Dien Li Posted November 6, 2012 Posted November 6, 2012 Hi, Brian I like to calculate solubility of TcO2 and other compounds in high pH solution. I used REACT to do it. I set the solution chemistry in the basic, and set the solids in reactant. However, it indicated first the charge imbalance, I added Cl for charge balance, but after several hundred runs, it could not converged, and said that the high residue for Cl. Attached please you can see the sample for quartz. It is the same for other solids. Could you help me? You can also reply me at dien.li@srs.gov. Thanks. DL Quartz solubility in pure water.rea
Brian Farrell Posted November 7, 2012 Posted November 7, 2012 Hello, Regarding the charge balance issue, there are a couple fixes. Since you're at high pH, there will be more OH- than H+ in solution. Additionally, some of the silica will be present as neutral SiO2(aq) and some will be H3SiO4-, so you'll actually have an excess of negative charge. Cl- is the default charge balancing ion, but you are free to use any other ion. Balancing on a cation, Na+ for example, makes your model converge. Another option is to simply turn charge balance off, as is sometimes done when making speciation diagrams across a range of pH values. Now that your model runs, you can take a look at the results. You'll find that the quartz quickly reaches saturation and ceases to dissolve, and that the concentration of all silica species remains constant after this. You should note that the time you specify here doesn't mean much, since you haven't specified a kinetic rate law by which quartz precipitates or dissolves. Making the run last 7 days simply specifies that the 2 g of quartz be added in even increments up to 7 days, but the time could change without affecting your results. Another way to create solubility diagrams is to set the fluid in equilibrium with your mineral, quartz here, at a certain pH, then using a sliding pH path to see how the solubility of the mineral changes with pH. If you look at the diagrams page of our website, you can scroll down to the React section and choose the "Mineral solubility" example. Click on the React icon to launch a precalculated model of mineral solubility. You might try experimenting with this model to start, and adjust it as you see fit. Hope this helps, Brian Farrell Aqueous Solutions
Dien Li Posted November 9, 2012 Author Posted November 9, 2012 HI, Brian Per your last email, the time set up is meaningless if no kinetic rate is set. The 2 gram quartz was assumed to be added invrementally over the setting time. Now, if I want to model the case: I have or add 2 gram quartz into the system once in the beginning, I like to model how solubility changes with time and second, with pH. Could you advise on how to set this model. Also, in my previous sample, I tried again, it looks that addting Na or Cl and keep it as a charge balance, the modeling is not convergent due to high residue. If I add Cl and keep balance off, it works. More comments about this. Thanks. DL
Brian Farrell Posted November 9, 2012 Posted November 9, 2012 Hi DL, In order to make your original script run, not only do you need to add Na+ to the Basis, but you must tell GWB to use it as the charge balancing ion, instead of Cl-. To do so, use the command "balance on Na+". I think that should solve your problem. To model dissolution of a mineral with time, you need to set a kinetic rate law. Since you want the 2 g of quartz in your system at the beginning, you should swap it into the Basis in place of SiO2(aq) and set its mass to 2 free g. Using the rate law and the current chemical composition, React will calculate how much quartz precipitates or dissolves at each step. To see how solubility changes with pH, again swap quartz into the Basis pane, set the initial pH, then move to the Reactants pane and specify a target value for a sliding pH path. The mineral solubility example I mentioned earlier demonstrates this. Hope this helps, Brian
Dien Li Posted November 16, 2012 Author Posted November 16, 2012 Hi, Brian How are you! I still need your help in modeling of TcO2 solubility in cement leachate solution of high pH. I input the cement leachate chemistry in BASIC and place TcO2 in REACTANT, and like to determine the aqueous Tc species concentration, and then solubility of TcO2. Is this a right logic to do it? The key problem is that it is not convergent as I ran this type of simple model. What is the general approach to this. Please you can see attached file. Please you can help run it to see how to make it work. This is urgent to me, I wish you can help me, thanks. DL Solubility of TcO2 in genuine cement leachate.rea
Brian Farrell Posted November 16, 2012 Posted November 16, 2012 Hi DL, It look like your problem is solving for the equilibrium state of the Initial system (before the TcO2© is even added). You should set the mass of solvent water as a free constraint (the default option) rather than a bulk constraint. This ensures all 1 kg of H2O at the beginning of the run is water, not distributed among the various aqueous species and minerals that are present. You should take a look at section 7.2 of the GWB Essentials Guide for more on free constraints. Since you set an Eh of -.4 at pH 12.3, you might want to swap out TcO4- for a more reduced form, such as Tc+++. This isn't too critical in this case, but it does make the initial portion of the path a little smoother, and it's a good thing to keep in mind when troubleshooting your runs. A simple Eh-pH diagram of TcO2 should help explain to you what I mean. You might even consider plotting the reaction trace on the Eh-pH diagram. Hope this helps, Brian
Dien Li Posted November 17, 2012 Author Posted November 17, 2012 HI, Brian Thanks you so much for your reply. I still have the troubleshooting problem for the not convergent runs. Could you have a quick trouble shototing for this data file. Thanks and ave a nice holidays. DL Solubility of TcO2-test.rea
Dien Li Posted November 17, 2012 Author Posted November 17, 2012 HI, Brian For the same topic, I used ACT2 to make phase diagram of Tc in pure water, I suppressed all Tc minerals, except for TcO2© or TcO2.1.6H2O, to determine the phase diagram at Eh = -0.4 V. They indicated that the solubility of TcO2.1.6H2O is higher in the magnitude of ~4 orders than that of TcO2©, which is what I believe. However, when I used REACT to calculate the solubility of TcO2© and TcO2.1.6H2O at pH 8 and Eh -0.4, their solubilities of these two are identical at ~9e-12 M. In fact, in all other aqueous systems, they showed similar solubility, in contrast to the phase diagram. Could you advise, thanks you so much. Is this is a right way to calculate the solubility of TcO2 or any other minerals? Thanks. Please see the attached files. DL Solubility of TcO2-in pure water.rea Phase diagram of TcO2.1.6H2O.ac2 Phase diagram of TcO2.ac2
Brian Farrell Posted November 19, 2012 Posted November 19, 2012 HI, Brian Thanks you so much for your reply. I still have the troubleshooting problem for the not convergent runs. Could you have a quick trouble shototing for this data file. Thanks and ave a nice holidays. DL Hi DL, To make this run converge at your pH of interest (12.66), I swapped out Al+++ (dominant Al phase at low pH) for Al(OH)4- (dominant Al phase at high pH). Since your concentration units are mg/kg, you should be sure to adjust your value to reflect the different molecular weights of Al+++ and Al(OH)4-. Hope this helps, Brian
Brian Farrell Posted November 19, 2012 Posted November 19, 2012 HI, Brian For the same topic, I used ACT2 to make phase diagram of Tc in pure water, I suppressed all Tc minerals, except for TcO2© or TcO2.1.6H2O, to determine the phase diagram at Eh = -0.4 V. They indicated that the solubility of TcO2.1.6H2O is higher in the magnitude of ~4 orders than that of TcO2©, which is what I believe. However, when I used REACT to calculate the solubility of TcO2© and TcO2.1.6H2O at pH 8 and Eh -0.4, their solubilities of these two are identical at ~9e-12 M. In fact, in all other aqueous systems, they showed similar solubility, in contrast to the phase diagram. Could you advise, thanks you so much. Is this is a right way to calculate the solubility of TcO2 or any other minerals? Thanks. Please see the attached files. DL Hi DL, Could you please post the thermo dataset you are using, so that I can take a closer look? Thanks, Brian
Dien Li Posted January 14, 2013 Author Posted January 14, 2013 HI, Brian In this same topic on the solubility colcaulation of Tc solids in cement leachates which have very high ion strength (Na, K and Ca...) and high pH under reducing condition, we used two databases: thermo.com.v8.r6+.com and thermo_phreeqc.dat. My questions are: 1. In general, what are the algebra or math descriptions of activity coefficients in these two models, how differet these descriptions are in the two models. 2. More specific, are the model desciriptions still valid in solutions of very high ion strength, high pH and under reducing conditions? Thanks you so much for your help. Dien
Brian Farrell Posted January 14, 2013 Posted January 14, 2013 Hi Dien, Like thermo.dat, thermo.com.v8.r6+.dat uses an extended form of the Debye-Huckel equation called the B-dot model. If thermo_phreeqc.dat is loaded, however, GWB uses the same activity model originally used in PhreeqC. As far as I can tell, this is the Davies equation (at least for charged species). Both of these activity models are described in Craig Bethke's Geochemical and Biogeochemical Reaction Modeling text, Section 8.1. The B-dot equation is valid to ionic strengths slightly higher than the Davies equation, but both work relatively poorly at high ionic strength. In this case, virial methods like the Pitzer equations may be useful (these have their own limitations, however, notably a lack of data for redox reactions and certain components, and a limited temperature range). You should take a look at Chapter 8 of that text, or Section 7.4 of the GWB Essentials Guide for more on the implementation of activity coefficients and activity models in GWB. As for the validity of your models at high pH or low redox potential, I don't think the activity model used is particularly important. Rather, the availability of data for appropriate species and minerals is likely to be your main concern. The various thermo datasets have their various strengths and weaknesses, and some are more complete with respect to certain components than others. Take a look at Section 2.3 (Uncertainty in geochemical modeling) of the GBRM textbook for an example. Hope this helps, Brian
Dien Li Posted January 15, 2013 Author Posted January 15, 2013 Hi, Brian Thanks you so much. Dien Hi Dien, Like thermo.dat, thermo.com.v8.r6+.dat uses an extended form of the Debye-Huckel equation called the B-dot model. If thermo_phreeqc.dat is loaded, however, GWB uses the same activity model originally used in PhreeqC. As far as I can tell, this is the Davies equation (at least for charged species). Both of these activity models are described in Craig Bethke's Geochemical and Biogeochemical Reaction Modeling text, Section 8.1. The B-dot equation is valid to ionic strengths slightly higher than the Davies equation, but both work relatively poorly at high ionic strength. In this case, virial methods like the Pitzer equations may be useful (these have their own limitations, however, notably a lack of data for redox reactions and certain components, and a limited temperature range). You should take a look at Chapter 8 of that text, or Section 7.4 of the GWB Essentials Guide for more on the implementation of activity coefficients and activity models in GWB. As for the validity of your models at high pH or low redox potential, I don't think the activity model used is particularly important. Rather, the availability of data for appropriate species and minerals is likely to be your main concern. The various thermo datasets have their various strengths and weaknesses, and some are more complete with respect to certain components than others. Take a look at Section 2.3 (Uncertainty in geochemical modeling) of the GBRM textbook for an example. Hope this helps, Brian
Brian Farrell Posted January 15, 2013 Posted January 15, 2013 Hi Dien, I forgot to add the link to the PHREEQC User's Guides from the USGS website. You might want to take a look at the "Activities and Mass-Action Equations" section. Cheers, Brian
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now