PhreeqcUsers Discussion Forum

Applications and Case Studies => Acid Drainage => Topic started by: rapheul on November 11, 2016, 08:34:26 AM

Title: convergence errors
Post by: rapheul on November 11, 2016, 08:34:26 AM
Hi, there,
I am not quite sure whether it is suitable to post my problem here, it is about piChem use in FEM which couples PHREEQC.
I am trying to predict the AMD evolution after mine closure, the acid drainage water (pH=2.1) was previously defined as pollutant source for the aquifers evolved. Reactions in aquifers and closed mined stope only include simple mixture, carbonates and gypsum dissolutions and cation exchanges. However, the PHREEQC calculation always failed with convergence errors. The attached are the files I obtained from the FEM model. The question is I do not understand the message and would like to know what may be the problem from it. Then I can modify the input files and reactions.

Some addtional information: each of the input files could run well in PHREEQC alone, but coupled in the 3D FEM transport model, after one reaction, the second step begin to produce error messages. In the error txt, the Raw number 179116 is the corresponding node ID in the model domain, the error position is below the AMD water body where a carbonate water is assigned. Besides, the FEM coupled with PHREEQC runs well if the AMD solution is not added. So I guess the problem probably is due to the adding of AMD that changes the related chemical field but don't know how to modify. Are there any hints from the error messages?

Thanks in advance!     
Title: Re: convergence errors
Post by: dlparkhurst on November 11, 2016, 04:12:36 PM
Try adding the following, either in the database, or in a file that is run by all of the IPhreeqc objects. I think you are at a point where the redox system is unpoised, that is, no pair of redox states has species present in significant quantities for both redox states.

H2O + 0.01e- = H2O-0.01
log_k -9
Title: Re: convergence errors
Post by: rapheul on November 11, 2016, 08:11:46 PM
Hi David,

Thanks so much! I wait to update my feedback until the model complete, and it works now after adding these lines!! That is really helpful!

Best regards
Title: Re: convergence errors
Post by: Pendegrass on October 13, 2017, 02:53:40 PM
I ran into the same issue and this works like a charm. Thanks very much Dlparkhurst.