Mechanism of macroscopic equilibration of isolated quantum systems

Research output: Contribution to journalArticlepeer-review

14 Scopus citations


We discuss the approach toward equilibrium of an isolated quantum system. For a wide class of systems, in particular, for chaotic systems, we argue that the expectation value of a local operator averaged over a time interval of length T is bounded by the so-called deviation function, which characterizes maximal deviation from the equilibrium of all states with a given value of energy fluctuations. This result applies to any initial state with a well-defined effective temperature. We provide numerical evidence that the bound is approximately saturated by the initial configurations with spatial inhomogeneities at a macroscopic level. In this way the deviation function establishes an explicit connection between the macroscopically observed timescales associated with the transport and the properties of microscopic matrix elements. The form of the deviation function indicates that among the "slowest" states which saturate the bound there are also those with arbitrarily long equilibration times.

Original languageEnglish
Article number224302
JournalPhysical Review B
Issue number22
StatePublished - Jun 6 2019

Bibliographical note

Funding Information:
I am grateful to Joel Lebowitz and David Huse for helpful discussions. I would also like to thank Tomaz Prosen, Hong Liu, Moshe Rozali, and Luis Pedro Garcia-Pintos for reading the manuscript and helpful comments. I acknowledge the University of Kentucky Center for Computational Sciences for computing time on the Lipscomb High Performance Computing Cluster. This work was supported by a grant of the Russian Science Foundation (Project No. 17-12-01587).

Publisher Copyright:
© 2019 American Physical Society.

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics


Dive into the research topics of 'Mechanism of macroscopic equilibration of isolated quantum systems'. Together they form a unique fingerprint.

Cite this