Macroscopic Length Correlations in Non-Equilibrium Systems and Their Possible Realizatons

Abstract

We consider general systems that start from and/or end in thermodynamic equilibrium while experiencing a finite rate of change of their energy density or other intensive quantities $q$ at intermediate times. We demonstrate that at these times, during which $q$ varies at a finite rate, the associated covariance, the connected pair correlator $G_{ij} = \langle q_{i} q_{j} \rangle - \langle q_{i} \rangle \langle q_{j} \rangle$, between any two (far separated) sites $i$ and $j$ in a macroscopic system may, on average, become finite. Once the global mean $q$ no longer changes, the average of $G_{ij}$ over all site pairs $i$ and $j$ may tend to zero. However, when the equilibration times are significant (e.g., as in a glass that is not in true thermodynamic equilibrium yet in which the energy density (or temperature) reaches a final steady state value), these long range correlations may persist also long after $q$ ceases to change. We explore viable experimental implications of our findings and speculate on their potential realization in glasses (where a prediction of a theory based on the effect that we describe here suggests a universal collapse of the viscosity that agrees with all published viscosity measurements over sixteen decades) and non-Fermi liquids. We discuss effective equilibrium in driven systems and derive uncertainty relation based inequalities that connect the heat capacity to the dynamics in general open thermal systems. These rigorous thermalization inequalities suggest the shortest possible fluctuation times scales in open equilibrated systems at a temperature $T$ are typically "Planckian" (i.e., ${\cal{O}}(\hbar/(k_{B} T))$). We briefly comment on parallels between quantum measurements, unitary quantum evolution, and thermalization and on how Gaussian distributions may generically emerge.