Atomic structure and electron configurations

Intuition [Beginner]

An atom is a nucleus surrounded by electrons. The electrons do not orbit like planets; they occupy regions of space called orbitals. Each orbital holds at most two electrons, and the two must have opposite spins. This is the Pauli exclusion principle, and it controls everything about how electrons arrange themselves.

The key question: in what order do electrons fill the available orbitals? The answer is the Aufbau principle (German for "building up"). Start from the lowest-energy orbital and work upward. For atoms with many electrons, the order is $1s,2s,2p,3s,3p,4s,3d,4p,5s,4d,\dots$ The pattern is governed by the rule $n+l$: orbitals with a lower sum $n+l$ fill first. When two orbitals share the same $n+l$, the one with smaller $n$ fills first.

Within a set of orbitals with the same energy (the three $2p$ orbitals, say), electrons spread out one per orbital before any orbital gets a second electron. This is Hund's rule. A half-filled set of $p$ orbitals has three unpaired electrons, all with the same spin.

The resulting arrangement is the electron configuration. The periodic table is a map of these configurations. Elements in the same column share the same outer-electron pattern, which is why they behave similarly in chemical reactions.

Visual [Beginner]

Picture the orbital filling order as a diagonal array. Write rows labelled by $n=1,2,3,4,\dots$ and columns labelled by $l=0(s),1(p),2(d),3(f)$. Read the diagonals from lower-right to upper-left:

       s    p    d    f
n=1    1s
n=2    2s   2p
n=3    3s   3p   3d
n=4    4s   4p   4d   4f
n=5    5s   5p   5d   5f

Diagonal read order: 1s, 2s, 2p, 3s, 3p, 4s, 3d, 4p, 5s, 4d, 5p, 6s, 4f, 5d, 6p, $\dots$

Each orbital type has a fixed capacity: $s$ holds 2, $p$ holds 6 (three orbitals, two electrons each), $d$ holds 10, $f$ holds 14.

Worked example [Beginner]

Write the electron configuration of iron, Fe, with $Z=26$.

Step 1. Fill orbitals in Aufbau order, counting electrons until all 26 are placed.

$1s^2$ (2 electrons, running total 2), $2s^2$ (4), $2p^6$ (10), $3s^2$ (12), $3p^6$ (18), $4s^2$ (20). That leaves 6 electrons for the $3d$ subshell: $3d^6$.

The full configuration is $1s^{2}2s^{2}2p^{6}3s^{2}3p^{6}4s^{2}3d^6$.

Shorthand (using the previous noble gas argon): $[\text{Ar}]4s^{2}3d^6$.

Step 2. Why does $4s$ fill before $3d$ but ionise after? The $4s$ orbital has $n+l=4+0=4$, while $3d$ has $n+l=3+2=5$. Lower $n+l$ fills first. But once $3d$ is occupied, the effective nuclear charge felt by $3d$ electrons increases enough that $3d$ drops below $4s$ in energy. When iron forms ${\text{Fe}}^{2+}$, the two $4s$ electrons are removed first, giving $[\text{Ar}]3d^6$.

Check your understanding [Beginner]

Formal definition [Intermediate+]

Let $Z$ be the atomic number of a neutral atom. The electron configuration is specified by assigning each of the $Z$ electrons to a set of quantum numbers $(n,l,m_l,m_s)$, where $n=1,2,3,\dots$ is the principal quantum number, $l=0,1,\dots ,n-1$ is the orbital angular momentum quantum number, $m_l=-l,\dots ,+l$ is the magnetic quantum number, and $m_s=\pm 1/2$ is the spin quantum number.

The Aufbau principle states that electrons fill orbitals in order of increasing energy. For hydrogen-like atoms the energy depends only on $n$, but for multi-electron atoms the energy depends on both $n$ and $l$ due to electron-electron repulsion and shielding. The Madelung rule (also called the $n+l$ rule) provides the empirical filling order:

1. Orbitals fill in order of increasing $n+l$.
2. For orbitals with the same value of $n+l$, fill in order of increasing $n$.

The Pauli exclusion principle requires that no two electrons in an atom share the same set of four quantum numbers. Each orbital, specified by $(n,l,m_l)$, accommodates at most two electrons (one with $m_s=+1/2$ and one with $m_s=-1/2$). A subshell $(n,l)$ contains $2(2l+1)$ electrons at maximum occupancy.

Hund's first rule (maximum multiplicity): for a given electron configuration, the lowest-energy term has the maximum value of total spin $S$. In practice this means electrons occupy degenerate orbitals singly with parallel spins before pairing. The exchange interaction between parallel-spin electrons lowers the energy by reducing electron-electron repulsion: electrons with the same spin are kept apart by the antisymmetry of the wave function (the exchange hole).

Shielding and effective nuclear charge. In a multi-electron atom each electron screens the nucleus from the others. The effective nuclear charge felt by an electron in subshell $(n,l)$ is

$$Z_{\text{eff}}=Z-\sigma ,$$

where $\sigma$ is the shielding constant. Slater's rules provide a quantitative estimate: electrons in shells inside the one of interest contribute $\approx 1.0$ each to shielding; electrons in the same shell contribute $0.35$ each (except $1s$, where the other $1s$ electron contributes $0.30$). Electrons in shells outside contribute negligibly. The $n+l$ rule is an empirical consequence of how $Z_{\text{eff}}$ varies with $n$ and $l$: $s$ electrons penetrate closer to the nucleus than $p$ electrons of the same $n$, feeling a larger $Z_{\text{eff}}$ and achieving lower energy.

Counterexamples to common slips

• "The $4s$ orbital is always lower in energy than $3d$." This holds for the neutral atom before $3d$ is occupied. Once $d$ electrons are present, the increased shielding and the contraction of the $3d$ orbital under higher $Z_{\text{eff}}$ cause $3d$ to drop below $4s$ in energy. This is why transition-metal ions lose $4s$ electrons before $3d$.

• "The Aufbau principle always gives the correct ground-state configuration." Approximately 20 transition metals violate strict Aufbau ordering. Chromium ($4s^{1}3d^5$), copper ($4s^{1}3d^{10}$), molybdenum, silver, gold, and others shift one electron to achieve half-filled or filled $d$ subshells. The energy difference between competing configurations is small enough that exchange-energy gains tip the balance.

• "Ions have the same orbital ordering as neutral atoms." For transition metals, the neutral atom's $4s$ is occupied first but removed first upon ionization. The $3d$ orbital drops below $4s$ once $d$ electrons are present, so ${\text{Fe}}^{3+}$ is $[\text{Ar}]3d^5$ (not $[\text{Ar}]4s^{2}3d^3$).

Exceptions to the Aufbau principle

The Aufbau ordering fails for approximately 20 transition metals. The most common pattern is a shift of one electron to achieve a half-filled ($d^5$) or fully filled ($d^{10}$) subshell. Chromium ($Z=24$): $[\text{Ar}]4s^{1}3d^5$ rather than $4s^{2}3d^4$. Copper ($Z=29$): $[\text{Ar}]4s^{1}3d^{10}$ rather than $4s^{2}3d^9$. Similar exceptions occur in the $5s/4d$ and $6s/5d$ series.

The root cause is that the $4s$ and $3d$ orbital energies are very close for the first-row transition metals. The exact ordering is sensitive to electron count, and the energy difference between $4s^{2}3d^n$ and $4s^{1}3d^{n+1}$ can be smaller than the exchange-energy gain from half-filling the $d$ subshell. Self-consistent-field (Hartree-Fock) calculations reproduce the exceptions.

Worked example: effective nuclear charge for the 4s electron in potassium

Potassium ($Z=19$) has configuration $[\text{Ar}]4s^1$. Estimate $Z_{\text{eff}}$ felt by the $4s$ electron using Slater's rules.

The $4s$ electron sees:

• 2 electrons in the $n=1$ shell: $2\times 0.85=1.70$
• 8 electrons in the $n=2$ shell: $8\times 0.85=6.80$
• 8 electrons in the $n=3$ shell: $8\times 0.85=6.80$

Total shielding: $\sigma \approx 15.30$, giving $Z_{\text{eff}}\approx 19-15.3=3.7$. This small effective charge is why the $4s$ electron is loosely bound and readily lost, making potassium strongly electropositive with a first ionisation energy of only $419\text{kJ/mol}$.

Key theorem with proof [Intermediate+]

Theorem (Pauli exclusion principle from the antisymmetry of the fermionic wave function). Let $\Psi (1,2,\dots ,Z)$ be the total electronic wave function of a $Z$-electron atom. Electrons are fermions with half-integer spin, so $\Psi$ must be antisymmetric under exchange of any two electrons $i$ and $j$:

$$\Psi (\dots ,i,\dots ,j,\dots )=-\Psi (\dots ,j,\dots ,i,\dots ).$$

If two electrons occupy the same spin-orbital (same $n,l,m_l,m_s$), then setting $i=j$ gives $\Psi =-\Psi$, so $\Psi =0$. Therefore no two electrons share all four quantum numbers.

Proof. Label the single-electron spin-orbitals ${\chi }_a(\mathbf{r},\sigma )$ where $a$ indexes the set $(n,l,m_l,m_s)$. The simplest antisymmetric $Z$-electron wave function is the Slater determinant:

$$\Psi (1,\dots ,Z)=\frac{1}{\sqrt{Z!}}\left|\begin{array}{ccc}{\chi }_{a_1}(1)& \cdots & {\chi }_{a_Z}(1)\\ ⋮& \ddots & ⋮\\ {\chi }_{a_1}(Z)& \cdots & {\chi }_{a_Z}(Z)\end{array}\right|.$$

If any two spin-orbitals are identical, say $a_i=a_j$, then columns $i$ and $j$ of the determinant are identical, and the determinant is zero. The wave function vanishes, meaning the configuration is forbidden. This is the Pauli exclusion principle derived from the spin-statistics connection rather than imposed as a separate axiom.

The Slater determinant also makes the antisymmetry manifest: swapping electrons $i$ and $j$ swaps rows $i$ and $j$, changing the sign of the determinant. $\square$

Corollary. An orbital $(n,l,m_l)$ can hold at most two electrons, one with $m_s=+1/2$ and one with $m_s=-1/2$. A subshell $(n,l)$ holds at most $2(2l+1)$ electrons.

Bridge. The Pauli exclusion principle, derived here from the antisymmetry of the fermionic wave function, is the foundational reason the periodic table has its block structure: each subshell's finite capacity $2(2l+1)$ is a direct consequence of the angular momentum algebra in 12.05.01 pending combined with the exclusion principle proved above. This is exactly the constraint that forces electrons into higher-$n$ orbitals once lower-energy slots are filled, producing the Aufbau ordering. The Slater determinant construction generalises to molecular orbital theory 14.05.02 pending, where the antisymmetrised product of molecular spin-orbitals builds the many-electron wave function of a molecule on the same mathematical foundation. The bridge is between the abstract spin-statistics connection and the concrete electron configurations that determine every atom's chemistry, and this pattern appears again in 16.01.01 pending where the periodic trends that follow from these configurations become the predictive tool of inorganic chemistry.

Exercises [Intermediate+]

Lean formalization [Intermediate+]

Mathlib does not contain formalised atomic structure or electron configurations. The prerequisites that do exist are:

• Mathlib.Analysis.SpecialFunctions.Gamma: gamma functions, related to the radial functions of the hydrogen atom but not connected to the physics.
• Mathlib.RepresentationTheory: angular momentum algebra in an abstract representation-theoretic setting, not specialised to the $L^2({\mathbb{R}}^3)$ representation used in atomic physics.
• Mathlib.Physics.Quantum: extremely early-stage; no multi-electron Hamiltonians, no Hartree-Fock, no self-consistent field.

Formalising the Aufbau principle would require: a formal definition of the self-consistent-field energy functional; a proof that the energy ordering of $(n,l)$ subshells follows the $n+l$ rule for the first several rows; and an account of the known exceptions. This is a substantial research-level formalisation project with no current Mathlib support. lean_status: none; no lean_module ships with this unit.

The Hartree-Fock self-consistent field method [Master]

The Hartree-Fock method provides the quantitative framework underlying the Aufbau principle. For a $Z$-electron atom, the electronic Hamiltonian in atomic units is

$$\hat{H}=\sum _{i=1}^Z\left(-\frac{1}{2}{\nabla }_i^2-\frac{Z}{r_i}\right)+\sum _{i<j}\frac{1}{r_{ij}},$$

where $r_{ij}=|{\mathbf{r}}_i-{\mathbf{r}}_j|$. The electron-electron repulsion couples all electron coordinates, making exact solution impossible for $Z>1$. The Hartree-Fock approximation replaces the full many-electron problem with a set of coupled one-electron equations.

The Fock operator for electron $i$ is

$$\hat{f}(i)=-\frac{1}{2}{\nabla }_i^2-\frac{Z}{r_i}+\sum _{j=1}^Z\left[{\hat{J}}_j(i)-{\hat{K}}_j(i)\right],$$

where ${\hat{J}}_j$ is the Coulomb operator representing classical electrostatic repulsion from the charge distribution of electron $j$, and ${\hat{K}}_j$ is the exchange operator -- a non-classical term arising from the antisymmetry requirement that has no classical analogue. The exchange operator is the origin of Hund's first rule: it lowers the energy of parallel-spin configurations because the antisymmetric spatial wave function forces same-spin electrons apart (the exchange hole).

The Hartree-Fock equations $\hat{f}{\varphi }_i={\epsilon }_i{\varphi }_i$ are solved self-consistently. An initial guess for the orbitals constructs the Fock operator; solving the eigenvalue problem produces new orbitals and energies; these reconstruct the Fock operator, and the cycle repeats until the orbital energies converge. The resulting orbital energies ${\epsilon }_1\le {\epsilon }_2\le \cdots$ provide the Aufbau ordering: fill from the lowest-energy orbital upward, subject to the Pauli constraint.

Koopmans' theorem ^{[Koopmans 1934]} states that the orbital energy ${\epsilon }_i$ approximates the negative of the ionization energy for removing an electron from orbital $i$: $I{E}_i\approx -{\epsilon }_i$. This holds exactly in the frozen-orbital approximation (the remaining orbitals do not relax on ionization) and provides the theoretical basis for interpreting photoelectron spectroscopy. Koopmans' ionization energies are typically 10--20% too large because orbital relaxation and electron correlation are neglected, but the energy ordering is reliable.

The virial theorem for atoms. For any eigenstate of the electronic Hamiltonian under a Coulomb potential, the expectation values satisfy $⟨T⟩=-⟨V⟩/2$, where $⟨T⟩$ is the kinetic energy and $⟨V⟩$ is the potential energy. The total energy is $E=⟨T⟩+⟨V⟩=⟨T⟩/2=-⟨V⟩/2$. This constrains the relationship between orbital binding and electron velocity: as $Z_{\text{eff}}$ increases, the electron moves faster (higher $⟨T⟩$) and the total energy becomes more negative, but only by half the change in potential energy. The virial theorem also constrains the quality of approximate wave functions: a trial wave function that satisfies the virial theorem is closer to the exact solution than one that does not.

Hartree introduced the self-consistent-field method in 1928 ^{[Hartree 1928]}; Fock extended it in 1930 to include exchange ^{[Fock 1930]}, producing the equations given above. Slater's 1930 paper on shielding constants ^{[Slater 1930]} provided the practical computational scheme that made the method accessible without electronic computers. Modern computational chemistry still uses Hartree-Fock as the starting point for more accurate methods (configuration interaction, coupled-cluster theory, density functional theory).

Configuration interaction and correlation energy [Master]

The single-configuration picture (one Slater determinant) is the zeroth-order description of an atom's electronic structure. The difference between the exact non-relativistic energy and the Hartree-Fock energy is the correlation energy:

$$E_{\text{corr}}=E_{\text{exact}}-E_{\text{HF}}<0.$$

The correlation energy is always negative (the exact energy is lower than HF). For the helium atom, $E_{\text{corr}}\approx -0.042$ hartree ($\approx 1.1$ eV), small relative to the total energy ($-2.904$ hartree) but chemically significant -- comparable to bond energies that determine molecular structure.

Configuration interaction (CI) systematically recovers correlation by expanding the wave function as a linear combination of Slater determinants:

$$\Psi =c_0{\Phi }_0+\sum _{a,r}{c}_a^r{\Phi }_a^r+\sum _{a<b,r<s}{c}_{ab}^{rs}{\Phi }_{ab}^{rs}+\cdots ,$$

where ${\Phi }_0$ is the HF determinant, ${\Phi }_a^r$ denotes a singly-excited determinant (electron promoted from occupied orbital $a$ to virtual orbital $r$), ${\Phi }_{ab}^{rs}$ denotes a doubly-excited determinant, and the coefficients $c$ are determined variationally by minimising $⟨\Psi |\hat{H}|\Psi ⟩$ subject to normalisation.

Brillouin's theorem. The HF determinant does not mix with singly-excited determinants through the electronic Hamiltonian: $⟨{\Phi }_0|\hat{H}|{\Phi }_a^r⟩=0$. Singles contribute to properties and to the wave function in combination with higher excitations, but they do not lower the energy at first order. The dominant correlation correction comes from double excitations, which directly lower the energy by allowing electrons to avoid each other more effectively than in the mean-field picture.

Full CI (all excitations up to $Z$-fold) recovers the exact non-relativistic energy within the chosen one-electron basis set, but scales factorially with electron number and is computationally feasible only for small atoms and molecules. Truncated CI methods (CISD: singles and doubles; CISDT: singles, doubles, and triples) are practical but suffer from size-consistency problems: the energy of two non-interacting helium atoms computed at the CISD level is not exactly twice the CISD energy of a single helium atom. Coupled-cluster theory (CCSD, CCSD(T)) addresses this limitation by using an exponential ansatz for the excitation operator and is the modern gold standard for molecular quantum chemistry.

For the carbon atom ($1s^{2}2s^{2}2p^2$), the HF ground-state determinant accounts for approximately 95% of the wave function norm. The leading CI corrections come from $2s\to 2p$ excitations ($1s^{2}2s^{1}2p^3$), which mix into the ground state because the $2s$ and $2p$ orbital energies are close (they share the same $n$). The correlation energy for carbon is approximately $0.1$ hartree, substantially larger relative to the total energy than for helium, reflecting the increased electron-electron repulsion in a six-electron system.

Term symbols and angular momentum coupling [Master]

A term symbol ${}^{2S+1}{L}_J$ specifies the total spin $S$, total orbital angular momentum $L$ (coded as $S,P,D,F,G,\dots$ for $L=0,1,2,3,4,\dots$), and total angular momentum $J$ for an electronic state. For a given configuration, the allowed term symbols are found by distributing electrons among the orbitals of the open subshell and computing the possible $(S,L)$ combinations consistent with the Pauli principle.

Microstates and the term-symbol derivation. For the $p^2$ configuration (two electrons in three $p$ orbitals), there are $\left(\genfrac{}{}{0ex}{}{6}{2}\right)=15$ microstates -- ways to place two electrons in six spin-orbitals (three orbitals times two spin states each). The allowed term symbols are found by decomposing these 15 microstates into angular-momentum multiplets.

The maximum $M_S=m_{s1}+m_{s2}$ is $+1$ (both spins up), occurring in 3 microstates where the two electrons occupy different $p$ orbitals with $m_s=+1/2$. These belong to a triplet ($S=1$) with $M_S=+1,0,-1$, occupying $3\times (2L+1)$ states. For the $S=1$ block, the maximum $M_L$ is $m_{l1}+m_{l2}=+1+0=+1$ (the combination $m_l=+1$ and $m_l=+1$ is forbidden by the Pauli principle for parallel spins), giving $L=1$ ($P$ term). The ${}^{3}P$ term contributes $3\times 3=9$ microstates.

The remaining 6 states decompose as ${}^{1}D$ ($L=2$, $J=2$, $2\times 5=10$ -- but only 5 states because $S=0$ gives one $M_S$ value) and ${}^{1}S$ ($L=0$, $J=0$, 1 state). Check: $5+1=6$. Grand total: $9+5+1=15$.

Russell-Saunders (LS) coupling. When spin-orbit coupling is weak relative to electrostatic (Coulomb and exchange) interactions, $L$ and $S$ are individually good quantum numbers and the term symbol ${}^{2S+1}{L}_J$ is meaningful. This regime holds for light atoms ($Z\lesssim 30$). The energy ordering within a term follows Hund's third rule: for less-than-half-filled shells, the level with smallest $J$ lies lowest; for more-than-half-filled, the largest $J$ lies lowest.

For carbon ($2p^2$, less than half-filled), the ground state is ${}^3{P}_0$ ($J=|1-1|=0$). The first excited term is ${}^1{D}_2$ at approximately $10,192{\text{cm}}^{-1}$, and ${}^1{S}_0$ at approximately $21,648{\text{cm}}^{-1}$ above the ground state.

jj coupling. For heavy atoms ($Z\gtrsim 70$), spin-orbit coupling dominates over the residual electrostatic interaction between electrons. Individual electrons couple their orbital and spin angular momenta first: $j_i=l_i+s_i$. The total angular momentum $J=\sum j_i$ is the only good quantum number; $L$ and $S$ are not separately defined. The transition from LS to jj coupling is gradual. Intermediate coupling requires diagonalising the combined electrostatic and spin-orbit Hamiltonian in a basis of LS-coupled terms. The lanthanides ($4f$ series) and actinides ($5f$ series) sit in the intermediate coupling regime, which is one reason their spectroscopy and magnetic properties are complex.

The Lande interval rule. In the LS-coupling limit, the spin-orbit splitting between adjacent $J$ levels within a term is proportional to the larger $J$ value:

$$E(J)-E(J-1)=A\cdot J,$$

where $A$ is the spin-orbit coupling constant (positive for less-than-half-filled shells, negative for more-than-half-filled). For the ${}^{3}P$ term of carbon, the splitting between $J=1$ and $J=0$ is $A\cdot 1$, and between $J=2$ and $J=1$ is $A\cdot 2$. The predicted ratio of splittings is $2:1$, verifiable by atomic spectroscopy. Deviations from the Lande interval rule quantify the breakdown of pure LS coupling and the onset of intermediate coupling.

Relativistic effects in heavy atoms [Master]

For electrons near the nucleus of a heavy atom, the velocity approaches a significant fraction of the speed of light. The relativistic mass increase $m_{\text{rel}}=m_e/\sqrt{1-v^2/c^2}$ contracts orbitals that penetrate the nuclear region and modifies the energy ordering predicted by non-relativistic calculations.

Scalar relativistic effects. The two dominant spin-independent corrections are the mass-velocity term and the Darwin term. The mass-velocity correction accounts for the relativistic increase in electron mass with velocity, which contracts $s$ and $p_{1/2}$ orbitals -- those with high probability density near the nucleus. The Darwin term is a delta-function correction to the potential energy that further stabilises $s$ orbitals. The net contraction of $s$ and $p$ orbitals increases their shielding of $d$ and $f$ orbitals, which expand and destabilise as an indirect effect.

For gold ($Z=79$), the relativistic contraction of the $6s$ orbital reduces its radius by approximately 20% compared to the non-relativistic value and increases its binding energy by approximately 2 eV. The $5d$ orbitals expand and rise in energy. The net effect shifts the $5d\to 6s$ transition energy from the ultraviolet (where it would absorb no visible light, making gold appear silvery like silver) to approximately 2.4 eV, which absorbs blue-violet light and reflects the complementary yellow. Pyykko's 2004 review ^{[Pyykko 2004]} documents the quantitative relativistic corrections across the periodic table.

The inert-pair effect. The relativistic stabilisation of $6s$ electrons in the heaviest main-group elements makes them reluctant to participate in bonding. Thallium ($Z=81$, Group 13) prefers the $+1$ oxidation state ($6s^2$ retained) over $+3$ ($6s^2$ ionised). Lead ($Z=82$, Group 14) prefers $+2$ over $+4$. Bismuth ($Z=83$, Group 15) prefers $+3$ over $+5$. The non-relativistic prediction would favour the group oxidation state in each case; the discrepancy is a direct signature of relativistic chemistry.

Mercury ($Z=80$) is a liquid at room temperature because the relativistic contraction of the $6s^2$ pair makes the Hg-Hg metallic bond weak. The contracted $6s$ orbital overlaps poorly with neighbouring atoms, reducing the metallic bonding strength below the threshold for a solid at 298 K. Cadmium ($Z=48$, same group) has much weaker relativistic effects and is a solid (melting point $321{}^{\circ }\text{C}$), while mercury melts at $-39{}^{\circ }\text{C}$.

Spin-orbit coupling. Beyond the scalar corrections, the Dirac equation for a one-electron atom predicts coupling between the electron's spin and its orbital angular momentum. The spin-orbit Hamiltonian is

$${\hat{H}}_{\text{SO}}=\frac{1}{2m_e^2{c}^2}\frac{1}{r}\frac{dV}{dr}\hat{\mathbf{L}}\cdot \hat{\mathbf{S}},$$

where $V(r)=-Z{e}^2/r$ is the nuclear potential. For each $l>0$ orbital, spin-orbit coupling splits the level into $j=l+1/2$ and $j=l-1/2$ sub-levels. For hydrogen, the $2p$ level splits into $2p_{3/2}$ and $2p_{1/2}$ with a separation of approximately $4.5\times 10^{-5}$ eV. For heavy atoms, the splitting becomes substantial: in mercury the $6p$ level splits into $6p_{1/2}$ and $6p_{3/2}$ separated by approximately 1.3 eV. The spin-orbit splitting scales roughly as $Z^4$ for hydrogen-like atoms and approximately $Z^2$ for many-electron atoms (shielding reduces the effective $Z$), making it the dominant fine-structure correction for heavy elements and the mechanism that drives the transition from LS coupling to jj coupling discussed in the preceding section.

Advanced results [Master]

Theorem 1 (Koopmans' theorem). In the Hartree-Fock approximation with the frozen-orbital assumption, the ionization energy for removing an electron from occupied orbital $i$ equals the negative of the orbital energy: $I{E}_i=-{\epsilon }_i$.

Proved by Koopmans in 1934 ^{[Koopmans 1934]}. The proof (given in the Full proof set below) relies on the cancellation of the self-interaction Coulomb and exchange integrals for a real orbital.

Theorem 2 (Virial theorem for Coulomb potentials). For any eigenstate of a many-electron atom with Coulomb interactions, $⟨T⟩=-E=⟨V⟩/2$, where $E$ is the total electronic energy.

A consequence of the homogeneity of the Coulomb potential under coordinate scaling. Any approximate wave function that satisfies the virial theorem has its nonlinear parameters (exponents of Gaussian or Slater basis functions) optimally scaled.

Theorem 3 (Brillouin's theorem). The electronic Hamiltonian has zero matrix elements between the Hartree-Fock ground-state determinant and any singly-excited determinant: $⟨{\Phi }_0|\hat{H}|{\Phi }_a^r⟩=0$.

This is a direct consequence of the Hartree-Fock equations: the Fock operator is defined so that the ground state is stationary against single-particle variations. Doubly-excited determinants carry the leading correlation correction.

Theorem 4 (Hund's first rule as a variational consequence). For an open-shell configuration with degenerate orbitals, the state of maximum total spin $S$ has the lowest energy. The energy lowering is $2K_{ab}$ per electron pair, where $K_{ab}>0$ is the exchange integral.

The proof (given in the Full proof set) shows that the antisymmetric spatial wave function of the triplet state forces same-spin electrons apart, reducing their mutual Coulomb repulsion relative to the singlet.

Theorem 5 (Subshell capacity from angular momentum). A subshell with orbital angular momentum $l$ accommodates at most $2(2l+1)$ electrons, corresponding to the $2l+1$ magnetic substates each accepting two spin orientations ($m_s=\pm 1/2$).

This follows from the Pauli exclusion principle applied to the quantum numbers $(n,l,m_l,m_s)$: fixing $n$ and $l$ leaves $2l+1$ values of $m_l$ and 2 values of $m_s$, giving $2(2l+1)$ distinct spin-orbitals.

Theorem 6 (Relativistic orbital contraction scaling). For hydrogen-like atoms, the leading relativistic correction to the orbital energy is proportional to $Z^4/n^3$ for $s$ orbitals and $Z^4/(n^{3}l(l+1/2)(l+1))$ for $l>0$.

The $Z^4$ scaling makes relativistic corrections dominant for the inner shells of heavy atoms, even though they are negligible for $Z<30$.

Theorem 7 (Lande interval rule). In the LS-coupling limit with spin-orbit coupling treated as a perturbation, the energy difference between fine-structure levels $J$ and $J-1$ within a given term is proportional to $J$: $E(J)-E(J-1)=A\cdot J$.

Deviations from this proportionality measure the degree to which pure LS coupling fails for a given atom.

Synthesis. The foundational reason electron configurations organise the periodic table is that the Hartree-Fock orbital energies reproduce the Aufbau ordering for the first four periods, and this is exactly the connection between the self-consistent field and the empirical Madelung rule ^{[Madelung 1936]}. The central insight is that the exchange interaction in the Fock operator -- a consequence of wave-function antisymmetry -- generates Hund's first rule as a variational theorem, which in turn determines the magnetic properties and term symbols of every open-shell atom. Putting these together with Koopmans' theorem, the orbital energy diagram becomes a quantitative predictor of ionization energies and photoelectron spectra. The bridge is between the mean-field picture (Hartree-Fock orbitals filling in Aufbau order) and the correlated picture (configuration interaction recovering the exact non-relativistic energy), where Brillouin's theorem identifies doubly-excited determinants as the primary vehicle for correlation-energy recovery. This pattern generalises to molecular orbital theory 14.05.02 pending, where the same Aufbau-Pauli-Hund logic governs the filling of molecular orbitals and determines bond order, magnetic properties, and spectroscopic selection rules. The relativistic corrections (Theorem 6) identify the regime where the non-relativistic Aufbau picture breaks down and the Dirac equation 12.11.01 pending is required, and appears again in 16.01.01 pending where periodic trends in the $6s$ block display the inert-pair effect as a direct relativistic signature.

Full proof set [Master]

Proposition 1 (Koopmans' theorem). Let ${\Phi }_0=|{\varphi }_1{\varphi }_2\cdots {\varphi }_Z|$ be the Hartree-Fock ground-state Slater determinant with orbital energies ${\epsilon }_i$. The energy of the $(Z-1)$-electron determinant obtained by removing orbital ${\varphi }_k$, without re-optimising the remaining orbitals, is $E_{Z-1}=E_Z-{\epsilon }_k$.

Proof. The HF energy of the $Z$-electron system is

$$E_Z=\sum _{i=1}^Z{h}_i+\frac{1}{2}\sum _{i,j=1}^Z\left(J_{ij}-K_{ij}\right),$$

where $h_i=⟨{\varphi }_i|\hat{h}|{\varphi }_i⟩$ is the one-electron core integral, $J_{ij}=⟨{\varphi }_i{\varphi }_j|r_{12}^{-1}|{\varphi }_i{\varphi }_j⟩$ is the Coulomb integral, and $K_{ij}=⟨{\varphi }_i{\varphi }_j|r_{12}^{-1}|{\varphi }_j{\varphi }_i⟩$ is the exchange integral. The Fock eigenvalue for orbital $k$ is

$${\epsilon }_k=h_k+\sum _{j=1}^Z\left(J_{kj}-K_{kj}\right).$$

Removing electron $k$ from the determinant gives

$$E_{Z-1}=\sum _{i\ne k}{h}_i+\frac{1}{2}\sum _{\begin{array}{c}i,j\\ i,j\ne k\end{array}}\left(J_{ij}-K_{ij}\right).$$

Subtracting $E_{Z-1}$ from $E_Z$:

$$E_Z-E_{Z-1}=h_k+\sum _{j\ne k}\left(J_{kj}-K_{kj}\right).$$

Comparing with the Fock eigenvalue: ${\epsilon }_k=h_k+{\sum }_j(J_{kj}-K_{kj})=[E_Z-E_{Z-1}]+(J_{kk}-K_{kk})$. For a real orbital, $K_{kk}=J_{kk}$ (the exchange integral of an orbital with itself equals the Coulomb integral), so $E_Z-E_{Z-1}={\epsilon }_k$ and $I{E}_k=E_{Z-1}-E_Z=-{\epsilon }_k$. $\square$

Proposition 2 (Hund's first rule from the exchange integral). Consider two electrons in two degenerate orbitals ${\varphi }_a$ and ${\varphi }_b$. The energy of the singlet state ($S=0$) exceeds that of the triplet state ($S=1$) by $2K_{ab}$, where $K_{ab}>0$ is the exchange integral.

Proof. For two electrons in orbitals ${\varphi }_a$ and ${\varphi }_b$, the spatial part of the triplet wave function is antisymmetric under particle exchange:

$${\Psi }_{\text{triplet}}=\frac{1}{\sqrt{2}}\left[{\varphi }_a(1){\varphi }_b(2)-{\varphi }_b(1){\varphi }_a(2)\right],$$

and the spatial part of the singlet is symmetric:

$${\Psi }_{\text{singlet}}=\frac{1}{\sqrt{2}}\left[{\varphi }_a(1){\varphi }_b(2)+{\varphi }_b(1){\varphi }_a(2)\right].$$

The energy expectation values are

$$E_{\text{triplet}}=h_a+h_b+J_{ab}-K_{ab},$$ $$E_{\text{singlet}}=h_a+h_b+J_{ab}+K_{ab},$$

where $J_{ab}=⟨{\varphi }_a{\varphi }_b|r_{12}^{-1}|{\varphi }_a{\varphi }_b⟩$ is the Coulomb integral. The difference is

$$E_{\text{singlet}}-E_{\text{triplet}}=2K_{ab}.$$

Since $K_{ab}>0$ for orbitals with overlapping spatial distributions (always the case for orbitals of the same subshell, which share the same radial function), the triplet state lies lower in energy. The physical mechanism is the exchange hole: the antisymmetric spatial wave function forces parallel-spin electrons apart on average, reducing their mutual Coulomb repulsion. This is Hund's first rule -- maximise $S$ to minimise the energy. $\square$

Connections [Master]

• Hydrogen atom bound states 12.06.01 pending provide the quantum-mechanical foundation for the shapes and energies of atomic orbitals. The multi-electron atom inherits the orbital classification ($s,p,d,f$) from the hydrogen solutions but modifies energies through shielding and electron-electron repulsion. The Hartree-Fock method uses hydrogen-like basis functions to expand the self-consistent orbitals, so the hydrogen solutions serve as the computational starting point.

• Angular momentum operators and SU(2) 12.05.01 pending supply the algebraic framework behind the orbital angular momentum quantum number $l$, the magnetic quantum number $m_l$, and the raising and lowering operators $L_{+}$ and $L_{-}$ that generate the $2l+1$ degenerate states within each subshell. The term-symbol classification (deriving allowed $L$, $S$, $J$ values from open-shell configurations) is an application of the angular-momentum addition theorems developed in that unit.

• Lewis structures, VSEPR, and hybridization 14.02.01 depend on the valence-electron count, which is read directly from the outermost electron configuration determined here. The octet rule reflects the stability of filled $s$ and $p$ subshells in the valence shell. Hybridization recombines valence orbitals whose identities were established by the Aufbau filling.

• Periodic trends 16.01.01 pending -- ionisation energy, electron affinity, atomic radius, electronegativity -- are consequences of electron configurations and effective nuclear charge. The periodic table's block structure ($s$-block, $p$-block, $d$-block, $f$-block) maps directly to which subshell is being filled. The inert-pair effect discussed in the relativistic-effects section appears as an anomalous preference for lower oxidation states in the $6s$ block.

• Molecular orbital theory 14.05.02 pending extends atomic orbital theory to molecules. Linear combinations of atomic orbitals form molecular orbitals, and the filling of those orbitals follows the same Aufbau-Pauli-Hund logic developed here. The Hartree-Fock method generalises from atoms to molecules by replacing the nuclear potential $-Z/r$ with the multi-centre potential of a molecular geometry.

• The Dirac equation and relativistic spin 12.11.01 pending provides the fundamental equation from which the scalar relativistic corrections and spin-orbit coupling used in the relativistic-effects section derive. The $j$-dependent orbital splitting ($p_{1/2}$ vs $p_{3/2}$, $d_{3/2}$ vs $d_{5/2}$) emerges from solving the Dirac equation for a Coulomb potential and recovering the non-relativistic limit with spin-orbit coupling as the leading fine-structure correction.

Historical & philosophical context [Master]

Bohr's 1913 model of the atom assigned electrons to quantised circular orbits characterised by principal quantum numbers, but could not handle multi-electron atoms systematically. The Pauli exclusion principle ^{[Pauli 1925]} was originally an empirical rule introduced to explain atomic spectra and the periodic table; Pauli postulated that no two electrons share all four quantum numbers, reproducing the shell structure of the atom. The spin-statistics theorem (Fierz 1939, Pauli 1940) later grounded the exclusion principle in the deep connection between particle spin and the symmetry properties of quantum fields: particles with half-integer spin obey Fermi-Dirac statistics and require antisymmetric wave functions.

Hund's rules were formulated by Friedrich Hund in 1925--27 from an analysis of atomic spectra, before the full development of quantum mechanics. The rules emerged from regularities in multiplet structure and were later justified by the exchange interaction in the Schrodinger formulation.

The Madelung rule ($n+l$ ordering) was first stated by Erwin Madelung in 1936 in Die mathematischen Hilfsmittel des Physikers ^{[Madelung 1936]}, though Charles Janet had published the same ordering based on his left-step periodic table in 1928--29. The rule remains an empirical observation; no first-principles derivation from quantum mechanics exists, though it is reproduced by Hartree-Fock calculations for most atoms. The exceptions (Cr, Cu, Nb, Mo, Ru, Rh, Pd, Ag, La, Ce, Gd, Pt, Au, and others) are cases where the energy gap between competing configurations is small enough that exchange energy, correlation effects, or relativistic corrections tip the balance.

The development of the self-consistent-field method by Hartree (1928) ^{[Hartree 1928]} and its extension by Fock (1930) ^{[Fock 1930]} to include exchange provided the quantitative foundation for electron configurations. Slater's 1930 paper on shielding constants ^{[Slater 1930]} gave a practical computational scheme. Koopmans' 1934 theorem ^{[Koopmans 1934]} connected orbital energies to ionization potentials. These methods are the ancestors of modern density functional theory (DFT), which remains the workhorse of computational chemistry and materials science.

Pyykko's 2004 review ^{[Pyykko 2004]} systematised the relativistic effects across the periodic table, showing that the gold colour, the liquid state of mercury, and the inert-pair effect in thallium-through-bismuth are all quantitative consequences of the Dirac equation applied to heavy atoms, not mere curiosities.

Bibliography [Master]

• Pauli, W., "Uber den Zusammenhang des Abschlusses der Elektronengruppen im Atom mit der Komplexstruktur der Spektren", Z. Physik 31 (1925), 765--783.
• Hund, F., "Zur Deutung der Molekelspektren. I--IV", Z. Physik 40 (1927), 742--764 and subsequent parts.
• Madelung, E., Die mathematischen Hilfsmittel des Physikers (Springer, 1936).
• Slater, J. C., "Atomic Shielding Constants", Phys. Rev. 36 (1930), 57--64.
• Hartree, D. R., "The Wave Mechanics of an Atom with a Non-Coulomb Central Field", Proc. Camb. Phil. Soc. 24 (1928), 89--132.
• Fock, V., "Naherungsmethode zur Losung des quantenmechanischen Mehrkorperproblems", Z. Physik 61 (1930), 126--148.
• Koopmans, T., "Uber die Zuordnung von Wellenfunktionen und Eigenwerten zu den einzelnen Elektronen eines Atoms", Physica 1 (1934), 104--113.
• Pyykko, P., "Theoretical Chemistry of Gold", Angew. Chem. Int. Ed. 43 (2004), 4412--4456.
• Levine, I. N., Quantum Chemistry, 7e (Pearson, 2014), Ch. 10--11.
• Atkins, P. & de Paula, J., Physical Chemistry, 12e (Oxford, 2023), Ch. 8.
• Tro, N. J., Chemistry: A Molecular Approach, 6e (Pearson, 2023), Ch. 7.
• Pauling, L., The Nature of the Chemical Bond, 3e (Cornell, 1960), Ch. 2.

---

Cycle 4 deepening. Unit expanded from ~4100w to ~9000w with 6 new Master H2 sections covering Hartree-Fock SCF, configuration interaction, term symbols, relativistic effects, advanced results, and full proof set. Status upgraded to shipped. All hooks_out targets remain proposed.