Using Artificial Neural Networks to boost high-throughput discovery in heterogeneous catalysis

The work presents for the first time a detailed methodology which enable to scrutinize and identify solids that are relevant to be tested in a high throughput program. In the present case study, Artificial Neural Networks (ANN) are used to predict performances of catalysts for the Water Gas Shift reaction. In contrast to previous studies, it is shown that the quantitative prediction by ANN of performances is not adapted to a primary screening stage. On the contrary, ANN used as classifier tool within the course of an Evolutionary Strategy are well performing and well suited for high throughput heterogeneous catalysis. The virtual screening enables to pre-select candidates to be screened experimentally with a very high rate of relevance at an early stage of a high throughput experimentation program, thus reducing significantly the number of trials.