Alcohol Doesn’t Really “Cook Out” of Food in Most Cases

Today I found out alcohol does not “cook out” of food in most cases.  The myth that alcohol does all cook out stems from the fact that alcohol has a much lower boiling point temperature (173° F / 78.5° C) than water (212° F / 100° C).  Thus, if the temperature is above 78.5° C, then the alcohol should boil off, right?

A group of researchers in 1992 at the U.S. Department of Agriculture, the University of Idaho, and Washington State University decided to find out.  In the end, what they discovered was that the “alcohol cooks out” assumption didn’t turn out to be correct for the vast majority of ways most people prepare food with alcohol.  In their study, they used a variety of recipes with various sources of alcohol and a variety of preparation types, including simmering, baking, flaming, refrigerating over night, etc.  What they found was that the amount of alcohol remaining after cooking was in the range of 4%-85%.  The variations weren’t just dependant on how long you kept the temperature above alcohol’s boiling point either.  They also found that the other ingredients made a difference in the alcohol retention rate.  The size of the cooking vessel also greatly affected the alcohol retention rates.  The smaller the vessel, the more alcohol will be retained given some set cooking time, due to the smaller surface area for evaporation.

In terms of preparation methods and times, their results were as follows (all of these assume the temperature is above 173° F, which is the boiling point of alcohol; also, the size of dish and contents of the food mixture affect the results, so this is just a general guideline):

  • The highest rates of retention were with alcohol added to boiling liquid and then shortly after removed from heat.  In this case, the alcohol retention rate was around 85%.
  • The second highest alcohol retention rate came when using the flaming method of cooking, which resulted in around a 75% retention level.
  • When using no heat and storing overnight, about 70% of the alcohol was retained.
  • When baked for 25 minutes with the mixture not being stirred, the retention rate was 45%.
  • When baked/simmered where the mixture is stirred, produced the following results:
    • 15 minutes 40%
    • 30 minutes 35%
    • 1 hour 25%
    • 1.5 hours 20%
    • 2 hours 10%
    • 2.5 hours 5%

Now, you might think from this that if you cooked the thing long enough, eventually the alcohol will all get cooked out.  From a practical standpoint, this is more or less true.  But if you are ever cooking for or are a recovering alcoholic, you’ll want to know, it’s not really true.  There will always be some alcohol remaining as long as there is still any kind of moisture in whatever you are cooking.   The reason behind this is that the alcohol binds with water and forms an azeotrope (mixture of two or more compounds where the ratio cannot be changed by simple distillation).  So as you boil the azeotrope, the ratio of alcohol in the compound stays the same throughout the boiling process.   So you will always retain some alcohol, unless you boil off all the liquid.

Bonus Fact:

  • Besides for flavor, alcohol is added to fondue because it lowers the boiling point of the cheese and, thus, helps to prevent curdling.
Expand for References
Share the Knowledge! facebooktwittergoogle_plusredditpinteresttumblrmailfacebooktwittergoogle_plusredditpinteresttumblrmail
Print Friendly
Enjoy this article? If so, get our FREE wildly popular Daily Knowledge and Weekly Wrap newsletters:

Subscribe Me To:  | 


  • While this may all be true, you also have to do some maths — for example, a dish that calls for two tablespoons of cooking wine and makes four servings ends up with the equivalent of one sip per person, spread through the entire serving; it’s even less if the wine’s in, say, a marinade which is largely discarded. Even for alcohol-sensitive, recovering alcoholic, or underage diners that’s not going to be a meaningful concentration even if not an ounce cooks off.

    • As someone who does not want to drink ANY alcohol, I find even a “sip” as unacceptable. ty

      • @foxxtrot You do realize that ethanol occurs naturally in fruits and vegetables? There’s even an ISO standard to determine the ethanol content in a standardized manner (ISO 2448). So are you going to give up all plant matter in your diet due to some bizarre moral disdain for “ANY alcohol”?

    • I’ve been using antibuse for 7+ years. my wife considers wine to be a condiment. I have had no reactions.

  • Actually this person is about 90% wrong…i have been apart of 2 experiments dealing with this…and in most pf the tests the alcohol burned out …and the alcohol thats was left i. Some foods was very very little…

    • Daven Hiskey

      @patrick: which experiments? Please cite the papers showing the results. Also, to make a more compelling argument, you should probably explain why the U.S. Department of Agriculture, the University of Idaho, and Washington State University’s study’s results were wrong on this.

  • Contradictary and misleading. How can you say it doesn’t cook out, when clearly it cooks out. You might say it does not completely cook out 100%. But it is cooking out.

  • this is actually true. As someone with a liver illness, I’ve been directed not to eat food that has been cooked with alcohol as it does not cook out fully. Even a sip could harm me.

  • So when adding alcohol to food I add it at the last possible moment and then remove it from the heat. ( You cook your way, I’ll cook mine)

    • Yes of course it is up to you how you cook. The only difference is now you are aware that you (and those dining with you) will be consuming alcohol.

  • This post definitely led me to do a bunch of digging, since I cook with beer a lot (mostly for braising or in stews). There’s been some more research since the Augustin studies; one study in particular (Mateus et al 2011, in Food Chemistry) found that 70% of alcohol was cooked out of a fish stew (when alcohol was added with cold ingredients and cooked, covered, for 45min) and 95% was cooked out of a beef stew (when added to boiling ingredients and cooked, covered, for 1hr).

    Based on the Augustin and Mateus papers, it seems like most of the alcohol does actually cook out of food (in a lot of common preparations), but that it’s also tricky (impossible?) to cook out 100% of the alcohol without REALLY cooking the hell out of it.

  • What about like using beer as a marinade, lets say soaking a steak for an 1 hr period, take the meat out and grill it. Not cooking in beer.

    • Very minimal unless you inject it. The searing would be hot enough to burn out most of it. You can always switch to O’douls if it worries you.

  • I’ve done a little checking into this too. and I think the part that’s missing is what constitutes ‘alcohol’. In neither of the studies I’ve looked at (the two mentioned above), has the proof of the alcohol being used been clarified.

    I find it hard to believe that if I cooked a stew in scotch for 3 hours that the amount of alcohol remaining would be the same as if I’d cooked it in beer or cider.

    If I’m right, then it makes all of this less than useful and jut conjecture, when it comes to cooking.

    All the best,


  • Richard,
    All of these quantities are relative to the original amount of alcohol used, not absolute. They are percentages, so the math still works.

  • As a recovering alcoholic, I’ve written a similar article citing the statistics from the USDA.

    How one cooks for their home or establishment is fine, but if you use alcohol in the prep, please list it.

    Even simple places like NOODLES use wine and sherry in their preps. Many can be made with out them.

    To those of you who think we are being “Silly” over our concern….. would you tell someone with seafood allergies “there is only a little bit in here, go ahead and eat it” or those with peanut allergies “It was only cooked with a little bit”

    A little bit is enough to KILL.

    Perhaps the alcoholic content may not be much, but the flavor is there, and that can be enough to trigger. I recall checking a menu in an irish place very carefully, and when my meal came the beef had been brazed in Guiness. It scared the heck out of me.

    Thank you.

  • “Temperature is about 75 degrees Celsius, which is hotter than is strictly comfortable, but would still be manageable.” That comes out to 167 degrees Fahrenheit–not “strictly comfortable,” indeed. Alcohol boils at 173.

  • They are just saying temperature was above 173 F. I would assume also the rate of evaporation of alcohol would be temperature dependent. May be kind of a faulty chart without temperature listed.

Leave a Reply

Your email address will not be published. Required fields are marked *