(456a) Braskem Award Lecture: Engineering the Crystallization and Semicrystalline State of Polymers through Molecular Modeling

Crystallization is a fundamental thermodynamic phase change. Yet, in polymers crystallization rarely proceeds to completion. The result is a heterogenous material comprising domains of crystalline and noncrystalline order, and the opportunity to influence material performance through the engineering of these domains. For this purpose, the engineer needs quantitative process-structure-property relationships. The challenge lies in identifying the essential structural elements that modulate properties at a molecular level, and the mechanisms by which they form. Questions of this nature have been studied for decades, but remain sources of confusion and debate, largely due to a lack of experimental tools with the spatial and temporal resolution to address them definitively. In their absence, hypotheses and theories have proliferated. Resolution of these questions is essential to the intentional design of polymers for specific applications, and the processes by which they are manufactured. This is as true for today’s sustainable polymers as for the petroleum-based polymers that have dominated the field to date.

Molecular modeling offers an alternative approach with the necessary resolution and accuracy to construct quantitative relationships for the formation and properties of semicrystalline polymers. In this presentation, we survey the state of the art in this field, with an emphasis on results from our own lab. We first identify the semicrystalline phase as a thermodynamic state of matter subject to well-defined constraints and amenable to a statistical mechanical description. This approach leads to relatively simple descriptions for the topological state of bridges, loops and tails that make up the so-called “amorphous” or noncrystalline domain of a semicrystalline material. From this point, a model of the semicrystalline material can be constructed, and “in silico experiments” performed to obtain quantitative measures of material properties.

Next, the process of crystallization is examined using molecular simulations, with the result that kinetics for each step of the phase transformation, nucleation and growth, are characterized thermodynamically. Such relationships enable detailed process modeling of polymer crystallization, while a structural analysis of the models themselves offers insight into the operative mechanisms and permit testing of hypotheses and simplifying assumptions. In this vein, we present results of simulations of homogeneous nucleation, heterogeneous (or surface) nucleation, and nucleation accelerated by flow (flow-enhanced nucleation).

In a final section, we show how models of crystallization derived from atomistic simulations can be combined with coarse-grained models of entanglement dynamics like the slip-link model to describe the complex constitutive behavior of a polymer melt operative during processing, wherein the melt rheology influences crystallization kinetics and the evolution of crystallinity in turn influences the melt rheology, in a dynamical two-way exchange across length and time scales. We close with a look at where we have come, and what challenges may lie ahead.