Jahn–Teller Effect

The Jahn–Teller effect (also known as Jahn–Teller distortion) is a fundamental quantum-mechanical phenomenon of molecular and solid-state physics formulated in 1937 by the German-British physicist Hermann Arthur Jahn and the Hungarian-American physicist Edward Teller. Using group-theoretical symmetry arguments, Jahn and Teller proved that any non-linear molecule or coordination complex in a spatially degenerate electronic state is structurally unstable with respect to at least one asymmetric vibrational normal mode that lowers the molecular symmetry, removes the orbital degeneracy, and reduces the total electronic energy of the ground state.

While applicable to organic radical ions (such as the benzene radical cation C₆H₆^•+ or cyclobutadiene) and fullerene anions (C₆₀ⁿ⁻), the Jahn–Teller theorem finds its most prominent applications in transition metal coordination chemistry, mineralogy, and bioinorganic biochemistry. In octahedral (O_h) transition-metal complexes, asymmetric occupation of the strongly σ-antibonding e_g orbital set (d_z² and d_x²−y²)—most notably in high-spin d⁴ (Cr²⁺, Mn³⁺), low-spin d⁷ (Co²⁺, Ni³⁺), and d⁹ (Cu²⁺) electronic configurations—drives a tetragonal distortion (O_h → D_4h) in which two axial metal–ligand bonds elongate while the four equatorial bonds contract.

Visualize Coordination Geometry: Explore octahedral and distorted coordination geometries in 3D inside our interactive VSEPR Lab simulation!

Linear Vibronic Coupling Along a Normal Coordinate

In the standard Born–Oppenheimer approximation, nuclear motion is separated from electronic motion because nuclei are thousands of times heavier than electrons. However, when two or more electronic wavefunctions are degenerate (or nearly degenerate) at a high-symmetry nuclear geometry (Q = 0), electronic and vibrational motions couple strongly—an interaction termed vibronic coupling.

Consider a high-symmetry octahedral complex with a doubly degenerate electronic state (E_g) that interacts with a symmetry-lowering normal vibrational coordinate Q (such as the tetragonal stretching mode Q_θ of e_g symmetry, which lengthens the two z-axis bonds while compressing the four x,y-plane bonds). Expanding the potential energy surface E(Q) to second order in Q yields two split adiabatic potential branches:

E_±(Q) = ½ K Q² ± F |Q|

where ½K Q² is the harmonic elastic restoring energy of the metal–ligand σ-framework (with force constant K > 0, which opposes distortion) and ±F|Q| is the linear Jahn–Teller electronic splitting term (with vibronic coupling constant F). Because the linear term −F|Q| scales with the first power of |Q| while the elastic restoring penalty ½K Q² scales with the second power, for small displacements around Q = 0 the linear electronic energy lowering always outweighs the quadratic elastic penalty.

Differentiating the lower branch E₋(Q) = ½K Q² − F|Q| with respect to Q and setting dE₋/dQ = 0 reveals that the symmetric geometry (Q = 0) is a local energy maximum (a conical intersection in two dimensions), and the new equilibrium minimum occurs at the distorted displacement Q₀:

Q₀ = ± F / K     and     E_JT = F² / (2K)

where E_JT is the Jahn–Teller stabilization energy (the depth of the energy trough below the undistorted O_h point). When both components of the degenerate e_g vibration (Q_θ and Q_ε) are plotted in polar coordinates, the two-sheeted potential energy surface forms the famous "Mexican hat" (sombrero) potential, featuring a central peak at Q = 0 surrounded by a circular trough of minimum energy at radius ρ₀ = F/K. Include higher-order anharmonic terms, and the circular trough warps into three equivalent energy wells corresponding to tetragonal elongation along the x-, y-, or z-axes.

Ligand Field Splitting Under Tetragonal Elongation (Oh → D4h)

In an undistorted octahedral (O_h) ligand field, the five degenerate d orbitals of a transition-metal ion split into two sets separated by the octahedral crystal-field splitting energy Δ_o:

• The t_2g triplet (d_xy, d_xz, d_yz): Directed between the Cartesian x-, y-, and z-axes where the six ligands reside, these orbitals are non-bonding (or weakly π-bonding/antibonding). Uneven electronic occupation of the t_2g set—occurring in d¹ (Ti³⁺), d² (V³⁺), high-spin d⁶ (Fe²⁺), and high-spin d⁷ (Co²⁺)—produces only a weak Jahn–Teller effect (bond length changes < 0.03 Å) because moving the ligands slightly toward or away from the metal barely alters the energy of orbitals pointing between the bond axes.
• The e_g doublet (d_z², d_x²−y²): Directed head-on along the Cartesian axes directly at the six donor lone pairs, these orbitals are strongly σ-antibonding. Uneven occupation of the e_g orbitals— specifically one electron (e_g¹) or three electrons (e_g³)—produces a strong Jahn–Teller effect (bond length changes of 0.20 to 0.45 Å and E_JT ≈ 10 to 40 kJ mol⁻¹).

Consider a copper(II) (d⁹) ion with configuration t_2g⁶e_g³ in an octahedral complex such as [Cu(H₂O)₆]²⁺. In O_h symmetry, the three e_g electrons can be arranged either as (d_z²)²(d_x²−y²)¹ or as (d_z²)¹(d_x²−y²)², a doubly degenerate ²E_g ground state. If the complex undergoes tetragonal elongation (z-out distortion, lowering symmetry from O_h to D_4h), the two axial ligands along the z-axis move farther from the Cu²⁺ nucleus while the four equatorial ligands in the xy-plane move closer:

1. Because electrostatic crystal-field repulsion and σ-antibonding overlap along the z-axis decrease, the d_z² orbital drops in energy by −½δ₁ (and within t_2g, the d_xz and d_yz orbitals drop by −⅓δ₂).
2. Simultaneously, because the four equatorial ligands move closer in the xy-plane, the d_x²−y² orbital rises in energy by +½δ₁ (and d_xy rises by +⅔δ₂), preserving the barycenter (center of gravity) of each orbital set.
3. Placing two electrons in the stabilized lower d_z² orbital and one electron in the destabilized upper d_x²−y² orbital yields a net electronic stabilization energy of 2(−½δ₁) + 1(+½δ₁) = −½δ₁ (whereas the fully filled t_2g⁶ shell contributes 4(−⅓δ₂) + 2(+⅔δ₂) = 0).

In crystalline copper(II) salts and aqueous [Cu(H₂O)₆]²⁺, this tetragonal elongation produces four short equatorial Cu–O bonds at 1.96 Å and two long, labile axial Cu–O bonds at 2.38 Å. In the extreme limit where the axial ligands are pushed out to infinity (infinite tetragonal elongation), the d_z² orbital drops so far in energy that even a d⁸ metal ion (such as Ni²⁺ with strong-field CN⁻ ligands, Pd²⁺, Pt²⁺, and Au³⁺) pairs both of its upper electrons in d_z², leaving d_x²−y² completely empty and adopting a diamagnetic square-planar (D_4h) geometry.

Fluxional Pseudorotation vs. Lattice Trapping

Because an octahedral complex has three equivalent Cartesian axes (x, y, and z) along which tetragonal elongation can occur, the Mexican hat potential energy surface contains three identical energy wells separated by small saddle barriers (typically ΔE_barrier ≈ 1 to 5 kJ mol⁻¹). Whether an experimental technique observes a distorted or symmetric molecule depends on the thermal energy (k_BT) and crystal environment:

• Dynamic Jahn–Teller Effect: In solution or at elevated temperatures where k_BT exceeds the barrier between the three wells (or via quantum tunneling at cryogenic temperatures), the complex rapidly interconverts (pseudorotates) the elongated axis among the x-, y-, and z-directions on a picosecond timescale (~10⁻¹¹ s). Fast spectroscopic probes such as X-ray diffraction or Electronic-Spin-Echo Envelope Modulation see an instantaneously distorted geometry, whereas slower techniques such as room-temperature NMR or isotropic Electron Paramagnetic Resonance (EPR) observe a time-averaged octahedral signal.
• Static Jahn–Teller Effect: When the complex contains chemically inequivalent ligands (e.g., trans-[CuCl₂(H₂O)₄]), is cooled below the barrier height in a rigid crystal lattice, or experiences cooperative crystal packing forces (as in perovskite manganites LaMnO₃), one of the three wells is permanently stabilized below the other two, locking the molecule into a static tetragonal geometry.

Bioinorganic Catalysis: Plastocyanin and the Photosystem II Mn₄CaO₅ Cluster

In bioinorganic chemistry, enzymes exploit the Jahn–Teller effect to control electron-transfer reorganization energies and ligand-exchange rates:

1. The Entatic State in Blue Copper Proteins (Plastocyanin): In aqueous solution, reducing tetragonally distorted Cu²⁺ (d⁹) to Cu⁺ (d¹⁰, which prefers a tetrahedral geometry with no Jahn–Teller distortion) requires a large inner-sphere reorganization energy (λ_in ≈ 1.2–1.5 eV) to rearrange the bond lengths and angles, resulting in slow self-exchange electron transfer. In photosynthetic plastocyanin and bacterial azurin, the rigid protein fold constrains the copper center (coordinated by His37, His87, Cys84 at ~2.1 Å, and a distant axial Met92 at 2.82 Å) into a flattened trigonal-pyramidal geometry intermediate between Cu²⁺ and Cu⁺—an energized "entatic" (rack-induced) state that suppresses the Jahn–Teller rearrangement upon reduction, lowering λ to ~0.7 eV and accelerating electron transfer by 10⁴-fold.
2. Photosystem II Water Oxidation (Mn³⁺ in the Oxygen-Evolving Complex): Within the Mn₄CaO₅ cluster of photosynthesis, oxidation of Mn²⁺ (d⁵, symmetric) to Mn³⁺ (high-spin d⁴, t_2g³e_g¹) during the Kok cycle induces a strong Jahn–Teller elongation axis on the dangling Mn4 and cubane Mn1 ions. This elongated Jahn–Teller axis weakens the axial Mn–OH₂ bond, creating an open coordination site and promoting rapid substrate water binding and deprotonation prior to O–O bond formation in the S₃ → S₄ → S₀ transition.

Related concepts include Crystal Field Theory and Molecular Geometry.

Why cannot linear molecules undergo a Jahn–Teller distortion?

Jahn and Teller proved explicitly in their 1937 paper that linear molecules (belonging to point groups C_∞v or D_∞h, such as CO₂, HCN, or •NO) are exempt from the first-order Jahn–Teller theorem. In a linear molecule, spatially degenerate electronic states have nonzero orbital angular momentum along the internuclear axis (such as Π states with Λ = ±1 or Δ states with Λ = ±2), whereas the asymmetric bending vibration has Π symmetry (angular momentum l = ±1). Group-theoretical evaluation of the linear vibronic matrix element ⟨Ψ_Λ | (∂H/∂Q_Π) | Ψ_−Λ⟩ shows that the integral vanishes identically because the change in electronic angular momentum (ΔΛ = ±2 for a Π state) cannot be matched by a single quantum of the bending vibration (Δl = ±1). Instead, linear degenerate molecules exhibit the Renner–Teller effect, a weaker quadratic (Q²) vibronic coupling that alters the bending force constant (and splits the potential into two curves with different curvatures) without shifting the energy minimum away from Q = 0 unless the quadratic coupling exceeds the harmonic restoring force.

Why is tetragonal elongation (z-out) much more common than tetragonal compression (z-in) in d⁹ Cu²⁺ complexes?

At the simple linear harmonic level of theory, tetragonal elongation (two long axial bonds, four short equatorial bonds; hole in d_x²−y²) and tetragonal compression (two short axial bonds, four long equatorial bonds; hole in d_z²) yield the same first-order stabilization energy E_JT. In real complexes, two higher-order effects overwhelmingly favor tetragonal elongation: (1) Anharmonic bond-repulsion asymmetry, because pushing four equatorial ligands closer to the metal core incurs far greater Born–Mayer core–core Pauli repulsion than pushing only two axial ligands closer; and (2) 4s–3d_z² orbital mixing: in D_4h symmetry, both the 3d_z² orbital and the empty valence 4s orbital belong to the totally symmetric a_1g irreducible representation, whereas 3d_x²−y² belongs to b_1g. Mixing the higher-lying 4s orbital into 3d_z² lowers the energy of 3d_z² even further, stabilizing the (d_z²)²(d_x²−y²)¹ elongated ground state.

How does the Jahn–Teller effect explain the rapid water-exchange rate of [Cu(H₂O)₆]²⁺?

Across the first-row divalent transition-metal aqua ions [M(H₂O)₆]²⁺, the rate constant for exchanging an inner-sphere water molecule with bulk solvent generally decreases from left to right as the ionic radius shrinks and Ligand Field Activation Energy increases: V²⁺ (d³, k_ex ≈ 87 s⁻¹) and Ni²⁺ (d⁸, k_ex ≈ 3.2 × 10⁴ s⁻¹) exchange relatively slowly. In stark contrast, [Cu(H₂O)₆]²⁺ (d⁹) exchanges water almost at the diffusion limit with k_ex ≈ 4.4 × 10⁹ s⁻¹—over 10⁵ times faster than Ni²⁺. Because [Cu(H₂O)₆]²⁺ has two elongated, weakly bound axial water ligands at 2.38 Å that rapidly pseudorotate with the four 1.96 Å equatorial ligands on a ~5 × 10⁻¹² s timescale, bulk water molecules easily substitute at the labile axial position, and pseudorotation immediately sweeps the newly bound ligand into an equatorial position.

1. Jahn, H. A., & Teller, E. (1937). Stability of polyatomic molecules in degenerate electronic states. I—Orbital degeneracy. Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, 161(905), 220–235. https://doi.org/10.1098/rspa.1937.0142
2. Bersuker, I. B. (2006). The Jahn–Teller Effect. Cambridge University Press. https://doi.org/10.1017/CBO9780511524769
3. Solomon, E. I., Szilagyi, R. K., DeBeer George, S., & Basumallick, L. (2004). Electronic structures of metal sites in proteins and models: Contributions to function in blue copper proteins. Chemical Reviews, 104(2), 419–458. https://doi.org/10.1021/cr0206317
4. Halcrow, M. A. (2013). Jahn–Teller distortions in transition metal compounds, and their importance in functional molecular and inorganic materials. Chemical Society Reviews, 42(4), 1784–1795. https://doi.org/10.1039/C2CS35253B