Recently, machine learning(ML) has become a widely used technique in materials science study. Most work focuses on predicting the rule and overall trend by building a machine learning model. However,new insights are o...Recently, machine learning(ML) has become a widely used technique in materials science study. Most work focuses on predicting the rule and overall trend by building a machine learning model. However,new insights are often learnt from exceptions against the overall trend. In this work, we demonstrate that how unusual structures are discovered from exceptions when machine learning is used to get the relationship between atomic and electronic structures based on big data from high-throughput calculation database. For example, after training an ML model for the relationship between atomic and electronic structures of crystals, we find AgO2 F, an unusual structure with both Ag3+and O22à, from structures whose band gap deviates much from the prediction made by our model. A further investigation on this structure might shed light into the research on anionic redox in transition metal oxides of Li-ion batteries.展开更多
Material identification technique is crucial to the development of structure chemistry and materials genome project. Current methods are promising candidates to identify structures effectively, but have limited abilit...Material identification technique is crucial to the development of structure chemistry and materials genome project. Current methods are promising candidates to identify structures effectively, but have limited ability to deal with all structures accurately and automatically in the big materials database because different material resources and various measurement errors lead to variation of bond length and bond angle. To address this issue, we propose a new paradigm based on graph theory(GTscheme) to improve the efficiency and accuracy of material identification, which focuses on processing the "topological relationship" rather than the value of bond length and bond angle among different structures. By using this method, automatic deduplication for big materials database is achieved for the first time, which identifies 626,772 unique structures from 865,458 original structures.Moreover, the graph theory scheme has been modified to solve some advanced problems such as identifying highly distorted structures, distinguishing structures with strong similarity and classifying complex crystal structures in materials big data.展开更多
The widely used density functional theory(DFT)has a major drawback of underestimating the band gaps of materials.Wannier–Koopmans method(WKM)was recently developed for band gap calculations with accuracy on a par wit...The widely used density functional theory(DFT)has a major drawback of underestimating the band gaps of materials.Wannier–Koopmans method(WKM)was recently developed for band gap calculations with accuracy on a par with more complicated methods.WKM has been tested for main group covalent semiconductors,alkali halides,2D materials,and organic crystals.Here we apply the WKM to another interesting type of material system:the transition metal(TM)oxides.展开更多
An octahedral Nb6 structural unit with space aromaticity is identified for the first time in a transition–metal monoxide crystal Nb3O3 by ab initio calculations.The strong Nb–Nb metallic bonding facilitates the form...An octahedral Nb6 structural unit with space aromaticity is identified for the first time in a transition–metal monoxide crystal Nb3O3 by ab initio calculations.The strong Nb–Nb metallic bonding facilitates the formation of stable octahedral Nb6 structural units and the release of delocalization energy.Moreover,the Nb atoms in continuously connected Nb6 structural units share their electrons with each other in a continuous space of framework,so that the electrons are uniformly distributed.The newly discovered aromaticity in the octahedral Nb6 structural units extends the range of aromatic compounds and broadens our vision in structural chemistry.展开更多
基金supported by the Director, Office of Science (SC), Basic Energy Science (BES), Materials Science and Engineering Division (MSED), of the US Department of Energy (DOE) under Contract No. DE-AC02-05CH11231 through the Materials Theory program (KC2301) under Contract No. DE-AC02-05CH11231financially supported by the National Key R&D Program of China (2016YFB0700600)+1 种基金Shenzhen Science and Technology Research Grant (ZDSYS201707281026184)Guangdong Keylab Project (2017B0303010130)
文摘Recently, machine learning(ML) has become a widely used technique in materials science study. Most work focuses on predicting the rule and overall trend by building a machine learning model. However,new insights are often learnt from exceptions against the overall trend. In this work, we demonstrate that how unusual structures are discovered from exceptions when machine learning is used to get the relationship between atomic and electronic structures based on big data from high-throughput calculation database. For example, after training an ML model for the relationship between atomic and electronic structures of crystals, we find AgO2 F, an unusual structure with both Ag3+and O22à, from structures whose band gap deviates much from the prediction made by our model. A further investigation on this structure might shed light into the research on anionic redox in transition metal oxides of Li-ion batteries.
基金supported by the National Key R&D Program of China (2016YFB0700600)the National Natural Science Foundation of China (21603007, 51672012)+1 种基金Soft Science Research Project of Guangdong Province (2017B030301013)New Energy Materials Genome Preparation & Test Key-Laboratory Project of Shenzhen (ZDSYS201707281026184)
文摘Material identification technique is crucial to the development of structure chemistry and materials genome project. Current methods are promising candidates to identify structures effectively, but have limited ability to deal with all structures accurately and automatically in the big materials database because different material resources and various measurement errors lead to variation of bond length and bond angle. To address this issue, we propose a new paradigm based on graph theory(GTscheme) to improve the efficiency and accuracy of material identification, which focuses on processing the "topological relationship" rather than the value of bond length and bond angle among different structures. By using this method, automatic deduplication for big materials database is achieved for the first time, which identifies 626,772 unique structures from 865,458 original structures.Moreover, the graph theory scheme has been modified to solve some advanced problems such as identifying highly distorted structures, distinguishing structures with strong similarity and classifying complex crystal structures in materials big data.
基金L.-W.W.is supported by the Director,Office of Science,the Office of Basic Energy Sciences(BES),Materials Sciences and Engineering(MSE)Division of the U.S.Department of Energy(DOE)through the theory of material(KC2301)program under Contract No.DEAC02-05CH11231This work is also financially supported by National Materials Genome Project of China(2016YFB0700600)Shenzhen Science and Technology Research Grant(Nos JCYJ20160226105838578 and JCYJ20151015162256516).
文摘The widely used density functional theory(DFT)has a major drawback of underestimating the band gaps of materials.Wannier–Koopmans method(WKM)was recently developed for band gap calculations with accuracy on a par with more complicated methods.WKM has been tested for main group covalent semiconductors,alkali halides,2D materials,and organic crystals.Here we apply the WKM to another interesting type of material system:the transition metal(TM)oxides.
基金financially supported by National Key R&D Program of China(2016YFB0700600)Soft Science Research Project of Guangdong Province(2017B030301013)Shenzhen Science and Technology Research Grant(ZDSYS201707281026184).
文摘An octahedral Nb6 structural unit with space aromaticity is identified for the first time in a transition–metal monoxide crystal Nb3O3 by ab initio calculations.The strong Nb–Nb metallic bonding facilitates the formation of stable octahedral Nb6 structural units and the release of delocalization energy.Moreover,the Nb atoms in continuously connected Nb6 structural units share their electrons with each other in a continuous space of framework,so that the electrons are uniformly distributed.The newly discovered aromaticity in the octahedral Nb6 structural units extends the range of aromatic compounds and broadens our vision in structural chemistry.
