Development of a new sample preparation method based on liquid-liquid-liquid extraction combined with dispersive liquid-liquid microextraction and its application on unfiltered samples containing high content of solids.

A new sample preparation method based on liquid-liquid-liquid extraction combined with dispersive liquid-liquid microextraction followed by gas chromatography-flame ionization detection has been reported for the extraction/preconcentration and determination of trace levels of twelve pesticide residues from different samples with high content of solids without filtration. This method consists of a three-phase system including an aqueous phase (sample solution), acetonitrile, and hexane. The extraction mechanism is based on different affinities of the substances from the sample matrices towards each of the involved phase, which provides a high selectivity to the process. In other words, interfering hydrophobic compounds are transferred into hexane and will not be present in the final extract. Furthermore, ionic and polar compounds are retained in the aqueous phase. Therefore, only semi-polar compounds such as the studied pesticides are extracted into acetonitrile. In this method, a homogeneous solution of the aqueous phase and acetonitrile (a water-soluble extraction solvent) forms two clearly separated phases in the presence of sodium sulfate (as a phase separation agent) and simultaneously the analytes are extracted into the fine droplets of the acetonitrile collected on the surface of the aqueous phase. To achieve high enrichment factors, the acetonitrile phase is mixed with 1,2-dibromoethane (as a preconcentration solvent) at µL-level to perform the following dispersive liquid-liquid microextraction procedure. Several parameters that can affect extraction efficiency including kind and volume of extraction solvent, type and concentration of phase separation agent, hexane volume, kind of preconcentration solvent, and ionic strength were studied and optimized. Under the optimal conditions, extraction recoveries were obtained in the range of 53-93% and the calibration curves were linear in wide ranges with correlation coefficients ≥ 0.9983. Intra- (n = 6) and inter-day (n=5) precisions of the method were satisfactory with relative standard deviations less than or equal to 7% (at two concentrations of 10 and 50µgL-1 of each analyte). Moreover, the detection limits and enrichment factors of the target analytes were obtained in the ranges of 0.43-1.2µgL-1 and 264-464, respectively. Finally, the proposed method was applied on different fruit and vegetable samples and diazinon was determined in apple sample at a concentration of 0.030µgg-1 .