High-performance diffusion model for inverse design of high T_c superconductors with effective doping and accurate stoichiometry

Abstract

The pursuit of designing superconductors with high T_c has been a long-standing endeavor. However, the widespread incorporation of doping in high T_c superconductors significantly impacts electronic structure, intricately influencing T_c. The complex interplay between the structural composition and material performance presents a formidable challenge in superconductor design. Based on a novel generative model, diffusion model, and doping adaptive representation: three-channel matrix, we have designed a high T_c superconductors inverse design model called Supercon-Diffusion. It has achieved remarkable success in accurately generating chemical formulas for doped high T_c superconductors. Supercon-Diffusion is capable of generating superconductors that exhibit high T_c and excels at identifying the optimal doping ratios that yield the peak T_c. The doping effectiveness (55%) and electrical neutrality (55%) of the generated doped superconductors exceed those of traditional GAN models by more than tenfold. Density of state calculations on the structures further confirm the validity of the generated superconductors. Additionally, we have proposed 200 potential high T_c superconductors that have not been documented yet. This groundbreaking contribution effectively reduces the search space for high T_c superconductors. Moreover, it successfully establishes a bridge between the interrelated aspects of composition, structure, and property in superconductors, providing a novel solution for designing other doped materials.

Keywords

diffusion model / generative model / high T_c superconductors / inverse design / machine learning

References

[1] a) Larbalestier D, Gurevich A, Feldmann DM, Polyanskii A. High Tc superconducting materials for electric power applications. Nature. 2001;414(6861):368-377. b) Huse DA, Fisher MP, Fisher DS, et al. Are superconductors really superconducting. Nature. 1992;358(6387):553-559.

[2] a) Hosono H, Yamamoto A, Hiramatsu H, Ma Y, et al. High Tc superconducting materials for electric power applications. Mater Today. 2018;21(3):278-302. b) Hao Z, Clem JR, McElfresh M, Civale L, Malozemoff A, Holtzberg F. Model for the reversible magnetization of high-k type-II superconductors: Application to High Tc superconductors. Phys Rev B. 1991;43(4):2844. c) Yao C, Ma Y, et al. Superconducting materials: Challenges and opportunities for large-scale applications. Iscience. 2021;24(1):101956.

[3] a) Aranda MAG. Crystal structures of copper-based high-Tc superconductor. Adv Mater. 1994;6(12):905-921. b) Aswathy P, Anooja J, Sarun P, Syamaprasad U. An overview on iron based superconductors. Supercond Sci Technol. 2010;23(7):073001. c) Dasenbrock-Gammon N, Snider E, McBride R, et al. Evidence of near-ambient superconductivity in a N-doped lutetium hydride. Nature. 2023;615(6500):244-248.

[4] a) Zhou X, Lee W-S, Imada M, et al. High-temperature superconductivity. Nature Rev Phys. 2021;3(9):462-479. b) Flores-Livas JA, Boeri L, Sanna A, Profeta G, Arita R, Eremets M. A perspective on conventional high-temperature superconductors at high pressure: Methods and materials. Phys Rep. 2020;856:1-78. c) Ginzburg VL. High- temperature superconductivity-A dream or reality. Uspekhi Fizicheskikh Nauk. 1976;118(2):315-330.

[5] a) Dai P. Antiferromagnetic order and spin dynamics in iron-based superconductors. Rev Mod Phys. 2015;87(3):855-896. b) Demura S, Mizuguchi Y, Deguchi K, et al. New member of BiS-Based Superconductor NdO_1-xFxBiS₂. J Phys Soc Jpn. 2013;82(3):033708. c) Kuo H-H, Chu J-H, Palmstrom JC, Kivelson SA, Fisher IR. Ubiquitous signatures of nematic quantum criticality in optimally doped Fe-based superconductors. Science. 2016;352(6288):958-962.

[6] a) Li Y, Zhang J, Zhang K, Zhao M, Hu K, Lin X. Large data set-drive machine learning model for accurate prediction of the thermoelectric figure of merit. ACS Appl Mater Interfaces. 2022; 14(48):55517-55526. b) Zhang J, Zhang K, Xu S, et al. An integrated machine learning model for accurate and robust prediction of superconducting critical temperature. J Energy Chem. 2022;64:1-10. c) Zhang J, Zhu Z, Xiang X-D, et al. Machine learning prediction of superconducting critical temperature through the structural descriptor. J Phys Chem C. 2022;126(1):102-109. d) Zhang X, Jin K-H, Mao J, Zhao M, Liu Z, Liu F. Prediction of intrinsic topological superconductivity in Mn-doped GeTe monolayer from first-principles. npj Comput Mater. 2021;7(1):1-7. e) Zhong C, Zhang J, Lu X, et al. Deep generative model for inverse design of High-Temperature superconductor composition with predicted Tc>77K. ACS Appl Mater Interfaces. 2023;15(1):101-108. f) Hutcheon MJ, Shipley AM, Needs RJ. Predicting novel superconducting hydrides using machine learning approaches. Phys Rev B. 2020;101(14):144501.

[7] Xie SR, Quan Y, Hire AC, et al. Machine learning of superconducting critical temperature from Eliashberg theory. npj Comput Mater. 2022;8(1):1-9.

[8] Davies DW, Butler KT, Jackson AJ, et al. Computational screening of all stoichiometric inorganic materials. Chem. 2016;1(4):617-627.

[9] Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks. Commun ACM. 2020;63(11):139-144.

[10] Mao S, Cheng L, Zhao C, Khan FN, Li Q, Fu HY. Inverse design for silicon photonics: from iterative optimization algorithms to deep neural networks. Appl Sci. 2021;11(9):11.

[11] Zhao Y, Al-Fahdi M, Hu M, et al. High-throughput discovery of novel cubic crystal materials using deep generative neural networks. Adv Sci (Weinh). 2021;8(20):e2100566.

[12] Hellenbrandt M. The inorganic crystal structure database (ICSD)—present and future. Crystallogr Rev. 2004;10(1):17-22.

[13] Dan Y, Zhao Y, Li X, Li S, Hu M, Hu J. Generative adversarial networks (GAN) based efficient sampling of chemical composition space for inverse design of inorganic materials. npj Comput Mater. 2020;6(1):1.

[14] Miao H, Fabbris G, Koch RJ, et al. Charge density waves in cuprate superconductors beyond the critical doping. npj Quantum Mater. 2021;6(1):1-7.

[15] a) Wang K, Gou C, Duan Y, Lin Y, Zheng X, Wang F-Y. Generative adversarial networks: introduction and outlook. IEEE/CAA J Autom Sin. 2017;4(4):588-598. b) Saxena D, Cao J. Generative adversarial network challenges, solution and future directions. ACM Comput Surv (CSUR). 2021;54(3):1-36.

[16] Mi L, Shen M, Zhang J. A probe towards understanding gan and vae models. arXiv Preprint arXiv:1812.05676. 2018.

[17] Cao H, Tan C, Gao Z, Chen G, Heng P-A, Li SZ. A survey on generative diffusion model. arXiv Preprint arXiv:2209.02646. 2022.

[18] a) Dhariwal P, Nichol A. Diffusion model beat gans on images synthesis. Adv Neural Inf Process Syst. 2021;34:8780-8791. b) Kingma D, Salimans T, Poole B, Ho J. Variational diffusion models. Adv Neural Inf Process Syst. 2021;34:21696-21708.

[19] Huang EW, Mendl CB, Liu SX, et al. Numerical evidence of fluctuating stripes in the normal state of high-Tc cuprate superconductors. Science. 2017;358(6367):1161-1164.

[20] Canfield PC, Bud'Ko SL. FeAs-based superconductivity: a case study of the effects of transition metal doping on BaFe₂As₂. Annu Rev Condens Matter Phys. 2010;1(1):27-50.

[21] The Inorganic Crystal Structure Database. Accessed January 10, 2023. https://icsd.nist.gov/

[22] Turian J, Ratinov L, Bengio Y. Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics. 2010.

[23] Siddique N, Paheding S, Elkin CP, Devabhaktuni V. U-net and its variants for medical image segmentation: a review of theory and applications. IEEE Access. 2021;9(7):82031-82057.

[24] Cowles MK, Carlin BP. Markov chain Monte Carlo convergence diagnostics: a comparative review. J Am Stat Assoc. 1996;91(434):883-904.

[25] Van der Maaten L, Hinton G. Visualizing Data using t-SNE. J Mach Learn Res. 2008;9(11):2579-2605.

[26] Zeng S, Zhao Y, Li G, Wang R, Wang X, Ni J. Atom table convolutional neural networks for an accurate prediction of compounds properties. npj Comput Mater. 2019;5(1):1.

[27] Helm T, Kartsovnik M, Bartkowiak M, et al. Evolution of the Fermi surface of the electron-doped high-temperature superconductor Nd_2−xCe_xCuO₄ revealed by Shubnikov–de Haas oscillations. Phys Rev Lett. 2009;103(15):157002.

[28] Schmalian J, Baumgärtel G, Bennemann K-H. Doping dependence of local magnetic moments and antiferromagnetism in high-Tc superconductors: asymmetry between electron and hole doping. Solid State Commun. 1993;86(2):119-122.

[29] Goodall REA, Lee AA. Predicting materials properties without crystal structure: deep representation learning from stoichiometry. Nat Commun. 2020;11(1):6280.

[30] Zhao J-C. Combinatorial approaches as effective tools in the study of phase diagrams and composition–structure–property relationships. Prog Mater Sci. 2006;51(5):557-631.

[31] a) Levy O, Chepulskii RV, Hart GL, Curtarolo S. The new face of rhodium alloys: revealing ordered structures from first principles. J Am Chem Soc. 2010;132(2):833-837. b) Quan Y, Pickett WE. Van Hove singularities and spectral smearing in high-temperature superconducting H₃S. Phys Rev B. 2016;93(10):104526. c) Labbé J, Barišić S, Friedel J. Strong-coupling superconductivity in V₃X type of Compounds. Phys Rev Lett. 1967;19(18):1039-1042.

RIGHTS & PERMISSIONS

2024 2024 The Authors. InfoMat published by UESTC and John Wiley & Sons Australia, Ltd.