Jurnal Talanta
Jurnal Talanta
Jurnal Talanta
Talanta
journal homepage: www.elsevier.com/locate/talanta
a r t i c l e i n f o a b s t r a c t
Article history: Pesticides residues in aquatic ecosystems are an environmental concern which requires efficient
Received 22 January 2011 analytical methods. In this study, we proposed a generic method for the quantification of 13
Received in revised form 18 May 2011 pesticides (azoxystrobin, clomazone, diflufenican, dimethachlor, carbendazim, iprodion, isoproturon,
Accepted 11 June 2011
mesosulfuron-methyl, metazachlor, napropamid, quizalofop and thifensulfuron-methyl) in three envi-
Available online 17 June 2011
ronmental matrices. Pesticides from water were extracted using a solid phase extraction system and
a single solid–liquid extraction method was optimized for sediment and fish muscle, followed by a
Keywords:
unique analysis by liquid chromatography coupled with tandem mass spectrometry (LC–MS/MS). Limits
Multiresidue analysis
Pesticides
of quantification were below 5 ng L−1 for water (except for fluroxypyr and iprodion) and ranged between
Fish 0.1 ng g−1 and 57.7 ng g−1 for sediments and regarding fish, were below 1 ng g−1 for 8 molecules and were
Sediments determined between 5 and 49 ng g−1 for the 5 other compounds. This method was finally used as a new
Water routine practice for environmental research.
LC–MS/MS © 2011 Elsevier B.V. All rights reserved.
0039-9140/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.talanta.2011.06.023
A. Lazartigues et al. / Talanta 85 (2011) 1500–1507 1501
Table 1
Use and properties of the 13 target pesticides.
Compounds Actiona Cropb Molar massc (g mol−1 ) Solubility in waterc (mg L−1 ) Vapor pressurec (Pa, 20 ◦ C) Log Pc
−10
Azoxystrobin F Wheat/barley 403.39 6.7 1.1 × 10 2.50
Carbendazim F Rape seed 191.19 8 9.0 × 10−5 1.56
Clomazone H Rape seed 239.70 1100 1.9 × 10−2 2.54
Diflufenican H Wheat/barley 394.29 0.05 3.1 × 10−5 4.90
Dimethachlor H Rape seed 255.74 2300 1.5 × 10−3 2.25
Fluroxypyr H Wheat/barley 255.03 6500 3.8 × 10−9 2.00
Iprodion F Rape seed 330.17 13 5.0 × 10−7 2.99
Isoproturon H Wheat/barley 206.28 70 3.3 × 10−6 2.50
Mesosulfuron-methyl H Wheat 503.51 483 1.1 × 10−11 −0.48
Metazachlor H Rape seed 277.75 450 4.7 × 10−5 2.13
Napropamid H Rape seed 271.35 70 5.3 × 10−4 3.36
Quizalofop H Rape seed 344.8 0.31 1.1 × 10−7 4.66
Thifensulfuron-methyl H Maize 387.39 2240 7.5 × 10−9 −1.70
a
Action is symbolized by H for herbicide and F for fungicide action.
b
Examples of crop are presented for illustrate pesticide use [27].
c
Data come from pesticide databases (data confirmed by 4 web databases: SIRIS, Footprint, Agritox and HSDB).
properties (Table 1), among the most commonly applied on thifensulfuron-methyl to 4.9 for diflufenican. Consequently, devel-
cereals or oilseeds crop [27]—i.e. azoxystrobin, clomazone, oping one analytical method for all of them was a real challenge.
diflufenican, dimethachlor, carbendazim, iprodion, isoproturon, The method was developed for 3 environmental matrices:
mesosulfuron-methyl, metazachlor, napropamid, quizalofop, and water, sediment and biota (muscle of 2 fish species). Extraction
thifensulfuron-methyl. The physicochemical characteristics of the consisted in a SPE for freshwater and a unique solid–liquid extrac-
selected pesticides are presented in Table 1. Among the 13 pes- tion for sediments and fish muscle, all followed by a LC–MS/MS
ticides, three of them present a fungicide action (azoxystrobin, analysis (Fig. 1). Finally, the method was successfully applied to
carbendazim, and iprodion) and the other ones an herbicide action. 10 samples of sediments and muscles of 50 carp and 43 perch. To
Their physicochemical properties vary significantly. For example, our best knowledge, this is one of the few publications presenting
the water solubility of fluroxypyr is 6.5 g L−1 at 20 ◦ C, whereas multi-residue methods with ng L−1 and ng g−1 reported limits for
the diflufenican is practically insoluble in water (0.05 mg L−1 ). the analysis of pesticides belonging to a wide variety of chemical
On the other hand, the log P of molecules vary from −1.7 for families in three different complex matrices (Fig. 2).
Fig. 1. Chromatogram of spiked fish with the 13 pesticides (300 ng g−1 ): 1 – carbendazim, 2 – fluroxypyr, 3 – thifensulfuron-methyl, 4 – mesosulfuron-methyl, 5 – isoproturon,
6 – metazachlor, 7 – dimethachlor, 8 – clomazone, 9 – quizalofop, 10 – azoxystrobin, 11 – napropamid, 12 – iprodion, and 13 – diflufenican.
1502 A. Lazartigues et al. / Talanta 85 (2011) 1500–1507
Fig. 2. Scheme of analytical strategy for pesticides from water, sediment and fish muscle samples.
2. Materials and methods period were considered per site). Sediments or surface water sam-
ples were dark conditioned in glass bottles and frozen at −20 ◦ C.
2.1. Chemicals and reagents Sediments samples were not dried to a better taking account of
environmental expectations evaluating pesticides also in accessible
Dimethyl-sulfoxyde (DMSO), dimethyl-formamide (DMF) and part of sediments (i.e. pore water).
acetone were of high purity (99%) and purchased from Riedel-de Fish species were carp (Cyprinus carpio) and perch (Perca fluvi-
haen (Seelze, Germany). HPLC-grade solvents (methanol, cyclo- atilis), collected during emptying periods of pond. Lipid contents
hexane, and ethyl acetate) used for sample preparation were in muscle were 0.5 ± 0.2% and 4 ± 3% for perch and carp, respec-
purchased from Sigma–Aldrich (Saint Quentin Fallavier, France), tively. A three months depuration in clean water was performed
Fisher (Strasbourg, France) and VWR (Fontenay-sous-Bois, France) for some of them in laboratory, for recovery studies and validation
respectively. For LC–MS/MS analysis, LC–MS-grade solvent, ace- (three months corresponded to three times the maximum half-life
tonitrile was obtained from Sigma–Aldrich like anhydrous MgSO4 . observed in fish muscle for these compounds). Others (without
Azoxystrobin, clomazone, diflufenican, dimethachlor, carben- depuration) were environmental samples. Fish were killed by a
dazim, iprodion, isoproturon, mesosulfuron-methyl, metazachlor, blow to the head, following anaesthesia in ice bath. Then, mus-
napropamid, thifensulfuron-methyl (Sigma–Aldrich), quizalofop cles were quickly removed and frozen (−20 ◦ C). All fish treatments
(Supelco) and carbendazim-d4 , isoproturon-d3 (Cluzeau) were used to obtain fish collection were in accordance with the gen-
powder conditioned with a purity upper than 97%. Ultra pure (UP) eral guidelines of the Council of European Communities (1986, No.
water was produced from a Direct-Q-system (Millipore, Molsheim, 86/609/CEE) [28] and the French Animal Care Guidelines.
France). Stock solutions were prepared in brown glass bottles and
frozen at −20 ◦ C. Concentration of each compound was 1.0 mg mL−1 2.3. Extraction method
(in acetonitrile) except for carbendazim (0.1 mg mL−1 , in methanol,
because of limited solubility). 2.3.1. Water
Water samples were thawed out at +4 ◦ C. Solid-phase extraction
2.2. Sample collection (SPE) was performed with Strata X column (200 mg, 33 m, 3 mL;
Phenomenex, Torrance, California) conditioned successively with
Water, sediment and fish were collected in five dam ponds acetonitrile (3 mL), methanol (3 mL) and ultrapure water (3 mL).
in the East of France (Lorraine region), between 2007 and 2009. 400 mL of water sample was used. Column was rinsed with water
These ponds, established on clay soils, dammed a stream and were (5 mL) and dried during 20 min. Pesticides were extracted with
strongly connected to their agricultural watersheds. Five sites had acetonitrile (3 mL) and then ethyl acetate (3 mL). 50 L of DMSO
0, 25, 45, 75 and 85% of cultivated parcels on the watershed area, and 50 L of DMF were added (Fig. 2). After evaporation (nitrogen
respectively. flux, 40 ◦ C), residue was dissolved in 2900 L of acetonitrile/water
Water samples were collected (10–15 cm depth) during spray- (10:90, v/v) and dark stored (−20 ◦ C).
ing periods (autumn and spring) at the level of the dam. Water
parameters were measured at each sample time: temperature was 2.3.2. Sediments and fish
ranged from 8 to 21 ◦ C, dissolved oxygen was ranged from 5 to Sediments and fish samples were thawed out at +4 ◦ C and
15 mg L−1 , pH was ranged from 7.5 to 9.5. Sediments (0–4 cm depth, grinded. 4 g of sediments or 3 g of muscle were sampled in 50 mL
pH was 8 ± 1) were sampled during the emptying period of pond in tubes (734-0492, VWR). A single extraction method was applied for
five sedimentation areas of each pond. They were homogenized the two matrices. A high-performance disperser, Ultra-turrax (IKA,
to have a single sample per site and per period (two emptying Lille, France), was used to grind the sample in tubes maintained
A. Lazartigues et al. / Talanta 85 (2011) 1500–1507 1503
Table 2
LC–MS/MS conditions: retention time, MRM transitions selected, declustering potential (DP), collision energy (CE), applied in positive ESI mode.
in water at 20 ◦ C with 10 mL of acetonitrile/water (50:50, v/v) ing potential (DP) and collision energy (CE) of the main transitions
and 4 g of MgSO4 anhydrous. Surrogate standards, carbendazim- were optimized from a continuous flow of a standard injection
d4 and isoproturon-d3 , were added to check extraction recovery. (1 mg L−1 solution at 10 L min−1 ). Two multiple reaction monitor-
The tube was shaken 10 min and then centrifuged (12,000 × g, ing (MRM) transitions were acquired for each analyte, using the Q/q
10 min, 25 ◦ C). The upper phase was put in a glass tube with 100 L intensity ratio as confirmatory parameter. LC–MS/MS conditions
DMSO. A second solid–liquid extraction was performed with ethyl are presented in Table 2.
acetate/cyclohexane (75:25, v/v), 10 min agitation and centrifuga-
tion (12,000 × g, 10 min, 25 ◦ C). The upper phase was remote in 2.5. Quantification and validation
a glass tube with 100 L DMF. After evaporation (nitrogen flux,
40 ◦ C) of each tube, residue was dissolved in 900 L of acetoni- 2.5.1. Instrument performance
trile/water (10:90, v/v). Residues of the two tubes (first and second Intra-day and inter-day precision were determined by five
centrifugation) were mixed in one and dark stored at −20 ◦ C (Fig. 2). repeated injections of standards at 5 g L−1 in the same day and dif-
ferent days [24]. Eight calibration standards from 0.01 to 100 g L−1
2.4. LC–MS/MS analysis were injected to evaluate the linearity of the instrumental method
and the instrumental detection limit (ILOD). The latter corresponds
Analyses were performed on Agilent 1100 series HPLC (Agilent to the analyte concentration that produced a chromatographic peak
Technologies, Heilbronn, Germany) equipped with a reversed- signal of three times the background noise (Table 3).
phase Zorbax XDB-C18 (Agilent, 50 mm × 2.1 mm, 1.8 m). Mobile
phases were 0.0125% (v/v, pH = 4.04) acetic acid (A) and 100% ace- 2.5.2. Method performance
tonitrile (B). Pesticides were separated following gradient program: To determine recoveries, freshwater, sediment and fish sam-
gradient was from 15% B to 60% B in 1 min and secondly, from 60% ples were spiked before and after extraction in triplicate. Recoveries
B to 100% B in 6 min. Solvents were maintained to 100% B for 5 min were evaluated at 3 concentrations, 1, 2.5 and 5 ng L−1 for water and
and then, returned to initial conditions with an equilibration of 1.5, 15 and 30 ng g−1 for solid and biotic matrices except diflufeni-
7 min. Column temperature was 60 ◦ C, flow rate was 0.3 mL min−1 . can, iprodion, quizalofop and fluroxypyr (2.5, 5 and 50 ng L−1 and
The sample injection volume was set at 10 L. 15, 30 and 300 ng g−1 ). Recoveries were calculated by the ratio of
The mass spectrometer used was a 3200 Q-Trap system (Applied the areas of each analyte in the sample spiked before extraction
Biosystems, Foster City, CA, USA) equipped with an ESI (TurboV) and in the sample spiked after extraction. Repeatability is associ-
source operated in positive ionization mode. The operating para- ated with the intra-day precision. To estimate it, the samples need
meters were: ionspray voltage 5500 V, curtain gas 20 (arbitrary to be spiked, extracted and analyzed in the same conditions, by the
units), nebulizer gas 50 (arbitrary units), auxiliary gas 60 (arbitrary same manipulator and on the same day. Therefore, each level of
units), probe temperature 600 ◦ C. For each compound, decluster- concentration was repeated three times. The intra-day precision
1504 A. Lazartigues et al. / Talanta 85 (2011) 1500–1507
RSD (%)
measurements [29].
0.5
Limits of quantification (LOQ) were determined as the analyte
4
6
2
21
6
5
22
4
3
4
3
5
concentration that produced a peak signal of three and 10 times
the background noise from the chromatogram respectively.
recovery (%)
Fish muscle
Extraction
Quantification, based on peak areas, was performed by exter-
nal calibration concerning water samples and matrix-matched
calibration regarding solid matrices (sediment and fish muscle).
71
82
85
36
80
115
52
90
89
82
67
68
93
Eight-point calibration curves (0.01–100 g L−1 ) were generated
using linear regression analysis. The linearity was qualified by lin-
ear correlation coefficient, r2 .
(ng g−1 )
Extraction parameters (recovery rate and coefficient of variation CV, n = 3) and limits of detection for instrument (ILOD) and limit of 24 quantification of the method (LOQ) in each matrix.
0.2
29.9
0.3
0.4
0.1
0.2
17.9
49.0
24.0
5.0
LOQ
C18 (Agilent, 50 mm × 2.1 mm, 1.8 m) for its good retention of
hydrophobic and hydrophilic compounds, even with a short size.
102
68
95
35
80
40
125
61
83
118
115
50
61
74.6
0.4
0.2
0.3
0.2
4.2
0.3
41.0
3.2.1. Water
0.1
0.1
0.2
1.8
0.1
15.4
82.6
0.2
0.1
0.1
0.1
1.6
0.1
obtained with spiked water did not show any significant break-
through before 500 mL. Nevertheless the volume was finally set at
Repet (RSD %)
Napropamid
Metazachlor
Diflufenican
Compounds
Isoproturon
Clomazone
Fluroxypyr
Quizalofop
high speed caused a too much increase of heat that could induce except iprodion. The percentage varies from −94 to 198% for
degradation of pesticides. Middle speed (3/5) was a better compro- fish samples. Thus, to compensate matrix effect and avoid any
mise if sample tube was maintained at 20 ◦ C in a water bath. Water under/over estimation of pesticides, a matrix-matched calibration
in solid matrices was trapped by anhydrous MgSO4 . NaSO4 was also in eight points was used, for sediment and fish matrices.
tested but produced a strong aggregation with sediments and did
not enable centrifugation. 3.4. Method performance and validation
Choice of solvents was based on miscibility of pesticides in order
to increase recoveries. Ethyl acetate, cyclohexane, acetone, ace- Instrument performance evaluation is detailed in Section 2.5.1.
tonitrile were tested in two successive centrifugations in different Limits of detection for instrument (ILOD) thus determined are
combinations (Table 4). Better recoveries were generally observed reported in Table 3: they are all inferior to 2 pg injected except
with ethyl acetate/acetonitrile (50:50, v/v) conditions, except for fluroxypyr (15.42 pg injected) and iprodion (82.64 pg injected).
for diflufenican, clomazone, and quizalofop which had very low Reproducibility and repeatability expressed as relative standard
recoveries. These three compounds had a better recovery with ethyl deviation, RSD, were lower than 18% for intra-day and 27% for
acetate/cyclohexane (75:25, v/v). For this reason two successive inter-day analysis, respectively.
solvent extractions were chosen to final conditions: firstly with The performance of the method was evaluated through estima-
ethyl acetate/acetonitrile (50:50, v/v) conditions and then, ethyl tion of recovery, intra-day precision, and sensitivity of the method.
acetate/cyclohexane (75:25, v/v). Recoveries were increased for all Regarding water samples, optimized extraction using Strata X
compounds. cartridges gave high recovery values, presented in Table 3. The
Centrifugation parameters were also tested to get a good separa-
tion solid/liquid for sediments and muscle. Temperature and speed
of rotation were tested in order to evaluate a good phase separation Table 4
(visually) and recuperation of solvents (measured volume). Three Comparison between recovery rates obtained with different solvents of extraction
in the case of spiked muscle (4 g) of Cyprinus carpio (300 ng g−1 for each compound
temperature levels (5, 20 and 25 ◦ C) and three rotation rates (4000;
except for carbendazim and carbendazim-d4 , 30 ng g−1 ).
10,000 and 12,000 × g) were tested. Below 25 ◦ C condition, recu-
peration of solvents was not enabled because of insufficient phase Compounds Extraction recovery (%) (n = 6)
100
>LOQ
<LOQ
80
Contaminated samples (%)
60
40
20
0
Quizalofop
Quizalofop
Quizalofop
Iprodion
Iprodion
Iprodion
Azoxystrobin
Fluroxypyr
Azoxystrobin
Fluroxypyr
Azoxystrobin
Fluroxypyr
Metazachlor
Metazachlor
Metazachlor
Mesosulfuron-methyl
Mesosulfuron-methyl
Mesosulfuron-methyl
Carbendazim
Carbendazim
Clomazone
Isoproturon
Carbendazim
Clomazone
Isoproturon
Clomazone
Isoproturon
Thifensulfuron-methyl
Thifensulfuron-methyl
Thifensulfuron-methyl
Diflufenican
Diflufenican
Diflufenican
Dimethachlor
Napropamid
Dimethachlor
Napropamid
Dimethachlor
Napropamid
sediments (n = 10) carp (n = 50) perch (n = 43)
Fig. 4. Frequency of contaminated samples (%) for sediments, carp and perch studied.
obtained values range between 42% (fluroxypyr) and 125% (ipro- diflufenican, dimethachlor, carbendazim, iprodion, isoproturon,
dion) with satisfactory intra-day precision (RSD lower than 23%). mesosulfuron-methyl, metazachlor, napropamid, quizalofop and
Extraction recoveries obtained on 3 g fish muscle and 4 g sedi- thifensulfuron-methyl) from water, sediments and fish muscle. The
ments, with the fully optimized method, are presented in Table 3. developed extraction step is an easy and relatively quick method
The obtained recoveries are satisfactory for such complex matrices using few solvent quantities. Two extraction methods were devel-
between 40 and 125% for sediment and 52 and 115% for fish muscle. oped: a solid-phase extraction for water and a unique solid–liquid
Only the diflufenican, the more hydrophobic compound in the list, extraction method for solid matrices, all followed by a single quan-
presented a recovery of 35% in sediment and 36% in fish muscle. tification method by LC–MS/MS using a matrix matched calibration.
The recoveries varied significantly depending on compound and LOQ were below 75 ng L−1 , 58 ng g−1 and 49 ng g−1 for water, sedi-
matrix. But, considering fish, RSD were less than 6.2% for all the ment and fish muscle respectively. But, for most compounds, LOQ
compounds in both matrices (except for diflufenican (20.8%) and were below 1 ng g−1 and recoveries were higher than 80%.
iprodion (22.1%)), which is considered as good method precision. To our knowledge, it is the first time that one method is
The limits of quantification in fish and sediment were lower performed to extract these 13 pesticides from freshwater and
than 5 ng L−1 and 1 ng g−1 for most of the compounds. The method several types of sediments and fish muscles. It allows a routine
provides lower sensitivity for diflufenican, fluroxypyr, iprodion, practice for environmental research, i.e. monitoring and under-
quizalofop and thifensulfuron-methyl (5.03 < LOQ < 49.02). standing pesticide behavior. This kind of development is of great
importance for two reasons. First, it allows a global view of aquatic
3.5. Method application ecosystems permitting a better understanding of contamination
of fishes. Secondly, reliable analytical methods are needed for
Method was applied to 10 samples of sediments and muscles of bioaccumulation studies which we have planned to undertake in
50 carps and 43 perches. Each sample was extracted and analyzed another work.
in duplicate. These environmental matrices were few contami-
nated; molecules were detected at concentration inferior to LOQ
Acknowledgments
in most samples (Fig. 4). Isoproturon and carbendazim were the
most frequently detected or quantified in sediments or fish mus-
This work was funded by the Agence de l’Eau Rhin-Meuse and the
cle. Maxima concentrations concerned isoproturon with 9.5, 0.50
Zone Atelier Moselle which are gratefully acknowledged.
and 0.85 ng g−1 for sediments, carp and perch muscles, respectively.
Maximum value in sediments is higher than concentration mea-
sured in Garonne river [3]. Carbendazim was measured respectively References
at 0.92, 0.34 and 0.21 ng g−1 in sediments, carp and perch muscles.
The other molecules were probably not persistent in these environ- [1] T. Reemtsma, M. Jekel, Organic Pollutants in the Water Cycle, Wiley-VCH Verlag
GmBH & Co. KGaA, Weinheim, 2006.
mental matrices. Comparison with literature is difficult because, to [2] R. Loos, B.M. Gawlik, G. Locoro, E. Rimaviciute, S. Contini, G. Bidoglio, Environ.
our knowledge, there is a lack of data about levels of these com- Pollut. 157 (2009) 561–568.
pounds in environmental matrices. [3] D.A. Devault, M. Gérino, C. Laplanche, F. Julien, P. Winterton, G. Merlina, F.
Delmas, P. Lim, J. Miguel Sánchez-Pérez, E. Pinelli, Sci. Total Environ. 407 (2009)
2659–2665.
4. Conclusion [4] S. Bony, I. Gaillard, A. Devaux, Int. J. Environ. Anal. Chem. 90 (2010) 418–421.
[5] J. Marchand, A. Tanguy, G. Charrier, L. Quiniou, E. Plee-Gauthier, J. Laroche, Mar.
Biotechnol. 8 (2006) 275–294.
This work proposed a generic multiresidue analysis to [6] P.J. John, A. Prakash, Bull. Environ. Contam. Toxicol. 70 (2003) 1013–1016.
extract and quantify 13 pesticides (azoxystrobin, clomazone, [7] F. Sun, S.S. Wong, G.C. Li, S.N. Chen, Chemosphere 62 (2006) 674–680.
A. Lazartigues et al. / Talanta 85 (2011) 1500–1507 1507
[8] R.P. Schwarzenbach, T. Egli, T.B. Hofstetter, U. von Gunten, B. Wehrli, Annu. Rev. [20] F. Sun, S.S. Wong, G.C. Li, S.N. Chen, J. Food Drug Anal. 13 (2005)
Environ. Resour. 35 (2010) 109–136. 151–158.
[9] J.L. Martinez Vidal, A. Belmonte Vega, F.J. Arebola, M.J. Gonzales Rodriguez, M.C. [21] D.F.K. Rawn, J. Judge, V. Roscoe, Anal. Bioanal. Chem. 397 (2010) 2525–
Morales Sanchez, A. Garrido Frenich, Rapid Commun. Mass Spectrom. 17 (2003) 2531.
2099–2106. [22] M. Anastassiades, S.J. Lehotay, D. Stajnbaher, F.J. Schenck, J. AOAC Int. 86 (2003)
[10] B. Erkmen, D. Kolankaya, Int. J. Environ. Anal. Chem. 86 (2006) 161–169. 412–431.
[11] V. Nardelli, D. DellOro, C. Palermo, D. Centonze, J. Chromatogr. A 1217 (2010) [23] A. Mekebri, D.B. Crane, G.J. Blondina, D.R. Oros, J.L. Rocca, Bull. Environ. Contam.
4996–5003. Toxicol. 80 (2008) 455–460.
[12] S.H.G. Brondi, A.N. de Macedo, G.H.L. Vicente, A.R.A. Nogueira, Bull. Environ. [24] A. Jelic, M. Petrovic, D. Barcelo, Talanta 80 (2009) 363–371.
Contam. Toxicol. 86 (2011) 18–22. [25] H. Berrada, F. Borrull, G. Font, R.M. Marcé, J. Chromatogr. A 1208 (2008) 83–89.
[13] L. Kalyoncu, I. Agca, A. Aktumsek, Chemosphere 74 (2009) 885–889. [26] C. Sparr-Eskilsson, E. Björklund, J. Chromatrogr. A 902 (2000) 227–250.
[14] V. Nardelli, D. dell’Oro, C. Palermo, D. Centonze, J. Chromatogr. A 1217 (2010) [27] A. Couteux, Index Phytosanitaire, ACTA Editions, Lille, 2008.
4996–5003. [28] Council of European Communities (1986), Council Directive 86/609/EEC
[15] M. Petrovic, M.D. Hernando, M.S. Diaz-Cruz, D. Barcelo, J. Chromatogr. A 1067 of 24 November 1986 on the approximation of laws, regulations and
(2005) 1–14. administrative provisions of the Member States regarding the protec-
[16] M. Gros, M. Petrovic, D. Barcelo, Anal. Bioanal. Chem. 386 (2006) 941–952. tion of animals used for experimental and other scientific purposes,
[17] A. Ranz, E. Maier, H. Motter, E. Lankmayr, J. Sep. Sci. 31 (2008) 3021–3029. http://europa.eu/legislation summaries/environment/nature and biodiversity/
[18] M. Petrovic, M. Farré, M.L. De Alda, S. Perez, C. Postigo, M. Köck, J. Radjenovic, l28104 en.htm (accessed 30.12.2010).
M. Gros, D. Barcelo, J. Chromatogr. A 1217 (2010) 4004–4017. [29] C. Mansilha, A. Melo, H. Rebelo, I. Ferreira, O. Pinho, V. Domingues, C. Pinho, P.
[19] M. Kuster, M.J. Lopez de Alda, M.D. Hernando, M. Petrovic, J. Martin-Alonso, D. Gameiro, J. Chromatogr. A 1217 (2010) 6681–6691.
Barcelo, J. Hydrol. 358 (2008) 112–123.