Flemish Red-Brown Beer

Flanders Red Ale is a classic beer style that is produced only by a small handful of breweries in West Flanders, Belgium today. Some American breweries aim to produce comparable beers . These beers, red to brown in color, are characterized as being sour and sometimes sweet, with malt flavors and fruity complexity from the mixed fermentation and hints of oak. They have often been described as "wine-like", and have previously been distinguished from their close cousin, the Oud Bruin, brewed in East Flanders. Rodenbach is the most well known Flanders red producer . Classic Belgian examples of Flemish red have been flash pasteurized or sterile filtered and are not alive in the bottle (~35 minutes in) (also ref Mad Fermentationist dregs list).

Style Guidelines

BJCP Guidelines

See category 23B. Flanders Red Ale.

Brewers Association Guidelines

Belgian-Style Flanders Oud Bruin or Oud Red Ale are copper to very dark. SRM/EBC color values can be misleading because the red spectrum of color is not accurately assessed using these procedures. Chill haze is acceptable at low serving temperatures. Some versions may be more highly carbonated and, when bottle conditioned, may appear cloudy when served. Roasted malt aromas including a cocoa-like character are acceptable at low levels. Brettanomyces produced aromas may be completely absent or very low. Fruity-ester aroma which is often cherry-like is apparent. Hop aroma is not perceived. Roasted malt flavors including a cocoa-like character are acceptable at low levels. A very low degree of malt sweetness may be present and in balance with the acidity produced by Lactobacillus activity. Hop flavor is not perceived. Hop bitterness is perceived to be very low to medium-low, though acidity and wood aging (if used) may mask higher bitterness unit levels. Overall balance is characterized by slight to strong lactic sourness, and with "Reds" sometimes a balanced degree of acetic acid. Brettanomyces produced flavors may be absent or very low. Fruity-ester flavor which is often cherry-like is apparent. Oak-like or woody characters may be pleasantly integrated into overall palate. Residual wine or distilled spirits flavors associated with used barrels should not be evident. Bottle conditioned versions are often blended old with new before packaging in order to create the brewer’s intended balance of characters. Body is described as a refreshing mouthfeel.

• Original Gravity: (ºPlato) 1.044 - 1.056 (11 - 13.8)
• Apparent Extract/Final Gravity: (ºPlato) 1.008 - 1.016 (2.1 - 4.1)
• Alcohol by Weight (Volume): 3.80% - 5.20% (4.80% - 6.60%)
• Bitterness (IBU): 5 - 18
• Color SRM (EBC): 12 - 25 (24 - 50)

Microbes and Flavor Compounds

Initial Study

Interestingly, and perhaps frustratingly, Flanders red and brown ales have had far fewer published studies than Belgian lambic beers. The following information is based off of the Martens et al. study from 1997 (see references).

Introduction

While most beer styles are fermented using one culture of Saccharomyces cerevisiae or S. pastorianus, Flanders Red Ales are fermented with a mixed culture fermentation. At one brewery (presumed to be Rodenbach) studied by Martens et al., two beers were produced using mixed fermentation and blended together. The first "light beer" was 11°P and was less acidic, while the second "heavy beer" was 13°P and served unblended as an Old Ale. Both beers were inoculated with an acid washed yeast slurry that was harvested from a previous fermentation of the "light beer". The yeast slurry contained about 5% lactic acid bacteria after the acid wash. The fermentation of these beers had three stages:

1. A seven day ethanol fermentation dominated by Saccharomyces.
2. A four to five week lactic acid fermentation dominated by Lactobacilli.
3. A twenty to twenty-four month fermentation dominated by Brettanomyces, Lactobacilli, Pediococcus, and acetic acid bacteria (Acetobacter).

The development of the third stage with Brettanomyces and Pediococcus was similar to the development of these microbes in Lambic fermentation. The "light beer" was never allowed to go through the third phase of fermentation, and was instead chilled to 0°C and then used to blend with previous batches of the "heavy beer" .

Primary Fermentation

In the brewery studied by Martens et al., the "light beer" was inoculated with a harvested yeast slurry of multiple strains of S. cerevisiae at a rate of 8x10⁶ CFU/mL, and the "heavy beer" was inoculated with 1x10⁷ CFU/mL. Small numbers of Candida guillermondii and Candida datilla were reportedly in the yeast slurry, but their identification was questioned in the study and they were not found during primary fermentation. One interesting finding was that the S. cerevisiae strains used at this brewery formed sexual spores (ascospore), which is quite unusual for brewing yeasts. The yeast in the "heavy beer" grew slower (3 days) and reached an overall cell count that was lower than the "light beer", which reached maximum cell count in 2 days. Yeast slurries with more lactic acid bacteria are generally used to inoculate the "heavy beer", and this may retard the yeast growth in the "heavy beer". The harvested slurry is always taken from the "light beer", which may be less adapted to the fermentation of the "heavy beer". Also, after 1 week the yeast flocculated and settled out better in the "light beer" than they did in the "heavy beer". Although lactic acid bacteria were in the yeast slurry, their growth started only after 4 days into the primary fermentation. No enterobacteria or acetic acid bacteria were found during this first phase of fermentation .

The lactic acid bacteria found in the yeast slurry consisted of 18 strains of L. delbruekii ssp delbruekii and 12 strains of L. delbruekii ssp bulgaricus. Small amounts of Pediococcus were also identified in the slurry, but were impossible to isolate with enrichment until they were found in the primary fermentation. During primary fermentation, one strain of L. plantarum, two strains of L. brevis, and one strain of Pediococcus parvulus were identified .

Secondary Fermentation

After the 7 days of the primary fermentation, the beer was transferred to a secondary fermenter and remained there for four to five weeks. Both the yeast and bacteria populations saw a decline during the transfer, and then a small and gradual growth in secondary with the final yeast count being 4.3x10⁵ in the "heavy beer" and 6.4x10³ in the "light beer". Lactic acid bacteria grew much faster and became dominate in the "light beer", whereas in the "heavy beer" they grew more slowly and yeast remained the dominate microbe. This was explained by there being more sugars in the "heavy beer", which gave the yeast the advantage. Acetic acid bacteria were still not detected during secondary fermentation . Lactic acid began to be produced as well during secondary fermentation, with about a third of it being L-lactic acid and two thirds of it being D-lactic acid .

During secondary fermentation, L. delbruekii ssp delbruekii strains dominated over the other lactic acid bacteria found. Additional strains of L. plantarum and L. coryneformis, and an additional strain of L. brevis was found in the "heavy beer". Other than the dominating L. delbruekii ssp delbruekii strains, only a few strains of L. brevis were found in the "light beer" . The "light beer" appeared to have a a smaller diversity of microbes.

Tertiary Fermentation

The same study by Martens et al. looked at two casks during the third fermentation. The "heavy beer" was transferred from the secondary fermentation vessel to the casks to age for two years. At the beginning of the third phase of fermentation, Saccharomyces cell counts began to drop while the appearance of Brettanomyces began. After 10 weeks in the casks, Lactobacilli greatly decreased, giving rise to strains of Pediococcus parvulus. After 12 weeks for Cask A and 18 weeks for Cask B, the beer no longer contained Saccharomyces, and Brettanomyces dominated. Specifically, B. lambicus (now classified as a strain of B. bruxellensis ) and B. bruxellensis were the dominate species, but much smaller counts of B. intermedius (now classified as B. anomala) and B. custersianus were also found. Brettanomyces continued to be the primary microorganisms for 36 weeks in Cask A and 50 weeks in Cask B. After this time period, P. parvulus began to dominate. Acetic acid bacteria also began to make an appearance in the casks, being detected at 27 weeks in Cask A and 40 weeks in Cask B. The exact numbers of the acetic acid bacteria were not reported by this study since they mostly grow on the surface of the beer inside the cask, and possibly on the walls of the cask where diffused oxygen could reach them more easily, and samples were not taken from these sections of the casks. The difference in the time periods for the microbial populations was determined to be affected by the casks themselves, which differed in age, size, and possibly different microbe colonization inside them before they were filled . During the third fermentation phase, lactic acid greatly increased from ~600 ppm to ~4500 ppm after 35 weeks, and continued to slowly increase to ~5200 ppm at 60 weeks, at which time L-lactic was only slightly less (~48%) than D-lactic acid . Acetic acid levels reached 1600 ppm by the end of the third phase. The final pH of the "heavy beer" was around 3.2-3.5 .

Comparison to Lambic

Lambic is a similar beer produced in Belgium, but is fermented using spontaneous fermentation. Enterobacteria were not found in the Flanders red brewery that Martens et al. studied since spontaneous fermentation was not used. However, after the enterobacteria and primary Saccharomyces fermentation phases of lambic brewing are complete, the microbial populations of lambic and Flanders red/brown beers are similar during their aging processes. Both beers display a dominance by Brettanomyces and Pediococcus during the aging phases. Flanders Red Ales differ by having a large portion of the acid production performed by Lactobacilli, where as in lambic the acid production is performed by Pediodoccus damnosus. Flanders red ales are also characterized by having P. parvulus instead of P. damnosus (although this may have been misidentified; see Modern Analysis below), however Martens et al. noted that the two species have no clear difference as far as their effects on fermentation go. Martens also noted that Brettanomyces began to disappear from old English Porter when the beer moved from wooden casks to metallic ones. It is thought then that the wooden casks are vital to Brettanomyces in Flanders Red Ale brewing, perhaps due to the presence of Pediococci, with which Brettanomyces "cooperates" to ferment dextrins in the beer during the aging phase .

Belgian brewers have even married the Flanders red ale and lambic by blending the two beers together. The Flanders "acid beer" is fermented with cherries, and later on lambic is added and the blend is allowed to referment in the bottle, creating something truly special .

Modern Analysis

A more recent study by Snaewaert et al. (see reference ) looked at the microbial and metabolic composition in the finished beer of the same brewery as Martens et al., as well as two other Flanders red ale breweries using "state of the art" DNA sequencing methods. Three samples were analyzed (A, B, and C) from each brewery (1, 2, and 3). As expected, there were both similarities and differences between the three breweries tested, as well as some differences between the individual beers from each brewery.

Microbial Populations

Each brewery had their three samples compared, and then also compared to each other. The bacterial populations for Breweries 1 and 3 were similar across all of their own samples, whereas Brewery 2's samples had a larger difference between the three samples of its own beers. The yeast populations were similar between all samples for Brewery 1, whereas Breweries 2 and 3 had a larger difference in yeast populations between their three samples of beer . In summary, Brewery 1 (presumed to be Rodenbach) had similar populations of bacteria and yeast across all samples of their beer, whereas Brewery 3 only had similar populations of bacteria but not yeast across all samples of their beer, and Brewery 2 had the least amount of similarity in yeast and bacteria populations between their beer samples. This indicates that the beers from Brewery 2 were less consistent.

Overall, the samples were mostly dominated by Pediococcus damnosus and Brettanomyces bruxellensis, except for two samples of Brewery 2 that were dominated by Acetobacter, and one sample from Brewery 2 and two samples from Brewery 3 that were dominated by an unidentified yeast. Note that these findings conflicted with the findings of Martens et al. who identified P. parvulus as the dominating Pediococcus species. These two species are closely related, and an explanation of this discrepancy between the two findings was never found. Acetobacter numbers were higher in Brewery 1 and 2, and lower in Brewery 3, and were associated with the high acetic acid amounts found in the beers. Smaller amounts of Acetobacter in Brewery 1 were identified as Acetobacter pasteurianus, and might be the same species in Breweries 2 and 3. Small numbers of Lactobacillus were found in all samples. Weissella and Leuconostoc were found in some samples from Brewery 2, whereas only trace amounts of Wessella were found in one samples from Brewery 3. A significant amount of Pichia yeast were found in two of Brewery 2 samples, and trace amounts in the third sample from that brewery. Only trace amounts of Pichia yeast strains were found in Breweries 1 and 3. Brewery 1 had some amount of Candida yeast in all samples. Brewery 3 had some amount of Candida in one sample, but only trace amounts in the other two samples, and Brewery 2 had only trace amounts of Candida across all three samples. Although detected with DNA sequencing methods, Lactobacillus and Candida along with other yeast could not be isolated and grown under lab conditions .

Only one OTU was assigned to Kregervanrija, Debaryomyces, Priceomyces, Hyphopichia, and Wickerhamomyces yeasts, consisting of 0.24, 0.08, 0.07, 0.03, and less than 0.01% of the DNA sequencing reads, respectively . (This wiki entry needs clarification.) This means that only an extremely small amount of these microbes were present.

Metabolic Composition

Snaewaert et al. also looked at the metabolic composition of finished beers in the three breweries before they were packaged. Glucose was completely gone from the samples, but substantial concentrations of maltose, maltotriose, maltotetraose, maltopentaose, and maltohexaose were still present. This contrasts with the relatively high glucose and fructose found in finished bottled versions of these beers, which indicates that the beers were back-sweetened with young beer or with residual sugar or even possibly some form of sugar at bottling time .

Overall the flavor of these beers was dominated by ethanol, lactic acid, acetic acid, ethyl acetate. Across the samples there was also a presence of isoamyl alcohol (31-150mg/L) and isoamyl acetate (1.99-6 mg/L), and an absence of 2-phenyl ethanol and 2-phenylethyl acetate in both the matured beers and the bottles versions. Small amounts of propionic acid, isobutyric acid, ethyl hexanoate, and ethyl octanoate were found. Higher levels of ethyl acetate were found compared to the Martens et al. study, and no ethyl decanoate was found, which is a typical ester found in gueuze. Just as the microbial populations of Brewery 2 differed from Breweries 1 and 3, so did its overall metabolite content. Higher levels of acetic acid were found in Brewery 2, which was attributed to high levels of Acetobacter populations. Additionally, all three of the samples from Brewery 2 differed from each other as far as metabolic content, which was also attributed to the microbial population differences between each of Brewery 2's samples .

See Also

Additional Articles on MTF Wiki

• Oud Bruin
• Mixed Fermentation
• Blending
• Lambic
• Saccharomyces
• Brettanomyces
• Lactobacillus
• Pediococcus

External Resources

• Rudy Ghequire from Rodenbach on the Sour Hour
• Interview with Rudi Ghequire on Belgian Smaak podcast.
• Designing and Brewing a Flanders Red Ale, by Matt Miller of Sour Beer Blog.
• CRAFT Beer: Quantitation of hte Major Components of Commercially Available Sour Beers Using Time-Domain CRAFT Analysis of ¹H NMR Spectra.