Raman spectroscopy for spinline crystallinity measurements. II. Validation of fundamental fiber-spinning models

The original Doufas-McHugh two-phase microstructural/constitutive model for stress-induced crystallization is expanded to polyolefin systems and validated for its predictive capability of online Raman crystallinity and spinline tension data for two Dow homopolymer polypropylene resins. The material parameters--inputs to the model--are obtained from laboratory-scale material characterization data, that is, oscillatory dynamic shear, rheotens (melt extensional rheology), and differential scanning calorimetry data. The same set of two stress-induced crystallization material/molecular parameters are capable of predicting the crystallinity profiles along the spinline and fiber tension very well overall for a variety of industrial fabrication conditions. The model is capable of predicting the freeze point, which is shown, for the first time, to correlate very well with the measured stick point (i.e., the point in the spinline at which the fiber bundle converts from a solid-like state to a liquid-like state and sticks to a solid object such as a glass rod). The model quantitatively captures the effects of the take-up speed, throughput, and melt flow rate on the crystallization rate of polypropylene due to stress-induced crystallization effects. This validated modeling approach has been used to guide fiber spinning for rapid product development. The original Doufas-McHugh stress-induced crystallization model is shown to be numerically robust for the simulation of steady polypropylene melt spinning over a wide range of processing conditions without issues of discontinuities due to the onset of the two-phase constitutive formulation downstream of the die face, at which crystallization more realistically begins. Because of the capturing of the physics of polypropylene fiber spinning and the very good model predictive power, the approximations of the original Doufas-McHugh model are asserted to be reasonable.