[OLD] Convergence problem

[webmaster]

From: Biniam Zerai

Subject: Converegence problem

I am trying to model using GWB version 3.2 a possible reaction for on going experimental work. My main input are swapped CO2(g), mineral Chrysotile, and a brine composed of 0.5 kg H2O, Oxalic_acid(aq), H3PO4(aq), HPO4. When I run in REACT it won't converge. I tried different amount of the brine main composition but still won't converge. Am I missing something? Anybody with suggestion would be greatly appreciated. I include the input with this email.

INPUT

time start = 0 days, end = 100 days

temperature = 80

swap Oxalic_acid(aq) for O2(aq)

swap H3PO4(aq) for H+

swap CO2(g) for HCO3-

swap NaCl(aq) for Cl-

total kg H2O = .5

total g/kg Oxalic_acid(aq) = 8.87

total g/kg H3PO4(aq) = 14.24

fugacity CO2(g) = 68

total g/kg NaCl(aq) = 1

balance on HPO4--

total g/kg HPO4-- = 1

total g/kg SiO2(aq) = 1

total g/kg Mg++ = 1

total g/kg Na+ = 1

react 25 gram of Chrysotile

kinetic Chrysotile rate_con = 4e-17 surface = 46000

 

From: Craig Bethke

Subject: Re: Converegence problem

It's hard to know exactly what you're trying to do, but the root of the convergence problem is the way you are constraining pH. In this input file, you are effectively fixing pH by ionic charge balance. This is theoretically possible, but you must be very careful to set precise concentrations for each of the basis entries. The reason for this is that you are fixing a small number (the concentration of H+) in terms of the imbalance among a series of large numbers (the concentrations of the major components). The fact that you have set a number of components to a concentration of 1 g/kg suggests to me that you do not have constraints accurate enough to set pH in this fashion. Also, I note that you swapped the NaCl ion pair for Cl-. In your input, then, you specify a larger mole number of Na+ than Cl-. (And remember that the mole weights of Na+, Cl-, and NaCl(aq) differ, so 1 g/kg of each is a different mole number.).