Neat, or on the rocks? Or maybe just a touch of water? It’s a question long debated by whiskey lovers and one largely left to personal preference—until now. In a new study, food scientists have determined the exact point at which watering down whiskey ends up altering its unique aroma profile beyond recognition.
The study is a collaboration between five food scientists from Washington State University, Michigan State University, and Oregon State University, published in the peer reviewed, open-access journal Foods. In the study, a panel of 20 regular whiskey consumers trained for the experiment were less able to tell rye and Bourbon apart, or single malt versus blended Scotch whisky, once the whiskey was diluted with water by more than 20 percent.
The reason is something called Henry’s law, and it may contradict centuries of whiskey’s so-called best practices. How do smells work, and how do liquids like whiskey emit them?
The Whiskey Taste Test
For the study, the researchers surveyed everyday consumers about the aromas they detect from whiskey—similar to wine-tasting notes, like oak, peat, vanilla, and sawdust. From that survey, they made a list of 15 checkboxes that were offered to 20 panelists in the experiment. The panelists were trained to smell these notes as part of the experiment.
→ The 15 whiskey aromas tested: bacon, Band-Aid, brine, cedar, cornmeal/cooked polenta, hay, malt, oak, peat, smoke, pine, pome fruit, rubber, sawdust solvent/chemical, and vanilla.
There were 25 whiskey samples in the mix, including ones from the United States, Ireland, and Scotland. They were diluted to various strengths, from original 100 percent all the way down to 50 percent whiskey, 50 percent water. Each dilution was presented to each of the 20 panelists, who smelled them and checked off any of the 15 smell notes that applied.
“Smoke-related attributes of ‘Rubber’, ‘Bacon’ and ‘Peat smoke’ were highly predictive of peated single malt Scotch, while the descriptors of ‘Oak’ and ‘Vanilla’ were strong predictors of Bourbons. ‘Pome fruit’ was a strong predictor of blended Scotch whisky. In general, undiluted whiskies were more associated with ‘Chemical/Solvent’, ‘Vanilla’, ‘Oak’, and ‘Bacon’ aromas while diluted whiskies were associated with ‘Pome Fruit’, and ‘Cedar’ aromas,” the study reads.
How Whiskey Flavours and Aromas Change With Water
Smells are very contextual and reactive. To talk about why, we’ll need to get through some technical terms. The first major determination is whether molecules cling better to water or fat, something you may know about if you’ve had experience blooming different spices for cooking or even thinking about how your vitamin supplements are digested. Hydrophilic molecules love water, while lipophilic molecules love fat.
“The decrease in ‘Vanilla’ aroma with dilution would be expected due to the amphipathic nature of phenols such as vanillin and guaiacol,” the researchers suggest. Vanillin is the major component of what we think of as “vanilla” flavouring, and amphipathic, or amphiphilic, means the molecule likes both water and fat at the same time. Emulsifiers like mustard or egg yolk are amphiphilic, as are some chemicals used as part of soap.
“‘Oak’ aroma appears to evolve into a ‘Cedar’ aroma with dilution, possibly due to the hydrophobic nature of many of the oak lactones found in Bourbon,” they explain. Hydrophobic molecules don’t mix naturally with water, and will instead clump together or into friendlier other materials. In other words, the oak lactones flee from the growing amount of water and have less and less surface area from which to contribute to the aroma.
Overall, the dilution of hydrophobic (water-hating) or lipophilic (fat-loving) molecules explains a lot of why the aromas changed so noticeably. The whiskey varieties started to merge together starting at just 80 percent whiskey, 20 per cent water, as the non-hydrophilic (water-loving) molecules were pushed aside and crowded out. But whiskey is already mostly water, isn’t it? Why is this seemingly not-that-large increase in water content such a dealbreaker?
How Henry’s Law Applies to Whiskey
Henry’s law is a chemistry law about the amount of different gases—in this case, aroma compounds—that are found dissolved in and floating above a liquid. The space in your glass above your whiskey is called the “headspace,” and the proportion of a gas dissolved in a liquid is directly proportional to the amount of those gases that swirls in the headspace.
Whiskey may already have water in it, but that water content has been steeping in the same wooden barrel or other environs as the rest of the whiskey; it’s chockablock with water-loving aroma compounds. Then, whiskey is diluted from its “cask strength” of around 55 to 65 percent alcohol down to about 40 percent alcohol, which distillers have found is a sweet spot for what consumers actually like to drink. If adding more water was overall perceived by distillers to be a good thing, they surely would: they could sell even more bottles from the same cask size.
With all that in mind, it makes sense that a water content of just 20 percent is the suggested maximum. It could also be that people who prefer whiskey with water find it easier to enjoy, and therefore savour the aromas and flavour compounds, when the harsh alcohol flavour is diluted a bit more. (When whiskey is mixed with other, well, mixers—like ginger ale, vermouth, or even just soda water—the aromas are changed by additions, not simply diluted.)
So, What’s the Best Ice For Whiskey?
The corresponding author on this study, the improbably named Thomas Collins, says the study gives credence to people who have recommended using a single extra-large ice cube in whiskey. “This study helps to understand why those large, square ice cubes have become so popular because you can actually enjoy the whiskey before it gets diluted to the point that it’s not the same whiskey,” Collins said in a Washington State University press release.
An ice cube that’s three inches per side (imagine with me here!) has 54 square inches of surface area. The same amount of ice but in one-inch cubes, for a total of 27 cubes, has 162 square inches of surface area—that’s three times as much. So the big cube is better for dilution, but it’s still not the best best bet. That’s actually a sphere of ice which, with the same volume of 27 cubic inches, has a measly 43 square inches of surface area.
