A surge in global energy demand with limited fossil fuel resources and their severe environmental impacts necessitates the development of cost-effective catalysts to perform desired molecular transformations with utmost energy efficiency and minimal byproducts. However, the immense phase space of catalytic materials spanned by structural and compositional degrees of freedom precludes thorough screening, even with combinatorial quantum-chemical calculations and high-throughput experiments. With recent advances of fundamental catalysis, there has been increasing awareness of volcano-type constraints on attainable catalytic activity/selectivity due to linear scaling relationships among the energetics of key reaction species on a wide variety of materials, e.g., metals. Development of a theoretical framework that guides us to go beyond the scaling relations of adsorbate-surface interactions and strategically steer elementary reaction steps towards desired pathways still remains one of the greatest challenges faced by catalysis community today. In this Talk, we aim to tackle this challenge with explainable AI (XAI) that provides physical insights into governing factors of catalytic properties without sacrificing prediction performance of out-of-sample systems, thus accelerating the discovery of novel structural motifs beyond energy-scaling limitations with electrochemical nitrate reduction as the target reaction. The study aims to provide a theory-guided strategy to break the linear scaling relationships on intermetallic alloy catalysts by harnessing the adsorbate-specific interactions with metal d-states.

Explainable AI for catalysis beyond energy-scaling limitations

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.