Pseudo Jahn–Teller Effect

The pseudo Jahn–Teller (PJT) effect—also referred to in inorganic and solid-state chemistry as the second-order Jahn–Teller (SOJT) effect—is a universal quantum-mechanical mechanism of structural symmetry breaking in polyatomic molecules, coordination complexes, and crystalline solids that lack exact orbital degeneracy in their ground electronic state. While the classical first-order Jahn–Teller effect requires a spatially degenerate ground state (such as ²E_g in octahedral Cu²⁺ complexes), the Pseudo Jahn–Teller Theorem formulated and proven by Isaac B. Bersuker and coworkers demonstrates that vibronic coupling between two or more non-degenerate electronic states is the sole physical origin of structural instability and spontaneous symmetry lowering in non-degenerate high-symmetry polyatomic configurations.

Whenever a molecule in a closed-shell or non-degenerate high-symmetry reference geometry (such as planar NH₃ in D_3h, octahedral XeF₆ in O_h, or cubic BaTiO₃ in Pm3̄m) possesses a low-lying excited state of appropriate symmetry separated by an energy gap Δ, an asymmetric normal vibrational displacement Q can mix the ground and excited electronic wavefunctions. If the resulting negative vibronic curvature exceeds the positive harmonic restoring force constant K₀ of the σ-bond framework, the high-symmetry configuration becomes a saddle point on the adiabatic potential energy surface, driving pyramidalization, stereochemically active lone-pair distortion, or ferroelectric polarization.

Interactive VSEPR Lab: Test trigonal pyramidal (NH₃), seesaw (SF₄), and pentagonal bipyramidal (XeF₆) lone-pair geometries in 3D inside our VSEPR Lab!

The Pseudo Jahn–Teller Hamiltonian and Curvature Softening

Consider a molecule at a high-symmetry reference nuclear configuration (Q = 0) with a non-degenerate ground electronic state |Ψ₁⟩ of symmetry Γ₁ and a non-degenerate low-lying excited electronic state |Ψ₂⟩ of symmetry Γ₂, separated by the vertical excitation energy gap Δ = E₂(0) − E₁(0) > 0 (sometimes denoted 2Δ in alternative conventions). Let Q be a symmetry-lowering normal vibrational coordinate belonging to the irreducible representation Γ_Q. By group theory, linear off-diagonal vibronic coupling between |Ψ₁⟩ and |Ψ₂⟩ is nonzero if and only if the direct product Γ₁ ⊗ Γ_Q ⊗ Γ₂ contains the totally symmetric representation A₁ (or A_1g)—meaning Γ_Q = Γ₁ ⊗ Γ₂.

In the two-state basis {|Ψ₁⟩, |Ψ₂⟩}, expanding the electronic Hamiltonian along Q to second order yields the 2 × 2 pseudo Jahn–Teller vibronic matrix:

• Diagonal elements: H₁₁(Q) = −½Δ + ½K₀Q² and H₂₂(Q) = +½Δ + ½K₀Q², where K₀ = ⟨Ψ₁ | (∂²H/∂Q²)₀ | Ψ₁⟩ > 0 is the primary (bare) harmonic force constant arising from the fixed ground-state electron density. As proved by Bersuker, K₀ is strictly positive for any equilibrium internuclear distance—meaning a fixed electron cloud can never drive a symmetry-breaking distortion on its own.
• Off-diagonal coupling elements: H₁₂(Q) = H₂₁(Q) = F Q, where F = ⟨Ψ₁ | (∂H/∂Q)₀ | Ψ₂⟩ is the off-diagonal linear vibronic coupling constant (typically matching the symmetry of the highest occupied molecular orbital, HOMO, and lowest unoccupied molecular orbital, LUMO).

Diagonalizing this 2 × 2 secular determinant yields the two adiabatic potential energy branches E_±(Q):

E_±(Q) = ½ K₀ Q² ± ½ √(Δ² + 4 F² Q²)

The Instability Threshold (4F² / K₀ > Δ) and Double-Well Potential

Expanding the lower adiabatic potential energy surface E₋(Q) in a Taylor series around the high-symmetry origin Q = 0 reveals how excited-state mixing modifies the vibrational curvature:

E₋(Q) ≈ −½ Δ + ½ [ K₀ − (2 F² / Δ) ] Q² + (F⁴ / Δ³) Q⁴ − …

The effective harmonic force constant of the ground state along coordinate Q is therefore K_eff = K₀ − K_v, where K_v = 2F² / Δ (or 4F² / Δ when Δ is defined as the half-gap) is the negative vibronic contribution arising from relaxation of the electron cloud as |Ψ₂⟩ mixes into |Ψ₁⟩. Two distinct physical regimes emerge depending on the ratio of vibronic coupling to the energy gap:

1. Weak Pseudo Jahn–Teller Coupling (2F² / Δ < K₀, or Δ > 2F² / K₀): The effective curvature K_eff remains positive, so the high-symmetry geometry (Q = 0) stays a stable energy minimum, but the vibrational mode Q is strongly softened (its harmonic frequency ω_eff = √(K_eff / μ) drops substantially below the bare frequency), and the excited state E₊(Q) is correspondingly stiffened (K₊ = K₀ + 2F² / Δ).
2. Strong Pseudo Jahn–Teller Instability (2F² / Δ > K₀, or under the half-gap convention 4F² / K₀ > Δ): The negative vibronic term overwhelms the bare elastic stiffness K₀, rendering K_eff < 0. The high-symmetry point Q = 0 becomes a local maximum (transition state), and the molecule spontaneously distorts into two equivalent minima at Q₀ = ± √[ (F² / K₀²) − (Δ² / 4F²) ], separated by aPJT inversion barrier ΔE_PJT. In the limit where the energy gap vanishes (Δ → 0), the two-level PJT equations reduce smoothly to the linear Jahn–Teller effect with a conical intersection at Q = 0.

Umbrella Inversion of Ammonia (NH₃: D₃h → C₃v)

In introductory chemistry, the trigonal-pyramidal molecular geometry of ammonia (NH₃, H–N–H bond angle 107.8°, point group C_3v) is rationalized heuristically via VSEPR lone-pair repulsion or Walsh diagrams. In rigorous quantum chemistry, the pyramidalization of planar D_3h NH₃ is a textbook (A₁′ + A₂″) ⊗ a₂″ pseudo Jahn–Teller distortion:

• In planar D_3h NH₃, the HOMO is the non-bonding nitrogen 2p_z lone-pair orbital of a₂″ symmetry (giving a ¹A₁′ ground state), while the LUMO is the N–H σ-antibonding Rydberg-valence orbital of a₁′ symmetry (dominated by nitrogen 3s and hydrogen 1s character), producing a low-lying ¹A₂″ excited state.
• The out-of-plane umbrella bending mode Q₂ transforms as a₂″ (= A₁′ ⊗ A₂″). Displacing the nitrogen atom out of the H₃ plane along Q₂ mixes the a₂″ HOMO (2p_z) with the a₁′ LUMO (3s–σ*), creating a hybridized sp³-like directional lone pair and lowering the HOMO energy.
• Because 2F² / Δ exceeds the bare planar bending force constant K₀, planar D_3h NH₃ is unstable and relaxes into two symmetric C_3v pyramidal minima at Q₀ = ±0.38 Å, separated by an inversion barrier of 24.2 kJ mol⁻¹ (2,020 cm⁻¹) through which the protons quantum-mechanically tunnel at 23.786 GHz (the ammonia maser transition). In heavier pnictogen hydrides (PH₃, AsH₃, SbH₃), poorer np–ns spatial overlap or modified energy denominators yield even stronger pyramidalization (bond angles ~92–93.5°) and dramatically higher inversion barriers (~130–160 kJ mol⁻¹), suppressing Berry pseudorotation and quantum tunneling at room temperature.

Post-Transition Metal ns² Lone Pairs (Pb²⁺, Bi³⁺, Sb³⁺, XeF₆)

Heavy post-transition metal cations with formal ns²np⁰ valence configurations—such as In⁺, Sn²⁺, Sb³⁺, Te⁴⁺ (5s²) and Tl⁺, Pb²⁺, Bi³⁺ (6s²), as well as hypervalent noble-gas fluorides like XeF₆—exhibit a striking dichotomy: in some coordination environments the ns² electron pair remains spherically symmetric (stereochemically inert, as in octahedral Cs₂TeBr₆ or rocksalt PbS), whereas in others the cation shifts strongly off-center to form a hemidirected coordination sphere (stereochemically active lone pair, as in tetragonal litharge α-PbO, BiI₃, or C_3v-distorted XeF₆).

Because a pure spherical s² shell has a_1g symmetry in an octahedral (O_h) ligand field and cannot break symmetry on its own, its stereochemical activity is governed by a (A_1g + T_1u) ⊗ t_1u pseudo Jahn–Teller interaction strengthened by ligand-mediated orbital mixing:

1. Cation s– Ligand p Antibonding (HOMO): In an octahedral coordination complex such as [SbF₆]³⁻ or solid α-PbO, the filled cation ns orbital (a_1g) interacts in phase and out of phase with the symmetry-adapted a_1g combination of ligand 2p lone pairs. The resulting HOMO is the strongly σ-antibonding a_1g* (cation ns − ligand np) molecular orbital, while the empty cation np orbitals form the low-lying t_1u* LUMO.
2. Off-Center t_1u Distortion: An odd-parity polar displacement of the metal cation relative to the anion cage (mode Q_t1u) mixes the filled a_1g* HOMO with the empty cation np (t_1u*) LUMO. This second-order mixing localizes an asymmetric, lobe-shaped electron density on one side of the cation and converts three short covalent bonds and three long secondary bonds.
3. Why Fluoride and Oxide Activate the Lone Pair: When the ligands are electronegative, low-lying O²⁻ or F⁻ ions (as in PbO, Bi₂O₃, or SbF₃), their valence 2p orbitals match the energy of the relativistic 6s²/5s² cation shell closely, producing a high-lying a_1g* HOMO, a small HOMO–LUMO gap Δ, a large coupling constant F, and a strong PJT distortion. Conversely, with softer, more diffuse halides or chalcogenides (Br⁻, I⁻, Te²⁻) or in hexacoordinate [SbBr₆]³⁻, the ligand p orbitals lie higher in energy, weakening a_1g* antibonding, enlarging the effective gap Δ, and leaving the ns² pair closer to the regular octahedral center (or dynamically averaging across the eight C_3v trigonal wells, as in fluxional XeF₆).

Off-Center d⁰ Ti⁴⁺ Displacements in Perovskite Ferroelectrics (BaTiO₃)

One of the most technologically important applications of the pseudo Jahn–Teller theorem is the microscopic theory of displacive ferroelectricity in d⁰ perovskite oxides such as barium titanate (BaTiO₃), lead titanate (PbTiO₃), and lithium niobate (LiNbO₃). Above its Curie temperature (T_C = 393 K), BaTiO₃ adopts a centrosymmetric cubic perovskite structure (space group Pm3̄m) in which each Ti⁴⁺ ion (3d⁰, closed-shell ¹A_1g ground state) sits at the exact center of symmetry of a regular [TiO₆]⁸⁻ octahedron, yielding zero net electric dipole moment.

Despite having zero d electrons—and therefore zero first-order Jahn–Teller effect in Crystal Field Theory—the cubic [TiO₆]⁸⁻ unit undergoes a spontaneous (A_1g + T_1u) ⊗ t_1u (and T_1g + T_2g) pseudo Jahn–Teller instability:

• The top of the valence band (HOMO) consists of non-bonding and weakly π-bonding O 2p orbitals (transforming as t_1u, t_1g, and t_2u in O_h symmetry), while the bottom of the conduction band (LUMO, separated by Δ ≈ 3.2 eV) consists of empty Ti 3d (t_2g) orbitals.
• When the Ti⁴⁺ cation displaces off-center along an odd-parity polar mode Q_t1u toward one corner O²⁻ ion (tetragonal phase), two edge O²⁻ ions (orthorhombic phase), or three face O²⁻ ions (rhombohedral phase below 183 K), symmetry breaks, allowing the filled O 2pπ orbitals to mix covalently with the empty Ti 3d (t_2g) orbitals to form a short, strong Ti≡O multiple bond (~1.86 Å vs. ~2.17 Å trans).
• This eight-minimum PJT adiabatic surface creates an intrinsic local dipole on every TiO₆ octahedron even in the high-temperature paraelectric cubic phase (confirmed by X-ray absorption fine structure, EXAFS), and cooperative dipole–dipole interactions across the crystal lattice lock these displacements into sequential rhombohedral → orthorhombic → tetragonal → cubic phase transitions upon heating. Conversely, as soon as d electrons are added into the t_2g shell (e.g., d¹–d³ perovskites), Pauli exclusion blocks O 2p → M 3d(t_2g) PJT charge transfer, explaining why classical ferroelectricity in perovskites is overwhelmingly restricted to d⁰ transition metals (Ti⁴⁺, Zr⁴⁺, Nb⁵⁺, Ta⁵⁺, W⁶⁺) or ns² lone-pair cations (Pb²⁺, Bi³⁺).

Heme Iron Out-of-Plane Doming in Deoxyhemoglobin and Myoglobin

In biochemistry and respiratory physiology, the cooperative oxygen-binding switch of hemoglobin between the tense (T, deoxy) and relaxed (R, oxy) quaternary states relies on a pseudo Jahn–Teller displacement of the iron atom in the porphyrin macrocycle. In oxygenated oxyhemoglobin, low-spin Fe²⁺ (d⁶, S = 0) sits flush inside the plane of the four pyrrole nitrogens (Q_z ≈ 0.0 Å). Upon O₂ release (or photodissociation), the five-coordinate iron transitions to high-spin Fe²⁺ (d⁶, S = 2) and rapidly domes ~0.40–0.55 Å out of the porphyrin plane toward the proximal histidine (His F8), pulling the F-helix and triggering the R → T allosteric transition. While steric repulsion of the populated σ-antibonding d_x²−y² orbital contributes, Bersuker demonstrated that out-of-plane doming along the A_2u normal coordinate is driven by PJT vibronic coupling between the iron 3d_z²/d_π manifold and low-lying porphyrin π/π* states (A_1g ⊗ A_2u), providing the ultrafast femtosecond mechanical force that initiates allosteric signaling.

What is the difference between the Jahn–Teller effect and the pseudo Jahn–Teller effect?

The classical first-order Jahn–Teller effect applies strictly to non-linear molecules with an exact orbital degeneracy in their reference electronic state (energy gap Δ = 0 at Q = 0, such as the doubly degenerate ²E_g state of octahedral Cu²⁺). Because the linear splitting term −F|Q| always beats the quadratic elastic restoring energy ½K₀Q² for infinitesimal displacements, a degenerate state is unconditionally unstable (no threshold required). By contrast, the pseudo Jahn–Teller (PJT) effect applies to non-degenerate states separated by a nonzero energy gap Δ > 0. Because mixing requires overcoming the energy denominator Δ, a PJT distortion occurs only if the threshold condition 2F² / Δ > K₀ is satisfied; below this threshold, the high-symmetry geometry remains stable, though its vibrational mode is softened.

Why are d⁰ transition-metal cations like Ti⁴⁺, V⁵⁺, and Mo⁶⁺ so prone to off-center distortions?

Even though d⁰ ions (Ti⁴⁺, V⁵⁺, Nb⁵⁺, Mo⁶⁺, W⁶⁺) have completely empty valence d shells and zero first-order Crystal Field Stabilization Energy, their empty t_2g (d_xy, d_xz, d_yz) orbitals lie very close in energy to the filled 2pπ lone-pair orbitals of coordinated oxide (O²⁻) or nitride (N³⁻) ligands. This small HOMO–LUMO energy gap Δ (typically 2.5–4.0 eV) and strong spatial overlap between O 2p and metal nd orbitals produce a large off-diagonal vibronic coupling constant F along the polar t_1u displacement mode, satisfying 2F² / Δ > K₀ and driving off-center metal–oxo multiple bonding in ferroelectric perovskites (BaTiO₃), polyoxometalates, and vanadate/molybdate enzyme active sites.

How can spectroscopy detect a weak pseudo Jahn–Teller effect when the molecule does not permanently distort?

When the vibronic coupling strength F is substantial but remains below the instability threshold (2F² / Δ < K₀), the molecule retains its high-symmetry equilibrium geometry (Q = 0), yet the PJT interaction leaves two unmistakable spectroscopic signatures: (1) Anomalous vibrational frequency softening and positive anharmonicity in the ground electronic state (the coupling mode Q exhibits an unusually low IR/Raman fundamental frequency and a flat-bottomed, quartic potential well), accompanied by dramatic frequency hardening of the exact same mode Q in the coupled upper electronic state |Ψ₂⟩; and (2) Herzberg–Teller intensity borrowing, whereby the symmetry-lowering mode Q mixes the allowed electronic transition moment of |Ψ₂⟩ into a forbidden transition, producing strong vibronic sidebands.

1. Bersuker, I. B. (2013). Pseudo-Jahn–Teller effect—A two-state paradigm in formation, deformation, and transformation of molecular systems and solids. Chemical Reviews, 113(3), 1351–1390. https://doi.org/10.1021/cr300279n
2. Pearson, R. G. (1969). Symmetry rule for predicting molecular structures. Journal of the American Chemical Society, 91(18), 4947–4955. https://doi.org/10.1021/ja01046a001
3. Walsh, A., Payne, D. J., Egdell, R. G., & Watson, G. W. (2011). Stereochemistry of post-transition metal oxides: revision of the classical lone pair model. Chemical Society Reviews, 40(9), 4455–4463. https://doi.org/10.1039/C1CS15098G
4. Bersuker, I. B. (2006). The Jahn–Teller Effect. Cambridge University Press. https://doi.org/10.1017/CBO9780511524769