NUCLEAR PHYSICS
To Ruth
PHYSICS
by
IRVING KAPLAN
Department of Nuclear Engineering Massachusetts Znstitute of Technology
SECOND
EDITION
ADDISONWESLEY
PUBLISHING
COMPANY
Reading. Massachusetts . Menlo Park, California . London Sydney . Manila
THIS BOOK IS AN ADDISONWESLEY WORLD STUDENT SERIES EDITION
EIGHTH PRINTING 1977
A complete and unabridged reprint of the original American textbook, this World Student Series edition may be sold only in those countries t o which it is consigned by AddisonWesley or its authorized trade distributors It may not be reexported from the country to which it has been consigned, and it may not be sold in the United States of America or its possessions.
Copyright @ 1962 by AddisonWesley Publishing Company, Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying. recording, or otherwise, without the prior written permission of the publisher. Original edition printed in the United States Published simultaneou.sly in Canada. Library of Congress Catalog Card Number 629402.
PREFACE TO THE FIRST EDITIOX
.
I have tried to write an elementary, yet coherent, account of nuclear physics suitable as an introduction to this field. The book is elementary in the sense that the reader is assumed to have had no previous exposure to atomic and nuclear physics. It is supposed that he has had a twoyear college course in physics covering approximately the material contained in Sears' books: Mechanics, Heat, and Sound; Electricity and Illagnetism; and Optics, in the AddisonWesley Physics Series. It is also assumed that the reader is familiar with the differential and integral calculus; a oneyear course in general chemistry would be helpful, if not essential. This book should therefore be useful in an advanced undergraduate course in nuclear physics, to engineers interested in the largescale applications of nuclear physics grouped together under the name "nuclear engineering," or to anyone else with the indicated preparation who might be interested in nuclear physics. The concepts and techniques of nuclear physics are not elementary in the sense that they are easily mastered. They have evolved through a great deal of experimental and theoretical research and cmnot be expressed or explained in simple nontechnical terms. I have tried, therefore, to show how our present knowledge of atoms and nuclei has been developed, and how some of this knowledge is applied. In treating the experiments that yield information about atoms and nuclei and the ways in which these experiments are interpreted in terms of theory, I have tried to avoid both the devil of experimental complication and the deep blue sea of mathematical detail. The emphasis is on the principles underlying the experiments and on the experimental results rather than on the apparatus, on the physical ideas rather than on the details of the calculations. Thus, derivations have been included when they involve straightforward applications of physical principles in terms of mathematics not beyond the integral calculus. When it is necessary to use the results of more advanced methods, such as quantum mechanics, the details of the derivations are omitted; but the physical ideas contained in the results are discussed in some detail and related to the experimental data. The plan of the book is based on the advice of the King of Hearts: "Begin a t the beginning," the King said gravely, "and go on till you come to the end: then stop." The need for a reasonable limit on the size of the book makes it necessary, however, to define the beginning and end arbitrarily and to leave some things out on the way. The first part of the book, consisting of the first seven chapters, is devoted to the background
v
Vi
PREFACE
of nuclear physics and begins with the chemical foundations of atomic theory. The second part, Chapters 8 through 17, treats the physics of the nucleus in a way that seems to me to be logical, yet consistent with the elementary nature of the book. The third part, Chapters 18 through 22, deals with special topics and applications and includes neutron physics and nuclear fission, which do not fit conveniently into the scheme of the second part and a t the same time lead into the most spectacular application of nuclear physics"nuclear energy." The subjects of chargedparticle accelerators and isotope separation, although not really nuclear physics, are closely related to important branches of this field; they are included for this reason and because they form, along with nuclear energy, the main part of nuclear engineering. I have not discussed such subjects as cosmic rays, mesons, and nuclear moments because I think that they need a more advanced treatment than the material covered. I have included, for several reasons, a large number of literature references. For the skeptical or curious reader, the references will supply the experimental and theoretical details omitted from the text, as well as proofs of results that are stated without proof. A more important reason for including these references stems from the fact that atomic and nuclear physics are recently developed fields, and the original literature is more readily available than that of classical physics. The student can read for himself the original papers and books of the pioneers in this fieldRutherford, Bohr, Millikan, Moseley, Aston, Chadwick, Fermi, and others. The articles in which the basic discoveries were first published give, more than any textbook can, a feeling for the imagination and beauty inherent in modern physics. The current literature adds a feeling for the gradual, and sometimes painful, accumulation of experimental and theoretical information, and for the problems which have yet to be solved. This book has been developed from the notes for a course of lectures given at the Brookhaven National Laboratory. I am indebted to colleagues who attended the lectures or used the notes for many valuable comments and suggestions, and to many workers in the field of nuclear physics for permission to cite and use their published results. Above all, I find it hard to express adequately my indebtedness and gratitude to Jean ~ a r l e s for s her unfailing patience, good humor, and skill in preparing the manuscript.
,
IRVING KAPLAN
Emt Patchogue, New York
June, 1964
PREFACE TO THE SECOND EDITION
I have tried to bring the book uptodate and to improve the presentation of the material. The rapid accumulation of information during the past seven years has made it necessary to change many of the numbers and some of the ideas, while the use of the book as a text, by friends and colleagues as well as by myself, has resulted in many suggestions for improvement and clarification. For the convenience of those who used the first edition, the major changes are listed. Section 28, on the detection and measurement of radiation, has been expanded. In the treatment of special relativity, in Chapter 6, derivations of the Lorentz transformation equations and of the formula for the variation of mass with velocity have been included. Section 79 has been added to Chapter 7, in which the Schroedinger equation is applied to the problem of a particle in a onedimensional box; it is hoped that this addition will help the student develop a little more insight into some of the elementary ideas of wave mechanics. Two new sections have been added to Chapter 14: Section 148 deals with the neutrino and its detection; Section 149 with symmetry laws and the nonconservation of parity in betadecay. The last two sections of Chapter 15, dealing with gammadecay, nuclear energy levels and isomerism, have been completely rewritten. A short section, 167, on the limitations of the compound nucleus theory has been added to Chapter 16. In Chapter 17, the treatment of nuclear models has been extended; in particular, the shell and collective models are discussed in somewhat more detail than in the first edition, and the optical model for nuclear reactions is introduced. A short section, 1 7 4 , on the nuclear radius has been added. Chapter 19, on nuclear fission, has been modernized to include material which was classified a t the time the first edition was written but is now part of the open literature. Corresponding changes have been made in Chapter 20, with the result that some of the elementary properties of nuclear reactors can be introduced in a more satisfactory way; a short section on controlled thermonuclear reactions has also been added. To compensate in part for the new material, Chapter 22 on isotope separation has been dropped; this subject seems to attract little interest in courses in nuclear physics, and is now treated thoroughly in other books. Finally, I have tried to improve the problems by adding to them and by replacing some of the "substitution" problems by "thought" problems. I am indebted to many friends, colleagues, and students who have pointed out errors and suggested changes, and I regret that I cannot list
vii
VU1
...
PREFACE
them individually because of the lack of space. I have, however, exercised the author's privilege of disregarding some advice and criticism, and have only myself to blame for the shortcomings of the revised book. I am indebted to Rschel Sprinsky for her invaluable assistance in the preparation of the manuscript. IRVING KAPLAN Cambridge, Mass. January, 1&32?
CONTENTS
PART I
.
THE BACKGROUND OF NUCLEAR PHYSICS
Page
The laws of chemical combination . . . . . Dalton's atomic hypothesis . . . . . . . Avogadro's hypothesis and the molecule . . . Molecular and atomic weights of gaseous elements The standard atomic weight . . . . . . . Atomic weights of nongaseous elements . . . Weights and sizes of atoms and molecules . . . The periodic system of the elements . . . .
Faraday's laws of electrolysis and the electron . . The conduction of electaicity in gases . Cathode rays . . . . . . . . . The electron: the determination of its charge . . . Positive rays . . . . . . . . . . . Xyays . . . . . . . . . . . . . Radioactivity: alpha.. beta.. and gammarays . The detection and measurement of radiation . . .
. . .
.
.
. . .
. . .
. . .
. . .
19 20 20 24 31 31 32 35
31 32 33
34
The Thomson atom . . . . . . . . . . . . Rutherford's theory of the scattering of alphaparticles . The experimental test of the Rutherford scattering theory . . Some characteristics of the atomic nucleus .
51 53 58 62
41 42 43 44
Some properties of xrays . . . . . . . . . . The scattering of xrays by atoms and the number of electrons per atom . . . . . . . . The diffraction of xrays and Bragg's law . Characteristicx.rayspectra . Moseley'slaw .
. .
67
70 72 76
51
The failure of classical physics to describe atomic phenomena
i x
.
84
The emission and absorption of thermal radiation . The classical theory of thermal radiation . The failure of the classical theory of thermal radiation Planck's quantum theory of thermal radiation . 56 Quantum theory and the photoelectric effect .
52 53 54 55
. . .
86 89 93 96
99
The role of the special theory of relativity in atomic and nuclear physics . . . . . . . . . . 62 The ether and the problem of absolute velocity . 63 The MichelsonMorley esperiment . 64 The problem of the invariant form of physical theories 65 The special theory of relativity; the variation of mass with velocity . . . . . . . . . . . 66 The relationship between mass and energy . 67 The Compton effect . . . . . . . . .
61
. . . . . . .
104 105 107 110 113 124 127
. . . . . . .
Atomic spectra . . . . . . . . . . . . . . The Bohr theory of atomic spectra and atomic structure . . The stationary states of an atom . . . . . . . . Extension of the Bohr theory: elliptic orbits . . . . . Failure of the Bohr theory: wave mechanics and the correct quantum numbers . . . . . . . . . . . . . Atomic theory and the periodic table . Atomic theory and characteristic xray spectra . The basic ideas of wave mechanics . The solution of the Schroedinger equation: some useful examples and results . . . . . . . . . . .
PART I1
.
.
. .
135 139 145 149 154 156 159
.
.
.
164
171
.
.
.
THE NUCLEUS
The protonelectron hypothesis of the constitution of the nucleus The angular momentum of the nucleus; failure of the protonelectron hypothesis . . . . . . . . . . . 83 Nuclear transmutation and the discovery of the neutron . . . 8 4 The protonneutron hypothesis . . . . . . . . . . 85 Magnetic and electric properties of the nucleus . 86 Additional properties of atomic nuclei . .
81 82
183 185 189 191 193 195
91
Natural radioactivity and isotopes
.
.
197
.
CONTENTS
xi
Page
92 93
94
95
. . Positiveray analysis and the esistence of isotopes Isotopic masses and abundances: the mass spectrograph and mass spectrometer . . . . . . . . . The stable isotopes of the elements and their percentage abundances . . . . . . . . . . Atomic masses: packing fractions and binding energies .
. .
. .
198 203
. .
209 218
101 102 103 104 105 106
The basis of the theory of radioactive disintegration . . The disintegration constant. the half.life. and the mean life Successive radioactive transformations . . . . . . Radioactive equilibrium . . . . . . . . . . The natural radioactive series . . . . . . . . Units of radioactivity . . . . . . . . . . .
. .
. .
. . . . . .
. .
229 234 239 244 248 253
Transmutation by alphaparticles: alphaproton reactions The balance of mass and energy in nuclear reactions . . The neutron: alphaneutron reactions . . . . . . The acceleration of charged particles . . . . . . Transmutation by protons . . . . . . . . . Transmutation by deuterons . . . . . . . . . Transmutation by neutrons . . . . . . . . . Transmutation by photons . . . . . . . . . Nuclear chemistry: nuclear masses . . . . . . .
Thediscoveryofartificialradioactivity . . . . The artificial radionuclides. Electron and positron emission . Orbital electron capture . . . . . 123 The transuranium elements . . . . . . . 124 The artificial radionuclides: alphaemitters . . . . . . . . 125 Isotope tables and nuclide charts
121 122
. . . .
283
. . . . 285 . . . . 290 . . . . ' 295
. . . .
297
131 132 133 134
The velocity and energy of alphaparticles . . . . . . The absorption of alphaparticles: range. ionization. and stopping power . . . . . . . . . . . Rangecnergy curves . . . . . . . . . . . . Alphaparticle spectra. Longrange particles and fine structure
.
. .
304 307 317 319
.
xii
CONTENTS
135 Nuclear energy levels . . 136 The theory of alphadecay
. . . . . . . . . . . . . . . . . . . . . .
322 326
The velocity and energy of betaparticles . The absorption of beta.particles . Rsnge. ionization. and energy loss . . . . . . . . . . . . Rangeenergyrelationsforbetaparticles . Betaparticle spectra. The continuous spectrum . The theory of betadecay . Basis of the theory . The theory of betadecay . Results and comparison withexperiment . . . . . . . . . . . . Energy levels and decay schemes . . . . . . The neutrino . . . . . . . . . . . . Symmetry laws and the nonconservation of parity in w e c a y
. 340
.
.
. . .
.
346 350 352
358
363
371 374 381
of experimental and theoretical results . . . . . . . 157 The measurement of gammaray energies . . . . . . 158 Gammadecay: internal conversion . . . . . . . . 159 Gammadecay and nuclear energy levels: theory . . . . 1510 Gammadecay and nuclear energy levels: experimental results . . . . . . . . . . . and nuclear isomerism
151 The absorption of gammarays by matter: experimental data 152 The interaction of gammarays with matter . . . . . 153 Photoelectric absorption . . . . . . . . . . . 154 Compton scattering . . . . . . . . . . . . 155 Electronpositron pair formation . . . . . . . . 156 The absorption of gammarays by matter Comparison
.
Nuclear reactions.and escited states of nuclei . . . . . . 447 The compound nucleus . . . . . . . . . . . . 452 Cross sections for nuclear reactions . . . . . . . . . 457 Esperimental results . General considerations . . . . . . 463 Neutroninduced reactions . . . . . . . . . . . 465 166 Reactions induced by protons and alphaparticles . . . . . 477 167 Limitations of the compound nucleus theory . . . . . . 484 168 Deuteroninduced reactions: intermediate nuclei. high energy . 487 169 Reactions induced by gammarays . . . . . . . . . 491 1610 Reactions a t ultrahigh energies . . . . . . . . . . 496 1611 Reactions with light nuclei: energy levels of light nuclei . . . 496
161 162 163 164 165
.
CONTENTS
Xlll
...
Page
171
Nuclear binding energies and the saturation of nuclear forces Nuclear stability and the forces between nucleons . . . 173 Energy levels of light nuclei and the hypothesis of the charge independence of nuclear forces . . . . . . . . . 1 7 4 The nuclear radius . . . . . . . . . . . . . 175 The interaction between two nucleons . . . . . . . . . . . . 176 The status of the problem of nuclear forces 177 Nuclear models. The shell. or independent particle. model . 178 The liquid drop model and the semiempirical binding energy formula . . . . . . . . . . . 179 The collective nuclear model . . . . . . . . . . 1710 The optical model for nuclear reactions . . . . . . .
172 PART I11
.
SPECIAL TOPICS AND APPLICATIONS
181 182 183 184 185 186 187
The production of neutrons . . . . . . . . . . . 559 . . . . . . . . . . . 563 The detection of neutrons The interaction of aeutrons with matter in bulk: slowing down . 565 Thermal neutrons . . . . . . . . . . . . . The diffusion of thermal neutrons . . . . . . . . Cross sections for neutroninduced reactions: measurement of the total cross section . . . . . . . . . Scattering. absorption. and activation cross sections . . .
The discovery of nuclear fission . . . . . . . . Fission cross sections and thresholds . . . . . . The fission products . . . . . . . . . . . The mass and energy distributions of the fission products Neutron emission in fission . . . . . . . . . The energy distribution of the neutrons emitted in fission The energy release in fission . . . . . . . . . The theory of the fission process . . . . . . . .
. . . . 201 Nuclear fission as a source of energy 202 The chainreacting system or nuclear reactor . . 203 Thermal nuclear reactors . The neutron cycle . . 204 The calculation of the multiplication factor for a homogeneous thermal reactor . .
.
. .
.
. .
.
. .
.
xiv
CONTENTS
Page
205 206 207 208 209
. .
The heterogeneous thermal reactor . . The critical size of a thermal reactor . Power and breeding . . . . . . . . . Energy production in stars . Thermonuclear reactions Controlled thermonuclear reactions . .
.
.
656
660
.
.
.
664
.
666 671
The CockcrofbWalton machine . . . . . . . . . . The electrostatic generator or Van de Graaff machine . . . The cyclotron . . . . . . . . . . . . . . . The frequencymodulated cyclotron or synchrocyclotron . . . The acceleration of electrons. The betatron and the electron synchrotron . . . . . . . . . . . . The proton synchrotron ' . . . . . . . . . . . . Linear accelerators . . . . . . . . . . . . . The alternatinggradient synchrotron . . . . . . . .
. JOURNAL APPENDIX 1 1 1 ABBREVIATIONS FOR REFERENCES .
APPENDIX IV. THE RUTHERFORD SCATTERINGFORMULA . APPENDIX V
.
717 718
.
.
.
NUCLIDE CHART .
.
Inside back cover
The Background of Nuclear Physics
3. and so on. which introduced the concept of atomic weight. 11 The laws of chemical combination. The laws of chemical combination were unified at the beginning of the 19th century by Dalton's atomic theory.52 parts by weight of magnesium.12 parts of aluminum.96 Copper 31. 3.68 The same combining weights are obtained when the oxides are prepared by methods other than simple oxidation. as turns out to be more convenient. which has been found to be true for all chemical reactions. Two of the fundamental quantities of atomic and nuclear physicsatomic weight and atomic numberhad their origins in the correlation of the results of chemical experiments and in the systemization of the properties of the chemical elements. are expressed by the law of definite proportions.76 Tin 29. This constancy of the weight. 2.71 parts of tin. is expressed in the law o f conservation o f mass. If. The experimental information which gave rise to the atomic theory of matter can be summarized in a few basic laws. one part by weight of oxygen always combines with 1. the weight of the vessel and its contents before and after heating is the same.04 Aluminum 8.50 parts of calcium. Oxygen 8 Magnesium 12. The systemization of the atomic weights and properties of the elements then led to the formulation of the periodic system and to the concepts of atomic number and atomic structure.16 Calcium 20. who showed that when tin is made to react with air in a closed vessel. The first of these was deduced from investigations such as those of Lavoisier. the following combining weights are obtained. which states that a particular chemical compound always contains the same elements united together in the same proportions by weight. These experimental results.97 parts of copper. the weight of oxygen is taken to be 8. 3 . which statere that the mass of a system is not affected by any chemical change within the system. I t was found that when various metals are oxidized in excess air. and many others like them. 1. The development of methods for determining the atomic weights of the elements and the investigation of chemical reactions were major contributions of 19th century chemistry.CHAPTER 1 THE CHEMICAL FOUNDATIONS OF ATOMIC THEORY The foundations of modern atomic theory were laid in the late 18th and 19th centuries in the attempt to understand the chemical properties of matter.
1 It was also found that two elements can combine to form more than one compound.015) of sulfur.47 parts of calcium. and with 32. This law is illustrated by the reactions which take place between oxygen and sulfur. the &fferent weights of one which combine with the same weight of the other are in the ratio of small whole numbers. For example. In other words.69) parts of zinc combine with 32.38(= 2 X 32. 8 parts by weight of oxygen combine with 8. and 28. 65. 4 7 : ~ ~ or z = 20. and sulfur and zinc.53:8 = 7 1 . The law of reciprocal proportions states that the weights of two (or more) substances which react separately with identical weights of a third are also the weights which react with each other. Nitrogen 14 14 14 14 14 Oxygen 8 . or in some simple multiple of those proportions. or simple multiples of them.04 for calcium. If oxygen = 8 is taken as the standard. Thus. The further study of the quantitative relations between the elements led to a fourth law of chemical combination.69 parts of zinc to form an oxide of zinc. which states that if two elements combine to form more than one compound. .1 x 8 16=2X8 24=3X8 3 2 ~ 4 x 8 40 = 5 x 8 The different weights of oxygen which can combine with the same weight of nitrogen are integral multiples of 8. Another generalization deduced from the analysis of chemical compounds is that of the combining weight or chemical equivalent of an element. This concept was touched upon in the discussion of the oxides of various metals in connection with the law of definite proportions. 28. the amount of each of the other elements which combines with this standard amount of oxygen can be calculated. Results of this kind are summarized by the lato of multiple proportions. oxygen and zinc. at least five distinct oxides of nitrogen are known in which the relative proportions by weight of nitrogen and oxygen are listed in the table. Consider a number of chemical compounds such as those tabulated on the next page.53 parts by weight of oxygen combine with 71. in the formation of zinc sulfide.4 CHEMICAL FOUNDATIONS OF ATOMIC THEORY [CU. Thus.06(= 4 X 8. then A and B can combine with each other only in those proportions in which they combine with C.015 parts of sulfur to form an oxide of sulfur. if each of two substances A and B combines with a substance C.
The numbers obtained for the different elements are called the combining weights or equivalent weights of the elements. Silver iodide where x is the amount of the element. Silver chloride 6.04 Hydrogen 1. 88.008 Chlorine 35. and led to the atomic hypothesis proposed early in the 19th century by Dalton. particularly in weight or mass.008 parts by weight of hydrogen. that of magnesium is 12. in this case calcium. The chemical elements consist of discrete particles of matter. . 2. multiple proportions.76 Magnesium 25.9 The value obtained for magnesium by the above indirect process is the same as that obtained earlier from data on the direct oxidation of magnesium.23 Chlorine 74.008. When this procedure is continued. In hydrogen chloride.81:8 = 11. in magnesium chloride. it is found that a number can be assigned to each element which represents the number of parts by weight of the given element that can combine with 8 parts by weight of oxygen or 1.47 Hydrogen 11.16 Silver 107. definite proportions. Each element is characterized by the weight of its atom. The four laws of chemical combination. percent Calcium 7 1.26 Silver 45.06 Calcium oxide Water Hydrogen chloride Magnesium chloride 5. that of chlorine is 35.1 1.88 Iodine 126. 2. together with the concept of combining weights. and the combining weights of the elements represent the combining weights of their respective atoms. All atoms'of the same element are identical in all respects. if the weight of hydrogen is taken as 1.008 for hydrogen.47 Chlorine 24. or x = 1.94 / ~ e r c e n t l Oxygen 28. 3.81 Chlorine 97.71 Iodine 54. From the experimental data for water. Oxygen 8 Calcium 20. atoms.19 Hydrogen 2. 4.53 Silver 75. which combines with 8 parts of oxygen.45.16. conservation of mass.19:s. 12 Dalton's atomic hypothesis.45 Magnesium 12.53 Oxygen 88. if the weight of chlorine is 35. summarized the basic experimental facts of chemical combination as known about 1800. Dalton's atomic hypothesis was proposed (1803) to account for the facts expressed by the laws of chemical combination and was based on the following postulates: 1. and reciprocal proportions. different elements have atoms differing in weight.45. which cannot be subdivided by any known chemical process and which preserve their individuality in chemical changes.
6 CHEMICAL FOUNDATIONS OF ATOMIC THEORY [mu. m. p. Similarly. and compounds of the elements B and C contain z atoms of Band y atoms of C. If a. This treatment can be extended to any number of compounds formed by two elements. and q are all small integers. and b is the weight of an atom of B. Chemical compounds are formed by the union of atoms of different elements in simple numerical proportions. the products np and mq are therefore also small integers and the weights of B combining with a fixed weight of A are in the ratio of two small integers. Since atoms undergo no change during a chemical process. Compounds of the elements A and C contain p atoms of A and q atoms of C. Since all atoms of the same element are identical in weight. the result is the law of conservation of mass.9 or np:mq. Hence. b. giving the law of definite proportions. and one part of A combines with &/ma parts of B. and a compound is formed by the union of atoms of different elements in a simple numerical proportion. e. they preserve their masses. 1 3. then the first compound contains ma parts of A and nb parts of B. the proportion by weight of B and C is Hence. If a is the weight of an atom of A.g. and the mass of a compound is the sum of the masses of its elements.. then one part of A combines with &/ma parts of B. . in the second compound. 1:2. giving the law of multiple proportions. I t is easy to show that the laws o f chemical combination can be deduced from these postulates. Similar arguments lead to the law of reciprocal proportions. the weights of B combined with a fixed weight of A are in the proportion Z . 1:1. respectively. and C. the proportions by weight in which two elements are combined in a given compound are always the same. n. Consider next the case in which two elements A and B can form two different compounds. Suppose that the first compound contains m atoms of A and n atoms of B. and that the second compound contains p atoms of A and q atoms of B. one part of A combines with qb/pa parts of B. and care the weights of the atoms A. It follows that In the compound formed by B and C. B. m'p' According to the third postulate. 2:3. Compounds of the elements A and B are formed according to the scheme m atoms of A and n atoms of B. and qc/pa parts of C. 2:1.
Dalton was forced to make arbitrary assumptions. If W 1and W 2 are the weights of two elements which combine to form a compound. When can fix only the the ratio n l : n 2is known. and led to many difficulties in the application of his atomic theory to the rapidly growing field of chemistry. however. there is always a simple relation between the volumes of the interacting gases. + 2 volumes of ammonia. I t follows that if elements in a gaseous state combine in simple proportions by volume. and also of the products. 13 Avogadro's hypothesis and the molecule. z are the same as q. Water was regarded as a compound of one atom of hydrogen and one atom of oxygen. n or small integral multiples of them. then y. it contains one atom of each element. that if only one compound of two elements is known. because it provided no way of determining even the relative weights of the atoms of the different elements. then where Al and A2 are the atomic weights and nl and n2 are the whole numbers of atoms of each element which enter into combination. This difficulty arose because Dalton had no way of finding out how many atoms of each element combine to form a compound. This is the law of reciprocal proportions. Dalton's theory was incomplete. if these are gases. GayLussac showed (18051808) that when chemical reactions occur between gases. and the latter form the basis of an atomic theory of matter. To apply his theory. but wrong. he assumed.Since. then the numbers of atoms in equal volumes of the reacting gases . and if the elements also combine in simple proportions by atoms. for example. by the third postulate. all of the quantities in the last equation are small integers. One of the difficulties met by the atomic hypothesis arose as a result of studies of the combining properties of gases. the existence of hydrogen peroxide being unknown a t the time. When the reacting gases and the products are under the same conditions of temperature and pressure + 1 volume of chlorine 2 volumes of hydrogen + 1 volume of oxygen + 2 volumes of steam. 3 volumes of hydrogen + 1 volume of nitrogen 1 volume of hydrogen + 2 volumes of hydrogen chloride. the value of the ratio W 1 : W 2 ratio of the atomic weights. The laws of chemical combination have thus been deduced from Dalton's postulates. Dalton's assumption was simple.
each molecule containing one atom of hydrogen and one atom of chlorine. argon. when 1 volume (n atoms) of hydrogen reacts with 1 volume (natoms) of chlorine. neon. He then postulated that equal volumes of all gases under the same physical conditions contain the same number of moleczcles. and nitrogen contain two atoms. of atoms. then 2 volumes (2n "compound atoms") of hydrogen chloride are formed. Although the atoms cannot be split so that one atom of hydrogen or chlorine enters into the composition of two molecules of hydrogen chloride. it is possible to write for any gaseous substance . say n. a result which contradicts that postulate of Dalton's hypothesis according to which atoms cannot be subdivided by any chemical process. The relative density of a gas is defined as the ratio of the weight of a given volume of the gas to that of an equal volume of a standard gas measured a t the same temperature and pressure. and suppose that 1 volume of hydrogen or chlorine contains n molecules. I n view of Avogadro's hypothesis. Avogadro (1811) showed that the difficulty could be resolved if a distinction is made between elementary atoms and the small particles of a gas. He assumed that the latter are aggregates of a definite number of atoms. 1 must be simply related. Under this assumption. and xenon have one atom per molecule.8 CHEMICAL FOUNDATIONS OF ATOMIC THEORY [CHAP. This hypothesis made it possible to interpret GayLussac's law of combining volumes in t e r m s of the atomic hypothesis. chlorine. Dalton assumed that equal volumes of different gases under the same physical conditions contain an equal number. Avogadro's hypothesis allowed GayLussac's law to be reconciled with Dalton's atomic hypothesis. Every atom of hydrogen and chlorine must then be split in half to form two compound atoms of hydrogen chloride.    Molecular weight of gas Weight of any volume of gas Molecular weight of standard gas Weight of equal volume of standard Density of gas Density of standard gas . Assume that each molecule of hydrogen or chlorine consists of two elementary atoms. krypton. and called the aggregates molecules in order to distinguish them from the elementary atoms. oxygen. the inert gases helium. one molecule of hydrogen and one molecule of chlorine can be divided between two molecules of hydrogen chloride. In this way. These molecules react to form 2n molecules of hydrogen chloride. 1 4 Molecular and atomic weights of gaseous elements. The detailed analysis of many reactions between gases has shown that the molecules of the gaseous elements hydrogen. The relative molecular weights of gaseous substances can be determined from mearmrements of relative densities.
942 83.214 0.816 27. the results of the fifth column are obtained for the molecular weights relative to that of oxygen taken as 32.T. for example.977 7 1.08988 1. and the molecular weights relative to this standard are given in the fourth column. hydrogen. The choice of the standard gas is arbitrary and may be. The results for some molecular weight determinations based on this method are shown in Table 11.034 39.42904 1. fluorine. its relative atomic weight can be determined when the number of atoms per molecule is known.52 3. with its molecular weight arbitrarily taken as 32. If hydrogen is chosen as the standard. since a molecule of hydrogen contains two atoms.17847 0.851 Molecular weight M (hydrogen) = 2 M (oxygen) = 32 2.828 37. The densities of the gases are given in grams per liter mea?ured under standard conditions (O°C and 780 mm mercury pressure.000 31. and chlorine  .161 39.738 71. oxygen. Of the gaseous elements listed.690 82. 0.971 20.510 130.013 32. and if its atomic weight is arbitrarily taken to be unity. the relative molecular weight of a gas is Molecular weight = 2 X density of gas relative to that of hydrogen. oxygen.141 MOLECULAR A N D dTOMIC WEIGHTS O F GASEOUS ELEMENTS 9 Element Hydrogen Oxygen Nitrogen Fluorine Chlorine Helium Argon Krypton Xenon Density. hydrogen. g / I at N. Highprecision work on the combining volumes of .19 * Corrected for deviations from ideal gas laws. Oxygen.000 28.T.gases has shown that there are slight deviations from GayLussac's law which have been traced to deviations from the ideal gas laws.7837 3.708 5.696 3. The thiid column gives the molecular weight relative to that of hydrogen taken as 2.000 37.90035 1. abbreviated as N. When corrections are made for these deviations.996 20.97 3. may also be used as a standard. When the relative molecular weight of a gaseous element has been determined in this way.03 131. or air.).P.25055 1. nitrogen.P.02 Molecular weight M (oxygen) = 32* 2.
the relative atomic weights of the inert gases are the same as their molecular weights. Hence. consequently hydrogen was rejected in favor of oxygen. If the standard is chosen so that the atomic weight of hydrogen is equal to or greater than unity. weight. Now. then 1.504. Atomic weight = Valence.OW. 15 The standard atomic weight. The choice of the value of the atomic weight of oxygen was based on the desire for convenient values for the other atomic and molecular weights. or equivalent. The choice of oxygen was based on the criterion of convenience.10 CHEMICAL FOUNDATIONS OF ATOMIC THEORY [CHAP. values. not absolute. and the atomic weight of hydrogen would be 0. The choice of 16 for oxygen makes the atomic weight of hydrogen equal to 1. and it was thought that the determination of the atomic weight of an element should be connected with the standard as closely as possible. the atomic weight of an element is a small integral multiple of the equivalent weight. A possible choice for the atomic weight of oxygen is its combining. say grams. of hydrogen combine with 8 gm of oxygen. I f this choice were made. hydrogen is the lightest of all the elements.008 gm of hydrogen contain twice as many atoms as 8 gm of oxygen. Equivalent weight The equivalent weight and the valence are useful because their values can be determined by chemical methods and lead t o the determimation of atomic weights. 1 are diatomic and their relative atomic weights are half of the molecular weights. Hydrogen was sometimes used as a standard. Oxygen is a useful standard for chemists because nearly all the elements form stable compounds with oxygen. For practical purposes. . and this standard is taken as oxygen with an atomic weight of 16 and a molecular weight of 32. and also leads to relative weights of heavier atoms and molecules which are not inconveniently large. 1. . It has also been found that two atoms of hydrogen combine with one atom of oxygen to form one molecule of water. there will be no element with an atomic weight smaller than unity. it is necessary to fix a standard for atomic weights. I t has been emphasized that the values of the molecular and atomic weights determined from density measurements are relative. but it forms very few compounds with metals which are suitable for atomic weight determinations. 8 units. The atomic weight of oxygen is then equal to twice the equivalent or combining weight. Experimental evidence shows that 1.008 parts by weight.008 (of the same units) would be the weight of two atoms of hydrogen. and this integer is called the valence of the element:    . In general.
volatile compounds can be found with the aid of the rule of Dulong and Petit. The method of relative densities can be used to find the approximate atomic weight of a nonvolatile element which has gaseous compounds. with an average of about 6. The weights of the particular element in the molecular weights of the various compounds are found by chemical analysis. Hydrogen = 1 Molecular weight 16 30 46 74 78 28 44 Weight of carbon in one molecular weight o f compound 12 24 24 48 72 12 12 = = = = (2X (2 X (4X (6 X 12) 12) 12) 12)  16 Atomic weights of nongaseous elements. an approximate . or no. the results of Table 12 are obtained. In general.ATOMIC WEIGHTS O F NONGASEOUS ELEMENTS Compound Methane Ethane Alcohol Ether Benzene Carbon monoxide Carbon dioxide Relative density.4. and the relative density measurements are used only to decide between various possible molecular weights. According to this rule. The values given in the table are approximate. more refined measurements and corrections for deviations from the ideal gas laws must be made in order to get accurate values of the stomic weight. In the case of carbon. The molecular weights of as many volatile compounds of the element as possible are found from! the relative densities. The smallest weight of the element present in one molecular weight of all its known volatile compounds is assumed to be the atomic weight of the element relative to oxygen = 16. If the specific heat of a solid element is measured. the numerical value of the product is usually between 6 and 7. When the specific heat is expressed in calories per gram. accurate values of the molecular and atomic weights are found from careful chemical analysis of the compounds. and the atomic weight of carbon is taken to be 12. Approximate values of the atomic weights of solid elements which have few. a t least one of the weights of the element. These weights must be integral multiples of the atomic weight and if the number of compounds used is large enough. present in the molecular weights of its compounds will probably be the atomic weight itself. the product of the atomic weight and the specific heat is roughly the same for many solid elements.
Since 67. Various other methods have been developed for determining accurate values of molecular and atomic weights. and a grammolecular weight of hydrogen is 2.09 gm/liter a t N.2 parts of this element combine with 16 parts of oxygen. so that the grammolecular volume is . Since the atomic weight of hydrogen is taken to be 1.21 200.. 2'01G 0.09 .100 28. More refined chemical analysis gives the value 63. For example.38 value of the atomic weight is found.22.3 .07 65.09 107.. there can be little doubt that 63.6 than 127. the specific heat of copper is 0.4 0.12 CHEMICAL FOUNDATIONS OF ATOMIC THEORY [CHAP.3 is much closer to 63. it is possible to calculate the volume occupied a t standard temperature and pressure by a grammolecular weight of a gas. 17 Weights and sizes of atoms and molecules.54. When the standard for atomic and molecular weights has been chosen. and a list of the presently accepted values of the atomic weights of some of the elements is given in Table 13.T. by the number of grams of gas numerically equal to the molecular weight.095 cal/gm.6 is the atomic weight of copper. this volume is called the grammolecular volume or molar volume. 6.6 and 127.2.61 14. that is.000 39.016 gm.880 32..008.095 = 67.066 238.4 liters. is the approximate atomic weight of copper. its molecular weight is 2. The density of hydrogen (Table 11) is very nearly 0. This value can then be used to determine which of the different values represented by the combining proportions by weight is the best choice for the atomic weight. By the rule of Dulong and Petit. Chemical analyses of compounds of copper with oxygen show that 63.016.P. 1 TABLE 13 ATOMIC WEIGHTSOF COMMON ELEMENTS RELATIVE TO OXYGEN = 16 Element Aluminum Barium Boron Calcium Carbon Chlorine Copper Fluorine Gold Hydrogen Iron Atomic weight Element Lead Mercury Nitrogen Oxygen Potassium Silicon Silver Sulfur Uranium Zinc Atomic weight 207.008 16.
therefore. that the grammolecular weights of all gases a t N. a t the melting point. One grammolecular weight of water occupies 18 cm3 and contains 6. In the solid phase a t the melting point. 22.171 WEIGHTS AND SIZES OF ATOMS AND MOLECULES 13 Experiments with a large number of gases have shown that the same volume.0 X molecules.0249 X loz3.T.4 liters.37 X lo' cm and 1. the atoms are closely packed and the distance between'them is not much greater than the atomic dimensions. or mole. I t follows. which has a molecular weight of 18 and a density of one gram per cubic centimeter.7 X lo' cm. in the neighborhood of 1 X lo' cm to 3 X 10'cm. while that of the heaviest naturally occurring element. This number is also the number of atoms in a gramatomic weight of a gaseous element. or r = 1. uranium. According to Avogadro's hypothesis. An estimate of the sizes of molecules and atoms can be made by considering water. is occupied by one grammolecular weight. contain the same number of molecules. An estimate of the atomic radius of a solid element can be made from the atomic volume. The weight of the lightest atom.P. and hence of any element.9 cm3. are obtained for the molecular radii of most of the common gaseous elements and compounds. Values of the radii of gas molecules can also be calculated with the aid of the kinetic theory of gases from measured values of the viscosity.95 X gm. the volume of a gramatom of the element. The radii of hydrogen and oxygen molecules obtained in this way are 1. The radius of a water molecule is then given by the 2X relation 47rr3 = 2 x cm3. is 1. this volume contains the same number of gas molecules regardless of the chemical nature of the gas. hydrogen.67 X gm. their volume is about onethird smaller. that is. the volume of a single molecule would be 3 X cm3. of any gas a t N. is 3. The accepted value a t present is 6. and are quite close to the value of the radius of the water molecule given by the above rough calculation. The volume per atom is . this number is called the Avogadro number or Avogadro's constant. Solid lead. The weight in grams of a single atom or molecule can be found by dividing the gramatomic or grammolecular weight by the Avogadro number.81 x lo' cm.T. has a density of 11 gm/cm3 and its atomic volume' is 18. If the molecules are assumed to be spherical. lo' cm may be taken as the order of magnitude of atomic radii. Since the number of atoms per molecule of these substances is small. and the thermal conductivity. If the water molecules were cubic in shape so that they packed together without any free space. Its value has been determined by several independent methods. Similar values. respectively. or cm3. with results which agree very well. one method will be described in Chapter 2. the diffusion constant.P. a t its melting point.
The atomic weight of an element may sometimes be corrected with the aid of knowledge of the weights of adjacent elements. Carbon: 0. and attempts were made to classify the elements according to such similarities. cesium). show a periodicity of properties. rubidium. since elements of low.g. . The discovery of many yet unknown elements may be expected. and if the atom is assumed cubical (no free space). With better methods for determining the atomic radius. 18 The periodic system of the elements. and high atomic weight are included in the list.2 X cm.  .9 X lo' cm . the following values have been obtained for some of the solid elements. 4.14 CHEMICAL FOUNDATIONS OF ATOMIC THEORY [CHAP.45 X Sodium: 1. The elements which are the most widely distributed in nature have small atomic weights. and all the elements of small atomic weights are characterized by sharply aefined properties. Certain characteristic properties of the elements can be predicted from their atomic weights. the radius is about 1. it became apparent that there are similarities in the properties of certain groups of elements. This law may be summarized in the following statements made by Mendelkff.g. 8. 1.5 X cm These values are similar to those obtained for the atomic radii of the gaseous elements..6 X lo' cm. 3. . the atomic dimension would be about 3. Elements which are similar as regards their chemical properties have atomic weights which are either approximately the same (e.15 X cm3. elements analogous to aluminum and silicon. The elements.77 X Aluminum: 1. osmium) or which increase regularly (e. who proposed his periodic 2 m in 1869. The arrangement of the elements or of groups of elements in the order of their atomic weights corresponds with their valences. 2. 5. for instance. 6. 1 3. iridium. Cesium is the atom with the largest radius. The magnitude of the atomic weight determines the character of an element. intermediate. if arranged according to their atomic weights. They indicate that atoms have dimensions of the order of lo' cm regardless of the atomic weight. . potassium. If the lead atom is assumed to be spherical and the space between atoms is taken into account. Bismuth: 1. 7.. platinum. The most successful attempt was that of MendelBeff. As information about the chemical and physical properties of the elements was accumulated.
and others. In each period there is a definite and characteristic gradation of properties from one element to the next. is reversed so that these elements may fall into the places which correspond to their properties. although there is a steady but gradual variation with increasing atomic weight. such as that shown in Table 1 4 . For four pairs of elements. So far as was known in MendelQff's time. have rectified these errors. 18. argon and potassium. the alkaline earth metali 2a. The basis of Mendel&ffJs classification of the elements. and because of ignorance of the inert gases. atomic heats a t low temperature. 8. Later research has shown that the chemical atomic weights were not seriously in error. The vertical columns. compressibility. namely. ionic mobility.181 THE PERIODIC SYSTEM OF THE ELEMENTS 15 MendelQff's general conclusions have been verified a i d extended. and that the real property underlying the periodic classification is the atomic number. the elements have similar properties. heats of formation of oxides and chlorides. thorium and protoactinium. and inert gases group 0. 8. Modern versions of the periodic table. thermal expansion. the order assigned in the periodic table departs from that of the atomic weights. the requirement that the elements should be arranged in the order of increasing atomic weight. tellurium and iodine.Within a group. offered no explanation of the different features of the periodic table. who believed that the accepted values of the atomic weights were inaccurate. sometimes because of errors in the atomic weights. hardness. to which the atomic weight is approximately proportional. boiling point. that on the right side the "bn subgroup. In each subgroup. cobalt and nickel. as well as the properties of these elements. crystalline form. as determined by the atomic weights. Mendelkeff constructed a table in which the elements were arranged horizontally in the order of their atomic weights and vertically according to their resemblances in properties. The alkali metals constitute the subgroup la. magnetic susceptibility. 32. atoms of dif . This was done for the first three pairs by MendelQff himself. respectively.omic weight/density). Among the properties which show the kind of periodic variation noted by Mendelkff are atomic volume (at. 18. volatility. Groups I to VII are each divided into subgroups. melting point. In spite of these imperfections. containing 2. the subgroup on the left side of the column is usually called the "a" subgroup. valence. The early tables were imperfect. oxygenlike elements 6a. similarities of a minor character exist between the members of the a subgroup and those of the b subgroup. labeled by the Roman numerals I to VIII and by the letter 0 are called groups. refractive index. and 18 (incomplete) elements. Mendelkff was able to predict the discovery of certain elements. thermal and electrical conductivity. In each case the order. halogens 7a. The table consists of a number of horizontal rows called periods.
. the approximate atomic weight of the most stable isotope is given in brackets.4tomic weights are bawd on the most recent values adopted by the International Union of Chemistry. (For artificially produced clemenb.:I Lanthanide &en8 . 1 1 2 2 88..) VII I VIII I 0 Period &ria.92 "' I ".
1950. but new discoveries and techniques in the field of physics opened the way to the understanding of atomic structure and the atomic number. Chapters 1. 9. Mass. W. Molecules i n Motion. 24. for example. Mass. Kinetic Theory of Gases. 1960. 1955. G. lithium. Beginning with the element following an inert gas. AddisonWesley.. T. . Reading. 8. that of the structure of atoms. 22.REFERENCES 17 . The gradual changes of properties from group to group suggested that some unit of atomic structure is added in successive elements until a certain portion of the structure is completed. 25. D. it was evident that the atomic weights alone could not explain the features of the table. New York: Harper Torchbooks. C. New York: PrenticeHall. AddisonWesley. 4. 6.. perhaps. 6. Principles of Physical Science. Elemenk of A h i c Physics. MELLOR. G. COWLING. 7. Modern Inorganic Chemistry. PRESENT. and potassium atoms should have similar properties. PEASLEE. 8. and so on. revised and edited by G. D. D. Speculations like thew introduced a new idea. Foundations of M o d e r n Physical Science. But before this idea could have any real physical meaning. 2. sodium. GENERAL J. Green and Co. experimental information bearing on the possible structure of atoms was needed.. F. 5. a new portion of the structure is started. Parkes. ROLLER.. Chapters 3. Since the properties of an element are the properties of its atoms. 1958. New York: McGrawHill. 1951. PHILLIPS. 1958. Reading. it was difficult to see why. 1957. ferent elements had nothing in common and were unrelated. T. London: Longmans. BONNER and M. London: Hutchinson and Co. H. a condition which occurs. Chapters 5. D. The periodicity of the properties of the elements stimulated speculation about the structures of the atoms of different elements. in the inert gases. HOLTON and D. The methods and techniques of classical chemistry could not supply this information. Chapters 21. R. 23.
and the lowest weight of the element is found in a compound which contains 22.112. (d) the exact atomic weight. The molecular weights of the five oxides listed in the above problem are 44.01.56 74. 1 1. From information contained in the Hadbook.63 46. (c) the valence. The Handbook of Chemistry and Phyaice. Plot the melting points of the solid elements as a function of position in the periodic table.0324. What is the compound? 6.T. .02. A certain element X forms several volatile compounds with chlorine.2% of oxygen by weight. 30. Show how the five oxides of nitrogen listed below illustrate the law of multiple proportions. The specific heat of the metal is 0. calculate the atomic volumes of the solid elements. Ohio. what is the atomic weight of platinum? 5. weighs 6. If the atomic weight of chlorine is 35.33 63. 76.571% of chlorine. (b) the approximate molecular weight of the compound.44 25. contains tables of various properties of the elements.16 69.457. (c) the approximate atomic weight of X.P.37 63.02. What is the molecular weight of the oxide? What is the combining weight of the new element? Where does the new element fit into the periodic system? What element would it be espected to resemble in chemical properties? What is the approximate atomic weight of the new element? . The specific heat of platinum a t room temperature is 0. a  .14 gm.54% of X and 77. Cleveland. the oxide contained 36. What periodic properties are observed in the two figures? 7. and plot the values obtained as a function of the position of the element in the periodic table. The new element formed a volatile oxide whose vapor density was 9.46 parts of chlorine.84 30. (e) the exact atomic weight of X. (d) the valence of X. 46.01.04 * 2.18 CHEMICAL FOUNDATIONS OF ATOMIC THEORY [CHAP. A metal forms a chloride which contains 65. find (a) the equivalent weight of X. Which element is X? What is the com~ound? num combined with 35. Find (a) the approximate atomic weight of the metal.69 times that of oxygen gaa a t the same temperature and pressure. published by the Chemical Rubber Company.46% of chlorine.67 36. respectively. Percentage composition Compound ~itrousoxide Nitric oxide Nitrous anhydride Nitrogen tetroxide Nitric anhydride Nitrogen Oxygen 36. (b) the equivalent weight. I n 1947 i t was announced that one of the longlived radioactive products of the h i o n of U235 was a new chemical element different from any other elements then known.02. 3.93 53. calculate the atomic weight of nitrogen from these data and the data of Problem 1. (f) the exact molecular weight of the compound. One liter of this compound a t N. and 108.
say 107. it served to indicate the order of magnitude of the electronic charge. and is denoted by F. The chemical action of a current of electricity is directly proportional to the absolute quantity of electricity which passes through the solution. From the work of Faraday on the electrolysis of aqueous solutions of chemical compounds. ELECTRONS. so that F = Nae. Stoney's calculation was based on the fact that one faraday of electric charge liberates the number of atoms in one gramatom of a univalent element. 1. Faraday's results implied that there is an elementary unit. or atom. This idea. of electricity. or in some. this number is just Avogadrols number No. The weights of the substances deposited by the passage of the same quantity of electricity are proportional to their chemical equivalents. Careful measurements have been made of the amount of electricity needed to liberate one equivalent weight of a chemical substance. He suggested the name electron for this quantity. an ion. and tried to calculate the magnitude of the electronic charge. did not seem to fit in with other electrical phenomena such as metallic conduction. and both Faraday and Maxwell hesitated to accept the idea of the atomic nature of electricity. The current in an electrolytic process results from the movement of the ions through the solution.3 X lo'' electrostatic unit of charge (esu). or 35. with its charge. or group of atoms. it appeared that electricity is also atomic in nature. Faraday discovered two laws of fundamental importance. From the known value of F and the rough value of No available a t the time. Stoney obtained: e = 0. 19 .88 gm of silver." namely. and he called the atom. one atom of hydrogen or one atom of any univalent substance.490 coul or one jaraday. at one of the electrodes. 2. 1 . This amount is 96. AND RADIATIONS 21 Faraday's laws of electrolysis and the electron. cases by a group of atoms. It was seen in the first chapter that the combining properties of the elements can be interpreted in terms of an atomic theory of matter. that quantity of electricity which must pass through a solution in order to liberate. In 1874 Stoney ventured the hypothesis that there is a "natural unit of electricity. He thought of this charge as carried by the atom. Although this is only & of the presently accepted value. where e represents the electronic charge.CHAPTER 2 ATOMS. negative ions moving toward the anode and positive ions toward the cathode.008 gm of hydrogen. however.46 gm of chlorine. Faraday inferred from his experimental results that one and the same amount of electricity is associated in the process of electrolysis with one atom of each of these substances.
2 x 101° esu for air a t 15OC and 760 rnm of mercury. This quantity can be written as ne. the amount of electricity carried by the univalent ions in one grammolecular weight of a salt in solution is one faraday: F = 96. but if ionizstion can be produced in the gas. Under normal conditions a ga4 is a poor conductor of electricity. discovered by Roentgen in 1895. As a result of these studies. ELECTRONS. 23 Cathode rays.. this ratio is 22 The conduction of electricity in gases. For hydrogen.20 ATOMS. This result also points to the existence of a unit electrical charge.490 coul = 2. If the molecules of the gas are awumed to be singly ionized. but they do give precise information about the ratio of the ionic charge to the mass of the atom with which it is associated in a given solution. which is very close to the value found for the ionized gas. The value of ne.~ of electricity represented by one electron. and e. the same charge as that carried by univalent ions in solution. it becomes a conductor. is the average charge on a gaseous ion.2 X 101° cm3 X 2. xrays. Following this discovery.36 X 10' cm3 = 2. 2 Faraday's experiments alone yield no certain information about the quantit. The volume occupied by one grammolecular weight of gas under these conditions is The total amount of electric charge associated with one grammolecular weight of gas is then 1. where n is the number of molecules of gas per cubic centimeter under some standard conditions of temperature and pressure. was found to be 1. Now. the mechanism of gaseous conduction and the nature of gaseous ions were investigated on an increased scale by many workers.90 X 10" esu. the value of the quantity of electric charge associated with one cubic centimeter of a gas could be calculated. When a gas is enclosed in a .83 X 10" esu. Theoretical and experimental studies were made of the mobility of gaseous ions and the coefficient of diffusion of these ions. provided a convenient means of producing ionization. they seem to carry. AND RADIATIONS [CHAP. The experimental proof of the independent existence of particles of negative electricity followed from the study of the conduction of electricity in rarefied gases. on the average.
231
CATHODE RAYS
21
FIG.21. Diagram o f a tube for the production of cathode rays.
glass tube with two electrodes to which a sufficiently high potential is applied, electricity flows through the gas, which becomes strongly illuminated. When the pressure of the gas is reduced below 0.1 mm of mercury the illumination gradually disappears. At very low pressures, of the order of lom4 or mm of mercury, a fluorescent glow appears on the wall of the glass tube opposite the cathode. The presence of the glow was correctly attributed by early workers to rays from the cathode, and these rays were called cathode rays. A schematic diagram of a tube for the production of cathode rays is shown in Fig. 21; C is the cathode, A is the anode, and T is a side tube through which the gas can be exhausted. The cathode rays were found to have several interesting properties. 1. The rays travel in straight lines. When screens S1 and S2,pierced with holes, are put into the tube, the glow is confined to a spot D on the end of the tube; or, if a body is placed in the path of the rays, it casts a shadow on the fluorescent part of the glass. From the geometrical relations it can be inferred that the rays travel in straight lines. 2. The rays can penetrate small thicknesses of matter. When a "window" of thin aluminum or gold foil is built into the end of the tube struck by the rays, the passage of the rays through the foil is shown by the presence of luminous blue streamers in the air on the far side of the foil. 3. The rays carry a negative electric charge. When the rays are caught in an insulated chamber connected to an electroscope, the electroscope becomes negatively charged. 4. The rays are deflected by an electrostatic field. When two parallel plate electrodes with their planes parallel to the peth of the rays are introduced into the cathoderay tube, the rays are deflected toward the positively charged plate. 5. The rays are deflected by a magnetic field. When an ordinary bar magnet is brought up to the cathoderay tube, the fluorescent spot opposite the cathode moves in one direction or another depending on which magnetic pole is presented to the rays. 6. The cathode rays carry considerable amounts of kinetic energy. When a metal obstacle is placed in the path of the rays, it can quickly become incandescent.

.

ATOMS, ELECTRONS,
AND RADIATIO?;~
[CHAP.
2
FIG. 22. The apparatus of J. J. Thomson for measuring the ratio e/m for cathode rays.
The only reasonable interpretation of these experimental facts is that the cathode rays consist of negatively charged particles. Thomson, in his fundamental experiments (1897), confirmed the view that the cathode rays consist of negatively charged particles and determined the ratio charge/mass = e/m for the particles. For these experiments, the tube of Fig. 21 was modified so that the rays could be made to pass through an electrostatic field. In the apparatus shown schematically in Fig. 22, a potential difference of a few thousand volts is maintained between the anode A and the cathode C. A narrow beam of rays from the cathode passes through a slit in the anode and through a second slit in the metal plug B. The end of the tube is coated on the inside with a fluorescent material and the point of impact of the cathoderay stream appears as a bright spot. Inside the tube are two metal plates D and D' between which a vertical electric field can be set up. By means of an external electromagnet, a magnetic field can be established, perpendicular to the plane of the diagram, within the region indicated by the dotted circle. If the electric field is upward (D' positively charged), and if there is no magnetic field, the cathoderay stream is deflected downward. The electric and magnetic deflections can be made to cancel each other by the proper adjustment of the electric and magnetic fields. Suppose that the cathode rays consist of negatively charged particles of charge e and mass m, and let E be the magnitude of the electrostatic field, and H that of the magnetic field. If the electric and magnetic fields are so adjusted that their effects cancel, and the particle stream passes through the tube without being deflected, then the electric force on the particles must be equal in magnitude and opposite in direction to the magnetic force. The electric force is given by eE, and the magnetic force by Hev, where v is the velocity of the particles."'* Then, eE = Hev, or
v =
'
*
E
H'
(21)
* Superscript numbers in parentheses refer to the references at the end of the chapter.
.
231
CATHODE RAYS
23
FIG. 23. The deflection of cathoderay particles in an electrostatic field.

and this experiment can be used to measure the velocity of the cathoderay particles. If El el and H are expressed in electromagnetic units, the velocity has the units centimeters per second. Next, the magnetic field is removed and the deflection of the particles caused by the electrostatic field is measured as shown in Fig. 23. In this diagram, the distance P I P z is the deflection caused by the electrostatic field as observed on the fluorescent screen; D and D' are the parallel plates as before, and 1 is the length of the plates. As the particles traverse the field. thev are subjected to a vertical acceleration in a direction parallel to that of the field during the time interval l/v, so that the velocity produced in the vertical direition is (eE/m) (l/v). The horizontal velocity of the particles is not changed by the field. On leaving the field, the velocity of the particles has the components v in the horizontal direction and (eE/m) (l/v) in the vertical direction. Hence, if P2 is the deflected position of the ray, P l P 2 / M P 1 is the ratio of the vertical and horizontal velocities :
P
In Eq. (22), the quantities 2, MP1 and E are known from the experimental setup; the deflection P I P z is measured, and v is known from the experiment in which the effects of the electric and magnetic fields cancelled each other. Thomson's experiments showed that the cathoderay particles move with very high speedsof the order of a tenth of the velocity of light. For e/m, he found the value 1.7 X lo7 emu/gm or, since 1 emu of charge = 3 x 101° esu, e/m = 5.1 x 10" esu/gm. This value was independent of the cathode materials which he used (aluminum, iron, and platinum), and of the gas (air, hydrogen, and carbon dioxide) in the discharge tube.

24
ATOMS, ELECTRONS,
AND RADIATIONS
[CHAP.
2
The value of e / m for the cathoderay particles is about 1800 times as large as that found for hydrogen ions in electrolysis. I f the cathoderay particles were assumed to have the same mass as the smallest atom, hydrogen, they would have to carry a charge 1800 times as large as that on a hydrogen ion. Furthermore, since the value of e l m is independent of the materials in the cathode tube, and materials with different ionic masses were used, it would be necessary to imagine some mechanism which would make the charge on the particles directly proportional to the mass. If, as an alternative, the charge on the particle is assumed to be the same as that on the hydrogen ion, then the mass of the particle is about 1/1800 of that of the hydrogen ion. The latter assumption seemed more reasonable to Thomson, who therefore considered that the cathoderay particles are a new kind of negative corpuscle. Later work proved that Thornson's conclusion was correct.
2 4 The electron: the determination of its charge. Thornson's experiments on cathode rays demonstrated the independent existence of very small negatively charged particles, but the assumptia that the charge on one of these particles is the same as the charge on a univalent ion in solution and on a gaseous ion in a conducting gas presented a basic problem. Although the assumption seemed logical, a proof of the equality of the charges was needed. The direct way to solve this problem would be to measure the charge on the particles. This direct measurement could not be made because of the great speed of the cathoderay particles, but other sources of similar negatively charged particles were available. Toward the end of the 19th century it was found that when ultraviolet light falls on certain metals, negatively charged particles with small velocities are emitted (the photoelectric effect). Thornson determined the value of e/m for these particles and found it to be practically the same as that for the cathoderay particles. The same result was obtained for the negatively charged particles emitted by an incandescent filament (the thermionic effect). In view of the constancy of e l m for the negatively charged particles produced in different ways, it was concluded that all of these particles are identical. There still remained, however, the problem of proving experimentally that the charge on the different particles (monovalent electrolytic and gaseous ions, photoelectric particles, thermionic particles) is the same. Several measurements of the charge were made around 1900, and their results may be summarized briefly as follows. For the average charge on an electrolytic ion, the value of 2 X lo'' esu was obtained, based on an improved but still approximate value of the Avogadro number; later work (19051908) gave a value of 4.25 x lo'' esu. The average charge on gaseous ions was found to lie between 3 X lo'' esu and 5 X lo'' em,
.


241
THE ELECTRON: THE DETERMINATION
OF ITS CHARGE
25



and similar values were found for the photoelectric particles. Experimental evidence of this kind pointed to the identity of the various negative electrical charges, and this view was gradually adopted. The term electrm was applied t o the negative particles themselves, rather than to the magnitude of the charge, and this usage has become general. The work on cathode rays and on the photoelectric and thermionic effects also indicated that electrms are a constituent part of all atoms, and the electron was regarded as a fundamental physical particle. In all of the experimental work discussed so far, either values of elm or of the average charge of the various particles were measured. Although it was difficult to argue against the conclusion that electrons exist and have a certain elementary unit of charge, the correctness of this conclusion had not been proved experimentally beyond any doubt. The final proof of the atomic or particulate nature of electricity and the first really precise determination of the value of the electronic charge was supplied by the work of Millikan (19091917). Millikan's exmriments will be discussed in some detail because of their fundamental place in modern physics. The earlier experiments on the determination of the electronic charge were based on the fact that electrically charged atoms (ions) act as nuclei for the condensation of water vapor. Water vapor was allowed to condense on ions, forming a cloud, and the total charge carried by the cloud was measured with an electrometer. The number of water drops in the cloud wqs calculated by weighing the water condensed from the cloud and dividing by the average weight of a single drop. This average weight could be found by measuring the rate of fall of the drops through the air, because the weight of a drop and its rate of fall were known to be related through Stokes' law of The total charge on the cloud divided by the number of drops gave the average value of the charge per drop. This method involves several assumptions: (1) that the number of ions is the same as the number of drops, (2) that Stokes' law, which had never been tested experimentally, is valid, and (3) that the drops are all alike and fall at a uniform rate not affected by evaporation. These assumptions were all questionable, and their validity had to be tested in careful experiments. The above method, used by Townsend (1897) and Thomson (1896), was improved by Wilson (1903), who formed water clouds between two parallel plates which could be connected to the terminals of a battery. The rate of fall of the top surface of the cloud was first determined under the influence of gravity. The plates were then charged so that the droplets were forced downward by both gravity and the electric field, and the new rate of fall was measured. When the two rates were compared, a value of the electronic charge could be obtained. In Wilson's method, it was not necessary to assume that the number of drops is equal to the number
26
ATOMS, ELECTRONS,
AND RADIATIONS
[CHAP.
2
of ions. Only the rate of fall of the top of the cloud was observed and, since the more heavily charged drops were driven down more rapidly by the field than the leas heavily charged ones, the actual experiments were made on thc least heavily charged drops. Wilson obtained the value 3.1 X lo'' esu, but the other assumptions remained, and there was considerable uncertainty in this result. Millikan improved Wilson's method by using oil drops, thus avoiding errors caused by evaporation. He also noticed that a single drop could be held stationary between the charged plates by adjusting the voltage on them so that the weight of the drop was just balanced by the electrostatic field between the plates. From the relationship between the forces, the value of the charge of the electron could be obtained. In addition, while working with balanced drops, Millikan noticed that occasionally a drop would start moving up or down in the electric field. Such a drop had evidently picked up an ion, either positive or negative, and it was possible to determine the change in the charge carried by a drop without regard to the original charge on the drop. The essential parts of Millikan's apparatus are shown in Fig. 2 4 ; P I and P2are two accurately parallel horizontal plates, 22 cm in diameter and 1.5 cm apart, housed in a large metal box to avoid air currents. Oil is sprayed in fine droplets from an atomizm above the upper plate and a few of the droplets fall through a small hole in the plate. A beam of light is directed between the plates, and a telescope is set up with its axis transverse to the light beam. The oil drops, illuminated by the light beam, look like tiny bright stars when viewed through the telescope, and move slowly under the combined influence of their weight, the buoyant force of the air, and the viscous force opposing their motion. The oil droplets in the spray from the atomizer are electrically charged, probably because of frictional effects; the charge is usually negative. Additional ions, both positive and negative, can be produced in the space between the plates by f the upper plate is positively charged ionizing the air by means of xrays. I to a potential of about 1000 volts, and the lower plate negatively charged, there is a uniform electric field between the plates, and there is an electric force on the droplet which affects its motion. When a spherical body falls freely in a viscous medium, it accelerates until a terminal velocity is reached such that the net downward force on the body equals the viscous force. The latter is proportional to the terminal velocity of the particle, and w = kv,, (23) where m is the apparent mass of the body, g the acceleration due to gravity, k a proportionality constant, and v, the terminal velocity of free fall. If,

.

.
.
241
THE
ELECTRON: THE
DETERMINATION OF ITS CHARGE
27
FIG. 24. Millikan's oil drop experiment.
in addition, the body carries a charge q and is subjected to the action of a n electric field of intensity El a new terminal velocity v~ is reached, given by qE  mg = k v ~ , (24) when the field acts counter to gravity. The ratio of the velocities is
VQ 
UE
mg . ~Emg
(25)
The charge q is then
It is evident that q can be evaluated if m and E are known and if the ratio (u, u E ) / v Q is observed experimentally. The main difficulty lies in the determination of m, the mass of the drop, which is too small to be measured directly.
+
28
ATOMS, ELECTRONS,
AND RADIATIONS
[CHAP.
2
I t is now possible to settle a fundamental question, namely, that of the atomic nature of electricity. I f there is a real unit charge e, then the charge q on an oil drop must be some integral multiple of e. Suppose that a measurement is made, such as that leading to Eq. (26), and that the charge on the drop then changes from q t,o q' because another ion is caught. Then q' is given by

Experimentally, ions supplied by the ionization of the air by xrays can be caught by an oil drop whiie under observation, and the behavior of a in the single drop carrying different charges can therefore be o b s e ~ e d same electric field E. The change in charge is

Millikan found, in many such experiments, that Aq was never smaller than a certain minimum value, and that all values of Aq were exact integral multiples of this value. By taking the least common divisor of all the values of Aq for a large number of changes of charge on the same drop, Millikan obtained the d k s t value of the change in charge, and he identified this value with Stoney's electron. Millikan's results provide direct proof that the elementary electric charge, the electron, is not an average charge, but rather that all electric charges on ions have either exactly the same f this value is value or else small, exact integral multiples of this value. I denoted by e, then q = ne, where n is a small integer. Millikan also showed, by studying drops carrying positive ions, that the elementary unit of positive charge haa the same magnitude as that of the negative electronic charge, and that all static charges both on conductors and insulators are built up of units of the electronic charge. The precise evaluation of the electronic charge depends, as has been noted, on the mass of the oil droplet. The masa can be obtained when the value of the proportionality factor k in Eq. (23) is known, and k is given by Stokes' law of fall. From the equations of hydrodynamics, Stokes derived a theoretical expression for the rate of fall of a spherical body in a viscous medium,

*
where a is the radius of the body, p the density of the body, p' the density
241
THE ELECTRON: THE DETERMINATION OF ITS CHARGE
29
of the medium, r, the viscosity of the medium, and v, the terminal or equilibrium rate of fall of the body. Equation (29) represents the balance between the gravitational and viscous forces on the body and can be applied to the fall of an oil drop in air. The value of v, is measured experimentally, and a is given by
RO
that
m
=
4?ra3(p p') = 
When the expression for m is inserted into Eq. (26) for the charge on an oil drop, the result is
In the last equation, en has been written for q because the charge on the oil drop has been proven to be an integral multiple of the electronic charge. If many observations of the quantity (v, vE) are made for changes of charge on the drop, the greatest common divisor, denoted by (v, OE)~, should correspond to the value of the electronic charge, denoted temporarily by el, so that
+
+
Millikan found that t,he value of e l was constant for large drops, but increased as the drops grew smaller. Careful studies then showed that StokesJ law in the form Eq. (29s) does not hold for the free fall of very small drops. It is necessary in this case to replace Eq. (29a) by
where p is the pressure of the air, a is again the radius of the drop, and b is an experimentally determined constant. The electronic charge is then given by the expression:
Millikan obtained the value e = 4.774 X 10lo esu, which was accepted for some years. Later determinations of e by means of indirect but highly precise methods gave results which disagreed with Milliksn's value. The discrepancy was caused by an error in the value of the viscosity of air used by Millikan. The use of more recent data for this parameter leads to the value e = 4.8036 f 0.0048 X 10lo esu, which agrees well with the values of e obtained by other methods. The constants e, F, and No, together with other quantities which will be met later, are called the fundamental constants of physics because of their great importance. The measurement of these constants is one of the basic problems of physics, and many highly precise methods have been developed for measuring them. The constants are often related, for example, F = Noe; and statistical methods have been developed for the analysis of the experimental results. This analysis yields best values of the constants which are consistent with the entire body of present knowl3 ' a partial list of these values is given in Appendix 11. The followedge; ' ing best values have been obtained for constants which have been discussed so far: e = (4.80286 f 0.00009) X 10lOesu, e/m = (1.75890 f 0.00002) X 107emu/gm = (5.27305 f 0.00007) X 10" esu/gm, m = (9.1083 f 0.0003) X gm, F = (2.89366 f 0.00003) X 1014 esu/gmequivalent weight = (96521.9 f 1.1) coul/gmequivalent weight. The Avogadro number is
The mass of an individual hydrogen ion, which is usually referred to as a proton, is MH = (1.67239 f 0.00004) X grn, and the ratio of the mass of the proton to that of the electron is
The values quoted for elm and m are correct provided that the speeds of the particles are very small compared with the speed of light. It will be seen in Chapter 6 that the mass of the electron increases with its speed; the value given above for the mass of the electron is called the rest mass and is often denoted by mo.
261
XRAYS
31


25 Positive rays. The fact that rays consisting of negatively charged particles are produced in a gas discharge tube suggested that rays of positively charged particles are also formed. Rays of this kind were discovered by Goldstein (1886), who observed that when the cathode of a discharge tube was pierced by narrow holes, streamers of light appeared behind it. He assumed that the luminosity was caused by rays which traveled in the opposite direction to the cathode rays and passed through the holes in the cathode. These rays, which were called canal rays, were deflected by electric and magnetic fields, and from the directions of the deflections it was concluded that the rays consist of positively charged particles. This result gave rise to the more generally used term positive rays. As in the case of the cathode rays, it was possible to measure the velocity and chargetomass ratio of the positiveray particles by means of the deflections in electric and magnetic fields. The results were quite different from those for the cathoderay particles. Both the velocities and the elm values were found to be much smaller than those for electrons. The value of elm was found to depend on the atomic weight of the gas in the discharge tube, and to decrease with increase in atomic weight. The largest value of elm was obtained for hydrogen and was very close to the value obtained in electrolysis. These results suggested that the positive rays are streams of positive ions produced by the ionization of atoms or molecules in the strong electric field of the discharge tube. The positive ion is formed by the removal of one or more electrons from the neutral atom, and the mass of the ion is practically the same as that of the atom. Hence, any measurement of the maas of positiveray particles should yield direct information about the masses of atoms of elements or molecules of compounds. Furthermore, this information should refer to the atoms or molecules individually rather than give average values for a large number of particles. I t is for this reason that the accurate measurement of the masses of positiveray particles is of great importance in modern physics. In fact, measurements of this kind resulted in the proof of the existence of isotopes and in the determination of isotopic weights. The subject of positive rays will be treated in greater detail in Chapter 9 in connection with the determination of isotopic masses and abundances.
26 Xrays. In 1895, Roentgen discovered xrays, which are produced when a beam of cathode rays strikes a solid target. He found that the operation of a cathoderay tube produced fluorescence in a screen covered with barium platinocyanide and placed a t some distance from the tube. The effect was traced to radiation coming from the walls of the cathoderay tube. In studying the properties of this new type of radiation, Roentgen noticed that if materials opaque to light were placed between
32
ATOMS, ELECTRONS, AND RADIATIONS
[CHAP.
2
the tube and the screen the intensity of the fluorescence decreased but did not disappear, showing that xrays can penetrate substances which are opaque to ordinary light. I t was also found that the xradiation can blacken a photographic plate and produce ionization in any gas through which it passes; the last property is used to measure the intensity of the radiation. The xrays were found to travel in straight lines from the source and could not be deflected by electric or magnetic fields, from which it was concluded that they are not charged particles. Later work showed that xrays can be reflected, refracted, and diffracted, and there is convincing evidence that these rays are an electromagnetic radiation like light but of much shorter wavelength. Xrays have turned out to be a valuable tool in atomic research and a great deal of information about atomic structure has been obtained from studies of the scattering and absorption of xrays by atoms, as will be shown in Chapter 4. 27 Radioactivity :alpha, beta, and gammarays. In 1896, Becquerel discovered that crystals of a uranium salt emitted rays which were similar to xrays in that they were highly penetrating, could affect a photographic plate, and induced electrical conductivity in gases. Becquerel's discovery was followed by the identification by the Curies (1898) of two other radioactive elements, polonium and radium. The activity of radium as measured by the intensity of the emitted rays was found to be more than a millioi times that of uranium.
.
.
.
L
acti
I
FIG.25. The three radiations from radioactive materiala and their paths in a magnetic field perpendicular to the plane of the diagram (schematic).
271
RADIOACTIVITY:
ALPHA,
BETA, AND GAMMARAYS
33
It can be shown, by means of experiments in magnetic fields, that there are three kinds of radiations from naturally occurring radioactive substances. In these experiments a collimated beam of rays is supplied by a small piece of radioactive material a t the bottom of a long groove in a lead block, as showu in Fig. 25. A photographic plate is placed some distance above the lead block, and the air is pumped out of the chamber. A weak magnetic field is applied at right angles to the plane of the diagram, and directed away from the reader. When the plate is developed, two distinct spots are found, one in the direct line of the groove in the lead block, and one deflected to the right. The undeviated spot is caused by neutral rays called ?rays; the spot to the right is caused by rays of negatively charged particles, called &rays. If the weak magnetic field is replaced by a strong one in the same direction, the undeviated spot is again found together with a spot deflected to the left; the latter is caused by positively charged particles called arays. In the strong field, the path of the 8rays is bent so much that it doesn't reach the plate. The need for a strong field indicated that the mass of the positively charged particles is much greater than that of the negatively charged particles. The different electrical properties of the radiations are shown schematically in Fig. 25. I t has been shown that all three radiations are not emitted simultaneously by all radioactive substances. Some elements emit arays, others emit Brays, while %rays sometimes accompany one and sometimes the other. The activity of a given material is not affected by any simple physical or chemical process, such as change in temperature or chemical combination with nonradioactive substances. These results, together with Mme. Curie's demonstration that the activity of any uranium salt is directly proportional to the quantity of uranium in the salt, showed that radioactivity is an atomic phenomenon. The 8rays, being mgatively charged particles, could be deflected in electric and magnetic fields, and their velocities and e/m values were measured. The results, although somewhat more complicated than in the case of cathode rays, showed that the &rays consist of swiftly moving electrons and differ from cathode rays only in velocity. The @raysfrom radium were found to have velocities up to within a few percent of that of light, while cathode rays usually have velocities considerably less than half that of light. The elm values of the bparticles were of the same order of magnitude as those of the cathoderay particles, but decreased with increasing velocity of the particles. Even before the Millikan oildrop experiment, it was considered that electric charge is atomic in nature and should not vary with velocity, and it was concluded that the mass of a 8particle increases with the speed with which the particle is moving. The value of elm for the slower &particles was found to be almost identical with that obtained for cathoderay particles. The &rays are much more
ATOMS, ELECl'RONS,
AXD RADIATIONS
[CHAP.
2
penetrating than arays, being able to pass through one millimeter of aluminum, whereas the arays are completely stopped by 0.006 cm of aluminum or by a sheet of ordinary writing paper. The ionization density produced in a gas by &rays is much less intense, however, than that produced by arays. Alpharays, because of their positive electrical charge, can be deflected by electric and magnetic fields, and their velocities and elm values determined in the usual way. The velocities range from about 1.4 X lo9 to 2.2 X 10' cm/sec, or less than fa of the velocity of light. The value of e/m.is close to 4820 emu/gm = 1.45 X 1014 esu/gm, or just onehalf of the value, 2.87 X 1014 esu/gm, for the singly charged hydrogen ion. I t is evident that if the aparticle had the same charge as the univalent hydrogen ion, its mass would be twice that of a hydrogen ion or atom. If the orparticle carried a charge twice that of the hydrogen ion, its mass would be four times as great and would correspond closely to that of the helium atom, which has an atomic weight of 4 units. The exact nature of the aparticle could not be determined until either its charge or mass could be measured separately. Rutherford showed, in an ingenious experiment, that when a radioactive substance emits arays, helium is produced. A small quantity of radon, a gaseous radioactive element, was sealed in a small glass tube with extremely thin walls, so thin that the arays emitted could pass through them into an outer glass tube, while the radon was kept in. The outer tube was provided with electrodes, and the spectrum produced on passing a discharge was observed a t regular intervals. After a few days it was found that the spectral lines of helium appeared; control experiments showed that ordinary helium gas could not penetrate the thinwalled tube, proving that the helium was produced from the arays. Thus, the aparticles must be helium ions and, since their m a s is 4 times that of a hydrogen ion, the charge on an aparticle must be twice that on the univalent hydrogen ion. The aparticle is, therefore, a doubly ionized helium atom. The charge on an aparticle was also determined separately by counting the number of aparticles emitted in a known time interval from a known weight of radioactive material, and a t the same time measuring the total positive charge emitted. I t will be shown in the next section that methods have been devised for counting individual aparticles. Two independent experiments, in which different counting methods were used, gave 9.3 x esu and 9.58 X 10lo esu, respectively, for the aparticle charge, and these values are very close t o twice the electronic charge. The 7rays, which are not deflected by electric or magnetic fields, have been shown to consist of electromagnetic waves, and are therefore similar in nature to light and xrays. The w a y s are roughly 10 to 100 times as penetrating as the @rays,but produce less ionization than do the &rays.

'
.
281
THE DETECTION A N D MEASUREMENT
OF RADIATION
35

The ionization produced by a, 8, and 7rays is roughly in the order 10,000, 100, and 1, a useful rule of thumb. The detailed chemical study of the radioactive substances has shown that each such substance is an element, and that its radioactivity is caused by a spontaneous disintegration of radioactive atoms into other atoms. This process occurs according to definite laws. The transformation of one atom into another atom with the emission of either an aparticle or a 8particle, led to the idea that the atoms of the chemical elements are complex structures built of smaller particles. The study of atomic structure became an important branch of physics with the main object of discovering the laws according to which atoms are built up. The discovery of natural radioactivity, together with the discovery of xrays and the proof of the independent existence of the electron, provided new and powerful methods for attacking the problem of atomic structure.
28 The detection and measurement of radiation. The study of radioactivity and the successful use of radiations as research tools or for other purposes depend on the quantitative detection and measurement of radiation. The quantities most often needed are the number of particles (e.g., electrons, protons, aparticles) arriving a t a detector per unit time and their energies. When a charged particle passes through matter, it causes excitation and ionization of the molecules of the material. This ionization is the basis of nearly all of the instruments used for the detection of such particles and the measurement of their energies. Similar instruments can be used for uncharged radiations (xrays or 7rays) because these impart energy to charged particles, which then cause ionization. The different types of instrument differ in the material within which the ionization is produced and in the way in which it is observed or measured. Many instruments are based on the production of ionization in a gas. I t is necessary, in this case, to separate and collect the positive and negative ions because, if these remained close together, they would recombine and no electrical information as to their presence could be obtained. The separation and collection of the ions requires the presence of an electrostatic field, and different instruments result depending on whether the field is small, large, or intermediate in magnitude. The ionization may also be produced in a liquid or in a crystal. When it is produced in a gas supersaturated with vapor or in a photographic emulsion, the tracks of the particles can be made visible. When particles strike certain liquid or solid materials called phosphors, which have the property of luminescence, part of the energy used up in molecular excitation and ionization is reemitted as visible or ultraviolet light. Sometimes this light can be observed by eye, or it may be detected by means of more sensitive devices.


but consists of a large number of individual flashes. barium platinocyanide. primarily the alkali halides. ELECTRONS. This method was especially useful for counting aparticles in the presence of other radiations. Light from the phosphor strikes the cathode. can produce luminescence in zinc sulphide. The visual scintillation counter has the disadvantages that the flashes are quite weak and can only be seen in a darkened room. The photomultiplier tube. and solutions of organic compounds such as terphenyl dissolved in xylene. the method was used in many of the basic experiments of nuclear physics between 1908 and about 1935 when it was replaced by electrical methods. because the zinc sulfide screen is comparatively insensitive to & and rrays. for example. crystalline organic materials such as naphthalene. Nevertheless. which replaces the microscope and observer.and ?rays. The modern scintillation counter can detect and record many millions of flashes per second and can be used with intense radiations. when produced by aparticles. Careful experiments have shown that each aparticle produces one scintillation. that radiations. but with difficulty. Since 1947. and good precision can be obtained. AND RADIATIONS [CHAP. The tube has several electrodes called dgnodes to which progresssively higher  . on a slip of glass. 2 ** We shall consider some of the commonly used detection methods because of their great importance in nuclear physics. The scintillatiunmethod. sodium or potassium iodide activated with thallium.  . A screen can be prepared by dusting small crystals of zinc sulfide. which is usually made of antimony and cesium. The counting can be done with a microscope with a magnification of about 30 times. and diamond. and the development of highly efficient photomultiplier tubes. Such treatments are given in some of the references a t the end of the chapter. and ejects electrons (photoelectric effect). the use of scintillation counters has been revived and greatly extended because of the discovery of new phosphors which are also sensitive to 8. which can be seen under a magnifying glass. The phosphors include: inorganic salts. so that the number of aparticles which fall on a detecting screen per unit time is given directly by the number of scintillations counted per unit time. converts the scintillations from the phosphor into amplified electrical pulses which can be counted or otherwise analysed with suitable electronic equipment. It was found. This luminescence. 26. and that visual counting is limited physiologically to about 60 scintillations/min and is very tedious. containing a very small amount of copper impurity. is not uniform. containing small amounts of impurities as activators for luminescence. although a detailed treatment is beyond the scope of this book. about 1900. especially aparticles.36 ATOMS. anthracene and stilbene. A schematic diagram of a photomultiplier tube is shown in Fig.
The output current.27. and strike the dynode. There may be as many as ten secondary electrons for each electron which strikes the dynode. 27.  potentials are applied. A schematic diagram of a scintillation detector is shown in Fig. The accelerated electrons impart enough energy to electrons in the dynode to eject some of them. Photomultiplier tube (schematic). . The photoelectrons are accelerated in the electrostatic field between the cathode and the first dynode. or pulse. The electroscope. which is at a positive potential relative to that of the cathode. Electronic systems have also been developed which measure the energy of the incident particles. the resulting instrument is called a scintillation spectrometer. at the anode may be more than a million times as great as the current originally emitted from the cathode.26.Anode Paths o f electrofs +200v I +400v I +600v I +800v I t1000v Cathode +LU)"/ +300v + m u +700v + m v Dynode FIG. Schematic diagram o f a scintillation detector. and the pulses are fed to an electronic system where they are counted. Each particle incident on the phosphor produces a pulse. This process is repeated.Light pipe Highvoltage power supply FIG.281 THE DETECTION AND MEASUREMENT OF RADIATION 37 I . It has been mentioned that many of the instruments used for the measurement of radiation depend for their operation on the Light reflector 1 . and the electron current is amplified as the electrons are accelerated from dynode to dynode.
28. When the two elements are connected across a battery. This instrument is a condenser. 28. lead and the sample is placed outside the instrument.A T O M S . One of the first and simplest instruments used in work on radioactivity is the electroscope. ionization of a gas. &. Hence ture and pressure is directly proportional to the number o the extent of the ionization is a direct measure of the amount of radioactive substance. an example of which is shown in Fig. Since the 7rays are much weaker ionizing agents than the brays. the collapse of the leaves is caused almost entirely by the &rays. I f the radioactive substance emits a. one of whose elements is the outside case. since the arays are by far the strongest ionizing agents. The net charge on the elements and the potential daerence between them decrease. they all contribute to the ionization but. ELECTRONS. the radioactive sample must be surrounded by enough material (2 or 3 mm of lead) to absorb completely the a. B'. Electroscopes have been developed which are much more  . To measure 7ray activity alone. or surrounded by. and the leaves collapse to the position B. they become oppositely charged and the gold leaves (A and A') separate because of the electrostatic f the battery is removed and the gas in repulsion of their like charges. A b r a y electroscope is made by placing an aluminum sheet about 0. The rate of collapse of the gold leaves is a measure of the rate a t which the gas in the chamber is ionized and hence of the intensity of the ionizing radiation. A given sample of radioactive substance emits a definite number of a. The number of ions which these rays can produce in air or in another gas at a fixed temperaf particles.and &rays. gaseous ions migrate to the oppositely charged elements of the condenser. ioactive ce Diagram o f an aray electroscope. AND RADIATIONS Fro. . and the ionization observed is then caused by the /3. the other element ends in a pair of gold leaves.and %rays. and 7rays. I the condenser is ionized by a radioactive substance.01 cm thick over the active material to absorb all of the arays. &. or ?rays per unit time. . Sometimes the entire electroscope is built of. the instrument is really an aray electroscope.
The differences in the three systems can be explained with the aid of Fig. 29. 29. As the voltage is increased. The movement of the fiber is observed through a microscope with a scale in the eyepiece. or how many electrons reach the central electrode as the applied voltage is varied.  sensitive t. will reach the central electrode.han the one just described. At a voltage VI of. The chamber is filled with a gas a t a pressure of one atmosphere or less.THE DETECTION APCD MEASUREMENT OF RADIATION Insulator Central electrode Fro. and GeigerMueller counter. . and no charge will appear on the capacitor. less than 10. We suppose that some ionization occurs in the gas because of the passage of a charged particle. the Lauritsen electroscope has as its moving element a very fine metalcoated quartz fiber. Each of these three detectors is based on the production of ionization in a gas and the separation and collection of the ions by means of an electrostatic field. For convenience. there is competitionbetweenthe loss of ion pairs by recombination and the removal of ions by collection on the electrodes. the ions will recombine. how many ion pairs are collected. I f there is no voltage across the electrodes. For example. Electrical imtruments: imLization chamber. we assume first that 10 ion pairs are formed. which shows a cylindrical conducting chamber containing a central conducting electrode located on the axis of the chamber and insulated from it. say to a few volts. for a given initial ionization. the central electrode is at a positive potential relative t o that of the chamber wall. Some electrons. Schematic diagram for pulse operation of a gwfilled chamber. A voltage V is applied between the wall and the central electrode through the resistance R shunted by the capacitor C. Each ion pair consists of a positive ion and an electron and we want to know. 210 where curves of total ion collection are plotted as functions of applied voltage. The discussion is based on that of Wilkinson (see general reference at end of chapter). corresponding to curve a of Fig. the logarithm to the base 10 of the number of ion pairs n has been used as the ordinate. proportional counter. To make the discussion less abstract.
210. ELECl'RONS. . proportional. 2 ha. AND RADIATIONS [CHAP.40 ATOMS. Curves of total ion collection versus applied voltage to illustrate ionisation. and GeigerMueller regions of operation of electrical radiation detection instruments.
say lo5 ion pairs. the region above being that of the continuous discharge. I t is similar to curves a and b. the number of ion pairs collected is then proportional to the initial ionization. Above V4 the charge collected becomes independent of the ionization initiating it. The discharge with 100 initial electrons is affected before the one with 10 initial electrons and increases less rapidly than the latter. The region between V1 and V2. and curves a and b continue parallel with a difference between them of one unit on the log n scale. a t Vi instead of a t V3. If 100 ion pairs are formed initially. Between V 2 and V3. I t ends a t a voltage V5 where the discharge tends to propagate itself indefinitely. The behavior of the two curves above Vp is interesting.281 THE DETECTION AND MEASUREMENT OF RADIATIOX 41   perhaps. as more electrons produce avalanches. it is parallel to curve a in the ionization chamber region and is separated from it by one unit on the log n scale. curves a and b approach each other and eventually meet at V4. the latter begin to interact with one another. log n reaches the value unity. V5 marks the end of the useful voltage scale. avalanche. not being affected by the presence of the other electrons. If the initial ionization is large. and. n increases above 10 because of a phenomenon called gas multiplication or gas ampli$cation. the extent of that region depends on the initial ionization. curve b results. the proportional counter region ends sooner. . The region between V g and V4 is the region of limited proportionality. This is the region of proportiaal counter operation. and parallel to them between V2 and VQ. each electron acts independently and gives its owr. in which the number of ion pairs collected is independent of the applied voltage and the curve is horizontal. The gas multiplication increases the total number of ions to a value that is limited by the characteristics of the chamber and the external circuit. For some range of voltages. Thus. 10 volts the loss of ions by recombination becomes negligible. up to V3. Above V3. the positiveion space charge of one avalanche inhibits the development of the next avalanche. and 10 electrons reach the central electrode. n stays constant a t 10 until a voltage V2 is reached which may be some tens or hundreds of volts depending on the conditions of the experiment. The initial 100 electrons always yield 10 times more final electrons than do the 10 initial electrons. is called the ionizatiun chamber region. The electrons @eased in the primary ionization acquire enough energy to produce additional ionization when they collide with gas molecules. When the voltage is increased above V2. The region above V4 is the region of GeigerAlueller counter operation. curve c is obtained. the gas multiplication effect continues to increase very rapidly. and the curves a and b become identical. and n increases roughly exponentially with V. Each initial electron produces a small "avalanche" of electrons with most of these secondary electrons liberated close to the central electrode. As V is increased.
The use of this counter permits both the counting and the energy determination of particles which do not produce enough ions to yield a detectable pulse in the region VL to VP. three detection instruments have been developed. The numbers in Fig. is characterized by a gas multiplication independent of the number of initial electrons. The electrodes of an ionization chamber may also be parallel plates. I Highvoltage insulator I voltage couecthg I Guard ring Collecting electrode Guard ring ". The proportional counter.g.211. Subject to certain conditions. The proportional counter.42 ATOMS. 2. 1.    Housing r \ Highvoltage electrode . therefore. 211. and filled with gas to a pressure of a few centimeters of Hg (e. 2 As a result of the behavior of the ions of the gas in the electrostatic field. Nuclear RadMEion Deledion. ELECTRONS. an application for which ionization chambers are not sensitive enough. McGrawH i l l . of all the electrons initially liberated by the passage of the particle. AND RADIATIONS [CHAP. without gas amplification.01 cm in radius. it will give a pulse proportional to the number of these electrons. 29 would be appropriate for a counter with an outer cylinder 1 cm in radius. offers advantages for pulsetype measureaents of beta radiation. which operates in the voltage region Vz to V3. Hence.Lowleakage insulator To electrometer FIG. Price. argon a t about 6 cm of Hg plus a little alcohol). approximately. A schematic diagram of an instrument of this type is shown in Fig.) . Schematic diagram of a parallel plate ionization chamber (Reprinted by permission from W.1958. is characterized by complete collection.. The ionizalion chamber. J. which operates a t voltages in the range V1 t o VZ. although gas multiplication is utilized. The counter length would be about 10 cm. a wire central electrode 0. Figure 29 is a schematic diagram of a chamber whicb could be used as an ionization chamber between 10 volts and 200 volts. the pulse is always proportional to the initial ionization.
counter. The tube contains dry air a t atmospheric pressure. The principle of the GeigerMueller counter was first used by Rutherford and Geiger in 1908 to count aparticles. Diagram of a Geiger point counter. a mixture of 90% argon and 10% ethyl alcohol is suitable. is characterized by the spread of the discharge throughout the entire length of the counter. A potential difference. which may be between 800 and 2000 volts. It is especially useful for the counting of lightly ionizing particles such as Fparticles or Yrays. they are use'd mainly for 0and Yrays. I t has a smooth fine point P at the end of a fine wire supported inside a tube T by an insulating plug. is applied to make the t u h negative with respect to the wire. Diagram of a GeigerMueller counter tube.213. and instruments like it are still used to detect aparticles. . and Yrays.g. 212. is shown schematically in Fig. The GM counter usually consists of a fine wire (e.2 4 THE DETECTION AND MEASUREMENT OF RADIATION 43 FIG. and the particles enter through a thin foil window W. FIG. 212. 6. or GM. resulting in a pulse size independent of the initial ionization.. The GeigerMueller counter. as well as xrays. 213. or a metal cylinder may be supported inside a glass tube as in Fig. GeigerMueller counters have been designed for the measurement of a.   3. The tube may be a metal such as copper. also known as Geiger. In the form described. which operates in the voltage region V4 to V5. The instrument that they used. with the object of determining their charge. the Geiger point counter. tungsten) mounted along the axis of a tube which contains a gas a t a pressure of about 2 to 10 cm Hg. in part because of the difficulty of making tubes with windows thin enough to be penetrated by arays.
The cooling is more than sufficient to overcome the effect of the increase in volume. a particle causes a small amount of ionization. There are two types of cloud chamber differing in the method used for obtaining supersaturation.44 ATOMS. For large rates. For very slow rates (less than about one per second). is an electric device which selects precisely every mth pulse and passes it on to the recorder. is made to expand by the quick motion of a piston. AND RADIATIONS (CHAP. either multiplied in number or not. is based on the discovery by C. The response of an ionization chamber can be made sufficiently rapid so that aparticles arriving a t intervals of sec or less can be detected and the counting must then be made very efficient. Other electronic devices are also available for measuring the magnitude of the pulse. This instrument. The detector then forms part of a circuit with appropriate electronic instrumentation. T. Rutherford and Geiger showed that each aparticle that strikes the screen causes one scintillation. A scaling unit. the space between the electrodes in the ionization chamber. the operation of this counter tube is similar to that of the GeigerMueller counter. it is often desired not only to record the occurrence of a pulse but. and the air becomes supersaturated with water vapor. The expansion is adiabatic and lowers the temperature. are collected and produce a voltage pulse which may be as small as 10 microvolts. ELECTRONS. saturated with vapor. to sort pulses according to their size (by means of an electronic discriminator circuit) or to sort them according to the time intervals during which they arrive (by means of an electronic timing circuit). the pulses can be made to operate a relaytype recorder. By comparing the activity of a radioactive substance as measured with a Geiger point counter and with a zinc sulfide screen. '4' the gas. Wilson (1897) that ions act as nuclei for the condensation of supersaturated water vapor. Thus. . In the expansion chamber. or scaler. per unit time. mechanical devices either fail completely or miss many counts. also require electronic instrumentation. and the :    . which gives the energy of the incident particle. R. Scintillation detectors. first used in 1912.scaler. The cloud chamber. also. If an ionizing ray enters the chamber. the ions formed act as nuclei for the condensation of the vapor. The ions. The amplified voltage pulses must then be counted in some way so that their rate can be measured. of course. One of the most important instruments for basic research on radiations is the cloud chamber. The number of pulses indicated on the counter per unit time multiplied by the appropriate factor of the scaler gives the number of aparticles entering. The scaling ratio is usually a power of 2 or 10 and very high counting rates can be achieved with the aid of a . 2 In other respects. a fact which increases the usefulness of the screen for quantitative measurements. In the electrical instruments discussed. An electronic pulse amplifier accepts these small voltages and amplifies them to a level usually in the range of 5 to 50 volts.
in the presence of ionizing radiation. In most expansion chambers. in order to obtain large numbers of photographs. When the piston is pulled down. The tracks can be illuminated by light L from the side and viewed or photographed through the window. A simplified diagram to indicate the principle of the expansion chamber is shown in Fig. . the gasvapor mixture is air with water. A cylinder of glass C is closed a t one end by a glass window W and a t the other end by a metal piston P to form the chamber. When a cloud chamber is placed between the pole pieces of an electromagnet. A simple expansion cloud chamber. 214. I f two stereoscopic pictures are taken simultaneously. the sign of the charge and the magnitude of the momentum of the particle can be determined. a t atmospheric pressure. fog tracks are formed. The piston is returned to its original position and the chamber is ready for another burst of radiation. The ions can then be removed by means of an electric field between the piston and the metal ring R. A small amount of water in the chamber keeps the air saturated. *   . automatic arrangements have been made so that the expansion can be repeated rapidly and photographs taken continuously on motion picture film. From the curvature of the path of a particle in the magnetic field. For example. to photograph the actual paths of ionizing radiations.the path of the particle in space can be reconstructed. the air becomes supersaturated as described above and. Many modifications of the original cloud chamber have been made.214. The Wilson cloud chamber makes it possible to study the behavior of individual atoms. it is possible to distinguish between positively and negatively charged particles.s81 THE DETECTION AND MEASUREMENT OF RADIATION 45 FIG. and to analyze a t leisure the complicated interactions which may take place between charged particles and individual atoms. or argon with ethyl alcohol. path of the ray appears as a thin track of fog.
where it becomes supersaturated. beaded straight tracks. Thus.215. and Ellii. which is sensitive only for a short time h t e ~ a lafter the expansion. produce narrow. ELECTRONS. The vapor is usually methyl or ethyl alcohol. The latter. and diffuses downward continuously through a region in which a steady vertical temperature gradient is maintained by cooling the bottom of the chamber. Slow electrons produce narrow. straight line tracks with an occasional sharp. gives the diffusion chamber advantages over the expansion chamber. . This property. the warm end. Chadwick. A schematic diagram of a diffusion chamber is shown in Fig. to a cold region. slow particles such as aparticles produce broad. both light and heavy. Tracks of aparticles in air. Radialha from Radioactive Substances.ATOMS. An ionizing particle traveling through the emulsion of a photographic plate leaves a track containing a number of developable silver bromide grains. and hydrogen from 1020 atmospheres. Tracks of aparticles in air are shown in Fig. Photographic emu2sim.1 Different particles produce different types of tracks. the bottom photograph is an enlargement of the end portion of a track showing two deflections caused by collisions with atoms of the air. The top photograph shows a complete track. AND RADIATIONS FIG.   * . The diffusion cloud chamber '5' operates by the diffusion of a condensable vapor from a warm region. densely packed. 215. unlike the expansion chamber. small angle bend. 216. where it is not saturated. The vapor is introduced a t the top. [Reprinted by permission from Rutherford. whiie fast particles. Cambridge University Press (Macmillan Co. called nuclear . however. heavy. Special photographic emulsions. Air is satisfactory for the gas a t pressures from 20 cm H g t o 4 atmospheres. beaded. together with its relative mechanical simplicity. The diffusion chamber is continuously sensitive. is still used in a large portion of serious cloudchamber work because of the greater experience with it and because of certain other desirable properties which it has. the bottom photograph is an enlargement of the end portion of a track showing two deflections caused by tollisions with atoms of the air. especially near the end of the track. The top photograph is that of the complete track of an aparticle from radon.) 1930. tortuous tracks.
in the grain size. and a variety of information can be obtained from the study of the tracks. and in the thickness of the emulsion. which may be as much as four times as great as in photographic plates. and its sensitivity is permanent rather than restricted to infrequently repeated short intervals. emulsions. and energy of the particles. An emulsion is limited in the chemical elements which it can contain. other instruments such as the bubble counter and the Cerenkov detector have been developed for use with particles of very high energy (they are discussed in the book by Price listed in the references). On the other hand.216. charge. and study of the detailed structure of the tracks yields information about the mass. records the path of the particle. the cloudchamber expansions can be controlled by means of a counter and can be made to record specific events.281 THE DETECTION AND MEASUREMENT O F RADIATION 47 Pm cylinder Aleti11 plntc FIG. Finally. Schematic diagram of a diffusion cloud chamber. Like the cloud chamber. . the photographic plate. have been developed. The photographic emulsion offers advantages over the cloud chamber in that the emulsion is solid so that the tracks are short. upon development. They are distinguished from optical emulsions in their high giver bromide content. Counting of the individual paths gives a measure of the number of particles entering the plate.
Horo and S. 1953. W. 267. S. and Heat. RADIATION DETECTION H. Chapter 9. 1958. P. SHARPE. Photons. AND RADIATIONS [CHAP. Frisch. N. CROWE. STAUB. A. Cambridge: University Press. CROWTHER. Protons. KORFF. SNOWDEN. W. p. Reading. M. 1953. New York: Interscience. . K. Electron and Nuclear Counters. J. 363 (1955). 3. "The Diffusion Cloud Chamber" Progress in Nuclear Physics. 4. Chicago: University of Chicago Press.and J. WILKINSON. Electrons (+ and )." 1955. G. Electrons. KORFF. 1951. B. and J. New York: Van Nostrand. 8th ed. K. also N. Meaotrona. The "Particles" o f Modern Physics. 1949. "Detection Methods. and Comic Rays. W. R. 1950.48 ATOMS." Rws. M. I.The Principles o f CloudGhamber Technique. 1955. on the Wilson Cloud Chamber and its Applications in Physics. PARTICULAR 1.O. 1947.: AddisonWesley. 225 (1946). COHEN. Maas.and J. gen. M. A. New York: Van Nostrand. W. R. Nuclear Radidion Detecturs. DUMOND. M. ed. Wave Motion. Reading. J. Electron and Nuckar Physics. COHEN. R.. Waves and Particles. A. 27. Fundamentcrl Conatats o f Physics. MILLIKAN. cit. E. London: Pergamon Press. Phys. J. New York: Academic Press. E. H.Scintillation Counters. 1953. LAYTON. COHEN. ELECTRONS. FRENCH. Das GUPTAand S. ed. Mod. DUXOND. New York: Wiley. B. S. Cambridge: "A Report University Press. London: Pergamon Press.Nuclear Radiation Detection. A.. Mod. 1948. also E. 2. London: Edward Arnold. W. I ? . R O L L E ~ . ref. 1958. New York: Wiley. New York: McGrawHill. CROWE. R. D. H. 1955. 1958. J. 5. J. Ions. STRANATHAN. Ionization Chambers and Counters. Vol. WILSON. 1944. . Maas. CONSTANT. T b e t i c d Phyaies: Thermodynamics. J. Chapter 15. Mechanics. Electromaqnetian. D. R. J. New York: Wiley. PRICE. and Ionizing Radaktions. Segrb. 3. Philadelphia: Blakiston. op. E. F. A. 2 QENERAL J. Part I. 1958. London: Methuen.: AddisonWesley. M. GHOSH. Phys. 2nd ed. 3rd ed. DUMOND. SEARS. S. W. BIRKS. 1957. Vol. M. "Analyeis of Variance of the 1952 Data on the Atomic Constants and a New Adjustment." Experimental N u c h r Physics. Principles o f Modern Physics. J. Rws. K. 18.
75 cm. distance between plates = 2. electric field of magnitude E is produced in the space between parallel plates which are closely spaced! concentric. but only rays with a particular speed are desired. When cathode rays with a speed v move in a direction perpendicular to a uniform magnetic field of strength H. I n experiments with cathode rays. 3. and 1. 2. suppose that the elestric field is turned off instead of the magnetic field. on striking the screen. Show that the kinetic energy of a particle of . Then calculate (a) the number of grams of each element deposited by 1 amphr of charge. = 32. where E/H = v. the cathode rays passed between condenser plates 1 cm apart. the source often emits rays with a wide range of speeds.0 cm. 2. the potential across the plates was changed to 1350 volts. to permit only the passage of rays with a certain desired speed? I n the apparatus of Problem 2. (c) the charge in esu associated with 1 gm of each element. Show that the value of e/m can be obtained from the equation e/m = E/H2R.. the rays were displaced 19. indium. nickel.0 cm. . What is the radius of the path of the cathode rays? 5. What magnetic field was needed to keep the beam from being deflected? What displacement of the beam was observed when the magnetic field was turned off? 4. The effect of the electric field was just balanced by a magnetic field of 28 gauss.03731 chemical equivalents. 2. radially directed. In the experiment of Problem 2. suppose that the potential across the plates is fixed and that the magnetic field strength can be varied. how would you obtain a monoenergetic beam of rays with a speed of 2. Another measurement of e/m for cathode rays was made by means of two experiments. When the potential across the plates was 1100 volts and the magnetic field strength was 12. Take 1 faraday as 96. a Thomson apparatus with the following dimensions was used: distance from condenser plates to screen length of plates = 8. (b) the number of coulombs of charge needed to deposit 1 mg of each element.5 cm. With the apparatus of Problem 2. 2. cylindrical sections of mean radius R. mercury. In the first.500 coul and the valences of chromium.1.0 X lo9 cm/sec? 7. the rays were bent into a path of 12. iron. the cathode rays passed through the apparatus without being deflected. Show first that 1 amphr corresponds to 0. i. with a potential of 730 volts across them. What was the speed of the rays? What value was obtained for elm? 3. copper.e. from their undeflected path. and silver as 6. their path is a circle of radius R given by the relationship Hev = mv2/R.50 gauss. How could the Thomson apparatus be modified to serve as a velocity filter. In the second experiment.00 cm radius by a magnetic field of 12. 3. When the magnetic field was turned off. What was the speed of the rays? What was the value obtained for e/m? 6. A uniform. respectively. . In an attempt to measure the value of the chargetomass ratio of cathode rays.50 gauss.
radius of oil drop = 2. with the aid of Eq. Millikan observed the motion of an oil drop in a constant electric field. As the drop captured or lost ions. (28).34.7. Calculate the rate of fall. An oil drop of radius 2. + .85. Why does such a system provide an energy filter for charged particles? If the mean radius is 10 cm. the viscosity of air is 1. An oildrop measurement of the charge on the electron was done under the following conditions: distance between the condenser plates = 1.34. 2 . density of oil = 0. charge q. Eight successive time intervals noted were. Explain.08571 cm/sec. In one of his experiments.0. mass M . changed.9199 gm/cm3 falls in air a t 23OC and 76 cm Hg.005480 cm/sec.16.0012 gm/cm3. first X from the uncorrected Stokes formula. how these data show that the electric charge on an ion is an esact integral multiple of an elementary charge.45. ELECTRONS. and 21. Under these conditions. then from the corrected formula. the time needed for i t to traverse a given distance. the constant b is 0.21. in the latter. movlng in the space between the plates is 3qER. 0. 9. pressure = 76 cm Hg. potential difference across the plates = 5085 volts.5222 cm.60 cm.76 X lo4 cm and density 0.7.34. what field strength is needed to permit a beam of cathode rays with a speed of 1. (b) the corrected value e.9199 gm/cm3. The greatest common V E ) measured ~ for the rise of the drop through a certain fixed divisor (v.76 X lo4 cm. and speed v.832 cgs and its density is 0.85.0.7. Calculate (a) the uncorrected value el of the electron charge. air temperature = 23OC. distance was 0.0 X lo9 cm/sec to pass through the filter? 8. ASD RADIATIOSS [CHAP.50 ATOMS.5. 10.000625 (cm) (cm Hg). The speed of free fall of the drop was 0. 12. in seconds.
therefore. was like a plum pudding. I n the absence of information about the way in which the positive and negative charges are distributed in an atom. the total negative charge in an atom must be an integral multiple of the electronic charge and. No particular assumptions could be made about the nature of the positive charges because the properties of the positive particles from radioactive substances and from gas discharge tubes did not have the uniformity shown by the properties of the negative particles. the total mass of the electrons in an atom is only a very small part of the mass of the atom. which has an atomic weight very close to unity. throughout which was distributed an equal and opposite charge in the form of electrons. provided a starting point for theories of atomic structure. for the present. The fact that atoms of a radioactive element are transformed into atoms of another element by emitting positively or negatively charged particles led to the view that atoms are made up of positive and negative charges. He assumed that an atom consisted of a sphere of positive electricity of uniform density. to state that early experiments of this kind indicated that the number of electrons per atom is of the order of the atomic weight. based on the hypothesis that atoms are made up of electrons and positive charges.CHAPTER 3 THE NUCLEAR ATOM 31 The Thomson atom. I t was remarked that the atom. together with Thomson's proof of the independent existence of the electron. The discovery of radioactivity. Hence. and it was logical to assume that practically the entire mass of an atom is associated with the positive charge. the positive and negative charges must be numerically equal. with the negative 51  . and this problem will be treated in some detail in the next chapter. I t will suffice. The emission of electrons by atoms under widely different conditions was convincing evidence that electrons exist as such inside atoms. Thomson proposed a simple model. and (2) how are the electrons and the positive charges arranged in an atom? Information about the first question was obtained experimentally by studying the way in which xrays interact with atoms. Two important questions then arose: (I) how many electrons are there in an atom. If this view is correct. It was known that the mass of an electron is about one twothousandth of the mass of a hydrogen atom. since the atom is electrically neutral under normal conditions. The first modern theories of atomic structure were. under this assumption.
but the failure of the Thomson model in the particular case of the scattering of aparticles proved to be most profitable because it led to the concept of the nuclear atom. Alphaparticles are also scattered by a very thin film of matter such as gold or silver foil. If a very thin foil is placed in the path of the beam. the area over which the scintillations occur becomes larger. The scattering process is a result of the interaction between the rays of the beam and the atoms of the material. 3 electricity dispersed like currants in a dough of positive electricity. some of the rays are deflected. When the apparatus contained air. the magnitude found for the size of an atom. the image of the slit had sharply defined edges if the experiment was performed in an evacuated vessel. The diameter of the sphere was supposed to be of the order of cm. from their original direction. who found that when a beam of aparticles passed through a narrow slit and fell on a photographic plate. and a careful study of the process can yield information about the rays. When a parallel beam of rays from a radioactive substance or from a discharge tube passes through matter. The scattering of charged particles such as aparticles can be described qualitatively in terms of the electrostatic forces between the particles and the charges which make up atoms.  . showing again that some of the particles have been deflected from their original direction. Thomson was able to calculate theoretically how atoms should behave under certain conditions. I t became clear when this comparison was made that Thomson's theory was inadequate.52 THE NUCLEAR ATOM [CHAP. an aparticle is subjected to both repulsive and attractive electrostatic forces in passing through matter. it was shown that the average deflection caused by a single atom should be very small. When a particular atomic model is postulated. The magnitude and direction of these forces depend on how near the particle happens to approach to the centers of the atoms past which or through which it moves. . and its boundary is much less definite than in the absence of the foil.  . or scattered. the extent of the scattering of the aparticles can be calculated quantitatively and compared with experiment. The . In the case of the Thomson atom. and the theoretical predictions could be compared with the results of experiments. and the scattering of aparticles will therefore be discussed in some detail. the atoms. The scattering of aparticles was first demonstrated by Rutherford. or both. With this model. showing that some of the rays had been deflected from their original path by the molecules in the air. When a beam of particles passes through a small circular hole and falls on a zinc sulfide screen. scintillations are seen over a welldefined circular area equal to the cross section of the beam. This concept is fundamental to atomic and nuclear physics. the image of the slit on the photographic plate was broadened. S i c e atoms contain both positive and negative charges.
is the average deflection after pawing through the foil. If this value is used for & . the number of aparticles which should be scattered through an angle of 10" or more is NlO0 = ~ 0 e . or multiple. f the scatterer is a gold and t is the thickness of the foil in centimeters. according to Thomson's theory. scattering. (31) where 0 is the average deflection caused by a single atom. For example. that is.l " = IO'~N~. or greater than. according t o the Thomson theory. In an experiment. (32) for very small angles. then 4 out to be about 308. and if a is assumed to be lo' cm.( +2 l h ) (32) 2 112 + where No is the number of particles corresponding to 6 = 0." mean deflection of a particle in passing through a thin foil of thickness t should be. one out of about every 8000 aparticles was scattered through an angle greater than 90°. is the result of a relatively large number of small deflections caused by the action of a large number of atoms of the scattering material on a single aparticle. and it was necessary to look for a better model for the atom."' 4m = e ( m t ) . Ruther ford '*' (1911) proposed a new theory of the scattering of aparticles by . The scattering of aparticles by a thin foil. 32 Rutherford's theory of the scattering of alphaparticles. it is convenient to count the number of aparticles scattered through a certain angle. but the number of particles scattered through large angles was much greater'3' than that predicted by the Thomson theory. a is the radius of the atom in centimeters. Geiger found that the scattering agreed with that predicted by Eq. I n fact. I . Rutherford showed that the number of aparticles Ng scattered through an angle equal to. This process is called compound. for very small values of 6. it can be seen that the probability that an aparticle is scattered through a large angle becomes vanishingly small. the most probable angle of deflection of a beam of aparticles was about lo. (32). should be given by Ng = ~ ~ ~ . The experimental scattering of aparticles at large angles could not possibly be reconciled with the theoretical predictions based on multiple scattering by a Thomson atom. which means that a significant number of aparticles in a beam incident on a foil had their directions changed to such an extent that they emerged again on the side of incidence. n is the number of atoms per cubic centimeter. I t was found experimentally by Geiger '2' that when a gold foil 4 x cm thick was used. turns foil 4 X lo' cm thick. in Eq. and 4 .
He assumed that the positive charge of the atom. 31. an aparticle can penetrate very close to the nucleus before the repulsive force on it becomes large enough to turn it back. the magnitude of this force is 72 (33) In his first c & l a t i o n . it was necessary to suppose that there is an intense electric field near an atom. . On this model. is concentrated in a minute center or nucleus. In Fig. 31. the . so that the attractive forces exerted on the aparticle by the electrons can be neglected. For large angle scattering to be possible. With the above assumptions. and the formula for the deflection can be derived from geometrical and physical relationships. Photographs of the tracks of aparticles in a cloud chamber showed that aparticles often traveled in a straight line for a considerable distance and then were deflected suddenly through a large angle. that of the motion of a highly energetic particle under a repulsive inverse square law d force. and can result in a large deflection.   . Rutherford's suggestion was in agreement with this kind of experimental evidence. the calculation of the orbit of the aparticle is reduced to a familiar problem of classical mechanics. Rutherford suggested that the deflection of an aparticle through a large angle could be caused by a single encounter with an atom rather than by multiple scattering. it is relatively far from the negative charges. Rutherford treated the case of an atom sufficiently heavy so that the nucleus could be considered to remain a t rest during the scattering process. instead of being distributed uniformly throughout a region of the size of the atom.  22e2 F=. but the repulsive force can then be very large. and if r is the distance between the two charges. Rutherford proposed a simple model of the atom which could provide such a field. The first step in the derivation of the Rutherford scattering formula is t o write down some useful geometrical relationships. If the magnitude of the aparticle charge is 2e and that of the nucleus is Ze. this theory was based on a new atomic model and was successful in describing the experimental results. which are spread over a much larger volume. where Z is an integer. 3  matter. ' * This statement is proved in Appendix IV.* as shown in Fig.54 THE NUCLEAR ATOM [CHAP. when the aparticle is near the nucleus. where s second derivation of the Rutherford scattering formula is given which contains certain points of special interest. and that the negative charge is distributed over a sphere of radius comparable with the atomic radius. For purposes of calculation. Rutherford sssumed that the nuclear and &particle charges act as point charges and that the scattering is caused by the repulsive electrostatic force between the nucleus and the &particle. At the same time.'5) The orbit is one branch of a hyperbola with the nucleus of the atom as the external focus.
a. c = p/sin 8. the magnitude of s is Also. from analytic geometry.FIG.a2. The aparticle. when the latter is far from the nucleus. and the angle between the xaxis and the initial direction of the aparticle is denoted by 0.  where a is the major semiaxis of the hyperbola. which can be done by applying the laws of . let s denote the distance of the nucleus from the vertex of the orbit of the aparticle. so that p is the minor semiaxis. The eccentricity e is defined as the ratio e = c/a. to the line OA is called the impact parameter and is denoted by p. so that The next step is to find a relationship between the impact parameter p and the scattering angle 6. is along AO. the perpendicular distance from the nucleus. The scattering of an aparticle by the Rutherford nuclear atom. The vertex of the orbit is at x = .31. The initial path of a particle. one of whose branches is the path of the aparticle. . Now. origin of the coordinate system i s taken a t the center of the hyperbola. situated a t the external focus x = c. is deflected through the angle I# and approaches the asymptote OA'. I t can also be seen from the figure that p2 = c2 . on nearing the nucleus. The angle of deviation I# is then equal to ?r . The equation of the hyperbola is. while that of the second branch of the hyperbola is at x = a. one of the asymptotes of the hyperbola.28 radians.
its potential energy. where M is the ma&. since 4 = ?r . in the electric field of the nucleus. sin 0 It follows from the law of conservation of angular momentum that M V p = MVos. Equation (311) is the desired relation between the impact parameter and the scattering angle. the result is When the expression for s from Eq. I f the last equation is divided through by +MV2 and if a new quantity b is introduced. then The second term on the right represents the potential energy of the aparticle.is inserted into the last equation. a t the vertex of its orbit. This energy must also be equal to the total energy f the of the aparticle when it is just at the vertex of the hyperbola. its total energy is equal to its initial kinetic energy. 3 conservation of energy and angular momentum.Ze2/MV2. If the velocity of the aparticle is V at this large separation.56 THE NUCLEAR ATOM [CHAP.b tan 0 2 l and. or and When this value for (Vo/V)' is put into Eq.28. b = 4. MV2/2. which is assumed to be infinite. (37). is p m t i c d y zero. (35). the sum of the kinetic energy and the potential energy is constant. it is found that p = . When the aparticle is at a great distance from the nucleus.1P (1 + cos e) . I velocity a t this point is Vo. . b v: v 2. which is inversely proportional to this distance. According to the former.
13) 2 Similarly. the scattering was determined by counting the number of aparticles incident perpendicularly on a constant area of a zinc sulfide screen placed at a distance R from the foil. Suppose that the beam of aparticles is incident perpendicularly on a thin foil of material of thickness t. the beam passes perpendicularly through the foil. It is now possible to calculate the fraction of the aparticles scattered through a given angle 4. the probability of deflection through an angle between 4 and 4 d 4 is equal to the probability of striking between the radii p and p d p . Hence the fraction of the total number of aparticles deflected through an angle greater than 4 is obtained by inserting for p. cosec3 4 d+ f m t b 2 cos 2 2 = & m t b 2sin 4 cosec4 d o .ntb2 cosec4( 4 / 2 ) 16R2 If now. $ (314) In the experiments made to test the theory. and if Y is the number of aparticles scattered to unit area of the zinc sulfide . from Eq. Q is the total number of aparticles incident on the foil. where 4 is given by Eq. 4 cosec2 3 4 d4 dq = f m t b 2 cot 2 = Q. The fraction of the scattered aparticles falling on an element of area of the screen at a distance R is given by dq 2 a R 2 sin d4  . and 4.l l ) . containing n atoms per unit volume. Then. (311). I t is assumed that the foil is so thin that the particles pass through without any significant change in velocity and that. p = *?mntb2cot2 (3. with the exception of a few particles which are scattered through a large angle. ( 3 . and is given by + + dq = 2apnt dp. Then the chance that a particle passes within a distance p of a nucleus is A particle which moves so as to pass within a distance p of the nucleus is scattered through an angle greater than 4 .
to the reciprocal of the square of the initial energy of 3. (316) were held constant. the number of aparticles falling on a unit area of the zinc sulfide screen a t a distance R from the point of scattering should be proportional to 1. the thickness of the scattering material. I / ( M V ~ ) or the aparticle. 33 The experimental test of the Rutherford scattering theory. ~ . varying the angle of deflection. 4. R represents a radioactive substance which is the source of the aparticles. Colums I11 and V give the observed numbers N of scintillations for silver and gold respectively. one with a silver scattering foil. The number of scintillations counted. F is a very thin foil of scattering material. The source and foil were held fixed. The variation in the value of the ratio is very small  . is proportional to Y. the second column gives the corresponding values of cosec4 (4/2). the ratio N/cosec4 (4/2) should be constant for a given foil under the conditions of the experiment. the other with a gold foil. . where 4 is the scattenng angle. 3 screen placed at a distance R from the foil and a t an angle 4 with the original direction of the particles. The effect of varying the angle of deflection 6 was studied in the apparatus shown schematically in Fig. 32. while the screen and microscope could be rotated in an airtight joint. 2. Rutherford's nuclear theory of the scattering of aparticles was tested point by point in 1913 by Geiger and Mar~den. then . The number of aparticles reaching unit area of the screen in a chosen time interval was obtained by counting the scintillations. hence. are given in Table 31. In the experiment. 1. cosec4 (4/2). N. t.58 THE XUCLEAR ATOM [CHAP. columns IV and VI show the value of the ratio N/cosec4 (4/2).'~'The dependence of the scattering on the four quantities listed a t the end of the last section will be considered in order. The first column gives the values of the angle 4 between the direction of the incident beam of aparticles and the direction in which the scattered particles were counted. and S is a zinc sulfide screen rigidly attached to a microscope M. According to Rutherford's theory. The entire apparatus was enclosed in a metal box which could be evacuated. or to cosec4 (412). (Ze)'. the angle 4 was varied while all of the other variables in Eq. the square of the nuclear positive charge. The dependence of the scattering on the angle of &fiction. In the diagram. The results of two sets of experiments.
3 16.4 18.3 136 320 989 1760 5260 20.@) .8 23..2 18.331 TEST OF T H E RUTHERFORD SCAWERING THEORY 59 FIG.8 18.4 33.300 105.300 132.6 93.0 508 1710 19.5 21 1 477 1435 3300 7800 27. .3 19.1 43. * From Geiger and Marsden.0 21. .2 27.0 51.000 .9 69...8 20.7 18.0 47. Apparatus of Geiger and Marsden for testing the angular dependence of aparticle scattering.6 29.400 5. .6 33. ..@) I Angle of leflection I11 Silver IV I v Gold VI) + Number of Number of N scintillations scintillations N :osec4(+/2) N 22.4 30. 32.
it was found desirable to reduce the number of scintillations counted. 33. the number of scintillations was reduced by about 2500. The results of several experiments are shown in Fig. The variation of the scattering of aparticles with the thickness of the scattering material. The deviations of the ratio from constancy were thought t o be within the experimental error and it was concluded that the theory predicts the correct dependence of the scattering on the angle of deviation. the thickness of air which produces the same loss in energy of the aparticles traversing it as that produced by the material being studied. while the value of cosec4 (#/2) varies by a factor of 250. it is clear that the value d the ratio changes little over the entire range of values of $. 3 compared with that of cosec4 (+/2). 2.(6) . the value of the ratio was practically constant between 4 = 5" and 4 = 30".60 THE NUCLEAR ATOM [CHAP. .   . The equivalent path length in air often serves as a useful standard for comparison + = 150". The dependence of the scaUering on the UIicknessof the scatta'ng matffial. The dependence of the scattering on the thickness of the scattering material was tested by fixing the angle of deflection and using foils of difTerent thicknesses and also of different materials. in which the number of particles per minute scattered through an angle of 25" is plotted as ordinate and the thickness t of the scattering foil is plotted as abscissa.33. Thickness of foils in centimeters of air equivalent FIG. When the results are fitted to those for the larger angles. The thickness of the foil is expressed in terms of the equivalent length of path in air.000. for angles between 4 = 15" and For smaller angles. that is. The results for the smaller angles can be compared with those of the larger angles by noting that in the case of the gold foil.
or to the fourth power of the velocity. The product NV4 should be constant when V is varied. when aparticles of the ranges given in the second column were used.  where a is a known constant. could be determined in several ways. called the range of the aparticles. The straight lines show that for any one element. The fourth column gives the number N of scintillations per minute under fixed conditions. the thicker the screen. The dependence of the g nir etta c s on the energy or the velocity of the aparticles. 3. The results of a typical experiment are shown in Table 32. When the velocity of the aparticles which passed through a given thickness of mica screen had been obtained in this way. The relative values of l / V 4 are given in the Number of sheets o f mica 0 1 2 3 4 5 6 Range of aparticles Relative values of 1/v4 (4 Number N of scintillations per minute * From Geiger and Marsden. which will be discussed in Chapter 13. the number of particles scattered through a given angle per unit time is directly proportional to the thickness of the foil. Geiger and Marsden varied the velocity of the aparticles from a given source by placing screens of mica between the source and the scattering foil.331  TEST O F THE RUTHERFORD SCATTERING THEORY 61 " in work with aparticles. the scattering through a known angle was measured by counting the number N of scintillations. The range R was known to be related to the velocity by the empirical formula. The velocities of the particles were determined by finding how far they traveled in air. This distance.(@ . In a series of experiments. the slower were the particles which reached the scattering foil. the number of aparticles scattered through an angle 4 should be inversely proportional to the square of the energy of the particles. in agreement with the theory. According to the Rutherford scattering formula.
and copper foils. however. and 78 for the three elements.5. The nuclear charge. and the velocity of the incident aparticles are quantities which could be measured directly and their effect on the scattering determined. unlike the other parameters.62 THE NUCLEAR ATOM [CHAP. an equal and opposite negative charge in the form of electrons."' using improved scattering techniques. the positive charge Ze on the nucleus was approximately +Ae. It was not until 1920 that Chadwick. obtained by an entirely independent method. succeeded in measuring the nuclear charge with good precision. 3 4 Some characteristics of the atomic nucleus.4 f 1.7. as will be seen in the next chapter. and a direct comparison between theory and experiment could not be made in this case. These results are not precise enough t o determine unique.   . in some way. (316). silver. could not be measured directly. Thus. Since 1913 the atom has therefore been considered to consist of a minute. showing that the scattering varies inversely as the fourth power of the velocity. integral values of Z. platinum: Z = 77. The d q e d e m x of the ng on the nuclear charge. The scattering angle. Some information about 2 could also be obtained from experiments on the scattering by different foil materials. all four tests of the Rutherford scattering theory were met successfully and constitute the earliest if not the greatest single piece of experimental evidence for the nuclear model of the atom. From these two types of experiments it was found that for elements heavier than aluminum. as predicted by Rutherford's theory. The remarkably good agreement between the predictions of Rutherford's theorj and the experimental results was interpreted as establishing the correctness of the concept of the nuclear atom. that the value of Z can be found by counting the number of aparticles in the beam incident on the scattering foil and the number in the scattered beam under fixed geometric conditions. where A is the atomic weight and e is the electronic charge. 47. 4. The product NV' given in the last column is very nearly constant over the range of values of V studied. These experiments were not accurate enough to provide a reliable determination of the nuclear charge Z.3 f 0. that is. silver: Z = 46. Z = A/2. he obtained copper: Z = 29. 3 third column. the thickness of the scatterer.3 f 0. I t is evident from Eq. positively charged nucleus around which is distributed. but. For platinum. they agree well with the values 29.
(37a). I t might be expected that if an aparticle approaches more closely to a nucleus. for copper down to 1. It is clear that quantitative informat.341 SOME CHARACTERISTICS OF THE ATOMIC NUCLEUS 63 point. The results showed that for silver the Coulomb law held down to 2 X 10la cm. For any hyperbolic orbit.67)4(1*24)(l. and the scattering of aparticles should depart widely . (35). The first information of this kind was obtained from the experiments on the scattering of aparticles. the result for b is b= 4(32)(4. An aparticle has a mass four times that of the hydrogen atom. I f this happens. the quantity b.7 X lo" cm. The smallest value that s can have is that for a headon collision. this is the smallest value which s may have.ionis now needed about the size of the nucleus. the forces between the aparticle and the nucleus should begin to change very rapidly with the distance. With e = 4. and for gold down to 3.6 X lo9 cm/sec.6)2(1018) = 1. the velocity of the aparticle a t the turning point is just zero. when the aparticle is deflected through an angle of 180".5. Consider the case of a copper nucleus bombarded by aparticles from radon. the inverse square law would eventually break down.the velocity of an aparticle from radon is close to 1. that for Vo to vanish.2 X 10l2 cm.5. s must be equal to b. the distance of closest approach is s. since Vo can never be negative. it came from the consideration of the distance of closest approach of an aparticle to a nucleus. if the results of Geiger and Marsden for Z as given by scattering experiments are used.8) (l. the distance from the vertex of the hyperbola to the nucleus as given by Eq. Copper has an atomic weight of 63. The validity of this assumption was borne out by the agreement between the Rutherford scattering theory and the results of the experiments of Geiger and Marsden. In such a collision.2 X lo'' cm. The above calculation depends on the assumption that the Coulomb force law between the aparticle and the nucleus is still valid a t such small distances from the nucleus. and of the range of validity of the Coulomb force law. or 4 X 1. By using faster aparticles. at the same time. Rutherford and others extended the experiments to see how close to the nucleus the l/r2 force law holds. it is supposed to contain practically all of the mass of the atom.8 X 10lo esu. The magnitude of this distance can be estimated by calculating a typical value of b.67 X lo" gm. defined by gives the closest distance which an aparticle of velocity V can approach to a nucleus of charge 2. Hence. I t follows from Eq. then Z for copper is approximately half of 63. furthermore. or 32.
The study of the scattering of aparticles has continued to be an important source of information concerning the atomic n u c l e ~ s . the repulsion was smaller than that calculated from the Coulomb force alone.64 THE NUCLEAR ATOM [CHAP. then for the elements mentioned. It was shown that the data could be interpreted in a consistent way if the radius of the nucleus is assumed to be approximately proportional to the cube root of the atomic weight. I t will be seen in later chapters that other methods of estimating values of nuclear radii give results in good agreement with that just obtained. ' ~ ~ By '~~'~' 1935. it was expected that aparticles might come closer than lo" cm to light nuclei.4 to 1. if "' r = ro~'I3. (318) that for aparticles of a given energy the distance of closest approach is proportional to the nuclear charge 2. with its radius of cm.5 X 10l3 cm."' In the case of aluminum. On the basis of the finding that Z was approximately proportional to the atomic weight. the radii of the respective nuclei are smaller than the distances listed. these data could be used to make quantitative estimates of nuclear radii. specifically nuclear force. . In each case departures from Coulomb scattering were observed. where A is the atomic weight. and are indeed very small compared with an atom. Thus. Theoretical and experimental studies showed that such departures from the inverse square law do indeed exist. that is. The deviations from the inverse square law scattering showed that very close to the nucleus. data had been collected on the elastic scattering of aparticles from most of the light elements through aluminum. with similar results for other light elements. the inverse square law was found to break down at about 6 to 8 X lo'' cm. the nuclei of these elements are about 1012 cm in radius. and that it might be possible to find departures from the Coulomb force law. With the aid of newer theoretical methods. These results provided the first evidence of the existence of a nonelectrical. It is seen from Eq. If the nucleus is de$ned as the region of deviation from the Coulomb force law. This property will be discussed further in Chapter 13. 3 from the predictions of the theory. and ro = 1.
611 (1935). An aparticle counter with an aperture 1 cm2 in area is placed a t a distance of 10 cm from the foil in such a way that a line from the center of the counter to the center of the area a t which the beam strikes the foil makes an angle of $ degrees with the direction of the beam. E. 10. RUTHERFORD. GEIGER and E. 8. CHADWICK.5 gm/cm3. Oxford. CHADWICK. F.. 45. Roy." Proc. Validity of the Inverse Square Law for the Pt Atom. E. J. M. R. (London)." Phys. 1950. 11. K. 3. The Theory o f Atomic Collisions. . Chapters 8."The Scattering of the aParticles by Matter. 1930. 2." Revs. 30.3 gm/cm3 and 1 X loF6 cm thick. Calculate the number of counts per hour for $ = 5. Mag. gen. EVANS. Chapter 9. 15. 6. cm thick and of density 4.and 8Particles by Matter and the Structure of the Atom.. and 60'. E. Maas. Repeat the calculation of Problem 1 with a silver foil instead of a gold foil. EISBERG Mod. 1. 1960. CHADWICK and ELLIS. PARTICULAR 1. 492 (1910). 2nd ed. J. R. H..00Mev aparticles. "The Laws of Deflection of aParticles Through Large Angles. Mom and H. of intensity 10. PORTER." Proc. 604 (1913). "The Scattering of a. ''On the Scattering of Rapidly Moving Electrified Particles. 5. Chapter 3.30 Mev) from polonium. When $ = lo0. Mag. POLLARD. 9. SYMON. 25. 82A. THOMSON. The Atomic Nucleus. MARSDEN.: AddisonWesley. GEIGER. but with 8. 40. W. Cambridge Phil. MASSEY. R. Calculate the atomic number of copper from these data. 10.465 (1910). New York: Macmillan." Phil. In the setup of Problem 1 a copper foil 1 X 8. D. ref. H. Soc. is incident normally on a gold foil of density 19. "Scattering of Alpha Particles." Phil Mag. 7. A beam of aparticles (kinetic energy 5.op.000 particles/sec. 21. "Charge on the Atomic Nucleus and the Law of Force. Mechanics. RUTHERFORD. 15. 2. and C.. Appendix B. J. Rev. "Nuclear Potential Barriers: Experiment and Theory. and ELLIS.Radiations from Radioactive Substances. Roy. (London). New York: McGrawHill.PROBLEMS 65 GENERAL RUTHERFORD. Repeat the calculation of Problem 1. 3. the counting rate is 820 counts/hr.. Soc. 495 (1909).734 (1920). Phys." Proc. GEIGERand E. 1955. cit. MARSDEN.. 669 (1911). England: Clarendon Press. N. 33. 190 (1961). =A.. 9. Soc. "On the Diffuse Reflection of the aParticles. 47. S. 4." Phil. H.90 gm/cm3 is used instead of the gold foil. Reading. The density of silver is 10.
lead. and uranium? How close would 7. What is the general expression for the distance of closeat approach of a positively charged particle to a nucleus? What is the distance of cloeest approach of a ZMev proton to a gold nucleus? How does this distance compare with those for a deuteron and an aparticle of the same energy? 8. a gold foil.00Mev aparticles come to the same nuclei? 6. silver. What is the distance of closest approach of 5. A beam of protons of 5Mev kinetic energy traverse.30Mev aparticlea to nuclei of the following elements: gold.66 THE NUCLEAR ATOM [CHAP. 3 5. copper. What is the thickness of the foil? . What fraction of the aparticles of Problem 1 are scattered through an angle greater than 90°? 7. one particle in 5 X lo6 is scattered 8o as to hit a surface 0. Show that the fraction of aparticles scattered through an angle between 9 0 ' and 180' is given by ntrb2.5 cma in ares a t a distance 10 cm from the foil and in a direction making an angle of 80" with the initial direction of the beam.
the sudden stopping of an electron gives rise to a pulse of radiation which takes the form of xrays. Independent evidence for the number of electrons in the atom was supplied by the pioneer work of Barkla on the absorption and scattering of xrays by matter. a moving charged particle emits electromagnetic radiation when it is accelerated. e l Some properties of xrays.CHAPTER 4 XRAYS AND ATOMIC STRUCTURE  The work of Geiger and Marsden on the scattering of aparticles confirmed Rutherford's concept of the nuclear atom. the cathode is a wire which is heated to such a temperature that it emits thermoelectrons. In practice. 4I). because the atom is electrically neutral. The tube is evacuated until there is no appreciable amount of gas remaining. gasfilled cathoderay tube in which a metal anticaWIode is situated opposite the cathode. and showed that the number of elementary positive charges on the atomic nucleus is approximately equal to half of the atomic weight. with the work of Barkla and Moseley as examples. it is necessary to review some of the properties of xrays in somewhat greater detail than was done in the second chapter. The relationship between the nuclear charge and the position of an element in the periodic table was determined by Moseley's work on the characteristic xray spectra of the elements. xrays are sometimes produced in a low prwure. This number must be the same as the number of electrons in the atom. In the more frequently used Coolidge tube (Fig. because the energy carried by the xrays from heavy metals has been found to be greater than the energy of the xrays from light metals. their strong One of the most important properties o penetrating power. The anticathode serves as a target for the electrons emitted from the cathode and as the source of xrays. In order to do so. Xrays are produced when swiftly moving electrons strike a solid target. and will be d i s c 4 in the present chapter. The anticathode is usually a metal of high atomic weight such as tungsten. f xrays is. of course. The extent to which a beam of xrays will penetrate f the xrays as well as on that of into a substance depends on the nature o 67 . The study of atomic properties by means of xrays. so that all of the current through the tube is carried by the thermeelectrons. has been one of the most fruitful sources of information about atoms. According to classical electre dynamics.
and hard rays. The rays going out from the plate consist in part of primary xrays. A rough differentiation may be made into soft rays. which have relatively high penetrating f their absorption. I is the intensity after passing through a thickness d. With increasing thickness of the absorber the value of p continually decreases. Figure 42 is a schematic representation of what happens. The variation of p with thickness of the absorber indicates that the xrays from the tube are heterogeneous and only become approximately homogeneous when the softer radiations are filtered out of the beam. The absorption of a hontogeneous radiation can be described by the formula where I0 represents the observed intensity of the beam incident normally on the absorbing material. The absorption equation (41) is valid only for homogeneous xrays. and finally approaches a limiting value. which have only small penetrating power and are easily absorbed. for example. while the rest is transformed into heat. the substance. and to some extent xrays can be characterized by their penetrating power in a given material. and in part of rays excited by the primary beam and called collectively the d r y radiutiun. power. When a beam of primary xrays coming from the anticathode of an xray tube falls on a plate of some chosen element. which depends on the absorbing material and on the radiation but is independent of the initial intensity of the xrays. A distinction may also be made. and really defines what is meant by homogeneous xrays.XRAYS AND ATOMIC STRUCTURE FIG. for the radiation from an ordinary xray tube. In that case. The secondary radia . Coolidge xray tube. on the basis o between honzogeneous and heterogeneous xrays. I t does not hold.  . no constant value of p is obtained. called the absorption c o e j h k n t . .41. or into radiation of another sort. the transmitted beam. The process of absorption of xrays is complicated and involves several phenomena closely connected with the properties of atoms. part of the radiation goes on through the plate. and p is a constant.
the necessary condition that the radiation shall appear at all is. made detailed measurements of the absorption coefficients of the characteristic xradiation from various elements. They found that the elements have. (d) characteristic Prays. The @raysare not of interest for the present and will be discussed in a later chapter. The scattered xrays have very nearly the same absorption coefficient in a given material as the primary rays. the scattered radiation is also heterogeneous provided the absorbing plate is not too thick. (b) characteristic xrays. in addition. They always have the same properties for a given chemical element.and Lradiations of the different elements can be compared in terms of their absorption coefficients in aluminum. the Kradiation is the harder and may have an absorption coefficient several hundred times smaller than that of the Lradiation. The xrays of the secondary radiation have turned out to be a fruitful source of information about atoms. The characteristic xrays are so called because they are typical of the material of the plate and. are homogeneous. however. independently of the hardness of the primary rays. they seem to be primary rays which have merely had their direction changed by the material through which they pass. important types of radiation: (a) scattered xrays. two types of characteristic radiation which differ greatJy in their absorption coefficients. The characteristic K. that the primary radiation shall have a certain minimum hardness. the .411 SOME PROPERTIES OF XRAYS f 69 Primary'xray 42. in other words. Interaction between xrays and matter. (c) scattered 6rays. and their character is independent of the nature of the absorbing and scattering material. Since the primary rays which come from an xray tube are heterogeneous. Barkla and his coworkers. The absorption coefficient for each of the radiations decreases with increasing atomic weight. These are called the K.and Gradiations. in general. during the early years of the 20th century. tion contains four distinct.
911 x loe2' grn. aa a secondary beam.'" and can be written as   Here. then the If effective radius of the electron is given by .80 X 10lo esu. m = 0. The quotient 1 / 1 0 must have the dimensions of an area. When a beam of xrays strikes the atoms in the plate. for elements of relatively low atomic weight and xrays of moderate hardness. and the radiation produced by the latter may be neglected. I and c = 3. 4 characteristic radiation of the elements incresses in hardness with increasing atomic weight. The scattering of the primary xrays by a material plate was treated by Thornson on the basis of the classical theory of electrodynamics.it is often denoted by 40. The proceea takes place in such a way that. is also called the classical cross &ion j f the values e = 4. The electrons. the latter is accelerated and emits electromagnetic radiation. a fraction 90 is scattered. but I denotes the entire scattered energy. we can now consider some of the experiments with xrays which added to the knowledge of atomic structure. the scattered and incident. The magnitude of the energy scattered by a free electron was calculated by ~homson. lo is the intensity of the primary radiation and has the units of energy per square centimeter.  G 2 The scattering of xrays by atoms and the number of electrons per atom. According to this theory. and the electron may o t sedten'ng by a j r e e this reason. O f the itmident radiation that falls on unit area of a surface drawn perpendicular to the beam. The electrons seem to abstract energy from the primary beam and reradiate. are accelerated much more strongly than are the atomic nuclei. then is regarded as the effective crose section of the electron. and is called the elassiccrl 'ng coe&ied f o raj r e e eledron. the latter are accelerated. With the above information about xrays as background.70 XRAYS AND ATOMIC BTRUCTURE [CHAP. eWm. radiations differ only in direction. or scatter it. because of their small mass. radiation incident on a charged particle which can move freely exerts a force on the particle.0 X 10'' cm/sec are inserted.
*~' other processes besides scattering were present in these experiments. can be found from measurements of the scattering of xrays. Hence. n. some of the energy is removed from the beam. Experiments of this type were performed by Barkla and his collab~ r a t o r s . the number of atoms per cubic centimeter of an element is . (48).nNO0 Ax. then the fractional diminution . and the intensity of the beam is reduced. the scattering could be represented by (48) nN+. = 0 . (45) Since +0 is independent of z. When a beam of xrays paases through a thin sheet of material. (47). I f there are n electrons per atom. ' ~ actually. Barkla found that within certain limits of hardness of the xrays. 2 ~ ~ independently of the hardness and of the scattering material. N and are known. The quantity e2/mc2 = 2. but the necessary corrections could be made. In Eq.. I = I oenN+o= I t follows from the last equation that ( 4 4  In Eq. with ionization chambers. for example..$t Now. with the known value of.A I / I of the incident beam in going a distance Ax is given by A I 1. p is the density in grams per cubic centimeter. Hence. the number of electrons per atom. both sides o f this equation can be integrated directly giving. for the intensity I of the beam after traversing a thickness z of the material.82 x 10l3 cm is usually referred to as the clrrsgieal radius of the electron.621 THE SCATTERING OF XRAYS BY ATOMS 71 The effectiveradius of the electron is therefore of the same order of magnitude as the radius of the lighter nuclei. 10 and I can be measured experimentally. and z is the chosen thickness of the scatterer. and N atoms per unit volume. and if it is assumed that all of the electrons scatter independently.
Second. it has been assumed that the laws of classical electromagneticshold. von h u e and his cc+workers discovered that crystals act as gratings for diffracting xrays. Esrly work on xrays indicated that they are electromagnetic radiations with wavelengths several thousand times smaller than those of visible light. while the average distance between atoms in a solid is between lo' and lo' cm. some additional information was needed on the optical properties of xrays. the relationship (48) is approximate. the number of electrons per atom is very close to h a l f the atomic weight. the methods of ordinary spectroscopy could not be used to measure the wavelength or frequencies of xrays. Then. In 1912. This result must be considered to be approximate for several reasons. For these reasons. Diffraction occurs because ordinary xrays have wavelengths between and lo" cm."s5' suppose that a train of monochromatic xray waves strikes a crystal which consists of a regular arrangement of atoms  . The way in which this could be done was shown by Bragg. or at best semiquantitative. Hence. but before that work could be done. +3 The diffraction of xrays and Bragg's law. therefore. to use the diffraction of xrays by a crystal to make quantitative measurements of the wavelengths of xrays. an assumption which is valid only within certain limits of hardness of the xrays and for scattering materials of low atomic weight. result. the result that the number of electrons is half the atomic weight was a useful qualitative. I t was. 4 where N o is Avogadro's number = 6. First.72 XRAYS AND ATOMIC STRUCTURE [CHAP. in a crystal there must be some atomic or molecular unit arranged in a regular repeating order which results in the observed crystal symmetfy. and A is the gram atomic weight of the element. The fact that two completely independent methods gave the same result was a strong argument for its validity.02 x atoms per gram atomic weight. The strongest evidence came from Moseley's work on xray spectra. Furthermore. in agreement with the equally semiquantitative results for the charge on the nucleus obtained from ascattering me+uements. Following Bragg. : and (49) that is. The properties of xrays and crystals result in conditions analogous to those which occur when visible light traverses an optical gratingregularly spaced discontinuities separated by distances several times the wavelength of the incident radiation. It is possible. however.
The condition that there be a reinforced reflected beam is then n A = 2d sin 0. can be derived by referring to Fig. (410) where n is an integer. . and let a line drawn from O2 perpendicular to the planes cut the first plane at 0 . As the angle of incidence is changed. The crystal acta as a series of parallel reflecting planes. . tion o f the Bragg equation. Consider a ray which meets two successive crystal planes at 0 1 and 0 2 respectively. as in Fig. and the diffraction caused by the reflection of xrays from crystal plmes becomes apparent. If the incident or glancing angle 0 haa the appropriate relationship to the distance d between the reflecting planes and to the wavelength A of the incident waves. the Bragg equation. Let Oc be perpendicular to 0 1 0 2 . The condition for obtaining reflection maxima for xrays. then Ola = Olc and the difference in path is cO2b = 2d sin 0.The reflection o f xrays by crystal planes. If the difference in path length is equal to one wavelength. 43. The reflections are then said to be in phase. then there will be a reflected beam a t the position O1 which satisfies the condition X = 2d sin el. illustrating the derivaF I ~ 43. There will be reinforcement if the path 0 1 0 2 b taken by waves scattered at 0 2 is longer than the path Ola for waves scattered a t O1 by an integral number of wavelengths. Draw OIPl and 02Parepresenting rays reflected from the two planes. the latter interfere with one another and the resulting beam is weak. the reflections from the various planes reinforce each other and the resulting reflection is exceptionally strong. a series of reflections is seen which show alternate maxima and minima of intensity. and draw aOb perpendicular to OIPl and 02P2to represent a wave front of the reflected beam. or molecules. But if the angle 0 does not satisfy the condition for the different reflections to be in phase. 43.
was shown by an experiment The validity of Bragg's law. E in which the glancing angle I9 was varied. i wavelengths. If the difference in path length is 2. 3. 44. A2. which acts as a diffraction grating. are seen a t angles whose sines are twice those of the angles corresponding to the Grst order reflections. . . (4lo). (&lo). and C2. After dilTraction. ' 6 . The results of such an experiment are shown in Fig. (&lo). This beam gives the reflection or spectrum of the jirst order. 45. . . peaks. . . . the ionization passes through peaks a t certain sharply defined angles. Diagram of a Bragg xray spectrometer. and Ca belong to the second order spectrum. the angle between the ionization chamber and the primary beam was kept equal to 28 in order to receive the beam reflected from the crystal. by B r a g g . These considerations formed the bask for the design. the and C1 represent xray spectrum lines.ith orders will be found a t the positions 01. while All B2. the xrays enter an ionization chamber D. Al. and the ionization was measured as a function of 0. . . third. .4 4 . The angular position of c is read by means of a vernier. Secondary three peaks All B1. and then impinge on the crystal c. but their relative intensities also are in the same ratio as those of the corresponding lines in the first order. . which absorbs xrays strongly. 6 ' of an xray spectrometer. The electrometer E records the intensity of ionization in the chamber. . Not only are the angles a t which the lines of the second order spectrum are seen just what they should be according to Eq. . A schematic diagram of this instrument is shown in Fig. and C1 are lines of the first order spectrum. B2. B1. 02. Bi.74 XRAYS AND ATOMIC STRUCTURE [CHAP. 4 FIG. then reflections of the second. According to Eq. The xrays from a tube are collimated into a narrow beam by the slits S1 and Sa cut in lead plates. I t is seen that instead of varying uniformly with the glancing angle. q . filled with methyl iodide.
On the other hand. on the average. o ' cm and is The best value obtained in this way for NaCl is 2. Then. where p is the density.454 g m and contains N o molecules or 2No atoms.165 gm/cm3. The wavelengths of xray lines can now be and d = 2. because xray studies of this crystal show that the sodium and chlorine atoms occupy alternately the corners of the elementary cube of the lattice. one grammole of NaCl weighs 58.grating spacing d between the layers of atoms is different for different crystals. the same lines appear with about the same relative intensity. equal to 2. if the crystal is changed but the target is kept the same.Glancing angle 0 in degrees FIG. the wavelengths of the spectrum lines can be measured with good precision. where No is the Avogadro number. the lattice spacing of crystals can be determined experimentally. When the grating spacing d is known accurately. 45.454/2N0 gm.820 X l measured accurately with ruled gratings and. as is shown by the fact that if the anticathode in the xray tube is changed. The grating spacing of the sodium chloride crystal can be calculated. @)). to every atom there corresponds a volume d3 with mass p i 3 . The first xray spectrum (Bragg. showing that the. The average mass per atom is then 58. The xray spectra obtained in this way are characteristic of the target from which the xrays are emitted.and I I o ' cm. The spectral lines are superimposed on a continuous spectrum represented by the partially dotted line in the figure. but the angles at which they appear are changed.8197 X l . with the known wavelengths. Also. an entirely different line spectrum is obtained.
Moseley's law. He used an xray spectrometer with a crystal of potassium ferrocyanide. The results discussed in the last section showed that the dBraction method could be a p plied to the analysis of the xray spectra of the elements. Positions of the Ka and KOxray spectrum lines of a sequence of elements as obtained by Moseley. and the characteristic radiations could be described quantitatively by the wavelengths of the spectral lines.and G types of characteristic radiation previously observed by Barkla. Thirtyeight different elements were used as targets in his xray tube. I t was soon found that these radiations are more complex than was supposed from Barkla's absorption measurements.4 6 . Moeeley's photographs showed that the spectral limes emitted by these elements belonged to two different series which were identified with the K. 4 4 Characteristic xray spectra. i n e s . 4 in good agreement with the calculated value. and detected the reflected xrays with a photographic plate rather than with an ionization chamber. .76 XRAYS AND ATOMIC STRUCFURE [CEAP. but they could still be separated into a number of sharply defined l The first systematic study of the xray spectra of the elements was made by ~ o s e l e ~ ' ' 'in 19131914. Crystals other than NaCl can also be used and. in fact. The Kradiation was found to consist of two lines denoted by Ka and Kn. calcite (CaCOa) is now used as the standard crystal. thia   Wavelength FIG.03567 X lo* cm. it has a lattice spacing of 3.
22 92.54 65. The gap between the calcium and titanium lines represents the positions. and it is clear that the wavelengths decrease in .38 88.85 58.lines is shown in Table 41.7 106. The lines are those of the first sequence of elements studied by Moseley.97 28.lines of a sequence of elements are listed in the second column of Table 41.95 101. The wavelengths of the K.08 47. where v is the frequency of the spectrum line.result can be seen from Fig.096 40. X lo8 cm Atomic weight 26. Moseley's quantitative analysis of some of his experimental data on the K. which occurs between those two elements in the periodic system and on which no measurements were made.457 39. and c is the Element Aluminum Silicon Chlorine Potassium Calcium Titanium Vanadium Chromium Manganese Iron Cobalt Nickel Copper Zinc Yttrium Zirconium Niobium Molybdenum Ruthenium Palladium Silver Wavelength.94 58. which also indicates the relative positions in which these lines appeared in the photographic plates.92 91.69 63. equal to c/X.95 52.88 Atomic number.91 95.of the lines of scandium.06 35.93 55. the elements from calcium to copper.01 54. 46.90 50.7 107. Z 13 14 17 19 20 22 23 24 25 26 27 28 29 30 39 40 41 42 44 46 47 .a regular way as the atomic weight increases. The third column of the table gives the values of a quantity QK defined by the relatimnship QK = ( v / ~ v o ) ~ ' ~ .
called the Rydberg constant. Moseley used the quantity QL. v o is replaced by another constant R. The quantity QK then becomes The Rydberg constant will be discussed at greater length in Chapter 7 and the basis for Moseley's choice of the constant R will be shown. defined by instead of QK. In order to do so. In this case. for which Moseley used the value 109. Moseley concluded that the experiments led to the view that the number of unit charges on the nucleus is the same as the number of the place occupied by the element in the periodic system. for reasons which will be explained in Chapter 7.78 XRAYS AND ATOMIC STRUCTURE [CHAP. Moseley assigned a value of the atomic number to each element from aluminum to . He considered that this quantity could only be the positive charge on the nucleus. vo is a constant. Except in the case of nickel and cobalt. The same conclusions can be drawn from the variation of QL as were drawn in the case of QK. On the basis of these results. It will be remembered that the conclusion had been drawn from uparticle and xray scattering that the number of unit charges on the nucleus of an atom is approximately half the atomic weight. and that both of these numbers could be represented by a quantity which he called the atomic number. For the present purposes it is enough to note that QK is proportional to the square root of the frequency of the K 1' me. Although. important in the study of line spectra. Moseley saw that QK increases by a constant amount in passing fro. they do so in a much less regular way than QK. 4 velocity of light. Moseley concluded that there is in the atom a fundamental quantity which increases by regular steps from one element to the next. Moseley extended his experiments as far as gold.^ one element to the next when the elements are in their order in the periodic system. Moseley then noted that atomic weights increase by approximately two units. in going from element to element and this suggested to him that the number of charges increases from atom to atom by a single electronic unit. the atomic weights increase.so that experimental results were available for 38 elements between aluminum and gold. If the frequency is replaced by the wave number B (the reciprocal of the wavelength. on the average. which represents a certain fundamental frequency.lines of a series of elements are shown in Table 42. His results for the L. P = l/h). he worked with the lines of the L series because the wavelengths of the K series of the heavier elements became too small for precise analysis.720 cm'. with this exception. this order is the same as the order of the atomic weights of the elements listed in Table 41.
Aluminum. When elements were skipped in Moseley's work. was assigned the atomic number 13. because 12 elements were known to precede it in the system. cobalt has a greater atomic weight than nickel. but its chemical properties are such that it should precede nickel in the periodic system. Known elements were found to correspond with all the numbers between 13 and 79 except for four . corresponding gaps in the values of QK or QL were found. The order of the atomic numbers turned out to be the same as that of the atomic weights except where the latter were known to disagree with the order of the chemical properties of the elements. Element Zirconium Niobium Molybdenum Ruthenium Rhodium Palladium Silver Tin Antimony Lanthanum Cerium Praeeeodymium Neodymium Samarium Europium Gadolinium Holmium Erbium Tantalum Tungsten Osmium Iridium Platinum Gold Wavelength. For example. The relative wavelengths and values of QK (Table 41) show that cobalt should indeed precede nickel in the periodic system in spite of the discrepancy in atomic weights. X lo8 cm QL = (~/&po)''~ I1 Atomic weight Atomic number. Z 40 41 42 44 45 46 47 50 51 57 58 59 60 62 63 64 66 68 73 74 76 77 78 79 gold. the first element in the periodic system studied by Moseley.
72 and 75. is plotted against the atomic number in Fig.5.3 f 0. a few years later. These missing elements have since been discovered and identified by their xray spectra. respectively. for either the K aor Laline is plotted against the atomic number. The points. 4  numbers. a straight line should result. 43. respectively. In addition. . 61. if the square root of the frequency. It can be seen from Tables 41 and 42 that Thus. which corresponded to still undiscovered elements. 48. 47. the points (represented by crosses) show much greater scatter. Similar results are obtained for the Lalines in Fig. The quantity QK. Moseley's relationship between the atomic number and the square root of the frequency of the K 1 ' me.80 XRAYS AND ATOMIC STRUCTURE [CFIU. 47. When QK is plotted against the atomic weight. represented by the dots in the figure. and platinum turned out to be 29. and 77. The figures show in a striking way that the 2 0 40 I  Atomic weight ( A ) 60 I 80 I 1 Atomic number ( Z ) FIG. silver. or the wave number. 47. line. obtained directly from &scattering measurements the values 29. do indeed fall on a straight line. which is proportional to the square root of the frequency of the K. the atomic numbers of copper.4 f 1. 46.3 f 0. As mentioned in Section 33.7. and 78. Chadwick.
.: u = (A&)  u o ( z .1l2.48. . These relationships will be useful later for comparison with theory. and the methods of xray spectroscopy have been improved.correlation between the frequency of the characteristic xray lines and the atomic number is much more satisfactory than that between the frequency and the atomic weight. The frequencies may also be expressed by the empirical formulas K. Z x Q L vs. Moseley's work has been extended to other elements.  in terms of the wave numbers. A I I I I 50 60 70 Atomic number (2) 80 FIG. Additional types of charAtomic weight ( A ) 100 I 120 I 140 I 160 I 1 Q L vs. Moseley's relationship between the atomic number and the square root of the frequency of the L 1'me.
14. f & : Vd. G.. K.. "Die Radioelemente. COMPTON and S. Spektsoscopie det Rdgenstrahlen. BRAGQ and W.82 XRAYS AND ATOMIC STRUCMJRE [CHAP. 27. A89. 17. London: 5. C. gen ref. G. 8. K. The Spectroscopy o f XRays. Bmaa. I . ALLISON. as was thought by MendelQff. LAURITSEN. "The Absorption of Roentgen Rays. Bmaa. 21. The fact that the position of an element in the periodic system is more directly related to the atomic number than t o the atomic weight shows that the chemical properties are directly related to the nuclear charge. CLARK. Cambridge Phil. 3rd ed. The CrystaUine S G. VAN DEN BROEK. "Note on the Energy of Scattered XRadiation. Chapter 8. General Survey. cit. Chapter 111.. New York: McGrawHill. (London). SADLER.1949. Applied XRays. 1935. PARTICULAR 1. F. 17. COMPTON and S. BARKLA. H. J. L." Proc. A." Proc. Mag. 703 (1914). G. L. M. 2nd ed. New York: McGrawHill. XRays in Themy and Ezperiment. 1955. A. W. and T. A88. 1931. . such as the M and N series of lines. H. H. i t is also the same as the ordinal number of the element in the periodic system. G. KENNARD. The identity of the latter number with the nuclear charge number and the number of electrons had been suggested by van den Broek"' but could not be considered t o be proven until the work of Moseley. I&odE(etiun 10 Modetn Physics.428 (1913). 1024 (1913). Oxford: Oxford University Press. 1925. 1942. A. H.43 (1912). "The Reflection of XRays by Crystals. MOSELEY. The atomic number of a n element is the same as the number of unit charges on the nucleus. 2. 7. L. L. and is equal to the number of electrons in the atom. op. 26. 32 (1913). K. and atomic numbers have been assigned t o all of the chemical elements. RICHTMYER. which are even softer than the Gradiations. Chapter 3. Berlin: Julius Springer. H. C. Bmaa. 648 (1911)." Physik. SIEGBAHN." Phil. Soc." Phil. and are not determined by the magnitude of the atomic weight. 739 (1909).. M. 6. Mag. ALLISON. 246 (1914). "The High Frequency Spectra of the Elements. SIEGBAHN. E. W. New York: Van Nostrand. GENERAL A. W." Phil. Soc. Roy. 2nd ed. 4. Mag. The conclusions reached by Moseley have been confirmed. BARKLA and C. 4 acteristic radiation have been found. BELL.. "The DitTraction of Short Electromagnetic Waves by a Crystal. das periodische System und die Konstitution der Atome. 3. 5th ed. Z.
3216 A. and zinc.0' for firstorder Bragg reflection from a cleavage face of NaCl. 6.  1. copper.454. respectively. and 9. What is the lattice spacing of NaCl? Find Avogadro's number. xrays with a wavelength of 0.820 X reflection is observed a t a grazing angle of 8' 35'. calcite. 4. How many electrons are there per atom of beryllium (a) as calculated directly from the attenuation.30 cm thick and of density 1. The wavelength of the L. What is the wave length of the xray? What is the atomic number of technicium? 7. respectively.966 A. A beam of xrays is incident on a sodium chloride crystal (lattice spacing cm. 5.164 gm/cm3 and the molecular weight is 58.255 A. xray line from this element hss been determined by measuring the glwcing angle as 7. and with the known atomic numbers 47 and 78 for Ag and Pt. The firsborder Bragg reflection from a cleavage face of NaCl is found in an experiment a t an angle of 15' 53'. The grating spacings of calcite. The intensity of a beam of xrays is reduced to 90% of its initial value in passing through a slab of beryllium 0. An unknown substance emits L. iron. nickel.036 A. Moseley used a crystal of potassium ferrocyanide having a grating spacing of 8.963 A. lines listed in Table 41. Technecium is one of the new elements discovered among the products of the fission of uranium. Calculate the frequencies and the wave numbers of the K . respectively? 4. xray lines of Ag and P t are 4. and mica. determine the atomic number of the unknown substance. At what grazing angle would the first reflection of a beam of 1A xrays be observed in rock salt. (b) as calculated from Barkla's relationship. Eq. 1 X loWs cm = 1 angstrom). . and mica are 3.5412 A as measured with a ruled grating. xrays of chromium. In his experiments. quartz. The K .1538 A and 1. quartz.408 A. With these data. manganese. The wavelength of the K . cobalt. line for copper has a wavelength of 1. 8. given that the density of NaCl is 2. Find the positions of the first reflections of the K . (48)? (c) What is the value of the absorption coefficient? 2. The first order Bragg = 2. What is the wavelength of the xray? At what angles would the second and third order Bragg reflections occur? 3. respectively. respectively.84 gm/cm3.
so that the latter are directly related to the nuclear charge and the number of electrons in the atom. I f the electrons @reeach a t a distance a from the nucleus. but they all failed. I n the first four chapters. and tLermodynamics) could not explain or describe atomic phenomena. the fundamental experiments bearing on the problem of the nature of atoms were discussed. Hence. Attempts were made to construct a theoretical atomic model on the basis of classical physics. T o describe in detail how these properties depend on the nuclear charge. and it would be expected that such a model would include a description of the way in which electrons are arranged in the atom. These experiments were interpreted in terms of certain concepts which form the basis of present ideas of atomic structure. and the example can be extended to any atom. which represents the position of the element in the periodic system. the repulsive electrostatic force between the electrons is e2/4a2. the electrons would fall into the nucleus and there would be no mechanical stability. In 1913. each electron is attracted to the nucleus by a force equal to 2e2/a2. and it was realized eventually that classical physics (Newtonian mechanics. Suppose next that the electrons revolve in some way about the nucleus. a more detailed model of the atom was needed. Maxwellian electromagnetics. a t rest. The first example is that of an atom consisting of stationary positive and negative charges. The charge on the nucleus was known to be Ze. where Z is a positive integer and e is the magnitude of the electronic charge. . Consider an atom consisting of a nucleus with two positive elementary charges and with two electrons somewhere outside the nucleus. . and consider for simplicity the case of a hydrogen atom which has one 84    . surrounded by electrons distributed over a volume about lo' cm in radius. In this system. an atom was supposed to consist of a positively charged nucleus about 10l2 cm in radius.CHAPTER 5 THE QUANTUM THEORY OF RADIATION 51 The failure of classical physics to describe atomic phenomena. the elements are ordered according to their properties. This failure may be illustrated by means of two elementary examples. under the action of electric forces alone. eight times as great as the repulsive force. the number of electrons was known to be equal to Z because the atom is electrically neutral under normal conditions. This example is a special case of a theorem which states that an electric charge cannot be in equilibrium. Moseley's work showed that the number Z is also equal to the atomic number. and 2a from each other.
and that the velocity of the electron is such that the attractive force between the electron and the nucleus provides just the centripetal force required. nucleus and electron. however. it was necessary to develop a theory for the emission of radiation which was based on a concept entirely opposed to the ideas of classical physics. Bohr's quantum theory of atomic structure was the first successful attempt to fill the need. and although it eventually was found to be unsatisfactory. according to electromagnetic theory. classical physics could not explain the dependence of the intensity of the radiant energy emitted by a blackbody on the wavelength of the radiation. Moreover. the electron should radiate energy. Planck's quantum theory lies at the foundation of present ideas about atoms and nuclei. is subject to a constant acceleration toward the nucleus. helium gives another set. it became apparent that the description of atomic phenomena required a new kind of physical theory. To explain this phenomenon. for example. The system. In particular. by passing an electric discharge through the gas. It had long been known that when the atoms of an element are excited. and atoms seem to last for a long time. and it would be futile to try to continue into atomic and nuclear physics without first having some grasp of the basic ideas of this theory. Actually. in the case of hydrogen. After many attempts like those cited. Assume that the electron revolves about the nucleus in a circular orbit of radius a. The new concept. The energy of the system should then decrease. Hydrogen always gives a set of lines with the same wavelengths. it prepared the way for the modem. sodium still another. in spite of many attempts. emitting radiation of constantly increasing frequency. The Bohr theory was baaed on the quantum theory of heat radiation introduced by Planck in 1901 and applied to light by Einstein in 1905. the electron should gradually spiral in toward the nucleus. must be applied to all problems involving the emission and absorption . that of quanh or discrete corpuscles of energy. Theory has again failed to account for the stability cf atoms. classical physics could not account for the characteristic optical spectra of the elements. and should eventually fall into the nucleus. nonclassical concepts. These predictions of classical physics are in strong disagreement with the experimental facts. The quantum theory actually arose from the failure of classical physics to explain some of the experimental facts of thermal radiation. in contrast to the prediction of the emission of radiation of increasing frequency. should then be mechanically stable. one which could describe atomic phenomena in terms of new. and if the resulting light (radiation) emitted by the element is dispersed by a prism. nor could it account for xray spectra. and so on. The electron. a line spectrum characteristic of the element is observed.511 FAILURE OF CLASSICAL PHYSICS 85 electron and a nucleus with one positive charge. there is no atomic catastrophe. and. more successful theory of quantum mechanics.
and those shorter than 50 A but longer than about 0. Gammarays have still shorter wavelengths. and make clear what is meant by quanta of energy. and is abbreviated as A. approximately the temperature of an incandescent lamp filament.that the'body appears "white hot. At any temperature. the radiation cLntins enough of the shorter wavelengths so. and T is the absolute temperature in OK. I t is well known that a hot body emits radiation in the form of heat. (51)  ." The consideration of measurements by Tyndall on the radiation from hot platinum wires led Stefan (1879) to suggest an empirical rule that can be written in the form W = euT4. Thermal radiation consists of electromagnetic waves and differs from visible light and xrays in having longer wavelengths. or total emitlance. and this spectral distribution changes with temperature. the fraction of visible radiation increases until a t 3000OC. the rate of radiation is small and the energy is chiefly of relatively long wavelength (infrared radiation). which means that the distribution of energy among the different wavelengths has shifted so that a large enough portion of the radiant energy has wavelengths within the visible spectrum.86 THE QUANTUM THEORY OF RADIATION [CHAP. Thus. 5 of electromagnetic radiation. wavelengths between 4000 A and about 50 A constitute the ultraviolet radiation. W is called the total emissive power. because the human eye can see these wavelengths. bodies begin to radiate visible light.01 cm are the infrared or heat waves. radiations with wavelengths between 7000 X cm and 4000 X lo' cm are usually considered to be light. and this is the reason for the importance of quantum theory in these fields. The concept of quanta of energy is so different from classical physical ideas that only a careful study of the problem of thermal radiation can show the need for a quantum theory.  where W is the rate of emission of radiant energy per unit area and is expressed in ergs per square centimeter per second. The problem of thermal radiation will therefore be considered in some detail. The length lo' cm is called the angstrom unit. At temperatures between 500 and 550°C. As the temperature rises. The quantity e is called the emissivity of the surface and has a value between zero and unity. At low temperatures. Waves longer than 7000 A and shorter than 0. but these are the basic problems of atomic and nuclear physics. 52 The emission and absorption of thermal radiation. depending on the nature of the surface. the emitted heat energy is distributed over a continuous spectrum of wavelengths. u is a constant .01 A are usually clwified as xrays.
states that the ratio of the emissive power to the absorptivity is the same for all bodies a t the same temperature. incident on the surface of the body. which is absorbed. but some substances such as lampblack. I t follows from Kirchhoff's law that no body can emit radiant energy a t a greater rate than a blackbody. the difficulties to which it led gave birth to the greatest revolution in physical thought during the 20th century. is equal to unity. The emissivity of a blackbody. Their interest was concentrated on "explaining" or "underctanding" the variation in terms of the fundamental concepts and theories of physics. is unity. The spectral distribution of the energy radiated by a blackbody has. or asbestos board reflect . i. a t the temperature T I . A very simple relation has been shown to exist between the absorptivity of a body and its total emissive power. called the StefanBoltmann constant. namely that for a blackbody. a body which absorbs all of the radiant energy falling upon it. Physicists chose as the subject of their studies the thermal radiation from a blackbody. and is equal to the emissive power of a blackbody a t this temperature.521 EMISSION AND ABSORPTION OF THERMAL RADIATION 87 . one with absorptivity equal to unity. physicists were more concerned with the way in which the emissive power of a body varies with temperature and wavelength. is also the most efficient emitter.the quantity e in Eq. Boltzmann derived Eq. rough steel plate. Hence. This relation. A blackbody is. A blackbody does not really exist in nature. therefore. a special interest. In the late 19th and early 20th centuries. called Kirchhoff's law. by definition. and the equation is therefore known as the StefanBoltzmann law.e. modifying the theories so that they include the experimental data. this law holds for each wavelength. the blackbody. In 1884. By "explaining" or "understanding" is meant fitting the experimental information into the scheme of existing physical theories or. and any smaller value of the absorptivity necessarily implies a smaller value of the emissive power. and has been confirmed by experiment. if this cannot be done. T o understand what is meant by the term blackbody we first define the absorptivity of a body as the fraction of the radiant energy. Although this interest may not appear to be a matter of great practical importance. (51) from thermodynamics. the net rate of loss of energy by the body is given by Equation (52) is familiar because of its application to practical problems of heat transfer. and the total eznissive power of a blackbody depends only on the temperature and not on the nature of the body. is surrounded by walls a t the temperature T2. flat black lacquer. smaller than TI. which is the most efficient absorber of radiant energy.. (5l). for the maximum value of the absorptivity. If a body of emissivity e.
which consists of a h i i strip blackened. The monochromatic emissive power is related to the total emissive power by the equation w WA~A. As a source of blackbody radiation. therefore. Hence. or cavity. an electric furnace may be used. very thin metal strip with electrical connections. consisting of a long tube. . The ordinates are relative values of the monochromatic emissive power. 51 is obtained.is the emissive power per unit range of wavelength. the emissive power can be measured as a function of wavelength. The radiation falling on one junction of the thermopile causes the temperature a t that junction to rise above that of the second junction. denoted by WA. Another useful energymeasuring device is the line bolometer. The bolometer can detect very small changes in temperature and. uniformly heated. Rsdiation falling on the strip raises its temperature. preferably with blackened walls. a curve like that in Fig. as measured by a sensitive bridge galvanometer. . or fluorite. a t a fixed temperature. The spectral distribution of the radiation can be analyzed by means of an optical spectrometer with the photographic plate replaced by some energymeasuring device. or spectral emitianee. . T is the receiver of the radiations. This quantity. The abscissa is the wavelength in microns (1 micron = lo' cm). and the resulting electric current causes a deflection of the galvanometer needle. and is connected as one arm of a balanced Wheatstone bridge.88 THE QUANTUM THEORY OF RADIATION [CHAP. hollow enclmre. A small hole is made through the wall and the radiation coming through the hole is observed. so that it is a sensitive and useful instrument for measuring the spectral distribution of thermal radiation. The radiations from the blackbody are refracted by a salt prism which may be rock salt. The deflection caused by the absorption of radiation in a given range of wavelength depends on the emissive power of the radiator in that range. ' . a + =c It is seen from the figure that for very short and also for very long wave . is a good substitute for the radiation of a blackbody. The temperature of the central part of the interior is measured with a thermometer of some kind. quartz. When the emissive power of a heated body is measured as a function of wavelength. 5 s only a few percent of the incident radiation and approximate a blackbody. causing a change in electrical resistance. Salt is used instead of glass because glass absorbs too much infrared radiation. or the radiant energy emitted per unit area per unit time in the range of wavelengths between X and X dX. heated by an electric current flowing in a wire wound around the tube. The energymeasuring device may be a lineartype thermopile connected to a galvanometer with high voltage sensitivity. very small differences in radiant energy. It has also been shown that the radiation coming out of the small opening of an almost completely closed.
The curves are similar in shape. but as the temperature increases. The experimental results contained in the curves of Fig. The total radiant energy for a given temperature is represented by the area between the curve and the horizontal axis. K. New York: McGrawHill. Kennard. Measurements made a t different temperatures by Lummer and Pringsheim (1899) yielded a set of curves like that of Fig. The distribution of energy in the spectrum of a blackbody at different temperatures. H. . Introduction to Modern Phyaics. it has a maximum value a t a definite wavelength A . Wien (1893) showed that it was possible to predict some features of the spectral distribution of thermal radiation from the laws of classical physics.531 THE CLASSICAL THEORY OF THERMAL RADIATION 89 . His results will be discussed in some detail in order to understand the difficulties which arose and which could only be resolved by the quantum theory. 53 The classical theory of thermal radiation. 1947. (By permission from F. the height of the maximum increases and the position of the maximum is shifted in the direction of smaller wavelengths or greater frequencies.110 3   F g 15 SO i ? 90 2 7060504030 20  i i j i S S . The distribution of energy in the spectrum of a tungsten lamp. Richtmyer and E. 4th ed..51.) FIG. lengths the monochromatic emissive power is very small. In the intermediate wavelength range. Wavelength (microns) 10 1 L Wavelength (microns)  FIG. 52. 52. 52 are typical of the phenomena which must be explained by theory. This area increases as the fourth power of the absolute temperature. according to the StefanBoltzmann law.
dv is the radiant energy per unit volume in the frequency range from v to v dv. and where the bars indicate the absolute magnitude of the quantity. the amount of it which will strike unit cross section per second is the amount contained in a prism whose base is 1 cm2 in area and whose slant height along the direction of propagation is the velocity of the radiation c. By definition. + + where c is the velocity of light. 5 In the theoretical treatment of the problem of radiation it is convenient t o use a quantity ux. u.90 THE QUANTUM THEORY OF RADIATION [CHAP. The absolute magnitude must be used because the energy density is positive. We shall not need this relationship and shall therefore not show it explicitly. Hence. the proportionality factor contains only geometrical factors and the velocity of light. The energy density and the emissive power are related geometrically in a simple way. u. rather than ux to show that it can be expressed in terms of the frequency rather than of the wavelength. the amount of light falling on 1 cm2/sec can be found by integrating over all directions of propagation. if 9 is the angle between the direction of propagation and the normal to the surface) multiplied by the energy of the radiation per unit volume. and . u~dX is the radiant energy per unit volume in the dX.dv must be equal in magnitude to uxdX. but the energy contained in a cubic centimeter of volume. This amount is the volume of the prism (c cos 0 . called the monochromatic energy density. Now. since each frequency corresponds to one wavelength. The energy density represents. rather than the monochromatic emissive power Wx. From vX = c. (dX/dvl = X2/c. Since this amount is just the monochromatic emissive power. the latter is proportional to the energy density. if the energy density is known. If radiation is propagated in a definite direction. Frequency and wavelength are range of wavelength from X to X related by the expression vX = C . not the energy falling on one square centimeter per second. it will silffice to remember that the monochromatic energy density and the monochromatic emissive power can be used interchangeably so long as absolute values are not required. The energy density can also be written as u. The amount of radiation of a given wavelength falling on 1 cm2 of a body per second a t a given temperature is equal to the amount which 1 cm2 of surface of the body would radiate in one second a t the same temperature (Kirchhoff's law). neither of which depends on wavelength or temperature.
(56)and (58) hold simultaneously. .531 THE OLASSICAL THEORY OF THERMAL RADIATION 91 According to Maxwell's electromagnetic theory of light. Wien also obtained another important result. with the aid of Eq. u A 5 or ux/T6 must be a function of the product AT: where C is a constant and the function f(AT) is. namely. (56). radiation in an enclosure exerts a pressure on the walls of the enclosure which is proportional to the energy density of the radiation. are the values of the temperature and wavelength of the radiation before the adiabatic change. treated the radiation as a thermodynamic engine to which the first and second laws of thermodynamics could be applied. then the absolute temperature changes in the inverse ratio. Wien. undetermined. and T.wavelength. Equation (55) may be written in the more general form A T = constant. that the monochromatic energy density and temperature before and after the adiabatic change are connected by the relationship where the subscript zero again refers to conditions before the change. therefore. Equation (57) may also be written in the more general form TS or. He found that the wavelength and temperature before and after the adiabatic change would be connected by the relationship where To. as yet. U A . (56)  This equation expresses the fact that if radiation of a particular wavelength whose intensity corresponds to a definite temperature is changed adiabatically to another . He considered radiation of a single wavelength and analysed the problem of what would happen to this radiation in an adiabatic expansion of the enclosure that it occupied. A. (584 In order that Eqs.constant. A are the values after the change.
T and UA~T'. the fourth and fifth columns of the table give the experimental value of the products A. constitute Wien's displacement These predictions can be compared with the experimental data of Lummer and Pringsheim. and Wien's displacement law holds. (56). and if . or (&lo). Eq. 52) is obtained from the first one by displacing each abscissa toward the origin in the ratio of distances. It is evident that there is no significant variation in either product over a wide range of values of the different quantities. 5 Observed T:deg. the Wien displacement constant.. More recent and more precise experiments give for the best value of this quantity. " Maximum relative energy density. then. (510) can be written as . Fig.K Wavelength for maximum energy density.0006 cmdeg. Theory of Heat. Since.2940 cmdeg. A = constant/T.. the value 0. he reason for this name is that if the change from one temperature t o a higher one is supposed t o take place by means of an adiabatic compression. h . XmT = constant = a.)~ ~ in the first curve ratio T'/T.A abscissa of a maximum ordinate. X/Xo = To/T. (511) Equations (56) and (57). The average value of the product obtained by Lummer and Pringsheim was 0. shown in Fig. is the corresponds the maximum ordinate ukm of the second. from Eq.. The relation (510) has also been verified by the experiments of Lummer and Pringsheim. UA.. the new distribution curve (cf. Thus. and increasing the ordinate in the or ) (u~.Q  X 1 * From Preston. to a maximum ( W A ~ . by Wien's law.. ~Aln cmdeg " ' . The results of the comparison are listed in Table 51 .respectively.2884 f 0. 52./T' = constant = b.92 THE QUAIWUM THEORY OF RADIATION [CHAP.
53 is shown the composite curve obtained by combining the data taken a t three different temperatures. in Fig.FAILURE O F CLASSICAL THEORY OF THERMAL RADIATION 93 FIG. Kennard. K. (512). (512)the function that fixes the shape of the spectral distribution curve. New York: McGrawHill. body radiation. The theoretical prediction is seen to be fulfilled. Consequently. and any model can be pastulated which appears reasonable. 1947. H. The derivation of Wien's law was based on thermodynamics and did not involve the nature of the radiator or the mechanism of emission of thermal radiation. a single curve should result in accordance with the functional relationship (512). who assumed that the radiation was produced by emitters or oscillators . Wien's law gives the general dependence of the monochromatic energy density on wavelength and temperature. (510) or Eq. The experimental verification of the displacement law for 'black . Richtmyer and E. explicit form of the dependence is affected by the choice of a model for the radiator. This choice is arbitrary.) where C ' is a new constant. The detailed. and a single curve serves to represent the spectral distribution of blackbody radiation at all temperatures. (By permission from F. 54 The failure of the classical theory of thermal radiation. leads to a distribution law of the general form of Eq. The problem of the spectral distribution of blackbody radiation has therefore been reduced to the determination of the single function f(AT) in Eq. The first attempt t o determine the function f(AT) was by Wien (1896). 4th ed. Znltoduction to Modern Phyeica.53. and agrees with experiment. If the experimental values of the quantity udTs are plotted against the values of the product AT.
then. Wien derived the following expression for the distribution: u ~ = . but a t longer wavelengths i t predicts values of the energy density which are too small. therefore. (513) where cl and c l are constants. Each standing wave of radiation may be considered to be caused by an electric dipole acting as a linear harmonic oscillator with a frequency v. . Another attempt to obtain a distribution law resulted in the RayleighJeans formula. The thermal radiation consists. In such a cavity there can be standing electromagnetic waves. conf the average energy carried by each stitute the thermal radiation. the energy density of the radiation can be obtained. for convenience. assume that the cavity is rectangular. however. is so satisfactory at short wavelengths that any successful radiation law must reduce approximately to such a function in this region of the spectrum. The number of these oscillators with frequency between v and v dv is given by Eq. With these assumptions. Consider a hollow cavity with perfectly reflecting walls and.94 THE QUANTUM THEORY OF RADIATION [CW. in fact. these wavea. (514) and each oscillator has an energy t which may take on any value between 0 and oo. This formula was derived in a much more general way than Wien's expression. and if the number of standing waves with frequency between v and v dv can be calculated. however. There is. (513). and that the intensity in any particular wavelength range was proportional to the number of oscillators with the requisite energy.e A6 . which is not the case in Eq. A brief outline of the ideas involved follows. 5 of molecular size. a certain average value Z of the energy of the oscillator that can be obtained from classical statistical mechanics. The number of waves has been calculated and is equal to 8m2 dv . It was. The exponential function. just about the best that classical physics could do. N(v) dv = c3 (514)  . that the frequency of the radiation was proportional to the kinetic energy of the oscillator. of the electromagnetic waves emitted by a large number of such oscillators. +  The average energy carried by each wave can also be calculated. I wave is known. The distribution formula also fails at high temperatures because ux (and W A )should increase beyond all limits when the temperature increases to an unlimited extent. C1 c21AT . say from 1 to 3 X lo' cm. and was shown to be a necessary consequence of classical dynamics and statistics. According to + . This distribution has the required general form and was found to fit the experimental curves quite well a t short wavelengths.
in this region the intensity of the radiation increases with the square of the frequency. the energy radiated by a blackbody in a given range of wavelength increases rapidly as X decreases. for convenience. then  The energy density of the radiation is. (515).which would mean that the total energy radiated per unit time per unit area is infinite at all finite temperatures.. Equation (515) is the RayleighJeans formula. The average energy 2 is obtained by averaging over all values of e. (516). since for very small wavelengths the energy density (or monochromatic emissive power) actually becomes vanishingly small. the energy carried by all wavelengths would be  and the integral is infinite for any value of T other than T = 0. when the oscillators are at equilibrium. because the total energy radiated at any temperature is actually finite. with this weight factor. then. and approaches infinity as the wavelength becomes very small.541 FAILURE OF CLASSICAL THEORY OF THERMAL RADIATION 95 that theory. The experimentally observed radiation curve is in complete disagreement with this conclusion. the formula fails. Set j 3 = l/kT. and may also be written in the form The RayleighJeans formula agrees well with the experimental intensity distribution for long wavelengths. small values of X. the RayleighJeans formula also fails to account for the observed dependence of radiation on temperature. ie. however. Hence. For large values of the frequency. the value e for any energy of the oscillator occurs with the relative probability ef'kT. According to Eq. This conclusion is false. where k is Boltzmann's constant. Furthermore. as in Eq. .
or quantum. . called P h c k ' s conslant. Planck's distribution law for thermal radiation is therefore  or. At low . with the aid of Eq. . (515). It is interesting to compare Planck's radiation law with the Wien distribution Eq. but can only take on values equal to 0 or €0 or 2to or 3c0. . (&lo). or mo. . or T/v.96 THE QUANTUM THEORY OF RADIATION [CHAP. The average value is now The energy density is then. The problem of the spectral distribution of thermal radiation was solved by Planck in 1901 by means of a revolutionary hypothesis. (514). of energy and n is an integer. in terms of wavelength. Hence. Planck postulated that a linear harmonic oscillator like that d i s c 4 in the last section does not have an energy that can take on any value from zero to infinity. or where h is a new universal constant. This formula must have the same general form as Wien's law. 5 55 Planck's quantum theory of thermal radiation. or v/T. Eq. (513) and the RayleighJeans formula Eq. finite amount. where to is a discrete. and the temperature must appear in the combination AT. co must be proportional to v. The average energy of an oscillator is obtained in an analogous way t o that used in deriving the RayleighJeans law except that sums are used instead o f integrals.
and h is Planck's constant. known to be valid at long wavelengths. Thus. the product AmTis a constant. (521) can be neglected in comparison with the exponential. k is the Boltsmann constant. c. The value of . Planck's radiation formula reduces to forms that agree with experiment a t the extremes of the wavelength scale. the familiar fourth power law. At frequencies such that hv/kT > > 1. which is Wien's formula (513). the Planck distribution gives a maximum value for the energy density. (520) can be expanded. and . (521).T was known to be 0. and k. we get the result In Eq. c is the velocity of light. frequencies. the experimental value of the product A. and the exponential function in the denominator of Eq. It can also be shown that Planck's radiation law leads to the StefanBoltzmann law.294 from the experiments of Lummer and Pringsheim (Table 51). the 1 in the denominator of Eq. also. in agreement with experiment. with u~ given by Eq. The total energy density is given by This integral can be evaluated. In 1901. Then and a t low frequencies (long wavelengths) the Planck formula reduces to the RayleighJeans formula (515). The position of the maximum can be found from the condition dux/& = 0. Omitting the details of the calculations. (522). Hence. and the resulting expression for u is u = aT4. known to be valid at short wavelengths. the value of c was known quite well. In the intermediate wavelength range. with (523) Equations (522) and (524) are two relations containing the three constants. h. hv/kT < < 1.
and the best. Kennard. H.98 THE QUANTUM THEORY OF RADIATION [CHAP. More recent and more precise experimental determinations of h and k by a variety of methods give h = (6. the agreement is excellent over the entire wavelength range I I 1 I I 1 2 3 4 5 Wavelength (microns) Fro. The ordinate represents Wh on an arbitrary scale.) .346 X lo'' erg/deg. 54. 1947.T. of Boltzmann's constant. the shape of the spectral distribution curve is completely determined. K. (520) or Eq. With the known values of a. 54 for the case of blackbody radiation at 1600°K. and the circles represent experimental points. this is done in Fig.A comparison o f the Planck.55 X ergsec. and the solid line is the distribution according to the Planck radiation law.62517 f 0. 5 the constant a was known from experimental studies of the StefanBoltzmann law.00023) X ergsec. Z n l r o d w t h to Modetn Phyaics. Richtmyer and E. Planck's distribution law can then be compared directly with experiment. This was the first calculation of Planck's constant. 4th ed. When the above values are inserted into Eq. and k = 1. New York: McGrawHill. .061 X 10l5 erg/cm3(OK)'. up to that time. Wien. Planck obtained h = 6. and RayleighJeans radiation laws with experiment at 1600°K. The points are the experimental results for the energy distribution a t diierent wavelengths. (521). a = 7. and c. (By permission from F. X.
or any other physical system capable of emitting electromagnetic radiation. For comparison. For a gammaray with a wavelength erg. say X = 10 cm.99 X 1017 erg. however. and is a consequence of Planck's postulate of quanta of energy. For a short radio wave. For a visible light quantum of 10lo cm.5 6 1 QUANTUM THEORY AND THE PHOTOELECTRIC EFFECT 99 of these experiments. and there is no doubt that the Planck radiation law is the only one which satisfactorily accounts for the experiments. 56 Quantum theory and the photoelectric effect. These results are typical of those found over a wide range of temperatures.99 X cm. but represents a radical. not . where v is the frequency of the emitted radiation. We are thus led to the idea that an oscillator. in ergs. The magnitude of the energy associated with a quantum of radiation can be obtained as a function of wavelength. The idea of discrete energy levels will be seen to be a fundamental one in atomic and nuclear physics. empirical modification of classical ideas. and so on. Equation (519) can also be written Then eo is given. by . The theory was successful. in general. the theoretical predictions of the Wien distribution (513) and the RayleighJeans formula (516) are also included. the quantum is 1.0 X 10l2 with a wavelength of 5000 A = 5 X erg. the energy is 4. Emission can take place only when an oscillator makes a discontinuous transition from a state in which it has one particular energy to another state in which it has another energy different from the first by an amount which is an integral multiple of hv. then the emission and absorption of radiation must be discontinuous processes. a discrete set of possible energy values or levels. needed in order to bring theory and experiment into agreement. the quantum is 1. energies intermediate between these allowed values never occur. This idea is not an outgrowth of classical physics. If this is so. has. The revolutionary nature of Planck's theory is contained in the postulate that the energy of an oscillator can vary only in discrete jumps. where )i is expressed in centimeters. The concept of discrete energy states and quanta of energy was so different from the classical ideas to which physicists were accustomed that Planck's theory met with strong opposition.
2. The photoelectric effect was studied very carefully and the most important experimental results may be summarized as follows: f light of a given frequency can liberate electrons from a surface. bllt also in the solution of other problems for which classical physics failed. of course. According to Emstein. E i t e i n deduced his photoelectric equation w2 = hv . One of thew problems was that of the photoelectric effect. 3. concentrate enough energy on one electron to knock it out of the metal. and is independent of the intensity of the incident light. Light of frequency greater than this critical value causes the irnrnediate emission of electrons. accounts successfully for the photoelectric effect. in a very short time interval. I t is now known. 4. 5 only in explaining the phenomena of thermal radiation. I the electron current is proportional to the intensity of the light. the time interval between the incidence of the light on the metallic surface and the appearance of electrons is not more than 3 x aec. light of smaller frequency cannot eject electrons no matter how long it falls on the surface or how great is its intensity.100 THE QUANTUM THEORY OF RADIATION [CW. With this hypothesis. For a given metallic surface. 1. Furthermore. Hertz (1887) discovered that a metallic surface can emit electricity when light of very short wavelength falls on it. that the particles are electrons. Einstein proposed a mechanism for the photoelectriceffect. that could account for the experiments.A. He assumed that the energy emitted by any radiator not only kept together in quanta as it traveled through space. the wave theory of light waa unable to account for the fact that the maximum energy of the ejected electrons increases linearly with the frequency of the light but is independent of the intensity. The maximum kinetic energy of the emitted electrons is a linear function of the frequency of the light which causes their emission. (526) This equation states that the maximum kinetic energy of the emitted . There w which a train of light waves spread out over a large number of atoms could. These experimental facts could not be explained on the basis of the a s no way in classical electromagnetic wave theory of light. there is a smallest value of the frequency for which the incident light can liberate electrons. but that a given source could emit and absorb radiant energy only in units which are all exactly equal to hv. as w i l l be seen. In 1905. light itself consists of quanta or corpuscles of energy hv which move through space with the velocity of light. based on Planck's idea of quanta of energy. This hypothesis is completely a t variance with the wave theory of light but. and in 1898 Thomson showed that the e/m value of the emitted particles was the same as that for cathode rays.
 . no electrons can be emitted. of radiation is also called a photon. the result should be a straight line with a slope whose magnitude is h/e. If the frequency is above the threshold value. potassium.(A/e) . it loses a part A of this energy needed to remove it from the metal. The energy of the ejected electrons was measured by applying to the metallic surface a positive potential just strong enough to prevent any of the eIectrons from carrying a charge to an electrometer. then the kinetic energy is given by h u 2 = Ve. it was not verified quantitatively until 1916 when Millikan performed a very beautiful experiment. Thus. a photon of radiation of frequency v carries an amount hv of energy. for a given metal. (526) agrees qualitatively with the experimental facts. He studied the effect of light of a range of frequencies on sodium. A quantum of light or. in all cases. The number of electrons ejected is proportional to the number of incident light quanta and therefore to the intensity of the light falling on the metal. the photon may be described as a "bundlen or "particle" of radiation. (525). If the frequency is so small that hv 5 A. a light quantum striking the metal and colliding with one of its electrons can give up all of its energy to the electron and knock it out of the metal. The experimental results gave. E i t e i n ' s photoelectric law has also been found to be valid for the electrons ejected by xrays and 7rays. it was found to be in good agreement with the value obtained by Planck from the measurements on blackbody radiation. or V = (h/e) v .A . as predicted by Einstein's theory. Combining this relation with Eq. (527) If the potential V is measured as a function of the frequency of the incident light. and lithium. the value of h could be determined. a good straight line. where e is the electronic charge. so that the frequency has a threshold value.561 QUANTUM THEORY AND THE PHOTOELECTRIC EFFECT 101 electron is equal to the energy hv of the incident light quantum minus a quantity A which represents the amount of work needed by the electron to get free of the surface. and there is no doubt of its general validity. Although the Einstein photoelectric equation accounted for the experiments qualitatively. If this potential is equal to V volts. more generally. Since the value of e had already been determined by Millikan. But before the electron emerges. The energy of the electrons varies linearly with the frequency and is independent of the intensity. below which no photoelectric emission can occur. I t is evident that Eq. we get V e = hv .
2. where again there was experimental information which could not be explained by classical theory. Suppose that the blackbody of Problem 1 is surrounded by walls kept a t a temperature of O°C. 1956. Photons. aided its acceptance. Chapter 4. Use k = 1. Give the results in terms of ergs and calories. Park 111.. W ° C . The best value of the Wien displacement law constant is now thought to be 0. 1933. E. 300°C. 1955. 2000°C.3804 X 10l6 erg/OK and c = 2. PRESTON. Calculate the value of h. Chapter 6. New York: Interscience. New York: McGrawHill. A. New York: Academic Preae. Miller. F. 1. SOMMERFELD. 4th ed. Ezperimenful Atomic Physics. PLANCK. R. Derive an expression for the total energy density in a blackbody enclosure by integrating Planck's distribution law over all possible values of the frequency. Calculate the value of h. K. Sections 20.. 5th ed. Theory of Heat. 1951... use the values of k and c given in Problem 3. and T. RICHTMYER. ed. These applications of the quantum theory by Einstein helped establish the theory. A. K.99793 X 101° cm/sec. 21. R. H. London: Macmillan Co. Protone. and laid the groundwork for the application of quantum theory t o the problem of atomic structure. Heat and Thermodynamics. J. Glasgow: Blackie. HARNWELL and J. 1947. (+ and ). 4th ed. Chapters 20. G. 1000°C. 1929. Derive. J. GENERAL M. MILLIKAN. London: Macmillan Co. LAURITSEN. l W ° C . . Chicago: University of Chicago Press. LIVINGOOD. KENNARD. J. Calculate the total emissive power of a blackbody a t the following temperatures: O°C. P. Zntrodwtion to Modern Physics. the expression (522) for the wavelength a t which the radiation density has ita maximum value. Mesotrons. ROBERTS. Thermodynamics and ~tutistical Mechanics. and Cosmic Rays. 5 The quantum theory was also applied successfully by Einstein (1907) t o the problem of the specific heat of solids and polyatomic gases. Ekctrons. New York: McGraw=ll. Chapter 2. The StefanBoltzmann constant u has the value 0.563 X 10l5 erg/cm3.5669 X lo4 erg~m~eecl (OK)+. T.3. 35. . The Theory of Heat.102 THE QUANTUM THEORY OF RADIATION [CHAP. 4. 100°C. revised by A. Cotter. 1929. Note that 4 The best value for the total energy density is given by a = 7. from Planck's radiation law. R. IV.2884 cmOK. and 3000°C. Calculate the net rate of energy loss by the body to the walls a t each of the temperatures listed in Problem 1.
124 X 10l5 volt/cycle/sec. 5000 A. L i . The emission of xrays as a result of the incidence of highspeed electrons on a metal surface. 4 . The first precise determination of h was made by Millikan. 4. (a) What is the work function (in ev) of the surface? (b) What is the maximum wavelength that the light can have and eject any photoelectrons? (c) If an electrometer is connected across the photocell. Ni. Calculate the frequency and the amount of energy associated with a quantum of electromagnetic radiation a t each of the following wavelengths: 1 km. to what potential will i t be charged when ultraviolet light of wavelength 1000 A is used? 10. The photoelectric work functions for samples of some metals are Ag. 1 mm. 1 cm. who measured the energy with which photoelectrons are ejected as a function of the frequency of the incident light. 5 ev.PROBLEMS 103 5. 9. 10l5 cm..4. 1 m. 3 ev. What are the wavelength. 8. Ba. frequency. He found that the potential V needed to keep the electrons from sodium from reaching an electrometer increased with frequency by 4. Show that A. Show that the velocity of the fastest photoelectrons is related to the stopping potential by the equation where v.000 volts? At lo6 volts? At 5 X lo5 volts? At lo8 volts? . in volts. Give the energy in ergs and in electron volts. and energy (in ev) of the xrays from a Coolidge tube operated a t 50. 10l3 cm.803 X 10lo esu.0 ev. Pt. 1 A. when the samples are illurninatid with ultraviolet light of wavelength 2000 A? (c) What is the stopping potential. as in an xray tube. (A) = 12395/V (volts). 2 . 1 ev. What is the resulting value of h? Note that 300 volts = 1 esu of potential.. . may be regarded as an inverse photoelectric effect. 5. = eV. and use e = 4.2 . Photoelectrons emitted from a photocell by light with a wavelength of 2500 A can be stopped by applying a potential of 2 volts to the collector. 6. is in centimeters pitr second and V is in volts. lo4 cm. (a) What is the wavelength a t the threshold in each case? (b) What is the maximum velocity of the ejected photoelectrons in each case. where V is the anode voltage.5 ev. The maximum frequency of the emitted radiation is given by the relation hv. 1000 A. for the photoelectrons of maximum velocity? 7..
Such a discussion should dispel some of the aura of magic that surrounds the relativity theory and even its simplest applications. One idea is that of the variation of the mass of a particle with its velocity. The theory of relativity may be introduced in either of two ways. Two problems will be considered. The theory can then be applied to physical problems to see how it meets the test of comparison with experiment. and its use should be accompanied by some understanding of its origin and meaning. on occasion. leads to the special theory of relativity. (2) the problem of the invariant form o f physical theories. and to see how these dilemmas could be resolved only by the revolutionary ideas contained in the relativity theory. to modify the commonly accepted meanings of these ideas. The massenergy equation is used often in nuclear physics. together with certain postulates. the second is that of the proportionality between mass and energy. expressed by the equation E = me2. . as a consequence of this examination.has achieved a certain degree of notoriety in recent years and has even. The second method is to consider certain problems which led to serious dilemmas in physical thought at the end of the 19th century. The second great theory of 20th century physics that is indispensable to the development of atomic and nuclear physics is the special theory of relativity proposed by Einstein in 1905. One way is to examine very carefully what is meant by such concepts as space.CHAPTER 6 THE SPECIAL THEORY OF ~ E L A T M T Y 61 The role of the special theory of relativity in atomic and nuclear physics. The second method is the one that will be adopted here because it is more closely related to experiment and is leas abstract than the first method. . A real understanding of the relationship can come only from a careful study of the relativity theory. This logical process. time. these are (1) the ether and the problem of absolute velocity. occupied paeitions of prominence in the public press. and should allow the student to use mass and energy as interchangeable quantities without feeling uncomfortable. . 104 . The latter relationship. but some familiarity with the ideas involved in the massenergy equation can be obtained from a short discussion of the background of experiment and theory which ied to relativity. The first applications of the relativity theory which will be met in atomic and nuclear physics depend on two closely related ideas. and simultaneity and.
For these reasons. if light and the earth both move with definite velocities relative to the stationary ether. was supposed to be a weightleae substance permeating the entire universe. The ether was assumed to be stationary. and a body such as the earth was supposed to move through it without producing any disturbance. Because of the success of the wave theory of light in accounting for optical phenomena. called the ether. while the same distance in the opposite direction will require the time L/(c u). and that light waves travel through the ether with a velocity c. from which light is emitted.62 The ether and the problem of absolute velocity. the theoretical basis of the most famous one can easily be derived. Consider a source. The assumption that there is an ether raised problems in connection with the motion of material bodies. if a ray of light were to travel a distance L in the direction of the earth's motion. I t was known from astronomical data that light can pass through space practically devoid of matter. its velocity relative to the stationary ether. the velocity of the light relative to the earth will be c .v. involving the propagation of light. These assumptions were not contradicted by astronomical evidence.v ) . it should be poasible to devise an experiment. If the direction of propagation of the light is the same as that of the earth's motion.e. f i x e d on the earth. L + L = c+o c . Therefore. i. the time required would be + + tlj = 2cL . it was thought that the measurement of absolute velocity should depend on the effect of motion on some phenomena involving l i t . The interpretation of the phenomena of the reflection. physicists filt it necessary to postulate the existence of a medium in which light waves could be propagated even in the absence of matter. Many such experiments were proposed. Since the ether was supposed to be the medium for the propagation of light. from which the absolute velocity of the earth could be deduced. the ether gradually came to be accepted as a physical reality.. . Thus. Suppose that the earth travels through the stationary ether with a velocity v. and it was inferred that it should be possible to determine the absolute velocity of a body. The propagation of light waves seemed to require a medium. and then be reflected back and travel the same distance in the opposite direction. This medium. refraction. and diffraction of light gave rise to the wave theory of light. If the light travels in the direction exactly opposite to that of the earth's motion.o c2tr2 . its velocity relative to the earth is c v. The time taken for the light to travel a certain distance L in the direction of the earth's motion is L/(c . just as water or sound waves cannot be separated from their respective media.
t~ and the ratio t l l / t ~both depend on the quantity (u/c)l.. The orbital velocity of the earth is 30 km/sec. we have c2t2 = L2 v2t2. From an experimental viewpoint. Hence. take longer for light to traverse the reflected path in the direction parallel to that of the earth's motion than to traverse the same distance in the direction perpendicular to that of the earth's motion. the diiTerence between tII and t~ is important. and the effect should be diflicult to detect. If the time required is t. so that AA' = vt. The actual path of the light ray is AB'. therefore.106 THE SPECIAL THEORY OF RELATIVITY [CHAP. or t2(c2 . and the point B has moved to B'. While this ray travels a distance L. In the same time interval.e. equal to AB. 6 Consider next a ray of light propagated in a direction perpendicular to that of the earth's motion (Fig. A has moved to A' with the velocity v. This difference is If we neglect terms of order higher than v2/c2. so that v/c = lo' and (v/c)' = lo'.v2) = L2. In the rightsngled triangle AA'B'. same length of time will be needed for the return trip. i. then AB' = ct. to A". or t = ~ / d n The . the time required to travel a distance L and back in a direction perpendicular to that of the earth's motion is + The ratio of the times required is It should. The difference 111. we may w r i t e Hence. . 61). the earth has moved from A to A'.
B a s i s o f an experiment on the velocity o f the earth relative to the ether. according to Eq. The theoretical relations derived in the last section were tested experimentally by Michelson and Morley (1887) who used an ingenious instrument. the ratio of displacement to wavelength is given by (2L/k)(v2/c2). the other beam is reflected by P to the mirror S 2 and is then reflected back to P. light from a source Q is divided into two perpendicular beams by a halfsilvered mirror P. and v2/c2 = Fro. the two light beams should arrive a t F at slightly diiTerent times. shown schematically in Fig. the interferometer. the arm which was originally in the direction of the earth's motion is now perpendicular thereto.631 THE MICHELSONMORLEY EXPERIMENT 107 63 The MichelsonMorley experiment. I interferometer is rotated through an angle of 90°. The frequency of the light used in the experiment is v. (63). Then the ratio of the displacement to the period is given by (2L/cT)(v2/c2). . If the position of the pattern is observed during the rotation of the interferometer. The two reflected beams pass through P to a f the distances PS1 and PS2 are telescope or photographic plate F.9 X lo' cm. and the interference pattern should be displaced to the opposite side of the position corresponding to v = 0. One beam passes through P to a mirror S1 and is reflected back to P. FIG. and if one arm of the interferometer is parallel to the direction of the earth's motion. 61. I equal. In the experiment of Michelson and Morley. causing an interference pattern (interference fringes). the fringes should be shifted by an amount correspondiig to the time interval 2(L/c)(v2/c2). the path length L was 1. Since the wavelength A = cT. 62. as a result of several reflections of the light beam. The MichelsonMorley esperiment: schematic diagram. The wavelength of the light used was about 5. and its period (inverse frequency) is T = l / v .62. on the absolute velocity of the earth and should be displaced from the position f the corresponding to v = 0 by an amount proportional to (L/c) (v2/c2). In the MichelsonMorley experiment.1 X l o 3 cm. The position of the pattern should depend.
There were also experimental objections to the idea that the earth carries the ether along with it.108 THE SPECIAL THEORY OF RELATIVITY [aw. and the uniqueness of "the stationary ethern as the medium for the propagation of light would disappear. especially by astronomical studies of double star systems and stellar aberration. (2L/h)(v2/c2)= [2(1. the velocity of the earth with respect to the ether. there was no significant difference between the rates at which the light traveled in the two directions at right angles to each other. the velocity of light as measured by the assumed observer would depend on the velocity of its source. were moving toward the observer. therefore. The simplest interpretation of this result seemed to be that v. when the interferometer was rotated through 90°. Thus. In the experiment.37. The observed velocity of light is always the same regardless of whether or not the emitting source moves. to a serious contradiction with experiment. This conclusion is contradicted by ail observations on the velocity of light. and this fact made it very diflicult to explain the negative result of the MichelsonMorley experiment. Thh result follows from the general behavior of waves in a moving medium. the velocity of the light wave8 from the source would appear to be greater than if the earth were moving away from the observer. and that light waves from this source are transmitted through the ether and eventually reach the observer. If the earth. a conclusion which led to serious difficulties. He assumed that when a body travels in a direction + . and the MichelsonMorley experiment showed that the ratio of the times of travel in Eq. together with the'ether. or how it moves. That is. Thus. Suppoee that in 8 there is an observer who is looking at a light source on the earth. the interference fringes should have been displaced by more than of a wavelength. a displacement of onehundredth of this amount could have been observed."' The theoretical value of the ratio of displacement to wavelength was derived by a simple and straightforward application of some of the fundamental ideas of classical physics. the velocity of the waves relative to a stationary obeerver should be equal to the sum of the velocity of the waves in the medium and the velocity of the medium. in which case the vdocity of the earth relative to the ether should be zero. or in many later repetitions. is zero. The simplest explanation which could be proposed was that the earth carries the ether with it. (61) is unity.6 Hence.9x l o ' ] = 0. Fitzgerald (1893) introduced a hypothesis to account for the negative result of the MichelsonMorley experiment while preserving the idea of the stationary ether. but no significant displacement was observed in the original experiment. This explanation was unattractive because every moving body would presumably carry its own ether with it. and had to be rejected.1)103 X 10~]/[5. The assumption that the earth cames the ether along with it led.
but the length of the perpendicular arm remains equal to L. if the mass of a moving electron is m. who was a t the same time trying to develop a consistent electromagnetic theory of the electron. it led to theoretical results consistent with experiment. where tois the radius of the electron when it is a t rest. Lorentz assumed that a moving electron contracts in the direction of its motion by the Fitzgerald factor 4 1 . because a measuring instrument placed along the arm would shrink in the same way. and if the mass when the electron is a t rest is mo. (65). it follows that I According to Eq.631 THE MICHEL~NMORLEY EXPERIMENT 109 parallel to that of the earth's motion it contracts in length. Early experiments on the value of elm for 8rays indicated that the value of e/m decreases with increasing velocity of the rays. The light does not have to travel as far in the parallel direction as in the perpendicular direction. and its main object waa to preserve the concept of a stationary ether. Lorentz. it was concluded that the mass of the electron does indeed increase with its velocity. Since there was no reason to suppose that the charge varies with velocity. and e is its charge. But the hypothesis lacked a sound theoretical basis. and classical physics was once again confronted by a . although the LorentzFitzgerald contraction hypothesis seemed physically unreasonable. According to this hypothesis. The Fitzgerald contraction seems absurd at first sight because it is not the result of any forces acting on a body but depends only on the fact that the body is in motion. he also assumed that no such contraction occurs in a direction perpendicular to that of the earth's motion. the maw of an electron should increase with its velocity. the mass of an electron could be expressed by the formula . Fitzgerald's hypothesis was purely ad hoe.(v2/c2). found the hypothesis useful in h i work and incorporated it into his theories. Then. so that the radius becomes r o d l . and the difference is such as to make the theoretical value of the shiit of the interference pattern equal to zero. Thus. the arm of an interferometer that is parallel to the direction of the earth's motion contracts in length from L to ~ d1 (v2/c2).(v2/c2). The contraction in the length of one arm of the interferometer would not be observed.(v2/c2) when the electron moves with a velocity v. becomi~g shorter by the factor 4 1 . It had previously been shown that under certain assumptions. especially when the latter reaches a significant fraction of the velocity of light. so that the hypothesis is now known as the LorentzFitzgerald contraction hypothesis.
and so on."' In 1958. 64 The problem of the invariant form of physical theories.  . and when the appropriate mathematical operations are performed. 63). y. momentum. the calculated values of the measurable quantities should agree with the values measured in suitable experiments. of equations or formulas which relate measurable quantities such as distance and time to invented quantities such as energy. . Suppose that there is another observer who is at rest with respect to another system of reference S' which moves with uniform velocity v along the zaxis. other testa of the possible motion of the earth relative to the ether have also been made. potential. the direction of which coincides with the direction of the z'axis. . 6 problem (the negative result of the MichelsonMorley experiment) that it could not really solve. At time . For example. The concept of the invariant form of physical theories is a fundamental one in science. To see what this means. and of time intervals t. A physical theory consists. in an extremely sensitive test which inclock^. Numbers can be assigned to the invented quantities. The MichelsonMorley experiment has been repeated under different conditions and with modifications. field strength. essentially. No significant motion has been detected.03 km/sec. and when they are inserted into the formulas of the theory. or less than 0. Two Cartesian reference systems in uniform relative translatory motion.110 THE SPECIAL THEORY OF RELATIVITY [CHAP. y. suppose that there is an observer who is at rest with respect to some system of reference S (Fig.'^' it was found that the maximum volved the use of nwT velocity of the earth with respect to the ether is less than 1/1000 of the l l show earth's orbital velocity. FIG. It is then highly desirable that the formulas which make up a theory be the same for different observers. The measured values can usually be expressed in terms of positions in some coordinate system such as x . z.63. He measuMs position in terms of the coordinate system (z. let us consider what is meant by a physical theory. These results a that the hypothesis of a stationary ether through which the earth moves is not verified by experiment. z) and measures time intervals t on a clock.
the two reference frames are assumed to coincide. the substitutions (66) give a correct description of the same experiment in terms of the varisbles x . When the transformation (66) is applied to the theory of electromagnetics. (66) Equations (66) connect the variables of the two systems S and S' in such a way that if an experiment involving the motion of a body has been measured by an observer moving with the system S' and has been described in terms of the quantities z ' . and t will appear. used by an observer moving with the system S. y' = y. the quantities x'. y'. Each observer will make measurements appropriate to the experiment. y'. t' = t. z'. one of whom is moving with a uniform velocity v with respect to the other. the quantities z. In the formulas of the observer S. in those of S'. y'. z'. I t was found long ago that the laws of classical mechanics as expressed by the two observers S and S' have the same form provided that the coordinates used by the two observers are related by means of the transformation equations: 2 ' = z . 2' = Z. and will describe the reaults by means of formulas. Lorentz. time as measured by an observer in the 8' system is different from that measured by an observer in the S system. Now let both observers S and S' perform an experiment which involves the motion of a body. The system of transformation equations that he derived is known as the Lorentz tramfotmattion. the formulas of the theory are not invariant.vt. and depends on the velocity with which the 8' observer is moving with respect to the 8 observer as well as on the space coordinate . In more formal language. t. who developed a theory of electrons based on classical electromagnetics. z.641 THE INVARIANT FORM OF PHYSICAL THEORIES 111   t = 0. the laws of clsssical mechanics are said to be invariant with respect to the transformation (66). and is given by the following relationships: These equations are remarkable because they indicate that space and time are not independent entities but are related. t'. solved the problem of what transformation equations would lead to the same formulas for his electron theory for two observers. The second observer uses the spatial coordinates z'. y. the formulas of an observer in the moving system S' are not the same as those of an observer in the S system. z. and t' will appear. z' and the time coordinate t'. y.
and this velocity v is the same velocity with which S' moves with respect to S. however. does lead to the LorentzFitzgerald contraction. that is. when it is in uniform motion relative to a stationary observer. But what value does the observer S measure for the length of the body which is now moving with respect to him? He measures a value L given by L = 2 2 . The relation between z' and z.112 THE SPECIAL THEORY OF RELATIVITY [CHAP. k t it 6e set into motion relative to S a t such a speed that it is a t rest in S'. while its dimensions perpendicular to the direction of motion are unaffected. it moves with a velocity v relative to S. (67). The difficulty with accepting these results was caused by the lack of a consistent physical basis for them. as can be shown quite easily.(v2/c2). it gives just the contraction needed to account for the negative result of that experiment. The length'will then be Lo as measured in St. we get According to this relation. The S' observer measures the length as If we substitute for z i and z: the values calculated from the first of Eqs. appears contracted in the direction of its relative motion by the factor 4 1 .Zl. . Thus. the requirement that the equations of electrodynamics be invariant under the transformation equations (67) leads to the LorentzFitzgerald contraction. a rigid body. Suppose that a body. where zz and xl are the coordinates of the ends of the body in the S system. and it became necessary to revise some of the fundamental ideas of physics. Classical physicists found i t impossible to accept this relationship between the space and time variables because it was contrary to the fundamental classical idea of the independence of space and time.since the length must have the same value in any reference system in which the body is at rest. 6 z. has a length Lo in the direction of the xaxis. when at rest in S. When this relation is applied to the interferometer arms in the MichelsonMorley experiment.
651 THE VARIATION OF wss WITH VELOCITY 113 6 5 The special theory of relativity. Hence. In formulating the relativity theory. only relative velocities can be measured. it would be possible to ascribe some absolute meaning to different velocities. according to Einstein. He therefore took as his first postulate : POBTULATE 1. Einstein took: 2. and it is impossible to ascribe any absolute meaning to different velocities. . The laws of physical phenomena are the same when stated in terms of either of two reference systems moving at a constant velocity relative to each other. i. who proposed the special. The ether was as close to an "absolutely stationary" reference frame as classical physics could come. it is believed to represent an experimental fact. I t follows that the general laws of physics must be independent of the velocity of the particular system of coordinates used in their statement because. The general theory is not needed to describe most atomic and nuclear phenomena and therefore will not be discussed.the variation of mass with velocity. . The velocity of light in free space is the same for all POSTULATE observers. Einstein deduced the Lorents transformation equations as well as other kinematical relationships. The Michelson . and is independent of the relative velocity of the source of light and the observer. For his second postulate. But this is just another way of stating the requirement that the equations of a physical theory be invariant with respect to coordinate systems moving with different velocities.e. The term "special theory of relativity" refers to the restriction in the first postulate to reference systems moving at a constant velocity relative to each other. if this were not so. Einstein adopted this requirement of invariance as one of the two basic postulates of the theory. theory of relativity. or restricted.. Einstein interpreted the negative results of the MichelaonMorley experiment to mean that it is indeed impossible to detect any absolute velocity through the ether. From his two postulates. Einstein was restricted by mathematical difficulties to the consideration of reference systems moving a t a constant velocity relative to each other. In his first statement of the theory. The second postulate may be regarded as a result of the MichelsonMorley experiment and other optical experiments and of astronomical observations.    The problems raised by the MichelsonMorley experiment and the requirement of invariance were solved in 1905 by Einstein. I t was not until 1916 in the general theory of relativity that Einstein was able to show that physical laws can be expressed in a form that is valid for any choice of spacetime coordinates.
it is also desirable that the transformation equations be linear. (66) do not lead from Eq.114 THE SPECLAL THEORY OF RELATIVITY [CHAP. 6 . z = z' = y = y' = z = z' = 0. a. t' = az + bt. and another transformation is needed. + Z t 2 = . z' = z. (69) where r . so that for ordinary mechanical phenomena. The interrelationship between space and time coordinates also follows from the postulates and represents a more advanced treatment of these ideas than does the earlier classical theory. Suppose that the system St is moving a t uniform speed v in the xdirection relative to the system S.vt). The need for the relativity theory in atomic and nuclear physics comes in part from the fact that fundamental particles such as electrons can travel with velocities approaching that of light. (68a). that when v is much smaller than c. + y2 + z2 = c2t2.  Morley experiment could be interpreted directly in terms of these relationships and there was no need for the Fitzgerald contraction hypothesis or for the ether. the result in the S' system must be zt2 + y12 (68b) with c. the same in the two systems. the speed of light. The relativity theory d m not "overthrow" the classical theory but rather extends and modifies it. with the result that the kinematic and dynamical consequences of the Lorentz equations are important. the distance traveled by the light ray is measured by observers in the two systems. however. In the S system the result is z2  . I t is easily verified that Eqs. by two observers. The second and third equations are chosen . the classical or Newtonian theory is adequate." I t must be pointed out. The first relation is chosen because it is similar to the first of the classical equations (66) and should reduce to that equation when v is small. y' = y. The interdependence between the space and time coordinates means that time must be treated in the same way as the space coordinates. ( 6 w according to Einstein's postulates. (68b) to Eq.+. of the distance traveled by a light ray. As a more general transformation we try the equations x' = r(x . At some later instant. The Lorentz transformation equations can be derived quite easily from Einstein's postulates if we consider measurements. This result gave rise to the ideas of a fourdimensional spacetime continuum as well as to popular misconceptions about the "fourth dimension. b are constants to be determined. the Lorentz transformation equations (67) reduce to the classical equations (66). and that a t time t = t' = 0 a light ray starts from the (then coincident) origins of coordinates.
(612b). (612c). there is no reason to suppose that the relative motion in the zdirection should affect the y.A2 = 1. (69) into Eq. and A. If we now insert Eqs.y2v2 = c2. and p = v/c. b. Elimination of A between Eq. we get In finding r. we must have y2 . (612b) yields If we now substitute this result in Eq. This procedure yields When these relations are inserted into Eqs. we choose the signs of the square roota so that they are consistent with Eq. The fourth relation is chosen because it is linear in z and t. (611) On setting A = ac.651 THE VARuTIox OF MASs WITH VELOCITY 115 because. we get 7' .a2c2 = 1. we get the Lorentz .and zcoordinates. (68a). (612a) and Eq. ah2 + vy2 = 0. and lead to the necessary results z = z ' and t = t' when v = 0. (6fib) we get Since this equation is to be identical with Eq.  c2b2 . (69). (612c) gives and comparison with Eq.
(620). In Newtonian kinematics. if u : and v both approach c. we have. in the first of Eqs. with the result y = y'. and us. Because of the denominator in Eqs.116 THE SPECIAL THEORY OF RELATIVITY [CHAP. also . is replaced by v. I f we set with similar definitions for 4. u. the composition of velocities is more complicated. (620).ui.. The latbr can be solved for z and 1. 6 transformation equations (67). Equations (618) are exactly similar in form to Eqs. in relativistic kinematics. The transformation equations for the components of the velocity of a moving point can be found by taking the derivatives of the Lorentz transformation equations with respect to t and t'. relative velocities are found by simple addition or subtraction. Thus. z = z'. the sum of two velocities which are separately leas than the velocity of light c can never exceed c. (67) except that v. the velocity of St relative to S. The inverse transformation equations are again obtained by replacing v by v: The transformation equations for the components of the velocity lead to an important addition theorem for velocities. u. the velocity of S relative to St.
relative to 8. approaches ul v. the denominator approaches unity and u. considered relativistically. with S' moving at a constant velocity v. Elastic collision of two identical balls. Newtonian mechanics in certain fundamental ways. at first to be contrary to our physical intuition. and that B throws a ball with velocity U along the yfaxis. The derivation which follows is due to Borr~. they yield correct results when applied to such problems as the Doppler effect of a light wave and the oberration of starlight. but for values of u : and v small compared with c. I t is not surprising. + Before collision At instant of collision After collision FIG.651 THE VARIATION OF MASS WITH VELOCITY 117 approaches c.ivity theory is the deduction that the mass of a body varies with its velocity.along the xaxis. Imagine an observer A located on the yaxis of a reference system S. One of the most important results of the relat. The relationship between mass and velocity can be derived by considering a simple conceptual experiment of a type first discussed by Tolman. Although the relativistic equations seem.'~' and is highly instructive because it shows how some of the basic ideas of the relativity theory are used. . and another observer B on the y'axis of a second system Sf. in view of the interrelationships between space and time coordinates as expressed in the Lorentz transformation equations. Suppose that A throws a ball with velocity U along the yaxis. the classical expression for the addition of velocities.f 3 4 . that relativistic mechanics differs from classical. The ball thrown by A has.
We. be the corresponding velocities as measured by B. The components of the velocity of the first ball. UZ: = . parallel to the y.(v2/c2).o the z or z' direction. respectively. are The components of the velocity of the second ball relative to the S system can be obtained by using the first and second of the velocitytranaformation equations (620). denote the masses of the two balls. and relative to the S system. in fact. as shown in Fig. I perfectly smooth and to undergo an elastic collision. that is.: = U and u : = u. Suppose. I t will not be assumed. let ul and u2 be the velocities of the balls before the collision. that the two balls appear to that observer to have the same maas. and taking u i = u. and let u. 6 when at rest in the S system. has the components f the balls are assumed to be We now consider the effect of the collision. as in classical mechanics. the component parallel t. before the collision. by ml and m2. and the subscript t for the transverse component.or y'direction. as seen by the observer in the 8' system. (621~) We calculate next the total momentum of the two balls relative to the S observer.118 THE SPECIAL THEORY OF RELATIVITY [CHAP. the momentum is defined as the product of mass and velocity. they cannot exert tangential forces on each other. = 0. further. as measured by A. 64. and that the momentum vector has the same direction as the velocity vector. that the balls are thrown so that they collide. it will. as seen by the observer in the S system are then The components of the velocity of the second ball. and u. We shall assume that. therefore. the zcomponents of their velocities are . Let the subscript 1 denote the ball thrown by A and the subscript 2 the ball thrown by B. the same mass as the ball thrown by B when it is at rest in the S' system. before the collision. as calculated by A (the S observer). however.U d 1 . before the collision. We use the subscript 1 to denote the longitudinal component of a velocity. with the line of centers a t the instant of impact in the direction of the yaxis. Then the total momentum before the collision. The result is ~ 2= 1 V. be seen that their masses must appear different.
(v2/c2). by hypothesis. The components of the velocity of the second ball as calculated by A are. ii2t = odl . and the collision must take place in a symmetrical way. the first equation would be identically correct. (624a) PI = PilFill + 8 1 = iEIDlt $ ?ii2Zzt= %lg + ? i i 2 U d l . U and v are different from zero and positive.(v2/c2). If conservation of momentum is required. The first ball must take on a transverse velocity . (623~) The components of the total momentum after the collision as calculated by A are then. (620). this requirement is fundamental in relativistic mechanics. ZZzl = V. and comparison of Eqs. We shall not assume that the mass and velocity of either ball must be the same after the collision as before but shall see what the theory tells us. we must give up the idea that the mass of either ball is constant and replace it by the assumption that the mass is variable and depends on its velocity (since there is no . ?ii2Tiiipl = ?ii2v. in analogy with classical mechanics. but the second would lead to a contradiction because it would require that which is impossible because. from Eqs. The observer A must see his ball undergo the same change in motion that the observer B sees his ball undergo. we have u. while the second ball must take on a velocity equal in magnitude but opposite in direction to that of the first ball. (624b) e conserved in We now require that the total momentum o f the two balls b the collision.I f opposite in direction to its velocity before the collision. If we label quantities after the collision with bars. (622) and 624) yields I f the mass were constant. ml = m2 = El = ?ii2. and must be positive because of the symmetry of the collision. The longitudinal and transverse components of the momentum must be conserved separately.651 THE VARIATION OF MASS WITH VELOCITY 119 not changed by the collision.
The magnitude of a velocity u is given by The speeds of the two balls relative to the S system are then. we assume that the mass of a body as measured in a chosen reference system depends on the magnitude of the body's velocity relative to that system. then.(v2/c2) = 0.(vZ/c2) (627a) According to Eq.m 2 4 l .if the mass varies only with the velocity.120 THE SPECIAL THEORY OF RELATIVITY [CHAP. (626). the speed of the first ball relative to observer A is not changed by the collision. 6 other parameter in the problem on which it can depend). Equations (626s) and (626c) then show that u1 = Til. a2 = d v 2 + U2[i .(v2/c2). or ml . the mass m2 appears larger than ml to an observer stationary in S by the factor 1 / 4 1 . u1 = U. according to Eqs. El = (626c) (6264 Now Eq. This condition requires that from which it follows that U = If. If we now imagine that the velocity U with which the balls are thrown is made smaller.( v ~ / c ~ ) ] . appear to have different masses when one reference system moves with constant velocity v relative to the other. be written. the two balls. then m2 and X 2 can be equal only if the corresponding velocities u2 and E2 are equal. ml = ?5il. Equation (625) can. (626a) (626b) after the collision. therefore. that is. ul = 0 and u2 = v in the limit U = 0. which were assumed to have the same mass when a t rest in their respective reference systems. When the system S' moves with constant velocity v relative to S. (627a). (625a) requires that mz = FEZ. and the mass of the first ball relative to A is also not changed by the collision. before the collision. More precisely. hence. The . I t follows that m2 = m1 4 1 . If.
or greater than the velocity of light. it holds for any moving body. according to the relativity theory. prays from radioactive substances have been used to test the relativistic formula for the mass. as a function of the speed of the beta particles. elm. while ma is the mass which corresponds to velocity v. According to the relativity theory. One of the methods used to test the formula depends on the measurement of the value of the specific charge. Equation (62713) had been obtained earlier for the mass of an electron. The derivation has shown that when a body moves with constant velocity v with respect to a stationary observer.651 THE VARIATION OF MASS WITH VELOCITY 121  velocity U is made to approach zero only to simplify the derivation and is not a necessary assumption (see Problem 8). The variation in mass becomes important when the speed of the body begins to approach that of light. and . Consequently. and is denoted by m. finally. For this reason. the value of elm should depend on the velocity only because of the dependence of the mass on the velocity. For bodies moving at speeds up to about one percent that of light. the variation in mass is negligible. We have. Equation (627b) shows that no material body can have a velocity equal to. The mass ml. the electric charge is a quantity which is independent of velocity and has the same value for all observers. but in the case of more swiftly moving particles the correction is appreciable. which corresponds to zero velocity is called the rest mass and is denoted by m ~ . it has a mass which depends on  the velocity and is larger than the rest maas mo characteristic of the body. as is evident from the following values.
(v2/c2) should be a constant. R. The condition for escape is eE = evH. and the mass ratio m/mo is also obtained."' the most famous of which is probably that of Bucherer in 1909. Schematic view o electrons with different speeds. as indicated in Fig. Cohen. 6 where elm0 is the value corresponding to the rest m a s s of the electron. a small radioactive source S of 8rays was placed at the center between two very closely spaced metal discs forming an electrical condenser C. 65(a). and the velocity of an emerging &particle can be determined. the measured value of elm divided by dl . In this experiment. I f elm and v are measured. the &particle is deflected in the magnetic field alone.122 THE SPECIAL THEORY OF RELATIVITY [CHAP. 1957. New York: Interscience. and the result of many particles over a wide range of f Bucherer's apparatus for measuring e/m for FIG. K. 65(b). Fundamental Constunts of Phyeics. 65. (Reprinted by permission from E. or v = E/H. In the region of the electric field between the discs the electric and magnetic deflecting forces must balance each other exactly for a particle to escape. M. The box was placed in a homogeneous magnetic field H parallel to the plane of the discs. W. and only particles with a single speed v emerge at a given angle 0. After emergence from between the plates. Several experiments of this type have been made. as shown schematically in Fig. which is e/mo.) . M. The condenser was placed inside a cylindrical box which could be evacuated. Crowe. (a) elevation view (b) plan view (c) traces on photographic film. and J. and a photographic film P was wrapped around the inner surface of the box. Dumond. The speed of an emerging bparticle is a function of the azimuthal angle 0 at which it is emitted.
3 0. I I 1 I 1 1. then The field strengths E and H are known and r. e/mo is in units of lo7 emu/gm. can be obtained from an analysis of the photographic trace and geometrical relationships among the various parts of the apparatus.6 0.651 THE VARIATION OF MASS WITH VELOCITY 123  speeds and emerging a t different angles 0 is a continuous curve on the photographic film.5 0. A typical trace corresponding to half the circumference of the box is shown by one of the curves in Fig. The specific charge and mass of the electron as functions of velocity.4 0.O 0. .9 Velocity (v/c) . Verification of the relativistic mass formula.66. the radius of the circular path of a 8particle of speed v.7 0. no. the second curve on the film is formed when the electric and magnetic fields are reversed simultaneously. 65(c) .8 0. The value of e/m corresponding to a velocity v is obtained from the relation if we insert for v its value v = E/H.
and can be found in the usual way by calculating the work done in setting the particle into motion: On insertion of Eq. and the proportionality factor is the square of the velocity in m . There is no doubt that the theoretical predictions are confirmed for swiftly moving &particles. (62713) we get Substitution for mu dv from Eq. (634) Thus. 66. W The relationship between mass and energy. a change in kinetic energy dT is directly proportional to a change a s s dm. force is defined as the time rate of change of momentum. Differentiating both sides of Eq. as in classical mechanics. The solid curves represent the theoretical expressions for m and e/mo as functions of velocity. 6 Later experiments have used improved methods's*6' and some of the results obtained by Bucherer and others are shown in Fig. In classical physics. . (632) yields dT = c2 dm. The variation of mass with velocity leads to modifications of our ideas about energy. (633) in Eq. and the points represent the experimental results. but now m The kinetic energy T of a particle is defined in terms of force and distance.124 THE SPECIAL THEORY OF RELATIVITY [CHAP. The mass formula has also been confirmed for other particles. and dv d F=(mv)=mdt' dt since mass is constant. (630) for F. This definition of force is kept in relativistic a s s is also a variable and mechanics. there follows = mu dv + v2 dm.
The total energy E of the body is then the sum of the kinetic energy and the rest energy. there is an amount of energy mc2. This relationship may be interpreted as meaning that the rest mass mo is associated with an amount of energy moc2. which may be called the rest energy of the body. there corresponds a mass given by Another important consequence of the proportionality between mass and energy has to do with the momentum associated with the transfer of energy. to the energy and momentum carried by a light quantum (photon). then we can write for the magnitude of the associated momentum p This relationship can be applied. to an energy E .661 THE RELATIONSHIP BETWEEN MASS AND ENERGY 125 of light. the kinetic energy of a body can be expressed in terms of the increase in the mass of the body over the rest mass. conversely. I f we substitute for dm from Eq. or E =T + moc2 Hence. The energy carried by a photon is hv and its velocity is c. (632). The momentum is. associated with a mass m. for example. (633) in Eq. (638)) . we get Integration of both sides of the last equation gives for the kinetic energy According to this result. The integral form of this result can also be obtained easily. If a quantity of energy is transferred with a velocity v. by Eq.
and this mass is hv/c2. In atomic and nuclear calculations. otherwise the mass of a photon traveling with the velocity of light would be infinite. In an ordinary chemical process. It is therefore the practice to express energies in units of eledron voh. a decrease which cannot be detected by even the most sensitive chemical balances. and One gram of mass is associated with nearly lo2' ergs of energy. 6 The rest mass of a photon must be zero.126 THE SPECIAL THEORY OF RELATIVITY [CHAP.5 X 10" Btu. (627) becomes zero for v = c.602 X lo" erg. The interconversion of mass and energy can illustrated by several numerical examples which will be useful in later calculations.184 X 10' ergs) = 4. with the result that 10' calories of heat are liberated.65 X lo' gm. the erg is an inconvenient unit of energy because the amounts of energy involved in single events usually are only small fractions of an erg. mass is always converted into energy. the energy is in ergs. it is equivalent to 1. I f the mans is in grams and the velocity is in centimeters per second. the photon has a m a s associated with its kinetic energy. the massenergy balance is (1 cal = 4. (637). For example.65 x 10lo gm/gm of substance. It will be seen in later chapters that the msssenergy conversion can be detected in nuclear phenomena because the amounts of energy involved per atom are very much greater than those involved in chemical reactions. corresponding to this mass is the momentum hv/c. because the denominator of Eq. Although the rest mass of a photon is zero. For con . since 1 Btu = 1. It is useful to obtaih now some of the conversion factors which will be needed in later work. This is a very large amount of energy indeed. The electron volt is the energy acquired by any charged particle carrying a unit electronic charge when it falls through a potential of one volt. This is the reason why the maasenergy effect has not yet been observed in chemical reactions. if 100 gm of a chemical substance take part in a reaction. or about 8.055 X 10" ergs. abbreviated as ev. but the practical conversion of mass to energy is a difficult problem. The loss of mass equivalent to the energy involved in the reaction is 4. according to Eq. say 10l2 to erg.
equal to lo3 electron volts. Finally. Compton in 1923 during the course of his studies of the scattering of xrays by matter.602 X low6erg.61 X (64 1 4 E (kev) = m(gm) X 5. some of the ideas used by Compton .u2 = 3X 4 X 1. ' by Eq. an aparticle may have a velocity of 2 x 10' cm/sec. It'is often convenient to use the energy corresponding to one atomic mass unit.141. the kinetic energy of this aparticle is *m.108 X 1 0 ~ ~ energy is 0.511 Mev.602 X is equal to 1.61 X lo2'. equal to lo6electron volts. Some of the ideas treated in Chapter 5 and in the present chapter can be illustrated by discussing an effect discovered by A. two other units are also used. H. the rest mass is 1. The relativistic mass correction for this velocity is sufficiently small so that it may be safely neglected.141 Mev. ( 6 41b) gm. For example. is represented by kev and the other. The massenergy relation becomes = m(gm) X 5. One kev is therefore equal to 1. the rest The rest mass of an electron is 9. The theoretical treatment of the Compton effect involves Einstein's quantum theory of light as well as some of the ideas of the relativity theory. Hence.39 X .or 931. The rest mass m. The effect is also of practical importance in the absorption of xrays by matter and will be treated in further detail from that viewpoint in Chapter 15. of an aparticle is very closely four times that of a hydrogen atom.674 X gm and the corresponding energy is 939 Mev.671 THE COMPTON EFFECT 127 : venience. (641c).674 X erg X (2 X 10')~erg = 13.66 X gm. This mass is 1. (642)  The values of the kinetic energy which will be used most often range from electron volts to millions of electron volts. or about onefifteenth of the velocity of light. is abbreviated erg. E (Mev) = m(atomic mass units) X 931. and one Mev by Mev. one.67 The Compton effect. For a hydrogen atom.
being electromagnetic radiations like light. its effect is to change the direction of the incident radiation.128 THE SPECIAL THEORY OF RELATIVITY [CHAP. . they carry momentum hv/c in accordance with the ideas developed in the last section. When the angle between the incident and scattered radiation was 90°. Compton (1923) was able to show that when a beam of monochromatic xrays was scattered by a light element such as carbon. 67. the electron which scatters the photon must recoil with a momentum equal to the vector difference between that of the incident and that of the scattered photon. The wavelengths of the two lines were measured with a Bragg xray spectrometer. the second of slightly longer wavelength. the difference in wavelength was found to be 0. or longer wavelengththan that of the primary photon. The requirement that energy be conserved gives . it was assumed that the wavelength of the scattered radiation was the same as that of the incident radiation. Along with their energy. 6 in his explanation of the effect play an important part in the modem quantum theory of atomic structure and will be needed in Chapter 7. The require . the scattered radiation consisted of two components. independent of the wavelength of the primary beam and of the nature of the scattering material. the primary xrays. With improved techniques. This kind of scattering is called "Thomson" scattering. To account for the presence of the shifted component and for the amount of the shift. The energy of this recoiling electron is taken from that of the primary photon. In the treatment of the scattering of xrays by electrons in Chapter 4. The difference in wavelength AX between the two scattered radiations was found to vary with the scattering angle and to increase rapidly a t large angles of scattering. v that of the ray scattered by the electron. and that in this collision energy and momentum are conserved. Compton assumed that the scattering process could be treated as an elastic collision between a photon and a free electron. the recoil velocity of the electron is v.    where vo is the frequency of the incident xray. as in Fig. leaving a scattered photon which has less energyand hence a lower frequency. In order that momentum be conserved. The scattered quantum or photon moves in a different direction from that of the primary photon and carries a different momentum. are propagated as quanta with energy hv. it is convenient to discuss the Compton effect now. For all of these reasons. and the relativistic kinetic energy of the electron is used because the velocity of the electron may be great enough for relativistic effects t o be significant. According t o Einstein's quantum theory of the photoelectric effect.0236 X cm. one of the same wavelength as that of the incident beam.
We therefore introduce the wavelengths A. This gives For a particular angle of scattering 4. the last three equations contain three unknowns.671 THE COMPTON EFFECT 129 Scattered photon Momentum = hv/c Incident photon ! Momentum = hvo/c * / \ (644) Recoiling electron Momentum = mu/ FIG. For comparison with experiment.(v2/c2) mOV ycomponent: hv 0 = . and expressions for these quantities can be found by a straightforward solution of the equations. it is convenient to set 8 = v/c. one for the xcomponent of the momentum.(v2/c2) cos 0.sin 4 C sin e.  The Compton effect. and one for the ycomponent hv hvO cos 4 xcomponent: C C + 4 1 . v. 4 1 . and rewrite the last three equations. however. (645) To solve these equations. and 8.67. 8. ment that momentum be conserved gives two equations. we are more interested in the shift in wavelength. = Yo C and A =V C .
A X = 0. Eq. 6 of the incident and scattered radiations. the wavelength A of the scattered xray should be greater than that of the incident xray by the amount 0. Then Eqs.cos 9) A.0242(1 . ( 6 4 7 ) can be written On squaring.cos 6) A. The predicted dependence X on the angle g was also verified by experiment. For 9 = 90°.9107 X Therefore.624 X X 10" are inserted. we get 2h2 (cos g XoA  1) + 2mo& m = 0.0242 X 1 0 . (648) and (649) become : h i sin g = di=F sin 8. respectively. (653) Equation (653) states than when an incident xray of wavelength by a free electron. (651). of A A . For a given value of the scattering angle 6. then A X =X and c = 2.A.0242(1 . Squaring these two equations and adding.cos 4) cm = 0. we get Subtracting Eq.130 THE SPECIAL THEORY OF R E I A T I v r m [CHAP.0236 A.~ ( 1 .998 . we have ~imilarly. which agrees very well with the observed value of 0. is scattered through an angle + . the shift in wavelength is independent of the wavelength of the incident radiation. = 0.0242 A. (650) from Eq. If the values h = 6.
New York: Simon and Schuster. New York: Crown Publishers. and it is found that T = hvo 1 + a(1 . the electron acts like a free electron. and this accounts for the presence of the unshifted spectral line. It will be recalled that it was assumed that the electrons are free. Relativity: The Special and General Thewy. 1956. 1961. Princeton: Princeton Universits Press. Readable Relatimtu. Hence. does not have its wavelength changed. C. the rest mass mo of the electron in Eq. REFERENCES GENERAL A.cos 4)a . 15th ed. The more refined details of the Compton effect. It will be noted that in the Compton formula. a s d this is actually the case so far as the shift in wavelength is concerned. (653) must be replaced by the mass of the atom. Actually. but remains bound. A. but does not account for the presence of the unshifted line. London: Methuen. demand a rigorous relativistic treatment. Part 111. A. The theory just developed explains the observed shift in wavelength. however. INFELD. 5th ed. The Evolution of Physics. electrons are bound to atoms more or less tightly. London: G. EINSTEIN. The Meaning of Relativity. The recoil electrons were sought and found. (652)) the only trace of the relativity theory that appears is the zero subscript denoting the rest mass of the electron. If the collision is such that the electron is not detached from the atom. ). It would seem as though the same effect would have been obtained if the problem had been set up without taking into account the relativistic effects. and a certain amount of energy is needed to shake an electron loose from an atom. 1938.' . 1926: New York: Harper and Brothers ( ~ a r ~ e r " ~ oBooks/The rch Science ~ i b r a r ~ 1960. which is several thousand times greater. therefore. V. (653) is valid. and their observed energies agreed with the values predicted by the theory. even the more elementary problems are usually set up relativistically. and Eq. EINSTEIN and L. DURELL. 0354   where a = hv0/mOc2. Eq.REFERENCES 131 The kinetic energy T of the recoil electron can be calculated without much difficulty. A photon which collides with a bound electron. Bell and Sons. The calculated value of Ah then becomes much too small to be detected. If the amount of energy given to the electron by the photon is much larger than the work needed to detach it from the atom. 1954. EINSTEIN.cos +) (1 .
4th ed. H.. PAULI. C. New York: PrenticeP. COMPTON and S. Oxford University Press. Nuovo c i d . New 5. Cambridge: University Press. 1920. 6. Chapter 14. 342 (1958)." Phys.") 2. London: Pergamon Press. A. BLAND. Cit. H. M. DINOLE. and ROGERS. ROGERS. Relativity. W. ref.132 THE SPECIAL THEORY OF RELATIVITY [CHAP. COHEN. H. pp. New York: Dover Publishing Co. At what intervals will the signals appear to an observer a t rest in the system Sf moving along the zaxis with uniform velocity u with respect to the S system? Suppose next that there is a source of homogeneous xrays at rest in the Sf system. Classical Electricity and Magnetism. Theory o f Reldivity.Reldivity Physics. London: Methuen and Co. Rev. gen. Chapter 3. New York: Harper and Brothers (Harper Torch Books/The Science Library). 130142. Minkowski. ALLISON. 1411 (1957). Einstein. 1922. PHILLIPS. Chapter 7. What is the relationship between A' and A? . No. XRays in Theory and Ezperiment. K. Reading. Fundamental Constants of Physics. JANOSSY. PANOFSKY and M. Space. London: Methuen and Co. Letlers.: AddiaonWesley. H. 6 A. PARTICULAR 1. PANOFBEY and M.Themy o f Relativity.. 1955. Maas. by H. 1934.. and DUMOND. FARAQO and L.The Special Themy of Relativity. 1957. 1940. Weyl. New York: Wiley. "Review of the Experimental Evidence for the Law of Variation of the Electron Mass with Velocity". The Principle of Relativity. Dritte Auflage. 1958. 2nd ed. The Relativistic Mass of the Electron.. Vol. A. Introduction to the Themy o f Relativity. MCCREA. New York: Wiley. BORN. 3. 1954. 15. W. Reu. C. Oxford University Press. MOLLER. A Coktion of Memoirs on The Special and General Theory o f Rekztivity." Phys. 6. op. Berlin: Springer. "New Experimental Test of Special Relativity. CEDARHOLM. Section 7. W. R. London: Methuen and Co. CROWE. and TOWNES. PHILLIPB. "A Determination of the Masses and Velocities of Three Radium B &particles. P. 1942. S. and H. G. A. 1959. TOLMAN. K. and that the wavelength of the xrays as measured by the S ' and S observers is A' and X. 57. HAVENS. BEROMANN. EDDINGTON. Time and Grayitation. H. New York: Van Nostrand. Suppose that a clock a t rest in the reference system S emits signah a t time intervals At = t2 . and Comlogy. MCREYNOLDS. Lore&. 1. Hall. 1923. 379 (1940). K. York: Interscience.t l . respectively. Chapters 14. 1. 4. W. Thernwdymmh.Die Re2ativiliitsthmk Einsterterns und ihre Physikalische Grundlagen. 16. 1935. H. V. "The experimental basis for the theory of special relativity. 1952.
0. and u. synchrocyclotrons have been built which accelerate protons to kinetic energies of about 400 Mev. . 7.0. move off with equal speeds of magnitude 0. What is the velocity of the second electron relative to the first. Derive the following useful relationships concerning the relativistic properties of moving particles (the symbols are defined in Section 66). 9. that the kinetic energy can be written + + Find the values of v/c for which the second term of the expression is I%.PROBLEMS 133 2.5 Bev protons.9c. Show that if v < c. Show. Calculate the kinetic energy.0. Calculate the ratio of the mass of a particle to its rest mass when the particle moves with speeds that are the following fractions of the speed of light: 0. and 50%. and the Cosmotron a t the Brookhaven National Laboratory has produced 2. (627a).1. Consider a body with rest mass M in a given frame of reference. Suppose that two electrons. 5.2. of the particle of Problem 5 a t each speed. in ergs and in Mev.. the second parallel to the 2direction.  10.999. 0. Derive an expression for the quantity d2 = ui2 G2 ui2 in terms of u. 3. as measured in the S'system? 4.xdirection.995. one moving parallel to the 42direction. 0. What is their speed relative to each other in the S system? Suppose that the S' system moves at a speed v = 0 . In connection with the derivation of the formula for the variation of mass with velocity. Suppose that the body breaks up spontaneously into its two parts with velocities v l and v2. 9 ~in the . What is the mass of the protons relative to the rest mass in each of the three cases cited? What is the velocity relative to that of light? What is the total energy? 8. the body is composed of two parts with rest masses MI and M2. 6.99. of the first term. 0. then that it is a proton.9999.v2/c2)1/2 in powers of v/c. A constant frequency cyclotron can accelerate protons to kinetic energies of about 20 Mev.0.. u.9. 0.5. 0.7. and u = c. by expanding the expression (1 . show that Eq. respectively.. so that it keeps up with the second electron. sssume first that the particle is an electron. ejected by a heated filament stationary in S. then u' is also equal to c. for U # 0. respectively.95. Charged particles can be accelerated to high energies by appropriate machines (Chapter 21). is valid generally. lo%.
The scattered rays are observed a t an angle of 54" with the direction of the incident beam. and velocity of the recoil electron. (e) the angle a t which the electron recoils.0900 A is incident on a carbon scatterer. (b) Show that the total energies of the two parts are given by m 2 11. . 13.134 THE SPECIAL THEORY OF RELATIVITY [CWP. What are the greatest wavelength and energy a photon can have and be able to transfer half its energy to the recoil electron during a Compton collision? What are the direction and magnitude of the velocity of the recoil electron in this csee? How important is the relativistic variation of mass with velocity in this problem? 14. Find (a) the wavelength of the scattered rays. A particle of rest mass m l and velocity v collides with a particle of mass a t rest. (b) the energies of the incident and scattered photons. the energy of the (b) For very large energy of the incident photon. (c) the momenta of the incident and scattered photons. is the wavelength of radiation having a quantum energy equivalent to the rest energy of an electron. Show that the mass M and velocity V of the composite particle are given by 12. a > backscattered photon approaches 3 m2. (d) the energy. h/m@. Show that the following relations hold in a Compton scattering process. momentum. (a) The maximum energy transfer to the electron is given by and > 1. 6 (a) Write an equation for the conservation of ldal energy. after which the two particles coalesce. (c) The angles 8 and 4 which the directions of the recoil electron and scattered photon make with respect to that of the incident neutron. obey the relation (d) Show also that the C a p t o n rwoelenqtn. A beam of homogeneous xrays with a wavelength of 0.
respectively. The wave number is the number of waves in one centimeter of light path in vacuum and is expressed in reciprocal centimeters. In the visible and near ultraviolet regions. for the Ha. 'the lines coming closer together as the wavelength decreases until the limit of the series is reached a t 3646 A. This problem was discussed briefly in the fifth chapter. H. now called the Balmer series. The lines may be in the ultraviolet or infrared as well as in the visible range of wavelengths. parallel lines. . or mission. . em'. Rydberg. In 1885. . is dispersed by a prism or grating spectrometer.4 ' where B is a constant equal to 3646 A. lines. only a few colors appear. The wavelengths of the lines are characteristic of the element emitting the light. i t consists of a series of lines whose relative positions are shown in Fig. found that the lines of the visible and near ultraviolet hydrogen spectrum. in the form of bright. and n is a variable integer which takes on the values 3. which is a property of the atoms of that element. it will now be treated in greater detail. 71. A line. a few years later. There is an apparent regularity in the spectrum. Instead of a continuous band of colors. or from a f h e into which a volatile salt has been put. but for some of the heavier atoms the spectra may consist of hundreds of lines.CHAPTER 7 ATOMIC SPECTRA AND ATOMIC STRUCTURE 71 Atomic spectra.. can be described in a more useful way in terms of the wave number. For example. The lighter atoms such as hydrogen and helium yield fairly simple spectra with a relatively small number of lines. or reciprocal of the wavelength. line is 6562. the wavelength of the H. as in a rainbow or as in the spectrum from an incandescent solid. and each element has its own particular line spectrum. For this reason. The first problem that must be solved by a theory of atomic structure is that of the line spectra of the elements. Balmer showed that the wavelengths of the nine lines then known in the spectrum of hydrogen could be expressed by the very simple formula n2 X=B(71) n2 . 4. line spectra are also called atomic spectra. .8 X lo' 135 . 5. . Each line is an image of the spectrometer slit. The spectrum of atomic hydrogen is especially interesting for historicd and theoretical reasons. apedrum is produced when light from a gas through which an electric discharge is passing. . . Hg. isolated. deviated through an angle which depends on the frequency of the light forming the image.
H x a H .26 3735. cm.63 3750. .85 3751. 5.01 3771.47 4101.A = 6563.33 4340.7 Red Blue Violet Ultraviolet FIQ. the wave numbers and thus the wavelengths of the lines of the Balmer series are given with great accuracy.74 3751. and the wave number v' = 1 / A = 15237.93 3971.43 3723.5 41Q1.66 4862.06 & 6 H.69 4102. the wave numbers of the Balmer lines are given by where Ra is a constant with the dimension cm'.70 4862.19 3836.74 4341.19 3890.46 3723.3 4340.98 Meaeured wavelength  corrected to vacuum.95 3711.48 3798.03 Wavelength calculated from the Rydberg formula 6564.68 4341. A: A 6564.31 3735.25 3734. as can be seen COMPARISON OF OBSERVED AND CALCULATED WAVELENGTHS OF THE LINESOF THE BALMER SERIEIJ IN THE HYDROGEN SPE~UM Line n 3 4 5 6 7 8 9 10 11 12 13 14 Ha HP HT H6 H. and n takes on the values 3.98 3771.89 3971.40 3797. .71. H. 4.4 cm'.07 3889. The value of the Rydberg constant for hydrogen has been determined empirically from spectroscopic measurements and is with this value of RH. A: A 6562.1 4861. Wavelength measured in air.00 3713. .16 3836.85 4861. According to Rydberg.03 3713. 15 .51 3799.06 3835. Diagram of the lines o f the Balmer series of atomic hydrogen. . called the Rydberg constant jm hydrogen.23 3890.37 3721.73 4102.74 3970.91 3770.
are now known. r series. the index of refracvacuum by means of the relation A tion of air.. infrared) (Pfund series.4 . P = RH(h$)j 5 = n=2.7. where p. a series of spectral lines can always be represented by the general formula In this equation.. The results illustrate the accuracy with which spectroscopic measurements can be made as well as the remarkable agreement between the values of the wavelengths measured experimentally and those given by the simple Rydberg formula with the empirically determined value of the Rydberg constant for hydrogen. 5 . and found that although the spectra of the heavier elements are more complex than that of hydrogen. Thus. named after their discoverers. 5.. tabulated function of wavelength.. one fixed and one varying. R is a constant for the element.. a and b are characteristic n is an integer which is fixed for a given constants for a particular series. is a known. infrared) (Brackett series. R ) n = 3 .. n=4. The wavelength measured in air has been corrected to . and three in the infrared.i. The wavelength in vacuum is then compared with the value calculated from the Rydberg formula. visible and near ultraviolet) (Paschen series.5.. 4 .. in order to be acceptable.. In addition to the Balmer series. infrared) R~(&$). They also show that a theory of atomic structure..6. the following spectral series..3. for hydrogen.. n = 5..711 ATOMIC SPECTRA 137 from Table 71.8.7. (Lyman series. He looked for similar regularities in the spectra of elements other than hydrogen. Rydberg noted that for each line in the Balmer series (the only series known at the time) the wave number could be expressed as the difference between two terms. 5 = 5 = RH(&$)~ RH(&$)I . = fiX. the hydrogen spectrum contains four other series.6. and n is a varying integer whose values give the different lines of the . must predict the values of the wavelengths and the Rydberg constant with great accuracy. one of which is in the ultraviolet region... n = 6. ultraviolet) (Balmer series.
and s. the following four series are found. n = 3 . one of the chief problems of spectroscopy became the representation of the lines of a spectrum as differences between terms. are certain constants which are characteristic of the alkali atoms.. The latter are obtained when light with a continuous  . (75) For hydrogen the terms are given by For other elements. principal series: sharp series : diiuse series: Bergmann series: : P B = R R ( 1 . In each case. = + PP R R R (n + $I2 R R R 1 = ( 2 + ~ ) ~ (. the terms usually have a more complicated form. .+dl2  I I= (3 (n + +jI2 ' In these series.138 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP. . The lines of the spectra of other elements also can be represented as differences between terms. The emission spectrum of an alkali metal is an example of a spectrum of this type. the lines of the sharp series appear relatively sharp in a spectrograph picture..k ~ ) ~ (n+pI2 (2 J n = 2. and each line can be expressed as the differencebetween two terms. . 5.element consists of several series of lines. .T2. The line spectrum of any . The principal series is so called because it contains the brightest and most persistent lines. . 7 series.. d. For example... the elements also have characteristic absorption spectra.4. 6 . . and the formula for the wave number can be written generally as P = T I. which states that. 4 . . the difference between any two terms gives the wave number of a spectral line of the atom.. with certain limitations. 3 . p. The clarification of the meaning of the terms then became a major theoretical problem. n = 2 .. 4. . as n becomes very large the lines of the series converge to a linhit given by the constant term. while the lines of the diffuse series are relatively broad. j. R is the Rydberg constant for the element.3. Ritz (1908) extended Rydberg's formulation of wave numbers by means of his combination principle. the difference between Tgand T8 for hydrogen gives the fifth line of the Paschen series. n = 4 . 5 . with as few terms as possible. As a result of the discovery of this combination principle.. . In addition to emission spectra. and the first and second members of the formula are obtained from different series.
whether they are singlets or multiplets.721 THE BOHR THEORY OF ATOMIC SPECTRA 139   . The radiation emitted or absorbed in . spectrum (as that from a filament lamp) is passed through an absorbing layer of the element and then analyzed with a spectrograph. e. such as the changes in the spectrum of an element when the atoms radiating the light are in a magnetic or electric field. In such cases. Bohr applied the quantum theory of radiation as developed by Planck and Einstein to the Rutherford nuclear atom. and the spectra o f the elements have been analyzed in great detail. When a light source is brought into a magnetic field. triplets. 1. 72 The Bohr theory of atomic spectra and atomic structure.. An atomic system possesses a number of states in which no emission of radiation takes place. A splitting of the spectral lines is also observed if the radiating atoms are in an electric field. every emission line can also occur as an absorption line. in the spectra of the alkali metals. or groups o f even more components. These states are called the s t a t i m r y states of the system. so that six formulas instead of three are required for a complete representation of the series in the visible region. in principle. the simple way in which spectral series can be expressed points to the existence within atoms of a relatively simple and universsl method by which the characteristic atomic radiations are emitted. In a photograph o f an absorption spectrum. There are also other effects. His theory is based on the following postulates. Bohr developed a theory o f the constitution of atoms which accounted for many of the properties of atomic spectra and laid the foundations for later research on theoretical and experimental atomic physics. The extent of the splitting depends on the strength of the field and on the nature of the spectral lines. The field of atomic spectra is one of the great branches o f physics. although such an emission is to be expected according to ordinary electrodynamics. for example. The absorption lines coincide exactly in wavelength with corresponding emission lines and.g. each emitted spectral line is split into a number of components (the Zeeman effect). doublets occur. In many elements. the multipbt structure of spectral lines has not been mentioned. 2. each line in the spectrum is single. but more commonly the lines form doublets. But in spite of the vast amount o f information which has been collected about atomic spectra. light lines (absorption lines) appear *on a dark continuous background. Only some o f the simplest features of atomic spectra have been presented here. a separate Rydberg formula must be written for each component line. In 8 series of epochmaking papers written between 1913 and 1915. even if the particles are in motion relative to each other. Any emission or absorption of radiation will correspond to a transition between two stationary states.
3. but not during a transition from one such state to another. the fourth postulate.  . If Z = 1. did not exist so long as the electron remained in one of its allowed. as will be seen later. The angular momentum is said to be qwmlized in these orbits. the laws of classical mechanics were presumed to be valid for an atomic system in a stationary state. But. The emission of radiation was associated with a jump of the system from one stationary state (energy level) to another. Thus. Bohr's theory is a hybrid. and its frequency v is determined by the relation h ~ = Wl . . containing some classical ideas and some quanta1 ideas. the emission of radiation. which separates the allowed orbits of the electron from the forbidden ones by means of the quantum condition. The justification for the postulates was the astonishing success of the theory in accounting for many of the experimental facts about atomic spectra. (78). The dynamical equilibrium of the system in the stationary states is governed by the ordinary laws of mechanics. Starting with Bohr's postulates. in accordance with the ideas of Planck's quantum theory. Eq. Also. has been shown to be in harmony with Planck's condition for the energy of an oscillator.w 2 1 (77) where h is Planck's constant and W 1 and Wa are the energies of the system in the two stationary states. Finally. the system is the singly ionized helium atom He+. for Z = 2. The different possible stationary states of a system consisting of an electron rotating about a positive nucleus are those for which the orbits are circles determined by the relation where p is the angular momentum of the electron.140 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP. but these laws do not hold for the transition from one state to another. or stationary orbits. Bohr solved the problem of the stability of a system of moving electric charges simply by postulating that the cause of the instability. Consider an atom with nuclear charge Ze. the hybrid nature of the theory eventually led to serious difficulties. h is Planck's constant. usually called the quantum number. 7 a transition is homogeneous. where Z is the atomic number and e is the magnitude of the electronic charge in electrostatic units. for Z = 3. we can calculate the values of the energy that correspond to the stationary states of an atom consisting of a nucleus and a single electron. the system is the doubly ionized lithium atom Li++ . the atom is the neutral hydrogen atom. These postulates are a combination of some ideas taken over from classical physics together with others in direct contradiction to classical physics. and n is a positive integer. 4. .
For hydrogen. 9. then the potential energy of the electron is. If the state of zero energy of the electron is arbitrarily chosen as that state in which the electron is at rest a t a very great distance ("infinitely far") from the nucleus. r = 5. and increase in the proportions 1. given by where m is the mam of the electron. a value that is in good agreement with estimates of atomic radii obtained by other methods. v is the linear velocity of the electron. 4. and if we insert the known values of m (rest mass). and n is a positive integer. The velocity of the electron in the smallest orbit is about 2. Hence. the centripetal force on the electron is provided by the attractive electrostatic force on the electron: Equations (79) and (710) can be solved for the radius r and the velocity v.721 THE BOHR THEORY O F ATOMIC SPECTRA 141 In accordance with the fourth postulate. . for the smallest orbit in hydrogen (n = I).. so that m may be taken as the rest mass of the electron. . we get.29 X lo@ cm.2 X 10' cm/sec.. and decreases as the radius of the orbit increases. u = .Ze (712) r The kinetic energy of the electron is . from orbit to orbit. the electron revolves in a circle about the nucleus and can have the angular momentum p. Z = 1. and it is found that (71 la) and (7llb)  The radii of the allowed orbits are proportional to n2. e. 16. . by the usual electrostatic formula. the motion of the electron in an allowed orbit must satisfy the classical mechanical laws. . r is the radius of the orbit. The energy of the electron is partly kinetic and partly potential. and h. By the third postulate.
The quantum number n. (717) becomes . it will be remembered. the smaller numerically. Z = 1 and Eq. the values given by Eq. The total energy is W = T U or. 3 . may take on any integral value. (714) it is seen that the larger the value of n. 7 by Eq. . The difference in the two energies is emitted as a single quantum of radiant energy whose frequency is given by the condition . n = 1 . is that corresponding to the first orbit. (7lo). According to Bohr's second postulate. 2 . for the stationary state d e h e d by the quantum number n.142 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP. a relation typical of motion under an inversesquare law of force. is the energy of the system. From Eq.and W. (714) for the states with quantum numbers nl and n2. . This is known as the nurnuaZ or ground state of the atom. + this is the energy of the atom when the electron is in its nth stationary or quantum state. but the larger in algebraic value. a hydrogen atom radiates energy when the electron jumps from one stationary state to another state of lower energy. since it should be the most stable state and the one ordinarily occupied by the electron. as In the case of hydrogen. we get for the frequency of the l i e emitted when the electron jumps from state nl to state n2 In terms of wave numbers this becomes where we have defined the constant R.. . and T = 3U. The lowest value of W . .. If we i n s e r t for Wn.
 This equation has exactly the same form as the empirical equation (72).' 721 THE BOHR THEORY OF ATOMIC SPECTRA 143 = Since nl < nl. both particles will move about the center of mass of the two particles. Actually. Let r be the distance of the electron from the nucleus. which was found to represent the lines of the Balmer series of the hydrogen spectrum. can be calculated theoretically. The formula for the reduced mass is one which is important in many problems involving the motion of two particles.x). + . and r . has been used t. and it can be derived quite easily. The subscrivt a of the coefficient under this assumption. where M is the mass of the nucleus.the wave number is positive. mM/(M m). that it is infinitely heavy in comparison . the nucleus revolves about the center of mass in a smaller circle. Both particles must have the same angular velocity w. x is then the distance of the nucleus from the center of mass.x its distance from the center of mass. since the center of mass must always lie on the line between the two particles. if the electron revolves about the nucleus. If neither of the particles can be assumed to be at rest. and a crucial test of Bohr's theory is the agreement or d i 5 agreement between the theoretical value of the coefficient and the experimental value of the Rydberg constant. and it does have a slight motion. I t has been tacitly assumed in the derivation of the formula for the wave number that the nucleus does not move. Bohr's theory also gives the value of the constant coefficient R. The effect of the motion of the nucleus is to replace the mass m of the electron by the reduced mass. the motion of the nucleus cannot be neglected even though the nucleus is so much heavier than the electron. in terms of fundamental constants. The precision of spectmopic measurements is so great that in testing the theory. At the same time. we have Mx = m(r . it is necessary to correct for an approximation which has been made in the treatment so far. If we set nz 2. In the case of a oneelectron atom. its path in space will be a circle about the center of mass of the combined system. Before this comparison can be made. for hydrogen. Hence. Taking moments about the center of mass. R. the nucleus of the atom is known to be about 1840times as heavy as the electron rather than infinitely heavy.o denote the value with the electron.
The constant R. I presently accepted values of the constants e. For helium    . The two values agree to within 3 parts in 100.000. (721) the where M H is the mass of the nucleus. m (rest mass). A very important result is obtained when the experimental values of the Rydberg constants for hydrogen and ionized helium are compared. and is in good agreement with the theoretical value. h. was also successfully explained by the Bohr theory. 7 The total angular momentum of the system about the center of mass is M) is revolving in a The system acts as though a mass m' = mM/(m circle of radius r. which is considerably smaller than the uncertainty in the theoretical value. and this accounts for the small the factor 1/[1 variation in the Rydberg constant for different elements. (717) through (720) should be replaced therefore by the new constant + + + + f we insert into Eq. and in each series the wave numbers of the lines were found to be those predicted by t.he theory. and the reduced mass M). that is. (719)] suggested that for values of na different from 2 and 3. and m/MH. The search for these series yielded the Lyman series (1916). additional series should exist. The mass m of the electron must therefore be replaced by m'. 72 together with the transitions that give rise to typical lines of the different series in the hydrogen spectrum. we get RH (theor) = 109681 as compared with the experimental value of 109677. Some of the circular orbits of the electron in the hydrogen atom are shown on a relative scale in Fig. When the Bohr theory was proposed.144 ATOMIC SPECTRA AND ATOMIC STRUCNJRE [CHAP. only the Balrner and Paschen series for hydrogen were known. in Eqs. The spectrum of singly ionized helium.58. with an angular velocity o about the center of mass.269. This agreement represents a remarkable feat of theoretical physics and is one of the great triumphs of the Bohr theory. and the Pfund series (1924). the Brackett series (1922). The experimental value of the Rydberg constant for helium is RHe = 109722. Since M varies from atom to atom. of the electron is mM/(?n (m/M)] varies slightly. The theory [Eq. He+. m must be multiplied by the factor M/(m M). c.
73 The stationary states of an atom. (77) or Eq. and represents another triumph for the Bohr theory. (721) and (722) it can be shown that MH = m RH . The Bohr theory. Eq. in addition to accounting for the spectra of oneelectron atoms. on substituting the wave number I = v/c for the frequency. In the frequency condition. Eq. we get Comparison with the Rydberg series formula. provides a physical interpretation of the spectral terms. The . Scheme of transitions between stable Bohr orbits in the hydrogen atom (approximately to scale). 72. since the mass of the helium nucleus is very nearly equal to 4 times that of the hydrogen nucleus. shows that a term is equal to an energy state of a particular electron orbit divided by hc. (75).@na = 1841. From Eqs.RH (723)  This result is very close to that obtained in Section 24 by other methods. &ie . (715).731 THE STATIONARY STATES OF AN ATOM 145 Fro.
the terms are equal to the separation energies of the electron ) of the atom.602 X 10la erg by hc. If there is such a state.146 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP. an excitation energy of 4 the mercury atom. There is a line of wavelength 2535 A in the spectrum of mercury and the experimental result can therefore be interpreted in accordance with the Bohr theory.9 ev above the ground state. Hence.9 volts should correspond therefore to a wave 0 6 6 . an energy difference of 1 ev is equivalent to a certain wave number obtained by dividing 1 ev = 1. A transition from the ground state to an excited state can occur only when an atom receives an amount of energy equal to the excitation energy for the excited state. for example. 7 experimental determination of the spectral terms of an element. and Bohrls relationship between the excited states and the term values of the spectral lines has been completely verified.. therefore. the validity o f Bohr's assumption about stationary states has been verified by direct experiments. 9 ev was transferred to Thus. They observed that electrons could not transfer energy to the mercury atoms unless the former had a . which is. accordingly. the separation energy is called the ionization energy. Eq. This should mean. Apart from the factor he. which is a line in the ultraviolet region. Xow. Many similar results have since been obtained with other lines and gases. that the mercury atom has an excited stationary state 4. the result is 8066. An excitation energy of 4. according to the kinetic energy of a t least 4 theory.. gives the values of the energy for the stationary states of the atom. I t will be remembered that the energy scale was chosen in such a way that the electron has the energy W =o when it is completely removed from the nucleus.1 cml. the work that must be done in order to remove the This amount of work is electron from its nth orbit to infinity is W. or the ionization potential. For the lowest state (n = 1 state. . and when an electron jumped back to the ground state this excitation energy was emitted as a photon with the theoretically predicted wavelength. (77). In this way. . by electron impact. by electron bombardment in a gas discharge tube. the ground in the given states. and the concept of the stationary states of an atom has been strongly confirmed. called the separation energy and is a positive quantity. or a wavelength of X = 1/P number P = 4. 1 = 39523 ern'. the energy of the excited atom should be emitted as a light quantum whose frequency or wavelength is determined by the Bohr frequency condition. The states which correspond to quantum numbers greater than unity are called excited states of the atom because their energies are greater than those of the ground state. The connection between term values and energies was demonstrated experimentally (1914)by the work of Franck and Hertd2' on collisions between slow electrons and mercury atoms.9 X 8 = 2530 A. 9 ev. equal to the largest term value of the atom.
731 THE STATIONARY STATES OF AN ATOM n Wave numbel FIG. Energy level diagram of the hydrogen atom.73. .
7  The stationary states. the separation of the horizontal lines. the value zero corresponds to the complete separation of the nucleus and electron (n = GO). in agreement with experiment. but more complicated. photons with less energy. A rough indication of the intensity of the spectral lines may also be given by means of their thickness. to remove the electron from the ground state i. the absorption spectrum of hydrogen gas under ordinary conditions shows only the lines of the Lyman series. For example. In this figure. with n = 1. oo . 73. the terms. diagrams. Only if the gas is initially excited. increasing from top to bottom. as all those corresponding to lines in the Balmer series. 13. therefore. only the lines of the Lyman series should appear in the absorption spectrum of normal hydrogen. decreases and converges to the value zero as n . The Bohr theory can account for this experimental result in the following way. . Under normal conditions. The Bohr model can account for the lines of absorption spectra as well as for those of emission spectra if it is assumed that an orbital electron can absorb a light quantum only if the energy so absorbed raises the electron into another allowed orbit. (n = 1) to infinity (n = GO). Every emission line. The electronvolt scale can be used directly to obtain the ionization potential of a given level.597 ev are needed.  . Toward the top of the diagram. and there is an appreciable number of electrons in orbits with n = 2. The theoretical explanation for the continuous spectrum will be given later. as shown in Fig. the atoms in hydrogen gas are in the most stable state (ground state). Thus. The length of the line connecting the two levels is directly proportional to the wave number of the spectral line (righthand scale). and the energy levels or terms R/nZ are drawn as horizontal lines. the ordinates give the energy. A spectral line results from the transition of the atom from one energy level to another and is represented in the energy diagram by a vertical line joining the two levels. and the spectral lines of hydrogen can be represented graphically in a very useful way by means of an energy level diagram. cannot be absorbed because they cannot make the electron go even from the innermost (n = 1) orbit to the second orbit. At the left is a scale in electron volts beginning with t b ground state as zero. The energy differences between that level and all other levels correspond to lines in the Lyman series. should .e. A continuous spectrum joins the term series here. to ionize the atom in its ground state. however. According to the theory. indicated by crosshatching. which represent the energy levels or terms. the wave number or term scale is given in units of cm'.148 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP. The absorption of light is then the exact inverse of emission and every absorption line should correspond exactly to an emission line.. is not found to correspond t o an absorption line. At the right. Spectra of other elements can be represented by similar.
the distance of the electron from the nucleus. the existence of fine structure means that instead of a simple energy level corresponding to a given value of the quantum number n. i. as does Planck's constant.. the angular momentum of any isolated system is a constant. This condition is expressed mathematically by the relation where pi is the momentum associated with the coordinate q. lr '4 = 2 ~ # J and . Sommerfeld succeeded in part in accounting for these levels and for the resulting fine structure of the spectral lines of hydrogen and He+ by postulating the existence of elliptic orbits as well as circular orbits.. In polar coordinates these are r. extended over one period of the motion.. and 4. When it is assumed that an electron can travel in an elliptic orbit.. which is at the focus of the ellipse. /PI d4 = p. and two quantum conditions are needed to fix the orbit. for the stationary states the socalled action integral. Refined spectroscopic analysis showed.e. Balmer lines are found in the absorption spectra of some stars in which the temperature of the stellar atmospheres is so high that an appreciable fraction of the hydrogen atoms is in the first excited state (n = 2).e. ergseconds. Sommerfeld and Wilson introduced a new and more general postulate than the original one of Bohr. there are actually a number of energy levels lying very close to one another. must be an integral multiple of h. and by taking into account the relativistic variation of the electron mass with velocity. The quantum condition that has previously been discussed is not enough to fix unambiguously both axes of the ellipse. The Bohr theory in its simplest form predicted with great accuracy the positions of the spectral lines of the neutral hydrogen atom and the singly ionized helium atom. pi is the orbital angular momentum of the system. the azimuthal angle. The two coordinates represent two degrees of freedom to be quantized. they consist of a number of component lines lying close together. Then. When q j = & the angle of revolution. The integral has the units of action. denoted by p+ to distinguish it from the radial momentum p. According to classical mechanics. that these lines are not simple but have what is called fine structure. there are two coordinates which vary. 7 4 Extension of the Bohr theory: elliptic orbits. In terms of energy levels. namely. however. i.741 EXTENSION OF THE BOHR THEORY: ELLIFTIC ORBITS 149 absorption lines corresponding to the Balmer series appear.
and n. so that n. or about the energies in the orbits. . This case was excluded in the Bohr theory as having no physical meaning. that is. . . (728) and (726).150 ATOMIC SPECTRA AND ATOMIC S T R U ~ U R E [CHAP. 3.e2 can take on values only between 0 and 1 because of the geometrical properties of the ellipse. + and n is called the pnpnncipal quantum number. 3. 3. The quantity n. n. 2. is not constant and the integral in the second quantum condition. b = 0. The . n. e is the eccentricity of the ellipse with semimajor and semiminor axes a and b.e2 = 1. and the ellipse becomes a straight line. is given by n . but has not told anything about the size of the ellipses.. may have the values 0. respectively. is called the radial quantum number. may have the values 1. . For 1 . n. and n. The sum n. .n .. . 2. . called the azimuthal quantum number. From Eqs. can take on the values '1. The radial mgmentum p. has been used instead of n for reasons which will soon become apparent.it is therefore set equal to another integer n. Hence. . The latter quantities are determined. and the ellipse becomes a circle. 3. . it follows that The eccentricity of the allowed ellipses is thus determined by the ratio of the quantum numbers n. 2. e = 0. and n. as in the case of the circular orbits. may have the values 1. (727) can be calculated only by more complicated methods. the last result means that the electron would oscillate on a straight line through the nucleus. n. Physically. . . The quantity 1 . .e2 = 0. . The quantity n. b = a. . was not allowed to be equal to zero. by combining the quantum conditions with the expression for the centripetal force provided by the attractive electrostatic force on the electron. 2. 1. 7 This condition is the same as that used by Bohr for his circular coordinates. The discussion so far has clarified the relations between the quantum numbers needed for the elliptical orbits and has defined the eccentricity of the allowed orbits. = 0. a and b. . . n. /pr dr = n ~ . For 1 . For any value of n. The result obtained is In this equation.
and are the same as the values for the circular orbit of radius a and quantum number n. and the number of possible orbit shapes of the same energy is increased over the case in which only circular orbits are considered. one quantum number suffices to determine the permitted energy values. Figure 74 shows the circular and elliptic orbits (drawn to scale) for hydrogen with various n values. the mass of the electron varies in the orbit. the energy .EXTENSION O F THE BOHR THEORY: ELLIPTIC ORBITS 151 FIG. are determined by the principal quantum number n. the previously equivalent states could take on different energies when the special theory of relativity was applied t o the motion of the electron. The results are    where m' is the reduced mass of the electron. According to this model of the atom. and smaller when it is relatively far away. the degeneracy could be removed. If this relativistic effect is taken into account. The velocity of an electron in an elliptic orbit varies (in a circle it is constant).) analysis is more complicated than in the circular case and the details are omitted. a given spectral line may be produced by a number of different but equivalent transitions between which no distinction can so far be made. The shape of the orbit is determined by the ratio nr/n. being greater when the electron is near the nucleus. Although the system has two degrees of freedom. for nr = 1. 2. Relative positions and dimensions of the elliptical orbits corresponding to the first four values o f the quantum number n. The values of the semimajor axis a and the energy W . Sommerfeld showed that in the case of a oneelectron atom. In accordance with the special theory of relativity. and 3. (Sommerfeld's modification for the hydrogenatom. That is. 7 4 . Such a system is said to be degenerate.
. including fine structure.ATOMIC SPECTRA AND ATOMIC STRUCTURE Fro. Energy level diagram of the hydrogen atom. 75.
For example (cf. rather than a single line. with the result that in the hydrogen spectrum. When the predictions of Sommerfeld's theory were compared with the experimental results on the resolution of the Balmer lines and of several He+ lines. The dependence on n* is small because of the smallness of the finestructure constant. for radiation of very small frequency or very long wavelength. Hence. while states with equal values of  .. the behavior of the atom tends to that which would be expected on the basis of classical electrodynamics. classical radiation theory was satisfactory. and the relativistic treatment of the elliptic orbits has the effect of replacing a single energy level by several levels lying close together. The mathematical analysis is again quite complicated. it was found that the theory predicted too many components for the fine structure. But classical theory made possible the calculation of allowed transitions. 73. the levels with different values of n. although it had the disadvantage of not being an integral part of the Bohr theory. the selection rules calculated classically. the frequency of the radiation emitted in the transition of an electron from a level given by n2 to the level nl will be slightly different according to the various values of n* which are possible. The fine structure of the lines of the hydrogen spectrum can be illustrated in an energy level diagram somewhat more complicated than that of Fig.1: An* = f 1.'~' I t was known that for very large quantum numbers n. and valid for very large quantum numbers. In the present case. a group of slightly separated lines appears. They were not an intrinsic part of the Bohr theory but could be obtained by means of Bohr's correspondence p~inciple. Chapter 5). are now drawn side by side a t the same height. The small variations in energy will result in small energy and frequency differences. Agreement between theory and experiment was obtained by the introduction of a selection rule which limited the number of transitions. In Fig. while the Bohr atomic theory did not. were also assumed to be correct for small quantum numbers.741 EXTENSION OF THE BOHR THEORY: ELLIPTIC ORBITS 153  of the electron is found to depend on both n and n*. . According to the correspondence principle. the selection rule stated that only those transitions are possible for which the quantum number n+ changes b y + I o r . (734) Rules like this one were found to be highly important in the theory of atomio spectra. the result is Sommerfeld's formula where rx = 27re2/h = 1/137 is thejine structure constant. 75. This method for deriving the selection rules worked.
73 to represent transitions from one value of n to another. . was quantized according to the condition  where m could have the values .154 ATOMIC SPECTRA AND ATOMIC S T R U ~ U R E [CHAP. 1. it was necessary t o introduce a third quantum number. . As a result. or spatial. According to the Bohr theory. . Hence. In the presence of a magnetic field. . In addition to circular and elliptic orbits. For twoelectron atoms such as neutral helium and singly ionized lithium. serious difficulties were met in the further application of the Bohr theory. 1. threedimensional. it is possible for electrons to have hyperbolic orbits. To show the fine structure lines. . . elliptic orbits in the Bohr atom helped explain the fine structure of the spectra of hydrogen and singly ionized helium. or 2n+ 1 values in all. This component. of f l . . Although the use of relativistic. quantization was necessary. its orbit is a hyperbola. Radiation is emitted when an electron with positive energy jumps to a lower state of positive energy or to a quantized state with negative energy. extending from the limit of the discrete energy levels. these are associated with the continuous spectrum. there is a continuous region of positive energy values. there is a component of the angular momentum in the direction of the field. m. by classical mechanics. The mathematical treatment of the problem of the + + . only the circular and elliptic orbits are quantized. If the free electron has kinetic energy. For a given value of n. there are oblique lines corresponding to changes in n. 7 no and different values of n are drawn above one another. and the series of discrete lines in the spectrum is followed by a continuous spectrum. although an electron does not radiate in any orbit.no. n . . each level has n sublevels corresponding to the n values which n. the energy and wave number scales would have to be very large. 0. The energy of a oneelectron atom has been defined in such a way that it has the value zero when the electron is a t rest completely removed from the nucleus. To try to account for the splitting of spectral lines in a magnetic field. the magnetic quantum number. can have. all positive values of the energy are possible and. whose direction may be taken to be parallel to the zaxis of a threedimensional coordinate system. the new quantum number was related t o the orientation of the elliptic orbits in space.n. The wave number for a transition is still given by the vettical distance between two levels. there were serious discrepancies between theory and experiment. because of the very small spacing between the sublevels. . its total energy is positive and. 75 Failure of the Bohr theory: wave mechanics and the correct quantum numbers. In place of the vertical lines of Fig.
It gradually became apparent that the difficulties of the theory were caused by an essential failure of the atomic model that was being used. These are the principal quantum number n. Multiplets are different in character from the finestructure lines in that the multiplets can be quite widely separated. The wellknown sodium lines with wavelengths of 5895. however. triplets are found. The difficulties discussed above (and others not mentioned) could not be resolved by further changes in the Bohr theory. The solution to the problem came in 1925 and the years following in the form of wave mechanics or quantum mechanics. had t o be introduced. still only partial. Some of the basic ideas of the newer theory will be discussed later in this chapter.96 A. In the magnesium spectrum. For the present it will suffice to state that wave mechanics provides a theory which does not depend on the classical picture of electrons revolving in planetary orbits around the nucleus. In the absence of a magnetic field.which replaces m.751 FAILURE OF THE BOHR THEORY 155 stationary states again led to too many energy levels and the selection rule.93 A and 5889. One of the most important results of the new theory has to do with the quantum numbers. the explanation of the emission and absorption of radiation in terms of transitions between stationary states. The values which the new numbers can take on are    . respectively. The new theory yields all of the correct ideas and results of the Bohr theory. Moreover. (736) Am = 0. wave mechanics solves many of the problems which could not be handled by the Bohr theory. and the RydbergRitz combination principle for spectral lines. the orbital quantum number 1. It is found that in the case of the oneelectron atom treated threedimensionally. Additional difficulties arose in connection with the multiplet structure of the spectral lines of many elements. are an example of a doublet. and the lines become more complex as the elements move to the right in the periodic table. and the magnetic quantum number ml. The success of the theory in dealing with the effect of a magnetic field on spectral lines was. which takes the place of n.. three quantum numbers appear in a direct way in the mathematical solution of the problem. multiplets could not be accounted for on the basis of the quantum numbers n and n*. as are the other members of the series to which the D lines belong.. fI. . such as stationary states with the correct energy values.
an arbitrary addition to wave mechanics. Bohr's mode of description. Dirac succeeded in obtaining an equation which satisfied the requirements of the special theory of relativity. it does not allow the structure of the atom to be depicted in a familiar way. from which the essential features of atomic phenomena can readily be made out. First. 3. The spin angular momentum of the electron was represented by a quantum number that could have the values +3 or 3. the fundamental equation of wave mechanics. The electron spin was. One of the most important applications of modern atomic theory is to the interpretation of the structure of the periodic table in terms of the physical ideas that have been    . can have the values 3. and not an intrinsic part of the theory. ( h / 2 r ) 9 where m. a t first. was not relativistically invariant. (738) spin angular momentum = m. The concepts from which the theory is built are more abstract than those of the Bohr theory and it is quite difficult to get an intuitive feeling for the problem of atomic structure.156 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP. It is possible to go quite a long way with the Bohr model without corning into du3sgreement with facts. To account for multiplet spectra and for the socalled "anomalous Zeeman effect. It had been known that the Schroedinger equation. but this difficulty was removed in 1928 by D i m . the Bohr theory may be considered to be a first approximation to wave mechanics. Wave mechanics has two disadvantages." wave mechanics was supplemented by the electron spin hypothesis of Uhlenbeck and Goudsmit (1926) according to which the electron was assumed to have a spin of magnitude 3h/27r and a magnetic moment equal to ( e / 2 m ) (h/2?r). Thus. and showed that all four quantuni numbers followed from his theory without the introduction of any arbitrary sssumptions. In this sense. 7 4 Atomic theory and the periodic table. 7 The necessary selection rules also turned out to be a direct result of the mathematical treatment and did not have to be introduced in an artificial way. Hence. consistent theory of atomic spectra and of many other atomic properties. four quantum numbers are needed to account for atomic spectra. provided that the quantum numbers of the wave mechanical theory are used and that corresponding corrections are made in the conclusions resulting from the Bohr theory. The result was a complete. is still often used. although both its conceptual basis and its mathematical methods have been replaced by those of the newer theory. Second. the mathematical methods are considerably more advanced than those of the older theory. and it was necessary to introduce the new quantum condition.
First. four 2p electrons as Zp4. The chemical properties of an atom are completely determined by the number of outer electrons. The basic idea underlying this description is that the number of electrons in an atom is equal t o the atomic number. For the second or Lshell. ATOMIC THEORY AND THE PERIODIC TABLE 157   . it is possible to assign to any electron in an atom a value for each of the four quantum numbers n. 4. and so on. Orbital angular momentum quantum number ( I ) : Electron designation: Atomic atate: 0 1 2 3 4 5 6 a p d f g h i S P D F G H I . which states that no two electrons in an atom can have the same four quantum numbers. and an electron can be identified by the particular set of values that it has. the letter indicating the value of 1. and it is usual to denote the number of such electrons in an atom by an index. the number giving the value of n. developed about atomic structure. 3. for n = 1. . The first or Kshell is full when it has two electrons. It follows from the second rule that this shell can have no more than two electrons with 1 = 0. Consider first the case n = 1. two deductions can be made. M. subgtovps can form within each shell. n = 2 and the shell can contain up to 8 electrons (2n2 = 8). the Padi exclusion principle. As a result of the second rule. . . 1. and so on. 2. The possible quantum numbers that an electron can have are limited by a fundamental rule.. and m. ml. As a result of the wave mechanical theory of the atom. An electron with n = 3 and 1 = 2 is called a 3d electron. Thus. All the electrons belonging to the same subgroup are called equivalent electrons. . there can he no more than 2n2 electrons with the same quantum number n. When this principle is applied to the electrons in an atom.761 . Any atomic electron structure is formed by + . . The only possible value of 1 is zero. L. all of the electrons in a subgroup having the same values of n and 1. . . Second. N . the shells being denoted by K. so that there can be a t most two electrons with n = 1. The subgroups are designated by letters. two 1s electrons are written as ls2. All the electrons which have the same total quantum number n are said to belong t o a shell. seven 4d electrons as 4d7. and no more than 6 electrons with 2 = 1. . as in Table 72. may be either +3 or 3. and r n t must also be zero. These two rules can be shown to account completely for the form of the periodic table. there can be only 2(2E 1) electrons with a given value of 2. The quantum number m. and this number is in turn fixed by the nuclear charge. The structure of the periodic table can now be described in terms of the possible numbers of electrons in the subgroups. respectively.
the electron configuration for this atom is IS%. and the Kshell is closed. which has one electron. at krypton the first two subgroups of the Nshell are filled. both of which can go into Is orbits. closes the first two periods of the periodic table (cf. and at xenon the first two subgroups of the 0shell are filled. a t neon. but the arrangement of subgroups explains the observed spectra of the elements. the electron can occupy one of the higher orbits (states). The values of the potential energy can be calculated from the wave mechanical theory of the atom.and Gshells at helium and neon. At argon the first two subgroups of the Mshell are filled. . This state is the one that. it is clear from Table 73 that this electron in the ground state occupies a Is orbit. with four electrons. The quantum numbers chosen for the added electron are such as to place this electron into the most tightly bound state possible. as is the selectron subgroup of the Gshell. Not only are the chemical properties of the atoms accounted for. The process can be continued until the entire periodic table is accounted tor. the rare gases are associated with the completion of either a shell or a subgroup.  .the Kshell is full. as required by the chemical properties of the elements. 7 Shell n K L M 3 38 3p 3d 1 la 2 2s 2p Possible electrons Maximum number o f electrons Total. The subgroups of electrons which are possible are shown in Table 73 for values of n up to 4. the entire Lshell is filled. respectively. if the atom is excited. the electron configurations of the elements are listed in Table 74. Thus. After each rare gas a new subgroup begins. making the potential energy a minimum. Beryllium.158 ATOMIC SPECl'RA AND ATOMIC STRUCIWRE [CHAP. for the given number of electrons. has the configuration 1s22s2. leads to the smallest value for the total potential energy of the atom. The process of adding electrons continues until. Helium has two electrons. 2n2 2 2 6 2 6 1 0 18 2 8 the addition of a single electron to the electronic structure characteristic of the preceding atom in the periodic table. The filling of the K. The configuration of the normal helium atom is then ls2. Table 14).  . The third electron needed for the formation of the lithium atom goes into a 2s orbit (to keep the potential energy a minimum). In the case of hydrogen.
Table 74 shows that each alkali metal has one or more closed shells and subgroups and a single outer electron. of the chemical properties of the elements. His results can be interpreted in terms of modem atomic theory and.: r = R(Z .6 ev for hydrogen. 4s orbits for Li. with a potential of 5.112 . the Bohr theory can be used to give a very neat picture of the production of the characteristic xray lines.. consequently. K. begins. in fact. The last column in Table 7 4 gives the experimental values of the ionization potential or separation energy. (t Equation (739) suggests that a K. in basic physical terms. The alkali earths have two outer electrons.4 ev. It can now be said that modern atomic theory has solved the problem of the interpretation. the ~or~struction of the second. Na.:2). The electron occupies the 28. and so on through the columns of the periodic table. A vacancy in the Kshell can occur if one of the two electrons in that shell is struck by an energetic cathode ray and knocked either out of the atom or into an unfilled outer shell. loosely bound. With this electron. Such a jump can occur only if one of the two electrons in the Kshell is missing. and all have similar doublet spectra. 3s. All of the alkalis have the same chemical valence and similar chemical properties. these values offer clear evidence for the structure of the electron shells. l i e is the result of the jump of an electron from the Gshell (n = 2) to the Kshell (n = 1). The second electron in the helium atom has an ionization energy of 24. The trend of the experimental values of the ionization potential is exactly that predicted by the theory.5 ev as compared with 13. and the helium electrons must therefore be much more closely bound than the single electron in the hydrogen atom. The importance of Moseley's work on the characteristic xray spectra of the elements was discussed at some length in the fourth chapter.5 ev. it must be at a considerably greater distance from the nucleus and is. It was seen in Section 44 that the wave numbers of the K. lines of the xray spectrum are given by the formulas K. This shell is filled in the neon atom with an ionization potential of 21. Subsequently. as expected from the electron configuration. an electron from the Gshell may drop into the vacancy in the Kshell with the emission of a quantum . or Lshell. The third electron which is built into the lithium atom has the small ionization potential of 5.1 ev. respectively. and L. 77 Atomic theory and characteristic xray spectra.771 ATOMIC THEORY AND CHARACTERISTIC XRAY SPECTRA 159   For example. and the Mshell starts with sodium.
138 7.56 6.01 15.54 13.Element Ionization potential.595 24.149 10.984 8.339 6. .296 11.320 8.614 17. ev 13.418 21.74 .83 6.264 14.755 4.580 5.357 13. H H e Li Be B C N 0 F Ne Na Mg A1 Si P 8 C1 Ar K Ca Sc T i v .559 5.55 10.111 6.644 5.390 9.
(Continued)  .
7 .ATOMIC SPECI'R.4 ABD ATOMIC STRUCTURE [CHAP.
.
76. An Mshell electron is screened from the nucleus by the other 7 ell electrons and the Kelectrons. This concept arose as an extension of the treatment of light sometimes as waves and sometimes as particles. line. in principle. Simple term scheme of xray levels showing transitions corresponding to xray lines.1)e. This electric field is equivalent to that of a positive charge of magnitude (2. 7 Fra. The design and operation of optical and xray spectrometers as well as the interpretation of the results obtained with these instruments depend on properties   . the same as that for the K. and diffraction. Light and xrays have been treated as waves during the discussion of experiments involving reflection. The fundamental idea of wave mechanics is that of the wave nature of matter.ATOMIC S P E m R A Ah'D ATOMIC STRUCTURE [CHAP. Just as for the optical spectra. 76. The explanation of the experimental results for the La l i e is. refraction. and a simple example of such a diagram is shown in Fig. The single Kshell electron tends to screen the nucleus and makes its effective nuclear charge about one unit less than the actual charge. 78 The basic ideas of wave mechanics. the stationsry energy levels and the spectral lines can be represented graphically in a term diagram. The electron which goes from the Lshell to the Kshell moves in the electric field of the positive nuclear charge and the negative charge of the electron remaining in the Kshell. Some of the basic ideas of wave mechanics are applied often in nuclear physics and a brief discussion of these ideas is in order. of radiation whose frequency is that of the Ka line. This time the vacancy is in the Lshell and the radiation is emitted when an electron from the Mshell (n = 3) drops into the vacancy. and the nuclear charge is effectively reduced by about 7 units.
and equations are written for the balance of energy and momentum. and e are in cgs units and V is in volts. while for other experiments radiation must be regarded as particlelike. This fact suggested that the existence of electron waves might be shown by using crystals as diffraction gratings. This duality is shown in the expression for the momentum of a photon  The momentum (a particle property) is related directly to the wavelength (a wave property). typical particle properties.781 THE BASIC IDEAS OF WAVE MECHANICS 165  typical of waves. '~' that matter also has a dual (particleIn 1924. of the order of magnitude of the distances between atomic planes in crystals. it was amenable to experimental tests. light must be assumed to consist of particles (quanta. and the wavelength was determined from the Bragg equation. He assumed that the relation (741) holds for electrons and other particles as well as for radiation. On the other hand. as is done for xrays."' On bombarding a single crystal of nickel with an electron beam. The analysis of the experiments was similar to that used for xray diffraction by crystals. and experiments based on this idea were done in 1927 by Davisson and Germer. de B r ~ ~ l i esuggested like and wavelike) character. (741) and (742) are combined. then  where m.2 X lo' cm. Suppose that the particles are electrons which have been accelerated to a velocity v under the action of a potential difference V. in the photoelectric effect. they found that the reflected beam showed maxima and minima which could be explained in terms of the diffraction of electron waves. The results of the calculation are expressed in terms of the shift in wavelength because this is the quantity that is measured experimentally. In the Compton effect. I t is clear that some experiments involving radiation can only be interpreted on the hypothesis that radiation is wavelike in nature. Although this assumption was a bold and surprising one to make. If V is 100 volts. The . v. The wavelength associated with the electron is. the interaction between aqxray and an electron is treated as an elastic collision between two particles. when Eqs. If v is small compared with the velocity of light. the value of A should be 1. or photons).
166 ATOMIC B P E AND ~ ATOMIC ~ U C I W R E [CHAP. have also been found to show diffraction effects. and v is the frequency. such as helium atoms and helium molecules. we may write *in the form = +e2rdd (745) * * where $ is a function of z. If the expression for is inserted into E q . In view of the wave nature of particles. t) is the amplitude of the wave 8[180~iated with the particle and u is its velocity. Particles of atomic size. The usual threedimensional wave equation i n rectangular Crvteaian coordinates is where (z. I f we desire a aolution that will represent standing waves (such as those on a string fastened at both ends). z. and z. moving in an electroetatic field of force. 77."' This equation can be derived quite easily for the case of a particle. The dual wave and particle nature of matter can be expreseed mathematically by means of a wave equation first derived by Schroedinger. we get * where X = u/v% the wavelength. Diffraction patterns were also obtained"' when electrons passed through thin films of matter such as metal foil or mica. (744). and the patterns n experiments on the dsraction were exactly analogous to those obtained i of xrays. In Fig. or a beam of particles all with the same energy. 7 quantitative agreement between the p d c t i o n s of the de Broglie hypothesis and the experimental results was remarkably good. we may write for the energy W and the momentiun p of a particle and the wavelength is then We assume now that the speed v of the particles is much less than that of . y. the similarity between the effect of xrays and that of electrons is so striking as to prove conclusively the wave properties of electrons. y. but not of the time t. and there is no doubt about the correctness of de Broglie's hypothesis.
The kinetic energy of the particle is the difference between the total energy W and the potential energy U. which gives the wave function aa a function of the coordinates. (Courteay of Dr. Cornell University.055 A. .) light. Karnmerer. (Courtesy of Mr.U. 0. Brookhaven.U) and.) (b) DXraction pattern obtained from a gold cry~tsl with electro~w of momentum corresponding to a wavelength of 0. and we can write *nw2 = W .77. we get Equation (750) is the S c h r d i g e r equation. F. Bauer. the potential energy is also a function of the coordinates. from Eq. (747c). 8. (746). (a) Diflraction pattern obtained from a gold crystal with xrap of wavelength 0. Then p = nw = 4 2 m ( W (748) .709 A.Nationa1 hboratory. On substituting this expression for in Eq.781 THE BASIC IDEAS OF WAVE MECHANICS 167  FIG.
These values are the same as those found for the stationary states in the Bohr theory. Representation o f a onedimensional standing electron wave in a The mathematical problem which must be solved consists of inserting the appropriate expression for the potential energy into the Schroedinger equation and solving for the wave function $. In order that the electron wave may just fill the circumference of a circle. Bohr's quantum conditions also follow from the idea of the wave nature of matter. Bohr orbit. the Schroedinger equation would be written in spherical polar coordinates in this case. The oneelectron at. and Eq. or (752) 2 m = nX.ATOMIC BPECTRA AND ATOMIC BTRUCTURE FIG.78. but the result is obtained from the mathematical formulation of the experimentally proven idea of the wave nature of matter rather than from a hybrid set of classicalquanta1 assumptions. . where r is the radius of the circle and n is an integer. X = h/p = h/mv. consider the electron in the hydrogen atom as a standing wave extending in a circle around the nucleus. as in Fig.(752) becomes  But w is the angular momentum of the electron regarded as a particle. As a simple example. actually. where r = d x 2 " y 9.om is treated by setting U = Ze2/r. the circle must contain an integral number of wavelengths. I t is found that physically significant solutions for $ for bound states of the atom exist only when W has the values + where n is an integer. 78. By de Broglie's hypothesis.
The meaning of the uncertainty principle can be illustrated by an idealized experiment. I t has been shown in wave .78 J THE BASIC IDEAS OF WAVE MECHAXICS 169 and we see that the wave mechanical picture leads naturally to Bohr's postulate that the angular momentum is equal to an integral multiple of h/27r. The question arises: If an electron must be described sometimes as a wave and sometimes as a particle. and the location of an electron. These quantities can be determined only with accuracies limited by the relation AzAp . AtAE . the less precisely . According to classical mechanics.h.can the momentum be measured. the difficulties inherent in Bohr's postulates are all satisfactorily removed by wave mechanics. the less accurately can the energy change of the atomic system be determined. in principle. the mechanism of emission does not have to be introduced as a separate postulate. and conversely. The concept of the dual particle and wave nature of matter leads to another important result. (755) The condition of Eq.'8' which states that the position and momentum of a particle cannot be determined simultaneously with any arbitrary desired accuracy. and the latter appear as a consequence of the condition for a standing wave rather than as arbitrary assumptions. I t can also be shown that in wave mechanics the emission of a photon of frequency v is a logical consequence of an energy change of the atom. (754)  where Ax is the error in the determination of the position and Ap is the error in the momentum. seems to present some difficulty. The product of the two errors is at least of the order of magnitude of Planck's constant. Consider the following attempt to measure simultaneously the position and momentum of an electron with an imaginary microscope of very high resolving power. But a wave is extended throughout a region of space. (754) means that the more precisely the position of a particle is determined. is it possible to know exactly where an electron is in space at some given instant? The answer to this question is given by Heisenberg's uncertainty principle. considered as a wave. A similar relation holds for the energy of a particle and the time. and conversely. a particle has a position and momentum which can. The concept of the electron as a standing wave rather than as a particle revolving in an orbit also removes the difficulty that in the Bohr theory the electron travels in its orbit without radiating. The last condition means that the more accurately the time a t which an atomic event occurs is determined. be determined with any desired accuracy. This type of proof can be extended to the other quantum conditions. Thus. as in the Bohr theory.h.
79. ($la dz dy dz. to make Az as amall as possible the wavelength must be very small and either xrays or 7rays must be used. which is the Heisenberg relationship. The uncertainty in the momentum is of the order hv sin A. a momentum of the order of hv/c.170 ATOMIC SPECl'RA AND ATOMIC STRUCTURE [CHAP. them are mathematical  .79. Then Ax is the uncertainty in the determination of the position of the electron. because of the Compton effect.h. as in Fig. FIG. Thus. Other phyeical result8 can be expressed in terms of Qpeddbn uaZues. it follows that Ap Ax . 2). (756) and (757). This discussion shows that one of the causes of the uncertainty relationship is the fact that the process of measurement itself perturbs the particle whose properties are being measured. Ap (757) N From Eqs. in the momentum of the electron cannot be determined exactly because the scattered photon can enter the microscope anywhere within the angle A. From an operational viewpoint. Ap. and A is the halfangle of the lens used. The interpretation of the Schroedinger wave function 9 presented a problem until Born (1926) showed that logical consistency and agreement between theory and experiment could be obtained provided that 9 is regarded as a merrsure of the probability of finding a particle in a given region. is the probabiity that the electron will be found in the volume element & dy dz about the point (2. 7 . The change. y. h is the wavelength of the light used. But when this photon is s c a t t e d it gives to the electron. This result gives a clue to the failure of Newtonian mechanics when it is applied to atomic physics. . Now the position of an electron can be'measured with a microscope only if at the very least one photon is deflected from ita initial direction into the microscope. and there is no real point in thinking of an atom in the aame way that we think about a macroscopic planetary system. the exact location of an electron has therefore no real meaning. where the vertical lines denote the absolute magnitude of 9. optics that the resolving power of a microscope is given approximately by where Ax is the distance between two points which can just be resolved by the microscope. Imaginary experiment in connection with the uncertainty principle.
1. The solution of the Schroedinger equation for significant physical problems usually involves mathematical methods beyond the scope of this book. for the results of a sbgle measurement. There are. or for the average of a large number of experiments. This interpretation is in accord with the statistical nature of the results of experiments on atomic systems. statistical mechanics. but is reflected when it strikes either end of the range. a few simple problems whom solutions show some of the basic differences between classical mechanics and quantum mechanics. Hence. " We consider a particle of maw m which can move freely over the range 0 < x < a. One of these problems is that of a f ~ e particle e in a "onedimensional box. the theory of chemical reaction rates. in the sense of probability theory. and many electric and magnetic phenomena can also be treated by means of the ideas and techniques of wave mechanics. The success of wave mechanics is not limited to the examples that have been discussed here. 70 The solution of the Schmdnger equation: some useful examples and d t s . The Schroedinger equation for the particle is then  If we set  the general solution is #(z) = A sin Icz + B cos kx. = 0. and the condition at z = a requires #(a) = A sin krr (761) We cannot take A = 0 because there would then be no solution. or . we must have sin ka = 0. The quantitative treatment of many problems in nuclear phyaics depends on the application of wave mechanics.191 THE SOLUTION O F THE SCHROEDINQER EQUATION 171 expectations. the physics of the solid and liquid states. Molecular physics. (760) The boundary condition # = 0 a t z = 0 requires B = 0. however. These conditions can be represented in Schroedinger's equation by aasuming that the potential energy U is zero for 0 < z < a. Many examples of these applications will be found in the books cited in the general references a t the end of this chapter. Particle in a box. and that the wave function # vanishes a t z = 0 and at z = a.
is called an eigenjunction or proper function. it is easily verified that while for n # m. The wave functions have another important property. that of orthogonality: for n = m. . there is a momentum pn given by  n h n h a n . (763). each value of Wn is called an eigewalue or proper value.2a = a 2a a The first three wave functions are plotted in Fig. p n = .(z) = A sin 1 a (763) with n related to k by Eq. (765). in a given state of energy.. for each energy value Wn. it comes from the physical boundary conditions and gives rise to the quantized values Wn of the energy. (762). We can calculate the probability that the particle. 710. The momentum is also quantized. will be located within any element dz of the onedimensional box.172 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP.= . That relation and Eq.. and the corresponding wave function +. The interpretation of the wave function requires that + . and to a wave function given by Eq. or that A = G. and The wave functions are said to be normalized when they satisfy Eq.nh. Equation (764) is a quantization condition which introduces the quantum number n. (759) require that the energy W can have only the values An energy level Wn corresponds to a value of n. 7 The only allowed solutions of the wave equation are the functions nm +.
a a (769) which is a wavelike function.sin dx. the quantum mechanical result is very different from the classical one for which the probability is the same for any element dx. a t x/a = Q and x/a = 3 in this case. it is shown graphically in Fig. y. box. where the probability is zero. Thus. b. and zaxes. 711. c. There are actually nodes. According to quantum mechanics. 711 for the state defined by n = 3. If the particle is confined within a threedimensional box with edges of lengths a. The probability distribution for finding a particle of given energy at points in a onedimensional box (case n = 3). this probability is 2 2n7rx l$(x) 1 dx = . The first three wave functions for a particle in a onedimensional FIG. . 710. parallel to the x. and W 3 = 9h2/8ma2. n2. n~ denote any set of three positive integers. the normalized wave functions for the stationary states and the energy levels are found to be where nl. respectively.THE SOLUTION OF THE SCHROEDINGER EQUATION 173 Fro.
has the magnitude. an electron also has a spin angular momentum S characterized by a spin quantum number which can take on the values f+ h / 2 r . namely. that the energy has the eigenvalues given by Eq. (737) defines a limited number of orientations which L may take up with respect to a specified axis. L. has magnitudes which are given by + According to quantum mechanics. in Fig. The projectian of L along a specified axis.e2/r into the Schroedinger equation (750). The magnetic quantum number mr. orbital angular momentum is said to be spatially quantized in terms of the quantum numbers mr. These orientations are shown. or coupled. 7 2. which may take on the values given in the second of Eqs. . the component of the spin angular momentum along a specified axis S is quantized according to the rule The orbital and spin motions of the electron are combined. L = m h / 2 ~ The . Another very important result involves the angular momentum and its quantization. as well as its magnitude. which approaches zero in the l i t of large values of I. the magnitude of the spin angular momentum is  . 712. in this case. The angular momentum vector L can never lie entirely along a chosen axis but only a t a minimum inclination given by cos' (1/1 1)'12. one of the results obtained has already been mentioned. (737). for 1 = 3. The angular momentum is a vector quantity and its direction must be specified. given by where 1 is the orbital quantum number. The hydrogen atom is treated in quantum mechanics by putting U = . (751). I t is found that the angular momentum L (vector quantities are indicated by boldface letters).174 ATOMIC SPECTRA AND A M M I C STRUCl'URE [CW. More precisely. The solution of the equation is complicated. usually taken as the zaxis. whose possible values are given in the third of Eqs. The hydrogen atom: a n g h r momaturn. parity.
depends on the coupling between the orbital and angular momenta of the individual electrons. the parity of a wave function #. such as energy and angular momentum. which is related to the symmetry properties of # and has no . and J. according to the rules of vector addition to give a resultant angular momentum J characterized by a quantum number j. Spatial quantization of the orbital angular momentum. The study of atomic spectra has shown that the predominant type of coupling is that in which all the orbital angular momentum vectors Liof the electrons combine to form a resultant L and. The total angular momentum is then given by the relation  This type of coupling is called LS.712.ors Si combine to form a resultant S. or RussellSaunders coupling after its discoverers. When changes in atomic states are produced by the action of two or more electrons.791 THE SOLUTION OF THE SCHROEDINGER EQUATION m = 175 4 FIG. j can take on only the values I =t3. We have seen that the quantummechanical study of the hydrogenlike atom has led to certain quantum numbers which apply to physical quantities. S. denoted by J. There is another quantum number. the value of the total angular momentum of these electrons. all the spin angular momentum vect. well known in classical mechanics. independently.  for a given value of I. The total angular momentum is quantized and there are appropriate rules for the possible values which may be taken on by the quantum numbers corresponding to L.
3rd ed. and T. 1955. 1958. New York: McGrawHill. which is a purely quantummechanical concept. Chapters 28. and J. 8. when P = 1. Practical Spectroscopy. B. LORD. . 1948. C. A. The parity is then quantized and has the quantum number P. LOOFBOUROW. 7. P. 7. 7. Phyeiee. Chapters 5. z + 2. 6. y + y. FRENCH. As for other quantum numbers. KENNARD. 1948. x + x. when P = 1. New York: McGrawHill. GENERAL F. J. P. 8. The parity of a wave function refers to its behavior under a simultaneous reflection of the space coordinates. EXPERIMENTAL TECHNIQUES G. B. Y . the system is said to have even parity. Since the parity is defined in terms of the Schroedinger wave function $. 1959. G. there are selection rules involving the change of parity in a transition. R. 2) = $(x. Y. HARRISON. Chapters 3. Principles o f Modern Phyaks. Inbodudion lo Moden. New York: Van Nostrand. 1933. KORFF. Phyaks. H. Chapters 5. We can treat P as a quantum number. it can have no classical analog. HARNWELL and J. K. RICHTMYER.176 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP. R. the system is said to have odd parity. (778a) + the wave function may be invariant under the reflection. New York: Wiey. HOAG and S. Ezperimental Atomic Physies. it is found that the parity is related to the orbital quantum number 1. 5th ed. LEIQHTON. LIVINQOOD. New York: PrenticeHall. E. J. R. LAURIT~EN. Electron and Nuclear Physics. Equations (778) can be combined in the form where P = 1. or i t may change sign. R. and the property it defines is called the parity of the system. 7. Prineiph of Moden. In the case of a hydrogenlike atom. We may distinguish between two ideal cases #(x. A. 7 relation to a classical physical quantity. z). Chapters 5. New York: McGrawHill.
H. Reading. New York: PrenticeHall. E. 1957. PAULING and S. 1951.Quantum Mechanics: NonRelativiatic Theory. 1959. N. New York: McGrawHill. BOHR. GOUDSMIT. New York: Academic Press. WITTKE. F. 11. Quantum Theory of Atomic Structure. 4th ed. 1951. New York: McGrawHill. New York: McGrawHill. New York: Academic Press. Zntroductim to Atomic and Nuclear Physics. STEPHENS. London: Butterworth. BACHER and S. Oxford: Clarendon Press. BOHM. M. J. The Spectrum of Atomic Hydrogen. 1958. L. Atomic Theory and the Description of Nature. G. HERZBERG. Vol. WHITE. I. zntroduction to A h i c Spectra. 1934 M. 1957. W. SCHIFF. G. SLATER. 1958. 1955. W. New York: PrenticeHall. Cambridge: University Press. HEITLER. 1930. The Principles of Quantum Mechanics. 1948. P. H. 1957.: AddisonWesley. Quantum Theory of Matter. 2nd ed. Quantum Mechanics. AtOmic Energy States. DIRAC. Atomic Spectra and Atomic S h t u r e . E. Atomic Physics. 1955. F. 1958. L.REFERENCES ATOMIC SPECTRA L. 1955. New York: McGrawHill. Reading. . SERIES. New York: McGrawHill Vol. A. C. R. POWELL and B. R. 1961. KAUZMANN. New York: McGrawHill. W w e Mechanics and Its Applications. D. Quantum Mechanics. Mass. N. J. LANDAU and E. G. 1956. Mass. Mom. Atomic Physics. BORN. New York: Wiley. C. J. IntrOdUctiOlL to Quantum Mechanics. D. U. F. PEASLEE. 1932. MOWand I. Reading. 1946. DICKE and J. 2nd ed. GOUDSMIT. 1958. New York: Rinehart. C. 1954. Quantum Mechanics. Basic Physics of Atoms and Molecules. Cambridge: University Press. HARNWELL and W. SEMAT. 11. 1960. I. Quantum Chemistry. P. D. LIPSHITZ. MANDL. Elements of Wave Mechanics. M. Oxford: Clarendon Press. L. Quantum Theory. 2nd ed. Elements of Atomic Physics. FANO. FANO and L.: AddisonWesley. N. 1960. 1960. P. SLATER. ATOMIC THEORY AND ITS APPLICATIONS N. 1934. Mass. 6th ed. Oxford: Clarendon Press. London and Glasgow: Blackie. Elementary Wave Mechanics. CRASEMANN. New York: Hafner. SNEDDON. Oxford: Clarendon Press. London: Pergamon Press. W.: AddisonWesley. New York: Dover. New York: McGrawHill. The Structure of Atomic Spectra.
512 (1914). 1924. BOER. The Physical Principles of the Qwrntum Theory. in wavenumber units. DAVISSON Nickel. and 3662. cit." Phys. then calculate the difference between the wavelengths of the first three Balmer lines of hydrogen and deuterium. 476. P. has a spectrum in which the Balmer lines are displaced slightly from the lines of ordinary hydrogen (mass 1). 3. BRILLOUIN. 7 1. Chicago: 8. ergs. 117 (1923). 2. Roy. Sn'. HERTZ.n2 = 1. Part I . 3. respectively.576 cmI and RHe = 109722. 14. Physik 13.. FRENCH. Soc.576.267 cml. 30. op. G. The Fundamental Postulates. Cambridge: University Press. P. THOMSON. J. 619 (1928). DE BROGLIE and L. Ges. BOAR. FRANCK also A. C. measured in air. DE BROQLIE. Selected Papers on Wave Mechanics. 16. phys. N. "On the Quantum Theory of Radiation and the Structure of the Atom. What is the relationship between the emitted and orbital frequencies when nl and n2 are both very large? 4.178 PARTICULAR ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP.24 A. electron volts. with nl . gen. "Experiment on the Diffraction of Cathode Rsys. deuterium. in L. Verhandlungen. "Diffraction of Electrons by a Crystal of 5.. 7.457. 6. and the index of refraction of air as 1. 1929. (London). 600 (1928)." ibid. Rev.. and G. 663 (1928). and calories per gram. W. Find the value of the Rydberg constant for deuterium. HEISENBERG. GERMER. New York: Dover. A119. 5. National Acad. University of Chicago Press. 217.  1. ref. E. On the Application of the Quantum Theory lo Atomic Structure. The hydrogen isotope of mass 2." Proc. also Z. 394 (1915). 1.43 A. A117." Phil. The best values of RE and Rae are now considered to be Ra = 109677. Calculate the energy of each of the five lowest stationary s t a t a (energy levels) of the hydrogen atom. What is the value of n for each line? Take RH as 109. SCHROEDINGER. Show that the frequency of revolution of an electron in a Bohr orbit with quantum number n is 4 ~ ~ r n ' e ~ ~ ~ vorb = h3n3 Show next that if an electron makes a transition from a state for which n = nl to a state for which n = n2.677. L. Proc. 705 (1927). London: Blackie. H. London: Blackie. 4. and L. Three lines in the Balmer series of hydrogen have the wavelengths 3682.. 857 (1913).0002837. 2. 26. N. Calculate accurately the value of the ratio . 1929. Collected Papers on Wave Mechanics. 3669. Mag.82 A. G. then the frequency v of the emitted radiation is intermediate between the frequencies of orbital revolution in the initial and final states. 30. "On the Constitution of Atoms and Molecules. 1929. Chapter 12.
Co. The noble gases Ne. If V exceeds a few thousand electron volts. Rb. hydrogen wrts bombarded by electrons with known kinetic energy. What is the wavelength of electrons accelerated through n.19 ev. 6. Prove that b/a = k/n. In an experiment of the FranckHertz type. What is its magnitude? Compute the wavelength of electrons for each of the speeds listed in Problem 5. 100.08 ev. Calculate the wavelengths of the K . Show that the wavelength of an electron can be written in the form The quantity h / m w is called the Compton wavelength of the electron. From the constancy of angular momentum a t kh/2r and the constancy of total energy at 2r2m'e4Z2/n2h2. 10. Cs. and Ag. Prepare a table showing the number of electrons in each shell and each orbit for these elements. Br. Na. 13. Ni. the effect of the relativistic increase of mass with velocity must be considered.05 ev. (See Born. Appendix XIV. 13. Show that the wavelength is given by where m/mo is the ratio of the moving mass of the electron to its rest mass. 1000 volts? 12. 1 0 ' . Kr. 13. xray lines of Al. K. 12. Show that the wavelength of electrons moving a t a velocity very small compared with that of light and with a kinetic energy of V electron volts can be written What is the wavelength of electrons with energies of 1. The temperature of the hydrogen was high enough so that the gas was dissociated into atomic hydrogen. Add to this table the electron configurations attributed to the alkalis.PROBLEMS 179 of the mass of the hydrogen atom to that of the electron. lo6. potential of lo4 volts? 106. Use the result to calculate the value of e / m for the electron. Mo. and to the halogens. and give the value of n and 1 associated with each shell and each orbit. I. I t was observed that radiation was emitted when the electrons had the following discrete energies: 10. Chapter 6. K. What is the Compton wavelength of a proton? What is the wavelength of protons a t each of the above speeds? 11.75 ev. and compare the values with those listed in Table 41. 12. where m' is the reduced mass of the electron.) 8. Atomic Physics. How is the occurrence of 15 rare earth elements with similar chemical properties explained? 9.31 em'. and Rn have electron configurations in which the s and p orbits of the outer shell are filled. Fe. What is the wavelength of the radiation corresponding to each electron energy? To which spectral series do the lines belong? What are the quantum numbers of the initial and final states in the transition represented by each spectral line? 7.31 ev. 10. and Fr. Use R = 109737.12 ev. A. and At. find the semiaxes of the Sommerfeld elliptic orbits in terms of k and n. Cu. F. Xe. C1.
drops into the smallest orbit permitted by quantum theory? + . (a) Show that the energy levels of positronium are analogous to those of hydrogen but that their magnitudes are smaller by the factor +[1 (m/MH)]. 4 X lo5 cm/sec? (b) a rifle bullet weighing 2 . initially a t rest infinitely far from the proton. (a) How do the mesonic orbits compare with those of the corresponding electronic orbits? What is the radius of the mesonic orbit for n = l ? (b) What is the wavelength of the electromagnetic radiation emitted when a negative xmeson. called positronium. What relationship exists between the emission spectrum of singly ionized helium and that of atomic hydrogen? What are the wavelengths of the first four lines of the series which includes transitions to the state n = 2 of singly ionized helium? How does the spectrum of doubly ionized lithium compare with that of hydrogen? What is the radius of the first Bohr orbit of doubly ionized lithium? 15. its charge is the same as that of the electron and its mass is 273 times that of the electron. A positron and an electron can form a system. with a very short. What is the de Broglie wavelength of (a) a hydrogen molecule moving a t a speed of 2 . with the same mass as that of the electron and with a charge equal in magnitude but opposite in sign. (c) What is the ground state energy of positronium in electron volts? What is the radius of the first Bohr orbit? 16. (b) Show that the radii of the Bohr orbits of positronium are larger than those of hydrogen by a factor which is the reciprocal of the factor of part (a). It is known that a particle exists. but measurable. A proton and a negative *meson can form shortlived stationary states analogous to those of the hydrogen atom. A particle called a negative xmeson has been shown to exist. 0 gm. analogous to hydrogen. 7 13.180 ATOMIC SPECTRA AND ATOMIC STRUCTURE [CHAP. and moving a t a speed of 400 m/sec? 14. called the positron. halflife.
Part II The Nucleus .
.
and &rays. 81 The protonelectron hypothesis of the constitution of the nucleus. (2) radioactivity. The density of matter in the nucleus offers a clue to the source of the difficulty. There is now available a large amount of experimental information about nuclei. A density of this magnitude is inconceivably large. In addition. it was shown that the application of quantum theory to the nuclear model of the atom led to the development of a satisfactory theory for those properties of atoms that depend on the extranuclear electrons. and it is clear that in the atomic nucleus.CHAPTER 8 THE CORSTITUTIOR OF THE NUCLEUS I n the last chapter. the mass of one of the lighter atoms is about grn. such as spin and magnetic moment. As early as 1816. some new and unfamiliar properties of the nucleus will be considered. the problem of the constitution of the nucleus will be discussed. (5) the direct measurement of certain nuclear properties.on the basis of the small number of atomic weights then known. natural and artificial. (3) the artificial transmutation of nuclei by bombardment with particles from radioactive substances or with highspeed particles produced by laboratory methods. Now. the interpretation of the nuclear properties of atoms in terms of a theory of nuclear structure presents great problems. and some ideas will be developed which will be useful in the interpretation of experimental data in the following chapters. so that the volume of the nucleus is of the order of cm3 or less. that they might be integral multiples of the atomic 183 . led to the idea that atoms are built up of elementary constituents. The main problems of nuclear physics are the collection and correlation of the experimental facts and their interpretation in terms of a theory or model of the nucleus. derived from work in several fields: (1) the precise measurement of the masses of atoms. Consequently. The development of a theory of the nucleus is a more difficult problem. (4) optical spectroscopy in the visible and ultraviolet regions. with the result that the density of the nucleus is a t least 1 0 " gm/cm3. Prout suggested that all atomic weights are whole numbers. and it is almost all concentrated in the nucleus. The fact that certain radioactive atoms emit a. both of which are corpuscular in nature. The work of Rutherford and his colleagues on the scattering of aparticles showed that the atomic nucleus has a radius of 10l2 to 10l3 cm. In this chapter. matter is put together in a way which may not be amenable to ordinary experimental and theoretical methods of analysis.
respectively. and 208. as shown in Chapter 1. The nuclei of these isotopes. The combination of the whole number rule and the special properties of the hydrogen nucleus led to the &ump .54).6% have the greater mass. Most elements are mixtures of isotopes. 212. and the fractional atomic weights determined by chemical methods are caused by the presence of two or more isotopes each of which has a nearly integral atomic weight. I t was found that there are atomic species which have different masses in spite of the fact that they belong to the same element and have the same atomic number and chemical properties. 207. Analogous results were obtained for copper. According to this rule. so many elements have atomic weights which are very close to whole numbers that there seemed to be some basis for Prout's hypothesis. four radioactive isotopes of lead were discovered with atomic weights of 214. It was found that thii is indeed the case. 75.98. The proof of the existence of isotopes in the radioactive elements led to experiments to test whether some of the ordinary elements also consist of a mixture of isotopes. 8 weight of hydrogen. the different species belonging to the same element are called isotopes. Nevertheless.4% of the chlorine atoms have the smaller mass. as found in nature. The idea that all elements are built up from one basic substance received new support during the early years of the 20th century when the study of the radioactive elements led to the discovery of isotopes. and in all the work of thii kind the lightest positively charged particle that was ever found had the same mass as the hydrogen atom. but of opposite sign. Chlorine. 211. The fact that the atomic masses of the isotopes of an element are close to whole numbers led Aston to formulate his whole number &.184 THE CONSTITUTION OF THE NUCLEUS [c. and that all elements might be built up of hydrogen. This particle is evidently the nucleus of a hydrogen atom and.46 of chlorine. and consequently their spectra have the same general structure. For example. as well as three nonradioactive isotopes with atomic weights of 206. which is really a modified form of Prout's hypothesis. those of chlorine (35. and the atomic masses of the isotopes are very close to whole numbers. they are distinguished from one another by their different atomic masses. varying in mass from 206 to 214. and 210.98 and 36. Prout's hypothesis was discarded when it was found that the atomic weights of some elements are fractional. as for example. Much of the experimental work on isotopes involved the analysis of the positive rays from different substances. show a wide range of stability as measured by the extent of their radioactivity. has a mass very close to one atomic mass unit. although all have the same charge. and carried one positive charge equal in magnitude to the electronic charge. The different isotopes of an element have the same number and arrangement of extranuclear electrons.46) and copper (63. while 24. all atomic weights are very close to integers. and this distribution explains the atomic weight 35. has two isotopes with atomic weights of 34.
and the hydrogen nucleus was given the name protun to indicate its importance as a fundamental constituent of all atoms. The aparticles could exist as such. failure of the protonelectron hypothesis. with which is associated a magnetic moment. it was found that each of these components is split into a number of lines lying extremely close together. was the result of the detailed study of spectral lines. I t was also reasonable to assume that aparticles could be formed in the nucleus by the combination of four protons and two electrons. holding to an accuracy of about 1 part in 1000. When individual components of multiplet lines were examined with spectral apparatus of the highest possible resolution. but would make the charge equal to Z. these would contribute a negligible amount to the mass of the nucleus. or spin. The interpretation of certain generalizations about radioactivity in terms of the nuclear atom showed that both the a. To get around this difficulty.Z electrons surrounded by Z extranuclear electrons. nearly the same as the atomic weight and not equal to the atomic number Z. the charge on the nucleus would be equal to A. it eventually led t o contradictions and had to be abandoned. + . To account for the mass of a nucleus whose atomic weight is very close to the integer A. One of the failures of the hypothesis was associated with a hitherto unknown property of the nucleus. as required. The most precise experiments show that there are small but systematic departures from this rule over the whole range of elements. this further . and the presence of electrons in the nucleus made it seem reasonable that under the appropriate conditions one of them might be ejected. 82 The angular momentum of the nucleus. The whole number rule is actually an approximation. it was necessary to assume that the nucleus contained A protons. it was assumed that in addition to the protons.821 THE ANGULAR MOMENTUM OF THE NUCLEUS 185 tion that atomic nuclei are built up of hydrogen nuclei. The protonelectron hypothesis of the nucleus seemed to be consistent with the emission of aand @particles by the atoms of radioactive elements. I t was thus possible to consider the atom as consisting of a nucleus of A protons and A . of the atomic weight. The discovery that the atomic nucleus has an angular momentum.and @particleswere ejected from the nuclei of the atoms undergoing transformation. atomic nuclei contained A . The number A is called the mass number and is the integer closest to the atomic weight. Although the hypothesis that nuclei are built up of protons and electrons had some satisfactory aspects.Z electrons. which is half. I t will be shown in the next chapter that these variations are of great importance in adding to our knowledge of the structure of nuclei. But if this were the case. or they might be formed at the instant of emission. or less. the angular momentum.
I t was shown in Chapter 7 that because of the simultaneous motion of nucleus and electron around the common center of gravity. I. which involve radiofrequency spectroscopy. therefore. if an element has more than one isotope. Hence. The multiplicity and the relative spacings of the lines can be derived theoretically and depend on the magnitudes of the nuclear angular momentum and magnetic moment. each isotope has a slightly different value of the Rydberg constant. Newer methods. The two magnetic moments interact. microwave spectroscopy. that the nucleus has an angular momentum. In many cases. as Pauli did in 1924. of magnitude which defines the greatest possible component of I along a specified axis. A magnetic moment is associated with the angular momentum of the nucleus just as one is associated with the angular momentum of an electron. The properties associated with the hyperfine structure are the mass and angular momentum of the nucleus. and a large body of experimental information has been built up concerning these nuclear properties. a splitting of the atomic levels. The nuclear anguler momentum has quantum mechanical properties analogous to those of the angular momentum of the electron. such as bismuth. which gives rise to the hyperfine structure of the lines of the atomic spectrum. however. and the theoretical predictions have been confirmed. the hyperfine structures can be accounted for quantitatively if it is assumed. I t is a h/27r. and it was necessary t s assume. This effect has been found experimentally. or the deflection of molecular beams in magnetic fields. and the interaction energy perturbs the total energy of the electrons. the isotope effect is not enough to explain the hyperfine structure because the number of components is greater than the number of isotopes. 8 splitting is called hyperfine stmture. where I is the quantum number vector. there is. The nuclear angular momentum can then be deduced from the experimentally determined multiplicity and relative spacings. The hyperfine structure could not be accounted for in terms of the extranuclear electrons. as for the extranuclear electrons. is only about 2 cm' or less. and corresponding spectral lines of different isotopes have slightly different wave numbers. in units of wave number. according to the rule h I z = I 2. (81) * The value of I has been found experimentally t o depend on the mass . The total splitting. In these cases. have been developed for measuring the nuclear magnetic moment. Elements which have only one isotope.186 THE CONSTITUTION OF THE NUCLEUS [CHAP. that it is related to properties of the atomic nucleus. the Rydberg constant of an element depends on the mass of the atomic nucleus. also show hyperfine structure.
. an odd halfintegral multiple of h/2r. 1. 2. I is an integer or zero. The contribution of the protons to the angular momentum should be an integral multiple of h/27r1 whether the angular momentum of a proton is an integral or odd halfintegral multiple of h/2r. Nitrogen has an atomic number of 7 and a mass number of 14. this quantity is called the Bohr mugneton and has the value 0. Measurements of the magnetic moments of many nuclei have given values which are only about 1/1000 of the value of the magnetic moment of the electron. in contradiction to the value predicted by the hypothesis. if A is odd.. An electron has spin 3(h/27r) so that its total angular momentum is always an odd halfintegral multiple of h/2r (see Section 79).821 THE ANGULAR MOMENTUM OF THE NUCLEUS 187  . the . I has an odd halfintegral value ( h a l f of an odd integer). Experiments show that the proton. 3. the isotopes of cadmium ( 2 = 48). 3. Thus. they have been found experimentally to be odd halfintegers. . like the electron. The angular momenta of these nuclei should. The contribution of 7 electrons is. All measured nuclear magnetic moments are of the order erg/gauss. if A is odd. and its nucleus would have 14 protons and 7 electrons under this hypothesis. . and their values can be expressed appropriately in terms of the quantity in which the electron mass has been replaced by the proton mass. and it is possible with this additional information to predict the angular momenta of many other nuclei on the basis of the protonelectron hypothesis. and the hypothesis again predicts values which disagree with the experimental results. . I may have one of the values: 0. if the mass number of the nucleus is even. The rules just cited lead to one of the failures of the protonelectron hypothesis for the constitution of the nucleus. mercury ( 2 = 80). 3. I may have one of the values 3. if A is even. an integer. But the angular momentum of the nitrogen nucleus has been found experimentally to be I = 1. . The magnitude of the latter is where m is the mass of the electron. and the total angular momentum of the nitrogen nucleus should be an odd halfintegral multiple of h/27r. . . In other words.92 X erg/gauss. The protonelectron hypothesis also fails to account for the order of magnitude of nuclear magnetic moments. has an intrinsic spin of 3(h/27r). and lead (2= 82) with odd mass numbers should each have an odd number of electrons and an even number of particles all together. therefore. number A of the nucleus. therefore. be zero or integral.
a free electron confined within a space as small as the nucleus would have to have a kinetic energy of the order of 60 Mev. contain an odd number of electrons).3 X 1 0 . &Ap h. Section 78). which and can consequently be neglected. This value is much greater than the term mic4 = (9 X 10~')'(3 x 1010)3.v lo".3 X 1015. From Eqs. There is also a wavemechanical argument against the existence of free electrons in the nucleus. where mo is the rest mass of the electron. (84)  where & is the uncertainty in the position of a particle and Ap is the uncertainty in its momentum. therefore.Z is odd (and which would.0001 nuclear magnetons.999~. If it is now assumed that the momentum p of the electron is no larger than the values just found for the uncertainty Ap. and the proton has a magnetic moment of 2.7926 f 0. Measured values of the nuclear magnetic moment vary from zero to about 5 nuclear magnetons. and a velocity greater than 0. The fact that nuclear magnetic moments are only of the order of magnitude of the nuclear magneton is. According to this result.v 3. Relativistic formulas must be used because the electron moves very rapidly in the nucleus. as will be seen. which is assumed here to be 2 x lo" cm. however. 8 quantity PN is called the nuclear magneton. (636) and (638). Then From the uncertainty in the momentum. the electrons . we should expect to find nuclear magnetic moments of the order of magnitude of the Bohr magneton.Experimentally. then p . Then E 2 = EE lo' erg = lo' 6X 1. a t least in those nuclei for which A .' ~ ) ~ ( 3 X 1010)2= erg. According to the uncertainty principle (cf. and c is the velocity of light. it is possible to get a rough estimate of the energy of an electron in the nucleus. on the protonelectron hypothesis. The uncertainty & in the position of an electron is roughly the same as the diameter of the nucleus. and p2c2 = (3.6 X 1012   lo7 ev = 60 Mev. the total energy of a particle can be expressed in terms of the momentum. Suppoae that the principle is applied to an electron in the nucleus. another strong argument against the existence of electrons inside the nucleus.188 THE CONSTITUTION OF THE NUCLEUS [CHAP. I f electrons were present in the nucleus.
as in the case of the electron. F i y . The presence of a neutron.  This value is only slightly greater than the rest energy of the proton.5 X erg = 1. and in view of the large discrepancy it seems improbable that nuclei can contain free electrons. similar results are obtained from more rigorous calculations. Although the calculation is a rough one. Hence. the kinetic energy of a proton in the nucleus is of the order of a few Mev. would be very hard to detect. 83 Nuclear transmutation and the discovery of the neutron.6 X 1012 lo' = 9.was that an electron and a proton might be so claeely combined as to form a neutral particle. For the proton in the nucleus. that the momentum is of the order of the uncertainty in the momentum. and now the first term may be neglected in comparison with the second. which is 1. It was proposed. Ax is also about 2 X 10l2 cm.3 X lo'' ergsec/cm.831 NUCLEAR TRANSMUTATION 189 emitted by radioactive nuclei have never been found to have kinetic energies greater than about 4 Mev. which had been suggested by Rutherford as early as 1920. and this hypothetical particle was given the name neutron. and Ap . in 1932. therefore. One possibility. and many unsuccesdul attempts were made to find neutrons. then from Eq. I f it is assumed. all of the methods that have been discussed so far for detecting particles of nuclear size depend on effects of the particle's electric charge. as deflection in magnetic or electric fields and ionization.008 X 931 = 938 Mev. (85). and it should be possible for a free proton to be contained in the nucleus.4 X 10' ev = 940 Mev. Chadwick demonstrated the existence of neutrons. as one of the results of research on the disintegration or transmutation of nuclei by aparticles. . which haa no charge.67 X lo" gm) is nearly ! X M M times as great as the electron rest mass.v 1. Then E . that the electrons are bound to the positively charged particles and have no independent existence in the nucleus. This discovery opened up a vast field for further experimental work and led to the presently f the nucleus: that it is built of protons accepted idea of the constitution o and neutrons.v 3. The above argument does not apply to a proton in the nucleus because the proton mass (1. The failure of the protonelectron hypothesis of the nucleus was related to the properties of the free electron. or at least an order of magnitude smaller than that calculated from the uncertainty principle. Now.
or their isotopes. are very close to whole numbers led to the hypothesis that the atomic nucleus is a composite structure made up of an integral number of elementary particles with masses close to unity. helium. together with the extremely high penetrating power of the particles. and of the other rapidly moving atoms. produce no tracks in the cloud chamber and no ionization in the ionization chamber. for example. Since the new . It was thought that this radiation might be a form of 7ray of very high energy. 0. Bothe and Becker (1930) discovered that these elements emitted a highly penetrating radiation when so bombarded. I f so. This explanation corresponds to the formation from nitrogen (atomic number 7. They found that when nitrogen was bombarded with aparticles. These results could not be explained under the assumption that the new radiation consisted of highenergy 7rays. Chadwick finally (1932) proved that the energies of the protons ejected from hydrogenous materials. Closer investigation of the bombardment of boron and beryllium by aparticles gave some additional and unexpected results. Li. Cloudchamber studies of the process showed that the aparticle was captured by the nitrogen nucleus. and then the proton was ejected. i. show that the charge of the latter must be zero. 8 It has been mentioned that the fact that the masses of the atoms of all elements.190 THE CONSTITUTION OF THE NUCLEUS [CHAP. it should be possible by suitable means to break down the nucleus and to change one element into another. Curie and Joliot (1932) found that when the radiation was allowed to fall on substances containing hydrogen. lithium. a t least. Rutherford and his colleagues used energetic aparticles from radioactive substances as projectiles with which to bombard atomic nuclei in the attempt to produce artificial nuclear disintegrations. The protons were identified by magnetic deflection measurements. nitrogen. could only be explained on the view that the "rays" from bombarded beryllium actually consisted of particles with a mass close to that of the proton. He. and their energies were considerably greater than those of the bombarding aparticles. energetic protons were obtained. C. Radioactivity is a naturally occurring process of this kind. carbon. the oxygen isotope of mass 17. The bombardment with aparticles was found by Rutherford to cause emission of protons from all the elements of atomic number up to 19 with the exception of hydrogen. Be.. The most marked transmutation effects occurred with boron. and aluminum. of their energy was derived from the internal energy of the nucleus. and N. atomic weight 14) of an atom of atomic number 8 and atomic weight 17. and beryllium. They must have been shot out from the struck nucleus in such a way that part.e. it caused the production of highly energetic protons. These facts. unlike protons. Chadwick (1931) was also able to show that the rays emitted from bombarded beryllium gave rise to rapidly moving atoms when allowed to fall on other substances. These particles.
*  . I f both proton and neutron have spin ) then. the total number of elementary particles in the nucleus. the nucleus is transformed into a different one with one proton more and one neutron less. the atomic weight is therefore very close to a whole number. neutrons and protons. as well as the ~jroton. according to quantum theory. This hypothesis was used for the first time as the basis of a detailed theory of the nucleus by Heisenberg in 1932. the argument showing that protons can be contained within the nucleus is also valid for neutrons. will be an integral or halfintegral multiple of h / 2 r according to whether A is even or odd. it must be concluded that in pradioactivity. or it may be formed at the instant of emission. and both experimental and theoretical evidence offer strong support for this view.00758 amu. Later measurements have shown that the mass of the neutron is 1. protons and neutrons together. but not very different in magnitude. and an electron is emitted. led to the assumption that every atomic nucleus consists of protons and neutrons. it was identified with Rutherford's neutron. The value of the magnetic moment of the neutron is close to 2 nuclear magnetons.and the number of neutrons is A . the evidence is now convincing that the spin of the neutron is indeed ) h / 2 ~ . In Pradioactivity. It may exist as such in the nucleus. The discovery of a particle. The neutronproton hypothesis is consistent with the phenomena of radioactivity. the neutron. so that it is slightly heavier than the proton. 84 The protonneutron hypothesis. the latter possibility is now regarded as more likely.841 1 THE PROTONNEUTRON HYPOTHESIS 191 particle was found to be neutral and to have a mass close to unity. an electron. it is opposite in sign to that of the proton.has a halfintegral spin. This event is regarded arj the result of the change of a neutron within the nucleus into a proton. the resultant of the spins of A elementary particles. Finally. The empirical rule connecting mass number and nuclear angular momentum can be interpreted as showing that the neutron. is equal to the mass number A of the nucleus. and a new particle called a neutrino. then. Under the protonneutron hypothesis. The number of protons is given by the nuclear charge 2. The new nuclear model avoids the failures of the protonelectron hypothesis. This conclusion is in accord with all the existing observations of nuclear angular momenta.00898 amu. The values for both the proton and neutron are consistent with those measured for many different nuclei. since the mass of the neutron is very close to that of the proton. with a mass of 1.  . An aparticle can be formed by the combination of two protons and two neutrons. with an atomic weight very close to unity and without electric charge. the electron is created in the act of emission.2. Since there are several reasons why electrons cannot be present in the nucleus.
. they are shortrange forces. and the Coulomb repulsive forces responsible for the scattering of. aparticles predominate. Much of the research in nuclear physics is aimed a t the clarification of the laws of interaction between nuclear particles. and on the transformtion of one nuclear species into another by the rearrangement of its constituent nucleons. but it will be discussed in a later chapter. These attractive forces are the specifically nuclear forces between a proton and a neutron. there must be repulsive. and is sometimes referred to among physicists as nuclear chemistry. i.of the proton. if stable complex nuclei are to exist. The magnitude of the nuclear forces is such that the work required to divide a nucleus into its constituent particles (the binding energy of the nucleus) is very much greater than the work needed to separate an extranuclear electron from an atom. Whereas the latter is of the order of electron volts. quantitative way in this book because of its complexity. The neutron is a fundamental particle in the same sense that the proton is. 8  I t must be emphasized. . energy changes in the nucleus are of the order of millions of electron volts. or between two protons. the neutron is not regarded as a composite system formed by a proton and an electron. however. The two are sometimes called nucleons in order to indicate their function as the building blocks of nuclei.e. there must be attractive forces in the nucleus strong enough to overcome the repulsive forces. One of the main problems of nuclear physics is that of understanding the nature of the forces holding the protons and neutrons together. They seem to be more complex than the gravitational or electromagnetic forces of classical physics. Outside the nucleus. It is apparent from the small size and great density of the nucleus that these forces must be very large in comparison with the forces between the nucleus and the extranuclear electrons. Because of the positive charge . For the present. electrostatic forces between the protons tending t o push the nucleus apart. The information accumulated will be considered and interpreted in terms of the protonneutron hypothesis. they decrease very rapidly. it will suffice t o point out some of the qualitative properties of the forces between the protons and neutrons in the nucleus. Hence. This problem will not be treated in a detailed. between two neutrons. The nuclear attractive forces must be very strong a t distances of the order of the nuclear radius. and for the rise of the new field of nuclear engineering. that in considering the nuclei of different elements as being built of protons and neutrons. The emphasis in the following chapters will be rather on the facts about nuclei. I t is the magnitude of the energies associated with nuclear transformations that is responsible for the largescale applications of nuclear physics. This aspect of nuclear physics is analogous t o the application of atomic physics to the chemical properties of the elements.192 THE CONSTITUTION OF THE NUCLEUS [CHAP.   .
denoted by I. each nucleon may be pictured as having an angular momentum associated with orbital motion within the nucleus. These properties are important in the interpretation of many nuclear phenomena and in the theory of the nucleus. For nuclei containing more than one particle. Nevertheless. and an odd halfintegral multiple when A is odd. and i is therefore halfintegral. in agreement with experimental results. Each proton and each neutron in the nucleus has an angular momentum which may be pictured as being caused by the particle's spinning motion about an axis through its center of mass. According to quantum theory. as well as a brief discussion of some magnetic and electric properties of the nucleus. say the zaxis. is either 3h/27r or +h/2a. the resultant total angular momentum of the nucleus is then where L is the total orbital angular momentum. The component of the spin along a given direction. the orbital angular momentum is a vector whose greatest possible component in any given direction is an integral multiple of h/2a. They are. The spin of any single nucleon can add or subtract 3h/27r depending on its orientation with respect to the axis of reference. and the scalar quantity I is defined as the maximum possible component of I in any given direction. it is customary to write corresponding relationships between the momenta in capitals. and an odd halfintegral multiple if the number of particles is odd. and they occupy a prominent place in nuclear physics. and S is the total spin angular momentum. . I t will be convenient to include in this chapter a few more remarks about the angular momentum of the nucleus. S is an even halfintegral multiple of h/27r if the number of nuclear particles is even. some familiarity with the ideas and terminology of these matters will be helpful to the reader. The wavemechanical properties of an angular momentum of this kind are such that its orientation in space can be described by only two states: the spin axis is either "parallel" or "antiparallel" to any given direction. The orbital angular momentum L is an integral multiple of h/27r. among the less elementary aspects of the subject and will not be treated in detail in this book. Hence. The total angular momentum is actually a vector. In addition. I is an integral multiple of h/27r when A is even. however. with where 1 is the orbital angular momentum and s is the spin angular momentum. Each nucleon has a total angular momentum i about a given direction. 85 Magnetic and electric properties of the nucleus. The magnitude of this spin angular momentum is 3h/27r.
8 There are two possible sources of confusion which arise from careless usage. This quantity. This incorrect usage was introduced before the problem of the internal structure of nuclei had attained its present importance and has been continued. The magnetic moment of a nucleus can be represented as where 71 and g~ are defined by Eqs. is a measure of the deviation f a nucleus is imagined to be an of a nucleus from spherical symmetry. for example. Another property which is highly important in connection with the shape of the nucleus is the eleclric qucrdrupok moment. (83) and (88)and are called the nwlear gyromagnetie ratio and n w b w gfactor. the total orbital angular momentum of a nucleus is often denoted by 1 rather than by L. Significant conclusions as to the structure of the nucleus may be obtained from data on nuclear spin and magnetic moment. for a treatment of the methods and results. It has been found. and if the electric charge density is assumed . I ellipsoid of revolution whose diameter is 2b along the symmetry axis and 2a a t right angles to this axis. In addition. I t follows from Eq. and has the numerical value 5. The meaning of the term "spin" and the symbol 1 can usually be inferred without difficulty from the context in which they appear. which cannot be discussed in a simple way. have proved very useful in the study of nuclear structure and other properties. respectively. The quantity which measures the magnitude of PI. (83). and certain patterns have been observed in the experimental results. and this has been found experimentally to be the case. (89) that a socalled eveneven nucleus should have no magnetic moment. and is called the nuclear magnetic moment PI. The spins and magnetic moments of many nuclei have been measured. particularly for the ground state of the nucleus.194 THE CONSTITUTION OF THE NUCLEUS [CEAP. The term "spinn is often used for the total angular momentum of a nucleus rather than for the spin S alone.04929 X erg/gauss. The interested reader is referred to the works by Ramsey and Kopfennann listed a t the end of the chapter. is The details of the methods for measuring nuclear spins and magnetic moments will not be discussed in this book. T h i s generalization. PN is the nuclear magneton defined by Eq. just as knowledge of the arrangement of the extranuclear electrons was obtained from information about their angular momenta. that I = 0 for nuclei containing even numbers of protons and neutrons. and others.
deuterium. The FermiDirac statistics apply to systems of particles for which the wave function of the system is anti%ymmetrical. It follows from this property of the wave function that each completely specified quantum state can be occupied by only one particle. i. In addition to their electric and magnetic properties. nuclei have certain properties which are not obviously physical in nature. Thus.e. electrons. and two new forms of statistics have been devised. The question of which form of quantum statistics applies to a system of particles of a given kind is related to a particular property of the wave function which describes the system. Many nuclei have been found to have quadrupole moments. say of two protons in a nucleus. and neutrons obey . the quadrupole moment Q is given by Q = 3 Z(b2 . protons. while an isotope of lutecium cona Qvalue of $0. based on quantum mechanics rather than on classical mechanics. the extent of the deviation from spherical symmetry. it changes sign when all of the coordinates (three spatial and one spin) of two identical particles are interchanged. thus. the Pauli exclusion principle applies to particles obeying the FermiDirac statistics.a2). the reader ahould at least know of the existence and usefulness of the concepts involved. as can many other macroscopic properties of gases. whether it is + h / 2 r or 3h/2?r. The investigation taining 176 nucleons has a Qvalue of 7 X of quadrupole moments along with spins and magnetic moments has led to important developments in the theory of nuclear structure. in general. and the parity. the sign may be positive or negative. The properties which will be discussed very briefly are the statistic8 to which nuclei are subject. 86 Additional properties of atomic nuclei. that is. and atomic nuclei cannot.861 ADDITIONAL PROPERTIES OF ATOMIC NUCLEI 195 uniform throughout the ellipsoid. The properties of assemblies of photons. neutrons.00274 X cm2. It has been deduced from experiments that electrons. . and the magnitude of the charge. The concept of statistics in physics is related to the behavior of large numbers of particles. a nucleus with one proton and one neutron. has cm2. This property has to do with the effect on the wave function of interchanging all of the coordinates of two identical particles. the distribution of energies or velocities among the molecules of a gas can be described by the classical MaxwellBoltzmann statistics. A nucleon is described by a function of its three space coordinates and the value of its spin. These are the BoseEinstein statistics. and the FermiDirac statistics. be described on the basis of classical statistics. (&lo)  Its magnitude depends on the size of the nucleus.. as will be shown in Chapter 17. Although these properties will not be treated in detail. protons.
H. 1955. U.2 ) . 1953. CONDON McGrawHill. and odd parity when the sum is odd. When reflection changes the sign of the spatial part of the wave function. Two or more particles may be in the same quantum state. the wave function of a nucleus may be expressed as the product of a function of the space coordinates and a function depending only on the spin orientation. GENERAL R. the selection rules for many nuclear transitions involve conditions on the parity. Part 111. Part 9. eds. . Oxford: Clarendon Press. The Atomic Nucleus. Segrb. SCHAWLOW. The last property that will be mentioned is the parity. 8 the FermiDirac statistics. F. if L is even. A system of particles will have even parity when the sum of the numerical values for L for all its particles is even. In the BoseEinstein statistics. i. z system of axes. Although the parity seems to be an abstract sort of property. There is a direct correlation between the total angular momentum of a nucleus and its statistics: FermiDirac particles (odd A ) have total nuclear angular momenta which are odd halfintegral multiples of h/27r. TOWNES McGr~wHill. 6. This transformation of coordinates is equivalent to a reflection of the nucleus' position about the origin of the z.. 52. y." E z p e r i d Nuclear Physies. as well as on the total angular momentum. it does not change sign when all the coordinates of any pair of identical particles are interchanged. London: Pergamon Press. New York: McGrawHill. K. H. if L is odd. and A. Microwwe Spectroscopy. Nudecu Mmnenta (Engliih version prepared from the second German edition by E. p. SMITE.MdeetclOl Beams. L. The motion of a nucleus is said t o have wen parity if the spatial part of its wave function is unchanged when the space coordinates (z. while BoseEinstein particles (even A ) have momenta which are integral multiples of h/27r. It has been shown that the parity of a nucleus in a given state is related to the value of the orbital angular momentum L. as do all nuclei of odd mass number A . New York: E. E. N. F. F. Handbook of Physice.1955. 4. Schneider). 1957. Chapters 16. ODISEAW. To a good approximation. Part 7. New York: Wiley.z) are replaced by (z. D. the motion of the nucleus is said to have odd p a d y . "Nuclear Momenta and Statistics. EVANS. E.196 THE CONSTITUTION O r THE NUCLEUS [CHAP.y." Progress in Nueleat Physics. All nuclei having even mass number A obey the BoseEinstein statistics. New York: Academic Press. N. .. 1958. the parity is even. 5. Vol. RAYsEY. New York: C. ed.e.R. the parity is odd. Vol. . y. KOPFERMANN. by various authors. 1958. 0. Frisch. "Nuclear Momenta and Spins. and H. I. the wave function is symmetrical.  . RAMSEY. .. Chapters 2. Chapter 3. ed. 1956.
i. Radium. This element emits an aparticle. Consider what happens as a result of the radioactive decay of the element called uranium I (UI) with an atomic number of 92 and an atomic weight (or mass number) of 238 units. a shiit of two places to the left in the periodic table. . the emission of an aparticle causes a decrease of two in the atomic number. At the same time. suggested as a name for them the word isotopes.. the two elements when mixed could not be separated chemically even with extremely sensitive methods. and according to the displacement law. which was called uranium X1 (UXl). I t was found that generally the product of radioactive decay is itself radioactive. the charge of the nucleus of the radioactive atom decreases by two units. and that each of the decay products behaves chemically in a different manner from its immediate parent and its daughter product. The systematic study of the radioactive elements showed. forming the rare gas radon. the atomic mass decreases by four units. the product. The discovery of isotopes was one of the results of work on the radioactive elements. but the mass is practically unchanged. a metal with an atomic weight of 226.e. Rsdiurn in turn emits an aparticle. The chemical properties of ionium turned out to be similar t o those of thorium and. in fact. The emission of a Bparticle causes an increase of one in the atomic number. is known as the displacement law o f radioactivity. Each time an aparticle is emitted. was shown to be the product of the adisintegration of an element that was given the name ionium. This law. however. meaning the same place in the periodic table. that the atomic weight of ionium must be 230. Thus.. Each time a &particle is emitted. and Soddy. Uranium X1 emits a &particle giving the product uranium X 2 (UX2) with an atomic 197    . while that of thorium was known t o be 232. i. the nuclear charge increases by one unit because one negative charge is removed. ionium and thorium were found to be spectroscopically identical.CHAPTER 9 ISOTOPES 91 Natural radioactivity and isotopes.e. since the aparticle carries a positive charge of two units. Other pairs of elements were also discovered which were chemically identical but differed in atomic weight. The way in which isotopes arise in the radioactive elements can be understood in terms of the effect of radioactive decay ow the atomic number and atomic weight. a shift of one place to the right in the periodic table. first deduced by Soddy and Fajans in 1913. in 1913. Moreover. has an atomic number of 90 and a mass number of 234. which also is radioactive.
the cathode K has a very small hole through which the positive ions pass. J. After the existence of isotopes had been proven for radioactive elements. and has the mass number 232. precise measurements of this quantity w One form of apparatus used by J. they are isotopes. Hence. Now. Naturally occurring thorium also has the atomic number 90. The anode is an aluminum rod D. The pressure in the tube is adjusted so that the discharge potential is 30. Aston showed that most of the elements are mixtures. The gas t o be studied is allowed to leak in through a fine glass capillary tube L. The most important uses to which the mass spectrograph has been put by Aston and others is the identification of h t o p e s and the precise determination of their masses.000 volts. and after circulating through the apparatus is pumped off a t F. protoactinium ( 2 = 91). The cathode is kept cool during the discharge by means of the water jacket 3. and are called isobars. There are many other groups of isotopes among the radioactive elements. respectively. in turn. 9 number of 91 and a mass number of 234. They are chemically different elements with the same atomic weight or mass number. thorium. Hence. proved that isotopes also occur among ordinary elements. the mixture constitutes one element. UX2. UX2. UI and UII have the same atomic number 92. and UII. Thornson"' for the analysis of positive rays is shown schematically in Fig. and the importance of highly i l l become evident. and they will be discussed in greater detail in the next chapter. !J2 Positiveray analysis and the existence of isotopes. It follows that these two elements must be entirely identical in all their chemical properties. The elements UXI. with an atomic number of 92 and a maas number of 234.000 to 50. all have mass numbers of 234. UII emits an aparticle giving the product ionium. The discharge by which the positive rays are produced takes place in a large tube A. and UXI are also isotopes. who developed a new type of positiveray apparatus which he called a mass spectrograph. The positive ions emerge from the end of the cathode as a narrow beam of ions whose velocities depend on their charge and mass and on the dis   . The isotopic mass is one cf the directly measurable quantities essential to an understanding of the atomic nucleus. From a chemical point of view. and uranium (2 = 92). with an atomic number of 90 and a mass number of 230. by deflection experiments on positive rays. which are isotopes of thorium ( 2 = 90). He was also able to obtain fairly accurate measurements of the relative abundances of the isotopes of the different elements. 91. Thomson (1913). and in fact it is impossible to separate these two elements from each other by chemical means. ionium. having in some cases as many as nine or ten isotopes. emits a &particle giving the element uranium I1 (UII).198 ISOTOPES [CHAP. Thomson's work was extended and improved by Aston.
The beam is analyzed by causing it to pass between the pieces of soft iron P1 and P2. charge potential. P1 and Pa can be maintained at any desired potential difference.91. A positive ion moving from left to right in this spwe undergoes a deflection in the plane of the paper because of the electric field. 80 that in the space between them there can be parallel magnetic and electric fields. (a) Diagram o f the apparatus. and perpendicular to the plane of the paper because of the magnetic field. which are placed between the poles M and N of an electromagnet. P1 and Pa are the pole pieces of the magnet but are electrically insulated from M and N by insulating strips I.Ion Y  FIG. Thomeon's method for the d y e i e of positive rap. the deflected particle moves in the . Thua. After leaving the space between PI and P2. (b) Formation o f Thomeon's parabolas.
The relation between z and y will therefore be the same as that between d and y'. the beam of ions will strike the plate a t a point in line with the hole through the cathode called the undeflected spot . which consequently d e w a deflection d in the zdirection. let L be the length of the path through the fields. Equation (93) represents a parabola.the point where the particle strikes the plate will have coordinates y and z which are proportional to y' and d l respectively. Suppose that an ion of mass M.  where C is a constant which depends on the dimensions of the apparatus. I f the distance of the photographic plate from the fields is large compared with L. and assume that the latter are constant and terminate sharply. and it involves the chargebmass ratio of the ions in the beam. respectively. Ions with different values of q/M produce different parabolas. A soft iron shield W is placed between the magnet and the discharge tube to prevent the stray magnetic field from interfering with the discharge in the tube A. The electrostatic force on the ion is qE. ions with dilTerent values of the velocity v but with the same value of q/M should make a parabolic trace on the photographic plate. (41) and (92). For constant values of the fields E and H. the particle moves in a straight line. given by After pansing through the fields. formed a pattern of parabolas on a photographic plate. after passing through the fields. so that the ion is subjected to an acceleration qE/M in the direction of the field. as Thomson found. The latter relation is obtained by eliminating v from Eqs. the deflection y' from the xaxis is The magnetic field exerts a force Hqv on the particle. If there are no fields.fieldfree. Thomson found that the positive ions. From the usual formula for the distance traveled by a particle under a constant acceleration. and velocity u enters the space P I P 2along a path coincident with the zaxis of a system of rectangular coordinates with origin in the space P I P 2 . The time taken to pass through the field is L/v. evacuated "cameran C and falls on the fluorescent screen or photographic plate S. charge q. Let E and H be the electric and magnetic field strengths. The equation of one of these parabolas can easily be derived.
(93). O+. 92(a) is obrved. the latter ratio can be meaaured on the plate. Positiveray parabolas obtained by the Thomson method. (a) Schematic diagram o f the arabolaa.421 POSITIVERAY ANALYSIS AND THE EXISTENCE OF ISOTOPES 201 FIG. T h e s e ions could then be used as standards from which the masses of ions p m f the positive rays consist of ducing other trscea could be determined. P a r a b o l a s corresponding to H+. (b) Actual parabolea obtained by Thomeon . Nea i than the upper. but with a m a s s M' greater than M . I a mixture of diierent ions.92. If ions are preaent with the same value of q. then M' can be determined in the following way. I accurately. By a systematic analysis of the relative positions of the various parabolic traces appearing on a series of plates. The magnetic field is reversed after half the time interval of the emosure of the plate.~ e f b Neaa. a Merent parabola B will be seen. 03. the relative intensities of the individual parab . CO+. and other ions with known masses were identified. giving the mass M' in terms of the standard atomic mass M . the midline Oy is also obtained in this way. a parabola like that denoted by A in Fig. The lower picturs had a longer exposure (1813). NeaO. it follows that M'/M = ( u d / b ~ )and ~ . I f the ions in a positiveray beam have the same value o f q/M but travel with different velocities. and brought out the weak Nez2trace that is barely visible in the upper picture. This parabola will be lower than A became the magnetic displacement is smaller for the f the mass M of the ions giving the parabola A is known heavier ions. From Eq. H%. Thomson was able to determine the origin of the traces. and a pair of curves C and D ia obtained which L e the mirror images o f A and B.
since (9 X 20) (1 X 22) = 20. He passed a sample of neon through a pipeclay tube.  . 10 + The suggestion that neon has two isotopes was so striking that Aston looked for further evidence that might bear on the problem. collected the portion of the gas which diffused through and allowed this to diffuse once more.2.15 and 20. Thoms~n'~' analyzed the positiveray parabolas formed when neon was used as the source of the ions [Fig. and Aston used this property of gases in an attempt to achieve a partial separation of the two constituents of neon. respectively. The experimental facts suggested that neon could exist in two forms which could not be distinguished chemically. The changes in intensity cormponded approximately to the degree of enrichment or depletion of the mass 20 constituent. After repeated diffusion and redifiusion. The fraction with the smaller atomic weight was supposed to contain more neon of atomic weight 20 than ordinary neon. The same two curves. but with m e r e n t atomic weights. the other with atomic weight 22. a t a position corresponding to the atomic weight 20. it was difficult to avoid the conclusion that there are two isotopes of neon. The chemical atomic weight 20. A much leas intense trace was also found corresponding to the atomic weight ! 2 2 . An intense parabola was found. 9 olas compond to the relative abundance5 of the various ions in the mixture. The changes in the proportions of the two constituenta were large enough to produce appreciable changes in the relative intensity of the two traces (for the massea 20 and 22) in the positiveray photograph. I t was well known that a light gas di81uses through a porous partition more rapidly than a heavier gas. The atomic weight of neon had been determined as 20.20. calculated from their densities. No element of atomic weight 22 was known. Although the proof was not complete. were obtained with neon samples of different purity and for Merent conditions of discharge and gas pressure. and neon was the lightest element whoee atomic weight differed significantly from a whole number. the parabolas corresponding to masses differing by 10% could be clearly distinguished. With the apparatus that he used. and so on. as expected. one with atomic weight 20. the trace could not be identified with any known molecular ion. with the same d a t i v e intensities.20 would reault if there were present nine times as many neon atoms with atomic weight 20 as with atomic weight 22. !+2(b)]. of 20. the fraction with the greater atomic weight was suppoeed to contain more neon of atomic weight 22 than ordinary neon. he obtained from 100 cm3 of ordinary neon gas two extreme fractions of 2 to 3 cm3with atomic weights.28. In 1912.202 ISOTOPES [CHAP.
the amount of the deflection being greater the smaller the velocity of the ions.. therefore. indicated by the circle a t 0. To meet this need. the more slowly.e. The magnetic field causes deflections. A schematic diagram of Aston's first mass spectrograph is shown in Fig. ions having the same value of q / M but slightly different energies can be brought to a single focus on a photographic plate. The Fro. 93. Other ions with the same range of energies but a different value of q / M are brought to a focus at a different point on the photographic plate. The narrow beam contains particles with a wide range of velocities. 93. this diverging stream of ions enters a magnetic field. but it could not yield precise values of isotopic masses and abundances. The paths of the slowmoving ions. greater separation of ions of different masses) and all ions with a given value of q / M were brought to a focus instead of being spread out in a parabola. as shown in the diagram. (1919) designed the mass spectrograph. .931 ISOTOPIC MASSES AND ABUNDASCES 203  + ! 3 Isotopic masses and abundances :the mass spectrographand mass spectrometer. After passing through D. with the result that it is broadened as it passes through the field. By these means. The Thomson parabola method of analyzing positive rays was adequate for a general survey of masses and velocities. The further quantitative study of the constitution of the elements required the determination of isotopic masses with a precision of a t least one part in a thousand. greater sensitivity and precision were attained. intersect those of the fastermoving ions at some point F. An electric field between these plates causes a deflection of the ions toward P2. The positive rays from a discharge tube pass through two very narrow parallel slits S1 and S2 and enter the space between the metal plates P I and Pa.moving ions being deflected more than the faster ones. A group of these particles is selected by means of the relatively wide diaphragm D. Diagram o f Aston's first mass spectrograph (1920). perpendicular to the plane of the paper. If the instrument is properly designed. Aston's method of analysis was an improvement on that of Thomson in that greater dispersion was achieved (i.
two lines were seen.000. bzrt not quite. With this new instrument.204 ISOTOPES [CHAP. Because of this similarity.46. Jordon.0. Mattauch. the series of lines is called a mass spectrum and the apparatus a mass spectrograph. Later instruments designed and built by Aston. the positions of the lines caused by the masses in question are compared with the positions of the lines caused by standard substances whose masses are accurately known. With this instrument. respect i ~ e l ~ . and others have yielded isotopic masses with accuracies approaching 1 part in 100. This result showed that chlorine has two isotopes of nearly integral atomic mass. Dempster. Bainbridge.0 and 37.000. and the result of an analysis is a series of lines reminiscent of an optical line spectrum.46) of chlorine. as well as their abundances. Some typical mass spectra are shown in Fig. the former being the more intense line. 9 focus for a particular value of q/M is actually a line.0 and 37. for example. when the maw of oxygen is taken as 16. 9 4 . and in his second in~trument'~ isotopic ' masses could be determined with an accuracy of 1 part in 10. and ordinary chlorine is a mixture of these two kinds of atoms in such proportions that the chemical atomic weight is 35. the isotopes of chlorine were found to have the masses 34. The newer spectrometers yield high dispersion. . and showed that the masses of atoms are very nearly. Mass spectra obtained by Aston with his first spectrograph (1920). Modem mass spectroscopic measurements are based on the massdoublet technique in which the quantity actually determined is the difference in mass between two ions of the same mass number but having slightly different masses.980 rather than 35. Instead. In one method.integers. 'He ~ ' also showed that in the mass spectrum of chlorine there was no line corresponding to the chemical atomic weight (35. 9 4 . respectively. corresponding very closely to the masses 35. Aston was able to prove beyond a doubt that neon has two isotopes with masses very close to the integers 20 and 22. FIG.983 and 36.0. Aston improved the design of the mass spectrograph. For example. Aston determined the masses of the isotopes of a large number of elements. The isotopic masses can be determined quantitatively in several different ways.
431 ISOTOPIC MASSES AND ABUNDANCES 205 : . Authoritative compilations of atomic masses usually combine results obtained with both method^. One method is to measure the following massdoublet differences: (016)2 . that is. the carbon isotope of mass number 12. whemlls the precision which can be obtained with the mass spectrograph decreases with increasing mass. The mass of one of the memben of the doublet must be known accurately. the hydrogen isotope of mass number 1. and the result is 0 1 6= 16 atomic  More complicated cycles involving a larger number of atoms can also be used."^^^' Other techniques have also been developed'22' which will only be mentioned here. and highly precise results can be obtained. Hydrocarbon compounds (which give molecular ions) are used as sources of reference masses because of the relative ease with which fragments of almost any mass number can be obtained for comparison with other ions of the same mass number. ~ values ' of atomic masses determined by this method. The masses of C12. values of massdoublet differences have been ~ o m ~ i l e d . are used as secondary standards.5 3 2 = a. This time is proportional to the mass of the ion.'&"' It will be seen in Chapter 11 that information from nuclear reactions can also be used to determine atomic masses with precision comparable to that obtained with mass spectroscopic methods. with the result that the precision is practically constant for all masses. the distance between the two lines of a doublet are relatively large.~ ~ ' the time needed for ions to describe a number of revolutions in a uniform magnetic field. with mass units (exactly). The three equations are solved simultaneously. Another device which depends on the angular . with 0'" the primary standard. The chrondron of G o ~ d s m i t ' ~ ~ measures . and H1. The masses of C12 and H' relative to that o f 016can be determined in a number of different ways. 'and ~ . and differences in mass can be measured with high precision.
it is essential to know not only the isotopic masses.95. The mass spectrograph can be used for making abundance measurements. a mass spectrograph is inconvenient because a photographic plate is used to record the different isotopic ions. Msae apectrum of potassium (Dempster. but also the relative numbers of atoms of each isotope of an element. 1922).90. and indeed many of the isotopes now known were first discovered and their abundances measured by kston by means of this instrument. In the systematic study of isotopes. Diagram of Dempter's ma= spectrometer. however.'"' which has given good results. motion of ions in a known magnetic induction is the mczss qpzhtomekr. Atomic weight FIQ.FIG. and the procedure of determining abun . When used for abundance measurements. Microwave spectrosc~py'~" has also been used successfully for the measurement of atomic masses.
A slit S1in the plate C allows a narrow bundle of ions to pass into the region of the magnetic field H. where M is the mass of the positive ion and v its velocity on emerging from the slit in C. Dempster'28' built an instrument which was basically simpler and which was well suited for making abundance measurements.97. . Ions of the element to be analyzed are formed by heating a salt of the element. A schematic diagram of one of Dempster's spectrometer models is shown in Fig. About the time of the development of the mass spectrograph by Aston.* 931 ISOTOPIC MASSES AND ABUNDANCES 207 r  dances from the plate traces is both more tedious and less reliable than the direct measurements made with a somewhat simpler instrument. Upon emerging from the source A. consequently. Mass spectrum of mercury (Nier. the ions are accelerated through a potential difference V of about 1000 volts by an electric field maintained in the region between A and C. or by bombarding it with electrons. This energy may also be represented by 3Mv2. 95. the mass spectrometer. although it could not be used for making accurate mass measurements. 1937). In passing from A to C. If the magnetic field is perpendicular to the plane of the paper. the positive ions carrying a charge q acquire energy equal to qV. the ions Atomic mass units FIG. I t was called a mass spectrometer because the ion current was measured electrically rather than recorded on a photographic plate.
A typical curve obtained by Dempster FIG. Since the radius of the circle must have a certain definite value in order that the ions enter the slit S2 and be detected by an electrometer a t P. the current can be plotted against the atomic weight. By varying the potential difference V. ions with different values of q/M are. will be received for a given combination of accelerating potential and magnetic field. since each accelerating potential corresponds to a definite mass of particle reaching the electrometer. The current recorded by the electrometer is proportional to the number of positive ions reaching it per unit time and. 9 will be forced to move along a circular path defined by the relationship where R is the radius of the circle. Schematic diagram of a Nier mass spectrometer (Xier(30)). say q/M1. made to pass through the second slit Sz to the collector plate P. it is clear that ions with only one particular value of q/M. 98. . in turn.208 ISOTOPES [CHAP.
The beam of ions passes into a wedgeshaped magnetic field in which they suffer a deflection of 60")rather than 180" as in the Dempster spectrometer. In recent years. An element like beryllium or aluminum.e. the term nuclide has been widely accepted for a species of atom characterized by the constitution of its nucleus.'29' which had extremely high resolving power and sensitivity and was especially designed for searching for rare isotopes and measuring relative abundances.. i. rather than a single stable isotope. and the isotope abundances are closely proportional to the magnitude of the peaks of the positive ion current. Dempster obtained 18: 1. An isotope is then one of a group of two or more nuclides having the same number of protons or. Before discussing these results. in other words. the former being much more abundant. 9 4 The stable isotopes of the elements and their percentage abundances. in the symbol zSA. Ions are produced by the electron bombardment of the gas under investigation. owing largely to improved vacuum techniques and the development of new methods of electrical measurement. for the ratio of the abundances. A nuclide is usually indicated by the chemical symbol with a subscript a t the lower left giving the atomic number. since the word isotope implies more than one species occupying the same place in the periodic system.  . every radioactive species is a radioactive nuclide or radionuclide. of which only one species exists in nature. One of the most useful instruments was that of Nier. The figure shows a "mass spectrogram" of mercury.941  THE STABLE ISOTOPES OF THE ELEMESTS 209 *  for potassium with a slightly modified apparatus is shown in Fig. In the diagram the ion beam is shown broken into two parts which fall on separate collectors and are measured by separate amplifiers. and are accelerated by a potential drop of about 1000 volts. and a superscript at the upper right giving the mass number.'31+3233' The results obtained for the isotopes of 83 elements are collected in Table 91. by the numbers of protons and neutrons it contains.  . There are two isotopes with atomic weights 39 and 41 units. Z is the atomic number. Similarly. mass spectrometry has advanced rapidly. This deflection makes it possible to obtain high resolution with a simple magnet. It was a modification of the Dempster apparatus and gave results like those shown in Fig. 98. some remarks about terminology are in order. During recent years. is said to form a single stable nuclide. This instrument has been used to get highly precise isotopic abundances. 96. The mass spectra of the elements have been investigated in detail and the isotopic composition of the elements has been determined. Thus. A more recent form of spectrometer designed by h'ie~'~O' is shown in Fig. 97. the atomic species listed in Table 91 may be referred to as the naturally occurring stable nuclides. having the same atomic number.
4519.759 0. A 1 Relative abundance.204 100 90.892 1.634 0. 9 5ym Atomic lumber. bundance.210 ISOTOPES [CHAP.92 0.366 99.55 98.3681. Z 1 bol Masa lumber.9861 1.037 0.64 lO. Z 13 14 A 1 Si M e s a Relative >umber.8.13 11.48 100 P 15 16 31 32 33 34 36 35 37 36 38 40 39 40 41 40 42 43 44 S L i Be B 3 4 5 C1 .52 92.17 Sc 17 A 18 C N 6 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 K 19 7 Ca 20 0 8 F Ne 9 10 46 48 21 22 45 46 47 48 49 50 Ti Na Mg 1 1 12 v 23 50 51 .108 99.70 10.257 8.01390.82 100 78. A % 27 28 29 30 D He 1 2 2 3 4 6 7 9 10 1 1 106104 100 7. sym H % M.0151 Y bol Atomic aumber.984099.
6 20.23 1.40 2.58 23.55 56.54 7.02 7.354 2.46 11.15 27.04 15.33 100 67. Z 24 Mass umber.54 49.56 9.11 17.87 9.52 27.90 17.84 9.85 0. % 50.51 1.76 100 0.38 100 5. Z 35 Mass umber.37 72.1 30.11 18.63 5.80 100 15.86 7.78 9.66 2.72 16.53 9.52 49.61 18.89 27.02 82.19 0. A 50 52 53 54 55 54 56 57 58 59 58 60 61 62 64 63 65 64 66 67 68 70 69 71 70 72 73 74 76 75 74 76 77 78 80 82 Relative rbundance.07 31.62 17.76 36.81 4. A 79 81 78 Relative bundance.72 12.08 69.19 bol Stomic umber.88 26.THE STABLE ISOTOPES OF THE ELEMENTS TABLE 41 (Continued) hbol Stomic umber.55 2.4 39.56 100 51.31 83.82 9.87 12.23 17.56 0.56 11.46 0.46 23.27 11.66 1.9 48.58  Br Kr 36 80 82 83 84 86 Mn Fe 25 26 Rb 37 co Ni 27 85 87 84 86 87 88 89 28 Sr 38 Cu Zn 29 30 Y 39 40 Zr 90 91 92 94 96 Nb Mo 41 93 92 94 95 96 97 98 100 96 98 99 100 101 102 104 Ga Ge 31 32 42 As Se 33 34 Ru 44 (Continued) .76 9.82 91.43 7.19 3.62 60. 5 y m  cr % 4.
(Continued) Atomic Mass Relative number. abundance Z A % Rh % 3ymbol 45 46 103 102 104 105 106 108 110 107 109 106 108 110 111 112 113 114 116 113 115 112 114 115 116 117 118 119 120 122 124 121 123 120 Te 52 Pd 126 128 130 I 53 54 i n 124 126 128 129 130 131 132 134 136 133 130 132 134 135 136 137 138 138 139 136 138 140 142 141 142 143 144 145 146 Xe Ag 47 48 Cd Ca 55 Ba 56 In 49 50 Sn La Ce 57 58 Sb Te 51 52 Pr Nd 59 60 i n 123 124 125  . Z A Relative abundance. number. number.ISOTOPES TABLE 91 SFbol Atomic Maas number.
018 1.18 100 0.3 62.41 37. abundance.15 14.72 a.73 97. 1 26.number.84 12.68 24.64 13.50 27.87 21.64 28.18 5.41 22.75 35.71 47.THE STABLE ISOTOPES OF THE ELEMENTS TABLE 91 (Continued) sym bol Atomic Mass number.0127 0.24 100 0.40 2 .03 14. z A 62 144 147 148 149 150 152 154 151 153 152 154 155 156 157 158 160 159 156 158 160 161 162 163 164 165 162 164 166 167 168 170 169 168 170 171 Relative abundance. 6 0 0.14 13.18 0. 9 33.20 2.07 62.09 14.53 24.90 100 0.93 0.59 1. Sm % 3.07 14.135 26.47 15.88 25.7 0.135 3. 5 6 33.0 37.20 18.31 sym.8 25.83 7.3 7.21 (Continued) Lu Hf Eu 63 64 Gd Ta W Tb DY 65 66 Re 0s Ho 67 68 Er Ir Pt Tm Yb 69 70 .41 14.94 27.88 100 0.3 1 6 .13 31.97 11.78 3 2 .44 26.4 41. bol A % Atomic Mass Relative Yb z 172 173 174 I76 175 176 174 176 177 178 179 180 181 180 182 183 184 186 185 187 184 186 187 188 189 190 192 191 193 190 192 194 195 196 198 21.0524 0. number.73 20.0902 2.40 30.82 16. number.24 13.294 18.97 28.82 52.136 1 .
Hydrogen has two isotopes.with the occasionally used symbol D. has a mass about double that of the common isotope. In1''.2. . I t is therefore more convenient to include them with the stable elements than with the radioactive ones.3 100 100 0. all the others consist of two or more isotopes. z number. Not all of the nuclides listed in Table 91 are actually stable. however. number. Table 91 contains 284 stable nuclides divided among 83 elements. number. Sm14'. and the symbol for a nuclide is then shortened to SA. so that the difference in mass is as great as the mass of the ordinary hydrogen atom itself. This relationship between the masses is an exceptional one and. At least nine naturally occurring isotopes of "stable" elements show feeble radioactivity: K40. These nuclides are distinct from the families or chains of the heavy naturally occurring radionuclides and are much feebler in activity. Sometimes it is not necessary to show the number of protons explicitly. of course. equal to A . are single species.015%. This rare isotope. .ISOTOPES TABLE91 Atomic Mass A 197 196 198 199 (Concluded) Atomic Mass Relative number. With the exceptions noted. deuterium. abundance. . the differences between the properties of the two isotopes are more marked than in any other pair of isotopes. L u ' ~ Re1" ~.2739 79 80 200 201 202 204 81 203 205 A is the mass number. and S is the chemical symbol.0056 0. it is one of the ten isotopes of tin. The number of neutrons is. about onefourth in all. and PtlQO.7205 99. La'38. the symbol 5oSn120represents the nuclide with 50 protons and a mass number of 120. The hydrogen isotope of mass 2 has therefore been given its own name. Thus. Thorium and uranium are radioactive. but they occur in sufficient amounts and with sufficiently weak activity so that they can be handled in the same way as the stable elements. as a result. RbS7.6 52. Ndl4'. Twenty elements.6 22. z A % 1.48 23. the one with mass number 2 having a relative abundance of only about 0.
only beryllium has an even value o f 2. has three isotopes. The nuclides lsK40 and 7 1 L ~ 1 7 have 6 odd Z and even A but are weakly radioactive. Apparently the tin nucleus cannot contain less than 62 neutrons nor more than 74 neutrons and still remain stable. This property of the stable nuclides is shown in Fig. with the exception of H1and He3. hydrogen lithium. An examination of the values o f Z and A in Table 91 shows that in the stable nuclei.941 THE STABLE ISOTOPES OF THE ELEMENTS 215 I Odd Z Number 40 43 83 elements of I Odd A 53 57 110 I Even 4 8 I / Total 61 223 284  A v e " ~ number of isotopes 1. while z3V5' and 57La138 are very rare. while oxygen has three. and the only isotope with an even mass number is the weakly radioactive K40. while the other 19 have odd 2. tin with an atomic number o f 50 has neutron numbers from 62 to 74. while xenon has nine. The number of neutrons which can be included in a stable nucleus with a given number of protons is limited. and the mass numbers of the tin isotopes lie between 112 and 124. 5B10. The numbers of nuclides with the various combinations of even and odd atomic and mass numbers are listed in Table 92. and several elements have seven. Nuclides of even Z are much more numerous than those of odd Z. and nuclides of even A are much more numerous than those of odd A . cadmium and tellurium have eight each.2 3. For example. and 7N 14.Z is plotted against the number of protons. in which the number of neutrons A . There are some striking regularities in Table 91. boron. the only common exceptions being 1H2. Nearly all nuclides with even A have even 2 . Nineteen elements with odd Z have two isotopes apiece. Of the 20 elements which have only a single nuclide. potassium. 99. the number of neutrons is always greater than or equal to the number of protons. The four common elements which have odd values of Z and A. Tin has the greatest number o f isotopes. and each of these nuclides has odd A . ten. The elements which have more than two isotopes (apart from potassium) all have even values of Z . two of which are rare. There is always at least one neutron for each proton. One element with odd Z. For the other elements (except .5 5.4 Even Z Total 166 174    Carbon and nitrogen also have two isotopes. and nitrogen. 3Li6. have equal numbers of protons and neutrons. two o f these have odd mam numbers.
99.FIG. Neutronproton plot of the stable nuclides. .
The regularities that have just been noted are closely connected with the problem of nuclear stability and must eventually be accounted for in terms of the protonneutron theory of the nucleus. this corresponds to a value of 0. The relative abundances of H1 and H2 depend somewhat on the source. In nonradioactive ores. The properties of this nucleus yield information about the force between a proton and a neutron. and deuterium has a prominent place in theoretical nuclear physics. and it has been suggested that He3 can be formed as a result of the transformation of various nuclides in the air and on the ground by cosmicray bombardment. the abundance of Hea is approximately ten times as great in atmospheric helium as in well helium. In general. ice. some exceptions which. to cite the abundance ratios when measurements involving this standard are reported.0139 to 0. the He3 content varies widely. the relative content of C12 seems to be somewhat greater in plant material than in limestone. although interesting. This assumption is true. The range 0. The range of abundances of the boron isotopes shown in the table is equivalent to a variation in the B1'/BIO ratio from 4. neutrons.27 to 4. There are. electrons. Cosmic ray? are highly penetrating radiations which originate outside the earth and consist of protons. independent of the source of the sample that is measured. the hydrogen isotope of mass two (deuterium) is especially important because its nucleus consists of one proton and one neutron. in terms of the forces between nucleons.941 I THE STABLE IBOTOPES OF THE ELEMENTS 217    xenon) this range is smaller and the limits for the existence of stable nuclei are narrower. There are wide variations in the ratio of He3 and He4 abundances. and the value of the 0 1 6 / 0 1 8 ratio in nature has a spread of about 4%. The ratio of hydrogen to deuterium abundance has been determined as 6700 f 50 for samples of tap water from London.0151% includes the variation over a wide range of substances such as water. The abundance of the carbon isotopes cited in Table 91 are those found in limestone and correspond to a C'2/C13 ratio of 89.8. that is. it affects the use of certain boron standards in nuclear physics. Osaka. and various cities in the United States. and such helium is all He4. and other particles.0149% for the relative abundance of H2 in tap water.42 or about 3%. do not affect any of the conclusions previously drawn from the consideration of abundances. In the preceding discussion it was assumed that the relative abundance of an isotope is constant in nature. Although this variation seems small. and animal and mineral materials from many sources. organic compounds. it is necessary.2. Helium is formed in radioactive minerals because of alphadecay. however. in coal the ratio is 91. From this standpoint. photons. snow. for example. The values of . The relative abundances of the oxygen isotopes also vary. in general. and the spread in the values listed in Table 91 shows that the variation is small. in particular. therefore.
and these are usually associated with the radioactive sources from which the different lead samples are derived. which is a mixture of isotopes.16.004840. or else they do not seriously affect further work in nuclear physics.00000 is associated with ordinary atmospheric oxygen. 19. if the latter are to be made to more than five significant figures. most abundant isotope. either there are no significant variations.759% of the isotope of mass 16. The standard of mass used here is slightly different from that used for the chemical atomic weights. Some atomic masses(ls. Then Physical atomic weight .204% of the isotope of mass 18. 95 Atomic masses : packing fractions and binding energies. whereas on the mass spectrographic.000279. for the remaining elements. and this is the atomic weight of atmospheric oxygen on the physical scale. it is not possible to give exact isotopic abundances without specifying the source of the material.2% of the oxygen atoms. Even for common lead. The weighted mean of these values is 16.00000 on the chemical scale. the number 16. the most abundant of which has the mass number 16. atomic weight scale.OOOOOO . together with 0. The values used for the abundances of the oxygen isotopes affect the value of the factor for converting atomic weights from the physical scale to the chemical scale (see Section 95).004462.037% of the isotope of mass 17. this is taken as the isotopic weight of the single. as compared with the postulated value of 16. T h i s value is also correct for the oxygen from limestone.2 f 0.004529. The abundance variations which have been discussed are the most important ones known and. but for oxygen from water or iron ores the ratio may be 4% higher.00000 for the weight of the common isotope of oxygen. Chemical atomic weight l6.7. or physical.the abundances quoted in the table are those for atmospheric oxygen and correspond to a value of the 0''/018 ratio of 489. the isotopic composition of the oxygen used as a reference must be specified. and 0. Atmospheric oxygen consists of 99. Wide variations in the abundances of the lead isotopes are also found.00000. The relationship between the chemical and physical atomic weight scales may be determined in the following way. The other two isotopes together constitute only about 0. The variation in the value of the conversion factor may be large enough to affect the precision of atomic weight determinations and. In the latter case. In the determination of isotopic weights by the mass spectrograph it is the practice to take as the standard the value of 16. the values in the table are for Great Bear Lske galena. The weights so obtained differ slightly from those based on the ordinary chemical atomic weight scale.004462 = 1. I t is seen from Table 91 that oxygen actually has three isotopes.36) are given in Table 93.
0 1803.14 1020.75 160.06 8.68 7.0 6.55 8.3 0.50 8.7 5.70 8.5 1560.3 3.63 127.47 7.62 7.5 6.4 8.25 8.92 7.52 8.7 2. Mev lucleon.7 2.ATOMIC MASSES 219  Number of ieutrom.225 28.19 92.1 1 551.92 236.3 7.77 7.57 8.88 271.75 7.13 845.6 3.Z 1 0 1 2 4 5 6 6 7 7 8 9 10 10 10 14 14 16 16 18 20 22 20 30 34 42 50 56 66 70 76 82 90 104 110 118 124 142 146 Pecking laction.13 317.2 2.24 58.113 7.7 9.7 2.58 131.6 3.33 8.97 7.93 7.92 492.51 282.2 8.46 6.78 8.22 652.73 8.80 147.47 7.09 104.49 8.  Binding energy Per x 104 total.07 5.73 139.2 0.48 8.1 1473.1 1.4 1.8 4.1 0.52 8.3 5.7 6.15 76.7 26.58 .6 5.23 768.8 81.61 6.1 1419.14 97.2 1096.5 1237.0 7.03 8.45 8.8 5.4 0 2.39 8.44 8. A .3 5.62 224.01 7.0 4.4 1.71 341.75 8.4 73.0 1623.2 5.7 11.63 8.59 8. Mev 89.7 1768.74 298.73 988.00 343.2 6.79 8.3 5.0 16.1 5.88 7.29 39.6 1141.7 5.
The packing fractions. Aston expressed these divergences in the form of a quantity called the packing fraction defined by Packing fraction = a s s . I t is seen from Table 93 that the isotopic masses are indeed very close to whole numbers. +lo. with the exception of those for He4. isotopic weights obtained by means of the mass spectrograph must be divided by 1.000279 in order to convert the results to the chemical atomic weight scale. fall on or near the solid curve. For Ole. This precision is essential in the study of nuclear reactions and transformations. the packing fraction becomes negative. 9 Hence. as will be seen in later chapters. becoming positive again a t values of A of about 180. It . is expressed in units of lo' amu. The packing fraction was very useful in the study of isotopic mssses. (97) + Sample values of the packing fraction are listed in Table 93.220 ISOTOPES [CHAP. The atomic mass of a nuclide can be understood in terms of the manses of its constituent particles and a quantity called the binding energy. but it does not have a precise physical meaning. by definition. given in parentheses. C12. The conversion factor is used only when it is necessary to compare mass spectrographic results with weights obtained by chemical methods. and the variation of the packing fraction with mass number for a larger number of nuclides is shown graphically in Fig. Estimates of the errors of the msss spectrographic measurements have been included in order to indicate the high precision with which atomic masses can now be determined. the value is zero. in nuclear work the atomic masses on the physical scale are used. then M Z . As A increases. I t is also important in the precise calculation of packing fractions and binding energies. and A is the mass number. If the packing fraction is denoted by f. The error.Mass number Atomic m M a s s number is the actual weight of a nuclide on the physical atomic weight where M Z S A scale. I t seemed clear to the early workers in this field that the systematic study of the divergences of the masses of nuclides from whole numbers was an important problem. passes through a rather flat minimum and then rises gradually. The explanation for its usefulness will appear from the discussion of the binding energies of nuclei. as will now be shown.and Ole. The values are high for elements of low mass number. apart from the nuclides mentioned. A = A(1 f).
and electrons combine to form an atom. it is the amount of mass which would be converted to energy if a particular atom were to be assembled from the requisite numbers of protons. derived from the special theory of relativity. is called the ma98 defect. AM. neutrons. then the above principle states that an amount of energy equal to is released in the process. and electrons.ATOMIC MASSES 22 1 0 20 40 60 80 100 120 140 160 180 200 220 240 Mass number (A) f mass number. Fro. The difference in mass. neutrons. Packing fraction as a function o  might a t first be supposed that the mass of an atom should be the sum of the masses of its constituent particles. A survey of the atomic masses shows. however. To account for this difference in mass. that the atomic mass is less than the sum of its constituent particles in the free state. The same amount of energy would be . is used. If AM is the decrease in mass when a number of protons. 410. the principle of the equivalence of mass and energy.
the mass of the neutron.. The mass of the constituent particles is the sum of Z proton masses.0089830 mass units.2 ) . The mass defect can then be written AM = Z ~ H( A .Mz. The proton and electron masses can be combined into the mass of Z hydrogen atoms because the minute change in mass which may accompany the formation of a hydrogen atom from a proton and electron is negligible. msss of the hydrogen atom is 1. Then Since one atomic mass unit is equivalent to 931. and A . Some values of ..0081437mass unita and m. is 1.A.145 Mev.Zneutrons. l a la.. and the energy equivalent of the mass defect is therefore a measure of the binding energy of the nucleus.911. (99) + where m ~ the . la M ~ s number s la I& & A . Zelectrons. FIG. needed to break the atom into its constituent particles.a Binding energy per nucleon as a function o f mase number.%! A a s. the bin* energy of the nucleus is given by The average binding energy per nucleon is obtained by dividing the total biding energy of the nucleus by the mass number A.
i. One pound of an atomic species contains 453..u per pound.6/M gram atomic weights ( M is the atomic weight). C". The subject of atomic masses should not be left without a brief discussion of a proposal'34) to replace the mass standard. and the binding energy is still 8. as is shown clearly by the values in the last column of Table 93. This enormous value of the energy that would be needed to dissociate a nucleus into its constituent protons and neutrons is another indication of the magnitude of the nuclear forces. and reaches a maximum value of about 8. O".73 X 1 0 ' ~nucleons. In recent years. These matters will be discussed in detail in later chapters. and the masses and biding energies of nuclei yield useful information about the constitution and stability of nuclei.451 ATOMIC MASSES 223 . The unit "1 Mev/ nucleonn is then practically the same as 4. and 01'.4 Mev at about A = 140. 453. and a graph of the binding energy per nucleon as a function of the mass number is shown in Fig.8 Mev in the neighborhood of A = 50. as well as that of its application to the problems of nuclear physics. the binding energy obtained by the method just outlined are listed in the last two columns of Table 93.15 X 10'' Btu/pound. the value decreases to about 7. however. The validity of the relativistic massenergy relationship has been proven. so that one pound contains 2. the analysis of nuclear transformations has provided an independent means of determining atomic masses which is a t present even more powerful than the mass spectrographic method. as can be shown by converting from Mev per nucleon to more familiar units such as Bt. The application of the principle of the equivalence of mass and energy and the introduction of the concept of binding energy have. thus far in our treatment. are very small. One Mev is equivalent to 1.8 Mev/nucleon or 350 billion Btu/pound. The binding energy of a nucleus is. With the exception of He4. and 0'' are considerably greater than those of their neighbors.6 Mev at uranium. In addition. only a theoretical basis. the binding energy is about 8. The binding energies of some of the very light nuclides. such as H2. a quantity with real physical meaning. The magnitude of the binding energy is enormous. In the neighborhood of the maximum of the biding energy curve. many nuclear transmutations have been accurately studied and careful measurements have been made of the changes of mass and energy in these reactions. by c". For higher mass numbers.e. or X A / M nucleons. C". The binding energy per particle rises sharply.519 X lo'' Btu.023 X The mass number A and the atomic weight M are practically the same. therefore. The binding energies of He4. . 911. the values of the binding energy per particle lie on or close to a single curve.6 X 6. The maximum is quite flat.
doubly charged ions of heavier elements could be compared directly with singly charged ions of the type (C'2). this property would permit many more doublet comparisons to be made directly with the reference nuclide than are now possible.145 Mev. carbon forms many compounds with hydrogen. and 3. as compared with 1 awu (016) = 931. The values of binding energies are unchanged. Carbon forms many more chemical compounds which can provide molecular ions for use in a spectrometer than does oxygen. thus. for values of A between 120 and 240.. Throughout this book. providing easy reference lines for doublets with masses up to 120 units. respectively.00031792 amu (016). up to 10 or more.. Thus. so that C12 would be a convenient standard for atoms of low mass number. .2247 Mev when either mass standard is used. The conversion factor from mass to energy is 1 amu (C12) = 931. triply. Nuclide Ole Standard CI2 Standard 1 The relation between the atomic mass units on the two scales is 1 amu (C12) = 1. the binding energy of the deuteron is 2. No other element besides carbon forms molecular ions containing as many atoms of one kind. A table of atomic masses has been prepared'35' based on Ct2 as the standard and the following results have been obtained.0000000 units. and quadruply charged ions of CL2occur at integral mass numbers and can be paired in doublets with ions o f mass number 6. The use of C12 as the standard would have several advantages in mass spectrometry.441 Mev. but the reader should be aware of the possibility that the standard may be changed within the next few years.4. or (C12). we shall use 0 1 6 as the mass standard and the atomic masses given in Table 93. (H1). Doubly.to assign to C12 the atomic mass 12. and would yield masses with increased precision a t intermediate and large values of A.
Mod. Phys.18. Vol. E . QUISENBERRY.21. 717 (1951)." Phys." Phys. 29. 13. Rev. ton." Phil."Phys." Phys. "Atomic Masses in the Region about Mass 40 and from Titanium through Zinc. 3. 88. Philadephia: Blakis. 39. F. 14. E. Phys. Washington. National Academy of SciencesNational Research Council. SCOLMAN. NIER. 225 (1911). 2.86. "Atomic Masses of the Stable Isotopes. INGHRAM. QUISENBERRY. J. 102. Atomic Masses. snd NIER. 9. L. Mag. 209.C. 712 (1958). THOMSON. London: Longmans. 1942. R w . Relative Isotope Abundances of the Elements. Ma88 Spectra and Zsotopes. "Charged Particle Dynamics and Optics.463 (1954). H.. C12. ASTON. 107.E. op. S32. 1944.GIESEand J. Mag. 666 (1952)." Revs. Rev.. 449 (1920). COLLINS. W.. C. Vol. ref. New York: Academic Press. 10 5 4 30. R.408 (1952).20. 669 (1912).Mass Spectra and Isotopes. < . PARTICULAR    1 . HOQQ. DUCKWORTH. Rays of Positive Electricity and Their Application to Chemical Analyses. and BENSON. 2nd ed.. B E N ~ O N Manganese. gen. Soc. M. GIESE. DUCKWORTH. "Atomic Masses of H1. Table VI : Mass Differences and Ratios..HALSTEAD. QUISENBERRY. "Atomic . D.." Phil.GENERAL F. 24. Mag. J. STRANATHAN.. (London). Nuclear Data Project. M. London: Edward Arnold. 1076 (1956).. 84." Phys. W. Rev. 4. 8. A115. New York: Wiley..2nd ed. Chapter 4. 1948." Phys. W. F. 13. and NIER. "Atomic Masses of H1. and 11." Advances in Electronics. 102. 767 (1957). 110. "A New Mass Spectrograph and the Whole Number Rule. Roy. F.Chapter 5 K. 1065 (1956). Differences. cit. W. 657 (1908). F. D. SIR J. Mod. The "Pmtkles" of Modern Physics. Masses from Phosphorus Through 12. 26.ASTON. ASTON." Phys. 821 (1909). 7. G. R w . E. J. 1 . 561 (1907). D2. C12." Experimental Nuclear Physics. 752 (1910). 707 (1919). Proc. 102." 5.. I. 1953. F. 1921. Segrc?.and PENNINQTON. T."The Constitution of Atmospheric Neon.487 (1927)." Rws. and S32. 10. Table IV: Relative Isotopic Abundances. KETTNER. Rev. "Mass Spectroscopic Atomic Mass Differences 11.."Atomic Masses from C12 to Ne22. W. Green and Co. J. THOMSON.16. Phil.ed. SCOLMAN. BAINBRID~E. 1st ed. 38. "Modern Mass Spectroscopy. Rev. "Mass Spectroscopic Atomic Mass 6.: United States Atomic Enexgy Commission. 1913. ASTON. 1071 (1956). "Atomic Masses from Palladium through Xenon. 1664 (1957). "A Positive Ray Spectrograph. ASTON. Papers on positive rays and Isotopes. and JOHNSON. 1959 Nuclear Data Tables.
Berlin: Springer Verlag.. Vol. G. Table V. p. Government Printing Ofice. 20. Phys. 111. 18. London: Pergamon Press. "Masses of Atoms of A < 40. HINTENBERQER. HOQQ. "Determination of Atomic M~eees by Microwave Methods. 86. 617 (1952). Rcldioc Isdopic Abundanecs of the 31. Division of Mathematical and Physical Sciences of the National Reeearch Council. E. "Mass Spectrograph Studies of the Isotopes of Various Elements. London: Pergamon Press. HAYS. Rev. Rev. Nuclear Science Series. 33.: U. A.. 85.. 1957. Sci." Phys. NIER.824 (1951). pp. April 1959.and STAMFORD. "Some New Values of Atomic Masses." Phys. Berlin: Springer Verlag.. 1969 Nuclear Data TaMcs. E.ed. Rco. . 6688. and A. "Msss Measurementu with a Magnetic TimeofFlight Msss Spectrometer. DUCKWORTH. p. 20. London: 22. Frisch.and GOUDSYIT. 11. Rev. "Some New Atomic Mass Measurements and Remarks on the M a s Evidence for Magic Numbers. 1. National Academy of SciencesNational Research Council. Preliminary Report No.. Ameriean Institute of Phyaica Handbook. 38. WAPSTRA. 469 (1953). S. 25. GESCHWIND. STAMFORD. 233. L. S. K.. 0. 398 (1947). 18. Vol.. "A TimeofFlight Mass Spectrometer. Washington.. Rev. D. "Atomic Msases of Nuclides." Phya. p." Rco. 21. 622 (1948). Rev.." Phys. 38. 1958. and GOUDSMIT. BAINBRIDGE Elemmk.15. p.  27. Rev." Prograd in Nuclear Physics. Rev. 933 (1937). Nuclear Maaaes and Meir Determination.NIER. R. DEMPSTER. Part 1. H. 38. EVERLINQ. 6." Handbuch der Physik. DUCKWORTH." Handbuch der Phyaik. ed. HAYS.. Vol. 1958.468 (1951). New York: McGrawHi. 50.C. 316 (1918). 74. SMITH. 138."Masses of PbZoS. Vol. "A Mass Spectrometer for Isotope and Gas Analysis. A. 24. Calcium and Zinc.. 86.. 17. 82."Measurements of Light Msssea with the Maes Synchmmeter." Phys. 16. Part 1. Thz32.." Phys. 1606 (1958). A. 83. S. Washington. WAY. Principally in the Region of 82 Neutrons. GOUDSMIT. 9. 30. 0. 26. Rev.RICHARDS." Phys. 0. p. "Masses of A t o m of A > 40. 1957. RICHARDS. H. Znsfr. 25. MATTAUCH and F. and U238.and GEIGER. 630 (1952). H. 52.. A. DUCKWORTH. Physies. DUCKWORTH and R. NIER. ed. Pergamon Press (1957).. T. 6. 1065 (1952). Frisch.. 29. KEQLEY.S. 1114 (1951). U234.OLSON. "Masses of Lead and Bismuth. K. 0. J. "PositiveRay Analyeis of Potassium.11 Phys. 23. J. 631 (1922). 28. ed. Rev. 81. 85." Phys." Phys. H. R. 1041 (1936). 0. 1957. "A New Method of PositiveRay Analyeis." Progress in Nuclear 19. PRESTON.
see Table 111. Fl9. in grams and atomic mass units? (c) the mass number of the ion? 2. and is deflected into a circular path 18.0" = 0. a singly charged positive ion (q = 1. EVERLING). WAPSTRA. V in volts. I t then travels through a uniform magnetic field for which H = 1000 gauss. Suppose that. 35. KOHMAN. NeZO. 1431 (1958). Hz. in Problem 8.042298 amu C1?H1)r . Suppose that singly charged ions with masses close to 12 and 14 amu are accelerated through a potential difference of 1000 volts and then travel through a magnetic field of 900 gauss. and WAPSTRA. A 36. clides. H in gauss. "New Reference Nuclide. "Atomic Masses of Nu70. N13. Use the formula to check the result of Problem 1.0000. Plot the number of stable nuclides against the neutron number A .and Si2*. At what values of Z are maxima observed? 5. Rev.036390 amu Calculate the atomic masses of H1.000000. 9. C12 = 12.0015483 amu (H2)3 .000000 were taken as the mass standard. Compare the results with the average binding energy per nucleon. the following relation holds: MV = 4. Show that for a singly charged ion in a mass spectrometer. From the values of the atomic masses listed in Table 93. MATTAUCH. 342 (1960).. Suppose that Cia." Phys. H2 and C12. 529 (1960). M n r r m c ~ and .PROBLEMS 227 34. 3. BHANOT. "Atomic Masses in the Heavy Mass Region. At what values of A . and 0'81 . 235 (1960). < 1. and NIER. with Ole = 16. calculate the binding energy of the last proton in C12. and 2 1 1 is in atomic mass units relative to O l e = 16. 18. What would be the doublet values and the massea of H1." Science 127. 7. The following values of mass spectroscopic doublets were obtained. What is (a) the speed of the ion? (b) the maw of the ion. where R is expressed in centimeters. Plot the number of stable isotopes per element against the atomic number Z. KONIG).H2 = 0. 15. how much energy would be liberated in each case? (For the mass of N13. JOHNSON. (H1)z . N14. 120. Where should collector plates for the two different ions be located? 4.Z are masima observed? 6.826 X 106H2R2.) 8.602 X emu) is accelerated through a potential daerence of 1000 volts.Z. and 016 were formed by combining the a p propriate atoms with C12.2 cm in radius." Nuclear Physics.f C12 = 0. In a mass spectrometer. N14.
on the physical scale. (99). The following doublet values were obtained with O l e = 16.10. [Hint: Note that Eq.9 coul/gmmolecular weight. Starting with Eq.000000. The best value of the faraday is considered to be 96521. (99) may be written 12. show that the average binding energy per nucleon for all but the lightest elements is close to 8 Mev. What is the corresponding value on the chemical scale? . What values are obtained for the masses of C12 and S3'? 11.
1&1 The basis of the theory of radioactive disintegration. The investigation of the scattering of aparticles by atoms led to the idea of the nuclear atom. is an important source of information about the structure of the nucleus. The first problem which will be considered is that of the quantitative description of radioactive growth and decay. but they are also valid for the artificial radionuclides. The importance of radioactivity depends to a large extent on the ability to measure radioactive changes with high precision. which is fundamental to all of atomic theory. which deals with the identification and classification of these levels. and are fundamental to a large part of the work to be discussed in the remaining chapters of this book. They can be applied. or induced. The analysis of the chemical relationships between the various radioactive elements resulted in the discovery of isotopes. and this led in turn to the discovery of the neutron and to the current theory of the composition of the nucleus. This discovery of artificial. therefore. . Nuclear spectro. It has been seen that the emission of a.scqpy. The laws of radioactive change were developed from information about the natural radioelements. radioactivity by Joliot and Curie. . It will be shown in a later chapter that the transmuted atoms resulting from this kind of bombardment are often radioactive. A clue to the way in which one radio229 . and hundreds of radioactive nuclides have now been made by various methods. to any radioactive transformation. in 1934.and &particles by certain atoms gave rise to the idea that atoms are built up of smaller units. and to describe them quantitatively by means of a straightforward theory. The bombardment of atoms with swift aparticles from radioactive substances was found to cause the disintegration of atomic nuclei.CHAPTER 10 NATURAL RADIOACTMTY AHD THE LAWS OF RADIOACTIVE TRANSFORMATION Many of the ideas and techniques of atomic and nuclear physics are based on the properties of the radioactive elements and their radiations. and i t is impossible to conceive of nuclear physics as something separate from radioactivity. Thus. and to the concept of atomic structure. and the study and use of radioactivity are essential to nuclear physics. started a new line of research. radioactivity has been intimately connected with the development of nuclear physics. The investigation of the radiations from the natural and artificial radionuclides has shown that the nucleus has energy levels analogous to the atomic energy levels discussed in Chapter 7.
Becquerel.230 NATURAL RADIOACTIVITY [CHAP. The active substance was contained in the reaidue and was given the name . which contained practically all the uranium. 4 8 1 2 1 6 2 0 2 4 2 8 Time (days) The decay of thorium X activity and the recovery of thorium T i e (days) FIG.1&1. therefore. and Rutherford and Soddy. It appeared."' Crookes (1900) found that if a uranium salt was precipitated from solution by the addition of ammonium carbonate. activity. activity. The product obtained by evaporating the solution. The decay of uranium X activity and the recovery of uranium active substance is produced from another was provided by early experiments of Crookes. was found to be highly radioactive. a small residue was left. 10 " FIG. when removed from the solution. but by another substance which could be separated from the uranium. that most of the observed activity of compounds of uranium was not cauaed by that element. had very little activity. 102. and then redissolved in excess of the reagent. This residue.
measured by one of the methods discussed in Chapter 2. and Eq. while the thorium. .5 days. 102.(t) is the activity after a time t. The activity.e. and the main body of the thorium was purctically inactive. which was used only to distinguish between the X body and its parent substance. was recovering its activity. 101. and A is a constant. the activity could be expressed as a function of time by the equation where Azo is the initial activity of the ThX. Thorium X loses half of its activity in about 3. and obtained the curves shown in Fig. The recovery curve for the thorium was found to fit the formula in which the constant A has the same value as in Eq. so that the value of A is greater for ThX than for UX. while that of the uranium fraction increased. Then N ( t ) = NoeW" (104) . These experimental observations enabled Rutherford and Soddy to formulate a theory of radioactive change. uranium X (UX) to distinguish it from uranium. Rutherford and Soddy studied quantitatively the rate of decay of the ThX activity and the rate of recovery of the thorium activity. They suggested that the atoms of radioactive elements undergo spontaneous disintegration with the emission of a. called the disintegration constant. Rutherford and Soddy (1902) '2' obtained similar results with thorium salts. i. the activity of the UX decreased. an active material which was called thorium X was separated. (103) The notation may be simplified by dropping the subscript x. may then be replaced by the number of atoms N. (101) may be written N.or &particles and the formation of atoms of a new element. which has been called the activity. it was noticed that the thorium X was losing its activity.(t) = N. The results for UX and U are shown in Fig.. (101). is proportional to the number of atoms which disintegrate per unit time. while UX loses half of its activity in 24 days.oeU. A. The experimental decay curve for the ThX was exponential in nature. A. Becquerel then found that if the uranium X and uranium fractions were allowed to stand sep arately for some time. the decay and recovery curves are therefore symmetrical. Then the intensity of the radioactivity. which had been freed from thorium X. The curves are similar to those for the thorium bodies except that the time scale is different. After a few days.
and with two assumptions.AN. 10 is the equation which represents the change with time of the number of atoms of a single decaying radioactive substance. write it in the form and multiply through by eAf. (104) gives dN . To integrate Eq. which is characteristic of a particular radioactive species. With this equation. (105). the second term gives the rate of disappearance of UX atoms because of their radioactive disintegration. it was possible to account for the growth of activity in the thorium or uranium fractions from which the ThX or UX had been removed. The first term on the right side of Eq. The proportionality factor is the disintegration constant. the decrease per unit time in the number of atoms of a radioactive element because of disintegration is proportional to the number of atoms which have not yet disintegrated. Suppose that Q atoms of UX are produced per second by a given mass of uranium. The assumptions are (1) that there is a constant production of a new radioactive substance (say UX) by the radioactive element (uranium). DitTerentiation of both sides of Eq.232 NATURAL RADIOACTIVITY [CHAP. (105).AN. * where N(t) has been abbreviated as N. (106) gives the rate of formation of U X atoms from U atoms. Equation (105) is the fundamental equation o f radioactive decay. and let N be the number of atoms of UX present a t time t after the complete removal of the initial amount of UX.Q . (106). Then The last equation csn be integrated directly to give C is an integration constant determined by the condition that N = 0 . According to Eq. and (2) that the new substance (UX) itself disintegrates according to the law of Eq. Then the net rate of increase of UX atoms in the uranium fraction is dNdt .
and that the probability p for an atom to disintegrate in a time interval At is independent o f the past history of the atom and is the same for all atoms of the same type. where A is the disintegration constant characteristic of the particular radioactive substance. for sufficiently short intervals. the probability of survival is [I . the probability of survival is (1 . or as n becomes very large. Now.when t = 0. and with No = &/A. Fractions of atoms of UX which decay per second A The exponential law o f decay was deduced by E.X(t/n)ln. The probability of disintegration then depends only on the length of the time interval and.AAt)". He assumed only that the disintegration of an atom of a radioactive element is subject to the laws of chance. is proportional to At. Then p = AAt. then the fraction remaining unchanged after a time t is N / N o = e". then the probability that it will not disintegrate in a second time interval At is again 1 . If the atom has survived this interval. Equation (107) also shows that the number of UX atoms in the mass of uranium approaches an equilibrium value for large values of t given by the ratio Q Number of atoms of UX produced from U per second .AAt.A A ~ for )~.AAt. I f the total time nAt is set equal to t. . von Schweidler (1905) without any special hypothesis about the structure of the radioactive atoms or about the mechanism o f disintegration. Equation (107) is the same as the recovery equation (102) so that the theory gives the correct result for the growth of activity in the uranium or thorium after the removal of the X body. where N is the number of unchanged atoms at time t. The probability that the given atom will survive both the first and the second intervals is (1 . n such intervals. The probability that the atom will remain unchanged after time t is the limit of this quantity as At becomes vanishingly small. one of the definitions of the exponential functions is I from which it follows that The statistical interpretation of this result is that if there are initially a large number No of radioactive atoms. This condition gives C = &/A.p = 1 . The probability that the given atom will not disintegrate during the short interval At is 1 .
say in counts per minute. from Eq. it follows that and the detection coefficient cancels out. on the average. and any one of three quantities. i. Hence. that each disintegration is detected. and the mean life. 102 The disintegmtion constant. The disintegra . XINl = h a 4 . that the detection coefficients are all equal to unity. XN. are not necessarily equal. ' ~ ' The number of radioactive atoms N and the activity A have been used interchangeably so far on the grounds that the latter is proportional to the former. or the mean life. the disintegration constant. the use of N and A as equivalent quantities usually leads to no confusion in the case of a single substance. If the number of atoms which disintegrate per unit time is the same for both substances. A radioactive nuclide may be characterized by the rate a t which it disintegrates. depends on the nature and efficiency of the detection instrument and may vary considerably from one radioactive substance to another.respectively.e. A = AN. but the number which break up in any second shows fluctuations around this value.234 NATURAL R A D I O A C I T ~ [m. But. the quantity in which we are usually interested is the ratio of the number of atoms at two values of the time. may be used for this purpose. which is sometimes called the &kction c eienl. and the statistical considerations involved are important in the design and interpretation of experiments having to do with the measurement of r a d i ~ a c t i v i t ~ . This condition usually cannot be achieved in practice. For a given radioactive substance. 10 The law of radioactive decay is thus a statistical l a w and is the result of a very large number of events subject to the laws of probability. the halflife. For any one substance. but it is adopted here to simplify the discussion. When two different substances are considered. (10Sa). with occasional exceptions.. the halflife. The magnitude of these fluctuations can be calculated with the aid of the theory of probability.unless otherwise noted. but the measured activities. The number of atoms which disintegrate in one second is. the measured activitiesare A1 = clAINl and A2 = c2X2N2. The activity will then be equal to the number of atoms disintegrating per unit time. the two quantities are actually connected by the relationship u e d The proportionality factor c. They are equal only if the detection coefficients are equal. It will be assumed.
103. a straight line should result whose slope is equal to 0. tion constant X can be determined experimentally. X is then 0. and Eq. Hence.0186 min'. .3026 times the logarithm to the base 10. with the time expressed in minutes. (104).  .4343Xt.00808.4343Xt1 (1010) No since the logarithm to the base e is equal to 2. " and Eq. if the logarithm of the measured activity is plotted against the time. the plot is made on semilog paper. The number of atoms N(t) is proportional to the measured activity A(t).4343A. or 3. with the help of Eq. for convenience. the slope is 0.60 120 180 240 Time (min) Fra. 103. (1010) may be written (1011) log A(t) = log A. An example of this method of determining X is shown in Fig. which may be written where the symbol "ln" represents the natural logarithm. (109) becomes log N(t) = 0. so that N(t)/No = A(t)/Ao. denoted by "log. In the example shown. The latter can be transformed to the ordinary logarithm.10 X sec'. Graphical method for determining the value of the disintegration constant.0. in many cases.
236 NATURAL RADIOACTIVITY .
Mathematically. the mean life is given by and is simply the reciprocal of the disintegration constant. it can be determined experimentally by measuring the activity as a function of the time. This fraction never reaches mro. and it follows from Eq. and the halflife can be read off directly.5 = 0. of the atoms of a radioactive species.4343XT. usually denoted by r. is given by the sum of the times of existence of all the atoms. The number of atoms which decay between t and t dt is + but the number of atoms still existing a t time t is so that Since the decay process is a statistical one. after seven halflives the activity is 1/128. T. the activity has fallen to 1/1024 or about 0. divided by the initial number. (101 1) that log 0. The relationship between the activity and the halflife is illustrated in Fig. or less than one percent of the initial activity. .693/X. From Eqs. After ten halflives.1% of the original amount.Another quantity which is used to characteriee a radionuclide is the halflife. (1012) and (1013). any single atom may have a life from 0 to oo.5.3010. or average l i e expectancy. After one halflife. (1012)  . Hence.4343XT or log 2 = 0. but it becomes very small. it is found in the following way. The mean life. as in the case of the d i e integration constant. and is usually negligible in comparison with the initial value. N(T)/No = 0. halflife = T = 0. It is also possible to determine the mean life. After n halflives (t = nT).6937. a straight line is obtained. it follows that the halflife and the mean life are proportional quantities: T = 0. the time needed for half of the radioactive atoms to diiintegrate. (1014)  I f the halflife of a single radioactive species has a value in the range from several seconds to several years. 104. When the activity is plotted against the time on semilog paper. Since log 2 = 0. the fraction of the activity remaining is (3)".
238 NATURAL RADIOACTIVITY [CHAP. When the total activity is plotted against the time on semilog paper. any complex decay curve can be analyzed into its components. If the activities are independent. Kennedy. only the longestlived activity will remain.. FIG. Introdvclion to Radioehemistry. The curve is concave upward because the shorterlived components decay relatively rapidly. New York: Wiley. Friedlander and J. and the value of its halflife can be read from the late portion of the decay curve. the various activities can sometimes be distinguished. in which case the observed activity is the sum of the separate activities. and if the values of the activity given by the line are subtracted from the total activity. a curve like the solid one in Fig. (Reprinted by permission from G.) . the other with a halflife of 8 hours.e.0 hr). i. and the separate halflives determined. 105 is obtained. It often happens that two or more radioactive species are mixed together. which will be a straight line. (c) Shorterlived component (T+ = 0. eventually leaving the longlived components. (a) Composite curve. If this straightline portion is extrapolated back to t = 0. W.8 hr). The example in Fig. 10 . After a sufficiently long time. in principle.8 hour. 105 is a mixture of two activities. 0 2 4 6 8 10 Time (hours) 12 14 f a composite decay curve. practical difficulties may limit the usefulness of the method to three com . one with a halflife of 0. 1949. The curve for the total activity is the sum of the two straight lines which represent the individual activities. the curve that remains will represent the decay of all the components of the mixture except the longestlived. Analysis o (b) Longerlived component (T+ = 8. Although. one component of the mixture does not give rise to another. 1b5.
The experimental activity. decays into a stable end product (subscript 3). halflives as short as sec have been determined. If the halflife of a radionuclide is either very long or very short. and so on.1031 SUCCESSIVE RADIOACTIVE TRANSFORMATIONS 239  ponents. and X3. A . Then X can be found from the relation provided that N. The system is described by the three equations   These equations express the following facts: the parent nuclide decays according to the basic law Eq. and values of halflives up to the order of 10'' years have been determined. and the disintegration constants are XI.where c gives the fraction of the disintegrating atoms detected by the measuring device. In each series. respectively. the parent nuclide decays into a daughter nuclide. atoms of the second kind are formed . I t was found experimentally that the naturally occurring radioactive nuclides form three series. X is very small. is equal to CAN. and even a twocomponent mixture may be hard to resolve if the halflives differ by less than a factor of about two. The numbers of atoms of the three kinds a t any time tare denoted by N1. N2. the discussion of which is beyond the scope of this book. which decays in turn. 103 Successive radioactive transformations.. methods different from those discussed so far must be used.e. in turn. In the study of radioactive series. This method has been used successfully for longlived aemitters. it is important to know the number of atoms of each member of the series as a function of time. For very short halflives. the number of atoms of the nuclide in the sample. (105). The procedure may be illustrated by treating the case of a radioactive nuclide. When the halflife is very long. X2. and N3. and that c is known as a result of an appropriate calibration. i. until finally a stable end product is reached. is known. other methods '4' must be used. it may not be possible to detect a change in activity during the course of the measurement. the latter.. denoted by the subscript 1. which decays into another radioactive nuclide (subscript 2). The answer to a problem of this kind can be obtained by solving a system of differential equations.
this gives The last equation can be integrated directly to give where C is a constant of integration. This expression for N1 is inserted into Eq. atoms of the stable end product appear a t the rate A2N2 as a result of the decay of atoms of the second kind. The value of the integration constant is determined by noting that when t = 0. (1015b). or N2 = Nil with N i equal to a constant.240 NATURAL RADIOACTIVITY [CHAP. 10 a t the rate AlNl because of the decay of parent atoms. The number of atoms Nl can be written down immediately. (1016) where # is the number of atoms of the first kind present at the time t = 0. and disappear a t the rate X2N2. we obtain the solution for N2 as a function of time: . and gives  Multiply Eq. Then Inserting this value into Eq. Multiplying through by e'2' gives . I t is instructive to solve this system of equations in detail because the procedure is one which is often used. (1018) and rearranging. N1(t) = Nye"'. (1017) through by eb'. the number of atoms of the second kind has some constant value.
although slowly at first. the constants N i and N i are both equal to zero. the result is  Equations (1016). The ordinates of the curves represent the relative numbers of the substances 1. The number of atoms. and passes through a maximum after about three hours. . This condition gives D = N : + N! + N?. since eventually all the atoms of the substance 1 will be converted to atoms of the stable end product. the corresponding values of the disintegration constants are X1 = 0. and 3 as functions of the time when the initial number of atoms of the substance 1 is taken as Ny = 100. 2. N3 approaches 100. and (1021) represent the solution of the problem. 106 show what happens in this case if it is assumed that the halflives of the active species are T1 = 1 hour and T 2= 5 hours. and the solutions for Np and N3 reduce to  The curves of Fig. One of the cases met most often in practice is that in which only radioactive atoms of the first kind are present initially. of substance 1. (1020).693 hr'. respectively. respectively. In this case. (1016) with a halflife of 1 hour. When this expression for D is inserted into Eq. (1015c) and integrating. Nz is initially zero.1031 SUCCESSIVE RADIOACTIVE TRANSFORMATIONS 241 The number o f atoms o f the third kind is found by inserting this expression for N2 into Eq. and X2 = 0. increases. when t becomes very large. and the solution of this problem is often useful. and then decreases gradually. which gives where D is an integration constant. determined by the condition N3 = N: a t t = 0. The treatment just discussed can be extended to a chain of any number of radioactive products. Nl.1386 hr'. The number of atoms N3 of the stable end product increases steadily with time. (1019). decreases exponentially according to Eq.
10 Time (hours) FIG.242 NATURAL RADIOACTIVITY [CHAP.XnNn. The procedure is similar to that of the special case already considered except that the mathematics becomes more tedious and the expressions for the numbers of atoms become more complicated as the length of the chain increases. dNn dt The solution of this system of equations under the assumption that a t . and the third member is stable.= XnlNnl . the daughter has a halflife of 5 hr. A radioactive series with three members: only the parent (T+= 1hr) is present initially. 106. The differential equations of the system are .
7 min. (1024) to (1027).and a certain number of atoms of the decay product of radon. 107.).8 min. + cne'"'. Chadwick. which decays. ~ ~(1026) ~ An example of the application of the Bateman equations.'5' The initial conditions are The number of atoms of the nth member of the chain is given by Nn(t) = ~.1031 SUCCESSIVE RADIOACTIVE TRANSFORMATIONS 243 Time (min) Fro.05 min were deposited on the test body.'+ ~~ ' ~ +e . the end product RaD or Pb210. The last halflife is sufficiently long so . Radiations jrom Radioactive Substances. into RaC (Bi214) with a halflife of 19. Eqs. Chadwick. with a halflife of 22 years. is shown in Fig.1 t = 0 only the parent substance is present was derived by Bateman. [Reprinted by permission from Rutherford. 107. Finally. taken from Rutherford. RaA or Po218. with a halflife of 3. . The decay of Radium A.e'l' with + ~ ~ e . and Ellis' book Radiations jrom Radioactive Substances. is formed. in turn. 1930. The curves were obtained under the conditions which follow. The RaA decays into RaB (Pb214)with a halflife of 26. Cambridge University Press (Macmillan Co. and Ellis. A test body was exposed for a few seconds to radon (Em222).
and then decreases with time. Solutions can be obtained for other problems with different initial conditions.241 NATURAL RADIOACTIVITY [CHAP. passes through a maximum about 10 min later. which is a contradiction. the number of atoms N1 can be taken to be constant. The number of RaA atoms decreases exponentially. in other words. The number of RaC atoms passes through a maximum after about 35 min. This . (1028) is then a very good approximation. dNn/dt are all equal to zero. reaching a maximum when the RaA and RaB have d i s appeared. Chadwick. This condition is satisfied by the naturally occurring radioactive chains. and the value of X I is very much smaller than that of any of the other X's in the chain. . . and Ellis for additional examples of the application of the theory. if the parent has a halflife very long compared with that of any of its decay products. The first of Eqs.. or that the number of atoms of any member of the chain is not changing. The conditions for equilibrium are then These conditions cannot be satisfied rigorously if the parent substance is a radioactive substance because the first of Eqs. if the parent substance decays much more slowly than any of the other members of the chain. Eventually the RaD would decay exponentially with a halflife of 22 years. The number of RaD atoms increases. It is possible to achieve a state very close to equilibrium. The number of RaB atoms is initially zero. When this condition is applied to the members of a radioactive chain described by Eqs. (1028) implies that X1 = 0. (1024). Uranium I has a halflife of 4. 10 that the number of radium D atoms which disintegrate may safely be neglected. and the reader is referred to the book by Rutherford. . The term equilibrium is usually used to express the condition that the derivative of a function with respect to the time is equal to zero. 104 Radioactive equilibrium. the initial number of atoms of RaA. and the rest of the conditions are rigorously valid. dN~/dt. however. i t means that the derivatives dNl/dt.5 X 10' years and the fraction of UI atoms transformed during the life of an experimenter is indeed negligible. In such a case. The sum of all the atoms present a t any time is No.
8 X 10' N2 X 1620 = 4. Consequently. that no decay products have been removed or allowed to escape for a long enough time for secular equilibrium to be established. and the condition (1029) is satisfied.8 X lo6 atoms of uranium I. It is only necessary to be sure that the material haa been undisturbed. The halflife of radium is known from direct measurements to be 1620 years. The term AINl in the second of Eqs. Equation (1032) may be rewritten .5 X log years. in terms of halflives. < h z . and (1031) N1 = N:. The relationships (1030) can be used to find the halflife of a radionuclide whose halflife is very long.1041 RADIOACTIVE EQUILIBRIUM 245 . For example. (1024) is then constant. Then (1016) and (1022) if it is noted that A1 = 0. Secular equilibrium can also be attained when a radioactive substance is produced a t a steady rate by some artificial method. and satisfies the condition or. that is. so that the latter is initially pure. The mathematical expressions for the number of atoms of parent and daughter may be obtained from Eqs. and A1 < e .~ ~ t 1. It is assumed that the daughter has been separated from the parent. The relationships (1029) and (1030) may be applied whenever several shortlived products arise from successive decays beginning with a relatively longlived parent. uranium minerals in which secular equilibrium has been established have been shown to contain one atom of radium for every 2. the halflife of uranium I I T I= N T2 = 2. We consider next an example of the approach to secular equilibrium. type of equilibrium is called seculut equilibrium.  . then a t equilibrium N 1 / N 2 = 2. If uranium I is denoted by the subscript 1 and radium by the subscript 2.8 X lo6. such as a nuclear reaction in a cyclotron or chainreacting pile. The case is that of a longlived parent (T = GO) and a shortlived daughter. as in the case of a very longlived parent.
1949. satisfying the condition for secular equilibrium. 10 0 2 4 6 7 8 i O i 2 Time (hours) PIG. in terms of the (constant) activity of the parent.) The last equation gives the activity of the daughter as a function of the time. (1016) and (1022). called transienl equilibrium. New York: Wiley. but the halflife of the parent is not very long. results if the parent is longerlived than the daughter (XI < X2). the numbers of atoms are again given by Eqs. For comparison. The total activity is given by A b t . Friedlander and J. e'2' becomes negligible compared with e'~'. W. ~= X ~ N ? XzNz = 2 2 ~ ~ ~ ~1N!e'~'. In this case.108.80 hr). the approximation X I = 0 may not be made. If the parent and daughter are separated so that the parent can be assumed to be initially pure. Kennedy. 108. (Reprinted by permission from G . These relationships are shown graphically in curves a and b of Fig. Introduction to Radiochemistry. (a) Daughter activity growing in freshly purified parent fraction. (d) Decay of freshly isolated daughter fraction (T4 = 0. curve d shows how a freshly isolated daughter fraction would decay. X2Nz approaches hIN:. : (1034) + Equation (1033) shows that as t increases the activity of the daughter increases. and after several halflives. and the number of  . Secular equilibrium.246 NATURAL RADIOACTIVITY [CHAP. After t becomes sufficiently large. curve c gives the total activity. (b) Activity of parent (T4 = w ) . A somewhat different state of affairs. (c) Total activity of an initially pure parent fraction.
Kennedy. W.) atoms of the daughter becomes Thus. (c) Total activity of an initially pure parent fraction.80 hr). (Reprinted by permission from G. New York: Wiley. . the daughter eventually decays with the same halflife as the parent. 109.0 hr).hl). (e) Total daughter activity in parentplusdaughter fractions. are shown graphically in Fig. which are characteristic of transient equilibrium between parent and daughter atoms.1041 RADIOACTIVE EQUILIBRIUM 247 Time (hours) FIG. i t follows from Eq. Since N:e'l' = N 1 . (d) Decay of freshly isolated daughter fraction (Ti = 0. (b) Activity of parent (T+= 8. Introduction to Radiodremistry. Friedlander and J. The above results. 1949. 109. (1035) that The ratio of the measured activities a t equilibrium is Xa/(Az and the daughter activity is greater than that of the parent by the factor . Transient equilibrium. (a) Daughter activity growing in freshly purified parent fraction.
the head of the series. There + + + Radioactive speeiea Type of di8iJltegnation Q Uranium I (UI) uranium XI(UXl) Ursnium Xn (UXn) Uranium z (UZ) Uranium I1 (UII) Ionium (10) Radium (Rs) Ra Emsnation (Rn) Radium A (RaA) Radium B (RaB) Astatine218 (AtH8) Radium C (RaC) Radium C' (RaC') Radium C" (RaC") Radium D (RaD) Radium E (RaE) Radium F (RaF) Thallium206 (T1206) Radium G (RaG) B B B Q Q Q Q Q. then as the parent decays.B Q B B B Q B Stable . As a result of physical and chemical researcL on the naturally occumng radioactive elements. In the actinium and thorium series. pass through a maximum. f the parent and daughter are s e p no state of equilibrium is attained. it was proved that each radioactive nuclide is a member of one of three long chaine. I arated initally. where n is an integer. and the uranium aeries is sometimes called the "4n 2" series. 10 When the parent has a shorter halflife than the daughter (A1 > X2). and eventually decay with the halflife of the daughter. stretching through the last part of the periodic system. the mass number of each member can be expressed in the form (4n 2).B B Q Q.248 NATURAL RADIOACTIVITY [CHAP. These series are named the uranium. the mass numbers are given by the expressions 4n 3 and 4n. actinium. In the uranium series. after elements at. respectively. or radioactive series. or near. respectively. the number of daughter atoms will increase. 1&5 The natural radioactive series. and thorium series.
9 m 3.68d 0. 1011 give the corresponding information about the actinium series. hours.@ .lm 10~8 2. The halflife may be in years. minutes. historical names of the radionuclides. The members of the uranium series are listed in Table 101 together with the mode of disintegration. dl h.52 s 4.@ 84Poa1' a 81~1307 @ 8zPbZo7 Stable 11.92 s 8m 1. or seconds. + Radioactive speciea Actinouranium (AcU) U d u m Y @Y) ProtOBetinium (Pa) Actinium (Ac) Nuclide rype o f disinte gration Halflife Dieintegration constant.. Both are included because both are found in the literature of physlcs. Mev P ~ U o0ThZ3' o'Pa2a' 89AcZz7 ~ ~ ~ Radi~tinium(RdAc] poTh'2' 87Fr223 Actinium K (AcK) Actinium X (AcX) Aetatine219 Ac Emanation (An) Biimuth215 Actinium A (AcA) Actinium B (AcB) Astatine215 Actinium C (A&) Actinium C' (AcC') Actinium C" (AcC") Actinium D (AcD) a . 1010. The mode of decay.@ s2Pbz" @ 85At215 a 8sBi2" a.79 m 8 ..'6' 'The changes in atomic number and mass number are shown in Fig.P02'6 a. 1012 are for the thorium series. while Table 103 and Fig. days.'6' the disintegration constant.83 X 36. while the second column gives the modern symbol. and the maximum energy of the emitted particles. m. the modern symbol gives the atomic and mass numbers. the value of the halflife (or disintegration constant). The older name usually indicates the series to which a radioactive substance belongs and gives some idea of the relative position in the series.@ 8. Table 102 and Fig. the halflife.1051 THE NATURAL RADIOACTIVE SERIES 249 is no natural radioactive series of nuclides whose mass numbers are represented by 4n 1.At2'Q a. sec' Particle energy. respectively. and it is helpful to have a code for translation readily available.Em2'g a 88Bi216 a. The first column of each table gives the oldfashioned. and the energy of the emitted particle characterize the radioactivity. s. abbreviated as y.15 m 0.
33 Particle ene'w. Radioactive species Thorium (Th) Memthoriuml(MsTh1) Memthorium2(MsTh2) Radiothorium (RdTh) Thorium X (ThX) Th Emanation (Tn) Thorium A (TM) Thorium B (ThB) Astatine216 (AtZ1e) Thorium C (ThC) Thorium C' (ThC') Thorium C" (ThC") Thorium D (ThD) Type o f Nuclide dirrintegrstion ooThZS2 ssRaZZ8 apA~"~ poThZ2a siRa2'4 &mz2P s4PoZ16 szPb"2 ssAtZ1e ssBizlz a Halflife Diaintegrstion conetsnt.91 x lo' S ~ P O ~ Q' ~ ~1TlZ'J~B szPbZos Stable .82 X 106 2.20 X 1.39 X 1Ol0y 6. sec' 1. 10 Atomic number FIG.16 s 10. Mev B B a Q Q Q.250 NATURAL RADIOACTIVITY [CHAP.5 m 100 106 108 106 102 1.15 X 2. 1010.3 X 103 1.28 X 3.5 s 0.910 y 3.7 y 6.14 x 1.B B a QI B 1.6 h 3 X lO's 60. The uranium (4n + 2) series.58 X 3.64 d 51.13 h 1.34 X 4.
The thorium (4n) series.1051 THE NATURAL RADIOACTIVE SERIES 251 Atomic number FIG.1012. .1011. The a c t i n i u m (4n + 3) series. Atomic number FIG.
RaC'. ThC emits a &particle to form ThC' (Po2") or an aparticle to form ThC1' (T1208). ThC' then emits an aparticle to give stable pb208 (ThD). AcB. giving one a. and ThD all have the atomic number 82 and are isotopes of lead. Similarly. while ThC1' emits a &particle and also forms stable PbZo8. The case of ThC ( B i 2 " ) is especially interesting because 66.or @particleand one atom of the product nuclide. 10 In the case of 8particles.7% by aemission. RaB." Figures 1010 through 1012 illustrate Soddy's displacement law discussed in Section 91. + +  . ThA.3% of the disintegrations are by @emission. and ThC' all have the atomic number 84 and are isotopes of polonium. ThB. the halflife is T = 0. Bi214(RaC). the aparticles are monoenergetic. in which case the greatest value is listed and denoted by "m. the atoms break up in two different ways.Ac. RaF. and 33. and by (AcC). A&). On the other hand.693/X = 0. Esch of t h e nuclides can decay either by aemiasion or by &emission. by Ac"~. . AcA.252 NATURAL RADIOACTIVITY [CHAP. Po2'' (AcA). All of the elements in any one column haye the same atomic number and must occupy the same place in the periodic table. &A. 87Fr223 (AcK). In each branched decay. and a 8transition by a horizontal arrow pointing to the right. For example.693/(X. for others. and 8 5 ~ t 2in Po216 (ThA) and Bi212 (ThC) in the thorium series. In most cases. For some radionuclides. therefore. the energy value listed i s that of the mast energetic particles. giving rise to two products with different properties. Series schemes like those of Figs. however. I t will be remembered that the emission of an aparticle decthe charge of a nucleus by two units and the mass number by 4 units. with one percent or less of the atoms decaying by emission. an aparticle may have one of several energy values. in contrast to the other cases mentioned. In most of the disintegration processes that make up a radioactive series. an disintegration is represented by an arrow sloping downward and to the left. and Bi2l0 (RaE) in the uranium series. RaC. AcD. RaG. the product atoms decay in turn to give the same nuclide. AcC'. Bi211 1 0the actinium series. and AcC decay almost entirely by Mecay. In some cases. each of the radionuclides breaks up in a definite way. RaD. one mode of decay is much more probable than the other. with only a small fraction of one percent of the atoms disintegrating by &emission. The probability of disintegration is the sum of the separate probabilities and X = A. 1010 through 1012 led Soddy (1913) to the discovery of the existence of isotopes. while the emission of a &particle increases the charge of the nucleus by one unit and leaves the mass number unchanged. RaE. For example. Radium A and AcA decay almost entirely by aemission. A&. In the figures. This kind of disintegration is called branching decay and is illustrated by the decay of Po2" (RaA).
The disintegration constant is I f W is the unknown weight.from its halflife of 26. Po215. At2''.70 X 10" disintegrations/sec. consider the problem of calculating the weight in grams of 1 curie and 1 rd of RaB (Pb214). ~ t ~ " or. which is onethousandth of a curie. or in terms of the number of emitted particles counted per unit time by a detector. for a substance with a short.respectively. defined as that amount of a radioactive substance which gives 10' disintegrations/sec. As in other branches of physics. the standard unit is the curie.8 min. namely. a new absolute unit of radioactive disintegration rate has been recommended. there has been some confusion about the use of the curie as the standard unit.lCr6l UNITS OF RADIOACTIVITY 253 1 M Units of radioactivity. For various reasons. mainly historical. To get an idea of some orders of magnitude. the amount of material which gives a curie of activity is almost unimaginably . it is useful to have a standard quantitative unit with an appropriate name. equal t o onemillionth of a curie. and the microcurie. these would correspond t o amounts of active materials giving 3.7 X 104 disintegrations/%~. then For 1 rd. then I f dN/dt = 3. The millirutherford (mrd) and microrutherford (prd) correspond to lo3 disintegrations/sec and 1 disintegration/sec.7 X 10' and 3.70 X 10" disintegrations/sec (1 curie). at present defined as that quantity of any radioactive material giving 3. Thus. In radioactivity. Consequently. When the halflife is a small fraction of a second (Po2". Po2"). but not very short. the rutherford (rd). are also useful units. The millicurie. The intensity of radioactivity has so far been considered in terms of the number of atoms which disintegrate per unit time. respectively. halflife very little material is needed to provide a curie of activity.
5. Phil. 585 (1958). E.) G.. New York: Academic h. W.423 (1910). New York: Wiley.. J. STROYMOER. 576 (1903).9 X lo'' sec'. R ~ T H E ~and RD F. op. G. HOLLAX~ and E R SEABORO. 1955." J. REFERENCES GENERAL R~ER~OR CHADWICK. X = 4. 26. E. (Reprinted with corrections. 7. 321. gen. No." Am. 4 N u c k Phyaica. Then For 1 curie of activity. KENNEDY. Radidion Doaimtry. Chapter 9. Chapters R. PANETE. E. Soc. In the case of a nuclide with a very long halflife. Phya. 3.. 816. cit. gen. L." Proc. NU& d Radhhtt*. New York: McGrawHill.and GammaRay Spec&ompy. 2. "Table of Isotopes. &. and more than three metric tons are needed. such ae UZ3' (T = 4. The Atomic Nuclew. 837 (1902). 111. PUTNAM. . 1930. S. Chapter 26. H. 3. 1951. 15. Decay. ref. D. Segr&. and ELLIS. D. S ~ a a h"Radioactive . cit. 1959. 6." h a . Mod. Rombadioc St&mue$. Pert IX. 1956. G. 16. PARTICULAR 1. KENNEDY. New York: Macmillan. 569 (1902)." Be&. JAUNCEY. 1955. "Early Yeam of Radioactivity. E. L. 2nd ed. G. HEVESY and F. 2 (Sept. A Manual of Radiooctioiry. New York: Interscience. 1955. M.. ed. "The Solution of a System of Merential Equations Occurring in the Theory of Radiobctive Transformations. G." E z p c n . BROWNELL. and J. 30.small. eds. 1938. 1948). EVANS. New York: Wiey. W. 81. 25. SODDY. 226 ( 1 9 w . "The Cause and Nature of Radioactivity. pp. BATEMAN. Chcm. K. ROWLANDS. A. Phya." Nudumicr 3. Vol. ed. 4. Cambridge Phil. 14. ~ I E D L A N D E Rand J. H ~ and E G. Mag. "Methods of M e a s u r i n g Very Long and Very Short HalfLivea. Amsterdam: North Holland Publishing Co. 370.. J . S a a d . ibid 5. ref. E. Oxford University Prese. Siegbahn. . "Measurement of Disintegration Rate. ~ E D L A N D E B 4. op.Rad&&ma fta.50 X 1 0 ' y). J.
100 mg was found to undergo 4. Find the values of the disintegration constant and halflife of a radioactive substance for which the following counting rates were obtained a t different times.0 3.5 985 2. how many atoms will disintegrate during the first day? During the fifth day? During the tenth day? 4. rate. HOWlong after a sample of UXI has been isolated wiU i t take for 90% of it to change to UX2? 950/0? 99%? 63. initially. What are the weights of (a) 1 curie and 1 rd of Ra? (b) 1 curie and 1 rd of Rn? (c) 1 microcurie and 1 rd of RBA? (d) 1 microcurie and 1 rd of R a C ? 6.0 530 7345 10. What is the halflife of this nuclide? Ionium is formed by the adecay of UZ3'. 3.0 3500 0. 10 days? If the sample contains.0 1835 9080 8. The halflife of UX1 is 24. How many disintegrations per second would the sample undergo immediately after separation? How many after 10. A freshly separated sample of RaF (Po210) contains 1. What fraction of a freshly separated sample of this nuclide will disintegrate in one day? In 2. Time.00 X that nuclide. The halflife of radon is 3. rate. 70.100mg sample of ThZaO? gm of 7.1 days. Plot the decay curve on semilog paper and determine the halflives and initial activities of the component activities.2yo? 5.82 days.0 290 4985 12.5 2520 14500 5.0 0. 300 . hr hr counts/min counts/min What would have been the counting rate a t t = O ? 2. HOWmany rutherfords of U234 would be needed to produce a 0. Counting Counting Time.32 X lo6 disintegrations/min. 30. 50. A sample of ionium (ThZ3O) weighing 0.PROBLEMS 1. Counting Counting Time.0 1. The counting rates l i s t e d below were obtained when the activity of a certain radioactive sample was measured a t different times.0 3.O 11410 6. 4. 5. Time. 100. one microgram of radon.0 1.  19100 4.
15. Consider a radioactive series whose first two members have halflives of 5 hours and 12 hours. the relative activities of the initially pure U236 fraction. Natural uranium contains 0. (a) What are the partial disintegration constants for adecay and Bdecay? (b) What are the halflives for adecay alone and for w e c a y alone? (c) At what rates will a freshly separated sample containing lo' gm of Bi212 emit aparticles? 0particles? (d) What will be the rates after three hours? 11. as functions of time. the UX1 decays with a halflife of 24. all the Th234 from 100 gm of uranium is removed. The nuclide Th234 (UX1) is normally present in uranium compounds in secular equilibrium with its parent U2381which has a halflife of 4. Assume that there are initially lo6 atoms of the first member. the U23s loses eight aparticles. if any.00 X 10lo gm of that nuclide a t time t = 0.256 NATURAL RADIOACTIVITY [CHAP. which decays with a halflife of 1.18 min into U234 (UII). Find the atomic concentrations. After how many hours will the number of atoms of the second member reach its masimum value? 9. 10 days? How many curies of activity would the sample contain a t theae times? How many rutherfords? 8. Plot. and none of the second and third members. The nuclide U235 has a halflife of 7. The nuclide Bi212 (ThC) decays both by aemission and &emission. all radioactive nuclides produced from U238 and U235 are in secular equilibrium with these parent nuclides.6 hours. the Th231 growing in the freshly purified U235 fraction. the latter has a halflife of 2. Natural samarium has been found to emit aparticles a t a rate of 135 particles/gm/sec.10 X lo8 years. if any? 13. (b) a 10day period. with a halflife of one day or more. A freshly purified sample of RaE contains 2. and of the U235 itself. 'In normal uranium ores. Plot the activities (in rutherfords) of Th234 and Pa234 as functions of time for (a) a 10min period.1 days to Pa234 (UXz). Neglect the less frequent type of decay when branching occurs. Suppose that the Th231 is separated chemically from a sample of U235. If the sample is allowed to stand. The Pb206 (RaG) content of a uraniumcontaining mineral can be used to determine the age of the mineral. . If the formation of UZ is neglected. I n the transition from U238 to PbZo6.0%) is responsible for the activity. its daughter Th231 has a halflife of 24. what is its halflife in years? 12. of each daughter nuclide of uranium. Plot. Discuss the kinds of equilibrium involved. 10. respectively.50 X lo5 years. the isolated Th231 fraction.72% by weight of U235 and 99.50 X lo9years. what will be the greatest weight of R a F that i t will ever contain? At what time will that amount be present? At that time. The isotope Sm"' (abundance 15. 14. what will be the aactivity in disintegrations per second? The @activity? How many rutherfords of RsF will be present? Plot the aand 8activities as functions of time. the numbers of atoms of the three kinds. in parts per billion. as functions of time. What kind of equilibrium is reached.28% of U238. while the third member is stable. Suppose that st time t = 0.
and if the decrease in the amount of U238 is neglected.5 X lo9 years.P. A given sample of a uraniumbearing mineral was found to contain 5. the helium can be collected and its volume measured. . by the formula Age = (He/U) X 8. approximately. leads to 2.0(He/Ra) years. per gram of uranium. Most of the helium produced in a uraniumbearing mineral is trapped and retained as gaseous helium. or 1.33 X of uranium? If the decresse in the amount of uranium because of disintegration is taken into account. Show that the age may also be written Age = 3.58Pb206) X 7. and 1.1 X cm3 of helium per gram of mineral.5 X lo9 years.16 X lo' cm3 of helium a t N. When a sample of the mineral is dissolved or fused. where the quantity in brackets represents the quantity of helium in cubic centimeters a t N.05 X 10l3 gm of radium per gram of mineral.PROBLEMS 257 : Show that if the presence of U235 is neglected (and this has been shown to be valid). where the quantity in parentheses is the weight in grams of Pbm6 per gram of uranium. Show that the disintegration of one gram of uranium per year. Show that the age of the mineral is given.07 X lo" gm of helium. +  How much difference does this correction make in the ages obtained in (a) and (b)? 16. What is the age of a uraniumbearing rock which contains (a) gm of PbZo6per gram of uranium? (b) 0. then the age can be represented approximately by the formula Age = (Pb206/U) X 7.1 gm of Pb206 per gram 1.P.6 X lo6 years. the last formula may be replaced by the better approsimation Age = (Pb206/U 0. What is the approximate age of the sample? .T. under the assumptions of the last problem.T.
just outside the opening. and Ellis. one o f the latter might penetrste into a nucleus and cause a disruption. The source of the aparticles was radium C placed on a small disc D.). and is shown schematically in Fig. In one end E of a box B was cut an opening which was covered by a silver foil F. and aparticles seemed moat likely to be effective because of their relatively great energy and momentum. The apparatus that Rutherford used was simple but sensitive. be reduced and the probability of disintegration increased by using some of the lighter atoms as targets. It seemed possible that if atoms were bombarded with energetic particles. however. 111. whose distance from S could be varied.CHAPTER 11 ARTIFICIAL NUCLEAR DISINTEGRATION 111 Transmutation by alphaparticles : alphaproton reactions. Different gases could be introduced into the box and removed through the side tubes T. The extent of the scattering could. [Reprinted by permission from Rutherford. When the box was filled with oxygen or carbon dioxide FIG. The first disintegration based on these ideas was made by Rutherford (1919)~"' who showed that the nuclei of nitrogen atoms emit swift protons when bombarded with aparticles from radium C. The radiations from the natural radionuclides could be used as projectiles.1 258 . it was apparent that the probability of causing a nuclear disintegration would be small. thereby reducing the magnitude of the repulsive Coulomb forces between the target nuclei and the aparticles. 111. Chadwick. The fact that certain atoms undergo spontaneous disintegration led to speculation about the possibility of causing the disintegration of the ordinary inactive nuclides. 1930. Since most of the bombarding aparticles would probably be scattered. The silver foil F was thick enough to absorb the aparticles from the source. Diagram o f Rutherford's apparatus for the disintegration o f nitrogen nuclei by aparticles. Cambridge University Press (Macmillan Co. Radiations from Radioactive Substances. and scintillations on the screen were obsemed by means of a microscope M. A zinc sulfide screen was placed a t S.
by means of careful experiments. that as rr result of a disintegration. there should have been four tracks rather than the three actually . and a recoil nucleus. The choice bestudied ' tween the two possibilities was settled in 1925 when B l a ~ k e t t ' ~ the tracks produced by aparticles passing through nitrogen in a cloud chamber. a t atmospheric pressure.1111 TRANSMUTATION BY ALPHAPARTICLES 259  . Two hypotheses were suggested as to the nature of the nuclear process leading to the emission of the proton. Measurement of the magnetic deflection of the particles suggested that they were protons. (b) The aparticle is captured by the nucleus of the atom it hits. as can be seen in Fig. Rutherford and Chadwi~k'~' extended the work on nitrogen to other elements and found evidence of the disintegration of all of the light elements. When the gas in the box was nitrogen. with the exception of carbon and oxygen. They also found that in some cases the energy of the ejected protons was greater than that of the bombarding aparticles. If the disintegration process had been the result simply of a disruption leading to the emission of a proton from the nitrogen nucleus. a proton. or compound. This thickness z of gas was great enough to absorb all of the aparticles from radium C even without the silver foil. 112. The absence of a track corresponding to an aparticle after the collision proved that the aparticle disappeared completely. however. the extra energy being acquired in the accompanying nuclear rearrangement. the only tracks which could be seen were those of the incident aparticle. and concluded that artificial disintegration of nitrogen atoms had taken place. and one result w a s the emission of a highly energetic proton by the nitrogen nucleus. Since it was knawn that aparticles from radium C could not penetrate 40 cm of air. one proton was produced for about one million aparticles passing through the gas. The experimental results also showed that the probability of disintegration was very small. and this surmise was confirmed by more precise work. Rutherford ruled out. scintillations were observed on the screen when the source of the aparticles was as much as 40 cm away. and the new. nucleus emits a proton. . Rutherford concluded that the scintillations were caused by particles ejected from the nitrogen nucleus by the impact of an aparticle. the possibility that the protons came from hydrogen present in the nitrogen as an impurity. He showed. This result provided additional evidence that the protons were emitted as the result of a disintegration process. while in case (a) it should still exist after the collision. from boron t o potassium. They were: (a) The nucleus of the bombarded atom simply loses a proton as the result of a collision with a swift aparticle. The disintegration was caused by the aparticles from the radium C. no scintillations were seen on the screen with the source 7 cm or more away. These two hypotheses for the disintegration process could be subjected to experimental test because in case (b) the aparticle should disappear.
the symbols on the left stand for the reacting nuclides.h. 112.p)OL7. This transmutation may also be rcprescntcd by the nblxeviatcd notation . this kind of nucleus is often called a compound nucleus.260 ARTIFICIAL NUCLEAR DISISTEGRATION [CHAP. I n other words.The transmutation of other nuclides by . A cloud chamber photograph showing the disintegration of a S 1 4 nucleus by an aparticle with the formation of 017 and a proton. (Blackett.L4(a. therefore. are on the right sidc of the equation. 11 FIG. The disintegration of a nitrogen atom by a n aparticle may be rcprcsented by a n equation analogous to those used for chemical reactions. I t was concluded.C3)) seen. a n atom of nitrogen has been transformed. that the aparticle entered the nucleus of the nitrogen atom with the formation of a n unstable system which immediately expelled a proton. The symbol in brackets stands for the unstable nucleus formed as thc result of the capture of the aparticle by the nitrogen nucleus. I n this equation. The charge and mass numbers must be the same on the two sides of the equation. or transtnutcd. into an atom of oxygen. so that the nitrogen nuclcus (Z = 7) must he transformed into the nuclcus of an isotope of oxygen. Thc long track of the proton and the short track of the recoiling oxygen nucleus can be seen. the products of the reaction. The emitted proton and the final nucleus.
therefore. x the bombarding particle. respectively. A nuclear reaction such as that represented by Eq. Thus. + m. (1 131 . the charge of the nucleus is increased by one unit and the mass is increased by three units. 112 The balance of mass and energy in nuclear reactions.c2) + Nxc2 = (EY+ 1Jfyc2)+ ( E y + my^^). In other reactions. The alphaproton reaction may.p) reactions. apart from the nuclei involved. product. and Cn represent the target. be written in the form where X. Y the product nucleus. to be discussed in later sections of this chapter.aparticles from radioactive substances may be represented by equations similar to that for nitrogen. I t is often convenient to refer to a nuclear reaction in terms of the incident and emitted particles. Since the total energy of a particle or atom is the sum of the rest energy and the kinetic energy. the reactions mentioned so far are examples of (a. and y the product particle. x is an aparticle. In the only type of reaction considered so far. (111) can be analyzed quantitatively in terms of the masses and energies of the nuclei and particles involved. Y. and compound nuclei. The analysis is similar to that used for chemical reactions except that the relativistic relation between mass and energy must be taken into account. I t will be assumed that the target nucleus X is initially a t rest so that it has no kinetic energy. the statement that the total energy is conserved in the nuclear reaction means that (E. and y is a proton. In each case. other bombarding and product particles will be met. Consider a nuclear reaction represented by the equation  where X is the target nucleus. The analysis of nuclear reactions is one of the main sources of information about nuclear properties and will therefore be discussed in some detail before more reactions are considered.
. but it can FIG.E. 11 I n Eq. (114). it can be determined either from the energy difference or from the mass difference in Eq. (116) that the analysis of nuclear reactions involves information about nuclear masses and particle energies. . (116) represents the recoil (kinetic) energy of The t the product nucleus. the reaction is then said to be ezothermie or exoergie. depending on what information is available. Q = E y + E.E N .m. the Qvalue. e r m E y in Eq.. I t is usually small and hard to measure. therefore. It is apparent from Eq.113. m. (114) can be expressed in terms of the masses because of the relationship (1l 3 ) . EY 4. We now introduce the quantity Q. and My represent the masses of the incident particle. I f the value of Q is negative. be used to obtain information about the masses of nuclei. respectively .m.. Conservation of momentum in nuclear reactions. (1141 The quantity on the right side of Eq. + (115) The quantity Q is called the energy balance of the reaction or. from Eqs. Nuclear reactions can. The outcoming particles are viewed at an angle of 90' with the direction of the incident particles. more commonly. The total mass of the reactants is greater than that of the products in this case. target nucleus.My .. the kinetic energy of the products is greater than that of the reactants. product particle.. or about Qvalues. Hence. (113). and which quantities can be measured. If the value of Q is positive. which represents the difference between the kinetic energy of the products of the reaction and that of the incident particle.)c2.262 ARTIFICIAL NUCLEAR DISINTEGRATION [CHAP. about particle energies. the reaction is endothermic or endoergic. (114) and (115). and product nucleus. the E's represent kinetic energies.Ez = (Mx ml .. Mx.
for simplicity. the result is Setting E. But not all of that kinetic energy is available for excitation because some is used to . Since the resultant momentum after the collision must also be directed along the zaxis. This energy must be supplied in the form of kinetic energy of the incoming particle. Before the collision.. of course. = MyVy cos $. Equation (1110) reduces. E. the energy Q is needed to excite the compound nucleus sufficiently so that it will break up. where v. to Eq. the ejected particle. (119) when e = 90'. it follows that (117) m. E Y = +MyVg. as in Fig. therefore.. (116) gives The value of Q can be determined. we get  Insertion of this expression for E y into Eq. m@. the special case in which the outcoming particles are observed a t an angle of 90' with the direction of the beam of projectiles. = &n&. = &n&. and the recoil nucleus. and $ is the angle between the xaxis and the direction of recoil of the product nucleus. if the energies of the incident and ejected particles are measured and if the mass number of the product nucleus is known. (119) and (1110). = M Y Vy sin 4. The importance of the last term decreases as the mass of the target nucleus increases. the momentum vector is directed along the xaxis. taken as the zaxis. it can be shown that (see Problem 19 a t the end of the chapter). and Vy are the velocities of the incident particle. In the general case of a reaction in which the outcoming particle is observed a t an angle e with the direction of the projectile beam.be eliminated by taking into account the conservation of momentum.v. a good approximation to the value of Q can be obtained by using the mass numbers instead of the actual masses in Eqs. v. Consider. In an endoergic reaction. and solving for EY. I f the rnasaes are not known accurately. 113. respectively. When the last two equations are squared and added.
they may be replaced by the masses of the neutral atoms.(1111). conservation of momentum requires that The part of the kinetic energy of the incident particle needed for excitation of the compound nucleus is But. however. For a reaction induced by Yrays. = m(H1) = 1.008144 amu . The masees which appear in the equations of this section are nuclear masses. As an example of the use of Eq. we must supply some energy in addition to Q.p)017 reaction. The following values of the atomic massea are obtained from Table 111.impart momentum to the compound nucleus. The amount of energy needed for an endoergic reaction is called the threslrold w g p and can be calculated easily. = M(He4) = 4.. I f we let Mc and Vc denote the m a s s and velocity of the compound nucleus. The electrons which must be added to the nuclei to form the neutral atoms cancel in the equation for a nuclear reaction because the number of electrons is the same on the two sides of the equation.003873 amu My = M(Oi7) = 17. m.004529 amu m . (116) consider the N14(a. Mc = Mx + m. for Q to be available for excitation of the compound nucleus. = 0. and the of Q from E threshold energy is just Q. In actual calculations. Section 119. and The threshold energy is then i t can be determined experimentally and the result used to find the value q .m. Consequently. this momentum is then distributed among the products of the reaction.
I t was shown from absorption measurements that the energy of the "photons" should be about 7 Mev. far example.   113 The neutron : alphaneutron reactions. the total energy available for the reaction is 15. On the assumption that the protons were liberated by elastic collisions with ?ray photons. The capture of an aparticle by a nucleus does not always result in the emission of a proton by the compound nucleus.00387. to difficulties. It was assumed that the radiation consisted of ?ray photons produced by the reaction This assumption led. however. When beryllium was bombarded by aparticles. calculations showed that each photon must have had an energy of about 50 Mev. I f the kinetic energy of the aparticle is assumed to be 5 Mev. Inserting the values M x = 9.001282 amu = 1. we get Q = 10. mz = 4. the threshold energy is 1. It is the sum of the Qvalue for the reaction (1112) and the kinetic energy of the incident aparticle minus the recoil energy of the product nucleus. a serious contradiction between the values of the photon energy given by the two methods. the mass of the photon. but these radiations were able to knock protons out of hydrogenous materials and to impart to those protons energies of about 5 Mev. it will now be treated in somewhat greater detail. and no protons were observed.My)c2.04 Mev has been obtained experimentally from energy measurements and agrees very well with the value calculated from the maLws.19 Mev and the reaction is endoergic. one of the products of the reaction seemed to be a very penetrating kind of radiation. There was.. The discovery of the neutron has been discussed briefly in Section 83. The greatest possible energy of . The equation for a nuclear reaction sometimes includes Q.16 f 0. +  since m. therefore. The Qvalue is given by the formula Q = (Mx mz . a value much greater than that deduced from the absorption measurements. The energy that can be attained by the assumed ?radiation can also be computed from known masses and energies.00748. is zero. part of which is retained as recoil energy of the product nucleus.1131 THE NEUTRON: ALPHANEUTRON REACTIONS 265 : The value of Q as calculated from the masses is Q = 0. Although this is not necessary. (11ll). according to Eq.5 Mev. and M Y = 13. A Qvalue of 1.01504. it is a convenient way of showing the energy balance.5 Mev.49 Mev.
therefore. by means of the experiments on the collisions of neutrons with protons and nitrogen nuclei. Chadwick showed that all of the difficulties disappeared when the hypothesis was made that the radiation coming from beryllium bombarded with Qparticles consists of particles (neutrons) with a mass very nearly equal to that of the proton.5 Mev. Chadwick's method was based on the fact that in a collision between a moving and a stationary particle.266 ARTIFICIAL NUCLEAR DIBINTEQRATION [caa~. the velocity imparted to the latter is greatest in a headon collision. one in which the struck particle moves in exactly the same direction as that in which the incident particle approached it. which is again much smaller than 50 Mev. The penetrating nature of the neutrons follows from the absence of a chsrge. The problem of the identity of the products of the bombardment of beryllium was solved by Chadwick. Chadwick also proved that the mass of the neutron is approximately equal to that of the proton. Since the neutron is not a charged particle. less than 15."' as was mentioned in Chapter 8. Suppose that a particle with mass ml and velocity v collides with a stationsry particle of mass ma. I f the energy of the recoil nuclei was assumed to be produced by elastic collisions with photons. and the energies of the recoil atoms in Chadwick's experiments could be completely accounted for on the basis of elastic collisions with an energetic particle of unit mass. while recoiling nitrogen nuclei had energies which led to a photon energy of about 90 Mev. it turned out that the energy that had to be attributed to the photon increased with the mass of the recoil atoma requirement that is contrary to the principles of conservation of energy and momentum in elastic collisions. but having no charge. An expression for the maximum velocity can be derived from the equations of conservation of kinetic energy and momentum in a headon collision. The equation of conservation of energy is . he showed that if these rays were assumed to be %rays then the reaults of the experiments led to values for their energy which depended on the nature of the recoil nucleus. In a aeries of experiments on the recoil of nuclei which were struck by the rays coming from the bombarded beryllium. The nuclear reaction which produces these neutrons is where on1 is the symbol for the neutron. protons ejected from paraffin had energies which led to a photon energy of 55 Mev. 11 the assumed %radiation is. its mass could not be determined from deflection measurements in electric and magnetic fields. and indirect methods had to be used. For example. and let the velocity of the incident particle after the collision be vl and that of the struck particle be va.
7 X lo8. This result.3 X 10' and v3 = 4. Taking ma and ma to be the masses of the proton and nitrogen nuclei. (1116) by Eq. its mass ml can be calculated from Eq. and v3 are known.3 X 10' ml 1 v3 4. the mass ml of the neutron can be found.15. ma. (1116) if the mass m2 and the maximum speed 02 of the recoiling nucleus are known. equal to 1 amu and 14 amu respectively. however. we get from Eq. we get  I f ma. and with the corresponding values v2 = 3. the maximum velocity imparted to the latter is given by  where ml and v are the same as in Eq. The recoils of nitrogen nuclei after being struck by the same neutrons were observed in cloudchamber experiments and their maximum velocity was found to be 4. I f a neutron with the same velocity v collides with a third particle of mass m3.7 x 108' ml + + and ml = 1. The maximum velocity imparted to hydrogen nuclei ejected from paraffin wax by the neutrons emitted by beryllium bombarded with polonium aparticles was found to be 3. did not have reliable information about the speed of the neutrons and had to use a less direct method.  I f the incident particle is a neutron with known velocity v.1131 THE NEUTRON: ALPHANEUTRON REACTIONS 267 and the equation of conservation of momentum is The velocity v l can be eliminated from the last two equations. Chadwick. and a relationship is obtained between v2 and v. (1116). Upon dividing Eq.3 X lo9 cm/sec. showed that the mass of the neutron is roughly the same as that of the proton.7 X los cm/sec. (1117). although approximate because of errors in the determination of the maximum recoil velocities. Chadwick obtained a somewhat better value of the neutron mass by considering the reaction . v2. (1118) 14 = 5 = 3.
deuterons. Furthermore. (116). however. and of the neutron (estimated from the maximum recoil energy of a proton after being struck by a neutron). He4. might yield interesting results when used as projectiles. (1113). [15P31]. [13A127]. It will be seen in the next chapter that the products of these nuclear reactions are sometimes radioactive. Among these are When these reactions are compared with those mentioned in Section 111.005 and 1. 11 The Qvalue was found from the kinetic energies of the incident aparticles. as will be seen later. and even 7rays.and A27. and it was . e.008983 amu.68 Mev (RaC'). and transmutations by these particles were possible only with the lighter elements. protons.008 atomic mass units. giving a different product in the two cases.. The development of the field of nuclear transmutation was limited. it seemed possible that other particles. and N14. The proof of the existence of the neutron and the development of methods for analyzing nuclear reactions in which one of the products is a neutron led to the discovery of other reactions similar to that represented by Eq.268 ARTIFICIAL NUCLEAR DISINTEGRATION [CHAP. of the nitrogen nuclei (measured in cloudchamber experiments). In this way. [QF"].emit either a proton or a neutron when bombarded with aparticles. From the Qvalue and the known atomic masses of B". The early artificial disintegrations were produced by bombarding materials with aparticles from natural radioactive substances. such as N". The compound nucleus. 1 1 4 The acceleration of charged particles. The best methods now available for determining the neutron mass give 1. because aparticles could be obtained only in beams of low intensity and with energies not greater than 7. Chadwick found a value between 1. i t is seen that some atoms. NaZ3. the mass of the neutron can be obtained with the aid of Eq. may disintegrate by means of either process. The neutrons from the bombardment of beryllium with aparticles were also used as projectiles and. Protons and deuterons with a single positive charge would experience smaller repulsive Coulomb forces than aparticles in the field of the nucleus. have produced many artificial disintegrations.g.
a large variety of nuclear reactions can be made to occur. to develop laboratory methods of accelerating charged particles to high energies. The further development of these machines and the design of new accelerators such as the frequencymodulated cyclotron and the synchrotron made it possible to obtain particles with energies up t o hundreds of Mev. and several successful methods were invented about 1930. highly energetic xrays are produced. It became important. For the purposes of the present chapter. The particle accelerators used to produce projectiles for nuclear reactions will be discussed in Chapter 21.7 Mev. The particles were proved to be aparticles by photographing their tracks in a cloud chamber. The cloudchamber pictures showed that two aparticles leave the  .1151 TRANSMUTATION BY PROTONS 269   thought that this effect might make disintegrations possible with lowenergy protons. By the suitable choice of a machine and its operating conditions. and the cyclotron of Lawrence and Livingston. With the available particles and accelerators. which can also be used as projectiles. and recently built machines have raised the energies to 25 Bev. Among the earliest particle accelerators were the voltage multiplier of Cockcroft and Walton. Chapter 20)) have supplied highly intense sources of neutrons for nuclear bombardment. and that these projectiles can cause many different kinds of nuclear reactions. Finally.5 billion electron volts. and the ideas involved can then be extended to the more complex reactions which occur with particle energies of the order of 100 Mev. protons. When these electrons strike a metal target. Electrons can be accelerated to energies up to 300 Mev in a betatron.1 to 0. it is enough t o realize that various bombarding particles of almost any desired energy can be obtained with the help of suitable devices. Cockcroft and Walton'" bombarded lithium with protons accelerated to energies of 0. the discovery of nuclear fission and the development of chainreacting piles. Scintillations caused by particles ejected from the lithium were observed on a zinc sulfide screen placed a short distance away.'6' and the lithium isotope that was disintegrated was proved to be Li7. therefore. These reactions will serve to show the most important features of artificial disintegration. or aparticles with energies as high as several Mev could be obtained by 193G. The rest of this chapter will be limited t o the commonest reactions which take place with particle energies up to about 20 Mev. The "Cosmotron" a t the Brookhaven Kational Laboratory has produced protons with energies of 2. The first case of a nuclear disintegration brought about entirely by artificial means was one for which protons were used as the projectiles. 115 Transmutation by protons. the electrostatic generator of Van de Graaff. or nuclear reactors (cf. deuterons. or in a linear accelerator.
from Table 111. 11 point of disintegration and proceed with equal energies in opposite directions. obtained from the energies of the incident protons and the emergent aparticles.018222 arnu.34 Mev. The proof will be repeated here.l ~ ~ 2 ~ ~e4. many nuclear transformations have been studied in d e w and the results invariably agree with the relationships deduced from the Einstein equation.33 Mev. +2~e4. . I t was a good reaction for this purpose because the energies of the products could be measured precisely. The value of Q obtained from the masses is Q . The best experimental value of Q.n o ARTIFICIAL NUCLEAR DISINTEGRATION [CHAP. and that there was a genuine release of energy from the lithium atom a t the expense of its mass. 4 ~ e+ 8 . The disintegration that has just been discussed is an example of the general type z ~ + A 1 ~+ 1 [ z + l ~ n A + l+ ] z . agrees with experiment.01860 arnu = 17. [ 5 ~ 1 0 ] 3Li6 + 2He4. with newer values of the various quantities. + 1H1 + [&12] + 4 ~ e + 8 2~e4. Since 1932. The values of the atomic masses are. The reaction may be represented by the equation This reaction has a certain historical interest because it provided one of the earliest quantitative proofs'7' of the validity of the Einstein masse n e k relationship. + Other examples are a ~ i 6 lH1 + [4Be7]+ 2He3 + e ~ l+ e lH1 l s ~ l+ " +H1 a ~ l l + 4 ~ e e lH1 + + + + 2He4. = 0. is 17. [14~i28] 1 2 ~ g Z +4 2He4. [ l o ~ e 2 0 ] 8016 + 2He4. Mx = M(Li7) = 7. + + + An interesting reaction occurs when the target is B". This agreement shows clearly that the theoretical expression for Q.He4 Besis highly unstable and breaks up into two more aparticles. and the masses were known. based on the massenergy relationship.
13 Mev. respectively."' The source of the protons was an electrostatic generator. The Qvalue is obtained from Eq. Proton bombardment can also result in nuclear reactions in which one of the products is a neutron.164 f 0. The reaction is of the type The effect of the transmutation is to increase the charge on the nucleus by one unit. the mass number is not changed. but becomes less unstable by emitting a "/ray photon rather than a neutron or an aparticle. + (1121) and some examples are + 6C12 + + 13~12 + 7 3~i7 Q ~ l Q 1 ~ + 1 1 ~ + 1 + + [7N13] + Y. 2 9 ~ ~ 5 8 + + on1. Then Q = E th X = 2. The experimental results in the neighborhood of the threshold energy were then analyzed. In some cases the bombarding proton is simply captured by a nucleus. [ 1 4 ~ i 2 8 ] 14si2' + Y. so that the reactions are endoergic.01 Mev.Be8] + 4 ~ e 8 Y.H' . and the value of the threshold energy was found to be Eth = 2.164 X = 2. The compound nucleus which is formed is again unstable. may be replaced by the mass numbers 65 and 1. [29~~5Q] + on1. moving it one place to the right in the periodic table. and the reaction may be regarded as a case of multiple particle production. the mass change is usually negative. The reaction is of the type zxA + 1 ~ + 1 [z+l~n*+l] + Z + ~ Y ~ + ' Y. In the (p. Examples of this reaction are 5 ~ 1 1 + + + [6~12 + ] BC" [Q~'Q + ] QF" + + + 28~i58 + + + on1. [ 3 0 ~ n 6 6 ] 30zns5 + on1  is an example of a case for which the threshold energy has been measured. The reaction 2QCue5 . and the relative number of neutrons produced was determined as a function of the energy of the bombarding protons.n) reactions.1151 : TRANSMUTATION BY PROTONS 27 1 The final result of the bombardment is three aparticles. [. [ l o ~ e 2 0 ] l o ~ e 2+ 0 Y. (11ll). ~ ' + 7 ~ + + + . in which the masses M X and m.
The general (d.272 ARTIFICIAL NUCLEAR DISINTEGRATION [CHAP. far more energetic than the 2. be used to produce nuclear disintegrations. the deuterons have been accelerated up to energies of several Mev in a cyclot. In most of these cases.2 Mev. [ I ~ s ~ ~ ] + [?N"] + 6 c 1 3 * + 15p3' 1 ~ ' . + + Other examples of (d. + + The Qvalues for these reactions are usually positive.~e'] + 2He4 2~e4. leaving the charge unchanged. such as + iH2 l l ~ a 2+ 3 + 6 ~ 1 2 + 1 ~ + 2 ~+ ' 15p31 1 + 1H1. A great many nuclear reactions have been observed with high energy deuterons as the bombarding particles. so that the reactions are exoergic.a) reactions are The mass change is usually positive so that the Qvalues are positive. Examples of this reaction are 116 Transmutation by deuterons. .ron or in an electrostatic generator. 11 The %ray photons emitted in these reactions are often very energetic and can. in turn. The bombardment of lithium yields photons with an energy of 17. disintegrations produced by deuterons do not always yield aparticles. The general reaction of this type is As in the case of proton bombardment. Deuteronproton reactions are often found. One of the first deuteroninduced reactions studied was again that on a lithium atom.p) reaction is (1123) zxA 1 ~ + 2 [ z + ~ c ~ ~ ++ zx ~A ] + l 1H1.'g' 3~i6 1 ~* 2 [. The result of these transformations is t o increase the mass of the nucleus by one unit. Another type of disintegration caused by protons is that in which deuterons are produced. and the reactions are exaergic. [ l 2 ~ ~ l l~~ a ~ 2+ ] 4 lH1.6Mev photons which are the most energetic available from the natural radioactive nuclides.
the average energy of the neutrons is reduced to a few hundredths of an electron volt. but slowly moving neutrons are also extremely effective. 117 Transmutation by neutrons.he development o f the chainreacting pile. In the first. The helium isotope of mass 3 is stable and is found in nature. and has been given the name tritium. deuterons. Before t. a proton and a new isotope of hydrogen are formed. Neutrons have proved to be especially effective in producing nuclear transformations. 1 ~ 2 1 ~ + 2 [2~e4 + ] 1 ~ 31 ~ 1 . or nuclear reactor.n)He3. as in the following reactions: The general (d. Since they have no electric charge. or aparticles. Not only are highly energetic neutrons capable of causing nuclear reactions. they are not subject to repulsive electrostatic forces in the neighborhood of a positively charged nucleus. One of the most interesting cases of deuteron bombardment is that in which the target contains deuterons.BeQ(d. These slow neutrons  . the main sources of neutrons were reactions such as H2(d. The new hydrogen isotope has a mass very nearly equal to 3 atomic mass units. Both the (d.n) reactions have been observed.n) reaction is represented by the equation .  The excited compound nucleus [2He4]can disintegrate in two ways.n)BIO and BeQ(a. A neutron gives up a large fraction of it.p) and the (d. a neutron and an isotope of helium are formed. after many collisions.TRANSMUTATION BY NEUTRONS 273 : Neutrons are often produced as a result of deuteron bombardment. Deuterium targets have been made by freezing deuterium oxide (D20"heavy water") onto a surface kept cold by liquid air. in the second. Because of these properties of the neutron many more disintegrations have been produced with neutrons than with any other particle.n)C12. The fast neutrons produced were slowed down by allowing them to pass through some hydrogencontaining substance such as water or paraffin. 1 ~ + + 2 1 ~ + 2 [2He4]+ + 2 ~ e+ 3 on1. and are therefore more likely to penetrate nuclei than are protons. I t is unstable and has a halflife of about 12 years.s energy in a collision with a hydrogen nucleus and.
2n) reaction leaves the charge of the nucleus unchanged and decreases the mass number by one unit. In one method. . but the charge is decreased by one unit and the atom is moved one place to the left in the periodic table. a proton. 11 have been found to be especially useful from a practical standpoint. one neutron is captured by the nucleus and two neutrons are emitted. The reaction between a neutron and a nucleus gives rise in most cases t o an aparticle. with heavier nuclei.[ 3 0 ~ n e 5 + ] . a ?ray photon.p) reaction is described by the equation The effect of this reaction is to replace a proton in the nucleus by a neutron. which is detected by the ionization it produces in the chamber. the mass number is not changed.274 ARTIFICIAL NUCLEAR DISINTEGRATION [CHAP. as in the case of AlZ7 and in the reaction . The (n. or to two neutrons. The ( n p ) reaction(lO'is represented by the general equation Among the most interesting examples of this reaction are The first two reactions have a relatively high yield and are therefore often used to detect neutrons. more energetic neutrons must be used. and their interactions with nuclei are particularly interesting. The capture of a neutron by an atom of the B1° isotope causes the liberation of an aparticle. the compound nucleus formed by the capture of a neutron emits a proton.. In some cases. usually in the form of a compound. The (n. The result is an isotope of the target nucleus with a mass number one unit smaller.cua4 'HI. an iodization chamber is lined with boron. Some examples of this reaction are The first of these reactions can be induced by slow neutrons. 3 0 ~ n 6 4 on'  + + In another type of reaction.
1181
TRANSMUTATION BY PHOTONS
275
.
An example of this reaction is
The mass change in the (n,2n) reaction is always negative; the Qvalue is negative and fast neutrons are needed to bring about this reaction. The commonest process which results from neutron capture is radiative capture," l' represented by

The compound nucleus emits one or more %ray photons, and the final nucleus is an isotope of the target nucleus with a mass number one unit ) has been observed in nearly all of the elements. greater. The ( n , ~ process The Qvalues are always positive, the excess energy being carried away by the %rays. The simplest ( n , ~ ) reaction with slow neutrons occurs with hydrogen as the target nucleus,

The product of the reaction is deuterium. When deuterium is bombarded with slow neutrons, tritium is formed,
Other typical ( n , ~ reactions ) are
+ + 4Q~n115 + on1 + [4Q~n116] 491n116 + 7, + on1 + [ Q 2 ~ 2 3 Q ] + Y.
1~~ 1] 3 ~ 1 ~ ' Y, 13~12' on1 + [ 1 3 ~ ) . +
+
Q2~238
+ 9
2
~
~
~
'
The radiative capture of slow neutrons often results in product nuclei which are radioactive, and this reaction is one of the most important sources of artificial radioactive nuclides.
118 Transmutation by photons. Atomic nuclei can also be disintegrated by bombardment with highenergy photons, a process which is usually called photodisintegration. Since the photon has no mass, it can supply only its kinetic energy to a nuclear reaction. This energy must be a t least as great as the binding energy of a nuclear particle before such a particle can be ejected from a nucleus. Photodisintegration reactions are, therefore, endoergic and usually have threshold energies of the order of 10 Mev. With only two exceptions, photodisintegration does not occur with 7rays from natural radioactive substances. These exceptions are the deuteron, which has a binding energy of only 2.2 Mev, and the nuclide 4BeQ,

276
ARTIFICIAL NUCLEAR DISINTEGRATION
[CHAP.
11
in which one neutron is loosely bound. In the case of the deuteron,(12' the reaction is 1H2 Y [ 1 ~ 2+ ] 1H1 On1,
+
+
+
and is the reverse of the radiative capture of a neutron by a proton. In beryllium, the reaction is 4~e'
+r
+
[&eQ] + 4Be8
+ on1,
with a Qvalue of 1.67 Mev. I t will be recalled from Section 115 that when Li is bombarded with protons, 17Mev Yrays are produced. These have been used successfully to cause the photodisintegration of other nuclides, as in the (Y,n) reaction
The ( 7 , ~ ) reaction requires still higher energies, and has been observed with highenergy photons from a betatron.
119 Nuclear chemisty : nuclear masses. The transmutations which have been discussed are the commonest examples of a large variety of nuclear reactions brought about by charged particles, neutrons, and photons. Many other reactions have been discovered and analyzed. Some have not been considered here because they are relatively rare; others have been omitted because they take place only a t very high energies, and their importance and interest lie outside the limits of elementary nuclear physics. I t has been seen that nuclides can be transmuted in a number of different ways. A given nuclide can be transformed to any one of several other nuclides, the nature of the newly formed substance depending on the particle used to produce the transmutation and upon the particle given off during the reaction. For example, the nuclide 13A127has been made to undergo the following transformations.


.
1191
NUCLEAR
CHEMISTRY: NUCLEAR
MASSES
277

The reactions numbered (lo), (15), and (16j are included to give some idea of transmutations which are possible a t higher energies. The protons in (10) and the ?rays in (15) and (16) have energies of about 50 Mev. At still higher energies, still more particles are given off and the term atom smashing is really appropriate. In the 16 reactions listed, 11 different nuclides are formed. One nuclide, llNa24, is formed from 13A127in 4 different ways, in reactions (6), (lo), ( l l ) , and (16). It can also be formed in 5 more reactions.
Thus, the nuclide 1Na24 can be formed in a t least 9 different ways. These examples show the tremendous variety of nuclear reactions which can be induced with the nuclides, particles, accelerators, and other sources of projectiles now available. The study of these reactions is one of the great branches of nuclear physics and because of the problems involved in separating and identifying the products, this field is aptly called 'buclear chemistry. " One of the important applications of nuclear reactions is in the determination of nuclear, or atomic, masses, a problem which has a direct analog in chemistry. The determination of atomic masses by means of the mass spectrograph has been discussed in Chapter 9 and the results obtained by that method can now be combined with values obtained from nuclear reactions. According to Eq. (116), a Qvalue can be expressed
Nuclide Atomic mass Atomic mass

Nuclide
Nuclide MgZ3 MgZ4 MgZ5 Mg26 MgZ7
~ 1 2 5
Atomic m a s s
~ 1 2 7
A128 SiZ7 Si28 Si2Q Si30 Si3 1
P28
P30 P3 P32
s32
s33 s34
s35
C135 c13a C137
'From Mattauch and Everliig.(4')
as the difference between the masses of the reactants and products of a nuclear reaction. Many Qvalues have been measured with high precision in recent years,"3114) and it is now possible t o obtain the masses of light nuclei directly in terms of the mass of 0 1 6 without the use of any massspectroscopic result^."^*'^' The most recent compilations of atomic masses combine massspectroscopic and Qvalue data. Nuclear reactions are, of course, a n essential source of information concerning the masses of nuclides which do not occur naturally and can be obtained only in nuclear reactions. The atomic masses of nuclides"" up t o C13' are listed in Table 111. Both naturally occurring and artificially produced nuclides are included in the table. The uncertainty, given in parentheses, is expressed in units of lo' amu.
GENERAL

RUTHERFORD, CHADWICK, and ELLIS,Radiations from Radioactive Substances. New York: Macmillan. 1930, Chapter 10 (reprinted with corrections 1951). M. S. LIVINGSTON and H. A. BETHE, "Nuclear Physics, C. Nuclear Dynamics, Experimental," Revs. Mod. Phys. 9, 245390 (July 1937). J. D. STRANATHAN, The "Particles" o f Modern Physics. Philadelphia: Blakiston, 1944, Chapters 10, 11. R. T. BEYER, ed., Foundations o f Nuckar Physics; Facsimiles of Thirteen Fundamental Studies as they were originally reported in the Scientific Journals, with a Bibliography. New York: Dover, 1947. R. D. EVANS, The Atomic Nucleus. New York: McGrawHill, 1955, Chapters 12, 13.
PARTICULAR

1. E. RUTHERFORD, "Collision of aParticles with Light Atoms, Part IV." An Anomalous Effect in Nitrogen, Phil. Mag.37, 581 (1919). (In Beyer.) 2. E. RUTHERFORD and J. CHADWICK, "Artificial Disintegration of Light Elements," Phil. Mag.42,809 (1921); 44,417 (1922); Proc. Phys. Soc. (London) 36, 417 (1924). "Photography of Artificial Disintegration Collisions 3. P. M. S. BLACKETT, and Accuracy of Angle Determinations," Proc. Roy. Soc. (London) A134, 658 (1932). 4. J. CHADWICK, "The Existence of a Neutron," Proc. Roy. Soc. (London) A136, 696 (1932). (In Beyer.) 5. J. D. COCKCROFT and E. T. S. WALTON, "Experiments with High Velocity Positive Ions." 11. "The Disintegration of Elements by High Velocity Protons," Proc. Roy. Soc. (London) A137, 229 (1933). 6. P. I. DEEand E. T. S. WALTON, "Transmutation of Lithium and Boron," Proc. Roy. SOC. (London) A141, 733 (1933).
280
44, 123 (1933).
ARTIFICIAL
NUCLEAR DISINTEGRATION
[CHAP. 11
7. K. T. BAINBRIDGE, "The Equivalence of Mans and Energy," Phys.
Rar.
8. SHOUPP, JENNINGS, and JONES, "Threshold for the ProtonNeutron Reaction in Copper," Phys. Rev. 73, 421 (1948). 9. LEWIS,LIVINGSTON, and LAWRENCE, "The Emission of aParticles from Various Targets Bombarded by Deuterons of High Speed," Phys. Rev. 44, 55 (1933). 10. N. FEATHER, "Collisions of Neutrons with Nitrogen Nuclei," Proc. Roy. Soc. (London) A136, 709 (1932). 11. FERMI,AMALDI, D'AGOSTINO, RASETTI,and SEQRE,"Artificial Radioactivity Produced by Neutron Bombardment," Proc. Roy. Soc. (London) A146, 483 (1934). 12. J. CHADWICK and M. GOLDHABER, "Nuclear Photoelectric Effect," Nature 134, 237 (1934); Proc. Roy. Soc. (London) A151, 479 (1935). 13. W. W. BUECHNER, "The Determination of Nuclear Reaction Energies by Deflection Methods," Progress in Nwlear Physics, 0. R. Frisch, ed., Vol. 5, p. 1, 1956. 14. D. M. VAN PATTERand W. WHALING, "Nuclear Disintegration Energies." I, Revs. Mod. Phys., 26, 402 (1954); 11, ibid., 29, 757 (1957). 15. LI, WHALING, FOWLER, and LAURITSEN, "Masses of Light Nuclei from Nuclear Disintegration Energies," Phys. Rev. 83, 512 (1951). 16. C. W. LI, "Nuclear Mass Determinations from Disintegration Energies: Oxygen to Sulphur," Phys. Rev. 88, 1038 (1952). 17. J. M a r r a u c ~ and F. EVERLING, "Masses of Atoms of A < 40," Progress in Nuclear Physics, 0. R. Frisch, ed., Vol. 6, p. 233 (1957).
1. Complete the reactions listed below. Rewrite them in a way which shows the balance of atomic and mass numbers, and indicate the compound nucleus in each case: H1(n,7) ;H2(n,7); Li7(p,n) ;Li7(p,a) ;Be9(p,d); Beg(d,p) ; Be9(p,a); Bl1(d,a); CI2(d,n); N14(a,p); N15(p,a); 016(d,n). 2. Calculate the Qvalues of each of the reactions of Problem 1. Which reactions are esoergic? Which are endoergic? Use the masses listed in Table 111. 3. In the four reactions which follow, the mass of the target nucleus A1 is given as 26.99007 amu, and the Qvalue of each reaction is given. Calculate the mass of the product nucleus.
(a) (b) (c) (d)
A127(n,7); A12'(d,p); Ai27(p,a); AI2'(d,a) ;
Q Q Q Q
= = = =
7.722 Mev 5.497 Mev 1.594 Mev 6.693 Mev

4, One of the reactions which occurs when boron is bombarded with 1.510Mev deuterons is B"(dJa)Be9. The aparticles coming off a t an angle of 90" with the direction of the deuteron beam have an energy of 6.370 Mev. What is the Qvalue of the reaction? 5. Under the conditions of Problem 4, the reaction B11(d,p)B12 was also found to occur, and to have a Qvalue of 1.136 Mev. What is the energy of the protons observed a t 90°? 6. The reactions Be9(p,n)Bg, C13(p,n)N13, and 018(p,n)FlS were found to have the threshold energies 2.059 Mev, 3.236 Mev, and 2.590 Mev, respectively. What are the Qvalues for the reactions? 7. Calculate, from the atomic masses, the threshold energies for the reactions B1l(p,n)C1l, 018(p,n)F18, and Na23(p,n)Mg23. 8. Write ten nuclear reactions which might be expected to take place with C12 as the target. Which of these reactions would be expected to be exoergic? Endoergic? 9. Write ten nuclear reactions which should give the product nucleus C12. Which reactions should be exoergic? Endoergic? 10. A sample of silicon is bombarded with 1.8Mev deuterons. Show that the following reactions can produce charged particles with energies greater than 4 Mev: Si28(d,p)Si29;Si29(d,p)Si30;Si29(d,a)A127;Sia0(d,p)Si3l; Si30(d,a)A128. 11. The reactions listed below have been found to have the iqdicated Qvalues (a) 016(d,a) N14; Q = 3.112 Mev (b) N14(d,p) NI5; Q = 8.615 Mev (c) N15(dJa) C13; Q = 7.681 Mev (d) C13(d,a) Dl1; Q = 5.160 Mev (e) Bll(p,a) Bes; Q = 8.567 Mev Q = 0.089 Mev (f) Bes(,),;
282
ARTIFICIAL NUCLEAR DISINTEGRATION
[CHAP. 11
Show that these data lead to a value of 4.003869 amu for the mass of He4 if the mass of deuterium is taken as 2.014723 amu. 12. Suppose that, in separate experiments, C12 is bombarded with neutrons, protons, deuterons, and aparticles, each having kinetic energy of 10.00 Mev. (a) What would be the excitation energy of the compound nucleus in each case if i t is assumed that the recoil energy of the compound nucleus can be neglected? (b) What is the recoil energy of the compound nucleus in each case when the recoil energy is not neglected? 13. With the aid of the data of Table 111, calculate the biding energy of the "last neutron" in each of the following sets of isotopes: (a) C12, C13, C14; (b) N14, N16, Nl6; (d) Ne20, Ne21, Ne22. (c) 016,017, 0"; 14. With the aid of the data of Table 111, calculate the biding energy of the last proton in each of the following sete of isotopee: (a) C", C12, C13; (c) 014,016, 016,0;'; (b) N13, N14, N16; (d) Nelg, Neao, Ne21.
15. What hypotheses would you propose to account for those value of the binding energy in Problems 13 and 14 which are particularly small, i.e., smaller than 5 Mev? 16. Use the values of the mass diflerences: neutronproton, and Be7Li7 to determine the minimum proton energy needed to obeerve the reaction Li7 (p,n)Be7. 17. Calculate the diierence between the msases of A128 and Sia8 from the following data: Al2?(d,p)AlZ8, Q = 5.48 Mev; Si28(d,p) S i 2 @ , Q = 6.25 Mev; Si2g(d,a)~127, Q = 5.99 Mev; 2H1  HZ = 1.584 rnillimass units; 2H2 He4 = 25.600 millimass unite. 18. From'the nuclear mssses involved, calculate the range of energy of the neutrons produced when Be is bombarded with 5.30 Mev aparticles from Po210. 19. Derive Eq. (1110).

CHAPTER 12
ARTIFICIAL RADIOACTMTY
121 The discovery of artificial radioactivity. The discovery by Curie and Joliot, in 1934, that the products of some induced nuclear transmutations are radioactive opened a new era in nuclear physics. The study of the radioactivity of the artificially made atomic nuclei has done much t o clarify concepts of the constitution and stability of nuclei. Many of the nuclear reactions discussed in the last chapter yield radioactive nuclides and about 1200 radioactive species have been made up to 1058. These nuclides decay spontaneously according to the same laws that govern the disintegration of the naturally occurring radioactive elements, and the term arti$cial radioactivity refers to the way in which the new radionuclides are produced rather than to their decay. Artificial radioactivity was discovered by Curie and Joliot while they were studying the effects of aparticles on the nuclei of light elements. When boron, magnesium, and aluminum were bombarded with aparticles from poIonium, protons and neutrons were observed, as expected from the known (alp) and (a,n) reactions. In addition to these particles, positive electrons or positrons were also observed."' The positive electron is a fundamental particle whose rest mass is the same as that of the electron, and whose charge has the same magnitude but opposite sign from that of the electron. It was discovered by Ander~on'~' in 1932 in the course of a study of photographs of cosmicray tracks in a Wilson cloud chamber. In the presence of a magnetic field, some tracks were seen which could have been caused only by particles of electronic mass and charge, but from the direction of the curvature of these tracks it was evident that the particles producing them must have been positively charged. Anderson called these particles positrons. Throughout the rest of this book the word electron will be used to denote the negative electron and the word positron will be used for the positive electron. This wage is common and more convenient than the use of the terms "negative electron" and "positive electron." The word "negatronn which has been suggested for the negative electron has not been generally adopted. The source of the positrons observed by Curie and Joliot was not clear a t first and several hypotheses were suggested which proved to be incorrect. For example, it was suggested that the positrons might result directly from the disintegration by the aparticles, with the formation of a neutron and instead of the usual proton. Subsequent experiments by Curie +owed, however, that the lightelement targets continued to
283
.

284
ARTIFICIAL RADIOACTIVITY
[CHAP.
12

emit positrons after the source of the aparticles had been removed; the rate of emission of the positrons gradually decreased with time, finally approaching zero. When the rate was plotted against the time after the removal of the aparticle source, an exponential curve was obtained for each target similar to those obtained in the pdecay of natural radionuclides. The act,ivity could be expressed accurately by the equation A = AoeXf,and characteristic halflives were found, namely, 14 min, 23 min, and 34 min for boron, magnesium, and aluminum targets, respectively. The explanation offered by Curie and Joliot was that the product nucleus formed in the (a,n) reaction in each case was an unstable nuclide, which then disintegrated with the emission of a positron. The nuclear reactions when boron and aluminum were the targets were postulated to be
7 ~ 1 3 +
&I3
+ lea; + lea;
T
= 14 min,
l s ~ 30, 1 4 ~ i 3 0
T = 36 min.
The symbol or +leo is used to denote the positron, since its charge is the same as that of the proton, and its mass number is zero. The halflife of N13 was later found to be 10 min and that of P30 23 min, rather than 14 min and 34 min, respectively, as first reported by Curie and Joliot. The nuclides N13 and P30 are not stable, as can be verified from Table 91, but the nuclides which remain after the emission of the positron are stable. The explanation offered by Curie and Joliot was tested chemically"' by separating the target element from the product element; radioactivity was found only in the latter. In the boron reaction, the target, boron nitride BN, was irradiated with aparticles for several minutes and then heated with caustic soda, liberating all the nitrogen as gaseous ammonia. The ammonia was found to have all the radioactivity. The halflife was the same as that found with other boron targets, and was not found with nitrogen targets. Hence, the boron must have been changed into a radioactive isotope of nitrogen. An aluminum target, after being irradiated with aparticles, was dissolved in hydrochloric acid, liberating any phosphe rus present as the gas phosphine (pH3), while the aluminum remained behind in the solution. The solution was then evaporated to dryness and the residue tested for positron activity. No activity was found. The gas, which contained the phosphorus, showed the characteristic positron activity, and the activity was therefore associated with an isotope of phosphorus. These tests provided the first definite chemical evidence forJhe artificial transformation of one element to another. The ~ h e @ 2 ' , & ~ &
.
1221
THE ARTIFICIAL RADIONUCLIDES
285
fication of the product nucleus, together with the determination of the nature of the ejected particles by measurements of ionization in a cloud chamber and of the deflection in a magnetic field, proved the correctness of the equations proposed for the disintegrations. After the discovery that the bombardment of light nuclides with aparticles can lead to radioactive products, it was found that nuclear reactions induced by protons, deuterons, neutrons, and photons can also result in radioactive products. As in the case of the natural radionuclides, an artificial radionuclide can be characterized by its halflife and by the radiation it emits. These properties have helped greatly in determining the chemical nature of the products of nuclear reactions. When the products are radioactive, they can be traced in chemical reactions by means of their characteristic halflives or disintegration products. Otherwise they could not be traced because of the very small amounts in which they are present, and their chemical nature could only be inferred from indirect physical evidence, such as mass and energy balances. The special branch of chemistry that deals with these and similar problems is called radiochemistry, and has become an important part of nuclear science. By radiochemical methods, the active products of nuclear reactions can be separated in forms suitable for the study of their properties. It is possible to determine the element with which a particular product is isotopic, and to assign a mass number. The latter assignment often depends on crossbombardment, that is, the particular nuclide is obtained by several different nuclear reactions, from which the mass number of the given active nuclide can be inferred. When the nuclide has been obtained in suitable form, its decay rate can be determined, as well as the properties of its radiations. About 1200 artificially radioactive nuclides have been identified and their properties determined by combined chemical and physical techniques.

122 The artificial radionuclides. Electron and positron emission. Orbital electron capture. The constitution of the known nuclides, both natural and artificial, can be shown by means of a graph in which the number of neutrons is plotted against the number of protons, as in Fig. 121. The stable nuclides are indicated by black squares (as in Fig. 99). Alphaemitters are indicated by open triangles and squares; the latter are &table, while the former may also undergo pdecay. Electronemitters are indicated by open circles and positron emitters by crosses. A cross in an open circle indicates that the nuclide decays by 8emmion, @+emission, and electron capture. The diagonal lines are lines of constant mass number, A , with A = Z N; nuclides which lie on the same diagonal line are called isobars because they have the same mass number and, therefore, very nearly the same atomic weight. For the most part, the
+
286
ARTIFICIAL RADIOACTIVITY
[CHAP. 12
Proton number (2)
FIG. 121. Chart of the known nuclei. (Reprinted by permission from S. E. Liverhant, Elementcrry Introduction to Nzlclear Re& Phyaies. New York: Wiley, 1960)
1221
THE ARTIFICIAL RADIONUCLIDES
287
.


artificially radioactive nuclides lie above or below the stable nuclides; they have either too many neutrons (not enough protons) or too few neutrons (too many protons) to be stable. A nucleus with too few protons tends to become stable by increasing the nuclear charge, while a nucleus with too many protons tends to become stable by decreasing the nuclear charge. One way by which the nuclear charge can be increased is by electron emission, as was seen during the treatment of natural radioactivity. The nuclear charge can be decreased by emission of a positron. Consequently, positron activity is usually seen in the isotopes of an element having mass numbers smaller than those of the stable isotopes, while electron activity usually occurs in isotopes with mass numbers greater than those of the stable isotopes. The charge on the nucleus is also reduced if the nucleus captures an orbital electron. The extranuclear electrons, in the course of their motions, often approach close to the nucleus and, according to wave mechanics, may even penetrate it. The electrons which are most likely to do this are the Kelectrons in atoms with a large value of 2. These electrons may be thought of as available for capture by the nucleus, especially if the latter is protonrich. If an electron from the Kshell is captured, the process is called Kcapture; less often an Lelectron is captured. The vacancy in the Kshell or Lshell is usually filled by an electron from an outer shell with the emission of a K or L xray, respectively, characteristic of the product nucleus. In this process, no charged particle is emitted and the process can be observed only because of the xray emission. Orbital electron capture is sometimes accompanied by the emission of electrons from the extranuclear structure. These electrons, called Auger electrons after their discoverer, result from what might be described as an internal photoelectric effect. For example, the emission of a K xray may be replaced by the ejection of an Lelectron with a kinetic energy equal to the difference between the K xray energy and the binding energy of the Lelectron. The three types of disintegration are isobaric transformations because they involve changes in nuclear charge but not in mass number. As in the case of natural radioactivity, some artificially radioactive nuclides emit 7rays as well as electrons or positrons. About onethird of the active nuclides are simple in the sense that they give off only particles; the remainder also emit w a y s . The question of whether an artificial nuclide will decay by electron emission, positron emission, or orbital electron capture can be discussed in terms of the energy available for the disintegration. Consider first the case of electron emission, in which the reaction is
288
ARTIFICIAL RADIOACTIVITY
[CHAP.
12
The energy balance for this reaction is given by the expression
Q/c2 =
Afn(zXA)  M n ( z + l Y A )  m,
(1221
where the M,'s represent the nuclear masses of the artificial nuclide and its decay product, and m is the electron mass. The last equation can be written in terms of the atomic masses if it is noted that
M a ( z X A )= M n ( z X A )
+ Zm,
where Ma is the atomic mass; the binding energy of the electrons is then included but leads t o no difficulty because it always cancels. Then Eq. (122) becomes
The condition that electron emission be energetically possible is that the value of Q be positive, or
i.e., the atomic mass of the artificial nuclide must be greater than that of its isobar with nuclear charge one unit greater. In the case of positron emission, the reaction is

and the Qvalue is given by
For positron emission to be energetically possible, the atomic mass of the artificial nuclide must be greater than the atomic mass of its isobar with nuclear charge one unit smaller by a t least 2 electron masses, or 1.02 Mev,
The reaction in the case of orbital electron capture is
1221
THE ARTIFICIAL RADIONUCLIDES
and the energy available is
The condition that orbital electron capture be energetically possible is therefore M,(zXA) > M&iYA), (1210) i.e., the mass of the artificial nuclide must be greater than that of its isobar with nuclear charge one unit smaller. Comparison of Eqs. (127) and (1210) shows that the capture of an electron by a nucleus is an energetically more favorable process than the emission of a positron. But the capture of even a Kelectron depends on the small probability that the electron be very close to, or "within," the nucleus. The result is that as soon as the energy available exceeds 2mc2 positron emission tends to happen more often than orbital electron capture. Hence, Kcapture has been observed only when the condition (1210) is satisfied, but it is not always observed when that condition is fulfilled. There are many examples of the above rules. Thus, the nuclides C1l, N13, 015, F1', Nelg, Na2l, Mg23, and A125a11 decay by positron emission. Each has a mass number smaller than those of the stable isotopes of the element, and is consequently protonrich. In each case, the mass of the radioactive nuclide is greater than that of the decay product by more than 2 electron masses. For example, according to Table 111, C1' has an atomic mass of 11.014924 amu, and B1' has a mass of 11.012795 amu; the diflerence is 0.002129 amu, considerably greater than 2 electron masses, which are equal to 0.001098 amu. The nuclides C14, N16, F20, Ne23 , K a 24 , MgZ7,and A12' decay by electron emission. Each has a mass number greater than those of the stable isotopes of the element and is therefore neutronrich. I t can be seen from Table 111 that, in each case, the mass of the active nuclide is greater than that of the product. For example, the mass of C'4 is 14.0076845 amu, while that of N14 is 14.0075179 amu. The isotopes of an element of intermediate atomic weight, such as iodine, show all three types of decay. Iodine has only one stable species, 531127, containing 53 protons and 74 neutrons. There are, however, 17 artificially radioactive isotopes, containing from 68 to 86 neutrons. Of these, 1121 and I'22,with the smallest number of neutrons, decay by positron emission. The nucleus captures an orbital electron, while sometimes captures an orbital electron and sometimes emits a positron; 1'" captures an orbital electron; IlZ6 sometimes captures an orbital electron and sometimes emits a negative electron; I'27 is stable; 112' decays by emission of an electron 95'3, of the time and either by orbital electron capture or positron emission 5y0 of the time. The eleven remaining neutronrich
(n.n) reactions. Examples are Positron emission is common for activities produced by (p. (d. There is some correlation between the radioactivity of an artificial nuclide and the transformation by means of which it is produced. Examples are The (alp). The decay products of some of the iodine isotopes are also unstable and decay in turn.r). which increase the chargetomass ratio of the nucleus. ~ ) reaction.a).n). The most fruitful source of artificial radionuclides is the radiative capture of neutron^.iodine isotopes.'^' the ( n . Electron emission is common for activities produced by (n. which is unstable and decays by electron emission into stable C S ' ~ ~ . since these reactions decrease the chargetomass ratio. decays into Xe133. (n. and (r. which usually yields an electronemitting product. The charge and mass changes are such that the target and product nuclei are generally both within the band of stable nuclides. decay by the emission of an electron.a). (p.n). Fermi'6' suggested the possibility that the bombardment of uranium with neutrons might result in the production of elements with nuclear charge greater .p). as in the examples 123 The transuranium elements. In 1934. and (d.r). and (d.n).p) reactions. (p. (a. from to 113' inclusive.a) reactions usually lead to stable products.
w nuclide. four different 8activities were detected. Soon after the f 93Np239. an isotope of a hitherto unknown element. in the systematic production and study of the transunrnium elements. In 1940. If UZ3' were to capture a neutron. $otape of plutonium is manufactured in large amounts in nuclear n m t m and can be used for the production of nuclear power or atomic is an aemitter with a halflife of 24. Np23Q. f Np239yields an isotope of a new element with Z = 94. and has a halflife of 4 : .l."' The element with Z = 93 was given the name neptunium.1231 .Z5. the jd. THE TRANSURANIUM ELEMENTS 29 1 than 92. .400 years. + T = 2. *. emits an electron and has a halflife ys. one of which was presumably was possible that the new eleassociated with Q ~ U It~ ~ also ~ considered .Sherproduct Np238emits an electron to form PuZ3'. . ment with Z = 93 might decay by electron emission to form an isotope of another new element with Z = 94. Any elements which might have values of Z greater than 92Pwere called transuranium elements because they would lie beyond uranium in the periodic table. many attempts were made to make and identify transuranium elements. Between 19:34 and 1939. there was difficulty with their identification.3 d. two other isotopes of neptunium were made. I t decays into an isotope of a new th Z = 93 which. When oxide is bombarded with fast deuterons from a cyclotron.Q) ' + T = 23 min. the following reaction should occur: if the U239were then to decay by electron emission. The research on these elements led to the discovery of nuclear fission. it was proved that the bombardment of uranium with slow m t K I l l s does indeed produce a new isotope U239which has a halflife of # 1and is an electron emitter. and the work which followed this discovery resulted.results from the decay Q2U23Q + Q3Np239 lea. among other things. . Early experiments showed that when uranium was bombarded with neutrons.2n) reaction takes place: [ I 92U238 + lH2 4 Q3Np238 on1 + + on1. but although some doubtless were produced. in turn. called " 93Np239 + 9 4 P ~ 2 3 Qlea. the result would be a nuclide with Z = 93.
Isotopes of elements 95 (americium) and 96 (curium) can be made by bombarding PuZ3' with neutrons. on bombardment with aparticles. Some of the isotopes and the reactions by which they are made are In all.0 days. 12 2. + + + The isotope 92U237 is an electron emitter with a halflife of 6. 96Cm242 is formed."1) The new nuclides are identified. or even 9 neutrons are ejected. gives an isotope of element 97 (berkelium). reactions occur in which 4.n243 which decays to P u by~ aemission. by separating them chemiallp and studying their radioactive properties toget her with those of their &cay products.. When uranium is exposed to fast neutrons. g2U237 on1 on1.  ~u~~~ ~ + Y. and decays into Kp237. an (n12n) reaction occurs. Isotopes of neptunium can also be obtained by bombarding uranium with deuterons with energies of about 100 Mev. Alphaparticles can also be used. 9."~' and e l a r reactions were used in the first preparation and identification d an imtope of californium.292 ARTIFICIAL RADIOACTIVITY [CHAP.2 X 10' years.and Avalues are determined. The latter is an aemitter with a halflife of 2. pu2'0 + ~n ~ +n g4pu241 5 g5~m2". in. These reactions were used in the discovery of berkeli~m. fieir 2. 9 2 ~ 2 3 8 on1 . 9sCf244. At the same time.8 days. 6. 10 isotopes of neptunium are now known. ~ ~ Bk243also emits an aparticle (q vith ' . The reactions are P U .1Bk~'~ decays by orbital electron capture to g6Ch. . and is the longestlived of the known transuranium nuclides. ~u~~~ + Y. The nuclide Am2". Thus. with mass numbers fmm 231 to 240.
.
.
one species of mendelevium ( 2 = 101) and one (or perhaps 2) species of element 102.'31' Work on the transuranium nuclides has so far resulted in the identification of 10 isotopes of neptunium ( 2 = 93).1241 THE ARTIFICIAL RADIONUCLIDES: ALPHAEMITTERS 295 of a second isotope of element 102 by bombardment of a mixture of curium isotopes with C13 nuclei has also been reported. only 24 aemitters were known and these were members of the naturally occurring radioactive series. or neptunium. There are several socalled collateral series. 15 isotopes of plutonium (Z = 94). and additional aemitters are found among the neutrondeficient isotopes of bismuth (Z = 83). I t was mentioned f the isotopes of the transuranium elements in the last section that some o decay by emission of an aparticle. The members of the neptunium ~ e r i e s ' ~ and ~ . 1 2 4 The artificial radionuclides : alphaemitters. ~their ~ ' properties are listed in Table 121. and francium ( 2 = 87). The neptunium (4n + 1) series. the 4n series. 8 isotopes of fermium (Z = 100). 10 isotopes of americium ( 2 = 95)) 13 isotopes of curium (Z = 96). 11 isotopes of californium (Z = 98). 124. 10 isotopes of einsteinium (Z = 99). emanation (Z = 86). 8 isotopes of berkelium ( 2 = 97). 1. polonium (2= 84). At the present time. astatine (Z = 85). about 150 adecaying nuclear species are known. others form part of a new radioactive series.'a2' About 50 of these are among the transuranium elements. . In 1939. and the relationships between the members are + Atomic number FIG.
even if such a series did exist a t one time. The halflife of Npa3?. that neptunium was formed a t the same time as the earth. sec1 particle energy. a a 8: 0. Like the naturally occurring radioactive series.99 0. then many halflives have elapsed for this nuclide and the amounts still present would be so minute as to be beyond the possibility of detection.05 a: 5. The absence of a naturally occurring 4n 1 series can therefore be understood. the neptunium series shows branched disintegration near the end.33 m 7.35 1.39 B B Stable 8. The stable end product is the ordinary bismuth species of mass number 209. thorium. Mev a: 4.870 0.535 m 0.02 m 0. is 2. a a 8: 1. and actinium series. as is probable.893 m B.021 5. its members would long since have decayed to Bi2O9 + . I f it is assumed. the longesblived member of the series. 124.568 m 4. I t is generally considered that the earth is about 5 x lo9 years old. rather than an isotope of lead as in the uranium.248 4.635 shown graphically in Fig.2 X 10' years.32 5.816 5.ARTIFICIAL RADIOACFIVITY TABLE121 THE NEPTUNIUM SERIES Halflife disintegration constant.86 B a B a a B a a a 8.818 m 6.
the type or types of decay. it is seen that the new series runs parallel to part of the actinium series. and artificial nuclides are differentiated by the color or shading of the square. + 1 mass type).79 m I f this series is compared with the actinium (4n 3) series in Fig.'35' These chains include series collateral to each of the natural radioactive series. and this is the reason for the name collateral series. 125. 1011. As an example of such a series.2. A . U2284 Th224_t Ra220+ Em216(4nmass type). pazzs + A C 4~ FrzZ0 ~ 4 ~ At216(4nmass type). as well as the artificial neptunium (4n 1) series.formed by bombarding thorium with 80Mev deuterons. both stable and unstable. the halflife. There is a series starting with U227which is also collateral to the actinium series. The most convenient method for quick reference is that of the nucltik chart. and actinium (4n 3). and stable. a portion of which is shown in Fig. naturally radioactive. In this type of chart the atomic number Z is plotted against the number o f neutrons. the reaction is Th232(d. and compilations of nuclear data are available like that prepared by the National Bureau of Standards. as well as the abundance of the isotope if it is stable. Uzz94 ~h~~~ + bzzl + Em217(4n  125 Isotope tables and nuclide charts. and the energy of the emitted particles . thorium (4n). Tables of isotopes are published periodically.The nuclide Pa227 decays in the following way: + + + ) (stable). The vast amount of information which has been collected about the different nuclear species.7n)Pa227.1251 ISOTOPE TABLES AND NUCLIDE CHARTS 297 In addition to the neptunium series. The symbol and mass number are shown in each case. s 3 ~ i 2 1 1 ( ~2 ~ C ) 8 1 ~ 1 2 0 7 ( ~ ~ ~ "pb207 2. makes it necessary to have some convenient methods for recording nuclear data. the uranium (4x1 2). Each nuclide occupies a square. artificial aemitters have been made which form chains collateral to the heavy radioactive series. such as those cited in the general references a t the end of this chapter. consider the decay of lPa227. this chain decays by successive aemissions as follows: + The other collateral series are Pazz64 Aczz2+ Fr218 4 At214(4n + 2 mass type).16 m 4.
12 .298 ARTIFICIAL RADIOACTIVITY [CHAP.
A complete nuclide chart is inserted a t the back of the book. LIVINGSTON and H. JOLIOT. New Haven. and SEABORG. G. The discussion of artificial radioactivity in this chapter is far from complete. and nuclei with the same mass number (isobars) lie on a 45' diagonal line running from upper left to lower right. 125 is of this type. "The Transuranium Elements. 198.78Mev 7rays. no Yrays are emitted. C.: AddisonWesley. 494 (1933). 585 (1958). PARTICULAR  1. These properties will be treated in the next three chapters. 205. Nuclear Dynamics. The information shown in the squares of Fig. 9. the isotopes lie along a diagonal line and isobars lie in the same vertical column.which is a positron emitter with a halflife of 2. HYDE d m Physik.: Yale University Press. which has a halflife of 2.n) reaction on A127is P30. REFERENCES GENERAL M. T.REFERENCES 299 and %rays. HOLLANDER. CURIE and F. along with 1. rend. K. Phys. phys.3 min. it should be apparent that the nuclide chart is very useful to anyone who needs easily available data about atomic nuclei. Mod." Handbuch E. 3." Revs. 245390 (July 1937). Similarly. In some charts. .J. The Tramuranium Ehnents. The example of Fig. Sometimes all nuclides on the same horizontal line have the same atomic number. Even with the information shown. the atomic mass and other data are also included. Only the most common modes of decay have been discussed. 1885 (1933).3 Mev. el radium 4. 125. the effect of a (d. S. 30. A. STROMINGER.p) reaction on A12' is seen immediately to be A12'. 43. 1958. 125 has been limited to keep the figure clear.SEABORG. in this case. I. and emits electrons with a maximum energy of 2. Mod. The nuclide chart can be used to obtain quickly information about the products of artificial nuclear disintegrations. 254 (1934). Berlin: Springer Verlag (1958). Conn. 196.87 Mev. and G. Vol. Mass." Revs. SEABORG. The chart gives the energy of the swiftest positrons emitted. Experimental. "Nuclear Physics. which is. One reason is that further discussion of the artificial nuclides and the information they give about nuclear stability and nuclear structure will require knowledge of the properties of the radiations from radioactive substances. "Table o f Isotopes. Reading. In other charts. such as the nuclear spin and magnetic moment. Phys. The displacements in charge and mass caused by various disintegrations are usually shown as in the lower righthand corner of Fig. p. l'Electrona Produced by Artificial Disintegration?" Compt. BETHE. I t can be seen immediately that the product of an (a.5 min. T.
and WAHL. Rev. 1 5. 12 2. 94. 10. ANDERSON. 555 (1946). FERMI. . 251. 107."Possible Production of Elements of Atomic Number Higher than 92. MECH. Rev. FRIED. ROURKE and COLLINS. FERMI. SEGR~ and . 14." Phys. 93. 198. KIRSCH.PYLE. 1081 (1954). 57. GHIORSO. 102. C. MAGNUSSON. Rev. THOMPSON.300 Rev.THOMPSON. "Transplutonium Elements in Thermonuclear Test Debris." Phys. 16. SEABORG. BUTLER."Properties of 94 (239). J. "The P d t i v e Electron." Phys. KENNEDY." Phys. 250." Phys. MECH. "Nuclear Properties of some Isotopes of Californium. and Californium. "Radioactive Element 93. GHIORSO. THOMPSON. 1635 (1957). Rev. 3." Compt. CURIE and F. PYLE. MAGNUSSON. phys. Rev.MANNING. 153 (1934). 1048 (1955). 11.HUIZENGA." Phys. JOLIOT. 908 (1954). JACKSON. I. PYLE. EASTWOOD. 9. 19. Rev. EASTWOOD. STUDIER. GHIORSO. Rev. I." Science 76. "Transcurium Isotopes Produced in the Neutron Irradiation of Plutonium. 366 (1946). THOMPSON. 94. 70. STREET. (London) A146." Phys." Phys. 8. MANNING.STEVENS. 1083 (1954).and SPENCE. Rev. 781 (1950). FIELDS. ''New Elements 99 and 100. Phys. 99. 1428 (1954)." Phys. "Chemical Separation of New Radioactive Ele9 8 . and JACKSON. SCHUMAN."Radioactive Element 94 from Deuterons on Uranium. JONES. HIGGINS. SELLERS. Roy." C m p l .INGHRAM. Rev. ABELSON. 790 (1950). and 252 from PileIrradiated Plutonium. 80. H. rend. 18.D'AGOSTINO.SCHUMAN. Berkelium. 254 (1934).STEVENS.THOMPSON. AMALDI. KENNEDY. GHIORSO.FRIEDMAN. SMITH. and CHOPPIN. DIAMOND. HARVEY. "Elements 99 and 100 from PileIrradiated Plutonium. HARVEY. 4. and SEABORG. Rev. 559 (1934). WAHL. CHOPPIN. Rev. FRIEDMAN." Phys.201 (1934)." Proc. MCMILWNand P. et radium 5 . D." Nature 133. FIELDS. 1185 (1940). "Isotopes of Einsteinium and Fermium Produced by Neutron Irradiation of Plutonium.898 (1934). HIGGINS. and HUIZENGA. 13. rend. ments.SEABORG. THOMPSON. Rev. 17. Elements 99 and 100.203 (1956). "The New Element Californium (Atomic Number 98). CURIEand F. "New Isotopes of Americium.  . MCMILLAN. FIELDS.and S E G R ~ "Artificial . COWPER. 93. STUDIER. GHIORSO. 80.11Phys. 20. E. 1080 (1954). DIAMOND. DIAMOND. "A New Type of Radioactivity. BUTLER. 12. M. CHOPPIN. "Identification of Californium Isotopes 249.STEVENS. and HARVEY. Radioactivity Produced by Neutron Bombardment. 7. CABELL. 491 (1933). SEABORG. "Isotopes of Berkelium and Californium Produced by Neutron Irradiation of Plutonium. JOLIOT. and SEABORG. G B I O R ~and . FRIED." Phys." Phys. DIAMOND. MECH.STUDIER. 238 (1932). Soc. ARTIFICIAL RADIOACTIVITY [CHAP. 94. and HUIZENGA. THOMPSON. RASETTI. "The New Element Berkelium (Atomic Number 97). STUDIER. Rev. Nature 133. 69. E. Rev. GHIORSO. 102.FIELDS. 15. 180 (1956)." Phys. 43. 6. 483 (1934). and SEABORG. MECH.
31. HAYMOND. 1518 (1955). 151 (1958). 257 (1954). Rev." Phys. "Heavy Ion Reactions with Heavy Elements. 225 (1959). THOMPSON. 24. which of the following nuclides are stable." Phys. "Artificial Chains Collateral to the Heavy Radioactive Families. HOLM and ~ % S T R ~ M . "The (4n 1) Radiosctive Series: The Decay Product of U233. V48. Rev. 782 (1951). 1460 (1957).256 (1954)." see Ref. FORSLING. STROMINOER." Phys." Phys. 85. GHIORMJ. 93. Leths. + + 1. "Reactions of U238 with CyclotronProduced Nitrogen Ions. XelZ3." Phys. "The Acceleration of Nitrogen14 (*) Ions in a 60inch Cyclotron. HAOEMANN. 34. Rev. G. ENGLISH. ROSSI. Rev. HARVEY. STREET." Phys. Geneva. 102. GHIORSO. Irle7. 30. GHIORSO.PROBLEMS 301 '    21. STUDIER. 15.JELLEY. W. and THOMPSON. FLEROV. 80.HAMILTON. Xe137. 33. AT~ERLINO. Rev. 81. 72. and SEABORG. KATZIN. and HSTROM. GHIORSO and T. 585 (1954)." Phys. Check your results by comparison with a nuclide chart. which of the following nuclides are electron emitters and which are positron emitters: Ga73. and which are positron emitters: NaZ3. Rev. 253 (1947). Rev. HOLLANDER. ." Phys."Californium Isotopes from Bombardment of U r a ~ u m with Carbon Atoms. R o s s ~ JONES. 26.HARVEY. 585 (1958). PUTNAM. 29. and HAMILTON. Check your results by comparison with a nuclide chart. Zns3. HINCKS. P/2299. P32. 32. 93. 14. 35. GHIORSO. THOMPSON. from the masses involved (Table 11l). S C ~AgllO. and MAY. Br78." Proceedings of the Second Zntemcrtional Conference on the Peaceful Uses of Atomic E w g y . Phys. FRIEDMAN. "Table of Isotopes. He6. "Heavy Ion Reactions. SIKKELAND. f. Physik. 95. CRANSHAW. MILSTED. A. HOLLANDER. SIKKELAND. Phys. Show. MEINKE. "The (4n 1) Radioactive Series. MgZ7. 17. 98. CHOPPIN. Arkiv. 22. and SEABORO. Atomic Number 101. ATTERLINQ.F18. and R o s s ~ . Rev. 30. 252 (1947). 158 (1958). .NaZ2. 486 (1950). Nbe6. P/2440. GHIORSO. 429 (1952). 1. Deduce." J. and SEABORG. 14. "New Element Mendelevium. 2." Revs. 18 (1958). 542 (1949). Rev. Rev. WALTON. Rev." Phys.MELANDER." Phys. 81."Acceleration of Stripped C12 and C13 Nuclei in the Cyclotron. RUBINSON. Mod. N. 29. "On the Production of Element 102. HARVEY. and S E A B O R "Element ~. FIELDS. 23. MILLER." Phys. HOLM. and SEABORO. No. Cs127. Si31. which are electron emitters. 25. Belo. 27. Phys. and S~a~o~o.GHIORSO. 1958. 107. DEMERS. from the d a b in the table of stable nuclides (Table 9I). . Chem. "The Equations of Radioactive Transformation in a Neutron Flux. 154 (1951). 28. AulQs. ~ . 72. FORSLINO. Rev."Element 100 Produced by Means of CyclotronAccelerated Oxygen Ions.
7) AulQ8. Plot the number of Mn6%toms present in the sample as a function of time. a new nuclide is formed as the r e s d of an (a. Which isotope of iron is the nuclide? 9.2n) reaction on germanium. The following atomic masses are given. Calculate the Qvalue in Mev for each of the following nuclear changes: He6@)Lie. Identify the unknown nuclide. from the time the cyclotron is turned on until 40 hr later. Derive a condition for the occurrence o f adecay analogous to those for electron emission and positron emission. . how long would the irradiation have to be continued until the number of AulW atoms reaches 95% of the value a t secular equilibrium? Suppose that the sample ia exposed for this length of time and then removed from the neutron beam. and for 15 days thereafter. the 46day activity is observed again. What would be the number of atoms of MnS6if secular equilibrium were attained? What fraction of this number is actuallv reached? 6. 12 3. 8. Hollander. The Mn56 ia an electron emitter with a halflife of 2. When a zinc target is bombarded with aparticles. and a radionuclide with a halflife of 46 days is obtained. The nuclide Aulg8 decays by electron emission with a halflife of 2. Na2'@)Mga4 . C14@)N14. A sample of iron is bombarded with deuterons in a cyclotron.p)Mn56 reaction. F18(B+)018.p) reaction on the germanium isotope of mass number 70 gives a nuclide with a halflife of 11.302 ARTIFICIAL RADIOACTIVITY [CHAP. F20(fi)NeZO. When a cobalt target is bombarded with neutrons of moderate energy. The same activity is found 8s a result of a (dl%) reaction on gallium. A sample of manganese is bombarded for 20 hr with deuterons in a cyclotron under conditions such that 5 X 108 atoms of Mn56 are formed per second as a result of the Mn65(d. A128(fl)Si28. Na22(B+) Ne22. This nuclide ia a positron emitter with a halflife of 1. and as a result of an (n. and Seaborg.n) reaction.58 hr. 10. 5. N13(B+)C13. How many a t o m of Aulg8 will be present after 100 hr? After 10 days? How many atoms of HglW will be present a t these times if it is assumed that the neutron beam doea not d e c t this nuclide? 7.4 days. Plot the value of the number of atoms of Aulg8 during the period of the irradiation.70 days. 4. The (d. A sample of gold is exposed to a neutron beam with an intensity such that 101° neutrons are absorbed per second because of the reaction Au197(n.65 days. Under the conditiom of Problem 6. Chemical analysis shows that the nuclide is an isotope of iron. Which of these nuclides should decay by flemission? By fl+emission? By orbital electron capture? Explain your answers and compare them with data in the "Table of Isotopes" of Strominger.
Ca40: 39. K40: 39. Gammarays are also emitted.975230. Which of these nuclides would you expect to be radioactive and how would you expect'them to decay? 13. The 7'rays follow electron capture in K40. .012 atom percent. Sulfur Argon 35.97974 35. and compare them with data in the "Table of Isotopesn of Strominger. One gram of potassium emits 29 @particles/sec owing to the decay of K40. 14.PROBLEMS 303 11. What is the halflife of K40? 15. and Seaborg. The following atomic masses are given: A40: 39.975050.Sc40: 39. 16.990250. which has an abundance ratio of 0. Explain. and the ratio of gammas to betas is 0. Calculate the partial disintegration constant and halflife for each mode of decay.97892  Which of these nuclides should decay by @emission? By @+emission? By orbital electron capture? Explain your answers. with the aid of Qvalues.positron emission (19y0)and orbital electron capture (42yo). Hollander. The nuclide Cus4 with a halflife of 12.97844 35. The following atomic masses are given. and one photon is emitted for each orbital capture. and Be8 are not found in nature. Show that Li6 must be stable by considering all the possible ways into which i t might be imagined to split. Li5.976653.8 hr decays by electron emission (39y0). why He5. 12.12.
i t is possible to determine nuclear masses and nuclear energy levels. and actinium.CHAPTER 13 ALPHADECAY The properties of the radiations from radioactive substances have been studied diligently since the discovery of radioactivity.". from the Qvalues. the methods for determining the energies of aparticles are also used for protons and deuterons.   . its orbit is a circle whose radius is determined by the relation . and the combination of the experimental and theoretical knowledge of these processes forms one of the great branches of nuclear physics. and 7rays have been developed. Accurate measurements of the energies of the radiations emitted by the natural radionuclides led to the idea of nuclear energy states analogous to atomic energy states. and this led to the discovery that some radionuclides actually emit a spectrum of aparticles.2. This interest has been extended to the information that the radiations give about the nucleus and the energy changes involved in its transformations. Fourth. and the study of the radiations emitted by both the natural and artificial radionuclides has resulted in the accumulation of a large amount of information about nuclear levels. The determination of the velocity and energy of aparticles will be discussed in some detail for several reasons. The method that gives the most precise results for the velocity and energy of aparticles depends on the measurement of the deflection of the paths of the particles in a magnetic field. thorium. Second. 8. First.3' When a charged particle moves in a magnetic field. The first interest in these radiations was in connection with the series of transformations of uranium. Theories of the emission of a. Third. These three charged particles are involved in many artificial disintegrations. and the accurate measurement of their energies yields accurate Qvalues. 131 The velocity and energy of alphaparticles. the accurate measurement of aparticle energies made it possible to determine energies which differ only by small amounts. accurate values of aparticle energies are needed in the development and use of the theory of adecay. knowledge of the energies of the components of aspectra makes it possible to assign certain nuclear energy levels with confidence.
Equation (131) may also be written 9 Hr. The mass of the aparticle is equal to the atomic mass of the helium atom . in emu. The aparticles from a radioactive source emerge in a narrow beam through a defining slit.1311 THE VELOCITY AND ENERGY OF ALPHAPARTICLES 305 FIG. The charge q is. and q/M in emu per gram. acting in a direction perpendicular to the plane of the diagram. In the figure. q and M are the charge and mass of the particle. The instrument is called a magnetic spectrograph and is designed on the principle of semicircular magnetic focusing. Particles with the same velocity have semicircular paths of the same radius. they may be detected with a photographic plate or with counters. the aparticles can be bent into a semicircle of 40 or 50 cm maximum radius. v =Af (132)  The velocity can be determined if the strength of the magnetic field is known and if the radius of the orbit is measured. respectively. and walls of the chamber. In most instruments of this type. since the value of the chargetomass ratio is well known. 131. A magnetic field of known strength. bends the aparticles through an angle of 180". The velocity v is obtained in centimeters per second when H is expressed in gauss. bottom. 131. the paths of two groups of particles with different velocities are shown. where H is the field strength. Schematic diagram of the deflection chamber o f a magnetic spectrograph for aparticles. and the radius of the path measured. and r is the radius of the orbit. The chamber slits help to reduce the scattering of aparticles from the top. A schematic diagram of an apparatus based on this principle is shown in Fig. r in centimeters.
vary from about 1. 6 4 3 ) 1 0 . (133) and ( 1 3 4 ) are . In this case.2 X 10' cm/sec.7801 6.0026 f 0. cm/sec X Energy. = 4.5 Hr. 13 Radionuclide Magnetic deflection.000 to 500. the relativistic mass correction is small and may often be neglected.5 emu/gm. At these velocities.779 5.003873 .306 ALPHADECAY [CHAP. and the values of Hr are in the range 300.00278 amu gm.074 X 1 0 .wn A 0.3007 8.6 X l o g cm/sec to about 2.6596 X loM2' = 6.685 f 0.013 f o. or M Then = 4.003 f 0.000 gauss.0466 5.0. which can be conveniently measured by this method.0007 f 0.y correction must be taken into account.4861 5.6430 X . Mev 6. (134) where v is given by Eq.004 0.0013 f 0.4823. the magnetic field strength used is generally of the order of 10.' ~ v ~ Mev.005 f 0.00278 X 1. M and v (cm/sec) = 4823. the kinetic energy E may be taken equal to +Mu2 and is given by E = +Mu2 = 3 ( 6 . Hr: gausscm X Velocity.000 gaussem.4226 4. The velocities.2823 5..620 6. (133).010 + less two electron masses. When the re1ativit. Eqs.6814 4. as in highly accurate work.~ ~ erg v~ = 2. ( 1 33) For aparticles from the natural radionuclides.0011 f 0.001098 = 4.
I f the relativistic correction were not made. with the known value of v . it is found that their number stays practically constant up to a certain distance R from the source. 132. Alphaparticles are easily absorbed. and stopping power. or Th232. and this method is used often."~) are listed in Table 131. (135) . This distance R is called the range of the particles.039 Mev in the case of ThC'. Although the magnetic deflection method gives highly accurate results. I t is relatively expensive. it requires samples with a relatively large amount of activity.004 cm thick. The uncertainty in the velocity is one unit in the fourth decimal place. If the particles emitted by a source in air are counted by counting the number of scintillations on a zinc sulfide screen.'4v5' Before the methods and the results are discussed. 132 The absorption of alphaparticles : range. The energies of charged particles. The source is placed on a movable block . from a critical ~urvey. and then drops rapidly to zero. and cannot be used successfully with substances such as UZ3'.1321 THE ABSORPTION OF ALPHAPARTICLES 307 replaced by the more complicated formulas and . When the value of Hr is known. The uncertainty in the measured value of the magnetic deflection is small enough so that the relativistic correction must be made. the resulting rangeenergy curve can be used to find unknown energies from measured ranges. or by several centimeters of air. can be determined from measurements of their absorption by matter. Precise measurements of ranges in air can be made with an apparatus'5' like that shown in Fig. (134). and more convenient (if less accurate) methods of determining aparticle energies have also been developed. U235. Some typical results for natural aemitters. and is related to the initial energy of the particles. ionization. including aparticles. those emitted in radioactive disintegrations can generally be absorbed by a sheet of paper. the kinetic energy is then given by Eq.015 Mev in the case of ThX and 0. where T represents the relativistic kinetic energy. If measured ranges are plotted against energies determined by magnetic deflection methods. or 0. the effect would be to reduce the energy by 4 to +0Jo. I t is usually easier and cheaper to measure ranges than energies. some of the features of the absorption of aparticles will be treated. the velocity is obtained from Eq. by an aluminum foil 0.
FIG. 132. An apparatus (Holloway and Li~ingston(~)). Curve D: End of a specific ionization curve for a single particle of mean range.870 cm. screenwalled ionization chamber 1 t o 2 mm deep. and reaches the detector. with the mean range = 3. Range curves for the aparticles from Po212 (Holloway and Livingston(5)). 133. Curve C: End of a specific ionization curve with the ionization extrapolated range Ri = 3. = 3. The counting rate is then determined a s a function of the distance between source and . Curve B: Differential range curve. Ions are formed in pulses in the chamber when individual aparticles pass through it. Curve A: Numberdistance curve with the extrapolated range R. A narrow beam of particles emerges through a collimating slit. The latter is a thin. whose distance from a detector can be varied. The voltage pulses induced on the chamber electrode are amplified electronically and counted. passes through a known thickness of air. .897 cm.842 cm.ALPHADECAY Screenwalled ionization chamber Collimated source on adjustable block for precise measurements of the range of aparticles FIG.
If the derivative of the numberdistance curve (curve A ) is computed a t different distances from the source and then plotted against the distance. 133. The ordinate is the relative number of particles. only the results near the end of the path are shown. [Reprinted by permission from Rutherford. is obtained by drawing the tangent to the curve a t its inflection point and noting where the tangent crosses the distance axis. and the abscissa is the distance from the source.75 cm from the source. Re. and then decreases somewhat more slowly t o zero.897 cm for the extrapolated range. Curve A shows the fraction of particles in the beam detected a t various distances from the source. 1930. A cloud chamber photograph showing the straggling of aparticles from Thorium C'. Chadwick.1321 THE ABSORPTION OF ALPHAPARTICLES 309 FIG.88 cm. New York: Macmillan. Radiations from Radioactive Substances. and Ellis. The results of a n experiment with aparticles from polonium are shown in Fig. 134. . The last three centimeters of range are shown. the fraction detected falls rapidly to 0:2 a t 3.1  detector. A quantity called the extrapohted range. All of the particles which pass through the slit are counted until the detector is about 3. The dotted line in the figure is the tangent and gives the value 3.
and is related to the energy and range. but vary around an average value. This curve gives the relative number of particles stopping a t a given distance as a function of the distance from the source. The values of the ranges are usually given for air a t 15°C and 760 mm Hg. the unit of the ordinate is so chosen that the area under the differential range curve is unity and all the particlesare accounted for. The amplifier can be designed so that the voltage height of its output pulse is very nearly proportional to the number of ion pairs formed in the chamber. the value of Ri for Po210was found to be 3. Specific ionizationdistance curves are shown in Fig. and then drops sharply to zero. and is shown clearly in curve C of Fig. which is a cloud chamber photograph of aparticles from ThC'. The ionizdiun extrapolated range Ri is defined as the value of the abscissa a t which the tangent to the curve a t its inflection point crosses the horizontal axis. and the intensity of the ionization caused by the particles is expressed by the speci@ ionimtion. 135 for the particles from RaF (Po210)and RaC' (Po214). the actual range of an aparticle is not definite. In the case considered. and to avoid this indefiniteness either the extrapolated or the mean range is used. The maximum ordinate of the differential range curve occurs at a value of the abscissa which is called the mean range. defined so that half the particle track lengths exceed it. reaches a maximum. E. giving rise to straggling. while half are shorter. 132 can be used to measure the relative specific ionization produced by a beam of aparticles a t different distances from the source. Alphaparticles lose a large fraction of their energy by causing ionization along their paths. 13 a diflerential range curre (curve B) is obtained. defined as the number of ion pairs formed per millimeter of beam path. An electron and the positive ion which results from its removal from an atom form an ion pair. The effect of straggling is seen near the end of the range. This effect is called straggling. The results show that the track lengths of the particles in the beam are not all the same. a t first quite slowly and then more rapidly. the actual distance from the source a t which their energy is completely used up is somewhat different for different particles. Hence. Because of straggling. The same type of curve is obtained for any gas and for any group of monoenergetic particles even though the magnitudes of the ionization and of the range vary considerably. 1 3 4 . . Some particles hit more and others hit less than the average number of molecules in passing through a centimeter of air. the relative specific ionization increases. The apparatus of Fig. As the distance of the aparticles from the source increases. The extent of the ionization mused by an aparticle depends on the number of molecules it hits along its path and on the way in which it hits them. The ionization caused by a beam of aparticles can be measured. and is illustrated in Fig. 133.870 cm.310 ALPHADECAY [CHAP.842 cm. the mean range is 3.
Specific ionization of a beam of aparticles as a function of distance from the source.1 'c' 1. It is possible to correct the relative ionizationdistance curve of a beam of aparticles for the effect of straggling. Radiotima from Radioactive Substances. Specific ionization of a single particle of mean range as a function of distance from the source and from the end of the track (Holloway and Livingst~n'~)). Chadwick.0.* 5 0. and the bottom abscissa scale gives the distance from the end of the track. New York: Macrnillan. while the top scale gives the distance of the detector from the source. and Ellis. 1930. 136.1321 THE ABSORPTION OF ALPHAPARTICLES 1.4 .6 d e . 135.1 Distance of the detector from the source (em) Distance from the end of the track (em) FIG. a curve of this kind is shown in Fig.8 $ 0. The specific ionization is usually measured from the end of the track.2 s 2 0 0 1 2 3 4 5 6 77.. and to obtain a specific ionization curve for a single particle of mean range.136.5 Distance in cm in air at 760 mm and 15°C Fro.0 7 2 0. [Reprinted by permission from Rutherford. A specific ionization curve for a single aparticle can be  .
(Reprinted by permission of the United States Atomic Energy Commission. 136. 137. and the probability of producing ion pairs increases. Analysis of aemitters in a multichannel pulse analyzer: radiothorium and its descendants. The shape of the ionization curve depends on the change in speed of an aparticle as it traverses its path. The specific ionization finally decreases very rapidly in the last few millimeters of its range. and is 2700 ion pairs/mm 3. The total number of ions produced in a gas by the complete absorption of an aparticle of known energy can be estimated. I n producing ion pairs. The ionization increases very slowly a t first as the aparticle loses energy. This effect accounts for the increase in specific ionization as the particle moves farther from the source.312 ALPHADECAY [CHAP. I t follows that the particle loses about 35 ev. When it moves more slowly it spends more time in the neighborhood of the air molecules it encounters."' the results for the aparticles from Po2'' are plotted as in Fig.(')) .95 cm produces a total of 2. for each ion pair formed.0 cm from the end of the range. It then increases more and more rapidly and reaches a maximum of about 7000 ion pairs/mm when the particle is 4 or 5 mm from the end of its range. 5. The ionization per millimeter of path for the average aparticle from RaC' has been deduced from the ionization curve of Fig.42 Mev . Channel number FIG. 135. the particle loses energy and its speed decreases. When the particle is just starting it produces about 2200 ion pairs/mm. on the average. TRdTh.2 x lo5 ion pairs in air at 15'C and 760 mm of Hg before being stopped. An aparticle from RaC' (Po2'') whose energy is 7. 13 obtained directly by studying the photographic density of tracks in a cloud chamber.68 Mev and whose ionization extrapolated range is 6.
has been very useful in a great deal of the work on artificial aemitting radionuclides discussed in the last chapter. uniform foils are placed over an aemitting sample. Rapid advances in the design of electronic instruments pushed photographic emulsions into a secondary position.lo' '~~ The output of the ionization chamber is amplified and then fed into a number of different pulseselector circuits. studies of the ionization produced in argon have recently"" helped to clarify some problems. and the thickness needed to absorb the aparticles completely is a measure of the range. If the entire path of an aparticle is contained in an ion chamber and the total number of ion pairs formed is measured. Until the develo~mentof electrical counting instruments. An instrument of this kind. the emulsion was one of the most important tools in the field of radioactivity. so that pulses of a number of different sizes are recorded simultaneously. A plot of energy against channel number is made by using samples of aemitters whose energies are known from magnetic deflection methods. electrons are captured by the aparticle and a neutral helium atom is formed which can no longer cause ionization."" with 48 channels. these emulsions are called nuclear emukions. the curve was taken in two stages. since the average number of Instead of electron volts per ion pair is known for various gases.'7~s911' measuring the amount of charge collected as a result of the ionization caused by each aparticle. the whole range of pulse heights may be recorded a t once. RdTh to ThA in the first stage and ThC' in the second stage. in fact. The channel number represents the size of the pulse (which depends on the energy). Each circuit is set to select pulses of a different height. Other gases can be used and. but recent improvements in the preparation of emulsions have made them once again important tools in radioactive mea~urements."3' Because of their usefulness in nuclear physics."~' Measurements of track lengths in emulsions can be correlated with particle energies. An example of the analysis with this instrument of the ionization produced in argon by a sample containing RdTh (Th228) and its descendants is shown in Fig. and this plot serves as a calibration. The total ionization caused by an aparticle can also be used to determine the energy of the particle. the energy of the particle can be determined. Thin. The number of aparticles emitted by each nuclide is proportional to the area under the peak. differentialpulse height s e b c t o ~ s . giving rangeenergy curves for different charged particles. Range measurements can also be made with solid foils as absorbers. . the voltage pulses formed are analyzed by means of ingeniously designed electronic circuits called multichannel. If there are enough of these channels. 137. The discussion of the absorption of aparticles has so far been limited to absorption by air.1321 THE ABSORPTION OF ALPHAPARTICLES 313   Finally. Photographic emulsions have always been useful in the study of the radiations from radioactive substances.
the range in the substance can be deduced from Eq. I t makes predictions which are in good agreement with the results of experiments on the ionization produced by fast charged particles.'^' The theory of the stopping power depends on knowledge of the behavior of electrons in atoms. and the range of the particle is given by where Eois the initial kinetic energy. the particles from RaC' which have an extrapolated range of 6. the stopping power can be obtained from the relation One of the reasons for the importance of the stopping power is that it can be calculated theoretically both from classical mechanics and quantum mechanic^. basis for the determination of charged particle energies from range and ionization measurements. as well as empirical. Another quantity of great value in the treatment of the absorption of charged particles by matter is the stopping power. aluminum. The stopping power varies with the energy of the particle. the usefulness of the concept can be illustrated by discussing one of its applications. The range of aparticles in such materials is very small. When the energy loss is found in this way for different initial velocities."^. .00140 cm in gold. The energy lost by a nonrelativistic charged . The theory provides an understanding of the empirical information concerning the absorption of aparticles by matter. 0. and on the range.0036 cm in mica. and there is a sound theoretical.00406 cm in aluminum.314 ALPHADECAY [CHAP. 13 Mica.g.953 cm in air have extrapolated ranges of 0. Although a detailed treatment of the stopping power is beyond the scope of this book.  where E is the classical kinetic energy. and 0. The stopping power S(E) of a substance can be determined experimentally by measuring (e.. on the energy loss. defined as the energy lost by the particle per unit path in the substance. If the range is known as a function of energy. (138) as a function of the initial energy. or gold foils are used commonly because they can be prepared easily in different thicknesses with reasonable uniformity. by magnetic deflection) the energy of the particles which have gone through a certain thickness of the substance.
called the average excitation potential of an atom of the substance. (1310) may be rewritten "  The quantity on the left.1321 THE ABSORPTION O F ALPHAPARTICLES 315 particle per unit length of its path in a given substance can be expressed in the following form derived from theory. it can be calculated and measured for different values of the atomic number Z of the absorbing substance. 138. For an aparticle. Z is the atomic number of the substance. as can be seen from Fig. me is the mass of an electron. so that a graph of S' against Z should give a curve which is not quite a straight line. where ze and v are the charge and the speed of the particle. The variation of the stopping power with atomic number. respectively. z = 2. FIG. 138. which must be obtained experimentally. the circles represent experimental results. among other things. Calculations of ranges from Eqs. N is the number of atoms per cubic centimeter of the substance. called St for convenience. and I is a quantity. The quantity I depends. and Eq. (138) and (1310) give values in good agreement with experiment. The agreement between theory and experiment is very good. is proportional to the stopping power. the solid curve represents the theoretical prediction. on Z. .
33 1. from Eqs. em equivdent to power 1 cm of air t. it is convenient to consider the ratio of the stopping power to that of a standard substance.45 1. The extrapolated range is used here because this is the quantity for which experimental values are given in the table.00406 0.00140 1930 1700 3800 4950 2. is often used. The values given are for aparticles from RaC'. of material needed to absorb the aparticles. (1313) This quantity gives the mass per unit area. or the thickness."'$. a ratio approximately independent of the velocity.3 27. Relative stopping power = Range of aparticle in air (1312) Range of aparticle in substance  For some purposes it is useful to express the stopping power of s substance in an alternative form. by the range of the aparticles in air. 13 Since the stopping power varies with the velocity of the particle.1 11. The relative s t o p ping power o f a substance.8 2. (1313). defined by Eq.316 ALPHADECAY [CHAP.57 2.70 8. which have an extrapolated range of 6.1 Mica Aluminum Cower Gold 0.0 16. The thickness that is equivalent in stopping power to one centimeter of air is obtained by dividing the equivalent thickness. Equivalent in mg/cm2 stopping gm/cm3 range.89 . Thickness in mg/cm2 equivalent to 1 cm of air = Density x 1000 Relative stopping power Values of the quantities discussed are listed in Table 132 for some of the most frequently used foil materials.93 19. Thickness Extrapolated Relative Density. (1310) and (13ll). Then."z 10.0036 0.00183 0.35 3. defined as the ratio of the range in air to the range in the substance.953 cm in air at 15OC and 760 mrn Hg. Air at 15OC and 760 rnm Hg is usually taken as the standard. the e q u i v W thickness in units of mg/m2 defined by the relationship Equivalent thickness in mg/cm2 = Range X Density x 1000.
15 cm for particles with a mean range of 11 cm. The mean range has the advantage of being somewhat less dependent on the particular experi Mean range in air.780 11. The ranges in air of the aparticles from some natural aemitters are listed in Table 133 together with the energies measured by the magnetic deflection method.542 5.318E3I2. the empirical equation = 0.64 3. the difference depends on the magnitude of the mean range and varies from about 0. The ranges are quoted from Reference 5 and the energies from Reference 2b.with 7i in centimeters of air a t 15OC and 760 mm Hg. holds fairly well.685 6.897 5. Mev .643 4. The extrapolated range. The values show clearly the differences between the mean and extrapolated ranges as well as the general relation between range and energy.035 5.997 7.1331 RANGEENERGY CURVES 317 133 Rangeenergy curves.15 11.496 5. is greater than the mean range. It is easier to determine the extrapolated range experimentally.839 9. as obtained from the numberdistance curve.769 6. The range increases monotonically with the energy.796 8. and for mean ranges between 3 and 7 cm.891 9.076 4.616 9.312 5.676 9.05 cm for particles with a mean range of 4 cm to about 0.722 6.870 Energy. dirhance distribution.727 6.272 5. cm cm 5.595 5.778 8.462 6. and the mean range can then be calculated quite from the extrapolated range. and E in Mev.503 6.714 4.57 3.015 5.953 7.672 4.841 11.641 5.09 11.713 4. cm Range from extrapolated Range from numberextrapolated ionization.073 5.109 4.053 5.
or of particles produced in nuclear reactions. Information at lower energies can be obtained either by studying the range and energy of particles which have been partially slowed down. The energy of the aparticle is 2. The first of these is with a Qvalue of 4.010 Mev and the range is 1.ALPHADECAY FIG. Rangeenergy curves for aparticlea (Bethe(16)). Two reactions producing aparticles have been found to be valuable for the rangeenergy relation. The data of Table 133 would give a rangeenergy curve down to an energy of only about 5 Mev. When the mean range is plotted against the energy.02 cm.023 Mev. 139. the resulting rangeenergy curve can be used to find the energy of arparticles whose range in air is measured.057 0.04 f 0. . which would limit the usefulness of the curve. The second reaction is B'O+n+Li7+He4+Q. mental setup. and is now generally used in range tables and for rangeenergy curves.788 f 0.
1341 ALPHAPARTICLE SPECTRA 319  with a Qvalue of 2. A sample of ThC contains ThC' and acts as a source of aparticles with both of the ranges cited. The energy as determined from total ioniation measurements is 4.492 Mev group. Later studies with the cloud chamber. and magnetic spectrograph indicated that ThC' emits two groups of longrange particles with energies of 9. Radium C' and ThC' emit a few longrange particles having energies considerably greater than that of the main group of particles.02 cm.015 cm.422 Mev group. Longrange particles and fine structure.02 cm. Rutherford and Wood found that a few particles from such a source were able to paea through an absorbing screen sufficiently thick to stop all particles with a range of 8.543 Mev group.   " . 1 3 4 Alphaparticle spectra. and the energy from the curve is 4.99 Mev. According to curve I1 of Fig. With the points provided by these nuclear reactions. and with recent improvements in the knowledge of the relationship between ionization and energy loss. emits aparticles with a mean range of 8. The most recent magnetic spectrograph measurements show the presence of a third group with an energy of 10.780 Mev for the particles in the main group.015 Mev."~~.6 cm.25 Mev.474 + 0.006 Mev.720 f 0.180 f 0. A recent measurement of the mean range of aparticles from u~~~ (UI) gave a value of 2. respectively. rangeenergy curve may be illustrated by some simple examples. ionization chamber. The use of th. similar curves are available for ~rotons and deuterons. For each million particles of the main group there are about 40 particles in the 9. that a given nuclide often emits particles with a number of different energies. The energy of the aparticle is 1. 20 in the 10.02 Mev. The existence of longrange particles was first observed by Rutherford and Wood. this range corresponds to an energy of 4."'A curve for aparticles is shown in Fig. and it was clear that they were different from the normal particles emitted by ThC and ThC'. it has been implied that all aparticles from a given active nuclide have the same initial velocity and energy.57 cm."" who studied the absorption of the arays from a sample of ThC.513 Mev. For Tha3'. Careful experiments have shown. 139.492 Mev and 10.004 Mev and its range is 0. Thorium C emits aparticles with a mean range of 4. When more absorbing material was used the penetrating particles were found to have a range of about 11 cm in air. and the agreement is very good. and 170 in the 10.49 f 0.316 f 0. it also decays by Bemission to ThC' which. in turn. 139.73 cm. as compared with an energy of 8. and the agreement is not bad. In the treatment of the energy and range of aparticles. the energy determined from the total ionization is 3. the range was found to be 2.422 Mev.70 f 0. however. it has been possible to prepare more extensive and more accurate rangeenergy ~urves.
138 2. (London) A122.608 1. 1310.696 2.014 2.512 2.000 0.438 2. Proc.266 2. A cloud chamber photograph of aparticles from RaCf showing two longrange particles.281 1. Mev Difference of alphadisintegration energy from that of main group.663 1.320 ALPHADECAY [CHAP. cm Alphaparticle energy' Mev Relative number of particles Alphadisintegration energy. Roy.879 .842 2.411 1. Mev 0. Soc. 13 FIG. (Nimmo and Feather. Mean range.
Radium (Ra226)has three groups with energies of 4. about 30 longrange particles are emitted per million of normal range. 4. emits 12 groups of longrange particles with energies ranging from 8.3%). whose main group of aparticles has an energy of 7. and 4. Thus.936 Mev(24%).1341 ALPHAPARTICLE SPECTRA 321  Measurements with the magnetic spectrograph have also shown that RaC'.94.000 0.7%). For this reason the different groups could be separated only in a magnetic spectrograph of high resolving power.492 0.506 Mev.473 0.2 69.777 Mev(. A great deal of research has been done on aparticle spectra since Rosenblum's discovery and the spectra of many natural and artificial aemitters are now known.615 Mev(23.016 6. Difference of alphadisintegration energy from that of a0 group. closely spaced components of the arays are said to form a spectrum.661 Mev and relative abundances varying from 25 to 0. Ionium (Th230) haa a t least four groups.994 Mev(76%) and 3.874 5.15 1. THE ALPHAPARTICLE S P E ~ U OF M ThC Alphaparticle energy. Mev 0.436 Mev(0.(lg' Altogether.4%). Another quantity.682 Mev(76. 1310. it will be discussed in Section 135. with energies of 4.329 0.070Jo). A cloudchamber picture of typical longrange particles from RaC' is shown in Fig.163 5. Alphadieintegration energy.711 5.593 Mev(5.680 Mev. it may have several other groups with very small abundances. is also listed. 4. or to show line structure. 4.7 0. The velocities and energies of the different groups differ so little that the ranges of all the particles lie within the region of straggling. with energies 3. The discrete.040 0.1 0. ~h~~~has two groups. In that year it was shown by ~ o s e n b l u m ' ~ magnetic deflection experiments that the normal aparticles emitted by some active nuclides fall into several closely spaced velocity groups. while R ~ A c ( T ~ ~ ~ ? ) has 14 groups with energies between 6.Ol%).584 . The longrange aparticles from RaC' are listed in Table 1 3 4 .34 Mev(O.1%.277 Mev to 10.2yo) and 4.619 Group % 27.9 1. the aparticles emitted by an active nuclide all had the same in~very ' careful initial energy. the alpha disintegration energy. Mev Relative number of particles.471 Mev(0.3%). Until 1930 it was thought that with the exception of the rare longrange particles.030 Mev and 5.
RaF. only a very small fraction (lo' to lo') of the number of particles in the main group. As in the c m e of atoms. by the emission of radiation. AcX. Careful measurements of the aparticle energies and the Yray energies led to the conclusion that the ?rays are emitted by the product nucleus that has been left in an excited state after the emission of an aparticle. the process of aemission sometimes leaves excited nuclei which then emit radiation. it must be possible to determine energy levels in the nucleus experimentally. Thus.322 ALPHADECAY [CHAP. 135 Nuclear energy levels. 13 The aparticle spectrum of ThC (Bi212)is given in detail in Table 135. or of only a slightly different. Spectra consisting of two or more discrete. Rn. and the way in which the energies of the %rays and the aparticles can be correlated in terms of nuclear energy levels was shown by gar no^. the latter quantity has been called the . Spectra consisting of a main group and groups of much higher energy (longrange) particles. The discovery of aspectra gave rise to the problem of why aparticles should be emitted by a given nucleus only with certain discrete energies. nuclei must be excited to emit radiation so that the energy levels in the nucleus may be deduced from the frequencies of this radiation. An. It is supposed that there is a number of discrete energy levels in the nucleus. ThX. 2. for example." for example. closely spaced (in velocity or energy) components with intensities of the same. the latter containing. i. A nucleus in such an ' excited state would be expected to give up its energy by some emission process. however. The information now available about aparticle spectra leads to the conclusion that these spectra may be divided into three groups: 1. Pa..e. The values of the energy used are averages of two sets of measurements given in the isotope table of Hollander.tions having more than the normal amount of energy. It was known that those aemitters that have discrete aparticle spectra also emit 7rays.   . order of magnitude. in configura. The third kind of spectrum occurs only in two extremely shortlived nuclides RaC' and ThC'. for example. and Seaborg. In view of the successful interpretstion of atomic spectra in terms of discrete electronic energy levels. Spectra consisting of a single group or "line. 3. RaA. as in the Bohr theory of the atom and in wave mechanics. it is necessary to die tinguish between the kinetic energy of an aparticle and the total energy change in an adecay process. but under certain conditions it may exist for short times in excited states. ThC. To apply these ideas. RdAc. Perlman.'^^' Before proceeding further with the discussion. and that a nucleus is normally in its lowest energy state. it seemed reasonable to try to account for discrete aspectra in term8 of nuclear energy levels.
Where relativistic effects are neglected. The principle of conservation of momentum requires that . = (M/M. nucleus recoils. carrying with it a certain amount of energy. the radiation should appear in the form of a %ray with an energy of 0. The adisintegration energy is the sum of the kinetic energies of the aparticle and the product nucleus. the product. the adisintegration energy E. I t is supposed that a ThC nucleus always releases 6.emits six groups evels can be illustrated by the case of T ~ C ( B ~ ~ 3f aparticles with energies listed in Table 135. 323 adisintegration energy. and that this amount of energy is associated with one of the most energetic aparticles. (1315)) v. and is found as follows. is the mass of the product nucleus. The ThCN nucleus. which retains the 0. or ground state. respectively. I f this is the case. Gamma .040 Mev.040 Mev less than the total available adisintegration energy. should be left in an excited state. Suppose now that a ThC nucleus emits an aparticle of the orlgroup for which the disintegration energy is 6.203 Mev of energy when it decays by aemission to form a nucleus of ThC"(TlZo8). The interpretation of complex aspectra in terms of nuclear energy which ' ~ ) . therefore. v. simply by multiplying the kinetic energy of the aparticle by the quantity . All of the disintegration energy has gone into the kinetic energy of the aparticle and the recoil energy of the ThC" nucleus. or residual. and The adisintegration energy is obtained. is its velocity.163 Mev or 0.086 Mev corresponding to an adisintegration energy of 6.)v.203 Mev.040 Mev of energy. where M.NUCLEAR ENERGY LEVELS . The emission of one of these particles is assumed to leave the product nucleus in its lowest energy state. and the quantities without subscripts refer to the aparticle. it might be expected to undergo a transition to the ground state by emitting electromagnetic radiation (in analogy with the transitions undergone by atoms which have electrons in excited states). is given by From Eq. The most energetic particles have a kinetic energy of 6. When an aparticle is emitted. mass of aparticle + mass of product nucleus  The adisintegration energy is listed in the fifth column of Tables 134 and 135 for the aparticle groups from RaC1 and ThC.
288. Experimentally.472 Mev.040. goes adecay. 0. he scheme for the decay of ThC to ThC" is shown in Fig. 0. By extending this analysis we can postulate the existence o f a t least five excited states of the ThC" nucleus and we would expect to find additional transitions from more excited states to less excited states. The diagonal lines represent the different aparticle groups observed in the decay of ThC. 13 ThC"(TP@) ThC (Biz12) FIG. and 0. 0. Each of these 7rays can be matched (within the uncertainties of the measurements) t o an energy difference between two excited states. As a result of the correlation between the energies of the aparticles and the 7rays. eight Yrays have so far been identified. Decay scheme for the adecay of ThC to ThC". in fact. The level diagram for the daughter nucleus is usually shown together with the decay data for the parent nucleus and the result is called decay scheme.432. the disintegration energies are also shown in this example.144.452. 0.1311. These transitions should involve the emission of 7rays. found experimentally when ThC under. 131 1. it is possible to construct nuclear energy level diagrams analogous to the atomic energy level diagrams shown in Chapter 7. with energies of 0. 0. rays of this energy are. The a .324 Energy above ground state of ThC" (Mev) ALPHADECAY [CHAP.328. and the energies of the aparticle groups and the %rays form a consistent scheme. respectively.164. 0.
1312. and energy level schemes have been worked out. which may make them hard to detect. This procedure represents a . vertical lines represent the 7rays found experimentally.NUCLEAR ENERGY LEVELS ThD (Pbm) Fra. There is a theory which accounts in a satisfactory way for the radiations from excited nuclei and the selection rules involve the angular momentum and parity of the initial and final states. + ThD decays. within the limits of experimental error. The absence of these Yrays may be caused either by their small intensity. Energy level diagram for the ThC + ThC' showing the origin of the longrange aparticles o f ThC'. I t is evident from the diagram that Yrays are not observed for all transitions which seem to be possible. although there is some ambiguity in the case of the 0.144Mev 7ray which may correspond. The aspectra of a large number of natural and artificial aemitters have now been correlated with the Yray spectra of the decay products. to either of two pwible transitions. Each w a y corresponds to the energy difference between two states. or by nuclear selection rules which may make the probability of these transitions very small.
A factor of two or three in the energy corresponds to a factor of 10'' in the halflife and disintegration constant. and the extra energy of the aparticle measures the excitation energy of the initial nucleus. The ratio of the longest to the shortest range is about 3. When the range. the newly formed excited nucleus is an aemitter. and disintegration conhant of the natural aemitters are compared.39 X 10" years for the longestlived nuclide to 3. the disintegration constant varies extremely rapidly with small changes in energy. Other methods will be discussed in later chapters. These empirical facts were first correlated in the GeigerNuttall'22' rule. I t has been found in the study of the emission of 8. an approximately straight line was obtained. halflife. before an adisintegration. The lines for the three natural series turned out t o be practically parallel. the ranges lie between 2. The origin of the longrange particles emitted by RaC' and ThC' can also be interpreted in terms of energy levels. With the exception of the rare longrange particles. 136 The theory of alphadecay. 13 I start on the general problem of nuclear energy levels because it provides one of the methods by which levels can be determined. The existence of long range aparticles is thus explained as being caused by the decay of an excited nucleus. to be in an excited state. The longestlived nuclides emit the least energetic aparticles. and the adisintegration energies vary from 4.31 X 10' sec'.and 7rays from nuclei that the emission of a nuclear Bray sometimes leaves the newly formed nucleus in an excited state. however.326 ALPHADECAY [CHAP. In some cases. the excitation may result from a previous disintegration. But the halflives vary from 1. In other words. and gets rid of its excess energy by emitting aparticles with greater energy than that of the normal particles.5 cm and 8.6 cm. and the disintegration constants vary from 1.05 Mev to 8. certain remarkable facts stand out.58 X lo'' sec' to 2. which are illustrated by the data listed in Table 136. An energy level diagram for the T h G ThC' decays showing the origin of the long range aparticles of ThC' is given in Fig.95 Mev. In most cases the excited nucleus goes to its ground state by emitting a Yray of the proper energy. It is possible for a nucleus.0 x lo' sec for the shortest. while the shortestlived nuclides emit the most energetic particles.5 and the ratio of the greatest to the least energy is between 2 and 2. energy.5. (1317)  . The rule may be represented by the formula log X = A log R + B. which stated that when the logarithm of the disintegration constant (in sec') was plotted against the logarithm of the range (in cm of standard air) for the aemitting members in the same radioactive series. This process will be discussed in greater detail in the next chapter. 1312.
where E is the kinetic energy of the aparticle. Alphacm of disintegration standard air energy. The GeigerNuttall rule is illustrated in Fig.23 X 2. until the .10 X 3.THE THEORY OF ALPHADECAY 327 Nuclide Th232 ~ ~ 2 2 6 (Ra) Th22S (RdTh) Em222 (Rn) Po2lS (RaA) Po216 (ThA) po214 ( h c ' ) Po212 (ThC') Mean range. are unable to penetrate close enough to the nucleus to show departures from the Coulomb law. (1318) + * which expresses the relationship between the logarithm of the disintegration constant and that of the energy. Mev Halflife constant. The GeigerNuttall rule may then be written in the form log X = A' log E B'. the potential energy is about 9 Mev.78 X 4. 1313.36 X 1. and incresses a t least up to some maximum value for smaller values of r. In Section 133 it was noted that the mean range was found empirically to be proportional to E312.16 X 2. sec1 1.31 X 10l8 lo" 106 lo3 lo6 The constant A is the slope of the line and has nearly the same value for the three series. Consider the case of emission and scattering by U238. I t follows that a t least up to a distance of about 3 X 10l2 cm from the nucleus.33 4. In experiments on the scattering of aparticles. but B has different values.58 X 1. The facts of orparticle emission presented a difficult theoretical problem which is brought out sharply when the spontaneous emission of aparticles by a nucleus is compared with the scattering of aparticles by the same nucleus. it was found that the fastest natural aparticles available. the potential energy of the aparticles in the field of the nucleus is still expressed by the Coulomb formula For r = 3 X 10l2 cm. those of ThC' with energies of nearly 9 Mev.
in the figure represents the adisintegration energy when the particle is far from the . The interactions between a radioactive nucleus and an *particle inside and outside the nucleus may then be represented by a potential energy curve such as that in Fig.328 ALPHADECAY [CHAP. The interaction between the nucleus and the aparticle in the region of uncertainty may be represented by a constant attractive potential Uo. 1314. This type of potential is spoken of as a potential well of depth U o and width or range ro. 5 12 4 g l o * 8 :1 FIG. Since radioactive uranium nuclei emit aparticles. the shape of the potential energy curve is not known with certainty. and may be considered to form a potential barrier. This repulsive potential prevents the incident aparticle from entering the nucleus. Close to the nucleus and inside it. (The horizontal line E. the aparticle could exist inside the potential well. the latter may be supposed to exist. Coulomb law breaks down. The rising portion of the curve from r2 to r1 indicates increasing repulsion of an aparticle as it approaches the nucleus. with a kinetic energy equal to E. The GeigerNuttall rule. within these nuclei. a t least for a short time before emission. 13 22 20  18  + * C 8 1614  c 8 . . Under classical mechanics. exerted over a distance t o called the eflective radiw of the nucleus.U . 1313. but the Coulomb potential must break down and be replaced by an attractive potential.
the maximum value of the potential energy curve corresponds to the value of the kinetic energy which an aparticle must have according to classical mechanics to get into the nucleus from outside. emits an aparticle with an energy of about 4 Mev. but . Similarly. Cbaical physics provides no solution to this problem. The mathematical treatment will not be given. nucleus. Thus.THE THEORY O F ALPHADEC. and it is very difficult to understand how the particles contained in the inside of the nucleus can go over a potential barrier which is more than twice as high as their total energy. however. 1314.4Y FIG.e. The uranium nucleus. the GeigerNuttaU rule. Thii maximum value must be greater than about 9 Mev. Alphaparticle emission: potential energy of a nucleus and an aparticle as a function of their separation. The paradox was resolved when the problem of a potential barrier such as that involved in aemission or scattering was analyzed by the methods of wave mechanics. i. an aparticle approaching the nucleus from outside could penetrate the nucleus only if it had enough kinetic energy to get over the potential energy barrier. nor can it account for the extremely large variations in halflife corresponding to small changes in energy.. since aparticles from ThC' are scattered by uranium.) To escape from the nucleus. the particle would have to have a kinetic energy at least as great as the value of the energy at the maximum of the potential energy curve. or to escape from inside the nucleus.
. The wavemechanical treatment of adecay yields an expression for the disintegration constant which depends on the permeability or transparency of the bamer. showed how this comes about. and is equal to v [I (M/M. it is found that there is a firmite probability that the particle can leak through the barrier even though its kinetic energy is less than the potential energy represented by the height of the barrier. the petmeability of a simple barrier is given mathematically by an expression of the type where h = h/27r. if the Schroedinger equation for an aparticle is solved. but may pass through the barrier and continue its forward motion. The probability that an aparticle can leak through the barrier (the "tunnel effect ") can be calculated. a relatively small . where v is the measured velocity of the aparticle and M.2) is the charge on the product nucleus. va is the velocity of the +particle relative to the nucleus. One. Since X is proportional to an exponential function. is the mans of the recoil nucleus. va/ro. more strictly. M is the mass of the aparticle. Equation (1320) has some interesting properties. Condon and Gurney. the exponential factor gives the probability that the aparticle will leak through the bamer.'24)in 1928.330 ALPHADECAY [CHAP. In other words.e. i. and the integral is taken over the entire region in which U(r) > Em.)J. a particle impinging on the barrier will not necessarily be reflected. An approximate form of the result is where is the radius of the nucleus or. gives in a rough way the number of times per second an aparticle inside the nucleus collides with the potential barrier. The expression for the disintegration constant may be considered to consist of two factors. If the motion of a particle in the neighborhood of a potential barrier is treated wavemechanically. The quantity Ea is the alphadisintegration energy given by to + and (Z . gar no^'^^' and. the "effective radius of the nucleus for adecay". independently. 13 the results and their implications will be discussed.
4?re2(2 . and the theoretical expression agrees in a general way with the observed facts. if the logarithm of the disintegration constant is plotted against the reciprocal of the . log A = a .sin a0 cos ao). so that the value of the factor is of the order of magnitude 100. All of these terms may be lumped into one term which is practically a constant. (1321) log X = log . Equation (1320) represents a highly important result of the application of wave mechanics to nuclear physics.2)/hv. Then (2.054 x Values for the velocities of aparticles range from 1.). the greater is the value of the disintegration constant. 2 + and. in the exponential (the function a0 . so that e2 = 2. vary relatively little. The argument of the exponential depends on the velocity of the aparticle and a small change in the velocity affects the value of X strongly. Consider the factor 8e2(Z . 8e2@ . v./ro is small because both ro and v.ro 2. The theory also offers a clue to the meaning of the GeigerNuttall rule. It is also seen that the greater the velocity of the aparticle.80 X lo" esu.303h [(Z .303hv.304 x 10'~. to a good approximation.2 X logcm/sec. A small change in the value of va or v will. log X = log . (1322) are small. The value of e is 4. This effect may be seen more directly when some numerical values are inserted.sin a 0 cos a0 = terms of higher order.303 enters because of the use of common logarithms.~ ) T ~ M $]. Equation (1320) becomes.2) . when the common logarithm of both sides is taken.2 = 90.3O'dhva the factor 2. (1322) The variation of log v. therefore.ro v. According to this equation.4 to 2. I t can be shown that ?r a 0 .2) 8e + 2. the factor is 1." ~. have a large effect on the value of the disintegration constant.sin a 0 cos a 0 is of the order unity and varies slowly with v. and the higher order terms in Eq. h = For Z .b va  where a and b are constants. 2.1361 THE THEORY OF ALPHADECAY 331 change in the argument of the exponential may cause a large change in the value of A.2) (ao .
1315 that when log X is plotted against l/v. the slopes of the lines are in good agreement with the values predicted by the theory. Disintegration constant X in secI. and the last digit in the mass number of the aemitter is given beside each point. A comparison has been made by Kaplan. straight lines are indeed obtained. I t can be seen from Fig. the points are experimental. 13 Reciprocal alpha velocity FIG. The results can also be shown in a graph like that of Fig. A squareroot scale is used for Q. 1315 so that there are more experimental data.. Logarithm of the disintegration constant us. very good straight lines are obtained. reciprocal of the velocity. The figure contains information more recent than that of Fig.ALPHADECAY [CHAP. again.ff. for even Z and A . 1316 in which the logarithm of the halflife is plotted against the "effective total decay energy.'25' who showed that aemitters with even values of A and Z give results which are in good agreement with the theory. avelocity for the aemitting isotopes of a radioactive element (Z constant). velocity in units of log cm/sec(25). the result should be a straight line." the latter is just the adisintegration energy with a small correction for the screening effect of the atomic electrons. 1315. for constant Z. The equation may be regarded as a theoretical form of the GeigerNuttall rule and may conveniently be compared wjth experiment. The relationships between the .
and odd Z and odd A is less straightforward. although it has achieved a n important degree of success. The application of the theory t o aemitters with odd Z and even A . The above discussion has been limited to the case of the eveneven aemitting nuclides in order to avoid complications which are beyond the scope of this book. The numbers give the last digit of the mass number of the alphaemitter. 1316. even Z and odd A. (1320) can be used t o calculate the value of ro. for example. For this reason and others. and most of the values of the momenta are not yet known when either A or Z is odd. The energy scale is linear in Qdi includes the electron screening correction.logous curves for even2oddA nuclides.(26) _ logarithm of the halflife and the adisintegration energy of the elements with even values of Z are shown in Fig. There are ana. 4 on the uranium scale refers to U234. and for oddZ nuclides. 1317. Eq.1361 THE THEORY OF ALPHADECAY 333  FIG. I t must be pointed out that the . the effective radius of the nucleus for aemission. If the energy and the disintegration constant are known for a given nucleus. the theory of adecay cannot yet be considered a closed subject. Halflife for alphadecay as a function of the square root of the effective total decay energy. Qdr. in part because the probability of aemission depends on the difference in the angular momenta of the parent and daughter nuclei.
The radii are found to conform to the formula ro = RoA1I3 X 10l3 cm. the values obtained from different processes agree quite well. Fortunately. as can be seen from Table 137. cm X 1013 radius of the nucleus is a quantity which cannot be measured directly and is obtained by calculation from some formula which also contains measured quantities. and lie between 10l3 and 10l2 cm. . 13 Parent nucleus Product nucleus Rsdius of product nucleus. The heavy aemitting nuclei have radii close to 10l2 cm.334 A L P H A D E C A Y [CHAP. The particular formula used depends on the experimental process under consideration. and the value of the radius obtained must be regarded as an eflective value for the given process. and were obtained from a more rigorous form of the theory. The values of the radii given are those of the product nuclei.
Ghiorso.5 7. Relationship between halflife and adisintegration energy for some eveneven aemitters (Perlman.  where Ro is a constant closely equal to 1.5 I I I 9.1361 THE THEORY OF ALPHADECAY 335 40 I I I I I t I .0 Alpha disintegration energy (Mev) 8. 1317.0 8.0 45 5. . a fact which will be useful in later work. on the regularities among the properties of aemitters. finally. a considerable amount of work. and Seaborg(27)). . It is well summarized in the general references by Perlman and Rasmussen and by Hanna. 7. This work. The radius turns out to be proportional to the cube root of the atomic weight.0 55 0. which has only been touched upon.0 6. sometimes referred to as the "systematics of aemitters" has for its object the correlation of the available experimental information about ruemission.53 and A is the atomic weight. There is.3 FIG.
RASMUS~EN. 8. Mod. S. Manhattan District Declassified Document MDDC1336." Revs. Soc. The Actinide Elements. New York: Macmillan. J. et radium 6. 183 (1936). National Nuclear Energy Series. 456. Book 2. 14B. SegrB. Heavy Elements. eds. FEATHER and R. ROBINSON. JAFFEY. 1930. R. 349 (1932). Berlin: Springer Verlag. WARD. 10S204. Vol. 1957. 0. RadMtiuns from Radioactive Substances. S. 1955. New York: McGrawHill. I. "The Ionization Curve of an Average aParticle. (London) A139. 26. M. 139 (1928). p. C. PERLMAN and J. D. (b) "Energies of Natural aParticles. and WYNNWILLIAMS. 1958. 1957. 14A. Roy." Proc. 1947."Parcours des Rayons a de l'Ionium". H. New York: Wiley. (London) A136. TSIEN. N. BRIGGS. "A 48Channel PulseHeight Analyzer for AlphaEnergy Measurements. (London) A157.. Chapter 16. ed. 10. ed. 26. 617 (1933)." Experimental Nuclear Physics. Cambridge Phil. 1954. LEWIS. I.and Fission Measurements. LIVINGSTON. HANNA. E. "Analysis of aRays by an Annular Magnetic Field. SEABORG. and WEISSBORD. Soc. ASAROand I. and M. Roy. "Table of AlphaDisintegration Energies of the F. pp. M. (a) G. E. paux Groupes des Rayons a. 6. Phys." Experimental Nuclear Physics. T.and BOWDEN. Vol." Revs. et radium 4. and Manning. PARTICULAR 1. ed. RUTHERFORD. 1226. ." The Transuranium Elements. 831. 1953. 4. 30." Proc. Vol. Soc. 1959. by E. G . STROMINGER. JR. 6. 22. "Range and Specific Ionization 5. Chapters 16. New York: McGrawHill. A. 111. "Alpha Radioactivity. Seaborg and J." Revs. BETHE and J. 2. ROSENBLUM and G. 7. NIMMO. b y Seaborg. RUTHERFORD. phys. "Pulse Height Distribution Analyzer. 162 (1945). H. PERLMAN. "The Ranges of the aParticles from the Radioactive Emanations and 'A' Products and from Polonium. 29.262 (1933). EVANS. also in National Nwlear Energy Series. T. H. 585904. Vol. phys. "Table of Isotopes. A. 1949. HOLLOWAY of AlphaParticles.  . HOLLANDER and G. W." Proc. 42. J." 4. WYNNWILLIAMS. ASHKIN. 54. "Passage of Radiation Through Matter. The Atomic Nucleus. and ELLIS. DUPOUY. Part IX. Feb. 24. "Mbure Absolue des Vitesses des Princi3.JAFFEY. 1954. Roy. Chapters 2. "Radiochemical Assay by a."A Determination of the Absolute Velocity of the aParticles from Radium C'. 3." Phys. D. J. G.GENERAL RUTHERFORD. 9." Handbuch der Physik. T. Plutonium Project Record. Vol. Phys." Nucleonics 4 50 (June 1949). 18 (1938). GLENN. New York: McGrawHill." J. Soc. Mod. Part 11. Rev. Katz. New York: Wiley. R. 5. Katz. CHADWICK." Proc. SegrB. CURIEand S. G. GHIORSO. 7. 1 (1954). Mod. "Alpha Radioactivity. I. Phys.
BLEULER. 782 (1950). 20. k BETHE. "Quantum Mechanics and Radio24. U. S.. SOC.438 (1930). GALLAOHER. Mod. Physics."LongRange AlphaParticles from 18. 379 (1916). Phys. 26 (1950). Section XVI. . B." Phil."Nuclear Physics.''The RangeEnergy Relations for Slow AlphaParticles and Protons in Air. H. 81." Phys. Experimental. Manhattan District Declassified Document MDDC1175. C. GURNEY active Disintegration. GAMOW." in Lecture Series in Nuclear 14." 2. ROSENBLUM. Lectures 11. 1 (1950). Chem. 439 (1912). RUTHERFORD Thorium. P. J. 126 (1953). 12." Nature 122. 23. New York: 12. 16. "An Analysis of the Fine Structure of 19. Factor Calculations. 0. Inorg. V. Instr." Proc. GAILAR. "The Ranges of the aParticles from 22. Roy.REFERENCES 337    . "RangeEnergy Curves for AlphaParticles and Protons. W." Phys. Rev. 21. 127 (1929). and H. Phys." Phil. Mag. "The Ionization in Argon and in Air by Single aParticles as a Function of their Energy. "Fine Structure of aRays." Phys. Nuclear 15. WOOD. Nucl. "The Theory of Stopping Power. Rw. phys. and SEABORG. Rev. "RangeEnergy Relations for 13. "Progr&s RBcentg dam 1'Etude du Spectre MagnCtique des Rayons a. 439 (1928). W. "Systematics of AlphaRadioactivity. 51. H. and TENDAM. and J." Rws. . 11. SADAUSKIS. Mag. Part C. GAMOW. 24. RASMUSSEN. BOWDEN. 24. Phys." Revs. Sci. 22. JESSE and J."The Systematics of EvenEven AlphaEmitters. 235 (1934).FORSTAT. and E. 647 (1912). 33. and J. 22. Rev. 213 (1950). F. 9. GHIORSO. BETHE. M. 77. JESSE." Phys. BLOCH. 1949. the aParticle Groups from ThC and of the LongRange Groups from ThC'. M.Phys. (London) A145. 333 (1957). 397 (1930). YAGODA." Rev. H. and A. and SADAUSKIS.204 (1928)." Nature 126. PERLMAN. 23: G. LEWISand B. 78. LIVINGSTON Dynamics. KAPLAN. E. SEIDLITZ. R. et radium 1. 613 (1911). Wiley. A. B." J . Mod. "Zur Quantentheorie des Atomkernes. W. 77. CONDON. G. 245 (1937). 31." J . JR. I. S. "AlphaDecay Hindrance 26. NUTTALL. AlphaParticles and Deuterons in the Kodak NTB Emulsion. Rev. 27. 25. GEIGER Various Radioactive Substances and a Relation Between Range and Period of Transformation. 962 (1951). 17. Radioactive Measurements with Nuclear Emulsions. 3.
the kinetic energy in ergs and Mev. The nuclide RnZ1' emits three groups of aparticles. relative to that of the rest mass. The nuclide AmZ4 emits six groups of aparticles.25 X log. 7.534 Mev.378 Mev. and 5. 3. and 10 different groups of aparticles. (b) the relative stopping power in aluminum. 0. 5.0428 Mev.582 Mev. 1. with kinetic energies of 4. 10 Mev.2. 4. respectively. 5 Mev. in kilogausscm: 393. 325. The nuclide U233 emitg six groups of aparticles. 9. and 5.0 X loQ. The aparticles coming off a t 90° were analyzed in a 180' magnetic spectrograph. The density of air may be taken as 0. 1. Find the Qvalue corresponding to each group of aparticles.5 X lo8. each with a different Hzvalue. and 0. Hollander. and 0. the range in another substance. 139 to find the mean ranges in air of the aparticles of Problems 1 and 2. In an investigation of the AlZ7(d.) 5. Use the resulting curve and those of Fig.0 X lo9. respectively. Gammarays with energies 0. and Seaborg (gen. 5 X lo8. 5. 379.655 Mev.a)Mg2' reaction. 6.435 Mev.613 Mev. 8 Mev. 369.773 Mev.477 Mev. 5. to a first approsimation.311 Mev. 4.0561 Mev. and 290.5 X lo9. 354. Find the speeds of aparticles with energies of 1 Mev. 7. 4.10Mev deuterons. The Hrvalues were. ref.717 Mev. an aluminum target was bombarded with 2.75 X lo9. 2. 8. 13 1. (c) the equivalent thickness in mg/cm2. Calculate (a) the range in aluminum of the aparticles from each isotope of polonium listed in Table 133 (use the mean range in air). Compare the results with the esperimental values given in Table 132. 2. 1 X log. Compare it with the scheme given by Strominger.489 Mev. Plot the energy in Mev against the value of Hr in gausscm.238 Mev.0 X 10lO.779 Mev. 2 Mev. .0 X 1Ol0. with kinetic energies of 5. is the atomic number of the substance z . 305. 4 Mev. and the value of Hr for aparticles moving a t each of the following speeds (in cm/sec): 1 X 108.). Calculate the mass.001226 gm/cm3 and the "atomic weight of air" as 14. (The interpretation of the occurrence of the different aparticle groups is treated in Chapter 16.4. Associated with the aparticles are 7rays with energies of 0. If the range in air is known.338 ALPHADECAY [CHAP. 6 Mev.847 Mev.169 Mev. Construct a decay scheme based on these data. with kinetic energies of 5. 5. from the empirical formula where the N's represent the number of atoms per cubic centimeter and 2. 329. Find the speed and Hrvalue of the aparticles emitted by each of the isotopes of polonium listed in Table 133. which fall within the energy limits covered by the curves. 5. were observed.0687 Mev.816 Mev. may be obtained. 5.500 Mev. 0.099 Mev have also been reported.0 X lo9.50 X log. 4. and 2. respectively. (d) the thickness in mg/cm2 equivalent to one cm of air. 323. Plot the values of the kinetic energy listed in Table 133 against those of the mean range in air. 1. denoted by the subscript z. 345. 1. and 4. 4. 3. Construct a decay scheme based on these data.
0. Sn112.0264 Mev.Ca40.. 0. The aenergy in the case of Th224is 7. Calculate the height of the barrier for aparticles and the nuclei NezO. what is the atomic mass of Po210? + .105. Gammarays are found. with energies of 0.and Th232. I t has been shown (25) that the aemission of the thorium isotopes can be described by the relation + log X = 56. Assume that the nuclear radius is given by the formula r = 1. 10lo). Construct a decay scheme based on these data. 10.Yb174. and Seaborg.0596 Mev. 0. 0. Trace the series by means of a graph of A against Z (cf. 0. The height U of the Coulomb barrier around a nucleus of charge Zle and radius r for a particle of positive charge 2 2 may be found to a first approximation by calculating the energy of Coulomb repulsion at a distance equal to the radius of the nucleus.and show its relationship to the uranium series. 0.33 Mev. Repeat the calculation for protons and deuterons.5 X 1013 cm. U = Z1Zze2/r.13 .159 Mev. The artificially produced radioactive nuclide Pa230 may be considered the parent of a collateral (4n 2) series.PROBLEMS respectively.0555 Mev. Hollander. .103 Mev.07/v. what is the halflife of Th224ag estimated from the above relation? 13. Znss.0435 Mev. Fig. where A is the mass number of the nucleus. Find its radioactive properties from the nuclide chart or the "Table of Isotopes" of Strominger.0386.0332 Mev. . What is the Qvalue of the reaction Po210 + Pbzo6 He4? If the atomic mass of Pbzos is 206. 12. 11.
This study is an important part of nuclear spectroscopy. the value of the chargetomass ratio is much greater for electrons than for aparticles and much smaller magnetic fields may therefore be used to deflect &particles. 141 The velocity and energy of betaparticles. information about the energy levels and decay schemes of light and intermediate weight nuclides as well as those in the region of the natural radioactive elements can be obtained by studying their &radiations. to serious theoretical problems. and in the theory of 8decay and its relation to other fundamental nuclear problems. Many more nuclides decay by electron emission. thorium. but also because of the conceptual problems involved. and the @decayprocesses yield information about hundreds of nuclear species not limited to those with large masses. The velocity.CHAPTER 14 BETADECAY The study of the properties of the &particles emitted by the natural and artificial radioactive nuclides has added greatly to our knowledge of the structure and properties of atomic nuclei. of @particlescan be measured by means of the deflection of the path of the particles in a magnetic field. One kind of instrument which is often used is the semicircularfocusing. the main interest is in the information that &particle emission gives about nuclear energy levels and decay schemes. These problems have been treated with considerable success.000 gauss 340 . Betadecay is important. At present. and orbital electron capture than by aparticle emission. or momentum. The most characteristic feature of the spontaneous 8disintegration of a nucleus is the continuous distribution in energy of the emitted electrons. 131). For example. The continuous energy distribution led. The details of the design of the two instruments are different because of the differences between the properties of a. as will be seen. positron emission. but only by means of a theory involving new and radical ideas. magnetic spectrograph similar in principle to the instrument used for aparticles (Fig. therefore. the early interest was centered on the series of transformations of uranium. As in the case of aparticles. not only because of its relationship to the practical problems of nuclear physics. Instead of fields of about 10. Consequently. which is in sharp contrast to the line spectra observed for aparticles. or 180°. and actinium. The emission of @particles differs from that of aparticles in respect to the spectrum of the energies of the emitted particles.and 8particles.
of the Pparticles emitted by a radioactive source can be measured with high precision with these instruments. or momenta. and when the velocities are known.99793 X 10" cm/sec. In the earlier instruments.75888 X lo7 emu/gm and the velocity of light is 2. are . while nearly all of the aparticles have energies greater than 4 Mev.!+particles. is elmo = 1. or "shaped. the 8particle.1411 . both positive and negative. The pparticles must be treated relativistically because of their large velocities.99 that of light. At the same kinetic energy. travels much faster than the aparticle. In general. THE VELOCITY AND ENERGY OF BETAPARTICLES 341  as in the case of the aparticle spectrograph. the energies of the ." field. where mo is the rest mass of the electron and e is the magnitude of the electronic charge.995 that of light. the energies can be computed. because of its much smaller mass. (135) and (136) for aparticles. the magnetic field was usually uniform. An aparticle with an energy of 4 Mev has a velocity about 1/20 that of light. are smaller than those of the aparticles emitted by radioactive nuclides.'2' The velocities. in electromagnetic units. I t is useful to express the quantities Hr and T as functions of the ratio v/c and Eq. but a 4Mev electron would have a velocity close to 0. I t was found in early experiments that &particles emitted by naturally radioactive nuclides may have velocities up to about 0. fields of 1000 gauss or less are strong enough. The formulas for the velocity and kinetic energy. analogous to Eqs. (141) may be written The specific charge of the electron. Then . Most of the 8particles have energies smaller than 4 Mev."' but in the more recent designs the focusing of the particles has been improved by using a suitably inhomogeneous.
BETADECAY .
999 The quantity moc2 is the energy associated with the rest mass of the electron and is 0.20 0.15 0. v/c AND KINETIC ENERGY OF Kinetic energy.99 0.94 0.30 0.55 0.996 0.40 0.90 0.995 0.80 0.85 0.997 0. the kinetic energy is * Values of HT and T are listed in Table 141 for different values of v/c.65 0.50 0.96 0.35 0.92 0.70 0.1411 THE VELOCITY AND ENERGY OF BETAPARTICLES 343 TABLE 141 MAGNETIC RIGIDITY (Hr) Velocity.25 0. and the kinetic energy T is plotted as a function of Hr in Fig. 141. . Hr: gausscm 0 0.98 0.45 0.998 0.75 0. T: Mev Magnetic rigidity.10 0. Hence.60 0.5110 Mev.
If the 8particles. 143.] The results obtained when the 8particles (electrons) from a radioactive nuclide are studied with a magnetic spectrograph are more complex than those obtained with aparticles. This fogging is continuous. are allowed to fall on a photographic plate.BETADECAY Kinetic energy of the betaparticles. Rev. showing the line spectrum. Betaparticle spectrum of Aulg8. 142. a fogging of the plate is observed which extends for a definite distance and then ceases a t a point characteristic of the &emitting nuclide. after being deflected. Phys.several sharp lines are sometimes seen on the plate a t positions characteristic of the emitting substance. but of nonuniform intensity along the plate. T (Mev) Fro. 87. 258 (1952). [Fan. These . FIG. In addition to this continuous pn'muty spectrum. The &particle spectrum of RaE (Nearycg)).
an e ~ a m p l e ' ~ "is ~ ' shown in Fig. 1 4 4 . The shape of the curve a t the lower energies is not known with certainty because i t is hard to make accurate measurement on the lowenergy particles. the magnetic field is varied.17 Mev. Betaspectra will be discussed in greater detail in Section 1 4 4 . The field forces the particles to move along helical paths defined by a series of disks and rings. the height and position of which depend on the nucleus emitting the particles. The solenoid produces a homogeneous magnetic: field whose direction is parallel to the axis of the cylinder. this upper limit. of the continuous spectrum is different for different nuclides. and when the emitted particles are electrons or positrons.1411 THE VELOCITY AND ENERGY OF BETAPARTICLES 345    . The continuous spectrum of @particle energies found for RaE is shown in Fig. I t consists of a long solenoid with the source of &radiation on the axis near one end and a thinwalled GeigerMueller counter a t the other end. but not all. or endpoint. 143. Another type. the numbers of particles corresponding to different energy values are obtained. about 90 cm from the source. often used. or line spectrum. These properties of continuous spectra are observed for both natural and artificial @emitters. is the solenoidal magnetic spectrometer which is a variety of magnetic lens spectrometer.e. The semicircularfocusing magnetic spectrometer is just one of several types of instruments for the analysis of betarays.. There is also a definite upper limit of energy for the particles emitted by a nuclide. when this is done. Every continuous @spectrum hrts a definite maximum. The velocity can be calculated from the . In this case no line spectrum is found and the upper limit or maximum energy is a t 1. When a line spectrum is also present. By adjusting the current in the solenoid. and the number of @particlesreaching the counter per unit time is obtained for different values of H. The counter is placed in a fixed position (i. 8particles from the source can be focused so that they make a single turn of the helix and just reach the counter. to determine the numbers of Pparticles with different energies from the darkening of the photographic plate. A velocity distribution curve (or an energy distribution curve) can be obtained by using a Geiger counter as the @particledetector instead of a photographic plate. observed for many. and hence the magnetic field." lines form the socalled secondary. the instrument is more properly called a spectrometer. as in Fig. I t is difficult. For the present it will be noted that the spectrum of RaE shows the most important features of continuous spectra. because of the continuous spectrum. the lines appear as distinct peaks superimposed upon the continuous distribution curve. 142. Since each value of H corresponds to a different value of H r and hence to a different value of the kinetic energy. @emitters. which act as baffles. For a given value of the magnetic field only those particles which have a particular velocity will reach the counter. the value of r is fixed).
but may be stronger in the region around the axis than a t the axis. the particles are penetrating enough so that solid absorbers can be used. and energy loss. ionization. the relative numbers of & particles with different velocities can be determined. instruments of this type are usually calibrated by means of electrons of known speed.8 cm and produces about 4000 ion pairs/mm of path.'5. 14 l. The energies of 8particles can be determined from measurements of their absorption in matter. A 3Mev 8particle has a range in air of over 1000 cm and produces only about 4 ion pairs/mrn of path. but velocity and energy distributions can again be obtained with the help of suitable calibration. Range. Because of the difficulty of producing a really uniform magnetic field. Aluminum foils are the most frequently used absorbers. In this way.W Counter l m ~ s / / Shutter control . the absorber foils are .BETADECAY [CHAP. or may vary along the axis. the socalled short lens magnetic spectrometer is often used. uniform magnetic field and is called a long las magnetic spectrometer. An aparticle with a kinetic energy of 3 Mev has a range in standard air of about 2. the reader is referred to a recent review. The instrument just described has a relatively long."' 142 The absorption of betaparticles. Schematic diagram o f a solenoidal magnetic magnetic field and the dimensions of the apparatus. Betaparticles are much more penetrating than aparticles and there are marked differences in the methods used for measuring the absorption in the two cases. Although the use of air as an absorbing medium for &particles is therefore impractical. The analysis of the motion of the &particles is somewhat more complicated than in the case of the uniform field. 10 cm Fro. the magnetic field is not uniform. though gold and mica are often used. In absorption experiments. For further details about these and other instruments.u. 144.6' as was the case with aparticles. In practice. however. In this case. and the velocity can then be determined directly from the current through the solenoid.
In other words. to a good approximation. A typical absorption curve is shown in Fig. and the . and the counting rate is determined as a function of the absorber thickness. and the point at which it meets the background is called the range Rb of the &particles. since it also includes the effects of the continuous energy distribution of the &particles. I t is possible t o obtain rangeenergy curves for electrons in different absorbers. The activity does not decrease to zero as the absorber becomes very thick but becomes practically constant a t a value which represents the socalled "background. or activity. placed between the active source and a thinwindowed detector. Diagram of an experimental arrangement for measuring the absorption of 8particles (Glendenid6)). decreases linearly. A ( z ) is the activity observed through a thickness s. usually a Geiger counter or an ionization chamber.145. The range RB is the distance traversed by the most energetic particles emitted. over a large fraction of the absorber thickness. Near its end. plotted on a logarithmic scale. the number of 8particles decreases exponentially with absorber thickness. A convenient experimental arrangement is shown schematically in Fig. 146 for the particles which form the continuous 8spectrum. 145. or very nearly so. and p is the absmption c o e m t . The exponential form of the curve is accidental. the absorption curve deviates from the exponential form. and of the scattering of the particles by the absorber. without absorber." There is always some radiation present which contributes t o the counting rate even though it does not represent 8particles from the source. The counting rate.1421 THE ABSORPTION OF BETAPARTICLES 347 A1 shelf arrangement Support for murce FIG. and the absorption may be represented by the formula   where A0 is the counting rate. and corresponds to the energy a t the endpoint of the continuous spectrum.
i. Furthermore. In the case of aluminum. A typical absorption curve for &particles in aluminum. and the weight divided by the cross section is equal to the product of the density and the thickness. This result follows from the fact that the range of a given j%particle depends not only on its initial energy. Thicknesses in absorption measurements are often given. This usage has certsin practical advantages. Although this ratio decreases from the light to the heavy elements.!%particles depends. FIG. 146. with Z/A = 13/27 = 0. &particles are accompanied by 7'rays which form part of the background. on Z / A . 146. is used as a measure of the thickness. as indicated in Fig.348 BETADECAY . and the resulting quantity. on the ratio of the atomic number to the mass number. but it gives results of fair accuracy and has the advantages of simplicity and speed. the range is nearly independent of the nature of the absorber.. it turns out that if the amount of absorber is expressed as the product of the density and thickness. but also on the number of electrons with which i t collides in passing through the absorber. endpoint energies of 8spectra can then be determined by absorption measurements.48. 147. which may be called the superficial density. but their weight and cross section can be measured easily. The absorption method does not give results comparable in accuracy to those obtained with the magnetic spectrometer. the effect of this variation on the thickness needed to stop /%particles is not large. in units of grams per square centimeter or milligrams per by the square centimeter of absorber.e. rm') (linear scale) Total alwxber (mgjcrn2) A typical absorption curve for monoenergetic /%particles (Glendenin(5)). The FIG. the range .\l~sorber thickness (rng. therefore. The actual thickness is mult~plied density. I t is often inconvenient to measure the thickness of very thin foils. The latter number depends on the density of electrons in the absorber or on the number of electrons per unit mass. The ability of an element to stop .
If the counting rate is plotted against the absorber thickness on ordinary graph paper. T o find where the curve meets the background many accurate experimental points are needed close to the range thickness. .  * . The particles which form a line in the discrete line spectrum often seen in &decay are monoenergetic 8particles and their absorption has practical importance. The absorption of monoenergetic &particles is somewhat different from that of the particles which constitute the continuous spectrum.40. it is probably the least reliable.2 Mev is practically linear. In any case in which the curve approaches the background very slowly it is difficult to obtain good accuracy. as in Fig. For monoenergetic electrons with energies less than 0. I t is possible to correct for these differences. with a small tail a t the end of the curve. The absorption of monoenergetic electrons with energies greater than about 0. a method that is called the visual or inspection method. the absorption curve deviates more and more from a straight line as the energy decreases. of a 1Mev &particle is about 400 mg/cm2. In the case of @+radiation the shape of the absorption curve is much the same as for @particles with the exception of certain differences in the background. comparison methods have been developed"' in which the range of the @particlesfrom a given nuclide is measured in terms of the range of a standard emitter. which is not very different from the range in aluminum.. T o avoid the difficulties of the visual method. The linear portion of the curve can be extrapolated to cut the thickness axis a t a value called the extrapolated or eflective range. the range is about 500 mg/cm2. and ranges of @+particlescan be determined with the methods used for 8particles. Quantitative results for ranges of 8particles in aluminum will be given in Section 143 in the course of the treatment of the rangeenergy curve for electrons. in gold. the curve closely approximates a straight line over most of the absorber thickness. and the absorption curve of the nuclide whose range is sought is compared in a detailed way with the curve of the standard substance. 147. The standard emitter is one with an absorption curve that is particularly favorable for the visual method. Still other methods of analyzing absorption curves have been de~eloped.1421 THE ABSORPTION OF BmAPARTICLES 349  . with Z/'4 = 79/197 = 0. and it becomes increasingly difficult t o determine the effective range by absorption methods. The range may be obtained from the absorption curve simply by inspecting the curve.2 Mev.'~' these depend on fitting the experimental absorption curve with an empirical formula which includes either the range or the endpoint energy as a parameter. The discussion so far may be applied to both 8particles (electrons) and (positrons). and this kind of absorption curve is met often. Although this method is the simplest way of analyzing the absorption curve.
and a formula has been derived for the energy lost by the particles because of ionization of absorber atoms. Betaparticles are also scattered much more easily by nuclei than are aparticles. The differences are caused. Consequently. The rangeenergy relation for aluminum has been determined and can be repre . so that their paths are usually not straight. therefore. is initially monoenergetic. The ratio of the two lossea is given approximately by  . a less precisely defined quantity for electrons than for aparticles or protons. The range is. with the result that straggling is much more marked in the case of electrons than with heavierparticles. The theory predicts in a satisfactory way the dependence of the energy loss on the energy of the @particle and on the atomic number of the absorbing element.d as distinct from (dT/dz)ioni. When an electron passes through the electric (Coulomb) field of a nucleus. the straggling and even if a beam of @particles scattering make possible widely different path lengths for particles p a w through the same thicknesses of absorber. A &particle may lose a large fraction of its energy in a single collision with an atomic electron. a rangeenergy relation can be used to obtain the maximum energies of &spectra. among other things. the energy loss by ionization.. where T is the @particleenergy in Mev and Z is the atomic number ofthe absorber. l&3 Rang~nergy relations for betaparticles. 14 The details of the paasage of @particles through matter are more complicated than is the case with aparticles. (13lo). the processes are well understood and there is good agreement between theory and experiment. When the ranges of /%particles are known from absorption experiments. an additional mechanism for losing energy must be taken into account. For high energy @particles. in aluminum (Z = 13) the radiation loss amounts to only a few percent for the energies available from e m i t t e r s . The energy loss per unit path length because of radiation may be denoted by (dT/dx). This energy appears as a continuous xray spectrum called bremsslrahlung or braking radiatiun. Eq. In general. In heavy elements such as lead (Z = 82). it loses energy by radiation. it can be said that even though the interaction between &particles and matter is considerably more complicated than is the case for aparticles. the energy loss by radiation is significant even for a 1Mev particle. because relativistic effects must be taken into account.350 BETADECAY [CHAP. by the much smaller maas of the @particleand its greater speed. The energy lost by an electron per unit path length can be calculated from theory. I t is more complicated than the corresponding formula given for aparticles.
28 (1952).RANGEENERGY RELATIONS FOR BETAPARTICLES Nuclide RaZz8 Rbg7 Nbe5 Lu'76 Co6" ZrB5 Belo I131 Endpoint energy. Modern Phys. 24. Mev Range. mg/cm2 SblZ4 Mn56 AuleS C" Ba140 Mn56 Cd115 Bizlo (RaE) N13 NaZ4 NaZ4 Srse P3Z TelZ9 MgZ7 MgZ7 Homogeneous rays YgO Bi212 (ThC) Rh'06 Pa234 Sb124 As76 Rhlo4 Mn5" Cu6Z Homogeneous rays Pr144 Rh'06 *Katz and Penfold: Revs. .
When a range is determined by an absorption measurement. and it is clear that there is good agreement among them. and R = 5 3 0 7 ' 0 .5 Mev. either by emitting a ?ray with an energy hv equal to the difference between the energies of the two nuclear states. . L. . A nucleus in an excited state can pass spontaneously to a state of the same nucleus. is now known to consist of extranuclear electrons ejected from the atom by a process called internal conversion. 148. and the corresponding range+nergy curve is shown in Fig. G. and the main features of these spectra will therefore be discussed in some detail. but of lower energy. The relation between range and energy can also be expressed by the empirical formulas where T o is the endpoint energy. .E L . energy hv .S. the energy can be obtained either from the graph or from the appropriate empirical formula. . ~ h . . .E K . (148) In Eqs. the ?'ray then ejecting a photoelectron just as it would from any atom through which it might pass. The values of the energy were determined with magnetic spectrometers and are taken from N. (147)and (148).5 Mev. shells. where EK. 1 4 4 Betaparticle spectra. . hv . . The line spectrum.TO is in Mev and R in mg/cm2. 5 Mev. The problems set by kparticle spectra are among the most important in nuclear physics. The continuous spectrum.106. The electron is ejected with kinetic K. To > 2. . (147)represents a good fit to the experimental results for endpoint energies less than 2 . . are the binding energies of the electron in the K.352 BETADECAY (CHAP. I t has been shown that this explanation is incorrect and that the process of internal conversion is caused by the direct inter * . and Eq. I t was thought for some time that the energy of the nuclear transition was first emitted as Yradiation. . . The discussion will be limited to the continuous spectrum because the pparticles which form this spectrum are the primary disintegration particles. or by giving the energy to an electron in the . Eq. .B. process is called intenal conversion by analogy with the ordinary photoelectric conversion of the energy of the Yray in an atom other than the one from which the 7ray was emitted. or secondary spectrum. (148) represents a good fit above 2.'^' The results of accurate determinations of some energies and ranges in aluminum are listed in Table 142. The differently marked points represent ranges obtained by different workers who used various methods. As noted. respectively. shell of the same atom. EL. . circular 499: Nuclear Data. 1 4 sented either by a graph or by empirical equation^.
the ordinate is proportional to the number of @particlesdetected per unit range of Hr. 149(e) for RaE with Fig. Rangeenergy curve for @particles(Katz and Penfold@)). so that this process is not a form of @decay. The momentumdistribution curve has the same properties as the energy distribution curve and is used often because it represents the actual experimental results more directly. therefore. such a spectrum is called simple and .ra are shown in Fig. 148. and the monoenergetic electrons which make up the lines found with the magnetic spectrograph will be disregarded in the rest of this chapter. Some typical &spect. the curve passes through a maximum and then decreases to zero a t a value of the abscissa which represents the endpoint energy. They can be converted to energy distribution curves by converting the values of Hr to energy values with the aid of Fig. and therefore a single endpoint energy. 142. Since Hr is proportional to the momentum. The electron line spectrum is. related to the emission of 7radiation rather than to the process of &decay and will be treated more fully in the next chapter. the nuclear charge does not change in internal conversion. THE CONTINUOUS SPECTRIJM 353 Range (mgirm') Fro. In any case. 141. The relationship between the two distribution curves may be seen by comparing Fig. the terms "@spectrum" and "continuous @spectrumvare synonymous. In each case. the curves are momentumdistribution curves. In view of the above discussion.a single continuous spectrum is found. action of the nucleus with its surrounding electrons.1441 BETAPARTICLE SPECTRA. In each case. and the abscissa is Hr. In some cases of @emission. 149.
149.354 BETADECAY [CHAP. 14 Hr (gausscm) IC) P32 FIG. (continued on nezt page) .
Fig. (d) In114 (Lawson and Cork(17). (c) P32 [Warshaw. THE CONTINUOUS SPECTRUM 355 . emits %rays as well as &particles. Rev. Math. FIG. Radium E. 149(d). (b) N13 [Siegbahn and Slatis. Fig. 149 are simple.1443 BETAPARTICLE SPECTRA. Continuous bparticle spectra.) is analogous to the emission of a single group of aparticles by an cuemitter. Fig. Thus NaZ4. as seen from the figure.149. Rev. is a naturally occurring radioactive nuclide. is a positron emitter and. 149(a). is especially interesting because of its use as a biological tracer material. (a) NaZ4[Siegbahnl 127 (1946)l. Phosphorus 32. and Appleton. Chen. Phys. Ark. 288 (1950)l. is interesting for historical reasons which will be  . 149(e). All of the spectra shown in Fig. 70. Fig. 80. 9 (1945)l.Conversion lime I loo0 5000 Hr (gaussrm) 9000 (d) 1~114 HI (gaussrm) (e) RaE  Phys. Fig. The nuclide N13. 149(c). Fysik 32A. 149(b). its continuous spectrum is similar in its general features to the other spectra.) (e) RaE (Nearycg). while the other nuclides do not emit %rays. No. while the other nuclides are produced artificially. Ast. and In1'*. The emission of the @particles may or may not be associated with the emission of 7rays.
The endpoint energies are 4. respectively. The spectrum of C13' is shown in Fig. 1. The total energy is found by integrating the product of the number of particles with a given energy and that energy. 1410.68 Mev (26%). 0. One of the isotopes of antimony (Sb12') has an even more complex spectrum made up of f i e simple spectra with endpoints of 2. The number of particles is deter . and 1. 2. together with the three simple spectra of which it is composed. 1. 16%. respectively.11 Mev.39. 14 FIG.69 Mev (7%). a t least six different 7'rays are observed. which means that there are two or more spectra with different endpoint energies and intensities. The average energy of the &particles which form the continuous spectrum can be calculated from the energy distribution curve.15 Mev.17 Mev. respectively.20.95 Mev (7%). In1".1410. The curves are again momentumdistribution curves rather than energydistribution curves.70. and RaE are 1.77 Mev.50 Mev (39y0). h'a13. one with an energy of 2. 1. These spectra can be separated by analyzing the distribution curve obtained with the magnetic spectrometer. Two ?rays have been observed along with the &particles. P32. 50 (1950)l. 77. the other with an energy of 1. and 1. Phye. 0.291 Mev (21%). The @particle spectrum of C138 [Lsnger. Many &emitters have complez spectra. mentioned later. and 31%. Rev.98.356 BETADECAY [CHAP. 1.81 Mev. The endpoint energies for Xa2'. and the relative intensities are 53%. and 0.60 Mev.
196 Mev. this number is just the area under the distribution curve.792 Mev) to an excited state of ThD. According to the result just obtained. 8. T is the energy. The latter then decays by two successive %ray emissions with a total energy of 3.195 Mev.34 Mev. 1. since no 7ray is emitted. Consider the decay of ThC to the ground state of ThD.340 Mev.792 3.15 = 0.946 = 11. the average energy found in this way is 0. This process can take place in either of two ways. and three independent observer^"^^^^' have obtained values of 0. The latter energy is 0.250 8. It can be shown that when a 8transformation takes place.20 = 11. On the other hand.946 Mev) to the ground state of ThD(Pb208). number of particles with energies between T and T with an endpoint energy of 1. the energy of the nucleus decreases by an amount just equal to the maximum or endpoint energy of the emitted spectrum.250 Mev).1443 BETAPARTICLE SPECTRA. Experiments of this kind have been performed with RaE.155 Mev or 0. there would be no such agreement. and by measuring the total heat energy produced by a known number of disintegrations. The total energy change is 2. Thorium C (Bi212)can decay by electron emission (maximum energy.17 Mev. THE CONTINUOUS SPECTRlJM 357  mined by integrating the number of particles with a given energy with respect to the energy. For RaE. The total energy change must be the same in the two cases.20 Mev to the ground state of ThD.17 Mev. If the average energy were used. respectively. and N(T) dT is the d T . The total energy change in the second mode of decay is 6. ThC" can then decay by Bemission (maximum energy. provided that the maximum &decay energy is used in the calculations. as will now be shown. to ThC'. so that the average energy is only about onethird of the maximum energy. 6. 1.000166 amu.020 Mev. 0. the average energy is     where T o is the endpoint energy. which decays by aemission (adisintegration energy. Expressed mathematically. 2.34 Mev obtained from the distribution curve.35 f 0. Another example is the decay of C14 to N" by electron emission. The average energy associated with a disintegration can be determined directly by placing a /%emitter inside a calorimeter designed to absorb all the known disintegration products. The difference in + + + + .04 Mev.203 Mev).155/931. and 0.337 f 0. These calorimetric results are in excellent agreement with the value 0.203 1. and it is. the difference between the masses of C" and N14 should be just equal to the mass corresponding to the endpoint energy of the &spectrum of C14. ThC can decay to ThC" by aemission (adisintegration energy.
It was postulated that an additional particle."" Part of the energy. Neutrinos had not been detected experimentally. In 1934. is not equal to the difference between the energy of the nucleus before and after emission.628/931.0. is produced in @decay and carries away the missing energy. .0. the neutrino. and the principle of conservation of energy seems to be violated. In view of the fact that this principle is basic to atomic and nuclear physics. which is just the same as the mass difference given by the maximum energy of the &spectrum.!?decay.15 = 1. the possibility that it might not hold for &decay presented a serious challenge to physical theory. but may have any value between zero and the maximum energy of the continuous spectrum. Many other similar cases have been studied.M(N") = m (neutron) . the concept of conservation is practically an article of faith to physical scientists and almost any other way out of a difficulty would be preferable to giving up this concept. which was applied successfully by Fermi in his theory of &decay.000165 amu. for which the value of Q has been found experimentally to be 0.000674 = 0. therefore. however. no doubt that the maximum energy of the spectrum is the one which enters into the balance of mass and energy in nuclear reactions.008983 . One way was to give up the idea that energy is conserved during a @decayprocess. (116). 145 The theory of betadecay. is based on the neutrino hypothesis first suggested by Pauli. It was. The above results lead to the following paradox: a parent nucleus in a definite state of energy emits an electron and leaves a product nucleus which is also in a definite energy state. therefore. Basis of the theory. The second solution.358 BETADECAY [CHAP. There is.1.m (proton) . From Eq. the mass difference is given by M(C14) . . and their properties must be such as to make them very hard to detect.628 f 0.004 Mev. The energy of the emitted electron. seems to have diappeared. This solution was not regarded with favor because the principle of conservation of energy had been uniformly successful in all of its previous applications and it was difficult to accept the idea that it should fail in . there seemed to be only two ways of accounting for the difficulties raised by the continuous &spectrum. as well as t o all other branches of physics and chemistry. +p + N1' + n + Q.008144 . approximately twothirds in fact. and in each case the same result is obtained. the average energy of the emitted particle is only about onethird of the maximum energy. 14 mass between C" and N ' ~ can be obtained independently by considering the threshold reaction C1' . Furthermore.
1451 THE THEORY OF BETADECAY.so that this property is not an arbitrary one. The absence of charge is in accord with the requirement of conservation of charge during @decay. Betatransformations may then be represented by the following processes: . Although it may. the piinciples of conservation of energy and momentum do not determine uniquely the energy and momentum of each particle. as they do in twobody problems. which will be discussed later in this section. and the existence of a maximum in the curve. it must account for the shape of the distribution curve. I t has been proved in analytical mechanics that when energy is divided among three particles. and the number of neutrons is decreased by one. while its mass is practically unchanged.. To be acceptable. seem strange and arbitrary to postulate the existence of an undetected particle. it must account for certain correlations. a t first. BASIS OF THE THEORY 359  *  . a theory of @decaymust accomplish more than this. each @decay process is accompanied by the release of an amount of energy given by the endpoint of the spectrum. The theory must also give the correct relationship between the average energy and the endpoint energy. however.&particle energies. When the ejected &particle is an electron. This work has finally succeeded. Under the neutrino hypothesis. based on the neutrino hypothesis. The Fermi theory is based on the following ideas. the neutrino. very small compared with that of the electron. the hypothesis that a neutrino is also emitted makes it possible to account in a general way for the continuous spectrum of .  postulated further that the neutrino is electrically neutral and has a very small mass. this hypothesis has led to a highly successful theory of @decay. and the successful experiments will be discussed in Section 148.e. When this result is applied to @decayit means that different nuclei of the same species distribute their equal available energies among the product particles in a continuous range of different ways. i. the existence of an endpoint energy. Finally. The success of the theory may be regarded as presumptive evidence for the existence of the neutrino. In positron emission the number of protons decreases by one and the number of neutrons increases by one. have succeeded in accounting for all of the features of odecay just listed. and has led to a great deal of experimental and theoretical work on attempts to detect neutrinos directly. the neutrino mass may even be zero. This disintegration energy is shared among the &particle. and the recoil nucleus. When a nucleus emits a @particleits charge changes by one unit. between the endpoint energies and the decay constants in @disintegrations. Thus. The Fermi theory of /3decay and its extensions. the number of protons in the nucleus is increased by one.
I t is shown in wave mechanics that the second contribution. . can only be zero. the resultant of these three contributions must be equal to the angular momentum of the initial nucleus A. a free particle but is bound in the nucleus by the nuclear forces. The total angular momentum of a system of even mass number would be an integral multiple of h/2r for the parent nucleus and a halfintegral multiple of h/2* for the product nucleus and &particle together. leaving a product nucleus B. which is a rotational angular momentum.360 BETADECAY [CHAP. For a system of odd mass number. (1411). the total angular momentum would be halfintegral for the parent nucleus and integral for the product nucleus and &particle together. their decay is a property of the nucleus as a whole and is not to be confused with the decay of the free neutron. or is leas than the sum by an integral multiple of h/2r. The neutron is not to be regarded as composed of a proton. in units  + . It was seen in Chapter 8 that the measured values of the total angular momentum of different nuclei can be accounted for on the hypothesis that nuclei contain only protons and neutrons. when one nucleus changes to another by emitting a &particle. + . the resultant of the spins of the electron and the neutrino is integral (I or 0. however. it is found that the total angular momentum of the system (B 8) must differ from that of A by an odd multiple of 3(h/2r). Finally. of course. Another property of the neutrino can be deduced from Eq. and that no neutrino is emitted. it is assumed that a neutrino with spin 3(h/21r) is emitted. When the three contributions are added in accordance with this rule. Similarly. (14lo). an electron. or an integral multiple of h/2r. The neutron or proton that is transformed is not. the neutrons cannot act as free particles. and a neutrino. For the total angular momentum to be conserved. The angular momentum of the system (B 8) consists of three parts: the angular momentum of B. and the spin of 8. 14 where v represents the neutrino. A difficulty with the angular momentum arises. this property is the intrinsic angular momentum or spin. or differ from it by an integral multiple of h/21r. If. the resultant is either equal to the sum of the components. but is considered to be transformed into those three particles a t the instant o f &emission. I t has been found" l' that the free neutron undergoes H e c a y according to Eq. however. Suppose that a nucleus A emits a &particle 8. when two or more angular momenta are added. and has a halflife of about 12 min. and this phenomenon will be discussed in more detail later on. In a more complex radioactive nucleus. According to wave mechanics. the proton is transformed a t the instant of &emission. the rotational angular momentum of B and B about their common center of mass. the 6rst contribution (angular momentum of B) must be equal to the angular momentum of A. Since the mass number of the nucleus is not changed by the decay process. the spin of the @particle (electron or positron) is known to be 3(h/2r). (1410) or Eq.
BASIS O F THE THEORY 361 / of h/27r). the atomic number of the nucleus is changed from Z t o Z . Without the neutrino. The total energy released. AE. (1412) + +   It was seen in Section 122 that orbital electron capture can occur provided that the mass of the initial atom is greater than that of the final atom. since neutron. I t also follows from Eq. is then integral or halfintegral according to whether that of the parent nucleus is integral or halfintegral. E. which is equal to A E . E. is divided between only two particles. and neutrino. . (1412) that the total angular momentum and statistics are conserved only if the spin of the neutrino is 3(h/27r) and if the neutrino obeys the FermiDirac statistics. while the binding energy of the electron and the recoil energy of the nucleus are very much smaller.E. The resultant angular momentum of all particles left after the decay. either from FermiDirac to BoseEinstein or vice versa.E. The neutrino hypothesis and the theory developed by Fermi apply also to the process of orbital electron capture. of the nucleus. the number of particles in the system differs by one unit before and after the &transformation. The number of particles then changes by 2 units and. must be the sum of the binding energy. In the process. very small mass. The energy corresponding to this mass difference is released. the initial and final nuclei obey the same kind of statistics. which may be represented by the equation lH1 lea + on1 V .. possibly zero. This kind of nonconservation is also unpleasant for the physicist to contemplate. The neutrinos emitted during orbital electron capture all have the same energy. and is avoided by assuming that the neutrino obeys the FermiDirac statistics. since no 8particle or 7ray is observed. only in this way can energy be conserved. of the orbital electron. Nearly all of the available energy must be carried off by the neutrino.. proton.1451 .E. THE THEORY O F BETADECAY. But A E as measured may be of the order of Mev. the recoil energy. and xrays are emitted which are characteristic of the product nucleus.1. The properties assumed for the neutrino may then by summarized as follows: no charge. product nucleus. the outer electrons rearrange themselves. FermiDirac statistics. electron. An analogous argument can be made for the statistics of a system undergoing 8decay. The last result follows from the fact that the energy AE . and the total angular momentum of the system can be conserved. spin equal to 3(h/27r). and the kinetic energy of the neu trino. whose energy and momentum are uniquely determined from simple kinematics. This difference implies that the statistics would change. electron and neutrino are FermiDirac particles. at least very small compared with the electron mass. .
about 100 times less likely than one on the upper curve. These lines. In each " . 1 4 FIG. 1411.BETADECAY [CHAP. and the lower curve is said to represent forbidden transitions. therefore. and the two curves do not correspond each t o one radioactive series. Before the Fermi theory is discussed. represent empirical rules which are analogous t o the GeigerNuttall rule in adecay. a transformation on the lower curve is. In contrast with the results for adecay. 1411. Sargent diagram for natural @emitters. 1411 gives a value of A which is about 100 times as great as the corresponding value on the lower curve. The terms allowed and forbidden are to be considered in a relative sense. called Sargent curves.(13' It is necessary to consider separately the curves for nuclides of small. the upper curve in Fig. most of the points fell on or near two straight lines. and large atomic numbers. Sargent"*' found that when the logarithm of the disintegration constant was plotted against the logarithm of the endpoint energy for the naturally occumng /%emitters. as shown in Fig. The log of the decay constant is plotted against the log of the maximum energy of the bparticles (Sargent(12)). Sargent's idea of plotting log A against log To has been extended t o artificially made @emitters and a large number of Sargent curves has been obtained. in @decaythe disintegration constant does not vary so rapidly with the energy. it will help to consider certain correlations which have been observed between the endpoint energy and the halflife or disintegration constant. and stand for different degrees of probability of spontaneous disintegration. These correlations have an important place in the systematics of piecay. The upper curve is said to represent allowed transitions. For a given value of the endpoint energy. For a given value of the energy. intermediate. .
The factor [MI2 is the square of the absolute a t r i x element. This fraction is +    Equation (1413) will be discussed in some detail.1461 THE THEORY OF BETADECAY 363 case several curves are obtained. These regularities in the experimental results mean that the theory of pdecay must account for the different relative probabilities of Bemission as well as for the properties of the energy spectrum. convert a neutron into a proton. one of which is considered to represent allowed transitions. (1413) form a statistical factor which describes how the energy T o is shared among the electron. The problem of the theory of 8decay is to calculate the probability of the processes of Eqs. F is unity. for the fraction of the nuclei which disintegrate per unit time by emitting a &particle with kinetic energy between T and T dT. . The statistical factor is responsible for the general properties of the continuous &spectrum. In most cases. The Fermi t h e ~ r ~ " ~yields . T o is the energy available for the disintegration and mo is the rest mass of the electron. the others represent transitions with different degrees of forbiddenness.' ~ ' a formula for the energy distribution of the emitted 8particles. The quantity G is a natural constant which represents the strength of the Fermi force. M cannot be calculated exactly..very slowly as Z increases up t o about 20. Results and comparison with experiment. To make the calculations it is necessary to introduce a force which can produce the changes expressed by those equations. Fermi introduced such a force and treated its operation mathematically by methods analogous to those used in the quantum mechanical treatment of the emission of electromagnetic radiation. called a m which is a measure of the relative probability that the nucleus will undergo a @transformation. and a t the same time produce an electron or positron and a neutrino. and the product nucleus. and negative when positrons are emitted. it changes . i. For Z = 0.T) is a complicated function which describes the effect of the Coulomb field of the nucleus on the emitted &particle. (14lo). (1411)) and (1412). and its predictions compared with experimental results. F(Z. Electron and positron emission are treated by taking Z positive in the formula for F when electrons are emitted.. The other quantities on the right side of Eq. magnitude of a wave mechanical quantity M. after which it varies more rapidly. i.e. but general considerations about it make possible theoretical distinctions between allowed and forbidden transitions. or vice versa.e. 146 The theory of betadecay. the neutrino. its actual magnitude will not be important in the following discussion.
F(Z. the value i s zero when T = 0. and the statistical factor shows what fraction of the available disintegration energy is carried off by the bparticle. (1413) or Eq. and it is easier to discuss than Eq... relatively short halflife and large disintegration constant. (1413) may then be described as follows. F may be taken equal to unity and M may be taken to be independent of the energy. and F1'. the average value of the energy of the emitted bparticle can be found by integration and is in good agreement with measured values of the average energy. 01'. The factor G2(MI2is a measure of the relative probability that the nucleus may emit a &particle and a neutrino. For a given value of To. increases as T increases. I t is found that. Ha. N13. : . (1413). passes through a maximum. Equation (1414) holds for such &emitters as the neutron. to a first approximation.. It shows that for T = 0 and T = To the probability of Bdecay vanishes.T) shows the effect of the Coulomb field of the nucleus on the emission of the &particle. C". Eq. When P(T) is plotted against T. and then decreases to zero a t T = To. the integration can be done analytically and A can be expressed directly in terms of the endpoint energy To. while for values of T between 0 and To. With the above approximation. (1414) over the possible values of the kinetic energy: . the theoretical decay probability may be written  where G' is a new constant. The distinction between allowed and forbidden transitions will be made clearer later in this section. the disintegration constant is proportional to the fifth power of the endpoint energy x = k ~ .(lO' The theoretical expression for the spectrum in an allowed transition thus gives correctly the most important features of experimental hpectra.e. The theoretical values of the disintegration constant A and the mean life 7 are obtained by integrating Eq. He6. w In the case of the simple allowed spectra represented by Eq. an allowed transition should be regarded simply as one with a relatively high probability.. All of the factors together give the required probability that a &particle will be emitted with the appropriate energy. Until then. P(T) is positive.364 BETADECAY [CHAP. i. (1413) may be simplified. 14 The structure of Eq. (1414). For transitions of the allowed type in nuclides with low values of Z.
as in allowed transitions of nuclei with greater values of 2. and should represent. since the slopes of the Sargent lines are approximately 5. The fraction P(T) dT is proportional to the number N(T) dT of &particles with energies between T and T dT. This method of comparing theory and experiment. It will be assumed that M is independent of the energy and can be regarded as a constant. (1413). or Fermi plot. therefore. or a t E = Eo. Equation (1413) may then be replaced by + + +  The Fermi theory predicts. the quantitative comparison between the theoretical and experimental spectra is made in an ingenious way involving the socalled Kurie plot. so that P(T) dT = P(p) dp. and ax ton. since moc2(= 0. suggested by Kurie. (1413) before discussing this method. Eq. is replaced by the momentum distribution + . If a straight line is obtained. The number of processes is the same. If N(T) is taken to be the number of &particles observed experimentally. it is good evidence for the validity of the theory.T is equal to Eo . The last equation may be written log X = log k + 5 log To. by definition. I t is convenient to rewrite Eq. In practice. or in forbidden transitions. then the plot should be a test of the Fermi theory. I t is found that this is the case. or N(T) dT = N ( p ) dp. I t can be shown that in relativistic mechanics. for corresponding intervals. at least roughly. It is then found that the energy distribution. the comparison is less direct and will be discussed later.E. that if the quantity [N(T)/FEp]'12 is plotted against T or against E. the momentum distribution of the emitted . the Sargent curves. The expression (T moc2) is just the total energy E of the emitted electron. Richardson.1461  THE THEORY OF B ~ A . A function P(p) can be defined so that P(p) dp is the fraction of disintegrations that give pparticles with momenta between p and p dp. the expression (T2 2moc2T)112 is equal to the "momentumn pc. and the Kurie plot then provides a straightline extrapolation method for finding the endpoint energy."^' has the advantage that it is independent of the actual measurement of To. the result should be a straight line which intersects the energy axis at T = To.%particles is often used.D E C A Y 365 where k is a constant. To each energy interval dT there corresponds a momentum interval dp. In practice.T) and M must be taken into account. When the effects of F(Z.511 Mev) is the rest energy of the electron. The difference To .
) FIG.BETADECAY Kinetic energy (kev) 0. (cmditd on nezf page) .1412.
(a) The F I ~ 1412. neutron (Robson(")). tritium (H3). 015. the function [N(p)/Fp2]'I2 can be plotted against T. Phys. together with many others like them.013 Mev. yields an endpoint energy of 0.782 f 0. where N(p) is the number of 8particles counted which have a momentum p. which is 0.782 Mev. (d) In"* (Lawson and Cork(")). Tritium is the simplest electronemitting nuclide. The result should again be a straight line. Kurie plots are shown in Fig. 1339 (1957)l. 1412 for some @emittersof special interest. as well as an accurate value of the endpoint energy. plots for allowed /3transformations. This value is just the one predicted from the difference between the masses of the neutron and the proton. . 1412(a). These results. (c) 015 [Kistner et al. A straight line is obtained in each case. Fig. or Ferrni.and InH4. The Kurie plot for the neutron. the neutron.00084 amu or 0. In the Kurie plot. (b) Tritium: H3 (Langer and MoffatQ7)). provide strong evidence for the validity of the Fermi theory of allowed spectra. apart . 105.. Rev.1461 THE THEORY OF BETADECAY 367 Kinetic energy (hlev) (4 h e ' Total P ~ ~ I E EV(units m i 2 ) I Kinetic energy (hlev) (4 Kurie.
therefore. The theory again provides an answer. Higher degrees of forbiddenness may be arrived at by continuing this argument. by another factor of about 100. and hence its relation to the problem of allowed and forbidden transformations. These deviations have been accounted for in a satisfactory way as a result of further development of the theory and refinements in the experimental methods. and is smaller than in the allowed case by a factor of about 100. as do P(T) and A. If the first term is not zero. the magnitude I of the total angular momentum of the nucleus . In a nuclear transformation. Fig. and A are still smaller. Fig. Fig. (MI2then has a certain value. the matrix element M also depends on the energy and affects the shape of the spectrum.010 Mev. The values of P(T) and X are also correspondingly smaller than in the allowed case. 14 from the neutron. practically the same as that of the first nonvanishing term. The question immediately arises as to why the first term. The plot for 015. The magnitude of IMI2 is. will be discussed next. Similarly. or any number of leading terms in the expansion for /MI2 should vanish. its Kurie plot. the endpoint energy is 1. I t is found that the second term is smaller than the first by a factor of about 100. is an example of the spectrum of an allowed positron transformation. For certain highly forbidden transformations. This value and the plot shown are important because. and the second term does not. IMI2. as will be discussed later. gives an endpoint energy of 17. the third term is smaller than the second by about the same factor. 1412(b). and the transformation is called first forbidden or once jotbidden. or the first and second. if the third term of the series is the first nonvanishing one. good agreement is obtained between theory and experiment.95 f 0. . It turns out that in these cases. the succeeding terms may be neglected. Finally. they make possible an estimate of the mass of the neutrino. and so on. The value of IMI2 is given by the second term of the series. The theory makes it possible to express IMI2 as an infinite series. the third and succeeding terms may be neglected. part of its effect on A may be estimated without detailed calculations. for which the halflife is relatively long and the disintegration constant small.736 f 0.368 BETADECAY [CUP. and to estimate k a rough way the relative magnitudes of the successive terms of the series. Although the magnitude of [MI2cannot be calculated accurately in most cases. was the first case"" for which unambiguous confirmation of the Fermi theory of allowed transformations was obtained. the Kurie plot for In"'." 1920' The effect of IMI2 on the disintegration constant and the halflife. When this additional energy dependence is taken into account. the shape of the spectrum departs considerably from that predicted by the simpler forms of the theory. and the transformation is called dowed. 1412(d). If the first term of the series vanishes. P. and the transformation is called second forbidden or trtrice forbidden. 1412(c). .10 kev.
for which IMI2 is independent of the energy. The selection rules derived from the theory depend on the way in which the special interaction introduced into the theory is defined.1461 THE THEORY OF BETADECAY 369 _   . Sections 73. is. at most. and there is no a pn'ori physical reason for choosing one definition over another. 4). f 1. There is. The most recent work in this branch of the field of @decayindicates that the actual interaction must be a combination of the Fermi and GamowTeller types. therefore. some arbitrariness in the theory which still remains to be removed. The first. from Eqs. and the transition is allowed. If a particular transformation obeya the selection rule. The problem of the exact definition of the interaction that leads t o bdecay has not yet been completely solved. with the difference A I appearing. a t most. and simplest. This interaction can be defined in several different ways. definition of the interaction was suggested by Fermi and led to the selection rule A I = 0 for allowed transformations. If it is obeyed. may change by an integral multiple of h/2u. This result follows from the way in which the angular momenta of the initial and h a 1 nuclei are worked into the expression for 1M12by the theory. if not. Depending on which selection rule is obeyed. and Gamow and Teller used a different definition of the interaction. the first term in the series for ]MIZ does not vanish. As in the case of atomic transitions (cf. Inconsistencies were found between theory and experiment. second forbidden. for example. first forbidden.  The integral in the last equation is a function only of Z and To. Its value may be computed by analytical or numerical integration depending on whether or not approximate expressions for F may be used. as orbital angular momentum of the electron and the neutrino. There is also a selection rule for first forbidden transitions. but nearly all of the experimental facts of bdecay can be explained by the theory. (1413) and (1415). a btransformation may be an allowed one. In either . The mean life of an allowed transition. or at least semiquantitatively. there is a selection rule which defines the values that A I may have. the transition actually is a first forbidden one. the first term in the series for IMI2 vanishes and the transition can be. which led to the selection rule AI = 0. If the selection rule is not obeyed. it is possible to treat differences between allowed and forbidden transitions quantitatively. and the problem mentioned seems to be well on the way to solution. the transition can be. or may have a particular degree of forbiddenness.'20' cannot ( ~ be calculated I n spite of the fact that the matrix element ( L I ~ accurately in most cases.
zoCa39.17C133. The halflife.7 to 3. The emission of a positron makes the product nucleus have one more neutron than proton.7 and 3.21).7. the logarithm of the ftvalue is usually considered rather than the jtvalue itself. ' . the initial relationships between the numbers of protons and neutrons. H3 (log ft = 3. In the latter nuclides.TO).1. In practice. In these cases. Since the value of IMI2 for a first forbidden transition is smaller by a factor of about 100 than the value for an allowed transition. This group includes the lightest @emitters. SO that the values of the product fTl12 should be considered. llNa21. and the number of protons exceeds the number of neutrons by one. and the product jt is called the comparative halflife. loNelg. the product fr may be used as an indication of the degree of forbiddenness. while the product nucleus 5B1 has five protons and six neutrons.18A35.74). the product of the function j and the halflife is usually calculated. If the dependence of IMI2 on energy is neglected in the case of forbidden transitions. The smallest values of the comparative halflife are found for a group of light nuclei for which log jt = 2.370 BETADECAY [CHAP. the last equation may be applied also to those transitions. with jTl12 = 0.15P2'. (1619) may be written (1420) 1/r = G21M12f(Z. I t also includes the positron emitters: 5C1l . or fr = 1 / G 2 1 ~ ( 2 . 14SiZ7. 14 case. 7N 13.22' TheYge from about 1000 sec to about 10" sec and. Thus. for convenience. 19K37. 1eS31. is often denoted by t. the mass number A is equal to 2 2 . and Eq. &I7.such as the neutron (log ft = 3. The halflife has been denoted by TlI2. the integral may be represented by a function f(Z.To). in the reaction the parent nucleus has six protons and five neutrons. Although some error is involved because IMI2 for forbidden transitions does depend to some extent on the energy. zlSe4l.693fr. however. The ftvalues of several hundred b emitters have now been determined'21' and analyzed. 12Mg23. (1421) The product jr is a quantity whose numerical value depends only on the atomic number and the measured values of the endpoint energy and the mean life. The emission of a positron has a particularly high probability in the mirror nuclides and the ftvalues are especially low. and z2Tid3. it would be expected that the value of fr would be about 100 times as large for a first forbidden t m s formation as for an allowed one. The transformations of this group with log jt between 2.7 are described as favored allowed transitions.'20.06)) He6 (log jt = 2. or the ftvalue.
.
2. (c) Mg27. (b) O14. (a) FZ0. Decay mhemes for &mitbm. (1.63 Mw) 8.] . Phyu. [Strominger.I ) z .77 Mev ( 1 6 : 8. Y M 3 6 . 8 . and S e a b ~ R .52 2.60 FIG. 1. Hollander.1413.16 .81 M w . 4. 585 (1958). 1. 30. Mod.15 Mev and lower 3.75 2. (a) Clis.11 M w 3 3.41 2. ~u.00 2.08 1. (el (0 CuM.
In 39% of the disintegrations.1471 ENERGY LEVELS AND DECAY SCHEMES 373 .02 Mev 2. an electron is emitted.59 Mev.30 Mev = 5.834Mev ?ray belong to the same transition. m 1. and is not associated with the emission of either the electron or the positron. There is. The N14 nucleus passes to its ground state by emitting a 2. an excited level of Nie4 1. listed among the general references a t the end of the chapter. a Kelectron is captured. In 42% of the disintegrations. disintegration energy is 1.015Mev yray belong to the same transition.78Mev pray and the 0. In nearly all of the captures. at least four groups of electrons are involved and a t least nine yrays have been observed. therefore. The nuclide Cus4 is a particularly interesting case of /ldecay because it emits both electrons and positrons and also undergoes orbital electron capture. The direct transition from the ground state of Mg2' to the ground state of A12' by electronemission is evidently highly forbidden. the product nucleus Nis4 is formed in its ground state. and the decay scheme is shown in Fig. they are coincidat). In 19% of the disintegrations.16 Mev.18 Mev is obsewd. Coincidence experiments show that the 1.86 Mev is the difference in energy between the ground state of 0 1 4and the excited state of the product nucleus N14. varying in complexity and in the detail with which they have been worked out. A decay scheme consistent with all of these data is shown in Fig.e.34 Mev. 1413(b).and that the 1. Several hundred decay schemes have now been determined. shown in Fig.78 Mev and about 30y0 to an endpoint energy of 1. two 7rays have been observed.66 Mev. but in a small fraction of the Kcaptures. In about 0. The decay scheme is shown in Fig. The electron decay of C13' has three groups of electrons. 1413(f). The product nucleus Zne4 is formed in its ground state.015 Mev. 1413(d). The decay scheme of Cus4 is shown in Fig.30Mev ?ray. ?'rays are observed with energies of 0.834 Mev and 1.6% of the disintegrations 014undergoes a transition directly to the ground state of N14 by emitting 4.57 Mev.59Mev &ray and the 1. The examples of decay schemes that have just been discussed represent a small fraction of the schemes that have been analyzed. + + . 1413(c).. and the reader is referred to that by Strominger et al. a 7ray with an energy of 0. A still more complicated scheme is that of 5. I t has been shown that the Yray is observed only in coincidence with the orbital electron capture. 1413(e) . a w a y is observed with an energy of 1.1Mev positrons. the product nucleus Nie4 is formed in its ground state. respectively. (i. as discussed in Section 144. Compilations of these schemes are useful in many phases of nuclear physics.34 Mev above the ground state.. The electrondecay of Mg2' is more complicated: about 70% of the disintegrations correspond to an endpoint energy of 1.84 Mev of which 2. and in less than one percent of the disintegrations. a positron is emitted with an endpoint energy of 0.La"'O. the 13spectrum is simple and the endpoint energy is 0.
imply that the interaction between neutrinos and matter is extremely small. determine4 from Qvalues for nuclear reactions or from atomic masses. or an (n. Ektimater.r)c2. a nuclear reaction induced by neutrinos was observed.M(He3) between the two equa . in 1956. ! x where mo is the rest mass of the electron or positron. of this mass can be obtained from two types of experiments.p) reaction is combined with a ~+ernission..0181 f 0. This success encouraged attempts to show experimentally that neutrinos exist.or' the momentum and energy balances in &decay. For example. an upper limitfor the rest mass of the neutrino. To. To = (AM . The greatest kinetic energy with which a &particle can be emitted is. (1422a) I  where AM is the difference in mass between parent and daughter atoms and p is the rest mass of the neutrino. an inverse Bdecay. for electrons. We consider first the vroblem of the mass of the neutrino. and the experimental search for the neutrino has ended successfully.n)He3 with Q = . We have elimination of the difference M(H3) . comparison of observed maximum energies of &ray spectra with known available decay energies and the shape of p r a y spectra near the endpoint. It has also been possible t o deduce. in particular the zero charge and very small mass.2mo . For pwitron emission the relation ' To = (AM . and it is not surprising that attempts to detect ionization caused by neutrinos failed. Comparie~of t observed endpoints. Finally. from nuclear data. been sought from stud. Additional indirect evidence for the existence of neutrinos has. BETADECAY [CEAP~~ The success of the Fermi theory of &1ecay in accounting for the properties of 8transformations provides indirect evidence of the existence of neutrinos.p)c2.n) reaction is combined with a 8emission.0.'23' But the properties which the neutrino must have to make it fit the requirements of the theory. therefore. with atomic mass differences AM.764 f 0.001 Mev can be combined with the @decay of H3 with To = 0. can be used to estimate the neutrino mass. The smallness of this interaction is emphasized by the fact that the discovery of the neutron with a mass very much greater than that assumed for the neutrino presented some difficulty. the reaction H3(p. The most direct comparison is that with a reaction which is just the reverse of a &decay: a (p.374 148 The neutrino.0002 Mev.
Sj THE NEUTRINO 375 tions and insertion of the values of Q. !The simplest reaction in which to study the nuclear recoil is orbital electron capture.03mo or about 1.5 kev. About a dozen similar examples involving light nuclei can be used. There are a number of reactions in which indirect evidence of the mesence of the neutrino has been obtained from the momentum given to 1.V. velocity. p represents momentum. however. The velocity with which the nucleus recoils is small because "of its enormous mass relative to that of the neutrino. With the assumption that a single neutrino is emitted the reaction is z + l X A ]e" . z Y A t. must be treated relativistically because its very small mass must be associated with high speed to give kinetic energies of the order of Mev. and E. The neutrino. To and the neutronproton mass difference gives p2= 0 0. One of the objects of recoil experiments has been to test the hypothesis that a single neutrino is emitted in a @decayprocess. In Fermi's formulation of the theory of @decay.14. its total energy Eu can be then written [see Problem 9(d) of Chapter 61 By Eq.Olmo. = so that E. recoiling nucleus. or p = 0 f 0. pv being that of the neutrino.03mo. Analyses of this kind'24' yield p 5 5 X 104m0. are the mass. The precise theoretical shape of an allowed @ spectrum in the neighborhood of the endpoint depends on the actual value of the rest mass of the neutrino. all of which yield limits below p = O. ( 1423) + If the initial nucleus is a t rest. = [ ( p ~ ' ) ~ the kinetic energy of the recoil nucleus is given by a + and . and kinetic energy of the recoiling nucleus.0015 Mev. with the limits depending on the uncertainty in the Qvalues and the endpoint energies. v. or a rest energy of 250 ev or less. conservation of momentum requires that where M. respectively. and its kinetic energy is just +Mu2. In this case the upper limit for the rest mass of the neutrino is 0. and comes from the uncertainty in the Qvalue. and comparison of theoretical and experimental spectra near the endpoint can be used to obtain estimates of the rest mass of the neutrino. the only condition on the rest mass of the neutrino was that it be small compared to the rest mass of the electron. p. (1424).
Eq.004 Mev for this mass difference has been obtained from the Qvalue of the C137(p. If the assumption that a single neutrino is emitted is correct. pcZ is generally muoh smaller than Ev. A number of orbital electron capture experiments have been made in which the energy spectrum of the recoiling nucleus has been m e p ~ r e d .67 f 0.M(Y).C137 mass difference.or ell. The recoil energy has been r r e a ~ u r e d ' ~ in~several ' ~ ~ 6 ) 0.~ nuclides ~' studied were Be7.Conservation of energy in orbital electron capture requires that where BKJ. on the other hand. to a good approximation. and note that mo = 5.n)A37reaction and the neutronproton mass difference. in units of moe2 = 0.L so that. E.08 ev.05 e ~ . The energy E. A37 and Cd1O7. is the binding energy of the electron in the K.816 f 0. ' ~ respectively.a value of 0. there should be a continuous distribution of recoil energies. and E. (1426) becomes It is convenient to express E. since the entire energy of the transformation is carried off by the nucleus and the neutrino. 0428) + The mass difference can be obtained from Qvalues for nuclear reactions or mass spectroscopic data. two or more neutrinos are emitted.6 and 9. = 9. Ev = M(X) .2 edZ8' independent experiments with results of 9. (1428) becomes where M is the atomic mass of the recoiling nucleus in amu. ' ~the ~.7 f 0. Equation (1431) then gives E. the recoiling atom should have a single energy. and the k t equation may be written If we now express M in atomic mass units. In most neutrino experiments. is given by the A3' . If.488 X amu. ~ ' These results show clearly that .65 rt 0.511 Mev. > > Er BK. is then known. ' 9.8 e ~ . which decays exclusively to the ground state of C137. and Eq. The simplest of these cases is that of A37.
Li7 is left in an excited state in about 10% of the disintegrations and a 0. i.31) and 55. can be written . The successful detection of neutrinos was made possible by the availability of a source of enormous numbers of neutrinos (a nuclear reactor) and of extremely sensitive radiation detectors. (14lo).478Mev 7ray is emitted.0 e ~ . The interpretation of the continuous spectrum. (1410) by transposing the electron (lea) to the left side of equation.3 f 0. and electrons which have not yet been discussed and it will be necessary to use some information from later chapters. The details of the experiment to be described involve some properties of neutrons. was complicated by experimental difficulties. In &particle emission the momentum conditions are more complicated than in electron capture. These electronneutrino angular correlation experiments yield information concerning the form of the interaction involved in @decay.6 f 1.5 ev energy if a single neutrino is emitted. The inverse of the @emission process. the results were consistent with that hypothesis. e.864 f 0. noting that . (14ll). because the recoil momentum is determined by the proportion in which the decay energy is shared between the electron and the neutrino and by the angle between the two.Li7 mass difference is 0. The Be7 . the force which converts the neutron into a proton and a t the same time produces an electron or positron and neutrino.. from conservation relations.e. recoil energy spectra were found with maxima of 56. however.(leo) = leal and reversing the direction of the reaction. Eq. ' in ~ ~ good ' agreement with 57. The 10% branch going to the excited state of Li7 should show a continuous recoil spectrum extending from nearly zero to 57.'33' This problem is beyond the scope of this book and will not be discussed further. Eq. the result is The inverse of the 8+emission process..003 Mev. The nuclear reactions which appeared most likely to permit the detection of neutrinos are the inverse @decayprocesses. Experimentally. I t is possible to measure the angle between the @particle and the recoil nucleus and.'  1483 THE NEUTRINO 377  a _ monoenergetic single neutrinos are emitted. in goy0 of the decays the recoils should be monoenergetic with 57.9 f 1.3 f 0. The recoil experiments with Be7 and Cd107 are complicated by the emission of 7rays. although it did not prove conclusively that only one neutrino is emitted in orbital capture in Be7.3 ev and resulting from the nearly simultaneous emission of a neutrino and a 7ray.0 ev'30.g. to deduce the angle between electron and neutrino. is obtained from Eq.5 ev. positrons.
They therefore decay. containing too many neutrons for the number of protons. The spin and momentum together define the sense of a screw. in accord with relativistic quantum mechanics. (1433c) The antineutrino is emitted in @decay. (1433s) orbital electron capture: p+8+n+v. (14lo). I t was suggested. and (1432) @emission : n + P t 8. the equations are written in simpler notation than that of Eqs. reactions of the form of Eq. liberating antineutrinos in the process.~~~~~~~~' work which will be discussed in the next section has shown that a distinction must indeed be made depending on the relation between the spin and momentum of the neutrino. the neutrino is emitted in @+decay and orbital electron capture. and the antimutrim (v*) as a particle with spin opposite in direction. are split into two fragments of intermediate atomic mass (Chapter 19). The choice shown in the following equations for the direct pprocesses is the one which has been generally adopted. The fragments are highly unstable. (1434a) should occur.BETADECAY and the inverse of orbital electron capture can be written Before proceeding with the discussion of the neutrino detection experiments we must consider an additional complication. nuclei of a fissionable material. or antiparallel. (14ll). The neutrino ( v ) is now defined as a particle with its spin vector parallel to its momentum vector. U235.v*.4. that the neutrinos involved in @decay are different from those involved in @+decay and orbital electron capture. to stable or longlived fission products. vkn+p+@. inverse orbital electron capture: In a nuclear reactor. The choice of which particle is called neutrino or antineutrino is an arbitrary one. to its momentum. The inverse reactions are inverse 8: inverse @+: v* +p + n + @+1 (1434a) (1434b) (1434c) v+n+~+@. the neutrino acting like a righthanded screw and the antineutrino like a lefthanded screw. by several successive @emissions. Recent theoretical and experi~nental'~'. for example. If these antineutrinos are permitted to bombard a hydrogenous material. and .
the annihilation of the positron produced is signalled by a prompt 7ray pulse. A delayed pulse should appear several microseconds after the prompt pulse because of the capture of the neutron in the cadmium after being slowed down. and coworker~'~~ for .88 f 0. When it is slowed down it can be captured in an (n. The liquid scintillator was surrounded with 110 photomultiplier tubes connected in parallel in each tank. A typical event is shown in the figure. liquid. The time required for the slowing down of the neutron can be calculated and the pulses due to the prompt disappearance of the positron and the delayed capture of the neutron are counted by a delayed coincidence method. This mutual annihilation takes place in a very short time in matter and prompt ?rays are formed.22 counts/hr with a signaltobackground ratio of three to one. This detection is possible because of certain properties of neutrons and positrons. and that the delayed pulses were due only to the capture of neutrons in cadmium. ) The experiments of Reines. this result has been accepted as convincing evidence of a nuclear reaction induced by antineutrinos. A prompt pulse appears in both the top and central scintillation counters because of the two 7rays from the annihilation of the positron a t point 2. A schematic diagram of the detector is shown in Fig. hydrogenous scintillation counter (terphenyl dissolved in triethylbenzine) containing cadmium. ~ reactions. the two interact in a very special way. 1414. The final result was a maximum neutrino signal rate of 2. Cadmium is such a heavy material with a very high probability for capturing slow neutrons in ( n . When a positron approaches an electron. is captured in cadmium a t point 3 with the appearance of several capture 7rays and a delayed pulse.1481 THE NEUTRINO 379  " ' the problem is to detect the simultaneous production of a neutron and a positron. which can be observed with a scintillation detector. . An antineutrino induces an inverse 8decay a t point 1. they disappear and two 7rays (photons) appear in their place (Chapter 15). When an antineutrino interacts with a proton. which is close to point 1. especially if a relatively heavy material is present. In the later experiment'3B'the detector consisted of a multiplelayered arrangement of three scintillation counters and two target tanks with one of the latter consisting of a polyethylene box containing a water solution of cadmium chloride. to show that the prompt pulses were due only to the annihilation of positrons. The effect of fast neutrons from the reactor had to be ruled out along with other possible spurious effects. Cowan.r) reaction. ~ the " detection of antineutrinos were based on the use of a large. after undergoing several collisions. To show that the prompt and delayed pulses were due to a reaction induced by antineutrinos it was necessary to relate the pulse rate to the power level of the reactor and to the number of protons in a target box. A neutron produced in a hydrogenous material undergoes elastic collisions with protons and loses energy (Chapter 18). The neutron.
A small measured volume of argon gas was introduced into the tanks before sweeping to serve as a carrier and to measure the recovery of argon. were exposed to antineutrinoa from the same reactor used in the experiment of Reines and Cowan.BETADECAY . 3000 gallons of carbon tetrachloride. 38. however. (1434~) shows that this reaction would be the inverse of orbital electron capture in A3' if the neutrino and antineutrino were identical. theae particlea are Merent and the antineutrinos from a nuclear reactor should not induce the reaction (1435). is much less probable than the reaction ~ * ( ~ . The value found'42' for the . At the end of the irradiation the CCll was swept with 12. quantitative measure of nuclear reactions is the moss seetion (see Section 163) which has values of the order of lo'' cm2 for most nuclear reactions.000 liters of helium gas to remove argon. The results showed that the induction of inverse Bdecay in C13' by antineutrinos.41' in which an attempt has been made to detect the reaction Comparison with Eq. free of argon. and its &activity was counted in a small Geiger counter. the reaction (1435).1414. According to theory. An experiment complementary to that on the detection of antineutrinos has been in progress for some time"0. In the most recent version of the experiment.* Target tank with cadmium FIG. n ) @ The + . The argon was then separated from the helium in a charcoal trap cooled with liquid nitrogen. Schematic diagram of the antineutrino detector used in the experiment of Ref.
depend explicitly on L. and to the relationship between mass and energy.2 X lo'' cm2. into the very details and language of atomic physics. a symmetry or invariance principle (see Sections 64 and 65) generates a conservation law. the quantum numbers which designate the states of an atomic system are often closely related to the numbers which represent the symmetries of the system. n ) reaction p+ is (11 =t value of the cross section was found to have an upper limit of 0. We shall follow the discussion by Yang. the invariance under a rotation in space has as a consequence the law of conservation of angular momentum. These results show that the probability of antineutrinoinduced inverse orbital electron capture is much smaller than that of the absorption of antineutrinos by protons.'") that. t o the neutrino hypothesis and eventually to the detection of the neutrino. Since one electron cannot be distinguished from . In recent years it has been discovered that an important symmetry law. A second example involves a kind of symmetry typical of quantum mechanics. breaks down in certain reactions. the 2. They show also that the probability for antineutrinoinduced reactions is about twenty orders of magnitude smaller than that for ordinary nuclear reactions. to the variation of mass with velocity."^' Thus. The development of quantum mechanics extended the use of symmetry principles still further. that of rightleft symmetry. During the first half of the 20th century it was recogni~ed'~" in general.for the reaction (1435). as we have seen. 149 Symmetry laws and the nonconsemation of parity in &decay. and the requirement that the equations of a physical theory satisfy this type of symmetry principle became a fundamental condition on a physical theory.5) x ~ * ( ~ . which embody the conservation of angular momentum. Thii new type of invariance is related to the impossibility of detecting absolute motion. In the theory of relativity the concept of invariance was extended to invariance under the Lorentz transformation. the invariance of physical laws under a displacement in space has as a consequence the law of conservation of momentum. One example is that of the total orbital angular momentum quantum number L of the electrons in an atom. To see how this comes about it is necessary that we consider the relationship between symmetry or invariance principles and conservation laws. and orbital angular momentum is conserved. The atom is symmetrical with respect to a threedimensional rotation about the nucleus.1401 SYMMETRY LAWS AND NONCONSERVATION OF PARITY 381 cm2. For example. and the consequences of these principles appear explicitly in the methods and results of quantum mechanic^. the socalled weuk interactiuns. the Schroedinger equation for the atom must be invariant under such rotations and the detailed solutions of that equation. of which @decayis an example. The problem of the conservation of momentum and energy in Bdecay led.
It will be recalled from Section 79 that parity is a quantum number which takes on the value +1 or 1 depending on whether the Schroedinger wave function is unchanged or changes sign under a reflection through the origin of the spatial coordinates. Another basic type of symmetry is that between right and left. we expect that the laws of nature are invariant under reflection.g. the Dirac equation. or symmetry under a reflection. an experiment made in a laboratory can also be made in the way it appears as seen in a mirror. the concept of parity and its conservation was taken over into nuclear physics. Since rightleft symmetry was unquestioned in interactions other than electromagnetic. and experience seems to support this idea. antineutron. the motion of any object as seen in a mirror is also a motion which would be permitted by the laws of nature. must be invariant under the Lorentz transf~rmation. i. More precisely.48' e. Experimentally (in the study of atomic spectra) it has been found'50' that in an atomic transition with the emission of a photon the parity of the initial state is equal to the total parity of the final state.382 BETADECAY [CHAP. In fact. an atom is symmetrical with respect to the interchange of electrons.e. and their discovery testifies to the power and beauty of the invariance principles. In quantum mechanics it has been shown"Q' that the consequence of the rightleft symmetry (invariance under reflection) of the electromagnetic forces in the atom appears as the conservation of parity. in particular into nuclear reactions and Ddecay. The principle of invariance involved may be stated in the following way: Any process which occurs in nature can also occur as it is seen reflected in a mirror. "the Schroedinger equation for an nelectron atom is invariant under the symmetric group of order n. The existence of the positron was anticipated theoretically by the requirement that the quantum mechanical equation for the electron." and the degree or order n of this invariance is closely related to the multiplicity of the spectral lines. 14 another. to the product of the parities of the final atomic state and the photon emitted. The combination of relativity and quantum mechanics leads to an especially striking example of the fundamental role of symmetry laws. has a similar theoretical interpretation. In more formal terms.'~~' The existence of other antiparticles. This invariance turns out to be related to the multiplicity of the spectral terms of the atom. the mirror image of any object is also a possible object in nature.'47. and into the analysis of reactions . and any resulting effect will be the mirror image of the actual effect... the law of force between the nucleus and the electrons. antiproton. the general structure of the periodic system (because of the way it is built up in terms of quantum numbers) may be said to be a consequence of the spatial symmetry of Coulomb's law. antineutrino. The Schroedinger equation which describes an n4ectron atom must therefore be invariant under a transformation which involves any permutation of its n electrons.
these forces are referred to as strong interactions. I t is convenient to use a dimensionless quantity to compare the magnitudes of forces. Among these particles are the T and 9 mesons which have been discovered in recent years. The force involved is that between charged particles and is proportional to e2. electron. This surprising possibility led them t o investigate the experimental information concerning parity conservation. the concept of parity and the law of conservation of parity proved to be very fruitful. or approximately lo'. They found impressive evidence for the validity of this law in atomic physics. Lee and Yang(34) showed that the results would fall into a consistent scheme only if parity were not conserved in these decay processes. It has been mentioned that Bdecay is an example of a weak interaction and we must define what is meant by this description. the domain of electromagnetic interactions. and whose properties have been the subject of intensive study. where g is a constant related to the constant G of Section 146. There were. (where h = h / 2 n ) . those linking the neutron. To do SO. we must compare the forces involved in &decay with those involved in other reactions. only if these processes were not invariant under reflection. and it is for this reason that they are called weak interactions. the analogous coupling constant of Newtonian gravitational forces is of the order Weak interactions occur not only in Bdecay but also in the decay of certain particles found in cosmic rays and nuclear reactions at high energies (several hundred Mev and higher). may be characterized by the coupling constant g2/hc = 1013. The nuclear forces may be characterized by a coupling constant f2/hc = 1. and the one which enters naturally into atomic theory is e2/hc = 1/137. I n these developments.04.1491 SYMMETRY LAWS AND NONCONSERVATION OF PARITY 383 involving new particles such as mesons and hyperons (which we do not treat in this book). The interactions with which we are most familiar so far are those in which electromagnetic radiation is emitted (see Sections 72 and 78). and in nuclear physics when strong interactions were concerned." The forces binding neutrons and protons together in the nucleus are much stronger than the electromagnetic forces and also much more complicated (they will be treated in Chapter 17).. proton. and neutrino (or antineutrino). This quantity is spoken of as the "coupling constant for electromagnetic interactions. but detailed analyses indicated that the final states had opposite parities. and the success accompanying their use was taken as support for the validity of rightleft symmetry. no experimental data relevant to the question of the conservation of parity in . They are very small compared to the nuclear and electromagnetic forces. i. however.e. The forces in pdecay. where f is a constant analogous to e. Results were obtained which seemed paradoxical in that all the physical properties of the parent particle in decay and Sdecay seemed to be identical. For comparison.
1415.'~" and the nuclei are said to be polarized.BETADECAY Mirror Counter Fro. At very low temperatures. the nuclear spins are oriented in all directions because of the thermal motion and the electron8 are emitted in all directions. and which contain weak interactions. schematic diagram of an experiment to test the conservation of parity in &decay (Yang(43)). less than O. however. weak interactions. the nuclei can be oriented in a particular direction. 1415. Lee and Yang proposed some experiments. the external field can force the spins into a given dire~tion. as shown in Fig. This can be done because the nuclear spin gives rise t o a magnetic moment which can be acted upon by a magnetic field external to the nucleus.l°K. The practical feasibility of such an experiment depends on the following facts. one of which involved the Bdecay of Co60. all the nuclei can be forced t o align their axes of rotation parallel to a given direction so that they all rotate in the same sense. The basic idea of this experiment is t o build two sets of experimental arrangements which are mirror images of each other. The experiment then consists in determining whether the two arrangements always give the same results in terms of the readings of their meters (or counters). The low temperature *   . The cobalt nucleus has a spin angular momentum. it rotates with a welldefined angular momentum when it is in its normal state. to test invariance under reflection. Under certain conditions. no experiment had ever been designed specificially to test invariance under reflection in the Wecay of nuclei or elementary particles. In a piece of CosOunder normal conditions.
The circles represent samples of Cow and the arrows represent electrons. The nuclei in a given sample are aligned so that they spin in the sense indicated by the curved arrow. In the image in the horizontal mirror. the electrons are emitted isotropically. if there is no preferential emission of @particles.the mirror image would give the same result.1416. 1416. and then see whether the Pparticles were emitted preferentially in one direction or the other along the axis. Cobaltfidecay: possible electron emission patterns and their mirror images.1491 SYMMETRY LAWS AND NONCONSERVATION OF PARITY 385 Mirror Y FIG. the sense of rotation is the same and the electron distribution is . the mirror image of the experiment would give preferential emission in the opposite direction. In case (a) there is no preferred direction of emission. is needed to reduce the thermal motion of the nuclei to a minimum and make the orderly alignment possible. with the arrowheads showing the direction of emission. Possible emission patterns are shown in Fig. The experiment proposed by Lee and Yang was simply to line up the spins of the electronemitting nuclei of Co60 along the same axis. If there is preferential emission in one direction.
The electron emission is anisotropic and the mirror image shows more electrons being emitted in the direction .'36' Cobalt60 emits an electron of energy 0. Ambler. and Hudson in their demonstration of the asymmetric 8particle distribution from polarized cobdt nuclei.Reentrant vacuum space / Anthracene crystal FIG. Hoppes. It was known'51' that Co60 nuclei can be polarized in cerium magnesium (cobalt) nitrate and the degree of polarization determined by  .(36) again isotropic. Caws (a) and (a') show the results expected with righeleft symmetry and conservation of parity.BETADECAY Lucite rod . The experimental arrangement used by Wu. more electrons are emitted in the direction opposite to that in which the nuclei are aligned. This kind of result would be expected if parity is not conserved. The CoaO decay experiment was made by Wu and coworkers. Hayward. The image in the vertical mirror looks just like the original turned upside down. electron emission is again isotropic and the mirror image of the experiment is identical with the experiment itself.1417. the direction in which the nuclei are aligned is rotated through 180". In case (b).312 Mev in over 99% of its decays and two other electrons in the remaining decays. In (a'). several ?rays are also emitted.Pumping tube for . The mirror image experiment would then give results different from those of the original experiment.of the nuclear spins.
1418. 1417. memuring the anisotropy of the w a y s . additional difficulties were introduced by the need to put a &particle counter into the cryostat in which the ample was to be cooled. two KaI w a y scintillation counters were used. a  T i e (min) FIG.1 p and down 8asymmetr~ (At pulse height 10 v)Exchange gas in w  ma  1 0.2 0.(a6) . a thin anthracene crystal was located inside the vacuum chamber near the cobalt source. and to locate the nuclei to be polarized in a thin surface layer. A schematic diagram of the part of the cryostat of interest is shown in Fig.1491 SYMMETRY LAWS AND NONCONSERVATlON OF PARITY 387 Gamma anistropy (a) Fhuatorial counter 7 0. T o detect the &particles. The cooling is accomplished by a special method of lowtemperature physics called adiabatic demagnetization. To measure the extent of the polarization of the Coo' nuclei. In the @decayexperiment. The results of the experiment of Wu et al. one in the equatorial plane and one near . The scintillations were transmitted to a photomultiplier located a t the top of the cryostat. . on the asymmetric distribution of the pparticles from polarized cobalt nuclei.
  .388 BETADECAY [CHAP. the principle of rightleft symmetry is violated. 1418. In other words. . the electrons are emitted in a preferred direction. The observed 7ray anisotropy provided a measure of the polarization and hence of the temperature of the sample. 14 the polar position. i. Many experiments have now been made on Bdecay and on the decay of mesons and other fundamental particles'37' with results similar to that of the Co60 experiment.  . and parity is not conserved. after about 8 min the nuclei have become sufficiently warm so that their thermal motion causes the alignment to become random.e. and there was no doubt about the result. The breakdown of the law of conservation of parity has led to a modified theory of the n e ~ t r i n o ' ~ and ~~~ to * extensive ' research into some of the basic concepts of especially into the concepts of symmetry and invariance and their relation to conservation laws. the experimental result corresponds to case (b) of Fig. Samples were made by taking good single crystals of cerium magnesium nitrate and growing on the upper surface an additional crystalline layer about 0.Ol°K. The time scale is actually a temperature scale measuring time from the instant when cooling is stopped. the external field is used to line up atoms (provided by the cerium magnesium sulfate). The cerium magnesium sulfate is needed because the field produced by an external magnet cannot. the electron intensity in one direction along the axis of rotation was found to be 40% greater than in the opposite direction. it indicates the number of electrons emerging in the direction of the magnetic field (the direction in which the nuclei are aligned). with special properties. by itself. The results of the experiment are shown in Fig. The curve labelled "gammaanisotropy " is a measure of the extent of polarization. and the magnetic field produced by these atoms in turn. the magnet used in the cooling process was turned off. in the direction opposite to that in which the nuclei are aligned. After the material in the cryostat was cooled to about O. 1416.002 in thick.. and the number emerging in the opposite direction. The size of the effect is remarkable. a vertical solenoid was raised around the lower part of the cryostat within 20 sec after the demagnetization. containing a few microcuries of Co60. The curve labelled "B asymmetry" is the significant one. lines up the nuclei. align the nuclear magnetic moments due to the spin. and the counting was started. and the 7anisotropy disappears. It is evident from the figure that more electrons are emitted in the direction opposite that of the magnetic field.
Mod. Rev. Fierz and V. Rev. 11. J. Weisskopt. R. D. STROMINGER. KOFOEDHANSEN. J. "BetaRay Spectrometers." 3. 11. L. "An Electron Lens Type of BetaRay Spectrometer.and GammaRay Spectroscopy." Advances in Electronics. ELLISand W. New York: Wiley. KING. NEARY. 26. Phys.REFERENCES GENERAL  RUTHERFORD. "The Average Energy of Disintegration of Radium E. E. New York: The Macrnillan Co. "The Neutrino".(London) A117." 2." Nucleonics 2. 83. L. 9.. and HINE. 21. New Jersey: Princeton University Press. MARINELLI. C. Roy. COOK. Phys. 12 (Jan. (London) A175. J."A High Resolution Nuclear Spectrometer. Emitted by Radioactive Isotopes. GLENDENIN. A Memorial Volume to Wolfgang Pauli. D. Vol. ORTHMANN. New York: McGrawHill. 585 (1958). 32 (1941). 1055. Vol. Princeton. New York: Academic Press (1953). Physik. K. KATZand A. 29. Theoretical Physics i n the Twentieth Century. LANGER and C. 97. ALLEN. M. W. 257 (1948).. 24. and SEABORG. New York: Interscience Publishers Inc. A. Amsterdam: NorthHolland Publishing Co. 14. 31. Phys. 773 (1957). SIEGBAHN. Znstr." Proc. "The &Ray Spectrum of Radium E. "The Radioactive Decay of thc Neutron. "BetaRays. CHADWICK." Phys. Sci. 19. "Determination of the Energy of BetaParticles and 5. 1960.. . Mod. 17. R. M. M. 327 (1954). SOC. The Neutrino. The Atomic Nucleus. BRINCKERHOFF." Proc. Radiations from Radioactive Substances. "Table of Isotopes. 2. Chapters 12. 1958. 1948). "Table of Total BetaDisintegration Energies. 4. Wu. C. R. "Uber eine Absolute Bestimmung der 8. New York: Interscience. 349 (1951). 143 (1930). Chapters 6. 60. LIDOFSKY. HOLLANDER. W. and ELLIS. "Average Energy of BetaRays 10. 19. ~ ' ~ E I T N E and Energie der primaren j3Strahlen von Radium E. 30." Revs. L. "RangeEnergy Relations for Electrons and the Determination of BetaRay EndPoint Energies by Absorption. Soc. 20. PARTICULAR 1." Rev. 28 (1952). Beta. Mod. Part XI. E. 109 (1927). ed. 1930. Phys. eds. Photons by Absorption. WITCHER. Mod." Rev. PENFOLD. Segrh. 6 (1940). Mod. L. G. M. 71 (1940). Znstr. p. 1959. 60. 7. S. J. 18." Experimental Nuclear Physics." Revs. W. L." Revs. R W.. Phys. 19. 25 (1947). ROBSON. 1955. LAWSON and A. 111. S. 13. Roy." Revs. HAYWARD. 5. S. WOOSTER. TYLER." Revs. ed. F."The Design of a Magnetic BetaRay Spectrometer. J. S. Sci. Phys. 6. C. "Table of Total BetaDisintegration Energies. DEUTSCH and 0. EVANS. L.
" Phys. 161 (1934). ALLEN. Wu.. TRIOG. T. ponent Theory of the Neutrino. D." Phys. I. see gen.1045 (1954). R. "The Radioactive Isotopes of Indium. Cambridge: Univensity Press. "The Experimental Clarification of f Nuc. Phys. Rw. Chapter 3. Soc. P L E A S O N ~ "Spectrometry. The Neutrino. M. LAWSON and J." Phys.390 BETADECAY [CW. S. S. cit. The Neutrino. Interactions. H. KONOPINSKI and L. 386 (1950). Rw. E. and J. 13. Siegbahn. H. Nuclear Stability Rules. 254 (1956). .. KONOPINSKI. 9 6 . B. M. ALLEN. 1396 (1955). Mod. J. Rev. 982 (1940). M. SNELL and F. 0." op. 1950. "Neutrino Recoil Experiments." Phys. ALLEN. ref. Phys." Phys. Mod. R."The Interpretation of Comparative HalfLives in the Fermi Theory of BetaDecay. LEE and C. P. Phys. "Theory of BetaDecay. 105. N. KOFOEDHANSEN. 5." Betaand GammaRay Spectroscopy. 1953. ed. RODEBACK and J. Rev. 17. RICHARDSON. 18." Revs. 381 (1951)." Revs. 967 (1952). FERMI. 25. "The Maximum Energy of the &Rays from Uranium X and other Bodies." Revs. Wu. 278 (1948). 22. "BetaDecay. JR. C. S. N. ref. Rev. R. ref.. Investigation of A3'. 81. 27. S. CORK.368 (1936). ALLEN. lium 7. Rw. (London) A139. A. FEINOOLD. YANG. Chapter 3. The Neutrino. YANG. 2." Revs. Mod. KOFOEDHANSEN. E. 86. 49.692 (1942). 115. 61. "Question of Parity Conservation in Weak 34. p. "Recent Investigation of the Shapes of j3Ray Spectra. 209 (1943). Mod. "Table of ftValues in BetaDecay. SARGENT. New York: Academic Press. Vol. T.. 19. ALLEN.of N. LEE and C. 100. 1 "Parity Nonconservation and a TwoCom35. Inc. 26."Versuch einer Theorie der flStrahlen. VQ~. G. DAVIS. Stanford: Annual the theory of BetaDecay. o Reviews. and PAXTON. Rev. J. Rev. 21. J. Recoils from Neutrino Emission in Argon37." Revs. Rw. S. W. W.. S." Progress in Nuclear Physies. B. 57. 23. 14." Phys. 14 12. see gen." Phys. "The Neutrino. 86. 88." Ann. A. 33. J. Roy. FEATHER. 1952.. Chapters 4. S. N. 0. Chapter 2." Proc." Z. K. 1 5 . J. 399 (1950). 10 (1951) 22." Phys. T.. 28. "The Energy and Momentum Relations in BetaDecay and the search for the Neutrino." Phys. FEENBERO and G. Rw. E. SMITH of Be7. 246 (1955). 1671 (1957). "Nuclear Recoil Following Neutrino Emission from Beryl32.L'Neutrin~ Recoils following the Capture of Orbital Electrons in A37. 2 2 . 659 (1933). 20. Rev. 1. S. p. E. ref. 446 (1952). The Upper Limit and Shapes of the &Ray Spectra from Several Elements. "Nuclear Recoils h u l t i n g from the Decay 31. C. 97. J. see gen. 0 4 . "Neutrino Recoil Spectrometer. 29. 20. 30. Sci. W. SKYRME. 15. CRANE. Physik. see also. 23. LINGER. A. 24.. gen. 16. Mod. Phys. 357." Phys. ALLEN. ref. D. H. "The Radiations Emitted from Artificially Produced Radioactive Substances. Phys. "Experimental Evidence for the Existence of a Neutrino. KURIE. see gen.
e)A3' Reaction Induced by Reactor Antineutrinos." Prog. DIRAC.HALBAN." Phil. COWAN. ed. 164 (1956). (7) 44. M.. DAVIS." unpublished. GRACE. Wu. T. S E O R ~ 48. "An attempt to Observe the C13' 41. COWAN. Phys. 565 (1958). Rev. 1958. 46. 117. (b) Radioisotopes in Scientific Research. N. P. "Table of Properties of the 'Elementary' Particles. Conf. F. G.. 31. Oxford: Clarendon Press." Science 127. Sci. 43. Frisch. 52. Paris. C." Revs. 1957). AMBLER. "Weak Interactions and Nonconservation of Parity. Proceedings of the Second International Conference on the Peaceful Uses of Atomic Energy. "Antinucleons. Polarization of CosO. 521 (1949). C. of Nuc. Vol." Science 127. Rev. p. P. S. Part 1. Mod. (v*. Rev. 18." Proc. Chapter 12. A. JR. S. 105. London: Pergamon Press. and HUDSON. 49. p. The Free Antineutrino Absorption Cross Section.HARRISON. Vol. and D."Invariance in Physical Theory. WICK. Sci. Free Neutrino. R. Am. BLINSTOYLE. DAVIS. Nuc. DURAND. Vol. 30. 42. REINES. E. 0. 569 (1958). "Experimental Test of Parity Conservation in Beta Decay. "Nuclear 51. "Invariance Principles of Nuclear Physics. 28. HAYWARD.The Principles of Quantum Mechanics. London: Pergamon Press. of Atomic Spectra." . M. C. REINES. Mod." (a) Phys." Phys." Revs. L. 93. 766 (1955). 47. 8. 783 (1959)."Law of Parity Conservation and other Symmetry Laws. HOPPES. Mag. 4th ed." Science 124. Rev. 103 (1956). 159 (1960). and HALBAN. "Detection of the 38. E. Geneva: United Nations. HARRISON. YANO. Chapters 17. WIONER. 37. AMBLER. Rev. P. 1 (Proc. p. of Nuc. "Parity Experiments in Beta Decays.REFERENCES 391 36. New York: Academic Press. KURTI. Ann. 97. First UNESCO Int. Group Theory and its Application to the Quantum Mechanics 45. a Confirmation. 1953. A. 1413 (1957). 8.JR.and KRUSE. "Oriented Nuclear Systems. 50."An Attempt to Observe the Capture of Reactor Neutrinos in CP7. SHAPIRO. 728. LEE. 1 (1958). Free Antineutrino. HARMER. 253." Ann. . Wu. D. COWAN. 63. Phil. R. Phys. REINESand C. 1958. Soc. KRUSE." Phys. MCGUIRE. 216 (1953). Phys."Detection of the 39. 40. 127 (1958). 1958. and JOHNSON. 44. Meaaurement of the Cross Section by Protons. E. R.and MCGUIRE. WIONER. GRACE. 1959. 3.
. Show that the magnetic rigidity for electrons may be written in the form where A = 3335. by inspection. 5 0.35 0.32 Find the range by inspection.989 X Me~~/(gausscm)~. mg/cm2 Activity.160. The absorption of the @raysfrom P32 was measured with aluminum and the following results were obtained. the range of the @particles. An absorption curve of a sample emitting @particles was taken with aluminum as the absorber and the following data were obtained.33 0. (147).00 Find. mg/cm2 Relative intensity 0 100 200 250 300 350 400 450 500 loo0 600 375 250 165 110 65 37 18 8 3 . [Hint: U s e the result of Problem 12 of Chapter 6. and T is the kinetic energy of the electron. mg/cm2 Relative intensity Absorber thickness. counts/min Absorber thickneae. mg/cm2 Activity.511 Mev.75 0.(b) Feather's empirical formula R = 543T . counts/min 0. and b = 8. Absorber thickness.50 0. 1 Absorber thickness.1 2 .40 0. [ H i d : Use the result of Problem 11 of Chapter 6 .375 10. 5 1 .00 13.8 gausscm/Mev.375 16000 6100 2100 520 170 4.875 7. What is the maximum energy of the kparticles? What is the halfthickness of A1 for these particles? 4. Show that the kinetic energy may be writFn in the form where a = 0. 1 3.1. then find the maximum energy from (a) Eq.where R is in mg/cm2 and T is in Mev.
94813. 63. In a spectrometer study of the @spectrumof P32.n)F17 (Q = .645 Mev. relative units 91 82 68 49 31 12 3 500 lo00 1500 2000 2500 3000 3500 2 7 42 58 76 91 94 5500 6000 6500 7000 The intensity is proportional to the number of emitted electrons per unit H. Calculate the unknown mass.1. the following data were obtained. the endpoint energy of the transition is given together with the mass of the initial or final atom. and Fi7@)01' with endpoint energy of 1. gausscm 4000 4500 5000 Intensity.a)Si2Q.94932. The reactions P31(d.2. Find the difference in mass (in Mev) between a neutron and a hydrogen atom from (a) the masses. respectively.364 Mev.PROBLEMS 393 5. (d) a Kurie plot of the energy distribution. respectively. The atomic masses of NiM. and Si30(d.94994. In the following examplea of @decay.p) Si31 have the Qvalues 8. Hr . CuM.748 Mev. In (c) and (d). assume that the Coulomb factor is constant.p)Si30. 1 6 and F18(@)018with an endpoint energy of 0. 63.n)F18(Q = . Mass of Transition Mass of " 7. (b) the endpoint of the @spectrum of the neutron. ZnM. and GaM are 63.447 Mev). (c) the reactions 01s(p. 8.624 Mev). 9. (a) Which of these nuclides are stable? Which are radioactive? (b) For those nuclides which can decay by .158. and 63. (b) the energy distribution of the electrons.95710 amu.917 Mev). (d) the reactions 0 (d. Plot (a) the momentum distribution of the electrons. relative units 14 Hr . O16(d.p)O17 (Q = 1. What is the extrapolated endpoint energy in each of the graphs? 6.388. Calculate the energy available for the @decayof Si31. (c) a Kurie plot of the momentum distribution. and 4. 8. gausscm Intensity. Si29(d.r interval. aa given in Table 111.
In the nuclear fission of U235. 14 &emission. 0.27. Hollander. 1.) 16. How much power is carried away by neutrinos from a nuclear power plant which liberates heat a t the rate of 100.40 4.000 kw? 14. and 0. respectively. and 7rays with energies 0. respectively.24 1.172. with energies 3.72. get the maximum energy of the particles emitted by (halflife of 7. Devise a decay scheme to fit these data.98 1. a t what rate in watts does the sample emit energy? 13.643.57 cm3 of tritium a t N.18 2. A sample of hydrogen containing 2.86 4. The result is a Sargent curve for allowed. Compare your scheme with that given in the "Table of Isotopes" of Strominger.99 3.94 Nuclide 0. 0.0697.3 sec) and C133 (halflife of 2. favored Pdecay.9 4. 0.34 Mev. 7rays are found.97.8 Endpoint Halflife. 1.83.0 days and the average energy of the &particles emitted is 0. when the energy distribution is given by a weak @emitter (To < the Fermi formula of Eq.13 Mev.2 1.41 Mev.561.72 2. Plot log X against log To for the positron emitters listed below.65. . Halflife. 0.T. and 0. ref. sec 11.36 Mev. respectively.76. (b) the average energy of the &particles emitted.40. (c) the ratio of average energy to maximum energy.545. with endpoint energies of 0. was found to produce 0. and Sesborg (gen.1032. and 0. 17.874 Mev. Find (a) the disintegration rate.P.0 sec).00 mg.85.68 1.87 Endpoint energy1 Mev 2. Devise a decay scheme to fit these data. A sample of RaE contains 4. From the curve.615 Mev. and 0. Show that the ratio of the average energy to the maximum energy of < moc2) is +.48 3.200. (Note that the 7ray energies are known with higher precision than the /liray energies. 0.9 4. 7rays with energies of 0. and 2. 1. respectively.1909 cal/hr of heat. 12.4 22. The halflife of tritium is 12. The nuclide As76 eplita four groups of &rays with endpoint energies of 2.46 years. The nuclide SmlS3 emits four groups of brays. 11.05 Mev.).41. are also emitted. (1414). 0. The nuclide TcW decays by orbital electron capture and B+emission. Construct a decay scheme to fit these data.73.6 3. sec 1200 606 122 70 18. about 9% of the total energy liberated comes from the @decayof fission products. 1. as well as positrons with an endpoint energy of 2.86 0.6 3.52 Si2' Sc41 15. 2. If the halflife is 5. what are the endpoint energies of the spectra and the @values of the decay reactions? 10. 2.394 BETADECAY [CEAP.
the change in intensity of the beam as it passes through the slab is proportional to the thickness and to the incident intensity where the proportionality constant p is called the absorption coejhent.CHAPTER 15 GAMMARAYS AND GAMMADECAY The importance of ?rays as a source of information about nuclear energy levels has been shown in Chapters 13 and 14 in connection with adecay and @decay. however. the emission of 7rays was regarded simply as a means by which a nucleus can pass from an excited state to a less excited one. and then the subject of 7decay will be taken up. and the integration of Eq. Since ?rays are electromagnetic radiations and have no electric charge. 151 The absorption of gammarays by matter :experimental data The basic property of the absorption of 7rays is the exponential decrease in the intensity of radiation as a homogeneous beam of w a y s passes through a thin slab of matter. Consequently. If the 7ray photons all have the same energy. it will be necessary to examine first the interactions between 7rays and matter. (151) yields Equation (152) gives the intensity of radiation I after a beam of initial 395 . a subject of great theoretical and practical importance and will therefore be discussed in greater detail in this chapter. The study of 7decay depends on the ability to measure the energy of ?rays with high precision. Because of these properties of 7rays. In those chapters. they cannot be deflected by magnetic or electric fields. direct measurements of their energies with a magnetic spectrometer are not possible. but the problem of measuring these energies is different from that presented by charged particles. The methods used for measuring ?ray energies will then be treated because they are based on these interactions. Gammadecay is in itself. p is independent of x. as is indicated by the very much greater penetrating power of %rays. When a beam of 7rays o f intensity I is incident on a slab of thickness As. The mechanism of the absorption of %rays by matter is also different from that of charged particles.
396
GAMMARAYS AND GAMMADECAY
[CHAP.
15
FIG. 151. Schematic plan view of the apparatus used for the measurement of total Yray absorption coefficients (Davisson and Evans(l1).
intensity I,, has traversed a thickness z of a given material. The intensity may be written I = Bhv, (153) where B is the number of photons crossing unit area in unit time, and hv is the energy per photon; B is often called the fEw,defined as the number of photons per square centimeter per second, and I is the corresponding energy flux. Equation (152) may then be written
.
The equations which describe the absorption of 7rays are the same as those for xrays as discussed in Section 41. This result is not surprising because both kinds of rays are electromagnetic radiations. In fact, it is not possible t o differentiate between ?rays and xrays on the grounds of differences in their properties, since no differences have been found. The terms Yrays and xrays are used now chiefly to distinguish between the sources of the rays; Yrays come from nuclei, while xrays are the high energy radiations resulting from jumps of extranuclear electrons in atoms, or produced by artificial sources such as a Coolidge tube or a betatron. The discussion of absorption in this section applies to both Yrays and xrays. I t is necessary to define precisely the conditions under which Equations (15l), (152), and (151) are valid. These conditions are (1) the Yrays
1b l ]
THE ABSORPTION OF GAMMARAYS B Y MATI'ER

2
0
1
2
3
4
5
0
7
8
Thickness of lead (cm)

FIG. 152. Experimental verification of the narrowbeam attenuation formula for 7rays, and measurement of the absorption coefficient in lead. Source, 1.14 Rlev Yrays from Zna5.
are monoenergetic, i.e., the beam is homogeneous; (2) the beam is collimated and of small solid angle; (3) the absorber is "thin." An experimental arrangement for studying the decrease in intensity, or attenuation, of a 7ray beam under these conditions is shown in Fig. 151. This arrangement provides a narrow collimated beam and is spoken of as "good geometry." The thick lead shielding blocks, separated by a narrow space, act a s the collimator. The counter used to detect the 7rays is surrounded by a lead shield to absorb extraneous 7rays and electrons which might otherwise be scattered into i t and counted. I n a "good geometry" experiment, photons are removed from the beam either by absorption or by scattering. A photon which undergoes a scattering collision with an atom in the absorber slab may be deflected enough so that it does not reach the detector and is lost from the beam just as though it had actually been absorbed. The decrease in the intensity of the beam in passing through the slab is a measure of the combined effects of outright absorption and deflection. The term absorption is often used as synonymous with attcnuation, but this practice causes less confusion than might be expected bccnuse the meaning is usually clear from the context. The validity of Eqs. (152) and (153) can be shown by a "good geometry" experiment. The results of such a n experiment are given in Fig. 152; the absorber is lead, and the source is ZnG5,which emits 7rays with a n energy of 1.14 Mev. When the fractional transmission is plotted on a logarithmic scale against the absorber thickness on s linear scale, the result is n straight line in agreement with the equations. The slope of the line gives the value p (cm') = 0.7068 f 0.0051. Many experimental detcrminations of absorption coefficients have been made in this way and it has been found that the value of the absorption coefficient depends on the nature of the absorber and on the initial energy of the ?rays. Some
398
GAMMARAYS A N D GAMMADECAY
[CHAP.
15
values obtained by Davisson and Evans"' are listed in the third column of Table 151. I t is evident that for a given element, the absorption coefficient decreases as the energy of the 7rays increases. The value of p also varies from element to element and is, in general, greater for heavy elements than for light elements. Theoretical values are listed for future reference. The thickness x may be expressed in cm or in gm/cm2, as was done in Chapter 14 in the discussion of the absorption of &particles. Since the product px must be dimensionless, p may be correspondingly expressed in cm' or cm2/gm. It is sometimes convenient to express the absorber thickness in atomq/cm2 or electrons/cm2 rather than in gm/cm2; the absorption coefficient then has the units cm2/atom or cm2/electron. The absorption coefficient is usually denoted by p when the units are cm', by p/p for cm2/gm, by ,p for cm2/electron, and by .p for cm2/atom. The subscripts e and a are placed a t the lower left in order to avoid confusion later on. In terms of ,p and .p, the other coefficients are
where Z is the atomic number, A is the atomic weight, N is Avogadro's number, and p is the density in gm/cm3. In Eq. (157), the product pN/A .Z is the number of electrons/cm3 of absorber and, since ,p has the units cm2/electron, p comes out in cm'. The coefficient .p has the units cm2/atom, and when i t is multiplied by pN/A, the number of atoms/ cm3, the result is again p in cm'. The coefficients .p and .p are often called the "cross section per electronn and the "cross section per atom," respectively, because of their units. The attenuation of a 7ray beam may also be expressed in terms of a quantity called the halfthickness, i.e., the thickness of absorber needed to reduce the intensity to half its initial value. Equation (152) may be written in terms of the common logarithm log ( I / I o ) =  0 . 4 3 4 3 ~ . When Illo=
3,
1511
THE ABSORPTION OF GAMMARAYS BY M A ~ E R
399
Source
Absorber
Experimental absorption coefficient, b: cmI
0.1823 0.5782 0.4683 1.210 0.9368
& 0.0003
Theoretical absorption coefficient,
0.1820 0.5718 0.4628 1.228 0.9256 0.1559 0.4914 0.3858 0.9536 0.7057 0.1001 0.3273 0.2692 0.6467 0.4644
b: cm1
Mn54
(0.835 Mev)
f 0.0013 f 0.0014 & 0.004 f 0.0041
Zne5
(1.14 Mev)
NaZ4 (2.76 Mev)
where zllz is the halfthickness. In the example of Fig. 152, the halfthickness of lead for 1.14 Mev w a y s is 0.98 cm. A useful rule of thumb may be deduced from the absorption coefficients. In Eqs. (156) and (157), the ratio Z/A changes very slowly as Z increases and, as will be shown later,',p is approximately the same for all elements in a certain energy region. Hence, the absorption coefficient shows the smallest variation from element to element when expressed as ,p or p/p. From Eq. (159),
and, since p/p varies slowly with Z, the product (zllz)p also varies slowly from element to element. It follows from this result that the greater the density the smaller the thickness needed of a given material to decrease the ?ray intensity to a specified extent. For this reason, heavy metals such as iron, and especially lead, are used for shielding against 7rays and xrays. The approximate constancy of p/p and (xl12)pmeans that the weights of different materials needed to decrease the intensity of the radiation by a certain fraction are very nearly the same. But for sub
400
GAMMARAYS AND GAMMADECAY
[CHAP.
15
stances of higher density, the volume, and hence the thickness, will be less than for materials of lower density.
152 The interaction of gammarays with matter. The interaction of 7rays with matter is markedly different from that of charged particles such as a or @particles. The difference is apparent in the much greater penetrating power of ?'rays and in the absorption laws. Gammarays and xrays, which are both elect,romagnetic radiations, show a characteristic exponential absorption in matter, and have no definite range such as is found for charged particles. Charged particles, especially heavy ones, lose their energy during the course of a large number of collisions with atomic electrons. The energy loss occurs in many small steps and the particle gradually slows down until it is stopped altogether and absorbed. When a beam of ?'ray photons is incident on a thin absorber, however, each photon that is removed from the beam is removed individually in a single event. The event may be an actual absorption process, in which case the photon disappears, or the photon may be scattered out of the beam. The one shot nature of the removal process is responsible for the exponential absorption. For, the number of photons that can be removed in passing through a thickness Ax of absorber is proportional to Ax and to the number of photons reaching Ax; this kind of dependence leads directly to the exponential absorption law, as shown by Eqs. (151) and (152). Three processes are mainly responsible for the absorption of 7rays. These are (1) photoelectric absorption, (2) Compton scattering by the electrons in the atoms, and (3) production of electronpositron pairs as a result of the interaction between 7rays and the electric fields of atomic nuclei. Quantum mechanics makes it possible to derive formulas for the probability of each process and the probability can be expressed as an absorption coefficient or as a cross section. The total absorption coefficient, which appears in Eqs. ( 1 5 1 ) and (152), is the sum of the absorp tion coefficients for the three different processes. The problem is further complicated by the fact that the absorption coefficient depends on the energy of the incident 7rays as well as on the nature of the absorbing material. Consequently, the absorption of 7rays cannot be described by a single formula, or by a rangeenergy curve. Each partial cross section or absorption coefficient must be evaluated as a function of energy for a given material, and tables or sets of curves can then be prepared from which values of the total absorption coefficient of a given material can be obtained. Thus, I@) = 7(E) K(E), (151 1)
+
+
where 7, u, and K denote the photoelectric, Compton, and pair formation coefficients, respectively, and each absorption coefficient is related to the
1521
I X T E R A ~ I O N OF GAMMARAYS WITH MATTER
401
corresponding cross section by Eq. (157). In the following sections, formulas for the cross sections for the three processes will be written down, discussed, and illustrated by means of curves and tables. Before the detailed treatment is begun, the processes will be discussed in a general way. In the photoelectric process, a l l the energy hv of the incident photon is transferred to a bound electron which is ejected from the atom with a kinetic energy T = hv  I, where I is the ionization potential of the electron. The electron may be ejected from the absorber or, if the absorber is not too thin, is more likely to be reabsorbed almost immediately because of the short range of an electron in a solid. At low photon energies, below 50 kev for aluminum and 500 kev for lead, the photoelectric effect gives the chief contribution to the absorption coefficient. As the energy of the radiation increases, Compton scattering replacgs the photoelectric effect as the chief means of removing photons from the initial beam. It was seen in Section 67 that in Compton scattering the incident photon is scattered by one of the atomic electrons and the latter is separated from its atom. The photon moves off a t an angle with its original direction and with less energy than it had initially. The change of direction serves to remove the photon from the incident ?ray beam. Compton scattering gives the main contribution to the absorption coefficient between 50 kev (0.050 Mev) and 15 Mev for aluminum, and between 0.5 and 5 Mev for lead. In these energy ranges, the energy of the incident Yray is much greater than the binding energy of the atomic electrons. The process may therefore be regarded as the scattering of a photon by a Free electron initially a t rest and, since each electron in the atom scatters "ldependently, the Compton absorption coefficient per atom is proportional the atomic number Z . 4 t high enough energies, both the photoelectric absorption and the npton scattering become unimportant compared with pair formation. ,he latter process, in the Coulomb field of an atomic nucleus, a 7ray I sufficient energy disappears and an electron and positron are created. : total energy of the pair is equal to the energy hv of the incident Yray; kinetic energy T of the pair is
s,
T = hv  2moc2,
(1512)

d the small recoil energy of the nucleus is neglected. For pair production to occur, hv must be greater than 2moc2 or 1.02 Mev. Pair production cannot occur for hv < 2moc2because this amount of energy is needed to supply the rest energy of the two particles. A pair of particles, with opposite charge, must be formed if charge is to be conserved. At photon energies greater than 5 Mev for lead and 15 Mev for aluminum, the probability of pair production is greater than that for Compton scattering and continues to increase with increasing energy.
402
GAMMARAYS AND GAMMADECAY
[CHAP.
15
In addition to the three effects discussed, there are some minor effects which may contribute to the attenuation of a %ray beam. The most important of these is coherent scattering by whole atoms or molecules, which may add several percent to the absorption coefficients of highZ materials a t low 7ray energies. Among the other effects are (1) the nuclear photoelectric effect, in which highenergy photons eject neutrons from the nuclei of highZ materials, and (2) Thomson and Compton scattering by nuclei rather than by electrons. For most practical purposes, all of these effects may be neglected. The results of the theory of the interaction between w a y s and matter have been collected and compared with experiment in the excellent review arti~le'~ by ' C. M. Davisson and R. D. Evans. Many tables and curves rye given there which present much of the presently available information in useful form.
153 Photoelectric absorption. Formulas for the probability that a photon of energy hv will undergo photoelectric absorption have been derived by quantum mechanical methods. Several different formulas must be used if the probability, expressed as an absorption coefficient, is to be calculated over the range of photon energies from 0.1 Mev to 5 or 10 Mev. The reason for this is that the theory is complicated and different assumptions are made in different energy ranges in order t o ease the mathematical difficulties. The main features of the dependence of the absorption coefficient on the atomic number of the absorber and on the energy of the photon are shown by the simplest of the formulas. I f the energy of the photon is sufficiently small so that relativistic effects are not important but large enough so that the binding energy of the electrons in the Kshe may be neglected, the cross section per atom , T for photoelectric absorpti is

where
I n Eq. (1513), hv is the energy of the incident photon, moc2 is the res; energy of the electron, and Z is the atomic number of the absorbing material; is a convenient unit for measuring the cross section and represents the cross section for the scattering of low energy photons by a free electron a t rest. This kind of scattering, called Thom.wn scattering, was discussed in Section 42. The simple equation (1513) applies only t o the ejection of electrons from the Kshell of the atom, which accounts for about 80% of the photoelectric effect.
+,,
1531
PHOTOELECTRIC ABSORPTION
403

.
"
The most important property of T . is the strong dependence on the atomic number and on the energy of the incident photon; ,T is proportional to Z5 and inversely proportional to (hv)'I2. The Z5 dependence means that, for a given photon energy, the process of photoelectric absorption is much more important in heavy metals such as lead than in light ones such as aluminum. The energy dependence shows that for a given element, the effect is much greater a t small photon energies than a t large ones. I t follows that the photoelectric effect is especially important in the absorption of low energy photons by heavy elements. In the more rigorous formulas, the dependence of , T on Z and hv is not T quite the same as that shown in Eq. (1513). Thus, the variation of , with Z does not follow the simple fifth power law, although for practical purposes the deviation from a power law is small. The exponent giving the best agreement with theory increases with 7ray energy and is between 4 and 5 for energies above 0.35 Mev, being approximately 4.5 a t 1.13 Mev and 4.6 a t 2.62 Mev. The rigorous dependence of , T on the energy of the Yrays also cannot be described by a simple power law. The cross section for ~ decreases rapidly with increasing energy, approximately as ( h ~ ) energies less than 0.5 Mev, and as (hv)' for energies greater than 0.5 Mev. Values obtained from the more rigorous theory are given in Table 152. For purposes of comparison and ease of use, the values of the quantity
Energy of Yrays z = moc2/hvl Mev
*Numerical values of .r/Z5n in units of
cm2.
404
GAMMARAYS AND GAMMADECAY
[CHAP.
15
FIG. 153. The absorption coefficient per centimeter as a function of the photon energy for photoelectric absorption in aluminum, iron, tin, and lead.
1541

COMPTON SCATTERING
405
"
,s/Z5n are listed, where n is the ratio moc2/hv. Increasing values of n correspond to decreasing w a y energies. For a particular absorber, say lead, the values of ,r are obtained for different %ray energies by multiplying the values listed in the table first by Z5 = (82)', and then by the appropriate value of n. Curves obtained in this way are shown in Fig. 153 for aluminum (Z = 13), iron (Z = 26), tin (Z = 50), and lead (Z = 82). I t will be noted that the values of .7/Z5n listed in the table cm2. Values of ,r for energies and atomic numbers are in units of other than those listed can be obtained by plotting the appropriate curves and interpolating. The values for Z = 0 are included to allow interpolation for elements with Z < 13.
154 Compton scattering. The main features of Compton scattering were discussed in Section 67, but it will be useful to repeat some of the results obtained there in somewhat different form. The energy lost by a photon in a single Compton scattering process can be obtained from Eq. (652), which gives the change in wavelength of the photon,
In Eq. (1515), Xo is the wavelength of the photon before collision, A is the wavelength after the collision, and 4 is the angle between the initial and final directions o f the photon as shown in Fig. 67. Since A = c/v, Eq. (1515) may be written
 Inversion of both sides of the last equation gives
V
=
1
1 vo
+
h (1  cos 4) m0c2
If the top and bottom of the right side are multiplied by vo, and if both sides o f the resulting equation are multiplied by h, the result .is hv = 1 hvo + moc 7 (1  cos 4) hvo
406
GAMMARAYS AND GAMMADECAY
[CHAP.
15
Equation (1516) gives the energy of the scattered photon in terms of the initial energy and the scattering angle. The recoiling electron has a kinetic energy given by

hvo (1  cos 4) m0c2
The kinetic energy of the electron has its maximum value when cos $I = 1 or 4 = 180°, and the photon is scattered directly backward. The electron energy, in this case, is
The electron receives the least energy in a grazing collision in which the photon continues with its initial frequency in the forward direction, and the electron is ejected with very nearly zero velocity in a direction perpendicular to that of the photon path. The discussion so far refers only to a single Compton scattering process. In order to treat the contribution of the Compton effect to the attenuation of a beam of 7rays in matter, it is necessary to calculate the probability that such a scattering process will occur. This probability was calculated on the basis of relativistic quantum mechanics by Klein and Nishina. Although the details of the theory are complicated, the results may be expressed in straightforward formulas with which calculations can be made quite easily. A formula was obtained for the cross section per electron p for the removal of photons from the incident beam by scattering:
(1519) where a has been defined as
and +o is given by Eq. (1514). When p is multiplied by pN(Z/A), the result is the Compton absorption coefficient a (em'),
1541
COMPTON SCATTERING
407
The coefficienta is a measure of the probability that a photon is scattered out of the beam per centimeter of absorber, and since the beam is initially homogeneous, u is also a measure of the total amount of energy removed from the beam per centimeter of absorber. I t is sometimes necessary to know the amount of energy carried off by the scattered photons or the amount of energy absorbed by the recoil electrons. These amounts may be expressed in terms of two quantities denoted by ,a, and ,a,. The former, ,a,, may be called the "Compton scattering cross section per electron for the energy of the scattered photon." I t is given by
The quantity ,aa is the "Compton cross s~ction per electron for the energy absorbed." The coefficient ,a is simply the sum of ,a, and ,a,:
SO
that when .a and ,a, are known, ,u, is obtained by subtraction. In analogy with Eq. (1521),
Finally, the sum
=
6s
+ aa,
shows how the total Compton coefficient is broken down into a coefficient for the energy scattered out of the beam and a coefficient for the energy removed by the recoil electrons. The electrons, because of their very short range, are often absorbed and their energy may appear in the form of heat. In some practical problems it is necessary to know how much heat is produced as a result of the attenuation of a beam of 7rays in passing through a given absorber. The contribution of Compton scattering to this heat can be obtained when ,a, is known for the particular Yray energy. It is convenient to tabulate the values of ,a, g,,and ,aa,i.e., the values in cm2/electron, because these are independent of the properties of the absorber. The values for a particular absorber are then obtained by using Eqs. (1521) and (1524). Table 153 gives the cross sections for the Compton effect g , ,a,, and ,aa in units of cm2/electron. The values of the absorption coefficients per centimeter, a,, a,, and a , for the special case of lead ( Z = 82) are plotted in Fig. 154.
GAMMARAYS A N D GAMMADECAY
[CHAP.
15
COMPTON SCATTERING
Initial en gY
a = hv/moc2
Mev
0.0128 0.0256 0.0383 0.0511 0.0767 0.102 0.128 0.153 0.204 0.255 0.307 0.358 0.409 0.460 0.511 0.767 1.022 1.278 1.533 1.789 2.044 2.300 2.555 3.066 3.577 4.088 4.599 5.108 6.132 10.22 15.33 25.55 35.77 51.10
*In units of
cm2/electron.
410
GAMMARAYS AND GAMMADECAY
[CHAP.
15
Equations (1519) and (1522) show that the Compton scattering per electron is independent of Z,so that the scattering per atom is proportional to Z. The mass scattering coefficient u/p is given by
and varies only slowly with Z. For the light elements Z / A is closely equal to 3, so that for a given photon energy u/p is practically constant for these elements. The total scattering coefficient per electron, 8,decreases with increasing photon energy, as can be seen from the values in Table 153. The decrease is quite slow a t low values of the energy, and for energies above 0.5 Mev, ,a is roughly proportional to (hv)'. Thus Compton scattering decreases much more slowly with increasing energy than does photoelectric absorption; even in heavy elements, it is the most important process in the energy range from about 0.6 to 2.5 Mev. In the treatment so far, it has been assumed that a singlyscattered photon was removed from the 7ray beam. This assumption may be considered to define a thin absorber. In a thick absorber some of the singlyscattered photons which have been considered lost will be scattered and rescattered and may eventually reach the detector. The problem of repeated, or multiple, Compton scattering of 7rays is important in connection with the shielding of personnel against the 7radiation emitted by accelerators or nuclear reactors. Considerable progress has been made in the theoretical and experimental treatment of this problem, and the reader is referred to the book by H. Goldstein listed in the general references a t the end of the chapter.
155 Electronpositron pair formation. The third mechanism by which electromagnetic radiation can be absorbed is the production of electronpositron pairs. This process, which has no analogy in classical physics, may be accepted as a strictly experimental phenomenon. The discovery of the positron and of pair formation, however, represents one of the great triumphs of modern physical theory, and for this reason a short digression seems worth while. I t was mentioned very briefly a t the end of Section 75 that the Schroedinger equation, the fundamental equation of wave mechanics, did not satisfy the requirement of invariance postulated by the thoery of special relativity. When the Bohr theory of the atom was replaced by wave mechanics, the problem of the relativistic treatment of the electron still remained. This problem was solved by Dirac, who derived an equation, the Dirac electron equation, which is one of the most important developments in modern physics. Dirac required that the equation which represents the motion of an electron be invariant under the Lorentz transformation (see Sections 6 4 and 65).
.
.
=
.
or "Dirac hole. but that they are all usually occupied. I n other words. Possible values of the energy of an electron according to the Dirac theory. while electrons in states of negative energy should have properties which have no classical analogy. the value of the energy of an electron could change only in a continuous way. however. According to quantum theory.155. This empty place. The empty state. since a particle of negative energy and negative charge is absent. But to solve the Dirac equation i t was necessary t o assume that the electron can exist in two sets of quantum states. an electron can make a discontinuous transition from one energy state to another. according t o Dirac. where the shaded regions are those in which values of the energy exist. would appear as a particle with positive energy and positive charge. 155. Electrons in states of positive energy behave in the usual manner of electrons that are ordinarily observed. Dirac first assumed  . in that case. so that there is no way of ruling out a jump from a positive energy state to a negative energy state. The solution of the Dirac equation led in a natural way to the formula for the fine structure of spectral lines and to the spin of the electron. and a n electron in a positive energy state could not bridge the discontinuous gap between a state of energy +moc2 and one of energy moc2. The electrons which are ordinarily observed are those in positive energy states. Suppose that one electron is missing in the distribution of negative energy states. a state of negative energy would have no real physical meaning. and the other of negative energy.ELECTRONPOSITRON PAIR FORMATION $ 1 t L~ e ~ a t i electron ve  FIG." would therefore behave like a positively charged particle. one of positive energy (including the rest energy). This state of affairs is shown in Fig. and that no possible energies for the electron exist between these two limits. For. These electrons could simply be neglected if they were considered from a classical viewpoint. I t was found that the possible values of the energy of a free electron are either greater than +moc2 or smaller than moc2. Dirac avoided the difficulty by assuming that the states of negative energy are real.
New York: Wiley.412 101 GAMMARAYS AND GAMIILADECAY J 1 2 5 10 20 Photon energy (hv/&) 50 100 Fig. Vol.K in units of 6.K/Z2. 1953. d 6 1 . I. . Pairformation cross section per atom as a function of the photon energy.. Segr*. from Ezperinzental Nuclear Phyeica. (By permission. in units of cm2/atom .) Photon energy. 156. E.. ed.
because some body must be given recoil energy and momentum in order that energy and momentum be conserved in the system. Thus. for pair production can be listed for certain photon energies and curves for the variation of K . SO that. At these energies the electronpositron pair is probably formed some distance from the nucleus. for a given photon energy. at higher energies (up to 25 Mev) the values for lead are somewhat smaller than those for air or aluminum. at still greater energies. the production of a positron is interpreted as follows. The cross section is zero for photon energies less than 1. it is necessary to take into account an effect called screening. the appearance of an electron in a positive energy state means the appearance of an ordinary electron.K/c$ as a function o f a = hv/moc2 are shown in Fig. it increases at first slowly. . pair formation increases quite rapidly with atomic number. Pair production cross sections are usually expressed in units o f the quantity Curves of .02 Mev. A photon of energy greater than 2moc2can raise an electron from a state of negative energy to a state of positive energy. Consequently. then more rapidly. observed particles with the same mass as that o f the electron and with an electric charge equal in magnitude but opposite in sign to that of the electron.02 Mev. In terms o f Dirac's theory.1551 ELECTRONPOSITRON PAIR FORMATION 413  that it represented a proton. the screening effect becomes more marked. during the course of a study of cosmic rays by means of a cloud chamber.or 1. which means the appearance of a positron. K / Z ~ for A1 and Pb are listed at several photon energies in Table 154. values o f . . with energy can be given. 156 for air and lead. The disappearance of an electron from a negative energy state leaves a hole. the field in which the pain are created is smaller than that of the nucleus alone and the probability of pair formation is reduced somewhat. but this assumption had to be discarded because it could explain neither the difference in mass between a proton and an electron nor the stable existence of the proton. and will not be written down here. It is proportional to Z2. The dilemma was resolved in 1933 when Anderson.K. Values of the cross section per atom.the photon must have an energy of at least 2moc2. ~ / 4and . The formulas for the probability o f pair formation are more complicated than those for photoelectric absorption and Compton scattering. for greater energies. This particle was given the name positron and identified with the Dirac hole. When this distance lies outside some o f the electron shells. The creation process usually takes place in the electric field in the neighborhood of a nucleus. Since each particle must have a rest mass equal to moc2. At photon energies above about 10moc2or about 5 Mev. a pair of particles is created.
p or p can now be obtained from the values of the partial coefficients given in Tables 152. Values of . 2. Two photons. 153.511 Mev.c( is then The mass absorption cciefficient p/p is The absorption coefficient p is . A positron. and the absorption of 7rays by the pairproduction process is always complicated by the appearance of this lowenergy secondary radiation. and multiplied by 6 = 5.414 GAMMARAYS A N D GAMMADECAY [CHAP.K/+ are obtained from Table 154 or Fig. 156 The absorption of gammarays by matter. rather than one. 156 for the same values of n(= l/a) used in the first two steps. Values of the Compton cross section per atom . Values of . The two particles disappear.75 Mev.7/Z5n for different values of n are found from Table 152 or from plots of the data in the table. and two photons appear moving in opposite directions. that of electronpositron annihilation. pair formation predominates. The following procedure is used: 1. 3. The contribution of pair formation t o the total absorption coefficient C( for lead is equal to that of the Compton effect a t about 4. equal to the rest energy of an electron. The total absorption coefficient per atom . 15 It follows from the way in which the cross section for pair formation depends on Z and on the energy that this process is important for high energies and for heavy elements. Above this energy. The process of pair production is closely related to the reverse process. each with an energy of 0. and 154 and from graphs of t h w data. Values of the photoelectric cross section per atom a~ are obtained on multiplying by Z5n.796 X 1028Z2 to give the pair production cross section per atom . after being formed. The photons which appear on the annihilation of an electronpositron pair are called annihilation radiation.K. Theoretical values of the absorption coefficient . is slowed down by collisions with atoms until it is practically a t rest. Comparison of experimental and theoretical results. a are obtained from the data of Table 153 on multiplying the values of 8 by Z. are needed to allow momentum to be conserved if the annihilation occurs away from nuclei. I t then interacts with an electron which is also practically a t rest.
3313 1.869 4.25 25.633 3. and the general features of the absorption of ?rays can be seen by considering these curves. P/P.41 0. . 157.122 25.2 35.31 3.607 0.3405 75.T. attenuation by the Compton effect becomes relatively more important. but decreases rapidly with increasing energy.87 15.00 23.68 1.086 5.675 6.28 14.a. Their values for lead are listed in Table 155 together with the values of p/p and p.39 2.771 0.4086 47.469 0.1948 0.32 0.86 1.45 18. aluminum.7 0. .31 1.119 14.651 7.130 10. and lead).5 51.K. The values of the cross section per atom are in units of cm2/atom. Mass absorption coefficients p/p ( ~ m ' / ~ mare ) listed in Table 156 for four commonly used absorbers (water. iron.620 0.5 Mev.499 0.04 18.08 2.8 0.5108 27. cm' cm2/gm 59.247 3.7 0.4 6.043 2.065 0. . At energies slightly less than 1 Mev most of the attenuation is Photon mergy.47 21.9 33.24 1822 1023 500 192 103.1022 1782 0.19 14.p have been prepared by Davisson and Evans"' for 24 elements ranging from hydrogen (2 = 1) to bismuth (2 = 83).507 5.1277 985 0. At very low energies.316 0.6811 14.5 1.206 15.16 0.108 0. Mev Photoelectric a7 Pair iormation OK Total aP Mass Coefficient :oeficient per cm.560 9. until the two effects are equal at about 0.2 23. I ( .6 16. As it decreases.698 0. absorption by photoelectrons predominates.022 6.3 73.362 3. The total and partial absorption cross sections per centimeter for lead are plotted in Fig.2554 161 0. and .93 18.533 2.54 0.491 0.835 *In units o f cm2/atom.22 0.1561 THE ABSORPTION OF GAMMARAYS BY MA'ITER 415 Tables of values of .42 1.1703 465 0.
Compton scattering. Photon energy.#) FIG.GAMbfARAYS A N D GAMhfADECAY [CHAP. showing the contributions of photoelectric absorption. 15 Photon energy (hv/m. Total absorption coefficientof lead. Water Aluminum Iron Mev . 157. and pair formation.
05 Mev <0. tin. The total absorption coefficients of aluminum. caused by the Compton effect. The production of artificial radionuclides has provided sources of monoenergetic w a y s with energies in the range from .5 5 Mev Pair formation >15 Mev > 5 Mev Aluminum Lead The variation of the absorption coefficient us. and lead in Fig. Photoelectric effect <0. on the purity of the absorber. copper. the particular energy a t which one process or another is important varies from element t o element.0515 Mev 0. 158. and lead. The simplest method of testing the theory of the absorption of w a y s is to measure the total absorption coefficient by the method discussed in Section 151. Absorption by pair formation starts a t about 1 Mev and increases while the other effects decrease. While these features are characteristic of the absorption curves for all elements.1561 THE ABSORPTION OF GAMMARAYS BY MATTER 417 Photon energy (hv/+) FIQ. energy curve from element to element is shown by the curves for aluminum. the experimental results depend on the knowledge of the energy of the photons emitted by the source. Although meastuements of this kind are basically simple. copper.5 Mev Compton effect 0. The following table indicates the energy ranges in which the various effects make their greatest contributions in a light element (aluminum) and in a heavy element (lead). tin. The curve of total absorption coefficient us. and on many other experimental details. until a t high energies the absorption is almost completely by pair production. energy has a minimum at energies where the Compton effect and pair production become comparable. 158.
and hence less extensive. the most direct method is to determine the wavelength and hence the energy by using a crystal as a diffraction grating (Section 43). The agreement has been found for a large number of elements in the periodic table from carbon to lead. and monoenergetic 7rays beams of higher energy have been obtained as products of nuclear reactions. 159. and pair production. Thus.8 Mev. Several general methods can be used to measure the energies of 7rays. in a line a t 8. The coilnting rate of the detector is determined as a function of the source position as the source is moved around the focusing circle. the theory of photoelectric absorption. ' * . Compton effect. the KleinNishina formulas for the Compton effect. This method can give results with very high precision.0 m). are in good agreement with the experimentally measured values.. A highprecision instrument called a eumerlcrystalfoamingtype spectrometer has been used by D u M ~ n d . than those of the total absorption coefficient. Since 7rays are electromagnetic radiations. of the experimental results. The radius of curvature of the crystal is then equal to the diameter of the focusing circle F shown (2. 15 0. including many experimental curves. It has been found that in the energy range from 0. The partial absorption coefficients have also been measured experimentally. (4lo). The reader is again referred to the review article by Davisson and Evans for a detailed discussion. If the source of ?rays is a t the real focus R. the rate has a sharp maximum a t the point for which a strong diffracted beam occurs.418 GAMMARAYS AND GAMMADECAY [CHAP.g. is satisfied. when extended. The theory of these processes and the experimental verification are treated in detail in Heitler's book. although measurements of this kind are more difficult.5 to 2. The position of this point determines the wavelength and hence the frequency and energy of the %rays. For each different wavelength there is a particular position of the source somewhere on the focusing circle which yields a diffracted beam. Eq.1 Mev to 6 Mev the theoretical values of the absorption coefficient as calculated from the present theories of the photoelectric effect. A partial comparison of theory and experiment is given in Table 151. there is a diffracted beam which enters the detector D as if i t came from a virtual source a t V. the energy of . The quartz diffracting crystal C is bent and clamped so that the diffracting planes meet. e.'~' 157 The measurement of gammaray energies. and the theory of pair formation have all been shown to predict the correct results. and if the Bragg condition. ~ and ' the principle on which it is designed is shown schematically in Fig. The Quantum Themy of R~diation. The theory of the individual effects has been found to give cross sections or partial absorption coefficients in good agreement with the experimental results. the Bragg angle is denoted by 8. normal to the plane of the figure.
969 0. The two numbers quoted can be used to calculate the possible difference in mass between the positron and the electron. the measuremenLs become more difficult and less precise as the energy of the 7rays increases and the wavelength decreases. DuMond's ?ray spectrometer was also used to measure the energy of the annihilation radiati~n. First. Schematic diagram of a crystal ?ray spectrometer (DuM~nd'~. The "best va1ue"for the electron rest energy a t the time of the experiment was 510.015 kev.1571 THE MEASURMENT OF GAMMARAY ENERGIES 419 FIG. they require highly active sources. however. Second.~)). the %ray associated with the &decay of AulS8 was determined as 411. with the result that they cannot be carried much above 1 Mev.067 kev. have two disadvantages. and it was found that to a part in lo4 there is no evidence .770 f 0.036 kev. and these are often not available.'~) The photons from the annihilation of positrons from CuS4were found to have an energy of 510.941 f 0. and results with similar precision have been obtained  for the energies of many other nuclear Crystal diffraction methods. so that the measured value of the energy of the annihilation radiation agreed very well with the value predicted by the theory of pair formation and annihilation. 159.
described in Section 141. D i m of an arrangement for measuring the energies of %rays from a sourcc of 6. In the comparison e~periment. it has been included here because of the way in which it supplements the annihilation radiation experiment. .Absorber . together with a recent direct comparison of the ratio m/e for the electron and positron. a recent measurement'32' of the deuteron binding energy gave the result 2. The particles followed identical trajectories in opposite directions in the spectrometer.420 GAMMARAY8 AND GAbfMADECAY [CHAP. 1510.and 7rsye. Thus. T h i s result has special interest because. Although the comparison experiment is not directly related to 7rays.0015 Mev.2255 & 0. 15 for any differencein mass between positive and negative electrons. The magnitude of the electric field needed for focusing was taken as a measure of m/e for the particle involved. . and the denominator is the accepted value of the ratio for the electron. both the mass and the specific charge of the positron and electron have been shown to be equal to within very small amounts.Source of B and yrays Counter FIG. The result obtained was where the ratios with the minus and plus signs refer to the values obtained by focusing the particles in the experiment. The energies of the ejected electrons are measured with a magnetic spectrometer. it provides independent experimental proof that the electron and positron are identical except for the sign of the electric charge. As an example of the precision that can be obtained.'~' electrons from a hot filament and positrons from Na22were analyzed in a mass spectrometer with a reversible electric field and a fixed magnetic field. Radiator .'~~*~ The curved crystal spectrometer has been improved and ~ i m ~ l i f i e dl' and applied to problems such as the study of the 7rays from neutron capture. Gammarays of moderate energy are most often studied by means of the photoelectrons and Compton recoil electrons which they eject from a suitable material called a radMtor.
and for spectrometers of very high resolution even M electrons may be resolved. then A schematic diagram of an arrangement for measuring the energies of 7rays from a source which emits both 19particles and w a y s is shown in Fig. and . the photoelectrons ejected from it form line spectra. and hv is the desired ?ray energy [this amounts simply to dropping the subscript zero in Eq. is the maximum is related to the 7ray energy by Eq. If T electron energy. the source is enclosed in an absorber of material of low atomic number such as aluminum. are focused in the spectrometer and form a continuous spectrum with a fairly sharply defined upper limit. from the Hr value) and the ?ray energy is obtained by adding the biding energy of the electron in the shell from which a s ejected to the measured electron energy.1b 7 ] THE MEASUREMENT OF GAMMARAY ENERGIES When the Compton effect is used. The energy of the electrons in a given line can be determined from the position of the line (i. (a) Photoelectrons and Compton recoil electrons ejected from a lead radiator by %rays from Md2. (15la)]. The Compton electrons ejected from the radiator. The upper limit corresponds to electrons ejected in the forward direction from the surface of the absorber. Lines appear corresponding t o electrons from the Kshell and the Gshell.e. 1511. 1510. (b) Compton recoil electrons ejected from a copper radiator by %rays from Mn52 (Peacock and Deutsch@)). thick enough to stop all the primary electrons or other charged particles but not the 7rays. the peaks are caused by photoelectrons. it w Some of the &particle spectrometers discussed in Section 141 have been used to determine 7ray energies by measuring the energies of Compton  FIG. If the radiator is a thin foil of material of intermediate or high atomic number. (1518). which is usually a thin foil.. .
Figure 15ll(a) shows the photoelectrons and Compton recoil electrons ejected from a lead radiator by 7rays from Mn5'. while the Compton and photoelectric cross sections decrease.and 7rays. Deutsch.015 Mev.46 f 0. The lowest energy ?ray is doubtless annihilation radiation. and measured with that instrument. Schematic diagram of a pair spectrometer (Walker and Mc .422 GAMMARAYS AND GAMMADECAY [CHAP. 15 electrons and photoelectrbns. because the probability of pair formation increases with increasing photon energy. and ~vans"' have treated the theory.02 Mev. 1511. At higher energies. A collimated beam of ?rays falls on a radiator and 512. 0."' are shown in Fig. and use of a short magnetic lens spectrometer for both 8. There are four pronounced peaks caused by +rays of energies 0. I)). and 1.94 f 0. 7rays are best studied with a pair spectrometer. above 2 or 3 Mev. Figure 15ll(b) shows the spectrum of Compton recoil electrons produced in a copper radiator indicating the same four 7rays.510 f 0. design.734 f 0. produced by ^/rays from Mn5'.03 Mev. 0. Elliott.01 Mev. since Mn52is a positron emitter. The spectra of 8particles from two radiators.
I t is necessary to detect simultaneously the electron and positron produced by one %ray. and the unknown energy can then be found from the measured value of the absorption coefficient. The energies of the paired electron and positron are determined from their Hr values. Unfortunately. the absorption coefficient is known as a function of energy either from theory. Electronpositron pairs are produced which. For a given absorber. With ionization second. a measurement of the attenuation of a %ray beam by a known absorber in a "good geometry"experiment would give a value of the absorption coefficient. The electron and the positron may each fall into one of four counters placed in the plane of the radiator foil.1571 THE MEASUREMENT OF GAMMARAY EXERGIES 423  ejects electronpositron pairs from it. go very nearly in the forward direction. 157 and 158. smaller resolving times (lo' sec) can be achieved. with Geiger counters. that is. I t is possible. but from the radioactive atom itself. A minor complication is introduced by the fact that a given absorption coefficient may correspond to two different photon energies because of the minimum in the absorppon coefficientenergy curve. A parallel beam of 7rays falls on a thin radiator foil placed perpendicular to the beam. The electrons and positrons are focused separately in a uniform magnetic field into a number of counters. not from a radiator. Two events occurring within a time interval less than the resolving time and detected by separate counters are said to be in coincidence. The correct energy can be found by obtaining absorption curves in two different materials. with suitable electronic circuits. Energies of Yrays can also be obtained from measurements of the energies of internal conversion electrons. and the sum of the energies together with the rest energies (1. A uniform magnetic field is applied perpendicular to the plane of the paper. The counters are connected into coincidence circuits in pairs. for w a y s of fairly high energy ( > 3 Mev). ' ~ . 1512. the "good geometry" method requires highly active sources because only a small fraction of the ?rays emitted by the source .02 Mev) gives the energy of the w a y . less than chambers and electron multipliers. ' ~ ' In the coincidence method only the simultaneous detection of two events in different counters is recorded. in such a way that pairs produced by %rays of the same energy are counted and recorded. Absorption methods can also be used to determine the energies of Yrays. The process of internal conversion has been discussed very briefly in Section 1 4 4 . or as a result of measurements with 7rays of known energy. In principle. and this can be done by using coincidence t e ~ h n i ~ u e s . to achieve a "resolving time" of a fraction of a microsec. in this process the secondary electrons come. A schematic diagram of a pair spectrometer used by Walker and Mc~aniel"" is shown in Fig. as shown in Figs. The electron energies are determined by magnetic analysis and appear as the secondary or line spectrum of Bemittem.
(12' The last method to be mentioned is that of the scintillation spectrometer. The energy may be obtained from a half thickness curve of known energy such as that for aluminum shown in Fig. 145 is feasible. It has been discussed in detail and extended to higher energies by Fowler. An arrangement aimilar to that of Fig. replaced by a %counter. it decreases rapidly with increasing absorber thickneas and becomes zero when the endpoint of the Compton electrons is reached. of c o r n . and Lauritaen. The coincidence rate is plotted againat the absorber thickness. It is also paasible to determine ?ray energies by measuring the maximum range of Compton electrons ejected from a radiator by the mays. Under these conditions and if the 7radiation is homogeneou8. but with the absorbera placed against the counter and with the source placed ss far away from the counter se the intensity of the source allows. In this method. :  ."'*'" the basis of which was discussed in Section 28. No. and absorbera are placed between the two countera. E. a scintillation counter acts as a proportional counter for measuring the energies of the electrons produced by the 7radiation. 1513. When weak sources only are available. and hence the energies of the 7rays. 1 (1948)l. ie used in the experiment. an exponential curve (straight line on semilogarithmic graph paper) is obtained over a range of a t leaat 3 or 4 halfthickneseea. The range measurements may be made in a cloud chamber or with coincidence methods. The &particle counter of the figure is. Nuc2conics. 1513. the Compton electrons pass through two thinwalled Geiger countera connected to a coincidence circuit. Thalliumactivated sodium iodide is most often ueed as the phogphor becauee a large sodium iodide crystal has a :  . This method can be used for energies up to about 5 Mev.Q A m . Lauritaen.R A Y S AND GAMMADECAY Fxa. 2. absorption experiments with "poor geometry" may be used. In the latter method. and this method yields fairly accurate energy values below 1 Mev. Halfthickness values in aluminum for loweneqy 7ray8 [L. Glendenin.
158 Gammadecay : internal conversion.1581 GAMMADECAY: INTERNAL CONVERSIOK 425 high efficiency for the absorption of 7rays. two of which will be described briefly. is a nearly linear function of the energy of the 7ray. as well as the pulse height. I t was shown that with this supposition it is possible to account for the energies of the 7rays associated with adecay and @decay. independent proofs were developed.8 minutes. The amount of light resulting from the capture of a Way.or @particle. After the internal conversion electrons are emitted."~'The electron shell in which the conversion took place was identified from the characteristic xradiation. Since the decay of RaB causes a change of atomic number from 82 to 83. The secondary electrons produce pulses which and electron~positron can be displayed on an oscilloscope screen and photographed. One method. Several different. with a halflife of 26.or bparticles. The result is a spectrum whose maxima give the energies of the 7rays. and the @particlespectrum shows a number of lines corresponding to internal conversion in various electron shells. emits electrons and 7rays. The sum of the kinetic energy of the electron and its  . I t was supposed in Sections 135 and 147 that a nucleus which has been left in an excited state following adecay or @decaycan pass to a less excited state either by emitting a 7ray or by internal conversion. Table 157 gives the energies of a group of @raylines of RaB. and to write down consistent decay schemes. the result showed that internal conversion takes place after the change in atomic number. used by Rutherford and ~ o o s t e r . and considerable work was done in the 1920's to prove this. The emission of an internal conversion electron after a radioactive decay implies that a reorganization of the product nucleus takes place after the decay process has occurred. as measured by Elli~. Rutherford and Wooster determined the wavelengths of some of the L xray lines which accompany the @decayof RaB and found that the wavelengths agreed excellently with the accepted values for atomic number 83. The high atomic number of iodine permits the measurement of "lines" due mainly to photoelectrons pairs. " ~depended ' on the measurement of the secondary xray lines which accompany the decay of RaB. so that both processes represent the same reorganization of the product nucleus. Radium B. and the binding energies of the electrons in the different shells were known. and the counting rate can be determined as a function of pulse height or energy. I t was not always clear that the emission of the ?rays follows that of the a. the vacancies in the electron shells are filled (see Section 77) with the emission of characteristic xrays. and the emission of the %ray must therefore follow that of the a. It was proved unambiguously that the amount of the energy released in internal conversion is the same as that released in 7ray emission.
66 51.99 3. with energies hv .L R ~ .KPt.83 52.hu .63 48. Since the emission o f the internal conversion electron follows the radioactive disintegration.426 GAMMARAYS AND GAMMADECAY [CHAP. kev 53.3 kev for the energy of a %ray associated with the decay of RaB.10 49.Kpt.93 0.KRaB. Another ingenious experiment on the time o f emission o f the conversion electrons was that of Ellis and W~oster. . Independent measurements of this type gave the value 53. . B.01 52.67 13.34 15. 15 &Ray energy. the same value.74 37. where Kpt is the binding energy of the electron in the Kshell of platinum.38 3.84 52.20 Conversion level LI LII %Ray energy. where K R is~the ~ binding . .91 kev. . on the outside of which was placed a thin coating of RaB.90 52.84 52. 52. At the same time the 7rays from the source on the outside of the tube were accompanied by internal conversion electrons with energies hv . kev 36. The difference in energy o f corresponding groups of electrons is K R ~ B .9 kev inferred from the p r a y lines.37 39. L R a ~ Lpt .68 3.83 52.85 49.LPt. as described in Section 157. of the photoelectric spectrum and of the internal conversion spectrum. proving that the emission of a 7ray and of an internal conversion electron represent the same nuclear reorganization. .64 Binding energy forZ = 83 16. and photographs were taken simultaneously. .08 53.91 L~II MI M I I MIII NI 0 Average binding energy was taken to be the energy of the converted %ray. Similar agreement was obtained for many 7rays between direct measurements of their energies and inferences from the &ray lines.17 0."~' They placed a source of RaB inside a thick platinum tube. and in the example shown. The energies o f the 7rays emitted by a given nuclide can also be measured directly with a crystal spectrometer. . The ?rays from the inside source ejected phot* electrons from the platinum. The electrons were studied by magnetic deflection. and on the same photographic plate. and so on. hu . in good agreement with the value of 52. for the ?ray energy was inferred from all of the %ray lines. the emission of the 7ray must also follow the particle decay process. within the experimental error. .04 53.78 52. .
and the experimental values of the energy differences could be obtained from the spacing between lines (Hr values) as seen on the photographic plate. requiring the use of large electronic computing machines.8 min. Values of the internal conversion coefficients can be calculated from quantum mechanics. In another case. i t h which 7emission and internal conversion The relative frequency w occur in a given nuclide is a matter of great interest from the standpoint of theory and is expressed by a quantity called the internal conversion coeficient.8 min. and the practice arose and has been continued of calling these ?rays the rrays of RaB rather than those of RaC. The product nucleus ThC" emits a t least eight different w a y s which can be fitted into a consistent decay scheme as in Fig. These results showed again that the internal conversion electron and. with a halflife of 26.K78 or Kg3 . depending on whether the internal conversion electrons and 7rays were assumed to come from the parent nucleus RaB (s2Pb214) or from the and compared with experimental results. the %ray intensity decays with the same halflife as that of the parent nucleus. the 7ray lifetime for these nuclei is of the order of magnitude of 10l3 to lo" sec. the */ray intensity also decays with a halflife of 26. According to theory. Other experiments were done with both aemitters and & emitters. but rigorous calculations are very laborious. these %rays are usually spoken of as the %rays of ThC and are so listed in tables and charts of nuclides. and it has not been possible to measure the time interval between the two decay processes. etc. This procedure was used for several different lines and in each case the experimental result agreed with the value calculated for Z = 83 and not with the value calculated for Z = 82. In the case of RaB. The proof that the 7rays are emitted by the product nucleus and not the parent nucleus must now be followed by a brief discussion of a generally accepted confusion in terminology. emitting six groups of aparticles. the %ray. 1311. therefore. The binding energies of the electrons in the K. Consequently. approximately as Z3 over part of the range. The coefficient should increase rapidly with the atomic number. Nevertheless.K78. The %rays which have been considered up to now are all emitted immediately after the a.1581 GAMMADECAY: INTERNAL CONVERSION 427 energy of an electron in the Kshell of RaB. is emitted after the @particle. This coefficient is defined as the ratio of the number of conversion electrons to the number of Yray photons emitted and there are different coefficients for the dqerent electron shells.and Lshells of platinum (Z = 78) and of the elements with Z = 82 and 83 were well known.or 0particle. and in each case the Yrays were shown to be emitted after the partide disintegration. product nucleus RaC (s3Bi214). ThC goes by adecay to ThC" and haa a spectrum with fine structure. because it is too short. The energy difference for the Klines was calculated theoretically as Ksz . and .
will be seen to be analogous to Eqs. the combination of experimental results with the available theory makes possible the determination of angular momenta and parities and the construction of energy level and decay schemes. internal conversion is relatively infrequent. a third depends on the angular distribution of the 7rays emitted by nuclei aligned in a magnetic field (see Section 149). effects which did not appear in the relatively simple cases of adecay and Bdecay considered earlier.5 Mev) 7rays are emitted. but these forces are not yet understood. therefore. therefore. the properties of nuclear levels and the transitions between levels cannot be predicted accurately and in detail. however. Some information about the angular momenta and parities of the levels can be obtained by combining experimental results with theoretical relationships between the decay probability and the energy of the 7rays. 159 Gammadecay and nuclear energy levels: theory. (1320) and (1322) for adecay and to Eqs. One such method is the quantitative study of internal conversion coefficients. 15 decreases rapidly as the 7ray energy increases. energy. A knowledge of nuclear forces would be needed to formulate the problem. Consequently. A complete theory of 7decay would require the use of the quantum theory of radiation together with a detailed description of the quantum states of the nucleus.another method depends on the angular correlation between successive 7rays emitted by an excited nucleus. internal conversion is found most often in heavy elements when lowenergy (<0. In the lighter elements and with higher energy 7rays. but some of the detailed predictions of the theory have been verified by means of ingenious experiments. Equations (1540) and (1541) below."" In general. and parity. the theory must be based on a rough model of the nucleus and on a method of treating the interaction between the nucleus and radiation. More precise information can be obtained with the aid of certain specialized methods.428 GAMMARAYS AND GAMMADECAY [CHAP. The study of the 7rays and internal conversion electrons emitted in transitions from excited levels of nuclei to less excited or stable levels is one of the main sources of our knowledge of low energy nuclear levels. discuss the theory briefly with the object of arriving at equations relating the disintegration constant for ?decay and the energy of the ?rays. angular momentum (spin). These levels are characterized by their stationary properties. The energies of the 7rays or conversion electrons determine the energies of the transitions which give. . Nevertheless. The conversion coefficient is a difficult quantity to determine experimentally with any accuracy. Even an elementary discussion of ?decay is complicated. In the present state of our knowledge of nuclear structure. We shall. the energy differences between levels. by the need to show explicitly the effects of angular momentum and parity. (1413) and (1414) for Bdecay. in turn.
the experimental material in atomic transi . The successive terms in the series decrease very rapidly and the decay probability is given almost completely by the first nonvanishing term in the series. ~ ~ (1532) where Ro is a constant with a value of about 1. and E is generally about 1 or 2 Mev. but the discussion which follows should lead to some familiarity with the nomenclature of rdecay and may serve to introduce the reader to some of the ideas underlying the theory. All3 has values up to about 6. In addition.~ / l O = ~ which is much smaller than the values for nuclei. and Now. Although the ratio ( R / x ) ~is small for most of the interesting nuclear transitions. the forbidden transitions are still important in rdecay. (1540) and (1541) is beyond the scope of this book. the succeeding terms. while X is a quantity which characterizes the radiation and is the wave length divided by 2r. The expression for the decay probability (disintegration constant) is given by theory as an infinite series in powers of (R/Q2. This term gives rise to an allowed transition. One reason is that the quantity observed in experiments is the transition rate and. the value of R/X is of the order l ~ .1591 OAWDECAY AND NUCLEAR ENERGY LEVELS 429 The derivation of Eqs. X= E (Mev) lg7 x 10l3 cm. Section 136) R=R ~ x A ~ cm. The wavelength of the rradiation is given by When E is in Mev and X in cm. and A is the mass number of the nucleus. even very low transition rates can be measured. so that the ratio R/X is nearly always less than 0. For atoms. at nuclear energies. so that the forbidden atomic transitions are relatively more forbidden than the nuclear ones. as in the case of @decay (Section 146). The nuclear radius can be written in the form (cf.1. which are much smaller. This situation is in strong contrast with that of atomic transitions where light is emitted or absorbed.5. It has proved to be very useful to express the interaction (force or potential) between the nucleus and electromagnetic radiation in terms of an infinite series of powers of the ratio R/X. give rise to forbidden transitions. where R is a linear dimension which characterizes the nucleus and is taken as the radius of the nucleus.
and even parity when L is odd. Magnetic multipole radiation has odd parity when L is even. The partial probabilities will be seen to depend very strongly on the angular momentum carried off by the photons. For L = 1. the radiation has been shown by theory to correspond to that emitted by a vibrating dipole. called electric and magnetic multipole radiations. In the case of electric multipole transitions of order it has been shown"Q' that the decay probability can be written in the following form when the independent particle or shell model of the nucleus (see Chapter 17) is used:  . or there may be a change in the parity. and odd parity when L is odd. these are shown in Table 158 together with the symbols used for the different types of radiation. there are selection rules which must be obeyed. nucleus and radiation. for L = 3 to that from an octupole. Photons can have only integral values of L. The selection rules for emission of %rays (or for their absorption) define those combinations of L and parity which give nonvanishing values of the transition probability. which is adapted from classical electromagnetics. which diiTer in their parity. separates the radiation from a nucleus into distinct types called multipole radiations. for a given %ray energy. given by 2=. An electric multiple has even parity when L is even. each term in the expansion of the decay probability must be examined to see what it means physically and under what conditions it becomes important. The expansion procedure. for L = 2.GAMMARAYS AND GAMMADECAY [CW. it corresponds to the radiation from a vibrating quadrupole. predominates. This separation corresponds to sorting the emitted ?rays (photons) according to the amount of angular momentum L A carried off by each photon. The decay probability is the sum of the partial probabilities for the emission of these different types of 7rays. For each value of L there are two classes of radiation. but in this sum one term. where +1 means even parity and 1 means odd parity. parity of electric multiple = (I)=. in either case. the parity of the wave function of the entire system. 15 tions consists of spectral lines.1)L. parity of magnetic multipole = (. A given rdecay process may leave the parity of the wave function of the nucleus unchanged. As in atomic transitions. and the intensities of the forbidden lines are so small as to make these lines very difficult to observe. photons can be emitted which differ in their angular momentum. . and so on. is conserved. the first nonvanishing term. the value L = 0 is ruled out as a consequence of the fact that electromagnetic waves are transverse in nature. Since the forbidden transitions are important in rdecay. The multipole radiation is characterized by its order. In other words. and this is one reason for the usefulness of the procedure.
v is the frequency of the emitted radiation. in units of A L ( I symbol Parity change of nuclear wave function No for L even Yes for L odd Yes for L even No for L odd Yes No No Yes Yes No No Yes Electric 2=pole EL I Magnetic 2Lpole Electric dipole Magnetic dipole Electric quadrupole Magnetic quadrupole Electric octupole Magnetic octupole Electric 16pole Magnetic 16pole I L I ML here hgL is the disintegration constant (or reciprocal mean life) for %decay with the emission of EL radiation. electric multi .1591 GAMMADECAY AND NUCLEAR ENERGY LEVELS 431 I I I CIassiScation I Change in angular momentum of nucleus. 9 The probability of emission of magnetic multipole radiation is given by where M is the mass of a nucleon. For a given value of L. The quantity S depends on L and is given by it has the following values. and will not be confused with the wavelength X.
Since the energy of the Yrays is a t most a few Mev. with we have used h/Mc = 0. The results show that when the angular momentum camed off by the ?ray is small (L = 1 or 2 ) . that is. for example. (1540) and (1541) together with the values of 8. . Eqs. (1532) and (1533) note that e 2 / h = 1/137.4 All3 X 1013 which is considered to be suitable for rdecay calculations. the decay probability is large and the mean life is very short. the probability of Ydecay should increase with increasing values of the energy of the 7rays. disand and The frequency v is proportional to the energy so that. values of the mean life (the reciprocal of the disintegration constant) for rdecay with emission of different types of multipole radiation are given for a nucleus with a radius of 6 X 10l3 cm ( A = 80). cm. when the change in the angular momentum of the nucleus is small. we get where RoA1I3 is just the radius of the nucleus in units of Equation (1537) becomes cm.that the probability of ?decay should decrease very rapidly with increasing values of L. It is evident from Eqs.211 X 10l3 cm and R = 1. 15 pole radiation is more probsble than magnetic multipole radiation. Equations (1540) and (1541) are convenient forms for calculation and for comparison with experimental results. the decay probability may become very small and the mean life may become long.432 OAMIIURAYS AND GAMILUDECAY [CHAP. on multiplying the numerator and denominator of the right side of the last equation by h. But when the change in the angular momentum of the nucleus is large (3 or more units). and inserting numerical values. (1540) and (1541)show that. (1535) and (1537) or from Eqs. The theoretical results are illustrated by the numerical examples given in Table 159. for a given value of L. The integration constants can be expressed in terms of the %ray energy the atomic mass of the nucleus if we use Eqs.
05 Mev El MI E2 M2 E3 M3 E4 M4 E5 M5 2 2 3 3 4 4 5 5 3x 3X 3X 3X 5X 6 1x 1X 4X 4x 1044 10l2 108 106 102 105 107 10'2 1014 The values given by Eqs.20 Mev I 0. But the theory should give the general trends of the decay probability with angular momentum. (1540) and (1541) are rough estimates because of the approximate nature of the theory.'20' The total decay rate X of an excited state is given by . Most known 7decay rates have been determined by the direct measurement of the lifetimes T of the excited states involved.1510] GAMMADECAY AND NUCLEAR ENERGY LEVELS 433 Type o f radiation Change in angular momentum o f nucleus. 1510 Gammadecay and nuclear energy levels : experimental results and nuclear isomerism. 1. the independent particle model. for example. . energy. and angular distributions) is still more complicated than that relating decay constant and energy and will not be discussed. The theory of the more specialized methods mentioned at the beginning of this section (internal conversion.= X = A. The use of a particular nuclear model. L: in units of fi 1 1 Mean life. sec Gammaray energy M~~ 3 X lol6 3 x 1044 7 x 10l2 8 X 10lo 6 X lov7 7 X lo5 6X 7 8 X lo4 9 X lo6 0. Measured lifetimes and nuclear isomerism. and mass number. a factor which depends on the actual value of the angular momentum of the initial nucleus is not included. 7 1 + A. may be a more serious limitation. angular correlations. and its usefulness will depend on how well the theoretical predictions agree with experiment.
16 Mev (280/0).13 Mev (13%). Both UX2 and UZ grow out of UXl (goTh234) by 8decay. 0.50 Mev (9%). pairs o f nuclear species exist which have the same atomic and mass numbers. Nearly 100% of the UX2 atoms decay directly to UII by electron emission. These "delayed" transitions are called isomeric transitions and the states from which they originate are called isomeric states or isomeric levels.7 hours. Although most %ray lifetimes have been found to be very short. 1. they are isotopic and isobaric. however. In113" represents an isomeric level of In113. Uranium X2 has a halflife of 1. UII (92U234) by electron emission. .. An isomeric state is usually denoted by the letter m attached to the mass number of the nuclide. UZ emits four groups of j 3 particles with endpoint energies of 0. It must be noted. The first case of nuclear isomerism to be discovered was that o f UX2 and UZ. before we make this comparison. but have different radioactive properties.58 Mev (1%).18 minutes. Uranium 2. 0.53 Mev (27%).31 Mev (90%).e.. these nuclides are called nuclear isomers and their exisfence is referred to as nucbar isomerism.  . i.15% of the UX2 nuclei decay to the ground state UZ by emitting a %ray with an energy then decays to of 0. is the rate of emission of photons and A. can be measured. which have the same atomic number and the same mass number but have different halflives and emit different radiations.32 Mev (3201. The transition rate by rdecay alone is then and it is this experimental value which is to be compared with theory.394 Mev. thus. is the rate of emission of conversion electrons.). that the isomeric property of a state is not absolute but is only relative to properties of states of lower energy. The internal conversion coe5cient a = A&. and both have the atomic number 90 1 = 91. and about 0. more than 250 cases are known in which the lltetime for decay is measurable. it emits three groups of electrons with endpoint energies of 2. ' + . which gives them the same mass number 234. and 0. Because of these delayed transitions. and has a halflife of 6. Uranium X2is more highly excited than UZ. On the other hand. 15 where X.'21' or it can be calculated the~retically'~~' by means of a method which is independent of the calculation of A. and 1. with observed halflives from 10lOsec to many years. The theoretical values of lifetimes shown in Table 159 indicate that some mean lives for Ydecay may be long enough to be measurable.434 GAIWMRAYS AND GAMMADECAY [CHAP. we must look more closely at the experimental lifetimes.the ground state of 91Pa234.'23' The existence of excited states with long lifetimes makes possible the direct study of such stationary properties as the angular momentum and magnetic moment. But. .
which is 35Brs0in its ground state. and the 18 min and 4. The first wellestablished case of artificially produced isomers was found in bromine. then decays by electron emission (92%) to 36Kr80.1510) GAMMADECAY AND NUCLEAR ENERGY LEVELS 435 This example of isomerism was unique for many years. each additional unit of spin change could increase the . 4. It has been found possible to separate the bromine isomers chemically. This result was surprising because there are only two stable isotopes of bromine. until the discovery of artificial radioactivity. Von Weizslicker showed that while rray transitions requiring only one or two units of spin change could be expected to occur with halflives of the order of 10l3 sec. and 4. only two radioactive products would be expected and one of the new bromine isotopes seemed to have two halflives for &decay.4 hr isomer decays into the 18 min isomer by emitting two 7rays in cascade.4 hr halflives were observed in both cases.4 hr. the product was found to show three different halflives for bdecay: 18 min.which is stable. with the following reactions:   Two products were obtained. Since BrsOwas formed in both cases.049 Mev and 0. He suggested that an isomer is an atom whose nucleus is in an excited state and has an angular momentum which differs by several units from that of any lower energy level including the ground state. and 34 hr. and by positron emission (3%) and orbital electron capture (5%) to the ground state of 34Ses0. therefore.4 min. 6. An explanation of isomerism was proposed by von Weizsacker in 1936'~~'. those with larger spin change might result in longer and measurable halflives. 18 min. with energies of 0. When a target containing bromine was bombarded with slow neutrons. and it has been shown that the 4. With bromine. again with three halflives.9 hr. led to the discovery of many isomeric pairs. respectively. and in this region of the periodic table the usual reaction with slow neutrons is the (n. r) process with the formation of the next higher isotope of the element in question. the two halflives were attributed to two isomeric states of BrsO. The identity of this isotope was determined by bombarding bromine with 17Mev rrays. and especially the production of many new radionuclides by neutron bombardment.037 Mev. Chemical analysis identified the carrier of each of these activities as bromine. Br7' and Brsl. The 18 min isomer.
(1548) is estimated to be less than 30%. Later. are not detailed enough to test the dependence of lifetime on mass number. Equation (1540) for electric multipole radiation may be written  Hence. I t has been found experimentally'26' that the measured lifetimes for L = 2.5 .(2L + 1) log E. 3.AZLI3) against log E should be a straight line with slope . 15 halflife of a transition of given energy by a factor of the order of lo6. as well as experimental results have verified von Weizsiicker's hypothesis. The measured quantities in studies of 7decay are A.7 . (1540) and (1541). therefore.(2L 1). A is a parameter which varies frcim nuclide to nuclide.  L = 2: log 7. more detailed calculations. Similarly. + . (sec) = 27.5log E (kev) (1547s) (1547b) (1547~) L = 3: log 7. with this factor the mean deviation of the experimental lifetime from the value given by Eq. The factor (2Zi l).9 log E (kev) The general form of the dependence of the experimental lifetime on the 7ray energies is. the values of the constants were actually obtained from an improved version'25' of Eqs. (1540) and (1541) explicitly for the different types of 7radiation. that predicted by theory. and 4 fit the following equations. such as those discussed in Section 159. we have + . (1546) The comparison of experimental and theoretical results can be discussed more conveniently if we write out Eqs. (or r7) and E . (1547). . The best quantitative agreement between theory and experiment is obtained for M4 transitions for which the experimental data conform to the relation in this case the dependence of lifetime on both energy and mass number is just as predicted by theory.7 log E (kev) L = 4: log 7. 2. a graph of log (r. although the results. as expressed by Eqs. Experimental relations between lifetime and energy. for magnetic multipole transitions. log ) = constant . (sec) = 17. for a given type of electric multipole radiation.436 GAMMARAYS AND GAMMADECAY [CHAP. (sec) = 4 . These equations are listed in Table 1510. where Zi is the angular momentum (in units of rlr) of the isomeric state is a statistical weight factor.
during the last few years. Isomers. been used to improve the model. 8X 1 . The study of isomers has been especially useful in this way and. 0X 1 . The experimental lifetimes for MI. Similar results have been found for the electric multipole transitions.1510] GAMMADECAY AND NUCLEAR ENERGY LEVELS 137 Type o f gammaradiation El Disintegration constant for gammadecay. model. and M3 transitions also depend on energy and maas number in the general way predicted on the basis of the independent particle model. are now playing an increasing role in the investi . the decay rates are generally smaller than those predicted by theory. modifications have been made in it and their validity tested by comparison with experiment. 1X 1 . the independent particle. or shell. 5X 1 . therefore. 5X 2 . For E3. 6X 1. M2. although it is not satisfactory in all its details. the general form of the dependence of lifetime on energy and mass number agrees with theory.2 X 1 . and E5 transitions. which have considerably higher transition probabilities than predicted by theory. 5x 10l4A2I3E" 1013E3 1O8A4I3E5 108A213E5 102A2E7 1O2A4I3E7 105A813EQ 104A2EQ 1011A1013Efl 1 0 4 ~ 8 1 3 ~ ~ ~ which represents good agreement for this kind of nuclear data. especially in certain eveneven nuclei. E4. The experimental data have been interpreted as showing that this model provides a useful basis for the prediction of electromagnetic transitions in nuclei. A. but there are many E2 transitions. which were a t first of interest as a semiempirical branch of radioactivity. 8X 5 . This agreement is especially important because most longlived isomers emit ?radiation of the M4 type. there has been a gradual shift of emphasis.: cmI M1 E2 M2 E3 M3 E4 M4 E5 M5 1 . but the actual numerical values of the lifetimes may be two or three orders of magnitude greater than the theoretical values. The discrepancies between theory and experiment have. 6X 7 . I t will be remembered that the theory mentioned refers to a particular model of the nucleus.
(a) Mg27. [Strominger.has ~ ~ been ' the accumulation of evidence in favor of a more highly developed model of the nucleus.1514. Rws. Hollander.] gation of nuclear structure. Energy level and decay schemes with emission of 7rays from shortlived excited states. One r e s ~ l t ' ~ ' . (b) Lal*O. 30. the colhtive model which will be discussed in Chapter 17. The interest has shifted from the general relations between lifetimes and energies to details of the nuclear structure which determine the multipole character and probability of ?ray transitions. Mod. . and Seaborg. Phys. 15 FIG.438 GAMMARAYS A N D GAMMADECAY [CHAP. 585 (1958). (c) Fk236.
. Bi206 decays by orbital electron capture to Pb206. as in the case of the ground state of Mg27 in (a). If the angular momentum is determined uniquely by other methods. The decay of both members of an isomeric pair may be more complex. Values of the angular momentum and parity assigned to a level are indicated a t the left end of the horizontal line representing the level. The angular momenta of the stable and isomeric levels have both been measured in this case. Energy levels and decay schemes. 3 for the first excited state of A127. it has not yet been possible to assign any values of angular momentum and parity. Some decay schemes are very complicated. 1515(a). Values of the energy. it is designated by the quantum number without any other modification. may be radioactive and decay by @emission. and parity have been established for 10 of the levels. Figures 1514 (a) and (b) contain information not given for and Ce140 in Fig. When the excited level decays only by remission to the ground + +. the 3. for example. BrSom decays to BrSO by emitting two 7rays in cascade. For some states. emits a ?ray with an energy of 0. e. In some isomeric pairs the ground state. indicated by a plus sign (even parity) or minus sign (odd parity). which then decays by electron emission to the ground state of Gas'. with a halflife of 104 min. Thus. Parity assignments. and extensive and careful analysis is needed to work them out.. the energy above that of the ground level is given a t the right end of the lines. and by positron emission and orbital electron capture to SeSO. BraO then decays by electron emission to KrsO.1510] GAMMADECAY AND NUCLEAR ENERGY LEVELS 439    3. Probable values of the angular momentum are indicated with parentheses. it is denoted by underlining that quantum number in the decay scheme. Some decay schemes involving the emission of 7rays by shortlived excited states are shown in Fig. for the ground state of A12' in (a). e. depend heavily on theory and the underlining and parentheses do not apply to them. the 28 observed 7rays have been fitted into a decay scheme which meets all the experimental and theoretical requirements. instead of being stable. as in the case of BrsO. 1515. as shown in Fig.. 1514.g. 1413. which is stable: In 13". angular momentum. 1515(b). Ground states of eveneven nuclei are known to have zero angular momentum and often fall into this category. 3 . and probable values assigned to the other two. In the simplest mode of decay the isomeric level decays by remission to the ground level. When the angular momentum of a state has been measured directly. Thus. the decay scheme is shown in Fig. 14 hr Zns9" emits a 0.and 12 excited levels and 28 7rays have been found so far. they differ by four units and the transition has been shown to be of the M4 type.g. 1515(c).392 Mev and becomes stable InH3.438 Mev ?'ray and goes to the ground state of Zns9.'2'' Examples of energy level and decay schemes for nuclear isomers are shown in Fig.00 Mev level of Ce140. these schemes show that members of isomeric pairs can decay in different ways.Fig.
Mod. (b)Zn6gm. (a) In113=. Hollander. and Seaborg. 30. (e) Cd"". 585 (1958). (d) Ge77m. [Strorninger. Phys. 15 FIG. (c) Braom. (f) Sb12'"1 and Sb124m2.1515.] . Decay schemes of nuclear isomers. Revs.440 GAMMARAYS AND GAMMADECAY [CHAP.
New York: Wiley. New York: McGrawHill. 1959. Triple isomerism has also been observed and Fig.C. 1108. E. Part X."Passage of Radiations Through Matter. ref. 1957. The more excited member of an isomeric pair may decay in two ways: Ge77mdecays t o GeT7by emission or' a ?'ray. 1953 (Vol. 1959. Chapters 12. H. Mass. 1515(f) shows a n example. 42 (Nuclear Reactions 111). R." Revs." Experimental Nuclear Physics.. XRay AbEMtion C o e j m from 10 kev to 100 Nev. 23.S. Reading. 111. as in the above cases. New York: The Macrnillan Co. and SEABORQ. Sept. i t is especially interesting became two of the energy levels of the product nucleus. Phys. 1955.. Sb124ma. 1960.MCGINNIS and VAN LIESHOUT. New York: Wiley. EVANS. Atomic Energy Commission Document TID5300. the ground level of Ge77 is unstable and decays to excited levels of As77 by electron emission.S. it can also decay by electron emission to the ground level of As77 or t o a n excited level.).3 min. 585 (1958). Parts A. .. GRODSTEIN.KING. HOLLANDER. Vol. Radiatim from Radioactive Substances. 15. 1957. 25. The ground state of Sb124is unstable and also decays by @emission. Nuclear Level Schemes. Washington. GOLDSTEIN. 1955. New York: Academic Press. The decay scheme is shown in Fig.REFERENCES 44 1 level. SIEQBAHN. Chapters 5.S. K. Berlin: Springer. W. Vol. Mod. KOFOEDHANSEN. Nuclear Physics.: AddisonWesley. Fundamental Aspeek o f Reactor Shielding. GENERAL RUTHERFORD. Nuclear Spectroscopy. 1930. ed. BETHEand J. ALBURGER. CHADWICK. In116.A. National Bureau of Standards Circular 583.. Government Printing Office. and both have complex &spectra. the decay scheme for Ge77m and GeT7is shown in Fig. "GammaRays. U. ed.." Encyclopedia of Physics.: U. pp.with a halflife of 1. decays both by remission and @emission. DEUTSCHand 0. ASHKIN. 30. the members of the isomeric pair are said to be genetidly relrrted. A = 40 WAY. 1515(d). 24. The members of a n isomeric pair may decay independently by @emission: Cd116" and Cd115 are not genetically related. STROMINGER. I. D. G." Experimental M. For a detailed summary of isomeric decay schemes the reader is referred t o a recent review by Alburger (see gen.. H. F. Segd. A. E. The Atomic Nuclew. D. U. 1515(e).and GammaRay Spectroscopy. ed. while Sb124m1. 1955. "Table of Isotopes. ed.. New York: Interscience. E. and ELLIS. AJZENBERQSELOVE.with a halflife of 21 min. Segrh. Beta. Amsterdam: NorthHolland Publishing Co. D. decays by remission to the ground state. form a n isomeric pair. B. "Nuclear Isomerism. Part 11). to A = 92.
BUTT and V. T h e t i c a l Nuclear Phyaks. D. Sei.  . and GUNST. 494. Revs. 132 (1947). Znatr. .404 (1951). Phys. G."Theory." Revs. M d 2 and V48."Measurement of Short Lifetimes of Excited States: Measurement by Delayed Coincidence. 12. Mod. DEUTSCH. EVANS. E. C. p. 69." Proc. 4. 163 (1958). copy of 7Radiation. D. (London) A143. WlS7."Direct Comparison of m/e for the Positron and the Electron. W. 18. Siegbahn. LAURITSEN. 1 19. aiso. M."Measurements of GammaRay Absorption Coefficients. 626 (1947). RaTh. M. Oxford University Press. 24. CurvedCrystal Focussing Spectrometer for Short WaveLength XRays and GammaRays. 5. Cs13'. WALKERand B." Revs. W." Beta.  + + . W. W. "High Resolving Power. 88. Phys. Rn.KLEIN. 178 (1944). C. ''The Atomic Number of a Radioactive Element a t the Moment of the Emission of the 7Rays. Siegbahn. and LAURITSEN. ed." Proc. "GammaRay Spectroscopy by Direct Crystal Diffraction. 20. Aul9S. C.''The 7Rays of Radium (B C) and of Thorium (C C')." Phy8." Rev. HOYT.STEHLE. 81. Mod. Roy. 79107. FOWLER. 20. 306 (1946). Chapter 12." Beta. 1954. D. DUNWORTH. 74." Proc. DAVISSON and R.and DUMOND. DEUTSCH. 1952. L. Phys. 296 (1948). 315 (1948). Mod. 844 (1925). A. 2. J. Cambridge Phil. 9 (1952). "Coincidence Methods in Nuclear Physics.). 6. R. 3. 9. DAVISSON and R. ref. R. MCDANIEL. EVANS. Rev.442 PARTICULAR GAMMARAYS AND GAMMADECAY [CHAP. The Quantum Theory of Radiation. 775 (1952)." Physiea 1 8 . 11.350 (1934). 16. 8 . 18. 167 (1940). (see gen. "Precision Measurements of Nuclear GammaRay Wavelengths of IrLB2."Phys. Soc." Revs. J. J. K. M. C a m m e Phil. "Disintegration Schemes of Radioactive 8. C. Sci." Phys."Scintillation Counter as GammaRay Spectrometer.14. 834 (1925). 20. 89. Sci. WEISSKOPF. P. "GammaRay Absorption Coefficients. 13." Phys." Rev. PAGE. C.3rd ed. Design and Applications of a Short Magnetic Lens Electron Spectrometer. Rev. IX. ERIKSEN. 1273 (1953). Zmtr. 11. BELL." Revs." Phys. M. R. G. and SUNDE. and EVANS. Znstr. HEITLER."GammaRay Spectrometer Measurements of Fluorine and Lithium Under Proton Bombardment. WOOSTER. (see gen. 15. Phys. D. 236277 (1948). 17. TaLS2. D." Ann. Rev. MITCHELL. ed. 133. ref. "GammaRadiation from Excited States of Light Nuclei. and M. 10. PEACOCK Substances. "Interaction of %Rays with Matter and the Spectros9 . Soc. Rev. BELL. JENSSEN. p. C. MULLER. "The Scintillation Method. DUMOND. 22. D. Mod. RUTHERFORD and W.and GammaRay Spectroscopy. K . WOOSTER. F." Rev. New York: Wiey. Comparison Methods. 7. ELLIOTT. Sn'. Nucl. ELLISand W. E. LATYSHEV. "The Natural XRay Spectrum of RaB. "The Use of Coincidence Counting Methods in De termining Nuclear Disintegration Schemes. . Soc. ELLIS.and GammaRay Spectroscopy.). 1 5 . 1952.15 1. A. Rev. and Annihilation Radiation. 22. A.
"Theory of Multiple Radiation. C. E. "Energy Levels in Pb206 from the Decay of Bi206.'' Beta." Phys. 83. M. 22. . H. 1310 (1961). and MARK.and GammaRay Spectroscopy. "GammaRays." 26. 29. A. 360. SUNYAR. "Classification of Nuclear Isomers. 983 (1960). 30." J. 1482 (1954). Energy. VARSHALOVICH. JOPSON." Rev. 10. 396. p. GOLDHABER Phys." Beta.). 27. 25. 21. and MARK. ROSE. D. M. ALBURGER and M. Siegbahn. M. CHUPP." Experimental Nuclear Physics. GORDON. Translated from Atomnaya Enerpiya 5. N w l . 906 (1951).. VON WEIZSACEER. HELMHOLZ. Rev. W. Scd."Measurement of the Deuteron Binding Energy using a BentCrystal Spectrograph. 170 (1959). 23. (see gen. 31. ed. Part A: Reactor Science. Mod. KOFOEDHANSEN.and GammaRay Spectroscopy. S E G R and Phys. 24. 112. PRYCE. L L E l e ~ t r ~ m a g Transitions neti~ in Isomeric Nuclei. 271 (1949). KAZI. E. DEUTSCH and 0. and A. ref. "Metastabile Zustande der Atomkerne. Segrk. 432 (1958).RASMUSSEN. 453. and MARK. R u s r ~ o v and D." Phys. A. p."Precise Determination of Nuclear Energy Levels in Heavy Elements. "Nuclear Isomerism.RASMUSSEN. W. M. Inair. 32. K. F. ref. 518 (1958)." Revs. 95. "Theory of Internal Conversion. S. L. ~ A. Rev.REFERENCES 443  21. Rev.). L. Siegbahn." Naturwiss. E. DUMOND. Rev. "Chssification of Nuclear Isomers. ed. 373." Beta.) p. ed. ref. K. p. 31." Phys. ref. 813 (1936)."SixMeter Radius BentCrystal Spectrograph for Nuclear Gamma Rays. 123. C. I. (see gen. KAZI. 28. E. 24. MOSZKOWSKI. GOLDHABER and A. (see gen. SUNYAR.and GammaRay Spectroscopy (see gen.).
095 0.308 0.14 Mev = 2. and 25 Mev.161 0. K for pair formation in aluminum and lead a t photon energies of 1. 20.010 0.634 0.r for photoelectric absorption in aluminum and lead a t the photon energies listed in Table 152. born the curve of Fig. graphically. graphically.444 GAMMARAY8 AND GAMMADECAY [CHAP.0017 Energy = 1. 10.0090 0. the value of the absorption coefficient ~ ( c m . 4.75 Mev Energy . Calculate the valuee of the cross sections u.065 0. Calculate the values of the cross section .043 0. the value of p(cml) in each case and compare the results with the values given in Table 151.026 0.76 Mev pass through varying thicknesses of aluminum.00 0. A heterogeneous beam of 'Yrays from a radioactive source is allowed to fall on an aluminum radiator and the electrons ejected are snalyaed in a magnetic spectrograph. respectively. 5.020 0..675 0. and 2. ~ r s Thicknem of aluminum.. 2. the intensity varies with the energy of the rays and the thickness shown in the following table. 1. cm Relative intensity Energy = 0. calculate the values of the c m section . and lead..102 0.835 Mev 1. 15 1. and 2750 gausscm.0042 0. Plot the resulting values against the photon energy in Mev.14 Mev. from theory. Plot the values obtained againat the photon energy in Mev. 1. Calculate. 15. The electrons fall into three group6 with maxima a t 11960.. Compare the results with the values given in Table 151. 5845. Compare the results with the theoretical and experimental values given in Table 151.0041 0.835 Mev. What are the energies of the three 7rays? .456 0.0019 Find.00 0.254 0. a. When a beam of Yrays passee through aluminum under "good geometry" conditions.I ) for copper for ?rays of energy 0. the intensity varies with the thickness of each substance aa shown in the table on the next page.I ) in each case.141 0. 7. 3. for Compton scattering in aluminum a t 10 values of a which cover the energy range of Table 153. tin. the values of ~ ( c m . and u.76 Mev. of the aluminum. 6 . When 'Yrays of energy 2. Find.208 0. 5. 156.401 0..
586 0. cm 0 2. platinum.1. and 194.154 0. and the mass of the recoil nucleus in m u . the rays are allowed to fall on a lead radiator.0. In an experiment on the Yrays from Ba131.0782. Two of the 7rays emitted by the latter nuclide give rise to an internal conversion line spectrum with the following energies: 32.036 0.263 0.86. corresponding to Hr values of 1250.387 0. and the shell and Yray with which each conversion line ia associated.041 0.5 10. Find the energies of the two 7rays.respectively.014 0. lead. and uranium. What are the energies of the Yrays? 9. and 0. 2050.622 0.449 0.84.093 0. cm Intensity 1.1178 Mev.788 0.4. The binding energy of the Kshell electrons in lead is 0. 11. respectively. 78. Ellis measured the energiea of the photoelectrons ejected from radiators of wolfram.240 0.0891 Mev. In each caee.4 kev. 1445.022 11 Tin Thickness. express the energy of the Yray in kev. The binding energies for K.008 Lead Thickness.00 0. and 2520 gaUB&em. 0. Derive an expression for the kinetic energy (in ev) with which the nucleus recoils when a Yray is emitted.0891. and 0.069 0. three groups of Kelectrons were observed with the following energies.150 0. 1. in Mev. .5 15 20 25 30 35 40 Intensity 1. Terbium160 decays by @emissionto excited states of dysprosium160. 143. respectively. Find the mean energy and the wave length of each of the three Yrays.0 12.7.0.0693.9.0 7. L.490 0. M.0.305 0.7 kev. 8.9. In an early experiment on the Yrays from RaB. Four groups of photoelectrons are observed.6.058 0. 10.PROBLEMS Aluminum Thickness.118 0. cm Intensity  8. 0. 187. and Nshell electrons in dyeprosium are 53.00 0. The binding energies of the Kelectrons in the four radiators are 0.5 5.
respectively. The nuclide Cr51 decays by orbital electron capture to V51 and the Qvalue for the ground state transition is 0. Two Yrays are observed. Construct a decay scheme to fit these data. and 0. belong? 13.061 Mev. respectively. Construct an energy level and decay scheme to fit these data. and 0. while the (a.75 Mev.33 Mev.7min level has an excitation energy of 0. To what nuclide. are observed. one with a halflife of 10.4 min. The 10. Gammarays with energies of 0. with energies of 1.31 Mev.n)Mn. 15.8 days.39 Mev.n) reaction on vanadium gives a product with a halflife of 310 days. The reactions Cr(p.446 GAMMARAYS AND GAMMADECAY [CHAP. 1. The irradiation of Co50 with neutrons produces a pair of isomers of CosO.325 Mev. 15 12. and Fe(d. or nuclides.54 Mev.320 Mev. The Qvalue for the transition from the isomeric state of CosO to the ground state of NisO is 2. with energies of 0. Construct an energy level and decay scheme based on these data.876 Mev.17 Mev and 1. Three Yrays are observed. and 0.089 Mev.2 years. do the two shorter activities . in cascade. the other with a halflife of 5.a)Mn both lead to activities of 21. . 0. The ground state level of Coso decays by @emission with an endpoint energy of 0.73 Mev.059 Mev and transforms mainly to the ground state of CosO by internal conversion. 14. emitting three groups of particles with endpoint energies of 1. The nuclide Cr40 decays by positron emission. the first two of which are in cascade.15 Mev.65 Mev respectively. 0. i t also decays by emitting @particles with an endpoint energy of 1.54 Mev.3 min and 5.
a wealth of experimental data for the field of nuclear spectroscopy and for the theory of nuclear structure.g.. The study of nuclear reactions is also important for other reasons. the nuclear reactions considered were of the type or. or they may themselves be nuclei.CHAPTER 16 NUCLEAR REACTIONS In the discussion of nuclear reactions in Chapter 11. The quantitative aspects stressed were those of the mass and energy balance. (161) does not cover all the reactions of interest. neutron. X* Z + x. The equation and the notation both mean that particle x strikes nucleus X to produce nucleus Y and particle y. Any nuclear transmutation process may lead to an excited state of the product nucleus. In the general case. In Chapter 11. The incident and emitted particle may even be of the same kind. + z. y)Y. aparticles or deuterons. etc. The great number of nuclear reactions which are possible provide. The transmutation represented by Eq. 161 Nuclear reactions and excited states of nuclei. For the purpose of this chapter. The particles x and y may be elementary particles or ?'rays. and the decay of the excited state then gives information about energy levels and decay schemes. deuteron. e. X(x. It is from this standpoint that nuclear reactions will be considered in the present chapter. the nuclear reactions to be considered may be represented by the set of equations x + x. The process may be caused by any kind of incident particle and may result in the emission of any kind of particle (proton. the emphasis was on the production of new nuclear species and the interest was mainly in the change of one nucleus into another. Information about the relative probabilities of different reactions provides clues to the problem of nuclear structure and offers a testing ground for ideas about nuclear forces. . uparticle. as in scattering processes. in more compact notation. more than one particle may emerge. or ?'ray). therefore. or the outgoing particle may be the same as the incident particle.
state as before the collision. The other reactions of the set represent different possible nuclear transmutations in which the product nuclei may be formed in their ground states or. respectively. and the total kinetic energy of the system is decreased by the amount of the excitation energy given to the target nucleus.30 Mev. two groups of emergent protons were found a t right angles to the incident aparticle beam. The energy balance of the reactions. one or more groups of protons. A proton group of lower energy gives a lower Qvalue. 16 In the first two reactions of the set. the outgoing particle is of the same The kind as the incident particle. when the latter was bombarded with aparticles for each of several from RaC'.". The second reaction represents inelastic ng. When boron was bombarded with aparticles from polonium with a range of 3. is the same before the collision as after. in excited states. Some kinetic energy is usually transferred from the projectile to the target nucleus. The excited product nucleus. and the process is called sm&ring. These ideas and their application involve some of the basic procedures of nuclear physics and will be illustrated by an analysis of two experiments. When a given light element was bombarded with monoenergetic natural aparticles. p) reactions on light elements. four groups of protons were ob~erved'~' different energies of the incident aparticles. or nuclear. or Qvalue. The proton group with the greatest energy gives the greatest Qvalue. of the system. which the target nucleus X is raised into an excited state X*. A collision between two billiard balls.usually decays very quickly to the ground state with the emission of Yrays. these protons had ranges of 20 cm and 50 cm. but the latter is left in the same internal. more often. each containing particles of the same energy. or an energy of 5. and the difference between the greatest Q and a smaller one gives the excitation energy of the product nucleus after emission of the lower energy proton.448 N U C L U R REACTIONS [CW. can be calculated for each energy group by the methods of Section 112. and this value is supposed to correspond to the ground state of the product nucleus. was observed for a particular direction of emission. The fact that the product nucleus in a transmutation process can be left in an excited state was discovered by measuring the energies of the protons emitted in (a. Similar results were obtained with other light elements such as fluorine and aluminum.8 cm. on the reaction . in which neither ball is damaged or otherwise changed. in which the total kinetic energy first reaction represents elastic ng." The existence of the distinct proton groups was demonstrated by their different ranges. the first nuclear reactions studied in detail. projectile plus target. is an example of an in elastic collision.
Mev 9. The results.34 6. and 2530 cm. There are. In one experiment. 1. obtained from rangeenergy curves for protons.4 .65 Mev. respectively. 40. since the angle a t which the protons are observed is 0°.24.65 1. which are collected in the first part of Table 161. are in good agreement with those of earlier experiment^.06.44.22 0. Mev emitted at 90°.'^*" It waa also found in this experiment EXCJTEDSTATES OF SiS0 FROM THE BOMBARDMENT OF ALUIUINUM BY ALPHAPARTICLES EzperimetlJ 1 (') Energy of aparticlea = 7.1611 NUCLEAR REACl'IONS AND EXCITED STATES O F NUCLEI 449 which are representative of work relating nuclear reactions to energy levels. 5. and 4.3 Mev.3 Mev Energy of emitted at oO.54 Mev Energy of protons Energy of excited state. as are the mass numbers of the initial and final nuclei and. 60. The energies of the incoming +particle and the outgoing proton are known for each group.24. The ranges of the protons emerging at an angle of 0" with the direction of the incident beam were measured by counting coincidences between protons and w a y s . Four groups of protons were observed with ranges in air of 101.44 2.98. The ground state energy is taken as the zero point of the energy scale for the various states.06 I Ezperiment 8(@ Energy o f aparticles = 21.6 cm.22 Mev. the three lowerenergy proton groups give Qvalues of 0. Mev 5. which corresponds to the formation of Si30 in its ground state. three excited states of Si30 with energies 2. Qvalues can be calculated from Eq. MeV 2. The corresponding energies. 6."' an aluminum foil was bombarded with aparticles accelerated in a cyclotron to an energy of 7.55.6 Mev above that of the ground state.4 Mev. are 9.28. therefore.8 cm. Mev Energy of excited state.55 4. and about 4. (1110).34.66.8 cm. 3. The greatest proton energy gives Q = 2. and about 2. Mev Qvalue. respectively.98 I Qvalue.
the number of protons counted in a particular group showed a peak a t the end of the range so that when the number of counts is plotted against the thickness of aluminum. Under the conditions of the counting procedure. the lowerenergy protons are indeed associated with excited states of the Si30nucleus. 161. In another e ~ ~ e r i m e n t . and are listed in the   . that no 7rays accompanied the highestenergy group of protons. Proton groups from the reaction A12'(a. lbl. The change of the counting rate with the thickness of the aluminum absorber is shown in Fig.m g / d of A 1 Wo. p)Si30 (Brolley et a1. while the lowerenergy protons did show coincidences with 7rays.C8)).NUCLEAR REACl'IONS Total absorption in proton beam.54 Mev. The protons emerging a t an angle of 90' with the direction of the incident beam were counted with two proportional counters placed in series and arranged for coincidence counting. The ranges of the proton groups in aluminum were measured by allowing the protons to pass through aluminum foils before entering the counting system. each peak in the curve corresponds to the range or energy o f a different proton group. The energies of the proton groups were obtained from the ranges in aluminum by means of the known theoretical relationship between range and energy in that material. while the highestenergy group is associated with the ground state. a ' ~ thin ' aluminum foil waa bombarded in vacuum with aparticles accelerated in a cyclotron to an energy of 21.
The range of the most energetic proton group could not be determined well in this experiment because the counting rates became too low. and it is probable that there are more which have not been excited by the (a. as might be expected on intuitive grounds. and 6.6 Mev observed in the first experiment was not observed in the second experiment. however.161) NUCLEAR REACTIONS AND EXCITED STATES O F NUCLEI 45 1 second part of Table 161.66 Mev and 3. p) reactions when the energy of the incident aparticles was varied. and was markedly lower in the energy range between any two of these peaks. In each case. p) reactions. 5.49 Mev correspond to the same excited state. . therefore. p) reaction. Another important feature of many nuclear reactions was discovered during the investigation of (a. The level at about 4. 5.22 Mev was used for the reaction leading to the ground state of Si30. Resonance is important in nuclear reactions because knowledge of the resonance energies in a reaction yields information about certain energy levels of the nucleus referred to in Chapter 11 as the compound nucleus. and lead to the same excited states of the product nucleus.86. There are.0. The emission of groups of particles is not limited to (a. but other work has established the existence of a level a t 4.49.75. and Qvalues can be calculated from the known resonance energy and the corresponding energies of the proton groups. It was found in the early experiments that when the energy of the incident aparticles was increased the yield of protons did not increase monotonically. The energy level values a t 3.25. The Qvalues are obtained from Eq. and gave the energy levels listed in the table.75 Mev. The experimental results obtained a t the resonance energies are often more useful than those obtained a t other energies because more events occur at the resonance energies and more precise information can be obtained about the Qvalues and excited levels in the compound nucleus.'3' In the bombardment of A1 by aparticles.61 Mev. so the value Q = 2. and the particular energies a t which the maxima occur are called resonam energies. as shown in the table. but is also found in many other reactions. the lowerenergy groups corresponding to excited states which decay to the ground state by emitting 7rays. at least 9 excited energy levels of Si30. The resonance phenomenon does not. add anything new about the energy levels of the product nucleus. the yield of protons had peaks a t aparticle energies of 4. 4. the different energy groups correspond to different states of the product nucleus. These Qvalues are the same as those obtained for nonresonance *particle energies. 4. At each resonance energy of the incident aparticles the different proton groups are observed. The occurrence of maxima in the reaction rate at different energies is called resmnce. (119). the discrepancy being experimental. The reason for the use of these terms will be discussed later. but showed sharp maxima at certain discrete values of the energy.
According to Bohr." of the compound nucleus. Bohr's assumption is also in accord with the picture of the nucleus as a system of particles held together by very strong shortrange forces. etc. which has been extremely useful in the correlation and interpretation of nuclear reactions. the new nuclide may be Mg2'. 16 The transmutations discussed in Chapter 11 were written in a way which indicates that the incident particle combines with the target nucleus to form an intermediate compound nucleus. 1 6 2 The compound nucleus. and it turns out that the existence of resonance energies is related to the energy levels of P31. and the state of the compound nucleus is then independent of the way it was formed. Incident particle initial nucleus + compound nucleus. several different new nuclides can be formed. The two steps of the reaction can then be considered separate processes:    1. which is nearly equal to the kinetic energy of the captured particle plus . In 1936. or Na2'. neutron. or target. + + Bohr's assumptions are in accord with many of the facts of nuclear transmutation. Bohr assumed that a nuclear reaction takes place in two steps: (1) The incident particle is absorbed by the initial. Si2'. or into Si2' and a 7ray. When A1 is bombarded with aparticles. nucleus. leaving the final. The basic ideas of this theory will therefore be considered before a detailed discussion of nuclear reactions is undertaken. When the incident particle enters the nucleus. the interaction between A12' and a proton is assumed to give the compound nucleus [14Si28]. or even into Na2'. Compound nucleus + product nucleus outgoing particle. That this conclusion is reasonable can be shown by the following arguments. On being captured. Bohr assumed also that the mode of disintegration of the compound nucleus is independent of the way in which the latter is formed. or into Si2' and a neutron. Resonance is not limited to (a.) or a Yray. its energy is quickly shared among the nuclear particles before reemission can occur. When A12' is bombarded with protons. 3 protons. such as its energy and angular momentum. aparticle. or product.   . 2.452 NUCLEAR REACTIONS [CHAP. the incident particle makes available a certain amount of excitation energy. It was shown in Section 119 that when a given nuclide is bombarded with particles of a single type. and depends only on the properties of the compound nucleus itself. Bohr proposed his theory". nucleus to form a unnpound nuelezrs. the compound nucleus is [P3']. (2) The compound nucleus disintegrates by ejecting a particle (proton. and a neutron. Si2'. which can then disintegrate in any one of several ways: into Mg2' and an aparticle. p) reactions but is also found in other nuclear reactions.
and compound nucleus. Immediately after the formation of the compound nucleus. and the disintegration + . During its relatively long lifetime.1b2] THE COMPOUND NUCLEUS 453 its binding energy in the compound nucleus. I f the incident particle is a 1Mev neutron. a t a later time it may be shared by other nucleons.00898and 26. the time required for a 1Mev neutron to cross the nucleus is of the order of sec. At a given instant. 1 The masses of the interacting neutron and nucleus are 1.99008 amu.72 Mev. Consider. t o which must be added the kinetic energy of the incident neutron. 1 3 ~ 2 7 + on' . The magnitude of the excitation energy can be calculated from the masses of the incident particle. the latter has a lifetime which is relatively long compared with the time that would be required for a particle to travel across the nucleus. and the kinetic energy of the incident particle. the excitation energy is nearly 7. but this additional energy is rapidly distributed among the other particles in the compound nucleus as a result of interactions among the nuclear particles. Even a slow neutron with a velocity of lo5 cm/sec would need only about lo'' sec to cross the nucleus. that of the compound nucleus is 27. Since the diameter of the nucleus is of the order of 10l2 cm. sometimes called the natural nuclear time. companding to 7. the excitation energy may be shared among several nucleons. or a combination of nucleons. or it may eventually again become concentrated on one nucleon or combination of nucleons. its speed is about lo9 cm/sec. I n the case of a slow neutron. The distribution presumably takes place in a random way.72 Mev and the energy of the compound nucleus [18Ala8] is greater than the energy of the ground state of Alls by this amount. the excitation energy may be considered t o be concentrated on the captured particle.99906 amu. target nucleus.72 1 = 8.99077 amu.[I~AP]. and is usually about 8 Mev. the compound nucleus "forgets" how it was formed. the capture of a neutron by Ala7 to form the compound nucleus [ ~ l ~ ~ ] . . is of the order of magnitude of the diameter of the nucleus divided by the speed of the incident particle. one nucleon. which is long compared with the natural nuclear time. and the compound nucleus disintegrates into the product nucleus and outgoing particle. It will be seen below that the lifetime of a compound nucleus may be as long as 1015 or lo" sec. The latter time interval. The mass excess is 0. the kinetic energy may be neglected. if the excitation energy is large enough.. If the incident neutron has a kinetic energy of 1 Mev.00829 amu. As a result of the random way in which the excitation energy is distributed in the compound nucleus. The energy that must be concentrated on a single nuclear particle or group of particles in order to separate it from the compound nucleus is called the separatim or dis8ocMtion energy. for example. may escrtpe. . or a total of 27. In the latter case.
The fraction of the particle energy that must be added to the binding f energy in order to get the excitation energy can be derived as follows. There are many ways in which the excitation energy of the compound nucleus can be divided among the nuclear particles and.454 NUCLEAR REACl'IONS [CHAP. V. incident particle plus target nucleus. 16 is independent of the mode of formation.68 Mev. Then . the wave and circuit are in resonance and the reception ia good. and MCN are the speed of the incident particle. .62 Mev. The excitation energy of a level cannot be obtained just by adding the resonance energy to the biding energy. there are many possible virtual levels of the compound nucleus. it can still disintegrate by ejecting one or more nucleons. When the frequency (energy) of the incoming wave is equal to that of the circuit. p)Si30 reaction discussed above. because that would neglect the motion of the compound nucleus. the reception is poor.  . . I the target nucleus is assumed to be a t rest before the collision. the probability that the compound nucleus will be formed is much greater than if the energy falls in the region between two levels. In the AlZ7(a. where v. These quasistationary states are usually called virtual states or virtual levels in contrast to bound states or bound levels. the momentum of the incident particle must be equal to that of the compound nuecleus. These levels are closely related to the f the energy of phenomenon of resonance.95 Mev to 8. is equal to the excitation energy of one of the levels of the compound nucleus. The system is analogous to that of a radio wave and a tuned receiver circuit. The occurrence of a resonance peak in the rate of a nuclear reaction when the energy of the incoming particles is varied shows that the compound nucleus has an energy level whose excitation energy is very nearly the sum of the binding energy of that particle and its kinetic energy. when the two frequencies (energies) are not equal. 18 mnances have been observed a t aparticle energies from 3. I t is reasonable to assume that i the incident particle is such that the total energy of the system. Some of the energy of the incident particle is used in supplying kinetic energy to the compound nucleus and is not available for excitation. since each distribution corresponds to a virtual level. and the speed and mass of the compound nucleus. which can decay only by emitting ?radiation. respectively. the binding energy of the aparticle in the compound nucleus [P3'] is 9. The compound nucleus may be said to exist in a "quasistationary * state. which means that although it exists for a time interval which is very long compared with the natural nuclear time.
The reactions A12'(a..1621 THE COMPOUND NUCLEUS 455 The portion of the energy of motion of the incident particle that goes into exciting the initial nucleus will be called E. The excitation energies of these levels are considerably greater than the dissociation energy (8 Mev) for a particle and the compound nucleus can disintegrate by particle emission so that the levels are virtual. on the average. there is a certain period of time. and is given by  In calculations of this kind. Mx Mx) + . p)Si30 and A12?(a. whether bound or virtual. the masses may be replaced by mass numbers without introducing any serious errors. The reciprocal of the mean life is the disintegration constant. the accuracy with which the energy and time can be determined for a quantized system such as an atomic nucleus is limited by the relationship . According to this principle. In the discussion of energy states excited by nuclear reactions it is customary to use. which gives the probability per unit time of the emission of a particle or ?ray.. n)P30 have been observed. a quantity proportional to it. and its use is based on an application of the Heisenberg uncertainty principle (Section 78).   since Mclv = Mx m. called the level width. during which the nucleus remains in a given excited state before decaying by emission of either a particle or a ?ray. Equation (163) may also be written = Ez (m. and defined by the relation + The level width has the units of energy. a result which is consistent with the idea that the compound nucleus [P3'] formed by the bombardment of aluminum with aparticles can disintegrate in more than one way. The energies of the excited levels of [P31]which correspond to aparticle resonances are given in Table 162. instead of the disintegration constant. Each excited state of the compound nucleus. has a certain mean lifetime T .
The mean lifetime of an excited state may be identified with the Z C in the energy. in energy. for each decay product. and so on. . For each possible mode of decay. the radiotion width. where r. The uncertainty At corresponding to an uncertainty A latter is defined as the width I'. therefore. Mev where AE is the uncertainty in the energy and At ia the uncertainty in the time. giving Eq.456 NUCLEAR REACl'IONS [CHAP. Mev Energy to be added to the biding energy. A state with a very short mean lifetime is poorly defined in energy and the width r is relatively large. while a longlived state is sharply defined in energy and the width is relatively amall.. a diflerent partial width r. of the excited level. (165). 16 aparticle mnance energy. r.'I is the aparticle width. there is a different probability of decay and. is a measure of the probability per unit time for emission of a 7ray. The total width r of an energy level is then the sum of the individual partial widths.. Mev Excited level of [F11. is the protun width. .
Along with the level width.6 X sec. When the narrow level is that of a compound nucleus. 163 Cross sections for nuclear reactions.6 X 10l5 sec. usually denoted by o with an appropriate subscript. can be obtained from resonance measurements and is an important quantity in nuclear spectroscopy. for a wide level. and can be applied to both bound and virtual levels. or mean distance between levels. as will be shown later.1 ev has the relatively long mean life of 6.60 X 1012 " Level widths are usually given in ev.1631 CROSS SECI'IONS FOR NUCLEAR REACTIONS 457 The concept of level widths is useful because values of these widths can often be obtained from measurements of resonances. it is necessary to have a quantitative measure of the probability of a given nuclear reaction. as mentioned earlier in this section. and these probabilities also yield information about nuclear structure. the lifetime of that level can be as long as 10l5 sec or lo" sec. The term "cross sectionn has already been met in connection with the attenuation of a %ray beam. When the total width is known.06 x I' (ev) x 1. The dependence of the width and spacing on the mass number and the excitation energy of the nucleus provides a test of theories of nuclear reactions and nuclear models. the cross section was regarded there as a form of absorption coefficient. while a sharp level with a I 'of 0. The partial widths of a level of the compound nucleus give the relative probabilities for different modes of disintegration. the level spacing D. The level width and level spacing are useful not only for characterizing compound nuclei. To study nuclear reactions in detail. The quantity that is most often used for this purpose is the moss section of a nucleus for a particular reaction. and the lifetime is 6. This quantity must be one which can be measured experimentally and calculated in such a way that the theoretical and experimental values can be compared readily. I' may be of the order of 10' ev. but for any excited nuclear states. as in the case of light nuclei. In the discussion of nuclear re ' : . The study of resonance phenomena in nuclear reactions can now be seen to have great importance because it provides information about the width and spacing of the energy levels of the compound nucleus. the value of the mean lifetime is given by h (ergsec) 7 (sec) = 2 1 7 I '(erg)  1. when it was used as a m e a s u r e of the probability that a photon is removed from the beam.
I R cm and the incident particles as point projectiles. . is equal to Nt. Equation (1610)then shows that the cross section for collision is the number of collisions per unit volume per second for unit incident flux and unit nuclear density.. a more detailed consideration of the idea of cross sections will prove useful. The concept of a nuclear cross section can be most easily visualized as the crosssectional area. 162.of each nucleus is given by o = uR2 cm2. (Hughes(12)). 162). nvNt (1610) The quantity nv. or target area. (169) A particle which passes normally through a thin aheet of material of area A containing NT nuclei has a probability N . the number of particles pansing through the sheet is nv per square centimeter per second. A the collision cross section is then .  u= collisions per square centimeter per second. that N w / A is small ( F i g . where N is the number of nuclei per cubic centimeter and t is the thickness of the sheet. or cross section u. p / A of colliding with a nucleus provided that there is no "overlappingn of the nuclei. For an incident beam containing n particles per cubic centimeter. is called the p a r t ~ j r u z .m N t . moving with velocity v. presented by a nucleus f the nuclei are considered as spheres of radius t o an incident particle.e.NUCLEAR REACTIONS nuclei = NT Nuclear density = N Number of FIG. and the collision rate can be expressed as collisions per square centimeter per second = nv NTu . The quantity N T / A . i. which is the number of particles in the incident beam crossing one square centimeter of area each second. then the target area. Cross eections for nuclear reactions as geometrical areas actions. which is the number of nuclei per square centimeter or the surface density of nuclei.
absorptions.the total cross section can be found by measuring the transmission of a particle beam in a "good geometry" experiment. per cubic centimeter per second for unit incident flux and unit nuclear density. and u(p. There is a direct analogy between ut and the total absorption cof square centimeter per atom.5~" X ~10I3 cm. There are also cross sections for individual reactions. The collision cross section. where A is the atomic weight. It is possible to speak of a cross section a. expressed in units o is the sum of a cross section for photoelectric absorption. and may consist of two parts. In the discussion of the theory of d e c a y . This product represents the cross section. the cross section for the particular process is then given by a relationship of the type of Eq. As in the case of Yradiation. contains the product Nui. and the term "collision" can be replaced by the appropriate term for any nuclear process. it was found that the radii of aemitting nuclei may be represented by the formula R=1. for the ith process. a. If it is assumed that this formula may be applied to all nuclei. for the absorption of a particular kind of particle..p). a cross section for Compton scattering. n). A rough idea of the magnitude of cross sections for nuclear reactions can be obtained from Eq. defined as the number of particles that are absorbed. ' ~ . watterings. or other processes) is counted. is usually called the total cross section. which corresponds to the effect of all possible processes. then u =T R = ~ ~ ( 1 . 5 )X~ ~ ~ cm2. a). In an experiment. I t is sometimes called the macroscopic cross section and denoted by xi= Nuj. (1610). and the quantities nu and N are measured.1631 CROSS SECTIONS FOR NUCLEAR REACTIONS 459 The above discussion can be applied to any nuclear reaction. say the ith kind. Thus. which efficient for %rays. the cross section for elastic scattering. particles can be removed by being absorbed in nuclear reactions or by being scattered out of the beam. su~h as @(a. u. the number of events of a given kind (collisions. o f all the atoms in a cubic centimeter of material. crt and is the sum of the cross sections for the individual reactions. and a cross section for pair formation. (169) for the geometrical cross section of a nucleus. the cross section for inelastic scattering. or disappear. u(a. may denote the cross section for scattering o f a given kind of particle. it has the unit cm'. The number of particles removed from the beam is obtained by adding the numbers removed by all the procesaes which can take place. and ui. The expression for the number of processes of a given kind. which will be shown to be a reasonably good approximation.
A given nucleus can have widely different cross sections for different nuclear reactions and the values represent the relative probabilities of those reactions. and to get a rough idea of its magnitude by calculating the geometrical cross section. and these quantities are not essential to the nuclear reaction itself. say A = 125. '  . In many c a s e s . This kind of information is needed in many practical problems.460 NUCLEAR REACTIONS For a nucleus of intermediate mass. or of the number of outgoing particles. (1610) and the s u b quent discussion. Nuclear cross sections are found to have values ranging from small fractions of a barn to hundreds of thousands of barns. it is possible to predict the number of reactions that will take place when a sample of the nuclide is exposed to a known flux of those particles for a given length of time. as expressed by Eq. and cross sections are a valuable form of nuclear data.  . The experimental meaning of the cross section comes from its use as a measure of the number of nuclear events which occur under a given set of experimental conditions. The relative intensities of different groups of product particles can also be measured by methods which are inde . and this unit is called the barn. Under certain special conditions scattering cross sections and geometrical cross sections can be related directly. Although it is convenient and simple to introduce the cross section as a target area. The elastic scattering of neutrons with energy greater than 10 Mev is a case in point. If the main interest is in the determination of the energies of the different groups of particles emitted. and these values often differ greatly from the geometrical cross section. * . The advantage of using the cross section lies in the fact that its value is independent of the flux of incident particles and the density of the material used for the target. without knowing exactly how many processes have taken place. The results of an experimental study of a nuclear reaction can be expressed in terms of the number of processes which take place under the conditions of the experiment or in terms of the cross section for the reaction. as in the manufacture of artificial radionuclides. this procedure must not be taken seriously. and the geometrical cross section of a nucleus is of the order of lo'' cm2. Cross sections for nuclear reactions are often expressed in terms of the unit cm2. and the measured values of the scattering cross sections are then used to determine the values of the nuclear radii. abbreviated as b.the results of the study of a nuclear reaction are expressed directly in terms of the number of processes that take place. it is enough to measure the energies by measuring the ranges or by magnetic deflection. When the cross section for a particular reaction between a given particle and a given nuclide is known. and will be discussed in Section 165E.
As the value of E departs from Eo. The concepts of cross section and level width can be applied to resonances in a quantitative way. In the particularly important csse of resonance processes. according to wave mechanics. and the cross section then has its greatest value. y). Eo is the energy a t the peak of the resonance. and on the charge and mass of the target nucleus. The first is a measure of the probability of forming a compound nucleus and. One of the problems of nuclear theory is the calculation of go(%)./r. . r. Then In general. E is the energy. the values of the cross sections and level widths depend on the energy of the incident particle. and the I"s are the widths already defined. According to the twostep view of nuclear reactions.1631 CROSS SECTIONS FOR NUCLEAR REACTIONS 461 pendent of the number of processes. r. the denominator increases in value. is proportional to h2.. (1611)  where a&) is the cross section for the formation of the compound nucleus in a collision between the particle x and the target nucleus X. The probability of the reaction X(x. where r. as can the energies of resonances in nuclear reactions. but the physical meaning of the formula can be discussed qualitatively. In its simplest form. a theoretical formula for the cross section was derived by Breit and Wigner. The second factor. The relative probability of the emission of the particle (or Yray) y is just r. u(x. Under these conditions. I is the mathematical expression for the resonance property and may be called the resonance factor. is the partial level width for y and r is the total level width. Equation (1613) may be regarded as containing three factors. and therefore of u(x. the formula is  where A is the de Broglie wavelength of the incident particle defined in Section 78. y) = uC(x) (relative probability of the emission of y). y)Y may be denoted by the cross section u(x.'g*'o' The rigorous derivation of the BreitWigner formula is a difficult problem. the BreitWigner formula gives the value of the cross section in the neighborhood of a single resonance level formed by an incident particle with zero angular momentum. The denominator of this factor has its smallest value when E = Eo. y).
he two points at which the cross section is u0/2. Its actual value in any particular case will. (1613) that the "full width a t halfmaximum. of course.462 NUCLEAR REACTIONS [CHAP. depend also on the other quantities in Eq. the values of the partial widths can be obtained by further experiments. The de Broglie wavelength of a particle is defined as where h is Planck's constant and mu is the momentum of the particle." that is. Eo. Since u(x. I t is found that the values of resonance cross sections vary over a wide range. It is convenient to express X in terms of the kinetic energy rather than the momentum and. An idea of the magnitudes of X and u may be obtained by considering the incident particle to be a neutron.87 X 10l2 cm. when the numerical values of h and m are inserted. ' I and r. . 16 and the resonance factor and cross section decrease.. is just the total width r. The values of uo and Eo can be deduced from the transmission curve.resonance energy. I t can be shown from Eq. while a 1Mev neutron has a wavelength of 2. the result is X (cm) = 2.87 x 10I cm. and it is typical of such applications that the wavelength of the incident particle appears in the formula.87 X am lo' A neutron with a kinetic energy of 10' ev has a wavelength of 2. or may sometimes be inferred with the aid of certain theoretical considerations. The total width can thus be determined directly from the actual observed width of the cross section resonance. (1613). y) (the cross section a t the peak of the resonance). and E is expressed in electron volts. Values of uo(x. The third factor is the probability for definite types of disintegration of the compound nucleus and is expressed by the partial widths . The BreitWigner formula is an example of the application of wave mechanics to nuclear physics. and r can be obtained from experiments in which the particle transmission through the target is measured ss a function of the energy of the incident particles in the neighborhood of the . since E = &m2. y) is proportional to X2. from fractions of a barn to the order of lo6 barns. its value should be of an order of magnitude similar to that obtained from the geometrical estimate in this section. the energy interval between t. When the total width is known.
tion of nuclear reactions serves to separate them according to their general character. General considerations. of order unity. This classifica. and the probability of forming a compound nucleus is relatively . and their energy distribution. Although the distinctions between the different groups are not clearcut. and will be followed during the remainder of this chapter. can penetrate relatively easily into a nucleus. Very high energies: 10 Mev < E < 50 Mev. I t also takes into account the fact that when two or more resonances are close together. C. The target nuclei may be divided into three groups according to the values of the mass number A : A. Heavy nuclei: 80 < A < 250. because of its lack of charge. and the atomic weight of the target nucleus. it is necessary to classify the reactions in some consistent scheme. Blatt and Weisskopf"" have shown that nuclear reactions can be classified into different groups characterized by the nature of the incident particle. The range of energy E of the incident particle may be divided roughly into five regions: I. its energy. Intermediate energies: 1 kev < E < 500 kev. Light nuclei: 1 < A < 25. as well as other properties. Ultrahigh energies: E > 50 Mev. 111. V.1641 EXPERIMENTAL RESULTS. 11. 164 Experimental results. It contains an additional factor. reactions initiated by Yrays will also be treated. the particles emitted. they are sufficiently marked to make the experimental results fall into definite patterns. The incident particles to be considered include neutrons. different characteristics from those initiated by charged particles. The neutron. Low energies: 0 < E < 1000 ev. GENERAL CONSIDERATIONS 463 The rigorous form of the BreitWigner formula is considerably more complicated than Eq. which depends on the angular momenta of the particles involved in the reaction. the onelevel formula is adequate. B. It is not surprising that reactions initiated by neutrons have. and deuterons.5 Mev < E < 10 Mev. aparticles. High energies: 0. for many practical purposes. Their classification and general treatment have proved successful. protons. Fortunately. 1V. (1613). Intermediate nuclei: 25 < A < 80. To discuss the experimental information about nuclear reactions in an orderly way. in general. the simple "onelevel formula" (1613) is no longer correct because the resonances may "overlapJ1and affect each other.
(Reprinted by permission from Themslical Nucluar Physics by J. and the energy mnga (row& The order of symbols in w h p u p corrssponds r o d y to the order of the yields of the corresponding reactions. The skstic scattering of charged particlea in omitted. F.Ructiom whom yield is ususlly lem than about lo' of the lending one an omitted Abbreviations: el = elantic. inel = inekstic. since it o c c m only with 8 few of the hwviest elements Hwv P n (el) 7 Incident puticle Energy of incident particle I. Fiaaion i n . since it cuurot e d l y be ssprnted from the nonnuclear Coulomb wttaring. Inc.610 Mev n (el) n (inel) a P (red for lower energim) IV. the type of incident puticle (columns). Intermediate: 1400 Lev v (real 1 1 1 . High: 0. . Very high: 1 0 4 0 Mev (inel) (ma for lower energies) an 2n n P (inel) np 2p P an pn ~ I I 2n n (inel) n( 4 P "P 2p a n (ind) n (el) P pn a an an a0 n p (inel) np 2p ap a n P SP an zp P P t h m or more particlea three or mom particlea n P np 2p a (inel) three or more puticles d (inel) triton8 t h m or mom partiela a three or mom puticlea three or mom particlea (inel) three or mom puticla nP 3n d (ind) tritonr tlum or mom puticled . Low: 01 kev (m) I n (el) n f I: P (ma) n P a (el) (=) 1 1 .) Thie table lids the nucleer reactions occurring in each group. John Wiley and Sons.Weiaeko~f. The abbreviation (red) refera to d reactioru in the box. The symbolslisted refer to the emergiw particle in a reaction c h t e r l e d by the type of taget. rea = mmnoes. M.1952.bo omitted.Bhtt and V.
charged particles with energies less than 500 kev (corresponding to the energy groups I and 11) cannot penetrate to the nucleus in groups B and C. The character and yield of most of the reactions seem to depend strongly on the structure of the individual nucleus.165) NEUTRONINDUCED REACTIONS 465 high. etc. they cannot be fitted into any general scheme. The reactions in each group are listed in the order of their importance. In energy region I11 charged particle reactions become important. and separation energies and Qvalues differ by much larger amounts among the light elements than within any other group. Reactions with light nuclei are not listed because. Consequently. or they must have enough energy to go over the top of the barrier. For example. they must either penetrate the Coulomb barrier. as discussed in Chapter 13. Finally. (p. In region I (low energy) resonance capture of neutrons by heavy nuclei predominates. it is a usefu! one.(l0' the different reaction types are arranged according to the groups in which they appear.]. In general. The reactions with intermediate or heavy nuclei are of different character in the different energy regions. in region V. and many more neutron reactions have been observed than charged particle reactions. a neutron can cause a reaction when it has a very low kinetic energy. the energy of the incident particle is so high that many nucleons can be ejected by a single incident particle. The light nuclei (group A) are characterized by the fact that it is extremely difficult to describe nuclear reactions in that group in terms of general rules. In Table 163. Zn). while charged particles need much higher energies. Experimental methods have been devel . although there are exceptions to this scheme. while in energy region I1 the most important reaction is the resonance elastic scattering of neutrons. the residual nucleus can be left in several excited states and the emitted particles can form groups with different energies. taken from Blatt and Weisskopf. In the following sections of this chapter. and the theory of the compound nucleus breaks down. the reaction which has the larger cross section precedes the one with lower yield and. Reactions induced by neutrons will be discussed first because of their great importance in both "pure" and "applied" nuclear physics. 165 Neutroninduced reactions. the different groups of reactions will be discussed in a general way and illustrated by specific examples. np). Positively charged particles must overcome the repulsive Coulomb forces exerted by the positively charged nucleus. In region IV the energy is high enough for reactions in which more than one particle is emitted for each incident particle [(p. The latter can be found in the literature references. The emphasis will be on the use of nuclear reactions as sources of information about nuclei rather than on the experimental techniques or the theoretical details. and neutron reactions are practically the only ones observed. as indicated above.
A. These methods form a highly specialized branch of nuclear physics and have been treated in detail by Hughes. and radiative Y). The most import a i t reactions between neutrons of low or intermebte energy and nuclei of intermediate atomic weight a ~ e l a s t i c scattering (n. it will simply be stated that beams of monoenergetic neutrons can be obtained at energies ranging from small fractions (lo') of an electron volt t o many millions of electron volts. intermediate nuclei. Low and intennediate energy. 16 Neutron energy (kev) F I ~ 163. n). for detecting them. For the purposes of the present chapter.(1z' they will be discussed in Chap ter 18. which is devoted to the subject of neutron physics. The total cross section ut and the .466 NUCLEAR REACTIONS [CHAP. and that the cross sections or yields of the different reactions induced by these neutrons can be measured by means of appropriate devices. oped for obtaining neutrons of different energies. . and for measuring the cross sections of the Merent neutroninduced reactions. esting example of these reactions. The bombardment of A12' by neutrons provides an intercapture (n. Total and capture cross sections of aluminum for neutrons in the intermediate energy region (Henkel and Barschall(13)).
Some of the results(13' are shown in Fig. the total cross section is practically equal to the elastic scattering cross section. or three orders of magnitude smaller. The experimental results show that the neutron widths are in the range 520 kev. At neutron energies where there is no resonance. while the capture cross sections are of the order of lo'' cm2. this cross section represents the probability that the neutrons are scattered without the formation of a compound nucleus. and it has been shown that for neutrons with energies up to about 1 Mev the scattering cross section is u. The scattering cross section has an appreciable value between resonances. because the force between the nucleus and the neutrons . since A1 has only one stable isotope. as distinct from reso : nance scattering. varies from 10 kev to 90 kev. and from the BreitWigner formula that the scattering or neutron width r. they must correspond to excited levels of A12'. This result would be expected according to the compound nucleus picture and serves as a confirmation of that picture. I t follows from the values of the cross section. as calculated from the difference mass of A12' + mass of neutron .8 Mev and. with an average value of about 40 kev. Y) have been measured as functions of the incident neutron energy. The capture cross section shows resonances a t the same energies as the total cross section.mass of A12'.1651 NEUTRONINDUCED REACTIONS 467 radiative capture cross section u(n. the nucleus acts like a hard sphere of radius R. 163. They are listed in Table 1 6 4 . together with the corresponding excited levels of the compound nucleus [A12'] computed as shown in Section 162. the distance between neighboring resonances. Over twenty resonances have been found for neutron energies in the range from 0.724 Mev. in which the total and radiative capture cross sections are plotted as functions of energy in the range from 10 to about 500 kev. is about 1000 times as great as the radiation width r. The values of the total cross section lie between 1 and 12 X cm2. so that the radiation widths are of the order of 10 ev.. being still of the order of cm2. = 4?rR2. In this case. Since the only reactions that have been observed when A12' is bombarded with intermediateenergy neutrons are elastic scattering and radiative capture. This type of scattering is called potential scattering. The observed level spacing for A12'. The capture cross section was obtained by making use of the fact that the product A12' is an electron emitter and its yield can be determined by measuring the induced radioactivity.009 Mev to 0. The total cross section was obtained from transmission measurements. The binding energy of the neutron in this nucleus is 7.
the thermal neutron energy region is in the neighborhood of 0.025 ev. it may be applied in the thermal energy region in the absence of a resonance if i t is assumed that the energy  . has a special property in the case of the (n.724 Excited level of energy in second column. r) reaction. Mev + can be described mathematically by a potential function analogous to the Coulomb potential in the scattering of aparticles by nuclei. Mev Energy to be added to the binding energy. One part of the low energy region I.NUCLEAR REACTIONS Neutron resonance energy. Although the BreitWigner formula is strictly valid only near a resonance. A beam of thermal neutrons can be obtained from a neutronproducing apparatus in which the neutrons are slowed down in some medium and emerge with energies in equilibrium with those of the thermal motion of the atoms of the medium. Mev [A128]= 7. At room temperature. the socalled thermal energy region. The potential which describes the neutron scattering is not electrical in nature. but is a characteristically nuclear potential.
with a few exceptions. and the neutron width r. and the curve is very different in character from the curve for Ag1Og.ance. which shows the variation of the total cross section of AglOg in the energy range from one ev to about 5000 ev. There are some . 164. which is proportional to the neutron velocity v. Lowenergy.r) = .1651 NEUTRONINDUCED REACTIONS 469 Eo is the resonance energy nearest to the thermal region. a t a neutron energy of 0. no resonances have been found. The formula for the (n. The total cross section"" of an exceptional nucleus BiZoglbetween one ev and 10.24 X lo" cm2. When the positive charge of the nucleus is large. which is proportional to l/v. I t turns out that all of the factors in the BreitWigner formula are constant compared with r. These nuclides are now thought to have a special structure interpreted in terms of the shell model of nuclear structure. 7)reaction then reduces to u(n.. heavy nuclei. In the thermal energy region. especially lead and bismuth. In heavy elements such as silver. the total cross section is approximately equal to the (potential) scattering cross section 47rR2. which will be discussed in the next chapter."~' . the energy resolution which can be achieved is shown by the triangles.important exceptions to this rule.025 ev. elastic scattering and radiative capture. Y) cross section is 0. the value of the (n. B. the effect of the Coulomb barrier prohibiting the emission of charged particles of low energy is even greater than with intermediate nuclei. and is much greater than the values in the resonance region. Figure 166 shows a lowthe solid line was calculated : energy capture resonance in ~admium. the resonances are mainly capture rather than scattering resonances."" At the higher energies. the spacing between resonances is very large (many kev). Neutron capture resonances have been analyzed extensively. and have been used to test the BreitWigner formula. the radiative capture of neutrons by aluminum is described by the "l/v law". 165. The level distances are often of the order of 10 to 100 ev and the excited states of the compound nuclei in this region are usually close together. 7)near reson. between resonances.. is small compared with the radiation width r. An example of a heavy nucleus with resonances very close together is given in Fig. is shown in Fig. and A.1 constant v and the cross section for radiative capture is inversely proportional to the neutron velocity.000 ev. The total cross section is practically equal to the capture cross section a(n. The only reactions possible for lowenergy neutrons on heavy nuclei are. In certain nuclides. the resonances cannot be resolved well because of experimental difficulties. only two widely spaced resonances are seen. and in others. the cross sections often show resonances very close together in energy. In reactions between lowenergy neutrons and heavy nuclei.
NUCLEAR REACTIONS .
NEUTRONINDUCED REACTIONS .
NUCLEAR REACTION8 .
as in Fig. values of the level spacing can be obtained. At thermal energies. but if the cross section is plotted against the number of neutrons in the nucleus. 168. = 7200 b. Individual resonances cannot be resolved and the experiments give information about cross sections averaged over many resonances. this energy lies in region 1 1 1 (high energy) rather than in region I1 (intermediate). and 126.1651 NEUTRONINDUCED REACTIONS 473  from the onelevel formula (1613) with the constants a. The radiative capture cross section has been measured for a large number of nuclides by Hughes(16' for 1Mev neutrons. and the circles represent the experimental points. The radiation widt. is smaller than for intermediate nuclei. As the mass number increases. a(n. Resonance scattering is more important than resonance capture and r. heavy nuclei. In the scheme of Section 164. Eo = 0. C. and if r. 82. is measured. r ) reactions a t neutron energies of about 1 Mev. The agreement between the formula and the experimental results is excellent and provides a quantitative proof of the validity of the formula. the cross section does not seem to behave regularly. but it is more convenient to discuss them here. In the intermediateenergy region.h r. it is found that the averaged (n. These reactions will be discussed here because they are a borderline or overlapping case which could be discussed in either place. At higher values of the mass number. > r. (1618). it is evident that the irregularities are caused by sharp drops in the value of the cross section that occur a t neutron numbers of 50. The spacing between resonances. decreases quite rapidly as the number of particles in the nucleus increases. It is hard to explore the intermediateenergy region because of experimental difficulties with neutrons of these energies. The measured values of the capture cross section are plotted against the mass number of the target nucleus in Fig. r ) follows the l / v law. When the BreitWigner formula (1613) is averaged over many resonance levels and applied to (n. The result is that the level spacing Dl from Eq.115 ev. is calculated theoretically. t. the capture cross section increases until A is about 100. but less rapidly than the cross section increases. r) cross section is given by  where D is the level spacing. however. and r = 0. both of which .he cross sections are given in units of millibarns (thousandths of a barn). as was the case with intermediate nuclei. Intermediateenergy. If ~ ( n7) . 167. according to theory. These exceptionally low values of the cross section correspond to unusually large level spacings snch as those mentioned previously for lead and bismuth.. in the absence of resonances.176 ev. the reactions are similar in nature to those with intermediate nuclei. decreases. These cross sections in turn yield information about the average level spacing.
167. Neutron capture cross sections a t an effective energy of 1 Mev (Hughes(12)). .168. . . 120 1 FIG. . . 40 . 50 . 30 . . 0 FIG. . I 0 I 20 . 60 70 80 90 1 0 0 1 1 0 Number of neuhom . .NUCLEAR REACI'IONB loOD 8  IQO dmbl 1 . Neutron capture cross sections a t 1 Mev as a function of neutron number (Hughes(12)).
or 126 neutrons have much greater level spacings than neighboring nuclides.because of the effect of the Coulomb barrier. and only a relatively small amount of information has been obt. the total cross values of the nuclear r a d i ~ s . The values of the level spacing obtained in this way vary from about 5 X lo4 ev for Na23to about 1 ev for the ordinary heavy nuclides. the other with 25Mev neutrons. Very high energy. The emitted neutrons can leave the residual nucleus in excited states so that some of the scattering is inelastic rather than elastic. The emission of neutrons is more probable than that of charged particles because of the Coulomb barrier.1651  NEUTRONINDUCED REACTIONS 475 " have 126 neutrons. and the values of nuclear radii conform with the formula (1620) . The quantity Ro is defined by the relationship R = R~A'I~ x 10l3 cm. one with 14Mev neutrons. I and should be a constant if the nuclear radius is proportional to the cube root of the nuclear mass. The excited states. D. When the energy of the incident neutron is about 1 Mev or higher. heavy and intermediate nuclei. 82. new types of processes can occur. and information about them can be obtained from the energies of the '+rays and from the energy distribution of the scattered neutrons. the values of R were obtained from the rigorous formula. and the value of the cross section given by Eq. E. decay by remission. " ~ . and it will be seen in the next chapter that these neutron numbers seem to be associated with particular nuclear structures. The determination of inelastic neutron scattering cross sections is a difficult experimental problem. (1619) is an asymptotic one which is reached only for extremely high energies. Highenergy. The scattering of neutrons with energies greater than 10 Mev has been used to determine ' to theory. especially when the target nucleus has a high value of 2. are listed in Table 165. ' ~According section at high energies approaches the value   so that if at is determined by transmission measurements. heavy and intmnedicrte nuclei. which are bound states. The cross sections for the reactions in which charged particles are emitted are much smaller than those for inelastic scattering and radiative capture. R can be obtained.ained in this way. The rigorous relationship between ut and R is more complicated than the last equation. The results of two sets of experiments. The results show that Ro is indeed practically constant. The nuclides with 50.(lg' The formula for at actually includes a term which depends on the energy of the neutrons. such as inelastic scattering and the emission of charged particles.
5 8.05 5.85 1 .8 6 .60 1 .35 3. 10l3 cm Be B C 0 A 1 S C1 Fe Cu Zn Se Ag Cd Sn Sb Au HI3 Pb Bi 2 .25 4. These forces should act differently on a neutron approaching the nucleus and an aparticle leaving it. 16 Element Energy.4 and agrees reasonably well with the value 1.8 7. 5 8 1 . 9 6. The fact that the values obtained by different indirect methods for the dimensions of atomic nuclei agree as well as they do represents a triumph for nuclear theory. The quantity R.53 obtained from adecay data in Section 136. 2 7 . R. Mev 14 14 25 25 14 25 14 25 14 25 14 14 14 25 14 14 14 14 14 25 14 14 Observed at.70 4.4 3 . is the distance a t which nuclear forces begin to act.3 6. 9 7 . 1 6 1. The two results s h ~ u l dnot be expected to be the same because they are not the results of direct measurements of the nuclear radius.03 3. 5 5 . 8 8 2. l o * ' cm2 0. 3 4. Values of nuclear radii have also been obtained recently from studies of the scattering of neutrons with energies in the neighborhood of 1 kev.3 8. as it appears in the theory.64 5.75 2.29 1.52 4.92 1.9 over the entire range of the elements. 4 3.82 3.25 5. The average value of Ro is close to 1.476 NUCLEAR REACTIONS [CHAP. 8 4 . 4 7 . and it is not surprising that different values are obtained for nuclear radii from the analysis of different kinds of experiments.6 4 .'*O'   .65 1 .68 5.50 3. 1 4 . with the possible exception of the very lightest ones.17 Nuclear radius.25 4.4 7. 7 5 .6 4. 3 7 . 6 5 .
protons and aparticles may be scattered elastically or inelastically. or they may cause the emission of a charged particle. . the emission of neutrons is more likely than that of charged particles. 3n). the incident particle has enough energy to cause the emission of more than one particle from the compound nucleus. This result is expected because. with Ro = 1. When the energy of the incident particles is high enough.Si28.1661  REACTIONS INDUCED BY PROTONS AND ALPH. 2n) reaction has been studied in a number of cases. the cross section increases with energy.~~ energy ' range from 1. a t every resonance obtained with the (p. The reaction A12'(p. About 100 resonances have been found a t proton energies between 0.. reaction. an aluminum target was bombarded with protons accelerated in a Van de Graaff electrostatic generator. (n. 2n). _  + . A large number of very sharp resonances has been found. 7). 169.'23' some of which are shown in Fig.'z1' and its threshold is usually in the neighborhood of 10 Mev. H i g h energy. according to the compound nucleus theory. is 11. In these experiments. a ) reactions. a)Mg24 has also been studied and more than 30 resonances have been o b s e r ~ e d ' ~in ~ the .6 Mev. reactions.35.'~ The ~ ' binding energy of the proton as obtained from the mass difference. the Coulomb barrier prevents any appreciable interaction between positively charged particles and nuclei. These resonances also give information about the excited levels of [Si2'] and should occur a t the same energies as those of the (p.7 and 15. In the very high energy region. The sharpness of the resonances (the low value of the total width) shows that the lifetimes of the excited states are relatively long. 166 Reactions induced by protons and alphaparticles. The cross sections of nuclear reactions induced by protons or aparticles are extremely small for particle energies below 0. there is an elastic scattering resonance. The elastic scattering of protons by aluminum also shows resonances and.'22' reaching a maximum somewhere near 20 Mev. the first one studied in detail. with few exceptions. Again. and the %ray yield was measured as a function of the proton energy.2 Mev and 4M~V.4PARTICLES 477 The interpretation of the experimental results is more complicated than a t the higher energies. AlZ7 p . y)Si2' yields information about the excited levels of the compound nucleus [Siz8]. may be expected. np). Above the threshold. and (n. but the results are similar. The (n. and then decreases with increasing energy of the incident neutrons. and the excited levels lie between 11. Y) or (p. At low or intermediate energies.4 Mev to 4 Mev. The properties of reactions induced by protons and aparticles can be illustrated by the reactions of protons with aluminum. because of the Coulomb barrier. The (a. (n. they may undergo radiative capture. etc.59 Mev. has already been discussed in Section 161. The reaction Alz7(p. p) reaction. A.1 Mev except for a few cases among the very light nuclei.
16 ton energy 3 kev FIG. aa a function of the proton energy (Brostrom. Yield of the (p. Yield of the (p. n) reaction in C13'. Photon energy (kev) FIG. 169. ( ~ ~ ) ) . Y) reaction in aluminum. as a function of the proton energy (Schoenfeld et . et ~ d .478 NUCLEAR REACTIONS [CHAP. 1610.
In this way. A sample of the experimental results in the range from 1. Similarly.95 Mev above the ground state.'26' They are bound levels and decay by %emission.he threshold to still higher values. Between this energy and 2. The method generally used for studying the inelastic scattering of protons is to bombard the scatterer with a collimated beam of monoenergetic protons. the neutronproton mass difference. to determine the energy distribution of the protons scattered through a known angle. the threshold proton energy for neutron emission is always greater than 782 kev. The scattered protons usually appear in several groups with different energies. the relative importance of neutron emission increases with increasing energy of the bombarding particle. protons were usually accelerated in an electrostatic generator and the neutron yield was measured. When the energy of the incident protons or aparticles is high enough. Below the threshold for neutron emission.84 Mev and 5.'29' corresponding . In these experiments.80 Mev to 1.64 Mev. and the emission of charged particles are important. Threshold energies have been detem~ined'~'' for a number of (p. n) reactions among the light and intermediate nuclides. A127 has been found to have more than 20 excited levels a t energies between 0. the threshold for the reaction C137(p. The inelastic scattering of protons by aluminum shows resonances and gives information about excited states of the compound nucleus [Si2']. neutron emission becomes the dominant reaction. n) reactions are threshold a neutron may be emitted. and then.51 Mev over 100 resonances have been found'2s' corresponding to excited states of the compound nucleus [A3']. if the target nucleus is stable. and the corresponding nuclear energy levels can be obtained from the equations of conservation of energy and momentum. Above the threshold.n)A3' is at 1. The resonances were resolved well enough so that an average separation between levels of about 5 kev was observed. it also provides an important method for determining the level spectrum of the turget nucleus A12'. This reaction may be described by the equations where the asterisk denotes an excited state of A127. by means of range or magnetic deflection measurements.1661 REACTIONS INDUCED BY PROTONS AND ALPHAPARTICLES 479  any state of the compound nucleus formed by proton capture must be capable of decay by emission of a proton with the full energy available.94 Mev is shown in Fig. scattering. In the case of C13'. n) and (a. The (p. reactions. n)Mn531 261 resonances have been observed. radiative capture. One effect of the Coulomb barrier is to increase t. and when the compound nucleus has enough excitation energy to emit a neutron o f 1 Mev or more. 1610. in the reaction Cr53(p.
three neutrons may be emitted. it is very difficult to resolve the resonances. 2n) or (a. n) reactions. These results are an indication of the complexity of nuclear spectra and of the great difficulty of accounting in detail for these spectra on the basis of a theory of nuclear structure. n) reactions on a series of nuclides.'30' These values of Ro are in good agreement with the values obtained from adecay and neutron scattering so that the application of the approximate theory of Feshbach and Weisskopf to (p. for protor.9 kev. When bismuth . more than one particle can be emitted. as in the heavierintermediate and heavy nuclei. n) and @(a. B. in nearly every case. The circles represent experimental points for the reactions Aglo7(p. the detailed resonance structure is not seen. 1611 are obtained. If the latter are denoted by uc(p) and uc(a). n)Cdlo7 and AglOQ(p. 2n) reactions. among others.480 NUCLEAR REACTIONS [CHAP. the most common reactions of this type are the (p.5 was used.3and 1.n) averaged over many resonances. the latter depend on the value chosen for the nuclear radius. n) and (a. and experimental measurements give values of u(p. One method of avoiding this problem is to develop an approximate theory which does not attempt to reproduce individual resonances but simply tries to describe the general trend of the magnitude and energy dependence of nuclear cross sections on the basis of a very rough picture of nuclear structure.47 Mev.5. but the residual nucleus may be left in a sufficiently excited state to emit a second neutron. This agreement is typical of results obtained for (p. the emission of a neutron is the dominant reaction.and the dashed curves are theoretical results. with an average spacing of 3. A schematic theory of this kind has been proposed by Feshbach and W e i s s k ~ ~ f "and ~ ' has been applied to the (p.n)CdlOQ. n) and (a. 16 to excited levels of the compound nucleus [Mn5']. then When the energy levels are very close together. n) reactions seems to be successful.42 Mev and 2. energies between 1. If the resolution is sufficiently poor. Very high energy. Good agreement was obtained when the formula (1620) was used for the radius with Ro = 1. When the energy of the incident particle is high enough. At still higher energies. The approximate theory can then be applied to the calculation of the cross sections averaged over many resonances. When the energy of the incident particles is greater than 10 Mev. and the other reactions are negligible in comparison. and curves such as those in Fig. giving the values for the cross sections for the formation of the compound nuclei. The compound nucleus decays most often by emitting a neutron. the experimental results could be accounted for by the theory provided that a value of Ro between 1. n) and (a.
n)Cdlo7 and AglOg(p..FIG. 3n) reaction is first observed. The cross section for a reaction of the type A a + [C*] + B b may be written + + : where ua(c) is the cross section for the absorption of the particle a with kinetic energy e by the target nucleus A to form a particular state [C*] of the compound nucleus. for convenience. the (a. Y ~ ( Eis ) the probability of disintegration of [C*] into the final nucleus B and the particle b. ' ~ ~ )  is bombarded by arparticles. (1612). The circles are experimental points and the dashed curves are the theoretical results of S h a p i r ~ . Crow sections for the reactions Ag107(p. where j 3 . n)CdlOg. As the energy is increased above 30 Mev. and reaches its maximum value of about one barn a t about 30 Mev. G h ~ s h a l ' ~ has ~ ' made a set of measurements which serves as an interesting test of the theory of the compound nucleus./r may be replaced. while ~ ( a3n) . 1611. 2n) reaction is first observed when the aparkicle energy is about 20 Mev. increases to I f barns.'31' the (a. For a given excited state of a compound nucleus r is a constant and the ratio r. The excitation energy E of the compound state [C*] is E = e pa. and reaches a value of onequarter of a barn at about 40 Mev. The idea underlying the experiment can be developed from Eq. is the binding energy + . 2n) decreases. u(a. At this energy. The cross section increases with increasing energy. In this energy region. by 7.
for Cue3. Eqs. but by the reaction between a different initial particle and target nucleus. If the compound nucleus [C*] is formed in the same state of excitation. (1625). (1625) provides a direct test of the validity of Bohr's theory of the compound nucleus. Since the excitation energy is assumed to be the same in the two cases. but yb(E) is the same in the two cases because of Bohr's basic assumption that the mode of decay of the compound nucleus [C*] is independent of the way in which it is formed. I t can be seen from the  . The energy (mass) difference between the two target nuclei is such that a proton energy e and an aparticle energy e' = e 7 Mev produce the compound nucleus [Zne4]in the same state of excitation. with the ratios @(a. According to the analysis leading to Eq. Ghoshal measured the cross sections of the reactions  + . t h e ratios ~ ( an): u(a. where the six cross sections are plotted as functions of the kinetic energy of the aparticles and protons. 1612. Ghoshal bombarded Cue3 with protons and NieOwith aparticles. If [C*] disintegrates in a different way to give the particle d and the residual nucleus D. but the binding energies of the incident particles are different. . e' is different from e. (1621) and (1622) are replaced by and I t follows from the last four equations that  An experimental test of Eq. producing the compound nucleus [Zne4] in both cases.482 NUCLEAR REACI'IONB [CHAP. + + where e' is the kinetic energy of the incident particle a'. 16 of the particle a to the target nucleus A. The proton energy scale has been shifted by 7 Mev with respect to the aparticle energy scale in order to make the peaks of the proton curves correspond with those of the aparticle curves. within the limits of experimental errors. :~ ( apn) .n) :~ ( a2n) . then the cross section for the a' + [C*] + B b is reaction A' . The experimental results are shown in Fig. pn) for Ni60 should agree. 2n): u(a.
3. it is directly related to the transition probability.l . (a. 1. and the efficiency of the detector. 2n). and the decay can be described in terms of appropriate multipole radiation. and is determined from the particle beam current. . pn) reactions on Cus3 and for the (a. 2. A cross section for the production of the Yray can be measured. the latter is subjected to a rapidly varying electric field. figure that the ratios agree quite well. 1612. (p. n).'33o34) When a charged particle (aparticle or proton) with a few Mev of energy moves past a target nucleus. C.7 . providing evidence for the validity of the compound nucleus hypothesis. pn) reactions on NisO as functions of the proton and aparticle energies. This field may give rise to transitions from the ground state to excited states of the target nucleus. n).  : . Energy of protons (htev) . respectively ( G h o ~ h a l ( ~ ~ ) ) . Experimental cross sections for the (p. 2n). . there can be inelastic scattering and excitation of the nucleus because of the Coulomb forces.3. A reaction different in nature from those discussed so far can occur when a charged particle interacts with a nucleus without specific nuclear forces coming into play. . Coulomb Excitation.. 2. An excited state produced in this way then decays by emission of Yrays or conversion electrons. and (p. 1. 12 16 20 24 28 32 Energy of a (Mev) 36 40 FIG. the thickness of the target. . and (a.
a useful tool for the study of lowlying states.484 NUCLEAR REACTIONS [CHAP. 16 FIG. Lowlying energy states of Inuh showing data obtained by Coulomb excitation. The existence of sharply defined resonances (in agreement with . The Coulomb excitation reaction is. as in collisions of protons or aparticles of intermediate energy with intermediate or heavy nuclei. Hollander. and can be observed only when the latter are extremely improbable. [Strominger. new information has been obtained for previously known levels.1613. Because of the large Coulomb forces. Revs.] The cross section for electric excitation is small (about or em2) compared with those for direct nuclear transformations. and the w a y s emitted in this case are of the E2 type. the classical distance of closest approach (cf. 167 Limitations of the compound nucleus theory. Mod: Phys. and Seaborg.'35' The nuclear states most strongly produced in thew reactions are the lowlying states (1 Mev or lower) induced by the electric quadruple field of the incident particles. Coulomb excitation has been studied in many nuclei. 585 (1958). therefore. 30. The information concerning nuclear reactions discussed in Chapter 11 and so far in this chapter is qualitatively in harmony with the theory of the compound nucleus. and many new levels have been identified. 1613. A decay scheme including data obtained by means of Coulomb excitation is shown in Fig. Chapter 3) is then large compared to the nuclear radius and the probability of direct nuclear reaction is very small. the arrows pointing upward represent the transitions excited in the reaction. and the theory of the process has been developed extensively.
Eq. however. independence is related to the sharing of the excitation energy by all the particles in the compound nucleus. Certain experimental results have also led to a further examination of the assump . reactions for which the independence hypothesis seems to break down in that the decay probabilities of the compound nucleus depend upon the way in which the compound nucleus is formed. But. discussed in the last section. proton emission was found to be more probable than neutron emission when the reaction was initiated by protons. Ghoshal's ex~eriments. At these higher energies the independence hypothesis is hard to test quantitatively. (166)] may be large enough to overlap several states. when the energy of the incident particle is high. There are. form one of a few such tests. The uncertainty [cf. a comparison has been made'36) of the relative probabilities of emission of protons and of neutrons from compound nuclei formed either with protons or with neutrons. in a direct and general way. because it is not easy to reach the same range of excitation energy in the compound nucleus with different reactions.1671 LIMITATIONS OF THE COMPOUND NUCLEUS THEORY 485  ' 5 theory based on Bohr's ideas) shows that the assumption that the mode of disintegration of the compound nucleus is independent of the way in which the latter is formed is valid within a certain limited region of excitation energy. the direction of incidence or the position in the nucleus of the particle initiating the reaction should have no effect on the decay of the compound nucleus. There is also indirect evidence against the independence a s s ~ m ~ t i o n . several states of the compound nucleus may be excited. and the independence assumption requires further testing. ' ~ ' According to Bohr's picture of nuclear reactions. For example. The decay of these states may then depend on the mode of excitation. When nuclei with mass numbers in the range 4871 were bombarded with 21Mev protons and 14Mev neutrons. and their relative excitation may depend on how the nucleus is excited. the independence assumption is justified. When the energy of this incident particle is within the region of sharp and welldefined resonances and if the energy coincides with or is near to a resonance. But reaction products are sometimes emitted with much greater energy than would be expected if the energy of the incoming particle were shared among all the constituents of the nucleus. ~ h results k show that the independence &sumption seems to be valid for a compound nucleus with an excitation energy of 15 t o 40 Mev when produced by protons on CU" or by aparticles on NisO. and the properties of a given quantum state are independent of the way it is produced. and the reaction products often show an angular distribution with more of them in the direction of the incident particle than would be expected. The nuclear reaction then produces only one quantum state of the compound nucleus. In these cases the compound nucleus seems to have more memory of the initial process than is compatible with the independence hypothesis.
NUCLEAR REACI'IONS .
neutrons. The reaction A12'(d. Up to now. the discussion of deuteroninduced reactions will be limited to intermediate nuclei and the high energy region 111. the most important results from the standpoint of nuclear spectroscopy have been obtained with the light and lighterintermediate nuclides in the highenergy region. The proton tracks were counted under a microscope in small strips of plate material and the number of tracks in each strip was plotted against the product HT. The product particles. Nuclear emulsion plates were used for recording the charged particles. Because of the special nature of the light nuclei. in his original statement of the hypothesis. A sample of the results is shown in Fig. and proton. This treatment will be discussed in Chapter 17 after the subject of nuclear models has been introduced. and that some photoninduced reactions may occur through a mechanism other than the formation of a compound nucleus. protons.1 Mev. and it has been found that this assumption must be changed. I t will also be seen in later sections of this chapter that certain reactions initiated by deuterons are not adequately described in terms of Bohr's hypothesis. These impurities were known and their peaks were identified by comparison with the results obtained :  : .p)AlZ8 reaction are designated by capital letters. 168 Deuteroninduced reactions : intermediate nuclei. pointed out that his twostep analysis of nuclear reactions would not apply to light nuclei. Bohr. There are additional limitations of the compound nucleus hypothesis. the proton peaks that have been assigned to the AlZ7(d. The target was placed between the pole pieces of a 180' magnetic spectrograph. At excitation energies above the region of sharply defined resonances. Other peaks caused by the presence of impurities were also observed.'37' an aluminum target was bombarded with deuterons accelerated in an electrostatic generator. often appear in groups with different energies and are related to the lowlying excited levels of the product nucleus. or aparticles. it has therefore become necessary to introduce a modified picture of nuclear reactions. and charged particles leaving the target at 90' with respect to the incident beam were analyzed in the spectrograph. Reactions induced by deuterons have been a fruitful source of information about the energy levels of nuclei.where H is the magnetic field strength and T is the radius of the path of the protons in the magnetic field. In one experiment. and aparticle tracks were recognized by the track lengths. there are too few particles in such nuclei to make possible the large number of energy exchanges needed for the independence hypothesis to be valid.1681 DEUTERONINDUCED REACTIONS 487  '  tion that the compound nucleus is formed immediately after the incident particle enters the nucleus. p)A12' is an example of the way in which deuterons have been used in the study of nuclear levels. high energy. deuteron. 1614 for incident deuterons with an energy of 2.
p)A12' reaction discussed above. 100 proton groups were resolved with nearly all of the new groups having energies lower than those shown in Fig. in this case [A12']. and their energies measFIQ. These states decay by Yemission and are bound together with the virstates of AlZS. and the binding and excitation energies are practically identical. The ?rays resulting from the decay of excited states of A12' could not be studied under the conditions of the AlZ7(d. Fifty different proton groups were found corresponding to the ground state and 49 excited states of A12'. The new proton groups correspond to energy levels of A12' about 6 Mev or more above the ground state. In a later e~periment.'~'' 7. They have been obsewed in studies of the 7rays from excited states of A12' formed as a result of neutron capture in A127. About 30 7rays have been obsewed. The compound nucleus. its kinetic energy is negligible compared with its binding energy in the compound nucleus. they indicate how the spectrum of a lighterintermediate nucleus looks.0Mev deuterons were used as well as a magnetic spectrograph with greater resolution. When the captured neutron is a slow one.(37)).1615. . An energy level diagram of these states is shown in Fig. 1615. Energylevel diagram ured with a pair spectrometer. 1614.'39' for A12' (Enge el d. tual levels listed in Table 163.NUCLEAR REACTIONS with other target materials. decays by remission through excited states of the product nucleus A1281and the energies of the %rays give information about these excited states.
5.1681 DEUTERONINDUCED REACTIONS 489 Energy spectrum of A128 Aln(n.p)AlB 2.I. 4. 3. The lower diagram shows the states as deduced from the (dl p) studies and indicates the relative proton yields. The inelastic scattering of deuterons.07. Y) studies. The states inferred in this way are compared in Fig. gives information about the levels of the target .88 f 0. 5.0 . The levels given by the two methods are in good agreement.O . The levels of MgZ6obtained from the two reactions agree very well.n)A127. An example"" is the reaction MgZ6(d.07. X . 0 I 1.32 f 0. I I : .O 6 11 A127(d.1 40 Capture state : i . Energylevel diagram for A128 with relative intensities (Enge  : The most energetic ?ray had an energy of 7. These levels are in good agreement with levels found from the inelastic scattering of protons. Excited states of Mg25can also be obtained from the reaction Mg2"(d. and the maxima and minima in the relative intensities have the same locations in the two diagrams.07.07.65 f 0.72 Mev.07. The (dl a) reaction has been used in a similar way as.~)AIB(Kinsey) I (24) 11 . for example. and there can be little doubt that this ?ray represents the direct transition from the capture state to the ground state. and the height of the lines represents the relative intensities of the w a y s . The upper part of the diagram was constructed from (n. 1 . et d. I . : . 1616 with those given by the (dl p) reaction. which gives excited states of AlZ7 at 0. The (dl n) reaction often yields groups of neutrons whose energies lead to levels of the product nucleus.(37)).128. like that of protons. 1.1 Mev.75 f 0. in the reaction A12'(d. The differences in energy between this 7ray and those with less energy can be related to bound states of A.0 I . 1.0 II 20 Ed = I 3. p)Mg25 and the results of a study'400'of the two reactions are listed in Table 166.1616.07 Mev..07.92 f 0. which is just the binding energy of the captured neutron. 0 = 90' Excitation energy FIG. a)Mg25.81 f 0.33 f 0. and 5. 2.
3.97. Because of the .76 Mev.. Finally. and 5. According to the cbmpound nucleus theory. These processes result from the following properties of the deuteron: (a) the deuteron is a very loosely bound structure with a binding energy of only 2. . the reaction Sia(d. A) bombarded by deuterons should form the compound nucleus [C(Z 1. (b) its charge distribution is very unsymmetric.2 Mev.NUCLEAR REACTIONS *Excitation energy in Mev. '  + + The special properties of the deuteron make it possible for other processes to take place in addition to those of the scheme (1626). 2. A 2)]. nucleus. the center of mass and the centerof charge (the do not coincide as they do in the aparticle.17.74.39. The experimental error in this case was about 0. i. 4. ' in ~~ agreement ' with the two lowest levels obtained by the three other methods. One reason is that new types of processes occur with deuterons which cannot be accounted for on the basis of the compound nucleus picture.e.23 Mev as compared with 28. a)AlZ7gives levels a t 0. the levels agree with some of those obtained from the inelastic scattering of protons by All7. and within the limits of this error. a nucleus X(Z. which can then decay in different ways: .3 Mev for the aparticle and 7 or 8 Mev for the average particle in a nucleus. Levels of AlZ7have been found in this way"2' with excitation energies of 0. The theoretical treatment of deuteroninduced reactions is more dif5cult than the treatment of the reactions discussed earlier in this chapter.01 M ~ v . p) reaction on Mg26. and from the (d.84 and 1.
or (Y. I Coulomb field of the nucleus keeps the proton away and the neutron is f the second constitmore likely to enter the nucleus than the proton. neutron. The additional processes which can take place with deuterons make the cross sections for these reactions greater than those for reactions with other charged particles of similar energy.p). The nuclear interaction energies (average binding energy per nucleon) are much higher than the binding energy of the deuteron. and the constituent of the deuteron that arrives first a t the nuclear surface is quickly separated f the kinetic energy of the deuteron is not very high. The simplest example of this type of reaction is the photodisintegration of the deuteron. This reaction has a threshold a t a 7ray energy just equal to the binding energy of the deuteron. A if the second nucleon (proton) misses the nucleus. and the investigation of these reactions has been of special use in nuclear spectroscopy. or by decay. a (Y.1691 REACTIONS INDUCED BY GAMMARAYS 491 finite distance between the proton and neutron in the deuteron one of these particles may reach the nuclear surface before the other. and very careful determinations of the threshold energy have been made because of the great impor . The characteristics of stripping reactions are closely related to the properties of the state of the product nucleus. But nucleus [C(Z 1. (1627) X(z. I uent of the deuteron hits the nucleus an instant later. The disintegration of nuclei by ?rays occurs when a photon is absorbed whose energy hv is greater than the separation energy of a particle (proton.n). the process that takes place is I)] p. the reaction (1627) is called the "OppenheimerPhillips process" after the two workers who first described it in 1935. At relatively low energies. (Y. deuteron reactions are of great practical importance in the production of radioactive nuclides and in the study of nuclear reactions. When this condition is satisfied. As a result.a) reaction is observed.A f the proton penetrates the Coulomb other particles. A + + + + + I)] may decay by emission of some The compound nucleus [C1(Z. the term nuclear photoe&ct is sometimes used in analogy with the atomic photoelectric effect. or aparticle). A process of this type is often called a photodisintegration. A ) d + [C'(Z.'"*5g' 169 Reactions induced by gammarays. the compound 2)] is formed just as in the scheme (1626). I bamer and hits the target nucleus first. the from its partner. the process may be + The processes (1627) and (1628) are called stripping reactions because a nucleon is stripped out of the deuteron by the target nucleus.
the threshold for the photodisintegration of beryllium was also determined.002 Mev. the energy of the electrons is practically identical with that of the most energetic xrays. the binding energy of the last neutron lies between about 5 and 13 Mev. d) reaction or a (dl H3) reaction. The measurement of neutron binding energies is one method of establishing mass differences with great accuracy and also provides information about nuclear structure.003 Mev.'"" It was found that . The energy released on the addition of a neutron to a stable nucleus is given by the Yraysfrom the capture of slow neutrons (n. and if the electron energy is measured very carefully. The open circles give the neutron count. this quantity should be proportional to the energy above threshold. 1617. and the photoneutron threshold was obtained by extrapolating the neutron yield versus energy curve to zero yield. The binding energy of the deuteron was determined as 2. the one removed from a stable nucleus or added to a stable nucleus. The neutron whose binding energy is determined is the "lastn neutron.'"" With the exception of deuterium and beryllium. the electrons from a cathoderay "gunnwere accelerated in an electrostatic generator. According to theory. that of the xrays is known with high precision. The energy needed to remove a neutron from a stable nucleus is given by the threshold of the (7. The neutrons resulting from the reaction d r. Photoneutron thresholds have been measured for many (Y. I n a highly precise meas~rement"~'of the threshold. The reaction in this case is + + and the threshold energy gives the binding energy of the neutron in the Begnucleus as 1. or by the Qvalue of the groundstate group from a (p.492 NUCLEAR REACTIONS [CHAP.666 f 0. n) or (n. The experimental values of the binding energy have been tabulated and analyzed as a function of the number of neutrons in the nucleus.p n were counted as a function of the electron energy. In the neighborhood of 2 Mev.226 f 0. The results of the experiment are shown in Fig. chiefly to get neutron binding energies. The shape of the yield curve is typical for photodisintegration reactions a t energies just above the threshold. 7) or can be obtained from the Qvalue of the groundstate proton group from a (dl p) reaction. These reactions furnish neutron binding data for the target nucleus. In the experiment just described. the dots represent the square root of the difference between the total neutron count and the background count. 16 tame of the biding energy of the deuteron in the theory of the nucleus. that is.n) reactions'*'" over the entire range of elements. deuterium was bombarded with xrays from the action of electrons on a gold target. 2n) reaction. The straight lines show that this is the case and allow an accurate extrapolation to zero yield.
there is a break a t 82 neutrons and probably one a t 28 neutrons. 82. These results indicate that the nuclei with 28. A theory of nuclear structure which accounts for the particularly stable nature of  . the special nature of nuclei with 50. nuclei with 28. 50. 1617. Thus the binding energy of the 127th neutron is about 2. 82. similarly. 82.1691 REACTIONS INDUCED BY GAMMARAYS 493 Electron energy (Mev absolute) ho. The binding energy of the 50th neutron is about 2 Mev greater than that of the 51st neutron. and 126 neutrons have special properties. and 126 neutrons are markedly more stable than their neighbors. In Section 165C. The photoneutron yield from deuterium near threshold (Mobley and La~benstein(~~)). 50. and 126 neutrons was shown by the low values of their capture cross sections for 1Mev neutrons.2 Mev less than that of the 126th neutron. in each case there is a sharp break in the binding energy of the neutron added to the nucleus.
Because of the blocking effect of the Coulomb barrier. for example. n) and (7. protons are expected to be emitted much more rarely. The yield was determined as a function of 7ray energy by measuring the activity of the S C ' ~ produced. n) reactions in that the cross section versus energy curves have the same general shape.2n) and (7. the ~ determined ~ with ntensity of the 160kev %ray characteristic of S Cwas . In many cases. and reaches a value of nearly cm2 in bismuth. The (7. by recent experiment^'^^' on the reaction V5'(7. p) reactions. the relative yield of (7. p) reaction^'^^' show that they aresirnilar in nature to the (7. n) and (7.'50' It is supposed that the energy of the 7ray photon is absorbed by a single proton in the nucleus which is then emitted before the energy can be shared with the other nucleons. and (2) the decay of the compound nucleus by the emission of any one of several different particles. as is shown.p) cross sections have been interpreted in terms of a direct photodisintegration process.r a ~ sfor ' ~ a~ series ' of elements covering the periodic table. (7. 16 these nuclei will be discussed in the next chapter.5 to 25 Mev. reaching a maximum somewhere near 20 Mev. A sample of vanadium was bombarded with ?'rays with energies which could be varied from 10. These results provide another example of the way in which the study of nuclear reactions is related to the problem of nuclear structure. p) cross section appreciably larger than the one expected from the compound nucleus alone. it seems natural to assume that the Bohr theory of the compound nucleus should apply. n) cross section has been measured as a function of the energy of the ~ . a)Sc4'. the (r. and that the process is divided into two parts: (1) the absorption of the 7ray photon forming a compound nucleus with the excitation energy E = hv. and there seems to be evidence supporting the idea of a direct photoproton effect. it is of the order of cm2 for the lighterintermediate nuclei such as aluminum. p) cross sections decrease. n) reaction. a)reactions can be interpreted adequately a t present in terms of the compound nucleus theory. Studies of (7. p) reactions does not agree with these ideas.494 NUCLEAR REACTIONS [CHAP. At ?ray energies in the neighborhood of 20 Mev. These unexpectedly large (Y. An interesting theoretical problem arises because of the relative magnitudes of (7. n) and (7. p) cro& sections. Theoretical calculations show that only a small number of protons would have to be ejected by this direct process to give rise to a (7. In the intermediate and heavy nuclei. In reactions involving the intermediate and heavy nuclei. and proton emission has been found to be much stronger than expected. while the (7. the cross section increases quite rapidly above the threshold energy. In contrast with (7. The maximum value of the cross section increases with increasing mass number. the most probable process is the emission of a neutron leading to a (7. np) reactions are observed and their cross sections increase with increasing energy.
Excitation function for the reaction V5'(7. It is reasonable to suppose that the excitation energy brought by the incident Yray is shared among many particles and that. Vanadium consists almost entirely (99. under favorable conditions. The (7. a ) reactions in mediumweight nuclei. four nucleons fuse to form an aparticle in a kind of internal conversion process. p) reactions on that nuclide are stable. a ) reaction.8 millibarns a t 23 Mev and agrees with the theoretical curve based on the compound nucleus theory. Hence those reactions.1f591 REACTIONS INDUCED BY GAMMARAYS 495 F I ~ 1618. and the products of the (7. . which are more probable than the (7. a ) reaction was not observed below 15. aparticles probably do not exist as such in the nucleus so that direct ejection of an aparticle is very unlikely. This result seems to be a general property of (7. . It is not surprising that the idea of the compound nucleus applies to these reactions. The emission of the aparticle from the excited compound nucleus is then analogous to the emission of aparticles from radioactive nuclei. as shown in Fig. a ) S ~ ~ ' . 1618. ( ~ l ) 3 a NaI(T1) scintillator and a pulseheight analyzer.5 Mev because of the effect of the Coulomb barrier on the emission of aparticles. n) and (7. The excitation function (curve of yield or cross section against energy) has a maximum of 0.75%) of V5'. do not interfere with the analysis of the latter.
Some of the most recent cornpilati on^'^^. 16 1610 Reactions at ultrahigh energies. I t is no longer possible to separate the reaction into two independent phases and it is necessary to treat separately the interaction of the incoming particle with the individual nucleons of the target nucleus. By means of this scattering process. The complexity of both the experimental information and the theoretical ideas is such that further discussion of ultrahigh energy nuclear physics is beyond the scope of this book. has made it possible to locate these states with high precision. the first 10 Mev of excitation have been studied thoroughly enough so that most of the stationary states are probably known. In the majority of the nuclei in this region. Although the reactions do not fall into the kind of patterns that were found for intermediate and heavy nuclei. i t may be absorbed to form a very highly excited compound nucleus. a dozen or more nucleons may escape and many combinations of protons and neutrons have been observed. one or a few nucleons may get a large fraction of the energy of the incident particle. The theory of the compound nucleus does not apply to these reactions because there is not enough time for the excitation energy to be shared among all the nucleons before particles start being emitted. The latter nucleon or group of nucleons may be ejected from the nucleus without further collision. In most collisions with ultrahigh energy particles. however. the energy levels provide a  . The progress made in recent years in the methods of studying excited states. including the production of new kinds of particles called mesons. I t is more likely.    . At energies of 100 Mev or more. The region of ultrahigh energy also includes reactions induced by cosmic rays. At energies of about 50 Mev and up.496 NUCLEAR REACTIONS [CHAP. The interested reader ~ *relatively ~~' easy treatments of these is referred to the l i t e r a t ~ r e ' ~for subjects.^" include more than 400 energy levels of the 40odd nuclei of atomic number less than 11. nuclear reactions are quite different in character from those a t lower energies. Reactions of this kind are described by the term spallation. together with the establishment of an accurate and consistent set of nuclear masses (Table 11I). A great deal of information about the excited states of light nuclei has now been accumulated from nuclear reactions. the target nuclei are excited to energies far greater than the binding energies of individual nucleons. Nucleons or groups of nucleons such as aparticles or lithium nuclei escape until the residual nucleus is left in a relatively stable state. or they may share their energy with a few other nucleons. When a particle with ultrahigh energy strikes a nucleus. another'25' includes many levels of nuclei with Z between 11 and 20. that the particle will collide with one or more nucleons and then escape with part of its original kinetic energy. 1 6 1 1 Reactions with light nuclei : energy levels of light nuclei.
p)C14. The latter quantities can really be understood only in terms of quantum mechanics. All of the data which will be considered have been taken from the compilation by Ajzenberg and Lauritsen referred to a b ~ v e . 7)N15. In the discussion of light nuclei.   body of experimental data in which important clues to the problems of nuclear forces and nuclear structure can be found. Lowlying levels (bound states) are given by the proton groups from the reactions . t)C12. the reactions are of the types which have already been discussed. N14(n. With a few exceptions. Altogether. d)N15. O16(n. The energy levels of the light nuclei are sufficiently well defined and separated so that it is possible to present the level structure for each nucleus in the form of a diagram showing the known levels and indicating the nuclear reactions in which they are involved. In addition to the position of the level. 015(j3+)N15 0 1 6 ( ~p)N15. p)N15. therefore. N15 is the product nucleus. The way in which the energylevel diagram is constructed can be illustrated by treating a typical nucleus in some detail. t)C12. The energy level diagram for N15 is shown in Fig. the emphasis will be on the levels rather than on the reactions. 25 reactions among various light nuclei provide information about the excited states of N15. pn)N15. and O1'(d. ' ~ where " the references to the original sources are given. C12(a. N14(d. C13(d. n)N14. . it will be seen to provide a concise and easily interpreted summary of the available data.~)N~~. and their effect on nuclear reactions has therefore not been mentioned in this chapter.(=~~(P. N14(n. such as the width or lifetime of an excited state. 018 (P. When the values of these parameters for a state are known. . a)Bl1. C12(t. and in 13 it is the compound nucleus. 2n)N13. p)C14.1611] REACTIONS WITH LIGHT NUCLEI . and the treatment of the light nuclei in this section can be regarded as a further application of the methods and ideas which have been developed in the earlier sections of this chapter. Information about the angular momentum and parity can also be obtained from the probability of 8decay and that of 7decay. Although it appears formidable a t first sight. n)N14. The nucleus N15 has been chosen because of the large . they are also shown in the level diagram. These are: BH(a. other information is also desired in nuclear spectroscopy. p)C14. They influence the angular distribution of the reaction products. N14(n. C14(d. and information about them can be obtained from the way in which the yield of the product particle varies with the angle between the direction of emission of the product particle and the direction of the incident particle. C15(/3)N15. N14(n. n)N15. In 12 reactions. C13(d. p)N15. 1619. p)C14. N14(n. Two additional parameters that are important in the description of a nuclear state are the angular momentum quantum number and the parity. a)B1l. C13(d. a)N15. N14(n. n)N14. N14(t.number of reactions in which it is either the compound nucleus or the product nucleus. R1l(a. n)N14. C14(p. C13(d.~)N~~.
there is good agreement among the values obtained from the different reactions.28 Mev) of the lowest excited level and for the fact that this level is a doublet with a spacing of 30 kev. The N15 spectrum is remarkable for the rather great excitation (5. by the neutron groups from the reaction C14(d. Although no single reaction gives all the levels. p)N15. 1619 shows the energy states of N15. p)N15 and C1*(a. The numbers indicate the energy in Mev above the ground state of N15. and by the Yrays from the reaction N14(n.498 NUCLEAR REACTIONS [CHAP. The middle column of Fig. n)N15. These levels are listed in Table 167. D FIG. 16 c"*. r)N15. which serves as the zero of the energy scale for the whole diagram.1619. Energy levels of N15 (AjsenbergSeloveand Lauritsen(57)). The angular momenta are indicated by the Jvalues as obtained from the analysis of the angular distribution of . N14(d.
TABLE 168 * Excitation energy in Mev.305 6.315 NM(n.309 8.164 7. p)N15 proton groups 5.328 7. n)N15 neutron groups * Excitation energy in Mev.278 9.1611] REACTIONS WITH LIGHT NUCLEI 499 NM(d.275 6.276 5.Y)N15 Yrays 5. p)N15 proton groups C14(d.164 7. .156 C12(a.325 7.356 8.
a)B" are indicated to the left of the upper part of the central column. Virtual levels of N15. Even parity is indicated by a plus sign. which is 8. are listed in Table 168. 1619. The binding energy of the last neutron in [N15]is 10.1620. the reactions N14(n. Nl4(nl p)C14. When two numbers are given for the angular momentum. Energy levels of N15 and 015. obtained from resonances in these reactions. p)N15 is shown to the left of the central column together with the Qvalue for the reaction. odd parity by a minus sign. together with the binding energy of the incident particle in the compound nucleus and a sketch of the cross section curve for each reaction. The formation of N15 by the 0+decay of 015and the 0decay of C15 are also shown. n)N14. it means that the assignment is not yet certain but the choice has been narrowed down to the values given. The reactions which give the lowlying levels are indicated a t either side of the central column. 16 the protons from the reaction N14(d.609 Mev.(58) .500 NUCLEAR REACTIONS [CHAP. Thus. The reactions which give rise to these resonances are shown on the sides of the diagram. About 30 levels. and NN(n. are shown in the upper part of the central column of Fig. as given by six different reactions. For example.834 Mev and this value serves as FIG. together with the binding energy of the incident particle in the compound nucleus. p)N15. the reaction NM(d.
New York: Intencience Publishers. 40. BREIT. Nuclear Reactions I . 1407. these levels'51' are shown for N15 and 015. G. 40. in fact. B. WILSON. M. 1 1 . Chapters 8. 1959. Levels. B. P." 1. New York: Wiley. 10. Nuclear Reactions I . G. F. 202372. The Atomic Nucleus. 1201. and Spectra of Light Nuclei. I n Fig." Vol." Vol. 9. WEISSKOPF. 10. pp. RAINWATER. Levels. Holland Publishing Co. 9. BURCHAM." Vol. "A Survey of Nuclear Reactions. Results like these provide evidence on the relative magnitudes of the force between two protons and that between two neutrons and. EVANS. 5. D. M. 13. New York: Academic Press. 1952. Nuclear Structure. ed. 4. New York: Wiey. Physics. pp. pp.. New Jeney: Princeton University Press. 1960. GENERAL  R. J. E. "The Nuclear Photoeffect. F. ~ ' for example. Part VI. pp. J. 1953. "Nuclear Reactions at High Energies. and Spectra of Heavy Nuclei. pp. BLATT and V. 450537. 3."Theory of Resonance Reactions and Allied Topics. The values of the resonance levels as given by the different reactions are in excellent agreement. MORRISON.has ~ ~ been ~ ~found. WATTENBERG. W." Experimental Nuclear P. "Nuclear Reactions. WIGNER. 1958. "Resonance Processes by Neutrons. KINSEY. 309361.." Vol. Nuclear Spectroscopy. 41. 19571959. Themetical Nuclear Physics. Nuclear Reactions I ." Vol. 2. ARTICLES IN "The Encyclopedia of Physics" (Handbuch der Physik) Berlin: Springer. Nuclear Reactions I I I . ed. Vol. L. Nuclear Reactions. Chapters 4. Vol. "  . Segrh. Princeton. Nuclear Reactions I. R. AJZENBERQSELOVE. "Nuclear Reactions. that the distribution of lowlying levels in mirror nuclei is similar. 40. A. pp. 42. that is. to have a particularly large I'. EISENBUD and E. The energy of the excited level is obtained from the binding energy and the resonance energy as shown in Section 162. Amsterdam: NorthP. while the latter has 8 protons and 7 neutrons. ' ~ I ~t . support the hypothesis that these two forces are equal. 1620. The comparison of the energy level diagrams of the light nuclei shows that there are similarities in some of the level spectra which have a bearing on the problem of nuclear f o r ~ e s . eds. Chapters 12. DEMEUR. The former nucleus has 7 protons and 8 neutrons. 40. New York: McGrawHill. BISHOP and R. The levels which are crosshatched are known to be particularly broad. E. 14. ENDT and M. 373409. 6.REFERENCES 501  a zero point for the energies of the particles in the various neutroninduced reactions. Nuclear Reactions 11.
CONSTABLE. SAMPSON. M. 5 .Chapter 8. CHADWICK. Roy. ENDT to Z = 20). 463 (1931). R. ed. D. Government Printing Office. (London) A130. M." Proc." Proc. 49. BARSCHALL. FRIEDMAN and V. Mod.S. "The Compound Nucleus. New 10. 29. Mod. 77 (1955). KLYVER. 73. F. 134161. DUNCANSON and H. M. WEISSKOPF. 10 above. Soc. CONSTABLE. Phys. C. "Energy Levels of Light Nuclei. BLATT and V. 1952. 7 (1948). BRAAMS. 1955. "Artificial Disintegration by aParticles. Pile Neubon Research. 80. and VAN LIESHOUT. Roy. L. "Artificial Disintegration by RaC' aParticles: Aluminum and Magnesium. R. 9 . BREITand E. B u r r and V. F. B. Nucleat E w g y Levels ( A = 40 6."Nuclear Physics. and J. HUGHES." Proc. "Table of Isotopes. Phys." Revs. MILLER. "Artificial Disintegration by aParticles. J. HOLLANDER and SEABORG. Mod." Revs. F. J.16 COLLECTIONS OF ENERGY LEVEL DATA 1 . LAURITSEN. BOHR. Phys. 11. York: Wiley.C." Niels Bohr and The Development of Physics. F. ENDTand C. 9. 26. D. 24. Mass. U. P. G. STROMINGER. 76. Theoretical Nuclear Physics. 1955. J. 519 (1936). Nuclear Dynamics. "Energy Levels of Light Nuclei. F. M. 13." Phys." Revs. 30. LAURITSEN. 11." Revs. V. A." Phys."Neutron Capture and Nuclear Constitution. 1953. F. 321 (1952).KING. (London) A135. 4 . "Energy Levels of Light Nuclei (Z = 11 to Z = 20).pp. 396 (1931). 2 . AJZENBERGSELOVE and T. WIONER. Rev. Reading. IV.502 NUCLEAR REACTIONS [CHAP. and POLLARD. 2. H." Phys. (London) AIM. 48 (1932). 683 (1957). Atomic Energy Commission Document TID5300. AJZENBERG and T. 1340 (1959). HENKEL and H. Washington. P.S. 3 . WEISSKOPF. W. Experimental. Roy. W. 6. 344 (1936). 585 (1958). AJZ