Fractal Screw Designs
A fractal screw design has been conceived to enable single pellet plastics extrusion. The reason is that extrusion screws have changed relatively little in the past 100 years. Conventional feed screw designs (shown at top of figure) use a channel depth that decreases from the feed zone to the metering zone. While the channel pitch and depth may vary along the screw axis, design thinking is dominated by the concept of compression ratio, which is defined as the ratio between the swept volume (proportional to pitch times depth) in the feed zone to that of the metering zone. The historical trend has been to increase throughput using longer screws and higher length:diameter (L:D) ratios wherein the feed, transition, and metering zones each represent roughly one-third of the active screw length.
Perhaps the single most significant advance in screw design in the past 50 years has been the development of the barrier screw, designed to provide more efficient melting than single-flighted conventional screws. As shown in the figure at center right, the barrier screw design introduces a secondary flight to increase shearing of the material and associated melting of the pellets. The secondary flight also acts as a “barrier” between the melted polymer and unmelted pellets, thereby avoiding the formation of a melt pool that can act as an insulating layer to prevent pellets from melting efficiently.
Reflecting on the dynamics in the feed and early transition zones, it is evident that very little work is performed on the pellets in the solidified bed, while the depth of the solidified bed also limits heat conduction. The single primary channel, in both conventional and barrier screws, also results in “surging” of the volumetric flow rate as the pellets at the feed throat enter the feed section and are driven up the screw. This phenomenon results in melt pressure and temperature fluctuations that causes melt instability and lower output quality.
Accordingly, the research is investigating a “fractal” screw design to enable single pellet plastics extrusion. The proposed fractal design is composed of multiple channels in the metering zone that combine to form larger channels in the transition and/or feed zones. For example, the fractal design (shown in the figure at bottom) splits the feed channel into three sub-channels, each of which are later split into two sub-channels in the transition zone. The two different locations for the channel divisions are provided for different reasons. An early split in the feed section is provided to physically break up the solidified bed while the later split in the transition section is provided to physically align and serially process individual pellets. The design is intended to provide the same throughput at a given L:D while reducing and narrowing the residence time distribution, reducing energy consumption, improving melt dispersion, and increasing melt homogeneity.
Three different screw design concepts: (top) conventional screw, (center) barrier screw, (bottom) fractal screw
Particle-Based Process Simulation
This project is exploring particle-based modeling of thermoviscoelastic flows in polymer processing to better model the transport, deformation, and melting of pelletized feedstocks so as to enable serial processing of individual pellets for improved efficiency, consistency, and quality. The proposed paradigm leverages recent advances in smoothed particle hydrodynamics (SPH) and parallel computing to provide many significant advantages over Eulerian approaches including (1) modeling of multiple materials and flow types, (2) inherent particle tracking to assist viscoelastic modeling, and (3) free surface modeling to model granular pellet flow, pellet breakup, and melt flow in the screw, leakage over flights, and through the die.
Concurrent with material characterization, the researchers are investigating SPH models of the polymer behavior within capillary and parallel plate rheometers; such modeling has not before been attempted, but it is a vital step in validating the constitutive models and numerical implementation before simulating more complex polymer processing systems. Then, the validated simulations will be used to optimally engineer a fractal screw design with multiple channels to break up the solidified bed and match the transient melting of the pelletized feedstock while providing multiple flow channels to dampen output variation. Optimized conventional and fractal extrusion screws will be machined, tested, and scaled to production applications at the industry liaison. We thus envision single pellet plastics extrusion with greatly improved performance and control of polymer states.