Nuclear Shell Model

Definition

The nuclear shell model describes the nucleus the way the periodic table describes the atom: as a collection of particles filling discrete, quantized energy levels — shells. The particles here are nucleons (protons and neutrons), and the headline result is that a nucleus is unusually stable whenever its proton count or neutron count exactly fills a set of shells.

Those special counts are the magic numbers:

2, 8, 20, 28, 50, 82, 126

A nucleus where both Z (protons) and N (neutrons) are magic is called doubly magic and is the nuclear equivalent of a noble gas — chemically and dynamically inert, tightly bound, hard to excite. Lead-208 (Z=82, N=126) is the heaviest stable doubly-magic nucleus known.

How it works

The central simplification is bold: instead of tracking the messy two-body forces between every pair of nucleons, you assume each nucleon moves independently in an average potential created by all the others — a self-consistent mean field. This is the same trick that makes atomic physics tractable, where each electron sees an averaged Coulomb potential.

Solving the quantum problem in that potential gives a ladder of energy levels, each labelled by quantum numbers. Two rules decide how nucleons populate them:

• The Pauli exclusion principle. Protons and neutrons are both spin-½ fermions, so no two identical nucleons can share the same quantum state. They stack into levels from the bottom up, just like electrons. (See Pauli Exclusion.)
• Two independent ladders. Protons fill their own set of levels and neutrons fill theirs, because they are distinguishable particles. That is why magic numbers apply to Z and N separately.

Where the model gets interesting is in the shape of the potential. Try the obvious candidates and the first cracks appear:

• A 3D harmonic oscillator (V ∝ r²) gives shell closures at 2, 8, 20, 40, 70, 112 — the first three are right, the rest are wrong.
• An infinite square well gives 2, 8, 18, 20, 34, 40, 58 — also wrong above 20.
• A realistic Woods-Saxon potential (a smoothly rounded well matching the nuclear density profile) sits between these, and still misses 28, 50, 82, 126.

No reasonable shape of the central potential alone reproduces the higher magic numbers. For nearly two decades this was the model's fatal flaw.

The spin-orbit fix (Mayer-Jensen, Nobel 1963)

In 1949 Maria Goeppert Mayer in Chicago and, independently, Otto Haxel, Hans Jensen and Hans Suess in Germany found the missing ingredient: a strong spin-orbit coupling. They added a term to the Hamiltonian proportional to the dot product of a nucleon's orbital angular momentum l and its spin s:

V(r) = V_central(r) + V_so(r) · (l · s)

This splits every level with orbital angular momentum l > 0 into two sublevels by total angular momentum j:

j = l + 1/2   →  pushed DOWN (lower energy)
j = l - 1/2   →  pushed UP   (higher energy)
splitting ∝ (2l + 1)        →  bigger for high-l levels

Two features make this decisive. First, the sign is the opposite of the weak atomic spin-orbit effect — in nuclei the higher-j (aligned) level drops. Second, the splitting is large, scaling with (2l+1), so a high-l intruder level can plunge all the way down into the shell below it. When you re-count the levels after this rearrangement, the gaps land exactly at 28, 50, 82, and 126. The model snapped into agreement with the data.

It was a startling result for the time — Goeppert Mayer's male colleagues nicknamed her "the Madonna of the onion" for the layered shell picture — and it earned Mayer and Jensen a share of the 1963 Nobel Prize in Physics (the other half went to Eugene Wigner). Goeppert Mayer was only the second woman to win the physics Nobel, after Marie Curie.

Worked example: filling up to oxygen-16

Let's fill the neutron ladder and watch the magic numbers fall out. Each level holds 2j+1 nucleons (the number of m-states). After spin-orbit splitting the order of the lowest levels is:

Level (n l j)	Capacity 2j+1	Running total	Gap after?
1s1/2	2	2	MAGIC — big gap
1p3/2	4	6	small
1p1/2	2	8	MAGIC — big gap
1d5/2	6	14	small
2s1/2	2	16	small
1d3/2	4	20	MAGIC — big gap
1f7/2	8	28	MAGIC — spin-orbit intruder!

Take oxygen-16: 8 protons and 8 neutrons. Both fill cleanly through the 1p_1/2 level — both are at the magic number 8. It is doubly magic. The payoff is measurable: oxygen-16's first excited state sits at 6.05 MeV, an enormous gap, whereas its neighbour oxygen-18 has a first excited state near 1.98 MeV. That factor-of-three difference is the energy gap above a closed shell made visible.

Now add one neutron to make oxygen-17. The eight paired neutrons contribute nothing; the lone ninth neutron must go into the next level, 1d_5/2. The shell model predicts the ground state therefore has spin-parity 5/2+ — and that is exactly what experiment measures. One unpaired nucleon, one prediction, confirmed.

Reading the binding-energy curve

The clearest fingerprint of shells is in two-nucleon separation energies — the energy to remove a pair of neutrons (S_2n) or protons. Plot S_2n against N and it falls smoothly, then drops sharply right after each magic number, because the next pair has to sit above a large shell gap. The same magic numbers show up as:

• Peaks in natural abundance. Tin (Z=50) has 10 stable isotopes — more than any other element — because 50 is a magic proton number.
• Spikes in first-excited-state energy. Hard to excite a closed shell.
• Kinks in nuclear radii and alpha/beta decay energies.
• Extra binding of a few MeV beyond the smooth liquid-drop prediction — the "shell correction".

Nucleus	Z	N	Magic in	Signature
Helium-4 (⁴He)	2	2	both — doubly magic	Binding 7.07 MeV/nucleon, no bound excited state
Oxygen-16 (¹⁶O)	8	8	both — doubly magic	First excited state at 6.05 MeV
Calcium-40 (⁴⁰Ca)	20	20	both — doubly magic	Spherical, stable, high binding
Calcium-48 (⁴⁸Ca)	20	28	both — doubly magic	Stable despite N/Z = 1.4; very long-lived
Tin-100 / Tin-132	50	50 / 82	both — doubly magic	Anchors of the tin isotope chain
Lead-208 (²⁰⁸Pb)	82	126	both — doubly magic	Heaviest stable doubly-magic nucleus

Variants and regimes

• Independent-particle (extreme single-particle) model. The simplest version: ignore residual interactions, fill the levels, read off spins. Works astonishingly well for nuclei one nucleon away from a closed shell.
• Interacting shell model (configuration interaction). Restore the residual forces between valence nucleons and diagonalize within a model space. This is the workhorse of modern ab initio nuclear structure — accurate but limited by combinatorial explosion of configurations.
• Nilsson (deformed) shell model. Far from closed shells, nuclei deform into ellipsoids. The spherical levels split further with deformation, and rotational bands appear. Magic numbers themselves can shift in exotic nuclei.
• Macroscopic-microscopic models. Combine the smooth liquid-drop energy with shell corrections — the basis of Strutinsky's method and predictions for superheavy elements.
• The island of stability. Extrapolating the shell gaps points to a doubly-magic superheavy nucleus near Z=114 (or 120/126) and N=184, where fresh shell closure should lengthen otherwise vanishing half-lives.

Common pitfalls and misconceptions

• "Nucleons orbit the centre like planets." There is no central body — nucleons collectively generate the very potential they move in. The shells are an emergent, self-consistent mean field, not orbits around a nucleus-within-the-nucleus.
• "Magic numbers match the noble gases." They do not. Atomic shell closures are 2, 10, 18, 36, 54, 86; nuclear ones are 2, 8, 20, 28, 50, 82, 126. The potentials and the spin-orbit strength are different.
• "One ladder for all nucleons." Protons and neutrons fill separate ladders. Calcium-48 is doubly magic precisely because Z=20 and N=28 are both magic on their own ladders.
• "The model works everywhere." It is excellent near closed shells and fails for strongly deformed mid-shell nuclei (rare earths, actinides), where collective rotation and vibration dominate. There you need the Nilsson or collective models.
• "Spin-orbit is a minor correction." In atoms it is. In nuclei it is the load-bearing term — without it you simply cannot get 28, 50, 82, or 126. Half the magic numbers exist only because of it.
• "Magic numbers are eternal." In neutron-rich exotic isotopes far from stability, traditional shell gaps can disappear and new ones (e.g. N=16, N=34) appear — "shell evolution" driven by the tensor force.

Applications

• Predicting stability and decay. The shell model tells you which isotopes will be long-lived (near magic numbers) and which spin-parities to expect — essential input for nuclear data tables and reactor physics.
• Nucleosynthesis and astrophysics. The rapid neutron-capture (r-process) in supernovae and neutron-star mergers piles up at the magic neutron numbers N=50, 82, 126, producing the abundance peaks we see in the solar system at mass ~80, ~130, ~195.
• Superheavy element hunting. Every search for new elements is steered by the predicted island of stability around N=184.
• Medical and detector isotopes. Choosing isotopes with favourable spins and lifetimes for imaging and dosimetry leans on shell-model spin/parity assignments.
• Fundamental tests. Doubly-magic nuclei like ⁴⁸Ca and ¹³⁶Xe are prime candidates for neutrinoless double-beta-decay experiments because their structure is cleanly calculable.

Why it works: a physicist's accounting

Why should a mean-field, independent-particle picture work at all, when the nucleus is a dense soup of strongly interacting nucleons mere femtometres apart? The answer is the Pauli principle again. Most low-energy two-nucleon collisions would scatter particles into already-occupied states — forbidden. With those collisions blocked, a nucleon travels a long mean free path between effective interactions, behaving almost as if it were free in an average field. The strong residual force largely renormalizes into the shape of that field rather than randomizing trajectories.

The numbers tell the story. The spin-orbit splitting of the high-l 1f_7/2 – 1f_5/2 pair in the mass-50 region is several MeV — comparable to a whole oscillator shell spacing (ℏω ≈ 41·A^−1/3 MeV, about 11 MeV for A=50). That is why a single intruder level can migrate down a full shell and create a brand-new gap. The model is not a loose analogy to atoms; it is a quantitative bookkeeping of angular momentum, parity, and energy that reproduces ground-state spins of hundreds of nuclei, the exact location of all seven magic numbers, and the few-MeV binding bonuses of every doubly-magic nucleus from helium-4 to lead-208.