PCE Oxidation by Sodium Persulfate in the Presence of Solids

Nov 11, 2010 - Batch reactor experiments were performed to determine the effects of solids on ... Sodium persulfate is a water-soluble oxidant used in...
0 downloads 0 Views 326KB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 9445–9450

PCE Oxidation by Sodium Persulfate in the Presence of Solids ˜ O,‡ J E D C O S T A N Z A , * ,† G R E T E L L O T A N JOHN CALLAGHAN,§ AND KURT D. PENNELL| School of Civil and Environmental Engineering, Georgia Institute of Technology, 311 Ferst Drive NW, Atlanta, Georgia 30332-0512, United States

Received March 29, 2010. Revised manuscript received October 22, 2010. Accepted October 27, 2010.

Batch reactor experiments were performed to determine the effects of solids on the oxidation of tetracholoroethylene (PCE) by sodium persulfate in aqueous solution. Based on the rates of PCE degradation and chloride formation, PCE oxidation by heat-activated sodium persulfate at 50 °C in the presence of solids ranged from no detectable oxidation of PCE to the levels observed in water-only reactors. Repeated doses of sodium persulfate, undertaken to overcome the inherent solids oxidant demand, improved the rate and extent of PCE oxidation in reactors containing reference solids; however, no improvement was observed in reactors containing field soils. Additionally, no improvements in PCE oxidation were observed after pretreating Great Lakes and Appling soils with ca. 15 g/kg of sodium persulfate or 30% hydrogen peroxide to remove oxidizable fractions, or acetic acid to remove the carbonate fraction. Based on these results, in situ treatment of Great Lakes and Appling soils with heat-activated sodium persulfate is not anticipated to result in substantial PCE oxidation, while in situ treatment of Fort Lewis soils is anticipated to result in PCE oxidation. This work demonstrates the need to perform soil-specific contaminant treatability tests rather than soil oxidant demand tests when determining oxidant dosage requirements.

Introduction Sodium persulfate is a water-soluble oxidant used in a variety of applications including initiation of polymerization reactions, bleaching of paper and metals, determination of the total organic carbon content of water samples (i.e., TOC analysis), and more recently, in situ chemical oxidation (ISCO) of contaminants in subsurface environments (1). Due to the stability of sodium persulfate, it can be injected into the subsurface as an aqueous solution for subsequent activation, which is most commonly accomplished by follow-on injection of a ferrous iron solution. Other activation methods include the injection of alkaline solutions, hydrogen peroxide, or the introduction of steam to increase subsurface temperatures (2). Although increasing temperature (i.e., heat activation) has been studied in bench scale tests as a method * Corresponding author phone: (202)564-8526; e-mail: [email protected]. † Present address: U.S. Environmental Protection Agency, Washington, D.C. ‡ Present address: DuPont Engineering, Houston, TX. § Present address: CALMAR Associates LLC, Dorothy, NJ. | Present address: Department of Civil and Environmental Engineering, Tufts University, Medford, MA. 10.1021/es100997a

 2010 American Chemical Society

Published on Web 11/11/2010

for activating sodium persulfate (1, 3, 4), it has received limited attention for field application. When compared to chemical activation of sodium persulfate (e.g., ferrous iron), one advantage of heat activation is the increase in aqueous phase contaminant concentrations caused by increasing subsurface temperatures (5). Given that sodium persulfate oxidation occurs in the aqueous phase, an increase in aqueous phase contaminant concentration caused by heating could result in greater amounts of contaminant oxidation compared to chemical activation, which occurs at ambient temperatures (i.e., 16 to 25 °C). Additionally, sodium persulfate could be utilized at sites undergoing in situ thermal treatment, where sodium persulfate injection during the period of declining contaminant mass recovery has the potential to provide a cost-effective polishing step which could reduce overall treatment times. Heat-activated sodium persulfate decomposition involves destabilizing the dioxygen bond between the two sulfate groups, yielding two sulfate radical anions as shown in Scheme 1.

SCHEME 1

Persulfate decomposition initiates the formation of sulfate radical anions (SO4- · ) leading to a series of free-radical reactions involving the formation of other radical species (i.e., propagation) (6). While persulfate decomposition is a first-order reaction at constant pH, it also depends on hydrogen ion concentrations (i.e., second order) (7), which explains the observed increase in the rate of persulfate decomposition via acid catalysis in unbuffered systems (8). During decomposition, the sulfate radical anions react with oxidizable compounds at differing rates, resulting in competition for radical anions (8). For example, the presence of chloride at concentrations greater than 20 µM was found to interfere with the oxidation of dissolved organic carbon (9). Thus, when applied to subsurface environments, sulfate radical anions will initially react with the most rapidly degradable compounds, which may not be the target organic chemicals. To overcome this limitation, the total oxidant demand of soils, as measured during bench tests, is thought to represent the oxidant dosage sufficient to degrade all oxidizable compounds present, including the target organic contaminant (10, 11). Sodium persulfate has been shown to degrade a range of organic contaminants in aqueous solution; however, few experimental results are available with solids present. Thus, this study was performed to determine the rate and extent of PCE oxidation by sodium persulfate in solids-containing batch reactors. Experiments were completed with 10 solids ranging from glass beads, to reference sands and clays, to field soils. After determining the rate of PCE oxidation by heat-activated sodium persulfate, subsequent experiments involved repeatedly dosing vials with sodium persulfate solution to determine if the rate of PCE oxidation could be increased. Field soils were pretreated with excess sodium persulfate and also treated to remove carbonate and readily oxidizable fractions, in an effort to improve the rate of PCE oxidation. VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9445

Experimental Section Materials. Tetrachloroethylene (HPLC grade, >99.9%) was obtained from Aldrich (Milwaukee, WI) and used as received. Stock aqueous PCE solutions were prepared by adding small quantities (98%) was obtained from Sigma-Aldrich (St. Louis, MO) and was used to prepare a 32.14 mM (7.65 g/L) solution. This solution was maintained at pH 7.0 using a phosphate buffer comprising 30.5 mL of 0.2 M Na2HPO4 and 29.5 mL of 0.2 M NaH2PO4 (certified A.C.S., >99%) obtained from Fisher Scientific (Fair Lawn, NJ), and diluted to 100 mL with DI water (12). Solids. Borosilicate glass beads with a diameter of 3 ( 0.5 mm were obtained from Kontes (Vineland, NJ). Soda-lime glass beads with diameters ranging from 0.3 to 0.45 mm (40-50 mesh) were obtained from Potter Industries (Apex, NC). Two reference sands were obtained from U.S. Silica (Berkeley Springs, WV): F-70 Ottawa sand with grain diameters between 0.35 to 0.075 mm (50 to 200 mesh) and ASTM 20-30 mesh Ottawa sand with grain diameters between 0.85 to 0.6 mm (20 and 30 mesh). Great Lakes soil samples were obtained from a former dry cleaning facility located at the Naval Training Center Great Lakes (Site 22), Great Lakes, IL by personnel from by TetraTech NUS, Inc. (Pittsburgh, PA). Soil cores were collected from a single borehole (F3) between 8 and 12 feet below ground surface using a steel tube and then extruded into a plastic zip lock bag. The soil, classified as Ozaukee silty clay loam (fine, illitic, mesic Oxyaquic Hapludalfs), was gray, very sticky and plastic, and strongly effervescent upon addition of mild acid. Fort Lewis soil samples were collected from a single borehole (RS0047b), 28 to 30 feet below ground surface, at the East Gate Disposal Yard, Fort Lewis, WA. The soil consisted of well-graded gravel in a matrix of sand, silt, and clay that was deposited as glacial till and outwash with total carbon content (TOC) of 0.069 ( 0.005% or 0.69 g TOC/kg soil and a specific surface area of 8.8 ( 0.7 m2/g (13). Appling soil was collected at the University of Georgia Agricultural Experiment Station located near Eastville, GA. This loamy coarse sand is from the Appling series (clayey, kaolinitc thermic Typic Hapludult) and contained 0.77% organic carbon (OC) or 7.7 g OC/kg soil (14). Reference clays, including montmorillonite (SAz-1), kaolinite (KGa-1), and Illite-smectite (ISCz-1), were obtained from The Clay Minerals Society (Chantilly, VA). Batch Reactivity Experiments. The rate of PCE oxidation was determined using 8 mL glass vials sealed with Teflonlined silicone septa affixed with screw-top caps (Kimble Glass, Inc., Vineland, NJ). Water-only experiments consisted of filling each of 12 vials with 7 mL of PCE stock solution. After preheating the capped vials for at least 5 min in a recirculating water bath (Neslab, RTE-111), six vials received 1 mL of 32.14 mM phosphate buffered persulfate solution (e.g., 7.65 g/L), and the remaining six vials received 1 mL of DI water, which served as the controls. The vials were returned to the water bath, and pairs of vials (one containing persulfate and one control) were removed at regular intervals. After cooling for 5 min in an ice bath, a 1 mL aqueous sample was collected from each vial using a gastight syringe (Hamilton, Reno, NV) and transferred to a 22 mL headspace vial followed by a 0.1 mL sample that was transferred to a 1.5 mL ion chromatography vial. These water-only experiments were completed at 31.1, 40, 50, 60, and 70 °C to determine the rate of PCE oxidation as a function of temperature. The solids containing experiments consisted of loading 12 vials with ca. 6 g of airdried solids followed by 3.5 mL of PCE stock solution. The contents of the vials were then mixed for 30 s at 2600 rpm 9446

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

using a Fisher Scientific touch mixer. The Great Lakes soil experiments were completed using 8 g of a soil and groundwater slurry (1:1 mass ratio), where the soil was previously contaminated with PCE. The vials were preheated in the recirculating water bath operated at 50 °C for 5 min, after which time 0.5 mL of 32.14 mM phosphate-buffered persulfate solution was added to six vials (i.e., 0.64 g/kg dosage) and 0.5 mL of DI was added to each of the remaining six vials, which served as controls. The dosage for vials containing Great Lakes soils was higher; 1.5 mL of 32.14 mM phosphate buffered persulfate solution was added making the dosage 2.9 g/kg. The vials were then returned to the 50 °C water bath, and two vials (one persulfate and one control) were removed after 10 min and then quenched in an ice bath. Pairs of vial were subsequently removed after 20, 35, 55, 80, and 120 min and placed in the ice bath. After centrifuging at 1000 rpm for 10 min, a 1 mL aqueous sample was collected from each vial using a gastight syringe (Hamilton, Reno, NV) and transferred to a 22 mL headspace vial, followed by a 0.1 mL sample that was transferred to a 1.5 mL ion chromatography vial. Persulfate Oxidation of Field Soils. To determine if sodium persulfate demand limited the destruction of PCE in batch reactor experiments, Great Lakes, Fort Lewis, and Appling soils were repeatedly treated with sodium persulfate. The treatments consisted of dispensing 200 g of each soil into separate glass jars, adding 100 mL of the 7.65 g/L sodium persulfate solution buffered to pH 7.0, and heating each jar in a water bath operated at 90 °C. The jars were incubated until there was no sodium persulfate remaining as determined by sodium thiosulfate titration (see Section S.1, Supporting Information), which required ca. 12 h per treatment. This treatment was executed four sequential times, resulting in cumulative sodium persulfate treatments of 15.37, 15.25, and 15.34 g/kg for Great Lakes, Fort Lewis, and Appling soil, respectively. Analytical Methods. Aqueous samples were analyzed for PCE, trichloroethylene (TCE), cis-1,2-dichloroethylene (cDCE), and vinyl chloride (VC) content using a Hewlett-Packard gas chromatograph (GC) 6890 equipped with a tekmar HT3 headspace autosampler (Teledyne Technologies, Inc., Mason, OH) and a 30 m long by 0.25 mm outside diameter (OD) DB-5 column (Agilent, Santa Clara, CA) with 0.25 µm film thickness connected to a flame ionization detector (FID). The headspace autosampler was programmed to heat samples for 30 min at 70 °C prior to transferring 1 mL of headspace gas into the GC inlet through heated silcosteel tubing. The GC oven was maintained at 35 °C for 7 min and then increased at a rate of 10 °C/min to 60 °C. Calibration standards were prepared by injecting small volumes ( 79%, Table S.1, Supporting Information), the reactors were sequentially treated with additional 0.64 g/kg doses of persulfate solution (2.9 g/kg for Great Lakes soils) to determine if the total oxidant demand VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9447

FIGURE 3. The amount of PCE degraded and amount of persulfate consumed after 80 min at 50 °C for reference sands and clays, glass beads, and field soils.

FIGURE 4. Extent of PCE oxidation with sequential persulfate treatment after 80 min at 50 °C. Only the aqueous phase was replaced between subsequent treatments. could be satisfied and thereby improve the extent of PCE oxidation relative to that observed in the experiment with borosilicate glass beads. As shown in Figure 4, the extent of PCE oxidation after 80 min at 50 °C increased with each subsequent treatment for reactors containing lime glass beads, reference sands (F70 and 20-30 mesh Ottawa sands), and for two of the reference clays (kaolinite and montmorillionite). The extent of oxidation with lime glass beads and reference sands equaled the extent observed in experiments completed with borosilicate glass beads after 2 and 4 persulfate treatments, respectively. Therefore, under these reaction conditions (i.e., 50 °C and 80 min duration), sodium persulfate dosages of 1.3 g/kg for glass beads and 2.6 g/kg for the reference sands were necessary to overcome the inherent oxidant demand. Although the extent of PCE oxidation in experiments with kaolinite and montmorillionite increased with subsequent persulfate treatments, the extent of PCE oxidation did not reach the level observed with borosilicate glass beads. Sequential treatment of the field soils and Illite-smectite failed to yield similar improvements in PCE destruction as with the reference solids. With Fort Lewis soils, the extent of PCE degradation at 80 min averaged 92 ( 3% during all four treatments suggesting that the oxidant demand had been met after the initial sodium persulfate dose of 0.64 g/kg. This is consistent with the measured TOC of 0.69 g/kg for the Fort Lewis soil. A similar trend was noted for Illite-smectite where 9448

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

PCE degradation averaged 69 ( 5% over the four treatments. Thus, meeting the oxidant demand of Fort Lewis soil and Illite-smectite was not limiting the extent of PCE degradation. PCE oxidation was not detected in reactors with Great Lakes and Appling soils suggesting insufficient reaction time or temperature, or that insufficient sodium persulfate dosages were applied to satisfy the oxidant demand of these soils. To address the possibility of insufficient sodium persulfate dosage, each soil was treated with ca. 15 g/kg sodium persulfate over a 4 day period. Assuming an average molecular weight of 5000 g/mol for the NOM contained in these soils and a stoichiometry of 2:1 sodium persulfate to NOM indicates that Appling soil was treated with 21 times excess sodium persulfate and Fort Lewis soil was treated with 235 times the amount of sodium persulfate required to oxidize the NOM. Batch reactor experiments completed using these pretreated soils yielded results that were similar to those observed in Figure 4. There was no observable difference between the concentration of PCE in sodium persulfate treated and control vials containing Great Lakes and Appling soils, and 87% of the PCE was degraded after 80 min in reactors with Fort Lewis soil. Thus, even after treatment by excess sodium persulfate, the presence of Great Lakes and Appling soils prevented the oxidation of PCE by sodium persulfate. Field Soil Fraction. To investigate the fraction of Appling and Great Lakes soils that was inhibiting sodium persulfate from oxidizing PCE, the soils were treated to remove the carbonate and readily oxidizable fractions (see Section S.5, Supporting Information for methods and results). After removing carbonates from the Great Lakes and Appling soils using three sequential treatments of acetic acid, there was no detectable decrease in PCE concentrations after treatment with sodium persulfate at 50 °C (see Figures S.5 and S.6, Supporting Information). These data suggest that the presence of soil carbonates did not inhibit sodium persulfate oxidation of PCE, leaving soil NOM as a potential inhibiting fraction. The humic fraction of the Great Lakes and Appling soils was extracted following the International Humic Substances Society method (see Section S.5.2, Supporting Information for methods and results). The Great Lakes soil extracts were found to contain fulvic and humic acids at concentrations of 16.8 mg/L and 0.6 mg/L, respectively, while Appling soil extracts contained fulvic acids at 5.3 mg/L and humic acids at 27.8 mg/L. Thus, both of these soils contained appreciable humic substance fractions that will increase the soil oxidant demand. To demonstrate this, an experiment was completed using sodium persulfate to oxidize PCE in aqueous solutions containing fulvic and humic acid extracts obtained from the Great Lakes and Appling soils and in solutions containing Suwannee River fulvic and humic acids (see Section S.5.2, Table S.2, Supporting Information). The vials containing humic substances were dosed with sufficient sodium persulfate (>2000 g/kg) to result in their complete destruction in addition to the destruction of PCE. However, after 55 min at 50 °C, PCE degradation was less than 26% with no PCE degradation detected in vials containing Suwannee River fulvic or dissolved humic acids. The presence of humic substances had little effect on the amount of sodium persulfate consumed in these water-only experiments with less than 2.6% consumed based on sulfate concentrations and less than 0% based on thiosulfate titration. Given that only a small fraction of sodium persulfate had reacted at 50 °C, increasing the temperature was expected to increase the amount of sodium persulfate reacting. These experiments were repeated at 90 °C resulting in an increase in the degradation of PCE with greater than 96% destruction with Suwannee River humic substances (see Section S.5.2, Table S.3, Supporting Information) after 55 min. There were also

substantial increases in the amount of persulfate consumed with greater that 65.8% based on sulfate concentrations and 68.5% by thiosulfate titration. These findings indicate that sodium persulfate dosage is an important consideration for PCE oxidation when NOM is present, but temperature or the rate of sodium persulfate decomposition may also play an important role. Since the humic substances in the Great Lakes and Appling soils inhibited PCE oxidation by sodium persulfate, removal of the NOM fraction from these soils was expected to increase the rate of PCE oxidation by heat-activated sodium persulfate. To demonstrate this, Great Lakes and Appling soils were repeatedly treated with a 30% hydrogen peroxide solution until effervescence was no longer observed, a standard end point for removing the readily oxidizable fraction of soil (15). After removing the readily oxidizable soil fraction with hydrogen peroxide treatments, batch reactor experiments were completed using these pretreated Great Lakes and Appling soils. The results of the batch experiments were similar to those observed in Figure 4 in that there was no detectable decrease in PCE concentrations following the addition of sodium persulfate at 50 °C (see Figures S.7 and S.8 Supporting Information).

Discussion Oxidation of PCE by heat-activated sodium persulfate in water-only reactors was a first-order reaction with the rate being a function of temperature at constant sodium persulfate concentration. Introduction of solids into the aqueous medium in which the reaction takes place slowed the rate of reaction with the exception of borosilicate glass beads. Slowing of the reaction rate may have been due to the presence of oxidizable species other than PCE that were introduced with the solids (i.e., inherent oxidant demand). However, the solids may have also affected the reaction rate in other ways such as by sequestering the target organic through adsorption or by interfering with the frequency of collision between the oxidant and PCE. Repeatedly dosing the reference solids (i.e., glass beads, sands, and clays) with sodium persulfate increased the rate and extent of PCE oxidation. This indicates that overcoming the inherent oxidant demand for these reference materials led to improvements in the rate of PCE oxidation by heatactivated sodium persulfate. However, this was not the case with the field soils where there was little improvement in the oxidation rate after repeated sodium persulfate doses. Pretreating the field soils with ca. 15 mg/kg of sodium persulfate or 30% hydrogen peroxide failed to improve the rate of PCE oxidation. This result suggests that overcoming the inherent oxidant demand for these field soils did not limit the oxidation of PCE by heat-activated sodium persulfate. Regarding sequestration of PCE by adsorption, the concentration of PCE in reactors with Great Lakes and Appling soils represented 77 ( 3% of the concentration of PCE added to these reactors. Likewise, the concentrations of PCE in the reactors with humic substances were within 7% of those in the control reactors. While a fraction of PCE was not available for reaction, the majority of PCE remained in the aqueous phase. The concentration of PCE in all experiments was determined by headspace analysis which consisted of heating a 1 mL aqueous sample to 70 °C for 30 min followed by automatically collecting a gas sample from the sample vial. Given that the temperatures used in these experiments were similar to that of the headspace analysis, the detection of PCE in the solids containing control vials demonstrates that PCE was available in the aqueous phase under reaction conditions and that solid-phase adsorption of PCE was not responsible for reducing the rate of heat-activated sodium

persulfate oxidation of PCE in reactors containing Great Lakes and Appling soils. Based on these results, in situ treatment of Great Lakes and Appling soils with heat-activated sodium persulfate is not anticipated to result in substantial PCE oxidation, while in situ treatment of Fort Lewis soils is anticipated to result in PCE oxidation. This work demonstrates the need to perform soil-specific contaminant treatability tests rather than soil oxidant demand tests when determining oxidant dosage requirements at potential field sites.

Acknowledgments The authors thank Kyra P. Lynch and Jeff Powers, United States Corps of Engineers, Seattle District, and Robert Davis, Jr., Tetra Tech NUS, Inc. for providing access to soil and groundwater samples. Support for this research was provided by the Strategic Environmental Research and Development Program (SERDP) under contract W912HQ-05-C-008 for Project ER-1419, “Investigation of Chemical Reactivity, Mass Recovery and Biological Activity During Thermal Treatment of DNAPL” and by a GeorgiaTech Presidential Undergraduate Research Award to John Callaghan. This work has not been subject to SERDP review, and no official endorsement should be inferred.

Supporting Information Available Arrhenius rate plot, similarity between degradation kinetics obtained from sacrificial and stirred batch reactors, PCE degradation and persulfate consumption during the initial persulfate treatment, solids oxidant demand, soil fractionation methods and results, and first-order kinetic model plots for each sequential persulfate treatment. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Huang, K. C.; Zhao, Z.; Hoag, G. E.; Dahmani, A.; Block, P. A. Degradation of volatile organic compounds with thermally activated persulfate oxidation. Chemosphere 2005, 61, 551–560. (2) Pound, M.; White, B.; Wong, R.; Abrams, S.; Block, P. Bench and pilot scale treatment of trichloroethylene by activated persulfate. Presented at Remediation of Chlorinated and Recalcitrant Compounds, The 5th International Conference, Monterey, CA, 2006. (3) Waldemer, R. H.; Tratnyek, P. G.; Johnson, R. L.; Nurmi, J. T. Oxidation of chlorinated ethenes by heat-activated persulfate: kinetics and products. Environ. Sci. Technol. 2007, 41, 1010– 1015. (4) Johnson, R. L.; Tratnyek, P. G.; Johnson, R. O. Persulfate persistence under thermal activation conditions. Environ. Sci. Technol. 2008, 42, 9350–9356. (5) Costanza, J.; Pennell, K. D. Distribution and Abiotic Degradation of Chlorinated Solvents in Heated Field Samples. Environ. Sci. Technol. 2007, 41, 1729–1734. (6) Peyton, G. R. The free-radical chemistry of persulfate-based total organic carbon analyzers. Mar. Chem. 1993, 41, 91–103. (7) Kolthoff, I. M.; Miller, I. K. The chemistry of persulfate. I. The kinetics and mechanism of the decomposition of the persulfate ion in aqueous medium. J. Am. Chem. Soc. 1951, 73, 3055–3059. (8) Goulden, P. D.; Anthony, D. H. J. Kinetics of uncatalyzed peroxydisulfate oxidation of organic materials in fresh water. Anal. Chem. 1978, 50, 953–958. (9) Alken, G. R. Chloride interference in the analysis of dissolved organic carbon by the wet oxidation method. Environ. Sci. Technol. 1992, 26, 2435–2439. (10) Haselow, J. S.; Siegrist, R. L.; Crimi, M.; Jarosch, T. Estimating the total oxidant demand for in situ chemical oxidation design. Remediation 2003, 13 (4), 5–16. (11) Johnson, R. L.; Tratnyek, P. G.; O’Brien-Johnson, R. Persulfate persistence under thermal activation conditions. Environ. Sci. Technol. 2008, 42, 9350–9356. VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9449

(12) Perrin, D. D.; Dempsey, B. Buffers for pH and metal ion control. Chapman and Hall, Wiley (13) Costanza, J.; Fletcher, K. E.; Loffler, F. E.; Pennell, K. D. Fate of TCE in heated Fort Lewis soil. Environ. Sci. Technol. 2009, 43, 909–914. (14) Karagunduz, A.; Pennell, K. D.; Young, M. H. Influence of a nonionic surfactant on the water retention properties of

9450

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

unsaturated soils. Soil Sci. Soc. Am. J. 2001, 65, 13921399. (15) Jackson, M. L. Soil Chemical Analysis: Advanced Course, Revised 2nd ed.; Parallel Press, University of Wisconsin-Madison Libraries: Madison, WI, 2005.

ES100997A