Author: Greg Foster
The results of the prior oxygenation xBmts were confounding to me, as I’d been convinced oxygen (O2) played an integral role in healthy yeast growth and fermentation. While this definitely got me thinking, it wasn’t enough to stop me from using pure O2 in my brewing. And it’s worked fine, my beers have been good, though that skeptical voice in my head continues to nag me– is it really making a difference?
Following the Shaken vs. Pure O2 xBmt, we received many suggestions to focus these types of tests on beers purported to require higher concentrations of O2 in order to ensure the best fermentation, particularly high OG styles known for stalling early. While preparing to get this going, I embarked on a search for valid evidence demonstrating the positive impact of O2 that wasn’t focused solely on the commercial brewing scale, though what I discovered certainly didn’t confirm my presumptions. About a year ago, an article by Tracy Aquilla called Oxygen’s Role in Beer Fermentation appeared on MoreBeer.com that seemed to imply O2 isn’t nearly as important as I’d come to believe. From the author:
It is true that aerating or oxygenating wort is generally beneficial to fermentation, but it is untrue to say that yeast requires oxygen to reproduce or that yeast uses oxygen to respire during fermentation.
Generally beneficial? Yeast doesn’t require oxygen to reproduce? I highly recommend everyone read the article, it provides a truly thought provoking look at the biochemistry of yeast. Changing such a strongly held belief isn’t easy and, despite the evidence Tracy shared, the part of me that finds comfort doing what I’ve been told, appealing to authority, has convinced me there must be situations where wort oxygenation makes a difference.
| PURPOSE |
To evaluate the differences between a split batch of the same beer where half was oxygenated with pure O2 while no attempt to oxygenate was made on the other half.
| METHODS |
I designed my recipe with one goal in mind– go big! I wanted to achieve a momentus 1.100 starting gravity without a ton of roasted malts, so American Barleywine seemed like an excellent choice. I started with a simple Sierra Nevada Bigfoot clone, mixed up the grain bill a wee bit, then finished my recipe off with some dry English yeast. The dextrose and DME weren’t a part of my original recipe, the reasons for their use will become clear shortly…
American Barleywine
| Batch Size | Boil Time | IBU | SRM | OG | FG | ABV |
| 5.5 gal | 140 min | 114 | 14 | 1.100 SG | 1.020 SG | 10.38% |
Fermentables
| Name | Amount | % |
| Domestic 2-Row Malt | 17 lbs | 81 |
| Aromatic Malt | 1 lb | 4.8 |
| Crystal 60 | 1 lb | 4.8 |
| Dextrose (Corn Sugar) | 1 lb | 4.8 |
| CaraMunich | 8 oz | 2.3 |
| Extra Light DME | 8 oz | 2.3 |
Hops
| Name | Amt/IBU (Tinseth) | Time | Use | Form | Alpha % |
| Centennial | 43 g/47 IBU | 60 Min | Boil | Pellet | 9.9% |
| Cascade | 43 g/32 IBU | 60 Min | Boil | Pellet | 6.7% |
| Chinook | 43 g/24 IBU | 20 Min | Boil | Pellet | 9.0% |
| Chinook | 43 g/12 IBU | Flameout w/ 20 min steep | Boil | Pellet | 9.0% |
Yeast
| Name | Lab | Attenuation | Ferm Temp |
| WLP007 Dry English Ale Yeast | White Labs | ~79% | 65°F |
Two days prior to brewing, I spun up a couple 1.5 liter starters using previously canned wort then split a single vial of WLP007 Dry English Ale yeast between them, they were propagated on stir plates for 24 hours before being placed in my fridge to cold crash.
With the yeast taken care of, it was time to get brewing! I mashed in at my desired 154°F and set my RIMS to hold this temperature for 10 minutes.
It was at this point a strange brewing neurosis overcame me and I decided to lower the mash temp to 152°F for the remaining 50 minutes. Once the saccharification rest was complete, I collected the first runnings, performed a routine batch sparge, then transferred the rest of the sweet wort to my kettle and began to heat it up. The boil proceeded as expected with hops added per the recipe. At 15 minutes left until flameout, I checked the gravity and realized some shitty errors I’d made had resulted in the SG being lower than expected. This was supposed to be a big Barleywine! In an attempt to remedy this issue, I added the 8 ounces of DME I had on-hand to the boiling wort. Presuming this wouldn’t be enough to get me where I wanted to be, I desperately threw in 1 lb of dextrose. Before I could take another SG reading, the time had come to chill. I was super excited to try out my new Hydra IC from JaDeD Brewing.
I couldn’t believe how fast this bad boy worked! Equal amounts of the chilled wort were transferred to two 3 gallon Better Bottles, filling from the bottom up as gently as possible in order to limit the introduction of any oxygen for the no O2 batch.
I decanted the starters then pitched the remaining slurry into each of the carboys. It was time to get to the crux of the xBmt. While one of the carboys was left completely alone, the other was hit with a 5 minute dose of O2 dispersed using a 0.5 micron stone. The regulator was set to 1 liter per minute for the first 3 minutes, after which I cranked it up to 2 liters/minute for the remainder of the time.
I took a hydrometer reading at this point and was a little miffed to find I didn’t quite achieve the magical 1.100 OG I was hoping for, but rather came in a few points shy at 1.096 SG.
The carboys were placed in my fermentation chamber controlled to 65°F. I eagerly watched over them the following week to see if there were any observable differences in fermentation. Fermentation kicked off within 6 hours for both beers and appeared pretty much identical to start, then at 28 hours in the kräusen on the pure O2 batch began to creep slightly higher than the no O2 batch, to the point of creating a minuscule blow off.
After 3 days of fermentation, I noticed the kräusen on each beer had completely dropped and signs of activity were all but absent, so I decided to take an initial hydrometer reading.
There was actually a difference! The pure O2 batch had dropped crystal clear and finished at an acceptable 1.021 SG while the murky looking no O2 sample had only managed to ferment down to 1.026 SG. I checked on the no O2 carboy and was surprised to discover small bubbles popping on the surface of the beer, possibly just CO2 off-gassing, but I figured it’d be best to give both batches a bit more time to be sure. When I returned for a follow-up hydrometer measurement a week later, I found the no O2 batch dropped to 1.024 SG, just a fewpoints higher than its pure O2 counterpart. Both batches were cold crashed, fined with gelatin, then kegged and force carbonated. They were looking good when it came time to present them to tasters for evaluation later that week.
| RESULTS |
A total of 36 people participated in this xBmt, a panel consisting of numerous BJCP judges, experienced homebrewers, and long-time craft beer drinkers attending Strand Brewers Club and Pacific Gravity meetings.
Each participant, blind to the nature of the xBmt, was served 1 sample of the pure O2 beer and 2 of the no O2 beer then asked to select the unique sample. In order to achieve statistical significance given the sample size, 17 (p<0.05) tasters would be required to correctly identify the odd-beer-out. In the end, a total of 16 people (p=0.079) correctly selected the pure O2 sample as being different, allowing for the interpretation that using pure O2 on a high OG beer doesn’t appear to produce a reliably distinguishable difference from the same beer that wasn’t oxygenated at all.
My Impressions: I tested my ability to distinguish between the beers before I presented them to any participants. I was at an obvious advantage since on top of being aware of the nature of the xBmt, I poured myself large samples of each beer and compared them side-by-side rather than in a triangle. My hopes weren’t high, but to my surprise, I perceived a difference in aroma alone. For the sake of confirmation, I “blindly” tested myself 6 more times by marking one of the glasses and closing by eyes while whiffing. I remained able to identify the pure O2 beer every time, which I perceived as having a sweeter and less muddled malt aroma than the no O2 sample. Likewise, my results were the same when I focused on flavor alone, though I experienced more difficulty with this than with aroma. It’s possible that my performance was a result of bias, but to my taste buds the no O2 batch had a very slight yeasty off-flavor while the pure O2 batch was clean and smooth. Don’t get me wrong, these beers were incredibly similar, but personally, I preferred the pure O2 batch.
| DISCUSSION |
Yet again, my belief that using pure O2 has a significant impact on the flavor of beer has been shaken, tasters could not reliably perceive a difference between a beer dosed with “proper” amounts of oxygen and one that wasn’t oxygenated at all. As biased as I was, even I detected only the subtlest of differences. Taken alone, these results, when combined with those of prior xBmts, suggest homebrewers may not have to worry about oxygenation nearly as much as some of us do.
But this wasn’t the only data gleaned from this xBmt.
The objectively measurable differences I observed throughout the process lead me to conclude that oxygenation may in fact play a positive role, particularly when it comes to fermentation. The fact the pure O2 batch reached terminal SG nearly a week prior to, and finished .003 points lower than, the no O2 beer supports the notion that oxygen does have an arguably beneficial effect on fermentation and overall yeast health. Moreover, the pure O2 batch dropped clear much quicker than the no O2 batch, even without the assistance of gelatin. To me, this alone is enough evidence to keep me pumping my wort, especially those of higher OG, full of oxygen.
Finally, I’m curious how this xBmt might have played out had I utilized a vitality starter, as I’ve heard this method compensates for a lack of oxygen and contributes to better yeast performance. A future xBmt, indeed.
If you have any thoughts about this xBmt, please don’t hesitate to share them in the comments section below!
Support Brülosophy In Style!
All designs are available in various colors and sizes on Amazon!
Follow Brülosophy on:
FACEBOOK | TWITTER | INSTAGRAM
If you enjoy this stuff and feel compelled to support Brulosophy.com, please check out the Support Us page for details on how you can very easily do so. Thanks!
33 thoughts on “exBEERiment | Wort Aeration: Impact Aeration With Pure Oxygen Has On An American Barleywine”
Thanks for all the great experiments Greg. As a microbiologist (bacteria not yeast) I assumed the aeration in beginning was to allow aerobic growth of yeast (more efficient use of sugars..more ATP for energy, more biomass per gram of sugar) which eventually is used up after building biomass (unless you are stirring vigorously though not as much O2 gets in as pure O2 in the beginning.
Roy M. Ventullo, Ph.D.
Professor and Department Chair
Myrle M. Burk and Vera Burk Will Chair in Biology
Director of Undergraduate Research
Biology Department
Wartburg College
Waverly, Iowa 50677
Abbracciare il lievito
Thanks for all the great experiments Greg. As a microbiologist (bacteria not yeast) I assumed the aeration in beginning was to allow aerobic growth of yeast (more efficient use of sugars..more ATP for energy, more biomass per gram of sugar) which eventually is used up after building biomass (unless you are stirring vigorously though not as much O2 gets in as pure O2 in the beginning).
Roy M. Ventullo
Biology Department
Wartburg College
Waverly, Iowa 50677
Abbracciare il lievito
Hi. do you have any scientific evidence to support the use of 5 minutes of pure oxygen? I usually use one minute at flow rate of 1 liter per minute for regular beers and 2 minutes to the high gravity beers just to make sure. The use of 1 minute is the recommended by the Yeast book.
I’ve read conflicting info on this. There was a study a while back where they measured DO in wort and it required a lot more time, I’ll try to find it. In hindsight, I probably should have stuck with the 1-2 minutes Chris White suggests. I really need to find a DO meter…
There’s a difference in the level of dissolved oxygen you can obtain when using air vs pure oxygen. This may be where part of the confusion comes from.
Another confusing aspect is that it seems most homebrewers don’t use a regulator that meters out a specific flow rate.
I’ve wondered if some of these “rules” date back to when our yeast choices were more limited (less healthy?), tended to be dry yeast, making starters wasn’t common, etc.
ie does making a properly sized starter make up for oxygen, trub, quick chilling, etc?
Great experiment. I have to ask, do you test for normal errors or equal variance? We call anything under p=.10 a “strong trend”, and is typically repeated. .07 isn’t actually that bad, especially given your N (though a distinguished N, at that).
While it could be chance, I’m inclined to continue oxygenating my wort…
It would be interesting to compare the reduction from OG over the first several days of fermentation in a bunch of different samples of oxygenated and non-oxygenated wort, as an objective test of fermentability.
Very interesting. I was a diffusing stone and pure O2 guy for a long time. Now I just use a cheap plastic aeration piece at the end of my hose when I transfer chilled wort to fermentor. Works fine and never had a stalled fermentation.
I dig your RIMS. Would love to see a Xbmt involving RIMS or HERMS vs standard igloo cooler mash.
I wonder if a couple of your exBEERiments which showed no statistical significance when combined would result in beer that tasted different. For example, nobody noticed a difference in mash temperature or oxygenation, or using a starter when you did all of those experiments. What if you compared a beer mashed high, using a starter, which was oxygenated to a beer that was mashed very low, had NO starter, and was not oxygenated. I bet they could all add up to statistical significance.
Nice work Greg! I love all these great experiments you and Marshall are doing. I’m thinking that by using a starter, you’re already exposing the yeast to the majority of O2 they will need to complete the fermentation. Most commercial breweries (and myself as a home brewer) simply re-pitch slurry from the last batch — which has depleted its O2 reserves from the fermentation. I bet you’d see a major difference using o2 depleted yeast in a non-aerated wort.
Or maybe trying the same with dry yeast.
I always use yeast that I have harvested from an earlier batch, 6 half pints from the original, and 6 again from each of those. I never make starters an have never had a problem. This is just one case, but…
Hi Greg:
Lesson learned about transferring knowledge of bacteria metabolism to yeast. Did some research and seems while some bacteria (and yeast in low sugar media) switch between aerobic respiration and fermentation mode depending on O2 concentration, in wort (high sugar concentration) that is not the case for yeast. Thus my idea about respiration/fermentation switch with added O2 not valid. p.s I have been using O2 for starters and wort for couple years and it seems to be working well.
It depends:
http://www.mbaa.com/publications/tq/tqPastIssues/2003/Abstracts/1014-01.htm
and this:
http://www.mbaa.com/publications/tq/tqPastIssues/1999/Abstracts/tq99ab22.htm
Jason.
found this after posting-
From the conclusion section of this paper: http://www.mbaa.com/publications/tq/tqPastIssues/2008/Abstracts/TQ-45-4-0352.htm
“It is widely known that oxygenation of wort provides essential cofactors for lipid synthesis in growing yeast cells. Without adequate oxygen, yeast cells become weak, more susceptible to stress, and less able to carry out a fermentation in which conditions are unfavorable. The findings of this study clearly show that an increase in available oxygen in the wort results in a higher budding percentage and, therefore, higher cell growth. The control fermentations, with no oxygen added, never reached the yeast cell mass that was obtained in the oxygen-enhanced fermentations. As shown in Figure 2, the fermentation with inadequate DO never reached the cell mass of the other fermentations. This is particularly important to brewers during yeast harvest, because insufficient oxygen supply in the initial generation may result in not enough yeast being produced to collect for subsequent fermentations”
My understanding, from a Charlie Bamforth Brewing Network podcast I heard awhile back (if I recall correctly), is that the majority of O2 that the wort absorbs actually happens at the surface of the wort. In other words, putting an O2 stone the wort and bubbling O2 through it is mostly useless and just creates an opportunity for sanitation problems. That O2 just goes to the surface and blankets the wort, where it is absorbed over time. Also, if you add so much O2 that it begins pushing out of the top of the fermentor you’re just wasting the O2. So basically, the best practice, from what I understand, is to just fill the surface area of the fermentor with O2 and let it absorb over the next few hours. Easy! Anyone else have thoughts on that?
Interesting. I wonder if you could run a stone and tube threw bung and just set your O2 as low as possible and leave it for 4-6 hours to keep a blanket on top.
Is there a way to discover mutation rates of the oxygenated yeast cake? Would it be more likely with pure O2 to get fewer mutations and thereby making the yeast more optimal for re-use multiple generations?
Maybe I misunderstand your method, but how is fewer responses than chance statistically significant? 18/36 would be pure chance, yet you claim only 17 were needed to be statistically significant @ p=0.05
Actually, 12/36 would be chance for a triangle test. 17 responses is significant with 36 participants.
Ah, then you are calculating the probability of selecting the right beer (1/3) rather than a simple yes/no if they selected the different beer (1/2). Is that correct? Are you calculating the significance for each experiment or just using a critical numbers chart?
That’s right. We use a one-tailed binomial proportions test for each xBmt.
I love these experiments. One thing I wondered as I read this one is whether an attempt to induce at least some oxygenation through shaking and splashing of the wort inside the fermenter would have done enough additional oxygenation to make the two essentially identical.
Not complaining about the method–the attempt was to create the greatest difference in oxygenation possible, to see if there still was a difference in outcome. The results suggest maybe, maybe not; if maybe, the difference doesn’t appear great.
I suspect that shaking and splashing would further reduce whatever differences were detected.
I’m surprised no one has mentioned/suggested the idea of “insurance”? To me it seems fairly clear that it does help (albeit only slightly) the higher gravity beers. But also; It gives you an extra level of confidence that all is going to go well. Given you’ve done all that work, it’s always a shame when it goes wrong at the final stage!
Some good information from Wyeast that backs your results, especially relating to high-gravity beer.
Oxygenation for Home Brewers:
https://www.wyeastlab.com/hb_oxygenation.cfm
Oxygenation for Home Brewers – High Gravity:
https://www.wyeastlab.com/hb_highgravitybrewing.cfm
Oxygenation/Aeration for Commercial Use
https://www.wyeastlab.com/com_w_oxygenation.cfm
Methods of Aeration / Oxygenation
Homebrewers have several aeration/oxygenation methods available to them: siphon sprays, whipping, splashing, shaking, pumping air through a stone with an aquarium pump, and injecting pure oxygen through a sintered stone. We have tested all of these methods using a dissolved oxygen meter and have found that, when using air, 8 ppm of oxygen in solution is the best that you can achieve. Injecting oxygen through a stone will allow much higher dissolved oxygen levels. The chart below shows methods tested and the results.
Method DO ppm Time
Siphon Spray 4 ppm 0 sec.
Splashing & Shaking 8 ppm 40 sec.
Aquarium Pump w/ stone 8 ppm 5 min
Pure Oxygen w/ stone 0-26ppm 60 sec (12ppm)
Somewhere there is an actual journal article on wort aeration. It identified the three main types of aeration (shaken, not stirred, pouring back and forth from bucket to bucket, and Oxygen through a stone). If my memory serves me correctly, the Oxygen stone and the bucket to bucket method were considered equal while the shaking was considered less effective. You mentioned, in an earlier Xbeerment about the potential that the starter had enough oxygen to keep the yeast going. How about repeating the Oxygen vs. no aeration test with only rehydrated dry yeast, no starter? I don’t have an Oxygen system and pouring 5-6 gallons of wort from bucket to bucket 3 times for every brew day gives my back a good workout. It’d be nice to not have to do that at all (while I wait for my beer budget to build up enough for an Oxygen stone etc.).
Here’s the article: http://www.brewangels.com/Beerformation/AerationMethods.pdf
In the article (https://hwcdn.libsyn.com/p/6/8/b/68bd31b8e7d7cb4a/AerationMethods.pdf?c_id=1452668&cs_id=1452668&expiration=1585427610&hwt=3017f7d0dcd2c84b62c7a6e6dec85aa6) “. . .methods that expose wort to open air for an extended time, such as by repeated pouring between containers or using stirring devices in an open container, were not evaluated.” However, rocking / shaking was the most effective method.
Enjoyed the read. Seems interesting. I think you, along with other Brulosophy, need to rethink how you interpret the p statistics (p=.079), to me does not show that the beers are the same. It shows that you can’t reject the null hypothesis. You might want to read The earth is round (p<0.05) by Cohen.
I’m a big fan of your scientific approach to brewing. If nothing else it puts a lot of things into perspective. However, since all of your experiments test only a single variable at a time, you miss a real possibility — that each individual variable tested makes a difference that’s too small to be detected on its own but that combining multiple variables together could make a detectable difference. So, what I would like to see next is an experiment that puts together multiple variables that gave insignificant results when tested individually. Perhaps the best place to start would be a “widely considered best practice” beer vs. one that switches the values of all the variables that didn’t make an individual difference in your experiments. If even this results in beers that are statistically indistinguishable then you can feel much more confident that those variables can all be ignored.
So what was the mistake that caused you to miss your OG? I brewed a barleywine this weekend aiming for 1.100 and also undershot by .010.
Love your exbeeriments! In reading the Wyeast article on oxygenating, it appears that a high-trub wort can get by with less oxygen. Here is the quote from that article:
“…The unsaturated fatty acids found in wort trub can be utilized by yeast for membrane synthesis. If wort trub levels are low, yeast will need to synthesize more of these lipids and therefore will require more oxygen…”
I’d suspect if you ran the exbeeriment again trying to limit the trub, you may see even a larger difference. Maybe a 3 way comparison Low-Trub+LowO2, Low-Trub+HighO2, HighTrub+LowO2? Should see a difference between the Low-Trub samples and maybe no difference between the LowTrub+HighO2 and HighTrub+LowO2.
Thanks again for the effort, it really helps newbs like me.
Don Peacock
Lapeer Mi
I love reading these experiments. In this case I wonder why you chose a barleywine. I understand from Chris White’s book that oxygen levels can influence ester and fusel alcohol production. I.e. it is not just about yeast growth and attenuation, but also influences on alternative metabolic pathways. Barleywines have strong flavours that would hide any off flavours caused by low O2 levels. I am curious what you would find if you repeated this xBmt with a light lager where off flavours have nowhere to hide. Personally I stir my cooled wort vigorously for 2 minutes using a large perforated stainless steel paddle. This leaves a heavy froth on top several inches thick. Therefore, I aerate rather than oxygenate. Although this works great for ales, my lagers often have hints of clove. I am wonder If low starting O2 is the cause (my other suspect is water chemistry because ours is carbonate rather than sulphate). Cheers!