Oxygen is critical for fermentation! In a side by side test, this experiment tests the implications of an additional forced O2 additions at 24 hours in a traditional session mead fermented with US-05. How important is O2 for fermentation, and what are the implications of not adding it?

Yeast requires a great deal of oxygen during the “lag phase” when it is rapidly multiplying as it is critical for cell wall health. Yeast needs between 8 and 10 parts per million (ppm) of oxygen.  A level of 8 ppm is achievable using air alone (which is 21% oxygen), but a higher level can only be achieved using pure oxygen.

There are three basic methods for aerating wort:

1. Shaking – Splashing the wort around in the fermentor can add sufficient oxygen. As long as there is enough headspace, you can get up to 8 ppm of oxygen for shaking aggressively for two minutes.
2. Stirring – Agitation is done by stirring rapidly with a whisk for several minutes, often making a vortex and splashing. Generally, a sterilized wine degasser whip is best if you have open access to the wort.  There is some concern that you can only get up to 6 ppm of oxygen, but it will depend on time, etc.
3. Injection – There are many ways to inject air or oxygen directly into the must, using bubbling or a carbonation stone or aeration stone at the end of the tube and wand.  You can use an inexpensive aquarium pump with an inline sterile filter or an actual oxygen bottle and regulator to inject oxygen.  Both can achieve the 8 ppm aeration level.

I prefer fermenting in buckets or in wide mouth 7 gallon wide mouth plastic carboys and agitate aggressively with a wine whip when mixing. Then, I agitate aggressively with the lid off to degas, which mixes in more 02 over the first two to three days. I’ve had great success with this method. Initially, I degassed less and added a short pure 02 addition at 24 hours, but I got away from this to avoid the cleaning of the O2 wand step but still made sure to agitate when degassing and didn’t notice much of a difference between using a stone or not. However, for many of the experiments, I ferment smaller batches in plastic water jugs. This does not allow for aggressive agitation with a wine whip or oxygen transfer, so instead I shake the jugs for two minutes at the start to try to get the 02 into solution. Reportedly, this is supposed to add sufficient 02 to the solution. I wasn’t using pure O2 because I got away from it when using buckets. However, I began noticing some small lager-like sulfur character in these carboy ferments. I designed this experiment to tests whether it could be coming from the oxygenation methods.

In this bench trial experiment, a forced 02 addition at 24 hours was tested for its flavor and aroma contributions in a 5% ABV, carbonated, dry traditional mead fermented using US-05. Both meads were aerated before pitch by aggressively stirring with a wine whip for three minutes. The treatment was an additional O2 addition using pure O2 and a diffusion stone at 24 hours. All other variables were kept the same. Both meads were presented in triangle tests in front of two judges. Judges were asked to determine the odd mead out and provided preference and tasting notes.

Recipe: 4.5%, Dry Traditional Short Mead, 2021, two 10.5 liter musts in 15 water jugs

• 9-liters of spring water
• 1.17 kg of white honey, clover and alfalfa, from Peace Valley Apiaries
• 2 x green 15-gallon spring water jugs
• 0.8 grams of calcium chloride
• 0.3 grams of Himalayan sea salt
• 1/5 packet or 4.9 grams of US-05 yeast

Treatment:

• 2 min of forced 02 at 24 hours with a diffusion stone, low flow, no big bubbles

Nutrients (calculated using The MeadMakr BatchBuildr):

Mixing Notes

• Mixed honey, water, salts aggressively for two minutes with wine whip, aerating with whip
• Sprinkled yeast on top of must, swirled in after 20 minutes

Fermentation Notes

The inorganic nutrients were split into three additions and given at 4, 24 and 48 hours.

Regular temperature, pH, gravity and aroma tests were taken. The table below summarizes the observations.

Fermentation Comments

I’m actually a bit surprised how similar these fermentations were. There were minor differences in the pH and activity, within the range of measurement error. That said, the mead with 02 treatment did seem to maintain a higher pH. Only on day four did I start to detect any differences in aroma. The mead with treatment kept smelling of bright and fresh throughout, whereas the mead without 02 starting smelling a bit more muddled and less bright. It did not, however, give off any strong off-flavor aromas.

Secondary

The meads for testing were racked into 1 gallon carboys on day 9. They were bottled at 4 weeks from pitch, using honey to 2.5 vol. The meads had dropped clear. The mead with O2 had notes of very light honey and was clean. The mead with no 02 also had light honey character but was somehow less sweet and some lager-like sulfur character.

Tasting Notes

Meads were tested at 8 months. Two judges were blind to the treatment, and the meads were served in six consecutive triangle tests in random order, and with a random odd mead out, in identical cups. Judges were asked to guess the odd mead out and provide brief tasting notes and preference rankings.

There was a significant difference between the two meads. In 10 of the 12 triangle tests, participants could identify the odd mead out. The null hypothesis that the results were from random guessing is rejected with 99.99 percent confidence. Both participants preferred the mead with 02 treatment, but the sample size to too small to tests for significance of preference. Here is a summary of the results:

Both judges, thought the mead with treatment was clean and honey like. One judged noted the floral character. The first judge thought the no 02 mead had lager-like aroma, and sulfur – like egg – was noted after swallowing. The second judge thought the no 02 mead was stinky and not good.

Both judges were mostly making their choices mainly by smelling. The second judge was preferring not to drink the no-02 mead, describing it as rank. The sulfur was quite strong, and the second judge was noting that it was lingering in their nose and resulting in having a harder time telling the clean mead apart.

Final Notes
I tasted the meads when serving the triangle tests. The mead with the 02 treatment was a perfect mead for me, and I really liked the subtle honey-like character and was very clean. It would also make a really good base mead for other flavors. I dumped the no O2 mead. I was surprised how much sulfur was in there during pouring, since it did not seem that dramatic during fermentation or bottling.

This experiment was really insightful for me. When I think of sulfur off-flavors, I would think about yeast stress from improper YAN/ over pitching, or temperature drops. Now I know that it can also arise from improper oxygenation, another critical nutrient. Now I can provide more informed feedback when judging.

Note that I did not shake the jugs for two minutes at the beginning, but just aerated with a wine whip. Shaking is known to add more 02 into solution. However, as I have also got slight sulfur character when shaking for two minutes, I am beginning to think that there needs to be later introductions of 02 during fermentation as the 02 is not available for later generations of yeasts. I am starting to either open ferment for the first two or three days or adding a pure 02 addition at 24 hours. However, the plan is to put this to a more inclusive bench trial to test when to add and when is too much.

I also did not add Go-ferm. Yeast can substitute the amino acids in Go-ferm for oxygen in building cell walls. I’ve noted in other bench trails that the use of Go-ferm eliminates the slight sulfur character when fermenting in carboys without the 24 hour O2 addition. It seems that there could be some interplay between the amino acids in Go-ferm and 02 additions. At least in this experiment, I can see that the later 02 addition gave a very clean mead even when I did not use Go-ferm.

After this experiment, I tried a few batches where I added 02 when using buckets/ wide mouth jugs, but then starting getting far too much yeast growth when also degassing and aerating with a wine whip resulted in some sluggish ferments. I’ve since gone back to only degassing and aerating with a wine whip when using buckets/ wide mouth jugs and adding the 02 when fermenting in carboys. Maybe this is why mead makers often prefer to ferment in buckets (or course, other than cleaning and handling fruit)?

Take away: pay attention to 02. Now, for my session meads recipes, I recommend:

1. if using a bucket/ large mouth carboy sufficient 02 can be added by taking the lid off and aggressively mixing with a wine whip two or three times a day for the first two to four days. If you’re not degassing aggressively or open fermenting, it’s best to give a pure 02 addition at 24 hours.
2. if using jugs for primary fermentation, shaking each of them for 2 minutes aggressively at start. Then either open ferment by covering the opening with a paper towel or coffee filter and elastic band for the first three or four days or add a pure 02 addition at 24 hours.

That said, given how important later O2 additions seem to be, more bench trials are warranted.

In the article on Tailored Additional of Nutrients with Go-ferm (TANG 2.0), I documented that adding Go-ferm at the recommended 1.25 grams per gram of dry yeast can quickly result in excess nutrients for session meads if the pitch rate is much above the recommended 1 gram per gallon of dry yeast. Then, in the high vs low pitch rate with Go-ferm triangle test experiment, it was found that a five gram per gallon pitch rate with the recommended amount of Go-ferm resulted in significant off-flavors when compared to a 1 gram per gallon pitch rate. However, this begged an important question: what is the pitch rate threshold for which Go-ferm off-flavors are perceived?

In this bench trial experiment, alternative amounts of Go-ferm and pitch rates were testing for their flavor and aroma contributions in a 4.5% ABV, carbonated, dry traditional mead fermented using S-04. Five Go-ferm amounts/ pitch rates were tested on identical musts. Each batch was fermented side-by-side and treated identically. Given the number of meads, all five meads were presented in a bench trials in front of two judges. Judges were asked to provided feedback and rank the meads.

Recipe: 4.5%, Dry Traditional Short Mead, April. 2021, 8 liter must, split into 1.7 liter batches

• 7-liters of spring water
• 1 kg of white honey, clover and alfalfa, from Peace Valley Apiaries
• 0.5 g of cal. chloride
• 0.25 g of gypsum
• 0.2 g of Himalayan sea salt
• S-04 yeast

Go-ferm variations:

• 2.5 gram per gallon yeast pitch rate, no Go-ferm
• 2.5 gram per gallon yeast pitch rate, 1.25 g/gal Go-ferm
• 2.5 gram per gallon yeast pitch rate with 1.25 grams per gram yeast pitch rate of Go-ferm
• 3.5 gram per gallon yeast pitch rate with 1.25 grams per gram yeast pitch rate of Go-ferm
• 4.5 gram per gallon yeast pitch rate with 1.25 grams per gram yeast pitch rate of Go-ferm

Nutrients (calculated using The MeadMakr BatchBuildr):

At pitch

• Mixed honey, water, salts
• Transferred to fermentors, added needed Goferm.
• Shook jugs for two minutes
• Sprinkled yeast on top of must
• First nutrient addition at 3 hours

Fermentation Notes

The inorganic nutrients were split into four additions and given at 3, 18, 36 and 48 hours.

Regular temperature, pH, gravity and aroma tests were taken. The table below summarizes the observations.

Time	2.5 g pitch, no go-ferm	2.5 g pitch, 1.25 g/gal go-ferm	2.5-gram pitch rate w go-ferm	3.5-gram pitch rate w go-ferm	4.5-gram pitch rate w go-ferm
+24h am 63.7°F pm 63.1°F	Unfermented honey	Unfermented honey	Unfermented honey	Unfermented honey	Unfermented honey/yeast
+48h am 63.5°F pm 65.2°F	Slight phenolic/ mothball honey 1.023 3.85 pH	Clean, honey, muddled apple 1.024 3.90 pH	Clean, honey, yeasty 1.022 3.93 pH	Yeasty 1.019 3.74 pH	Yeasty 1.019 3.93 pH
+3d am 65.8°F pm 64.9°F	Clean, honey, muddled apple 1.013 3.39 pH	Clean, honey, bright apple 1.013 3.38 pH	Clean, honey, apple, yeasty 1.009 3.43 pH	Clean, honey, muddled apple, slight yeasty, slight alcohol 1.006 3.40 pH	Apple, slight go-ferm, slight alcohol 1. 006 3.56 pH
+4d am 63.9°F pm 67.4°F	Clean, bright apple 1.006 3.24 pH	Pear, slight alcohol 1.008 3.35 pH	Pear, slight alcohol 1.005 3.38 pH	Pear, clean, yeasty 1.002 3.39 pH	Low pear, go-ferm, 1.002 3.53 pH
+5d pm 65.8°F	Apple, honey, clean 1.003 3.32 pH	Pear, slight alcohol, 1.003 3.32 pH	Pear/apple, alcohol, 1.002 3.38 pH	Pear, honey, clean, 0.999 3.44 pH	Pear, alcohol 0.999 3.53 pH
+7d am 66.2°F pm 67.3°F	Apple, clean 1.002 3.30 pH	Apple/pear, slight alc 1.002 3.32 pH	Pear, clean 1.000 3.38 pH
+8d pm 67.7°F	Apple, honey, clean 0.999 3.32 pH	Apple, honey, clean 0.999 3.32 pH	Apple, honey, clean, alcohol 0.999 3.40 pH

Comments on Fementation

I really got my nose in during the fermentation to smell for aroma. Interesting, no Go-ferm had some phenolics at day 2, but it quickly cleaned up. The 1.25 g/gal Go-ferm has pretty clean aroma descriptors throughout, whereas the higher pitch rates did have yeastiness described at some point. The 3.5 and 4.5 pitch rates finished fermentation on day 5 but the lower pitch rates/ Go-ferm finished on day 8. I am a bit surprised that the first three finished at the same time.

Secondary

• Most meads had dropped clear so transferred directly from primary to bottling
• Carbonated with 3.6 grams/ 500ml of clover honey

Tasting Notes

Meads were tested at 4 months. Judges were blind to the treatment, and the meads were served all at once in random order in identical looking cups.

Judges were asked to provide brief tasting notes and rank from best to worst.

Time	2.5 g pitch, no go-ferm	2.5 g pitch, 1.25 g/gal go-ferm	2.5-gram pitch rate w go-ferm	3.5-gram pitch rate w go-ferm	4.5-gram pitch rate w go-ferm
Judge 1	Clean, apple, slightly lager-like   Ranked 2nd	Apple, clean   Ranked 1st	Apple, clean, slightly bitter   Ranked 3rd	Tart, yeasty, bitter   Ranked 4th	Tart, bitter, Umami   Ranked 5th
Judge 2	Flat, not tart   Ranked 3rd	Balanced, neutral   Ranked 1st	Balanced, neutral   Ranked 2nd	Yeasty, tart   Ranked 4th	Really yeasty, tart   Ranked 5th

Both judges ranked the 1.25 g/gal Go-ferm first in preference. This amount is consistent with the Scott Labs 1 gram per gallon pitch rate recommendation at this ABV level for wine yeasts. Judges were split between the no Go-ferm and the 2.5 gram pitch rate with the recommended amount of Go-ferm. The first judge thought the no Go-ferm was slightly better then the 2.5 gram pitch rate with the recommended amount of Go-ferm, whereas the other judge ranked those two in reverse order. The mead without the Go-ferm was also described as flat and being slightly lager like, which I interpret as a very slight sulfur character. The 3.5 and 4.5 grams per gallon pitch rate was described as bitter, yeasty, and umami flavors. Bitterness was also described for the 2.5 gram pitch rate with the recommended amount of Go-ferm by one Judge. Interestingly, tartness was described for the high pitch rates despite them finishing at a higher pH. I wonder if the meads picked up souring bacteria because of high residual YAN. It would be interesting to test the YAN and organisms hypothesis in a lab.

Final Notes

In the high vs low pitch rate with Go-ferm experiment, it was found that a five gram per gallon pitch rate with the recommended amount of Go-ferm resulted in significant off-flavors when compared to a 1 gram per gallon pitch rate. Here, the evidence suggests that the pitch rate threshold for the off-flavor occurred at pitch rates above 2.5 grams per gallon.

A couple of things I would do differently next time. First, I provided a medium level of YAN. However, another experiment showed that S-04 presented better with a low-level of compared to a high level of nutrients. I would like to try this again if it is found that a low nutrient requirement is also preferred to medium levels of YAN. Second, I gave four nutrient additions before 48 hours, and the experiment on nutrient timing that was done after showed that three additions are preferred to four.

Take away: for me, 1.25 g/gal Go-ferm showed preference over no Go-ferm and using the recommended amount for Go-ferm at normal ale yeast pitch rates and higher rates. It seems that at pitch rates above and maybe around 2.5 grams per gallon you can start to detect off-flavors from Go-ferm. Now, for my session meads recipes, I recommend adding 5 grams of Go-ferm in my 4 gallon batches.

These findings provide continued evidence that online calculators need to change. The MeadMakr BatchBuildr recommends rounding to 5 gram packets and adding 6.25 grams of Go-ferm per gallon for one gallon batches at this ABV level which is clearly inconsistent with recommendations and these findings. Moreover, allowing for high pitch rates and still recommending the standard amount of Go-ferm should probably come with a warning on Mead Made Right (current pitch rates range from 1-10 grams per gallon of dry yeast). Both of these websites are great, I use them and recommend them all the time, but there is still room for improvement.

To follow up, this experiment should be repeated with other yeasts, with fully organic nutrients, and at different ABV levels. My hypothesis for fully organic nutrients would be that the pitch rate threshold is even lower because Go-ferm already contains organic sources nutrients.