Trying to set up the flavanolic phases during grape seed ripening: A spectral and chemical approach

•Hyperspectral imaging is used for classifying seeds with different ripening degrees.•Seven groups of seeds are formed from NIR hyperspectral imaging, during ripening.•Phloroglucinolysis and oligomers HPLC-MS provided an accurate flavanol profile.•Three analytical stages for seed ripening were suggested by Canonical Biplot/MANOVA.

Grape seeds were collected in ten different dates and classified in seven groups according to their individual hyperspectral imaging characteristics.Proanthocyanidin composition was studied using HPLC-MS for oligomers and acid catalyzed cleavage for polymers characterization. The combination of both analysis provided a complete description of the flavanols. Chemometric analysis was performed to summarize the analytical results. None of the considered variables presented statistical differences among all groups. From one to five groups were found for each variable, while three was the most frequent value, consequently three putative stages might be considered the real number of different analytical stages since it is the number of statistically significant groups for the majority of the compounds. This classification could be considered as the first step to optimize the use of seeds in winemaking to minimize the gap between sugar and phenolic maturities, consequence of the global climate change, mainly observed in warm climate.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slide