Title: “Elucidating catalytic mechanisms using DFT, informatics, and data science”
More than 1 trillion USD of fuels, chemicals, and functional materials require a catalyst, often an inorganic solid (a heterogeneous catalyst) for their synthesis. Elucidating the mechanistic underpinnings of a catalytic system – i.e. where and how reactions happen on a catalyst – can provide valuable insights on how to optimize them further, particularly for energy-intensive chemical processes. To this end, we aim to develop a rigorous understanding of the active sites and reaction mechanism of heterogeneous catalytic chemistries, employing a slew of computational tools ranging from electronic structure theories to informatics, data science, and mathematical optimization. In this talk, I will discuss two problems from my group’s research. First, I will consider the problem of CO2 hydrogenation in sour gas mixtures on a transition metal sulfide catalyst to show how a close combination of reaction kinetic experiments and density functional theory (DFT) calculations can allow for rigorous identification of active sites and reaction mechanism. Many catalytic chemistries are complex in that they comprise of thousands of intermediates and reactions. State-of-the-art computational approaches of developing purely DFT-based models are not scalable to such large systems. Borrowing concepts from cheminformatics, graph theory, data science, and optimization, our group has been developing methods to scalably identify plausible reaction mechanisms for such complex reaction systems. In the second part of the talk, I will discuss the application of these methods to an example from the domain of biomass conversion, viz. polyol decomposition.