Designing Materials for Catalysis via Systematic Experiments and Artificial Intelligence

The identification of key physicochemical parameters that correlate with the performance can accelerate the development of improved materials and unveil the relevant underlying physical processes. However, the analysis of correlations is often hindered by inconsistent data. Besides, nontrivial, yet unknown relationships may be important, and the intricacy of the various processes may be significant. Here, we discuss how these challenges can be tackled in heterogeneous catalysis via a combination of rigorous experiments and artificial intelligence. In particular, we apply the symbolic-regression sure-independence-screening-and-sparsifying-operator (SISSO) approach to identify an analytical expression describing the performance of perovskite oxides in the catalytic CO oxidation.[1] We assess the performance of the SISSO model for the design of new, more complex materials.

[1] G. Bellini et al., Angew. Chem. Int. Ed., DOI: 10.1002/anie.202417812.