Weighing the Merits of Slow Dilution

I can’t remember where or when I first heard about it—must have been 10 or 12 years ago—but, at some point, someone introduced me to the idea of slowly diluting spirits before packaging.

Years later, here I am: an adherent to a production philosophy that, in my humble opinion, very few people properly understand—including its most vocal supporters.

[PAYWALL] Making spirits takes time. I’ve known many distillers who take months obsessing over the final character of a vodka, to say nothing of the time it takes to produce aged spirits such as whiskey, rum, and brandy. And what are distillers doing with all that time? They’re pulling on the proverbial levers and tweaking the metaphorical knobs to dial in flavors, aromas, and mouthfeel to meet their established standards.

Typically, the last step in producing a spirit is diluting it to package strength. Much of the industry does that in one quick, calculated addition, just before bottling. After all, you’ve waited months or years to produce that liquid you’re about to bottle, and you’ve got customers ready to start sipping. It’s just water, right? Boring, neutral-ish, probably deionized, flavorless, and devoid of aromas. Just add the amount you need, check to make sure the proof is legally correct for bottling, and run with it.

What could possibly go wrong? It depends on who you ask.

Wait, Slow Down: Why Slow Dilution?

In 2013, I was fortunate enough to take a workshop led by Hubert Germain-Robin, the Cognac-born, California-based brandy distiller.

Hailing from a long line of cognac distillers, Germain-Robin came to the United States in the 1980s to see whether he could find a place to ply his trade outside the more restrictive production rules of his home country; his story is a fascinating bit of American liquor history. By the time I took his workshop, he was well established as a distilling legend.

I picked up a wealth of information from that workshop, but one of the most interesting tidbits was the French practice of slow dilution. I’d been trained that dilution was the final act of spirits production and that, outside of standard compliance concerns, all we had to do was add a precise amount of water.

However, instead of waiting until the last minute to dilute to bottle strength, cognac producers traditionally add small amounts of water over extended periods of time—often months, sometimes years. It sounds agonizingly long, but there is reason behind the approach.

Hubert pointed out that diluting too fast risks saponification of the spirit. Saponification occurs when a base converts an ester back into its constituent parts: a carboxylic acid and an alcohol. If the word “saponification” makes the word “soap” pop into your skull, it’s no coincidence: Saponification is the chemical reaction behind soap production. In soapmaking, you have fats (lipids) that contain a number of ester linkages that can be attached by the addition of lye (sodium hydroxide, or NaOH), a heavy base.

Many components that come from the simple saponification reaction taste very bitter. Besides those of us who cherish the intensely bitter flavors of various amaros, most people don’t have a positive view on bitterness—certainly not in tipples such as whiskey, brandy, rum, or gin.

Slow dilution has its fair share of skeptics. Eight years ago, I was out with a group that included John Glaser, founder of Compass Box Whisky. Glaser is a dead-set legend, in my eyes, and when the topic of slow dilution came up, he asked my opinion. I told him that I hadn’t seen any research supporting it, but also that the idea made sense. He politely disagreed with me, and I learned that Compass Box, apparently, never bothered with slow dilution—time is money, after all. And, I must admit, I’ve never detected a bitter note in their bottlings.

Not long after that night, I found myself judging whiskey for a national spirits competition. Joining me at the judges’ table were a few industry heavy-hitters—including my friend Dan Farber, owner and master distiller at Osocalis in Soquel, California. Farber also had learned from Germain-Robin and was a vocal adherent of slow dilution.

As we judged, we actually encountered a few spirits that we all decided were slightly to moderately saponified—they had a soapy, slightly oily, bitter note on the end of the palate, which felt like a dead giveaway after a few different samples had passed our lips… Or, perhaps these were just poorly distilled spirits with too many fatty acids from the tails still lingering about.

Therein lie some of the complications.

What of the Science?

There is a lot of art and tradition in what distillers do, backed by centuries of learning what works. That’s not the same as science.

“Science hasn’t been how the distilling industry has made decisions in the past,” says Aaron Hyde, director of distilling at RahrBSG, “and I’m not sure if thorough sensory analysis has been either. ... So, here we are.”

That’s a powerful statement, and it’s one that applies to a number of distilling concepts—including the idea that dilution risks saponification.

For years, I’ve searched academic papers, looking for anything to back up the anecdotal evidence—and I’ve largely come up empty. Yet, as Hyde’s statement implies, the lack of publicly available research on a given distilling topic shouldn’t be all that surprising.

There is certainly a wealth of data on saponification reactions in other industries, and it may be possible to draw the occasional parallel. However, until we see hard data on the topic, it can feel more than a little anarchic to jump to whichever conclusion you prefer.

Views from the Field

Of course, the lack of hard scientific evidence doesn’t seem to bother the growing number of slow-dilution practitioners.

While researching this piece, I cast a net into my wide network of distilling pros, asking for their opinions on the subject. I expected a healthy mix of both hardcore believers and skeptics. Instead, the skeptics seemed to be on vacation that week, and what I got was a wealth of responses in avid support of keeping things low and slow.

“I can remember my first time proofing a tote full of gin,” says Tony Gugino, owner of Eighth District Distilling in Manchester, Connecticut. “I weighed my alcohol, weighed my water, combined the two and... louche city had arrived. From that day forward, I have always added my proofing water slowly and made sure the solution was in movement, and it eventually led to the method I am [using] today.”

Gugino also notes that there’s an exothermic reaction when proofing down spirits, especially at larger volumes—and that can create off-flavors “due to the tearing of those ethanol bonds,” as well as a louche spirit.

Martin Weber, owner and distiller of Edelbrand Pure in Marthasville, Missouri, says he’s “totally convinced that a slow dilution process, in addition [to] a slow distilling method, makes a very beneficial difference in the quality of our fruit brandies and the gin we just came out with.” He says they divide their spirits equally into three-gallon glass carboys for dilution.

Jack Scardino, distiller at Black Button in Rochester, New York, also describes his carboy-based process: “I typically figure out how much dilution water I need to add per carboy to the desired final strength of the spirit,” he says. “I then divide the dilution process into 10 steps. I start with 15 percent of that volume in the first step, and then go to 13, 12, 11, 10, 9, 8, 7, 6, 5, and finally to 4 percent. I let the spirit rest for at least two to three days between each step. The whole process takes about three to four weeks.”

Scardino says he adds the dilution water slowly, “using finely metered glass cylinders, without any splashing or creating any air bubbles. [It] usually takes about 15 to 20 minutes, depending on each step and volume used.”

For me, it’s fascinating to read the processes, reasoning, and experiences of other distillers. At first glance, they all read fairly similarly to each other. Look a little closer, however, and you’ll note differences. Notably, Gugino at Eighth District aims to keep the spirit and water matrix in constant movement, while Scardino at Black Button avoids liquid agitation as much as possible.

Is one method better than the other? I’d wager a simple “no” here—it comes down to what works best for you and your distillery to achieve the character you want.

Throughout my career and interactions with various distillers, I’ve heard and seen a myriad of other slow-dilution protocols, all dancing around a similar theme: They add the water over a period of several days, at least, if not weeks. And, as I mentioned previously, some people—often chasers of French élevage techniques—may dilute over years.

Dialing in Your Dilution

This all begs the question: Assuming that saponification is a real threat, how long should the dilution process take to avoid it? Facing a lack of scientific data, we have to put on our thinking caps and reason through this the old-fashioned way.

For the saponification reaction to occur, we need esters, a base, and a little heat. I’ve long suspected that the bitter, soapy taste we attribute to saponified spirits would come from long-chain fatty acids. Logically, then, the basic, low-numbered carbon esters such as ethyl acetate aren’t the culprits here—they wouldn’t break down into long enough acid chains to cause bitterness.

So, let’s walk through the process. We’ve just decanted our brandy, gin, rum, whiskey or whatever into a tank, and we’re ready to dilute it down to package strength. Assuming there are long-chain fatty acid esters in the spirit matrix—they’re more common in heavier spirits meant for long-term cask maturation—then we’re looking at the potential for saponification to occur.

Unlike in soapmaking, where they use lye as the base, we’re using water. (Without getting into the heady world of organic chemistry, water can act as either a “base” or an “acid” in all sorts of reactions.) We also know that the addition of water to ethanol releases heat—that’s the exothermic reaction. (Simple experiment: Take a measure of 96 percent ABV neutral spirit, mix it with an equal volume of water, measure the temperature change during mixing—it’s significant.)

And there you go: We have our ester substrate, our base, and some heat to encourage the reaction. If we can remove one of those items then we could, in theory, limit the reaction to some extent.

We could get rid of the fatty acid esters by using chill filtration and/or carbon treatment. However, if filtration isn’t your bag, then the only other option is limiting the heat buildup. We can do that by using temperature-controlled tanks, or by adding the water slowly in increments at such a rate that the temperature doesn’t rise more than a few degrees.

If my own methods are anything to go by—and I make no claim here other than they work for my own projects—then a drop of no more than 5 percent ABV per day is fine.

Questions to Explore

Now, there may be those out there with pitchforks and torches in hand, ready to castigate me for my theories here. Before you do, remember that they are only theories—nothing more, nothing less. To harp on the recurring theme, there just aren’t any solid data—not that have been made public, at least—to back this up one way or the other.

“I have no doubt that local concentration gradients and heat could drive saponification, but I haven’t seen data in support of this,” says Brad Berron, research director for the James Beam Institute at the University of Kentucky. He shares my curiosity on this subject.

“If the heat drives this reaction, would slow mixing be more important in the summer and less important in the winter?” he asks. “Are these same distillers controlling product temperature all the way to store shelves? Slowing the water addition is really allowing the water to diffuse away before adding more. Will we also avoid saponification by getting really aggressive with mixing?”

(Leave it to a scientist of Berron’s caliber to add even more questions that need settling—I mean that as a compliment.)

Above, we’ve mostly considered slow dilution through the lenses of firm believers in its utility. What about the nonpractitioners out there? Are these crash dilutors just pumping out tanker-loads of saponified spirit? Almost certainly not, or else the public would have more to say about soapy-tasting brands from a few large distilling firms.

So, how is that possible?

Indulge in some more theorizing. There are a few things—and maybe more—that could potentially be at play:

• Perhaps they’re diluting in carefully controlled and monitored tanks, keeping temperature swings in check.
• Perhaps they’re removing much of the fatty-acid ester content using the chill filtration or carbon-addition practices that many large spirit houses—including those making American whiskey—regularly employ.
• Perhaps their distillates never contained much in the way of fatty-acid esters in the first place.
• And, continuing from that possibility, maybe the issue of bitter, soapy-tasting spirit is less about dilution and saponification and more about poor fractioning of distilled spirit.

So, where have we landed on the question of whether slow proofing is worthwhile? I’m not entirely sure. I can tell you that with a lack of science as a guiding light, I personally err on the side of caution, diluting my distillates down over several days, or two weeks at most. Others will say it takes months or even years to give the diluted distillate the proper amount of finesse and depth. And still others will say we’re all a bunch of crazy people who should just add the damn water and get on with our days.

Which technique is right? It’s the one that gives you the spirit character that you want in the time that you want it. For now, that’s about all we can say definitively for any of this stuff.