View
4
Download
0
Category
Preview:
Citation preview
Applic
atio
nNO
TE
Introduction Introduction
There are over 800 registered pesticides in use worldwide, and an ever-increasing requirement to monitor the levels of these compounds in foodstuffs. Under EU legislation, maximum residue limits (MRLs) are specified for pesticides considered a potential health risk; guidelines for developing analytical methods must be followed.
The analytical challenge is to maximize the number of pesticides, minimize the variety of methods, keep run times short and achieve limits of detection (LoDs) at or below the reporting level. Multi-residue methods are efficient for analysis of pesticide residues. For methods with a very wide scope, generic sample preparation procedures are employed. Inherent to this approach is that clean up of extracts is only possible to a limited extent. When applying such methods to complex matrices like baby food, herbs, spices and tobacco, enhanced selectivity in detection is required to make up for the low selectivity in sample preparation.
Whereas single quadrupole and ion trap MS instruments are suitable for simple matrices where LoDs of >0.01 mg/kg are required, these detection
systems provide insufficient selectivity for complex food matrices, such as baby food, garlic, ginger, herbs and spices. In this work, gas chromatography (GC) with tandem quadrupole mass spectrometric detection (GC/MS/MS) was used to achieve the required selectivity.
Requirements of the analytical method:
• A simple and generic, rapid extraction method, allowing recovery of multiple classes of pesticides (OP, OC, pyrethroids, etc.)
• Analysis of 100 or more pesticides in 1 run
• LoDs at or lower than the reporting level
• Need for targeting multiple compounds in a variety of produce/matrices
• Efficient use of time/instrumentation/personnel (Multiresidue methods)
• Selective MS detection method (required to compensate for less selective sample prep)
• Adequate sensitivity to keep amount of (dirty!) matrix introduced into the GC system as low as possible
David Douce1, Peter Hancock1, Stephanie Dudd1, Hans G.J. Mol 21Waters Corporation, Manchester M22 5PP, UK
2TNO, Zeist, The Netherlands
APPLICATION OF GC-TRIPLE QUADRUPOLE MS/MS FOR MULTI-RESIDUE ANALYSIS OF
PESTICIDES IN COMPLEX MATRICES
Figure 1. Waters® Micromass® Micromass® ® Quattro micro® Quattro micro® ™ GC. ™ GC. ™
Applic
atio
nNO
TE
Instrumentation
GC/MS/MS: Waters® Micromass® Quattro micro™ GC
GC: Agilent 6890 with PTV injector
Injection: 2 µl, solvent vent injection
Column: 30 m x 0.25 mm ID, 0.25 µm DB5MS
Carrier gas: Helium, 1.0 mL/min (constant flow)
Temperature program: 50 °C (2 min)
ramp 1 @25 °C/min – 150 °C ramp 2 @5 °C /min – 280 °C (4 mins) Total run time 36 mins
MS/MS: Waters Micromass Quattro micro GC operated in Multiple Reaction Monitoring (MRM) mode
Ionization: EI positive ion
Q2 pressure: 2.5*10-3 mbar (argon)
Collision energy: see Table 1
Sample preparation
Sample amount: 2.5–25 g
Extraction: 50 ml ethyl acetate
Clean up: PSA and GCB (dispersive)
Final extract: 0.05–0.5 g/ml
Method development
In order to maximize the response of the instrument for each residue the choice of precursor ion, product ion and collision energy were optimized.
Initially, the pesticide mixture was analysed in full scan mode (Figure 3).
QUADRUPOLE(MS1)
HEXAPOLECOLLISION CELL
QUADRUPOLE(MS2)
PHOTOMULTIPLIER
PHOSPHOR
PRE-FILTERPOST-FILTER
CONVERSIONDYNODE
Detector
REMOVEABLEION SOURCE
GC INTERFACE
GC OVEN
ISOLATIONVALVE
REMOVABLEINNER SOURCE
Figure 2. Waters Micromass Quattro micro GC instrument schematic.
200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 1050 1100 1150 1200 1250 1300Scan0
100
%
Figure 3. GC/MS full scan chromatogram (solvent standard of 100 pesticides).
Applic
atio
nNO
TE
dichlorvos 7,08 185 93 12 0,998 3mevinphos-cis 8,68 192 127 10 0,994 1mevinphos-trans 8,72 192 127 10 0,995 1ethoprophos 11,60 158 114 5 0,999 4chlorpropham 11,93 213 171 5 0,998 4cadusafos 12,40 159 131 5 0,998 5phorate 12,51 260 75 5 0,906 26HCH-alfa 12,65 181 145 10 0,997 8hexachlorobenzene 12,73 284 249 15 0,999 2dimethoate 13,05 229 87 5 0,996 12carbofuran 13,26 164 149 8 0,998 3HCH-beta 13,48 181 145 10 0,999 4dimethipin 13,48 118 58 7 0,997 25quintozene 13,51 295 237 15 0,997 11lindane 13,70 181 145 10 0,997 5diazinone 14,12 179 137 15 0,995 15chlorothalonil 14,22 266 133 25 0,996 5pentachloroaniline 15,14 165 194 20 0,995 5phosphamidon 15,43 264 127 10 0,995 8chlorpyriphos-methyl 15,66 286 93 18 0,995 4vinclozolin 15,82 212 145 20 0,998 6parathion-methyl 15,89 263 127 8 0,990 6alachlor 15,93 188 160 8 0,994 3carbaryl 16,10 144 115 20 0,997 2heptachlor 16,12 272 237 10 0,994 2pirimiphos-methyl 16,73 290 125 15 0,999 3MPCPS 16,84 296 263 12 0,998 2malathion 17,16 173 99 10 0,997 3chlorpyriphos 17,34 197 169 20 0,994 8aldrin 17,36 263 193 22 0,991 7fenthion 17,50 278 109 12 0,929 10parathion 17,64 291 109 10 0,986 21fenpropimorf 17,66 303 128 5 0,985 6triadimefon 17,75 208 181 6 0,996 6chlordane, oxy 18,76 185 121 10 0,998 5fipronil 18,76 367 213 25 0,996 5heptachlor-epoxide 18,90 353 257 15 0,994 11chlorfenvinphos 18,94 325 269 10 0,991 8phenthoate 19,10 274 121 7 0,995 5quinalphos 19,13 146 118 7 0,996 2triadimenol 19,27 168 70 5 0,990 31methidathion 19,59 145 85 5 0,996 1chlordane-alfa 19,61 373 266 15 0,995 5tetrachlorvinphos 19,87 329 109 18 0,994 3endosulfan-alfa 20,06 237 143 25 0,998 16chlordane-gamma 20,07 373 266 15 0,995 24prothiofos 20,66 267 239 5 0,989 6profenofos 20,85 337 267 8 0,992 5DDE, p,p' 20,95 246 176 22 0,997 1
Pesticide Retention Precursor Product Collision Corr.coef. LOD time ion ion energy (V) (r2) (pg)
dieldrin 21,04 237 165 20 0,976 5TDE, o,p' 21,15 235 165 15 0,973 3cinerin-I 21,47 123 81 7 0,997 4endrin 21,78 263 193 25 0,994 17nitrofen 21,81 283 253 10 0,998 30endosulfan-beta 22,20 237 143 25 0,983 22fensulfothion 22,28 292 109 10 0,993 9TDE, p,p' 22,45 235 165 15 0,999 1DDT, o,p' 22,54 235 165 15 0,981 4ethion 22,55 231 129 20 0,993 2jasmolin-I 22,80 164 109 8 0,996 17triazophos 23,12 257 162 5 0,994 11carbophenothion 23,47 342 157 10 0,991 8endosulfan-sulphate 23,60 387 241 15 0,986 44DDT-p,p' 23,83 235 165 20 0,971 4propargite 24,51 135 107 12 0,998 9piperonylbutoxide 24,79 176 131 10 0,993 4pyrethrin-I 25,20 133 105 8 0,999 12phosmet 25,49 160 77 20 0,990 3bromopropylate 25,71 341 183 15 0,996 2fenpropathrin 26,09 181 152 20 0,996 4tetradifon 26,60 229 201 12 0,999 4phosalone 26,86 182 111 10 0,994 5phenothrin 26,86 183 153 12 0,993 11azinphos-methyl 26,95 160 132 5 0,987 11cinerin-II 27,05 169 107 5 0,997 90cyhalothrin-lambda 27,73 197 141 10 0,996 3azinphos-ethyl 28,17 160 132 5 0,992 3acrinathrin 28,17 181 152 18 0,994 6jasmolin-II 28,25 163 121 6 0,995 23permethrin-cis 29,23 183 153 12 0,996 4permethrin-trans 29,50 183 153 12 0,997 3prochloraz 29,51 180 152 20 0,989 47cyfluthrin I 30,33 163 127 5 0,997 2cyfluthrin II 30,55 163 127 5 0,996 1cyfluthrin III 30,64 163 127 5 0,994 3cyfluthrin IV 30,75 163 127 5 0,993 5cypermethrin I 30,97 163 127 5 0,993 2cypermethrin II 31,18 163 127 5 0,996 3cypermethrin III 31,28 163 127 5 0,995 2flucythrinate I 31,32 199 157 5 0,994 1cypermethrin IV 31,38 163 127 5 0,997 2etofenprox 31,59 163 135 10 0,997 1flucythrinate II 31,73 199 157 5 0,997 1fenvalerate I 32,71 167 125 8 0,997 2fluvalinate,tau- 33,03 250 200 20 0,998 17fenvalerate II 33,18 167 125 8 0,997 4fluvalinate,tau- 33,22 250 200 20 0,994 4deltamethrin 34,48 181 152 18 0,998 6
Pesticide Retention Precursor Product Collision Corr.coef. LOD time ion ion energy (V) (r2) (pg)
Table 1. Precursor and products ions, linearity and LoDs.
Applic
atio
nNO
TE
The precursor ion was selected from the full scan The precursor ion was selected from the full scan spectra based on its relative abundance, e.g. Endosulfan-beta (MW = 404; Figure 4).
Method Performance
For the majority of pesticides, LoDs were below 10 pg on column (based on S/N >3:1; Table 1).
The ability to quantify closely eluting peaks is a prerequisite of multiresidue methods. Therefore, the dwell time allocated for each transition must be sufficient to ensure that at least 10 data points are acquired for accurate quantification. To assess the effect of short dwell times on data quality a standard solution of hexachlorobenzene was acquired using a range of dwell times.
Figure 6 illustrates that the signal intensity is unaffected by the shorter dwell time. With a 10 ms dwell time, the S/N measured is sufficient for quantification of the target compounds.
60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440m/z0
100
%
194.9
192.9
169.9158.9
101.9
89.0
84.9
75.062.9
61.0
74.1
99.0
89.9
103.0121.0
120.0
119.0
109.0
142.9
140.9156.9
171.9
190.9
172.9
236.8
234.8
196.9
206.9
228.8208.9
226.9
224.8
240.9
242.9
266.8
262.8
244.9
260.9
251.7
268.9
276.8278.8
338.7
336.8
306.8294.9
280.8
308.8 322.8
310.8325.3
340.7
405.7342.7
357.6 403.6359.8370.7
407.7
Figure 4. Full scan background subtracted spectrum of Endosulfan-beta.
The collision energy for the product ion was optimised using a range of collision energies between 5–30 eV (Figure 5).
50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240m/z0
100
%
0
100
%
0
100
%
0
100
%
0
100
%
0
100
%
Test011 85 (17.211) Cm (82:88) 6: Daughters of 237EI+ 2.61e4118.9
116.9105.9
94.769.967.6
63.852.1 89.777.682.398.1
107.9108.8
142.8140.8
131.9
129.8126.7
132.9
166.9164.9
143.3 162.9156.3 167.9 191.3175.6183.8 225.9203.0 216.5211.1 237.7231.6
Test010 86 (17.220) Cm (82:88) 6: Daughters of 237EI+ 4.95e4142.9
118.9116.9
Figure 5. MS/MS optimization.
10.20 10.40 10.60 10.80 11.00 11.20 11.40 11.60 11.80 12.00 12.20 12.40 12.60 12.80 13.00 13.20 13.40Time0
100
%
0
100
%
0
100
%
0
100
%
0
100
%
TNO028 2: MRM of 5 Channels EI+ 283.81 > 249
2.00e4x10 S/N:PtP=59.63
TNO028 2: MRM of 5 Channels EI+ 283.79 > 249
1.92e4x10 S/N:PtP=50.71
TNO028 2: MRM of 5 Channels EI+ 283.8 > 249
2.13e4x10 S/N:PtP=47.76
TNO028 2: MRM of 5 Channels EI+ 283.82 > 249
1.92e4x10 S/N:PtP=37.18
TNO028 2: MRM of 5 Channels EI+ 283.78 > 249
2.55e4x10 S/N:PtP=38.14
50 ms
40 ms
30 ms
20 ms
10 ms
Figure 6. 2 µl standard solution of hexachlorobenzene (10 pg/µl) acquired using a range of dwell times, from 10–50 ms.
Applic
atio
nNO
TE
To assess instrument robustness, hexachlorobenzene To assess instrument robustness, hexachlorobenzene (20 pg on-column) was analyzed repeatedly (n = 10) at a range of different dwell times, from 10–50 ms. The % RSD shown in Table 2 illustrates the repeatability of injection was less than 5%.
Hexachlorobenzene (20 pg)
Dwell (ms) % RSD (n = 10)
10 3.0 20 4.4 30 3.9 40 2.5 50 2.8
Table 2. Effect of dwell time on repeatability.
Figure 7. TIC of 100 pesticides in MRM mode. Each color corresponds to a different function window.
Compound name: DeltamethrinCorrelation coefficient: r = 0.998918, r^2 = 0.997838Calibration curve: 14.0112 * x + 125.819Response type: External Std, AreaCurve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: None
pg/µl0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500
0
1000
2000
3000
4000
5000
6000
7000
Compound name: Pirimiphos-methylCorrelation coefficient: r = 0.999511, r^2 = 0.999022Calibration curve: 20.2691 * x + -27.0132Response type: External Std, AreaCurve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: None
pg/µl0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
Re
spo
nse
Re
spo
nse
A
B
Figure 8. Calibration curves for A) Pyrimiphos-methyl (TrA) Pyrimiphos-methyl (TrA) Pyrimiphos-methyl (T = 16.73 mins) and r = 16.73 mins) and rB) Deltamethrin (TrB) Deltamethrin (TrB) Deltamethrin (T = 34.48 mins) showing r = 34.48 mins) showing rlinearity over the concentration range.
For 100 pesticide residues, the optimum experimental setup utilized 84 MRM transitions in 14 MRM function windows, with 4 to 8 transitions in each window (Figure 7). With this setup, at any point in time, the maximum number of transitions acquired was 15.
The linearity of the system is illustrated in Figure 8.
Results
The analytical method was applied to the analysis of a range of complex food extracts, including fresh produce, baby food (matrix equivalent = 0.5 g/ml), dried herbs, spices, tobacco, ginkgo and cannabis (matrix equivalent = 0.1 g/ml).
Applic
atio
nNO
TE
Figure 9 compares chromatograms from complex and non-complex matrices, showing the importance of high selectivity for the analysis of these types of sample. Figure 10 shows the GC/MS/MS chroma-tograms of selected pesticide residues in herbal tea.
In all cases (Table 3), a high percentage of the spiked pesticides were detected and measured, despite the highly complex matrices, generic sample preparation and limited clean up employed.
6.000 8.000 10.000 12.000 14.000 16.000 18.000 20.000 22.000 24.000 26.000 28.000 30.000 32.000rt0
100
%
6.744305 10.197
719
8.796551
8.471512
9.413625
13.7001139
23.3332294
19.0461780
15.8021391
15.4431348 17.979
1652
17.4201585
21.9742131
25.8762599
25.2092519
28.2362882
28.4112903
28.5782923
6.000 8.000 10.000 12.000 14.000 16.000 18.000 20.000 22.000 24.000 26.000 28.000 30.000 32.000rt0
100
%
22.4962026
10.204552
9.903516
8.761379
11.321686 19.260
163812.689850
15.0491133 17.451
1421
21.7791940
22.7632058
25.2652358
25.4402379
Figure 9. GC full scan chromatograms (TIC) of herbal tea, 0.2 mg on column (red trace) and strawberry, 10 mg on column (blue trace).
Figure 10. GC/MS/MS of selected pesticides in herbal tea (0.3 mg/kg dry).
Table 3. Summary of performance in terms of selectivity.
Product Pesticide fortification Compounds spiked Compounds found % found
Baby food 0.01 mg/kg 93 81 87Baby food 0.05 mg/kg 93 92 99Herbal tea 0.03 mg/kg 93 88 95Herbal mix 0.03 mg/kg 93 87 94Curry 0.15 mg/kg 93 90 97Massala 0.15 mg/kg 93 87 94Tobacco 0.15 mg/kg 93 87 94Ginko 0.02–4.0 mg/kg 52 44 85Cannabis 0.02–4.0 mg/kg 52 41 79
Conclusions
• A multiresidue method has been developed for the surveillance monitoring of 100 pesticide residues in a range of food matrices.
• Sufficient selectivity has been achieved to allow generic sample clean up, even for very complex food matrices.
• The Waters Micromass Quattro micro GC can be used in routine analysis at 10 ms dwell times. Subsequent expansion of this method to include an increased number of MRM transitions would be possible without a loss in sensitivity.
• The system provided good sensitivity (1–10 pg on column) for the majority of pesticide residues, allowing the amount of matrix introduced into the GC system to be minimized.
Applic
atio
nNO
TE
Waters, Micromass and Quattro micro are trademarks of Waters Corporation.All other trademarks are property of their respective owners.©2004 Waters Corporation Printed in the U.S.A. October 2004 720000987EN LL-PDF
WATERS CORPORATIONWATERS CORPORATION34 Maple St.Milford, MA 01757 U.S.A.T: 508 478 2000F: 508 872 1990www.waters.com
Sales Offices:AUSTRIA 43 1 877 18 07
AUSTRALIA 61 2 9933 1777
BELGIUM AND LUXEMBOURG 32 2 726 1000
BRAZIL 55 11 5543 7788
CANADA 800 252 4752 X2205
CZECH REPUBLIC 420 2 617 11384
DENMARK 45 46 59 8080
FINLAND 358 9 506 4140
FRANCE 33 1 3048 7200
GERMANY 49 6196 400600GERMANY 49 6196 400600GERMANY
HONG KONG 852 29 64 1800
HUNGARY 36 1 350 5086
INDIA 91 80 2837 1900
IRELAND 353 1 448 1500
ITALY 39 02 27 4211ITALY 39 02 27 4211ITALY
JAPAN 81 3 3471 7191
KOREA 82 2 820 2700
MEXICO 52 55 5524 7636
THE NETHERLANDS 31 76 508 7200
NORWAY 47 6 384 6050NORWAY 47 6 384 6050NORWAY
PEOPLE’S REPUBLIC OF CHINA 86 10 8451 8918
POLAND 48 22 833 4400
PUERTO RICO 787 747 8445
RUSSIA/CIS 7 095 931 9193
SINGAPORE 65 6278 7997
SPAIN 34 93 600 9300
SWEDEN 46 8 555 11 500
SWITZERLAND 41 62 889 2030
TAIWAN 886 2 2543 1898
UK 44 208 238 6100
US 800 252 4752
Recommended