北京－食品安全中日韓論壇－－潘燦平：農(nóng)藥殘留檢測技術(shù)及其應用－－從取樣到結(jié)果的判定課件

1、農(nóng)藥殘留檢測技術(shù)及其應用從取樣到結(jié)果的判定潘燦平中國農(nóng)業(yè)大學農(nóng)業(yè)部農(nóng)產(chǎn)品質(zhì)量監(jiān)督檢驗測試中心（北京）Email： TEL: 010 627332192008.1.11 中日韓食品安全分析技術(shù)與應用學術(shù)報告會Pesticides are chemicals that we use to kill undesireable organisms. 對農(nóng)藥的認識：1） 定義的拓展2） 化學品使用必不可少3） 安全、合理使用（環(huán)境、食品、人）農(nóng)業(yè)現(xiàn)狀我們生活在一個人口爆炸的時代，對糧食的需求也就不可避免的會很大。WHO預測，2025年全球人口將超過80億，到2050年將變成100億。同時，人均可耕種的土

2、地將從1960年的1.5公頃減少到2015年的0.2公頃。農(nóng)用化學品使用必不可少病蟲草害使得作物減產(chǎn)，其中雜草為1525，病害1550，蟲害2035。如果農(nóng)民停止使用農(nóng)藥，水果將減產(chǎn)78，蔬菜將減產(chǎn)54，谷物將減產(chǎn)32。由于作物的病蟲草害防治不當，將使產(chǎn)量減少平均達30以上。 生物燃料 專家預測， 在未來50年100年現(xiàn)有的石油燃料將會耗盡。美國已經(jīng)頒布總統(tǒng)法令，要求2015年以前實現(xiàn)生物燃料消耗占到總?cè)剂舷牡?0。雖然有人聲稱美國沒有足夠的土地來實現(xiàn)此目標，然而2007年美國的玉米種植面積還是達到歷史新高。按照美國政府的計劃，2008年30的玉米將用于發(fā)展乙醇工業(yè)，在20022003年的水

3、平上增加了50。到2010年，美國的玉米種植面積將擴大至1億英畝（4200萬公頃）。對生物燃料的需求發(fā)展是如此之快，以致很多國家還沒來得及準備好應對由于日益減少的農(nóng)田資源而引發(fā)的利益沖突。農(nóng)民自然愿意種植價格更高的燃料作物而非糧食作物，就如在墨西哥，美國燃料作物需求的增加帶動了玉米價格的上升，使得墨西哥的農(nóng)民更愿意出口他們的農(nóng)產(chǎn)品而不愿意在本國市場上將其作為糧食產(chǎn)品出售。 美國用于生產(chǎn)生物燃料作物情況如何正確理解農(nóng)藥殘留？讓人們知道食品中含有哪些化學物質(zhì)，該怎樣檢測農(nóng)藥含量和哪種情況下是安全的食物中的天然化學物質(zhì)有時遠比合成化學品危險？ Pesticide residueAny specifi

4、ed substance in food, agricultural commodities or animal feed resulting from the use of pesticide.The term includes any derivatives of a pesticide, such as conversion products, metabolites, reaction products and impurities considered to be of toxicological significance. 注意： 殘留的定義很重要， 不同場合要求分析的殘留對象可能

5、不同！CCPRDetermine PrioritiesReview Residue Data and Propose MRLsReview Toxicology DataEstimate Short-term and Chronic Dietary ExposuresEstimate Values for the ADI and ARfDAre the Estimated Dietary Intakes less than the Toxicological Benchmarks?CCPRConsider Proposed MRLs in the context of the Risk Ass

6、essmentFAO Residue ExpertsWHO Toxicology ExpertsJMPR RISK ASSESSMENTMRL的設(shè)定模型TMDI的計算是根據(jù)推薦的MRL值， 假設(shè)人們每日攝入的食品中都含有MRL水平的農(nóng)藥殘留，然后根據(jù)人們的膳食結(jié)構(gòu)來計算每日理論攝入量，即：TMDI Fi x Mi 其中Fi 表示平均每天攝入某種食品的量， Mi表示相應食品的MRL。 由于TMDI比較保守，因此WHO在1997年頒布了比較準確的EDI方法來估算人們的每日攝入量， EDI STMRi x Ei x Pi x Fi； 其中， STMRi 是殘留中濃度（STMR)。Ei某食品可食部分系

7、數(shù)，Pi為食品的加工系數(shù)，F(xiàn)i平均每天攝入某種食品的量。 優(yōu)良農(nóng)業(yè)措施（GAP）GAP應包括最佳使用量范圍、使用次數(shù)、每次使用的間隔期、安全采收期（PHI) 等。在該過程中，在首先考慮有效性的同時，應同時考慮盡量少施藥以減少農(nóng)藥的浪費和對環(huán)境的污染。MRL的主要作用 作為一種法定限量其主要作用是用以檢驗農(nóng)作物和食品生產(chǎn)過程是否嚴格執(zhí)行優(yōu)良農(nóng)業(yè)措施，幫助經(jīng)農(nóng)藥處理的農(nóng)產(chǎn)品在全球進行進出口貿(mào)易。MRL可保障食品安全？ MRL不是一個安全限量，農(nóng)業(yè)殘留超標的食品并不一定必須意味著對人體健康有影響。ADI是通過一系列毒理學研究而得到的一個安全限量，也就是說估計攝入量是否超過由ADI計算的TMDI理論攝

8、入量，才是判斷食品是否安全的標準。ADI反應了長期評價；最近還關(guān)注急性毒性方面的安全。 Results from the 1989, 1990 & 1991 FDA Total Diet Studies作物組合? 解決全面制定MRL的鑰匙？將在生物學，分類學，和園藝學特征上接近的作物組合起來，再選出代表作物。用代表作物的殘留數(shù)據(jù)來建立全組的殘留指標。Preparation of the analytical sample美國、日本、德國等多殘留分析方法日本殘留農(nóng)薬多成分一斉分析法展望LC/MS(SCAN)分析ENVI-Carb SPE乙腈提取ODS SPE塩析PSA SPEGC/MS(SCAN

9、)分析日本positive list 對應分析方法(約350農(nóng)薬)(約100農(nóng)薬)日本農(nóng)藥多殘留分析方法丙酮提取ODS凈化CH2Cl2轉(zhuǎn)移QMA+NH2凈化(SAX+PSA)GC/MS分析Luke法（FDA)乙腈提取ODS凈化塩析HPLC分析CDFA法GC/MS分析乙腈提取ODS凈化塩析秋山法PSA凈化NH2凈化GC-FPDGC-ELCDDispersive-SPE Why use an SPE apparatus for “chemical filtration?” Dispersive-SPE involves the mixing of the sorbent with the extr

10、act in a mini-centrifuge tube to retain matrix interferants, but not analytes.樣品前處理流程圖15g 粉碎 樣品+15ml 0.1% 醋酸/乙腈溶液+6.0g 無水硫酸鎂+1.5g 醋酸鈉取2.0ml 上清液至盛有300mg無水硫酸鎂100mg PSA和100mgC18離心試管中5000轉(zhuǎn)/分鐘離心1.0分鐘劇烈振蕩1.0分鐘旋渦混合器混合1.0分鐘5000轉(zhuǎn)/分鐘離心1.0分鐘取1.0ml 上清液至進樣器小瓶中，用氮吹儀在室溫條件下濃縮至小于0.8ml分別加入100ul TPP和100ul 分析保護劑混合溶液，用乙

11、睛定容至1.0ml技術(shù)特點：1. 無水硫酸鎂代替無水硫 酸鈉和氯化鈉脫水2. 分散固相萃取代替固相 萃取柱凈化3. 添加分析保護劑提高檢 測靈敏度保護劑D05(20mg/ ml) +B06(1.0mg/ ml) 對峰形和峰面積的影響 A: 基質(zhì)提取+分析保護劑 B: 基質(zhì)提取 C: 標樣+分析保護劑 D: 標樣(黃瓜,蘋果,韭菜混合基質(zhì))ABCDABCDMethamidophos m/z 94Acephate m/z 136Omethoate m/z 156Carbofuran m/z 164Mevinphos m/z 127Fenvalerate m/z 167QUECHERS 方法參考論文

12、色譜法測定農(nóng)產(chǎn)品中農(nóng)藥殘留時的基質(zhì)效應，農(nóng)藥學學報，2005年第4期氣質(zhì)聯(lián)機分析蔬菜中農(nóng)藥多殘留及基質(zhì)效應的補償，高等學?；瘜W學報，2006年第2期應用分析保護劑補償基質(zhì)效應與氣相色譜質(zhì)譜快速檢測果蔬中農(nóng)藥多殘留，分析測試學報，2006年第3期。California Department of Food and Agriculture Multi-Residue MethodThree PDP laboratories (California, Texas, and Washington) use MRM CDFA method： Fifty grams of homogenized samp

13、le is extracted withacetonitrile. Sodium chloride is added to the sample/acetonitrile mixture and agitated. All samples aresubjected to an initial C-18 solid phase extraction (SPE) clean-up step. The acetonitrile is evaporatedfrom one aliquot of each sample, which is reconstituted with acetone for o

14、rganophosphate analysis.Additional aliquots are cleaned up using florisil or aminopropyl SPE columns depending on the type ofcompound or the instrumental analysis to be conducted.Florida Department of Agriculture and Consumer Services Multi-Residue MethodThe Florida laboratory uses an acetonitrile-b

15、ased extraction with aminopropyl solid phase extraction (SPE) cleanup procedures to extract low-fat, high moisture commodities (i.e., fruit and vegetables) for pesticide screening. Fifty grams of homogenized sample is shaken with 50 mL acetonitrile. Sodium chloride is added to the sample/acetonitril

16、e mixture and agitated. Samples for gas chromatography (GC) analyses are cleaned up using an aminopropyl SPE cartridge, eluted with 25% toluene in acetone and reconstituted in iso-octane for instrumental analysis. Samples for liquid chromatography (LC) analyses are cleaned up using an aminopropyl SP

17、E cartridge, eluted with 3% methanol in acetone and reconstituted in methanol for instrumental analysis.major pesticide classes (e.g., organophosphates, organochlorines, carbamates, triazines, triazoles, pyrethroids, neonicotinyls, strobilurinsModified Luke Multi-Residue MethodOne PDP laboratory (Oh

18、io) uses a modified version of the Luke method：100 g of homogenized sample is extracted with acetone. An aliquot of the filtrate is partitioned with methylene chloride and petroleum ether. Two additional methylene chloride partitions are performed. The final 10 mL extract is reconstituted in acetone

19、 and split for separate analysis using various instruments.New York Modified SPE MethodThe New York Department of Agriculture and Markets laboratory uses a method based on the Agriculture and Agri-Food Canada solid phase extraction (SPE) method4 with some improvements based on the Luke extraction.50

20、g of sample is extracted with 5% ethanol in acetonitrile and 15 grams sodium chloride. The organic layer is dried with sodium sulfate and cleaned up using a combination of SPE cartridges: graphitized carbon (Envi-carb), strong anion exchange (SAX), and weak anion exchange (PSA). Compounds are eluted

21、 with 3:1 mixture of acetonitrile:toluene, reconstituted in the appropriate solvent, and analyzed by the following instrument/detector combinations: gas chromatography (GC)/flame photometric detector (FPD), GC/tandem mass spectrometry (GC/MS/MS), and high performance liquid chromatography (HPLC)/tan

22、dem mass spectrometry (LC/MS/MS).NPDFPD/PFPDMSD / ITDGaschromatographic AnalysisECDFraction IECDFraction IIECDFraction IIIClean-up on silica gelGPC clean upDrying, Evaporation, FiltrationLiquid/liquid partition with NaCl and Cyclohexane/EthylacetateExtraction with Acetone/WaterPRINCIPLE OF THE PRINC

23、IPLE OF THE “S19S19”MULTI RESIDUE METHOD MULTI RESIDUE METHOD NPDFPD/PFPDMSD / ITDGC AnalysisECDFraction IECDFraction IIECDFraction IIICleanup on silica gelGPC cleanupDry, Evaporate, FilterLiquid/liquid partition with NaCl and Cyclohexane/EtAcExtraction with Acetone/WaterPRINCIPLE OF THE PRINCIPLE O

24、F THE “S19S19”MULTI RESIDUE METHOD MULTI RESIDUE METHOD Slide from Gnter Lach of EurofinsFruits and vegetablesExtraction: Module E 9 Sample weight: 50 gExtraction with ASEGPC with TACSDetection ECD and NPDGC column: DB 5, length 30 m, i. D. 0,25 mm, film thickness 0,25 m Relative frequency distribut

25、ion of chlorpyrifos residues in apple samples36Conclusions The typical sampling uncertainty For medium size crops with sample size 10CVS,med = 28% For small crops with sample size 10CVS,small CVS,medFor large crops with sample size 5CVS,large is about 40% (temporary estimate) 37SamplingConsideration

26、 of the spread and variability of the residues helps to avoid misleading interpretations of small differences or drawing definite conclusions from a single calculated value. Careful attention to the details of sampling is essential if representative samples are to be obtained. Preparation of the ana

27、lytical sampleThe preparation of the analytical sample from which the analytical portion is withdrawn consists of two distinct procedures: sample preparation and sample processing.Sample preparation and processing must be carried out according to the aim of analysis.For instance, to provide data for

28、 the estimation of maximum residue limits, the Codex Alimentarius Commission specifies the portion of commodities to which Maximum Residue Limits (MRLs) apply. Sample preparationAccording to the Codex Alimentarius, sample preparation is the procedure used, if required, to convert the laboratory samp

29、le into the analytical sample by removal of parts (soil, stones, bones) not to be analysed.For instance:for lettuce: remove dirty and withered leavesfor oranges: remove leaves and stemsThe method of sample preparation may be the source of substantial systematic and random errors, which cannot be est

30、imated.However, it is possible to maintain this uncertainty component to a minimum level by strictly following sample preparation procedures, as described in the relevant standard operating procedures (SOPs) attached to the study plan. HomogenisationHomogenisation implies a mechanical size reduction

31、 with suitable equipment, like mincing machines, choppers, blenders, or grinders.If available, it is advisable to add dry-ice during homogenisation in order to improve the efficiency of sample processing, while reducing the risk of thermal degradation of pesticides. The homogenisation in the presenc

32、e of dry-ice produces a fine powder, which is easy to mix The sampling constant (KS) is the weight (W)of a single increment that must be withdrawn from a well-mixed material to hold the relative sampling (withdrawing and processing) uncertainty to 1% with 68% level of confidence取樣系數(shù)與不確定度Apple sample

33、sConfidence intervals The uncertainty of the analytical result comprises:The expression can be modified to incorporate any additional stagesUncertainty components:The efficiency and uncertainty of sample processing can be tested utilising the “sampling constant”, KSStability of residues Systematic s

34、tudies revealed substantial decrease (40-70%) of the concentration of several analytes (captan, captafol, folpet, chorothalonil, dichlofluanid) during processing at ambient T The extent of loss was dependent on both pesticide and commodity, and even among varietiesRecent study on stability of residu

35、es during cryogenic processing : 94 of 106 pesticides tested were stablelosses of several pesticides -bitertanol (95%), heptenophos (50%), isophenphos (40%), and tolylfluanid (48%)- reported at Tamb did not occur during cryogenic processing Still needs to be demonstrated that has no detrimental effe

36、ct on some pesticidesExtraction of pesticide residues and metabolites農(nóng)藥殘留分析中包括分離步驟， 即把分析目標物質(zhì)和干擾物質(zhì)分離的過程。 多數(shù)操作中， 第一步是提取過程。 different extraction liquids (solvents), extraction equipment and the importance of determining extraction efficiency and method recovery. Filtration and concentration of the samp

37、le extracts標記同位素在分析中的應用測試的基本步驟： preparation of the samples, homogenization, extraction, clean-up and injection into the adequate equipment for chromatographic separation and measurement.測定各個步驟的不確定度是GLP要求。 The uncertainty of each step of the analytical procedure can also be determined.Radiolabeled pe

38、sticides can help:研究回收率和提取效率 check the analytical recovery of residues,研究殘留物質(zhì)的穩(wěn)定性 check the stability of residues, and研究單元操作的不確定度 estimate the uncertainty of processes. Estimation of the uncertainty of processes目前還沒有一種分析方法， 可以采用一種提取方法， 一種凈化過程， 一種檢測手段一次分析所有基質(zhì)類型中的所有可能農(nóng)藥殘留。 這樣的方法是理想的， 也是不存在的。-Pan C.檢測與

39、確證手段GCMSLCMSMSMSSPR？ 快速、無損incorrectly identified or false positive resultsIn pesticide analysis, analytes in samples are often incorrectly identified or false positive results are reported. This may be due to:the presence of natural co-extractives or other contaminants, which mimic the characteristi

40、cs of some pesticides, or the lack of proper confirmation of the analyte identity.Retention time GLC analysisMass spectrometerHPLC and LC-MS/MS analysisDerivatisationTLC analysisDetermination criteria ELISADerivatisationEU對質(zhì)譜確證方法的要求全掃描：當用單級質(zhì)譜記錄全掃描圖譜時，至少要有四種離子的相對豐度大于等于基峰的10%。如果分子離子峰在參考圖譜中的相對豐度10%，則必須

41、包括在內(nèi)。 選擇離子監(jiān)測質(zhì)譜方法時要求: 96/23/EC指令附錄 I A組 4識別點 96/23/EC指令附錄 I B組 3識別點 對質(zhì)譜確證方法的要求MS技術(shù)每種離子的識別點低分辨質(zhì)譜1.0LR-MSn母離子1.0LR-MSn子離子1.5HRMS2.0HR-MSn母離子2.0HR-MSn子離子2.5質(zhì)量碎片類型和識別點的關(guān)系 對質(zhì)譜確證方法的要求 對于每個診斷離子，信噪比 3:1 測定至少一對離子豐度比相對離子豐度最大容許偏差 不能超過下表相對豐度（%基峰）EI-GC-MS（相對）CI-GC-MS、GC-MSnLC-MS、LC-MSn（相對）50%10%20%20%到50%15%25%10

42、%到20%20%30%10%50%50%對其他確證方法的要求僅對于 B 組物質(zhì)適用 （限用物質(zhì)）對于 A 組物質(zhì)有1個識別點 LC 或 GC 帶 IR檢測器LC帶全掃描 UV/VIS檢測器LC帶熒光檢測器2-D HPTLC帶全掃描 UV/VIS對其他確證方法的要求僅對于 B 組物質(zhì)適用 （限用物質(zhì)）對于 A 組物質(zhì)有1個識別點GC-ECD (用二根不同極性的色譜柱)LC-酶免測定 (2 種不同的色譜分離或第二種測定方法)LC-帶 UV/VIS 單波長 (2 種不同的色譜分離或第二種測定方法)Quality control for pesticide residue analysis What

43、is it?QC proceduresSample handlingSamplingSample transportationSamples at laboratorySample storagePesticide standardsStock and working standardsExtractionContaminationInterferenceControl instrumental analysisDead volume peak shape testSolvent peak shape testSystem suitability tests (SSTs)Analytical

44、calibrationMatrix effectsAnalytical methods and recovery determinationsSpike or Recovery samplesDuplicate analysisRecovery determinationBlind samplesProficiency testingConfirmation of resultsReporting of resultsTools for IQCQC activitiesSample storageDuring the storage in the laboratory, the samples

45、 must be protected against physical, chemical and mechanical influence, which can lead to changes in the samples and consequently to the analytes.Examples are:air oxidation or other chemical influence,evaporation of water and other volatile components,enzymatic degradation,microbial activity,polluti

46、on from the container. ExtractionAnalytical portions should be disintegrated thoroughly before or during extraction to maximise extraction efficiency. Over-heating during extraction must be avoided to minimise solvent or pesticide losses. Temperature, pH, etc., must be controlled if they are known t

47、o affect extraction efficiency and/or pesticide stability.System suitability tests (SSTs)To verify the applicability of the chromatographic system to sample analysis or to reveal degradation in system performance, a system suitability test mixture (SST) is periodically injected in the chromatographi

48、c system. The SST mixture contains a series of compounds that have a range of functional groups and selected characteristics. For multiresidue analysis of pesticides the test mixture should contain pesticides covering the retention time range of the compounds that are analysed. GC SST mixture for NP

49、D, FPD, PFPD and MSEPTCNeff, Sensitivity, Selectivity S to C (for PFPD) PropoxurInertness Tributylphosphate (TBP)Sensitivity, Selectivity P to CDimethoateAsymmetryPirimicarbSensitivity, Selectivity N to CChlorpyrifos methylResolution with parathion methylParathion methylResolution with chlorpyrifos

50、methylCarbarylInertnessChlorpyrifos ethylReference for RRTsQuinalphosResolution with methidathionMethidathionResolution with quinalphosHydrocarbon SelectivityPhosaloneNeff, RRTNPD SSTOptimized NPD SST NPD SST after injection of 40 sample extractChange of peak shape and RRT123Chlorpyrifos EthylAsym 1

51、.21 1.22 1.60123Dimethoate RRT Asym 0.690 1.56 0.691 1.82 0.708 3.33123EPTC RRT Asym 0.403 0.92 0.403 1.01 0.402 0.951- peak shape after the system maintenance;2 after injection of 20 sample extracts;3 after injection of 40 sample extractsVariation of P selectivity to carbon and MDQ of P in TBP duri

52、ng 3 year periodBead replaced, optimisedColumn cut, liner replacedCalibration: Matrix enhancement effectTools for IQCControl chartsControl charting is a suitable way of documenting IQC in analytical laboratories: it is used to monitor the stability of the measurement process and the analytical perfo

53、rmance over time. A control chart is a graphical plot of test results with respect to time or sequence of measurements, with established limits (upper and lower action/warning limits). A procedure is in statistical control when results consistently fall within the established control limits.Control

54、charts can be of different types:Charts for mean valuesCharts for individual valuesCharts for rangesInternal quality controlRe-analyse regularly analytical portions of positive samples. For 2 replicatesFor 3 replicatesSummary of most commonly used QC activitiesAnalysis of reagent blankAnalysis of blank samplesDuplicate analysisSystem suitabili