And, the shrinkage has nothing to do with electron orbits shrinking.
It has to do with the motions of the molecules themselves - they don’t move as fast when they are cold (pretty much the definition of “temperature”), and so there is less pressure trying to expand a mass of water. When the water starts to crystalize, it forms crystals that are just slightly less compact than the water around them.
This is the key here. Water expands when it gets hot and contracts when it gets cold, just like most other things. When it actually freezes though, it forms crystals, and those crystals aren’t as compact as the loose water molecules are. That’s why freezing water expands and breaks street surfaces, forming potholes, and why glass bottles of water will shatter in the freezer. It’s also why ice floats on water.
Water isn’t unique in this, either. Plenty of other materials, like silicon, for example, will also form crystals which are less densely packed together than the material in its liquid form.
First of all … liquid water does shrink as it cools until it reaches 4ºC at standard pressure … then it expands as it cools to 0ºC … indeed an exotic property of a rather mundane substance … liquid water molecules at 0ºC can actually get closer than in ice, once water freezes the molecule must align themselves in a crystal structure which requires the molecules to be further apart … thus water greatly expands as it freezes … again another exotic property of a rather mundane substance …
That’s not how potholes form … if we don’t compact our gravel base properly, then voids begin to form between the gravel and asphalt … which crumbles the asphalt and then the proto-pothole fills with water … the next tire that rolls over the place splashes the water out PLUS pieces of gravel and asphalt … each time making the pothole bigger … perhaps you’re confusing this with the weathering action of water freezing/thawing and the above mentioned expansion as water freezes …
The distance electrons orbit the nucleus doesn’t change in the way you seem to think they do … instead it is the speed at which they move around that set orbit that changes with temperature … also how much the molecule vibrates in space increase with temperature (and here I’m using the “classical” definition of temperature) … the more the molecule vibrates, the more space it creates for itself within the fluid medium … meaning higher temperatures allow the fluid to expand …
If we can think of a water molecule as an oxygen atom with two bare protons sticking out of it … we’ll have a little proton/proton repulsion going on … 4ºC is the point that the vibrations are overcoming this repulsion the most … less vibration (hence less temperature) allows the repulsion to come back into play and start expanding the fluid … yes, even though temperature is going down …
Counter-intuitive? … sure … water has a lot of exotic properties and it’s fair to say it is an exotic substance … it just happens to be rather common is all …
Water is weird stuff. Attach some carbon to those molecules and it gets about 25% less dense even though each molecule is heavier. Dissolve salt in water instead and it’s density increases. It’s witchcraft I tells ya!
But you guys haven’t said why it expands in the last 4 C before freezing, rather than just at freezing. Or at least didn’t make it clear. The answer is that it starts to form the bonds that make it a crystal about at that temperature. Here is a good expanation for it.
Ice would form, but it would sink to the bottom of the lake. Ice floating on the top of lakes insulates the water from the colder air above it in winter. Without that insulation, lakes would freeze from the bottom up, and eventually would freeze solid. In many climates, the duration and temperature of summer would only melt a few inches at the top of the lake and in winter it would again freeze completely. This would make aquatic life impossible, so it’s a good thing for us that ice is less dense than liquid water.
IIRC, water stats to expand at about 4C and ice stops expanding somewhere around -20C. One suggestion is this is why ice is slippery - pressure, especially for concentrated so high pressure like a narrow skate blade, can melt a thin layer of the ice underneath so the skate or whatever is riding on a thin layer of water between the blade and ice. Once you get below 20C (or about 0F) ice is nowhere near as slippery.
Potholes, and natural erosion, can happen because liquid water gets into cracks in asphalt or rock. Then it freezes -and expands, expanding the crack it’s in. Rinse and repeat so the cracks expand and get longer and deeper. Each freeze-thaw cycle even more water means an even bigger crack. In areas with frequent freeze-thaw cycles, this can be particularly serious. For potholes, the pounding of traffic and the water seeping through to underneath and softening the road base under the gravel help deteriorate the road.
Another phenomenon is how rocks come to the surface every spring, a common problem in places like Ireland or New England - fields had to be cleaned of new rocks every spring (hence the quaint rock fences around the fields). Water pools uner the rocks, then freezes forcing them upward (path of least resistance). When melt happens, accumulations of maller dirt run under the rock and hold it up before the ice completely melts. As a result, the rocks work their way up in the soil over time.
Thought experiment: what if I took a perfectly rigid container, filled it completely with liquid water at 4 °C, screwed the lid on tight, and then stuck it in a deep freezer? Would the water freeze?
When I was a kid, there used to be doorstep milk delivery. ('member dat?) I recall a few times where the milk was left on the doorstep in freezing temperatures and you’d see a column of frozen milk pushed out the mouth of a gallon jug, probably an inch or more high.
Basically, pressure goes up until something gives - the container cracks, the top pops off, the container stretches/bulges, etc.
(in the case of the milk jug, you have the same lake-ices-over process working. The coldest liquid milk rises, being less dense below 4C; it freezes in the neck, and is then pushed up as the liquid underneath freezes, and pushes up more. Basically, extruding freezing milk.)
I recall seeing pictures of a heavy steel container that burst open from water freezing for just this purpose. I have no idea how strong a container you would need to resist the force of the expanding water without any deformation. I forget what happens to the freezing point of water under pressure. Here’s a link toPhysics Stack page that might help. Don’t have time to figure it out myself right now (if I even can, it looks all mathy).
The real answer is there is no such thing as a perfectly rigid solid, but for the purposes of your question it’s probably equivalent to consider an extremely strong container, made for example of high strength steel several inches thick. The water would still freeze, but it would be under high pressure, and depending on the exact temperature and pressure, it might be a different phase than normal ice (ice I). Water’s phase diagram shows the different phases of water at various temperatures and pressures.
This theory of why ice skating is possible dates from the 19th century. Although you still sometimes hear it today, it is obsolete and has been debunked. The change in ice’s melting point induced by the weight of a skater is far too small to produce a significant layer of water.
For example, this article in the American Journal of Physics says
