Nuclear physics 

Nucleus · Nucleons (p, n) · Nuclear matter · Nuclear force · Nuclear structure · Nuclear reaction 

In nuclear physics, atomic physics, and nuclear chemistry, the nuclear shell model is a model of the atomic nucleus which uses the Pauli exclusion principle to describe the structure of the nucleus in terms of energy levels.^{[1]} The first shell model was proposed by Dmitry Ivanenko (together with E. Gapon) in 1932. The model was developed in 1949 following independent work by several physicists, most notably Eugene Paul Wigner, Maria Goeppert Mayer and J. Hans D. Jensen, who shared the 1963 Nobel Prize in Physics for their contributions.
The shell model is partly analogous to the atomic shell model which describes the arrangement of electrons in an atom, in that a filled shell results in greater stability. When adding nucleons (protons or neutrons) to a nucleus, there are certain points where the binding energy of the next nucleon is significantly less than the last one. This observation, that there are certain magic numbers of nucleons (2, 8, 20, 28, 50, 82, 126) which are more tightly bound than the next higher number, is the origin of the shell model.
The shells for protons and for neutrons are independent of each other. Therefore, "magic nuclei" exist in which one nucleon type or the other is at a magic number, and "doubly magic nuclei", where both are. Due to some variations in orbital filling, the upper magic numbers are 126 and, speculatively, 184 for neutrons but only 114 for protons, playing a role in the search for the socalled island of stability. Some semimagic numbers have been found, notably Z = 40 giving nuclear shell filling for the various elements; 16 may also be a magic number.^{[2]}
In order to get these numbers, the nuclear shell model starts from an average potential with a shape something between the square well and the harmonic oscillator. To this potential, a spin orbit term is added. Even so, the total perturbation does not coincide with experiment, and an empirical spin orbit coupling must be added with at least two or three different values of its coupling constant, depending on the nuclei being studied.
Nevertheless, the magic numbers of nucleons, as well as other properties, can be arrived at by approximating the model with a threedimensional harmonic oscillator plus a spinorbit interaction. A more realistic but also complicated potential is known as WoodsSaxon potential.
This section's tone or style may not reflect the encyclopedic tone used on Wikipedia. (September 2019) 
Consider a threedimensional harmonic oscillator. This would give, for example, in the first three levels ("l" is the angular momentum quantum number)
level n  l  m_{l}  m_{s} 

0  0  0  +1⁄2 
1⁄2  
1  1  +1  +1⁄2 
1⁄2  
0  +1⁄2  
1⁄2  
1  +1⁄2  
1⁄2  
2  0  0  +1⁄2 
1⁄2  
2  +2  +1⁄2  
1⁄2  
+1  +1⁄2  
1⁄2  
0  +1⁄2  
1⁄2  
1  +1⁄2  
1⁄2  
2  +1⁄2  
1⁄2 
We can imagine ourselves building a nucleus by adding protons and neutrons. These will always fill the lowest available level. Thus the first two protons fill level zero, the next six protons fill level one, and so on. As with electrons in the periodic table, protons in the outermost shell will be relatively loosely bound to the nucleus if there are only few protons in that shell, because they are farthest from the center of the nucleus. Therefore, nuclei which have a full outer proton shell will have a higher binding energy than other nuclei with a similar total number of protons. All this is true for neutrons as well.
This means that the magic numbers are expected to be those in which all occupied shells are full. We see that for the first two numbers we get 2 (level 0 full) and 8 (levels 0 and 1 full), in accord with experiment. However the full set of magic numbers does not turn out correctly. These can be computed as follows:
In particular, the first six shells are:
where for every l there are 2l+1 different values of m_{l} and 2 values of m_{s}, giving a total of 4l+2 states for every specific level.
These numbers are twice the values of triangular numbers from the Pascal Triangle: 1, 3, 6, 10, 15, 21, ....
We next include a spinorbit interaction. First we have to describe the system by the quantum numbers j, m_{j} and parity instead of l, m_{l} and m_{s}, as in the hydrogenlike atom. Since every even level includes only even values of l, it includes only states of even (positive) parity. Similarly, every odd level includes only states of odd (negative) parity. Thus we can ignore parity in counting states. The first six shells, described by the new quantum numbers, are
where for every j there are 2j+1 different states from different values of m_{j}.
Due to the spinorbit interaction the energies of states of the same level but with different j will no longer be identical. This is because in the original quantum numbers, when is parallel to , the interaction energy is positive; and in this case j = l + s = l + 1⁄2. When is antiparallel to (i.e. aligned oppositely), the interaction energy is negative, and in this case j=ls=l1⁄2. Furthermore, the strength of the interaction is roughly proportional to l.
For example, consider the states at level 4:
The harmonic oscillator potential grows infinitely as the distance from the center r goes to infinity. A more realistic potential, such as WoodsSaxon potential, would approach a constant at this limit. One main consequence is that the average radius of nucleons' orbits would be larger in a realistic potential; This leads to a reduced term in the Laplace operator of the Hamiltonian. Another main difference is that orbits with high average radii, such as those with high n or high l, will have a lower energy than in a harmonic oscillator potential. Both effects lead to a reduction in the energy levels of high l orbits.
Together with the spinorbit interaction, and for appropriate magnitudes of both effects, one is led to the following qualitative picture: At all levels, the highest j states have their energies shifted downwards, especially for high n (where the highest j is high). This is both due to the negative spinorbit interaction energy and to the reduction in energy resulting from deforming the potential to a more realistic one. The secondtohighest j states, on the contrary, have their energy shifted up by the first effect and down by the second effect, leading to a small overall shift. The shifts in the energy of the highest j states can thus bring the energy of states of one level to be closer to the energy of states of a lower level. The "shells" of the shell model are then no longer identical to the levels denoted by n, and the magic numbers are changed.
We may then suppose that the highest j states for n = 3 have an intermediate energy between the average energies of n = 2 and n = 3, and suppose that the highest j states for larger n (at least up to n = 7) have an energy closer to the average energy of n1. Then we get the following shells (see the figure)
and so on.
Note that the numbers of states after the 4th shell are doubled triangular numbers plus two. Spinorbit coupling causes socalled 'intruder levels' to drop down from the next higher shell into the structure of the previous shell. The sizes of the intruders are such that the resulting shell sizes are themselves increased to the very next higher doubled triangular numbers from those of the harmonic oscillator. For example, 1f2p has 20 nucleons, and spinorbit coupling adds 1g9/2 (10 nucleons) leading to a new shell with 30 nucleons. 1g2d3s has 30 nucleons, and addition of intruder 1h11/2 (12 nucleons) yields a new shell size of 42, and so on.
The magic numbers are then
and so on. This gives all the observed magic numbers, and also predicts a new one (the socalled island of stability) at the value of 184 (for protons, the magic number 126 has not been observed yet, and more complicated theoretical considerations predict the magic number to be 114 instead).
Another way to predict magic (and semimagic) numbers is by laying out the idealized filling order (with spinorbit splitting but energy levels not overlapping). For consistency s is split into j = 1/2 and j = 1/2 components with 2 and 0 members respectively. Taking leftmost and rightmost total counts within sequences marked bounded by / here gives the magic and semimagic numbers.
The rightmost predicted magic numbers of each pair within the quartets bisected by / are double tetrahedral numbers from the Pascal Triangle: 2, 8, 20, 40, 70, 112, 168, 240 are 2x 1, 4, 10, 20, 35, 56, 84, 120, ..., and the leftmost members of the pairs differ from the rightmost by double triangular numbers: 2  2 = 0, 8  6 = 2, 20  14 = 6, 40  28 = 12, 70  50 = 20, 112  82 = 30, 168  126 = 42, 240  184 = 56, where 0, 2, 6, 12, 20, 30, 42, 56, ... are 2 × 0, 1, 3, 6, 10, 15, 21, 28, ... .
This model also predicts or explains with some success other properties of nuclei, in particular spin and parity of nuclei ground states, and to some extent their excited states as well. Take (oxygen17) as an example: Its nucleus has eight protons filling the three first proton "shells", eight neutrons filling the three first neutron "shells", and one extra neutron. All protons in a complete proton shell have zero total angular momentum, since their angular momenta cancel each other. The same is true for neutrons. All protons in the same level (n) have the same parity (either +1 or 1), and since the parity of a pair of particles is the product of their parities, an even number of protons from the same level (n) will have +1 parity. Thus the total angular momentum of the eight protons and the first eight neutrons is zero, and their total parity is +1. This means that the spin (i.e. angular momentum) of the nucleus, as well as its parity, are fully determined by that of the ninth neutron. This one is in the first (i.e. lowest energy) state of the 4th shell, which is a dshell (l = 2), and since , this gives the nucleus an overall parity of +1. This 4th dshell has a j = 5⁄2, thus the nucleus of is expected to have positive parity and total angular momentum 5⁄2, which indeed it has.
The rules for the ordering of the nucleus shells are similar to Hund's Rules of the atomic shells, however, unlike its use in atomic physics the completion of a shell is not signified by reaching the next n, as such the shell model cannot accurately predict the order of excited nuclei states, though it is very successful in predicting the ground states. The order of the first few terms are listed as follows: 1s, 1p3⁄2, 1p1⁄2, 1d5⁄2, 2s, 1d3⁄2... For further clarification on the notation refer to the article on the RussellSaunders term symbol.
For nuclei farther from the magic numbers one must add the assumption that due to the relation between the strong nuclear force and angular momentum, protons or neutrons with the same n tend to form pairs of opposite angular momenta. Therefore, a nucleus with an even number of protons and an even number of neutrons has 0 spin and positive parity. A nucleus with an even number of protons and an odd number of neutrons (or vice versa) has the parity of the last neutron (or proton), and the spin equal to the total angular momentum of this neutron (or proton). By "last" we mean the properties coming from the highest energy level.
In the case of a nucleus with an odd number of protons and an odd number of neutrons, one must consider the total angular momentum and parity of both the last neutron and the last proton. The nucleus parity will be a product of theirs, while the nucleus spin will be one of the possible results of the sum of their angular momenta (with other possible results being excited states of the nucleus).
The ordering of angular momentum levels within each shell is according to the principles described above  due to spinorbit interaction, with high angular momentum states having their energies shifted downwards due to the deformation of the potential (i.e. moving from a harmonic oscillator potential to a more realistic one). For nucleon pairs, however, it is often energetically favorable to be at high angular momentum, even if its energy level for a single nucleon would be higher. This is due to the relation between angular momentum and the strong nuclear force.
Nuclear magnetic moment is partly predicted by this simple version of the shell model. The magnetic moment is calculated through j, l and s of the "last" nucleon, but nuclei are not in states of well defined l and s. Furthermore, for oddodd nuclei, one has to consider the two "last" nucleons, as in deuterium. Therefore, one gets several possible answers for the nuclear magnetic moment, one for each possible combined l and s state, and the real state of the nucleus is a superposition of them. Thus the real (measured) nuclear magnetic moment is somewhere in between the possible answers.
The electric dipole of a nucleus is always zero, because its ground state has a definite parity, so its matter density (, where is the wavefunction) is always invariant under parity. This is usually the situations with the atomic electric dipole as well.
Higher electric and magnetic multipole moments cannot be predicted by this simple version of the shell model, for the reasons similar to those in the case of deuterium.
For nuclei having two or more valence nucleons (i.e. nucleons outside a closed shell) a residual twobody interaction must be added. This residual term comes from the part of the internucleon interaction not included in the approximative average potential. Through this inclusion different shell configurations are mixed and the energy degeneracy of states corresponding to the same configuration is broken.^{[4]}^{[5]}
These residual interactions are incorporated through shell model calculations in a truncated model space (or valence space). This space is spanned by a basis of manyparticle states where only singleparticle states in the model space are active. The Schrödinger equation is solved in this basis, using an effective Hamiltonian specifically suited for the model space. This Hamiltonian is different from the one of free nucleons as it among other things has to compensate for excluded configurations.^{[5]}
One can do away with the average potential approximation entirely by extending the model space to the previously inert core and treat all singleparticle states up to the model space truncation as active. This forms the basis of the nocore shell model, which is an ab initio method. It is necessary to include a threebody interaction in such calculations to achieve agreement with experiments.^{[6]}
In 1953, the first experimental examples were found of rotational bands in nuclei, with their energy levels following the same J(J+1) pattern of energies as in rotating molecules. Quantum mechanically, it is impossible to have a collective rotation of a sphere, so this implied that the shape of these nuclei was nonspherical. In principle, these rotational states could have been described as coherent superpositions of particlehole excitations in the basis consisting of singleparticle states of the spherical potential. But in reality, the description of these states in this manner is intractable, due to the large number of valence particlesand this intractability was even greater in the 1950s, when computing power was extremely rudimentary. For these reasons, Aage Bohr, Ben Mottelson, and Sven Gösta Nilsson constructed models in which the potential was deformed into an ellipsoidal shape. The first successful model of this type is the one now known as the Nilsson model. It is essentially the harmonic oscillator model described in this article, but with anisotropy added, so that the oscillator frequencies along the three Cartesian axes are not all the same. Typically the shape is a prolate ellipsoid, with the axis of symmetry taken to be z. Because the potential is not spherically symmetric, the singleparticle states are not states of good angular momentum J. However, a Lagrange multiplier , known as a "cranking" term, can be added to the Hamiltonian. Usually the angular frequency vector ? is taken to be perpendicular to the symmetry axis, although tiltedaxis cranking can also be considered. Filling the singleparticle states up to the Fermi level then produces states whose expected angular momentum along the cranking axis is the desired value.
Igal Talmi developed a method to obtain the information from experimental data and use it to calculate and predict energies which have not been measured. This method has been successfully used by many nuclear physicists and has led to deeper understanding of nuclear structure. The theory which gives a good description of these properties was developed. This description turned out to furnish the shell model basis of the elegant and successful interacting boson model.
A model derived from the nuclear shell model is the alpha particle model developed by Henry Margenau, Edward Teller, J. K. Pering, T. H. Skyrme, also sometimes called the Skyrme model.^{[7]}^{[8]} Note, however, that the Skyrme model is usually taken to be a model of the nucleon itself, as a "cloud" of mesons (pions), rather than as a model of the nucleus as a "cloud" of alpha particles.