1D_PhreeqC vs 1D-iCP (-water, diffusivity)

More
6 years 10 months ago #57 by JLP
Hi there,
I have set up a 1D-iCP model that is based on one of mine 1D-PhreeqC models. The PhreeqC model well alighns with experimental data (Sorption on goethite, CD-MUSIC model).

By comparing iCP with PhreeqC model results I learned 2 things:

1) More a comment:
By defining the sites in the chemical domain files as
-sites DENSITY # [moles/L]
the line
- water 0.3 # porosity
hat not effect!

I had to specify the sites as
-sites DENSITY # [moles/(L*porosity)]
in order to get the correct results.

2) The steepness of the PhreeqC BTC is slightly larger (dahed line in attached *.png) than that from the iCP result (blue asterisk), though I do used the same values for the diffusivity. I have no explanation for that.
It is not a mesh effect. This I tested.

Thanks for your comments!
JLP
Attachments:

Please Log in or Create an account to join the conversation.

More
6 years 10 months ago #58 by andres.idiart
Hi,
Regarding comment 1, keep in mind that the Phreeqc file that you use when running iCP should include converted queantities for the solids (including the sites density). The values in Phreeqc should be converted from mol/kgw to mol/liter medium.
That is why we specify the amount of water (in your case 0.3) per liter medium (in a fully saturated case this value corresponds to the porosity of the system), and then the sites density also in mol/liter medium.

About comment 2, these differences in the time evolution in concentration could be in part due to the tolerances in Comsol. Check (and adjust to more accurate) tolerances, both relative and absolute.

Regards
The following user(s) said Thank You: JLP

Please Log in or Create an account to join the conversation.

Time to create page: 0.170 seconds
Powered by Kunena Forum