AI and the World's Hard Problems · Physics · arXiv · 2026-10-19
Automated end-station at the soft X-ray beamline featuring robotic air-to-ultrahigh-vacuum sample transfer, scanning probe spectroscopy, and machine learning analysis
The authors developed an automated end-station at the soft X-ray beamline BL08U of the NanoTerasu synchrotron that can run without on-site staff. Two robots transfer samples from air into an ultrahigh-vacuum measurement stage, and spectra are acquired automatically by scanning a micro-focused beam. Machine-learning methods such as PCA and UMAP analyse the data, which are stored in a database for remote examination by users, so that beamline staff can devote time to other work.