Sniffing out the Tempranillo

Chemometrics and mass spectrometry can be used to classify wine according to geographic region. Proof of principle was undertaken with the new electronic nose carried out successfully with Tempranillo wines from Spain and Australia.

Wies Cynkar, Paul Smith, and Daniel Cozzolino of The Australian Wine Research Institute, in Glen Osmond, South Australia and colleague Robert Dambergs at the University of Tasmania, point out that the authentication of food and drink products is of major importance for consumers, producers, and regulators alike. They suggest that it is even more pertinent to wine, given that the exact composition of a wine influenced by many factors such as production area, grape varieties, soil and climate and wine making practices all affect quality and the wine-tasting experience. "In many wine producing countries strict rules apply for defining regionality of the wine," the team says.

One might imagine that adulteration of wine especially with regard to specific grape varieties and regions of origin is an ongoing issue facing the industry, Indeed, the problem has been studied extensively the team explains, because it is such a simple matter to produce an adulterated product that would not be readily detectable. "Today, the determination of food authenticity and the detection of adulteration are major issues in the food industry attracting increased attention of wine producers, researchers and consumers," the team adds. Wines from particular regions often attract a higher price than wines from less desirable areas, so there are now compelling reasons to find straightforward and inexpensive methods for detecting wine adulteration and to be able to identify the region of origin of a wine.

Researchers have developed various sensor devices and approaches that are often known colloquially as an "electronic nose" for testing food and drink. The majority are based on the direct analysis of the volatile compounds present in the vapour above a sample, the headspace, and commonly use a gas sensor array. Chemometrics are then applied to study the sensor responses and to interpret and classify the headspace and to correlate the pattern obtained from each test product against standards. There are, say the researchers, many different types of electronic nose, some based on metal oxide sensors, others on conducting polymers. There are quartz crystal membrane sensors and mass spectrometers used as an electronic nose are also commercially available.

Systems for wine have been lacking because of the overwhelming presence of ethanol in the headspace. Various preparative and evaporative strategies have been tried to circumvent that problem but they compromise the advantages of using an electronic nose in the first place, the researchers add.

About a decade ago, an electronic nose based on mass spectrometry was developed that is operated in a similar setup to gas chromatography-MS but without the GC! Fragments due to ethanol in the mass spectra are easily ignored. As such, MS-EN has been used in the analysis of food and alcoholic beverages ever since. It is particularly suited to determining specific components of wine or determining the maturation age of spirits in oak barrels.

"Multivariate statistical techniques such as principal component analysis (PCA) or discriminant analysis (e.g. linear discriminant analysis, LDA; or partial least squares discriminant analysis PLS-DA), provide the possibility to extract and use information based on the overall properties of the sample and perform a classification without the need for additional compositional data," the team reasoned.

They have now applied this chemometrics approach to MS-EN in order to analyse the popular Tempranillo wine with a view to defining the origin of given samples from Spain and Australia.

The team analysed the MS-EN data and used PCA, PLS-DA, and stepwise LDA with full cross validation (leave-one-out method). SLDA correctly classified wine origin in 86% of samples while PLS-DA was 85% accurate. "The relative benefits of using MS-EN will provide capability for rapid screening of wines," the team concludes.