Logo: to the web site of Uppsala University

In this thesis, the main concern has been to improve the reliability of different parts of the analytical workflow (Paper I, II, IV &V). Additionally, one of the resulting optimized methods was used in a real application (Paper III).

Paper I-II concerned the extraction of acrylamide (AA) from foods. In Paper I different parameters such as sample particle size, extraction solvent, extraction time and extraction temperature were optimized, leading to a method that showed good agreement with the assigned AA levels of several proficiency test samples. Later, after the publication of the paper, the method showed good performance in a collaborative trial validation, in terms of trueness, repeatabil­ity and reproducibility figures. It was labeled “undoubtedly fit for the purpose”.

In Paper II, it was shown that the ‘extra’ amounts of AA obtained during extraction of foods with an alkaline aqueous solution was not due to improved extractability of AA from the food matrix. Strongly alkaline conditions seemed to rather induce net formation of AA from water-soluble precursors formed during thermolysis. This phenomenon should therefore be regarded as an extraction artifact.

Paper III was an application of the optimized method from Paper I, where it was used to study the reduction of AA in potato chips (crisps) by using pre-treatments and frying at reduced pressure. There were significant reductions in AA, down to below the limit of quantification (5 µg/kg) for the method.

Paper IV-V concerned analysis of anthocyanins (AC) in red onion. In Paper IV, a new separation method using capillary electrophoresis was developed, and its rapidness combined with an acidic background electrolyte helped in preventing AC degradation. Furthermore, its alternative separation mechanism is a complement to that of the more commonly used liquid chromatography technique. In Paper V, simultaneous extraction and degradation of anthocyanins from red onion was studied in a static batch reactor at 110ºC. The extraction and degradation kinetics were successfully separated, and an ideal theoretical extraction curve was constructed by compensating mathematically for degradation effects, showing that more anthocyanins, 21 to 36% depending on different species, could be extracted if no degradation occurred. The results give important information about the different kinetics competing during an extraction procedure, and also show that quantitative extraction is not to recommend in the batch system used in the study.