Fermentation Temperature Experiment: Results

This post has been a long time coming. Right before I left NYC for Pennsylvania, I had the opportunity to teach one more yeast class and brew an experiment beer to share. The focus of this experiment was the effect of fermentation temperature on the final product – beer. My previous post a couple of months ago,  which describes the effects different fermentation temperatures can have on beer, can be found here. To summarize, high uncontrolled fermentation temperatures can lead to several faults in beer, including increased fruity esters, thin and watery beer, and over attenuation. Too cold and you risk a sluggish fermentation and poor flavors.

While temperatures effects on beer is well-known, I wanted to see for myself and it provided a great teaching point for the yeast class. A Belgian blonde ale was fermented at three different temperatures: 60°F, 72-74°F, and 86°F. I chose to ferment with Wyeast 1388 since this Belgian ale is characterful and has wide suggested fermentation temperature range. Each beer was bottle condition for two weeks and served in the class. Each student was instructed to sample the beer and score what flavors he/she thought was present (no limit to score).

Results:

The results were somewhat predictable, but there was some surprises as well. As expected, the beer fermented at 86°F was the fruitiest beer, although not by much since the strain itself produces fruity beers. The warmer fermented beer had the highest score for being thin/watery and almost everyone noticed the green apple aroma and taste as well. Surprisingly, the lower temperature fermented beer was extremely bitter and was picked up by most people in the yeast class. Phenolics (spicy/band-aid/sharp) and sulfur was also attributed to the beer fermented at 60°F. The control beer, which hit the suggested fermentation temperature range at 72°F – 74°F was the maltier of the three and had slightly more fusel alcohols.

I decided not to measure gravity over time since I did not have the time (I was away for three days in the middle of fermentation). However, the biggest surprise came when I checked the final gravity:

• 60°F: 1.009
• 86°F: 1.006
• 73°F: 1.001

Conclusions and Discussion:

The predictions of high fermentation temperatures held true: the beer had more esters (fruity), was thinner, and had more green apple flavor (indicative of acetaldehyde). With higher temperatures comes higher metabolic activity. More glucose (which originally came from other worts sugars such as sucrose and fructose) is pushed into the glycolytic pathway to produce excess pyruvate. The pyruvate, if not completely used in yeast biochemistry to produce ethanol, will breakdown into acetaldehyde and leak out of the cell (this can sometimes break down to diacetyl as well). More acetyl CoA is also produced which helps with the formation of esters by bonding with an alcohol. The 86°F beer also had a vigorous fermentation, which undoubtedly helped scrub the beer into a thinner form and reduce bitterness. As I bottled the 60°F sample, it was still a hazy from yeast in suspension, indicative of a sluggish fermentation. Surprisingly, this beer was noticeably bitter and had a unique sulfur component to the nose. My running hypothesis is that the fermentation was not vigorous enough to blow off any residual sulfur generated during fermentation. This often occurs with lager yeasts and why lagering needs time and proper attention to temperatures. The control beer was spot on for a clean Belgian pale ale. The slight fusel score may be indicative of strain itself.

On an interesting note was the odd finishing gravities, with the control beer almost completely fermenting out while the other two beers did not. I was expecting to have a lower gravity in the 86°F sample compared to other brews. It’s tough for me to explain this, but my gut feeling is the both the colder and warmer beer are stressed to the point that fermentation stops prematurely. This suggests that the temperature range given by the manufacture holds true as the optimal temperature range. Having said that, the class unanimously agreed that all three beers were good and drinkable. While most people could pick out the higher temperature beer, most students had trouble identifying the control beer from the 60°F (including myself). Similar to using different pitching rates, this suggests that playing around with vastly different fermentation temperatures represents another tool in the brewer’s toolbox. Of course, this statement only holds true if you adhere to healthy yeast and pitching principles.

Future:

Ah… So much to do. Similar to my pitching rate experiment, it would be great to have a panel of different yeast tested at different temperatures. Although interesting, at some point I would be re-inventing the wheel by repeating what major yeast manufacturers have already tested. More interestingly, I would like to see if pitching yeast at different temperatures has any effects. For example, a yeast starter that is cold versus warm. Another idea would be to keep the pitching yeast constant in temperature but vary the wort temperature. Pitching into warm versus cold wort for example. Lastly I would like to point out that these experiments, although interesting, are not very quantitative. Putting some actual metrics on some of these results is something I need to do, but now they reflect small-scale experiments that non-scientific audiences can sink their teeth into. For those interested in serious data analysis in homebrewing experiments, definitely refer to Braukasier’s blog. Some of his experiments are fascinating and he has the statistics to back it up.