Dense Nonaqueous Phase Liquid Saturation Research Articles

Aquifer remediation by pumping: A model for stochastic‐advective transport with nonaqueous phase liquid dissolution

The coupled effects of advection and rate‐limited mass transfer to the aqueous phase from residual dense nonaqueous phase liquid (DNAPL) in the saturated zone are investigated within an analytical stochastic‐advective framework. The transport domain is assumed to consist of a source zone where mass transfer takes place and a down‐gradient plume zone where transport is by advection only. The mass flux in the aquifer is evaluated from an analytical solution to coupled advection‐reaction and nonlinear mass transfer equations. The results emphasize the importance of the parameters associated with the source zone: its length in the mean flow direction, the residual DNAPL saturation, and the mass transfer rate coefficient. The sensitivity to the hydraulic conductivity variance is low for significantly rate‐limited mass transfer. It is found that advection in the plume zone has small effect on mass fluxes and cleanup times for the considered lengths of the source and plume regions.

Influence of Viscous, Gravitational, and Capillary Forces on DNAPL Saturation

AbstractFour dense nonaqueous phase liquids (DNAPLs)—bromoform, chlorobenzene, tetrachloroethylene, and trichloro‐ethylene—were used to investigate the influence of viscous, gravitational, and capillary forces on DNAPL saturation in a natural aquifer sand. The relative magnitudes of these forces are expressed in terms of two dimensionless groups, the Capillary Number (NCa), defined as the ratio of the viscous force to capillary force, and the Bond Number (NBo), defined as the ratio of the gravitational force to capillary force.Nondimensionalization of the equations governing two‐phase flow suggests that DNAPL saturation should be a function of a linear combination of the Capillary and Bond Numbers (NCa/krw— NBo), provided the permeability to water (krw) in the presence of discontinuous DNAPL is considered. Experimental studies in which DNAPL saturations were measured over a range of Capillary and Bond Numbers for upward, horizontal, and downward displacement of DNAPL by water corroborate the results of the nondimensionlization. DNAPL saturations generally decreased with increasing Capillary Number and with decreasing Bond Number until NCa/krw— NBo was greater than approximately 1 × 10‐5 at which point residual saturation was attained. For the DNAPLs used in this study, with adhesion tensions on the order of 26 dynes/cm and Bond Numbers ranging from 1.3 × 10‐7 to 2.4 × 10‐6, residual saturation was attained at Capillary Numbers greater than approximately 5 × 10‐5. These results provide a means of estimating the system conditions under which the DNAPLs studied achieve residual saturation in aquifer material.

Observed Migration of a Controlled DNAPL Release by Geophysical Methods

AbstractSeven hundred seventy liters of a dense nonaqueous phase liquid (DNAPL), tetrachloroethylene (PCE), were released into an isolated volume of a completely saturated natural sandy aquifer. The release was monitored over a period of 984 hours with a variety of geophysical methods including ground penetrating radar, time domain reflectometry, in situ resistivity, and a neutron soil moisture probe. The PCE formed a pool on a low permeability layer at approximately 1 m depth and spread over an area exceeding 32 m2. In its course of downward migration, the PCE subsequently formed eight smaller pools. At the end of the experiment an estimated 41 percent of the total PCE volume remained trapped in the upper pool.The PCE mass and its spatial moments were calculated from radar reflection amplitudes. Between 48 and 100 percent of the PCE mass was accounted for by radar measurements. The center of mass moved a total of 0.5 m south southeast and 1.3 m downward. Spatial variances showed that the greatest lateral spreading occurred in the east‐west direction. The results demonstrate that natural heterogeneities, even in a relatively homogeneous aquifer, can cause DNAPLs to spread laterally over large areas in the subsurface. This experiment also demonstrated that while the ability of geophysics to uniquely measure the presence of DNAPL is limited, certain techniques are well‐suited to monitoring changes in DNAPL saturation.

Chemistry and analysis of the permitted coal-tar food dyes

Direct pumping and enhanced recovery of coal tar and creosote dense, non-aqueous phase liquids (DNAPLs) from the subsurface have had mixed results because these DNAPLs are viscous fluids that can potentially alter aquifer wettability. To improve the inefficiencies associated with waterflooding, the research presented here considered the use of a polymer solution that can be added to the injected flood solution to increase the viscosity and decrease the velocity of the flooding solution. Results from one-dimensional, vertically oriented laboratory column experiments that evaluate the recovery of coal-derived DNAPL with both water and polymer flooding solutions are presented. The final DNAPL saturation remaining in the column was assessed in water and oil-wet systems for three viscous DNAPLs. Adding polymer to increase the aqueous solution viscosity did not have a significant impact in water-wet systems. A final DNAPL saturation of approximately 19% was achieved for both water and polymer floods. In contrast, the addition of polymer significantly improved recovery in oil-wet systems. The final saturation was over 40% in oil-wet systems after waterflooding, but approximately 19% with a polymer flushing solution. Although the final saturation produced with polymer flooding was similar between the oil- and water-wet systems, differences in the relative permeability and distribution of DNAPL in the porous matrix caused the DNAPL recovery to be much slower in the oil-wet system.