Parts | Days | Selection | Search | Updates | Downloads | Help
TT: Fachverband Tiefe Temperaturen
TT 34: Nonequilibrium Quantum Many-Body Systems 2
TT 34.2: Talk
Tuesday, April 2, 2019, 14:15–14:30, H22
Quantum thermalization in isolated ultracold gases — •Marvin Lenk1, Anna Posazhennikova2, Tim Lappe1, and Johann Kroha1 — 1Physikalisches Institut, Universität Bonn, Germany — 2Royal Holloway, University of London, United Kingdom
Quantum thermalization, i.e., how an isolated quantum system can dynamically reach thermal equilibrium behavior, is a long-standing problem of quantum statistics. The eigenstate thermalization hypothesis (ETH) poses that, under certain conditions, the long-time expectation value w.r.t. a typical energy eigenstate is indistinguishable from a microcanonical average. By contrast, thermal behavior is reached generally in a non-integrable quantum many-body system alone due to the vast size of the Hilbert space dimension D. In any realistic experiment, only a small subset of the quantum numbers defining a pure state can be measured, if D is sufficiently large. The Hilbert space spanned by the undetermined quantum numbers is traced over and dynamically forms a grand-canonical bath [Ann. Phys. 530, 1700124 (2018)]. We identify this mechanism in a generic system of N interacting bosons in M single-particle levels by computing numerically exactly the time evolution of the reduced densitiy matrix, the entanglement entropy for the observed subsystem as well as expectation values, fluctuations, thermalization times and the distribution function. The thermalizing quantities are, thus, defined by the measurement itself and not restricted to local observables. For N≈ 25 and M≈ 5, D is large enough for thermalization to occur dynamically. By contrast, ETH requires microcanonical initial conditions implying stationary time dependence.