IPA Experiment



The chloride-to-sulfate ratio in your brewing liquor influences the perception of your beer. Adding chloride to your water, usually in the form of calcium chloride, makes your beer more “rounded” and accentuates the malt. Adding sulfate, usually in the form of calcium sulfate (from gypsum), accentuates the hop bitterness. We’ve designed an experiment to test if this is true, if you can adjust this ratio in finished beer and which ratio works best in an IPA (or other hoppy ale). The last one is the biggie, here. The first idea is well-accepted and there’s anecdotal evidence that the second is true.

Brewing the Base Beer

The experiment is simple, brew 5.0 gallons (19 L) of IPA with a 1:1 chloride-to-sulfate ratio, then adjust the ratio as you bottle. To do this, on brewday, start with distilled water, RO water or tap water that has been diluted with distilled water to 25 ppm bicarbonate or less. Then, add calcium to your water by adding both calcium chloride and gypsum. For 10 gallons (38 L) of water, add 4.0 g (a little over 1 tsp) of calcium chloride and 4.0 g (about 1 tsp) of gypsum to get to 50 ppm calcium, with a 1:1 chloride-to-sulfate ratio. Make the wort, ferment it and get ready to bottle.



When you bottle, you will bottle 1.0 gallon (3.8 L) of beer at the existing 1:1 chloride to sulfate ratio. For each subsequent gallon (3.8 L) you bottle, you will add gypsum to yield bottles at 1:2, 1:3, 1:4 and 1:5 ratios, respectively. Here’s how to do that. Bottle the first gallon (3.8 L) as you normally would, and label these bottles with a “1.” Then, add 1.5 g of calcium sulfate to the remaining 4.0 gallons (15 L) of beer. To prevent foaming, wet the gypsum before adding it (see below for more). Stir to dissolve the gypsum into the batch, but not so much that you splash and aerate the batch. Let the beer sit for 2 minutes (so the dissolved gypsum evens out), then bottle the second gallon. Label these bottles with a “2.”

For the third gallon (3.8 L), add 1.0 g of gypsum and label the bottles “3.” For the fourth and fifth gallons, add 0.75 g and 0.33 g of gypsum, respectively. Label the bottles “4” and “5,” as appropriate.


The Solubility of Gypsum

Measuring the tiny amounts of gypsum required could be a problem. One way around this would be to make a known gypsum solution and add this. However, the solubility of gypsum is around 2.4 g/L in pure water. As such, you’d need add a total of just less than 2.0 qts. (2.0 L) of this solution to the beer by the end of bottling. So, when adding gypsum, just do your best to measure out the small amount. Then, wet it into a slurry. This will still likely raise some foam (and lower your level of dissolved CO2) when you stir it in, but it won’t dilute the beer as much. Incidentally, the solubility of calcium sulfate increases in colder water, contrary to how most solids behave, so don’t try to use hot water to make the slurry. The solubility of gypsum is increased in acidic solutions, so it’s possible that you could make an acidified known solution of gypsum, but then you’d also be dosing the beer with whatever acid you used.



If you’re making an extract beer, you won’t know the initial chloride-to-sulfate ratio in the dissolved extract. You can still try the experiment by brewing an extract beer using distilled water to dissolve the malt extract. You can use water treated with an equal amount of calcium chloride and calcium sulfate (gypsum) if you do a partial mash. Add the same amount of gypsum as called for above. You won’t know the exact ratio of your beers, but you will have a series of beers with increasing percentages of calcium sulfate.



You can brew any IPA you’d like, or any other hoppy beer that you like, for this experiment. Here’s the recipe James and I will be brewing, if you’d like to join us. (Ignore the water recommendations in the recipe, obviously, and follow the experimental protocol above.)



We’ll post a form later to fill out, but basically the analysis will involve tasting the beers and giving your impressions of each. We’ll collect the results of participants through this Halloween (Oct 31, 2013).



There are a couple criticisms that could made of this experimental design. Although the chloride-to-sulfate ratio varies among the experimental treatments, so does the level of calcium and carbonation. The higher numbered beers will have more minerals dissolved in them and (likely) less carbon dioxide from foaming when the gypsum is stirred in. This could be corrected by brewing five separate batches of IPA. Of course, that’s a lot more work and we think the experiment will still provide valuable information (and perhaps inspiration to run larger scale experiments in the future).

You can hear James and I discuss this experiment on his Basic Brewing Radio podcast, September 25th, 2013 episode.


  1. What about a little easier one for those of us with kegging abilities. Make the base beer, and start by racking it to 2L bottles. Get 4 2L worth. Add Gypsum to each 2L (0.2 grams, 0.4g, 0.6g, 0.8g) and cap with a carbonator cap. Carbonate with CO2 per normal method. This will remove the loss of CO2 with each addition in the bottling scheme, and leaves you with 3 gallons of base beer left in the keg. Compare the 4 2L batches and the kegged batch, take notes, find the best one, then scale up the results of the experiment on the kegged portion to have the best Sulfate:Chloride ratio for the remaining beer.

    • Chris Colby says:

      That would also work (and, as you mention, solve the carbonation problem). I might get a couple more carbonation caps and try it that way.

      Note: you could also carbonate the beer first, then move it 2 L bottles. You could then use the carbonator caps simply to touch up the carbonation levels. (Of course, you might get more foaming when you add the gypsum.)

  2. Eric from Long Island says:

    I have another idea to avoid the solubility and calcium issue.
    1 batch – as big as you can boil safely
    split into 2 fermenters (A & B)
    add extra gypsum to fermentor B to get the 1:5 ratio in fermentor B (mix well pre fermentation)

    Important to make sure both fermenters get the same yeast pitch and are fermented side by side. Need to check OG and FG for both fermenters.

    at bottling bottle blend the beers to produce 5 different batches:
    100% A
    75% A + 25% B
    50% A +50% B
    25% A + 75% B
    100% B

    the process would be to add beer from fermenter A and B to bottling bucket. Add priming sugar. Stir to mix. Bottle. Mark. Rinse and sanitize bottling bucket, then repeat until all 5 batches are done…

    Extra credit: I believe it would be possible to reduce the CaCl2 addition in the base beer, then add CaCl2 to fermenter A at same time CaSO4 is added to fermenter B. This would allow you to keep the Ca levels pretty close across the 5 beers.

    What do you think? I’d be happy to do the experiment either way it sounds like fun.

    • Chris Colby says:

      Yes, I think this would work. Yet another way would be to split the wort into five one-gallon jugs (or small carboys). The wort could initially be made with just enough gypsum added to give add the minimal amount of calcium. Then calcium chloride and gypsum could be added to each jug (perhaps pre-fermentation) in such a way that they varied with respect to the chloride-to-sulfate ratio, but had the same calcium levels (basically the same amount of minerals overall, just in different ratios).
      The experimental outline in the article is the simplest way to do it (and with the least equipment; no multiple carboys needed, etc.), but there are other ways of doing it — including brewing five separate batches — that yield a better controlled experiment.
      Good idea.

  3. I wonder if you could do a less scientific experiment on commercial beer (especially if you could get the ratio from the bewery), to see if adding gypsum to it in the glass would make a difference.

  4. You wouldn’t want to add the salts after the fact of them being “useful”. They would need to be added to water before the mash. They are necessary for mashing, useful in the boil, and during fermentation.

    Could you overcome the solubility issue by dissolving in ethanol and then adding that to water? It might help or it might precipitate.

    • Chris Colby says:

      In this case, we’re adding the salts for the chloride-to-sulfate balance, not to adjust mash pH (and I specified adding a small amount of calcium first for that purpose). So, adding the mineral after the mash makes sense, given the objective for adding them. Otherwise, we’d have to brew five separate beers (from 5 separate mashes) to do the experiment.

    • John Fitts says:

      in answer to your second comment, it sounds like a good idea, but it turns out that most salts have next to zero solubility in ethanol and gypsum is no exception.

      As a side note, i think i might spread my batch out into 5 secondary fermenters (1 gallon jugs) and add the gypsum there so that I don’t have to stir it in after fermentation is completely done and have to worry about different levels of carbonation. Just add the appropriate amounts of gypsum to each jug and rack 1 gallon n of beer on top, let it sit till cleared, then bottle.

  5. Brandon Dufala says:

    I just wanted to make sure of the the numbers. I’ve never used chemical additions. At the top of the article it says for 10gals of water. But the additions are really small. I wanted gone sure that I was supposed to add 4g of each per 10gal of water.

    And to be clear … This is all before mashing? Correct?

    • Chris Colby says:

      Yes, the mineral additions for the brewing water are for 10 gallons. 4 g each of calcium chloride and gypsum. This is a small amount, but the target is only roughly 50 ppm for [Ca 2+], [Cl -], and [SO4 2-]. And yes, those initial additions are made before the mash.

      • Brandon Dufala says:

        Do I need to add any other chemicals to the distilled water for pH or anything like that?

        I’m excited to do this experiment.

  6. I posted this on Facebook, but just to get it on the record here…my experiments have me pretty much convinced that the absolute amounts of sulfate and chloride make a lot more difference than the ratio between them. I reached that conclusion after doing some experimenting with increasing the amount of sulfate in my Rye IPA, a recipe I know very well. I also was conscious of Martin Brungard’s advice to keep chloride levels under 100 in general. To get amount of sulfate I wanted in my IPA while maintaining the “perfect” ratio would have put my chloride well over 100. So I simply started increasing the sulfate while maintaining the chloride at previous levels. It got me exactly what I was looking for and subjective, non scientific tasting told me there were no ill effects from not maintaining the ratio.

    • Adding some additional info here…here’s a citation from Martin re: ratios and absolute chloride levels…

      “Remember, you can’t just add chloride and sulfate to your heart’s desire. At some point, you’ll have a minerally beer on your hands. I feel that keeping chloride to less than 100 ppm is always good and you can vary sulfate from 0 to around 300 ppm with no problem. However when you really boost sulfate, chloride should be reduced well below 100 ppm. If you want minerally beer flavor, boost chloride above 150 ppm and sulfate above 300 ppm. ”


  7. I recently got my hands on a 1/100 g resolution scale and measured the grams per volume of various salts. I did get “a little over a tsp” for 4 g CaCl, but gypsum was “a little more over a tsp.” My exact measurements are posted at http://lifefermented.wordpress.com/2013/10/02/water-treatment-for-brewing-3-6-brewing-salt-gtsp-conversion/

    I’m not saying Chris’ approximations are wrong- it could be the difference between brands of salts, the ambient humidity, etc etc.

    • Chris Colby says:

      Measuring the weight of solids is always better than volume-based approximations. The approximations I used are from John Palmer’s water spreadsheet. In volume for weight approximations, how tightly the solid is packed makes a big difference.

  8. Brandon Dufala says:

    I finished my experiment IPA last night around 3:00am. 3oz magnum. 2oz cascade. Plus 2 more oz at dry hop.

  9. Eric from Long Island says:

    I finished brewing my batch Saturday (10/5). Columbus – Otter Smash. 1.060 OG / 61 IBU Calculated with total of 4oz hop additions at FWH/60/FO and DH.

    I deviated from the experiment by producing a base beer with 40 ppm calcium, 79 ppm sulfate and 29 ppm chloride. At bottling I will make 5 “priming sugar solutions” which will include 1 cup water, 23 grams corn sugar, plus salts (gypsum and/or calcium chloride). For example the 1:1 ratio salt addition is 0.4 g CaCl2, the 2:1 addition will be 0.2 g CaCl2 + 0.3 g gypsum and so on. The priming sugar solutions will be boiled and stirred, transferred warm into the bottling bucket. I will then autosiphon into the bottling bucket to the 1 gallon mark and gently stir to combine. Fill bottles, cap bottles, mark bottles, then rinse bottling bucket with star san and repeat process till done. This will give me the required set of ratios with each beer having about the same calcium concentration (range is 69 to 96 ppm), and the same manipulation in the bottling bucket. The salts may or may not dissolve in the priming sugar solutions but should be sufficiently suspended to evenly dissolve in the beer and I will bottle fast after the gentle stir so salts in suspension stay that way at least into the bottles.

    I am marking the bottles with symbols: e.g square, circle, triangle, star, hash mark to facilitate blinded tastings.

    I’m now looking for ideas about how to conduct the tastings. Awaiting the form… Has anybody thought about how to avoid pallet fatigue in tasting a series of very similar IPAs? I am wondering if I should get multiple tasters and give them the samples in different orders.

    Maybe you can see I’m having fun with this. Thanks Chris for creating this opportunity!

  10. Brandon Dufala says:

    I’m bottling my IPA tomorrow (Tuesday 10/22/13). I can’t wait to see how this turns out.

  11. Brandon Dufala says:

    Chris Colby. When are you going to post the analysis form?

    I bottled the other day and will begkn checking carbonation after 7 days.

  12. Brandonlovesbeer says:

    Will the additions also contribute to the perception of Hop flavor and aroma? Or just the Hop bitterness?

  13. Brandon Dufala says:


Speak Your Mind