In reading back over this thread I note a lot of concern about the precipitation of flakes of calcium carbonate over time to the point that some have abandoned the scheme for management of calcium chloride proposed in this thread. Early in the thread concern seems to have been focused on the fact that exactly what this material is was unknown an secondarily on that clearly as something is coming out of solution something must be being lost. Common sense says that the material is calcium carbonate produced by carbon dioxide in the air and experiments by several indicate that this is the case as the material fizzes when exposed to acid. I had originally opined that the reaction involved is
Ca++ + 2Cl- + CO2 + H2O ---> CaCO3 + 2H+ + 2Cl-
but we note the release of hydrogen ions implying a reduction in pH. This reaction can indeed tale place but once the pH has been lowered to 8.2 no more CO2 will dissolve as atmospheric CO2 (assumed to be at 0.0004 bar) will be in equilibrium with the precipitated chalk so we doubt this is the responsible mechanism. A little investigation revealed that most CaCl2 produced is as a byproduct of the Solvay process in which limestone and salt (NaCl) are feedstocks for the production of soda ash (NaCO3). This process is carried out at high pH and as a consequence it is not surprising that the CaCl2 is 'contaminated' with some (OH)- ions. We also found a specification for FCC grade CaCl2 (what we buy from the LHBS) which promised alkalinity of less than 0.3% w/w as Ca(OH2). Thus our CaCl2 solution is likely to contain some 'lime water' and we all remember from high school chemistry that lime water is used to detect CO2:
Ca++ + 2(OH)- + CO2 ---> CaCO3 + H2O
It is thus quite reasonable for us to expect that a strong solution of FCC CaCl2 is, if, as inevitably it will be, it is exposed to air, going to precipitate some chalk.
Some may want to know how much of this will precipitate. To start lets look at 100 grams of 'CaCl2' powder. If the hydroxyl contamination level is 0.3 % that means it is equivalent to 300 mg. As the molecular weight is 74.093 that implies the equivalent of 300/74.093 = 4.05 mmol of Ca(OH)2. or 8.1 mEq of (OH)- ions in the solution which, according to the equation above, will precipitate 4.05 mmol of calcium. The molecular weight of anhydrous calcium chloride is 110.98 so nominally 100 grams of that is 1000*100./110.98 = 901.063 mmol and thus our calcium loss would be 4.05/901.063 = 0.45% or 1.5 times the specified alkalinity of the product if that alkalinity is expressed w/w as Ca(OH)2 which in all liklihood it is going to be though in fact not all the hydroxyl ions are going to be paired with calcium though, of course, most of them will be. Given the feedstocks for the Solvay process there will doubtless be magnesium, sodium, potassium and probably other cations present as well.
As, in general, we don't have the alkalinity for the product we bought from the LHBS there may be those that want to be able to obtain some indication as to how much hydroxyl contamination is actually present the the product they are holding in their hands. There is a simple test which tells us how much (OH)- is in a sample and that is one with which we are well familiar: the alkalinity test. Most here are aware that in that test acid of known strength is added to a solution and the pH monitored until the pH is reduced to 4.5 (ISO test). The alkalinity of the solution is simply the number of mEq of acid required to bring a liter of the solution to that pH. The number, called the M alkalinity (or sometimes T for 'total') is often multiplied by 50 to give units of 'ppm as CaCO3' but we won't do that here. What some may not realize is that the test is actually done in two steps with the first being a titration to pH 8.3. The amount of acid required to get to pH 8.3 is called the P alkalinity. By making a solution of the product in question and determining its alkalinity, P and M, we can determine (OH)- concentration.
Let's assume we start, as suggested above, with 100 grams of 'CaCl2' in liter of DI water. We've calculated that for 0.3% alkalinity we would have 8.1 mEq of (OH)- ions in the solution. This implies that the pH of such a solution is going to be 14 + log(0.0081) = 11.91. Thus the first clue that your solution is going to eventually precipitate chalk is a high pH like this. Were the alkalinity 0.15% the (OH)- concentration would be 4.1 mmol/L and the pH =14 +log(0.0041) = 11.61 i.e. it drops by 0.3 for each halving of the alkalinity. Were the alkalinity 0.03% as opposed to 0.3% the pH would drop 1 unit to 10.91 i.e. by 1 unit for each factor of 10 reduction in the alkalinity. So for 0.003% contamination pH would be 9.91 etc. so that it is pretty clear that a pH < 10 means insignificant loss of calcium to precipitation. Thus one really does not really need to do a titration. A simple pH measurement is really sufficient to tell you whether you need to worry about precipitation or not or indeed whether you can expect to even have it.
But as this is the Brewing Science forum, here goes:
The amount of acid required to move a liter of solution from pHs (its sample pH) to pH = 8.3 is the P alkalinity and is, approximately
P = [(OH)-] + Ct*(-1 - Q(pHs)) = [(OH-] - Ct*(1 + Q(pHs))
P is in units of mEq acid added per liter to reach pH = 8.3
[(OH)-] is the concnetration of hydroxyl ions in mEq/L
Ct is the total concentration of carbon atoms in CO3--, HCO3- and CO2 in solution in units of mmol/L
Ct is found from the approximation
Ct ~ (M - P)
i.e. the difference between the M and P alkalinities. It should be small or 0 and is there only if the CaCl2 is contaminated with chalk which, again, given the feedstocks, is reasonable to assume but as it is so insoluble there shouldn't be much.
Q(pHs) can be found from the curve in the Palmer Kaminsky book but is easily calculated from
r1 = 10^(pHs - 6.38)
r2 = 10^(pHs - 10.38)
f0 = 1/(1 + r1 + r1*r2)
f1 = r1*f0
f2 = r2*f1
Q = -(f1 + 2*f2)
Typical value: Q(10.38) = -1.5
Then
[(OH-] = P + Ct*(1 + Q(pHs))
If M and P are close, as we would expect
[(OH-] ~ P
but you can also, in this case, calculate it from
[(OH-] = 1000*10^(pHs - 14)
To sumarize: the stuff is chalk (CaCO3). It is harmless. Very little of it precipitates thus very little calcium is lost. If the pH of a solution of 100 grams/L is below 10 the loss is likely to be less than 0.0045% of the dissolved calcium ion. The solution SG will not be appreciably effected.