Tuesday Phase-Transition Blogging

Why does yesterday's salt-based beer chilling tip work? I'd assumed it was simply because the salt changed the melting point of the ice, allowing water to get colder than it normally does, but Midwest Product says there's more:

The presence of the salt does allow the water temperature to drop below 32 degrees, but this trick also works in part because the salt speeds up the melting of the ice. The phase transition from solid to liquid absorbs heat, which helps keep the water extra cold.

That's why calipygian's trick is related; it simply uses the transition from liquid to gas rather than from solid to liquid. Just like melting ice, evaporating water absorbs heat from its surroundings (in his example, the warm bottle).

Fascinating.

Comments

While not a thermodynamicist, I have some experience melting ice, water, and salt and measuring the temperature. I can assure you the addition of salt in rational quantities is of minimal consequence and is certainly not having a large effect on the final temperature (the time saved by not going to get and add salt could be offset by the extra time in the bucket of ice water. I'm also assuming you don't need the beer to be exactly 32F or lower). The keys to getting this to work are:

1) plentiful ice and water. Not enough ice and it's not cold enough, not enough water and there's poor conductance to the container of beer.
2) agitation. Mixes the liquid surrounding the beer so that warmed water is not touching container, also mixes beer in container to achieve a more even temperature.
3) beer container type. Aluminum conducts heat much better than glass and beer in a can will cool much faster than beer in a bottle.

Posted by Crusty Dem | June 17, 2008 8:40 AM

Oh well--

There's actually two things going on here-- 1) chilling the beer i.e., extracting a certain amount of heat energy (which you can calculate-- delta temperature times the beer volume times the specific heat of beer), and 2) doing it fast, which depends on the thermal conductivity of the beer and the temperature difference between the beer and the cold bath. So, you have to have a cold temperature 'reservoir' with enough capacity to take sufficient heat from the beer, -and- the cold reservoir has to be significantly colder than the beer for it to happen quickly. The salt helps with both effects.

Posted by MattF | June 17, 2008 8:48 AM

Midwest is right - which is kind of obvious if you look at it the right way. If you have ice right at the freezing point and some salt at room temperature, then why would you expect to get anything colder than than 32 F or 0 C by mixing the two?

What happens is that the ice and salt in contact at their surfaces begin to mix (due to their intrinsic thermal motion), but as they mix, the presence of the salt (which breaks into Na+ and Cl- components) destabilizes the ice (which has a complicated network structure at the molecular level), turning it into water. But, this process is endothermic, that is, it costs energy. Not a problem, because the ice is below room temperature, so the room is happy to lend the energy required.

Viewed the other way around, the ice and salt mixing to form salty water ends up cooling the surroundings.

Ironically, if you put the ice and salt together in an environment that is sufficiently cold, then the environmental energy needed to drive the salt-ice mixing isn't available, which means the process doesn't go forward and draw out any heat.

To put it another way, you can cool beer faster with salt and ice in a room temperature environment than you can in a freezer!

Posted by Ben V-L | June 17, 2008 8:50 AM

I saw this on MythBuster's awhile ago.

Posted by Jay | June 17, 2008 8:52 AM

Midwest is right - which is kind of obvious if you look at it the right way. If you have ice right at the freezing point and some salt at room temperature, then why would you expect to get anything colder than than 32 F or 0 C by mixing the two?

Because ice is typically not at the freezing point. A home freezer should be set at 0-5 F for max. efficiency. I reckon store-bought ice would be at a similar temp.

If you just add water to cold ice, the water freezes, staying at 32 F until it's all frozen, so you're capped at 32 F unless you have comparatively little water*. Adding salt to the mix allows the water to cool further before it freezes. So the final temp will be

* once it's all frozen, then the temperature can decrease further, but your beer is encased in ice, so what good is that?

Posted by Matt B | June 17, 2008 9:12 AM

Remedial thermodynamics is a necessity to understanding this dilemma. Briefly, temperature is not absolute. Ice and water can coexist at one temp, the melting point, but the water containers more heat than ice (a lot more) which is why you drop ice in your drink to cool it, rather than cold water.. Salt lowers the temperature at which ice melts, but it takes a lot to drop the temp by a noticable amount.

Posted by Crusty Dem | June 17, 2008 9:30 AM

I am a chemist and we use this trick all the time.

We will fill a bucket with ice and dump salt in, we reach about -10 deg C in a very short time.

Ben V-L is correct the Enthalpy of Solvation for NaCl is endothermic, so throwing table salt on your sidewalks in the winter does nothing.

If you really want to cool quicker add a little isopropanol to lower the freezing point a little more.

Then again you could plan ahead and keep your fridge stocked like I do :)

Posted by Brian | June 17, 2008 9:32 AM

I am a chemist and we use this trick all the time.

We will fill a bucket with ice and dump salt in, we reach about -10 deg C in a very short time.

Ben V-L is correct the Enthalpy of Solvation for NaCl is endothermic, so throwing table salt on your sidewalks in the winter does nothing.

If you really want to cool quicker add a little isopropanol to lower the freezing point a little more.

Then again you could plan ahead and keep your fridge stocked like I do :)

Posted by Brian | June 17, 2008 9:33 AM

Matt B, evidently I wasn't very clear. I was trying to show why Matt Y's formerly-held belief (that it is just an effect of lowering the freezing point by adding salt) can't be adequate.

So take the ice at whatever temperature you want between 0 F and 32 F. Take salt at the same or warmer temperature.

Now the question is: Why should mixing these two substances be able to make both of them colder? It should be clear that the answer has to involve more than lowering the freezing point of the water.

Posted by Ben V-L | June 17, 2008 9:54 AM

Brian, the suggestion of alcohol instead of salt is a good one. It's also beneficial to use small chunks of ice, or even a slurry of finely chopped ice. I'd recommend buying some inexpensive dilute ethanol, and instead of salt for increasing the ionic strength of the solution, add sugar. However, the sugar solution can be a bit hard to clean up, so I'll add an acid to the solution, lime juice works pretty well. If you've done it right, just pour into glasses and give the warm beer to your guests who insist on it. If they want it cold, just give them a wet sock (for calipygian's trick).

Posted by Crusty Dem | June 17, 2008 10:39 AM

But the primary element of why this is better than just chucking a beer in the freezer, is the thermal conductivity of water is much higher than air, and thus draws heat away from the beer much faster. (For the same reason, you can die of hypothermia in 55 degree water but generally not 55 degree air). The salt is just a marginal improvement.

Posted by Kolohe | June 17, 2008 11:21 AM

"Now the question is: Why should mixing these two substances be able to make both of them colder? It should be clear that the answer has to involve more than lowering the freezing point of the water."

The phase transition from ice at 0C to water at 0C
is indeed the answer to why everything gets colder.
There's latent heat of melting, or in simpler
terms, energy has to be added to break up the solid ice crystals and convert them into liquid water.

But then that begs the question of why the phase
transition happens at all ? And the answer to that
is entropy: the solid ice crystals are in a
highly-ordered (and thus low entropy) state; the
liquid water is less-ordered (and thus high entropy). Since we're also breaking up the
highly-ordered salt crystal lattice, we get an
entropy gain from that as well (I'm not sure
whether the dissolving is endothermic or
exothermic though).

So the change *will* happen precisely
when the increase in system entropy due to the
dissolution of the low-entropy ice and salt crystals exceeds the loss of entropy due to lower temperature.

The same effect can be observed with a sufficient
number of warm toddlers and a paddling pool.
Order dissolves quickly, and after a while the
toddlers get completely wet and cold and start shivering ...

Posted by Richard Cownie | June 17, 2008 11:36 AM