Signature of collective elastic glass physics in surface-induced long-range tails in dynamical gradients

Asieh Ghanekarade, Anh D. Phan, Kenneth S. Schweizer, David S. Simmons

Research output: Contribution to journalArticlepeer-review

Abstract

Understanding the underlying nature of dynamical correlations believed to drive the bulk glass transition is a long-standing problem. Here we show that the form of spatial gradients of the glass transition temperature and structural relaxation time near an interface indeed provide signatures of the nature of relaxation in bulk glass-forming liquids. We report the results of long-time, large-system molecular dynamics simulations of thick glass-forming polymer films with one vapour interface, supported on a dynamically neutral substrate. We find that gradients in the glass transition temperature and logarithm of the structural relaxation time nucleated at a vapour interface exhibit two distinct regimes: a medium-ranged, large-amplitude exponential gradient, followed by a long-range slowly decaying tail that can be described by an inverse power law. This behaviour disagrees with multiple proposed theories of glassy dynamics but is predicted by the ‘elastically collective nonlinear Langevin equation’ theory as a consequence of two coupled mechanisms: a medium-ranged interface-nucleated gradient of surface-modified local caging constraints, and an interfacial truncation of a long-ranged collective elastic field. These findings support a coupled spatially local–nonlocal mechanism of activated glassy relaxation and kinetic vitrification in both the isotropic bulk and in broken-symmetry films.

Original languageEnglish (US)
Pages (from-to)800-806
Number of pages7
JournalNature Physics
Volume19
Issue number6
DOIs
StatePublished - Jun 2023

ASJC Scopus subject areas

  • General Physics and Astronomy

Fingerprint

Dive into the research topics of 'Signature of collective elastic glass physics in surface-induced long-range tails in dynamical gradients'. Together they form a unique fingerprint.

Cite this