Markov Chain Monte Carlo optimization of visible light-driven hydrogen production

This paper uses Markov Chain Monte Carlo techniques to draw out insight in optimizing the visible light-driven hydrogen production process of Tolod et. al. by specifying which reactions to promote in order to increase the rate of hydrogen gas production from a water-based solution. The objective is to minimize the time it takes to form a new hydrogen gas molecule through controlling the initial concentration of the reactants. An agent-based program has been created to model and simulate the reactor setup. Gibbs sampling has been used to determine a unique and stationary distribution suggesting that reactions producing carbon dioxide and hydronium, when increased, will yield faster hydrogen production for the reactor under study.