Determination of Persistent Organic Pollutants in Solid Environmental Samples Using Accelerated Solvent Extraction and Supercritical Fluid Extraction. Exhaustive extraction and sorption/desorption studies of PCBs

Sammanfattning: Human activity is constantly causing environmental problems due to production and release of numerous chemicals. A group of compounds of special concern is persistent organic pollutants (POP). These toxic, lipophilic chemicals have a high chemical and biological stability, and tend to accumulate in the lipid phase of living organisms. A major sink for POPs are sediments, and consequently these are important for the distribution of POPs in the aquatic environment. Traditionally, determination of POPs relay on exhaustive extraction using liquid extraction techniques (e.g. Soxhlet extraction developed in the late 19th century) followed by gas chromatographic analysis. Since liquid-solid extraction normally requires large volumes of organic solvents in combination with long extraction times and extract clean-up, there has been an increasing demand for improved technology. This should result in reduced organic solvent consumption and sample preparation time, at the same time improving the environment and cutting costs for POP monitoring. In this thesis two modern techniques with capability of fulfilling at least one of these goals have been investigated: (1) Supercritical Fluid Extraction (SFE), and (2) Accelerated Solvent Extraction (ASE). Polychlorinated biphenyls (PCBs) were chosen as model compounds in all experiments performed on environmental matrices, since they cover a relatively large range of physiochemical parameters. Important parameters influencing the overall extraction efficiency in ASE and SFE, are discussed and illustrated for a large number of sediments. It was demonstrated that, by careful consideration of the experimental parameters, both techniques are capable of replacing old methods such as Soxhlet extraction. ASE is somewhat faster than SFE, but the extracts generated in SFE are much cleaner and can be analyzed without sample clean-up. Consequently the overall sample preparation time may be substantially lower using SFE. However, ASE is important to verify that the developed SFE methods are exhaustive. It is quite clear though that there is no reason to continue using for example Soxhlet extraction, which should be replaced in routine laboratories. The possibility of utilizing SFE as a selective tool in sorption/desorption studies of POPs in natural sediments was also addressed. This second objective was proven successful and sediments could be characterized in terms of resistance towards desorption of bound analytes under supercritical conditions. These data could be correlated to desorption processes occurring under natural conditions. Additionally it could be verified that sorption of POPs from water to sediment is a very slow process requiring months or even years. This supports recent research results, demonstrating that distribution coefficients many times are underestimated since the system has not reached equilibrium.