The Interfaces in Multiphase Flows Laboratory develops and applies advanced computational methods to investigate interfacial phenomena in multiphase flows. Examples include the consequences of electrostatic interactions between droplets in emulsion flows, the effects of surfactants, and separations driven by electrokinetic phenomena.
More details about my past and current research projects can be found in my publication list.
When carbon dioxide dissolves into an aqueous suspension of particles, some of it reacts to form a pair of ions, H+ and HCO3-. As these ions diffuse into the suspension, they attract/repel suspended particles to/from the source of CO2 in a phenomenon called diffusiophoresis. This effect has been exploited in the design of a continuous process for colloidal particle filtration. One of the advantages of using carbon dioxide or other gases to create ion concentration gradients is that the added component can be easily removed, in this case by de-gassing, after the separation is performed.
Ongoing research is examining the optimization and applications of this process, different systems involving CO2, the use of other soluble gases, and more generally a variety of multiphase flows where solutes react and diffuse, creating concentration gradients that drive useful particle motion.
When the droplets of a flowing emulsion collide, the fluid between them compresses into a thin film. If the collision lasts long enough, the film ruptures and the two droplets merge (coalesce). Simulations of droplet collisions and coalescence are challenging because they must resolve these films that are several orders of magnitude smaller than the droplets. An understanding of the conditions when droplets coalesce is needed, for example, to predict how droplet size distributions change when emulsions are produced and flow through pipes and processing equipment. I use the phase field (free energy) lattice Boltzmann method (LBM) to study such droplet flows. I am continuing to develop a custom parallel (MPI+CUDA) code, which to date I have run on over 60 GPUs in a cluster. I am also exploring the use of other new parallel computing platforms.
Phase field LBM simulations are also useful for studying the breakup of droplets, and in general multiphase flows in which the fluid phases undergo topological changes.
Much of the energy consumed during the production of emulsions (emulsification) is lost to friction rather than being used to create new surfaces between the liquids. However, when thin films of oil spread on surfactant solutions, they can spontaneously break up into lenses. This process creates complex and visually-striking patterns, such as the one shown on the left. An understanding of the dynamics of the growth of the holes in the film and breakup of the ridges around the holes into lenses is needed to predict the resulting lens size distribution. The lens size distribution in turn determines the droplet size distribution of emulsions produced in this way. I am working on simulations of the hole opening and ridge breakup processes.
Surfactants, or surface active agents, are molecules that alter the properties of the interfaces between fluids. For example, when the amount of surfactant varies along a surface, the non-uniform surface tension creates a stress on the adjacent fluids, inducing them to flow or resisting their motion. I recently studied how random initial distributions of surfactants on a fluid surface spread out to achieve a uniform distribtion and the flow that occurs beneath the surface. When we account for the inertia of the sub-surface flow and consider a sinusoidal initial distribution, we obtain a surprising result: the spreading surfactant changes direction three times before relaxing monotonically to a uniform distribution.