No abstract

a b s t r a c tAn analytical methodology based on headspace solid phase microextraction (HS-SPME) combined with comprehensive two-dimensional gas chromatography-time-of-flight mass spectrometry (GC × GC-ToFMS) was developed for the identification and quantification of the toxic contaminant ethyl carbamate (EC) directly in fortified wines. The method performance was assessed for dry/medium dry and sweet/medium sweet model wines, and for quantification purposes, calibration plots were performed for both matrices using the ion extraction chromatography (IEC) mode (m/z 62). Good linearity was obtained with a regression coefficient (r 2 ) higher than 0.981. A good precision was attained (R.S.D. <20%) and low detection limits (LOD) were achieved for dry (4.31 g/L) and sweet (2.75 g/L) model wines. The quantification limits (LOQ) and recovery for dry wines were 14.38 g/L and 88.6%, whereas for sweet wines were 9.16 g/L and 99.4%, respectively. The higher performance was attainted with sweet model wine, as increasing of glucose content improves the volatile compound in headspace, and a better linearity, recovery and precision were achieved. The analytical methodology was applied to analyse 20 fortified Madeira wines including different types of wine (dry, medium dry, sweet, and medium sweet) obtained from several harvests in Madeira Island (Portugal). The EC levels ranged from 54.1 g/L (medium dry) to 162.5 g/L (medium sweet).

Wine aroma is the result of complex interactions between volatile compounds and non-volatile ones and individual perception phenomenon. In this work, an aroma network approach, that links volatile composition (chromatographic data) with its corresponding aroma descriptors was used to explain the wine aroma properties. This concept was applied to six monovarietal wines from Bairrada Appellation (Portugal) and used as a case study. A comprehensive determination of the wines’ volatile composition was done (71 variables, i.e., volatile components), establishing a workflow that combines extraction techniques and gas chromatographic analysis. Then, a bipartite network-based approach consisting of two different nodes was built, one with 19 aroma descriptors, and the other with the corresponding volatile compound(s). To construct the aroma networks, the odor active values were calculated for each determined compound and combined with the bipartite network. Finally, the aroma network of each wine was compared with sensory descriptive analysis. The analysis of the specific aroma network of each wine revealed that Sauvignon Blanc and Arinto white wines present higher fruity (esters) and sweet notes (esters and C13 norisoprenoids) than Bical wine. Sauvignon Blanc also exhibits higher toasted aromas (thiols) while Arinto and Bical wines exhibit higher flowery (C13 norisoprenoids) and herbaceous notes (thiols), respectively. For red wines, sweet fruit aromas are the most abundant, especially for Touriga Nacional. Castelão and Touriga Nacional wines also present toasted aromas (thiols). Baga and Castelão wines also exhibit fusel/alcohol notes (alcohols). The proposed approach establishes a chemical aroma fingerprint (aroma ID) for each type of wine, which may be further used to estimate wine aroma characteristics by projection of the volatile composition on the aroma network.

No abstract