RESUMO
A study was conducted to evaluate the efficiency and reproducibility of a diethyl ether-based soxhlet extraction procedure for faecal sterols occurring from catchment waters. Water samples spiked with a mixture of faecal sterols were filtered and analytes were extracted using the diethyl ether-based soxhlet method and the Bligh and Dyer chloroform extraction process. For diethyl ether-based soxhlet extraction procedure, solvent extracts were saponified with 100 microL of 10% KOH in methanol (100 degrees C/120 min) and then acidified with 60 microL of 6M HCl. Lipid contents were extracted by ethanol (0.5 mL) from the saponification products. The lipid extracts were then reacted with 100 microL of bis(trimethyl)trifluoroacetamide (BSTFA) containing 1% trimethyl chlorosilane (100 degrees C/60 min) to form the trimethylsilyl (TMS) derivatives. The derivatised extracts were then analyzed by gas chromatography-mass spectrometry. For sterol concentrations ranging from 35 to 175 microg mL(-1), the soxhlet-based extraction process yielded the following recovery efficiencies for coprostanol (101%), epicoprostanol (97%), cholesterol (97%), dihydrocholesterol (97%) and 5alpha-cholestane (111%), whereas the Bligh and Dyer process yielded recoveries of 32, 41, 0, 36 and 51%, respectively. The results suggested that the diethyl ether-based soxhlet extraction method was more efficient and reproducible than the Bligh and Dyer chloroform extraction process for the analyses of trace levels of faecal sterols from water samples. Moreover, it was revealed that the diethyl ether-based soxhlet extraction method used less solvent and was logistically easier.