Whiskey is a distilled spirit beverage made from grain-based fermentations or mashes, produced in regions where grain is grown commercially. The price of whisky can range from tens to thousands of dollars for a 750 mL bottle depending on a variety of factors such as the quality of grains used, and variations applied during the distillation process. Due to the value of high-end quality bottles, they are often the target of fraud.
Vodka and neutral grain spirits are distilled to 190 proof (95 % v/v ethanol) to produce a spirit that has minimal character (hence the name “neutral”), while whiskies are generally distilled to 160 proof (80 % v/v) to retain some character from the grains and from fermentation1. Additionally, whiskies are generally expected or required to be aged in wooden casks for a minimum period of time, often two or three years, and may be aged in cask for 10 to 15 years or longer. While in cask, the whiskey will take on aromatic and flavored compounds from the wood. As the casks provide a somewhat oxidative environment, a range of oxidation reactions can occur involving compounds present in the new spirit as well as compounds extracted from the cask2. Much of the character of the final spirit comes from the imparted compounds and reactions that occur during the cask aging of the spirit.
Whiskey analysis; quality assurance, process improvement, and authenticity verification
Given the complexity of both the distillation and aging processes, there has always been a demand for analytical techniques to monitor and help us to improve these processes. Large distilleries have routinely used gas chromatography (GC) coupled with flame ionization detection (FID) and, more recently, with mass spectrometry (MS), to analyze newly distilled spirits for quality assurance purposes3. GC has also been applied to the analysis of aged whiskies, both for quality assurance and process improvement, but also as a tool for verifying the authenticity of aged spirits.
Given the value of high-quality aged whiskies in the marketplace, these products are frequently the target of a range of fraudulent activities. A common type of fraud is whiskey substitution; the refilling of more expensive, authentic bottles with younger, lower quality whiskies that are cheaper and more readily available. While the average drinker may not recognize the fraud, particularly when the whiskey is in a mixed drink, more discerning whiskey consumers are often able to detect poorly made or fraudulent products.
A range of approaches are used for the detection of fraudulent whiskies, many of which rely on the comparison of a genuine example of the product with the suspected fraudulent product. GC-FID and GC-MS profiles can be used for this purpose, as can UV-Vis spectrophotometric techniques. The State of New Jersey, for example, conducted a series of raids on restaurants and taverns in 2013 in response to a number of consumers alleging that fraudulent activities were taking place in these establishments (Operation Swill)4. The regulatory agencies used a portable UV-Vis spectrophotometer, comparing spectra from suspected bottles with authentic spectra results from the onboard library. Products which failed this field test were then subjected to further analysis using additional techniques. For example, precise isotope ratio measurement of elements such as carbon or oxygen with a specialized mass spectrometer, can reveal the geographic location where a whiskey was produced as these ratios vary slightly among regions. Electrospray ionization Fourier transform ion cyclotron mass spectrometry (ESI-FT-ICR/MS) is a high-resolution MS technique requiring long spectral acquisition times and providing a chemical profile that can be used for authentication.
Developing a new analytical approach
As much of the character of a well-aged whiskey is derived from components extracted from barrels, including less-volatile components not readily analyzed using GC-MS techniques, we developed a new approach using ultra-high-pressure liquid chromatography (UHPLC) coupled with accurate mass quadrupole-time-of-flight mass spectrometry (QTOF-MS) along with multivariate statistical tools for the comparison of results across the sample sets5. The advantage this method offers over the previously used ESI-FT-ICR/MS, is that the chromatographic separation ahead of the mass spectrometer provides the ability to differentiate among compounds which have the same mass-to-charge ratio (m/z), but different retention times (RT), which is important for the characterization of these products. Total per-sample analysis time is twelve minutes, which compares favorably with the analysis times for many of the GC-MS profiling methods available.
A comprehensive software suite is essential to the application of this method given the complexity of the data sets generated by the analysis of the large number of samples required to adequately characterize the range of whiskies currently in the marketplace. To maximize the advantages of the additional information about isomers provided by the chromatographic separation, mass and RT, data must be aligned across the samples in the sample sets, using appropriate mass and RT windows to allow for expected variation in those parameters.
Aligned mass and RT (molecular feature) data sets can be further evaluated using multivariate statistical techniques including principal component analysis or discriminant analysis to explore the relationships among samples and compounds which are grouped together in each data set. These techniques can provide insight into which compounds are characteristic of particular whiskey types. Using LC/Q-TOF MS to profile whiskies for North America and Scotland, distinction between these spirits are clearly demonstrated (Figure 1). Soft modeling techniques such as SIMCA (soft independent modeling by class analogy) can also be applied to evaluate whether a sample of unknown origin is consistent with authentic examples of whiskies.
Figure 1: Principle component analysis (PCA) of the aligned chemical profiles in Bourbon whiskies. Credit: Adapted from reference 5, property of Agilent.
Just as libraries of authentic samples have been generated for GC-MS and UV-Vis spectroscopy, this new UHPLC/Q-TOF MS technique provides an orthogonal approach to analysis of whiskies. Compared with GC-MS , it focuses on less-volatile wood-derived components (from casks) in the product rather than volatile composition, while providing more detailed information about composition than can be generated using UV-Vis spectroscopy. In addition to the potential of this approach for the detection of fraudulent whiskies, these techniques can be applied to process improvement research at commercial distilleries, offering an objective approach to the evaluation of in-house research projects including, for example, cask-aging studies, blending trials, and consumer sensory projects, among others.
1. CFR; U.S. Code of Federal Regulations, Title 27 Part 5.22; (http://www. ecfr.gov/cgi-bin/text-idx?SID=422646116a9adddfbe43ec813964aaec&node=27:18.104.22.168.22.214.171.124&rgn=div8).
2. MacNamara, K., Dabrowska, D., Baden, M., & Helle, N. (2011). Advances in the ageing
chemistry of distilled spirits distilled in oak barrels. LC/GC, 14(3), 6–22.
3. Saxberg, B. E. H., Duewer, D. L., Booker, J. L., & Kowalski, B. R. (1978). Pattern recognition and blind assay techniques applied to forensic separation of whiskies. Analytica Chimica Acta, 103, 201–212.
4. Twenty-Nine Establishments Raided by Division of Alcoholic Beverage Control in “Operation Swill” for Allegedly Switching Premium Alcohol for Cheaper Substitutes (https://nj.gov/oag/newsreleases13/pr20130523b.html).
5. Collins, T.S., Zweigenbaum, J., and Ebeler, S.E. (2014) Profiling of non-volatiles in whiskeys using ultra high pressure liquid chromatography quadrupole time-of-flight mass spectrometry (UHPLC-QTOF MS). Food Chemistry, 163, 186-196.