Predicting The Effective Temperaure of a Glass

TL;DR

This paper demonstrates that for strongly correlating liquids, the effective temperature of a glass depends solely on the final density or temperature jump, allowing predictions from a single equilibrium simulation.

Contribution

It introduces a predictive method for the effective temperature of glasses in strongly correlating liquids based on initial equilibrium data, validated through simulations.

Findings

Effective temperature depends only on final density or temperature jump.

Prediction method works for strongly correlating liquids like Lennard-Jones.

Method does not apply to non-strongly correlating liquids.

Abstract

We explain the findings by Di Leonardo et al. [Phys. Rev. Lett. 84, 6054 (2000)] that the effective temperature of a Lennard-Jones glass depends only on the final value of the density in the volume and/or temperature jump that produces the glass phase. This is not only a property of the Lennard-Jones liquid, but a feature of all strongly correlating liquids. For such liquids data from a single quench simulation provides enough information to predict the effective temperature of any glass produced by jumping from an equilibrium state. This prediction is validated by simulations of the Kob-Andersen binary Lennard-Jones liquid and shown not to apply for the non-strongly correlating monatomic Lennard-Jones Gaussian liquid.