PubMed2000
The success of air sparging as a remedial technology for treatment of contaminated aquifers is well documented. However, there is no consensus, to date, on the mechanisms that control the flow of injected air through the saturated ground. Currently, only qualitative results from laboratory experiments are available to predict the zone of influence of a sparging well. Given that the patterns of air flow through the soil will ultimately determine the efficiency of an air sparging treatment, it is important to quantify how sparged air travels through a saturated porous medium. The main objective of this research is to develop a model that describes air transport through saturated porous media. This paper presents results from an ongoing study that employs centrifuge modeling to reproduce in situ air sparging conditions. Centrifuge testing is an experimental technique that allows reduced-scale duplication, in the laboratory, of the stresses and pressure distributions encountered in the field. In situ conditions are critical in the development of actual air flow patterns. Experiments are being conducted in a transparent porous medium consisting of crushed borosilicate glass submerged in fluids of matching indices of refraction. Air is observed as it flows through the porous medium at varying gravitational accelerations. Recorded images of experiments allow the determination of flow patterns, breakthrough velocities, and plume shapes as a function of g-level and injection pressure. Results show that air flow patterns vary from fingering, at low g-levels, to pulsing at higher accelerations. Grain and pore size distribution of the porous medium do not exclusively control air flow characteristics. Injector geometry has a definite effect on breakthrough velocities and air plume shapes. Experiments have been conducted to compare the velocity of air flow through the saturated porous medium to that of air in pure liquids. Results show that the velocity of air through the medium is lower than that in the pure fluid, as expected. At high g-levels however, plume breakthrough velocities are proportional to the velocity of the air in the pure fluid.