Precipitating minerals in a complex system

[Olasz Tamas]

Dear Community,

I am working with a red mud sample and adding acids to it. I am using the llnl database since I have many different elements (even some REE) included. I am using REACTION to add acids to the sample, usually 100 moles of reaction (HNO3, HCl, H2SO4, H3PO4) in 50 steps.

How I approach minerals precipitation is I run the model, I select which minerals have positive SI, and after that I define these minerals in the EQUILIBRIUM_PHASES block as for example: Wollastonite 0 0. My problem is, as I add the acids the pH changes significantly, and the mineral equilibria with it. There is no way I can check all the 50 steps in my models for all the positive SI-s. It just occured to me that for the H3PO4 model, Hopeite precipitates just for a couple reaction steps, and if I dont go check that particular step and define Hopeite 0 0, than my end results will be different (not what I want in this case).

Is there an option in PHREEQC (or in geochemical practice) to solve this issue? What I would need is PHREEQC to always just print out the mineral phases that have a positive SI, and their amounts precipitated. Or is this a bad approach due to practical/theoretical reasons, and if so, can you help me understand why?

Thanks a lot,
Tamás

[dlparkhurst]

It is possible to add an inclusive list of all minerals that you think might precipitate in EQUILIBRIUM_PHASES with 0 SI and 0 moles of each. The stable phase assemblage should be found at each reaction step.

That said, it is unlikely that some minerals will precipitate in reality, even though they are calculated to be in the stable phase assemblage. Hematite is the least soluble iron oxyhydroxide, but would probably not form in your experiment. Rather, ferrihydrite or goethite would be more likely to form. Similarly, many primary rock minerals are unlikely to form under the conditions of your experiment.

Here is a Basic script that extracts all of the minerals appropriate to the system defined by SOLUTION. It may help you sort through all of the many possible minerals; you could also use it to define an EQUILIBRIUM_PHASES data block.

Code: [Select]

SOLUTION 1
Ca 1
P  1
Si 1
USER_PRINT
-start
10 max_si = SYS("phases", count, name$, type$, phase_si)
20 FOR i = 1 TO count
30   PRINT PAD(name$(i),20), STR_F$(phase_si(i),10,2), \
               " ", CHR$(35), " ", PHASE_FORMULA(name$(i))
40 NEXT i
-end
END


[Olasz Tamas]

Thank you Mr. Parkhurst for your answer. I decided to take your advice and include the minerals which I can expect in my system, and built a model with those. I think the results makes sense, but I will also play around with your script, which looks extremely useful. Now I have a better understanding of these more complex systems.