Term Symbols for d-Electron Configurations

Term symbols are not a historical curiosity confined to atomic physics — they are the backbone of every modern discussion of transition-metal electronic structure. Tanabe–Sugano and Orgel diagrams, used daily to interpret UV-vis spectra of coordination complexes in synthetic inorganic and bioinorganic laboratories, are literally plots of how each free-ion term (derived exactly as above) splits and shifts in energy as octahedral or tetrahedral ligand-field strength increases; you cannot construct or read one without first knowing the parent term list. In transition-metal catalysis and materials design, predicting whether a $d^n$ metal center will be high-spin or low-spin, and hence how reactive or inert it will be, starts from identifying the ground term (the lowest-energy term among the derived list, governed by Ground-State Terms and Hund's Rules) and tracking how the ligand field mixes or splits it (the subject of Splitting of Terms in an Octahedral Field). Instrumentation such as electron paramagnetic resonance (EPR) spectroscopy and SQUID magnetometry directly measures quantities — effective magnetic moments, zero-field splittings — that are only interpretable once the correct term (not just the configuration) is known; a $d^8$ Ni²⁺ center behaving as $^3F$-derived $^3A_{2g}$ in an octahedral field has starkly different EPR behavior than the same ion in a term-mixed low-symmetry site. In bioinorganic chemistry, the pale color and slow ligand-substitution kinetics of high-spin Mn²⁺ and Fe³⁺ centers in metalloproteins (superoxide dismutases, ribonucleotide reductases) trace directly to the $^6S$ ground term derived above — an all-spins-parallel, $L=0$ state that makes d-d transitions both spin- and Laporte-forbidden, giving these ions famously weak, sharp absorption features used as spectroscopic handles by researchers studying enzyme active sites. In phosphor and pigment chemistry, lanthanide and transition-metal dopants in LEDs and lasers are selected in part by engineering which terms lie energetically accessible, exploiting exactly the same $(L,S)$ bookkeeping extended to $f^n$ configurations.

Ask the tutor: For a $d^6$ configuration like Fe²⁺, can you derive the full term list by hole equivalence with $d^4$, and verify the degeneracy sum equals $\binom{10}{6}=210$?

Consider five degenerate d orbitals, labeled by their magnetic quantum number $m_l = +2, +1, 0, -1, -2$. Each orbital can hold two electrons of opposite spin, $m_s = +\tfrac12$ or $-\tfrac12$. A microstate is one specific, legal way of placing all $n$ electrons among these ten spin-orbitals (5 orbitals × 2 spins), consistent with the Pauli exclusion principle: no two electrons may share both the same $m_l$ and the same $m_s$. For a given configuration $d^n$, the total number of microstates is a combinatorial count,

because we are choosing $n$ of the 10 available spin-orbitals with no further restriction (Pauli exclusion is already built into "choosing distinct spin-orbitals"). For $d^1$, $N=10$; for $d^2$, $N=45$; for $d^3$, $N=120$; for $d^5$, $N=252$. Every one of these microstates is a valid quantum state of the free ion, and every one must appear inside some term. Each microstate carries two additive quantum numbers,

obtained just by summing the individual orbital and spin projections of the occupied electrons. These are good quantum numbers for any single microstate, but a term is characterized by the total $L$ and $S$, which are the maximum values of $|M_L|$ and $|M_S|$ that appear within a self-contained multiplet: a term with quantum numbers $(L,S)$ contributes exactly $(2L+1)(2S+1)$ microstates, one for every combination $M_L = -L,\dots,+L$ and $M_S = -S,\dots,+S$. The derivation procedure exploits this multiplicity directly. Build a microstate table: rows indexed by $M_S$ (from $+S_{max}$ down to $-S_{max}$), columns indexed by $M_L$ (from $+L_{max}$ down to $-L_{max}$), and each cell filled with the number of microstates having that particular $(M_L, M_S)$. Then repeatedly extract the term with the largest $M_L$ present at the largest $M_S$ present: that $(M_L^{max}, M_S^{max})$ pair identifies $L = M_L^{max}$ and $S = M_S^{max}$, and the term $^{2S+1}L$ must account for exactly one microstate in every cell $(M_L, M_S)$ with $|M_L|\le L$ and $|M_S| \le S$. Subtract that whole rectangular block from the table and repeat until the table is empty. Every subtraction step corresponds to identifying and removing one full term.

Notice the pattern of degeneracies: $^1G$ contributes $(2\cdot4+1)(2\cdot0+1) = 9$ states, $^3F$ contributes $(2\cdot3+1)(2\cdot1+1)=21$, $^1D$ contributes $5$, $^3P$ contributes $9$, and $^1S$ contributes $1$. These sum to $9+21+5+9+1=45$, exactly the total microstate count — a mandatory internal check that catches almost every bookkeeping error. The letter code for $L$ follows the historical atomic spectroscopy convention, directly analogous to the s, p, d, f orbital labels but capitalized and shifted in meaning (they now label total orbital angular momentum of the many-electron system, not a single orbital):

Do not confuse the term symbol letter "$S$" (meaning $L=0$) with the spin quantum number $S$ — the notation is unfortunately overloaded, and the multiplicity prefix $2S+1$ (spin) must be read separately from the italic $L$-letter (orbital). A second physical shortcut, electron–hole equivalence, halves the labor for configurations beyond the middle of the shell. A shell missing $n$ electrons ($d^{10-n}$, i.e., containing $n$ holes) generates exactly the same set of terms as a shell containing $n$ electrons, because the Coulomb repulsion pattern among $n$ holes mirrors that among $n$ electrons (a hole is simply the absence of an electron, and pairwise repulsion energies among holes track those among particles by symmetry of the interaction). Thus $d^1$ and $d^9$ share terms ($^2D$ only), $d^2$ and $d^8$ share terms, $d^3$ and $d^7$ share terms, and $d^4$ and $d^6$ share terms. Only $d^5$ stands alone at the half-filled midpoint.

The total number of microstates for $d^n$ is

Where: $10$ is the number of spin-orbitals available (5 orbitals × 2 spin states), and $n$ is the number of d electrons. Reading: this is not an approximation — it is an exact count of Pauli-allowed arrangements, and it must equal the sum of $(2L+1)(2S+1)$ over every term you extract. The degeneracy of a single term is

where: $2L+1$ counts the orbital sublevels ($M_L = -L,\dots,+L$) and $2S+1$ counts the spin sublevels ($M_S = -S,\dots,+S$) that remain degenerate in the absence of spin-orbit coupling and external fields. Reading: this is the term's contribution to the microstate ledger, and summing it over all terms of a configuration must reproduce $N(d^n)$ exactly — the master consistency check of the whole method. The total orbital and spin angular momentum projections of any microstate are

where: the sums run over the $n$ occupied electrons, each carrying its own $m_l \in \{-2,-1,0,1,2\}$ and $m_s \in \{-\tfrac12,+\tfrac12\}$. Reading: these projections are simply additive — no vector coupling algebra is needed to build the table, only arithmetic; the coupling machinery (Clebsch–Gordan-like reasoning) is hidden inside the extraction step, where the maximum $M_L$ observed at the maximum $M_S$ tells you the true $(L,S)$ of a term.

Example 1 — Foundational: $d^1$ (Ti³⁺, V⁴⁺). A single d electron has $n=1$, so

With only one electron, $M_L$ can be any of $2,1,0,-1,-2$ and $M_S$ can be $+\tfrac12$ or $-\tfrac12$, independently — every one of the $5\times2=10$ combinations occurs exactly once. The maximum $M_L$ is $2$, occurring at both $M_S=+\tfrac12$ and $M_S=-\tfrac12$, so $L=2$ and $S=\tfrac12$: a single term, $^2D$, with degeneracy

which exactly exhausts all ten microstates — no other term exists. This is the simplest possible check of the method: one electron, one term, and the multiplicity count closes immediately. By hole equivalence, $d^9$ (Cu²⁺) also has the sole term $^2D$; as a limiting-case sanity check, setting $n\to 10-n$ with $n=1$ gives $n=9$, and indeed both configurations independently enumerate to the single term $^2D$ — the electron-hole symmetry is not assumed but reproduced by direct counting.

Example 2 — Applied: $d^8$ (Ni²⁺, a common octahedral and square-planar center). Rather than building a 45-row table for $\binom{10}{8}=45$ microstates directly, invoke hole equivalence: $d^8$ has $10-8=2$ holes, and 2 holes generate the identical term set as 2 electrons, i.e. exactly the $d^2$ result derived in full above:

This is not a coincidence to be taken on faith — the same microstate-counting exercise carried out for two holes in the ten spin-orbitals gives $\binom{10}{2}=45$ hole-microstates, and grouping them by total hole angular momentum reproduces precisely the same $(L,S)$ combinations, because the pairwise Coulomb repulsion structure that determines allowed $(M_L,M_S)$ combinations is symmetric under particle–hole exchange in a closed formal sense. The practical payoff: for Ni²⁺ (d⁸), the ground term (by Hund's rules, treated fully elsewhere) is $^3F$, the same symbol that anchors the ground state of $d^2$ ions like V³⁺ — which is exactly why V³⁺ and Ni²⁺ complexes show qualitatively parallel (inverted-order) Tanabe–Sugano diagrams and are treated together in ligand-field correlation diagrams.

Example 3 — Competition-level: $d^5$ (Mn²⁺, Fe³⁺), the self-paired midpoint case. This configuration has no hole-equivalence partner (it partners only with itself, since $10-5=5$), so it must be derived directly, and it is the most demanding case because $\binom{10}{5}=252$ microstates must be sorted. Begin at the top of the table. The maximum $M_S$ is $\tfrac52$, requiring all five electrons to have parallel spin — which forces them into all five different orbitals (Pauli), i.e. $m_l = 2,1,0,-1,-2$ each singly occupied with spin up. There is exactly one such microstate, and its $M_L = 2+1+0-1-2 = 0$. So the entire $M_S=\tfrac52$ row has a single entry, at $M_L=0$. This immediately identifies the term: $L=0$ (since $M_L=0$ is the only value present at the maximal spin) and $S=\tfrac52$, giving $^6S$, with degeneracy

This matches the physical intuition that a half-filled shell with all spins parallel is a single, uniquely symmetric state — the famous "extra stability" configuration behind Mn²⁺'s large positive $\Delta_{oct}$ resistance and its weak, spin-forbidden d-d bands (Mn²⁺ complexes are famously pale, because every excited term has a different $S$, making transitions from the $^6S$ ground term spin-forbidden as well as Laporte-forbidden). Removing this term's contribution (which only touched $M_S=\tfrac52$ and $M_S=-\tfrac52$ rows, one entry each, and $M_S=\tfrac32,\tfrac12,-\tfrac12,-\tfrac32$ each lose one entry at $M_L=0$ as part of the same $S=\tfrac52$ multiplet — recall the full $^6S$ block spans $M_S = -\tfrac52 \ldots \tfrac52$ all at $M_L=0$), the remaining table must be worked through the same maximum-$M_L$-at-maximum-remaining-$M_S$ procedure. Carrying this out completely (a longer but entirely mechanical repetition of the same six-step pattern used for $d^2$ and $d^3$) yields the full known term list for $d^5$:

As a check on just the doublet-and-sextet multiplicities without re-deriving every step, sum the degeneracies:

which equals $\binom{10}{5} = 252$ exactly, confirming the full sixteen-term list (twenty counting repeats). This is the largest and most term-rich configuration in the entire $3d$ series — a direct consequence of the combinatorial maximum of $\binom{10}{n}$ occurring at $n=5$ — and it is why Mn²⁺/Fe³⁺ spectra, though individually weak, contain the richest set of possible spin-forbidden transitions of any dⁿ ion.

The formal machinery of term symbols descends from atomic spectroscopy of the early twentieth century, but the version used throughout inorganic chemistry — coupling to build $L$ and $S$ separately before invoking spin-orbit coupling — is named for the physicists who established when this coupling scheme is valid.