This study integrates first-principles calculations,computational chemistry,system simulations,experiments,and machine learning to identify redox perovskite oxides for solar thermochemical hydrogen production.Using tw...This study integrates first-principles calculations,computational chemistry,system simulations,experiments,and machine learning to identify redox perovskite oxides for solar thermochemical hydrogen production.Using two random forest regressions and one classification model,the approach predicts materials’stability and the enthalpy of oxygen vacancy formation(Δh_(o)),a critical property for selecting materials for thermochemical hydrogen production.B-site composition significantly influencesΔho predictions.The methodology led to the discovery of Ba_(0.875)Ca_(0.125)Zr_(0.875)Mn_(0.125)O_(3)(BCZM),which reduces at temperatures up to 250°C lower than CeO_(2)and is expected to outperform other perovskites in water splitting.However,CeO_(2)remains the benchmark for solar thermochemical hydrogen production.The combined use of machine learning and DFT calculations refined 4ho predictions and provided insights into experimental results.This framework not only enhances database creation for material screening but also establishes a novel approach for perovskite discovery for hydrogen production applications.展开更多
基金funding from ARENA as part of ARENA’s Research and Development Program–Renewable Hydrogen for ExportAC and RP gratefully acknowledge the financial support by Programa Atracción de Talento Fellowship of“Comunidad de Madrid”(2022-T1/AMB-23875)+2 种基金J.C.,A.B.,and A.C.thank Comunidad de Madrid”for the financial support to ACES4NET0-CM project(TEC-2024/ECO-116),through the R&D activities programme“Tecnologías 2024”.SD acknowledges support from the University of Newcastle Priority Research Centre for Frontier Energy Technologies and UtilizationM.V.G.P.thanks for the Grant PID2021-128915NB-I00 funded by MCIN/AEI/10.13039/501100011033 and by ERDF,UE.A.B.and L.M.gratefully acknowledge the financial support by Programa de Atracción de Talento Fellowship of“Comunidad de Madrid”(2020-T1/AMB-19884)A.B.also acknowledge the funding from Consolidacion Investigadora Fellowship(CNS2023-144887).We thank B.Baldassarri for helpful discussions.Computer time provided by the RES(Red Española de Supercomputación)resources at MareNostrum(BSC,Barcelona)node is acknowledged.
文摘This study integrates first-principles calculations,computational chemistry,system simulations,experiments,and machine learning to identify redox perovskite oxides for solar thermochemical hydrogen production.Using two random forest regressions and one classification model,the approach predicts materials’stability and the enthalpy of oxygen vacancy formation(Δh_(o)),a critical property for selecting materials for thermochemical hydrogen production.B-site composition significantly influencesΔho predictions.The methodology led to the discovery of Ba_(0.875)Ca_(0.125)Zr_(0.875)Mn_(0.125)O_(3)(BCZM),which reduces at temperatures up to 250°C lower than CeO_(2)and is expected to outperform other perovskites in water splitting.However,CeO_(2)remains the benchmark for solar thermochemical hydrogen production.The combined use of machine learning and DFT calculations refined 4ho predictions and provided insights into experimental results.This framework not only enhances database creation for material screening but also establishes a novel approach for perovskite discovery for hydrogen production applications.
