Developing highly efficient and enantioselective reactions and understanding their mechanism are highly attractive.As the solvent molecules and counterions interact with the central metal in which the chiral environme...Developing highly efficient and enantioselective reactions and understanding their mechanism are highly attractive.As the solvent molecules and counterions interact with the central metal in which the chiral environment of the active site is very complicated during the reaction,the origin of many organometallic catalytic reactions such as Cu(II)/TOX[TOX:trisoxazoline]catalyzed Friedel–Crafts reaction remains elusive despite years of investigation both by experiments and theoretical calculations.Here,to the best of our knowledge,we employed a machine learning(ML)-based global potential energy exploration approach for the first time and combined it with control experiments to resolve the origin of asymmetric induction.By exploring thousands of likely transition state(TS)configurations,we revised the previous coordination mode of Cu(II)catalyst proposed in the Friedel–Crafts alkylation,where both the side-arm and the trifluoromethanesulfonate anion(OTf^(−))coordinated weakly with Cu(II)at the enantio-determining step.The dynamic coordination of the side-arm and the OTf−introduced a rich variation of the coordination environment in different solvents,leading to the variation in ee values.With the designed control experiments,we further confirmed the theoretical model,and thus,proved the immense potential of ML-atomic simulations in unraveling the catalytic intrinsic essence of structurally complex systems in organic chemistry.展开更多
基金supported by the National Key Research and Development Program of China(2018YFA0208600)the National Science Foundation of China(21773032,21972023,21533001,22022301,91545107,91745201)。
基金the National Science Foundation of China(grant nos.12188101,22033003,91745201,and 91945301)the Fundamental Research Funds for the Central Universities,China(grant no.20720220011)+3 种基金the National Key Research and Development Program of China(grant no.2018YFA0208600)Science and Technology Commission of Shanghai Municipality,China(grant no.23JC1404500)the Strategic Priority Research Program of the Chinese Academy of Sciences(grant no.XDB0610000)the Tencent Foundation for XPLORER PRIZE,Beijing,China.
文摘Developing highly efficient and enantioselective reactions and understanding their mechanism are highly attractive.As the solvent molecules and counterions interact with the central metal in which the chiral environment of the active site is very complicated during the reaction,the origin of many organometallic catalytic reactions such as Cu(II)/TOX[TOX:trisoxazoline]catalyzed Friedel–Crafts reaction remains elusive despite years of investigation both by experiments and theoretical calculations.Here,to the best of our knowledge,we employed a machine learning(ML)-based global potential energy exploration approach for the first time and combined it with control experiments to resolve the origin of asymmetric induction.By exploring thousands of likely transition state(TS)configurations,we revised the previous coordination mode of Cu(II)catalyst proposed in the Friedel–Crafts alkylation,where both the side-arm and the trifluoromethanesulfonate anion(OTf^(−))coordinated weakly with Cu(II)at the enantio-determining step.The dynamic coordination of the side-arm and the OTf−introduced a rich variation of the coordination environment in different solvents,leading to the variation in ee values.With the designed control experiments,we further confirmed the theoretical model,and thus,proved the immense potential of ML-atomic simulations in unraveling the catalytic intrinsic essence of structurally complex systems in organic chemistry.
