66

In the Netflix documentary Salt Fat Acid Heat, Samin Nosrat says this when she is making sofrito and she is about to put olive oil into the pan:

This is one of those important things that I think home cooks always forget, is how important it is to pre-heat the pan. You have to heat the pan before you heat the oil.

Why do you need to heat the pan before putting the oil in?

Josh Withee
  • 771
  • 1
  • 5
  • 7
  • 21
    This is, as far as I know, a myth. When I find backing for it, I'll post an answer. – FuzzyChef Jun 14 '19 at 16:22
  • The only logic I can come up with is to start heating the pan before adding the oil, because the it all takes longer to warm up than you think and that gains you a few moments of heating. – Chris H Jun 14 '19 at 16:25
  • I've heard two reasons for it, but I don't know if either one's actually true -- 1. it reduces the chance of food sticking (which *is* true for preheating the pan, but I don't know about waiting to add the oil); 2. it lets the pan expand and close any micro-fissures that it may have, preventing the oil from seeping in and making it harder to clean. But of course, this doesn't apply to teflon pans -- always heat those up with oil or water in them, never dry. – Joe Jun 14 '19 at 16:55
  • Joe: both of those are unfounded myths, see my answer below. – FuzzyChef Jun 14 '19 at 17:08
  • 3
    Possible duplicate of [Do you heat the pan first, then add oil? Or put the oil in and heat up with the pan?](https://cooking.stackexchange.com/questions/2690/do-you-heat-the-pan-first-then-add-oil-or-put-the-oil-in-and-heat-up-with-the) – moscafj Jun 14 '19 at 17:35
  • 3
    Moscafj: the question is a duplicate, but the accepted answer on the older question is wrong. So please don't close this one. – FuzzyChef Jun 14 '19 at 17:43
  • 4
    @FuzzyChef: you can close old questions as duplicates of new questions that have better / more-detailed answers. Your answer here looks like a good canonical duplicate for any other questions about this topic, nice job :) – Peter Cordes Jun 16 '19 at 03:05
  • 3
    @moscafj We should close the other way around, keeping this one. – David Richerby Jun 16 '19 at 12:40
  • 2
    Voted to close the other question as a duplicate. I agree that this one should stay open. Great answer @FuzzyChef! – Cindy Jun 16 '19 at 18:12

5 Answers5

115

TL;DR: heating the pan before the oil has no useful effects in most cases.

While this is a duplicate of another question, I'm going to answer it again because that question's accepted answer provides zero evidence or citations to back itself up. Which is important, because the accepted answer is wrong.

The popular myth is "Cold oil in hot pan and food won't stick". Like most such cooking myths, this one is nonsense; as Kitchen Myths points out:

What you really want is “hot pan, hot oil” and that’s what you are actually getting because the cold oil heats up almost instantly when added to the hot pan. You’ll get the same results if you heat the oil along with the pan rather than adding the oil at the last minute. In fact some cooks prefer this technique because the appearance of the oil in the pan can give you some indication of when the pan has reached the proper temperature.

Serious Eats says the same thing:

The thing is, only raw proteins will form this bond. Heat causes proteins to fold in on themselves, or even to break down and form all new compounds. Once in their folded or rearranged form, they no longer stick. So the goal is to get the meat to cook before it even comes into contact with the metal by heating oil hot enough that it can cook the meat in the time it takes for it to pass from the air, through the film of oil, and into the pan.

So, you want a hot pan with hot oil. Most of the time, this means that you want to preheat the oil with the pan, not add oil to a preheated pan, although the latter doesn't do any harm. It just doesn't provide any benefit.

You'll notice I said most of the time, though. There are times when you do want to preheat the cooking vessel before adding fat, and both of those times have to do with needing to heat the metal hotter than the smoke point of the oil you are using.

  1. If you are cooking with a wok, getting proper "wok hei" (searing) requires heating the wok above the smoke point of vegetable oil, dry, a technique called "long yao" (video, skip to 3:22). Classic cast-iron steak cooking uses a similar technique, heating the pan to 250 °C/500 °F before adding the oil or meat.
  2. When cooking with butter as your fat (or a few other low-temperature oils like unrefined coconut oil), the burning point of the fat is sometimes a lower temperature than you want to cook at. If so, the only way to get the pan hot enough is to preheat at dry plan, add the butter or fat, and then quickly add the food before the butter burns.

Neither of these cases has anything to do with preventing sticking, though. They are both about not burning the cooking fat. And the first technique only makes sense if you are using cast iron or carbon steel; it can damage other types of cookware.

You might ask: doesn't this apply to cooking sofrito, though? And the answer is no. Filtered olive oil has a smoke point of 210 °C, which is plenty hot enough for the very wet ingredients in a sofrito, which will drag the real pan temperature down to 105 °C or so in a few seconds anyway. Further, a sofrito is made up entirely of non-starchy vegetables and aromatics, which means that sticking isn't a serious concern.

So, myth busted!

Chronocidal
  • 103
  • 3
FuzzyChef
  • 58,085
  • 18
  • 142
  • 218
  • 36
    The only other reason that I can think of is if the pan is wet. Water in oil would cause the oil to splatter. Heating the pan first would dry it. – MaxW Jun 14 '19 at 21:30
  • 23
    True. That's a pretty specialized reason, though, and a towel would work as well. – FuzzyChef Jun 14 '19 at 23:09
  • 9
    There's also another version of this myth, with cast iron but occasionally with other metal types: that it "opens the pores" of the metal to let it absorb oil, or that it "closes the microfissures in the metal". Neither one of these versions of the myth has a shred of scientific evidence to back it up. – FuzzyChef Jun 14 '19 at 23:13
  • 3
    Hot pan cold oil can reduce sticking precisely by leveraging your caveat #1 - storing up as much heat as possible in the pan, then adding cold oil and food quickly, extends the time (or the quantity of food) that you can sear to a non-sticking state before the pan temperature crashes. The extra heat from an overheated pan can make all the difference between sticking food and not sticking food. – J... Jun 15 '19 at 13:14
  • I know comments aren't supposed to merely say “good” or “bad”, but I feel it has to be said that this answer is really _excellent_. – leftaroundabout Jun 15 '19 at 18:31
  • 5
    @FuzzyChef Heating the pan will make any cracks wider. Suppose the material looks like `##_##`. If you heat it, it expands and looks like `####__####`. I see no reason to believe that this would have any effect on cooking, but it's clear that the crack gets wider. – David Richerby Jun 15 '19 at 20:45
  • David: you have a link for that? I'm curious, more for blacksmithing than for cooking ... – FuzzyChef Jun 16 '19 at 05:20
  • lefaroundabout: thank you, I was really trying to make it a canonical answer to a fairly common question. – FuzzyChef Jun 16 '19 at 05:24
  • 4
    @FuzzyChef The other answer has a link to a Physics SE question about this; I don't have a better link. But, if you think about it, the only way the crack could close would be if the material behind its walls got wider. Well, that material _does_ get wider, but the material underneath it gets wider by exactly the same amount, so the walls of the crack don't get any closer. Additionally, the material at the floor of the crack gets wider, so the crack spreads. – David Richerby Jun 16 '19 at 12:39
  • 5
    @FuzzyChef CTE scales the entire object with temperature, [including holes in the object](https://physics.stackexchange.com/q/12599/44080). – J... Jun 16 '19 at 20:05
  • 1
    There are many other important cooking styles and fats that burn at temperatures discussed here that should be taken into account. Most people who want to brown meats or get crispy skins or char or saute or brown even scallops will usually cook at temperatures that would affect fats if they sit in a preheating pan for long. – Behacad Jun 16 '19 at 23:44
  • 1
    Behacad: really, they don't. Refined vegetable oils heat up to 200-220C, which is plenty hot enough for searing seafood, chicken, or most other foods. And, per my comment on your other answer, heating pans above those temperatures can damage many popular types of cookware. – FuzzyChef Jun 17 '19 at 04:53
  • @FuzzyChef: The question specifically mentions Olive Oil. – Jack Aidley Jun 17 '19 at 19:52
  • I think this is a case where new cooking methods and older or "traditional" cooking methods collide. But in this case it's more like new cooking equipment and older cooking equipment have different approaches. – JG sd Jun 20 '19 at 12:14
  • 1
    @DavidRicherby Love the graphics! – gnicko Jun 20 '19 at 21:10
  • 1
    @JackAidley filtered, cleaned olive oil is a refined vegetable oil, and has a smoke point of 210C ... per the answer. If the OP was using specifically raw, unfiltered olive oil then case (2) would apply, just like using butter. – FuzzyChef Jun 25 '19 at 04:05
  • 2
    Based solely on this answer, I think the proper conclusion should be to heat the pan before adding oil. You state there is no harm in heating the pan first. Then you state there are sometimes benefits to heating the pan first. Why isn't the best and simplest set of instructions to always heat the pan first? – mattm Jun 25 '19 at 15:03
  • There are some advantages to heating the oil in the pan, such as being able to tell the temperature of the pan by the behavior of the oil, and that for a nonstick pan it's a really bad idea to heat it empty. There's also the question of knowing how long to wait after oil hits a hot pan so that it's hot but not burning. Really, though, I was just trying to dispell the myth that heating the pan first conveyed some kind of general advantage, not suggest that you *shouldn't* do it. I mean, sometimes I heat the pan first, it depends. – FuzzyChef Jun 28 '19 at 03:11
  • @FuzzyChef - while the answer is admirable and you correctly scorch the various ridiculous myths, one could say you need to emphasize the concept of point 2. Say you are cooking with oil. If you put oil in a cold pan, and heat the pan, you will end up with **extremely hot oil** (as hot as the pan, more or less). However, if you heat the pan and then put in the oil, **you can choose the temperature of the oil** ie, oil heating for a short time (before the ingreds) you get only slightly hot oil; oil heating for a longer time (before the ingreds) you get hotter oil (but not pan-temp scalding). – Fattie Jun 03 '21 at 21:38
  • Fattie: that would be true only if "choosing the temperature of the oil" conveyed some kind of advatage in most recipes, which it doesn't. I'll point out that heating the pan, then adding the oil, takes more time and uses more energy, so it's a *disadvantage* if there's not some specific reason you need the oil to be cooler than the pan ... something already discussed in my answer. – FuzzyChef Jun 04 '21 at 18:09
  • `So, you want a hot pan with hot oil. Most of the time, this means that you want to preheat the oil with the pan, not add oil to a preheated pan`. But you will still get a hot pan with hot oil by adding oil to a preheated pan, since your answer also says `What you really want is “hot pan, hot oil” and that’s what you are actually getting because the cold oil heats up almost instantly when added to the hot pan. You’ll get the same results if you heat the oil along with the pan rather than adding the oil at the last minute.` – pacoverflow Sep 25 '22 at 15:40
16

Regarding the (mis)conception that heating the pan will close its microfissures so micropieces of food will not fall into the small holes sticking to them: wrong. Check the response to the more general problem of thermal expansion of a solid with a hole

https://physics.stackexchange.com/questions/12599/will-a-hole-cut-into-a-metal-disk-expand-or-shrink-when-the-disc-is-heated

The only meaningful reason to heat the pan before the oil is to dry the pan, so the water droplets will not mix with the oil, then starting splattering around hot oil drops when the water-oil mixture gets heated to T>100C. As @MaxW in the comments noted, a towel would do the same.

EarlGrey
  • 261
  • 1
  • 4
  • I agree with the physics, when you heat a stuck nut on a car it's because the whole nut expands, including the inside, making it big enough to spin off. – GdD Jun 17 '19 at 08:43
  • But what if the bottom of the pan is significantly hotter than the top? For a thick pan, that could cause slight bending that would make holes on the top surface smaller. Most pans are designed not to bend when they are heated, though. :) – Justin Jun 17 '19 at 15:08
  • 1
    @Justin Heat conduction through a metal pan is good enough that you're not going to get any real bending in normal situations, especially given that a thicker pan is more rigid. Rather, the temperature gradient will work to widen the cracks even more: the top of the crack is cooler, so expands less than the hotter base. – David Richerby Jun 17 '19 at 15:46
13

Recipes often call for a warm up for a reason similar to preheating an oven. The recipe author has no idea how heavy you pan is. What the author does know is that for just about any hot pan, 2Tbs olive oil will heat within a few seconds, where you can saute or whatever for a specified duration.

Stuart W
  • 139
  • 2
  • Hey, that's the most legit reason I've heard. Thanks! – FuzzyChef Jun 17 '19 at 04:54
  • 5
    Why does the recipe writer care whether I heat the oil slowly as the pan warms up, or quickly by pouring it into a hot pan? As long as the pan and oil are hot enough when I start sauteing, it doesn't matter what happened in the past. – David Richerby Jun 17 '19 at 15:40
  • Perhaps the recipe author is trying to help a less-experienced cook get the pan to the correct temperature. "Heat the pan until a drop of water evaporates immediately" is hard to mess up, and not safe to do with oil already present. – arp Jun 17 '19 at 16:01
  • 1
    @DavidRicherby I think this answer is conflating *starting cooking* pre/post heat with *placing the oil* pre/post heat. In the "preheating an oven" scenario, then what this answer points out is correct. You can instruct "pre-heat to 200C then cook for 15 minutes" but you cannot write "Cook for 20 minutes in a cold oven set to 200C" because you don't know how long it will take to warm up. But that doesn't apply to oil, because the duration commences when you place the food in the pan, not when you place the oil in the pan. – JBentley Jun 17 '19 at 22:37
  • 1
    We aren't talking about a recipe author though. We are talking about a professional chef who believes the pan should be hot before putting the oil in ... If _her_ real concern is making sure that the oil is fully warmed up for cooking, I think she would have made that her point. – Josh Withee Jun 19 '19 at 12:54
5

Samin Nosrat writes in her book Salt Fat Acid Heat:

Preheat the pan to reduce the amount of time fat spends in direct contact with the hot metal, minimizing opportunity for it to deteriorate. As oil is heated, it breaks down, leading to flavor degradation and the release of toxic chemicals. Food is also more likely to stick to a cold pan—another reason to preheat. But exceptions to the preheating rule exist: butter and garlic. Both will burn if the pan is too hot, so you must heat them gently. In all other cooking, preheat the pan and then add the fat, letting it too heat up before adding any other ingredients.

For a heavy pan and an underpowered stove, the difference can be few minutes versus few seconds for fat at gradually rising temperature. How much of a difference this makes depends on the oil, but various processes (e.g. cis-trans isomerization) involved in burning fat vary their speed with temperature, so longer time spent in the let's say range of 180-200°C really results in more degradation.

Source: my wife's a chemist

Jan Pulmann
  • 51
  • 2
  • 2
    Thanks for that reference filling in Nosrat's thinking. I would love a chance to argue this out with her, given the citations in my answer; one simply isn't eating refined oils to degredation temperatures unless one has forgotten the pan on the stove. 200C is a higher temperature than most sauteeing takes place at. Anyway, a valuable point. – FuzzyChef Jun 26 '19 at 02:07
2

The major reason I can think of for adding oil to a pan after it is hot is safety: if you are heating a pan of oil and then forget it could start a pan fire. If you add the oil only when you are ready to cook you don't have that risk.

I often add the oil when I first heat the pan because I can watch the oil and I can tell when it's hot by looking at it, but then sometimes I heat the pan dry and use the back of my hand to test for heat. Either method works just fine.

GdD
  • 74,019
  • 3
  • 128
  • 240
  • How would you forget though? If you're deep frying then sure, because that can take a long time. For sauteeing though your pan heats up in a few tens of seconds, so you don't have an opportunity to wander off. And even if you did manage to forget it, you've only got a tablespoon of oil so you won't get it to light off in the same way. – Graham Jun 17 '19 at 09:58
  • @Graham It happens. I've done it. More than once. – Summer Jun 17 '19 at 10:04
  • Kids are amazing distractions @Graham. You hear a nasty thump and a scream and next thing you know you're racing upstairs. – GdD Jun 17 '19 at 10:35
  • 2
    @GdD But even then, you don't risk a pan fire. There simply isn't enough oil. If there is then you're frying in some depth of oil, not sauteeing, and then the temperature of the pan is completely immaterial, because even the heaviest pan doesn't retain enough heat to get that much oil up to cooking temperature. So if you've been heating up the pan thinking that it'll change the temperature of a hundred ml or more of oil, then you're doing it wrong to start with. ;) – Graham Jun 17 '19 at 11:11
  • 2
    @GdD Besides which, if you leave a stove on and unattended with kids around, there are *way* more serious safety issues. Maybe I'm a slightly cold-blooded parent, but triage says if they're screaming then they're still alive, whereas if they catch their clothes in a gas flame then they won't be. So I shout "I'm coming" so they know I've heard, and spend a few seconds to make sure what I'm leaving is safe. Ditto lawnmowers, live electrics, etc... – Graham Jun 17 '19 at 11:19
  • 3
    @Graham - It's more the quick things near the kitchen that are a problem. When your two year old is crying for a cup of milk, say. You can pour it in less than 10 seconds, but then you knock over the cup...onto your computer...and next thing you know the pan is smoking. – Justin Jun 17 '19 at 15:11
  • 2
    But, conversely, if you're heating a dry non-stick pan and get distracted, you can end up with extremely toxic fumes. If there was oil in the pan, the pan wouldn't be able to get much hotter than the boiling point of the oil until all the oil had gone. – David Richerby Jun 17 '19 at 15:43
  • 2
    My preferred method to "tell when it's hot" is to place a very small piece of the thing I'm cooking (e.g. onions) in the pan. It's hot when I hear it start to sizzle (to me a clearer sign than looking at the oil), and it won't make any difference to the outcome that the one small piece got a few seconds head start. Of course this too requires the oil to go in before it is heated. – JBentley Jun 17 '19 at 16:46
  • 1
    Heating a dry pan and forgetting about it is also fairly dangerous. – FuzzyChef Jun 20 '19 at 22:12