Project POCI-01–0145-FEDER-006984 – Associate Laboratory LSRE-LCM and by Project
UID/QUI/00616/2013 – CQ-VR both funded by FEDER - Fundo Europeu de Desenvolvimento Regional through COMPETE2020 - Programa Operacional Competitividade e Internacionalização (POCI) – and by national funds through FCT - Fundação para a Ciência e a Tecnologia, Portugal. Strategic funding of UID/BIO/04469/2013 unit is also acknowledged. Nuno Rodrigues thanks FCT, POPH-QREN and FSE for the Ph.D. Grant (SFRH/BD/104038/2014).
Table olives are highly appreciated and consumed worldwide. Different aspects are used for trade category classification being the sensory assessment of negative defects present in the olives and brines one of the most important. The trade category quality classification must follow the International Olive Council directives, requiring the organoleptic assessment of defects by a trained sensory panel. However, the training process is a hard, complex and sometimes subjective task, being the low number of samples that can be evaluated per day a major drawback considering the real needs of the olive industry. In this context, the development of electronic tongues as taste sensors for defects' sensory evaluation is of utmost relevance. So, an electronic tongue was used for table olives classification according to the presence and intensity of negative defects. Linear discrimination models were established based on sub-sets of sensor signals selected by a simulated annealing algorithm. The predictive potential of the novel approach was first demonstrated for standard solutions of chemical compounds that mimic butyric, putrid and zapateria defects (≥93% for cross-validation procedures). Then its applicability was verified; using reference table olives/brine solutions samples identified with a single intense negative attribute, namely butyric, musty, putrid, zapateria or winey-vinegary defects (≥93% cross-validation procedures). Finally, the E-tongue coupled with the same chemometric approach was applied to classify table olive samples according to the trade commercial categories (extra, 1(st) choice, 2(nd) choice and unsuitable for consumption) and an additional quality category (extra free of defects), established based on sensory analysis data. Despite the heterogeneity of the samples studied and number of different sensory defects perceived, the predictive linear discriminant model established showed sensitivities greater than 86%. So, the overall performance achieved showed that the electrochemical device could be used as a taste sensor for table olives organoleptic trade successful classification, allowing a preliminary quality assessment, which could facilitate, in the future, the complex task of sensory panelists.
