A Combined Theoretical and Experimental Study of Oxygen Vacancies in Co$_3$O$_4$ for Liquid-Phase Oxidation Catalysis

TL;DR

This study combines spin-polarized DFT+U, ab initio MD, and X-ray absorption spectroscopy to elucidate the behavior of oxygen vacancies in Co$_3$O$_4$ under liquid-phase ethylene glycol oxidation. A vacancy induces two high-spin Co$^{2+}$ centers in five-fold coordinated, distorted octahedral sites and narrows the band gap, with these states remaining stable at room temperature. Experimental O K-edge and Co K-edge XAS show vacancy-like features in fresh catalysts that shift toward an oxidized spinel reference after reaction, indicating vacancy healing driven by the reaction environment; magnetometry supports a reduction in Co$^{2+}$ content after catalysis. These findings imply a redox interplay in which ethylene glycol reduces the surface while O$_2$ reoxidizes it, and highlight the catalyst’s stability and reusability under alkaline, oxygen-rich liquid-phase conditions.

Abstract

In the present work, we investigate oxygen vacancies (V$_\mathrm{O}$) in Co$_3$O$_4$, both in the bulk phase and under liquid-phase ethylene glycol oxidation, by combining theoretical and experimental techniques. Density functional theory calculations for bulk Co$_3$O$_4$ show that introducing an oxygen vacancy reduces two adjacent Co$^{3+}$ ions to Co$^{2+}$ and narrows the band gap. The newly formed Co$^{2+}$ ions adopt high-spin configurations in distorted octahedral sites and remain stable in this state in ab initio molecular dynamics simulations at $300$ K. Computed O and Co K-edge X-ray absorption spectra for ideal and vacancy-containing Co$_3$O$_4$ show excellent agreement with the experimental data and serve as references to analyze the liquid-phase ethylene glycol oxidation. The comparison with experimental O K-edge spectra of fresh and post-reaction catalysts shows that fresh samples resemble the vacancy-containing reference, whereas post-reaction spectra shift toward the ideal reference. These results suggest that under liquid-phase ethylene glycol oxidation conditions, Co$_3$O$_4$ becomes more oxidized rather than reduced, by refilling preexisting oxygen vacancies. This is further supported by the observation that higher O$_2$ pressures increase the conversion and that the catalyst remains stable and active over several cycles.

Figures (7)

• Figure 1: Co$_3$O$_4$ spinel ($2\times 2\times 2$) supercell. Co$^{2+}$ cations (green) occupy tetrahedral sites and Co$^{3+}$ cations (purple) occupy octahedral sites. The right side of the figure illustrates the corresponding crystal-field splittings: the lower-right diagram shows tetrahedrally coordinated Co$^{2+}$ in a $d^7$ configuration, and the upper-right diagram shows octahedrally coordinated Co$^{3+}$ in a $d^6$ configuration in the low-spin state.
• Figure 2: Local structural and electronic response of Co$_3$O$_4$ to the creation of an oxygen vacancy. Top: atomic configuration around the O site before (left) and after (right) removing the central lattice O atom. In the ideal case, the O atom is coordinated to three low-spin Co$^{3+}$ (purple, $0~\mu_\mathrm{B}$) and one high-spin Co$^{2+}$ (green, $2.6~\mu_\mathrm{B}$). After introducing the O vacancy, two of the neighboring Co$^{3+}$ ions are reduced to Co$^{2+}$, as indicated by their change to green and the appearance of finite magnetic moments, demonstrating local charge and spin redistribution. Bottom: spin-resolved, orbital-projected density of states (s, p, d) summed over all atoms. Solid lines denote the spin-up channel and dashed lines of the same color denote the spin-down channel (only spin-up is listed in the legend). In ideal Co$_3$O$_4$ a clear band gap is present. After the vacancy is created, additional Co 3$d$ character appears closer to the Fermi level, spin splitting becomes visible, and the band gap is narrowed.
• Figure 3: Charge density difference $\Delta \rho = \rho_{\mathrm{V_O}} - \rho_{\mathrm{ideal}}$ for Co$_3$O$_4$ with a single oxygen vacancy. Cyan isosurfaces indicate negative charge depletion at the vacancy site and along the former Co–O bonds, while yellow isosurfaces show negative charge accumulation on neighboring Co cations and nearby O atoms, showing redistribution of the electrons left behind by the removed oxygen.
• Figure 4: Time evolution of the two Co ions next to the oxygen vacancy, followed over a $50~\mathrm{ps}$ab initio MD trajectory at $300~\mathrm{K}$. Panels (a) and (b) track all Co–O distances for Co 9 and Co 10. Each ion holds on to five short Co–O bonds in the range of $1.9$–$2.2~\text{\AA}$, while all remaining oxygen atoms are found beyond a distance of $3~\text{\AA}$. Neither site ever regains an octahedral environment; the five-fold coordination imposed by the vacancy remains intact. The coordination number plots in panels (c) and (d), obtained by counting O atoms within $2.3~\text{\AA}$ cutoff, show the same scenario: both Co$^{2+}$ centers fluctuate around a value of $5$, with only momentary drops to $4$ and never any increase to reach $6$. This configuration is therefore structurally stable at $300~\mathrm{K}$. Panels (e) and (f) follow the magnetic moments. Co 9 stays essentially locked in its high-spin state at roughly $2.5~\mu_\mathrm{B}$ for the entire simulation. Co 10 shows occasional, very short-lived excursions down to $1.6$–$2.0~\mu_\mathrm{B}$, but always returns to the same high-spin value, and never displays anything resembling a spin collapse. Overall, these results show that the oxygen vacancy produces two Co$^{2+}$ ions that remain both structurally and magnetically robust over the full $50~\mathrm{ps}$ trajectory at $300~\mathrm{K}$.
• Figure 5: Effect of O2 pressure and catalyst reusability for the hard-templated Co$_3$O$_4$ sample in the liquid-phase oxidation of ethylene glycol. The left part shows the EG conversion and the corresponding yields/selectivities of glycolic acid (GA), formic acid (FA), and oxalic acid (OA) after $6~\mathrm{h}$ at $120\,^{\circ}\mathrm{C}$, $0.325~\mathrm{M}$ EG, and $0.65~\mathrm{M}$ KOH for O2 pressures of 5, 10, and 15 bar. The right part shows EG conversion and GA/FA/OA yields/selectivities for three consecutive runs at 10 bar O2 under the same conditions, which illustrates the stability and reusability of the hard-templated sample.