Water expands as it freezes. If we placed it in a container made of hypothetical steel where the water has zero room to expand at its freezing point, would it still freeze?
In other words, a perfect gallon of water is in a perfect gallon space where it can’t expand to its 1.09 gallon volume of ice as it freezes. Does the water still freeze? Do we end up with some exotic form of ice?
Can’t think of any way this could happen in nature (maybe at the center of a Jupiter sized mass? How much pressure does water exert as it turns into ice?) much less a lab, but just a thought popping in my head as I froze a bottle of water yesterday.
You may want to take a look at the phase diagram of water. That expansion you refer to is for “normal” ice (Ih in the diagram), and it’s at the triple point. If your container is strong enough to prevent water from expanding we’re talking about higher pressures: for some values of pressure and 0ºC you’ll have liquid water, but make the pressure high enough and you will get different forms of ice - you can even get ice at temperatures much higher than 100ºC, but it will be, in your terminology, “exotic forms”.
That picture comes from lab studies. No perfect gallons of water, but there’s an enormous amount of work that has been and keeps being done on the physics and chemistry of water.
There are different forms of ice with different crystal structures. I imagine the water would form one of these. Would the freezing point change in this case?
If you examine the phase diagram referenced above, you see that there is a slight slope to the “freezing” line: the freezing temperature changes with pressure. The crystal structure isn’t dramatically different: that would cause a sharp angle or a split (two different phases).
The slight slope leaves open the possiblity that “ice” consists of a mixture of two distinct crystal structures, with the proportions changing with preasure and temperature. …But I don’t think that ice has a very regular crystal structure anyway, I think it just packs in at higher pressures and lower temperatures.
OP: note that the phase diagram lines do not extend to infinity, or to the margins. There is a point at which high-pressure low-temperature water is indistinguishable from ice. It’s not possible to say that it’s “frozen”, or just “cold and under pressure”.
Practical real-life effect: If you put a (plastic!) bottle of water in the freezer, it will freeze…but there will be a particular moment when it is still liquid, but below freezing temperature. It is kept liquid because it’s under pressure.
If you open the bottle…it will suddenly, and with a lovely crackling sound, freeze right up! This is because the pressure has been released.
A plastic water bottle doesn’t remain liquid because of the pressure. It remains liquid because it hasn’t had a chance to nucleate and start freezing. Opening it is likely to nucleate it, but then, so is just shaking it.
I believe it needs to be distilled. I freeze bottles at night for work the next day (spring water with minerals galore), and they’re always solid when I open the freezer door the next day.
I suspect your bottles aren’t sealed perfectly, and allow the pressure out. If it is not allowing the pressure out, then I have to wonder why your bottles aren’t cracking from that pressure.
No, distillation is not required. I have personally seen this effect with orange juice, soda, wine, milk, whatever. (Where “this effect” means that it is still liquid when taken out of the freezer, and freezes to slush when the bottle is opened.)
Echoing Chronos, it’s not the pressure that causes bottled water to remain liquid below freezing. It’s because the water is very still and there are few nucleation sites.
Proof: Pick up one of your chilled sub-32F bottles and tap it on the side, then watch as it freezes in your hand, while still sealed.
And it doesn’t crack because water bottles, unlike pop bottles, are usually designed with features like ridges that allow a little expansion (enough to accommodate freezing).
It will still freeze. As the water tries to expand, the pressure increases until it reaches 43,500 psi then it freezes and forms II. Lower temperatures and higher pressures cause the ice II to transform to other phases of ice.
mixdenny: Wow, damn near the exact phrasing I used, guess there really is nothing new on the internet.
I like the note at the very bottom of that article that in the terapascal pressure range ice could become metallic. Can’t even imagine a terapascal (that’s about 145-million psi? for comparison, if xkcd is right, earth’s core is only at 30 million psi).
43,500 psi is several times the pressure at the bottom of the ocean. I guess the resistance water can exert outwards by way of it’s innate electromagnetic structure is why it is considered virtually incompressible.
Can’t it be both? (Or either?) I know this has happened with a can of Coke: I left the can right next to the vent in the fridge. When I popped the tab fluid almost immediately started to flow out the top; after a short time (30 seconds?) most of the Coke left in the can was frozen. The expanding ice pushed the fluid out the top.
A can is more likely to have pressure issues than a plastic water bottle. In fact, we already know it’s slightly pressurized because you hear the hiss when it opens. But the carbonation in a soda also explains pushing fluid out the top a lot better than expansion of freezing water does. I’ve seen this happen myself with a soda that never froze past a slush phase, and too much of the contents were displaced for freezing to be the explanation.
Besides all that, when you look at that phase diagram, pressure has a pretty small change on the temperature of freezing. To change the freezing temperature significantly would be enough pressure to rupture the can.
If someone has a cite explaining what’s happening in the can I’ll concede but otherwise I disagree. The Coke was coming out the top too slowly (IMO) for carbonation to be the explanation. Given that there was a large chunk of ice left in the can it seems obvious that its formation was the reason for the overflow.
From the original post, water expands by 9% as it freezes. Keeping in mind that soda cans are not 100% full to begin with, does that explain the amount you saw pushed up?
As for speed, the slushy soda I’m talking about was pushing up half-frozen slush for more than a minute.
I don’t know about a cite, but one possibility is that the chunk that was frozen served to provide a million nucleation sites on the jagged crystal edges for the CO2 to come out of suspension as a bubble and foam out of the top of the can. If you dropped something in there after you opened it, you’ll get some fizz. If you drop something with a ton of nucleation sites (say a pinch of some powder or granular substance) you might get enough to get a thick foam going.
SOURCE: I homebrew beer…you do NOT add hops or sugar or anything else to an already-carbonated keg of beer. EDIT: You can, actually, just be prepared…it will likely foam. I was thinking of a different situation that’s actually dangerous. But people do this all the time…but sometimes they’re surprised if they’re not expecting the result…
So, if you had this steel container, and made it really cold, and ended up with this different-kind-of-ice, then opened up the container, what would happen?
(a) in super-cold air?
(b) in room temperature air?
If you do this experiment, and the steel box is completely sealed, how would you know what happened inside it? If there was a window built in, how could you even tell the water had changed to ice?