From electrons to phase diagrams with machine learning potentials using pyiron based automated workflows  

作　　者：Sarath Menon Yury Lysogorskiy Alexander L.M.Knoll Niklas Leimeroth Marvin Poul Minaam Qamar Jan Janssen Matous Mrovec Jochen Rohrer Karsten Albe Jörg Behler Ralf Drautz Jörg Neugebauer 

机构地区：[1]Max-Planck-Institut für Nachhaltige Materialien GmbH,40237 Düsseldorf,Germany [2]ICAMS,Ruhr-Universität Bochum,44801 Bochum,Germany [3]Lehrstuhl für Theoretische Chemie II,Ruhr-Universität Bochum,44780 Bochum,Germany [4]Research Center Chemical Sciences and Sustainability,Research Alliance Ruhr,44780 Bochum,Germany [5]Technische Universität Darmstadt,Fachbereich Material und Geowissenschaften,Fachgebiet Materialmodellierung,64287 Darmstadt,Germany [6]Max-Planck-Institut für Eisenforschung GmbH,40237 Düsseldorf,Germany

出　　处：《npj Computational Materials》2024年第1期454-468,共15页计算材料学(英文)

基　　金：The workflows,potentials,and results presented here were obtained in the framework of the POTENTIALS collaboration and scientific network“Assessment of atomistic simulations”with funding from the German Science Foundation(DFG)(grant number 405602047);S.M.acknowledges funding by the Deutsche Forschungsgemeinschaft(DFG,German Research Foundation)under the National Research Data Infrastructure-NFDI 38/1-project number 460247524;J.B.acknowledges funding by the DFG(project number 405479457 as part of PAK 965/1);A.K.acknowledges funding by the Studienstiftung des Deutschen Volkes(doctoral scholarship);N.L.and J.R.acknowledge funding by the Deutsche Forschungsgemeinschaft(DFG,German Research Foundation)under grant number 405621137;K.A.acknowledges funding from the the DFG undergrant number 405621160;M.M.and R.D.acknowledge funding by the German Science Foundation(DFG),projects 405621081 and 405621217.R.D.and Y.L.acknowledge computation time by Center for Interface-Dominated High Performance Materials(ZGH)at Ruhr-Universität Bochum,Germany;J.J.and J.N.acknowledge funding by the DFG under grant number 405621217.M.P.and J.N.acknowledge funding from the DFG under grant number 405621160.

摘　　要：We present a comprehensive and user-friendly framework built upon the pyiron integrated development environment(IDE),enabling researchers to perform the entire Machine Learning Potential(MLP)development cycle consisting of(i)creating systematic DFT databases,(ii)fitting the Density Functional Theory(DFT)data to empirical potentials orMLPs,and(iii)validating the potentials in a largely automatic approach.The power and performance of this framework are demonstrated for three conceptually very different classes of interatomic potentials:an empirical potential(embedded atom method-EAM),neural networks(high-dimensional neural network potentials-HDNNP)and expansions in basis sets(atomic cluster expansion-ACE).As an advanced example for validation and application,we show the computation of a binary composition-temperature phase diagram for Al-Li,a technologically important lightweight alloy system with applications in the aerospace industry.

关 键 词：alloy phase NEURAL 

分 类 号：TG1[金属学及工艺—金属学]