Impact of size polydispersity on the nature of Lennard-Jones liquids

TL;DR

This study investigates how size polydispersity affects the fundamental properties of Lennard-Jones liquids, demonstrating that their simple correlated behavior persists even at high polydispersity levels and that isomorph theory applies well to these systems.

Contribution

It shows that the Roskilde-simple nature of Lennard-Jones liquids remains intact at high polydispersity and extends isomorph theory to polydisperse systems.

Findings

Roskilde-simple behavior persists at >40% polydispersity

Isomorphs approximate well in polydisperse LJ liquids

Theory of isomorphs extends to multi-component systems

Abstract

Polydisperse fluids are encountered everywhere in biological and industrial processes. These fluids naturally show a rich phenomenology exhibiting fractionation and shifts in critical point and freezing temperatures. Here, we study the impact of size polydispersity on the basic nature of Lennard-Jones (LJ) liquids, which represent most molecular liquids without hydrogen bonds, via two- and three-dimensional molecular dynamics computer simulations. A single-component liquid constituting spherical particles and interacting via the LJ potential is known to exhibit strong correlations between virial and potential energy equilibrium fluctuations at constant volume. This correlation significantly simplifies the physical description of the liquid, and these liquids are now known as Roskilde-simple (RS) liquids. We show that this simple nature of the single-component LJ liquid is preserved even for very high polydispersities (above 40% polydispersity for the studied uniform distribution). We also investigate isomorphs of moderately polydisperse LJ liquids. Isomorphs are curves in the phase diagram of RS liquids along which structure, dynamics, and some thermodynamic quantities are invariant in dimensionless units. We find that isomorphs are a good approximation even for polydisperse LJ liquids. The theory of isomorphs thus extends readily to multi-component systems and can be used to improve even further the understanding of these intriguing systems.