C-----Version: 11.02.2015 Changes in THERIAK-DOMINO version 11.02.2015 -------------------------------------------- Solution models White et al. 2014 included in database. Warning: Equation of state for melts is not working properly. Is OK up to medium pressures. {??) Changes in THERIAK-DOMINO version 03.01.2012 -------------------------------------------- Coefficients for the asymmetric van Laar model, according to Holland and Powell (2002): If a0 is negative, the asymmetry size parameter used is the relative volume of the end-members at P & T. (For fluids) Equation of state according to Holland an Powell (2011): new volume function (Holland and Powell 2011) V11 Bragg-Williams (Holland and Powell 2011) BW1 Landau (Holland and Powell 2011) LA1 dk0dt for liquids (Holland and Powell 2011) DK1 EOS for water: K.S. Pitzer and S.M. Sterner, 1994 PS94H2O EOS for CO2: K.S. Pitzer and S.M. Sterner, 1994 PS94CO2 In some cases where an explicit Volume is not required to calculate ∆G, the volumes were previously not calculated. Now all volumes are calculated explicitly or, in some cases with (dG/dP). The number of endmembers in the external subroutine is no longer restricted to 10. (is now defined as 15, but can "easily" be changed. The phases used for a calculation can be defined in the input line for the database. Example: Enter [ "?" | CR | "files" | database filename ] ? JUN92.bs Albite Jadeite a-Quartz b-Quartz Of all phases possible for the given bulk composition, this will use only Albite, Jadeite, a-Quartz and b-Quartz. (useful to calculate e.g. a single reaction. Note the two blanks between items) Provision to use initial guesses for the compositions of stable solutions. Still experimental. Filneame=inimini example: SOLUTION ClAMP tremolite tschermakite2 pargasite2 glaucophane2 cummingtonite2 grunerite2 acam bcam mgriebekite 7.76376D-01 3.88098D-02 7.53481D-02 2.17005D-02 5.63273D-03 1.00090D+00 -5.50374D-01 -3.68393D-01 0.000000D0 4.21530D-02 3.04170D-02 4.10399D-03 7.39931D-03 4.11153D-02 3.63996D-01 1.87278D-01 3.23537D-01 0.000000D0 In domino: smaller step along grid (ptdist) may be defined (not recommended) Changes in THERIAK-DOMINO version 010809 ---------------------------------------- Simplified input for isolines Addition of "REF" in drv-files. New program plotxy for tab-files. Excluding phases and monitoring their chemical potential with EXCL and $ Definition of thermal gradient in input. Models including site-margules, new section in database. Changes in THERIAK-DOMINO version 150508 ---------------------------------------- possibility to distinguish solution phases by the dominant endmember. (key "-1" in labeling of reactions) minor changes, e.g. "elements in stable phases" in output of Theriak. Distribution contins Theriag (see special documentation). Changes in THERIAK-DOMINO version 111207 ---------------------------------------- Made compilable with fortran90. (gfortran): Comments with "!", continuation lines with "&" at end of line. Changed loops with real DO indices to "DO WHILE" loops. Changes in THERIAK-DOMINO version 200307 ---------------------------------------- Several changes were made in the algorithm, to improve robustness in complex solutions. As a result, the calculations become considerably slower than with previous versions. I consider robustness far more important than speed, and because computers are becoming faster each year, this slowing is of minor concern. Solutions may include an asummetry van Laar parameter, according to holland and Powell(2002). Solutions that may have negative concentrations of their components must now be marked with a "-" in the solution definition line. Default is: all endmembers are >= 0. Lists or phases etc. are now in alphabetical order. In the section "activity of all phases", not all unstable solution phases are printed. The "pix" function in domino has now be implemented. This allows to calculate phase diagrams on a fixed grid (e.g. 100x100) and then to plot graymap images of any desired variable. The program to produce the grey-scale images, "makemap" is included. Changes in THERIAK-DOMINO version ....06 ---------------------------------------- Significant contributions to increase user friendliness and code portability were made by K. Petrakakis. The most important changes are: - Help function. (Interactive input by the user includes a '?', which shows detailed explanations for each option. The text of this help-function is stored in the file thhelp.txt) - Flexible file-assignments. (most file names for input and output are defined in the file theriak.ini. Users may change these to their taste) - A few platform-dependent subroutines (e.g. the setting of environmental variables) are provided in the files platfxxx.f. The rest of the programs and subroutines are completely platform independent. This structure increased portability of the programs to the most common operating systems allowing for preparation of binaries for download. Further changes in code include: - Input numbers (e.g. in the chemical formulas or COM-line) may include fractions (x/y). e.g. the definition of ordered biotite may be written as: COM phlogopite[2/3]annite[1/3] - The Database now includes the stoichiometric oxygen numbers based on the commonly used standard oxides. This allows to define the total amount of O (Oxygen) in bulk composition definition (in dat-file) as O(?). In this case the standard oxidation state as defined by the stoichiometric oxygen amount defined in the database is used to calculate the total O, e.g. AL(2)SI(1)O(?) is equivalent to: AL(2)SI(1)O(5). - The calculation parameters (like TEST or LO1MAX) are now defined in the file theriak.ini. Earlier they were defined in the dat-file (e.g. THERIN). - The parameter PGAS (ratio (solid pressure)/(liquid pressure) can be defined in the dat-file as optional input following P. This was already implemented in earlier versions. - Definition of solutions: The site multiplicity m of a solution is now defined as a number directly following the right bracket of the model-keywords. The default is m = 1, but may also be a fraction like 1/3 etc. Example of current garnet model definition: GARNET (MARGULES,IDEAL)3 M(3):Ca,Mg,Fe,Mn. Now the keyword IDEAL means strictly that activity(i) = molefraction(i)**m. (xi = concentration of endmember i, m = multiplicity as defined above) Note: this is a major change. In older versions ideal solutions with a multiplicity other than one had to be defined as "SITE" or "IDEAL" with one site defined, for eample as SOL-NAME (IDEAL) M1(3):Fe,Mg These definitions must now be changed to: SOL-NAME (IDEAL)3 M1(3):Fe,Mg. The site definition in "IDEAL" solutions is ignored for the calculation. If present, it will be used to print site occupancies in the output. Changes in THERIAK-DOMINO version 140205 ---------------------------------------- The minimization routine has been adapted to allow for some endmembers of certain solutions to have negative concentrations. Details: If the activity of an endmember is not zero although its concentration is zero, this endmember is assumed to describe ordering and thus may have negative concentrations. This definition allows the program to handle solution models with "proportions" (as defined by R. Powell and T. Holland for THERMOCALC) as normal endmember solutions. Warning: This feature is not extensively tested. Users are advised to think. Changes in THERIAK-DOMINO version 200704 ---------------------------------------- Increased robustness of code. Added a Newton-Raphson to the gradiant search. Check rounding errors in finding minimum assemblage. Result of these changes is a substancially increased computing time. (ca. 2x). Considering the advantages of the increased robustness and the increasing speed of newer computers this is a minor concern.