Antisymmetry and the Slater Determinant

Every production-grade electronic structure code — Gaussian, ORCA, Q-Chem, Psi4, VASP, CP2K — represents its reference wavefunction as one or more Slater determinants, whether the ultimate method is Hartree–Fock itself, post-Hartree–Fock correlation methods (MP2, coupled cluster, configuration interaction), or Kohn–Sham density functional theory, whose auxiliary orbitals are likewise packaged into a determinant to guarantee the correct kinetic energy and exchange behavior of a non-interacting reference system. Multireference methods (CASSCF, CASPT2), essential for describing bond-breaking, transition metal spin states, and photochemical excited states, are built as linear combinations of many Slater determinants spanning an active space — the determinant remains the indivisible unit even when a single one is inadequate. The Fermi hole concept underlies exchange-correlation functional design in DFT: functionals like B3LYP and PBE0 mix in a fraction of "exact exchange" computed exactly as the $K_{ij}$ integral derived here, because pure density functionals systematically underestimate the energetic benefit of same-spin electron avoidance. In materials science, antisymmetry drives the exchange interaction responsible for ferromagnetism (the Heisenberg exchange coupling in solids traces back to exactly the same $K_{ij}$-type integral), explaining why iron is magnetic and why certain molecular magnets retain spin alignment. In spectroscopy, Hund's-rule ordering of atomic and molecular term energies — essential for interpreting UV-Vis and EPR spectra of transition metal complexes and lanthanide ions — is a direct macroscopic signature of the exchange energy quantified in Example 2. Even qualitative organic chemistry concepts, like the preference for triplet diradicals or the electronic structure of high-spin versus low-spin transition metal complexes in bioinorganic chemistry (e.g., iron centers in hemoglobin and cytochromes), trace their energetic ordering to this same determinant-level exchange bookkeeping.

Ask the tutor: How do the Slater–Condon rules change when the two determinants differ by two spin-orbitals instead of one, and why does that case introduce both a Coulomb and an exchange term?

Electrons are indistinguishable fermions: there is no experiment, even in principle, that can track "electron 1" as opposed to "electron 2" through a molecule. Quantum mechanics encodes this indistinguishability through the behavior of the wavefunction under particle exchange. Define the permutation operator $\hat P_{ij}$, which swaps the full coordinates (spatial and spin) of particles $i$ and $j$ inside the wavefunction. Because the physical probability density $|\Psi|^2$ must be unchanged by any such relabeling, we need

Applying $\hat P_{ij}$ twice returns the original labeling, so $\lambda^2 = 1$, giving $\lambda = \pm 1$. Nature splits particles into two exhaustive classes: bosons, with $\lambda = +1$ (symmetric wavefunctions), and fermions, with $\lambda = -1$ (antisymmetric wavefunctions). Electrons, having half-integer spin, are fermions — this is the content of the Pauli exclusion principle in its most general form, and it is an experimentally verified postulate of relativistic quantum field theory (the spin–statistics theorem), not something derivable from non-relativistic quantum mechanics alone. Now consider building an $N$-electron wavefunction out of one-electron spin-orbitals $\chi_i(\mathbf{x})$, where $\mathbf{x}$ bundles spatial coordinates $\mathbf{r}$ and spin coordinate $\omega$. The naive Hartree product

assigns electron 1 to orbital $\chi_1$, electron 2 to orbital $\chi_2$, and so on — a labeling scheme that silently assumes electrons are distinguishable billiard balls. Swapping electrons 1 and 2 gives $\chi_1(2)\chi_2(1)\chi_3(3)\cdots$, which is neither $+\Psi^{HP}$ nor $-\Psi^{HP}$ in general. The Hartree product has no definite exchange symmetry, so it cannot be the wavefunction of a system of identical fermions — full stop, regardless of how good the individual orbitals are. The remedy is to antisymmetrize by hand: sum over all $N!$ permutations of electrons among orbitals, with a sign $(-1)^p$ for each permutation $p$ (even permutations get $+$, odd get $-$). This sum is exactly the determinant of the matrix whose $(i,j)$ entry is $\chi_j(i)$ — orbital $j$ evaluated for electron $i$:

Rows index electrons, columns index spin-orbitals (or vice versa — the convention only affects an overall sign, never physics). Two immediate determinant properties carry the entire physical content:

• Row swap ⇒ sign flip. Interchanging rows $i$ and $j$ of a determinant multiplies it by $-1$. Rows correspond to electron labels, so swapping electrons $i$ and $j$ automatically antisymmetrizes $\Psi$. This is exactly the fermion exchange condition $\hat P_{ij}\Psi = -\Psi$.
• Repeated column ⇒ determinant vanishes. If two spin-orbitals are identical, $\chi_k = \chi_l$, two columns of the matrix are identical, and the determinant is exactly zero. No two electrons can occupy the same spin-orbital — this is the Pauli exclusion principle, derived rather than assumed.

The figure below shows the $2\times2$ case explicitly — the smallest nontrivial Slater determinant, for two electrons in orbitals $\chi_1$ and $\chi_2$.

A subtlety worth making explicit: antisymmetry applies to the complete coordinate, space and spin together. Two electrons can share the same spatial orbital $\phi(\mathbf r)$ as long as they carry opposite spin, because then the spin-orbitals $\phi(\mathbf r)\alpha(\omega)$ and $\phi(\mathbf r)\beta(\omega)$ are still distinct columns — this is exactly the familiar picture of two electrons paired in one orbital, one spin-up and one spin-down, that underlies closed-shell chemistry. The second physical consequence of antisymmetry is more subtle and often underappreciated: even in the complete absence of Coulomb repulsion, antisymmetry alone forces electrons of the same spin to avoid each other spatially. Expand $|\Psi(1,2)|^2$ for the two-electron determinant above with same-spin orbitals $\phi_1(\mathbf r)\alpha$ and $\phi_2(\mathbf r)\alpha$; integrating out spin leaves a spatial pair density containing a cross term $-\phi_1^*(\mathbf r_1)\phi_2^*(\mathbf r_2)\phi_2(\mathbf r_1)\phi_1(\mathbf r_2)$ that vanishes whenever $\mathbf r_1 = \mathbf r_2$. This is the Fermi hole: a region of suppressed same-spin probability density around each electron, purely a consequence of the determinant's structure, distinct from and additional to ordinary Coulomb (electrostatic) repulsion. The energy lowering associated with this hole is the exchange energy — the direct physical payoff of insisting on antisymmetry rather than a raw Hartree product.

The antisymmetry postulate itself is the starting axiom:

where: $\hat P_{ij}$ exchanges the full coordinates (space + spin) of electrons $i$ and $j$; the minus sign holds for every pair $i\neq j$ in an $N$-electron fermionic wavefunction.

Reading: this is not derivable from non-relativistic wave mechanics — it is an empirical postulate (justified more deeply by the spin–statistics theorem in relativistic quantum field theory) that must be imposed on any trial wavefunction for electrons. The normalized Slater determinant compactly enforces this postulate for orbital-product wavefunctions:

where: the sum runs over all $N!$ permutations $p$ of the electron labels; $(-1)^p$ is $+1$ for even permutations and $-1$ for odd ones; $\hat P$ generates the permuted product; $1/\sqrt{N!}$ normalizes the determinant assuming the $\chi_i$ are themselves orthonormal.

Reading: the determinant is the antisymmetrizer applied to the Hartree product — a single closed-form expression replacing an explicit sum over $N!$ terms. A shorthand notation packages this even more tightly. Since only the diagonal of the matrix (the "natural" assignment) needs to be written, chemists use the ket notation:

where: the ordering of spin-orbitals inside the ket is significant — swapping the order of any two labels inside the ket is understood to multiply the whole ket by $-1$, exactly mirroring the row-swap property of the determinant.

Reading: this notation lets one write and manipulate many-electron wavefunctions without redrawing the full determinant every time, while the antisymmetry is carried silently by the ordering convention. For matrix elements of operators between determinants, two central algebraic facts (Slater–Condon rules) do essentially all the work in Hartree–Fock theory and beyond. For a one-electron operator $\hat h = \sum_i \hat h(i)$,

where: the sum runs over all occupied spin-orbitals in $\Psi$; cross terms between different orbitals vanish because of orthonormality and the antisymmetrized structure.

Reading: the total one-electron energy of a determinant is just the sum of one-electron expectation values over occupied spin-orbitals — as intuitive as the Hartree-product result, but now rigorously justified. For the two-electron operator $\hat g = \sum_{i<j} 1/r_{ij}$ (Coulomb repulsion),

where: $\langle \chi_i\chi_j|\chi_i\chi_j\rangle = \iint \chi_i^*(1)\chi_j^*(2)\,\frac{1}{r_{12}}\,\chi_i(1)\chi_j(2)\,d\mathbf{x}_1 d\mathbf{x}_2$ is the Coulomb integral (classical electrostatic repulsion between charge clouds $i$ and $j$); $\langle \chi_i\chi_j|\chi_j\chi_i\rangle$ is the exchange integral (no classical analogue), nonzero only when $i$ and $j$ share the same spin.

Reading: the exchange integral is the direct fingerprint of antisymmetry inside the energy expression — it is exactly what the Hartree-product energy is missing, and exactly what lowers the energy of same-spin pairs by keeping them apart (the Fermi hole made quantitative).

Example 1 — Foundational: writing and checking the ground-state determinant of helium. Helium has two electrons. In the simplest (minimal-basis, restricted) approximation both occupy the same spatial orbital $\phi_{1s}(\mathbf r)$, one with $\alpha$ spin and one with $\beta$ spin. The spin-orbitals are $\chi_1 = \phi_{1s}\alpha$ and $\chi_2 = \phi_{1s}\beta$. The Slater determinant is

Check antisymmetry directly: swap electrons 1 and 2 in the final expression. The spatial factor $\phi_{1s}(1)\phi_{1s}(2)$ is symmetric (unchanged), but the spin factor $\alpha(1)\beta(2)-\beta(1)\alpha(2)$ becomes $\alpha(2)\beta(1)-\beta(2)\alpha(1) = -[\alpha(1)\beta(2)-\beta(1)\alpha(2)]$, so the whole wavefunction flips sign — antisymmetry confirmed, even though the spatial part alone looks deceptively symmetric. Check exclusion: if we tried $\chi_2=\phi_{1s}\alpha$ as well (both electrons spin-up in the 1s orbital), the two columns of the determinant become identical and $\Psi\equiv 0$ — exactly why helium's ground configuration is $1s^2$ with paired opposite spins and no third electron can join this orbital at all, foreshadowing the aufbau principle for heavier atoms. Using the energy formula derived above with $\chi_1,\chi_2$ carrying opposite spin, the exchange integral $K_{12}$ vanishes (its spin overlap integral $\int \alpha^*(2)\beta(2)\,d\omega_2 = 0$), so paired 1s electrons experience Coulomb repulsion $J_{12}$ but zero exchange stabilization — consistent with the fact that exchange only acts between same-spin electrons.

Example 2 — Applied: exchange energy lowering in the triplet excited state of helium. Consider the $1s^12s^1$ configuration of helium with parallel spins (a triplet state, physically realized in helium's lowest triplet term, $^3S$). Take spin-orbitals $\chi_1 = \phi_{1s}\alpha$ and $\chi_2 = \phi_{2s}\alpha$ — both spin-up, different spatial orbitals. Now the exchange integral does not vanish, since both spin functions are $\alpha$ and the spin overlap integral is 1. Using representative (illustrative, not spectroscopically exact) hydrogenic-orbital-scale integral values for helium's effective nuclear charge, suppose numerical evaluation of the relevant integrals gives (in atomic units, hartree):

(These are illustrative magnitudes consistent with the known ordering $J > K > 0$ for real hydrogenic 1s/2s pairs; the qualitative conclusion below is what matters and is exact.) For the triplet (same-spin, exchange active) determinant, the two-electron repulsion energy is

For the corresponding singlet (opposite-spin, no exchange term between 1s and 2s) configuration, built as a determinant with $\chi_1=\phi_{1s}\alpha$, $\chi_2=\phi_{2s}\beta$,

The triplet lies lower in electron–electron repulsion energy by exactly $K_{1s,2s} = 0.0440$ Ha $\approx 1.20$ eV — the same-spin electrons' Fermi hole keeps them further apart on average, reducing Coulomb repulsion. This reproduces, from first principles, Hund's rule: for a given orbital occupation, the higher-spin-multiplicity state is lower in energy, because exchange stabilization only operates between same-spin electron pairs. (The true experimental singlet–triplet splitting for He $1s2s$ also involves a modest change in the spatial orbitals themselves between the two spin states, so real spectroscopic splittings differ somewhat from this simplified fixed-orbital estimate, but the sign and mechanism — exchange stabilizing the triplet — are exactly right.)

Example 3 — Exam/competition level: parity of a 4-electron determinant under a non-adjacent transposition, and a Slater–Condon matrix element. Consider a four-electron closed-shell determinant built from spin-orbitals $\chi_1,\chi_2,\chi_3,\chi_4$ (say, $1s\alpha, 1s\beta, 2s\alpha, 2s\beta$ for beryllium), written in ket shorthand as $\Psi = |\chi_1\chi_2\chi_3\chi_4\rangle$. (a) Sign of a cyclic permutation. Suppose we relabel the electrons by the cyclic permutation $(1\,2\,3\,4)\to(2\,3\,4\,1)$, i.e., electron originally in position 1 moves to position 4, and each other electron shifts one slot left. A 4-cycle decomposes into 3 transpositions (a $k$-cycle is $(k-1)$ transpositions), so its parity is $(-1)^3 = -1$ — odd. Directly: writing the cycle as transpositions, $(1\,2\,3\,4) = (1\,4)(1\,3)(1\,2)$ — three swaps, confirming oddness. Therefore this relabeling multiplies the determinant by $-1$, exactly as required for consistency with the antisymmetry postulate applied electron-by-electron: any permutation's effect on $\Psi$ is $(-1)^p\Psi$ where $p$ counts the transposition parity, and a direct row-by-row check (each individual transposition contributes one sign flip, three flips multiply to $-1$) agrees. (b) A Slater–Condon "singles" matrix element. Now consider a second determinant $\Psi' = |\chi_1\chi_2\chi_3\chi_4'\rangle$, identical to $\Psi$ except spin-orbital $\chi_4$ is replaced by a different spin-orbital $\chi_4'$ (a "singly excited" determinant relative to $\Psi$, differing in exactly one spin-orbital). Slater's rules state that for a one-electron operator $\hat h=\sum_i\hat h(i)$,

i.e., all spin-orbitals that are the same in both determinants (here $\chi_1,\chi_2,\chi_3$) contribute exactly zero to the matrix element — only the one differing pair survives, because the $(N-1)!$ permutations involving the matched orbitals integrate to a Kronecker-delta orthogonality unless the unmatched orbital is also matched, which it is not here. If instead $\chi_4'$ differed by both spatial and spin part from $\chi_4$ — say $\chi_4=2s\beta$ but $\chi_4'=2p_z\beta$ — and if $\langle 2s|\hat h|2p_z\rangle = -0.0850$ Ha (an off-diagonal core-Hamiltonian element, nonzero because $\hat h$ contains the nuclear attraction operator, which is not diagonal in this basis for a non-hydrogenic ion), then

This single number is the entire matrix element between two four-electron, 24-term (4!) determinants — the Slater–Condon rules collapse what looks like an intractable sum into one small integral, which is exactly the computational leverage that makes determinant-based methods (Hartree–Fock, CI, coupled cluster) tractable at all.

John Clarke Slater (1900–1976) was an American physicist whose career sat squarely at the intersection of quantum theory and chemical structure — a rare figure equally at home deriving group-theoretic term symbols and running some of the earliest large-scale numerical calculations in physics. Trained at Harvard and a postdoctoral visitor to the great European centers of the new quantum mechanics (Cambridge, Copenhagen, and importantly Zurich, where Pauli's exclusion principle was being absorbed into the emerging formalism of many-electron quantum mechanics), Slater returned to build one of the strongest theoretical physics groups in the United States at MIT. He had an unusual talent for converting abstract symmetry requirements into concrete, computable rules — the determinant construction bearing his name is the purest expression of that talent: a single line of linear algebra that automatically enforces a subtle physical postulate that had otherwise required careful, error-prone bookkeeping over permutations.