CShort Posted November 8, 2016 Share Posted November 8, 2016 I'm working with a relatively simple X1T model in which water 1 is injected into a sandstone containing water 2 for 22 years then chased through the system with water 1. During transport, the only attenuation process operating is an activity-KD retardation model. I'm trying to run this model using different porosity's, assuming that the rest of the volume is just non-reactive quartz. This should be a simple model. This model runs for certain values of concentrations and Kd's, but not others, although the model is fundamentally unchanged. The biggest problem is that the model is unable to initialize because the "Newton-Raphson did not converge...". I've been able to duplicate this issue in React. One issue I've noticed causes this problem is declaring a porosity. The modelled sandstone has a porosity of 0.25. When I run my react code using a blank porosity, blank bulk volume, and 0 inert volume (the default values), it's able to initialize. If I declare a porosity of 0.25, it is unable to initialize. Do you have any thoughts on what might be going on here or what I should look out for when trying to declare porosity values? I've referenced both the reaction guide and the reference guide and have not had much luck sorting this out. Thank you! Clark React_Issue.rea Kd_Ambrosia_Por=25_KD=0.029.sdat Quote Link to comment Share on other sites More sharing options...
Brian Farrell Posted November 9, 2016 Share Posted November 9, 2016 Hi Clark, Can you double-check whether the Kd for Pb in your surface dataset is correct? I ask because it's orders of magnitude higher than any of the other sorbing species. This is certainly possible, but you end up with a huge amount of sorbed Pb. The surface reactions aren't actually accounted for in the default setup with 100% solution, since no mineral mass is implied, but by setting porosity to .99, for example, you imply a small sediment mass which allows the surfaces reactions to be accounted for. The system needs a lot of Cl- to balance out the positive charge of the >Pb++ species. I think the problem is related to this. Like you, I'm not able to run the calculation with a porosity of .25, which is presumably the only correct porosity for your Kd. Another question, do the concentrations that you specify on the Basis pane refer to the concentrations measured in the fluid, or for the whole system (i.e. in the fluid and on the mineral surfaces)? If I assume the latter (using the "sorbate include" option), the calculations works just fine, but I'm not sure whether or not that's appropriate for your system. I'll continue to think on this, but if you could check into my questions that would help a lot. Regards, Brian Farrell Aqueous Solutions LLC Quote Link to comment Share on other sites More sharing options...
Brian Farrell Posted November 9, 2016 Share Posted November 9, 2016 Also, are there any more ions (particularly anions) in the fluid that might also be sorbing to your sediments? Quote Link to comment Share on other sites More sharing options...
CShort Posted November 10, 2016 Author Share Posted November 10, 2016 Hi Bryan, Yes that Pb KD is both correct and absurdly high. The measurements are just for the fluid - if the sorbate option is checked, then that's a mistake - it's something I played around but, as you mentioned, isn't appropriate for my model. I too noticed that I can run it using the sorbate option and thought that was curious. I'm not sure I have data handy to add sorbing anions, but that's a good idea. I'm not at all concerned with what the Cl concentration is doing - Na/Cl are both in there just because GWB likes charge balanced solutions. I'm thinking a little bit about creating alternate thermo & surface databases where Pb is not charged to avoid this issue. I'll mess around with sorbing anions as well - maybe I can create a junk anion species with the same KD, concentration, and opposite charge to avoid the issue. I understand these assumptions would misrepresent reality, but the purpose here is to emulate a much older model 1D transport model that doesn't have all the bells and whistles of GWB... actually it has no bells or whistles at all really. It was called SOLUTE ONE-D. It certainly didn't care about charge balancing anything - it didn't even have parameter specific variables. Thank you for your thoguhts Bryan. I'll be messing around with it more today. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.