Determination of hormones, a plasticizer, preservatives, perfluoroalkylated compounds, and a flame retardant in water samples by ultrasound-assisted dispersive liquid–liquid microextraction based on the solidification of a floating organic drop

• Applicability of DLLME-SFO to multiresidue analysis was evaluated.
• A DLLME-SFO method for the analysis of 15 organic pollutants was developed.
• Recoveries were highly dependent on log Kow values.
• Recoveries of homologs were linearly correlated to their log Kow values.
• Good precision and low matrix effect were obtained.

Dispersive liquid–liquid microextraction based on the solidification of a floating organic drop (DLLME-SFO) is a novel extraction technique commonly applied for the extraction on a specific group of compounds. In this paper, the applicability of ultrasound-assisted DLLME-SFO for multiresidue extraction has been evaluated. A method for the simultaneous extraction of four hormones (17α-ethinylestradiol, 17β-estradiol, estriol and estrone), a plasticizer (bisphenol A), three preservatives (methyl-, ethyl- and propylparaben), six perfluoroalkylated compounds (perfluorooctane sulfonic acid and five perfluoroalkyl carboxylic acids, from C4 to C8), and a brominated flame retardant (hexabromocyclododecane) has been developed and validated for their extraction from surface water and tap water. Determination was carried out by high-performance liquid chromatography–tandem mass spectrometry in negative ionization mode. Recoveries of the target compounds were highly dependent on their log Kow values. Linear relationship between recoveries and log Kow values was observed for compounds from the same group (hormones, preservatives and perfluoroalkylated carboxylic acids). The lowest recoveries were obtained for the less hydrophobic compounds (estriol (43%), methylparaben (32%), ethylparaben (45%) and the perfluorinated compounds of shorter alkyl chain (C4: 17%, C5: 41% and C6: 57%)). Recoveries of the other pollutants were higher than 80%. Precision, expressed as relative standard deviation, was in the range from 1% to 16%. Method detection limits were in the range 0.001–1.126 µg L−1, for surface water, and 0.001–1.446 µg L−1 for tap water. No important matrix effect was observed.

Figure optionsDownload as PowerPoint slide