E-noses have grown in popularity in recent years as a result of their wide range of useful uses, which include sensing when crops are ready for harvest to identifying food products that are about to expire. It’s somewhat unexpected that numerous e-noses have been designed to assess alcoholic beverages, such as whiskey, which had a global market worth an estimated $58 billion in 2018.
With just a smell of a dram, a whiskey enthusiast can identify the brand, area, and style of whiskey at hand. But, in terms of distinguishing the attributes of a whiskey, how do our human noses compare to electronic nostrils?
A novel e-nose that is astonishingly accurate at assessing whiskies—and can identify the brand of whiskey with more than 95 percent accuracy after only one “whiff”—was described in a study published in IEEE Sensors Journal on February 1st.
“This lucrative industry has the potential to be a target of fraudulent activities such as mislabeling and adulteration,” explains Steven Su, an associate professor at the Faculty of Engineering and IT, University of Technology Sydney. “Trained experts and experienced aficionados can easily tell the difference between whiskies from their scents. But it is quite difficult for most consumers, especially amateurs.”
As a result, Su and his colleagues decided to modify one of their e-noses to examine some critical characteristics of whiskey. Their first e-nose was created to detect illegal animal parts traded on the black market, such as rhino horns, and they’ve since adapted it for breath analysis and food quality assessment.
E-Nose, their newest e-nose for sniffing whiskey, has a small vial where the whiskey sample is added. The whiskey’s smell is injected into a gas sensor chamber, which detects the different odours and delivers the information to a computer for analysis. Machine learning algorithms are then used to extract and analyse the most essential fragrance components in order to recognise the brand, region, and type of whiskey.
The researchers put Nos.e to the test by analysing a panel of six whiskies, ranging from Johnnie Walker Red and Black label brands to Macallan’s 12-year-old single malt Scotch whisky. They also used a state-of-the-art gas chromatography (GC) instrument to evaluate the whiskey samples for comparison.
The results, which indicated that Nos.e was on par with the GC gadget in terms of detecting whiskey brands, astounded Su’s team.
“Given the average accuracy of over 95%, the e-nose [could possibly] be trained to assess the quality of other popular liquors that can be counterfeited, such as Australian red wine, Japanese whisky, or Chinese white wine,” Su says.
Some liquor businesses are thinking along the same lines. Su claims that a Chinese liquor business has approached his team about applying the technology to their drinks, albeit the proposal has been put on hold for the time being owing to the pandemic.
Su also points out that his team’s new e-nose could not be up to pace with other analytical approaches like computer vision analysis. “The features of present gas sensors, which are sensitive to environmental changes such as humidity and temperature,” he continues, “create this constraint [with e-noses].” “As a result, we intend to apply algorithms to address these issues in the future.”