HTSC-2025: A benchmark dataset of ambient-pressure high-temperature superconductors for AI-driven critical temperature prediction

doi:10.1088/1674-1056/adf042

中国物理B ›› 2025, Vol. 34 ›› Issue (10): 100301-100301.doi: 10.1088/1674-1056/adf042

HTSC-2025: A benchmark dataset of ambient-pressure high-temperature superconductors for AI-driven critical temperature prediction

Xiao-Qi Han(韩小琪)^1,2, Ze-Feng Gao(高泽峰)^1,2,†, Xin-De Wang(王馨德)^1,2, Zhenfeng Ouyang(欧阳振峰)^1,2, Peng-Jie Guo(郭朋杰)^1,2, and Zhong-Yi Lu(卢仲毅)^1,2,3

1 School of Physics and Beijing Key Laboratory of Opto-electronic Functional Materials & Micro-nano Devices, Renmin University of China, Beijing 100872, China;
2 Key Laboratory of Quantum State Construction and Manipulation (Ministry of Education), Renmin University of China, Beijing 100872, China;
3 Hefei National Laboratory, Hefei 230088, China

收稿日期:2025-06-10 修回日期:2025-07-14 接受日期:2025-07-16 发布日期:2025-09-29
通讯作者: Ze-Feng Gao E-mail:zfgao@ruc.edu.cn
基金资助:
Project supported by the National Natural Science Foundation of China (Grant Nos. 62476278, 12434009, and 12204533), the National Key R&D Program of China (Grant No. 2024YFA1408601), and the Innovation Program for Quantum Science and Technology (Grant No. 2021ZD0302402).

HTSC-2025: A benchmark dataset of ambient-pressure high-temperature superconductors for AI-driven critical temperature prediction

Xiao-Qi Han(韩小琪)^1,2, Ze-Feng Gao(高泽峰)^1,2,†, Xin-De Wang(王馨德)^1,2, Zhenfeng Ouyang(欧阳振峰)^1,2, Peng-Jie Guo(郭朋杰)^1,2, and Zhong-Yi Lu(卢仲毅)^1,2,3

1 School of Physics and Beijing Key Laboratory of Opto-electronic Functional Materials & Micro-nano Devices, Renmin University of China, Beijing 100872, China;
2 Key Laboratory of Quantum State Construction and Manipulation (Ministry of Education), Renmin University of China, Beijing 100872, China;
3 Hefei National Laboratory, Hefei 230088, China

Received:2025-06-10 Revised:2025-07-14 Accepted:2025-07-16 Published:2025-09-29
Contact: Ze-Feng Gao E-mail:zfgao@ruc.edu.cn
Supported by:
Project supported by the National Natural Science Foundation of China (Grant Nos. 62476278, 12434009, and 12204533), the National Key R&D Program of China (Grant No. 2024YFA1408601), and the Innovation Program for Quantum Science and Technology (Grant No. 2021ZD0302402).

摘要/Abstract

摘要： The discovery of high-temperature superconducting materials holds great significance for human industry and daily life. In recent years, research on predicting superconducting transition temperatures using artificial intelligence (AI) has gained popularity, with most of these tools claiming to achieve remarkable accuracy. However, the lack of widely accepted benchmark datasets in this field has severely hindered fair comparisons between different AI algorithms and impeded further advancement of these methods. In this work, we present HTSC-2025, an ambient-pressure high-temperature superconducting benchmark dataset. This comprehensive compilation encompasses theoretically predicted superconducting materials discovered by theoretical physicists from 2023 to 2025 based on BCS superconductivity theory, including the renowned $X_2Y$H$_6$ system, perovskite $MX$H$_3$ system, $M_3X$H$_8$ system, cage-like BCN-doped metal atomic systems derived from LaH$_{10}$ structural evolution, and two-dimensional honeycomb-structured systems evolving from MgB$_2$. In addition, we note a range of approaches inspired by physical intuition for designing high-temperature superconductors, such as hole doping, the introduction of light elements to form strong covalent bonds, and the tuning of spin-orbit coupling. The dataset presented in this paper is openly available at ScienceDB. The HTSC-2025 benchmark has been open-sourced on Hugging Face at https://huggingface.co/datasets/xiao-qi/HTSC-2025 and will be continuously updated, while the Electronic Laboratory for Material Science platform is available at https://in.iphy.ac.cn/eln/link.html#/124/V2s4.

关键词: benchmark, superconductors, artificial intelligence

Abstract: The discovery of high-temperature superconducting materials holds great significance for human industry and daily life. In recent years, research on predicting superconducting transition temperatures using artificial intelligence (AI) has gained popularity, with most of these tools claiming to achieve remarkable accuracy. However, the lack of widely accepted benchmark datasets in this field has severely hindered fair comparisons between different AI algorithms and impeded further advancement of these methods. In this work, we present HTSC-2025, an ambient-pressure high-temperature superconducting benchmark dataset. This comprehensive compilation encompasses theoretically predicted superconducting materials discovered by theoretical physicists from 2023 to 2025 based on BCS superconductivity theory, including the renowned $X_2Y$H$_6$ system, perovskite $MX$H$_3$ system, $M_3X$H$_8$ system, cage-like BCN-doped metal atomic systems derived from LaH$_{10}$ structural evolution, and two-dimensional honeycomb-structured systems evolving from MgB$_2$. In addition, we note a range of approaches inspired by physical intuition for designing high-temperature superconductors, such as hole doping, the introduction of light elements to form strong covalent bonds, and the tuning of spin-orbit coupling. The dataset presented in this paper is openly available at ScienceDB. The HTSC-2025 benchmark has been open-sourced on Hugging Face at https://huggingface.co/datasets/xiao-qi/HTSC-2025 and will be continuously updated, while the Electronic Laboratory for Material Science platform is available at https://in.iphy.ac.cn/eln/link.html#/124/V2s4.

Key words: benchmark, superconductors, artificial intelligence

中图分类号: (Tunneling, Josephson effect, Bose-Einstein condensates in periodic potentials, solitons, vortices, and topological excitations)

03.75.Lm

07.05.Mh (Neural networks, fuzzy logic, artificial intelligence) 84.35.+i (Neural networks) 05.70.Jk (Critical point phenomena)

Xiao-Qi Han(韩小琪), Ze-Feng Gao(高泽峰), Xin-De Wang(王馨德), Zhenfeng Ouyang(欧阳振峰), Peng-Jie Guo(郭朋杰), and Zhong-Yi Lu(卢仲毅). HTSC-2025: A benchmark dataset of ambient-pressure high-temperature superconductors for AI-driven critical temperature prediction[J]. 中国物理B, 2025, 34(10): 100301-100301.

参考文献

[1] Onnes H K 1911 Commun. Theor. Phys. 120
[2] arXiv https://arxiv.org/
[3] Coombs T A, Wang Q, Shah A, Hu J, Hao L, Patel I, Wei H, Wu Y, Coombs T and Wang W 2024 Nat. Rev. Electr. Eng. 1 788
[4] Yao C and Ma Y 2021 iScience 24 102541
[5] Pustogow A, Luo Y, Chronister A, et al. 2019 Nature 574 72
[6] Bruzzone P, Fietz W H, Minervini J V, Novikov M, Yanagi N, Zhai Y and Zheng J 2018 Nucl. Fusion 58 103001
[7] Han X Q,Wang X D, Xu M Y, Feng Z, Yao BW, Guo P J, Gao Z F and Lu Z Y 2025 Chin. Phys. Lett. 42 027403
[8] Choudhary K and Garrity K 2022 NPJ Comput. Mater. 8 244
[9] Choudhary K and DeCost B 2021 NPJ Comput. Mater. 7 185
[10] Cerqueira T F T, Fang YW, Errea I, Sanna A and MarquesMA L 2024 Adv. Funct. Mater. 34 2404043
[11] Han X Q, Ouyang Z F, Guo P J, Sun H, Gao Z F and Lu Z Y 2024 Chin. Phys. Lett. 42 047301
[12] Ouyang Z F, Yao B W, Han X Q, Guo P J, Gao Z F and Lu Z Y 2025 Phys. Rev. B 111 L140501
[13] Gibson J B, Hire A C, Dee P M, Barrera O, Geisler B, Hirschfeld P J and Hennig R G 2025 npj Comput. Mater. 11 7
[14] Li J, Fang W Q, Jin S J, Zhang T D, Wu Y L, Xu X D, Liu Y and Yao D X 2025 AI Sci. 1 015001
[15] Han X Q, Guo P J, Gao Z F, Sun H and Lu Z Y 2025 arXiv:2505.09203 [cond-mat.mtrl-sci]
[16] Abramson J, Adler J and Dunger J 2024 Nature 630 493
[17] Dolui K, Conway L J, Heil C, Strobel T A, Prasankumar R P and Pickard C J 2024 Phys. Rev. Lett. 132 166001
[18] Li B, Fan Y X, Zhai J J, Cao Z S, Zhu C, Cheng J, Liu S L and Shi Z X 2025 J. Phys. Chem. C 129 7098
[19] Liu S M, Shi J J, He Y, Tian C, Zhu Y H, Wang X Q and Zhong H X 2024 Adv. Funct. Mater. 34 2315386
[20] Tian C, He Y, Zhu Y H, Du J, Liu S M, Guo W H, Zhong H X, Lu J, Wang X Q and Shi J J 2024 Adv. Funct. Mater. 34 2304919
[21] He Y, Lu J, Wang X Q and Shi J J 2023 Phys. Rev. B 108 054515
[22] Cerqueira T F T, Sanna A and Marques M A L 2024 Adv. Mater. 36 2307085
[23] Wines D and Choudhary K 2024 Mater. Futures 3 025602
[24] Li B, Zhu C, Zhai J J, Yin C H, Fan Y X, Cheng J, Liu S L and Shi Z X 2024 Phys. Rev. B 110 214504
[25] Quan H, Shi X B, Han Y L, Zhang P and Wang B T 2025 Phys. Rev. B 111 134509
[26] Wang H, Tse J S, Tanaka K, Iitaka T and Ma Y M 2012 Proc. Natl. Acad. Sci. USA 109 6463
[27] Ma L, Wang K, Xie Y, Yang X, Wang Y Y, Zhou M, Liu H Y, Yu X H, Zhao Y S, Wang H B, Liu G T and Ma Y M 2022 Phys. Rev. Lett. 128 167001
[28] Li Y W, Hao J, Liu H Y, Tse J S, Wang Y C and Ma Y M 2015 Sci. Rep. 5 9948
[29] Troyan I A, Semenok D V, Kvashnin A G, Sadakov A V, Sobolevskiy O A, Pudalov V M and Oganov A R 2021 Adv. Mater. 33 2006832
[30] Chen W H, Semenok D V, Huang X L, Shu H Y, Li X, Duan D F, Cui T and Oganov A R 2021 Phys. Rev. Lett. 127 117001
[31] Somayazulu M, Ahart M, Mishra A K, Geballe Z M, Baldini M, Meng Y, Struzhkin V V and Hemley R J 2019 Phys. Rev. Lett. 122 027001
[32] Drozdov A P, Kong P P, Minkov V S, Besedin S P, Kuzovnikov M A, Mozaffari S, Balicas L, Balakirev F F, Graf D E, Prakapenka V B, Greenberg E, Knyazev D A, Tkacz M and Eremets M I 2019 Nature 569 528
[33] Li B, Fan Y X, Yin C H, Zhai J J, Zhu C, Cao Z S, Cheng J and Liu S L 2025 Phys. Rev. B 111 014510
[34] Ding H B, Feng Y J, Jiang M J, Tian H L, Zhong G H, Yang C L, Chen X J and Lin H Q 2022 Phys. Rev. B 106 104508
[35] Li X, Yong X, Wu M, Lu S Y, Liu H Y, Meng S, Tse J S and Li Y W 2019 J. Phys. Chem. Lett. 10 2554
[36] Duan Q Z, Zhan L H, Shen J Y, Zhong X and Lu C 2024 Phys. Rev. B 109 054505
[37] Geng N, Hilleke K P, Zhu L,Wang X Y, Strobel T A and Zurek E 2023 J. Am. Chem. Soc. 145 1696
[38] Tomassetti C R, Kafle G P, Marcial E T, Margine E R and Kolmogorov A N 2024 J. Mater. Chem. C 12 4870
[39] Chen C, Miao J Y, Zuo J N, He S and Lu C 2024 Phys. Rev. B 110 174514
[40] Jiang Q W, Chen L, Ma H, Li C D, Duan D F and Cui T 2025 Inorg. Chem. 64 400
[41] Di Cataldo S and Boeri L 2023 Phys. Rev. B 107 L060501
[42] Han Y L, Jiang K Y, Wang B T, Zhang P and Lu H Y 2025 Phys. Rev. B 111 094520
[43] Tomassetti C R, Gochitashvili D, Renskers C, Margine E R and Kolmogorov A N 2024 Phys. Rev. Mater. 8 114801
[44] Liu H D, Fu X P, Fu Z G, Lu H Y and Zhang P 2025 Phys. Rev. B 111 184502
[45] Nagamatsu J, Nakagawa N, Muranaka T, Zenitani Y and Akimitsu J 2001 Nature 410 63
[46] Singh S, Romero A H, Mella J D, Eremeev V, Muñoz E, Alexandrova A N, Rabe K M, Vanderbilt D and Muñoz F 2022 npj Quantum Mater. 7 37
[47] Jiang J J, Xue Y M, Zha L L, Yao S W, Wang B, Hu W J, Peng L, Shi T T, Chen J, Liu X L and Lin J 2025 J. Mater. Chem. C 13 9799
[48] Liu W Y, Wang X J, Li Y Q, Wei Y H, Zhong M M and Kuang M Q 2025 Supercond. Sci. Technol. 38 045024
[49] Jiang B, Luo X S, Sun Y, Zhong X, Lv J, Xie Y, Ma Y M and Liu H Y 2025 Phys. Rev. B 111 054505
[50] Wang X Y, Zhang C Q, Wang Z Y, Liu H Y, Lv J, Wang H, E W N and Ma Y M 2025 arXiv:2502.16558 [cond-mat.supr-con]
[51] Wang V, Xu N, Liu J C, Tang G and Geng W T 2021 Comput. Phys. Commun. 267 108033
[52] Quantum ESPRESSO https://www.quantum-espresso.org/
[53] Perdew J P, Burke K and Ernzerhof M 1996 Phys. Rev. Lett. 77 3865
[54] Hamann D R 2013 Phys. Rev. B 88 085117
[55] Methfessel M and Paxton A T 1989 Phys. Rev. B 40 3616
[56] Baroni S, de Gironcoli S, Dal Corso A and Giannozzi P 2001 Rev. Mod. Phys. 73 515

HTSC-2025: A benchmark dataset of ambient-pressure high-temperature superconductors for AI-driven critical temperature prediction

HTSC-2025: A benchmark dataset of ambient-pressure high-temperature superconductors for AI-driven critical temperature prediction

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	Shi Huang(黄石) and Xi Luo(罗熙). Topological superconductors with spin-triplet pairings and Majorana Fermi arcs[J]. 中国物理B, 2024, 33(8): 87301-087301.
[2]	Yubo Yang(杨玉波), Jizhe Zhao(赵吉哲), Yinjie Liu(刘胤洁), Xiayang Hua(华夏扬), Tianrui Wang(王天睿), Jiyuan Zheng(郑纪元), Zhibiao Hao(郝智彪), Bing Xiong(熊兵), Changzheng Sun(孙长征), Yanjun Han(韩彦军), Jian Wang(王健), Hongtao Li(李洪涛), Lai Wang(汪莱), and Yi Luo(罗毅). Advances in neuromorphic computing: Expanding horizons for AI development through novel artificial neurons and in-sensor computing[J]. 中国物理B, 2024, 33(3): 30702-030702.
[3]	Yue Zhao(赵玥), Sihan Liu(刘思涵), Jiao Liu(刘骄), Tingting Gu(顾婷婷), Jian Hao(郝健), Jingming Shi(石景明), Wenwen Cui(崔文文), and Yinwei Li(李印威). Design of superconducting compounds at lower pressure via intercalating XH₄ molecules (X= B, C, and N) into fcc lattices[J]. 中国物理B, 2024, 33(12): 127101-127101.
[4]	Stephen J. Pennycook, Ryo Ishikawa, Haijun Wu(武海军), Xiaoxu Zhao(赵晓续), Changjian Li(黎长建), Duane Loh, Jiadong Dan, and Wu Zhou(周武). Physics through the microscope[J]. 中国物理B, 2024, 33(11): 116801-116801.
[5]	Yuan-Fang Yue(岳远放), Zhong-Bing Huang(黄忠兵), Huan Li(黎欢),Xing Ming(明星), and Xiao-Jun Zheng(郑晓军). Hole density dependent magnetic structure and anisotropy in Fe-pnictide superconductor[J]. 中国物理B, 2023, 32(9): 97403-097403.
[6]	Conghao Lin(林丛豪), Chuanshuai Huang(黄传帅), and Xiancong Lu(卢仙聪). Customizing topological phases in the twisted bilayer superconductors with even-parity pairings[J]. 中国物理B, 2023, 32(8): 87401-087401.
[7]	Di Liu(刘迪), Xingyu Wang(王兴玉), Zezhong Li(李泽众), Xiaoyan Ma(马肖燕), and Shiliang Li(李世亮). Focused-ion-beam assisted technique for achieving high pressure by uniaxial-pressure devices[J]. 中国物理B, 2023, 32(4): 47401-047401.
[8]	Zi-Yi Chen(陈子逸), Fan-Kai Xie(谢帆恺), Meng Wan(万萌), Yang Yuan(袁扬), Miao Liu(刘淼), Zong-Guo Wang(王宗国), Sheng Meng(孟胜), and Yan-Gang Wang(王彦棡). MatChat: A large language model and application service platform for materials science[J]. 中国物理B, 2023, 32(11): 118104-118104.
[9]	Chunyu Zhang(张春雨), Kai Wang(王凯), Ran Si(司然), Jinqing Li(李金晴), Changxian Song(宋昌仙), Sijie Wu(吴思捷), Bishuang Yan(严碧霜), and Chongyang Chen(陈重阳). Benchmarking calculations of excitation energies and transition properties with spectroscopic accuracy of highly charged ions used for the fusion plasma and astrophysical plasma[J]. 中国物理B, 2023, 32(11): 113102-113102.
[10]	Chen-Yi Huang(黄晨猗) and Shi-Bin Zhang(张仕斌). A backdoor attack against quantum neural networks with limited information[J]. 中国物理B, 2023, 32(10): 100306-100306.
[11]	Geng Li(李更), Shiyu Zhu(朱诗雨), Peng Fan(范朋), Lu Cao(曹路), and Hong-Jun Gao(高鸿钧). Exploring Majorana zero modes in iron-based superconductors[J]. 中国物理B, 2022, 31(8): 80301-080301.
[12]	Hong-Lin Zhou(周宏霖), Yu-Hao Zhang(张与豪), Yang Li(李阳), Shi-Liang Li(李世亮), Wen-Shan Hong(洪文山), and Hui-Qian Luo(罗会仟). Growth and characterization of superconducting Ca_1-xNa_xFe₂As₂ single crystals by NaAs-flux method[J]. 中国物理B, 2022, 31(11): 117401-117401.
[13]	Zhen-Zhen Huang(黄真真), Xiong-Tao Peng(彭雄涛), Wan-Sheng Wang(王万胜), and Jin-Hua Sun(孙金华). Kondo screening cloud in a superconductor with mixed s-wave and p-wave pairing states[J]. 中国物理B, 2022, 31(10): 107101-107101.
[14]	Yu Liu(刘宇), Xiao-Fan Gou(苟晓凡), and Feng Xue(薛峰). Barrier or easy-flow channel: The role of grain boundary acting on vortex motion in type-II superconductors[J]. 中国物理B, 2021, 30(9): 97402-097402.
[15]	Zhaohui Cheng(程朝晖), Bin Lei(雷彬), Xigang Luo(罗习刚), Jianjun Ying(应剑俊), Zhenyu Wang(王震宇), Tao Wu(吴涛), and Xianhui Chen(陈仙辉). Revealing the A_1g-type strain effect on superconductivity and nematicity in FeSe thin flake[J]. 中国物理B, 2021, 30(9): 97403-097403.