DFTB-MD Simulations and Quantum Chemical Investigations of Reaction Mechanisms of Water-Reactive Molecules (SiH <b> ₂ </b> Cl <b> ₂ </b> , PCl <b> ₃ </b> , and SOCl <b> ₂ </b> )

Elucidating reaction mechanisms in complex solvation environments presents a significant challenge in computational chemistry. This study establishes a robust computational protocol that combines density functional tight-binding molecular dynamics (DFTB-MD) simulations with high-level quantum chemical calculations to investigate the reactions of water-reactive molecules (SiH₂Cl₂, PCl₃, and SOCl₂) under realistic solvation conditions. The protocol addresses limitations of conventional approaches by incorporating explicit solvation with numerous water molecules surrounding reactants, enabling the identification of frequently occurring reaction pathways through DFTB-MD simulations. Subsequently, density functional theory and domain-based local pair natural orbital coupled cluster singles and doubles with perturbative triples calculations provide quantitative evaluations of energetic values. Validation using SiH₄, a Cl-free analogue, demonstrates the protocol's ability to distinguish reactivity differences. The results reveal that Cl atoms bonded to central atoms (Si, P, and S) act as effective electron acceptors, facilitating electron transfer from H atoms of coordinated H₂O molecules and significantly enhancing the reactivity. For PCl₃, phosphorous acid (H₃PO₃) formation in both isomeric forms (P(OH)₃ and HPO(OH)₂) is observed, while for SOCl₂, sequential SO₂ formation followed by H₂SO₃ production is captured, demonstrating excellent agreement with experimental behavior. All reactions are spontaneous and strongly exothermic, producing hydrochloric acid. Rate constants calculated using transition-state theory and compared with diffusion-controlled limits or experimental data confirm that our solvation model accurately reflects bulk-liquid-phase conditions. The established computational protocol successfully reproduces experimental observations by accurately reflecting realistic reaction conditions, demonstrating its potential for broader application to complex chemical reactions in diverse solvent environments, with significant practical implications.