'dumping' minerals during mixing of fluids

[Jeff]

Hi, I'm attempting to using GWB to model the mixing of hydrothermal fluid with seawater. To start with, I am simply mixing a hot, low salinity fluid high in iron and sulfide with an oxic, high salinity, cold fluid. As expected, rhombic sulfur and pyrite saturate very quickly, but the model allows them to stay in the mixture (i think) and they end up re-dissolving. The model then gets stuck in a loop where it can't decide whether to swap Fe(II) for pyrite or visa verca. Is there a way of 'dumping' mineral phases as they form? i.e. in the context of hydrothermal mixing - form solid sulfide structures which do not remain in the package of fluid which we are considering?

 

Thanks,

Jeff

 

(i've tried to attach my .rea file)

plume mixing 17Nov.rea

[Jeff]

By the way, I realise that I will eventually need to apply a kinetic model to this, as allowing each iteration to reach thermodynamic equilibrium isn't appropriate in this context, and time is very important in hydrothermal mixing.

 

Thanks,

Jeff

[Brian Farrell]

Hi Jeff,

 

It's probably a good idea to start with an equilibrium model, as you are doing, and then add complexity (like kinetic rate laws) as you gain more understanding of your system. As for "dumping" minerals to prevent them from redissolving, there are several reaction configurations within React that may prove useful. There is in fact a "dump" configuration, but I think the flow-through option would be best for the problem you are describing. From Craig Bethke's Geochemical and Biogeochemical Reaction Modeling text:

 

"In a flow-through reaction path, the model isolates from the system minerals that form over the course of the calculation, preventing them from reacting further... In the dump option, once the equilibrium state of the initial system is determined, the minerals in the system are jettisoned. The minerals present in the initial system, then, are not available over the course of the reaction path. The dump option differs from the flow-through model in that while the minerals present in the initial system are prevented from back-reacting, those that precipitate over the reaction path are not."

 

You should take a look at Section 2.2.5 of GBRM or Chapter 3 of the GWB Reaction Modeling Guide (Release 8) for more discussion and examples.

 

Hope this helps,

 

Brian Farrell

Aqueous Solutions

[Jeff]

Thank you Brian, that was very helpful. I also found a hydrothermal example in the textbook you mentioned - and with a combination of the two answers I solved that problem. I've also had some success with Fe++/Fe+++ kinetics. How would you go about using a kinetic redox reaction at the same time as adding a reactant? For example, can you set up a rate law and then run a simulation where you add 1kg of reactant seawater to the basis every 1 minute, and only allow equilibration over this time? (rather than thermodynamic equilibrium) I'm imagining some sort of loop function where you instruct the program to add 1kg reactant, react for 1 minute, plot the results, and loop.

 

Let me know if you have any ideas!

 

Jeff

[Brian Farrell]

Hi Jeff,

 

Assuming you've equilibrated 1kg of a fluid, picked it up and made it a reactant, and are mixing that into your other fluid, just add the line reactants times 10, for example, to the command line (or from the GUI), set the end time to 10 minutes, perhaps, and the fluid will be added continuously at a rate of 1 kg per minute. Just specify a kinetic rate law for Fe++ oxidation, and you should be all set. The program will move toward overall equilibrium with each step of the reaction path, incrementally adding your fluid, evaluating your kinetic rate law for Fe++, and attaining equilibrium for the rest of the species at each step.

 

Hope this helps,

Brian

[Jeff]

Thank you,

that seems like a really silly question now!

 

Jeff