Rum Super Geekdom
Talking to Bryan Davis about spirits can sometimes be like an episode of NOVA or The Big Bang Theory. Not only is he incredibly passionate about his craft, he's also eager to share everything he knows about making spirits. To give you an example, look at this response to an email I wrote him earlier this week, asking about the production specs of his new Lost Spirits Navy Style Rum.
Everyone wants to talk about age, but in truth the barrel should only represent the final step that catalyzes a chain of chemical reactions and brings all the work together from each step of the spirit production process.
To understand how I approach spirits is as follows.
Hgh density of long chained esters. + Benzaldehyde (a chemical that does not come out of the oak in quantity for a very long time under normal cellaring conditions).
- High density of free acids (hot spirits, that need to age longer to convert the acids to esters)
- Low density of acids or esters (light and boring spirits)
- High density of short-chained esters (smooth but only half finished with heavy solvent / paint thinner notes).
The beauty of making spirits is that each step presents an opportunity to intensify the density along the way.
Basic distilled spirit chemistry 101:
- Esters (the aromas of fruits, flowers, and spices) are made from chemically bonding alcohols to acids.
- Short chained esters are made from chemically bonding one acid and one alcohol together. They smell solvent like in high concentrations and fruity in low concentrations. These are intermediaries in the spirit maturation process. They will eventually form long chained esters in the cask.
- Long chained esters (our goal) are made by bonding multiple acids of different types and alcohols together.
To make esters you need acids and alcohols. Since we are making distilled spirits alcohol is a given. So where do the acids come from?
There are two types of acids in distilled spirits: Carboxylic acids and Phenolic acids and they can come from many places in the production process.
- Yeast produce carboxylic acids during fermentation. Primarily acetic acid but depending on strain many other acids are possible.
- Bacteria produce carboxylic acids during fermentation, varying wildly depending on stain employed.
- Toasted, burned, caramelized Lignin (a common polymer in all plant life) yields phenolic acids when burned (charred/toasted oak is a major source of this, but there are other sources in the process)
- Toasted, burned, caramelized Hemicellulose (the other common polymer in plant life) yields carboxylic acids and plant sugars (charred/toasted oak is a major source of this, but there are other sources in the process).
Now that we have covered the basics, it is time to being dissecting the entire process and trying to optimize each step to produce, first, the highest density of acids possible, then use our oak to both add new acids as well as catalyze the esterification process and convert all those acids into delicious long chained esters.
Step One… The Raw Material
Raw materials present an opportunity to gain precursor acids. Burnt, caramelized, smoked, or toasted raw materials contain free acids which can do chemistry. All the raw materials contain lignin and hemicellulose. When the lignin or hemicellulose is burned it breaks down into free acids. When the molasses are caramelized they release phenolic acids from the lignin in the sugar cane. Toasting grains, roasting coffee, roasting nuts, all function on the same principal.
Unlike roasting foods, in distilled spirits making we take things a big step forward and take the chemical products of the roasting and then put them back together in a new order that suits our desires (barrel aging).
By choosing / making raw materials that have the right aromatic acids in them from the start we boost our overall end ester count in the process. For the rum this means finding molasses that had a maximum amount of free phenolic acids without other bad tasting things that might pass through the still. The key, for me, is not the terroir of the sugar cane, the key is the cooking process and the amount and characteristics of the phenolic compounds generated.
Step Two… Dunder
Dunder is a mysterious substance added to the fermentation in high ester rum production. Dunder is sometimes made from overripe fruits, rotten fruits, and sometimes a special soup of decomposing bats, and waste from the last distillation.
Dunder is made in pits or caldrons and is sometimes ripened for up to a year before use. Though it may sound like voodoo there is actually a good reason for this substance. When the fruit, molasses waste, or bats undergo bacterial fermentation the bacteria produce carboxylic acids as a byproduct. These acids are responsible for the "rotting smell" but remember we are going to chemically bond them to acids later to make esters. The final esters will smell and taste completely different from the acids they are made from.
A carefully made “dunder” can yield more carboxylic acid than many years in a barrel. In my case this means overripe bananas which are a component of the yeast starter.
Yeast make carboxylic acids. Yeast also make short-chained esters from free acids present in our raw materials and “dunder.”
Yeast make acids to inhibit bacteria growth. It is a defense mechanism for them. If you want them to make more acid than usual you need to first select a yeast strain that makes more acids than normal. You can also stimulate the process by stressing the yeast into a defense posture. This can be accomplished by introducing controlled amounts of bacteria (see dudnder). This is also done, by manipulating the biochemistry of the yeast. In my case depriving them of nitrogen to create weak cell walls (another way of stressing them).
Getting the yeast to then convert those acids, and the dunder acids into esters is a clever trick. The yeast make esters as a means of removing alcohol from the solution. It’s a defense mechanism to prevent them from dying of alcohol poisoning. Again under the right stress conditions we can force them to convert a significant portion of the free acids into esters. properly managed the yeast can produce as many short chained esters as the first few years in a cask.
Optimizing the fermentation is not about mimicking the cask aging process. It is about going into the cask with a big dose of the carboxylic acids and esters. The logic works like this, if you have a big dose of acids and esters going into the cask you don’t have to wait from them to leach out of the oak. Furthermore, you can control which esters your making far more carefully through fermentation and raw materials selection, than you can by charing oak and waiting to see what the wood gives you.
When you approach spirits making this way, the time on oak is mostly about transforming the short-chained esters from the fermentation into complex esters.
There are two approaches to distillation. The more common approach, called high rectification, is to ferment with a lot of bad stuff in the mix and then use the still to clean the booze up. Some people also choose to ferment very neutral booze and rely on the barrel to do all the work. This is the bubble plate pot / column strategy.
The other strategy is to manage the fermentation carefully so that you can distill without plates and capture as much of the fermentation character as possible. This is a low rectification pot still strategy, and it is the one I favor because I have tight control over my fermentation and raw material selection.
If we had the right yeast, and the perfect dunder, and a perfect fermentation, it would all be ruined by excessive rectification. Our goal is to distill as much of the good stuff into the resulting spirit, with the alcohol, through low rectification.
It is all about the char. Oak on its own is chemically stable. It is only through the charring / toasting process that the lignin and hemicellulose in the oak become unstable yielding carboxylic acids, wood sugars, and phenolic acids. The acids coming from the oak are not only adding building blocks for ester making, they are also catalysts triggering the ester formation in the barrel.
Furthermore as the cask grows very old the lignin begins to decompose into the spirit yielding the all important holy grail of benzoic acid and benzaldehyde. These compounds are responsible for the sweet, "wet wood" character of the very oldest spirits.
We use a controlled charring process incorporating heat, flame, and even special frequencies of light to break the compounds we want out fast.
After that it is about manipulating the environment to make the catalyst from the oak do its job. I won't disclose all my secrets but in truth the aging process should be seen as the last step in a long line of process decisions that create a given spirit.
While it is true that you need the aging to complete the reactions and make the long-chained esters. Honestly, in the industry, far too much attention is paid to this final step, and not enough attention is paid to all the decisions that lead up to the aging as a final columniation.
P.S. The oak is Oloroso sherry seasoned New American Oak.
I think that clears it up, Bryan! Thanks for the detailed response!