Y. S. Hu

Crystallization and phase separation in blends of high stereoregular poly(lactide) with poly(ethylene glycol)

The effect of cooling rate on crystallization and subsequent aging of high stereoregular poly(lactide) (PLA) blended with poly(ethylene glycol) (PEG) was studied by thermal analysis and by direct observation of the solid state structure with atomic force microscopy (AFM). Blending with PEG accelerated crystallization of PLA. When a PLA/PEG 70/30 (wt/wt) blend was slowly cooled from the melt, PLA crystallized first as large spherulites followed by crystallization of PEG. The extent of PLA crystallization depended on the cooling rate, however, for a given blend composition the PEG crystallinity was proportional to PLA crystallinity. The partially crystallized blend obtained with a cooling rate of 30 °C min−1 consisted of large spherulites dispersed in a homogeneous matrix. The blend was not stable at ambient temperature. With time, epitaxial crystallization of PEG on the edges of the spherulites depleted the surrounding region of PEG, which created a vitrified region surrounding the spherulites. Further from the spherulites, the homogeneous amorphous phase underwent phase separation with formation of a more rigid PLA-rich phase and a less-rigid PEG-rich phase. Decreasing the amount of PEG in the blend decreased the crystallization rate of PLA and increased the nucleation density. The amount of PLA crystallinity did not depend on blend composition, however, PEG crystallinity decreased to the extent that PEG did not crystallize in a PLA/PEG 90/10 (wt/wt) blend.

Correlation of fatigue and creep crack growth in poly(vinyl chloride)

Slow crack growth in PVC pipe was studied in order to develop a methodology for predicting long-term creep fracture from short-term tension-tension fatigue tests. In all cases, the crack propagated continuously through a crack-tip craze. In fatigue, the density of drawn craze fibrils gradually increased with decreased frequency and increased temperature. At the lowest frequency, 0.01 Hz, the fibril density in fatigue approached that in creep. The kinetics of fatigue and creep crack growth followed the conventional Paris law formulations with the same exponent 2.7, da/dt = AfΔKI2.7, da/dt = BKI2.7, respectively. The effects of frequency, temperature and R-ratio (the ratio of minimum to maximum stress intensity factor in the fatigue loading cycle) on the Paris law prefactors were characterized. Comparison of frequency and R-ratio tests revealed that the fatigue contribution depended on strain rate. Therefore, at each temperature, crack growth rate was modeled as the product of a creep contribution that depended only on the maximum stress intensity factor and a fatigue contribution that depended on strain rate: (da/dt) = BKI,max2.7 (1 + Cε), where B is the prefactor in the Paris law for creep and C is a coefficient defining the strain rate sensitivity. A linear correlation allowed extrapolation of the creep prefactor (B) from fatigue data. The extrapolated values were systematically higher than the values measured directly from creep and only converged at Tg. The difference was attributed to damage of the craze fibrils during crack closure upon unloading in the fatigue cycle.

Effect of impact modification on slow crack growth in poly(vinyl chloride)

The effect of impact modification on slow crack growth in a poly(vinyl chloride) (PVC) compound was examined in order to test a methodology for predicting long-term creep fracture from short-term tension-tension fatigue tests. In all cases the crack propagated in a stepwise manner through a crack tip craze zone. Step length was analyzed in terms of the Dugdale model for a crack tip plastic zone. The overall crack growth rate in fatigue and creep followed the conventional Paris power law with the same power 2.7, da/dt = Af ΔKI2.7 and da/dt = BKI2.7,respectively. The effects of frequency, temperature, and R-ratio (the ratio of the minimum to maximum stress intensity factor in the fatigue loading cycle) on the Paris prefactor were determined. Crack growth rate was modeled as the product of a creep contribution that depended only on the maximum stress intensity factor and a fatigue contribution that depended on strain rate da/dt = BfKI,max2.7 (1 + C ε, where C is a coefficient defining the strain rate sensitivity. A linear correlation allowed for extrapolation of the creep prefactor Bf from fatigue data. Impact modification decreased Bf but had no effect on C.