404
Ind. Eng. Chem. Res. 2011, 50, 404–411
Fe/Ni Bimetallic Particles Transport in Columns Packed with Sandy Clay Soil S. Harendra* and C. Vipulanandan Department of CiVil and EnVironmental Engineering, UniVersity of Houston, Houston, Texas 77204-4003, United States
Over the past years, there has been an increase in concern regarding colloidal particle transport in soil. To date, there exists limited research on the fate and transport of these particles in the environment. In this study, a series of laboratory miscible displacement experiments were done using Fe/Ni particles. There is not enough data available in the literature about the transport mechanism of Fe/Ni particles. Although Fe/Ni particles have high reactivity with chlorinated solvents and metal ions, their transport properties are poorly understood, and the other uniqueness of this study was that it was done in clay soil and soil contaminated with perchloroethylene (PCE). These particles undergo collisions with soil grains and are subject to adsorption, dispersion, and retardation. Fe/Ni particles transport was studied in detail, and it was found that all mechanisms involved in particles transport in soil, which is the major contribution to environmental remediation. The Fe/Ni particle concentrations that were used were similar to those used in field studies for in situ treatment of chlorinated compounds. Fe/Ni particles were produced using solution method and characterized using SEM and XRD. The SEM micrograph of the Fe/Ni particles showed that the particles gave the mean size of Fe/Ni nanoparticles as a bimodel distribution of 249 and 3410 nm and were spherical in shape. Retardation factor and dispersion coefficients of Fe/Ni colloids in various clay soils were determined. The miscible displacement breakthrough curves indicated chemical nonequilibrium transport. The chemical nonequilibrium, miscible displacement model (two-site kinetic model) was used to describe the column breakthrough curves. This model resulted in excellent descriptions of the data. 1. Introduction Studies at many sites in the U.S. have led to the recognition that even small chlorinated solvent source zones can generate extensive dissolved plumes, and traditional pump-and-treat techniques designed to provide treatment of contaminant plumes often require long-term operation. The aggregate financial burden for site cleanup is colossal. Fe-Ni particles hold a potential to cost effectively address some of the challenges of site remediation. Delivering zerovalent Fe-Ni particles to contamination sources in soil and groundwater is an interesting challenge for in situ remediation. The particles are typically injected as an aqueous slurry, and they must travel distances of meters through soil or aquifers to reach the contamination source zones.1-5 In this study, a detailed study of nano and micro Fe-Ni transport through sandy columns was performed. DNAPLs, or dense nonaqueous phase liquids, such as chlorinated solvents and creosote, are groundwater contaminants commonly encountered throughout industrial areas of North America. Fine Fe/Ni bimetallic particles represent a new generation of environmental remediation technologies that could provide cost-effective solutions to some of the most challenging environmental clean-up problems such as DNAPLs.6-12 Those particles have large surface areas and high surface reactivity.12-14 This leads to greater remediation of chlorinated compounds present in soil and aquifer. Newer techniques such as the injection of reactive slurries or suspended solids may serve to overcome some of the limitations so that contaminants at greater depths can be reached and more directly targeted.15-18 Although Fe-Ni particles have high reactivity with chlorinated solvents and metal ions, their transport properties are poorly understood. These particles undergo collisions with soil grains and are subject to adsorption, dispersion, and retardation. * To whom correspondence should be addressed. Tel.: (713) 8902503. E-mail:
[email protected].
The transport and deposition of colloidal particles during flow through porous media have been studied extensively using wellcharacterized model systems.19-24 Filtration theories have been developed from theoretical considerations and experimental results for the transport of monodisperse microspheres through columns packed with spherical grains. However, application of filtration theory to colloid transport in subsurface systems seems limited by the fact that most natural porous media have wide pore and particle size distributions, complex pore geometry, and rough surfaces with considerable surface charge heterogeneities. Moreover, current theories often do not accurately predict surface effects even in simple model systems.24 Evaluating the mobility of colloids in a particular subsurface porous medium therefore strongly relies on empirical results. Early research was driven by the need to understand the performance of deep bed filters used in chemical engineering and wastewater treatment.22,25-28 More recently, the focus of interest has shifted to the transport of mobile colloids in natural subsurface porous media. The primary mechanisms controlling the transport of colloidal particles in subsurface porous media are particle advection, dispersion, and deposition (filtration). These are influenced by the surface chemical characteristics of the natural porous media and colloids, size and morphology of colloidal particles and granular porous media, solution chemistry, and the fluid flow field.25-28 There were studies conducted on iron transport on model soils and glass beads.25-30 Because the pore size is much bigger in packed models soils and glass beads, particles move much more easily, and there is no aggregation of fine particles. This is not the real scenario in fields where any remediation sites have a certain amount of clay present, which affects the characteristics of soil and mobility of particles. In this study, Fe/Ni particles transport was studied in sandy clay soil. Bimetallic particles were transported through contaminated
10.1021/ie101546k 2011 American Chemical Society Published on Web 11/23/2010
Ind. Eng. Chem. Res., Vol. 50, No. 1, 2011
405
Table 1. Transport Parameters of Fe/Ni Particles in Soil Column
Fe/Ni Fe/Ni Fe/Ni Fe/Ni Fe/Ni
(pure sand) (100% kalonite) (100% bentonite) (sand 82.5%, kao 17.5%) (sand 95%, ben 5%)
R
D (cm/ min2)
mean square error (MSR)
4.43 16.83 infinite 17.38 35.61
2.26 1.05 0 2.45 5.47
0.002 0.002 0 0.11 0.0005
(perchloroethylene (PCE)) and uncontaminated soil. The retardation factor (R) and dispersion coefficient (D) of particles were determined. The overall objective of this study was to determine the transport parameters of Fe/Ni bimetallic particles in packed soil. The specific objectives are as follows: (1) determination of transport parameters of Fe/Ni particles in soil, kaolinite, and bentonite in column studies, (2) the concentration profiles of Fe/Ni particles along the depth of the column studies after the breakthrough studies, and (3) Tthe efficiency of Fe/Ni particles in degrading PCE in contaminated soil.
Figure 1. Grain size distribution of sand, kaolinite, and bentonite.
2. Materials and Methods UH biosurfactant was produced in the lab.31-33 Cationic surfactant cetyltrimethyl ammonium bromide (CTAB) was purchased from Sigma-Aldrich, St. Louis, MO. Iron(II) chloride (98%) and sodium borohyride powder (98%) were also purchased from Sigma-Aldrich (St. Louis, MO) for producing the iron nanoparticles using the solution method. Nickel(II) chloride hexa hydrate (Reagent plus), iron(11) sulfate heptahydrate (99%, certified A.C.S.), and ferric chloride were also purchased from the Sigma Aldrich Co. for the preparation of iron and Fe-Ni bimetallic particles using the solution method. The blasting sand was obtained from TEC minerals of Eagle Lake, TX. Sand particles are made mostly of quartz and feldspar and have a very stable tetrahedral structure. The particle size ranged from 0.07 to 5 mm. The specific gravity was 2.65. Clays have marked effects on both the physical and the chemical properties of the soils because of their comparatively large surface area and permanent negative surface charge. Hydrite-R kaolinite was purchased from D & F Distributing Inc. as its properties are listed in Table 1 as specified by manufacture’s data sheet. The isomorphous substitution of Al(III) for Si(IV) and Mg(II) substituting for Al(III) gives rise to a net negative charge on its surface. The species of SiO2, Al2O3, H2O, Fe2O3, TiO2, MgO, CaO, K2O, and Na2O are present in the kaolinite clay. The specific surface of kaolinite clay is in the range of 5-30 m2/g, and their cation exchange capacity is in the range of 15-20 mequiv/kg. Pure bentonite was used as packing medium. The packing medium consists of sand with a specific gravity of 2.65, kaolinite containing a specific gravity of 2.623, and bentonite containing a specific gravity of 2.89. Grain size distribution of soil samples was determined using sieve analysis for sand and hydrometer tests (ASTM D 422-63) for kaolinite and bentonite. The grain size distribution is shown in Figure 1. Characterization. To characterize and analyze the particles and/or surfactants, the following test methods were adopted. a. Scanning Electron Microscopy (SEM). The particles synthesized using the solution method were characterized using the JEOL 2000FX scanning electron microscope (SEM) and Siemens D5000 powder X-ray diffraction (XRD). The bimetallic particles were deposited on carbon grid for SEM characterization. Morphology of the particles was determined using the SEM image. The SEM micrographs of Fe-Ni bimetallic particles are shown in Figure 3. The particles were spherical, and the size varied from 100 nm to 3 µm.
Figure 2. Schematic drawing of the experimental setup used for breakthrough experiments.
Figure 3. SEM image of Fe-Ni particles produced by solution method.
b. pH/ORP. The ORP was measured to determine the electron activity in the solutions. The ORP indicates the relative tendency to accept or transfer electrons. Positive values of ORP imply oxidizing conditions in the solution, while negative values indicate reducing conditions. The ORP was measured using an Eh/pH (model Orion 290 A) probe according to ASTM D 1498. The ORP was measured to an accuracy of (0.2 mV or (0.05% of reading, whichever was greater. A portable pH/ISE meter (Orion Research Inc., model 290 A) was used to measure the changes in pH in effluents. The pH was measured to an accuracy of 0.005. The pH meter was calibrated on a regularly basis using the standard solutions of pH of 4, 7, and 10.
406
Ind. Eng. Chem. Res., Vol. 50, No. 1, 2011
c. Dynamic Light Scattering (DLS). Dynamic light scattering is also called the photon correlation spectroscopy. This is used to measure the size of the particles. A laser beam was passed through spherical particles in solution, which are in Brownian motion; this caused a Doppler shift and thus changed the wavelength of the light. This change in wavelength was related to the size of the particles. Using the autocorrelation function and measuring the dispersion coefficient, it was possible to determine the spherical particle size and describe the motion of the particles in the medium. The advantage of this method was that it could measure a wide range of particles. The dynamic light scattering was performed using an ALV-5000 (ALV-Laser, Germany) with a green line from an argon laser. The correlator was equipped with 256 channels. d. X-ray Diffraction (XRD). The XRD pattern of the particles was obtained by using Siemens D5000 powder X-ray diffraction. The Siemens D5000 instrument has a large diameter goniometer (600 mm), low divergence collimator, and soller slits. Attachments include sample spinner stages, reflection/ transmission holders, incident or diffracted beam monochromators, and zero background holders. The instrument is useful for studying both powder and bulk materials. This diffractometer is best utilized for high-precision work. Data collection is performed under computer control using the MDI “Data Scan” software. e. UV Spectrophotometer. A UV spectra meter, Cecil (CL1000) scanning spectrophotometer, was used to determine the response of the Fe/Ni particles solution with time. The absorption spectra of the Fe/Ni particles were obtained by subtracting the spectrum of the blank solution of water from that of the sample solution containing Fe/Ni particles. Fe/Ni particles were oxidized using strong oxidant agent H2SO4. 1,10Phenanthroline is one of the most selective chelating agents used for the spectrophotometric determination. The formed Fe/Ni1,10-phenanthroline complex was analyzed using a UV spectrophotometer at 510 nm. Synthesis of Fine Particles. Fe-Ni Particles. To produce the bimetallic particles, 6.15 g of FeSO4 · 7H2O and 1.5 g of NiCl2 · 6H2O were mixed in 50 mL of water using a magnetic stirrer in 200 mL bottles for 30 min, and 1.3 g of NaBH4 was added into the solution for preparation of Fe-Ni bimetallic particles. The entire system was kept under nitrogen atmosphere to prevent the oxidation of Fe-Ni bimetallic particles, and the proposed reaction is as follows: 3FeSO4 + NiCl2 + 8NaBH4 + 24H2O f 3Fe-Ni V + 28H2 v + 8B(OH)3 + 2NaCl + 3Na2SO4
(1)
3. Experiment Batch Studies. Blasting sand, bentonite clay, and kaolinite clay were used for adsorption and desorption study of fine metallic particles. Batch sorption experiments were conducted at room temperature. Colloidal particles of different types were prepared in different concentrations. Those solutions were transferred to 40 mL EPA vials with sand. After 14 days of end-over-end rotation of tumbler, the incubation vials were centrifuged, and subsequently the clear supernatant was analyzed for the total aqueous colloid concentration by UV spectroscopy using the 1,10-phenanthroline method. Tracer Experiments. Tracer breakthrough curves were analyzed to determine Peclet number and column porosity. From the first and second moments of the Br- breakthrough curve, the average travel time of the tracer and the dispersion coefficient were calculated. Fitting the convective-dispersive
Figure 4. Schematic diagram of two-site equilibrium/kinetic model.
transport equation to the Br- breakthrough data using the nonlinear least-squares procedure gave very similar results (Figure 7). The porosity varied from 0.4 to 0.5. Column Studies. The experimental setup for the column studies is the column having 5.1 cm diameter and 50 cm long made out of glass, which is equipped with an air compressor and a pressure regulator unit that is used to pump the solutions. Two columns were used for bentonite and kaolinite clay. Yet in both cases the same type of blasting sand was used (Figure 2). Advective-Dispersive Equation. Consider a laboratory column containing a uniform natural porous medium at saturated flow conditions and constant solution composition. Transport of colloidal particles can be described by accounting for particle advection, hydrodynamic dispersion, and deposition (filtration). The concentration of suspended colloidal particles c(x,t) at a column depth x and time t follows the one-dimensional advection dispersion equation with a sink term for particle deposition: dc d2c dc ) D 2 - V - kc dt dx dx
(2)
where V is the interstitial velocity of the colloidal particles, D is the hydrodynamic dispersion coefficient, and k is the particle deposition rate coefficient. The transport equation assumes particle deposition to follow first-order kinetics and to be irreversible. Both assumptions are justified at sufficiently low particle concentration and for moderate to high ionic strengths where particle release is negligible as compared to particle deposition. Two-Site Equilibrium/Kinetic Model. Consider a soil system composed of a liquid phase involving convective and dispersive transport and of a solid phase subject to chemical sorption or exchange (Figure 4). Because the two-site model itself has been explained in detail by Van Genuchten and Parker,26,27 only a summarized version is presented here. Following the notation of Van Genuchten and Parker,26,27 we have at equilibrium for the Type-1 (equilibrium) and Type-2 (kinetic) sites, respectively: S1 ) fkC
(3)
S2 ) (1 - f)kC
(4)
where f is the fraction of exchange sites assumed to be at equilibrium, C is concentration, θ is the volumetric water, k is an empirical distribution coefficient, µ is a first-order decay coefficient, R is a first-order kinetic rate coefficient, F is the soil bulk density, and the subscripts 1 and 2 refer to the two
Ind. Eng. Chem. Res., Vol. 50, No. 1, 2011
407
Type-1 and Type-2 sorption sites, respectively. Total adsorption, S, is given by S ) S1 + S2
(5)
which is at equilibrium. Figure 2 shows schematically a soil made up of the liquid phase, a Type-1 solid phase, and a Type-2 solid phase. Both Ja1 and Ja2 are the sorption rates from the liquid into the Type-1 and Type-2 regions, respectively. Mass transport in the liquid phase given by Ja now becomes the sum of Ja1 and Ja2. The mass balances for the Type-1 and Type-2 are represented by F
∂S1 ) Ja1 - Fµs1S1 ∂t
(6)
F
∂S2 ) Ja2 - Fµs2S2 ∂t
(7)
and
The mass transport equation for the system as a whole follows by adding the contributions of eqs 6 and 7, and noting that Ja ) Ja1 + Ja2: ∂C ∂ θD - qC ∂(S1 + S2) ∂(θC) ∂x ) - θµ1C +F ∂t ∂t ∂x Fµs1S1 - Fµs2S2
(
)
Figure 5. XRD analysis of Fe/Ni particles produced using solution method.
(8)
Because Type-1 sites are always at equilibrium, sorption onto these sites is also given by the time-derivative, and it becomes: ∂S1 ∂C ) fk ∂t ∂t
(9)
By using a first-order kinetic sorption rate law, the mass balance equation for the Type-2 sites becomes: ∂S2 ) R[(1 - f)kc - S2] - µs2S2 ∂t
(10)
From the above equations, the equilibrium sorbed phase degradation term leads to: ∂(θ + fFk)C ) ∂t
(
∂ θD
∂C - qC ∂x - RF[(1 - f)kC - S2] ∂x θµ1C - fFkµs1C (11)
)
Hence, the complete two-site model is given by eq 11. It can be rearranged as ∂2C Ffk ∂C ∂C RF )D 2 -V 2 [(1 - f)kc - S2] 1+ θ ∂t θ ∂X ∂x fFkµslC (12) µlC θ
(
)
The above-mentioned basic, one-site kinetic model, two-site kinetic model equations can be applied to different initial and boundary conditions to get the different final solutions. For steady state flow in a soil, transport of a linearly adsorbed solute is given by a two-site nonequilibrium transport model, which makes a distinction between type 1 (equilibrium) and type 2 (first-order kinetics) adsorption sites.28 This model was verified and used in many past research works. f is the fraction of exchange sites that are always at equilibrium, and (1 - f)
represents the sites that are first-order kinetics. If f ) 0, it will fall into the one-site kinetic model. One- and two-site kinetic models have been applied to fit and predict breakthrough curves from column experiments. The two-site model fit all breakthrough curves satisfactorily, accounting for the skewness of the rising limb as well as for the smooth transition of the declining limb to the tail of the breakthrough curve. The one-site kinetic model does not follow the curvature of the breakthrough tail, leading to an overestimation of the inactivation site. The resulting two-site kinetic model has been successfully used to describe the transport of a number of solutes that interact with a solid phase composed of such different constituents as soil materials, organic matter, and various oxides. This model is widely used in experiments such as viruses, bacteria, protozoa, and colloidal particles.26-28 The particles often exhibit a typical skewness. The typical response of such a breakthrough has a slowly rising limb and smooth transition of a declining limb to a very long tail. One-site kinetic models fail to fit the rising and declining limbs together with the tail satisfactorily.26-28 4. Results and Discussion Fe/Ni particles were produced using the solution method. The transportation of Fe-Ni particles was carried out in sandy clay soil. The SEM analysis of Fe/Ni particles prepared by the solution method showed spherical shaped particles (Figure 3). XRD analysis was done on Fe/Ni particles prepared by the solution method (Figure 5). The DLS analysis of Fe/Ni bimetallic particles showed that the sizes of those particles varied from nanoscale to microscale (Figure 6). The Fe/Ni particles showed no notable peak, indicating that no crystalline phase was present. These fine particles were then heated to 200 °C in an inert atmosphere for one-half of an hour. With the heat treatment, Fe/Ni became crystalline and showed peaks of Fe and Ni (Figure 5). In these bimetallic particles, peaks of other compounds were not detected. Around 0.3 g of the particles prepared was used for the XRD analysis.
408
Ind. Eng. Chem. Res., Vol. 50, No. 1, 2011
Figure 6. DLS analysis of Fe/Ni nanoparticles prepared by the solution method.
Figure 9. Breakthrough curve of Fe-Ni (solution method) in (A) 82.5% sand + 17.5% kaolinite, and (B) 95% sand + 5% bentonite soil columns.
Figure 7. Breakthrough of tracer (Br-) in soil column.
Figure 10. pH, ORP variation of Fe-Ni (solution method) nanoparticle transportation in (A) 82.5% sand + 17.5% kaolinite, and (B) 95% sand + 5% bentonite soil columns.
Figure 8. Transport of Fe/Ni particles in (A) 100% sand, (B) 100% kaolinite, and (C) 100% bentonite soil columns.
a. Fe/Ni Particle Transport in Pure Sand, Kaolinite, and Bentonite. The Fe/Ni particles have transported through soil column. The retardation factor and dispersion coefficient were 4.43 and 2.26 cm/min2, respectively, whereas for kaolinite they were 16.83 and 1.05 cm/min2. Fine bimetallic Fe/Ni particles were not transported through pure bentonite. It was concluded that bentonite was a good filtering medium for Fe/ Ni particles (Figure 8). b. Fe/Ni Particle Transport in Sand (82.5%) and Kaolinite (17.5%) Mixture. The pH of effluent varied from 4.5 to 6.33, whereas ORP varied from 189 mV to 90.3 mV (Figure 10). This indicates the particles are positively charged. The remaining Fe/Ni particle concentration profile from the top
Figure 11. Fe-Ni concentrations along soil depth (solution method) in (A) 82.5% sand + 17.5% kaolinite, and (B) 95% sand + 5% bentonite soil columns.
layer to the bottom layer varied from 0.29 to 0.0012 mg Fe-Ni/g soil (Figure 11). The retardation factor, dispersion coefficient, and advective velocity were 17.38, 2.45 cm/min2, and 1.77 cm/min, respectively (Figure 9). c. Fe/Ni Particle Transport in Sand (95%) and Bentonite (5%) Mixture. The breakthrough curve of Fe//Ni particles produced using the solution method is shown in Figure 9. The pH of effluent varied from 3.72 to 7.62, whereas ORP varied from 208.4 mV to -37.8 mV (Figure 10). This indicates
Ind. Eng. Chem. Res., Vol. 50, No. 1, 2011
409
Figure 13. UH biosurfactant breakthrough curve in PCE contaminated column. Figure 12. Particle size distribution in effluent in (A) 82.5% sand + 17.5% kaolinite, and (B) 95% sand + 5% bentonite soil columns.
the particles are positively charged. The remaining Fe/Ni particle concentration profile from the top layer to the bottom layer varied from 0.20 to 0.09 mg Fe-Ni/g soil (Figure 11). The retardation factor, dispersion coefficient, and advective velocity were 35.61, 5.47 cm/min2, and 2.18 cm/min, respectively. The average particle size in the effluents is shown in Figure 12. Transport results showed that most of the remaining particles were found on the top part of the columns (around 1/3 of the columns) after breakthrough studies. Agglomeration of these bimetallic particles takes place primarily through interparticle interactions such as van der Waals forces and magnetic interactions. Further, once aggregation occurs, the aggregates are retained at the column inlet, subsequently attracting additional particles. The possibility of much more enhanced transport of Fe/Ni particles in clay soil is using stabilizers.34 A stabilizer can enhance dispersion or reduce agglomerations of fine bimetallic particles through (a) electrostatic repulsion (i.e., adsorption of charged stabilizer molecules to the metal core results in an enhanced electrical double layer and, thus, increased Columbian repulsion between the capped particles) and (b) steric hindrance (i.e., coating the metal core with sterically bulk stabilizers such as polymers impedes particle attractions).34 The results showed that transport of Fe/Ni bimetallic particles through bentonite clay was much less as compared to kaolinite clay where it was observed visually during column experiments. Permeability was drastically reduced with time in columns packed with only with pure bentonite material, and no Fe/Ni particles were observed at the outlet. The unique characteristic of bentonite material is swelling. During breakthrough studies, bentonite swelled and thereby filled up the pore spaces of packed columns. It created a better impermeable zone, which prevented particle migration (Figure 8). This is called self-sealing. This was not the case either for pure kaolinite or for pure sand, and the permeability remained the same (Figure 8). Most of the studies were conducted alone either on surfactant flushing or colloid particles transport.16,23 In this study, both effects of surfactants flushing on contaminated soils and Fe/Ni particles transport to degrade PCE in contaminated soil were investigated in the same conditions to illustrate the whole spectrum clearly. Breakthrough studies were conducted in PCE contaminated soil. The goal of this study was to understand the recover efficiency of PCE using water, surfactants, and nanoparticles. CTAB and UH biosurfactant were chosen to represent anionic and biosurfactant in contaminated column studies. The soil (82.5% blasting sand and 17.5% kaolinite) was contaminated to 100 000 mg/kg and aged for 3 months. Two columns were
Figure 14. CTAB breakthrough curve in PCE contaminated column.
packed with 3 month aged soil in columns. The critical micelle concentration (CMC) of CTAB and UH biosurfactant was 0.4 and 0.7 g/L, respectively. The breakthrough studies were done using water, 5 g/L CTAB surfactant solutions as continuous input, and PCE concentration was measured in effluents. After passing of many pore volumes of surfactant solutions, 200 g/L Fe/Ni particles was injected as pulse input to one column. After passing 5 g/L CTAB surfactant solutions through the column, the remaining effluent PCE concentration was measured (Figure 13). In the first set of column studies, the PCE recovery efficiency was 3.55% from contaminated soil when water alone was used. When there was no PCE present in effluent, the remaining column was flushed using surfactant, and PCE recovery was increased. In case of UH biosurfactant and CTAB, the efficiency was increased to 4.36% and 3.95%, respectively (Figures 13 and 14). In the second set of column studies, surfactant was alone used in breakthrough studies. It yielded the PCE recovery of 8% and 7.9% for UH biosurfactant and CTAB, respectively. The third column studies Fe/Ni particles that were injected as a pulse input under surfactant as background solutions, and PCE recovery was measured. In case of UH biosurfactant as a background solution, the PCE recovery was 0.6%, whereas for CTAB, it was 0.8%. This indicated the Fe/Ni particles degrade PCE in contaminated soil. The results are shown in Table 2. Higher nanoparticle concentrations would be needed to achieve appreciable contaminant removal over an area. Nanoscale particles were effective for the transformation of PCE. Results from this work show that the nanoparticles technology is suited as an effective, portable, and flexible remedial technique for amenable groundwater contaminants such as chlorinated hydrocarbons. The ability to deliver nanoparticles directly to
410
Ind. Eng. Chem. Res., Vol. 50, No. 1, 2011
Table 2. PCE Recovery Efficiencies
water alone water + surfactant surfactant nano iron
UH biosurfactant
CTAB
0.0355 0.0436 0.0800 0.0057
0.0353 0.0395 0.0787 0.0085
contaminant areas, and the relative simplicity of the process, could make the nanoparticle technology effective for many applications. The transport parameters of nanoparticles in various soil columns determined from our experiments show the potential of this remedial technology in sandy clay soils and long-term evaluations of destruction of chlorinated compounds in subsurface and groundwater. 5. Conclusions The transport parameters of Fe/Ni particles were determined in sandy clay soil. Those numbers are mainly critical for in situ treatments, which are mainly governed by the sorption mechanism. PCE in the contaminated soil was degraded by Fe/Ni particles during breakthrough studies. The eluted pore volumes’ pH increased from low to high and ORP decreased from high to low in sandy clay soil in which the particles were negatively charged. To further investigate these mechanisms, more attention should be directed to the fundamentals of nanochemistry in the environment. Because these studies were carried out on the lab scale, more field scale studies should be carried out to investigate the reactions, transport mechanisms, and the potential and limits of this technology. Through improved understanding of geochemistry of nanoparticles in both basic and field studies, the success of this prospect technology will be further exploited. Acknowledgment This study was supported by the Center for Innovative Grouting Materials and Technology (CIGMAT) at the University of Houston with funding from the Texas Hazardous Waste Research Center and Texas Higher Education Coordinating Board (THECB). The contents do not necessarily reflect the views and policies of the funding agencies. Nomenclature CTAB ) cetyltrimethylammonium bromide SEM ) scanning electron microscopy EDS ) energy dispersive X-ray spectroscopy XRD ) X-ray diffraction UH biosurfactant ) University of Houston biosurfactant DNAPL ) dense nonaqueous phase liquid
Literature Cited (1) Zhang, W. Nanoscale iron particles for environmental remediation: An overview. J. Nanopart. Res. 2003, 5, 323–332. (2) Brooks, M. C.; Annable, M. D.; Rao, P. S. C.; Hatfield, K.; Jawitz, J. W.; Wise, W. R.; Wood, A. L.; Enfield, C. G. Controlled release, blind test of DNAPL remediation by ethanol flushing. J. Contam. Hydrol. 2004, 69, 281–297. (3) Sabatini, D. A.; Knox, R. C.; Harwell, J. H. Emerging Technologies Surfactant Enhanced Subsurface Remediation; ACS Symposium Series 594; American Chemical Society: Washington, DC, 1995; pp 2-8. (4) Chu, W. Remediation of contaminated soils by surfactant aided soil washing. Practical Periodical of Hazardous; Toxic and Radioactive Waste Management: 2003; Vol. 7, pp 19-24. (5) Cohen, R. M.; Mercer, J. W. DNAPL Site EValuation; Lewis Publishers: Chelsea, MI, 1993.
(6) Alessi, D. S.; Li, Z. Synergic effect of cationic surfactants on perchloroethylene degradation by zerovalent iron. EnViron. Sci. Technol. 2001, 35, 3713–3717. (7) Boronina, T.; Klabunde, K. J.; Sergeev, G. Destruction of organohalides in water using metal particles: carbon tetrachloride/water reactions with magnesium, tin and zinc. EnViron. Sci. Technol. 1995, 29, 1511–1517. (8) Cho, S.; Shahin, A.; Long, G.; Davies, J.; Liu, K.; Grandjean, F.; Kauzlarich, S. Magnetic and mossbauer spectral study of core/shell structured Fe/Au nanoparticles. Chem. Mater. 2006, 18, 960–967. (9) Doong, R.; Chen, K.; Tsai, H. Reductive dechlorination of carbon tetrachloride and tetrachloroethylene by zerovalent silicon-iron reductants. EnViron. Sci. Technol. 2003, 37, 2575–2581. (10) Gillham, R. W.; O’Hannesin, S. F. Enhanced degradation of halogenated aliphatics by zero-valent iron. Ground Water 1994, 32, 958– 967. (11) Huh, C. Formation of a middle phase from a lower or upper phase micro emulsion. J. Colloid Interface Sci. 1984, 97, 201–219. (12) Jawitz, J. W.; Annable, M. D.; Rao, P. S. C.; Rhue, R. D. Field implementation of a winsor type 1 surfactant/alcohol mixture for in situ solubilization of a complex LNAPL as a single phase micro emulsion. EnViron. Sci. Technol. 1998, 32, 523–530. (13) Tee, Y.; Grulke, E.; Bhattacharyya, D. Role of Ni/Fe nanoparticle composition on the degradation of TCE from water. Ind. Eng. Chem. Res. 2005, 44, 7062–7070. (14) Schrick, B.; Blough, J. L.; Jones, D. A.; Mallouck, T. E. Hydrodechlorination of trichloroethylene to hydrocarbons using bimetallic nickeliron nano particles. Chem. Mater. 2002, 14, 5140–5147. (15) Kretzschmar, R.; Barmettler, K.; Grolimund, D.; Yao, Y.; Borkovec, M.; Hans, S. Experimental determination of colloid deposition rates and collision efficiencies in natural porous media. Water Resour. Res. 1997, 33, 1129–1137. (16) Lindstrom, F. T.; Boersma, L. Analytical solutions for convective dispersive transport in confined aquifers with differential initial and boundary conditions. Water Resour. Res. 1989, 25, 241–255. (17) Mackay, D. M.; Cherry, J. A. Groundwater contamination: limits of pump-and-treat remediation. EnViron. Sci. Technol. 1989, 23, 630–636. (18) Harendra, S.; Vipulanandan, C. Degradation of high concentrations of PCE solubilized in SDS and biosurfactant with Fe/Ni bi-metallic particles. Colloids Surf., A 2008, 322, 6–13. (19) Nurmi, J. T.; Tratnyek, P. G.; Sarathy, V.; Baer, D. R.; Amonette, J. E.; Pecher, K.; Wang, C.; Linehan, J. C.; Matson, D. W.; Penn, R. L.; Driessen, M. D. Characterization and properties of metallic iron nanoparticles: spectroscopy, electrochemistry and kinetics. EnViron. Sci. Technol. 2005, 39, 1221–1230. (20) Hong, Y.; Honda, R. J.; Myung, N. V.; Walker, S. L. Transport of iron-based nanoparticles: Role of magnetic properties. EnViron. Sci. Technol. 2009, 43, 8834–8839. (21) Schijven, J. F.; Hassanizadeh, S. M.; Bruin, R. H. Two site kinetic modeling of bacteriophages transport through columns of saturated dune sand. J. Contam. Hydrol. 2002, 57, 259–279. (22) Elliott, D. W.; Zhang, W. Field assessment of nanoscale bimetallic particles for groundwater treatment. EnViron. Sci. Technol. 2001, 35, 4922– 4926. (23) Elimelech, M.; O’Melia, C. R. Effect of particle size on collision efficiency in the deposition of brownian particles with electrostatic energy barriers. Langmuir 1990, 6, 1153–1163. (24) Elimelech, M.; Gregory, J.; Jia, X.; Williams, R. Particle Deposition and Aggregation, Measurement, Modeling, and Simulation; Elsevier Publishers: New York, 1995; pp 1-441. (25) Grolimund, D.; Borkovec, M.; Barmettler, K.; Sticher, H. Colloid facilitated transport of strongly sorbing contaminants in natural porous media: A laboratory column study. EnViron. Sci. Technol. 1996, 30, 3118– 3123. (26) Van Genuchten, M. T.; Wagenet, R. J. Two site, two region models for pesticide transport and degradation: Theoretical development and analytical solutions. Soil Sci. Soc. Am. J. 1989, 53, 1303–1310. (27) Van Genuchten, M. T.; Parker, J. C. Boundary conditions for displacement experiments through short laboratory soil columns. Soil Sci. Soc. Am. J. 1984, 48, 703–708. (28) Zhou, L.; Selim, H. M. Solute transport in layered soils: Non linear and kinetic reactivity. Soil Sci. Soc. Am. J. 2001, 65, 1056–1064. (29) Zhan, J.; Zheng, T.; Piringer, G.; Day, C.; Mcpherson, G. L.; Lu, Y.; Papadopoulos, K.; John, V. T. Transport characteristics of nanoscale functional zerovalent iron/silica composites for in situ remediation of trichloroethylene. EnViron. Sci. Technol. 2008, 42, 8871–8876. (30) Zhuang, J.; Flury, M.; Jin, Y. Colloid facilitated Cs transport through water saturated Hanford sediment and Ottawa sand. EnViron. Sci. Technol. 2003, 37, 4905–4911.
Ind. Eng. Chem. Res., Vol. 50, No. 1, 2011 (31) Hariharan, S. Production of biosurfactant from used vegetable oil and use for lead precipitation. Master’s thesis, Department of Civil and Environmental Engineering, University of Houston, TX, 1996. (32) Li, F.; Vipulanandan, C.; Mohanty, K. M. Microemulsion solution approaches to nanoscale iron production for degradation of TCE. Colloids Surf., A 2003, 223, 103–112. (33) Ren, X. Application of a biosurfactant in naphthalene, phenanthrene and TCE remediation. Doctoral dissertation, Department of Civil and Environmental Engineering, University of Houston, TX, 1998.
411
(34) He, F.; Zhao, D.; Liu, J.; Roberts, C. B. Stabilization of Fe-Pd nanoparticles with sodium carboxymethyl cellulose for enhanced transport and dechlorination of trichloroethylene in soil and groundwater. Ind. Eng. Chem. Res. 2007, 46, 29–34.
ReceiVed for reView July 19, 2010 ReVised manuscript receiVed October 4, 2010 Accepted November 8, 2010 IE101546K