I was surfing around moviemistakes.com just now (a good way to waste some time 
) and found this quote on the X-Men 2 movie:
OK, all cool. But then a correction from another user:
I don’t get it. 
Is (s)he right?
It depends on the ice you’re forming, but under normal atmospheric conditions water has a maximum density at around 4 C. From that point, it will expand either on heating or cooling (to a point).
Water does expand when it’s frozen. The way the molecules settle together actually has a greater volume than when it’s in liquid state; that’s why ice floats. It will also expand if you heat it, since you’re imparting more energy to the molecules and they’re bouncing around more, creating a larger volume.
As for the movie, I really can’t say. If I remember correctly, he freezes the cup as well, and it happens really quickly. So if anything, I’m surprised the cup didn’t break just from the stress induced by going from hot to cold in such a short time.
In simple terms, everything expands when you heat it and contracts when you cool it. However, water is a special case that basically does that but in addition…when it freezes its molecules form a crystal structure that takes up more room than the liquid water did a few degrees above freezing. The expansion from this crystallization effect overrides the normal contraction from its getting colder.
As mentioned, water, like pretty much anything else, expands when heated. Of course, it also expands when it freezes, I think the minimum point (max density)is just over 0 degrees C. Now, the teacher is dead wrong in thinking that hot tea might be the same density as ice. This is trivially tested by putting an icecube in boiling hot water.
<goes to boil some water>
Tested… ice floats in boiling hot water, therefore, ice should take up more space than hot tea.
Gee, didn’t need to break out those pesky coefficients of expansion to figure this out. Could have relied on personal experience, but a nice fresh test is always worth it.
Doesn’t water expand due to gases in solution? Can we make a water with nothing in solution that wouldn’t expand?
Well, to be fair, most ice cubes contain a significant amount of voids (frozen air bubbles) that make it less dense still, but yes, water ice (and liquid water below 4[sup]o[/sup]C) expands rather than contracts. This is unusual, and for us, very fortunate, because it means that frozen water (ice) stays on the top of a pond, insulating the thermal mass of the water and creating convection currents that probably had a fair amount to do with life being created and maintained waybackwhen.
Water behaves this way because of the dipole nature of the H[sub]2[/sub]O molecule, which changes its angle from about 104.5[sup]o[/sup] when fluid (liquid or gas) to around 108[sup]o[/sup] when solid. The reasons it does this are very complex and beyond my grasp of electrochemistry to explain clearly, but this is what causes water ice to float, and what leads to the hexigram shape of water crystals, as explained here.
I believe water is the only “simple” molecule that does this, although I’m morally certain that many of the complex organic polymers show these kinds of properties. Perhaps some chemist Doper will come along and clarify/correct/vilify my explaination. There’s quite a good book about the interesting properties and history of water; Life’s Matrix: A Biography of Water by Philip Ball, that you might want to check out if you are interested on the topic.
Stranger
And the great majority of the expansion comes in the transition from liquid at 0 C to solid at 0 C.
My source gives the specific gravity of water at 4 C as 1.0000; at 0 C 0.9999 (liquid); and at 0 C (solid) 0.9150
To answer your question, no, you couldn’t make water that doesn’t expand when heated. Say you have a glass of pure H[sub]2[/sub]O. Each of those molecules are bouncing around a bit. As you heat up the water, you’re imparting more energy to those molecules. They start to bounce around with more speed, thus widening the area around each individual. Think of a 250 lb football player at a folk show. Maybe he’s swaying side to side a little, but on the whole, the people can pack in pretty tight there. Now put him at a Tool show. The harder he starts to mosh, the more space is going to clear around him. So the overall “volume” of the crowd expands, just as the overall volume of the water expands.
If there are gases dissolved in the water, like CO[sub]2[/sub] or something, those will also expand and probably cause the overall volume to increase, but the precise effect is complicated due to the interactions of two different molecules.
Elemental gallium displays the behavior of expanding on transition from liquid to solid. Since the freezing point is just above room temperature, this is actually a consideration in the storage of the stuff.
One doesn’t necessarily conclude the other. Another rationale is that the force of rising air bubbles in boiling water is great enough to overcome the weight of an ice cube, therefore keeping it afloat.
Okay, I knew that the crystals in frozen water take up more room than liquid water does. I didn’t know they were “approximately tetrahedral crystals that give the characteristic hexagonal shape to ice crystals,” until I looked at Stranger’s link. However, one of my nephews once asked me a question about those expanded crystals that I still can’t answer. He wanted to know. "What fills those teensy spaces in the crystals?"
Electric and magnetic fields?
The water was just off the boil, rather than currently boiling, just like a fresh cup of hot tea 5 seconds before the teabag is added.
We recently had a very entertaining discussion about this, re: Why can’t my hand go through my desk?. I think the most relevent posts to your question are here and here.
Short Answer: it’s all space that’s filled up with fields.
Long Answer: Start with Introduction To Modern Physics and Introductory Electrochemistry and go on from there.
Stranger
Hmmm… sorry to have induced you into making such an obvious answer. I meant freezing the water, since I imagined that we were all in general agreement that virtually everything expands when heated. My bad.
I don’t think I understand your question.
Water contracts, slightly, between 0[sup]o[/sup] and (roughly) 4[sup]o[/sup] at standard atmospheric pressure, and then goes on to behave like a perfectly normal substance after that. This is due to its peculiar molecular configuration, and has nothing to do with gases in solution (though gases in solution will resolve out as water transitions into solid state, resulting in the voids that makes most ice translucent. Ice without inclusions or bubbles is clear (slightly blueish, actually), but is still buoyant in liquid water.
Stranger
Those aren’t air bubbles. Those are water vapor bubbles.
To be fair, we don’t know just how cold that tea-ice was. After it freezes, ice, like most substances, will contract with lowering temperatures. So is there any sufficiently low temperature (at standard pressure) where ice would have the same density as hot water?