Van Vleck's 1958 Faraday Discussion paper on the magnetic behaviour of regular and inverted crystalline energy levels closed with four open questions about the non-cubic (trigonal/tetragonal) parts of the crystal field acting on 3d ions: (i) deriving realistic values of the low-order angular coefficients a2 and a4 for specific cation–ligand clusters — the splittings inferred for the alums implied a ratio Δ4/Δ5 as large as ∼10 that no plausible electrostatic cluster model could reproduce; (ii) the magnitude and ion-dependence of the cation 'back action' (ligand polarization and structural response); (iii) the quantitative interplay of Jahn–Teller and spin–orbit coupling, including the asserted one-tenth reduction of linear Jahn–Teller matrix elements within the spin–orbit manifold of the Fe²⁺ ⁵DΓ5 triplet; and (iv) the coordination-geometry dependence of the harmonic coefficients and radial integrals. What survives as open is a common core of these questions: a quantitative microscopic determination of the non-cubic ligand-field coefficients and Jahn–Teller coupling constants for concrete materials — including the alum systems that originated the questions, for which no determination free of empirically fitted parameters is known — with their cation, temperature, and coordination dependence, consistent with the magnetic and spectroscopic data from which these quantities are inferred.

Background

Crystal-field theory describes how the electric field of the surrounding ligands lifts the degeneracies of a transition-metal ion's d-electron levels. In an octahedral environment the five 3d orbitals split into a lower t2g triplet and an upper eg doublet, and the free-ion orbital terms split correspondingly into components labelled by irreducible representations Γ of the cubic group. Van Vleck's 1958 paper contrasted the magnetic consequences of 'regular' versus 'inverted' orderings of these cubic components across the 3d series: which component lies lowest controls how strongly spin–orbit coupling (the relativistic coupling between the electron's spin and its orbital motion, of strength λ) mixes neighbouring levels, and hence how anisotropic the ion's magnetic response is. The observed pattern — anisotropies of tens of percent for Fe²⁺, Co²⁺, and Cu²⁺ versus fractions of a percent for Cr³⁺ and Ni²⁺ — was a central test of the theory.

Real sites are never exactly cubic. In the alums, M⁺M³⁺(SO₄)₂·12H₂O, the trivalent ion sits in a slightly trigonally distorted [M(H₂O)₆]³⁺ octahedron, and the non-cubic part of the ligand field is conventionally expanded in angular harmonics of second and fourth rank with coefficients a2 and a4, multiplied by the radial moments ⟨r²⟩ and ⟨r⁴⟩ of the 3d electrons. This field splits the cubic components further: van Vleck denoted by Δ5 the splitting of the orbital triplet of the cubic-field-split D term (d¹/d⁶ ions such as Ti³⁺ and Fe²⁺; ²D for d¹, ⁵D for d⁶) and by Δ4 that of the Γ4 triplet of the F term (d²/d⁷ ions such as V³⁺ and Co²⁺; ³F for d², ⁴F for d⁷), with linear relations Δ5 = a2⟨r²⟩ + a4⟨r⁴⟩ and Δ4 = −(1/10)a2⟨r²⟩ − (3/2)a4⟨r⁴⟩ (up to convention-dependent factors; tetragonal geometries give −2/5 and −3/4). Electron paramagnetic resonance (EPR) — the microwave spectroscopy of the spin sublevels of a paramagnetic ion — probes these splittings through the anisotropic g-tensor (the orientation-dependent effective gyromagnetic factor linking the ion's magnetic moment to its spin) and through the trigonal-field parameters of the Abragam–Pryce spin-Hamiltonian, the standard effective Hamiltonian written in spin operators that parametrizes EPR spectra. Two further mechanisms complicate the picture: the Jahn–Teller (JT) effect, by which an electronically degenerate ion distorts its surroundings to lift the degeneracy, and 'back action' — the polarization and structural response of the ligand shell to the cation, which modifies a2 and a4 beyond any fixed point-charge model.

In the 1958 paper van Vleck identified four quantitative gaps: (i) no electrostatic cluster model could reproduce the trigonal splittings inferred for the alums — the inferred ratios Δ4/Δ5 of order 10 would require the 3d electron cloud to extend a full metal–ligand distance; (ii) the magnitude and ion-dependence of back action was unquantified, though then-available pressure-dependence, nuclear-quadrupole, and axis-reorientation data showed it was real; (iii) the interplay of JT and spin–orbit coupling needed microscopic calculation, illustrated by an asserted one-tenth reduction of linear JT matrix elements within the spin–orbit manifold of Fe²⁺'s ⁵DΓ5 triplet; and (iv) the coordination-geometry dependence (tetrahedral vs octahedral, and distortions thereof) of the harmonic coefficients and radial integrals was uncharted. The subsequent six decades brought covalent molecular-orbital ligand-field theory, the dynamic Jahn–Teller theory of paramagnetic crystals (whose reduction factors quantify how coupling to lattice vibrations renormalizes, and typically suppresses, electronic matrix elements such as spin–orbit and orbital operators), the superposition and exchange-charge models that parametrize non-cubic fields in terms of intrinsic ligand parameters and geometry power laws, and ab initio ligand-field theory, which extracts ligand-field parameters from multireference first-principles wavefunctions.

Problem Statement

The problem is to give a quantitative microscopic account of the non-cubic ligand field and vibronic coupling of 3d ions in specific cation–ligand clusters, comprising the four questions with which van Vleck's 1958 paper closed, all four of which are required; the concrete systems to be treated are the alum systems that originated the questions, the Fe²⁺-oxide systems named under (iii), and tetrahedrally coordinated 3d systems, to be selected by the solver with cited measured splittings, for the coordination-number comparison required by (iv). (i) Derive realistic values of the low-order angular coefficients a2 and a4 together with the 3d radial moments ⟨r²⟩ and ⟨r⁴⟩ for concrete complexes, so that the non-cubic splittings Δ5 (d¹/d⁶ ions) and Δ4 (d²/d⁷ ions) and their ratio are predicted rather than fitted. The historical electrostatic models failed this outright, and the 1960 dissolution of the Δ4/Δ5 ∼ 10 puzzle — an artifact of comparing data taken at different temperatures and of different types under a fixed-electrostatic-field premise — explains the failure but does not itself supply the derivation; current determinations still fit at least one parameter to experiment. (ii) Quantify the cation 'back action' — the polarization and structural distortion of the ligand shell induced by the central ion — and its ion-dependence, which enters a2 and a4 beyond fixed point-charge or fixed-geometry models; existing treatments absorb it into fitted covalency or exchange-charge parameters rather than predicting it. (iii) Compute microscopically the interplay of Jahn–Teller and spin–orbit coupling for concrete materials (the Ti³⁺ and V³⁺ alums; Fe²⁺ in octahedral oxides such as FeO): the numerical Jahn–Teller coupling constants and their reduction within spin–orbit manifolds (of which the archetype is the factor A1/a = 1/10 for ⁵DΓ5, the ratio of the linear Jahn–Teller coefficient in the J' = 1 spin–orbit manifold to its unprojected value, as derived by van Vleck in his 1960 follow-up), the crossover criteria between spin–orbit-dominated and Jahn–Teller-dominated regimes, and the temperature dependence of the effective splittings through the materials' structural dynamics. General model-level criteria and semi-empirical vibronic treatments exist, but first-principles Jahn–Teller constants for these systems do not. (iv) Give a systematic quantitative account of how the harmonic coefficients and radial integrals change with coordination number and ligand geometry; superposition- and exchange-charge-model workflows now derive geometry power-law exponents per system but retain fitted parameters. Part (iv) is the derivation of parameter-free coefficients, radial moments, and per-system geometry power-law exponents for the six-coordinate and tetrahedral systems, reproducing those systems' measured splittings within the uncertainties reported by the sources cited for them. The items under (i)–(iii) are derived parameter-free and together reproduce the measured data cited for the systems treated within the stated uncertainties: for (i), the coefficients a2 and a4 together with the radial moments ⟨r²⟩ and ⟨r⁴⟩, reproducing the observed Δ5, Δ4, and their ratio; for (ii), the magnitude and ion-dependence of the back action; and for (iii), the first-principles Jahn–Teller coupling constants with their spin–orbit reduction factors, the crossover criteria, and the temperature dependence of the effective splittings. Part (iv) presupposes (i) and does not discharge it. Because the sought coefficients and coupling constants are not directly observable — the historical values, including the alum ratio Δ4/Δ5 ∼ 10, were inferred from EPR g-tensors and spin-Hamiltonian parameters, magnetic susceptibilities as functions of temperature, and optical spectra — derived values must reproduce such data, within the uncertainties of the original measurements, for the systems treated, beginning with the alum systems that originated the questions; no first-principles treatment of the alums' non-cubic fields or Jahn–Teller couplings is known.

References

  1. J. H. van Vleck. II. Energetics of complexes. The magnetic behaviour of regular and inverted crystalline energy levels. Discussions of the Faraday Society 26, 96 (1958). DOI: 10.1039/df9582600096. URL: https://doi.org/10.1039/df9582600096.
  2. A. Abragam, M. H. L. Pryce. The theory of paramagnetic resonance in hydrated cobalt salts. Proceedings of the Royal Society of London. Series A 206, 173–191 (1951). DOI: 10.1098/rspa.1951.0063. URL: https://doi.org/10.1098/rspa.1951.0063.
  3. H. A. Jahn, E. Teller. Stability of polyatomic molecules in degenerate electronic states. I. Orbital degeneracy. Proceedings of the Royal Society of London. Series A 161, 220–235 (1937). DOI: 10.1098/rspa.1937.0142. URL: https://doi.org/10.1098/rspa.1937.0142.
  4. U. Öpik, M. H. L. Pryce. Studies of the Jahn-Teller effect. I. A survey of the static problem. Proceedings of the Royal Society of London. Series A 238, 425–447 (1957). DOI: 10.1098/rspa.1957.0010. URL: https://doi.org/10.1098/rspa.1957.0010.
  5. H. C. Longuet-Higgins, U. Öpik, M. H. L. Pryce, R. A. Sack. Studies of the Jahn-Teller effect. II. The dynamical problem. Proceedings of the Royal Society of London. Series A 244, 1–16 (1958). DOI: 10.1098/rspa.1958.0022. URL: https://doi.org/10.1098/rspa.1958.0022.
  6. A. Bose, A. S. Chakravarty, R. Chatterjee. The nature of the crystalline fields in Ti3+ alum. Proceedings of the Royal Society of London. Series A 255, 145–151 (1960). DOI: 10.1098/rspa.1960.0059. URL: https://doi.org/10.1098/rspa.1960.0059.
  7. J. H. van Vleck. Note on the gyromagnetic ratio of Co++ and on the Jahn-Teller effect in Fe++. Physica 26, 544–552 (1960). DOI: 10.1016/0031-8914(60)90107-5. URL: https://doi.org/10.1016/0031-8914(60)90107-5.
  8. A. D. Liehr. Ionic Radii, Spin-Orbit Coupling and the Geometrical Stability of Inorganic Complexes. Bell System Technical Journal 39, 1617–1626 (1960). DOI: 10.1002/J.1538-7305.1960.TB01618.X. URL: https://doi.org/10.1002/J.1538-7305.1960.TB01618.X.
  9. A. Bose, R. Chatterjee, R. Rai. Molecular orbital theory of the ligand field in hydrated V3+ salts. Proceedings of the Physical Society 83, 959–965 (1964). DOI: 10.1088/0370-1328/83/6/308. URL: https://doi.org/10.1088/0370-1328/83/6/308.
  10. G. F. Dionne. Electron paramagnetic resonance experiments with Ti3+ ions in RbAl(SO4)2·12H2O. Canadian Journal of Physics 42, 2419–2428 (1964). DOI: 10.1139/P64-220. URL: https://doi.org/10.1139/P64-220.
  11. M. C. M. O'Brien. The dynamic Jahn-Teller effect in octahedrally co-ordinated d9 ions. Proceedings of the Royal Society of London. Series A 281, 323–339 (1964). DOI: 10.1098/rspa.1964.0185. URL: https://doi.org/10.1098/rspa.1964.0185.
  12. H. M. Gladney, J. D. Swalen. Theory of EPR of Ti3+ in Trigonal Environments. The Journal of Chemical Physics 42, 1999–2010 (1965). DOI: 10.1063/1.1696237. URL: https://doi.org/10.1063/1.1696237.
  13. M. D. Sturge. The Jahn-Teller Effect in Solids. Solid State Physics 20, 91–211 (1968). DOI: 10.1016/S0081-1947(08)60218-0. URL: https://doi.org/10.1016/S0081-1947(08)60218-0.
  14. A. Bose, U. S. Ghosh. Dependence of ligand fields in paramagnetic crystals on thermal changes in their structures. Radiation Effects 4, 275–279 (1970). DOI: 10.1080/00337577008242014. URL: https://doi.org/10.1080/00337577008242014.
  15. F. S. Ham. Jahn-Teller Effects in Electron Paramagnetic Resonance Spectra. In: Electron Paramagnetic Resonance, pp. 1–119. Springer, Boston, MA (1972). DOI: 10.1007/978-1-4899-5310-0_1. URL: https://doi.org/10.1007/978-1-4899-5310-0_1.
  16. C. A. Bates. Jahn-Teller effects in paramagnetic crystals. Physics Reports 35, 187–304 (1978). DOI: 10.1016/0370-1573(78)90122-9. URL: https://doi.org/10.1016/0370-1573(78)90122-9.
  17. J. K. Beattie, S. P. Best. Structures and spectroscopy of hexaaquametal(III) ions. Coordination Chemistry Reviews 166, 391–415 (1997). DOI: 10.1016/S0010-8545(97)00049-0. URL: https://doi.org/10.1016/S0010-8545(97)00049-0.
  18. L. Dubicki, M. J. Riley. Paramagnetism of caesium titanium alum. The Journal of Chemical Physics 106, 1669–1675 (1997). DOI: 10.1063/1.473320. URL: https://doi.org/10.1063/1.473320.
  19. P. L. W. Tregenna-Piggott, M. C. M. O'Brien, J. R. Pilbrow, H. U. Güdel, S. P. Best, C. Noble. Paramagnetism of caesium titanium alum and the Jahn-Teller interaction. The Journal of Chemical Physics 107, 8275–8291 (1997). DOI: 10.1063/1.475029. URL: https://doi.org/10.1063/1.475029.
  20. A. R. Lim, Y. Paik, K.-Y. Lim. NMR relaxation study of the phase transitions and relaxation mechanisms of the alums MCr(SO4)2·12H2O (M=Rb and Cs) single crystals. Journal of Solid State Chemistry 184, 1367–1372 (2011). DOI: 10.1016/j.jssc.2011.04.009. URL: https://doi.org/10.1016/j.jssc.2011.04.009.
  21. Fang Wang, Heng-Jie Chen, Chong-Geng Ma, Wen-Chen Zheng, Da-Xiao Yang, Rong-Li Gao. Systematic analysis of the fine structure of energy levels and spin-Hamiltonian parameters of V3+ ion in corundum with dynamic Jahn-Teller effect. Journal of Luminescence 208, 273–278 (2019). DOI: 10.1016/j.jlumin.2018.12.031. URL: https://doi.org/10.1016/j.jlumin.2018.12.031.
  22. S. V. Streltsov, D. I. Khomskii. The Jahn-Teller effect and spin-orbit coupling: friends or foes?. Physical Review X 10, 031043 (2020). DOI: 10.1103/PhysRevX.10.031043. URL: https://doi.org/10.1103/PhysRevX.10.031043.
  23. M. G. Brik. 3d Ions in Solids and Microscopic Crystal Field Effects: Theoretical Analysis and Relations with Experimental Spectroscopic Data. IOP Conference Series: Materials Science and Engineering 835, 012032 (2020). DOI: 10.1088/1757-899x/835/1/012032. URL: https://doi.org/10.1088/1757-899x/835/1/012032.
  24. X.-K. Hu, B. Wu, Y. Yang, Y. Y. Yeung, C.-G. Ma, M. G. Brik. An old system revisited: Al2O3:Ti3+ — Microscopic crystal field effects explored by the crystal field and first-principles calculations. Journal of Alloys and Compounds 847, 156459 (2020). DOI: 10.1016/j.jallcom.2020.156459. URL: https://doi.org/10.1016/j.jallcom.2020.156459.
  25. S. V. Rao, D. Maganas, K. Sivalingam, M. Atanasov, F. Neese. Extended Active Space Ab Initio Ligand Field Theory: Applications to Transition-Metal Ions. Inorganic Chemistry 63, 24672–24684 (2024). DOI: 10.1021/acs.inorgchem.4c03893. URL: https://doi.org/10.1021/acs.inorgchem.4c03893.
  26. V. R. M. Nielsen, M. Grasser, B. Le Guennic, T. J. Sørensen. Neodymium(III) Aqua Ion as a Model System in Ab Initio Crystal Field Analysis Beyond Point Charges and Crystal Field Theory. Inorganic Chemistry 64, 3463–3475 (2025). DOI: 10.1021/acs.inorgchem.4c05044. URL: https://doi.org/10.1021/acs.inorgchem.4c05044.