💎 Crystal Field Theory — d-Orbital Splitting & Transition Metal Colors

Crystal field theory describes how the electric field created by the ligands surrounding a transition-metal ion splits the five, otherwise degenerate, d-orbitals into two (or more) energy levels. Orbital lobes that point directly at approaching ligands feel stronger electrostatic repulsion and rise in energy; lobes that point between ligands feel less repulsion and drop in energy, relative to a hypothetical spherical (free-ion) field.

In an octahedral complex, six ligands approach along the ±x, ±y and ±z axes. The dz² and dx²−y² orbitals point directly along those axes, so they are pushed up into the higher-energy eg set. The dxy, dxz and dyz orbitals point between the axes, are repelled less, and are pushed down into the lower-energy t2g set — three orbitals stabilized by −0.4Δo each, two destabilized by +0.6Δo each.

It's a competition between the crystal field splitting energy Δo and the pairing energy P, the extra energetic cost of forcing two electrons into the same orbital. When Δo is smaller than P, electrons spread out into eg before pairing (high-spin, maximizing unpaired electrons). When Δo exceeds P, electrons pair up in t2g first because that costs less energy than jumping to eg (low-spin). This choice only exists for d4–d7 configurations.

The spectrochemical series is an empirically-derived ranking of common ligands from weak-field to strong-field, based on how large a Δo splitting each induces: roughly I⁻ < Br⁻ < Cl⁻ < F⁻ < H2O < NH3 < en < CN⁻ ≈ CO. It's used to predict whether a given metal-ligand combination will be high-spin or low-spin, and to estimate the color of the resulting complex.

Δo often corresponds to the energy of a visible-light photon, so a d-electron can be promoted from t2g to eg by absorbing a specific wavelength of visible light. The complex then appears the complementary color of whatever wavelength it absorbed. Main-group compounds have no partially filled d-shell and no such low-energy transition available, so they typically don't absorb visible light and appear colorless.

A d0 ion has no d-electrons to promote, and a d10 ion has no empty eg orbital to promote an electron into — in both cases, no t2g→eg transition is possible at all. Since Δo excitation is essentially the only source of visible-light absorption available to a simple transition-metal complex, d0 and d10 complexes are typically colorless or only very weakly colored.

For d1–d3, there are three empty t2g orbitals and not enough electrons to force a choice — they simply fill singly by Hund's rule. For d8–d10, the t2g set is already completely full (6 electrons) no matter what, so the only question is how the remaining electrons fill the two eg orbitals, which again follows Hund's rule without any pairing-vs-splitting trade-off. Only d4–d7 have enough electrons to create a genuine high-spin/low-spin choice.

CFSE quantifies the net energy lowering a configuration gains from occupying the split t2g/eg levels instead of a hypothetical undivided level, using CFSE = −0.4·n(t2g) + 0.6·n(eg) in units of Δo. Configurations with large negative CFSE (like low-spin d6) are especially stable, which helps explain trends in hydration enthalpies, lattice energies, and reaction rates across the transition series.

Any configuration with unpaired electrons is paramagnetic (weakly attracted into a magnetic field); a configuration with all electrons paired is diamagnetic. Crystal field theory reaches this conclusion via electrostatic orbital splitting, a completely different mechanism from the molecular-orbital-diagram picture used elsewhere on this site, where paramagnetism instead comes from unpaired electrons in bonding/antibonding molecular orbitals — yet both predict the same magnetic behavior for many real molecules.