Why ice is slippery

Step onto a frozen path in flat-soled shoes and you are down before you have decided anything. You didn't press hard, you didn't skate, you barely shifted your weight — and still the ground was already waiting to slide out from under your foot. That already is the whole puzzle. Most other slippery things you have to make slippery: you wet the floor, you oil the pan. Ice arrives lubricated. And the reason most of us were handed for it in school turns out to be, if not quite wrong, then almost beside the point.

The schoolbook story is about pressure, and it leans on a genuinely strange fact about ice. Water is one of the very few substances that expands when it freezes — ice is less dense than the liquid it came from, which is why ice floats and why a forgotten bottle in the freezer splits. A consequence of that expansion is that squeezing ice nudges it back toward the liquid: pile on enough pressure and the melting point drops below zero, so ice that was solid at a given cold temperature can melt under the load. Put a skater's whole weight on the thin steel edge of a blade, the story goes, and the pressure under that edge is enormous. It melts a microscopic film of water; the skater glides on the film; when the blade moves on, the water refreezes. It is a tidy picture, and it quietly flatters us — we melt the ice ourselves, just by bearing down on it.

Then you run the numbers and the picture falls apart. It takes a great deal of pressure to move ice's melting point by even a little: roughly a hundred and thirty atmospheres to lower it a single degree. A sharp skate under a heavy person does generate high pressure, but not that high — the best estimates buy you a fraction of a degree, perhaps a degree or two if you are generous with the arithmetic. Meanwhile outdoor ice at thirty below zero is perfectly skateable, and no plausible weight on any blade will melt ice that cold by pressure alone. Worse for the theory, you don't need a blade at all. A person standing still in wide, flat boots — spreading their weight over a hundred times the area, at a hundredth the pressure — slips just as readily. And a hockey puck, which barely presses on the ice at all, slides for the length of the rink. Whatever makes ice slippery is mostly already present before any weight arrives, and pressure-melting is too weak, and too dependent on a sharp edge, to be it.

There is a second, better answer that handles the moving cases: friction. Drag anything across ice fast enough and the rubbing generates heat, and that heat, not the pressure, melts a thin lubricating layer on the fly. This is real, and for a sprinting skater or a bobsled it does much of the work; it is why the ice under a fast blade is measurably wetter than the ice beside it. But friction-melting has the same blind spot as pressure. It needs motion. It cannot explain the ice that is slippery the instant your still foot touches it, before you have rubbed anything against anything. The slipperiness that dumps you on the path is there at rest.

Which leaves the answer that was actually proposed first, long before the other two were ruled out, and has quietly outlived them both. In 1850, Michael Faraday guessed that the surface of ice is not truly solid — that ice wears a permanent skin of water, a film only molecules thick, present even when the ice is well below freezing and touched by nothing. His evidence was a homely experiment anyone can repeat: press two ice cubes together and hold them a moment, and they weld into one. Faraday's explanation was that each cube carries its liquid skin, but when the two skins are pressed into the interior of the new, joined block, they are no longer at a surface — so they freeze, gluing the cubes together. He was arguing, against colleagues who backed the pressure theory, that the film comes first and the freezing is what needs explaining, not the melting.

He was right, and it took modern surface science more than a century to prove it and to say why. At the surface of the crystal, the outermost water molecules have neighbors below them but open air above, so they are bonded on one side only. Held that loosely, they cannot lock into the rigid lattice the way molecules deep inside can; they jostle and slip in a disordered, liquid-like layer that sits on top of the solid ice with no pressure and no rubbing required. This quasi-liquid layer has been measured directly now — it is real, it is there far below zero, and it grows thicker as the temperature climbs toward the melting point, which is exactly why ice gets more treacherous, not less, on a day that is merely a little cold rather than bitterly so. Faraday's welded cubes were the first sight of it.

So the honest answer to why ice is slippery is that it is all three, in their seasons — a surface already wet on its own before you arrive, thickened by any pressure you bring and melted further by any motion you add — but the one nobody can take away, the one that is there when you are standing still on a windless day at twenty below, is the skin the ice cannot help growing. And it is worth saying plainly that this is not a closed question. Researchers are still arguing about how that surface layer really behaves under a sliding foot — whether it acts like ordinary water or like something stranger, its weakly-held molecules rolling and tumbling more like tiny ball bearings than like a puddle. The thing under your feet that seems most simple and settled is, at the scale where the slipping actually happens, still partly a mystery.

What is settled is the reversal. We picture the slipperiness as something we do to the ice — our weight, our motion, melting a path. Mostly we do nothing of the kind. The water is there ahead of us, a film the ice grows because a surface is a place where the crystal runs out of ways to hold itself together. We don't make the ice slippery. We find it that way, and fall.