ANNANDALE DISTILLERY IS COMMITTED TO RESPONSIBLE DRINKING

Barley – What makes it so special?

Chill Filtration - Sacrilege Or Saviour?

Copper - It's Effect On Single Malt Scotch Whisky Flavour

Click here to browse the second Technical Notes page - Or follow the links below.

Single Cask-Single Malt Scotch Whisky vs Vatted Single Malt Scotch Whisky

Small Casks - Why Don't We Use Them?

STR's - Shaved, Toasted & Re-Charred Wine Casks

Making Single Malt Scotch Whisky is part science and part art. As a scientist, I rejoice in the artisan aspect of whisky-making and sincerely hope that the human senses and the human mind will forever be the final arbiter of sensory maturity and sensory excellence rather than some mechanical device. Nonetheless, over the past 50 years especially, science has revealed some of the mysteries of whisky making and exploded some of the unhelpful myths without, I hope, ruining the mystique and the majesty of Scotland’s Single Malts.

This section of Annandale’s website comprises a series of short Technical Notes on various aspects of Single Malt Scotch Whisky production, from barley through to the mature product. The topics are presented alphabetically, and the series will be expanded over the coming months.

I’m deeply grateful to the late Dr Jim Swan and to Dr Gordon Steele for educating me in the science of whisky making and to Mark Trainor (Head of Production at Annandale Distillery) for schooling me in the practical subtleties of making the very finest Single Malt Scotch Whisky.

Professor David Thomson

Co-Founder – Annandale Distillery

Scotland produces two types of Scotch Whisky; Scotch Grain Whisky and Scotch Malt Whisky. There are two principal distinctions between these two types of Scotch: First of all, Scotch Grain Whisky is produced from various cereals, typically wheat, barley, malted barley and sometimes maize (depending to some extent on the prevailing cereal prices), whereas Scotch Malt Whisky must be made solely from malted barley (Hordeum vulgare). Secondly, Scotch Grain Whisky is distilled in a patent still (continuous process) whereas Scotch Malt Whisky is distilled in copper pot stills (batch process). The fundamental differences between patent stills and copper pot stills, and the consequent effect on flavour, will be addressed elsewhere in this series of technical notes. Suffice to say that patent stills tend to produce Scotch Whisky that’s generally smoother and less characterful than the Scotch Malt Whisky that’s distilled in copper pot stills.

The bulk of Scotch Whisky sold worldwide is Blended Scotch Whisky (e.g. Johnnie Walker, Chivas Regal, Bells, Ballantine’s, Famous Grouse, Grants Standfast, etc., etc.). These are typically produced from unique blends of Grain and Malt Scotch Whiskies, formulated by very experienced ‘master blenders’ to deliver the flavour profile that characterises the brand. Most of the Malt Whisky produced in Scotland is used as an ingredient in Blended Scotch Whisky. Back in the 1960s, very little Malt Whisky was sold as Single Malt (the word ‘Single’ indicating that it’s the product of just one distillery). William Grant should be given credit for creating a worldwide market for Single Malts via Glenfiddich Single Malt Scotch Whisky. Today, the global market for Single Malts has become well established and it’s growing in volume, value and kudos. In 2018, Scotch Malt Whisky exports were valued at £1.3 billion, constituting ~27% of total Scotch Whisky exports.

Much of the appeal and interest in Single Malts derives from the fact that there are over 100 Single Malt distilleries in Scotland, each typically producing a number of expressions differing in age and type of maturation cask. This creates a wide selection of sensory profiles for consumers to compare and contrast. There’s also huge interest in the provenance of the individual distilleries.

There’s little doubt that the principal sensory differences between Scotch Grain Whisky and Scotch Malt Whisky derive from the distillation processes rather than the various different type(s) of cereal used in their production.  That’s not to say that the compositional differences between barley and other cereals are unimportant in whisky flavour production, but the impact of these cereal-derived differences is modest. It seems much more plausible that the stipulation that Scotch Malt Whisky must be produced solely from malted barley would have been a pragmatic commercial decision intended to create a clear legal distinction between the two types of Scotch Whisky. If this was the case, then those who sought to create this distinction should be congratulated for their foresight in providing a platform from which the burgeoning Single Malt Scotch Whisky category has since developed. However, this doesn’t explain why malted barley is often preferred by beer brewers. Nor does it explain why Scotch, Irish, American and Canadian grain whiskies invariably include a significant proportion of malted barley (which is relatively expensive) in their mash bills.

Surely, there must be another reason!

In nature, barley grains are simply the seeds of barley plants but, from the distiller’s perspective, they’re the source of carbohydrate (starch) from which alcohol is produced. To fully appreciate the ‘specialness’ of barley, it’s important to know something about the structure of barley grains.

Barley grains comprise 3 principal botanical features: the germ, endosperm and bran.

The bran layer is essentially an outer sheath that encloses the grain. It accounts for 10% of the dry weight of the grain and comprises three distinct layers; the husk, pericarp and testa. The husk is a leaf-like structure that’s physiologically external to the grain. Barley is one of only four commercial cereals that retains its husk after harvest (along with rice, oats and millet). The pericarp and testa form an integral part of the grain known as the seed coat.

The germ constitutes about 3% of the dry weight of the barley grain. It is the reproductive part of the seed (i.e. the embryo) that eventually grows into a barley plant after germination.

The endosperm is the starchy food store upon which the developing seedling depends for sustenance, until such time that it can self-nourish via photosynthesis. Endosperm cells contain large and small starch granules, embedded in a protein matrix, all of which is enclosed within a cell wall. The endosperm, which accounts for 75% of the dry weight of the barley grain, comprises mainly starch (85%) and protein (10%). The remaining 5% is mostly cell wall material.

The entire endosperm is enclosed within a membrane-like structure known as the aleurone layer, which comprises 12% of the dry weight of the grain. Whilst in most cultivated cereals (wheat, rye, oats, rice, etc.) the aleurone layer is just one cell thick, in barley it’s multicellular. For reasons that will be explained shortly, it’s the multicellular nature of its aleurone layer that makes barley so very special!

Bearing in mind that it’s sugar rather than starch that’s fermented by yeast into alcohol, the distiller’s challenge (or more correctly, the brewer’s challenge) is to release the starch from within the endosperm cells and then break it down into fermentable sugar. This is a 3-stage process that starts with malting, followed by milling and finally by mashing. The end-product, known as wort, is a golden-brown liquid that’s rich in fermentable sugars and other materials that are essential for making good-quality Single Malt Scotch whisky.

Malting, milling and mashing are dealt with in much greater detail elsewhere in this series of technical notes. Save to say at this stage, that malting is essentially induced and carefully controlled germination of barley seeds. The principal objective of malting is to break down the walls of the endosperm cells and thereafter to release the starch granules therein from the adherent protein matrix in which they’re embedded. This process is known as modification. Mashing is the process whereby starch released from the endosperm cells during malting is broken down into fermentable sugars.

Three principal types of enzyme are involved in malting: Enzymes (largely glucanases and pentosanases) responsible for breaking down the endosperm cell walls, thereby allowing the cell contents (i.e. starch granules embedded in protein) to spill out. Enzymes (known as endo-proteases) that break down the endosperm cell proteins thereby exposing the starch granules to the third type of enzyme (known as amylolytic enzymes) that ultimately break down starch molecules into fermentable sugars (largely maltose).

The ‘trick’ with malting is to allow the process to continue to the point where cell wall and protein matrix break down (modification) is optimal and amylolytic enzyme potential is maximised, whilst conversion of starch into sugar (and subsequent consumption of sugar by the growing seedling) is minimal. Malting is arrested by drying the germinating barley in a malt kiln. At the end of the malting process, only about 10% of the starch should have been converted into sugar but this is enough to make the malted barley grains taste slightly sweet.

Stopping the malting process at exactly the right point requires a huge amount of skill and experience. If it’s stopped too soon, modification will be incomplete, and a significant proportion of the starch granules will be inaccessible to the amylolytic enzymes. Too late and conversion of starch into sugar will be too advanced. Either way, the yield of alcohol will be reduced, with all the attendant economic consequences. It’s for this reason that most distillers leave malting to the experts and buy-in their malt from commercial maltsters.

Going back to the ‘specialness’ of barley, all three enzyme types mentioned above are either released from or synthesised in the aleurone layer that surrounds the starchy endosperm in all cereal grains. However, in most cereals, the aleurone layer is just one cell thick whereas in barley it’s multicellular. This means that the enzymes required for modification of the endosperm cells and for the subsequent conversion of starch into fermentable sugars, are produced much more rapidly and more proficiently in barley than in other cereals……

……and it’s this, more than anything else, that makes barley so special!

Including malted barley in the mash bill of grain whiskies means that mashing will be more rapid and more likely to go to completion, with all that this implies for efficiency, yield and cost.

David Thomson

Co-Founder – Annandale Distillery

When whisky is chilled, either through storage at low temperatures or the addition of ice in-glass, it may form a cloudy haze. This is considered undesirable by some whisky drinkers, although not all. Chill filtration removes the compounds responsible for haze, leaving the whisky crystal clear.

To understand how haze is formed, and subsequently removed, it’s necessary to know something about two types of chemical compound formed in whisky; fatty acids and esters.

Fatty acids

Yeast and lactic bacteria (to a lesser extent) produce fatty acids during the fermentation of wort (the sugary solution produced when starch is converted into sugar during the mashing of barley) into wash (alcoholic beer-like liquor). Fatty acids, along with another substance called glycerol are the principal constituents of fat. The ‘body’ of a fatty acid is formed from carbon atoms linked together in a chain. These chains of carbon atoms can be short, long or somewhere in between.

Esters

Esters are formed when a molecule of alcohol and a molecule of acid react together (known as esterification). Esters are fruity in character and make a significant contribution to the sensory characteristics of Scotch whisky.

Two basic types of ester occur in Scotch Whisky: Acetate esters formed by esterification of acetic acid and alcohol, and ethyl esters formed from ethanol and a fatty acid. As ethanol (ethyl alcohol) is the principal alcohol formed during brewing, ethyl acetate is the most commonly occurring acetate ester in whisky. This has a characteristic sweet, pear drops smell. As most of the alcohols formed during fermentation have short carbon chains, the corresponding acetate esters also have short carbon chains and therefore are not greatly implicated in the formation of haze. However, because ethyl esters are formed from long-chain fatty acids, they can be quite large molecules. As the chain length of a fatty acid or an ester increases, volatility decreases. Consequently, fewer of these larger fatty acids and esters will make their way into the distilled spirit. Nonetheless, some do, and this is how the problem of hazing arises.

Micelles

Molecules of long chain fatty acids and esters have polar (charged) and non-polar (non-charged) entities, due to the nature of the various different functional groups that form part of the molecule’s structure. Polarisation causes the molecules to clump together to form structures known as micelles, where the non-polar (neutral) parts of the molecules aggregate towards the centre of the micelle and the polar (charged) parts project out into the solution. As the proportion of water in whisky is increased (the critical point seems to be about 46% alcohol by volume), the micelles begin to come out of solution and are precipitated. It’s these precipitated micelles of long chain fatty acids and esters that form haze in whisky. Lowering the temperature of the whisky further reduces the solubility of the micelles which exacerbates the problem of haze. Consequently, adding a few chunks of ice to a non-chill filtered cask-strength whisky could reduce the ABV below 46% and chill the liquid sufficiently to cause the micelles to come out of solution and cause haze. The most important esters in haze formation are ethyl laurate, a C14 (14 carbon chain) molecule, and ethyl palmitate and ethyl-9-hexadeanoate, both of which are C18 molecules. Removing these could potentially diminish the flavour of the whisky, albeit to a very modest extent

Process

As the name suggests, chill filtering removes haze from whisky by first chilling it (between ambient and -10oC) and then filtering it through a plate and frame filter containing filter sheets made of low-calcium kieselguhr (a type of diatomaceous earth). Chill filtering is actually a very mild process that removes about 80% to 90% of the fatty acid/ester micelles from Scotch whisky without causing any collateral damage.

To chill filter or not to chill filter?

With distilleries such as Annandale, that produce Single Cask, Single Malt Scotch Whiskies at cask strength, chill filtering is utterly pointless and not something we are likely to entertain. People that buy Annandale’s whiskies generally prefer truly authentic whisky and they’re unlikely to be put-off by haze. Indeed, for many, it is a hallmark of authenticity.

However, chill filtering isn’t quite the ‘evil’ that some people make it out to be, especially when implemented properly. By virtue of its very purpose, chill filtering will inevitably remove some of the fruity esters and this could potentially impact on the fruity character of the whisky, but probably only to a very modest extent. It shouldn’t remove colour. Nonetheless, as haze is construed as a fault or an impurity by some whisky drinkers, chill filtering is probably justifiable for mass market blended Scotch whiskies and vatted Single Malts……but not for Single Cask, Single Malt Scotch whiskies such as Annandale’s Man O’Sword or Man O’Words.

Rule of thumb

When buying Single Malt Scotch whisky, check the alcohol by volume (ABV) on the label. If the ABV is 46% or above, it probably hasn’t been chill filtered. If it’s below 46% ABV it almost certainly has been chill filtered. Most Single Malts are sold at either 40% ABV or 43% ABV, the latter in the USA and duty free. These will have been chill filtered. Most Single Casks Single Malts are sold at cask strength (typically 55% - 61% ABV, depending on age) and will not have been chill filtered (as with Annandale).

The choice is yours!

I always buy non-chill filtered whisky and add a small amount of water and/or ice to reduce the ABV to approximately 35%. Watching the whisky form a haze is, for me, part of the sheer pleasure of drinking Single Malts.

David Thomson

Co-Founder – Annandale Distillery

Acknowledgement:             Dr Gordon Steele

Copper pot stills are the iconic distillation vessels found in all Single Malt Scotch Whisky distilleries. The choice of copper as a construction material for the stills and condensers was initially governed by its malleability, which makes it relatively easy to form into the complex shapes, and by its ability to conduct heat efficiently. The fact that it happens to play an important role in spirit quality is a happy coincidence. This positive effect on flavour is attributed to the ability of copper to reduce the level of sulphur containing compounds in the distillate and to the act as a catalyst in the formation of esters (flavour compounds that contribute positively to the flavour of Scotch whisky). Surprisingly, given the importance of copper, details of how it influences whisky flavour are not fully understood. However enough is known for distillers to appreciate the importance of copper during distillation, in terms of determining and refining spirit flavour and also in terms of the consistency of spirit quality.

Sulphur containing compounds, at high levels, are generally considered to give the spirit unpleasant, undesirable odours often described as vegetable, rotten egg, grassy, rubbery or struck matches. Importantly not all sulphur compounds are undesirable; at lower levels they can make a positive contribution to the complexity of the whisky improving the feintyness*, and meaty aromas. (*’Feints’ is the name given to the third fraction of distillate produced in the spirit still. It has a characteristic smell and flavour {‘feintyness’} which can be desirable, but only in very small amounts. Excessive feintyness is very unpleaant.)

The most common sulphur based compounds in whisky are dimethyl sulphide (DMS), dimethyl disulphide (DMDS) and dimethyl trisulphide (DMTS), although there are others. All three of these sulphur compounds have an undesirable odour. They are formed when methanethiol reacts with hydrogen sulphide. These compounds derive from the amino acids methionine and cysteine, respectively, which ultimately derive from the yeast and barley used in fermentation.

Copper is involved in both increasing and decreasing the levels of sulphur compounds, although it is the latter that’s most important. Copper salts, which are produced by corrosion of copper on the inside surfaces of the stills, promotes the formation of DMTS from methionine. Conversely, native (i.e. uncorroded) copper lowers the concentration of DTMS. Perhaps the simplest way in which copper removes undesirable sulphur compounds is by absorbing them onto the inner surfaces of the stills and condensers during distillation and subsequently releasing them when exposed to air as the distillation equipment is emptied at the end of the process. Of greater significance, sulphur compounds are also reduced via chemical reactions that transform them into less flavoursome substances or by forming complexes with copper which don’t ultimately distil over into the new make spirit.

The role of copper in the formation of esters is not well understood. The majority of esters in Scotch whisky are produced by yeast during fermentation. They are formed when alcohol (principally ethanol in this case) reacts with carboxylic acids (principally fatty acids but also acetic acid) and impart flowery and fruit flavours to the whisky. Copper is a catalyst for ester formation which occurs when the reactants meet native copper on the inner surfaces of the stills and condensers. It is well known that after cleaning the inside of copper stills and condensers, undesirable flavour changes may occur until a patina has built up on the copper surface which changes the properties of the copper as a catalyst.

It is important to recognise that while copper plays a crucial role in influencing the quality of new make spirit, flavour reactions can only occur when the liquid is in contact with the copper. Thus, it is important during the short period of copper exposure to maximise the desired effects of reducing sulphur containing compounds and esterification. This is achieved via process management techniques and by optimising various distillery design features. Process management parameters include distillation time, reflux rates (the average number re-condensation - re-volatilisation cycles achieved before the spirit leaves the still), boil rates, cut points and condenser temperatures, all of which control the amount of time the liquid remains in contact with copper. These parameters are used to optimise new make spirit flavour and thereafter, by adhering to the optimised recipe and process conditions, ensuring production of a consistent spirit with the desired favour profile.

Distillery design, particularly the design of the copper stills and the copper innards of the condensers, is fundamental to spirit quality (as described above). Once built, making changes is prohibitively expensive so, from the moment of conception, it’s important to design the stills and the condensers to produce the desired style of spirit. Equipment designed to give enhanced contact with copper tends to produce lighter, fruitier spirit. Lesser contact with copper yields spirit which is heavier and meatier in character. Tall stills with long lyne arms (the wide pipe that connects the top of the still to the condenser) and tube and shell condensers, all will give lighter, less sulphury and fruity spirits.  In contrast, short stills with traditional worm tub condensers which give more complex spirits with a heavier, meatier character. Using twin spirit stills to distil the same volume of spirit (as with Annandale Distillery) produces increased copper contact because the surface area to volume ratio reduces as still size increases. Consequently, two smaller stills will give greater copper contact versus one big still with the same combined volume.

The location of copper in the distillation apparatus is also important for reducing sulphur compounds. Interestingly, there are opposite effects in the wash still and spirit still. With the wash still, the copper pot is least effective in reducing sulphur compounds whilst the copper innards of the condenser have a much greater influence. The opposite happens in the spirit still, where the pot has the greatest influence and the condenser has the least. Consequently, the most effective sections of the copper distillation apparatus are the wash still condenser and the spirit still pot. It is noteworthy that these tend to be areas of higher copper corrosion, possibly due to a relatively acidic environment. Higher acidification also facilitates greater removal of sulphur compounds. It may be that the environment in these sections of the distillation apparatus makes them relatively susceptible to acid corrosion which in turn improves their ability to remove sulphur compounds.  Once again having more but smaller stills (as with Annandale) will improve these processes by effectively increasing the surface area to volume ratio. It’s noteworthy that distilleries renowned for their light fruity spirit (e.g. Glenfiddich) have chosen to increase the number of stills rather than the size of the individual stills, when increasing production capacity.     

David Thomson

Co-Founder – Annandale Distillery

Acknowledgement:             Dr Gordon Steele

Single Cask-Single Malt Scotch Whisky vs Vatted Single Malt Scotch Whisky

In this technical note we consider the differences between:

Single Cask-Single Malts – non-chill filtered and bottled at cask strength, typically 55% - 60% ABV (alcohol by volume), as presented in Annandale Distillery’s Man O’ Sword and Man O’ Words collection.

Vatted Single Malts – chill filtered and bottled at either 40% or 43% ABV, as typically purchased in retail and duty-free shops, bars and restaurants.

To produce Vatted Single Malts, a relatively large number of barrels of mature whisky, all from the same distillery, are blended together in a vat to achieve a flavour profile that’s consistent with consumers’ expectations of that brand. Consequently, every bottle of Glenfiddich 12 year old (for example) should taste essentially the same no matter where and when it’s purchased. Achieving this degree of consistency requires a vast inventory of maturing whisky to select from, and the services of a very skilled whisky blender (more of which later). The age declaration, should there be one, refers to the youngest whisky included in the vatting.

Assuming that the bulk of whiskies in a particular vatting were around 12 years old (again, as an example), it’s likely that the average alcohol content would be approximately 55 ± 2% ABV. To achieve a bottling strength of 40% or 43% ABV, the whisky is diluted with water. Ideally the dilution water would be drawn from the distillery source, although treated mains water is often used instead.

There are two principal reasons why Single Malts are diluted to either 40% or 43% ABV prior to bottling:

Economic – the excise duty on a 70cl bottle of 40% ABV Single Malt (£8.04 in the UK) is considerably less than on a 55% ABV cask strength bottling (£11.06). Unsurprisingly, the water used for dilution is much less expensive than whisky. These two factors make 40% ABV Single Malt significantly cheaper than cask strength whisky, allowing it to be accessible to a wider socio-economic group than might otherwise be the case.

Sensory – adding water decreases the pungency of the alcohol. This makes it easier to appreciate the underlying sensory characteristics of the whisky and otherwise, less challenging for some consumers.

Making Single Malt Scotch Whisky more accessible in both of these respects has helped the industry to expand enormously over the last 30 - 40 years.

However, there’s a catch!

When whisky is diluted below ~46% ABV, it will often become cloudy/hazy if subsequently chilled and diluted in-glass by the further addition of cold water and/or ice. This effect is caused by the clumping together of naturally occurring fatty acids and esters to form insoluble micelles. Cloudiness is considered undesirable by some, if not all, whisky drinkers. To prevent this happening, fatty acids and esters are removed using a process known as chill filtration (see Technical Note on Chill Filtration for further details). Almost all single malts bottled at either 43% or 40% ABV will have been chill filtered. Conversely, those bottled at 46% ABV and above are likely to be non-chill filtered.

Chill filtration is often demonised by whisky aficionados in the belief that it defiles the simple, natural purity of Single Malt Scotch Whisky by unnecessarily removing something of its very essence. In practice, only the largest (longest chain) fatty acids and esters are actually removed and these typically have little or no impact on flavour. Nonetheless, chill filtration is a complete anathema to Single Malt purists. Paradoxically, the emergence of a cloudy haze when cask strength Single Cask-Single Malt is diluted and chilled in-glass, is considered by some to be indicative of its unsullied purity!

With non-chill filtered Single Cask-Single Malts bottled at cask strength, the whisky is simply disgorged from its cask into a tank, passed through a relatively coarse membrane to remove particles of carbon (from the charred inside surface of the cask) and then bottled. Whilst it is permissible to add caramel as a colourant, the colour of a Single Cask-Single Malt usually derives solely from pigments extracted from the oak barrel staves. Nothing is added and nothing is taken away other than carbon particles. This is authentic, unsullied and uncomplicated whisky at its absolute purest!

The other defining feature of a Single Cask-Single Malt is the very fact that it derives from just one cask (the clue’s in the name.) As there’s no blending (vatting) of multiple casks to achieve a standardised flavour profile, the flavour of the whisky comes to depend on four factors; the spirit, the cask, the maturation environment (microclimate inside the maturation warehouse) and the length of maturation (i.e. its age). Let’s consider each of these factors in turn:

Spirit – most distilleries strive to produce spirit that’s consistent from batch-to-batch, although some extent of seasonal, ingredient and process related variation is inevitable. Whilst this might ultimately affect the flavour profile of the mature whisky, the unique and defining character of the distillery should still be apparent over and above any such background variation.

Casks – it’s a legal requirement that Scotch whisky should be matured in oak casks. However, oak casks aren’t all the same! For example, they may differ in the species of oak used in their construction; principally American versus French/Spanish oak. The more-open grained Southern European oaks permit easier ingression of the spirit into the wood and consequently, more rapid extraction of oak-derived flavour components. The closer-grained American oak (Quercus Alba) is generally considered to be optimal, even for sherry butts coopered in Spain!

The size of the cask, or more precisely the ratio of the inner surface area of the cask to the volume of liquid it contains, influences the interaction between spirit and oak and consequently, flavour development. (Refer to Technical Note on Small Casks for further details).

Almost all of the casks used in the maturation of Scotch Whisky will have been used previously to mature other spirits (principally bourbon but also other American whiskies, cognac, rum and tequila), wines of various types (mainly sherry but also port, madeira and a variety of red and white wines) and even beer. The extent to which the previous occupant impacts on the flavour of the maturing spirit, depends on whether it’s the first time the cask has been used for Scotch Whisky maturation (known as a ‘fresh’, or ‘once-used’ cask) or alternatively, if the cask has been re-used on several occasions (known as a ‘refill’ cask). The more often a cask has been refilled with Scotch Whisky, the less potent the effect of both the oak and the previous occupant and consequently, the longer the maturation time.

Maturation Environment – when the spirit inside a cask warms up due to diurnal and/or seasonal temperature fluctuations, it expands. This increases the pressure inside the cask, causing expulsion of air through the microscopic gaps that exist between cask staves. The expelled air (the angels’ share) comprises a mixture of water vapour, alcohol vapour and various volatile flavour compounds. When the liquid cools it draws air back into the cask through the same gaps. The greater the temperature fluctuations, the greater this effect. It’s not too difficult to imagine that the cool, moist, salt-laden air drawn into casks maturing on the island of Islay, might have a different effect on flavour development versus the drier, less salty and often cooler air inside a Speyside maturation warehouse. And what about Kavalan Distillery on the island of Taiwan, where the average daytime temperature is consistently around 30+oC but the night-time temperature may drop by as much as 20oC?

The practice of transporting bulk spirit from a local distillery and maturing it in a distant, centralised bonded warehouse containing whiskies from other distilleries, inevitably neutralises the effect of the local micro-climate on flavour development. For those who care about the authenticity of their Single Malt, this practice is regrettable, to say the very least!

Length of maturation – it’s a common misconception that older whiskies are ‘better’ than younger whiskies. Whilst immature whisky typically lacks complexity, depth of flavour and balance, and it may even be harsh and insipid, it’s also possible to over-mature whisky by leaving it in-cask for too long. Should this occur, the sensory profile of the whisky will be dominated by flavour components extracted from the cask staves (known as ‘cask effect’). If excessive, this will mask the more subtle distillery-specific sensory characteristics (distinctive distillery character).

The age at which a particular whisky reaches optimum maturity depends on the nature of the spirit (whisky from some distilleries matures faster than others), the type of cask (‘fresh’ casks mature spirit more rapidly than ‘refills’) and the maturation environment. Also, what constitutes optimum maturity depends on the sensory preferences of individual whisky drinkers (which may be very different).

This suggests that maturation should be determined not by age, but by a combination of several key sensory criteria:

Presence of……                                                                              Absence of……

Flavour complexity                                                                         Excessive cask effect

Depth of flavour                                                                            Harshness

Distinctive distillery character                                                          Insipidness

Balance

We could describe whiskies that satisfy all of these criteria as having reached ‘sensory maturity’ irrespective of age. This allows for the possibility of ongoing flavour development/change with age, provided that none of these sensory criteria are violated. It also allows for the possibility that the maturing whisky could pass through a sequence of several different sensory optima, the longer it stays in-cask.

For any particular distillery, the nature of the spirit produced and the maturation environment are essentially constants, in so far as the spirit should be consistent and all casks should be subject to more or less the same diurnal and seasonal temperature variations during maturation (although the exact location within the warehouse may affect the ambient temperature in the immediate environs of any particular cask). Consequently, the principal determinants of flavour variation for a particular distillery are the nature of the maturation cask combined with length of time in-cask (i.e. its age). This implies that the sensory profile of each cask is likely to be unique, to some extent, but still within the ‘universe’ of what constitutes characteristic flavour for that particular distillery.

Now for the slightly trickier part!

Let’s imagine that we’re going to plot all of the maturing casks produced by a particular distillery (Distillery X) on a theoretical sensory map (Figure 1), where those casks that are most similar in flavour are located near to each other on the map, and those that are more different are further apart (i.e. the greater the sensory differences in the maturing whisky, the greater the distance apart on the map).

Assuming that Distillery X is producing consistent spirit and maturing it in similar casks of common provenance, we’d expect to find a large and dense cluster of casks located on the map (not necessarily at the centre), with the other casks spread around the outside of the main cluster. Some of these other casks would be clumped together in smaller clusters and some would be singletons. Of these smaller clusters/singletons, some would be located quite close to the main phalanx, whereas others would be further afield (because their sensory profiles are rather different). A small number of singletons or tiny clusters of casks would also be positioned towards the various extremities of the map (sensory outliers).

Each cask would have a unique identity in terms of date distilled, type of cask/cask provenance and length of time the whisky has been in-cask (i.e. its age). Whilst it might be anticipated that casks of a similar type/provenance, aged for a similar amount of time should produce Single Malts with very similar sensory profiles (and therefore should locate proximally on the map), this is by no means a given. Indeed, part of the mystique of Single Malt production is that spirit produced in the same batch, filled into seemingly identical casks and stored side-by-side in the same bonded warehouse for the same length of time, will sometimes produce noticeably different whiskies.

Let’s now imagine that Distillery X wishes to produce a 12 year old vatted Single Malt at 40% ABV. The first thing to do would be to edit our theoretical sensory map by notionally deleting all of the casks aged for less than 12 years (Figure 2). It might be imagined that the topography of the edited map would still resemble that of the original sensory map, but the size, shape and density of the cask clusters would perhaps change, and some of the outliers would probably disappear.

Inevitably, the character of the vatted 12 year old would be dominated by the sensory characteristics of Single Malts located in the main cluster because these constitute the bulk of the maturing stock. However, rather than including casks drawn solely from the main cluster, the master blender might choose to make the sensory profile of the vatted Single Malt more complex and/or more balanced (for example) by introducing a selection of Single Malts from the more outlying regions of the sensory map. Bearing this in mind, if the sensory profile of this 12 year old vatted Single Malt was subsequently plotted onto the original sensory map, it would perhaps be positioned away from the centre of the main cluster; its location reflecting some of the sensory nuances introduced by the master blender.

If we further imagine that Distillery X also wanted to produce a 16 year old vatted Single Malt, a similar theoretical process could take place. Presumably, the sensory profiles of the 12 and 16 year old vattings would be different so they’d be located somewhat apart on our theoretical sensory map.

As part of the vatting process, treated water is added to reduce the alcohol content to either 40% or 43% ABV, prior to chill filtration and bottling.

Does this all sound rather theoretical and far-fetched? Perhaps it does, but this is essentially what goes on in the mind of the master blender. They hold in their memory, a mental representation of the target sensory profile of the vatted Single Malt, along with mental representations of the universe of sensory profiles of the maturing Single Malts typically produced by Distillery X (i.e. a mental representation of the above mentioned sensory map). On selecting and nosing samples of maturing stock, the master blender will decide whether or not the whisky in each of these casks is suitable for inclusion in the vatting. In doing so, they systematically build-up a mental representation of the vatting that they’re creating, decide what else needs to be added to the blend to achieve the target sensory profile and then they’ll sniff and select appropriate casks from their inventory to finalise the blend. Exactly the same process is used for creating Blended Scotch Whiskies.

Although most master blenders can readily describe what they’re doing and why, blending is essentially a non-verbal, non-conscious process (even if the blender might think otherwise). Now you know why master blenders are such ‘rare beasts’, why it takes them so long to train and become proficient and why huge Scotch whisky producers such as Diageo, Chivas Brothers, William Grant, Edrington, Beam Suntory, etc., are all obliged to rely upon just a few amazingly skilled and talented individuals!

But where do Single Cask-Single Malts fit into this picture?

Cast your mind back to the theoretical sensory map of Single Malts from Distillery X (Figure 1). Now delete from the map, all of those casks that have not yet attained sensory maturity (as previously defined). The mission hereafter of the whisky/sensory expert (if not the master blender) is to identify two ‘types’ of Single Cask from the depleted sensory map:

1/ Single Malts from casks drawn from the centre of the main cluster (as described above), which exhibit the fundamental sensory character of Distillery X (Figure 3). In practice, these will probably be several-times-used ex-bourbon casks (refills) where the cask effect is somewhat muted, or once-used (fresh) ex-bourbon casks where the Single Malt has reached sensory maturity without exhibiting excessive influence from the oak (or its previous occupant). Obviously, the latter are likely to reach sensory maturity more quickly than the former.

2/ Single Casks that are characteristic of Distillery X but exhibit interesting deviations. These could be ex-bourbon casks of one type or other which, for whatever reason, have delivered a sensorially mature Single Malt that’s outside the main cluster. However, it’s more likely that these will be the product of some less run-of-the-mill casks, as detailed previously. It’s important that these shouldn’t be construed as outlandish, but interesting and slightly more unusual whiskies from within the sensory universe of Distillery X (Figure 3).

At Annandale Distillery we don’t currently produce vatted Single Malts. Whilst we have total respect for Master Blenders and the wonderful blended and vatted Scotch Whiskies that they create, we prefer to select Single Casks that exemplify the unfettered character of our peated or unpeated Annandale Single Malts, or slightly unusual departures from the centre-ground of Annandale’s peated and unpeated sensory maps. We don’t chill filter because there’s no point and we also prefer to bottle our Single Malts at cask strength, allowing the whisky drinker to dilute-to-taste.

Our Rare Vintage 2014 and Vintage 2015 are exemplars of ‘main cluster’ Single Cask-Single Malts whereas our Founders’ Selection range are ‘interesting deviations’ from the main cluster. Indeed, one of the greatest delights that comes from owning a distillery is in sampling and tracking the maturing stock in our bonded warehouses and singling out those casks that satisfy the stringent standards we set for Annandale’s Single Cask-Single Malt bottlings.

As our guarantee of uniqueness and provenance, every bottle of Annandale’s Single Cask-Single Malt is identified by cask number (e.g. 2014/98) and sequential bottle number (e.g. 98 of 236). We hope they’ll delight your senses and your emotions!

Single Cask-Single Malt – authentic, unsullied, uncomplicated Scotch Whisky at its absolute purest!

David Thomson

Annandale Distillery