Controle de qualidade de azeites a partir de análises espectroscópicas, sensorial e ferramentas estatísticas

Abstract

Olive oil is obtained by extracting the oil from the olive tree fruit. According to the current legislation, there are virgin olive oils, refined oils, olive oils and crude olivepomace oil. Virgin olive oils are very much used in cooking because of their nutritional properties and taste appreciation. As it is a more refined product and produced in a smaller scale, besides having greater commercial value with respect to the other vegetal oils, it is a great target in adulterations. The objective of the present proposal is to characterize soybean, canola, sunflower, corn and olive oil oils obtained in commercial establishments through spectroscopic techniques and to detect adulterations in olive oil using spectroscopic techniques coupled with chemometric tools and by means of analysis sensory. The work presents an innovative proposal to join modern techniques of infrared spectroscopy with Fourier transform (FTIR), near infrared spectroscopy (NIR), ultraviolet-visible light absorption spectroscopy (UV-Vis). The rapidity of processing large amounts of data using statistical tools used in data characterization was principal component analysis (PCA), and classification models were constructed using partial least squares regression (PLS) and regression by major components (PCR ). In addition, the multiple comparison test was used to determine the sensorial sensitivity of the evaluator to the adulteration of olive oil with soybean oil. From the spectral data (FTIR and UV-Vis light absorption) and the statistical analysis, PCA, it was possible to characterize the samples as to the origin of each one of them, as well as to identify the existing adulterations. The PLS and PCR models, using UVVis light absorption data and the dependent variable (5801 cm-1) showed a good correlation between predicted and meditated data. The results obtained through the multiple comparison test in the sensorial analysis were satisfactory, since they determined that from 20% the evaluator can detect the adulteration of the olive oil with soybean oil.