Insights into Cation Ordering of Double Perovskite Oxides from Machine Learning and Causal Relations

作     者：Ghosh, Ayana Palanichamy, Gayathri Trujillo, Dennis P. Shaikh, Monirul Ghosh, Saurabh 

作者机构：SRM Inst Sci & Technol Dept Phys & Nanotechnol Kattankulathur 603203 India Argonne Natl Lab Xray Sci Div Lemont IL 60439 USA Oak Ridge Natl Lab Ctr Nanophase Mat Sci Oak Ridge TN 37831 USA Oak Ridge Natl Lab Computat Sci & Engn Div Oak Ridge TN 37831 USA 

出 版 物：《CHEMISTRY OF MATERIALS》 (材料的化学性质)

年 卷 期：2022年第34卷第16期

页      面：7563-7578页

核心收录：

学科分类：08[工学] 0805[工学-材料科学与工程（可授工学、理学学位）] 080502[工学-材料学] 0703[理学-化学] 

基　　金：U.S. Department of Energy (DOE) , Office of Science, Office of Basic Energy Sciences Data Dr. Sergei V. Kali n i n (UT-Knoxville) DST-INSPIRE, Govt. of India [CRG/2018/001728] DST-SERB Core Research Grant [IF170335] DST National Supercomputing Mission, Government of India [DST/NSM/R&D _HPC _ Applications/Sanction/2021/34] 

主　　题：Machine learning 

摘      要：This work investigates origins of cation ordering in double perovskites using first-principles theory computations combined with machine learning (ML) and causal relations. We have considered various oxidation states of A, A , B, and B from the family of transition metal ions to construct a diverse compositional space. A conventional framework employing traditional ML classification algorithms such as Random Forest (RF) coupled with appropriate features including geometry-driven and key structural modes leads to accurate prediction (similar to 98%) of A-site cation ordering. We have evaluated the accuracy of ML models by employing analyses of decision paths, assignments of probabilistic confidence bound, and finally a direct non-Gaussian acyclic structural equation model to investigate causality. Our study suggests that structural modes are crucial for classifying layered, columnar, and rock-salt ordering. The charge difference between A and A is the most important feature for predicting clear layered ordering, which in turn depends on the B and B charge separation. We have also designed mathematical relationships with these features to derive energy differences to form clear layered ordering. The trilinear coupling between tilt, in-phase rotation, and A-site antiferroelectric displacement in the Landau free-energy expansion becomes the necessary condition behind formation of A-site cation ordering.