Method Validation Using Normal and Weighted Linear Regression Models for Quantification of Pesticides in Mango (Mangifera indica L.) Samples

A fast and efficient method was developed and validated for the determination of pesticides in mangos, which uses the QuEChERS citrate and gas chromatography-mass spectrometry (GC-MS) techniques. A detailed statistical analysis was performed to study the matrix effect. The calibration model using the method of weighted least squares is used in cases, where heteroscedasticity is observed. The matrix effect was observed for most studied compounds using analytical curves based on a spiked matrix. The limits of detection were 0.0025-0.01 mg kg(-1), and the limits of quantification (LOQ) were 0.008-0.03 mg kg(-1). The LOQ values were minor or equal to the established MRLs by major regulatory agencies in Brazil (ANVISA), the United States (US-EPA), and Europe (CE). The compounds showed acceptable recovery levels of 71-109% with a standard deviation less than 15%. The method was applied to determine pesticide residues in mango samples. For 12 samples, five compounds (chloroneb, propachlor, alpha-chlordane, chlorpyrifos, DCPA, chlorobenzilate, and trans-permethrin) were detected, with contents of 0.004-0.042 mg kg(-1). For chloroneb, propachlor, and alpha-chlordane, the found concentrations were above the maximum permitted residue limit, according to data from the European Commission.