| 688 | 4 | 212 |
| 下载次数 | 被引频次 | 阅读次数 |
理论晶体结构预测可以在给定化学组分的条件下确定材料的晶体结构，已成为材料科学研究的重要工具。然而，该方法一直面临计算成本高的瓶颈问题。近年来，新兴机器学习方法在传统科学计算上展现了广阔的应用前景，逐渐被引入到晶体结构预测领域。本文主要讨论机器学习方法在理论晶体结构预测领域的最新研究进展，分别从加速晶体结构的能量计算和势能面的探索两个方面介绍领域的最新成果，并对未来研究可能的发展提出抛砖引玉的见解。
Abstract：Crystal structure prediction is a powerful theoretical simulation tool, which can determine the crystal structure of materials with the given information of chemical composition. However, its application is severely limited due to the highly computational cost. In recent years, the state-of-art machine learning methods reveal a promising prospect in accelerating the conventional scientific computing,thus introducing the methods into the crystal structure prediction. This review briefly introduced recent progress on the application of machine learning for the crystal structure prediction. Two aspects were discussed, i.e., accelerating the energy evaluation and enhancing the potential energy surface sampling. In addition, some insights into the future development in this aspect were also suggested.
[1] STILLINGER F H. Exponential multiplicity of inherent structures[J].Phys Rev E, 1999, 59(1):48–51.
[2] STILLINGER F H. Inherent structures enumeration for low-density materials[J]. Phys Rev E, 2001, 63(1):011110.
[3] WALES D J, DOYE J P K. Global optimization by basin-hopping and the lowest energy structures of lennard-jones clusters containing up to110 atoms[J]. J Phys Chem, 1997, 101(28):5111–5116.
[4] GLASS C W, OGANOV A R, HANSEN N. USPEX—Evolutionary crystal structure prediction[J]. Comput Phys Commun, 2006,175(11/12):713–720.
[5] OGANOV A R, GLASS C W. Crystal structure prediction using ab initio evolutionary techniques:Principles and applications[J]. J Chem Phys, 2006, 124(24):244704.
[6] WANG Y, LV J, ZHU L, et al. Crystal structure prediction via particle-swarm optimization[J]. Phys Rev B, 2010, 82(9):094116.
[7] WANG Y, LV J, ZHU L, et al. CALYPSO:A method for crystal structure prediction[J]. Comput Phys Commun, 2012, 183(10):2063–2070.
[8] PICKARD C J, NEEDS R J. Ab initio random structure searching[J]. J Phys Condens Matter, 2011, 23(5):053201.
[9] PICKARD C J, NEEDS R J. High-pressure phases of silane[J]. Phys Rev Lett, 2006, 97(4):045504.
[10] MA Y, EREMETS M, OGANOV A R, et al. Transparent dense sodium[J]. Nature, 2009, 458(7235):182–185.
[11] LV J, WANG Y, ZHU L, et al. Predicted novel high-pressure phases of lithium[J]. Phys Rev Lett, 2011, 106(1):015503.
[12] WANG H, TSE J S, TANAKA K, et al. Superconductive sodalite-like clathrate calcium hydride at high pressures[J]. Proc National Acad Sci,2012, 109(17):6463–6466.
[13] PENG F, SUN Y, PICKARD C J, et al. Hydrogen clathrate structures in rare earth hydrides at high pressures:Possible route to room-temperature superconductivity[J]. Phys Rev Lett, 2017, 119(10):107001.
[14] SCHMIDT J, MARQUES M R G, BOTTI S, et al. Recent advances and applications of machine learning in solid-state materials science[J].Npj Comput Mater, 2019, 5(1):83.
[15] WANG J, WANG Y, CHEN Y. Inverse design of materials by machine learning[J]. Materials, 2022, 15(5):1811.
[16] WANG Y, LV J, GAO P, et al. Crystal structure prediction via efficient sampling of the potential energy surface[J]. Accounts Chem Res, 2022,55(15):2068–2076.
[17] YIN X, GOUNARIS C E. Search methods for inorganic materials crystal structure prediction[J]. Curr Opin Chem Eng, 2022, 35:100726.
[18] KOCER E, KO T W, BEHLER J. Neural network potentials:A concise overview of methods[J]. Annu Rev Phys Chem, 2022, 73(1):1–24.
[19] BEHLER J. Four generations of high-dimensional neural network potentials[J]. Chem Rev, 2021, 121(16):10037–10072.
[20] FUHR A S, SUMPTER B G. Deep generative models for materials discovery and machine learning-accelerated innovation[J]. Front Mater,2022, 9:865270.
[21] CHEN L, ZHANG W, NIE Z, et al. Generative models for inverse design of inorganic solid materials[J]. J Mater Inform, 2021, 1:4.
[22] HOHENBERG P, KOHN W. Inhomogeneous electron gas[J]. Phys Rev,1964, 136(3B):B864–B871.
[23] KOHN W, SHAM L J. Self-consistent equations including exchange and correlation effects[J]. Phys Rev, 1965, 140(4A):A1133–A1138.
[24] LENNARD-JONES J E. Cohesion[J]. P Phys Soc, 1931, 43(5):461–482.
[25] BUCKINGHAM R A. The classical equation of state of gaseous helium, neon and argon[J]. Proc Royal Soc Lond Ser Math Phys Sci,1938, 168(933):264–283.
[26] STILLINGER F H, WEBER T A. Computer simulation of local order in condensed phases of silicon[J]. Phys Rev B, 1984, 31(8):5262–5271.
[27] DAW M S, BASKES M I. Embedded-atom method:Derivation and application to impurities, surfaces, and other defects in metals[J]. Phys Rev B, 1984, 29(12):6443–6453.
[28] THOMPSON A P, AKTULGA H M, BERGER R, et al. LAMMPS-a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales[J]. Comput Phys Commun, 2022,271:108171.
[29] BEHLER J. Perspective:Machine learning potentials for atomistic simulations[J]. J Chem Phys, 2016, 145(17):170901.
[30] TONG Q, XUE L, LV J, et al. Accelerating CALYPSO structure prediction by data-driven learning of a potential energy surface[J].Faraday Discuss, 2018, 211(0):31–43.
[31] GUBAEV K, PODRYABINKIN E V, HART G L W, et al. Accelerating high-throughput searches for new alloys with active learning of interatomic potentials[J]. Comp Mater Sci, 2019, 156:148–156.
[32] BLANK T B, BROWN S D, CALHOUN A W, et al. Neural network models of potential energy surfaces[J]. J Chem Phys, 1995, 103(10):4129–4137.
[33] BEHLER J, PARRINELLO M. Generalized neural-network representation of high-dimensional potential-energy surfaces[J]. Phys Rev Lett, 2007, 98(14):146401.
[34] BEHLER J. Atom-centered symmetry functions for constructing high-dimensional neural network potentials[J]. J Chem Phys, 2011,134(7):074106.
[35] KHALIULLIN R Z, ESHET H, KüHNE T D, et al. Nucleation mechanism for the direct graphite-to-diamond phase transition[J]. Nat Mater:2011, 10(9):693–697.
[36] ARTRITH N, URBAN A. An implementation of artificial neural-network potentials for atomistic materials simulations:Performance for TiO2[J]. Comp Mater Sci:2016, 114:135–150.
[37] BEHLER J. Four generations of high-dimensional neural network potentials[J]. Chem Rev, 2021, 121(16):10037–10072.
[38] BARTóK A P, KONDOR R, CSáNYI G. On representing chemical environments[J]. Phys Rev B, 2013, 87(18):184115.
[39] BARTóK A P, PAYNE M C, KONDOR R, et al. Gaussian approximation potentials:the accuracy of quantum mechanics, without the electrons[J]. Phys Rev Lett, 2010, 104(13):136403.
[40] DERINGER V L, CSáNYI G, PROSERPIO D M. Extracting crystal chemistry from amorphous carbon structures[J]. ChemPhysChem,2017, 18(8):873–877.
[41] DERINGER V L, CSáNYI G. Machine learning based interatomic potential for amorphous carbon[J]. Phys Rev B, 2017, 95(9):094203.
[42] ROWE P, DERINGER V L, GASPAROTTO P, et al. An accurate and transferable machine learning potential for carbon[J]. J Chem Phys,2020, 153(3):034702.
[43] DERINGER V L, PICKARD C J, CSáNYI G. Data-driven learning of total and local energies in elemental boron[J]. Phys Rev Lett, 2018,120(15):156001.
[44] SZLACHTA W J, BARTóK A P, CSáNYI G. Accuracy and transferability of Gaussian approximation potential models for tungsten[J]. Phys Rev B, 2014, 90(10):104108.
[45] DRAGONI D, DAFF T D, CSáNYI G, et al. Achieving DFT accuracy with a machine-learning interatomic potential:Thermomechanics and defects in bcc ferromagnetic iron[J]. Phys Rev Mater, 2018, 2(1):013808.
[46] DERINGER V L, BARTóK A P, BERNSTEIN N, et al. Gaussian process regression for materials and molecules[J]. Chem Rev:2021,121(16):10073–10141.
[47] SHAPEEV A V. Moment tensor potentials:A class of systematically improvable interatomic potentials[J]. Multiscale Model Sim, 2016,14(3):1153–1173.
[48] NOVIKOV I S, GUBAEV K, PODRYABINKIN E V, et al. The MLIP package:moment tensor potentials with MPI and active learning[J].Mach Learn Sci Technology, 2021, 2(2):025002.
[49] CHEN C, DENG Z, TRAN R, et al. Accurate force field for molybdenum by machine learning large materials data[J]. Phys Rev Mater, 2017, 1(4):043603.
[50] LI X-G, HU C, CHEN C, et al. Quantum-accurate spectral neighbor analysis potential models for Ni-Mo binary alloys and fcc metals[J].Phys Rev B, 2018, 98(9):094104.
[51] DENG Z, CHEN C, LI X G, et al. An electrostatic spectral neighbor analysis potential for lithium nitride[J]. NPJ Comput Mater, 2019, 5(1):75.
[52] WOOD M A, CUSENTINO M A, WIRTH B D, et al. Data-driven material models for atomistic simulation[J]. Phys Rev B, 2019, 99(18):184305.
[53] ZHANG L, HAN J, WANG H, et al. End-to-end Symmetry Preserving Inter-atomic Potential Energy Model for Finite and Extended Systems[C]//Proceedings of the 32nd International Conference on Neural Information Processing Systems, Montréal, CANADA, 2018:4441–4451.
[54] HAN J, ZHANG L, CAR R, et al. Deep potential:A general representation of a many-body potential energy surface[J]. Commun Comput Phys, 2018, 23(3):629–639.
[55] ZHANG L, HAN J, WANG H, et al. Deep Potential molecular dynamics:a scalable model with the accuracy of quantum mechanics[J]. Phys Rev Lett, 2018, 120(14):143001.
[56] WEN T, ZHANG L, WANG H, et al. Deep potentials for materials science[J]. Mater Futur, 2022, 1(2):022601.
[57] JIA W, WANG H, CHEN M, et al. Pushing the limit of molecular dynamics with ab initio accuracy to 100 million atoms with machine learning[J]. Arxiv, 2020. Doi:10.48550/arXiv.2005.00223.
[58] XIE T, GROSSMAN J C. Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties[J].Phys Rev Lett, 2018, 120(14):145301.
[59] TAKAHASHI K, TAKAHASHI L. Creating machine learning-driven material recipes based on crystal structure[J]. J Phys Chem Lett, 2019,10(2):283–288.
[60] DERINGER V L, CARO M A, CSáNYI G. Machine learning interatomic potentials as emerging tools for materials science[J]. Adv Mater, 2019, 31(46):1902765.
[61] OUYANG R, XIE Y, JIANG D. Global minimization of gold clusters by combining neural network potentials and the basin-hopping method[J]. Nanoscale, 2015, 7(36):14817–14821.
[62] JINDAL S, CHIRIKI S, BULUSU S S. Spherical harmonics based descriptor for neural network potentials:Structure and dynamics of Au147 nanocluster[J]. J Chem Phys, 2017, 146(20):204301.
[63] CHIRIKI S, JINDAL S, BULUSU S S. c-T phase diagram and Landau free energies of(AgAu)55 nanoalloy via neural-network molecular dynamic simulations[J]. J Chem Phys, 2017, 147(15):154303.
[64] CHIRIKI S, JINDAL S, BULUSU S S. Neural network potentials for dynamics and thermodynamics of gold nanoparticles[J]. J Chem Phys,2017, 146(8):084314.
[65] HAJINAZAR S, SANDOVAL E D, CULLO A J, et al. Multitribe evolutionary search for stable Cu–Pd–Ag nanoparticles using neural network models[J]. Phys Chem Chem Phys, 2019, 21(17):8729–8742.
[66] THORN A, ROJAS-NUNEZ J, HAJINAZAR S, et al. Toward ab initio ground states of gold clusters via neural network modeling[J]. J Phys Chem C, 2019, 123(50):30088–30098.
[67] WANG H, ZHANG Y, ZHANG L, et al. Crystal structure prediction of binary alloys via deep potential[J]. Front Chem, 2020, 8:589795.
[68] ZHANG L, LIN D-Y, WANG H, et al. Active learning of uniformly accurate interatomic potentials for materials simulation[J]. Phys Rev Mater, 2019, 3(2):023804.
[69] DERINGER V L, PICKARD C J, CSáNYI G. Data-driven learning of total and local energies in elemental boron[J]. Phys Rev Lett, 2018,120(15):156001.
[70] DERINGER V L, PROSERPIO D M, CSáNYI G, et al. Data-driven learning and prediction of inorganic crystal structures[J]. Faraday Discuss, 2018, 211(0):45–59.
[71] BERNSTEIN N, CSáNYI G, DERINGER V L. De novo exploration and self-guided learning of potential-energy surfaces[J]. Npj Comput Mater, 2019, 5(1):99.
[72] TONG Q, GAO P, LIU H, et al. Combining machine learning potential and structure prediction for accelerated materials design and discovery[J]. J Phys Chem Lett, 2020, 11(20):8710–8720.
[73] PODRYABINKIN E V, TIKHONOV E V, SHAPEEV A V, et al.Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning[J]. Phys Rev B, 2019, 99(6):064114.
[74] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al.Generative Adversarial Networks[C]//Advances in Neural Information Processing Systems 27, Montréal, CANADA, 2014:2672–2680.
[75] KINGMA D P, WELLING M. Auto-encoding variational bayes[J].Arxiv, 2014. Doi:10.48550/arxiv.1312.6114.
[76] REZENDE D J, MOHAMED S. Variational Inference with Normalizing Flows[C]//Proceedings of the 32nd International Conference on Machine Learning, Lille, France, 2015:1530–1538.
[77] HO J, JAIN A, ABBEEL P. Denoising Diffusion Probabilistic Models[C]//Advances in Neural Information Processing Systems 33,Virtual, 2020:6840–6851.
[78] SOHL-DICKSTEIN J, WEISS E A, MAHESWARANATHAN N, et al.Deep Unsupervised Learning using Nonequilibrium Thermodynamics[C]//Proceedings of the 32nd International Conference on International Conference on Machine Learning, Lille,France, 2015:2256–2265.
[79] RYAN K, LENGYEL J, SHATRUK M. Crystal structure prediction via deep learning[J]. J Am Chem Soc, 2018, 140(32):10158–10168.
[80] NOUIRA A, SOKOLOVSKA N, CRIVELLO J-C. CrystalGAN:Learning to Discover Crystallographic Structures with Generative Adversarial Networks[C]//Proceedings of the AAAI 2019 Spring Symposium on Combining Machine Learning with Knowledge Engineering, California, USA, 2018:doi:10.48550/arxiv.1810.11203.
[81] KIM S, NOH J, GU G H, et al. Generative adversarial networks for crystal structure prediction[J]. Acs Central Sci, 2020, 6(8):1412–1420.
[82] ZHAO Y, AL-FAHDI M, HU M, et al. High-throughput discovery of novel cubic crystal materials using deep generative neural networks[J].Adv Sci, 2021, 8(20):2100566.
[83] REN Z, TIAN S I P, NOH J, et al. An invertible crystallographic representation for general inverse design of inorganic crystals with targeted properties[J]. Matter, 2022, 5(1):314–335.
[84] HOFFMANN J, MAESTRATI L, SAWADA Y, et al. Data-driven approach to encoding and decoding 3-D crystal structures[J]. Arxiv:2019, doi:10.48550/arxiv.1909.00949.
[85] NOH J, KIM J, STEIN H S, et al. Inverse design of solid-state materials via a continuous representation[J]. Matter, 2019, 1(5):1370–1384.
[86] KIM B, LEE S, KIM J. Inverse design of porous materials using artificial neural networks[J]. Sci Adv, 2020, 6(1):eaax9324.
[87] COURT C J, YILDIRIM B, JAIN A, et al. 3-D inorganic crystal structure generation and property prediction via representation learning[J]. J Chem Inf Model, 2020, 60(10):4518–4535.
[88] LONG T, FORTUNATO N M, OPAHLE I, et al. Constrained crystals deep convolutional generative adversarial network for the inverse design of crystal structures[J]. NPJ Comput Mater, 2021, 7(1):66.
[89] XIE T, FU X, GANEA O-E, et al. Crystal Diffusion Variational Autoencoder for Periodic Material Generation[C]//The Tenth International Conference on Learning Representations(ICLR 2022),Virtual, 2022. Doi:10.48550/arxiv.2110.06197.
[90] KIM S, NOH J, GU G H, et al. Generative adversarial networks for crystal structure prediction[J]. Acs Central Sci, 2020, 6(8):1412–1420.
[91] ARJOVSKY M, CHINTALA S, BOTTOU L. Wasserstein GAN[J].Arxiv, 2017. Doi:10.48550/arxiv.1701.07875.
[92] GULRAJANI I, AHMED F, ARJOVSKY M, et al. Improved Training of Wasserstein GANs[C]//Advances in Neural Information Processing Systems 30, Los Angeles, USA, 2017:5767–5777.
[93] REN Z, TIAN S I P, NOH J, et al. An invertible crystallographic representation for general inverse design of inorganic crystals with targeted properties[J]. Matter, 2022, 5(1):314–335.
[94] RONNEBERGER O, FISCHER P, BROX T. U-Net:Convolutional Networks for Biomedical Image Segmentation[C]//Medical Image Computing and Computer-Assisted Intervention, Munich, Germany,2015:234–241.
[95] OKTAY O, SCHLEMPER J, FOLGOC L L, et al. Attention U-Net:learning where to look for the pancreas[J]. Arxiv, 2018.Doi:10.48550/arxiv.1804.03999.
[96] SCARSELLI F, GORI M, TSOI A C, et al. The graph neural network model[J]. Ieee T Neural Networ, 2009, 20(1):61–80.
[97] GILMER J, SCHOENHOLZ S S, RILEY P F, et al. Neural Message Passing for Quantum Chemistry[C]//Proceedings of the 34th International Conference on Machine Learning-Volume 70, 2017:1263–1272.
[98] BATTAGLIA P W, HAMRICK J B, BAPST V, et al. How good is my GAN?[J]. Arxiv:,2018. Doi:10.48550/arxiv.1807.09499.
[99] THOMAS N, SMIDT T, KEARNES S, et al. Tensor field networks:Rotation-and translation-equivariant neural networks for 3D point clouds[J]. Arxiv, 2018. Doi:10.48550/arxiv.1802.08219.
[100] FUCHS F B, WORRALL D E, FISCHER V, et al. SE(3)-Transformers:3D Roto-Translation Equivariant Attention Networks[C]//Advances in Neural Information Processing Systems 33, Virtual, 2020:1970–1981.
[101] SATORRAS V G, HOOGEBOOM E, WELLING M. E(n)Equivariant Graph Neural Networks[C]//Proceedings of the 38th International Conference on Machine Learning, Virtual, 2021:9323–9332.
[102] KLIPFEL A, BOURAOUI Z, FRéGIER Y, et al. Equivariant graph neural network for crystalline materials[C]//Proceedings of the 1st International Workshop on Spatio-Temporal Reasoning and Learning,Vienna, Austria, 2022:doi:10.48550/arxiv.2102.09844.
[103] WANG S, GUO X, ZHAO L. Deep Generative Model for Periodic Graphs[J]. Arxiv, 2022. Doi:10.48550/arxiv.2201.11932.
[104] JAIN A, ONG S P, HAUTIER G, et al. Commentary:The materials project:A materials genome approach to accelerating materials innovation[J]. Apl Mater, 2013, 1(1):011002.
[105] YUE L, BIRU G, XINXIN Z, et al. Machine learning assisted materials design and discovery for rechargeable batteries[J]. Energy Storage Materials, 2020, 31:434–450.
[106] SIQI S, ZHANGWEI T, XINXIN Z, et al. Applying data-driven machine learning to studying electrochemical energy storage materials[J]. Energy Storage Sci Technol, 2022, 11(3):739.
[107] YUE L, XINXIN Z, ZHENGWEI Y, et al. Machine learning embedded with materials domain knowledge[J]. J Chin Ceram Soc, 2022, 50(3):863–876.
基本信息:
DOI：10.14062/j.issn.0454-5648.20220835
中图分类号:O469;TP181
引用信息:
[1]罗啸山,王振雨,高朋越,等.机器学习加速理论晶体结构预测研究进展[J].硅酸盐学报,2023,51(02):552-560.DOI:10.14062/j.issn.0454-5648.20220835.
基金信息:
国家自然科学基金(91961204,12034009,11974134,11904129); 吉林省教育厅科学技术研究项目(JJKH20211042KJ)
2023-01-17
2023-01-17
2023-01-17