Spin Crossover

Spin crossover (SCO)—sometimes termed spin transition or spin equilibrium—is a reversible quantum-mechanical switching phenomenon in which a transition metal coordination complex transitions between a low-spin (LS) electronic ground state and a high-spin (HS) electronic state in response to external perturbations such as temperature, hydrostatic pressure, light irradiation, magnetic fields, or guest/substrate binding. First discovered in 1931 by Live Cambi and Luigi Szegö during magnetic susceptibility measurements of iron(III) tris(dithiocarbamate) complexes, spin crossover occurs predominantly in first-row octahedral (O_h) complexes with d⁴, d⁵, d⁶, and d⁷ electronic configurations positioned near the vertical spin-pairing boundary on a Tanabe–Sugano diagram.

In the language of Crystal Field Theory and Ligand Field Theory, an octahedral complex adopts a high-spin state when the ligand-field splitting energy Δ_o is smaller than the interelectronic spin-pairing energy P (Δ_o < P), and a low-spin state when Δ_o exceeds P (Δ_o > P). When intermediate-field ligands (most famously nitrogen-donor chelates coordinated to Fe²⁺ d⁶ or Co²⁺ d⁷) tune Δ_o so that the zero-point energy difference between the low-spin and high-spin potential wells is comparable to thermal energy k_BT (ΔE°_HL ≈ E_HS − E_LS ≈ 100 to 2,000 cm⁻¹), the complex becomes bistable. Switching from low spin to high spin transfers one or two electrons from the non-bonding t_2g orbitals into the strongly σ-antibonding e_g* orbitals, simultaneously changing the magnetic moment, optical color, metal–ligand bond lengths (by 0.08 to 0.22 Å), and Jahn–Teller activity of the molecule—making spin crossover central both to molecular spintronics and to the allosteric mechanics of hemoglobin oxygen transport.

Interactive VSEPR Lab: Examine 6-coordinate octahedral Fe(II)/Co(II) coordination spheres in 3D inside the VSEPR Lab!

Enthalpy (ΔH > 0) vs. Vibrational & Electronic Entropy (ΔS > 0)

In nearly all thermally driven spin-crossover complexes, the low-spin (LS) state is the true thermodynamic ground state at cryogenic temperatures (T → 0 K), whereas heating the sample above a characteristic transition temperature T_½ (where the molar fractions of high-spin and low-spin molecules are equal, γ_HS = γ_LS = 0.5) converts the ensemble to the high-spin (HS) state. This universal thermal directionality follows directly from the Gibbs free energy of the LS → HS equilibrium:

ΔG_HL = ΔH_HL − T ΔS_HL

At the transition temperature T_½, ΔG_HL = 0, so T_½ = ΔH_HL / ΔS_HL:

1. Enthalpy Favors the Low-Spin State (ΔH_HL > 0, typically +6 to +25 kJ mol⁻¹): In the low-spin state (e.g., Fe²⁺ d⁶, t_2g⁶e_gESCAST⁰, ¹A__1g), all six d electrons reside in the π-bonding/non-bonding t_2g shell and the σ-antibonding e_gESCAST orbitals are completely vacant, producing short, strong Fe–N bonds (~1.97 Å). Converting to the high-spin state (t_2g⁴e_gESCAST², ⁵T_2g) promotes two electrons into the σ-antibonding e_gESCAST orbitals, weakening the metal–ligand bonds, lengthening all six Fe–N distances by Δr__HL ≈ +0.18 to +0.22 Å (to ~2.18 Å), and costing net bond enthalpy.
2. Entropy Strongly Favors the High-Spin State (ΔS_HL = ΔS_el + ΔS_vib > 0, typically +40 to +85 J K⁻¹ mol⁻¹): The total entropy gain upon LS → HS conversion is the sum of two positive terms:

• Electronic Spin & Orbital Entropy (ΔS_el): Determined by the increase in spin and orbital degeneracy, ΔS_el = R ln[(2S_HS + 1)g_orb,HS / (2S_LS + 1)g_orb,LS]. For Fe²⁺ (d⁶, S_LS = 0 → S_HS = 2), the spin degeneracy jumps from 1 to 5, contributing R ln 5 = 13.4 J K⁻¹ mol⁻¹ (or up to R ln 15 = 22.5 J K⁻¹ mol⁻¹ including ⁵T_2g orbital degeneracy).
• Vibrational Entropy (ΔS_vib, ~70–80% of total ΔS_HL): Because populating the σ-antibonding e_g* orbitals lengthens the metal–ligand bonds by ~0.20 Å, the harmonic force constants k of the 15 intramolecular skeletal vibrational modes of the ML₆ core drop substantially (typical Fe–N stretching frequencies decrease from ~400–500 cm⁻¹ in LS down to ~220–280 cm⁻¹ in HS). Softer vibrational modes have much denser vibrational energy levels at room temperature, generating a huge vibrational entropy gain of ΔS_vib ≈ +35 to +65 J K⁻¹ mol⁻¹ that drives the complete thermal conversion to high spin above T_½.

Conversely, applying hydrostatic pressure (P) favors the low-spin state because the low-spin octahedron has a significantly smaller molecular volume (ΔV_HL = V_HS − V_LS ≈ +20 to +35 ų per molecule, or +12 to +22 cm³ mol⁻¹), shifting T_½ upward via the Clausius–Clapeyron relation dT_½/dP = ΔV_HL/ΔS_HL > 0.

Coupled Spin Crossover and the Jahn–Teller Effect in Co(II) d⁷ and Mn(III) d⁴

Spin crossover and the Jahn–Teller effect are intimately linked whenever one of the two spin states possesses an asymmetrically occupied e_g* shell:

• Cobalt(II) (d⁷, ²E_g ↔ ⁴T_1g): In octahedral Co²⁺ complexes (such as [Co(terpy)₂]²⁺ or vitamin B₁₂ cob(II)alamin models), the low-spin state (t_2g⁶e_g*¹, S = ½, ²E_g) has a single electron in the σ-antibonding e_gESCAST shell and undergoes a strong tetragonal Jahn–Teller elongation (four short equatorial Co–N bonds at ~1.92 Å and two long axial Co–N bonds at ~2.12–2.20 Å). Upon thermal conversion to the high-spin state (t_2g⁵e_gESCAST², S = 3/2, ⁴T_1g), the e_gESCAST shell becomes symmetrically half-filled (dz²¹dx²−y_²¹), quenching the strong e_g Jahn–Teller distortion so that the equatorial bonds expand while the axial bonds change much less (average Δr_HL ≈ +0.09 Å).
• Manganese(III) (d⁴, ³T_1g ↔ ⁵E_g): Conversely, in Mn³⁺ spin-crossover complexes, the low-spin state (t_2g⁴e_g*⁰, S = 1, ³T_1g) has an empty e_gESCAST shell and a nearly regular octahedral geometry, whereas switching to the high-spin state (t_2g³e_gESCAST¹, S = 2, ⁵E_g) turns on a powerful e_g¹ Jahn–Teller elongation along one axis.

Slichter–Drickamer Elastic Cooperativity and Thermal Hysteresis

In liquid solution, spin-crossover molecules interact weakly with their neighbors, so the high-spin molar fraction γ_HS(T) follows a smooth, gradual Boltzmann sigmoid spanning 100–200 K. In the crystalline solid state, however, the ~10% volume expansion of the coordination sphere (Δr_HL ≈ +0.20 Å in Fe²⁺) generates an elastic pressure wave that propagates through π–π stacking, hydrogen-bonding networks, or covalent polymeric bridges—an elastic feedback mechanism closely analogous to the cooperative Jahn–Teller effect.

Described thermodynamically by the Slichter–Drickamer mean-field model, adding an inter-site cooperative interaction parameter Γ (in J mol⁻¹) modifies the excess free energy of mixing:

ΔG(T, γ_HS) = γ_HS ΔH_HL + Γ γ_HS(1 − γ_HS) − T [ γ_HS ΔS_HL − R (γ_HS ln γ_HS + (1 − γ_HS) ln(1 − γ_HS)) ]

• When Γ < 2RT_½, the spin transition remains continuous, becoming progressively steeper as Γ approaches 2RT_½.
• When Γ > 2RT_½ (strong elastic cooperativity, as in the classic benchmark complex [Fe(phen)₂(NCS)₂] at T_½ = 176 K or 1D triazole polymers Fe(Htrz)₂(trz) above room temperature), the transition becomes abrupt and first-order, opening a thermal hysteresis loop (ΔT_hyst = T_½↑ − T_½↓ up to 30–70 K wide). Within the hysteresis window, the crystal exhibits molecular memory: depending on its thermal history, it remains locked in either the diamagnetic purple low-spin state or the paramagnetic white/yellow high-spin state at the exact same temperature.

Light-Induced Excited Spin-State Trapping (LIESST)

In 1984, Stefaan Decurtins, Philipp Gütlich, and coworkers discovered that a low-spin Fe²⁺ spin-crossover crystal held at cryogenic temperatures (T < 50 K)—such as Fe(ptz)₆₂ (ptz = 1-propyltetrazole)—can be quantitatively switched to a long-lived metastable high-spin state by irradiating it with green light (~514 nm), a phenomenon termed Light-Induced Excited Spin-State Trapping (LIESST):

1. Photoexcitation and Double Intersystem Crossing: Green light excites the spin-allowed low-spin d–d band ¹A_1g → ¹T_1g (or MLCT band). Within sub-picoseconds (< 200 fs), the excited complex undergoes ultrafast spin–orbit-mediated intersystem crossing through intermediate triplet states (³T_1g / ³T_2g) into the high-spin ⁵T_2g potential well.
2. Franck–Condon Vibronic Barrier Trapping: Because the high-spin ⁵T_2g minimum is displaced horizontally along the breathing coordinate by Δr_HL = +0.20 Å from the low-spin ¹A_1g minimum, relaxation back to the ¹A_1g ground state requires overcoming a significant thermal activation barrier across the potential surface intersection (or slow quantum-mechanical nuclear tunneling at T < 20 K). Below the critical LIESST relaxation temperature (T_LIESST ≈ 50–130 K), the photo-generated high-spin state remains trapped for days or weeks, and can be erased on demand (Reverse-LIESST) by irradiating the high-spin band with near-infrared light (~820 nm, ⁵T_2g → ⁵E_g).

Perutz's Stereochemical Trigger in Hemoglobin Oxygen Binding

Nature exploits Fe²⁺ (d⁶) spin crossover as the mechanical actuator driving cooperative oxygen binding in vertebrate hemoglobin (Hb):

• Deoxyhemoglobin (T-State, High-Spin S = 2): In deoxygenated hemoglobin, each Fe²⁺ ion is 5-coordinate in a square-pyramidal molecular geometry bound to four pyrrole nitrogens of protoporphyrin IX and the axial imidazole N_ε of the proximal histidine (His F8). Because this 5-coordinate ligand field is relatively weak, Fe²⁺ adopts a high-spin d⁶ (S = 2, t_2g⁴e_gESCAST²) state. Population of the σ-antibonding dx²−y_² orbital directed directly at the four porphyrin nitrogens expands the Fe²⁺ ionic radius (0.78 Å) and lengthens the Fe–N_porphyrin bonds to 2.06 Å, forcing the iron atom to dome ~0.40 to 0.55 Å out of the porphyrin plane toward the proximal His F8.
• Oxyhemoglobin (R-State, Low-Spin S = 0): When a strong-field π-acceptor O₂ molecule binds at the sixth axial coordination site on the distal side of the heme, the octahedral splitting Δ_o jumps above the spin-pairing threshold. The iron center undergoes an instantaneous high-spin (S = 2) → low-spin (S = 0, t_2g⁶e_g*⁰) spin crossover, vacating the σ-antibonding d_x²−y² and d_z² orbitals. As Max Perutz demonstrated in 1970, emptying e_g* shrinks the Fe²⁺ radius by 0.17 Å (to 0.61 Å, Fe–N_porphyrin = 1.98 Å), allowing the iron ion to snap squarely into the plane of the porphyrin ring. This 0.40 Å inward mechanical displacement pulls the attached proximal His F8 and the F-helix with it, rupturing inter-subunit salt bridges at the α₁β₂ interface and switching the tetramer from the low-affinity Tense (T) state to the 300-fold higher-affinity Relaxed (R) state.

Substrate-Triggered Spin Crossover in Cytochrome P450 Monooxygenases

A complementary Fe³⁺ (d⁵) spin-crossover gate prevents wasteful consumption of cellular NADPH and O₂ in cytochrome P450 heme-thiolate monooxygenases:

1. Resting Substrate-Free State (Low-Spin S = ½, E°′ = −300 mV): In the resting enzyme, the ferric heme Fe³⁺ (d⁵) is 6-coordinate, bound axially to an anionic cysteine thiolate (Cys) below and a hydrogen-bonded water molecule above. This 6-coordinate ligand field keeps Fe³⁺ in the low-spin (S = ½, t_2g⁵e_g*⁰) state with a very negative reduction potential (−300 mV vs. SHE) that cannot be reduced by cytochrome P450 reductase (NADPH/FAD/FMN at −260 mV).
2. Substrate-Bound State (High-Spin S = 5/2, E°′ = −170 mV): When a hydrophobic substrate (such as camphor in P450_cam) enters the active-site pocket, it displaces the axial water ligand, converting the iron center to a 5-coordinate square-pyramidal complex. The drop in axial ligand field triggers a low-spin (S = ½) → high-spin (S = 5/2, t_2g³e_g*²) spin crossover (shifting the Soret UV–Vis band from 417 nm to 391 nm), moves Fe³⁺ out of the heme plane toward the Cys thiolate, and raises the Fe^3+/2+ reduction potential by +130 mV (to −170 mV)—acting as a redox switch that permits electron transfer and O₂ activation only when substrate is present. Conversely, electron-transfer proteins like cytochrome c lock their iron center into a permanent low-spin entatic state to avoid this 0.20 Å spin-crossover reorganization barrier.

Why is Fe(II) (d⁶) the most prolific spin-crossover ion in coordination chemistry?

Octahedral Fe²⁺ (d⁶) undergoes the largest simultaneous change in electronic and structural properties upon spin crossover: switching from low-spin ¹A_1g (t_2g⁶e_g*⁰, S = 0) to high-spin ⁵T_2g (t_2g⁴e_g*², S = 2) transfers two full electrons directly from the non-bonding t_2g orbitals into the strongly σ-antibonding e_gESCAST orbitals. Consequently, all six metal–ligand bonds expand isotropically by ~0.20 Å (~10% of the bond length), producing a strong vibrational entropy driving force (ΔS__HL ≈ 50–80 J K⁻¹ mol⁻¹), strong elastic cooperativity in crystal lattices, a complete switch between diamagnetism (0 unpaired electrons) and paramagnetism (4 unpaired electrons, μ_eff ≈ 5.2 μ_B), and a sharp visual color change.

What is the difference between spin crossover and resonance (spin-state mesomerism)?

In chemical resonance (such as the Kekulé structures of benzene), the true ground state is a single stationary quantum-mechanical superposition at a single nuclear geometry, and the individual contributing structures do not exist as separate physical species. In spin crossover, the low-spin and high-spin states are two distinct adiabatic electronic states separated by a real potential energy barrier and possessing different equilibrium nuclear geometries (e.g., Fe–N = 1.97 Å vs. 2.18 Å) and different spin multiplicities (e.g., S = 0 vs. S = 2). Mössbauer spectroscopy (timescale ~10⁻⁷ s) and X-ray diffraction directly resolve the distinct low-spin and high-spin populations coexisting in thermal equilibrium near T_½.

How can chemists synthetically tune the spin-crossover temperature T_½ of an Fe(II) complex?

Because T_½ = ΔH_HL / ΔS_HL depends directly on the zero-point energy gap ΔE°_HL = Δ_o − P, chemists tune T_½ across a 400 K range using Ligand Field Theory and steric design: (1) Electronic tuning, where replacing pyridine rings with stronger π-acceptor pyrazine or 1,10-phenanthroline rings increases Δ_o and shifts T_½ to higher temperatures; (2) Intraligand steric crowding (such as placing methyl groups at the 2,9-positions of phenanthroline adjacent to the donor nitrogens), which prevents the Fe–N bonds from contracting to the 1.97 Å low-spin distance and stabilizes the high-spin state (lowering T_½); and (3) Crystal engineering with hydrogen-bonded counterions or solvates to modify the lattice elastic cooperativity Γ.

1. Gütlich, P., Hauser, A., & Spiering, H. (1994). Thermal and optical switching of iron(II) complexes. Angewandte Chemie International Edition in English, 33(20), 2024–2054. https://doi.org/10.1002/anie.199420241
2. Halcrow, M. A. (2011). Structure:function relationships in molecular spin-crossover complexes. Chemical Society Reviews, 40(7), 4119–4142. https://doi.org/10.1039/C1CS15046D
3. Perutz, M. F. (1970). Stereochemistry of cooperative effects in haemoglobin. Nature, 228(5273), 726–734. https://doi.org/10.1038/228726a0
4. Decurtins, S., Gütlich, P., Köhler, C. P., Spiering, H., & Hauser, A. (1984). Light-induced excited spin state trapping in a transition-metal complex: The hexa(1-propyltetrazole)iron(II) tetrafluoroborate spin-crossover system. Chemical Physics Letters, 105(1), 1–4. https://doi.org/10.1016/0009-2614(84)80403-080403-0)