Crystal Field Theory Revisited: The Point-Charge Model

Introduction

Crystal field theory is often taught as a finished energy diagram. At university level, the diagram should be traced to a controlled approximation: ligands are represented as point charges or dipoles fixed around a metal ion, and metal d electrons experience their directional electrostatic field. The model predicts the pattern of splitting well, but its numerical energies and ligand trends require caution because actual bonds are not point charges.

Core explanation

For an isolated spherical metal ion, five d orbitals are degenerate within the one-electron central-field approximation. Put six identical negative ligand charges on the +x, −x, +y, −y, +z and −z axes. The d(x²−y²) and d(z²) lobes face these charges and experience relatively strong electron–electron repulsion. The d(xy), d(xz) and d(yz) lobes lie between the axes, so their repulsion is relatively smaller. The first pair becomes the higher e g set and the second triplet the lower t₂g set. The gap is Δₒ.

The electrostatic field raises the d-electron energy relative to a remote free-ion reference, but CFSE uses a different reference: the spherical average of the ligand perturbation. Relative to that barycentre, the lower three orbitals sit at −0.4Δₒ each and the upper two at +0.6Δₒ. The coefficients follow from 3E(t₂g)+2E(e g)=0 and E(e g)−E(t₂g)=Δₒ. Preserving this weighted mean isolates directional splitting; it does not say that bringing ligands near a metal has zero total energy change.

The point-charge calculation assumes a chosen geometry, identical fixed ligands and an electrostatic interaction independent of metal–ligand orbital overlap. It treats d electrons as mostly metal-centred and commonly uses one-electron orbital energies plus a separate pairing parameter. This simplification yields clear predictions for high- versus low-spin fillings and relative CFSE, but does not explicitly explain σ-bonding, π donation or π back-bonding.

One immediate warning is the spectrochemical series. Neutral CO and CN⁻ are often strong-field ligands, while some anionic halides are weak field. A charge-only ranking cannot account for that pattern. Molecular-orbital ligand-field theory explains how ligand orbitals of suitable symmetry mix with metal d-like orbitals and alter Δₒ. The point-charge diagram remains a valuable effective description of the resulting level order, but should not be mistaken for a complete bonding mechanism.

Geometry changes the answer. Four tetrahedral ligands approach between axes, making the two e orbitals lower and three t₂ orbitals higher, with a smaller gap Δ t. Removing or moving octahedral ligands can create tetragonal or square-planar patterns. A point-charge model can predict qualitative changes from positions, but real bond distances, covalency and distortions alter numerical values.

Finally, a split d-orbital diagram is not a full many-electron spectrum. Electron–electron repulsion creates terms, and optical transitions connect many-electron states. Selection rules affect which bands are intense. Pure CFT provides a scaffold for that analysis, not a direct one-to-one mapping from every observed absorption peak to Δₒ.

Step-by-step reasoning

State geometry and ligand directions. Compare each d orbital’s lobes with those directions to rank electrostatic repulsion. Use symmetry to group equal-energy orbitals, then place their weighted barycentre and calculate relative level shifts. Fill d electrons with an explicit pairing assumption, and name omitted covalent and many-electron effects before interpreting measured data.

Visual explanation

Draw an octahedral x/y/z axis cross with ligand charges at six ends. Shade d(x²−y²) and d(z²) lobes pointing at charges; draw d(xy), d(xz), d(yz) between them. Beside it show the fivefold line splitting into three-low and two-high lines about a dashed barycentre.

Real-world analogy

People standing directly in front of six loudspeakers hear a stronger signal than people standing between them. Direction relative to the sources matters even when every speaker has the same power. Orbital shapes similarly sample an anisotropic ligand field differently.

Real-world example

Octahedral [Ti(H₂O)₆]³⁺ is d¹. Its single d electron occupies a t₂g-like orbital, and an electronic excitation toward an e g-like state provides a comparatively simple illustration of the field gap. More-electron ions require additional term analysis.

Why?

Why do three orbitals move down by 0.4Δₒ while two move up by 0.6Δₒ relative to the barycentre? Their degeneracies differ. Three smaller downward shifts must balance two larger upward shifts so the weighted average stays at the chosen reference.

Common misconception

“Point-charge CFT says the metal–ligand bond is entirely ionic.” It is an idealised model of one electronic effect. Actual coordination bonds can have substantial covalent character, which ligand-field theory represents explicitly.

Worked example

Assume an ideal octahedral d² ion with Δₒ=18,000 cm⁻¹. Its electrons occupy two different t₂g orbitals, so orbital CFSE=2(−0.4Δₒ)=−0.8Δₒ=−14,400 cm⁻¹. This is relative to the spherical-field barycentre, not a prediction that complex formation releases exactly 14,400 cm⁻¹ per ion. A measured absorption band must also be assigned to an allowed many-electron transition.

Quick check

1. Which octahedral orbitals point most directly at axial ligand charges? Answer: d(x²−y²) and d(z²), the e g pair.

Exam focus

State assumptions before using the diagram: ideal geometry, equal ligands and electrostatic approximation. Distinguish level ordering from a complete prediction of bonding or spectral intensity.

Advanced insight

In group-theoretic language, an octahedral perturbation preserves equality only among orbitals belonging to the same irreducible representation. The two-plus-three split reflects symmetry as well as the intuitive orientation argument.

Summary

The point-charge model explains octahedral e g/t₂g ordering through directional repulsion and the barycentre coefficients through degeneracy. Its simplicity supports CFSE reasoning while leaving covalency and detailed spectra to richer models.

Practice questions

1. Why does octahedral d(xy) lie below d(x²−y²) in point-charge CFT? Answer: d(xy) points between x and y ligand axes, whereas d(x²−y²) points directly along those axes toward ligand charges and experiences greater repulsion. 2. Derive e g energy if t₂g is −0.4Δₒ. Answer: The gap is Δₒ, so E(e g)=−0.4Δₒ+Δₒ=+0.6Δₒ; the weighted mean also checks as 3(−0.4)+2(+0.6)=0. 3. Name a ligand trend that pure charge-based CFT struggles to explain. Answer: Neutral CO is often strong field despite having no formal negative charge, while halide anions can be comparatively weak field; covalent π interactions matter. 4. Does the barycentre rule imply zero total metal–ligand binding energy? Answer: No. It only balances the directional d-level shifts relative to a spherical ligand-field reference.