I saw ingeniously simple "cooler box" yesterday - a double-walled plastic box with some liquid between the walls. You freeze it, and it keeps temperature near the freezing point of that liquid, until the liquid melts, while you can store dry foods inside.

Now, the efficiency of such a box will depend on volume of the liquid, to a lesser degree - how far below its melting point it was frozen, and on latent heat of its melting. (and insulation of course, but that's an entirely different matter.)

While I suspect the liquid used in the box I saw was just water with a little coloring added, I wonder if I could do better than that. Is there a substance that can be reasonably obtained by a DIY tinkerer, that has a latent heat of melting significantly higher than water, and a melting point at normal pressure somewhere "within reason" - achievable by normal kitchen freezer, and not above "food inside will spoil"? (bonus points for it not being overly dangerous...) Or maybe is there some way to increase latent heat of melting water ice by adding something to it?

  • 3
    $\begingroup$ There are not that many (common) chemicals with a latent heat of fusion above that of water. Check out this list for example: engineeringtoolbox.com/latent-heat-melting-solids-d_96.html ---- therefore it will be very hard (maybe not possible) to find a liquid with the melting point roughly equal to water and a higher latent heat of fusion. $\endgroup$
    – Michiel
    May 22, 2013 at 10:53
  • $\begingroup$ If you find one, give me a call! I definitely want to be in on the kind of business we could make out of that. But i think what your asking for is some kind of free lunch. And the basic rule is: There's no such thing. $\endgroup$
    – Karl
    Aug 25, 2015 at 13:01
  • $\begingroup$ Good answer by @Jerry, and comments about the uniquely high latent heat of fusion of water. Depending on the plastic, you could also use a mixture of dry ice chunks/powder and isopropanol. Isopropanol freezes about 10C below the sublimation point of dry ice, so you aren't likely to freeze it. Iit will just be a very cold bath until the dry ice sublimates and the solution heats up. Extra insulation, inside and out, would be in order to keep things from getting too cold and to make it last much longer. But step one would be testing that the plastic is OK with isopropanol. Then reuse the isoprop. $\endgroup$
    – airhuff
    Feb 6, 2017 at 23:41

1 Answer 1


Indeed, it will be hard to find a substitute for water. You might want to look into how ice packs work.

Some of the things that are added to water in such packs include substances to prevent bacterial growth and to prevent the water to become fully liquid. The water thus become a gel.

Gel packs may be made by adding hydroxyethyl cellulose (Cellusize) or vinyl-coated silica gel.

One factor that you might consider, is that if there is more mass (for example, dissolving something in the water) will provide more substance for the any heat to go to, so that the temperature rise will be slower.

Another, is hydrogen bonds, which are very present in water molecules and which explains its high specific heats (latent heat and heat capacity).

Unfortunately, adding in stuff to the water will tend to disrupt the hydrogen bonds. I haven't tried it, nor can find something to support my next sentence, but I would expect ammonia to be a possible additive. It has a higher specific heat capacity than water and dissolves easily as well since it does not disrupt hydrogen bonds that much due to its molecular structure.

Glucose is also a potential additive since it doesn't disrupt the hydrogen bonds much, but its specific heat is apparently a bit lower than ammonia. I found a site where I could input some parameters and it outputs the heat capacity of the sugar water and the outputs often say that the specific heat capacity is lower than that of ammonia alone (from the same site that michielm mentioned, ammonia has a specific heat capacity of a bit less than 4.6 at $0^oC$).

  • $\begingroup$ The problem with ammonia (besides its smell and toxicity) is its melting point is -77C, so while I'd be increasing specific heat of the mix, I think I'd be reducing its latent heat at "home freezer temperature" - only after crossing -77C I'd get summarily more "cold stored" (heat removed) than I would in case of pure water. $\endgroup$
    – SF.
    May 22, 2013 at 14:19
  • $\begingroup$ @SF the ammonia will be dissolved in the water, so its melting point is not relevant. And since it will be dissolved, the smell will be significantly less (about 40 times less according to a pdf doc. The latent heat of water should not change by much since ammonia bonds well with water. And on the other side ammonia will absorb the heat as well all the way through the temperature changes. Meaning that when water will melt, more heat will be required to melt it since ammonia is appropriating part of that supplied heat. $\endgroup$
    – Jerry
    May 22, 2013 at 15:21
  • $\begingroup$ Yes, but the volume is my limit. To make room for adding 1cc of pure liquid ammonia that will only absorb heat proportional to its specific heat, I have to remove 1cc of water, which would at the same time go through phase change and absorb massive amounts of heat proportional to its latent heat, on top of its similar (if lesser) heat absorption. Unless the bonding somehow reduces the volume of the mix... does it? $\endgroup$
    – SF.
    May 22, 2013 at 16:07
  • $\begingroup$ @SF You don't have to remove the 1 cc of water. You can add 1200 cc of pure ammonia and the 1 cc of water will remain the same. Or alternatively, if you have pure liquid ammonia, replacing 1 cc of water by 1 cc of it will actually equate to adding back 1 cc of water together with 1200 cc of ammonia (1200 volumes of ammonia dissolves in 1 cc of water at 0C). There is enough intermolecular space to fit ammonia between water molecules. I looked for some actual experiments where heat absorbed and temperature change were plotted for water+ammonia and water alone, but without success. :( $\endgroup$
    – Jerry
    May 22, 2013 at 17:02
  • $\begingroup$ Wait wait wait. I have the dual-walled box with 100cc of room for coolant liquid. I fill it up with 99cc of water. How much pure liquid ammonia can I add before the mixture spills over? $\endgroup$
    – SF.
    May 22, 2013 at 19:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.