Rapid Large Volume InjectionIGC-MS Analysis Food Prepared .

Rapid Large VolumeInjectionIGC-MS Analysisof Pesticides in Food Preparedby the QuEChERS e t h o d2Shmd9JEuropa GmbH, Ouisbwrg, Gemmy.In m r i y a d p a l e a r t h e r e k a t o s a w t i m esamples.inFor the prognmmbletmpwetture vaporizationm p k p a i w r . n h(PrW)g i h dnhrtsp&atg,k,tinerswre wed.,Theythe wl13;mQuECW4 {Qukk,Cheap, Effldmt,do rmt haw any firling mafeial, pmming decmpdtianRugged and SafdqJmethod hqs k e n applied In manyof fragile wciks.Capadty forthe ta7gemwdweI a h a t w b . This M d l ysamp;le prepamtbni p j d m lsachtedhyttre rough inner surfaceof theeffort whep ampared with the fwme tyd methad with linen. Thb surface was 'SlLfEi dexhttd, To m?mapiea final gel pmwtimchfmatqaphy (GPC)dean-upt k w h d e p m s aheraline wasdir&(che&d by astep. Incon&&. when hrjeching thewtracts prepad byspecial dqadation r&tW the UN Xautomatic timQuEthERS mnymtrhsignals an h &serwd irl them c h p was imkdhd ,andafter shut 80 injdonsGT-AA5 chmmtcgm, Full %anmodes ate, therefwe,the system peffomedthe lime n g e a w p o l l y .t w w s a y t o p m t f a l s e pcrcridverrr falseThe compound &ks mpdse af awr 500 pwidds.dminatim of targetk t muld easily w rI l n W o n of the tar@ mnpmmd was prbmd byw h d n gthe gas &mnaatography-m s s p c b m e y d w c h g full mcsgechaand by &ng tinear retention' d k e s a checkedl l y as an ac4dkbat filmmter(GC-MS) q s t m in the more sensitive selPonmonitoring (SIM). To reach a high sensitiuityfm routinew f k ih il%ascan mode, firstty, ihG C 4 S s y s m h dSample Preparatianbe d RigXI mHity instrurnenkandm d i y , a largeThe procedure inr0IM the extraction of 10 g sample W hv o l w kfectbnh&r imthe limk of quaMtcat:bn 10 mL wetonltiile. fdtowed by a liquid- uid-partitiming(WJ.r epWdb addrk14ganMm*MW4W1gIn this &de a methad wiled rapid large d u r nNaCl, 1 g Ma&& and 0.5 g Na2Mtrat. The mpki n j e c h was 4wkh a PW injection port (Optic 3, AT#dwn-up was p d m w i using a rapid p d u r e calledGI. Intertratioral). Wolumes up to 50 pl. were hrJectedwithdispersive solid-pk &maionIRwhkh 1% mgsubmyent full scan GC-MS runs and the q r c a t i w,anhydfw MgSb4and 25 mg pffmry m # a t y aminefPSA)sorb&rrt are m h d with I rntad6nWe extract. konwasdxckdbyrm1ysIngrwndmbjnWt mk&i(w,IContact authorE-mall.Hans-Ulrich BaiershirnadzuOshimadruau

Friedrichs, Winkeler and BaierAfter a second mixing and centrifugation step, the extractwas transferred to autosampler vials for concurrent analysisby large volume GC-MS.Experimental ConditionsThe instrumentation was a GCMS-QP2010 Plus (ShimadzuEuropa GmbH) with Optic 3 injector (ATAS GL International),AOC-5000 autoinjector (Shimadzu Europa GmbH) with anautomatic glass liner exchanger option (LINEX, ATAS GLInternational). The chromatographic conditions were:VF-5-MS EZ guard column 30 m x 0.25 mm, 0.25 pmwith an integrated retention gap of 10 m. The columntemperature was set to 50 "C for 1 min (hold) then with40 "Clmin to 150 "C followed by 4.6 "Clmin to 280 "C for28.24 min, with a mean linear velocity of 30 cmls (He). Forthe Optic 3 injector the temperature was set to 55 "C duringthe period of the solvent venting time and then rampedwith 15 "CIS to 280 "C for the rest of the analytical run(59.75 min). The solvent venting at low temperature (55 "C)was optimized and finally set to 38 s at a split ratio of 50:l.The split was further programmed to transfer the analytes tothe column.This was achieved by closing the split after the ventingtime for 2 minutes (analyte transfer). The split was thenreopened to 10:l to purge residual solvent out of the liner.The injection volume was finally set to 30 pL. The Optic 3is heated by direct ohmical heating. This leads to liner innerdiameters of this PTV of about 3.4 mm, corresponding totypical hot splitlsplitless liner dimensions.It is possible to ramp the Optic 3 up to a maximum of30 "CIS even using these liners, which is in contrast toconventional PTVs that have indirect resistive heaters andcorrespondingly have typical inner diameters of about1-2 mm. This has a strong influence on the methoddevelopment in large volume injections. For the Optic 3 theinjection speed up to about 100 pL is not as critical and,therefore, the liquid can be injected rather quickly [rapidlarge volume injection (RLVI)] while in the latter case a speedcontrol of injection is important.The mass spectrometer was operated in full scan modeto minimize false positive or false negative identification.The scan range was set to 50-550 mlz. The ion sourcetemperature and the interface was set to 200 "C and320 "C, respectively.The Columnwww.thecolumn.eu.comMarch 2009ResultsThe correct liner choice is critical to the success of any pesticideanalysis using PTV injection. The liner must be thoroughlydeactivated or many labile pesticides may decompose oradsorb in the inlet. For large volume injections the capacityof the glass insert is crucial. Any filling material such as glasswool or TENAX used in classical large volume injections thatincreases the injection volume capacity has to be avoidedeven if deactivated. For this reason, a glass insert with arough surface (sintered glass liner, ATAS GL International)waschosen. With these glass liners the inner surface is covered bymany small quartz beads to have a larger surface. When usinga syringe with a side hole needle the liquid injected will besprayed onto the wall surface of the liner. External experimentsshowed that even a 50 pL acetonitrile injection does notresult in dropping any liquid out of the liner. These liners weredeactivated by a double SILTEK (Restek) deactivation process.In Figure 1 such a liner is shown. The inertness of theglass insert after subsequent injections of pesticide matrixFigure 1: Sintered glass insert with glass beads on inner surfacefor Optic 3 with SILTEK deactivation.

March 2009Friedrichs, Winkeler and Baierwww.thecolumn.eu.com The ColumnWe sort thingsout0.0was checked by a degradation of dichloro-diphenyltrichloroethane (DDT) (also used in EPA 8270). Thedegradation of DDT must be below 20%. This was checkedautomatically in batch runs.After optimization of the Optic 3 injector parameters theanalytical performance of the repeatability was studied.Pesticide free extracts were spiked with a pesticides mixture(10-200 pg/L) and 10 subsequent 30 pL injections weremade; quantification was fully automated. The result of therelative standard deviations (RSDs) was calculated from theconcentration and was about 4.3% or better.In the following steps the linearity of response wasstudied with standard solutions prepared in matrix extracts.The calibration curves generated from the matrix matchedIFigure 2: Multiple pesticide-residuecalibration run of a mixture ofmore than 50 pesticides on a VF-5-MS 30 m x 0.25 x 0.25 pmwith an i n t n r a t rptpntinndnan nf 10 m.and savcfourmoney!Check out our monthlyoffers!IShodex "1Capture the essence1.Figure 3: Strawberry sample (Germany) from the field. Thepesticides cyprodinil 0.016 mglkg, fludioxonil 0.022 mglkg,trifloxystrobin 0.005 mglkg, fenhexamid 1.078 mglkg, boscalid0.032 mglkg and azoxystrobin 0.189 mglkg.1I

Friedrichs, Winkeler and Baier1-The Column-rTable I:Analytical data of a multi-residue pesticide yn213Spiroxamin 1100Fenpropidin98Spiroxamin ifloxystrobin116Quinoxyfen237Endosulphansulphate 50.01050.0530.06150.050.0245 0.0160.062400.05480.0510.230.05050.1261 0.0220.05050.05650.052650.049050.055350.06550.1875 0.0050.026250.04545www.thecolumn.eu.comMarch 2009No. RTSubstancem/z RIConc.363738394041FenhexarnidHexazinonPropagite 1TebuconazolPropagite 2HaloxyfopethoxyethylsterTriphenylphoshat(TPP - uinAcrinathrinCyfluthrin IFenbuconazolCyfluthrin 2Cyfluthrin 3Cyfluthrin 4Boscalid 750.0290000.0594 0.0320.12750.1584 0.18926.57826.82527.0627.10627.11927.23842 .77341.268Strawmglkg berrysamplestandards were used for quantification, so that possibleerrors as a result of enhancementlsuppression caused by thematrix effects could be minimized. All results were calculatedusing triphenylphosphate (TPP) as internal standard.For the analysis of carrot extracts, for example, thecorrelation coefficients obtained for the calibration plotsof all analytes were in the range 0.989-0.999 in theconcentration range of 0.002-1.3 mg1L.Applying this method, the lowest detection limit (LOD)for more than 500 analytes were in the range of 0.0020.020 mglkg depending on the substance.Figure 2 shows the calibration run of a multiplepesticide-residue standard spiked with a blank strawberrysample. In Table 1 the retention times, linear retentionindices (LRI) and concentrations are given.

M a h 2009wnmrthmmlurnn.muxom Thm ColmnFigure 3 shows the TIC chromatogram and the ion setsof cyprodinil, fludioxonil, trifloxystrobin and boscalid of astrawberry sample from the field (Germany). The target ionsshown in this figure correspond to a range of the LOQ of0.002-0.02 mglkg. The LOD for this strawberry example isin the range of 0.0005-0.001 mg/kg.Another critical point is the number of possible injectionsbefore the SILTEK deactivation is no longer stable enoughto give reliable quantitative data. For this purpose, adegradation check standard was analysed to check thecondition of the glass liner. A mixture of 5 ng pp'-DDTand endrin was periodically injected and analysed. Thebreakdown of ppl-DDT and endrin must be below 20%.pp'-DDD, pp'-DDE, endrin aldehyde and endrin-ketoneare the metabolites from endrin and pp'-DDT. The checkformula used was to calculate the total target areas of pp'DDT pp'-DDE ppX-DDDItotaltarget areas of pp'-DDE pp'-DDD x 100%. In a batch series every fifth injection wasdone with the degradation check standard diluted in matrixextram (for example apple matrix) and degradation checkvalues were calculated. SummaryThe quantitative determination of rnuttiresidue pesticidesin food matrix according to the QuEChERS method can besuccessfully combined with a rapid large volume injection(RLVI) and a full scan GC-MS method.The lowest determination limits (LOQ) were 0 . 0 0 2mg/kg. The observed RSDs of 4.3% and below indicate a highprecision in routine work. Up to 80 injections of RLVI of 30 pLinto a SILTEK deactivated sirrtered glass liner were possible.Referen1.2.Anastassiadeset al., h m l of A O A C l n t e m a I860).,(2003).T.Cajkaetal.,J.Sep.Sri.,28,104&1060 005).Karin Friedrichs holds a chemical engineering degreefrom Munster Technical University. Between 1981 and1985, she worked at Essen University's organic chemistryinstitute on structural determination of organic compoundsusing spectrometrical methods. Since 1985, she has beenworking at the Chemical and Veterinary Investigation Officein Detmold, Germany. She is responsible for the GC-MSanalysis and sample preparation of contaminants in food.Website: www.shirnadzu.euFriedrichs, Winkeler and BaierHeineDi%ter Winkeler obtained his dtxtor's degree atPaderborn University in 1983. Since 1990, he has beenhead of the laboratory of residue analysis at the Chemicaland Veterinary Investigation Office in Detrnold, Germany.His research interests include the development of analyticalmethods in the field of organic trace analysis using GC-MSand LC-MS-MS.Hans-Ulrlch Ealer studied physia and achieved his PhDdegree in 1991. He has been working for Shimadzu EuropaGmbH since 1992. He started as a product specialist forsurface analysis and mass spectrometry. Since 1996, he hasbeen working as a product specialist for GC/GC-MS.

development in large volume injections. For the Optic 3 the injection speed up to about 100 pL is not as critical and, therefore, the liquid can be injected rather quickly [rapid large volume injection (RLVI)] while in the latter case a speed control of inject