Understanding Stokes-Einstein Relation in Supercooled Liquids using Random Pinning

TL;DR

This paper investigates the breakdown of the Stokes-Einstein relation in supercooled liquids by using random particle pinning and molecular dynamics simulations to analyze cluster structures and their percolation.

Contribution

It introduces a systematic way to control and study the Stokes-Einstein violation via particle pinning and links cluster fractal properties to fractional relation exponents.

Findings

Percolation of slow particle clusters correlates with the onset of Stokes-Einstein breakdown.

Fractal dimensions of clusters relate to fractional Stokes-Einstein exponents.

Random pinning effectively tunes the dynamical heterogeneity in supercooled liquids.

Abstract

Breakdown of Stokes-Einstein relation in supercooled liquids is believed to be one of the hallmarks of glass transition. The phenomena is studied in depth over many years to understand the microscopic mechanism without much success. Recently it was found that violation of Stokes-Einstein relation in supercooled liquids can be tuned very systematically by pinning randomly a set of particles in their equilibrium positions. This observation suggested a possible framework where breakdown of Stokes-Einstein relation in the dynamics of supercooled liquids can be studied with precise control. We have done extensive molecular dynamics simulations to understand this phenomena by analyzing the structure of appropriately defined set of dynamically slow and fast particles clusters. We have shown that the Stokes-Einstein breakdown actually become predominant once the cluster formed by the slow particles percolate the entire system size. Finally we proposed a possible close connection between fractal dimensions of these clusters and the exponents associated with the fractional Stokes-Einstein relation.