Materials Science Simulating the diffusion of small molecules in a polymer melt is a difficult task on account of the multitude of length and time scales that must be accommodated, ranging from the fast motions of individual chain segments and pendant groups to the much slower fluctuations of the polymer backbones. Fritz et al. show that a hierarchical model can be used to calculate the diffusion coefficients and excess chemical potentials (solubilities) for ethylbenzene diffusion in atactic polystyrene melts. Diffusion coefficients were first obtained using molecular dynamics simulations of coarse-grained (CG) models, wherein each CG bead represented 5 to 10 atoms and the interactions between beads were guided by potentials derived from all-atom simulations, thus reducing the computational time by several orders of magnitude. By comparing the CG results with those from high-temperature, all-atom simulations, in which equilibration occurs quickly, the authors could extrapolate time scales to lower temperatures using a Volger-Fulcher relationship. The excess chemical potential was determined by thermodynamic integration of the work needed to perturb the system from state A to state B, thereby coupling the solute-polymer interactions. The authors believe that their methods can be applied for fast, quantitative calculations in the limiting case where penetrating molecules are similar in size to the polymer repeat unit. Soft Matter 5 , 10.1039/b911713j (2009).