Research Overview

The Computational Multiphase Flow group is engaged in the development of numerical methods for direct simulations of multiphase flows and the use of those methods to study various aspects of multiphase flows. Projects include bubbly flows, suspensions, boiling, atomization, drop inpact, and solidification. Past and present sponsers include NSF, NASA, AFOSR, DARPA, ONR, GRI, and DoE.

Multi-phase and multi-fluid flows are common in many natural and technologically important processes. Rain, spray combustion, spray painting and boiling heat transfer are just a few examples. While it is the overall, integral characteristics of such flow that are of most interest, these processes are determined to a large degree by the evolution of the smallest scales in the flow. The combustion of sprays, for example, depends on the size and the number density of the drops. Generally, these small-scale processes take place on a short spatial scale and fast temporal scale, and in most cases visual access to the interior of the flow is limited. Experimentally, it is therefore very difficult to determine the exact nature of the small-scale processes. Full numerical simulations, where the governing equations are solved exactly on a computer, offer the potential to gain a detailed understanding of the flow. Such full simulations—where it is necessary to account accurately for inertial, viscous and surface tension forces in addition to follow the motion of a deformable interface between the different phases—are one of the most difficult problems in computational fluid dynamics.

We have developed a numerical method that holds considerable promise in this area. By combining a front tracking technique, where separate computational elements are used to explicitly represent the interface between the different phases, with a relatively conventional finite volume method we are able to accurate simulate fully three-dimensional systems with, for example, many bubbles and drops for a long time, thus studying the small-scale dynamics in detail. We have used this technique to look at several problems, including collision and coalescence of drops, the rise of buoyant bubbles, including the effect of surfactants and vertical shear, the dynamics of bubble clouds, thermal migration of many drops, cavitating bubbles and solidification. These simulations have shown several unexpected effects and provided a detailed understanding of other known, but poorly understood effects.

We are currently developing the numerical technique further, both making it more efficient and applicable to problems with more complicated physics, as well as using it to examine several multi-phase problems. The goal here is to develop a sufficient understanding of the "elementary" processes in multiphase flows to aid in the development of engineering models for these flows.

Maintained by webmaster@wpi.edu
Last modified: Aug 06, 2009, 15:07 EDT