Regensburg 2025 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
O: Fachverband Oberflächenphysik
O 51: Poster Electronic Structure of Surfaces: Spectroscopy, Surface States
O 51.4: Poster
Tuesday, March 18, 2025, 18:00–20:00, P2
Next-Gen ARPES: AI-controlled beam polarization through Graphene — •Ridha Eddhib1, Balasubramanian Thiagarajan2, and Jan Minar1 — 1New Technologies-Research Centre, University of West Bohemia, 30100 Pilsen, Czech Republic — 2MAX IV Laboratory, Lund University, Lund, 22100, Sweden
Angle-Resolved Photoemission Spectroscopy (ARPES) stands as a cornerstone technique in condensed matter physics, offering deep insights into electronic structures and quantum states. However, the precision of ARPES measurements critically depends on the exact calibration of the photon beam polarization, a challenge that often confronts experimentalists. This study introduces an innovative application of artificial intelligence (AI) to revolutionize ARPES experiments by enabling precise calibration and tuning of photon beam polarization through graphene ARPES cross section, aiming for the ideal of 100% circular polarization. At the heart of our methodology is a neural network model, meticulously trained on datasets generated by the sophisticated one-step model of the SPRKKR [1] (Spin-Polarized Relativistic Korringa-Kohn-Rostoker) code, renowned for its accurate ARPES simulation capabilities. This study leverages the ase2sprkkr package for streamlined interfacing with SPRKKR, enriching our AI model training data with high-fidelity simulations. By implementing this AI-driven methodology, researchers can dynamically adjust their ARPES setups, ensuring that each measurement is conducted under optimal polarization conditions. [1] Ebert, H., Koedderitzsch, D., & Minar, J. (2011). 74(9), 096501.
Keywords: Graphene; ARPES; Polarization; Machine Learning; SPRKKR