Exploring the Relationship Between Softness and Excess Entropy in Glass-forming Systems

TL;DR

This paper investigates how a machine-learned structural measure called softness relates to excess entropy in supercooled liquids, revealing a strong quantitative link that improves understanding of particle rearrangements in glassy systems.

Contribution

It demonstrates that computing excess entropy over softness-based groupings provides a better prediction of particle rearrangement barriers in supercooled liquids.

Findings

Softness correlates with local excess entropy in supercooled liquids.

Excess entropy computed over softness groups predicts rearrangement barriers effectively.

The approach enhances understanding of glassy dynamics beyond traditional scaling laws.

Abstract

We explore the relationship between a machine-learned structural quantity (softness) and excess entropy in simulations of supercooled liquids. Excess entropy is known to scale well the dynamical properties of liquids, but this quasi-universal scaling is known to breakdown in the supercooled and glassy regimes. Using numerical simulations, we test whether a local form of the excess entropy can lead to predictions that derive from softness, which has been shown to correlate well with the tendency for individual particles to rearrange. To that end, we explore leveraging softness to compute excess entropy in the traditional fashion over softness groupings. Our results show that by computing the excess entropy over softness-binned groupings, we can build a strong quantitative relationship between the rearrangement barriers across the explored systems.