Steady shear flow thermodynamics based on a canonical distribution approach

TL;DR

This paper develops a thermodynamic framework for steady shear flows using a canonical distribution approach, linking shear flow properties to classical mechanics and providing stability conditions supported by molecular dynamics simulations.

Contribution

It introduces a novel non-equilibrium thermodynamics for shear flows based on a canonical distribution, connecting shear rate, energy, and viscosity through a new theoretical approach.

Findings

Derived a thermodynamic first law for shear flow involving Helfand's moment

Established stability conditions linked to correlation functions

Supported theoretical results with molecular dynamics simulations

Abstract

A non-equilibrium steady state thermodynamics to describe shear flows is developed using a canonical distribution approach. We construct a canonical distribution for shear flow based on the energy in the moving frame using the Lagrangian formalism of the classical mechanics. From this distribution we derive the Evans-Hanley shear flow thermodynamics, which is characterized by the first law of thermodynamics $dE = T dS - Q d\gamma$ relating infinitesimal changes in energy $E$, entropy $S$ and shear rate $\gamma$ with kinetic temperature $T$. Our central result is that the coefficient $Q$ is given by Helfand's moment for viscosity. This approach leads to thermodynamic stability conditions for shear flow, one of which is equivalent to the positivity of the correlation function of $Q$. We emphasize the role of the external work required to sustain the steady shear flow in this approach, and show theoretically that the ensemble average of its power $\dot{W}$ must be non-negative. A non-equilibrium entropy, increasing in time, is introduced, so that the amount of heat based on this entropy is equal to the average of $\dot{W}$. Numerical results from non-equilibrium molecular dynamics simulation of two-dimensional many-particle systems with soft-core interactions are presented which support our interpretation.