Quantum-Chemistry Study of the Hydrolysis Reaction Profile in Borate Networks: A Benchmark

This investigation involved an ab initio and Density Functional Theory (DFT) analysis of the hydrolysis mechanism and energetics in a borate network. The focus was on understanding how water molecules interact with and disrupt the borate network, an area where the experimental data are scarce and unreliable. The modeled system consisted of two boron atoms, bridging oxygen atoms, and varying numbers of water molecules. This setup allows for an exploration of hydrolysis under different environmental conditions, including the presence of OH- or H+ ions to simulate basic or acidic environments, respectively. Our investigation utilized both ab initio calculations at the MP2 and CCSD(T) levels and DFT with a range of exchange-correlation functionals. The findings indicate that the borate network is significantly more susceptible to hydrolysis in a basic environment, with respect to an acidic or to a neutral pH setting. The inclusion of explicit water molecules in the calculations can significantly affect the results, depending on the nature of the transition state. In fact, some transition states exhibited closed-ring configurations involving water and the boron-oxygen-boron network; in these cases, there were indeed more water molecules corresponding to lower energy barriers for the reaction, suggesting a crucial role of water in stabilizing the transition states. This study provides valuable insights into the hydrolysis process of borate networks, offering a detailed comparison between different computational approaches. The results demonstrate that the functionals B3LYP, PBE0, and wB97Xd closely approximated the reference MP2 and CCSD(T) calculated reaction pathways, both qualitatively in terms of the mechanism, and quantitatively in terms of the differences in the reaction barriers within the 0.1-0.2 eV interval for the most plausible reaction pathways. In addition, CAM-B3LYP also yielded acceptable results in all cases except for the most complicated pathway. These findings are useful for guiding further computational studies, including those employing machine learning approaches, and experimental investigations requiring accurate reference data for hydrolysis reactions in borate networks.