Comprehensive Non-Targeted Analysis of Contaminated Groundwater

Aug 14, 2009 - Germany and the Office of Water and Waste Management. (AWA), Water and Soil Protection Laboratory (WSPL),. Schermenweg 11, 3014 Bern, ...
0 downloads 0 Views 3MB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 7055–7061

Comprehensive Non-Targeted Analysis of Contaminated Groundwater of a Former Ammunition Destruction Site using 1H-NMR and HPLC-SPE-NMR/TOF-MS M A R K U S G O D E J O H A N N , †,* L E A H E I N T Z , † CRISTINA DAOLIO,† JEAN-DANIEL BERSET,‡ AND DANIEL MUFF‡ Bruker BioSpin GmbH, Silberstreifen 4, D-76287 Rheinstetten, Germany and the Office of Water and Waste Management (AWA), Water and Soil Protection Laboratory (WSPL), Schermenweg 11, 3014 Bern, Switzerland

Received April 8, 2009. Revised manuscript received June 30, 2009. Accepted July 27, 2009.

The aim of the present study was to explore the capabilities of the combination of 1H NMR (proton nuclear magnetic resonance) mixture analysis and HPLC-SPE-NMR/TOF-MS (highperformance liquid chromatography coupled to solid-phase extraction and nuclear magnetic resonance and time-of-flight mass spectrometry) for the characterization of xenobiotic contaminants in groundwater samples. As an example, solidphase extracts of two groundwater samples taken from a former ammunition destruction site in Switzerland were investigated. 1H NMR spectra of postcolumn SPE enriched compounds, together with accurate mass measurements, allowed the structural elucidation of unknowns. This untargeted approach allowed us to identify expected residues of explosives such as 2,4,6-trinitrotoluene (2,4,6-TNT), Hexogen (RDX) and Octogen (HMX), degradation products of TNT (1,3,5trinitrobenzene (1,3,5-TNB), 2-amino-4,6-dinitrotoluene (2-A4,6-DNT), 3,5-dinitrophenol (3,5-DNP), 3,5-dinitroaniline (3,5-DNA), 2,6-dinitroanthranile, and 2-Hydroxy-4,6-dinitrobenzonitrile), benzoic acid, Bisphenol A (a known endocrine disruptor compound), and some toxicologically relevant additives for propelling charges: Centralite I (1,3-diethyl-1,3-diphenylurea), DPU (N,N-diphenylurethane), N,N-diphenylcarbamate (Acardite II), and N-methyl-N-phenylurethane. To our knowledge, this is the first report of the presence of these additives in environmental samples. Extraction recoveries for Centralite I and DPU have been determined. Contaminants identified by our techniques were quantified based on HPLC-UV (HPLCultraviolet detection) and 1H NMR mixture analysis. The concentrations of the contaminants ranged between 0.1 and 48 µg/L: assuming 100% recovery for the SPE step.

* Corresponding author phone: ++4972151616313; fax: ++4972151616297; E-mail: [email protected]. † Bruker BioSpin GmbH. ‡ Office of Water and Waste Management (AWA), Water and Soil Protection Laboratory (WSPL). 10.1021/es901068d CCC: $40.75

Published on Web 08/14/2009

 2009 American Chemical Society

Introduction In Switzerland, the Susten-Stone Glacier (Steingletscher) used to be an important ammunition disposal site where tons of ammunition were stored and destroyed. Previously, old ammunition had been disposed of in several lakes (1). In 1992 over 300 tons of ammunition were stored in a cavern at Susten-Steingletscher. In November that year a huge detonation occurred and all of the ammunition was destroyed. Six people were killed and the reasons for the uncontrolled detonation still remain unknown. After having disposed of 5600 tons of ammunition, activities were finally stopped at Susten-Steingletscher in 1997. It is well-known that such sites remain highly polluted with explosives, frequently with 2,4,6-trinitrotoluene (TNT) and its metabolites, as well as with heavy metals and other not yet identified compounds (2–4) and might contaminate nearby waters. The presence of such compounds is of environmental concern as they exhibit genotoxic and carcinogenic properties (5, 6). Toxicological guidelines have been established in the EU to protect the drinking water supply (7). For this reason, systematic analysis of groundwater samples in the vicinity of former ammunition destruction sites is of great concern. High-performance liquid chromatography (HPLC) coupled to tandem mass spectrometry (MS/MS) using atmospheric pressure ionization (API) turns out to be the method of choice for targeted detection and quantification of explosive residues (8–12). As well as the frequently used electrospray ionization (ESI) (3), atmospheric pressure chemical ionization (APCI) and atmospheric pressure photoionization (APPI) have been used for determining trace levels of explosives and related compounds in water (13). Recently, TOF and hybrid quadrupole-TOF (Q-TOF) mass spectrometers have emerged as powerful detection systems in the elucidation of unknown contaminants in environmental samples (14, 15). Contrary to triple quadrupole instruments which provide highest sensitivity in multiple reaction monitoring mode rather than in full scan mode, time-of-flight mass spectrometers are mostly used in a nontargeted approach: they are able to scan the whole mass range with high mass accuracy and high sensitivity. The high mass accuracy of these instruments allows the calculation of empirical formulas of the unknown analytes and of their fragments if Q-TOF/MS experiments are done. Due to these properties, TOF instruments have been used successfully to confirm the presence of microcontaminants such as pharmaceuticals, pesticides, explosives, etc. (16). Nuclear magnetic resonance gives information complementary to that obtained from mass spectrometry. NMR is usually used for the structural elucidation of organic compounds. Parallel to this qualitative information, the 1H NMR spectrum of a sample yields a quantitative overview of proton carrying constituents (17–19) without the need of the existence of analytical reference standards. The hyphenation between chromatographic techniques, NMR and MS helps to overcome the problem of superimposition of NMR resonance lines in complex mixtures and enables the fast structure elucidation of complete unknowns. Typical fields of application are found in pharmaceutical or natural product analysis (20–22). The coupling of HPLC with NMR alone and with MS in environmental analysis has already been reported for structure confirmation, elucidation, and quantification of explosives and other contaminants together with their degradation products (23–27). LC-SPE-NMR is a relatively new concept, enabling the NMR detection of single chromatographic peaks trapped on online SPE cartridges postcolumn to an HPLC separation. During the trapping experiVOL. 43, NO. 18, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7055

FIGURE 1. 1H NMR spectra of SPE extracts reconstituted in acetonitrile-d3. The upper spectrum originates from the extract of sample 1, the lower spectrum is from the extract of sample 2. The integrals and chemical shifts of the NMR signals allow an estimation of the concentration and of the chemical class of compounds present in environmental samples. Protons from nitroaromatic compounds typically resonate above 8 ppm. ment the effluent from the chromatography is diluted with an aqueous phase to increase the retention of the analytes on the stationary phase. Each time a peak is detected by UV or MS, a set of valves direct the chromatographic flow into an SPE cartridge. After the chromatography is finished, the cartridges are dried with a stream of nitrogen gas to remove residual protonated solvent which would produce huge signals in the NMR spectrum and hence obscure the resonance signals originating from the contaminants. Elution of the trapped compounds is done using deuterated organic solvents with very low amounts of residual water in the sample. The use of cryogenically cooled NMR probe heads increases the signal-to-noise of the NMR experiment by a factor of 4. In combination with the SPE approach discussed above, the amount of analyte needed to get the structure elucidated decreases drastically when compared to traditional hyphenated techniques (28, 29). Integration of a high-resolution time-of-flight instrument (by splitting the chromatographic flow between the mass spectrometer and the SPE device in a 5:95 ratio) enables additional structural evidence to be found. The empirical formula calculated from the mass spectrum can be unequivocally correlated to the NMR spectrum acquired after elution of the peak of interest. For small molecules in particular this is sufficient to come up with a tentative structure proposal which can easily be validated by comparison with reference standards or chemical shift values documented in the literature.

Experimental Section Chemicals and materials used, site description and sampling, water samples, solid-phase extraction procedure, 1H NMR mixture analysis, liquid chromatography-SPE-1H NMR spectroscopy/TOF-mass spectroscopy, and 1H NMR analysis of individual peaks are described in the Supporting Information. 7056

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 18, 2009

Results and Discussion Qualitative Analysis of SPE Extracts by 1H NMR. After reconstitution of SPE extracts of groundwater samples (Sample 1 was obtained in July 2008 and sample 2 was obtained in October 2008) in 650 µL of deuterated acetonitrile, the 1H NMR mixture spectra revealed the presence of aromatic compounds as shown in Figure 1. Resonance lines with chemical shift values above 8 ppm can originate from nitroaromatic compounds. Singlets at 9 ppm (and 2.7 ppm for the methyl group) and 9.35 ppm are due to the presence of 2,4,6-TNT and 1,3,5-TNB, respectively. The equivalent protons for the explosives RDX and HMX resonate around 6 ppm. In addition to these resonance lines typically observed with well-known contaminants from former ammunition sites (17), significant NMR signals in the high field region of aromatic protons give an expectation of contaminants carrying aromatic moieties lacking an electronwithdrawing group. Between 4.4 and 3.4 ppm a number of quartets can be identified. They belong to X-CH-CH3 groups with X ) O, N, S together with triplets around 1.1 ppm. Singlets between 2 and 3 ppm originate from methyl groups attached to an aromatic ring. Structure Elucidation of Contaminants by HPLC-SPENMR/TOF-MS. In contrast to mass spectrometry, NMR is a nondestructive spectroscopic technique. This allows the sample mixture to be injected into HPLC columns after NMR measurement without additional preparation. Further investigations of the samples employed HPLC-SPE-NMR/TOFMS. The HPLC system is coupled to an automated SPE device which allows trapping of peaks after chromatographic separation. This is followed by elution into an NMR probe head using deuterated organic solvents. For the current investigations the postcolumn peak trapping process was carried out dependent on the response of the diode array detector at 254 nm because the results of the NMR inves-

FIGURE 2. HPLC-UV (top) at 220 nm and HPLC-MS (bottom) base peak chromatogram in negative ionization mode of the groundwater extract from sample 1. Dotted vertical lines (1-18) indicate the start and end of the trapping process on SPE cartridges. The peak at 54 min represents 2,4,6-TNT. tigation revealed the presence of aromatic compounds in the sample. Figure 2 shows the resulting chromatograms indicating 18 peaks trapped on SPE cartridges. Inspection of the NMR spectra obtained after the elution of the 18 cartridges into the NMR probe head yielded 16 spectra showing resonance lines not originating from chemical background, with interpretable signal-to-noise. Figure 3 shows the resulting NMR spectra acquired in methanol-d3. The comparison of these spectra to the mixture spectrum shown in Figure 1 proves that all contaminants present in significant concentration have been captured and measured by this approach. Due to the direct coupling of the HPLC-UV to a timeof-flight mass spectrometer, the MS response in both positive and negative ionization mode could be obtained for most of the peaks trapped. Experimental mass spectra, together with theoretical patterns for all contaminants identified during this study, are displayed in Figure S1. These complementary spectroscopic data usually suffice to come up with a proposed structure. The analytical strategy first requires an NMR spectrum of a trapped peak, where protons of the contaminant can be detected with sufficient signal-to-noise ratio, to enable the evaluation of spectroscopic information, e.g., number of protons, chemical shift values, and spin-spin coupling constants. Comparison to the mixture 1H NMR spectrum of the extract allows an estimation of the concentration level of the individual compound. The second step is the extraction of high-resolution mass spectroscopic data from the original chromatogram. This step is critical as the response of an analyte in the MS strongly depends on its ionization efficiency in the ion source, in contrast to the NMR response. To get the MS response of unknown analytes, each sample has to be measured in both positive and negative ionization mode as some compounds might only be detectable in one mode. If more than one mass-to-charge ratio can be extracted from the chromatogram where the peak was trapped, it is important to calculate the sum formula for each potential mass and to compare the number of protons with those obtained from the NMR spectrum. Only compounds with exact mass information and corresponding

NMR spectrum were considered to be tentatively identified in this study. This is why only 15 out of 18 peaks trapped on cartridges and eluted to the NMR could be assigned to a proposed structure. This takes into account that some peaks are composed of more than one compound eluting from the column. Once the contaminant is tentatively identified, its commercial availability as a pure chemical needs to be evaluated. To unequivocally identify the contaminant, a simple proton NMR spectrum of the reference standard diluted in acetonitrile-d3 needs to be recorded and compared to the spectrum of the groundwater extract as shown for Centralite I and diphenylurethane in Figures S2 and S3. Some contaminants identified have been described already in the literature (13, 14), including NMR spectroscopic information. For 4,6-dinitroanthranile, 3,5-dinitrophenol, 3,5-dinitroaniline, N-methyl-N-phenylurethane, and N,N-diphenylcarbamate, structures were tentatively identified solely based on MS and NMR spectroscopic information. The presence of these contaminants would need to be confirmed by comparison with reference standards. Table 1 compiles the structures not related to nitroaromatic compounds that were identified using this approach. Diethyl-N,N′-diphenylurea, also known as Ethyl Centralite or Centralite I, is a compound used as a stabilizer in gunpowder and military nitrocellulose. Bisphenol A is also known to be added to ammunition to stabilize the propellant charge. Quantitative Determination of Contaminants Present in the Extracts. Besides the qualitative information, the NMR spectrum inherently contains comprehensive quantitative data. The integral for the proton signal of one mol of protons remains constant regardless of the multiplicity of the resonance and the chemical class of the contaminant detected by NMR. If one compound in the mixture is quantified externally by validated methods, e.g., LC-UV or LC-MS using reference material for calibration, each contaminant identified in the mixture showing an NMR signal free of superimposition by other signals can be quantified without the need for reference material (17). VOL. 43, NO. 18, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7057

FIGURE 3. 1H NMR spectra of peaks trapped and eluted from SPE cartridges to the NMR probe head with methanol-d3. The spectra from bottom to top compared to the peaks trapped shown in the chromatogram displayed in Figure 2. Solvent signals from methanol at δ ) 3.3 ppm and from water at δ ) 4.7 ppm were removed by double solvent suppression. Moreover, the quantification of one known component of the mixture can also be performed externally provided the concentration of the external standard is known and the NMR parameters are kept identical. In this case, the integrals obtained from the reference standard of known concentration can be compared to the integral of the same resonance line detected in the extract. The NMR quantification method is well-known to provide accurate and precise results. A detailed validation of this method, using nitroaromatic compounds as an example, is described elsewhere (19). Centralite I was chosen as the reference standard for NMR quantification of all contaminants identified (Figures S4 and S5). Recovery values for Centralite I together with diphenylurethane were experimentally obtained at two spiking levels of 10 and 1 µg/L as no data were available from the literature. Recovery values for both analytes were not significantly different from 100%, as shown in Table S1. In addition, Figure S6 proves that no background is present for these analytes when a blank cartridge is prepared and measured using the same parameters as for the spiked samples. Quantification using LC-UV at 254 nm with external calibration was compared to the results obtained by the external NMR quantification as described above. From the LC-UV method, a concentration of 140 mg/L of Centralite I was obtained for the extract of sample 1 using an external 10 points calibration. For the extract of sample 2, the concentration calculated was 60 mg/L. Taking into account the enrichment factor for the SPE step of 3077 for the extract 1 and 1538 for the extract 2, the original concentration in the groundwater could be calculated to be 47 µg/L for sample 1 and 39 µg/L for sample 2, assuming 100% recovery, which is in good agreement with recovery data given in the literature for explosives and related compounds (19) and with our experimental results obtained for Centralite I and DPU. The 7058

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 18, 2009

enrichment factor was calculated as the ratio of the sample volume before SPE extraction, to the volume of the extract after reconstitution with deuterated acetonitrile. The external calibration using the NMR approach yielded nearly the same quantitative results as the LC-UV method. A concentration of 48 µg/L for sample 1 and 35 µg/L for sample 2 was determined with minimized time consumption for the preparation of the stock standard in deuterated solvent and the measurement of the proton reference experiment. The concentrations obtained for all contaminants identified using this approach are listed in Table 2. Environmental Significance. The results obtained during this study of military hot spot samples revealed not only high concentrations of traditional explosive residues such as TNT, RDX, HMX, and corresponding metabolites of TNT, but also elevated amounts of stabilizers of powders and propellants. These constituents were introduced in formulations to catch nitrogen-oxides formed during degradation of explosives and therefore to prevent chemical changes of the energetic compounds over a reasonable period of time. To our knowledge, very little is known about the toxicity of these molecules. Among them, mainly diphenylamine (DPA) and its nitrodiphenylamine transformation products have been found at former military sites (30, 31). N,N-dimethyl-N,Ndiphenylurea (Centralite II), a similar compound to N,Ndiethyl-N,N-diphenylurea (Centralite I) found in the extracts of the Susten-Stone Glacier, was detected in gunshot residues (32). The remaining stabilizers N,N-diphenylurethane (DPU), N,N-diphenylcarbamate, and N-methyl-N-phenylurethane have not previously been reported as occurring at such sites. Several databases were searched for eco-toxicity data of these compounds (33–36). Data sets were found only for the structurally similar compound N,N-diphenylamine (DPA). EC50 (half maximal effective concentration) values ranged between 0.05 and 2 mg/L for algae and 2 and 5 mg/L for fish,

TABLE 1. Compilation of Structures for Contaminants Not Related to Nitroaromatic Compounds That Were Identified in the Groundwater Extracts of the Former Ammunition Destruction Site

TABLE 2. Qualitative and Quantitative Results for Compounds Identified in Groundwater Extracts of the Former Ammunition Disposal Site (Note that Hexogen and Octogen are Ionizing As Chlorine Adducts in Negative Ionization Mode) t Ra m/z, m/z, peak (min) z ) -1 z ) +1

formula

1 2 3 4 5 6 6 7

39.1 41.1 42.7 44.6 45.5 46 46 47.5

121.031 182.019 223.996 223.997 257.003 331.014

8

47.8

207.999

C7H2N3O5

8.76 (d), 8.23 (d) 6.08 (s) 6.03 (s) 7.39 (t), 7.26 (m), 2.74 (s) 8.02 (d), 7.87 (d), 4.37 (m), 3.82 (t) 8.5 (d), 8.12 (d)

8 9 10 11 12 13 14

47.8 48.3 49.2 49.4 49.8 51.2 51.4

183.004

C6H3N2O5

8.47 (t), 7.99 (d)

182.019 227.106

C6H4N3O4 C15H15O2

8.11 (t), 7.79 (d) 7.04 (m), 6.67 (m), 1.59 (s) 9.35 (s) 7.8 (d), 7.74 (d), 2.27 (s) 7.38 (t), 7.27 (m), 4.15 (q), 2.29 (s), 1.24 (t)

15 16 17

52.8 54.1 60

18

61.4

227.119

196.035

C 7H 5O 2 C6H4N3O4 C7H2N3O6 C7H2N3O7 C3H6ClN6O6 C4H8ClN8O8 C14H15N2O

NMR signals (ppm)b

180.102

C7H6N3O4 C10H14NO2

242.123

C7H4N3O6 C15H16NO2

269.166

C17H21N2O

226.01

8.04 (d), 7.61 (t), 7.49 (t)

8.98 (s), 2.66 (s) 7.37 (t), 7.27 (m), 4.2 (q), 1.23 (t) 7.04 (t), 6.95 (t), 6.75 (t), 3.63 (q), 1.12 (t)

conc. conc. sample 1 sample 2 c (µg/L) (µg/L)d

compound

confirmation

benzoic acid unknown unknown 4,6-dinitroanthranile Hexogen (RDX) Octogen (HMX) N,N-diphenylcarbamate unknown

standard, 1H NMR

6.2

0.6

literature (17) literature (17)

0.3 6.0 2.3 0.1

0.4 5.4 1.4 0.2

2-hydroxy-4,6dinitrobenzonitrile 3,5-DNP unknown 3,5-DNA Bisphenol A 1,3,5-TNB 2-A-4,6-DNT N-methyl-Nphenylurethane unknown 2,4,6-TNT N,N-diphenylurethane

literature (18)

1.9

1.1

0.3

0.2

standard, 1H NMR literature (17) literature (17)

3.4 13 5.1 2.6 1.5

3.1 12 4.1 2.9 1.7

literature (17) standard, 1H NMR

47 16

61 18

Centralite I

standard, 1H NMR

48

35

a Retention time. b s: singlet, d: doublet, t: triplet, q: quartet, m: multiplet; referenced to CD3OD at 3.3 ppm. c Based on NMR, enrichment factor 3077, referenced to Centralite I, assuming 100% recovery. d Based on NMR, enrichment factor 1538, referenced to Centralite I, assuming 100% recovery.

VOL. 43, NO. 18, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7059

whereas NOECs (no observed effect concentration) varied between 0.02-0.37 mg/L for algae. Therefore, a QSAR (quantitative structure-activity relationship) model was applied to estimate the LC50 (concentration which kills 50% of a sample population) values for the fathead minnow (Pimephales promelas) for Centralite I and DPU and to compare them to DPA (37). For DPA the mode of action in fish was described as baseline-narcosis (narcosis I) (37). In this context more lipophilic compounds exhibit a higher toxicity. Based on LC50 values for the fathead minnow and log Pow values, Russom et al. (37) developed a QSAR which allowed the calculation of LC50 values of Centralite I and DPU according to the following equation: log molar LC50 ) -0.94logPOW + 0.94log(0.000068POW + 1) - 1.25 EC50 values of the compounds were derived from a second QSAR as described by Walter (38) for the estimation of EC50 values for the green algae, Scenedesmus vacuolatus, assuming the same mode of action: log EC50(mol/L) ) -0.81logPOW - 0.87 Results obtained are summarized in Table S2 in the Supporting Information. As can be seen, LC50 and EC50 values of Centralite I and DPU are somewhat lower compared to DPA, which means that these compounds might be more toxic than DPA. Finally, a PNEC (predicted no effect concentration) of 0.4 µg/L was obtained for DPA based on the lowest NOEC (no observed effect concentration) for algae and applying a safety factor of 50. Although no PNECs could be calculated for Centralite I and DPU, the presented data suggest that these compounds might well have a negative impact in the aquatic environment at rather low concentrations. This statement is based on the data obtained for the reference compound DPA and the POW values of the other compounds under consideration. Therefore, stabilizers as characterized in this study, should be included in future investigations of military hot spots.

Acknowledgments We thank Joerg Mathieu of Armasuisse (Federal Department of Defence Civil Protection and Sport) for providing reference standards of stabilizer compounds for 1H NMR comparison studies, Oliver Steiner of AWA for the site description of the Susten-Stone Glacier and the sampling procedure, and Marion Junghans for providing toxicological data of stabilizers. David Kilgour is kindly acknowledged for proofreading this manuscript.

Supporting Information Available Additional information mentioned within this article. This information is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Ochsenbein, U.; Zeh, M.; Berset, J. D. Comparing solid phase extraction and direct injection for the analysis of ultra-trace levels of relevant explosives in lake water and tributaries using liquid chromatography-electrospray tandem mass spectrometry. Chemosphere 2008, 72 (6), 974–980. (2) Eisentraeger, A.; Reifferscheid, G.; Dardenne, F.; Blust, R.; Schofer, A. Hazard characterization and identification of a former ammunition site using microarrays, bioassays, and chemical analysis. Environ. Toxicol. Chem. 2007, 26 (4), 634– 646. (3) Schmidt, A. C.; Niehus, B.; Matysik, F. M.; Engewald, W. Identification and quantification of polar nitroaromatic compounds in explosive-contaminated waters by means of HPLCESI-MS-MS and HPLC-UV. Chromatographia 2006, 63 (1-2), 1–11. 7060

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 18, 2009

(4) Spiegel, K.; Headley, J. V.; Peru, K. M.; Haidar, N.; Gurorasard, N. P. Residues of explosives in groundwater leached from soils at a military site in Eastern Germany. Commun. Soil Sci. Plant Anal. 2005, 36 (1-3), 133–153. (5) Grummt, T.; Wunderlich, H. G.; Chakraborty, A.; Kundi, M.; Majer, B.; Ferk, F.; Nersesyan, A. K.; Parzefall, W.; Knasmu ¨ ller, S. Genotoxicity of nitrosulfonic acids, nitrobenzoic acids and nitrobenzylalcohols, pollutants commonly found in groundwater near ammunition facilities. Environ. Mol. Mutagen. 2006, 47 (2), 95–106. (6) Neuwoehner, J.; Schofer, A.; Erlenkaempfer, B.; Steinbach, K.; Hund-Rinke, T. K.; Eisentraeger, A. Toxicological characterization of 2,4,6-trinitrotoluene, its transformation products, and two nitramine explosives. Environ. Toxicol. Chem. 2007, 26 (6), 1090–1099. (7) Wollin, K.-M.; Dieter, H. H. Toxicological guidelines for monocyclic nitro-, amino- and aminonitroaromatics, nitramines and nitrate esters in drinking water. Arch. Environ. Contam. Toxicol. 2005, 49 (1), 18–26. (8) Yinon, J. Mass spectrometry of explosives: Nitro compounds, nitrate esters, and nitramines. Mass Spectrom. Rev. 1982, 1, 257–307. (9) Yinon, J.; McClellan, J. E.; Yost, R. A. Electrospray ionization tandem mass spectrometry collision-induced dissociation study of explosives in an ion trap mass spectrometer. Rapid Commun. Mass Spectrom. 1997, 11 (18), 1961–1070. (10) Yinon, J.; Zhao, X. Characterization and origin identification of 2,4,6-trinitrotoluene through its by-product isomers by liquid chromatography-atmospheric pressure chemical ionization mass spectrometry. J. Chromatogr. A 2002, 946 (1-2), 125–132. (11) Yinon, J.; Zhao, X. Identification of nitrate ester explosives by liquid chromatography-electrospray ionization and atmospheric pressure chemical ionization mass spectrometry. J. Chromatogr. A 2002, 977 (1), 59–68. (12) Xu, X.; van de Craats, A. M.; de Bruyn, P. C. Highly sensitive screening methods for nitroaromatic, nitramine and nitrate ester explosives by high performance liquid chromatographyatmospheric pressure ionization - mass spectrometry (HPLCAPI-MS) in forensic applications. J. Forensic Sci. 2004, 49 (6), 1171–1180. (13) Crescenzi, C.; Albin ˜ ana, J.; Carlsson, H.; Holmgren, E.; Battle, R. On-line strategies for determining trace levels of nitroaromatic explosives and related compounds in water. J. Chromatogr. A 2007, 1153 (1-2), 186–193. (14) Ma, W. T.; Steinbach, K.; Cai, Z. Analysis of dinitro- and aminonitro-toluenesulfonic acids in groundwater by solid-phase extraction and liquid chromatography-mass spectrometry. Anal. Bioanal. Chem. 2004, 378 (7), 1828–1835. (15) Iba´n ˜ ez, M.; Sancho, J. V.; Pozo, O. J.; Niessen, W.; Herna˜ndez, F. Use of quadrupole time-of-flight mass spectrometry in the elucidation of unknown compounds present in environmental water. Rapid Commun. Mass Spectrom. 2005, 19 (2), 169–178. (16) Williamson, L. N.; Bartlett, M. G. Quantitative liquid chromatography/time-of-flight mass spectrometry. Biomed. Chromatogr. 2007, 21 (6), 567–576. (17) Preiss, A.; Levsen, K.; Humpfer, E.; Spraul, M. Application of high-field proton nuclear magnetic resonance (1H-NMR) spectroscopy for the analysis of explosives and related compounds in groundwater samples - a comparison with the highperformance liquid chromatography (HPLC) method. Anal. Bioanal. Chem. 1996, 356 (7), 445–451. (18) Preiss, A.; Elend, M.; Gerling, S.; Tra¨nckner, S. Analysis of highly polar compounds in groundwater samples from ammunition waste sites. Part I-Characterization of the pollutant spectrum. Magn. Reson. Chem. 2005, 43 (9), 736–746. (19) Godejohann, M.; Preiss, A.; Levsen, K.; Wollin, K.-M.; Mu ¨ gge, C. Determination of polar organic pollutants in aqueous samples of former ammunition sites in Lower Saxony by means of HPLC/ photodiode array detection (HPLC/PDA) and proton nuclear magnetic resonance spectroscopy (1H-NMR). Acta Hydrochim. Hydrobiol. 1998, 26 (6), 330–337. (20) Yang, Z. Online hyphenated liquid chromatography-nuclear magnetic resonance spectroscopy-mass spectrometry for drug metabolite and nature product analysis. J. Pharm. Biomed. Anal. 2006, 40 (3), 516–527. (21) Corcoran, O.; Spraul, M. LC-NMR-MS in drug discovery. Drug Discov. Today 2003, 8 (14), 624–631. (22) Jaroszewski, J. W. Hyphenated NMR methods in natural products research, part 1: direct hyphenation. Planta Med. 2005, 71 (8), 691–700.

(23) Godejohann, M.; Preiss, A.; Mu ¨ gge, C.; Wu ¨ nsch, G. Application of on-line HPLC-1H NMR to environmental samples: analysis of groundwater near former ammunition plants. Anal. Chem. 1997, 69 (18), 3832–3837. (24) Godejohann, M.; Preiss, A.; Mu ¨ gge, C. Quantitative measurements in continuous-flow HPLC/NMR. Anal. Chem. 1998, 70 (3), 590–595. (25) Godejohann, M.; Astratov, M.; Preiss, A.; Levsen, K. Application of continuous-flow HPLC-proton-nuclear magnetic resonance spectroscopy and HPLC-thermospray-mass spectroscopy for the structural elucidation of phototransformation products of 2,4,6-trinitrotoluene. Anal. Chem. 1998, 70 (19), 4104–4110. (26) Schrader, W.; Geiger, J.; Godejohann, M. Studies of complex reactions using modern hyphenated methods: alpha-pinene ozonolysis as a model reaction. J. Chromatogr. A 2005, 1075 (1-2), 185–196. (27) Preiss, A.; Elend, M.; Gerling, S.; Berger-Preiss, E.; Steinbach, K. Identification of highly polar nitroaromatic compounds in leachate and ground water samples from a TNT-contaminated waste site by LC-MS, LC-NMR and off-line NMR and MS investigations. Anal. Bioanal. Chem. 2007, 389 (6), 1979–1988. (28) Exarchou, V.; Godejohann, M.; Van Beek, T.; Gerothanassis, I. P.; Vervoort, J. LC-UV-solid phase extraction-NMR-MS combined with a cryogenic flow probe and its application to the identification of compounds present in Greek oregano. Anal. Chem. 2003, 75 (22), 6288–6294. (29) Godejohann, M.; Tseng, L. H.; Braumann, U.; Fuchser, J.; Spraul, M. Characterization of a paracetamol metabolite using on-line LC-SPE-NMR-MS and a cryogenic NMR probe. J. Chromatogr. A 2004, 1058 (1-2), 191–196.

(30) Bausinger, T.; Dehner, U.; Preuβ, J. Bestimmung von Nitrodiphenylaminen und verwandten Verbindungen im Sickerwasser einer Ru ¨ stungsaltlast. Z. Umweltchem. Oekotox. 2005, 17 (1), 7–12. (31) Folly, P.; Ma¨der, P. Propellant Chemistry. Chimia 2004, 58 (6), 374–382. (32) Wu, Z.; Tong, Y.; Yu, J.; Zhang, X.; Pan, C.; Deng, X.; Xu, Y.; Wen, Y. Detection of N, N’-diphenyl-N, N’-dimethylurea (methyl centralite) in gunshot residues using MS-MS method. Analyst 1999, 124 (11), 1563–1567. (33) European Commission, Joint Research Center, Institute for Health and Consumer Protection. ESIS: European chemical Substances Information System; http://ecb.jrc.ec.europa.eu/ esis. (34) U.S. Environmental Protection Agency. ECOTOX Database; http://cfpub.epa.gov/ecotox. (35) PAN Pesticides Database. Chemicals; http://pesticideinfo.org/ Search_Chemicals.jsp. (36) EPA DSSTOX Structure Browser v2.0; http://www.epa.gov/ dsstox_structurebrowser. (37) Russom, C. L.; Bradbury, S. P.; Broderius, S. J.; Hammermeister, D. E.; Drummond, R. A. Predicting modes of toxic action from chemical structure: acute toxicity in the fathead minnow (Pimephales promelas). Environ. Toxicol. Chem. 1997, 16 (5), 948–967. (38) Walter, H.; Consolaro, F.; Gramatica, P.; Scholze, M.; Altenburger, R. Mixture toxicity of priority pollutants at no observed effect concentrations (NOECs). Ecotoxicology 2002, 11 (5), 299–310.

ES901068D

VOL. 43, NO. 18, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7061