| CONDENSED MATTER: STRUCTURAL, MECHANICAL, AND THERMAL PROPERTIES |
Prev
|
|
|
Machine learning-assisted optimization of MTO basis sets |
| Zhiqiang Li(李志强), and Lei Wang(王蕾)† |
| Key Laboratory of Quantum Materials and Devices of Ministry of Education, School of Physics, Southeast University, Nanjing 211189, China |
|
|
|
|
Abstract First-principles calculations based on density functional theory (DFT) have had a significant impact on chemistry, physics, and materials science, enabling in-depth exploration of the structural and electronic properties of a wide variety of materials. Among different implementations of DFT, the plane-wave method is widely used for periodic systems because of its high accuracy. However, this method typically requires a large number of basis functions for large systems, leading to high computational costs. Localized basis sets, such as the muffin-tin orbital (MTO) method, have been introduced to provide a more efficient description of electronic structure with a reduced basis set, albeit at the cost of reduced computational accuracy. In this work, we propose an optimization strategy using machine-learning techniques to automate MTO basis-set parameters, thereby improving the accuracy and efficiency of MTO-based calculations. Default MTO parameter settings primarily focus on lattice structure and give less consideration to element-specific differences. In contrast, our optimized parameters incorporate both structural and elemental information. Based on these converged parameters, we successfully recovered missing bands for CrTe2. For the other three materials — Si, GaAs, and CrI3 — we achieved band improvements of up to 2 eV. Furthermore, the generalization of the machine-learned method is validated by perturbation, strain, and elemental substitution, resulting in improved band structures. Additionally, lattice-constant optimization for GaAs using the converged parameters yields closer agreement with experiment.
|
Received: 07 May 2025
Revised: 19 August 2025
Accepted manuscript online: 23 September 2025
|
|
PACS:
|
63.20.dk
|
(First-principles theory)
|
| |
89.20.Ff
|
(Computer science and technology)
|
|
| Fund: This project was supported by the National Key Research and Development Program of China (Grant Nos. 2023YFA1406600 and 2021YFA1202200). |
Corresponding Authors:
Lei Wang
E-mail: wanglei.icer@seu.edu.cn
|
Cite this article:
Zhiqiang Li(李志强), and Lei Wang(王蕾) Machine learning-assisted optimization of MTO basis sets 2026 Chin. Phys. B 35 016301
|
[1] Hohenberg P and Kohn W 1964 Phys. Rev. 136 B864
[2] Burke K and Wagne L O 2013 Int. J. Quantum Chem. 113 96
[3] Perdew J P, Ruzsinszky A, Tao J, Staroverov V N, Scuseria G E and Csonka G I 2005 J. Chem. Phys. 123 062201
[4] Chachiyo T 2016 J. Chem. Phys. 145 021101
[5] Jitropas U and Hsu C H 2017 Jpn. J. Appl. Phys. 56 070313
[6] Perdew J P, Chevary J A, Vosko S H, Jackson K A, Pederson M R, Singh D J and Fiolhais C 1992 Phys. Rev. B 46 6671
[7] Becke A D 1988 Phys. Rev. A 38 3098
[8] Langreth D C and Mehl M J 1983 Phys. Rev. B 28 1809
[9] Topp W C and Hopfield J J 1973 Phys. Rev. B 7 1295
[10] Kresse G and Furthmüller J 1996 Phys. Rev. B 54 11169
[11] Kresse G and Hafner J 1993 Phys. Rev. B 47 558
[12] Blaha P, Schwarz K, Tran F, Laskowski R, Madsen G K H and Marks L D 2020 J. Chem. Phys. 152 074101
[13] Soler J M, Artacho E, Gale J D, García A, Junquera J, Ordejón P and Sánchez-Portal D 2002 J. Phys.: Condens. Matter 14 2745
[14] Artacho E, Anglada E, Diéguez O, Gale J D, García A, Junquera J, Martin R M, Ordejón P, Pruneda J M, Sánchez-Portal D and Soler J M 2008 J. Phys.: Condens. Matter 20 064208
[15] García A, Papior N, Akhtar A, et al. 2020 J. Chem. Phys. 152 204108
[16] Ozaki T 2003 Phys. Rev. B 67 155108
[17] Ozaki T and Kino H 2004 Phys. Rev. B 69 195113
[18] Ozaki T and Kino H 2005 Phys. Rev. B 72 045121
[19] Chen M H, Guo G C and He L X, 2010 J. Phys. Condens. Matter 22 44550
[20] Li P F, Liu X H, Chen M H, Lin P Z, Ren X G, Lin L, Yang C and He L X 2016 Comput. Mater. Sci. 112 503
[21] Kühne T D, Annuzzi M, Ben M D, et al. 2020 J. Chem. Phys. 152 194103
[22] Andersen O K 2012 2012 Correlated Electrons: From Models to Materials 2nd edn. (Germany: Forschungszentrum Jülich) pp. 55–84
[23] Zurek E, Jepsen O and Andersen O K 2005 arXiv:cond-mat/0504374 [cond-mat.str-el]
[24] Andersen O K 1975 Phys. Rev. B 12 3060
[25] Andersen O K, Jepsen O and Krier G 1994 Lectures on Methods of Electronic Structure Calculations (Singapore: World Scientific Publishing Co) pp. 63–124
[26] Ando T 1991 Phys. Rev. B 44 8017
[27] Xia K, Zwierzycki M, Talanana M, Kelly P J and Bauer G E W 2006 Phys. Rev. B 73 064420
[28] Wesselink R J H, Gupta K, Yuan Z and Kelly P J 2019 Phys. Rev. B 99 144409
[29] Starikov A A, Liu Y, Yuan Z and Kelly P J 2018 Phys. Rev. B 97 214415
[30] Wang L, Min T and Xia K 2021 Phys. Rev. B 103 054204
[31] Nomura M and Shibata M 2024 arXiv:2402.01373 [cs.NE]
[32] Akiba T, Sano S, Yanase T, Ohta T and Koyama M 2019 in Proceedings of the 25th ACM SIGKDD Int. Conf. Knowledge Discovery & Data Mining KDD’19, ACM, New York, NY, USA, pp. 2623–2631
[33] Blöchl P E 1994 Phys. Rev. B 50 17953
[34] Kresse G and Joubert D 1999 Phys. Rev. B 59 1758
[35] Perdew J P, Burke K and Ernzerhof M 1996 Phys. Rev. Lett. 77 3865
[36] Liu Y, Kwon S, de Coster G J, Lake R K and Neupane M R 2022 Phys. Rev. Mater. 6 084004
[37] Mcguire M A, Dixit H, Cooper V R and Sales B C 2014 Chem. Mater. 27 612
[38] von Barth U and Hedin L 1972 J. Phys. C: Solid State Phys. 5 1629
[39] Katsura T and Tange Y 2019 Minerals 9 745
[40] Dufek P, Schwarz K and Blaha P 1993 Phys. Rev. B 48 12672
[41] Drittler B, Weinert M, Zeller R and Dederichs P H 1990 Phys. Rev. B 42 9336
[42] Zhang Y C 2017 J. Chem. Phys. 146 194705 |
| No Suggested Reading articles found! |
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
Altmetric
|
|
blogs
Facebook pages
Wikipedia page
Google+ users
|
Online attention
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
View more on Altmetrics
|
|
|
