Alcohol types

e-Nose to distinguish different types of premium alcoholic beverages

Irish or Scottish, blended or single malt? While a whiskey lover may be able to tell a good drink from a mediocre drink just by smelling it, most drinkers depend on the label.

The electronic nose was tested and presented at CEBIT Australia. Image credit: UTS

Whiskey is one of the most sought-after alcoholic beverages internationally and, with some top brands costing five or six figures, it’s also a prime target for deception.

To combat this, scientists have created an electronic nose (e-nose) capable of differentiating between different brands, styles and origins by “sniffing” alcohol.

The study was led by Associate Professor Steven Su, who holds a Ph.D. students Taoping Liu and Wentian Zhang, from the University of Technology Sydney (UTS), in partnership with chemists Dr Maiken Ueland and Professor Shari Forbes.

Until now, detecting differences between whiskeys required either a trained whiskey connoisseur, who could still be wrong, or complex and time-consuming chemical analysis by scientists in a laboratory. So, having a quick, easy-to-use, real-time whiskey assessment to identify quality and uncover any tampering or fraud could be of great benefit to wholesalers and high-end buyers.

Associate Professor Steven Su, Study Leader, University of Technology Sydney

Researchers used a new prototype electronic nose (called NOS.E), created at UTS, to recognize differences between six whiskeys by their brand names, styles and regions in less than 4 minutes.

The experiment took samples of three single malt whiskeys and three blended malts, including Chivas Regal, Ardberg, Johnnie Walker Black and Red label whiskey and a 12-year-old Macallan’s whiskey.

The study recently appeared in the journal IEEE sensors. It demonstrated that the electronic nose achieved 100% accuracy for region identification, 96.15% accuracy for brand name, and 92.31% accuracy for style.

NOS.E is designed to mimic the human olfactory system, with eight gas sensors to identify odors in a container of whiskey. The sensor array produces the proprietary signal matrix based on the different odor molecules it is exposed to.

It then transmits the information to a computer for review, with a machine learning algorithm programmed to identify the whiskey’s characteristics.

The scientists verified the NOS.E results using advanced laboratory tests on the whiskey samples: time-of-flight mass spectrometry coupled with two-dimensional gas chromatography, which produced similar results.

This technology is not only used in the alcohol sector, with drinks such as cognac and wine in addition to whiskey, but also for other products facing counterfeiting such as high-end perfume.

Electronic nose technology has also been used to identify prohibited animal parts traded on the black market, for example black rhino horns, and has considerable potential for disease detection and health applications.

Journal reference:

Zhang, W. et al. (2022) The use of the electronic nose for the classification of blended and single malt Scotch whisky. IEEE sensors.