Yeast needs to breathe

Extracorporeal Membrane Oxygenation deviceAs readers may have guessed from previous posts, my brewing interests are minimally conventional. Fortunately, the Basic Brewing Radio podcast seems to regularly expand well beyond the usual “fermented malt flavored with a tisane of hops” thing (I need to try to make my own “Ginger Beer Plant” from scratch one of these days…). A couple of weeks ago, they did an episode covering an experiment on aeration methods which was very interesting. It does my ego good to know that I correctly guessed how the results would turn out. You can get a copy of the nice write-up of the experiment itself here, but here’s the simple version:

It’s usual to start brewing projects by mixing your ingredients together in a big pot and boiling them. In addition to potential effects on the flavor this also serves to sterilize what you’re about to ferment, which helps prevent having things growing in it besides the yeast (and possibly bacteria) that you intentionally put into it. Unfortunately, this also gets rid of most of the oxygen in your brew since gases don’t stay dissolved well in hot watery liquids. In order to allow your yeast to grow enough before settling down into the “fermentation” stage of the process, the yeast rely on oxygen to provide biological power through respiration plus a bit more oxygen which is used to generate cell membrane material in a separate process. That means that after boiling, you normally need to aerate your liquid somehow once it cools down.

Traditionally, this would be done by shaking or stirring the brew, or possibly by pouring it back and forth between two containers a few times. Humans being fundamentally lazy animals, however, a lot of people seem to like using air pumps instead. Fred Johnson decided to compare aeration with air pumps (with and without an “air stone”) with aeration by shaking.

I don’t think I can explain the results without giving away the punchline, so here goes: shaking works much better than air pumps. I recall reading the reason for this years ago in a book on keeping aquarium fish; the “air pump” is not actually putting substantial amounts of oxygen into the water. The reason the air pump results in aeration of the water is because the bubbles make the water circulate, drawing the water with less oxygen from the bottom of the tank to the top where it can exchange gases with the air across the surface of the water. This would also explain why the experiment seemed to show that just pumping the boiled test water into the container seemed to cause a substantial increase in the dissolved oxygen all by itself. In short, the more of the surface of the water that is exposed to air, the faster it will reach its normal oxygen saturation level.

Because I am extremely jealous of Fred Johnson’s ability to get his hands on peristaltic pumps and a dissolved oxygen meter while at the moment I can’t even afford a pH meter (let alone the microscope setup I want […sniffle…]), I’d like to take a couple of paragraphs to pick at a couple of possible nits in the experimental design to make myself feel better. At least one of those nits gives me an idea though…

First, after being cooled the boiled water in the experiment was pumped into the containers for testing through silicone tubing. Silicone is possibly the most oxygen permeable “rubber” in existence[1][4] – so much so that there are standard medical devices which act as artificial lungs, pumping blood through silicone membranes in air to oxygenate it[2]. If it works for blood, perhaps this principle might also be useful for other liquids. I’d love to see some experiments to determine how quickly water pumped through a length of silicone tubing is aerated (I would also hypothesize that putting lots of bends in the tubing would induce turbulence and increase the oxygenation rate over pumping through a straight or gently coiled set of tubing). I’d also like to see the same experiment done with butyl rubber tubing for comparison – butyl rubber is what we used for the stoppers on anaerobic culture tubes, since it’s got very low gas permeability by comparison[3].

I would also be curious as to whether sparging with nitrogen would be more effective at removing oxygen from the water than boiling would. This would probably have to be done with smaller samples, though – in the lab we were bubbling nitrogen through each small, narrow 10ml tube of culture medium for about ten minutes before capping the tube with a butyl-rubber stopper to do anaerobic culturing. I suspect it’d be a lot more difficult to scale the same method up to 6 gallons…

Silicone TubingAnyway, here’s something that might be worth playing with if silicone is permeable enough for the purpose: one might get their hands on a long spool of silicone “air-line tubing” (smaller tube = greater surface/volume ratio and hypothetically faster gas exchange), then rig up a way to transfer your wort/must/whatever directly from the boiling pot to the fermenter through the coiled tubing to accomplish the aeration without having to shake, bubble, or otherwise disturb the liquid, thus reducing the threat of contamination.

a sprinkler headA simpler hack might be to install a sprinkler head on the outlet end of the transfer tubing. Simpler still, perhaps crimping flat (but not closing entirely) the outlet end of the tubing so that the emerging liquid emerges spread out into a flat sheet rather than a stream. Either way, this ought to give a big increase in the amount of liquid surface exposed to air on the way out of the tubing and down into the fermenter, resulting in a substantial amount of incidental aeration without any additional labor or motorized gadgets.

There, aren’t I a know-it-allsmartass guy?

[1] Zhang H, Cloud A: “The Permeability Characteristics of Silicone Rubber
“; IN Sampe Fall Technical Conference Global Advances in Materials and Process Engineering: 38th International Sampe Technical Conference; Society for the Advancement of Material & Process Engineering ; November 9, 2006 ISBN 0938994727
(Article accessible as PDF from here as of 20080910)
[2] Carlson RG, Landé AJ, Ivey LW, Starek PJ, Rees JR, Subramanian VA, Twichell J,
Baxter J, Bloch JH, Lillehei CW: “Total cardiopulmonary support with disposable membrane oxygenator during aortocoronary artery-vein graft operations.”; Chest. 1972 Oct;62(4):424-32.
[3] Hungate RE, Smith W, Clarke RT: “Suitability of butyl rubber stoppers for closing anaerobic roll culture tubes.”; J Bacteriol. 1966 Feb;91(2):908-9.
[4] AZo Journal of Materials Online: “Silicone Rubber”; (accessed 20080910)

Published by


The Author is (currently) an autodidactic student of Industrial and Environmental microbiology, who is sick of people assuming all microbiology should be medical in nature, and who would really like to be allowed to go to graduate school one of these days now that he's finished his BS in Microbiology (with a bonus AS in Chemistry). He also enjoys exploring the Big Room (the one with the really high blue ceiling and big light that tracks from one side to the other every day) and looking at its contents from unusual mental angles.

4 thoughts on “Yeast needs to breathe”

  1. James Spencer pointed me to your blog, and I must respond since I discovered an error in my report after the podcast aired and after the manuscript was “published”. I reported the error to James Spencer and he played it on a subsequent podcast (, Sept. 4, 2008 podcast). I discovered that my DO meter was giving me erroneously high values at the low end of the dissolved oxygen scale, so I don’t really know what the oxygen content of the water was immediately after delivering it to the fermentor. However, it was likely much closer to zero than I reported initially. When I boiled the water or used metabisulfite to completely eliminate oxygen from the water, the meter was still reading dissolved oxygen levels over 3.5 mg/L.

    I doubt there was very much oxygen picked up during the transfer from the use of silicone tubing. At one time in my career, I was performing ex vivo liver perfusion experiments in which I used very thin-walled silicone (Silastic) tubing to oxygenate the blood in the perfusion system (as you alluded to), but it was crucial that the wall thickness be very thin for good oxygen exchange. The 1/6 inch thick silicone tubing I used may pass oxygen, but not rapidly enough to be an effective single-pass oxygenator.

    (If I ever fix my DO meter/probe problem, I’ll repeat the experiments that were wrongly published.)

    Fred L Johnson

  2. FYI. The silicone tubing I used to oxygenate blood in an artificial had a wall thickness of 0.25 mm.

    Fred L Johnson

  3. Oh, thanks – I had forgotten that you’d mentioned using the sodium metabisulfite to eat up the oxygen.

    Thanks also for the information about the silicone tubing permeability also – I found pretty easily that it was “very permeable” but hadn’t yet had time to try to dig up a reference that explained exactly how permeable it was.

    I’d still be curious about how permeable “air-line tubing” or an equivalent type of silicone tube would be – I wonder if it might be useful to have a large loop of it through which the fermenting wort/must/whatever could be recirculated in the open air. Ideally, there’d be enough permeability to provide oxygen for cell membrane manufacture but not enough to support respiration, keeping the yeast healthier throughout the fermentation process. (I have no idea at this point whether that would help the flavor or not, but I haven’t spotted anyone trying something like that yet…)

    Looking forward to the continued experiments – thanks for doing them!

Leave a Reply