Quick insights into whisky - investigating rapid and efficient methods for sensory evaluation and chemical analysis

Abstract

The evaluation of aroma properties of beverages, both analytically as well as with human sensory studies, is a challenging task and most often related to time- and cost-intensive analyses. Whisky is a spirit offering a wide variety of aroma impressions caused by a complex mixture of aroma active compounds. In the present study, methods for the efficient evaluation of aroma characteristics are evaluated for 16 whisky samples of different origins (Scotch and American). Rate all that apply (RATA) was applied as a rapid method for the sensory evaluation of whiskies. Sensory evaluation of the samples led to the determination of eight significant aroma attributes: caramel-/cream caramel-/toffee-like, vanilla-like, (canned) peach-like, phenolic, smoky, fruity, flowery and (fermented) apple-/cider-like. Chemical analysis was conducted by stir bar sorptive extraction (SBSE) coupled to gas chromatography-mass spectrometry in combination with an in-house data processing tool for semi-automated analyte detection. Through chemical analysis of the whisky samples and automated compound detection, we report over 200 mostly aroma-active volatiles. To test both approaches for their potential for sample classification, a simple classification problem (Scotch vs. American) was applied. Linear discriminant analysis (LDA) indicates both that sensory evaluation by RATA (97.86%) and the applied analytical procedure (96.94%) are suitable for the distinction between the two whisky types. In both approaches, potential markers were determined for the classification. These investigations build a solid foundation for the implementation of a versatile platform facilitating rapid and efficient aroma evaluation in various foodstuffs and beverages.

Keywords: Automated compound detection; Classification; Gas chromatography-mass spectrometry; Human sensory; Whisky.

Fig. 1

Sensory mapping of 16 whisky (9 Scotch, red and 7 American, blue) samples at two different ABV levels (20%, dot and 40%, circle) based on the evaluation of significant sensory attributes (main descriptors, black circles)

Fig. 2

Linear discriminant analysis of the sensory data of American and Scotch whisky samples at 20 % and 40 % ABV (alcohol by volume). Left: relevance of single sensory attributes as predictors for the categorization. The delta predictor indicates the influence of a single attribute on the differentiability between classes. Right: differentiation between Scotch and American whisky. Validation accuracy shows one value for one set of 13 samples (12/13 correctly predicted equals 92.3%, etc.) in 5000 training repetitions with cross-validation. Test accuracy shows the same for 5000 repetitions of testing on a set of 3 samples (e.g., 2/3 correctly predicted equals 66.6% test accuracy)

Fig. 3

Correctly assigned analytes in model whisky 1. All analytes designated as “best match” present in model whisky 1 as well as one analyte suggested as 2^nd best match are shown together with the respective 2^nd best matches, chromatogram was considered before blank subtraction. If analytes were detected at several retention times, matches with highest peak areas were considered. Analytes present in the model whisky that have not been detected, are indicated at the bottom

Fig. 4

Determination of potential predictors for the classification of Scotch and American whisky based on the delta predictors of an LDA using all detected whisky compounds. A total of 10 potential predictors were determined: (E)-2-octenal, (E)-dec-2-enal, eugenol (2-methoxy-4-prop-2-enylphenol), whisky lactone (5-butyl-3-methyloxolan-2-one), β-damascenone ((E)-1-(2,6,6-trimethylcyclohexa-1,3-dien-1-yl)but-2-en-1-one), 3-methylbutyl decanoate, citronellol (3,7-dimethyloct-6-en-1-ol), menthol (5-methyl-2-propan-2-ylcyclohexan-1-ol), myrtenol ((6,6-dimethyl-2-bicyclo[3.1.1]hept-2-enyl)methanol), and methyl decanoate

Fig. 5

Linear discriminant analysis for the classification of 16 whisky samples based on analytical data, 5000 repetitions using a training set of 13 and test set of 3 samples. A without and B with preceding PCA keeping only 4 main components, leading to accuracies of 80.74% and 96.98%, respectively

Fig. 6

Linear discriminant analysis for the classification of 16 whisky samples based on analytical and sensory data in combination, 5000 repetitions using a training set of 13 and test set of 3 samples. Indication of potential predictors based on chemical analysis ((E)-2-octenal, (E)-dec-2-enal, eugenol (2-methoxy-4-prop-2-enylphenol), whisky lactone (5-butyl-3-methyloxolan-2-one), β-damascenone ((E)-1-(2,6,6-trimethylcyclohexa-1,3-dien-1-yl)but-2-en-1-one), 3-methylbutyl decanoate, citronellol (3,7-dimethyloct-6-en-1-ol), menthol (5-methyl-2-propan-2-ylcyclohexan-1-ol), myrtenol ((6,6-dimethyl-2-bicyclo[3.1.1]hept-2-enyl)methanol) and methyl decanoate), and sensory evaluation (caramel, peach)