ECD)

Talanta 72 (2007) 612–619

Extraction and determination of trace amounts of energetic compounds in blood by gas chromatography with electron capture detection (GC/ECD) Baohong Zhang ∗,1 , Xiaoping Pan 1 , Jordan N. Smith, Todd A. Anderson, George P. Cobb The Institute of Environmental and Human Health, and Department of Environmental Toxicology, Texas Tech University, Lubbock, TX 79409, USA Received 30 October 2006; received in revised form 16 November 2006; accepted 18 November 2006 Available online 21 December 2006

Abstract This paper describes an efficient and sensitive method for determining five energetic compounds at trace levels (ng/mL) in blood by gas chromatography with electron capture detection (GC/ECD). For seven test concentrations (1–1250 ng/mL), the average recoveries (%) were 104 ± 16, 108 ± 22, 105 ± 14, 100 ± 22 and 108 ± 16 for hexahydro-1,3,5-trinitroso-1,3,5-triazine (TNX), hexahydro-1,3-dinitroso-5-nitro-1,3,5-triazine (DNX), hexahydro-1-nitroso-3,5-dinitro-1,3,5-triazine (MNX), hexahydro-1,3,5-trinitro-1,3,5-triazacyclohexane (RDX) and 2,4,6-trinitrotoluene (TNT) (n = 84), respectively. Analysis of DNX and RDX produced lower precision than other energetic compounds. Acetonitrile extracts of blood samples should be analyzed immediately as the test compounds can transform into unknown compounds, which lowered the recovery by 0–45% within 10 days at room temperature (∼20 ◦ C). Maintaining sample extracts at 4 ◦ C decreased loss of test compounds. The method described herein was validated by different analysis teams on different days. Two-way ANOVA indicated that there was no significant difference between analysis teams or days of analysis. The method was successfully employed in the analysis of blood samples from a mouse dosing study involving TNX and RDX. © 2006 Elsevier B.V. All rights reserved. Keywords: Energetic compound; Explosive; Blood; Gas chromatography; Electron capture detection

1. Introduction In the past century, energetic compounds, such as 2,4,6-trinitrotoluene (TNT), hexahydro-1,3,5-trinitro-1,3,5triazacyclohexane (RDX), and octahydro-1,3,5,7-tetranitro1,3,5,7-tetraazacyclooctane (HMX), were produced as explosives and widely used for military and civil purposes around the world [1,2]. In some instances, these activities released energetic compounds into the environment, contaminating water and soil [1,3]. An estimated 12,000 sites across the U.S. are contaminated by energetic compounds; TNT, RDX and HMX are the most prevalent of these contaminants [2,4,5]. Laboratory and field studies indicate that these energetic compounds are toxic at relatively low concentrations to microorganisms [6–8], plants [9,10], invertebrates [4,5,11–14], birds [15], rats [16], and humans [17,18]. Energetic compounds

∗ 1

Corresponding author. Tel.: +1 806 885 4567; fax: +1 806 885 4577. E-mail address: [email protected] (B. Zhang). Co-first authors.

0039-9140/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2006.11.026

inhibit growth and development of organisms [19], cause seizures in humans [20–23] and rats [24], and may cause genotoxicity, cancer, or death [17,25]. To better understand the environmental and biological fate of energetic compounds and to better understand the potential health risks, sensitive analytical methods must be available to measure these compounds and their biotransformation products [1,3,26]. Although some methods have been developed for the determination of energetic compounds using HPLC [27] and GC, the majority of these methods were developed for analyzing explosive residues in water [28–36], soil [1,31,36–41] and plant samples [1]. Little information exists in the literature describing methods that can be used on animal tissues and biological fluids. This could hamper investigations on the effect of energetic compounds on animals [42,43]. Biological tissues and fluids are complex mixtures that contain many endogenous compounds (such as proteins, lipids) that can interfere with analyte determination and obstruct the separation and analysis of energetic compounds. Animals are usually indirectly exposed by drinking contaminated water or by eating contaminated food, so provided the contaminants

B. Zhang et al. / Talanta 72 (2007) 612–619

are usually not biomagnified, concentrations of energetic compounds and their biotransformation products should be lower than in the primary contamination source. Thus, more efficient methods are needed to detect energetic compounds and their biotransformation products in animal tissues and fluids for risk assessment. Unfortunately, analytical methods specifically developed for detecting energetic compounds and their biotransformation products in animal tissues are limited [44]. In this study, we describe a rapid and sensitive method to determine RDX, TNT, and three transformation products of RDX [45–47] in blood samples using gas chromatography with electron capture detection (GC/ECD). This method should facilitate risk assessments of explosives by allowing for the determination of exposure and distribution of explosive residues in humans and wildlife. 2. Experimental

613

by dilution of the stock standard solution with acetonitrile. All stock and working standard solutions were stored at 4 ◦ C prior to analysis. Working standard solutions were individually analyzed (described below) from low concentration to high. Each standard was analyzed three times and the average peak area was computed. A graph of analyte concentration versus GC–ECD response was constructed. The method detection limits (MDLs) of the analytes were determined in the test blood matrix described below, and calculated according to EPA guidelines using a 2 ng/mL standard of each analyte [48] and the following equation: MDL = standard deviation of seven replicates × Student’s t-value at 99% confidence level (t = 3.14 for n = 7).

2.1. Chemicals and reagents

2.3. Blood sample treatment and extraction procedure

TNT (CAS: 118-96-7), RDX (CAS: 121-82-4), MNX (CAS: 5755-27-1), DNX (CAS: 80251-29-2), and TNX (CAS: 13980-04-6) (Fig. 1) were analyzed in this study. RDX (purity > 99%) was purchased from Supelco (Bellafonte, PA). TNT (purity > 99.9%), MNX (purity = 98.4%), DNX (purity = 67%), and TNX (purity > 99.9%) were obtained from SRI International (Menlo Park, CA). HPLC-grade acetone and acetonitrile were purchased from Fisher Scientific (Pittsburg, PA). Ultra-pure water (>18 M) was prepared by a Barnstead NANOpure infinity ultrapure water system (Dubuque, IA). Glassware was washed with phosphate-free detergent and rinsed with deionized water, acetone, and acetonitrile.

The blood used for method development and validation was porcine blood, which was collected during ongoing slaughter operations that were unrelated to our study. Collected pig blood samples were placed into 50 mL centrifuge tubes containing heparin and stored at 4 ◦ C prior to treatment and analysis. A 1 mL aliquot of blood was placed into a 15 mL glass centrifuge tube and spiked with a standard solution of the test compound to final concentrations of 1, 5, 10, 20, 50, 250, and 1250 ng/mL. Spiked samples were mixed with a vortex-mixer for 1 min. Blank (untreated) blood samples were similarly prepared by amendment with the same volume of acetonitrile. Blank and spiked blood samples were stored overnight at 4 ◦ C before extraction and GC–ECD analysis. Liquid extraction coupled with sonication was employed for extracting explosives from blood samples. Briefly, 7 mL of acetonitrile was added to the 1 mL blood sample, followed by mixing with a vortex-mixer for 1 min. Samples were sonicated using an ultrasonic water bath (Branson, Danbury, CT) at 50 ◦ C. During sonication, the samples were mixed periodically with a vortex-mixer for 1 min. After liquid extraction for 2–3 h, the blood samples were centrifuged (3500 rpm) using a Beckman Allegra 6R centrifuge (Palo Alto, CA) for 10 min. The supernatants were collected and cleaned using Florisil solid-phase extraction (SPE) cartridges according to the following procedure: Florisil SPE cartridges were first placed on a 24-port manifold (Supelco, Bellafonte, PA) and were conditioned with acetonitrile (3 × 10 mL). Samples were then loaded, and eluates were collected. The Florisil cartridges were rinsed three times with acetonitrile (3 × 1 mL). The collected samples from SPE cartridges were concentrated to 1 mL under nitrogen using N-EVAPTM 111 nitrogen evaporator (Organomation Associates Inc., Berlin, MA), and filtered (0.2 ␮m) prior to GC analysis.

2.2. Standard solution, calibration curve, and method detection limits A stock standard solution for each of the test compounds was prepared in acetonitrile at a concentration of 1000 mg/L. Working standard solutions of each energetic compound (0, 1, 2, 5, 10, 20, 50, 100, 200, 500, 1000 and 1250 ng/mL) were made

2.4. Sample analysis by GC–ECD

Fig. 1. Structures of energetic compounds and their biotransformation products.

Analyses were performed using an HP 6890 gas chromatograph equipped with an HP 6890 autosampler and an electron

614

B. Zhang et al. / Talanta 72 (2007) 612–619

capture detector (Agilent, Palo Alto, CA). Separation was performed with a 30 m × 0.25 mm i.d. × 0.25 ␮m film thickness HP-5 column from Hewlett-Packard (Wilmington, DE). Helium (99.999% purity) served as carrier gas at a constant linear velocity of 80 cm/s. Argon:methane served as make-up gas for the detector. The oven temperature program began at 90 ◦ C, held for 2 min, increased to 130 ◦ C at a rate of 25 ◦ C/min, then made a 10 ◦ C/min ramp to 200 ◦ C, finally increased to 250 ◦ C at a rate of 25 ◦ C/min. The injection port temperature was 170 ◦ C, while the detector was 270 ◦ C. A 2 ␮L standard or sample was injected in splitless EPC mode. A splitless inlet liner (4.0 mm i.d.) with glass wool was used in this analysis. The septum and inlet liner were replaced after every 60 injections. The ECD was operated in constant current mode. A set of working standards was used to construct a calibration curve based on average peak area. One calibration standard (100 ng/mL) was also injected after every 10 samples to insure that the calibration was maintained; the response of that check standard was incorporated into the existing calibration curve. If the calibration check failed, a new calibration curve was constructed prior to any further analyses. Typically, more than 100 samples could be run (for all analytes except RDX) before a calibration failure occurred. This frequency of standard analysis ensured that analyte and detector stability were maintained during instrumental analysis. In each sample batch (n = 60), three blank (untreated blood) samples were also analyzed. The measurement of intra- and inter-day variability was used to determine the precision of the developed method. Precision estimates were based on relative standard deviation (R.S.D.) of the analyses. 2.5. Method validation, reproducibility, and stability test Two research teams were employed to validate and test the stability and reproducibility of the method during seven consecutive days. These tests were performed with pig blood samples spiked with RDX, MNX, TNX, DNX, or TNT (12 replicate samples at 50, 250, and 1250 ng/mL. 2.6. Application of the developed method The developed method was used to analyze blood samples from deer mice (Peromyscus maniculatus) obtained from an ongoing toxicity study with TNX and RDX. In the toxicity study, deer mice were dosed daily with 10 or 100 ng/mL TNX- or RDXcontaining water for more than 1 month. Deer mice blood sample extraction and analysis followed the protocol described above. 2.7. Statistical analysis Recovery data were processed using standard statistical software (SigmaPlot, Version 8.0, SPSS, Chicago, Illinois, USA). Two-way ANOVA was employed to compare potential differences among days and teams. A significance level of α = 0.05 was used in all comparative statistics.

2.8. Safety consideration RDX and TNT are explosives and can only be received in milligram quantities without permit. Thus, RDX, TNT, and their biotransformation products must be carefully handled. RDX, TNT, and RDX biotransformation products are also potentially toxic and carcinogenic compounds [42,43]. Personnel involved in work with explosives residues should wear protective gloves and goggles (ANSI Z 87.1-2003), especially when handling neat explosives. All waste solutions containing these energetic materials and their biotransformation products should be collected and discarded appropriately. 3. Results 3.1. Method development Each test compound was efficiently separated (baseline resolution) from endogenous compounds in hog blood and from each other (Fig. 2). The GC–ECD response versus concentration was best fitted to a quadratic model (y = ax2 + bx + c) with excellent correlation coefficients (>0.99) for each of the energetic compounds over the tested concentration range (1–1250 ng/mL) (Table 1). We also attempted to analyze HMX (another important explosive) using this method, however, the detector response was low. In addition, the low volatility of HMX precluded elution in a reasonable time without thermal degradation. Although several modifications to the method were attempted (inlet temperature, shorter column), significant progress was not made in HMX analysis by GC/ECD. The GC/ECD method was sensitive to TNX, DNX, TNT, MNX, and RDX; method detection limits (MDLs) in pig blood samples were 0.05, 0.10, 0.20, 0.25, and 0.50 ng/mL for TNX, DNX, TNT, MNX, and RDX, respectively. In addition, high recoveries of the analytes of interest were achieved at all of the concentrations tested (Table 2). However, different energetic compounds gave slightly different recoveries; the recovery range for DNX and RDX was wider than the other test compounds. At all tested concentrations, the recovery for MNX was 99–111%, while it was 97–114%, 97–125%, 98–130%, and 84–129% for TNX, TNT, DNX, and RDX, respectively. However, statistical analyses indicated that there was no significant difference among recoveries. This suggests that the method for extracting and analyzing explosive residues from blood was not analyteor concentration-dependent. Table 1 Parameters of the standard calibration curves for five analytes Analytes

a

b

c

r2

Concentration range (ng/mL)

TNX DNX TNT MNX RDX

0.2134 0.0064 0.1081 0.0056 0.0062

33.3 17.5 12.8 8.41 2.72

−2.78 −16.56 −11.25 −8.79 −1.67

0.9987 0.9989 0.9921 0.9985 0.9998

1–1250 1–1250 1–1250 1–1250 1–1250

The GC–ECD response versus concentration was best fit to a quadratic model (y = ax2 + bx + c) with excellent correlation coefficients.

B. Zhang et al. / Talanta 72 (2007) 612–619

615

Fig. 2. Representative GC/ECD chromatograms. (A) Standard solution (10 ng/mL) of a mixture of explosives; (B) blank blood sample; (C) pig blood sample spiked with a mixture (30 ng/mL) of explosives.

The precision ranged from 2.5 to 27% for all the analytes at all tested concentrations (1–1250 ng/mL). At relatively high concentrations (1250, 250, and 50 ng/mL), the precision ranged from 2.5 to 5.1, 3.8 to 7.7, 6.2 to 11.7, and 4.0 to 13.0 for TNX,

DNX, TNT, and MNX, respectively; the accuracy [accuracy (%) = (observed concentration−spiked concentration)/(spiked concentration) × 100] ranged from −2.7 to 2.2, −1.7 to 1.5, −2.9 to 6.4, −0.1 to 6.0, and −4.4 to 0.1 for TNX, DNX,

616

B. Zhang et al. / Talanta 72 (2007) 612–619

Table 2 Recovery and precision for energetic compounds in pig blood by GC–ECD (%) Compound

n

Recovery

Precision

Level: 1250 ng/mL TNX DNX TNT MNX RDX

12 12 12 12 12

99 102 106 100 100

± ± ± ± ±

2.5 4.4 8.7 4.0 3.0

2.48 4.31 8.15 4.04 3.30

Level: 250 ng/mL TNX DNX TNT MNX RDX

12 12 12 12 12

102 100 97 106 97

± ± ± ± ±

3.3 3.8 11 6.6 7.8

3.26 3.82 11.75 6.25 8.05

Level: 50 ng/mL TNX DNX TNT MNX RDX

12 12 12 12 12

97 98 98 105 96

± ± ± ± ±

5.0 7.6 6.1 14 23

5.14 7.69 6.16 13.02 23.98

Level: 20 ng/mL TNX DNX TNT MNX RDX

12 12 12 12 12

114 118 125 99 84

± ± ± ± ±

19 27 8.1 15 14

16.45 22.99 6.49 14.69 17.28

Level: 10 ng/mL TNX DNX TNT MNX RDX

12 12 12 12 12

110 115 113 102 96

± ± ± ± ±

19 34 14 14 19

17.20 29.34 12.03 13.22 19.89

Level: 5 ng/mL TNX DNX TNT MNX RDX

12 12 12 12 12

108 113 116 111 123

± ± ± ± ±

29 30 7.6 22 33

26.68 26.99 6.54 19.47 26.68

Level: 1 ng/mL TNX DNX TNT MNX RDX

12 12 12 12 12

106 130 114 107 129

± ± ± ± ±

26 29 31 17 35

24.88 22.17 26.69 15.93 27.02

Average TNX DNX TNT MNX RDX

84 84 84 84 84

104 108 108 105 100

± ± ± ± ±

16 22 16 14 22

15.14 20.07 14.87 13.10 21.69

Table 3 Method validation by different research teams on different days using pig blood samples spiked with 50, 250, and 1250 ng/mL of each test compound (recovery %) Compound

Day I

Day II

Team I

Team II

Team I

Team II

Level: 1250 ng/mL TNX 102 DNX 104 TNT 104 MNX 103 RDX 98

± ± ± ± ±

4.9 9.2 9.3 9.8 3.7

105 98 112 108 105

± ± ± ± ±

6.4 19 14 13 8.4

98 105 114 105 97

± ± ± ± ±

4.1 0.7 2.0 3.7 12

96 101 106 103 96

± ± ± ± ±

0.9 3.2 5.0 4.5 7.0

Level: 250 ng/mL TNX 100 DNX 100 TNT 95 MNX 110 RDX 101

± ± ± ± ±

2.6 19 23 13 22

104 101 93 106 90

± ± ± ± ±

15 4.4 9.7 19 15

106 102 100 109 106

± ± ± ± ±

0.8 4.2 13 5.8 13

107 101 96 110 94

± ± ± ± ±

9.5 2.6 13 3.9 8.0

Level: 50 ng/mL TNX DNX TNT MNX RDX

± ± ± ± ±

5.7 12 3.5 21 19

94 105 100 101 91

± ± ± ± ±

9.1 16 14 24 32

99 95 97 100 97

± ± ± ± ±

2.4 14 2.8 17 29

101 99 99 108 96

± ± ± ± ±

15 16 6.6 9.6 16

95 96 96 106 96

Six replicates for each data group.

3.2. Method validation

Precision was expressed by relative standard deviation (R.S.D.) (%). R.S.D. (%) = (S.D./mean) × 100%.

TNT, MNX, and RDX, respectively. Precision and accuracy increased slightly as concentration increased for most energetic compounds. Precision and accuracy for DNX and RDX were the lowest of the tested compounds possibly due to the instability of DNX and the relatively high detection limit for RDX. However, we considered both the precision and accuracy for this type of analysis in a blood matrix to be acceptable.

An interlaboratory study was employed to determine the stability and reproducibility of the developed extraction and analysis method using pig blood samples spiked with RDX, TNX, DNX, MNX, and TNT at three concentrations (50, 250, and 1250 ng/mL). Each treatment had high recovery (91–114%) (Table 3). Two way ANOVA indicated no significant difference between analysis teams or days of analysis (p < 0.001). 3.3. Application of the developed extraction and analysis method The developed method was employed to analyze blood samples obtained from TNX or RDX-exposed deer mice. We did not detect TNX in blood samples from the control group (n = 4) (Fig. 3A). Trace amounts of TNX (0.14–0.53 ng/mL) were detected in the low dose group (n = 9). Relatively high concentrations of TNX (0.63–43 ng/mL) were detected in the high dose group (n = 5) (Table 4). Similar results were obtained from blood samples collected in the RDX-dosing study (Fig. 3C). These Table 4 Measured TNX concentrations in blood samples obtained from deer mice exposed to 0, 10, and 100 ng/mL TNX for 30 days Dose (ng/mL)

Number of deer mice

TNX concentration in blood (mean) (ng/mL)

0 10 100

4 9 5

N.D. 0.14–0.53 (0.29) 0.63–42.99 (13.35)

N.D.: not detected.

B. Zhang et al. / Talanta 72 (2007) 612–619

617

Fig. 3. Representative chromatograms of blood samples obtained from TNX or RDX-dosed deer mice. (A) Control; (B) TNX-treated; (C) RDX-treated. The slight retention time shift compared to the previous figure was due to the use of a different analytical column. The peak tailing was caused by the overused column.

data indicate that this method can be used in basic toxicology and toxiokinetic studies. 3.4. Stability of energetic compounds in extracted solution The stability of each energetic compound (RDX and TNT) and RDX biotransformation products (MNX, DNX, and TNX) in acetonitrile extracts (n = 3) was monitored for 10 days using

GC/ECD analysis. All energetic compounds and RDX biotransformation products were unstable in the acetonitrile extracts at room temperature. After 10 days, 15–40% of the extracted compounds were lost (transformed into other unknown compounds) (Fig. 4). RDX and MNX easily degraded into other compounds. After 10 days in the extract, 64 and 60% of RDX and MNX were still present. TNT and TNX were more stable than other energetic compounds tested; only 16 and 22% of parent compound

618

B. Zhang et al. / Talanta 72 (2007) 612–619

Fig. 4. Stability of energetic compounds and their transformation products in acetonitrile extracts of blood. The Y-axis was calculated as the ratio of the concentration of energetic compounds to the concentration at time 0. Error bars indicate the standard deviation for triplicate samples.

was lost in extracts after 10 days. Analyte concentrations only slightly affected the degradation rate of energetic compounds in acetonitrile extracts. Low temperature, such as maintaining extracts in a refrigerator (4 ◦ C), slowed the transformations (data not shown). RDX can be biotransformed to MNX, DNX, and TNX by some bacteria under anaerobic conditions [45–47]. However, no studies have found that RDX can be transformed to TNX under aerobic conditions. TNX, DNX, MNX or any other identifiable compound was not observed in extracts of RDX-spiked blood samples stored at room temperature for 10 days, although 45% of RDX was lost (degraded) during that time; the transformation pathway(s) remains unclear.

ment in environmental laboratories. Determination of explosive residues using GC–ECD is advantageous due to the lower detection limits and improved chromatographic resolution. Because of its low vapor pressure and thermal lability, RDX is difficult to quantify by gas chromatography. Thus, HPLC is used in the majority of methods for detecting RDX. Our study showed that injection port temperature is important for determining RDX and its biotransformation products using GC–ECD. The optimal injection port temperature was 160–170 ◦ C. Higher or lower inlet temperatures significantly reduced GC–ECD response to RDX and its N-nitroso-metabolites [49].

4. Discussion

A simple, efficient, and sensitive GC–ECD method for determination of explosives or their biotransformation products in blood samples was developed with detection limits <0.5 ng/mL. This method gave high recovery, precision, and accuracy within a concentration range of 1–1250 ng/mL. This method was validated by different analysis teams on different days and successfully employed to monitor explosive residues in blood samples from TNX- and RDX-exposed deer mice.

Several laboratory and field studies have indicated that RDX can be biotransformed into a series of N-nitroso metabolites (MNX, DNX, and TNX), which have also been detected in groundwater [28]. To better understand the potential health effects and environmental fate of these energetic compounds, sensitive analytical methods must be available to measure these compounds and their biotransformation products [1,3,26]. Based on our best knowledge, there has been only one report on analytical methods for detecting explosive residues in blood samples. Ozhan et al. (2004) developed an HPLC–UV method for detecting RDX in human blood. However, the method was only capable of detecting 5 ng/mL RDX in human plasma based on 3:1 signal-to-noise ratio [44]. GC–ECD is a common instru-

5. Conclusions

Acknowledgments The authors would like to thank Dr. J.M. Hellman of the Animal Care and Use Committee, Texas Tech University for kindly providing pig blood samples. This work is funded by the U.S. Department of Defense contract CU1235, through the

B. Zhang et al. / Talanta 72 (2007) 612–619

Strategic Environmental Research and Development Program (SERDP). An EPA STAR fellowship supported J.N.S. References [1] A. Halasz, C. Groom, E. Zhou, L. Paquet, C. Beaulieu, S. Deschamps, A. Corriveau, S. Thiboutot, G. Ampleman, A. Dubois, et al., J. Chromatogr. A 963 (2002) 411. [2] S.S. Talmage, D.M. Opresko, C.J. Maxwell, C.J. Welsh, F.M. Cretella, P.H. Reno, F.B. Daniel, Rev. Environ. Contam. Toxicol. 161 (1999) 1. [3] J.I. Steinfield, J. Wormhoudt, Annu. Rev. Phys. Chem. 49 (1988) 203. [4] P.Y. Robidoux, C. Dubois, J. Hawari, G.I. Sunahara, Ecotoxicology 13 (2004) 603. [5] P.Y. Robidoux, G.I. Sunahara, K. Savard, Y. Berthelot, S. Dodard, M. Martel, P. Gong, J. Hawari, Environ. Toxicol. Chem. 23 (2004) 1026. [6] D. Juck, B.T. Driscoll, T.C. Charles, C.W. Greer, FEMS Microbiol. Ecol. 43 (2003) 255. [7] R.J. Spanggord, K.R. Stewart, E.S. Riccio, Mutat. Res. 335 (1995) 207. [8] G.I. Sunahara, S. Dodard, M. Sarrazin, L. Paquet, G. Ampleman, S. Thiboutot, J. Hawari, A.Y. Renoux, Ecotoxicol. Environ. Saf. 39 (1998) 185. [9] A.J. Palazzo, D.C. Leggett, J. Environ. Qual. 15 (1986) 49. [10] L.E. Winfield, J.H. Rodgers Jr., S.J. D’Surney, Ecotoxicology 13 (2004) 335. [11] R.G. Kuperman, R.T. Checkai, M. Simini, C.T. Phillips, J.E. Kolakowski, C.W. Kurnas, G.I. Sunahara, Pedobiologia 47 (2003) 651. [12] P.Y. Robidoux, J. Hawari, G. Bardai, L. Paquet, G. Ampleman, S. Thiboutot, G.I. Sunahara, Arch. Environ. Contam. Toxicol. 43 (2002) 379. [13] B.H. Zhang, C.M. Freitag, J.E. Canas, Q.Q. Cheng, T.A. Anderson, Environ. Pollut. 144 (2006) 540. [14] P.Y. Robidoux, C. Svendsen, J. Caumartin, J. Hawari, G. Ampleman, S. Thiboutot, J.M. Weeks, G.I. Sunahara, Environ. Toxicol. Chem. 19 (2000) 1764. [15] R.M. Gogal, M.S. Johnson, C.T. Larsen, M.R. Prater, R.B. Duncan, D.L. Ward, R.B. Lee, C.J. Salice, B. Jortner, S.D. Holladay, Environ. Toxicol. Chem. 22 (2003) 381. [16] S. Homma-Takeda, Y. Hiraku, Y. Ohkuma, S. Oikawa, M. Murata, K. Ogawa, T. Iwamuro, S. Li, G.F. Sun, A. Kumagai, et al., Free Radic. Res. 36 (2002) 555. [17] B. Lachance, P.Y. Robidoux, J. Hawari, G. Ampleman, S. Thiboutot, G.I. Sunahara, Mutat. Res. 444 (1999) 25. [18] D. Bruns-Nagel, S. Scheffer, B. Casper, H. Garn, O. Drzyzga, E.V. Low, D. Gemsa, Environ. Sci. Technol. 33 (1999) 2566. [19] J.A. Steevens, B.M. Duke, G.R. Lotufo, T.S. Bridges, Environ. Toxicol. Chem. 21 (2002) 1475. [20] Y. Kucukardali, H.V. Acar, S. Ozkan, S. Nalbant, Y. Yazgan, E.M. Atasoyu, O. Keskin, A. Naz, N. Akyatan, A. Gokben, et al., J. Intensive Care Med. 18 (2003) 42.

619

[21] L.J. Burdette, L.L. Cook, R.S. Dyer, Toxicol. Appl. Pharmacol. 92 (1988) 436. [22] D.J. Goldberg, S.T. Green, D. Nathwani, J. McMenamin, N. Hamlet, D.H. Kennedy, J. R. Soc. Med. 85 (1992) 181. [23] B. Harrell-Bruder, K.L. Hutchins, Ann. Emerg. Med. 26 (1995) 746. [24] S.A. Meyer, A.J. Marchand, J.L. Hight, G.H. Roberts, L.B. Escalon, L.S. Inouye, D.K. MacMillan, J. Appl. Toxicol. 25 (2005) 427. [25] M.E. Honeycutt, A.S. Jarvis, V.A. McFarland, Ecotoxicol. Environ. Saf. 35 (1996) 282. [26] D.S. Moore, Rev. Sci. Instruments 75 (2004) 2499. [27] USEPA, SW846 Method 8330, Nitroaromatics and Nitramines by High Performance Liquid Chromatography [HPLC], US EPA, Washington, DC, 1994. [28] H.R. Beller, K. Tiemeier, Environ. Sci. Technol. 36 (2002) 2060. [29] T.L. Buxton, P.D. Harrington, Appl. Spectrosc. 57 (2003) 223. [30] T.M. Chow, M.R. Wilcoxon, M.D. Piwoni, N.R. Adrian, J. Chromatogr. Sci. 42 (2004) 470. [31] A. Hilmi, J.H. Luong, A.L. Nguyen, J. Chromatogr. 844 (1999) 97. [32] F. Monteil-Rivera, C. Beaulieu, S. Deschamps, L. Paquet, J. Hawari, J. Chromatogr. A 1048 (2004) 213. [33] F. Monteil-Rivera, C. Beaulieu, J. Hawari, J. Chromatogr. 1066 (2005) 177. [34] M.E. Walsh, T.F. Jenkins, Anal. Chim. Acta 231 (1990) 313. [35] J. Yinon, J. Chromatogr. A 742 (1996) 205. [36] R.W. Bishop, M.A. Hable, C.G. Oliver, R.J. Valis, USACHPPM, J. Chromatogr. Sci. 41 (2003) 73. [37] M.E. Walsh, Talanta 54 (2001) 427. [38] S. Campbell, R. Ogoshi, G. Uehara, Q.X. Li, J. Chromatogr. Sci. 41 (2003) 284. [39] C.A. Groom, S. Beaudet, A. Halasz, L. Paquet, J. Hawari, J. Chromatogr. 909 (2001) 53. [40] B.H. Zhang, X.P. Pan, G.P. Cobb, T.A. Anderson, J. Chromatogr. B 824 (2005) 277. [41] X.P. Pan, B.H. Zhang, S.B. Cox, T.A. Anderson, G.P. Cobb, J. Chromatogr. A 1107 (2006) 2. [42] ATSDR, Toxicological Profile for RDX, US Department of Health and Services. http://www.atsdr.cdc/gov/toxprofiles/tp78.pdf, 1995. [43] ATSDR, Toxicological Profile for HMX, Agency for Toxic Substances and Disease Registry, US Department of Health and Human Services. Atlanta, GA, 1997. [44] G. Ozhan, S. Topuz, B. Alpertunga, Farmaco (Societa Chimica Italiana: 1989) 58 (2003) 445. [45] N.R. Adrian, C.M. Arnett, Curr. Microbiol. 48 (2004) 332. [46] J. Hawari, A. Halasz, T. Sheremata, S. Beaudet, C. Groom, L. Paquet, C. Rhofir, G. Ampleman, S. Thiboutot, Appl. Environ. Microbiol. 66 (2000) 2652. [47] J. Hawari, S. Beaudet, A. Halasz, S. Thiboutot, G. Ampleman, Appl. Microbiol. Biotechnol. 54 (2000) 605. [48] EPA, SW846 Test Methods, US EPA, Washington, DC, 2000. [49] X.P. Pan, B.H. Zhang, G.P. Cobb, Talanta 67 (2005) 816.