MrHomeScientist - 12-4-2013 at 08:07
Here's a (sort of) thought experiment for everyone:
I was talking to a friend recently, and we came up with an idea for a "phase change" fountain. The layout would be something like a standard tiered
garden water fountain, but instead of water it would use a gas (and would need to be placed in a glass box to contain the gas, of course). Each tier
of the fountain would be held at a different temperature. This gas would exit at the top of the fountain, and be cooled by the first tier to condense
into a liquid, which falls down to the second tier as in a normal fountain. The second tier would be further cooled to freeze the liquid into a "snow"
or small crystals, something that could still flow easily. This snow would build up and fall off the second tier down to the bottom. At the bottom,
the solid would be collected and heated back up to transition it back to a gas, then pumped back out the top.
Another possibility would be to do away with the fountain altogether, and just have a tall glass box with a pipe in the middle, and a temperature
gradient across the height of the box. The gas would exit at the top, condense as rain in the middle, and freeze and fall as snow near the bottom.
Simpler conceptually, but probably much harder to pull off practically.
So the question I pose to you, is what substances could be used in such a fountain? Ideally, it would be a chemical that has an extremely narrow
liquid range somewhere near room temperature, to make the design at least vaguely achievable in practice.
=======================
As I was typing this, I thought to myself: Why not just use water? 
Steam out
the top, a cool first tier to condense into cold water, and an extremely cold lower tier to flash freeze into snow (LN2? dry ice?). I think the tricky
part, no matter what substance is used, is freezing it rapidly so it doesn't just solidify into a block.
Even if plain 'ol water turns out to be the quick solution, I'd still like to hear other people's ideas!
Finnnicus - 12-4-2013 at 08:39
Bromine? Fits well, but er... Yeah.
Edit: A suggestion.
-Atomize the chilled liquid with high pressure through a very very cold copper tube (dry ice/metho?) to create 'snow', maybe go one better and put a
grate over the end of the tube.
[Edited on 12-4-2013 by Finnnicus]
12AX7 - 12-4-2013 at 09:22
You want something with a triple point close to STP. Water isn't bad, but needs a reasonable vacuum, which will be challenging to build a glovebox to
hold. The reduced pressure also means snowflakes (if the formed, which they wouldn't, as described; snow is sublimated, not frozen) fall
ballistically rather than gracefully.
The only materials on Wikipedia's list of triple points which are nearby are I2 and UF6. 
Tim