Title: Can Packing Efficiency Predict Glass-Forming Ability?

Author (Poster): Kai Zhang, Yale University

Abstract:

Glass transition and jamming are often thought to be intimately related. For hard spheres, beyond the glass transition packing fraction φg, the trajectory of the system is no longer ergodic but constrained in one branch of the many metastable trajectories. Further compression of the glass state brings the system to the jamming point φJ, at which the system rests on the inherent structure of the corresponding glass. The glass-forming ability (GFA) is measured by the critical cooling rate Rc (or compresstion rate in the case of hard spheres), below which crystallization cannot be avoided. The lower Rc, the better GFA for slower quenching allows the glassy material to be casted in larger thickness. For bulk metallic glasses (BMGs), it is believed that good GFA is associated with efficient packing of the amorphous structure. The underlying argument is that higher packing fraction at jamming implies a larger free volume at lower densities close to the glass transition. With fast compression molecular dynamics simulations, we produce random close packing of binary and ternary hard sphere mixtures and find that the best packing does not coincide with the best glass-forming compositions. We think the packing efficiency of competing crystals can also play a role. With genetic algorithm, we explore binary hard sphere crystals with optimal packing efficiency, which indicates special glass-forming regimes.