TDVP is extended from matrix product states to arbitrary tree tensor networks, and used to time-evolve Fork Tensor Product States with off-diagonal hybridizations.
Magnetization and entanglement after a geometric quench in the XXZ chain
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
We investigate the dynamics of the XXZ spin chain after a geometric quench, which is realized by connecting two half-chains prepared in their ground states with zero and maximum magnetizations, respectively. The profiles of magnetization after the subsequent time evolution are studied numerically by density-matrix renormalization group methods, and a comparison to the predictions of generalized hydrodynamics yields a very good agreement. We also calculate the profiles of entanglement entropy and propose an ansatz for the noninteracting XX case, based on arguments from conformal field theory. In the general interacting case, the propagation of the entropy front is studied numerically both before and after the reflection from the chain boundaries. Finally, our results for the magnetization fluctuations indicate a leading order proportionality relation to the entanglement entropy.
fields
cond-mat.str-el 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Time Dependent Variational Principle for Tree Tensor Networks
TDVP is extended from matrix product states to arbitrary tree tensor networks, and used to time-evolve Fork Tensor Product States with off-diagonal hybridizations.