Machine learning hydrogen adsorption on nanoclusters through structural descriptors

作     者：Marc O.J.Jäger Eiaki V.Morooka Filippo Federici Canova Lauri Himanen Adam S.Foster 

作者机构：Department of Applied PhysicsAalto UniversityP.O.Box 1110000076 AaltoEspooFinland Nanolayers Research Computing Ltd15 Southgrove RoadSheffield S102NPUK WPI Nano Life Science Institute(WPI-NanoLSI)Kanazawa UniversityKakuma-machiKanazawa 920-1192Japan Graduate School Materials ScienceStaudinger Weg 9Mainz 55128Germany 

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

年 卷 期：2018年第4卷第1期

页      面：346-353页

核心收录：

学科分类：081704[工学-应用化学] 07[理学] 070304[理学-物理化学(含∶化学物理)] 08[工学] 0817[工学-化学工程与技术] 0703[理学-化学] 

基　　金：The work was supported by the World Premier International Research Center Initiative(WPI) MEXT Japan and the European Union’s Horizon 2020 research and innovation program under grant agreement no.676580 NOMAD a European Center of Excellence and no.686053 CRITCAT 

主　　题：adsorption clusters molybdenum 

摘      要：Catalytic activity of the hydrogen evolution reaction on nanoclusters depends on diverse adsorption site *** learning reduces the cost for modelling those sites with the aid of *** analysed the performance of state-of-the-art structural descriptors Smooth Overlap of Atomic Positions,Many-Body Tensor Representation and Atom-Centered Symmetry Functions while predicting the hydrogen adsorption(free)energy on the surface of *** 2D-material molybdenum disulphide and the alloy copper–gold functioned as test *** energy scans of hydrogen on the cluster surfaces were conducted to compare the accuracy of the descriptors in kernel ridge *** having recourse to data sets of 91 molybdenum disulphide clusters and 24 copper–gold clusters,we found that the mean absolute error could be reduced by machine learning on different clusters simultaneously rather than *** adsorption energy was explained by the local descriptor Smooth Overlap of Atomic Positions,combining it with the global descriptor Many-Body Tensor Representation did not improve the overall *** concluded that fitting of potential energy surfaces could be reduced significantly by merging data from different nanoclusters.