Accelerated identification of equilibrium structures of multicomponent inorganic crystals using machine learning potentials
Source abstract
Abstract
Abstract The discovery of multicomponent inorganic compounds can provide direct solutions to scientific and engineering challenges, yet the vast uncharted material space dwarfs synthesis throughput. While the crystal structure prediction (CSP) may mitigate this frustration, the exponential complexity of CSP and expensive density functional theory (DFT) calculations prohibit material exploration at scale. Herein, we introduce SPINNER, a structure-prediction framework based on random and evolutionary searches. Harnessing speed and accuracy of neural network potentials (NNPs), the program navigates configurational spaces 102–103 times faster than DFT-based methods. Furthermore, SPINNER incorporates algorithms tuned for NNPs, achieving performances exceeding conventional algorithms. In blind tests on 60 ternary compositions, SPINNER identifies experimental (or theoretically more stable) phases for ~80% of materials. When benchmarked against data-mining or DFT-based evolutionary predictions, SPINNER identifies more stable phases in many cases. By developing a reliable and fast structure-prediction framework, this work paves the way to large-scale, open exploration of undiscovered inorganic crystals.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Machine Learning in Materials Science · X-ray Diffraction in Crystallography · Electronic and Structural Properties of Oxides
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (84)
Show source-provided references
1. FIZ Karlsruhe. Inorganic Crystal Structure Database (Karlsruhe, 2006).
2. Kamaya, N. et al. A lithium superionic conductor. Nat. Mater. 10, 682–686 (2011).10.1038/nmat3066 ↗
3. Nomura, K. et al. Room-temperature fabrication of transparent flexible thin-film transistors using amorphous oxide semiconductors. Nature 432, 488–492 (2004).10.1038/nature03090 ↗
4. Drozdov, A. P., Eremets, M. I., Troyan, I. A., Ksenofontov, V. & Shylin, S. I. Conventional superconductivity at 203 kelvin at high pressures in the sulfur hydride system. Nature 525, 73–76 (2015).10.1038/nature14964 ↗
5. Kamihara, Y., Watanabe, T., Hirano, M. & Hosono, H. Iron-based layered superconductor La[O1−xFx]FeAs (x = 0.05−0.12) with Tc = 26 K. J. Am. Chem. Soc. 130, 3296–3297 (2008).10.1021/ja800073m ↗
6. Drozdov, A. P. et al. Superconductivity at 250 K in lanthanum hydride under high pressures. Nature 569, 528–531 (2019).10.1038/s41586-019-1201-8 ↗
7. Slavney, A. H., Hu, T., Lindenberg, A. M. & Karunadsa, H. I. A bismuth-halide perovskite with long carrier recombination lifetime for photovoltaic applications. J. Am. Chem. Soc. 138, 2138–2141 (2016).10.1021/jacs.5b13294 ↗
8. Somayazulu, M. et al. Evidence for superconductivity above 260 K in Lanthanum superhydride at megabar pressures. Phys. Rev. Lett. 122, 027001 (2019).10.1103/PhysRevLett.122.027001 ↗
9. Oganov, A. R., Pickard, C. J., Zhu, Q. & Needs, R. J. Structure prediction derives materials discovery. Nat. Rev. Mater. 4, 331–348 (2019).10.1038/s41578-019-0101-8 ↗
10. Van de Krol, R. A faster path to solar water splitting. Matter 3, 1389–1402 (2020).10.1016/j.matt.2020.10.017 ↗
11. Yu, X.-Y. et al. Facet-dependent electrochemical properties of Co3O4 nanocrystals toward heavy metal ions. Sci. Rep. 3, 2886 (2013).10.1038/srep02886 ↗
12. Hou, J. Advanced electrochemical materials in energy conversion and storage (CRC Press, Boca Raton, 2022).10.1201/9781003133971 ↗
13. Sun, W. et al. A map of the inorganic ternary metal nitrides. Nat. Mater. 18, 732–739 (2019).10.1038/s41563-019-0396-2 ↗
14. Nishijima, M. et al. Accelerated discovery of cathode materials with prolonged cycle life for lithium-ion battery. Nat. Commun. 5, 4553 (2014).10.1038/ncomms5553 ↗
15. Hinuma, Y. et al. Discovery of earth-abundant nitride semiconductors by computational screening and high-pressure synthesis. Nat. Commun. 7, 11962 (2016).10.1038/ncomms11962 ↗
16. Gautier, R. et al. Prediction and accelerated laboratory discovery of previously unknown 18-electron ABX compounds. Nat. Chem. 7, 308–316 (2015).10.1038/nchem.2207 ↗
17. Wang, J. et al. Discovery of hexagonal ternary phase Ti2InB2 and its evolution to layered boride TiB. Nat. Commun. 10, 2284 (2019).10.1038/s41467-019-10297-8 ↗
18. Duan, D. et al. Pressure-induced metallization of dense (H2S)2H2 with high-Tc superconductivity. Sci. Rep. 4, 6968 (2014).10.1038/srep06968 ↗
19. Peng, F. et al. Hydrogen clathrate structures in rare earth hydrides at high pressures: possible route to room-temperature superconductivity. Phys. Rev. Lett. 119, 107001 (2017).10.1103/PhysRevLett.119.107001 ↗
20. Liu, H. et al. Dynamics and superconductivity in compressed lanthanum superhydride. Phys. Rev. B 98, 100102(R) (2018).10.1103/PhysRevB.98.100102 ↗
21. Youn, Y. et al. Large-scale computational identification of p-type oxide semiconductors by hierarchical screening. Chem. Mater. 31, 5475–5483 (2019).10.1021/acs.chemmater.9b00816 ↗
22. Kang, Y., Youn, Y., Han, S., Park, J. & Oh, C.-S. Computational screening of indirect gap semiconductors for potential photovoltaic absorbers. Chem. Mater. 31, 4072–4080 (2019).10.1021/acs.chemmater.9b00708 ↗
23. Oganov, A. R. Modern methods of crystal structure prediction (WILEY-VCH, Weinheim, 2011).10.1002/9783527632831 ↗
24. Deringer, V. L., Pickard, C. J. & Csányi, G. Data-driven learning of total and local energies in elemental boron. Phys. Rev. Lett. 120, 156001 (2018).10.1103/PhysRevLett.120.156001 ↗
25. Podryabinkin, E. V., Tikhonov, E. V., Shapeev, A. V. & Oganov, A. R. Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning. Phys. Rev. B 99, 064114 (2019).10.1103/PhysRevB.99.064114 ↗
26. Tong, Q., Xue, L., Lv, J., Wang, Y. & Ma, Y. Accelerating CALYPSO structure prediction by data-driven learning of a potential energy surface. Faraday Discuss 211, 31–43 (2018).10.1039/C8FD00055G ↗
27. Cheng, B., Mazzola, G., Pickard, C. J. & Ceriotti, M. Evidence for supercritical behaviour of high-pressure liquid hydrogen. Nature 585, 217–220 (2020).10.1038/s41586-020-2677-y ↗
28. Tahmasbi, H., Goedecker, S. & Ghasemi, S. A. Large-scale structure prediction of near-stoichiometric magnesium oxide based on a machine-learned interatomic potential: Crystalline phases and oxygen-vacancy ordering. Phys. Rev. Mater. 5, 083806 (2021).10.1103/PhysRevMaterials.5.083806 ↗
29. Khajehpasha, E. R., Goedecker, S. & Ghasemi, S. A. New strontium titanate polymorphs under high pressure. J. Comput. Chem. 42, 699–705 (2021).10.1002/jcc.26490 ↗
30. Rostami, S., Ghasemi, S. A. & Goedecker, S. Novel polymorphs and polytypes of lithium chloride from structure predictions based on charge equilibration via neural network technique. Phys. Rev. Mater. 5, 123603 (2021).10.1103/PhysRevMaterials.5.123603 ↗
31. Eivari, H. A. et al. Two-dimensional hexagonal sheet of TiO2. Chem. Mater. 29, 8594–8603 (2017).10.1021/acs.chemmater.7b02031 ↗
32. Leoffler, T. D. et al. Active learning a neural network model for gold clusters & bulk from sparse first principles training data. ChemCatChem 12, 4796–4806 (2020).10.1002/cctc.202000774 ↗
33. Cao, L. et al. Artificial neural network potential for Au20 clusters based on the first-principles. J. Phys. Condens. Matter 34, 174005 (2022).10.1088/1361-648X/ac4f7d ↗
34. Cao, L.-Z., Wang, P.-J., Sai, L.-W., Fu, J. & Duan, X.-M. Artificial neural network potential for gold clusters. Chin. Phys. B 29, 117304 (2020).10.1088/1674-1056/abc15d ↗
35. Ouyang, R., Xie, Y. & Jiang, D. Global minimization of gold clusters by combining neural network potentials and the basin-hopping method. Nanoscale 7, 14817–14821 (2015).10.1039/C5NR03903G ↗
36. Jindal, S., Chiriki, S. & Bulusu, S. S. Spherical harmonics based descriptor for neural network potentials: Structure and dynamics of Au147 nanocluster. J. Chem. Phys. 146, 204301 (2017).10.1063/1.4983392 ↗
37. Jindal, S. & Bulusu, S. S. Structural evolution in gold nanoparticles using artificial neural network based interatomic potentials. J. Chem. Phys. 152, 154302 (2020).10.1063/1.5142903 ↗
38. Thorn, A., Rojas-Nunez, J., Hajinazar, S., Baltazar, S. & Kolmogrov, A. N. Toward ab initio ground states of gold clusters via neural network modeling. J. Phys. Chem. C. 123, 30088–30098 (2019).10.1021/acs.jpcc.9b08517 ↗
39. Faraji, S., Ghasemi, S. A., Parsaeifard, B. & Goedecker, S. Surface reconstructions and premelting of the (100) CaF2 surface. Phys. Chem. Chem. Phys. 21, 16270–16281 (2019).10.1039/C9CP02213A ↗
40. Hong, C. et al. Training machine-learning potentials for crystal structure prediction using disordered structures. Phys. Rev. B 102, 224104 (2020).10.1103/PhysRevB.102.224104 ↗
41. Behler, J. & Parrinello, M. Generalized neural-network representation of high-dimensional potential-energy surfaces. Phys. Rev. Lett. 98, 146401 (2007).10.1103/PhysRevLett.98.146401 ↗
42. Hautier, G., Fischer, C., Ehrlacher, V., Jain, A. & Ceder, G. Data mined ionic substitutions for the discovery of new compounds. Inorg. Chem. 50, 656–663 (2011).10.1021/ic102031h ↗
43. Glass, C. W., Oganov, A. R. & Hansen, N. USPEX−Evolutionary crystal structure prediction. Comput. Phys. Commun. 175, 713–720 (2006).10.1016/j.cpc.2006.07.020 ↗
44. Wang, Y., Lv, J., Zhu, L. & Ma, Y. CALYPSO: a method for crystal structure prediction. Comput. Phys. Commun. 186, 2063 (2012).10.1016/j.cpc.2012.05.008 ↗
45. Kresse, G. & Furthmüller, J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. 6, 15–50 (1996).10.1016/0927-0256(96)00008-0 ↗
46. Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865 (1996).10.1103/PhysRevLett.77.3865 ↗
47. Su, C. et al. Construction of crystal structure prototype database: methods and applications. J. Phys. Condens. Matter 29, 165901 (2017).10.1088/1361-648X/aa63cd ↗
48. Bushlanov, P. V., Blatov, V. A. & Oganov, A. R. Topology-based crystal structure generator. Comput. Phys. Commun. 236, 1–7 (2019).10.1016/j.cpc.2018.09.016 ↗
49. Zhong, Y. et al. Pauling’s rules guided Monte Carlo search (PAMCARS): a shortcut of predicting inorganic crystal structures. Comput. Phys. Commun. 256, 107486 (2020).10.1016/j.cpc.2020.107486 ↗
50. Amsler, M. & Goedecker, S. Crystal structure prediction using the minima hopping method. J. Chem. Phys. 133, 224104 (2010).10.1063/1.3512900 ↗
51. Novikov, I., Grabowski, B., Kormann, F. & Shapeev, A. Magnetic Moment Tensor Potentials for collinear spin-polarized materials reproduce different magnetic states of bcc Fe. npj Comput. Mater. 8, 13 (2022).10.1038/s41524-022-00696-9 ↗
52. Eckhoff, M. & Behler, J. High-dimensional neural network potentials for magnetic systems using spin-dependent atom-centered symmetry functions. npj Comput. Mater. 7, 170 (2021).10.1038/s41524-021-00636-z ↗
53. Youn, Y. et al. AMP2: a fully automated program for ab initio calculations of crystalline materials. Comput. Phys. Commun. 256, 107450 (2020).10.1016/j.cpc.2020.107450 ↗
54. Kim, S. et al. A band-gap database for semiconducting inorganic materials calculated with hybrid functional. Sci. Data 7, 387 (2020).10.1038/s41597-020-00723-8 ↗
55. Jain, A. et al. The Materials Project: a materials genome approach to accelerating materials innovation. APL Mater. 1, 011002 (2013).10.1063/1.4812323 ↗
56. Yang, J. H., Kitchaev, D. A. & Ceder, G. Rationalizing accurate structure prediction in the meta-GGA SCAN functional. Phys. Rev. B 100, 035132 (2019).10.1103/PhysRevB.100.035132 ↗
57. Sun, J., Ruzsinszky, A. & Perdew, J. P. Strongly constrained and appropriately normed semilocal density functional. Phys. Rev. Lett. 115, 036402 (2015).10.1103/PhysRevLett.115.036402 ↗
58. Zhang, Y. et al. Efficient first-principles prediction of solid stability: towards chemical accuracy. npj Comput. Mater. 4, 9 (2018).10.1038/s41524-018-0065-z ↗
59. Wu, Y., Lazic, P., Hautier, G., Persson, K. & Ceder, G. First principles high throughput screening of oxynitrides for water-splitting photocatalysts. Energy Environ. Sci. 6, 157–168 (2013).10.1039/C2EE23482C ↗
60. Zhu, Y., He, X. & Mo, Y. Strategies based on nitride materials chemistry to stabilize Li metal anode. Adv. Sci. 4, 1600517 (2017).10.1002/advs.201600517 ↗
61. Narayan, A. et al. Computational and experimental investigation for new transition metal selenides and sulfides: the importance of experimental verification for stability. Phys. Rev. B 94, 045105 (2016).10.1103/PhysRevB.94.045105 ↗
62. Miao, N. et al. Computational prediction of boron-based MAX phases and MXene derivatives. Chem. Mater. 32, 6947–6957 (2020).10.1021/acs.chemmater.0c02139 ↗
63. Kvashnin, A. G., Tantardini, C., Zakaryan, H. A., Kvashnina, Y. A. & Oganov, A. R. Computational search for new W−Mo−B compounds. Chem. Mater. 32, 7028–7035 (2020).10.1021/acs.chemmater.0c02440 ↗
64. Davies, D. W. et al. Computer-aided design of metal chalcohalide semiconductors: from chemical composition to crystal structure. Chem. Sci. 9, 1022–1030 (2018).10.1039/C7SC03961A ↗
65. Luo, D., Qiao, X. & Dronskowski, R. Predicting nitrogen-based families of compounds: transition-metal guanidinates TCN3 (T = V,Nb,Ta) and ortho-nitrido carbonates T’2CN4 (T’ = Ti, Zr, Hf). Angew. Chem. Int. Ed. 60, 486–492 (2021).10.1002/anie.202011196 ↗
66. Sun et al. The thermodynamic scale of inorganic crystalline metastability. Sci. Adv. 2, e1600225 (2016).10.1126/sciadv.1600225 ↗
67. Perdew, J. P. et al. Restoring the density-gradient expansion for exchange in solids and surfaces. Phys. Rev. Lett. 100, 136406 (2008).10.1103/PhysRevLett.100.136406 ↗
68. Zhao, X. & Vanderbilt, D. First-principles study of structural, vibrational, and lattice dielectric properties of hafnium oxide. Phys. Rev. B 65, 233106 (2002).10.1103/PhysRevB.65.233106 ↗
69. Hicks, D. et al. AFLOW-XtalFinder: a reliable choice to identify crystalline prototypes. npj Comput. Mater. 7, 30 (2021). As far as we can test, the AFLOW prototype encyclopedia does not contain all prototypes derived from ICSD yet. Therefore, we compare the similarity between identified structures and all the compounds in ICSD having the same composition and Z number using ‘structure comparison’ feature in XtalFinder.10.1038/s41524-020-00483-4 ↗
70. Kallenbach, P., Sröbele, M. & Meyer, H.-J. Missing carbodiimide and oxide carbodiimide of scandium: Sc2(CN2)3 and Sc2O2(CN2). Z. Anorg. Allg. Chem. 646, 1281–1284 (2020).10.1002/zaac.202000212 ↗
71. Havlák, L., Fábry, J., Henriques, M. & Dusek, M. Structure determination of KScS2, RbScS2 and KLnS2 (Ln = Nd, Sm, Tb, Dy, Ho, Er, Tm and Yb) and crystal-chemical discussion. Acta Crtyst C71, 623–630 (2015).
72. Avery, P. & Zurek, E. RandSpg: an open-source program for generating atomistic crystal structures with specific spacegroups. Comput. Phys. Commun. 213, 208–216 (2017).10.1016/j.cpc.2016.12.005 ↗
73. Yoo, D. et al. Atomic energy mapping of neural network potential. Phys. Rev. Mater. 3, 093802 (2019).10.1103/PhysRevMaterials.3.093802 ↗
74. Angelis, F. D., Valentin, C. D., Fantacci, S., Vittadini, A. & Selloni, A. Theoretical studies on anatase and less common TiO2 phases: bulk, surfaces, and nanomaterials. Chem. Rev. 114, 9708–9753 (2014).10.1021/cr500055q ↗
75. Yoo, D., Jung, J., Jeong, W. & Han, S. Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials. npj Comput. Mater. 7, 131 (2021).10.1038/s41524-021-00595-5 ↗
76. Lee, K., Yoo, D., Jeong, W. & Han, S. SIMPLE-NN: an efficient package for training and executing neural-network interatomic potentials. Comput. Phys. Commun. 242, 95–103 (2019).10.1016/j.cpc.2019.04.014 ↗
77. Jovan Jose, K. V., Artrith, N. & Behler, J. Construction of high-dimensional neural network potentials using environment-dependent atom pairs. J. Chem. Phys. 136, 194111 (2012).10.1063/1.4712397 ↗
78. Plimpton, S. Fast parallel algorithms for short-range molecular dynamics. J. Comput. Phys. 117, 1–19 (1995).10.1006/jcph.1995.1039 ↗
79. Parsaeifard, B., De, D. S., Finkler, J. A. & Goedecker, S. Fingerprint-based detection of non-local effects in the electronic structure of a simple single component covalent system. Condens. Mater. 6, 9 (2021).10.3390/condmat6010009 ↗
80. Ko, T. W., Finkler, J. A., Goedecker, S. & Behler, J. A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer. Nat. Commun. 12, 398 (2021).10.1038/s41467-020-20427-2 ↗
81. Oganov, A. R. & Valle, M. How to quantify energy landscapes of solids. J. Chem. Phys. 130, 104504 (2009).10.1063/1.3079326 ↗
82. Lyakhov, A. O., Oganov, A. R., Stokes, H. T. & Zhu, Q. New developments in evolutionary structure prediction algorithm USPEX. Comput. Phys. Commun. 184, 1172–1182 (2013).10.1016/j.cpc.2012.12.009 ↗
83. Kang, S. et al. Accelerated identification of equilibrium structures of multicomponent inorganic crystals using machine learning potentials. figshare https://doi.org/10.6084/m9.figshare.14910495 (2021).10.6084/m9.figshare.14910495 ↗
84. SNUMAT: SNU Material Data Center (Seoul National University, 2021); https://www.snumat.com.