基于人工智能的描述符预测合金催化材料形成能

doi:10.3969/j.issn.0253-2778.2020.06.018

中国科学技术大学学报 ›› 2020, Vol. 50 ›› Issue (6): 844-851.DOI: 10.3969/j.issn.0253-2778.2020.06.018

基于人工智能的描述符预测合金催化材料形成能

李健聪，王泰然，舒武，胡素磊，欧阳润海，李微雪

1.中国科学技术大学合肥微尺度物质科学国家研究中心，安徽合肥 230026; 2.中国科学技术大学化学与材料科学学院化学物理系，安徽合肥 230026; 3.上海大学材料基因组工程研究院，上海 200444

收稿日期:2020-03-27 修回日期:2020-05-18 接受日期:2020-05-18 出版日期:2020-06-30 发布日期:2020-05-18
通讯作者: 李微雪
作者简介:李健聪，男，1995年生，硕士生. 研究方向：量子计算化学. E-mail: jcli@mail.ustc.edu.cn
基金资助:
科技部重点研发计划(2018YFA0208603)，中国科学院前沿重点项目(QYZDJ-SSW-SLH054)，国家自然科学基金重点项目(91645202, 91945302)资助.

AI-based descriptor for predicting alloy formation energy

LI Jiancong, WANG Tairan, SHU Wu, HU Sulei, OUYANG Runhai, LI Weixue

1. Hefei National Laboratory for Physical Sciences at the Microscale, University of Science and Technology of China, Hefei 230026, China; 2. Department of Chemical Physics, School of Chemistry and Materials Science, University of Science and Technology of China, Hefei 230026, China; 3. Materials Genome Institute, Shanghai University, Shanghai 200444, China

Received:2020-03-27 Revised:2020-05-18 Accepted:2020-05-18 Online:2020-06-30 Published:2020-05-18

摘要/Abstract

摘要： 合金材料因其丰富可调的几何结构和电子性质，在催化和材料科学领域得到了广泛的应用.其中合金形成能作为一个重要的物理量，对合金材料的形成和催化活性有重要影响.近年来，随着人工智能和数据库的发展，利用机器学习的方法研究和设计新的材料成为新的研究焦点.基于此，通过人工智能的多任务压缩感知算法，结合AB2合金形成能数据库展开了合金形成能描述符和预测研究. 首先建立了相应合金形成能的通用描述符，并展开了特征敏感性分析，揭示出合金材料组分的电子性质和几何性质的影响及其相互依赖关系.研究结果显示，该描述符的预测误差低于8.10 kJ·mol-1，具有清晰的物理可阐述性，并预测了大量未知合金材料的形成能.

关键词: 合金形成能, 数据库, 机器学习, 描述符, 敏感性分析

Abstract: Because of their rich geometric structure and electronic properties, metal alloys have been widely used in catalysis and materials science. Among them, alloys formation energy has an important influence on the formation and catalytic activity of metal alloys. With the development of artificial intelligence and databases in recent years, machine learning has been used to rationally design new materials. Based on the multi-task compressed sensing algotithm in artificial intelligence, the alloy formation energy descriptor of the AB2 alloy formation energy database was investigated. A universal descriptor of the corresponding alloy formation energy was established, and the sensitivity analysis of features revealed the importance of electronic and geometrical properties of metal alloys. The results show that this descriptor has a prediction error lower than 8.10kJ·mol-1 and a better physical interpretation. Finally, the formation energy of a large number of unknown metal alloys was predicted.

Key words: alloy formation energy, database, machine learning, descriptor, sensitivity analysis

中图分类号:

O642

李健聪，王泰然，舒武，胡素磊，欧阳润海，李微雪. 基于人工智能的描述符预测合金催化材料形成能[J]. 中国科学技术大学学报, 2020, 50(6): 844-851.

LI Jiancong, WANG Tairan, SHU Wu, HU Sulei, OUYANG Runhai, LI Weixue. AI-based descriptor for predicting alloy formation energy[J]. Journal of University of Science and Technology of China, 2020, 50(6): 844-851.

参考文献

［1］
MITTASCH A, FRANKENBURG W. Early studies of multicomponent catalysts [J]. Advances in Catalysis, 1950, 2(6): 81-104.
[2] SINFELT J H. Catalysis by alloys and bimetallic clusters [J]. Accounts of Chemical Research, 1977, 10(1): 15-20.
[3] SINFELT J H. Structure of metal catalysts [J]. Reviews of Modern Physics, 1979, 51(3): 569.
[4] XU F Y, DENG S B, XU J, et al. Highly active and stable Ni-Fe bimetal prepared by ball milling for catalytic hydrodechlorination of 4-chlorophenol[J]. Environmental Science & Technology, 2012, 46(8): 4576-4582.
[5] HUANG X, LI Y, et al. Synthesis of PtPd bimetal nanocrystals with controllable shape, composition, and their tunable catalytic properties[J]. Nano Letters, 2012, 12(8): 4265-4270.
[6] PARAMSOTHY M, GUPTA M, SRIKANTH N. Processing, microstructure, and properties of a Mg/Al bimetal macrocomposite[J]. Journal of Composite Materials, 2008, 42(24): 2567-2584.
[7] GAO W L, JIN R C, CHEN J X, et al. Titania-supported bimetallic catalysts for photocatalytic reduction of nitrate[J]. Catalysis Today, 2004, 90(3/4): 331-336.
[8] FERNNDEZ J L, WALSH D A, BARD A J. Thermodynamic guidelines for the design of bimetallic catalysts for oxygen electroreduction and rapid screening by scanning electrochemical microscopy. M-Co (M: Pd, Ag, Au)[J]. Journal of the American Chemical Society, 2005, 127(1): 357-365.
[9] STAMENKOVIC V R, FOWLER B, MUN B S, et al. Improved oxygen reduction activity on Pt3Ni(111) via increased surface site availability[J]. Science, 2007, 315(5811): 493-497.
[10] CRISAFULLI C, SCIRE S, MAGGIORE R, et al. CO2 reforming of methane over Ni-Ru and Ni-Pd bimetallic catalysts[J]. Catalysis Letters, 1999, 59(1): 21-26.
[11] DE S, ZHANG J, LUQUE R, et al. Ni-based bimetallic heterogeneous catalysts for energy and environmental applications[J]. Energy & Environmental Science, 2016, 9(11): 3314-3347.
[12] LEE J H, KATTEL S, JIANG Z, et al. Tuning the activity and selectivity of electroreduction of CO2 to synthesis gas using bimetallic catalysts[J]. Nature Communications, 2019, 10(1): 1-8.
[13] SCHWAB G. Some new aspects of the strength of alloys[J]. Transactions of the Faraday Society, 1949, 45：385-396.
[14] WANG C P, LIU X J, OHNUMA I, et al. Formation of immiscible alloy powders with egg-type microstructure[J]. Science, 2002, 297(5583): 990-993.
[15] COLINET C. High temperature calorimetry: Recent developments[J]. Journal of Alloys and Compounds, 1995，220(1/2): 76-87.
[16] TOPOR L, KLEPPA O J. Standard molar enthalpy of formation of LaB6 by high-temperature calorimetry[J]. The Journal of Chemical Thermodynamics, 1984, 16(10): 993-1002.
[17] KIM G, MESCHEL S V, NASH P, et al. Experimental formation enthalpies for intermetallic phases and other inorganic compounds[J]. Scientific Data, 2017, 4(1): 170162.
[18] MIEDEMA A R, CHTEL P F D, BOER F. Cohesion in alloys — fundamentals of a semi-empirical model[J]. Physica B+C, 1980, 100(1): 1-28.
[19] MIEDEMA A R, BOER F R, CHATEL P F. Empirical description of the role of electronegativity in alloy formation[J]. Journal of Physics F: Metal Physics, 1973, 3(8): 1558.
[20] ZHANG R F, RAJAN K. Statistically based assessment of formation enthalpy for intermetallic compounds[J]. Chemical Physics Letters, 2014, 612: 177-181.
[21] ZHANG R F, ZHANG S H, HE Z J, et al. Miedema calculator: A thermodynamic platform for predicting formation enthalpies of alloys within framework of Miedema’s Theory [J]. Computer Physics Communications，2016, 209: 58-69.
[22] PALINA N, SAKATA O, KUMARA L, et al. Electronic structure evolution with composition alteration of RhxCuy alloy nanoparticles[J]. Scientific Reports, 2017, 7(1): 1-9.
[23] FOILES S M, BASKES M I, DAW M S. Embedded-atom-method functions for the fcc metals Cu, Ag, Au, Ni, Pd, Pt, and their alloys[J]. Physical Review B, 1986, 33(12): 7983.
[24] JOHNSON R A. Alloy models with the embedded-atom method[J]. Physical Review B, 1989, 39(17): 12554.
[25] LEI Y W, SUN X R, ZHOU R L, et al. Embedded atom method potentials for Ce-Ni binary alloy[J]. Computational Materials Science, 2018, 150: 1-8.
[26] WILLIAMS A R, GELATT C D, MORUZZI V L. Microscopic basis of miedema’s empirical theory of transition-metal compound formation[J]. Physical Review Letters, 1980, 44(11): 394-395.
[27] ZHANG Y, KRESSE G, WOLVERTON C. Nonlocal first-principles calculations in Cu-Au and other intermetallic alloys[J]. Physical Review Letters, 2014, 112(7): 75502.
[28] FARSI L, DESKINS N A. First principles analysis of surface dependent segregation in bimetallic alloys[J]. Physical Chemistry Chemical Physics, 2019, 21(42): 23626-23637.
[29] REITH D, PODLOUCKY R. First-principles model study of the phase stabilities of dilute Fe-Cu alloys: Role of vibrational free energy[J]. Physical Review B, 2009, 80(5): 54108.
[30] OU L. The origin of enhanced electrocatalytic activity of Pt-M (M= Fe, Co, Ni, Cu, and W) alloys in PEM fuel cell cathodes: A DFT computational study[J]. Computational and Theoretical Chemistry, 2014, 1048: 69-76.
[31] GUBAEV K, PODRYABINKIN E V, HART G L, et al. Accelerating high-throughput searches for new alloys with active learning of interatomic potentials[J]. Computational Materials Science, 2019, 156: 148-156.
[32] DOAK J W, HAO S, KIRKLIN S, et al. Computational prediction of nanostructured alloys with enhanced thermoelectric properties[J]. Physical Review Materials, 2019, 3(10): 105404.
[33] JAIN A, ONG S P, HAUTIER G, et al. Commentary: The materials project: A materials genome approach to accelerating materials innovation[J]. APL Materials, 2013, 1(1): 11002.
[34] SAAL J E, KIRKLIN S, AYKOL M, et al. Materials design and discovery with high-throughput density functional theory: The open quantum materials database (OQMD)[J]. JOM, 2013, 65(11): 1501-1509.
[35] WU X D, ZHU X Q, WU G Q, et al. Data mining with big data[J]. IEEE Transactions on Knowledge and Data Engineering, 2013, 26(1): 97-107.
[36] ROSCHER R, BOHN B, DUARTE M F, et al. Explainable machine learning for scientific insights and discoveries[J]. IEEE Access, 2020, 8: 42200- 42216.
[37] LI Z, WANG S W, CHIN W S, et al. High-throughput screening of bimetallic catalysts enabled by machine learning[J]. Journal of Materials Chemistry A, 2017, 5(46): 24131-24138.
[38] GHIRINGHELLI L M, VYBIRAL J, LEVCHENKO S V, et al. Big data of materials science: Critical role of the descriptor[J]. Physical Review Letters, 2015, 114(10): 105503.
[39] RUBAN A, HAMMER B, STOLTZE P, et al. Surface electronic structure and reactivity of transition and noble metals[J]. Journal of Molecular Catalysis A: Chemical, 1997, 115(3): 421-429.
[40] ZHAO Z J, LIU S H, ZHA S J, et al. Theory-guided design of catalytic materials using scaling relationships and reactivity descriptors[J]. Nature Reviews Materials, 2019, 4: 792-804.
[41] CHOI K, JANG Y, CHUNG D Y, et al. A simple synthesis of urchin-like Pt-Ni bimetallic nanostructures as enhanced electrocatalysts for the oxygen reduction reaction[J]. Chemical Communications, 2016, 52(3): 597-600.
[42] LI H, WEN P, LI Q, et al. Earth-abundant iron diboride (FeB2) nanoparticles as highly active bifunctional electrocatalysts for overall water splitting[J]. Advanced Energy Materials, 2017, 7(17): 1700513.
[43] YUAN W Y, ZHAO X S, HAO W J, et al. Performance of surface-oxidized Ni3B, Ni2B, and NiB2 electrocatalysts for overall water splitting[J]. ChemElectroChem, 2019, 6(3): 764-770.
[44] ZALUSKA A, ZALUSKI L, STRM-OLSEN J O. Synergy of hydrogen sorption in ball-milled hydrides of Mg and Mg2Ni[J]. Journal of Alloys and Compounds, 1999, 289(1/2): 197-206.
[45] ZHANG L T, CAI Z L, YAO Z D, et al. A striking catalytic effect of facile synthesized ZrMn2 nanoparticles on the de/rehydrogenation properties of MgH2[J]. Journal of Materials Chemistry A, 2019, 7(10): 5626-5634.
[46] OUYANG R H, CURTAROLO S, AHMETCIK E, et al. SISSO: A compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates[J]. Physical Review Materials, 2018, 2: 083802.
[47] OUYANG R H, AHMETCIK E, CARBOGNO C, et al. Simultaneous learning of several materials properties from incomplete databases with multi-task SISSO[J]. Journal of Physics: Materials, 2019, 2(2): 24002.
[48] BARTEL C J, MILLICAN S L, DEML A M, et al. Physical descriptor for the Gibbs energy of inorganic crystalline solids and temperature-dependent materials chemistry[J]. Nature Communications, 2018, 9: 4168.
[49] BARTEL C J, SUTTON C, GOLDSMITH B R, et al. New tolerance factor to predict the stability of perovskite oxides and halides[J]. Science Advances, 2019, 5(2): evva0693.
[50] ANDERSEN M, LEVCHENKO S V, SCHEFFLER M, et al. Beyond scaling relations for the description of catalytic materials[J]. ACS Catalysis, 2019, 9(4): 2752-2759.
[51] OUYANG R H. Exploiting ionic radii for rational design of halide perovskites[J]. Chemistry of Materials，2020, 32(1): 595-604.
[52] HAYNES W M. CRC Handbook of Chemistry and Physics[M]. Boca Raton, FL: CRC Press, 2014.
[53] MEDVEDEV V A, COX J D, WAGMAN D D. CODATA Key Values for Thermodynamics[M]. New York: Hemisphere Publishing Corporation, 1989.
[54] BLUM A L, LANGLEY P. Selection of relevant features and examples in machine learning[J]. Artificial Intelligence, 1997, 97(1/2): 245-271.
[55] BREIMAN L, SPECTOR P. Submodel selection and evaluation in regression. The X-random case[J]. International Statistical Review, 1992, 60(3): 291-319.
[56] RAO R B, FUNG G, ROSALES R. On the dangers of cross-validation. An experimental evaluation[C]// Proceedings of the 2008 SIAM International Conference on Data Mining. Society for Industrial and Applied Mathematics, 2008：588.
[57] GHIRINGHELLI L M, VYBIRAL J, AHMETCIK E, et al. Learning physical descriptors for materials science by compressed sensing[J]. New Journal of Physics, 2017, 19(2): 23017.
[58] MA X F, LI Z, ACHENIE L E K, et al. Machine-learning-augmented chemisorption model for CO2 electroreduction catalyst screening[J]. The Journal of Physical Chemistry Letters, 2015, 6(18): 3528-3533.
[59] 吴春峰, 李慧改, 郑少波, 等. 二元合金热力学模型—Miedema模型[J].上海金属, 2011, 33(4): 1-5.
[60] WATSON R E, BENNETT L H. Optimized predictions for heats of formation of transition-metal alloys II [J]. Calphad, 1984, 8(4): 307-321.

()
()

[1]	高翔，陈力. 分组随机梯度下降法：掉队和延迟的平衡[J]. 中国科学技术大学学报, 2020, 50(8): 1156-1161.
[2]	李鹏，郑宇，张谈贵. 基于视觉显著性的无人机目标跟踪[J]. 中国科学技术大学学报, 2020, 50(8): 1162-1169.
[3]	龙飞, 黄琨, 李峰. 一种机器学习与相变之间的新型映射[J]. 中国科学技术大学学报, 2020, 50(1): 18-28.
[4]	栾艳红，孙雪姣. 基于CP-net偏好的关系数据库的Top-k实现[J]. 中国科学技术大学学报, 2019, 49(2): 93-99.
[5]	谭嘉莉，何钰，吴燕晶，孙广中. 应用于校园心理咨询的对话匹配度预测模型[J]. 中国科学技术大学学报, 2018, 48(9): 739-747.
[6]	李勇明，肖洁，王品，颜芳. 基于封装式偏差回归的心脏生理和病理年龄估计算法[J]. 中国科学技术大学学报, 2018, 48(9): 762-769.
[7]	林宇鹏，谢智歌，徐凯，陈飞，刘利刚. 基于超限学习机的快速癌症检测方法[J]. 中国科学技术大学学报, 2018, 48(2): 154-160.
[8]	张莉，卢星凝，陆从林，王邦军，李凡长. 支持向量机在高考成绩预测分析中的应用[J]. 中国科学技术大学学报, 2017, 47(1): 1-9.

基于人工智能的描述符预测合金催化材料形成能

AI-based descriptor for predicting alloy formation energy

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价