tetrazocine by Hybrid Poplar Trees - American Chemical Society

measured log Kow for HMX was 0.19, less than other explosives,. TNT, and RDX. However ... day exposure using radiochromatography of plant tissue extra...
3 downloads 0 Views 910KB Size
Environ. Sci. Technol. 2002, 36, 4649-4655

Uptake and Leaching of Octahydro-1,3,5,7-tetranitro-1,3,5,7tetrazocine by Hybrid Poplar Trees JONG MOON YOON, BYUNG-TAEK OH, CRAIG L. JUST, AND JERALD L. SCHNOOR* Department of Civil and Environmental Engineering, 4112 Seaman’s Center (SC), The University of Iowa, Iowa City, Iowa 52242

The feasibility of remediating a high explosive, octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX), using hybrid poplar trees (Populus deltoides × nigra, DN34) was investigated. The fate, transport, and toxicity were determined. HMX was taken up by poplar cuttings from hydroponic solutions in long-term experiments (65 days) without evidence of toxicity. HMX was not toxic to actively growing hybrid poplar cuttings, even under saturated conditions. The measured log Kow for HMX was 0.19, less than other explosives, TNT, and RDX. However, the calculated transpiration stream concentration factor (TSCF) and root concentration factor (RCF) for HMX from an uptake study using radiolabeled [U-14C]HMX were 0.21 ( 0.07 and 5.55 ( 1.78 mL/ g, respectively, both of which were intermediate between the values for TNT and RDX in previous reports. A 70% uptake of [U-14C]HMX was translocated and accumulated in leaves, and no metabolites were observed during a 65day exposure using radiochromatography of plant tissue extracts. Most of the accumulated HMX (57%) in dried (fallen) poplar leaves was leached by deionized water after 5 days. Bioaccumulation in poplar trees and resolublization of HMX from leaves would be of significant ecological concern, and phytoremediation may not be warranted as a treatment option unless other processes occur under field conditions that degrade HMX to innocuous end products (e.g., photolysis, hydrolysis, or microbial degradation).

Introduction Concern regarding military ammunitions waste has increased because of potential environmental and health impacts. Groundwater and soils from closed military bases, explosive production plants, and weapon dismantling facilities are contaminated with explosives, such as 2,4,6-trinitrotoluene (TNT), hexahyro-1,3,5-trinitro-1,3,5-triazine (RDX), and octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX). HMX has replaced TNT and RDX in numerous military applications because of its higher chemical yield and stability. Figure 1 shows the chemical structure of HMX and, for comparison, RDX and TNT. HMX is a heterocyclic aliphatic explosive compound with an eight-membered ring and stable “crown” conformation (1). The concentration of HMX has decreased in groundwater at the Joliet Army Ammunition Plant (JAAP), but it remained above the desired level of concentration in the JAAP soils (2) * Corresponding author phone: (319)335-5649; fax: (319)335-5660; e-mail: [email protected]. 10.1021/es020673c CCC: $22.00 Published on Web 10/05/2002

 2002 American Chemical Society

FIGURE 1. Chemical structures of (a) HMX, (b) RDX, and (c) TNT. prior to a removal action and windrow composting. Simini et al. reported that the concentration of HMX ranged from 1 to 3055 mg/kg in 13 out of 41 JAAP soils, which had been earlier the site of explosives incineration, munitions production, and assembly operation (3). Even though chronic carcinogenicity and toxicity thresholds for HMX to humans is not well-determined, HMX is reported to be toxic to some aquatic organisms, mice, and rats. HMX toxicity to the earthworm was investigated, but growth inhibition and reproduction of adult earthworms occurred at the high concentration (4). Several studies of treatment of HMX have been reported. They focused on alkaline hydrolysis (5), denitrification followed by hydrolysis (6), and bioremediation (7). Recently, the metabolism pathway of HMX using microbes in municipal anaerobic sludge was suggested by Hawari et al. (8). They identified nitroso derivatives as well as methylenedinitramine and bis(hydroxymethyl)nitramine intermediates. The intermediates were degraded further into other products: CO2, N2, nitrous oxide (N2O), and formaldehyde (HCHO). This suggests that microbial biotransformation routes of HMX are similar to that of RDX (8, 9). Phytoremediation is an innovative technology using plants to clean up contaminated soil and groundwater pollutants because it is cost-effective and environmentally friendly (10). Several studies have been conducted with explosives. Thompson et al. showed the uptake and translocation of RDX (11) and TNT (12) by hybrid poplar trees. Bhadra et al. investigated the phytoremediation of RDX and HMX using aquatic plants (Myriophyllum aquaticum) and hairy root cultures of Catharanthus roseus (13). Uptake and translocation were evident in these studies, but the long-term fate of HMX in intact plant tissues was not evaluated. The goal of this research was to investigate the feasibility of hybrid poplar trees to uptake, translocate, and transform HMX because the long-term fate of HMX in hybrid poplar trees, the model plant system of choice in many phytoremediation applications, has not been evaluated previously.

Materials and Methods Imperial Carolina hybrid poplar cuttings (Populus deltoides × nigra, DN-34) were obtained from Hramoor Nursery (Manistee, MI). The cuttings were cut to 25 cm long and grown for 28 days with half-strength Hoagland nutrient solution (14) adjusted to pH 6.8-6.9 with 1.0 N NaOH. The cuttings were transferred into 250-mL Erlenmeyer flasks containing 200 mL of aqueous solution before starting experiments. The flasks were wrapped in aluminum foil to inhibit algae growth and photodegradation of HMX. The reactors also were sealed with the modified Teflon septa and hole caps or covered with cardboard to reduce evaporation. They were placed randomly under fluorescent lighting with a light intensity between 120 and 180 µmol m-2 s-1 and a photoperiod of 16 h a day. Chemicals. Unlabeled HMX was provided by the Iowa Army Ammunition Plant (99% pure), and external standards VOL. 36, NO. 21, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4649

for HMX were purchased from Chemservice (West Chester, PA). For labeled HMX, the powder form of [U-14C]HMX obtained from NEN (Boston, MA) was 97.0% pure, and it exhibited a radioactivity of 6.84 mCi/mmol. All other chemicals used were reagent grade or better. HPLC Analysis. HMX concentrations in hydroponic solution were analyzed using high-performance liquid chromatography (HPLC) with a variable wavelength UV detector (HPLC/UV, Hewlett-Packard Series 1100) at 230 nm. A mobile phase consisting of 60% deionized water and 40% methanol (vol %) was used with a flow rate of 1 mL/min. A Supelcosil LC-18 column was used for separation. The injection volume was set to 100 µL using an autosampler. All samples were filtered through 0.2-µm microfilters and diluted 1:1 with acetonitrile before injection. HMX Uptake Study. An HMX uptake study was performed to investigate the rate of decrease of HMX concentration in solution with respect to time (impulse input of HMX on day 0). Poplar cuttings were exposed to different concentrations of HMX for 21 days. One glass control group without cuttings was included in the experiment in triplicate. To estimate the transpiration stream concentration factor (TSCF), the gravimetric determination method of Thompson et al. (12) was used. Hydroponic solutions were sampled for HPLC analysis of HMX either every day or every 2 days. Batch-Fed Chemostat for Phytotoxicity. For a toxicity test, poplar cuttings were placed in nutrient solution containing various HMX concentrations for 28 days (0 for control and approximately 1.5, 3, and 5 mg/L HMX). To keep a constant mass of HMX, the hydroponic solutions were sampled daily, and the amount of HMX taken up by trees was replenished to the target concentration with HMX stock solution or half-strength nutrient solution. [U-14C]HMX Translocation in Poplar Trees. To obtain the mass balance, [U-14C]HMX was used like the uptake study. After 21, 30, 38, 50, and 65 days, plants were harvested and separated into leaves, new stems, top old stems, bottom old stems, and roots. Each part was weighed and then divided into two samples for biooxidation and extraction. Tissue samples less than 1.0 g were combusted for 4 min in a biooxidizer (RJ Harvey model OX-600, Hillsdale, NJ). Ten milliliters out of 15 mL of RJ Harvey 14carbon scintillation cocktail captured 14CO2 was transferred to a glass scintillation vial for analysis with a Beckman LS 6000IC scintillation counter (Fullerton, CA). Mannitol spiked with 0.027 µCi of [U-14C]atrazine (Sigma Chemical, 97.1% purity, 10.9 mCi/ mmol specific activity) was combusted and compared to a known standard of similar activity to determine the combustion efficiency. Plant Tissue Extraction. A total of 3-5 g of wet tissue sample was chopped with pruning shears and then placed into 50-mL polypropylene centrifuge tubes filled with 40 mL of acetone. The samples were homogenized with a biohomogenizer (except stems) and sonicated for 18 h. The sonicator was connected with a circulating water bath to keep the temperature of the sonicator at 20 °C. Then the samples were passed through 4.25 cm diameter glass fiber filters (Whatman 934-AH) using ceramic filtration funnels. A 100-µL aliquot of the filtrate was used for liquid scintillation analysis, and filter remnants were air-dried for 2 days prior to combustion analysis. Ten milliliters of deionized water was added to the filtrate, and then the mixture was distilled with a Buchi rotary evaporator at 65 °C for 3-4 min to remove the acetone. A portion of the water was transferred to 1.5-mL tubes and centrifuged for 10 min at 12 000 rpm. The supernatants were filtered and analyzed by HPLC with radiochemical detection (12). Log Kow Evaluation. To determine log Kow, 3.6 mg/L of HMX solution in deionized water was mixed with 1-octanol 4650

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 21, 2002

FIGURE 2. (a) First-order uptake plot of average ln C/C0 vs time and (b) second-order uptake normalized for biomass of octahydro-1,3,5,7tetranitro-1,3,5,7-tetrazocine (HMX) in hydroponic solutions over time. Cuttings were exposed to 4.70 (n ) 3), 2.84 (n ) 2), and 1.37 mg/L (n ) 3) initially. The initial HMX concentration in glass controls (O) were 1.77 mg/L. Error bars indicate (1 SD. n ) 3 for glass controls. (99% HPLC grade, Aldrich) in 25-mL vials. To minimize disturbance of the solution during sampling, a 60-mL separation funnel was also used. The volume of each octanol and water containing HMX was 5 mL in vials and 10 mL for separation funnels (1:1 vol %). The vials and funnels were vigorously shaken manually for 4 min. After being equilibrated for 1 h, samples were taken from both organic and aqueous phases over time and analyzed by HPLC. Both phases were diluted with acetonitrile (1:1 vol %) before injection. Leaching Experiment. A preliminary leaching experiment was performed to determine if any parent HMX could be resolubilized from poplar tissues. Poplar cuttings exposed to radiolabeled HMX produced withered and fallen leaves naturally during the course of the experiment. These leaves were collected and dried in air at room temperature. The dried, fallen leaves were chopped with scissors, wrapped with Whatman No.1 filters, and placed in the bottom of 20mL vials. Deionized water (15 mL) was added slowly to prevent the floating of the wrapped leaves. Three sets of triplicates were prepared. Two sets were placed on the shaker at 160 rpm, and the other set was unshaken. The initial radioactivity of the fallen leaves was determined by the biooxidizer method, and the radioactivity of the liquid phase

TABLE 1. Results of Linear Regression for Each Tree Sample and Estimated TSCF (n ) 8) tree

slope (k)

intercept

r2

cumulative transpiration (mL)

daily transpiration rate (mL/day)

1 2 3 4 5 6 7 8 average standard deviation coefficient variance 95% confidence

-0.02363 -0.03389 -0.03292 -0.02857 -0.04483 -0.03402 -0.02573 -0.03632 -0.03249 0.00667 -0.20515 0.00557

-0.04078 -0.00483 -0.01202 0.01763 0.01184 -0.03191 -0.02823 -0.01423

0.85360 0.93488 0.90080 0.91101 0.97918 0.97580 0.95017 0.95368

498.73 761.79 879.49 567.86 1441.95 844.17 567.25 853.45

23.75 36.28 41.88 27.04 68.66 40.20 27.01 40.64 38.18 14.21

in the vials was measured over time using a scintillation counter. Following the preliminary leaching experiment, a definitive leaching experiment was performed that resulted in a complete mass balance. The liquid phase was drained and measured for volume. The residues were analyzed by the biological oxidizer for the mass balance.

Results and Discussion HMX Uptake Study. An uptake study using unlabeled HMX was conducted for 21 days and for all trees excluding one that did not survive to the end of the experiment. No toxic effects were evident. Each initial concentration (triplicate) was 1.37, 2.84, and 4.70 mg/L, respectively. After 21 days, 44.83 ( 8.50% of the initial HMX in hydroponic solution was uptaken by the hybrid poplar plants. The glass control group without cuttings indicated that 5% of the initial concentration (1.77 mg/L) was removed from solution. Results of uptake of HMX in solution from the root zone are shown in Figure 2a, indicating that HMX is taken up continuously by hybrid poplar trees from hydroponic solution over a 21-day experiment. Bhadra et al. found that the uptake of HMX in aquatic plants (M. aquaticum) showed no significant difference between live plants and controls and that 25% of HMX was removed by hairy root cultures of C. roseus for 90-day exposure (13). From the results of the uptake study, the normalized uptake rate depended on daily transpiration. The TSCF for HMX was calculated from the following model (15):

dC ) -kC dt

(1)

and

k)

TSCF × Tr V

(2)

where k is the first-order reaction rate constant (day-1), TSCF is the transpiration stream concentration factor (fractional efficiency of uptake), Tr is the daily transpiration rate (mL/ day), V is the water volume in reactors (mL), and C is the concentration of HMX in solution (mg/L). Assuming an average transpiration rate for 21 days, the results of linear regressions of ln C/C0 versus time for eight reactors are presented in Table 1. Figure 2a shows the average of all eight reactors (a summary of the results in Table 1). Equation 2 was rearranged to solve for TSCF. TSCF values are presented in the last column of Table 1. The estimated TSCF for HMX was 0.18 ( 0.03 according to eq 2. This indicated that HMX is moderately available for uptake by the plant roots (19).

TSCF 0.20 0.19 0.16 0.21 0.13 0.17 0.19 0.18 0.18 0.03 0.14 0.02

Another conceptual model for plant uptake is secondorder uptake. Uptake of chemical by the plant is directly proportional to the aqueous chemical concentration and to the total plant biomass. The second-order rate constant (k′) is simply the first-order rate constant in eq 2 normalized to the total biomass of the plant. The biomass was taken as an average during the course of the 21-day experiment:

dC ) -k′BC dt

(3)

where k′ is the second-order rate constant (mL g-1 day-1), and B is the biomass (g mL-1). The second-order rate constant was determined from a plot of ln C/C0 divided by the biomass concentration on the y-axis (Figure 2b). In Figure 2b, the R2 of linear regression for eight samples was 0.8901, indicating that eq 3 is also a good model to normalize the uptake for all these experiments, especially when biomass varies significantly from reactor to reactor. Batch-Fed Phytotoxicity. Batch-fed chemostats were used to keep the contaminant concentration constant for a continuous toxicity test. This experiment was different than the uptake study where unlabeled HMX was spiked as an initial concentration on the first day, and it was diluted by additional nutrient solution only as the trees transpired water. In this experiment, the mass of HMX and water transpired was added back each day, holding the HMX exposure concentration roughly constant for 28 days. Each group consisted of three trees exposed to various HMX solutions and a group not exposed to HMX as a control. During the toxicity experiment period, one cutting exposed to approximately 3 mg/L died after 12 days. Also, one cutting exposed to about 5 mg/L was dead after 21 days. The dead trees did not show any toxic symptoms such as leaf abscission or chlorosis. They were the only two cuttings to die during these experiments out of 12, and it is suspected that they died from random, background causes. Vigorous growth of the plants was evident throughout the toxicity experiment. The biomass of each group after 28 days increased by 19 g for control group and by 22, 23, and 18 g for trees exposed to approximately 1.5, 3.0, and 5.0 mg/L HMX concentration, respectively. Biomass data suggested that up to 5 mg/L HMX concentration was not toxic to the poplar trees. The growth rates for all groups were not significantly different. A reported threshold concentration showing toxic effects of TNT to poplar trees was 5 mg/L (16). For RDX, Chen showed that concentrations up to 21 mg/L were not toxic to maize or wheat (17), and Thompson did not observe toxicity of RDX to poplar trees at 21 mg/L (16). These threshold concentrations of TNT and RDX are below their water solubilities at 20 °C: 130 mg/L for TNT and 38.4 mg/L for RDX (18), respectively. HMX was not as toxic as TNT or RDX to hybrid poplar trees, even at supersaturated VOL. 36, NO. 21, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4651

FIGURE 3. Photos of hybrid poplar cuttings exposed for (top) 21 and (bottom) 30 days to 8.22 mg/L HMX on day 0. Trees were actively growing, and there were no visible signs of toxicity. concentrations. In this research, hybrid poplars exposed to 8.22 mg/L (1.64 times saturation in deionized water) were actively growing during the course of the experiments and did not show signs of toxicity (Figure 3). HMX was not toxic to hybrid poplar cuttings even at saturated concentrations in these experiments. [U-14C]HMX Translocation in Poplar Trees. Plant uptake and translocation of HMX in poplar trees was further evaluated using radiolabeled [U-14C]HMX. Twelve poplar cuttings were transferred into reactors containing HMX solution mixed with labeled and unlabeled compounds. The total radioactivity was approximately 12 µCi per reactor. The ratio of the labeled to unlabeled chemical was 1.41:1, and total concentration was initially 8.22 mg/L by HPLC, exceeding the solubility limit of HMX reported by Heilmann et al. 4652

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 21, 2002

(5). The slight supersaturation of HMX in the nutrient solution seemed to be achieved by acetone used as a cosolvent for [U-14C]HMX. Two cuttings were not included in the mass balance calculation because of the death and mishandling. Another six reactors were spiked with only radiolabeled HMX. The six cuttings spiked with 10 µCi per reactor were sacrificed after 38 and 65 days for analysis of metabolites in tissue samples and for mass balance calculations. The remaining radioactivity of [U-14C]HMX in solution was less than 35% of the applied radioactivity when the trees were sacrificed. The mass balance recoveries from biooxidation are shown in Table 2 with an average recovery of 87.69%. The majority (69.6%) of the uptaken [U-14C]HMX was translocated into the leaves. About 6.8%, 2.5%, 13.9%, and 7.2% of the uptaken label was distributed in roots, new stems, bottom old stems,

TABLE 2. Mass Balance: Percent Distributions of Applied 14C in Uptake Experiments Using a Biooxidizer time (days)a

leaves

new stems

old stems (top)

old stems (bottom)

roots

remaining in solution

removed by sampling

total recovery

21 (n ) 3)b 30 (n ) 4)b 38 (n ) 3)c,d 50 (n ) 3)b 65 (n ) 3)c

24.3 ( 9.5 46.9 ( 3.5 51.4 ( 15.2 47.8 ( 21.8 59.9 ( 2.8

1.12 ( 0.16 1.36 ( 0.24 1.80 ( 0.88 1.63 ( 0.70 1.84 ( 0.32

3.15 ( 0.62 4.01 ( 0.87 6.95 ( 0.77 4.13 ( 1.13 4.59 ( 0.36

6.46 ( 1.10 7.38 ( 0.17 13.8 ( 3.3 8.38 ( 1.19 6.02 ( 2.08

2.89 ( 1.44 4.03 ( 0.54 3.13 ( 0.50 3.46 ( 0.98 3.27 ( 0.81

35.4 ( 5.7 20.1 ( 1.9 14.7 ( 6.2 15.1 ( 5.1 5.49 ( 0.84

4.52 ( 0.05 5.31 ( 0.24 5.34 ( 0.54 6.36 ( 0.34 5.97 ( 1.07

77.8 ( 13.3 89.1 ( 3.4 97.1 ( 5.3 86.9 ( 17.6 87.1 ( 3.7

a n is the number of cuttings used for analysis. used for 38-day mass balance.

b

The initial radioactivity was 12 µCi. c The initial radioactivity was 10 µCi.

d

Dried leaves were

TABLE 3. Comparison of log Kow and Calculated and Experimental Root Concentration Factors (RCFs) and Transpiration Stream Concentrations (TSCFs) for Explosives by Hybrid Poplar Cuttingsa calculated valueb Briggs log Kow TNTc 1.90d RDX 0.90e HMX 0.19

Burken

RCF RCF (mL/g) TSCF (mL/g) TSCF 1.70 0.96 0.86

0.78 0.57 0.28

3.46 3.10 3.04

experimental value RCF (mL/g)

TSCF

0.66 49.0 ( 30.0d 0.46 ( 0.19d 0.28 1.30 ( 0.73e 0.16 ( 0.06e 0.10 5.55 ( 1.78 0.21 ( 0.07

a n ) 10 for the estimation of RCF and TSCF for HMX. b The equations used for calculated values of TSCFs and RCFs are as follows: TSCF ) 0.784 exp{-(log Kow - 1.78)2/2.44}, log(RCF - 0.82) ) 0.77 log Kow 1.52 (Brigg) and TSCF ) 0.756 exp{-(log Kow - 2.50)2/2.58}, log(RCF - 3.00) ) 0.65 log Kow - 1.57 (Burken). c TNT is rapidly uptaken by hybrid poplar roots, but it does not translocate to shoots much. d Data for TNT from Thompson et al. (12). e Data for RDX from Thompson et al. (11).

FIGURE 4. (a) Mass balance from extraction and (b) percent distributions of applied 14C-radiolabel determined from extraction experiment after 30 days; n ) 3 for each group. Error bars indicate (1 SD. E and UE represent extractable and unextractable portions in roots, stems, and leaves. Unextractable portions were quantified by combustion and recovered as 14CO2. and upper old stems, respectively. The results of HMX biooxidation showed similarity with RDX (10). Thompson found that 60% of the uptaken [U-14C]RDX was in the leaves of poplar trees after 9 days. In the case of TNT, 78% of the radiolabel was absorbed (59% of the applied label) into root tissues after 1.7 days (12). The efficiency of biooxidation ranged from 92% to 99% at the beginning and end of the experiment in this research (11). Mass balance and distribution of 14C via extraction of plant tissues is shown in Figure 4. The distribution by extraction showed a similar trend to the results from biooxidation. Most of applied 14C (47.38%) remained in leaves when the cuttings were exposed for 30 days. As an extraction solvent for HMX from plant tissues, acetone was used because it was reported to be a better extractant than methanol (16). When the distilled acetone was analyzed with a liquid scintillation

counter, no radioactivity was measured. Mass balance recoveries averaged over 80% by extraction and 77-97% by biooxidation/combustion (Table 2). The log Kow of HMX was determined using a “shake flask” method (19), and it was 0.19, lower than those of TNT and RDX. There was no difference in the log Kow value using either vials or separation funnels. Results did not vary with sample times between 3 h and 7 days. Leo et al. reviewed partition coefficients and the experimental methods (19), suggesting that 100 inversions in 5 min were enough to produce consistent results. The log Kow values of explosives measured in this research are presented in Table 3. The experimentally determined and calculated values (20, 21) for TSCF and RCF are also shown in Table 3 for TNT (12), RDX (11), and HMX. Both TSCF and RCF for HMX were intermediate between those for TNT and RDX. Experimentally determined TSCF for HMX were within the range of the values predicted by the Burken and Schnoor model (20) and Brigg’s equation (21). Fate of [U-14C]HMX. Plant tissues were extracted and analyzed to quantify metabolites. Using radiochromatography, no detectable metabolites of HMX in poplars were observed. In Figure 5, radiochromatograms are shown from distilled extracts of roots, stems, and leaves after 50 days. Also, similar results and radiochromatograms were obtained from extracts of cuttings sacrificed after 30, 38, and 65 days. From the radiochromatography, the chemical in plant tissues that remained as parent compound, HMX, and plant metabolites were not detected in any samples even after 65 days. Accumulation of HMX in poplar leaves without transformation is a potential concern from an environmental standpoint. These results agree with those reported for rye grass (Lolium perenne) by Groom et al. (22). HMX and its reduced chemical, octahydro-1-nitroso-3,5,7-trinitro-1,3,5,7tetrazocine (MN-HMX), were translocated into the blade tissue of rye grass, but HMX was not degraded. However, VOL. 36, NO. 21, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4653

FIGURE 5. Radiochromatograms of extracts from (a) leaves, (b) stems, and (c) roots in 50 days. Panel d is radiochromatograms of liquid phase from the leaching experiment with deionized water for 3 days. The peak of an unknown polar compound was observed and more obvious by purging (solid line) with nitrogen gas than without purging (dash line). The peak was not detected by the same treatment with acetonitrile. they found metabolite products of RDX and HMX in the soils contaminated with explosives (22). From the preliminary leaching experiment, it was shown that HMX can be resolubilized from poplar tissues using deionized water. The initial radioactivity of withered leaves was 0.12 µCi per dried leaf (about 0.1 g). A total of 57% of the radioactivity in leaf tissues was released into water after 2 days on the shaker, but it took 5 days to release the same amount of radioactivity without shaking (Figure 6). In the definitive leaching experiment, dried leaves (about 0.3 g, radioactivity 0.7-1.3 µCi/g by combustion) were used to determine the mass balance. Fifty-two percent of the total radioactivity in the leaves was leached out, and the residual portion was 41%. The total recovery of 14C was 93.32 ( 5.30% from biooxidation (BO) analysis and liquid scintillation (LSC). The liquid phase was analyzed by radiochromatography. Most radioactivity in the liquid phase (79% of total peak area) existed as parent [U-14C]HMX and an unknown compound (12%) with a shorter retention time than HMX (Figure 5d). The peak of the unknown compound was not detected in the liquid phase by extraction with acetonitrile and purging with nitrogen gas. Poplars are deciduous, and when leaves laden with HMX fall from the tree, there will be potential concern of contamination of the ecosystem and food chain transfers. However, in field studies high concentrations of explosives have not been measured in plant tissues at explosives contaminated sites (23). This may indicate that 4654

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 21, 2002

FIGURE 6. Leaching of HMX from dried, fallen leaves in deionized water. Two triplicates (b and 2) were shaken at 160 rpm. The others (9) were in unshaken condition. The initial radioactivity was 0.12 µCi/0.1 g based on dried weight; 0.1 g was used for (9) and (b), and 0.2 g was used for (2). Error bars indicate (1 SD. processes at work in the field (photolysis, hydrolysis, or microbial degradation) may play a greater role than in laboratory hydroponic studies.

In conclusion, HMX is taken up by hybrid poplar trees readily, and it is not toxic to the plant. However, the compound is not metabolized or transformed in plant tissues under a hydroponic system, and it can be leached from fallen leaves. Therefore, applications of phytoremediation (other than phytoextraction) may not be warranted at this time.

Acknowledgments The research was supported by an award from the U.S. Army, DAAA09-99-0049. We thank Cyril Onewokae, Industrial Operations Command. We also thank Benoit Van Aken, Jeremy Rentz, Jason Christopherson, and Dong-Koo Park for technical assistance and comments on the experiments.

Literature Cited (1) Iqbal, Z.; Bulusu, S.; Autera, J. R. J. Chem. Phys. 1974, 60, 221230. (2) Pennington, J.; Harrelson, D. Technical Report EL-98-8; U.S. Army Industrial Operations Command: Rock Island, IL, 1998. (3) Simini, M.; Wentsel, R. S.; Checkai, R. T.; Phillips, C. T.; Chester, N. A.; Major, M. A.; Amos, J. C. Environ. Toxicol. Chem. 1995, 14, 623-630. (4) Robidoux, P. Y.; Hawari, J.; Thiboutot, S.; Ampleman, G.; Sunahara, G. I. Environ. Pollut. 2001, 111, 283-292. (5) Heilmann, H. M.; Wiesmann, U.; Stenstrom, M. K. Environ. Sci. Technol. 1996, 30, 1485-1492. (6) Zoh, K.; Stenstrom, M. K. Water Sci. Technol. 1997, 36, 47-54. (7) Boopathy, R.; Manning, J.; Kulpa, C. F. Water Environ. Res. 1998, 70, 80-86. (8) Hawari, J.; Halasz, A.; Beaudet, S.; Paquet, L.; Ampleman, G.; Thiboutot, S. Environ. Sci. Technol. 2001, 35, 70-75. (9) Halasz, A.; Spain, J.; Paquet, L.; Beaulieu, C.; Hawari, J. Environ. Sci. Technol. 2002, 36, 633-638.

(10) Schnoor, J. L.; Licht, L. A.; McCutcheon, S. C.; Lee Wolfe, L.; Carreira, L. H. Environ. Sci. Technol. 1995, 29, 318A-323A. (11) Thompson, P. L.; Ramer, L. A.; Schnoor, J. L. Environ. Toxicol. Chem. 1999, 18, 279-284. (12) Thompson, P. L.; Ramer L. A.; Schnoor, J. L. Environ. Sci. Technol. 1998, 32, 975-980. (13) Bhadra, R.; Wayment, D. G.; Williams, R. K.; Barman, S. N.; Stone, M. B.; Hughes, J. B.; Shanks, J. V. Chemosphere 2001, 44, 1259-1264. (14) Epstein, E. Mineral nutrition of plants: Principles and perspectives; John Wiley & Sons: New York, 1972. (15) Burken, J. G.; Schnoor, J. L. J. Environ. Eng. 1996, 122, 958-963. (16) Thompson, P. L. Phytoremediation of munitions (RDX, TNT) waste at the Iowa Army Ammunition Plant with hybrid poplar trees. Ph.D. Thesis, University of Iowa, 1997. (17) Chen, D. Plant uptake and soil adsorption of RDX. Master’s Thesis, University of Illinois, 1993. (18) Talmage, S. S.; Opresko, D. M.; Maxwell, C. J.; Welsh, C.; Cretalla, F. M.; Reno, P. H.; Daniel, F. B. Rev. Environ. Contam. Toxicol. 1999, 161, 1-156. (19) Leo, A.; Mansch, C.; Elkins, D. Chem. Rev. 1971, 71, 525-616. (20) Burken, J. G.; Schnoor, J. L. Environ. Sci. Technol. 1998, 32, 3379-3385. (21) Briggs, G. G.; Bromilow, R.; Evans, A. A. Pestic. Sci. 1982, 13, 495-504. (22) Groom, C. A.; Beaudet, S.; Halasz, A.; Paquet, L.; Hawari, J. J. Chromatogr. 2001, 909, 53-60. (23) Schneider, J. F.; Zellmer, S. D.; Tomczyk, N. A.; Rastorfer, J. R.; Chen, D.; Banwart, W. L. Final Report SFIM-AEC-ET-CR-95013; U.S. Army Environmental Center: Gaithersburg, MD, 1995.

Received for review April 4, 2002. Revised manuscript received August 9, 2002. Accepted August 12, 2002. ES020673C

VOL. 36, NO. 21, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4655