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
Guo-Fang Pang,Chun-Lin Fan,Yan-Zhong Cao,Fang Yan,Yan Li,Jian Kang,Hui Chen,Qiao-Ying Chang
DOI: https://doi.org/10.5740/jaoacint.15021
Abstract: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.