Identifying Whiskey Counterfeits

The mid-IR region is the best spectral region for identification of materials. Since this region is greatly obscured in whisky by water, it is difficult to use for determination of other ingredients. It is possible to decrease the intensity of the water bands (by working in the near-IR region), but this would also sacrifice the intensity from the other ingredients. So how can accurate identification compounds within predominately aqueous liquids be achieved?

Image Credit: PerkinElmer
Figure 6: Spectra of whisky residue (top) and ­caramel sample (bottom).

Using a heated ATR accessory (heated to 65 degrees Celsius) allows for the evaporation of the water and ethanol to leave behind a residue of the other ingredients within the whisky. This residue can then be analyzed without the strong inferences afforded by the present of water and ethanol. The residue spectrum obtained appears very similar to the spectrum of the caramel ingredient used in these whiskies as shown in Figure 6.

The addition of E150 a caramel is permitted within Whisky legislation to achieve consistency of color within whiskies. The ATR technique should be capable of identifying the type of caramel additive used.

Study of Whisky Blends

As IR sampling and measurements are easy and fast, the technique lends itself to rapid screening testing. The ATR spectra of a series of commercially available blends were measured to determine if it would be possible to differentiate the blends using their spectra, Figure 7.

Image Credit: PerkinElmer
Figure 7: Black – Blend A. Red – Blend B.
Green – Blend C. Blue – Blend D.

As seen in Figure 7, the spectral shapes are similar, but the overall intensities differ among blends. This suggests different amounts of caramel and/or other dissolved non-volatiles are present in the different blends. Blends A and B differ only by alcohol content. Blend C and Blend D appear the most similar. Applying chemometrics to the spectral data would allow for qualitative identification of the blends. A soft independent modeling by class analogy (SIMCA) algorithm has been applied to this blend data and shows that the different blends separate out into different classes of materials within the model, see Figure 8.

Image Credit: PerkinElmer
Figure 8: SIMCA plot showing separation of the whisky blend samples.

Measuring the different blends by UV-visible spectroscopy (UV-vis) showed very similar results to those generated by IR; with different absorbance intensities recorded, see Figure 9, which also translated into separate groups when the SIMCA plot was applied. UV-vis could offer a potentially faster route to results, but its real advantage over IR is in the measurement and quantification of sample color. Color is often added to counterfeit spirits and the ability for UV-vis to differentiate between real and fake products provides a valuable tool in the arsenal to fight fraud.

Image Credit: PerkinElmer
Figure 9: Second derivative spectra of Blends A (black), B (red), C (green), D (blue).

As seen in this article, it is possible to distinguish between different blends of Scotch whisky with simple ATR IR measurements. Utilizing the same methodology, it’s also possible to differentiate between French, Scotch, and Spanish whiskies, and whisky from other non-Scotch spirits. Accordingly, it’s feasible to distinguish samples that have been diluted with water and or ethanol, offering a robust solution for the authentication of whisky.

Dr. Vosloo is the senior leader of strategy and global applications at PerkinElmer. Reach her at [email protected].