Paul T. Imhoff

An experimental study of complete dissolution of a nonaqueous phase liquid in saturated porous media

The attenuation of gamma radiation was utilized to measure changing residual trichloroethylene (TCE) saturation in an otherwise water-saturated porous medium as clean water was flushed through the medium. A front over which dissolution actively occurred was observed. Once developed, this front varied in length from ≈11 mm to ≈21 mm, lengthening as it moved through the porous medium. Gamma attenuation measurements and analyses of effluent water samples indicate that there was minimal if any transport of TCE as colloidal droplets. Even as trapped TCE ganglia decreased in size due to dissolution, there is no evidence that they became mobile and advected downgradient. An extraction of the porous medium at the completion of one experiment indicated that less than 0.002% of the original TCE mass remained, suggesting that minimal amounts of separate phase TCE remained trapped within the medium after flushing with 290 pore volumes. Mass transfer rate coefficients were computed and are shown to be a function of Darcy flux, TCE volumetric content, and distance into the region of residual TCE.

Multiphase flow and transport modeling in heterogeneous porous media: challenges and approaches

Abstract We review the current status of modeling multiphase systems, including balance equation formulation, constitutive relations for both pressure-saturation-conductivity and interphase mass transfer, and stochastic and computational issues. We discuss weaknesses and inconsistencies of current approaches based on theoretical, computational, and experimental evidence. Where possible, we suggest new or evolving approaches.

Cosolvent-enhanced remediation of residual dense nonaqueous phase liquids: experimental investigation.

The removal of denser than water nonaqueous phase liquids (DNAPLs) trapped at residual saturation is an important problem at many contaminated groundwater sites. Because pump-and-treat technologies have been ineffective in removing DNAPLs, alternative strategies have been suggested, one of which is enhancing the mobilization and dissolution of DNAPLs by flushing with a cosolvent. Tetrachloroethylene (PCE)/methanol/water systems were studied to evaluate the effect of methanol on the remediation of PCE-contaminated porous media. Experimental measurements of interfacial tension, equilibrium phase composition, and phase density at various methanol/water fractions were combined with other published properties to characterize these systems. In methanol flushing experiments, PCE mobilization, non-equilibrium PCE dissolution, and flow bypassing were all observed. The results demonstrate that (a) small-scale heterogeneities may lead to locally high residual DNAPL saturations that are more easily mobilized than DNAPL residuals in homogeneous media ; (b) mass transfer rate coefficients for PCE/methanol/water systems can be predicted to within 30% using an existing correlation developed for systems with similar NAPL emplacement procedures ; and (c) flow bypassing, due to nonuniform distributions of DNAPL residual or dissolution fingering, can occur in even small-scale experiments.

Dissolution fingering during the solubilization of nonaqueous phase liquids in saturated porous media 2. Experimental observations

Nonaqueous phase liquids (NAPLs) are a common source of contamination at polluted groundwater sites, where they frequently remain trapped within the pore space at residual saturation and reduce the permeability of the medium to aqueous phase flow. The model presented in a companion paper [Imhoff and Miller, this issue] suggested that when fluid flow is imposed on such a system, the aqueous phase may interact with dissolution-induced permeability changes, and lead to fingered patterns. In this investigation, a two-dimensional flow cell was used to study the effects of porous medium structure, Darcy flux, initial residual NAPL saturation, median particle diameter, gravity, and NAPL composition on dissolution fingering. Fingering occurred when two conditions were met: (1) 11 to 80 e-fold times had elapsed, where e-fold time is the time required for the instability to grow by a factor e and was predicted from the linear stability analysis in the companion paper; and (2) the length of the dissolution front before finger development was smaller than the zone of NAPL residual. Where fingers formed, finger structure was similar and showed no systematic variation within the parameters investigated. Observed finger wavelengths compared well with model predictions. A single experiment in a three-dimensional cell, 1 m long, demonstrated that fingers can grow to at least 30 cm in length. When experimental observations in this cell were compared with predictions of NAPL dissolution based on models that did not include fingering, the measurements of changing NAPL saturation differed significantly from model predictions.

Dissolution Fingering During the Solubilization of Nonaqueous Phase Liquids in Saturated Porous Media: 1. Model Predictions

The dissolution of nonaqueous phase liquids (NAPLs) trapped at residual saturation is an important problem at many contaminated groundwater sites. It is well known that NAPL ganglia trapped within the pore space reduce the permeability of the medium to aqueous phase flow. When fluid flow is imposed on such a system, the aqueous phase may interact with the dissolution-induced permeability changes, leading to fingered patterns. This mechanism is very similar to that of mineral dissolution instabilities, which are a particular example of reactive infiltration instabilities. Extending that literature, we present a nonlinear model describing the dissolution of NAPL ganglia and perform a linear stability analysis of the resultant moving free boundary problem, demonstrating that instabilities may develop from a planar dissolution front. Predicted finger wavelengths are a function of both residual NAPL saturation and the imposed aqueous phase flow rate; they range from centimeters to meters. Experimental observations of dissolution fingering are presented in a companion paper [Imhoff et al., this issue] and are compared with predictions from this model. Dissolution fingering may affect the solubilization of NAPL ganglia in natural environments and in experimental studies of NAPL dissolution intended to quantify mass transfer rates.

Effect of liquid distribution on gas-water phase mass transfer in an unsaturated sand during infiltration

Abstract Gas-water phase mass transfer was examined in a homogeneous sand with both the gas and water phase mobile: water was infiltrated from the top of the sand column while benzene-laden air flowed upward from the bottom. Mass-transfer limitations for this situation may be important for applications of bioventing, where water and nutrients are added at the ground surface simultaneously with induced air movement to carry oxygen and volatile organics to microbial populations. Gas- and water-phase samples indicate that gas-water phase mass transfer was sufficiently fast that equilibrium between gas and water phases was achieved at all sampling locations within the porous medium. Lower-bound estimates for the gas-water mass-transfer rate coefficient show that mass transfer was at least 10–40 times larger than predictions made from an empirical model developed for gas-water phase mass transfer in an identical porous medium. A water-phase tracer test demonstrates that water flow was much more uniform in this study than in those earlier experiments, which is a likely explanation for the differing rates of gas-water phase mass transfer. It is hypothesized that the liquid distribution in previous laboratory experiments was less uniform because of preferential flow paths due to wetting front instabilities. Gas-water phase mass-transfer rate coefficients reported in this investigation are for an ideal situation of uniform water infiltration: mass-transfer rates in field soils are expected to be significantly smaller.

Development of a correlation for aqueous-vapor phase mass transfer in porous media

Abstract In many situations vapor-phase extraction procedures (e.g., soil venting, air sparging, and bioventing) may be suitable methods for remediating porous media contaminated by volatile organic compounds. This has led to increased study of operative processes in these systems, including aqueous-vapor phase mass transfer. Past work has shown the importance of the flow regime on this process, but a quantitative estimate of mass-transfer coefficients is lacking, especially for systems not confounded by uncertainties involving interfacial area between the phases. An experimental investigation was conducted to isolate the resistance to aqueous-vapor phase mass transfer at the phase boundary, using an ideal porous medium system. Mass-transfer coefficients were measured for toluene for a wide range of Reynolds numbers. An empirical model was fit to the data in dimensionless form. The mass-transfer model was coupled with an available interfacial area model, yielding a dimensionless expression for the mass-transfer rate coefficient. This expression was used to compare results from this work to three other experimental studies reported in the literature. These comparisons showed that for experiments where infiltrating water flowed uniformly within the porous medium, the predicted mass-transfer coefficients were within a factor of 5 of the measured coefficients. Mass transfer was significantly slower than the rate predicted, using the results from this work, in experiments where infiltrating water flowed nonuniformly.

Quantifying Factors Limiting Aerobic Degradation During Aerobic Bioreactor Landfilling

A bioreactor landfill cell at Yolo County, California was operated aerobically for six months to quantify the extent of aerobic degradation and mechanisms limiting aerobic activity during air injection and liquid addition. The portion of the solid waste degraded anaerobically was estimated and tracked through time. From an analysis of in situ aerobic respiration and gas tracer data, it was found that a large fraction of the gas-filled pore space was in immobile zones where it was difficult to maintain aerobic conditions, even at relatively moderate landfill cell-average moisture contents of 33-36%. Even with the intentional injection of air, anaerobic activity was never less than 13%, and sometimes exceeded 65%. Analyses of gas tracer and respiration data were used to quantify rates of respiration and rates of mass transfer to immobile gas zones. The similarity of these rates indicated that waste degradation was influenced significantly by rates of oxygen transfer to immobile gas zones, which comprised 32-92% of the gas-filled pore space. Gas tracer tests might be useful for estimating the size of the mobile/immobile gas zones, rates of mass transfer between these regions, and the difficulty of degrading waste aerobically in particular waste bodies.