100 Pesticides in Fruit and Veg
-
Upload
magicianchemist -
Category
Documents
-
view
14 -
download
2
Transcript of 100 Pesticides in Fruit and Veg
RAPID COMMUNICATIONS IN MASS SPECTROMETRY
Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
) DOI: 10.1002/rcm.3268
Published online in Wiley InterScience (www.interscience.wiley.comScreening and confirmation of 100 pesticides in food
samples by liquid chromatography/tandem mass
spectrometry
Imma Ferrer1*, E. Michael Thurman2 and Jerry A. Zweigenbaum3
1Pesticide Residue Research Group, University of Almeria, Almeria, Spain2Department of Environmental Engineering, University of Colorado, Boulder, CO, USA3Agilent Technologies, Inc., LFS, Wilmington, DE, USA
Received 27 July 2007; Revised 11 September 2007; Accepted 13 September 2007
*CorrespoUniversitE-mail: if
An analytical method for screening, quantitation and confirmation of a group of 100 pesticides in
vegetable and fruit samples was developed using liquid chromatography coupled with tandemmass
spectrometry (LC/MS/MS). The pesticides studied belonged to different chemical families
of herbicides, insecticides and fungicides; some degradation products were included as well. A
thorough optimization was performed for each analyte to achieve individual optimum fragmentor
and collision energy voltages. Two transitions per parent compound were monitored in a single
chromatographic run containing two time segments. A small particle size C18 column (1.8mm) was
used for the chromatographic separation of the mixture, providing very narrow peaks and allowing
an excellent separation of all the analytes in a 30-min period formaximumpeak capacity. Themethod
was validated with blank matrices of green pepper, tomato and orange spiked from 0.1 to 100mg/kg
with the pesticide mix. Quantitation was carried out using matrix-matched standard calibration and
linearity of response over 3 orders of magnitude was demonstrated (r > 0.99). Limits of detection
based on two transitions and ion-ratio requirements ranged between 0.3 and 50mg/kg. In general, the
sensitivity obtained meets the maximum residue levels (MRLs) established by the European Union
regulation for food monitoring programs. The analytical performance of the method was evaluated
for different types of vegetables and fruits, showing little or no matrix effects, and examples of
screening and confirmation of pesticides in these samples are shown here. Copyright # 2007 John
Wiley & Sons, Ltd.
It is well known that the presence of pesticide residues in
food can affect human health. Different organizations have
set stringent regulatory controls on pesticide use in order to
minimize exposure of the general population to pesticide
residues in food. For example, the European Union (EU) has
set newDirectives for pesticides at low levels in vegetables in
order to meet health concerns.1 For fruits and vegetables
intended for production of baby food, a maximum residue
limit (MRL) of 0.01mg/kg is applicable for all pesticides, and
compounds without a stated regulation have the lowest
MRLs at 0.01mg/kg as well.2
The low MRLs have encouraged the development of more
sensitive analytical methods to meet the requirements in
complex samples. Therefore, sensitive and reliable con-
firmatorymethods are required tomonitor pesticide residues
in foods. In this sense, liquid chromatography/tandemmass
spectrometry (LC/MS/MS) with a triple quadrupole in
multiple reaction monitoring (MRM) mode has become, so
far, the most widely used technique for monitoring and
quantitation of pesticides in food, as reported extensively in
ndence to: I. Ferrer, Pesticide Residue Research Group,y of Almeria, Almeria, [email protected]
the literature.3–13 The advantages of tandem mass spectrom-
etry are based on high sensitivity, reduction of sample-
treatment steps and reliable quantitation and confirmation at
the low required concentrations.11 The simplicity of meth-
odologies using the triple quadrupole as a detection
technique together with the low limits of detection achieved,
and the MS/MS capability, makes this technique a valuable
and unique tool for routine monitoring programs established
in regulatory official laboratories. The ease of use is some-
times an essential for these types of regulatory agencies,
which lack the highly skilled personnel required for more
sophisticated techniques.
The recent developments in stationary column phases,
such as the use of smaller particle sizes (1.7 and 1.8mm),11,14
have allowed improved peak resolution and, therefore,
increased sensitivity in chromatographic separations. The
van Deemter equation indicates that, as the particle size
decreases to less than 2.5mm, there is a significant gain in
efficiency and that efficiency does not diminish at increased
flow rates or linear velocities.15 This is especially useful when
Copyright # 2007 John Wiley & Sons, Ltd.
3870 I. Ferrer, E. M. Thurman and J. A. Zweigenbaum
the number of compounds is high since it allows the
separation and detection of all the compounds present in a
complex sample. The only requirement when coupling to
tandem mass spectrometry is to achieve rapid data
acquisition, so the improved resolution is not degraded.11
Currently more than 900 pesticides are used worldwide,
both legally and illegally, on food products and in the treat-
ment of soil and crops. Most of these pesticides have MRLs
for both food and water to protect consumers. The MRL
concentrations have to be monitored as part of the quality
control of food, especially fruits and vegetables; thus,
large-scale multi-residue methods with hundreds of pesti-
cides are needed for quality control. However, the ability to
monitor hundreds of pesticides in a single analysis is a
challenging problem both for chromatography and mass
spectrometry. Herewe evaluate a tandemmass spectrometry
methodology to not only screen but also to quantitate and
confirm 100 pesticides in a single analysis using a
combination of the new 1.8mm LC columns (for maximum
peak resolution) and two time segments with 100 transitions
per segment in order to have both a quantifier ion and
a qualifier ion, which satisfies the EU specifications for
unequivocal identification and confirmation by mass spec-
trometry.16 A validation study was carried out using
matrix-matched samples for quantitation and as alternative
to compensate effects of suppression or enhancement of
signal due to the matrix.
EXPERIMENTAL
Chemicals and reagentsPesticide analytical standards were purchased from both
Sigma (St. Louis, MO, USA) and Dr. Ehrenstorfer (Ausburg,
Germany) by Agilent and kindly supplied to us. Individual
pesticide stock solutions (approximately 1000mg/mL) were
prepared in pure acetonitrile or methanol depending on the
solubility of each individual compound, and stored at�188C.From these stock solutions, working standard solutions
were prepared by dilution with acetonitrile and water.
HPLC-grade acetonitrile and methanol were obtained from
Merck (Darmstadt, Germany). Formic acid was obtained
from Fluka (Buchs, Switzerland). A Milli-Q-Plus ultra-pure
water system from Millipore (Milford, MA, USA) was used
throughout the study to obtain the HPLC-grade water used
during the analyses.
Sample preparationVegetable and fruit samples were obtained from local
markets. ’Blank’ vegetable and fruit extracts were used to
prepare the matrix-matched standards for validation pur-
poses. In this way, two types of vegetables and one fruit
(green peppers, tomatoes and oranges) were extracted using
the QuEChERS method described previously.17,18 The final
vegetable extracts were spiked with the mix of standards at
different concentrations (ranging from 0.1 to 100mg/kg) and
subsequently analyzed by LC/MS/MS. The scope of this
work was simply to develop a method for the screening,
quantitation and confirmation of a large number of pesticides
in vegetable and fruit matrices, not to study the recovery of
this type of compounds from food matrices. This has already
Copyright # 2007 John Wiley & Sons, Ltd.
been reported in various studies6,11,12 and, as the response of
every pesticide in the raw sample highly depends on the
recovery obtained, we decided that to spike directly the
matrices and to study the behavior of all the analytes was a
more fair, clear and broad comparison for this specific study.
LC/MS/MS analysesThe separation of the analytes was carried out using an
HPLC system (consisting of vacuum degasser, autosampler
and a binary pump, Agilent Series 1200; Agilent Technol-
ogies, Santa Clara, CA, USA) equipped with a reversed-
phase C18 analytical column of 150mm� 4.6mm and 1.8mm
particle size (Zorbax Eclipse SB-C18). Column temperature
was maintained at 258C. The injected sample volume was
10mL. Mobile phases A and B were acetonitrile and water
with 0.1% formic acid, respectively. The optimized chroma-
tographic method started at an initial mobile phase
composition of 10% A and gradually increased to 98% in
28min, then to 100% A in a further 2min, and finally held at
100% for 1min. The flow rate used was 0.6mL/min. A low
flow rate and a wide diameter column were used in order to
stretch the run time and allow improved separation between
the 100 compounds. A 10-min post-run time was used after
each analysis. This HPLC system was connected to a Agilent
6410 triple quadrupole mass spectrometer (Agilent Tech-
nologies) equipped with an electrospray ionization (ESI)
interface operating in positive ion, using the following
operating parameters: capillary voltage: 4000V; nebulizer
pressure: 40 psig; drying gas: 9 L/min; gas temperature:
3508C; dwell time: 10ms. The data recorded was processed
with the Mass Hunter software.
RESULTS AND DISCUSSION
Optimization of LC/MS/MS conditionsA preliminary study of the optimal MRM conditions for
every compound was carried out by injecting individual
standard solutions (1mg/mL) in acetonitrile/water (50:50).
All the compounds analyzed were detected with the ESI
source in positive ion mode. Full-scan spectra were acquired
first to optimize the collision-induced dissociation (CID)
fragmentation applied at the source via the fragmentor
voltage in order to obtain the maximum sensitivity for the
protonated molecule. Typically, the protonated molecule
was used as a precursor ion, except for one case (aldicarb, for
which a fragment ion had to be selected instead due to the
lack of protonated molecule in the spectrum). Secondly, MS/
MS spectra in product ion mode of operation were acquired
to obtain information on fragment ions. Once the product
ions (two) had been selected for every analyte, an MRM
experiment was carried out to select the optimum collision
energy for every specific transition. Various collision energies
(from 5 to 30 eV) were investigated. The optimum energies
were those ones that gave the best sensitivity for every
transition, and they were selected as optimum ones. An
example of this optimization is shown in Fig. 1 for linuron.
The product ion at m/z 182 was most sensitive at 15 eV and
the ion at m/z 160 was slightly more sensitive at 20 eV. This
process was carried out for all the other analytes studied and
specific information was obtained for each transition. Table 1
Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Figure 1. Example of optimization of the collision energy for linuron.
Screening of 100 pesticides in food samples by LC/MS/MS 3871
summarizes the most relevant MS settings such as fragmen-
tor voltage and collision energies used for each one of the 100
compounds investigated, as well as all the MRM transitions
selected for screening, quantitation and confirmation. These
were the MRM transitions selected in the validation study
with real matrices. The first transition shown was used for
quantitation (calibration curves and reproducibility) and the
second transition was used for confirmatory purposes and to
calculate limits of detection as well.
Chromatographic separation and dwell timeA slow gradient starting with 10% acetonitrile up to 100% in
30min was applied, which was first developed by our
group19 and has proven to be successful for the separation of
pesticide compounds. The use of a small particle size column
leads to very narrow peaks (approx. 5 s), in comparison with
conventional columns of 5mmwith peak widths of 15–20 s.20
The analytical column packed with non-porous 1.8mm
particles enabled elution of sample components in much
narrower, more concentrated, bands, resulting in better
chromatographic resolution and increased peak height. The
typical peak width was 5–10 s at base, thus permitting very
good separation of all compounds in only 30min. The
optimized gradient adequately resolved the geometric
isomers of bromuconazole, difeconazole, dimethomorph
and propiconazole (see Table 1).
Two different time segments were recorded in the
chromatographic run (each one of them containing about
half of the pesticides studied). In this way, a total of 100
transitions were monitored in each segment (50 compounds
per segment with two transitions each). The time segment
change was set up at 20min. Figure 2 shows the
Copyright # 2007 John Wiley & Sons, Ltd.
chromatogram corresponding to the analysis of a tomato
sample spiked with the 100 pesticide mix, at a concentration
equivalent to 100 pg on column for all analytes (or 0.01mg/
kg). Extracted ion chromatograms are overlaid for each one
of the target analytes according to their respective MRM
quantifying transition.
In this study, the effect of reduced dwell times on mass
spectral quality and sensitivity was assessed in the devel-
opment of an analytical method for quantitative and
confirmatory purposes as well. Different dwell times were
tested between 5 and 20ms (Fig. 3). When higher dwell times
of 50 and 100ms (results not shown) were used the number
of data points were not enough to define a good chromato-
graphic peak. When the dwell time was between 5 or 10ms
the peak shape improved becoming Gaussian. Usually, a
chromatographic peak needs at least 10 data points to be
Gaussian. If we take into account that a peak is 10 s wide by
default, then a dwell time of 10ms is theoretically needed for
a total of 100 transitions to have an average of 10 data points
in each peak (1 data point per second). For this reason, a
common dwell time of 10ms was chosen for all the
transitions monitored.
MRM ion ratiosConfirmation of positive identifications in real samples
requires the additional second MRM transition and the
evaluation of ion ratios between the two monitored
transitions as compared to a reference standard. The con-
firmation criteria using tandem mass spectrometry cover a
range of maximum permitted tolerances for relative ion
intensity, expressed as a percentage of the intensity of the
most intense transition.21 Figure 4 shows the ion ratios for
Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Table 1. Retention times, MRM transitions and MS operating parameters selected for analysis of the 100 pesticides
Compound RT (min) Fragmentor voltage MRM transition (m/z) Collision energy (eV) LOD� (mg/kg)
Acetamiprid 12.2 80 223> 126 15 0.3223> 56 15
Acetochlor 23.1 120 270> 224 10 0.8270> 148 10
Alachlor 23.1 80 270> 238 10 0.8270> 162 15
Aldicarb 14.3 80 116> 89 5 2116> 70 5
Aldicarb sulfone 7.9 80 223> 76 5 5223> 148 5
Aldicarb sulfoxide 6.1 80 207> 89 5 2207> 132 5
Atrazine 17.5 120 216> 174 15 0.4216> 132 20
Azoxystrobin 21.3 120 404> 372 10 0.3404> 344 15
Benalaxyl 24.4 120 326> 148 10 0.5326> 294 5
Bendiocarb 16.5 80 224> 109 10 1224> 167 5
Bensulfuron-methyl 19 120 411> 149 20 0.4411> 182 15
Bromoxynil 17.9 120 278> 199 30 40278> 223 30
Bromuconazole 21.5þ 22.5 80 376> 159 20 1376> 70 20
Buprofezin 26.6 120 306> 201 10 0.7306> 116 15
Butylate 27.7 120 218> 57 10 5218> 156 10
Carbaryl 17.4 80 202> 145 5 10202> 117 10
Carbendazim 7.1 80 192> 160 15 0.5192> 132 20
Carbetamide 13.9 80 237> 118 10 0.5237> 192 5
Carbofuran 16.6 120 222> 165 10 0.9222> 123 15
Chlorfenvinphos 23.7 120 359> 155 10 2359> 127 15
Chlorotoluron 16.8 120 213> 72 20 0.3213> 140 20
Chlorpyrifos methyl 25.9 80 322> 125 15 10322> 290 15
Cyanazine 15.3 120 241> 214 15 2241> 174 15
Cyproconazole 20.3 120 292> 70 15 0.5292> 125 15
Cyromazine 3.4 120 167> 85 25 10167> 125 20
Deethylatrazine 11.2 120 188> 146 15 1188> 104 20
Deethylterbuthylazine 15.4 120 202> 146 15 0.8202> 110 20
Deisopropylatrazine 8.7 120 174> 96 15 4174> 132 15
Diazinon 25.3 160 305> 169 20 0.3305> 153 20
Dichlorvos 15.4 120 221> 109 15 5221> 145 15
Difeconazole 24.7þ 24.9 160 406> 251 20 0.3406> 337 15
Difenoxuron 18 120 287> 72 20 0.6287> 123 15
Diflubenzuron 22.3 80 311> 158 10 6311> 141 15
Dimethenamide 21.2 120 276> 244 10 0.4
(Continues)
Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
3872 I. Ferrer, E. M. Thurman and J. A. Zweigenbaum
Table 1. (Continued)
Compound RT (min) Fragmentor voltage MRM transition (m/z) Collision energy (eV) LOD� (mg/kg)
276> 168 15Dimethoate 11.8 80 230> 199 5 0.7
230> 171 10Dimethomorph 19.2þ 19.6 120 388> 301 20 0.6
388> 165 25Diuron 17.1 120 233> 72 20 0.8
233> 160 20Ethiofencarb 17.9 80 226> 107 5 0.7
226> 164 5Fenamiphos 20.8 120 304> 217 20 0.6
304> 234 15Fenuron 11.2 120 165> 72 15 1.5
165> 120 15Flufenacet 23.0 80 364> 152 10 0.5
364> 194 5Flufenoxuron 27.6 80 489> 158 10 5
489> 141 15Fluometuron 17.9 120 233> 72 20 1
233> 160 20Fluroxypyr 14.9 80 255> 209 10 10
255> 181 15Hexaflumuron 25.1 120 461> 158 10 7
461> 141 20Hydroxyatrazine 8.1 120 198> 156 15 4
198> 86 20Imazalil 18.5 160 297> 159 20 10
297> 255 20Imazapyr 9.2 160 262> 217 15 0.7
262> 234 15Imazaquin 15.4 160 312> 199 25 0.6
312> 267 20Imidacloprid 11.4 80 256> 175 10 4
256> 209 10Ioxynil 19.6 120 372> 118 30 20
372> 245 30Iprodione 22.6 120 330> 245 10 12
330> 288 10Irgarol 1051 19.2 120 254> 198 15 0.8
254> 156 20Irgarol metabolite 13.6 120 214> 158 15 1.2
214> 110 20Isofenphos 26.4 80 346> 217 20 1
346> 245 10Isoproturon 17.7 120 207> 72 15 1.3
207> 165 15Lenacil 15.5 80 235> 153 10 8
235> 136 15Linuron 20.7 120 249> 160 20 1
249> 182 15Lufenuron 26.8 80 511> 158 10 3
511> 141 20Malathion 22.7 80 331> 99 10 0.8
331> 127 5Mebendazole 14.8 120 296> 264 20 0.6
296> 105 25Metalaxyl 17.7 120 280> 192 15 1
280> 220 10Metamitron 10.6 120 203> 175 15 0.9
203> 104 20Methidathion 20.8 80 303> 85 10 0.7
303> 145 5Methiocarb 20.4 80 226> 121 10 0.8
226> 169 5Methiocarb sulfone 13.2 80 258> 122 5 30
258> 217 10Methomyl 8.6 80 163> 88 5 0.8
163> 106 5
(Continues)
Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Screening of 100 pesticides in food samples by LC/MS/MS 3873
Table 1. (Continued)
Compound RT (min) Fragmentor voltage MRM transition (m/z) Collision energy (eV) LOD� (mg/kg)
Metolachlor 23.2 120 284> 252 10 0.4284> 176 15
Metolcarb 15.3 80 166> 109 5 2166> 91 10
Metribuzin 15.9 120 215> 187 15 1215> 131 20
Molinate 22.2 120 188> 126 10 2188> 83 15
Monuron 14.9 120 199> 72 15 1.5199> 126 15
Nicosulfuron 13.7 120 411> 182 15 0.8411> 213 10
Nitenpyram 11.0 120 271> 225 10 0.7271> 99 15
Oxadixyl 14.9 80 279> 219 10 5279> 102 10
Parathion ethyl 24.6 120 292> 236 10 5292> 264 5
Pendimethalin 28.5 80 282> 212 5 4282> 194 10
Phosmet 21.2 80 318> 160 5 6318> 133 5
Prochloraz 23.2 80 376> 308 10 5376> 266 10
Profenofos 26.6 120 373> 303 15 5373> 345 10
Promecarb 20.9 80 208> 109 10 0.7208> 151 5
Prometon 14.0 120 226> 142 20 2226> 184 20
Prometryn 18.3 120 242> 158 20 0.9242> 200 20
Propachlor 19.1 80 212> 170 10 1212> 152 15
Propanil 19.8 120 218> 127 20 0.8218> 162 15
Propiconazole 23.7þ 24.0 120 342> 159 20 0.7342> 69 20
Prosulfocarb 27.1 120 252> 91 15 0.6252> 128 10
Simazine 14.9 120 202> 132 20 0.7202> 124 20
Spiromesifen 30.1 80 371> 273 5 7371> 255 20
Sulfosulfuron 18.4 120 471> 211 10 0.8471> 261 15
Teflubenzuron 25.6 80 381> 158 10 9381> 141 15
Terbuthylazine 20.5 120 230> 174 15 0.3230> 132 20
Terbutryn 18.6 120 242> 186 15 1242> 71 20
Thiabendazole 7.8 120 202> 175 30 6202> 131 30
Thiacloprid 14.0 120 253> 126 15 2253> 186 10
Thiocyclam 6.3 120 182> 137 10 50182> 73 20
Triazophos 22.9 120 314> 162 20 0.6314> 286 10
Triclocarban 25.2 120 315> 162 15 2315> 128 15
Trifloxystrobin 26.1 120 409> 186 15 0.4409> 206 10
Triflumizole 24.9 80 346> 278 5 3346> 73 10
RT: retention time.�LODs were calculated for all 100 compounds spiked in a green pepper matrix sample.
Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
3874 I. Ferrer, E. M. Thurman and J. A. Zweigenbaum
Figure 2. (a) Total ion chromatogram (TIC) corresponding to the analysis of the mix of 100 pesticides in a tomato matrix
(concentration: 0.01mg/kg or 100 pg on-column) using a 1.8mmC18 column. (b) Extracted ion chromatograms for the quantifying
MRM transition of each analyte.
Screening of 100 pesticides in food samples by LC/MS/MS 3875
benalaxyl in solvent and in an extract of green pepper spiked
with the pesticide mix at 50mg/kg (500 pg on-column). The
m/z 148 ionwas used for quantitation and them/z 294 ionwas
used as the qualifier ion, with a window set at �25% for an
ion ratio of 45. As shown in Fig. 4, in the two ion profiles
benalaxyl was easily identified in this complex matrix due to
the selectivity of the MRM transitions and instrument
sensitivity. Confirmation of the identity of the pesticides
in real samples is usually based in ion-ratio statistics for
the transitions monitored.22 Table 2 shows the ion ratios
obtained for all the compounds studied in solvent (expected
Copyright # 2007 John Wiley & Sons, Ltd.
ratio) and in a green pepper matrix (observed ratio) at
50mg/kg concentration. As can be observed in this table
almost all the compounds can be confirmed in a vegetable
matrix sample by LC/MS/MS at the low-mg/kg level, being
the ion ratios in vegetable matrices within the tolerances
specified in the EU directives.22 The exceptions in the table
(marked with asterisks) are those cases where the analyte
was not sensitive enough (such is the case of bromoxynil,
ioxynil, lufenuron and methiocarb sulfone). In the case of
carbofuran the problem was the presence of an interferent
peak at the same retention time and with the same qualifying
Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Figure 3. Effect of dwell time on atrazine.
Figure 4. Ion ratios for benalaxyl in (a) solvent and (b) green pepper matrix. Concentration: 50mg/kg.
3876 I. Ferrer, E. M. Thurman and J. A. Zweigenbaum
ion transition. For prochloraz the problem was the poor
chromatography that this compound experiences in C18
columns due to the basicity of the protonated molecule.
Analytical performance
(a) Quantitation and confirmationQuantitation was performed using calibration with matrix-
matched standards to prevent slight variations in the signal
for some analytes and possible enhancement or suppression
of the signal from vegetable samples as compared to pure
Copyright # 2007 John Wiley & Sons, Ltd.
solvent.19 For quantitation and confirmation purposes the
peak areas of both transitions (quantifier and qualifier) were
measured using the automated Mass Hunter quantitation
software. Using this approach, samples can be quantified
automatically by the use of batches, which include the files of
the calibration standards selected. An example is shown in
Fig. 5 for the neonicotinoid pesticide acetamiprid in a green
pepper matrix. As can be observed in this figure, a range of
values (retention times, peak areas and calculated concen-
trations) are obtained aswell as information on ion ratios and
calibration curve data. Values in light blue in the software
Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Table 2. LC/MS/MS ion ratios (intensity of qualifying ion/intensity of quantifying ion) for the 100 pesticides spiked into solvent and
in a green pepper matrix sample at 50mg/kg
Compound Expected ratioa Observed ratiob Acceptable limit (%)c
Acetamiprid 35 37 25Acetochlor 47 45 20Alachlor 94 93 20Aldicarb 74 88 20Aldicarb sulfone 77 73 20Aldicarb sulfoxide 50 59 20Atrazine 20 19 25Azoxystrobin 19 19 30Benalaxyl 45 45 25Bendiocarb 50 51 25Bensulfuron-methyl 26 23 25Bromoxynil� 67 44 20Bromuconazole 18 17 30Buprofezin 54 58 20Butylate 5 3 50Carbaryl 6 7 50Carbendazim 6 5 50Carbetamide 35 38 25Carbofuran� 24 78 25Chlorfenvinphos 80 69 20Chlorotoluron 7 6 50Chlorpyrifos methyl 8 11 50Cyanazine 10 10 30Cyproconazole 24 25 25Cyromazine 58 55 20Deethylatrazine 19 18 30Deethylterbuthylazine 6 5 50Deisopropylatrazine 76 72 20Diazinon 55 52 20Dichlorvos 3 4 50Difeconazole 8 9 50Difenoxuron 47 44 25Diflubenzuron 49 53 25Dimethenamide 39 36 25Dimethoate 81 85 20Dimethomorph 42 43 25Diuron 4 4 50Ethiofencarb 7 7 50Fenamiphos 12 11 30Fenuron 2 2 50Flufenacet 51 47 20Flufenoxuron 14 16 30Fluometuron 4 5 50Fluroxypyr 97 90 20Hexaflumuron 70 65 20Hydroxyatrazine 45 44 25Imazalil 14 19 30Imazapyr 20 18 30Imazaquin 58 53 20Imidacloprid 76 69 20Ioxynil� 20 31 30Iprodione 6 5 50Irgarol 1051 5 5 50Irgarol metabolite 3 3 50Isofenphos 32 37 25Isoproturon 8 7 50Lenacil 2 1 50Linuron 76 67 20Lufenuron� 70 41 20Malathion 45 53 25Mebendazole 21 20 25Metalaxyl 63 60 20Metamitron 38 37 25Methidathion 55 51 20Methiocarb 80 79 20
(Continues)
Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Screening of 100 pesticides in food samples by LC/MS/MS 3877
Table 2. (Continued)
Compound Expected ratioa Observed ratiob Acceptable limit (%)c
Methiocarb sulfone� 5 0.2 50Methomyl 46 48 25Metolachlor 25 27 25Metolcarb 5 5 50Metribuzin 16 16 30Molinate 41 46 25Monuron 5 4 50Nicosulfuron 15 19 30Nitenpyram 61 58 20Oxadixyl 18 20 30Parathion ethyl 13 12 30Pendimethalin 27 28 25Phosmet 1 0.3 50Prochloraz� 20 38 30Profenofos 12 15 30Promecarb 61 56 20Prometon 39 37 25Prometryn 30 29 25Propachlor 6 7 50Propanil 94 107 20Propiconazole 53 47 20Prosulfocarb 10 8 30Simazine 82 76 20Spiromesifen 93 81 20Sulfosulfuron 75 73 20Teflubenzuron 40 50 25Terbuthylazine 10 9 50Terbutryn 6 6 50Thiabendazole 63 70 20Thiacloprid 4 3 20Thiocyclam 58 49 20Triazophos 3 3 50Triclocarban 89 73 20Trifloxystrobin 15 14 30Triflumizole 55 57 20
a Ratio determined from analysis of standard solutions (n¼ 3).b Ratio determined from analysis of spiked samples of green pepper (n¼ 3).c Reference 22 sets the criteria for the observed ratio as follows: expected ratio >50%, observed ratio should be within �20%; expected ratio>20–50%, observed ratio should be within �20%; expected ratio >10–20%, ratio should be within �30%; expected ratio �10%, ratio should bewithin �50%.�Compounds that exceeded the criteria.
3878 I. Ferrer, E. M. Thurman and J. A. Zweigenbaum
(shown in white here) denote values between the expected
ion-ratio ranges, whereas values in red in the software
(shown in dark grey here) represent outlier values. In this
figure for example the outlier value of 50.1 corresponds to
the ion ratio that exceeded the theoretical value at this
low concentration. The expected ratio should be 35 for
acetamiprid, so the tolerated values (according to the 25%
tolerance) would be between 26.2 and 43.7.
(b) Linearity and limits of detectionLinearity was evaluated by analyzing the standard solutions
at six different concentration levels in the range from 0.1 to
100mg/kg. For the calibration curves only the area of the
quantifying transition was taken into account. As an
example, the calibration curves generated for four com-
pounds are shown in Fig. 6. As can be observed in this figure,
the linearity of the analytical response across the studied
range is excellent, with correlation coefficients higher than
0.998. In general, matrix-matched calibration curves were
linear between the concentrations studied with correlation
coefficients higher than 0.99 for all the matrices tested, with
Copyright # 2007 John Wiley & Sons, Ltd.
the exception of those analytes with higher limits of detection
(LODs).
The LODs were estimated from the injection of matrix-
matched solutions at concentration levels corresponding to a
signal-to-noise ratio of about 3 for the quantitation ion and
presence of the confirmatory ion aswell. The results obtained
for all the standards spiked in a green pepper matrix are
included in Table 1. These LODs are higher and more reali-
stic than others reported for only one transition23 since it not
only takes into account the main transition, but also confirms
the compound with the second MRM transition. The best
LODs were obtained for acetamiprid, azoxystrobin, chloroto-
luron, diazinon, difeconazole, and terbuthylazine (0.3mg/kg
or 3pg on-column) and the highest LODs were for bro-
moxynil, ioxynil, methiocarb sulfone and thiocyclam (above
20mg/kg or 200 pg on-column). These are compounds that
do not ionize well under electrospray conditions in positive
ion mode. In some cases (i.e. imazalil, oxadixyl) the shape of
the peak from the separation in the C18 column is also the
cause for the poor LODobtained. In general, the LODs for the
100-pesticide mix met the requirements regarding the MRLs
Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Figure
5.Example
ofautomatedquantitationfortheneonicotinoid
pesticideacetamipridin
greenpeppermatrices(areasandexperimental
concentrationsare
shown).Ionratiosandcalibrationcurveare
shownaswell.
Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Screening of 100 pesticides in food samples by LC/MS/MS 3879
Figure
6.Calibrationcurvesforcarbofuran,deethylterbuthylazine,fenuronandparathion-ethylin
orangematrix
usingalinearfitwithno
weightingandnoorigin
treatm
ent.
Copyright # 2007 John Wiley & Sons, Ltd. Rapid Commun. Mass S
3880 I. Ferrer, E. M. Thurman and J. A. Zweigenbaum
pectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
Figure 7. MRM extracted chromatograms of a tomato sample for (a) buprofezin and (b) cyproconazole. Ion ratios are included.
Screening of 100 pesticides in food samples by LC/MS/MS 3881
imposed by the existing EU regulations which are close
to 10mg/kg.
The precision and repeatability of the method was
assessed using a 0.05mg/kg matrix-matched standard. Peak
area values were obtained for the quantifying ion, yielding a
relative standard deviation of 6% (median RSD) and a mean
RSD of 6.7% (n¼ 5).
(c) Matrix effectsMatrix effects are common in vegetable and fruit samples
due to the presence of natural products in such samples.
Matrix effects typically mean suppression; however, they
also mean matrix interferences that are present in the sample
and, hence, they show up in the detector. These effects can be
acute when using full-scan mass spectrometry techniques
such as the case of time-of-flight mass spectrometry and, in
fact, they have been reported in our previous work.19 When
using tandem mass spectrometric techniques these effects
are further minimized due to the higher selectivity of the two
MRM transitions.6,11 The presence of interfering peaks in the
chromatogram is diminished under these conditions. In this
study the calibration curves in solvent were plotted against
those in matrix and no significative differences in slope were
observed. This demonstrates that the use of LC/MS/MSwith
two MRM transitions is a highly selective technique that
discriminates any interferences present in the matrix. So,
standards in pure solvent could, in principle, be used for
quantitation. However, we prefer the use of matrix-matched
standards for quantitation in order to match as closely as
possible the samples and the standards used for quantitation.
Application to vegetable matricesTo confirm the suitability of the method for analysis of real
samples, several vegetable and fruit samples were analyzed
(green peppers, tomatoes and oranges). Positive findings for
three or more pesticides were obtained for all the samples
analyzed, but always at concentrations below 10mg/kg.
Figure 7 shows the analysis of a tomato sample analyzed by
Copyright # 2007 John Wiley & Sons, Ltd.
the methodology described in this work. As can be observed
in the twoMS/MS extracted product ion chromatograms, for
buprofezin and cyproconazole, these two compounds could
be easily identified in these complex matrices due to the
selectivity of the two MRM transitions, thus fulfilling the
regulation limits imposed by the EU directives. Furthermore,
confirmation of these analytes was possible according to
themaximumpermitted tolerances for relative ion intensities
(ion ratios) as shown in this figure.
CONCLUDING REMARKS
A study to evaluate the effectiveness of LC/MS/MS for
screening, quantitation and confirmation of different families
of pesticides in vegetable and fruit samples was carried out.
The methodology developed in this work allowed the
screening of a large-scale number of pesticides, commonly
used in agricultural practices. Two MRM transitions were
monitored for each pesticide and ion-ratio information was
acquired for confirmatory purposes. We have demonstrated
that the sensitivity and selectivity achieved with this
methodology are appropriate for large-scale multi-residue
analysis of pesticides in food samples according to the
requirements imposed by the EU regulations. The potential
of the proposed method was established by analysis of
vegetable and fruit samples showing the presence of several
pesticides in these samples.
AcknowledgementsAnalytical standards, chromatographic columns and sol-
vents were kindly provided by Agilent Inc. All the analytical
work described in this manuscript was carried out by the
authors at the Agilent facilities in Little Falls (Wilmington,
Delaware, USA) in May of 2006. Travel funding from
Almeria to Wilmington was provided by Dr. Jerry Zweigen-
baum. Paul Zavitsanos is thanked for helpful discussions
and support with the instrument and Yanyan Fang for help
in sample preparation.
Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm
3882 I. Ferrer, E. M. Thurman and J. A. Zweigenbaum
REFERENCES
1. Available: http://europa.eu.int/comm/food/index_en.htm.2. Commission Directive 2003/13/EC of 10 February 2003
amending Directive 96/5/EC on processed cereal-basedfoods and baby foods for infants and young children, OfficialJournal L41/33.
3. Alder L, Greulich K, Kempe G, Vieth B. Mass Spectrom. Rev.2006; 25: 838.
4. Hernandez F, Pozo OJ, Sancho JV, Bijlsma R, Barreda M,Pitarch E. J. Chromatogr. A 2006; 1109: 242.
5. Kovalczuk T, Jech M, Poustka J, Hajslova J. Anal. Chim. Acta2006; 577: 8.
6. Leandro CC, Hancock P, Fussell RJ, Keely BJ.J. Chromatogr. A 2006; 1103: 94.
7. Pizzutti IR, de Kok A, Zanella R, Adaime MB, HiemstraM, Wickert C, Prestes OD. J. Chromatogr. A 2007; 1142:123.
8. Hetherton CL, Sykes MD, Fussell RJ, Goodall DM. RapidCommun. Mass Spectrom. 2004; 18: 2443.
9. Jansson C, Pihlstrom T, Osterdahl BG, Markides KE.J. Chromatogr. A 2004; 1023: 93.
10. Wang J, Cheung W, Grant D. J. Agric. Food Chem. 2005; 53:528.
11. Pozo OJ, Barreda M, Sancho JV, Hernandez F, Llibreria JL,Cortes MA, Bago B. Anal. Bioanal. Chem. 2007; on-line firstin press.
Copyright # 2007 John Wiley & Sons, Ltd.
12. Hiemstra M, de Kok A. J. Chromatogr. A 2007; 1154: 3.13. Klein J, Alder L. J. AOAC Int. 2003; 86: 1015.14. Nguyen DTT, Guillarme D, Rudaz S, Veuthey JL. J. Sep. Sci.
2006; 29: 1836.15. Swartz ME. LC-GC (June) 2005; 5.16. Ibanez M, Sancho JV, Pozo OJ, Hernandez F. Anal. Chem.
2004; 76: 4349.17. Lehotay SJ, de Kok A, Hiemstra M, van Bodegraven P.
J. AOAC Int. 2005; 88: 595.18. Lehotay SJ, Mastovska K, Lightfield AR. J. AOAC Int. 2005;
88: 615.19. Ferrer I, Garcıa-Reyes JF, Mezcua M, Thurman EM, Fernan-
dez-Alba AR. J. Chromatogr. A 2005; 1082: 81.20. Ferrer I, Fernandez-Alba A, Zweigenbaum JA, Thurman
EM. Rapid Commun. Mass Spectrom. 2006; 20: 3659.21. Andre F, De Wasch KKG, De Brabander HF, Impens SR,
Stolker LAM, van Ginkel L, Stephany RW, Schilt R,Courtheyn D, Bonnaire Y, Furst P, Gowik P, Kennedy G,Kuhn T, Moretain JP, Sauer M. Trends Anal. Chem. 2001; 20:435.
22. Commission Directive 2002/657/EC of 12 August 2002implementing Council Directive 96/23/EC concerning theperformance of analytical methods and the interpretation ofresults, Official Journal L221/8.
23. Ferrer I, Thurman EM, Yanyan F, Zavitsanos P, Zweigen-baum JA. Agilent Application Note 5989-5469EN.
Rapid Commun. Mass Spectrom. 2007; 21: 3869–3882
DOI: 10.1002/rcm