Sciencemadness Discussion Board

Mixed salt bath instead of oil bath for heating?

Polysialate - 30-8-2019 at 15:38

I had the idea the other day to use* a saturated NaCl/KCl/MgCl2/CaCl2 solution as a heat-transfer medium between a hot pan and glassware instead of an oil bath. Just checking wikipedia it seems like CaCl2 or MgCl2 hydrates will remain liquid (and not evap) even at temps of ~150°C. I figure I can throw some NaCl and KCl in there as well just for a wider working temperature range.

My issue with oil baths are that they have flammability issues and stink up the place if the oil gets old. Plus it's a bitch to get old tar'd up oil off of beakers and such. Does this seem viable?

EDIT: *forgot a word

[Edited on 8-31-2019 by Polysialate]

Pumukli - 31-8-2019 at 12:22

I prefer dry heat transfer materials, even if they are obviously not as effective as liquids. I simply dislike any liquid that is dripping from hot glass so you have to soak it up with a paper towel, rag, anything before you could put down the glass on anything. Unless you want to risk slippery fingers, spots on your clothing, lab bench, lab notes, keyboard, etc...

You can use finely ground table salt eg, but it is a bit hygroscopic and tends to clump together.

What I use is even more "eco-friendly": sieved sand from the garden. It only smokes a bit at the first time when you "burn it out", later it is not a problem. (In case you keep it clean.) It has another advantage: does not splatter if you accidentally pour something liquid on it. It is definitely not so frightening in this respect as e.g. hot oil. It can't catch fire and fairly inert chemically. It doesn't clump. It can be used to put out (small) fires. It can clean your bathroom. (Well, it is not THAT versatile, but almost!) ;)

Polysialate - 31-8-2019 at 14:26

A granulated material like that should be a good insulator, rather than acting as a heat-transfer agent... no? I suppose it would mean more of the surface's heat went to the glass instead of the air, but for minimizing heat gradients (I use crappy "glassware", AKA empty liquor bottles) I think I'll have to go with a liquid transfer agent.