Microfluidic Droplets

This work was predominantly funded by an Australian Research Council Linkage (CSIRO) grant, `Drop deformation in confined microfluidic geometries' (LC0348317).

A parametric study of droplet deformation through a microfluidic contraction-expansion

Dalton Harvie, Malcolm Davidson, Justin Cooper-White, Murray Rudman and Gary Rosengarten

Section under construction

Viscoelastic droplets

Dalton Harvie, Malcolm Davidson and Justin Cooper-White

We have also studied the behaviour of viscoelastic droplets that pass through microfluidic contraction-expansions. The particular simulation shown below is based on an experiment performed in a thin planar contraction, in which a small `packet' of surrounding continuous phase Newtonian fluid was encapsulated within the viscoelastic PEO droplet after it had passed through the 4:1:4 contraction (see the publication).

Above: Simulation of a viscoelastic droplet that is surrounded by a Newtonian fluid and passes through a planar contraction. The simulations are performed in 2D Cartesian coordinates and use an Oldroyd-B `dumbbell' model to capture polymer rheology. Numerically a coupled VOF method was used to simultaneously advect the fluid volume fractions and dumbbell orientation tensors (more details in numerical methods). The shading represents the nondimensional length of the polymers (the square root of the trace of the orientation tensor). The `vectors' indicate the average length and principal orientation of the polymers.

The 2D simulations (above) capture the formation of a precursory forked tail on the droplet while it is within the contraction. They also predict the formation of a stress field, created by the extension and orientation of the polymers, that supports the encapsulation of continuous phase fluid from the rear of the droplet forward, along its centreline. Substantial encapsulation of the surrounding fluid is not predicted by the simulations: This is most probably because of the 2D Cartesian approximation employed to make the problem computationally tractable.

Work is ongoing to simulate this physical process in cylindrical and Cartesian 3D coordinates.

Publications