How Scientists Assess the Age and Authenticity of Old Whisky

Scientists assess old whisky by testing when its alcohol originated and whether its chemical composition fits the whisky it claims to be. Radiocarbon dating can narrow the distillation period for some whiskies, while chemical analysis can expose counterfeit contents; neither should be treated as a universal test that reads an age statement or settles every question of authenticity.[1][8][10]

Two questions, not one

The first question is chronological: does the alcohol carry a carbon signature consistent with the claimed distillation period? The second is compositional: does the sample contain the compounds, or combinations of compounds, expected in authentic whisky? The Scotch Whisky Research Institute, or SWRI, describes a suite of authentication methods rather than a single decisive instrument. Its approach includes tests for alcohol strength, sugars, flavourings and compounds associated with oak maturation.[1][10]

That distinction matters when interpreting a report and assessing warning signs and authentication limits. A radiocarbon result addresses the alcohol's date signature; a chemical profile addresses its composition. Neither result should be expanded into a claim that the method was not designed to establish. SWRI's warning that there is no “one tool fits all” approach is a useful starting point.[1][10]

How radiocarbon dating works

Barley takes up radiocarbon as it grows, leaving a date signature in the whisky made from it. Atmospheric nuclear testing from the 1950s increased radiocarbon levels; after the 1963 test-ban treaty, those levels declined. This rise and fall provides a reference against which scientists can compare a whisky sample.[2]

Gordon Cook, Elaine Dunbar, Brian Tripney and Derek Fabel at SUERC published a whisky-specific calibration curve in 2020. Their study used 221 single malt and new make spirit samples of known distillation year, covering 1950 to 2015. Diageo, Edrington and Rare Whisky 101 supplied samples; the authors reported no external funding.[1]

A calibration curve connects the measured carbon signature with possible dates. For certain periods from 1955 onwards, the authors report dating precision of 1 to 3 years. That is a conditional result, not a promise that every old whisky can be assigned a single year, and it does not apply to pre-1950 samples.[1]

There is another wrinkle: the curve often produces two possible date ranges. The researchers found that plotting another carbon isotope measurement, δ13C, against distillation year could often distinguish between them. Their whisky curve also runs about one year behind the Northern Hemisphere NH1 bomb curve, which they attribute to the interval between barley harvest and distillation.[1]

For whisky from before 1950, the paper uses five-year IntCal13 calibration data, with much broader dating results than those available for suitable nuclear-era samples. The physical Carbon-14 half-life used in the paper is 5,730 years. The power of the later calibration comes from the atmospheric nuclear-testing signature, not an ability to count individual years of cask maturation.[1]

From a small sample to a date

Radiocarbon analysis requires liquid from the bottle. Cook described taking about one millilitre with a hypodermic needle. In the published SUERC method, the laboratory redistils the whisky to remove oak-derived components and concentrate ethanol before preparing its carbon for measurement by accelerator mass spectrometry. These are specialist laboratory procedures, not a home authentication method.[1][4]

Previous cask contents might seem an obvious complication. The SUERC authors note that casks can retain 2 to 10 litres of their previous contents, but found the effect on apparent age “negligible”. That finding belongs to their dating research: it should not be turned into a general claim that previous contents have no effect on whisky.[1]

Carbon isotopes can also help answer a different question about ingredients. The paper describes δ13C as a way to separate C3 grains, including barley and wheat, from C4 plants such as maize and cane. This is evidence about the source of the alcohol, distinct from the radiocarbon evidence used to assess its date.[1]

What chemical authentication tests look for

SWRI's published account of routine authentication shows why laboratories use several techniques. Some measurements screen for conspicuous inconsistencies; others examine particular components in more detail. Alcohol strength is one part of that work, not a certificate of authenticity. The institute also describes confirmatory tests, including trace metal analysis, for brand-related investigations.[10]

These abbreviations describe different analytical techniques, rather than competing verdicts on a bottle. SWRI's examples assign them specific tasks: detecting added sugars is not the same job as examining oak-derived compounds. Its account also describes a UV-Vis instrument as very good at identifying gross counterfeits, but explicitly treats it as a screening tool.[10]

Can scientists test whisky through the bottle?

Yes, some research methods can examine whisky through glass. A 2019 study by Manchester, Liverpool and SWRI researchers used handheld Raman spectroscopy with a 1064 nm laser to investigate fake spirits and measure methanol. SWRI supplied 144 blind-coded samples, including 40 counterfeits. Results varied across the seven commercial bottles tested, showing that container effects can limit the method.[6]

In 2020, University of St Andrews physicists reported an axicon-based Raman system designed to read whisky through glass while avoiding fluorescence from the bottle wall. The significant point for an old, sealed bottle is that this approach does not require opening it. It remains a different kind of investigation from radiocarbon dating of a liquid sample.[7]

A separate Strathclyde study in 2011 used mid-infrared spectrometry with fibre-optic probes on 17 blended whisky samples. Measurements of ethanol and colourant correctly separated all eight authentic and nine counterfeit samples in blind tests. That is a clear result for the samples studied, not evidence that every counterfeit can be recognised with the same success.[5]

Why recognition is not an age statement

Nuclear magnetic resonance, or NMR, offers another view of whisky's chemistry. In 2019, Edinburgh and SWRI researchers studied 148 authentic samples obtained directly from producers and 32 known counterfeits using proton NMR. They identified 25 compounds in Scotch whisky. Counterfeits contained added sugars, glycerol and vanillin at levels or ratios not found in the authentic samples.[8]

Yet the same study's models for age and origin were not statistically significant, with p values of 0.244 and 0.263 respectively. That distinction is central to reading authentication research: a method can reveal suspicious composition without reliably establishing how old a whisky is or where it came from. Those are separate claims requiring their own evidence.[8]

The much-reported “artificial tongue” illustrates the same need for precision. Glasgow and Strathclyde engineers developed a gold and aluminium sensor using plasmonic resonance. The university reported greater than 99% accuracy in distinguishing whiskies from Glenfiddich, Glen Marnoch and Laphroaig, and separation of the same whisky at 12, 15 and 18 years.[9]

That finding concerns discrimination between the whiskies tested. It is not a general demonstration that an instrument can read any bottle's age statement. Read alongside the NMR study, it shows why the test population and the precise question matter as much as an impressive accuracy figure: recognition, dating and authentication are not interchangeable tasks.[8][9]

How to read a laboratory conclusion

Start with what was measured and what the result actually supports. For a radiocarbon report, look for the possible date ranges and the calibration used. For a chemical study, look for the authentic comparison samples, the counterfeit samples and the specific differences detected. These are the foundations of the published work, rather than decorative technical detail.[1][8]

Keep the conclusion at that level. A narrow date range is not available for every period; successful classification of a research sample set is not universal proof of identity. SWRI's combined approach provides the clearest practical lesson: the question determines the tests, and the evidence from those tests determines how firmly an authenticity claim can be made.[1][5][8][10]

Where to go next

• Fake Whisky: Warning Signs and the Limits of Authentication: for the broader warning signs and the limits of a verdict.
• Famous Whisky Fakes: How They Were Exposed: for documented cases and the evidence behind their exposure.
• Whisky Provenance and Documentation: for the documentary side of an authenticity assessment.

Sources

[1] Cook, Gordon; Dunbar, Elaine; Tripney, Brian; and Fabel, Derek, SUERC. Using Carbon Isotopes to Fight the Rise in Fraudulent Whisky. Radiocarbon, volume 62, issue 1, 2020.
https://eprints.gla.ac.uk/204495/7/204495.pdf

[2] University of Glasgow. Fake whisky exposed, headline abbreviated to comply with site policy, 20 December 2018.
https://www.gla.ac.uk/news/archiveofnews/2018/december/headline_627779_en.html

[4] South China Morning Post. Around 40 per cent of vintage whisky bottles are fake, but now the Verifier app and an artificial tongue can date an old Laphroaig, Talisker or Ardbeg single malt accurately. Publication year not established in the supplied fact sheet.
https://www.scmp.com/magazines/style/leisure/article/3095524/around-40-cent-vintage-whisky-bottles-are-fake-now-verifier

[5] ScienceDaily, reporting University of Strathclyde research. How to tell real whiskey from fake, faster. 25 July 2011.
http://www.sciencedaily.com/releases/2011/07/110725123709.htm

[6] Ellis, Muhamadali, Xu, Eccles, Goodall and Goodacre. Rapid through-container detection of fake spirits and methanol quantification with handheld Raman spectroscopy. Analyst, volume 144, 2019.
http://biospec.net/pubs/pdfs/Ellis-Analyst2019.pdf

[7] Fleming, Chen, Bruce and Dholakia, University of St Andrews Research Portal. Through-bottle whisky sensing and classification using Raman spectroscopy in an axicon-based backscattering configuration. Analytical Methods, 7 October 2020.
https://research-portal.st-andrews.ac.uk/en/publications/through-bottle-whisky-sensing-and-classification-using-raman-spec/

[8] Kew, Goodall and Uhrín. Analysis of Scotch Whisky by 1H NMR and chemometrics yields insight into its complex chemistry. Food Chemistry, volume 298, 2019.
https://www.pure.ed.ac.uk/ws/files/97956620/20190625_Uhrin_NMR_Whisky.pdf

[9] University of Glasgow. Artificial tongue could have whisky counterfeiting licked. 6 August 2019.
https://www.gla.ac.uk/news/archiveofnews/2019/august/headline_659435_en.html

[10] Scotch Whisky Research Institute, Shona Glancy. Developing rapid analysis methods in the spirit drinks sector. Presentation, 21 June 2016, hosted by GOV.UK.
https://assets.publishing.service.gov.uk/media/5a81870240f0b62305b8f6f0/Developing_Rapid_Analysis_Methods_S_Glancy.pdf

This desk-researched article was prepared from the site's verified fact sheet TOP-A549 and its cited research papers, institutional accounts and reporting.

Last checked: November 2026