Q. X1t, Residuals too large, 833-th interation

[Donghun Pak]

Dear GWB member.

This is Donghun

I have a stupid question. Here is my X1t data file, from this modeling, I'd like to make a diffusivity of cobalt in the groundwater system with major reactants such as quartz and K-feldspar. But here are the error message.

Solving for composition of fluid "inlet".
Residuals too large, 689-th interation
Largest residual(s):
                       Resid     Resid/Totmol   Cbasis
---------------------------------------------------------
 Cl-                  0.007018   1.961e+200   3.564e-203
 

What should I do to keep calculating for my modeling? Please let me know.

Thanks a lot

Donghun

X1t Question.x1t

[Jia Wang]

 Hello Donghun,

The error message means the program can't calculate equilibrium state of your "inlet" fluid. You can try troubleshooting in React for just the fluid chemistry first. Right-click and drag the Fluids pane from X1t to a new instance of React. When I disabled the charge balancing ion and ran the calculation in React, the results show a charge imbalance with an excess of anion. You would likely want to pick an abundant cation (like Na+) instead of an anion for charge balancing.

I also noticed that all you use relative units for all components except for cobalt concentration. Perhaps this was a mistake? If you use absolute units for the total concentration of a component, the amount of that solute won't scale with the total amount of fluid in your system.

I think the more dominant species of cobalt is HCoO2-, given the pH in your system. You can try swapping to the predominant species to help the program converge.

Additionally, I also noticed a couple other things in your input file that you might want to check.

In the Reactants pane, a sliding pH path increments the pH from what you set in the system initially to a final value. This is not typically used in reactive transport models and you have it set to the same pH value as the start of the simulation. Is this what you had intended?

I also suggest that you check your mineral volume and porosity. I noticed that with the volume of mineral and porosity, you have specified over 100% of your domain. This doesn't seem correct.

I noted that you have very large specific surface areas (reactive surface area per gram of mineral) for kinetic minerals. The built-in rate law in the GWB uses the specific surface area to calculate the surface area used to calculate the kinetic rate. It would be good to check that your specific surface areas and the rate constants are correct. You can find the built-in rate law for kinetic reactions in section 4.2 Kinetics of precipitation and dissolution of the GWB Reaction Modeling User Guide.

Hope this helps,
Jia Wang
Aqueous Solutions LLC