Machine learning assisted prediction of high entropy alloys catalysts for hydrogen evolution reaction

Abstract

Innovation in the field of catalysis is of utmost importance to reduce reaction time and reaction efficiency. This requires sophistication in existing techniques and investigation of new, suitable materials for efficient catalytic design. Alloys as heterogeneous catalysts can be employed in comparison to pure elements. The Hydrogen Evolution Reaction (HER), a key step in the electrolysis of water, gives Hydrogen which can be considered as a very clean renewable energy resource compared to conventional fossil fuels. HER needs economic catalysts because of expensive pure element platinum electrocatalysts. In this study, we’ll find out suitable High Entropy Alloys as the highly active catalysts for HER. To do so, we perform adsorption energy calculations for preparing training data from few selected surface microstructures using DFT with the PBE functional and GPAW. We wish to develop a simple model for the full distribution of adsorption energies involving random combinations of lattice positions using the prediction of a Machine Learning Algorithm. The optimization of the alloy then needs to be done so as to maximize catalytic activity using a suitable regression programming technique in the model algorithm. My aim is to engineer economically feasible catalysts for HER which in turn increases the prospects of Hydrogen as a clean renewable resource of energy.