100% Brettanomyces Fermentation

From Milk The Funk Wiki
Revision as of 02:24, 10 June 2019 by DanABA (talk | contribs)
Jump to: navigation, search

100% Brett fermentations are beers that are fermented with only Brettanomyces and no other microbes such as S. cerevisiae, Lactobacillus, or Pediococcus.

General Information

The method of fermenting wort with only Brettanomyces was pioneered by Tomme Arthur from Pizza Port/Lost Abbey, and Peter Bouckaert from New Belgium in 2004 with their 100% Brettanomyces fermentented Mo' Bretta, and Vinnie Cilurzo of Russian River with Sanctification later that same year [1]. Avery Brewing Co. and Jeff O'Neal from Ithica Beer Co. also produced early 100% Brettanomyces beers [2]. This method was further popularized by Chad Yakobson's Brettanomyces Dissertation on the Brettanomyces Project blog, and by his brewery, Crooked Stave Artisan Beer Project. While primary fermentation with Brettanomyces is a complex subject due to the wide range of characteristics of different species and strains of Brettanomyces, in general beer that is fermented with Brett in primary usually produces a surprisingly clean, lightly fruity beer (see Chapter 8 in American Sour Beers by Michael Tonsmeire for a full description of 100% Brettanomyces fermented beers). For this reason, wort that is characteristic of American IPA has been found to create wonderful beers when primarily fermented with Brett.

Typical characteristics of Brett primary fermentations (these are generalizations, and may not be true for every strain):

  • Initially subdued "horsey", "funky", "barnyardy" flavors due to the lack of Saccharomyces esters/phenols (see the Brettanomyces Metabolism page for more information). However, this is a generalization and some brewers have reported getting some "funkier" flavors out of some strains.
  • Light fruit characteristics.
  • A longer lasting hop aroma and flavor due to Brett's ability to constantly metabolize micro-oxygenation.
  • A lack of glycerol, which is a compound that Saccharomyces produces which gives beer it's slick mouthfeel. Malts such as oats or flaked wheat are often used to make up for the lack of glycerol. However, the role of glycerol in creating mouthfeel is debatable in the wine world [3].
  • Slightly longer primary fermentation in general (3-6 weeks), although some people have reported faster fermentations between 1-3 weeks for some strains and conditions (lower starting gravity beers, for example) [4].
  • Perceived bitterness may be quite a bit lower than the same wort fermented with a clean ale yeast.

Brewing Techniques

Obtaining a clean culture and true attenuation ability

Please note that many of the Brettanomyces cultures sold are contaminated with Saccharomyces [5][6]. Generally, Brettanomyces ferments slow and a fermentation could take considerably longer to ferment out compared to a fermentation containing Saccharomyces. Besides Brettanomyces has a somewhat limited metabolism and the apparent attenuation would be quite a bit lower compared to what a comparable Saccharomyces fermentation would showcase. Additionally, many strains of Brettanomyces, especially B. anomalus, cannot efficiently ferment maltose in brewing conditions, and therefore are not good candidates for 100% Brettanomyces beers (see Brettanomyces carbohydrate metabolism for details). If a Brettanomyces culture fully ferments out a beer in less than a month, then it may have a Saccharomyces contamination, however there are exceptions to this (see Fermentation Characteristics below).

Starter Information

When relying on a Brettanomyces culture for primary fermentation, a starter will often be necessary due to the fact that most yeast labs provide a small cell count for their Brettanomyces cultures. See the Brettanomyces Starter Information section for more information on Brett starters. About 500ml starter per 25 liters of wort seems to be the current best practice. Data from Thomas Hübbe supports that the initial pitching rate doesn't have a great effect on the final cell count in pure Brettanomyces starters or beer, indicating that Brettanomyces is fairly forgiving in regards to small initial cell counts [7].

See also Brettanomyces pitching rates.

Wort Production

American IPA or American Pale Ale recipes are a tried and true general approach to making wort that is favorable to 100% Brett fermentations. Fruitier hops such as citra, amarillo, galaxy, etc. tend to compliment the light fruity characteristics of a Brett primary fermentation. The addition of body-increasing malts such as oats, unmalted barley, rye, wheat, or carapils may assist with the lack of glycerol that is typical for Brett [8], but isn't always necessary. Otherwise, wort production can remain the same as it is for an American IPA/Pale Ale recipe. Aeration of the wort before fermentation starts should be done. This will greatly increase cell growth (see the Brettanomyces Propagation Experiment). As far as we know, acetic acid is a byproduct of ethanol production by Brettanomyces and not the prior lag phase, so as long as ethanol is not already being produced then acetic acid production is not a concern [9]. Examples of commercial 100% Brettanomyces beers that receive the same amount of initial aeration that would be typical of ales of their respective gravities are "Sanctification" from Russian River and "Mo’ Betta Bretta" from Lost Abbey [10].

Fermentation Characteristics of Individual Species and Strains

Not all species of Brettanomyces are effective at efficiently attenuating wort on their own. Additionally, some strains and species may produce better results flavor-wise than others.

  • Some microbiologists have witnessed that B. claussenii is very slow to ferment wort by itself. If fermentation finishes in two weeks, this might be due to contamination of another yeast [11][12].
  • Not all strains can ferment maltose, which is almost 50% of the sugar composition of wort. These strains should be avoided for 100% Brettanomyes fermentations. See Brettanomyces carbohydrate metabolism for more details.
  • Chad Yakobson's thesis showed that WLP645, WLP650, WLP653, WY5112, WY5526, and WY5151 were not able to attenuate wort more than 50% within 35 days (these were pure cultures). BSI Drie was the only strain tested that was able to attenuate wort at levels similar to brewers yeast. All strains that he tested were able to utilize maltose, however some less efficiently than others. More time may or may not have resulted in further attenuation. Contamination with another yeast is one explanation for why brewers are able ot use these cultures from labs to fully attenuate wort (Yakobson used purified isolates for this research).
  • Nick Mader of Fremont Brewing (2017 Master Brewers Conference Presentation) observed that 100% BSI Drei fermentation resulted in around 77% attenuation (3.17°P final gravity), while co-fermentation with different pitch rates of a saison yeast resulted in around ~90% attenuation (~1.5°P final gravity). The esters were generally lower than when cofermented with the saison yeast, but ethyl decanoate (apple, brandy) was considerably higher with the 100% fermentation with BSI Drei. 4-ethyl phenol concentrations with 100% Drei were around the same as when cofermented with the saison yeast. See also cofermentation with Saccharomyces.
  • Mark Trent's Brettanomyces propagation experiment tested his house strain of Brettanomyces (originally isolated from Orval), which fully attenuated wort under different different conditions within 6 days. So, there are strains that are faster fermenters, but they appear to be the exception to the rule.
  • In general a broad selection of various Brettanomyces yeasts and a few months of time is a safe bet to make sure fermentation carries through.

Impact of Fermentation Temperature

In Saccharomyces species, higher fermentation temperature has been associated with faster fermentation, higher growth rates, and ester formation. Tyrawa et al. (2019) set out to explore the impact of fermentation temperature on 7 beer strains (including BSI Drei and several isolates from commercial sour/saison/lambic beers) and 2 wine strains of B. bruxellensis (Fermentis US-05 was used as a control). Each strain was fermented in autoclaved 100% 2-row malt wort at a starting gravity of 1.050 and fermented at 15°C (59°F) versus 22.5°C (72.5°F) for 28 days. The pitching rate was 1.2 x 107 cells/mL. Each strain was genetically tested to ensure their species was correctly identified and that they were genetically distinct from each other [13].

Their results showed that there is a vast diversity in how temperature effects attenuation for different strains of B. bruxellensis. In general, the cooler 15°C (59°F) fermentation temperature slowed the attenuation rate for most strains. The US-05 attenuated the most at both temperatures, with only one saison strain matching that attenuation level when fermented at 22.5°C (72.5°F). This same strain, which was isolated from a commercial USA saison beer, and the BSI Drei strains had fast attenuation rates that were comparative to the US-05 fermentation at both temperatures, while the other strains had lag times of 8-10 days at 15°C (59°F) or 2-4 days at 22.5°C (72.5°F) [13].

At 15°C (59°F), none of the Brettanomyces strains could match the US-05 attenuation, with most of them falling to around 50-25% less final attenuation after 28 days, and one of the wine strains and one of the USA saison strains fermented almost nothing at all. Still, this data showed that some beer strains of B. bruxellensis can ferment at lower temperatures. Interestingly, one of the wine strains was almost unaffected by the difference in fermentation temperature; it only lagged for a couple of days longer in the colder 15°C (59°F) fermentation temperature versus the warmer 22.5°C (72.5°F) fermentation temperature [13].

At 22.5°C (72.5°F), all of the Brettanomyces strains fermented much better, although their final attenuation numbers for some strains were significantly less than other strains and quite varied, with only one strain (the previously mentioned strain that was isolated from a commercial USA saison beer) attenuating at levels that matched the US-05 control. This indicates that most B. bruxellensis strains are not as efficient at fermenting wort by themselves as Saccharomyces cerevisiae ale strains [13].

The effect on phenol production, 4-ethylguaiacol (clove) and 4-ethylphenol (barnyard), was relatively the same and above flavor threshold for both fermentation temperatures for all of the B. bruxellensis strains tested, although some strains had slightly more or less of these phenols produced at the different fermentation temperatures. The temperature of the fermentation had a larger impact on the amount of esters produced. Ethyl acetate (pineapple/pear) was significantly higher in the warmer fermentation temperature of 22.5°C (72.5°F) than the cooler temperature of 15°C (59°F) for all strains, with one saison strain producing significantly more ethyl acetate and another saison strain producing significantly less ethyl acetate than the other strains. As expected, the US-05 produced higher amounts of 4-vinylguaiacol (clove) and isoamyle acetate (banana) at 22.5°C (72.5°F) and lower amounts at 15°C (59°F). The US-05 produced comparably high amounts of phenethyl alcohol (dried rose), phenethyl acetate (honey/rose pedal), and isoamyl alcohol (banana/oily) at both temperatures. These were generally not produced at more than low levels by the Brettanomyces strains, except for 4-vinylguaiacol which was produced more at the lower temperature by BSI Drei and one of the saison strains, indicating that the lower fermentation temperature slowed the process of these strains to convert the 4-VG to 4-EG, and they also produced the lowest amount of the ethyl phenols compared to the other Brettanomyces strains. The Brettanomyces strains produced other fatty acid esters at significant levels above tastes threshold that the US-05 produced below tastes threshold. These esters include ethyl caproate (pineapple/apple), ethyl caprylate (pineapple), ethyl decanoate (brandy/apple) and ethyl nonanoate (fruity/rose/waxy). In general, a higher amount of these esters were produced at the higher fermentation temperature, although there were exceptions. Several of the saison strains and the lambic strain produced higher amounts of esters than the BSI Drei control, especially when fermented at the warmer temperature, demonstrating the the amount of esters produced is highly variable amount different strains of B. bruxellensis, particularly when fermented at 22.5°C (72.5°F) rather than the lower fermentation temperature of 15°C (59°F). Interestingly, the two wine strains of Brettanomyces bruxellensis did not produce above threshold levels of any of these esters at either fermentation temperature (the wine strains produced the highest levels of decanoic acid, which were elevated at the higher fermentation temperature) [13].

Aging

In general, 100% Brettanomyces beers are not aged for more than 2-3 months (sometimes less). The beer can be packaged when it reaches a stable final gravity (see Packaging). "Brett IPA's", for example, are often not aged since this leads to a decline in hop flavor and aroma. However, there is no "rule" against aging them if the brewer chooses to. Due to the potential for acetic acid development when exposed to oxygen over time, care should be taken when aging any beer with living Brettanomyces. See mixed fermentation aging.

Styles

Questioning Conventional Wisdom

About Trois

Up until April 9, 2015, "WLP644 Brettanomyces bruxellensis Trois" was thought to be a Brettanomyces species. Following the analysis of the genetics of Trois by Lance Shaner and several other members of MTF that showed this strain to be S. cerevisiae, White Labs released a statement saying that their DNA analysis also showed that Trois was a Saccharomcyes species, but they did not specify the species of Saccharomyces [14][15]. Beer fermentations with the this strain (now labeled as "WLP644 - Saccharomyces brux-­like Trois") are no longer considered to be 100% Brettanomyces fermentations. While this strain does produce a lot of fruity esters, it does not produce phenols, which is a signifying characteristic of Brettanomyces fermentations. Trois fermentations are therefore not representative of the flavor profile of true Brettanomyces fermentations, and this has become a common misconception because of the popularity of Trois and the misclassification. See this MTF thread for links to the details about the efforts to identify WLP644 as S. cerevisiae from various independent sources.

When using WLP644, it is recommended to make a 1 liter starter for 36-48 hours due to the extremely small cell count of the vials [16].

Are 100% Brett Beers Really Cleaner?

A lot of the conventional wisdom listed above regarding 100% Brettanomyces fermentations is anecdotal information derived from Trois fermentations. As explained above, Trois is not actually Brettanomyces, and so conventional wisdom regarding 100% Brettanomyces beers has been brought into question. One particular area of question is the conventional wisdom that Brettanomyces requires phenols from POF+ Saccharomyces strains in order to convert 4-vinyl phenols into 4-ethyl phenols, and that 100% Brettanomyces fermentations are therefore "less funky".

There is surprisingly little data to back this idea up outside of the anecdotal information gathered from brewers fermenting with 100% Trois, which was once thought to be Brettanomyces [17]. One controlled experiment by Lance Shaner of Omega Yeast Labs and Richard Preiss of Escarpment Labs showed that the levels of 4-ethyl guaiacol and 4-ethyl phenol produced by Brettanomyces did not depend on the amount of their 4-vinyl precursors, suggesting that Brettanomyces is capable of producing 4EP and 4EG de novo (without being dependent on precursors produced by Saccharomyces). In addition to this, the possibility that brewers and even some yeast labs have Saccharomyces contamination issues in their Brettanomyces products complicates the issue. This is only one data point, however, and more data needs to be researched.

A study [18] conducted by Caroline Tyrawa and Richard Preiss measured, among other things, the 4-ethyl guaiacol in 100% Brettanomyces bruxellensis. It shows significant levels of 4-ethyl guaiacol in wort fermented by various strains of the before-mentioned yeast. A somewhat speculative conclusion of this might be that the high ester levels of 100% Brettanomyces fermented beers might mask the funk (4-ethyl guaiacol, 4-ethyl phenol, etc) produced. As esters tend to be chemically unstable (ref?) the fruity character of a Brettanomyces beer will fade over time allowing the funk a more prominent role. This is also supported by a study that looked at 4-ethyl phenol and 4-ethyl guiacol levels in one strain of B. bruxellensis when fermented alone and when co-fermented with a wine strain (EC1118); they found that there were about 20% more phenols in the 100% B. bruxellensis fermentation than there were when the B. bruxellensis was co-fermented with EC1118 (this might have been because the wine strain uesd, EC1118, can metabolize hydrocinnamic precursors differently and reduce the 4-vinyl levels [19]) [20].

Thomas Hübbe's masters thesis also supports the hypothesis that Brettanomyces produces more esters other than ethyl acetate when it is not co-fermented with S. cerevisiae, specifically because it has better growth without competition from S. cerevisiae. Although below threshold, the esters ethyl caprylate, ethyl caprate, ethyl dodecanoate, and ethyl tetradecanoate were significantly lower when Brettanomyces was co-fermented with S. cerevisiae and Lactobacillus than when it was fermented with only Lactobacillus. Ethyl acetate (still under threshold levels) was higher when Brettanomyces was fermented with Lactobacillus but without S. cerevisiae, and significantly higher when it was fermented with both Lactobacillus and S. cerevisiae [7]. This seems to support the idea that, with the exception of the ester ethyl acetate, 100% Brettanomyces fermentations are not necessarily less phenolic, but that they are more fruity probably due to higher growth without competition from S. cerevisiae (although phenols were not measured in Hübbe's study) [21].

See Also

Additional Articles on MTF Wiki

External Resources

References

  1. American Sour Beers. Michael Tonsmeire. July 2014. Pg 189.
  2. Yakobson, Chad. Interview on Craft Commander. 12/20/2016. Retrieved 12/20/2016. (~21 mins in)
  3. Tim Patterson. "Many Roads to Mouthfeel". Wines & Vines Magazine. Nov 2009. Retrieved 03/23/2018.
  4. Conversation on MTF regarding how long 100% Brett ferments can take. 10/04/2015.
  5. MTF discussion on 2016-04-30
  6. MTF Discussion, 2016-05-19
  7. 7.0 7.1 Effect of mixed cultures on microbiological development in Berliner Weisse (master thesis). Thomas Hübbe. 2016.
  8. Conversation with Tom Belgrano on MTF. 11/12/2015.
  9. Conversation with Richard Preiss on MTF about oxygenating wort that will receive 100% Brett. 12/30/2015.
  10. BYO Magazine. Brettanomyces. Steve Piatz. October 2005.
  11. Conversation with Lance Shaner on MTF regarding B. claussenii attentuation. 04/06/2016.
  12. Conversation with Brian Martyniak regarding general Brettanomyces sugar utilization. 08/24/2016.
  13. 13.0 13.1 13.2 13.3 13.4 The temperature dependent functionality of Brettanomyces bruxellensis strains in wort fermentations. Caroline Tyrawa, Richard Preiss, Meagan Armstrong, George van der Merwe. 2019. DOI: https://doi.org/10.1002/jib.565.
  14. Archive of MTF discussions regarding Trois genetic analysis results.
  15. White Labs Blog article. April 9, 2015.
  16. Conversation with Lance Shaner on MTF. 12/30/2015.
  17. Conversation with Lance Shaner on MTF. 02/05/2016.
  18. "Funky can be Great: Brettanomyces bruxellensis Beer Fermentations" (poster for study). Caroline Tyrawa, Richard Preiss, and George van der Merwe. 2017.
  19. Richard Preiss. Statements about the Kosel et al. study. Milk The Funk Facebook gruop. 07/26/2017.
  20. The influence of Dekkera bruxellensis on the transcriptome of Saccharomyces cerevisiae and on the aromatic profile of synthetic wine must. Janez Kose, Neža Čade, Dorit Schulle, Laura Carret, Ricardo Franco-Duarte Peter Raspor. 2017.
  21. Comments by Richard Preiss regarding Thomas Hübbe's masters thesis. 09/15/2016.