Field Investigations of Natural Attenuation and Trench Application in a Heterogeneous Shallow Contaminated Aquifer with Free-Phase LNAPLs
This study focused on the evaluation of natural attenuation of groundwater and the applicability of trench method for intercepting the contaminant plume in a highly contaminated aquifer with dissolved hydrocarbons and free phase LNAPLs. Several times of groundwater sampling and subsequential geochemical analyses were carried out over a period of 6 years. The selected aquifer is composed of a high permeability gravel layer that acts as a preferential pathway of LNAPLs when the groundwater fluctuates by 1-2 m. The results of field investigations show that the NA of hydrocarbon contaminants had sufficient ability to mineralize the contaminants but the remediation rate was lower than that observed at the other sites because the LNAPL source has not completely removed yet. After the installation of the trench, long term monitoring of contaminant concentration indicates that the expanding of the contaminant plume is greatly controlled by the trench. The Mann-Kendall trend analyses show that BTEX contaminant plume in the source area has reached "STABLE" with high contaminant concentration for the whole period but the contaminant plume in down-gradient area was "DECREASING" with low contaminant concentrations due to the role of the trench. Approximately 98% of the contaminants were intercepted by the trench and the intercepted rates of the contaminant were approximately 1,469 ~ 2,393 g/year.
Effect of Surface Wettability on DNAPL Migration in a Rough-Walled Fracture
To assess the influence of the surface wettability on DNAPL migration in a rough-walled fracture, DNAPL migration experiments were carried out with the rough-walled glass fracture, where intermediate and oil-wetted surfaces were made by the exposure by gasoline and creosote. Migration paths and processes of DNAPL highly depend on the surface wettability. DNAPL migrated through water-wetted surface with snap-off (disconnected) migration pattern, but it passed in a continuous form on the intermediate-wet and oil-wet surfaces. DNAPL migration passed fastest through the fracture with intermediate-wet surface, where a near 90„a contact angle makes the capillary pressure no longer dominant compared to gravity and viscous pressure of flowing groundwater. The migration was highly delayed in the hydrophobic surfaces due to the fact that the capillary barrier of larger apertures causes DNAPL to migrate through the smaller apertures, as opposed to the migration characteristics in water-wetted surface. This indicates that DNAPL recovery, based on waterflood, will be inefficient in rough-walled fracture with oil-wetted surfaces. However, as the flow becomes nonlinear, the DNAPL migrated downward along the groundwater flow in all surface wettability conditions, where the inertial pressure, which is proportional to the square of groundwater velocity, overwhelms the capillary pressure. In the nonlinear flow regime, DNAPL migration process is not surface wettability-dependnt, which gives an important implication of NAPL recovery from rough-walled fractures with oil-wetted surfaces, otherwise NAPL tends to be highly retained in the linear flow regime.
Characterization of Gas Flow During the Slow Expansion of Discontinuous Gas due to DNAPL Partitioning
Recent studies have shown that the partitioning of volatile DNAPL constituents to discontinuous gas in the vicinity of DNAPL sources results in the expansion and mobilization of the gas phase. This spontaneous expansion and mobilization of the gas phase can change the mass transfer from the surface of DNAPL pools, and produce aqueous-phase DNAPL constituent concentrations different from those expected based on typical DNAPL-water only models. Although macroscopic measurements of gas-phase extent have been presented in the literature along with qualitative observations of pore-scale effects, little quantitative information is available concerning the rate of gas generation or the distribution of gas clusters produced by this mechanism. This study used small-scale (10 cm x 8 cm x 1 cm) flow cells to carefully measure gas expansion rates above tetrachloroethene (PCE) and trans-1,2-dichloroethene (tDCE) pools by directly measuring water displaced from the cell during expansion. These rates were found to be very slow, and capable of producing discontinuous gas flow in most sands. Additional air-injection experiments were performed at these very slow rates to characterize the size of the resulting disconnected gas clusters in various porous media. A novel combination of pressure measurements and image analysis was used to measure the height of gas clusters at the moment of fragmentation and mobilization. These experimental results suggest that any macroscale effect of the spontaneously expanding gas will depend on the relative rates of dissolution and expansion between gas clusters above and connected to the DNAPL surface, respectively. They also suggest that considerable mass transfer enhancement may occur in finer media, where longer gas clusters may remain attached to the DNAPL surface.
Removal of Trichloroethylene and Heavy Metals by Zerovalent Iron Nanoparticles
Heavy metals combined with chlorinated solvents are one class of mixed waste found at various hazardous waste sites in North America. Nano zerovalent iron (nZVI), an emerging technology, is being successfully used for treating chlorinated solvents and heavy metals independently, however comparatively little research has investigated the remediation of the wastes when they are present in the same mixture. The remediation of trichloroethylene (TCE)/heavy metal waste mixtures via nZVI has been investigated in the present study. Results suggest that some metals are reduced by nZVI to their zerovalent state and thus precipitate on nZVI particles. This improves the contaminant removal performance of nZVI by forming bimetallic iron nanoparticles. Other metals are directly precipitated or adsorbed on the nZVI particles in their original oxidation state and are rendered immobile. In some cases the presence of the heavy metals in the waste mixture enhanced the dechlorination of TCE while in other cases it did not. This study suggests that nano zerovalent iron particles can be effectively used for the remediation of mixed contamination of heavy metals and chlorinated solvents. Results have been supported by a variety of techniques including X-ray photoelectron spectroscopy (XPS) analysis.
Mobility and modeling of nano-scale zero valent iron in porous media with residual NAPL
Nano-scale zero valent iron (nZVI) has shown significant potential for source zone remediation in laboratory
experiments but unresolved challenges remain prior to widespread field application. One significant challenge
is limited nZVI subsurface mobility. Polymer modification for particle stability has provided some novel
generations of nZVI, the most promising of which use environmentally benign polymers that limit the
aggregation of nZVI particles during synthesis and provide delivery solution stability for several days. Many
efforts investigating the mobility of polymer modified nZVI have emphasized environmentally relevant porous
media conditions (ionic strength, presence of strong cations, heterogeneity and natural soils.), however much
of the work published in the literature has focused on flow rates that are greater than expected at remediation
sites and nZVI dosages lower than those applied in the field. In this study a series of one dimensional column
experiments were performed under conditions representative of groundwater remediation sites (eg:
representative flow rates, DNAPL at residual saturation) to improve understanding of the conditions that control
mobility. The ability of a numerical simulator to predict observed laboratory behaviour is also assessed.
The Wettability of a Multi-Component DNAPL on Quartz and Iron Oxide Sands
Dense nonaqueous phase liquids (DNAPLs) released to the subsurface often contain a variety of chemical constituents, via either co-disposal or intentional modification to increase their industrial efficacy. These additional constituents are often surface active compounds (surfactants)that partition to soil surfaces. The role that these surface active compounds that sorb to soil surfaces have on DNAPL migration is still poorly understood despite an increasing amount of work in the area. Most studies have focused on the role surface active chemicals play in altering the wettability of quartz sands. This research aims to extend the understanding of multi-component DNAPL transport to other porous media and under a variety of pH conditions. Specifically, the objective of this study was to compare the changes in the wettability of quartz and iron oxide sands in a tetrachloroethylene (PCE)/water system spiked with dodecylamine, a representative cationic surfactant. Wettability was assessed through: (i) contact angles measured on representative quartz and iron oxide-coated plates as well as (ii) contact angles measured directly on sands using an Axial Drop Symmetrical Analyzer apparatus; and (iii) capillary pressure-saturation relationships obtained via multi-step outflow experiments. In addition, two-dimensional sandbox experiments explored the influences of iron oxide and quartz sands on multicomponent DNAPL migration. Results suggest that quartz and iron oxide-coated sands exhibit different wetting characteristics under similar subsurface conditions.
Numerical Modelling of Smouldering Combustion as a Remediation Technology for NAPL Source Zones
Smouldering combustion of non-aqueous phase liquids (NAPLs) is a novel concept that has significant potential for the remediation of contaminated industrial sites. Many common NAPLs, including coal tar, solvents, oils and petrochemicals are combustible and capable of generating substantial amounts of heat when burned. Smouldering is a flameless form of combustion in which a condensed phase fuel undergoes surface oxidation reactions within a porous matrix. Gerhard et al., 2006 (Eos Trans., 87(52), Fall Meeting Suppl. H24A) presented proof-of-concept experiments demonstrating the successful destruction of NAPLs embedded in a porous medium via smouldering. Pironi et al., 2008 (Eos Trans., 89(53), Fall Meet. Suppl. H34C) presented a series of column experiments illustrating the self-sustaining nature of the NAPL smouldering process and examined its sensitivity to a variety of key system parameters. In this work, a numerical model capable of simulating the propagation of a smouldering front in NAPL-contaminated porous media is presented. The model couples the multiphase flow code DNAPL3D-MT [Gerhard and Grant, 2007] with an analytical model for fire propagation [Richards, 1995]. The fire model is modified in this work for smouldering behaviour; in particular, incorporating a correlation of the velocity of the smouldering front to key parameters such as contaminant type, NAPL saturation, water saturation, porous media type and air injection rate developed from the column experiments. NAPL smouldering simulations are then validated against the column experiments. Furthermore, multidimensional simulations provide insight into scaling up the remediation process and are valuable for evaluating process sensitivity at the scales of in situ pilot and field applications.
Enhanced transport of biodegradable polymer-coated nanoiron particles in sand columns
The use of nanoscale zerovalent iron has shown promise as a technology for remediation of subsurface contamination by chlorinated solvents. However, the delivery of nanoiron particles to target contaminated subsurface zones is hindered by the aggregation of particles due to magnetic attraction. To overcome the limitations of aggregation and increase nanoiron mobility in porous media, nanoiron particles have been coated with various polymers. Polymer adsorption onto nanoiron particles provides electrosteric stabilization, increases the mobility, and decreases the attachment onto the soil surface. Various polymers were investigated in this study, including carboxylmethyl cellulose (CMC) and guar gum, both of which are biodegradable. In sand column experiments the transport of nanoiron particles was investigated as a function of type of electrolyte, ionic strength, flow velocity, and nanoiron particle concentration. Settling curves showed the enhanced stability of polymer-coated nanoiron particles compared to bare commercial nanoiron particles (bare RNIP-10DS). A newly developed nanoparticle transport numerical model was used to quantify the attachment efficiency, as well as investigate dominant nanoparticle transport and removal mechanisms. Finally the particle-collector interaction energy was predicted using DLVO (Derjaguin-Landau-Verwey-Overbeek) theory.