Data-Driven High-Throughput Screening of High-Performance Single-Atom Catalysts for Hydrogen Evolution and Hydrogen Sensing

Abstract

The exploration of high-performance catalytic materials has attracted significant attention due to their substantial economic value. However, the vast material search space and inherent limitations of conventional experimental trial-and-error methods pose significant challenges in exploring these catalytic materials. Herein, we propose a data-driven high-throughput approach for screening high-performance single-atom catalysts (SACs) suitable for hydrogen evolution reactions (HER) and hydrogen sensing applications. This methodology integrates density functional theory (DFT) calculations and a graph neural network (GNN)-based machine learning algorithm. Our results indicate that this data-driven approach effectively predicts SACs for HER and hydrogen sensing applications. This integrated framework significantly accelerates the discovery and development of high-performance catalytic materials, thereby advancing hydrogen-related technologies.