Bandgap prediction by deep learning in configurationally hybridized graphene and boron nitride

查看全文 作　　者：Yuan [1]Dong;Chuhan [2]Wu;Chi [1]Zhang;Yingda [2]Liu;Jianlin [2]Cheng;Jian [1]Lin

机构地区：[1]Department of Mechanical&Aerospace Engineering,University of Missouri,Columbia,MO 65211,USA;[2]Department of Electrical Engineering&Computer Science,University of Missouri,Columbia,MO 65211,USA

出　　处：《npj Computational Materials》索引2019年第1期,共8页

基　　金：J.L.acknowledges financial support from University of Missouri-Columbia start-up fund,NASA Missouri Space Consortium(Project:00049784);Unite States Department of Agriculture(Award number:2018-67017-27880);This material is based upon work partially supported by the Department of Energy National Energy Technology Laboratory under Award Number DE-FE0031645;J.C.acknowledges National Science Foundation(Award numbers:DBI1759934 and IIS1763246);The computations were performed on the HPC resources at the University of Missouri Bioinformatics Consortium(UMBC),supported in part by NSF(award number:1429294). 摘　　要：It is well-known that the atomic-scale and nano-scale configuration of dopants can play a crucial role in determining the electronic properties of materials.However,predicting such effects is challenging due to the large range of atomic configurations that are possible.Here,we present a case study of how deep learning algorithms can enable bandgap prediction in hybridized boron–nitrogen graphene with arbitrary supercell configurations.A material descriptor that enables correlation of structure and bandgap was developed for convolutional neural networks.Bandgaps calculated by ab initio calculations,and corresponding structures,were used as training datasets.The trained networks were then used to predict bandgaps of systems with various configurations.For 4×4 and 5×5 supercells they accurately predict bandgaps,with a R^(2) of >90% and root-mean-square error of~0.1 eV.The transfer learning was performed by leveraging data generated from small supercells to improve the prediction accuracy for 6×6 supercells.This work will pave a route to future investigation of configurationally hybridized graphene and other 2D materials.Moreover,given the ubiquitous existence of configurations in materials,this work may stimulate interest in applying deep learning algorithms for the configurational design of materials across different length scales. 关键词：MATERIALS CONFIGURATION BORON

相关文献

参考文献（58）
引证文献（8）
耦合文献（12）

参考文献(58)

引证文献(8)

耦合文献(12)

互联网出版许可证新出网证(渝)字10号全国400电话 - 免长途话费

Bandgap prediction by deep learning in configurationally hybridized graphene and boron nitride

参考文献(58)

引证文献(8)

耦合文献(12)

用户提示