Every good experiment begins with a question. A chance conversation with a professional mead maker prompted this one. He was concerned about shelf stability and had looked into pasteurization. We talked about some of the research and anecdotal evidence. Pasteurization has been part of commercial-beer brewing for quite a while, and even some craft breweries are using it. Shelf stability is not typically an issue for homebrewers, but it might be helpful for wild beers. But how would it affect the beer?
Most of the online literature about beer pasteurization dates back to a 1955 study, which identified that American commercial breweries at the time applied 2.4 to 45.6 pasteurization units (PUs) to their beers, with an average of 14.8 PUs. Pasteurization units are used to rate pasteurization effectiveness; a single PU is equivalent to holding the beer at 60°C (140°F) for one minute. There’s a formula that maps time and temperature to PUs:
PU = t x 1.393 ^ (T - 60)
where t is the time in minutes and T is the temperature in Celsius.
References don’t offer much detail, but this gives a feel for the target range of time and temperature to apply.
While pasteurization might be most useful for wild beers, they didn’t seem like a good target style for this test. They can have a lot going on, and bottle-to-bottle variation might make it harder to judge the impact of pasteurization. Instead, I chose two unpasteurized commercial beers with a clean flavor and distinctive character: Sierra Nevada Pale Ale and Deschutes Fresh Squeezed IPA.
To maximize the chance of detecting an effect, I intentionally went well over the target range from the 1955 study. I held the beer at 66°C (150°F) for 20 minutes, which is equivalent to 128 PUs.
To evaluate the beers, I ran a triangle test with a panel of four skilled and unskilled judges. I also tasted the beers separately by myself, to see what I could taste knowing which beers were treated.
Pasteurizing the beer at home was easy. I heated a large pot of water to 76.5–82°C (170–180°F). I placed bottles of each beer into the pot so the water was above the level of the beer in the bottle. I also placed an open test beer in the pot that I could use to track the beer temperature. I probably could have saved some beer by filling a beer bottle with water, but it was a sacrifice for science.
I let the beer temperature rise to 66°C (150°F), and then pulled the pot off the burner. The beer stayed at that temperature without any additional effort. I let it sit for 20 minutes, then removed the bottles from their water bath.
I let the bottles cool to room temperature, then refrigerated them and their untreated siblings. The treated bottles were marked, so I had no trouble keeping track of them.
Safety notes: Protective equipment (gloves, eyewear) is a good idea. Also, do not place the hot beers directly into the refrigerator to cool because the bottles may shatter.
The judging was arranged as a pair of triangle tests. The first round was the Sierra Nevada Pale Ale and then they had the Deschutes. For each test, every judge was presented with three uniquely tagged glasses of beer. Two of the three were identical, but each judge got his/her own set in random order, so one might have two pasteurized beers while another had only one. I literally rolled dice to determine the composition and order of each evaluator’s set.
The judges were given a form that asked which two beers were the same, what differences they found, and which they preferred. I didn’t tell them what the beers were or that we were testing the effects of pasteurization.
All four judges thought that it was particularly hard to differentiate the odd beer with the Sierra Nevada. Even knowing it, I couldn’t tell. Two judges correctly guessed, which is better than random, but they weren’t very confident, and their comments didn’t quite sync up. The BJCP National judge thought both were equally bitter, but the pasteurized version had a fuller, maltier aroma, while an untrained judge thought that the pasteurized one was less bitter. They both preferred the pasteurized beer.
The judges were more confident about their choice with the Deschutes Fresh Squeezed IPA, but it didn’t change the number of correct guesses: still two out of four. In this case, the amateur judge who guessed correctly thought the pasteurized beer was maltier. The BJCP judge noticed a richer floral citrus hops aroma on the unpasteurized version. Still, they both liked the pasteurized beer better. My own observations were in line with the BJCP judge’s take: I liked the fuller hops and citrus expression on the unpasteurized Deschutes.
It was also interesting that the same untrained judge did the best, picking out the odd beer correctly each time.
With only four judges, it’s unreasonable to expect conclusive results. Going into this, I would have guessed that pasteurization would affect aroma the most and that the treated beer would lose subtlety. That seemed to be truer with the Fresh Squeezed IPA, but the Sierra Nevada Pale Ale apparently had a better aroma afterward. Interestingly, the pasteurized beers were perceived as maltier, so it may affect hops character more heavily.
Applying 128 PUs was fairly extreme, and it does seem to have generated a real difference. I’d like to run the test again, aiming for 30–45 PUs to see whether the Deschutes would still stand out. I’ll also ensure a larger sample size next time, perhaps seeing if judges can repeat their guesses in subsequent rounds.