Viscoelastic model hierarchy for fiber melt spinning of semi-crystalline polymers

TL;DR

This paper develops a hierarchy of viscoelastic fiber models for semi-crystalline polymers in melt spinning, balancing accuracy and computational efficiency, and introduces a novel stress-averaged model that performs well in low crystallization regimes.

Contribution

It presents a new stress-averaged fiber model that simplifies computations while maintaining accuracy, extending existing models for better process simulation.

Findings

The stress-averaged model is computationally efficient.

The models effectively capture crystallization effects.

Simulation results validate the model hierarchy.

Abstract

In the fiber melt spinning of semi-crystalline polymers, the degree of crystallization can be non-homogeneous over the cross-section of the fiber, affecting the properties of the end product. For simulation-based process design, the question arises as to which fiber quantities and hence model equations must be resolved in radial direction to capture all practically relevant effects and at the same time imply a model that can be computed with reasonable effort. In this paper, we present a hierarchy of viscoelastic two-phase fiber models ranging from a complex, fully resolved and highly expensive three-dimensional description to a cross-sectionally averaged, cheap-to-evaluate one-dimensional model. In particular, we propose a novel stress-averaged one-two-dimensional fiber model, which circumvents additional assumptions on the inlet profiles needed in the established stress-resolved fiber model by Doufas et al.\ (2001). Simulation results demonstrate the performance and application regime of the dimensionally reduced models. The novel stress-averaged variant provides fast and reliable results, especially in the regime of low flow-enhanced crystallization.