Abstract
The performance of two different analytical methodologies to investigate the presence of glyphosate (GLY) and aminomethylphosphonic acid (AMPA) residues in wine samples was evaluated. Transformation of compounds in their fluorene-9-methyloxycarbonyl derivatives permitted their separation under reversed-phase liquid chromatography with tandem mass spectrometry (LC-MS/MS) determination. Although the wine matrix severely impaired the efficiency of GLY derivatization, this drawback was solved using a molecularly imprinted sorbent for the previous, selective extraction of GLY and AMPA from wine. Alternatively, the use of a strong anionic exchange, polyvinyl alcohol-based LC column, turned to be the most effective alternative for direct determination of both compounds in diluted wine samples. The chromatographic behavior of this column and the magnitude of matrix effects observed during analysis of diluted wine samples were significantly affected by the composition of the mobile phase. Under final working conditions, this column permitted the separation of AMPA and the fungicide fosetyl (which shows common transitions in tandem MS/MS methods), it improved significantly the sample throughput versus extraction-derivatization-purification method, and it allowed the use of solvent-based calibration standards. Both analytical procedures provided similar limits of quantification (LOQs) for GLY (0.5-1.0 ng mL-1), while the multistep method was 8 times more sensitive to AMPA than the direct procedure. GLY residues stayed above method LOQs in 70% of the processed wines; however, concentrations measured in 95% of positive samples remained 100 times below the maximum residue limit (MRL) set for GLY in vinification grapes.