Re: PHREEQCi

["David L Parkhurst" <dlpark@xxxxxxxx>]

> I am using PHREEQCi (1.3) to help elucidate solute provenance and
potential
chemical reaction processes from chemical analysis of glacial meltwaters
collected at the snouts of various glaciers.

> Having used PHREEQCi for a while for inverse modelling calculations, I
have
been trying to add my own phases, i.e. (i) different mineral stoichiometry,
and (ii) different phase reaction steps. Although I am probably incorrect,
I
have wanted to do this because I wanted to simulate particular reactions,
not just to get an idea of the final mole transfers from particular
minerals. For example the weathering of Anorthite to Kaolinite, NOT to the
end member H4SiO4.

> e.g. CaAl2Si2O8 + 2H2CO3 + H2O = Ca+2 + 2CO3-2 + 2H+ + Al2Si2O5(OH)4

> not as defined in MINTEC database : CaAl2Si2O8 + 8H+ = Ca+2 + 2Al+3 +
2H4SiO4

In this case, you do not need to redefine the PHASES data for either
inverse or forward modeling. In inverse modeling, you need only include
Anorthite and Kaolinite (without quartz or chalcedony) in the
INVERSE_MODELING -phases, and the models will have to form kaolinite. For
forward modeling, the equilibrium is independent of the precise reaction
that is used to define Anorthite (provided the log K's are internally
consistent).

You are getting an error because PHASES requires dissociation reactions
that have the mineral as the first entity on the left-hand-side of the
equation, and aqueous species for the rest of the reaction. So you can't
use kaolinite, but you can use any aqueous Al and Si species, provided you
have a log K for the anorthite reaction that you specify. However, you
shouldn't have to redefine the phases unless you want to change the
stoichiometry of the mineral or change the log K for the reaction.

For inverse modeling, only the stoichiometry of the mineral is important,
not the equation or the log K. In the equation you were trying to redefine,
the H4SiO4 is dissolved silica. There is a mole-balance equation for
silica, so if you add Si from dissolution of anorthite, more than is in the
final solution, then the excess silica must go into another silica phase.
In the extreme, if you constrain kaolinite as the only silicate that can
precipitate, then the silica has to go into kaolinite.

David

David Parkhurst (dlpark@xxxxxxxx)
U.S. Geological Survey
Box 25046, MS 413
Denver Federal Center
Denver, CO 80225