Tracciabilità geografica di prodotti alimentari tradizionali attraverso profili multi-elementali e spettroscopici combinati alla chemiometria

The concept of traceability is a complex and interdisciplinary field having economic and environmental implications and encompassing different actors: governments, producers and consumers. The development of an effective traceability system involves different research areas both in the food industry and in the academic world and focuses on very different objectives. In general, a good traceability system needs full cooperation between the above-mentioned actors and an effective legislative contribution. In this context, the role of analytical chemistry is mainly addressed to food authentication and control, in which the conformity of given products to the label specifications as well as the presence of contaminants or hazardous substances must be verified. In recent years, the research in food analysis has mainly focused on the identification of new and efficient indicators of quality and authenticity, through the development of sensitive, reliable analytical methods compatible with routine control analysis and to be integrated into the supply chain. The different case studies reported in this thesis concern typical regional foodstuffs grown in small-scale farms located in restricted Italian areas. To these products is, nowadays, recognized high traditional worth and market price, a value that is protected by certified trademarks. The high commercial value and the limited availability make these products vulnerable to fraud and food adulteration for profit-making purposes. In this context, small producers may not have the financial possibilities to invest in implementing an effective traceability system to protect their reputation. Thus, this work aims at developing and validating analytical methods useful for the authentication and geographical discrimination to safeguard some niche high-value foods, produced in different Italian regions. In the field of food analysis, fingerprinting techniques should be preferred to the so-called targeted analysis; the former measures the composition of foodstuffs in a non-selective way providing a wide outlook about the food matrix. This approach produces a large number of variables that allow a complete characterisation of the complex food matrix but collects, also, non-informative variables that may obscure the contribution of those useful for solving the specific problem. In this thesis, two main untargeted approaches were employed. The first exploits analytical techniques capable of generating a multi-elemental profile, providing valuable information about the origin of the product, as the observed multi-elemental pattern is directly linked to the soil and climatic conditions of the production area. The second approach involves the use of spectroscopic techniques in the UV-Vis and IR spectral ranges. Both experimental responses have been correlated with the food geographical origin through appropriate multivariate statistical processing. The employed chemometric approaches included exploratory multivariate analysis and different single and multi-block supervised pattern recognition methods. The excellent results obtained in the applications supported the effective and indispensable role of chemometric processing in the field of food control to handle complex matrices, like those provided in food analysis. Therefore, this study could be a starting point for putting in place experimental/chemometric methods that could be integrated into small-scale traceability systems aimed at protecting local markets.