Bio-sketch:
Dr. S. Shambhawi is a Cambridge India Ramanujan Trust Scholar with a PhD in Chemical Engineering from University of Cambridge. She completed her BTech in Chemical Engineering from IIT Roorkee. She has experience working on theoretical catalyst design, optimization, and reaction kinetic modelling. Her publications are widely based on data analysis and data science applications for catalyst design. Shambhawi currently works as an AI Research Scientist for Chemical Data Intelligence in London. She is also a visiting scientist at the University of Cambridge where she leads group projects. Based on her research in Cambridge, she has co-founded a tech-based start-up called GreenCAT LLC. that provides catalyst solutions to industries.
Abstract:
Catalyst materials are known for facilitating reactions by lowering the activation barriers of reaction steps. Computational design of these catalytic materials is a high-dimensional structure optimization problem that is limited by expensive quantum mechanical computation tools, non-availability of standard catalyst database, limited catalyst search spaces, highly interdependent catalyst performance factors and expensive high-throughput investigation facilities. Current implementations of methods of catalyst design are very data-hungry, problem-specific, and confirmatory in nature. However, they can be made less data-dependent, transferable, and exploratory by developing forward and inverse catalyst-to-performance mapping tools that either employ inexpensive empirical/semi-empirical formalisms or are based on relevant descriptor analysis, or both. This thesis investigates the application of such tools and proposes a workflow for developing a generalized catalyst design scheme. Starting with a detailed analysis of the existing catalyst search tools and their implementations, it then introduces a forward and reverse catalyst optimization problem and demonstrates them using sample reaction systems. It compares the two optimization approaches and outlines a methodology for catalyst search and design. This includes a reverse catalyst optimization formulation and catalyst screening methods based on scaling relations. Lastly, it summarizes the proposed workflow and highlights possible future advancements that might be initiated via the current study.