Surprising Salifert result after boiling

[Hogarth]

For my last brew I tried boiling the liquor rather than adding CRS to reduce the alkalinity. I boiled it for 30 mins and let it cool overnight. The next day there was a lot of visible Calcium Carbonate at the bottom, but according to a Salifert test my Alkalinity was still at 160mg/l of CaCO3, down from around 230. I'd expected it to be much lower. Should I boil for longer, maybe? Or add gypsum before boiling? I'm a bit out of my depth here. Any advice much appreciated!

[Eric]

Same happened for me the last attempt at boiling, down to 170 ppm from possibly 240 as I didn't take an initial measurement. Put it down to either not getting out all CO2 in solution or it being reabsorbed to dissolve the precipitated carbonate.
I've managed in the past the get alkalinity down to 70 mg/l CaCO3 but more often than not it's been above 100. On that last occasion I used my last 5 ml of CRS and some citric acid crystals and brewed a darker beer.

[orlando]

Very interesting. I know boiling is not an exact science but that suggests it is problematic for people who use it and go on to rely on GW's water calculator that assumes 20 ppm IIRC.

[keith1664]

If you're leaving the precipitate in the water overnight then it's quite likely that it's reacting with co2 and going back into solution. I believe adding gypsum makes the precipitate collect into bigger lumps which drop out of solution quicker.

[keith1664]

Also the practical minimum when boiling is 50 mg/l.

[chastuck]

My water out of the tap has nearly 250 alkalinity and by boiling I have never got this below 80. I rack off within 2 hours to avoid reabsorption.

[Hogarth]

Ah, thanks, didn't know about reabsorbtion. Next time I'll try racking it off sooner.

[Hogarth]

Update: as an experiment I put two saucepans to boil on the stove for 30mins, one containing two litres of water and the other containing two litres of water plus 1/4tsp gypsum. I measured the alkalinity a few minutes after the end of boil (cooling the sample) and also the next day after leaving the water in the saucepans overnight. The results were (in mg/l of CaCo3):

Immediately afterwards: Plain water 60; water+gypsum 40.
The next day: Plain water 63; water+gypsum 35.

Again from a starting alkalinity of 230.

It wasn't a particularly accurate experiment -- I found it hard to keep the two saucepans boiling equally -- but my tentative conclusions would be:

* It makes no difference to the alkalinity level if you siphon the water off the precipitate shortly afterwards or the next day.
* Gypsum helps.
* I can't explain the discrepancy between these readings and my reading of 160 for the 30-litre batch I did previously, but perhaps I didn't boil the latter vigorously enough.

[EoinMag]

From what I understand boiling only precipitates temporary hardness.


That might be an explanation?

[Eric]

From what I understand boiling only precipitates temporary hardness.


That might be an explanation?

Alkalinity is, for all practical purposes, temporary hardness.

Yes Hogarth, again similar to my experience on the hob. An hour on the hob would boil my pans dry while in the boiler only 10-20% would boil off. Another thing is that the deposit quickly sticks to the metal pan while my boiler is plastic and the only place it bonds is on the element leaving the rest sitting around the side on the waterline.
I've just topped up with 2 litres of acids, but they don't come without problems either.

Eric.

[Graham]

Update: as an experiment I put two saucepans to boil on the stove for 30mins, one containing two litres of water and the other containing two litres of water plus 1/4tsp gypsum. I measured the alkalinity a few minutes after the end of boil (cooling the sample) and also the next day after leaving the water in the saucepans overnight. The results were (in mg/l of CaCo3):

Immediately afterwards: Plain water 60; water+gypsum 40.
The next day: Plain water 63; water+gypsum 35.

Again from a starting alkalinity of 230.

It wasn't a particularly accurate experiment -- I found it hard to keep the two saucepans boiling equally -- but my tentative conclusions would be:

* It makes no difference to the alkalinity level if you siphon the water off the precipitate shortly afterwards or the next day.
* Gypsum helps.
* I can't explain the discrepancy between these readings and my reading of 160 for the 30-litre batch I did previously, but perhaps I didn't boil the latter vigorously enough.

There is the issue of micro-particles of carbonate remaining in suspension and not settling out. This can give artificially high readings with the test kit. The alkalinity of both my primary tap water and boiled brewing liquor lowers if I shove it through a water filter before testing. I am sure that this is due to particles in suspension being taken out. Others have suggested adding precipitated chalk to the boil to provide nucleation sites for the micro-particles to aggregate on and thereby precipitate out. It is okay in theory, but I have not tried it because it might negate the workings of the next paragraph.

Gypsum helps because of Common Ion effect, sometimes known as the solubility product of ions. Calcium carbonate has a solubility of about 47 ppm at room temperature, but to maintain that solubility the calcium ions and the carbonate ions must be present in a fixed relationship to each other to maintain equilibrium; adding more calcium in the form of calcium chloride or calcium sulphate upsets the equilibrium and more carbonate must precipitate in the form of calcium carbonate.

Also bear in mind that titration test kits that use a fixed indicator, such as the Salifert, are not particularly accurate at low alkalinities because the end-point of the solution is at a different pH at lower alkalinities.

[orlando]

The comments so far underline the point that boiling is at best a simple approach to carbonate reduction but still leaves a great deal of uncertainty over its precise effects and of course the extra use of energy.

ps Nice to see you back Graham, what with you and Runwell Steve popping up again maybe we were a little premature with our lament about losing some "old hands".

[mabrungard]

As Eric mentions, alkalinity can be equal to temporary hardness and is in many waters. However if the water has elevated Mg or Na content, then its possible for the temporary hardness to be less than the alkalinity. This same factor could be a reason why the boiling treatment conducted by Hogarth was not as expected.

For instance, Munich water has high alkalinity and hardness. Part of that hardness is supplied by the Mg as shown in it's profile below.

Ca 77
Mg 17
Na 4
SO4 18
Cl 8
HCO3 295

The temporary hardness in this case is equal to its alkalinity (242 ppm as CaCO3). But the secondary factor that limits the ability to decarbonate this water by boiling, is the minimum solubility of the calcium ion. That minimum Ca solubility concentration is somewhere between 12 and 20 ppm. Even though we have a lot of bicarbonate in the water, boiling cannot drive as much of it off since it has to be paired with calcium ions in order to precipitate as chalk. In the case of this Munich water, the calcium can only be reduced to 12 ppm and the bicarbonate content can only fall to 100 ppm. That resulting boiled profile is shown below. This is far less than the 40 to 60 ppm that can be expected as a result of boiling a water with high temporary hardness.

Ca 12
Mg 17
Na 4
SO4 18
Cl 8
HCO3 100

So, possibly this limiting ion is why Hogarth is not seeing the reduction expected. Another factor already mentioned is leaving the boiled water in contact with the sediment too long. Decanting off the sediment should be done as soon as possible after the water has cleared. One technique for enhancing the clearing of boiled water is to add powdered calcium carbonate to the water. That calcium carbonate will not dissolve, but it does provide nucleation sites that help the water's original calcium carbonate that is precipitating out of solution to grab onto those powder granules and flocculate out more rapidly and completely.

[Eric]

As Eric mentions, alkalinity can be equal to temporary hardness and is in many waters. However if the water has elevated Mg or Na content, then its possible for the temporary hardness to be less than the alkalinity. This same factor could be a reason why the boiling treatment conducted by Hogarth was not as expected.

From what you say Martin, assuming Hogarth's water was similar for both trials, would this not suggest that his water has not exceptional magnesium or sodium content?

On our little island I believe exceptions might be the exception.

[WallyBrew]

As Eric mentions, alkalinity can be equal to temporary hardness and is in many waters. However if the water has elevated Mg or Na content, then its possible for the temporary hardness to be less than the alkalinity. This same factor could be a reason why the boiling treatment conducted by Hogarth was not as expected.

From what you say Martin, assuming Hogarth's water was similar for both trials, would this not suggest that his water has not exceptional magnesium or sodium content?

On our little island I believe exceptions might be the exception.

Correct

The magnesium content of London mains water is between 4 and 5 which would equate to an alkalinity, when expressed as calcium carbonate, of between 17 and 21.

The sodium content of London mains water is also balanced with some of the chloride, the remaining chloride being balanced by the potassium.

From memory the solubility of calcium carbonate is 22mg/L so adding this to the 17 to 21 from the magnesium indicates that Hogarth has reached a limiting value.