Brian Farrell

Hi Sam, Sorry your post escaped out notice. I've moved it from the archive of old posts to the GWB forum front page. It looks like you didn't use the entire set of input commands. The partial script below only defines the surface to be used in your model. If you go to Craig's GBRM text, you'll find that the next few paragraphs are used to build the model step by step. First, Na+ and Cl- are added to form a simple electrolyte solution. Next, a "sliding pH" path is defined. You set an initial and final pH value - React will then incrementally increase the pH and repeatedly solve for the distribution of mass at each step. This produces the model you see in Figure 14.8. This just demonstrates how the surface sites protonate and deprotonate with changes in pH. Finally, a more complex fluid with more cations and anions is defined and the simulation repeated, producing the model shown in Figures 14.10 and 14.11. The specific error you mentioned arises because there is no Cl- (chloride) in your system. This is the default ion used to enforce change balance. You can choose to turn charge balance off, or to balance on a different major ion. See the "balance" command in the GWB Reference Manual for more information. As for considering sorption of additional species, you'll need to check the surface datasets (and possibly the literature) to see whether any data exists. Ca++ is accounted for, and I believe there is estimated data for Mg++ sorption to HFO, but you may need to look for Al+++ sorption data. Hope this helps, Brian Farrell Aqueous Solutions LLC

Hi, It looks like GSS doesn't like some of the analytes being listed as elemental equivalents. In your case, the polyatomic ions like B(OH)3, SeO4--, etc. are fine. You can define them "as B" or "as Se," but the single element ions, like Pb++, cause the problem when defined "as Pb." For now, you'll have to change the "as..." setting for the appropriate analytes. We'll take a look at the code regarding this issue. Hope this helps, Brian

Dear GWB users, Just a gentle reminder: October 8 is the last day to early register for the Reactive Transport Modeling short course to be held November 29-30 in Seattle. Register by Monday and save $100! Going to AGU? Swing through Seattle and join us! Regards, Brian Farrell Geochemist

Dear GWB users, We are pleased to announce our latest maintenance release, GWB 9.0.2. 9.0.2 features numerous improvements to the GWB plug-in, including wrappers for C++, Fortran 90, Python, Perl, and Java. Update from 9.0 or 9.0.1 at no charge to ensure you have all the newest features and bug fixes. Regards, Brian Farrell Aqueous Solutions LLC

Hi, We'll take a look at this. Would you mind attaching the GSS file itself? Thanks, Brian Farrell Aqueous Solutions LLC

Hi Gus, Are you adding component/ species concentrations, or actually specifying/ calculating the alkalinity? Would you mind attaching your GSS spreadsheet, so that we can take a look? You can also send it to support@gwb.com if you'd like. Sorry I didn't notice your post earlier. I've moved it from the archive of old posts to the main forum page. This helps with automatic email notifications, and puts new questions on the front page for people to see. Regards, Brian Farrell Aqueous Solutions LLC

Hi Sam, If you attach your modified dataset I can take a look, but I don't think you want to add H2S(g) or CuS as new basis species. In fact, H2S(g) and Covellite (CuS(s)) are already included in thermo.dat under gases and minerals, respectively. You typically wouldn't want to add these as new basis species anyway, since they can be formed from combinations of the existing basis species H+, SO4--, O2(aq), Cu+, and H2O. Instead, you would add reactions for new redox species (a basis species in an alternative redox state), aqueous species, minerals, or gases. If you need to modify a reaction log K, you can do so from the Config menu - Alter log Ks... Not sure why your script with thermo.dat only included one mineral, although it's possible under the specified conditions that is the only stable mineral. If you attach your script I can take a look. You might also take a look at Figures 5.2 or 5.5 in the GWB Essentials Guide for example redox-pH diagrams in multi-element systems. Hope this helps, Brian

Hi Sanjoy, If you'd prefer, you can send your script to support@gwb.com. We're currently teaching a Reactive Transport Modeling short course in Australia, but we'll try to take a look at this as soon as we can. Cheers, Brian Farrell Aqueous Solutions LLC

Hi Sam, You should double check the modifications you've made, ensuring they follow the pattern of other datasets and the format detailed in the Appendix to the GWB Reference Manual. An error might be something as simple as not enough numbers after a decimal point. If you post your thermo dataset I can take a look at it. If you can't get your own dataset going right away, you might use one of the defaults like thermo.dat or thermo.com.v8.r6+. You can always use the "alter" and "suppress" commands to suit your needs. Hope this helps, Brian Farrell Aqueous Solutions LLC

Hi Hubert, Can you attach a script (or just type in the commands, since there won't be many) because I'm getting a slightly different result. The concentration of the H+ component is negative, but not the free concentration of the H+ species (which is small but still positive). Type "print species = long" to see the concentration of all species (under "Aqueous species"), not just those present at high concentrations. I think you're looking at the section "Original basis total moles" which refers to total component concentrations, not actual species. If you're unsure how a component can have negative mass, consider a simple system with only the components H2O and H+. The species that can be formed from this basis set include H2O, H+, and OH- (OH- = H2O - H+). Since the pH is higher than 7, there will be more OH- than H+, and the overall solution is described by a positive amount of water component and negative amount of H+ component. Chapter 3 of the Geochemical and Biogeochemical Reaction Modeling text will be helpful here. Cheers, Brian

Glad that helped. The pressure you see in the output file is not really a calculated result. Rather, it is simply the pressure specified in the thermodynamic dataset which corresponds to the current temperature. Typically, thermodynamic data is collected at 1 atm (1.013 bars) up to 100 C. Above this the pressure refers to the vapor pressure of water. Brian

Okay, it looks like everything should be set up on the Basis pane. I would probably swap 1 free g Calcite for HCO3- and balance on Ca++. Then just set pH to the measured initial value, and .1 mol/l each of NH3 and Cl-. Then create a polythermal path from the starting value to 70. Hope this helps, Brian

Hi Hubert, What exactly is known about your system? Swapping CO2(g) in for HCO3- is often a good idea if the reactor is open to the atmosphere, but may or may not be in a closed reactor, since the atmosphere is a much larger reservoir than the headspace of a reactor and its CO2(g) partial pressure is more likely to remain unchanged. Do you have a chemical analysis of the reactor fluid (or a recipe to make it)? A pH measurement? Since you have Calcite in your system, you might swap that in for either Ca++ or HCO3-. Another question, how would you best describe your system? Are you dropping Calcite grains into a 0.1 M NH4Cl solution, or adding some volume of 0.1 M NH4Cl into a solution saturated with Calcite? Regards, Brian

Hi Susan, The "pressure" command only applies to the Act2 and Tact programs. It will affect the position of the water stability limits and the fugacity of gases. You can't set pressure by entering a value directly in React or the other programs. The stability of minerals and aqueous species is commonly not greatly affected by pressure under near-surface conditions. Gases, on the other hand, are different. You can set their fugacities (like partial pressure) directly by swapping the gas into the basis (ex. swap O2(g) for O2(aq) and set fugacity to 0.2, the partial pressure in the atmosphere). This is more work, but you might attempt to create your own thermodynamic dataset with data corrected for pressure. Hope this helps, Brian Farrell Aqueous Solutions LLC

Hi DL, It's hard to say what the problem is here. On the one hand, complexes can increase solubility (carbonate reacts strongly with uranium to form aqueous complexes). On the other hand, I know people who have done experimental work with microbial uranium reduction, and they needed to limit the amount of phosphorous in their growth media to prevent precipitation. Uranium chemistry is really complex and an area of active research. It's very possible that important minerals and species are not included in some of the thermodynamic datasets. If possible, I would try to find out what phase or phases people are identifying under similar conditions in the literature. Then check to see whether they are included in any of the datasets, and whether the logK values for their reactions seem appropriate. Hope this helps, Brian Farrell Aqueous Solutions LLC

Hi, Assuming you're using a thermodynamic dataset like thermo.dat, NO3- is the default basis species for N. If you take a look at the entry for NH4+ in thermo.dat you'll see that in order to form NH4+, you'll need NO3-, H+, H2O, and O2(aq) in your system. Alternatively, just go to program Rxn and type in "react NH4+" and it will spit out the reaction: NH4+ + 2 O2(aq) = NO3- + 2 H+ + H2O. Since you don't want to consider oxygen in your system, or redox reactions between NH4+ and NO3-, you need to decouple the redox pair NH4+/NO3-. This will allow you to add NH4+ to your basis without swapping it for NO3- and adding O2(aq). As long as you don't additionally add NO3- to your basis, no N species in that redox state will be considered in your calculations. The thermo.com.v8.r6+ dataset, on the other hand, uses NH3(aq) to represent N species. In this case, there is no need to decouple the reaction between NH3 and NO3- for your particular simulation. You should look into section 2.4 Redox couples of the GWB Essentials Guide or chapter 7 in the Geochemical and Biogeochemical Reaction Modeling text for more on redox disequilibrium. Hope this helps, Brian Farrell Aqueous Solutions LLC

Hello Kaizen, I would probably not remove anions to achieve charge balance when Fe and Al are excluded. You can just turn charge balance off. The lack of Fe and Al species may be the bigger problem here. If you can't find more datasets/ data in the literature, I would recommend creating separate runs using the Debye-Huckel and virial datasets distributed with GWB. Try to compare your results, keeping in mind that with a D-H dataset you'll probably have more coverage of all the components in your system, but with the virial datasets you'll likely have a better handle on the solubility of the minerals that are included. Regards, Brian

Hi Kaizen, The availability, accuracy, and comprehensiveness of thermodynamic datasets will affect how you set up your simulations. Depending on the extent of evaporation, there is a potential to reach very high ionic strengths which might necessitate use of the Pitzer/HMW/virial methods instead of the standard B-dot variation of the Debye-Huckel equation. The thermo_hmw and thermo_phrqpitz are the two datasets distributed with the GWB recommended for use. These datasets do not contain data for many elements, including Cu and Zn. Unless you have access to alternate datasets, you'll need to assess what is most important in your model. If you're interested in those metals, you may need to use a standard dataset like thermo.dat and live with the likelihood of error in estimating activity coefficients. If you'd like to fix oxidation state, that are a few options available. You can add O2(aq)to the Basis (assuming it's in the thermo dataset) and then swap it out for e-. Set Eh to 600 mV, then go to the Reactants pane to "fix Eh." Alternatively, if your water is exposed to the atmosphere, you might swap in O2(g) for O2(aq) and set a fugacity (like a partial pressure) of 0.2, or whatever it may be in your system. You can then fix this gas buffer over the course of your run. Depending on your situation, oxygen gas may be a better constraint than Eh, since there are well-known issues with Eh electrodes. Once your simulation is up and running, you can view how SI, aqueous species, mineral mass, etc. vary as the reaction path proceeds. This may be more useful than a simple Eh-pH diagram, especially since you want the oxidation state to remain fixed. If you post your React script (and thermo dataset if its not from GWB) I can take a look at your specific error. Hope this helps, Brian Farrell Aqueous Solutions LLC

Hi Bron, We're taking a look at this problem. Unfortunately switching thermodynamic datasets in an open script or spreadsheet is not a trivial matter, since there are more differences than just what minerals are included and their stabilities. Different datasets will sometimes choose different basis species to represent the same component (HCO3- vs. CO3-- or Fe++ vs. Fe+++) and can define redox coupling in very different ways. My advice would be to create separate GSS spreadsheets from scratch for different thermo datasets. Once you add in your samples and analytes, pasting in values from your original spreadsheet shouldn't take too long. We're working on improving this for current and future versions of the software. Hope this helps, Brian Farrell Aqueous Solutions LLC

Hi Bron, GWB makes no such correction to the Saturation Index at this time. SI values are reported in the standard fashion as the log (Q/K). I imagine this is partly to avoid confusion, and partly because there are a number of pitfalls (besides the stoichiometry of the mineral formula chosen) in comparing saturation indices of different minerals, at least as an indicator of the amount of mineral likely to form at equilibrium. We can look into whether this might be a beneficial option, however. The issue of interpreting saturation indices is discussed further in Chapter 6 of Geochemical and Biogeochemical Reaction Modeling. Regards, Brian Farrell Aqueous Solutions LLC

Hi Elisabeth, We're hoping to come out with 9.0.2 within the next few weeks. Btw, there are two options for multicolor lines in 9.0.1. You can check the "multicolor" option which is an icon with 6 different colors, or you can select individual lines and choose any color. To use the second option, right-click your line, choose "Edit Appearance..." then next to Line color you can choose colors for lines one-by-one. Regards, Brian

Hello all, Just one more thing to add here. 500.0000 is not a cutoff value for GWB in deciding whether or not to consider a mineral. Entering exactly 500.0000 for one of the principal temperatures indicates to the program that data does not exist for that particular temperature. 500.0001, on the other hand, will be interpreted as a valid equilibrium constant. Hope this helps, Brian

Hi Elisabeth, You're in luck. Since you're using 9.0, I would recommend you download the free update to 9.0.1 to set line colors individually and take advantage of many other improvements and fixes. You can automatically color all lines with a standard 6-color palette, or select the color of individual lines manually. I looked at saving the image as an svg (and other file types) and I'm getting the same error that you are (GWB 8 seems to work okay). We'll look into this and hopefully get it fixed before the next maintenance release. Do you have access to Microsoft PowerPoint? That works pretty well for editing plots. I think the Open Office version will work too. Hope this helps, Brian Farrell Aqueous Solutions LLC

Hi Frank, We're looking into this. Thanks for bringing it to our attention. Best, Brian Farrell Aqueous Solutions LLC

Hi Fernando, I was able to make an Act2 diagram with Enargite using the data you supplied. Perhaps you did not add it to the dataset correctly, or maybe you did not include in your Act2 basis all of the components necessary to form Enargite? If you look at the thermodynamic information you provided above, you'll see the 6 species needed to form it (Cu+, As(OH)4-, SO4--, H2O, O2(aq), and H+). All of those must be present to form Enargite. For example, if I diagram Cu+ on Eh-pH (or O2(aq)-pH) axes, I would additionally need to add As(OH)4- and SO4-- under the "in the presence of" field for Enargite to be considered. If you're still having trouble, you can post your thermo dataset and Act2 script so that I can take a look? You can also email them to me at support@gwb.com. Regards, Brian