Single Cask Collection | 66.7% ABV

What’s the highest strength whisky you’ve tried?

I didn’t need to add another bottle to my collection a year ago, but when I saw the %ABV or proof of this sherry nuke, I knew I needed to have it. I couldn’t let the chance to try some non-bourbon uber-high Scotch malt whisky pass me by. The fact that it was an unpeated Bunnahabhain had me sold. Take my money.

Then I got thinking… in Scotchland, the angel’s share usually drops the %ABV because the relatively high atmospheric humidity forces the bulk of the losses to occur from the alcohol. In drier climates, the opposite is generally true where water evaporates from the cask and the alcohol remains.

So, how did this 12yo end up higher than the typical industry-standard 63.5 %ABV filling strength for most barrels in Scotland? Was it filled higher than the norm? Did it lose water and thereby increase in strength? If it was the former option and if it was filled higher, how would that affect the spirit?

To answer these questions for myself, I undertook a mini literature review. Now fair warning here, I’ve written an engineering doctorate thesis and authored many technical peer-reviewed journal articles in my professional life so the following may read quite dry and technical, but it’s the way my brain processes this information and stores it.

Don’t fret though, I will do my best to give a TL;DR synopsis at the end if you want to Skip to Go and collect your $200 on %ABV’s impact on barrel extractives.

It’s important to set the stage here as we will be talking about extraction of compounds from the barrel, so let’s go over a short version of barrel making so we are on the same page. Oak casks are toasted, charred, or both during the coopering process. These high temperature processes fundamentally alter the relatively unappealing and tasteless wood structure/compounds into tasty or repulsive compounds.

The cellulose and hemicellulose sugars caramelise (caramel, butter, candy notes) and break down into various other compounds (typically nuts) while the sugar-less lignin produces spicy, vanilla, smoky, medicinal, fruity, and “green” notes. These congeners, born by way of flame and heat, are able to be extracted by both water and alcohol; however their extraction rate tends to be unique to the chemical compound and the ethanol/water ratio (ie., filling strength).

“According to the experience of some distilleries, the lower the strength, the sweeter and smoother the taste of a whisky. On the other hand, the higher the alcohol content, the woodier and spicier the taste. A higher filling strength also results in a lower content of colour and volatile acids in the matured whisky later on. In addition, very high-proof spirits mature more slowly… Several investigations with different filling strengths led to the results that colour, solids, volatile acids, fusel oils and tannins decrease with increasing filling strength. One exception worth mentioning is the fruity esters, which remain constant over the maturation period and do not seem to be influenced by the filling strength.”
- Weinberger [1]

Alright, armed with a foundation of cask compounds, let’s see what the literature might suggest.

Clyne et. al studied the effect of 1989-distilled Longmorn malt whisky at two filling strengths (67.5 and 63.4 %ABV, apparently two common strengths at the time) in both uncharred and charred American oak barrels, and sampled at 3-to-36 mo of age at 6 mo intervals [2]. Triplicates were utilised to temper the effects of barrel variations (12 barrels total), with gas and liquid chromatography and a 20-person sensory panel utilised to determine any variations.

No significant effect on volatile compounds was observed with barrel char, filling strength, and maturation length. For non-volatile compounds, the charred wood had significantly higher phenols and syringaldehyde (wood spice) than the un-charred barrels. The effect of ethanol concentration (strength) was not clear or consistent, outside of a decrease in gallic acid (bitter or drying effects) and vanillic acid (sour) extraction at the lower strength while an increase in vanillin (vanilla) was observed at higher strength. The sensory panel found that barrel filling strength made virtually no discernable difference, with the exception that the high strength un-charred barrel samples were consistently ranked lower.

In an adjacent study, Connor et. al and Piggott et. al studied the effect of 63.4 and 67.5 %ABV filling strengths of Longmorn single malt on a first fill ex-bourbon cask (previously held bourbon), second fill ex-bourbon cask (previously held Scotch whisky), and an exhausted cask after 36 months of maturation [3,4]. In a first-fill scenario, the higher filling strength extracted more vanillin (vanilla), gallic acid (bitter), vanillic acid (sour), and interestingly less syringaldehyde (wood spice, wood smoke) compared to the lower strength. A second-fill barrel, which is now the barrel’s third time in contact with ethanol, the higher strength extracted more gallic acid, vanillin, and vanillic acid compared to the lower strength although the concentration differences are smaller than the first-fill barrels. Interestingly, the syringaldehyde extraction was reversed with the second-fill barrels, with the lower filling strength extracting less than the higher strength studies. No reasons for this variance were provided by the authors however it is possible that barrel-to-barrel variations of extraction, ethanolysis/hydrolysis, or oxidation of when the samples were drawn relative to environmental fluctuations/effects may have influenced the study.

Mosedale, in a summary of wood’s role in the maturation of whisky and other spirits, summarised several historical author’s work on the effect of filling strength [5]. In general, the extractive potential of water and ethanol can be at odds with each other and as such, the filling strength can influence the overall extraction of various compounds. Hydrolytic reactions require water while the solubility of degraded or extracted compounds require alcohol. Or in another way, the optimal extraction rate (ie., highest rate = faster maturation or wood-based flavours) will occur when the %ABV or ethanol/water concentration is optimised. Rather than re-writing a well laid out review paper, I’ve copied the below extract:

“However, Maga (1989a) found that maximum lactone (BB: fruits, coconut) extraction was obtained with 40 percent alcohol, while Puech (1984) found that the optimum extraction levels of tannins and lignin by Armagnac occurred at 55 percent. It is likely that similar variation may be found for other types of extractives. If the alcohol content rises above approximately 60 per cent, then the rate of extraction for colour, solids, tannins, and volatile acids have all been found to decrease.”
- Mosedale [5]

Over a 6 year period, Reazin found that filling at 63.5% or lower produced a more stable maturation profile and increase in more desirable colour compounds, congeners and solids (namely sugars) being extracted at increasing rates at proofs much lower than 63.5%, partially summarised in Figure 1 [6].

So that’s the literature review portion, things that are peer reviewed and published in international journals. I was quite surprised that the literature was quite limited on the effect of filling strength. So when forced to look for additional details, one of the best stones to turn over for anecdotal fountains of knowledge is those who make the stuff!

“I don’t think there are significant differences between spirits filled into the cask at 63.5% abv against spirit filled at full strength 71.0- 72.5% abv. The 63.5% abv was really an agreed filling strength used by the major distillers when reciprocal filling contracts were being agreed.“If it was a quality issue it’s beyond belief that all distillers just happened to arrive at the same conclusion that 63.5% was best, what about 64.1% or 62.9% abv, never seen that in 40 years.”
- Jim McEwan

“These (63.5% and 68.5%) are our traditional filling strengths for as far back as we can see in our records and, once it’s a tradition, it’s a very bold move to change it. Every cask at either filling strength could be used in any expression. Glenfiddich 12 Years Old, for example, comprises casks filled at both strengths,”
- Brian Kinsman, William Grant & Sons master blender

“The level of vanillin can be greater in casks filled at higher strength, but once diluted for bottling there’s no difference in the level of vanilla notes on a sensory evaluation, and the level measured in PPM (parts per million) is the same.”
- Brian Kinsman

“We sometimes fill Ardbeg above 63.5% ABV, which promotes a smokier whisky; filling below 63.5% ABV makes it fruitier”
- Brendan McCarron, Glenmorangie’s head of maturing whisky stock

“Perhaps Torabhaig, operational since 2017, will provide answers. They have been filling new-make spirit at various strengths, including 58%, 61.5% and 63.4% ABV, and according to Neil Mathieson, whisky maker at Torabhaig, each cask is labelled and set up in such a way that the team can directly compare casks where the only variable is the filling strength: “There are distinct patterns in the flavour profiles gained at each strength... But we can only muse on the results until we have actual empirical data.”
- Ian Wisniewski, Whisky Magazine

Now if you skipped to this part, I won’t blame you. Here’s the short of it:

The effect of fill strength has a minor impact when looking at minute changes in strength… the difference between 63.5% and 65%? Small. The difference between 53%, 63.5%, and still-strength 70%? Yup, there’s going to be some differences in the final matured whiskies based on barrel extractives. But don’t sweat it.

The rule of thumb that higher fill strength whiskies take “longer to mature” is somewhat complex and simple at the same time. More bitter notes and less sugars will have been extracted over the same time period compared to a “standard” entry proof; however the higher initial proof will enable the whisky to achieve greater esterification in the long term, ultimately producing a more complex whisky.

Additionally, the higher proof will “hold off” the woodiness of the barrel as less water-soluble compounds will be initially extracted. I believe this is the main driver for the “longer to mature” statement where some of the sweeter notes are present in insufficient quantities to balance out the more bitter components, potentially producing an imbalanced whisky at younger ages (imbalanced is can be hard to define based on sensory analysis however).

The oft-quoted 63.5% filling strength of Scotch whisky is likely an experience-derived initial starting point for maturation, aiming for optimal cask extraction at a desired return period, thus balancing out angel’s share and product portfolio unique to the distillery and region. If you live in regions with a significantly different climate, it is likely that your optimal entry proof should be different to accommodate your unique situation and desired return period (ie., bottling age, product portfolio, desired whisky flavour characterstics, etc.).

Objectively, these published works are focussed on short-term impacts and do not consider any long term oxidation or esterification effects that happen with more headspace (those greedy angels!) and longer maturation time. They also do not consider the long-term drop in alcohol over the entire maturation cycle, with more hydrolysed compounds being taken into the whisky the longer it is aged. And there is a simple answer for this: money.

It takes a lot of money to conceptualise, conduct, analyse, and publish any research findings and typical funding cycles last for 2-5 years. So it is virtually impossible to conduct a 10, 15, or 20 year study on these other variables without a very significant investment of money and time from the researchers. So take my above synopsis with a grain of salt, knowing that it is but one tiny snapshot of the magical chemical history swirling in your glass.

Review

Bunnahabhain 12yo, A.D. Rattray Single Cask Collection, 1st fill sherry butt, #900037, distilled 02.03.2009, bottled, 23.09.2021, 555 bottles, 66.7% ABV
CAD$225 (£108) paid

The fun of cask strength whisky is that we can play with the dilution on our own terms according to our flavour preferences. So I’ve set out some incremental dilution targets, ending at 46.3 %ABV for a direct comparison against an official bottling of Bunnahabhain 12.

Given the high price for a 12 yo, when I equate this down to the same official bottling %ABV, it ends up around $155 for 46.3 %ABV, or roughly 1.8x the price of a standard Bunna 12 in my area. So it wasn’t a cheap bottle and could end up more as a novelty than anything significantly better than the admittedly quite good official bottling. We shall see though.

I poured and sampled all four side-by-side to reduce any day to day variation in my senses.

I definitely couldn’t operate heavy machinery after this session!

Nose

66.7 %ABV

Wood varnish, darkly roasted walnuts, dark toffee. Whiff of fruitcake, singed orange peel, vanilla, and light roast espresso powder. It smells like the %ABV and colour would suggest. - Unofficial score: 7/10

56.5 %ABV (halfway to the 46.3 %ABV official bottling)

Fruitcake, salted toffee, light molasses, cinnamon, plush leather chairs, orange peel pulled out of an old fashioned. I like this %ABV the best, the perfect blend of high proof olfactory density with the ease of smelling common to lower proofs. - Unofficial score: 8/10

46.3 %ABV

Fruitcake, candied orange peel, red apple crumble drizzled with caramel sauce, old library and leather bound books. Sweet and enticing, devoid of any alcoholic prickle. - Unofficial score: 7/10

Control: Bunna 12 OB

More vanilla and more jagged smelling than the AD Rattray. More ozone/sea air and smashed sea shells on wet rocks than the AD Rattray as well. Touch of walnuts. It smells thinner and less rich. - Unofficial score: 6/10

Palate

66.7 %ABV

Hits your mouth with the force of Thor’s hammer. Salted toffee, dark espresso, dark chocolate, roasted hazelnuts and walnuts, cinnamon hit me right out of the gate before raw ginger and black pepper abrade your taste buds. It’s so dense that it takes some time for my synapses to wade through it, it’s best sipped in 2-3 mL increments. Successive sips reveal orange furniture polish, leather, and Demerara sugar, before the raw ginger comes back around. - Unofficial score: 7/10, trending towards 8/10 but the %ABV is too distracting, especially when drinking the 46.3% diluted version after a sip of this Bunna bomb.

56.5 %ABV (halfway to the 46.3 %ABV official bottling)

The %ABV makes itself felt over the more benign diluted 46.3% version and even the cask strength pour. The cinnamon, nutmeg, and raw ginger bite leads the way, followed by walnut and dark chocolate enrobed coffee bean bitterness, before leading into orange chocolate, leather, and toffee. There is a not so insignificant peppery bite here, detracting from the experience and score. - Unofficial score: 6/10

46.3 %ABV

Oh my. Thick, luscious, and dangerously drinkable, if a bit less diverse than the OB which makes sense since this is a single cask and not a vatting. Sugared orange peel, caramel sauce, muscovado sugar, fresh baked cinnamon apple danish with vanilla icing. Retronasal is an orange and citrus fruit jelly filling bomb. Overall, an improvement in mouthfeel, finish length, and experience than the OB, clinging to my palate for many minutes. - Unofficial score: 8/10

Control: Bunna 12 OB

The palate is better than the nose would leave you to believe. Thicker and richer than the nose. Honey, vanilla, toasted oats, caramel sauce, roasted walnuts dusted with cinnamon and black pepper. Nice and smooth. I forgot how good this stuff is. When I come back to this to check how the AD Rattray’s are doing, the much higher amount of vanilla is very evident, along with the thinner mouthfeel. - Unofficial score: 6/10

The Dregs

I’ll spare you a long trail-off here, you’ve been through a lot to get to this point.

As you can see, my “unofficial” scores varied significantly depending on the dilution ratio and the nose/palate experience. I’ve decided to go with the score 8/10 because that was objectively the highest experience throughout and when combined with the DIY aspect of this high proof whisky, means you could likely find your unique optimum strength and I’m sure you would arrive at an 8/10 as well.

A big surprise here was the 46.3 %ABV comparison to the OB. The AD Rattray was significantly oilier and gloopier, with very tenacious tendrils clinging to the sides of the glass. The OB Bunna was positively watery in comparison, coming through on the palate as well. Very interesting!

I can see the allure of the quality 12yo official bottling from Bunnahabhain. Sweet and smooth on the palate but quite lacking in unique features that this AD Rattray has in spades regardless of the dilution ratio. Alone, the OB is great, but side-by-side?

Not even a fair fight.

PS: this was the highest %ABV dram in my collection however I’ve recently acquired a 68.9 %ABV 10 yo Craigellachie. I’m excited to explore and keep playing with the addition of water! Long live cask strength releases!

Score: 8/10

References

[1] H. Weinberger, Guest Blog: Cask Filling Strength - Do Variations Matter?, GlenAllachie, https://theglenallachie.com/guest-blog-cask-filling-strength-do-variations-matter/, 2021.

[2] Clyne, J., Conner, J.M., Paterson, A., Piggott, J.R., The effect of cask charring on Scotch whisky maturation, Int. J. of Food Science and Tech., vol. 28, pp 69-81, 1993.

[3] Conner, J.M., Paterson, A., Piggot, J.R., Analysis of Lignin from Oak Casks Used for the Maturation of Scotch Whisky, J. Sci. Food Agriculture, vol. 60, pp 349-353, 1992.

[4] Piggott, J.R., Conner, J.M., Paterson, A., Clyne, J., Effects on Scotch Whisky Composition and Flavour of Maturation in Oak Casks With Varying Histories, lnt. J.of Food Sci. and Tech., vol. 28, pp 303-318, 1993.

[5] J.R. Mosedale, Effects of Oak Wood On the Maturation of Alcoholic Beverages With Particular Reference to Whisky, Forestry, vol. 68, no. 3, pp 203-230, 1995.

[6] G. Reazin, Chemical Mechanisms of Whiskey Maturation, American Journal for Enology and Viticulture, vol. 32, pp 283-289, 1981.

Tried this? Share your thoughts in the comments below. BB

Other opinions on this:

Whiskybase

Got a link to a reliable review? Tell us.