2019 Fiscal Year Research-status Report
Designing Thermal Functional Materials via Materials Informatics
| Project/Area Number |
19K14902
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
鞠 生宏 東京大学, 大学院工学系研究科(工学部), 客員研究員 (30809645)
|
|---|
| Project Period (FY) |
2019-04-01 – 2022-03-31
|
| Keywords | マテリアルズ・インフォマティクス / 伝熱機能材料 |
|---|
| Outline of Annual Research Achievements |
This materials informatics (MI) research work went smoothly in 2019. We have successfully applied MI to design materials for thermal radiation and magnetic tunnel junctions (MTJs). The main achievements are summarized as follows:
1. Through Bayesian optimization, LASSO technique, and the first-principles, the disordered-MgAl2O4 structures were successfully designed which gives the largest tunnel magnetoresistance ratios in MTJs.
2. The thermal photonic structures for radiative cooling applications were successfully designed by a method combining the rigorous coupled wave analysis and Bayesian optimization.
3. The transfer learning technology has been successfully applied to screen high thermal conductivity crystals.
|
| Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
This materials informatics related research project goes smoothly as expected in the proposal. Up to present, we have applied the materials informatics method to design thermal conduction, thermal radiation as well as magnetic properties. Our work has shown great advantage in designing materials or structures with optimal transport properties for different energy carriers including phonons, photons and magnons. The mentioned research works were published in three important international journals as research papers.
|
| Strategy for Future Research Activity |
Following the research proposal, we will push forward the work related with: (1) explore high thermal conductivity crystals via transfer learning, and (2) design novel functional materials using various materials informatics method including the Bayesian optimization, the Monte Carlo tree search, and the quantum annealing.
|
| Causes of Carryover |
We hope to be able to continue to use the amount unused from previous fiscal year.
We plan to used those amount to buy some machine learning or materials related books (around 114,511円), to pay for open access journal paper publication fees (around 300,000円), and to support one international conferences (MRS or APS, around 300,000円).
|
Research Products
(11 results)
