Dresden 2020 – wissenschaftliches Programm
Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...
TT 23.4: Vortrag
Dienstag, 17. März 2020, 10:15–10:30, HSZ 204
A memory truncation scheme to investigate long-time dynamics in correlated systems — •Antonio Picano and Martin Eckstein — Friedrich-Alexander-Universität Erlangen-Nürnberg
We present an approach to follow the evolution of many-body systems up to previously unaccessible long time scales. Provided only that the system of interest shows a self-energy that is short-range in time, its time evolution can be determined by solving a simplified version of the full Kadanoff-Baym equations. The computational effort scales only linearly with the number of time-steps, and the computer memory is independent of the propagation time.
We have applied the method to investigate the dynamical phase transitions from antiferromagnetic to paramagnetic states driven by an interaction quench in the fermionic Hubbard model, using the nonequilibrium dynamical mean-field theory. We have observed the presence of two dynamical transition points: one is related to the thermal phase transition, the other is connected to the existence of a transient nonthermal antiferromagnetic order above the thermal critical temperature. The non-thermal order displays a slow decay, which is followed by a faster thermalization process, and thermalization is significantly delayed by the trapping of the system in the nonthermal state.