An alternative method for analysis of food taints using stir bar sorptive extraction

Abstract

The determination of taints in food products currently can involve the use of several sample extraction techniques, including direct headspace (DHS), steam distillation extraction (SDE) and more recently solid phase microextraction (SPME). Each of these techniques has disadvantages, such as the use of large volumes of solvents (SDE), or limitations in sensitivity (DHS), or have only been applied to date for determination of individual or specific groups of tainting compounds (SPME). The use of stir bar sorptive extraction (SBSE) has been evaluated as a quantitative screening method for unknown tainting compounds in foods. A range of commonly investigated problem compounds, with a range of physical and chemical properties, were examined. The method was optimised to give the best response for the majority of compounds and the performance was evaluated by examining the accuracy, precision, linearity, limits of detection and quantitation and uncertainties for each analyte. For most compounds SBSE gave the lowest limits of detection compared to steam distillation extraction or direct headspace analysis and in general was better than these established techniques. However, for methyl methacrylate and hexanal no response was observed following stir bar extraction under the optimised conditions. The assays were carried out using a single quadrupole GC–MS in scan mode. A comparison of acquisition modes and instrumentation was performed using standards to illustrate the increase in sensitivity possible using more targeted ion monitoring or a more sensitive high resolution mass spectrometer. This comparison illustrated the usefulness of this approach as an alternative to specialised glassware or expensive instrumentation. SBSE in particular offers a ‘greener’ extraction method by a large reduction in the use of organic solvents and also minimises the potential for contamination from external laboratory sources, which is of particular concern for taint analysis.