Soil Properties and Design Factors Influencing Free-Phase Hydrocarbon Cleanup

Abstract

Light nonaqueous phase liquid (LNAPL) saturation and movement in the subsurface are controlled by soil capillary characteristics, permeability, and fluid properties. Where free-product occurs in monitoring wells, LNAPL saturations in the formation vary significantly as a function of the observed thickness in the monitoring well and soil type. Fine-grained soils generally exhibit lower LNAPL saturations than coarse-grained material for the same observed thickness in a monitoring well. MAGNAS3 [MAGNAS3. HydroGeoLogic, Inc., Herndon, VA, 1992; Huyakorn, P. S.; Panday, S.; Wu, Y. S. J. Contam. Hydrol. 1994, 16, 190−130], a three-dimensional, finite-element model that can simulate movement of three active phases (air, water, and LNAPL), was used to investigate LNAPL recovery in three different soil types and using several recovery designs to examine the effect of these factors on LNAPL recovery. The results of this analysis show that, because the relative mobility of LNAPL decreases with decreasing saturation and because the intrinsic permeability of fine-grained soil is less than that of coarse-grained soil, free-product pumping or skimming has less likelihood of success in fine-grained soil. Recovery in fine-grained soils was minimal, with significant reductions in LNAPL saturation occurring only within about 10−15 ft of the well. Recovery of LNAPL in coarse-grained soils was simulated to be much more successful, with ap proximately 95% of the original hydrocarbon recovered through fluid pumping. The results of the modeling further shows that, for any soil type, recovery decreases LNAPL saturation and mobility near the well with the effect diminishing with distance. Therefore, a zone of decreased LNAPL permeability is formed near recovery wells that acts to impede additional recovery from greater distances. Increases in the hydraulic recovery rate (i.e., not considering volatilization) can be realized in all of the soils studied through vacuum enhanced free-product recovery.

Numerical investigation on light non–aqueous phase liquid flow in the vadose zone considering porosity effects on soil hydraulic properties

“…It can be concluded that, given the same leak source, coarse-grained soil (sand) could cause a much larger LNAPL volume than fine-grained soils (silt F I G U R E 1 6 Influence of layering obtained through Series B4: vertical front depth and total light non-aqueous phase liquid (LNAPL) volume and clay). This is mainly attributed to the different intrinsic permeabilities of these soils, as explained by some previous researchers (Beckett & Huntley, 1998;.…”

“…For example, in the design of petrol and gas stations as well as industrial sites, heavy compaction is desired to control potential LNAPL contamination. Furthermore, soil type is generally considered in the studies on the flow and distribution of LNAPL (Beckett & Huntley, 1998; Lari et al., 2018; Oostrom & Lenhard, 1998). This is not sufficient because the porosity of a given soil can vary widely, depending on the geological history and loading path.…”

“…For instance, Lenhard and Parker (1990) found the influence of water/LNAPL retention behavior on LNAPL flow upon the leakage of LNAPL. Beckett and Huntley (1998) simulated the LNAPL remediation in coarse-grained and fine-grained soils. They found a higher remediation efficiency in the coarse-grained soil due to its higher intrinsic permeability.…”

Characteristics of light nonaqueous phase liquid recovery in the presence of fine‐scale soil layering

“…They used Monte Carlo simulation to compare spatial moments of the LNAPL saturation in the plume where permeability was lognormally distributed and autocorrelated while constitutive relationships were also dependent on permeability. Beckett and Huntley [1998] examined different homogeneous soil classes, while Kaluarachchi [1996] included stochastic spatial variability of permeability in comparing recovery schemes with those which assume homogeneous conditions [ Cooper et al , 1995, 1998]. The model used by Kaluarachchi [1996] was a vertically averaged model, so only lateral heterogeneity was considered.…”

Evaluation of LNAPL distribution and mobility in heterogeneous porous media under water table fluctuations

“…This discrepancy 303 between different methods may be also due to poor well development or other 304 artifacts, as it has been documented in the case of comparisons between slug tests and 305 pumping tests in groundwater systems (Butler and Healey, 1998). The spatial scale of 306 the selected methods is determined by their radius of capture, typically larger for 307 coarser-grained sediments (Beckett and Huntley, 1998). 308 at research area B), although it showed higher LNAPL saturations and b n .…”