422 Acta Chim. Slov. 2005, 52, \72-\7& Scientific Paper MSPD Combined With Fast GC for Ultratrace Analysis of Pesticide Residues in Non-Fatty Food Milena Dömötörová,a Eva Matisová,a* Michal Kirchner,a and Jaap de Zeeuwb "Faculty of Chemical and Food Technology, Slovak University of Technology, Radlinského 9, 812 37 Bratislava, Slovakia, E-mail: eva.matisova@stuba.sk b Varian International B.V., Herculesweg 8, P.O. Box 8033, 4330 EA, Middelburg, The Netherlands Received 30-03-2005 Abstract Matrix solid phase dispersion (MSPD) sample preparation method was combined with fast gas chromatography (GC) to determine pesticide residues of different volatility and polarity at ultratrace concentration level. Apples as representatives of a non-fatty food were chosen as a matrix; they are also a common raw material for baby food production. At fast GC conditions with electron capture detection (ECD) several parameters of MSPD procedure were optimised. Sample is homogenized with sorbent Florisil, pesticides are eluted with the optimised volume of etylacetate. After evaporation of solvent to dryness, reconstitution of the rest to toluene follovvs and the final extract is injected utilising splitless injection. These optimised procedure leads to recoveries > 90% (at concentration level of 60 /xg kg-1) and limits of quantification (LOQs) < 47 /xg kg-1 (except of diazinon) utilizing ECD. Except of dimethoate ali LOQs are lower than related maximum residual limits (MRLs) set for commodity apple. Also the possibilities of mass spectrometric (MS) detection were studied. Even at the half pre-concentration factor LOQs for ali pesticides were lower than 10 /xg kg-1, the MRL for baby food. Acceptable recoveries were obtained even at concentration level of 5 /xg kg-1 (> 79%). Key words: MSPD, fast GC, pesticide residues, apples, matrix-effects Introduction Analysis of pesticide residues has recently become one of the priorities in the food chain control. The usage of chemical pesticides is known to have a significant positive impact on crop yields; however, the residues of pesticides have a negative impact on human health. Therefore, European Commission strictly regulates the level of pesticide residues in various food commodities through the maximum residue limits (MRLs).1 The most suitable approaches in determination of pesticide residues content at ultratrace level (<1 mg kg-1, or <0.0001%, w/w)2 in food material are chromatographic methods connected with various sample preparation methods. Especially fast GC techniques provide faster, cost-effective analytical answer.3'4 Practically, the advantage of increased throughput of fast separation method can be fully taken only in combination with fast and effective sample preparation method, therefore, matrix solid-phase dispersion (MSPD) was chosen as sample preparation technique for the analvtes of apple samples. MSPD offers several advantages over other sample preparation methods including simplicity, relativeh/ low consumption of organic solvents. It performs celi disruption, homogenization, extraction and clean-up in one step, what is related with better precision.5'6 In this work, MSPD sample preparation procedure was connected with fast GC utilizing selective detections: electron capture (EC) and mass spectrometric (MS) detectors. Several parameters of sample preparation method were optimized at on the most common raw plant material for baby food production - apple. The objective of this work is to reach satisfactory limits of quantification (LOQs) and recoveries meeting the requirements of Council Directive 91/414 EEC (particularly the guidance documents establishing MRLs1 for selected fruits and establishing requirements on validation7). Experimental 1. Reagents and materials The pesticides diazinon, terbuthylazine, dimethoate, pyrimethanil, chlorpyrifos-methyl, fenitrothion, chlorpyrifos, cyprodinyl, penconazole, methidathion, kresoxim-methyl, myclobutanil, tebuconazole, phosalone, bitertanol, cypermethrin, Dömötörová et al. Analysis of Pesticide Residues in Non-Fatty Food Acta Chim. Slov. 2005, 52, 422–428 423 etofenprox were of > 95% purity from various sources (Table 1). Bitertanol and cypermethrin consist of isomers. Stock solution of pesticides with concentration of 0.5 jig L-1 was prepared by dissolving 5 mg of each compound in 10 mL of toluene (Suprasolv, Merck, Darmstad, Germany) and was stored at -18 °C. Standards were weighted on Sartorius Analytic MCI balances (Sartorius, Götingen, Germany) with a precision of + 10 jig. The stock solution of pesticides was diluted with acetone (Suprasolv, Merck, Darmstad, Germany) to get appropriate pesticide standard solutions for the preparation of spiked samples and matrix-matched standards. Ethyl acetate and dichloromethane were of gas chromatography grade (Suprasolv, Merck, Darmstad, Germany). Magnesium sulfate (anhydrous powder) was from Lachema (Neratovice, Czech Republic). The sorbent used was Florisil (60-100 mesh) from Rotichrom, Roth, Karlsruhe, Germany. Apples were mixed with blender Braun MX 2050 (Kronberg, Germany). 2. Sample preparation The sample used for this study was homogenously mixed chemically untreated apples (with peel) which were stored at -18 °C in a refrigerator. The whole optimised process of sample preparation is presented in Figure 1. Solutions of matrix-matched standards were prepared by adding the appropriate volume of the pesticide stock solution to an extract of a blank apple sample prepared by the MSPD method. 3. Instrumentation GC-ECD GC measurements were performed on a HP 6890 gas chromatograph (Hewlett Packard, Avondale, PA, USA) equipped with an electron capture detector (ECD) operated at 320 °C. The rate of data acquisition was set to 50 Hz. The splitless injector with 2 mm i.d. liner (Agilent Technologies, USA) operated at 250 °C (split vent 100 mL min1, 1 min) was used for sample introduction. Injections were carried out by an autosampler using a 10 /jlL syringe (Hamilton, Reno, Nevada, USA). Hewlett Packard GC Chemstation software was used to control analysis, collect and process the data. For the experiments, a 25 m x 0.15 mm i.d. CP Sil 13 CB fused silica capillary column with a 0.4 /im 14% phenyl, 86% dimethylpolysiloxane stationary phase (Varian, Middelburg, The Netherlands) was used and it was coupled with a 1 m x 0.32 mm i.d. non-polar deactivated fused silica pre-column (Supelco, Bellefonte, PH, USA). For a pre-column - analytical column connection a glass press-fit connector (0.32/0.20, Agilent Technologies, USA) was utilized and sealed with a polyimide resin according to the manufacturers instructions (Supelco, Bellefonte, USA). Measurements were performed under temperature-programmed conditions (initial temperature 100 °C, initial time 1 min, heating rate 65 °C min1, and final temperature 290 °C (4 min)). Hydrogen (purity >99.99%, Linde, Technoplyn, Bratislava, Slovak Republic) was used as a carrier gas. Carrier gas flow programming was used: 2.3 mL min1 (5.5 min), 2 mL min2, 3.4 mL min1; an electronic pressure control was employed. Table 1. List of pesticides, chemical classes, sources of pesticide standards and monitored ions utilizing GC-MS in SIM mode. Pesticide* Chemical class Source of pesticides Monitored ions in SIM mode Target ion Diazinon Organophosphoras Argovita 276, 304 a Terbuthylazine Triazine Ciba-Geigy, Basel, Switzerland 214, 229 a Dimethoate Organophosphoras Cheminova Agro, Denmark 87,125 a Pyrimethanil Anilinopyrimidine Schering, Germany 198, 199 a Chlorpyrifos-methyl Organophosphoras Dr. Ehrenstorfer, Germany 286, 288b Fenitrothion Organophosphoras Sumimoto Chemical Co., Japan 260, 277b Chlorpyrifos Organophosphoras Dow Chemical Company, USA 286, 314b Cyprodinil Anilinopyrimidine Ciba-Geigy, Basel, Switzerland 224, 225c Penconazole Triazole Ciba-Geigy, Basel, Switzerland 248, 250c Methidathion Organophosphoras Ciba-Geigy, Basel, Switzerland 145, 302c Kresoxim-methyl Oximinoacetate BASF, Germany 131, 132 d Myclobutanil Triazole Dow Agro Science, USA 179, 245 d Tebuconazole Triazole Argovita 250, 252e Phosalone Organophosphoras Dr. Ehrenstorfer, Germany 182, 367f Bitertanol Triazole Bayer, Germany 168, 170g Cypermethrin Pyrethroid Argovita 163, 181g Etofenprox Non-ester pyrethroid Mitsui Toatsu Chemicals, Japan 163, 376 g * pesticides are arranged according to retention times; a, b, c, d, e, f, g denote ions monitored in one SIM group; Target ion - used for quantification. Dömötörová et al. Analysis of Pesticide Residues in Non-Fatty Food 424 Acta Chim. Slov. 2005, 52, 422–428 iniection of 2 fiL 1. Blending reconstitution with E l.SmL toluene > , => 3. Evaporation on rotary vacuum evaporator final extract eluat glass wool 4. Evaporation under N2 to final volume 2. Elution Figure 1. Scheme of the whole apple sample MSPD preparation process. GC-MS GC-MS measurements were performed on an Agilent 6890N GC coupled to 5973 MSD (Agilent Technologies, Avondale, PA, USA) equipped with a programmed temperature vaporizer (PTV) and autoinjector Agilent 7683. MS with electron impact ionization (EI) mode (70 eV) was operated in SIM mode; for each pesticide two specific ions were selected and sorted into groups (maximal number of ions in one group was 8); the used dwell time was 10 ms. PTV was operated in cold splitless mode. An injection volume was 2 /aL. Helium with purity 5.0 (Linde Technoplyn, Bratislava, Slovak Republic) was used as a carrier gas. Narrow bore chromatographic columns CP-Sil 8 CB (Varian, Middelburg, The Netherlands) with 5% diphenvl 95% dimethylsiloxane stationary phase 15 m x 0.15 mm I.D. x 0.15 /im was utilised. It was connected to a non-polar deactivated pre-column. Constant pressure mode 363.5 kPa was used until the elution of the last analvte (ethofenprox, 7.90 min), additional pressure ramp (1000 kPa min-1, 685 kPa) was used to speed-up elution of higher boiling matrix co-extractives. Chromatographic separation was performed under a temperature program, 130 °C (1.13 min), 27.25 °C min-1, 290 °C (8 min). PTV conditions: temperature program, 150 °C, 400 °C min-1, 300 °C (2 min), 400 °C/min, 350 (5 min); split vent open time 1.13 min. Results and discussion Pesticides belonging to the different chemical classes to include different chemical properties and polarities were selected (Table 1). For the separation of pesticides, capillary GC columns coated with non-polar to middle polar stationary phases have been utilized. A capillary with the internal diameter of 0.15 mm was chosen instead of 0.1 mm for the reason of the increased column capacity and/or carrier gas flow through the column. This column dimension can be used in the majority of GC instruments and offers more flexibility with respect to flow, loadability and practical operation. Conditions of fast GC-ECD were optimised with the mixture of selected pesticides and n-alkanes C10-C28 in hexane at the concentration level of 10 mg L * of each compound.8 The influence of the injection volume, type of liner, injection technique, retention gap length, type of solvent, oven temperature and column length on peak areas, peak shape and peak broadening of analvtes was investigated. At these conditions aH peaks were separated with the resolution > 1.4. In the èase of fast GC-MS analysis, the chromatographic method was developed with respect to the separation/speed trade-off based on key principles proposed by Klee at al.:9 speed-optimised flow (SOF), and optimal temperature ramp rate 10 °C/tM, tM denotes void time. Development of sample preparation method A critical aspect of pesticide residue analysis is the sample extraction and purification, which are required to isolate the residues from the matrix components. The MSPD technique seems to be superior to other techniques used for this purpose.10 For purposes of development of MSPD sample preparation method in apples as non-fatty food representatives, fast GC-ECD configuration was utilised. Fruit sample free of pesticides were used for the preparation of a matrix matched standard prepared in blank extracts. The ECD shows that blanks were free of the selected pesticides (Figure 2A). First two solvents of different polarity - ethyl acetate and dichloromethane Dömötörová et al. Analysis of Pesticide Residues in Non-Fatty Food Acta Chim. Slov. 2005, 52, 422–428 425 were tested as eluents (80 mL). After eluent evaporation the final volume used for the residues reconstitution was 1 mL of toluene. Lhe recoveries of pesticides using fortified apples at the level of 120 jig kg-1 ranged between 97-113% for ethyl acetate and 83-97% for dichloromethane. Lhe difference in background of chromatograms for the two solvents was not significant; therefore, ethyl acetate was used as eluting solvent in further experiments. A min B 10 5 Hz 1400 1200 1000 800 600 400 200 0 5 Hz 1400 1200 1000 800 600 400 200 0 5 Hz 1400 1200 1000 800 600 400 200 0 3 4 5 6 min Figure 2. Chromatograms of apple sample extracts prepared by MSPD (A) blank apple sample; spiked apple samples: (B) 120 fig kg_1 with final volume of solvent 1 mL used for reconstitution of residues and (C) 60 fig kgJ with final volume 0.5 mL, 1: diazinon, 2: dimethoate, 3: chlorpyrifos-methyl, 4: fenitrothion, 5: chlorpvrifos, 6: penconazole, 7: methidathion, 8: kresoxim-methvl, 9: mvclobutanil, 10: phosalone. min C 4 6 8 7 10 to reach peaks of the same height, the background increased moderately related to the peak heights. Lhe recoveries were found over 90% (except diazinon) (Lable 2). Further reduction of the final volume leads to significantly lower precision and higher background leading to problems with matrix induced enhancement effect. Lherefore, the final volume adjusted to 0.5 mL is the best choice. Lhe choice of injection volume of samples with difficult matrices, such as plant extracts must be a compromise betvveen: • Required LOQs • Degree of chromatographic system contamination. Lhe injected matrix significantly affects chromatographic system and subsequently analytical results (matrix effects).11 1400 1200 1000 800 600 400 200 0 1p.L ¦ 1 ¦ 5 D 20 ¦ 55 0 90 10500 9000 7500 6000 4500 3000 1500 0 ¦ 1 ¦ 5 D 20 ¦ 55 0 90 Figure 3. Graph of the dependence of peak areas of pesticides injected by autosampler on the number of performed injections studied/retention gap at the concentration of matrix matched standard 0.625 mg L–1 (corresponds to concentration 125 mg kg–1). Lhen the eluting volume was tested and the volume of 60 mL is considered as sufficient to reach satisfactory recoveries (95%-105%). Finally the final volume of toluene (Figure 2) was adjusted. In Figure 2 there is the comparison of two chromatograms with the final volume of 1 and 0.5 mL. Although the spiking concentration level for the final volume of 0.5 mL was two times lower then for lmL 90 injections of 1 and 5 /jlL of matrix matched standards (1 mg L-1) were performed to study the changes in the peak areas and widths with the number of injections. Lhe experiment started utilizing a new pre-column, inlet liner and septum. In general, peak areas increased with the number of injections in both cases (Figure 3). For the injection volume of 5 /jlL, the increase was more noticeable. Lhe change in the Dömötörová et al. Analysis of Pesticide Residues in Non-Fatty Food 426 Acta Chim. Slov. 2005, 52, \72-\7& first five injections presents up to 12% increase of peak areas for the injection of 1 /jlL and mostly over 120% for 5 /jlL injection volume. The changes in peak areas in the whole range of 90 injections ranged from 6.9% for phosalone to 344% for kresoxim-methyl for injected volume of 1 /aL and from -2.3% for phosalone to 1196% for penconazole for injected volume of 5 /jlL. The dependence of peak widths on the number of performed injections is presented in Figure 4. Evident increasing trend for aH compounds except of diazinon was observed for the higher injection volume, for 1 /jlL injected the changes in the peak widths represent up to 20%. Therefore, 2 /aL was a good choice as a compromise as the injection volume for the combination of fast GC and MSPD. To minimize the impact of the matrix effect five injections of a blank apple sample were injected before alternation of injections of spiked samples and matrix matched standards. In this study, several other ways to reduce matrix effects were practised: addition of the final isothermal part of the oven temperature program to remove the high boiling components from a column and the utilization of a pre-column to protect the analytical column from an excessive contamination. 2.5 2 1.5 0.5 0 1HL ¦ 1 ¦ 5 D 20 ¦ 55 0 90 Method validation GC-ECD For the recovery calculation responses of pesticide residues of matrix matched standards and spiked samples were used. The recovery data and LOQs under optimal conditions are shown in Table 2. Except of dimethoate ali LOQs of the studied pesticides were lower than the required MRL established by European Commission for commodity apple.1 Good recoveries ( > 90%) were obtained, what is in an agreement with directives laid down by the European Commission (pesticides recoveries should be in 70-110% range with relative standard deviations (RSD) < 20%).7 Repeatability of peak areas measurement by fast GC-ECD for ali pesticides expressed as RSD (n=5) was in the range of 5.8-16.1% at the concentration level of 60 l±g kg-1. The obtained MRLs are, however, not sufficient for baby food control. Therefore, further research was performed. Table 2. The GC-ECD results of the recovery and limits of quantification (LOQs) of selected pesticide residues from apples at the concentration level of 60 fig kgJ. Pesticide residue Recoverv % LOQ a MRL b Hg kg-1 ng kg-1 Diazinon 90 100.6 300 Dimethoate 98 37.1 20 Chlorpyrifos-methyl 92 13.8 500 Fenitrothion 96 46.2 500 Chlorpyrifos 91 8.8 500 Penconazole 92 2.0 200 Methidation 93 29.3 300 Kresoxim-methyl 90 12.8 200 Myclobutanil 93 31.2 500 Phosalone 92 30.8 2000 "(LOQ) = 0-10 , s 0 - standard deviation of noise (peak height), š - detector response (height), h MRL maximum residual limit for commoditv apple.1 2.5 t 2 1.5 1 0.5 0 D 1 D 5 D 20 ¦ 55 0 90 Figure 4. Graph of the dependence of peak widths at half height (wi/2) of pesticides injected by autosampler on the number of performed injections studied/retention gap at the concentration of matrix matched standard 0.625 mg L-1 (corresponds to concentration 125 fig kgJ). GC-MS LOQs may be further reduced by utilising MS detection technique in SIM mode by measuring specific ions what significantly improves selectivity of detection. Extracted chromatograms of target ions of pesticides at concentration level of 10 jig kg-1 are shown in Figure 5. Table 3 presents LOQs obtained by GC-MS analysis in SIM mode as well as the results from the fortification experiment of apples. Recoveries at three spiking levels of 5, 10 and 100 jig kg-1 are presented. As the table shows, LOQs obtained with MS detection are 3-290 times lower compared to ECD (Table 3) and are lower than the required MRL for baby food 10 jig kg-1. The recovery results fell within the commonh/ accepted range 70-110% recovery and < 20% RSD1 (except of tebuconazole at 5 jig kg-1 level with RSD 54% Dömötörová et al. Analysis of Pesticide Residues in Non-Fatty Food Acta Chim. Slov. 2005, 52, 422–428 427 Table 3. Limits of quantification (LOQs), average recoveries (K) and relative standard deviations (RSDs) of fortified pesticides in apples at different concentrations 5, 10 and 100 fig kgJ from the MSPD method with GC-MS analyses Spiking level (ug kg"1) LOQ ug kg"1 5 10 100 Pesticide R% RSD " % R% RSD"% R% RSD a % Dimethoate 2.67 84 4.2 107 14 99 2.2 Terbuthylazine 0.25 101 7.7 99 1.7 96 0.8 Diazinon 0.36 93 5.4 91 6.6 93 8.0 Pyrimethanil 0.07 93 0.4 94 3.1 95 0.9 Chlorpyrifos-methyl 0.06 94 0.5 95 3.8 93 3.8 Fenitrothion 0.16 98 4.8 95 3.9 95 1.6 Chlorpyrifos 0.30 90 1.6 93 0.7 93 2.6 Cyprodinyl 0.65 92 2.3 90 1.0 94 0.7 Penconazole 0.60 86 4.5 99 0.9 96 0.2 Methidathion 2.76 99 0.9 106 1.7 96 0.5 Myclobutanil 0.68 96 9.3 99 0.9 96 0.2 Kresoxim-methyl 2.00 99 8.9 71 18 93 1.9 Tebuconazole 0.59 79 54 92 0.6 97 2.3 Phosalone 0.57 88 18 105 16 95 0.5 Bitertanoll 0.24 95 1.6 94 3.2 95 0.9 BitertanoU 2.25 92 12 96 3.3 91 0.5 Cypermethrinl 1.42 108 2.2 92 13 92 5.9 Cypermethrin2 2.64 79 16 91 20 90 0.5 Cypermethrin3 5.03 79 17 96 1.0 95 7.2 Etofenprox 1.17 85 4.5 91 6.3 93 2.7 ' RSD (n=2) of recovery experiments were calculated according to Eckschlager et al.12 due to interference). These satisfactory results were obtained even without final evaporation step (Figure 1); the evaporation on rotarv vacuum evaporator was followed by reconstitution in 1 mL of toluene, what saves tirne and simplifies the overall sample preparation procedure. Besides that the burden on chromatographic svstem decreases. Abundance 3000 2000 1000 -0 Abundance 4000 3000 2000 1000 - 0- 13 14 17 (I / 15 16 // Time--> 3.80 4.00 4.20 4.40 4.60 4.80 Time--> 6.40 6.60 6.80 7.00 7.20 7.40 7.60 7.80 Figure 5. Extracted chromatograms of target ions of pesticides in matrix matched standard solution from fast GC-MS measurements; concentration of pesticides 10 fig kg-1. 1: dimethoate, 2: terbuthylazine, 3: diazinon, 4: pyrimethanil, 5: chlorpyrifos-methyl, 6: fenitrothion, 7: chlorpyrifos, 8: cyprodinyl, 9: penconazole, 10: methidathion, 11: kresoxim-methyl, 12: myclobutanil, 13: tebuconazole, 14: phosalone, 15: bitertanol, 16: cypermethrin, and 17: etofenprox. Abundance 2000 - 1500 - 1000 - 500 - 0-Time--> 10 11 12 4.80 Lil y/Vw—-—~f^c^\^__J——J ^—~—- -7—^—i-------¦------7~"~r~~l'i-------¦-------T 5.00 5.20 5.40 5.60 5.80 The linearitv of response of GC-MS in SIM mode was checked with calibration matrix matched standards in black extracts. Duplicate determinations at three concentration levels (0.025-0.5 mg L-1, what corresponds to 5-100 jig kg-1 in apple sample) were carried out according to European Commission directives.13 The obtained coefficients of determination R2 were in the range of 0.9995-1, except of kresoxim-methyl (0.9985), tebuconazole (0.9911), and cvpermethrinl (0.9594). 5 6 8 9 7 Dömötörová et al. Analysis of Pesticide Residues in Non-Fatty Food 428 Acta Chim. Slov. 2005, 52, 422–428 Conclusions References Application of fast GC in combination with MSPD sample preparation to ultratrace analysis of pesticides in apple samples (non-fatty food) was performed. Several parameters of MSPD procedure were optimized utilising ECD detection. The elution solvent ethyl acetate with elution volume of 60 mL was selected for extraction purposes of pesticides from the homogenized sample with sorbent Florisil. The final volume of 0.5 mL was adjusted by evaporation of reconstituted solution under N2. At these conditions recoveries in the range of 90-98% were obtained at the concentration level of 60 jig kg-1. LOQs reached were betvveen 2-100.6 jig kg1. Except of dimethoate LOQs of aH pesticides did not exceed MRLs set by European Commission for commodity apple. In the èase of GC-MS measurements in SIM mode, the rest after the evaporation to dryness by rotary vacuum evaporator was reconstituted in 1 mL toluene, so the step of evaporation under N2 was omitted. Good recoveries of pesticides at the concentration levels of 5, 10 and 100 jig kg-1 were achieved at aH concentration levels. LOQs fulfil even the requirement on the analysis of baby food; the values are lower than 5 jig kg-1 (MRL set for baby food is 10 jig kg1). Good linearity (R2 between 0.9995-1) is reached in the range of 0.025-0.5 mg L-1, what corresponds to 5-100 jig kg-1 in apple sample. Acknowledgements This work was supported by the NATO project No. SfP 977 983 and the Slovak Grant Agency (Vega, project No. 1/2463/05). 1. Cuncil Directives 76/895/EEC, http://europa.eu.int/ comm/food/plant/protection/resources/mrl_pesticide.xls, (accessed: 2005-06-29). 2. F. A. Settle, in: F. A. Settle (Ed.): Handbook of Instrumental Fechniques for Analvtical Chemistrv; Prentice Hali PFR, New Jersev, 1997, pp. 10. 3. E. Matisová, M. Dömötörová, /. Chromatogr. A 2003, 1000, 199-221. 4. P. Korvtár, H. G. Janssen, E. Matisová, U. A. Fh. Brinkman, Trends Anal. Chem. 2002, 21, 558-572. 5. S. A. Barker,/. Chromatogr. A 2004, 1055, 159-168. 6. B. Morzvcka, Chem. Anal. (Warsaw) 2002, 47, 571-583. 7. Council Directives 94/43/EC, Off. J. Eur. Com. 1994, L227, 31. 8. M. Kirchner, E. Matisová, M. Dömötörová, J. de Zeeuw, /. Chromatogr. A 2004, 1055, 159-168. 9. M. S. Klee, F. M. Blumberg,/. Chromatogr. Sci. 2002, 40, 234-247. 10. M. Kirchner, E. Matisová, Chem. Listy 2004, 98, 396^-05. 11. J. Hajšlová, J. Zrostlfková,/. Chromatogr. A 2003, 1000, 181-197. 12. K. Eckschlager, I. Horsák, Z. Kodejš, in: E. Hugová (Ed): Evaluation of Analvtical Results and Methods, SNFF, Praha, 1980, pp. 25. 13. SANCO/3029/99 REV.4, Working document, http:// europa.eu.int/comm/food/plant/protection/evaluation/ guidanceAvrkdocl2_en.pdf, (accessed: 2005-06-29). Povzetek Za kromatografsko analizo ostankov pesticidov razliènih polarnosti in hlapnosti smo uporabili predpripravo vzorca s pripravo mešanice matrice s sorbentom v trdni fazi. Kot primer nemastnega prehrambenega pridelka smo kot vzorec izbrali jabolka. Za plinsko kromatografsko detekcijo smo uporabili detektor na zajetje elektronov. Optimizirali smo veè parametrov procesa predpriprave vzorca, kot je kolièina etilacetata, ki ga uporabimo za spiranje sorbenta Florisila, ki ga pripravimo v mešanici z vzorcem. Po odparevanju topila do suhega smo preostanek raztopili v toluenu in injecirali brez deljenja vzorca. Izkoristki ekstrakcijskega postopka so nad 90% (pri vsebnosti ~60 µg kg–1), meje doloèitve pa pod 47 µg kg–1 (razen za diazinon). Razen za dimetoat so vse meje doloèitve manjše kot so odgovarjajoèe najvišje meje za preostanke pesticidov doloèene za jabolka široke potrošnje. Preuèili smo tudi možnosti masno-spektrometriène doloèitve. Tudi pri poloviènem predkoncentracijskem kvocientu so bile meje doloèitve manjše kot 10 µg kg–1, kar je meja za ostanke pesticidov v otroški hrani. Sprejemljive izkoristke smo dobili tudi pri vsebnostih ~5 µg kg–1 (?79%). Dömötörová et al. Analysis of Pesticide Residues in Non-Fatty Food