JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hess, T. F.
Right arrow Articles by Schrader, P. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hess, T. F.
Right arrow Articles by Schrader, P. S.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Hess, T. F.
Right arrow Articles by Schrader, P. S.
Related Collections
Right arrow Bioremediation and Biodegradation
Right arrow Organic Compounds
Right arrow Water Pollution
Journal of Environmental Quality 31:736-744 (2002)
© 2002 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America

TECHNICAL REPORTS
Bioremediation and Biodegradation

Coupled Abiotic–Biotic Mineralization of 2,4,6-Trinitrotoluene (TNT)

Thomas F. Hess* and Paul S. Schrader

Center for Hazardous Waste Remediation and Research, Univ. of Idaho, Moscow, ID 83844-0904

* Corresponding author (tfhess{at}uidaho.edu)

Received for publication May 14, 2001.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Munitions wastes such as TNT are widespread contaminants in soils and ground waters. We investigated a coupled abiotic–biotic treatment scheme for remediation of aqueous solutions of TNT. Mineralization of aqueous TNT (0.22 mM) was initially optimized with minimum reactant use (Fe3+ and H2O2) in light-assisted and dark, modified Fenton reactions at acidic and neutral pH. Complete TNT degradation occurred under all reaction conditions within 24 h. Using the optimum reactant concentrations, coupled abiotic–biotic reactions showed an increase in TNT mineralization, from 47 to 80%, after biomass addition to the acidic, dark Fenton-like reaction. Comparable increases of TNT mineralization were observed under neutral pH with similar reaction conditions. In light-assisted Fenton-like reactions at neutral pH, no increase in cumulative TNT mineralization (66%) was seen in coupled abiotic–biotic reactions. Abiotic photo-Fenton-like reactions alone, at acidic pH, produced complete TNT mineralization and required no biotic assistance. While light-enhanced Fenton reactions alone can provide high levels of TNT mineralization, the dark abiotic–biotic reaction scheme has perhaps a wider use due to a similar extent of TNT mineralization in the absence of light, leading to possible applications in soil slurry and in situ processes in the subsurface.

Abbreviations: DNP, 2,4-dinitrophenol • NTA, nitrilotriaceticacid • TNT, 2,4,6-trinitrotoluene • WAS, waste-activated sludge


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
WASTE GENERATED DURING the past manufacture of 2,4,6-trinitrotoluene (TNT) and decommissioning of military ordnance has contributed to widespread environmental contamination at many current and former Department of Defense (DOD) facilities located throughout the United States (Urbanski, 1964; Spalding and Fulton, 1988). More than 1200 explosives-contaminated DOD sites have been identified with almost 90% of these sites containing TNT-contaminated ground water (Schmelling and Gray, 1995). Because TNT is acutely toxic to humans and animals even in low concentrations (Watts, 1998), mutagenic in the Ames test (Won et al., 1976), and listed as a priority pollutant by the USEPA (Schuster and Gratzfeld-Huesgen, 1993), remediation of TNT-contaminated soils and ground waters is legally mandated.

The numerous technologies investigated for remediation of waters contaminated with TNT and related compounds principally have been based on chemical or biological processes. Studies of chemical treatment have focused on advanced oxidative processes (AOPs), while those related to biological transformation have used either bacterial or fungal systems under aerobic or anaerobic conditions. The AOPs previously applied to the treatment of TNT include ozone-catalyzed decomposition of TNT (Lang et al., 1998), TiO2–mediated photocatalysis (Schmelling et al., 1996; Schmelling and Gray, 1993, 1995), and Fenton chemistry (Li et al., 1997a,b).

Fenton's reaction is the catalyzed decomposition of hydrogen peroxide by transition metals, resulting in the generation of hydroxyl radicals (Haber and Weiss, 1934). The standard Fenton reaction proceeds by adding dilute hydrogen peroxide to a degassed solution of iron(II), resulting in nearly stoichiometric generation of hydroxyl radicals. Most environmental applications of Fenton chemistry involve reaction modifications, including use of higher concentrations of hydrogen peroxide, phosphate-buffered medium, heterogeneous catalysts, or iron. These conditions, although not as stoichiometrically efficient as the standard Fenton's reaction, are often necessary to treat sorbed contaminants in soils and ground water (Tyre et al., 1991).

Hydroxyl radical has high reactivity with many environmental contaminants at or near diffusion-controlled rates (>109 M-1 s-1). The degradation of xenobiotic chemicals by hydroxyl radical is typically due to either hydroxylation or hydrogen atom abstraction. Some biorefractory compounds such as perchloroethylene, hexachlorocyclopentadiene, and hexachlorobenzene have been effectively destroyed by hydroxyl radicals within minutes (Leung et al., 1992; Sato et al., 1993; Watts et al., 1994). The use of Fenton's reagent (29.4 mM H2O2 and 40 mM Fe2+) for the destruction of TNT in aqueous solutions (0.31 mM) resulted in complete degradation within 8 h and 40% mineralization within 24 h (Li et al., 1997a).

Anaerobic and aerobic biodegradation processes have been investigated for the destruction of TNT. While aerobic TNT biodegradation has been demonstrated, problems such as accumulation of metabolic intermediates (Vorbeck et al., 1994, Ramos et al., 1995), inhibitory intermediate compound formation (Michels and Gottschalk, 1995), and low mineralization (Fernando et al., 1990) limit process effectiveness. Additionally, reductive microbial transformation of TNT is the thermodynamically favorable pathway due to the aromatic ring stabilization and electron-withdrawing effect of the nitro groups (Bruhn et al., 1987). The consensus of recent research is that anaerobic biological processes hold the most promise for stand-alone bioremediation of TNT (Funk et al., 1993; Preuss and Rieger, 1995, Crawford, 1995; Regan and Crawford, 1994; Lewis et al., 1997). The research has indicated, however, that substantial mineralization of TNT may not be achieved although the parent compound can be entirely transformed to intermediary metabolites (Crawford, 1995).

Combined technologies for the destruction of hazardous wastes have received research attention (Carberry and Benzing, 1991; Koyama et al., 1994; Scott and Ollis, 1995; Ravikumar and Gurol, 1991). In these studies, the authors investigated sequential processes, using abiotic reactions as a pretreatment step for a separate, follow-on biological reaction. Such technologies were developed to overcome the biorecalcitrance of a particular compound inherent with stand-alone biological processes. Our own research into sequential, coupled processes has indicated that TiO2–mediated photocatalysis followed by biological degradation (Hess et al., 1998) and coexistent abiotic and biotic transformations (Buyuksonmez et al., 1998, 1999; Howsakeng et al., 2002) can be used to treat biorefractory compounds.

This research was designed to test the efficacy of coupled abiotic–biotic reactions for a high extent of TNT destruction in simple, aqueous matrices, thus avoiding complications of soil solutions, yet provide a basis for future work in soil. A modified Fenton's system (Fe3+ catalyst) was used in the abiotic reaction and two different, uncharacterized aerobic biomasses were tested for their potential to degrade any Fenton-degradation products. The specific objectives of this research were fourfold: (i) explore the efficacy of the chemical treatment under a variety of environmental conditions, (ii) optimize the chemical mineralization of TNT under these conditions, (iii) explore the use of an aerobic microbial biomass as a subsequent treatment, and (iv) determine the kinetics of each treatment.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Chemical Reagents
Iron(III) sulfate pentahydrate (97%) and nitrilotriaceticacid (NTA, 99%) were purchased from Aldrich Chemical Company (Milwaukee, WI). Reagent-grade H2O2 (30% v/v) was obtained from J.T. Baker (Phillipsburg, NJ). The 2,4,6-trinitrotoluene (99%) was purchased from Chem Service (West Chester, PA). Ecolite (+) scintillation cocktail was purchased from ICN Biomedicals (Costa Mesa, CA). Uniformly ring-labeled 2,4,6-trinitrotoluene with a specific activity of 2.18 MBq mM-1 (>99%) was synthesized by Dr. Stefan Goszczynski of the Environmental Biotechnology Institute, University of Idaho (Moscow, ID). All other chemicals used in the project were of the highest available purity. Double-deionized water (>18 M{Omega}-cm) was used for preparation of all chemical solutions.

Experimental Design
Several sets of experiments were conducted for this research: abiotic, optimization experiments to determine maximum TNT mineralization occurring at minimum reactant concentrations; combined abiotic–biotic kinetic experiments (conducted at optimum reactant concentrations determined above) to determine any increase in TNT mineralization due to biotic reactions; abiotic degradation experiments conducted to determine extent of TNT degradation using optimum reactant concentrations; and abiotic experiments to determine the effect of metal chelate concentrations on overall TNT mineralization (Fig. 1) . Modified Fenton reactions were used for all abiotic TNT mineralization and degradation studies.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 1. Diagram of experimental setup showing incremental use of results (arrows) between various stages of experimentation.

 
Optimization experiments were originally conceived as two-level, rotatable, central composite designs (Table 1) but later converted to factorial designs when early data (dark-Fenton reactions, pH 3) did not show rotatability (Table 2) (Cochran and Cox, 1992). The two-level designs (Tables 1 and 2) included Fe3+ and H2O2 concentrations as experimental variables with TNT mineralization as the response. The factorial experimental design was used to quantify the effects of the individual experimental variables, iron and hydrogen peroxide, on the response. This was done by analyzing the response when testing one experimental variable over a range of concentrations while leaving the other constant. The same was then done for the other variable. Optimality was determined graphically based on the maximum response achieved using the lowest reactant concentrations.


View this table:
[in this window]
[in a new window]
  		Table 1. Experimental designs and results of 14C-labeled experiments for investigation of the effects of iron and hydrogen peroxide on the mineralization of aqueous TNT (0.22 mM) in dark-Fenton reactions at pH 3. Optimum iron and hydrogen peroxide concentrations that produced maximum TNT mineralization are in italic type.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Experimental design and results of 14C-labeled experiments for investigation of the effects of iron and hydrogen peroxide on the mineralization of aqueous TNT (0.22 mM) in dark-Fenton reactions at pH 7 and light-Fenton reactions at pH 3 and pH 7. Experiments at pH 7 used NTA to Fe molar ratios of 10:1 to chelate iron. Optimum iron hydrogen peroxide concentrations that produced maximum TNT mineralization are in italic type.

 
Abiotic TNT Mineralization Experiments
Abiotic TNT mineralization was measured in both optimization and kinetic experiments by capture and quantification of 14CO2 produced during the modified Fenton reactions. All experiments were conducted with 30-mL solutions of TNT (0.22 mM) using mixtures of nonlabeled TNT and enough 14C uniformly labelled TNT to provide approximately 1.67 x 103 Bq per flask. Mineralization reactions were conducted in 500-mL biometer flasks each sealed with a rubber stopper and containing a glass cup (holding base solution) suspended in the atmosphere of the flask and a piece of glass tubing extending from the atmosphere of the flask through the stopper. Attached to the outside end of the glass tubing was an expandable bladder used to hold the gasses evolved from the reaction and allow free exchange with the flask atmosphere. The glass cup contained 1 mL of a 0.1 M NaOH solution used to capture CO2 from the flask atmosphere. For optimization experiments, the 1-mL NaOH sample was collected, as well as two successive 1-mL H2O-rinsates, at the end of the experiment. For kinetic experiments, the sample and rinsates were collected at timed intervals during the course of the experiment. The samples were added directly to 15 mL of Ecolite (+) liquid scintillation cocktail and analyzed by scintillation counting as described below to quantify the amount of 14CO2 generated from the reaction. Knowing the original quantity of 14C-TNT in solution, we determined the extent (percent) of TNT mineralization. A 1-mL sample of the aqueous portion of the reaction was collected at the end of the experiment and analyzed similarly to determine a mass balance on 14C.

Procedurally, the mineralization experiments were set up by initially adding Fe3+ to the TNT solution and the Fenton-like reaction was then begun by adding H2O2 to the TNT–Fe3+ solution. Hydrogen peroxide and Fe3+ concentrations varied between 15 and 294 mM (1%) and 0.05 and 20 mM, respectively, for optimization experiments (Tables 1 and 2) and were constant for kinetic experiments, based on results found during optimization (values listed in italic type in Tables 1 and 2). The iron used in neutral pH reactions was chelated with nitrilotriaceticacid (NTA) at either equimolar concentrations (1:1) or ten times more NTA than iron (10:1), depending on the experiment. The pH of the NTA–Fe3+ solution was adjusted to 7.0 using NaOH. All experiments were monitored for pH using a meter and probe, and calibrated prior to each use with standard buffer solutions (Accument Basic; Fisher Scientific, Pittsburgh, PA). Experiments performed in the dark (dark-Fenton) were conducted after covering the reaction vessels in aluminum foil. Those experiments run in the light (light-Fenton) were put under a light box containing six, 24-inch full-spectrum (380–750 nm, peak intensity at 610 nm) 20-watt light bulbs, 16-inches above the flasks, giving a light intensity of 54.2 cd m-2.

Biotic Reactions
Kinetic experiments that received biotic treatment subsequent to the Fenton reaction were first brought to neutral pH by the addition of 1 mL of M9 salts (Provence and Curtiss, 1994). The addition of the salts resulted in a final concentration of Na2HPO4 (42.3 mM), KH2PO4 (22 mM), NaCl (8.5 mM), and NH4Cl (18.7 mM). Two uncharacterized biomasses from aerobic, bench-scale sequencing batch reactors (SBRs), described previously (Hess et al., 1993), were used in separate experiments. The first SBR was seeded with waste-activated sludge (WAS) from the Pullman, Washington Wastewater Treatment Facility, fed daily a synthetic waste (Kennedy et al., 1990) with an organic carbon content of approximately 130 mg L-1, and maintained at an average total suspended solids (TSS) concentration of 2800 mg L-1. The WAS was added to the biometer flasks at 24 h after the initiation of the abiotic, modified Fenton reaction (after solution neutralization) at an average concentration of either 467 or 93 mg L-1, depending on the experiment, and allowed to react for an additional 6 d with base samples, containing 14CO2, taken at regular intervals. The second SBR was seeded with a consortium of 2,4-dinitrophenol (DNP)–degrading bacteria, fed daily with a DNP–glucose waste (Hess et al., 1990) with an organic carbon content of approximately 45 mg L-1, and maintained at an average TSS concentration of 2000 mg L-1. This DNP biomass was added to biometer flasks similar to the WAS biomass at concentrations of 333 and 67 mg L-1, depending on the experiment.

Liquid Scintillation Analysis
All samples from mineralization studies were counted with a liquid scintillation analyzer (Tri-Carb Model 2100TR; Packard Bioscience, Meriden, CT) using a 14C protocol. Counts per minute were converted to disintegrations per minute by using an efficiency plot for known 14C quench standards and appropriate blanks to eliminate background chemiluminescence.

TNT Degradation Sample Quenching and Preparation
For analyses of each TNT degradation experiment, a 2-mL sample of the modified Fenton reaction mixture was combined with 100 µL of sodium bicarbonate (approximately 0.9 M) in a test tube in order to stop the reaction (Glaze and Kang, 1988). A 1-mL aliquot of the quenched material was prepared for high performance liquid chromatography (HPLC) analysis by filtering through a 0.2-µm nylon filter and placed into 1.5-mL amber vials with Teflon-lined septa.

Chemical Analyses
The TNT concentrations were determined using HPLC (Model 1090, Series II; Hewlett Packard, Palo Alto, CA) equipped with a security guard column containing a C18 (ODS octadecyl) filter connected to a C18 reverse-phase column (250 mm x 2.0 mm x 5 µm; Phenomenex, Torrance, CA). A binary solvent, gradient elution methodology was used and consisted of (i) acetonitrile and (ii) 0.5 mM lithium phosphate buffer, pH 4.0 ± 0.1, at a flow rate of 0.22 mL min-1. Initial conditions were 5% acetonitrile (0 to 3 min), to 51% acetonitrile (3 to 21 min, held 12 min), to 70% acetonitrile (21 to 29 min, held 4 min), with a return to initial setup conditions by 32 min. A 10-µL injection of each sample collected was analyzed with the temperature of the HPLC column held constant at 40°C. The HPLC was equipped with a diode array UV/Visible light detector (DAD) monitoring A230 with continuous scanning of the absorption spectrum of each peak from 190 to 600 nm. Compounds detected were identified by a comparison of their retention times and UV/Visible light spectra with those of authentic standards.

Identification of nitrate and oxalate ions was determined by ion chromatography using conductivity detection (Dionex Corporation, Sunnyvale, CA). A binary solvent, isocratic elution methodology was used for separation and consisted of 89.5% double distilled deionized water and 11.5% 100 mM NaOH solution. Total run time for each sample was 9 min. An IonPac AG-11 (Dionex) guard column connected to an IonPac AS11 column with an IonPac ATC-1 filter for carbonate removal were used for all separations. Anions detected were identified and quantified based on comparison with known reference standards.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Two major groups of experiments were conducted in this project: (i) optimization experiments, using abiotic Fenton reactions, designed to find minimum reactant concentrations resulting in maximum TNT mineralization; and (ii) kinetic experiments, using abiotic Fenton reactions (conducted at reactant concentrations as found in optimization experiments) coupled with follow-on biotic reactions, to determine the increase in TNT mineralization due to biological treatment. Both groups of experiments were conducted with four different reaction conditions: light-Fenton and dark-Fenton reactions at pH 3 and light-Fenton and dark-Fenton reactions at pH 7. Two additional minor groups of experiments were conducted, based on optimization results: (i) determination of the effects of iron chelate concentrations on TNT mineralization extent and (ii) overall TNT degradation kinetics. All experiments used 0.22 mM aqueous TNT solutions with varying concentrations of Fe+3 and H2O2 in the modified Fenton reactions. The biotic portion of kinetic experiments used varying concentrations of two uncharacterized biomasses, previously described.

Abiotic TNT Mineralization Optimization
Optimal concentrations of reactants, Fe+3 and H2O2, in aqueous, modified Fenton reactions were determined from the results of experiments in factorial designs (Tables 1 and 2) for the four reaction conditions described above. Initial experiments were based on rotatable, central composite designs (dark-Fenton reactions at pH 3) but, upon failing to show rotatability, were converted to factorial designs more appropriate for investigating the resulting maximal responses (Chochran and Cox, 1992). Iron(III) and hydrogen peroxide were tested over concentration ranges of 0.05 to 20 mM and 15 to 294 mM, respectively. Optimal reactant concentrations, the lowest concentrations of reactants producing the highest extent of TNT mineralization (shown in italic type in Tables 1 and 2) were determined graphically (not shown) for each experiment. These reactant concentrations were used for all subsequent experimentation. Experimental error for the data (Tables 1 and 2) was approximately 0.8% mineralization, based on the standard deviation of the mean of center point replicates of the central composite experiment (Table 1). Carbon balances, based on recovery of 14C (Table 3), were done for all optimization data (Tables 1 and 2) and averaged approximately 90% recovery for each reaction condition.


View this table:
[in this window]
[in a new window]
  		Table 3. Percent recovery of 14C in optimization (Tables 1 and 2) and kinetic (Fig. 3–5) experiments related to aqueous TNT (0.22 mM) mineralization. Data are means ± standard error (P <= 0.05).

 


View larger version (34K):
[in this window]
[in a new window]
  		Fig. 3. Mineralization of aqueous TNT (0.22 mM) in dark-Fenton reactions comparing abiotic with coupled abiotic–biotic treatments at high and low biomass (waste-activated sludge [WAS] or 2,4-dinitrophenol [DNP] biomass) concentrations: (A) pH 3, WAS biomass; (B) pH 3, DNP biomass; (C) pH 7, WAS biomass; (D) pH 7, DNP biomass. Abiotic reaction (•); high biomass concentration, either 467 mg WAS L-1 or 333 mg DNP L-1 ({circ}); low biomass concentration, either 93 mg WAS L-1 or 67 mg DNP L-1 ({blacktriangledown}). Abiotic, pH 7, reactions conducted with Fe3+ to nitrilotriaceticacid (NTA) ratio = 1:1. Biotic reactions for (A) and (B) were conducted after solution neutralization. Error bars on symbols indicate standard error of means (n = 3, P <= 0.05); where absent, bars fall within symbols.

 


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 5. Mineralization of aqueous TNT (0.22 mM) in light-Fenton reactions at pH 7 comparing abiotic with coupled abiotic–biotic treatments at high and low biomass (waste-activated sludge [WAS] or 2,4-dinitrophenol [DNP]) concentrations: (A) WAS biomass; (B) DNP biomass. Abiotic reaction (•); high biomass concentration, either 467 mg WAS L-1 or 333 mg DNP L-1 ({circ}); low biomass concentration, either 93 mg WAS L-1 or 67 mg DNP L-1 ({blacktriangledown}). Abiotic reactions conducted with Fe3+ to nitrilotriaceticacid (NTA) ratio = 1:1. Error bars on symbols indicate standard error of means (n = 3, P <= 0.05); where absent, bars fall within symbols.

 
Several phenomena were observed upon examination of the results. In all of the optimization experiments, TNT mineralization extent declined with reactant concentrations above optimal (Table 1), presumably due to competitive reactions between the reactive oxygen radical species and excess Fe+3 (De Laat and Gallard, 1999). Greater TNT mineralization extent in light-Fenton reactions (Table 2) than in dark-Fenton reactions (Table 1), under similar conditions, was attributed to photo-assisted Fenton oxidation, mechanistically described by others (Sun and Pignatello, 1993; Li et al., 1997a). Overall TNT mineralization extent in dark-Fenton reactions was considerably less under neutral pH conditions than under acidic conditions. This was attributed to either or both (i) a lowered efficiency of hydroxyl radical formation at neutral pH or (ii) a higher reactivity of hydroxyl radical with the iron–chelate complex than with TNT. The chelating agent (NTA) was required to prevent formation of iron oxide precipitates at neutral pH, thus removing the soluble iron catalyst from solution and limiting the Fenton-like reaction.

Effect of Chelate Concentration
As mentioned above, the efficiency of Fenton reactions with TNT at neutral pH was lowered, presumably due to the use of a chelate. We therefore explored the effects of different chelate (NTA) concentrations on TNT mineralization in light- and dark-Fenton reactions at neutral pH. Optimal concentrations of iron(III) and hydrogen peroxide (Tables 1 and 2) were used in these experiments. A 1:1 molar ratio of chelate to iron was the lowest concentration of NTA that maintained iron in solution. The light-Fenton reaction with a 1:1 NTA to iron ratio resulted in the greatest TNT mineralization extent of 66%, whereas the same reaction with a 10:1 ratio produced 44% TNT mineralization (data not shown). Similarly, with dark-Fenton reactions, the lower molar ratio of NTA produced 45% TNT mineralization versus 28% for the reaction with a 10:1 molar ratio of NTA to Fe (data not shown).

TNT Degradation
Degradation of TNT has been a common benchmark for success of a treatment process (Schmelling and Gray, 1993; Schmelling et al., 1996) and a good complement to measuring the mineralization of TNT. By using both measurements, one may determine how much of the TNT parent molecule remains as well as the fraction mineralized. Degradation experiments were performed for the four light- and dark-Fenton reactions using the optimal reactant and chelate concentrations defined previously. Two additional light- and dark-Fenton reactions, at neutral pH, were also conducted with optimal reactant concentrations and excess chelate to iron concentrations (10:1). It was observed (Fig. 2) that 100% of the TNT was transformed under each reaction condition within 24 h and, in most cases, within 6 h. The effect of lowering the chelate concentration was seen in these degradation experiments. The dark-Fenton reaction at neutral pH, and containing a 10:1 chelate to iron ratio, required nearly 24 h for complete TNT degradation as compared with 10 h for an experiment with similar reaction conditions, but with a 1:1 chelate to iron ratio. Photo-Fenton effects were also observed (Fig. 2). More time was required for complete TNT degradation in all dark-Fenton reactions (6–24 h) compared with light-Fenton reactions (2–6 h).



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2. Degradation of aqueous TNT (0.22 mM) in dark- or light-Fenton reactions: dark, pH 3 (•); light, pH 3 ({triangledown}); light, pH 7 (Fe3+ to nitrilotriaceticacid [NTA] ratio = 1:1) ({blacksquare}); dark, pH 7 (Fe3+ to NTA ratio = 1:1) ({circ}); light, pH 7 (Fe3+ to NTA ratio = 1:10) ({square}); dark, pH 7 (Fe3+ to NTA ratio = 1:10) ({blacktriangledown}). See text for description of dark- and light-Fenton reactions. Error bars on symbols indicate standard error of means (n = 3, P <= 0.05); where absent, bars fall within symbols.

 
Coupled Abiotic–Biotic TNT Mineralization Kinetics
The kinetics of a treatment process must be known prior to assessing its overall viability, cost, and usefulness in an engineered system (Grady et al., 1999). We therefore conducted several sets of experiments to determine (i) the kinetics of the abiotic, modified Fenton reaction and (ii) the kinetics of a combined, sequential abiotic–biotic reaction scheme using modified Fenton reactions followed by biological treatment. All Fenton reactions were investigated using the optimal reactant concentrations previously determined. The abiotic reaction was run for 24 h at the pH tested with subsequent addition of biomass after solution neutralization.

Capture and analysis of 14CO2 resulting from biological or chemical transformation of radiolabeled compounds is widely regarded as incontrovertible evidence of compound mineralization (Bartha and Pramer, 1965; Code of Federal Regulations, 1996). Using modified biometer flasks, described above, we conducted kinetic studies on the abiotic mineralization of TNT in dark-Fenton reactions, at both pH 3 and 7. The cumulative TNT mineralization achieved in each reaction condition, 47 and 45% respectively, was approximately 85% complete within 24 h (Fig. 3) . The extent of TNT mineralization in dark-Fenton reactions at neutral pH (45%) was greater than that seen in the optimization experiments (25%) due to the lowered NTA to Fe ratio (1:1) found during chelate optimization studies. Kinetic studies using dark-Fenton reactions with a NTA to Fe ratio of 10:1 (similar to the initial optimization studies), at neutral pH, yielded an ultimate TNT mineralization extent of 28% (data not shown).

In an effort to increase TNT mineralization using a coupled abiotic–biotic scheme, experiments were conducted with the addition of biotic treatment after the abiotic, modified Fenton treatments previously investigated. Two different biomasses were tested, a waste-activated sludge (WAS) biomass and a 2,4-dinitrophenol (DNP)–degrading biomass, both cultivated in bench-scale sequencing batch reactors (Hess et al., 1993). Experiments were conducted with the four reaction conditions based on pH, acidic and neutral, and presence or absence of light during the abiotic Fenton reactions. The timing of biomass addition to the abiotic TNT reaction was critical and several time intervals after abiotic reaction initiation (6, 10, and 14 h) were initially tested. In all tests, TNT mineralization increased with the addition of biomass (data not shown). However, presumably due to abiotic reaction quenching by the biotic, cellular constituents (catalase and peroxidase), the overall coupled abiotic–biotic TNT mineralization extent increased with increasing time interval prior to biomass addition, indicating improper timing of the addition. We therefore conducted further experiments with biomass addition 24 h after abiotic reaction initiation (Fig. 3). Overall TNT mineralization increased from 47%, in the dark-Fenton reaction, pH 3, to 75 and 80% in the coupled abiotic–biotic system with WAS and DNP biomasses, respectively (Fig. 3A,B). In reactions conducted at neutral pH, an increase in TNT mineralization from 45% in the dark-Fenton reaction to 78% in both of the coupled abiotic–biotic systems (WAS or DNP biomasses) was also seen (Fig. 3C,D). Direct TNT mineralization due to biological action was negligible (less than 2%) based on results from biotic controls (Fig. 4) . Analysis of 14C for all kinetic experiments indicated adequate carbon balances (Table 3) with most experiments achieving at least 90% recovery of radioactivity. Lower recovery percentages (>90%) in some experiments were attributed to experimental error.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 4. Biotic mineralization of aqueous TNT (0.22 mM) with two different biomasses, 467 mg waste-activated sludge (WAS) L-1 (•) and 333 mg 2,4-dinitrophenol (DNP) L-1 ({circ}). Error bars on symbols indicate standard error of means (n = 3, P <= 0.05); where absent, bars fall within symbols.

 
The mechanism of increased TNT mineralization in the coupled abiotic–biotic system over that of the abiotic reaction alone was probably due to biotic assimilation of the abiotic TNT transformation products. As recently reported by others (Li et al., 1997a), dicarboxylic acids, principally oxalic acid, and nitrate are the primary end products of dark-Fenton reactions with TNT. We also found accumulation of oxalate and nitrate in solution after abiotic reactions in approximate 1.4 and 3.1 molar ratios to TNT, respectively (data not shown). In such reaction conditions (low pH, dark reaction), oxalate is essentially unreactive with hydroxyl radical (KOH = 4.7 x 107 mol s-1; Buxton et al., 1988) and would persist in solution. Oxalate is readily usable by many aerobic organisms, entering their metabolic pathways after breakdown via oxalate decarboxylase (Tanner and Bornemann, 2000; Hokama et al., 2000). In our abiotic–biotic experiments, there was no statistical difference (P <= 0.05) between ultimate TNT mineralization extent (approximately 78 ± 3%) for WAS or DNP biomasses, at both acidic and neutral pH and low biomass concentrations (Fig. 3A,B and Fig. 3C,D, respectively), indicating that both consortia of organisms probably used a common enzyme and/or pathway to degrade the byproducts of Fenton-mediated transformation of TNT. Analysis of the data in two experiments (Fig. 3B,C), however, showed statistical differences between mineralization kinetics in the biotic reaction for low and high concentrations of similar biomasses. In both experiments, the low biomass concentrations (67 or 93 mg L-1 as TSS) produced a higher extent of TNT mineralization in a shorter length of time than did the higher concentrations of biomass (333 or 467 mg L-1 as TSS). This phenomenon may have been due to quenching, by the biomass, of the Fenton reaction, only approximately 85% complete (Fig. 3), thereby slowing the biotic reaction and lowering the overall TNT mineralization extent. Such solution interactions probably also resulted in coexistent abiotic and biotic reactions, recently shown to exist in combined Fenton–biological reactions degrading perchloroethylene and its transformation products (Buyuksonmez et al., 1999) and perchloroethylene and oxalate (Howsawkeng et al., 2002).

Results of kinetics studies on TNT mineralization in light-Fenton reactions were quite different from those in dark-Fenton reactions. Abiotic, modified Fenton reactions at pH 3, using optimal reactant concentrations (Table 2) previously found, resulted in near-complete mineralization of TNT (>95%) within 48 h (data not shown). The high extent of TNT mineralization, in photo-assisted reactions, was similar to that found by Li et al. (1997a), with the exception of our lower reactant concentrations. Because of the essentially complete mineralization of TNT at acidic pH when photo-assisted, abiotic Fenton reactions were used, no follow-on biotic reactions were necessary. However, at neutral pH, the abiotic light-Fenton reaction (using optimal reactant concentrations from Table 2 and a 1:1 ratio of NTA to Fe+3) resulted in only approximately 65% TNT mineralization (Fig. 5) , presumably due to lowered efficiency of hydroxyl radical formation as compared with acidic pH. Surprisingly, when biomass was added to the abiotic reaction mixture after 24 h of reaction (at approximately 90% of reaction completion), no significant increase in TNT mineralization occurred. As seen in Fig. 5, two different concentrations of WAS biomass and DNP biomass, respectively, failed to appreciably alter the abiotic TNT mineralization kinetics.

The lack of biotic activity after abiotic light-Fenton reactions was probably due to photo Fenton–induced destruction of oxalate (or other Fenton transformation products of TNT), the only microbially assimilable carbon in solution. Other researchers have found high a extent of oxalate mineralization in photo-Fenton systems (Zuo and Hoigne, 1992; Li et al., 1997a). Such phenomena are reportedly due to photodissociation of the chelate in solution by ligand-to-metal charge transfer excitation followed by reaction of the photoreduced iron with peroxide in the Fenton reaction (Sun and Pignatello, 1993).


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Abiotic reactions were optimized for maximum TNT mineralization using minimum reactants, Fe3+ and H2O2, under acidic pH and neutral pH conditions in modified light- and dark-Fenton reactions. We demonstrated that 100% degradation could be obtained under each reaction condition, but the observed mineralization varied widely. At acidic pH, light-Fenton reactions yielded near complete mineralization of TNT (>99%) versus 43% in dark-Fenton reactions. At neutral pH, light- and dark-Fenton reactions produced a TNT mineralization extent of 36 and 25%, respectively. Upon reducing the chelate to iron molar ratio (from 10:1 to 1:1) in solution, TNT mineralization was increased to 66% and 45% in the light- and dark-Fenton reactions, respectively.

Addition of biomass to the abiotic reaction products increased overall TNT mineralization extent at both acidic and neutral pH in the modified dark-Fenton reactions. Using this coupled reaction process, abiotic (dark-Fenton reaction) TNT mineralization at pH 3 was increased from 47 to 75 or 80% using WAS biomass or DNP biomass, respectively. Under neutral pH conditions (dark-Fenton reaction), TNT mineralization was increased from 45 to 78% by both biomasses. The increase in TNT mineralization was probably due to the abiotic formation of organic acids as intermediate products that were subsequently transformed by microorganisms.

Kinetics of the coupled abiotic–biotic reactions were comparatively rapid. The abiotic reaction required 24 h while the addition of the biomass required an additional 4 d including the 10-h lag between the time of addition of the biomass and an increase in the mineralization of TNT. These kinetics are competitive with chemical treatments and much faster than most stand-alone biological treatments.

Application of the coupled abiotic–biotic process may have the most utility when using the dark-Fenton reaction. In aqueous, reactor-based processes, a high extent of TNT mineralization is achievable, comparable with photo-Fenton reactions, with no additional power input or concern for light penetration. A wider use of the process, however, may be for the remediation of soils contaminated with TNT, in both slurry reactor and in situ treatments, where photo-Fenton processes are not applicable. We are currently investigating the use of these coupled processes for remediation of contaminated soils.


   ACKNOWLEDGMENTS
 
Appreciation is expressed to NSF EPSCoR Cooperative Agreement OSR-9350539 for funding assistance on this project.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
P. S. Schrader and T. F. Hess
Coupled Abiotic-Biotic Mineralization of 2,4,6-Trinitrotoluene (TNT) in Soil Slurry
J. Environ. Qual., July 1, 2004; 33(4): 1202 - 1209.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hess, T. F.
Right arrow Articles by Schrader, P. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hess, T. F.
Right arrow Articles by Schrader, P. S.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Hess, T. F.
Right arrow Articles by Schrader, P. S.
Related Collections
Right arrow Bioremediation and Biodegradation
Right arrow Organic Compounds
Right arrow Water Pollution

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Vadose Zone Journal Journal of Plant Registrations
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal