How is CO2 absorbed during natural carbonation?

Moving this from the Beginners Forum, but wasn't sure where to put it. It could go to Fermantation and Yeast or even the Bottling Forums, but in any case I thought it deserved its own thread.

There have been two schools of thought on how CO2 gets into our beer during natural carbonation. There was a thread here on HBT a while back and a BYO article that promote the idea that the CO2 goes into the head space and then is absorbed into the beer until it reaches equilibrium for that temperature. The other is best represented by this quote from Bobby_M.

As yeast are releasing CO2, it's one molecule at a time. A single CO2 molecule in liquid is already considered dissolved. Because the source of CO2 in bottle conditioning is in the liquid already the liquid will always have the higher concentration of CO2 compared to the headspace until no more CO2 is being produced. At that point equilibrium is reached and the concentration will be equal.

Click to expand...

Both make sense to me, but are contradictory, so I wanted to try to repeat the experiment with some better controls in place. For now I have a temperature probe attached to the side of the steel container. I don’t think that there is too much heat being produced by this level of fermentation, and from experience, I don’t think the liquid temperature will be significantly different from the reading at the surface of the container. Still, depending on how this goes, I may repeat it with a temperature controlled bath.

My biggest drawback is that I don’t have a way to monitor the experiment other than visual observations. Still, what I’m looking for is a peak in pressure and then a leveling. If I get this, I’ll know that further trials are needed.


Any comments or suggestions are welcome.




Edit to say:
It's been one week since starting this experiment. The pressure went up to about 20 psi over a few day and (accounting for temperature) has stay there. That combined with the logic and science of Bobby, ajdelange and others has me convinced that there is no build up of CO2 in the headspace prior to absorbsion.



The setup

As CO2 leaves the yeast cell it will move along any gradient of chemical potential. Assuming the yeast cells are uniformly dispersed throughout the fermenting beer there is no gradient (other than locally i.e. in the vicinity of each cell) but there is a gradient at the surface and so CO2 will move to the headpace but it will only continue to do so as long as the chemical potential of CO2 in the headspace is less than the chemical potential of CO2 in the beer. If the headspace is full of air then there is no gradient until the partial pressure of CO2 in the beer surpasses 0.0004 atm. Once it does that, however, an appreciable potential difference will exist and appreciable flow across that interface takes place in an attempt to reach equilibrium. If the fermentor is open to the air CO2 leaving the solution will accumulate (assuming a closed but vented container like a carboy with an air lock or a cylindroconical) until all the air is displaced at which point there will be 1 atm of CO2 pressure in the headspace and 1 in the beer. Further evolved CO2 molecules may leave the solution or they may stay and some from the headspace will redissolve but the net will be an outflow from the beer. This is what makes the airlock bubble. At cool temperature and 1 atm (absolute) CO2 pressure about 1 volume of CO2 is dissolved. Obviously, for the airlock to bubble the CO2 pressure in the carboy must be more than 1 atmosphere and in this case we can estimate how much that must be as the water level must rise about half an inch to let the gas escape. The pressure difference is this about half an inch of water (~0.5/12/30 atm.). Similarly there must be a potential difference between the beer and air and it is also of low magnitude but I'd have no idea how to measure it other than perhaps with an ion selective electrode.

If you close up the container after all the air is purged (or even if it isn't) the yeast will continue to produce CO2 and as long as the partial pressure in the beer is a tiny amount higher than in the headspace gas will flow to the headspace. This will continue until the pressure on the yeast renders them dormant or they consume all the fermentables. At this point, if the vessel is leakproof, the pressure will stop increasing.

You can, therefore, expect the pressure to rise and then stop rising. I'm not sure what that tells you. I suppose it does tell you that the mechanism isn't escape to the headspace followed by re-dissolution if that's what you are trying to demonstrate but it's pretty obvious that that is not the mechanism involved.

ajdelange – thanks much for your description of the process. I’m not sure I’m reading the last paragraph correctly. I think during bottle conditioning equilibrium is reached for the roughly 70 degree conditioning temperature – then when it is refrigerated, the solubility is increased, and some CO2 dissolves back into the beer. Is that correct?

If Bobby_M is correct, you should be able to bottle condition in growlers, right?

The arguement against that is that growlers are rated to hold carbonated beer, but not beer that is carbing. I believe I'm understanding you correctly Bobby, that you feel there isn't a spike in psi during bottle carbing, but just a gradual increase to equilibrium?

AZ_IPA said:

. . . growlers are rated to hold carbonated beer, but not beer that is carbing.

Click to expand...

This is one of my reasons for trying to get this straight. I'd bought into the spike theory and even brought it up when people asked about carbing in a growler. This science removes (or lessens) that threat. Once you fill your growler and quickly cap it, the pressure should match the volumes of CO2 in the beer.

Even with this information, the two things against carbing in a growler would be the chance of using too much priming sugar when bottle conditioning and also that once a growler is filled from a keg it is normally kept chilled which has a big effect on the pressure. If you let it raise to room temperature or higher there may be a chance of a bomb.

ajdelange said:

I suppose it does tell you that the mechanism isn't escape to the headspace followed by re-dissolution if that's what you are trying to demonstrate but it's pretty obvious that that is not the mechanism involved.

Click to expand...

Not so obvious to me. Should have paid more attention during my science classes. :cross:

Not everyone else either. You should check out this article from BYO. Goes to show that you can't believe everything you read.

Natural conditioning relies on residual yeast suspended in the beer that didn’t settle out during fermentation. The yeast consumes the priming sugar and releases carbon dioxide. Because the bottle or keg is capped, the carbon dioxide is not able to escape. Over time the trapped carbon dioxide dissolves into the beer, naturally carbonating it. Typically the carbonation process requires two to three weeks to complete.

Click to expand...

Awesome! So the carbon dioxide that is being produced is already in solution. As the partial pressure exceeds the partial pressure in the headspace it will diffuse into the headspace until the pressures equalize.

When you keg it’s the exact opposite, except the gas diffusing into the liquid is a lot slower process.

So when bottles take forever to carbonate, it’s because the yeast are being poky eating the priming sugar.

Here’s a link explaining why soft drinks spew when you open a shaken bottle. It might help explain some of the chemistry. Hint :

You might think the pressure inside the bottle increases when shaken, but it does not.

Click to expand...

- http://www.wonderquest.com/spewing-soda-can.htm

ncbrewer said:

I think during bottle conditioning equilibrium is reached for the roughly 70 degree conditioning temperature – then when it is refrigerated, the solubility is increased, and some CO2 dissolves back into the beer. Is that correct?

Click to expand...

Yes. Same physics though. Here I'd expect the head space to cool faster than the beer (less thermal mass) so that the potential of CO2 on the liquid side would be higher and gas would initially migrate to the head space. But then as the beer cooled the potential on the beer side would go down and some of the gas would move back to the beer.

I missed this thread for some reason but I wanted to thank anoldUR for taking the initiative to follow through with the experiment. I would love to copy it just for some more confirmation if I had time. I'm glad AJ chimed in too, although in a language few mortals may ever decipher. It is quite a interesting topic in that science heavy folks (not me) don't even know it's a questioned phenomenon while typical homebrewers accept the overpressurized headspace assumption. I'm sure the myth will still spring up from time to time, and it would be neat to send a letter to the editor (BYO) to challenge their version of the story.

As for carbing in growlers, for one thing, they're only designed to be filled with cold beer, stay cold, open and drink. Just a simple look at a chart says a growler only has to hold about 15psi max (even higher carbed beers will lose some pressure in the pour). Take that same 13psi/40F beer and let it warm up to room temp in a growler and you get 30 psi. This is also what it would have to hold during a carbonation to 2.6 or so volumes. Now, maybe many growlers can hold 20/30/40 psi all day long while others can't. The second challenge is carbing in larger vessels and getting an accurate priming solution.

This is a great thread. Thank you guys. I have bottle conditioned in growlers before and it has turned out quite well. I used a 6 gallon bucket with priming sugar and filled 12oz, 22 oz, and growlers and the growlers carbonated very well at about 70 degrees. After opening they do not keep their carbonation for more than a day or so, though.