Recoveries of four representative organophosphorus pesticides from 18 plant products belonging to different botanical categories: implications for matrix effects.

A study of the matrix effect was performed in order to evaluate the influence of some matrices on the gas chromatographic responses of representative organophosphorus insecticides. In particular, three fortification levels and the maximum residual level of four organophosphorus pesticide standard solutions were added into 18 matrices (15 belonging to five specific botanical categories and three botanically unclassified). The recoveries of the examined pesticides were determined using a simple official multiresidue method of detection, without any additional clean-up step. Recoveries were estimated based on standards prepared in solvent. The recovery percentages of the most polar pesticides, especially methamidophos (20.2-288.4%), were much higher than those of non-polar ones (chlorpyrifos and methidathion 25.7-136.4 and 37.6-292.6%, respectively), except for quince. Pronounced matrix effects (>120%) were observed to the lowest fortification levels (maximum residual levels established by the European Union), of dimethoate and methamidophos (i.e. the most polar analytes) in the majority of product extracts. Furthermore, it was proved that there is no correlation between the classification of plant products, according to their botanical characteristics, and the recoveries of pesticides in the equivalent extracts. Indeed, there were significant differences in the recoveries of pesticides in extracts, derived from matrices belonging to the same botanical group (especially in the categories of pome fruits and citrus). Consequently, it was shown that, under the examined conditions, obtaining recovery data, derived from only one representative matrix, with the purpose to validate an official method in its botanical category is, possibly, an erroneous practice.