The Physics of Batch Column Stills and Bubble Plates

Until recently the graceful pot stills of Scotland were the most familiar images people had when you talk about distilling. Now batch column stills, also known as hybrid, reflux, or Lomond stills, common in the eau de vie and craft distilling industries, are far more recognizable than they used to be. These occupy an important middle ground between simple pot and more complex continuous stills, representing an evolution of the double retort pot still. In the simplest way of thinking about them, batch column stills are a way to create more refined spirits while using a smaller footprint by increasing the number of times that vapor is condensed back into liquid between the pot and the condenser.

New Deal Distillery's hybrid still

I've previous written about the physics of pot stills, which gives important background explaining concepts such as separation/resolution and reflux. Pot stills come in a bewildering array of shapes that are designed to influence reflux, but columns put all of that complexity inside far more plain-looking tubes. All of them operate by putting some kind of material in the path that the vapor travels from the pot to the condenser to create more opportunities for the vapor to condense and flow back towards the pot. The simplest method involves packing the column with small, high surface area items made of copper or a non-reactive material. In the same way that vapor condenses on the walls and flows back into the pot in a pot still, packing a column with material creates even more surface area for that process. Because the surface area is so high in a packed column, the vapor stream will also exchange material with the wetted surfaces in the column, further enriching the vapor stream with lower boiling constituents and depositing higher boiling constituents in the liquid. One of the significant challenges in setting up a packed column is ensuring that there is still enough void space for vapor and liquid to pass through the material without creating an unsafe amount of pressure. Another downside is that packed columns often have to be disassembled and emptied to clean the packing material thoroughly, which can be a significant challenge with larger columns. While these materials give a very high amount of reflux, care must be taken with the amount and type of packing material. A highly packed column may only be suitable for the production of neutral spirits because the product coming off the still will be almost flavorless. Using copper mesh also creates more opportunities for the metal to catalyze chemical reactions, which may be good for reducing sulfur compounds in the vapor, but can also lead to higher levels of acetaldehyde in general or the carcinogen urethane from fruit mashes with high amounts of cyanide in them.

The art of constructing hybrid stills is figuring out how to create a greater amount of reflux than a simple pot still while still producing flavorful spirits. This generally requires some kind of plate design, which has the advantages of letting distillers more precisely control their reflux ratio.

The simplest setup in this category is the sieve plate, which is just what it sounds like - a perforated plate, usually made from copper. The size of the plate and the ratio of holes to surface determine the amount of reflux each plate generates. Basic sieve plates have two significant advantages - first, they are simple to manufacture and thus cheap and second, they can be installed in the column on pivots that allow them to be turned 90º so that they create a minimal amount of reflux. This allows a distiller to tune the amount of reflux in the column for the type of spirit that they want to create, which is analogous to being able to change the shape and height of a pot still. This flexibility is especially important for craft distillers who want to produce multiple types of spirit on the same equipment. A larger multi-plate column can be fully engaged for making vodka, while most of the plates can be disengaged for more flavorful spirits like whiskey or brandy. Alternatively, creative plumbing can allow multiple columns to be used in series and selectively bypassed, so both a short and a taller column can be used for lighter spirits while the shorter column alone can be used for more flavorful spirits.

Sieve plates with sufficiently small holes operate with a layer of condensed liquid on top of them that is kept from falling back through the holes by vapor pressure. If the vapor pressure is not maintained at an adequate level liquid can 'weep' through the holes, reducing the efficiency of the plate. Some plate column stills have sufficiently high reflux ratios that liquid will collect on the plates passively, others require the plates to be preloaded with wash or water before the run, and many will use a dephlegmator, which is a partial condenser at the top of the column, to build up liquid on the plates. The plates will also need something called a downcomer (see diagram at right), which is a tube that allows liquid to drain from one plate to the plate below it. This pipe is built with a fixed or variable amount of height above the plate to ensure that the liquid level doesn't drop to zero. Similarly the lower end of each downcomer pipe is surrounded by a weir, which prevents vapor from traveling up the downcomer and bypassing the plates. The arrangement of the downcomers forces the liquid to flow across the plate from one side to the other, ensuring good contact between the vapor and the liquid as it proceeds back down to the pot. This setup means that vapor passing through each plate will exchange its heat with the liquid, depositing lower boiling compounds in the liquid phase and vaporizing higher boiling compounds to proceed upwards in the enriched vapor. In essence each plate becomes a small pot still, with the vaporization and condensation processes happening multiple times in miniature. This can be seen as an evolution of single- or double-retort pot stills (primarily found in rum distilleries) and thumpers (primarily used in bourbon distilleries) where the output of one pot still is passed through liquid in a subsequent pot still, then some of the liquid content of the pot is passed back to the previous still.

Bubble cap and valve plates are the next step up, meant to more effectively maintain the liquid level on the plate. While there is significant variation in design, all consist of small pipes with caps or valves on top of them. The liquid on the plate is prevented from passing through the pipe by the pressure on the cap or valve, while vapor can flow up and around to pass through into the liquid. This design allows the still to operate at a lower vapor flow rate than a simple sieve plate because the design reduces or eliminates the chance of weeping.

With all of this careful engineering, what's the point? There are any number of factors that can be pointed at, ranging from smaller footprints (no need for a huge pot still to produce light spirits when you can do the same thing with a more compact column), to efficiency (high reflux columns can be run harder than a pot still without loss of separation), to control and flexibility (pot stills only have two axes of control - heat input and condenser cooling water input). A classic example comes from the world of unaged fruit brandies or eau de vie, which were some of the first major users of batch column stills. This is because they found that the products from single pass distillation in batch columns were significantly different than double distillation in simple pots for certain types of fruit. For instance, one study found that while total ester levels were higher in pot distilled cider brandies, the levels of higher alcohols were elevated in the reflux column distillates.

The utility of batch column stills is even more clear for the craft distilling industry. They are faced by an array of challenges stemming from the huge amounts of capital that are needed to start a commercially viable distillery. Batch column stills present solutions to many of those problems. While they are more expensive than simple pot stills, they are far less expensive than continuous stills or multiple pot stills. They can be configured to produce an array of different spirits from the same system, allowing a new distillery to make lighter unaged spirits that can be sold immediately as well as heavier spirits that are designed for aging in casks. The increased efficiency and smaller footprint both help to save money and maximize the utilization of valuable space, especially for distilleries located in urban areas with higher real estate prices. Last, but not least, they allow the dynamics of distillation runs to be radically altered in comparison to pot stills.

Diagram of New Deal Distillery's hybrid still

The combination of a dephlegmator and a plate column allows for almost unprecedented control over how a distillation run proceeds. With a simple pot still, the only choices available are how to input heat into the pot, the rate that cooling water flows into the condenser, and where the cuts are made. While these tools are obviously sufficient to create some of the best spirits on earth, they require a lot of trial and error to perfect. With a batch column still equipped with a multi-plate column and a dephlegmator, a skilled operator can do something completely impossible with a simple pot still - establish, albeit temporarily, equilibrium. With 100% reflux the pot and column effectively become a closed system. The components of the heads will be compressed into the vapor phase near the top of the column, while the tails are all firmly in the pot. By reducing the flow of the dephlegmator a bit, the heads will leave the column in a comparatively small volume with very little alcohol. With further reduction in dephlemator flow the hearts will then come in behind at a constant ABV, unlike the steadily declining ABV of the hearts fraction from a pot still. Further tweaks will also compress the tails fraction so that very little of the fusel oils contaminate the hearts. This allows the creation of a very clean hearts fraction that can be bottled directly or that needs very little aging to round off its remaining rough edges.

An example of this flexibility is Westland Distillery in Seattle. Though they run a standard double distillation process for the majority of their spirit, their wash still is a batch column still. It is primarily used as a pot still with the dephlegmator turned off and open drains (see below) that empty the plates back into the pot, giving a greater amount of copper contact but with essentially pot still characteristics. Its full capacity is used to redistill the combined heads and tails from previous runs on the spirit still, with the dephlegmator fully engaged and the drains closed to keep the plates flooded to compresses the heads, then slowly reducing the cold water feed into the dephlegmator to extract a cleaner set of hearts, albeit with different character from their standard double distilled spirit.

Westland's wash still - see plate drains on the left side of column

There has been a long-running debate within the spirits community about whether batch column stills count as pot stills. On the one hand, they both operate in batch mode, which influences how the spirit is shaped through cuts. On the other hand, a fully equipped hybrid still with a column and dephlegmator can manipulate the process of distillation in ways that are simply impossible with pot stills, producing very different sorts of spirits. Being able to produce clean, high-proof spirit in a single pass is fundamentally different than simple pot stills, which is demonstrated by their use in the fruit brandy industry. At the same time, the line between the two is blurred by double retort pot stills, which are widely acknowledged to be pot stills, but share characteristics with batch column stills utilizing only a few cross-flow plates. While there is a hard break between batch columns and continuous stills, what the debate really comes down to is mystique - are people willing to pay more for pot still spirits because they expect a certain level of quality and character? I would argue that batch columns can produce spirits of equal quality, as long as the distiller is skilled enough to use their equipment to its maximum potential. If the reputation of batch columns has become tarnished, that is because of the people using them, not because of the technology itself.

The Physics of Double Retort Pot Stills and Thumpers

Pot stills are the oldest form of distillation and continue to be used across the world, but one of their main limitations is that the maximum ABV that can be attained in a single distillation is ~45% from standard 8-10% ABV wash. The small amount of high proof spirit from the first distillation will also be highly contaminated with low boiling compounds that range from unpleasant to unsafe, so it has traditionally required at least two pot still distillations to produce flavorful, drinkable spirit.

But double distillation is slow and expensive. Each run requires charging the still, consuming lots of fuel to heat it up, and cleaning out the remains in the pot after a run is complete. It also requires complex logistics to balance the flow of raw material through mashing, fermentation, and distillation so that equipment is being efficiently utilized. Distilling has always been a volume-driven business, so the more time it took to produce marketable spirit the less money a distiller was making on the vast amount of capital they had sunk into their plant, inputs, and labor. Somewhere in the 17th or 18th century distillers had the clever idea of hooking multiple pot stills together to perform multiple distillations simultaneously in series.

These types of stills are now uncommon, but can still be found in a number of rum distilleries across the Caribbean such as the double retort systems at DDL in Guyana (both the Port Mourant wooden 'double' pot still and John Dore high ester still), Appleton, Hampden, and Worthy Park in Jamaica, and Foursquare and Mount Gay on Barbados. They can also be found in many bourbon distilleries coupled to column stills under the title of 'doubler' or 'thumper'. All perform a secondary or tertiary distillation to boost the ABV of the output without having to manually perform a second or third distillation.

I've written before about the physics of pot stills and that background will be important for understanding what happens when they are connected to a retort. In essence all of this comes down to a bit of plumbing - while the lyne arm of a traditional pot still is connected directly with a condenser, a retort pot still passes the lyne arm into a additional pot still. This can either direct the hot vapor into liquid where it bubbles through and heats the contents through residual heat or the vapor can first be condensed then passed into the next pot where it is heated again and undergoes another distillation. In either case some portion of the liquid has to be passed back to the previous pot to maintain the liquid level as water and feints are left behind from the increasingly enriched vapor. Importantly, when this is a batch process being fed by a pot still all that is being changed is how many times the vapor is being redistilled. The distiller still makes heads, hearts, and tails cuts just like with a simple pot still.

Double retort pot still with rectifying column at the Worthy Park distillery from The Floating Rum Shack

One of the most important parts of this process is what goes into the retort. If you put pure water in the retort the ABV of the output will not be significantly boosted, but some of the more water-soluble compounds may be scrubbed out, kind of like a hookah or bong. At many distilleries that use these systems, the retorts are loaded with what are called 'low wines' and 'high wines' (see labels on retorts in photo of Hampden Estate below), which are respectively the tails and heads from previous distillations diluted to differing degrees depending on the desired output. Others, such as DDL, combine the heads and tails together before loading them into the retort. This replicates the practice in many distilleries with simple pot stills of recycling feints back into the wash still for redistillation. A visual description of that process can be found here. For more flavorful spirits, stillage or dunder (what remains in the pot after a previous run) can also be charged into the retorts to boost the ester content in the Cousin's process (this is a sufficiently complex topic that it will get its own post at a later date).

To cite one example of how a retort pot still operates, this report claims that Appleton's double retort pot still starts with 8% ABV wash that is converted into roughly 30% ABV output, which goes through the first retort charged with 30% ABV low wines and is converted into roughly 60% ABV output, which goes through the second retort charged with 75% ABV high wines to give a final product at 80-90% ABV.

Double retort pot still at Hampden Estate from Leonardo Pinto

While the dynamics of retorts fed with the condensed output from the previous still (doublers in bourbon parlance) are basically the same as any other pot still, a vapor feed creates far more complex dynamics. What happens to the vapor bubbling through the liquid in the retort is dependent on a large number of influences that will shape the output. Thanks go out to user The Black Tot from the Rum Project forums, who did a pretty thorough job of thinking through what's happening in a retort.

Vapor from the pot still emerges into the liquid in the retort, initially at a much higher temperature than the liquid. The height of the liquid in the retort creates pressure that compresses the bubble. These forces will make the bubble partially or completely collapse as the temperature drops and the pressure rises, driving the vapor within the bubble below its condensation point. The heat from the vapor, both from its initial temperature and the gas to liquid phase change, will be added to the liquid. That process will be more or less complete depending on the temperature of the liquid, the pressure in the liquid where the bubbles emerge, and the size of those bubbles. Low temperature liquid with a lot of depth and small bubbles will encourage complete collapse, while higher temperature liquid without much depth and larger bubbles will be more likely to reach the surface of the liquid and burst. The first case will give better separation as the liquid is gently heated, while the second case will give less separation as the liquid is quickly heated and boils turbulently, mixing up heavier and lower boiling components.

The interplay between the size of the retort and the volume of the charge in it play an important role in determining how much heat will be lost from the system through radiant cooling and influence how much reflux is generated in the retort. A larger retort with a smaller charge will result in more cooling and more reflux, while a smaller retort with a larger charge will result in less cooling and less reflux. The charge will be influenced by how the stills are set up to handle the mass balance of the system - vapor enters the retort, gives up its heat, and the alcohol is preferentially vaporized again. The enriched vapor stream leaves water behind, which will tend to increase the amount of liquid in the retort. This is usually dealt with by passing some of the liquid back to the previous pot, but that can be plumbed in different ways. An outlet with a vapor lock part way up the wall of the retort can help to maintain a constant liquid level, while one leaving at the bottom will have a flow dependent on relative pressures in each vessel, though this can also be controlled with a valve if the distiller wants to vary the conditions over the course of a run. In some ways this is also analogous to a purifier pipe in the lyne arm of a pot still, passing material back to be redistilled and giving a greater amount of total reflux through the system.

All of these parameters give a distiller multiple ways to control the process and output, resulting in full-bodied 'pot still' spirits in a single run that would take a standard pot still two to three distillations to match. In my next article in this series I will describe how this concept was transformed into the batch column stills that have become so common in the craft distilling industry.

Why Sherry Bodegas and Whisky Distillers Want Very Different Casks

During conversations on Twitter, I have seen confusion about what constitutes a 'good sherry cask'. There was some discussion in the spring 2014 issue of Whisky Advocate, but it seemed worthwhile to elaborate on the subject.

Demonstration casks at Springbank distillery

There are a number of different ways in which 'aging' occurs when an alcoholic liquid is placed in a cask.

The first, and most obvious, is that compounds are extracted from the wood by the alcohol. This is influenced by a number of factors:

•The range of compounds that can be extracted from the wood - which is influenced by the type of wood, where it was grown, how long it was seasoned (left out to dry before being shaped into a cask), how the cask was toasted or charred (more heat breaks bigger compounds into smaller ones, producing new compounds), and how many times and for how long the cask has held other liquids before (hence the new wood/first fill/refill/etc terminology one finds in whisky info).

•The ABV, as some compounds will be more soluble in ethanol while others will be more soluble in water, so the strength will shift the sets of compounds that are extracted, everything else being equal.

•The size of the barrel, as the interaction between spirit and cask is limited by the surface area. So generally a smaller barrel will increase the rate of extraction as there is a higher surface area:volume ratio (hence why some distillers use smaller casks to 'speed up' maturation), while the opposite is true for bigger casks.

•Temperature fluctuations will cause the spirit to expand and contract, pushing and pulling it out of the wood. So a climate with broad temperature extremes will increase the rate of extraction, while a climate with narrow temperature extremes will slow down the rate of extraction. This is why some rum distillers in the Caribbean will actually heat their warehouses, to prevent the barrels from cooling down at night and increasing the rate of extraction. On the flip side, this is one reason why cool, maritime Scotland tends to have lower rates of wood extraction.

This can generally be thought of as 'additive' aging, whereby new compounds are added to the array present in the liquid when it is first placed in the cask.

Second, compounds are extracted out of the spirit by the wood, more so in casks that are heavily charred, by the layer of charcoal on the inside of the barrel. Some compounds will be absorbed into that layer of charcoal in the same fashion as household water purifiers. This is one form of 'subtractive' aging, whereby compounds that are present in the liquid when it is added to the cask are removed.

Third, compounds evaporate from the cask as it interacts with air. This is one reason why Diageo's notorious 'cling-film' experiment never went very far - there needs to be a certain amount of interaction with air to allow high boiling compounds that made it past the foreshots cut to evaporate. Otherwise the whisky would be left with more 'immature' and off-putting odors and flavors. Additionally, water and alcohol also evaporate, depending on environmental conditions, changing the volume and ABV of the liquid inside, which will influence its extractive potential as noted in point 1.

Fourth, compounds react, both with the wood, with other compounds within the liquid, and with the oxygen in the air.

•One of the main ways in which compounds within the liquid react with each other is via the formation of esters. Put simply, an ester is a combination of an acid and an alcohol that gives off water in the condensation process. Some esters are created during fermentation or the distillation process, but they can also be created from free volatile acids and alcohols that react as the spirit matures. Additionally, recombination will happen, especially as ethanol displaces other alcohols in esters via mass action. Additionally, as ethanol is oxidized to acetic acid, acetate esters will also become more common. Putting the two together, ethyl acetate tends to be the dominant ester in all spirits. This will be influenced both by the concentration of alcohol (primarily ethanol), temperature, and the rate of oxidation.

•Oxidation will transform molecules within the spirit. Alcohols will become aldehydes and ketones, then aldehydes will become acids. Unsaturated compounds will be cleaved into aldehydes and ketones. This is influenced by the oxygen tension in the cask, the rate of gas exchange, and the ambient temperature (a general rule of thumb for chemical reactions is that they will go 2x faster for every 10º C that the temperature is raised). But as these reactions are generally uncatalyzed and molecular oxygen is not a particularly effective oxidizer of organic molecules on its own, they will be rather slow.

•Alcohol will help to break down the macromolecules that make up the wood, increasing the range of substances that can be extracted into the spirit. This is influenced largely by the concentration of alcohol in the spirit.

Sherry butts at Bruichladdich

Now back to the question posed at the beginning. What it comes down to is that sherry bodegas and whisky distillers want to focus on different axes of the aging process.

Sherry begins as a white wine, produced largely from palomino grapes, which is then fermented to dryness at ~15% ABV. Most sherries are then fortified to between 15.5% and 20% ABV with neutral grape spirit.

Aging sherry focuses primarily on the recombination of compounds already within the liquid and, for some varieties, on oxidation. This means that the casks are basically inactive, acting as contains rather than as direct participants in the process. If you've tried sherry before, you will have noticed that it doesn't have the tannic notes of, say, a California cabernet. This is because the casks used by sherry bodegas are first seasoned with lower quality wines that are later used for making sherry vinegar and the like, to extract the bitter tannins before they are used for higher quality sherry. That is not to say that the casks play no role - storing sherry in truly inactive containers of glass or stainless steel would not produce the same product.

The casks are host to microbial flora that interact with the wine and are critical in the formation of flor - a waxy layer of yeast that forms on top of sherry when the concentration of alcohol is around 15-16% ABV. It acts to exclude gas exchange, protecting the sherry from oxidation. This is critical for fino and manzanilla sherry to retain their freshness, even after prolonged time in the cask. On the flip side, amontillado and oloroso sherries rely on the gas exchange afforded by casks, aging oxidatively. These sherries are fortified to 17-20% ABV, which kills the flor, allowing oxidation to occur. This develops color and new flavors in the sherry that are not found in fino and manzanilla sherries.

In addition to primary cask aging, sherries are blended in a process called a solera. This is formed from layers of casks - wine for bottling is withdrawn from the bottom level, which contains the oldest wine. The casks on the bottom are then refilled from the level above, continuing upwards until the top layer is filled with new wine. This is a process of fractional blending, where some of the old wine always remains in the solera, adding complexity to the finished product. Solera casks also tend to be extremely inactive - if they contained significant amounts of extractable compounds, the final product would become intolerably bitter as more tannins were leached into the wine.

Duty paid sample sherry casks at Lagavulin

Now let's contrast that with whisky. In many ways, aging whisky is a more complex process as it is operating on every one of the aforementioned axes - extraction, subtraction, and reaction. While subtraction and some reaction can occur in inactive casks, extraction necessitates very different casks than those used by sherry bodegas.

In fact, as noted by Whisky Advocate in their article about sherry, there was a period in the 1980s and 1990s when distillers were buying solera casks from the bodegas. The wood was, as noted above, rather inactive, so these casks would have added a layer of sherry flavor on top of the whisky, but would not have contained the other extractable compounds that distillers seek. In addition, they likely would have been leaky, necessitating extra work by the coopers.

So what constitutes a good sherry cask for a distiller? For much of the 19th and 20th centuries, these were casks left over from transporting sherry from Spain to England, where it was bottled by British firms for British consumption. These could range from local grocers buying a cask or two, to large firms like Harveys, who would bottle cases upon cases. After the sherry was dumped, there were a lot of casks left over. It didn't take long for Scottish distillers to realize that not only were these cheap containers for storing their spirit, but they also made it tastes a whole lot better. The critical element for distillers, beyond price and availability, was that the transport casks often would have been new wood, rather than the inactive casks preferred by the bodegas. So the fresh wood would be impregnated with sherry for a relatively short amount of time before being turned over to the distillers. Transport casks eventually ceased to be an option, due to sherry producers beginning to bottle their own products in the late 19th and early 20th century and the eventual ban of transporting bulk wine over 15.5% ABV within Europe in 1981 (I've seen it stated as 1986 elsewhere).

In the early 20th century, DCL figured out that they could 'improve' the process by adding a thick, syrupy form of sherry called paxarete to a cask, then subjecting it to high pressures and temperatures, to artificially 'inject' sherry into the wood. Especially after Prohibition was lifted in America and ex-bourbon barrels became extremely cheap due to regulatory requirements, this became a way to create a new 'sherry cask'. This was helpful, both because it was even cheaper than transport casks (by that time it would have been clear that they still had value) and consistency, both in terms of the output and in ensuring a steady supply, due to the decreasing availability of transport casks. The practice of using pax was fairly common from roughly the 1920s until the 1980s, when the Scotch Whisky Association banned it. Again, this would often be carried out on relatively new wood, either a freshly made cask or an ex-bourbon barrel that still contained a lot of extractable compounds.

Used sherry casks at Springbank

Currently, most sherry casks used by distillers are custom made. These are either built to order by contractors such as Toneleria del Sur or, like some distillers are now doing with their bourbon casks, specifically coopered by the distillers, then 'loaned' to the bodegas for aging sherry destined for vinegar or distillation. The switch to custom casks has also involved a switch to European oak (it contains flavor compounds that distillers want), whereas the bodegas tend to favor American oak as it is both cheaper and easier to work with (the bodegas want neutral casks, so the differences in flavoring compounds in the wood is largely moot). A point to note is that these custom casks are, in many respects, very similar to the transport casks that distillers were so fond of a century ago. In both cases, the wood will be new when sherry is added - you can see how new the wood is when the casks arrive at the distillery in this picture from Springbank. The sherry that comes out of the custom casks, often having spent years seasoning the wood, will likely not be fit for drinking (this is part of why custom casks tend to be so expensive), as it will have pulled a significant amount of tannins and other bitter compounds out of the wood. However, that is good for the distiller, as those compounds will not be present when the cask is filled with whisky.

One of the most important aspects of making good whisky is achieving the proper balance of extractable compounds in the wood. New wood (sometimes known as virgin oak) is, with rare exceptions, considered to be too active for scotch, being used only as a finish for whisky that spent most of its maturation in casks that had previously held some other liquid. Distillers usually seek a balance of extractable compounds - enough to impart flavors of vanilla and coconut (these tend to be the dominant elements of American oak) or spices (these tend to be dominant in European oak) to the spirit without completely overwhelming it. The sweet spot is an ex-bourbon or ex-wine cask that is being filled with whisky for the first time (a slightly misnomered 'first-fill' cask) or second time (also slightly confusing 'refill' or 'second-fill' cask). First-fill casks are perfect for whiskies that will be aged a relatively short time, say 8-15 years. The more active first-fill casks will impart their flavors more quickly, adding a significant amount of richness to the spirit, but there is also the risk of going too far and making the spirit overly oaky. Refill or second-fill casks are more suited for longer periods of time, where the wood will impart flavors, but then hit a point where the wood has given up all it can, allowing the other axes of maturation to proceed without overwhelming the spirit with extracted flavors. There are always exceptions to these rules of thumb - first-fill casks will not always stamp a heavy mark on the spirit or may be well-suited to a particular distillery with intrinscially weighty or flavorful spirit (Mortlach or Ledaig, for instance), while refill casks will on occasion provide more richness than a first-fill cask. Additionally, distillers will often continue to use a cask for 3-5 fills, especially for lower quality or grain whiskies that are destined for less refined blends, though these also sometimes end up in the warehouses of independent bottlers. But increasing attention is being paid to getting the right amount of extraction out of wood, providing the right amount of flavor from the wood to balance the character of their spirit. For instance, Laphroaig will discard their casks after a single fill because the heavy character of their spirit requires active casks to balance it out (though there are also cases where they may have overshot the mark).

The take-away from all of this is that aging alcohol in oak casks requires an understanding of what the wood will or will not impart to the liquid, given its state and the amount of time it will be spending in the cask. Aging is a complex process and focusing on one element or another will require different sorts of wood. A good distiller or venenciar will know how to use the casks to achieve the final product they desire.

The Physics of Pot Stills

Pot stills are one of the oldest methods on earth for concentrating the alcohol in fermented beverages (freeze distillation may have come first, but it's hard to know). In essence, a pot still is any system for boiling an alcoholic liquid, then directly condensing the vapors to generate a liquid with a higher alcohol concentration. While they all follow this basic model, there have been historical developments that improved on the basic setup over time.

The simplest stills were literally pots, often earthenware, with a surface above it that would allow the vapors to condense and then run down into another vessel or a wide lip around the edge of the pot. As an open system, they were extremely inefficient as vapors could easily escape without condensing. Additionally, there was no easy way to remove heat from the condensing apparatus, which significantly reduced their efficiency.

The first major development in design was the retort, which is a container, usually made of glass, with an arm extending off to the side, where the vapor would collect, (hopefully) condense, and then be drain in another vessel. Assuming good rates of condensation, this was a major step forward, but there was still no good way to remove heat from the arm of the vessel.

An adaptation of the retort was to extend the side arm into a coil, which would then be immersed in a container of water, called a worm tub. The high heat capacity of water gives it a great ability to absorb heat from the condensing vapor, which made these stills much more efficient. Some industrial stills use roughly this same design, though the worm tub has water piped through it to maintain a constant cooling bath. While possible to construct a worm tub condenser from glass, those tend to be very delicate. So they were generally made from copper, which was ductile, making it easy to work with, and stronger.

The final development of the pot still came from the addition of the column or shell condenser. This is a large tube with numerous copper pipes running through it that carry cooling water. This is the most efficient cooling system for pot stills and is now the primary condensation method.

The simplest form of distillation when spirits are produced is as a wash still or beer still, which operates largely to concentrate the alcoholic portion of the fermented liquid being distilled. A lot of chemistry goes on as well, but here we're going to focus on the physical process of distillation. The wash still will convert wash at 5-8% ABV into 'low wines' at an average strength of ~25% ABV, which thus concentrates the liquid 3-4X, which a concomitant decrease in volume. Additionally, any compounds with boiling points significantly above that of water will be left behind in the pot, along with any residual solids that made it through the filtration of the mash or that were produced during distillation (especially when heat is supplied by a naked flame rather than by steam coils).

© 2001 The International Centre for Brewing and Distilling

You can see from the phase diagram above that when 8% ABV wash is heated to its boiling point of 94.5º C, it vaporizes, then recondenses to a liquid of 45% ABV. Distillation proceeds as the temperature rises, the ABV of the wash decreases, and the concentration of alcohol in the condenser approaches zero. In the other direction you should note the point where the vapor and liquid lines meet, which is the point when ethanol and water form an azeotrope - a mixture that has a single boiling point. This is why ethanol cannot be distilled above 95.5% ABV.

So what's really going on? At a basic level, the purpose of any distillation is to separate different compounds by their boiling points. The components with lower boiling points will vaporize first, then so on down to the highest boiling components. The separation between different compounds is described technically as resolution. In double (or triple) distillation setups, this is carried out in a second pot still called a low wines or spirit still.

Let's consider two cases. In each diagram, the X-axis represents temperature (or time, it doesn't matter too much if you assume a flow of heat into the system), while the Y-axis represents the concentration of each compound as measured at the condenser. First, you have a mixture of two liquids with almost exactly the same boiling point. When you heat the mixture, both will vaporize at roughly the same time and the concentration of each compound in the condenser over time will rise and fall in tandem.

Next a mixture of two liquids with moderately different boiling points. The liquid with the lower boiling point (in red) will begin to vaporize first, with the output of the condenser consisting entirely of the lower boiling liquid at first. As time goes on and more heat is added to the system, the liquid with the higher boiling point (in blue) will begin to vaporize, with the output of the condenser consisting of a mixture of the two. As more heat is added, the compound with the lower boiling point will have been almost entirely removed from the pot, so the output of the condenser will be almost all the higher boiling liquid.

Last, a mixture of two liquids with very different boiling points. This case will be very similar to the last one, but the period where both liquids are coming out of the condenser will be much smaller. This is the resolution that I talked about earlier - the greater difference in boiling points increases the separation between the two peaks, making it easier to 'resolve' one from the other.

Now obviously even the simplest distilled spirits are a much more complicated case, as there are hundreds if not thousands of different compounds that are going through the same vaporization-condensation process alongside the main constituents - water and alcohol. But the principles remain the same, you just have to imagine almost countless overlapping distributions.

There are a number of different ways to influence the resolution in a pot still.

One is the rate of heating the contents of the pot. This is described by distillers as running the stills 'fast' or 'slow'. Adding heat quickly will cause more compounds to vaporize at the same time. In contrast, slow heating will have the opposite effect, as the temperature rises slowly and low boiling compounds can vaporize and enter the condenser before the pot becomes hot enough to vaporize the high boiling components.

The second major component of resolution is the amount of reflux in a still. Put simply, reflux is the tendency of vapors to recondense before reaching the condenser, which makes them fall back into the pot, where they can be reheated and vaporize again. This usually happens because the vapors contact the material of the still and transfer some of their heat to it, which removes that heat from the vapor and reduces its temperature. This forces the higher boiling components to go through the kind of separation described above again and again, significantly increasing the resolution. On a more macro level, this also affects the strength of the spirit that can be obtained in a single distillation, as a greater degree of reflux allows for greater separation between alcohol and water, producing a spirit of a higher strength.

Adapted from Ethanolator

The above diagram shows the equilibrium curve between liquid and vapor mixtures of ethanol and water. This is a different way to visualize the temperature:concentration phase diagram shown above. This chart assumes that the temperature is at the boiling point for a given mixture, then displays the relationship between the concentration of ethanol in the liquid and vapor phases. The hashed line is where the liquid and vapor phases have equal concentrations. Drawing a vertical line from the starting concentration in the liquid up to the equilibrium line gives the maximum concentration that will be achieved in the vapor phase. A horizontal line would be drawn from that point going to the equal liquid/vapor concentration line in an ideal case where there is 100% reflux and everything coming out of the condenser is returned to the pot, which results in the liquid and vapor phases establishing a stable equilibrium. But in the case of an operating still, where mass is removed from the condenser and the system is not in perfect equilibrium, as represented by the lower solid curve (not actual data, but suggestive of a normal pot still), the liquid in the condenser will have a slightly lower maximum proof, so the horizontal line is drawn to the curve rather than the dashed equality line. This procedure is repeated as the mixture is repeatedly distilled, up to the point where ethanol and water become an azeotrope. This kind of analysis can thus tell you how many distillations will be required to reach a particular maximum strength. This is why double distillation usually reaches a maximum of ~70% ABV, but triple distillation can reach ~80% ABV.

The amount of reflux is primarily dependent on the shape of the still. The simplest factor is still height - the shorter the vertical distance between the pot and the lyne arm that leads to the condenser, the less reflux there will be. Conversely, the taller the still, the more reflux. Next, the smaller the neck of the still, the more reflux, as this increases the surface area:volume ratio of the neck of the still, increasing the likelihood that vapor will come in contact with the material of the still. Conversely, a wider neck will decrease the amount of reflux, as there are more paths for the vapor to travel that do not bring it in contact with the material of the still, increasing the likelihood that it will reach the condenser as vapor. The angle of the lyne arm influences reflux as well - a lyne arm that is roughly at a 90º angle from the neck is considered to be neutral as vapors that condense inside the neck can run either back into the pot or forward into the condenser. A lyne arm that has an angle greater than 90º, rising from the neck, will increase the amount of reflux as it effectively raises the height of the still and vapors that condense inside the neck will run back down into the pot. A lyne arm that has an angle less than 90º, descending from the neck, will decrease the amount of reflux as vapors that condense inside it will tend to run into the condenser, instead of back into the pot. Last, but not least, the amount of reflux can be increased by creating constrictions and bulges in the neck of the still. This exploits the properties of gases as they are compressed and expand. As the vapors emerge from the constriction, they will expand experience a corresponding decrease in temperature, which will increase the amount of reflux. So a still with a ball or lamp glass shape will have more reflux than a still with a plain neck. Another trick to increase reflux is to run a pipe from the lyne arm back into the pot, which will return any material that condenses in the lyne arm for redistillation.

Left - plain still with ascending lyne arm (Bowmore); Center - ball still with descending lyne arm (Kilchoman); Right - lamp glass still with flat lyne arm (Jura)

We can see  two extremes. First, Glenmorangie's spirit stills, which are very tall and narrow, with a ball shape, and rising lyne arms, which are designed to produce an extremely refined spirit with low congener content. Second, Lagavulin's stills, which are short and squat, with a plain shape and steeply descending lyne arms, which are designed to produce very heavy spirit with an extremely high congener content.

It is in the second distillation that the concept of 'cuts' comes in. While different compounds are hitting the condenser at different times, they will all be collected unless the output is diverted into different receptacles. Most of those early peaks are things that you do not want to be drinking, like methanol, acetaldehyde, and other toxic compounds. So the person running the still will send some of the early output, known as foreshots, to the 'feints receiver'. Visually this can be thought of as making a vertical slice through one of the peaks - everything before goes to one place, everything after elsewhere. What comes next is the 'heart' of the run, usually running between 70% and 60% ABV, which contains the bulk of the alcohol and congeners that make the spirit taste good. As alcohol concentrations drop, water concentrations rise, and other higher boiling congeners that are also undesirable begin to hit the condenser. A second 'cut' is made at this point, sending the 'tails' or feints into the feints receiver as well.

© 2001 The International Centre for Brewing and Distilling

That should cover most of the basics. If you have any more questions, feel free to ask them in the comments and I'll do my best to answer them.

Why is Alcohol Toxic? - Part I: Cellular Damage

Every drinker has at some point had more than it was wise to have and suffered for it. Most of the time, that leads to a hangover that is unpleasant, but passes. But either consistent heavy drinking or indulging to the point of alcohol poisoning can have far more serious consequences.

While there appear to be benefits to moderate drinking, ethanol is at its root a toxin. This is due to three factors. First, its basic chemical properties, second, the ways in which it is metabolized, and third, the effects that it has on the brain.

Let's look at its chemistry first. Ethanol is similar to water in that it has a hydrogen bound to an oxygen. This allows it to interact with many hydrophilic compounds (those that dissolve well in water), because the hydrogen-oxygen bond is polarized in a similar fashion. However, instead of a second hydrogen, ethanol has a short hydrocarbon chain. This makes it more like a volatile solvent, e.g. gasoline, and allows it to interact with hydrophobic compounds as well. Having both of these features means that it can act a bit like a detergent. Detergents, like soap, also have both polar and non-polar features, which allow them to solubilize greasy molecules in water. While ethanol is not a strong detergent, at high concentrations it does have the ability to disrupt the structures of macromolecules like proteins and lipids in the cell - this is part of why ethanol is a good disinfectant.

Thankfully, for most of us, our bodies are equipped to handle limited amounts of alcohol. Even if you have never touched a drop, there will be trace amounts of ethanol in your blood from the metabolism of various compounds we consume, like ethyl esters. To keep it from accumulating to levels where it would significantly disrupt normal biological processes, most humans possess an enzyme called alcohol dehydrogenase (ADH). This enzyme catalyzes the first step in the conversion of ethanol to acetic acid, which is then fed into the citric acid cycle, where it produces energy for the cell. All well and good, on the surface.

There are a number of wrinkles in this process. First, the method by which energy is extracted from the oxidation of ethanol, is primarily by the transfer of electrons to nicotinamide adenine dinucleotide (NADH), the main carrier of reducing equivalents in the cell. This changes the redox balance in the cell, which has both direct and indirect effects. Indirectly, this can decrease the oxidation of fatty acids, as they go through a very similar oxidation process, so high levels of NADH will suppress utilization of fat. Additionally, the acetic acid formed as an end product of ethanol oxidation can be converted into acetyl-CoA, which is the starting material for fatty acid synthesis. The dual push of decreased fat burning and increased fat synthesis are two of the main drivers of alcoholic steatosis, also known as fatty liver. Taken to an extreme, this can lead to cirrhosis and liver failure.

From Wikipedia

Second, during the conversion process of ethanol to acetic acid, there is an intermediate product called acetaldehyde. This is the source of much of ethanol's toxicity and carcinogenicity. Aldehydes are very reactive, especially with amines such as those found in the amino acids that make up proteins or the nucleic acids that form DNA. The reaction between acetaldehyde and amines, especially in the reducing environment (which, remember, is enhanced by the metabolism of alcohol) found inside cells, will form adducts, disrupting the structure and functions of proteins and nucleic acids. In the case of nucleic acids, these adducts can lead to mutations that may increase the risk of cancer. Alcohol consumption is a well-known risk factor for cancer, especially esophageal cancer, which is directly related to the exposure to acetaldehyde. In small quantities, cells can remove damaged proteins and repair DNA adducts, but if errors build up they can lead to mutations or cell death.

This is particularly dangerous for people who have variations in either ADH or acetaldehyde dehydrogenase (ADLH), which catalyzes the second half of alcohol metabolism. These variations are more broadly known as 'alcohol flush syndrome', which is prevalent in people of Asian ancestry. This is caused either by a variation in ADH that results in extremely rapid conversion of ethanol to acetaldehyde, potentially coupled with a less active version of ADLH. In either case, acetaldehyde builds up to dangerous levels very quickly and the conversion to acetic acid happens slowly or not at all. The high acetaldehyde levels, in the short term, produce the characteristic flush, but in the long term produce significantly higher risks for cancer, especially of the esophagus.

Similarly, the reason why other alcohols like methanol, rubbing alcohol (isopropanol), and anti-freeze (ethylene glycol) are even more toxic is that they are oxidized to the aldehyde or ketone stage, but cannot be oxidized further to acids. This is also why the treatment for methanol or anti-freeze poisoning is to drink ethanol - ethanol binds more effectively to ADH, which fully engages the enzyme in ethanol oxidation and gives the body time to flush out the other alcohols.

Even worse compounds are produced after alcohol tolerance has built up. This is because ADH and ADLH levels remain the same, even after high chronic levels of alcohol consumption, which necessitates the liver inducing new pathways (essentially calling in for backup) for detoxification. These pathways are primarily enzymes called cytochrome P450s (CYPs). There are a whole host of these enzymes, which perform many different types of oxidations on different kinds of compounds. The common thread is that they are all designed to metabolize foreign substances into compounds that can more easily be eliminated from the body.

The primary CYP induced by alcohol consumption is called CYP2E1. While the mechanism of alcohol oxidation used by ADH is comparatively benign, that of CYP2E1 is not, as it is both less specific and more powerful. CYP2E1 directly oxidizes alcohol via the activation of molecular oxygen. A byproduct of CYP2E1 oxidation is superoxide, which is a powerful oxidizer in its own right. That can be partially ameliorated by the enzyme superoxide dismutase, but the product of that reaction is hydrogen peroxide. Anyone who has poured a bit of hydrogen peroxide on their skin to disinfect a wound knows its effects and it does not take much imagination to figure out that it's not a good thing for your cells to be producing, especially in large quantities. All of these oxidizing byproducts can go on to damage cellular components such as proteins (including CYP2E1 itself) and lipids. Lipid peroxidation has been implicated in cell wall breakdown and DNA adduct formation, which increases the risk of cancer-causing mutations.

CYP2E1 oxidation cycle via NIAAA

Adding to the danger is the fact that CYP2E1 is also implicated in the oxidation of many other drugs. The most important of these is acetaminophen, the active ingredient in Tylenol. Normally the liver deals with acetaminophen by conjugating it with an acidic sugar, glucaronic acid, which targets it for excretion in the urine. However, when concentrations are high, the glucaronidating enzymes will become saturated, which increases the concentration of free drug. CYP2E1 can oxidize acetaminophen to a compound called n-acetyl para-benzoquinone imine (NAPQI). Much like acetaldehyde, NAPQI can react with many other cellular components, especially proteins and nucleic acids, leading to protein inactivation and DNA mutations. Normally NAPQI is mopped up by conjugating with a cellular compound called glutathione, producing a harmless byproduct. However, glutathione also reacts with acetaldehyde. So high alcohol consumption will reduce the levels of free glutathione, leaving less available for detoxifying other drugs. This is why giving acetaminophen to alcoholics, especially those in the very early stages of recovery, can potentially be fatal. CYP2E1 is induced by chronic alcohol use, thus increasing the production of NAPQI. Glutathione levels will be low, which decreases the ability to detoxify NAPQI. Put together, these significantly increase the toxicity of acetaminophen. While acetaminophen is the most well-studying case of alcohol-induced CYP2E1 toxicity, a large number of other compounds that are either broken down or activated to toxic byproducts are also metabolized by CYP2E1 and will have their metabolism changed by either acute or chronic alcohol use.

This covers most of the ways in which alcohol is directly toxic to your body, primarily your liver. In the next half, I'll cover alcohol's effects on the brain and another way in which chronic alcohol consumption can be particularly dangerous.

When Whisky Was (Possibly) (Slightly) (More) Carcinogenic

Let's be honest with ourselves. When we drink alcohol, especially distilled spirits, we are drinking poison.

Now, as Paracelsus noted, "The dose makes the poison" and in moderate amounts, alcohol may have beneficial effects that outweigh its downsides. But there was a brief period in the late 1970s and early 1980s when there was worry that beverages made from malt, including beer and whisky, contained dangers above and beyond their standard risks.

The fear was caused by a compound called N-nitrosodimethylamine (NDMA). Nitrosamine compounds form when nitrogen oxides react with amines. For instance, nitrosamine levels used to be fairly high in meats preserved with sodium nitrite, such as bacon, and are still rather high in tobacco products.

Beginning in the late 1950s and into the 1960s, evidence began to accumulate that nitrosamines could lead to cancer. Studies in rats showed that administration of NDMA led to liver cancer and there was an incident in Norway where pigs fed herring (fish tends to have high levels of amines - hence its smell) preserved with sodium nitrite developed liver diseases, including cancer. At this point, while it was known that nitrosamines were dangerous, analytical techniques were unable to detect it in human foodstuffs.

That made it very alarming when studies released in 1979 found that beer and malt whisky contained detectable levels of NDMA. While concentrations were low, as little as 0.4 to 0.7 parts per billion (PPB), this was still unsettling as some studies on rats had concluded that even 10 PPB were enough to triple lung cancer rates. It is known that nitrosamines can react with DNA to form adducts, which is a plausible mechanism for much of their carcinogenicity.

How did this happen? While nitrosamines were likely always present in malt to one degree or another, increasing levels came about from the advancing technology used in the process of drying malt. Heat is used to arrest germination and dry the malt to preserve it in a stable form. Though peat and coal had been historically used all over Scotland, they were being phased out in favor of gas burners, which are more flavor-neutral sources of heat. These appeared to burn cleanly and dry the malt without imparting any flavor, making it easier to produce the unpeated malts distillers needed for the making lighter, more cleanly flavored whisky.

However, the temperatures produced by gas flames were significantly higher than those of peat or even coal and oil, which increased the formation of nitrogen oxides (primarily dinitrogen trioxide and dinitrogen tetroxide) from the nitrogen present in air. Those nitrogen oxides would then react with nitrogen-containing compounds in the malt to produce nitrosamines. As noted, the most common nitrosamine is NDMA. This is formed primarily from hordenine, a dimethyl derivative of tyramine (itself a derivative of the amino acid tyrosine), which is at its peak concentration early in the kilning process. The nitrogen oxides in the hot air act both to cleave dimethylamine from hordenine and convert it into NDMA.

Formation of NDMA from hordenine

As many of these nitrosamines have boiling points comparable to phenolic compounds (~150º C vs ~180º C) that also find their way into malt whisky, they can be carried through the production process and ended up in the final product.

Thankfully solutions to this problem were found fairly rapidly. The simplest was to heat the malt indirectly rather than directly. Heat from a gas burner is fed into an exchanger, which transfers that heat to clean air, which is passed through the malt. This is the process now used in almost all maltings, especially larger ones.

From Shimadzu News 3/2005

However, some maltings still use direct heat in the form of burning peat. While peat fires are generally used to generate smoke rather than heat, per se, they can still produce nitrogen oxides. This makes it important to avoid flaring while burning peat, as that will increase the production of nitrogen oxides. To prevent this from happening, sulfur is burned alongside the peat, which forms sulfur dioxide, which reduces the pH of the malt and inhibits the formation of nitrosamines. This process also occurs naturally in kilns heated by coal and oil burners, as these fuels contain fairly high levels of sulfur. In comparison, natural gas contains very low levels of sulfur, which likely contributed to the formation of nitrogen oxides when it was used to dry malt.

The kiln at Springbank distillery

Alongside the changes to production techniques, analytical techniques have also improved over the decades since this issue was first brought to light, which makes the routine analysis of malt for nitrosamine content relatively simple. This ensures that quantity of nitrosamines in malt used for brewing and distilling are far below the level that would do you any harm. So you only have to worry about what the alcohol itself is doing to you.

Why Long Fermentation Times are Important for Ester Formation in Malt Whisky

One fact I noticed during my trip to Scotland was that the average weekday fermentation time at scotch whisky distilleries was about 55 hours, but some went as short as 48 hours while others went as long as 75 (weekend fermentations sometimes reaching 120 hours).

It was generally the case that the larger distilleries used shorter fermentation times while the smaller ones had longer fermentation times (Caol Ila used to be an outlier, but appears to have shortened their fermentation times since replacing their old wooden washbacks with stainless steel a few years ago). In addition to Caol Ila, other distilleries such as Ardbeg have also decreased their fermentation times over the last decade, likely in an effort to increase the output of the distillery. But will shorter fermentation times produce the same kind of spirit?

After doing a bit of reading on the subject, I'm willing to say that the answer is probably 'no'. While shorter fermentation times can extract the same amount of alcohol out of a mash as a longer fermentations, there are other processes that need more time.

Lets begin with what happens during the production of malt whisky.

Malted barley is ground in a mill to produce grist, a mixture of flakes, finer flour, and hulls. This is then added into the mash tun, where it is mixed with hot water to extract the simple sugars from the grain. after soaking for some time, the liquid is drained off and progressively hotter water is added each time, usually three or four times total. The water is rather hot, with the first water added at ~65º C, the second at ~75º C, and subsequent waters are between 85-95º C, which extract the last bits of sugar from the grist and are generally recycled to be used for subsequent first and second waters.

Semi-lauder mash tun at Auchentoshan Distillery

The key point in that process is that while temperatures are high, they are not, unlike mashing done at breweries, heated above 100º C. This means that while the microbial cultures living in the malt are significantly thinned by the heat, they are not all killed.

The sugary liquid from the first two waters is cooled to 18ºC and piped over to the washbacks, where cultured yeast is added. As the dissolved oxygen in the wort is quickly consumed, the yeast begin to grow and divide anaerobically, converting the sugars in the liquid into alcohol, carbon dioxide, and other compounds (for more details, see Whisky Science). Because the yeast is pitched at fairly high concentrations, it can out-compete the remaining residual bacteria for the first 30-40 hours of the fermentation. At that point, the yeast begin to run out of steam as they start to choke on their own waste products - alcohol and heat.

Highly active fermentation at Laphroaig Distillery

While the starting temperature of 18-22º C is a bit below the optimal temperature for yeast, it is necessary to start that low because no whisky distillery I have seen has active cooling systems in its washbacks. This means that the liquid will absorb all of the heat produced by the yeast as they multiply and divide, which, over time, ends up being a lot of heat. After 48 hours, the temperature of the wort can rise by 10-20º C. While wine yeasts are sometimes tolerant up to 32º C, the S. cervisiaie strains used by distilleries will begin to suffer above 25º C or so.

Additionally, the end product of fermentation, ethanol, is toxic to the yeast that produce it. Final alcohol concentrations range from 5-8%, which is approaching the upper limit of survivability for S. cervisiaie. While the yeast will attempt to sequester the alcohol by converting it into esters, this is not a long-term strategy.

Both heat and alcohol end up creating the conditions for autolysis. While you may have heard of this process as something that brewers attempt to prevent, it may actually be an important step in developing the flavors of malt whisky (and champagne). As the yeast become stressed, they begin, in essence, to digest themselves. Cells are exquisitely organized to keep different functions in distinct compartments. When those compartments begin to lose coherence, degradative enzymes are loosed upon the rest of the cell, leading to almost complete breakdown. Large polysaccharides, including the major constituents of the yeast's cell wall, are broken down into smaller mono- and oligosaccharides; proteins are broken down into peptides and free amino acids; triglycerides are broken down into free fatty acids and glycerol.

All of those compounds released during yeast autolysis provide fodder for the bacteria that have been lurking in the background during the initial phases of fermentation. A study by van Beek & Priest (2002) found that bacterial communities, primarily lactic acid bacteria, only begin to thrive after 30-40 hours of fermentation and hit their maximum growth after 70 hours.

van Beek (2002)

While the bacteria are important in and of themselves, the intermediate period between 30 and 50 hours is critical because the yeast begin to defend themselves against the growing bacterial communities. Yeast and bacteria have coexisted for billions of years and yeast have developed a number of defensive mechanisms to suppress bacterial competition for resources. One of these defense mechanisms is the synthesis of acids.

van Beek (2002)

As you can see from the table above, acetic acid concentrations rise almost 10X between 40 and 50 hours. From the previous figure you can see that this is where the bacterial community enters an exponential growth phase and starts to present real competition to the yeast. In response, the yeast produce acetic acid to suppress that growth. As noted above, this is also partially a strategy to reduce the concentration of alcohol by converting it into ethyl acetate, though that never rises above mg/L concentrations. The rise in acetic acid has effectively ceased by 65 hours, at which point the yeast have almost all undergone autolysis and the bacteria are dominant.

The major constituent of the bacterial communities during malt whisky fermentation are strains of Lactobacilli. As the name suggests, these bacteria tend to produce lactic acid. This is the end produce of lactic acid fermentation, which breaks down sugars anaerobically. Why is this important to the flavor of whisky? Lactic acid can form esters, primarily ethyl lactate, which has a creamy or buttery flavor. Additionally, Wanikawa et. al. found that lactic acid bacteria hydroxylate unsaturated fatty acids from yeast, which can be esterified into lactones, which have fruity or coconut odors and flavors. Lactic acid bacteria also continue the process of ester synthesis started by the yeast, producing new acetate derivatives of fusel oils. Additionally, the action of lipases continuing to break down the triglycerides from the yeast to produce free fatty acids, which are then available for esterification and the production of fusel oils is continued from the free amino acids released by yeast autolysis via the Ehrlich pathway, which provide the two necessary raw materials for esterification.

To add to the importance of lactic acid bacteria, Simpson et. al. found that there are differences in the strains of bacteria present in the worts of different distilleries in Scotland. These populations are relatively stable, though they do change to some degree depending on time of year and the types of malt being brought into the distillery. Especially in distilleries with longer fermentation times, these bacterial communities may represent one part of their 'terroir'.

Microorganisms growing on the washbacks at Springbank Distillery

So while yeast may get center stage when it comes to the production of whisky, there are other microorganisms that also play important roles in developing the flavors we associate with the spirit but need more time than is usually given to exert their influence. This does not bode well for distilleries that have reduced their fermentation times over the last decade in an effort to increase output. They may find that this spirit is less complex and flavorful than it was before.