Thomas Collins,* Caroline Merrell, and James Harbertston 
*Viticulture and Enology Program, Washington State University,
359 University Drive, Richland, WA 99354 (tom.collins@wsu.edu)

The hydrophobicity of a set of 15 bourbon whiskeys and 16 Scotch whisky samples was determined by octanol-water partitioning modified from published methods (Mueller-Harvey et al. 2007, Zanchi et al. 2009). Whiskey, water, and octanol were combined in a 15 mL centrifuge tube and mixed for 30 sec. The tubes were centrifuged to separate the layers and an aliquot of each layer was analyzed via HPLC. HPLC analysis was carried out on an Agilent 1200 HPLC system (Santa Clara, CA). Detection used a diode array detector at 280 nm; total peak area was determined for each phase. The octanol/water partition coefficient (Kow) was calculated for each whiskey by dividing the total peak area of the octanol phase by the peak area of the water phase. The partition coefficients for the bourbon whiskeys spanned a narrow range between 1.3 and 1.6. The partition coefficients for Scotch whiskies were more variable, but generally lower than for the bourbon whiskeys. The composition of the water and octanol phases was also characterized using UHPLC/QTOF-MS. This analysis was conducted on an Agilent 1290 UHPLC/6545 QTOF-MS system using electrospray ionization (ESI) in negative mode. The MS data was analyzed using the Agilent MassHunter and Mass Profiler Professional software. The water phase contained lignan glycosides, while the octanol phase was characterized by lipids, phenylpropanoids, lignans, and triterpenoids. Triterpenoid glycosides were present in both phases. Using principal component analysis, bourbon and Scotch whisky samples were readily differentiated based on the composition of their octanol phases. Both the partition coefficients and the composition of the two phases can be related to the use of Sherry butts versus American oak casks and the number of fills for American casks.

Funding Support: WSU start up funds