Modeling of Crystallization

To generate fundamental understanding of the molecular mechanisms underlying polymer crystallization from dense melt, Monte Carlo (MC) and molecular dynamics (MD) simulations are developed to study the structure and energetics of the crystal nucleation phenomenon in polyethylene and low molecular weight n-alkane analogues. We used n-octane as a prototype system to demonstrate that both MC and MD are capable of investigating the nucleation events. A realistic force field has been adopted such that the simulated crystal structure and melting point satisfactorily match experimental measurements. Using this force field, the nucleation process was observed by brute-force MD. We observed extended-chain crystal nuclei in n-octane system. A biased MC simulation sampled the formation free energy of the crystal nucleus. Visualization of the critical nuclei in both MD simulation and MC simulation suggest that a cylindrical nucleus model is a better description for the chain molecule system than a spherical nucleus model. By fitting the free energy curve to a cylindrical nucleus model, we calculated the melt-crystal interface free energy of the end surface and the lateral surface of a cylindrical nucleus, independently. We are currently investigating longer chain systems, where one molecule can be part of both a crystal nucleus and the amorphous melt.

Compared to quiescent crystallization, the nucleation density of the crystal phase is known to be enhanced by several orders of magnitude in process flows. The nucleation and crystallization of polymer chains during fiber spinning are preceded by an oriented precursor melt phase. Consequently, to understand the processes of fiber spinning and drawing, a complete description of the characteristics of the oriented polymeric melt is invaluable.

By using Semi Grand Canonical MC simulations we can create equilibriated oriented polymer melts at the real melt densities. We have demonstrated that such systems show ordering behavior within the timescales of MD simulation. The orientation relaxation competes with the relaxation behavior of the polymer chains. Once the orientation work is performed on the system, we observe the system relaxation into ordered and disordered phases by MD simulation. With this methodology, we can characterize orientation-induced nucleation in polyethylene and also observe the critical nucleus in oriented polymer melts. This method can be used to study the role and molecular level behavior of long and short chain molecules in flow-induced crystallization experiments. Ultimately, our molecular model will lead to better understating of orientation-induced crystallization, which is an industrially relevant problem.