Predicting superconducting temperatures with new hierarchical neural network AI model
Shaomeng Xu1,2, Pu Chen2, Mingyang Qin2(), Kui Jin3, X.-D. Xiang2()
1. School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, China 2. Department of Materials Science and Engineering, Southern University of Science and Technology, Shenzhen 518055, China 3. Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China
Superconducting critical temperature is the most attractive material property due to its impact on the applications of electricity transmission, railway transportation, strong magnetic fields for nuclear fusion and medical imaging, quantum computing, etc. The ability to predict its value is a constant pursuit for condensed matter physicists. We developed a new hierarchical neural network (HNN) AI algorithm to resolve the contradiction between the large number of descriptors and the small number of datasets always faced by neural network AI approaches to materials science. With this new HNN-based AI model, a much-increased number of 909 universal descriptors for inorganic compounds, and a dramatically cleaned database for conventional superconductors, we achieved high prediction accuracy with a test R2 score of 95.6%. The newly developed HNN model accurately predicted of 45 new high-entropy alloy superconductors with a mean absolute percent error below 6% compared to the experimental data. This demonstrated a significant potential for predicting other properties of inorganic materials.
Statistical properties of elements (17 × 53 = 901)
Total
909
Tab.2
Fig.2
Descriptors’ name
Abbreviation
Frequency
AP_ave_BCCfermi
MD1
1055
ave_FirstIonizationEnergy
MD2
960
AP_min_GSestBCClatcnt
MD3
944
ave_NValance
MD4
914
AP_min_thermal_conductivity
MD5
902
Tab.3
Fig.3
Fig.4
Fig.5
1
Bardeen J., N. Cooper L., and R. Schrieffer J., Microscopic theory of superconductivity, Phys. Rev. 106(1), 162 (1957) https://doi.org/10.1103/PhysRev.106.162
2
Karakozov A., Maksimov E., and Mashkov S., Effect of the frequency dependence of the electron−phonon interaction spectral function on the thermodynamic properties of superconductors, Zh Eksp Teor Fiz 68, 1937 (1975)
Dolgov O., I. Mazin I., A. Golubov A., Y. Savrasov S., and G. Maksimov E., Critical temperature and enhanced isotope effect in the presence of paramagnons in phonon-mediated superconductors, Phys. Rev. Lett. 95(25), 257003 (2005) https://doi.org/10.1103/PhysRevLett.95.257003
V. Sadovskii M., Limits of Eliashberg theory and bounds for superconducting transition temperature, arXiv: (2021)
7
Wang Z., Chaudhary G., Chen Q., and Levin K., Quantum geometric contributions to the BKT transition: Beyond mean field theory, Phys. Rev. B 102(18), 184504 (2020) https://doi.org/10.1103/PhysRevB.102.184504
8
Chávez I.Salas P.A. Solís M.de Llano M., A Boson−Fermion theory that goes beyond the BCS approximations for superconductors, Physica A 607, 128167 (2022)
R. Xie S., Quan Y., C. Hire A., Deng B., M. DeStefano J., Salinas I., S. Shah U., Fanfarillo L., Lim J., Kim J., R. Stewart G., J. Hamlin J., J. Hirschfeld P., and G. Hennig R., Machine learning of superconducting critical temperature from Eliashberg theory, npj Comput. Mater. 8, 14 (2022) https://doi.org/10.1038/s41524-021-00666-7
11
Konno T., Kurokawa H., Nabeshima F., Sakishita Y., Ogawa R., Hosako I., and Maeda A., Deep learning model for finding new superconductors, Phys. Rev. B 103(1), 014509 (2021) https://doi.org/10.1103/PhysRevB.103.014509
12
Roter B., Ninkovic N., and V. Dordevic S., Clustering superconductors using unsupervised machine learning, Physica C 598, 1354078 (2022) https://doi.org/10.1016/j.physc.2022.1354078
13
Zhong C., Zhang J., Lu X., Zhang K., Liu J., Hu K., Chen J., and Lin X., Deep generative model for inverse design of high-temperature superconductor compositions with predicted Tc> 77 K, ACS Appl. Mater. Interfaces 15(25), 30029 (2023) https://doi.org/10.1021/acsami.3c00593
14
Zhong C., Zhang J., Wang Y., Long Y., Zhu P., Liu J., Hu K., Chen J., and Lin X., High-performance diffusion model for inverse design of high Tc superconductors with effective doping and accurate stoichiometry, InfoMat 6(5), e12519 (2024) https://doi.org/10.1002/inf2.12519
15
O. Owolabi T., O. Akande K., and O. Olatunji S., Estimation of superconducting transition temperature Tc for superconductors of the doped MgB2 system from the crystal lattice parameters using support vector regression, J. Supercond. Nov. Magn. 28(1), 75 (2015) https://doi.org/10.1007/s10948-014-2891-7
16
Choudhary K. and Garrity K., Designing high-Tc superconductors with BCS-inspired screening, density functional theory, and deep-learning, npj Comput. Mater. 8, 244 (2022) https://doi.org/10.1038/s41524-022-00933-1
17
Zhang J., Zhu Z., D. Xiang X., Zhang K., Huang S., Zhong C., J. Qiu H., Hu K., and Lin X., Machine learning prediction of superconducting critical temperature through the structural descriptor, J. Phys. Chem. C 126(20), 8922 (2022) https://doi.org/10.1021/acs.jpcc.2c01904
Stanev V., Oses C., G. Kusne A., Rodriguez E., Paglione J., Curtarolo S., and Takeuchi I., Machine learning modeling of superconducting critical temperature, npj Comput. Mater. 4, 29 (2018) https://doi.org/10.1038/s41524-018-0085-8
20
Hamidieh K., A data-driven statistical model for predicting the critical temperature of a superconductor, Comput. Mater. Sci 154, 346 (2018) https://doi.org/10.1016/j.commatsci.2018.07.052
21
J. García-Nieto P., Garcia-Gonzalo E., P. Paredes-Sanchez J., and of the critical temperature of a superconductor by using the WOA/MARS Prediction, Lasso and Elastic-net machine learning techniques, Neural Comput. Appl. 33(24), 17131 (2021) https://doi.org/10.1007/s00521-021-06304-z
22
Zhang J., Zhang K., Xu S., Li Y., Zhong C., Zhao M., J. Qiu H., Qin M., D. Xiang X., Hu K., and Lin X., An integrated machine learning model for accurate and robust prediction of superconducting critical temperature, J. Energy Chem. 78, 232 (2023) https://doi.org/10.1016/j.jechem.2022.11.047
23
Ward L.Agrawal A.Choudhary A.Wolverton C., A general-purpose machine learning framework for predicting properties of inorganic materials, npj Comput. Mater. 2, 16028 (2016)
24
Materials Information Station, URL: supercon.nims.go.jp/index_en.html (accessed 20th October, 2021)
Ward L., Liu R., Krishna A., I. Hegde V., Agrawal A., Choudhary A., and Wolverton C., Including crystal structure attributes in machine learning models of formation energies via Voronoi tessellations, Phys. Rev. B 96(2), 024104 (2017) https://doi.org/10.1103/PhysRevB.96.024104
27
Miracle D., V. Louzguine-Luzgin D., V. Louzguina-Luzgina L., and Inoue A., An assessment of binary metallic glasses: correlations between structure, glass forming ability and stability, Int. Mater. Rev. 55(4), 218 (2010) https://doi.org/10.1179/095066010X12646898728200
28
R. Lide D., CRC Handbook of Chemistry and Physics, Vol. 85, CRC Press, 2004
29
R. De Boer F., et al.., Cohesion in Metals. Transition Metal Alloys, Netherlands: N. p., 1988
30
H. Holland J., Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence, MIT Press, 1992
31
Michalewicz Z. and Schoenauer M., Evolutionary algorithms for constrained parameter optimization problems, Evol. Comput. 4(1), 1 (1996) https://doi.org/10.1162/evco.1996.4.1.1
32
Gu S., Cheng R., and Jin Y., Feature selection for high-dimensional classification using a competitive swarm optimizer, Soft Comput. 22(3), 811 (2018) https://doi.org/10.1007/s00500-016-2385-6
33
Lecun Y., Bottou L., Bengio Y., and Haffner P., Gradient-based learning applied to document recognition, Proc. IEEE 86(11), 2278 (1998) https://doi.org/10.1109/5.726791
34
Chen T.Guestrin C., XGBoost: A scalable tree boosting system, arXiv: (2016)
35
Kitagawa J., Hamamoto S., and Ishizu N., Cutting edge of high-entropy alloy superconductors from the perspective of materials research, Metals (Basel) 10(8), 1078 (2020) https://doi.org/10.3390/met10081078
