In a Facebook group I follow, an interesting problem came up. Somebody had brewed a mixed-fermentation Berliner Weisse with brewer’s yeast, lactobacillus brevis and brettanonymces claussenii. They didn’t own a saccharometer, but instead determined their OG with a refractometer. They now wanted to know whether fermentation was finished, and used the refractometer as well to measure the beer that had been in the fermenter for several weeks.
Before I continue, a quick excursion into how refractometers work. Every translucent material bends light to a certain extent, the light gets refracted. To describe the extent with which the light is bent, the so-called refractive index is used. The refractive index n is defined as n = c / v, where c is the speed of light in a vacuum, and v is the speed of light in the particular medium. Water for example has a refractive index of 1.333. When we dissolve sugar in water, the refractive index of the solution is increased. The refractive index of a 10% glucose solution in water is 1.347, for example. This change in the refractive index can be used to indirectly measure the OG, by looking at the difference between the expected refraction of water vs the measured one, i.e. by how much more light gets refracted.
In fermented beer, this gets trickier, because due to fermentation, the resulting liquid contains ethanol. Ethanol has a refractive index of 1.361, which skews the overall measurement. Ethanol also skews the measurement when determining the FG of a beer, as its specific gravity is just 0.79. In typical fermented beers, the amount of ethanol is large enough that we need to correct our measurements to be able to estimate the actual FG. This is all fine.
The mixed fermentation opens up another problem, though. It has a refractive index of 1.427 which significantly higher than that of either water or ethanol, but in typical fermented sour beers, its content by weight is fairly low compared to e.g. ethanol. According to this presentation, typical Berliner Weisse contains 2 to 4 g/L of lactic acid, i.e. 0.2 to 0.4% by weight.
When I read about the issue of seemingly underattenuated Berliner Weisse, one of the things that came to my mind was exactly whether the lactic acid from the mixed fermentation skewed the measurement enough to cause such a large disparity that a beer that is expected to be overattenuated to come up with an apparent attenuation of just 68%.
So I asked myself the question: if I added the typical concentration of lactic acid in a Berliner Weisse (i.e. 2 to 4 g/L) to distilled water, by how much would my refractometer be skewed?
I have a refractometer at home, 2 litres of distilled water, a big bottle of 80% lactic acid, and pipets with which I can measure out millilitres of lactic acid. But how many grams is a millilitre of lactic acid? A millilitre of pure lactic acid weighs 1.357 grams, therefore a millilitre of an 80% solution would weigh would weight about 1.285 grams. Since I want to test a whole range of lactic acid content, just measuring out by ml is good enough for me.
I poured 1 litre of distilled water into a clean vessel, and calibrated my refractometer so that it shows exactly 0°Brix. I then added 1 ml of lactic acid (= 1.285 g), stirred it well, and measured again with the refractometer. I measured X°Brix. I then repeated this to up to 5 ml of lactic acid (= 6.425g), and got the following measurements:
- 1 ml (~ 1.3g): 0°Brix
- 2 ml (~ 2.6g): 0.2°Brix
- 3 ml (~ 3.8g): 0.4°Brix
- 4 ml (~ 5.1g): 0.4°Brix
- 5 ml (~ 6.4g): 0.6°Brix
This was actually less skew than what I had expected. For a quick counter-check, I added 80% lactic acid on the refractometer, and the measurement was off the scale.
Luckily, I have a Berliner Weisse maturing at home. I brewed it a few months ago, mixed fermentation with S-04, Lacto brevis and Brett bruxellensis. I simply lautered and sparged 30 litres of wort from a 50% Pilsner/50% pale wheat malt mash, which turned out at 11°P OG (unboiled, of course). On my refractometer, I measured 5.2°Brix, while with my saccharometer, I measured 2.0°P. I used calculators to get the expected attenuation based on the OG (in Plato) and FG (in Plato) resp. refractometer reading (in Brix). For both values, I got almost the same level of attenuation (81.82 vs 82.11 apparent attenuation) as well as almost the same ABV (4.7% vs. 4.72%). This is fairly consistent with what I’ve measured earlier, namely that the amount of lactic acid in Berliner Weisse has very little impact to skew a refractometer measurement.
Nevertheless, I think it goes without saying though that a brewer should never ever rely on a refractometer alone. While I use one during my home-brewing, I only ever employ it to measure sugar content ad hoc during the brew day: it’s useful to observe saccharification of your mash, the strength of your first runnings, as well as the sugar content in your final runnings, or to get a good idea about the pre-boil gravity and post-boil gravity of your wort. It’s a tool that has its place, but for observing the progress of fermenting beer, I think it’s a much better idea to just use saccharometers. Even quite precise ones, with a scale down to 1/10 of a degree Plato, and thermometers for further correction of the measurement, are not exactly expensive.