Residuals too large error (X2t)

[tylerp]

Hi All,

 

I have trying to use X2t to simulate reactions of brine along fracture. There are errors "Solving for composition of inlet fluid, Residuals too large". I tried to run both 2 solutions using SpecE8, there were no problem with charge balance (error is about 0.5%). The solution is acidic (HCl added), and the acidity was added under HCO3- species. How do I enter HCO3- species using acidity?

 

Can someone please help take a look at the file attached?

 

Thanks,

Tyler

brine.x2t

[tylerp]

the concentration of chloride in the solution is very high because of the HCl added. I also tried to set Na+ to be the balance species, but the same error occurred. Thanks!

[Brian Farrell]

Hi Tyler,

 

When you are constraining your system, especially under extreme conditions, it sometimes helps the program converge when the species that are swapped into the basis are somewhat predominant. For example, at very low pH, you might want to swap in CO2(aq) for HCO3-, or at high pH, swap in CO3-- for HCO3-.

 

Hope this helps,

 

Brian Farrell

Aqueous Solutions LLC

[tylerp]

Hi Brian,

 

Thank you very much for your suggestion and quick response. I will try it out.

 

Best regards,

Tyler

[tylerp]

Hi,

 

This is to follow up. I have swapped CO2(aq) for HCO3- and also converted the script to X1t instead as a simple 1D model would work for my application.

 

Unfortunately, there is still error below:

 

Solving for composition of inlet fluid.

N-R didn't converge after 999 its., maximum residual = 1.#R, Xi = 0.0000

Largest residual(s):

Resid Resid/Totmol Cbasis

---------------------------------------------------------

---------------------------------------------------------

Can someone please help? Brian Farrell? The revised script is attached.

Thanks,

Tyler

Brine.x1t

[Brian Farrell]

Is your measured acidity a "total acidity"? In addition to the water acidity (H⁺ + OH^- = H₂O) and carbonate acidity (CO₂ + OH^- = HCO₃^-) of a fluid, "mineral acidity" can sometimes be quite substantial (for example, Al⁺⁺⁺ + 4 OH^- = Al(OH)₄^-). What you enter for the HCO₃^- component (whether CO₂ is swapped into the basis or not) should be the sum of the concentration of carbonate species (CO₂ + HCO₃^- + CO₃^-- + NaHCO₃ + ...). If you're entering acidity titration directly for a fluid with substantial mineral acidity, that could be the source of your problem.

 

Section 15.1 of the Geochemical and Biogeochemical Reaction Modeling textbook describes an example in which a titration model is used to interpret an alkalinity measurement and determine the carbonate component concentration, which is what you want. You could do the same thing to interpret an acidity measurement.

 

Finally, you're using alkalinity units for the HCO₃^-/CO_2, but at a pH of 2 you should basically have no alkalinity. You should enter the carbonate concentration in normal concentration units, like moles/kg or mg/kg.

 

Hope this helps,

Brian

[tylerp]

Thanks, Brian.

 

Yes, the measured acidity is total phenolpthalein acidity (titrated o pH 8.3) in which mineral acid (HCl) dominated (I added concentrated HCl to the solution).

 

Apparently, the measured acidity might not be correct (acidity = 1186 mg/L as CaCO3 which is equivalent to ~23.72mmol NaOH). However, I run the React module to titrate this solution (pH 2) with NaOH. It requires about 100 mmol to bring pH to 8.3. In this run, I assume CO2aq is in equilibrium with atmospheric CO2 (during titration) so I used HCO3-/CO2 = 1.2E-05 mol/l, which is calculated from Henry's law.

So it comes to this question:

(a) The solution is in equilibrium with N₂ filled in head space during the experiment. Is it correct (acceptable) to assume the CO_2(aq) is in equilibrium with very low P_CO2, then calculate CO2aq = K_CO2xP_CO2 according to Henry's law? this method should give HCO3-/CO2 = <1.2E-05 mol/l.

OR

( should I calculate CO_2aq from CO_{2 (aq)} + H₂O --> H⁺_(aq) + HCO3^-_(aq), assuming HCO3-= H+ = 0.01 mol/l? this method gives HCO3-/CO2 = 0.01^2/4.45e-7 = 224.7 mol/L (which is too high, is it odd?)

Thanks,

Tyler

[tylerp]

I think the acidity is not correct because the measured pH (pH 2) was measured a few days prior to titration for acidity. So I used HCO3-/CO2 = 1.2E-07 mol/l (solution in equilibrium with pCO2 = 1% of pCO2 in the atmosphere) and successfully run the reactions of the this solution with some minerals using React.

 

I also used the same solution and minerals to simulate kinetic reactions along surface of fracture with fluid flow using X1t, the simulation ran properly (at 25oC, fixed constants). Now I just need to make some changes to add activation energy and pre-exponential factor so kinetic reactions can be calculated for higher temperatures.

 

Thank you very much for your help, Brian.

Tyler

[Brian Farrell]

Hi Tyler,

 

It sounds like you have this figured out and working to your satisfaction. I'll leave you with one last tip for now that may be useful for simulating titrations or other reaction processes in the future. Sometimes one sets the initial fugacity of a gas then lets it evolve during a reaction path. Other times it's desirable to "fix" the fugacity of a gas if the external reservoir is sufficiently large and maintains equilibrium with the fluid over the course of the reaction path. You can take a look at 15.3, Gas buffers, in the GBRM text for more info.

 

Best,

Brian

[tylerp]

Hi Brian,

 

Thank you very much for your help. I find the GBRM book very useful.

 

Best wishes,

Tyler