I was reading instructables a few weeks ago and I saw this:

http://www.instructables...-using-Peltier-Coolers/

I suddenly had an idea:
would it be possible to use the same method to freeze precipitate spice from naptha at an unheard-of efficiency?
The "Naphtha-Freezinator 2000" was quickly born.

The idea is simple. We use 2 peltier units sequentially to cool down an insulated glass chamber to very low temperatures (-40c if not more).
Ideally, we would place a naptha-filled glass in the cooling chamber, where it will quickly precipitate most if not all the spice dissolved in it, then take it out, pour the naptha and scrape the spice.
The cooling chamber will be partially filled with alcohol to facilitate the heat transfer and have a larger heat mass.

This is just an idea at the moment, but I'd like to hear you guys' opinion if there's a point in building such a machine. At the very worst, it will be used to cool down beers very quickly

Well that's great, although bear in mind that better crystals will be formed if the naphtha is allowed to cool more slowly. How might you turn your creative imagination be turned to that aspect?

I like the beer idea just as much, although that can be achieved with a damp sock, or ice and salt if you want it to be quicker.

Thermoelectric cooling

I'm sure it's possible to make a simple circuit that very slowly increases the wattage fed to the thermoelectric elements, causing a slow and measured change in temperature.

My suggestion;

Do some manual recrystalizations to really learn the finer aspects. Check my detailed posts in this thread https://www.dmt-nexus.me/forum/default.aspx?g=posts&t=73610
Once you think you have a pretty good grasp of it, pair the peltier with a temperature probe, hook it all up to a computer, and write a program to control temperature over time. I have a lot of patience, but not as much as a computer. This could potentially yield some large single polygon crystals.