ = Ĥ − λ Q Q̂ , (7.36)

where Q̂ is the operator of the quadrupole moment Q̂ = 16π 5
Q̂ 20 . λ Q is the
Lagrange multiplier and is determined such that the expectation value of the quadru-
pole moment becomes equal to the given value Q:

9
Q̂ = Q ∼ A R02 β. (7.37)
5π
It is therefore a function of the magnitude of the quadrupole moment Q or the

deformation parameter β, that is, λ Q = λ Q (β).9 In the method of quadratic constraint,
one performs variational calculations to make
1
E ≡ Ĥ + C( Q̂ − μ)2 (7.38)
2
minimum. This method corresponds to taking λ Q = C(μ − Q̂) in the linear con-
straint method. If one takes a large enough value for C, E gives the minimum energy
for the state with the deformation μ when μ = Q̂.
The single-particle states and the corresponding energies εi (β) are determined
by performing mean-field calculations under these constraints. In the Hartree–
Fock+BCS theory, one needs to further determine the u and v factors. There are
practically two ways to that end. The one is a simplified method, which treats Δ as
a deformation independent parameter. The other is to determine Δ as a function of
9 Inrecent years, the study of unstable nuclei which are far from the β-stability line such as those
in the vicinity of drip lines is extensively going on. One of the main concerns there is to elucidate
the difference of the deformations of the proton and neutron distributions inside a nucleus. In such
studies, one performs calculations by imposing separate constraints on protons and neutrons.
7.6 Theoretical Prediction of Nuclear Shape 185
β by solving the gap equation (6.134) for each value of the deformation parameter
β, and further to determine the corresponding u and v factors. As the Nilsson levels
show, the distribution of single-particle energy levels strongly depends on β, so that
the latter is, of course, much superior. As we discuss later, it is sometimes crucially
important to follow the latter procedure in order to determine the shape of nuclei. In
that case, one determines the strength of the pairing correlation G in advance by some
way, and assumes that it does not depend on the nuclear deformation. This method
is often called the constant G method. By comparison, the simplified method which
treats Δ as a deformation independent constant is called the constant Δ method.
Figure 7.8 shows the energy surface for 184 Hg obtained by the relativistic mean-
field (RMF) calculations using the method of quadratic constraint as a function of
the deformation parameter β [20]. The figure shows that the ground state of 184 Hg
is oblate and suggests that there appears prolate states at low excitation energies.
Actually, these results well agree with the experimental data, i.e., that for light Hg
isotopes there appears a rotational band with prolate shape at low excitation ener-
gies above the ground-state rotational band with oblate shape as we mentioned in
connection with Fig. 7.3. The RMF calculations were performed by assuming the
NL1 force [21], and the BCS calculations in the constant G method. In the latter, the
magnitude of G has been determined in such a way that the average value of the gap
parameter
Δ̄n = 4.8/N 1/3 , (7.39)
which has been obtained from the systematic data analyses for many nuclei [22], can
be reproduced for spherical nuclei, instead of using the average value mentioned in
Sect. 6.4.4 as it is for the magnitude of G. Using thus obtained G n and the energies
εk (β) of the single-particle levels obtained by the constrained RMF, the gap equation
Fig. 7.8 The energy surface

as a function of β. Taken
from [20]
Fig. 7.9 The deformation

dependence of the gap
parameter. Taken from [20]
2 k
max
1
= (7.40)
Gn k=0
(εk (β) − λ)2 + Δ2n (β)
has been solved for each value of β and the value of the gap parameter Δn (β)
has been determined. The G p and Δp (β) for protons have also been determined by
using Δ̄p = 4.8/Z 1/3 and by the same procedure. Figure 7.9 shows thus obtained gap
parameters as functions of the deformation parameter [20]. The figure clearly shows
that the value of the gap parameter, hence the superfluidity of a nucleus, changes
drastically with the deformation of the nucleus. This can be easily expected from
the facts that the value of the gap parameter strongly depends on the level density
in the vicinity of the Fermi surface as Eq. (7.40) shows, and that the distribution of
the nucleon energy levels strongly reflects the degree of deformation as the Nilsson
diagram studied in Sect. 7.4 shows. As mentioned in Sect. 7.3, 184 Hg is one of the
exceptional nuclei which are oblate in the ground state. If we use the constant Δ
method, 184 Hg is predicted to be prolate. This suggests that the pairing correlation
plays a decisive role in determining the shape of nuclei, and that it is crucially
important to correctly take into account the deformation dependence of the pairing
correlation in order to discuss the shape of nuclei. It will be worth mentioning that
this calculation well reproduces the experimental data of the isotope variation of the
charge radii of Hg isotopes over a wide range of the mass number.
7.7 Column: Superdeformed States 187
7.7 Column: Superdeformed States

The amount of deformation of rare-earth nuclei and actinides in their ground
states is about δ ∼ 0.3. By comparison, the deformation of light deformed nuclei
such as 8 Be is large.
Today, many states with large deformation have been found in the excitation
energy region in the medium-mass and heavy nuclei as well. They are called
superdeformed states and superdeformed rotational bands built on them, and
have been extensively studied together with high spin states [23].
The fission isomers mentioned in Sect. 2.3.3 are typical examples of superde-
formed states. Figure 7.10 shows the distribution of superdeformed bands and
fission isomers on the nuclear chart.
We learnt in Sect. 7.3 that the deformation of nucleus is intimately related to
the shell structure of single-particle levels. In this connection, Fig. 7.11 shows
the distribution of single-particle energy levels in the deformed shell model. The
figure shows that there appears a distinct shell structure, i.e., the degeneracy of
levels, similar to that for δ = 0, when the ratio between the long and short axes
Rz : R⊥ becomes, e.g., 2:1. This happens because in that case the ratio of the
magnitudes of the energy quanta in the direction parallel to the symmetry axis
and in the perpendicular direction ωz : ω⊥ becomes 1:2 due to the consistency
condition between the potential for nucleons and the nucleon density. In this
case, the deformation δ (SD) of the superdeformed state becomes δosc (SD)
= 0.6 in
terms of δosc defined by Eq. (7.25).
Fig. 7.10 The distribution of superdeformed states, taken from [24]

Fig. 7.11 The energy levels in the deformed harmonic oscillator model. Taken from [25]
References
1. A.B. Balantekin, N. Takigawa, Rev. Mod. Phys. 70, 77 (1998)

2. M. Dasgupta, D.J. Hinde, N. Rowley, A.M. Stefanini, Annu. Rev. Nucl. Part. Sci. 48, 401
(1998)
3. K. Hagino, N. Rowley, A.T. Kruppa, Comput. Phys. Commun. 123, 143 (1999)
4. H. Esbensen, Nucl. Phys. A 352, 147 (1981)
5. K. Hagino, N. Takigawa, Butsuri 57, 588 (2002)
6. K. Hagino, N. Takigawa, Prog. Theor. Phys. 128, 1001 (2012)
8. K. Heyde, J.L. Wood, Rev. Mod. Phys. 83, 1467 (2011)
9. A. Messiah, Quantum Mechanics, vol. II (Elsevier, Amsterdam, 1962)
10. A. Bohr, B.R. Mottelson, Nuclear Structure, vol. I (Benjamin, New York, 1969)
11. F. Iachello, A. Arima, The Interacting Boson Model (Cambridge University Press, Cambridge,
1987)
References 189
13. A. Bohr, B.R. Mottelson, Nuclear Structure, vol. II (Benjamin, New York, 1975)
14. S.G. Nilsson, Mat. Fys. Medd. Dan. Vid. Selsk. 29(16), (1955)
15. C. Gustafson, I.L. Lam, B. Nilsson, S.G. Nilsson, Ark. Fys. 36, 613 (1967)
16. D.L. Hendrie et al., Phys. Lett. B 26, 127 (1968)
17. V.M. Strutinsky, Yad. Fiz. 3, 614 (1966); Sov. J. Nucl. Phys. 3, 449 (1966); Nucl. Phys. A 95,
420 (1967); Nucl. Phys. A 122, 1 (1968)
18. P. Möller, J.R. Nix, W.D. Myers, W.J. Swiatecki, At. Data Nucl. Data Tables 59, 185 (1995)
19. P. Ring, P. Schuck, The Nuclear Many-Body problem (Springer, Berlin, 1980)
20. S. Yoshida, N. Takigawa, Phys. Rev. C 55, 1255 (1997)
21. P.-G. Reinhard, M. Rufa, J. Maruhn, W. Greiner, J. Friedrich, Z. Phys. A 323, 13 (1986)
22. P. Möller, J.R. Nix, Nucl. Phys. A 536, 20 (1992)
23. P.J. Nolan, P.J. Twin, Annu. Rev. Nucl. Part. Sci. 38, 533 (1988); R.V.F. Janssens, T.L. Khoo,
Annu. Rev. Nucl. Part. Sci. 41, 321 (1991)
24. B. Singh, R.B. Firestone, S.Y. Frank Chu, Table of Superdeformed Nuclear Bands and Fission
Isomers, WWW Edition, Updated June 1997; An updated version of the figure can be found
in B. Singh, R. Zywina, R.B. Firestone. Nucl. Data Sheets 97, 241 (2002)
25. I. Ragnarsson, S.G. Nilsson, R.K. Sheline, Phys. Rep. 45, 1 (1987)
Chapter 8
Nuclear Decay and Radioactivity
Abstract Many nuclei with finite lifetimes are produced by nuclear reactions and/or
nuclear decays, though the number of stable nuclei is limited to less than 300 if we call
those nuclei whose lifetime is longer than or nearly equal to the age of the universe,
i.e., about 13.8 × 109 years, as stable nuclei. Also, all the excited states of nuclei
eventually decay into other nuclei via α or β decay or fission, or make transitions
to lower energy levels of the same nucleus by emitting γ rays. The decay called
cluster decay or heavy particle decay or cluster radioactivity, where a nucleus decays
by emitting a nucleus heavier than α particle such as 14 C, has also been actively
studied since 1980s (see Rose and Jones, Nature, 307: 245 (1984), [1]). In 2000s,
the study of one-proton and two-proton radioactivities of proton-rich nuclei has also
become active (see Blank and Borge, Prog Part Nucl Phys, 60: 403 (2008); Blank
and Płoszajczak Rep Prog Phys, 71: 046301 (2008), [2]). In this chapter we learn the
α-decay and electromagnetic transitions among the decay of nuclei and radioactivity.
We also briefly discuss recent developments concerning fission.
8.1 Alpha Decay
Unstable nuclei with large atomic number decay primarily by emitting an α particle
(α-decay).

+
210
84 Po 0g.s. → 206
82 Pb + α (τ1/2 = 138.38 days) , (8.1)

+
93 Np 5/2g.s. → 91 Pa + α (τ1/2 = 2.14 × 10 years) ,
237 233 6
(8.2)
are the examples. Each of the total numbers of protons and neutrons is preserved
in the decay. The nucleus before and after the decay is called the parent and the
daughter nucleus, respectively. 210 Po and 206 Pb are the parent and daughter nuclei,
respectively, in the decay of 210 Po. Both of the states of the parent nucleus prior to
the decay and of the daughter nucleus after the decay are not necessarily the ground
states, and are also not necessarily unique. The decay of 210 Po mentioned above
represents the decay of the ground state of 210 Po. The half-life of this state is 138.38

DOI 10.1007/978-4-431-55378-6_8
192 8 Nuclear Decay and Radioactivity
+
days, and the decay occurs to the ground state of 206 Pb (0g.s. ) with probability of
more than 99% and the remaining part decays to the first excited state (21+ state) of
206
Pb. In the case of the α-decay of the ground state of 237 Np, the branching ratios
to many different states of the daughter nucleus are nearly equal and the final states
diverge.
The amount of energy given by
Qα = B(A − 4, Z − 2) + B(4, 2) − B(A, Z) (8.3)
is released in the decay from the ground state of the parent nucleus to the ground
state of the daughter nucleus. Qα is called the Q-value of the α-decay between the
ground states.
The α-decay takes place only when Qα is positive. Using Eq. (8.3), the condition
can be approximately represented as BA < B(4,2)
4
− A dA ( A ). Figure 2.6 suggests that
d B
−3
( ) is about −7.5×10 MeV. On the other hand, the binding energy of α particle
d B
dA A
is 28.3 MeV. Hence the condition for the α-decay to occur becomes
B
< 7.08 + 7.5 × 10−3 A MeV . (8.4)
A
By adding the right-hand side of Eq. (8.4) to Fig. 2.6, we find that the Q-value of
the α-decay gets positive, and hence the nucleus becomes in principle unstable with
respect to α-decay, when the mass number exceeds about 150.
8.1.1 Decay Width
There exist various theoretical methods to estimate the α-decay width. Here, we
introduce some of them.1
8.1.1.1 The Method to Obtain the Resonance Formula from the Pole of
the S-Matrix
One of the simple methods is the potential model. In this method, one considers the
α-decay to be the process that a metastable state in a potential field VDα , which we
write simply as V in the following, of the daughter nucleus for the α particle decays
with time by a quantum tunneling. The metastable state corresponds to the state of the
parent nucleus. V consists of the nuclear and Coulomb potentials2 and will behave as
shown in Fig. 8.1. Figure 8.2 is its simplification. The abscissa r denotes the distance
1 As a related topic, the study of resonance states by using the so-called complex scaling method is
actively going on [3]. It is related also to the study of unstable nuclei.
2 Here, we consider the decay of the s-wave.
8.1 Alpha Decay 193
Fig. 8.1 The interaction V(r)

potential between α particle
and nucleus
VCB
E
r3 r2 rB r1 r
Fig. 8.2 The interaction V(r)

potential between α particle
and nucleus. The Gamow
model VCB
E
r3 r2 r1 r
–V0 (rB)
between the centers of the α particle and the daughter nucleus. For simplicity, we
assumed that the daughter nucleus is spherical and correspondingly that V is a central
potential.
The information on the metastable state can be obtained by studying the pole of the
S-matrix for the scattering if one takes the view that the α-decay is the inverse process
of the resonance scattering between the α particle and the daughter nucleus by the
potential V . Although numerical calculations are needed in general to that end, one
can derive formulae to determine the energy E, the width Γ of the metastable state
and hence the lifetime of the α-decay τ = /Γ with intuitive physical understanding
if one uses a semi-classical theory. In that connection, Fig. 8.1 shows the position of
the metastable state, the Q-value of the α-decay (E = Qα ) and the three classical
turning points r1 , r2 , r3 where the potential energy coincides with the Q-value of
the α-decay: V (ri ) = E(i = 1, 2, 3). Leaving the details of the derivation of the
formulae to [4–6], here we quote the results:
r2
1 1 1
k(r)dr = n + π + argN ∼ n + π (n : integer), (8.5)
r3 2 2 2
r1
Γ 1 1
= exp −2 κ(r)dr = t (WKB) , (8.6)
T r2 T

k(r) = 2μ(E − V (r))/ , κ(r) = 2μ(V (r) − E)/ , (8.7)
r2 r2
dr dr
T =2 √ =2 . (8.8)
r3 2μ(E − V (r))/μ r3 v(r)
μ is the reduced mass for the relative motion between the α particle and the daughter
nucleus. Equation (8.5) is the quantization condition to determine the position E
of the metastable state and E has to agree with the Q value of the α-decay. The
N is a factor related to the quantum tunneling. The resonance position reflects the
properties of the potential barrier through the argument of the factor N. However,
one can set argN to be 0 provided the α-decay takes place at an energy well below
the potential barrier. The exponential factor of the second term of Eq. (8.6) is nothing
but the tunneling probability t (WKB) estimated in the WKB approximation. On the
other hand, as the definition of T , Eq. (8.8), suggests, the factor 1/T in front of the
exponential function is the attempt frequency per unit time with which the α particle
tries to decay.
8.1.1.2 Gamow Factor
The potential V (r) is often assumed to be given by

−V0 for r < rB
V (r) = ZD Zα e2 (8.9)
r
for r ≥ rB ,
as shown in Fig. 8.2 for the discussion of α-decay. ZD and Zα are the atomic numbers
of the daughter nucleus and the α particle, respectively. In the following, we call
the model which assumes Eq. (8.9) the Gamow model. Since the condition for the
applicability of the WKB approximation is not satisfied in this case, one cannot
use Eq. (8.6) for the estimate of the decay width as we learn later (see Eqs. (8.13)–
(8.17)). However, it ris useful for various discussions to evaluate the exponential factor
t (WKB) = exp(−2 r21 κ(r)dr) for this case by denoting r2 and r1 as R and re , where
re is the external classical turning point of the tunneling region. With the substitution
r = re cos2 θ the integral can be easily evaluated to give

r1
(WKB) R R R
tGM ≡ exp −2 κ(r)dr = exp −4η cos−1 − 1− ,
r2 re re re
(8.10)
where η is called the Sommerfeld parameter, and is defined by η ≡ ZD Zα e2 /v,

where v is the speed in the asymptotic region.
Suppose that the Q-value is much smaller than the height of the potential barrier
VCB = ZD Zα e2 /R. Since re R in this case, the exponential factor becomes
√
(WKB)
tGM ∼ FG e4 2ηkR
, (8.11)
FG ≡ e−2πη , (8.12)
8.1 Alpha Decay 195
where k is the wave number in the asymptotic region. The FG is called the Gamow
factor, and plays the principal role in governing the probability of quantum
√ tunneling
when the decay energy is small. The contribution of the second factor e4 2ηkR is much
smaller than that of the Gamow factor. However, it cannot be ignored or set equal to
1 for the α-decay of heavy nuclei in order to keep the quantitative accuracy. In fact,
it is larger than the pre-exponential factor which will be defined later.
8.1.1.3 Direct Method Based on the Gamow Model
In the case of the Gamow model, the wave functions are given by the spherical
Bessel functions of the first kind in the potential region and by the Coulomb wave
functions in the external region. Hence one can derive the formulae for the decay
width by using them and the asymptotic form of the Coulomb wave function G 0 [7].
Leaving the derivation to Appendix A.8, here we quote the result for the case when
E VCB . Letting T = 2R/v , where v = K/μ is the speed of the α particle inside
the potential well, i.e., at r < rB , we have
Γ 1 (D)
= tGM , (8.13)
T
(D) k 1
tGM ∼4 , (8.14)
K G 20 (η, kR)
1/2 √
1/2
1 2η −2πη 4 2ηkR 2η (WKB)
∼ e e ∼ tGM , (8.15)
G 0 (η, kR)
2 kR kR
(D) k re 1/2 (WKB) (D) (WKB)
tGM =4 tGM = AGM tGM , (8.16)
K R
K ≡ 2μ(E + V0 )/, k = 2μE/ . (8.17)
(D)
tGM is the probability of the tunneling effect given by Eq. (A.127). Note that it differs
from the value given by Eq. (8.11), which is obtained by simply applying the WKB
(D)
approximation, by the factor AGM .
8.1.1.4 Time-Dependent Perturbation Theory Based on the Gamow

Model: Two-Potential Approach
The two-potential approach [8] is also often used for the calculation of the α-decay
width. In this method, one introduces the potential V1 (r), which has an infinitely thick
potential barrier, i.e., V1 (r) = −V0 for r < R and V1 (r) = VCB for r ≥ R, and the
2
potential V2 (r), which has only the Coulomb potential, i.e., V2 (r) = VC (r) = ZD Zrα e .
The Gamow potential can be decomposed as

V1 (r) for r < R
V (r) = (8.18)
V2 (r) for r ≥ R ,
by using these potentials. The α-decay width can be obtained by calculating the
transition probability from a bound state φn (r) with energy En in the potential V1 (r)
to a continuum state φE (r) with energy E close to En in the potential V2 (r). Since the
perturbation Hamiltonian is V (r) − V1 (r), the transition probability per unit time is
given by
2
Γ 2π
= φE (r)[V (r) − V1 (r)]φn (r)dr .
∗
(8.19)

One finally obtains3

re
Γ k1 2 k1 2 −2πη 4√2ηkR k1 2 (WKB)
= D exp −2 κ(r)dr = D e e ∼ D tGM ,
μ R μ μ
(8.20)

k1 2 2 2 1/2 E + V0
D = (VCB − E)1/2 , (8.21)
μ R μ VCB + V0
1/2
2 K
D≡ ; k1 = 2μ(VCB − E)/ . (8.22)
R k12 + K 2
According to [8], the pre-exponential factor ( kμ1 )D2 represents the probability that
the α particle appears at the nuclear surface per unit time.
8.1.1.5 Comparison of Three Methods
The formulae for the decay width in the three methods discussed in Sects. 8.1.1.1,
8.1.1.3 and 8.1.1.4 contain the Gamow factor in common. However, the pre-
exponential factor depends on the way of derivation of the formulae in each method.
The formulae introduced in Sect. 8.1.1.1 are superior in the sense that they can
be applied to smooth realistic potentials, since they have been derived based on the
WKB approximation. However, one has to be careful when they are applied to the
Gamow model as they are, and when the tunneling probability t (WKB) in Eq. (8.6) is
replaced by the Gamow factor e−2πη or by the formula multiplied with the correction
factor Eq. (8.11). Since the discussions in both Sect. 8.1.1.3 and 8.1.1.4 are based on
the Gamow model, their formulae must yield the correct result within the Gamow
model, and should in principle give the same decay width.
As an example, let us consider the α-decay of 210 Po. As we discuss in Sect. A.8,
there exists an ambiguity for the potential describing the α-decay, and there are
3 See [8] for the details of the derivation and physical interpretation. See also [9].
8.1 Alpha Decay 197
various potentials ranging from a shallow to a deep potentials.4 Here, we assume

that the wave function of the relative motion between the daughter nucleus and the
α particle has 11 nodes by considering the Pauli exclusion principle. In this case, the
potential becomes a deep potential, and V0 = 108 MeV, R = 7.92 fm as we show in
Sect. A.8. VCB is about 30 MeV, while Q-value is 5.4 MeV. Hence it holds that
V0 VCB Eα . (8.24)
In this case, the pre-exponential factor kμ1 D2 in Eq. (8.20) approximately becomes
4 ( ) . The result of Sect. 8.1.1.3 given by Eqs. (8.13), (8.15), and (8.20) in
1 kα re 1/2
T K R
Sect. 8.1.1.4 thus become practically equivalent for the model which assumes a deep
potential for the potential between the daughter nucleus and the α particle.5
4 Let us take a view of cluster model that a nucleus consists of two constituent clusters A and B. In
the case of α-decay, the nucleus corresponds to the parent nucleus, and the clusters A and B to the
daughter nucleus and the emitted α particle, respectively. Accordingly, we first represent the total
wave function as the product of the intrinsic wave functions of the clusters A and B and the wave
function for their relative motion. We then antisymmetrize with respect to the whole nucleons. This
is called the resonating group method (RGM). If we represent the intrinsic wave functions of two
clusters in the harmonic oscillator model with the same oscillator parameter, then there exist in
general special states called either the Pauli forbidden states or the redundant states for the wave
function of the relative motion because of the constraint that the total wave function has to be
antisymmetric with respect to the exchange of any two nucleons. They are the states, for which
the total wave function vanishes as the result of antisymmetrization. Consequently, there appears
ambiguity in the wave function of the relative motion concerning the admixture of the Pauli forbidden
states, and it turns out that there exist several equivalent wave functions with different number of
nodes, which we denote by nr = 0, 1, . . . by excluding the node at the origin. Correspondingly,
there appears an ambiguity for the potential describing the relative motion ranging from shallow to
deep potentials. For example, if we represent 8 Be as the product of two α particles and their relative
motion, then the 0s and 1s states are the Pauli forbidden states for the relative motion with angular
momentum 0. By counting the quantum number in the harmonic oscillator shell model, one can
write the condition for the Pauli forbidden states as
(SM) (SM) (SM)
2nr + < NC − NA − NB , (8.23)
(SM) (SM) (SM)
when the angular momentum of the relative motion is . Here, NC , NA and NB are the total
quantum number of the whole nucleus (C) and of the constituent clusters (A) and (B) when they
are represented by the harmonic oscillator shell model. For example, in the case of α–α scattering
(SM) (SM) (SM)
or in the case of the α-decay of 8 Be, NA = NB = 0, NC = 4. Equation (8.23) is called
the Wildermuth condition. In the case of the α-decay of 210 Po, the right-hand side of Eq. (8.23) is
(5 + 6) × 2, so that the states with nr = 11 and/or larger nr are the allowed states. Incidentally,
the existence of the Pauli forbidden states is also the origin of the repulsive core in the potential
between complex nuclei. In the α–α scattering, the 0s and 1s states are the redundant states for
= 0 and the 0d state is the redundant state for = 2. Hence there exists repulsive core in the
potential for these partial waves. On the other hand, there exists no redundant state for = 4 and
higher partial waves, so that there is no repulsive core in the potential for those high partial waves.
5 Various methods to calculate the decay width have been numerically compared for the proton
decay [10].
8.1.1.6 Preformation Factor—R-Matrix Theory
All the three methods so far discussed presume that the α particle with the same
structure as that of the α particle in nature exists inside the parent nucleus in advance
of the α-decay. Let us call such a point of view the potential model. On the other
hand, one can make a detailed description of the state of the parent nucleus using the
wave functions of nucleons based on, e.g., the shell model or the cluster model. In
such calculations the α particle is not presumed to exist in advance inside the nucleus
with the same structure as that in nature. For example, a simple shell model is based
on the thought that each nucleon inside a nucleus moves independently from each
other in the mean field except for the constraint due to the Pauli exclusion principle.
In order to evaluate the α-decay width based on the study of nuclear structure by a
many-body theory such as the shell model, it is necessary to estimate the formation
probability of α particle inside the nucleus prior to the decay. The formalism based
on the R-matrix theory6 is one of the representatives of such approaches which takes
into account the preformation probability, where the decay width is given by the
following formulae:
Γ = 2krc γL2 (rc )vL (rc ) = 2PL (rc )γL2 (rc ) , (8.25)
1 1
vL (rc ) ≡ 2 ∼ 2 , (8.26)
FL (rc ) + G 2L (rc ) G L (rc )
2 1/2

γL (r) = YL (r) , (8.27)
2μr
YL (r)
= φα φD YLM (r̂)|ψP int,r̂ . (8.28)
r
The FL and G L are the Coulomb wave functions. rc is chosen such that the effect
of short-range force between the daughter nucleus and the emitted particle, i.e., the
α particle in the case of alpha decay, is absent in the region r > rc , and is called
the channel radius. vL (rc ) is called the penetration factor, where the index L shows
that the decay takes place for the partial wave L.7,8 The ψP is the wave function
of the parent nucleus, and φD , φα are the intrinsic wave functions of the daughter
nucleus and the α particle, respectively. They are determined based on appropriate
6 The R-matrix theory is a theory to describe resonances in nuclear reactions. It describes the reaction
by dividing the space into the internal region and the external region, i.e., the asymptotic region
where the nuclear force between the emitted nucleus and the remaining nucleus can be ignored. See
[11–15].
7 As the asymptotic formula G + iF ∼ exp[i(ρ − η ln(2ρ) − Lπ/2 + σ )], where ρ = kr and σ
L L L L
being the Coulomb phase shift, indicates, the wave function which becomes the outgoing wave of
unit magnitude is given by ψL (r) = FL (r) + iG L (r) in the region r ≥ rc for the Gamow model.
Hence the vL (rc ) ≡ F 2 (r )+G
1
2 (r ) gives the magnitude of the penetration probability from rc to the
L c L c
outside of the potential barrier.
8 As it was shown by Eq. (8.14) in Sect. 8.1.1.3, v and the transmission coefficient t in the Gamow
0
(D)
model are related as tGM ∼ 4 Kk G 2 (kr
1
)
∼ 4 k
K v 0 (kr c ).
0 c
8.1 Alpha Decay 199
many-body theories such as the shell model, cluster model or mean-field theory. The
lower index int, r̂ of Eq. (8.28) means to integrate over the intrinsic coordinates and
the angular coordinates of the relative motion. The YL (r)/r is therefore the radial
wave function of the relative motion between the daughter nucleus and the α particle.
γL2 (r) has the dimension of energy and called the reduced width. It gives a measure of
formation probability of the α particle at the surface (r = rc ) of the parent nucleus.
It is often expressed as
32
γL2 (rc ) ≡ θ 2 (rc )γW2 (rc ) , γW2 (rc ) = . (8.29)
2μrc2
γW2 (rc ) is the value of γL2 (rc ) in the case when the initial state of the α-decay in the
parent nucleus has such a structure that an α particle is clusterized and the radial
wave function of the relative motion YL (r) = YL (r)/r distributes uniformly in the
region 0 ≤ r ≤ rc . It is called the Wigner limit [16].9 θ 2 is the reduced width in units
of the Wigner limit.
Exercise 8.1 Noting that G 0 (ρ = 0) = 1/C0 (η) and C02 (η) = 2π η(e2πη − 1)−1 ,
confirm that 1/G 20 is proportional to the Gamow factor, which is the principal factor
to govern the tunneling probability, if η 1.
Let us investigate the relation between the formulae for the α-decay width
discussed in Sect. 8.1.1.1, 8.1.1.3, and 8.1.1.4, and Eq. (8.25). If we denote the
momentum for Y (r) in the region r < rc by pin , and estimate it as pin rc ∼
based on the uncertainty relation, we have
γW2 3 2 1 pin 3
= ∼ . (8.30)
2 μ rc 2 T
Hence Eq. (8.25) can be expressed as
Γ 1
= t fP (8.31)
T
by writing the penetration probability as t, which is tentatively identified with PL (rc ).
fP ∝ θ 2 represents the probability that the α particle exists inside the parent nucleus
in advance of the decay. Let us therefore refer to it as the preformation factor.
The value of θ 2 provides important information on the extent of clusterization in
light nuclei [18]. A large value of the reduced width for the α-decay corresponds to a
large preformation factor of the α particle, and suggests that the picture of the alpha
cluster structure holds. The preformation factor is ignored in some analyses of the
α-decay. In that case, the effect of the preformation factor is effectively incorporated
in the parameters such as the depth V0 and the range parameter R of the potential.10
Reference [20] is an example of the calculation of θ 2 based on the shell model.
9 See [17] for the connection to the Teichmann–Wigner sum rule.

10 Here, we quote [19] as references for the analysis of α-decay.
Incidentally, for heavy particle decay such as the 14 C emission, the possibility that
the species of the emitted particle and its structure change on the way of decay due
to the channel coupling effects has also been discussed.
8.1.2 The Geiger–Nuttal Rule
The α-decay width strongly depends on the Q-value of the decay and the atomic
number of the parent or daughter nuclei through the tunneling probability. Their
principal effects are expressed by the Gamow factor. Let us therefore express the
decay width, more precisely, the decay rate per unit time as
Γ
= Ae−2πη (8.32)

by separating out the Gamow factor. Taking the logarithm of both sides and using
η ≡ ZD Zα e2 /v, we obtain
Γ
log10 = log A − CZD Eα−1/2 , (8.33)
10
e2 μc2
C ≡ 2π Zα log10 e ≈ 1.72 . (8.34)
c 2
The factor A does not depend so much on the decaying system as long as we
consider nuclei in a region of similar mass number. Hence Eq. (8.33) suggests that
the experimental data distribute almost along a straight line if we plot them by taking
−1/2
log10 Γ / as the ordinate and ZD Eα as the abscissa [12, 21]. This property is
called the Geiger–Nuttal rule.
Incidentally, detailed information can be obtained from the analysis of the factor
A, since it contains the information on the system such as the potential and the
preformation factor.
8.2 Fission
In Chap. 2 we studied fission somewhat in detail with respect to nuclear instability.

Here, we briefly discuss recent developments of the study of fission. As shown
in Fig. 2.19 in a simplified fashion, fission is, in a general term, one of the decay
processes of a metastable state, and occurs by quantum tunneling if the temperature
of the system is 0 or low, while occurs thermally, i.e., by thermal hopping, if the
temperature becomes sufficiently high. Figure 8.3 schematically shows this situation.
The critical temperature TC which divides the quantum decay and the thermal decay
is given by
8.2 Fission 201
Fig. 8.3 The temperature or

energy dependence of the
decay mechanism and the
width of a metastable state
Ω
TC ∼ , (8.35)
2π
if we denote the curvature of the potential barrier by Ω (Here, we consider in terms

of the corresponding energy) [22, 23]. If we take a simple view of a one-dimensional
process and express the curvature of the potential, in which the metastable state is
made, around the potential minimum by Ω0 , then the decay width, more precisely,
the number of decay per unit time is given by
Ω
0 1
2π 1+exp( Ω
2π
Vf )
for T < TC ; Quantum decay
Γ ∼ Ω0 −Vf /T
(8.36)
2π
e for T ≥ TC ; Thermal decay.
In each equation, the first factor Ω

2π
0
represents the frequency of the metastable state to
attempt decay and the second factor the success probability. Vf , which corresponds
to Ef in Sect. 2.3.4, is the height of the fission barrier. We used the formula of the
uniform approximation for the tunneling probability and approximated the potential
barrier by a quadratic function. e−Vf /T is a characteristic factor for thermal decay,
and is called the Arrhenius factor.
The spontaneous fission is an example of the decay by quantum tunneling. A
characteristic is that it is a quantum tunneling in a multi-dimensional space. Many
experimental data such as those of the mass and energy distributions of the fission
fragments have been accumulated. On the other hand, detailed theoretical calcula-
tions of the potential energy in the multi-dimensional space, where the mass ratio
and the deformation parameters of the fission fragments are variables, have been per-
formed by using, e.g., the macroscopic–microscopic method, and the search for the
decay path and the calculation of the fission lifetime with, e.g., the multi-dimensional
escape path method [24] have been developed [25, 26].11
11 Not necessarily limited to the spontaneous fission, it has been experimentally reported that there
are cases where there exist multiple decay paths with different barrier heights for the fission of the
same nucleus. They are called bimodal fission. The bimodal fission appears as the coexistence of the
symmetric and asymmetric fission for, e.g., 226 Ra(3 He,df). On the other hand, both are symmetric
fission, but differ in the width of the mass distribution and the distribution of the kinetic energy of
fission fragments in the case of the spontaneous fission of some of the actinides with large mass
number such as 258 Fm.
The induced fission corresponds to the thermal decay if the excitation energy is
high as it is the case for the fission induced by heavy-ion collisions. The formula
for the thermal decay given by the lower line of Eq. (8.36) is known as the formula
of the transition state theory by Bohr and Wheeler [27]. The Kramers formula [28]
which takes into account the effect of friction,
γ
e−Vf /T
Vf
γ when < T
(the case of weak friction)
Γ ∼ Ω0
T 2Ω
γ
Vf
(8.37)
2π
Ke−Vf /T when 2Ω
> T
Vf
(the case of strong friction) ,
where

1 1 γ
K= Ω2 + γ2 − (8.38)
Ω 4 2
is also often used as the basic formula for the decay width of the thermal fission
in calculating the competition between fission and the emission of particles such as
neutrons by viewing the fission as a kind of Brownian motion [29, 30] in the external
field and hence by formulating it in terms of the Fokker–Planck equation in the phase
space. γ is the friction or viscosity coefficient in the Fokker–Planck equation. K is
the factor which takes into account the effect of friction on the decay rate and is
called the Kramers factor.12
The studies of the reaction mechanism of heavy-ion reactions and of various
physical quantities governing the collision have significantly advanced since 1970s.
In this procedure, various concepts have been introduced. They are, e.g., the complete
fusion reaction, where the colliding nuclei completely fuse and decay after a long
time, and the quasi-fission, which resembles the complete fusion concerning the
mass distribution of the reaction products, but differs in the angular distribution and
the deep inelastic collision (DIC), which is supposed to have a short reaction time
because of the feature that the masses of the reaction products stay near the masses of
the projectile and the target nuclei even though a large energy loss from the relative
motion is involved. Many experimental data have been accumulated in these studies
concerning the mass and kinetic energy distributions of the reaction products such
as the fission fragments, and the number of emitted neutrons associated with fission.
The theoretical approach which follows the time evolution of the system based on
the Langevin equation by simulating the reaction as a random walk (or stochastic
process, or drunker’s walk or Wiener process) in a multi-dimensional space has been
developed as a powerful method to analyze these data by paying attention to the
dissipation and associated fluctuation properties [35] (see also [36]).
12 There appears in addition a multiplicative correction factor which takes into account the quantum
fluctuation around the classical path if we relate the decay width to the imaginary part of the
Helmholtz free energy [31] and evaluate the partition function by using the path integral method
[32–34].
8.3 Electromagnetic Transitions 203
8.3 Electromagnetic Transitions
A nucleus in the ground state makes transitions to excited states by absorbing photon,
while a nucleus in an excited state makes transition to the ground state or to lower
excited states by emitting photon. One can extract much information on the structure
of the nucleus such as the excitation energy, the spin and parity of each energy
level, the nuclear shape and the collectivity of each excitation motion through these
transitions.
8.3.1 Multipole Transition, Reduced Transition Probability
The probability of the electromagnetic transition can be given to a good accuracy by

the Fermi golden rule in the first order perturbation theory,
2π
Tfi = |f |Ĥint |i|2 g(Ef ) . (8.39)

Ĥint is the interaction between the nucleus and the electromagnetic field, and is given
by Eq. (A.173) as will be described in Appendix A.9. If we denote the initial and the
final states of the nucleus by |ψi and |ψf , respectively, and express the state of the
radiation field in the occupation number representation (see Appendix A.9.3), then
the change of the state can be schematically written as

|f = |ψf | · · · nλ − 1 · · · for absorption,
|i = |ψi | · · · nλ · · · → (8.40)
|f = |ψf | · · · nλ + 1 · · · for emission.
Also, it holds that kc = |Ef − Ei | because of the energy conservation if we denote
the wave length of the photon by λ = 2π/k, and the energy of the initial and final
states of the nucleus by Ei and Ef , respectively.
g(Ef ) is the number of final states of the radiation field per unit energy. If we
consider the magnetic radiation as an example, we have
1 π
0 = jI (kn R) ∼ sin kn R − I (8.41)
kn R 2
from the boundary condition given by Eq. (A.167). Here, we considered R → ∞.

Hence it follows that
π
kn R = I + nπ, n = 0, ±1, ±2, · · · . (8.42)
2
The density of the final state of the radiation field g is thus given by
1 1 1 R
g(k) = = = π = . (8.43)
ΔE(k) (Δk)c Rc cπ
Using the explicit expressions of the vector fields, Eqs. (A.164) and (A.165), the
expression of the normalization constant Eq. (A.169), and the formula for the level
density Eq. (8.43), we can write the transition probability per unit time by emitting
one photon of the (λ, k, I, M) type as
2
8π k 1
Tfi (E, kIM) = f j · ∇ × jI (kr)YIIM dr i , (8.44)
ck
2

8π k 1
Tfi (M, kIM) = f j · j (kr)Y dr i . (8.45)
c
I IIM
Let us write the Tfi , in general, as

2I+1 2
8π(I + 1) Eγ
Tfi (λ, kIM) = f |Mˆ(λ, kIM)|i (8.46)
I[(2I + 1)!!]2 c
by using the multipole transition operator Mˆ(λ, kIM).

In nuclear physics, the long wave length approximation usually well applies.
Namely, it well holds that
kR 1 (8.47)
if we denote the nuclear radius by R. Equation (8.47) is equivalent to13
c c −1/3
Eγ = kc = A ≈ 197A−1/3 MeV . (8.48)
R r0
In the long wave length approximation, using

(kr)I 1 (kr)2
jI (kr) ∼ 1− + ··· , (8.49)
(2I + 1)!! 2 2I + 3
we can show that the multipole transition operators are nothing but the electric
multipole and the magnetic multipole operators introduced in Sect. 4.2 as

Mˆ(E, kIM) ≈ ρr I YIM dr = Q̂IM , (8.50)

1
Mˆ(M, kIM) ≈ (r × j) · ∇(r I YIM )dr = M̂IM . (8.51)
c(I + 1)
13 Particle decay usually occurs with a higher probability when a high energy γ ray is emitted.
Exercise 8.2 Show that the wave length of photons of 1 MeV is about 400π fm.
One usually does not distinguish the difference of the orientation of the angular
momentum of the nucleus except for the experiments related to polarization. By
taking average with respect to the initial value Mi of the z component of the angular
momentum of the nucleus, and sum over the value Mf in the final state, the total
electric or magnetic transition probability with multipolarity I is given by
1
Tfi (λ, I) = Tfi (λ, kIM) , (8.52)
2Ii + 1 M ,M ,M
i f
where Ii is the angular momentum of the initial state of the nucleus. Performing the
sum in Eq. (8.52) by using the Wigner–Eckart theorem and the properties of the 3j
symbol or the Clebsch–Gordan coefficients, we obtain
2I+1
8π(I + 1) Eγ
Tfi (λ, I) = B(λI; Ii → If ) . (8.53)
I[(2I + 1)!!]2 c
Here, B(λ) is called the reduced transition probability, and is given by
1
B(EI; Ii → If ) = |f Q̂I i|2 (8.54)
2Ii + 1
and
1
B(MI; Ii → If ) = |f M̂I i|2 (8.55)
2Ii + 1
for the electric and magnetic transitions, respectively.
8.3.2 General Consideration of the Selection Rule and the

Magnitude
If one denotes the angular momentum and parity of the initial and final states by
Ii , Πi and If , Πf , then the selection rules,
|Ii − If | ≤ I ≤ Ii + If , Πi ΠλI Πf = 1 , (8.56)
hold due to the structure of the transition matrices If Mf |Q̂IM |Ii Mi , If Mf |M̂IM |Ii Mi .
ΠλI is the parity of the multipole radiation, and is given by
ΠEI = (−1)I , ΠMI = (−1)I+1 , (8.57)
because the interaction with the radiation field is given by Ĥint ∝ A · j.

Table 8.1 Summary of the electromagnetic transition probability. The unit of each quantity is:
B(Eλ): e2 fm2λ ; B(Mλ): μ2N fm2λ−2 ; Eγ : MeV; T : s−1 . The “yes” and “no” in the column of “Parity
change” mean that the transition can take place only when the parities of the initial and final states,
Πi and Πf , are different or the same, respectively. The γ ray transition is absolutely forbidden if
Ii = If = 0, since the intrinsic spin of photon is 1
Type Angular momentum Parity change Transition probability T (s−1 )
selection rule
E1 |Ii − If | ≤ 1 ≤ Ii + If yes (Πi = Πf ) T (E1) = 1.59 × 1015 Eγ3 B(E1)
E2 |Ii − If | ≤ 2 ≤ Ii + If no (Πi = Πf ) T (E2) = 1.22 × 109 Eγ5 B(E2)
E3 |Ii − If | ≤ 3 ≤ Ii + If yes (Πi = Πf ) T (E3) = 5.67 × 102 Eγ7 B(E3)
M1 |Ii − If | ≤ 1 ≤ Ii + If no (Πi = Πf ) T (M1) = 1.76 × 1013 Eγ3 B(M1)
M2 |Ii − If | ≤ 2 ≤ Ii + If yes (Πi = Πf ) T (M2) = 1.35 × 107 Eγ5 B(M2)
M3 |Ii − If | ≤ 3 ≤ Ii + If no (Πi = Πf ) T (M3) = 6.28 × 100 Eγ7 B(M3)
The transition probability Tfi is proportional to Eγ2I+1 as Eq. (8.53) shows. Also,
the radiation of the γ ray is more strongly hindered with increasing multipolarity I.
Table 8.1 summarizes the properties of the electromagnetic transition.
8.3.3 Single-Particle Model Estimate: Weisskopf Units and

Experimental Values
The electromagnetic transitions give information on the lifetime of the excited states
and the strength of the photo-absorption as the inverse process. They provide infor-
mation also on the nuclear structure such as the nuclear shape, and also on the motion
of nucleons inside a nucleus, especially on collective motions. In this connection, it
is convenient to have an estimate of the value for the case when the single-particle
model holds as a standard value to be compared.
As an example, let us consider the electric transition probability of an odd nucleus.
The reduced transition probability is given by
2
1
B(Eλ; Ii → If ) =
λ
ψf (If ) ρr Yλ dr ψi (Ii ) (8.58)
2Ii + 1
A 2
1
λ

= ψf (If ) ek rk Yλ (θk , ϕk ) ψi (Ii ) . (8.59)
2Ii + 1
k=1
In the single-particle model, one considers that the electromagnetic transition occurs
by the transition of the valence nucleon from the point of view of the shell model.
If we denote the initial and final states by the quantum numbers of the shell model
(nj)i,f , then we have
1 2
Bsp (Eλ; Ii → If ) = (nj)f eeff r λ Yλ (nj)i , (8.60)
2Ii + 1
which becomes

2λ + 1 1 1 2
Bsp (Eλ; Ii → If ) = e2eff j λ 0 j
4π 2 2
i f
2 1 + (−1)i +f +λ
× (nj)f |r λ |(nj)i (8.61)
2
after some calculations. The eeff , named the effective charge, effectively takes into
account the effect of the nucleons other than the valence nucleon due to, e.g., the
core polarization. Let us consider especially the case when ji = I + 1/2, jf = 1/2,
and approximate the Clebsch–Gordan coefficient by the asymptotic value:

1 1 1 2(I + 1) 1
ji λ 0 j f = (−1) √
I
∼ (−1)I √ , (8.62)
2 2 2I + 1 2I + 1 2I + 1
and furthermore estimate f |r I |i as

∞
R0
r I r 2 dr 3
f |r |i =
I
Rf (r)r Ri (r)r dr = 0 R0
I 2
= RI , (8.63)
0 r 2 dr I +3 0
0
where Rf (r) and Ri (r) are the radial wave functions of the final and initial states, by
assuming that the nucleon uniformly distributes inside the region of radius R0 . We
then obtain 2
1 3
BW (EI) = e2 R02I . (8.64)
4π I + 3
We have set eeff to the elementary charge e in order to have a measure of the single-
particle transition. By introducing similar approximations, and further approximating
I 2 (gs − 2(I + 1)−1 g )2 ≈ 10 [37], we obtain
2
10 3
BW (MI) = R02I−2 μ2N (8.65)
π I +3
for the magnetic transition [37]. The values given by Eqs. (8.64) and (8.65) are called
the Weisskopf unit and play an important role in judging the properties of the nucleus,
especially concerning whether a certain motion of the nucleus is of single-particle
nature or of a collective nature.
For even–even nuclei as well, by taking the initial and final states to be
+
|0g.s. = |(nj)2π ; 0+ , |21+ = |(nj)2π ; 21+ , (8.66)
Table 8.2 E2 transition probability B(E2) in units of e2 fm4

Nucleus Initial Final Bexp. (E2) Bsp (E2) Effective
configuration configuration charge eeff /e
17 O s1/2 d5/2 6.3 35 0.42
8
17 F s1/2 d5/2 64 43 1.2
9
41 Ca p3/2 f7/2 66 40 1.3
20
41 Sc p3/2 f7/2 110 40 1.7
21
−1
207 Pb
82 f5/2 p−1
1/2 70 81 0.9
209 Bi
83 f7/2 h9/2 40 ± 20 2.3 4 ± 1.5
we can derive
5 2 3 2 2
+ +
B E2; 0g.s. → 21 ∼ e R (8.67)
4π 5 0
as an estimate based on the simple shell model. The index π in Eq. (8.66) refers to
proton. Like the case for odd nuclei, we assumed that j is large, and used j|r 2 |j =
3R02 /5.
Incidentally, note that in general
1
B(E2; 21+ → 0g.s.
+
)= +
B(E2; 0g.s. → 21+ ) . (8.68)
5
Table 8.2 compares the experimental values of the quadrupole transition proba-
bility from the first excited state to the ground state for several doubly magic ±1
nuclei with the single particle estimate given by Eq. (8.61). In the latter, we supposed
the configurations for the valence nucleon in the initial and final states to be those
given in the second and third columns, respectively, from the experimentally known
spin and parity and the shell model. The last column is the effective charge which
is required to reproduce the experimental value with the single-particle model. The
table shows that the single-particle model well reproduces the experimental value.
However, a large effective charge is needed for Bi.
On the other hand, Fig. 8.4 shows the experimental values of the deformation
parameter β, which have been deduced from the experimental values of the quadru-
pole transition probability B(E2) from the ground state to the first excited 2+ state
based on the collective model (Eq. (8.76), which will be discussed later), in the ratio
to the estimate of the single-particle model βsp for many even–even nuclei in the
wide range of the atomic number [38]. The equivalent radius has been assumed to be
R0 = 1.2A1/3 fm in estimating β. Also, the formula βsp = 1.59/Z, which is obtained
by replacing B(E2) in Eq. (8.76) with the single-particle estimate Eq. (8.67), was used
for βsp . In contrast to the case for doubly magic±1 nuclei shown in Table 8.2, the
experimental values for the nuclei far from the magic numbers significantly deviate
from the estimate of the single-particle model. The experimental values are much
more enhanced than the estimates of the single-particle model especially for the
nuclei in the region of mass number between 150 and 200 such as rare-earth nuclei,
and for the nuclei whose atomic number exceeds 86 such as Rn, U and Th.
Fig. 8.4 The atomic number dependence of the deformation of even–even nuclei
8.3.4 Connection Between Electromagnetic Transitions and

the Shapes and Collective Motions of Nuclei
The nuclei with large deformation, i.e., with large quadrupole transition probability,
which have been pointed out regarding Fig. 8.4, well coincide with those nuclei
which are expected to be statically deformed in their ground state and to be easily
excited to the rotational excited states from the energy level structure in the vicinity
of the ground state. In addition, the experimental values of the quadrupole transition
probability systematically show significantly larger values than the prediction of the
single-particle model for the nuclei whose mass number is around 120 such as Cd
isotopes with the atomic number 48 as well. The structure of the energy spectrum
suggests that they are the nuclei which are expected to be easily excited to the
vibrational excited states of the quadrupole type, although they are spherical in the
ground state. These nuclei can be more naturally described in the collective model
than in the single-particle model.
Here, we mention some basic results concerning the collective model, i.e., the
geometrical collective model, which is often called the Bohr–Mottelson model.
The geometrical collective model uses the deformation parameters αλμ introduced
by Eq. (7.2) as dynamical variables. Correspondingly, the density distribution of
protons is assumed to be given by
ρ̂p (r) = ρp(0) Θ[R̂(θ, ϕ) − r]

⎡ ⎛ ⎞ ⎤
∞
λ
= ρp(0) Θ ⎣R0 ⎝1 + α̂00 + ∗
α̂λμ Yλμ (θ, ϕ)⎠ − r ⎦ (8.69)
λ=2 μ=−λ
using the step function Θ(x).14 The operator of the electric transition is then given
by
3 λ
Q̂λμ = e ρ̂p (r)r λ Yλμ (θ, ϕ)d3 r ∼ eZ R0 α̂λμ . (8.70)
V 4π
Here, we truncated the sum by the first order of the operator of the deformation
parameter.15
For spherical nuclei, if we approximate the surface vibration by the harmonic
oscillator model, then the quadrupole transition probability from the first excited
state to the ground state is given by

3 2 2
B(E2; 21+ → +
0g.s. ) = eZ R , (8.71)
4π 0 2B2 Ω2
if the first excited state is the one phonon state with the angular momentum and parity
2+ . The Ω2 and B2 represent the excitation energy and the mass parameter of the
quadrupole vibration, respectively.16
Exercise 8.3 Derive the expressions of B(E2) from the two phonon states 02+ , 22+ , 41+
of the quadrupole vibration to the one phonon state 21+ and discuss their ratios to
B(E2; 21+ → 0g.s.
+
).
On the other hand, for axially symmetric even–even nuclei with quadrupole defor-
mation, the Euler angles ω = {φ, θ, ψ} which specify the directions of principal axes
become the variables for the rotational motion and the total wave function is given
by
1/2
2I + 1 " I #
|ΨIMK = DMK (ω)ΦK (q) + (−1)I+K DM−K
I
ΦK̄ (q) ,
16π 2 (1 + δK0 )
(8.72)
where DMK I
is the Wigner D function, and K is the component of the angular momen-
tum along the direction of the symmetry axis. ΦK (q) is the wave function of the
intrinsic motions, and ΦK̄ (q) is the wave function which is obtained by rotating
14 Θ(x) = 1 (x ≥ 0), Θ(x) = 0 (x < 0).

15 Seep.13 of [39] for the expression including higher order terms.
16 The values of Ω and B can be theoretically estimated in the liquid-drop model which views
2 2
the nucleus as an irrotational incompressible fluid. However, the resulting values are quantitatively
in significant disagreement with the experimental data. Instead, more reliable estimates can be
obtained phenomenologically from the experimental values of the excitation energy of the first
excited 2+ state E2 and of B(E2; 21+ → 0g.s.
+ ).
ΦK (q) by 180◦ around the second axis when the symmetry axis is taken to be the
third axis. The operator of the electric quadrupole moment and the E2 transition is
given by

5 1/2 λ
Q̂λμ ∼ eQ0 Dμ0 , (8.73)
16π

16π 1/2 3 2
Q0 = Z R β, (8.74)
5 4π 0
where λ = 2. Q0 is the intrinsic quadrupole moment.

Using Eqs. (8.72)–(8.74) and the properties of the D function, one can show
that the electric quadrupole transition probability within a rotational band with the
quantum number K is given by
5 2 2
B(E2; KI1 → KI2 ) = e Q0 I1 2K0|I2 K2 . (8.75)
16π
Especially, the value of B(E2) from the first excited 21+ state to the ground state 0g.s.
+
of the ground state rotational band with K = 0 of even–even nuclei is given by

2
1 3
B(E2; 21+ → +
0g.s. ) = ZeR02 β2 . (8.76)
5 4π
Equation (8.76) shows that the magnitude of B(E2) increases proportionally to the
square of the deformation β.17,18 Note that the electromagnetic transition probabili-
17 One can estimate the deformation parameter β, more precisely βR2 , from the experimental data of
0
B(E2) by using Eq. (8.76). However, B(E2) gives only the magnitude of the deformation parameter.
One of the standard methods to determine the sign is to use the so-called reorientation effect
in the Coulomb excitation. Recently, it has been attempted to precisely determine the deformation
including the sign through the analyses of heavy-ion fusion reactions at energies below the Coulomb
barrier [40–45].
18 The transition from high excited states to lower states within a nucleus can take place not only
by the electromagnetic transition which emits a photon, but by ejecting an electron with the energy
released in the decay. The latter process is called the internal conversion (see [46] for details).
Hence the total transition probability is given by
w = wγ + we = (1 + α)wγ , (8.77)
α ≡ we /wγ , (8.78)
if we denote the transition probability by emitting photon by wγ , and the transition probability
by the internal conversion process by we . α is called the internal conversion coefficient. If the
transition energy is large to some extent, then α is small, so that one can ignore the contribution of
the internal conversion. On the other hand, the internal conversion becomes the main decay process
if the excitation energy of the first excited 21+ state is small as in the case of 238
92 U, where it is about
45 keV. Equation (8.76) holds in this case as well. However, care should be taken when one tries to
estimate the magnitude of β from the lifetime of the 21+ state.
ties between different states within a rotational band are related to each other through
the Clebsch–Gordan coefficients as is given by Eq. (8.75) in the case of ideal rota-
tional motion, where the deformation does not change with the angular momentum,
namely, where the intrinsic state does not change in the ground and excited states.19
References
1. H.J. Rose, G.A. Jones, Nature 307, 245 (1984)

2. B. Blank, M.J.G. Borge, Prog. Part. Nucl. Phys. 60, 403 (2008); B. Blank, M. Płoszajczak.
Rep. Prog. Phys. 71, 046301 (2008)
3. K. Kato, K. Ikeda, Butsuri 61, 814 (2006); T. Myo, Y. Kikuchi, H. Masui, K. Kato. Prog. Part.
Nucl. Phys. 79, 1 (2014)
4. D.M. Brink, N. Takigawa, Nucl. Phys. A 279, 159 (1977)
5. S.Y. Lee, N. Takigawa, Nucl. Phys. A 308, 189 (1978)
6. S.Y. Lee, N. Takigawa, C. Marty, Nucl. Phys. A 308, 161 (1978)
7. M. Abramowitz, I.A. Stegun, Handbook of Mathematical Functions with Formulas, Graphs,
and Mathematical Tables (Dover, London, 1978)
8. M. Nogami, Nuclear Physics (Shoukabou, Tokyo, 1973). Japanese edn.
9. D.F. Jackson, M. Rhoades-Brown, Ann. Phys. (NY) 105, 151 (1977); S. A. Gurvitz, G. Kalber-
mann, Phys. Rev. Lett. 59, 262 (1987); S. A. Gurvitz. Phys. Rev. A 38, 1747 (1988)
10. S. Åberg, P.B. Semmes, W. Nazarewicz, Phys. Rev. C 56, 1762 (1997)
11. J.M. Blatt, V.F. Weiskopf, Theoretical Nuclear Physics (Wiley, New York, 1952)
12. M.A. Preston, Physics of the Nucleus (Addison-Wesley, London, 1962)
13. K. Kikuchi, M. Kawai, Nuclear Matter and Nuclear Reactions (North-Holland, Amsterdam,
1968)
14. S. Yoshida, M. Kawai, Nuclear Reactions (Asakura, Tokyo, 2002). Japanese edn.
15. A.M. Lane, R.G. Thomas, Rev. Mod. Phys. 30, 257 (1958)
16. E.P. Wigner, Phys. Rev. 70, 606 (1946); E.P. Wigner, L. Eisenbud. Phys. Rev. 72, 29 (1947)
17. T. Teichman, E.P. Wigner, Phys. Rev. 87, 123 (1952)
19 Here, we add comments on the liquid-drop model. The experimental data show systematics given
by
Z2 $ %
E2+ B(E2; 21+ → 0g.s.
+
) ≈ (25 ± 8) MeV e2 fm4 ∝ Z 2 /A . (8.79)
1 A
On the other hand, the liquid-drop model leads to
Ω2 B(E2) ∝ Z 2 R04 /B2 ∝ Z 2 R04 /AR02 ∝ Z 2 A2/3 /A ∼ Z 2 A−1/3 (8.80)
for vibrational excitations.

The prediction of the liquid-drop model Eq. (8.80) has a different mass number dependence from
that of the experimental data given by Eq. (8.79). One reason of the discrepancy is the quantum
mechanical effect, which plays important roles in nuclear physics. It is necessary to treat nucleus as
a quantum liquid in order to handle nucleus in the liquid-drop model in a quantitatively proper way.
For example, the classical liquid-drop model evaluates the potential energy by taking into account
the effect of deformation only in the ordinary space in deriving the Hamiltonian for collective
motions. Accurately, however, one needs to consider the effects of deformation in the momentum
space as well as suggested by the uncertainty principle in quantum mechanics. This effect shows
up, for example, in the vibrational excited states such as the giant quadrupole resonance (GQR)
[39]. The discrepancy between the experimental data and the prediction of the liquid-drop model
appears also regarding the moment of inertia of deformed nuclei.
References 213
18. A. Arima, H. Horiuchi, K. Kubodera, N. Takigawa, Adv. Nucl. Phys. 5, 345 (1972)
19. B. Buck, A.C. Merchant, S.M. Perez, Phys. Rev. C 45, 2247 (1992); B. Buck, J. C. Johnson,
A. C. Merchant, S. M. Perez. Phys. Rev. C 53, 2841 (1996)
20. I. Tonozuka, A. Arima, Nucl. Phys. A 323, 45 (1979)
22. P. Hänggi, P. Talkner, M. Borkovec, Rev. Mod. Phys. 62, 251 (1990). and references therein
23. U. Weiss, Quantum Dissipative Systems (World Scientific, Singapore, 1999)
24. P. L. Kapur, R. Peierls, Proc. Roy. Soc. London A 163, 606 (1937); A. Schmid, Ann. Phys.
(NY) 170, 333 (1986)
25. T. Kindo, A. Iwamoto, Phys. Lett. B 225, 203 (1989)
26. N. Takigawa, K. Hagino, M. Abe, Phys. Rev. C 51, 187 (1995)
27. N. Bohr, J.A. Wheeler, Phys. Rev. 56, 426 (1939)
28. H.A. Kramers, Physica 7(284), 284 (1940)
29. R. Kubo, M. Toda, N. Hashitsume, Statistical Physics II, Non-equilibrium Statistical Mechanics
(Springer, New York, 1985)
30. H. Risken, The Fokker-Planck Equation (Springer, Berlin, 1989)
31. J.S. Langer, Ann. Phys. (NY) 41, 374 (1967); I. K. Affleck. Phys. Rev. Lett. 46, 388 (1981)
32. R.P. Feynman, A.R. Hibbs, Quantum Mechanics and Path Integrals (McGraw-Hill, New York,
1965)
33. H. Grabert, P. Olschowski, U. Weiss, Phys. Rev. B 36, 1931 (1987)
34. N. Takigawa, M. Abe, Phys. Rev. C 41, 2451 (1990)
35. Y. Abe, C. Gregoire, H. Delagrange, J. Phys. 47, C4 (1986); T. Wada, Y. Abe, N. Carjan, Phys.
Rev. Lett. 70, 3538 (1993); Y. Abe, S. Ayik, P.-G. Reinhard, E. Suraud, Phys. Rep. 275, 49
(1996)
36. P. Fröbrich, R. Lipperheide, Theory of Nuclear Reactions (Clarendon Press, Oxford, 1996)
38. S. Raman, C.W. Nestor Jr., P. Tikkanen, At. Data Nucl. Data Tables 78(1), 1 (2001)
40. A.B. Balantekin, N. Takigawa, Rev. Mod. Phys. 70, 77 (1998)
41. M. Dasgupta, D.J. Hinde, N. Rowley, A.M. Stefanini, Ann. Rev. Nucl. Part. Sci. 48, 401 (1998)
42. K. Hagino, N. Rowley, A.T. Kruppa, Comp. Phys. Commun. 123, 143 (1999)
43. H. Esbensen, Nucl. Phys. A 352, 147 (1981)
44. K. Hagino, N. Takigawa, Butsuri 57, 588 (2002)
45. K. Hagino, N. Takigawa, Prog. Theor. Phys. 128, 1001 (2012)
46. K. Yagi, Nuclear Physics (Asakura, Tokyo, 1971). Japanese edn.
Chapter 9
Synthesis of Elements
Abstract It is a long-standing intriguing problem to explore the synthesis of

elements, and extensive studies are going on (See Clayton (Principles of Stellar
Evolution and Nucleosynthesis, The University of Chicago Press, Chicago, 1968,
[1]), Rolfs and Rodney (Cauldrons in the Cosmos, The University of Chicago Press,
Chicago, 1988, [2]), Thompson and Nunes (Nuclear Reactions for Astrophysics:
Principles, Calculation and Applications of Low-Energy Reactions, Cambridge Uni-
versity Press, Cambridge, 2009, [3]), VHS Element Genesis—Solving the Mystery,
2001, [4]). Roughly speaking, the synthesis of elements can be divided into the pri-
mordial nucleosynthesis, which is also called the Big Bang nucleosynthesis, and the
stellar nucleosynthesis. The light elements such as deuterons, He and Li have been
synthesized by nuclear reactions within 3–15 min after the Big Bang. This is the Big
Bang nucleosynthesis. It terminates at the elements of mass number 7, since there
is no stable nucleus of mass number 8. Stars started to be formed about one billion
years later. Then, thermal nuclear reactions took place inside stars, and nuclei up to
Fe, which has the largest binding energy per nucleon, have been successively syn-
thesized depending on the mass of each star. (Precisely speaking, the nucleus which
has the largest binding energy per nucleon is 62 28 Ni as remarked in Chap. 2.) Nuclei
beyond Fe are synthesized either slowly by the neutron capture reactions called slow
process (s-process) inside red giant stars, or synthesized by the explosive astrophys-
ical phenomenon called rapid process (r-process). Nuclei with extremely large mass
number such as U are thought to be synthesized at the supernovae explosion, which
is one of the last stages of stars. In this chapter we learn some basics concerning the
nuclear reactions related to nucleosynthesis (See Clayton (Principles of Stellar Evo-
lution and Nucleosynthesis, 1968, [1]), Rolfs and Rodney (Cauldrons in the Cosmos,
1988, [2]), Thompson and Nunes (Nuclear Reactions for Astrophysics: Principles,
Calculation and Applications of Low-Energy Reactions, Cambridge University
Press, Cambridge, 2009, [3]) for details).
9.1 The Astrophysical S-Factor and Gamow Factor
The nuclear reactions between two charged particles in stars occur with the collision
energy much lower than the Coulomb barrier like the proton–proton reactions in the
DOI 10.1007/978-4-431-55378-6_9
216 9 Synthesis of Elements
Sun discussed in Sect. 1.4. Hence it is customary to discuss not the cross section
σ (E), but S(E), which is defined by
S(E)
σ (E) = exp(−2π η) , (9.1)
E
√
Z 1 Z 2 e2 Z 1 Z 2 e2 μ 1
η= = √ √ , (9.2)
v 2 E
by taking out the trivial exponential factor (Gamow factor) exp(−2π η), where η is
the Sommerfeld parameter. S(E) is called the astrophysical S-factor.
Since the Gamow factor has been taken out, S(E) is expected to vary gently as
a function of energy unless resonance phenomena are involved. The energy of the
Gamow peak (see the next section), which plays a central role in nuclear reactions
in stars, is much lower than the lowest energy where experiments can be performed
in many cases. The cross section at the Gamow peak is therefore often estimated
by extrapolating the experimental data at high energies, where experiments can be
performed. The gentle energy dependence of S(E) is convenient in this connection,
too.
Figure 9.1 shows the experimental data of S(E) factor for the 3 He(3 He,2p)4 He
reaction. It is remarkable that the experiments have been done down to energies
as low as the Gamow peak in the Sun for this reaction. The figure shows that the
experimental value of S(E) is systematically larger than the value obtained by the
extrapolation of high energy data. This situation is not special to this reaction, but
is common to many other systems so far studied, and the enhancement of the cross
Fig. 9.1 The S(E) factor for

the 3 He(3 He,2p)4 He
reaction. U0 is the screening
energy. Taken
from [5]
9.1 The Astrophysical S-Factor and Gamow Factor 217
section has been attempted to be explained in terms of the screening effect due to,
e.g., the electrons of the target nucleus.
9.2 Gamow Peak
If we denote the reaction rate per pair of colliding nuclei by λ,1 then it is given by

λ = σ v = σ (E)v(E)φ(v)dv . (9.3)
Here, φ(v) is the normalized Maxwell–Boltzmann distribution, and is given by

2 √
φ(v)dv = √ β 3/2 exp(−β E) EdE , (9.4)
π
with β = 1/kT , where T and k are the temperature of the star and the Boltzmann
constant, respectively.
Inserting Eqs. (9.1) and (9.4) into Eq. (9.3), and using the definition of the
Sommerfeld parameter Eq. (9.2), we obtain

8 3/2 ∞
λ= β S(E) exp −β E − bE −1/2 dE , (9.5)
μπ 0
√
π Z 1 Z 2 e2 2μ
b= . (9.6)

b2 is called the Gamow energy.

The exponential factor in the integrand of Eq. (9.5) is sharply peaked at an energy
because of the competition between the Gamow factor and the Maxwell distribution.
Since S(E) is a gentle function of energy unless resonances are involved, one can
apply the method of steepest descent to calculate the integral with little error. From
the condition
d
β E + bE −1/2 = 0 , (9.7)
dE
the saddle point energy can be obtained as
2/3
b
E0 = . (9.8)
2β
E 0 is called the Gamow peak. Finally, the reaction rate is given by

√
4 2 1 1/2
λ ≈ √ √ E 0 β S0 e−3β E0 , (9.9)
3 μ
where S0 = S(E 0 ).
1 The actual reaction rate per pair is given by multiplying the pair density.
Exercise 9.1 Prove Eq. (9.9) based on the formula of the method of steepest descent

2π
g(x)e− f (x) dx ≈ g(x0 )e− f (x0 ) ,
f (x0 )
where x0 and f are the position of the saddle point and the second derivative of f ,
respectively.
9.3 Neutron Capture Cross Section
We show in this section that the capture cross section of s-wave neutrons, i.e., the
cross section of low-energy reactions such as the reactions by thermal neutrons, is
proportional to 1/v (1/v-law), where v is the velocity of the neutron relative to the
target nucleus, and that the v dependence changes as the energy increases.
To make the argument simple, we approximate the cross section of the capture
reaction by the absorption cross section by a strongly absorbing square well poten-
tial. Furthermore, we represent the strong absorption effect by the incoming wave
boundary condition at the potential surface at r = a. If we express the wave function
of the relative motion as R(r ) = u(r )/r , and name the region of r ≤ a the region I,
and that of r > a the region II, then
u I (r ) = Ae−iK r + BeiK r , (9.10)

−ikr
u I I (r ) = e − Se ikr
, (9.11)

2μ(E + V0 )
K = , (9.12)
2

2μE
k= . (9.13)
2
V0 is the depth of the potential, which is assumed to be a real number.

B = 0 if we assume the strong absorption and impose the incoming wave bound-
ary condition. In this case, the continuity conditions of the wave function and of its
derivative at r = a lead to
K − k −2ika
S= e . (9.14)
K +k
Hence the cross section of the capture reaction is given by

π
σ = (1 − |S|2 )
k2
π 1
= 2 4k K . (9.15)
k |K + k|2
9.3 Neutron Capture Cross Section 219

Since K ≈ K 0 ≡ 2μV0 /2 at low energies, we have
π 1 1
σ ∼ 4k K 0 ∝ . (9.16)
k 2 K 02 k
Equation (9.15) shows also that the v dependence of the capture cross section varies
with the energy of collision (see Ref. [1]).
Exercise 9.2 Discuss the v dependence of the capture cross section of p-wave neu-
trons by heavy nuclei.
9.4 Synthesis of Heavy Elements: S-Process and R-Process
As mentioned at the beginning of this chapter, nuclei with large mass number beyond
Fe are believed to be synthesized by either the so-called s-process (slow process) or
by the r-process (rapid process).
In the abundance curve shown in Fig. 1.7, the right peak (indicated by “s”) in
each double peaks at A ∼ 80, 130, and 200 is produced by the s-process. This peak
corresponds to nuclei with a neutron magic number (N = 50, 82, or 126). The left
peak (indicated by “r”) is produced by the r-process.
Figure 9.2 conceptually shows the paths of the synthesis of elements by the s- and
r-processes. It shows how heavy nuclei are built up by successive neutron captures,
interspersed with β-decays. While the s-process nucleosynthesis proceeds almost
along the stability line, the r-process nucleosynthesis rapidly proceeds via the region
of neutron-rich unstable nuclei.2 The study of r-process is therefore extensively going
on in these days in connection with the study of unstable nuclei. For the r-process, it
is suggested that the neutron number stays unchanged for a while when it coincides
with one of the magic numbers, and nuclei with various atomic numbers with the
fixed neutron number are synthesized.3
2 Large neutron fluxes are expected in supernovae explosions. Consequently, neutron-rich nuclei
are successively synthesized by (n,γ ) reaction. The binding energy of neutron becomes small if
the number of excess neutrons gets too large. The (n,γ ) reaction then becomes balanced with the
inverse (γ ,n) reaction, and the synthesis does not proceed further. Eventually, the unstable neutron-
rich nucleus either proceeds towards stability line via β-decays, or repeats (n,γ ) reactions after
increasing the atomic number via β-decays to synthesize nuclei with still larger proton and neutron
numbers. The r-process path is thus determined.
3 The regions where the neutron number does not change for a while in the r-process nucleosynthesis
are called waiting points and the corresponding nuclei are called waiting point nuclei. It is considered
that the peaks on the left side in Fig. 1.7 appear reflecting the waiting points.
Fig. 9.2 Conceptual illustration of the synthesis of elements by s- and r-processes. Taken from [6]
9.5 Column: Overview of the Synthesis of Elements
Figure 9.3 summarizes the various paths of the synthesis of elements hitherto
described. The rp-process, which has been added in the figure, is the process,
where proton-rich nuclei with the mass number up to ∼ 100 are synthesized
by (p,γ ) and (α,p) reactions.4 In addition, there exists the p-process, called
sometimes γ process, where neutron-deficient nuclei with the mass number
larger than that for the nuclei made by the rp-process are synthesized by photo-
disintegration such as (γ ,n), (γ ,p), and (γ , α) reactions [3]. Nucleosynthesis
by neutrinos associated with supernovae explosions is also one of the current
topics [7].
In parallel with the research to uncover the synthesis of elements, various
studies of astrophysical phenomena such as the physics of neutron stars from
the points of view of nuclear physics are very actively ongoing. It forms an
important branch of current nuclear physics.
4 Recent research on neutron stars suggests that the rp-process ends at 105
52 Te.
9.5 Column: Overview of the Synthesis of Elements 221
Fig. 9.3 Paths of nucleosynthesis. After [4] with some modifications
References
1. D.D. Clayton, Principles of Stellar Evolution and Nucleosynthesis (The University of Chicago
Press, Chicago, 1968)
2. C.E. Rolfs, W.S. Rodney, Cauldrons in the Cosmos (The University of Chicago Press, Chicago,
1988)
3. I.J. Thompson, F.M. Nunes, Nuclear Reactions for Astrophysics: Principles, Calculation and
Applications of Low-Energy Reactions (Cambridge University Press, Cambridge, 2009)
4. VHS Element Genesis—Solving the Mystery, Sci. Eds. Y. Motizuki, I. Tanihata, Y. Yano, R.
Boyd (RIKEN & Image Science, Inc., 2001)
5. M. Junker et al., Phys. Rev. C 57, 2700 (1998)
6. M. Arnould, S. Goriely, K. Takahashi, Phys. Rep. 450, 97 (2007)
7. J.N. Bahcall, Neutrino Astrophysics (Cambridge University Press, Cambridge, 1989)
Appendix A
Important Formulae and Their Derivation
Abstract This chapter collects appendices to the previous chapters.
A.1 Basics of Scattering Problems
Here we summarize some basics for scattering problems such as the partial wave
expansion of the scattering amplitude.1
A.1.1 Partial Wave Expansion
(1) The asymptotic form of the radial wave function for the partial wave (the case
of a short range force; k is the incident wave number, S is the scattering matrix,
i.e., the S-matrix, and δ is the phase shift)
1 −i(kr − 1 π )
− S()ei(kr − 2 π )
1
R (r ) ∼ e 2 (A.1)
kr
1 1
∼ sin kr − π + δ . (A.2)
kr 2
(2) Partial wave expansion of the elastic scattering amplitude2 (θ is the scattering
angle, and α is the index of each state)
1 We consider the scattering of particles without spin. See [1] for the scattering of particles with
spin.
2 When one takes the partial wave sum in the case of the scattering problem involving Coulomb
force, it is traditional to separate out the Rutherford scattering amplitude.

DOI 10.1007/978-4-431-55378-6
224 Appendix A: Important Formulae and Their Derivation
∞
1
f (el) (θ ) = (2 + 1) Sα,α () − 1 P (cos θ ) , (A.3)
2ik =0
Sα,α () = e2iδ . (A.4)
(3) Partial wave expansion of the inelastic scattering amplitude (α and β are the
state indices)
1
f β(inel)
=α (θ ) = (2 + 1) Sβ,α ()P (cos θ ) . (A.5)
2ik
(4) Cross sections of the elastic, inelastic and total inelastic scattering
π
σ (el) = σ(el) = 2
(2 + 1)|Sα,α () − 1|2 , (A.6)

k α
(inel) π
σβ(inel)
=α = σβ=α () = 2 (2 + 1)|Sβ,α ()|2 , (A.7)

k α
(inel) π
σ (inel) = σβ=α = 2 (2 + 1) 1 − |Sα,α ()|2 , (A.8)

β=α
kα
(5) Unitarity condition

|Sβ,α ()|2 = 1 . (A.9)
β
(6) Total cross section (ReS is the real part of the S-matrix)
2π
σ (total) = σ (el) + σ (inel) = (2 + 1)(1 − ReS) . (A.10)
kα2
(7) Optical theorem (Im f (el) is the imaginary part of f (el) )
4π
σ (total) = Im f (el) (θ = 0) . (A.11)
kα
A.1.2 Sommerfeld–Watson Transformation
For high energy scattering, it is more convenient to use the integral representation
of the scattering amplitude instead of the partial wave expansion. The integral repre-
sentation sometimes becomes useful for scattering at low energies as well. Starting
from the integral representation of the scattering amplitude and deriving character-
istic closed form expression through some suitable approximation scheme such as
the stationary phase approximation or the method of steepest descent, one can often
Appendix A: Important Formulae and Their Derivation 225
Fig. A.1 The contour used

for the Sommerfeld–Watson
transformation
obtain intuitive understanding for complicated patterns appearing in the differential

cross section in terms of, e.g., the interference of various scattering waves or the
caustics such as the rainbow scattering and the glory scattering [2–7].
The transformation from the partial wave sum to the integral representation can
be achieved by using the knowledge of the contour integral in the complex plane.
We first represent the scattering amplitude as

1 1 1
f (θ ) = − d e iπ
+ [S(k, ) − 1] P (cos θ ) (A.12)
2k C sin π 2
by using the integral along the contour C in the complex plane shown in Fig. A.1.
The transformation from Eq. (A.3) to Eq. (A.12) is known as the Sommerfeld–Watson
transformation. One can easily confirm that Eq. (A.12) reduces to Eq. (A.3) by noting
that the integrand has a pole of order one at each integer value of and that the
corresponding residue is given by
− 0 1
lim = (−1)0 (0 : integer). (A.13)
→0 sin π π
A.1.3 Poisson Sum Formula
The integral in Eq. (A.12) is a contour integral in the complex plane. Let us attempt to
convert it into an integral along the real axis. To that end, we first rewrite Eq. (A.12) as

i eiπ 1
f (θ ) = − d + [S(k, ) − 1] P (cos θ ) . (A.14)
k C eiπ − e−iπ 2
We then remark that
eiπ eiπ(+iε)
= lim (A.15)
eiπ − e−iπ ε→0+ eiπ(+iε) − e−iπ(+iε)

= −e2iπ 1 + e2iπ + e4iπ + · · · (A.16)
on the contour C+ above the real axis, and that
eiπ eiπ(−iε)
= lim (A.17)
eiπ − e−iπ ε→0+ eiπ(−iε) − e−iπ(−iε)
= 1 + e−2iπ + e−4iπ + · · · (A.18)
on the contour C− below the real axis, hence

eiπ eiπ
d iπ h() = + d h()
C e − e−iπ C+ C− eiπ − e−iπ
∞
∞
= d e2iπm h() , (A.19)
m=−∞ −1/2
for a function h(). We finally obtain

∞ ∞
i
f (k, θ ) = − (−1)m dλe2πimλ λ [S(k, λ − 1/2) − 1] Pλ−1/2 (cos θ )
k m=−∞ 0
(A.20)
by introducing the Langer replacement + 1/2 = λ, which is referred to also as the
Langer correction or the Langer modification. Equation (A.20) is called the Poisson
sum formula. In the semi-classical theory for scattering [8], m represents the number
of times which the incident particle circulates the target particle. The m = 0 term
governs the scattering at high energies or if the attractive force between the projectile
and target particles is weak. However, the m ≥ 1 terms also play important roles in
the scattering at low energies if a strong attractive force is involved [6].
We obtain the following Poisson sum formula for the cross section of the total
inelastic scattering by applying the similar procedure to Eq. (A.8),
∞ ∞
2π
σ (inel) = (−1) m
dλe 2iπmλ
λ 1 − |Sα,α | 2
(A.21)
kα2 m=−∞ 0
∞ ∞
= 2π (−1)m db b e2iπmkb 1 − |Sα,α (b)|2 (A.22)

m=−∞ 0
∞

≈ 2π db b 1 − |Sα,α (b)|2 (for high energy). (A.23)

0
The relation between the angular momentum and the impact parameter λ = kb has
been used in transforming Eqs. (A.21) to (A.22).
A.2 Basics of Semi-classical Theory I: WKB

Approximation
Here, let us first study the WKB approximation3 among various semi-classical meth-
ods. The WKB approximation is the method to approximately obtain the solution
of the Schrödinger equation in the limit of taking the Planck constant to be zero.
Physically, it is valid when the change of the potential energy within a wave length
can be ignored compared with the magnitude of the local kinetic energy. Hence, it
can be said to be a short wave length approximation. In this approximation, the wave
function and the phase shift as well as the scattering matrix for scattering problems
are given in terms of the action integrals in classical mechanics.
A.2.1 Wave Function
As an example, let us consider a problem where the potential is given by Vn (r ) +

VC (r ), the former and the latter being the nuclear and the Coulomb potentials, re-
spectively, and approximately obtain ψ (r ) when the radial wave function R (r ) of
an eigenstate of energy or the scattering problem with the orbital angular momentum
is expressed as R (r ) ≡ ψ (r )/r . The ψ (r ) obeys the following equation:

2 d 2
− + V (r ) − E ψ (r ) = 0 , (A.24)
2μ dr 2
V (r ) = Vn (r ) + VC (r ) + V (r ) , (A.25)
2 ( + 1/2)2 2 λ 2
V (r ) = 2
= . (A.26)
2μ r 2μ r 2
We introduced the Langer replacement ( + 1) → ( + 1/2)2 ≡ λ2 in Eq. (A.26).

In the WKB approximation, we express the wave function in the form of the
product of the amplitude and the phase factor, and write
ψ(r ) = A(r )eiS(r )/ . (A.27)
We leave out the index in Eq. (A.27) and throughout this section. Making the
substitution Eq. (A.27) in Eq. (A.24) and treating A(r ), S(r ) and V (r ) as quantities
independent of and collecting the terms with the same powers of , we obtain
3 Wenzel–Kramers–Brillion approximation. It is also called the JWKB approximation by including

the name of Jefferey.
2
1 dS i d A dS d2 S 1 d2 A 2
+ V (r ) − E A0 − 2 + A 2 1 − =0.
2μ dr 2μ dr dr dr 2μ dr 2
(A.28)
By considering the limit of → 0, and setting the coefficients of the zeroth and the
first order power of to be zero separately, we obtain
2
1 dS
+ V (r ) − E = 0 , (A.29)
2μ dr

d 2 dS
A =0. (A.30)
dr dr
Equation (A.29) corresponds to the Hamilton–Jacobi equation in classical mechanics,

and suggests that S is given by the action integral in classical mechanics

S=± 2μ(E − V (r ))dr = ± k(r )dr . (A.31)
Equation (A.30) denotes the conservation of the current density. Together with
Eq. (A.31), it shows that the amplitude A is inversely proportional to the square
root of the wave number, i.e.,
1
A∝ √ . (A.32)
k(r )
Finally, the wave function is given by

r r
1 1
ψ(r ) = A+ √ exp i k(r )dr + A− √ exp −i k(r )dr ,
k(r ) r0 k(r ) r0
(A.33)
where A+ and A− are constants. In this way, the wave function in the WKB approx-
imation is given by replacing the wave number times displacement in the arguments
of the exponential functions for a free particle with the action integral using the local
wave number. This is a natural consequence, since the WKB approximation is based
on the assumption that the spatial change of momentum is small.
The 2 term in Eq. (A.28) has to be sufficiently small compared with the other
terms in order for the WKB approximation to be applicable. Equations (A.31) and
(A.32) lead to
2 2
d2 A A dS 3 1 1 dV
∼ + ··· . (A.34)
dr 2 2 dr 16 ( k (r )/2μ) k (r )
2 2 2 2 dr
Thus, the condition for the WKB approximation to be valid is that the amount of
change of the potential energy within a wave length is much smaller than the kinetic
energy at that position.
A.3 Basics of Semi-Classical Theory II: Method

of Comparison Equation
In this section, we formulate the semi-classical theory, i.e., the WKB method, using
the method of comparison equation, which is different from the standard method
described in Sect. A.2.
A.3.1 Principle of the Method of Comparison Equation
Turning points play important roles in describing the reflection of a particle in clas-
sical mechanics. Since the turning points are defined as the points where the energy
of the particle matches the value of the potential energy, namely, as the points where
E = V (r ), they become in general complex numbers if the potential is a complex
potential as in the case of scattering by an optical potential. There exist many turning
points in the complex r plane even if V (r ) is a real potential. For example, when
the potential between the scattering nuclei has the form shown in Fig. 8.1, and when
the scattering energy is slightly above the potential barrier, there appears a turning
point, whose real part is the barrier position, at each side of, i.e., above and below,
the real axis in the complex plane.
Exercise A.1 Let us express the potential near the barrier position r B as V (r ) =
VCB − μΩ 2 (r − r B )2 /2 when the potential looks like Fig. 8.1. Discuss the change
of the positions of the classical turning points in the complex r plane as the scattering
energy E varies from below to above the barrier height VCB .
We therefore make an analytic continuation of the Schrödinger equation to com-

plex plane and seek the solution of the second order differential equation of a complex
variable r :
2
d
+ χ (r ) ψ(r ) = 0 , (A.35)
dr 2
2μ
χ (r ) = 2 (E − V (r )) . (A.36)

V (r ) can have in general both the real and the imaginary parts; V (r ) = VR (r ) +
iVI (r ), where both VR (r ) and VI (r ) are analytic functions and are real functions
when r is real.
In the method of comparison equation, we introduce another space of a com-
plex number, which we denote by σ , besides the complex r space, and introduce a
subsidiary second order differential equation in that space:

d2
+ Γ (σ ) φ(σ ) = 0 , (A.37)
dσ 2
which is called the comparison equation.

The function Γ (σ ) is chosen to have the same topological structure as that of

χ (r ). In simpler words, it is chosen to have the same number of important zero
points, that is, the same number of turning points. Moreover, it is so chosen that
the exact solutions of Eq. (A.37) are functions whose properties are well known. The
basic idea of the method of comparison equation is to obtain an approximate solution
ψ(r ) of the original Eq. (A.35) in terms of the solutions φ(σ ) of Eq. (A.37), whose
properties are well known, by introducing suitable comparison equation.
The mapping between the two spaces is made in the following way. We first map
the important zeros in the r space onto the zeros in the σ space by one to one mapping,
then define the mapping between the two spaces so that

dσ χ (r ) 1/2
= . (A.38)
dr Γ (σ )
Since the integral form of Eq. (A.38) reads

r σ
[χ (r )]
1/2
dr = [Γ (σ )]1/2 dσ , (A.39)
r0 σ0
Equation (A.38) means that the mapping is made to conserve the action integral in
two spaces.
One can show that in this case it holds
−1/2
dσ
ψ(r ) ≈ φ(σ (r )) (A.40)
dr
in the limit of → 0.
In fact, by changing Eq. (A.40) into an identity equation, and inserting it into
Eq. (A.35), one obtains
3/2 −2
dσ d2 φ 1 dσ
+ 2 Q(r ) φ(σ ) + f (σ )φ(σ ) = 0 , (A.41)
dr dσ 2 dr
where −3/2
1 d dσ d2 σ
f (σ ) = − . (A.42)
2 dr dr dr 2
In Eq. (A.41), we expressed as χ (r ) = Q(r )/2 in order to make the role of clear.
Equation (A.41) shows that Eq. (A.37) is obtained in the limit of → 0 if one defines
2
φ(σ )
the mapping by Eq. (A.38). The error is given by ( dσ
dr
)−3/2 f (σ )φ(σ )/ d dσ 2 .
A.3.2 Derivation of the WKB Wave Function
As a simple example, let us obtain the wave function at a place far from any classical
turning points. In this case, the Γ can be chosen as
Γ (σ ) = 1 . (A.43)
The solution of the corresponding comparison Eq. (A.37) is given by
φ(σ ) = Aeiσ + Be−iσ . (A.44)
On the other hand, Eq. (A.39) leads to

σ = χ 1/2 dr = k(r )dr . (A.45)
Since χ (r ) = k 2 (r ), we finally obtain

r r
ψ(r ) ∼ [χ (r )]−1/4 A exp i k(r )dr + B exp −i k(r )dr . (A.46)
The right-hand side agrees with Eq. (A.33) obtained in Sect. A.2, and is nothing but
the wave function in the WKB approximation.
Incidentally, the WKB formula for the tunneling probability can be derived by
separately treating the two turning points on both sides of the potential barrier as an
isolated turning point, and using σ as Γ for each region, and connecting the wave
functions on both sides by using the asymptotic form of the Airy function, which
is the corresponding φ. On the other hand, the formula of the barrier transmission
probability in the uniform approximation (2.77) can be obtained by considering the
potential barrier basically as a parabolic barrier and hence by taking Γ = σ 2 /4 − ε
and by studying the asymptotic behaviours of the corresponding eigenfunctions φ,
which are the Whittaker functions [9, 10]. The ε is determined so as to conserve the
action integral between the two turning points at the potential barrier in the r and
σ spaces. Also, the S matrix, which is the basis of Eqs. (8.5)–(8.8) can be derived
by mapping the inner region including r3 (see Fig. 8.1) by a linear function and the
barrier region by a quadratic function, determining approximate wave function for
each region and by connecting them using their asymptotic forms. In the connection,
it is crucial to use the correct asymptotic forms of the Whittaker and Airy functions
depending on the phase of the arguments.
A.4 Eikonal Approximation
A.4.1 Scattering Amplitude
The scattering amplitude by a potential V at high energies can be well given in the
eikonal approximation as [11]
∞
f (k, θ ) = −ik db b J0 (kbθ ) e2iΔ(b) − 1 , (A.47)

0
+∞
−m
Δ(b) = δeikonal () = V ( b2 + z 2 )dz , (A.48)
2k2 −∞
and the cross section as

∞
(inel)
σeikonal ≈ 2π db b 1 − e−4Δ I (b) (for high energy), (A.49)

0
Δ I (b) = ImΔ(b) . (A.50)
A.4.2 Glauber Theory
Here, we derive the formulae of the Glauber theory in the optical limit given by
Eqs. (2.34)–(2.37).
The starting point is Eqs. (A.47)–(A.49) of the eikonal approximation. In order to
apply these equations to the nucleon–nucleus and nucleus–nucleus scattering, let us
microscopically represent the scattering potential in terms of the sum of the nuclear
force between individual nucleons v N N (ri P − r jT ),

V (R) = v N N (ri P − r jT ) (A.51)
i P , jT

= dr P dr T v N N (r P − r T + R)ρ P (r P )ρT (r T ) . (A.52)
In these equations, R is the coordinate of the relative motion, R ≡ R P − R T , between

the centers of mass of the projectile and target nuclei, R P , R T . ri P (r jT ) is the
coordinate of the i-th nucleon in the projectile ( j-th nucleon in the target) measured
from the origin of the space fixed coordinate system. r P and r T are the intrinsic
coordinates of the nucleons in the projectile and target nuclei measured from R P and
R T , respectively. ρ P (r P ) and ρT (r T ) are the nucleon densities of the projectile and
target nuclei at r P and r T , respectively. In transforming fromEqs. (A.51) to (A.52),
we used that the nucleon densities are given by ρ P (r P ) = i P δ(ri P − R P − r P ),
and ρT (r T ) = jT δ(r jT − R T − r T ), like Eq. (5.2). v N N (r) is the nuclear potential
between two nucleons, which are separated by r. The potential given by Eq. (A.52)
is called the double folding potential, and is often used for the interaction between
two nuclei.
Since the nuclear force is of short range, we approximate it as
v N N (r) = (vRe + ivIm )δ(r). (A.53)
The imaginary part (ivIm ) has been introduced to take into account the other processes
than the elastic scattering. The potential is then given by

V (R) = (vRe + ivIm ) drρ P (r − R)ρT (r) . (A.54)
By inserting the result of Eq. (A.54) into Eq. (A.48) and using Eq. (A.49), we obtain
∞
(inel) 2μnn
σ ≈ 2π db b 1 − exp vIm Ov (b) (for high energy) , (A.55)
0 knn 2
where the overlap function Ov is given by Eq. (2.36). μnn and knn are the reduced
mass and the wave number in the nucleus–nucleus scattering, respectively.
We next use the Born approximation and the optical theorem in order to relate
the strength of the nuclear force to the total cross section in the nucleon–nucleon
scattering. By applying the formula in the Born approximation,

1 2m
f (1) (θ ) = − e−iq·r V (r)dr ,
4π 2
to the nucleon–nucleon scattering, and assuming the limit of the short range force
given by Eq. (A.53), we obtain
μN N
fN N = − (vRe + ivIm ) (A.56)
2π 2
for the scattering amplitude f N N . μ N N = M N /2 is the reduced mass in the nucleon–
nucleon scattering. If we use here the optical theorem, we obtain
2π 2 k N N (total)
vIm = − σ (k N N ) . (A.57)
μ N N 4π N N
k N N is the wave number of the relative motion in the nucleon–nucleon scattering,

σ N(total)
N (k N N ) is the total cross section when the nucleon–nucleon collision takes place
at the wave number k N N . By inserting the results of Eq. (A.57) into Eq. (A.55), we
obtain
∞
μnn k N N (total)
σ (inel) ≈ 2π db b 1 − exp − σ N N (k N N )Ov (b) . (A.58)
0 knn μ N N
In order for Eq. (A.58) to agree with Eq. (2.34), it is required that
μnn k N N
=1. (A.59)
knn μ N N
This is equivalent to the condition (2.37). Equations (2.34)–(2.37) are derived in this
way. Since the proof used the same reaction cross section irrespective of the nucleon
species, it will be adequate to use the average cross section σ̄ N N .
A.5 Non-local Potential
A characteristic of nuclear physics is that a non-local potential is often involved.

The Fock potential in the Hartree–Fock approximation for nuclear structure, and the
exchange potentials which appear when one microscopically describes the scattering
between complex nuclei such as the scattering between α particle and nucleus and
between two heavy nuclei by taking into account antisymmetrization among nucleons
based on, e.g., the resonating group method are typical examples.
On the other hand, one usually assumes a local potential for phenomenological
analyses of nuclear structure and nuclear reactions. In this section, we review some
standard methods to convert the Schrödinger equation involving a non-local potential,
hence an integro-differential equation, into an equivalent local differential equation,
and interpret the characteristic effects due to non-locality in other physical terms, such
as the effective mass, which are relevant to the description by using local potential,
hence by a local differential equation [1].
A.5.1 Integro-Differential Equation
We consider a one-dimensional problem and represent the local and the non-local
potentials by U and V , respectively. In this case, the wave function obeys the integro-
differential equation:

2 d2 ψ(x)
− + U (x)ψ(x) + V (x, x )ψ(x )dx = Eψ(x) . (A.60)
2m dx 2
A.5.2 Equivalent Effective Local Potential: WKB

Approximation
One useful method to study the characteristic physical effects of a non-local potential
is the method based on the WKB approximation. To that end, we first rewrite the
term of the non-local potential into

V (x, x )ψ(x )dx = V (x, x )e (x −x) p̂ψ dx ψ(x) .
i
(A.61)
The subscript ψ of the momentum operator p̂ has been attached in order to indicate
that it operates on the wave function ψ. The essence of the WKB approximation
is to represent the wave function in the form of a plane wave by using the local
momentum. Also, it is based on the assumption that the spatial variation of the wave
number is small. Hence,
p̂ψ ψ(x) ∼ p(x)ψ(x) , (A.62)
where p(x) is the local momentum at the position x.
A.5.2.1 WKB Approximation I: Local Momentum Approximation,

Energy Dependent Potential
By applying the approximation (A.62) to Eq. (A.61), one obtains the following equiv-
alent effective local potential:

V (x, x )e (x −x) p(x) dx
i
Ueff (E; x) = U (x) + (A.63)
√

V (x, x )e (x −x) 2m(E−Ueff (E;x))
dx
i
= U (x) + (A.64)
√

V (x, x )e (x −x) 2m(E−U (x))
dx .
i
∼ U (x) + (A.65)
Equation (A.65) is an approximate equation which is valid when the contribution

from the non-local potential is small compared with U (x). As these formulae show,
the equivalent effective local potential depends on the energy.
The energy dependence discussed here originates from non-locality. The potential
has in addition another energy dependence due to the polarization or the channel-
coupling effects such as the phonon excitation on the way of collision. The latter is
often called the intrinsic energy dependence in order to distinguish from the energy
dependence due to non-locality [1].
A.5.2.2 WKB Approximation II: Effective Mass
Let us next expand the translation operator exp[ i (x − x) p̂ψ ] and keep up to the
second order terms,

i i 1 1
exp (x − x) p̂ψ ≈ 1 + (x − x) p̂ψ − (x − x)2 p̂ψ2 . (A.66)
2! 2
Equation (A.60) is then approximated as
2 d2 ψ(x) (0) d
− 2
+ Ueff (x)ψ(x) + V (1) (x) ψ(x) = Eψ(x) , (A.67)
2m eff dx dx
where

1 1 1
= − 2 (x − x)2 V (x, x )dx , (A.68)
m eff m

(0)
Ueff (x) ≡ U (x) + V (x, x )dx , (A.69)

V (1) (x) ≡ (x − x)V (x, x )dx . (A.70)
In particular, suppose that the non-local potential is given by a separable type such
that [12]

x + x
V (x, x ) = F G(x − x) (A.71)
2

x + x 1 1 −[(x −x)/λ N L ]2
=F √ e (A.72)
2 π λN L
1 1 −[(x −x)/λ N L ]2
∼ F(x) √ e . (A.73)
π λN L
λ N L is a parameter to measure the degree of non-locality. The degree of non-locality

is larger for larger values of λ N L . Equation (A.73) leads to
1 1 1 λ2
= − 2 F(x) N L , (A.74)
m eff (x) m 2
(0)
Ueff (x) ≡ U (x) + F(x) , (A.75)
(1)
V (x) = 0 . (A.76)
Equation (A.74) indicates that Eq. (A.67) is valid when the degree of non-locality is
small, and that the mass parameter becomes position dependent due to the non-local
potential. In addition, Eq. (A.75) shows that the static part of the potential is also
modified by the non-local potential.
A.5.2.3 WKB Approximation III: Wigner Transformation
Here we derive an alternative expression of the equivalent local potential based on

the Wigner transformation. Let us first rewrite the non-local potential term by using
the Wigner transformation of the non-local potential as

V (x, x )ψ(x )dx

1 −i p(x −x)/ i(x −x) p̂V /2 i(x −x) p̂ψ /
= dx d pe e VW (x, p)e ψ(x) .
2π
(A.77)
p̂V is an operator which operates on x in VW (x, p). If we assume

ei(x −x) p̂V /2 VW (x, p) ≈ VW (x, p) , (A.78)
we obtain

V (x, x )ψ(x )dx = VW (x, p̂ψ )ψ(x) (A.79)
∼ VW (x, p(x))ψ(x) . (A.80)
The local momentum approximation (A.62), which is based on the WKB approxi-
mation, has been used in transforming from Eq. (A.79) to (A.80). Finally, we obtain
(W )
Ueff = U (x) + VW (x, p(x)) , (A.81)
for the equivalent effective local potential. The right-hand side of Eq. (A.81) agrees
with the equivalent potential in the phase space which appears in the classical theory
in the presence of a non-local potential [13].4
A.6 Tensor Algebra
A.6.1 Definition of Tensor Operators
The tensor property of a physical quantity or an operator represents their transfor-

mation property under a rotation of either the coordinate system or the object and is
a unified generalization of scalar, vector and tensor by introducing the index named
rank. The tensor property is one of the fundamental properties of each physical quan-
tity like the spatial and/or temporal dimension related to units, and plays an important
role, e.g., in evaluating transition matrices and in deriving selection rules.
4 Thanks to the large effective mass and hence the short wave length property, the classical theory
which uses the concept of classical trajectory is useful for heavy-ion collisions, and has been
successfully applied to analysing fusion reactions and deep inelastic collisions involving a huge
amount of energy dissipation from the relative motion to intrinsic motions and the associated
fluctuations [14, 15]. In that connection, it is an interesting question how to incorporate the exchange
effects of nucleons in the classical theory. The semi-classical treatment of a non-local potential has
been discussed also in [16].
Let us consider a translation before we discuss a rotation. In that case, there are two
alternative ways. In the one, one keeps the object as it is, and performs a translation
of the coordinate system from the old one K (x, y, z) to a new one K (x , y , z ).
In the other, one translates the object by keeping the coordinate system as it is. Here,
we take the latter. Let us consider the translation of an object by a. If we express the
wave function before and after the translation as ψ(x) and ψ (x), respectively, then
ψ (x) = ψ(x − a)
d 1 d2
= ψ(x) − a ψ(x) + ψ(x) + · · ·
dx 2! dx 2
d
= e−ia i dx / ψ(x) = e−ia p̂/ ψ(x) . (A.82)
Equation (A.82) shows that p̂ is the generator of the infinitesimal translation in space.
Equation (A.82) is expressed as
|ψ = T̂ (a)|ψ , (A.83)
−ia p̂/
T̂ (a) ≡ e , (A.84)
in the Dirac notation. T̂ (a) given by Eq. (A.84) is the unitary operator which causes
the translation by a.5 Also, if we define the operator Ô for an operator Ô repre-
senting some observable in such a way that the expectation value is conserved under
transformation, namely, by
ψ| Ô|ψ ≡ ψ | Ô |ψ , (A.85)
then by rewriting the right-hand side of Eq. (A.85) as ψ | Ô |ψ = ψ|T̂ † Ô T̂ |ψ,

we find that Ô is given by
Ô = T̂ (a) Ô T̂ † (a) = T̂ (a) Ô T̂ −1 (a) . (A.86)
Equation (A.86) shows that the operator is transformed by the same unitary operator
as that for the state vector ψ.
Exercise A.2 Show
x̂ ≡ T̂ (a)x̂ T̂ † (a) = x̂ − a , (A.87)

ψ |x̂|ψ = ψ|x̂|ψ + a , (A.88)
by using Eq. (A.84). Equation (A.88) agrees with that |ψ is the state, where the
object is translated by a.
is the operator to cause the translation by −a if one takes the point of view of fixing the object
5 It
and transforming the coordinate system.

Analogously, by considering infinitesimal rotations, one can show that the gen-
erators of the infinitesimal rotations are the angular momentum operators Jˆα (α =
x, y, z). For example, the rotation of an object through an angle α about the z-axis
ˆ
is given by the unitary operator R̂z = e−iα Jz . By extending this result, the unitary
operator for a rotation in the three-dimensional space is in general given either as
R̂(w) = e−iw·Ĵ (A.89)
by using the vector w, which specifies the direction and the magnitude of rotation,
or as
ˆ ˆ ˆ
R̂(φ, θ, ψ) = e−iφ Jz e−iθ Jy e−iψ Jz (A.90)
by using the Euler angles [17].

(i) First Definition of a Tensor Operator: When 2k + 1 operators T̂q(k) (q =
−k, −k + 1, . . . , k − 1, k) transform to each other following

k
(T̂q(k) ) = R̂ T̂q(k) R̂ −1 = T̂q(k) (k)
Rq q (φ, θ, ψ) (A.91)
q =−k
under a rotation, T̂ (k) is called an irreducible tensor of rank k, and T̂q(k) the q-
component. The rotation matrix Rq(k) q (φ, θ, ψ) is defined by
Rq(k)
q (φ, θ, ψ) ≡ kq | R̂(φ, θ, ψ)|kq . (A.92)
[N.B.] Equation (A.92) obeys [17]. It is the same as D function in Ref. [18]:
Rq(k) (k)
q (φ, θ, ψ) = (Dq q (φ, θ, ψ))BS . It is related to the D function in Ref. [19] as
Rq(k) (k) ∗
q (φ, θ, ψ) = (Dq q (φ, θ, ψ))BM . Since Ref. [20] transforms coordinate system
rather than the object, the sign of the arguments of the D function is inverted:
Rq(k) (k) (k)
q (φ, θ, ψ) = (Dq q (−φ, −θ, −ψ))RS . In this book, the definition of Dqq follows
Ref. [19].
(ii) Second Definition of a Tensor Operator: Since the generators of the infini-
tesimal rotations are the angular momentum operators, one can alternatively define
the irreducible tensor of rank k as the set of operators which satisfy the following
commutation relations:

Jˆz , T̂q(k) = q T̂q(k) , (A.93)

Jˆ± , T̂q(k) = (k ∓ q)(k ± q + 1)T̂q±1
(k)
. (A.94)
In order to show the equivalence between the two definitions, we show here that
Eqs. (A.93) and (A.94) can be derived from Eq. (A.91). Let us now assume that R̂
is the infinitesimal rotation operator (1 − iα Jˆλ ). In this case, from the definition
Eq. (A.92), we have
Rq(k) ˆ ˆ
q = kq |(1 − iα Jλ )|kq = δq q − iα kq | Jλ |kq . (A.95)
On the other hand,
R̂ T̂q(k) R̂ −1 = (1 − iα Jˆλ )T̂q(k) (1 + iα Jˆλ ) . (A.96)
Hence from Eq. (A.91)

(1 − iα Jˆλ )T̂q(k) (1 + iα Jˆλ ) = δq q − iα kq | Jˆλ |kq T̂q(k)
. (A.97)
Thus it follows
[ Jˆλ , T̂q(k) ] = kq | Jˆλ |kqT̂q(k)
. (A.98)
q
By putting here Jˆλ = Jˆz and Jˆλ = Jˆ± and noting

kq | Jˆz |kq = qδq q , k q ± 1| Jˆ± |kq = (k ∓ q)(k ± q + 1) , (A.99)
we obtain Eqs. (A.93) and (A.94).
A.6.2 Examples of Irreducible Tensor
(i) Spherical Harmonics: Ykq (θ, ϕ).

(ii) Tensor of Rank 1: Vector Operators A vector is an irreducible tensor of rank
1 and its 1, 0, −1 and x, y, z components are related as [18],
1
V±1 = ∓ √ (Vx ± iVy ) ,
2
V0 = Vz . (A.100)
Equation (A.100) can be understood from the fact that the x, y and z components of
the position vector r can be expressed as

3 1
r Y1±1 = ∓ √ (x ± iy) ,
4π 2

3
r Y10 = z, (A.101)
4π
by using the explicit expressions of the irreducible tensor of rank 1, Y1q .

A.6.3 Wigner–Eckart Theorem
The following formula holds6
1
α J M|Tq(k) |α J M = √ J k M q|J M α J T (k) α J , (A.102)
2J + 1
for the q component of a tensor of rank k, Tq(k) . The α and α are the quantum numbers
other than the angular momentum to specify states.
Equation (A.102) is called the Wigner–Eckart theorem, and shows that the matrix
element of a tensor operator can factor out the dependence on the z-component of the
angular momentum in terms of the geometrical factor given by the Clebsch–Gordan
coefficient. The factor α J T (k) α J is called the reduced matrix element. The
Wigner–Eckart theorem allows to automatically determine all the matrix elements
based on the Clebsch–Gordan coefficient if the reduced matrix element is once deter-
mined by calculating the matrix element for one combination of M , q, M for which
the Clebsch–Gordan coefficient is finite.
Especially when T is 1 or the angular momentum operators or the spherical
harmonics, the reduced matrix elements are given by
√
α J 1α J = δαα δ J J 2J + 1 , (A.103)

α J Jα J = δαα δ J J J (J + 1)(2J + 1) (A.104)

(k) − −k (2k + 1)(2 + 1)
Y = (−1) k 00|0
4π

(2k + 1)(2 + 1)(2 + 1) k

= (−1) , (A.105)
4π 00 0

k
where is a 3 j symbol.
00 0
A.6.4 Projection Theorem
Using the Wigner–Eckart theorem, one can show that the matrix element of any
vector operator A can be expressed as
1
α J M|A|α J M = α J M|J|α J M α J J |(A · J)|α J J . (A.106)
J (J + 1)
√
6 There exists an alternative definition, which does not explicitly take out the factor 1/ 2J + 1 on

the right-hand side, but includes it in the factor α J T α J .
Equation (A.106) is called the projection theorem or the Landé formula, and is use-
ful, e.g., in deriving the Schmidt values for the magnetic dipole moment given in
Sect. 5.4.1.
A.6.5 Relation Between the Scalar Product and the Tensor

Product of Rank 0
One can define a combined tensor operator of two tensor operators by taking an
appropriate sum of the product of their components using the Clebsch–Gordan
coefficients as follows

(K )
VQ(K ) ≡ T (k1 ) × U (k2 ) Q ≡ k1 k2 q1 q2 |K QTq(k1 1 ) Uq(k2 2 ) . (A.107)
q1
In the case of a combined tensor of rank 0, one can use the scalar product defined
by (k)
S ≡ (T (k) · U (k) ) ≡ (−1)q Tq(k) U−q (scalar product), (A.108)
q
for V0(0) . Especially, the scalar product of two vectors is given by
S = (V(1) · W(1) ) = Vx Wx + Vy W y + Vz Wz . (A.109)
Exercise A.3 Show that the scalar product and the tensor of rank 0 are related as
√
S = (−1)k 2k + 1 V0(0) (A.110)
when one combines two tensors of rank k.
A.7 Relation Between the Quadrupole Moment

and Intrinsic Quadrupole Moment
Here, we prove Eq. (4.47), which relates the experimentally observed quadrupole
moment for a deformed nucleus to the intrinsic quadrupole moment.
We first remark from the general transformation formula of a tensor, Eq. (A.91),
that the multipole moment operator Q̂ intr.
λμ given by the intrinsic coordinates and the
corresponding operator Q̂ lab.
λμ which is related to the experimental measurement and
is defined by the coordinates in the laboratory system are related as

λμ =
Q̂ intr. Dμλ∗ μ (ω) Q̂ lab.
λμ , λμ =
Q̂ lab. λ
Dμν (ω) Q̂ intr.
λν , (A.111)
μ ν
through the Euler angles ω = (φ, θ, ψ) which specify the orientation of the principal
axes. We used the properties of the D function in transforming from the first to the
second equations. Since the wave function for an axial-symmetrically deformed state
is given by Eq. (8.72), the matrix element of the operator is given in general by
2I1 + 1
1/2
2I2 + 1
1/2
lab.
Ψ I1 M1 K 1 Q̂ λμ Ψ I2 M2 K 2 =
ν
16π 2 (1 + δ K 1 0 ) 16π 2 (1 + δ K 2 0 )
λ I2
intr.
× D MI11 K 1 Dμν D
M2 K 2 Φ K 1 Q̂ λν Φ K 2
λ I2
intr.
+ (−1) I2 +K 2 D MI11 K 1 Dμν D
M2 −K 2 Φ K1 Q̂ λν Φ K̄ 2

+ (−1) I1 +K 1 D M11 −K 1 Dμν
I λ I
λν Φ K 2
D M22 K 2 Φ K̄ 1 Q̂ intr.
λ I2
intr.
+ (−1) I2 +K 2 +I1 +K 1 D MI11 −K 1 Dμν D
M2 −K 2 Φ K̄ 1 Q̂ λν Φ K̄ 2 . (A.112)
In the following, we consider the quadrupole moment. Since the quadrupole mo-
ment is defined by Ψ I M=I K | Q̂ lab.
λ=2,μ=0 |Ψ I M=I K , one can easily show that the second
and third terms on the right-hand side of Eq. (A.112) are zero for axial-symmetrically
deformed even–even nuclei7 and for odd nuclei in the strong coupling state with
I = K = 1/2 or K ≥ 3/2 by noting the properties of the D function, |ν| ≤ 2, the
addition rule of two angular momenta and the fact that K = 0, 2, 4 · · · for collective
motions of axial-symmetrically deformed even–even nuclei. In addition, the fourth
and first terms become equal. It therefore follows
2I + 1
Ψ I I K | Q̂ lab.
20 |Ψ I I K = 2
D M=I
I
K |D0ν |D M=I K Φ K | Q̂ λν |Φ K (A.113)
2 I intr.
ν
8π
= 2I 0I |I I 2I 0K |I K Φ K | Q̂ intr.
20 |Φ K . (A.114)
We then obtain Eq. (4.47) by using the explicit expression of the Clebsch–Gordan
coefficient.
A.8 Derivation of the Formula for the α-Decay Width

Based on the Gamow Model: Direct Method
Here, we derive the formulae (8.13)–(8.17) for the α-decay width. We first note that
in the Gamow model the wave function is given by
7 This excludes the case when K 1 = K 2 = 0. The second and third terms can be combined with the
first and/or fourth terms when K 1 = K 2 = 0.

φ I (r ) = N j0 (K r ) for r < R
φ(r ) = (A.115)
φ I I (r ) = (μ/k) [G 0 (η, kr ) + iF0 (η, kr )] /r for r ≥ R ,
1/2
where K and k = kα are given by Eq. (8.17). The continuity conditions at r = R

require that

μ
N sin KR = K [G 0 (η, k R) + iF0 (η, k R)]
k

μ
N cos KR = k G 0 (η, k R) + iF0 (η, k R) , (A.116)

k
where G 0 and F0 are the derivative of G 0 (η, ρ) and F0 (η, ρ) with respect to ρ, where
ρ = kr . These equations lead to
k G 0 (η, k R) + iF0 (η, k R)

tan KR = . (A.117)
K G 0 (η, k R) + iF0 (η, k R)
On the other hand, the decay width Γ is given by the outgoing flux per unit time at
the external turning point of the potential barrier in the case when the wave function
inside the potential is normalized to be unity,
Γ flux at the external turning point

≡ . (A.118)
probability to be inside the potential
Since Eq. (A.115) normalizes the external flux to be unity, we have

R re
= |φ I (r )| r dr +
2 2
|φ I I (r )|2 r 2 dr (A.119)
Γ 0 R

μ 1 G 20 (η, k R) + F02 (η, k R) sin 2KR
= 2 kR 1−
k 2 sin2 KR 2KR
2η
+ (G 20 (η, ρ) + F02 (η, ρ))dρ . (A.120)
kR
One can determine the energy E, i.e., the Q-value for the α-decay, and the decay
width Γ by using Eqs. (A.117) and (A.120) if the depth of the potential V0 and the
range of the nuclear force R are known in advance. Conversely, one can determine
the depth of the potential V0 and the range of the nuclear force R for the α-decay from
the Q-value and the decay width using Eqs. (A.117) and (A.120) if they are known
experimentally. However, as discussed in Sect. 8.1.1, there remains an ambiguity of
potential because of the presence of Pauli forbidden states, in other words, redundant
states. In order to remove the ambiguity, one needs to specify the number of nodes in
the wave function for the relative motion. As one example of such analyses, Table A.1
+ ) → 206 Pb(0+ ) + α [21]

Table A.1 The potential for the α-decay 210 Po(0g.s. g.s.
Number of nodes R (fm) V0 (MeV) KR/π
13 7.91 150 13.62
12 7.91 128 12.63
11 7.92 108 11.64
10 7.93 89.4 10.65
9 7.94 72.5 9.660
8 7.95 57.2 8.675
7 7.97 43.6 7.693
6 7.99 31.7 6.714
5 8.03 21.5 5.739
4 8.08 13.0 4.768
3 8.15 6.16 3.804
2 8.25 0.95 2.846
1 8.43 −2.65 1.894
0 8.77 −4.69 0.9461
shows the results of the analysis to determine the potential for the α-decay from the
ground state of 210 Po to the ground state of 206 Pb given by Eq. (8.1) by using the
experimental data for the Q-value and the decay width. As the Table shows, there
exist a wide range of potentials from shallow to deep ones depending on the number
of nodes in the radial wave function of the relative motion. The number of nodes is
11 for the α-decay from the ground state of 210 Po to the ground state of 206 Pb if we
count the quantum number in the harmonic oscillator shell model.
The last column of Table A.1 gives the magnitude of the action integral inside the
potential area in units of π . When the number of node n is large, including the case
of n = 11, the value of the action integral inside the potential region becomes

1
KR ≈ n + π, (A.121)
2
and well agrees with the quantization condition (8.5) derived based on the semi-
classical theory.
In the following, suppose that the Q-value of the α-decay is small, so that 2η
ρ = k R. In that case, the following asymptotic formulae hold for the Coulomb wave
functions F0 (η, k R), F0 (η, k R), G 0 (η, k R) and G 0 (η, k R) [22]:
1 α 1
F0 ∼ βe ; F0 ∼ β −1 eα ; G 0 ∼ βe−α ; G 0 ∼ −β −1 e−α , (A.122)
2 2
α = 2 2ηρ − π η , β = (ρ/2η)1/4 . (A.123)
Let us now relate the decay width to the tunneling probability by using Eq. (A.121).
We set sin 2KR to 0 based on Eq. (A.121). Equations (A.122) and (A.123) suggest
|F0 (η, k R)| |G 0 (η, k R)|, so that we can ignore F0 in Eq. (A.120). Also, we ignore
the integration term as well. The decay width is then approximately given by
22 k 1
Γ ∼ . (A.124)
μR G 0 (η, k R)
2
On the other hand, the tunneling probability t is given by the ratio of the outgoing
flux JI(+) (+)
I to the flux J I which enters the potential barrier at r = R in the outward
direction from inside. Since JI(+) and JI(+)
I are given by

1 K N 2 2 K N 2

JI(+) = j I(+) dS
= r dΩ = 4π 2K , (A.125)
r 2 μ 2K μ

1 2
JI(+)
I = j I(+)
I dS = r dΩ = 4π , (A.126)
r2
the tunneling probability becomes
(D) JI(+) 4k sin2 KR 4k 1

tGM ≡ I
∼ ∼ . (A.127)
JI(+) K G 20 (η, k R) K G 20 (η, k R)
Equations (A.124) and (A.127) lead to Eq. (8.13).

Thus one can obtain Eqs. (8.13)–(8.17) by using the asymptotic forms of the
Coulomb wave functions given by Eqs. (A.122) and (A.123) [22], which hold when
k R 2η = kre , i.e., when R re or when the Q-value is small.
A.9 Basics of Electromagnetic Transitions
A.9.1 Hamiltonian of the Total System
To discuss the electromagnetic transitions of a nucleus, it is necessary to treat the

combined system consisting of the nucleus, i.e., a many-body system of nucleons,
and the electromagnetic fields. The Hamiltonian of the total system is given by
Ĥtot = Ĥnucl + Ĥfield + Ĥint . (A.128)
Ĥnucl is the Hamiltonian of an isolated nucleus, and consists of the kinetic energy of
nucleons and the strong interaction among nucleons. Alternatively, it is expressed
in an appropriate model such as the shell model or the collective model. We assume
that the eigenstates and the corresponding eigenvalues are known and express as
Ĥnucl Ψα = E α Ψα . (A.129)
The Hamiltonian of the electromagnetic fields Ĥfield and the interaction Hamiltonian
between the nucleus and the electromagnetic fields Ĥint are given by

1 2
Ĥfield = E (r, t) + B2 (r, t) dr , (A.130)

8π

1
Ĥint =− jμ Aμ dr (A.131)
c

1
= ρ(r, t)Φ(r, t) − j(r, t)A(r, t) dr , (A.132)
c
where E, B, ρ and j are the electric field, the magnetic field, the charge and current
densities of the nucleus, respectively.
From the point of view of the original many-body system of nucleons, ρ is given
by
A
1
ρ(r, t) = e − t3(i) δ(r − ri (t)) , (A.133)
i=1
2
while j by
j(r, t) = c∇ × μ(r, t) (A.134)
using the magnetic moment μ(r, t). Since the magnetic moment originates from the
orbital motion of protons, and also from the spins of protons and neutrons, j is given
by a sum of two parts. If we denote j originating from the orbital motion of protons
by jc , then
A
1 (i) 1
jc = e − t3 [vi δ(r − ri (t)) + h.c.] . (A.135)
i=1
2 2

vi is given by vi = i Ĥnucl , ri , and becomes vi = pi /M N if the potential in Ĥnucl
does not depend on the momentum. On the other hand, if we denote the magnetic
moment originating from the spin of nucleons by μs , then it is given by

A
e 1 1
μs (r, t) = δ(r − ri (t)) − t3(i) g p + + t3(i) gn si (A.136)
i=1
2M N c 2 2
using the g-factors of protons and neutrons: gp = 2.792847 × 2 ≈ 5.586, gn =

−1.9130427 × 2 ≈ −3.826.
A.9.2 Wave Function of Photon: Vector Spherical

Harmonics
A.9.2.1 Coulomb Gauge
It is convenient to adopt the Coulomb gauge, known also as the transverse gauge or
the radiation gauge [23]:
∇·A=0 (A.137)
in order to discuss the electromagnetic transitions. In this case, the Hamiltonian of

the radiation field is given by

1 1 2
Ĥfield = Ȧ + (∇ × A) 2
dr , (A.138)
8π c2
and the vector potential A obeys the wave equation:
1 ∂2
ΔA − A=0. (A.139)
c2 ∂t 2
If we put the time dependence as e−ickt , then we obtain the Helmholtz equation
(Δ + k 2 )A = 0 . (A.140)
Note that Eq. (A.137) indicates that the vector potential A is perpendicular to the di-
rection of the propagation of the wave (k ⊥ A), i.e., that the wave is a transverse wave.
A.9.2.2 Rotational Transformation of Vector Wave Function

and the Spin and Intrinsic Wave Function of Photon
When one uses the spherical polar coordinates and makes the partial wave expansion,
the solution of the Helmholtz equation (A.140) for the partial wave is nothing but
j (kr ) provided the field is a scalar field. In reality, photons are described by a vector
field, so that they have spin 1 as we show later. It is therefore necessary to extend the
2 2 2
wave function so as to be a simultaneous eigenfunction of L̂ , Ŝ , Ĵ and Jˆz .
(i) The Transformation Property of a Vector Function: Let us first study how a
vector function V(r, θ, ϕ) is transformed under rotation. One needs to consider both
the transformation of vector components and that of coordinates under rotation. For
example, when one rotates the coordinate system through α about the z-axis, the
vector components in the new coordinate system are given by
⎛ ⎞ ⎛ ⎞⎛ ⎞
Vx (r, θ, ϕ) cos α sin α 0 Vx (r, θ, ϕ + α)
⎝Vy (r, θ, ϕ)⎠ = ⎝− sin α cos α 0⎠ ⎝Vy (r, θ, ϕ + α)⎠ . (A.141)
Vz (r, θ, ϕ) 0 0 1 Vz (r, θ, ϕ + α)
For small α, Eq. (A.141) can be represented as
V = (1 + iα Ĵz )V + O(α 2 ) , (A.142)
where Jˆz is given by
∂
Jˆz = −i + iez × (A.143)
∂ϕ

∂ ∂
= −i x −y + iez × (A.144)
∂y ∂x
= L̂ z + iez × . (A.145)
ez × means to take the outer product between the unit vector in the z-axis ez and the
vector V which appears on the right side in Eq. (A.142).
Exercise A.4 Prove Eq. (A.143) by noting that Eq. (A.141) leads to
∂
Vx (r, θ, ϕ) = Vx (r, θ, ϕ) + α Vx + αVy , (A.146)
∂ϕ
∂
Vz (r, θ, ϕ) = Vz (r, θ, ϕ) + α Vz , (A.147)
∂ϕ
and that
ez × V = e y Vx − e x Vy . (A.148)
(ii) Spin of a Photon: Equation (A.145) shows that the intrinsic spin operator of a
photon is given by
Ŝx = ie x × , Ŝ y = ie y × , Ŝz = iez × . (A.149)
Also, the total angular momentum is given by
Ĵ = L̂ + Ŝ . (A.150)
Exercise A.5 Show that Ŝ defined by Eq. (A.149) satisfies the commutation relations
of the angular momentum operator.
(iii) The Eigenvectors of the Intrinsic Spin of Photon: If we define
1
e+1 = − √ (e x + ie y ) ,
2
e0 = e z ,
1
e−1 = √ (e x − ie y ) , (A.151)
2
in terms of the unit vectors in the x, y, z direction by following the general rule that
relates the 1, 0, −1 components of a tensor of rank 1 to the x, y, z components of a
vector, we can show that
2
Ŝ eq = 2eq , Ŝz eq = qeq (q = ±1, 0) . (A.152)
Equation (A.152) shows that the magnitude of the intrinsic spin of photon is 1 and
that the eigenvector corresponding to Sz = q is eq .
A.9.2.3 Vector Spherical Harmonics
(i) Definition: From the above arguments, it follows that the wave function of photon,
2 2 2
which is the simultaneous eigenfunction of L̂ , Ŝ , Ĵ and Jˆz , is given by

Y J M (θ, ϕ) = 1 mq|J MYm (θ, ϕ)eq (A.153)
mq
by using the addition rule of two angular momenta. Y J M (θ, ϕ) are called vector
spherical harmonics.
One can easily show that the vector spherical harmonics have the following prop-
erties,
2
Ĵ Y J M (θ, ϕ) = J (J + 1)Y J M (θ, ϕ) , (A.154)
Jˆz Y J M (θ, ϕ) = MY J M (θ, ϕ) , (A.155)
2
L̂ Y J M (θ, ϕ) = ( + 1)Y J M (θ, ϕ) , (A.156)
2
Ŝ Y J M (θ, ϕ) = 2Y J M (θ, ϕ) , (A.157)
and the orthogonality,

dΩY∗J M (θ, ϕ) · Y J M (θ, ϕ) = δ J J δ δ M M . (A.158)
Note that there exist three kinds of vector spherical harmonics Y J J M (θ, ϕ) and
Y J J ±1M (θ, ϕ), which have different values of the orbital angular momentum for each
set of J, M. They can be classified into the following two groups depending on the
difference of parity:
• Y J J M (θ, ϕ): Parity (−1) J
This corresponds to the magnetic field for the electric multipole radiation and to
the electric field for the magnetic multipole radiation.
• Y J J ±1M (θ, ϕ): Parity (−1) J +1

This corresponds to the electric field for the electric multipole radiation and to the
magnetic field for the magnetic multipole radiation.
(ii) The Method to Construct from the Ordinary Scalar Spherical Harmonics:
The vector spherical harmonics can be constructed from the ordinary scalar spherical
harmonics by the following operations:

Ym (θ, ϕ) = ( + 1)Ym (θ, ϕ) , (A.159)
1/2
r + 1 1/2
Ym = − Y+1m (θ, ϕ) + Y−1m (θ, ϕ) , (A.160)
r 2 + 1 2 + 1
(iii) Gradient Formulae:

+ 1 1/2 d
∇Φ(r )Ym = − − Φ(r )Y+1m (θ, ϕ)
2 + 1 dr r
1/2
d +1
+ + Φ(r )Y−1m (θ, ϕ) , (A.161)
2 + 1 dr r
1/2
d +2
∇ × (Φ(r )Y+1m ) = i + Φ(r )Ym (θ, ϕ) . (A.162)
2 + 1 dr r
A.9.2.4 Eigenfunctions of Photon
(i) General form: As described in Sect. A.9.2.3, there exist three independent solu-
tions of the Helmholtz equation which differ in the orbital angular momentum when
the total angular momentum of photon is given, which we denote by I . They have to
satisfy in addition the transversality condition ∇ · A = 0 in order to be the wave func-
tion of photon. Equation (A.159) shows that j I (kr )Y I I M (θ, ϕ) satisfies this condi-
tion. The other wave functions, which are orthogonal to this wave function, are given
by taking linear combinations of j I +1 (kr )Y I I +1M (θ, ϕ) and j I −1 (kr )Y I I −1M (θ, ϕ).
Among them, the unique wave function which fulfills the transversality condition is

i I I +1
∇ × j I (kr )Y I I m = j I +1 (kr )Y I I +1m − j I −1 (kr )Y I I −1m .
k 2I + 1 2I + 1
(A.163)
We distinguish them with the names of the magnetic and electric radiations, and
represent as
A Mk I M (r) = N j I (kr )Y I I M (θ, ϕ)

N 1
=√ (r × ∇) j I (kr )Y I M (θ, ϕ) , (A.164)
I (I + 1) i
i
A Ek I M (r) = (∇ × A Mk I M (r))
k
N 1 ∂
= −√ ∇ Y I M (θ, ϕ) r j I (kr ) + k r j I (kr )Y I M (θ, ϕ) ,
2
I (I + 1) k ∂r
(A.165)
by using the indices M and E, respectively. N is the normalization constant.

(ii) Quantization of the Wave Number: The wave number is quantized by con-
sidering a perfectly conducting sphere of radius R,8 and by imposing the following
boundary conditions on the surface [23],
E = 0 , B⊥ = 0 . (A.166)
Since E = − 1c ∂A
∂t
, B = ∇ × A in the Coulomb gauge, the magnetic and electric
radiations are quantized respectively as

j I (kn R) = 0 magnetic radiation,
(A.167)
∂
[r j I (kn r )] r =R = 0 electric radiation,
∂r
based on Eqs. (A.164) and (A.165).

(iii) Normalization: The normalization constant is determined so that

R
2π c
2
r dr dΩA∗λ k I M (r) · Aλk I M (r) = δλλ δkk δ I I δ M M , (A.168)
0 k
where λ = M, E. We obtain
4π ck
N = (A.169)
R
by considering the asymptotic form of j (kr ) and using the boundary conditions
(A.167) in the limit of R → ∞.
A.9.3 Multipole Expansion and Quantization
The most general solution of the Helmholtz equation which satisfies the condition
of a transverse wave is given by

Â(r, t) = Aλk I M (r)e−ickt âλk
†
IM + A∗
λk I M (r)e ickt
â λk I M . (A.170)
λk I M
8R is introduced for the sake of quantization. It is taken to be much larger than the nuclear radius,
and is eventually taken to be the limit of R → ∞.
The coefficients â † and â are independent variables to describe the electromagnetic

field, and are assumed to obey the following commutation relations of the Bose–
Einstein statistics,

âλk I M , âλ k I M = 0 , âλk I M , âλ† k I M = δλλ δkk δ I I δ M M . (A.171)
The Hamiltonian of the radiation field and that of the electromagnetic interaction
are given by

† 1
Ĥfield = kc âλk I M âλk I M + , (A.172)
λk I M
2

1 † −ikct
Ĥint =− âλk IM dr j(r, t) · Aλk I M (r)e + h.c. , (A.173)
c λk I M
in terms of â † and â. The eigenstate of the radiation field is given as
| · · · n λk I M · · · (A.174)
in the occupation number representation. Since it holds that
| · · · n λk I M − 1 · · · |âλk I M | · · · n λk I M · · · |2 = n λk I M , (A.175)
†
| · · · n λk I M + 1 · · · |âλk I M | · · · n λk I M · · · | = n λk I M + 1 ,
2
(A.176)
the creation and annihilation of a photon take place via â † and â in Eq. (A.173). On
the other hand, the nucleus changes its state through j(r, t).
A.10 Notation for Relativistic Equation of Motion

and Dirac Equation
Here, we summarize the notations of the Dirac equation used in this book and their
basic functions:
x μ = (ct, x) , xμ = (ct, −x) , (A.177)

p μ = (E/c, p) , pμ = (E/c, −p) , (A.178)

μ ∂ ∂ ∂ ∂
∂ ≡ = , −∇ , ∂μ ≡ μ = , ∇ , (A.179)
∂ xμ c∂t ∂x c∂t
⎛ ⎞
1 0 0 0
⎜ 0 −1 0 0 ⎟
g μν = gμν = ⎜⎝0 0 −1 0 ⎠ ,
⎟ (A.180)
0 0 0 −1
a · b = aμ bμ = a μ gμν bν = a 0 b0 − a · b , (A.181)
∂ 2
∂ μ ∂μ = − ∇2 , (A.182)
c2 ∂t 2

0 σ 1 0
α= , β= , (A.183)
σ 0 0 −1

0 σ
γ0 = β , γ = βα = , γ μ = (γ 0 , γ ) , (A.184)
−σ 0
ψ̄ = ψ † γ 0 , (A.185)

∂ ∂ ∂ ∂ ∂
p̂ μ = i = i , i , i , i (A.186)
∂ xμ ∂(ct) ∂ x1 ∂ x2 ∂ x3

∂ ∂ ∂ ∂
= i∇ μ = i , −i , −i , −i (A.187)
∂(ct) ∂x ∂y ∂z

∂
= i , −∇ . (A.188)
∂(ct)
A.11 Derivation of the Pairing Energy
In this section, we outline the derivation of Eqs. (5.67)–(5.69) for the pairing energy
due to the residual interaction of the δ-function type given by Eq. (5.66).
We first remark that the antisymmetrized wave function of two-nucleon system is
given by Eq. (5.63). We assume that each wave function |( jα jβ )J M is normalized.
The normalization constant N is then given by
√
1/ 2 for (n α α jα ) = (n β β jβ )
N= (A.189)
1/2 for (n α α jα ) = (n β β jβ ) .
Since the residual interaction is independent of spin, it is convenient to transform

the wave function from the j j-coupling scheme to the L S-coupling scheme. It can
be achieved by means of a recoupling coefficient as

|( jα jβ )J M = (α 1/2) jα , (β 1/2) jβ ; J M (A.190)

= (α β )L , (1/2 1/2)S; J M
LS
!
× (α β )L , (1/2 1/2)S; J (α 1/2) jα , (β 1/2) jβ ; J , (A.191)
where the recoupling coefficient is given by [24]

!
( j1 j2 )J12 , ( j3 j4 )J34 ; J ( j1 j3 )J13 , ( j2 j4 )J24 ; J
⎧ ⎫
⎨ j1 j2 J12 ⎬
= [(2J12 + 1)(2J34 + 1)(2J13 + 1)(2J24 + 1)]1/2 j3 j4 J34 . (A.192)
⎩ ⎭
J13 J24 J
Similarly,

|( jβ jα )J M = (β 1/2) jβ , (α 1/2) jα ; J M (A.193)

= (β α )L , (1/2 1/2)S; J M
LS
!
× (β α )L , (1/2 1/2)S; J (β 1/2) jβ , (α 1/2) jα ; J . (A.194)
The recoupling coefficients in Eqs. (A.191) and (A.194) can be related to each
other as
(β α )L , (1/2 1/2)S; J |(β 1/2) jβ , (α 1/2) jα ; J

⎧ ⎫
⎨ β α L ⎬
= [(2L + 1)(2S + 1)(2 jβ + 1)(2 jα + 1)]1/2 1/2 1/2 S (A.195)
⎩ ⎭
jβ jα J
!
= (−1)σ1 (α β )L , (1/2 1/2)S; J (α 1/2) jα , (β 1/2) jβ ; J , (A.196)
where
σ1 = 1 + α + β + jα + jβ + L + S + J (A.197)
by using the following exchange property of the 9 j symbol,

⎧ ⎫ ⎧ ⎫
⎨ j1 j2 J12 ⎬ ⎨ j2 j1 J12 ⎬
j3 j4 J34 = (−1)Σ j4 j3 J34 , (A.198)
⎩ ⎭ ⎩ ⎭
J13 J24 J J24 J13 J
where
Σ = j1 + j2 + j3 + j4 + J12 + J34 + J13 + J24 + J . (A.199)
The antisymmetrized wave function is now given by

!
|( jα jβ )J Ma = N (α β )L , (1/2 1/2)S; J (α 1/2) jα , (β 1/2) jβ ; J
LS

× (α β )L , (1/2 1/2)S; J M + (−1)σ (β α )L , (1/2 1/2)S; J M ,

(A.200)
where
σ = α + β + L + S . (A.201)
The energy shift due to a residual interaction V is given by
ΔE((n α α jα , n β β jβ )J ) = a ( jα jβ )J |V |( jα jβ )J a (A.202)
in the first order perturbation theory. Inserting Eq. (A.200), we have

ΔE (n α α jα , n β β jβ )J
!
(α β )L , (1/2 1/2)S; J (α 1/2) jα , (β 1/2) jβ ; J
2
= 2N 2
LS
!
× (α β )L , (1/2 1/2)S; J |V |(α β )L , (1/2 1/2)S; J
!
+ (−1)σ (α β )L , (1/2 1/2)S; J |V |(β α )L , (1/2 1/2)S; J . (A.203)
We now remark that

1
δ(r1 − r2 ) = δ(r1 − r2 )δ(cos θ1 − cos θ2 )δ(ϕ1 − ϕ2 ) (A.204)
r1 r2
for the residual interaction of the δ-function type, and that the two-body wave function
is given by
|(α β )L , (1/2 1/2)S; J

= L S M L M S |J M α β m α m β |L M L |S M S
ML MS m α m β
1 1
× u n (r1 ) u n β β (r2 )Yα m α (Ω1 )Yβ m β (Ω2 ) . (A.205)
r1 α α r2
It is then straightforward to obtain

!
(α β )L , (1/2 1/2)S ; J M δ(r1 − r2 )(α β )L , (1/2 1/2)S; J

(2α + 1)(2β + 1) α β L 2 1
2
= δ SS δ L L u n α α (r )u n β β (r ) dr ,
4π 0 0 0 r 2
(A.206)
and
!
(α β )L , (1/2 1/2)S ; J M δ(r1 − r2 )(β α )L , (1/2 1/2)S; J
2
α +β +L (2α + 1)(2β + 1) α β L 1
=δ SS δ L L (−1) [u n (r )u n β β (r )]2 dr .
4π 0 0 0 r2 α α
(A.207)
Equations (A.203), (A.206), and (A.207) lead to

!
(α β )L , (1/2 1/2)S; J (α 1/2) jα , (β 1/2) jβ ; J
2
= −V0 2N 2
LS

(2α + 1)(2β + 1) α β L 2 1
2
× [1 + (−1) S ] u n α α (r )u n β β (r ) dr .
4π 0 0 0 r 2
(A.208)
The factor [1 + (−1) S ] shows that only S = 0 state contributes. This is a consequence
of the δ-force for the two-body system of identical nucleons. The isospin wave
function for the system of identical nucleons is symmetric. On the other hand, the
δ-force has a finite effect only in the positive parity states, i.e., spatially symmetric
states. Consequently, the δ-force affects only the spin singlet state. Equation (A.208)
now becomes

!
= −V0 4N 2 (α β )J, (1/2 1/2)0; J (α 1/2) jα , (β 1/2) jβ ; J
2

(2α + 1)(2β + 1) α β J 2 1
2
× u n α α (r )u n β β (r ) dr . (A.209)
4π 0 0 0 r 2
One has to use various properties of the 3 j, 6 j and 9 j symbols, which can be found
in [17, 24], in order to reach the final result given by Eqs. (5.67)–(5.69). The relevant
formulae in the present context are
⎧ ⎫
⎨ j1 j2 J ⎬ ( )
1 j j J
j1 j2 J = √ (−1)− j2 −J − j1 −k 1 2 (A.210)
⎩ ⎭ (2J + 1)(2k + 1) j2 j1 k
k k 0
and
( )
j1 j2 J J j1 j2 J
= − (21 + 1)(22 + 1) 1 2 . (A.211)
1/2 −1/2 0 0 0 0 2 1 1/2
Using these formulae, one can rewrite Eq. (A.209) into Eqs. (5.67)–(5.69).
A.12 Errata to the Japanese Edition
In Table A.2, we list the errata for the original Japanese edition.
Table A.2 Errata for the Japanese edition

Page Place Original Correction
8 Sect. 1.5, 5th line 92 roughly speaking 92
20 Eq. (2.24) 0.14 fm−3 0.14–0.17 fm−3
21 Eq. (2.28) 1.18∼1.2 fm 1.1–1.2 fm
(n−1)(n+2) 2 (n−1)(n+2)
2(2n+1) αn 2(2n+1) αn
45 Eq. (2.67) 2
n=4,5...
5(n−1) 2 5(n−1) 2
45 Eq. (2.68) α
(2n+1)2 n n=4,5... (2n+1)2 αn
72 Footnote amnd and
72 Footnote potetial potential
79 4th line (3.87) (3.86)
g2 2
4π gπ
79 Eq. (3.95) 4π
3μ2
μc2 4ππ c m c2
μ3 3 c π
83 Eq. (3.103) f gπ
96 Below Eq. (4.38) electromagnetic electromagnetic multipole moments
moment
104 Fig. 5.1 occupation number degeneracy
104 Fig. 5.1 connection lines from N = 5 and 6, respectively
between H.O. and ISQ
for 1h, 1i
120 Eq. (5.68) Rn α α , Rn β β u n α α , u n β β
120 Below Eq. (5.69) Rn (r ) u n (r )
131 Below Eq. (6.32) Eq. (6.32) Eq. (6.31)
134 2nd line 0.14 fm−3 0.17 fm−3
2 1
163 Eq. (7.18) 9 9
172 Fig. 7.12 superdeformed rotors superdeformed rotational bands
174 1st line 13.7 billion years 13.8 billion years
176 Fig. 8.1 positions of r2 , r3 turning point position for each
√ √
177 (8.11), below (8.12) e2 2ηk R e4 2ηk R
√ √
178 Eqs. (8.15), (8.20) e2 2ηk R e4 2ηk R
3 2 1 pin 3 2 1 pin
181 Eq. (8.30) 2 μ rc 2 μ rc 2
199 2nd line 60 100
205 Eq. (10.31) k k
√ √
222 Eq. (10.123) α= 2ηρ − πη α = 2 2ηρ − πη
√ √
222 Below Eq. (10.127) α= 2ηρ − πη α = 2 2ηρ − πη
222 Below Eq. (10.127) (10.123) (10.122), (10.123)
228 Eq. (10.167) r j I (kn R) r j I (kn r )
228 Eq. (10.168) A∗λk I M (r) Aλk I M (r)
References
1. G.R. Satchler, Direct Nuclear Reactions (Clarendon Press, Oxford, 1983)

2. K.W. Ford, J.A. Wheeler, Ann. Phys. 7:259 (1959); 287 (1959)
3. W.E. Frahn, Ann. Phys. 72, 524 (1972); W.E. Frahn, M.S. Hussein, Nucl. Phys. A 346, 237
(1980)
4. K.W. McVoy, Phys. Rev. C 3, 1104 (1971)
5. N. Rowley, H.C. Doubre, C. Marty, Phys. Lett. B 69, 147 (1977)
6. N. Takigawa, S.Y. Lee, Nucl. Phys. A 292, 173 (1977)
7. M. Ueda, M.P. Pato, M.S. Hussein, N. Takigawa, Phys. Rev. Lett. 81, 1809 (1998)
8. D.M. Brink, Semi-Classical Methods for Nucleus–Nucleus Scattering (Cambridge University
Press, London, 1985)
9. D.M. Brink, N. Takigawa, Nucl. Phys. A 279, 159 (1977)
10. S.Y. Lee, N. Takigawa, C. Marty, Nucl. Phys. A 308, 161 (1978)
11. J.J. Sakurai, Modern Quantum Mechanics (Addison-Wesley, New York, 1985)
12. W.E. Frahn, R.H. Lemmer, Nuovo Cimento 5, 1564 (1957); 6, 664 (1957); F.G. Perey, B. Buck,
Nucl. Phys. 32, 353 (1962)
13. N. Takigawa, K. Hara, Z. Phys. A 276, 79 (1976)
14. D.H.E. Gross, H. Kalinowski, Phys. Lett. B 48, 302 (1974); Phys. Rep. 45, 175 (1978)
15. P. Fröbrich, R. Lipperheide, Theory of Nuclear Reactions (Clarendon Press, Oxford, 1996)
16. H. Horiuchi, Prog. Theor. Phys. 63, 725 (1980); 64, 184 (1980)
17. A. Messiah, Quantum Mechanics, vol. II (Elsevier, Amsterdam, 1962)
18. D.M. Brink, G.R. Satchler, Angular Momentum (Oxford University Press, Oxford, 1962)
19. A. Bohr, B.R. Mottelson, Nucl. Struct. vol. I (Benjamin, New York, 1969)
21. DSc thesis of Yoshihiro Nozawa, Tohoku University (2002)
22. M. Abramowitz, I.A. Stegun, Handbook of Mathematical Functions with Formulas, Graphs,
and Mathematical Tables (Dover, London, 1978)
23. J.D. Jackson, Classical Electrodynamics, 3rd edn. (Wiley, New York, 1998)
24. A. de-Shalit, I. Talmi, Nuclear Shell Theory (Academic, New York, 1963)
Index
Symbols 14 C, 71, 191, 200

1/v-law, 59, 218 14 N, 33, 71, 119
Δ(1232), 1, 3 15 N, 25, 71
particle, 3, 9 14 O, 71
Σ particle, 3, 9 15 O, 25, 71
particle, 9 16 O, 1, 33, 38, 108, 117, 173
α particle, 15, 38 20 Ne, 173
α-decay, 191, 243 24 Mg, 38, 173
α-nucleus, 33 48 Ca, 130
(α,p) reaction, 220 45 Sc, 2, 119
α-particle model, 38, 179 56 Fe, 40, 215
β-decay, 6, 45, 58, 219 62 Ni, 33, 40, 215
β-stability line, 22 Ge, 173

γ process, 220 90 Sr, 60
(γ , α) reaction, 220 Tc, 33

(γ ,n) reaction, 219 Cd, 209
(γ ,p) reaction, 220 112 Cd, 172
μ particle, 3 105 Te, 220
ω-meson, 85, 88, 125 137 Cs, 60
π -meson, 3, 85 150 Ce, 44
virtual, 1, 5 Pm, 33
ρ-meson, 3, 85, 87, 125 150 Sm, 44
σ -particle, 88, 125 166 Er, 171
σ ωρ model, 155 167 Er, 171, 182

3 H, 120, 122 180 Ta, 33
He, 215 189 Os, 120

3 He, 89, 120, 122, 142 Au, 15, 20
4 He, 108 Hg, 173, 180
Li, 215 182 Hg, 173
6 Li, 33 184 Hg, 173, 185
6,7 Li, 23 186 Hg, 173
8,9 Li, 23 208 Hg, 44
11 Li, 22, 23 206 Pb, 191
8 Be, 33, 63, 173, 179, 187, 197 208 Pb, 22, 44, 117
10 B, 33 209 Pb, 117
12 C, 33, 38, 173, 180 210 Pb, 127, 164, 168, 171
13 C, 120 209 Bi, 33, 118, 119

DOI 10.1007/978-4-431-55378-6
262 Index
210 Po, 15, 122, 127, 171, 191, 196, 197, 245 Binding energy, 32, 110
Rn, 208 BNL, 11
226 Ra, 201
Bohr magneton, 4
230 Th, 61
Bohr radius, 30
231 Th, 61
Bohr, N., 46, 117
U, 208, 215 Boltzmann constant, 217
235 U, 46, 49, 55, 57, 58
Bonn potential, 88
236 U, 55
Born approximation, 19, 82, 233
238 U, 49, 55, 57, 63, 211
Breathing mode of vibration, 146
239 U, 54, 55
Brink–Boeker force, 95
237 Np, 60, 192
Broken spatial symmetry, 39
239 Pu, 46
Brownian motion, 202
240 Pu, 63
Brueckner theory, 93
Cm, 130
Cf, 130
258 Fm, 201
C
Nh (Nihonium), 131 Capture reaction, 218
Fl (Flerovium), 132 Caustics, 225
Mc (Moscovium), 132 Central potential, 72
Lv (Livermorium), 132 CERN, 11
Ts (Tennessine), 132 Chain reaction, 57, 58
Og (Oganesson), 132
Channel-coupling effects, 235
Channel radius, 198
Charge distribution, 22
A
Charge-exchange operator, 70
Absorber, 16
Abundance of elements, 9 Charge independence, 71, 80
Actinides, 173, 187 breaking, 88
Action integral, 230 Charge symmetry, 71
Airy function, 231 breaking, 89
Alkali metals, 117 Chiral perturbation theory, 88
Allowed states, 197 Classical liquid, 23, 212
Alvarez, L., 36 Classical optics, 16
Anomalous magnetic moments, 4, 5, 99, Classical orbits, 14
122, 123 Classical trajectories, 15
Argonne potential, 88 Classical turning points, 54
Arima, A., 122 Clebsch–Gordan coefficient, 105, 127
Arrhenius factor, 201 Closed form expression, 224
Astrophysical S-factor, 216 Cluster decay, 191
Asymmetric fission, 46, 108 Clusterization, 199
Atomic model, 15 Cluster model, 38, 179, 197
Atomic number, 1 Cluster structure, 38, 95, 126
Attenuation method, 26 CNO cycle, 34
Coester line, 156
Coherence length, 169
B Cold fusion, 130
Barrier transmission probability, 231 Collective model, 98
Bartlett operator, 70 Collective motion, 212
Baryons, 1 Commutation relation, 30
BCS, 164 Comparison equation, 229
Bernoulli’s formula, 149 Complete fusion reaction, 202
Bethe–Goldstone equation, 93 Complex scaling method, 192
Big Bang nucleosynthesis, 215 Compound nucleus reactions, 107, 117
Bimodal fission, 201 Compton wavelength, 66, 87
Index 263
Configuration mixing, 120, 121 Double folding potential, 233

Constant Δ method, 185 Doubly-closed shell nucleus, 117
Constant G method, 185 Doubly-magic ±2 nuclei, 164
Constrained Hartree–Fock, 184 Doubly-magic nucleus, 117
Control rod, 58 D quark, 9, 123
Convection current, 99 Dtμ molecule, 37
Core polarization, 120, 121, 207 D3h symmetry, 38
Correlation length, 169
Coulomb barrier, 7
Coulomb energy, 42 E
Coulomb excitation, 211 E 4 /E 2 ratio, 173
Coulomb gauge, 248 Effective charge, 207
Coulomb scattering, 14 Effective many-body interaction, 141
Coulomb trajectory, 17 Effective mass, 57, 136, 138, 139, 144, 147,
Coulomb wave functions, 245 154, 234, 235
Critical mass, 58 e-mass, 148, 155
Critical temperature, 200 k-mass, 155
Cumulant expansion method, 27 ω-mass, 155
Effective range, 79, 80
Effective range theory, 78, 79
Effective sharp radius, 23
D Eikonal approximation, 27, 232
D’Alembert equation, 66 Electric charge distribution, 4
Daughter nucleus, 48 Electric dipole moment, 98
De Broglie wavelength, 17 Electric multipole operator, 102
Debye–Hückel formula, 31 Electric multipole radiation, 251
Decay channel, 38 Electric quadrupole moment, 98, 105, 171,
Decay width, 38, 192 211
Deep inelastic collision (DIC), 202 Electric transition, 205
Deep potential, 197 Electromagnetic field, 97
Deformation Electromagnetic force, 17
oblate, 106 Electromagnetic interaction, 6
prolate, 106 Electromagnetic moments, 97
Deformation parameter, 106, 174 Electromagnetic transitions, 191, 209, 246
αλ , 49 Electron, 3
αλμ , 49, 174 Electron capture, 29, 41, 45
β, γ , 176 Elliott’s SU (3) model, 39
δ, 174 Energy density, 145
Deformed shell model, 177 Energy dependent potential, 235
Degeneracy pressure, 29, 41 Energy problem, 36
Degree of non-locality, 236 Enriched uranium, 58, 59
Delayed neutrons, 58 Equation of state, 141, 147, 148
Density distribution, 20 Equilateral triangle configuration, 38
Density of nuclear matter, 22 Equivalent local potential, 136, 139, 234,
Deuterium, 32 237
Deuteron, 1, 8, 32, 74, 76, 215 Equivalent radius, 23
Diffraction, 16 Escape path method, 201
Diffraction effect, 17 Euler angles, 175, 243
Diffraction pattern, 20 Euler–Lagrange equation, 86, 157
Diffractive eikonal approximation, 27 Evaporation theory, 59
Dirac equation, 4, 99 Even–odd property, 33
Dissipation, 202 Exchange force, 81
Distance of closest approach, 15 Exchange operator, 68
Distortion effects, 21 Exchange potentials, 234
264 Index
F Gravitational force, 6
Faddeev equation, 142 Green function, 67
Fast reactor (fast neutron reactor), 59
Fermi, E., 58
Fermi energy, 29, 94, 110, 167 H
Fermi-gas model, 28 Hadrons, 1
Fermi momentum, 28 Hahn, O., 46
Fermi’s golden rule, 203 Half-life, 6
Fermi velocity, 29 Halo nucleus, 23
Feshbach, H., 94 Hamada–Johnston potential, 88, 89
Fissility parameter, 49–51 Hard core, 41, 93
Fission, 41, 46, 57, 107, 117, 200 Hard-core potential, 88
decay width, 56 Harmonic oscillator, 111
energy release, 47 Hartree–Fock–Bogoliubov (HFB), 164
half-life, 55 Hartree–Fock calculations, 96
threshold energy, 55 Hartree potential, 140
Fission barrier, 48, 54, 62 Hasegawa–Nagata force, 95
Fission coordinate, 49 Heavy-ion fusion reactions, 171, 211
Fission isomer, 60, 62, 63, 187 Heavy particle decay, 191, 200
Fission lifetime, 201 Heisenberg operator, 70
Fock potential, 234 Heisenberg valley, 43–45
Fokker–Planck equation, 202 Hikosaka, Tadayoshi, 116
Form factor, 20, 24 Hofstadter, R., 17
Fourier transformation, 67 Horie, H., 122
Fragmentation reactions, 147, 152 Hot fusion, 130
Fraunhofer diffraction, 20, 26 Hoyle-state, 38
Fraunhofer scattering, 25 Hulthén, L., 77
Fresnel diffraction, 16 Hybrid model, 39
Fresnel scattering, 16 Hyperdeformed states, 62
Friction, 202 Hyperfine structure, 100
Fusion reactions, 33, 41 Hypernuclei, 9, 17
Hyperons, 151
G
Gamow energy, 217 I
Gamow factor, 195, 199, 200 Ikeda diagram, 38
Gamow model, 193, 198, 243 Incoming wave boundary condition, 218
Gamow peak, 216, 217 Incompressibility, 142, 145
Gap equation, 167 Independent particle model, 116
Gap parameter, 167, 186 Induced fission, 46, 47, 202
Gaussian type, 23 Inelastic scattering, 15
Geiger, H., 13 Inertial confinement, 36
Geiger–Nuttal rule, 200 Infinite square-well potential, 110
Geometrical collective model, 175, 209 Integral representation, 224
g-factor, 99, 122, 247 Integro-differential equation, 234
Giant quadrupole resonance (GQR), 212 Interacting boson model (IBM), 175
Giant resonances, 17 Interference, 225
Glauber theory, 27, 232 Internal conversion, 211
Glory scattering, 225 Internal conversion coefficient, 211
G-matrix, 92 Intrinsic electric quadrupole moment, 105
Gogny force, 96 Intrinsic energy dependence, 235
Gradient formulae, 251 Intrinsic quadrupole moment, 211, 242
Gravitational collapse, 41 Intruder states, 116
Index 265
Irreducible tensor, 239 microcluster, 117

Irrotational incompressible fluid, 210 Magnetic charge distribution, 4
Isobaric multiplet, 71 Magnetic confinement, 36
Isobars, 8 Magnetic dipole moment, 3, 98, 105
Isoscalar monopole vibration, 146 Magnetic effects, 24
Isospin, 2, 3 Magnetic moment, 4, 75, 97, 119, 247
Isospin-exchange operator, 70 Magnetic multipole operator, 102
Isospin operators, 2 Magnetic multipole radiation, 251
Isospin singlet, 43 Magnetic radiation, 203
Isospin space, 2 Magnetic transition, 205
Isospin triplet, 43 Magnetization current, 99
Isotones, 8 Magneton, 3
Isotopes, 8 Main sequence stars, 33
Isotope variation of the charge radii, 186 Majorana exchange term, 95
Majorana force, 82
Majorana operator, 70
J Marsden, E., 13
Jahn–Teller effect, 40, 179 Mass density, 29
Jensen, J.H.D., 117 Mass distribution, 26
J-PARC, 9 Mass formula, 42
Massive star, 40
Mass number, 1
K Maxwell–Boltzmann distribution, 217
Kinetic energy density, 144 Mayer, M.G., 117
Klein–Gordon equation, 66 Mean field, 109
K quantum number, 40 Mean life, 3
Kramers factor, 202 Mean square charge radius, 3
Kramers formula, 202 Mean-square radius, 23
Mechanical instability, 153
Meitner, Lise, 46
L
Meson effect, 122
Ladder scattering, 93
Meson exchange current, 100
Landé formula, 242
Meson theory, 5, 65, 85
Lane potential, 125
Metal microclusters, 40
Langer replacement, 226
Method of steepest descent, 217, 224
Langevin equation, 202
Minnesota force, 95
Lattice QCD, 88
Level density parameter, 155 Mirror nuclei, 25, 71
LHC, 11 Molecular dynamics calculations, 38
Linear accelerator, 17 Molecular point of view, 38
Liquid-drop model, 28, 42, 47, 107, 117, Molecular structure, 38
210, 212 Moment of inertia, 212
Liquid–gas phase transition, 152 Mott scattering, 20
Local density approximation, 137 Multiple scattering, 93
Local momentum approximation, 235, 237 Multipole moments, 171
Long wave length approximation, 204 Multipole transition operator, 204
Muon-catalysed fusion (μCF), 36
Muonic-atom, 25
M
Macroscopic–microscopic method, 57, 182,
201 N
Magic numbers, 9, 74, 107, 108, 112, 116, Nagaoka, Hantaro, 13
117, 219 Natural radioactive nucleus, 15
atom, 108, 117 Neutrino, 34, 220
266 Index
Neutrino oscillation, 36 Odd–even mass parameter, 33

Neutron, 1, 2 Odd–odd nuclei, 33
Neutron absorber, 57, 58 Oklo, 60
Neutron capture, 215, 218, 219 One-pion exchange potential (OPEP), 87, 92
Neutron-deficient nuclei, 108, 220 Onion layer structure, 40
Neutron distribution, 22 Optical limit, 232
Neutron drip line, 22, 23 Optical limit formulae, 27
Neutron flux, 219 Optical potential, 229
Neutron halo, 22, 23 Optical theorem, 224, 233
Neutron-induced fission, 46 Oscillator parameter, 112
Neutron moderator, 57, 58 Overlap function, 27
Neutron number, 2
Neutron-rich nuclei, 219
Neutron-rich unstable nuclei, 108 P
Neutron skin, 22 Pairing
Neutron star, 6, 146, 151, 163, 220 monopole, 163, 165
(n,γ ) reaction, 219 multipole, 163
Nijmegen potential, 88 quadrupole, 163
Nilsson levels, 180 Pairing correlation, 33, 42, 55, 58, 127, 128,
Nix, J.R., 49 130, 162
Noble gases, 107, 108 Pairing energy, 254
Nolen–Schiffer anomaly, 89
Parahydrogen molecule, 79
Noncentral potential, 73
Paris potential, 88
Non-locality, 235
Partial wave expansion, 223
Non-local potential, 136, 234
Particle–vibration coupling, 155
Nuclear chart, 8, 9, 73, 173, 187
Pauli exclusion principle, 29, 93, 109, 179,
Nuclear energy, 57
197
Nuclear force, 7, 41, 65
Pauli forbidden states, 197, 244
charge dependence, 89
Pauli, W.E., 4
density dependence, 94
Penetration factor, 198
exchange property, 95
Perfect rigid sphere model, 141
mesons, 91
Periodic table of the elements, 107, 108, 117
radial form factor, 95
(p,γ ) reaction, 220
range, 65
Phase diagram of nuclear matter, 11
state dependence, 80, 92
Nuclear interaction, 17 Phase shift, 79, 83
Nuclear magneton, 4 Phase shift analysis, 78
Nuclear matter, 11, 137 Photo-disintegration, 41, 220
Nuclear potential, 109 Plum pudding model, 13
Nuclear power, 57 Point group, 38
Nuclear radius, 20 Poisson equation, 19, 30
Nuclear reaction in matter, 38 Poisson sum formula, 26, 225
Nuclear waste, 60 Polarization, 85, 235
Nucleon, 2 Polarized electrons, 4, 22
Nucleon–nucleon scattering, 74, 78 Polarized projectile, 78
Nucleon–nucleus reaction, 27 Polarized protons, 22
Nucleosynthesis, 42 Polarized target, 78
Nucleus, 1, 13 Pole of the S-matrix, 192
Nucleus–nucleus reaction, 27 Polyatomic molecule, 40
Number density, 28 Pp collision in the Sun, 7
Pp-chain (proton–proton chain) reactions,
34
O PpI-chain reactions, 34
Octupole vibration, 175 PpII-chain reactions, 34
Index 267
PpIII-chain reactions, 34 Reduced matrix element, 241

P-process, 220 Reduced transition probability, 203, 205
Preformation factor, 199 Reduced width, 199
Pressure, 145, 149 Redundant states, 197, 244
Presupernova star, 40 Refraction effect, 17
Primordial nucleosynthesis, 215 Reid potential, 88, 91
Projection operator, 68, 70 Relativistic effects, 20
Projection operator method, 94 Relativistic mean-field theory, 29, 125, 155
Projection theorem, 76, 120, 241 Reorientation effect, 211
Prompt neutrons, 58 Repulsive core, 83, 88, 92, 197
Proton, 1, 2 Residual interaction, 121, 128, 254
Proton decay, 197 Resonance, 115
Proton distribution, 22 Resonating group method (RGM), 197, 234
Proton–nucleus scattering, 26 Respiration, 108
Proton radioactivity, 191 Rest energy, 18
Proton-rich nuclei, 108 RHIC, 11
R-matrix, 115, 198
Root-mean-square radius, 23
Q Rosenfeld force, 95
QCD phase diagram, 11 Rotation, 238
Quadrupole moment, 76, 242 Rotational band, 38, 63, 173
Quadrupole vibration, 175 Rp-process, 220
Quantization condition, 194, 245 R-process, 41, 42, 215, 219
Quantum chromodynamics (QCD), 1, 65, 85 Rutherford, E., 13
Quantum decay, 200 Rutherford model for atom, 14
Quantum fluctuation, 202 Rutherford scattering, 13, 14
Quantum liquid, 23, 212
Quantum mechanics, 1
Quantum tunneling, 8, 48, 200 S
Quark cluster model, 88 Saddle point, 51
Quark–gluon plasma, 11 Saturation property, 95, 140, 148
Quark model, 3, 5, 123 binding energy, 33, 107
Quarks, 1 density, 23, 28, 107
Quasi-fission, 202 Saturn model, 13
Quasi-spin formalism, 130, 164 Scalar product, 242
Quenching effect, 100, 122 Scattering length, 79, 80
Q-value, 47 Schmidt lines, 120
Schrödinger equation, 18
Scission point, 46
R Screened Coulomb potential, 30
Radiation gauge, 248 Screening effect, 30, 36, 217
Radiative neutron capture, 97 Screening energy, 216
Radioactivity, 191 Screening length, 31
Radius parameter, 23 Segré chart, 8
Rainbow scattering, 225 Segré, E., 14
Raisin bread model, 13 Selection rules, 58, 104, 205
Random walk, 202 Self-consistency condition, 177
Rapid neutron capture, 41 Semi-classical theory, 26, 193, 227
Rare-earth, 169, 187, 208 Semi-Empirical mass formula, 42
Reactivation probability, 37 Seniority, 164
Reactor, 54, 57, 59 Separable type, 236
Recoupling coefficient, 254 Separation energy, 32
Red giant stars, 215 Serber force, 82, 95, 126, 140
268 Index
Shadow scattering, 25, 26 Supernova explosion, 29, 41, 146, 151, 215,
Shallow potential, 197 219
Shape coexistence, 173 Surface diffuseness parameter, 22
Shape transition, 173 Surface effect, 42, 116
Sharp-cut-off-model, 26 Surface energy, 54
Shell correction, 49 Surface tension, 42
Shell correction energy, 131 Surface term, 42
Shell effect, 8, 48 Surface thickness, 148, 154
Shell model, 22, 38, 98, 116 Symmetry, 38, 65
Shell structure, 107, 179 Symmetry energy, 42
SHIP, 131 Synthesis of elements, 6, 97, 215
Si-burning reaction, 40
Single particle model, 119
Single-Particle model estimate, 206 T
Skyrme force, 96, 141 Tani–Foldy–Wouthuysen transformation,
Skyrme, T.H.R., 141 160
Soft-core potential, 88 Teichmann–Wigner sum rule, 199
Solar neutrino problem, 36 Tensor force, 73, 75, 87, 92, 116, 125, 126
Sommerfeld parameter, 194, 216 Tensor operator, 73, 237
Sommerfeld–Watson transformation, 224, Thermal decay, 200, 202
225 Thermal fission, 202
Space-exchange operator, 70 Thermal hopping, 200
Spherical spinors, 159 Thermal neutron, 55, 79
Thermal nuclear reactions, 215
Spin of photon, 249
Thomas–Fermi approximation, 28, 31, 43,
Spin-exchange operator, 69, 70
144
Spin-flip, 85
extended, 152
Spinodal line, 148, 152, 154
Thompson, J.J., 13
Spin–orbit coupling, 160
Thompson model, 14
Spin–orbit density, 144
Three-body force, 142
Spin–orbit interaction, 39, 73, 114, 116, 125, Threshold energy, 115
142 Time-reversal operator, 165
Spontaneous fission, 46, 47, 201 T -matrix, 93
S-process, 45, 215, 219 Transition state theory, 202
S quark, 9 Translation, 238
Stability line, 43, 219 Transmutation, 60
Stationary phase approximation, 224 Transparency function, 27
Stellar nucleosynthesis, 215 Transverse gauge, 248
Stern, O., 4, 97, 100 Triaxial, 176
Sticking probability, 37 Tritium, 3, 32
Stopping power, 36 Triton, 32, 89, 142
Strangeness, 3 Tunneling probability, 194, 246
Strassmann, F., 46 Turning points, 54, 229
Strong absorption, 17 Two-Potential approach, 195
Strong interaction, 1, 6, 65
Structure of nucleon, 17
Strutinsky method, 182 U
Superdeformed states, 62, 187 Uniform approximation, 53, 231
Superfluidity, 186 Unitarity condition, 224
1 S , 163
0 Unitary model operator approach (UMOA),
3 P , 163
2 94
Superheavy elements, 8, 42, 130 Unstable nuclei, 22, 42, 108, 219
Superheavy nuclei, 8 U quark, 9, 123
Index 269
V Well-behaved potential, 140

van der Waals equation of state, 152 Wheeler, J.A., 46
Variational method, 95 Whittaker functions, 231
Vector function, 248 Wiener process, 202
Vector spherical harmonics, 250 Wigner D function, 210
Vector wave function, 248 Wigner–Eckart theorem, 104, 241
Vesman mechanism, 37 Wigner force, 82
Viscosity, 202 Wigner limit, 199
Volkov interaction, 95, 126 Wigner transformation, 236
Volume term, 42 Wildermuth condition, 197
WKB approximation, 137, 194, 227
Woods–Saxon, 22
W
Woods–Saxon potential, 110, 114
Waiting point nuclei, 219
Waiting points, 219
Weak interaction, 6, 8
Weisskopf units, 207 Y
Weizsäcker–Bethe mass formula, 42 Yamanouchi, Takahiko, 116
Weizsäcker correction, 152 Yukawa function, 87

Fundamentals of Nuclear Physics

Uploaded by

Copyright:

Available Formats

Fundamentals of Nuclear Physics

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fundamentals of Nuclear Physics

Uploaded by

Copyright:

Available Formats

What are the main subjects covered in nuclear physics?

What are the main subjects covered in nuclear physics?

What are some phenomena that can be understood from the point of view that a nucleus is a quantum many-body system of nucleons?

What are some phenomena that can be understood from the point of view that a nucleus is a quantum many-body system of nucleons?

Noboru

ISBN 978-4-431-55377-9 ISBN 978-4-431-55378-6 (eBook)

Printed on acid-free paper

This Springer imprint is published by Springer Nature

This book provides an introduction to nuclear physics. Research in nuclear physics

us opportunities to work in their institutions. We are grateful to H. Niko and

Sendai, Japan Noboru Takigawa

3 The Nuclear Force and Two-Body Systems . . . . . . . . . . . . . . . . . . . . . 65

5 Shell Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

6.3 Relativistic Mean-Field Theory (rxq Model) . . . . . . . . . . . . . . . . . 155

8.3.4 Connection Between Electromagnetic Transitions

Abstract It would be useful to have an overview of some fundamental aspects of

1.1 The Constituents and Basic Structure of Atomic Nuclei

© Springer Japan 2017 1

Fig. 1.1 Conceptual

1.2 Properties of Particles Relevant to Nuclear Physics

and, analogously to the Pauli spin operators σ̂ , τ̂ = 2t̂ by

γ 1− <6 × 10−23 stable

ρ ±,0 1 1− 775.5 4.5 × 10−24 (ud̄, dū)

Fig. 1.2 A physical nucleon

The anomalous magnetic moments of nucleons can be understood by either the

for the magnetic moment of a neutron. Assuming Cp = Cn , one obtains

1.3 The Role of Various Forces

Fig. 1.3 The β-decay of a

1.4 Useful Physical Quantities

It is sometimes worth making order-of-magnitude estimates of various physical quan-

c = 2.99792458 × 108 m/s ≈ 3.00 × 108 m/s, (1.9)

Fig. 1.4 Illustration of the V(r)

1.5 Species of Nuclei

Fig. 1.5 Nuclear chart.

Fig. 1.6 Three-dimensional

1.6 Column: QCD Phase Diagram of Nuclear Matter

1. W.-M. Yao et al. (Particle Data Group), J. Phys. G 33, 1 (2006)

2.1 Nuclear Sizes

2.1.1 Rutherford Scattering

Fig. 2.1 Connection

Using the relation

Fig. 2.2 Differential cross

Fig. 2.3 Trajectories of

2.1.2 Electron Scattering

2.1.2.1 The de Broglie Wavelength of Electron

pe2 = 2m e E e + E e2 /c2 . (2.8)

By inserting Eq. (2.8) into (2.6), we obtain

2.1.2.2 Form Factor

where ki is the wave-vector of the incident electron, kf is the wave-vector of the

in order to relate the scattering amplitude to the distribution of protons inside a

where F(q) is defined as

The differential cross section is therefore given by

Correctly, the expression of the differential cross section is obtained as

where j0 (x) is the spherical Bessel function of the first kind.

Exercise 2.4 Prove Eq. (2.20) in the following two methods:

2.1.2.3 Density Distribution

Fig. 2.4 The elastic

(2) The Woods–Saxon Representation of the Density Distribution Accurate infor-

R ∼ (1.1–1.2)A1/3 fm , a ∼ 0.6 fm , ρ0 ∼ (0.14–0.17) fm−3 , (2.24)

as parameters in the Woods–Saxon parametrization.9 The fact that the radius is