For a long time I’ve been a proponent of a very simple method of adding coffee to your beer.

My process is a long cold steep – 1 cup coarse ground coffee, 2-3 cups of filtered cold water. Mix in a French press, soak for 8 to 12 hours, press, filter and add to the beer in the keg/secondary.

I do this process a few times a week because it’s a way of making coffee that my un-caffeinated brain can handle. Press, pour over ice with a little water dilute, drink, repeat until brain screams for mercy. One pot of coffee concentrate can last for several days in the fridge.

As another advantage, using a long slow cold extraction process pulls a different set of flavors and aromas than the typical high speed, hot water extraction. Hot extracts grab acidic and acrid flavors from the coffee beans and blows out much of their aroma, the cold method avoids most of the a’s in favor of extracting the more volatile essential oils. The end result is more fragrant, richer and sweeter with a potent aromatic punch. But.. is there a dark side?

I first noticed it back in January of this year when I judged an attempt at a White Stout. (A style that made Denny say “man” about 100 times as I convinced him it belonged in the book) The color was pale golden, the beer was a little turbid, but it had a wonderfully rocky head. So far, so good!

Unfortunately, immediately on putting it to my nose – all I could smell was green pepper and jalapeño. A strong vegetal note sat on top of the beer. It hid most of the coffee and cacao nib flavors that the brewer worked to put in the beer. While I can’t say for certain that the brewer used cold steeped coffee, I believe they did. (see edit below!)

So how did they get such a flavor? I decided to do some experiments. I broke out my trusty grinder and a couple of different bean/roast types. Specifically, the experiment focused on lighter roast blends vs. darker roast blends. Think Kona vs. Columbian. (For the record, I tend to prefer a medium roast for my coffee to hit the flavors of light and dark roasts)

I prepared the same coffees the same way. 1 cup coarse grounds to 3 cups water. Sit for X hours at room temp, press and sample. I repeated the experiment with the same bean types sitting for 12, 24 and 36 hours.

When I sampled the resulting concentrates, I found the culprit. Each of the lighter roasts had a lighter aromatic profile that included some faint vegetal aromas. These green aromas were barely there when the grounds were soaked for 12 hours, but as soak time was extended, the flavor grew more dominant. The darker roasts on the other hand tended to become harsher, more acidic over time, but didn’t pick up any greenness.

The worst offender in the pepper derby was a bag of Safeway Select Kona Blend City Roast. Even smelled at 12 hour mark the aroma was Jalapeñoeriffic. Of course, in retrospect, this shouldn’t be a big surprise. In a light City roast, the beans aren’t nearly as heated through as the darker roasts, leaving behind more of the green compounds. During a hot extract process, those compounds are volatilized before you ever get to sip them. The darker beans get more throughly heated, losing a number of aromatics and other compounds (including caffeine).

Given the delicate nature of aromatic compounds, it’s reasonable to assume that they would burn up in the roaster. Doing some additional reading (google “coffee green pepper“) turns up a class of chemicals pyrazines. Basically it’s a set of compounds that are structured as rings, stabilized by Nitrogen.

In coffee and wine, Methoxypyrazines convey “green” flavors, particularly 3-isobutyl-2-methoxypyrazine, which smells exactly like green peppers and peas. Interestingly, at least some of the results seem to indicate that the roasting actually covers the aroma and not volatilize it. This is reinforced by another chart in the Clarke and Vitzthum book that shows the level pyrazines don’t change between raw and roasted coffee.

In corresponding experiments with an aroma model for the coffee brew the assessors did not notice the absence of 3-isobutyl-2-methoxypyrazine. This leads to the conclusion that roasting of raw coffee does not only produce the pleasant aroma, but, in addition, generates odorants which mask the ‘green peas’ note caused by methoxypyrazine. (Coffee: Recent Developments – Ronald Clarke, O. G. Vitzthum)

So here’s my recommendations for a cold steeped coffee addition: Test your beans first! I like a number of the flavors and the higher octane rating of a lighter roast, but for a beer addition, I might consider starting at a medium roast and moving up from there. Additionally, limit contact time or at least conduct the extract at a lower temp in your fridge.

Otherwise, I think cold extract is the way to go when wanting a little java jolt brew.

Edited to add: Heard back from the entrant who’s beer I judged and here’s what they did – 1oz of coarse ground coffee (Handsome Roasters’ Regalo de Popayan – a lighter roast coffee – note the “riesling” characters they reference? That’s the pyrazine!) in a growler for 48 hours. They submitted two different entries – one CO2 purged, one not. I only judged the un-purged entry which had the green pepper. The other ended up winning a 2nd place medal. Quoting from Ward who did some followup research (and he’s actually knowledgable in chemistry – unlike some of us here):

Looking at structures it doesn’t appear to be an oxidation related process, as the methoxypyrazines that are responsible for bell pepper flavors aren’t oxidation products of those that produce favorable aromas. Instead it looks like the result of different substitutions off of the pyrazine ring. So in short, your base pyrazine content might be a major deciding factor on tasty versus vegetal. Might explain why some of my coffee beers with super roasty coffees were delicious, and why the one I just made with delicious, mellow coffee was a bummer.