H54C-01 INVITED
Controlled placement of polyelectrolyte modified engineered nanomaterials in the subsurface: Correlating modifier layer properties and geochemistry with mobility
Nanoscale zerovalent iron (NZVI) particles are used for in situ remediation of contaminated groundwater. To be effective, NZVI must be reactive with target groundwater contaminants, and be mobile in the subsurface to enable placement. This is typically achieved with surface modification with polymers or surfactants. Concern over the potential toxicity of engineered nanomaterials is further motivation to understanding the factors controlling their mobility. Several hydrogeochemical properties control the mobility of nanoparticles in the subsurface including, the particle surface properties and polydispersity, groundwater geochemistry, the porewater velocity, and the grain size/type and the degree of sorting. Dynamic light scattering, sedimentation, and column transport studies were conducted on polyelectrolyte-modified NZVI under a variety of hydrogeochemical conditions, and with varying polydispersity, to determine the effect of each parameter on their mobility in water-saturated sand columns. It was shown that modifiers that provide electrosteric repulsion, which are less sensitive to changes in ionic strength, were needed to provide good mobility in saturated sand columns under geochemical conditions representative of groundwater conditions. At low particle concentration, mobility decreased with decreasing particle grain size, flow velocity, and in the presence of clay fines. This is consistent with colloid theory predictions. Mobility experiments conducted at high particle concentrations (1 to 6 g/L) and polydisperse samples, and at geochemical conditions relevant to particle injection (10 mM Na+, a pore water velocity of 3.2 x 10-4 m/s) showed that more polydisperse samples containing larger particles (several hundred nanometers) are less mobile than monodisperse samples containing only small particles (~100nm). The degree of deposition reversibility also decreased as polydispersity increases. Because the adsorbed layer properties and thus particle-collector interactions of polymer-modified NZVI are similar for all particle sizes used, the higher deposition rate of the polydisperse samples is attributed to increased aggregation from the magnetic attractive forces between particles which increase with r6. This study emphasizes an important role of geochemistry and particle-particle interaction (aggregation) in the transport of concentrated NZVI dispersions in porous media.
H54C-02
Development and Assessment of Oil-in-Water Emulsions for Encapsulation of Reactive Iron Particles for Subsurface Delivery
Reactive iron particles hold promise for use in the destruction of contaminants in the subsurface environment. Application of these nano- to submicron-scale particles, however, may be limited by poor subsurface transport and non-uniform distribution of the reactive material. Delivery issues are particularly important when evaluating the efficacy of iron-based technologies for treatment of dense non-aqueous phase liquid (DNAPL) source zones. Current approaches for the delivery of reactive iron particles within DNAPL source zones are hindered by particle agglomeration, flow bypassing, and presence of non-target reactions. Encapsulation of the reactive particles within an oil-in-water emulsion is a novel approach that may overcome these limitations. Development of kinetically-stable, iron-laden, oil-in-water emulsions commenced by identifying surfactant-based coatings to increase the stability of commercially-available iron particles within non-polar organic phases (e.g., soy oil). A phase inversion technique was employed to disperse approximately 10% wt of the iron-laden, organic phase within a continuous aqueous phase containing nonionic emulsifiers. Emulsions were designed to ensure emulsifier proportions yielded hydrophilic-lipophilic balances affiliated with oil-in-water emulsions. Micrographs of the oil-in-water emulsions suggest that the average diameter of the oil droplets is approximately one micrometer. The presence of iron within oil droplets was confirmed in the micrographs and supported by an absence of iron agglomeration within the continuous phase. Bulk characteristics of each emulsion (density and viscosity) were used in conjunction with interfacial tension measurements in total trapping number analyses to assess the propensity of these emulsions to mobilize an entrapped trichloroethene (TCE)-DNAPL. Results suggest that the emulsions described herein should not cause significant mobilization of entrapped TCE-DNAPL in fine-to-medium grain sandy media. Column experiments are being conducted to evaluate the transport of these emulsions through sandy media. Preliminary results from experiments with iron-free emulsions suggest conductivity reductions occurring during emulsion flushing are not the result of extensive pore-clogging but rather are due to viscosity changes (emulsion viscosities range from 2 to 10 cP). Current efforts are focused on assessing and comparing both transport and reaction of commercially available iron particles and iron-laden emulsions within sandy porous media.
H54C-03 INVITED
Effects of Aging and Colloids on Iron Nanoparticle Transport in Groundwater
Most studies of manufactured nanoparticle transport in aquifer materials have used "fresh" particles in homogenous solutions. However, the role of particle aging and the presence of other colloidal materials may have a significant impact on mobility. We have examined both of these processes using reactive nano-iron particles (RNIP), dynamic light scattering, column studies, and deep-bed filtration theory. In the absence of colloidal materials, RNIP are rapidly removed from porous media by coagulation and attachment to aquifer materials (due in part to their magnetic properties). However, upon aging (in the presence of oxidants ranging in strength from water to carbon tetrachloride), the outer shell of the RNIP will be converted to oxides that are less magnetic and have very different surface charges that result in different sticking coefficients. The presence of colloids (both organic and inorganic) may modify the surfaces of aquifer materials and/or bind with the nanoparticles, and as a consequence, aggregation and sticking coefficients of the nanoparticles can also be significantly altered.
H54C-04
Modeling the Coupled Effects of Pore Space Geometry and Velocity on Colloid and Nanoparticle Transport and Retention
Colloid and nanoparticle retention in porous media has traditionally been assumed to be controlled by chemical interactions between the particles and the solid interface. The influence of system hydrodynamics in classical retention models is only considered for the rate with which particles strike solid surfaces. Recent experimental and theoretical work, however, has demonstrated that hydrodynamics also plays an important role in particle retention under unfavorable attachment conditions. In this case, a balance of adhesive and hydrodynamic forces and torques acting on particles near solid surfaces indicates that particle retention will occur only in low velocity regions that are controlled by the pore space geometry and primarily occur in the smallest regions of the pore space. Computer models that consider the average pore-water velocity in the porous medium and a single attachment rate coefficient are therefore not always adequate to describe retention processes, which frequently produce depth-dependent particle retention profiles (non-exponential). In this work, we highlight various computer models that can be used to account for particle retention in the smallest regions of the pore space. The models may be based on: (i) physical and chemical nonequilibrium; (ii) dual permeability; and (iii) stochastic stream tubes. Applications, implications, and limitations of the various models to characterize particle transport and retention will be demonstrated and discussed.
H54C-05
Investigation of n-C60 Nanoparticle Transport and Retention in Saturated Porous Media
Buckminster fullerene (C60), a molecule composed of 60 carbon atoms arranged as a spherical cage, has recently gained wide application in many commercial products. Based on its widespread use, it is likely that C60 will be released into the environment during manufacture, transportation, and/or application. Possessing negligible solubility in water, C60 is capable of acquiring charge and form highly stable nano-scale aggregates (n-C60) in aqueous systems. While the toxicity of C60 nanoparticles has recently been explored, our current understanding of n-C60 fate and transport in subsurface environments is still quite limited. In this study, experimental and mathematical modeling studies were performed to investigate the transport and retention of n-C60 nanoparticles in water-saturated porous media. A series of transport experiments was conducted at several pore-water velocities in glass columns packed with various size fractions of Ottawa sand. A mathematical model that incorporates non-equilibrium attachment kinetics and a maximum retention capacity was developed to simulate n-C60 transport and retention in porous media. The numerical model is able to simulate both the effluent concentration and particle retention profiles. Experimental and simulation results suggest that the retention of n-C60 aggregates is strongly dependent on porous media surface area and flow rate. The extended Derjaguin-Landau-Verwey-Overbeek (DLVO) theory, which includes van der Waals, electro- static repulsion, and hydrophobic interaction forces, was used to evaluate potential mechanisms governing n- C60 attachment. This analysis suggests that a sizable energy barrier exists between n-C60 aggregates and the Ottawa sand surface, with a small secondary minimum attraction region. Attachment rate coefficients derived from secondary energy minimum theory were found to be in close agreement with those fit to retention data. . Maximum retention capacity was found to be correlated with system Peclet number and porous medium grain size.
H54C-06
Dynamic Retention and Release of Nanoparticles in Unsaturated Porous Media During Multiple Drying/Wetting Cycles
To better understand the transport of nanomaterials in the vadose zone, a series of experiments were conducted to study the retention and release of nanomaterials in unsaturated porous media during drainage and imbibition. The research makes use of a custom-designed membrane-based automatic system which allows the continuous real-time measurement of nanomaterials retained by unsaturated porous media during multiple drainages. As nanomaterials are retained in the porous media during drainage, their concentration in the pore solution decreases. For the experiments to be discussed, the concentration decrease of nanomaterials in the pore solution was tracked during multiple drainages, along with the simultaneous tracking of porous media capillary pressure and saturation. A continuous mass balance was used to calculate the mass of nanomaterials retained as a function of saturation. The release and re-retention of previously retained nanomaterials has been investigated through a series of wetting/drying cycles with changing nanomaterial concentrations. The air-water interfacial area formed during each drainage of multiple wetting/drying cycles was measured as a function of saturation to investigate the role of the air-water interface, and to elucidate possible mechanisms governing the unsaturated transport of nanoparticles. Manufactured nanoparticles selected for this study included tin oxide (SnO2), titanium dioxide (TiO2) and polystyrene latex. The dependence of retention/release on nanoparticle properties will be discussed.
H54C-07
Transport and Retention of Engineered Nanoparticles in Porous Media Column Experiments
The transport and interactions of engineered nanoparticles in the subsurface are key factors that govern environmental impact of nanoparticles, as well as their human and ecological exposure pathways. Three aspects of nanoparticle fate in the subsurface environment are considered here. First, a new analytical technique to quantitatively measure concentrations of nanosized iron oxide and copper oxide particles in solution, based on UV-vis spectroscopy is presented; this technique complements optical measurements. We then demonstrate that sorption of these nanoparticles on various artificial and natural porous media is essentially negligible, within experimental error; the addition of humic acids to the porous media marginally increases sorption, but yields remain less than 5%. Finally, the transport and distribution of nanoparticles in water-saturated columns packed with porous media is examined. We find that transport behavior is influenced strongly by the method of nanoparticle introduction at the column inlet. Application of dry nanoparticles at the column inlet, followed by periodic flow of water, leads to highly limited nanoparticle migration. Use of continuous water flow causes the nanoparticles to undergo migration, settling and retention, with small amounts of elution from the column outlet. In contrast, nanoparticles that enter the column in a flowing suspension are transported through the column with virtually no retention effects.