Environmental Pollution 139 (2006) 133e142 www.elsevier.com/locate/envpol
Development of an SPMEeGCeMS/MS method for the determination of pesticides in rainwater: Laboratory and field experiments Nathalie Sauret-Szczepanski a,1, Philippe Mirabel a, Henri Wortham b,* a
Equipe de Physico-chimie de l’Atmosphere, Centre de Ge´ochimie de la Surface (UMR 7517) 1, rue Blessig, 67084 Strasbourg Cedex, France b Universite´ de Provence, Laboratoire Chimie et Environnement, Case 29, 3 Place Victor Hugo, 13331 Marseille Cedex, France Received 31 July 2003; accepted 23 April 2005
Solid-phase microextraction efficiency of pesticides in rainwater was optimized. Abstract A solid-phase microextraction e coupled to a gas chromatography e ion trap tandem mass spectrometry (SPMEeGCeMS/MS) method was developed for the quantitative determination in rainwater of 8 pesticides amongst the most used in France and 3 triazines metabolites. The main factors affecting the SPME process were studied. Using a 3 mL sample, the method developed showed good linearity for concentrations ranging from 0.05 to 50 mg L1 with correlation coefficients between 0.997 and 0.9999 and relative standard deviations (% RSD) below 14%. The study of matrix effects showed that rainwater was too diluted to have any significant influence on the extraction efficiency. To validate the method, a field campaign was carried out on the rain events, which occurred in Strasbourg during a one-year period. The rain concentrations showed patterns of high pesticide concentrations during spring months, which were correlated to the spraying periods of most of these substances. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Pesticides; Rainwater; Solid-phase microextraction; Mass spectrometry
1. Introduction The extraction of semi-volatile compounds from aqueous samples is traditionally carried out with liquideliquid extraction (LLE). This technique is time consuming and uses a large volume of solvent. Moreover, a pre-concentration step is usually required before analysis. For the past 20 years, solid phase
* Corresponding author. Tel.: C33 4 91 10 62 44; fax: C33 4 91 10 63 77. E-mail addresses:
[email protected] (N. SauretSzczepanski),
[email protected] (H. Wortham). 1 Present address: Alcan International Limited, Arvida Research and Development Centre, Jonquie`re, Que´bec, G7S 4K8, Canada. 0269-7491/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2005.04.024
extraction (SPE) has replaced LLE in most sample preparations. The main drawbacks of using disposable SPE cartridges make it quite expensive and require several cleaning, conditioning and drying steps are required to obtain good blanks. In spite of these difficulties, SPE seems to be the pre-concentration method most often used for environmental samples (Aguilar et al., 1997; Sabik et al., 2000). The supercritical fluid extraction (SFE) has received increasing attention because it is fast and solvent-free (Camel, 1997; De Martinis and Lancas, 2000). Nevertheless, it has several disadvantages. Indeed, the most used fluid is CO2 but its non-polar characteristic excludes an efficient extraction of polar compounds such as oxygenated compounds. It is possible to combine the fluid with an organic solvent
134
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142
but the extraction selectivity decreases and a cleaning step is then required. Recently, a new technique, the solid-phase microextraction (SPME), has been devised by Zhang et al. (1994). It allows simultaneous extraction and preconcentration of analytes from sample matrix. This technique is based on extraction using a fused-silica fiber coated with polymeric liquid phase such as polyacrylate, usually housed in a modified syringe, which allows easy manipulation from sample extraction to final desorption in the inlet port of a gas chromatograph. The advantages of this method are numerous: - simple and fast to use - solvent free - single step sample preparation. Furthermore, because the SPME approach only requires a low sample volume, it is perfectly adapted to the extraction of compounds from rainwater because precipitation volumes are often insufficient to allow for LLE or SPE. SPME has an increasing interest for semi-volatile compounds analysis as shown in literature especially for organochlorine, organophosphorus, triazine and PAH compounds (Eisert et al., 1994; Graham et al., 1996; Zambonin and Palmisano, 2000; Lassagne and Baussand, 2000; Guehenneux et al., 2004). All kinds of aqueous matrices have been studied: ultrapure water (Barnabas et al., 1995; Sng et al., 1997), ground water and surface water (Boyd-Boland et al., 1996; Choudhury et al., 1996; Graham et al., 1996), river water (Dugay et al., 1998), waste water (Valor et al., 1997) and drinking water (Aguilar et al., 1998). But to our knowledge, no studies have been made on rainwater. In this paper, an in-depth study of SPME optimization and application has been made. Optimization of the variables involved in the SPME procedure has been carried out by a well-structured step-by-step approach using spiked water samples. Therefore, we offer an original multi-residual method for the determination of 8 selected pesticides widely used nowadays and 3 triazine metabolites in rainwater. This method combines SPME with an advanced analytical technique: gas chromatography coupled to an ion trap tandem mass spectrometry which provides a rapid, sensitive and accurate way to analyze pesticides (Sauret et al., 2000; Charreˆteur et al., 1996).
2. Material and method 2.1. Materials The 8 pesticides used in this study were supplied by several company members of the U.I.P.P. (Union des
Industries de la Protection des Plantes) and were 98e 99% pure. Atrazine, terbuthylazine and metolachlor were supplied by Syngenta (France), alachlor by Monsanto (France), diflufenicanil, isoproturon, iprodione, fenoxaprop-p-ethyl by Aventis CropScience (France) and the 3 triazines metabolites (de-ethylatrazine (DEA), de-isopropylatrazine (DIA) and deethylterbuthylazine (DET)) by Promochem. Stock solutions were prepared in ethyl acetate and stored at 4 C. These standards were used to prepare diluted standard aqueous solutions and spiked water samples. In the pesticide concentrations under study, the amount of acetyl acetate coming from the stock solution is less than 1&. It has been shown that above this level, its presence does not affect the extraction (Eisert and Levsen, 1995). De-ionized water was obtained from a Milli-Q water system (Millipore, France). Sodium chloride of quality O99.5% was purchased from Prolabo (France). The SPME apparatus consists of a manual reusable syringe assembly supplied by Supelco (France). The microextraction fibers used were coated with polydimethylsiloxane (PDMS) of 100 and 7 mm thickness, polyacrylate (PA) of 85 mm thickness and 65 mm thickness of carbowax/divinylbenzene (CW/DVB). Supelco supplied them all. The fibers were conditioned before their first use according to the manufacturer’s specifications. A magnetic stirring and heater unit for stirring samples during the SPME process was used. 2.2. Apparatus A Varian Star 3400 CX gas chromatograph equipped with a splitesplitless injector and coupled to a Saturn 4D Varian ion trap mass spectrometer was used. The two devices were connected to an acquisition system Saturn IV and the ion trap mass spectrometer was used in MSeMS mode. The gas chromatograph was fitted with a 30 m ! 0.32 mm i.d. J and W Scientific fusedsilica DB-5-MS capillary column with a film thickness of 0.25 mm. The carrier gas was helium at a inlet pressure of 12 psi corresponding to a flow rate of about 1 mL min1. The temperature program used to separate all the compounds under study was as follows: initial oven temperature was held at 60 C for 2 min, programmed with a gradient of 15 C min1 up to 145 C and maintained at 145 C for 3 min, then programmed with a gradient of 1 C min1 up to 151 C and finally raised at a rate of 11 C min1 up to 250 C and maintained at this temperature for 10 min. The injector, the transfer line and the manifold temperatures were kept at 290, 280 and 200 C, respectively. Linearity and sensitivity of the ion trap detector were optimized by adjusting various parameters such as filament emission current, voltage of the electron multiplier and the total number of ions in the trap
135
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142
(target). These adjustments and the improvement induced by the MSeMS mode compared to the single-MS mode were discussed in detail in a previous paper (Sauret et al., 2000). Briefly, the parameters which apply to the MSeMS mode are presented below. Based on the reference values obtained in single-MS, several parameters have to be adjusted in MSeMS: - The voltage of the electron multiplier was increased by 400 V, - The target value was decreased to 5000 ions, - In electronic impact mode (EI), the pre-scan ionization time had to be increased to 1500 ms and the filament current had to be raised to 80 mA. On the other hand, the ionization energy was not adjustable and was fixed at 70 eV. All these adjustments in MS/MS were recommended in the constructor’s manual but the experiments showed that a fluctuation of 10% of these values did not affect the analytical results. Finally, some specific MSeMS parameters had to be determined: the isolation window (3 m/z), the excitation time (5 ms) and the collision time (20 ms). 2.3. Detection optimization Before developing the MS/MS program, we had to study the fragmentation of the compounds under study in single-MS in order to choose their parent ions. For all the compounds, we preferred the electronic impact mode (EI) rather than the chemical ionization mode (CI) because we noticed that CI produced a higher background level than EI which induced, an increase of the detection limit. Once the parent ions had been isolated in the trap, they were accelerated to collide with helium molecules used as carrier gas in GC (collisioninduced dissociation, CID). This dissociation was optimized in order to give a sufficient number of daughter ions for identification and quantification. The MS/MS procedure was described in detail in a previous study (Sauret et al., 2000), and the optimum MS/MS parameters used for the pesticides under study are summarized in Table 1. 2.4. SPME procedures To develop a multi-residue procedure for the determination of pesticides with different physico-chemical properties, the choice of the fiber was essential in order to achieve the highest possible extraction efficiency. In a first approach, three of more popular fibers were studied: 85 mm PA, 7 mm and 100 mm PDMS. Each fiber was plunged for 1 h in 3 mL of an aqueous standard solution containing the 8 pesticides and 3 metabolites under study (1 mg L1 each). The samples were then homogenized by magnetic stirring and maintained at
Table 1 MS/MS parameters Compound
Parent ion (m/z)
RF storage level (m/z)
CID excitation voltage (V)
DIA DEA DET Atrazine Terbuthylazine Alachlor Isoproturon Metolachlor Diflufenicanil Iprodione Fenoxapropp-ethyl
173 172 186 200 214 188 191 162 266 314 361
57 61 61 66 71 62 63 53 88 104 120
36 49 49 55 56 45 40 45 81 69 75
Daughter ions used for quantification (m/z) 145, 79, 104, 94, 104, 104, 132, 117, 120, 218, 238, 245, 244, 261,
158 104 145 174 173 160 146 134 246 271 288
a temperature of 25 C. The fibers were then inserted into the injection port of the GC for 5 min, at a temperature allowing an optimal extraction of the pesticides adsorbed on each fiber. Using this method, the performances of the various fibers were compared under their optimal conditions. After choosing the fiber, the adsorption of the compounds on the fiber was optimized by studying the immersion time of the fiber in the sample. This operation determined the time needed for the pesticides under study to reach the equilibrium between the aqueous and polymeric phases. For this investigation, we followed the evolution of the chromatographic responses versus the adsorption time. During the tests, the experimental conditions of adsorption, injection and analysis where those previously described, only the adsorption time was changed from one experiment to another. The temperature of the sample was also an important parameter for the SPME method because it controls the diffusion kinetics of the molecules towards the fiber and the value of the equilibrium constant between the aqueous phase and the polyacrylate fiber. Therefore, it had to be chosen so as to improve the adsorption on the solid phase, and accurately controlled to minimize the uncertainties of the analysis. Three temperatures were tested in triplicate by using the protocols of extraction and analysis previously described and an extraction time of 40 min. Finally, the effect of the ionic strength on the extraction efficiency was studied by adding different quantities of NaCl (0% up to saturation) in aqueous standard solutions of pesticides. 2.5. Rain samples From August 2000 to August 2001, rain samples were collected on a weekly basis in the non-cultivated area of
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142 5e+5
(a)
4e+5 3e+5 2e+5 1e+5 0
(b)
5e+6 4e+6
area
the botanical garden of Strasbourg (400,000 inhabitants) near the historic center. Strasbourg is a polluted city, which has in its vicinity crops such as maize, corn and vine. An automatic wet-only rainwater collector was used to collect the samples and an open collector was used to allow direct measurement of precipitation levels (Sanusi et al., 2000). After each sampling collection, the funnel and the collection bottle were cleaned and rinsed thoroughly with Milli-Q water. Samples were stored in the dark in pre-cleaned glass bottles at 4 C prior to analysis.
area
136
3e+6 2e+6
3. Results and discussion
1e+6
3.1. SPME optimization
0
(c)
3.1.2. Adsorption time profile The experimental results are presented in Fig. 1. They correspond to the mean values of three replicate tests while the error bars correspond to the relative standard deviation. According to the results, the equilibrium water/polyacrylate was reached more or less quickly depending on the molecules under study. Compounds such as triazines reached the equilibrium after 1 h. On the other hand, alachlor and metolachlor needed only 15 min to reach the equilibrium. In order to develop an extraction method to analyze the 11 compounds under study at one time, a period of 40 min was a satisfactory compromise. It avoided a too long extraction period and gave a good extraction efficiency for all the compounds. Nevertheless, because this duration is below the equilibrium time of the analytes, the extraction time had to be rigorously identical from one sample to another in order to obtain comparable results.
3e+7
area
3.1.1. Fiber type comparison The study of the chromatograms obtained from the three types of fiber shows that the 7 mm and 100 mm PDMS are efficient to extract the less polar compounds but do not efficiently extract the more polar pesticides. On the other hand, the 85 mm PA fiber seems to be more adapted to the target analytes because more polar compounds are more efficiently extracted while no significant decrease in the extraction efficiency of the less polar pesticides is observed. To complete this comparative study, the 65-mm CW/DVB fiber was tested. The results were similar to those obtained with the 85-mm PA. But the low stability of this fiber in presence of salt (Hernandez et al., 2000) (only 15 extractions are possible with the same fiber instead of more than 100 with the other fibers) led to the selection of the 85 mm PA. The influence of various parameters such as adsorption time, sample temperature, ionic strength and pH are tested in order to determine the best conditions for using this fiber.
2e+7 1e+7 0 0
20
40
60
80
100
120
extraction time (min) Fig. 1. Effect of the extraction time on the equilibrium of the various pesticides between the aqueous and polyacrylate phases (n Z 7 in triplicate).
3.1.3. Sample temperature The results obtained with the three temperatures tested are reported in Fig. 2. The evolution of the chromatographic responses versus the sample temperature showed that, an increase of 25 C (to reach 50 C) improved the adsorption of the molecules on the fiber. Nevertheless, above 50 C the extraction efficiency decreased. These behaviors have already been described in previous studies (Valor et al., 1997) and were probably the result of a competition between the kinetics of adsorption and desorption from the fiber. Indeed, the principle of SPME is based on the equilibrium of the analytes between the solid and liquid phases. This equilibrium is ruled out by the kinetics of adsorption and desorption from the fiber. Because these kinetics are temperature dependent, the variations of the working temperature modify the solideliquid equilibrium and consequently the extraction efficiency. According to our results, it seems that, up to 50 C the increasing temperature is favorable to the kinetics of adsorption while above this value, the kinetic competition becomes favorable to the desorption mechanism. 3.1.4. Sample ionic strength and pH The evolution of the chromatographic response versus the percentage of NaCl in the sample is presented
137
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142 4e+3
(a)
(b)
1,4e+5
3e+3
1,2e+5
3e+3
1,0e+5
area
area
4e+3
2e+3
8,0e+4
2e+3
6,0e+4
1e+3
4,0e+4
5e+2
2,0e+4 0,0
0 20
30
40
50
60
70
80
90
20
30
(c)
6e+5
50
60
70
80
90
80
90
(d) 5e+6
5e+5
4e+6
4e+5
area
area
40
sample temperature [°C]
sample temperature [°C]
3e+5
3e+6 2e+6
2e+5
1e+6
1e+5 0
0 20
30
40
50
60
70
80
90
20
30
sample temperature [°C]
40
50
60
70
sample temperature [°C]
Fig. 2. Evolution of the chromatographic responses of the various pesticides under study versus the sample temperature (n Z 3 in triplicate).
in Fig. 3. Each result corresponds to the mean value of three replicate tests and the error bars correspond to the relative standard deviation. The results show that the addition of salt enhances the extraction of 9 of the 11 analytes under study. Only, the response for diflufenicanil and fenoxaprop-p-ethyl decreases slightly. These
1,2e+6
(a)
5e+4
1,0e+6
4e+4
8,0e+5
peak area
peak area
6e+4
results are in good harmony with the theoretical physico-chemistry of the aqueous phase. The increase of the ionic strength corresponding to the addition of salt, induces a decrease in the apparent polarity of the molecules of water and therefore, decreases the solubility of the polar molecules. This shifts the equilibrium
3e+4 2e+4 1e+4
(b)
6,0e+5 4,0e+5 2,0e+5
0
0,0 0
20
40
60
80
100
0
20
% of NaCl
60
80
100
% of NaCl
2,5e+6
6e+6
(c)
(d)
5e+6
peak area
2,0e+6
peak area
40
1,5e+6 1,0e+6 5,0e+5
4e+6 3e+6 2e+6 1e+6 0
0,0 0
20
40
60
% of NaCl
80
100
0
20
40
60
80
100
% of NaCl
Fig. 3. Evolution of the chromatographic responses of the pesticides under study versus the NaCl percentage in the sample (n Z 5 in triplicate).
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142
3.1.5. Desorption temperature Tests were carried out to optimize the desorption of the molecules from the fiber. This was done by varying the injector temperature between 250 C and 300 C. For all the triplicate tests, the fibre was kept in the injection chamber for 5 min. The effect of the injector temperature on the desorption efficiency is presented in Fig. 4. Except for diflufenicanil, for which a slight increase of the chromatographic response is observed, an increase of the injector temperature from 250 C to 270 C had no influence on the analysis. Above the threshold value, a slight decrease in the chromatographic response was observed for triazines and their metabolite. This could be explained by a thermal degradation beyond the temperature threshold. Nevertheless, no degradation product was observed. Before fixing an optimal temperature, it was necessary to verify that the desorption was complete in order to avoid a contamination of sample n by the sample n 1. To test this phenomenon and evaluate the percentage of nondesorbed pesticides, a blank injection of the fiber was carried out after each sample extraction. The results are summarized in Fig. 5. The molecules of DIA, DEA, DET, atrazine, isoproturon iprodione and fenoxapropp-ethyl give no signal in the blank indicating that these compounds are readily desorbed from the fiber during a 5 min injection period at 250 C. For the four other compounds under study (terbuthylazine, alachlor, metolachlor and diflufenicanil), a portion of them was non-desorbed but the percentage of non-desorbed molecules decreased when the injector temperature increased and dropped to a minimum when a temperature reached 290 C. As discussed above, slight decrease of the chromatographic response was obtained at this
peak area
8e+6
6e+6
4e+6
6e+7 5e+7
peak area
between the aqueous phase and the polyacrylate fibers in favour of the latter. On the other hand, for less polar molecules, such as diflufenicanil and fenoxaprop-pethyl, the decrease of the polarity of the solvent induces an increase of their solubility in the aqueous phase and consequently a decreased extraction efficiency. To obtain an efficient extraction for all the target compounds a percentage of 70% of NaCl in the sample seems to be a satisfactory compromise. Moreover, this large addition of NaCl makes it possible to carry out the extraction step of all the atmospheric water samples (rain, fog and snow) in identical conditions because the initial ionic strength of natural atmospheric samples is much lower. On the other hand, the pH of the rain samples could vary between 3 and 7 (Sanusi et al., 1996) and this could affect the extraction efficiency of the compounds under study. Thus, to standardize the extraction conditions, we adjusted the sample to a pH of 6, which is a value close to the pH value of pure water in equilibrium with the atmospheric CO2.
4e+7 3e+7 2e+7 1e+7 0
6e+5
peak area
138
4e+5
2e+5
0 240
250
260
270
280
290
300
310
Injector temperature (°C) Fig. 4. Evolution of the chromatographic responses of the pesticides under study versus the injector temperature (n Z 5 in triplicate).
temperature. Nevertheless, this condition was chosen because it minimized the fraction of non-desorbed pesticides. In brief, the optimal conditions for SPME applied to the 85 mm polyacrylate fiber to analyze all the compounds under study were the following: sample volume of 3 mL, adsorption time 40 min, temperature of adsorption 50 C, pH 6, NaCl concentration 70%, desorption time 5 min and temperature of desorption 290 C. 3.2. Calibration 3.2.1. Calibration curves, LODs and precision In order to quantify the pesticides in rain samples, we determined the range of concentrations for which a linear response of the detector was obtained (ranging
139
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142
Non desorbed pesticide (%)
Non desorbed pesticide (%)
0,8
0,6
0,4
0,2
0,0 240
250
260
270
280
290
300
310
Injector temperature (°C)
10 8 6 4 2 0 240
250
260
270
280
290
300
310
Injector temperature (°C)
Fig. 5. Evolution of the percentage of non-desorbed pesticides versus the temperature of the injector (n Z 5 in triplicate).
generally between 0.05 and 10 mg L1). Naphthalene d8 was chosen as the internal standard and was systematically added to the sample to reach a concentration of 4 mg L1. The standard solutions of the analytes were prepared by dilution of the stock solution in different amounts of Milli-Q water samples. Each standard was analyzed five times under the conditions described above in order to determine the relative standard deviation (% RSD) of the analytical protocol, which ranged between 8.5% and 14% depending on the pesticide under study. Moreover, a linear response (regression coefficient from 0.997 to 0.9999) was obtained for concentrations ranging between 0.05 and 50 mg L1. Finally, the detection limits corresponding to a signal three times higher than the background ranged between 0.01 and 0.05 mg L1 depending on the pesticide under study. These values are impressive considering the small volume of the samples (3 mL). This is a good example of the efficiency of this analytical protocol. 3.2.2. Matrix effects To complete the validation of this method, the potential matrix effect induced by the environmental sample had to be studied. To evaluate this matrix effect, it was possible to split a rain sample into several fractions and to spike them with increasing and known amounts of standard solution of the target pesticides. The analysis of these various fractions made it possible to build a calibration curve in a rainwater matrix. The initial concentrations of pesticides in the sample under study were obtained from the interception of the linear regression with the y-axis. The linearity of the regression curve and the comparison between the pesticide concentrations e measured by using the two analytical methods (direct analysis of an aliquot of rain sample without spiking and indirect analysis with standard addition) e gave information on the importance of the matrix effect. Indeed, as the amount of pesticide spiked increased, the matrix effect on the analytical results became negligible. This variability of the matrix effect
had an influence on the linearity of the regression curve and on the indirect analysis method. In the present work, in order to estimate the matrix effect, the rain sample chosen had to be large and strongly concentrated in organic compounds so as to carry out the test in the worst possible conditions. The rain sample collected during the week of October 1st to 8th, 2001 fulfilled these conditions and its large volume made it possible to split it into eight aliquots of 3 mL. In the first aliquot, the concentrations of the target compounds were determined using the SPME method described previously as well as the classical external standard method. The seven remaining aliquots were spiked with various amounts of standard solutions ranging from 0.05 to 10 mg L1 in order to determine concentrations of the target pesticide using the standard addition method. By this method, the initial concentrations were calculated from the interception of the linear regression with the y-axis. The results obtained for the various pesticides using the two analytical methods (direct determination and standard addition) are presented in Table 2. A comparative study showed that the concentrations determined with the direct calibration method were slightly higher than those calculated with the standard addition method. Nevertheless, the deviation remained in the relative standard deviation range (14%). We could therefore conclude that the rainwater matrix did not significantly affect the SPME extraction efficiency of the target compounds.
3.3. Application to rain samples The concentrations of the 11 compounds under study were measured in the rain events occurring in Strasbourg between August 3rd, 2000 and August 6th, 2001. The raw data obtained did not allow direct comparisons between samples because the precipitation levels induce a dilution more or less important of the compounds. To avoid this difficulty, the concentrations were normalized
140
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142
Table 2 Determination of the matrix effect of rain sample by comparison of the direct determination and the standard addition method (n Z 8) Concentration (mg L1)
Compound
DIA DEA DET Atrazine Terbuthylazine Alachlor Isoproturon Metolachlor Iprodione Fenoxaprop-p-ethyl Diflufenicanil a
Direct method
Standard addition method
NDa ND 0.100 0.076 0.241 0.176 ND 0.229 0.178 0.182 0.048
ND ND 0.089 0.068 0.279 0.165 ND 0.204 0.155 0.166 0.051
Deviation between the two methods (%) e e 11 10 13 6.7 e 10 13 8.5 6.5
ND, not detected.
in function of their volume according to the following equation: Cnormalized Z
herbicides on corn and soybeans. We observed the same seasonal pattern for triazines metabolites (DEA and DET) which appeared a few days after their parents. Nevertheless, it was shown that the reactions producing these degradation products could occur both in surface water and in the atmosphere and with the current data we have no way to estimate the relative importance of these two modes of formation. Moreover, because the same metabolites are produced in the two media, we cannot use the present results to demonstrate the efficiency of the atmospheric reactions on the degradation of atmospheric pesticides. Nevertheless, because the metabolite concentrations were in the same order of magnitude as the parent compounds, these molecules have to be considered in the determination of the pesticide contamination assessment. Also, high concentrations of diflufenicanil and isoproturon were observed during winter months according to their straying periods.
4. Conclusion
Csample !Hsample Hmean
ð1Þ
where Csample is the concentration measured in the rain sample (mg L1), Hsample is the precipitation level of the weekly sample (mm) and Hmean is the average weekly precipitation level (mm). The concentration characteristics of the target molecules in rain samples are presented in Table 3. The detection percentage calculation was based upon the 29 samples collected during the sampling campaign. Rain concentrations showed patterns of high concentrations during spring months (cf. Fig. 6). These maximal concentrations decreased rapidly down to a level of less than 0.1 mg L1 in the following months for triazines, chloroacetanilides and fenoxaprop-p-ethyl. Observed concentration peaks during the spring months corresponded to the spraying periods of these pre-emergent
The successful development of a SPMEeGCeMS/ MS method for the multi-residue analysis of commonly used pesticides in rainwater has been developed. This solvent-free method gave good accuracy, a wide range of linearity (above 3 orders of magnitude), and detection limits obtained were comparable or even better than those required by official methods established for pesticide analysis. Moreover, the study of matrix effects showed that they are negligible compared to the relative standard deviation of the method. Finally the SPME technique was used to carry out a field campaign in Strasbourg during which 10 of the 11 molecules under study were detected without ambiguity. These findings support the use of SPME, an extraction technique which, combined with GCeMS/ MS, is perfectly suitable for the analysis of rain samples.
Table 3 Concentration characteristics of the target molecules in rain samples from 08.03.00 to 08.06.01 Compound
Detection frequency (%)
Minimum (mg L1)
Maximum (mg L1)
Maximum date
Spraying period
DIA DEA DET Atrazine Terbuthylazine Alachlor Metolachlor Diflufenicanil Fenoxaprop-p-ethyl Iprodione Isoproturon Cymoxanil
0 50.0 54.2 75.0 70.8 100 100 100 79.2 66.7 8.33 100
e !LQ 0.038 !LQ !LQ !LQ !LQ !LQ !LQ !LQ 0.768 !LQ
e 0.498 0.525 0.794 1.06 3.169 1.443 2.823 0.300 0.124 1.273 14.78
e 28.03e12.04.01 28.03e12.04.01 16e23.03.01 28.03e12.04.01 15e21.05.01 28.03e12.04.01 31.01e21.02.01 24e30.04.01 28.03e12.04.01 20e31.10.01 12e17.04.01
e e e mid Aprileend of May MarcheApril end of Aprilemid May mid Aprileend of May NovembereFebruary MarcheApril AprileSeptember OctobereFebruary AprileAugust
LQ, quantification limit estimated at 10 times the signal/noise ratio.
141
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142
atrazine terbuthylazine
1,2 0,8 0,4 0,0
DEA DET
1,2 0,8
Concentration (µg L-1)
0,4 0,0 (3.25)
1,2
alachlor metolachlor
0,8 0,4 0,0 (2.89) diflufenicanil isoproturon
1,2 0,8 0,4 0,0
fenoxaprop-p-ethyl iprodione
1,2 0,8 0,4
03 /1 0 04 .08 /1 0 11 .09 /1 7 09 .09 /1 9. 20 10 /3 1 08 .10 /3 0 31 .01 /2 1 22 .02 /0 1 01 . /0 03 8. 09 03* /1 5 16 .03 /2 28 3. /1 03 2. 12 04* /1 17 7, /2 04 24 3,0 /3 4* 0, 31 04* /0 7 15 ,05 /2 1, 21 05 /3 0 30 ,05 /0 7 07 ,06 /1 1 11 ,06 /1 9 06 ,06 /0 9, 24 07 /3 31 1,0 /0 7 6, 08
0,0
Date Fig. 6. Concentrations of the target compounds (mg L1) in rain samples during the field campaign (n Z 24).
Acknowledgments We gratefully thank UIPP (Union des Industries pour la Protection des Plantes) and the ‘‘Agence De l’Environnement et de la Maitrise de l’Energie’’ (ADEME) for their financial and technical support.
References Aguilar, C., Borrull, F., Marce´, R.M., 1997. Determination of pesticides in environmental waters by solid-phase extraction and gas chromatography with electron-capture and mass spectrometry detection. J. Chromatogr. A 771, 221e231. Aguilar, C., Pen˜alver, S., Pocurull, E., Borrull, F., Marce´, R.M., 1998. Solid-phase microextraction and gas chromatography with mass
spectrometric detection for the determination of pesticides in aqueous samples. J. Chromatogr. A 795, 105e115. Barnabas, I.J., Dean, J.R., Fowlis, I.A., Owen, S.P., 1995. Automated determination of s-triazine herbicides using solid-phase microextraction. J. Chromatogr. A 705, 305e312. Boyd-Boland, A.A., Magdic, S., Pawliszyn, J., 1996. Simultaneous determination of 60 pesticides in water using solid-phase microextraction and gas chromatographyemass spectrometry. Analyst 121, 929e938. Camel, V., 1997. The determination of pesticide residues and metabolites using supercritical fluid extraction. Trends Anal. Chem. 16 (6), 351e369. Charreˆteur, C., Kerbaol, C., Pe´ron, J.J., 1996. Contribution to the analysis of triazines in water by gas chromatography and ion trap tandem mass spectrometry. Analusis 24, 336e346. Choudhury, T.K., Gerhardt, K.O., Mawhinney, T.P., 1996. Solidphase microextraction of nitrogen- and phosphorus-containing pesticides from water and gas chromatographic analysis. Environ. Sci. Technol. 30, 3259e3265.
142
N. Sauret-Szczepanski et al. / Environmental Pollution 139 (2006) 133e142
De Martinis, B.S., Lancas, F.M., 2000. An alternative supercritical fluid extraction system for aqueous matrices and its application in pesticides residue analysis. J. Environ. Sci. Health e Part B Pesticides, Food Contaminants and Agricultural Wastes 35 (5), 539e547. Dugay, J., Mie`ge, C., Hennion, M.-C., 1998. Effect of the various parameters governing solid-phase microextraction for the tracedetermination of pesticides in water. J. Chromatogr. A 795, 27e42. Eisert, R., Levsen, K., 1995. Determination of organophosphorus, triazine and 2,6-dinitroaniline pesticides in aqueous samples via solid-phase microextraction (SPME) and gas chromatography with nitrogenephosphorus detection. Fresenius J. Anal. Chem. 351A, 555e562. Eisert, R., Levsen, K., Wu¨nsch, G., 1994. Element-selective detection of pesticides by gas-chromatographyeatomic emission detection and solid-phase microextraction. J. Chromatogr. A 683, 175e183. Graham, K.N., Sarna, L.P., Webster, G.R.B., Gaynor, J.D., Ng, H.Y.F., 1996. Solid-phase microextraction of the herbicide metolachlor in runoff and tile-drainage water samples. J. Chromatogr. A 725, 129e136. Guehenneux, G., Wortham, H., Baussand, P., 2004. Determination of the detection limit of naphthalene using SPME coupled to a GC/ MS. Fresenius. Environ. Bull. 13 (11). Hernandez, F., Beltran, J., Lopez, F.J., Gaspar, J.V., 2000. Use of solid-phase microextraction for the quantitative determination of herbicides in soil and water samples. Anal. Chem. 72 (10), 2313e2322.
Lassagne, L., Baussand, P., 2000. A new way of measuring PAHs by solid phase microextraction. Fresenius Environ. Bull. 10, 381e385. Sabik, H., Jeannot, R., Rondeau, B., 2000. Multiresidue methods using solid-phase extraction for monitoring pesticides, including triazines and degradation products, in ground and surface waters. J. Chromatogr. A 885, 217e236. Sanusi, A., Wortham, H., Millet, M., Mirabel, Ph., 1996. Chemical composition of rain water and pollution origin in Eastern France. Atmos. Environ. 30, 59e72. Sanusi, A., Millet, M., Mirabel, P., Wortham, H., 2000. Comparison of atmospheric pesticide concentrations measured at three sampling sites: local regional and long-range transport. Sci. Total Environ. 263, 263e277. Sauret, N., Millet, M., Herckes, P., Mirabel, P., Wortham, H., 2000. Analytical method using gas chromatography and ion trap tandem mass spectrometry for the determination of S-triazines and their metabolites in the atmosphere. Environ. Pollut. 110, 243e252. Sng, M.T., Lee, F.K., Lakso, H.A., 1997. Solid-phase microextraction of organophosphorus pesticides from water. J. Chromatogr. A 759, 225e230. Valor, I., Molto´, J.C., Apraiz, D., Font, G., 1997. Matrix effects on solid-phase micro-extraction of organophosphorus pesticides from water. J. Chromatogr. A 767, 195e203. Zambonin, C.G., Palmisano, F., 2000. Determination of triazines in soil leachates by solid-phase microextraction coupled to gas chromatographyemass spectrometry. J. Chromatogr. A 874, 247e255. Zhang, Z., Yang, M.J., Pawliszyn, J., 1994. Solid-phase microextraction. Anal. Chem. 66 (17), 844Ae853A.
