Angular Momentum Coupling Schemes

An angular momentum coupling scheme is a choice of which angular momenta to add first and which commuting operators to use as labels. It is not a new Hilbert space. It is a basis and labeling strategy adapted to a Hamiltonian.

The basic rule is simple: couple first the angular momenta whose interaction is most important. If a Hamiltonian is dominated by electrostatic terms, spin–orbit terms, hyperfine terms, rotational terms, or external fields, different labels may be natural.

Canonical Split

This page is an overview of coupling schemes. The operator definition of total angular momentum is in Total Angular Momentum. The one-particle case $\mathbf J=\mathbf L+\mathbf S$ is in Addition of Orbital and Spin Angular Momentum. The basis dictionary is in Coupled and Uncoupled Bases. Numerical change-of-basis coefficients are in Clebsch–Gordan Coefficients. Recoupling between different orders is treated in Recoupling and Wigner Symbols.

Detailed atomic fine structure, hyperfine spectroscopy, and molecular Hund-case spectroscopy belong to dedicated atomic and molecular pages. Here the goal is the angular-momentum logic.

Coupling Scheme as a Good-Label Choice

A useful coupling scheme usually corresponds to a convenient complete or nearly complete set of commuting observables. For two angular momenta, the coupled basis diagonalizes

$$J_1^2,\quad J_2^2,\quad J^2,\quad J_z,$$

where

$$\mathbf J=\mathbf J_1+\mathbf J_2.$$

For several angular momenta, there are many possible intermediate sums. For example,

$$\mathbf J = \mathbf J_1+\mathbf J_2+\mathbf J_3$$

can be built by first forming

$$\mathbf J_{12}=\mathbf J_1+\mathbf J_2$$

or by first forming

$$\mathbf J_{23}=\mathbf J_2+\mathbf J_3.$$

Both are legitimate bases of the same space. The Hamiltonian decides which one is natural. A label is a good quantum number only when the corresponding operator is conserved or approximately conserved in the regime being described.

Hierarchy of Hamiltonian Terms

In spectroscopy one often has a hierarchy such as

$$H = H_{\mathrm{central}} + H_{\mathrm{ee}} + H_{\mathrm{SO}} + H_{\mathrm{hf}} + H_Z + \cdots .$$

The symbols here are schematic: central potential, electron-electron interactions, spin–orbit coupling, hyperfine coupling, Zeeman coupling, and smaller corrections. The best labels depend on the relative size of these terms.

If one term is much larger than the others, diagonalize that term first and treat smaller terms as perturbations. If two terms are comparable, simple labels may become approximate and one must diagonalize in a larger basis. This is the logic behind “intermediate coupling.”

LS Coupling

In LS coupling, also called Russell–Saunders coupling, one first adds all orbital angular momenta to total $\mathbf L$ and all spins to total $\mathbf S$:

$$\mathbf L=\sum_i\mathbf l_i, \qquad \mathbf S=\sum_i\mathbf s_i.$$

Then one adds

$$\mathbf J=\mathbf L+\mathbf S.$$

The corresponding labels are often written schematically as

$$\lvert \alpha,L,S;J,M\rangle,$$

where $\alpha$ denotes additional labels needed to distinguish states with the same $L,S,J,M$.

LS coupling is useful when spin-independent interactions organize the electron configuration more strongly than the spin–orbit interaction. This is often a good first approximation for light atoms, where spin–orbit effects are relatively small. In that regime $L$ and $S$ are useful approximate labels, and spin–orbit coupling splits levels according to $J$.

The familiar spectroscopic term symbol

$${}^{2S+1}L_J$$

belongs to this scheme. Atomic Term Symbols teaches the full notation and its atomic examples; this page records the angular-momentum coupling order.

Slater Determinants in Atoms shows how uncoupled atomic determinants are transformed into an $LS$-coupled configuration-state-function basis.

LS Coupling develops the atomic energy-scale criterion, isolated-term fine structure, Landé diagnostics, and intermediate-coupling failure modes.

jj Coupling

In jj coupling, each electron’s orbital and spin angular momenta are combined first:

$$\mathbf j_i=\mathbf l_i+\mathbf s_i.$$

Then the individual $\mathbf j_i$ are added to total $\mathbf J$:

$$\mathbf J=\sum_i\mathbf j_i.$$

A schematic two-electron basis is

$$\lvert (l_1s_1)j_1,(l_2s_2)j_2;J,M\rangle.$$

This scheme is natural when individual spin–orbit interactions are strong compared with residual interactions that would first organize total $\mathbf L$ and total $\mathbf S$. It is especially important in heavy atoms, where relativistic effects make spin–orbit splitting large.

The labels $L$ and $S$ are then no longer generally good labels. The total $\mathbf J$ may remain good if the Hamiltonian is still rotationally invariant.

jj Coupling develops the atomic relativistic-subshell basis, equivalent-electron Pauli restrictions, LS-to-jj $9j$ recoupling, and spectroscopic diagnostics for this limit.

Intermediate Coupling

Real systems often lie between the ideal LS and jj limits. Then neither set of labels is exact. One may expand an eigenstate as

$$\lvert\Psi;J,M\rangle = \sum_{\alpha,L,S} c_{\alpha LS} \lvert \alpha,L,S;J,M\rangle$$

or in a jj-coupled basis, and diagonalize the Hamiltonian numerically inside the fixed-$J,M$ sector.

The phrase “intermediate coupling” should not be used as a mystery label. It means the Hamiltonian mixes basis states that would be separately labeled in one limiting scheme. Which labels remain useful depends on the size of the mixing; see Approximate Symmetry.

Hyperfine Coupling

Hyperfine coupling adds nuclear spin. If the electronic angular momentum is $\mathbf J$ and the nuclear spin is $\mathbf I$, the total atomic angular momentum is

$$\mathbf F=\mathbf J+\mathbf I.$$

The coupled labels are

$$\lvert J,I;F,m_F\rangle.$$

For fixed $J$ and $I$, the allowed values are

$$F=|J-I|,|J-I|+1,\ldots,J+I.$$

Hyperfine labels are natural when the hyperfine interaction is resolved and external fields are weak enough that $\mathbf F$ remains useful. In strong magnetic fields, the Zeeman interaction can compete with or dominate hyperfine coupling, and uncoupled labels such as $m_J,m_I$ may become more useful. Zeeman Effect as a Perturbation Example applies this logic to weak-field, intermediate, and Paschen–Back regimes, while Zeeman Effect Revisited supplies the historical context.

Molecular Coupling Preview

Molecules add another layer because electronic orbital motion, electronic spin, vibration, and molecular rotation can all carry angular momentum. Depending on the molecule and energy scale, useful labels may involve:

• projection of electronic orbital angular momentum on the molecular axis;
• projection of spin on the molecular axis;
• rotational angular momentum of the nuclei;
• total angular momentum including electronic and rotational parts;
• nuclear spin when hyperfine structure is resolved.

The standard molecular Hund cases are coupling schemes for these different hierarchies. They are not different laws of quantum mechanics. They are different approximate bases adapted to different molecular Hamiltonians.

For the simplest rotor side of the story, see Rotational Spectra. Detailed molecular angular momentum requires additional molecular structure beyond this overview.

Recoupling Between Schemes

Different coupling schemes are related by unitary changes of basis. For three angular momenta, one common recoupling problem is

$$\left| (j_1j_2)j_{12},j_3;J,M \right\rangle \quad \longleftrightarrow \quad \left| j_1,(j_2j_3)j_{23};J,M \right\rangle.$$

The transformation coefficients are expressed using Wigner $6j$ symbols. For four angular momenta, $9j$ symbols appear naturally. The important point is conceptual: changing the coupling order is not changing the state space. It is changing the basis used to describe the same state space.

Choosing a Scheme

Use this decision sequence:

1. Identify all angular momenta in the model.
2. Identify the dominant Hamiltonian terms.
3. Choose the coupling order that diagonalizes or nearly diagonalizes the dominant terms.
4. Check which labels commute with the Hamiltonian or are approximately conserved.
5. Treat weaker terms as perturbations or diagonalize them inside fixed symmetry sectors.
6. Translate between schemes with Clebsch–Gordan, $6j$, or $9j$ coefficients when needed.

The hierarchy can change with external fields, atomic number, molecular species, or energy scale. A good coupling scheme is therefore a model statement, not a permanent property of the system.

Common Mistakes

• Treating LS and jj coupling as different physical Hilbert spaces rather than different bases.
• Calling $L$ and $S$ good labels after adding a strong spin–orbit term without checking commutators.
• Using hyperfine labels $F,m_F$ in a strong-field regime where $m_J,m_I$ are more nearly conserved.
• Forgetting extra labels such as $\alpha$ when several states share the same angular-momentum quantum numbers.
• Assuming “intermediate coupling” is self-explanatory without specifying which Hamiltonian terms are competing.
• Confusing molecular Hund-case labels with exact quantum numbers in regimes where the corresponding couplings are only approximate.

Cross-Links

• Atomic Term Symbols
• Slater Determinants in Atoms
• LS Coupling
• jj Coupling
• Total Angular Momentum
• Coupled and Uncoupled Bases
• Clebsch–Gordan Coefficients
• Clebsch–Gordan Tables and Conventions
• Addition of Orbital and Spin Angular Momentum
• Recoupling and Wigner Symbols
• Spin–Orbit Coupling
• Simultaneous Eigenstates and Good Quantum Numbers
• Approximate Symmetry
• Degeneracy Lifting
• Zeeman Effect Revisited
• Zeeman Effect as a Perturbation Example
• Rotational Spectra
• Addition of Angular Momentum Formula Card

References

• E. U. Condon and G. H. Shortley, The Theory of Atomic Spectra, Cambridge University Press, 1935.
• A. R. Edmonds, Angular Momentum in Quantum Mechanics, Princeton University Press, 1957.
• R. N. Zare, Angular Momentum: Understanding Spatial Aspects in Chemistry and Physics, Wiley, 1988.
• D. A. Varshalovich, A. N. Moskalev, and V. K. Khersonskii, Quantum Theory of Angular Momentum, World Scientific, 1988.
• J. J. Sakurai and J. Napolitano, Modern Quantum Mechanics, 3rd ed., Cambridge University Press, 2020.
• J. M. Brown and A. Carrington, Rotational Spectroscopy of Diatomic Molecules, Cambridge University Press, 2003.

Exercises

1. A single electron has orbital angular momentum $\ell=1$ and spin $s=1/2$. What total $j$ values are possible?

Solution

Add $\ell=1$ and $s=1/2$:

$$j = \left|1-\frac12\right|, \ldots, 1+\frac12.$$

Thus

$$j=\frac12,\frac32.$$

2. In a light atom, electrostatic interactions strongly organize the electron configuration and spin–orbit coupling is a smaller correction. Which limiting scheme is usually the better first label system: LS or jj?

Solution

LS coupling is the better first label system. One first forms total $\mathbf L$ and total $\mathbf S$ from the electrons, then treats spin–orbit coupling as a smaller interaction that splits levels according to total $\mathbf J=\mathbf L+\mathbf S$.

3. Suppose an atom has electronic $J=1/2$ and nuclear spin $I=3/2$. List the allowed hyperfine $F$ values and check dimensions.

Solution

The allowed values are

$$F = \left|\frac12-\frac32\right|, \ldots, \frac12+\frac32,$$

so

$$F=1,2.$$

The original product dimension is

$$(2J+1)(2I+1) = 2\cdot4 = 8.$$

The coupled dimensions are

$$(2\cdot1+1)+(2\cdot2+1) = 3+5 = 8.$$

4. Why is changing from LS coupling to jj coupling not by itself an approximation?

Solution

Both LS and jj coupling define bases of the same angular-momentum Hilbert space. Changing between them is a unitary basis change, expressible through recoupling coefficients. The approximation enters when one treats one set of labels as good or nearly good because a particular hierarchy of Hamiltonian terms is assumed.