Why Steam Cannot Get Hotter Than 100 Degrees

This page contains affiliate links. If you buy through one we may earn a commission, at no cost to you.

Because there is no other way to make steam hotter than 100 degrees, and a steamer that cannot get hotter than 100 degrees has no settings at all. The question treats pressure as something the appliance adds on top of steaming. It is not an addition. Put water in a sealed vessel and heat it, and the steam has nowhere to go, so the pressure rises on its own. The only decision available to the designer is where to let the valve open, and that decision is the entire temperature control.

At zero gauge pressure, steam is at 100.0 degrees. At 11.6 pounds per square inch, the top of the usual electric range, it is 117.3. That 17.3 degrees is worth between three and ten times the cooking rate, and it is the whole reason the lid locks.

Water and steam only coexist along one line

This is the fact the question runs into. Liquid water and its vapour can sit together at one temperature for any given pressure and no other, and the relationship between the two is fixed by physics rather than by design. It is the saturation curve, and you can compute any point on it from the latent heat of vaporisation.

Anchoring at 100 degrees and 101.325 kilopascals, with a latent heat of 2,257 kilojoules per kilogram, the Clausius Clapeyron relation gives 115.6 degrees at 10.2 pounds per square inch of gauge pressure, 117.3 degrees at 11.6 pounds, and 121.2 degrees at 15. Those agree with published steam tables to within a tenth of a degree.

So an open steamer basket over a pan is a 100 degree device. Not approximately: an exactly 100 degree device, because the steam cannot be anything else while liquid water is present and the lid is off. Turning the burner up makes more steam at 100 degrees. There is no dial. A sealed vessel is the only way to get a steamer with a temperature setting, and the setting is spelled in pounds per square inch.

Notice the shape of the curve as well, because it explains where the ceiling comes from. Each further degree needs more pressure than the last, and the vessel has to get heavier to hold it. That is why domestic cookers stop somewhere near 15 pounds rather than carrying on: the metal mass climbs faster than the temperature does, and the last few degrees are not worth what they take.

What 17 degrees is worth

Chemistry does not care much about temperature in a linear way. Reaction rates follow an Arrhenius dependence, exponential in inverse temperature, which in food work is usually written as a Q10: the factor by which a rate multiplies for every ten kelvin. For the reactions that matter in cooking, starch gelatinisation, collagen hydrolysis, pectin breakdown, Q10 runs somewhere between 2 and 4.

Run the 17.3 kelvin rise through that. A Q10 of 2 gives a rate multiplier of 3.25. A Q10 of 3 gives 6.47. A Q10 of 4 gives 10.56. At the 21.2 kelvin rise available from a 15 pound stovetop pot, the same three assumptions give 4.29, 10.05 and 18.38.

Which makes a claim in one of our own listings worth checking, because figures like it are usually indefensible. A Presto 01365 6 quart stainless steel pressure cooker states that it cooks meals "3 to 10 times faster than traditional methods". Set against the band above, that is inside the range and, for a stovetop pot, at the conservative end of it. Published multipliers are usually the weakest sentence in a listing. This one is the arithmetic.

The same Arrhenius pattern turns up elsewhere on this site in a completely different field. Our page on what a 100,000 hour LED rating actually promises works with the same law for lumen depreciation, where the chemistry is a phosphor degrading rather than a collagen unwinding. Same exponential, same habit of being written as a factor per ten degrees, opposite desired outcome.

The second reason the lid locks, which nobody mentions

Temperature is the headline, and there is a quieter reason that matters for an appliance running on a kitchen socket. Steam carries its latent heat with it. Every kilogram that leaves a pot takes 2,257 kilojoules out of the building.

Put a number on that. An electric multicooker's element delivers roughly a kilowatt. If all of it went into evaporation, that is 1,000 divided by 2,257,000, which is 0.443 grams of steam a second, or 1.60 kilograms an hour. An unsealed steamer is therefore a machine that boils a litre and a half of water away every hour and runs its element flat out to do it, because the energy is leaving as fast as it arrives.

Seal it and the steam stays. Once the vessel is at temperature the element has nothing to supply but the heat leaking through the walls, which on an insulated body is perhaps 50 watts. So sealing drops the steady draw by something like a factor of twenty, and that is the difference between an appliance that can hold a long programme on a 120 volt socket and one that cannot.

A pressure cooker is the one appliance that reaches the conduction floor

Here is the part worth knowing if you have read our page on why an air fryer browns the crust before the middle cooks. That page works out that every convection appliance is fighting a floor set by conduction through the food, and that no fan or wattage gets past it. A pressure cooker is the machine that actually arrives there.

The reason is condensation. When steam condenses on a cold surface it gives up all 2,257 kilojoules per kilogram at the surface, which makes the heat transfer coefficient enormous, in the region of 5,000 to 10,000 watts per square metre kelvin against the 70 of an air fryer. For a 40 millimetre potato with a conductivity of 0.55, that puts the Biot number between 182 and 364. Effectively infinite. All of the resistance is inside the food and none of it is at the surface.

Which has two consequences. The first is uniformity: condensing steam is isothermal, so every exposed surface in the vessel is at exactly the saturation temperature and there are no hot spots to manage. The second is that the remaining time is pure conduction, and conduction is a square law.

So for that 40 millimetre potato, with a thermal diffusivity of 1.4 times ten to the minus seven, the centre reaches 96 degrees in 34.0 minutes at 117.3 degrees and 62.5 minutes at 100. The pressure buys 1.84 times on the time to temperature, and then separately buys 3 to 10 times on the rate of everything that happens once it is there. Two different mechanisms, often conflated into one claim.

The minutes that are not in the programme

Every electric pressure cooker displays a cooking time, and that time begins when pressure is reached and ends when the programme ends. Both ends have a real duration and one listing in our catalogue is straight about it.

A Ninja HyperHeat 9-in-1 pressure cooker says in its own drawbacks that "Rice and grain times exclude pressure build/release". That is an honest and unusual disclosure, and it is worth computing what it excludes.

Building pressure means heating the contents and the pot from room temperature to 117 degrees. Take 1.5 litres of water, which needs 1.5 times 4,186 times 97, or 609 kilojoules, plus about a kilogram of stainless inner pot at 48.5 kilojoules, plus around 10 kilojoules to fill the headspace with steam. Call it 667 kilojoules. At a kilowatt that is 11.1 minutes.

And it scales with the element, which gives the one wattage in our catalogue something to do. An Instant Pot Duo 7-in-1 Mini publishes, among its drawbacks, that its "700-watt power may heat slower than higher-watt models". Divide the same 667 kilojoules by 700 watts and it is 15.9 minutes rather than 11.1. The drawback is correct, and in a 3 quart pot with less water in it the gap narrows again, which is the sort of thing a published figure lets you work out and an unpublished one does not.

The other end is longer and more consequential. Natural release means waiting for the vessel to cool from 117 degrees back to 100 so the lid will unlock, and that heat has to leave through the walls. The thermal mass is about 6.78 kilojoules per kelvin for the water and pot together, so 17.3 kelvin is 117 kilojoules, and at 50 watts of loss that takes 2,346 seconds. Thirty nine minutes, during which the food is sitting somewhere between 117 and 100 degrees and still cooking at between 1 and 10 times the rate it would at room temperature.

Quick release moves the identical 117 kilojoules out as steam instead of through the walls. At 2,257 kilojoules per kilogram that is 51.9 grams of steam, which at any plausible vent flow is about a minute. Same energy, same temperature drop, two routes, and a difference of thirty eight minutes in how long the food stays hot.

That is a genuinely useful thing to hold onto. The release method is not a convenience setting, it is a second cooking stage of about forty minutes that you either include or skip, and most recipes that specify natural release are relying on it without saying so.

The number the whole category does not publish

Everything above hangs on the operating pressure, and not one pressure cooker listing on this site states one. Not in pounds per square inch, not in kilopascals, not as a pair of high and low figures. The programmes are named after foods instead.

One of our own records says so, which is more useful than any of the others. A Cosori 6 quart pressure cooker lists among its drawbacks: "No explicit mention of pressure levels or wattage; check listing for details." That line is correct and it is the right complaint, because pressure is not a secondary specification on this appliance. It is the temperature, and the temperature is the product. A cooker at 10.2 pounds and one at 15 differ by 5.6 degrees, which on a Q10 of 3 is a rate ratio of 1.7, and no buyer can see which one they are getting.

Our electric pressure cooker coverage describes the programmes and the release step carefully and does not state a pressure either, because there is nothing to state it from. That is worth fixing upstream rather than arguing about here, and in the meantime the honest position is that a stated cooking time without a stated pressure is not a comparable figure.

The honest limit, and the free lever

Two soft spots. The Q10 band of 2 to 4 is wide, and it is wide because different reactions in the same pot have different activation energies: the collagen in a joint and the starch in a potato do not accelerate at the same rate, so a single multiplier is a convenience rather than a fact. And the 50 watts of wall loss behind the 39 minute natural release figure is an estimate for an insulated electric body, not a published number, so read that one as tens of minutes rather than as 39.

The saturation temperatures are not soft. They are a physical relationship, they agree with steam tables, and they are the part of this page that does not move.

Which leaves one lever, and it is free. Since the vessel is already at the conduction floor, the only remaining variable is the size of the pieces, and conduction is a square law. That 40 millimetre potato takes 34.0 minutes. Cut to 20 millimetres it takes 8.5. Four times faster for nothing, from a knife you already own, and no appliance in this category offers an improvement remotely that large. Our moka pot page makes the same point from the other direction about a device where the steam pressure is doing a mechanical job rather than a thermal one, and the same relation sets both.

And the second free move: decide the release method deliberately. If the food was done when the timer ended, vent it, because natural release is another forty minutes of cooking. If it needed longer, leave it, and stop adding minutes to the programme to compensate for a stage you were already getting.