hist-brewing: Water Chemistry
burningb at burningbridges.com
Fri Jan 14 15:09:18 PST 2000
Jeff Renner <nerenner at umich.edu> wrote, in reply to:
> OxladeMac at aol.com asks
> >Are there any chemists or advanced brewers out there that could field a
> >question on water chemistry adjustment?
> >City Ca++ CO3-- Cl- Mg++ Na+ SO4- HardnessTDS
> >London 90 125 20 5 15 40 235 300
> >Stillwater, OK 75 40 296 64 279 116 140 456
> Your best bet is to
> build the water from scratch using deionized/distilled water.
I definitely concur on that advice. However...
> You probably won't be able to completely duplicate the ions given for
> London water because they don't balance.
Are you sure? The Stillwater numbers don't balance, and the discrepancy is
way too big to be likely explained by extra common hydrologic ions like
silicate or fluoride ( The sodium number looks suspect at least since even
in brackish waters it usually closely follows chloride); but for the London
numbers a quick BOTE shows, in milliequivalents per liter:
Ca++: 4.50 CO3--: 4.16
Mg++: 0.42 Cl-: 0.55
Na+: 0.65 SO4--: 0.84
Within the limits of experimental error, and certainly within the limits of
natural variation encountered in water supplies (often +/- 10% for
individual ionic species), the numbers provided _do_ balance.
Of course, the quantity that's really in question here is the number for
carbonate; what does it really represent? It can't mean actual carbonate
(CO3--) concentration: the solubility product for calcium carbonate is
around 3.31*10^-9 (mol/l)^2, and since the molar concentration of both
calcium and carbonate is reported at around 2 millimolar, even taking into
account Debye-Huckel activity coefficients of around 0.5 for each species at
that ionic strength, the product of their concentrations exceeds the s.p. by
about 3 orders of magnitude. Also, at near neutral pH in dilute solutions
bicarbonate ion (HCO3-) will greatly dominate in equilibrium with
carbonate, so it wouldn't make any sense to report actual carbonate
concentration. My guess is that the carbonate number quoted represents
equivalent ionic CO3-- molarity, with each 2 moles of bicarbonate present
being equivalent to 1 mole of CO3--.
This is the amount of carbonate you would measure if you first standardized
the sample by heating it to 100C to disproportionate bicarbonate into
carbonate and CO2 and to drive off dissolved CO2 in the sample. It would be
a reasonable approach to get consistent results, since the amount of
dissolved CO2 in a sample is highly variable, depending on how long it has
been in contact with the air, at what temperature and the surface
area/volume of the sample-- even distilled water left out in contact with
the air will eventually have a pH of around 5.7 due to solution of the CO2.
On the basis of consistency I would conclude that the quoted carbonate
number must represent equivalent ionic carbonate molarity, not quantifying
neutral species CO2 or H2CO3.
That being said, the London water composition can be almost exactly achieved
on a per gallon basis with the following easily available additives:
[for one gallon distilled water:]
Epsom Salt (MgSO4.7H2O): 196 milligrams
Unhydrated Plaster of Paris (2CaSO4.H2O): 115 mg.
Table Salt (NaCl): 124 mg.
Baking Soda (NaHCO3): 32 mg.
Powdered Chalk (CaCO3): 772 mg.
In a later post, Jeff writes:
> You could add the chalk directly to the water rather than the mash and
> bubble CO2 through it to dissolve it. This mimics nature, where CO2
> saturated rainwater percolates through limestone.
Yes; a similar and easier way it to dissolve the above ingredients (times
how many gallons your batch will be) in a bottle of cold seltzer water
(making certain it's the unmineralized variety) and stored this way for
eventual use. Then just add the seltzer water with the distilled water into
the mash when it's needed.
> His recommendation for London Water (per US gallon, 3.78 l, of distilled
> English Ale -- A London well-water profile. 3.7 grams Epsom salt, 2.9
> grams baking soda, 2 grams chalk (add to mash), 2 grams canning salt, 0.3
> gram gypsum. Ca=46, SO4=85, Mg=19, Na=83, Cl=64, CO3=173, Hardness=195,
Those amounts look way too large to me. The numbers aren't internally
consistent. Consider just the calcium part:
46 ppm = 46 mg./l (*3.84) = 176.6 mg Ca / gal
But from the additive part:
2 g. CaCO3 (*40[Ca]/100[CaCO3]) = 800 mg Ca
300 mg. gypsum (*40[Ca]/172[CaSO4.2H2O]) = 70 mg Ca
(don't know if there is Ca in "canning salt"? Assume no-- is this just
Total: 870 mg Ca/gal from the formula.
Obviously there is a discrepancy here. My guess is that the quantities are
_not_ per U.S. gallon as stated, but rather for a 5 gallon batch. Including
the stated quantities on a per gallon basis would generate ionic levels more
in line with the Stillwater near-brackish (and probably not very good for
To unsubscribe from this list, send email to majordomo at pbm.com containing
the words "unsubscribe hist-brewing" (or unsubscribe hist-brewing-digest, if
you get the digest.) To contact a human about problems, send mail to
owner-hist-brewing at pbm.com
More information about the hist-brewing