Abstract

A catalyst alters the course of a chemical reaction without being consumed. Catalysts are the fairy dust that drives the chemical and petroleum industries, and, through their ability to improve reaction selectivity, enable alternate reaction pathways, and degrade unwanted side products, catalysts are a key component of society’s energy and environmental future. Fundamental understanding of catalytic phenomena requires knowledge at the molecular level. We learn about reaction pathways, i.e. the precise dance of the molecules, from chemical kinetic analysis, at both steady and unsteady state. To understand how a catalyst enables those pathways, however, we have to look into the reactor at the reacting catalytic site or at least characterize the chemical nature of the catalyst and its interaction with reacting molecules. We do this with an ever-expanding array of spectroscopic and microscopy tools. With the advent of reliable quantum level computation we now also have the ability to validate detailed molecule models of catalytic events. Thus, we are on the verge of making catalysis a science, i.e. of understanding the chemistry well enough to predict what chemical sites can catalyze particular reactions. This talk will illustrate how catalytic understanding has been gained in example systems including the water gas shift reaction, selective oxidation of propylene to propylene oxide, and olefin polymerization. A potential path forward to true catalyst design using a model-based approach we call Discovery Informatics will also be described.