TY - THES ID - 148643836 TI - Optimization of a novel LC-GCxGC-ToFMS/FID system for mineral oil analysis in food AU - Bauwens, Grégory AU - Purcaro, Giorgia AU - Sindic, Marianne AU - Maesen, Philippe AU - Danthine, Sabine AU - Focant, Jean-François AU - Panto, Sebastiano PY - 2020 PB - Liège Université de Liège (ULiège) DB - UniCat KW - MOSH KW - MOAH KW - Mineral Oil KW - LC-GC KW - GC×GC KW - MS KW - FID KW - column set KW - 2D quantification KW - Physique, chimie, mathématiques & sciences de la terre > Chimie UR - https://www.unicat.be/uniCat?func=search&query=sysid:148643836 AB - Mineral oil is an ubiquitous contamination, divided in two fractions MOSH (mineral oil saturated hydrocarbon) and MOAH (mineral oil aromatic hydrocarbon), for which there is currently a lack of toxicological data and therefore no legislation. The MOAH are strongly thought to be carcinogenic and genotoxic agents. The toxicology of the MOSH is still unknown, but they accumulate in the human body. The allowance of some kinds of mineral oil in the food industry complicates further the matter. The European Union has issued a call to collect more data on the topic in order to take decisions. The standard method currently used liquid chromatography-gas chromatography-flame ionization detector (LC-GC-FID) has the purpose of quantifying the contamination but does not give any information regarding the compounds and the classes. Moreover, the European Union requires the use of a confirmatory method for contaminants in foods. Comprehensive gas chromatography (GC×GC) and mass spectrometry (MS) are the methods currently proposed. Furthermore, GC×GC is useful to separate the different subclasses of the two fractions and this separation could be useful for the toxicological studies. 
The use of GC×GC as a confirmatory method is not currently really developed. In fact, it requires the use of twice the usual number of analysis because this method is currently not used to quantify. The purpose of this thesis is, first of all, to evaluate the possibility to quantify using GC×GC and compare the results obtained with those of the standard method in order to bypass this limitation. The results are encouraging even though some improvements are needed, especially concerning the removal of the peaks on top of the hump. 
Moreover, there is currently a discussion about the best set of columns to be used in order to separate the different subclasses of the two fractions. Different parameters were evaluated in this thesis and the best set seems to be a combination of a mid-polar column with a thin film and an apolar column (reverse set of column) 
The double detection of MS and flame ionization detector (FID) in order to quantify some regions or compounds identified thanks to the MS will also be briefly aborded. ER -