Information on the transport of dissolved gases in ground water is needed to design ways to increase dissolved gas concentrations in ground water for use in in situ bioremediation (e.g., O2 and CH4) and to determine if dissolved gases are conservative tracers of ground‐water flow (e.g., He). A theoretical model was developed to describe the effect of small quantities of trapped gas bubbles on the transport of dissolved gases in otherwise saturated porous media. Dissolved gas transport in porous media can be retarded by gas partitioning between the mobile aqueous phase and a stationary trapped gas phase. The model assumes equilibrium partitioning where the retardation factor is defined as R = 1 + H′(Vg/Vw) where H' is the dimensionless Henry's Law constant for the dissolved gas, and Vg and Vw are the volumes of the trapped gas and water phases, respectively. At 15°C and with Vg/ Vw= 0.05, the predicted retardation factors for He, O2, and CH4 are 5.8, 2.4, and 2.3, respectively. The validity of the model was tested for dissolved oxygen in small‐scale column experiments over a range of trapped gas volumes. Retardation factors of dissolved oxygen increased from 1 to 6.6 as Vg/Vw increased from 0 to 0.123 and are in general agreement with model predictions except for the larger values of Vg/Vw. The theoretical and experimental results suggest that gas partitioning between the aqueous phase and a trapped gas phase can greatly influence rates of dissolved gas transport in ground water.

Abstract. Oxygen is often the rate-limiting factor in aerobic in situ bioremediation. This paper investigates the degree to which air or oxygen gas can be emplaced into the pore space of saturated porous media and provide a significant mass of oxygen. Column experiments were performed to test three emplacement methods: direct gas injection, injection of water supersaturated with gas, and injection of a hydrogen peroxide solution. The direct gas injection method fills 14-17% of the pore space with trapped gas. Water supersaturated with gas fills 18-27% of the pore space with a trapped gas phase, and hydrogen peroxide solution injections emplaces trapped gas in 17-55% of the pore space. In addition to supplying oxygen, gas entrapment causes a decrease in hydraulic conductivity which could be an advantage by decreasing the flow of contaminants offsite. The relative hydraulic conductivity of porous media with a trapped gas volume of 14-55% was 0.62-0.05. IntroductionAlternative methods are needed for introducing oxygen into groundwater for aerobic in situ bioremediation of contaminants. In situ bioremediation is often limited by the amount of oxygen available to the microorganisms in the subsurface. To degrade a simple hydrocarbon (e.g., benzene), approximately 3.1 times more oxygen than contaminant in mass/volume is necessary to meet the stoichiometric requirements. Since the solubility of oxygen in water is low, it is difficult to significantly increase the oxygen mass in a contaminated aquifer by dissolving the oxygen in water first before transferring it to the aquifer. Twenty-eight times more oxygen per volume can be stored in the gas phase than can be dissolved in water, assuming equilibrium based on Henry's law at 15øC.We are investigating whether a wall or zone of trapped gas bubbles can be emplaced into an otherwise-saturated porous medium and provide a substantial source of oxygen for bioremediation of contaminated groundwater (Figure 1). If 15% of the pore space can be filled with oxygen gas, 20 times more oxygen will be emplaced compared to the case in which oxygen is dissolved in the pore water in equilibrium with air. Once a wall or zone of oxygen bubbles is emplaced into an aquifer, the oxygen will be dissolved by the water and be potentially available for use by microorganisms in biodegradation. When the oxygen is used up, it can be emplaced again. This cycle may be repeated until the contaminant is degraded to levels that meet the regulatory requirement. Because trapped gas emplacement does not require continuous injection, a bubble wall or zone may be constructed using modified sampling equipment (e.g., a Hydropunch TM or GeoprobeTM), which will allow for much closer spacing of injection points than injection through a well casing. Here we refer to oxygen because it is the gas that is most widely needed for bioremediation of contaminated groundwater, but other gases, such as hydrogen and methane, which have been shown to be effective in remediation of contaminated groundwater, could also be introdu...

No abstract

In situ bioremediation of contaminated aquifers is often limited by the concentration of dissolved oxygen in the ground water. Various methods have been used to increase dissolved oxygen concentrations in ground water, but the effect of a trapped gas phase on the distribution and transport of dissolved oxygen needs to be understood. The two‐dimensional transport of dissolved oxygen is investigated in experiments conducted in a large‐scale physical aquifer model (2 m × 4 m × 0.2 m) where a gas phase is trapped in the pore spaces of an otherwise‐saturated porous medium. The transport of dissolved oxygen is shown to be retarded up to 11.2 times the transport of the bulk water due to the mass transfer of oxygen between the aqueous phase and the trapped gas phase. The theoretical model for dissolved gas transport in the presence of a trapped gas phase is evaluated in a two‐dimensional ground‐water flow field using the U.S.G.S. numerical model MOC. The results show that dissolved oxygen transport can be modeled with the advection‐dispersion equation with linear equilibrium mass transfer but only when the longitudinal dispersion is increased compared to the value determined using a bromide tracer of the water flow. Increased longitudinal dispersion of the dissolved oxygen plume may be due to a temporally or spatially varying retardation factor or rate‐limited mass transfer. The presence of even a small amount of a trapped gas phase in an aquifer will significantly affect the distribution and transport of dissolved oxygen (trapped gas filling only 5% of the pore space will cause a retardation factor for oxygen of 2.6 at T = 15°C) and thus should be considered when designing ways to increase the dissolved oxygen concentration in ground water for in situ bioremediation.

An analytical solution is derived for the advection‐dispersion equation with rate‐limited desorption and first‐order decay, using an eigenfunction integral equations method. The model equations represent one‐dimensional solute transport in a homogeneous isotropic porous medium where the porous medium is saturated with the aqueous solution. The flow field is uniform. Rate‐limited desorption is described as a first‐order process where the rate is proportional to the difference in concentration between the sorbed phase and the aqueous phase. The solution was verified for the limiting case of equilibrium desorption using the solution of van Genuchten and Alves (1982). Example calculations are presented to show the effect of the desorption rate, decay rate, and distribution coefficient on the rate of contaminant removal from both the aqueous and sorbed phases of a groundwater aquifer. The solution quantifies the expected results, where the larger the desorption and decay rate and the smaller the distribution coefficient, the faster the rate of contaminant removal from the aqueous and sorbed phases.