Item

Abstract

We briefly survey recent developments in surface catalysis modeling. The differentiated view on required level of accuracy established in the wake of multi-scale modeling approaches led to the emergence of high-throughput computational screening approaches. The large amounts of data created this way are now increasingly mined with machine learning techniques. We discuss status and challenges in this exciting mix of methodologies that describe catalytic systems from the electrons to the reactor. Next to the traditional focus on understanding and predicting catalytic activity, we argue that approaches to dynamical catalyst restructuring, to concomitant heat management, and to catalyst lifetime are important themes for the decade to come.