On the nonlocal viscosity kernel of mixtures

TL;DR

This study uses molecular dynamics simulations to analyze how the nonlocal viscosity kernel varies with mixture composition in different liquids, revealing composition dependence in atomic mixtures and scale-dependent deviations in polymer melts.

Contribution

It provides new insights into the composition dependence of the nonlocal viscosity kernel across different liquid mixtures, including polymer melts.

Findings

Viscosity kernel depends on composition in atomic mixtures across studied wave vectors.

Polymer melt kernel is independent of composition at large wave vectors.

Deviations from ideal mixing vary among different mixtures, with a critical length scale identified for polymer melts.

Abstract

In this report we investigate the multiscale hydrodynamical response of a liquid as a function of mixture composition. This is done via a series of molecular dynamics simulations where the wave vector dependent viscosity kernel is computed for three mixtures each with 7-15 different compositions. We observe that the nonlocal viscosity kernel is dependent on composition for simple atomic mixtures for all the wave vectors studied here, however, for a model polymer melt mixture the kernel is independent of composition for large wave vectors. The deviation from ideal mixing is also studied. Here it is shown that a Lennard-Jones mixture follows the ideal mixing rule surprisingly well for a large range of wave vectors, whereas for both the Kob-Andersen mixture and the polymer melt large deviations are found. Furthermore, for the polymer melt the deviation is wave vector dependent such that there exists a critical length scale at which the ideal mixing goes from under-estimating to over-estimating the viscosity.