Biographical Information:
Claudia Draxl is professor at the Humboldt University of Berlin in theoretical condensed matter physics. She studied mathematics and physics at the University of Graz receiving her doctorate in 1987. She finished her habilitation at University of Graz in 1996, and then became a lecturer and an associate professor. She was the deputy director and director of the Institute of Theoretical Physics at the University of Graz. She was a university professor at the University of Leoben and had the chair for Atomistic Modelling and Design of Materials. Since 2011, she is a university professor at the Humboldt University of Berlin and has the chair of theoretical condensed-matter physics.Abstract:
Tackling excitations in complex and defected materials Many-body approaches to electronic excitations have become an indispensable tool for an in-depth understanding of the subtle processes that take place in a wide variety of materials. In particular, many-body perturbation theory (MBPT) allows us to treat the interplay between competing interactions of similar strength and on the same energy scale, which can give rise to exciting phenomena. Elucidating the underlying mechanisms and keeping up with the recent developments in experimental techniques, requires the development of advanced methodologies. In this talk, I will first show for selected oxides that only treating electron-electron interaction, electron-vibrational coupling, electron- hole correlation, and exciton-exciton coupling on the same footing, allows for a quantitative description of excitations in these materials. I will also report on our recent progress in keeping such calculations computationally affordable. However, disordered systems are still beyond the reach of even the fastest computers available. A classical approach to handling alloys and disordered systems is the cluster-expansion (CE) technique [2]. However, CE struggles to describe properties that exhibit a nonlinear dependence on material composition. By looking at CE through the lens of machine learning, we have resolved this severe problem [3]. I will show how this novel method can be used to predict nonlinear properties such as band gaps of materials with disorder and vacancies.
References
[1] W. Aggoune, A. Eljarrat, D. Nabok, K. Irmscher, M. Zupancic, Z. Galazka, M. Albrecht, C.
T. Koch, and C. Draxl, Commun. Mater., 3 (2022) 12.
[2] S. Rigamonti, M. Troppenz, M. Kuban, A. Hübner, and C. Draxl,
https://arxiv.org/abs/2310.18223.
[3] A. Stroth, C. Draxl, and S. Rigamonti, preprint (2024).