This work focuses on the experimental measurement and mathematical modeling of processes affecting the dissolution of nonaqueous phase liquids (NAPLs) entrapped in sandy porous media. Results of a series of laboratory‐scale one‐dimensional column dissolution experiments indicate that the length of time required to dissolve NAPLs and substantially reduce aqueous phase effluent concentrations is many times greater than predicted by equilibrium calculations. Experimental measurements clearly show an influence of both grain size and grain size distribution on the evolution of effluent concentrations. The longer cleaning times associated with coarse or graded media are attributed to the larger and more amorphous NAPL blobs associated with these media. A general correlation for transient dissolution rates is proposed which incorporates porous medium properties, Reynolds number, and volumetric fraction of NAPL. The model is calibrated with results from styrene dissolution experiments and is shown to adequately predict trichloroethylene dissolution rates in the same sandy media over the period of time required to dissolve the NAPL.

Results of an experimental investigation into steady state dissolution of nonaqueous phase liquids (NAPLs) entrapped within water saturated porous media are presented. The influence of porous media characteristics, NAPL type, and aqueous phase velocity on NAPL dissolution rates is explored through evaluation of a series of laboratory column studies. For many of the conditions tested, measured organic solute concentrations in the column effluents were below solubility limits, indicating nonequilibrium conditions. Entrapped NAPL distributions are examined and shown to depend upon porous media grading and mean grain size. Experimental results reveal a dependence of dissolution rate on the distribution pattern of entrapped NAPL, as well as upon aqueous phase velocity. A phenomenological model for the mass transfer process is developed which incorporates grain size parameters as surrogate measures of NAPL distribution. An additional set of column experiments, employing another porous medium and NAPL type, confirm the usefulness of the model as a predictor of steady state mass transfer rates in similar systems.

The potential for nonaqueous phase liquid (NAPL) mobilization is one of the most important considerations in the development and implementation of surfactantbased remediation technologies. Column experiments were performed to investigate the onset and extent of tetrachloroethylene (PCE) mobilization during surfactant flushing. To induce mobilization, the interfacial tension between residual PCE and the aqueous phase was reduced from 47.8 to 0.09 dyn/ cm by flushing with different surfactant solutions. The resulting PCE desaturation curves are expressed in terms of a total trapping number (N T ), which relates viscous and buoyancy forces to the capillary forces acting to retain organic liquids within a porous medium. The critical value of N T required to initiate PCE mobilization fell within the range of 2 × 10 -5 to 5 × 10 -5 , while complete displacement of PCE was observed as N T approached 1 × 10 -3 . The interplay of viscous and buoyancy forces during PCE mobilization is illustrated in horizontal column experiments, in which angled banks of PCE were displaced through the columns. These results demonstrate the potential contribution of buoyancy forces to PCE mobilization and provide a novel approach for predicting NAPL displacement during surfactant flushing.

No abstract

A coupled experimental and mathematical modeling investigation was undertaken to explore nanoscale fullerene aggregate (nC60) transport and deposition in water-saturated porous media. Column experiments were conducted with four different size fractions of Ottawa sand at two pore-water velocities. A mathematical model that incorporates nonequilibrium attachment kinetics and a maximum retention capacity was used to simulate experimental nC60 effluent breakthrough curves and deposition profiles. Fitted maximum retention capacities (S(max)), which ranged from 0.44 to 13.99 microg/g, are found to be correlated to normalized mass flux. The developed correlation provides a means to estimate S(max) as a function of flow velocity, nanoparticle size, and mean grain size of the porous medium. Collision efficiency factors, estimated from fitted attachment rate coefficients, are relatively constant (approximately 0.14) over the range of conditions considered. These fitted values, however, are more than 1 order of magnitude larger than the theoretical collision efficiency factor computed from Derjaguin-Landau-Verwey-Overbeek (DLVO) theory (0.009). Data analyses suggest that neither physical straining nor attraction to the secondary minimum is responsible for this discrepancy. Patch-wise surface charge heterogeneity on the sand grains is shown to be the likely contributor to the observed deviations from classical DLVO theory. These findings indicate that modifications to clean-bed filtration theory and consideration of surface heterogeneity are necessary to accurately predict nC60 transport behavior in saturated porous media.

Experimental and mathematical modeling studies were performed to investigate the transport and retention of nanoscale fullerene aggregates (nC60) in water-saturated porous media. Aqueous suspensions of nC60 aggregates (95 nm diameter, 1 to 3 mg/L) were introduced into columns packed with either glass beads or Ottawa sand at a Darcy velocity of 2.8 m/d. In the presence of 1.0 mM CaCl2, nC60 effluent breakthrough curves (BTCs) gradually increased to a maximum value and then declined sharply upon reintroduction of nC60-free solution. Retention of nC60 in glass bead columns ranged from 8 to 49% of the introduced mass, while up to 77% of the mass was retained in Ottawa sand columns. When nC60 suspensions were prepared in deionized water alone, effluent nC60 BTCs coincided with those of a nonreactive tracer (Br-), with minimal nC60 retention. Observed differences in nC60 transport and retention behavior in glass beads and Ottawa sand were consistent with independent batch retention data and theoretical calculations of electrostatic interactions between nC60 and the solid surfaces. Effluent concentration and retention profile data were accurately simulated using a numerical model that accounted for nC60 attachment kinetics and a limiting retention capacity.

The objective of this work is to assess the potential significance of deviations from local equilibrium for the exchange of mass between residual nonaqueous phase liquids and the aqueous phase in the saturated groundwater zone. A one-dimensional convection-dispersion mass balance equation incorporating a first-order interphase mass transfer rate relationship and temporal changes in blob configuration is used to model this system. Analytical and numerical rnethods are employed to examine the steady state and transient behavior of the system under a variety of hypothetical aquifer conditions and pumping remediation schemes. Sensitivity of the model to several parameters including mass transfer coefficient, blob size and shape, and Darcy velocity is explored. Results of the theoretical assessment indicate that nonequilibrium effects could play a significant role in some contamination scenarios, primarily for large blob sizes and relatively high velocities. Design of soil flushing techniques will be impacted by these conclusions. Uncertainty in several parameter values used in this analysis indicate the need for further experimental investigation of this process. INTRODUCTION Spills or leaks of organic chemicals to the environment frequently result in the contamination of subsurface soils and groundwater. Many of these pollutants are only slightly soluble in water and thus may exist as virtually immiscible or nonaqueous phase liquids (NAPLs). Documentation of the existence of NAPLs in aquifer systems is growing [Atwater, 1984; Cohen et al., 1987; Feenstra and Coburn, 1986; Schwille, 1988], and the need for an enhancement of our uMerstanding of the fate and transport processes associated with these pollutants is becoming more evident [U.S. Environmental Protection Agency, (USEPA), 1987; Abriola, 9891.Migration of NAPLs in subsurface systems is a complex process. Following a spill or leak, NAPLs generally migrate downward through the vadose zone due to gravitational forces. If the spill is sufficiently large and the NAPL less dense than water, it will eventually reach the water table, where it will spread laterally in the capillary fringe zone. Alternatively, if the NAPL is heavier than water, it will continue to migrate vertically, displacing aquifer pore water [Schwille, 1988;Mackay et al., 1985]. Interfacial forces acting between the water phase or air phase and the NAPL will cause residual "blobs" of the organic phase to be retained within the unsaturated and saturated zones, as depicted schematically in Figure 1. Migration of the bulk organic phase as such will eventually cease when all of the fluid becomes trapped as discontinuous blobs, or when the NAPL encounters a low permeability stratum [Kueper et al., !989] and has insufficient pressure to force the nonwetting NAPL into the small pores of this layer. In this case a pool of NAPL collects on the low permeability stratum. The presence of NAPL in the subsurface represents a potential long-term source of pollution [Pinder, 1982; $chwille, 1988; Baehr, 1987]. In ...

Knowledge of perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) accumulation at the air−water interface is critical to understanding the fate and transport of these substances in subsurface environments. The surface tension of aqueous solutions containing PFOA and PFOS at concentrations ranging from 0.1 to >1000 mg/L and with dissolved solids (i.e., cations and anions) commonly found in groundwater was measured using the Wilhelmy plate method. The surface tensions of solutions containing dissolved solids were lower than those for ultrapure water, indicating an increase in the surface excess of PFOA and PFOS in the presence of dissolved solids. An equation for the surface excess of PFOA and PFOS with total dissolved solids was developed by fitting the measured surface tension values, which ranged from 72.0 to 16.7 mN/m, to the Szyszkowski equation. On the basis of mass distribution calculations for a representative unsaturated, fine-grained soil, up to 78% of the PFOA and PFOS mass will accumulate at the air−water interface, with the remaining mass dissolved in water and adsorbed on the solids.