Surface not using ferrihydrite concentration

[aleoconn]

Hello,

What I'm troubleshooting here is one step in my model. The overall goal is to model how iron and arsenic in a groundwater system would response to increasing pe. I realize that fixing pe in models is controversial; however, it makes the most sense for this proof-of-concept that we're doing.

What I would like the model to do is for each step up in pe, to calculate the speciation of As and Fe, precipitate Fe(OH)3 (if saturated), adjust the HFO amount accordingly, and calculate As sorption.

In the attached code, I'm attempting to accomplish this using these steps:

1) Define a starting concentration of Fe(OH)3 that can only precipitate (we know some small amount exists throughout the system)
2) Populate an exchange surface according to measured aqueous concentrations
3j) Define the amount of exchange surface using Fe(OH)3

In the output, this all makes sense. However, when I specify a reaction with a new equilibrium phase that fixes pe, the exchange surface doesn't seem to find any Fe(OH)3 to use as HFO, despite it having been previously defined (see attached screenshot). I've also attached the code so you can see the full results. Right now, the pe in the reaction is fixed as the same pe specified for the solution for troubleshooting purposes. In the full code, the last simulation is replicated, but the pe is fixed at a new value.

Thank you so much for any help! Please let me know if I can clarify in any way.

[dlparkhurst]

The initial solution is undersaturated, as is your reacted solution, even after all your Fe(OH)3(s) dissolved.

It is ok to fix the pe by adding O2. In general, I would prefer fixing the O2 log10 partial pressure. The stability field of water runs from about log10(P(O2)) -40 to 0.

[aleoconn]

Thank you for the answer Dave! I would have thought specifying `precipitate_only` in the EQUILIBRIUM_PHASES block would have prevented Fe(OH)3 dissolution, even though it is undersaturated. And, in the phase assemblage result, it shows that there is Fe(OH)3 in the "Final" column. Is there another way to prevent the dissolution?

[dlparkhurst]

Sorry, I did not look closely enough. You are right, you do have the expected amount of Fe(OH)3; it did not dissolve.

I guess the precipitate_only and SURFACE-related equilibrium phase do not work together. Not sure I ever tried it. The logic is a bit tangled for each type of calculation.

As I understand it, to get exactly what you described, you would have to make two calculations, one without the dissolve only condition, and one with the fixed number of sites (without Fe(OH)3 in the SURFACE definition) and without Fe(OH)3 in EQUILIBRIUM_PHASES. You would have to choose which calculation to use based on whether Fe(OH)3 dissolves.

[aleoconn]

Hi all,

I noticed something interesting while working thorough my solution (based on your suggestion to define the exchange phases differently whether or not Fe(OH)3 is saturated).

When the surface is first defined, the number of moles of Fe(OH)3 and of HFO are equal. However, it appears that when I redefine the equilibrium phases to fix pe, the amount of HFO ends up being defined as the change in the Fe(OH)3 moles rather than the total amount. I've attached two screenshots of the results and included the input file to show how the code is set up.

Any thoughts on why this might be happening? I'm guessing it has something to do with how I'm setting up the reaction. Thank you!

[dlparkhurst]

I think you need to remove the "precipitate_only" directives. A surface related to a precipitate only mineral does not work.