Imperial 24yo
Single Malts of Scotland | 42.9% ABV
Give thanks to the angels
What do we normally think about when it comes to “mature” whisky? Age is the most obvious metric. Some people focus on colour and/or level of cask flavour. For some, it’s the distinct absence of distillate character; the antithesis of any new make properties. Well, these are all useful, but I’d like to add another distinct metric for your consideration; oxidation.
We all know that part of maturing whisky is the so-called “angel’s share”; the loss of liquid volume is replaced with a headspace, and some of this headspace is spirit vapour, while the rest is atmospheric ingress, generally through the pores of the oak and from around the seal of the bung.
Part of this atmosphere, oxygen, interacts with the liquid and slowly undergoes redox reactions with a number of the compounds in the spirit. When the bung is dry (ie when maturation occurs on its side, as is typical in Scotland’s dunnage or racked warehouses) 30%-60% of the cask’s atmospheric oxygen ingress is via the bung (i.e. an imperfect seal) after which the principal source of dissolved oxygen in the liquid is via diffusion across the gas-liquid interface [5].
We won’t delve into these generalised oxidative reactions too much, especially given I’ve covered them somewhat in previous reviews (such as this SMWS Islay Festival release and this Octomore deep dive). There are two reiterations I will make though; ethanol oxidation to acetic acid (via acetaldehyde as the intermediary product) and increase in total organic acids (principally acetic) from wood extract.
Acetyl units are speculated to be derived from hemicellulose degradation [4]. Both of these increase the proportion of acetic acid in solution, the result being esterification with available alcohols (dominantly ethanol), lending to an increase in perceived fruitiness. The other oak derived acids (minor quantities) are listed in the below table [4,6], many of which will also undergo progressive esterification with alcohols over time.
It should also be noted that the oxidation processes referenced above are dynamic in nature, dependent on a number of factors. For instance, the geometry of the cask and the resultant liquid-gas interface at various fill volumes; we’ll touch back on this later. Oxidation rate is also affected by environmental conditions; the generalised rate of reactions is temperature dependent as per the Arrhenius equation, which essentially dictates an inverse logarithmic relationship between temperature and rate of reaction.
Thus the warmer the liquid, the greater the reaction rate. Greater thermal energy leads to more frequent chemical collisions, as well as those collisions having a greater kinetic energy, both of which increase the likelihood of a collision leading to a reaction. The rate of oxidation is also mediated by barrel extract, with greater extract leading to higher rates of spirit oxidation [2,7] and resultant acid/ester development.
Philip et al [9] speculate that hydrolysable tannins may be of particular interest so far as cask extractive solids, given that their interaction with copper ions and oxygen produce peroxide ions and active oxygen, which may be catalysts for oxidative reactions. This is complicated by the work of McPhail et al [10] who established the same hydrolysable tannins as having antioxidant properties; thus further elucidation is necessary. It may be that the net effect on maturing whisky is conditional (i.e. a dynamic equilibria) to a number of factors.
The liquid’s pH is also a factor in esterification/hydrolysis equilibria, with a decrease in pH due to organic acids favouring esterification. Likewise, the progressive change of ABV dictates esterification rate and equilibria points too; decreasing ABV (ie more water relative to system) favours hydrolysis of esters, whereas ABV increasing favours esterification. This may be part of the reasoning for anecdotal reports that whisky matured in warmer/drier climates (where ABV increases over time, such as parts of India) developing fruity/estery qualities rapidly. In any case, given the steady increase in acids and decreasing pH expected of maturing whisky in a (generally) ethanol dominant environment, we expect to see ester levels trend upwards overall in the majority of instances [7].
The other component of increased atmospheric interaction (both atmospheric ingress and egress) lending to mature character is the capacity for volatile compounds to evaporate into the headspace in a cask, and then subsequently out of the cask as part of the evaporative losses. This is one of two principal mechanisms by which maturing whisky experiences reduction in sulfidic character over time, the other being adsorption to the interior surface of the cask.
Regardless of the specific role of hydrolysable tannins, total extracted solids certainly correlate to higher rates of increase for both organic acids and esters, including by oxidative pathways, thus the greater the extract potential of the wood, the greater the rate of acquired maturation as per these oxidative/acidifying/esterifying metrics. Specifically, extract potential will largely relate to coopering of the cask, including cask seasoning, thermal degradation processes, as well as the degree and nature of previous uses of the cask. For more information, refer to my cask deep dive article. The rate of extract is also dependent on initial conditions, such as cask entry proof (see Broddy’s excellent piece for more on this) and initial headspace volume coupled with the resultant surface area of the liquid-gas interface. Again, we’ll touch on this geometry shortly.
Right, enough recapping and chemistry. So what are we talking about relative to today’s whisky then? Well, if you look closely at the label, you’ll notice that the number of bottles is quite low coming from a barrel, even at 24 years of age; just 104, or the equivalent of around 73 litres. Even assuming “Barrel” refers to an American Standard Barrel (approx. 200L) then that’s an angel’s share of around 59% (allowing roughly 10L loss to what soaked into the oak). Back calculating for this rate of loss over 24 years gives us the equation 0.59 = (1-x)^24, thus x is approximately 2.17% per annum. Usually we assume that 2% is an upper bound approximation for the angel’s share in Scotland, and that the angel’s share rate generally decreases over time.
This makes sense, given that the angel’s share typically causes a decrease in ABV over time in Scotland, thus a decrease in volatility of the liquid leading to a lower rate of evaporation into the headspace and out of the cask. In other words, either this was either a particularly porous cask, or there was a leak at some point in its life.
In the former case, there are a number of possibilities; one of the first that comes to mind is a re-coopered cask. Conner [1] posits that rejuvenation processes involving the removal of tired wood (i.e. scraping/shaving) from the interior of the stave increases the porosity of casks, leading to greater angel’s share rates. Likewise, the increased porosity could simply be due to individual oak physiology or initial coopering processes; for instance, finer grained wood has been shown [3] to have a greater relative porosity. Elsewise, individual cooperages use differing stave thickness, which we might logically extend from Conner’s observations also impacts porosity.
At the end of the day, we’ll never know what the cause was for this cask yielding so few bottles, but I’m very thankful regardless. Due to the greater angel’s share, this liquid has spent a greater proportion of its maturing life with a greater headspace volume than usual, as well as a larger surface area to volume ratio between the gas-liquid interface and whisky volume (respectively). Thus the total rate of oxygenation and oxidation reactions contributing to maturation character appears greater than many other malts I’ve had at similar ages.
But Tyree, how large an impact could we reasonably expect headspace volume to really play in the oxidative potential of a whisky? Well, I’m glad you asked, because the maths involved in trying to answer that question is really quite something. Firstly, one needs to assume the general shape of a cask; that is, what is the curvature of the staves relative to the other cask dimensions? If we assume that the curvature is roughly parabolic, then using the difference in radius between the cask’s heads (the h term) as well as the internal radius at the bung (the b term, easily measured using a chain or weighted string as a dipstick) then we can determine this curvature as per the derivation in the below image. Combining these with the length of the cask (the L term) we can derive the volume. All due apologies; this was my first draft of the derivation, and I have always been messy in my maths! The circled formula at the bottom is equivalent to the volume.
P.S. - If anyone feels like double checking any of the following maths, please go for gold. I stupidly omitted to have it verified prior to publication!
Considering the ullage of a cask, Miller [11] derives the following unpleasantness;
Calculating the surface area as a function of this ullage as applied to a parabolic curvature quickly becomes, to use a technical term, rather icky. The cylindrical approximation is a much nicer perspective, though the error committed by the approximation is difficult to evaluate due to the initial ullage formula as given above. The surface area, volume and ratio between the two are given below, though practically the formulas are of limited use in application.
If we substitute in a couple of quick values, we can roughly see the impact this geometry has. For a small volume loss (i.e. when the whisky is still young) the angle might be around 0.2 radians, and let us assume the radius of the cylinder is approximately 35cm. The resultant surface area to volume ratio yields approximately 0.002cm^-1. The same cask, when half full (i.e. quite old) has an angle of pi radians, which yields a surface area to volume ratio of 0.0364cm^(-1). That is to say, over 18 times the surface area to volume ratio. Again, this is a very coarse approximation using the cylindrical substitution, but it serves adequate demonstrable purposes.
The ratio continues to grow rapidly, asymptotically approaching a theoretical maximum as the angle tends to 2*pi radians (ie the cask being empty).
The upshot of all this is simply to say that the greater the angel’s share, particularly early on to generate atmospheric headspace in the cask, the greater the capacity for oxidation by geometric arguments.
Review
Imperial 24yo, Single Malts of Scotland, 42.9% ABV
AUD$500 (£250) retail, AUD$400 (210) paid.
Having had fairly few Imperials to date (at least that I can recall) I had very little in the way of expectations tasting this. The bottle was purchased, as with the majority of these premiumised bottles, to be shared through my whisky club, thus significantly diluting the cost. Reading other reviews from folks like Serge, there seems to be a general sense of something waxy/oily and with floral/fruity qualities. Sounds fun, let’s find out!
Nose
Evident maturity. There are the usual markers with some vanilla, light baking spice and other cask extractives, but they’re notably gentle. More pronounced are the estery fruits, particularly dried peach and apricot slices, sweet pickled mango chunks and subtle tropical gum. There’s a touch of heady, sweet florals in the mix too, as well as floral honeys, further lending to the elegance. Can I still get away with my lilac and gooseberries reference, or too soon sans Henry Cavill? Regardless; perhaps a soft sense of some oiliness in the mix too.
Palate
There’s almost an element of old bottle effect to this, further leading me to believe O.B.E. is largely a product of extended chemical equilibria coupled with oxidative ageing. No data, just a hunch. Anyway…
The fruits continue here beautifully, coupled with a subtle fatty/oily quality not dissimilar to Clynelish, though more restrained. I feel I’m starting to understand the profile comparisons now.
Mango chutney, papaya and more various dried orchard fruits; flits back and forth nervously between orchard and tropical varieties in a very pleasing way. Perhaps even a little gewurztraminer-esque lychee. The perfumed and floral elements are here again, mingling nicely with the fruit. The finish extends with some pleasant white pepper and a tinge of nutty maltiness, rooting us back to the fact that this is malt whisky.
One could be forgiven for thinking the low ABV would make this feel thin on the palate or otherwise somehow sub-par, but that just isn’t the case. It’s not a viscous dram, but there is a lovely texture concomitant with the overall poise and maturity. What a belter!
The Dregs
Reading back, I feel my notes don’t do an adequate job of describing how well composed this whisky is. The spirit is evident from start to finish, unencumbered by any surplus of oak, simply morphed over time and oxidation to be elegant and prominently fruity. That said, there is still a sense of what I imagine is the distillery character (namely some malt and that subtle waxy/oily note) injecting some unique personality too.
Love it; not cheap, but not exorbitant for the specs, especially at the price paid. Score unadjusted by pricing then.
Score: 8/10
References
[1] John Conner, Reducing the need for new wood by rejuvenating and re-using casks, World Distilled Spirits Convention 2011 (Article Chapter 6)
[2] John Conner, John Ramsay, Brewer & Distiller International, Volume 4, Issue 9, September 2008
[3] Whisky and Other Spirits, Edited by Inge Russell, Graham G. Stewart and Julie Kellershohn, 3rd Edition 2021, chapter 16 (maturation).
[4] Nishimura et al in Flavour of Distilled Beverages: Origin and Development, 1983, pages 241-255.
[5] Maria del Alamo-Sanza, Ignacio Nevares, Oak wine barrel as an active vessel: A critical
review of past and current knowledge, Critical Reviews in Food Science and Nutrition 2018.
[6] Lalli Nykanen, Aroma compounds liberated from oak chips and wooden casks by alcohol. In: Nkanen et al, Proceedings of the Alko symposium on flavour research of alcoholic beverages, 1984.
[7] M.C. Brockman, Relationship between acids, esters and solids during the ageing of whisky, Journal of the Association of Official Agricultural Chemists, 1950, Vol.33 (1), p.127-129
[8] Olatunde et al, Rates and equilibria of ester hydrolysis: Combination of slow and rapid reactions, Chemical Engineering and Processing - Process Intensification, Volume 50, Issue 7, July 2011, Pages 665-674
[9] J.M. Philip et al, Scotch whisky flavour development during maturation, Proceedings of the second Aviemore conference on malting, brewing and distilling, Institute of Brewing, 1986
[10] D.B. McPhail et al, assessment of the antioxidant potential of Scotch whiskies by electron spin resonance spectroscopy: relationship to hydroxyl containing aromatic compounds, Journal of Agricultural and Food Chemistry, 1999.
[11] Gregory H. Miller, Whisky Science, A Condensed Distillation, Springer publications, 2019
Tried this? Share your thoughts in the comments below. TK