Efficient exploration method for novel oxides using robot-assisted collaborative synthesis experiments and synthesis condition recommender system

2022 Fiscal Year Final Research Report

• Project/Area Number: 20H02423
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 26010:Metallic material properties-related
• Research Institution: Kyoto University
• Principal Investigator: HAYASHI Hiroyuki 京都大学, 工学研究科, 助教 (50727419)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 無機化合物探索 / 酸化物合成 / 合成条件推薦システム / 並列合成実験

Outline of Final Research Achievements

In this study, we demonstrated that a large number of synthesis experiments can be carried out using parallel automated experiments with robots, and the resulting synthesis data can be organized into a database and used to predict the synthesis results for completely untested conditions using a recommendation system, which is one of the machine learning methods. We actually discovered two new materials and reported their crystal structures. In addition, we focused on coordination polymers as a precursor synthesis method for rapid synthesis of a large number of oxide compounds, and verified under which conditions simple and complex oxide compounds can be synthesized. We developed a fast precursor preparation method compared to techniques such as sol-gel and complexation polymerization that require heating and drying, and demonstrated the synthesis of complex oxide compounds.

Free Research Field

無機物質合成

Academic Significance and Societal Importance of the Research Achievements

本手法により合成に失敗するであろう合成条件を避けて効率的に探索実験を進めていくことができ、研究開発者の労力やコスト削減だけでなく、資源の浪費を抑えることができるようになる。また、機械学習手法がどのように予測しているかということを解析し、合成条件間の類似性などを定量的に評価できるため、合成研究の理解の深化が期待できる。