The design and optimization of heterogeneous catalysts using computational methods

The computational design of catalytic materials is a high dimensional structure optimization problem that is limited by the bottleneck of expensive quantum computation tools. Current implementations of first principles computational models for catalyst design are data-hungry, problem-specific and confirmatory in nature. However, they can be made less data-dependent, more transferable and exploratory by developing both forward and inverse catalyst mapping tools that are either inexpensive correlations, like scaling relations, or regression models that are based on relevant descriptors analysis. This work reviews the current application and the possible landscape for future advancements of such tools for developing generalized schemes for catalyst design and optimization. Computational design of catalytic materials is a high dimensional structure optimization problem that is limited by the bottleneck of expensive quantum computation tools. An illustration of interaction of different factors involved in the design and optimization of a catalyst.