Discrimination of Arabica natural, Arabica washed and Robusta coffee through NIR spectroscopy and artificial tongue analysis

The objective of this study is to investigate the potential of non-destructive techniques associated with chemometric data analysis for discriminating between Arabica and Robusta species coffee samples, and between Arabica coffee treated through two different processes. A total of 227 green coffee samples were analyzed by a FT-NIR spectrometer; electronic tongue investigation was simultaneously achieved on the same samples roasted and extracted in hot water. The spectroscopic technique was used to investigate differences between beans chemical composition, while electronic tongue evaluated diversity in the extract sample taste profile. Principal Component Analysis (PCA) was applied to spectral and electronic tongue data, as an exploratory tool to uncover correlations between samples. Linear discriminant analysis is used to classify coffee samples on the basis of the species and of the different treatments for Arabica samples. The results shows the ability of near-infrared methodology to classify these three typology of coffee, while electronic tongue approach generates a whole depiction of the taste profile. The spectroscopic methods result to be a reliable, cheap and fast classification tool, not requiring chemical analyses for discrimination among species; at the same time electronic tongue profile can be elaborated with the aim of better characterize coffee chemical compositio