Predicting defect behavior in B2 intermetallics by merging ab initio modeling and machine learning

作     者：Bharat Medasani Anthony Gamst Hong Ding Wei Chen Kristin A Persson Mark Asta Andrew Canning Maciej Haranczyk 

作者机构：Computational Research DivisionLawrence Berkeley National LaboratoryBerkeleyCA 94720USA Physical and Computational Sciences DirectoratePacific National Northwest LaboratoryRichlandWA 99354USA Biostatistics and BioinformaticsUniversity of CaliforniaSan DiegoCA 92093USA Department of Materials Science&EngineeringUniversity of CaliforniaBerkeleyCA 94720USA Energy Technologies Area DivisionLawrence Berkeley National LaboratoryBerkeleyCA 94720USA Energy Technologies AreaLawrence Berkeley National LaboratoryBerkeleyCA 94720USA Department of MechanicalMaterials and Aerospace EngineeringIllinois Institute of TechnologyChicagoIL 60616USA Materials Sciences DivisionLawrence Berkeley National LaboratoryBerkeleyCA 94720USA 

出 版 物：《npj Computational Materials》 (计算材料学（英文）)

年 卷 期：2016年第2卷第1期

页      面：1-10页

核心收录：

学科分类：08[工学] 080502[工学-材料学] 0805[工学-材料科学与工程（可授工学、理学学位）] 

基　　金：supported by the Office of Science of the U.S.Department of Energy under Contract No.DEAC02-05CH11231 

主　　题：intermetallic intermetallics defect 

摘      要：We present a combination of machine learning and high throughput calculations to predict the points defects behavior in binary intermetallic(A–B)compounds,using as an example systems with the cubic B2 crystal structure(with equiatomic AB stoichiometry).To the best of our knowledge,this work is the first application of machine learning-models for point defect *** throughput first principles density functional calculations have been employed to compute intrinsic point defect energies in 100 B2 intermetallic *** systems are classified into two groups:(i)those for which the intrinsic defects are antisites for both A and B rich compositions,and(ii)those for which vacancies are the dominant defect for either or both composition *** data was analyzed by machine learning-techniques using decision tree,and full and reduced multiple additive regression tree(MART)*** these three schemes,a reduced MART(r-MART)model using six descriptors(formation energy,minimum and difference of electron densities at the Wigner–Seitz cell boundary,atomic radius difference,maximal atomic number and maximal electronegativity)presents the highest fit(98%)and predictive(75%)*** model is used to predict the defect behavior of other B2 compounds,and it is found that 45%of the compounds considered feature vacancies as dominant defects for either A or B rich compositions(or both).The ability to predict dominant defect types is important for the modeling of thermodynamic and kinetic properties of intermetallic compounds,and the present results illustrate how this information can be derived using modern tools combining high throughput calculations and data analytics.