Large pesticide multiresidue screening method by liquid chromatography-Orbitrap mass spectrometry in full scan mode applied to fruit and vegetables

The present work is focused on evaluating the main operational parameters for multiresidue screening of an Orbitrap mass spectrometer for pesticide residue analysis in fruits and vegetables. Operational parameters such as resolution, software for the automatic detection, mass tolerance and retention time extraction window, along with the analytical performance, were evaluated in an updated UHPLC-Orbitrap-mass spectrometer working in full scan mode. The evaluation was performed using QuEChERS extracts of tomato, pepper, orange and green tea. The extracts were spiked with 170 selected pesticides at four concentration levels (10 mu g/kg, 50 mu g/kg, 100 mu g/kg and 500 mu g/kg). Extracts were diluted 5 fold before injection. Three different resolution settings (17,500, 35,000 and 70,000) were evaluated at various concentration levels. At 10 mu g/kg, using a resolution of 17,500 and 5 ppm of mass tolerance, the detected pesticide rates were from 91% in tomato, to 83% in green tea. These percentages increased at higher resolution values. A resolution of 70,000 was adequate for such analysis even when a small percentage of false detect at low concentration was obtained. The rates of detected compounds increased and were from 98% in tomato to 88% in green tea. Mass tolerance of 5 ppm was the most adequate for screening purposes. The observed false negative detects were mainly a consequence of a lack of compound sensitivity exacerbated by ion suppression effects in the experimental conditions applied. With reporting limits of 10 mu g/kg, reproducibility improved with resolution levels of 35,000 or higher. Linearity was investigated in the 2-100 ng/mL (equivalent to 10-500 mu g/kg in the sample) range. Particularly good automatic screening effectiveness was obtained using the selected settings in the analysis of real samples where no false negatives detects and 5% of false positives detects were obtained. (C) 2014 Elsevier B.V. All rights reserved.