Targeted design of advanced electrocatalysts by machine learning

Exploring the production and application of clean energy has always been core sustainable development. As a technology, electrocatalysis receiving widespread attention. It is crucial to achieve efficient, stable cheap electrocatalysts. However, traditional “trial error” method time-consuming, laborious costly. In recent years, with significant increase in computing power, computations have played an important role electrocatalyst design. Nevertheless, it still difficult search for advanced electrocatalysts vast chemical space through density functional theory (DFT) computations. Fortunately, development machine learning interdisciplinary integration will inject new impetus into targeted design Machine able predict electrochemical performances accuracy close DFT. Here we provide overview design, including prediction structure, thermodynamic properties kinetic barriers. We also discuss potential explicit solvent model combined molecular dynamics this field. Finally, favorable circumstances challenges are outlined future electrocatalysis. The studies on processes by further realize high-efficiency