You are on page 1of 27

CHAPTER 12

The Structure of Atomic Nuclei
12.1 Introduction
12.2 Nuclear Properties
12.3 The Nuclear Force
12.4 Electrons versus Neutrons as Nuclear Constituents
12.5 The IPA Potential Energy for Nucleons
12.6 The Pauli Principle and the Symmetry Effect
12.7 The Semiempirical Binding-Energy Formula
12.8 The Shell Model
12.9 Mass Spectrometers (optional)
Problems for Chapter 12
In Parts I and II we have developed the two theories, relativity and
quantum mechanics, that revolutionized twentieth-century physics. In Parts III
and IV we shall describe some of the many applications of these two theories. In
Part III we discuss subatomic systems (nuclei and subnuclear particles), and in
Part IV, systems containing more than one atom (mostly molecules and solids).
As far as practicable, we have made the chapters of Parts III and IV independent
of one another, so that you can select just those topics that interest you. In
particular, you can, if you wish, omit Part III entirely and go straight to Part IV.
Part III contains three chapters, 12 through 14. Chapters 12 and 13
describe the properties of atomic nuclei, while Chapter 14 treats the subnuclear,
"elementary" particles. The division between the two chapters on nuclear physics
is that Chapter 12 is concerned with the properties of an isolated, static nucleus,
while Chapter 13 treats nuclear reactions, in which nuclei change their energy
and often their identity as well. These two chapters are not completely
independent of one another, but if your main interest is in nuclear reactions
(Chapter 13), you need to read only Sections 12.1 to 12.3 before skipping to
Chapter 13. If your main interest is in particle physics (Chapter 14), you could
read just Sections 12.1 to 12.3 and 13.1 to 13.4 before going directly to Chapter
14.

12.1 introduction
In discussing the properties of atoms, we needed to know surprisingly little about
the atomic nucleus. In fact, for most purposes we needed only three facts: The nucleus
carries a positive charge Ze; the nucleus is so much heavier than the electrons that we can
usually treat it as fixed; and the nucleus is so small that — for the purpose of atomic
physics — we can usually treat it as a point particle.
In the next two chapters we turn our attention to the internal structure of atomic
nuclei. We shall find that in some ways the motion of the protons and neutrons inside a
nucleus is analogous to the motion of the electrons in an atom. Many of the ideas
developed for atomic physics (the Schrödinger equation, the Pauli principle, spin, etc.)
are equally useful in nuclear physics.
Nevertheless, there are significant differences between atomic and nuclear
physics, of which perhaps the most important is the difference between the forces
involved. The force that holds electrons in their atomic orbits is the familiar electrostatic
force, which was well understood long before the advent of atomic physics. In nuclear
physics the situation was quite different, at least historically. As we shall argue in Section
12.3, the force that holds protons and neutrons together in a nucleus is not a force that
was known in classical physics, and, even today, we do not have a complete theoretical
understanding of nuclear forces. Therefore, the history of nuclear physics during the last
60 years is the story of a simultaneous effort to elucidate the nuclear force and to explain
the properties of nuclei. In this sense, the development of nuclear physics has been more
difficult than that of atomic physics. Nevertheless, we shall see that nuclear physicists
have achieved a remarkable understanding of nuclear properties. We shall find that this
understanding is both interesting for its own sake and of great practical importance.
12.2 Nuclear Properties
We begin our study of nuclear physics with a survey of several nuclear properties, some
of which we met briefly in Chapter 4. In later sections we shall see that many of these
properties can be understood with the help of the same quantum principles developed in
Chapters 7 through 11.
CONSTITUENTS

We have already stated that nuclei are made up of protons and neutrons. The
number of protons we denote by Z, and the number of neutrons by N. Since each proton
has charge + e, while neutrons are uncharged, the total charge of the nucleus is + Ze.
The mass of a nucleus is
mnuc  Zmp  Nmn 

B
,
c2

where B is the total binding energy, the energy required to pull the nucleus apart into its Z
+ N separate constituents. As discussed in Chapter 4, the binding-energy term, –B/c2, is
usually less than 1% of mnuc. Further, the difference between the masses mp and mn is only

1 1 This arrangement has changed from time to time. with The naturally occurring elements have atomic numbers in the range 1  Z  92 and neutron numbers with 0  N  146. 13 6 C7 and 14 7 N 7 . Hence we can neglect the last term in (12. it is usually not necessary to give all of the three numbers on these symbols. as illustrated in Table 12.1) and set mn  mp to give the approximation mnuc ~ ( Z  N ) mp  Amp . It is sometimes convenient to write the numbers A. where. 29 Cu 34 . with the same mass number are called isobars. Fortunately. The values of Z and N for all naturally occurring nuclei are plotted in Fig. and that the tendency for N to be somewhat bigger than Z is due to the electrostatic repulsion of the protons in nuclei with large Z.8. Another important tendency among nuclei is the observed preference for Z and N to be even. and two nuclei. from which one sees that for light nuclei (A 40) N and Z are nearly equal. 24 He 2 . 126 C6 .1.] 16 17 Two different nuclei. 238 92 U146 .6 that the symmetry effect results from the Pauli principle and the so-called charge independence of nuclear forces. This tendency is the result of the so-called pairing effect. which is the total number of nucleons (protons and neutrons) in the nucleus. As discussed in Chapter 4. such as 6 C8 and 7 N 7 . TABLE 12. 21 H1 . 2H. [This may be a good time to remind you that the first three nuclei in (12. such as 8 O8 and 8 O8 . the proton. and the nucleus 4He is called the alpha particle. A  Z  N. which have the same number of protons are called isotopes. 12. The tendency to have N  Z is called the symmetry effect.2) have their own special names: 1H is. and there are only four stable nuclei in which both Z and N are odd. The agreed placement around the chemical symbol X is now1 A Z XN . and is sometimes denoted D. Some examples are 1 1 63 H 0 . .5 Z in the heaviest nuclei. with N  1.1 part in 1000. the corresponding atoms are 14 14 chemically indistinguishable. or nucleon number. We shall show in Section 12. and we shall generally include only as many as seem necessary for the discussion at hand. of course. such as the same number of neutrons are called isotones. we have defined the mass number. denoted p. the nucleus of heavy hydrogen. The stable nuclei in which Z and N are both even outnumber comfortably those in which either Z or N is odd. or very nearly so. as before. Two nuclei. and N explicitly with the chemical symbol. Z. is called the deuteron. while for heavier nuclei N is a little larger than Z.1. and many older books use a different convention. denoted . which is described in Section 12.

3 where V0 is the constant 43  R0 . protons.3) is clearer if we consider the volume of the nucleus. and. V  43  R 3  43  R0 A 3  V0 A . and  particles to probe the size and shape of atomic nuclei. Evidently. It is found that the radii of all nuclei are given approximately by the simple formula R  R0 A1/ 3 . although in a few cases this difference can be as large as 30%. It is usually a reasonable approximation to treat nonspherical nuclei as spherical. or fermis (1 fm = 10–15 m). like most pumpkins). since they are ejected by many naturally occurring radioactive substances. (12. In nonspherical nuclei the longest diameter is typically a few percent greater than the shortest diameter. The radii of nuclei increase steadily from about 2 fm for light nuclei (helium. with radii of just a few femtometers. The significance of (12. like an American football) and some oblate (flattened. This means that the volume per . some being prolate (elongated. R0  1. (The main reason that he used  particles was simply that they were readily available. The majority of nuclei are spherical. for example) to about 7 fm for heavy nuclei (uranium. A number are slightly nonspherical. the number of constituent nucleons. for example). where A is the number of nucleons and R0 is a constant. More recent experiments have used artificially accelerated beams of electrons. we shall consider nuclei to be spherical.Numbers of stable nuclei Z N Number of stable nuclei Even Even 148 Even Odd 51 Odd Even 49 Odd Odd 4 SIZES AND SHAPES OF NUCLEI We described in Chapter 4 how Rutherford probed the atomic nucleus using energetic  particles.3) implies that the volume of a nucleus is proportional to A.07 fm. for the most part.) Rutherford's experiments established that nuclei are much smaller than atoms. The results of these experiments are summarized in the next few paragraphs.

This is evident in Fig. In heavy nuclei the level spacing is usually several tens of keV. because of their wave functions. and the definition assumed in (12. In no case is there a unique radius at which the density falls exactly to zero. quite like the molecules in a liquidand entirely unlike the electrons in an atom. that is. for instance). The usual definition of R. the only important difference being that the photons emitted by nuclei are usually far more energetic. as expected. 3 NUCLEAR ENERGY LEVELS Like atoms. whose volume is some 1014 times smaller than the atom (Problem 12. the electrons' probability distribution is spread out through the atom. the spacing of the levels in nuclei is many times larger than that in atoms.7). If a nucleus is raised to one of its excited states (when struck by an energetic proton.nucleon V0 is approximately the same in all nuclei. nuclei remain locked in their ground states during normal chemical processes.4). (Remember that water has density 103 kg/m3. which have a measurable size. This is analogous to the emission of photons in atomic transitions. This means that the nucleons are quite closely packed inside a nucleus. This does not change the fact that the electron itself is very small compared to the atom. current theory holds that the electron is a point particle — unlike nucleons.) The enormous density inside nuclei is easy to understand if we recall that almost all the mass of an atom is in the nucleus. Notice that.1 The radius of a nucleus is not an exactly defined quantity. . which shows the distribution of density in three representative nuclei. Measurement of their energy tells us the difference between the two nuclear energy levels involved. it can return to the ground state by emitting a photon. 12. 12. However. is the radius at which the density p(r) falls to half its maximum value p(0). In other words. In fact. Notice also that all three graphs fall to zero at about the same rate. Photons emitted in nuclear transitions (with energies ranging from keVs to MeVs) are called  rays. while in light nuclei it can range up to 10 MeV and more. some 14 orders of magnitude greater than the density of ordinary liquids or solids. nuclei are found to have quantized energy levels. the thickness of the surface region is about the same in an three nuclei.2 to be about 3  1017 kg/m . Therefore. This is why it is often possible in atomic physics to treat the nucleus as a structureless body with no internal motion. The actual density inside a nucleus is seen from Fig. Thus we can say that nucleons in a nucleus take up half of the total volume. the density of nucleons inside the nucleus is the same in all nuclei. The volume of a single isolated nucleon turns out to be about V0/2.3) and (12. 1 Of course.2. the nuclear density is 1014 times bigger than the atomic density. Because their excitation energies are so much larger than the energies involved in chemical reactions. the central density is the same in all three pictures. which is itself about the same as that of ordinary liquids or solids.

in the sense that it must more than balance the electrostatic repulsion of the protons. r2 The ratio of these forces is independent of r and is equal to Fgrav Felec  Gmp ke 2 2  (6. The gravitational attraction is Fgrav  Gmp 2 r2 while the repulsive electrostatic force is Felec  ke 2 . Second. and is always attractive.3 The Nuclear Force The force that holds electrons in their atomic orbits is the familiar electrostatic attraction between the positive nucleus and the negative electrons. there must be some other force between nucleons. EXAMPLE 12. charged or not. gravity is many orders of magnitude too weak. Since the gravitational force acts on all matter.7 1027 kg) 2  1036. Thus the total angular momentum J of a nucleus is the sum of the orbital angular momenta of all of its nucleons plus all of their spins.7 1011 N  m 2 / kg 2 )  (1. as the following example illustrates. with J  |J|  j ( j  1) h. The nuclear force must be stronger than the electrostatic force. .6 10 C) Clearly. First. the neutron is uncharged and feels no electrostatic force at all. 9 2 2 19 2 (9.1 Compare the gravitational attraction between two protons with their electrostatic repulsion. the magnitude of J is quantized. one might wonder if gravity could be the force that holds the nucleus together. the protons are all positive and the effect of the electrostatic force is therefore to push the protons apart. it should be clear that electrostatic attraction is not what holds the nucleons in the nucleus. and we call it the nuclear force. As with all angular momenta. the possible values are labeled by a quantum number j.0  10 N  m / C )  (1. For this reason the nuclear force is often called the strong force. 12.ANGULAR MOMENTA OF NUCLEI Like the electron. the gravitational force is far too weak to overcome the electrostatic repulsion of the protons and hold the nucleus together. On the other hand. Since neither the electrostatic force nor gravity holds the nucleus together. Unfortunately. both the neutron and proton have spin ½.

2  (940MeV)  (6 fm) 2 K min  That is. The easiest way to explain the observed constancy of nuclear densities is to assume that at very short distances the nucleons are held apart by a repulsive force. Inside 2 fm the force is attractive and the potential energy becomes negative. as we have just argued. it must also be of very short range. the conclusion stated is correct. The simplest argument for this last claim is that. dropping to a minimum of about — 100 MeV at 1 In fact. we can imagine any one nucleon moving inside a cubical box whose side a is of order 6 fm (the approximate diameter of a fairly light nucleus such as aluminum.3. the density inside a nucleus is approximately constant and is the same for all nuclei.2. . the constancy of the density in nuclei involves several other effects in addition to the short-range repulsion. Therefore. Those protons that pass by a nucleus at large distance are influenced only by the Coulomb repulsion between the proton's charge and the nuclear charge. In order that the nucleus hold together. indicating the presence of a strong. the force between nucleons must fall rapidly to zero as their separation increases. its total energy must be negative. It will be seen that the main features of this graph agree with our qualitative conclusions: The nuclear potential energy approaches zero rapidly beyond about 2 fm. and this proves to be the case. Only when the protons approach within 1 or 2 fm of the nuclear surface does the pattern of scattering change. Perhaps the simplest evidence for this comes from experiments in which a beam of protons is scattered by nuclei. However.1 Today. the nuclear force on any nucleon due to the other A – 1 nucleons must produce a potential-energy well of depth at least several tens of MeV. 27Al).A useful way to view the strength of the nuclear force is in terms of the energy of a nucleon in the nucleus. 12. The main conclusions of these experiments are summarized in Fig. that the minimum kinetic energy for such a particle is 3h2 2 3 2 (hc) 2  2ma 2 2mc 2 a 2 3 2 (200 MeV  fm) 2   20 MeV. While the nuclear potential well must be very deep. much as Rutherford's  particles were scattered by the nuclei in a metal foil. as we saw in Section 12. the minimum kinetic energy of any one nucleon in a nucleus is of order 20 MeV. which shows the potential energy of the nuclear force between two nucleons. At separations of 1 or 2 fm the force between nucleons is strongly attractive. Even protons passing within 2 or 3 fm of the nuclear surface scatter in the pattern predicted on the basis of a simple Coulomb repulsion. Nevertheless. As a crude model. We saw in Chapter 9. plotted as a function of their separation r. If the nuclear force were purely attractive. much of the best evidence for the details of the nuclear force comes from nucleon-nucleon scattering experiments. We conclude that the nuclear force between nucleons is negligibly small when they are more than 2 or 3 fm apart. attractive nuclear force. at separations less than about 1 fm the nuclear force becomes strongly repulsive. in which a beam of neutrons or protons is scattered by the protons in a hydrogen target. that is. we would expect nucleons at the center of a large nucleus to be forced inward more tightly than in a small nucleus. equation (9.32).

Apart from this obvious difference. Further evidence comes from the study of energy levels in nuclear isobars. the pair of nuclei 3 Li 4 and 7 4 Be3 . even when the electrons penetrate deep into the nucleus. two or more nuclei that have the same total number of nucleons and differ only in 7 their numbers of protons and neutrons. The only difference between these is that one neutron in 73 Li 4 has been replaced by 7 a proton in 4 Be3 Figure 12. the levels of the two nuclei are almost identical.6. Much detailed evidence for the charge independence of nuclear forces comes from studies of the interactions between two nucleons (p-p. it does not act on electrons at all — in just the same way that the electrostatic force does not act on particles that are electrically neutral. . The close correspondence of the energy levels (both in excitation energy and angular momentum) is strong evidence that the nuclear force acts equally on neutrons and protons. this was not always obvious. The force between nucleons depends somewhat on their angular momentum.7 MeV due to the additional Coulomb repulsion.4 Electrons versus Neutrons as Nuclear Constituents With our general knowledge of nuclear properties and forces we are now ready to examine several questions in more detail. or both neutrons.about 1 fm. since its interaction with the nucleons is just the well-understood electromagnetic interaction. This property is called the charge independence of nuclear forces — the nuclear force on a nucleon is independent of whether the nucleon is charged (a proton) or uncharged (a neutron). Consider. The first question that we address concerns the constituents of nuclei. Because 7 Be has one extra charge. the nuclear force between two nucleons is the same whether they are both protons. The principal evidence for this claim comes from experiments in which high-energy electrons collide with nuclei.1 CHARGE INDEPENDENCE OF NUCLEAR FORCES Two final properties of the nuclear force should be mentioned. there is no evidence that they are affected by the strong.3 is that for two nucleons with zero angular momentum. or one proton and one neutron. Although we now know that nuclei consist of protons and neutrons. It is charge independence that is responsible for the observed tendency for nuclei to have equal numbers of neutrons and protons as we describe in Section 12. ELECTRONS AND THE NUCLEAR FORCE Although the nuclear force is so crucially important for neutrons and protons. That nuclei contain protons was established in 1919 when Rutherford 1 Strictly speaking. the force is repulsive and the potential energy rapidly becomes positive and large. 12. that is.4 shows the lowest five energy levels of each of these two nuclei. This makes the electron a very useful probe of nuclei. p-n. but for our present purposes the potential energy shown is the most important. it is found that for a given quantum state. 12. all of its levels are raised by 1. First. nuclear force. n-n) in collision experiments. Inside this minimum. the potential energy sketched in Fig. for example.

and because the electron has charge – e.observed protons ejected from collisions between  particles and nitrogen nuclei. but we need to be careful because the electron mass is only 0. Because me s so small. as in the following example. THE ENERGY OF NUCLEAR CONSTITUENTS We have seen that the kinetic energy of a proton or neutron inside a nucleus is at least several MeV. in (12. The picture of the nucleus as made of protons and electrons had two very appealing features.1 Therefore. We saw in Section 9.2 Assuming that an electron could be confined inside a nucleus of radius about 3 fm (for example. the mass would still be about Amp. px = h/2a. and the same minimum applies to py and pz well. Second. The question that we address here is this: How do we know that the picture of nuclei as made of protons and neutrons is correct and the proton-electron picture incorrect? We present two arguments. that the nucleus contained electrons.9) reflects that the maximum possible wavelength  is  = 2a. Therefore.5 MeV/c2 so we must anticipate that its motion may be relativistic. and we have seen that the observed nuclear force is sufficiently strong. Suppose. however. since electrons and protons have opposite charges. we shall start from the general relation E  ( pc) 2  ( mc 2 ) 2 and find the minimum possible value of p2. the minimum possible value of p2 is 1 The minimum. to give it charge Ze. Recall that the two basic facts about nuclei are that they have charge Ze and mass Amp (approximately). 2a where a is the length of the cube's edge. First. it was conceivable that the nucleus was held together by electrostatic attraction. the same two facts could be equally well explained by supposing that the nucleus has A protons and A – Z electrons. This requires that the nuclear force be strong enough to confine nucleons with these energies. and A – Z neutrons (with mn  mp). as observed. . We can do the same thing for an electron. The neutron was not discovered for another 12 years. the A – Z electrons would reduce the total charge to Ze. and at first. Today we explain these by saying that the nucleus has Z protons. However.3 that the requirement that the wave function vanish at the walls of the box led to a minimum for each component of momentum: px  hk x  h . EXAMPLE 12. it was economical: If correct. It is a simple matter to estimate the minimum kinetic energy of an electron in a nucleus. to give it total mass Amp. protons and electrons. 27Al estimate the electron's minimum kinetic energy. it meant that all matter was made of just two kinds of particle. it was natural to suppose that nuclei were made of protons and electrons. To estimate the kinetic energy of nucleons in a nucleus we treated the nucleus as a rigid cubical box.

However.2 10 4 MeV 2 . This energy is negative and would tend to bind the electron.p 2  px2  px2  px2  3h . there is no known force that could confine an electron with this much kinetic energy. 2a Setting a  6 fm (the diameter of the nucleus) we find that ( pc ) 2  3(hc) 2 3  (1240 MeV  fm) 2   3. Therefore.25) MeV  180 MeV. 4a 2 4  (6 fm)2 Substituting into (12. MAGNETIC MOMENTS OF NUCLEI Our second argument against nuclei made of protons and electrons concerns their magnetic moments. but comparison with (12. the kinetic energy. and we conclude that there can be no electrons bound inside a nucleus. For 27Al. and since we now know that the nuclear force does not act on electrons.23) U  9 MeV. Note that this answer is much larger than mc2 = 0. and hence that the electron would indeed be moving relativistically. K = E – mc2. In the 1920s the only known possibility was the electrostatic Coulomb force.) In the same way we would expect the nucleons' magnetic moment to be of order N  eh . (The exact magnetic moment depends on the orbital quantum number l but is always equal to B times a number of order 1. the Coulomb force is still the only candidate.8). We saw in Chapter 10 that the electron has a magnetic moment whose magnitude is given by the Bohr magneton. It is easy to estimate the electrostatic potential energy of an electron inside any given nucleus. for example. is close to the total energy and K min  180 MeV. 5 MeV. we obtain for the minimum energy E  ( pc) 2  ( mc 2 ) 2  (3 104 )  (0.10) shows that it is hopelessly too small to overcome the minimum kinetic energy of the electron. 2me where me is the electron's mass. one finds (Problem 12. This example shows that an electron confined inside a nucleus would have kinetic energy of order at least 100 MeV. 2mN . B  eh .

and it is the motion of these charges that gives the neutron its magnetic moment. The dashed curve shows a smoothed. Figure 12. which consists. because x — unlike r — can be positive or negative. one simple reason is just that the system in question is a quantum system. some of which depend on subtle cancellations between the attractive and repulsive parts of the nuclear force.12) is that. we found the possible states of each separate electron and hence those of the atom as a whole. Working within this approximation. there are charges inside the neutron. corresponding to the nucleon being to the right or left of its fixed companion. one might expect it to have no magnetic moment. Thus if there were any electrons in the nucleus. of protons and neutrons. the Bohr magneton B is 2000 times greater thanN. The question naturally arises whether the corresponding approximation would be useful in nuclear physics. since me is about 2000 times smaller than mN. however.5(b) shows the potential energy of one nucleon in a one-dimensional nucleus containing four other nucleons.5(a) is a schematic representation of the potential energy of a nucleon moving along the x axis and interacting with just one other nucleon. In this approximation each of the Z electrons was assumed to move independently in the field produced by the average distribution of the other Z – 1 electrons.) Figure 12. 12. The important difference between B in (12. which is therefore called the nuclear magneton. This proves to be correct: Both the proton and neutron1 have magnetic moments whose magnitudes are close to the value (12. In the early days of nuclear physics it was generally believed that the answer would be no. However. But quantum mechanics tells us that the four nucleons are themselves spread out in 1 Since the neutron is chargeless.5 The IPA Potential Energy for Nucleons In studying atomic structure we relied heavily on the independent-particle approximation. Figure 12. average potential energy. each at a definite position in the nucleus. In this way we were led to the shell structure of atoms and to a remarkably successful account of the periodic properties of the elements.11) and N in (12. 12. their large magnetic moment would dominate.3 folded over. . However.22) it is clear that there are no electrons inside the atomic nucleus. we now know that although its total charge is zero. abbreviated IPM. and the total magnetic moment would be of order B.21 and 12. In fact.where mN denotes the mass of the nucleon. or IPA2. the measured magnetic moments of an nuclei are far smaller than this and are all of order N — in keeping with the proton-neutron theory. and at first sight it would seem an extremely bad approximation to replace the rapidly fluctuating solid curve by the smooth dashed curve. For these and other reasons (Problems 12. instead. In fact.5(b) represents the potential energy of a nucleon in the field of four classical nucleons. There are several reasons for this. the effect of the rapidly fluctuating nuclear potential energy is approximated surprisingly well by its smoothed IPA average.12). The main reasons were that the nuclear force is so strong and of such short range. 2 Nuclear physicists often call the IPA the independent-particle model. however. (This is just the graph of Fig.

is spherically symmetric and is sketched in Fig.13) we see that.accordance with their wave functions. .7 and is maximum at the center. 12. For protons in heavy nuclei this is certainly not a reasonable approximation. The range of the potential well is given approximately by R (force range)  (1. r Inside the nucleus it has the shape of the inverted parabola shown in Fig.27 and 12. To a good approximation this is the potential energy of a proton in the electric field of a uniform sphere of charge (Z – 1)e. as one would expect.6 is the averaged nuclear potential energy seen by each nucleon. and we shall have to use the total potential energy. we must add the potential energy of the electrostatic Coulomb force. 2 R The Coulomb potential energy (12.8.17 fm) A1/ 3 . which shows the IPA potential well for a proton in the middle-mass nucleus 120Sn (tin). We have seen that the nucleons are distributed nearly uniformly through a sphere of radius R (nuclear surface)  (1.16) is easily calculated for any given nucleus (Problems 12. Figure 12.6. while the real nucleus is. 12.5 times its value at the nuclear surface (Problem 12. since the force can "reach out" 1 fm or so beyond the distribution of nucleons. 12. In the case of protons. three-dimensional.6. this IPA potential energy is the same for neutrons as for protons. however. 12. 12. of course. The potential-energy function in Fig. This quantum spreading of the other particles is an important factor in the success of the IPA in describing the motion of nucleons in nuclei. the force range is a little larger than the nuclear radius. This function is plotted in Fig. the average potential energy seen by each nucleon is found by averaging the nucleon-nucleon potential energy over this sphere.28): 3 ke 2 U Coul (0)  ( Z  1) .07 fm) A1/ 3 . U(r).5 is one-dimensional. where it is 1. For protons in light nuclei it is therefore a fair approximation to ignore UCoul compared to the nuclear potential energy of Fig.28). The answer varies from 1 or 2 MeV for the lightest nuclei to about 30 MeV for the heaviest. Therefore. The depth of the IPA wen is about 50 MeV and is approximately the same for most nuclei. U (r )  U nuc ( r )  U Coul (r ). Outside the nucleus (r > R) this is U Coul (r )  ( Z  1) ke 2 . Comparing this with (12. The resulting IPA potential energy. Because of the charge independence of the nuclear force. as it is also for the case of electrons in an atom.

Using the IPA potential-energy functions one can calculate the energy levels of each nucleon in a nucleus. though not the details. The ground state of a nucleus is found by assigning its Z protons and N neutrons to their lowest-possible states consistent with the Pauli principle.5. similarly. but we have shown their actual observed values — two for the lowest level and four for the next.It is often convenient to display the IPA potential-energy functions for protons and neutrons side by side. To see how this works. 12. consider the nucleus 5 B7 with five protons and seven neutrons.6 The Pauli Principle and the Symmetry Effect We saw earlier. especially when A is small. not as good for heavier nuclei. It is important to understand that the Pauli principle only restricts the states of two identical particles.8. (This will be a good approximation for light nuclei.9. Like electrons. which also shows the ground states of 6C6 and 7N5. in Fig. As we describe in Section 12. we can now show that the symmetry effect is a simple consequence of the Pauli principle and the charge independence of nuclear forces. of the energy levels of the wells shown in Fig. Figure 12. For example. the nuclear shell model was developed about 1950 and is very successful in explaining many (though certainly not all) nuclear properties. 12. The ground state of 5 B7 is shown on the left of Fig. is a nucleus with Z = N. let us first ignore the Coulomb repulsion of the protons. but Z  N.10(b) shows the resulting ground 12 state of the nucleus 6 C6 . so that the "wells" in which the protons and neutrons move are identical. quite analogous to the atomic shell model described in Chapter 11. no two neutrons can occupy the same quantum state. the same is true of the levels.9.) 12 The example shown in Fig. Since the wells are identical. (The precise degeneracies of the levels do not matter for the present discussion.1. 12. In this way we can construct a shell model of the nucleus. they describe the potential energy of a proton and a neutron in one and the same nucleus. Comparing the ground state of 5B7 with . Before taking up the shell model we discuss two other ideas that depend on the existence.11. 12. This tendency is often called the symmetry effect. Thus it is perfectly possible to have a proton and neutron both occupying exactly the same quantum state. The Pauli principle requires that we place the Z protons in the lowest proton levels. 12. One convenient way to do this is to draw them "back to back" as in Fig. That is. for comparison. along with their degeneracies. and similarly with the neutrons.10(a) shows the two wells and the lowest few levels in each.10(b). Using the ideas of Section 12.) Figure 12. 12. both protons and neutrons obey the Pauli exclusion principle. one of the 12 12 isobars of 6 C6 . that there is a tendency for stable nuclei to have nearly equal numbers of protons and neutrons. It is important to remember that although we have drawn these two functions separately. no two protons can occupy the same quantum state and. Let us now consider a nucleus with the same number of nucleons. with the number of protons in a level never exceeding that level's degeneracy. 6 C6 . that is. it places no restrictions on the states occupied by two particles of different type.

Based on these arguments (in which we have so far ignored the Coulomb energy of the protons) we can state the following general conclusion: Among any set of isobars (nuclei with the same total number of nucleons) the nucleus (or nuclei) with Z closest to N will have the lowest total energy. all occupying the lowest neutron level. 12. Binding energies of different nuclei are a useful measure of their relative stabilities as we shall see later. In Fig. the stable nuclei. However. If we move still farther away from Z = N (to 8O4. For light nuclei the difference between the proton and neutron wells is small. Under these conditions the majority of elements that make up our world would not exist. Similar arguments can be applied to any set of isobars. the nucleus with greater energy is unstable and will eventually convert itself into the isobar with lower energy.4.1. and 7N5 has higher total energy than 6C6. as we move to heavier nuclei. If the neutron well is appreciably lower than the proton well. as indicated schematically in the figure. It is perhaps worth reiterating the important role of the Pauli principle in the symmetry effect.that of 6C6 we see that because of the Pauli principle the seventh neutron in 5B7 has to occupy a higher level than any of the neutrons or protons in 6C6. This is shown in Fig. there may be several neutron levels below any of the proton levels. which shows the two wells and a schematic set of energy levels. Therefore. The process by which this occurs is called  decay and is discussed in Section 13. If a nucleus has appreciably more energy than a neighboring isobar. but for now the important point is this: Any nucleus with Z much different from N is unstable and will convert itself into a neighboring isobar with Z closer to N and hence lower total energy. as is confirmed by the data in Fig.5. The seventh proton of 7N5 must occupy a higher level than any of the nucleons in 6C6. the nucleus 5B7 has higher total energy than 6C6. In this section we discuss how the binding energy . should all have Z  N. Therefore. and our original conclusion that stable nuclei should have Z  N is unaltered.12(a) it is clear that without the Pauli principle. we need to anticipate a property of radioactive decay. 12.12(a). In this case the most stable isobar will have more neutrons than protons. we find a similar situation. for example). 12. In discussing the symmetry effect we have so far ignored the electrostatic energy of the protons. the total energy has to be even higher.7 The Semiempirical Binding-Energy Formula The binding energy B of a nucleus is the total energy needed to separate completely its A individual nucleons. the state of lowest energy would consist in all (or almost all) the nucleons being neutrons. and hence the nuclei normally found to occur naturally. 12 12 If we look on the other side of 6 C6 at 7 N 5 . which we discuss in Chapter 13. 12. This is easily taken into account: As discussed in Section 12. the Coulomb energy simply raises the proton well relative to the neutron well. the Coulomb energy steadily rises and the ratio N/Z should increase slowly. To understand what this conclusion implies.

N. This means that we can rewrite (12. B   17(1. We start with a brief discussion of the measurement of binding energies.969 u]c 2 1 There is. we can add Zme to mnuc in (12.19) the required binding energy is B   17mH  18mn  matom ( 35 Cl) c 2 or. This replaces Zmp by Z (mp  me )  ZmH . nuclear binding energies are large enough to cause an observable difference (of order 1 %) between the mass of a nucleus and the sum of the separate masses of its Z protons and N neutrons: mnuc  Zmp  Nmn  B . EXAMPLE 12. rather than that of the bare nucleus. and A = Z + N. The mass matom is equal to mnuc plus the mass of the Z atomic electrons1 matom  mnuc  Zme .5. As mentioned in Section 4.18) if we also add Zme to the term Zmp.969 u)  c 2  [35. Now. . with the data from Appendix D (rounded to five significant figures). According to (12.depends on the nuclear parameters Z. since it is matom that is usually measured. a small correction due to the binding energy of the electrons. while nuclear binding energies are several MeV. Given accurate measurements of the masses involved. this correction is almost always negligible. where mH is the mass of the hydrogen atom (1H).18) as B  ( ZmH  Nmn  matom )c 2 and work entirely with atomic masses. we can use this relation to find B as B  ( Zmp  Nmn  mnuc )c 2 . c2 where mnuc denotes the mass of the nucleus concerned.0087 u)  (34.289 u  32. Most mass tabulations. including that in Appendix D.0078 u)  18(1. give the mass matom of the atom. Since average electron binding energies are at most several keV. of course.3 Use the atomic masses in Appendix D to find the binding energy of the 35 nucleus 17 Cl18 .

N) completely equals the energy to pull off one neutron plus the energy to dismember the remaining nucleus (Z. (0. the energy to dismember (Z. but the values of the coefficients that appear are generally found by fitting the formula to the measured binding energies. after its inventor. Notice also that (12. All of the terms in this formula have at least partial theoretical justifications.6 MeV)  131. For example. we can use (12. N  1). one can measure the energy needed to separate one neutron from a nucleus. It is easy to see that the binding energy of the nucleus (Z. is seldom possible. or the Weizsacker semiempirical formula. one can find the binding energy of a nucleus if one knows the corresponding atomic mass accurately. N – 1). This difference is always of order 1% or less. or vice versa. Notice in (12.22) the required binding energy is B ( 17 O)  S n ( 17 O)  B( 16 O)  (4. As the example above shows. S n ( Z . Thus if the binding energy of (Z. Here we needed five significant figures in the masses to get three in B. N – 1) is known.8 MeV. If we insert the formula for B into . However. it is often possible to find the binding energy of one nucleus in terms of the known binding energy of some other nucleus.9). What is the binding energy of 17 8 O9 ? According to (12.2 MeV is needed to separate a neutron from 17 8 O9 . Using a mass spectrometer it is possible to measure atomic masses very accurately (see Section 12. For this reason the formula is called the semiempirical binding-energy formula. N. N )  energy to separate one neutron from nucleus (Z .6 MeV. The direct measurement of binding energies.20) that the mass of the separate constituents is greater than that of 35Cl confirming that we must add energy to puff the nucleus apart. N )  B ( Z . and this is therefore one of the best ways to find nuclear binding energies. N ). N). by complete separation of a nucleus.22) to find the binding energy of (Z. EXAMPLE 12.2 MeV)  (127. Thus we must know the masses to at least two significant figures more than are required in our answer. This is called the neutron separation energy. that is. N )  S n ( Z . It is found that 4. In the remainder of this section we derive a formula that gives the binding energy of nuclei as a function of the numbers Z. and A = Z + N. N) is B ( Z .4 The binding energy of 16 8 O8 is known to be 127.5 MeV/c 2 1u  298 MeV.329 u)  931.20) gives B as the difference of two masses that are very nearly equal.

5 R Since R = R0A1/3 this implies a correction of the form BCoul   aCoul Z2 . It follows that the number of neighbors to which each nucleon is bound is approximately the same for all nuclei. For most nuclei.13) and are less tightly bound than those in the interior. where the surface correction Bsurf negative and has the form Bsurf   asurf A2 / 3 . (12. the average binding energy of each nucleon is the same in all nuclei. those neighbors within a sphere of radius about 2 fm.the equation mnuc = Zmp + Nmn – B/c2. A further correction is needed because of the electrostatic repulsion of the protons. of its nucleons. each nucleon is bound to only a fraction of the other nucleons. the resulting equation is called the semiempirical mass formula. which is smaller than most nuclear sizes.13. There is an immediate correction to the approximation (12. That is.23). then. 4R2 = 4R02A2/3. with asurf a positive constant. To an excellent approximation the binding energy of a nucleus is found to be proportional to the number. so that the total binding energy is proportional to A. 12. A. this approximation is called the volume term and is written B  Bvol  avol A where avol is a positive constant. A1/ 3 where aCoul 3 ke 2  .23) must be reduced by an amount proportional to A2/3 . The number of these surface nucleons is proportional to the surface area of the nucleus. This result is easily understood by the following classical argument: We have seen that the nuclear force has a range of about 2 fm.37) U Coul  3 k ( Ze) 2 . and our next approximation is B  Bvol  Bsurf . 12. Those nucleons near the nuclear surface have fewer neighbors to attract them (Fig. as shown in Fig. This reduces the binding energy by an amount equal to the potential energy of the total nuclear charge. 5 R0 . We have also seen that the density of nucleons is about the same inside all nuclei. Accordingly. namely. Since A is proportional to the volume of the nucleus. The potential energy of a uniform sphere of charge Ze is (Problem 12.

if Z and N are both odd.11) is therefore smaller. there are 100 nuclei with Z even but N odd. We mentioned in Section 12. or vice versa.    . Since B gets less as Z – N increases in either direction. 12.75 MeV asurf  17. this is not the exact value of aCoul (Problem 12. and there are only 4 in which both Z and N are odd.2 that the majority of stable nuclei (148. For now we simply state that this effect gives a correction to the binding energy with the form Bpair   apair A1/ 2 where apair is some constant and  1    0  -1  if Z and N are both even. in fact) have both Z and N even. the nucleus with Z = N has highest binding energy. and the difference of energy between neighboring isobars (shown as E in Fig. A The factor of A in the denominator reflects that for larger nuclei the level spacing is smaller.2 MeV. sym 1/ 3 A A A1/ 2 where  is defined in (12. where we saw that for any set of isobars. This preference for Z and i to be even is explained by a property of the nuclear force called the pairing effect.37). The observed binding energies of all nuclei are best fitted by a correction of the form Bsym   asym (Z  N )2 .8 MeV  aCoul  0. if Z or N is even and the other is odd. The final term in our formula for the binding energy reflects a phenomenon that we shall discuss in more detail in the next section. Putting all five of these contributions together.711 MeV asym  23.29).Since the nucleus is not precisely a uniform sphere of charge. this suggests a negative term proportional to (Z – N)2. we obtain the semiempirical binding-energy formula B  Bvol  Bsurf  BCoul  Bsym  Bpair  avol A  asurf A2 / 3  aCoul a Z2 (Z  N )2  a   pair . The five coefficients are generally adjusted to fit all the measured binding energies as well as possible.7 MeV  a pair  11. The next correction arises from the symmetry effect discussed in Section 12. One set of values that gives a good fit is as follows: avol  15.6.

12. Nevertheless. Clearly.5  62.30) and is seen to give an excellent fit for all but the lightest nuclei (those with A below about 20). being greatest when the electrons exactly fill a closed shell. as shown in Fig. B/A slopes gently down to about 7. occupy energy levels that fall into distinct shells. while fusion provides the energy of the stars. this decrease occurs because almost all nucleons in a small nucleus are close to the surface and the negative surface correction is proportionately large. the binding energy per nucleon.5 Use the binding-energy formula (12.30) does not work well for the light nuclei.30). in which the quantization of energy levels is proportionately more important. In this section we discuss the corresponding shell properties of nuclei. into two lighter nuclei.7  0) MeV =297 MeV.14. with the maximum value of 8.21). when A decreases below about 20. . and dropping steeply just beyond each closed shell. it is a smoothed.8 MeV for iron (Fe). 12. 12. Z = 17. The lower values of B/A when A is very small imply that energy is also released when two light nuclei fuse to form one heavier nucleus. like atomic electrons. and the coefficients (12.30) we find that a Z2 ( Z  N )2  a   pair sym 1/ 3 A A A1/ 2  (551. The most obvious feature of Fig. Inserting the values A = 35. It is not surprising that the smooth function (12. including our own sun. On the other side of 16Fe. EXAMPLE 12. even when A is small.8  0. the behavior of B/A. The lower values of B/A when A is large imply that energy is released when a heavy nucleus splits. is of great practical significance. Recall that atomic binding energies fluctuate periodically with Z.Because the dominant behavior of B is B  A.6 MeV for 238U. N = 18.14 is that B/A is close to 8 MeV for almost all nuclei.30) to predict the binding energy of 35Cl and compare the result with the measured value found in (12.2  190. Nuclear fission is the source of energy in nuclear power stations. Figure 12.14 shows the measured value of B/A for a large number of nuclei. Beyond iron. the general trend of B/A is correctly given by (12. The smooth curve is the value of B/A predicted by (12.8 The Shell Model The semiempirical binding-energy formula gives a remarkably accurate picture of the general trends of nuclear binding energies. Both fission and fusion are used in nuclear weapons. average picture and shows none of the fluctuations that one would expect if nucleons. B/A falls rapidly to zero for 1H (which has no binding energy). Nevertheless. or fissions.3 1) into the formula (12. Notice that the pairing term is zero in this example because Z is odd and N even. B  avol A  asurf A2 / 3  aCoul which agrees very well with the measured value of 298 MeV. it is usual to discuss B/A. this decrease is due mainly to the increasing importance of the Coulomb repulsion of the protons.

One simple procedure is to eliminate one of these variables by an appropriate averaging. The nuclear separation energies depend on two variables. Nuclei in which Z is magic tend to have larger numbers of stable isotopes than average. with Z equal to the magic number 50.11) shows a maximum at each of the closed shells. nuclei in which N is magic tend to have more stable isotones than average. 54. (This is easily discernible in the case of 2He2 on the graph of B/A in Fig. (We have not included data for Z < 20. 8O8. 10. and are harder to plot. that nuclei with Z = 126 have not been observed. for example. and even more so if both are. there is no experimental evidence for a proton magic number in this region. with an abrupt drop immediately thereafter. 12.15(a) shows the proton separation energy predicted by the binding-energy formula.33) consists of many observations. 20Ca20. The magic numbers are the same for protons as for neutrons. The binding energy per nucleon. then for each value of Z we can compute the average S p ( Z ) for all stable isotopes with this Z. has 10 stable isotopes.32).14. By the late 1940s. 20.1 On the other hand. A nucleus is especially stable if Z or N is equal to one of these magic numbers. Similarly. Z and N.15(a) and. 28. 82. reflecting the great difference between the dominant forces in nuclei and atoms. since our averaging procedure obscures shell effects for very light nuclei. as one would expect from the charge independence of nuclear forces. 12. however. 11.) The dashed curve in Fig. . in each of the "doubly magic" nuclei 2He2. as they came to be called. Recall that a plot of atomic ionization energy against number of electrons (Fig. 50. the proton separation energy SP(Z. 20Ca28. These are the energies required to remove one proton or one neutron from a nucleus and are analogous to the ionization energy of an atom. has pronounced drops immediately after each of the magic numbers indicated. and that these magic numbers. 36. 18. the nuclear magic numbers are quite different from the atomic closed-shell numbers (12. If we consider. The evidence for the magic numbers (12. tin.) Perhaps the best evidence for the magic numbers (12. This reproduces the general trend very well but entirely misses the fluctuations associated with the magic numbers. were: nuclear magic numbers: 2. with 7 stable isotones. 12. 1 Note. 86. A plot of Sp ( Z ) against Z is shown in Fig.33) concerns the separation energies Sp and Sn.THE MAGIC NUMBERS The total numbers of electrons in the closed-shell atoms are known to be atomic closed-shell numbers: 2. 8. substantial evidence had accumulated that nuclei have analogous closed-shell numbers which lead to unusual stability. as expected. and 82Pb126 is markedly higher than the prediction of the binding-energy formula. N = 82 holds the record. 126. B/A. few of which are conclusive by themselves. more than any other element. Therefore. N). For example.

the IPA potential well of Fig. then the nucleon has an additional potential energy that depends on the orbital angular momentum L and its orientation relative to the spin l. although not quite.6. 50. .. are indicated by the code letters s. The correct explanation was discovered independently in 1949 by Mayer and by Jensen and co-workers. in order of increasing energy. and 82. L  l (l  1) h.33). and so on. The first three of these numbers agree exactly with the observed magic numbers. and the total numbers of protons or neutrons needed to produce closed shells are indicated on the right.and so on. 12. but beyond this all of the predicted closed-shell numbers turn out to be wrong. as in Fig. 12. It turns out that the nuclear force depends strongly on the relative orientation of each nucleon's spin and orbital angular momentum.15(b) shows the average neutron separation energies S n ( N ) . For each value of l. the lowest d state 3d. and also at N = 126. 2. . l – 1.Figure 12. 2. . – l) and two possible values of S z   12 h . f. . 3p. . 2p. but if l  0. 4.16 are labeled according to the scheme normally used in nuclear physics. The energy levels for this well can be calculated numerically.1 The levels (or subshells) shown in Fig. p. Just as in atomic physics.15(a). there are 2l + 1 possible values of Lz  mh (with m = l. On the other hand.. . however. no reasonable well gives the observed numbers. 12. . 3. d. all of the parameters associated with angular momentum are the same as in atomic physics. In fact. 3.16 are grouped into well-defined energy shells. 28. 3d. . Thus the levels with l = 1 are labeled 1p.5 and sketched in Fig. . each level is identified by two quantum numbers. Evidently. The levels shown in Fig. Specifically.. 12. . The degeneracy of each level is therefore 2(2l + 1). . The IPA potential energy felt by any one proton or neutron was described in Section 12. the same as that used in atomic physics. This scheme is very convenient in the hydrogen atom (where it means that all states of given n have the same energy) but has no particular merit in nuclear physics. that is. n and l. At first it would seem natural to assume that a different shape for the IPA potential well might produce the correct magic numbers. . g. each averaged over all stable isotones of a given N. 20. 1 Recall that in atomic physics the lowest p state is called 2p. This scheme is nearly. where  is the angle between S and L. . physicists naturally hoped to explain them in terms of nucleon energy levels and the grouping of those levels into energy shells. . . the levels with l = 2 are 1d.16. and the ordering of the levels is shown in Fig. 12. The possible values l = 0. ORIGIN OF THE MAGIC NUMBERS Having established the existence of the nuclear magic numbers (12. 1. nuclear physicists define the quantum number n so that the levels of a given l are simply numbered n = 1. . This shows abrupt drops beyond the same magic numbers. This additional energy U is found to have the form U  S L cos . 12. The number l identifies the magnitude of the nucleon's orbital angular momentum. .6 is the correct potential-energy function for a nucleon with l = 0. . 2d. . .

J j ( j  1) h. Since the spin-orbit energy of each nucleon is proportional to S•L. 1. The effect of the spin-orbit energy on the levels of Fig. .) A spin-orbit splitting of this kind is familiar in atomic physics. 12. 12. then j  l  12 (S and L parallel). As one would expect. Rather. If l  0.17.17 the value of j in each level is shown by a subscript following the code letter for l. the effect of the spin-orbit energy increases with increasing l: The 1s level. 12.16 (with l  0) is split into two levels according to the two possible orientations of S relative to L. for a nucleon in a nucleus the corresponding state has much lower energy. since the electron has a magnetic moment   S.. which shows the resulting levels (or subshells) with their degeneracies and their grouping into energy shells. It can be seen that as a result of these splittings. If its orbital angular momentum is zero (l = 0). like all s levels. we must discuss briefly the total angular momentum J of a nucleon. J  L  S. Thus the original lp level. shown on the left. each energy level of Fig. It is found that the magnitude of J is given by a similar formula. and hence j = ½. like all p levels (and also d levels). but the resulting two levels are still in the same shell.U  S  L . . 12. and that of S is S  s ( s  1) h with s always equal to ½. is not split at all. this leads to a magnetic energy – •B  S•L. In Fig. the energy U is called the spin-orbit energy. To understand the labeling and degeneracies of the levels in Fig. 12. is really two levels with . where it produces the so-called fine structure of atomic spectra. the closed-shell numbers agree perfectly with the observed magic numbers. depending on the orientation of S relative to L: If S is parallel to L. The magnitude of L is L  l (l  1) h with l = 0. the nuclear spin-orbit splitting is not a magnetic effect. but if S is antiparallel to L. and all levels with l  3. However. the nucleon's total angular momentum is just the spin. The nuclear splitting is much larger than the atomic effect. (Recall that S has just two possible orientations relative to any direction.17. As described in Section 10. it is found that there are two possible values of j. is split. an orbiting electron sees a magnetic field B  L. The first clear evidence for this nuclear spin-orbit energy came from studies of nuclear shell structure.16 is shown in Fig. . but scattering experiments with separate spin-up and spin-down beams of nucleons have since confirmed its existence and properties. For this reason. 2. and is in the opposite direction: For an electron in an atom the state with S and L parallel has slightly higher energy. the splitting is so large that the resulting two levels belong to different shells.7. the 1p level. for the 1f level. it is a new property of the nuclear force. then j  l  12 (S and L antiparallel).

j  1.55) that for any given values of n and l. 12. That is. that would have existed in the absence of the S•L energy. with lp3/2 somewhat lower in energy. These are shown on the right of Fig. as indicated to the right of each level. which are given by J z  m j h. there are several possible orientations of J. 12. We first note that any filled level of given j contributes zero total angular momentum. We can now apply similar reasoning to nuclei.17 as 1p3/2 and lp1/2. It can be shown quite generally (Problem 12.6 we saw that one can predict the angular momentum of many atoms once one knows the order in which the electron energy levels are filled. ANGULAR MOMENTA OF NUCLEI In Section 11. The required degeneracies are 1 p3/ 2 : degeneracy  2 j  1  (2  32 )  1  4 and 1 p1/ 2 : degeneracy  2 j  1  (2  12 )  1  2. nl. there would have been a single 1 p level whose degeneracy would have been 2(2l + 1) = 6 (that is.) This implies that any nucleus in which all of the occupied levels are closed (no partially filled levels) should have zero total angular momentum. . Show that the sum of these is the same as the total degeneracy of the original lp level on the left of Fig. In the absence of the S•L energy.17. Thus each of the levels of Fig.17 has a degeneracy of (2j + 1).38) and (12.    .  j . all with the same energy. EXAMPLE 12. Thus the effect of the S•L term is simply to redistribute the degeneracies so as to produce the observed magic numbers. (This is because for every nucleon with J z  m j h there is a second with J z  m j h . the sum of the degeneracies of the two corresponding levels (with j = l ± ½) is equal to the total degeneracy of the single level. This is equal to the sum of (12.j  l  12  1  12  3 2 or 1 2 . where mj can have any of the 2j + 1 values m j  j . For each magnitude of the total angular momentum. 12. These correspond to the different possible values of Jz. two orientations of the spin times three of L).39).6 Find the degeneracies of the lp3/2 and lp1/2 levels.

the angular momentum of the whole nucleus is just that of the final. for example. in 6C6 the protons and neutrons both completely fill their 1s1/2 and 1p3/2 levels. unpaired nucleon.17 we see that just above the 1d5/2 level is the 2s1/2. Let us consider next a nucleus with a single nucleon outside otherwise filled subshells. and 20Ca28 should all have jtot = 0. Thus we predict that 8O9 should have jtot = 5/2. it turns out that the state in which the two angular momenta are antiparallel (jtot = 0) has significantly lower energy. the angular momenta of its nucleons can combine to give various different total angular momenta. in 8O9 the eight protons and eight of the nine neutrons fill the levels 1s1/2. we just saw that the ground state of 8O9 has its last neutron in the 1d5/2 level. We could. If the number of nucleons in the level is even. zero total angular momentum. we can predict in the same way that the closed-subshell nuclei 8O8. odd neutron. the ground state of 8O10 has jtot = 0. the pairing effect implies that in the state of lowest energy the nucleons will be arranged. and in every case this is found to be correct. and is typically a few MeV.J tot  jtot ( jtot  1) h in this case. Throughout this section we have been discussing the ground states of nuclei. and this proves to be correct. Whenever a level is partially filled. The filled levels all have zero angular momentum. The extra binding energy in the jtot = 0 state is called the pairing energy. 12. This energy is the origin of the term Bpair. and should have jtot = 1/2. For example. For example. In Fig. Hence we would expect that the first excited state of 8O9 would be obtained by lifting the last neutron to the 2s1/2 level. 14Si16. jtot = 0. as far as possible. and lp1/2.17. It is found quite generally that any pair of identical nucleons in the same level has lowest energy when their total angular momentum is zero. but since each of the last two neutrons has j = 5/2 there are several possible values of jtot depending on the relative orientation of the two separate angular momenta. However. in pairs with zero angular momentum. and does have. Therefore. Therefore. 12. For example. (12. but we can also predict the properties of some excited states. PAIRING ENERGY Let us next consider a nucleus with two identical nucleons (two protons or two neutrons) outside otherwise filled subshells. These properties lead us to the following useful rules: . 1p3/2. consider 8O10 with two neutrons in the 1d5/2 level. but the ninth neutron must occupy the 1d5/2 level all by itself Since the closed levels all have zero angular momentum. if the number of nucleons is odd. This is observed to be the case.28). Referring to Fig. this means that the state of lowest energy has jtot = 0. the state of lowest energy has all but one nucleon arranged in j = 0 pairs and the total angular momentum is just that of the odd. However. 6C6 should have. in the semiempirical binding-energy formula.

12.7. the mass spectrometer has developed into a powerful tool of chemical analysis.17). Finally.7 What are the quantum numbers jtot. A. unpaired nucleon. In the arrangement shown in Fig.9 Mass Spectrometers (optional) In this section we describe mass spectrometers.) The use of both rules is illustrated in the following example. 12. mass spectrometers can measure masses with an accuracy of order 1 part in 108. Second.The ground state of any nucleus in which both Z and N are even (an "even-even" nucleus) has zero total angular momentum and If a nucleus has A odd (that is. First.18. revealed the existence of isotopes. as we describe shortly. 12. for which the ordering of Fig. the devices used to measure nuclear masses. which could distinguish masses differing by a few percent. the protons have zero total angular momentum. the ions are next accelerated by a potential difference V0. 12. 12. revealed several properties of the nuclear force. of radius R. As we saw earlier. Since Z is even. The rule (12. measurements of nuclear masses contributed to our understanding of nuclear structure in two important ways. by a 1 The difficulty is to predict which relative orientation of the angular momenta of the last proton and last neutron gives the lowest energy. Hence the nucleus 43Ca should.4 1). correctly predicts the angular momentum of most odd-A nuclei. the only exceptions involve nonspherical nuclei. as we discussed in Section 12. and does. Z is even and N odd. The total angular momentum of the neutrons is just that of the twenty-third neutron. early measurements. which causes some atoms to lose one electron and become positive ions of charge e. they are bent into a circular path. which allows binding energies of typical nuclei to be calculated within 1 keV or so. The nucleus 43Ca has Z = 20 and N = 23. The variation of binding energy with mass number. The rule (12. together with the level ordering shown in Fig.53. (See Problem 12. In practice. The substance to be analyzed is introduced into an electric discharge. EXAMPLE 12. . more precise measurements were able to determine the small relativistic shifts (about 1%) in nuclear masses due to the binding energy. the total angular momentum of its ground state is just that of the last. Nuclei in which both Z and N are odd are harder to analyze.17. or vice versa). Most recently.40) is true for all known even-even nuclei. Today. it is generally the mass of an ionized atom (or molecule) that is measured in a mass spectrometer. which is in the 1f7/2 level (Fig.17 is not always applicable. for the total angular momenta of the nuclei 22Ne and 43Ca? Since 22Ne has Z = 10 and N = 12. it is an even-even nucleus and hence has jtot= 0. have jtot= 7/2.1 but there are only four such stable nuclei in any case.

In addition. mass spectrometers are widely used in chemical analysis to help determine the composition of molecules. When the ions enter the magnetic field B. which shows the spectrum obtained when the simple diatomic compound HCl is introduced into the ion source of a mass spectrometer.44). after initial calibration with ions of known mass. Figure 12. K  V0 e .18. they follow a circular path with radius given by (3. we find that m eB 2 R 2 . where m is the mass of the ion. Since chlorine has two stable isotopes. which is usually an excellent approximation (see Problem 12. they gain a kinetic energy. the computer can display the results directly in atomic mass units. which displays its results directly as a graph of number of particles against mass. Today.60). These occur because the electric discharge in the ion source dissociates some of the molecules into H and Cl atoms. eB Substituting (12. and the location of an electronic ion detector.42) and solving for m. When an element is analyzed in a mass spectrometer. we note that when the ions are accelerated by the potential difference V0.48) as R p . . its various isotopes are separated because of their different masses. (2) the molecule has fragments with 1 We are treating the motion as nonrelativistic. there are two large peaks with masses close to 36 u (1H35Cl) and 38 u (1H37Cl). Since m is the mass of the ion. smaller peaks can be seen at the positions one would expect for single atoms of 35Cl and 37Cl. and measurement of R lets one find m from (12. To calculate the mass. Measurement of V0. R was measured by locating the spots where the ions struck a photographic film. 12.19 shows the spectrum of masses obtained when natural lead is analyzed in a modern spectrometer. its composition could be deduced from the following observations: (1) the total molecular mass has two values. 12. the field B. as in Fig. Their momentum p is then1 p  2mK  2mV0 e . that of the bare nucleus is found by subtracting (Z– 1)me. If the HCl analyzed in the spectrometer were an unknown substance.magnetic field B. m.20. and B allows one to calculate the ions' mass. 2V0 If B and V0 are fixed. This is illustrated in Fig. me. In most early mass spectrometers. different masses will have different radii. 35Cl and 37Cl. In modern spectrometers a computer monitors the accelerating voltage V0. R. 36 and 38. as we shall see directly. the mass of the corresponding neutral atom is calculated by adding one electron mass.

masses of 35 and 37.20 and 12.2 1 (b).21 shows that each is accompanied by several smaller peaks. The mass spectrum obtained for a more complex molecule is shown in Fig. the peaks labeled 61 and 88 arise because the molecule breaks easily at the point indicated by the arrow in Fig. The masses of the various segments of this complex organic molecule provide valuable clues to its structure. 12. and (2) about 1% of the carbon atoms in any natural sample are the isotope 13C (as opposed to the dominant 12C). The details of these subsidiary peaks are additional clues to the composition of the molecule.21 for almost any substance that needs to be analyzed.2 1. As was the case with HCl. The molecule shown. For example. together with a diagram of the molecule's structure. 12. Current research in molecular structure has extended mass spectroscopy to molecules with masses of thousands of atomic mass units. 12. and (3) the ratio of the peaks at 36 and 38 (and those at 35 and 37) is the same as that of naturally occurring chlorine. Close inspection of the major peaks in Fig. producing fragments of masses 61 u and 88 u. has mass 149 u and produces the peak labeled 149 in the spectrum. These smaller peaks differ from the dominant peaks by one or two mass units and come about in two ways: (1) a molecule. Mass spectrometers with this capability are commercially available and can quickly and automatically produce spectra like those in Figs. or molecular fragment. methionine. 12. can gain or lose one or two hydrogen atoms in the ion source. and it is these fragments that produce the several peaks at lower masses. the molecule can be broken into smaller fragments in the ion source. .