It is well known that the melt rheology (and the consequent processing stresses) of polymers largely dictate the semi-crystalline morphology that develops in products such as films (blown and cast) wherein the melt is solidified under the influence of an external strain. The concentration of the longest molecules present is also recognized to be critical contributing factor. Various research groups (JA Kornfield, Caltech; BS Hsiao, SUNY) have examined and reported on the critical influence exerted by the presence of even very small amounts of very long molecules on the structure development during controlled flow experiments. Consequently, a framework for flow-induced crystallization of polymers is slowly emerging. However, one molecular architectural feature not considered in past investigations is the influence exerted by the presence of crystallizable defects (short branches) on the longest molecules.

In this study, we have created two polymers/blends that are very similar in terms of their molecular weight distribution, "long" molecule concentration and size, total branch (from 1-hexene comonomer) content, and shear rheological characteristics. However, these polymers differ in terms of the architecture of the longest molecules. While the "long" molecules in one of them is strictly linear (no short branches or crystallizable defects), the "long" molecules in the other contain a few (~ 5 per 1000 backbone carbon atoms) short branches. Oriented cast films were produced using the above polymers by maintaining constant extrusion and film drawdown conditions. Because of the similarity in their shear rheological characteristics and in the processing variables, we expect the stresses experienced by the two polymers during the cast film extrusion to be very similar. However, we have discovered some dramatic differences in the semi-crystalline morphology of the two films that have consequences on their mechanical properties as well. These results indicate that the architecture of the "long" is very important in any flow-induced crystallization considerations. Based on our experimental observations, we will also propose a mechanistic model for flow-induced crystallization.