Liquid lace

Under-extrusion is a common problem in additive manufacturing processes. Forming solid printed layers requires careful calibration of flow rate, nozzle speed, and layer height. If not enough material is supplied, adhesion will start to fail, creating holes that tend to propagate to subsequent layers. In this project, we study how intentionally under-extruding can lead to regular patterns, where the extrudate either sticks to the layer below (we call this a blob) or stretches in the air (we call this a thread).

Our thermography (above) suggests the underlying mechanism of pattern formation, where rapid solidification of molten fluid upon surface contact effectively pins an unsustainable volume of fluid in a blob, leading to the formation of a thread that stretches and thins to conserve mass. While the thread stretches, it also cools, and a new droplet of molten polymer is able to grow at the nozzle. The process restarts when the new droplet is large enough to touch the surface. While the pattern forms spontaneously on a flat surface, the characteristically uneven shape consisting of blob mounds and thread valleys reinforces a natural wavelength.

“Twisted towers” 

Beyond 1D stacked rows, we probe the same mechanism with circular toolpaths. Without a continuous substrate beneath the path, the extrudate can miss contact, form long free spans, and anchor only at occasional adhesion sites. This intermittent anchoring, together with the pattern’s natural layer-to-layer phase shift, produces twisted-tower-like forms.

Continuing work

This project is ongoing; current directions include:

• Minimal models of pattern formation. We are developing reduced models that capture onset, natural wavelength selection, and nonlinear limit-cycle behavior. We identify the chiefly geometric parameters that govern how the pattern stretches, how quickly it recovers from perturbations, and when frustration nucleates topological defects.
• Viscoelastic mechanics during rapid cooling. We are interested in quantifying how a spanning thread transitions from viscous to elastic as it cools, using the printer as a controlled platform. We aim to map temperature-dependent modulus/viscosity (Deborah number), relate this evolution to necking and flow cut-off into the thread, and explain the emergence of droplets.
• Micro-CT as a tracer of process history. We have observed spatial radiodensity patterns in X-ray CT images. We would like to test whether they proxy extensional flow (via pigment transport/segregation), record cooling history, or reflect some other mechanism at play in this system.