The Heat a Cast Iron Remembers

I have just lifted a piece of steak off a cast-iron skillet and set it on a rack to rest, and I have turned the burner off, and the pan is now sitting on the dead burner with nothing in it but the small fond of meat and the last of the rendered fat. The flame is off. The thermostat behind the wall is silent. By every conventional reading the heat in the pan should be on its way down. And yet, with my hand held a few inches above the metal, I can feel that the heat in the pan is not on its way down in any particular hurry. It is going to stay close to where it is for the next four or five minutes. It will be hot enough to crisp the edge of a piece of bread in two minutes. It will be hot enough to wilt a handful of spinach in three. It will still be warm enough at fifteen minutes to be useful for setting a sauce, and it will still be too warm to put a bare hand on at twenty. The pan does not know the burner is off. The pan is doing what cast iron does, which is to hold the heat it was at and let go of it slowly, and the cook who plans for that release gets ten more minutes of useful pan out of the same fifteen minutes of gas.

I do not pretend to have the thermodynamics of cast iron pinned in any laboratory sense. My best guess after years of cooking on heavy pans is that what is happening is straightforward at the bulk level even if I would not care to defend the numbers. Cast iron has a high mass and a relatively low conductivity compared to copper or aluminium. The high mass means that the pan, when it is hot, contains a lot of stored thermal energy per square inch of cooking surface. The low conductivity means that the pan does not give that energy back to the room or to the food quickly. The combination is a kind of thermal slow-release — the burner takes a long time to bring the pan up to temperature, and the pan takes a long time to come back down. I have heard it described as a thermal flywheel, which I think is the right metaphor: once it is spinning, it keeps spinning, and the cook who learns to read its momentum can stop adding energy several minutes before the work is finished and still have the pan do the rest of the work for them. I would not swear to specific time-to-temperature curves and would not want to. What I would say is that the principle is reliable across pans, across burners, across burner outputs, and across cooks, which is more than I can say for most things.

The pivot, then, is the moment the burner can come off. I have come to think of cooking on cast iron as a planning exercise as much as a cooking exercise. Before I start I am estimating how much heat the pan needs to be at for the longest piece of work I am about to do, and once I have it there, I am estimating how much heat will leave the pan between the longest work and the shorter pieces of work that come after it. A steak, in my experience, takes about three minutes on a side on a properly preheated cast-iron pan; the pan, having seared the steak, is now somewhere between 150 and 200 degrees centigrade depending on what kind of mood my best guess is in, and that heat is enough to do the next several minutes of work with no flame underneath it. I will lay a halved tomato cut-side down on the pan after the steak comes off, with no burner, and the tomato will blister at the edges and warm through to its centre by the time I am ready to plate. I will toast a piece of bread on the pan in the next four minutes, again without lighting the burner. I will fold a handful of greens in the residual fat in the seventh or eighth minute. I will set a small sauce on the pan in the twelfth minute, where it will hold warm without reducing. None of these tasks need the burner. They need the memory in the pan, and the pan is happy to provide it.

This matters because the most common failure of cooking on a heavy pan at home is the one where the cook never trusts the residual heat. The cook sears the steak, turns the burner off, sets the steak to rest, and then turns the burner back on under the same pan to wilt some greens, on the theory that wilting greens requires heat and the pan was hot a minute ago but might be cooling. The greens go on. The pan, which already had ample residual heat to wilt them, now has ample heat plus a flame underneath it, and the greens are scorched at the edges within a minute. Or the cook leaves the pan on the burner because the burner is convenient, even after the burner is off, and the next thing the cook does is reach for the pan with a bare hand because the cook has, unconsciously, decided the pan must have cooled by now if the burner has been off for five minutes. The pan has not cooled. The cook gets a bad burn on the palm. The cast iron is doing exactly what cast iron does, and the cook is failing to plan around it. The cook who learns the flywheel — who gets the pan to temperature, does the longest work on it, then runs the residual heat through three or four more small tasks with the burner off — is the cook who uses less gas, ends up with less burnt food, and avoids the kind of palm burn that takes a week to heal.

I should name the caveats. The thermal memory is not infinite. A small cast-iron pan — a six-inch skillet — has noticeably less mass and gives back its heat in maybe seven or eight minutes of useful work; a twelve-inch pan can carry twenty. A thinner cast-iron pan, of the kind some modern brands have made to compete with carbon steel, is closer to carbon steel in behaviour and the memory is shorter than the cook expects; the cook who is used to a heavy old pan will misjudge the new one for the first few weeks. The starting temperature matters. If the pan was at full sear heat — smoking, almost too hot for the meat — the residual will run hotter and longer than if the pan was at a medium roast heat. The pan in the open air gives off heat more slowly than a pan with food on it; the moment the cook puts something cold on the surface, the temperature drops faster, because the food is taking the heat where the air was not. None of these break the principle, which is that the pan remembers, the cook plans, and the burner can usually come off earlier than the cook expects.

I am at minute six now. The tomato is blistered. The bread is toasted. The greens have just gone into the pan in the last of the rendered fat and are wilting at the same rate they would have on a low flame. The pan is still warm under my hand at a respectful distance. The steak is rested. I will lift everything off, slice the steak across the grain, plate it with the tomato and the greens beside it and a piece of bread underneath, and the pan will sit there for another fifteen minutes, slowly forgetting the heat that did the work, before I rinse it and dry it and oil it and put it away. The flywheel will have wound down by then. The next time I light a burner under it, I will start the spin again.