Precise understading of the kinetics for catalytic activities has long been essential to the in-depth mechanistic study of catalysis and recently to set the basis for catalyst screening. However, commonly used methods suffer from low accuracy. This talk reports an enhanced approach to accurately identify the kinetic parameters from a kinetic dataset using the machine learning (ML) surrogate model via open source scikit-learn code package. The CO hydrogenation to methanol over metal (M)-doped Cu(111) model catalysts was taken as a case study. Our model captures not only the contribution from individual elementary step, but also the interaction between relevant steps within a reaction network, which was found to be essential to achieve the accurate predicibility. More importantly, going beyond the purely mathematic ML model, the catalytic role of each identified parameter can be revealed by using the model-agnostic interpretation tools, which enhances the mechanistic insight into the promoting effect of alloying. The trained ML model outperforms the conventional derivative-based methods in terms of both the predictability and mechanistic understanding.

Enhanced kinetic understanding of catalytic processes using data-driven framework

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.