The flames may disappear in seconds. The alcohol is another matter.
There are few moments in restaurant service quite like a tableside flambé. A pan is brought to the table, the spirits are added, and suddenly a blue flame rises above the food. Conversations stop. Heads turn. For a few seconds, the dining room has a little theater of its own.
I spent a good part of my restaurant career performing that theater. At the Ship's Tavern in the Moana Hotel, I prepared Crêpes Suzette tableside, one of those classic desserts that combined cooking, service, and a little showmanship. Later, at Hy's Steak House in Waikīkī, I created my own flambé preparations, including Sinatra's Strawberry Flambé, named after my Doberman. I always enjoyed that moment when the flames rose and the guests leaned forward. Even after years of doing it, there was something satisfying about turning dessert into an occasion.
But there's a question I don't recall guests asking very often. Once the flames disappear, what happens to the alcohol? Most people assume it has burned away. After all, they just watched it burn. The answer is a little less spectacular than the performance.
What Actually Burns During a Flambé?
When brandy, rum, or another spirit is heated in a pan, some of its alcohol evaporates into the air above the liquid. When we ignite the spirit, it is primarily that alcohol vapor that catches fire, not the liquid itself. The familiar blue flames are the visible result of alcohol vapor burning above the food. As the vapor is consumed and the conditions become less favorable for combustion, the flames eventually disappear.
What doesn't happen is the instantaneous disappearance of all the alcohol in the pan. The flame creates a convincing illusion because we see something burning and naturally assume that whatever was burning must now be gone. But the liquid remaining in the pan is still a mixture of water, alcohol, sugars, and other compounds. Ignition removes some alcohol, while much of the rest remains in the food.
And the amount remaining is surprisingly substantial.
The Numbers Behind the Fire
Research published in the Journal of the American Dietetic Association in 1992 examined alcohol retention in foods prepared by different cooking methods. The findings helped establish the alcohol-retention figures subsequently published by the USDA.
According to those figures, a flambé can retain approximately 75 percent of the alcohol originally added. In other words, the dramatic burst of flame may remove only about one-quarter of it. Adding alcohol to boiling liquid and then removing the preparation from the heat can leave even more behind, approximately 85 percent.
Cooking for longer periods makes a substantial difference. In the USDA reference figures, a dish baked or simmered with alcohol for 15 minutes retains approximately 40 percent of the original alcohol. After 30 minutes, the figure falls to 35 percent. An hour of cooking leaves approximately 25 percent, while two hours reduces it to 10 percent. Even after two and a half hours, however, about 5 percent may remain.
These figures are drawn from specific preparation conditions, not guarantees for every recipe. The ingredients, cooking vessel, temperature, and amount of evaporation all influence the results. Still, they illustrate something the flames alone cannot tell us: a few seconds of fire do not accomplish what prolonged cooking can.
Even these percentages require a little interpretation. Retaining 75 percent of the original alcohol does not mean the finished dish contains 75 percent alcohol by volume. If a small amount of brandy is divided among four servings of dessert, the actual quantity in each portion may be relatively modest, even though much of the alcohol survives. The percentage retained tells us how much of the original alcohol remains, not how alcoholic the finished dish is.
Does Lighting the Pan Actually Make a Difference?
A study published in the International Journal of Gastronomy and Food Science examined the effects of heating and flaming separately. Researchers compared preparations that were heated and ignited with similar preparations that were heated without ignition. They wanted to know how much of the alcohol loss, browning, and sensory change could actually be attributed to the flame.
In an experiment using a caramel sauce made with butter, sugar, and vodka, the researchers found no statistically significant difference in ethanol loss between the flaming and non-flaming preparations. The losses were approximately 13 to 14 percent in both treatments. The dramatic burst of flame had contributed remarkably little to the disappearance of the alcohol.
In a simpler experiment using vodka alone, the researchers measured greater alcohol loss in the ignited preparation, but heating still accounted for most of the loss. The broader conclusion was that heating, rather than combustion, was responsible for most of the alcohol removed during flambéing.
The temperature measurements were equally revealing. During 15 seconds of flaming, the researchers recorded flame temperatures reaching 532°C, or nearly 990°F. Yet just one centimeter above the surface of the pan, the maximum measured temperature was only 67°C, approximately 153°F. That's below the boiling point of water and well below the temperatures normally associated with significant browning reactions over such a short period.
It seems extraordinary that a flame approaching 1,000 degrees can look so impressive while transferring relatively little heat to the food beneath it. The researchers also found no significant difference in measured browning between the flamed and unflamed caramel preparations.
That doesn't make flambé pointless. The presentation is part of the pleasure, and the researchers found that flaming could produce detectable sensory differences under certain conditions. But it does challenge the assumption that setting a preparation on fire is a particularly effective way to eliminate alcohol.
For those of us who spent years entertaining guests with a flaming pan, it's a rather amusing discovery. Apparently, we were better showmen than alcohol-removal specialists.
Why Time Matters More Than Flames
Alcohol has a lower boiling point than water, approximately 173°F (78°C) compared with water's 212°F (100°C) at sea level. That fact has contributed to another common misconception: once a dish gets hot enough, the alcohol simply boils away.
But alcohol and water don't behave as two completely separate liquids when they're mixed together. Both evaporate during cooking, although alcohol generally escapes more readily. It doesn't all vanish the moment a sauce reaches alcohol's boiling point. Instead, the alcohol concentration changes gradually as the preparation cooks and liquid evaporates.
A shallow sauté pan, a deep stockpot, and a covered braising vessel all create different conditions. The original concentration of alcohol, the ingredients, and how much liquid remains also influence the result. A vigorously reduced sauce may lose alcohol quite differently from a stew that barely simmers.
Research published in Food Chemistry in 2017 examined alcohol loss in meat stocks prepared with wine and beer. The researchers found that ethanol concentration was closely related to the proportion of liquid remaining, rather than cooking time alone. Their mathematical model described how alcohol concentration declined as the stock reduced, with the relationship holding across different pot sizes and cooking temperatures.
One finding was particularly unexpected. When researchers covered the pot, the alcohol concentration decreased more steeply relative to the reduction in liquid volume. That doesn't mean putting a lid on a stew is automatically the fastest way to remove alcohol. It means the relationship between alcohol evaporation and water evaporation is more complicated than the familiar advice to leave the lid off.
Consider the difference between a quick wine reduction and a traditional braise. When a chef adds wine to a hot sauté pan and reduces it vigorously, the goal is often to concentrate flavor while allowing volatile compounds, including alcohol, to escape. In a braise, the wine may be combined with stock, vegetables, meat, and aromatics and cooked gently for several hours.
Both preparations lose alcohol, but neither guarantees that every trace disappears.
What About Beef Bourguignon, Coq au Vin, and Other Wine-Braised Dishes?
Consider beef bourguignon, one of the classic examples of cooking with wine. A generous quantity of red wine becomes part of the braising liquid, contributing acidity, fruit character, and complexity as the meat slowly cooks. Over several hours, the wine becomes integrated with the stock, meat juices, vegetables, and aromatics. By the time the beef is tender, the liquid bears little resemblance to the wine that went into the pot.
Much of the original alcohol may have evaporated during that time. The USDA figures suggest that extended cooking can reduce the amount substantially, but even after two and a half hours, some alcohol may remain. The exact percentage depends on the preparation, so the reference figures should not be treated as a promise that every long-cooked dish will end up with precisely five percent of its original alcohol.
The finished dish is not the equivalent of drinking the wine that went into the pot. The alcohol is distributed throughout the preparation, and its concentration is generally much lower than that of the original wine. But calling it alcohol-free would be another matter.
The same reasoning applies to beer-braised meats, wine sauces, and desserts prepared with liqueurs. A spoonful of flambéed dessert sauce and a serving of beef bourguignon may both contain residual alcohol, but their actual quantities could be quite different. The recipe, serving size, and preparation method matter far more than whether the chef happened to set something on fire.
Can You Taste the Alcohol That Remains?
Not necessarily. Alcohol contributes to aroma, flavor perception, and the way certain ingredients are experienced, but the unmistakable sharpness of an alcoholic beverage can soften during cooking. Wine also contributes acidity, sweetness, tannin, and aromatic compounds. Spirits bring their own character: the warmth and vanilla notes of brandy, the molasses richness of dark rum, or the distinctive flavors of whiskey.
As a sauce cooks, those flavors change. Some aromas disappear, others become more prominent, and the remaining ingredients begin to dominate. What began as a noticeably alcoholic preparation may eventually taste beautifully integrated.
That integration is one of the pleasures of cooking with wine and spirits, but it can also be misleading. The disappearance of an obvious alcoholic taste is not proof that the alcohol itself is gone. A sauce can taste balanced, mellow, and entirely free of alcoholic sharpness while still containing ethanol.
Our palates are remarkably useful instruments in the kitchen. Unfortunately, they aren't laboratory equipment.
When Does the Remaining Alcohol Matter?
For many adults who are comfortable consuming alcohol, the residual quantity in a serving of a cooked dish may be relatively small. But there are circumstances in which even a small amount matters. Someone avoiding alcohol for religious reasons, pregnancy, medication interactions, recovery from alcohol dependence, or personal preference may reasonably expect a genuinely alcohol-free preparation when told that the alcohol has cooked away.
In those situations, the responsible answer is not an assurance based on the appearance of a flame or the length of time a dish has simmered. Cooking generally reduces alcohol, sometimes substantially, but does not reliably eliminate it. If complete avoidance is important, a preparation made without alcoholic ingredients is the more dependable choice.
From a hospitality standpoint, this matters. A guest asking whether a dish contains alcohol deserves an accurate answer, not a reassuring guess. We should be just as careful with ingredients that guests choose to avoid as we are with those they cannot safely consume.
There is a difference between explaining that most of the alcohol has evaporated and promising that none remains. Good hospitality requires knowing the difference.
The Flame Was Never the Whole Story
Looking back on my years of tableside service, I don't think knowing these percentages would have made the flambé any less enjoyable. Quite the opposite. It adds another layer to something I already appreciated. From Crêpes Suzette at the Ship's Tavern to the strawberry flambé I named after Sinatra, the pleasure was always about more than the flames themselves.
There was the preparation, the aroma, the movement of the pan, and the anticipation as the guests watched. The flame was the climax of a small performance, but the experience began long before the match was struck. And, of course, we still had to make something delicious.
Guests weren't ordering a chemistry demonstration. They were participating in a restaurant tradition that made dinner feel like an occasion. A good tableside preparation brought the cooking into the dining room and made the guest part of it, something I believe we've lost a little of in modern restaurant service.
What the science tells us is that the performance and the chemistry are two different things. Fire creates the spectacle. Heating and evaporation determine much of what remains. Neither takes anything away from the other, provided we understand what the flame actually accomplishes.
The next time someone tells you that the alcohol burns off because the chef set the pan on fire, you can appreciate the show while knowing a little more about what's happening. The flame may be gone, but Brandy hasn't left the building.
Research and Sources
Augustin, J., Augustin, E., Cutrufelli, R. L., Hagen, S. R., & Teitzel, C. (1992). Alcohol retention in food preparation. Journal of the American Dietetic Association, 92(4), 486–488.
https://pubmed.ncbi.nlm.nih.gov/1556354/
USDA Agricultural Research Service. (2007). USDA Table of Nutrient Retention Factors, Release 6.
https://www.ars.usda.gov/ARSUserFiles/80400535/Data/retn/retn06.pdf
Hansen, C. E., Kwasniewski, M. T., & Sacks, G. L. (2012). Decoupling the effects of heating and flaming on chemical and sensory changes during flambé cooking. International Journal of Gastronomy and Food Science, 1(2), 90–95.
https://doi.org/10.1016/j.ijgfs.2013.04.001
Snitkjær, P., Ryapushkina, J., Skovenborg, E., et al. (2017). Fate of ethanol during cooking of liquid foods prepared with alcoholic beverages: Theory and experimental studies. Food Chemistry, 230, 234–240.
https://doi.org/10.1016/j.foodchem.2017.03.034
If this essay resonates, Hospitality Between the Lines is just below.
