Doctor of Philosophy (PhD)


Petroleum Engineering

Document Type



This study investigates foam delivery mechanisms in vadose-zone remediation by using Method of Characteristics (MoC). In such applications, dry foams are introduced into a porous medium which is initially at low saturation of water (Sw) containing pollutants such as metals and radionuclides. For vadose-zone remediation processes to be successful, the injected aqueous phase should carry chemicals to react with pollutants and precipitate them for immobilization and stabilization purposes. Typical remediation techniques such as water and surfactant injections are not applicable, because of the concerns about downward migration. As a result, understanding foam flow mechanism in-situ is key to the optimal design of field applications. This study mainly consists of two parts: Part 1, formulating foam model mathematically using method of characteristics (MoC) and fractional flow analysis; and Part 2, using the model to fit to experimental data. Results from Part 1 show that foam delivery mechanism is indeed very complicated, making the optimum injection condition field-specific. The five major parameters selected (i.e., initial saturation of the medium, injection foam quality, surfactant adsorption, foam strength, and foam stability) are shown to be all important, interacting with each other linearly and non-linearly. In addition, the presence of water bank ahead of stable foams conjectured in previous studies is confirmed. Results also imply that although dry foam injection is generally recommended, too dry injection condition is found to hurt this process due to slow foam propagation. The results from Part 2 reveals a few important insights regarding foam-assisted deep vadose zone remediation: (i) the mathematical framework established for foam modeling can fit typical flow experiments matching wave velocities, saturation history and pressure responses; (ii) the set of input parameters may not be unique for the fit, and therefore conducting experiments to measure basic model parameters related to relative permeability, initial and residual saturations, surfactant adsorption and so on should not be overlooked; and (iii) gas compressibility plays an important role for data analysis, thus should be handled carefully in laboratory flow experiments. Foam kinetics, causing foam texture to reach its steady-state value slowly, may impose additional complications.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Kam, Seung