A multiscale approach to structural relaxation and diffusion in metallic glasses

Metallic glasses are promising materials with unique mechanical and thermal properties, but their atomic-scale dynamics remain challenging to understand. In this work, we develop a unified approach to investigate the glass transition and structural relaxation in CoCrNi,

,

, and

metallic glasses. Molecular dynamics (MD) simulation is employed to analyze the radial distribution function at different temperatures and accurately determine the glass transition temperature. We then combine this temperature with the Elastically Collective Nonlinear Langevin Equation (ECNLE) theory to predict the temperature dependence of the structural relaxation time, ${\tau }_{\alpha }\left(T\right)$. By connecting ${\tau }_{\alpha }\left(T\right)$ to the diffusion constant, the ECNLE predictions of ${\tau }_{\alpha }\left(T\right)$ can be compared with those calculated from MD simulations or estimated based on the diffusion constant. By combining atomistic simulation with force-level statistical mechanics, our multiscale approach offers deeper insights into relaxation dynamics and diffusion across various timescales. The relationship between the glass transition and the liquidus temperature is elucidated. This study enhances understanding of the glassy dynamics and properties in complex amorphous materials.