Conceptual Models > Program coupling

Inhibiting Redox Reaction

<< < (3/3)

dlparkhurst:
It does not seem logical to me to define both kinetics and equilibrium for the same reaction. Unfortunately for you, PHREEQC operates the way I think is logical. You have to learn the way I think, which may be unfair, but you are kind of stuck with it, or move on.

Just to conclude, defining redox disequilibrium for multiple redox reactions is arduous with PHREEQC. If that is really what you want, you should probably look elsewhere. However, redox reactions, mediated by bacteria, must follow thermodynamics so that the bacteria obtain energy. It is easy in PHREEQC to add an electron donor (or acceptor) kinetically and let thermodynamics determine which electron acceptor (or donor) is most favored. That is what the bacteria are doing, so it is not that unreasonable an assumption. If you get down to details, you probably do not  have the information to determine the rates of each redox couple individually, but you can add as much complexity as you want.

GeeqC:
Your previous comment provided further clarity. If I understand it correctly, the kinetics module in Phreeqc serves the purpose of adding or removing componds with a specific rate from and to the system (i.e. the solution), while the compounds in the solution remain at equilibrium.

This is of course a valid approach, which allows to accurately simulate partial systems. Nevertheless, this does not exactly represent the view of an inhabited planet where life locally reduces entropy, maintaining a state of strong disequilibrium (Lovelock's view).

From a conceptual point of view, simulating a solution at disequilibrium would not even be so complex. The the thermodynamic equilibria could still be used, just that the reactions would proceed with a rate that depends on the departure from equilirium.

So far, the option of decoupling redox states, as you suggested above, is feasible. The question is only, how I can retrieve the departure from equilibrium. This would require the difference of the existing concentrations from the concentrations that would exist at equilibrium.

dlparkhurst:
I assume the Bethke paper provides affinity terms for the reactions they consider.

For minerals, the saturation ratio (1 - SR) is used as the deviation from equilibrium. There would be analogous terms for the deviation from equilibrium for the master species of two valence states. At equilibrium,


--- Code: ---logK = log(a(Fe+3)) + log(a(e-)) - log(a(Fe+2))

--- End code ---

You will have to define the system with something like Fe_tri and Fe_di, the deviation from equilibrium will be related to the quantity log(a(Fe_tri+3)) + log(a(e-)) - log(a(Fe_di+2)) - logK or something like 1 - [a(Fe_tri+3)a(e-)]/[K a(Fe_di+2)].

Now, the a(e-) is a curious thing. It is probably related to the oxidant or reductant that is actually reacting with Fe_di and Fe_tri. But is it a pe  derived from O2, where you could replace a(e-) with  a(O2) using a relation like 2H2O = O2 + 4H+ + 4e-? Does that mean that all redox active species should use a pe derived from O2? Or perhaps a times it should be the pe derived from NO3-/NO2-. And what is uranium doing at the same time? To me, this soon becomes intractable, so I leave the puzzle to you. I will continue to add O2 (or NO3-) to the solution kinetically and let Fe(2) (and uranium) oxidize thermodynamically at the rate that O2 (or NO3-) is added.

GeeqC:
Yes, indeed, the departure from equilibrium of a reaction can be derived directly from the equilibrium constant.

I had to think for a moment about your concern with respect to the pe. Please correct me if I am wrong, but the pe should be a concept that is based on equilibrium, i.e. based on those redox couples that are actually equilibrating. Compounds that do not spontaneously react (e.g. H2 + O2 = 2H2O, in absence of a catalyst), would not be included in the calculation of the pe. Whether a calculated pe is realistic or not depends on how the solution is defined and which redox couples are selected to participate.

In the extreme case, one may inhibit all redox reactions, in which case the pe does not change. Presumably, it would assume a default value from the initial solution, without further meaning.

The latter case is not so far fetched. It occurs when calculating the initial solution. This shows that switching off redox reactions in Phreeqc would be in principal possible. It would be very practical, if redox reactions during RunCells could be switched off by a simple RedoxOn/Off switch. Then, complex renaming of redox couples in the database could be avoided. This would make PhreeqcRM more universally applicable to simulate aqueous solutions under near-Earth-surface conditions.

dlparkhurst:
Sorry, pe cannot be switched off in PHREEQC. Perhaps you want Geochemist's Workbench; I believe it is straightforward to decouple redox states.

I'm about done with this thread. You know PHREEQC's approach and capabilities.

Navigation

[0] Message Index

[*] Previous page