Phenolics are an important group of wine compounds which have been linked with key sensory attributes, including the color and astringency of wines. For winemakers, accurate determination of phenolics is an increasingly important component of the quality monitoring process. One promising approach for the quantification of phenolics is through the use of ultraviolet-visible (UV-Vis) spectroscopy, which is inexpensive and requires minimal sample preparation. Previous research has explored some parts of the UV–vis spectrum for various types of red wines, though none have extensively explored Pinot noir wines, which have a lower average phenolic content than other common varieties such as Cabernet Sauvignon and Merlot. UV-Vis spectra were obtained alongside reference determinations of phenolic compounds using high-performance liquid chromatography, for a set of 155 Pinot noir wines from New Zealand. Partial least squares (PLS) regression models, using the UV-Vis spectra and the reference HPLC measurements, were constructed. Overall, the PLS regression models indicated that phenolic compounds with high concentrations and groups of compounds could be determined with a high degree of accuracy, while for phenolic compounds with lower concentrations the prediction performance was generally less accurate.
Read full abstract