High Throughput Analytical Techniques for the Determination and Confirmation of Residues of 653 Multiclass Pesticides and Chemical Pollutants in Tea by GC/MS, GC/MS/MS, and LC/MS/MS: Collaborative Study, First Action 2014.09.

Thirty laboratories from fom North and South America, Europe, and Asia participated in this AOAC collaborative study (15 from China; five from Germany; two each from Italy and the United States; and one each from the Republic of Korea, Canada, Spain, Japan, Belgium, and India). Participants represented government regulatory, commercial testing, university, research institute, and private laboratories. The single-laboratory validated (SLV) tea method was evaluated in the collaborative study to determine the recovery and reproducibility of the method under multilaboratory conditions. Since there were no restrictions regarding the type of analytical instrumentation to use for the analyses, laboratories used a combination of equipment that included GC/MS, GC/MS/MS, and LC/MS/MS instruments from 22 different manufacturers, 21 brands of GC and LC columns, 13 different GC temperature programming profiles, 11 LC gradient elution programs, and six different vendor manufactured SPE cartridges. Even though all the analytical performance parameters for all the 653 compounds had been determined in the SLV study, guidance was obtained from an expert review panel of the AOAC Method-Centric Committee on Pesticide Residues to conduct the multilaboratory collaborative study based on 20 selected compounds that can be analyzed by GC/MS and 20 compounds that can be analyzed by LC/MS/MS. Altogether, 560 samples covering the 40 selected pesticides were analyzed in the study. These samples included green tea and oolong tea samples fortified typically at the European Union maximum residue limit for regulatory guidance and compliance, aged tea samples incurred with 20 pesticides, and green tea and oolong tea samples incurred with five pesticides. The analysis of the 560 samples generated a total of 82 459 test results by the 30 participating laboratories. One laboratory failed to meet the proficiency requirements in the precollaborative study. Therefore, its data submitted for the collaborative study were excluded from further analysis and interpretation. The results presented are therefore the 6638 analytical results obtained from the 29 remaining laboratories, which included 1977 results generated by GC/MS, 1704 results by GC/MS/MS, and 2957 results by LC/MS/MS. It was determined after application of the Grubbs and Dixon tests for outliers to the data sets that there were 65 outlier results from the 1977 GC/MS results (3.3%), 65 outlier results from the 1704 GC/MS/MS results (3.8%), and 57 outlier results out of 2957 LC/MS/MS results (1.9%), representing 0.98, 0.98, and 0.86%, respectively, of the 6638 results generated in the study. Analysis with the AOAC statistical software package also confirmed that the method is rugged, and average recovery, average concentration, RSDr, RSDR, and HorRat values all meet recovery and reproducibility criteria for use in multiple laboratories. The Study Director is recommending this method for adoption as an AOAC First Action Official MethodSM.

High-throughput GC/MS and HPLC/MS/MS techniques for the multiclass, multiresidue determination of 653 pesticides and chemical pollutants in tea.

An efficient and sensitive method has been established for simultaneous determination of 653 pesticides in teas by GC/MS and HPLC/MS/MS. The method involved extraction with acetonitrile followed by cleanup using Cleanert-TPT SPE and subsequent identification and quantitation of 490 pesticides by GC/MS and 448 pesticides by HPLC/MS/ MS. The LODs for pesticides determined by GC/MS were between 1.0 and 500 microg/kg, and those determined by HPLC/MS/MS were between 0.03 and 4820 microg/kg. At the low fortification levels of 0.01-100 microg/kg, the average recoveries of 94% of the pesticides determined by GC/MS were between 60 and 120%, 77% of which had an RSD below 20%. For 91% of pesticides determined by HPLC/MS/MS, the average recoveries were between 60 and 120%, 76% of which had an RSD below 20%. The paper also reports a novel SPE column, Cleanert TPT, which comprised graphitized carbon black (PestiCarb), polyamine silica, and amide polystyrene for purifying the tea samples. The results indicated good repeatiblity and reproducibility.

Analysis method study on 839 pesticide and chemical contaminant multiresidues in animal muscles by gel permeation chromatography cleanup, GC/MS, and LC/MS/MS.

The paper reports the study of gel permeation chromatography (GPC), gas chromatography/mass spectrometry (GS/MS), and column chromatography tandem MS (LC/MS/MS) for 839 pesticides and chemical contaminants, through which a GPC data bank has been established for 744 pesticides, a GC/MS data bank for 541 pesticides, and an LC/MS/MS data bank for 464 pesticides. On the basis of this study, a new method for quantitative determination of 587 pesticide residues in animal muscles such as beef, mutton, pork, chicken, and rabbit has been established using GPC cleanup followed by GC/MS and LC/MS/MS. In the method, 10 g animal samples were mixed with 20 g sodium sulfate and extracted twice with 35 mL cyclohexane-ethyl acetate (1 + 1) by blender homogenization followed by centrifugation, filtration, and concentration. An equivalent of 5 g sample was injected into a 400 x 25 mm S-X3 GPC column, with cyclohexane-ethyl acetate (1 + 1) as the mobile phase at a flow rate of 5 mL/min. The 22-40 min fraction was collected for subsequent analysis. For the 478 pesticides determined by GC/MS, the portions collected from GPC were concentrated to 0.5 mL and exchanged twice with 5 mL hexane. For the 379 pesticides determined by LC/MS/MS, the portions collected from GPC were dissolved with acetonitrile-water (60 + 40) after taking the extract to dryness with nitrogen gas. At the limit of quantification (LOQ) and 10 LOQ fortification levels of 0.1-16 000 microg/kg, recoveries were within 40-130%, among which 563 pesticide recoveries were between 60 and 130%, accounting for 96% of the compounds; 24 analytes were recovered between 40 and 60%, accounting for 4% of the compounds. The relative standard deviation was below 30% for all 587 pesticides. The limits of detection for the method were 0.1-1600 microg/kg. In comparison with GC/MS, LC/MS/MS increased the detection sensitivity 2-1000 times for 236 pesticides.

Validation study on 660 pesticide residues in animal tissues by gel permeation chromatography cleanup/gas chromatography-mass spectrometry and liquid chromatography-tandem mass spectrometry.

A new method using gel permeation chromatography (GPC) cleanup followed by gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-tandem mass spectrometry (LC-MS-MS) has been established for quantitative determination of 437 pesticide residues in animal tissues such as beef, mutton, pork, chicken, and rabbit. Based on an appraisal of the characteristics of both GC-MS and LC-MS-MS, validation experiments were conducted for 660 pesticides. In the method, 10 g animal samples were mixed with 20 g sodium sulfate and extracted with 35 mL of cyclohexane+ethyl acetate (1+1) twice by blender homogenization, centrifugation, and filtration. Evaporation was conducted and an equivalent of 5 g sample was injected into a 400 mm x 25 mm S-X3 GPC column, with cyclohexane+ethyl acetate (1+1) as the mobile phase at a flow rate of 5 mL/min. The 22-40 min fraction was collected for subsequent analysis. For the 368 pesticides determined by GC-MS, the portions collected from GPC were concentrated to 0.5 mL and exchanged with 5 mL hexane twice. For the 69 pesticides by LC-MS-MS, the portions collected from GPC were dissolved with acetonitrile+water (60+40) after taking the extract to dryness with nitrogen gas. In the linear range of each pesticide, the correlation coefficient was r > or = 0.98, exceptions being dinobuton, linuron, and fenamiphos sulfoxide. At the low, medium and high three fortification levels of 0.2-4800 microg/kg, recoveries fell within 40-120%, among which 417 pesticides recoveries between 60% and 120%, accounting for 95%, 20 analytes between 40% and 60%, accounting for 5%. The relative standard deviation was below 28% for all 437 pesticides. The limits of detection for the method were 0.2-600 microg/kg, depending on each pesticide.

Determination of residues of 446 pesticides in fruits and vegetables by three-cartridge solid-phase extraction-gas chromatography-mass spectrometry and liquid chromatography-tandem mass spectrometry.

A method was developed for determination of residues of 446 pesticides in fruits and vegetables through the use of cleanup by a 3-cartridge solid-phase extraction-gas chromatography/ mass spectrometry (GC/MS) and liquid chromatography/tandem mass spectrometry (LC/MS/MS). Fruit and vegetable samples (20 g) were extracted with 40 mL acetonitrile, salted out, and centrifuged. Half of the supernatant was passed into an Envi-18 cartridge, eluted with acetonitrile, and cleaned up with Envi-Carb and aminopropyl Sep-Pak cartridges in series after concentration of the eluates. Pesticides were eluted with acetonitrile-toluene (3 + 1, v/v), and eluates were concentrated to 0.5 mL and then added into internal standards after solvent exchange with 2 mL hexane and used for determination of 383 pesticides by GC/MS. The other half of the supernatant was concentrated to 1 mL and cleaned up with Envi-Carb and aminopropyl Sep-Pak cartridges in series. Pesticides were eluted with acetonitrile-toluene (3 + 1, v/v), and the eluates were concentrated to 0.5 mL, dried with nitrogen gas, diluted to 1.0 mL with acetonitrile-water (3 + 2, v/v), and used for determination of 63 pesticides by LC/MS/MS. The limit of detection for the method was 0.2-600 ng/g depending on the individual pesticide. In the method, fortification recovery tests at high, medium, and low levels were conducted on 6 varieties of fruits and vegetables, i.e., apples, oranges, grapes, cabbage, tomatoes, and celery, with average recoveries falling within the range of 55.0-133.8% for 446 pesticides, among which average recoveries between 60.0-120.0% accounted for 99% of the results. The relative standard deviation was between 2.1-39.1%, of which a relative standard deviation of 2.1-25.0% made up 96% of the results. Experiments proved that the method was applicable for determination of residues of 446 pesticides in fruit and vegetables.

Simultaneous determination of 16 sulfonamides in honey by liquid chromatography/tandem mass spectrometry.

A method is described for the determination of 16 sulfonamides in honey. Samples are dissolved in phosphoric acid solution (pH2), cleaned up with 2 solid-phase extraction (SPE) cartridges, an aromatic sulfonic cation-exchange cartridge and an Oasis HLB SPE cartridge, and analyzed both qualitatively and quantitatively by liquid chromatography/tandem mass spectrometry (LC/MS/MS) under the selected conditions. Without exception, calibration curves were linear (r = > 0.995), when sulfamethizole was between 1.0 and 25.0 microg/kg; sulfacetamide, sulfapyridine, sulfadiazine, sulfachloropyridazine, sulfamethoxazole, sulfamerazine, sulfisoxazole, sulfamonomethoxine, and sulfadoxine were between 2.0 and 50.0 microg/kg; sulfamethoxypyridazine, sulfadimethoxine, and sulfathiazole were between 4.0 and 100.0 microg/kg; sulfamethazine and sulfameter were between 8.0 and 200.0 microg/kg; and sulfaphenazole was between 12.0 and 300.0 microg/kg. Average recoveries at 4 fortification levels in the range of 1.0-300 microg/kg in honey were 70.9-102.5%, and relative standard deviations were 2.02-11.52%. The limits of quantitation for the 16 sulfonamides were between 1.0 and 12.0 microg/kg, with the LC/MS/MS method.

Evaluation of analyte stability and method ruggedness in the determination of streptomycin residues in honey by liquid chromatography with post-column derivatization.

This study demonstrated that streptomycin in honey is quite stable, and the results showed no obvious differences for 3 samples containing incurred analyte during continuous testing for 4 months. Fifteen laboratories evaluated method performance at 4 fortification levels ranging from 0.010 to 0.100 mg/kg; the recoveries ranged from 73.7 to 78.5%, the reproducibility relative standard deviations ranged from 5.76 to 15.85%, and the repeatability relative standard deviations ranged from 1.64 to 3.80%. In 1999-2002, the method was used to determine streptomycin residues in 5106 lots of honey samples from >20 provinces all over China. All of the honey samples were found to be in conformity with the requirements of customs clearance for exports to Europe, the United States, and Japan. The continuous 4-year quality analysis also found that C18 solid-phase extraction cartridges should be standardized to ensure that the analytical results are accurate when different lots of cartridges are used.

Determination of clopidol residues in chicken tissues by liquid chromatography: collaborative study.

Eighteen laboratories participated in a collaborative study on the determination of clopidol residues in chicken muscle tissues by liquid chromatography. Of these, results from 16 laboratories which rigorously followed the method were subjected to statistical analysis. The method performance was assessed by all participants using 14 samples of chicken muscle fortified at concentrations ranging from 0.1 to 5.0 mg/kg. In addition, 9 participants each reported results for 6 clopidol-incurred samples in chicken muscle. Test portions were extracted with acetonitrile, and the extracts were purified with alumina and anion exchange resin solid-phase extraction cartridges in sequence. Clopidol was separated by reversed-phase liquid chromatography and quantified at 270 nm. Average recoveries ranged from 81.8 to 85.4%, reproducibility relative standard deviation (RSDR) ranged from 11.9 to 22.6%, and repeatability relative standard deviation (RSDr) ranged from 9.9 to 15.1%. For clopidol-incurred samples at concentrations of 0.100-0.687 mg/kg, the mean determination value range was 0.099-0.659 mg/kg; RSDR was 12.6-19.8%, RSDr was 3.1-8.5%; and HORRAT values were 0.7-1.1. The accuracy and precision of the method are in conformity with the requirements specified by AOAC INTERNATIONAL. The method was adopted Official First Action in April 2003.

Simultaneous determination of 405 pesticide residues in grain by accelerated solvent extraction then gas chromatography-mass spectrometry or liquid chromatography-tandem mass spectrometry.

A new method has been established for simultaneous determination of 405 pesticide residues in grain, using accelerated solvent extraction (ASE), solid-phase extraction (SPE), and GC-MS and LC-MS-MS. The method was based on appraisal of the GC-MS and LC-MS-MS characteristics of 660 pesticides, their efficiency of extraction from grain, and their purification. Samples of grain (10 g) were mixed with Celite 545 (10 g) and the mixture was placed in a 34-mL cell of an accelerated solvent extractor and extracted with acetonitrile in the static state for 3 min with two cycles at 1,500 psig and at 80 degrees C. For the 362 pesticides determined by GC-MS, half of the extracts were cleaned with an Envi-18 cartridge and then further cleaned with Envi-Carb and Sep-Pak NH2 cartridges in series. The pesticides were eluted with acetonitrile-toluene, 3:1, and the eluates were concentrated and used for analysis after being exchanged with hexane twice. For the 43 pesticides determined by LC-MS-MS the other half of the extracts were cleaned with Sep-Pak Alumina N cartridge and further cleaned with Envi-Carb and Sep-Pak NH2 cartridges. Pesticides were eluted with acetonitrile-toluene, 3:1. After evaporation to dryness the eluates were diluted with acetonitrile-water, 3:2, and used for analysis. In the linear range of each pesticide the linear correlation coefficient r was equal to or greater than 0.956 and 94% of linear correlation coefficients were greater than 0.990. At low, medium, and high fortification levels, at the limit of detection (LOD), twice the LOD and ten times LOD, respectively, recoveries ranged from 42 to 132%; for 382 pesticides, or 94.32%, recovery was from 60 to 120%. The relative standard deviation (RSD) was always below 38% and was below 30% for 391 pesticides, or 96.54%. The LOD was 0.0005-0.3000 mg kg(-1). The proposed method is suitable for determination of 405 pesticide residues in grain such as maize, wheat, oat, rice, and barley, etc.

Liquid chromatography-fluorescence detection for simultaneous analysis of sulfonamide residues in honey.

A rapid, accurate LC analytical method has been developed for determination of eight sulfonamides (sulfacetamide, sulfapyridine, sulfamerazine, sulfamethoxypyridazine, sulfameter, sulfachloropyridazine, sulfamethoxazole and sulfadimethoxine) in honey. The sample was dissolved in phosphoric acid solution (pH 2). After filtration, the sample solution was cleaned by use of two solid-phase extraction (SPE) cartridges-an aromatic sulfonic cation-exchange cartridge and an Oasis HLB cartridge. The eight sulfonamides were then derivatized with fluorescamine and the derivatives were determined by LC with fluorescence detection at excitation and emission wavelengths of 405 and 495 nm, respectively. Average recoveries at three fortification levels in the range 0.02-0.50 mg kg(-1) in twelve different kinds of honey were 73.5-94.1% with coefficients of variation of 4.35-16.60%. The limit of detection (LOD) was 0.002 mg kg(-1) for sulfacetamide, sulfapyridine, sulfamerazine, and sulfamethoxypyridazine; that for sulfameter, sulfachloropyridazine, sulfamethoxazole and sulfadimethoxine was 0.005 mg kg(-1). The limit of quantitation (LOQ) was 0.005 mg kg(-1) for sulfacetamide, sulfapyridine, sulfamerazine, and sulfamethoxypyridazine; that for sulfameter, sulfachloropyridazine, sulfamethoxazole, and sulfadimethoxine was 0.010 mg kg(-1). The method is suitable for determination of multiresidue sulfonamides in the various kinds of honey.