thank for your answer. Unfortunately GWB Pro license is not available in my company at the moment, so I cannot try out. Solving this principal problem might however result in purchasing the license.
I think, that the solution you suggest would basically simulate flushing the soil profile by rainwater with a given amount of oxygen initially dissolved. I have set up similar simulation using React in "flush" configuration. The problem is, that the amount of oxygen brought into the system just by the rainwater is not high enough. The resulting sulphate concentration are however unrealistically low (see Model 1 attached). Another problem with this model is, that a lot of rainwater would be necessary to oxidize the pyrite in 1dm3 of the waste rock material (taking it about 1000 years at a given annual rainfall).
I was able to obtain more realistic concentration, while setting the pyrite dissolution rate and oxygen dissolution rate in the "single cell" flush model (see model 2).
However in the kinetic reactive transport approach there seems to be a principle problem how to deal with the O2(g) fugacity buffer in the column representing the soil profile. The oxygen fugacity buffer in the soil profile would decrease from the atmospheric value at the top (0.2) to a very low value at the bottom. This fugacity profile would result from the relative proportion between the rate of O2(g) diffusion in the pore air from atmosphere and oxidation rate of pyrite.
Now the question is, how to set up the diffusion of oxygen in the soil profile pore air? How is it going to influence the drainage chemistry? I suppose, that the weathered zone thickness should increase over time.
It is a bit complex problem and I hope I didn't missed my point.