[3C01-05] Data-Driven Surface Science 

9:30 AM - 10:00 AM JST(12:30 AM - 1:00 AM UTC)

[3C02【依頼講演】] Robotics- and machine-learning-driven materials exploration with the PIONEER system at NanoTerasu BL08U

*Keita Hiromori¹, Masashige Miyamoto¹, Masafumi Horio¹, Masahito Niibe¹, Takeru Nakashima², Yasunobu Ando², Masato Kotsugi³, Hirokazu Fukidome⁴, Takahiro Kondo⁵, Iwao Matsuda¹ (1. The Institute for Solid State Physics, The University of Tokyo (Japan), 2. Institute of Science Tokyo (Japan), 3. Tokyo University of Science (Japan), 4. Tohoku University (Jersey), 5. University of Tsukuba (Japan))

We developed the PIONEER (Process Informatics Operando Novel Experiment for Electronic Research) system at NanoTerasu BL08U to automate soft X-ray absorption spectroscopy experiments. The system integrates atmospheric and vacuum robotic arms, automated valve control, beamline control, and a scanning microscopic XAS chamber, enabling unattended spatially resolved measurements. As a demonstration, autonomous scanning XAS was performed on boron-based materials including h-BN, c-BN, and hydrogen boride sheets. The obtained large-scale spectra were analyzed by UMAP and k-means clustering to visualize and classify electronic-state variations and chemical domains without expert bias. This robotics- and machine-learning-driven workflow improves throughput, reproducibility, and spatial characterization, providing a first step toward autonomous materials exploration at soft X-ray beamlines.

Password required to view