Material Design for Strongly-Correlated Nonlinear Photonics with Advanced Data-Driven Science and Quantum Measurements

2023 Fiscal Year Final Research Report

• Project/Area Number: 19K05303
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 30020:Optical engineering and photon science-related
• Research Institution: University of Fukui
• Principal Investigator: Makino Takayuki 福井大学, 遠赤外領域開発研究センター, 准教授 (70311363)
• Project Period (FY): 2019-04-01 – 2024-03-31
• Keywords: 励起子

Outline of Final Research Achievements

In this study, we investigated the impact of local Coulomb energy, electron-phonon coupling, and hopping integrals on the nonlinear optical properties of quasi-stable rare-earth-based strongly correlated thin films. These materials are attracting attention in various academic fields, including electronic engineering and applied physics. Additionally, we used machine learning to extract underlying correlations behind the nonlinear optical data, contributing to the development of nonlinear materials and advancements in materials informatics. By performing first-principles electronic structure calculations and optimizing local Coulomb energy, we obtained results that reproduce the dielectric function. Furthermore, we achieved a highly sensitive measurement system for nonlinear spectroscopy, and these diverse achievements were published in original research papers.

Free Research Field

光物性物理学

Academic Significance and Societal Importance of the Research Achievements

機械学習・深層学習と電子状態計算を統合することで、大量のデータに依存する従来のアプローチを超え、少量の実験データから外挿的な物性予測が可能になった。これにより、新材料の探索と設計が効率的に行えるようになった。
学習データから外れた領域でも適切に予測ができる新たな深層学習の基礎技術を適用でき、これにより、未知の新物質に対しても物性を高精度予測することが可能となり、高機能な物質や学問的に興味深い性質を持った物質の探索・設計に応用することができた。
従来の深層学習技術の弱点を深層学習と理論の融合で克服することで、材料開発関連する分野で新たな成果を生み出すことが期待できる。