Model calibration: Ca2+ does not increase when increasing Sm and k25 of calcite in React

[Zhidi Wu]

Hello!

I’m trying to calibrate the model parameters Sm (surface area) and k25 by matching fluid components (cation concentrations) up to lab measurements. But I found that the Ca2+ concentration neither changes much when increasing Sm and k25 of calcite, nor suppressing all minerals from precipitation. However, the saturation index of calcite in output is negative (unsaturated). To my understanding, this means Ca2+ should be below the solubility of calcite and is supposed to increase with Sm and k25. How to explain these opposing observations and increase the Ca2+ concentration (~153 ppm) to match up with lab measurements (~1143 ppm)? Similar observations were also seen when I tried to decrease  Mg2+ by decreasing Sm and k25 of dolomite. 

Many thanks,

Zhidi

[Brian Farrell]

Hi Zhidi,

It's hard to say without seeing your input. You can attach input files here if you'd like someone to take a look.

Regards,

Brian Farrell
Aqueous Solutions LLC

[Zhidi Wu]

Hi Brian,

I attached the file here. Thank you for looking into this in advance, and I look forward to your insights.

Zhidi

Hd_batch_optimized_SuppressAll_1h.rea

[Jia Wang]

Hello Zhidi,

Thank you for attaching your input file. Could you also attach your thermodynamic dataset so we can run your input file?

Best regards,
Jia Wang
Aqueous Solutions LLC

[Zhidi Wu]

Hi Brian,

Of course. Here you go, thanks.

Zhidi

AGI_20e_7-24-2015.dat

[Jia Wang]

Hello Zhidi,

Thanks for attaching your input file and your thermodynamic dataset.

If I render the results in Gtplot and plot Mineral saturation vs. Time from your simulation, I find that Calcite is initially undersaturated with respect to your fluid but very quickly becomes oversaturated, with a Q/K value hovering right above 1. If I plot the amount of calcite (in Gtplot Variable type -> Mineral) available, it shows that there's an initial dip which indicates that calcite dissolved initially and then slowly increases, which means the mineral is precipitating. In this case, Ca is only added to the fluid from calcite in the first instances of the simulation and later removes Ca due to precipitation.

Also, note that Dolomite and Anhydrite are both Ca containing minerals. These two minerals remain undersaturated with respect to the fluid and as they dissolve, they release Ca into the solution. In a multi-component system like this, the overall component concentration reflects the total effect of all reactions.

The default built-in rate law for kinetic mineral reactions is r = A_s * k * (1-Q/K), where A_s is the surface area, k is the rate constant, Q is the ion activity product, and K is the equilibrium constant. In the GWB, a positive r means dissolution and negative r represents precipitation. When you are adjust the specific surface area or intrinsic rate constant in your rate law, the rate is linearly affected. The term Q/K determines whether the reaction is saturated, undersaturated, or in equilibrium. If your mineral is undersaturated, (Q/K is less than 1), then r is positive and the mineral dissolves. For more information on kinetic reactions, please see section 4 in the GWB Reaction Modeling User Guide. I think it would help you understand how the software handles kinetic reaction. If you make an example with just 1 kinetic mineral and look at how the results change with your simulation, without the effects from other minerals.

The question of how to get your concentrations of your simulation to match that of your simulation is rather difficult to say without knowing anything about your system. You want to check that the kinetic constraints you are putting in are reflecting on what is actually occurring in your simulation. Maybe some requires a different rate law for constraining its kinetic reaction. Or maybe some minerals that are dissolved very fast can just be added as simple minerals. I am not really sure what is the most appropriate for your system.

Best regards,
Jia