Electrolytic Redox and Electrochemical Generated Alkaline Hydrolysis

Jul 13, 2009 - Croce , M.; Okamoto , Y. Cationic micellar catalysis of the aqueous ...... Ali Ciblak , Joniqua Howard , Ingrid Padilla , Rita Loch-Car...
0 downloads 0 Views 3MB Size
Environ. Sci. Technol. 2009, 43, 6301–6307

Electrolytic Redox and Electrochemical Generated Alkaline Hydrolysis of Hexahydro-1,3,5-trinitro-1,3,5 triazine (RDX) in Sand Columns DAVID B. GENT,† ALTAF H. WANI,‡ JEFFREY L. DAVIS,† AND A K R A M A L S H A W A B K E H * ,§ Environmental Laboratory, U.S. Army Engineer Research and Development Center, 3909 Halls Ferry Road, Vicksburg, Mississippi 39180, Environmental Resources Management (ERM), 15810 Park Ten Place, Suite 300, Houston, Texas 77084, and Department of Civil and Environmental Engineering, Northeastern University, 360 Huntington Avenue, Boston, Massachusetts 02115

Received December 16, 2008. Revised manuscript received June 19, 2009. Accepted June 25, 2009.

Sand-packed horizontal flow columns (5 cm i.d. × 65 cm) were used in laboratory experiments to simulate in situ electrolytic and alkaline hybrid treatment zone for aqueous phase decomposition of RDX. An upgradient cathode and downgradient anode, spaced 35 cm apart, were used to create alkaline reducing conditions followed by oxic, acidic conditions to degrade RDX by combination of alkaline hydrolysis and direct electrolysis. A preliminary experiment (25 mg/L RDX influent) with seepage velocity of 30.5 cm/day and current density of 9.9 A/m2 was used to determine the treatment feasibility and the aqueous products of RDX decomposition. Three additional column experiments (0.5 mg/L RDX influent) under the same conditions as the preliminary column were used to observe the treatment process repeatability and the alkaline treatment zone development. The results demonstrated approximately 95% decomposition of RDX in the column with an applied current density of 9.9 A/m2. Aqueous endproducts formate, nitrite, and nitrate were detected in the effluent. Approximately 75% of the RDX was destroyed near the cathode, presumably by electrolysis, with 23% decomposed downstream of the cathode by alkaline hydrolysis. The preliminary column pseudo first order alkaline hydrolysis rate coefficient of 1 0.7 × 10-3 min-1 was used to estimate a treatment zone length less than 100 cm for RDX treatment below the EPA drinking water lifetime health advisory of 0.002 mg/L.

Introduction Hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) forms the basis for many common military explosives: Composition A, Composition B, Composition C, HBX, H-6, Cyclotol, and Amatex. Manufacturing, assembling and loading, storage, disassembly of out-of-date munitions, demolition, and open burning of nitroaromatic and nitramine explosive com* Corresponding author e-mail: [email protected]; phone: 617 373 3994. † U.S. Army Engineer Research and Development Center. ‡ Environmental Resources Management. § Northeastern University. 10.1021/es803567s CCC: $40.75

Published on Web 07/13/2009

 2009 American Chemical Society

pounds may have contributed to contamination of soil and groundwater. The United States Environmental Protection Agency (U.S. EPA) lists RDX as a contaminant under the Unregulated Contaminant Monitoring Regulation for Public in List 2. The U.S. EPA recommends a drinking water lifetime health advisory of 0.002 mg RDX per liter for exposure to RDX (1). Remediation technologies used in treating RDX-contaminated groundwater historically focused on extraction and ex situ treatment or in situ manipulation of oxidation/ reduction (redox) conditions by injecting electron-donor to facilitate bioremediation, abiotic treatment, or a combination of abiotic/biotic treatment. Processes used to treat RDX include granular activated carbon (GAC) adsorption (2), alkaline hydrolysis (3-6), zerovalent iron reduction (7-9), and electrochemical reduction (10-12). In situ processes include biological treatment (13-17) and sequential electrolytic redox (18, 19). Direct electrolytic treatment of dinitrotoluene (DNT), 2,4,6 trinitrotoluene (TNT), and RDX showed that decomposition rates increased with increased current (20). RDX-contaminated water can be electrochemically reduced to hexahydro1-nitroso-3,5-dinitro-1,3,5-triazine (MNX) followed by complete transformation of RDX to small molecules such as formaldehyde, methanol, nitrite, and nitrous oxide (Table 1, eqs 1a and 1b) (12). Wani et al. (18) demonstrated 70% transformation of unlabeled RDX in a single pass through a closely spaced electrochemical barrier (e- barrier) in a sand column. Approximately 20% of radiolabeled 14C-RDX mineralized to 14CO2 and 43% of the total activity in the dissolved phase was unidentified transformation products with no nitroso intermediates (MNX, hexahydro-1,3-dinitroso-5nitro-1,3,5-triazine, DNX, and hexahydro-1,3,5-trinitroso1,3,5-triazine, TNX) present with an activity balance of 92%. In sequential electrochemical redox (18, 19) the role of alkaline hydrolysis at the cathode was ruled out because the electrodes were closely spaced (2.5 cm spacing) and the process maintained relatively neutral groundwater pH conditions. Alkaline hydrolysis can cause breakdown of nitramine explosives (21). Nitramines readily undergo base hydrolysis at pH greater than 10.5; but, the reaction kinetic rates are much faster when the pH exceeds 11.5 (6). Nitroso-derivatives, MNX, DNX, and TNX intermediate products associated with anaerobic biodegradation (28) are not formed with alkaline hydrolysis decomposition of RDX. Identified products of RDX alkaline hydrolysis are nitrite, nitrous oxide, ammonia, formate, and formaldehyde (Table 1, eq 2; 3, 4, 22). Formaldehyde, subject to a Cannizzarro-reaction at elevated pH, oxidizes to formate (23) (Table 1, eq 3). Effective treatment (99% removal) of RDX by alkaline hydrolysis was demonstrated in a continuous stirred tank reactor (CSTR) with nitrite and formate end products (6). In-situ development of alkaline conditions can be achieved by application of low-level direct current (dc). Electrolysis of flowing groundwater results in the generation of an alkaline (reducing environment) at the cathode and an acid (oxidizing environment) at the anode (Table 1, eqs 4 and 5). Consequently, the cathode pH may increase to greater than 12 and the anode pH may decrease to less than 3, depending on electric current and rate of hydraulic flow. Redistribution of the acid and base environments will occur in groundwater because of ion migration, chemical diffusion, and groundwater flow (24). By placing the cathode upstream of the anode, an alkaline front will travel downstream toward the anode by advection, dispersion, and ion migration (Figure 1). The generation of a highly reducing zone beginning at the cathode VOL. 43, NO. 16, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6301

TABLE 1. Electrolytic and Redox Reactions Involved in Reduction of RDX

will initiate decomposition of aqueous RDX. Electrons added to an RDX molecule during reduction destabilize it, resulting in ring cleavage to small organic and inorganic ions (12). At the cathode, nitrite may convert to ammonia and nitrous oxide (Table 1, eqs 6 and 7). Formaldehyde will convert to formate by cathodic oxidation and to formic acid by anodic oxidation (Table 1, eqs 3 and 8). Additional oxidation reactions at the anode can mineralize formate to carbon dioxide (Table 1, eqs 9 and 10), and nitrite to nitrate (Table 1, eq 11). This study reports the role of electrolytic degradation and electrolytic generated alkaline hydrolysis in remediation of RDX-contaminated water in porous media. This process was accomplished by separating the electrodes and measuring the changes in concentration between the anode and cathode wells under flow conditions. Test setups consisted of horizontally mounted clear polyvinyl chloride (PVC) columns (65 cm × 5 cm i.d.) containing a single set of electrodes. Both ends of the column were closed with PVC blind-flange fittings screened with porous (100 µm) polyethylene (PE) material. The electrodes were installed in PVC #10 slotted (0.254 mm openings) well screens (2.1 cm diameter) and separated by 35 cm of sand. The electrodes consisted of mixed metal oxide catalyst sintered to an expanded titanium mesh (Corrpro Companies, Inc., Medina, OH). The electrode wells occupied 51% of the column cross-sectional area with open areas remaining on either side of the electrode well allowing some of the flow to bypass the electrode (Figure 2, blow-up diagram). The washed well-pack silica sand (20 × 40 Oglebay Norton Industrial Sand, Brady, TX) had an effective size between 45 and 48 mm and was poorly graded (according to the Unified Soil Classification System) with a uniformity coefficient of 1.25, and a D50 of 0.65 mm. Aqueous samples were collected from the influent, effluent, electrode wells, and porous Teflon filter ports inserted along the length of the column (Figure 2). Measurements of the physical properties of the sand included constant head hydraulic conductivity (ASTM D2434) and chloride tracer tests (Table 2). Chloride tracer tests 6302

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 16, 2009

evaluated the hydrodynamic properties, dispersivity and porosity, using a one-dimensional steady-state solution to the convective-dispersive transport equation (25). For the tracer tests, a 50 mg/L chloride concentration was pumped through the column and the effluent was periodically sampled with an Eldex UP-50 universal fraction collector (Eldex Laboratories, Inc., Napa, CA). The sand had a dispersivity of 20 cm, porosity of 0.37, and a hydraulic retention of 50.8 h based on analysis from the tracer tests. Individual column flows and seepage velocities are presented in Table 2. RDX-contaminated water was prepared by spiking organicfree reagent-grade water (6 µS/cm, pH ) 6.8) with RDX stock solution procured from Holston Ammunition Plant (Kingston, TN). The influent water contained 100 mg/L of calcium chloride as chloride with either 0.5 or 25 mg/L of RDX with a target flow of 9.3 mL/h. The electric power input for the preliminary high-concentration experiment was set at 4.9 and 9.9 A/m2 for 60 and 30 days, respectively (Table 2). The dc power supply used in both high-concentration experiments was an Agilent model E3612A (60-120 V/.5-.25 A Dual) operated in constant current mode. The power input for the 0.5 mg/L RDX influent concentration experiments was set at a current density of 9.9 A/m2 using two dc power supplies (Sorensen DCS 300 V-3.5A) connected in series. Each experiment began by pumping the influent water through the sand-filled columns until the effluent RDX concentration matched the influent concentration with a valveless piston pump attached to a variable speed L/S Masterflex drive. After RDX breakthrough, the treatment process began by activating the power supplies. Two experiments conducted with a high RDX concentration (25 mg/L) determined treatment efficiency and aqueous products of RDX decomposition and separated electrolytic reduction from alkaline hydrolysis. The high RDX concentration was required to ensure that end-product concentrations would be detectable by the analytical methods available. The first of these experiments used two levels of current density (4.9 and 9.9 A/m2) to assist in the detection of RDX

FIGURE 1. A schematic of electrochemical barrier: (a) in situ application of a direct current across a cathode and anode; (b) placement of the cathode upstream (groundwater flow) of the anode; (c) RDX destruction at the cathode by direct electrolysis; (d) indirect electrolysis of RDX downstream of the cathode by alkaline hydrolysis migrating toward the anode; and (e) oxidation at the anode.

FIGURE 2. Drawing of experimental column. decomposition products by electrolytic and alkaline hydrolysis. In the second experiment, 0.0032 M sulfuric acid was injected into the column at a rate of 1.3 mL/h 2 cm downstream of the cathode well. The low pH from the sulfuric acid addition quenched potential transformation of RDX by alkaline hydrolysis. Weekly sample analysis from all ports included pH, RDX, nitroso decomposition produces, formate, nitrite, and nitrate. Three low-RDX-concentration (0.5 mg/L) column experiments were conducted under the same conditions as the high-

concentration experiments. The low influent concentration represents concentrations usually found in aquifers contaminated with RDX. These experiments were to show treatment repeatability and observe the alkaline front development with time. The hydroxide transport through the column was measured in-line with pH electrodes (Sensorex S200CD) housed in low-flow Teflon mounts cross-circulated from the front to the back of the column with low flow peristaltic pumps (Ismatec IP, Glattbrugg Switzerland). The pH readings were collected every 10 min with a Fuji Electric PHF Chart Recorder. VOL. 43, NO. 16, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6303

TABLE 2. Column Operation Parameters low concentration (0.5 mg/L) experiment current (mA) current density (A/m2) column flow (mL/h) Seepage velocity (cm/h) duration (d)

25 mg/L (two current settings) 10 4.9 9.3 ( 0.3 1.31 60

20 9.9 10.6 ( 0.8 1.49 30

acidified column 25 mg/L

A

B

C

20 9.9 10.6 ( 0.8 1.49 62

20 9.9 9.5 ( 0.3 1.23 44

20 9.9 9.6 ( 0.3 1.21 82

20 9.9 9.5 ( 0.3 1.23 82

RDX, MNX, DNX, and TNX calibration standards were obtained from Supelco (Boston, MA). A modified EPA Method 8330 using a Dionex HPLC system (Sunnyvale, CA) consisted of a P580 fluid pump, ASI-100 autosampler, and UVD340U absorbance detector for the analysis of RDX and its nitrososubstituted transformation products. The injection volume was 25 µL and analyte absorbance was monitored at 254 nm. Chemical separation was achieved with a Supelco CN reversephase HPLC column (250 × 4.6 mm). Cumulative effluent samples (∼500 mL) analyzed using solid-phase extraction lowered the laboratory reporting limits from 0.02 mg/L of RDX to 0.002 mg/L. Inorganic and organic aqueous phase concentrations of nitrite, nitrate, and formate were analyzed using a Dionex (Sunnyvale, CA) Ion Chromatograph. Chemical separation and detection were achieved using an Ionpac AS11 analytical column and a Dionex conductivity detector. The pH of the samples was measured with Sensorex S200CD epoxy body combination electrodes. Oxidationreduction potential (ORP) was measured in the electrode wells, with the power off, using a Cole-Parmer combination Ag/AgCl redox electrode. ORP and pH measurements were taken with an Oakton WD-35100-00 model pH/ORP controller (Cole-Parmer, Vernon Hills, IL) with a measurement range of 0 to 14 for pH and -1250 to 1250 mV ORP. The Eh values were calculated by adding the standard potential of the reference electrode (202 mV at 25 °C) to the measured potential.

Results and Discussion High Concentration Experiment. The alkaline front developed in all experiments and migrated under hydraulic flow from the cathode to the anode. The pH values of the influent, cathode well, and effluent averaged 6.2 ( 1.2, 12.1 ( 0.4, and 3.2 ( 0.52, respectively. Adjusting the current density from 4.9 to 9.9 A/m2 increased hydroxide production at the cathode causing the pH in the column to rise from 11.5 to 11.8 (Figure 3a). The increase in current density resulted in an 80% increase in the hydroxide concentration from 6.3 to 8.7 mM. The strongest reducing and oxidizing conditions for all columns developed at the cathode (Eh ) -0.72 ( 0.09 V) and the anode (Eh ) 1.23 ( 0.07 V). These Eh values are consistent with electrode potentials reported by electrolysis under similar conditions by refs 18, 19, 26, and 27. The redox potential in the electrode wells remained steady even though the current density changed because the electric potential supplied to the electrodes was sufficient to drive the fluid to the lower and upper limits of water stability, Table 1, eqs 4 and 5. The measured cathode well redox values were within 92% of the theoretical values based on a modified Nernst equation (Eh ) E0 + 0.059 pH) and the actual cathode well pH. In the first experiment, RDX influent concentration (25.7 ( 1.0 mg/L) decreased to 3.4 ( 1.1 and 2.6 ( 1.8 mg/L in the effluent under the two current settings (Table 3) with removal percentages of 86.42 ( 4.2 and 89.7 ( 7.3. The RDX passing the cathode well untreated decreased from 6.9 ( 1.8 to 2.7 ( 1.3 mg/L in the column effluent apparently by alkaline 6304

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 16, 2009

FIGURE 3. Profile of (a) pH and (b) RDX vs distance from inlet by selected day. hydrolysis. The increased RDX concentration (Figure 3b) in the first port immediately after the cathode well shows that some of the contaminant bypassed the electrode well untreated. The estimated RDX decomposed by the cathode was 72.9 ( 6.9% of the influent with the remaining 22.3 ( 9.7% removed between the electrodes presumably by alkaline hydrolysis. Weekly RDX analysis from the ports between the electrodes and the anode well were fitted with a two-parameter pseudo first order decay model to calculate the reaction rate coefficients for both current settings. These rate coefficients with units of cm-1 were converted to a per time basis using the one-dimensional steady-state solution to the convectivedispersive-reactive transport equation (25) to determine a treatment zone length required to decompose RDX from 0.5 mg/L to the EPA drinking water health advisory (0.002 mg/ L) at a seepage velocity of 30.5 cm/day. The average rate coefficients for the 10 and 20 mA data sets are 1.4 ( 0.3 × 10-3 and 1.8 ( 0.5 × 10-3 min-1. Based on these data, treatment zone lengths of 89 and 69 cm would be required

TABLE 3. Concentration of Initial and Effluent RDX, Removal Efficiency, Initial and Final Voltage Electric Potential, and Volt Gradient for Each Column 25 mg/L

RDX (mg/L) RDX removal (%) electric potential (V) voltage gradient (V/cm)

influent effluent initial final initial final

low concentration (0.5 mg/L)

10 mA

20 mA

acidified column 25 mg/L

A

B

C

26.0 ( 1.3 3.4 ( 1.1 86.8 ( 4.2 136.6 46 4.1 1.4

25.5 ( 0.7 2.6 ( 1.8 89.7 ( 7.3 76 132 2.4 4.0

25.1 ( 0.5 5.3 ( 1.5 74.6 ( 8.5 132.4 65 113.8 24.8

0.50 ( 0.01 0.01 ( 0.00 98.7 ( 0.01 605 162 18.3 4.9

0.51 ( 0.01 0.03 ( 0.01 94.4 ( 2.4 605 87 18.3 2.6

0.539 ( 0.00 0.04 ( 0.01 93.2 ( 2.4 603 100 18.3 3.0

to decompose RDX from 0.5L to 0.002 mg/L at a seepage velocity of 30.5 cm/day. Intermediate and End Products. As expected, nitrososubstituted transformation products, MNX, DNX, and TNX were not detected in the column or effluent by either liquid samples or solid phase extraction under either current setting. Various concentrations of formate, nitrite, and nitrate were detected in the column as RDX decomposed. Figure 4a,b,c illustrates formation of formate, nitrite, and nitrate axially

along the column from 4 selected days. The average effluent concentrations of formate, nitrite, and nitrate were 2.9 ( 0.6, 1.5 ( 0.8, and 7.5 ( 2.3 mg/L, respectively. The increase in formate at the anode well (Figure 4a) indicates incomplete conversion of formaldehyde to formate via the Cannizzarroreaction in the alkaline region of the column between the electrode wells. More complete oxidation of formaldehyde to formic acid occurred under acidic conditions produced by the anode. Nitrite averaged 2.6 ( 0.9 mg/L in the influent

FIGURE 4. Profiles of (a) formate, (b) nitrite, and (c) nitrate with distance from inlet. VOL. 43, NO. 16, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6305

TABLE 4. Molar Ratios of RDX End Products from High RDX Concentration Column formate

nitrite

nitrate

influent cathode port between C & A anode well effluent

10 mA 0.0 ( 0.1 0.2 ( 0.0 0.2 ( 0.0 0.9 ( 0.2 0.5 ( 0.1

0.5 ( 0.2 0.6 ( 1.0 0.7 ( 0.1 0.2 ( 0.1 0.3 ( 0.2

0.1 ( 0.1 1.5 ( 2.0 0.1 ( 0.1 3.3 ( 2.4 1.1 ( 0.4

influent cathode port between C & A anode well effluent

20 mA 0.1 ( 0.1 0.2 ( 1.0 0.0 ( 0.0 0.9 ( 0.2 0.7 ( 0.2

0.5 ( 0.1 0.6 ( 1.0 0.7 ( 0.1 0.4 ( 0.1 0.4 ( 0.1

0.1 ( 0.1 0.1 ( 1.0 0.1 ( 0.1 1.8 ( 0.9 1.2 ( 0.3

(Figure 4b), increased to an average of 4.0 ( 0.7 mg/L in the pore fluid between the electrodes because of RDX decomposition by both direct electrolysis and alkaline hydrolysis. The decrease in nitrite to 1.0 ( 0.5 mg/L in and after the anode well may have been from anodic oxidation to nitrate. Nitrate not detected in the column prior to the anode increased from zero to 10.3 ( 3.0 mg/L in the anode well confirming the oxidation of nitrite to nitrate (Figure 4c). Unidentified gaseous decomposition products may have included carbon dioxide, nitrous oxide, and nitrogen as well as formaldehyde. The average end-product molar ratio results are listed in Table 4 by column region and current setting. The molar ratios presented here are similar to those reported from alkaline hydrolysis literature (5, 22, 28). Acid Injection Column to Distinguish Electrolysis from Alkaline Hydrolysis. The second high RDX concentration experiment evaluated the difference between electrochemical reduction and alkaline hydrolysis. In this experiment, 0.0032 M sulfuric acid was injected into the column at a rate of 1.3 mL/h at a point 2 cm downstream of the cathode well. The low pH from sulfuric acid addition quenched further transformation of RDX by alkaline hydrolysis. One-way analysis of variance statistical analysis showed no statistical difference among the RDX concentrations from the weekly column port samples downstream of the cathode well (P ) 0.599). Analysis indicates that 74.6 ( 8.5% of the RDX transformed in the cathode well and the 2-cm zone between the cathode and acid injection port with no additional RDX transformed in the column. Calculations using kinetic data from alkaline hydrolysis of RDX under similar flow experiments (29) estimate that 8.3% of the RDX removed from the 2-cm region between the cathode well and the acid injection port may be attributed to alkaline hydrolysis. These results concur with the results from ref 18 where 70% RDX transformed in a 10-cm diameter column experiment by electrolysis alone using an applied current of 20 mA and a current density of 2.5 A/m2. Aqueous end products could not be determined from this experiment because of the high sulfate concentration from the acid addition. Three additional column experiments determined the validity and repeatability of the treatment process using RDX concentrations (0.5 mg/L) comparable to those found in actual contaminated aquifers. The hydroxide generated at the cathodes migrated through the columns by electrochemical (ion) migration and hydraulic flow toward the anodes (Figure 5). The hydroxide migrated past each sampling port (with a pH < 11) on an average 1.8, 3.6, 4.3, and 0.3 times faster than the measured hydraulic flow (Figure 5). The hydroxide transport past Port 4 in each column was retarded by hydronium ion production at the anodes migrating toward the cathode. Figure 5 illustrates alkaline front migration over the initial 80 h for Column A. The alkaline 6306

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 16, 2009

FIGURE 5. Port pH vs time in initial 80 h of treatment for triplicate Column A. front eventually overcame the acid front forcing it back toward the anode wells after 240 h for Column A and approximately 144 h for both Column B and C. After 250 h, the pH within each column stabilized at 11.5 or higher at all ports for the remainder of the experiments. The effluent pH decreased initially from 6 to 3 after 36 h and remained between 2.8 and 3.2 over the remainder of each experiment. The initial electric potential 600, 421, and 337 V for each column decreased to less than 160 V after 44 days of operation. The difference among the startup voltages was because each column contained unique properties such as dispersivity, porosity, and flow. Column A experiment ended after 44 days while Column B and C experiments ended after 82 days with final voltages of 160, 87, and 100 V, respectively. The initial and final voltage gradients for Columns B and C averaged 11.5 and 3 V/cm (Table 3). The initial high electric potentials for all experiments decreased with time as the fluid became more ionized increasing the electrical conductivity. The effluent analysis after 18 days was performed by solid phase extraction (SPE) because RDX concentrations dropped below 0.02 mg/L. Average RDX results by column were (A) 0.007 ( 0.004, (B) 0.034 ( 0.012, and (C) 0.016 ( 0.036 mg/L with removal percentages of 98.7, 94.3, and 93.2 (Table 3).

Summary An average of 95% of the influent RDX concentration (0.5 mg/L) can be decomposed under an electric current density of 9.9 A/m2 in a horizontal sand column with a seepage velocity of 30.5 cm/day. The results from low current setting in the high RDX concentration experiment were ineffective in aiding in the detection of nitroso-substituted RDX transformation products. Aqueous end products detected from RDX decomposition were formate, nitrite, and nitrate. Alkaline hydrolysis may be responsible for 23% of the RDX decomposed in an electrically charged system. The hydroxide ions migrated though the column at rates up to four times faster than under hydraulic flow. These results confirm the potential application of an electrolytic generated alkaline system for complete decomposition of RDX in an in situ application. Further research will be required to optimize the electrode spacing, power requirement, alkaline hydrolysis reaction rates and to delineate portions of degradation because of electrolytic and hydrolytic processes. The application of this technology may be limited to aquifers containing low dissolved solids because of potential precipitation clogging pore spaces. The effect of the process on minerals dissolution and/or precipitation,

which may be an issue in the presence of clay minerals, was not evaluated.

Acknowledgments This research work was conducted at the Environmental Laboratory, U.S. Army Engineer Research and Development Center (ERDC), Vicksburg, MS. We gratefully acknowledge the financial support provided by U.S. Army Environmental Quality Technology Research and Development Program. The findings and conclusions reported in this publication are exclusively of the authors and do not necessarily reflect the views of U.S. Army. The use of trade and product names is for descriptive purposes only and does not constitute endorsement by the U.S. Government.

Literature Cited (1) U.S. Environmental Protection Agency. Edition of the Drinking Water Standards and Health Advisories; EPA 822-R-04-005; Office of Water: Washington, DC, 2004. (2) Wujcik, W. J.; Lowe, W. L.; Marks, P. J. Granular activated carbon pilot treatment studies for explosives removal from contaminated groundwater. Environ. Prog. 1992, 11, 178l 89. (3) Balakrishnan, V. K.; Halasz, A.; Hawari, J. Alkaline hydrolysis of the cyclic nitramine explosives RDX, HMX and CL20: New insights into degradation pathways obtained by the observation of novel intermediates. Environ. Sci. Technol. 2003, 37, 1838– 1843. (4) Croce, M.; Okamoto, Y. Cationic micellar catalysis of the aqueous alkaline hydrolyses of 1,3,5-triaza-1,3,5-trinitrocyclohexane and 1,3,5,7-tetraaza-1,3,5,7-tetranitrocyclooctane. Org. Chem. 1979, 44 (13), 2100–2103. (5) Heilmann, H. M.; Wiesman, U.; Strenstrom, M. K. Kinetics of alkaline hydrolysis of high explosives RDX and HMX in aqueous solution and adsorbed to activated carbon. Environ. Sci. Technol. 1996, 30, 1485–1492. (6) Hwang, S.; Min, K.-S.; Davis, J. L. Comparative assessment of nucleophilic alkaline hydrolysis for remediation of high explosives-contaminated groundwater. Environ. Eng. Res. (Korean) 2004, 9 (1), 13–22. (7) Wanaratna, P.; Christodoulatos, C.; Sidhoum, M. Kinetics of RDX degradation by zero-valent iron (ZVI). J. Hazard. Mater. 2006, 136 (1), 68–74; Selected Proceedings of the Seventh Biennial Protection and Restoration of the Environment International Conference. (8) Singh, J.; Comfort, S. D.; Shea, P. J. Iron-mediated remediation of RDX-contaminated water and soil under controlled Eh/pH. Environ. Sci. Technol. 1999, 33 (9), 1488–1494. (9) Wildman, M. J.; Alvarez, P J. RDX degradation using an integrated Fe(0)-microbial treatment approach. Water Sci. Technol. 2001, 43 (2), 25–33. (10) Bunce, N. J.; Merica, S. G.; Lipkowski, J. Prospects for the use of electrochemical methods for the destruction of aromatic organochlorine wastes. Chemosphere 1997, 35, 2719–2726. (11) Rodgers, J. D.; Bunce, N. J. Electrochemical treatment of 2,4,6trinitrotoluene and related compounds. Environ. Sci. Technol. 2001, 35, 406–410. (12) Bonin, P. M.; Bejan, D.; Schutt, L.; Hawari, J.; Bunce, N. J. Electrochemical reduction of hexahydro-1,3,5-trinitro-1,3,5triazine in aqueous solutions. Environ. Sci. Technol. 2004, 38, 1595–1599.

(13) McCormick, N. G.; Cornell, J. H.; Kaplan, A. M. Biodegradation of hexahydro-1,3,5-trinitro-1,3,5-triazine. Appl. Environ. Microbiol. 1981, 42, 817–823. (14) Halasz, A.; Spain, J.; Paquet, L.; Beaulieu, C.; Hawari, J. Insights into the formation and degradation mechanisms of methylenedinitramine during the incubation of RDX with anaerobic sludge. Environ. Sci. Technol. 2002, 36, 633–638. (15) Waisner, S.; Hansen, L.; Fredrickson, H.; Nestler, C.; Zappi, M.; Banerji, S.; Baijpai, R. Biodegradation of RDX within soil-water slurries using a combination of differing redox incubation conditions. J. Hazard. Mater. 2002, 95 (1-2), 91–106. (16) Wani, A. H.; Felt, D. R.; Davis, J. L. RDX biodegradation column study: Extent of RDX mineralization and influence of temperature on rate of RDX biotransformation. Environ. Eng. Sci. 2005, 22 (3), 310–323. (17) Davis, J. L.; Wani, A. H.; O’Neal, B. R.; Hansen, L. D. RDX biodegradation column study: comparison of electron donors for biologically induced reductive transformation in groundwater. J. Hazard. Mater. 2004, 112 (1-2), 45–54. (18) Wani, A. H.; O’Neal, B. R.; Gilbert, D. M.; Gent, D. B.; Davis, J. L. Electrolytic transformation of ordnance related compounds (ORCs) in groundwater: Laboratory mass balance studies. Chemosphere 2006, 62 (5), 689–698. (19) Gilbert, D. M.; Sale, T. C. Sequential electrolytic oxidation and reduction of aqueous phase energetic compounds. Environ. Sci. Technol. 2005, 39 (23), 9270–9277. (20) Doppalapudi, R.; Palaniswamy, D.; Maloney, S. W.; Sorial, G. Electrochemical Reduction of Energetically Contaminated Wastewater: Development and Testing of Pilot Scale Reactor; Report ERDC/CERL TR-00-38; U.S. Army Engineer Research and Development Center, Construction Engineering Research laboratory, Champaign, IL, November 2000. (21) Hwang, S.; Felt, D. R.; Bouwer, E. J.; Brooks, M. C.; Larson, S. L.; Davis, J. L. Remediation of RDX-contaminated water using alkaline hydrolysis. J Environ. Eng. 2006, 132, 256. (22) Hoffsommer, J. C.; Kubose, D. A.; Glover, D. J. Kinetic isotope effects and intermediate formation for the aqueous alkaline homogenous hydrolysis of 1,3,5-triaza-1,3,5-tribitrocyclohexane (RDX). J. Phys. Chem. 1977, 81 (5), 380–385. (23) Beyer, H. Lehrbuch der Organischen Chemie, 18th ed.; Hirzel: Stuttgart, Germany, 1976. (24) Acar, Y. B.; Alshawabkeh, A. N. Principles of electrokinetic remediation. Environ. Sci. Technol. 1993, (27), 2638–2647. (25) Van Genuchten, M. Th. Alves, W. J. Analytical Solutions of the One-Dimensional Convective-Dispersive Solute Transport Equation; Technical Bulletin 1661; U.S. Department of Agriculture, Agricultural Research Service, 1982. (26) Alshawabkeh, A. N.; Sarahney, H. Effect of current density on enhanced transformation of naphthalene. Environ. Sci. Technol. 2005, 39, 5837–5843. (27) Sarahney, H.; Alshawabkeh, A. Effect of current density on electrolytic transformation of benzene for groundwater remediation. J. Hazard. Mater 2007, (143), 3, 649–654. (28) Hawari, J.; Halasz, A.; Sheremata, T.; Beaudet, S.; Groom, C.; Paquet, L.; Rhofir, C.; Ampleman, G.; Thiboutot, S. Characterization of metabolites during biodegradation of hexahydro1,3,5-trinitro-1,3,5-triazine (RDX) with municipal anaerobic sludge. Appl. Environ. Microbiol. 2000, 66, 2652–2657. (29) Gent, D. B. Electrolytic Alkaline Hydrolysis for in-situ Decomposition of Hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) in Groundwater. PhD thesis, Department of Civil and Environmental Engineering, Northeastern University, 2007.

ES803567S

VOL. 43, NO. 16, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6307