The Angels’ share is an expression we hear extremely often about the maturation of whisky, with just a quick explanation of the angels’ share being a portion (or share) of the volume of whisky (or other spirits) that is lost to evaporation during the aging in oak barrels. But what is it exactly? (Note: this piece was mostly written for the Amrut Greedy Angels review but deserve its place in the All about whisky category by itself).
So, what’s the Angels’ share?
There are two answers to this question: it’s part of whisky maturation, and a very fun film by Ken Loach about a mythical cask of Malt Mill to be sold on an auction at the Balblair Distillery. Go watch the latter if you didn’t already (Wait! Please finish reading this first!), and let’s talk about the former answer.
The Angels share is the term commonly used to refer to the loss of liquid from a cask of maturing spirit due to evaporation. I use the term “maturing spirit” and not just “maturing whisky” as all spirits (whisky, brandy, cognac, wine…) matured in oak casks will be subjected to this evaporation. Wood is a porous material that contracts and expands with the seasons, depending on humidity, temperature, airflow, the quality of the cask, the skill of people involved, and many other factors.
The X Factors
The influence of the oak
The skill of the cooper who made the cask will have an effect, with the staves of the cask being more or less tight depending on the skill. If the joins of the casks and there’s too much space between the staves, or even leeks, the losses of liquid will be greater than for a tight cask. Then the warehouse man has an influence on the cask, as casks turned regularly will not see tiny cracks appear where the oak is not in contact with liquid, making it dry.
Oak is also a factor by itself, as it is a natural substance with quality and properties that may vary from one cask to another: casks made with oak coming from the same forest by the same cooper will be different as oak trees are not all with the same quality.
And the influence of the climate
But that’s not all. New make spirit is, to simplify, around 65% pure alcohol and 35% water when the cask is filled. Water and alcohol are not drawn through the wood and evaporated for the angels to enjoy at the same rate according to the environmental conditions (humidity, temperature, airflow) in which the casks are stored. Alcohol is more volatile than water so it evaporates more easily from the cask, but even the evaporation is influenced by the environmental conditions.
When humidity is high in the warehouse, the water will evaporate more slowly and it will be mostly alcohol that will be lost. In Scotland, where the relative humidity is around 80 to 90%, this is the case and it is mostly alcohol that will evaporate, diminishing the ABV of the maturing spirit over the years. However, in drier climates like Kentucky for example, water can evaporate faster than alcohol, resulting in an increase in the alcohol by volume during the maturation of the spirit.
Controlled airflow is important too. If the airflow is too significant within the warehouse, then the concentration of water and alcohol in the atmosphere will be lower, and thus the evaporation will be sped up.
Then, the higher the temperature in the warehouse, the faster the evaporation of alcohol and water will be. That means that in some warehouses where casks are stored on several stories, the ones stored on the top, where the ambient temperature is the highest, will suffer from a higher share taken by the angels than the casks stored at the bottom with cooler temperatures. However, with hotter temperatures, the liquid expands, which augments the contact with the wood, allowing more absorption of flavours.
The Angel is in the (ware)house
The location and style of the warehouse greatly affect these atmospheric conditions and thus the angels’ share. In Scotland, many distilleries have old-style “dunnage” warehouses. These are traditional warehouses, not very tall and quite small compared to other warehouses. They’re usually built with an earthen floor, stone walls and a slate roof, and barrels are stacked on top of each other with three casks high. Since they’re not tall, the temperature difference between the casks on the bottom and the ones on the top is very limited. And since they usually have an earthen floor, they tend to have a higher humidity allowing moisture to seep out, and a higher humidity results in a slower maturation. However, there is no heavy machinery to move and rotate the casks, so it’s more labour intensive for the warehouse man to rotate and switch the casks.
Some distilleries, on the other hand, use racked warehouses. They have thinner walls and tin roofs, they’re a lot taller than dunnage warehouses, allowing from 8 to 12 stories of pallets 3 barrels high to be stacked on top of each other, with steel shelves. The walls are usually made from brick or concrete, and the floor will be concrete too. The atmosphere at the bottom of a racked warehouse is cool and moist, and the molecule of water being very small, water from the surrounding moisture can penetrate the cask, lowering the concentration of alcohol over time. On the other hand, the top of a rackhouse has a very hot and dry atmosphere, allowing the water to evaporate faster than the alcohol, which has much larger molecules, and the ABV inside will go up.
Both dunnage and racked warehouses have their pros and cons, and it’s difficult to say which one is the best for maturing whisky. What’s important to observe though is that many distilleries in fact use both dunnage and racked warehouses to mature their whisky.
Are all angels equally thirsty?
I won’t detail the atmospheric differences about all the countries producing whisky, but let’s take a closer look to four of them with data from a city from each of those countries: Edinburgh for Scotland, Louisville for Kentucky (USA), Taipei for Taiwan and and Bangalore for India, where Amrut is based. I chose those cities as enough detailed climate data was available.
In Scotland, humidity is high all year long. On average, November to February are the dampest months (83-84%), while April to June are the least damp with 75-76%. The average humidity in Edinburgh is 80.1% along the year.
Regarding temperatures, the annual mean temperature is quite cool at 8.4°C (47.1°F). July is the warmest month, with an average temperature of 14.5°C (58.1°F) while January is the coolest month, having a mean temperature of 3.2°C (37.8°F). The variation of mean diurnal temperatures is of 7.4°C (13.4°F) which means the mean difference between the night and the day is 7.4°C on average along the year.
Now let’s move to Louisville, Kentucky, as 95% of the world’s bourbon is made in Kentucky. The humidity is way lower than in Edinburgh (who would have known!) with an annual average of relative humidity of 57%, and the average monthly relative humidity ranges from 52% in September to 68% in January.
The annual mean temperature of Louisville is mild at 13.3°C (55.9°F). July is like Edinburgh the warmest month with a mean temperature of 25.1°C (77.2°F) while January is the coolest month with a mean temperature of -0.2°C (31.6°F). The variation of diurnal average temperatures is higher than Edinburgh, with 11.1°C (20°F).
For its part, Taipei (Taiwan) has a slightly higher humidity than Edinburgh with an annual average of 81.6% of relative humidity, going from 78% in July to 84% in January:
Regarding Taipei’s temperatures, they are really warm, with a mean annual temperature of 21.6°C (70.9°F) and an average daily variation of 7.6°C (13.7°F). The hottest month is July with a staggering average temperature of 28.5°C (83.3°F) while the coldest month is February with an average temperature of 15°C (59°F), a wee bit more than Edinburgh’s hottest month!
And in India
Finally, let’s see what the usual climate is in Bangalore, India, where Amrut is from. The average annual relative humidity of 65.2% is not that high, 15% less than Edinburgh average humidity, with March being very dry at 45% and August quite damp at 79% on average:
The average annual temperature in Bangalore is fairly hot at 24.1°C, hotter than Taipei and way hotter than Louisville and Edinburgh. The variation of daily average temperatures is also quite important with 10.4°C (18.8°F) on average between night and day. April is the hottest month with an average of 28°C (82.4°F) while the coldest month, December, is still quite warm with a mean temperature of 21.1°C (70°F).
Let’s stack the data to ease the comparison
Now if we compare the humidity of the four cities in a single graph for readability purposes. We can see here that Edinburgh and Taipei both have similar high relative humidity, while Bangalore as a huge variation and Louisville being more regular but the dryer on average:
Regarding temperature, the differences are very visible here, and you can imagine the huge differences in evaporation due to the temperatures in those cities:
Now how does that translate in the average angels’ share percentage for these regions? The average angels’ share in Scotland is usually said to be around 2% and the alcohol by volume will lower over time, while in Kentucky the angels’ share is said to be around 5%, with the ABV slightly higher than what it was when the cask was filled after a few years. In Taiwan, the angels’ share is between 10 and 12%, while in Bangalore they also lose close to 12% of the spirit per annum due to evaporation and the ABV after 3 years will be more than 2% higher than the initial filling ABV:
Now there must be a scientific explanation for the difference of volumes lost to the Angels’ share, and that’s where people working at the Rum distillery Santiago de Cuba, a professor of chemistry from the Faculty of Chemical Engineering at the University of Santiago de Cuba and a Research Group of Applied and Analytical Chemistry from the Hasselt University in Belgium come in.
Calculation of the volume lost during spirit ageing
In 2019, the MDPI (Multidisciplinary Digital Publishing Institute, a publisher of peer-reviewed, open access journals and publications in Switzerland) published a special issue: “Distilled Beverages: Science and Technology Across the Supply Chain“. One of the three publications in this issue was titled “Boltzmann-Based Empirical Model to Calculate Volume Loss during Spirit Ageing” and was written by Noemi del Toro del Toro and other people (see the complete list of authors in the publication).
I won’t try to paraphrase the abstract of the study, so here it is:
The Boltzmann equation is applied to fit data of volume loss for evaporation (in %) during spirit ageing in northern white oak (Quercus Alba) standard barrels of 205 L (+/− 10 L) using a temperature and humidity-controlled cellar. The Boltzmann equation satisfactory fitted to the experimental data of the volume loss against temperature at constant humidity. Two parameters of the Boltzmann equation showed a linear dependency on the relative humidity of the air, while the other two parameters exhibited a constant value independently of the air humidity. The found empirical mathematical model can be used to calculate the volume loss for evaporation of spirits (40% v/v of ethanol) during ageing in terms of relative humidity (range: 40%–95%) and temperature (range: 10–30 °C) with significant accuracy.Del Toro del Toro, et. al, 2019
The study introduces the angels’ share existence in rum and wants to obtain an empirical Boltzmann-based mathematical model to describe the volume loss during ageing in a conditioned cellar. The model is used to calculate the annual percent volume loss during a 40% abv spirit ageing in standard white oak barrels under controlled conditions of temperature and air humidity, the only two variables considered in this study. As I’ve written earlier, both alcohol and water evaporate at the same time, so the study talks about total volume loss and not just alcohol or water.
The experimental data
The experimental data was provided by the rum producer Santiago de Cuba and the volume loss per year (in percents) at different conditions of air temperature and relative humidity was presented in the table below:
The Boltzmann equation, devised by Ludwig Boltzmann in 1872, describes the statistical behaviour of a thermodynamic system not in a state of equilibrium. I’ll let you go to the Wikipedia page for this equation, but in a few words, it can be used to determine how physical quantities change when a fluid is in transport.
Del Toro del Toro and the other authors used the Boltzmann sigmoidal classical model to fit the experimental data.
One of their results was a figure showing the volume loss for evaporation as function of the temperature at a constant relative air humidity with 12 curves:
This figure shows that air humidity significantly influences the evaporation from the barrel. And the higher the water content in the air, the lower the volume loss during ageing. It also showed that an increment of the temperature generated an increase in the volume loss, and the effect of the temperature on the volume evaporated is more significant at a lower humidity.
After several other calculations, they came up with an Angels’ share equation. Obviously, the parameters of the equation change with the frequency of the calculation of the environmental measurements. So, here’s the Angels’ share equation:
The result of this study is that using a four-constants classical Boltzmann sigmoidal model you can measure the volume loss of spirit in the described conditions.
So, in the end, what’s the Angels’ share?
It’s the volume loss to evaporation of a spirit during ageing in oak casks, and it’s not the same everywhere, due to a shit-ton of reasons.