(6cz) Catalytic Processes: From Molecules to Complex Reaction Networks

Designing “optimal” processes in terms of yields, selectivity, reaction rates, economics, and environmental impact is an open challenge in the field of heterogeneous catalysis. This can involve multi-dimensional decision-making, including the choice of chemistry, catalyst, and the reactors.  My research aims to apply state-of-the-art computational tools and developing new ones for detailed modeling and design of heterogeneous catalytic processes. To this end, I propose to build and expand on my PhD experience at University of Minnesota under the co-advisorship of Profs. Daoutidis and Bhan on developing methods and software for elucidating complex reaction networks and my current research co-advised by Prof. Mavrikakis and Maravelias at University of Wisconsin focusing on using DFT and mathematical optimization for a first-principles-based approach to rigorously model and design catalytic systems.

In this poster, I will present a multi-pronged approach involving computational chemistry, microkinetic modeling, numerical optimization, cheminformatics, graph theory, and network analysis that allows for elucidating molecular events such as surface reactions and relating them to reactor-scale phenomena for systems spanning few tens of species and reactions to several thousands. Specifically, I will discuss: (a) identification of molecular structure-property relationships in competitive adsorption of organonitrogen compounds in gas oil hydrotreating using DFT and vdW-DF methods, (b) leveraging nonlinear programming methods for rational catalyst design, design of experiments, and parameter estimation, and (c) development of a new code, RING, for automated rule-based generation, mechanism analysis, and modeling of complex reaction networks.