In this paper, it is demonstrated for the first time and in the case of a complex sample such as red wine, that also quantification of integral parameters (colour indicators) and specific compounds (total anthocyanins and polyphenols) is possible through a multivariate analysis of CSPT fingerprints. Most of the properties of the method are due to the combination of light emission and absorption properties that are captured by the CSPT platform. Thanks to this combination, a CSPT fingerprint may contain a comparable amount of information with respect to standard spectrophotometers.
Wine measurements were performed on intact samples without the intervention of chemical mediators. The regression models, built by Partial Least Squares, obtained errors of estimation of colour parameters, total polyphenols and anthocyanines that are comparable with those typical of the standard methods in use.
Since computation functionalities, video capture and display are embedded in a steadily growing number of ubiquitously distributed equipments (from portable computers to cellular phones), these results indicate a viable methodology for low-cost and largely diffused analytical capabilities.