Determinação de resíduos de agrotóxicos em tomate para fins de acreditação

Abstract

The optimization and validation of an analytical method for 51 pesticides determination in tomato, using UPLC-MS/MS, was developed in order to fulfill the requirement 5.4.5 (Method Validation) of the NBR ISO/IEC 17025:2005 and following the orientation document DOQ-CGCRE-008 (Orientation of analytical methods validation). This item is very important to a laboratory to be accredited by INMETRO (Brazilian accreditation body). The method validation was carried out following all the validation parameters, such as linearity of the analytical curves, limits of detection and quantification, matrix effects, accuracy and precision of the method. The studied compounds in this work were determined by UPLC-MS/MS in the positive ESI ionization mode. The stock solutions of the pesticides were prepared in organic solvent compatible with the compound solubility. These solutions were prepared by two different analysts in order to validate the preparation by comparing the areas of the injections (n=7) of each individual solution. After evaluation, 43 from 51 studied solutions showed results between the acceptable range of ± 10% demonstrating the satisfactory quality of solutions preparation. The evaluated pesticides were extracted from tomatoes applying the mini Luke method with a mixture 1:1:1 (v/v/v) of each acetone/petroleum ether/dichloromethane solvent as extraction solvent. After extraction, the organic solvent extract was evaporate and reconstituted in appropriate solvent (acidified methanol with 0.1% acetic acid) for UPLC-MS/MS analysis. The recovery experiments were done at 10, 20 and 50 μg kg-1 spike levels, by two analysts in different days, and 7 replicates (n=7) for each spike level plus the blank matrix (without pesticides). The limits of detection and quantification for the instrument and method were estimated, as well as the linearity of the analytical curves evaluated based on determination coefficient (r²), dynamic linear range, accuracy by recovery experiments (%), precision (RSD%) and matrix effect. From the 51 studied compounds, 46 showed recoveries between the range of 81 115% for all the evaluated spike levels. About precision, 92% of the compounds showed RSD% below 18.7% in the lowest spike level. The quantification limit of the method, for 82% from all evaluated compounds, was the lowest spike level studied (10 μg kg-1). The matrix effect observed was between ± 20% for all evaluated compounds, showing that there is not considerable suppression or enhancement in the analytes signal. For all evaluated pesticides the intermediate precision was below 20% of RSD showing the good repeatability of the method.