Analytical study of volatile compounds in wine. Chemometric characterization of different denomination of origin (Spanish text).

摘要

The objective of this work is the comparative study of different extraction procedures for volatile compounds (VOCs) from wines. Several methods have been developed for the analysis of volatile components in wine.; The extraction procedures studied were: Discontinuous liquid-liquid extraction, Continuous liquid-liquid extraction, Microextraction, Solid-phase extraction (SPE) and Solid-phase microextraction (SPME). The advantages and disadvantages of the methods have been evaluated. Also, the sensitivity and reproducibility have been compared.; The study has been applied to 12 compounds present in wine: 3-methyl-butyl acetate (A3M1B), 3-methyl-1-butanol (3M1B), ethyl hexanoate (HE), hexanol (1H), ethyl octanoate (OE), diethyl succinate (SD), phenyl ethyl acetate (AFE), hexanoic acid (AH), geraniol (G), phenyl ethyl alcohol (2FE), octanoic acid (AO), decanoic acid (AD). This method was allowed to determine the VOCs identified in the wine samples providing good sensitivity and reproducibility. These compounds selected were identified by GC-MS analysis of real wine sample.; Solid-phase microextraction (SPME) in headspace mode coupled with gas-chromatography (CG) and a flame ionization detector (FID) is selected as a rapid method for the analysis of volatile compounds from wine samples. The optimised method was successfully applied to the analysis of wine samples from five Denomination of Origin (Rioja, Navarra, Valdepenas, Carinena and La Mancha).; Several types of fibres were studied and the extraction method was optimised. The fibre that proved to be the best to analyse this type of samples was Carbowax/Divinylbenzene (CW/DVB).; The chemometric tool is very appropriate in screening experiments where the aim is to investigate several possibly influential and interacting factors.; Multivariate chemometric techniques were used to construct sensitive and specific models for wines. Principal Component Analysis (PCA), Lineal Discriminant Analysis (LDA) and Cluster Analysis (CA) were applied to the data, revelling good separation between the groups whit LDA.; LDA show differences in wines according to the Denomination origin.