Lime Softening

[bierhaus15]

I've been brewing with RO water for years without issue, but my source of cheap RO water is no more and I've always wanted to try lime softening. As such, I recently treated 3 batches of my very hard well water and have seen a reduction in alkalinity, but I'm sure I could be doing things better. I am also using phosphoric acid to re-acidify and I have some concerns about the resulting phosphate content.

Here is my water report:
pH: 7.5
TDS: 348
Na: 11
Ca: 81
Mg: 26
Tot. Hard: 311
Sulfate: 5
Cl: 22
Bicarb: 360
Tot. Alk: 295

My process has been to add calcium hydroxide to 15 gallons of untreated well water until the pH reaches around 11.0 pH, measured via a HQ40d meter. Approximately 1 Tb. of chalk is added to help with calcium precipitation. Water sits overnight until all of the chalk precipitates out of solution. The clear water is siphoned into another stainless vessel and phosphoric acid (75%) is added to the water until the pH reaches 7 (about 3 ml). While adding the phosphoric acid I've noticed a milky-blue haze develop in the water. One time it congealed into a jelly-like mass that slowly sank to the bottom (apatite?) and two other times it started to form but eventually dissipated back into the water. I tested the resulting batches of water for total alkalinity (CaC03) and got 42, 57, and 69 ppm. I also tested for phosphate (total and reactive) but the results have been in the negatives (-0.98)? I'm using Hach TNT 870 and 844 kits.

I also tried bubbling C02 into the water with a 0.5 micron stone but the pH hardly dropped after 15 minutes.

Is there a better way to do this process, especially using phosphoric acid for acidification? I've read that apatite gel can form when adding phos acid, but it has not been consistent. Is it due to pH? That said, when I add calcium sulfate/chloride to the treated mash water there is a similar color change and the water goes opaque-blue/white. Lastly, I've seen references to adding untreated water as a means for reducing pH, but I am not sure what that does for the resulting calcium and alkalinity.

Suggestions or insight is appreciated.

 

[mabrungard]

Your water is well suited for Lime Softening. You would get more rapid treatment by targeting a slightly higher pH of about 12. Magnesium hydroxide can be precipitated as a gel if the pH is raised above 11. I'm not sure why you observed a gel when phosphoric acid was added.

Bubbling for 15 minutes with a little stone is no where near enough CO2 to reduce pH via that method. It would be hours. We use large air-stripping towers to impart enough CO2 when dealing with municipal water supplies.

Adding untreated water to the treated water is known as split-treatment, but that does mean that you'll have more of the ions that you're trying to remove. Split treatment is not recommended if you need the Ca and Mg levels to be low. However, brewing doesn't really require those ions to be that low.

 

[bierhaus15]

Thanks, so the 'gel' that forms is not apatite but magnesium hydroxide? Moreover, is there any issue using phosphoric acid to bring the pH down, or is the amt of phosphate added insignificant? I would suspect so given malt contributions are likely much higher.

Is there any benefit to precipitating magnesium hydroxide, given my Mg levels are not terribly high?

 

[mabrungard]

I'm not aware of a reactant that is gel when phosphoric acid is used, so it might be magnesium hydroxide. Still iffy about it either way.

26 ppm Mg isn't excessive in most styles, but it might be notable in more delicate styles.

 

[ajdelange]

If you take the pH high enouugh Mg(OH)2 gel will form and precipitate. If you decant at this point the Mg in that portion of the water you added the Ca(OHJ)2 to will be removed along with most of the bicarbonate/carbonate. What is left is excess Ca(OH)2. If you went as high as pH 12 the OH- concentration will be 10 mEq/L alkalinity (500 ppm) that must be neutralized. At pH 11, however, the OH- concentration is only 1 mEq/L (50 ppm alkalinity). At this point you have really achieved your goal. The water has been decarbonated (carbonate alkalinity reduced to near 0) and the total alkalinity is about 1 mEq/L. The pH is 11 but who cares? The pH of the water going into the mash is high but that doesn't matter as long as the alkalinity is low. I think a lot of people care because almost everyone neutralizes some of that OH- alkalinity. The acid used by most is HCO3- a solution of which you have very handy in more of your source water. The typical practice is to add more source water until pH falls back to 8.4. As long as there is plenty of calcium in that source water (you have a little over 4 mEq/L but your alkalinity is a little under 5 so you might want to add a bit of CaCl2 to help out) more chalk will precipitate. When finished you would wait for settlement and decant again. If you didn't want to get rid of the Mg (and at the level you have I see no reason to) there would only be one decantation. You will have water with about 1 mEq/L alkalinity. Note that this is similar to using carbonic acid (bubbling CO2) as when that is done the first reaction is CO2 + H2O + OH- --> HCO3- + H2O. The bicarbonate then neutralizes another OH- by HCO3- + OH- --> CO3-- + H2O, the CO3-- then being grabbed by a Ca++ ion and precipitated. Starting with CO2 you get two OH- neutralized as opposed to 1 using HCO3- as the acid but, as you have seen, getting enough CO2 into solution to do the job is problematical whereas there is plenty of HCO3- in your source water.

Another approach is to try to calculate exactly how much Ca(OH2) is necessary to precipitate the bicarbonate and to treat 3 samples with that amount and that amount ± 10% (this is the procedure as described by De Clerck). You then check the alkalinity of each and use the lime addition which was most successful. I usually don't advocate that method because for most home brewers 3 alkalinity checks would be too tiresome but as you have the TNT technology (more on that later) it would not be a problem for you.

Finally, there is the method you used: neutralizing the remaining alkalinity with another acid. If you add a little phosphoric acid to a high pH solution which also contains a fair amount of calcium the pH will only fall back a bit at so apatite (and several of the other myriad calcium phosphates) will precipitate. Depending on the chemistry the precipitate will redissolve as more acid is added and the pH falls back. So yes, the gunk you are seeing is doubtless largely apatite. I wouldn't worry about the added phosphate. You only have (from pH 11) 1 mEq/L OH- to neutralize and at pH 8 (no point in going lower than that) each mmol of phosphoric acid will neutralize 2 mEq of them so you only need 1/2 mmol/L. Some of that is going to precipitate and that which stays in solution will be swamped by the phosphate contained in the malt a kg of which carries about 20 grams of it. But you could, of course, use another acid such as hydrochloric which would be ideal as you are a bit shy on chloride.

In reactions where precipitation is involved it is often difficult to determine exactly what will happen but the rule of thumb here is that you can get down to 1 mEq/L which is pretty close to what you got i.e. 0.84, 1.14 and 1.38 mEq/L.

Finally a comment on the TNT tests. It is obviously fantastic where you need to run a quick alkalinity check (or checks as in the De Clerck method) but you know you have to give up something to get convenience like this and it is accuracy. You are probably aware of this but the RFID chip on the package contains calibration data for the lot and by waving the package in front of the instrument's sensor you cause it to be displayed on the screen. The accuracy implied is not that fantastic. Maybe I am deceiving myself that I am doing better with a digital titrator. In any case the TNT results are plenty good for checking on decarbonation by lime addition. As to the negative value for phosphate how can that be? Your test sample is showing absorption less than your blank? Something is wrong there.