Search

Aims: Fourier Transform-Near Infrared (FT-NIR) spectrum and instrumental texture parameters were assessed as total phenol content and extractability predictors in intact grape seeds. Methods and results:The study was carried out on Cabernet-Sauvignon seeds from grapes harvested at six different advanced physiological stages throughout ripening and calibrated by flotation to reduce the in-field heterogeneity inside each sample. Among the instrumental mechanical properties tested (i. e., break force, break energy, Young’s modulus of elasticity and deformation index), the seed Young’s modulus of elasticity showed an increase during the first four weeks of ripening. This parameter also showed significant correlations with phenol content and extractability, although with low R coefficients. These correlations highlighted that the springier seed tissues greatly increase phenol extractability. Nevertheless, the best prediction of seed phenol content, performed directly on intact seeds, was found using FT-NIR spectroscopy in transmittance mode. The standard error of prediction for total phenol content was less than 8 %, while that for phenol extractability was worse. Conclusion: On the basis of these results, the two analytical methods could be applied in oenology for the rapid monitoring of seed phenolic maturity. Significance and impact of the study: The phenolic composition of grapes at the harvest time is a key factor determining their quality, and thus the quality of the finished wine. The chemical methods used for the determination of seed phenol content and extractability are generally slow because they require a preliminary extraction. Therefore, a rapid evaluation of these parameters could be highly interesting for the oenological sector.