The 278th Special CMSM-DXMag joint seminar
Machine Learning-Assisted Synchrotron X-ray Analysis of Heusler Alloys
Dr. Takahiro Yamazaki
Department of Materials Science and Technology, Tokyo University of Science
Date & Time: 15:00 - 16:00, May 22nd (Thu), 2025.
Place: 2nd Conference Room, Main Bldg., Sengen.
Abstract:
Half-metallicity, nearly 100% spin polarization of conduction electrons, is key to future spintronic technologies like high-density magnetic storage and low-power memory. Co-based Heusler alloys, with their theoretical half-metallicity and strong magnetic ordering, are promising candidates. However, achieving room-temperature half-metallicity remains challenging as even minor defects can significantly affect spin polarization. In this seminar, we introduce a high-throughput approach that combines large-scale synchrotron radiation measurement data from compositionally graded thin film with machine learning analysis. By focusing on soft X-ray magnetic circular dichroism (MCD) and photoelectron holography (PEH), our method directly and accurately captures subtle differences in electronic, spin, and atomic configurations. We will compare these advanced techniques with conventional analytical methods, discussing the unique features and potential of both MCD spectral data and PEH images.
(Contact)
Yukiko K. Takahashi,
Director of Research Center for Magnetic and Spintronic Materials
E-mail:
TAKAHASHI.Yukiko[at]nims.go.jp