Tom Meuzelaar

Dear GWB Users: The University of Illinois and RockWare, Inc. are proud to announce the official release of GWB version 8! Version 8 promises to be our most exciting release to date, offering significant speed enhancements and dual porosity model in the Professional version, and the brand new Geochemist’s Spreadsheet (GSS) in the Essentials, Standard and Professional versions. Speed enhancements and dual porosity modeling GWB Professional version 8 comes with significant speed enhancements. The new version uses multi-threading / parallel routines to take advantage of the extra processing power that multi-processor computers offer. Reactive transport model simulations in v8 on multi-core machines run significantly faster than they would in v7 on single-core computers! GWB Professional version 8 also offers new dual porosity modeling capabilities GSS - THE spreadsheet for water data The Geochemist's Spreadsheet (GSS) is a great place to store all of your water data. GSS comes with GWB v8 (Essentials, Standard and Professional) and offers the following features and benefits: Fast speciation, mineral saturation and gas fugacities for multiple samples Convert units for numerous samples with one click Create cross plots, series and time series graphs for multiple analyses Intelligent cross plots- symbols, colors and marker size by analyte or sample Your favorite plots: triangular, Piper, Durov, Schoeller, Stiff, Radial, Bar Chart, Pie Chart Overlay GSS data on redox-pH, activity-activity diagrams in Act2 Compare replicates and check against standards Check analytes against regulatory limits Quick fluid mixing Sort spreadsheet data by name or value Launch SpecE8 or React for any sample, or for multiple samples in batch mode! Please visit the RockWare GWB web pages for additional details and pricing information. Feel free to contact me directly by email (gwb@rockware.com) or telephone (303 640 5526) with any questions. Regards, Tom Meuzelaar RockWare, Inc.

Dear users: If you use compiled rate law or heterogeneity functions, remember to relink your DLLs when you upgrade to GWB8. When you relink, be sure to: Since GWB8 is multithreaded, functions should be compiled with the /MD flag. Use the GWB8 header files (ratelaw_param.h, node_param.h, gwb_context.h), installed in \Program Files\GWB\src. Link with a current .lib file (react.lib, x1t.lib, or x2t.lib). These are in \Program Files\GWB. In functions to be used with X1t or X2t, avoid writing data to static or global variables in the C++ source. This avoids having different threads writing to and reading from the same memory location, which of course can have undesirable consequences. Regards, Tom Meuzelaar RockWare, Inc.

Hello: thermo_pitzer.dat is an outdated database that is included with GWB only for compatibility with early GWB releases. I recommend that you use the Harvey-Moller-Weare compilation, thermo_hmw.dat, which treats solutions in the Na-K-Mg-Ca-H-Cl-SO4-OH-HCO3-CO3-CO2-H2O system at 25C. You might also take a look at thermo_phrqpitz.dat, an expanded version of the H-M-W database, as implemented in the PHRQPITZ geochemical model, which has some provision for working at temperatures greater than 25C. I hope that helps, Tom Meuzelaar RockWare, Inc.

Hi Stefan: A-phi is a fitting parameter, not a fundamental constant. The value used for A-phi in the green book appendices is a value used to parameterize that method for calculating E-theta. The values in the thermo datasets correspond to the virial coefficients in those datasets, according to the source that compiled them. There's no need to worry about the fourth decimal place in a calculation that is not even accurate to two decimal places. As far as your temperature dependency question, Aphi = 2.303 adh/3 (p. 118 of the green book). As you noted, adh values are given in the thermo dataset for the 8 temperature values defined for the database. A polynomial fit through the 8 values allows GWB to compute values at other temperatures. I hope that helps, Tom Meuzelaar RockWare, Inc.

Hi Jose: I'm guessing the problem stems from the fact that the v5 installation is not 64-bit compliant, hence the pre-upgrade install is never valid to begin with. I'm going to send you a full installation executable for v7.0.5. I'll do this via personal email. Regards, Tom Meuzelaar RockWare, Inc.

Hi Fabio: Not sure I completely understand- but you can take the activity values predicted by the speciation calculation in React (look at the text output file) and use them to create a corresponding activity diagram with Act2. Is this what you are looking to do? Regards, Tom

Hello: All of the components you list are part of the default database, so no need for modification. Most of the components can be added directly via the add button in React: The only additional modifications you need to make are decoupling HS-/SO4-- redox couple (via the Config - Redox Couples menu) if you want to include both reduced and oxidized sulfur, representing a system in redox disequilibrium with respect to the sulfur component. You'll want to make the a similar decision for NH4 (swap for NO3- or decouple). There are a number of threads on this forum that discuss redox equilibrium (example), so you might refer to those, as well as Craig's book. Finally- you have three constraints for the carbonate system- pH, dissolved CO2 and bicarbonate. You'll want to pick two of those to constrain the carbonate system- three is over-constrained. If you choose pH and bicarbonate, for example, you can look at React's calculated speciation for HCO3- and CO2-- as a check for equilibrium. I hope that helps, Tom

Julie: It looks like the attachment upload functionality has been fixed. You should be able to upload any React script now. If you upload a thermodynamic database, change the file extension from .dat to .txt. Regards, Tom Meuzelaar RockWare, Inc.

Indeed, it looks like the attachment upload functionality has stopped working. Thanks for bringing this to our attention- I have asked our web server support technicians to address this. It looks like the problem with your script is the units you are using for some of your dissolved concentrations. For example, the concentration for HPO4-- is set to 406.77 g/kg, Ca++ is set for 265.87 g/kg etc. Did you mean for these to be mg/kg? I also recommend that you start with major anions and cations and add in your metal components one by one. That is, be sure your model runs for a simple system, and then add in complexity gradually. I hope that helps, Tom Meuzelaar RockWare, Inc.

Hi Julie: Can you attach your React script so I can take a look? Regards, Tom Meuzelaar RockWare, Inc.

Dear GWB users: We are pleased to announce a GWB workshop, this June, in Canberra, Australia for GWB Essentials, Standard and Professional users. The June 22-23 workshop is designed for: Current users of GWB Essentials, GWB Standard and GWB Professional who wish to become more familiar with the software's interface and features. Those interested in reviewing the basics of geochemical modeling (speciation models, activity models and activity diagrams) Those interested in learning detailed reaction path modeling (using redox disequilibrium, kinetics, and surface complexation) Those interested in learning the basics of 1D/2D reactive transport modeling Registration fees are $899. Students can register on stand-by for $299. For all the details, visit the Australia GWB workshop page on the RockWare website. Please don't hesitate to contact me with any questions. Regards, Tom Meuzelaar RockWare, Inc. 2221 East Street Golden, CO 80401 ph. 303 640 5526

Hi Jono: Faradays and equivalents are actually closely related: An equivalent is the amount of substance that will react with one mole of electrons (1 eq Na+ = 1 mol Na+, 1 eq Ca++ = .5 mol Ca++) A Faraday represents the amount of charge (in Coulombs) per mole of electrons. Therefore, 1 equivalent is often the same as 1 Faraday (or 1000 meq = 1 Faraday). The primary difference is that the more traditional cation-anion balance methods contrast anion and cation concentrations on a 'per kg solvent' or 'per kg solution' basis (usually meq/kg or meq/L). So for more concentrated solutions, or for systems where solvent mass is not 1 kg, you'll need to adjust units appropriately. When comparing charge balance differences reported by GWB vs. other methods, also keep in mind that the methodologies used to calculate anion and cation concentrations can vary significantly. I hope that helps, Tom Meuzelaar RockWare, Inc.

Hello: You can set silica activity as follows: If you want to model Amorphous Silica precipitation, you'll either want to create a kinetic model, or simply suppress the more stable silica polymorphs (Quartz, Tridymite, Chalcedony, Cristobalite) using the Config - Suppress option. I hope that helps, Tom

Hi Julie: Sorry for the delay- since I'm not much of a programmer, I had to consult the GWB developers on this. I'll relay their advice to the best of my understanding. You can accomplish this in one of two ways. The first is to set up a loop that cycles through all the minerals in the database and report those with saturation index reaching a certain criteria. For example, to report all minerals with SI > -5, use the following code: set minerals [eval report all_minerals] foreach min $minerals { set si [eval report SI $min] if { $si > -5.0 } { puts "$min $si" } } There is one problem with this- the control script seems to fail for multi-word minerals in the database such as "Sanidine high". One work-around would be to remove spaces from multi-word minerals in the database, but this is tedious and less desirable. A better approach would be to use GWB Professional's Remote Control feature. You can read about the Remote Control feature in the appendices of the GWB v7 Reference Manual. An example Tcl remote control script that accomplishes the same thing, but does not run into the multi-word issue: source RC_helper.tcl OpenGwbApplic {"/Program Files/gwb/spece8.exe"} SendCommand "read freshwater.sp8" SendCommand "report all_minerals" minerals foreach min $minerals { SendCommand "report SI \"$min\"" si if { $si > -5.0 } { puts "$min $si" } } SendCommand quit There's a higher learning curve involve with the latter approach, but in the long run this will open the door to extremely powerful applications. I hope that gets you going in the right direction! Tom

Hi Julie: My version definitely runs without the error. All of the minerals being flagged in your screenshot are Fe-bearing, and the mole weight errors all vary by exactly the same amount, so my suspicion is that your copy of thermo_phreeqc.dat was compromised. I would guess either the mole weight value for the Fe element or Fe++ basis species. Easiest fix is probably just to download a new copy of the database from the GWB thermodynamic database page. Let me know if that fixes the problem. Regards, Tom

Hi: I think this archived thread on this topic will answer your questions. If not, please let me know. Regards, Tom Meuzelaar RockWare, Inc.

Hi Julie: Nope- didn't see that message. Can you send me a screenshot showing the exact message? Thanks, Tom

Hello: I'll walk you through the steps for Erianite, so that you can do this yourself in the future. Stilbite and Laumontite are already in the extended version of the LLNL database: thermo.dat.v8.r6+.dat. First, note that there are many other discussions in this forum, in Craig's book, and elsewhere about the dangers of adding thermodynamic data from separate sources, or derived by variable methods, to a thermodynamic database. The LLNL database (thermo.dat) is widely considered to be the most internally consistent and reliable thermodynamic database for general use - add data at your own risk. Here's one way to add Erianite (K4Al4Si14O36:15H2O) to the database: Step 1: determine the number of basis components involved in the reaction Components in the "basis species" section (and sometimes "redox couples") of the database are used to construct the reactions for minerals, aqueous species and gases. Since Erionite has K, Al, Si, O and H2O, we'll need the following components: H2O, K+, SiO2(aq), Al+++ and probably H+ for charge balance. Step 2: write the reaction for primary components, then balance with H2O, H+... (be sure to balance for charge) Start with arranging the minimum required components: K4Al4Si14O36:15H2O -> 4K+ + 4Al+++ + 14SiO2(aq) + H2O Excess positive charge on the reactant side is balanced with H+ on the product side: K4Al4Si14O36:15H2O + 16H+ -> 4K+ + 4Al+++ + 14SiO2(aq) + H2O Adding up oxygen and hydrogens on both sides leads to the final stoichiometric coefficient in front of H2O: K4Al4Si14O36:15H2O + 16H+ -> 4K+ + 4Al+++ + 14SiO2(aq) + 23H2O There are 5 basis species in this reaction. IMPORTANT: before you start modifying the database, give it a new name, so you don't corrupt the default databases. In case you corrupt a database, don't worry, you can download original copies here. Step 3: find a reaction in the database with the same number of species in reaction, make a copy, and change the parameters to fit the new mineral (or aqueous species/gas). Be sure molecular weight is carried to 4 decimals (calculate based on mole weights given in the 'elements' or 'basis species' sections in the database). Basis components on the reactant side are negative. Step 4: change the log K values to fit the reaction. Note the log K values here still reflect the values copied from the mineral listed below the Erianite entry (Chinochl-14A). Step 5: increment the number of minerals (or aqueous species/gases) in the header section by the number of minerals you have added to the database. If you've done this correctly, the database will load without error. I've attached a modified copy of thermo.dat with the Erianite entry. Regards, Tom

Hi Julie: Turns out the solution is rather simple. In the default LLNL database element list (thermo.dat), Fe is termed 'Iron', whereas in the PHREEQC database element list, Fe is termed 'Fe'. So: - the command 'decouple Iron' works for thermo.dat - the command 'decouple Fe' works for thermo_phreeqc.dat I have changed the decouple command in your script and verified that it works. Hope that helps, Tom

Hi Julie: Can you attach your script so that I can take a look at it? Regards, Tom

Hello: All 3 of these are found in the extended LLNL database thermo.com.v8.r6+.dat, which you can add via the File - Thermo data... menu. I hope that helps, Tom Meuzelaar RockWare, Inc.

Hello: For starters, be sure you change the file format from .txt to .rea. If the file still will not run, please include it as an attachment and I'll be happy to take a look at it. Regards, Tom Meuzelaar RockWare, Inc.

Hi Julie: Very minor syntax error- try capitalizing iron: 'decouple Iron' should work. fyi- to test any of these commands, try typing them out in the Command pane of SpecE8 or React. Hope that helps, Tom Meuzelaar RockWare, Inc.

Hi: Have a look at the revised thermo database and compare to your earlier version. There were a number of issues with the original entries for Chabazite and Phillipsite, including: chemical reactions were unbalanced (both entries were missing the H+ component) molecular weights did not reflect the mineral dissolution equation one of the components used to balance the equations was not a database basis species (Al2O3) be sure to set the '# of species in reaction' modifier be sure that all numbers are carried out to 4 significant units (use appending zeroes if necessary) The equation I used for balancing Phillipsite is as follows: K3Al3Si5O16-6H2O + 12H+ -> 3K+ + 3Al+++ + 5SiO2 + 12H2O You'll still need to verify that the log K values reflect the actual equations used in the database, or your modeling results will be erroneous. I hope that helps, Tom Meuzelaar RockWare, Inc.

Hello: Can you attach the actual modified database so I can take a look at it? Regards, Tom Meuzelaar RockWare, Inc.