''Fake whisky'' says Prof. Dholakia, ''refers to the illegal manufacture of drinks that are sold as authentic single malt or blended whiskies. Fake whisky not only takes away money from the legitimate whisky market, it also poses potential risks to human health as there is no control over the additives or other compounds added to the counterfeit drinks available. Authentic Scotch whisky must include at least a 40% alcoholic strength by volume and can take upwards of 10 years and beyond before sales. Due to the large market and natural expense for such a specialist spirits, there is an opportunity for counterfeiters to create drinks that appear similar, but at a faction of the time and cost."
There are many types of whiskey that have a lower market value than Scotch whisky, and as such, the cheaper tipple could be counterfeited and sold as the more expensive drink. A keen whisky connoisseur can often taste the difference, but is there a way to tell the difference without taking a dram from every bottle? Professor Dholakia explains: "In principle, yes. Our method uses both Raman and fluorescence signatures. It is, in fact, largely the fluorescence signal used for distinguishing whiskies, based on data taken so far using Raman spectroscopy, enabling us to determine the alcohol content."
As is the case with other spectroscopic analysis, a database of results is required to fingerprint authentic Scotch whisky over lower value substitutes. “A database needs to be created," says Prof. Dholakia, "in order to train the chip to know if an unknown whisky sample is inserted." Japanese whisky is distilled in pot stills, similar to Scotch whisky; is there a way to successfully differentiate whiskies that are so similar? "Interestingly," Prof. Dholakia says, "we have differentiated some Japanese whiskies from Scotch whiskies, which shows great promise, and we can also tell the difference between malt and blended Scotch whisky."
Prof. Dholakia uses "Near infra-red spectroscopy on an optofluidic microchip, the size of half a credit card, for quality monitoring of single malt Scotch whisky. Fiber optics deliver and collect the length from the chip. Analysis on this alignment-free, portable chip may be performed in only two seconds with a sample volume of only 20 μl (the size of a teardrop). Using a mathematical approach (partial least square (PLS) calibration), we have demonstrated that the alcohol content in the beverage may be predicted to within a 1% prediction error. A statistical approach [principal component analysis (PCA)] was employed for successful classification of whiskies based upon their age, type and cask." This technology can also be translated to other drinks, and Prof. Dholakia is currently performing research on vodka and wine.
The future and handheld devices
Raman spectroscopy could have huge benefits industrially if it could be translated into a handheld device. "At the moment we have compressed the whole device to a footprint of around 0.5 m x 0.2 m x 0.35 m," Prof Dholakia describes. "It is very challenging, but possible, with advances in spectrometer technology and new approaches to hold and illuminate the whisky sample. We hope our method may be used for whisky authentication analysis on site, and we hope to assist this industry to ensure we are can all enjoy such spirits safe in the knowledge that we are drinking a real malt or blend."