Potential Energy Landscape in Lennard-Jones binary mixture model

TL;DR

This study investigates the potential energy landscape of the Lennard-Jones binary mixture model across a range of temperatures, revealing signatures of dynamical arrest through analysis of configurations, saddles, and vibrational states.

Contribution

It provides a detailed analysis of the energy landscape from liquid to supercooled regimes, highlighting features associated with dynamical arrest in the model.

Findings

Identification of saddle points and minima at various temperatures

Vibrational density of states shows signatures of approaching dynamical arrest

Number of negative eigenvalues correlates with dynamical slowdown

Abstract

The potential energy landscape in the Kob-Andersen Lennard-Jones binary mixture model has been studied carefully from liquid down to the supercooled regime, from T =2 down to T =0.46. One thousand of independent configurations along the time evolution have been examined at each investigated temperature. From the starting configuration we searched the nearest saddle (or quasi-saddle) and minimum of the potential energy. The vibrational densities of states for the starting and the two derived configurations have been evaluated. Besides the number of negative eigenvalues of saddle, other quantities show some signature of the approaching of the dynamical arrest temperature.