Predicting Grain Boundary Thermal Conductivitites from Local Atomic Environments

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K15034
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 26020:Inorganic materials and properties-related
• Research Institution: Osaka University (2021-2022); Japan Fine Ceramics Center (2020)
• Principal Investigator: Fujii Susumu 大阪大学, 大学院工学研究科, 助教 (90826033)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 無機材料 / セラミックス / 結晶粒界 / 熱伝導 / フォノン / 熱電変換材料 / 計算科学 / 機械学習

Outline of Final Research Achievements

Grain boundaries (GBs), that are ubiquitously formed between crystal grains of inorganic compounds, have a decisive effect on a variety of material properties. Here, we aimed to reveal the relationship between GB atomic structure and thermal conductivity and its underlying mechanisms. Systematic GB calculations showed that the dominant factor in GB thermal conductivity is the excess volume near the GBs in ionic MgO and SrTiO3, and the variance in bond angles in covalent Si. Using machine learning to the obtained computational data, a model was constructed to accurately predict GB thermal conductivity from the MgO GB structure. As a result, it was found that a small structural distortion effectively reduces GB thermal conductivity. This is a material design guideline to improve the performance of thermoelectric materials and thermal barrier coatings.

Free Research Field

無機材料科学

Academic Significance and Societal Importance of the Research Achievements

粒界は材料中に普遍的に存在し、様々な材料特性に影響を与える。例えば、粒界は熱伝導度を低下させる作用を持つ。熱伝導度は遮熱性が必要な熱電変換材料（熱を直接電気に変換可能な材料）やタービンエンジン等の遮熱コーティング、放熱性が必要なPC等の電子デバイスなど、熱に関連する機能性材料において極めて重要である。本研究では、材料中の粒界がどのように熱伝導に影響を与えるかを、計算科学と機械学習により初めて定量的に解明した。この知見を活かせば、将来的により高性能な熱機能性材料の作製が可能となる。