Evaporative cooling inside a freezer?
December 19, 2020 11:06 PM   Subscribe

If you want to make something cooler than its environment, you can wrap it in wet cloth and exploit evaporative cooling (at least, to the extent that the surrounding air can take up more water vapor). Is there a version of this trick that will work in my freezer?

That is, is there some similarly passive way to pump heat out of an item in my freezer so that it ends up colder than the other stuff in my freezer? (I know some areas of the freezer are colder than others, but let's assume I'm already putting the thing in the coldest part of the freezer, and I want to get it even colder.)

My motivation here is 40% getting my freezer-bowl ice cream maker to suck less, 60% satisfying some curiosity about thermodynamics.
posted by aws17576 to Science & Nature (21 answers total) 1 user marked this as a favorite
 
If you could do this, you'd violate "no net heat transfer from a colder thing to a hotter thing." The lazy thermodynamicist thinks in terms of isolated boxes which tend to the an average temperature over time.

You'd have to create a box in which the temperature of the ice-cream pail is warmer than some non-frozen sub-zero-celsius saltwater where the heat of the saltwater is itself continually cooling by evaporation. And yet, the overall box of the freezer is going to tend to an average level of heat energy so your ice-cream pail won't sustain being made colder.

Then, does water actually evaporate from the sub-zero-celsius saltwater? Your freezer should be colder than the triple point, at which we observe all three of water vapour, liquid water and ice, so maybe you won't get water vapour at all. Even if the zero-percent humidity made it plausible that some evaporation occurs, you can't use this to pump extra heat energy away.
posted by k3ninho at 1:08 AM on December 20, 2020


is there some similarly passive way to pump heat out of an item in my freezer so that it ends up colder than the other stuff in my freezer?

Yes there is. Evaporative cooling works at any temperature. The trick is finding a liquid that evaporates (or a solid that sublimes) in the range of temperatures you're interested in.

Liquid nitrogen will evaporate, and dry ice will sublime, at a lower temperature than any ever found inside a domestic freezer compartment, so either of those would work. In fact they'd work so well that you don't even need the freezer! Cooling ice cream with either of these substances works fine outside one.

In fact your freezer itself works by the exact same "passive" evaporative cooling mechanism, with liquid refrigerant evaporating to gas inside the tubing attached to the evaporator plate.

The only real differences between what goes on inside your freezer every day and typical wet cloth evaporative cooling are that (a) the freezer collects the evaporated refrigerant and pumps it around through a condenser for another go, instead of letting it all escape to atmosphere; (b) the boiling point of liquid refrigerant, at the pressure inside the evaporator plate tubing, is much lower than that of water, though not as low as either the boiling point of liquid nitrogen or the sublimation point of solid carbon dioxide; (c) the evaporator plate is typically not in direct physical contact with whatever it is you're trying to cool, which means that the cooling has to happen via the intermediary of cold air, whose thermal resistance is quite high.

You can lower the thermal resistance of air by making it move faster, so perhaps you might want to experiment with sticking a little battery-powered fan inside your freezer compartment.
posted by flabdablet at 1:57 AM on December 20, 2020 [2 favorites]


As I understand it, a frost-free refrigerator works by raising the temperature for a short period on some sort of schedule. If caught in one of those cycles, you ice cream tub is going to be a lot warmer than you want. I dont know any way to control, or even monitor, those cycles, but you might be able to protect against them some by burying the tub in big, heavy, and solidly frozen items.
posted by SemiSalt at 5:03 AM on December 20, 2020


Building on what flabdablet wrote, in the summer I use the freezer to quickly cool down cans of beer in the same way you mention: I wrap a can in a wet paper towel and leave it in the freezer for 15 minutes. By then the water has begun to freeze, so the evaporative cooking stops, but this is a great trick to quickly and dramatically chill canned and bottled liquids to temperatures well below refrigerator temperatures. The idea of putting a little fan in the freezer is a good one, as it would speed up the evaporative process to some degree.
posted by late afternoon dreaming hotel at 5:34 AM on December 20, 2020 [1 favorite]


If you could do this, you'd violate "no net heat transfer from a colder thing to a hotter thing."

It's not quite that simple, since a phase transition is involved. If the atmosphere was saturated with vapor of whatever substance you're using, then the wet-bulb temperature would be equal to the dry-bulb temperature and you could only cool the object to the freezer temperature. But because of the latent heat of vaporization, you can actually end up with a situation where an object can be cooled below ambient temperature.

Whether or not you can do this successfully do this in a home freezer without poisoning or incinerating yourself is another question. You're looking for something with a fairly high vapour pressure at -18°C (standard freezer temperature), which tells you (roughly) how quickly the stuff will evaporate. The standard undergraduate-chemistry-demonstration substance is ether, but that's not the sort of substance to be messing about with in your home freezer. The best bet might be plain old ethanol — get a bottle of Everclear of the highest proof you can find and see if it works.

Note that if it does work, you may need to find a way to ventilate the ethanol vapour from your freezer before you try it again. Otherwise, you'll end up with a situation where the air mass in your freezer is saturated with ethanol, and you won't get any evaporative cooling since no more ethanol can evaporate.
posted by Johnny Assay at 6:39 AM on December 20, 2020 [1 favorite]


(Disclaimer: I'm a physicist, not a chemist, and this stuff is a bit outside my realm of expertise; so I welcome correction about any of the above.)
posted by Johnny Assay at 6:41 AM on December 20, 2020


Cold air doesn't hold much moisture, so this should be somewhat effective. Water from the dishcloth or paper towel will evaporate, your ice cream will be colder. While making ice cream, set the freezer several notches cooler at least 12 hours prior, and chill the ingredients quite well.
posted by theora55 at 6:54 AM on December 20, 2020


You might be able to put a small solid state heat pump (Peltier) into the freezer to cool one smaller enclosure, however, the energy of this work will also have to be overcome. This could quickly overwhelm the relatively modest cooling capacity of the fridge itself, and lead to compressor damage.
posted by nickggully at 7:01 AM on December 20, 2020


Isn’t this what the ice-salt mixture is doing, but with melting providing the phase change instead of evaporation?

The risk is that if one area of the freezer is cooler than the rest and the thermostat finds out, it might stop trying to cool. Maybe you could put it in a cold insulated sub-container or something.
posted by Huffy Puffy at 7:01 AM on December 20, 2020


Isn’t this what the ice-salt mixture is doing, but with melting providing the phase change instead of evaporation?

Exactly right. The freezing point of plain water is 32F/0C. That is the equilibrium point between ice and the water it is floating in.

If you add salt to the water, you depress the equilibrium point. Depending on the salt concentration, you can get the equilibrium temperature down to about 0F/-18C.

The melting point of ice cream is just one or two degrees below plain water, mostly due to the sugar content. But you want to chill the ice cream as quickly a possible to that temperature so that the ice crystals are tiny and the ice cream is smoother. That's why you want a chiller down around 0F/-18C.

So you could just put your ice cream in your freezer at 0F/-18C and it would eventually freeze. But the texture might not be the best because of the large crystals. That is because transfer of heat from the cold air in the freezer is relatively slow.

So to get quick freezing, you are better off using the old standard ice cream mixer immersed in the ice and salt bath. The below freezing liquid at 0F/-18C will chill your ice cream faster than the air in your freezer at the same 0F/-18C. So chill your ice cream in the salt water mixture, and when frozen, transfer it to your freeze to keep.

Wrapping a wet towel around the container and putting it your freezer won't help. It will make it worse. First you have to freeze the towel at 32F/0C before it even begins to cool the ice cream.

What you could do to speed up the freezing is put a large container of very salty water in your freezer, so salty that it doesn't freeze at 0F/-18C. Then plunge your ice cream container in this container of very cold salt water. The salt water bath will conduct heat out of the ice cream much faster than just the cold air of the freezer.

But this probably isn't any faster than just using the ice cream maker as directed and using the salt water bath outside the freezer. And part of what makes the ice cream smooth is the cranking and stirring at the same time it is freezing, which you can't do if the whole thing is in your freezer.
posted by JackFlash at 11:57 AM on December 20, 2020 [2 favorites]


Response by poster: Thanks for the answers so far!

Just to clarify one thing, the "freezer bowl" style of ice cream maker doesn't use salt. It consists of a bowl which you bring to freezer temperature before use, plus a churn and motorized base. The bowl has some kind of fluid permanently trapped between its inner and outer walls with a low freezing point and high heat capacity, so it's supposed to be good at sucking heat out of the bowl contents. You make the custard for the ice cream, bring it down to fridge temperature, then pour it in the frozen bowl, put it on the motorized base, and churn it until the bowl and the custard equalize in temperature, at which point you stop because both are only going to get warmer from there. My problem is that my ice cream tends to barely frozen at that point. I figure if I can get the freezer bowl colder to start with, then it will be more effective.

Johnny Assay, your answer intrigues me. Regular-proof vodka already doesn't freeze at the temperature of my freezer, so I wonder if wrapping the bowl in a vodka-soaked paper towel would do the trick. If I can track down a bulb thermometer, there's an easy experiment to try here.
posted by aws17576 at 4:06 PM on December 20, 2020


Just getting the bowl a little colder probably isn't going to help much. What you really want to do is remove a lot of heat from the custard. Best way to do that is with a phase change solid to liquid, which your freezer bowl provides, but not enough.

So get yourself a second, spare freezer bowl (amazon, ebay). Now you have twice as much phase change material. Chill the custard in the first bowl for a few minutes then transfer to the second bowl for the final freeze.
posted by JackFlash at 4:52 PM on December 20, 2020


How long would you typically leave the freezer bowl in the freezer before using it to make ice cream?

I'm thinking that if your ice cream mixture it's not getting as cold as it should do after churning, then that means that the phase change material between the walls of the freezer bowl isn't soaking up as much heat as it's supposed to.

There are a few possible causes for that. One is, as JackFlash suggests, that there simply isn't enough phase change material in your freezer bowl to do the job. The fix for that is to churn smaller batches.

Another is that the freezer bowl isn't actually being left in the freezer for long enough to make all of the phase change material inside it actually change phase.

And a third is that the temperature inside your freezer is not actually below the freezer bowl's phase change temperature, which means that you're missing out on the magic of phase change heat removal altogether and just churning your ice cream in a cold bowl with a somewhat high thermal mass.

The third cause could certainly be addressed by using dry ice or liquid nitrogen to pre-chill the freezer bowl as well as or perhaps instead of just putting it in the freezer. But if the problem is actually one of the first two, this would make only a very slight difference.

You can find out whether you're actually getting some phase change out of your ice cream maker by monitoring the cooling-down process with a digital freezer thermometer.

You want the thermometer's sensor probe to be closer to the temperature of the freezer bowl's inner wall than that of the air inside the freezer. Easiest way I can think of to make that happen is to sink the probe in a liquid that won't throw off the measurements by doing its own phase change; high-proof alcohol should work. So stick the probe to the bottom of the bowl with a bit of chewy, pour in enough Everclear to cover it completely, put it in the freezer and log the temperature every five minutes.

If the freezer temperature is indeed set below the phase change point of the mystery fluid in the freezer bowl's walls, you should see the probe temperature come down steadily for a while and then stick at some very nearly constant temperature for a long time before finally dropping to the freezer's set point.

If you don't see that, then either your ice cream maker isn't designed around a phase change material, which means it kind of sucks, or your freezer temperature isn't set low enough to trigger the phase change.

If you do, then you'll know how long it's worth leaving the freezer bowl in the freezer before using it on a batch of ice cream.
posted by flabdablet at 8:08 PM on December 20, 2020 [1 favorite]


Another is that the freezer bowl isn't actually being left in the freezer for long enough to make all of the phase change material inside it actually change phase.

You should be able to tell by shaking it. I'm guessing that when liquid you should be able to detect a bit of sloshing which goes away when solid.

Another trick would be to get your ingredients to a lower temperature before mixing. Most refrigerators are 35F to 38F. If you just put your stuff in the refrigerator to pre-chill, everything is still several degrees above freezing. So you could put your liquid ingredients into the freezer for maybe 15 or 20 minutes before mixing to get them closer to 32F. Dry ingredients, like sugar, you could maybe pre-chill all the way to 0F in the freezer. And also pre-chill your mixer paddles.

It might take a little experimentation. You want your ingredients very close to freezing before you pour them into the mixing bowl, but not so cold that they freeze before you can get them properly mixed.
posted by JackFlash at 8:27 PM on December 20, 2020 [1 favorite]


We have that freezer bowl, and the most important thing is to get a thermometer for the freezer and make sure it really gets down to 0 degrees F.
posted by oneirodynia at 8:50 PM on December 20, 2020 [1 favorite]


Yes, the main thing with the (Cuisinart?) ice cream maker with the integrated freezer bowl & coolant is that the coolant must be fully frozen for it to work. If you shake the bowl and you hear sloshing, then your freezer isn't cold enough.

So now that I think about it more, I don't think evaporative cooling tricks are likely to help. If it is in fact fully frozen, then reducing the temperature of the frozen coolant isn't likely to do much; and if it's not fully frozen, then tricks with evaporative coolant aren't likely to help it freeze. Materials give up a huge amount of heat when they freeze, compared to the amount of heat required to change their temperature. For example, a given amount of water gives out more heat when you freeze it (starting with liquid at 0°C and going to ice at 0°C) than when you take that same amount of ice at 0°C and cool it down to -150°C (-240°F or so.)

So evaporative cooling is, I think, unlikely to be able to transport enough heat away from the freezer bowl if it's not freezing; and if bowl is freezing, it's unlikely to make a significant difference in how cold your ice cream gets. (My gut instinct is that evaporative cooling wouldn't be able to reduce the temperature of the bowl by more a few degrees, but that's really just a WAG.) As flabdablet points out, the magic is in the phase change.

Showing my work: the latent heat of fusion of water is 330 kJ/kg. The heat capacity of ice at 0°C is about 2.05 kJ/kg, and decreases as the temperature decreases. Thus, the heat transfer required to change the temperature of ice from 0°C to -150°C is less than 307.5 kJ/kg.
posted by Johnny Assay at 7:16 AM on December 21, 2020 [2 favorites]


My gut instinct is that evaporative cooling wouldn't be able to reduce the temperature of the bowl by more a few degrees

That really does depend on what it is you're evaporating and how much of it you use and how well the evaporation site is thermally coupled to the bowl. But just getting the freezer thermostat set right is generally going to be more convenient.

the magic is in the phase change

and that's exactly what evaporation is.
posted by flabdablet at 8:18 AM on December 21, 2020


Fair enough. I guess my implicit assumptions were that the amount of liquid being evaporated would be relatively small compared to amount of the coolant in the bowl, and that the coolant would be designed to have a relatively large specific latent heat of fusion (the better to freeze ice cream with). But who knows if either one is correct.
posted by Johnny Assay at 10:43 AM on December 21, 2020


I think those are all excellent assumptions, but failing to make them yields many more amusing possibilities for impractical and/or downright dangerous ice cream manufacturing technique.

Mmmm, butane.
posted by flabdablet at 10:55 AM on December 21, 2020


I don't have an ice cream maker but have read up on them a fair bit (I want one but it wouldn't be good for me if I got one). I'm pretty sure that after you make the ice cream you have to put it in the freezer for a while to further harden.
posted by any portmanteau in a storm at 1:54 PM on December 21, 2020


Oh hey I'm way late to this question, but I have another observation about this type of ice cream maker. I have two of them, both Cuisinarts. They use the same core, but one does a much better job at freezing the ice cream. The difference is that the more expensive one A) spins faster and B) has a closer fit between the dasher and the tub. These two effects mean that the ice cream is much better mixed. In the cheaper one, I get a thin layer of hard-frozen ice cream lining the inside of the tub, and liquidy stuff inside that. In the fancier one, I get uniform soft serve straight out of the churn. Both harden into decent ice cream after a night in the freezer, but the texture is better from the more expensive machine. So, there may be plenty of cold available, but it just isn't sufficiently mixed. Also, don't overfill it. The core should be only about half full when you start churning.
posted by agentofselection at 11:56 AM on December 29, 2020


« Older Webcomic about awesome Victorian names?   |   Just when I had started to believe that my needs... Newer »
This thread is closed to new comments.