High-spin Co³⁺ in cobalt oxyhydroxide for efficient water oxidation

Electrochemical water splitting is a promising route for hydrogen production but is limited by sluggish oxygen evolution reaction (OER) kinetics involving multiple coupled proton–electron transfer steps. Cobalt oxyhydroxide (CoOOH) is an earth-abundant, active OER catalyst whose performance has been tuned in prior work via morphology control, cation doping, oxygen vacancies, and heterojunctions. In conventional CoOOH, Co³⁺ (3d⁶) typically adopts a low-spin t2g6 configuration, where electron transfer proceeds via face-to-face t2g orbitals. Simulations suggest introducing high-spin Co³⁺ could boost OER by enabling faster electron transfer through apex-to-apex eg orbitals. Although high-spin Co has been engineered in other Co-oxide systems, they often reconstruct under alkaline anodic conditions into CoOOH, the true OER phase, and there has been no experimental demonstration of the effect of high-spin Co³⁺ within CoOOH itself. This study aims to synthesize CoOOH hosting high-spin Co³⁺, elucidate its origin, confirm its presence experimentally, and evaluate its impact on electron transfer kinetics and OER activity and stability.

Prior studies on CoOOH-based OER catalysts primarily enhanced low-spin Co³⁺ systems via morphology engineering, metal ion doping, introduction of oxygen vacancies, and constructing heterojunctions. Theory indicates faster electron transfer via eg orbitals compared to t2g, and spin-state engineering to high-spin Co³⁺ could enhance activity. High-spin Co has been reported in other cobalt oxides (e.g., perovskites, spinels) but such materials reconstruct to CoOOH under OER conditions, leaving a gap in understanding the direct role of high-spin Co³⁺ within CoOOH. The present work addresses this gap by experimentally creating and characterizing high-spin Co³⁺ in CoOOH and correlating it with OER performance.

Synthesis: S-COOOH (target high-spin CoOOH) was produced via a sulfurization–electro-oxidation route. Metallic Co was first electrodeposited on pretreated carbon cloth (1×2 cm) in a two-electrode cell (electrolyte: 0.15 M CoSO4·7H2O + 0.6 M H3BO3; 10 mA cm⁻², 1 h). The Co-coated substrate was sulfurized at 400 °C for 1 h in N2 to form Co3S4, washed and dried, then electrochemically oxidized at −1.58 V vs Hg/HgO for 5 h in 1 M KOH (three-electrode setup) to yield S-COOOH. The reference R-COOOH was prepared by hydrothermal synthesis of β-Co(OH)2 (Co(NO3)2·6H2O + hexamethylenetetramine at 120 °C for 6 h), followed by electro-oxidation to CoOOH under the same conditions. Electrolyte purification: Fe impurities were removed from 1 M KOH by H2 pre-purging, precipitation with Co(NO3)2·6H2O to form Co(OH)2, contacting purified KOH with Co(OH)2, settling 24 h, centrifugation, and decanting; final pH 13.92. Characterization: Phase and structure by XRD and Raman. Composition and chemistry by XPS and ICP-OES. Morphology and EDS by SEM, TEM/STEM. Magnetic properties by SQUID (M–H, M–T) and EPR at X-band. XAS: Co K-edge (transmission, SSLS XAFCA), Co L-edge and O K-edge (SUV beamline, TEY mode); EXAFS fitting with Athena/Artemis (FEFF6; K-range 3–13 Å⁻¹; R-range 1–2.85 Å; K-weights 1,2,3). Electrochemistry: Three-electrode cell in Fe-purified 1 M KOH, carbon cloth-supported samples as working electrodes, Pt counter, Hg/HgO reference; potentials converted to RHE. LSV at 0.1 mV s⁻¹ with 90% iR correction (Ru from EIS at OCP). Tafel analysis from η–log(J). ECSA from double-layer capacitance (Cs = 0.040 mF cm⁻²; CV 0.60–0.70 V vs RHE; scan 0.01–0.05 mV s⁻¹). Pulse-voltammetry (PV): cycles between El = 1.40 V (300 s) and Eh steps (6 s) from 1.42–1.58 V in 20 mV increments; charge normalized to ECSA. Chronopotentiometry (CP) at 10 mA cm⁻² for 200 h. Theory: DFT (VASP, PBE-GGA, PAW, DFT+U with U−J=3.52 eV for Co³⁺, D3 dispersion). Cutoff 500 eV; convergence 1e−5 eV, 0.03 eV Å⁻¹. k-mesh: 14×14×14 for β-CoOOH primitive cell DOS; 4×8×1 for edge model. Electronic analysis with VASPKIT; visualization with VESTA.

- High-spin Co³⁺ successfully introduced into CoOOH (S-COOOH) by creating coordinatively unsaturated edge Co sites via sulfurization–electro-oxidation. - Magnetic evidence: S-COOOH shows ferromagnetic hysteresis at 300 K with remanence; R-COOOH is paramagnetic at 300 K and antiferromagnetic below ~10 K (Néel temperature). Effective magnetic moments from Curie–Weiss fits: R-COOOH μeff ≈ 0.09 μB; S-COOOH μeff ≈ 0.76 μB, corresponding to ~15% high-spin Co³⁺. EPR peak at g ≈ 2.15 observed for S-COOOH (unpaired electrons in high-spin Co³⁺), absent in R-COOOH; both show g ≈ 2.00 signals attributed to oxygen vacancies, with higher concentration in S-COOOH. - XAS/EXAFS: Co K-edge white line of S-COOOH is broader and lower intensity, indicating 4p non-degeneracy; higher pre-edge intensity (lower centrosymmetry); Co–O bond length increases from 1.903 Å (R) to 1.910 Å (S); Co–O coordination number decreases to 5.6 in S-COOOH versus 6 in R-COOOH, evidencing unsaturated coordination; O K-edge and Co L-edge indicate increased 3d band splitting/broadening. A small red-shift (~0.47 eV) in S-COOOH suggests slightly lower Co valence consistent with high-spin formation. - DFT: Edge Co atoms with reduced coordination adopt high-spin and ferromagnetism; bulk/six-coordinated Co remains low-spin and non-magnetic. PDOS shows broadened, asymmetric eg and t2g* bands and increased states from −2 to 1 eV around EF for S-COOOH, consistent with enhanced electron transport. - Electron-transfer kinetics: Pulse-voltammetry shows higher QECSA–potential slopes for S-COOOH across potentials, indicating faster electron transfer from catalyst to circuit. - OER performance (Fe-free 1 M KOH, 90% iR-corrected): Overpotential at 10 mA cm⁻² is 226 mV (S-COOOH, Ru = 0.75 Ω) vs 374 mV (R-COOOH, Ru = 0.77 Ω), an improvement of 148 mV; Tafel slopes: 28 mV dec⁻¹ (S) vs 77 mV dec⁻¹ (R). Enhanced intrinsic activity confirmed after normalizing by ECSA (S: 256.25 cm²; R: 176.50 cm²) and by loading mass. Superior activity persists at higher current densities. - Stability: S-COOOH maintains ~constant potential over 200 h CP at 10 mA cm⁻²; morphology and Co K-edge XAS/FT-EXAFS remain essentially unchanged before/after durability test, indicating high structural stability.

The study directly demonstrates that inducing high-spin Co³⁺ at coordinatively unsaturated edge sites in CoOOH increases the electronic density of states near the Fermi level and enables faster electron transfer via apex-to-apex eg orbitals compared to face-to-face t2g transfer in low-spin CoOOH. This electronic reconfiguration accelerates deprotonation and charge transfer steps in OER, translating into substantially lower overpotential and improved Tafel kinetics. Despite constituting a minority (~15%) of Co sites, the high-spin edge Co³⁺ centers disproportionately enhance overall catalytic performance, underscoring their superior intrinsic activity. The combination of magnetic, spectroscopic, electrochemical, and DFT evidence coherently links spin-state engineering to catalytic enhancement while also showing that the structural framework remains robust under prolonged OER operation.

This work establishes a strategy to realize high-spin Co³⁺ within CoOOH by engineering coordinatively unsaturated edge sites through a sulfurization–electro-oxidation route. The high-spin configuration broadens and increases DOS near EF and promotes faster electron transfer via eg orbitals, delivering markedly enhanced OER performance (η10 = 226 mV; Tafel 28 mV dec⁻¹) and excellent 200 h stability in alkaline media. The results highlight spin-state control as a key lever for boosting the intrinsic activity of CoOOH and suggest extending this knowledge-driven design to other oxide-based electrocatalysts where edge coordination and spin transitions can be tuned for optimal OER kinetics. Future efforts could increase the fraction of high-spin sites, control edge densities and geometries, and generalize the approach across compositions and supports.

The fraction of high-spin Co³⁺ sites is relatively small (~15% by μeff estimation), concentrated at edge locations with unsaturated coordination; quantitative control over their density and distribution remains limited. While Fe impurities were rigorously removed from the electrolyte and structural stability was verified post-200 h OER, operando identification of the exact active sites and their evolution beyond Co valence/spin proxies is not fully resolved. The correlation between small Co valence shifts and spin state is inferred but could benefit from additional operando spectroscopies and site-selective probes.