Re: Nucleation kinetics in PhreeqC?

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

> I'm a PhD student in the Research Center Karlsruhe at the Institute of
Nuclear Waste Management. I'm working on a DFG project regarding the
bassanite-gypsum-transition together with Dr. Dirk Bosbach.

> We try to model the reaction kinetics of bassanite dissolution, gypsum
nucleation and gypsum growth to develop a predictive reaction model for
plaster setting with PhreeqC. We determined the kinetic constants for the
reaction rates by batch and flow-through experiments. Then we used PhreeqC
succesfully to model bassanite dissolution and gypsum growth, but the
nucleation is still missing. We have determined the nucleation rate.

> Now my question: Can you give me hint, how to integrate gypsum
nucleation/nucleation rate in PhreeqC? I've attached my input file and the
database we use.

In your input file you have M0 = 1e-20, which is then included several
places in the rate. This makes the rate very sensitive to M0, which I don't
think is a good idea. So you should reconsider how to take account of M0.

I don't know much about nucleation kinetics, so I can't give you a
definitive answer. It seems like you need a more complex rate expression.

Perhaps you want to keep track of the mass and surface area of a number of
particles. If you know the nucleation rate and the initial diameter of the
particles, you could add particles of a given diameter at a rate (probably
dependent on SI) and store them with PUT statements for each time step. I
guess you could just store particle diameter as PUT(n,diameter). You would
have to revisit previously nucleated particles to calculate rates and
update their diameters. It might be tricky to apportion the overall rate to
the collection of diameters.

A simpler approach would be to make the rate a function of total moles
precipitated (but not M0) and SI (or SR).

I know Dennis Eberl has a set of programs called Galloper (I think) that
implements a model of crystal growth, though not connected to a aqueous
model. It uses a law of proportionate growth for the particles. I have not
seen a mechanistic explanation for the applicability of the law, but it
seems to account for crystal sizes. You may want to try to implement his
approach with a number of particles.

David

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

Project web page: https://wwwbrr.cr.usgs.gov/projects/GWC_coupled