I have friends who almost never make a style-based beer - or, at least, that's what they think. The simple reality is that every beer has a style category in the Beer Judge Certification Program guidelines, but sometimes that style is this one: Category 34C, "Experimental Beers."
It's into this category that we venture today, because three times in the last week I've run across people asking about whether you can use fractional distillation (freezing off water out of a fermented beer to concentrate it) on styles other than Eisbock. The simple answer is that, yes, you can do anything, which doesn't mean that you should. In this case, though, I'll make the case that we should be using this process with abandon to play with the limits of what beer can do.
I wouldn't recommend this for every beer, but knowing how to build a recipe for fractional distillation (and, of course, how to use the process) and making a style an "Ice Whatever" is a great club to have in your bag.
STYLE
Not every beer needs to be iced. Can you make an Eis-Helles? Sure. Why would you want to? If you want a Helles with concentrated flavors and higher ABV, just increase the poundage of grain in the recipe and choose more-intense character grains/hops. The best styles for this process are those that already live at the upper limits of beer flavors and ABV.
The most famous ice beers in the world - Eisbocks - take a relatively strong lager and make it stronger, sometimes well into double-digits ABV. If we take other styles in that vicinity, we can likewise use fractional distillation to concentrate them into imperial versions of their former selves.
Strong Scotch Ale, a wide range of stouts, British or American Strong Ales, or (as we'll road-test below) Baltic Porter, among others.
RECIPE
There are definitely some recipe considerations here. Let's use my Ostseeküste Porter recipe as an example. The standard recipe calls for about 15 pounds (7.8kg) of base grains and about half a pound (0.2kg) each of three crystal/chocolate malts.
If I'm going to be concentrating flavors, I usually leave the base malts alone - they can get more intense by several orders of magnitude without blowing out anyone's palate. However, I'm going to take a good, hard look at those crystal and chocolate malts. The higher up the Lovibond scale we go, the more I'm going to think about decreasing the amount I'm using or pivoting to a less-assertive substitute.
Crystal 20? Leave it as-is. Crystal 65 (which we use in the Baltic Porter recipe) up to about Crystal 120? Probably OK, but I'd reduce it by whatever percentage I'm planning on concentrating. But once we get north of 120L I'm either eliminating the malt or replacing it with something milder.
So, in this case, instead of the Carafa Special II we can use Carafa Special I. In other recipes, where we're using husked chocolate malts, I always recommend pivoting to something dehusked (Chocolate Rye, for example). Any astringent flavors you'd pick up will get much more intense when concentrated.
You can leave your flavor hops alone, but reduce your IBU load by more than your anticipated concentration percentage. A linear reduction will likely lead to too-great an impression of bitterness. Instead, take your anticipated concentration, apply a 1.25 multiplier to it, and reduce your calculated IBUs by that much.
Finally, always choose a clean yeast for your ice beers. If it's a lager, you're already fine. If it's an ale, consider the American Ale strains or my venerable Wyeast 1007 (German Ale). If it's a Belgian…ask yourself why you're fractionally distilling a Belgian, and if you're satisfied with your own answer choose a relatively low-key Belgian strain like Belgian Ardennes from Wyeast.
PROCESS
Take every opportunity to make an attenuation-heavy beer here. Ice beers are the principal exception to my "mash everything at 152F/67C" rule (someone recently recommended I start wearing the number on a t-shirt to speed up the process when I'm asked about mash temps). Instead, go lower (148F/64C). Otherwise, your process is the same as it ever is, on the hot side.
On the cold side, manage your fermentation aggressively. Oxygenate well to get a clean start, start cool for the first 72 hours then ramp up the temperature to encourage full attenuation, and give yourself a high and long diacetyl rest. We want to minimize any and all off-flavors and, depending on the style, fermentation character. Post-fermentation, the beer should taste a bit thin in body and slightly sweet. Then the freezing begins.
There are two schools of thought here. One is that you should go very low - about 15F/-9C - and form your slush rapidly, potentially in just a few hours. The other is that you should take a slower path - about 22F/-5C - and take something more like 8-10 hours. I choose the second, slower path. It will probably mean a more-accurate concentration, especially as you repeat the process and/or recipe, and more control is always preferable for me as it means I can more-accurately dial in the result and make effective adjustments. In either case, generate your ice and rack out from underneath it before packaging.
Finally, bottle conditioning is absolutely still possible in ice beers, even without the addition of more yeast. It might take a little longer, but yeast are hardy, and even my most aggressive Eisbock perked up in just three weeks at room temperature. Having said that, I don't think it's ridiculous to do so here, or just force-carbonate (even if that's not your usual route).
IN CLOSING
You may notice some sharp flavors when your ice beers are young, but these are great candidates for extended lagering. Your best attempts will be drinkable straightaway, but don't expect that. Let them lager for eight weeks or so before evaluating, and consider cellaring a sample of bottles for six months to a year. Fire up that freezer and have some fun!