The Pauli Exclusion Principle

The exclusion principle is not a historical curiosity confined to introductory orbital diagrams — it is operative in essentially every area of modern chemistry and physics that involves more than one electron. In quantum chemistry software, every Hartree–Fock or density-functional-theory calculation used to design drugs, catalysts, or battery materials builds its wavefunction (or, in DFT, its effective single-particle framework) from Slater determinants precisely because antisymmetry must be enforced exactly — get this wrong and the computed energies and geometries of molecules are simply incorrect. In materials science, the Fermi energy and Fermi surface of a metal, central to understanding electrical conductivity, superconductivity, and semiconductor band structure, exist only because electrons fill available states from the bottom up, one (spin-paired) state at a time, obeying exclusion — this is literally why metals have a well-defined "sea" of conduction electrons up to a sharp energy cutoff rather than all electrons sitting in the lowest state. In astrophysics, electron degeneracy pressure — a direct macroscopic manifestation of exclusion forcing electrons into ever-higher momentum states as they are compressed — is what holds white dwarf stars up against gravitational collapse, and neutron degeneracy pressure (the same principle applied to neutrons) supports neutron stars, with the Chandrasekhar mass limit ($\approx 1.4\,M_\odot$) marking where even this quantum pressure fails and the star must collapse further. In magnetism and materials for data storage, the exclusion-driven preference for parallel-spin filling of degenerate orbitals (Hund's rule, itself an exclusion-plus-exchange-energy consequence) is the microscopic origin of ferromagnetism in iron, cobalt, and nickel, underlying everything from hard-disk storage to MRI machine construction. In spectroscopy and photochemistry, the allowed term symbols and selection rules for atomic and molecular transitions — used to interpret everything from flame tests to stellar absorption spectra to the color of transition-metal complexes in inorganic chemistry — are enumerated using exactly the same quantum-number-uniqueness bookkeeping worked through above. And in the emerging field of quantum computing with fermionic qubits, exclusion-respecting encodings (such as the Jordan–Wigner transformation) are required to simulate molecular electronic structure on quantum hardware, directly exploiting the antisymmetry postulate as a computational resource rather than a constraint to work around.

Ask the tutor: How does the antisymmetry requirement change if we go beyond the orbital approximation and include explicit electron correlation, and how is that reflected in a full Slater-determinant-based configuration interaction wavefunction?

Begin with the notion of indistinguishability. In classical mechanics you can, in principle, paint electron #1 red and electron #2 blue and track each trajectory. Quantum mechanically this is meaningless: electrons have no trajectories, only a joint probability amplitude $\Psi(1,2,\dots,N)$ where "1" denotes all the coordinates of electron one — its position $\mathbf{r}_1$ and its spin projection $m_{s,1}$. Because the electrons are physically identical, swapping the labels "1" and "2" everywhere in $\Psi$ cannot change any measurable prediction, since $|\Psi|^2$ is the only physically accessible quantity and it must be unchanged by a relabeling that corresponds to no physical difference. This symmetry requirement allows exactly two possibilities for how $\Psi$ itself (not just $|\Psi|^2$) behaves under exchange of two identical particles: it can stay the same (symmetric) or flip sign (antisymmetric). Relativistic quantum field theory (the spin-statistics theorem) proves that which behavior occurs is locked to the particle's intrinsic spin: integer-spin particles (bosons) are always symmetric under exchange, and half-integer-spin particles (fermions) are always antisymmetric. Electrons have spin $\tfrac12$, so they are fermions, and the wavefunction of any collection of electrons must satisfy

for every pair $i,j$. This is the antisymmetry postulate, and it is the actual content of the Pauli exclusion principle. Everything else — orbital diagrams, "up-down" arrows in boxes, the rule about quantum numbers — is downstream bookkeeping. Now watch what antisymmetry does mechanically. Suppose we try to build a wavefunction for two electrons from one-electron spin-orbitals $\chi_a(1)$ and $\chi_b(2)$, where a spin-orbital is a spatial orbital multiplied by a spin function, e.g. $\chi_a = \psi_{1s}\alpha$. A simple product $\chi_a(1)\chi_b(2)$ is not antisymmetric; exchanging labels gives $\chi_a(2)\chi_b(1) \ne -\chi_a(1)\chi_b(2)$ in general. The unique way (up to normalization and an overall phase) to build an antisymmetric combination out of two spin-orbitals is the determinant-like combination

If you now ask "what if $a = b$", i.e., both electrons are assigned to the same spin-orbital, the two terms become identical and cancel exactly:

A wavefunction that is identically zero describes a state with zero probability of occurring anywhere. That is the exclusion principle in its rawest algebraic form: two electrons cannot occupy the same spin-orbital, not because of some extra force pushing them apart, but because the only legal (antisymmetric) wavefunction for that assignment vanishes.

The next physical picture worth holding in mind is the Fermi hole. Antisymmetry does more than forbid identical spin-orbitals outright — even for different orbitals with the same spin, the wavefunction is constructed so that the probability of finding two same-spin electrons at the same point in space is forced to zero, and stays suppressed over a "correlation hole" of radius roughly one orbital size. This is why electrons of like spin instinctively avoid each other beyond what plain Coulomb repulsion would give — an effect called exchange correlation, and it is why parallel-spin configurations (as in Hund's first rule) are lower in energy than you might naively expect from orbital energies alone. Finally, connect the abstract antisymmetry postulate to the familiar orbital-counting rule. In the independent-particle (orbital) picture, a spin-orbital is labeled by $(n,\ell,m_\ell,m_s)$. Antisymmetry forces the many-electron wavefunction to be a Slater determinant of spin-orbitals; a determinant with two identical rows (two electrons assigned to the same spin-orbital, i.e., identical $n,\ell,m_\ell,m_s$) is identically zero, by the elementary linear-algebra fact that a determinant with two equal rows vanishes. That is precisely why chemists say "no two electrons in an atom can have the same four quantum numbers" — it is the orbital-language shadow of the antisymmetry postulate.

The master statement, valid for any number of identical fermions and any basis (not just orbital products), is the antisymmetry postulate itself:

where:

• $\hat{P}_{ij}$ is the operator that exchanges the complete coordinates (space and spin) of particles $i$ and $j$,
• $\Psi$ is the total $N$-electron wavefunction,
• the eigenvalue $-1$ (rather than $+1$) is the defining property of fermions, fixed by the spin-statistics theorem.

Reading: this is not a statement about orbitals at all — it holds for the exact, fully correlated many-electron wavefunction, long before you ever approximate it as built from one-electron orbitals. The orbital-level, practical consequence — built from a normalized Slater determinant of $N$ spin-orbitals $\chi_1,\dots,\chi_N$ — is

where:

• rows index electrons, columns index occupied spin-orbitals,
• $1/\sqrt{N!}$ normalizes the determinant given orthonormal spin-orbitals,
• swapping two rows (two electrons) swaps two rows of the determinant, which multiplies it by $-1$ — automatically satisfying antisymmetry,
• two identical columns (two electrons assigned the same spin-orbital) make the determinant vanish identically.

Reading: the Slater determinant is not an arbitrary trick — it is the unique, general construction that turns any set of one-electron functions into a properly antisymmetrized many-electron wavefunction, and it automatically enforces exclusion via ordinary linear algebra. The orbital-counting shortcut used constantly in general chemistry is the corollary:

where:

• $n$ = principal quantum number, $\ell$ = orbital angular momentum quantum number, $m_\ell$ = its projection, $m_s = \pm\tfrac12$ the spin projection,
• this restricts each spatial orbital $(n,\ell,m_\ell)$ to at most two electrons, and they must have opposite $m_s$.

Reading: this is the version every student first meets; it is exactly right for the independent-electron (orbital) model, and it is what makes shell filling ($2, 8, 8, \dots$ style counting) and the aufbau principle work. A quantitative measure of how strongly exclusion reorganizes electron density is the pair correlation built into the Slater determinant. For two electrons with the same spin in different spatial orbitals $\psi_a,\psi_b$, the probability of finding them at the same point vanishes:

where:

• this follows directly from the antisymmetric spatial combination $\tfrac{1}{\sqrt2}[\psi_a(\mathbf r_1)\psi_b(\mathbf r_2) - \psi_b(\mathbf r_1)\psi_a(\mathbf r_2)]$ vanishing at $\mathbf r_1=\mathbf r_2$,
• for opposite-spin electrons no such cancellation occurs, so they can coexist at the same point (subject only to Coulomb repulsion), which is why opposite-spin pairing in an orbital is allowed while same-spin double occupancy is not.

Reading: this is the mathematical seed of the Fermi hole and of exchange energy, both of which reappear whenever Hund's rules or magnetism are discussed.

Example 1 — Foundational: verifying antisymmetry and the vanishing determinant for He 1s². Consider the helium atom ground configuration, both electrons nominally in the $1s$ spatial orbital $\psi_{1s}$. Assign spin-orbitals $\chi_1 = \psi_{1s}\alpha$ and $\chi_2 = \psi_{1s}\beta$ (opposite spins — the only legal choice, as shown above). The two-electron wavefunction is

Expanding the determinant:

which is a legitimate, non-zero, properly antisymmetric wavefunction — the familiar $1s^2$ singlet ground state. Now try the illegal configuration: force both electrons into $\chi_1 = \psi_{1s}\alpha$ (same spatial orbital, same spin — this would be trying to write $1s^2$ with parallel spins, which violates the shell-filling rule taught in introductory courses). The determinant becomes

because the two columns are identical. There is no such state; helium's ground configuration with two parallel-spin electrons in $1s$ simply does not exist as a physical possibility — confirmed by the vanishing determinant, not asserted by fiat. Limiting-case check: if we instead let the spatial orbitals differ, say electron "$a$" in $1s$ and electron "$b$" in $2s$ (an excited configuration $1s^12s^1$), then parallel spin is allowed, since $\chi_a=\psi_{1s}\alpha \ne \chi_b = \psi_{2s}\alpha$ — the columns differ and the determinant is nonzero. This correctly reproduces the excited helium triplet state $1s^12s^1\,{}^3S$, which is real and spectroscopically observed, confirming the rule discriminates correctly between allowed and forbidden cases rather than banning all same-spin pairs universally.

Example 2 — Applied: predicting the ground-state electron configuration and term symbol contribution for nitrogen (Z = 7). Nitrogen's electron configuration is $1s^2\,2s^2\,2p^3$. Apply the exclusion principle orbital-by-orbital using asymmetric, explicit quantum numbers (not just shell counts). The $1s$ subshell ($\ell=0$, so $m_\ell=0$ only) holds two electrons: $(n,\ell,m_\ell,m_s) = (1,0,0,+\tfrac12)$ and $(1,0,0,-\tfrac12)$ — this exhausts the only two spin-orbitals available at $n=1,\ell=0$, and no further electron can be added there. The $2s$ subshell similarly holds $(2,0,0,+\tfrac12)$ and $(2,0,0,-\tfrac12)$ — full at two electrons. The $2p$ subshell has $\ell=1$, so $m_\ell \in \{-1,0,+1\}$, giving six available spin-orbitals: $(2,1,-1,\pm\tfrac12)$, $(2,1,0,\pm\tfrac12)$, $(2,1,+1,\pm\tfrac12)$. Nitrogen has exactly three electrons to place here. Exclusion forbids doubling up in any single spin-orbital, but says nothing yet about how three electrons should distribute among three spatial $2p$ orbitals with two spin choices each — that additional physics (minimizing electron repulsion by keeping electrons in different spatial orbitals with parallel spin where possible) is Hund's first rule, layered on top of exclusion. Exclusion alone tells us the three $2p$ electrons must occupy three different $(n,\ell,m_\ell,m_s)$ combinations. The lowest-repulsion arrangement consistent with this — verified by spectroscopy — places one electron in each spatial $2p$ orbital with parallel spins:

This is a perfectly legal assignment: all three quadruplets of quantum numbers are distinct (they differ in $m_\ell$), so no antisymmetry violation occurs, and the corresponding Slater determinant is nonzero — consistent with nitrogen's experimentally observed $^4S_{3/2}$ ground term with total spin $S = 3/2$ (three unpaired electrons). Limiting-case check: if nitrogen had four $2p$ electrons instead (as in an isoelectronic ion or the next configuration up the periodic table, oxygen's $2p^4$), the fourth electron cannot find a fifth empty spatial 2p orbital — there are only three — so it must pair up in an already-occupied one with opposite spin, giving one doubly occupied $2p$ orbital and two singly occupied ones. This correctly predicts oxygen's ground term has only two unpaired electrons ($S=1$, term $^3P$), matching its well-known paramagnetism (weaker than nitrogen's, stronger than neon's zero), and shows the exclusion-driven bookkeeping scales correctly as electron count changes.

Example 3 — Exam/competition level: counting the maximum electrons in the $n = 3$ shell using only quantum-number exclusion, and cross-checking against $2n^2$. A common competition-style question: derive, from Pauli exclusion and the allowed quantum number ranges, the maximum number of electrons the $n=3$ shell can hold — without simply quoting the formula $2n^2$. For $n=3$, the allowed $\ell$ values are $\ell = 0, 1, 2$ (i.e., $3s, 3p, 3d$). For each $\ell$, $m_\ell$ ranges over $2\ell+1$ integer values from $-\ell$ to $+\ell$, and each spatial orbital $(n,\ell,m_\ell)$ admits exactly 2 spin states ($m_s = \pm\tfrac12$) by exclusion (Example verified above: never more than 2, since only two spin values exist). Count spin-orbitals subshell by subshell:

Sum over all $\ell$ from 0 to $n-1=2$:

Cross-check against the general closed-form result. The total number of $(\ell,m_\ell)$ spatial orbital slots for principal quantum number $n$ is $\sum_{\ell=0}^{n-1}(2\ell+1) = n^2$ (a standard sum-of-odd-numbers identity: $1+3+5+\dots+(2n-1) = n^2$). Multiplying by 2 for the two spin states per spatial orbital (exclusion's contribution) gives the textbook shell capacity:

matching the subshell-by-subshell count exactly. This confirms both the individual subshell counting and the general $2n^2$ formula are the same statement, with the exclusion principle supplying the "times 2" and the geometry of angular momentum supplying the $n^2$. Limiting-case check: setting $n=1$ gives $N_{\max}(1) = 2(1)^2 = 2$, correctly reproducing helium's closed $1s^2$ shell as the smallest possible closed shell — and the subshell sum for $n=1$ trivially has only $\ell=0$ contributing $2$ spin-orbitals, matching exactly with no further terms, confirming the formula degenerates correctly at the smallest nontrivial case.