Instabilities, nucleation, and critical behavior in nonequilibrium driven fluids: theory and simulation

TL;DR

This thesis investigates instabilities and phase transitions in driven nonequilibrium fluids, introducing a realistic simulation model and analyzing granular gases through simulations and hydrodynamics to understand their critical behaviors.

Contribution

It presents a novel, realistic model for simulating anisotropic driven fluids and provides comprehensive analysis of clustering and phase separation in driven granular gases.

Findings

Developed a new simulation model for anisotropic fluids.

Validated hydrostatic predictions with molecular dynamics.

Analyzed clustering and symmetry breaking in granular gases.

Abstract

The main subject of this thesis rests on the study ---at different levels of description--- of instabilities in systems which are driven, i.e., maintained far from equilibrium by an external forcing. We focus here on two main classes, namely, driven--diffusive fluids and driven granular gases. A particular driven-diffusive lattice model, prototype for nonequilibrium phase transitions, is investigated. A well-known disadvantage of lattice models is that, when they are compared directly with experiment, often do not account for important features of the corresponding nonequilibrium phase diagram, such as structural, morphological, and even critical properties. Furthermore, theoreticians often tend to consider them as prototypical models for certain behavior, a fact which is in many cases not justified. This is discussed in the first part of this thesis, where we introduce a novel, realistic model for computer simulation of anisotropic fluids. The second class of systems we consider in this thesis concerns driven granular gases. We study clustering, symmetry breaking, and phase separation instabilities in two-dimensional driven granular gases, using both molecular dynamics simulations and granular hydrodynamics. Hydrostatic predictions are tested by comparing with molecular dynamics simulations. We are able to develop an effective Langevin description for close-packed macroparticle, confined by a harmonic potential and driven by a delta-correlated noise.