I’ve been growing salt crystals. Unlike a pet, salt crystals won’t tear up the furniture, do their business on the carpet or need to be walked in the rain. Really I was testing to see if I could do this as a time-lapse project and wanted to test how long it would take. And it has fulfilled Hofstadter’s law: it always takes longer than you expect, even when you take into account Hofstadter’s law.
I took the standard approach of heating some water and dissolving a bunch of (uniodized) salt in it, letting it cool and pouring it into a beaker. And I waited. And waited. Finally, after a few weeks:
One thing I should have anticipated is how long it takes. There was some salt left in the pot when I poured the solution into the beaker, so I thought the crystallizing would begin quickly, but it didn’t. Plus, the evaporation was slow. I knew that boiling point elevation and freezing point depression are colligative properties (they depend on the number of dissolved atoms) so I reasoned that evaporation rate should be as well. And it is — there is Raoult’s law
The vapour pressure of an ideal solution is dependent on the vapour pressure of each chemical component and the mole fraction of the component present in the solution.
The vapor pressure of salt is very low, so as its concentration rises the total vapor pressure of the solution drops, and so does the evaporation rate. Rather than evaporating fully, one would expect it to reach an equilibrium with the atmosphere which would depend on the humidity. In fact, if the salt concentration were high enough, one might expect it to dehumidify the air, which is precisely what some people do. Salt concentrations are used in dehumidifiers — you expose the solution to the air and let is “grab” some water, then heat it up (often solar, for a completely passive system) to let the excess water evaporate, and cool it again in a cycle. Or you can have a solution with some solute left in the container, and as you “grab” the water, you dissolve more of the salt, so it can continue doing its job as long as there is more salt that can dissolve.
Another unexpected event in all of this is that I was getting salt crystallizing on the surface of the water. A small “raft” would float there until it grew massive enough that it would sink (or someone poked it). I had thought the crystallization would just build on any crystal that started up, but there are lots of small cubes rather than just a few large ones. Still, the biggest cubes are perhaps 10-20x larger on a side than the original grains.
Hmm… Now I want to mess around with this myself. I grow (and study the optical properties of) chloride crystals for a living, but I’m always growing grouchy, unstable, wildly anisotropic ternary compounds rather than friendly little binaries like NaCl. Plus, I always do melt growth in ampoules, so my final crystals are rather boringly cylindrical. This is bound to be a lot more satisfying and less frustrating.
One thing to try would be to filter or decant away from the undissolved salt before cooling starts, then suspend a wire (people use platinum in research labs, but I bet copper would work) cut with a pointy end into the solution. I’d bet that your growth will nucleate on the sharp end of the wire, which may lead to a large single crystal. There’s probably a limit to how large a cube you could produce this way, since, without temperature control and the like, your growth will go dendritic at some point, but it would be interesting to see how far it goes.
Also, moving the beaker into the fridge will likely cause additional growth.
And as I think about it, I wonder if messing with the solvent could give better results. Perhaps adding a little bit of vodka every day, since salt is going to be a lot less soluble in ethanol?
Okay, it looks like I have a project to start…
Suspending a string for nucleation was something I considered and will do on my next attempt; if wire works better I’ll try that.
Using the fridge would would make filming problematic. I have to get a power supply, since battery power runs out too quickly.
http://scienceprojectideasforkids.com/wp-content/uploads/2010/05/Solubility-Curve-Sucrose-Sodium-Chloride.jpg
http://www.chemguide.co.uk/physical/phaseeqia/solcurve1.gif
http://dwb4.unl.edu/CHEM/CHEM869A/CHEM869AMats/SaltSol_files/SaltSolT.gif
http://www.docbrown.info/ks3chemistry/gifs/solubilityKS3.gif
http://en.wikipedia.org/wiki/Solubility_table
Grow crystals from water starting with a clean seed spot epoxied to the end of a hair (2 lb fishing line or pulled glass fiber if you are an optimist). Set up a solubility cycle with a tipped beaker mostly filled with water saturated with solute, and always with with excess solute at the low corner. Gently heat that corner to plume a warm more concentrated flow upwards. It cools and sinks, round and round. After equilibration, dip the seed in water to dissolve off the surface, then immediately place in the descending supersaturated plume. Give it a few days of circulation. Solute differential solubility with temp is now a moot point. Potassium ferricyanide (red, polyfaceted, photography supplies) is very nice!
If you have high standards, spot glue the distal end of the hair to the hub of a sacrificed battery-operated clock, mounted upside-down horizontal, so the seed is constantly turning in the descending plume.
If you silanize your glassware with Rain-X (for automobile windshields), you get no surface nucleation. Spray it on, swish. Suddenly the Rain-X no longer wets the glass. Pour out, rinse with rubbing alcohol, drain, dry at 80 C for 15 minutes.
http://faculty.washington.edu/annkurth/dyeingcrystals.html
Because the universe is an interesting place.
http://books.google.com/books/about/Crystal_Growth_in_Gels.html?id=pQ-JuNXEL6wC
So it’s insoluble – reactively grow it.
What Uncle Al is describing is hydrothermal growth, which is used industrially to produce single crystals of some substances; I didn’t know enough about it to describe a simple setup, but that sounds very doable.