- Email: [email protected]
Simultaneous determination of 2-naphthol and 1-hydroxy pyrene in urine by gas chromatography–mass spectrometry
Journal of Chromatography B, 879 (2011) 489–494
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Simultaneous determination of 2-naphthol and 1-hydroxy pyrene in urine by gas chromatography–mass spectrometry Ho-Sang Shin a,∗ , Hyun-Hee Lim b a b
Department of Environmental Education, Kongju National University, Kongju 314-701, Republic of Korea Department of Environmental Science, Kongju National University, Kongju 314-701, Republic of Korea
a r t i c l e
i n f o
Article history: Received 9 July 2010 Accepted 7 January 2011 Available online 14 January 2011 Keywords: GC–MS 1-Hydroxy pyrene 2-Naphthol Urine MTBDMSTFA silylation
a b s t r a c t A gas chromatography–mass spectrometric (GC–MS) method was developed for the determination of 2naphthol (2-NAP) and 1-hydroxypyrene (1-HOP) in human urine. Extraction from urine after the enzyme hydrolysis with -glucuronidase/arylsulfatase was achieved with a liquid extraction using 5 mL of pentane. After addition of 50 L of N-methyl-N-(tert-butyldimethylsilyl) trifluoroacetamide (MTBDMSTFA) to prevent the loss of 2-NAP during drying, the extract was completely dried and derivatized with MTBDMSTFA for 30 min at 60 ◦ C. The accuracies were in the range of 96–109% at a concentration of 0.5, 10 and 25 g/L and their precisions were less than 15%. Method detection limits of 2-NAP and 1-HOP were 0.07 and 0.01 g/L, respectively. This method was used to analyze twenty urine samples, and they were found in the concentration range <0.07–13.7 g/L (2-NAP) and <0.01–0.88 g/L (1-HOP). The concentrations of 2-NAP and 1-HOP were well correlated to those of naphthalene and pyrene in blood, respectively. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are chemical compounds that consist of fused aromatic rings, and they occur in oil and coal, and are produced during burning. PAHs are one of the most widespread organic pollutants. As pollutants, they are of concern because some of these compounds have been identified as carcinogenic, mutagenic and teratogenic [1,2]. PAHs are also found in some foods. Most food intake of PAHs comes from cereals, oils and fats [3]. Following exposure, PAHs are metabolized to hydroxy PAHs. Urinary 1-hydroxy pyrene (1-HOP) and 2-naphthol (2-NAP) have been proposed as biomarkers of PAHs exposure [4,5]. Many analytical procedures have been proposed for the determination of ng/L levels of these biomarkers of PAHs exposure. Almost all of these are based on the determination of 1-OHP and 2NAP by high-performance liquid chromatography [6–20]. Among these methods, fluorimetric HPLC methods [6–12] are the most conventional and most frequently used methods for the quantification of 1-OHP and 2-NAP in urine. Also, to date many analysis methods using liquid chromatography–tandem mass spectrometry [13–20] have been developed for the simultaneous determination of 1-OHP and 2-NAP in urine. Synchronous fluorescence determination [21–24] has proved to be a valuable technique owing to its high level of efficiency and low sample volume requirement. Also, chemiluminescent enzyme immunoassay [25] and voltammetric
determination methods [26] have been developed. But these methods are not sensitive for the measurement of 1-OHP and 2-NAP in population urines. Gas chromatographic methods have been published for the analysis of 1-OHP and 2-NAP in human urine; these methods involve electron impact ionization-mass spectrometry (GC–EI-MS) [27,28] and use high resolution mass spectrometry (GC–HRMS) [29,30]. Campo et al. [27] described a sensitive GC–MS method to determine monohydroxy metabolites of PAHs by derivatization with a silylating agent. This method had detection limits as low as 0.5 g/L and 0.9 g/L for 1-OHP and 2-NAP, respectively. These detection limits are insufficient to monitor background concentrations of 1-OHP and 2-NAP in population urines. 2-NAP is relatively volatile, and can be vaporized during the concentration and dryness after the extraction. A treatment needs to prevent the loss of 2-NAP during drying. We had tried to develop a precise and sensitive detection method of these compounds by using gas chromatograph/mass spectrometry (GC/MS) after the silylation with methyl-N-(tertbutyldimethylsilyl) trifluoroacetamide (MTBDMSTFA). The analytical method was applied to the quantification of 1-OHP and 2-NAP in population urines. 2. Experimental 2.1. Chemicals and reagents
∗ Corresponding author. Tel.: +82 41 850 8811; fax: +82 41 850 8998. E-mail address: [email protected] (H.-S. Shin). 1570-0232/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jchromb.2011.01.009
2-NAP (>99%), 1-OHP (98%), 1-OHP-d9 (99%), naphthalene (99%), pyrene (98%), pyrene-10 (99%), sodium acetate,
490
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
Table 1 The source of the biological samples. Sample no.
Age
Gender
Occupational exposure
Smoking status
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
59 68 52 41 21 57 62 65 52 31 62 43 54 33 36 57 57 54 67 21
Female Male Female Female Male Female Female Female Female Male Female Male Female Female Female Male Male Female Female Male
No No No No No No No No No No No No No No No No No No No No
S S S N S N N S N N N N N N S N N S N N
for 1 min, raised to 320 ◦ C at 20 ◦ C/min and held for 5 min. All mass spectra were obtained with an Agilent 5975 B instrument. The ion source was operated in the electron ionization mode (EI; 70 eV, 230 ◦ C). Full-scan mass spectra (m/z 50–800) were recorded for the identification of analytes at high concentration. Confirmation of trace chemicals was completed by three MS characteristic ions, the ratio of three MS characteristic ions and GC-retention time match to the known standard compound. The ions selected in this study were m/z 185, 201, 216 for TMS-2-NAP, m/z 259, 275, 290 for TMS1-HOP, m/z 267, 284, 299 for 1-HOP-d9-TMS, m/z 185, 201, 258 for TBDMS-2-NAP, m/z 259, 275, 332 for TBDMS-1-HOP, and m/z 267, 284, 341 for TBDMS-1-HOP-d9. 2.5. Calibration and quantification Calibration curves for 2-NAP and 1-HOP were established by extraction after adding 0, 2, 10, 20, 40, 100 and 200 ng of standards and 5.0 ng of internal standard in 2 mL of blank urine. 1-HOP-d9 was used as the internal standard. The ratio of the peak area of standard to that of internal standard was used in the quantification of the compound.
S, smoker; N, nonsmoker.
2.6. Extraction and determination of PAHs in blood -glucuronidase/aryl sulfatase, N-methyl-N-trimethylsilyl trifluoroacetamide (MSTFA) and N-methyl-N-(tert-butyldimethylsilyl) trifluoroacetamide (MTBDMSTFA) were purchased from Sigma (St. Louis, MO, USA), and n-pentane and acetone purchased from J.T. Baker (USA) were used as solvents. Water was purified in milli-Q (Millipore Corp., Milford, MA). Spot urine and blood samples were obtained from volunteers. The source of the urine and blood samples was described in detail in Table 1. Blank urine, in which the analytes were not detected with our method, was used. Solutions of standards were prepared at 1000 mg/L in methanol: methylene chloride (1:1) and were stored at −20 ◦ C. Working solutions were used after 500 times or more dilution of the standard solutions using methanol and prepared newly before the experiment. 2.2. Spiking Spiked samples of human urine (2.0 mL) were prepared with 20–100 L of standard solutions at concentrations of 0.1–1.0 mg/L and with 50 L of the internal standard solution at a concentration of 0.1 mg/L. 2.3. Enzyme hydrolysis, extraction and derivatization procedure A 2.0 mL sample of urine was placed in a 20-mL test tube. About 2.0 mL of sodium acetate buffer solution and 10 L of glucuronidase/aryl sulfatase were added to the sample solution, and the sample was incubated for 3 h at 37 ◦ C. 50 L of 1-HOP-d9 solution (0.1 mg/L in acetone) was added to the solution and the sample was extracted with 5.0 mL of pentane by mechanical shaking for 20 min. 50 L MTBDMSTFA was added to the solution, and the solution was evaporated in vacuum rotary to about 0.1 mL, and then almost dried with a nitrogen stream. The dry residue was dissolved with 50 L of MTBDMSTFA and the tubes were heated for 30 min at 60 ◦ C. A 2 L sample of the solution was injected into the GC system. 2.4. Gas chromatography–mass spectrometry The gas chromatograph used was an Agilent 7890 A with a split/splitless injector. The analytical column was a 30 m HP-5MS column (cross-linked 5% phenylmethylsilicon, 0.2 mm I.D. × 0.25 m F.T). Oven temperature program began at 80 ◦ C, held
A 0.5 mL sample of blood was placed in a 20-mL test tube. 50 L of pyrene-d10 solution (2.0 mg/L in acetone) was added to the solution and the sample was extracted with 5.0 mL of pentane by mechanical shaking for 20 min. The extract was evaporated in vacuum rotary and almost dried with a nitrogen stream. The dry residue was dissolved with 50 L of methanol. At appropriate times, a 2 L sample of the solution was injected into the GC system. Calibration curves for naphthalene and pyrene were established by extraction after adding 0, 0.1, 0.5, 1.0, 2.0, 5.0 and 10 ng of standards and 50 L of pyrene-d10 solution (2.0 mg/L in acetone) in 0.5 mL of blank blood. Pyrene-d10 was used as the internal standard. The ratio of the peak area of standard to that of internal standard was used in the quantification of the compound. The gas chromatograph used was an Agilent 7890 A with a split injector (split ratio = 5:1). The analytical column was a 30 m HP-5MS column (cross-linked 5% phenylmethylsilicon, 0.2 mm I.D. × 0.25 m F.T). Oven temperature program began at 50 ◦ C, held for 1 min, raised to 300 ◦ C at 20 ◦ C/min and held for 10 min. All mass spectra were obtained with an Agilent 5975 B instrument. The ion source was operated in the electron ionization mode (EI; 70 eV, 230 ◦ C). Full-scan mass spectra (m/z 50–800) were recorded for the identification of analytes at high concentration. Confirmation of trace chemicals was completed by three MS characteristic ions, the ratio of three MS characteristic ions and GC-retention time match to the known standard compound. The ions selected by SIM were m/z 128, 129 and 127 for naphthalene; m/z 202, 200 and 203 for pyrene and m/z 212 for pyrene-d10 (internal standard). 3. Results and discussion 3.1. Chromatography For the GC separation of TBDMS-1-OHP and TBDMS-2-NAP, the use of the nonpolar stationary phase was found to be efficient. The peaks of TBDMS-1-OHP, TBDMS-1-OHP-d9 and TBDMS-2-NAP were symmetrical and separation of the analytes from the background compounds in urine extract was more easily observed when compared to TMS-1-OHP, TMS-1-OHP-d9 and TMS-2-NAP (Fig. 1). The retention times of TBDMS-2-NAP, TBDMS-1-OHP-d9 and TBDMS-1-OHP were 7.8, 11.9 and 12.0 min, respectively. In a chromatogram of blank urine, there were no extraneous peaks observed in the time range of between 7.0 and 13.0 min.
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
491
Fig. 1. Chromatograms of standard (A and B) and volunteer sample (C). The standard solution was spiked to a concentration of 4.0 g/L for 2-NAP, 1 g/L for 1-HOP and 1.0 g/L for 1-HOP-d9 in blank urine. Sample was quantified as the concentration of 1.6 g/L for 2-NAP and 0.29 g/L for 1-HOP.
3.2. Mass spectrometry Mass spectra analyses of TBDMS-2-NAP, TBDMS-1-HOP and TBDMS-1-HOP-d9 were obtained (Fig. 2). TBDMS-2-NAP showed molecular ion at m/z 258 and the base peak at m/z 201 that cleavages t-butyl group from molecular ion and characteristic ions at m/z 127, 146 and 186. The ions at 186 and 146 were from the losses of one methyl group or [2CH3 + Si–3H] group from m/z 201 ion, and the
ion at 127 was a result of the loss of [TBDMSO] group from molecular ion. TBDMS-1-HOP revealed molecular ion at m/z 332 and the base ion at m/z 275, which was due to the loss of t-butyl group from molecular ion, and characteristic ions at m/z 217, 244 and 259. The ions at 259 and 244 were the result of the losses of [CH3 + H] or [2CH3 + H] from m/z 275 ion, and the ion at 217 was from the loss of [TBDMS] group from molecular ion. TBDMS-1-HOP-d9 showed molecular and base ion at m/z 341 and m/z 284, which is due to the
492
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
Fig. 2. Mass spectra of TBDMS-2-NAP (A), TBDMS-1-HOP (B) and TBDMS-1-HOP-d9 (C).
loss of t-butyl group from molecular ion, and characteristic ions at m/z 226, 252 and 267. The ions at 267 and 252 were the result of the losses of [CH3 + D] or [2CH3 + D] from m/z 284 ion, and the ion at 226 was due to the loss of [TBDMS] group from molecular ion. Selected ions for the quantization were m/z 258 for TBDMS-2-NAP, m/z 332 for TBDMS-1-HOP, and m/z 341 for TBDMS-1-HOP-d9.
3.3. Derivatization 1-OHP and 2-NAP have a polar hydroxy group, which can be silylated in displacement reaction. The silylating reagents, MSTFA and MTBDMSTFA, were compared to each other in terms of repeatability, sensitivity and stability of the derivatives.
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
493
Table 2 Within run-precision and accuracy of target compounds in urine (n = 5). Compound
Amount spiked (g/L)
Results (g/L) Calculated conc
Mean ± SD (RSD%)
Accuracy (%)
1-OHP
0.5 10.0 25.0
0.49, 0.52, 0.53, 0.52, 0.54 9.7, 9.8, 9.8 9.6, 9.7 24.5, 24.4, 23.7, 23.7, 24.5
0.52 ± 0.017 (3.3%) 9.7 ± 0.1 (0.7%) 24.2 ± 0.4 (1.7%)
104 97.0 96.8
2-NAP
0.5 10.0 25.0
0.51, 0.48, 0.38, 0.51, 0.50 12.5, 10.9, 11.3, 12.8, 7.1 24.8, 31.6, 19.0, 27.5, 21.8
0.48 ± 0.06 (12.3%) 10.9 ± 1.5 (14.1%) 24.9 ± 4 (14.8%)
96.0 109 99.6
SD, standard deviation; RSD, relative standard deviation (%).
2-NAP is fairly volatile and can be vaporized during concentration and drying. We tested the usability of MSTFA or MTBDMSTFA as a keeper to prevent the loss. After 20 ng of 2-NAP was spiked with 5 mL of pentane (n = 3), MSTFA or MTBDMSTFA was added in pentane solution. The mixture was shaken for 10 min mechanically, the solvent was dried completely. The residue was once more derivatized for 30 min at 60 ◦ C with 50 L of MSTFA or MTBDMSTFA. As a result, mean ± standard deviation of the area of the derivatives was 19,936 ± 16,347 (n = 3) without MTBDMSTFA and 26,531 ± 1881 with MTBDMSTFA, and 20,642 ± 15,430 (n = 3) without MSTFA and 25,648 ± 5137 with MSTFA. The recovery, that is the ratios between the area of 2-NAP detected after drying with and without MTBDMSTFA and MSTFA, were 1.33 and 1.24, respectively. 2-NAP-TBDMS and 2-NAP-TMS formed from this treatment turn out to be less volatile. Then these derivatives were detected in high peak area with small standard deviation. The derivatives were analyzed at reaction times of 30, 60, 90 and 120 min. The reaction rate of 1-OHP and 2-NAP with MSTFA and MTBDMSTFA was determined by detecting peak area of the products. 1-OHP and 2-NAP showed very rapid and complete reaction with both MSTFA and MTBDMSTFA. TBDMS-1-OHP, TBDMS-1OHP-d9 and TBDMS-2-NAP were detected as the products of 1-OHP, 1-OHP-d9 and 2-NAP with MTBDMSTFA in about 30 min at 60 ◦ C. To evaluate the stability of the derivatives, the experiment was repeated three times and the same samples were analyzed after 100 h. The stability was evaluated as area percentage of the derivatives obtained on the first day of analysis to that of the derivatives reacted newly after 100 h. The mean percentual stability after 100 h varied from 5.65% (1-OHP) to 7.3% (2-NAP). Otherwise, that of the TMS derivatives after 100 h varied from 11.5% (1-OHP) to 13.7% (2-NAP). As a result, TBDMS-1-OHP and TBDMS-2-NAP derivatized by MTBDMSTFA showed superior repeatability and stability compared to those derivatized by MSTFA.
3.4. Validation of the assay Despite the improvement of other alternative extraction techniques, liquid-liquid extraction (LLE) is still an efficient technique for the routinely performed analysis of 1-OHP and 2-NAP in urine. LLE was performed for the analysis of 1-OHP and 2-NAP in urine. Test samples at 10.0 and 25.0 g/L in urine were prepared and the relative recovery was calculated by percentage of the analytes recovered. The recoveries of the test compounds were about 107% for 2-NAP and about 97% for 1-OHP. Using the least-squares fit technique, an examination of typical standard curve was performed by computing a regression line of peak area ratios for 1-OHP and 2-NAP to the internal standard (1-OHP-d9) on the concentration. This examination demonstrated a linear relationship with correlation coefficients being greater than 0.990. Linear equations were y = 8.5142x + 0.0131 for 2-NAP and y = 7.9504x + 0.0068 for 1-HOP in the concentration range of 0.1–50.0 g/L. The method detection limit (MDL) was defined by 3.14 times of standard deviation for replicate determinations (n = 7) from samples spiked at the concentration of 0.01 g/L in blank urine [31], in which the concentration is a value that corresponds to an instrument signal/noise ratio within the range of 2.5. MDL was calculated 0.07 g/L for 2-NAP and 0.01 g/L for 1-HOP based upon an assayed sample of 2.0 mL. The range and standard deviation values for the within run-precision and accuracy are given in Table 2. For five independent determinations at 0.5, 10 and 25.0 g/L, the coefficient of variation was less than 15%. The construction of typical standard curves of naphthalene and pyrene was performed by above method. Linear equations were y = 0.3394x + 0.0011 for naphthalene and y = 2.1317x + 0.0238 for pyrene in the concentration range of 0.01–50 g/L. MDLs of naphthalene and pyrene were calculated as 0.01 g/L in plasma.
Fig. 3. Analytical results of 2-NAP and 1-HOP in 20 volunteer samples.
Naphthalene Conc. (µg/L)
494
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
60
4. Conclusions
y = 1.337x + 6.7612 R² = 0.2138
50 40 30 20 10 0
0
2
4
6
8
10
2-NAP Conc. (µg/L)
12
14
16
Fig. 4. Relationship of urinary 2-NAP concentration and the naphthalene concentration in plasma.
Pyrene Conc. (µg/L)
0.9
y = 0.9029x - 0.0072 R² = 0.6757
0.8
References
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 -0.1 0
For the enhancement of GC performance and sensitivity, the silylating reagents, MSTFA and MTBDMSTFA were compared to each other in terms of repeatability and stability of the derivatives. The result showed that TBDMS-1-OHP and TBDMS-2-NAP derivatized by MTBDMSTFA showed superior repeatability and stability compared to those derivatized by MSTFA. The peaks of TBDMS-2-NAP and TBDMS-1-HOP have good chromatographic properties by using HP-5MS column and show sensitive response for the GC–MS (SIM). Method detection limits of target compounds were 0.07 g/L for 2-NAP and 0.01 g/L for 1-OHP. Urine samples from 20 volunteers were analyzed by the developed method. The results showed that urinary 2-NAP and 1-HOP showed good correlation with the concentrations of naphthalene and pyrene in blood, respectively. The developed method was shown to be a useful method in the identification of human exposure to PAHs.
0.2
0.4
0.6
0.8
1
1-HOP Conc. (µg/L) Fig. 5. Relationship of urinary 1-HOP concentration and the pyrene concentration in plasma.
3.5. Sample analysis We analyzed the target compounds in 20 volunteer samples. 2-NAP and 1-OHP were found in the concentration range <0.07–13.7 g/L (2-NAP) and <0.01–0.88 g/L (1-OHP) and above the LOD in 12/20 and 8/20 samples (Fig. 3). Confirmation of trace 2-NAP and 1-HOP was performed by the ratio of two MS characteristic ions and GC-retention time match to the known standard compound. Otherwise, naphthalene and pyrene in plasma were found in the concentration range of <0.01–23.95 g/L and <0.01–0.75 g/L, respectively. The relationship of urinary metabolites concentrations and plasma PAHs concentrations was constructed. As shown in Fig. 4, urinary 2-NAP was found to be well correlated with the concentration of naphthalene in plasma (r2 = 0.21, P = 0.002). Also, urinary 1-HOP showed good correlation with the concentration of pyrene in plasma (r2 = 0.68, P = 0.013) (Fig. 5). It appeared that urinary 1HOP or 2-NAP is good indicator of naphthalene or pyrene levels exposed.
[1] B. Marczynski, R. Preuss, T. Mensing, J. Angerer, A. Seidel, A.E. Mourabit, M. Wilhelm, T. Brüning, Int. Arch. Occup. Environ. Health 78 (2005) 97. [2] International Program on Chemical Safety (IPCS), Environmental Health Criteria 202. Selected Non-heterocylic Polycyclic Aromatic Hydrocarbons, WHO, Geneva, 1998. [3] Polycyclic aromatic hydrocarbons—occurrence in foods, dietary exposure and health effects, European Commission, Scientific Committee on Food. December 4, 2002. http://ec.europa.eu/food/fs/sc/scf/out154 en.pdf. [4] F.J. Jongeneelen, R.B.M. Anzion, P.T.J. Scheepers, R.P. Bos, P.T. Henderson, E.H. Nijenhuis, S.J. Veenstra, R.M. Brouns, A. Winkers, Ann. Occup. Hyg. 32 (1988) 35. [5] M. Yang, M. Koga, T. Katoh, T. Kawamoto, Arch. Environ. Contam. Toxicol. 36 (1) (1999) 99. [6] T. Chetiyanukornkul, A. Toriba, R. Kizu, J. Chromatogr. A 961 (1) (2002) 107. [7] C.-T. Kuo, H.-W. Chen, J.-L. Chen, J. Chromatogr. B 805 (2) (2004) 187. [8] E. Eivor, Arch. Toxicol. 77 (4) (2003) 183. [9] H. Li, R.I. Krieger, X. LiQ, Sci. Total Environ. 257 (2000) 147. [10] C. Xavier, B. Maryline, R. Miguel, J. Sep. Sci. 32 (20) (2009) 3403. [11] S.M. Ramasamy, R.J. Hurtubise, A. Weston, Appl. Spectrosc. 51 (9) (1997) 1377. [12] R. Preuss, J. Angerer, J. Chromatogr. B 801 (2) (2004) 307. [13] L. Xu, D.C. Spink, J. Chromatogr. B 855 (2) (2007) 159. [14] D. Pigini, A.M. Cialdella, P. Faranda, Rapid Commun. Mass Spectrom. 20 (6) (2006) 1013. [15] R. Fan, Y. Dong, W. Zhang, J. Chromatogr. B 836 (2006) 92. [16] K. Kakimoto, A. Toriba, T. Ohno, J. Chromatogr. B 867 (2) (2008) 259. [17] S. Ferrari, F. Mandel, J.D. Berset, Chemosphere 47 (2) (2002) 173. [18] L. Yinghe, L.C. Austin, S. Haihong, Rapid Commun. Mass Spectrom. 19 (22) (2005) 3331. [19] A. Holm, P. Molander, E. Lundanes, J. Chromatogr. B 794 (1) (2003) 175. [20] P. Jacob III, M. Wilson, N.L. Benowitz, Anal. Chem. 79 (2) (2007) 587. [21] L. Michel, B. Rachid, F. de Violet Philippe, Anal. Bioanal. Chem. 376 (6) (2003) 816. [22] R.H. Kang, Y.S. Wang, H.M. Yang, Anal. Chim. Acta 658 (2) (2010) 180. [23] Y. Zhang, Y. Zhu, S. Wang, Bull. Environ. Contam. Toxicol. 69 (4) (2002) 494. [24] H.M. Yang, Y.S. Wang, J.H. Li, Anal. Chim. Acta 636 (1) (2009) 51. [25] R. Aldo, P. Simoni, M. Mirasoli, Anal. Bioanal. Chem. 372 (2002) 751. [26] K. Nicole, H. Kevin, C. Emsp, Electroanalysis 17 (7) (2005) 571. [27] L. Campo, F. Rossella, S. Fustinoni, J. Chromatogr. B 875 (2008) 531. [28] K. Desmet, B. Tienpont, P. Sandra, Chromatographia 57 (2003) 681. [29] Z. Li, L.C. Romanoff, D.A. Trinidad, Anal. Chem. 78 (16) (2006) 5744. [30] C.J. Smith, C.J. Walcott, W. Huang, J. Chromatogr. B 778 (2002) 157. [31] USEPA Method SW 846, Quality Control, 1992, p. 26 (Chapter 1).
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Simultaneous determination of 2-naphthol and 1-hydroxy pyrene in urine by gas chromatography–mass spectrometry Ho-Sang Shin a,∗ , Hyun-Hee Lim b a b
Department of Environmental Education, Kongju National University, Kongju 314-701, Republic of Korea Department of Environmental Science, Kongju National University, Kongju 314-701, Republic of Korea
a r t i c l e
i n f o
Article history: Received 9 July 2010 Accepted 7 January 2011 Available online 14 January 2011 Keywords: GC–MS 1-Hydroxy pyrene 2-Naphthol Urine MTBDMSTFA silylation
a b s t r a c t A gas chromatography–mass spectrometric (GC–MS) method was developed for the determination of 2naphthol (2-NAP) and 1-hydroxypyrene (1-HOP) in human urine. Extraction from urine after the enzyme hydrolysis with -glucuronidase/arylsulfatase was achieved with a liquid extraction using 5 mL of pentane. After addition of 50 L of N-methyl-N-(tert-butyldimethylsilyl) trifluoroacetamide (MTBDMSTFA) to prevent the loss of 2-NAP during drying, the extract was completely dried and derivatized with MTBDMSTFA for 30 min at 60 ◦ C. The accuracies were in the range of 96–109% at a concentration of 0.5, 10 and 25 g/L and their precisions were less than 15%. Method detection limits of 2-NAP and 1-HOP were 0.07 and 0.01 g/L, respectively. This method was used to analyze twenty urine samples, and they were found in the concentration range <0.07–13.7 g/L (2-NAP) and <0.01–0.88 g/L (1-HOP). The concentrations of 2-NAP and 1-HOP were well correlated to those of naphthalene and pyrene in blood, respectively. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are chemical compounds that consist of fused aromatic rings, and they occur in oil and coal, and are produced during burning. PAHs are one of the most widespread organic pollutants. As pollutants, they are of concern because some of these compounds have been identified as carcinogenic, mutagenic and teratogenic [1,2]. PAHs are also found in some foods. Most food intake of PAHs comes from cereals, oils and fats [3]. Following exposure, PAHs are metabolized to hydroxy PAHs. Urinary 1-hydroxy pyrene (1-HOP) and 2-naphthol (2-NAP) have been proposed as biomarkers of PAHs exposure [4,5]. Many analytical procedures have been proposed for the determination of ng/L levels of these biomarkers of PAHs exposure. Almost all of these are based on the determination of 1-OHP and 2NAP by high-performance liquid chromatography [6–20]. Among these methods, fluorimetric HPLC methods [6–12] are the most conventional and most frequently used methods for the quantification of 1-OHP and 2-NAP in urine. Also, to date many analysis methods using liquid chromatography–tandem mass spectrometry [13–20] have been developed for the simultaneous determination of 1-OHP and 2-NAP in urine. Synchronous fluorescence determination [21–24] has proved to be a valuable technique owing to its high level of efficiency and low sample volume requirement. Also, chemiluminescent enzyme immunoassay [25] and voltammetric
determination methods [26] have been developed. But these methods are not sensitive for the measurement of 1-OHP and 2-NAP in population urines. Gas chromatographic methods have been published for the analysis of 1-OHP and 2-NAP in human urine; these methods involve electron impact ionization-mass spectrometry (GC–EI-MS) [27,28] and use high resolution mass spectrometry (GC–HRMS) [29,30]. Campo et al. [27] described a sensitive GC–MS method to determine monohydroxy metabolites of PAHs by derivatization with a silylating agent. This method had detection limits as low as 0.5 g/L and 0.9 g/L for 1-OHP and 2-NAP, respectively. These detection limits are insufficient to monitor background concentrations of 1-OHP and 2-NAP in population urines. 2-NAP is relatively volatile, and can be vaporized during the concentration and dryness after the extraction. A treatment needs to prevent the loss of 2-NAP during drying. We had tried to develop a precise and sensitive detection method of these compounds by using gas chromatograph/mass spectrometry (GC/MS) after the silylation with methyl-N-(tertbutyldimethylsilyl) trifluoroacetamide (MTBDMSTFA). The analytical method was applied to the quantification of 1-OHP and 2-NAP in population urines. 2. Experimental 2.1. Chemicals and reagents
∗ Corresponding author. Tel.: +82 41 850 8811; fax: +82 41 850 8998. E-mail address: [email protected] (H.-S. Shin). 1570-0232/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jchromb.2011.01.009
2-NAP (>99%), 1-OHP (98%), 1-OHP-d9 (99%), naphthalene (99%), pyrene (98%), pyrene-10 (99%), sodium acetate,
490
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
Table 1 The source of the biological samples. Sample no.
Age
Gender
Occupational exposure
Smoking status
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
59 68 52 41 21 57 62 65 52 31 62 43 54 33 36 57 57 54 67 21
Female Male Female Female Male Female Female Female Female Male Female Male Female Female Female Male Male Female Female Male
No No No No No No No No No No No No No No No No No No No No
S S S N S N N S N N N N N N S N N S N N
for 1 min, raised to 320 ◦ C at 20 ◦ C/min and held for 5 min. All mass spectra were obtained with an Agilent 5975 B instrument. The ion source was operated in the electron ionization mode (EI; 70 eV, 230 ◦ C). Full-scan mass spectra (m/z 50–800) were recorded for the identification of analytes at high concentration. Confirmation of trace chemicals was completed by three MS characteristic ions, the ratio of three MS characteristic ions and GC-retention time match to the known standard compound. The ions selected in this study were m/z 185, 201, 216 for TMS-2-NAP, m/z 259, 275, 290 for TMS1-HOP, m/z 267, 284, 299 for 1-HOP-d9-TMS, m/z 185, 201, 258 for TBDMS-2-NAP, m/z 259, 275, 332 for TBDMS-1-HOP, and m/z 267, 284, 341 for TBDMS-1-HOP-d9. 2.5. Calibration and quantification Calibration curves for 2-NAP and 1-HOP were established by extraction after adding 0, 2, 10, 20, 40, 100 and 200 ng of standards and 5.0 ng of internal standard in 2 mL of blank urine. 1-HOP-d9 was used as the internal standard. The ratio of the peak area of standard to that of internal standard was used in the quantification of the compound.
S, smoker; N, nonsmoker.
2.6. Extraction and determination of PAHs in blood -glucuronidase/aryl sulfatase, N-methyl-N-trimethylsilyl trifluoroacetamide (MSTFA) and N-methyl-N-(tert-butyldimethylsilyl) trifluoroacetamide (MTBDMSTFA) were purchased from Sigma (St. Louis, MO, USA), and n-pentane and acetone purchased from J.T. Baker (USA) were used as solvents. Water was purified in milli-Q (Millipore Corp., Milford, MA). Spot urine and blood samples were obtained from volunteers. The source of the urine and blood samples was described in detail in Table 1. Blank urine, in which the analytes were not detected with our method, was used. Solutions of standards were prepared at 1000 mg/L in methanol: methylene chloride (1:1) and were stored at −20 ◦ C. Working solutions were used after 500 times or more dilution of the standard solutions using methanol and prepared newly before the experiment. 2.2. Spiking Spiked samples of human urine (2.0 mL) were prepared with 20–100 L of standard solutions at concentrations of 0.1–1.0 mg/L and with 50 L of the internal standard solution at a concentration of 0.1 mg/L. 2.3. Enzyme hydrolysis, extraction and derivatization procedure A 2.0 mL sample of urine was placed in a 20-mL test tube. About 2.0 mL of sodium acetate buffer solution and 10 L of glucuronidase/aryl sulfatase were added to the sample solution, and the sample was incubated for 3 h at 37 ◦ C. 50 L of 1-HOP-d9 solution (0.1 mg/L in acetone) was added to the solution and the sample was extracted with 5.0 mL of pentane by mechanical shaking for 20 min. 50 L MTBDMSTFA was added to the solution, and the solution was evaporated in vacuum rotary to about 0.1 mL, and then almost dried with a nitrogen stream. The dry residue was dissolved with 50 L of MTBDMSTFA and the tubes were heated for 30 min at 60 ◦ C. A 2 L sample of the solution was injected into the GC system. 2.4. Gas chromatography–mass spectrometry The gas chromatograph used was an Agilent 7890 A with a split/splitless injector. The analytical column was a 30 m HP-5MS column (cross-linked 5% phenylmethylsilicon, 0.2 mm I.D. × 0.25 m F.T). Oven temperature program began at 80 ◦ C, held
A 0.5 mL sample of blood was placed in a 20-mL test tube. 50 L of pyrene-d10 solution (2.0 mg/L in acetone) was added to the solution and the sample was extracted with 5.0 mL of pentane by mechanical shaking for 20 min. The extract was evaporated in vacuum rotary and almost dried with a nitrogen stream. The dry residue was dissolved with 50 L of methanol. At appropriate times, a 2 L sample of the solution was injected into the GC system. Calibration curves for naphthalene and pyrene were established by extraction after adding 0, 0.1, 0.5, 1.0, 2.0, 5.0 and 10 ng of standards and 50 L of pyrene-d10 solution (2.0 mg/L in acetone) in 0.5 mL of blank blood. Pyrene-d10 was used as the internal standard. The ratio of the peak area of standard to that of internal standard was used in the quantification of the compound. The gas chromatograph used was an Agilent 7890 A with a split injector (split ratio = 5:1). The analytical column was a 30 m HP-5MS column (cross-linked 5% phenylmethylsilicon, 0.2 mm I.D. × 0.25 m F.T). Oven temperature program began at 50 ◦ C, held for 1 min, raised to 300 ◦ C at 20 ◦ C/min and held for 10 min. All mass spectra were obtained with an Agilent 5975 B instrument. The ion source was operated in the electron ionization mode (EI; 70 eV, 230 ◦ C). Full-scan mass spectra (m/z 50–800) were recorded for the identification of analytes at high concentration. Confirmation of trace chemicals was completed by three MS characteristic ions, the ratio of three MS characteristic ions and GC-retention time match to the known standard compound. The ions selected by SIM were m/z 128, 129 and 127 for naphthalene; m/z 202, 200 and 203 for pyrene and m/z 212 for pyrene-d10 (internal standard). 3. Results and discussion 3.1. Chromatography For the GC separation of TBDMS-1-OHP and TBDMS-2-NAP, the use of the nonpolar stationary phase was found to be efficient. The peaks of TBDMS-1-OHP, TBDMS-1-OHP-d9 and TBDMS-2-NAP were symmetrical and separation of the analytes from the background compounds in urine extract was more easily observed when compared to TMS-1-OHP, TMS-1-OHP-d9 and TMS-2-NAP (Fig. 1). The retention times of TBDMS-2-NAP, TBDMS-1-OHP-d9 and TBDMS-1-OHP were 7.8, 11.9 and 12.0 min, respectively. In a chromatogram of blank urine, there were no extraneous peaks observed in the time range of between 7.0 and 13.0 min.
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
491
Fig. 1. Chromatograms of standard (A and B) and volunteer sample (C). The standard solution was spiked to a concentration of 4.0 g/L for 2-NAP, 1 g/L for 1-HOP and 1.0 g/L for 1-HOP-d9 in blank urine. Sample was quantified as the concentration of 1.6 g/L for 2-NAP and 0.29 g/L for 1-HOP.
3.2. Mass spectrometry Mass spectra analyses of TBDMS-2-NAP, TBDMS-1-HOP and TBDMS-1-HOP-d9 were obtained (Fig. 2). TBDMS-2-NAP showed molecular ion at m/z 258 and the base peak at m/z 201 that cleavages t-butyl group from molecular ion and characteristic ions at m/z 127, 146 and 186. The ions at 186 and 146 were from the losses of one methyl group or [2CH3 + Si–3H] group from m/z 201 ion, and the
ion at 127 was a result of the loss of [TBDMSO] group from molecular ion. TBDMS-1-HOP revealed molecular ion at m/z 332 and the base ion at m/z 275, which was due to the loss of t-butyl group from molecular ion, and characteristic ions at m/z 217, 244 and 259. The ions at 259 and 244 were the result of the losses of [CH3 + H] or [2CH3 + H] from m/z 275 ion, and the ion at 217 was from the loss of [TBDMS] group from molecular ion. TBDMS-1-HOP-d9 showed molecular and base ion at m/z 341 and m/z 284, which is due to the
492
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
Fig. 2. Mass spectra of TBDMS-2-NAP (A), TBDMS-1-HOP (B) and TBDMS-1-HOP-d9 (C).
loss of t-butyl group from molecular ion, and characteristic ions at m/z 226, 252 and 267. The ions at 267 and 252 were the result of the losses of [CH3 + D] or [2CH3 + D] from m/z 284 ion, and the ion at 226 was due to the loss of [TBDMS] group from molecular ion. Selected ions for the quantization were m/z 258 for TBDMS-2-NAP, m/z 332 for TBDMS-1-HOP, and m/z 341 for TBDMS-1-HOP-d9.
3.3. Derivatization 1-OHP and 2-NAP have a polar hydroxy group, which can be silylated in displacement reaction. The silylating reagents, MSTFA and MTBDMSTFA, were compared to each other in terms of repeatability, sensitivity and stability of the derivatives.
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
493
Table 2 Within run-precision and accuracy of target compounds in urine (n = 5). Compound
Amount spiked (g/L)
Results (g/L) Calculated conc
Mean ± SD (RSD%)
Accuracy (%)
1-OHP
0.5 10.0 25.0
0.49, 0.52, 0.53, 0.52, 0.54 9.7, 9.8, 9.8 9.6, 9.7 24.5, 24.4, 23.7, 23.7, 24.5
0.52 ± 0.017 (3.3%) 9.7 ± 0.1 (0.7%) 24.2 ± 0.4 (1.7%)
104 97.0 96.8
2-NAP
0.5 10.0 25.0
0.51, 0.48, 0.38, 0.51, 0.50 12.5, 10.9, 11.3, 12.8, 7.1 24.8, 31.6, 19.0, 27.5, 21.8
0.48 ± 0.06 (12.3%) 10.9 ± 1.5 (14.1%) 24.9 ± 4 (14.8%)
96.0 109 99.6
SD, standard deviation; RSD, relative standard deviation (%).
2-NAP is fairly volatile and can be vaporized during concentration and drying. We tested the usability of MSTFA or MTBDMSTFA as a keeper to prevent the loss. After 20 ng of 2-NAP was spiked with 5 mL of pentane (n = 3), MSTFA or MTBDMSTFA was added in pentane solution. The mixture was shaken for 10 min mechanically, the solvent was dried completely. The residue was once more derivatized for 30 min at 60 ◦ C with 50 L of MSTFA or MTBDMSTFA. As a result, mean ± standard deviation of the area of the derivatives was 19,936 ± 16,347 (n = 3) without MTBDMSTFA and 26,531 ± 1881 with MTBDMSTFA, and 20,642 ± 15,430 (n = 3) without MSTFA and 25,648 ± 5137 with MSTFA. The recovery, that is the ratios between the area of 2-NAP detected after drying with and without MTBDMSTFA and MSTFA, were 1.33 and 1.24, respectively. 2-NAP-TBDMS and 2-NAP-TMS formed from this treatment turn out to be less volatile. Then these derivatives were detected in high peak area with small standard deviation. The derivatives were analyzed at reaction times of 30, 60, 90 and 120 min. The reaction rate of 1-OHP and 2-NAP with MSTFA and MTBDMSTFA was determined by detecting peak area of the products. 1-OHP and 2-NAP showed very rapid and complete reaction with both MSTFA and MTBDMSTFA. TBDMS-1-OHP, TBDMS-1OHP-d9 and TBDMS-2-NAP were detected as the products of 1-OHP, 1-OHP-d9 and 2-NAP with MTBDMSTFA in about 30 min at 60 ◦ C. To evaluate the stability of the derivatives, the experiment was repeated three times and the same samples were analyzed after 100 h. The stability was evaluated as area percentage of the derivatives obtained on the first day of analysis to that of the derivatives reacted newly after 100 h. The mean percentual stability after 100 h varied from 5.65% (1-OHP) to 7.3% (2-NAP). Otherwise, that of the TMS derivatives after 100 h varied from 11.5% (1-OHP) to 13.7% (2-NAP). As a result, TBDMS-1-OHP and TBDMS-2-NAP derivatized by MTBDMSTFA showed superior repeatability and stability compared to those derivatized by MSTFA.
3.4. Validation of the assay Despite the improvement of other alternative extraction techniques, liquid-liquid extraction (LLE) is still an efficient technique for the routinely performed analysis of 1-OHP and 2-NAP in urine. LLE was performed for the analysis of 1-OHP and 2-NAP in urine. Test samples at 10.0 and 25.0 g/L in urine were prepared and the relative recovery was calculated by percentage of the analytes recovered. The recoveries of the test compounds were about 107% for 2-NAP and about 97% for 1-OHP. Using the least-squares fit technique, an examination of typical standard curve was performed by computing a regression line of peak area ratios for 1-OHP and 2-NAP to the internal standard (1-OHP-d9) on the concentration. This examination demonstrated a linear relationship with correlation coefficients being greater than 0.990. Linear equations were y = 8.5142x + 0.0131 for 2-NAP and y = 7.9504x + 0.0068 for 1-HOP in the concentration range of 0.1–50.0 g/L. The method detection limit (MDL) was defined by 3.14 times of standard deviation for replicate determinations (n = 7) from samples spiked at the concentration of 0.01 g/L in blank urine [31], in which the concentration is a value that corresponds to an instrument signal/noise ratio within the range of 2.5. MDL was calculated 0.07 g/L for 2-NAP and 0.01 g/L for 1-HOP based upon an assayed sample of 2.0 mL. The range and standard deviation values for the within run-precision and accuracy are given in Table 2. For five independent determinations at 0.5, 10 and 25.0 g/L, the coefficient of variation was less than 15%. The construction of typical standard curves of naphthalene and pyrene was performed by above method. Linear equations were y = 0.3394x + 0.0011 for naphthalene and y = 2.1317x + 0.0238 for pyrene in the concentration range of 0.01–50 g/L. MDLs of naphthalene and pyrene were calculated as 0.01 g/L in plasma.
Fig. 3. Analytical results of 2-NAP and 1-HOP in 20 volunteer samples.
Naphthalene Conc. (µg/L)
494
H.-S. Shin, H.-H. Lim / J. Chromatogr. B 879 (2011) 489–494
60
4. Conclusions
y = 1.337x + 6.7612 R² = 0.2138
50 40 30 20 10 0
0
2
4
6
8
10
2-NAP Conc. (µg/L)
12
14
16
Fig. 4. Relationship of urinary 2-NAP concentration and the naphthalene concentration in plasma.
Pyrene Conc. (µg/L)
0.9
y = 0.9029x - 0.0072 R² = 0.6757
0.8
References
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 -0.1 0
For the enhancement of GC performance and sensitivity, the silylating reagents, MSTFA and MTBDMSTFA were compared to each other in terms of repeatability and stability of the derivatives. The result showed that TBDMS-1-OHP and TBDMS-2-NAP derivatized by MTBDMSTFA showed superior repeatability and stability compared to those derivatized by MSTFA. The peaks of TBDMS-2-NAP and TBDMS-1-HOP have good chromatographic properties by using HP-5MS column and show sensitive response for the GC–MS (SIM). Method detection limits of target compounds were 0.07 g/L for 2-NAP and 0.01 g/L for 1-OHP. Urine samples from 20 volunteers were analyzed by the developed method. The results showed that urinary 2-NAP and 1-HOP showed good correlation with the concentrations of naphthalene and pyrene in blood, respectively. The developed method was shown to be a useful method in the identification of human exposure to PAHs.
0.2
0.4
0.6
0.8
1
1-HOP Conc. (µg/L) Fig. 5. Relationship of urinary 1-HOP concentration and the pyrene concentration in plasma.
3.5. Sample analysis We analyzed the target compounds in 20 volunteer samples. 2-NAP and 1-OHP were found in the concentration range <0.07–13.7 g/L (2-NAP) and <0.01–0.88 g/L (1-OHP) and above the LOD in 12/20 and 8/20 samples (Fig. 3). Confirmation of trace 2-NAP and 1-HOP was performed by the ratio of two MS characteristic ions and GC-retention time match to the known standard compound. Otherwise, naphthalene and pyrene in plasma were found in the concentration range of <0.01–23.95 g/L and <0.01–0.75 g/L, respectively. The relationship of urinary metabolites concentrations and plasma PAHs concentrations was constructed. As shown in Fig. 4, urinary 2-NAP was found to be well correlated with the concentration of naphthalene in plasma (r2 = 0.21, P = 0.002). Also, urinary 1-HOP showed good correlation with the concentration of pyrene in plasma (r2 = 0.68, P = 0.013) (Fig. 5). It appeared that urinary 1HOP or 2-NAP is good indicator of naphthalene or pyrene levels exposed.
[1] B. Marczynski, R. Preuss, T. Mensing, J. Angerer, A. Seidel, A.E. Mourabit, M. Wilhelm, T. Brüning, Int. Arch. Occup. Environ. Health 78 (2005) 97. [2] International Program on Chemical Safety (IPCS), Environmental Health Criteria 202. Selected Non-heterocylic Polycyclic Aromatic Hydrocarbons, WHO, Geneva, 1998. [3] Polycyclic aromatic hydrocarbons—occurrence in foods, dietary exposure and health effects, European Commission, Scientific Committee on Food. December 4, 2002. http://ec.europa.eu/food/fs/sc/scf/out154 en.pdf. [4] F.J. Jongeneelen, R.B.M. Anzion, P.T.J. Scheepers, R.P. Bos, P.T. Henderson, E.H. Nijenhuis, S.J. Veenstra, R.M. Brouns, A. Winkers, Ann. Occup. Hyg. 32 (1988) 35. [5] M. Yang, M. Koga, T. Katoh, T. Kawamoto, Arch. Environ. Contam. Toxicol. 36 (1) (1999) 99. [6] T. Chetiyanukornkul, A. Toriba, R. Kizu, J. Chromatogr. A 961 (1) (2002) 107. [7] C.-T. Kuo, H.-W. Chen, J.-L. Chen, J. Chromatogr. B 805 (2) (2004) 187. [8] E. Eivor, Arch. Toxicol. 77 (4) (2003) 183. [9] H. Li, R.I. Krieger, X. LiQ, Sci. Total Environ. 257 (2000) 147. [10] C. Xavier, B. Maryline, R. Miguel, J. Sep. Sci. 32 (20) (2009) 3403. [11] S.M. Ramasamy, R.J. Hurtubise, A. Weston, Appl. Spectrosc. 51 (9) (1997) 1377. [12] R. Preuss, J. Angerer, J. Chromatogr. B 801 (2) (2004) 307. [13] L. Xu, D.C. Spink, J. Chromatogr. B 855 (2) (2007) 159. [14] D. Pigini, A.M. Cialdella, P. Faranda, Rapid Commun. Mass Spectrom. 20 (6) (2006) 1013. [15] R. Fan, Y. Dong, W. Zhang, J. Chromatogr. B 836 (2006) 92. [16] K. Kakimoto, A. Toriba, T. Ohno, J. Chromatogr. B 867 (2) (2008) 259. [17] S. Ferrari, F. Mandel, J.D. Berset, Chemosphere 47 (2) (2002) 173. [18] L. Yinghe, L.C. Austin, S. Haihong, Rapid Commun. Mass Spectrom. 19 (22) (2005) 3331. [19] A. Holm, P. Molander, E. Lundanes, J. Chromatogr. B 794 (1) (2003) 175. [20] P. Jacob III, M. Wilson, N.L. Benowitz, Anal. Chem. 79 (2) (2007) 587. [21] L. Michel, B. Rachid, F. de Violet Philippe, Anal. Bioanal. Chem. 376 (6) (2003) 816. [22] R.H. Kang, Y.S. Wang, H.M. Yang, Anal. Chim. Acta 658 (2) (2010) 180. [23] Y. Zhang, Y. Zhu, S. Wang, Bull. Environ. Contam. Toxicol. 69 (4) (2002) 494. [24] H.M. Yang, Y.S. Wang, J.H. Li, Anal. Chim. Acta 636 (1) (2009) 51. [25] R. Aldo, P. Simoni, M. Mirasoli, Anal. Bioanal. Chem. 372 (2002) 751. [26] K. Nicole, H. Kevin, C. Emsp, Electroanalysis 17 (7) (2005) 571. [27] L. Campo, F. Rossella, S. Fustinoni, J. Chromatogr. B 875 (2008) 531. [28] K. Desmet, B. Tienpont, P. Sandra, Chromatographia 57 (2003) 681. [29] Z. Li, L.C. Romanoff, D.A. Trinidad, Anal. Chem. 78 (16) (2006) 5744. [30] C.J. Smith, C.J. Walcott, W. Huang, J. Chromatogr. B 778 (2002) 157. [31] USEPA Method SW 846, Quality Control, 1992, p. 26 (Chapter 1).