Machine learning prediction of activation energy in cubic Li-argyrodites with hierarchically encoding crystal structure-based(HECS)descriptors  

基于分层编码晶体结构(HECS)描述符的机器学习预测立方相锂-硫银锗矿电解质材料的激活能

作　　者：Qian Zhao Maxim Avdeev Liquan Chen Siqi Shi 赵倩;Maxim Avdeev;陈立泉;施思齐(Materials Genome Institute,Shanghai University,Shanghai 200444,China;Australian Nuclear Science and Technology Organization,New Illawarra Rd,Lucas Heights,NSW 2234,Australia;School of Chemistry,The University of Sydney,Sydney 2006,Australia;Key Laboratory for Renewable Energy,Institute of Physics,Chinese Academy of Sciences,Beijing 100190,China;School of Materials Science and Engineering,Shanghai University,Shanghai 200444,China)

机构地区：[1]Materials Genome Institute,Shanghai University,Shanghai 200444,China [2]Australian Nuclear Science and Technology Organization,New Illawarra Rd,Lucas Heights,NSW 2234,Australia [3]School of Chemistry,The University of Sydney,Sydney 2006,Australia [4]Key Laboratory for Renewable Energy,Institute of Physics,Chinese Academy of Sciences,Beijing 100190,China [5]School of Materials Science and Engineering,Shanghai University,Shanghai 200444,China

出　　处：《Science Bulletin》2021年第14期1401-1408,M0003,共9页科学通报（英文版）

基　　金：the National Key Research and Development Program of China(2017YFB0701600);the National Natural Science Foundation of China(11874254,51622207,and U1630134)。

摘　　要：Rational design of solid-state electrolytes(SSEs)with high ionic conductivity and low activation energy(Ea)is vital for all solid-state batteries.Machine learning(ML)techniques have recently been successful in predicting Li^(+) conduction property in SSEs with various descriptors and accelerating the development of SSEs.In this work,we extend the previous efforts and introduce a framework of ML prediction for E_(a) in SSEs with hierarchically encoding crystal structure-based(HECS)descriptors.Taking cubic Li-argyrodites as an example,an Ea prediction model is developed to the coefficient of determination(R^(2))and rootmean-square error(RMSE)values of 0.887 and 0.02 eV for training dataset,and 0.820 and 0.02 eV for test dataset,respectively by partial least squares(PLS)analysis,proving the prediction power of HECSdescriptors.The variable importance in projection(VIP)scores demonstrate the combined effects of the global and local Li^(+) conduction environments,especially the anion size and the resultant structural changes associated with anion site disorder.The developed E_(a) prediction model directs us to optimize and design new Li-argyrodites with lower Ea,such as Li_(6–x)PS_(5–x)Cl_(1+x)(<0.322 eV),Li_(6+x)PS_(5+x)Br_(1–x)(<0.273 eV),Li_(6+x)PS_(5+x)Br_(0.25)I_(0.75–x)(<0.352 eV),Li_(6+(5–n)y)P_(1–y)N_(y)S_(5)I(<0.420 eV),Li_(6+(5–n)y)As_(1–y)N_(y)S_(5)I(<0.371 eV),Li_(6+(5–n)y)As_(1–y)NySe_(5)I(<0.450 eV),by broadening bottleneck size,invoking site disorder and activating concerted Li+conduction.This analysis shows great potential in promoting rational design of advanced SSEs and the same approach can be applied to other types of materials.合理设计高离子电导率和低活化能(Ea)的固态电解质(SSEs)对全固态电池至关重要.近年来,基于各种描述符的机器学习技术成功地预测了锂离子在SSEs中的传导特性.本文建立了一个通用的基于HECS描述符的机器学习预测无机SSEs材料Ea的框架,并且以立方相锂-硫银锗矿型SSEs材料作为模型体系进行实例研究,采用偏最小二乘方法(PLS)建立了高精度(训练集:R^(2),88.7%;RMSE,0.02 e V;测试集:R^(2),82.0%;RMSE,0.02 e V)预测E_(a)的模型.变量投影重要性(VIP)分析表明了全局及局域离子传导环境对E_(a)的联合作用,其中平均阴离子尺寸以及与阴离子位置无序密切相关的结构的改变对激活能值的贡献尤为突出,这一发现有助于进一步指导发现或设计新的无机固态电解质材料.同时,对该模型进行的知识提取表明,可以通过增大瓶颈尺寸、引发阴离子位置无序、激活离子协同迁移等优化和设计出具有高离子传导性能的新的无机固态电解质材料.

关 键 词：Solid-state electrolytes(SSEs) Hierarchically encoding crystal structurebased (HECS)descriptors Predicting activation energy Cubic Li-argyrodites Machine learning 

分 类 号：TM912[电气工程—电力电子与电力传动] TB34[一般工业技术—材料科学与工程]