Unveiling the hidden reaction kinetic network of carbon in water with first-principles Markov state models

• DOI: 10.48550/arxiv.2401.07019
• Publication Date: 2024-01-01
• Authors: Chu Li; Yuan Yao; Ding Pan
• Keywords: Chemical Physics (physics.chem-ph); Statistical Mechanics (cond-mat.stat-mech); Physics - Chemical Physics; FOS: Physical sciences; Computational Physics (physics.comp-ph); Physics - Computational Physics; Condensed Matter - Statistical Mechanics

Dissolution of CO$_2$ in water followed by the subsequent hydrolysis reactions is great importance to global carbon cycle, and capture storage. Despite enormous previous studies, are still not fully understood at atomistic scale. Here, we combined ab initio molecular dynamics simulations with Markov state models elucidate reaction mechanisms kinetics supercritical both bulk nanoconfined states. The integration unsupervised learning first-principles data allows us identify complex coordinates pathways automatically instead a priori human speculation. Interestingly, our unbiased modelling found novel pathway dissolving CO$_2$(aq) under graphene nanoconfinement, involving pyrocarbonate anion (C$_2$O$_5^{2-}$(aq)) as an intermediate state. unexpected appearance pyrocarbonates related superionic behavior confined solutions. Our study highlights large oxocarbons aqueous reactions, implications for deep cycle sequestration CO$_2$.