KINETIC STUDY OF SUPERCRITICAL FLUID EXTRACTION OF ORGANIC CONTAMINANTS FROM HETEROGENEOUS ENVIRONMENTAL SAMPLES WITH CARBON DIOXIDE AND ELEVATED TEMPERATURES

摘要

Supercritical fluid extraction (SFE) rates of spiked polychlorinated dibenzo-p-dioxins (PCDDs) from Florisil, spiked [C-13]PCDDs and native PCDDs from fly ash, and spiked [H-2]polycyclic aromatic hydrocarbons (PAHs) and native PAHs from marine sediment and railroad bed soil were examined at 40, 120, and 200 degrees C, while constant fluid density (d = 0.67 g/mL) and now rate were maintained, Over 30 min (150 void volumes of CO2) was required to quantitatively remove both spiked 2,3,7,8-tetrachlorodibenzo-p-dioxin and 1,2,3,7,8-pentachlorodibenzo-p-dioxin from Florisil at 40 degrees C, while SFE at 200 degrees C significantly improved the elution rate so complete removal was achieved in similar to 10 min, Elution rates of spiked PCDDs from Florisil were slower with a 5-mL vessel (12 cm long) than a 0.5-mL vessel (6 cm long), Increasing the temperature from 40 to 120 and 200 degrees C enhanced the SFE rates of spiked [C-13]PCDDs and native PCDDs from fly ash, as well as [H-2]PAHs and native PAHs from marine sediment and railroad bed soil, In all cases, native analytes were extracted more slowly than spiked analytes, suggesting that additional processes affect SFE rates of native analytes, A kinetic model was used that could help distinguish between these processes and included terms for matrix-fluid mass transport, as well as partitioning and bulk mass transport in the supercritical fluid. Using a three-rate constant desorption model to describe mass transport, good correlations (r(2) > 0.9 in most cases) were obtained with experimental data for native analytes, and desorption rate constants suggest that analyte-matrix interactions are strong, The results of this study show that increasing the extraction temperature is a simple and effective method to increase SFE rates while still exploiting the advantages of supercritical CO2, and can be used regardless of whether slow SFE rates are due to poor partitioning into the fluid or limited by slow desorption due to strong analyte-matrix interactions.