Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published February 2018 | Submitted + Published
Journal Article Open

A real-time extension of density matrix embedding theory for non-equilibrium electron dynamics

Abstract

We introduce real-time density matrix embedding theory (DMET), a dynamical quantum embedding theory for computing non-equilibrium electron dynamics in strongly correlated systems. As in the previously developed static DMET, real-time DMET partitions the system into an impurity corresponding to the region of interest coupled to the surrounding environment, which is efficiently represented by a quantum bath of the same size as the impurity. In this work, we focus on a simplified single-impurity time-dependent formulation as a first step toward a multi-impurity theory. The equations of motion of the coupled impurity and bath embedding problem are derived using the time-dependent variational principle. The accuracy of real-time DMET is compared to that of time-dependent complete active space self-consistent field (TD-CASSCF) theory and time-dependent Hartree-Fock (TDHF) theory for a variety of quantum quenches in the single impurity Anderson model (SIAM), in which the Hamiltonian is suddenly changed (quenched) to induce a non-equilibrium state. Real-time DMET shows a marked improvement over the mean-field TDHF, converging to the exact answer even in the non-trivial Kondo regime of the SIAM. However, as expected from analogous behavior in static DMET, the constrained structure of the real-time DMET wavefunction leads to a slower convergence with respect to active space size, in the single-impurity formulation, relative to TD-CASSCF. Our initial results suggest that real-time DMET provides a promising framework to simulate non-equilibrium electron dynamics in which strong electron correlation plays an important role, and lays the groundwork for future multi-impurity formulations.

Additional Information

© 2017 Published by AIP Publishing. Received 7 November 2017; accepted 14 January 2018; published online 7 February 2018. This work was supported by the US National Science Foundation, through Nos. CHE-1265277 and CHE-1665333. We thank Miles Stoudenmire for help with ITensor. The PySCF modules used in this work were developed with support from the US National Science Foundation, through No. OAC-1657286.

Attached Files

Published - 1.5012766.pdf

Submitted - 1609.07678.pdf

Files

1.5012766.pdf
Files (2.7 MB)
Name Size Download all
md5:ad75b1edc3bf11d96a432bb15ba29eda
1.8 MB Preview Download
md5:add3b6ffee3a2f890cdcc3c78f42f583
916.8 kB Preview Download

Additional details

Created:
August 21, 2023
Modified:
October 23, 2023