Development of new analytical methods and application of chemometric tools in flavour research

In food science, there are several ways of defining quality, and perhaps there is no single universal definition that adequately satisfies all situations; in general terms, quality is defined as “the combination of attributes or characteristics of a product that have significance in determining the degree of acceptability of that product by the consumer”. The aim of this Thesis lies mostly on food quality field: indeed, one of the most important criteria for consumer acceptance of food is flavour. Due to the increasing demand of high quality and enjoyable foods, the need for methodologies enabling a better understanding of flavour has arisen: chemical analysis is a valuable way of studying volatile composition of different food matrices, and very sophisticated instrumentations are available nowadays, but almost always there is the need of clean-up/concentration steps before such analysis. Moreover, when volatile compounds are present in traces (ppb or even ppt), their detection and identification require the development of analytical tools which can tackle these difficult tasks: in the past, they were largely detected with insufficient performances or even undetected, due to the inadequacies in old-fashioned analytical techniques. Hence, in this work novel, simple, rapid, and “environmental friendly” sample preparation methods were implemented, which coupled really well with Thermal Desorption – Gas Chromatography – Mass Spectrometry (TD-GC-MS), following also the Green Chemistry requirements. It was investigated the use of Stir Bar Sorptive Extraction (SBSE) technique, showing that it is extremely suited for troublesome analytical challenges such as isolation of volatile compounds from several food matrixes, including both processed (beverages, yogurts) and fresh foods (fruits and vegetables); the application of this technique was thus broaden in new complex matrices (e.g. yogurts). This sorptive sample preparation showed good performances for practical, real-life analytical problems, which cannot easily be solved by alternative and time-consuming approaches: it allows an effective extraction of volatiles in the trace range, and a rich aromatic profile may be pursued using optimized conditions. SBSE has proven to be a technique that can offer high and reproducible extraction recoveries for the complex matrices under study; it was also characterized by ease of use, good repeatability and robustness. In addition, Headspace Solid Phase Micro-Extraction (HSPME) was used for sampling volatile from the headspace of olive oil samples: the point is that enrichment for gaseous samples is more difficult; particularly the analysis of polar compounds is an analytical problem that cannot be addressed adequately by conventional techniques, i.e. distillation, dynamic headspace, etc. Using this other solventless extraction technique, a reliable analytical method for the determination of a broad range of volatiles in olive oil was developed. Then, looking also at volatile release, which is a complementary task for a better understanding of interactions among food components, volatiles and in-vivo behavior, a real-time instrumentation, Proton Transfer Reaction - Mass Spectrometry (PTR-MS), was used to follow these rapid changing phenomena, without any need of sample preparation or calibration. The application of these analytical methods was then combined with the use of suitable chemometric techniques such as PCA (Principal Component Analysis), ASCA (Analysis of Variance combined to Simultaneous Component Analysis), MIXTURE DESIGN MODELING, and PLSR (Partial Least Squares Regression). We truly think that nowadays chemometrics is unavoidable and of utmost importance, when scientists have to deal with multivariate datasets, and when robust “chemical-driven” modeling of complex phenomena is needed.