Abstract

The discrete slip-link theory is a hierarchy of strongly connected models that have great success predicting the linear and nonlinear rheology of high-molecular-weight polymers. Three of the four parameters of the most-detailed model, which can be extracted from primitive-path analysis, give quantitative agreement with experimental data for all examined chemistries (polystyrene, polyisoprene, polybutadiene, and polyethylene). Here, we attempt to extract the remaining friction parameter from atomistic simulations. In particular, an available quantum chemistry-based force field for polyethylene oxide (PEO) was used to perform molecular-dynamics simulations of a 12 kDa melt. The Kuhn friction is obtained from the mean-squared displacement of the center-of-mass of the chains (MSD of COM) in the melt. The result is also corroborated using the relaxation modulus calculated through the Green-Kubo formula. Once the four parameters are determined for any chemistry, all parameters for all members of the slip-link hierarchy are determined. Then, using a coarser member of the hierarchy, the dynamic modulus of a 256 kDa PEO melt was predicted. The predictions are compared to experimental measurements performed at the same temperature. Unfortunately, the extracted friction is about 30% larger than the one observed in the experiment. However, two fundamentally different methods, one utilizing the MSD of COM and the other the relaxation modulus, gave consistent results for the extracted Kuhn friction. Therefore, the discrepancy presumably arises from insufficient accuracy in the force field. Nonetheless, the work demonstrates that theory predictions without adjustable parameters should be possible.