Trap Model and Slow Dynamics in Supercooled Liquids

The predictions of a class of phenomenological trap models of supercooled liquids are tested via computer simulation of a model glass-forming liquid. It is found that a model with a Gaussian distribution of trap energies provides a good description of the landscape dynamics, even at temperatures above $T_{c}$, the critical temperature of mode-coupling theory. A scenario is discussed whereby deep traps are composed of collections of inherent structures above $T_{c}$ and single inherent structures below $T_{c}$. Deviations from the simple Gaussian trap picture are quantified and discussed.