Environ. Sci. Technol. 2004, 38, 2715-2720
Nitrate Reduction by Zerovalent Iron: Effects of Formate, Oxalate, Citrate, Chloride, Sulfate, Borate, and Phosphate
nitrate concentrations greater than 10 mg of N L-1 in need of corrective action. It has been shown that zerovalent iron (Fe0) can serve as an electron donor to reduce nitrate (2-17). Two possible mechanisms are direct reduction of nitrate by Fe0 and its indirect reduction by hydrogen generated by iron corrosion (7, 8, 10). Nitrate reduction by Fe0 is a spontaneous process under acidic conditions:
CHUNMING SU* AND ROBERT W. PULS Ground Water and Ecosystems Restoration Division, National Risk Management Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency, 919 Kerr Research Drive, Ada, Oklahoma 74820
NO3- + 4Fe0 + 10H+ a NH4+ + 4Fe2+ + 3H2O
Recent studies have shown that zerovalent iron (Fe0) may potentially be used as a chemical medium in permeable reactive barriers (PRBs) for groundwater nitrate remediation; however, the effects of commonly found organic and inorganic ligands in soil and sediments on nitrate reduction by Fe0 have not been well understood. A 25.0 mL nitrate solution of 20.0 mg of N L-1 (1.43 mM nitrate) was reacted with 1.00 g of Peerless Fe0 at 200 rpm on a rotational shaker at 23 °C for up to 120 h in the presence of each of the organic acids (3.0 mM formic, 1.5 mM oxalic, and 1.0 mM citric acids) and inorganic acids (3.0 mM HCl, 1.5 mM H2SO4, 3.0 mM H3BO3, and 1.5 mM H3PO4). These acids provided an initial dissociable H+ concentration of 3.0 mM available for nitrate reduction reactions under conditions of final pH < 9.3. Nitrate reduction rates (pseudo-firstorder) increased in the order: H3PO4 < citric acid < H3BO3 < oxalic acid < H2SO4 < formic acid < HCl, ranging from 0.00278 to 0.0913 h-1, corresponding to surface area normalized rates ranging from 0.126 to 4.15 h-1 m-2 mL. Correlation analysis showed a negative linear relationship between the nitrate reduction rates for the ligands and the conditional stability constants for the soluble complexes of the ligands with Fe2+ (R2 ) 0.701) or Fe3+ (R2 ) 0.918) ions. This sequence of reactivity corresponds also to surface adsorption and complexation of the three organic ligands to iron oxides, which increase in the order formate < oxalate < citrate. The results are also consistent with the sequence of strength of surface complexation of the inorganic ligands to iron oxides, which increases in the order: chloride < sulfate < borate < phosphate. The blockage of reactive sites on the surface of Fe0 and its corrosion products by specific adsorption of the inner-sphere complex forming ligands (oxalate, citrate, sulfate, borate, and phosphate) may be responsible for the decreased nitrate reduction by Fe0 relative to the chloride system.
Introduction Nitrate contamination of groundwater is widespread due to agricultural land runoff, leaching of nitrogen fertilizers, concentrated animal feeding operations, food processing, and industrial waste effluent discharge (1). Numerous sites exist at government and private facilities with groundwater * Corresponding author phone: (580) 436-8638; fax: (580) 4368703; e-mail:
[email protected]. 10.1021/es034650p Not subject to U.S. Copyright. Publ. 2004 Am. Chem. Soc. Published on Web 03/20/2004
(1)
For the above reaction to proceed at a significant rate, the solution pH must be low (3, 7) or hydrogen must be supplied (4). Some researchers have reported insignificant nitrate reduction at pH values greater than ∼5 (3, 7), or in unbuffered, deionized water (5, 8), whereas others have demonstrated nitrate reduction in both buffered (4, 5, 8) and unbuffered (6, 10, 13) solutions at slightly alkaline pH values. As indicated by eq 1 and confirmed by many batch studies, ammonium ions in nearly stoichiometric amounts are the final end product (6-12), with a few studies reporting nitrite as an intermediate species (4, 6). Zerovalent iron based permeable reactive barrier (PRB) technology has been applied to remediate polluted groundwater with nitrate as a cocontaminant. For example, a Peerless Fe0 based PRB was constructed for groundwater nitrate and uranium remediation at the Oak Ridge Y-12 Plant site in November 1997 (18). Other contaminated sites may also have PRB listed as a potential choice for remediation. Successful implementation of a PRB requires a thorough understanding of the effects of geochemical variables such as chemical composition of the plume, pH, and redox potential on the behavior of the PRB (19). It is expected that major ions, organic and inorganic, will affect nitrate reduction kinetics through chemical interactions with Fe0 and its corrosion products (iron oxides and green rusts) in a subsurface PRB system, which may influence the effectiveness and longevity of PRB for contaminant removal. Simple aliphatic organic acids with one to three carboxylic groups such as formic, oxalic, and citric acids occur frequently in soil and subsurface environments from biological activities. They alter chemical processes in soils through complexation reactions with metal ions in solution and ligand exchange reactions at soil surfaces (20-22). Consequently, they may play an important role in nitrate reduction by Fe0 via adsorption and complexation with the surface sites of Fe0 and iron corrosion products. In addition, major anions such as sulfate, borate, and phosphate have been shown to decrease the removal kinetics of both arsenate and arsenite by Peerless Fe0 relative to chloride in both batch (23) and column tests (24). The interaction of phosphate with iron oxides is strong and well documented (22). Recent Fourier transform infrared (FTIR) spectroscopic evidence for sulfate adsorption reveals that one monodentate surface complex is dominant over a wide range of surface coverages and pH values in the case of adsorption on hematite (25, 26) and goethite (27). Another recent FTIR study shows that sulfate forms both outer-sphere and inner-sphere surface complexes on goethite at pH < 6; at pH > 6, sulfate adsorbs on goethite only as an outer-sphere complex (28). Borate is also known to form inner-sphere complexes with iron oxides (29). The influence of these organic and inorganic ligands on nitrate reduction by Fe0 has not been systematically studied and quantified. Consequently, our objective of this study was to evaluate the effects of representative organic and inorganic ligands on nitrate reduction by Fe0. Three organic acids (formic, oxalic, and citric) were compared with four inorganic VOL. 38, NO. 9, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2715
TABLE 1. Nitrate Reduction by Peerless Fe0 as Affected by Organic and Inorganic Ligandsa treatment
r2
initial pH
reaction pHb
k, h-1
kSA, h-1 m-2 mL
t1/2, h
tSA,1/2, h m-2 mL
3.0 mM formic acid 1.5 mM oxalic acid 1.0 mM citric acid 3.0 mM HCl 1.5 mM H2SO4 3.0 mM H3BO3 10 mM H3BO3 1.5 mM H3PO4 7.0 mM NaCl + 0.86 mM CaSO4
0.971 0.918 0.944 0.915 0.959 0.920 0.927 0.848 0.740
3.36 3.19 3.43 2.89 2.75 6.16 8.00 3.08 6.50
6.48 ( 1.08 7.48 ( 1.08 8.18 ( 0.97 6.02 ( 1.12 5.93 ( 0.46 8.15 ( 0.32 8.34 ( 0.21 7.69 ( 1.03 9.39 ( 0.20
0.0592 ( 0.0117 0.0256 ( 0.0033 0.00584 ( 0.0002 0.0913 ( 0.0028 0.0571 ( 0.0027 0.00904 ( 0.00148 0.00886 ( 0.00082 0.00278 ( 0.00031 0.00223 ( 0.00022
2.69 ( 0.53 1.17 ( 0.15 0.265 ( 0.010 4.15 ( 0.13 2.59 ( 0.12 0.411 ( 0.067 0.403 ( 0.037 0.126 ( 0.014 0.101 ( 0.010
12.0 ( 2.4 27.3 ( 3.5 119 ( 12 7.59 ( 0.23 12.2 ( 0.6 77.7 ( 6.3 78.5 ( 7.2 250 ( 28 313 ( 30
0.26 ( 0.05 0.60 ( 0.08 2.61 ( 0.10 0.17 ( 0.01 0.27 ( 0.01 1.71 ( 0.14 1.73 ( 0.16 5.51 ( 0.61 6.88 ( 0.67
a Pseudo-first-order equation coefficients of determination (r2), rate constants (k), surface area normalized rate constant (k ), calculated halfSA lives (T1/2), and surface area normalized half-lives (TSA,1/2), reported as mean ( sample standard deviation (n ) 2). b The average reaction pH was derived from 18 measured values at 3, 5, 9, 13, 25, 49, 73, 97, and 121 h.
acids (hydrochloric, sulfuric, boric, and phosphoric) at an equal initial dissociable H+ concentration of 3.0 mM available for nitrate reduction reactions under conditions of final pH < 9.3 in all systems. This allowed the solution pH to be kept at comparable levels for as many ligands as possible during the experiment without introducing a foreign and interfering ligand. We avoided using pH buffers because previous studies had shown that use of buffers such as HEPES at concentrations from 0.01 to 0.1 M in tests of nitrate reduction by Fe0 failed to control pH (8, 16) and also resulted in enhanced nitrate removal relative to the no-buffer treatment (8).
Materials and Methods To minimize material heterogeneity for the batch experiments, we used a coarser fraction (>0.5 mm) of Peerless Fe0 (Peerless Metal Powders & Abrasives, Detroit, MI) that was separated from the as-received material. The greater than 0.5 mm fraction had a BET N2 surface area of 0.55 ( 0.05 m2 g-1 determined with a Coulter SA3100 surface area analyzer. The pristine Peerless Fe0 is a low-grade steel that contains 90+% Fe, 2.50% C, 2.0% Si, 0.60% Mn, and 0.20% Cr (data from the manufacturer). All chemicals used were ACS reagent grade or analytical reagent grade as received. We chose an initial dissociable H+ concentration of 3.0 mM for testing nitrate reduction kinetics to avoid rigorous H2 gas generation, which occurs within a short time period when strong and high-concentration acids are reacted with Fe0. Consequently, the molar concentrations of the organic acids were 3.0 mM formic (HCOOH, pKa ) 3.75), 1.5 mM oxalic (HOOC-COOH, pKa1 ) 1.25, pKa2 ) 4.27), and 1.0 mM citric [(CH2)2COH(COOH)3, pKa1 ) 3.13, pKa2 ) 4.76, and pKa3 ) 6.40). The dissociation constants for H3PO4 are pKa1 ) 2.15, pKa2 ) 7.25, and pKa3 ) 12.38 [all the above pKa values were taken from the NIST Standard Reference Database 46 (30)]; therefore, only two hydrogen ions per H3PO4 molecule will dissociate in the Peerless Fe0 system when the final pH is near 9. The only measurable pKa for H3BO3 is 9.24; therefore, less than one hydrogen ion will dissociate from each added H3BO3 molecule in the Fe0 system when the final pH is near 9, whereas added HCl and H2SO4 (pKa1 ) -3, pKa2 ) 1.99) will completely dissociate in contact with the Peerless Fe0. Consequently, the molar concentrations of the inorganic acids were 3.0 mM HCl (one total hydrogen atom in its molecular formula), 1.5 mM H2SO4 (two total hydrogen atoms), 3.0 mM H3BO3 (three total hydrogen atoms), and 1.5 mM H3PO4 (three total hydrogen atoms). The acids contained 20.0 mg of N L-1 as NaNO3 (1.43 mM nitrate), and the initial pH values (before Peerless Fe0 contact) were 3.36 for the formic, 3.19 for the oxalic, 3.43 for the citric, 2.89 for the hydrochloric, 2.75 for the sulfuric, 6.16 for the boric, and 3.08 for the phosphoric acid solutions (Table 1). In a separate batch test, we monitored nitrate reduction in a background solution of 7.0 mM NaCl + 0.86 mM CaSO4 2716
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 9, 2004
(pH 6.5, adjusted with 0.1 M NaOH). This solution represents approximately the major ionic composition of the groundwater at a PRB site in Elizabeth City, NC, where a hexavalent chromium plume was intercepted and remediated through chemical reduction of Cr(VI) to form a mixed ferric and trivalent chromium hydroxide by a wall of Peerless Fe0 (31, 32). This batch test was also compared to that of a 10 mM H3BO3 solution (initial pH 8.0, adjusted with 0.1 M NaOH) due to its stronger pH buffering capacity. In general, kinetic studies (at reaction times of 3, 5, 9, 13, 25, 49, 73, 97, and 121 h, 1 h of which was the centrifugation time) were performed for the Peerless Fe0-ligand systems. Duplicate samples of 1.00 g of Peerless Fe0 were added to each 50 mL centrifuge tube to which 25.0 mL of a nitratecontaining acid solution was added subsequently. The tubes were shaken on a rotational shaker at 200 rpm at 23 ( 1 °C for the predetermined time period before being centrifuged for 1 h at 3600 rpm (2600g). An aliquot of 10 mL was filtered through a 0.22 µm filter membrane for determination of NO3-, NO2-, and NH4+. The dissolved NO3- and NO2- were measured by the hydrazine reduction method (USEPA, method 353.1), and NH4+ was determined by the automated phenate colorimetric method (USEPA, method 350.1). Selected solutions were also analyzed for total dissolved iron with inductively coupled plasma optical emission spectroscopy (ICP-OES). The remaining solution was then analyzed for the final pH with an Orion combination pH electrode and for redox potential (Eh) with an Orion platinum redox electrode. The measured Eh values were corrected to a standard hydrogen electrode (SHE) basis, i.e., true Eh values following procedures in the instrument manual. A pseudo-first-order kinetic model that was able to successfully describe dechlorination kinetics of trichloroethene with Peerless Fe0 (33) was also used to describe the nitrate disappearance over time in various Peerless Fe0ligand systems. For each experimental run, unless otherwise noted, ten ln([NO3-]t/[NO3-]0) points were plotted as a function of time, where [NO3-]0 is the initial nitrate concentration and [NO3-]t is the nitrate concentration at time t. The kinetic model is acceptable if the coefficient of determination (r2) is g0.585 (df ) 8, P ) 0.01). The surface area normalized kinetic parameters such as rate constants and half-lives were calculated on the basis of a 1.0 m2 surface area of Peerless Fe0 solids per 1.0 mL of aqueous solution.
Results and Discussion Averaged results from duplicate tests were reported. In general, relative sample standard deviations for duplicate tests were less than 15%. Effects of Organic Ligands. Nitrate concentration decreased exponentially with time with a concomitant increase in NH4+ concentration in the presence of formic acid (Figure
FIGURE 1. Nitrate reduction by Peerless Fe0 in 3.0 mM formic acid: (a) dissolved NO3-, NO2-, and NH4+, (b) Eh and pH. The control treatment was a nitrate solution of 20 mg of N L-1 + 3.0 mM formic acid without Fe0. Mean values of duplicate tests are shown.
FIGURE 2. Nitrate reduction by Peerless Fe0 in 1.5 mM oxalic acid: (a) dissolved NO3-, NO2-, and NH4+, (b) Eh and pH. 1), oxalic acid (Figure 2), and citric acid (Figure 3). Nitrogen mass balance, as defined as the percentage of the sum of measured NO3-, NO2-, and NH4+ over added NO3-, was generally in the range of 90-110%. From trace to small amounts of NO2- (up to 1.1 mg of N L-1) were also detected. This is consistent with some studies that show detectable
FIGURE 3. Nitrate reduction by Peerless Fe0 in 1.0 mM citric acid: (a) dissolved NO3-, NO2-, and NH4+, (b) Eh and pH. nitrite as an intermediate product of nitrate reduction by Fe0 (4, 6). The organic acids in the absence of Peerless Fe0 (the control treatments) did not change nitrate concentrations during the experiment (Figures 1-3). Abiotic nitrate reduction usually requires strong reductants and catalysts. Nitrate does not react with dissolved ferrous ions without catalysts of other transition-metal ions such as Cu2+ (34) and Ag+ (35), implying that abiotic reduction of nitrate occurred only at the surfaces of the iron metal in our study. The dissolved iron was generally below the detection limit of ICP-OES (0.035 mg L-1) largely due to iron precipitation after oxidation of ferrous ions during the filtration of supernatant solution in the open air. In all systems, pH increased with increasing time to reach steady levels (Figures 1-3), whereas Eh generally decreased with increasing time to reach steady state with positive values for the formic and oxalic acid treatments (Figures 1 and 2) and with negative values for the citric acid treatment (Figure 3) after 9 h of reaction. Nitrate reduction by Peerless Fe0 was enhanced in the presence of simple aliphatic acids compared to Peerless Fe0 in simulated Elizabeth City groundwater (initial pH 6.50) (Table 1). This is largely caused by the higher solution pH in the simulated Elizabeth City groundwater system during experiment (pH 9.39 ( 0.20, n ) 18) due to the absence of added acid and the low pH buffering capacity of the simulated groundwater. Despite the fact that the three organic acids used had similar initial pH values (3.2-3.4), from 3 to 121 h after initiation of the experiment, the order of pH values in the organic acid system became: formic < oxalic < citric (Figures 1-3), consistent with the order of calculated halflives and surface area normalized half-lives of nitrate reduction (Table 1). The final pH values at 121 h reached 7.88 for formic acid, 8.66 for oxalic acid, and 9.19 for citric acid treatments. The increase in pH was likely caused by a number of chemical processes. First, iron corrosion releases hydroxyl ions.
anaerobic corrosion: Fe0 + 2H2O a Fe2+ + H2 + 2OH- (2) VOL. 38, NO. 9, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2717
aerobic corrosion: 2Fe0 + O2 + 2H2O a 2Fe2+ + 4OH- (3) Second, subsequent precipitation of ferrous iron produces hydrogen ion.
hydrolysis: 4Fe2+ + O2 + 10H2O a 4Fe(OH)3(s) + 8 H+ (4) green rust formation: 3Fe2+ + Fe3+ + Cl- + 8H2O a Fe4(OH)8Cl
(s, chloride green rust) + 8H+ (5)
4Fe2+ + 2Fe3+ + SO42- + 12H2O a Fe6
(OH)12SO4(s, sulfate green rust) + 12H+ (6)
Under the highly reducing conditions found in groundwater remediation applications, the oxide formed on Fe0 is probably best modeled as a mixed-valent or pure Fe(II) phase such as magnetite, siderite, or green rust (36). Finally, adsorption of organic ligands to the iron corrosion products through ligand exchange reactions releases OH- into solution. Apparently, the net result of these different chemical processes was an increase in solution pH. The nitrate reduction rate in the 3.0 mM formic acid/Fe0 system is 10 times greater than that in the 1.0 mM citric acid/Fe0 system with overall reduction rates decreasing in the order: formic acid > oxalic acid > citric acid (Table 1). This sequence of reactivity may be explained by the relative strength of aqueous soluble complexes between Fe2+/Fe3+ and these organic ligands (Table 2). There is no reported stability constant value for the Fe2+-formate complex, and the other stability constants are available for only a few ionic strength levels (thermodynamic equilibrium constants are lacking). Yet, the available stability constants listed in Table 2 clearly show that the strengths of the aqueous soluble complexes follow the order: formate < oxalate < citrate. Stumm and Morgan (22) suggested that the same chemical mode of interaction occurs in solution and at the surface and that the linear free energy relation between the tendency to form solute complexes of metal ions and the tendency to form surface complexes on metal oxides may be used to predict (intrinsic) sorption constants from solute complex formation constants and vice versa. The sequence of nitrate reduction rates (Table 1) is opposite the sequence for the strength of surface adsorption and complexation of these organic ligands with the iron oxide (22), which are corrosion products of Fe0. It is apparent that the strength of surface adsorption and complexation increases as the number of carboxylic groups in each ligand molecule increases. Nitrate reduction thus can be retarded by the binding of strong organic ligands such as oxalate and citrate at the iron metal surfaces so that the availability of the active sites for nitrate reduction is decreased. Effects of Inorganic Ligands. In the case of HCl (Figure 4) and H2SO4 (Figure 5), after 49 h of reaction, nitrate concentrations were below the detection limit ( H2SO4 > H3BO3 > H3PO4 (Figures 4-7 and Table 1). The critical stability constants for soluble aqueous complexes between Fe2+/Fe3+ and inorganic ligands increase in the order: chloride < sulfate < phosphate (Table 2). It is well established that chloride does not sorb specifically to iron oxides in aqueous solution, whereas borate (29) and phosphate (22) form inner-sphere complexes with iron oxides. Borate also decreases dechlorination kinetics of chlorinated solvents for zerovalent iron (37). Recent studies show sulfate also forms inner-sphere complexes on iron oxides (25-27). This study shows that the sequence for the nitrate reduction rate for the inorganic ligands is also opposite the sequence for the strength or affinity of the surface complexation of these inorganic ligands with iron oxides. A higher concentration of boric acid (10 mM, initial pH 8.0) gave only a slight decrease in the nitrate reduction rate as compared to a 3.0 mM (initial pH 6.5) solution (Table 1). It is possible that sorption of borate (and perhaps other ligands as well) to zerovalent iron and iron corrosion products reached a level close to saturation of available sorption sites for the initial concentrations of ligands
FIGURE 5. Nitrate reduction by Peerless Fe0 in 1.5 mM H2SO4: (a) dissolved NO3-, NO2-, and NH4+, (b) Eh and pH.
FIGURE 6. Nitrate reduction by Peerless Fe0 in 3.0 mM H3BO3: (a) dissolved NO3-, NO2-, and NH4+, (b) Eh and pH. used in this study. This study supports the findings of earlier work (4-6, 8, 10, 13, 15) that nitrate reduction also occurs at pH values greater than 5. An attempt to explore the linear free energy relationships between nitrate reduction rates for the Peerless iron and the stability constants of aqueous Fe2+- and Fe3+-ligand complexes did not materialize due to the absence of thermodynamic data for most complexes of interest; however,
FIGURE 7. Nitrate reduction by Peerless Fe0 in 1.5 mM H3PO4: (a) dissolved NO3-, NO2-, and NH4+, (b) Eh and pH. limited information is available for the conditional (ionic strength dependent) critical stability constants taken from the NIST database (30) (Table 2). A linear relationship was evident between the surface area normalized rate constants for nitrate reduction by Peerless iron for different organic and inorganic ligands and the conditional critical stability constants for the complexes of the ligands with Fe2+ (R2 ) 0.701, df ) 3, P < 0.1) or Fe3+ (R2 ) 0.918, df ) 4, P < 0.01) ions (Figure 8). Both equations shown in Figure 8 exhibit similar intercept values at 3.65 and 3.82 h-1 m-2 mL for a zero metal-ligand stability constant. The linear relationships are statistically significant and may be improved if thermodynamic equilibrium constants (at zero ionic strength) become available. Implications for Nitrate Reduction at PRB Field Sites. Our laboratory results show that the overall nitrate reduction rate follows the order: HCl > formic > H2SO4 > oxalic > H3BO3 > citric > H3PO4. The presence of the two strongest ligands, phosphate and citrate, will definitely retard nitrate reduction by Peerless Fe0. The ligand effects should thus be taken into consideration when a PRB is designed for field applications. Consequently, a greater mass of Fe0 may be needed for sites where phosphate and organic acids are significant cocontaminants with nitrate in groundwater. Nitrate reduction by Fe0 (2-17) and by iron corrosion products such as sulfate green rust (38) produces ammonium ions under laboratory conditions. The reduction process is abiotic in the absence of soils and sediments. This may be a drawback for field application at some sites where the microbial denitrification process is limited. Several methods may be used to address this problem: selective photocatalytic oxidation of NH3 to N2 gas using platinized TiO2 in water (39), raising the pH and releasing NH3 from the remediated solution, coupling the iron reduction process with a biological process to convert NH3 to nitrogen gas (e.g, organic compost materials with active microbes). On the other hand, reaction pathways of a PRB with nitrate in a field setting are more likely to be complicated by microbial interactions, which could be beneficial for nitrate removal. The presence of Fe0 VOL. 38, NO. 9, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2719
anonymous reviewers helped improve an earlier version of this paper.
Literature Cited
FIGURE 8. Surface area normalized rate constant for nitrate reduction by Peerless Fe0 in the presence of different ligands as a function of (a) the Fe2+-ligand stability constant (log K, 25 °C) and (b) the Fe3+-ligand stability constant (log K, 25 °C). Stability constants are listed in Table 2. has been shown to enhance autotrophic denitrification (9), and to stimulate indigenous microbes including denitrifiers within and in the vicinity of a Peerless Fe0 PRB (18). Microbial population including sulfate-reducing and denitrifying bacteria within and in the vicinity of the Peerless Fe0 PRB at the Oak Ridge Y-12 Plant site was increased from 1 to 3 orders of magnitude relative to those found in the background soil/ groundwater samples (18). Under anaerobic conditions, the presence of both nitrate and ammonium may facilitate establishment of anaerobic ammonium-oxidizing (anammox) bacteria that directly oxidize ammonium to dinitrogen gas with nitrate (40) and nitrite (41) as the electron acceptor:
5NH4+ + 3NO3- f 4N2 + 9H2O + 2H+
(7)
NH4+ + NO2- f N2 + 2H2O
(8)
It remains to be seen if ammonium accumulation is an issue of concern for other PRB sites where nitrate is a major remediation target.
Acknowledgments Although the research described in this paper has been funded wholly by the U.S. Environmental Protection Agency, it has not been subjected to Agency review and, therefore, does not necessarily reflect the views of the Agency, and no official endorsement should be inferred. We acknowledge gratefully the analytical assistance of Mr. Jarrod A. Tollett, Ms. Sharon Hightower, and Ms. Lynda Pennington of ManTech Environmental Research Services Corp. Constructive comments from the Editor (Dr. J. L. Schnoor) and 2720
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 9, 2004
(1) Follett, R. F. In Nitrogen in the Environment: Sources, Problems, and Management; Follett, R. F., Hatfield, J. L., Eds.; Elsevier Science: Amsterdam, 2001; Chapter 2, pp 17-44. (2) Szabo, Z. G.; Bartha, L. G. Anal. Chem. Acta 1952, 6, 416-419. (3) Young, G. K.; Bungay, H. R.; Brown, L. M.; Parsons, W. A. J. Water Pollut. Control Fed. 1964, 36, 395-398. (4) Siantar, D. P.; Schreier, C. G.; Chou, C. S.; Reinhard, M. Water Res. 1996, 30, 2315-2322. (5) Cheng, I. F.; Muftikian, R.; Fernando, Q.; Korte, N. Chemosphere 1997, 35, 2689-2695. (6) Rahman, A.; Agrawal, A. Prepr. Ext. Abstr. 213th ACS Natl. Meet., Am. Chem. Soc., Div. Environ. Chem. 1997, 37 (1), 157-159. (7) Huang, C. P.; Wang, H. W.; Chiu, P. C. Water Res. 1998, 32, 2257-2264. (8) Zawaideh, L. L.; Zhang, T. C. Water Sci. Technol. 1998, 38, 107-115. (9) Till, B. A.; Weathers, L. J.; Alvarez, P. J. J. Environ. Sci. Technol. 1998, 32, 634-639. (10) Chew, C. F.; Zhang, T. C. Environ. Eng. Sci. 1999, 16, 389-401. (11) Devlin, J. F.; Eedy, R.; Butler, B. J. J. Contam. Hydrol. 2000, 46, 81-97. (12) Kielemoes, J.; De Boever, P.; Verstraete, W. Environ. Sci. Technol. 2000, 34, 663-671. (13) Choe, S.; Chang, Y. Y.; Hwang, K. Y.; Khim, J. Chemsphere 2000, 41, 1307-1311. (14) Schlicker, O.; Ebert, M.; Fruth, M.; Weidner, M.; Wust, W.; Dahmke, A. Ground Water 2000, 38, 403-409. (15) Alowitz, M. J.; Scherer, M. M. Environ. Sci. Technol. 2002, 36, 299-306. (16) Westerhoff, P. J. Environ. Eng. 2003, 129, 10-16. (17) Westerhoff, P.; James, J. Water Res. 2003, 37, 1818-1830. (18) Gu, B.; Watson, D. B.; Wu, L.; Philips, D. H.; White, D. C.; Zhou, J. Environ. Monit. Assess. 2002, 77, 293-309. (19) Liang, L.; Korte, N.; Gu, B.; Puls, R. W.; Reeter, C. Adv. Environ. Res. 2000, 4, 273-286. (20) Jones, D. L. Plant Soil 1998, 205, 25-44. (21) Strobel, B. W. Geoderma 2001, 99, 169-198. (22) Stumm, W.; Morgan, J. J. Aquatic Chemistry, Chemical Equilibria and Rates In Natural Waters, 3rd ed.; John Wiley & Sons: New York. 1996. (23) Su, C.; Puls, R. W. Environ. Sci. Technol. 2001, 35, 4562-4568. (24) Su, C.; Puls, R. W. Environ. Sci. Technol. 2003, 37, 2582-2587. (25) Hug, S. J. J. Colloid Interface Sci. 1997, 188, 415-422. (26) Eggleston, C. M.; Hug, S. J.; Stumm, W.; Sulzberger, B.; Afonso, M. D. S. Geochim. Cosmochim. Acta 1998, 62, 585-593. (27) Sugimoto, T.; Wang, Y. J. Colloid Interface Sci. 1998, 207, 137149. (28) Peak, D.; Ford, R. G.; Sparks, D. L. J. Colloid Interface Sci. 1999, 218, 289-299. (29) Su, C.; Suarez, D. L. Environ. Sci. Technol. 1995, 29, 302-311. (30) Martel, A. E.; Smith, R. M.; Motekaitis, R. J. NIST Critically Selected Stability Constants of Metal Complexes Database; NIST Standard Reference Database 46, Version 7; NIST: Gaithersburg, MD, 2003. (31) Puls, R. W.; Blowes, D. W.; Gillham, R. W. J. Hazard. Mater. 1999, 68, 109-124. (32) Wilkin, R. T.; Puls, R. W.; Sewell, G. W. Ground Water 2003, 41, 493-503. (33) Su, C.; Puls, R. W. Environ. Sci. Technol. 1999, 33, 163-168. (34) Buresh, R. J.; Moraghan, J. T. J. Environ. Qual. 1976, 5, 320-325. (35) Brown, L. L.; Drury, J. S. J. Chem. Phys. 1967, 46, 2833-2837. (36) Scherer, M. M.; Balko, B. A.; Tratnyek, P. G. ACS Symp. Ser. 1998, No. 715, 301-322. (37) Johnson, T. L.; Fish, W.; Gorby, Y. A.; Tratnyek, P. G. J. Contam. Hydrol. 1998, 29, 377-396. (38) Hansen, H. C. B.; Koch, C. B.; Nancke-Krogh, H.; Borggaard, O. K.; Sorensen, J. Environ. Sci. Technol. 1996, 30, 2053-2056. (39) Lee, J.; Park, H.; Choi, W. Environ. Sci. Technol. 2002, 36, 54625468. (40) Mulder, A.; Vandegraaf, A. A.; Robertson, L. A.; Kuenen, J. G. FEMS Microbiol. Ecol. 1995, 16, 177-183. (41) Dalsgaard, T.; Canfield, D. E.; Petersen, J.; Thamdrup, B.; AcunaGonzalez, J. Nature 2003, 422, 606-608.
Received for review June 24, 2003. Revised manuscript received January 30, 2004. Accepted February 9, 2004. ES034650P
