Astringency quantification in wine: comparison of the electronic tongue and FT-MIR spectroscopy

摘要

Electronic tongue based on potentiometric chemical sensors and Fourier transform mid-infrared spectroscopy (FT-MIR) were evaluated as rapid tools for the prediction of gelatin index of red, white and rose wines. Gelatin index was measured using modified procedure, which was applicable not only to red but also to white and rose wines. Relationship between gelatin index and a set of chemical parameters, electronic tongue and FT-MIR data have been investigated using multi-block analysis method called common component and specific weight analysis (CCSWA). Calibration models for gelatin index prediction were calculated using partial least square (PLS) regression. Variable selection for the PLS models was done using Variable Importance in Prediction criterion. Quantification of gelatin index in red and rose wine was possible using both electronic tongue and FT-MIR data with adjusted R~2 in prediction of 0.75 and 0.89 for rose and 0.83 and 0.87 for red wine, respectively.