Machine learning prediction of LDOS and adsorption energies of HER and OER multimetallic electrocatalysts
발표자
()
초록
내용
Multi-metallic alloy catalysts have gained attention as means of finding better alternatives to the currently known pure and bimetallic catalyst. As multi-metallic alloys possess a diverse set of surface sites, tuning the distribution of active adsorption and reaction site is key in changing the catalytic performance. However, combinatorial search of the multi-elemental space is expensive, with the different elemental selection and ratio of mixing being the cause. Herein, we propose a framework for efficient search of promising multi-metallic alloys using Graph Convolutional Neural Network (GCNN), targeting prediction of electrocatalytic activity related properties such as adsorption energy of key intermediates and the surface atoms' LDOS. We demonstrate the effectiveness of our method in prediction of homogeneous multimetallic catalyst system's activity towards Hydrogen evolution and Oxygen evolution reactions, two key reactions for green hydrogen production.