Pesticide Analysis EZ

Transcription

1 SPECIAL News from GERSTEL GmbH & Co. KG Eberhard-Gerstel-Platz 1 D Mülheim an der Ruhr Germany Phone Fax Intelligent Automation for GC/MS and LC/MS Pesticide Analysis EZ Fruit and Vegetables

2 Monitoring of Pesticides in Fruit and Vegetables One of the most important aspects of controlling pesticide exposure is to monitor pesticide residues in food, water and beverages. Many food samples require extensive sample preparation and sample clean-up in order to reach the required limits of determination and in order to keep the analysis system sufficiently stable for a routine analysis environment. Most established analytical methods are based on traditional liquidliquid extraction, gel permeation chromatography (GPC) or solid phase extraction (SPE) in combination with GC/MS or LC/MS. The benefits of automated sample preparation and automated matrix elimination for the determination of pesticides by GC/MS and LC/MS are demonstrated in the material shown in this GERSTEL Newsletter. GERSTEL offers a wide range of reliable and efficient automated sample preparation systems for GC/MS and LC/MS pesticide determination. Our solutions cover the full spectrum from aqueous through non-fatcontaining to fat containing samples: The GERSTEL Automated Liner EXchance ALEX technology for GC/MS helps to improve sample throughput in pesticide residue analysis especially when using the QuE- ChERS method for sample preparation. The GERSTEL SPE provides a fully automated solution including sample preparation and introduction to GC/MS or LC/MS. Disposable Pipette Extraction (DPX), a fast dispersive SPE technique that requires very little solvent, enables rapid and efficient extraction of analytes using only small amounts of solvent and, last, but not least, the SBSE technique (GERSTEL Twister ) is a well recognized technique for extracting and concentrating pesticides and other contaminants from aqueous samples. This special issue of the GERSTEL Solutions Worldwide Magazine brings you an overview of GERSTEL technology and applications in the field of pesticide analysis. Enjoy the magazine! Yours Ralf Löscher, Ph.D. International Sales Manager In this issue Automated Liner EXchange for GC Inlets Handling Dirty Samples Sample clean-up steps, which are needed in order to prepare for example environmental or food samples for pesticide analysis, are time-consuming and a potential source of errors. Simplification - or elimination - of such procedures is often the motivation when developing new analytical methods and new instrumentation. Unfortunately, analytical instruments normally do not tolerate introduction of dirty samples or even of extracts that have not been cleaned. For example, extracts containing suspended matter or high molecular weight compounds contaminate a GC inlet within a few injections, causing peak broadening or even loss of sensitive compounds. The GERSTEL Automated Liner EXchange enables routine analysis of dirty extracts without system down-time since dirty liners are automatically replaced. Page 3. Automated Solid Phase Extraction (SPE) Performing Solid Phase Extraction (SPE) manually can be time consuming and nerve-racking, especially when recovery and reproducibility are lacking due to sample variability. If SPE can be reliably automated, it becomes a much more efficient and reproducible process. The GER- STEL SPE provides several benefits compared with manual SPE: Improved recovery, precision and reproducibility as well as maximized sample throughput by performing SPE during GC or LC analysis of the preceding sample. In many cases this translates to more than 50 percent time saving for the overall analysis, compared to manual processing. Following the SPE cleanup steps, the MPS can introduce the sample extract directly to the LC/MS or GC/MS system for analysis. A method used for the determination of up to 185 different pesticides using GC/ MS and LC/MS is reported. Page 6. Automated Disposable Pipette Extraction (DPX) Disposable Pipette Extraction (DPX) is a fast and efficient SPE technique used for a wide range of applications such as pesticide monitoring. Only µl of sample is needed to reach the required limits of detection using a fully automated process. The extraction step is performed in seconds and the complete process including elution and rinse steps takes 3-6 minutes depending on the application. Additionally, automated DPX is performed during the GC or LC run of the preceding sample ensuring maximum throughput and best possible GC/MS or LC/MS system utilization. Elution requires only a small amount of solvent, which means that DPX effectively provides a concentration step. For many applications, such as pesticides in fruit and vegetables, solvent evaporation is not required - or it can be automated by using Large Volume Injection (LVI) techniques. Page 9. Imprint Published by GERSTEL GmbH & Co. KG Eberhard-Gerstel-Platz Mülheim an der Ruhr Germany Editorial Director Guido Deußing ScienceCommunication Neuss, Germany guido.deussing@t-online.de Translation and editing Kaj Petersen kaj_petersen@gerstel.de Scientific advisory board Eike Kleine-Benne, Ph.D. eike_kleine-benne@gerstel.de Oliver Lerch, Ph.D. oliver_lerch@gerstel.de Malte Reimold, Ph.D. malte_reimold@gerstel.de Contact gerstel@gerstel.com Design Paura Design, Hagen, Germany 06/09 ISSN

3 New technology for handling dirty samples Automated liner exchange for GC injectors Sample clean-up steps needed to prepare environmental or food samples for determination of pesticides are time-consuming and are potential sources of errors. Simplification or elimination of such procedures is often the motivation for development of new analytical methods and new instrumentation. Unfortunately, analytical instruments do not normally tolerate introduction of dirty samples or even dirty extracts. For example, extracts containing suspended matter or high-molecular-weight compounds contaminate a GC inlet after a few injections, causing peak broadening or even loss of sensitive compounds. Reducing or eliminating clean-up steps will result in dirty extracts and daily or even hourly maintenance of the GC system. The GERSTEL Automated Liner EXchange (ALEX) enables routine GC analysis of samples containing large amounts of matrix or other solid residue. After several injections, matrix material deposited in the GC inlet leads to adsorption and loss of active analytes, such as pesticides. ALEX replaces the GERSTEL CIS inlet liner at user-defi ned intervals, eliminating the need for time-consuming clean-up steps during sample preparation. System Design for Automated Liner Exchange A simple and automated liner exchange system is able to overcome most chromatographic problems caused by dirty samples in GC analysis. A solution is presented that uses a commercially available programmable temperature vaporizing (PTV) inlet in combination with an autosampler, which can automatically perform a liner exchange at any time during a sample sequence. Every liner is equipped with a transport adapter, which also allows liquid injection through a septum. Adapters fitted with liners are transported by means of the autosampler which also performs the liquid injection. The system is based on the Cooled Injection System (CIS 4) inlet and the MultiPur- Figure 1: Automated Liner Exchange (ALEX) installed on an Agilent 7890 GC equipped with a CIS 4 programmed temperature vaporization (PTV) type inlet. Detailed view of the ALEX System. 173

4 Figure 3: GC FID chromatogram of a Grob test mixture (1 µl splitless) injected into a CIS 4 equipped with the ALEX system pose Sampler (MPS 2) (GERSTEL, Germany). Instead of the septumless head normally used on the CIS 4 for liquid injection, a special support head is mounted. This support head seals the transport adapters, providing an un-compromised carrier gas flow throughthe adapter and liner. The support head and the transport adapters are conical. In order to provide a perfect seal, every transport adapter is fitted with two o- rings, between which the carrier gas inlet is placed. Such a sealing system has been proven through years of use in other systems where glass tubes are automatically exchanged, such as the GERSTEL Thermal Desorption System (TDS). In order to grip and transport the adapters, the autosampler has been modified slightly and fitted with an electrical gripper. Up to 97 conditioned liners are stored in a special tray; the transport adapters provide a gas-tight seal for contamination-free storage. For liquid injections, every transport adapter is equipped with a 5 x 3 mm septum that is commercially available and is, for example, used in Agilent s cool on-column inlet. The top of the injector, and the transport adapter in particular, remain cool during the analysis due to effective heat de- Figure 2 Left: Exploded view of transport adapter for automated liner exchange with liner, adapter, 3 x 5mm septum and septum screw; a hole for carrier gas entry can be seen between the o-rings of the transport adapter. Right: Assembled transport adapter with liner Weigh 10 g of sample > Add 10 ml of Acetonitrile (AcN) Shake vigorously 1 min > Add 4 g MgSO 4 and 1 gnacl Shake vigorously 1 min > Add internal standard solution Shake 30 sec and centrifuge Take Aliquot of supernatant > Add MgSO 4 and sorbent Shake 30 sec and centrifuge GC-MS and LC-MS Table 1: QuEChERS method sample preparation steps for multi-residue analysis of pesticides in non-fatty foods such as fruits and vegetables Figure 4. Screen shot of a control sequence for ALEX embedded in Agilent GC ChemStation and MS ChemStation. coupling between the body and the top of the PTV. As a result, no septum bleeding or bleeding of the o-rings of the adapters can be observed. The body of the injector is identical to the CIS 4 inlet and all commercially available types of liners for this inlet can be used (empty liners or liners filled with glass wool or adsorbents). The automated liner exchange head doesn t affect the analytical performance of the CIS 4 inlet. As an example, Fig. 3 shows a chromatogram of a Grob test mixture with uncompromised peak resolution and peak shapes. Tests with n-alkane mixtures proved that recovery of high boiling substances is comparable to a normal CIS 4 system. It is clearly seen that this new Automated Liner EXchange (ALEX) system has no influence on CIS 4 performance. Methods developed for CIS 4 system can be transferred to the ALEX system without any modifications. A software solution was developed that enables the user to exchange liners at any point in time during the analysis sequence. The software can be operated stand-alone or integrated into the Agilent GC or MS ChemSation. This means that only one sequence list is needed for the complete system. Fig. 4 shows a screen shot of such a sample sequence. Pesticide analysis of non-fatty foods with reduced sample preparation Recently a new multi-residue method for pesticide analysis in fruits and vegetables was presented (QuEChERS, Quick Easy Cheap Effective Rugged Safe) [1]. Compared to previous methods, the QuECh- ERS method enables rapid sample preparation for determination of pesticides such that 8 samples can be prepared in less than 30 minutes. Table 1 summarizes all necessary steps of the QuEChERS method. The main benefit of this sample preparation method is a less time-consuming analysis, which is also less error-prone. Unfortunately, extracts obtained following this procedure often have a high matrix content, which causes chromatographic problems for GC analysis due to residue build-up in the liner. Fig. 5 shows a picture of a 2 ml vial containing a bell pepper extract and a glass wool packed liner in which 5 µl of this extract has been injected. 184

5 Residue build-up in the GC liner very quickly affects the analysis of many pesticides, as can be seen in Fig. 6. A 5 µl standard solution made up in a bell pepper matrix was injected 20 times into a deactivated baffled liner. Peak area trends for three different pesticides are presented. For endosulfane sulphate and chlorothalonil, peak areas decrease over the course of the 20 injections. This can be explained with increasing matrix contamination of the liner, leading to loss of analytes. For dichlorophos the situation is different; the peak areas increase. This effect is described in the literature as matrix-induced chromatographic response enhancement [2]. This means that matrix components cover remaining active sites in the chromatographic system leading to higher response for sensitive analytes. Fig. 6 shows clearly that when analyzing extracts obtained with the QuEChERS method, a liner exchange is required after 10 or at least 15 runs for vegetables like bell peppers. In order to implement the QuEChERS method in a laboratory for a routine automated analysis, it is absolutely necessary to have the capability of exchanging liners automatically as is provided by the new GERSTEL ALEX system. Compound sd Compound sd 1 Cyhalothrin 9.5% Imazalil 7.2% 2 Cyhalothrin 6.7% Cresoxim-methyl 6.7% Atrazine 9.0% Methamidophos 6.5% Azoxystrobin 5.9% Permethrin 7.0% Bifenthrin 6.8% Permethrin 2 6.8% Carbaril 12.9% Procymidone 4.7% Chloropyriphos-methyl 6.9% Tebuconazol 6.8% Chlorpyrifos-ethyl 8.5% Thiabendazole 6.8% Chlorthalonil 29.3% Tolylfluanide 8.2% Cyprodinil 7.7% Trifluraline 6.9% Dichlorvos 12.0% Tritane 3.5% Endosulfan sulphate 12.7% o-phenylphenol 6.8% Ethion 8.0% p,p -DDD 6.1% Table 2 Standard deviations for different pesticides achieved under optimized injection conditions (see Table 3) for 10 injections in one liner. Even though only 10 injections per liner are performed, standard deviations are still high for several pesticides. Apart from the specifi c chemistry of some pesticides, this is due to the fact that 5 µl of the acetonitrile extract had to be injected into an empty liner. Acetonitrile is known not to be a suitable solvent for GC analysis. Normally for such a solution, and such an injection volume, a liner with glass wool should be used. Unfortunately, some substances are very sensitive and show discrimination on glass wool liners. On the other hand a 5 µl injection volume is necessary in order to meet required detection limits of 0.01 mg/kg. The following is an example of a sequence for routine analysis: Liner Injection Volume Injection Speed Injection Mode CIS 4 Temp. Program Empty baffled liner (deactivated) 5 µl (AcN solutions) 10 µl/sec Solvent vent for 15 sec (50 ml/min; 8.2 psi), Splitless sample transfer, Purge Flow sec 50 C (0.25 min) - 12 C/sec 280 C (30 min) Table 3: Injection conditions for Pesticide Analysis after QuEChERS sample extraction method, GERSTEL MPS 2 with ALEX system, GERSTEL CIS 4, Agilent 6890 GC, Varian FactorFour XMS column (30 m, 0.25 mm ID, 0.25 µm fi lm), Leco Pegasus 3 TOF-MS; Liner: Empty baffl ed liner, deactivated. Figure 5: 2 ml Vial containing a QuEChERS method bell pepper extract and a liner packed with glass wool in which 5 µl of this extract has been injected. Figure 6: Peak area trend (GC / TOF-MS) for 5 µl injections of a standard solution in matrix (bell pepper) into an empty, deactivated baffl ed liner 1 Liner Exchange 2 Standard Injections for recalibration (3 or 5 concentration levels) 3 Sample Injections (7 up to 10 runs) 4 Liner Exchange 5..Steps 2-4 are repeated.. Conclusions The system described herein for automatic exchange of PTV inlet liners enables automated GC analysis of samples or extracts with a high content of high boiling substances or suspended matter. The chosen application, determining pesticides in fruits and vegetables with QuEChERS sample preparation, demonstrates that a reduction of sample preparation steps combined with a GC system which tolerates injection of solutions with a high matrix content is a powerful solution. Laboratory time for sample preparation is reduced dramatically and at the same time a high sample throughput for the analytical instrument is ensured. References [1] M. Anastassiades, S. Lehotay, D. Stajnbaher and F. Schenck: Fast and easy multiresidue method employing acetonitrile extraction/ partitioning and dispersive solid-phase extraction for the determination of pesticide residues in produce. J AOAC Int 86(2) (2003) [2] M. Anastassiades, K. Mastovska, S.J. Lehotay: Evaluation of analyte protectants to improve gas chromatographic analysis of pesticides. J. Chromatogr. A 1015 (2003)

6 Pesticide analysis EZ When the sample matrix no longer matters Application specialists from TeLA GmbH have developed a new method that dramatically simplifi es LC/MS determination of pesticide levels, providing high-quality results independent of the sample matrix type and complexity. Norbert Helle, Ph.D. and Meike Baden, TeLa GmbH, Fischkai 1, Bremerhaven, Germany; Phone: / ; Fax: / ; postfach@tela-bremerhaven.de Fully automated Sample clean-up and Pesticide Screening with Agilent 6410 LC/MS QQQ online SPE System, Agilent ordering Number: EN 6

7 Pesticides, fungicides and herbicides are needed in order to provide an adequate supply of food to the ever-growing human population across the world. The other side of the coin is that residues of these types of compounds in foods cannot be allowed to endanger or affect the health of the consumer. To ensure that foods do not endanger us, maximum acceptable levels, sometimes referred to as tolerated levels, have been established for individual compounds according to the current state of scientific knowledge. If these levels are exceeded, it would be illegal to market the contaminated product in Europe. Corresponding laws were established by the EU. This legal basis must be, or, in some cases, already has been, adapted into National law by EU member states. In Germany, the details on maximum acceptable levels of residues can be found in the German LF- GB, acronym for the compendium of laws governing Food, Feed and various consumer products. As an aside, the term consumer products in this context spans a great variety of products ranging from packaging that comes into contact with food, feed or personal care products to personal care products themselves, such as cosmetics, tooth paste or shampoo or other personal care items that make more than brief contact with skin or mucous membranes. Rules, unless properly enforced, are of course worth less than the proverbial paper they were printed on. In other words trust is fine, but we should verify and if needed take corrective action to ensure compliance and best possible consumer safety. This requires a network of reliable laboratories, which is not a trivial matter, as can be seen a bit later in this text. World-wide, around 700 pesticides are in use, very few of which can be legally used throughout Europe. Various compounds classes have been established, but even these can cover a wide range of polarities, making it difficult to develop a fast all encompassing analysis method. Still, effective multi-residue methods are in use for the determination of pesticides, helping to ensure food safety. When fruits and vegetables are analyzed for pesticide residues, often several pesticides are found. The effects on human health have only been documented for very few of these compounds or compound groups. Tracking down pesticides using GC/MS and LC/MS Classical pesticide analysis relied on gas chromatography (GC) using an electron capture detector (ECD) or a nitrogen phosphorous detector (NPD). The most widely used detector today is the mass selective detector (MSD). In Germany, the analytes that are mainly in focus are those listed in the DFG S19 method, a multiresidue method for the determination of pesticides in food, which enjoys Europe-wide recognition. The analysis of the 270 compounds listed in the S19 method does, however, require significant sample preparation including a gel chromatography clean-up step to separate analytes from the matrix. Different analysis techniques are used for different types of pesticides. Liquid chromatography (LC) combined with a mass selective detector (MS) is used to determine polar to moderately apolar compounds. Gas chromatography (GC), most often in combination with a mass selective detector (MSD) covers apolar to moderately polar compounds. As can be seen from this description, there is some overlap between the techniques. Recently a new multi-residue method for the determination of pesticide levels in fruits and vegetables was presented (QuECh- ERS: Quick, Easy, Cheap, Effective, Rugged & Safe) [*]. Compared to previous methods, the QuEChERS sample preparation steps are much less time-consuming, enabling the preparation of 8 samples in less than 30 minutes. QuEChERS is a sample preparation method well suited for both GC, GC/MS and LC/MS analysis. The QuEChERS sample preparation steps are listed below. The main benefit of this sample preparation method is that the overall analysis is less time-consuming and less error-prone than more traditional approaches. Unfortunately, extracts obtained following this procedure often have a high matrix content, which causes chromatographic problems for GC analysis due to residue build-up in the liner unless an automated liner exchange system such as the GERSTEL ALEX is used. (Cf.: GERSTEL Solutions Worldwide Magazine No. 5 p. 18) ( QuEChERS method: Weigh 10 g of sample > Add 10 ml of Acetonitrile (AcN) Shake vigorously 1 min > Add 4 g MgSO4 and 1 gnacl Shake vigorously 1 min > Add internal standard solution Shake 30 sec and centrifuge > Take Aliquot of supernatant > Add MgSO4 and sorbent Shake 30 sec and centrifuge > Take Aliquot of supernatant > inject to GC-MS and LC-MS The results obtained using QuEChERS sample preparation are comparable to those reached using the S19 method. The QuEChERS method is much faster, requires much less sample preparation, covers a wider range of analytes and is more readily automated. In addition, much smaller volumes of partly toxic organic solvents are required, compared with other currently used methods for determining pesticides in fruits and vegetables. In addition to the financial benefits of a much higher laboratory throughput, the cost of materials at around one Euro per sample is relatively low. The limits of QuEChERS are encountered whenever samples with more complex matrices need to be analyzed, such as garlic, onion, artichoke or avocado Calibration curves for nine pesticides, determined using the TeLA GmbH SPE-LC-MS/MS pesticide multi-residue method. [*] M. Anastassiades, S. Lehotay, D. Stajnbaher and F. Schenck: Fast and easy multiresidue method employing acetonitrile extraction/partitioning and dispersive solid-phase extraction for the determination of pesticide residues in produce. J AOAC Int 86 (2) (2003)

8 Norbert Helle and Meike Baden in front of the SPE LC-MS/MS system used for the pesticide multi-residue method: Agilent Series LC 1200 and a GERSTEL SPE system mounted across an Agilent 6410 MS/MS Triple Quad. with much higher fat content. This can lead to problems with interferences, than can especially influence quantification unless further clean-up steps are performed. To enable reliable and rugged analysis independent of the sample matrix, we looked for a similarly effective alternative sample preparation procedure. We found that automated solid phase extraction (SPE) based on the GERSTEL MultiPurpose Sampler (MPS) provided an excellent solution. The GERSTEL SPE, we have previously used successfully for a number of applications, including aflatoxins, chloramphenicol and malachite green in foods. In summary, we can report that our automated SPE-LC-MS/MS-ESI multi-residue method reduces the number of manual steps required to a minimum while increasing laboratory throughput. The results are solid and reproducible combined with high sensitivity and good limits of determination. Instrumental requirements The GERSTEL SPE was fitted with an injection valve; sample introduction to the Agilent LC 1200 was performed directly by the SPE system; detection was performed using an Agilent 6410 MS/MS Triple Quad instrument. Sample Preparation: 15 ml of an acetonitrile/water mixture (80:20) was added to a five gram sample of fruit or vegetable for extraction. The SPE cartridge (M&N C-18ec, 6 ml, 1 g) was conditioned using 10 ml methanol (MeOH) and 10 ml water. All steps in the sample preparation procedure, including sample introduction were fully automated. 5 ml sample was added to the cartridge, which was subsequently rinsed with 5 ml water. Analytes were then eluted using an acetonitrile/water mixture added at a flow rate of 600µL/min. In contrast to most manual SPE methods, the liquid is not aspirated through the cartridge under vacuum, rather it is added under positive pressure using a syringe. This means that flows, and therefore also the elution speed, are accurately controlled and results more reproducible. This holds true even when sample matrix changes the restriction across the cartridge. The eluate was concentrated for six minutes at 50 C and the residual analytes taken up in 5 ml of a acetonitrile/formic acid mixture (30:70). Sample introduction and analyte separation: 20 µl of the cleaned-up extract was introduced directly to the LC/MS-MS System. The temperature of the column (ZorbaxXDB C x2.1 mm, 1.8 µm rapid resolution) was set to 50 C; flow rate: 0.5 ml/ min resulting in a column head pressure of approximately 420 bar. A solvent mixture of 5mM formic acid (A) and acetonitrile (B) was used as mobile phase based on the following gradient programming: 0 min (20 % B); 5 min (20 % B), 30 min (90 % B). Detection: Analytes were detected with positive Electron Spray Ionization (ESI) using the electron spray ion source or, alternatively, the Agilent Multimode ion source. Our experiments clearly showed that the Multimode source provided significantly lower detection limits for some pesticides than the ESI source. For other compounds, however, a lower response was obtained than with the ESI ion source. The settings for the ion source were optimized for the flow and eluent used. The following parameters were used: N 2 temperature: 340 C; carrier gas flow (N 2): 9 L/min; nebulizer pressure: 30 psi. The triple quadrupole instrument was operated in MRM mode, with 5 different time segments, monitoring two transitions for each pesticide. In each segment 40 to 50 analytes were monitored. The proof of the pudding When using the QuEChERS method, it is necessary to adapt the clean-up steps to the sample at hand. It has been clearly shown that for uncomplicated matrices, such as lettuce or cucumber, additional cleanup steps are not required following the acetonitrile/ water extraction. For complex matrices that contain fat and other challenging matrix components, further clean-up steps are of course needed. For this purpose we used the GERSTEL SPE system. Raw sample extracts were automatically loaded onto standard SPE cartridges and cleaned. A new cartridge was used for every sample to eliminate crosscontamination. Macherey-Nagel cartridges containing C18 reversed phase material were found to produce excellent, reliable results. Automated SPE clean-up as described in this article took around 20 minutes to complete. Apart from the first sample, the SPE process was performed dur 8

9 ing LC/MS or GC/MS analysis of the preceding sample, ensuring that the SPE step was performed without increasing the overall analysis time. Once the first sample had been prepared for analysis, the LC/MS or GC/MS system never had to wait idly for the next sample. An LC 1200 Rapid Resolution HPLC system from Agilent Technologies was used for the analysis. In order to achieve good separation combined with method ruggedness, the conscious decision was made to only seek a moderate reduction of the analysis time. The total analysis time required to determine around 140 compounds was in the order of 35 minutes. This time period was more than sufficient to prepare the following sample for just-in-time sample introduction to the LC/MS system. Sample clean-up using SPE contributes not only to the ruggedness of the method, it also improves reproducibility and linearity, among other things. To illustrate this, a bell pepper sample was spiked with a pesticide mixture and analyzed. Following SPE clean-up, retention times and peak areas of the analytes showed excellent reproducibility. The linearity was excellent, both for polar compounds like Carbendazim und Thiabendazole as well as for apolar pesticides like Diazinon und Pirimiphosmethyl. Orange oil samples were cleaned up using a slightly modified SPE method. The efficiency of SPE cleanup is illustrated by the fact that the intense yellow color of the sample was transferred to the cartridge while the resulting extract was a clear and colorless liquid. Recovery for the various compounds in this difficult matrix ranged from 70 to 90 % while recoveries from fruit and vegetable samples were mainly in the range from 80 to 100 %. It is worth noting that the Zorbax SB-C18 Rapid Resolution columns used provided excellent peak symmetry. One final comment: Every method must prove its worth in practice. The test, as always, is in the analysis of real world samples. To prove the validity of our Overlay medium polarity sections of 8 different chromatograms: 8 separate sample preparations and injections of a bell pepper sample spiked with a standard mixture of pesticides, 100 ng/ml each. The peaks shown are for the pesticides Terbutylazin, Cyprodinil, Prochloraz, Flusilazol and Fenoxycarb, all showing good reproducibility. method, we took part in a Europe-wide round robin with 46 participating laboratories. A vegetable sample (zucchini) had to be analyzed for 185 different pesticide residues. Out of 46 laboratories, TeLA GmbH was among the 12 that managed to correctly identify and quantify the analytes thus meeting the round robin requirements and passing the test. 128 of the 185 pesticides were determined using our SPE-LC-MS/MS pesticide multi-residue method. 90 of the 185 pesticides were determined using a GC/ MS system (GC 6890 / MSD 5973, both from Agilent Technologies) in combination with the GERSTEL MultiPurpose Sampler (MPS) using a Retention Time Locking (RTL) method. Carbendazim Determination of polar and apolar pesticides respectively in orange oil. Overlay chromatograms covering 9 different concentrations are shown. Thiabendazol 9

10 Disposable Pipette Extraction (DPX) Automated multi-residue Pesticide Analysis in Fruits and Vegetables by DPX-QuECHERS One of the most important aspects of reducing pesticide exposure is monitoring of pesticide residues in foods. A number of analytical methods have been developed, many of them based on traditional liquidliquid extraction, gelpermeation or solid phase extraction in combination with GCMS or LC-MS. Recently, the QuEChERS (quick, easy, cheap, effective, rugged and safe) sample preparation methods have been developed to help monitor pesticides in a range of food samples. These methods, however, still require many manual steps, such as centrifugation, leading to increased total analysis time. There is a need for a simple, reliable and readily automated technique to clean up QuEChERS type extracts in order to improve laboratory productivity for monitoring pesticide residues in foods. In this study, we present a novel solidphase extraction technique called disposable pipette extraction (DPX). The solid-phase sorbent contained in the DPX tip is loose, which AN/2009/1-2 permits mixing of solutions to provide unsurpassed extraction effi ciency and short equilibration times. DPX extractions are automated using the GERSTEL MultiPurpose Sampler (MPS), enabling efficient, high-throughput sample preparation. The GERSTEL DPX- Q and the DPX-Qg with graphitized carbon black, represent the only commercially available automated QuEChERS application for multi-residue analysis of pesticides. DPX is a fast and efficient solid phase extraction technique used for a wide range of applications such as drugs of abuse, therapeutic drug monitoring, comprehensive screening, pharmacology studies, as well as pesticides in fruits and vegetables. The DPX process is shown schematically in Figure 1. If needed, the sorbent is conditioned with solvent prior to extraction. The sample is then drawn into the pipette tip for direct contact with the solid phase sorbent. Turbulent air mixing creates a suspension of the sorbent in the sample ensuring optimal contact and highly efficient extraction. The extracted sample is discharged, typically after 30 seconds. If needed, the sorbent can be washed to remove unwanted residue. The extract is then eluted into a vial for subsequent LC or GC analysis. For sample cleanup methods, such as QuEChERS which focuses on removing fatty acids and water, the DPX method simply incorporates the steps of aspirating the sample solution, mixing with the sorbent, and dispensing the solution into the vial for analysis. There are no wash- or elution steps, the extractions can therefore take place in less than 1 minute. The GERSTEL MPS 2 with MAESTRO software control automates the entire process including sample introduction. In the following study, DPX-Qg tips are used to remove sample interferences from QuEChERS type extracts prior to GC/MS analysis. Experimental Instrumentation: Analyses were performed on a 7890 GC equipped with a 5975C MSD with triple axis detector (Agilent Technologies), PTV inlet (CIS 4, GERSTEL) and MPS 2 robotic sampler with 10 µl syringe (GERSTEL). Figure 2. Picture of extracts before and after clean-up. Standard preparation: A composite standard of organochlorine and organophosphate pesticides was prepared at a concentration of 1000 µg/l in acetonitrile. The standard was diluted to 20, 50, 100 and Automated DPX process All steps are performed automatically by the MPS. If needed, the sorbent is conditioned with solvent prior to the extraction process Sample is drawn into the pipette tip for direct contact with the solid phase sorbent. There is no contact between the sample and the syringe used to aspirate the sample and therefore no risk of cross contamination. Air is drawn into the pipette tip from below through the frit. Turbulent air bubble mixing creates a suspension of sorbent in the sample, ensuring optimal contact, highly efficient extraction, and high recovery. The extracted sample is discharged, typically after 30 seconds. If needed, the sorbent can be washed to remove unwanted residue. Extracted analytes are eluted using a suitable solvent, which is added from above for most efficient elution. The eluate is collected in a vial for subsequent sample introduction to LC/MS or GC/MS. The total time required for extraction in the examples shown in this article was always less than 6 minutes. Sample preparation and GC/MS or LC/MS determination can be performed in parallel for best possible throughput and system utilization. 10

11 250 µg/l. Twent y - f ive microliters of a matrix matching solution were added to the standards. Sample preparation: Processed fruit and vegetable extracts were provided by the University of South Carolina. The pesticide standard was diluted to obtain concentrations of 20 and 200 µg/l in 500 µl of the extracts. DPX extraction: 1 ml QuEChERS DPX tips were provided by DPX Labs, LLC. 500 µl of vegetable extract was manually transferred into a test tube. The extracts, 0.5 ml each, were drawn through the DPX tips 3 times. This process was automated using the GERSTEL MPS 2 autosampler with 2.5 ml syringe. The extract was then transferred into a 2 ml autosampler vial. 1 µl of eluent was injected into the GC. Results and Discussion The spinach extract was cleaned using a DPX Qg-1TA tip and the orange extract was cleaned using both the Q-1TA and Qg-1TA tips. The Qg-1TA tips contain graphitized carbon black which is more effective in removing chlorophyll from the extracts. Figure 2 shows a photograph of the spinach extracts before and after processing with the DPX tips. The green color is effectively removed from the spinach sample. The orange sample still had a mild orange tint when the Q-1TA tip was used. The Qg-1TA tip effectively removed all color and was used for all subsequent experiments. Figures 3 show chromatographically the effectiveness of the cleanup of the samples using DPX for the spinach sample, respectively. The DPX Qg-1TA tips effectively remove interferents, especially free acids in the minute retention time window. Each extract type was spiked at 20 and 200 ppb in triplicate and cleaned up using DPX. An example chromatogram for a spinach extract spiked at 200 ppb is shown in Figure 4. Table 1 shows the results for the DPX cleanup. The recoveries were calculated from an external six point calibration plot for each analyte. The external standards were matrix matched. The results show good recoveries for the mix of OC and OP pesticides used in this study. The average % RSDs (n=3) are less than 10 % for both matrices at both 20 and 200 ppb spike levels. The recoveries at the 200 ppb level range from % with an average value of 119 % for the orange extracts and range from % and average 91 % for the spinach extracts. These values could be improved with further optimization of the DPX automation method (number of extracts and aspiration speeds) and the use of internal standards in the GC/MS method. The recoveries with and without DPX cleanup are shown in Table 2 for the spinach and orange extracts, clearly showing the elimination of matrix interferences with the DPX Qg-1TA tip. Conclusions This study demonstrates the feasibility of using the DPX Qg-1TA for cleanup of QuEChERS type extracts Spinach Extracts Analyte % Recovery % RSD 20 ppb 200 ppb 20 ppb 200 ppb Dichlorvos Mevinphos Phorate BHC BHC Diazinon Methyl Parathion Ronnel Aldrin Trichloronate Heptachlor Epoxide t-chlordane Prothiofos Dieldrin Endrin ß-Endosulfan Fensulfothion Sulprofos DDT Endrin Ketone Average Table 1. Pesticide recoveries and % RSD. Abundance Time--> Abundance Time--> Figure 3. Chromatograms of spinach extract before (A) and after (B) clean-up. Abundance Dichlorvos Mevinphos Time--> Phorate -BHC Diazinon Figure 4. Chromatogram of spinach extract, spiked with 200 ppb, after DPX clean-up. -BHC Methyl Parathion Ronnel Aldrin Trichloronate Heptachlor Epoxide Chlordane Prothiofos Dieldrin -Endosulfan Fensilfothion Sulprofos DDT Spinach Compound No DPX DPX Dichlorvos Mevinphos Phorate BHC BHC Diazinon Methyl Parathion Ronnel Aldrin Trichloronate Heptachlor Epoxide t-chlordane Prothiofos Dieldrin Endrin ß-Endosulfan Fensulfothion Sulprofos DDT Endrin Ketone Average A B prior to GC/MS analysis. The DPX tips remove matrix interferents, leading to better analyte recovery and reducing the need for maintenance since there is less buildup of non-volatile material in the GC inlet. Full automation of the sample cleanup and injection is accomplished using the GERSTEL MPS 2. Table 2. Pesticide recoveries with and without DPX clean-up; spike level = 200 ppb Endrin Keton 11

12 About GERSTEL Pesticides in Water A method for fast screening of pesticide multi residues in aqueous samples using dual stir bar sorptive extraction (dual SBSE) - thermal desorption (TD) fast GC/MS has been developed. Recovery of 82 pesticides organochlorine, carbamate, organophosphorous, pyrethroid and others for the SBSE was evaluated as a function of octanol water distribution coefficients (log Ko/w: ), sample volume (2-20 ml), salt addition (0-30 % NaCl), and methanol addition (0-20 %). The optimized method consists of a dual SBSE performed simultaneously on respectively a 20 ml sample containing 30 % NaCl and a 20 ml sample without modifier (100 % sample solution). One extraction with 30 % NaCl is mainly targeting solutes with low Ko/w (log Ko/w < 3.5) and another extraction with unmodified sample solution is targeting solutes with medium and high Ko/w (log Ko/w > 3.5). After extraction, the two stir bars were placed in a single glass desorption liner and were simultaneously desorbed. The desorbed compounds were analyzed by fast GC/MS using a modular accelerated column heater (MACH). The method showed good linearity (r2 > ) and high sensitivity (limit of detection: < 10 ng/l) for most of the target pesticides. The method was applied to the determination of pesticides at ng/l levels in river water and brewed green tea. n Further information AppNote 12/2008 ( com/pdf/p-gc-an corrected.pdf)) GERSTEL develops, produces and supports solutions that include automated sample preparation and sample introduction for GC/MS and LC/MS. The available techniques include automated Solid Phase Extraction (SPE), which can be performed in combination with Standard Addition, Derivatization, and Eluate Concentration with or without Keeper Solvent. Additionally, in combination with a GC or GC/MS system, Automated Liner Exchange (ALEX) enables automated matrix elimination in the GC, allowing the user to dramatically reduce the amount of effort going into sample preparation, for example by using the QuE- ChERS sample preparation method for the determination of pesticides in non-fatty fruits and vegetables. Sample preparation is performed simultaneously with the LC/MS or GC/MS run of the preceding sample enabling highest possible productivity and system utilization. The system is controlled through the GERSTEL MA- ESTRO software in stand-alone mode or fully integrated with the Agilent Technologies GC/MS or LC/MS software. One method and one sequence table controls the complete system. G L O B A L A N A L Y T I C A L S O L U T I O N S GERSTEL GmbH & Co. KG Eberhard-Gerstel-Platz Mülheim an der Ruhr Germany gerstel@gerstel.com GERSTEL, Inc. 701 Digital Drive Suite J Linthicum, MD USA sales@gerstelus.com GERSTEL AG Enterprise Surentalstrasse Sursee Switzerland gerstel@ch.gerstel.com GERSTEL K.K Nakane, Meguro-ku Tokyo Dai-Hyaku Seimei Toritsudai Ekimae Bldg 2F Japan info@gerstel.co.jp Subject to change. GERSTEL, GRAPHPACK and TWISTER are registered trademarks of GERSTEL GmbH & Co. KG. Printed in Germany 0309 Copyright by GERSTEL GmbH & Co. KG