Advances in the Application of Machine Learning in Electrocatalytic Carbon Dioxide Reduction
In the context of continuous growth in global energy demand,limited fossil fuel resources,and the increasingly serious impacts of carbon dio-xide emissions on the climate,urgent reductions in carbon dioxide emissions are required.The electrochemical reduction of carbon dioxide(CO2 RR)based on clean electricity is an ideal way to alleviate fossil fuel consumption and greenhouse gas emissions.Traditional catalyst re-search and development models mainly rely on experimental trial-and-error methods,which are time consuming and limited in their ability to meet the needs for efficient catalysts.The rapid development of data science technologies,such as machine learning,has brought paradigm changing opportunities for catalyst research and development.High-throughput computing combined with machine learning has become an important ap-proach in the design of electrocatalysts in recent years.Thus,this paper reviews contemporary research on high-throughput computing combined with machine learning to guide catalyst development,including the principles of catalyst design,simulation calculation strategies,and the con-struction of machine learning models.The combination of high-throughput computing and machine learning should provide a new method for effi-cient screening and development of CO2 RR catalysts,which should broad the application of artificial intelligence in catalyst screening design.
CO2 RRdensity functional theory(DFT)machine learningdescriptorcatalyst screening
万方数据
维普
