Lecture 4Multi-phase Flow in Porous MediaNon-aqueous phase liquidsTypical densities of NAPLsFormation of residual and dissolutionFormation of residual and vapor plumeTypical residual saturation values“Fingering” of DNAPL in vadose zone“The spill question”DNAPL below the water tableLNAPL behavior at water tableFactors affecting NAPL movementNAPL Movement in Porous MediumWetting vs. non-wetting fluidsInterfacial tension force balanceInterfacial tension effectsInterfacial tension valuesInterfacial tension values (continued)Force balance on fluid elementForce balance on fluid elementLNAPL on waterWater on DNAPLCapillary pressureCapillary pressureCapillary force balanceRelationship between PC and sRelationship between PC and sPC depends on NAPL and porous mediumPC vs. water saturationPC vs. water saturation for PCEImplications of entry pressureNAPL mobilizationMobilization of entrapped NAPLMobilization of entrapped NAPLResidual saturation vs. Bond and Capillary NumbersResidual saturation vs. Capillary NumberHydraulic gradient to mobilize NAPLCan NAPL be mobilized?Can NAPL be mobilized?Lecture 4Multi-phase Flow in Porous MediaNon-aqueous phase liquidsNAPL = Non-Aqueous Phase LiquidDNAPL = Dense Non-Aqueous Phase LiquidChlorinated solventsCoal tarLNAPL = Light Non-Aqueous Phase LiquidPetroleum fuelsTypical densities of NAPLsDNAPLs LNAPLsPCE 1.62 Xylene 0.86Carbon tetrachloride 1.59 Toluene 0.87Chloroform 1.48 Benzene 0.88TCE 1.46 Gasoline 0.72-0.761,1,1-TCA 1.34 Jet fuel 0.76Methylene chloride 1.33 Diesel fuel 0.84-0.85Coal tar 1.01-1.05 Fuel oil 0.87-0.95See Figure XI.a in Friedrich Schwille, 1988. Dense Chlorinated Solvents in Porous and Fractured Media, Model Experiments. Lewis Publishers,Chelsea, Michigan.Laboratory experiments with PCE DNAPL in unsaturated and saturated mediaSee Figure 1 in M.M. Poulsen and B.H. Kueper, 1992. A Field Experiment to Study the Behavior of Tetrachloroethylene in Unsaturated Porous Media. Environmental Science &Technology, Vol. 26, No. 5, Pp. 889-895, May 1992.PCE penetration at Borden aquifer field siteShows structure (even very subtle) of porous medium may affect lateral movementSee Figure 2 in M.M. Poulsen and B.H. Kueper, 1992. A Field Experiment to Study the Behavior of Tetrachloroethylenein Unsaturated Porous Media. Environmental Science &Technology, Vol. 26, No. 5, Pp. 889-895, May 1992.PCE penetration at Borden aquifer field siteIn seemingly uniform sand, subtle structure caused lateral movementCharacter of release affected very penetrationFormation of residual and dissolutionLow permeability lenses may soak up NAPL, cause lateral spreadingResidual contamination in unsaturated soil acts as long-term source of contaminants to dissolve into infiltrating rechargeFormation of residual and vapor plumeAt same time, solvent may be vaporizing. Solvent vapor is denser than air, and will sink in unsaturated zone and can contaminate ground water.Typical residual saturation valuesMaterial Residual saturation, srTCE 0.15 – 0.2PCE 0.002 – 0.20Gasoline 0.12 – 0.60voids of volumeNAPL of volumesr=Residual saturation, (dimensionless)Good rough guess: sr= 0.2Source: Cohen, R. M., and J. W. Mercer, 1993. DNAPL Site Evaluation. C.K. Smoley, Boca Raton, Florida. After NAPL passes through soil, fraction remains behind as a residual.See Plate 1 in S.H. Conrad, J.L. Wilson, W.R. Mason, andW. J. Peplinski, 1992. Visualization of Residual Organic Liquid Trapped in Aquifers. Water Resources Research. Vo. 28, No. 2,Pp. 467-478. February 1992.Residual in etched glass “porous medium”Lab studies generally confirm the conceptual model of residual formation. Conrad et al. (1992) conducted two types of experiments. The first is show here. They created a planar porous medium with two plates of etched glass. Filled the pore space with distilled water, then injected red-dyed Soltrol (an oil) from the left (Figure a), and re-flooded with distilled water flowing from the right (in the opposite direction of the injection) (Figure b).See S.H. Conrad, J.L. Wilson, W.R. Mason, and W. J. Peplinski, 1992. Visualization of Residual Organic Liquid Trapped in Aquifers. Water Resources Research. Vo. 28, No. 2, Pp. 467-478. February 1992.Residual formed different shapes: singlets, doublets, etc. Physical nature of residual depended upon balance between viscous and inertial forces on residual.See Fig.5 in S.H. Conrad, J.L. Wilson, W.R. Mason, and W. J. Peplinski, 1992. Visualization of Residual Organic Liquid Trapped in Aquifers. Water Resources Research. Vo. 28, No. 2, Pp. 467-478. February 1992.Residual in sand porous mediumSecond experiment by Conrad et al. (1992) was to construct small water-saturated columns of sand, into which they injected a styrene monomer. The styrene solidified after injection. They then dissolved away the sand with hydrofluoric acid, leaving styrene casts.See Fig.6 in S.H. Conrad, J.L. Wilson, W.R. Mason, and W. J. Peplinski, 1992. Visualization of Residual Organic Liquid Trapped in Aquifers. Water Resources Research. Vo. 28, No. 2, Pp. 467-478. February 1992.NAPL will remain essentially immobile at residual concentration. It dissolves very slowly (for example, TCE has a solubility of 1,100 mg/L) and thus takes a long time to “wash” out. Infiltrating rain water dissolves out NAPL, becomes itself contaminated, and is likely to carry contaminants to the water table and ground-water system. The residual also acts as a long-term source of vapor in the subsurface.Small amount of PCE releasedLater time: more PCE released,Formation of fingersSee Figures XVIII.a and b in Friedrich Schwille, 1988. Dense Chlorinated Solvents in Porous and Fractured Media, Model Experiments. Lewis Publishers, Chelsea, Michigan.PCE penetration into unsaturated porous mediumIf small amount of NAPL is spilled to the ground (top figure), it will soak into the soil by capillary action and due to gravity. In the soil, it will reach residual concentration and travel no further as DNAPL. Thus, a small spill will not penetrate to the water table.If more is spilled on the same spot (bottom figure), NAPL will be displaced downward and the “front” of NAPL residual will advance.“Fingering” of DNAPL in vadose zoneSee Friedrich Schwille, 1988. Dense Chlorinated Solvents in Porous and Fractured Media, Model Experiments. Lewis Publishers, Chelsea, Michigan.An areally distributed spill may not infiltrate as a contiguous mass but tends to advance as