Crystal Malts Reduce Fermentability: Fact or Fiction?

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk:

We all know those topics that pop up time and time again on any home brewing forum. One of those topics is under-attenuation. Why didn't my beer attenuate? someone asks, followed by a string of replies asking how much crystal malt was used, after which ensues the explanation that the addition of unfermentable crystal malt sugars is to blame. I want to investigate whether or not this reasoning is actually sound, and what the effects of using crystal malts are on the fermentability of the wort.
As with any investigation, we don't know all the parameters and have to make assumptions. This is no exception, so let's start by taking it as fact that after the malting process, crystal malts contain a lot of unfermentable sugar. I've never seen a crystal malt specification sheet showing the breakdown of the different sugar molecules, but there seems to be evidence to support this, so let's roll with it. During the malting process, the grain essentially starts to germinate; the starch-converting enzymes start working away creating sugar. Before too much of the sugar is used by the seed for growth, the process is halted, typically by dry heat. This part of the process has the side effect of denaturing enzymes, notably the starch and sugar conversion enzymes (a- and B- amylase, limit dextrinase etc.). Now you have grain with high sugar content and no diastatic power. So it makes sense that if you steep crystal malt in the absence of starch/sugar-converting enzymes, then the resulting sweet liquor will contain a substantial amount of unfermentable sugar. Therefore, in extract plus specialty grains brewing the old theory seems reasonable.
What happens if you bring other malts in to the equation? Most of the time in all-grain brewing (or partial mash brewing for that matter), the crystal malts are mashed in with all the other malts. Modern base malts are high in diastatic power, so there are plenty of enzymes to go around, and any low or non-diastatic malts used in relatively small amounts convert just fine. But we're supposed to believe that the enzymes from the base malts aren't going to work on the sugars from the crystal malts? That seems pretty suspect to me. Are the enzymes xenophobic or something? Is there some sort of magical force field around the sugars from the crystal malt?
In my pursuit of knowledge I came across an experiment by fellow HBT-member nilo' that was aimed at getting to the bottom of my question. The experiment was documented in the author's blog Homebrewing Beer and discussed in detail in this HBT thread. In brief, Nilo made seven samples: three 100% crystal malt samples, each using a different crystal malt with a different colour rating between 10 and 120L; one 100% base malt sample; and three 50% base malt/50% crystal malt samples (the same crystal malts as the 100% samples). All samples were mashed or steeped at 155F (~68C) for 30 minutes. The original gravities (OGs) were recorded along with the apparent attenuations (AAs) as a function of time. All samples were fermented at 70F (~21C) with Fermentis Safale S-04 yeast.
Figure 1 shows the measured OGs of the samples along with the OG you would expect for the 50/50 samples by taking the average of the straight base malt and corresponding straight crystal malt sample. Figure 1 suggests that the enzymes in the base malt do in fact work on the crystal malts. This is illustrated by the fact that in each case the yield increased when 50% base malt was used, but the increase was more than that expected when simply averaging the extracts from the crystal and base malt.

Extract yield expressed as specific gravity for each sample. Samples were mashed at 155F (~68C) for 30 minutes. Source.

Figure 2
shows the AAs of the samples and illustrates two things. First, it supports the idea that crystal malts contain substantial amounts of unfermentable sugar after the malting process, and that steeping in the absence of diastatic malts results in a not very fermentable wort. Second, it supports the idea that adding diastatic power makes the resulting wort more fermentable. While the AAs of the 100% crystal malt samples were quite low (40-50%), the samples with 50% base malt showed 67-75% AA compared to the 100% base malt sample which exhibited 80% AA.

Measured apparent attenuation (AA) of each sample and expected apparent attenuation for the 50/50 base malt/crystal malt samples. Expected AA is the average of the corresponding 100% crystal malt sample and the 100% base malt sample. Source.
So yes, the crystal malt lowered the AA even when mashed with base malt. However, the presence of base malt had a large impact on the fermentability of extract obtained from crystal malt. Also, remember these were 50% crystal malt, which is unusually high for a beer grain bill. Using a linear interpolation between the 100% base malt (80% AA) and worst attenuating 50/50 sample (67% AA), I determined the following table of percentage crystal malt, AA, and corresponding FG from an OG of 1.050.

Expected apparent attenuation (AA) and expected final gravity for 1.050 OG samples consisting of various percentages of crystal malt mashed at 155F (~68C) for 30 minutes and fermented with S-04 yeast. Data linearly interpolated from experimental results shown in Figures 1 and 2.

So for a much more reasonable recipe using only 5-10% crystal malt, you would only expect the FG to be maybe a point higher compared to 100% base malt. However, there are so many other factors that affect the attenuation such as yeast health, pitching rate, nutrient content of the wort, etc., which most likely produce a larger change. This makes it very hard to realistically point the finger at crystal malt for causing major changes to the apparent attenuation in the context of all-grain brewing.
So, do crystal malts reduce fermentability? Yes. However, as always, the real world results aren't so straight forward, and a proper understanding is required to make a sensible assessment of the situation.
Many thanks to Nilo for doing the real work here and conducting the experiment. Happy brewing, everyone.
"During the malting process, the grain essentially starts to germinate; the starch-converting enzymes start working away creating sugar" I don't believe this is correct. My understanding is that crystal malts are simply base malts that have been further kilned, such that the starches in the barley are caramelized into sugar not by the enzymes, but by the heat. That's why the resulting sugars are "unfermentable." They're a different kind of sugar than the sugars produced by enzymatic conversion.
Hmm, so it looks like the answer is, "both?" The enzymes do indeed (apparently) convert the starches to sugars during malting, but the sugars are then caramelized by heat into (presumably) unfermentable sugars.
Yes, base malts that have been malted, which is the process that creates sugar, and then kilned or roasted wet to crystallize the sugars.
I don't know if the heat from the kilning/roasting process creates more sugars, but it does change them. It's the malting that creates the majority of the sugars. You can't simply wet kiln/roast unmalted grain and expect to get much in the way of sugars out of it.
I suppose that's a way of saying, I think the quote you cited is correct as written, more or less, and I'm not sure what part you think is mistaken.
Edit: Ninjaed! :)
Don't forget, too, that right after the malting process, the kilning of base malts is done gently enough to preserve the enzymes. So the later process of mashing releases yet more sugar from the grain. It's entirely possible that crystal and base malts have roughly similar amounts of sugar when you pull them from the bag, but that base malt has the potential to create more.
This is also why crystal malts will attenuate far more when mashed rather than used as steeping grains. There's still starch in there that can be converted, they just don't have any enzymes to do so themselves.
Another thing that confuses me: "Figure 1 suggests that the enzymes in the base malt do in fact work on the crystal malts."
Does it? The gravity reading doesn't necessarily mean the enzymes have carried out any conversion on the crystal malt - it only means that "stuff" has dissolved from the malts into the wort. It could be sugars, or it could be unconverted starches. The author seems to be equating gravity with sugars, and I think that's a false equivalency that ignores the fact that (unconverted) starches will affect gravity, too.
"Caramel Malts have undergone a special heat 'stewing' process after the malting which crystallizes the sugars. These sugars are caramelized into longer chains that are not converted into simple sugars by the enzymes during the mash."
-How to Brew, John Palmer
"Unfermentable sugars are highly caramelized sugars, like those in caramel malts, and long chain sugars referred to as dextrins. Dextrin malt and malto-dextrin powder have been previously mentioned in the ingredients chapters. Dextrins are tasteless carbohydrates that hang around, adding some weight and viscosity to the beer. The effect is fairly limited and some brewers suspect that dextrins are a leading cause of 'beer farts,' when these otherwise unfermentable carbohydrates are finally broken down in the intestines."
-How to Brew, John Palmer
  • Article Author
  • Article Author
  • #11
That's a good point. However, beers with plenty of crystal malt don't necessarily show signs of excess starch present. E.g. starch haze. I made the assumption that the extract = sugar because of this.
  • Article Author
  • Article Author
  • #12
As many have pointed out, my description of crystal malt was slightly incorrect. I was trying to keep things simple for the less scientifically inclined readers. Yes, some of the sugars in wort resulting from crystal malts are different to those from base malts. However, Nilo's data indicates that some of the sugars obtained from crystal malts do get converted, when mashed, into fermentable ones.
  • Article Author
  • Article Author
  • #13
Yes, that was a big assumption, but with only two ratios of base to crystal malt to work with I didn't have much choice. I chose the worst case scenario. I.e. the case that would give the largest difference in AA compared to 100 % base malt.
Nice job with your meta-analysis of Nilo's experiments. There are so many brewing "rules" and myths that get busted all the time. Thanks for sharing.
@kombat "My understanding is that crystal malts are simply base malts that have been further kilned"
You are describing kilned malts (biscuit, amber, brown, chocolate, etc.) which are different than crystal malts.
Crystal malts are pre-converted prior to heat exposure - sometimes by hot water immersion, or in the case of british malts, by stewing - (bonus: can be steeped and still add sugar to an extract beer), otherwise there would be no sugar to caramelize, only toasted starch and mailardized protein.
Kilned malts are unconverted when exposed to additional heat (and therefore must be mashed to contribute to gravity).
It seems to me that a 30 minute mash is kind of short to make assumptions about. I'd generally do a 60 minute mash and get even better conversion from a 90 minute. Especially on the mixed mashes you're dealing with reduced enzymes in the mix so I would expect conversion to take longer. I'd like to see the same experiment done with a 60-90 min mash to see if it changes much.
  • Article Author
  • Article Author
  • #17
As with any experiment, the results only tell you what they tell you. Nothing more, nothing less. If you're right about converison taking longer due to lower levels of enzymes, then mashing longer would just bring the 50/50 results closer to the 100 % base malt sample. In the end I don't think a 30 min mash is an issue because we're only comparing the samples to one another, not really looking at the absolute extract and attenuation. Therefore so long as all the samples were mashed in the same way, including duration, then the relative comparisons stand.
Great experiment! Very useful.
I can' t figure out why OG in 50/50 was higher than expected, as it was said above, OG shouldn't increment if there are more fermentable sugars, it should be similar.
Also, I was wondering if you know which crystal malts brand was used. I read that in case of weyermann's crystal malts the enzimes are 100% denatured, but I don't know if it is true.