webmaster

From: Tiziano Boschetti Subject: Processing multiple samples I'm using a GWB 4.0.3 version. I've a database of 200 water samples and I'd like to calculate the saturation indexes. It's possible to copy at the same time more of one sample data in the spreadsheet form (edit--> Paste From Spreadsheet)? It's possible to include in the dataset, other than basis concentrations, a parameter as density?" From: Tom Meuzelaar Subject: Re: Processing multiple samples The ability to process multiple samples is not available in GWB version 4, but is available in GWB version 6 via TCL scripting. In the GWB 6 reference guide which you can preview on the GWB support page at: http://www.rockware.com/catalog/pages/gwbs...ort.html#guides there is an Appendix entitled "Multiple Analyses", which gives a sample script using SpecE8 that speciates and calculates saturation indices for multiple samples which are initially stored in a spreadsheet. You can use the control scripting features with any of the GWB 6 applications. You can set your initial fluid density in React via the Config- Variables menu. The default density unit is g/cm3.

From: Andrew L. Scott Subject: ACT2 and REACT not in agreement I have been trying to model the Yucca Mountain J-13 well water in preparation for performing a sensitivity analysis for U speciation. Attached you will find the Statement of the Problem, some REACT output, the thermo dB, scripts for ACT2 and REACT and some calculation macros. I updated the thermo.com.v8.r6+ dB with the NEA 2003 numbers for U species only at 25 deg C. I hope someone is able to understand and help with my problem. I have tried to be as complete as possible with the materials I have provided, but if there is anything that I have not included or you do not understand something I have done, please feel free to ask. [A note from the admin: The files are not available From: Mark Logsdon Subject: Re: ACT2 and REACT not in agreement You certainly have been complete in your information. I regret that I don't have time to go throuygh all your calculations, but I have a quick question on the two bases you use. Your REACT basis gives Na at 2 mmolar and Cl at 0.2 mmolar. But the ACT2 basis has log a Na at - 3.5 and log a Cl at -2.8. Are you sure you are calculating the acivities properly? If you just take log C (as a quick consistency check) I get -2.7 and -3.7, respectively, so the log a values look funny to me - and in fact, their order doesn't look right either. I haven't tried to do this for other components or to check your spreadsheet. From: Tom Meuzelaar Subject: Re: ACT2 and REACT not in agreement I think I see the problem. If you allow NO3- to speciate over your activity diagram (in Act2, Basis tab, click the right-most arrow for the NO3- entry, and choose "Speciate over X-Y"), you'll get a very similar diagram to your Act2 plot (the "correct" plot) without NO3- (if you additionally suppress Coffinite, the diagrams look even more similar). In your original "incorrect" activity diagram, there is an implicit assumption that NO3- is stable over your entire Eh-pH range- React predicts, however, that most of your NO3- component will occur as the N2(aq) species- you can confirm this by looking at your React output file. By choosing the "speciate" option in Act2, you're allowing the program to predict which nitrogen species is stable- the outcome greatly effects the final diagram. From: Andrew L. Scott Subject: RE: ACT2 and REACT not in agreement Thanks for the feedback, particularly Tom Meuzlaar and Mark Logsdon. Based on their comments, I recalculated the basis and allowed speciation over the X-Y axis in ACT2, and now have perfect agreement (and a greater understanding of process) between ACT2 and REACT. ............... view attachments ........... http://gwb.eligeos.org/attachments/al-scott1.jpg http://gwb.eligeos.org/attachments/al-scott2.jpg ................................................... If the silicon species is not allowed to speciate, quartz will not precipitate, and allows the precipitation of Haiweeite, a uranyl silicate mineral, at equilibrium. This calls into question the kinetics of the two processes (quartz and Haiweeite precipitation/dissolution) with regard to the mobility of the U in the Yucca Mountain J-13 water (represented by the top-most green dot). ............... view attachments ........... http://gwb.eligeos.org/attachments/al-scott3.jpg http://gwb.eligeos.org/attachments/al-scott4.jpg ...................................................

From: Biniam Zerai Subject: Info on CO2(aq) I am trying to model the possible reaction of CO2 gas, albite, annite, siderite, calcite and dolomite in 2 molal of brine solution. I made the correction for fugacity of the gas using Duan and Sun equation of state but I am not sure what is the right way as how to correct the salting-out effect using GWB. I have been told to change the the ionic size in the thermodynamic database that contains the aqueous species, CO2(aq) to -0.5 in order to correct the effect of salinity on CO2(g) solubility. Is this the right way to do it? I did it using the above advise and observe a large difference. Any info, input or comment is highly appreciated. I am using GWB 3.2.2. From: Craig Bethke Subject: Re: Info on CO2(aq) The equations used to calculate activity coefficients for electrically neutral species are given in Chapter 7 of the "Geochemical Reaction Modeling" text. The special meaning of the ion size parameter for neutral species is described in the "Thermo Datasets" appendix to the GWB Reference Manual: The ion size parameter (ao) has special meaning for neutrally charged aqueous species in the thermo dataset. For neutral species with ao = 0, the species' activity coefficient is set to one. When ao = 1/2, the activity coefficient is calculated from the "CO2" coefficients in the data table section, according to equation 7.6 in the "Geochemical Reaction Modeling" text. When ao = 1, the logarithm of the activity coefficient is set to the product B(dot)x I, where B(dot) is given by the data tables above, and I is true ionic strength.

From: Tiziano Boschetti Subject: Correct formula for hydrated minerals I've noted that the water of hydrated minerals in the database thermo_phrqpitz is written as: H2O or ^H2O. So, if I add a new mineral, what's the correct form? From: Tom Meuzelaar Subject: Re: Correct formula for hydrated minerals As far as the software is concerned, the notation you use doesn't matter- what is important is that the formation reaction for the mineral is written only in terms of the 16 basis species in the thermo_phrqpitz database. Maybe another user can shed light as to whether or not there's any significance to the differing notation styles.

From: GOLDEN, D. C. Subject: Evaporation of saline solutions containing Al3+ and Fe3+ I am trying to simulate the evaporation of acidic aqueous solutions containing Na, K, Mg, Ca, Mg, sulfate, Al and Fe. I use the flow through model of the React2 in GWB 4.0.3. I tried to use the data base for the Harvie-Moller-Weare activity model provided in the above package. When I include Al and Fe in the Basis, it responds : Al3+ not in the database, and similarly for Fe3+. Is there a data base for high ionic product solutions which takes into account these ions? How do I update the existing database to include these ions? If any one can help or give suggestions, I will appreciate it very much.

From: Hlanganani Tutu Subject: Modelling geochemical speciation of uranium in sediments I need your help and advice on how I can do thermodynamic modelling of uranium speciation in sediments. Is it proper for instance if I measure the temperature, pH and redox potential of the moist sediment and then leach with deionised water (shake test) and measure sulphates and uranium in the leachate. Finally, I use these results to construct the Eh-pH diagrams. I'm studying the geochemical speciation of uranium in mine tailings and am using GWB release 4.0.1 for modelling. From: Armand R. Groffman Subject: Re: Modelling geochemical speciation of uranium in sediments I worked on a uranium mill tailings project in the USA (UMTRA) and here is what we did to determine U-speciation in pore fluids: Collect pore fluids with a suction lysimeter. If that is not possible performed DI batch tests. If sulphate is the overwhelming anion it is most likely the major complexing agent. Also check into alkalinity. If you need more detail then collect and analyze the pore fluids for organic carbon, major cations and anions (+ NO3), and trace elements including Fe, Mn, and sulfide. Look into the milling process to understand what residual organic products would be in solution and analyze for them or their degradation products. pH is easy. For redox use an ORP probe (a bit wonky) or if the tailings are not too acidic you may be able to determine redox by measuring Mn, Fe, or S-. This may be a better way to understand the redox potential of the system. Your model output will provide speciation and ppt products.

From: Andrew L. Scott Subject: Database Conversion Routines Does anyone have any conversion programs for converting some other database format to that of GWB? I am particularly interested in converting the NIST Standard Reference Database 46- Critically Selected Stability Constants of Metal Complexes- Ver 8, Martell, et al.

Since GWB does not natively run on Linux you have emulate Windows to get GWB running. So far I have tested two options: VMWare and Crossover Office. There is a slight difference between the two. In VMWare you install Windows as your virtual machine and then install GWB as you would in Windows. There are no major problems I know about, however, I recommend a fast machine for VMWare to operate fast enough. The cost is ~190USD (http://www.vmware.com). Crossover Office (based on Wine), on the other hand, will trick GWB (and a lot of other Windows programs) into believing it is being installed and run on Windows. While being a more efficient of the two, this option has more problems. As of now, it works but with numerous issues. It is possible that in the future (perhaps as soon as end of this year) GWB will run smoothly on COO/Linux. I am an official "advocate" for GWB at Codeweavers Compatibility Center. Please see http://www.asmirnov.org/os/gwb for details, screenshots and most up-to-date information. The cost is ~40 USD (http://www.codeweavers.com). Please let me know (alex[at]asmirnov.org) if you have any questions/suggestions/experience with anything mentioned above.

Edward Grew Subject: Cerium redox equilibria Is anybody aware of experimental work on CeO2 - Ce2O3 equilibria? I am looking for a T - f(O2) diagram for Ce oxides. From: J. Michael Palin Subject: Re: Cerium redox equilibria I seem to remember various multiple valence oxide pairs, including CeO4-Ce2O3 and Eu2O3-EuO, shown on a T-fO2 diagram in: Carmichael, I.S.E. & Ghiorso, M.S. (1990): The effect of oxygen fugacity on the redox state of natural liquids and their crystallizing phases. in ¨Modern methods of igneous petrology: Understanding magmatic processes¨, J. Nicholls & J.K. Russell, eds., Mineralogical Society of America, Reviews in Mineralogy, 24, 191-212. There may also be something particular to Ce in: H.D. Schreiber, H.V. Luer, T. Thanyasiri (1980) The redox state of cerium in basaltic magmas: an experimental study of iron-cerium interactions in silicate melts, Geochim. Cosmochim. Acta 44: 1599-1612. From: Kees Linthout Subject: Re: cerium redox equilibria For a logfO2 vs. T diagram of the CeO2/Ce2O3 equilibrium and its calculation from the delta G data, see Appendix 1 in Hanchar et al. (2001, Am. Min. 86, 667-680). From: Eric Essene Subject: Re: Cerium redox equilibria You don't need experiments, just the delta G data for Ce2O3 and CeO2. That is only a crude guide however for the oxidation state of Ce in silicate crystals or liquids, unfortunately.

From: Henry B. Kerfoot Subject: Request for kinetic data (calcite) I am looking for kinetic data on the reaction of calcite with low-pH water. In particular, I wish to be able to estimate an appropriate residence time for pH 3.5 water in commercially available limestone gravel.

From: Kirk J Cantrell Subject: Pitzer and React I am using React to calculate the solubility of a mineral at high ionic strength using the Phrqpitz thermo database. I modified the database to include a solubility constant for a mineral I am interested in, as well as some associated Pitzer parameters. Without giving you all the details, I have determined that the model is not using the new theta parameters for the interaction of two anionic species that I added to the Phrqpitz thermo database. The reason I know this is that at zero ionic strength, React is calculating the solubility correctly. When I go to 2 m NaCl, where the theta interaction parameters has a significant influence, the solubility calculated by React is less than half of what it should be (based on the model used to determine the solubility and Pitzer parameters for the mineral I am working with). If I go back and set the theta parameter in the thermo database to zero, I get the same solubility at 2 m NaCl. This suggests to me that for some reason, React is not including my newly added theta interaction parameter in the equilibrium calculation and that this is the reason the calculated solubility is too low. Do you have any clues as to why this could happen? To make locating my additions to the thermo database easier to find, I have indicated them here: In elements: Uranium (U ) mole wt.= 238.0290 In basis species: UO2(CO3)3---- charge= -4.0 ion size= 9.0 A mole wt.= 450.0554 3 elements in species 1.000 U 11.000 O 3.000 C In aqueous species: (UO2)3(CO3)6------ charge= -6.0 ion size= 20.0 A mole wt.= 1170.1386 g 3 species in reaction 3.000 H+ 3.000 UO2(CO3)3---- -3.000 HCO3- -19.5000 -19.5000 -19.5000 -19.5000 -19.5000 -19.5000 -19.5000 -19.5000 In minerals: Cejkaite type= formula= Na4UO2(CO3)3 mole vol.= 150.000 cc mole wt.= 542.0146 g 2 species in reaction 4.000 Na+ 1.000 UO2(CO3)3---- -4.0800 -4.0800 -4.0800 -4.0800 -4.0800 -4.0800 -4.0800 -4.0800 In beta virial coefficients: Na+ UO2(CO3)3---- beta0 = 0.61 beta1 = 18.2 beta2 = 0.0 cphi = 0.0 alpha1 = 2.0 alpha2 = 0.0 Na+ (UO2)3(CO3)6------ beta0 = 2.4 beta1 = 46.6 beta2 = 0.0 cphi = 0.0 alpha1 = 2.0 alpha2 = 0.0 In theta virial coefficients: Cl- UO2(C03)3---- theta = -0.13 SO4-- UO2(C03)3---- theta = -0.37 Note that (UO2)3(CO3)6------ is not required for the current calculations. This species is required for very high carbonate conditions and I intended to put estimates for its theta virial coefficients in the database later. Attachments to this post: http://gwb.eligeos.org/attachments/React-1...aiteSolPitz.rea http://gwb.eligeos.org/attachments/React-1...hrqpitz_kjc.dat From: Craig Bethke Subject: Re: Pitzer and React I think you will find that the cause of your trouble is that in entering theta values in the thermo dataset, you refer to species "UO2(CO3)3----" as "UO2(C03)3----". In other words, for these entries you've typed in C-zero-3, not C-Oh-3.

From: Burnol Andre Subject: Convert Phreeqc format to GWB format or EQ3 format I am wondering wether a code converting Phreeqc to EQ3 format for the thermodynamic database is available ? I know that, recently, thermo_phreeqc.dat - The thermodynamic database from PhreeqC release 2.8 - is ready for use with GWB. But I don't know how the conversion was made ?! if a code was developed for this purpose, it will be not very difficult for me to adapt this convertor if the sources are available. I am currently using a modified llnl.dat in Phreeqc format and I would like to convert it in EQ3 format. It will be for me a big help not to have to rewrite all my modifications in EQ3 format or to develop myself the convertor.

From: Andrew L. Scott Subject: 'Best' database I am performing uranium speciation of natural ground waters under various conditions, including speciation with organic and inorganic ligands. What are your opinions on the 'best' database to use for this application (please also consider databases that are not necessarily in GWB format, as I can convert if necessary). I am currently using thermo.com.v8.r6+, but want to ensure I am using the 'best' available data. I have available, for example, NIST standard reference database 46, critically selected stability constants of metal complexes v7.0, and could easily obtain v8.0 if necessary. From: J.L. Fernandez Turiel Subject: RE: 'Best' database You can try with the JNC database. The updates are recent. http://migrationdb.jnc.go.jp/english.html

From: Brian Gibney Subject: Fe/As redox and sorption I am modeling the sorption of arsenic to Fe(OH)3 under the following circumstances: 1) Arsenic consistently oxidized, Fe redox slides from reduced to oxidized 2) Fe consistently oxidized, As redox slides from reduced to oxidized The basis for my reaction has been set up as below: hematite/Fe+++ = 1 free gram e-/O2 = 0 Eh Na = 100 mmolal Cl- = 100 mmolal (charge balance) SO4- = 10 mmolal Ca++ = 10 mmolal As(OH)4- = 0.1 mmolal H+ = 7 (pH) Reactants: react 1 g Fe(OH)3 ppd slide Eh to 1 Plots of reactions don't show a decrease in As sorption at low Eh (as iron is reduced), as one would expect. Any suggestions?

From: "Brugger, Joel (SAM)" Subject: Speciate different components over x-y I'm trying to calculate a log f O2(g) vs pH diagram to show relationships between Fe-Ni sulfides. I therefore speciate SO4-- and Fe++ over x-y and make the diagram for Ni++. I get a nice diagram, but something disturbs me. The dashed boundaries correspond to the overlap of the boundaries for the pure Fe-O-H and S-O-H systems. It seems more appropiate that the boundaries for Fe should take into account Fe sulfides (e.g. pyrite), i.e. the Fe should be speciated in the Fe-S-O-H system. So my question is: is there any way to tell GWB to speciate S in the S-O-H system, and then Fe in the Fe-S-O-H system, and then to calculate the nickel speciate over these two diagrams? From: Craig Bethke Subject: Re: Speciate different components over x-y I think you'll find that the scheme you propose is not internally consistent and will not result in a unique solution. To convince yourself, try drawing such a diagram by hand. I think you will find that you end up with more than one way to balance the reactions among the Ni species.

From: Tom Meuzelaar Subject: conductive heating This message is being posted on behalf of Mike Adams, EGI, who is having trouble posting to the list. Message follows below... I am running a simple conductive heating model. When I run the model and allow precipitation at 20 degrees C, I get a pH of 6.909 and 60 grams of dolomite. When I run it from 20 to 200, I get the same answer at 20 C. If I run it from 20 to any temperature over 200, I get 6.866 and 66 grams of dolomite at 20 C. I am running REACT 5.0.3. The run file is listed below. Any suggestions as to why this is happening? # React script, saved Wed Sep 22 2004 by madams title = "Hay Ranch North for Mixing" data = "c:program filesgwbgtdatathermo.dat" verify temperature initial = 20, final = 200 decouple ALL 1.00000751 kg H2O .00561915555 mol HCO3- balance on Cl- .00187730745 mol Cl- 7.89540172e-6 mol F- .000398024745 mol H+ .00349790126 mol SO4-- .00243512974 mol Ca++ .00154700679 mol Mg++ .000221748772 mol K+ .00591566695 mol Na+ 1.61726201e-8 mol B(OH)3 1.44071459e-7 mol Li+ 3.02408076e-8 mol HS- 6.23340356e-8 mol CH4(aq) 8.32164143e-5 mol SiO2(aq) 1.11187295e-7 mol Al+++ suppress Dolomite-ord Dolomite-dis From: Craig Bethke Subject: Re: conductive heating When you run a React simulation, the program by default considers only those aqueous species and minerals whose thermodynamic stabilities are known across the range of temperatures considered in your run. Sounds like your run incorporates one or more species that have stabilities defined in the thermo dataset only from 0-200°C. When you expand the temperature range to 300°C, the species are not loaded. You can get around this by setting the extrapolate option, but of course this is a little dangerous. It would be better to identify the species in question and finding reasonable stability constants for them at high temperature.

From: Brian Gibney Subject: carbonate vs Eh I am modeling solubility of carbonate in a micro-aerobic to anaerobic groundwater system with a sliding Eh to 1.0. Groudwater conditions are as follows. add "H2O" 1 free kg "H2O" swap "e-" "O2(aq)" -.5 Eh "e-" add "H+" pH 6.8 add "SO4--" 2.38 mg/kg "SO4--" add "Fe++" 031 mg/kg "Fe++" add "F-" 09 mg/kg "F-" add "Na+" 5.95 mg/kg "Na+" add "K+" 1.57 mg/kg "K+" add "Ca++" 10.4 mg/kg "Ca++" add "Al+++" 018 mg/kg "Al+++" add "Mg++" 1.7 mg/kg "Mg++" add "SiO2(aq)" 25 mg/kg "SiO2(aq)" add "Cl-" 86 mg/kg "Cl-" balance on "Cl-" add "HCO3-" 54.6 mg/kg "HCO3-" The output is as follows Solving for initial system. Loaded: 130 aqueous species, 212 minerals, 7 gases, 0 surface species, 13 elements, 11 oxides. N-R converged in 602 its., resmax = 6.85e-012, Xi = 0.0000 Charge balance: Cl- molality adjusted from 2.426e-5 to .0009466 27 supersaturated phases, most = Muscovite Following reaction path. Swapping Al(OH)4- in for Al+++ Swapping CH4(aq) in for HCO3- Swapping H2S(aq) in for H+ Swapping Al(OH)2+ in for SO4-- Caution: molality of Cl- component forced negative by charge balance: -2.349e+008 *N-R didn't converge after 400 its., resmax = 1.6e+131, Xi = 0.0100 Cutting step size to find solution Swapping SO4-- in for Al(OH)4- Swapping HCO3- in for CH4(aq) Swapping Al+++ in for H2S(aq) Swapping H+ in for Al(OH)2+ Swapping H2S(aq) in for SO4-- Swapping CH4(aq) in for HCO3- Swapping Al(OH)4- in for Al+++ Residuals too large, 667-th interation -- Didn't wake up, abandoning path From: Craig Bethke Subject: Re: carbonate vs Eh There are several issues with this run: The endpoint Eh values you specify are far outside the stability limits of water, which makes it unlikely the program will converge. If you start at the reducing end, you need to set an appropriate basis (i.e., CH4(aq) instead of HCO3-, HS- instead of SO4--). It is probably easiest to start at the oxidized endpoint. I expect that you want pH to be held constant, rather than drift freely. If so, set a fixed pH reactant on the Reactant pane. Because your solution is very dilute, there aren't enough counterions to achieve charge balance over the entire range of possible Ehs. You will probably need to turn off charge balance to do this run.

From: Gregg Jones Subject: Dissolved Oxygen I am simulating the injection of potable water from a water treatment plant into an aquifer storage and recovery well. The storage zone is in the Suwannee Limestone in the Floridan Aquifer. Reducing conditions allow minute quantities of pyrite that occur in the limestone matrix to be stable in the presence of formation water. My goal is to determine the ratio of injection water to formation water that causes pyrite to become unstable. I am producing Eh/pH type diagrams to look at pyrite stability but instead of using Eh as my y axis, I am using log activity of the sulfate-sulfide ratio. I am assuming the sulfide concentration is virtually zero in the water coming out of the treatment plant. Prior to mixing, the sulfate and sulfide concentrations in the formation water are 1,526 mg/l and 14.5 mg/l respectively. When I run the simulation, I get the sulfate-sulfide ratio vs pH for each reaction step as I increase the ratio of injection water to formation water. I then plot this reaction path on the stability diagram and I get the ratio of injection water to formation water where pyrite becomes unstable. For my first attempt, I assumed that the treatment plant water is equilibrated with the atmosphere. This would give me a pH of approximately 5.6 and a dissolved oxygen concentration of approximately 7.5 mg/l. I revised the simulation when I received additional data on the chemical quality of the treatment plant water. The actual pH was 7.6 vs 5.6 and the actual DO was 15.4 mg/l vs 7.5 mg/l. When I ran the simulation using these values, I saw very little change in the reaction path and therefore, the ratio of injection water to formation water where pyrite became unstable changed very little. It seems to me that the higher DO concentration should cause pyrite to dissolve at lower injection water to formation water ratios. Why doesn't the model account for the change or am I missing something? I ran into a similar problem when I conducted additional runs where I systematically lowered the DO to see how low the DO would have to be in the injection water in order for pyrite to remain stable in the limestone. Even when the DO value in the injection water is as low as 0.25 mg/l (the level in the formation water), the reaction path looks the same as the path created when DO in the injection water was 15 mg/l. Also, the resulting species and their concentrations are not very different between the injection waters with 15.4 mg/l DO and 0.25 mg/l DO. Shouldn't the DO concentration have a big effect? I am using Release 3.2.1. I start by characterizing the injection water from the treatment plant. T = 25.8 TDS = 325 mg/l Ca++ = 88.9 mg/l Mg++ = 9.5 mg/l Na+ = 61 mg/l HCO3- = 81.9 mg/l S04-- = 114 mg/l Cl- = 28.7 mg/l Fe++ = 0.1 mg/l K+ = 1.9 mg/l PH = 7.6 DO = 15.4 mg/l Go Pickup reactants = fluid Reactants times 5 Now I characterize the formation water. T = 26.4 TDS = 2510 mg/l pH = 6.99 Ca++ = 399 mg/l Mg++ = 148.5 mg/l Na+ = 52.7 mg/l HCO3- = 129.5 mg/l SO4-- = 1526 mg/l Cl- = 97.2 mg/l Swap HS- for O2(aq) HS- = 14.94 mg/l Fe++ = 0.07 mg/l K+ = 5.2 mg/l Go From: Craig Bethke Subject: Re: Dissolved Oxygen I can't say I follow your logic, but I think there's an easier way to work this problem. If I were you, I'd run the mixing simulation in React and then plot the saturation state of pyrite against the mixing ratio (which is numerically equal to the kg of injected fluid added to the initial 1 kg of formation water). Your results should look something like this (see attached file).

From: Richard Laffers Subject: Pickup fluid problem I encountered a wierd problem with "pickup fluid" command, documented by the attached script: REACT calculates a hypothetical reaction path with sliding temperature. At the end of the React_output_1.txt (the last reaction step) I checked the bulk concentrations of basis species. There is a difference in number of total moles of Al+++ and number of moles of Al+++ in fluid, though the system contains no minerals (flow-through option). Why? Then the fluid is picked up. Here I noticed, that the amount of Al+++ is picked up in terms of free molality, in contrast to other elements (except for H+, which is OK). Why? If I now make equilibrium calculation, the resulting fluid should be the same as at the end of React_output_1.txt, but it is not. The difference is in Al+++ bulk concentration, ionic strength, and mineral saturations. Why "pickup fluid" did not pick up exactly the same fluid from the end of the reaction path? I noticed, that "pickup fluid" command works perfectly when flow-through option is off. React-GWB 4.0.3, Win XP, thermo.com.v8.r6+.dat data = "C:Program FilesGwbGtdatathermo.com.v8.r6+.dat" verify temperature initial = 300, final = 200 swap Quartz for SiO2(aq) 1 kg free H2O free gram Quartz = 5000 total g/kg Na+ = 40 balance on Cl- total g/kg Cl- = 50 total molality Al+++ = .001 pH = 5 dump flow-through printout surfaces = none gases = none elements = none suffix _1 go pickup fluid suffix _2 go

From: Richard Laffers Subject: Setting pH and Eh of solution reactant Dear users, using REACT, I troubled quite a time with this: On the reactants pane, I set a solution as reactant. I wondered, how could I specify the solution's pH, since the amount of any reactant can be constrained only in moles, grams or cm3 respectively. To find out number of moles of H+ needed for the desired solution's pH, I have done the following: I made an equilibrium calculation (using REACT) for the solution with the desired pH. In the React_output.txt I checked the bulk concentration of H+ (in the Original basis section). This value I used to constrain H+ on the reactants pane (along with setting the amount of H2O = 1 kg). Is there any more straightforward way to specify pH of solution reactant? Anyone experienced this problem? Similar problem arises when we try to specify the solution's Eh. From: James Cleverley Subject: RE: Setting pH and Eh of solution reactant The best place to get this answer is by looking at example 16.2 (p.237) in Geochemical Reaction Modelling (Craig Bethke, 1996), available through Rockware I believe. This example gives you all the information you need to help you understand how to set a fluid as a reactant in REACT. Also see 'picking up the results of a run' on page 145 of the GWB user's manual. From: Craig Bethke Subject: Re: Setting pH and Eh of solution reactant The easy way to do this is to set up an equilibrium model of the reactant fluid in the Basis pane, setting pH, Eh, etc., as usual. Run the model, then do a Run -> Pickup -> Reactants -> Fluid. Now, the fluid will be converted to a reactant. You can change the total amount of fluid to be reacted by setting a factor in the "reactants times" box on the Reactant pane. P.S. There are some specific examples of picking up fluids as reactants in the green book as well as the User's Guide. You might find these helpful. From: Richard Laffers Subject: Re: Setting pH and Eh of solution reactant Dear Craig, James, thank you for your advices. Picking up fluid as reactant would certainly works, nevertheless not in my case. I try to model mixing of 2 fluids. The composition of the first fluid is result of former reaction path modeling and it is picked up with "pickup system=fluid" command. That's why I cannot directly use the equilibrium calculation + "pickup reactants=fluid" to specify the other fluid - it would reset the initial system (specifying the first fluid), which cannot be set directly.

From: Gregg W. Jones Subject: Reaction Path Problems I am simulating the injection of potable water that is fresh out of a water treatment plant into an aquifer storage and recovery well. The storage zone is in the Suwannee Limestone in the Floridan Aquifer. I'm interested in the composition of the mixed fluid. I am assuming the treatment plant water is equilibrated with the atmosphere. I am interested in getting the sulfate/sulfide ratio at each reaction step. My intention is to plot the reaction path on a plot of the log activity of the sulfate/sulfide ratio against pH to determine at what mixture of injected fluid and formation water pyrite becomes unstable. I'm also trying to plot the reaction path between the two end members: the treatment plant water and the formation water. Pyrite is stable in the formation water but as I mix in larger volumes of treatment plant water, the H2S and HS- concentrations reported on the output quickly drop below 1X10-8 molal. The program no longer reports the concentrations of these species below that concentration. This is a problem in plotting the reaction path between the two end members because you don't get very far from the formation water on the reaction path before the H2S and HS- concentrations are no longer reported and your path comes to a dead end long before you get to the treatment plant water. I realize the concentrations of H2S and HS- get extremely small but is there any way to get the program to report them? Another problem is that as the volume of formation water gets very small with respect to the volume of the treatment plant water, the program starts to report larger and larger charge balance errors. Why is that happening and does it invalidate my results? Output is as follows. I'm using the latest release of the program. I start by characterizing the injection water from the treatment plant. Swap O2(g) for O2(aq) Swap CO2(g) for H+ f O2(g) = 0.2 log f CO2(g) = -3.5 T = 25.8 TDS = 325 mg/l Ca++ = 88.9 mg/l Mg++ = 9.5 mg/l Na+ = 61 mg/l HCO3- = 81.9 mg/l S04-- = 114 mg/l Cl- = 28.7 mg/l Fe++ = 0.1 mg/l K+ = 1.9 mg/l Go Pickup reactants = fluid Reactants times 5 (of course, this number gets larger and larger as I mix in more treatment plant water) Now I characterize the formation water. T = 26.4 TDS = 2510 mg/l pH = 6.99 Ca++ = 399 mg/l Mg++ = 148.5 mg/l Na+ = 52.7 mg/l HCO3- = 129.5 mg/l SO4-- = 1526 mg/l Decouple HS- HS- = 14.94 mg/l Cl- = 97.2 mg/l Swap HS- for O2(aq) HS- = 14.94 mg/l Fe++ = 0.07 mg/l K+ = 5.2 mg/l Go From: Craig Bethke Subject: Re: Reaction Path Problems To cause all species to be listed in React's printed output, regardless of concentration, do a Config -> Print Options, then for "Aqueous species" select "long". Run React again and you will see a complete list, not just those species with concentrations > 10^-8 molal. Regarding the charge balance error, although the absolute number of Faradays of charge imbalance increases, the imbalance per kg of fluid does not. So don't worry, be happy.

From: Gregg Jones Subject: Eh confusion I sent the message below to the users group on the 20th of January and received several helpful responses. The model ran just fine when I added a command to decouple HS-. Thanks to those who provided guidance. My problem now is with interpretation of the output. My intention was to plot the log activity of the sulfate/sulfide ratio against pH to determine at what mixture of injected fluid and formation water pyrite becomes unstable. I was able to do that but now I'd like to take the Eh and pH values provided in the output to determine pyrite stability. The problem is that for each mixing step, the output provides more than one Eh value. The first Eh value within the data for a given mixing step is always positive no matter what mixing ratio I use. That can't be right because I know pyrite is stable in my formation water so I should get negative Eh values when my mixture is mostly formation water. The other Eh value in a given mixing step is under the heading of Nernst Redox Couples. There are Eh values for 2 equations, one of the equations is: 8*e- + 9*H+ + SO4-- = 4*H2O + HS-. The Eh values for this equation are always negative, even when I have almost pure injection water which is highly oxygenated as you can see from my input. Shouldn't the Eh be positive for the injection water? So there's my dilemma. Any advice? Gregg I am trying to simulate the injection of potable water that is fresh out of a water treatment plant into an aquifer storage and recovery well. The storage zone is in the Suwannee Limestone in the Floridan Aquifer. I am interested in the composition of the mixed fluid. I am assuming the treatment plant water is equilibrated with the atmosphere. I am interested in getting the sulfate/sulfide ratio at each reaction step. I am assuming the sulfide concentration is virtually zero in the water out of the treatment plant. Prior to mixing, the sulfate and sulfide concentrations in the formation water are 1,526 mg/l and 14.5 mg/l respectively. I ran the model first without swapping in HS- and it ran fine. At the pH of the formation water, sulfide is present mostly as HS-. When I swapped HS- for O2(aq), the model didn't run and the message was initial solution is too supersaturated. Am I setting the problem up correctly and why does the model not run when I try to get sulfide into the basis? I'm using release 3. I start by characterizing the injection water from the treatment plant. Swap O2(g) for O2(aq) Swap CO2(g) for H+ f O2(g) = 0.2 log f CO2(g) = -3.5 T = 25.8 TDS = 325 mg/l Ca++ = 88.9 mg/l Mg++ = 9.5 mg/l Na+ = 61 mg/l HCO3- = 81.9 mg/l S04-- = 114 mg/l Cl- = 28.7 mg/l Fe++ = 0.1 mg/l K+ = 1.9 mg/l Go Pickup reactants = fluid Reactants times 5 Now I characterize the formation water. T = 26.4 TDS = 2510 mg/l pH = 6.99 Ca++ = 399 mg/l Mg++ = 148.5 mg/l Na+ = 52.7 mg/l HCO3- = 129.5 mg/l SO4-- = 1526 mg/l Decouple HS- HS- = 14.94 mg/l Cl- = 97.2 mg/l Swap HS- for O2(aq) HS- = 14.94 mg/l Fe++ = 0.07 mg/l K+ = 5.2 mg/l Go From: Craig Bethke Subject: Re: Eh confusion The first Eh value represents the redox potential of the couples that you left engaged: O2/H2O, Fe3+/Fe2+, HCO3/CH4, H+/H2, and so on. Since you set these couples in redox equilibrium, they all have the same redox potential. The Nernst Eh for the SO4--/HS- couple, which you have disengaged, reflects the sulfate/sulfide ratio you set for the fluid. You decoupled this reaction, so the redox potential will differ from the first value. This Nernst Eh will invariably be small because you maintain a certain amount of sulfide in the fluid. This sulfide cannot oxidize in your simulation because you have disengaged the redox couple. There is no oxidation reaction available to it. One way to allow the sulfide to oxidize would be to specify a kinetic redox reaction by which it reacts with an oxidant to form sulfate.

From: Katie Aguilar Subject: Leachate through a clay layer I have written the users group before for this same problem, but have moved on to a new avenue/ barrier to completion. As a refresher: I am modeling leachate through a clay layer. The contaminants of concern are metals, there are no organics in this problem. I am trying to model the natural attenuation in clay as the leachate runs through a set volume of clay with a CEC = 0.6 eq/g. I eventually want to include the ion exhange script and FeOH script to allow for ion exchange and adsorption. However, right now I am reconciling the basic model. I have brought my leachate to equilibrium, and have added the soil minerals. We are not sure what is out there exactly, but are assuming (reasonable, informed assumptions here) that kaolinite, quartz and goethite exist as soil minerals. Two problems: 1) When I run the system, it indicated a drop in sodium in the leachate, even when I add ion exchange. I know that the level of Na in my leachate should increase as the sodium ions in the soil are replaced with ions in the leachate. Does anyone have experience with this and have a suggestion? 2) I have quartz in my system. When I run the model, it has about 50 moles of quartz precipitating out. I can not suppress the species to prevent this from happening as it is in my basis. I know, though, that my time line does not allow for the generation of quartz in my system. Does anyone have experience at surpressing the precipitation of minerals that are in the basis? Attached is the script. I am using the thermodynamic database thermo_minteq_gwb4. I am running GWB (React) Release 4.0.2 on Windows 2000, and have attached the script. I am also using the default thermo dataset. Katie # React script, saved Fri Jan 16 2004 by kaguilar data = "C:Program FilesGwbGtdatathermo_minteq_gwb4.dat" verify temperature = 25 swap PbCrO4 for Pb++ swap Calcite for CO3-- swap Cerargyrite for Ag+ swap Barite for Ba++ swap Kaolinite for Al+++ swap Goethite for Fe+++ swap Quartz for H4SiO4 1.00000004 kg free H2O total mol AsO4--- = 8.07563367e-6 total mol SeO4-- = 2.52051463e-6 free mol PbCrO4 = 1.93505021e-6 total mol Hg(OH)2 = 2.49264669e-9 total mol H+ = .000811677887 free mol Calcite = 3.76854276e-5 total mol CrO4-- = 2.11436482e-5 total mol Mg++ = .00390753537 free mol Cerargyrite = 3.79105577e-6 free mol Barite = 1.77546743e-6 total mol O2(aq) = .00015625 total mol Cl- = .000814193318 balance on Na+ total mol Na+ = .0223492309 total mol K+ = .00749322285 total mol SO4-- = .0288026584 total mol Ca++ = .0107906579 total mol Cd++ = 1.08541505e-6 free kg Kaolinite = 2.61 free kg Goethite = .125 free kg Quartz = 3.69 react 2.61 kg of Kaolinite react .125 gram of Goethite react 3.69 gram of Quartz itmax0 = 5e4 From: Craig Bethke Subject: Re: Leachate through a clay layer Regarding your silica question, you need to determine whether or not your initial system is in equilibrium with quartz. If it is, but the mineral is unlikely to form to any significant extent over the reaction path, then on the Reactants pane set quartz as a kinetic mineral, using a small or zero-valued rate constant. If it is not, set the initial silica concentration by swapping a less stable mineral, such as tridymite, into the basis in place of aqueous silica. Or, constrain the basis using a measurement of silica concentration. Then you can suppress quartz.

From: Gregg W. Jones Subject: Injection of water into an aquifer I'm trying to simulate the injection of potable water that is fresh out of a water treatment plant into an aquifer storage and recovery well. The storage zone is in the Suwannee Limestone in the Floridan Aquifer. I am interested in the composition of the mixed fluid. I'm assuming the treatment plant water is equilibrated with the atmosphere. I am interested in getting the sulfate/sulfide ratio at each reaction step. I am assuming the sulfide concentration is virtually zero in the water out of the treatment plant. Prior to mixing, the sulfate and sulfide concentrations in the formation water are 1,526 mg/l and 14.5 mg/l respectively. I ran the model first without swapping in HS- and it ran fine. At the pH of the formation water, sulfide is present mostly as HS-. When I swapped HS- for O2(aq), the model didn't run and the message was "initial solution is too supersaturated". Am I setting the problem up correctly and why does the model not run when I try to get sulfide into the basis? I'm using release 3.2.1 I start by characterizing the injection water from the treatment plant. Swap O2(g) for O2(aq) Swap CO2(g) for H+ f O2(g) = 0.2 log f CO2(g) = -3.5 T = 25.8 TDS = 325 mg/l Ca++ = 88.9 mg/l Mg++ = 9.5 mg/l Na+ = 61 mg/l HCO3- = 81.9 mg/l S04-- = 114 mg/l Cl- = 28.7 mg/l Fe++ = 0.1 mg/l K+ = 1.9 mg/l Go Pickup reactants = fluid Reactants times 5 Now I characterize the formation water. T = 26.4 TDS = 2510 mg/l pH = 6.99 Ca++ = 399 mg/l Mg++ = 148.5 mg/l Na+ = 52.7 mg/l HCO3- = 129.5 mg/l SO4-- = 1526 mg/l Cl- = 97.2 mg/l Swap HS- for O2(aq) HS- = 14.94 mg/l Fe++ = 0.07 mg/l K+ = 5.2 mg/l Go From: Craig Bethke Subject: Re: Injection of water into an aquifer Your fluid is very supersaturated with respect to pyrite, which is causing your problems. Since in reality pyrite isn't likely to form in this system, I suggest you simply suppress it. Also, I think that instead of swapping sulfide for dioxygen, you want to decouple sulfide. This way, you can create a model in redox disequilibrium.

From: Alexander Neaman Subject: how to obtain mineral saturated states with React I am wondering how to obtain the mineral saturated states and the list of all supersaturated phases in the output of React. The option "mineral saturation states" (in "configurations - print options") is set on "long", but the program still does not give the mineral saturation states.I am using React 4.0.3, Windows XP Professional, and the database thermo_minteq_gwb4. From: Alexander Neaman Subject: Re: how to obtain mineral saturated states with React I apologize for multiple e-mails. Attached please find an input file. I was meaning "mineral saturation indexes" (not "mineral saturation states"). # React script, saved Mon Jan 26 2004 by Administrator data = "C:Program FilesGwbGtdatathermo_minteq_gwb4.dat" verify temperature = 30 swap "O2 (g)" for O2(aq) 1 kg free H2O fugacity "O2 (g)" = .2 total mmolar Na+ = 17.98 total mmolar NO3- = 17.6 total mmolar Mg++ = .62 total mmolar SO4-- = .62 balance on Cl- total mmolar Cl- = .48 total mmolar Citrate--- = .13 total mmolar CO3-- = .19 pH = 9 total mmolar Fe+++ = .05 total mmolar Ca++ = .455 total umolar PO4--- = .46 printout alphabetical species = long minerals = long basis = long reactions = long plot = character