Why a Stainless Pot Bottom Turns Rainbow

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

It is not a residue and nothing has been left behind on it. The rainbow is the steel's own protective film, grown thicker by heat until it became thick enough to interfere with visible light, and the whole thing is somewhere between forty and seventy nanometres deep. On a six quart pot base the entire rainbow weighs about ten milligrams.

That matters practically, because it changes what will take it off and it explains the thing people find strangest: why it appeared after a sauté and not after a month of pressure cooking.

The film has no colour of its own

Every piece of stainless steel is already covered in a chromium rich oxide, one to three nanometres thick, and that film is the entire reason the alloy is called stainless. At that thickness it is invisible, because it is far thinner than any wavelength of light.

Heat it in air and the film thickens. Nothing about it changes except its depth, and at some point the depth starts to matter optically. Light arriving at the surface splits in two: part reflects off the top of the film, and part goes through, bounces off the metal underneath and comes back out. The second part has travelled an extra distance of twice the film thickness, multiplied by the film's refractive index of about 2.5 because light moves more slowly inside it.

When that extra path amounts to half a wavelength, the two reflections cancel and that wavelength goes missing from what reaches your eye. So the first order condition is the thickness equal to the wavelength divided by four times the refractive index, and the arithmetic is a single division:

Remove blue at 450 nanometres and the film is 45 nanometres thick, and the surface looks straw or pale gold. Remove green at 550 and the film is 55, and the surface looks magenta or purple. Remove red at 650 and the film is 65, and the surface looks blue.

Which is the surprise. There is no colour at all until about forty nanometres, the whole visible sequence happens in the next thirty, and past about a hundred nanometres several orders of interference overlap, no wavelength is cleanly removed, and the surface just goes a washed out grey. A pot goes from looking new to looking ruined across a thickness change smaller than one three thousandth of a sheet of paper.

How little metal it took

Work out the mass, because the proportion is the useful thing to carry away.

Chromium oxide has a density of 5.22 grams per cubic centimetre and is 68.4 percent chromium by weight. A film 55 nanometres thick therefore weighs 0.2871 grams per square metre and holds 0.19637 grams of chromium per square metre.

Where did that chromium come from? From the steel directly underneath. A RIZIZMVG oversized 304 stainless steel cutting board is the one item in our catalogue that states its alloy and its thickness in plain prose, "food-grade 304 stainless steel" over a "heavy-duty 14-gauge core", and that lets this arithmetic be anchored rather than assumed. Grade 304 is 18 percent chromium at 8.0 grams per cubic centimetre, which is 1.44 million grams of chromium per cubic metre. Divide 0.19637 by that and the chromium came out of the top 136 nanometres of the steel.

Against a pot base a couple of millimetres thick, 136 nanometres is 68 parts per million. And spread over the 0.0356 square metre base of a six quart inner pot, which is the figure our own page on why a pressure cooker is tall and its pot is not works out from the capacity, the whole rainbow comes to about ten milligrams of oxide.

So the honest description of a tempered pot bottom is that about a seventh of a micron of its surface has rearranged itself into the compound it was already covered in.

Why it appeared after sautéing and not after pressure cooking

This is the part that makes it diagnosable, and it falls out of a temperature that is not free to vary.

While there is water in a sealed cooker the surface is pinned near the saturation temperature, which our page on why steam cannot get hotter than 100 degrees puts at 117 degrees Celsius at the usual operating pressure. That is far too cool to grow a visible oxide; temper colours on stainless want somewhere in the region of 200 to 400 degrees in air, for minutes. You could pressure cook for years and never see a rainbow.

Sauté has no liquid in it. Nothing pins the plate, so the only thing limiting its temperature is how fast the heat can leave, and on an empty base that is slow. An Instant Pot Pro 6 Qt 10-in-1 electric pressure cooker publishes 1200 watts, and spread over that same 0.0356 square metre base the flux is 33.7 kilowatts per square metre. With nothing absorbing it, that base is in temper colour territory within minutes.

Which means the function that produces the mark is named on every one of these records. An Instant Pot Duo 6 Qt 7-in-1 electric pressure cooker has an 18/8 stainless inner pot whose tri-ply bottom, says its own record, "spreads heat evenly and supports searing and sauteing right in the pot", and its performance text adds that "Browning works well on the saute function". Both statements are true and both describe the exact condition that tempers the steel. No record in the category joins the two.

And it slows down, which is why it seems to stabilise

People notice the rainbow after the first few uses and then report that it stops getting worse, which sounds like wishful thinking and is actually the kinetics.

The film grows by oxygen diffusing through the film that already exists, so the thicker it gets the slower it grows. Thickness goes as the square root of time, which means time goes as the square of thickness. Going from 45 nanometres of straw to 65 nanometres of blue is a factor of 1.44 in thickness and therefore 2.09 in time. Getting to 90 nanometres and grey takes 4.0 times as long as the straw took.

So the first short dry run gives you most of the colour you are ever going to get from ordinary use, and the same property that produced the mark, the film blocking its own growth, is the property that stops it. That is also, not coincidentally, how the alloy protects itself in the first place.

Two of our own pages, one half wrong and one right for the wrong reason

Our guide to stainless steel versus nonstick cookware gives this advice: "if you see cloudy white spots or a rainbow tint, a splash of vinegar or a stainless cleaner brings it back". That sentence has two soils and two reagents in it and they do not pair up the way the sentence implies.

Cloudy white spots are mineral deposit, mostly calcium carbonate dried out of the water. Acetic acid dissolves carbonate briskly, so vinegar is the correct reagent and the advice is right.

A rainbow tint is chromium oxide, and chromium oxide is one of the most acid resistant oxides in common use. That resistance is not incidental, it is the whole mechanism by which stainless steel survives a tomato sauce. Asking household vinegar to dissolve it is asking it to perform precisely the attack the alloy was formulated to shrug off. It will not work, and the reason it will not work is the reason you bought the pan.

The other half of that same sentence, "or a stainless cleaner", is correct, and it is correct because those powders are built around oxalic acid rather than acetic. Oxalic acid attacks iron and chromium oxides by binding the metal ion as it comes off, which is a different chemistry from simply donating a proton. Our page on why one cleaner cannot cut grease and dissolve limescale already sets out properly why a reagent for one soil does nothing for another, and this is the same argument on a third pair of soils. Credit where it belongs, and this page will not repeat the chemistry.

Our page on whether you can use metal utensils on stainless steel cookware suggests "a paste of baking soda" for "rainbow-colored discoloration". That one does work, and not for any chemical reason: sodium bicarbonate is an alkali and will not touch chromium oxide either. What it does is abrade. A paste of a mildly hard, finely divided powder removes a fifty nanometre film mechanically, which is the correct approach, and the page arrives at the right answer by the wrong route. Worth knowing which, because it tells you that pressure and a soft cloth matter more than dwell time, and that leaving the paste to sit will achieve nothing at all.

Telling it apart from the thing it is confused with

An iridescent film on stainless is not always temper colour. A thin mineral or protein film dried onto the surface can also produce interference colours, by exactly the same optics in a different material, and the two look similar in a photograph.

They are easy to separate by geography rather than by chemistry. Heat tint appears only where the metal got hot, so on a cooker base it follows the footprint of the element underneath and stops at its edge, usually as a ring or a disc with a defined boundary. A deposit film follows the water instead, so it appears at the old water line, in streaks down the sides, or over the whole wetted interior. If the mark has a sharp circular edge in the middle of the base, it is the steel. If it fades out at a horizontal line, it is not.

And one honest consequence of the thinness: a mark that comes off with light abrasion in a few passes was a deposit. A mark that needs real work was in the metal, and taking it out means removing that top layer of steel, which is why heavy polishing leaves a visibly different finish where you worked.

What the records do and do not say

A Presto 01365 6 quart stainless steel pressure cooker describes its material the way almost every listing does: stainless steel is "known for durability and resistance to staining and corrosion". That is accurate and it is also why a buyer has no way to anticipate this. Heat tint is not a stain, because nothing was deposited, and it is not corrosion, because nothing was lost. It is the protective film working, made visible. A sentence about staining and corrosion cannot warn you about it because it is neither.

The closest any record in our catalogue comes to naming the phenomenon is an All-Clad D3 stainless steel 12 inch frying pan, whose own drawbacks say it "Requires more maintenance to keep shiny". That is the honest version of what owners actually experience with bare stainless at searing temperatures, and it earns the product its place here, because it is the one listing that concedes the surface is a thing you will be managing rather than a thing that stays as supplied. Its performance text is also right that "High-heat methods like searing and sauteing are where the D3 shines", which is true and is the same condition described above.

What no record states, anywhere in the category, is a maximum dry surface temperature. Every cooker and pan publishes a wattage or an oven rating, and the oven rating is for a cavity rather than for a plate with a heating element pressed against it. The figure that decides whether a surface will temper is how hot the metal itself gets with nothing in the pan, and that is not a number anyone publishes, which is why the arithmetic on this page has to go through flux instead.

The honest limit

The optics are solid and the metallurgy is approximate.

The interference condition and the square root growth law are standard and they do not depend on anybody's pan. The refractive index of 2.5, the oxide density of 5.22 and the 68.4 percent chromium are handbook values for pure chromium oxide. Real temper films are not pure chromium oxide: they are mixed iron and chromium oxides whose composition shifts with temperature and time, with more iron in them at the hotter end. So read the 136 nanometres of depleted steel and the ten milligrams of film as the right order of magnitude rather than as measurements. If the film were half chromium oxide and half iron oxide, both figures would shift by a few tens of percent and nothing on this page would change.

The thirty nanometre window is the number to keep. It is the reason this particular cosmetic complaint is so common and so alarming: the material has no intermediate appearance between untouched and lurid, and the step between them is a distance you cannot see, cannot feel, and cannot avoid if you intend to brown anything in a bare steel pan.