Quantum correlations, separability and quantum coherence length in equilibrium many-body systems

TL;DR

This paper investigates quantum correlations in equilibrium many-body systems, introducing correlation functions that reveal a finite quantum coherence length distinct from classical correlation length, with implications for understanding quantum non-locality.

Contribution

It explicitly constructs two-point quantum correlation functions at finite temperature and demonstrates their role in defining a quantum coherence length that remains finite even when classical correlations diverge.

Findings

Quantum correlation functions can be experimentally accessed.

Quantum coherence length remains finite at finite temperature.

Quantum correlations exhibit a length scale distinct from classical correlation length.

Abstract

Non-locality is a fundamental trait of quantum many-body systems, both at the level of pure states, as well as at the level of mixed states. Due to non-locality, mixed states of any two subsystems are correlated in a stronger way than what can be accounted for by considering correlated probabilities of occupying some microstates. In the case of equilibrium mixed states, we explicitly build two-point quantum correlation functions, which capture the specific, superior correlations of quantum systems at finite temperature, and which are directly { accessible to experiments when correlating measurable properties}. When non-vanishing, these correlation functions rule out a precise form of separability of the equilibrium state. In particular, we show numerically that quantum correlation functions generically exhibit a finite \emph{quantum coherence length}, dictating the characteristic distance over which degrees of freedom cannot be considered as separable. This coherence length is completely disconnected from the correlation length of the system -- as it remains finite even when the correlation length of the system diverges at finite temperature -- and it unveils the unique spatial structure of quantum correlations.