(474d) Imaging and Modeling the Intermittent Flows of Attractive Nanoemulsions

Liquid-based soft material extrusion processes exhibit complex phase behavior and velocity distributions, driven by the interplay of pressure-driven flow and attractive forces amongst colloidal particles. However, shear flow and bulk rheology cannot fully capture the behavior in channel flow of non-Newtonian fluids, which lead to difficulties predicting the printability of liquid-based inks and other non-Newtonian fluids. We couple fast line-averaging and object tracking image processing with first principles transport modeling to capture the complex phase behaviors within gels passing through a cylindrical capillary at temperatures between 39°C to 60°C and radially averaged shear rates between 0.21 s^-1 to 6.2 s^-1. The liquid is a colloidal system containing thermoresponsive nanoemulsions of 20 vol% fluorescent poly(dimethylsiloxane) droplets suspended in a liquid precursor containing an anionic surfactant, 200 mM sodium dodecyl sulfate, and a thermal gelator, 33 vol% poly(ethylene glycol diacrylate). This system spontaneously self-assembles into gel networks with length scales from nanometers to microns at elevated temperatures due to interparticle attractions. When a 0.5-30 kPa pressure drop is applied to the nanoemulsions in a 1 mm cylindrical capillary tube, they flow and undergo shear thinning within the specified shear rates, which further undergo local cluster migration and lead to a highly heterogeneous flow profile that cannot be predicted using bulk rheological models. High-speed confocal microscopy shows the presence of intermittent flow profiles punctuated by syneresis at specific flow velocities, exacerbated by a large pairwise attraction between droplets. To predict the distribution of the oil phase during flow, we develop a transport model using the finite difference method, which discretizes the governing equations for momentum (advection) and species conservation (dispersion) within the capillary. By integrating a large dataset of dynamic images with the model, we capture the cluster diffusion coefficients and local velocity profiles needed for fast control of additive manufacturing processes through in situ monitoring methods.