Field notes on things that run themselves
It Floats on What It Is Losing
Flick water into a pan that is merely hot and it spreads, hisses, and is gone. Flick it into a pan that is very hot and it does something else: the drop draws itself into a bead and skates, whole, taking its time. A puddle a few millimetres across, on a plate at 300 °C, takes two to four minutes to go. On a cooler plate the same water is finished in seconds.
That is the first surprising thing, and it is not a trick of perception. Past a certain surface temperature, water on hot metal lasts longer than water on cooler metal. There is a temperature at which a drop dies fastest, and it is not the hottest one available.
What changes is that the drop stops touching the pan.
Underneath it is a gap, and the gap is not a slab of air. It is a shallow central pocket ringed by a thinner annular neck, and at 300 °C that neck is about sixty microns thick — thin enough that Leidenfrost himself, in 1756, held a candle behind a drop on a red-hot iron spoon and saw the light come through. Vapour pours out of the neck sideways at several metres a second. The pressure it develops is tiny, some five thousandths of an atmosphere, and it is enough, because a drop of water does not weigh much.
Now the loop. Every bit of that vapour is water that was the drop a moment ago. The film is not a barrier the drop is hiding behind; the film is the drop, leaving. And water vapour is a poor conductor of heat, so the film that the evaporation builds is precisely what slows the evaporation down. The drop buys its own insulation with its own substance, continuously, and cannot stop buying: let the loop pause and the cushion drains in milliseconds and the water is on the metal.
This makes it an odd relative of the other standing patterns here. A river wave (No. 2) and a lens-shaped cloud (No. 52) are shapes held open by matter passing through them; the shape outlives its contents. A Leidenfrost drop is the contents. It is not fed. It is spending.
The man whose name is on it had no such explanation. Johann Gottlob Leidenfrost heated an iron spoon “well polished and without rust” until it glowed, laid a drop in it, and timed the drop’s life with a pendulum — thirty-nine pages of a 175-page Latin treatise, under a heading that translates as On the Fixation of Water in Diverse Fire. He believed fire was converting the water into earth, and rejected the suggestion that the residue was ordinary airborne dust. That suggestion was Herman Boerhaave’s, who had described the levitating drop twenty-four years earlier and was right about the dust. Leidenfrost was not translated out of Latin until 1966. The name records who looked hardest, not who looked first, and not who understood.
The loop has a size limit. Above roughly a centimetre across, the film under a puddle goes unstable and vapour chimneys punch up through the middle. And it undermines itself from the far end: the drop’s top is evaporating too, into open air, which cools it, which sets up a surface-tension gradient, which drives a slow circulation inside the drop — and the shear of that circulation reaches the underside and squeezes the gap to about half the thickness it would otherwise have. The process holding the drop up thins its own footing.
None of which is the strange part. The strange part is that after two and a half centuries nobody can say where the effect begins. Reported Leidenfrost temperatures for plain water at ordinary pressure run from about 155 °C to over 300 °C — not sloppiness, but real dependence on the metal, its roughness, its cleanliness, the drop’s size, and the air pressure in the room. In 2021 a leading group in the field wrote that it is “still unclear why a minimum temperature is required before the effect manifests itself, what properties affect this temperature, and what physical principles govern it.” The geometry we can model to the micron; the threshold we cannot predict. It matters: quenching steel, and cooling a reactor whose cladding has reached 700 °C, both turn on knowing when the film will stop forming and let water in.
There is a last asymmetry, and it is the one worth keeping. A drop carefully coaxed into levitating below the usual threshold will stay there, apparently stable. Nudge it once so that it touches down, and it never comes back. The conditions that sustain the loop are not the conditions that start it.
A tardigrade’s tun (No. 70) halts completely, and its last act before halting is to manufacture the thing that lets it resume. A drop has no such provision. It has only the film, and the film is never older than the last instant.
One loop I’m watching
Next: a heap of rotting leaves that a bird uses as an incubator, and checks with its mouth. A male malleefowl rakes litter and sand into a mound, and the microbes working through it give off heat — the sun adds more — enough to hold the eggs buried inside near 33 °C, about where a body would have held them. Nobody sits on this nest. The bird spends most of the year opening and closing the mound instead: piling sand on, venting heat before dawn, testing the pile by taking a beakful and reading its temperature, and keeping the eggs within roughly a degree for two months.
Tip: the ← and → arrow keys move between issues.
Prefer feeds? Follow by RSS or JSON Feed.
Want something playable between issues? Thought Toys has its own email list and RSS feed.
New to The Standing Wave? Start here →