423181 Molecular Dynamics Melting Simulation of Isotactic Polypropylene Using a Defect-Induced Method

Thursday, November 12, 2015: 4:15 PM
251B (Salt Palace Convention Center)
Qin Chen, Chemical Engineering, Pennsylvania State University, University Park, PA, T.C. Mike Chung, Materials Science and Engineering, Pennsylvania State University, University Park, PA, Eric B. Sirota, Corporate Strategic Research, ExxonMobil Research and Engineering Company, Annandale, NJ and Scott T. Milner, Chemical Engineering, The Pennsylvania State University, University Park, PA

The equilibrium melting temperature Tm is a challenging experimental benchmark for molecular dynamics simulations of polymer melting and crystallization. Tm obtained from MD heating scans of crystalline polymers sometimes exhibits superheating of as much as 100ºC. This superheating has been attributed to the combined effects of periodic boundary conditions and ultra-fast heating rates, both of which inhibit nucleation of the melt. We have developed a simple method to minimize this superheating, without introducing artificial potentials or constructing crystal-melt interfaces. To promote melting, we insert a gap into the periodic crystal structure, thus cleaving the infinite system into finite crystal slabs. The exposed crystal faces can nucleate surface melting, instead of relying on nucleating melt droplets within a periodic crystal. For experimental comparison, we measured the melting temperatures of a series of low molecular weight iPP oligomers. These oligomers were synthesized using a metallocene catalyst and a potent chain transfer agent to produce short chain samples, which were further fractionated by crystallizing from solution. The experimental Tm values agree well with our gapped simulation results. We confirm that melting in the simulations of gapped systems initiates at the free surface; at finite heating rate, the interface first starts to advance into the crystal at Tm, with a velocity proportional to T-Tm. This results in a characteristic quadratic rise in the system energy versus T; when the entire crystal has melted, the energy resumes a linear rise with T indicating a constant heat capacity. We obtain the simulated melting temperature as the onset of the quadratic rise in system energy, which corresponds well to the experimental Tm. The same simulations give the crystal-vacuum interfacial free energy, from the difference in heat of fusion between systems with and without the gap.

Extended Abstract: File Not Uploaded