Developing machine learning frameworks with predictive power,interpretability,and transferability is crucial,yet it faces challenges in the field of electrocatalysis.To achieve this,we employed rigorous feature engine...Developing machine learning frameworks with predictive power,interpretability,and transferability is crucial,yet it faces challenges in the field of electrocatalysis.To achieve this,we employed rigorous feature engineering to establish a finely tuned gradient boosting regressor(GBR)model,which adeptly captures the physical complexity from feature space to target variables.We demonstrated that environmental electron effects and atomic number significantly govern the success of the mapping process via global and local explanations.The finely tuned GBR model exhibits exceptional robustness in predicting CO adsorption energies(R_(ave)^(2)=0.937,RMSE=0.153 eV).Moreover,the model demonstrated remarkable transfer learning ability,showing excellent predictive power for OH,NO,and N_(2) adsorption.Importantly,the GBR model exhibits exceptional predictive capability across an extensive search space,thereby demonstrating profound adaptability and versatility.Our research framework significantly enhances the interpretability and transferability of machine learning in electrocatalysis,offering vital insights for further advancements.展开更多
Electrocatalytic nitrogen reduction to ammonia has garnered significant attention with the blooming of single-atom catalysts(SACs),showcasing their potential for sustainable and energy-efficient ammonia production.How...Electrocatalytic nitrogen reduction to ammonia has garnered significant attention with the blooming of single-atom catalysts(SACs),showcasing their potential for sustainable and energy-efficient ammonia production.However,cost-effectively designing and screening efficient electrocatalysts remains a challenge.In this study,we have successfully established interpretable machine learning(ML)models to evaluate the catalytic activity of SACs by directly and accurately predicting reaction Gibbs free energy.Our models were trained using non-density functional theory(DFT)calculated features from a dataset comprising 90 graphene-supported SACs.Our results underscore the superior prediction accuracy of the gradient boosting regression(GBR)model for bothΔg(N_(2)→NNH)andΔG(NH_(2)→NH_(3)),boasting coefficient of determination(R^(2))score of 0.972 and 0.984,along with root mean square error(RMSE)of 0.051 and 0.085 eV,respectively.Moreover,feature importance analysis elucidates that the high accuracy of GBR model stems from its adept capture of characteristics pertinent to the active center and coordination environment,unveilling the significance of elementary descriptors,with the colvalent radius playing a dominant role.Additionally,Shapley additive explanations(SHAP)analysis provides global and local interpretation of the working mechanism of the GBR model.Our analysis identifies that a pyrrole-type coordination(flag=0),d-orbitals with a moderate occupation(N_(d)=5),and a moderate difference in covalent radius(r_(TM-ave)near 140 pm)are conducive to achieving high activity.Furthermore,we extend the prediction of activity to more catalysts without additional DFT calculations,validating the reliability of our feature engineering,model training,and design strategy.These findings not only highlight new opportunity for accelerating catalyst design using non-DFT calculated features,but also shed light on the working mechanism of"black box"ML model.Moreover,the model provides valuable guidance for catalytic material design in multiple proton-electron coupling reactions,particularly in driving sustainable CO_(2),O_(2),and N_(2) conversion.展开更多
Lithium-sulfur(Li-S)batteries are regarded as one of the most promising next-generation energy storage systems due to their high theoretical specific energy density and low cost.However,serious shuttle effect and slug...Lithium-sulfur(Li-S)batteries are regarded as one of the most promising next-generation energy storage systems due to their high theoretical specific energy density and low cost.However,serious shuttle effect and sluggish lithium polysulfides(LiPSs)redox kinetics severely impede the practical application of Li-S batteries.Employing polar sulfur hosts is an effective strategy to alleviate the above problems.Herein,the potential of two-dimensional(2D)Ti_(2)B-based sulfur hosts for Li-S batteries was comprehensively explored using first-principles calculations.The results show that functional groups of Ti_(2)B can significantly modulate its structural properties,thus affecting its interaction with sulfurcontaining species.Among S,Se,F,Cl,and Br elements,Ti_(2)B terminated with S and Se atoms possess stronger adsorption capability towards soluble Li_(2)S_(8),Li_(2)S_(6),and Li_(2)S_(4),obviously stronger than organic electrolytes,which indicates that they can completely suppress the shuttle effect.Besides,Ti_(2)BS_(2)and Ti_(2)BSe_(2)can powerfully expedite the electrochemical conversion of LiPSs.Moreover,the decomposition energy barrier of Li_(2)S and diffusion energy barrier of single Li ion on them are also fairly low,manifesting their excellent catalytic performance towards the oxidation of Li_(2)S.Finally,Ti_(2)BS_(2)and Ti_(2)BSe_(2)always keep metallic conductivity during the whole charge/discharge process.Taking all this into account,Ti_(2)BS_(2)and Ti_(2)BSe_(2)are proposed as promising bifunctional sulfur hosts for Li-S batteries.Our results suggest that increasing the proportion of S and Se groups during the synthesis of Ti_(2)B monolayers is greatly helpful for obtaining high-performance Li-S batteries.Besides,our work not only reveals the huge potential of 2D transition metal borides in Li-S batteries,but also provides insightful guidance for the design and screening of new efficient sulfur cathodes.展开更多
Rechargeable metal-air batteries generally require efficient,durable,and safe bifunctional electrocatalysts to simultaneously support oxygen reduction/evolution reactions(ORR/OER).Herein,we employed first-principles c...Rechargeable metal-air batteries generally require efficient,durable,and safe bifunctional electrocatalysts to simultaneously support oxygen reduction/evolution reactions(ORR/OER).Herein,we employed first-principles calculations to explore the structure-activity relationship between the composition control of metal atoms and the catalytic activity of Pt-doped Ti_(2-x) Mn_(x) CO_(2) single-atom catalysts(SACs).The research found a clear linear relationship between the proportion of Mn and bifunctional performance,which can effectively modulate catalytic activity.Additionally,it shows excellent bifunctional catalytic activity at medium concentrations,among which the catalyst of Pt-V_(O)-Ti_(0.89) Mn_(1.11) CO_(2) displays the lowest overpo-tential(η^(ORR/OER)=0.26/0.28 V).Attributed to the modulation of the average magnetism of Mn and the d-band center of Pt by different com ponents,the bonding strength of the active center of Pt to adsorption intermediates is changed,resulting in the enhancement of the catalyst activity.Crucially,the molecular orbital-level bonding between the active site Pt and the adsorbed intermediate OH is clarified,shedding light on the involvement of the partially occupied antibonding state of Pt’s d orbital in the activation process.The research extensively explores the control of catalyst activity through composition,offering strong support for designing and optimizing highly active Janus MXene-supported SACs.展开更多
基金supported by the Research Grants Council of Hong Kong(CityU 11305919 and 11308620)and NSFC/RGC Joint Research Scheme N_CityU104/19Hong Kong Research Grant Council Collaborative Research Fund:C1002-21G and C1017-22Gsupported by the Hong Kong Research Grant Council Collaborative Research Fund:C6021-19E.
文摘Developing machine learning frameworks with predictive power,interpretability,and transferability is crucial,yet it faces challenges in the field of electrocatalysis.To achieve this,we employed rigorous feature engineering to establish a finely tuned gradient boosting regressor(GBR)model,which adeptly captures the physical complexity from feature space to target variables.We demonstrated that environmental electron effects and atomic number significantly govern the success of the mapping process via global and local explanations.The finely tuned GBR model exhibits exceptional robustness in predicting CO adsorption energies(R_(ave)^(2)=0.937,RMSE=0.153 eV).Moreover,the model demonstrated remarkable transfer learning ability,showing excellent predictive power for OH,NO,and N_(2) adsorption.Importantly,the GBR model exhibits exceptional predictive capability across an extensive search space,thereby demonstrating profound adaptability and versatility.Our research framework significantly enhances the interpretability and transferability of machine learning in electrocatalysis,offering vital insights for further advancements.
基金supported by the Research Grants Council of Hong Kong (City U 11305919 and 11308620)the NSFC/RGC Joint Research Scheme N_City U104/19The Hong Kong Research Grant Council Collaborative Research Fund:C1002-21G and C1017-22G。
文摘Electrocatalytic nitrogen reduction to ammonia has garnered significant attention with the blooming of single-atom catalysts(SACs),showcasing their potential for sustainable and energy-efficient ammonia production.However,cost-effectively designing and screening efficient electrocatalysts remains a challenge.In this study,we have successfully established interpretable machine learning(ML)models to evaluate the catalytic activity of SACs by directly and accurately predicting reaction Gibbs free energy.Our models were trained using non-density functional theory(DFT)calculated features from a dataset comprising 90 graphene-supported SACs.Our results underscore the superior prediction accuracy of the gradient boosting regression(GBR)model for bothΔg(N_(2)→NNH)andΔG(NH_(2)→NH_(3)),boasting coefficient of determination(R^(2))score of 0.972 and 0.984,along with root mean square error(RMSE)of 0.051 and 0.085 eV,respectively.Moreover,feature importance analysis elucidates that the high accuracy of GBR model stems from its adept capture of characteristics pertinent to the active center and coordination environment,unveilling the significance of elementary descriptors,with the colvalent radius playing a dominant role.Additionally,Shapley additive explanations(SHAP)analysis provides global and local interpretation of the working mechanism of the GBR model.Our analysis identifies that a pyrrole-type coordination(flag=0),d-orbitals with a moderate occupation(N_(d)=5),and a moderate difference in covalent radius(r_(TM-ave)near 140 pm)are conducive to achieving high activity.Furthermore,we extend the prediction of activity to more catalysts without additional DFT calculations,validating the reliability of our feature engineering,model training,and design strategy.These findings not only highlight new opportunity for accelerating catalyst design using non-DFT calculated features,but also shed light on the working mechanism of"black box"ML model.Moreover,the model provides valuable guidance for catalytic material design in multiple proton-electron coupling reactions,particularly in driving sustainable CO_(2),O_(2),and N_(2) conversion.
基金supported by the Shanxi Province Science Foundation for Youth(202303021212150)1331 Engineering of Shanxi Province,Research Grants Council of Hong Kong(CityU 11306517,11305919,and 11308620)National Natural Science Foundation of ChinaeResearch Grants Council of Hong Kong Joint Research Scheme N_CityU104/19.
文摘Lithium-sulfur(Li-S)batteries are regarded as one of the most promising next-generation energy storage systems due to their high theoretical specific energy density and low cost.However,serious shuttle effect and sluggish lithium polysulfides(LiPSs)redox kinetics severely impede the practical application of Li-S batteries.Employing polar sulfur hosts is an effective strategy to alleviate the above problems.Herein,the potential of two-dimensional(2D)Ti_(2)B-based sulfur hosts for Li-S batteries was comprehensively explored using first-principles calculations.The results show that functional groups of Ti_(2)B can significantly modulate its structural properties,thus affecting its interaction with sulfurcontaining species.Among S,Se,F,Cl,and Br elements,Ti_(2)B terminated with S and Se atoms possess stronger adsorption capability towards soluble Li_(2)S_(8),Li_(2)S_(6),and Li_(2)S_(4),obviously stronger than organic electrolytes,which indicates that they can completely suppress the shuttle effect.Besides,Ti_(2)BS_(2)and Ti_(2)BSe_(2)can powerfully expedite the electrochemical conversion of LiPSs.Moreover,the decomposition energy barrier of Li_(2)S and diffusion energy barrier of single Li ion on them are also fairly low,manifesting their excellent catalytic performance towards the oxidation of Li_(2)S.Finally,Ti_(2)BS_(2)and Ti_(2)BSe_(2)always keep metallic conductivity during the whole charge/discharge process.Taking all this into account,Ti_(2)BS_(2)and Ti_(2)BSe_(2)are proposed as promising bifunctional sulfur hosts for Li-S batteries.Our results suggest that increasing the proportion of S and Se groups during the synthesis of Ti_(2)B monolayers is greatly helpful for obtaining high-performance Li-S batteries.Besides,our work not only reveals the huge potential of 2D transition metal borides in Li-S batteries,but also provides insightful guidance for the design and screening of new efficient sulfur cathodes.
基金supported by the Research Grants Council of Hong Kong(Nos.CityU 11305919 and 11308620)the NSFC/RGC Joint Research Scheme(No.N_CityU104/19)+1 种基金Hong Kong Re-search Grant Council Collaborative Research Fund(Nos.C1002-21 G and C1017-22 G)This research made use of the comput-ing resources of the X-GPU cluster supported by the Hong Kong Research Grant Council Collaborative Research Fund(No.C6021-19EF).
文摘Rechargeable metal-air batteries generally require efficient,durable,and safe bifunctional electrocatalysts to simultaneously support oxygen reduction/evolution reactions(ORR/OER).Herein,we employed first-principles calculations to explore the structure-activity relationship between the composition control of metal atoms and the catalytic activity of Pt-doped Ti_(2-x) Mn_(x) CO_(2) single-atom catalysts(SACs).The research found a clear linear relationship between the proportion of Mn and bifunctional performance,which can effectively modulate catalytic activity.Additionally,it shows excellent bifunctional catalytic activity at medium concentrations,among which the catalyst of Pt-V_(O)-Ti_(0.89) Mn_(1.11) CO_(2) displays the lowest overpo-tential(η^(ORR/OER)=0.26/0.28 V).Attributed to the modulation of the average magnetism of Mn and the d-band center of Pt by different com ponents,the bonding strength of the active center of Pt to adsorption intermediates is changed,resulting in the enhancement of the catalyst activity.Crucially,the molecular orbital-level bonding between the active site Pt and the adsorbed intermediate OH is clarified,shedding light on the involvement of the partially occupied antibonding state of Pt’s d orbital in the activation process.The research extensively explores the control of catalyst activity through composition,offering strong support for designing and optimizing highly active Janus MXene-supported SACs.
