Article pubs.acs.org/est
Field-Scale Transport and Transformation of Carboxymethylcellulose-Stabilized Nano Zero-Valent Iron Richard L. Johnson,†,* James T. Nurmi,† Graham S. O’Brien Johnson,† Dimin Fan,† Reid L. O’Brien Johnson,† Zhenqing Shi,† Alexandra J. Salter-Blanc,† Paul G. Tratnyek,† and Gregory V. Lowry‡ †
Institute of Environmental Health, Oregon Health & Science University, 20000 NW Walker Road, Beaverton, Oregon 97006, United States ‡ Center for Environmental Implications of NanoTechnology (CEINT) and Departments of Civil & Environmental Engineering, Chemical Engineering, and Biomedical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213-3890, United States S Supporting Information *
ABSTRACT: The fate of nano zerovalent iron (nZVI) during subsurface injection was examined using carboxymethylcellulose (CMC) stabilized nZVI in a very large three-dimensional physical model aquifer with detailed monitoring using multiple, complementary detection methods. A fluorescein tracer test in the aquifer plus laboratory column data suggested that the very-aggressive flow conditions necessary to achieve 2.5 m of nZVI transport could be obtained using a hydraulically constrained flow path between injection and extraction wells. However, total unoxidized nZVI was transported only about 1 m and 0.5 m. To achieve this hydraulic gradient over a 2.5 m distance, three parallel sets of injection and extraction wells were used (Figure 1A). The center injection well was used for the CMC/nZVI injection and the adjacent injection wells were used for a CMConly solution. The three injection wells were constructed of ∼2.5 cm I.D. (1 in. schedule-40) PVC pipe, positioned about 0.05 m from the wall of the physical model using direct push equipment (Geoprobe, Salina, KS). Each well had a 0.75 mlong screen (slot size 0.5 mm) and the screened interval straddled the sand layer between the silica flour and clay layers (Figure 1B). The flow rate into each injection well was ∼6 L min−1 (∼1.6 gallons min−1). A total of six extraction wells were used to minimize localized drawdown, and approximately 3 L min−1 were extracted from each well. The flow rate and total flow for both injection and extraction were monitored during the experiment using in-line flow transducers (Cole Parmer, Vernon Hills, IL). The hydraulic gradient was monitored using a pair of wells installed ∼0.5 m from the injection and extraction wells (i.e., 1.5 m apart) and equipped with submersible pressure transducers (Instrumentation Northwest, Kirkland, WA). These wells were constructed using ∼2.5-cm i.d. (1-in. SCH 40) PVC pipe with 15-cm screened intervals. The submersible pressure transducers were connected to a data acquisition 1575
dx.doi.org/10.1021/es304564q | Environ. Sci. Technol. 2013, 47, 1573−1580
Environmental Science & Technology
Article
Figure 2. Composite photograph of water samples taken from the first three sample wells over the time period of the injection test. Rectangular markers highlight the location of the two color transitions that indicate breakthrough of CMC/nZVIox (yellow) and CMC/nZVI (black).
by minimizing radial flow of injected nZVI. Immediately following nZVI injection, an additional 400 L of 4.8 g L−1 CMC-only solution was injected into each of the three injection wells, again at a rate of ∼6 L min−1 to further push the nZVI away from the injection well.
■
RESULTS AND DISCUSSION Breakthrough of CMC/nZVI. Figure 2 shows a composite photograph of water samples collected from the first three sampling wells (R, O, and Y) over the time period of the injection test. The time series of water samples from each well exhibits a transition from colorless, to yellow and then black; which is consistent with initial breakthrough of oxidized iron (nZVIOX, for example, formed by the reaction of nZVI with oxidizing constituents of the groundwater including water itself, dissolved oxygen, and matrix material), followed by the arrival of black nZVI. The quantity of oxidized iron, represented by the separation between the initial transition from colorless to yellow and the subsequent transition from yellow to black, reflects the “reductant demand” of the aquifer system. The time (volume of nZVI suspension) required to satisfy the reductant demand increased with distance from the injection well. This suggests that reductant demand caused significant oxidation of nZVI, a prospect that has been suggested previously,27 butto our knowledgehas not been quantified in a field-scale system. To quantify the breakthrough of unoxidized nZVI, we measured absorbance in samples collected from each well. Figure S6 (SI) shows that at 800 nm there is little absorption by CMC/nZVIOX but strong absorption by CMC/nZVI and, as a result, absorbance at 800 nm was used to detect the unoxidized nZVI. Background-subtracted absorbance measurements of collected water samples (Figure 3) show that at each sampling well there was an increase in 800 nm absorbance concurrent with the visual transition from yellow to black. To develop a quantitative relationship between absorbance and nZVI concentration, dilutions of the nZVI injection solution were analyzed. The data in SI, Figure S7 indicate a very linear relationship between concentration and absorbance, and allowed the concentration of nZVI in each sample to be directly determined. The concentration of nZVIOX in each sample was then calculated as the difference between the total
Figure 3. Absorbance (background-subtracted) at 800 nm vs time at sampling wells R, O, and Y. The shaded regions are drawn from the appearance of CMC/nZVI, based on visual inspection of the data in Figure 2. Transitions shown by the colored bars are defined by the results in Figure 2.
iron and nZVI concentrations, and the results for wells R, O, and Y are shown in Figure 4. The data from well Y indicate that the total iron was only ∼4% of the injection concentration, and perhaps more importantly, the concentration of unoxidized nZVI was less than 2% of the injected value. Resolving the nZVIOX and nZVI portions of the total iron breakthroughas shown in Figure 4demonstrates that oxidation has a significant impact on nZVI delivery and almost all the iron was oxidized at 1 m (Well Y). This apparent reductant demand could be due to reaction of CMC/nZVI with dissolved oxygen, dissolved or adsorbed contaminants, natural organic matter, exposed surfaces of the mineral grains that comprise the aquifer matrix, or water. The potential diversity of oxidizing aquifer materials suggests that the kinetics of CMC/ nZVI reaction with these materials may vary greatly, so the apparent reductant demand is likely to depend on contact time and other operational factors. In addition, the apparent reductant demand will depend on properties of the reductant: with stronger reductants (e.g., nano- vs micro-sized ZVI, or zerovalent zinc vs zerovalent iron) reacting faster and farther with the available oxidants. For both of these reasons, we 1576
dx.doi.org/10.1021/es304564q | Environ. Sci. Technol. 2013, 47, 1573−1580
Environmental Science & Technology
Article
Figure 4. Contributions of nZVI and nZVIOX to total iron (mg/L) breakthrough at the three wells as a function of injection time (calculated from absorbance at 800 nm, measured total iron concentration and the calibration approach described in the SI, Section S6.).
SC decreases occurred while CMC/nZVI injection was still ongoing. These changes are consistent with the decreases in total iron concentration observed at wells R and O (Figure 4). However, while decreases in iron concentration could have been due to either increased filtration of the nZVI (by the porous medium) or changes in flow, the decrease in SC can only be attributed to changes in flow. This result is similar to behavior we observed in a preliminary CMC/nZVI injection test (SI, Section S7). In that case, total iron and SC concentrations peaked and then dropped to near-background levels while CMC/nZVI injection was ongoing. To further demonstrate that changes in the flow field were responsible for the decrease in the preliminary injection experiment, a second conservative tracer, boron from the borohydride used in nZVI synthesis, was measured in sampling well R. The boron data (SI, Figure S8) match those of the iron and SC, which further supports the conclusion that flow changes resulted in the mixing of different parcels of water that led to decreases in iron concentrations in the current injection experiment. While these studies were not designed to provide a detailed characterization of changes in the flow regime, additional insight is provided by further consideration of the preliminary injection experiment results. In that case, following CMC/nZVI injection, H2-free water was flushed through the system, and a second fluorescein tracer experiment was conducted. This follow-up tracer test showed good agreement with the fluorescein test conducted prior to the preliminary CMC/ nZVI experiment, and indicated that flow changes caused by the CMC/nZVI injection were reversible. We believe that the most likely explanation for these results is the formation of hydrogen bubbles during the injection process, which could cause a temporary diversion of flow from the areas where nZVI concentrated. As discussed above, in the current injection experiment, the CMC/nZVI solution was sparged immediately before injection to remove all dissolved hydrogen. We believe this resulted in improved CMC/nZVI transport in the current experiment compared to the preliminary one; however, flow changes were observed in the current experiment as well (which we again believe were due to hydrogen bubbles generated in situ during the injection) which had a smaller but still significant effect on overall flow and nZVI transport. Indirect Measurement of nZVI Transport. The data in Figures 2−4 represent relatively direct indictors of CMC/nZVI breakthrough; however, most field studies rely heavily on indirect indicators, including DO, ORP, and pH. They are appealing because they are easily measured with simple commercially available electrodes. Their measurements are usually made off-line using grab samples, but they can also be made continuously with in-line flow cells. All three of these
anticipate that reductant demand will eventually prove to be closely analogous to oxidant demand (e.g., natural oxidant demand, NOD, as defined in applications of situ chemical oxidation (ISCO)). However, while the concept of oxidant demand has been developed and applied extensively in recent years,37−40 the significance of reductant demand to the performance of treatments by in situ chemical reduction (ISCR) has not yet been widely recognized. The specific conductance of the CMC/nZVI injection fluid was ∼3900 μS cm−1, the SC of the CMC-only solution used in the two adjacent injection wells was ∼1040 μS cm−1, and the background groundwater had an SC of ∼200 μS cm−1. The SC at each of the sampling wells (Figure 5) increased more-or-less
Figure 5. Specific conductance (μS cm−1) measured at the R, O, and Y sampling wells during CMC/nZVI injection.
concurrently with the appearance of nZVIOX. This shows that the conductivity measurements were not indicative of nZVI breakthrough. Given the composition of the injection solutions, the conductivity is likely dominated by Na+, SO42+, and borate. As a result, SC can be interpreted as a direct, conservative tracer of overall fluid transport during the experiment. The SC data indicate that, unlike the fluorescein tracer test (SI, Figure S3), concentrations at sampling wells R, O, and Y did not reach the concentration of the injected solution (3900 μS cm−1). In addition, arrival of the SC plume was somewhat slower than for the fluorescein tracer test. The latter may have been due, in part, to increased lateral flow caused by displacement of less-viscous water surrounding the primary flow path by the higher viscosity CMC solution. Of greater significance, however, are the decreases in SC that occurred over time at each location (Figure 5). For the R and O wells, 1577
dx.doi.org/10.1021/es304564q | Environ. Sci. Technol. 2013, 47, 1573−1580
Environmental Science & Technology
Article
sized, electrochemical flow cell were used to make continuous measurements at well O only (Figure 7B). In all cases, the ORP shows a sharp drop of 200−400 mV after breakthrough of nZVIox and coincident with breakthrough of nZVI (as evidenced by the changes in visible color and absorbance at 800 nm). In all cases, the ORP stabilized, briefly, at a minimum value (ORPmin) of roughly −200 mV (vs Ag/AgCl), which is considerably more positive than would be expected from a suspension of >100 mg/L fresh nZVI in a closed system.26 Therefore, absent other evidence, the measured ORP data are not sufficient to demonstrate breakthrough of unoxidized nZVI. After an additional 40 min, the Pt electrode in the microflow cell (Figure 7B) reached a more negative ORP that is typical of systems containing fully reduced nZVI (ca. −700 mV vs Ag/ AgCl, ref.26). However, this electrode response may not be (entirely) due to the direct effect of nZVI, because the high concentrations of H2 generated by corrosion of Fe0 can produce similar potentials on a Pt electrode.26 The GC carbon electrode, which should be relatively insensitive to H2, never gave an ORP below −200 mV, but instead showed a small (∼100 mV) rebound over the first 30 min after the initial ORPmin was reached. A similar rebound was seen in the ORP’s obtained with the Pt electrode in the milliliter flow cell (Figure 7A, see wells R and O. Well Y was not monitored long enough to see this effect). Even though these rebounds are relatively small features of the data, the three independent occurrences of the same effect suggest that it is a significant characteristic of the system. The most straightforward interpretation of the rebound in ORP measurements is that it is due to the same changes in flow that we invoked to explain the decreases in absorbance (Figure 3) and SC (Figure 5). However, to reconcile this with the decrease in ORP at the Pt electrode in the microliter flow cell requires additional considerations. Recall that the nZVI suspension was sparged to lower the dissolved H2 concentration immediately before injection (in order to minimize H2 bubble formation near the injection well), but reaction of nZVI and any residual borohydride with H2O are expected to increase [H2]. This process may be accurately reflected in the decrease in ORP at the Pt electrode in the microliter flow cell. The lack of such an effect on the conventional Pt electrode response could be because it had a relatively long response time
parameters were measured in this study; however, the pH data provided little insight and are not discussed here. In the current experiment, DO concentrations at the beginning of the injection test were 6−9 mg L−1 (Figure 6), which is consistent
Figure 6. Dissolved oxygen concentrations for at well R, O, and Y.
with aerobic oxygenated conditions in the aquifer. Arrival of unoxidized nZVI at each location coincided with sharp decreases in dissolved oxygen concentrations to below our detection limit (0.5 mg L−1) due to the rapid rate of oxygen reduction by nZVI. This suggests that the rate of oxidation by dissolved oxygen is much more rapid than with other potential oxidants including water and aquifer materials, which is consistent with previous reports using a different nZVI.41,42 The data also suggest that nZVI will not be present until after the DO is depleted. However, the presence of slower-reacting oxidants likely means that the observed depletion of DO is a necessary, but not sufficient, condition for demonstrating the arrival of nZVI, which may limit its usefulness as an indicator of nZVI transport. The results from two types of ORP measurements are shown in Figure 7. A conventional, combination Pt electrode in a milliliter-volume flow-through cell was used to make periodic measurements at wells R, O, and Y (Figure 7A); whereas, freshly polished Pt and GC working electrodes in a microliter-
Figure 7. ORP measurements using two electrode configurations: (A) traditional ORP data collected at the first three sampling wells measured with a Pt electrode used in a 2.5 cm diameter flow through cell. Symbols indicate discrete data collection. (B) ORP data at sampling well O using a Pt and GC working electrode in a micrometer sized flow cell configuration. Data points were collected every ∼14 s. 1578
dx.doi.org/10.1021/es304564q | Environ. Sci. Technol. 2013, 47, 1573−1580
Environmental Science & Technology
■
to H2determined in separate control experiments (not shown), and presumably because it could not be polished like we did for both electrodes used in the microflow cell. Overall, the ORP results illustrate the challenges and opportunities provided by indirect indicators for monitoring injection of nZVI: the primary effect is a sharp drop in ORP that corresponds well with other direct and indirect indicators of nZVI breakthrough. However, the primary effect can be obscured by secondary effects, which will make decreases in ORP alone insufficient for demonstrating breakthrough of nZVI. The converse, however, is almost certainly true: given the high reactivity of nZVI, the absence of strongly negative ORP values indicates that unoxidized nZVI has not arrived. Implications for Assessment of nZVI Injections. The chemical measurements presented above reflect a complex reactive transport process. Visual inspection and spectrophotometric data indicate that total iron measurements are not adequate for detecting the arrival of nZVI, due to the potential importance of reductant demand. Assessment of nZVI arrival can also be complicated because groundwater flow can change during the injection process due to hydrogen bubble formation. Bubble formation may, in part, be due to the reaction of the nZVI with water to form H2, but it may also be due to decomposition of residual borohydride from the nZVI synthesis process. In either case, it will not be possible to fully evaluate these effects using common indirect indicators such as DO and ORP. As a consequence, it is important to include a conservative tracer (e.g., bromide, specific conductance) in the injection fluids to directly evaluate changes in flow. However, the absence of a significant drop in DO and ORP is a good indication that unoxidized nZVI is not present at the sampling location. As part of our experimental work, we carried out a simple mathematical analysis of the reactive transport process (SI, Sections 8 and 9). However, given the complexities of bubble formation, aquifer heterogeneity, aggregation, nanoparticle filtration, and reaction, we believe that predictive modeling of the transport of nZVI and its effect on the flow field will not be possible in most field implementations. The implication of this is that robust diagnostic tools are critical for assessing the effectiveness of nZVI delivery. As described here, we believe that the combination of optical absorption measurements for nZVI and a conservative tracer represents a simple yet powerful approach that can both directly measure transport of the nZVI and provide insight into flow changes that could occur during transport. Geochemical characterizations that are indirect indicators of the effects of nZVI emplacement (e.g., ORP and DO) are valuable only when used as complementary information together with direct characterization methods.
■
Article
AUTHOR INFORMATION
Corresponding Author
*Phone: 503-748-1193; fax: 503-748-1464; e-mail: rjohnson@ ebs.ogi.edu. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was funded by the Strategic Environmental Research and Development Program (SERDP) as part of ER-1485 (Fundamental Study of the Delivery of Nanoiron to DNAPL Source Zones in Naturally Heterogeneous Field Systems), and their support is gratefully acknowledged.
■
REFERENCES
(1) Lowry, G. V. Nanomaterials for groundwater remediation. In Environmental Nanotechnology; Wiesner, M. R., Bottero, J.-Y., Eds.; McGraw Hill: New York, 2007; pp 297−336. (2) Geiger, C. L.; Carvalho-Knighton, K.; Novaes-Card, S.; Maloney, P.; DeVor, R. A review of environmental applications of nanoscale and microscale reactive metal particles. In Environmental Applications of Nanoscale and Microscale Reactive Metal Particles, ACS Symposium Series; Geiger, C. L., Carvalho-Knighton, K. M., Eds.; American Chemical Society: Washington DC, 2009; Vol. 1027, pp 1−20. (3) Tratnyek, P. G.; Johnson, R. L. Nanotechnologies for environmental cleanup. NanoToday 2006, 1, 44−48. (4) He, F.; Zhao, D.; Paul, C. Field assessment of carboxymethyl cellulose stabilized iron nanoparticles for in situ destruction of chlorinated solvents in source zones. Water Res. 2010, 44, 2360−2370. (5) Johnson, R. L.; Thoms, R. B.; O’Brien Johnson, R.; Nurmi, J. T.; Tratnyek, P. G. Mineral precipitation upgradient from a zero-valent iron permeable reactive barrier. Ground Water Monit. Remed. 2008, 28, 56−64. (6) Phillips, D. H.; Van, N. T.; Bastiaens, L.; Russell, M. I.; Dickson, K.; Plant, S.; Ahad, J. M. E.; Newton, T.; Elliot, T.; Kalin, R. M. Ten year performance evaluation of a field-scale zero-valent iron permeable reactive barrier installed to remediate trichloroethene contaminated groundwater. Environ. Sci. Technol. 2010, 44, 3861−3869. (7) Kouznetsova, I.; Bayer, P.; Ebert, M.; Finkel, M. Modeling the long-term performance of zero-valent iron using a spatio-temporal approach for iron aging. J. Contam. Hydrol. 2007, 90, 58−80. (8) Kirschling, T. L.; Gregory, K. B.; Minkley, E. G., Jr; Lowry, G. V.; Tilton, R. D. Impact of nanoscale zero valent iron on geochemistry and microbial populations in trichloroethylene contaminated aquifer materials. Environ. Sci. Technol. 2010, 44, 3474−3480. (9) Phenrat, T.; Kim, H.-J.; Fagerlund, F.; Illangasekare, T.; Tilton, R. D.; Lowry, G. V. Particle size distribution, concentration, and magnetic attraction affect transport of polymer-modified Fe0 nanoparticles in sand columns. Environ. Sci. Technol. 2009, 43, 5079−5085. (10) Kim, H.-J.; Phenrat, T.; Tilton, R. D.; Lowry, G. V. Effect of kaolinite, silica fines and pH on transport of polymer-modified zero valent iron nano-particles in heterogeneous porous media. J. Colloid Interface Sci. 2012, 370, 1−10. (11) Kanel, S. R.; Goswami, R. R.; Clement, T. P.; Barnett, M. O.; Zhao, D. Two dimensional transport characteristics of surface stabilized zero-valent iron nanoparticles in porous media. Environ. Sci. Technol. 2008, 42, 896−900. (12) Vecchia, E. D.; Luna, M.; Sethi, R. Transport in porous media of highly concentrated iron micro- and nanoparticles in the presence of xanthan gum. Environ. Sci. Technol. 2009, 43, 8942−8947. (13) Tosco, T.; Sethi, R. Transport of non-newtonian suspensions of highly concentrated micro- and nanoscale iron particles in porous media: A modeling approach. Environ. Sci. Technol. 2010, 44, 9062− 9068. (14) Phenrat, T.; Cihan, A.; Kim, H.-J.; Mital, M.; Illangasekare, T.; Lowry, G. V. Transport and deposition of polymer-modified Fe0 nanoparticles in 2-D heterogeneous porous media: Effects of particle
ASSOCIATED CONTENT
S Supporting Information *
Previous field studies, preliminary CMC-nZVI mobility measurements, previous CMC-nZVI injection results, transport characterization using fluorescein tracer, field synthesis of CMC-nZVI, colorimetric determination of CMC-nZVI, flow modeling; transport modeling, and parameters and methods used during the injection process can be found in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org. 1579
dx.doi.org/10.1021/es304564q | Environ. Sci. Technol. 2013, 47, 1573−1580
Environmental Science & Technology
Article
concentration, Fe0 content, and coatings. Environ. Sci. Technol. 2010, 44, 9086−9093. (15) Saleh, N.; Kim, H.-J.; Phenrat, T.; Matyjaszewski, K.; Tilton, R. D.; Lowry, G. V. Ionic strength and composition affect the mobility of surface-modified Fe0 nanoparticles in water-saturated sand columns. Environ. Sci. Technol. 2008, 42, 3349−3355. (16) Mackenzie, K.; Bleyl, S.; Georgi, A.; Kopinke, F. D. Carbo-iron An Fe/AC composite - As alternative to nano-iron for groundwater treatment. Water Res. 2012, 46, 3817−3826. (17) Sunkara, B.; Zhan, J.; He, J.; McPherson, G. L.; Piringer, G.; John, V. T. Nanoscale zerovalent iron supported on uniform carbon microspheres for the in situ remediation of chlorinated hydrocarbons. ACS Appl. Mater. Interfaces 2010, 2, 2854−2862. (18) Schrick, B.; Hydutsky, B. W.; Blough, J. L.; Mallouk, T. E. Delivery vehicles for zerovalent metal nanoparticles in soil and groundwater. Chem. Mater. 2004, 16, 2187−2193. (19) Zemansky, G. M.; Logar, A.; Manchester, K. R.; Zaluski, M. H.; Hogan, M.; Jaynes, N.; Petersen, S., Sand-tank test on injectability of nano-size zvi into saturated sand for mending an existing permeable reactive barrier in the 100-D area at the Hanford site. In International Environmental Nanotechnology Conference: Applications and Implications, (7−9 October 2008) Chicago, IL, EPA 905-R09-032; U.S. Environmental Protection Agency, 2009. (20) Zhong, L.; Oostrom, M.; Wietsma, T. W.; Covert, M. A. Enhanced remedial amendment delivery through fluid viscosity modifications: Experiments and numerical simulations. J. Contam. Hydrol. 2008, 101, 29−41. (21) Krug, T.; O’Hara, S.; Watling, M.; Quinn, J. Emulsified ZeroValent Nano-Scale Iron Treatment of Chlorinated Solvent DNAPL Source Areas, ESTCP Final Report, ER-0431-FR, 2010. (22) Henn, K. W.; Waddill, D. W. Utilization of nanoscale zerovalent iron for source remediationA case study. Biorem. J. 2006, 16, 57−77. (23) Wei, Y. T.; Wu, S. C.; Chou, C. M.; Che, C. H.; Tsai, S. M.; Lien, H. L. Influence of nanoscale zero-valent iron on geochemical properties of groundwater and vinyl chloride degradation: A field case study. Water Res. 2010, 44, 131−140. (24) Elliott, D. W.; Zhang, W.-X. Field assessment of nanoscale bimetallic particles for groundwater treatment. Environ. Sci. Technol. 2001, 35, 4922−4926. (25) Quinn, J.; Geiger, C.; Clausen, C.; Brooks, K.; Coon, C.; O’Hara, S.; Krug, T.; Major, D.; Yoon, W.-S.; Gavaskar, A.; Holdsworth, T. Field demonstration of DNAPL dehalogenation using emulsified zero-valent iron. Environ. Sci. Technol. 2005, 39, 1309−1318. (26) Shi, Z.; Nurmi, J. T.; Tratnyek, P. G. Effects of nano zero-valent iron (nZVI) on oxidation-reduction potential (ORP). Environ. Sci. Technol. 2011, 45, 1586−1592. (27) Borda, M. J.; Venkatakrishnan, R.; Gheorghiu, F. Status of nZVI technology: Lessons learned from North American and International implementations. In Environmental Applications of Nanoscale and Microscale Reactive Metal Particles, ACS Symposium Series; Geiger Cherie, L., Carvalho-Knighton, K. M., Eds.; American Chemical Society: Washington DC, 2009; Vol. 1027; pp 219−232. (28) Raychoudhury, T.; Tufenkji, N.; Ghoshal, S. Aggregation and deposition kinetics of carboxymethyl cellulose-modified zero-valent iron nanoparticles in porous media. Water Res. 2012, 46, 1735−1744. (29) Tratnyek, P. G.; Salter-Blanc, A. J.; Nurmi, J. T.; Amonette, J. E.; Liu, J.; Wang, C.; Dohnalkova, A.; Baer, D. R. Reactivity of zerovalent metals in aquatic media: Effects of organic surface coatings. In Aquatic Redox Chemistry, ACS Symposium Series; Tratnyek, P. G.; Grundl, T. J.; Haderlein, S. B., Eds.; American Chemical Society: Washington, DC, 2011; Vol. 1071, pp 381−406. (30) Sakulchaicharoen, N.; O’Carroll, D. M.; Herrera, J. E. Enhanced stability and dechlorination activity of pre-synthesis stabilized nanoscale FePd particles. J. Contam. Hydrol. 2010, 118, 117−127. (31) Fatisson, J.; Ghoshal, S.; Tufenkji, N. Deposition of carboxymethylcellulose-coated zero-valent iron nanoparticles onto
silica: Roles of solution chemistry and organic molecules. Langmuir 2010, 26, 12832−12840. (32) Daily, W.; Ramirez, A.; Johnson, R. L. Electrical impedance tomography of a perchloroethylene release. J. Explor. Eng. Geol. 1998, 2, 189−201. (33) Slater, L.; Binley, A. M.; Daily, W.; Johnson, R. L. Cross-hole electrical imaging of a controlled saline tracer injection. J. Appl. Geophys. 2000, 44, 85−102. (34) He, F.; Zhao, D. Manipulating the size and dispersibility of zerovalent iron nanoparticles by use of carboxymethyl cellulose stabilizers. Environ. Sci. Technol. 2007, 41, 6216−6221. (35) 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. (36) He, F.; Liu, J.; Roberts, C. B.; Zhao, D. One-step “Green” synthesis of Pd nanoparticles of controlled size and their catalytic activity for trichloroethene hydrodechlorination. Ind. Eng. Chem. Res. 2009, 48, 6550−6557. (37) Xu, X.; Thomson, N. R. A long-term bench-scale investigation of permanganate consumption by aquifer materials. J. Contam. Hydrol. 2009, 110, 73−86. (38) Xu, X.; Thomson, N. R. Estimation of the maximum consumption of permanganate by aquifer solids using a modified chemical oxygen demand test. J. Environ. Eng. 2008, 134, 353−361. (39) Urynowicz, M. A.; Balu, B.; Udayasankar, U. Kinetics of natural oxidant demand by permanganate in aquifer solids. J. Contam. Hydrol. 2008, 96, 187−194. (40) Johnson, R. L.; Tratnyek, P. G.; O’Brien Johnson, R. Persulfate persistence under thermal activation conditions. Environ. Sci. Technol. 2008, 42, 9350−9356. (41) Liu, Y.; Phenrat, T.; Lowry, G. V. Effect of TCE concentration and dissolved groundwater solutes on NZVI-promoted TCE dechlorination and H2 evolution. Environ. Sci. Technol. 2007, 41, 7881−7887. (42) Li, Z.; Greden, K.; Alvarez, P. J. J.; Gregory, K. B.; Lowry, G. V. Adsorbed polymer and NOM limits adhesion and toxicity of nano scale zerovalent iron to E. coli. Environ. Sci. Technol. 2010, 44, 3462− 3467.
1580
dx.doi.org/10.1021/es304564q | Environ. Sci. Technol. 2013, 47, 1573−1580
