A qualitative analysis of different kinds of green leaves brands employed in a popular Argentinean beverage-mate-was performed with an electronic nose (EN) based on quartz microbalance sensors. We employed a LibraNose from Technobiochip, which contains eight sensors. Two types of 'odours' are analysed for each sample corresponding to short-and long-term periods after the opening of the package. Principal component analysis (PCA) is applied to the sensor responses and it is possible to distinguish between different brands through their locus in the plane defined by the first two PCA components. Preliminary results confirm the usefulness of this EN for classification and future quality control purposes.