Download file (4.45 MB)
Download as zip (4.07 MB)

Collection

Citation

Description:

Stable proton configurations in solid-state materials are a prerequisite for the theoretical microscopic investigation of solid-state proton-conductive materials. However, a large number of initial atomistic configurations should be considered to find stable proton configurations, and relaxation calculations using the density functional theory approach are required for each initial configuration. Consequently, the determination of stable configurations is a difficult and time-consuming task. Furthermore, when the size of the simulation cells or the number of doped atoms increases, the number of initial configurations leads to a combinatorial explosion, rendering the computation infeasible. In this study, black-box optimization was combined with an Ising machine and density functional calculations to perform an efficient search for stable proton configurations. Scandium-doped barium zirconate, a typical high-proton conductive oxide, was selected as the model system. The Ising machine was able to rapidly select the initial atomistic configuration, ultimately leading to stable proton configurations after subsequent relaxation calculations. This optimization strategy should be able to solve various issues related to configuration optimization in solid-state materials, thereby promoting novel scientific discoveries.

Rights:

• Creative Commons BY Attribution 4.0 International
  

Keyword: Machine learning, Ising machine, Proton configuration

Date published: 2025-02-06

Publisher: American Chemical Society (ACS)

Journal:

• The Journal of Physical Chemistry C (ISSN: 19327447) vol. 129 issue. 5 p. 2332-2340

Funding:

• Ministry of Education, Culture, Sports, Science and Technology JPMXP1020230327
• Ministry of Education, Culture, Sports, Science and Technology JPMXP1122683430
• Core Research for Evolutional Science and Technology JPMJCR2234
• Japan Society for the Promotion of Science 21H01008

Manuscript type: Publisher's version (Version of record)

MDR DOI:

First published URL: https://doi.org/10.1021/acs.jpcc.4c07104

Related item:

Other identifier(s):

Contact agent:

Updated at: 2025-05-22 13:18:53 +0900

Published on MDR: 2025-05-22 16:32:07 +0900