Intracluster O–O Coupling Pathway Evidenced for an Anderson-Type Single-Cobalt Polymolybdate Water Oxidation Catalyst

Abstract

Toward the establishment of a carbon-neutral society, artificial photosynthesis, replacing fossil fuels with renewable energies (H2, HCOOH, CH3OH, etc.), has attracted great attention in recent years. One of the most important issues in this area has been to advance our knowledge and technical skills in handling water oxidation (WO) catalysis often considered as the bottleneck in the overall photosynthetic processes. The polyoxometalate water oxidation catalysts (WOCs) studied herein belong to one of the most highly active as well as the most highly robust family of WOCs. Nevertheless, due to their structural complexity and the high molecular weight associated with the high content of heavy elements, their mechanism of action remains largely unexplored from both theoretical and experimental viewpoints. To solve these problems, here we focus on one of the smallest and lightest polyoxometalate WOCs, single-cobalt polyoxometalate (NH4)3[CoMo6O24H6]·5H2O (Co-POM-Mo), which was previously demonstrated to be highly active for WO [Tanaka, S. , Chem. Commun. 2012, 48, 1653−1655]. The temperature dependence of the WO rate observed by mixing Co-POM-Mo and [RuIII(bpy)3]3+ using a stopped-flow technique revealed a positive entropy of activation (ΔS‡ = ca. 20–40 cal mol–1 K–1), revealing the promotion of intramolecular O–O coupling via the dissociative activation of the oxygens preinstalled in the cluster. The 17O NMR and 18O-labeling experiments, conducted to understand the source of oxygens in the O2 evolved, clarified the major contribution of the preinstalled 16O oxygens in the O–O coupling by Co-POM-Mo. The O2 evolved from the three catalysis cycles by Co-POM-Mo (100% in 16O) from an 18O-enriched aqueous solution (49% in 18O) resulted in the 16O2, 16O18O, and 18O2 abundances of 46, 40, and 14%, respectively. The Monte Carlo simulation reproduced the observed abundances under the assumption that the three consecutive O–O coupling proceed by only adopting a single μ3-OH site among the six available sites. Our simulation also supposed the O–O coupling process competes with the exchange of the catalytically active μ3-OH oxygen with the bulk water oxygen with the former twice higher in rate than the latter. Our DFT results also agree well with the promotion of O–O coupling between the inner μ3-OH oxygen bound to the central cobalt ion and the adjacent μ2-O oxygen constructing the Mo6(μ2-O)6 ring. This study strongly suggests that the O–O coupling among the preinstalled oxygens is the only available pathway to promote the WO by one of the highly active polyoxometalate WOCs, providing a new perspective on the catalysis by oxo clusters.