Distinguishment, identification and aroma compound quantification of Portuguese olive oils based on physicochemical attributes, HS-GC/MS analysis and voltammetric electronic tongue.

BACKGROUND: In this paper, various extra-virgin and virgin olive oils samples from different Portuguese markets were studied. For this purpose, a voltammetric electronic tongue (VE-tongue), consisting of two kinds of working electrode within the array, together with physicochemical analysis and headspace gas chromatography coupled with mass spectrometry (HS-GC-MS), were applied. In addition, preliminary considerations of relationships between physicochemical parameters and multisensory system were reported. RESULTS: The physicochemical parameters exhibit significant differences among the analyzed olive oil samples that define its qualities. Regarding the aroma profile, 14 volatile compounds were characterized using HS-GC-MS; among these, hex-2-enal, hexanal, acetic acid, hex-3-ene-1-ol acetate and hex-3-en-1-ol were semi-quantitatively detected as the main aroma compounds in the analyzed samples. Moreover, pattern recognition methods demonstrate the discrimination power of the proposed VE-tongue system. The results reveal the VE-tongue's ability to classify olive oil samples and to identify unknown samples based of built models. In addition, the correlation between VE-tongue and physicochemical analysis exhibits a remarkable prediction model aimed at anticipating carotenoid content. CONCLUSION: The preliminary results of this investigation indicate that physicochemical and HS-GC-MS analysis, together with multisensory system coupled with chemometric techniques, presented a satisfactory performance regarding olive oil sample discrimination and identification. © 2017 Society of Chemical Industry.

Emerging approach for analytical characterization and geographical classification of Moroccan and French honeys by means of a voltammetric electronic tongue.

Moroccan and French honeys from different geographical areas were classified and characterized by applying a voltammetric electronic tongue (VE-tongue) coupled to analytical methods. The studied parameters include color intensity, free lactonic and total acidity, proteins, phenols, hydroxymethylfurfural content (HMF), sucrose, reducing and total sugars. The geographical classification of different honeys was developed through three-pattern recognition techniques: principal component analysis (PCA), support vector machines (SVMs) and hierarchical cluster analysis (HCA). Honey characterization was achieved by partial least squares modeling (PLS). All the PLS models developed were able to accurately estimate the correct values of the parameters analyzed using as input the voltammetric experimental data (i.e. r>0.9). This confirms the potential ability of the VE-tongue for performing a rapid characterization of honeys via PLS in which an uncomplicated, cost-effective sample preparation process that does not require the use of additional chemicals is implemented.

Aging time and brand determination of pasteurized milk using a multisensor e-nose combined with a voltammetric e-tongue.

A combined approach based on a multisensor system to get additional chemical information from liquid samples through the analysis of the solution and its headspace is illustrated and commented. In the present work, innovative analytical techniques, such as a hybrid e-nose and a voltammetric e-tongue were elaborated to differentiate between different pasteurized milk brands and for the exact recognition of their storage days through the data fusion technique of the combined system. The Principal Component Analysis (PCA) has shown an acceptable discrimination of the pasteurized milk brands on the first day of storage, when the two instruments were used independently. Contrariwise, PCA indicated that no clear storage day's discrimination can be drawn when the two instruments are applied separately. Mid-level of abstraction data fusion approach has demonstrated that results obtained by the data fusion approach outperformed the classification results of the e-nose and e-tongue taken individually. Furthermore, the Support Vector Machine (SVM) supervised method was applied to the new subset and confirmed that all storage days were correctly identified. This study can be generalized to several beverage and food products where their quality is based on the perception of odor and flavor.