The Proof of the Potation Is In The Pyrotechnics

The Dallas Observer reports:

Now, I have always been told that liquors burn at 100 proof (50 percent alcohol) or higher. In fact, some sources claim fires only start around the 57 percent range. There are even stories from the past seeming to confirm this. Tavern owners in the 1700 and 1800s would set a match to samples of whiskey brought in by dealers. If they lit, it was “proof” of the alcohol content. If not, they considered the barrel to be watered down or otherwise inferior. However, the B-52 flares when bartenders touch a lighter to orange liqueur. Same when French restaurants torch crepe suzette: it’s Grand Marnier (40 percent) that bursts into flame. And one of Cassel’s recipes at a previous restaurant required him to reduce then ignite port wine.

Here’s what we found:

Grand Marnier indeed lights up quickly. The flame is low and steady, burning blue to yellow–and is easily extinguished. Other 40 percent spirits also ignite, though with less force. When poured onto a plate and lit, for instance, vodka (at room temperature) carries a small blue plume over about half its surface area. Same with tequila (100 percent agave reposado), although the flame is almost invisible. Both are rather unstable, the slightest movement or breath being enough to douse them. Yet even a liqueur rated at 38 percent burned momentarily.

Higher proof samples ignite instantly–and they resist attempts to extinguish the resulting blaze. For example, Chartreuse (55 percent) literally erupts, the burst from a tablespoon of the bitter liqueur drizzled on a plate reaches five or six inches in height. When I lit this next to a sink, my quick but clumsy effort to move the fiery plate under the faucet caused flickering liquid to splash around the drain… where it continued to burn. …

But the old stories are wrong. It’s not just the 100 proof stuff that burns.

VinePair elaborates on the science and reports further details:

Things are a little more scientific these days. We now know that it’s the alcohol vapor that catches fire, not the liquid, just like with gasoline. More vapor comes off the alcohol as the alcohol heats up, making it easier for it to light. Colder alcohol means less vapor, which means less chance for flames.

The vapor also impacts what you can use as a lighter. A match elicits less vapor from the alcohol than a torch, which is much hotter.

So how low can you go (in alcohol by volume) before the floater of alcohol on that would-be burning cocktail turns into a failed party trick? Here’s the guide, from top to bottom.

Everclear

Sells at either 75.5 percent alcohol by volume, or 95 percent (although the latter is illegal in some states). Everclear will maintain a consistent and difficult-to-put-out flame.

Lit Level: Dangerous.

Bacardi 151

It’s right there in the name: 151-proof, or 75.5 percent alcohol by volume. Once again, consistent flame to be expected.

Lit Level: Dangerous.

Absinthe

Absinthe can range from 45 to 74 percent alcohol by volume. It’ll catch on fire, but you don’t really want to ruin the taste of a good absinthe, right? Will light and maintain a flame, but it will go out easier.

Lit Level: Moderate to Dangerous.

Whiskey, Vodka, Tequila and Gin

Throw it back to history for this one. Cask strength whiskey will quickly light on fire and sustain a decent burn without being impossible to put out. The more common stuff, which averages around 40 percent alcohol by volume, will burn a small blue flame over the top of the drink that goes out with a light breath — same for vodka, tequila and gin.

Lit Level: Moderate.

Light Liqueurs, Wine and Beer

Don’t bother trying to light a floater of any alcohol under 40 percent alcohol by volume. You can get a little action in the 30-percent range, but it’s not worth the trouble.

Lit Level: Not Lit.

So the putative early proof-testing seem to have been binary: alcohol was assumed to burn if and only if it was at least 100 proof. Modern investigation elaborates that the appearance and durability of the flame are indicative of the percentage of alcohol content. A third variation of this procedure, however, is reported in the eighteenth century Polish work שו”ת הר הכרמל: the liquor is ignited and the alcohol burned off. Measuring the remaining quantity of liquid (and presumably comparing it to the initial quantity) enables the determination of the percentage of alcohol content:

מה שנוהגין לפעמים לבחון משקה יין שרף כמה מחזיק הוא ערך מים שבו ושופכין היין שרף על כלי בדיל ומדליקין היין שרף לראות כמה נשאר מים ושאל השואל אם יש לכלי זה על ידי כך דין נשתמש חמץ על ידי אור ובעי ליבון ואין לכלי זו תקנה לפסח או נימא דמקרי שלא על ידי אור ובהגעלה סגי:1

  1. שו”ת הר הכרמל או”ח סימן ה’, הובא בשערי תשובה (מרגליות) סימן תנ”א ס”ק כ”ז []