## Cooling liquid with gas

Marsh - 18-9-2008 at 17:02

Let's say I had a small tank of CO2, maybe 225g worth. Then I attached a cord from the bottle, going to a small copper or aluminum thermal transfer tube to cool a liquid. By small, I mean approximately 4mm diameter with thin walls. That tube would be spiral shaped, to provide a lot of surface area in the liquid it is submerged within, with an open end for gas exit. Now, if that metal tube was submerged in a liquid such as water at room temp, say 300ml, and enough bends were there to provide the surface area, could that CO2 be purged through the metal tube to cool the entire liquid surrounding it low enough to reach near freezing? Of course, the gas would be purged slowly as possible to aid the dispersion.

Would it transfer much of the gas temperature into the liquid, or would the gas simply escape the end without much thermal conduction?

I know that with enough CO2 it will work. I'm just trying to get an idea of how much it would take for that amount of liquid to be cooled.

I do realize there are plenty of variables involved, but I am roughly speaking.

[Edited on 18-9-2008 by Marsh]

roamingnome - 18-9-2008 at 17:27

or you can put 300 mls of water in your freezer and make ice cubes

seriously though ... gas can cool...

in new pebble bed nuclear reactors helium is the cooling gas in a closed loop... why helium will not form activation products from stray neutrons...

its seems that you describe dry ice evaporating threw a coil...
it would cool your system of choice....

but what are you really cooling, not that it matters per say

how putting your system in a dry ice acetone bath

quote: "how much it would take for that amount of liquid to be cooled."

300 divided by 18 grams/mole is your first mathematical operation...

497 - 18-9-2008 at 18:51

Hmm lets see...

225g/ 44g/mol = 5.1 mol CO2
5.1*15.33 kj/mol = 78.4 kj or 19 kcal

That is about equivalent to the energy absorbed by melting 235g ice. Not much really.
Sure it would work, but I see no advantage to it.

MagicJigPipe - 18-9-2008 at 18:59

And that's probably ice right at 0*C. Freezers usually run much cooler than that.

I think it's a waste of money and compressed CO2. The ONLY reason I could see this as a good idea is if it was used to reach lower temperatures than are practical with water ice or other substances/processes.

Marsh - 18-9-2008 at 19:23

That's what I was afraid of.

The idea was that it could be used to cool reactions from the center which are exothermic, rather than an ice bath on the exterior beaker surfaces only. This would allow you to progress more rapidly, without waiting for the ice to slowly cool the reaction.

I have some CO2 tanks, one is actually about 550g capacity. I was just curious if they could be put to any good use, but it doesn't sound like it.

[Edited on 18-9-2008 by Marsh]

12AX7 - 18-9-2008 at 20:02

What's the difference, then, between that and, say, refrigerated brine or ice-water?

Tim

Marsh - 18-9-2008 at 20:17

It's not to create a water bath, but rather sit in direct contact with the reactants, for temperature control during a reaction. The tube would have to be of suitable material or shielded of course. I was hoping it would be a useful idea that is a little quicker than an ice bath around a beaker. My mind was just pondering, since I have some tanks not being used.

[Edited on 18-9-2008 by Marsh]

not_important - 18-9-2008 at 21:58

A couple of points

Phase changes generally absorb or release much more energy than just heating or cooling something without a phase change.

The greater the temperature difference, the easier it is to move heat across a boundary. So CO2 expanding to give sub-zero temperatures may be more effective a near-zero coolant.

The larger the area for heat exchange the faster the transfer. How much area does the beaker present compared to the spiral tube?

Moving the medium over the heat exchanger will give better transfer, meaning that you should stir the reaction media in either case.

The phase change and area available for transfer relationships suggests you might do best with choosing a low boiling solvent or co-solvent and use a high area condenser so as to run the reaction in reflux, the boiling of the solvent removing heat efficiently and the big condenser presenting much more area for heat transfer than can be had in or around the reaction itself.

12AX7 - 19-9-2008 at 06:37

 Quote: Originally posted by Marsh It's not to create a water bath, but rather sit in direct contact with the reactants

Well you're not proposing to shoot this stuff directly into the reaction, you already stated you'd use a cooling tube. And you can just as well circulate refrigerated liquid through that tube. So can you find any advantage over that? I doubt it.

Better using CO2 for what CO2 is good for... seltzer, pressurizing, making dry ice (for those occasions when you want much colder acetone), etc.

Tim

watson.fawkes - 19-9-2008 at 15:08

 Quote: Originally posted by Marsh Let's say I had a small tank of CO2, maybe 225g worth. Then I attached a cord from the bottle, going to a small copper or aluminum thermal transfer tube to cool a liquid. By small, I mean approximately 4mm diameter with thin walls. That tube would be spiral shaped, to provide a lot of surface area in the liquid it is submerged within, with an open end for gas exit. [etc. ...]
I'm confused. How does the gas get cold so that it can cool the liquid?
Blind Angel - 19-9-2008 at 17:48

PV = nRT
P = Pressure
V = Volume
n = Number of mole
R = 8.31 kPa*L/mol*K
T = Temperature in K

If the volume of a gas change so it's pressure diminish and by changing pressure it cool down. Basicly, check for the thermodynamic of gaz, or just how a refrigerator work.

Edit: I'm assuming that the CO2 is under pressure here. If it's not then it's by simple heat transfer.

[Edited on 19-9-2008 by Blind Angel]

Panache - 19-9-2008 at 18:25

 Quote: Originally posted by not_important The phase change and area available for transfer relationships suggests you might do best with choosing a low boiling solvent or co-solvent and use a high area condenser so as to run the reaction in reflux, the boiling of the solvent removing heat efficiently and the big condenser presenting much more area for heat transfer than can be had in or around the reaction itself.

Butane can, sit in freezer til liquid, drill hole into top, use as your solvent, just let it boil as it likes, top up your reaction as needed. \$5 will get you 4 325ml cans. excellent toluene substitute for low temp reactions far better than cooling imo. The smelly additive is usually only around 10ppm and generally interferes in nothing much except your olfActory senses.

12AX7 - 19-9-2008 at 19:29

 Quote: Originally posted by Blind Angel PV = nRT Edit: I'm assuming that the CO2 is under pressure here. If it's not then it's by simple heat transfer.

Of note, CO2 isn't a very ideal gas and has a considerable Joule-Thompson effect. This, combined with a phase change (if using a dip tube or upside-down valve to release liquified CO2), gives a considerable cooling effect beyond mere expansion.

Tim

watson.fawkes - 19-9-2008 at 19:41

 Quote: Originally posted by Blind Angel [ideal gas law elided] If the volume of a gas change so it's pressure diminish and by changing pressure it cool down. Basicly, check for the thermodynamic of gaz, or just how a refrigerator work.
Uh, no. The Joule-Thomson effect was just mentioned, but J-T coefficients can be both positive (heating upon uncontrolled expansion) and negative (cooling). You have one process and two unconstrained variables; the J-T coefficient gives the exact relationship (it's defined as a derivative). CO<sub>2</sub> has negative coefficient at ordinary temperatures and pressures, so it does cool on expansion from a pressure tank, but not for the reason stated.

No, my question was, why is there any reason at all to assume that the gas is cold while in the heat-transfer coil? Is there an implicit orifice in the valve or the vicinity of the valve? Since it was capillary tubing that was mentioned (4 mm diameter), there's every reason to believe that the gas expansion happens at the end of the capillary, getting cold at the same time as it forms bubbles.

Marsh - 20-9-2008 at 05:59

Watson,

If the cooling tube has an open end on it, wouldn't the pressure drop as soon as the liquid CO2 leaves the valve, or as soon as it enters that tube? Say there was a hollow tube 100 meters long, it still is at standard atmosphere, because the end is open on it. This is assuming the liquid is purged out very slowly. So the liquid changes to gas long before it leaves the end, and has cooled.

If I had a condensing system, then I am back to something which also requires electrical power to run the pump. The idea was that this could be used anywhere.

watson.fawkes - 20-9-2008 at 07:42

 Quote: Originally posted by Marsh If the cooling tube has an open end on it, wouldn't the pressure drop as soon as the liquid CO2 leaves the valve, or as soon as it enters that tube?
Not necessarily. The J-T effect is about unconstrained expansion. If you're piping gas from a valve to a capillary tube, there's not going to be much expansion. Indeed, many home refrigerators use the junction of a capillary tube brazed into a larger tube as the expansion orifice.
 Quote: Say there was a hollow tube 100 meters long, it still is at standard atmosphere, because the end is open on it. This is assuming the liquid is purged out very slowly.
If it's at a constant pressure along its whole length, there's no flow. Is the valve closed?
 Quote: The idea was that this could be used anywhere.
If you're looking for this criterion, you want an ammonia absorption refrigerator. It runs on a heat input. Very simple versions (so called "cold ball") run in batch mode. More mechanically complex versions run continuously.