DPG Phi
Verhandlungen
Verhandlungen
DPG

Berlin 2024 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

DY: Fachverband Dynamik und Statistische Physik

DY 17: Many-body Systems: Equilibration, Chaos, and Localization (joint session DY/TT)

DY 17.9: Vortrag

Dienstag, 19. März 2024, 11:45–12:00, A 151

Floquet-Anderson localization in the Thouless pump and how to avoid itAndrás Grabarits1, 2, Attila Takács1, 3, Ion Cosma Fulga4, 5, and •János K. Asbóth1, 61Dept of Theor. Physics, Budapest University of Technology and Economics — 2Dept of Physics and Materials Science, University of Luxembourg — 3Universite de Lorraine, CNRS, Nancy, France — 4Leibniz Institute for Solid State and Materials Research, IFW Dresden, — 5Wurzburg-Dresden Cluster of Excellence ct.qmat, 01062 Dresden — 6Wigner Research Centre for Physics, Budapest

We investigate numerically how onsite disorder affects conduction in the periodically driven Rice-Mele model, a prototypical realization of the Thouless pump, when run at finite period time T. We find that at any fixed period time and nonzero disorder, increasing the system size L to infinity always leads to a breakdown of the pump by Anderson localization of the Floquet states. In a properly defined thermodynamic limit, where L/T^θ is kept constant, Anderson localization can be avoided, and the charge pumped per cycle has a well-defined value (as long as the disorder is not too strong). The exponent θ is not universal, rather, depends on the disorder strength. Our findings are relevant for practical, experimental realizations of the Thouless pump, for studies investigating the nature of its current-carrying Floquet eigenstates, as well as the mechanism of the full breakdown of the pump, expected if the disorder exceeds a critical value.

Keywords: Topological phases; Thouless pump; Finite-size correction; Anderson localization

100% | Mobil-Ansicht | English Version | Kontakt/Impressum/Datenschutz
DPG-Physik > DPG-Verhandlungen > 2024 > Berlin