Water Chemistry for Stout

[CPM]

One of my favorite beers is Oatmeal Stout. My recipe uses 5% Crystal Malt and 7% roasted UK malts. My new brewing location has different water chemistry, and my opinion is that the beer is not quite as good as it was. On the last batch, brewed this weekend, I was planning to use CaC03, but I read up a bit on its use, and changed my plans. Instead, I decided to monitor the mash pH and add nothing.
The mash pH started and ended at about 4.9.
The flavor difference, and reason I care, is that the new water doesn't seem to produce the rich, chocolatey flavor that the original beer had. The roastiness comes through (yeah, it ought to) but it has a somewhat hollow character and tastes a bit more burnt than browned. I never tested the pH of my mash at the old place.

What should I do to get my pH back in line?


***NEW (municipal) WATER*** Ward Labs Data:

pH                                      8.0
Total Dissolved Solids (TDS) Est, ppm   121
Electrical Conductivity, mmho/cm       0.20
Cations / Anions, me/L            1.8 / 1.7

                         ppm
Sodium, Na                24
Potassium, K               3
Calcium, Ca                8
Magnesium, Mg              3
Total Hardness, CaCO3     33
Nitrate, NO3-N           0.4 (SAFE)
Sulfate, SO4-S            11
Chloride, Cl              11
Carbonate, CO3         < 1.0
Bicarbonate, HCO3         37
Total Alkalinity, CaCO3   31
Total Phosphorus, P     0.25
Total Iron, Fe        < 0.01

Just for comparison...

***OLD (well) WATER*** Ward Labs Data:

pH                                      7.7
Total Dissolved Solids (TDS) Est, ppm   249
Electrical Conductivity, mmho/cm       0.41
Cations / Anions, me/L            4.6 / 4.5
                         ppm
Sodium, Na                10
Potassium, K               2
Calcium, Ca               61
Magnesium, Mg             12
Total Hardness, CaCO3    203
Nitrate, NO3-N           0.2 (SAFE)
Sulfate, SO4-S             3
Chloride, Cl               9
Carbonate, CO3         < 1.0
Bicarbonate, HCO3        244
Total Alkalinity, CaCO3  201
Total Phosphorus, P     0.02
Total Iron, Fe        < 0.01

 

[mabrungard]

To better duplicate your old water source, using pickling lime would add the alkalinity needed along with calcium. Unfortunately, dry pickling lime may not be as pure as assumed and it may not add the intended alkalinity or calcium since a portion is reverted to chalk. A surer way of employing pickling lime is to create a saturated solution of the lime in water and adding the saturated solution to the mashing water.

Another sure way to obtain the desired alkalinity is to use baking soda. However instead of calcium, you will be adding sodium and that can have some negative flavor effects when sodium content gets too high. Fortunately, you don't typically have to add much baking soda to the mash and the sodium content stays somewhat low.

 

[CPM]

Thank You, mabrungard! After reading about the difficulty with dissolving chalk and its questionable efficacy, I was thinking of using Baking Soda to adjust the water, maybe with some Gypsum.
I have been using Bru'n Water 1.16b to try out some ideas. It predicts a pH of 5.5 for no adjustment, though I ended up with 4.9. Should I aim for shifting that 5.5 -> 5.8 in order to achieve 4.9 -> 5.2? That's a 0.6g/gal addition of NaHCO3, and it only takes my sodium up to 67.4
It's not important to match the old water, Fixing the flavor is my goal.

 

[mabrungard]

Was a freshly-calibrated pH meter used to measure that 4.9 pH? If the new low-alkalinity water was used, the low pH sounds possible with that grist.

 

[CPM]

I used two. One was a lab grade piece of equipment that hadn't been calibrated recently, and the other was an el cheapo pen that was calibrated. Both measurements rounded to 4.9.

 

[ajdelange]

To better duplicate your old water source, using pickling lime would add the alkalinity needed along with calcium. Unfortunately, dry pickling lime may not be as pure as assumed and it may not add the intended alkalinity or calcium since a portion is reverted to chalk.

I've been thinking about this very subject a bit. At this point I've come up with the idea of using calcium carbonate but using it in the following fashion. Weigh out a gram of calcium carbonate. That's 10 mmol as the molecular weight of CaCO3 is 100 mg/mmol. Put that in a suitable container meaning a Pyrex test tube or crucible and heat to red heat over a propane or MAP torch flame or even a gas stove burner for about 5 minutes (should do). Cool. You now have 10 mmol CaO which will absorb 20 mEq protons
CaO + H2O + 2H+ --> Ca(OH)2 + 2H+ --> Ca++ + 2H2O

Recognize that CaO is 'quick' lime which is so called because a lot of heat is evolved when it is hydrated. At gram levels this shouldn't present a problem but if you need to make more be cautious when adding to water.

Verify that the CaCO3 has completely converted by putting a little of the CaO in some RO or DI water and measuring both the P (pHe = 8.3; phenolpthalein) and M (pHe = 4.3; Methyl orange) alkalinities. They should be the same.

 

[CPM]

ajdelange, do I understand correctly that this process would be to address the purity issue? One would still have to deal with carbonate precipitation if added to the hot liquor, right?

 

[zwiller]

First off, congrats on the new supply. Nice water. Martin already mentioned it, but I like using baking soda for dark beers. I find the sodium contribution works well in dark beers. Try adding a pinch to a pint and see if you like it. Then again, lime would be more neutral in flavor.

 

[ajdelange]

ajdelange, do I understand correctly that this process would be to address the purity issue?

Yes. The calcium carbonate is converted to calcium oxide:

CaCO3 --> CaO + CO2. When this is 'slaked'

CaO + H2O --> Ca(OH)2

you get pure (provided that all the caCO3 was converted) Ca(OH)2 and that absorbs protons

Ca(OH)2 + 2H+ --> Ca++ + 2H2O

One would still have to deal with carbonate precipitation if added to the hot liquor, right?

No. There is no carbonate to precipitate.

Now perhaps you are asking the very sensible question "If heating converts CaCO3 to CaO why can't I just heat my contaminated lime and convert the contaminating CaCO3 to CaO?" and the answer is that you certainly can do that but need to understand that not only will the CaCO3 convert to quicklime by emitting CO2 but that the Ca(OH)2 (slaked lime) will convert back to quicklime by giving off water. Thus you start with a mixture of Ca(OH)2 and CaCO3 and wind up with pure CaO. This implies that you need to do any weighing after heating (and cooling!) rather than before. Given that you have to go to the trouble (and expose yourself to the risks - be careful, wear proper safety equipment!) of heating some powder you might as well deal with chalk which you can get at any LHBS. And the calculations are simple. 100 mg of chalk treated in this way provides 100 mg alkalinity as CaCO3.

Let's say you start out with x mmol of pure Ca(OH)2 of which, after some time, a fraction, r, converts to CaCO3 from exposure to air. Then of the x mmols of calcium r*x are attached to carbonate and (1 - r)*x are attached to hydroxyl ions. If you put this powder into water and add strong acid to it (most kits use 0.1 N sulfuric) while monitoring pH (with a meter or phenolpthalein indicator) until pH reaches 8.3 you will have converted all the (OH)- ions to water requiring 2*(1-r)*x mEq of the acid. You will also have converted all the carbonate to bicarbonate thus consuming r*x mL additional for a total of P = 2*x -2*r*x + r*x = 2*x - r*x mEq. If you now add more acid until the pH reaches 4.3 - 4.5 you will convert the HCO3- ions to H2CO3 and in doing so consume r*x mEq more acid so that the total to pH 4.3 is M = 2*x -r*x + r*x = 2*x mEq. r, the contamination factor, is easily determined as

r = 2*(M - P)/M

The purity is (1 - r) and 100*(1 - r) is the purity of the Ca(OH)2 in %. Note that while we have used mEq as the units for M and P but as we are after a ratio it doesn't matter what the relationship between the mmol of calcium and the alkalinity measurement unit is (x cancels). Thus M and P can be in units of drops if a drop count kit is used, mL of acid of a particular strength, dH, °F or whatever your alkalinity test kit uses. The only requirement is that the test kit measure total (M) alkalinity and phenolpthalein (P) alkalinity. I don't know if the aquarium test kits permit this but I would think they would.

 

[ajdelange]

I got all hung up on the purity of Ca(OH)2 thing and never addressed the basic question so here goes:

What should I do to get my pH back in line?

Figure out the difference in proton absorptions of the two waters and add base to absorb the protons that the old water grabbed but the new water doesn't. Let's start with...

***OLD (well) WATER*** Ward Labs Data:
pH 7.7
Calcium, Ca 61
Total Alkalinity, CaCO3 201

The first thing we need to know is the total carbo content of the water this is Ct(201/50,7.7, 4.4) = 4.20687 mmol/L. Ct is a function that calculates this from the alkalinity of your water (201/50), it's pH and the end point used in the titration by the lab when it measures alkalinity. Ward labs uses 4.4 though the ISO standard calls for 4.5. Note that the carbo chart we'll get to in a minute uses 4.5. It doesn't make that much difference in most cases. The formula for Ct is given in Step 10 at https://www.homebrewtalk.com/showthread.php?t=473408 if you want to put it in a spreadsheet or it can be found from this chart.

Next you need to determine how much acid is required to take 4.21 mmol/L of carbo from the water's original pH of 7.7 to your mash pH. Let's assume that your mash pH was 5.5 for purposes of illustration here. Use the second chart below to find out how much acid is needed per mmol/L carbo to reach 5.5 from 7.7. That's about 0.838. Multiply this by 4.21 to find that you would need 3.528 mEq/L. Using the math the answer it howmuch(5.5,7.7,201/50,4.4) = 3.53347 mEq/L.

Now we do the same with the new water.

***NEW (municipal) WATER*** Ward Labs Data:
pH 8.0
Calcium, Ca 8
Total Alkalinity, CaCO3 31

Jumping directly to the acid requirement obtained in exactly the same way
howmuch(5.5,8,31/50,4.4) = 0.516151 mEq/L
and we see that you have 3.53347 - 0.516151 = 3.01732 mEq/L more acid in your old water mash than you need with the new water. To hit the same pH you need 3.0 mEq less acid per liter of the water. Assuming that you don't want to remove acid by removing dark grain you must remove it by adding proton absorbing base. Sodium bicarbonate is a candidate, of course, but adding 3 mEq of it per liter of water results in adding 3 mmol (69 mg) sodium per liter as well. I don't suppose that's too bad. Using lime you would add 3 mEq of calcium (60 mg/L). Calcium is less flavorful than sodium and we note that adding 60 mg/L calcium would get the calcium level of the treated new water close to what it was in the old.

My recipe uses 5% Crystal Malt and 7% roasted UK malts.
...
The mash pH started and ended at about 4.9.

Given the parameters of the new water and the grist description, non specific as it is, I don't for a minute believe your actual mash pH was as low as 4.9. The fact that it didn't change over the time of the mash and that the metering was questionable make me even more confident that this is the case. A reasonable mash pH for the grist and new water would be around 5.56. Using the same actual malts that I used to estimate 5.56 as the pH for the new water would give me an estimate of 5.69 as an estimate for the old. This is the effect of the 3 mEq/L additional proton absorption.

 

[ajdelange]

I really should point out that as we use approximately 90% of the alkalinity (in mEq/L) to reach typical mash pH from typical source water pH you can estimate the acid requirement difference in going between the two waters as
0.9*(201 - 31)/50 = 3.06. That's pretty close to the 3.017 calculated exactly. So you don't have to go through the whole process described in #10 but I think you'll have to admit the pictures are pretty cool.

 

[CPM]

Thanks, zwiller, I like the pinch in a pint idea. I have a good scale, so I'll probably make a solution and add an amount to the beer to simulate the amount that would be added to the strike water.

Thanks, ajdelange, for your generous help with this. It will take me a little while to get my head wrapped around your formulae, and to begin to think in mEq instead of just grams and liters and whatever the spreadsheet told me. I appreciate your walking it all the way through, and giving me a starting point.
I'll definitely calibrate that meter. I dont have an ATC probe for it, so I just cooled the sample to about 80F/27C in a water bath in the freezer.