research with: Prof. Clay Radke, Department of Chemical Engineering
Prof. Tad Patzek, Department of Material Science and Mineral Engineering
University of California, Berkeley
learn more about the Radke Group and the College of Chemistry

The goal of my research was to characterize the stability of foam in porous media. The utility of this project is based on the applications of foam as an agent in enhanced oil recovery and for use in underground gas storage and leak containment. An additional application of foam in porous media is to prevent water coning in a gas well.

After foam is generated, its texture will coarsen as a result of the diffusion of gas between adjacent bubbles through liquid films, or lamellae. The driving force for this diffusion is a pressure gradient across the film. Click below to see a photograph of bulk foam. You can see that small bubbles are at higher pressure than large bubbles by looking at the curvature of the lamellae. Small bubbles will shrink at the expense of large bubbles until the foam has totally disappeared.

As diffusion driven foam decay occurs, gas rearranges throughout the system. Therefore, the decay of foam can be quantified by tracking gas pressures as they change with time in stationary foam. I used both modeling and experiments to follow and understand the pressure changes and thus describe the stability of foam in porous media. Foam in a porous medium has the same basic behavior as bulk foam except that the final result of diffusion is not total decay. The final state of foam in a porous medium with no external disturbances is one with all the lamellae sitting in the pore throats and all bubbles at equal pressure.

For the purpose of simulating the decay of foam, a pore is approximated as an hourglass shape, generated by rotating a sphere about an axis of symmetry. A network of these pores is created with a specified distribution of pore sizes and bubbles are defined by distributing lamellae throughout these pores. As the gas diffuses across the lamellae, they move until they reach the pore throats. The motion of these lamellae is governed by sets of mass transfer and force balance equations. Solving these equations simultaneously allows us to predict the features of the foam diffusion process. I have written a simulation program that can track the gas diffusion for various size networks.

Interesting behavior of foam in porous media has been discovered. Due to the curvature of the pore surfaces, lamellae configurations tend to undergo sudden jumps. When all the lamellae in the porous medium are heading toward the pore throats (x=0), the lamellae may undergo a series of oscillations before proceeding to equilibrium.

Since we cannot measure the pressures in the individual bubbles within a foam, we must find an alternative quantity to track experimentally. We have built an apparatus to measure the "headspace" pressure adjacent to a porous medium as the foam within decays. The headspace can be thought of as a very large pore body, one with a volume much greater than the scale of the pores. This pressure increases with time as the foam decays and early simulation results suggest that the time scale for this process is of the same order as for the diffusion of bulk foam with a similar bubble size distribution.

In addition to tracking the "headspace pressure," I measure the pressure gradient created during the foam generation process and see how it evolves with time. The simulator can also predict the behavior of foam in a porous medium under the influence of a pressure gradient and the simulation results can be compared to experimental data.

The most recent portion of my thesis research involved prediction of the mobilization behavior of foam in porous media. A large fraction of foam is trapped under normal flow conditions, and the purpose of the work was to better understand the magnitude of the pressure necessary to lead to mobilization. A simple program allows the prediction of the fraction of foam which is mobilized as a function of the pressure drop for any defined pore size distribution. A resulting analytic expression is the first which gives the mobilized fraction and can be used in continuum models which exist for understanding flow of foam in porous media.