Comment on "Aging Effects in a Lennard-Jones Glass"

In a recent letter [PRL 78 (1997) 4581] Kob and Barrat reported results of their molecular dynamics simulation of the non-equilibrium dynamics of a binary Lennard-Jones glass. They propose an aging scenario in this system that is close to a t/t_w scaling behavior. In this comment we present new results and show that a different aging scenario, known under the name activated dynamics scaling [i.e. log((t+t_w)/tau)/log(t_w/tau)], describes the aging regime of this kind of glassy systems much better.