Dinitrogen activation under mild conditions is important but extremely challenging due to the inert nature of the N≡N triple bond evidenced by high bond dissociation energy(945 k J/mol) and large HOMOLUMO gap(10.8 e ...Dinitrogen activation under mild conditions is important but extremely challenging due to the inert nature of the N≡N triple bond evidenced by high bond dissociation energy(945 k J/mol) and large HOMOLUMO gap(10.8 e V). In comparison with largely developed transition metal systems, the reported main group species on dinitrogen activation are rare. Here, we carry out density functional theory calculations on methyleneboranes to understand the reaction mechanisms of their dinitrogen activation. It is found that the methyleneboranes without any substituent at the boron atom performs best on dinitrogen activation, which could be contributed to its small singlet-triplet gap. In addition, strong correlations are achieved on dinitrogen activation between the singlet-triplet energy gap and the reaction energies for the formation of the end-on products as well as the side-on ones. The principal interacting orbital analysis suggests that methyleneboranes can mimic transition metals to cleave the N≡N triple bond. Our findings could be helpful for experimental chemists aiming at dinitrogen activation by main group species.展开更多
This study explores the application of machine learning to predict the bond dissociation energies(BDEs)of metal-trifluoromethyl compounds.We constructed a dataset comprising 2219 metal-trifluoromethyl BDEs using densi...This study explores the application of machine learning to predict the bond dissociation energies(BDEs)of metal-trifluoromethyl compounds.We constructed a dataset comprising 2219 metal-trifluoromethyl BDEs using density functional theory(DFT).Through a comparative analysis of various machine learning algorithms and molecular fingerprints,we determined that the XGBoost algorithm,when combined with MACCS and Morgan fingerprints,exhibited superior performance.To further enhance predictive accuracy,we integrated chemical descriptors alongside multiple fingerprints,achieving an R^(2) value of 0.951 on the test set.The model demonstrated excellent generalization capabilities when applied to synthesized metal-trifluoromethyl compounds,highlighting the critical role of chemical descriptors in improving predictive performance.This research not only establishes a robust predictive model for metal-trifluoromethyl BDEs but also underscores the value of incorporating chemical insights into machine learning workflows to enhance the prediction of chemical properties.展开更多
基金Financial support by the National Science Foundation of China (No. 22073079)the Top-Notch Young Talents Program of China is gratefully acknowledged。
文摘Dinitrogen activation under mild conditions is important but extremely challenging due to the inert nature of the N≡N triple bond evidenced by high bond dissociation energy(945 k J/mol) and large HOMOLUMO gap(10.8 e V). In comparison with largely developed transition metal systems, the reported main group species on dinitrogen activation are rare. Here, we carry out density functional theory calculations on methyleneboranes to understand the reaction mechanisms of their dinitrogen activation. It is found that the methyleneboranes without any substituent at the boron atom performs best on dinitrogen activation, which could be contributed to its small singlet-triplet gap. In addition, strong correlations are achieved on dinitrogen activation between the singlet-triplet energy gap and the reaction energies for the formation of the end-on products as well as the side-on ones. The principal interacting orbital analysis suggests that methyleneboranes can mimic transition metals to cleave the N≡N triple bond. Our findings could be helpful for experimental chemists aiming at dinitrogen activation by main group species.
基金supported by the National Natural Science Foundation of China(Nos.22122104,22193012,and 21933004)the CAS Project for Young Scientists in Basic Research(Grant No.YSBR-095)the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDB0590000).
文摘This study explores the application of machine learning to predict the bond dissociation energies(BDEs)of metal-trifluoromethyl compounds.We constructed a dataset comprising 2219 metal-trifluoromethyl BDEs using density functional theory(DFT).Through a comparative analysis of various machine learning algorithms and molecular fingerprints,we determined that the XGBoost algorithm,when combined with MACCS and Morgan fingerprints,exhibited superior performance.To further enhance predictive accuracy,we integrated chemical descriptors alongside multiple fingerprints,achieving an R^(2) value of 0.951 on the test set.The model demonstrated excellent generalization capabilities when applied to synthesized metal-trifluoromethyl compounds,highlighting the critical role of chemical descriptors in improving predictive performance.This research not only establishes a robust predictive model for metal-trifluoromethyl BDEs but also underscores the value of incorporating chemical insights into machine learning workflows to enhance the prediction of chemical properties.
