Thermodynamic inference of correlations in nonequilibrium collective dynamics

TL;DR

This paper introduces a thermodynamic method to infer hidden correlations in nonequilibrium collective systems by analyzing apparent violations of the thermodynamic uncertainty relation, enabling experimental insights into interaction strengths.

Contribution

It proposes a novel protocol to bound correlations and the number of interacting currents in nonequilibrium systems using fluctuation relations, with analytical and numerical validation.

Findings

Lower bounds on correlations and number of currents derived from TUR violations

Protocol applicable to experimental systems like molecular motors

Conditions identified for tight bounds in collective dynamics

Abstract

The theory of stochastic thermodynamics has revealed many useful fluctuation relations, with the thermodynamic uncertainty relation (TUR) being a theorem of major interest. When many nonequilibrium currents interact with each other, a naive application of the TUR to an individual current can result in an apparent violation of the TUR bound. Here, we explore how such an apparent violation can be used to put a lower bound on the strength of correlations $C$ as well as the number $N$ of interacting currents in collective dynamics. This lower bound is a combined bound on $C(N-1)$ if only one current is measured, or a bound on $N$ if two currents are measured. Our proposed protocol allows for the inference of hidden correlations in experiment, for example when a team of molecular motors pulls on the same cargo but only one or a subset of them is fluorescently tagged. By solving analytically and numerically several models of many-body nonequilibrium dynamics, we ascertain under which conditions this strategy can be applied and the inferred bound on correlations becomes tight.