Failed mass balance for CO2 if "Two Fluid Phases conditions" employed

Hello everyone,

I am trying to run the geochemical equilibration of brine with a sandstone rock composed of several minerals (mostly quartz 70%, around 4% carbonates and some feldspars and clay minerals). I am using Toughreact 1.2 with ECO2N EOS module.

As long as I am using the "Single Fluid Phase" as initial condition (primary variables P, T, Xsm, XCO2aq=0), the mass balance of all chemical and physical processes involving CO2 works well. But if I change to "Two Fluid Phase" (primary variables: P, T, Sg=0, Xsm), a relatively large XCO2aq value appears, HCO3- also increases by a factor of 50 (thus also changing the pH value by one order of magnitude), a gas phase appears (although Sg was initially set to 0) and the overall mass balance fails by around 25%. This means, that much more CO2 appears in the aqueous solution or in the gas phase than it is released by dissolution of carbonate minerals.

Does anybody have an idea, where this artificial and unbalanced CO2 comes from and how this can be prevented in this conditions mode? Otherwise the subsequent simulation of a CO2 injection into such a system seem to be quite unprofitable, since the chemistry would be ruled by physically non-existing CO2?!

Additionally, I have one question related to this topic. How is the variable XCO2aq (from the TOUGH part of the simulation) related to the chemistry (the TOUGHREACT part) of the aqueous phase, i.e. HCO3- concentration? I have calculated a time dependent mass balance from one simulation on my own and found out, that the amount of CO2 given as XCO2aq and HCO3- (or rather the change of this amounts compared to the initial values at t=0s) appear to be more or less the same if the system is in equilibrium, but they are different, if reactions still take place. Any idea, too?

Thank all of you in advance for your support!