Publicação
Toward Port wine authenticity: from spectrum to signature
| dc.contributor.author | Ferreira, V. S. Fernandes | |
| dc.contributor.author | Lima, Manuel | |
| dc.contributor.author | Ribeiro, Natália | |
| dc.contributor.author | Ferreira, A. C. Silva | |
| dc.date.accessioned | 2026-09-04T16:10:45Z | |
| dc.date.available | 2026-09-04T16:10:45Z | |
| dc.date.issued | 2026-08-24 | |
| dc.description.abstract | Port wine is a fortified wine whose aging process involves marked molecular transformations that influence both sensory properties and commercial classification. Oxygen exposure and temperature are the main factors governing this evolution. Different oxidation regimes, from prolonged oxidative aging in wood to comparatively lower oxygen exposure in bottle, contribute to the differentiation of Port wine categories such as Colheita, Tawny, Vintage, and Late Bottled Vintage. At the molecular level, carbohydrates, amino compounds, and phenolic compounds are interconnected through aging-related oxidative reactions, including Strecker degradation, Maillard-type pathways, and the formation of furanic compounds derived from sugar degradation. These oxygen-dependent chemical trajectories may generate infrared spectral signatures that evolve predictably with age. In this study, FTIR spectroscopy was evaluated as a rapid and nondestructive approach for Port wine age estimation and authenticity assessment. A data set comprising 7281 spectra from four commercial categories was analyzed using classical chemometric methods and machine learning models. For category classification, a support vector machine with a radial basis function kernel achieved 95.88% accuracy and an ROC-AUC of 0.9945. For age prediction, a deep ensemble model achieved an R2 value of 0.9803 and a mean absolute error of 1.25 years. A hybrid two-stage random forest model, designed to account for category-specific aging regimes, achieved comparable performance, with an R2 of 0.9630 and a mean absolute error of 1.28 years, while requiring lower computational resources. Leveragebased prediction intervals provided per-sample uncertainty estimates with a 97.7% empirical coverage. Feature importance analysis identified three main spectral regions, around 1740, 1050, and 1600−1450 cm−1 , associated with ester-, alcohol-, and phenolicrelated functional groups, respectively. These regions are consistent with known chemical transformations occurring during Port wine aging. The results demonstrate that FT-IR spectroscopy, combined with appropriate modeling strategies, can support rapid, nondestructive, and scalable estimation of Port wine age, offering a promising complementary tool for certification and authenticity control. | eng |
| dc.identifier.doi | 10.1021/acsfoodscitech.6c00536 | |
| dc.identifier.other | d71eae36-5357-4afa-8854-a9079ca1c2a9 | |
| dc.identifier.uri | http://hdl.handle.net/10400.14/59274 | |
| dc.identifier.wos | 001857088300001 | |
| dc.language.iso | eng | |
| dc.peerreviewed | yes | |
| dc.publisher | American Chemical Society | |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | |
| dc.subject | FT-IR spectroscopy | eng |
| dc.subject | Port wine authentication | eng |
| dc.subject | Age prediction | eng |
| dc.subject | Chemometrics | eng |
| dc.subject | Deep learning | eng |
| dc.subject | Esterification | eng |
| dc.subject | Food fraud | eng |
| dc.subject | Nondestructive analysis | eng |
| dc.subject | Oxidative aging chemistry | eng |
| dc.subject | Polyphenol oxidation | eng |
| dc.subject | Venetian blind cross-validation | eng |
| dc.title | Toward Port wine authenticity: from spectrum to signature | |
| dc.type | research article | |
| dspace.entity.type | Publication | |
| oaire.version | http://purl.org/coar/version/c_ab4af688f83e57aa |
Ficheiros
Principais
1 - 1 de 1