Spinel nitride solid solutions: charting properties in the configurational space with explainable machine learning

Lists

Tools

Sanchez-Palencia, P., Hamad, S., Palacios, P., Grau-Crespo, R. ORCID: https://orcid.org/0000-0001-8845-1719 and Butler, K. T. (2022) Spinel nitride solid solutions: charting properties in the configurational space with explainable machine learning. Digital Discovery, 1 (5). pp. 665-678. ISSN 2635-098X

Text - Accepted Version
· Restricted to Repository staff only
2MB

Preview

Text (Open access) - Published Version
· Available under License Creative Commons Attribution.
· Please see our End User Agreement before downloading.
1MB

It is advisable to refer to the publisher's version if you intend to cite from this work. See Guidance on citing.

To link to this item DOI: 10.1039/D2DD00038E

Abstract/Summary

Ab initio prediction of the variation of properties in the configurational space of solid solutions is computationally very demanding. We present an approach to accelerate these predictions via a combination of density functional theory and machine learning, using the cubic spinel nitride GeSn2N4 as a case study, exploring how formation energy and electronic bandgap are affected by configurational variations. Furthermore, we demonstrate the utility of applying explainable machine learning to understand the crystal chemistry origins of the trends that we observe. Different configuration descriptors (Coulomb matrix eigenspectrum, many-body tensor representation, and cluster correlation function vectors) are combined with different models (linear regression, gradient-boosted decision trees, and multi-layer perceptron) to extrapolate the calculation of ab initio properties from a small set of configurations to the full space with thousands of configurations. We discuss the performance of different descriptors and models. SHAP (SHapley Additive exPlanations) analysis of the machine learning models highlights how values of formation energy are dominated by variations in local crystal structure (single polyhedral environments), while values of electronic bandgap are dominated by variations in more extended structural motifs. Finally, we demonstrate the usefulness of this approach by constructing structure-property maps, identifying important configurations of GeSn2N4 with extremal properties, as well as by calculating accurate equilibrium properties using configurational averaging.

Item Type:	Article
Refereed:	Yes
Divisions:	Life Sciences > School of Chemistry, Food and Pharmacy > Department of Chemistry
ID Code:	109245
Uncontrolled Keywords:	spinel, solar cells, cluster expansion, machine learning
Publisher:	Royal Society of Chemistry

Download Statistics

Downloads

Downloads per month over past year

Altmetric

Deposit Details

References

1. Hohenberg P, Kohn W. Inhomogeneous Electron Gas. Phys Rev. 1964;136(3B):864-871. doi:10.1103/PhysRev.136.B864 2. Kohn W, Sham LJ. Self-Consistent Equations Including Exchange and Correlation Effects. Phys Rev. 1965;140(4A):1133-1138. doi:10.1103/PhysRev.140.A1133 3. Midgley SD, Hamad S, Butler KT, Grau-Crespo R. Bandgap Engineering in the Configurational Space of Solid Solutions via Machine Learning: (Mg,Zn)O Case Study. J Phys Chem Lett. 2021;12(21):5163-5168. doi:10.1021/acs.jpclett.1c01031 4. Hart GLW, Mueller T, Toher C, Curtarolo S. Machine learning for alloys. Nat Rev Mater. 2021;6(8):730-755. doi:10.1038/s41578-021-00340-w 5. Yaghoobi M, Alaei M. Machine learning for compositional disorder : A comparison between different descriptors and machine learning frameworks. Comput Mater Sci. 2022;207:111284. doi:10.1016/j.commatsci.2022.111284 6. Askanazi EM, Yadav S, Grinberg I. Prediction of the Curie temperatures of ferroelectric solid solutions using machine learning methods. Comput Mater Sci. 2021;199:110730. doi:10.1016/j.commatsci.2021.110730 7. Pentyala P, Singhania V, Duggineni VK, Deshpande PA. Machine learning-assisted DFT reveals key descriptors governing the vacancy formation energy in Pd-substituted multicomponent ceria. Mol Catal. 2022;522:112190. doi:10.1016/j.mcat.2022.112190 8. Pei Z, Yin J, Hawk JA, Alman DE, Gao MC. Machine-learning informed prediction of high-entropy solid solution formation : Beyond the Hume-Rothery rules. npj Comput Mater. 2020;6:50. doi:10.1038/s41524-020-0308-7 9. Chandran M, Lee SC, Shim J hyeok. Machine learning assisted first-principles calculation of multicomponent solid solutions : estimation of interface energy in Ni-based superalloys. Model Simul Mater Sci Eng. 2018;26:025010. doi:10.1088/1361-651X/aa9f37 10. Morita K, Davies DW, Butler KT, Walsh A. Modeling the dielectric constants of crystals using machine learning. J Chem Phys. 2020;153:024503. doi:10.1063/5.0013136 11. Butler KT, Le MD, Thiyagalingam J, Perring TG. Interpretable, calibrated neural networks for analysis and understanding of inelastic neutron scattering data. J Phys Condens Matter. 2021;33:194006. doi:10.1088/1361-648X/abea1c 12. Mo S Di, Ouyang L, Ching WY, Tanaka I, Koyama Y, Riedel R. Interesting physical properties of the new spinel phase of Si3N4 and C3N4. Phys Rev Lett. 1999;83(24):5046-5049. doi:10.1103/PhysRevLett.83.5046 13. Zerr A, Miehe G, Serghiou G, et al. Synthesis of cubic silicon nitride. Nature. 1999;400:340-342. doi:10.1038/22493 14. Zerr A, Riedel R, Sekine T, Lowther JE, Ching WY, Tanaka I. Recent advances in new hard high-pressure nitrides. Adv Mater. 2006;18(22):2933-2948. doi:10.1002/adma.200501872 15. Boyko TD, Moewes A. The hardness of group 14 spinel nitrides revisited. J Ceram Soc Japan. 2016;124(10):1063-1066. doi:10.2109/jcersj2.16097 16. Hu H, Peslherbe GH. Accurate Mechanical and Electronic Properties of Spinel Nitrides from Density Functional Theory. J Phys Chem C. 2021;125(17):8927-8937. doi:10.1021/acs.jpcc.0c09896 17. Boyko TD, Hunt A, Zerr A, Moewes A. Electronic structure of spinel-type nitride compounds Si3N4, Ge3N4, and Sn3N4 with tunable band gaps: Application to light emitting diodes. Phys Rev Lett. 2013;111(9):097402. doi:10.1103/PhysRevLett.111.097402 18. Caskey CM, Seabold JA, Stevanović V, et al. Semiconducting properties of spinel tin nitride and other IV3N4 polymorphs. J Mater Chem C. 2015;3(6):1389-1396. doi:10.1039/c4tc02528h 19. Qu F, Yuan Y, Yang M. Programmed Synthesis of Sn3N4 Nanoparticles via a Soft Chemistry Approach with Urea: Application for Ethanol Vapor Sensing. Chem Mater. 2017;29(3):969-974. doi:10.1021/acs.chemmater.6b03435 20. Li X, Hector AL, Owen JR, Shah SIU. Evaluation of nanocrystalline Sn3N4 derived from ammonolysis of Sn(NEt2)4 as a negative electrode material for Li-ion and Na-ion batteries. J Mater Chem A. 2016;4(14):5081-5087. doi:10.1039/c5ta08287k 21. Wang J, Chen H, Wei SH, Yin WJ. Materials Design of Solar Cell Absorbers Beyond Perovskites and Conventional Semiconductors via Combining Tetrahedral and Octahedral Coordination. Adv Mater. 2019;31(17):1806593. doi:10.1002/adma.201806593 22. Rosenbrock CW, Gubaev K, Shapeev A V., et al. Machine-learned interatomic potentials for alloys and alloy phase diagrams. npj Comput Mater. 2021;7:24. doi:10.1038/s41524-020-00477-2 23. Kostiuchenko T, Körmann F, Neugebauer J, Shapeev A. Impact of lattice relaxations on phase transitions in a high-entropy alloy studied by machine-learning potentials. npj Comput Mater. 2019;5:55. doi:10.1038/s41524-019-0195-y 24. Batchelor TAA, Pedersen JK, Winther SH, Castelli IE, Jacobsen KW, Rossmeisl J. High-Entropy Alloys as a Discovery Platform for Electrocatalysis. Joule. 2019;3(3):834-845. doi:10.1016/j.joule.2018.12.015 25. Sun S, Tiihonen A, Oviedo F, et al. A data fusion approach to optimize compositional stability of halide perovskites. Matter. 2021;4(4):1305-1322. doi:10.1016/j.matt.2021.01.008 26. Kusne AG, Yu H, Wu C, et al. On-the-fly closed-loop materials discovery via Bayesian active learning. Nat Commun. 2020;11(1):1-11. doi:10.1038/s41467-020-19597-w 27. Sánchez-Palencia P, García G, Conesa JC, Wahnón P, Palacios P. Spinel-Type nitride compounds with improved features as solar cell absorbers. Acta Mater. 2020;197:316-329. doi:10.1016/j.actamat.2020.07.034 28. Hart JN, Allan NL, Claeyssens F. Ternary silicon germanium nitrides: A class of tunable band gap materials. Phys Rev B - Condens Matter Mater Phys. 2011;84(24). doi:10.1103/PhysRevB.84.245209 29. Dudiy S V., Zunger A. Searching for alloy configurations with target physical properties: Impurity design via a genetic algorithm inverse band structure approach. Phys Rev Lett. 2006;97(4):1-4. doi:10.1103/PhysRevLett.97.046401 30. Seminovski Y, Palacios P, Wahnn P, Grau-Crespo R. Band gap control via tuning of inversion degree in CdIn2S4 spinel. Appl Phys Lett. 2012;100:102112. doi:10.1063/1.3692780 31. Roychowdhury S, Ghosh T, Arora R, et al. Enhanced atomic ordering leads to high thermoelectric performance in AgSbTe2. Science (80- ). 2021;371(6530):722-727. doi:10.1126/science.abb3517 32. Nechache R, Harnagea C, Li S, et al. Bandgap tuning of multiferroic oxide solar cells. Nat Photonics. 2014;9:61-67. doi:10.1038/nphoton.2014.255 33. Wang Y, Kavanagh SR, Burgués-Ceballos I, Walsh A, Scanlon DO, Konstantatos G. Cation disorder engineering yields AgBiS2 nanocrystals with enhanced optical absorption for efficient ultrathin solar cells. Nat Photonics. 2022;16:235-241. doi:10.1038/s41566-021-00950-4 34. Lundberg SM, Lee SI. A Unified Approach to Interpreting Model Predictions. In: von Luxburg U, Guyon I, Bengio S, Wallach H, Fergus R, eds. NIPS’17: Proceedings of the 31st International Conference on Neural Information Processing Systems. Curran Associates Inc.; 2017:4768–4777. 35. Kresse G, Furthmüller J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput Mater Sci. 1996;6(1):15-50. doi:10.1016/0927-0256(96)00008-0 36. Kresse G, Furthmüller J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys Rev B. 1996;54(16):169-186. doi:10.1103/PhysRevB.54.11169 37. Blöchl PE. Projector augmented-wave method. Phys Rev B. 1994;50(24):17953-17979. doi:10.1103/PhysRevB.50.17953 38. Perdew JP, Burke K, Ernzerhof M. Generalized gradient approximation made simple. Phys Rev Lett. 1996;77(18):3865-3868. doi:10.1103/PhysRevLett.77.3865 39. Hubbard J. Electron correlations in narrow energy bands. Proc R Soc London Ser A Math Phys Sci. 1963;276(1365):238-257. doi:10.1098/rspa.1963.0204 40. Dudarev S, Botton G. Electron-energy-loss spectra and the structural stability of nickel oxide: An LSDA+U study. Phys Rev B. 1998;57(3):1505-1509. doi:10.1103/PhysRevB.57.1505 41. Heyd J, Scuseria GE, Ernzerhof M. Hybrid functionals based on a screened Coulomb potential. J Chem Phys. 2003;118(18):8207-8215. doi:10.1063/1.1564060 42. Feldbach E, Zerr A, Museur L, et al. Electronic Band Transitions in γ-Ge3N4. Electron Mater Lett. 2021;17(4):315-323. doi:10.1007/s13391-021-00291-y 43. Grau-Crespo R, Hamad S, Catlow CRA, De Leeuw NH. Symmetry-adapted configurational modelling of fractional site occupancy in solids. J Phys Condens Matter. 2007;19(25). doi:10.1088/0953-8984/19/25/256201 44. Schmidt J, Marques MRG, Botti S, Marques MAL. Recent advances and applications of machine learning in solid-state materials science. npj Comput Mater. 2019;5:83. doi:10.1038/s41524-019-0221-0 45. Faber F, Lindmaa A, Von Lilienfeld OA, Armiento R. Crystal structure representations for machine learning models of formation energies. Int J Quantum Chem. 2015;115(16):1094-1101. doi:10.1002/qua.24917 46. Rupp M, Tkatchenko A, Müller KR, Von Lilienfeld OA. Fast and accurate modeling of molecular atomization energies with machine learning. Phys Rev Lett. 2012;108:058301. doi:10.1103/PhysRevLett.108.058301 47. Huo H, Rupp M. Unified Representation of Molecules and Crystals for Machine Learning. Published online 2017. doi:10.48550/arXiv.1704.06439 48. Sanchez JM, Ducastelle F, Gratias D. Generalized cluster description of multicomponent systems. Phys A Stat Mech its Appl. 1984;128(1-2):334-350. doi:10.1016/0378-4371(84)90096-7 49. Troppenz M, Rigamonti S, Draxl C. Predicting Ground-State Configurations and Electronic Properties of the Thermoelectric Clathrates Ba8AlxSi46-x and Sr8AlxSi46-x. Chem Mater. 2017;29(6):2414-2424. doi:10.1021/acs.chemmater.6b05027 50. Tibshirani R. Regression Shrinkage and Selection Via the Lasso. J R Stat Soc Ser B. 1996;58(1):267-288. doi:10.1111/j.2517-6161.1996.tb02080.x 51. Friedman JH. Stochastic gradient boosting. Comput Stat Data Anal. 2002;38(4):367-378. doi:10.1016/S0167-9473(01)00065-2 52. Hornik K, Stinchcombe M, White H. Multilayer feedforward networks are universal approximators. Neural Networks. 1989;2(5):359-366. doi:10.1016/0893-6080(89)90020-8 53. Xu X, Jiang H. Cluster expansion based configurational averaging approach to bandgaps of semiconductor alloys. J Chem Phys. 2019;150(3). doi:10.1063/1.5078399 54. Liu J, Wang X, Borkiewicz OJ, et al. Unified View of the Local Cation-Ordered State in Inverse Spinel Oxides. Inorg Chem. 2019;58(21):14389-14402. doi:10.1021/acs.inorgchem.9b01685 55. Stevanović V, D’Avezac M, Zunger A. Simple point-ion electrostatic model explains the cation distribution in spinel oxides. Phys Rev Lett. 2010;105(7):11-14. doi:10.1103/PhysRevLett.105.075501 56. Seko A, Oba F, Tanaka I. Classification of spinel structures based on first-principles cluster expansion analysis. Phys Rev B. 2010;81:054114. doi:10.1103/PhysRevB.81.054114 57. Seko A, Yuge K, Oba F, Kuwabara A, Tanaka I. Prediction of ground-state structures and order-disorder phase transitions in II-III spinel oxides: A combined cluster-expansion method and first-principles study. Phys Rev B. 2006;73:184117. doi:10.1103/PhysRevB.73.184117 58. Wechsler BA, Navrotsky A. Thermodynamics and structural chemistry of compounds in the system MgOTiO2. J Solid State Chem. 1984;55(2):165-180. doi:10.1016/0022-4596(84)90262-7 59. Santos-Carballal D, Roldan A, Grau-Crespo R, De Leeuw NH. First-principles study of the inversion thermodynamics and electronic structure of FeM2X4 (thio)spinels (M=Cr, Mn, Co, Ni; X= O, S). Phys Rev B. 2015;91:195106. doi:10.1103/PhysRevB.91.195106 60. O’Neill HSC, Navrotsky A. Simple spinels: crystallographic parameters, cation radii, lattice energies, and cation distribution. Am Mineral. 1983;68:181-194. 61. Callen HB, Harrison SE, Kriessman CJ. Cation distributions in ferrospinels. Theoretical. Phys Rev. 1956;103(4):851-856. doi:10.1103/PhysRev.103.851 62. Grau-crespo R, Waghmare U V. Simulation of Crystals with Chemical Disorder at Lattice Sites. In: Rai B, ed. Molecular Modeling for the Design of Novel Performance Chemicals and Materials. CRC Press; 2012:319-342. doi:10.1201/b11590-12 63. Gautam A, Sadowski M, Prinz N, et al. Rapid Crystallization and Kinetic Freezing of Site-Disorder in the Lithium Superionic Argyrodite Li6PS5Br. Chem Mater. 2019;31(24):10178-10185. doi:10.1021/acs.chemmater.9b03852 64. Redfern SA, Harrison RJ, O’Neill HSC, Wood DR. Thermodynamics and kinetics of cation ordering in MgAl2O4 spinel up to 1600 C from in situ neutron diffraction. Am Mineral. 1999;84:299-310. doi:10.2138/am-1999-0313

University Staff: Request a correction | Centaur Editors: Update this record

University of Reading

CentAUR: Central Archive at the University of Reading

Accessibility navigation

Spinel nitride solid solutions: charting properties in the configurational space with explainable machine learning

Abstract/Summary

Downloads

Page navigation

See also

Footer navigation