Multi-residue Method for Pesticides Residue Analysis in Grapes by GCxGC-TOF-MS

Transcription

1 Multi-residue Method for Pesticides Residue Analysis in Grapes by GCxGC-TOF-MS Published in :J.Chromatography A,119 (28),35-357

2 Chapter -3 Multiresidue Analysis of Pesticides Residue in Grapes by GC x GC-TOF-MS Published in :Journal of Chromatography A,119 (28),35-357

3 ABSTRACT A comprehensive GC GC TOFMS method was optimized for multiresidue analysis of pesticides using a combination of a non-polar (RTX-5MS, 1.m.18mm.2µm) and a polar capillary column (TR-5MS, 1.m.1mm.1µm), connected in series through a dual stage thermal modulator. The method resolved the co-elution problems as observed in full scan one-dimensional GC MS analysis and allowed chromatographic separation of 51 pesticides within 24 min run time with library-searchable mass spectrometric confirmation. Four pesticides, viz. chlorpyrifos-methyl, vinclozoline, parathion-methyl and heptachlor could be baseline separated on GC GC, whichwere otherwise closely eluting and interfering each other s detection in 1D GC MS run. Similarly, it could be possible to separate myclobutanil, buprofezin, flusilazole and oxyfluorfen on GC GC. Although in 1D GC MS, these closely eluting compounds could be identified through deconvolution algorithm and peak-find option of the Chromatof software but the spectral purity significantly improved on GC GC analysis. Thorough optimization was accomplished for the oven temperature programming, ion source temperature and GC GC parameters like modulation period, duration of hot pulses, modulation-offset temperature, acquisition rate, etc. to achieve best possible separation of the test compounds. The limit of detection significantly improved by 2 12 times on GC GC TOFMS against GC TOFMS because of sharper and narrower peak shapes. The method was tested for grape matrix after preparing the samples using previously described method and recoveries of the entire test pesticides were within 7 11% at 1 ng/g level of fortification. GC GC TOFMS was found to be an excellent technique for library-based screening of pesticides with high accuracy and sensitivity. 3.1 INTRODUCTION Comprehensive two-dimensional gas chromatography coupled with time-offlight mass spectrometry (GC GC TOFMS) offers unprecedented separation power in multiresidue analysis. Combination of a long non-polar with a short and polar capillary column connected in series through a thermal modulator provides enormous peak capacity, which is utilized in separating mixture of large number of compounds in single chromatographic run. The TOF mass analyzer further enhances the 78

4 separation process on the basis of relative flight times of ions as decided by their mass/charge (m/z) ratio. Although quite a number of papers described the separation efficiency of this technique [1 4], the literature evaluating the quantitative performance of this technique in pesticide residue analysis in agricultural products is rather limited [5]. The commercial cultivation of grapes in India receives frequent application of a large number of pesticides throughout cropping season to control a variety of pests and diseases. Pesticide residue is a major concern for the grape industry, since a significant fraction of the total production is consumed as fresh fruit without any possibility of peeling or decontamination treatments. Since pesticide residues may also appear from indirect sources like soil, contaminated agro-inputs (e.g. manures, fertilizers, growth regulators, irrigation water, etc.), drift from adjoining fields of other agricultural crops, etc. it makes residue monitoring a challenging task and establishes the need for monitoring both the recommended as well as non-recommended chemicals [6,7] to ensure consumer safety. The quality regulations and food safety standards are becoming more stringent in most countries. Thus, target-oriented residue analysis using quadrupole or ion trap GC MS with selected ion monitoring (SIM) [8 1] or tandem mass spectrometry (MS/MS) [11,12] is not always sufficient to provide complete information about the contamination status of any food sample as these techniques involve targeted acquisition of only selected compounds included in the screening program and at non-target full scan mode, it may not be possible to achieve the desired level of sensitivity for all analytes. As in SIM or MS/MS modes, we lose valuable mass spectral information, the uncertainty level of analysis increases and library-based screening is not possible. On the other hand, in full scan mode, the closely eluting compounds may affect the quality of each other s mass spectra and this in turn may result in false positives/negatives. Co-elution may also lead to over-estimation or under-estimation of residues. Sometimes, use of long column with slow multi-step temperature programming is useful in resolving co-elution problems but the industry and the regulatory bodies expect a rapid turn-around time, and thus, we cannot afford to have a long GC run to separate multiple compounds. Unambiguous identifications of residues are thus challenging with 1D GC MS especially when sample history or contamination sources are unknown. Under such situation, GC GC TOF-MS provide novel solution in providing high peak capacity, adequate sensitivity in full 79

5 scan (mass range: 5 1 amu) due to high mass analyzer efficiency and acquisition rate as high as complete 5 spectra/s, generating large number of data points across a narrow peak. Objective In this work, we report the strategy by which we developed and optimized a GC GC technique for separation of 51 pesticides in grape matrix followed by TOF- MS analysis. This powerful technique facilitated automated library-based screening of the residues at 1 ng/g level. We thought to describe our endeavour in detail so that it will be useful for fellow researchers while planning to use this relatively new and novel technique in multiresidue analysis of pesticides. While optimizing the method, we followed the strategy described by Hoh et al. for polychlorinated dibenzo-p-dioxins and dibenzofurans [13]. 3.2 MATERIALS AND METHODS Chemicals and Apparatus Certified reference standards of all the test pesticides were of >98% purity and purchased from the Ehrenstorfer GmbH, Augsburg, Germany. The extraction solvent ethyl acetate was of specially dried residue analysis grade from Thomas Baker, India. Dispersive solid phase extraction (DSPE) material, viz. primary secondary amine (PSA, 4µm, Bondesil) was procured from the United Chemical Technology (UCT), PA, USA. Anhydrous sodium sulphate (Analytical Reagent grade) was purchased from Merck India Limited and was further purified by heating at 65 C for 4 h before use. An Agilent 689N GC system (Agilent Technologies, USA) hyphenated to Pegasus IV time-of-flight mass spectrometer (Leco, St. Joseph, MI, USA) was used for analysis. The GC system was equipped with a secondary column oven and nonmoving quad jet dual stage thermal modulator. During modulation, cold pulses were generated using dry nitrogen gas cooled by liquid nitrogen (INOX Air Product Limited, Mumbai, India), whereas dry air heated by a heating block was used for hot pulses. Ultra-pure grade helium (BOC India Limited) was used as the carrier gas. The other equipments used in analyses included a high speed homogenizer (DIAX- 8

6 9, Heidolph, Germany), low-volume concentrator (Turbo Vap LV; Caliper Life Sciences, Russelsheim, Germany), non-refrigerated centrifuge (Eltek, Mumbai, India) and a microcentrifuge (Microfuge Pico, Kendro D-3752, Osterode, Germany) Selection of Pesticides A total of 51 pesticides were considered for this study, which belonged to the list of chemicals to be monitored in Indian table grapes for export [6] and were amenable for GC analysis. The names of these pesticides along with their chemical classes are presented in Table 3.1 in alphabetical order Preparation of Standard Solutions The stock solutions of the individual pesticide standards were prepared by accurately weighing 1 mg (±.1 mg) of each analyte in volumetric flasks (certified A class) and dissolving the same in 1 ml ethyl acetate. These were stored in dark vials in a refrigerator at 2 o C. An intermediate stock standard mixture of 1 mg/l was prepared by mixing the appropriate quantities of the individual stock solutions followed by requisite volume makeup. A working standard mixture of 1 mg/l was prepared by diluting the intermediate stock standard solution, from which the calibration standards were prepared by serial dilution with ethyl acetate Calibration The calibration graphs (seven levels) for all the compounds were obtained by plotting the individual peak areas against the concentration of the corresponding calibration standards, viz. 1, 25, 5, 1, 25, 5 and 1 ng/ml in ethyl acetate. Matrix-matched standards were prepared using the ethyl acetate extract of fresh untreated grapes. The matrix extracts were at first analyzed by GC MS and LC MS/MS to confirm the absence of the test pesticides in them before spiking. For each pesticide, the mass fragment (m/z) with maximum S/N was selected as the quantification ion (Table 3.1) and its peak area for different concentrations were used to construct the calibration graph. The limit of detection (LOD) of the test compounds was determined by considering a signal to noise ratio (S/N) of 3 with reference to the background noise obtained from blank sample. 81

7 3.2.5 Sample Preparation for Multiresidue Analysis in Grapes Grape samples (2 kg berry only) treated with pesticide mixture were blended thoroughly under ambient condition. Blended sample (2 g) was further homogenized at high speed and from it 1 g sample was drawn for final analysis. The samples were extracted with 1 ml ethyl acetate (+1 g anhydrous sodium sulphate) and processed as per the method described earlier [14]. The final 1mL extract was cleaned by DSPE with PSA (25 mg), centrifuged at 1, rpm for 2 min and the supernatant was filtered through.2 µm polytetrafluoroethylene membrane filter and directly injected to the GC MS system Recovery Experiments The recovery experiments were carried out on fresh untreated grapes by fortifying the samples (1 g) in eight replicates with pesticide mixture at 1 ng/g level. All the treated samples were analyzed immediately GC GC TOF-MS Analysis The GC separation was performed by injecting 1 µl (splitless) through autosampler (CTC Combipal) on an RTX-5MS [Cross bond 5% diphenyl 95% dimethylpolysiloxane] capillary column (1 m.18 mm.2 µm, Restek, Germany) connected in series to a TR-5MS (5% Phenyl polysilphenylene-siloxane; 1 m.1 mm.1 µm, Thermo, USA) secondary column at the constant helium gas flow rate of 1mL/min. At constant flow mode, the flow rate of carrier gas remains constant throughout the chromatographic run and if the column resistance changes due to temperature programming, the column head pressure is adjusted to keep the flow rate constant. The injector port temperature was 25 o C with transfer line temperature 28 o C. The oven temperature was programmed from 7 o C (1 min hold) to 27 o C (3 min hold) with ramping at 1 o C /min. The secondary oven was kept at 1 o C higher temperature throughout the chromatographic run. The acquisition delay was 32 s. The modulation period was set at 4 s, with hot pulse.8 s and corresponding cold pulse time of 1.2 s with modulation-offset temperature 25 o C. The MS parameters included electron ionization at 7 ev with ion source temperature 24 C. The mass 82

8 spectrometer was tuned for the mass spectrum of PFTBA (perfluorotributylamine). The detector voltage was set at 175V and the data acquisition was carried out within the mass range of 5 6 m/z at acquisition rate of 1 spectra/s Data Processing and Quantification Chromatof 3.22 software was used for data processing. The chromatograms were processed with baseline offset.5 (computation through the middle of noise), peak find with S/N of the quantifier ion at least 3 and peak width.1 s. Minimum similarity match with regards to NIST library spectra was kept at 6 (reversed fit), whereas, the software combined the second dimension peaks of the same compounds from different modulation cycles on the basis of similarity set at 7. Quantification was done on the basis of single diagnostic ion (Table 3.1) and the peak assignments, integration and summation of modulations were automatically done through software. Quantifications on the basis of seven-point calibrations were also manually checked and corrected for any errors. On an average, the software took 2.2 min times for processing the data for single GC GC TOFMS file. 3.3 RESULTS AND DISCUSSION Optimization of Ion Source Temperature In EI-MS, the ion source temperature is an important parameter that influences the extent of analyte ionization and fragmentation. Different ion source temperatures, viz. 16, 18, 2, 21, 22, 23,24 and 25 o C were evaluated for their effect on the response of various pesticides. In general, the S/N of most of the pesticides increased by 2 3% with the increase in ion source temperature from 16 to 24 o C because of sharper peaks and less tailing [13]. For most of the organophosphorus pesticides, the S/N increased upto 23 o C and remained similar over C with the exception of ethion and fenchlorphos, where response increased by 13% when the temperature increased from 24 to 25 C. For methamidophos and buprofezin, the S/N were around 1 14% higher at 16 C in comparison to 24 C, which might occur due to their better thermal stability at lower temperature. Since 25 C is the maximum ion source 83

9 temperature recommended by the instrument manufacturer, we did not test the impact of any higher temperatures on S/N and decided to keep it at 24 C GC Oven Temperature Program Optimization Similar to the approach of Hoh et al. [13], at first, we attempted to optimize the temperature program for 1D GC separation before attempting GC GC. With linear temperature programming between 7 and 27 o C, we noted close elution of chlorpyrifos methyl, heptachlor, parathion-methyl and vinclozoline (Mix I) and also among myclobutanil, buprofezin, flusilazole and oxyfluorfen (Mix II).We compared a number of oven temperature programs for optimum separation of 51 analytes in terms of resolution (Rs) calculated as Rs = 2 t R / (W 1 + W 2 ), where tr represents difference in tr of two adjacent peaks and W 1 and W 2 are their width at base. The deconvolution algorithm and peak find options of the Chromatof software was utilized for processing of data for library searchable identification of the compounds [13, 15]. A temperature programming from 7 to 27 o C at the rate of 1 o C/min resulted in co-elution of dichlorvos with 4-bromo-2- chlorophenol (a metabolite of profenophos). A 1-min hold at 7 o C could separate them but the Rs between atrazine and -HCH was <.2 and in case of Mix I, parathion-methyl remained unidentified (Figure 3.1A). Similarly, components in Mix II co-eluted (Figure 3.2A). The best 1D separation could be achieved with an initial temperature of 7 o C (hold for 1 min); increased to 13 o C at 1 o C/min (hold for 2 min); to 16 o C at 1 o C/min (hold for 2 min); to 19 o C at 4 o C/min and finally increased to 27 o C at 1 o C/min with hold for 3 min (total run time = 32.5 min). The temperature of the secondary column was maintained at 1 o C above the primary column. In this program, the Rs between buprofezin and flusilazole was.62, whereas, for fluisilazole and oxyfluorfen, the Rs significantly improved to But for myclobutanil and buprofezin, and for chlorpyrifos-methyl and parathion-methyl, still the Rs were low (.24 and.7), indicating the limitations of 1D GC program to separate strongly co-eluting pesticides. We also tried different flow programs but it could not yield any significant improvement in Rs of above co-eluting compounds. 84

10 3.3.3 Optimization of GC GC Parameters The optimization of GC GC separation focused on further separation on the basis of the oven temperature program standardized as above. At first, we directly applied the best temperature program as decided for 1D separation and found that on second dimension, all the test compounds in mixture got completely separated at 4 s modulation period having hot pulse time.8 s and modulation offset temperature 25 o C. Since GC GC has significantly higher peak capacity (ideally the product of the primary and the secondary column), we decided to reduce the ramping steps and establish a relatively simpler program with shorter run time to optimize the distribution of peaks over the entire two-dimensional space. After several efforts, finally, we optimized a relatively simple temperature program with initial temperature of 7 o C (hold for 1min); increased to 27 o C at 1 o C/min (hold for 3 min) with the total run time of 24 min. Thus, in comprehensive two-dimensional method, we could save at least 8.5 min. time per run, which is quite significant considering the output of a laboratory in delivering time-bound test results. A further effort to reduce the run time through increase in ramping rate disturbed the separation of certain compounds, hence could not be adopted. The temperature difference between the primary and the secondary column was optimized by using offset temperatures of 1, 2 and 3 o C with regards to the primary oven temperature. The peaks for buprofezin and flusilazole were resolved when the secondary oven temperature was 1 o C higher than the primary column. But, when the secondary column temperature was kept at 2 and 3 o C higher levels, these peaks were co-eluting. Similar effect was observed for heptachlor and vinclozoline also. Thus, we decided to keep the secondary column at 1 o C higher level to optimize the best separation. In GC GC setup, the modulator plays a vital role in improving peak shape, sensitivity and separation through cryofocusing. We tried modulation periods of 2, 3, 3.5, 4, 4.5, 5, 5.5, 6 and 8 s with hot pulse duration remaining at 1/5 th of the modulation period. In general, the peak height increased when the modulation period increased. At short modulation period of 2 s, the peaks of many compounds got distributed into multiple modulation cycles and thus, appeared in multiple slices in the chromatogram. 85

11 Thus, such slicing significantly reduced the signal heights. As for example, methamidophos, being polar compound, tailed on the non-polar primary column and as a result, the peak appeared in two slices with the first one being the dominant slice. Similarly, diazinon appeared in four slices at 2 s modulation period but, at 4 s, the most part of the peak was accommodated in single modulation cycle and thus, there was significant enhancement in S/N by at least two times. For,,, and - HCH (hexachlorocyclohexane) too, the S/N was highest at 4 s (Figure 3.3). At modulation period of 6 s, -HCH sliced between two modulation periods resulting in reduction in S/N by 38%. Similar slicing occurred for -HCH at 6 s modulation period. For fenchlorfos, however, the little slicing at 4 s could be completely resolved at 6 s modulation periods; but it disturbed the separation of other compounds. A modulation period of 4 s also gave the best separation of the Mix I (Figure 3.1B) and Mix II (Figure. 3.1B). A longer modulation period could not preserve the separation achieved on the primary column in certain cases, e.g. o,p- DDE co-eluted with buprofezin at modulation period of 8 s and buprofezin remained undetected in peak table. Our next endeavour was to optimize the duration of hot pulses within the 4 s modulation period. We tried six hot pulse durations, viz..4,.6,.8, 1., 1.2 and 1.6 s. The corresponding cold pulse durations were automatically adjusted by the software as 1.6, 1.4, 1.2, 1.,.8 and.4 s to ensure two complete cycles within the 4 s modulation period. When the hot pulse duration was too short (.4 s), in general, there was loss in sensitivity, which might have occurred as a result of relatively longer retention in secondary column, slower transfer of analytes from modulator into the rest of the secondary column and resultant relatively broader peaks. The resolution of co-eluted compounds, e.g. heptachlor (tr = 2.32, 2.16 s) and vinclozoline (tr = 2.36, 2.16 s) was poor, when the hot pulse duration was.4 s. But, these two peaks were completely separated when the hot pulse duration was increased to.8 s. The influence of hot pulse time on the nature of peak cluster of Mix I was quite interesting to note (Figure 3.1B). As heptachlor has the minimum polarity out of these four pesticides, it eluted after the three other peaks from the non-polar primary column but retained for least duration in the relatively polar secondary column. But, when hot pulse duration was 86

12 shorter (like.4 s), the relatively longer cold pulse time lead to increased cryofocusing and as a result, heptachlor took relatively longer time to get into the secondary column after modulation. This obviously affected its resolution with the closely eluting compound vinclozoline. At hot pulse time of.8 s, the peaks of heptachlor and vinclozoline were narrower because of faster elution and thus, their resolution improved to.5 against.19 at.4 s cold times indicating more efficient transfer of masses from the modulator to the secondary column. For buprofezin, when the hot pulse time was.4s, a part of the peak sliced into the next modulation period because of its longer retention in secondary column. But, when the hot pulse time was increased to.8 s, almost the entire peak of buprofezin appeared in single modulation cycle with consequent increase in signal height by 1.5 times. There was significant improvement in spectral purity of all the compounds. The NIST library matching for heptachlor, chlorpyrifos-methyl, parathion-methyl and vinclozoline increased from 81, 62, 65 and 8% in 1D to 91%, 9%, 93% and 9%, respectively, in GC GC TOFMS. Similarly, for myclobutanil, buprofezin, flusilazole and oxyfluorfen, the NIST library matching improved from 69, 68, 63 and 71% in 1D to 87%, 9%, 89% and 92% when analyzed by GC GC TOFMS. The library matching of all the pesticide peaks in matrix further improved (>9%) when compared to the own reference mass spectra created with the standards in this equipment. Thus, although the deconvolution algorithm could identify all the test compounds on 1D GC MS, but automated library searching significantly improved on GC GC TOFMS and this could be effectively utilized in minimizing chances of false detections and associated uncertainties in quantitative analysis at residue level as low as 1 ng/g Setting of Modulation Offset Optimization of the modulation-offset temperature is important to ensure effective transfer of compounds from modulator to the rest of the secondary column. We tried three different modulation offset temperatures, viz. 15, 25 and 35 o C. The S/N was monitored in all these cases and compared. At 15 o C offset temperature, peak shape and separation of certain compounds (e.g. heptachlor and vinclozoline) were not satisfactory. When the offset temperature was increased to 35 o C, again the 87

13 separation was affected. The resolution between heptachlor and vinclozoline reduced from.68 to.48 when the offset temperature was increased from 25 to 35 o C. The temperature 25 o C gave the most satisfactory separation as well as maximum S/N for most of the analytes and thus, selected for the test pesticide mixture Effect of Acquisition Rate on Analysis In general, the peak height decreased as the acquisition rate increased; although the resolution was higher at a faster acquisition rate because of more number of data points across a peak. As for example, for α-hch, the S/N decreased by 22% when the acquisition rate increased from 2 to 2 spectra/s. On the other hand, the resolution between β-hch and atrazine increased from.34 at 2 spectra/s to.4 and.5 at 5 and 1 spectra/s, respectively. The acquisition rate of 1 spectra/s was most suitable for detection and quantification of all the test compounds at 1 ng/g level. Furthermore, in GC GC TOF-MS, there is an option for resample, where the software can re-process an already run sample at a lower acquisition rate to enhance the signal height and peak find possibilities. Thus, in Figure 3.4, we observe that dichlorvos, which was not identifiable at 2 spectra/s, could be very well identified at 1 spectra/s with NIST library match as high as >88%. At low acquisition rate, resolution was lower and so, dichlorvos was completely merged with a matrix compound. This finding is especially important as β-phenylethyl acetate, the interfering compound, is a commonly identifiable volatile compound in grape juice and wine [16, 17], which elutes at almost same tr as dichlorvos. Thus, we could run a sample at 1 spectras to achieve optimum resolutions with subsequent re-processing of the same acquired run at a lower rate like 2 spectra/s or even lower to assure desired sensitivity. The final GC GC TOFMS method has been described in Section Practical Application in Multiresidue Analysis of Pesticides in Grapes The method was applied for grape matrix and gave satisfactory performance for multiresidue analysis of pesticides (Figure 3.5). Linearity of the calibration curve was established and it ranged between 1 and 1 ng g -1 for all the test chemicals in 88

14 matrix solution. The correlation coefficient (R 2 ) of the calibration graph, both pure solvent-based as well as matrix matched were >.99 for all the compounds. The matrix-matched standards were used for all quantification purposes to avoid any ambiguity. The limit of quantification (LOQ) on the basis of S/N 1:1 could be achievable at 1 ng g -1 or lower level for all the test compounds, which complies with the international requirements for lowest MRL [18 2]. As the matrix components were well separated from the pesticide peaks on GC GC, the matrix influence on S/N was not significant. The polar secondary column played a vital role in separating the co-eluting matrix compounds from the pesticides, which was especially significant for the analysis of early eluting compounds. This also signifies potential application of this technique in analyzing complex food and environment matrixes. A good separation between parathionmethyl and chlorpyrifos-methyl was important as the first compound is sometimes applied by growers to control soil insects, whereas, the second one is not a commonly used chemical in India. Similarly, chromatographic separation of myclobutanil and buprofezin offered a significant advantage on regulatory point of view as both are frequently applied pesticides in Indian viticulture and thus, unambiguous identification of their residues in mixture is essential. The sensitivity of analysis significantly improved with GC GC TOFMS when compared to the GC TOFMS analysis (Table 3.1). The S/N in 1D at 5 ng ml -1 with 1 spectra/s was comparable to the S/N at 1 ng ml -1 with 1 spectra/s in 2D analysis. The width of the peaks got narrower on GC GC and there was significant reduction in noise level also. As for example, for -HCH at 1 ng ml -1, the peak width decreased from 4 s at 1D to.2 s in GC GC with corresponding reduction in the noise level from 8 to 1. Thus, the S/N increased from 8 to 426, accounting for >5 fold increase on GC GC, when compared to GC TOFMS analysis. An increase in the temperature of modulation offset and secondary column could further reduce the peak width in GC GC to less than.2 s but it disturbed the separation of other compounds. The recovery data for all the compounds at 1 ng g -1 level were within the range of 7 11% with acceptable repeatability (Table 3.1) as per the international requirement [21,22]. We propose to conduct complete single 89

15 laboratory validation of this method in different fruits and vegetable matrixes in future attempts. 3.4 CONCLUSION Multiresidue analysis by GC GC TOFMS gave distinct advantages over the GC TOFMS analysis. A good separation of all co-eluted as well as closely eluted compounds allowed high sensitivity analysis of 51 pesticides within 24 min. with library searchable mass spectral confirmations. There was significant improvement in sensitivity when the analysis was performed in two-dimensional mode at1 spectra/s with at least two times enhancement in S/N for all compounds. For some compounds, even >1 times enhancement in the LOD was recorded. Spectral deconvolution and peak find options allowed library-based full scan screening at trace level. The technique shows promise in regulatory system in resolving any conflicting results of different laboratories and issues related to false positives/negatives. 3.5 REFERENCES [1] Zrostlıkova J, Hajslova J, Cajka T, J. Chromatogr. A 119, , (23). [2] Dalluge J, Beens J, Brinkman U A Th, J. Chromatogr. A 1, 69-18, (23). [3] Adahchour M, Beens J, Vreuls R J J, Max Batenburg A, Brinkman U A Th, J.Chromatogr. A 154, 47-55, (24). [4] [4] Harynuk J, Vlaeminck B, Zaher P, Marriott P. J. Anal. Bioanal. Chem. 386, 62, (26). [5] Van der Lee M K, van derweg G, Traag W A, Mol H G J, J. Chomatrogr. A 1186, 325-, (28). [6] APEDA, Regulation of export of fresh grapes from India through monitoring of pesticide residues, Amendments in grape RMP-27, Amendment-5 (Revised Annexure: 7&11), February 7, 27 (accessed on 3 th Septeber 27), 26a. [7] APEDA, Regulation of export of fresh grapes from India through monitoring of pesticide residues, Annexure 1: Circular No. 91-4/95 PQD, February 29, 2, 5/RMPGrapes27 17 oct6 3.pdf, 26b. 9

16 [8] Wong J W, Hennessy M K, Hayward D G, Krynitsky A J, Cassias I, Schenck F J, J.Agric. Food Chem. 55, , (27). [9] Mastovsk a K, Hajˇslov a J, Lehotay S J, J. Chromatogr. A 154, , (24). [1] Wong J W, Webster M G, Halverson C A, Hengel M J, Ngim K K, Ebeler S E, J. Agric. Food Chem. 51, , (23). [11] Walorczyk S, Gnusowski B, J. Chromatogr. A 1128, , (26). [12] Vidal J L M, Li ebanas F J A, Rodr ıguez M J G, Frenich A G, Moreno J L F, Rapid Commun. Mass Spectrom. 2, , (26). [13] Hoh E, Maˇstovsk a K, Lehotay S J, J. Chromatogr. A 1145, , (27). [14] Banerjee K, Oulkar D P, Dasgupta S, Patil S B, Patil S H, Savant R H, Adsule P G, J. Chromatogr. A 1173, 98-19, (27). [15] De Koning S, Lach G, Linkerhagner M, Loscher R, Tablack P H, Brinkman U. A. Th, J. Chromatogr. A 18, , (23). [16] Tamborra P, Martino N, Esti M, Anal. Chim. Acta 513, , (24). [17] Masoud W, Poll L, Jakobsen M, Yeast 22, , (25). [18] Germany, Maximum residue levels according to German legislation, January 24, 28. [19] The Netherlands, Maximum residue limits of pesticides in the Netherlands, December, 27. [2] United Kingdom, December 2, 27. [21] European Union Commission Directorate of General Health and Consumer protection, Guidance document on residue analytical methods, SANCO/825/ rev. 6, June 2, 2. [22] Horwitz W, Albert R, J. AOAC Int. 89, , (26). 91

17 Sr. No. Table 3.1: Comparative Multiresidue Analysis of Pesticides in GC TOFMS and GC GC TOF-MS LOD (ng/g) S/N in matrix LOD Name of Pesticide (Class*) m/z enhanced 1D (5 2D (1 1D 2D (1D/2D) 1spectra/s) 1spectra/s) 1. Aldrin (I) ±6 Recovery 1 ng/g (n = 8) 2. Atrazine(II) ±1 3. Azinphos-Me (III) ±1 4. Buprofezin (IV) ±4 5. Chlorfenvinfos (III) ±6 6. Chlorpyrifos (III) ±5 7. Chlorpyrifos-Me (III) ±1 8. Chlorothalonil (I) ±6 9. cis-chlordane (I) ±3 1. Diazinon (III) ±8 11. Dichlorvos (III) ±8 12. Dieldrin (I) ±5 13. Dimethoate (III) ±3 14. Ethion (III) ±3 15. Fenarimol (V) ±6 92

18 Sr. No. LOD (ng/g) S/N in matrix LOD Name of Pesticide (Class*) m/z enhanced 1D (5 2D (1 1D 2D (1D/2D) 1spectra/s) 1spectra/s) 16. Fenitrothion (III) ±6 Recovery 1 ng/g (n = 8) 17. Fipronil (VI) ±4 18. Flusilazole (V) ±3 19. Heptachlor (I) ±8 2. Iprodione (VII) ±5 21. Kresoxim-methyl (VIII) ±4 22. Malathion (III) ±3 23. Metalaxyl (IX) ±5 24. Methamidophos (III) ±6 25. Mevinphos (III) ±3 26. Monocrotophos (III) ±6 27. Myclobutanil (V) ±6 28. o,p DDE (I) ±6 29. o,p DDT (I) ±4 3. Oxyfluorfen (X) ±5 31. p,p DDE (I) ±5 93

19 Sr. No. LOD (ng/g) S/N in matrix LOD Name of Pesticide (Class*) m/z enhanced 1D (5 2D (1 1D 2D (1D/2D) 1spectra/s) 1spectra/s) 32. p,p DDT (I) ±9 Recovery 1 ng/g (n = 8) 33. Parathion-ethyl (III) ±3 34. Parathion-methyl (III) ±3 35. Penconazole (V) ±4 36. Phorate (III) ±4 37. Phosalone (III) ±5 38. Phosphamidon (III) ±6 39. Procymidone (VII) ±5 4. Propargite (XI) ±5 41. Propiconazole (V) ±8 42. Quinalphos (III) ±4 43. Tebuconazole (V) ±5 44. trans-chlordane (I) ±3 45. Triazophos (III) ±6 46. Vinclozolin (VII) ±5 47. α-hch (I) ±3 94

20 Sr. No. LOD (ng/g) S/N in matrix LOD Name of Pesticide (Class*) m/z enhanced 1D (5 2D (1 1D 2D (1D/2D) 1spectra/s) 1spectra/s) 48. β-hch (I) ±1 Recovery 1 ng/g (n = 8) 49. γ-hch (I) ±1 5. δ-hch (I) ±5 51. λ-cyhalothrin (XII) ±5 * Pesticide class designations: I, organochlorine; II, triazine; III, organophosphorus; IV, chitin synthesis inhibitor; V, triazole; VI, phenylpyrazole; VII, dicarboximide; VIII, strobilurine; IX, acylamino acid; X, nitrophenylether; XI, sulfite ester; XII, synthetic pyrethroid. Pesticides are arranged in alphabetical order. 95

21 (A) Chlorpyrifos-Me Vinclozoline Parathion methyl Heptachlor (B) Figure 3.1 Chromatographic Separation of Heptachlor (I), Vinclozoline (II), Chlorpyrifos methyl (III) and Parathion-methyl (IV): (A) Co-elution of Four Pesticides in GC TOFMS Analysis and (B) Baseline Separation of Heptachlor, Vinclozoline, Chlorpyrifos-methyl and Parathion-methyl on GC GC TOFMS Analysis 96

22 (A) Buprofezine Myclobutanil Flusilazole Oxyfluorfen Figure 3.2 Chromatographic Separation of Myclobutanil, Buprofezin, Flusilazole and Oxyfluorfen: (A) Co-elution of Four Pesticides in GC TOFMS and (B) Baseline Separation of Oxyfluorfen (I), Buprofezin (II), Flusilazole (III) and Myclobutanil (IV) on GC GC TOFMS Analysis 97

23 Modulation period = 2 S st Time (s) 2nd Time (s) Modulation period = 4 S α-hch β-hch γ-hch δ-hch st Time (s) 2nd Time (s) Modulation period = 6 S st Time (s) 2nd Time (s) Figure 3.3 Effect of Modulation Period on,,, and -HCH at 5 ng/g 98

24 Figure 3.4 Effect of Acquisition Rate on Peak Finds: Identification of Dichlorvos at 1 spectra/s. (A) Dichlorvos is Masked by -phenyl ethyl acetate and Not Detectable at 2 spectra/s and (B) Dichlorvos Detected at 1 spectra/s Figure 3.5 : Multiresidue Chromatogram of 51 Pesticides in Grape Matrix at 5 ng/ml 99