Determination of volatile N-nitrosamines by gas chromatography--mass spectrometry with positive-ion chemical ionization/Gaasikromatograafiline massispektromeetria meetod N-nitrosoamiinide maaramiseks positiivse iooni keemilise ionisatsiooni abil.
N-nitrosamines (NA) are known as a widely acting group of animal carcinogens and are found, usually at trace levels, in diverse places in the environment. Volatile NA may be also formed in certain foods from naturally occurring amines present in them. These amines are nitrosated by agents derived from added nitrite or nitrogen oxides . Nitrosation reactions can be influenced by the presence of nitrosation inhibitors (redox compounds such as ascorbate, vitamin E, and alcohols) and catalysts (metal ions, phenols, nucleophilic anions such as [Cl.sup.-], [J.sup.-], [Br.sup.-], SC[N.sup.-], carbonyl compounds, and surface-active agents) . NA can be formed in food, water, air, and human organism. They constitute a large group of genotoxic chemical carcinogens, which occur in human diet and other environmental media, and can be formed endogenously in the human body.
Table 1 presents data on the toxicity and carcinogenicity of five volatile NA (NDMA = N-nitrosodimethylamine, NDEA = N-nitrosodiethylamine, NPYR = N-nitrosopyrrolidine, NPIP = N-nitrosopiperidine, NDBA = N-nitrosodibuthylamine) from the literature [2, 3].
Limit concentrations for two NA have been established in Estonia: the maximum allowed content of summed NDMA and NDEA in meat and meat products is 2 [micro]g/kg; in smoked meat products 4 [micro]g/kg .
NA may be formed from nitrate or nitrite used as preservatives in food, or from nitrogen oxides in combustion gases when direct heating, smoking, or drying processes are used. The concentration of NA depends on the cooking method, frying temperature and time, concentration of NA precursor, possibly smoking procedures, and ascorbate concentration .
Various methods have been used for the determination of volatile NA in diverse food matrixes: beer [6, 7], groundwaters and drinking waters [8, 9], fried bacon , cheeses , and nonfat dry milk . Sample preparation methods used are based on vacuum distillation , steam distillation [14, 15], mineral oil distillation , solid-phase extraction (SPE) [17, 18], and liquid-liquid extraction . Fidder & Pensabene  used a supercritical fluid extraction (SFE) method to analyse 10 different NA in fried bacon. The results were compared with mineral oil distillation, low-temperature vacuum distillation, and SPE methods. Analysis of NPYR data for different sample preparation methods used (SFE, SPE, low-temperature vacuum distillation, and mineral oil distillation) showed significant variation .
Analytical methods to detect volatile NA are based on gas chromatography (GC) with a thermal energy analyser (TEA) [16, 21, 22], with mass spectrometry (MS) by electron impact ionization (EI) [23-25], and with mass spectrometry by chemical ionization [26-29].
Planar chromatography , high pressure liquid chromatography (HPLC), flame thermionic detection, colorimetric detection , fluorescence determination , and fluorescence and chemiluminescence  are methods used less often to detect volatile NA in food.
The task of this study was to work out a rapid and efficient GC-MS method for the determination of volatile NA in food products. Sample cleaning optimization and selection of the most suitable reagent gas for positive-ion chemical ionization (PCI) were dealt with.
For sample cleaning SPE was used. SPE sorbents are available in a wide range of different surfaces, pore size, particle size, and base support. Which sorbent is optimal for extraction depends upon the properties of the NA, the sample/matrix composition, and other factors that are discussed below. To get the proper SPE product one has to take into account the mode of NA retention, sorbent chemistry, mass, and physical configuration.
Two SPE steps with Extrelut (Kieselguhr) and Florisil (85% Si[O.sub.2], 15.5% MgO, 0.5% [Na.sub.2]S[O.sub.4]) sorbents were used. The food sample (6.0 [+ or -] 1.0 g in 100 mL glass beaker) was mixed with 0.1 N NaOH (6 mL).
As the first step, about 6 g of Extrelut was placed at the bottom of the glass column (30 cm x 1.5 cm) and wetted with 20 mL hexane/dichloromethane 40 : 60 (v : v). After that the sample was eluted with two 20-mL portions of hexane/ dichloromethane solution. The eluate was collected in a 50-mL concentrator flask and evaporated in water bath at 60[degrees]C. As the second step, about 1 g Florisil was placed at the bottom of the Florisil cartridge (6.5 cm x 1.3 cm), wetted with 6 mL 95 : 5 (v : v) dichloromethane/methanol solution, and eluted with 6 mL dichloromethane/ methanol solution. The solution was evaporated at 60[degrees]C to 1 mL. Then 200 [micro]L of internal standard (NDPA, 1 [micro]g/mL) was added. The prepared solution was transferred to the GC injector vial. Extractions were performed in duplicate .
Methanol (J. T. Baker, Holland), dichloromethane (Sigma-Aldrich, USA), hexane (Rathburn, Scotland), and 0.1 N NaOH solution (Chemapo, Czechoslovakia) were used. As internal standard NDPA solution 100 [micro]g/mL in methanol from Ultra Scientific was used. NA mixtures in methanol were commercial products from Sigma except NDMA, which was from Aldrich. Extrelut was from Merck, Germany; Florisil 100/200 from Alltech, Belgium.
The following volatile NA were analysed: NDMA, NDEA, NDBA, NPYR, and NPIP.
1. [R.sup.1] = [R.sup.2] = C[H.sup.3] NDMA
2. [R.sup.1] = [R.sup.2] = C[H.sub.2]C[H.sub.3] NDEA
3. [R.sup.1] = [R.sup.2] = [(C[H.sub.2]).sub.2]C[H.sub.3] NDPA, internal standard
4. [R.sup.1] = [R.sup.2] = [(C[H.sub.2]).sub.3]C[H.sub.3] NDBA
[FORMULA NOT REPRODUCIBLE IN ASCII]
Gas chromatography--mass spectrometry determination
Analyses were carried out using a HP (Hewlett-Packard) 6890 Plus GC gas chromatograph with (MSD)--HP 5973 MSD mass selective detector.
Sample portions (2 [micro]L) were injected into a chromatograph column (30 m HP-1701 MS; 0.25 mm i.d., 0.25 [micro]m film thickness) containing 14% cyanopropylphenyl and 86% methyl polysiloxane. GC oven programme was set at 55[degrees]C for 2 min and then raised to 260[degrees]C at a rate of 10[degrees]C/min and held at the final temperature for 2 min; the velocity of He carrier gas (99.99%) was 1 mL/min. Oven conditions for GC and mass selective detector settings are given in Table 2.
PCI MS option was used in the selected ion-monitoring mode (SIM) with two reagent gases--methane and ammonia.
Mass spectra for NA were registered in SCAN and SIM modes. (SCAN refers to the sequence of control over operating parameters of a mass spectrometer that results in a spectrum of masses, velocities, momenta, or energies; SIM refers to the use of the instrument to record the ion current at selected masses characteristic of the compound of interest in an expected retention time window ). For sensitivity optimization the temperature of the ion source and the reagent gas flow rate were varied. The method may be further optimized for even more rapid analysis.
Recovery is the proportion of the amount of analyte present in or added to the analytical portion of the test material that is extracted and presented for measurement. In this work recoveries of five volatile NA were investigated by a standard addition experiment. The meat product was fortified with appropriate volumes of standard solutions in methanol to get recoveries at the levels 0.1-1 ppb. We used two SPE steps--Extrelut and Florisil. The sample was analysed by GC-MS and recovery was calculated by Eq. 1 .
Rec, % = [([C.sub.1] - [C.sub.2])/[C.sub.3]] x 100, (1)
where [C.sub.1] is the concentration determined in the fortified sample, [C.sub.2] is the concentration determined in the unfortified sample, and [C.sub.3] is the concentration of fortification.
The concentration of NA, [C.sub.NA] ([micro]g/mL), detected by GC, was calculated using the following equation :
[C.sub.NA] = [C.sub.NDPA] ([A.sub.NA]/[A.sub.NDPA]), (2)
where [C.sub.NDPA] is the concentration of the internal standard (NDPA) solution in the sample, [A.sub.NA] designates the area of NA peak, and [A.sub.NDPA] is the area of the internal standard peak.
The content of NA, X ([micro]g/kg), in a sample was calculated as follows :
X = [10.sup.5] [C.sub.NA] V/m Rec, (3)
where [C.sub.NA] stands for NA concentration by GC detection, m is the weight of the sample (g), V is the volume of the final concentrated extract (mL), and Rec is the recovery of NA (%). The peaks were integrated by a Hewlett-Packard 6890 Gas Chromatograph.
The ratio of the chromatogram peak area of the NA) ([A.sub.NA] to the peak area of the internal standard) ([A.sub.NDPA] in the spiked bacon sample was calculated. The linear relationship of this ratio to the concentration ratio is shown in Fig. 1. This calibration curve enables to calculate the NA concentration using the GC-MS peak area measurements. The correlation coefficient [R.sup.2] = 0.9998 for NDMA and [R.sup.2] = 0.9986 for NDEA.
[FIGURE 1 OMITTED]
Limit of detection of NA
Limit of detection (LOD) is the threshold concentration below which identification is unreliable .
The value of LOD is given by the equation:
LOD = [X.sub.bl] + K [S.sub.bl], (4)
where [X.sub.bl] is the mean of the blank measures, [S.sub.bl] is the standard deviation of the blank measures, and K is a numerical factor chosen according to the confidence level desired .
The most widely used method for the determination of volatile NA in meat products has been GC with different detectors. These detectors include a flame ionization detector , Coulson electrolytic detector , thermal energy analyser (TEA) [17, 19], and mass selective detector (MSD) [26, 28].
Prest & Hermann  developed a GC-MS method with a MSD by PCI using ammonia reagent gas and large volume injection (50 [micro]L) for the determination of NDMA in water. This method provides the lowest values for NA detection limit reported in the literature (1-3 ng/L). In this work the same temperature program was used as in the optimized method in  except the initial temperature, which was 55[degrees]C in our study to minimize the changes of analyte concentrations through possible solvent and matrix effects.
Thermal Energy Analyzer as a selective detector for N-nitroso compounds has been succesfully interfaced to gas  and liquid  chromatographs. In these publications the authors evaluated LOD for GC as 0.3 ppb and for LC, 1 ppb.
Unfortunately, the authors do not mention what kind of NA they studied. With a Coulson electrolytic detector LOD of 0.1-5 ppb was reached .
In this work five volatile NA were separated by GC with MSD and recognized by PCI using two different reagent gases, ammonia and methane. The sensitivity for NA increased and LOD was better when ammonia was used as reagent gas in the PCI experiment (Table 3). LOD was determined as the mean value of six independent measurements for each volatile NA. PCI with ammonia as reagent gas was superior to EI ionization for volatile NA in most cases giving up to [10.sup.3] times lower LOD, except for NPYR and NDMA, where the LOD values are almost equal.
Electron impact ionization
Ionization is achieved through the interaction of an analyte with an energetic electron beam. This results in the loss of an electron from the analyte and production of a radical cation. As samples are thermally desorbed to the gas phase and subjected to the high energy of EI, analytes must be both thermally and energetically stable. EI ionizing energy used is 70 eV to promote greatest sensitivity and to produce molecular and fragment ions used for chemical characterization and identification.
A summary of the mass spectra of the NA obtained by EI in this work is given in Table 1 in the Appendix and shown in Figs. 1 and 2 in the Appendix. The spectra of NDMA, NDEA, NDPA, NPYR, and NPIP are compared with those published in NIST Standard Reference Database  (see Table 1 in the Appendix). The mass spectrum of NDBA is not available in NIST and is compared to the spectrum published elsewhere .
Positive-ion chemical ionization
PCI is a soft ionization method developed to ionize compounds. The probability of proton exchange is based on the relative gas phase basicities (proton affinities) of the analyte and the reagent gas. Although PCI is capable of generating molecular weight information from labile species, it still requires the sample to be volatile, which could hinder the detection of thermally unstable analytes.
A summary of the mass spectra of NA studied by SIM-PCI using different reagent gases (methane, PA = 129.9 kcal/mol, and ammonia, PA = 204.0 kcal/mol ), is presented in Table 2 in the Appendix and shown in Figs. 3-6 in the Appendix.
PCI with methane as reagent gas
Table 2 in the Appendix gives simple PCI mass spectra for all NA measured, with the [(M+H).sup.+] ion accounting for almost all the ion current. Volatile NA molecules M for which proton affinity is greater than that of methane (i.e. PA(M) > PA(C[H.sub.4])) give mass spectra with (M+H)+ as the base peak, owing to protonation of the molecule, and a much smaller peak for the [(M+[C.sub.2][H.sub.5]).sup.+] adduct. Unfortunately, the proton affinities of these NA are not available in the literature.
PCI with ammonia as reagent gas
As expected, the use of ammonia as reagent gas led to lower fragmentation than observed with methane. The best sensitivity was achieved at ammonia flow of 1 mL/min. Usually, molecules where proton affinity is lower than the value for ammonia (i.e. PA(M) < PA(NH3)) give mass spectra with [(M+N[H.sub.4]).sup.+] as the base peak and a much smaller peak for the [(M+H).sup.+] adduct.
Recovery and content of volatile NA in some meat products
The recovery measurements of volatile NA in smoked meat samples are given in Table 4. Recoveries were calculated by Eq. 1. The differences in the recovery between NA can be explained most likely by different polarities, which affect their elution from extraction columns. Figure 2 presents as an example a chromatogram of a fried meat sample spiked with a standard solution containing five NA.
A combination of Extrelut-Florisil gave the highest recovery of NA compared to other cartridges that were investigated, such as Cyano, Silica, Amino and Alumino (Fig. 3).
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
Data on the content of volatile NA in some Estonian meat products measured in this work are presented in Table 5. All meat products studied were purchased from local market and stored at -20[degrees]C until analysis.
The highest NA levels were found in smoked and fried foodstuffs. The obtained results showed that the formation of NA in food is influenced by N[O.sub.2.sup.-] and N[O.sub.3.sup.-] concentrations, smoking and frying temperature regime and time, and storage of food. Addition of sodium nitrite to smoked chicken increased the exposure of NPYR. The phenols and nitrous gas in smoke and addition of nitrites may increase NA levels in smoked products. Nitrous gas in smoke accelerates nitrosation reactions . The highest level of NPYR was detected in pork fat, most likely due to its lipophilic properties.
In this work a method for separation and identification of volatile N-nitrosamines was developed. The advantages of this method are rapidity, effectiveness, and simplicity. The recovery of NA in food samples with different extraction methods is in the range 50-70%. The SPE method was found to be superior to other extraction methods used for high recovery and rapidity of analysis. The advantages of the proposed method are low time consumption for sample preparation and decreased possibility of pollution of the GC-MS system by additives.
Two different reagent gases, methane and ammonia, were tested. Methane causes higher fragmentation and can be used for NA with high molecular weight only. The required sensitivity cannot be reached for low-molecular weight NA. Ammonia as a reagent gas of relatively "soft" nature shows milder fragmentation but is a very specific reagent gas for NA. The limit of detection using positive-ion chemical ionization with ammonia was 0.01 ppb and the recovery was about 70-80%. As a conclusion, PCI with ammonia as reagent gas was the most selective technique distinguishing volatile NA on the basis of relative abundances of their adduct ions.
Five volatile NA of different molecular structures were determined in meat products. Comparison of the results of our analyses with the results obtained by other authors encourages us to use this highly sensitive method for identification and quantitative analysis of NA in addition to meat products also for other matrixes.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Table 1. Intensities (%) of the N-nitrosamine ions observed with electron impact ionization in the present work and in NIST Database (relative to the most intensive peak) N-nitrosamine Present work, NIST Database , m/z, m/z, Rel. intensity Rel. intensity (in parentheses) (in parentheses) NDMA 74(100), 42(98), 43(44) 74(100), 42(72), 43(45) NDEA 102(100), 44(25), 42(20), 102(100), 44(87), 42(83), 56(21) 56(54) NDPA 130(28), 42(65), 43(100), 130(22), 42(66), 43(100), 70(98), 101(13) 70(66), 101(13) NPYR 100 (100), 41(80), 100 (100), 41(69), 42(61), 68(18) 42(45), 68(14) NPIP 114(90), 41(25), 42(100), 114(100), 41(28), 42(94), 55(54) 55(83) NDBA 158(38), 84(100), 158(18), 84(100), 116(43), 99(36), 57(54) 116(35), 99(19), 57(70) Table 2. Intensities (%) of the N-nitrosamine ions observed with positive-ion chemical ionization with methane and ammonia used as reagent gases (relative to the most intensive peak) m/z, Rel. intensity (in parentheses) Methane [(M+[C.sub.2] [(M+[C.sub.3] N-nitros- M, [H.sub.5]) [H.sub.5]) amines g/mol [(M+H).sup.+] .sup.+] .sup.+] NDMA 74 75(100) 103(0) 115(4) NDEA 102 103(100) 131(10) 143(8) NPYR 100 101(100) 129(0) 141(0) NPIP 114 115(100) 143(7.2) 155(4.7) NDPA 130 131(100) 159(8) 171(0) NDBA 158 159(100) 187(0) 199(0) m/z, Rel. intensity (in parentheses) Ammonia N-nitros- [(M+[N.sub.2] amines [(M+H).sup.+] [(M+[NH.sub.4]).sup.+] [H.sub.7]).sup.+] NDMA 75(16.6) 92(100) 109(1.6) NDEA 103(27) 120(100) 137(0.6) NPYR 101(30.9) 118(100) 135(0.6) NPIP 115(32.6) 132(100) 149(0.6) NDPA 131(44) 148(100) 165(4.5) NDBA 159(86) 176(100) 193(1.3)
The authors are grateful to Tartu Laboratory of Health Protection Inspectorate for the possibility of making analyses.
Received 9 May 2002, in revised form 15 May 2002
[1.] Tricker, A. R. & Preussmann, R. Carcinogenic N-nitrosamines in the diet: occurrence, formation, mechanisms and carcinogenic potential. Mutat. Res., 1991, 259, 277-289.
[2.] Rubenchik, B. L. Carcinogens Formation from Nitrogen Compounds. Naukova Dumka, Kiev, 1990 (in Russian).
[3.] Fernlof, G. & Darnerud, P. O. N-nitroso compounds and precursors in food-level, intake and health effect data and evaluation of risk. Livmedelsverkets rapport, National Food Administration, Uppsala, 1996, 15, 79.
[4.] Vabariigi Valitsuse 12. jaanuari 2000. a maaruse nr 14 "Toidus lubatud saasteainete loetelu ja piirnormide toidugruppide kaupa kehtestamine" lisa 4.
[5.] Rubenchik, B. L. Nourishment, Carcinogens and Cancer. Naukova Dumka, Kiev, 1979 (in Russian).
[6.] Tricker, A. R. & Preussmann, R. Volatile and nonvolatile nitrosamines in beer. J. Cancer Res. Clin. Oncol., 1991, 117, 130-132.
[7.] Leppanen, O. & Ronkainen, P. A sensitive method for determination of volatile nitrosamines in alcoholic beverages. Chromatographia, 1982, 16, 219-223.
[8.] Tomkins, B. A. & Griest, W. H. Determinations of N-nitrosodimethylamine at part-per-trillion concentrations in contaminated groundwaters and drinking waters featuring carbon-based membrane extraction disks. Anal. Chem., 1996, 68, 2533-2540.
[9.] Jenkins, S. W., Koester, C. J., Taguchi, V. Y., Wang, D. T., Palmentier, J. F. & Hong, K. P. N-nirosodimethylamine in drinking water using a rapid, solid-phase extraction method. Environ. Sci. Pollut. Res. Int., 1995, 2(4), 207-210.
[10.] Gloria, M., Beatiz, A., Barbour, F. & Scanlan, R. A. Volatile nitrosamines in fried bacon. J. Agric. Food Chem., 1997, 45, 1816-1818.
[11.] Dellisanti, A., Cerutti, G. & Airoldi, L. Volatile N-nitrosamines in selected Italian cheeses. Bull. Environ. Contam. Toxicol., 1996, 57, 6-21.
[12.] Sen, N. P., Seaman, S. & Karpinsky, K. Determination of N-nirosodimethylamine in nonfat dry milk: collaborative study. J. Assoc. Off. Anal. Chem., 1984, 67(2), 232-236.
[13.] Telling, G. M., Bryce, T. A. & Althorpe, J. Use of vacuum distillation and gas chromatographymass spectrometry for determination of levels of volatile nitrosamines in meat products. J. Agric. Food Chem., 1971, 19(5), 937-940.
[14.] Gough, T. & Webb, K. Mass spectrometric determination of volatile N-nitrosamines after screening by the Coulson electrolytic detector. IARC. Environ. Carcinogens - Selected Methods Anal., 1980, 31, 141-150.
[15.] Zhukova, G. F., Luthelan, V. :. & Hotimchenko, S. :. Determination of N-Nitrosamines by Fluorescence and Chemiluminescence Methods. Institute of Nourishment, Moscow, 1998 (in Russian).
[16.] Fazio, T., Howard, J. W. & Havery, D. Analysis of volatile N-nitrosamines in food using flame thermionic detection and mass spectrometry confirmation. IARC. Environ. Carcinogens--Selected Methods Anal., 1980, 31, 83-96.
[17.] Raoul, S., Gremaud, E., Biaudet, H. & Tureski, R. Rapid solid-phase extraction method for the detection of volatile nitrosoamines in food. J. Agric. Food. Chem., 1997, 45, 4706-4713.
[18.] Pensabene, J. W., Fiddler, W. & Gates, R. A. Solid-phase extraction method for volatile N-nitrosamines in hams processed with elastic rubber netting. J. AOAC Int., 1992, 75(3), 438-442.
[19.] Sen, N. P. & Seaman, S. S. A rapid liquid-liquid extraction method for the determination of volatile N-nitrosamines in cooked-out bacon fat. J. Agric. Food Chem., 1982, 30(2), 364-367.
[20.] Fiddler, W. & Pensabene, J. W. Supercritical fluid extraction of volatile N-nitrosamines in fried bacon and its drippings: method of comparison. J. AOAC Int., 1996, 79(4), 895-901.
[21.] Havery, D. C., Hotchkiss, J. H. & Fazio, T. Rapid determination of volatile nitrosamines in nonfat dry milk. J. Dairy Sci., 1982, 65(2), 182-185.
[22.] Ostedahl, B. G. Volatile nitrosamines in foods on the Swedish market and estimation of their daily intake. Food Addit. Contam., 1988, 5(4), 587-595.
[23.] CD-ROM (Method 8070 A) Nitrosamines by Gas Chromatography. 1996.
[24.] Pensabene, J. W., Fiddler, W., Dooley, C. J., Doerr, R. C. & Wassermann, A. E. Spectral and gas chromatographic characteristics of some N-nitrosamines. J. Agric. Food Chem., 1972, 20(2), 274-277.
[25.] Pokrovsky, A. A., Kostyukovsky, Ya. L., Melamed, D. B. & Medvedev, F. A. Determination of N-nitrosamines in meat and meat products. Vopr. pitan., 1977, 2, 65-72 (in Russian).
[26.] Fang, Y., Ding, J. & Liu, S. Chemical ionization mass spectrometry of nitrosamines. Talanta, 1984, 31(8), 619-620.
[27.] Fish, R. H., Holmstead, R. L. & Gaffield, W. Chemical ionization mass spectrometry of N-nitrosamines. Tetrahedron, 1976, 32(22), 2689-2692.
[28.] Gaffield, W. Chemical ionization mass spectrometry of nitrosamines. Environ. N-nitroso Comp. Anal. Form., 1976, 14, 11-20.
[29.] Prest, H. P. & Hermann, R. E. An approach to the determination of N-nitrosodimethylamine at part-per-quadrillion levels using positive chemical ionization and large-volume injection. Hewlett-Packard Company, Application Note. 1999, 23, 5968-5747E (www.hp.com/go/chemsupplies).
[30.] Zhukova, G. F., Pimenova, V. V. & Medvedev, F. A. Use of thin-layer chromatography on polyamide for fluorescence determination of N-nitrosamines. Zh. anal. khim., 1979, XXXIV(8), 1621-1625 (in Russian).
[31.] Bellec, G., Cauvin, M. C., Calve, K. L., Dreano, Y., Gouerou, H., Menez, J. F. & Berthou, F. Analysis of N-nitrosamines by high-performance liquid chromatography with post-column photohydrolysis and colorimetric detection. J. Chromat. A, 1996, 727, 83-92.
[32.] Gracheva, I. N., Zhukova, G. F., Kovelman, I. R., Pimenova, V. V. & Tochilkin, A. I. Fluorescence determination of carcinogenic N-nitrosamines with N-aziridine. Zh. anal. khim., 1986, XLI(2), 356-359 (in Russian).
[33.] Kitson, F., Larsen, B. & McEwen, C. Gas Chromatography and Mass Spectrometry. Academic Press, California, 1996.
[34.] The Fitness for Purpose of Analytical Methods. A Laboratory Guide to Method Validation and Related Topics. First Internet version, 1998.
[35.] Eisenbrand, G., Ellen, G., Preussmann, R., Schuller, P., Spiegelhalder, B., Stephany, R. & Webb, K. Environmental carcinogens; selected methods of analysis, Vol. 6. N-Nitroso compounds. IARC Sci. Publ., 1983, 45, 181.
[36.] NMKL Procedure No. 4 "Validation of chemical analytical methods". The Nordic Committee of Food Analysis, Espoo, 1996, 1-24.
[37.] Bartmess, J. E. In NIST Chemistry WebBook, Nist Standard Reference Database Number 69 (Mallard, W. G., & Linstrom, P. J., eds.). National Institute of Standards and Technology, Gaitheburg MD, February 2000, 20899 (http://webbook.nist.gov).
[38.] Libbey, L. & Scanlan, R. Mass spectrometry of N-nitrosamines. J. Assoc. Off. Anal. Chem., 1981, 64, 1255-1256.
Sergei Jurtchenko (a) *, Toomas Tenno (a), Uldo Molder (b), Mari Reinik (c)
(a) Institute of Physical Chemistry, University of Tartu, Jakobi 2, 51014 Tartu, Estonia
(b) Institute of Chemical Physics, University of Tartu, Jakobi 2, 51014 Tartu, Estonia
(c) Health Protection Inspectorate, Tartu Laboratory, Pollu 1a, 50002 Tartu, Estonia
* Corresponding author, firstname.lastname@example.org
Table 1. Half lethal dose ([LD.sub.50]) and tumorgenic dose rate 50 ([TD.sub.50]) values of volatile N-nitrosamines [2, 3] Compound [LD.sub.50], [TD.sub.50], [micro]g/kg mg/kg body weight/day NDMA 40 0.01 NDEA 280 0.11 NPYR 900 2.10 NPIP 200 1.30 NDBA 1200 0.69 Table 2. Positive-ion chemical ionization with ammonia as reagent gas Injector Injection volume 2.0 [micro]L Columns Detector MSD Outlet pressure 54.8 kPa Flow (He carrier gas) 1.0 mL/min Average velocity 37 cm/s Inlets Heater 220[degrees]C Pressure 54.7 kPa Total flow 33.8 mL/min Injection pulse Pressure 90.0 kPa, 1.40 min Purge flow to split vent 30.0 mL/min, 1.20 min Gas saver 15.0 mL/min, 2.00 min Oven Heating time at intial temperature 2.00 min 55[degrees]C Heating time at 260[degrees]C 3.00 min Solvent washes Heptane MSD parameters Tune file PCINH3.U Acquistion mode SIM Run time 25.50 min Table 3. Limit of detection (ppb) reached for volatile N-nitrosamines with positive-ion chemical ionization (PCI) and electron impact ionization (EI) Compound * PCI with PCI with EI methane ammonia NDMA 0.5 0.1 10.0 NDEA 0.02 0.01 2.0 NPYR 0.2 0.2 2.0 NPIP 0.01 0.01 1.0 NDBA 0.01 0.01 1.0 * See Table 1 for abbreviations. Table 4. Recovery (%) of N-nitrosamines added in different concentrations to a spiked smoked meat sample (extraction by Extrelut-Florisil SPE) Concentration of NA added, NDMA NDEA NPIP NPYR NDBA ppb 0.1 61 65 73 64 60 0.5 59 68 74 66 59 1.0 68 74 78 76 67 Table 5. Mean level ([micro]g/kg) of volatile N-nitrosamines in some Estonian meat products Product No. of NDMA NDEA NPIP NPYR NDBA Total samples Fried meat 3 2.37 0.54 0.60 3.54 0.16 7.21 Ham 21 1.80 0.42 0.53 2.32 0.24 5.31 Smoked chicken 7 1.90 0.33 1.31 8.94 0.19 12.67