An Interview: How Does Whisky/Whiskey Get Its Flavour?

Below was conducted and an interview with Nikita Beverage Co. regarding whisky and its connection to vanilla flavour.

What is whisk(e)y?

Whisky is a spirit produced from cereals (primarily barley), water, and yeast. While this seems like a simple categorization, international definitions are actually quite convoluted. I will give you a little taste of this before we start to nail down terminology.

Historical debate aside, whisky is now primarily produced in Scotland, Ireland, America and Japan. That being said, new markets (e.g. Taiwan, France) are starting to produce some world class products, and “craft” whisky producers are becoming globally ubiquitous; similarly to the “craft beer” bourgeoning.

Whisky or whiskey?

One common syntactic difference you may encounter when exploring whisky is whether or not to include that extra “e”. Is it spelled whiskey, or whisky? Well, the Scots spell it whisky, and therefore so to do those countries which seek to emulate their styles (e.g. Japan). On the other hand, the Irish have that extra “e”, making it whiskey in Ireland. This Irish spelling is also seen in America, most likely an adoption from Irish immigrants in the 1700s. The difference in the spelling likely stems from the translation from the Scottish to the Irish Gaelic tongues.

What is Whisk(e)y?

European definitions

Essentially, each country has its own definition of what constitutes a whisky, so let me present how the EU defines it:

Whisky or whiskey is a spirit drink produced exclusively by: (i) distillation of a mash made from malted cereals with or without whole grains of other cereals, which has been: saccharified by the diastase of the malt contained therein, with or without other natural enzymes, fermented by the action of yeast; (ii) one or more distillations at less than 94.8 % vol., so that the distillate has an aroma and taste derived from the raw materials used, (iii) maturation of the final distillate for at least three years in wooden casks not exceeding 700 litres capacity.
The minimum alcoholic strength by volume of whisky or whiskey shall be 40 %.
No addition of alcohol, diluted or not, shall take place.
Whisky or whiskey shall not be sweetened or flavoured, nor contain any additives other than plain caramel used for colouring.
Ref: REGULATION (EC) No 110/2008 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL

How whisky is made?

There are of course tacit differences in the whiskies produced in these different countries, beyond the inclusion of the “e” in the spelling. These differences lie in the distillation process, the stills, the raw ingredients and the maturation process. For a general overview of how whisky is made I will use single malt Scotch whisky as the model, as it is that which most closely resembles the batch-wise process employed in home distilling.

If you’re interested in exploring the differences in the production of other styles, take a look into entry level overview books on whisky (e.g. Whiskey: The Definitive World Guide).

The first few steps of whisky production are actually quite similar to that of beer production; imagine producing an un-hopped wort, without having to boil it. Of course some of the parameters will be different, as the final objectives are drastically different.

Whisky Distillation

Batch Distillation Flow Diagram

It starts with malted barley

Malted barley is the principle cereal used in single malt Scotch whisky production. Americans, like Scotch grain whisky producers, also use corn and wheat, among others; while Irish use un-malted barley, in significant amounts.

Malted barley is used as a source of enzymes which catalyze the hydrolysis of starch (break them down into sugars), as well as being a source of starch itself. A whisky distiller uses barley cultivars of the same species as would a brewer (e.g. Hordeum vulgare L.), however distillers typically use varieties which deliver significant improvements in production performance. The reason being that distillers, in contrast to brewers, concern themselves more with maximizing fermentable extract from the malted barley to optimize alcohol yield; as opposed to yeast health, product clarity, mouthfeel, head-retention and malt flavours; which are parameters of more importance to the brewer.

The Wash

The first stage in the process, after the cereal has been milled, is the “mashing”; where the barley malt is added to water at a target temperature, and the starches are extracted from the grains along with the enzymes which will turn these into sugars. This extraction takes place in several stages, for varying reasons. The resultant sugars are required for fermentation; transformation of the sugars into the required alcohol. The length of this fermentation varies, however it typically only lasts for 24-72 hours.

The barley has been removed by this point and we’re left with a sugary liquid, called the “wash”. The whisky makers will add yeast to this liquid for the desired fermentation to begin, converting sugars into alcohol. At the end of this step, the wash will only contain 5-10% alcohol, by volume.

Distillation

The alcohol concentration is then further increased by means of subsequent batch distillations. The principles of distillation are beyond the scope of this article, but “in a nut-shell” the act of distilling involves the separation of volatile compounds from a mixture, based on the contrasting boiling points of said compounds.

The resultant liquid, obtained from distillation, is called “new-make” spirit. While the production of this “new-make” only takes a few days, the spirit still needs to age for several years before we can call it whisky. This elusive, technical, and slightly “magical” process of “maturation” is often not well understood, and it is here where I was asked to “shed some light”.

Does whisky actually have any flavour after distillation?

Most definitely! New make spirit is full of flavor! Some of the flavors are desirable, some are not; the overall expression however is generally considered to be undesirable, and referred to as “immature”. It is for this reason that we need to age the spirit in wood for several years; to render the spirit more palatable.

Before talking about flavor let me make some distinctions. When we talk about sensory evaluation we like to discern between the senses. “Flavor” typically references a compounding effect of taste and aroma, generally speaking. As there are so many more aroma receptors in your olfactory system, then there are taste receptors on your tongue, many of the nuanced “flavors” are typically coming from the smell. I could go on for ages about sensory science but let’s just leave it at that for now.

Nearly every parameter up until this point will affect the new-make spirit’s flavor: raw material selection (e.g. water hardness), wash preparation (e.g. mash temperature), yeast (e.g. physiological state), fermentation conditions (e.g. pH), and the distillation (e.g. still shape). New-make spirit can give an indication of potential spirit quality as well as the required maturation time. In fact, in the Scotch whisky industry new-make spirits are frequently traded prior to maturation, with new-make sensory character being the primary determinant of quality.

Ultimately, the flavors you are going to get with the new-make spirit will vary significantly, based on all of the parameters aforementioned. However, some flavor descriptors commonly associated with new-make spirits are: “soapy”, “mouldy”, “musty”, “rancid”, “cabbage”, “sulphury”, “meaty” and “feinty” – which are typically seen as undesirable – and “sweet”, “fruity” and “solvent-like” – which are often thought of as desirable.

If so, does it matter which yeasts or grains are used?

Big time! There are several schools of thought on this. Traditionally, in Scotland let’s say, the yeast strain was selected based on its ability to undergo a rapid, reliable, and high alcohol yielding fermentation. The selection was typically not made for its nuanced sensory impacts to the new-make spirit. However, leaving fermentation performance aside, in the metabolism of the wash sugars and amino acids, the yeast produces a variety of secondary metabolites, or flavor “congeners”, imposing a strong influence on the development of spirit flavor.

The same can be said for the cereal selection. As I previously alluded to in the introduction, this choice was – much like yeast strain selection – historically made with the efficiency of production being the primary driver, in lieu of sensory qualities. Cereal selection will not only directly influence flavor (e.g. “nutty” aroma associated with pyrazines created from the Maillard reactions imposed on the barley during malting), but it will influence it indirectly (e.g. certain cereals require mashing techniques which will ultimately lead to reduced ‘fruity’ ester formation during fermentation).

Couldn’t you just make whisky out of sugar beets?

Of course you can make a distillate from sugar beets; they are in fact an excellent substrate. However, as defined, the resultant spirit would not be a ‘whisky’. They are excellent candidates for vodka and gin production, however. Sugar beets lack the enzymes and yeast nutrients that you would get from a malted cereal.

Spirit “types” have traditionally evolved from specific regions, and from sugar sources native to those regions (e.g. rum is made from some form of sugarcane and is therefore typically made in locations which grow sugarcane).

What happens to whisky after distillation?

Following distillation the new-make spirit is matured, typically in oak barrels. This process lasts anywhere from 3 to 30, or more, years. The spirit maturing in each individual barrel will acquire unique nuances to them, and they are monitored accordingly along the way. The aim here is to turn the “immature”, colorless, new-make spirit into the visually appealing, “mature”, spirit desired by the consumer. It is the task of the “master blender” to judge the unique profile of each barrel, and decide which to blend, and when, to achieve a signature flavor.

Maturation is arguably the most important phase in the final flavor profile of whisky, and it is here where the whisky acquires its color. The chemistry involved in the maturation process is staggering, and its extended time frame has resulted in a lag in establishing a substantial body of research. At this point, whisky maturation is a somewhat of an art.

How are the wooden casks chosen?

This is a loaded question, with a very long, and possibly subjective, answer. A comprehensive answer would involve the chemistry of maturation, wood chemistry, cooperage techniques and some discussion on sensory science and artistry.

Can you give us an overview on some of these topics?

Certainly.

Oak seems to be the modern wood of choice for cask construction; and for good reason. Oak exhibits several physical characteristics which make it desirable for containing, storing, and maturing whisky, namely: high porosity, strength, durability and flexibility. The structure lends a low oxidation rate to its contents and it contributes a characteristic organoleptic quality to the spirit.

In a broad sense the wood is composed of two relevant fractions: the sapwood and the heartwood. The heartwood is the section which coopers (ones who build wooden casks) desire due to its additional hardness and flavor contributions. These two portions of the wood are separated by balloon like structures called tyloses; which block the conductive vessels running from the sapwood to the heartwood, lowering the permeability of the lumber when made into staves (individual longitudinal segments which make up a cask). The tyloses are one of the main limiting factors for selecting suitable wood for coopering, as they are only found in 20 varieties of oak; all of which are white oak.

The heartwood is composed of three main components: cellulose, hemicellulose and lignin. These are all polymeric structures that make up the cell walls, with lignin accounting for most of the intercellular space. Combined, these three components make up roughly 90% of the heartwood, with the remaining being “extractives” (e.g. volatile oils and acids, sugars, tannins, pigments); mostly in the form of ellagitannins and other volatile compounds which are free to diffuse into the maturing spirit over time. All of these components will play a role in the flavor development of the whisky.

American vs European cask construction

There are two prominent methods employed for cask construction: American and European. The main differences lie in the manner in which the staves are cut, dried, treated, and assembled. The European method involves air drying the wood and as a result takes drastically longer; in the time frame of several months to several years. In contrast, the American method typically only takes up to a couple of months; as kilns are used to do the bulk of the drying. Following drying, the staves are shaped, assembled into a cask, and bent into shape; with the quality and longevity being decided very much by the technique used.

When I say “treating” the cask, I am referring to either charring or toasting the inside of it. European coopers “toast” the inside of the cask, bringing the staves to around 200°C for about 20 minutes or longer, based on the level of toasting needed; light, medium or heavy toast. Conversely, Americans steam the cask at roughly 95°C for 10 to 20 minutes to soften the wooden fibers and bend it into shape. The American coopers will then “char” the still wet wood by exposing the inside to 230–260°C flames for 15 minutes. This is a crucial step, regardless of whether toasting or charring, as it will have a significant influence on the chemical makeup of the oak as it carbonizes the structure compounds via pyrolysis.

Maturation is intended to improve spirit quality through: additive (e.g. extraction of wood compounds), subtractive (e.g. adsorption onto char layer) and interactive reactions. The magnitude and rate of change is dependent on cask and process factors (see Maturation Parameters). The composition of the 10% of “free” extractives in the heartwood, which I had previously mentioned, has a major effect on maturation and varies between oak species, treatment technique, and the number and type of previous maturation.

The choice of cask, ideally speaking, will depend on matching these factors to the intended flavor outcome. Let me present some examples of some decision criteria for such matching:

European oak may produce whiskies with higher levels of color and extractives; with prominant vanilla, fruity and sweet qualities.
Toasted casks are known to generate higher levels of color and extractives than charred casks; as much of these compounds are lost due to carbonization and volatilization during the charring process.
Toasting is a common treatment for wine caskss, so if using an ex-wine cask you will likely get a higher level of extractives and color which will be compounded by the transfer of remnant wine components
Using a new-charred cask will likely impart “woody, “vanilla”, “coconut”, and “resinous” qualities. These are usually prominent and can actually mask many of the distillates other more nuanced character. The charred layer acts an activated carbon layer and physically absorbs many sulfurous compounds from the spirit.

This is of course idealistic and often flavor outcome is either not known, or needs is allowed to proceed organically to produce a certain product line.

Traditionally choice has come down to availability, such as is typically seen in the Scotch whisky industry; or legislative impositions, such as experienced in the US. American whiskey makers were legally required to use “new” casks for each spirit maturation, in what many expect was an attempt to keep cooperages in business.

This had a direct impact on the Scotch whisky producer’s cask selection. With American distillers essentially having to dispose of casks after their first use, this created an inexpensive supply of used whiskey casks for the Scotch whisky distillers. For this reason, and from a similar supply from Spanish Sherry bodegas, the choice of cask in the maturation of Scotch whisky has has been typically limited to ex-bourbon and ex-sherry casks. However, distillers are ever-more experimenting with casks from novel sources, for new flavour profiles (e.g. French wine barrels, Hungarian wine barrels).

What type of flavours develop during the ‘ripening’?

This answer is quite complex, so I will speak generally. Maturation delivers an improvement in flavor quality from the development of mellow, or mature, characteristics from the wood; and the loss of immature characteristic of the new-make spirit (e.g. “grassy”, “sour”, “oily”, “sulphury”). There is a wide array of flavors which develop, but some common descriptors are: “vanilla”, “spicy”, “floral”, “woody”, “oaky”, “smooth”, and “sweet”.

Can you explain how some of these flavors are formed?

As I mentioned earlier, maturation involves 3 separate categories of chemical and physical interactions acting in synergy:

additive (e.g. extraction of wood compounds and remnants of previous contents)
subtractive (e.g. evaporation, adsorption onto char layer, solution changes that affect the release of aroma compounds)
interactive reactions (e.g. acetal formation, oxidation, esterification)

Additive Reactions

Additive reactions involve three main groups of “extractives”:

the diffusion of those compounds remaining in the wood from its previous contents (e.g. sherry, bourbon, wine)
the diffusion of those volatile compounds from the heartwood which are “free”
those which originate from the structural components of the wood, made available through heat treatment (e.g. lignin degradation products, Maillard reaction products), and to a lesser extent through ethanolysis and oxidation

This diagram can help give you an idea of what kind of compounds are to be found in the wood.

For the purpose of our discussion, the most important groups here of which to take notice are the oak lactones and the lignin degradation products. The oak lactones impose the characteristic “oaky-ness” to the spirit. In contrast, lignin degradation products (e.g. vanillin) enter the maturing spirit primarily through heat treatment and to a lesser extend through oxidation, hydrolysis and ethanolysis. These compounds typically provide “vanilla”, “floral”, “spicy”, and “smooth” attributes to the flavour or the spirit.

Although European oak wood has more extractives, American oak is said to impart more flavor to a spirit. The reasons for this are complicating, but partially involve the lower aroma threshold of the cis- oak-lactone isomer; with American oak boasting higher levels of said isomer than the European counterparts.

Subtractive Reactions

The second category of reaction, subtractive, typically alters the new-make character through the loss, or suppression, of aroma compounds; involving:

physical reactions:
- evaporation of the low boiling point compounds through the wood
- absorption of compounds onto the wood and charred layer of the cask
chemical reactions which lead to the change in volatility of certain key aroma compounds

Generally, the physical reactions help to remove many “immature” qualities of the new-make spirit, such as the evaporation and absorption of sulfur compounds onto the charred layer; and with them those “sulfury”, “vegetable” and “rotten” qualities. Aside from the loss of certain flavor compounds, the evaporation loss will also see either the loss of alcohol, or water, to the atmosphere. In Scotland, as the ambient air is quite high in moisture, alcohol is lost to the atmostphere – they call this the “angel’s share”. This will change not only the concentration of alcohol in the cask, but the organoleptic quality of the spirit in-and-of itself; because as you know certain compounds are more, or less, soluble in alcohol than others and this will alter the balance of certain chemical equilibria (e.g. free aldehyde and acetals).

Interactive Reactions

The interactive reactions occurring throughout maturation are the most complex and might have the largest impact on the final spirit, however that would require another discussion! I can, however, present some highlights:

initially, ethanol is oxidized to acetaldehyde and acetic acid which causes a drop in pH; affecting the ionization state, and thus the volatility, of certain compounds.
some desirable “fruity” esters are lost and more generic ones are created from free acids and ethanols
the reduction in free acids helps to reduce the “sour” quality of the spirit
Aldehydes and acetals are formed which give desirable “fruity” quality
Oak tannins are oxidized to give add more “vanilla” quality

As you can see maturation is a very dynamic process, and it requires a large amount of experience, expertise, and skill; a knowledge of the science will only get you so far. A proper outcome will ultimately depend on many factors.

How does vanilla fit in this story?

Vanilla of course refers to the very quintessential flavor associated with the extract from the vanilla plant or, more precisely, the fruit of said plant. Well, this flavor can actually be attributed to a group of chemical compounds found within these fruits in high abundance; the most notable of which being vanillin and syringaldehyde.

Vanilla is a very interesting flavor, and it is one of the key characteristics attributed to some of the best whiskies in the world. How does this come to be? As you may have guessed from the previous question, it comes from the oak hardwood.

Vanillin is present in the “free” extractives of the oak heartwood, but only in small quantities. The main source is made available from the heat treatment during cooperage of the casks; specifically it is one of the lignin degradation products. While the free extractives in the wood can never be replenished, subsequent heat treatments will liberate more of the vanillin compounds from the lignin; and these heat treatments are therefore seen as “regenerative”.

Aside from heat treatment, smaller amounts of both vanillin and syringaldehyde are formed by oxidation and ethanolysis during the maturation process.

How did you become interested in food science/distillation?

I would say my interest was initially sparked from the scientific aspect, in both regards. Growing up we are taught to compartmentalize subjects in school, you know – math, chemisty, history, and so on; while my studies in chemical engineering really opened my eyes to the beauty of scientific consilience.

Distilling and fermentation were key focal points in my studies; distilling – being a well-studied phenomena combining physics, chemistry and mathematics – and fermentation – bringing biology into the picture – being somewhat on the fringes of scientific research, in many regards.

This, however, became more of a passion for me many years later. My work life began introducing me to many new areas of the world, opening me eyes to not only the vastness of the planet, but the different ways in which cultures approach its appreciation. Spending a large amount of time in France, in particular, really substantiated my passion for food. There is such a deep appreciation in rural France for not just meals, but the intricacies and nuances of individual ingredients.

Our connection with food seems as integral to our existence as the air we breathe, yet life is so busy now, and food is so readily available, that we can easily disconnect ourselves from it. This is such a shame because there is just so much excitement to be had there. From a plant creating food from the sun’s photons, to the proteins inter-linking to retain foam on my beer, to the cognitive effects from sharing a great meal with family and friends; there’s so much that one can appreciate. I know for me, just sitting down having dinner with my parents and sister is as archetypical as a fond memory gets.

The lives of all creatures revolve around food in some regard; perhaps advances in cooking techniques have been the driver for the evolution of our minds and our technology. Primatologist Richard Wrangham interestingly hypothesized that simply using fire to cook our food helped enable us to evolve our larger brains and cooking itself became a primary focus of human social activity; or “cooking made us the social and intelligent species we are today”, as he puts it. Maybe we just lucked out in our discovery of fire! Regardless, it’s fascinating to even look at the chemical changes involved in cooking a tomato!