720 DO NOT DELETE THIS LINE --------------------------------------------------------------------------- FOR MODEL DEFINITIONS AND REFERENCES READ THE MODEL COMMENTARY OR REFER TO THE GLOSSARY AT: www.perplex.ethz.ch/perplex_documentation.html#SOLUTION_MODEL_GLOSSARY --------------------------------------------------------------------------- DO NOT USE TABS IN PERPLE_X DATA FILES, TAB CHARACTERS ARE NOT INTERPRETED AS BLANK SPACES AND CAUSE FORMATTING ERRORS. --------------------------------------------------------------------------- Character data is format free. --------------------------------------------------------------------------- Comments can be placed between models provided, nothing is written in the first 10 columns. Comments may be placed after data if it is separated from the data by a '|' marker. --------------------------------------------------------------------------- VARIOUS DOCUMENTATION SECTIONS BELOW: I) Solution model types II) Subdivision schemes III) Species limits for order-disorder models IV) Templates for solution model types 2 and 39 --------------------------------------------------------------------------- I) Solution model types are: 2 - simplicial composition space. 6 - order-disorder, simplicial composition space. 7 - polytope composition space (reciprocal when correct). 8 - order-disorder, polytope composition space. 688 - generic format, i.e., encompasses all of the above. Special model types: 0 - internal (fluid) EoS 20 - Electrolyte (charge balance) model 26 - H2O-CO2-NaCl, Aranovich et al. (2010) 29 - BCC Fe-Si alloy, Lacaze & Sundman 1990 30-33 - FeSiC alloys (BCC/FCC/CBCC/HCP), Lacaze & Sundman 1990 39 - GFSM, generic (hybrid) fluid solution model. 40 - Silicate fluid (MRK) 41 - COH fluid (hybrid MRK) 42 - FeS liquid, Saxena & Eriksson 2015 with ECRG corrections --------------------------------------------------------------------------- II) SUBDIVISION SCHEMES define the discretization and range of compositions for a solution during a calculation. In many cases, particularly for chemically complex solutions, this range is restricted to reduce the time and memory required for calculations. If such a restriction is encountered (**warning ver991**) during the exploratory stage of a calculation the restriction may have consequences for the final auto-refine stage result. For this reason it is important for users to correct the scheme (unless the restriction is intentional). A resume of warnings written at the end of the exploratory stage is written to the *_auto_refine.txt file. This file should be examined before any results are accepted as final. There are only two subdivision schemes implemented in the current version of Perple_X (6.8.1+): Cartesian subdivision and non-linear subdivision. Cartesian is more robust and is therefore specified for most solution models by default. Non-linear subdivision allows for high resolution at low concentrations of a species. It is recommended for any compositional variable whose maximum anticipated value is less than the value of the exploratory stage initial_resolution (default = 1/16) and it is mandatory for any compositional variable whose maximum anticipated value is less than the value of the auto-refine stage final_resolution (default = 1e-3). For example, the mole fraction of Zr in granitic melts using the melt(HP) model is ~1e-5. If this melt model is run with the default option values, Perple_X will indicate that Zr is insoluble (i.e., below the limit of resolution). A brief description of the Cartesian and non-linear subdivision schemes follows (it reads worse than it is): If the solution model has a simplicial composition space, then each scheme is directly associated with an endmember (as identified in *_auto_refine.txt). For a solution with N endmembers, there are thus N-1 subdivision schemes. In general it is best to order the solution model endmembers so that the most important end- member is listed last and therefore determined by difference. If the solution model has a prismatic composition space (in Perple_X a polyhedral composition with more vertices than a simplex, e.g., a quadrilateral, is referred to as a prism), then a subdivision scheme is associated with the S-1 independent compositions of the T simplexes that comprise the prism (as identified in *_auto_refine.txt). Thus for prismatic composition spaces there are T*(S-1) subdivision schemes for the compositional variables X(1..T,1..S-1). X(1..T,S) is determined by difference and therefore then endmembers should be ordered so that X(1..T,S) corresponds to the most important compositional variable. The subdivision scheme for each independent compositional variable (X) is specified by four parameters that are entered as XMIN, XMAX, XINC, IMOD. The significance of these parameters is dicussed below. NOTE 1: In solution model type 688 format, if the composition space of the model involves at least one prism, the subdivision scheme parameters for every independent composition include a name for the compositional variable, i.e., NAME, XMIN, XMAX, XINC, IMOD, where NAME is an alphanumeric string used to identify the variable in output. NOTE 2: For purposes of this discussion XMIN, XMAX, and XINC are described as static variables as is indeed the case for calculations with CONVEX; however, during adaptive optimization calculations in VERTEX and MEEMUM these variables are rescaled to obtain higher resolution, see www.perplex.ethz.ch/perplex_options_body.html#Iteration for information on this scaling. IMOD - may be either 0 or 1: IMOD = 0 - is a cartesian scheme, in which the compositional variable is discretized with regular spacing. IMOD = 1 - is a conformal, non-linear, transformation scheme, in which the discretization of the compositional variable becomes finer toward zero. Use of this scheme is recommended if the maximum anticipated value of the compositional variable is < the initial_resolution option value. It is mandatory if the maximum anticipated value of the compositional variable is < the final_resolution option value. It should be used with caution if the maximum anticipated value of the compositional variable is >> initial_resolution. And it should NOT be used if the maximum anticipated value of the compositional variable is > 1/2. XMAX - is the maximum value of the compositional variable permitted by the scheme. This value may be relaxed if the hard_limits option is FALSE (default). XMIN - the exact significance of XMIN is dependent on the subdivision scheme: IMOD = 0 - XMIN is the minumum value of the compositional variable permitted by the scheme. This value may be relaxed if the hard_limits option is FALSE (default). IMOD = 1 and XMIN > 0 - XMIN is the smallest non-zero value of the compositon for statically generated compositions. Thus, the value of XMIN dictates the asymmetry of the discretization. XMIN should be of the order of magnitude the smallest anticipated value of X, values of XMIN more than an order of magnitude below this value are wasteful. During adaptive optimization (VERTEX/MEEMUM), XMIN does not limit X. During convexhull optimization (CONVEX), X < XMIN is allowed if the hard_limits option is FALSE (default). IMOD = 1 and XMIN = 0 - the asymmetry of the discretization is specified by the stretch_factor option. For IMOD = 1, compositions are always generated at X = 0, X = XMIN, and X = XMAX. XINC - controls the discretization (resolution) of the compositional variable its exact significance is dependent on the subdivision scheme or values of the initial_resolution option initial_resolution = [0 0] - XINC is read from the relevant solution model, regardless of the value of IMOD. IMOD = 0 - linear scheme - the number of compositions generated over the interval [XMIN XMAX] is 1/XINC + 1. XINC is equated to the appropriate (exploratory or auto-refine stage) value of the initial_resolution keyword, which defaults to [1/16 1/48]. IMOD = 1 - nonlinear scheme - 1/XINC + 1 is the number of compositions that would be generated over the interval [0 1]. The actual number of compositions generated depends on the asymetry of the scheme and the value of XMAX. Compositions are always generated at X = 0 and X = XMIN, additional compositions are generated by incrementing the conformal variable by XINC until the X > XMAX, this final value of X is then reset to XMAX. This algorithm has the consequence that the number of compositions generated is always 1/XINC + 1 if XMAX = 1; for values XMAX < 1, small values of XINC will cause more compositions to be generated over the interval [0 XMAX], the tables below illustrate the dependence of the number of compositions on XINC and XMIN. Because Perple_X adaptively refines compositions to a resolution comparable to the final_resolution option value, it is generally unnecessary to generate a large number of compositions at low X values as occurs if XINC is small. For adaptive optimization (programs MEEMUM and VERTEX) the default behavior is that XINC is read from the text of the relevant solution model. For convexhull optimization (program CONVEX) the default behavior is that XINC is equated to the appropriate (exploratory or auto-refine stage) value of the initial_resolution keyword, provided this value is non-zero. These defaults toggle when the non_linear_switch option is set to T (true). TIPS ON THE USE OF NON-LINEAR SUBDIVISION 1) If XMAX = 1, then the number of compositions generated by non-linear subdivision is dictated entirely by XINC. XMIN controls the asymmetry of the subdivision such that the compositions cluster more closely to XMIN as XMIN becomes smaller. Setting XMAX < 1 chops off the poorly resolved region of the composition space; by definition, few compositions are generated in this region, therefore reducing XMAX does not save much memory, this is illustrated in the 3rd table below by the difference in the number of compositions generated for XMAX = 0.1 and XMAX = 0.2. XMAX should be > the maximum anticipated value of the composition. If this is not the case Perple_X will write a warning to the my_project_auto_refine.txt file. Non-linear subdivision should NOT be used with XMAX > 1/2. 2) XMIN should be the same order of magnitude as the minimum anticipated value of the composition. If XMIN is much lower than this composition, compositional resolution will deteriorate more rapidly than is necessary toward XMAX. The minimum actual value of all compositional values is written to the my_project_auto_refine.txt file. 3) XMIN should be at least one order of magnitude < the exploratory-stage value of the initial_resolution option (typically 1/16 to 1/8), i.e., XMIN should be <= 1e-2 for most applications. 4) As a rule of thumb XINC should be chosen for adaptive optimization (MEEMUM/VERTEX) so that the number of compositions generated between XMIN and XMAX is equal to the number of decades of resolution between XMIN and XMAX. For example, if XMIN = 1e-4 and XMAX = 1e-1, the scheme must cover log(1e-1/1e-4) = 3 decades of resolution. The exact value of XINC necessary to achieve this is dependent on both XMAX and XMIN, but the results in the 3rd table below suggest that XINC = 0.2 should be adequate for most purposes. XINC = 1/48 = 0.002083... (default auto-refine stage initial_resolution) XMAX 0.1 0.2 XMIN ---------------------- 1d-2 10 17 1d-3 24 31 1d-4 32 37 1d-5 36 40 XINC = 1/16 = 0.0625 (default exploratory stage initial_resolution) XMAX 0.1 0.2 XMIN ---------------------- 1d-2 7 9 1d-3 11 12 1d-4 13 14 1d-5 14 15 XINC = 0.1 or 0.2 XMAX 0.1 0.2 XMIN ---------------------- 1d-2 4 5 1d-3 5 5 1d-4 5 6 1d-5 6 6 --------------------------------------------------------------------------- III) Species limits for order-disorder models, refer to: www.perplex.ethz.ch/perplex/faq/species_limit_expressions_for_type_6_and_8_solution_models.txt --------------------------------------------------------------------------- IV) Templates for solution model types 2 and 39, refer to: www.perplex.ethz.ch/perplex/faq/solution_model_type_2_(simple)_template.dat www.perplex.ethz.ch/perplex/faq/solution_model_type_39_(GFSM)_template.dat For additional information on generic fluid solution models (GFSM) refer to perplex.ethz.ch/Perple_X_generic_hyrbid_fluid_EoS.html --------------------------------------------------------------------------- begin_model | Ti-Fe-Mg-OH-O Chlinohumite, with ad-hoc Ti and Fe, assumes, somewhat improbably, that | Fe and Mg are disordered across MO and MB and that Ti is disordered on MB. this model | is an implicit o/d model because the ftchum endmember is specified as an independent | endmember. JADC, 5/20. Site: MB MO OH ____________________________ Mutliplicity 1 8 2 endmember ____________________________ type _________ _________ chum Mg Mg OH independent tchum TiMg Mg OOH independent fchum Fe Fe OH independent ftchum TiFe Fe OOH independent | chum is in holland & powell data files, as is an arbitrary make definition for | fchum (after fruh-green et al 94). this model requires make definitions (or real data) | for fchum, tchum, and ftchum, suggested make definitions are: | tchum = 1 chum - 1/2 br + 1/2 ru | DQF = -R*ln(2)*T | ftchum = 1 chum - 1/2 br + 1/2 ru - 1/6 ta + 1/6 fta - 4 fo + 4 fa | DQF = -R*ln(2)*T TiCh(PL) abbreviation TiCh full_name clinohumite 688 | model type: 688 format standard model 1 | number of polytopes 2 | number of simplices 2 2 | number of vertices on each simplex chum tchum fchum_q ftchum X_Ti 0. 1 .1 0 | imod = 0 -> cartesian subdivision X_M by difference X_Fe 0. 1 .1 0 | imod = 0 -> cartesian subdivision X_Mg by difference ideal 3 | number of sites in configurational entropy model MO | site name 2 8 8 | number of species, effective multiplicity, true multiplicity z(Mg,M) = 1 chum + 1 tchum z(Fe,M) = 1 fchum_q + 1 ftchum MB | site name 3 1 1 | number of species, effective multiplicity, true multiplicity z(Ti,M) = 1/2 tchum + 1/2 ftchum z(Mg,M) = 1 chum + 1/2 tchum z(Fe,M) = 1 fchum_q + 1/2 ftchum OH | site name 2 2 2 | number of species, effective multiplicity, true multiplicity z(OH,OH) = 1 chum + 1/2 tchum + 1 fchum_q + 1/2 ftchum z(O,OH) = 1/2 tchum + 1/2 ftchum [Si4O16] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(fchum_q) = 0 end_dqf_corrections end_of_model --------------------------------------------------------------------------- begin_model | Ti-Fe-Mg-F Chondrodite, with non-ideal F-OH interaction after Duffy & Greenwood, Am Min 64:1156-1174, 1979. | Ad-hoc Ti and Fe, assumes, somewhat improbably, that Fe and Mg are disordered across | MO and MB and that Ti is disordered on MB. this model is an implicit o/d model because | the tichdr endmember is specified as an independent endmember. JADC, 5/20. Site: MB MO OH ____________________________ Mutliplicity 1 8 2 endmember ____________________________ type _________ _________ chdr Mg Mg OH independent fchdr Mg Mg F independent tchdr Mg TiMg OOH independent ftchdr Mg TiMg OF dependent ichdr Fe Fe OH independent fichdr Fe Fe F dependent tichdr Fe TiFe OOH independent ftichdr Fe TiFe OF dependent This model requires data (or a make definition) for tichdr, my first guess would be to use tchdr + 4*(fa-fo) + (fta-ta). TiChdr(B) abbreviation TiCh full_name clinohumite 688 | model type: 688 format standard model 1 | number of polytopes 3 | number of simplices 2 2 2 | number of vertices on each simplex chdr fchdr tchdr ftchdr ichdr fichdr tichdr ftichdr X_F 0. 1 .1 0 | X(F), imod = 0 -> cartesian subdivision X_OH by difference X_Ti 0. 1 .1 0 | X(Ti), imod = 0 -> cartesian subdivision X_M by difference X_Fe 0. 1 .1 0 | X(Fe), imod = 0 -> cartesian subdivision X_Mg by difference begin_dependent_endmembers ftchdr = 1 tchdr + 1/2 fchdr - 1/2 chdr fichdr = 1 ichdr + 1 fchdr - 1 chdr ftichdr = 1 tichdr + 1/2 fchdr - 1/2 chdr end_dependent_endmembers begin_excess_function W(ichdr ichdr fchdr) -21572 W(chdr tichdr fchdr) -21572 W(tichdr tichdr fchdr) -5393 W(ichdr fchdr fchdr) -59120 W(fchdr fchdr tichdr) -29560 W(fchdr chdr ichdr) -43144.2 W(fchdr ichdr tchdr) -21572.1 W(fchdr ichdr tichdr) -21572.1 W(fchdr tchdr tichdr) -10786.05 W(chdr chdr fchdr) -21572 W(chdr tchdr fchdr) -21572 W(tchdr tchdr fchdr) -5393 W(chdr fchdr fchdr) -59120 W(fchdr fchdr tchdr) -29560 end_excess_function 3 | number of sites in configurational entropy model MO | site name 2 8 8 | number of species, effective multiplicity, true multiplicity z(Mg,M) = 1 fchdr + 1 chdr + 1 tchdr z(Fe,M) = 1 ichdr + 1 tichdr MB | site name 3 1 1 | number of species, effective multiplicity, true multiplicity z(Ti,M) = 1/2 tchdr + 1/2 tichdr z(Mg,M) = 1 fchdr + 1 chdr + 1/2 tchdr z(Fe,M) = 1 ichdr + 1/2 tichdr OH | site name 3 2 2 | number of species, effective multiplicity, true multiplicity z(OH,OH) = 1 chdr + 1/2 tchdr + 1 ichdr + 1/2 tichdr z(F,OH) = 1 fchdr z(O,OH) = 1/2 tchdr + 1/2 tichdr [Si4O16] | formula suffix, enter "none" for no suffix. end_of_model --------------------------------------------------------------------------- begin_model ternary feldsar (fuhrman & lindsley, am min, 1988) for binary plagioclse this model is identical to that of Newton et al. 1980, and for binary alkali feldspar it is identical to Haselton et al. (1983). CORRECTED FOR TYPO IN ORIGINAL PAPER. feldspar abbreviation Fsp full_name ternary-feldspar 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 3 | number of vertices on each simplex abh | endmembers on the vertices an san 0. 1 .1 0 | range and resolution for albite, imod = 0 -> cartesian subdivision 0. 1 .1 0 | range and resolution for anorthite, imod = 0 -> cartesian subdivision begin_excess_function w(abh abh san) 27320 -10.3 .394 * P_bar w(abh san san) 18810 -10.3 .394 * P_bar w(an an san) 52468 w(an san san) 47396 -.12 * P_bar w(an an abh) 28226 w(an abh abh) 8471 w(an abh san) 100045.5 -10.3 -0.76 * P_bar end_excess_function 2 | number of sites in configurational entropy model M | site name 3 1 1 | number of species, effective multiplicity, true multiplicity z(Na,M) = 1 abh z(Ca,M) = 1 an z(K,M) = 1 san T1 | site name 2 2 2 | number of species, effective multiplicity, true multiplicity z(Al,T1) = 1/2 + 1/2 an z(Si,T1) = 1/2 - 1/2 an [Si2O8] | formula suffix, enter "none" for no suffix. end_of_model --------------------------------------------------------- begin_model cAmph(G)_I => Green et al. (JMG, 2016) clinoamphibole. --------------------------------------------------------- Reformulated as a 4-subcomposition 688 format model. JADC, 3/20 The composition space of the complete (published) model can be represented as a mixture of three independent subcompositions tschermakitic amphibole: {Mg,Fe}*[Al,Fe3,Ti]M2*[M,C,N]M4 (18 endmembers) pargasitic amphibole: {Mg,Fe}*[K,N]A*[MAl,MFe3,MTi]M2*[M,C]M4 (36 endmembers) divalent amphibole: {Mg,Fe}*[M,C]M4 (4 endmembers) where T1 = f(N,M4), OH = f(Ti,M2) --------------------------------------------------------- NOTE: the above compositions can be combined to form the dependent subcomposition piglitic amphibole: {Mg,Fe}*[K,N]A*[Al,Ti,Fe3]M2*[N,CN,MN]M4 (36 endmembers) Because Perple_X does not currently allow dependent subcompositons, piglitic compositions can only be reached if the piglitic subcomposition is treated as an independent composition. This practice is extremely costly and for this reason the piglitic subcomposition has been commented out in the model text below. To restore the full model remove the comments and correct the subcomposition counter accordingly. The consequence of omitting piglitic compositions is that the site fraction of alkalis on the A-site (z[N+K,A]) is constrained to be less than one minus the site fraction of the alkalis on the M4 site (z[N,M4]), whereas in the full model z[N+K,A] may be as large as 1 - 1/2*z[N,M4]. If the piglitic subcomposition is restored, then tt is probable that VERTEX/MEEMUM will run out of dynamic memory. Should this occur, uncomment the low_reach option keyword at the end of this model. The low_reach option dramatically reduces the memory and time required for calculations, but may also causes a significant deterioration in phase diagram quality. --------------------------------------------------------- NOTE: to use this the following endmembers must be specified with make definitions in the thermodynamic data file mrbG = 1 gl - 1 gr + 1 andr DQF = 0 kprg = 1 mu - 1 pa + 1 parg DQF = -7060 + 20 * T_K tts = -2 dsp + 2 ru + 1 ts DQF = 95000 to avoid cluttering thermodynamic data files with the new plague of TC DQF "corrections", the following simple DQF's are specified in the make_definition section at the end of this model: DQF(gl) = -3000 DQF(parg) = -10000 DQF(ts_q) = 10000 DQF(grun) = -3000 For (nonsensical, but common) positive DFQ's this approach may lead to interference between the phase relations of the solution model and the un-DQF'd endmember. If such interference occurs: the DQF'd endmember must be renamed; its definition placed in the thermodynamic data file; and the un-DQF'd endmember excluded from the calculation. --------------------------------------------------------- independent endmembers and site populations: Sites A M1 M2 M4 T1 OH _______________________________________ Multiplicity 1 3 2 2 1(4) 2 _______________________________________ Ts V Mg Al Ca SiAl OH Parg Na Mg MgAl Ca SiAl OH Kprg K Mg MgAl Ca SiAl OH Tts V Mg Ti Ca SiAl O Gl V Mg Al Na Si OH mrbG V Mg Fe3 Na Si OH Tr V Mg Mg Ca Si OH Cumm V Mg Mg Mg Si OH Grun V Fe Fe Fe Si OH a V Mg Fe Fe Si OH b V Fe Mg Fe Si OH cross-site exchanges 2 NaCa-1(M4) + 2 SiAl-1(T1) = (gl-ts) 2 TiAl-1(M2) + 2 H-1(OH) = (Tts-Ts) NaV-1(A) + 1 AlMg-1(M2) = (parg-ts) = (parg-tr) AlMg-1(M2) + 1 NaCa-1(M4) = (gl-cumm)/2 intra-site exchanges NaK-1(A) = (parg-kparg) 2 CaMg-1(M4) = (Tr-Cumm) 2 Fe3Al-1(M2) = (Rb-Gl) 3 FeMg-1(M1) = (grun - a) 2 FeMg-1(M2) = (grun - b) 2 FeMg-1(M4) = (-grun + cumm + a + b) --------------------------------------------------------- cAmph_I(G) abbreviation Amph full_name clinoamphibole 688 | model type: 688 format standard model 3 | number of subcompositions | subcomposition names and composite composition space subdivision schemes Ts-Amph 0 .5 .25 0 | X([ ][M][T][M,C,N]) Pg-Amph 0 .5 .25 0 | X([A][M][MT][M,C,N]) |Pgl-Amph 0 .1 .1 0 | X([A][M][T][MN,CN,N]) Trem-Amph by difference | X([ ][M][M][MC]) | ------------------------------------------- | subcomposition 1 [Mg,Fe]*[ ]*[M]*[Ti,Fe3,Al]*[M,C,N] 3 | number of simplices 2 3 3 | number of vertices on each simplex | Simplex 1 {Mg,Fe} pseudo-site for the conditional cartesian product, defines bulk Mg/Fe | Simplex 2 T = Ti, Fe3, Al; z(T,M2) | Simplex 3 exchange vectors as below: | minor (0.09 bl478): vmtn_d vftn_d | [ ][M][Ti][N] Ti-glaucophane * mrbG frb_d | [ ][M][Fe][N] Fe3-glaucophane, aka riebeckite gl fgl_d | [ ][M][Al][N] Al-glaucophane mtts_d ftts_d | [ ][M][Ti][M] Ti-Ts mgrk_d fgrk_d | [ ][M][Fe][M] Fe3-Ts mts_d fts_d | [ ][M][Al][M] Al-Ts tts fctts_d | [ ][M][Ti][C] M-Ca-Ti_tschermaks mcgrk_d fcgrk_d | [ ][M][Fe][C] M-Ca-Fe3+_tschermaks (Na-free riebeckite) ts_q fcts_d | [ ][M][Al][C] Al-cTs | 1st simplex x_Mg,1 0 1 .25 0 | bulk Mg/M x_Fe,1 by difference | 2nd simplex z_Ti,1 0 .5 .25 0 | T on M2 z_Fe,1 0 1 .25 0 | z_Al,1 by difference | 3rd simplex z_Na,1 0 .5 .25 0 | M,C,N on M4 z_M,1 0 .5 .25 0 | z_Ca,1 by difference | ------------------------------------------- | subcomposition 2 [Mg,Fe]*[A]*[M]*[MT]*[M,C] 4 3 | number of simplices 2 3 2 2 | number of vertices on each simplex | Simplex 1 {Mg,Fe} pseudo-site that defines bulk Mg/Fe | Simplex 2 {Ti, Fe3, Al} on M2 | Simplex 3 {K, Na} on A | Simplex 4 {M,C} on M4 tmkpg_d tfkpg_d | [K][M][MTi][M] K-Ti-M-Pg omkpg_d ofkpg_d | [K][M][MFe][M] K-Fe3+-M parg kmparg_d kfparg_d | [K][M][MAl][M] K-Al-M parg tmpg_d tfpg_d | [N][M][MTi][M] N-Ti-M-Pg ompg_d ofpg_d | [N][M][MFe][M] Na-Fe3+-M parg mparg_d fparg_d | [N][M][MAl][M] Na-Al-M parg | minor (0.06 bl478): tcmkpg_d tcfkpg_d | [K][M][MTi][C] K-Ti-Ca parg ocmkpg_d ocfkpg_d | [K][M][MFe][C] K-Fe3-Ca parg kprg kfcprg_d | [K][M][MAl][C] K-Al-Ca parg tcmpg_d tcfpg_d | [N][M][MTi][C] N-Ti-Ca parg ocmpg_d ocfpg_d | [N][M][MFe][C] Na-Fe3-Ca parg parg fcparg_d | [N][M][MAl][C] Na-Al-Ca parg | 1st simplex x_Mg,2 0 1 .25 0 | bulk Mg/M x_Fe,2 by difference | 2nd simplex z_MTi 0 1 .25 0 | T on M2 z_MFe 0 1 .25 0 z_MAl by difference | 4th simplex z_K,2 0 .25 .25 0 | K,Na on A z_Na,2 by difference | 3rd simplex z_M,2 0 .5 .25 0 | M,C on M4 z_C,2 by difference | ------------------------------------------- | subcomposition 3 [Mg,Fe]*[N,CN,MN]*[K,N]*[Al,Ti,Fe3] |4 | number of simplices |2 3 2 3 | number of vertices on each simplex [Mg,Fe]*[N,CN,MN]*[K,N]*[Al,Ti,Fe3] |kmtn_d kftn_d | [K][M][MTi][N] K-Ti-N-Pgl |kmtmn_d kftfn_d | [K][M][Ti][MN] K-Ti-MN-Pgl** |kmtcn_d kftcn_d | [K][M][Ti][CN] K-Ti-CN-Pgl** |nmtn_d nftn_d | [N][M][MTi][N] N-Ti-N-Pgl |nmtmn_d nftfn_d | [N][M][Ti][MN] N-Ti-MN-Pgl** |nmtcn_d nftcn_d | [N][M][Ti][CN] N-Ti-CN-Pgl** |kmfn_d kffn_d | [K][M][MFe][N] K-Fe-N-Pgl |kmfmn_d kfffn_d | [K][M][Fe][MN] K-Fe-MN-Pgl** |kmfcn_d kffcn_d | [K][M][Fe][CN] K-Fe-CN-Pgl** |nmfn_d nffn_d | [N][M][MFe][N] N-Fe-N-Pgl |nmfmn_d nfffn_d | [N][M][Fe][MN] N-Fe-MN-Pgl** |nmfcn_d nffcn_d | [N][M][Fe][CN] N-Fe-CN-Pgl** |kman_d kfan_d | [K][M][MAl][N] K-Al-N-Pgl |kmamn_d kfafn_d | [K][M][Al][MN] K-Al-MN-Pgl** |kmacn_d kfacn_d | [K][M][Al][CN] K-Al-CN-Pgl** |nman_d nfan_d | [N][M][MAl][N] N-Al-N-Pgl |nmamn_d nfafn_d | [N][M][Al][MN] N-Al-MN-Pgl** |nmacn_d nfacn_d | [N][M][Al][CN] N-Al-CN-Pgl** | 1st simplex |x_Mg,3 0 1 .1 0 | bulk Mg/M |x_Fe,3 by difference | 2nd simplex |z_MN,3 0 1 .1 0 | MN,CN,N on M4 |z_CN,3 0 1 .1 0 |z_N,3 by difference | 3rd simplex |z_K,3 0 1 .1 0 | K,Na on A |z_Na,3 by difference | 4th simplex |z_Ti,3 0 .2 .1 0 | T on M2 |z_Fe,3 0 .7 .1 0 |z_Al,3 by difference | ------------------------------------------- | subcomposition 4 [Mg,Fe]*[ ][M][M][M,C] 2 | number of simplices 2 2 | number of vertices on each non-zero-d simplex cumm grun | M-Ca divalent amphibole tr ftr_d | First simplex x_Mg 0 1 .25 0 | X(1,1) = bulk Mg/M x_Fe by difference | Second simplex z_M 0 1 .25 0 | [ ][M][M][M] z_C by difference | [ ][M][M][C] begin_ordered_endmembers a = 3/7 cumm + 4/7 grun delta_g_of_ordering = -9486 | = DQF(A) -4/7 DQF(Grun) b = 2/7 cumm + 5/7 grun delta_g_of_ordering = -11657 | = DQF(B) -5/7 DQF(Grun) end_ordered_endmembers begin_dependent_endmembers |================================================== | dependent endmembers for Tschermakitic subcomposition | T = Al fcts_d = 1 ts_q + 1 grun - 1 a fts_d = 1 ts_q - 1 tr + 1 b mts_d = 1 ts_q - 1 tr + 1 cumm fgl_d = 1 gl + 1 grun - 1 a | T = Ti vmtn_d = 1 gl + 1 tts - 1 ts_q vftn_d = 1 gl + 1 grun - 1 a + 1 tts - 1 ts_q fctts_d = 1 tts + 1 grun - 1 a ftts_d = 1 tts - 1 tr + 1 b mtts_d = 1 tts - 1 tr + 1 cumm | T = Fe3 frb_d = 1 mrbG + 1 grun - 1 a mcgrk_d = 1 mrbG + 1 ts_q - 1 gl mgrk_d = 1 mrbG + 1 ts_q - 1 gl - 1 tr + 1 cumm fgrk_d = 1 mrbG + 1 ts_q - 1 gl - 1 tr + 1 b fcgrk_d = 1 mrbG + 1 ts_q - 1 gl - 1 a + 1 grun |================================================== | dependent endmembers for Pargasitic subcomposition | M or Ca on M4, Al on M2 fcparg_d = 1 parg + 3/2 grun - 1 a - 1/2 b fparg_d = 1 parg + 1/2 grun + 1/2 b - 1 tr mparg_d = 1 parg - 1 tr + 1 cumm kfcprg_d = 1 kprg + 3/2 grun - 1 a - 1/2 b kfparg_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr kmparg_d = 1 kprg - 1 tr + 1 cumm | M or Ca on M4, Ti on M2 tcmpg_d = 1 parg + 1/2 tts - 1/2 ts_q tcfpg_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 tts - 1/2 ts_q tmpg_d = 1 parg - 1 tr + 1 cumm + 1/2 tts - 1/2 ts_q tfpg_d = 1 parg + 1/2 grun + 1/2 b - 1 tr + 1/2 tts - 1/2 ts_q tcmkpg_d = 1 kprg + 1/2 tts - 1/2 ts_q tcfkpg_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1/2 tts - 1/2 ts_q tmkpg_d = 1 kprg - 1 tr + 1 cumm + 1/2 tts - 1/2 ts_q tfkpg_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr + 1/2 tts - 1/2 ts_q | M or Ca on M4, Fe3 on M2 ocmpg_d = 1 parg + 1/2 mrbG - 1/2 gl ocfpg_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG - 1/2 gl ompg_d = 1 parg - 1 tr + 1 cumm + 1/2 mrbG - 1/2 gl ofpg_d = 1 parg + 1/2 grun + 1/2 b - 1 tr + 1/2 mrbG - 1/2 gl ocmkpg_d = 1 kprg + 1/2 mrbG - 1/2 gl ocfkpg_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG - 1/2 gl omkpg_d = 1 kprg - 1 tr + 1 cumm + 1/2 mrbG - 1/2 gl ofkpg_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr + 1/2 mrbG - 1/2 gl |================================================== | dependent endmembers for Piglitic subcomposition | Na on M4 |kmtn_d = 1 kprg + 1/2 tts - 3/2 ts_q + 1 gl |kftn_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1/2 tts - 3/2 ts_q + 1 gl |kmfn_d = 1 kprg + 1/2 mrbG + 1/2 gl - 1 ts_q |kffn_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG + 1/2 gl - 1 ts_q |kman_d = 1 kprg + 1 gl - 1 ts_q |kfan_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1 gl - 1 ts_q |nmtn_d = 1 parg + 1/2 tts - 3/2 ts_q + 1 gl |nftn_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 tts - 3/2 ts_q + 1 gl |nmfn_d = 1 parg + 1/2 mrbG + 1/2 gl - 1 ts_q |nffn_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG + 1/2 gl - 1 ts_q |nman_d = 1 parg + 1 gl - 1 ts_q |nfan_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1 gl - 1 ts_q | MgNa on M4 |nmamn_d = 1 parg - 1 tr + 1/2 cumm + 1/2 gl |nmfmn_d = 1 parg - 1 tr + 1/2 cumm + 1/2 mrbG |nmtmn_d = 1 parg - 1 tr + 1/2 cumm + 1/2 tts - 1/2 ts_q + 1/2 gl |kmamn_d = 1 kprg - 1 tr + 1/2 cumm + 1/2 gl |kmfmn_d = 1 kprg - 1 tr + 1/2 cumm + 1/2 mrbG |kmtmn_d = 1 kprg - 1 tr + 1/2 cumm + 1/2 tts - 1/2 ts_q + 1/2 gl | Mg-MgCa on M4 |nmacn_d = 1 parg + 1/2 gl - 1/2 ts_q |nmfcn_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG - 1/2 ts_q |nmtcn_d = 1 parg + 1/2 tts - 1 ts_q + 1/2 gl |kmacn_d = 1 kprg + 1/2 gl - 1/2 ts_q |kmfcn_d = 1 kprg + 1/2 mrbG - 1/2 ts_q |kmtcn_d = 1 kprg + 1/2 tts - 1 ts_q + 1/2 gl | FeNa on M4 |nfafn_d = 1 parg + 1/2 grun + 1/2 b - 1 tr + 1/2 gl - 1/2 a |nfffn_d = 1 parg + 1/2 grun + 1/2 b - 1 tr + 1/2 mrbG - 1/2 a |nftfn_d = 1 parg + 1/2 grun + 1/2 b - 1 tr + 1/2 tts - 1/2 ts_q + 1/2 gl - 1/2 a |kfafn_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr + 1/2 gl - 1/2 a |kfffn_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr + 1/2 mrbG - 1/2 a |kftfn_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr + 1/2 tts - 1/2 ts_q + 1/2 gl - 1/2 a | Fe-FeCa on M4 |nfacn_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 gl - 1/2 ts_q |nffcn_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG - 1/2 ts_q |nftcn_d = 1 parg + 1/2 grun + 1/2 b - 1 tr + 1/2 tts - 1/2 ts_q + 1/2 gl - 1/2 a |kfacn_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1/2 gl - 1/2 ts_q |kffcn_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG - 1/2 ts_q |kftcn_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr + 1/2 tts - 1/2 ts_q + 1/2 gl - 1/2 a |================================================== | dependent endmembers for Tremolitic subcomposition ftr_d = 1 tr + 2 grun - 1 a - 1 b end_dependent_endmembers begin_excess_function W(tr ts_q) 20d3 W(tr parg) 25d3 W(tr gl) 65d3 W(tr cumm) 45d3 W(tr grun) 75d3 W(tr a) 57d3 W(tr b) 63d3 W(tr mrbG) 52d3 W(tr kprg) 30d3 W(tr tts) 85d3 W(ts_q parg) -40d3 W(ts_q gl) 25d3 W(ts_q cumm) 70d3 W(ts_q grun) 80d3 W(ts_q a) 70d3 W(ts_q b) 72.5d3 W(ts_q mrbG) 20d3 W(ts_q kprg) -40d3 W(ts_q tts) 35d3 W(parg gl) 50d3 W(parg cumm) 90d3 W(parg grun) 106.7d3 W(parg a) 94.8d3 W(parg b) 94.8d3 W(parg mrbG) 40d3 W(parg kprg) 8d3 W(parg tts) 15d3 W(gl cumm) 100d3 W(gl grun) 113.5d3 W(gl a) 100d3 W(gl b) 111.2d3 W(gl kprg) 54d3 W(gl tts) 75d3 W(cumm grun) 33d3 W(cumm a) 18d3 W(cumm b) 23d3 W(cumm mrbG) 80d3 W(cumm kprg) 87d3 W(cumm tts) 100d3 W(grun a) 12d3 W(grun b) 8d3 W(grun mrbG) 91d3 W(grun kprg) 96d3 W(grun tts) 65d3 W(a b) 20d3 W(a mrbG) 80d3 W(a kprg) 94d3 W(a tts) 95d3 W(b mrbG) 90d3 W(b kprg) 94d3 W(b tts) 95d3 W(mrbG kprg) 50d3 W(mrbG tts) 50d3 W(kprg tts) 35d3 end_excess_function 6 | number of sites in configurational entropy model (A, M1, M2, M4, T1, OH) A | site name 3 1 1 | number of species, effective multiplicity, true multiplicity z(Na,A) = 1 parg z(K,A) = 1 kprg z(Vac,A) = 1 cumm + 1 grun + 1 tr + 1 ts_q + 1 gl + 1 a + 1 b + 1 tts + 1 mrbG M1 | site name 2 3 3 | number of species, effective multiplicity, true multiplicity z(Fe,M1) = 1 grun + 1 b z(Mg,M1) = 1 cumm + 1 tr + 1 ts_q + 1 parg + 1 kprg + 1 mrbG + 1 gl + 1 tts + 1 a M2 | site name 5 2 2 | number of species, effective multiplicity, true multiplicity z(Fe,M2) = 1 grun + 1 a z(Fe3,M2) = 1 mrbG z(Al,M2) = 1 ts_q + 1 gl + 1/2 kprg + 1/2 parg z(Ti,M2) = 1 tts z(Mg,M2) = 1/2 kprg + 1/2 parg + 1 tr + 1 cumm + 1 b M4 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Mg,M4) = 1 cumm z(Na,M4) = 1 gl + 1 mrbG z(Fe,M4) = 1 grun + 1 a + 1 b z(Ca,M4) = 1 ts_q + 1 parg + 1 kprg + 1 tts + 1 tr T1 | site name 2 1 4 | number of species, effective multiplicity, true multiplicity z(Al,T1) = 1/2 ts_q + 1/2 parg + 1/2 kprg + 1/2 tts z(Si,T1) = 1/2 ts_q + 1/2 parg + 1/2 kprg + 1/2 tts + 1 gl + 1 mrbG + 1 tr + 1 grun + 1 a + 1 b + 1 cumm OH | site name 2 2 2 | number of species, effective multiplicity, true multiplicity z(O,OH) = 1 tts z(OH,OH) = 1 cumm + 1 grun + 1 ts_q + 1 tr + 1 parg + 1 kprg + 1 mrbG + 1 gl + 1 b + 1 a [Si4O22] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(tr) 1 alpha(ts_q) 1.5 alpha(parg) 1.7 alpha(gl) 0.8 alpha(cumm) 1 alpha(grun) 1 alpha(a) 1 alpha(b) 1 alpha(mrbG) 0.8 alpha(kprg) 1.7 alpha(tts) 1.5 end_van_laar_sizes begin_dqf_corrections DQF(gl) = -3000 DQF(parg) = -10000 DQF(ts_q) = 10000 DQF(grun) = -3000 end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers | use_model_resolution end_of_model -------------------------------------------------------- begin_model cAmph(G) => Green et al. (JMG, 2016) clinoamphibole. --------------------------------------------------------- Reformulated as a 2-polytope 688 format model. JADC, 11/19 The composition space of the complete (published) model corresponds to: [A][M][M,T,MT][M,C,N] where the brackets indicate the A, M1, M2, and M4 sites (for brevity the dependent T1 and OH site chemistry has been dropped) and A = Na, K, V T = Al, Fe3, Ti* MT = MAl, MFe3, MTi C = Ca M = Mg, Fe N = Na *The endmembers calibrated by Green et al. (2016) do not allow Ti-glaucophane The composition space of the model here has been simplified to [A][M][M,T,MAl][M,C,N] where the brackets indicate the A, M1, M2, and M4 sites and A = Na, K, V T = Al, Fe3, Ti* MT = MAl C = Ca M = Mg, Fe N = Na NOTE: the definitions for the MFe3 and MTi pargasite exchanges allowed in the full model are provided below but have been commented out below because they are costly and do not seem to have any significant effect on computed phase relations, this may also be true of the M-Ca-Fe3+_tschermaks [grk] exchange, although this has not been tested. It is probable that VERTEX/MEEMUM will run out of dynamic memory if the full model is restored, if this occurs uncomment the low_reach option keyword at the end of this model. The low_reach option dramatically reduces the memory and time required for calculations, but also causes a significant deterioration in phase diagram quality. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = Glaucophane = [ ][M][T'][N] => 2*2 Polytope 2 = ACMT-Amph = [A][M][M,T,MAl][M,C] => 2*2*6 where T is generalized Tschermaks [M][T][M,C] MT is generalized pargasite [M][MT][N] N is generalized glaucophane [M][T][N] M on M2 is generalized amphibole [M][M][M,C] --------------------------------------------------------- NOTE: to use this the following endmembers must be specified with make definitions in the thermodynamic data file mrbG = 1 gl - 1 gr + 1 andr DQF = 0 kprg = 1 mu - 1 pa + 1 parg DQF = -7060 + 20 * T_K tts = -2 dsp + 2 ru + 1 ts DQF = 95000 to avoid cluttering thermodynamic data files with the new plague of TC DQF "corrections", the following simple DQF's are specified in the make_definition section at the end of this model: DQF(gl) = -3000 DQF(parg) = -10000 DQF(ts_q) = 10000 DQF(grun) = -3000 For (nonsensical, but common) positive DFQ's this approach may lead to interference between the phase relations of the solution model and the un-DQF'd endmember. If such interference occurs: the DQF'd endmember must be renamed; its definition placed in the thermodynamic data file; and the un-DQF'd endmember excluded from the calculation. --------------------------------------------------------- independent endmembers and site populations: Sites A M1 M2 M4 T1 OH _______________________________________ Multiplicity 1 3 2 2 1(4) 2 _______________________________________ Ts V Mg Al Ca SiAl OH Parg Na Mg MgAl Ca SiAl OH Kprg K Mg MgAl Ca SiAl OH Gl V Mg Al Na Si OH mrbG V Mg Fe3 Na Si OH Tts V Mg Ti Ca Si O Tr V Mg Mg Ca Si OH Cumm V Mg Mg Mg Si OH Grun V Fe Fe Fe Si OH A V Mg Fe Fe Si OH B V Fe Mg Fe Si OH dependent exchange Mgrk V M Fe3 M SiAl OH dependent exchange Mts V M Al M SiAl OH dependent exchange Mpg Na M MAl M SiAl OH dependent exchange MKpg K M MAl M SiAl OH dependent exchange MTts V M Ti M SiAl OH 2 NaCa-1(M4) = (Gl-Ts) 2 CaMg-1(M4) = (Tr-Cumm) 2 Fe3Al-1(M2) = (Rb-Gl) 2 TiAl-1(M2) = (Tts-Ts) => coupled to T1 and OH. 3 FeMg-1(M1) = (grun - a) 2 FeMg-1(M2) = (grun - b) 2 FeMg-1(M4) = (-grun - cumm + a + b) --------------------------------------------------------- cAmph(G) abbreviation Amph full_name clinoamphibole 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes GlRb-Amph 0 1 .1 0 | X([ ][M][Fe3,Al][N]), Mg-Fe-Al-Fe3+ "glaucophane" ACMT-Amph by difference | X([A][M][M,T,MT][M,C]), K-Na-Mg-Fe-Al-Ti-Fe3+ | ------------------------------------------- | Polytope 1 [ ][M][T ][N] (glauc) 2 | number of simplices, then for each simplex 2 2 | number of vertices on each simplex | Simplex 1 Mg, Fe; is Mg/M [M1 site] | Simplex 2 Fe3, Al is Fe3+/T' [M2 site] | endmembers on the vertices, index on the first simplex | changes fastest, last slowest | => endmember 11 is [ ][Mg][Fe3][Na] | => endmember 21 is [ ][Fe][Fe3][Na] | => endmember 12 is [ ][Mg][Al ][Na] | => endmember 22 is [ ][Fe][Al ][Na] mrbG frb_d | Fe3-glaucophane, aka riebeckite gl fgl_d | Al-glaucophane | First (1d) simplex X_Mg1 0 1 .25 0 | X(1,1) = bulk Mg/M X_Fe1 by difference | Second (3d) simplex X_rb 0 1 .25 0 | X(2,1) = Fe3/T' on M2, T' = Al + Fe3+ X_gl by difference | ------------------------------------------- | Polytope 2 [A][M][M,T,MT][M,C] 3 | number of simplices 2 2 6 | number of vertices on each simplex, change to 2 2 10 for full model. | Simplex 1 is Mg/M | Simplex 2 is M/[Ca + M] on M4 | Simplex 3 is parg and M, Fe3, Ti, Al M2 site chemistry | endmembers on the vertices, index on the first simplex | changes fastest, last slowest mts_d fts_d | M-Ca-Al_tschermaks ts_q fcts_d mparg_d fparg_d | M-Ca-Na-Al_parg parg fcparg_d | ompg_d ofpg_d | M-Ca-Na-Fe3+ parg | ocmpg_d ocfpg_d | tmpg_d tfpg_d | M-Ca-Na-Ti parg | tcmpg_d tcfpg_d kmparg_d kfparg_d | M-Ca-K-Al_parg kprg kfcprg_d | omkpg_d ofkpg_d | M-Ca-K-Fe3+ parg | ocmkpg_d ocfkpg_d | tmkpg_d tfkpg_d | M-Ca-K-Ti parg | tcmkpg_d tcfkpg_d mtts_d ftts_d | M-Ca-Ti_tschermaks tts fctts_d mgrk_d fgrk_d | M-Ca-Fe3+_tschermaks (Na-free riebeckite) mcgrk_d fcgrk_d cumm grun | M-Ca divalent amphibole tr ftr_d | First (1d) simplex X_Mg 0 1 .25 0 | X(1,1) = bulk Mg/M X_Fe by difference | Second (1d) simplex X_M_M2 0 1 .25 0 | X(2,1) = M/[Ca + M] on M4 X_C_M2 by difference | Third simplex X_AlTs 0 1 .25 0 | X(3,1) = X(Al-Ts) X_NaPg 0 1 .25 0 | X(3,2) = X(Na-Al-Parg) | 0 .45 .1 0 | X(3,3) = X(Na-Fe3+-Parg) | 0 .3 .1 0 | X(3,4) = X(Na-Ti-Parg) X_KPg 0 1 .25 0 | X(3,5) = X(K-Al-Parg) | 0 .1 .1 0 | X(3,6) = X(K-Fe3+-Parg) | 0 .1 .1 0 | X(3,7) = X(K-Ti-Parg) X_TiTs 0 1 .25 0 | X(3,8) = X(Ti-Ts) X_FeTs 0 1 .25 0 | X(3,9) = X(Fe3-Ts) X_MCamph by difference begin_ordered_endmembers a = 3/7 cumm + 4/7 grun delta_g_of_ordering = -9486 | = DQF(A) -4/7 DQF(Grun) b = 2/7 cumm + 5/7 grun delta_g_of_ordering = -11657 | = DQF(B) -5/7 DQF(Grun) end_ordered_endmembers begin_dependent_endmembers fcts_d = 1 ts_q + 1 grun - 1 a fts_d = 1 ts_q - 1 tr + 1 b mts_d = 1 ts_q - 1 tr + 1 cumm fcparg_d = 1 parg + 3/2 grun - 1 a - 1/2 b fparg_d = 1 parg + 1/2 grun + 1/2 b - 1 tr mparg_d = 1 parg - 1 tr + 1 cumm |tcmpg_d = 1 parg + 1/2 tts - 1/2 ts_q |tcfpg_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 tts - 1/2 ts_q |tmpg_d = 1 parg - 1 tr + 1 cumm + 1/2 tts - 1/2 ts_q |tfpg_d = 1 parg + 1/2 grun + 1/2 b - 1 tr + 1/2 tts - 1/2 ts_q |ocmpg_d = 1 parg + 1/2 mrbG - 1/2 gl |ocfpg_d = 1 parg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG - 1/2 gl |ompg_d = 1 parg - 1 tr + 1 cumm + 1/2 mrbG - 1/2 gl |ofpg_d = 1 parg + 1/2 grun + 1/2 b - 1 tr + 1/2 mrbG - 1/2 gl kfcprg_d = 1 kprg + 3/2 grun - 1 a - 1/2 b kfparg_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr kmparg_d = 1 kprg - 1 tr + 1 cumm |tcmkpg_d = 1 kprg + 1/2 tts - 1/2 ts_q |tcfkpg_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1/2 tts - 1/2 ts_q |tmkpg_d = 1 kprg - 1 tr + 1 cumm + 1/2 tts - 1/2 ts_q |tfkpg_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr + 1/2 tts - 1/2 ts_q |ocmkpg_d = 1 kprg + 1/2 mrbG - 1/2 gl |ocfkpg_d = 1 kprg + 3/2 grun - 1 a - 1/2 b + 1/2 mrbG - 1/2 gl |omkpg_d = 1 kprg - 1 tr + 1 cumm + 1/2 mrbG - 1/2 gl |ofkpg_d = 1 kprg + 1/2 grun + 1/2 b - 1 tr + 1/2 mrbG - 1/2 gl fgl_d = 1 gl + 1 grun - 1 a frb_d = 1 mrbG + 1 grun - 1 a fctts_d = 1 tts + 1 grun - 1 a ftts_d = 1 tts - 1 tr + 1 b mtts_d = 1 tts - 1 tr + 1 cumm mcgrk_d = 1 mrbG + 1 ts_q - 1 gl mgrk_d = 1 mrbG + 1 ts_q - 1 gl - 1 tr + 1 cumm fgrk_d = 1 mrbG + 1 ts_q - 1 gl - 1 tr + 1 b fcgrk_d = 1 mrbG + 1 ts_q - 1 gl - 1 a + 1 grun ftr_d = 1 tr + 2 grun - 1 a - 1 b end_dependent_endmembers begin_excess_function W(tr ts_q) 20d3 W(tr parg) 25d3 W(tr gl) 65d3 W(tr cumm) 45d3 W(tr grun) 75d3 W(tr a) 57d3 W(tr b) 63d3 W(tr mrbG) 52d3 W(tr kprg) 30d3 W(tr tts) 85d3 W(ts_q parg) -40d3 W(ts_q gl) 25d3 W(ts_q cumm) 70d3 W(ts_q grun) 80d3 W(ts_q a) 70d3 W(ts_q b) 72.5d3 W(ts_q mrbG) 20d3 W(ts_q kprg) -40d3 W(ts_q tts) 35d3 W(parg gl) 50d3 W(parg cumm) 90d3 W(parg grun) 106.7d3 W(parg a) 94.8d3 W(parg b) 94.8d3 W(parg mrbG) 40d3 W(parg kprg) 8d3 W(parg tts) 15d3 W(gl cumm) 100d3 W(gl grun) 113.5d3 W(gl a) 100d3 W(gl b) 111.2d3 W(gl kprg) 54d3 W(gl tts) 75d3 W(cumm grun) 33d3 W(cumm a) 18d3 W(cumm b) 23d3 W(cumm mrbG) 80d3 W(cumm kprg) 87d3 W(cumm tts) 100d3 W(grun a) 12d3 W(grun b) 8d3 W(grun mrbG) 91d3 W(grun kprg) 96d3 W(grun tts) 65d3 W(a b) 20d3 W(a mrbG) 80d3 W(a kprg) 94d3 W(a tts) 95d3 W(b mrbG) 90d3 W(b kprg) 94d3 W(b tts) 95d3 W(mrbG kprg) 50d3 W(mrbG tts) 50d3 W(kprg tts) 35d3 end_excess_function 6 | number of sites in configurational entropy model (A, M1, M2, M4, T1, OH) A | site name 3 1 1 | number of species, effective multiplicity, true multiplicity z(Na,A) = 1 parg z(K,A) = 1 kprg z(Vac,A) = 1 cumm + 1 grun + 1 tr + 1 ts_q + 1 gl + 1 a + 1 b + 1 tts + 1 mrbG M1 | site name 2 3 3 | number of species, effective multiplicity, true multiplicity z(Fe,M1) = 1 grun + 1 b z(Mg,M1) = 1 cumm + 1 tr + 1 ts_q + 1 parg + 1 kprg + 1 mrbG + 1 gl + 1 tts + 1 a M2 | site name 5 2 2 | number of species, effective multiplicity, true multiplicity z(Fe,M2) = 1 grun + 1 a z(Fe3,M2) = 1 mrbG z(Al,M2) = 1 ts_q + 1 gl + 1/2 kprg + 1/2 parg z(Ti,M2) = 1 tts z(Mg,M2) = 1/2 kprg + 1/2 parg + 1 tr + 1 cumm + 1 b M4 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Mg,M4) = 1 cumm z(Na,M4) = 1 gl + 1 mrbG z(Fe,M4) = 1 grun + 1 a + 1 b z(Ca,M4) = 1 ts_q + 1 parg + 1 kprg + 1 tts + 1 tr T1 | site name 2 1 4 | number of species, effective multiplicity, true multiplicity | for some reason tts was "Al-free" here, but not in cAmph(G)_I | corrected, DEK 22/9/21 z(Al,T1) = 1/2 ts_q + 1/2 parg + 1/2 kprg + 1/2 tts z(Si,T1) = 1/2 ts_q + 1/2 parg + 1/2 kprg + 1/2 tts + 1 gl + 1 mrbG + 1 tr + 1 grun + 1 a + 1 b + 1 cumm OH | site name 2 2 2 | number of species, effective multiplicity, true multiplicity z(O,OH) = 1 tts z(OH,OH) = 1 cumm + 1 grun + 1 ts_q + 1 tr + 1 parg + 1 kprg + 1 mrbG + 1 gl + 1 b + 1 a [Si4O22] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(tr) 1 alpha(ts_q) 1.5 alpha(parg) 1.7 alpha(gl) 0.8 alpha(cumm) 1 alpha(grun) 1 alpha(a) 1 alpha(b) 1 alpha(mrbG) 0.8 alpha(kprg) 1.7 alpha(tts) 1.5 end_van_laar_sizes begin_dqf_corrections DQF(gl) = -3000 DQF(parg) = -10000 DQF(ts_q) = 10000 DQF(grun) = -3000 end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers | use_model_resolution end_of_model -------------------------------------------------------- begin_model Aq_solvent abbreviation Aq full_name special_purpose 20 | model type: electrolytic fluid 1 | = ns = number of solvent species H2O 1 | = nn = number of neutral solute species SiO2,aq 5 | = nq = number of charged solute species H+ HSiO3- Mg+2 MgOH+ OH- | ranges on the first ns-1 = 0 species | ranges (mol fraction) on the nn neutral solute species 1e-3 1e-2 .1 1 | ranges (mol fraction) on the first nq-1 charged solute species 1e-7 2e-6 .1 1 1e-10 1e-9 .1 1 1e-9 6e-8 .1 1 1e-7 3e-6 .1 1 end_of_model begin_model Aq_yuri abbreviation Aq full_name special_purpose 20 | model type: electrolytic fluid 3 | = ns = number of solvent species CO2 CH4 H2O 1 | = nn = number of neutral solute species SiO2,aq 6 | = nq = number of charged solute species HCO3- HS- OH- HSiO3- H+ Ca(HCO3) | ranges on the first ns-1 = 0 species 1e-5 5e-2 .1 1 | subdivision ranges, imod = 1 -> non-linear subdivision 1e-5 2e-4 .1 1 | subdivision ranges, imod = 1 -> non-linear subdivision | ranges (mol fraction) on the nn neutral solute species 1e-3 1e-1 .2 1 | ranges (mol fraction) on the first nq-1 charged solute species 1e-5 1e-1 .2 1 1e-5 1e-1 .2 1 1e-5 1e-1 .2 1 1e-5 1e-1 .2 1 1e-5 1e-1 .2 1 end_of_model Aq_solven0 abbreviation Aq full_name special_purpose 20 | model type: electrolytic fluid 3 | = ns = number of solvent species, water must be the last solvent species. CO2 CH4 H2O 2 | = nn = number of neutral solute species SiO2,aq Si2O4,aq 5 | = nq = number of charged species HCO3- Mg(HCO3) OH- HSiO3- Mg(HSiO3 | ranges on the first ns-1 species 1e-2 0.05 .1 1 1e-2 0.05 .1 1 | ranges (mol fraction) on the nn neutral solute species 1e-3 0.8d-2 .1 1 1e-3 0.8d-2 .1 1 | ranges (mol fraction) on the first nq-1 charged solute species 1e-7 1e-6 .1 1 1e-7 1e-6 .1 1 1e-7 1e-6 .1 1 1e-7 1e-6 .1 1 end_of_model -------------------------------------------------------- begin_model Clinopyroxene, Cpx_I(HGP) - Holland, Green, and Powell; Journal of Petrology, 2018. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus --------------------------------------------------------- Reformulated as a 2-subcomposition 688 format model. JADC, 9/19 The composition space of the published model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3, Cr, MTi on M1 C = Ca on M2 M = Mg, Fe A = Na, K on M2* *A is fully coupled to T by charge balance. The composition space of the model is formed as a mixture of 3 subcompositions (dropping the dependent T site notation): subcomposition 1 = [Al,Fe3,Cr]*[N,K,C] subcomposition 2 = [Mg,Fe]*[Al,Fe3,Cr,MTi]*[N,K] subcomposition 3 = [Mg,Fe]*[M]*[M,C] --------------------------------------------------------- First coded for Perple_X by Julien Cornet Sept 2018. Re-formulated as a triple simplex prism. JADC, Dec 11, 2018. Corrected cfm enthalpy and site limit expression for z(Mg,M1). JADC, April 29, 2019. NOTES: to use this model, the following endmembers must be specified with make definitions in the thermodynamic data file: --------------------------------------------------------- M1 M2 T ____________________________________ Multiplicity 1 1 2 -> fake T multiplicity = 1/2 ____________________________________ subcomposition 1: mkbuf_d MgTi K Si dependent mnbuf_d MgTi Na Si dependent mcbuf_q MgTi Ca AlSi _________ fkbuf_d FeTi K Si dependent fnbuf_d FeTi Na Si dependent fcbuf_d FeTi Ca AlSi dependent _________ crkjd_d Cr K Si dependent crjd_d Cr Na Si dependent crdi_q Cr Ca AlSi _________ kess_d Fe3+ K Si dependent ness_d Fe3+ Na Si dependent cess_q Fe3 Ca AlSi _________ kjd_q Al K Si jd Al Na Si cats Al Ca AlSi ____________________________________ subcomposition 2: mmbuf_d MgTi Mg AlSi dependent ffbuf_d FeTi Fe AlSi dependent cren_d Cr Mg AlSi dependent crfs_d Cr Fe AlSi dependent mess_d Fe3 Mg AlSi dependent fess_d Fe3 Fe AlSi dependent mats_d Al Mg AlSi dependent fats_d Al Fe AlSi dependent di Mg Ca Si hed_d Fe Ca Si dependent cen_q Mg Mg Si cfs_q Fe Fe Si _______________________________________________ Ordered species: cfm Mg Fe Si independent femg-1(M2) = cfm - cen_q femg-1(M1) = cfs_q - cfm mgca-1(M2) = cen_q - di feca-1(M2) = cfm - di --------------------------------------------------------- Cpx_I(HGP) abbreviation Cpx full_name clinopyroxene 688 | model type: 688 format standard model 2 | number of subcompositions | subcomposition names and composite composition space subdivision schemes [T][A,C,M] 0 .25 .1 0 [M][C,M] by difference | ------------------------------------------- | subcomposition 1 [T][A,C,M] 2 | number of simplices 3 4 | number of vertices on each simplex, change to 3 5 for full model mkbuf_d mnbuf_d mcbuf_q crkjd_d crjd_d crdi_q kess_d ness_d cess_q |fkbuf_d fnbuf_d fcbuf_d kjd_q jd cats | Simplex 1 X_K 0 1 .1 0 | X(1,1) = Na/[A + Ca] on M2 X_Na 0 1 .1 0 | X(1,2) = K /[A + Ca] on M2 X_Ca by difference | X(1,N) = Ca/[A + Ca] on M2 | Simplex 2 X_TiTs 0 1 .1 0 | X(2,1) = MgTi/T on M1 X_CrTs 0 1 .1 0 | X(2,2) = Cr/T on M1 X_FeTs 0 1 .1 0 | X(2,3) = Fe3+/T on M1 |0 1 .1 0 |X(2,4) = FeTi/T on M1 X_AlTs by difference | X(2,N) = Al/T on M1 | ------------------------------------------- | subcomposition 2 TM+MC 2 | number of simplices 2 6 | number of vertices on each simplex ffbuf_d mmbuf_d crfs_d cren_d fess_d mess_d fats_d mats_d cfs_q cen_q hed_d di | Simplex 1 X_Fe 0 .2 .1 0 | X(1,1) => Fe/M X_Mg by difference | X(1,N) = Mg/M | Simplex 2 X_MTiTs 0 .1 .1 0 | X(2,1) - M-buf(MgTi) X_MCrTs 0 .1 .1 0 | X(2,2) - M-Cr X_MFeTs 0 .1 .1 0 | X(2,3) - M-Fe3+ X_MAlTs 0 .2 .1 0 | X(2,4) - M-Ts X_MMPx 0 1 .1 0 | X(2,5) - M-M X_MCPx by difference | X(2,6) - M-Ca begin_ordered_endmembers cfm = 1/2 cen_q + 1/2 cfs_q delta_g_of_ordering = -4400 | DQF(cfm) - (DQF(ceng) + DQF(cfs_q))/2 end_ordered_endmembers begin_dependent_endmembers mkbuf_d = -1 cats + 1 mcbuf_q + 1 kjd_q mnbuf_d = -1 cats + 1 mcbuf_q + 1 jd |fcbuf_d = 1/2 cfs_q + 1 mcbuf_q - 1/2 cfm |fkbuf_d = 1/2 cfs_q - cats + mcbuf_q - 1/2 cfm + kjd_q |fnbuf_d = 1/2 cfs_q - 1 cats + 1 mcbuf_q + jd - 1/2 cfm ffbuf_d = -1 di + 1/2 cfs_q + 1 mcbuf_q + 1/2 cfm mmbuf_d = -1 di + 1 mcbuf_q + 1 cen_q crkjd_d = -1 cats + 1 crdi_q + 1 kjd_q crjd_d = -1 cats + 1 crdi_q + 1 jd cren_d = -1 di + 1 crdi_q + 1 cen_q crfs_d = -1 di + 1 crdi_q + 1 cfm kess_d = -1 cats + 1 cess_q + 1 kjd_q ness_d = -1 cats + 1 cess_q + 1 jd mess_d = -1 di + 1 cess_q + 1 cen_q fess_d = -1 di + 1 cess_q + 1 cfm mats_d = -1 di + 1 cats + 1 cen_q fats_d = -1 di + 1 cats + 1 cfm hed_d = 1 di + 1 cfs_q - 1 cfm end_dependent_endmembers begin_excess_function W(di cfs_q) 25.8d3 - 0.03 * p W(di cats) 13d3 - 0.06 * p W(di crdi_q) 8d3 W(di cess_q) 8d3 W(di jd) 26d3 W(di cen_q) 29.8d3 - 0.03 * p W(di cfm) 20.6d3 - 0.03 * p W(di kjd_q) 26d3 W(cfs_q cats) 25d3 - .1 * p W(cfs_q crdi_q) 38.3d3 W(cfs_q cess_q) 43.3d3 W(cfs_q jd) 24d3 W(cfs_q cen_q) 2.3d3 W(cfs_q cfm) 3.5d3 W(cfs_q kjd_q) 24d3 W(cats crdi_q) 2d3 W(cats cess_q) 2d3 W(cats jd) 6d3 W(cats cen_q) 45.2d3 - 0.35 * p W(cats cfm) 27d3 - .1 * p W(cats kjd_q) 6d3 W(crdi_q cess_q) 2d3 W(crdi_q jd) 3d3 W(crdi_q cen_q) 52.3d3 W(crdi_q cfm) 40.3d3 W(crdi_q kjd_q) 3d3 W(cess_q jd) 3d3 W(cess_q cen_q) 57.3d3 W(cess_q cfm) 45.3d3 W(cess_q kjd_q) 3d3 W(jd cen_q) 40d3 W(jd cfm) 40d3 W(jd kjd_q) 10d3 W(cen_q cfm) 4d3 W(cen_q kjd_q) 40d3 W(cfm kjd_q) 40d3 end_excess_function 3 | number of sites in configurational entropy model (M1, M2, T) M1 | site name 6 1 1 | number of species, effective multiplicity, true multiplicity z(Fe) = 1 cfs_q z(Mg) = 1 di + 1 cen_q + 1/2 mcbuf_q + 1 cfm z(Fe3+) = 1 cess_q z(Ti) = 1/2 mcbuf_q z(Cr) = 1 crdi_q z(Al) = 1 cats + 1 jd + 1 kjd_q M2 | site name 5 1 1 | number of species, effective multiplicity, true multiplicity z(Na) = 1 jd z(K) = 1 kjd_q z(Mg) = 1 cen_q z(Fe) = 1 cfm + 1 cfs_q z(Ca) = 1 di + 1 cats + 1 cess_q + 1 crdi_q + 1 mcbuf_q T | site name 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 cats + 1/2 mcbuf_q + 1/2 crdi_q + 1/2 cess_q z(Si,T) = 1/2 cats + 1/2 mcbuf_q + 1/2 crdi_q + 1/2 cess_q + 1 cfs_q + 1 cen_q + 1 di + 1 cfm + 1 jd + 1 kjd_q [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(di) 1.2 alpha(cen_q) 1 alpha(cfs_q) 1 alpha(jd) 1.2 alpha(kjd_q) 1.2 alpha(cats) 1.9 alpha(mcbuf_q) 1.9 alpha(cess_q) 1.9 alpha(crdi_q) 1.9 alpha(cfm) 1 end_van_laar_sizes begin_dqf_corrections dqf(cen_q) 3500 - 2 * T + 0.048 * P | cenjh dqf(cfs_q) 2100 - 2 * T + 0.045 * P | cfs_q dqf(cess_q) 3450 | cessh dqf(mcbuf_q) -4900 | mcbuf dqf(crdi_q) -4900 | crdi_q dqf(kjd_q) 11.7d3 + 0.6 * P | kdjh end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Cpx(HGP) - Clinopyroxene - Holland, Green, and Powell; Journal of Petrology, 2018. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus --------------------------------------------------------- Reformulated as a 2-polytope 688 format model. JADC, 9/19 The composition space of the published model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3, Cr, MTi on M1 C = Ca on M2 M = Mg, Fe A = Na, K on M2* *A is partially coupled to T by charge balance. The composition space of the model here has been simplified by eliminating the [FeTi][A,C] endmembers. The retained [FeTi][Fe] endmember is probably a waste of resources. NOTE: the definitions for the FeTi exchanges allowed in the full model are provided but have been commented out below. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = [T'][A,C] Polytope 2 = [T,M][M]+[M][C] where T' = Al, Fe3, MgTi on M1 --------------------------------------------------------- First coded for Perple_X by Julien Cornet Sept 2018. Re-formulated as a triple simplex prism. JADC, Dec 11, 2018. Corrected cfm enthalpy and site limit expression for z(Mg,M1). JADC, April 29, 2019. NOTES: to use this model, the following endmembers must be specified with make definitions in the thermodynamic data file: cen_q cfs_q kjd_q crdi_q cess_q mcbuf_q --------------------------------------------------------- M1 M2 T ____________________________________ Multiplicity 1 1 2 -> fake T multiplicity = 1/2 ____________________________________ Polytope 1: mkbuf_d MgTi K Si dependent mnbuf_d MgTi Na Si dependent mcbuf_q MgTi Ca AlSi _________ fkbuf_d FeTi K Si dependent fnbuf_d FeTi Na Si dependent fcbuf_d FeTi Ca AlSi dependent _________ crkjd_d Cr K Si dependent crjd_d Cr Na Si dependent crdi_q Cr Ca AlSi _________ kess_d Fe3+ K Si dependent ness_d Fe3+ Na Si dependent cess_q Fe3 Ca AlSi _________ kjd_q Al K Si jd Al Na Si cats Al Ca AlSi ____________________________________ Polytope 2: mmbuf_d MgTi Mg AlSi dependent ffbuf_d FeTi Fe AlSi dependent cren_d Cr Mg AlSi dependent crfs_d Cr Fe AlSi dependent mess_d Fe3 Mg AlSi dependent fess_d Fe3 Fe AlSi dependent mats_d Al Mg AlSi dependent fats_d Al Fe AlSi dependent di Mg Ca Si hed_d Fe Ca Si dependent cen_q Mg Mg Si cfs_q Fe Fe Si _______________________________________________ Ordered species: cfm Mg Fe Si independent femg-1(M2) = cfm - cen_q femg-1(M1) = cfs_q - cfm mgca-1(M2) = cen_q - di feca-1(M2) = cfm - di --------------------------------------------------------- Cpx(HGP) abbreviation Cpx full_name clinopyroxene 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [T'][A,C] 0 .25 .1 0 TM+MC by difference | ------------------------------------------- | Polytope 1 [T'][A,C] 2 | number of simplices 3 4 | number of vertices on each simplex, change to 3 5 for full model mkbuf_d mnbuf_d mcbuf_q crkjd_d crjd_d crdi_q kess_d ness_d cess_q |fkbuf_d fnbuf_d fcbuf_d kjd_q jd cats | Simplex 1 X_K 0 1 .1 0 | X(1,1) = Na/[A + Ca] on M2 X_Na 0 1 .1 0 | X(1,2) = K /[A + Ca] on M2 X_Ca by difference | X(1,N) = Ca/[A + Ca] on M2 | Simplex 2 X_TiTs 0 1 .1 0 | X(2,1) = MgTi/T on M1 X_CrTs 0 1 .1 0 | X(2,2) = Cr/T on M1 X_FeTs 0 1 .1 0 | X(2,3) = Fe3+/T on M1 |0 1 .1 0 |X(2,4) = FeTi/T on M1 X_AlTs by difference | X(2,N) = Al/T on M1 | ------------------------------------------- | Polytope 2 TM+MC 2 | number of simplices 2 6 | number of vertices on each simplex ffbuf_d mmbuf_d crfs_d cren_d fess_d mess_d fats_d mats_d cfs_q cen_q hed_d di | Simplex 1 X_Fe 0 1 .1 0 | X(1,1) => Fe/M X_Mg by difference | X(1,N) = Mg/M | Simplex 2 X_MTiTs 0 .1 .1 0 | X(2,1) - M-buf(MgTi) X_MCrTs 0 .1 .1 0 | X(2,2) - M-Cr X_MFeTs 0 .1 .1 0 | X(2,3) - M-Fe3+ X_MAlTs 0 .2 .1 0 | X(2,4) - M-Ts X_MMPx 0 1 .1 0 | X(2,5) - M-M X_MCPx by difference | X(2,6) - M-Ca begin_ordered_endmembers cfm = 1/2 cen_q + 1/2 cfs_q delta_g_of_ordering = -4400 | DQF(cfm) - (DQF(cen_q) + DQF(cfs_q))/2 end_ordered_endmembers begin_dependent_endmembers mkbuf_d = -1 cats + 1 mcbuf_q + 1 kjd_q mnbuf_d = -1 cats + 1 mcbuf_q + 1 jd |fcbuf_d = 1/2 cfs_q + 1 mcbuf_q - 1/2 cfm |fkbuf_d = 1/2 cfs_q - cats + mcbuf_q - 1/2 cfm + kjd_q |fnbuf_d = 1/2 cfs_q - 1 cats + 1 mcbuf_q + jd - 1/2 cfm ffbuf_d = -1 di + 1/2 cfs_q + 1 mcbuf_q + 1/2 cfm mmbuf_d = -1 di + 1 mcbuf_q + 1 cen_q crkjd_d = -1 cats + 1 crdi_q + 1 kjd_q crjd_d = -1 cats + 1 crdi_q + 1 jd cren_d = -1 di + 1 crdi_q + 1 cen_q crfs_d = -1 di + 1 crdi_q + 1 cfm kess_d = -1 cats + 1 cess_q + 1 kjd_q ness_d = -1 cats + 1 cess_q + 1 jd mess_d = -1 di + 1 cess_q + 1 cen_q fess_d = -1 di + 1 cess_q + 1 cfm mats_d = -1 di + 1 cats + 1 cen_q fats_d = -1 di + 1 cats + 1 cfm hed_d = 1 di + 1 cfs_q - 1 cfm end_dependent_endmembers begin_excess_function W(di cfs_q) 25.8d3 | changed from W(di cfs_q) = 25.8d3 - 0.03 * p by DB to match the ECRG 10/20 revision. W(di cats) 13d3 - 0.06 * p W(di crdi_q) 8d3 W(di cess_q) 8d3 W(di mcbuf_q) 8d3 | added by DB to match the ECRG 10/20 revision. W(di jd) 26d3 W(di cen_q) 29.8d3 | changed from W(di cen_q) 29.8d3 - 0.03 * p by DB to match the ECRG 10/20 revision. W(di cfm) 20.6d3 | changed from W(di cfm) 20.6d3 - 0.03 * p by DB to match the ECRG 10/20 revision. W(di kjd_q) 26d3 W(cfs_q cats) 25d3 - .1 * p W(cfs_q crdi_q) 38.3d3 W(cfs_q cess_q) 43.3d3 W(cfs_q jd) 24d3 W(cfs_q mcbuf_q) 24d3 | added by DB to match the ECRG 10/20 revision. W(cfs_q cen_q) 2.3d3 W(cfs_q cfm) 3.5d3 W(cfs_q kjd_q) 24d3 W(cats crdi_q) 2d3 W(cats cess_q) 2d3 W(cats mcbuf_q) 6d3 | added by DB to match the ECRG 10/20 revision. W(cats jd) 6d3 W(cats cen_q) 45.2d3 - 0.35 * p W(cats cfm) 27d3 - .1 * p W(cats kjd_q) 6d3 W(crdi_q cess_q) 2d3 W(crdi_q mcbuf_q) 6d3 | added by DB to match the ECRG 10/20 revision. W(crdi_q jd) 3d3 W(crdi_q cen_q) 52.3d3 W(crdi_q cfm) 40.3d3 W(crdi_q kjd_q) 3d3 W(cess_q jd) 3d3 W(cess_q cen_q) 57.3d3 W(cess_q mcbuf_q) 6d3 | added by DB to match the ECRG 10/20 revision. W(cess_q cfm) 45.3d3 W(cess_q kjd_q) 3d3 W(mcbuf_q jd) 16d3 | added by DB to match the ECRG 10/20 revision. W(mcbuf_q cen_q) 24d3 | added by DB to match the ECRG 10/20 revision. W(mcbuf_q cfm) 22d3 | added by DB to match the ECRG 10/20 revision. W(mcbuf_q kjd_q) 16d3 | added by DB to match the ECRG 10/20 revision. W(jd cen_q) 40d3 W(jd cfm) 40d3 W(jd kjd_q) 28d3 |changed from W(jd kjd_q) 10d3 by DB to match the ECRG 10/20 revision. W(cen_q cfm) 4d3 W(cen_q kjd_q) 40d3 W(cfm kjd_q) 40d3 end_excess_function 3 | number of sites in configurational entropy model (M1, M2, T) M1 | site name 6 1 1 | number of species, effective multiplicity, true multiplicity z(Fe) = 1 cfs_q z(Mg) = 1 di + 1 cen_q + 1/2 mcbuf_q + 1 cfm z(Fe3+) = 1 cess_q z(Ti) = 1/2 mcbuf_q z(Cr) = 1 crdi_q z(Al) = 1 cats + 1 jd + 1 kjd_q M2 | site name 5 1 1 | number of species, effective multiplicity, true multiplicity z(Na) = 1 jd z(K) = 1 kjd_q z(Mg) = 1 cen_q z(Fe) = 1 cfm + 1 cfs_q z(Ca) = 1 di + 1 cats + 1 cess_q + 1 crdi_q + 1 mcbuf_q T | site name 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 cats + 1/2 mcbuf_q + 1/2 crdi_q + 1/2 cess_q z(Si,T) = 1/2 cats + 1/2 mcbuf_q + 1/2 crdi_q + 1/2 cess_q + 1 cfs_q + 1 cen_q + 1 di + 1 cfm + 1 jd + 1 kjd_q [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(di) 1.2 alpha(cen_q) 1 alpha(cfs_q) 1 alpha(jd) 1.2 alpha(kjd_q) 1.2 alpha(cats) 1.9 alpha(mcbuf_q) 1.9 alpha(cess_q) 1.9 alpha(crdi_q) 1.9 alpha(cfm) 1 end_van_laar_sizes begin_dqf_corrections dqf(cen_q) = 3500 - 2 * T + 0.048 * P | cenjh dqf(cfs_q) = 2100 - 2 * T + 0.045 * P | cfsg dqf(cess_q) = -3450 | cessh dqf(mcbuf_q) = -16.2d3 - 1.2 * T - 0.005 * P | mcbuf dqf(crdi_q) = -4900 | crdih dqf(kjd_q) = 11.7d3 + 0.6 * P | kjdh end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Augite(G) => Green et al (JMG, 2016) Augite model. This model should not be used for "Na-rich" compositions. --------------------------------------------------------- Reformulated as a 2-polytope 688 format model. JADC, 9/19 The composition space of the model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3* on M1 C = Ca on M2 M = Mg, Fe A = Na on M2* *A is partially coupled to T by charge balance. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = [T][A,C] Polytope 2 = [T][M]+[M][C] --------------------------------------------------------- NOTE: to use this the following endmembers must be specified with make definitions in the thermodynamic data file cen_q cfs_q -------------------------------------------------------- M1 M2 T ____________________________________ Multiplicity 1 1 1/2 <- fake T multiplicity poytope II ____________________________________ Diopside Mg Ca Si Si cEnstatite Mg Mg Si Si Mats Al Mg AlSi AlSi dependent macm_d Fe3 Mg AlSi AlSi dependent Species: hed Fe Ca Si Si dependent cferrosilite Fe Fe Si Si fats Al Fe AlSi AlSi dependent facm Fe3 Fe AlSi AlSi dependent polytope I ____________________________________ Cats Al Ca AlSi AlSi cacm Fe3 Ca AlSi AlSi dependent Jadeite Al Na Si Si Acmite Fe3+ Na Si Si -------------------------------------------------------- Augite(G) abbreviation Cpx full_name clinopyroxene 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [T][N,C] 0 1 .1 0 TM+MC by difference | = [M,T][M,C] | ------------------------------------------- | Polytope 1 [T][N,C] 2 | number of simplices 2 2 | number of vertices on each simplex acm cacm_d jd cats | Simplex 1 X_Na 0 1 .1 0 | X(1,1) = Na/[Na + Ca] on M2 X_Ca1 by difference | Simplex 2 X_FeTs 0 1 .1 0 | X(2,1) = X(Fe3) on M1 X_AlTs by difference | ------------------------------------------- | Polytope 2 TM+MC 2 | number of simplices 2 4 | number of vertices on each simplex macm_d facm_d mats_d fats_d cen_q cfs_q di hed_d | Simplex 1 X_Mg .0 1 .1 0 | X(1,1) => bulk Mg/M X_Fe by difference | Simplex 2 X_MFeTs 0 1 .1 0 | X(2,1) - M-Fe3+ X_MAlTs 0 1 .1 0 | X(2,2) - M-Ts X_MMPx 0 1 .1 0 | X(2,3) - M-M X-MCPx by difference begin_ordered_endmembers ocats = 1 cats delta_g_of_ordering = -3800 + 2.8816 * T -0.01 * P | this undoes the DQF on cats fmc = 1/2 cen_q + 1/2 cfs_q delta_g_of_ordering = -4400 | DQF(fmc) - (DQF(cen_q) + DQF(cfs))/2 end_ordered_endmembers begin_dependent_endmembers hed_d = 1 di + 1 cfs_q - 1 fmc cacm_d = 1 acm + 1 cats - 1 jd mats_d = 1 cats + 1 cen_q - 1 di macm_d = 1 cats + 1 cen_q - 1 di + 1 acm - 1 jd fats_d = 1 cats - 1 di + 1 fmc facm_d = 1 cats - 1 di + 1 fmc + 1 acm - 1 jd end_dependent_endmembers begin_excess_function W(di cen_q) 29.8d3 -0.03 * p W(di cfs_q) 25.8d3 -0.03 * p W(di jd) 26d3 W(di acm) 21d3 W(di ocats) 12.3d3 -0.01 * p W(di cats) 12.3d3 -0.01 * p W(di fmc) 20.6d3 -0.03 * p W(cen_q cfs_q) 2.3d3 W(cen_q jd) 50d3 W(cen_q acm) 62d3 W(cen_q ocats) 45.7d3 -0.29 * p W(cen_q cats) 45.7d3 -0.29 * p W(cen_q fmc) 4d3 W(cfs_q jd) 60d3 W(cfs_q acm) 58d3 W(cfs_q ocats) 48d3 W(cfs_q cats) 48d3 W(cfs_q fmc) 3.5d3 W(jd acm) 5d3 W(jd ocats) 40d3 W(jd cats) 40d3 W(jd fmc) 40d3 W(acm ocats) 35d3 W(acm cats) 35d3 W(acm fmc) 60d3 W(ocats cats) 3.8d3 + 0.01 * p W(ocats fmc) 50d3 W(cats fmc) 50d3 end_excess_function 4 | 4 site entropy model (M1, M2, T1, T2) M1 | site name 4 1 1 | 4 species on m1, mult = 1 z(Fe) = 1 cfs_q z(Mg) = 1 di + 1 fmc + 1 cen_q z(Fe3+) = 1 acm z(Al) = 1 cats + 1 ocats + 1 jd M2 | site name 4 1 1 | 4 species on m2, mult. = 1 z(Na) = 1 acm + 1 jd z(Mg) = 1 cen_q z(Fe) = 1 fmc + 1 cfs_q z(Ca) = 1 di + 1 cats + 1 ocats T1 | site name 2 0.25 1 | 2 species on T1, effective mult. = 1/4 z(Al,T1) = 1/2 cats z(Si,T1) = 1/2 cats + 1 ocats + 1 fmc + 1 cfs_q + 1 acm + 1 jd + 1 cen_q + 1 di T2 | site name 2 0.25 1 | 2 species on T2, effective mult. = 1/4 z(Al,T2) = 1/2 cats + 1 ocats z(Si,T2) = 1/2 cats + 1 fmc + 1 cfs_q + 1 acm + 1 jd + 1 cen_q + 1 di [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(di) 1.2 alpha(cen_q) 1 alpha(cfs_q) 1 alpha(jd) 1.2 alpha(acm) 1.2 alpha(ocats) 1.9 alpha(cats) 1.9 alpha(fmc) 1 end_van_laar_sizes begin_dqf_corrections dqf(cen_q) = 3500 - 2 * T + 0.048 * P | cenjh dqf(cfs_q) = 2100 - 2 * T + 0.045 * P | cfs_q dqf(jd) = 2000 dqf(acm) = -5000 dqf(cats) = 3800 - 2.8816 * T + 0.01 * P end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Aqueous low density silicate fluid/melt - Holland, Green, & Powell; J Pet, 2018, 59:881 900. Intended to account for the solubility of silicates in water at P > 1 GPa. Use with caution. JADC 4/19 Modified to account for molecular CO2 by a CORK-like H2O-CO2 interaction term. This is not consistent with the a(H2O) ~ X^2 form of this model, not much else is either (i.e., it's a poor model for water-rich fluids). JADC, 9/5/2025 Corrections: woL_q DQF corrected, Debaditya Bandyopadhyay, 5/6/19. dqf(woL_q) = -0.22d3 -0.12*Pbar corrected to dqf(woL_q) = 0.22d3 -0.12*Pbar, JADC 11/27/2024 Add water vacancy site, JADC 11/27/2024 Aqfl(HGPK) abbreviation F full_name fluid 2 | model type: simplicial composition space 12 | number of endmembers foL_q faL_q jdL_q silL_q kjL_q woL_q ekL_q hmL_q tiL_q qL_q CO2 H2O 1 1 1 1 1 1 1 1 1 1 1 1 | endmember flags | non-linear subdivision is likely to be | preferable to cartesian for this model | see commentary in the header of this file for explanation of | endmember flags and subdivision schemes 0 1 0.1 0 | range and resolution of X(fo) 0 1 0.1 0 | range and resolution of X(fa) 0 1 0.1 0 | range and resolution of X(jd) 0 1 0.1 0 | range and resolution of X(sil) 0 1 0.1 0 | range and resolution of X(kj) 0 1 0.1 0 | range and resolution of X(wo) 0 1 0.1 0 | range and resolution of X(ek) 0 1 0.1 0 | range and resolution of X(hm) 0 1 0.1 0 | range and resolution of X(ti) 0 1 0.1 0 | range and resolution of X(q) 0 1 0.1 0 | range and resolution of X(CO2) begin_excess_function W(qL_q H2O) 59d3 0 -0.82 W(silL_q H2O) 57.6d3 0 -0.80 W(woL_q H2O) 72.2d3 0 -0.67 W(foL_q H2O) 71.7d3 0 -1.10 W(faL_q H2O) 71.7d3 0 -1.10 W(jdL_q H2O) 57d3 0 -0.79 W(hmL_q H2O) 73d3 0 -0.66 W(ekL_q H2O) 73d3 0 -0.66 W(tiL_q H2O) 75d3 0 -0.67 W(kjL_q H2O) 44.9d3 0 -1.19 W(CO2 H2O) 1d4 | closes regular solution solvus at ~602 K for x*ln(x) model [which this isn't] end_excess_function 2 | 2 sites 12 1 | molecular site z(fo) = 1 foL_q z(fa) = 1 faL_q z(jd) = 1 jdL_q z(silL_q) = 1 silL_q z(kj) = 1 kjL_q z(wo) = 1 woL_q z(ek) = 1 ekL_q z(hm) = 1 hmL_q z(ti) = 1 tiL_q z(q) = 1 qL_q z(CO2) = 1 CO2 2 1 | water-vacancy site z(H) = 1 H2O begin_dqf_corrections dqf(foL_q) = 8.59d3 - 0.136*Pbar dqf(faL_q) = 13.56d3 - 0.052*Pbar dqf(qL_q) = 2.1d3 - 0.051*Pbar dqf(hmL_q) = 4.05d3 - 0.077*Pbar dqf(kjL_q) = 12.88d3 - 0.227*Pbar dqf(jdL_q) = 12.32d3 - 0.099*Pbar dqf(ekL_q) = 24.75d3 + 0.245*Pbar dqf(tiL_q) = 5.6d3 - 0.489*Pbar dqf(silL_q) = 6.72d3 -0.313*Pbar dqf(woL_q) = 0.22d3 -0.12 *Pbar end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Generic Melt - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881-900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus. JADC 12/18 Modified after Heinrich & Connolly, GEOLOGY, 50:1101-1105. To account for molecular CO2 solubility based on the Morizet et al. (2010) synthesis for basaltic liquid. Use with caution. JADC 9/22. * DQF's changed from published values to the values in the TC files. JADC 4/23/19 * updated to 688 format, added previously missing van Laar size terms. Ben Klein, BC, 8/19/20. This model requires the following make definitions in the thermodynamic data file make_definitions section: foL_q = 2 foL faL_q = 2 faL qL_q = 4 qL jdL_q = 1 abL - 1 qL hmL_q = 1/2 hemL ekL_q = 1/2 eskL tiL_q = 1 ruL kjL = 1 kspL - 1 qL to avoid cluttering thermodynamic data files with the new plague of TC DQF "corrections", the following simple DQF's are specified in the make_definition section at the end of this model: dqf(silL_q) = 6.35d3 -0.320*Pbar updated to DQF 6.20 -0.318 dqf(woL_q) = -0.18d3 -0.118*Pbar updated to DQF -0.45 -0.114 dqf(wL_q) = 3.24d3 -3.9*TK 0.00085*Pbar dqf(CO2) = 54355. |based on Morizet 2010 for basalt data & rhyolite curve in Fig. 11, JADC-CAH apr22 melt(HGPH) abbreviation Melt full_name liquid 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 12 | number of vertices (endmembers) on each simplex foL_q faL_q jdL_q silL_q kjL_q woL_q ekL_q hmL_q tiL_q qL_q CO2 wL_q 0 1 .1 0 | range and resolution of X(fo) 0 1 .1 0 | range and resolution of X(fa) 0 1 .1 0 | range and resolution of X(jd) 0 1 .1 0 | range and resolution of X(sil) 0 1 .1 0 | range and resolution of X(kj) 0 1 .1 0 | range and resolution of X(wo) 0 1 .1 0 | range and resolution of X(ek) 0 1 .1 0 | range and resolution of X(hm) 0 1 .1 0 | range and resolution of X(ti) 0 1 .1 0 | range and resolution of X(q) 0 1 .1 0 | range and resolution of X(CO2) begin_ordered_endmembers | H(ctL) = -108.3d3 +55*TK +0.053*Pbar - dqfwoL_q - dqfsil + dqfq/4 ctL = 1 woL_q + 1 silL_q - 1/4 qL_q Delta(enthalpy) = -113995 + 55*TK+ .47025*P end_ordered_endmembers begin_excess_function W(qL_q silL_q) 9.5d3 -0.1 * Pbar W(qL_q woL_q) -10.3d3 W(qL_q foL_q) -26.5d3 -3.12 * Pbar W(qL_q faL_q) -12d3 -0.55 * Pbar W(qL_q jdL_q) -15.1d3 -0.13 * Pbar W(qL_q hmL_q) 20d3 W(qL_q tiL_q) 24.6d3 W(qL_q kjL_q) -17.8d3 -0.05 * Pbar W(qL_q ctL) -14.6d3 W(qL_q wL_q) 17.8d3 -0.61 * Pbar W(silL_q woL_q) -26.5d3 0.85 * Pbar W(silL_q foL_q) 2.2d3 W(silL_q faL_q) 2.5d3 W(silL_q jdL_q) 16.8d3 W(silL_q hmL_q) -5d3 W(silL_q tiL_q) 15.2d3 -0.04 * Pbar W(silL_q kjL_q) 7d3 W(silL_q ctL) 4d3 W(silL_q wL_q) 23.7d3 -0.94 * Pbar W(woL_q foL_q) 25.5d3 0.11 * Pbar W(woL_q faL_q) 14d3 W(woL_q jdL_q) -1.2d3 W(woL_q tiL_q) 18d3 W(woL_q kjL_q) -1.1d3 W(woL_q ctL) 9.5d3 W(woL_q wL_q) 40.3d3 -0.86 * Pbar W(foL_q faL_q) 18d3 W(foL_q jdL_q) 1.5d3 W(foL_q tiL_q) 7.5d3 W(foL_q kjL_q) 3d3 W(foL_q ctL) -5.6d3 W(foL_q wL_q) 9.4d3 -1.58 * Pbar W(faL_q jdL_q) 7.5d3 -0.05 * Pbar W(faL_q hmL_q) -30d3 W(faL_q tiL_q) 6.7d3 W(faL_q kjL_q) 10d3 W(faL_q ctL) -6.5d3 W(faL_q wL_q) 9.2d3 -1.58 * Pbar W(jdL_q hmL_q) 10d3 W(jdL_q tiL_q) 16.5d3 0.14 * Pbar W(jdL_q kjL_q) -5.9d3 W(jdL_q ctL) 7.6d3 W(jdL_q wL_q) -8.3d3 -0.06 * Pbar W(hmL_q kjL_q) 10d3 W(hmL_q wL_q) 60d3 -0.66 * Pbar W(ekL_q wL_q) 30d3 -0.66 * Pbar W(tiL_q kjL_q) 9d3 W(tiL_q wL_q) 30d3 -0.6 * Pbar W(kjL_q ctL) -5.6d3 W(kjL_q wL_q) -0.1d3 0.22 * Pbar W(ctL wL_q) 17.3d3 0.05 * Pbar end_excess_function 3 OH 2 2 1 | water-vacancy site, multiplicity 2 z(H) = 1 wL_q z(Vac) = 1 qL_q + 1 silL_q + 1 woL_q + 1 foL_q + 1 faL_q + 1 jdL_q + 1 hmL_q + 1 ekL_q + 1 tiL_q + 1 kjL_q + 1 ctL + 1 CO2 M 4 0 0 | Temkin M-site n(Mg) = 4 foL_q n(Fe) = 4 faL_q n(Ca) = 1 woL_q n(Al) = 1 silL_q F 11 0 0 | Temkin F-site, decomposing "AlSi2" into k/na-jd here eliminates the necessity of the A-site n(jd) = 1 jdL_q n(kj) = 1 kjL_q n(alsi) = 1 silL_q n(si0) = 1 woL_q n(ol) = 1 faL_q + 1 foL_q n(q) = 1 qL_q n(ek) = 1 ekL_q n(ti) = 1 tiL_q n(ct) = 1 ctL n(hm) = 1 hmL_q n(CO2) = 1 CO2 none | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(qL_q) = 1 alpha(silL_q) = 1.2 alpha(woL_q) = 1.4 alpha(foL_q) = 2.4 alpha(faL_q) = 1 alpha(jdL_q) = 1.2 alpha(hmL_q) = 1 alpha(ekL_q) = 1 alpha(tiL_q) = 1 alpha(kjL_q) = 1 alpha(ctL) = 1 alpha(CO2) = 1 alpha(wL_q) = 1 end_van_laar_sizes begin_dqf_corrections DQF(foL_q) = 8.67d3 - 0.131*Pbar DQF(faL_q) = 13.7d3 - 0.055*Pbar DQF(qL_q) = 0.22d3 - 0.059*Pbar DQF(kjL_q) = 11.98d3 - 0.210*Pbar DQF(hmL_q) = 3.3d3 - 0.032 *Pbar DQF(jdL_q) = 12.19d3 - 0.089*Pbar DQF(ekL_q) = 24.85d3 + 0.245*Pbar DQF(tiL_q) = 5.58d3 - 0.489*Pbar dqf(silL_q) = 6.2d3 - 0.318 *Pbar dqf(woL_q) = -0.45d3 - 0.114 *Pbar dqf(wL_q) = 3.2d3 -3.9*TK + 0.00087 *Pbar dqf(CO2) = 54355. |based on Morizet 2010 for basalt data & rhyolite curve in Fig. 11 ****ADDITION JADC-CAH apr22 |dqf(CO2) = 46994. |based on Morizet 2010 for the andesite curve in their Fig. 11 ****ADDITION JADC-CAH apr22 end_dqf_corrections begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. This | may be useful for testing purposes for systems in which the | the endmember compositions are not expected to be stable, i.e., | the stability of pure wL_q would alert the user to unexpected | behaviour. if/when endmember compositions of the solution are | expected, those endmembers should be removed from the list of | flagged endmembers. foL_q faL_q jdL_q silL_q kjL_q woL_q ekL_q hmL_q tiL_q qL_q wL_q CO2 end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Tonalitic melt, Green et al., JMG, 2016. JADC 9/16 * reformulated as standard O/D model, 12/18, JADC * converted to 688 format, 1/8/20, JADC * the current solver for this model (gpmelt) precludes stability of antiordered states, 6/27/20, JADC. This model requires the following make definitions in the thermodynamic data file make_definitions section: foL_q = 2 foL faL_q = 2 faL qL_q = 4 qL to avoid cluttering thermodynamic data files with the new plague of TC DQF "corrections", the following simple DQF's are specified in the make_definition section at the end of this model: dqf(silL_q) = -7800 dqf(woL_q) = 1300 melt(G) abbreviation Melt full_name liquid 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 8 | number of vertices (endmembers) on each simplex foL_q faL_q abL silL_q kspL woL_q qL_q wL_q 0 1 .1 0 | range and resolution of X(fo), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(fa), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(ab), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(sil), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(ksp), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(wo), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(q), 0 => cartesian subdivision begin_ordered_endmembers | H(oanL) = -46.5d3 - 0.25 * P - dqf(woL_q) 1300 - dqf(silL_q) -7800 oanL = 1 woL_q + 1 silL_q Delta(enthalpy) = -40000 - 0.25 * P end_ordered_endmembers begin_excess_function W(qL_q abL) = 12d3 - 0.4 * P W(qL_q kspL) = -2d3 - .5 * P W(qL_q woL_q) = -5d3 W(qL_q foL_q) = 42d3 + 1 * P W(qL_q wL_q) = 18.1d3 - 0.68 * P W(qL_q oanL) = -29.5d3 - .1 * P W(abL kspL) = -6d3 + 3 * P W(abL woL_q) = -12d3 W(abL silL_q) = 10d3 W(abL faL_q) = -30d3 + 0.8 * P W(abL foL_q) = -47.3d3 + 0.3 * P W(abL wL_q) = -4.4d3 - 0.17 * P W(abL oanL) = 8.6d3 + 0.4 * P W(kspL woL_q) = -13d3 W(kspL faL_q) = -11.3d3 W(kspL foL_q) = 6.8d3 W(kspL wL_q) = 10.4d3 - 0.39 * P W(kspL oanL) = -16d3 - 0.25 * P W(woL_q silL_q) = -1.6d3 W(woL_q faL_q) = 6.5d3 W(woL_q foL_q) = 4d3 W(woL_q wL_q) = 21d3 W(woL_q oanL) = 3.5d3 W(silL_q faL_q) = 12d3 W(silL_q foL_q) = 12d3 W(silL_q wL_q) = 11d3 - .5 * P W(silL_q oanL) = 6.4d3 W(faL_q foL_q) = 18d3 W(faL_q wL_q) = 29d3 W(faL_q oanL) = -43.5d3 - 0.95 * P W(foL_q wL_q) = 29d3 - .5 * P W(foL_q oanL) = -26d3 - 0.6 * P W(wL_q oanL) = 9.75d3 - .5 * P end_excess_function 3 | Configurational entropy: two non-temkin sites (Water, Melt) | and one temkin site (olvine). HP assume a fsp = ab + or "molecule" | with mixing on a temkin M site, but the math works out the same as | the endmembers are treated as separate endmembers and the M site | dropped. OH 2 1 1 | water-vacancy site z(OH) = 1 wL_q z(Vac) = 1 foL_q + 1 faL_q + 1 abL + 1 silL_q + 1 kspL + 1 woL_q + 1 qL_q + 1 oanL Ol 2 0 0 | Temkin olivine site n(Mg) = 4 foL_q n(Fe) = 4 faL_q M 8 0 0 | melt site, pseudo-Temkin z(q) = 1 qL_q z(ksp) = 1 kspL z(ab) = 1 abL z(sil) = 1 silL_q z(an) = 1 oanL z(wo) = 1 woL_q z(Ol) = 1 faL_q + 1 foL_q | this term was not counted prior to dec 12, 2018. z(OH) = 1 wL_q none | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(wL_q) = 0 dqf(silL_q) = -7800 dqf(woL_q) = 1300 dqf(foL_q) = -4d3 dqf(faL_q) = -8.2d3 - 1.4 * P dqf(qL_q) = 0 end_dqf_corrections begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. foL_q faL_q abL silL_q kspL woL_q qL_q wL_q end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model CHLORITE, White et al JMG 32:261-286, 2014 NOTES: * This is an oddball model because Al disorder is implicit while Fe-Mg disorder is explicit. The model should be reformulated so that clin is an explicit ordered species. * This model will only function in the Al-free and Mg-free limits if Al and Mg are retained as thermodynamic components. JADC 4/14 Mn endmembers entered, P.-H. Trapy, 5/16 Site limits corrected, JADC, 5/16 M1 M2+M3 M4 T2 ________________________________ Mutliplicity 1 4 1 2 ________________________________ 1 fames Al Fe Al Al_Al dependent 2 fafchl Fe Fe Fe Si_Si dependent 3 ff3cli Fe Fe Fe3+ Al_Si dependent 4 daph Fe Fe Al Al_Si 5 f3cl_q Mg Mg Fe3+ Al_Si 6 ames Al Mg Al Al_Al 7 afchl Mg Mg Mg Si_Si 8 clin Mg Mg Al Al_Si 9 mnchl Mn Mn Al Al_Si _______________________________ 13 och1 Mg Fe Fe Si_Si ordered 14 och2 Fe Mg Mg Si_Si ordered Chl(W) abbreviation Chl full_name chlorite 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [Mn][Mn] 0 .5 .1 0 | X(1) is [Mn][Mn] [M][M,T] by difference | ---------------------------- | Polytope 1 - 1 simplex 1 | number of simplices 1 | number of vertices on each simplex mnchl | endmembers on the vertices | ---------------------------- | Polytope 2 - 4x2 simplices 2 | number of simplices 4 2 | number of vertices on each simplex | endmembers on the vertices fames fafchl ff3cli daph ames afchl f3cl_q clin | subdivision model for (quaternary) chemical mixing space X_Mames 0 1 .1 0 | range and resolution of X(M-ames), imod = 0 -> cartesian subdivision X_Mafchl 0 1 .1 0 | range and resolution of X(M-afchl), imod = 0 -> cartesian subdivision X_Mf3cl_q 0 1 .1 0 | range and resolution of X(M-f3cl_q), imod = 0 -> cartesian subdivision X_Mclin by difference | subdivision model for (ternary) chemical mixing space X_Mg 0 1 .1 0 | range and resolution of X(Fe), imod = 0 -> cartesian subdivision X_Fe by difference begin_ordered_endmembers och1 = 1 afchl - 1 clin + 1 daph delta_g_of_ordering = 3d3 och2 = 1 afchl - 1/5 clin + 1/5 daph delta_g_of_ordering = 2.4d3 end_ordered_endmembers begin_dependent_endmembers fames = 1 ames + 1 daph - 1 och2 + 1 afchl - 1 clin fafchl = 1 och1 + 1 och2 - 1 afchl ff3cli = 1 f3cl_q + 1 daph - 1 clin end_dependent_endmembers begin_excess_function W(clin afchl) 17d3 W(clin ames) 17d3 W(clin daph) 20d3 W(clin och1) 30d3 W(clin och2) 21d3 W(clin f3cl_q) 2d3 W(clin mnchl) 6d3 | changed from 15d3 DEK 02-JUN-2021 W(afchl ames) 16d3 W(afchl daph) 37d3 W(afchl och1) 20d3 W(afchl och2) 4d3 W(afchl f3cl_q) 15d3 W(afchl mnchl) 23d3 | changed from 32d3 DEK 02-JUN-2021 W(ames daph) 30d3 W(ames och1) 29d3 W(ames och2) 13d3 W(ames f3cl_q) 19d3 W(ames mnchl) 17d3 | changed from 26d3 DEK 02-JUN-2021 W(daph och1) 18d3 W(daph och2) 33d3 W(daph f3cl_q) 22d3 W(daph mnchl) 4d3 | changed from 10d3 DEK 02-JUN-2021 W(och1 och2) 24d3 W(och1 f3cl_q) 28.6d3 W(och1 mnchl) 19d3 | changed from 25d3 DEK 02-JUN-2021 W(och2 f3cl_q) 19d3 W(och2 mnchl) 22d3 | changed from 31d3 DEK 02-JUN-2021 W(f3cl_q mnchl) 8d3 | changed from 17d3 DEK 02-JUN-2021 end_excess_function 4 | 4 site configurational entropy model: M1 4 1 1 | 4 species on 1 M1 site z(Al,M1) = 1 ames z(Fe,M1) = 1 daph + 1 och2 z(Mn,M1) = 1 mnchl z(Mg,M1) = 1 f3cl_q + 1 afchl + 1 clin + 1 och1 M23 3 4 4 | 3 species on 4 M2+M3 sites z(Fe,M23) = 1 daph + 1 och1 z(Mn,M23) = 1 mnchl z(Mg,M23) = 1 f3cl_q + 1 ames + 1 afchl + 1 clin + 1 och2 M4 4 1 1 | 4 species on 1 M4 site z(Fe3,M4) = 1 f3cl_q z(Fe,M4) = 1 och1 z(Mg,M4) = 1 afchl + 1 och2 z(Al,M4) = 1 daph + 1 clin + 1 ames + 1 mnchl T2 2 2 2 | 2 species on 2 T2 sites z(Al,T2) = 1 ames + 1/2 clin + 1/2 daph + 1/2 f3cl_q + 1/2 mnchl z(Si,T2) = 1 afchl + 1 och1 + 1 och2 + 1/2 clin + 1/2 daph + 1/2 f3cl_q + 1/2 mnchl [Si2O10(OH)8] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(mnchl) = -5670 | changed from -13030 DEK 02-JUN-2021 dqf(f3cl_q) = 2000 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model holland and powell '11 non-ideal cz-fep solution entered by Pierre-Henri Trapy, May 16, 2016 reformulated as an implicit O/D model to eliminate need to adjust the enthalpy of ordering with each revision of the TC data base. DEK/JADC, 8/10/2021. M1 M3 _____________ Mutliplicity 1 1 _____________ 1 cz Al Al Species: 2 fep Fe Fe _____________ Ordered Cpd: 3 ep Al Fe Ep(HP11) abbreviation Ep full_name epidote 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 3 | number of vertices on each simplex cz fep ep | endmember names 0 1 .1 0 | X(cz) 0 1 .1 0 | X(fep) begin_excess_function w(cz fep) 3d3 w(fep ep) 1d3 w(cz ep) 1d3 end_excess_function 2 | 2 site (M1, M3) configurational entropy model M1 | site name 2 1 1 | 2 species on M1, 1 site per formula unit. z(Fe,M1) = 1 fep z(Al,M1) = 1 cz + 1 ep M3 | site name 2 1 1 | 2 species on M3, 1 site per formula unit. z(Al,M3) = 1 cz z(Fe,M3) = 1 fep + 1 ep [Ca2Si3O12(OH)] | formula suffix, enter "none" for no suffix. end_of_model -------------------------------------------------------- begin_model Cpx(JH) => Jennings and Holland, J Pet, 56:869-892, 2015, intended for pressures of 0-60 GPa --------------------------------------------------------- Reformulated as a 2-polytope 688 format model. JADC, 9/19 The composition space of the published model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3, Cr on M1 C = Ca on M2 M = Mg, Fe A = Na on M2* *A is partially coupled to T by charge balance. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = [T][A,C] Polytope 2 = [T,M][M]+[M][C] --------------------------------------------------------- NOTES: The Cr/Fe3+-free version of this model is from Holland et al., 2013 or Green et al., 2012? Pierre Bouilhol & JADC, 12/16/15 Changed to true T site multiplicity, Oliver Shorttle, 3/16 Changed to fake T site multiplicity (1/4) with true Al site fraction, JADC 6/16 Changed to fake T site multiplicity (1/2) with true Al site fraction, Bob Myhill, 2/18 Formatted as prismatic model w/o Cr & Fe3+. Bob Myhill, 2/18 Corrected cess_q composition assumed in the model (mess was assumed previously). JADC, 6/18 Dual polytope 688 format, JADC, 9/19. M1 M2 T ____________________________________ Multiplicity 1 1 1/2 <- fake T multiplicity ____________________________________ Polytope 2 species: di Mg Ca Si independent cen_q Mg Mg Si independent mats_d Al Mg AlSi dependent cren_d Cr Mg AlSi dependent mcess_d Fe3+ Mg AlSi dependent hed_d Fe Ca Si dependent cfs Fe Fe Si independent fats_d Al Fe AlSi dependent crfs_d Cr Fe AlSi dependent fcess_d Fe3+ Fe AlSi dependent ____________________________________ Ppolytope 1 species: cats Al Ca AlSi independent crdi_q Cr Ca AlSi independent cess_q Fe3+ Ca AlSi independent jd Al Na Si independent crjd Cr Na Si dependent ness Fe3+ Na Si dependent ___________________________________ --------------------------------------------------------- Cpx(JH) abbreviation Cpx full_name clinopyroxene 688 | model type: 688 format standard model 2 | number of polytopes, subvision range for the composite space | polytope names and composite composition space subdivision schemes [T][N,C] 0 .25 .1 0 [M,T][M,C] by difference | ------------------------------------------- | Polytope 1 [T][N,C] 2 | number of simplices 2 3 | number of vertices on each simplex crjd_d crdi_q ness_d cess_q jd cats | Simplex 1 X_Na 0 1 .1 0 | X(1,1) = Na/[Na + Ca] on M2 X_Ca by difference | Simplex 2 X_CrTs1 0 1 .1 0 | X(2,1) = X(Cr) on M1 X_Fe3Ts1 0 1 .1 0 | X(2,2) = X(Fe3) on M1 X_AlTs1 by difference | ------------------------------------------- | Polytope 2 [M,T][M,C] 2 | number of simplices 2 5 | number of vertices on each simplex cren_d crfs_d mess_d fess_d mats_d fats_d cen_q cfs_q di hed_d | Simplex 1 X_Mg 0 1 .1 0 | X(1,1) => bulk Mg/M X_Fe by difference | Simplex 2 X_CrTs 0 .1 .1 0 | X(2,1) - M-Cr X_Fe3Ts 0 .2 .1 0 | X(2,2) - M-Fe3+ X_MTs 0 .2 .1 0 | X(2,3) - M-Ts X_MM 0 .9 .1 0 | X(2,4) - M-M X_MCa by difference begin_ordered_endmembers cfm = 1/2 cen_q + 1/2 cfs_q delta_g_of_ordering = -6.65d3 2.5 - 0.039 | DQF(cfm) - (DQF(cen_q) + DQF(cfs_q))/2 | -3 - (3.5 - 2T + 0.048P + 3.8 - 3T + 0.03)/2 | = -6.65 + 2.5T - 0.039P end_ordered_endmembers begin_dependent_endmembers crjd_d = 1 crdi_q + 1 jd - 1 cats cren_d = 1 cen_q + 1 crdi_q - 1 di crfs_d = 1 cfm + 1 crdi_q - 1 di ness_d = 1 cess_q + 1 jd - 1 cats mess_d = 1 cess_q - 1 di + 1 cen_q fess_d = 1 cess_q - 1 di + 1 cfm hed_d = 1 di + 1 cfs_q - 1 cfm mats_d = 1 cats + 1 cen_q - 1 di fats_d = 1 cats - 1 di + 1 cfm end_dependent_endmembers begin_excess_function W(di cfs_q) 20d3 W(di cats) 12.3d3 -0.1 * P W(di jd) 26d3 W(di cen_q) 29.8d3 -0.03 * P W(di cfm) 18d3 W(cfs_q cats) 25d3 -0.1 * P W(cfs_q jd) 36d3 W(cfs_q cen_q) 7d3 W(cfs_q cfm) 4d3 W(cats jd) 6d3 W(cats cen_q) 45.7d3 -0.29 * P W(cats cfm) 27d3 -0.1 * P W(jd cen_q) 40d3 W(jd cfm) 40d3 W(cen_q cfm) 4d3 W(di crdi_q) 8d3 W(di cess_q) 8d3 W(cfs_q crdi_q) 34d3 W(cfs_q cess_q) 34d3 W(cats crdi_q) 2d3 W(cats cess_q) 2d3 W(crdi_q cess_q) 2d3 W(crdi_q jd) 3d3 W(crdi_q cen_q) 48d3 W(crdi_q cfm) 36d3 W(cess_q jd) 3d3 W(cess_q cen_q) 58d3 W(cess_q cfm) 36d3 end_excess_function 3 | 3 site entropy model (M1, M2, T) M1 | site name 5 1 1 | number of species, effective multiplicity, true multiplicity z(Fe,m1) = 1 cfs_q z(Al,m1) = 1 cats + 1 jd z(Fe3,m1) = 1 cess_q z(Cr,m1) = 1 crdi_q z(Mg,M1) = 1 di + 1 cen_q + 1 cfm M2 | site name 4 1 1 | number of species, effective multiplicity, true multiplicity z(Na,m2) = 1 jd z(Mg,m2) = 1 cen_q z(Fe,m2) = 1 cfs_q + 1 cfm z(Ca,M2) = 1 di + 1 crdi_q + 1 cess_q + 1 cats T | site name 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 cats + 1/2 cess_q + 1/2 crdi_q z(Si,T) = 1/2 cats + 1/2 cess_q + 1/2 crdi_q + 1 cfs_q + 1 cfm + 1 jd + 1 cen_q + 1 di [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(di) = 1.2 alpha(cfs_q) = 1 alpha(cats) = 1.9 alpha(crdi_q) = 1.9 alpha(cess_q) = 1.9 alpha(jd) = 1.2 alpha(cen_q) = 1 alpha(cfm) = 1 end_van_laar_sizes begin_dqf_corrections dqf(cen_q) 3500 - 2 * T + 0.048 * P | cenjh dqf(cfs_q) 3800 - 3 * T + 0.03 * P | cfs dqf(cess_q) -6000 | cess dqf(crdi_q) -3000 | crdi end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Jennings and Holland, J Pet, 56:869-892, 2015 Mantle Melt Model, intended for pressures of 0-60 GPa Pierre Bouilhol & JADC, 12/16/15 Temkin M-site added 3/16, Oliver Shorttle Melt(JH) abbreviation Melt full_name liquid 2 8 jdL_q ctL_q fo1L_q fa1L_q diL_q hmL_q ekL_q q2L_q 0 0 0 0 0 0 0 0 0 1 .1 0 | jd 0 1 .1 0 | ct 0 1 .1 0 | fo 0 1 .1 0 | fa 0 1 .1 0 | di 0 1 .1 0 | hm 0 1 .1 0 | ek begin_excess_function W(q2L_q diL_q) 26d3 - .4 * P W(q2L_q jdL_q) -10d3 W(q2L_q ctL_q) -10d3 W(q2L_q fo1L_q) -25d3 - .1 * P W(q2L_q fa1L_q) -7d3 W(diL_q ctL_q) -2d3 W(diL_q fo1L_q) 24d3 + .2 * P W(diL_q fa1L_q) 17d3 W(ctL_q fo1L_q) -1d3 + .1 * P W(fo1L_q fa1L_q) 9d3 end_excess_function 2 only 2 sites contribute to configurational entropy 3 0 Temkin M-site has Fe, Mg, Ca n(Ca) = 1 diL_q n(Mg) = 1 diL_q + 2 fo1L_q n(Fe) = 2 fa1L_q 7 1 F site z(jd) = 1 jdL_q z(di) = 1 diL_q z(ct) = 1 ctL_q z(ol) = 1 fo1L_q + 1 fa1L_q z(q) = 1 q2L_q z(ek) = 1 ekL_q begin_dqf_corrections dqf(hmL_q) = 0 + 0.20 * p_bar dqf(jdL_q) = 18.7d3 - 0.16 * p_bar dqf(ekL_q) = 21d3 dqf(q2L_q) = 4d3 - 0.43 * p_bar dqf(fo1L_q) = 2d3 - 0.03 * p_bar dqf(fa1L_q) = -2.5d3 + 0.18 * p_bar dqf(diL_q) = -6.7d3 + 0.17 * p_bar dqf(ctL_q) = 2.5d3 + 0.29 * p_bar end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Jennings and Holland, J Pet, 56:869-892, 2015 Mantle Melt Model, intended for pressures of 0-60 GPa Pierre Bouilhol & JADC, 12/16/15 O(JH) abbreviation Ol full_name olivine 2 model type: simplicial composition space 2 2 endmembers fo fa 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision begin_excess_function W(fo fa) 9000 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(fe) = 1 fa end_of_model -------------------------------------------------------- begin_model Jennings and Holland, J Pet, 56:869-892, 2015 Mantle Melt Model, intended for pressures of 0-60 GPa Pierre Bouilhol & JADC, 12/16/15 Removed extraneous line in Sp(JH) excess definitions [which read w(sp herc) 7d2 0 0] Added z(al) = 2/3 sp + 2/3 herc definition to Sp(JH) configurational entropy site fractions. Oliver Shorttle, 3/18/16 Sp(JH) abbreviation Sp full_name spinel 2 | model type: simplicial composition space 4 | 4 endmembers sp mt picr herc 0 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(sp herc) 4d3 W(sp mt) 56d3 W(sp picr) 39d3 W(herc mt) 32d3 W(herc picr) 27d3 W(mt picr) 36d3 end_excess_function 1 5 3. cations disordered across all sites z(al) = 2/3 sp + 2/3 herc z(mg) = 1/3 sp + 1/3 picr z(fe3) = 2/3 mt z(cr) = 2/3 picr end_of_model -------------------------------------------------------- begin_model Jennings and Holland, J Pet, 56:869-892, 2015 Mantle Melt Model, intended for pressures of 0-60 GPa Pierre Bouilhol & JADC, 12/16/15 Pl(JH) abbreviation Pl full_name binary-feldspar 2 | model type: simplicial composition space 2 | 2 endmembers abh an 1 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function w(abh an) 22.4d3 end_excess_function 2 | 2 sites (O, T) kerrick and darkens Al-avoidance model: 2 1 | 2 species on O site, multiplicity = 1. z(Na) = 1 abh 2 2. | 2 species on T, mutiplicity = 2. z(Al) = 1/2 + 1/2 an begin_van_laar_sizes alpha(abh) 1 alpha(an) .39 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model Jennings and Holland, J Pet, 56:869-892, 2015 Mantle Melt Model, intended for pressures of 0-60 GPa Pierre Bouilhol & JADC, 12/16/15 X Y _____________ Mutliplicity 3 2 _____________ Dependent: fkho_d Fe Fe3+ Dependent: kho_d Mg Fe3+ andr Ca Fe3+ Dependent: fkno_d Fe Cr knor Mg Cr Dependent: ckno_d Ca Cr alm Fe Al py Mg Al gr Ca Al Grt(JH) abbreviation Gt full_name garnet 7 2 3 3 alm py gr fkno_d knor ckno_d fkho_d kho_d andr 4 fkno_d = 1 alm + 1 knor - 1 py ckno_d = 1 gr + 1 knor - 1 py fkho_d = 1 alm + 1 andr - 1 gr kho_d = 1 py + 1 andr - 1 gr 0 0 0 0 0 0 0 0 0 0 1 .1 0 0 1 .1 0 0 1 .1 0 0 1 .1 0 begin_excess_function W(py alm) 4d3 .1 * P W(py gr) 35d3 .1 * P W(py andr) 91d3 1.7 * T .032 * P W(py knor) 2d3 W(alm gr) 4d3 .1 * P W(alm andr) 60d3 1.7 * T .032 * P W(alm knor) 6d3 .01 * P W(gr andr) 2d3 W(gr knor) 47d3 -33.8 * T .221 * P W(andr knor) 101d3 -32.1 * T .153 * P end_excess_function 2 3 3. z(m1,fe) = 1 alm z(m1,mg) = 1 py + 1 knor 3 2. z(m2,cr) = 1 knor z(m2,fe3) = 1 andr unbounded_composition end_of_model -------------------------------------------------------- begin_model Jennings and Holland, J Pet, 56:869-892, 2015 originally from Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 Mantle Melt Model, intended for pressures of 0-60 GPa Pierre Bouilhol & JADC, 12/16/15 T-site mixing added 3/16, O. Shorttle. T-site mixing corrected for a false multiplicity of 1/4, with the fully disordered tetrahedral Al site fraction. JADC, 6/10/15 Changed fake T site multiplicity from 1/4 to 1/2 with true Al site fraction. Bob Myhill, 2/18 NOTE: to use this the following endmembers must be specified with make definitions in the thermodynamic data file: M1 M2 T _____________________ Mutliplicity 1 1 1/2 ______________________ en Mg Mg SiSi mgts Al Mg SiAl oen Fe3+ Mg SiAl cren_q Cr Mg SiAl Species: fs Fe Fe SiSi tsfs Al Fe SiAl dependent ofs Fe3+ Fe SiAl dependent crfs Cr Fe SiAl dependent Species: odi_q Mg Ca SiSi tsdi Al Ca SiAl dependent odif Fe3+ Ca SiAl dependent odicr Cr Ca SiAl dependent _____________ Internal: opx Mg Fe SiSi Opx(JH) abbreviation Opx full_name orthopyroxene 8 | prismatic composition space, order-disorder 2 | 2 independent composition spaces 3 4 | 3 vertices on first space, 4 on second | endmembers: odif oen ofs odicr cren_q crfs tsdi mgts tsfs odi_q en fs 1 | ordered species definition opx = 1/2 en + 1/2 fs Delta(enthalpy) = -6d3 6 | 6 dependent endmembers tsfs = 1 mgts + 1 opx - 1 en tsdi = 1 odi_q + 1 mgts - 1 en ofs = 1 oen + 1 opx - 1 en odif = 1 odi_q + 1 oen - 1 en crfs = 1 cren_q + 1 opx - 1 en odicr = 1 odi_q + 1 cren_q - 1 en 0 0 0 0 0 0 0 0 0 0 0 0 | endmember flags, indicate if endmember is part of the solution (i.e. iend = 0). | subdivision model for (ternary) site 1 (M1): 0 1 .1 0 0 1 .1 0 | subdivision model for (quaternary) site 2 (M2): 0 .1 .1 0 0 .1 .1 0 0 .2 .1 0 begin_excess_function W(en fs) 5.2d3 W(en opx) 4d3 W(en odi_q ) 32.2d3 0.12 * P_bar W(en mgts) 13d3 -0.15 * P_bar W(en cren_q) 8d3 W(en oen) 8d3 W(fs opx) 4d3 W(fs odi_q ) 24d3 W(fs mgts) 7d3 -0.15 * P_bar W(fs cren_q) 10d3 W(fs oen) 10d3 W(opx odi_q ) 18d3 W(opx mgts) 2d3 -0.15 * P_bar W(opx cren_q) 12d3 W(opx oen) 12d3 W(odi_q mgts) 75.4d3 -0.94 * P_bar W(odi_q cren_q) 30d3 W(odi_q oen) 30d3 W(mgts cren_q) 2d3 W(mgts oen) 2d3 W(cren_q oen) 2d3 end_excess_function 3 | 2 site (M1, M2, T1) configurational entropy model 5 1 | 5 species on M1, 1 site per formula unit. z(m1,fe) = 1 fs z(m1,cr) = 1 cren_q z(m1,fe3+) = 1 oen z(m1,al) = 1 mgts 3 1 | 3 species on M2, 1 site per formula unit. OS_EDIT-3/16 changed comment [2 species...] z(m2,ca) = 1 odi_q z(m2,fe) = 1 fs + 1 opx 2 .5 | 2 species on "1/2" T sites with true site fraction. z(t,al) = 1/2 mgts + 1/2 cren_q + 1/2 oen begin_van_laar_sizes alpha(en) 1 alpha(fs) 1 alpha(opx) 1 alpha(odi_q ) 1.2 alpha(mgts) 1 alpha(cren_q) 1 alpha(oen) 1 end_van_laar_sizes begin_dqf_corrections dqf(cren_q) = 3000 | crenJH dqf(odi_q) = 2800 + 0.005 * P | = odi, it's unclear why this model is retroactively modified. end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Ca-Fe2+-Mg-Al-Fe3+ Garnet, White et al., JMG 32:261-286, 2014. * In calculations that use this model, the andradite endmember ("andr") in the Holland and Powell data base must be excluded. This model also requires the following make definition for khoharite in the thermodynamic data file: * Mn added as in Gt(WPPH). JADC 4/14 * Mn excess energy updated, alpha(kho_q & spss) reduced to 1(formerly 2.7). Felix Gervais, 1/28/2015. X Y _____________ Mutliplicity 3 2 _____________ Dependent: fkho_i Fe Fe3+ kho_q Mg Fe3+ Dependent: fmn_i Mn Fe3+ Dependent: andr_i Ca Fe3+ spss Mn Al alm Fe Al py Mg Al gr Ca Al Gt(W) abbreviation Gt full_name garnet 7 | model type: prismatic (reciprocal) 2 | the number of independent subcompositions 4 2 | 4 species on site 1, 2 species on site 2. | M2 and M1 can be identified as sites 1 and 2, respectively. the | species that mix on site 1 are Mn-Mg-Fe-Ca and the species that mix on | site 2 are Al-Fe3+. spss alm py gr | endmember names fmn_i fkho_i kho_q andr_i 3 | number of dependent endmembers andr_i = 1 kho_q - 1 py +1 gr fkho_i = 1 kho_q + 1 alm -1 py fmn_i = 1 kho_q + 1 spss -1 py 0 0 0 0 0 0 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision (xmn) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision (xfe) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision (xmg) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision x(fe3+) on Y begin_excess_function W(py alm) 2.5d3 | as given by White et al 2014 and updated (post-GreenEtAl) metabasite set. W(py gr) 31d3 | DEK 2-Jun2021: was 30.1d3 0.164 * P_bar W(py kho_q ) 5.4d3 W(py spss) 2d3 W(alm gr) 5d3 W(alm kho_q ) 22.6d3 | DEK 2-Jun2021: was 21.63d3 0.168 * P_bar W(alm spss) 2d3 W(gr kho_q ) -15.3d3 | DEK 2-Jun2021: was -16.15d3 0.164 * P_bar w(spss kho_q ) 29.4d3 | DEK 2-Jun2021: was 28.43d3 0.168 * P_barend_excess_function end_excess_function 2 |2 site entropy model 4 3. |4 species, site multiplicity 3 z(x,mn) = 1 spss z(x,fe) = 1 alm z(x,ca) = 1 gr 2 2. |2 species, site multiplicity 2 z(y,al) = 1 spss + 1 alm + 1 py + 1 gr begin_van_laar_sizes alpha(py) 1 alpha(alm) 1 alpha(gr) 2.7 alpha(kho_q) 1 alpha(spss) 1 end_van_laar_sizes begin_dqf_corrections dqf(kho_q) 27d3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Leonya's melt model Melt(A) abbreviation Melt full_name liquid 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 3 | number of vertices (endmembers) on each simplex abL wL_q hltL 0 1 .1 0 | range and resolution of X(ab) 0 1 .1 0 | range and resolution of X(h2o) begin_excess_function |W(abL wL_q) -1.5d3 -0.3 * P_bar W(abL abL wL_q) -1.5d3 -0.3 * P_bar |Wabw*xab^2*xw W(abL hltL wL_q) -3d3 -0.6 * P_bar |2*Wabw*xab*xhlt*xw W(abL wL_q wL_q) -1.5d3 -0.3 * P_bar |Wabw*xab*xw^2 W(abL hltL hltL) 75340.63 |d*xab*xhlt^2 W(hltL hltL wL_q) -3560.6 |e*xhlt^2*xw W(hltL hltL) 75340.63 |d*xhlt^2 W(hltL wL_q) -3560.6 |e*xhlt*xw end_excess_function 2 | Configurational entropy: quadratic water, linear salt V 2 1 1 | water-vacancy site z(H,V) = 1 wL_q z(V,V) = 1 abL + 1 hltL M 3 0 0 | temkin melt site z(ab,M) = 1 abL z(hlt,M) = 1 hltL z(H,V) = 1 wL_q none | formula suffix, enter "none" for no suffix. begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. This | may be useful for testing purposes for systems in which the | the endmember compositions are not expected to be stabLe, i.e., | the stability of pure wL_q would alert the user to unexpected | behaviour. if/when endmember compositions of the solution are | expected, those endmembers should be removed from the list of | flagged endmembers. wL_q hltL end_flagged_endmembers begin_dqf_corrections dqf(wL_q) = 0 dqf(hltL) = 14072.2 + 13.85077 * T_K -0.830043 * P_bar end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Melt(B) abbreviation Melt full_name liquid 2 model type: simplicial composition space. 2 number of endmembers abL wL_q 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution of X(ab) begin_excess_function W(abL wL_q) -1.5d3 -0.3 * P_bar end_excess_function 2 | Configurational entropy: two non-temkin sites (Water, Melt) 2 1 | water-vacancy site z(H) = 1 wL_q 2 0 | melt species site, treating ksp and abl as separate species n(ab) = 1 abL n(w) = 1 wL_q begin_dqf_corrections dqf(wL_q) = 0 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Melt, White et al., JMG 32:261-286, 2014. JADC 4/14 reformulated as ordinary configurational entropy model. JADC 12/18. This model requires the following make definitions in the thermodynamic data file make_definitions section: siL_q = 8/5 silL_q foL_q = 2 foL faL_q = 2 faL qL_q = 4 qL WARNING 3: the (stabilizing) dqf corrections made to the foL_q, faL_q, and siL_q endmembers make the haplogranite melt model inapplicable to melts where these endmembers are present in high concentrations, to model such situations or to reproduce the published fo-fa-q or sil-q melting phase relations, the dqf corrections (below) should be set to zero. WARNING 5: The melt model incorrectly predicts a high pressure-low temperature stability field for water-silica rich melts. at 10 kb this field extends to ca 750 K and at 3 kb to ca 550 K. To eliminate this artifact set T_melt in perplex_option.dat. melt(W) abbreviation Melt full_name liquid 2 model type: simplicial composition space. 8 number of endmembers foL_q faL_q abL siL_q anL kspL qL_q wL_q 0 0 0 0 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution of X(fo), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(fa), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(ab), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(sil), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(an), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(ksp), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(q), 0 => cartesian subdivision begin_excess_function W(qL_q abL) 12d3 -0.4 * P_bar W(qL_q kspL) -2d3 -0.5 * P_bar W(qL_q anL) 5d3 W(qL_q siL_q) 12d3 W(qL_q foL_q) 12d3 -0.4 * P_bar W(qL_q faL_q) 14d3 W(qL_q wL_q) 17d3 -0.5 * P_bar W(abL kspL) -6d3 3 * P_bar W(abL siL_q) 12d3 W(abL foL_q) 10d3 W(abL faL_q) 2d3 W(abL wL_q) -1.5d3 -0.3 * P_bar W(kspL anL) 0d3 -1 * P_bar W(kspL siL_q) 12d3 W(kspL foL_q) 12d3 W(kspL faL_q) 12d3 W(kspL wL_q) 9.5d3 -0.3 * P_bar W(anL wL_q) 7.5d3 -0.5 * P_bar W(siL_q foL_q) 12d3 W(siL_q faL_q) 12d3 W(siL_q wL_q) 11d3 W(foL_q faL_q) 18d3 W(foL_q wL_q) 11d3 -0.5 * P_bar W(faL_q wL_q) 12d3 end_excess_function 3 | Configurational entropy: two non-temkin sites (Water, Melt) | and one temkin site (olvine). 2 1 | water-vacancy site z(H) = 1 wL_q 2 0 | temkin olivine site n(Mg) = 4 foL_q n(Fe) = 4 faL_q 7 0 | melt species site, treating ksp and abl as separate species n(q) = 1 qL_q | accomplishes the same thing as adding them to form fsp and then n(ksp) = 1 kspL | adding a fsp temkin site. n(ab) = 1 abL n(sil) = 1 siL_q n(an) = 1 anL n(ol) = 1 foL_q + 1 faL_q | this term was not counted prior to dec 2018. n(w) = 1 wL_q begin_dqf_corrections dqf(wL_q) = 0 DQF(qL_q) = 0 DQF(siL_q) = -23d3 DQF(foL_q) = -10d3 DQF(faL_q) = -9d3 -1.3*P_bar end_dqf_corrrections end_of_model -------------------------------------------------------- begin_model Mica(W), White et al., JMG 32:261-286, 2014. This model requires the make definitions: fmu = 1 mu + 1/2 andr - 1/2 gr 25d3 0 0 ma1_dqf: replaced, see Jan 31, 2026 note below. in the thermodynamic data file (e.g., hp11ver.dat). In thermocalc, different dqf's are used for "muscovite", "margarite" and "paragonite" described by this model. this treatment is clearly inconsistent and not followed here. The dqf's given above are for the thermocalc "muscovite" phase. Therefore this model cannot be used to predict margarite-rich phase relations. JADC 2014 Jan 31, 2026: the ma1_dqf has been replaced by an internal DQF, with the value of 5 kJ/mol; this differs from the published value but is used in current TC files. It is not clear when the modification was made. JADC May 1, 2019: W(pa ma) and W(mu ma) adjusted slightly to match the values used in the Holland et al (2018) TC files (3/19). It is not known when the change was introduced in TC. Change noted by Debaditya Bandyopadhyay A M2a M2b T1 M1 ___________________________________ Mutliplicity 1 1 1 2 1 ___________________________________ 1 mu K Al Al AlSi _ 2 pa Na Al Al AlSi _ 3 ma Ca Al Al AlAl _ 4 cel K Mg Al SiSi _ 5 fcel K Fe Al SiSi _ 6 fmu K Al Fe3+ AlSi _ Mica(W) abbreviation Mica full_name white-mica 2 | model type: simplex 6 | 6 endmembers mu pa ma cel fmu fcel 0 0 0 0 0 0 0 0 | endmember flags | subdivision model 0 1 .1 0 | range and resolution of X(mu), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(pa), imod = 0 -> cartesian subdivision 0 1 .1 1 | range and resolution of X(ma), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(cel), imod = 0 -> cartesian subdivision 0 .3 .1 0 | range and resolution of X(fmu), begin_excess_function W(mu cel) 0 + .2 * P_bar W(mu fcel) 0 + .2 * P_bar W(mu pa) 10.12d3 + 3.4 *T_K + .353 * P_bar W(mu ma) 35d3 | changed from 34d3, 5/1/19. W(cel pa) 45d3 + .25 * P_bar W(cel ma) 50d3 W(fcel pa) 45d3 + .25 * P_bar W(fcel ma) 50d3 W(pa ma) 15d3 | changed from 18d3, 5/1/19. W(pa fmu) 30d3 W(ma fmu) 35d3 end_excess function 4 | Configurational entropy: 4 sites, A, M2a, M2b, T1. 3 1 | 3 species on A, 1 site per formula unit. z(a,k) = 1 mu + 1 cel + 1 fcel + 1 fmu z(a,na) = 1 pa 3 1 | 3 species on M2a, 1 site per formula unit. z(m2a,al) = 1 mu + 1 pa + 1 ma + 1 fmu z(m2a,mg) = 1 cel 2 1 | 2 species on M2b, 1 site per formula unit. z(m2b,fe) = 1 fmu 2 2. | 2 species on T1, 2 sites per formula unit. z(t,al) = 1/2 mu + 1/2 pa + 1/2 fmu + 1 ma begin_van_laar_sizes alpha(mu) .63 alpha(pa) .37 alpha(ma) .63 alpha(cel) .63 alpha(fmu) .63 alpha(fcel) .63 end_van_laar_sizes begin_dqf_corrections dqf(ma) 5000 | published as 6500 kj/mol end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Fe3-Fe-Mg Ctd, White et al., JMG 32:261-286, 2014. This model requires the make definitions: ctdo = 1 mct + 1/4 andr - 1/4 gr 25d3 0 0 JADC, 4/14 Mn added DEK June 2021. Ctd(W) abbreviation Ctd full_name chloritoid 2 | model type: simplicial composition space 4 | 4 endmembers ctdo fctd mctd mnctd 0 0 0 0 | endmember flags | Note restricted range on X(Mn) 0 1 .1 0 | range and resolution for X(Fe3), imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | range and resolution for X(Fe), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision begin_excess_function W(mctd fctd) 4d3 W(mctd mnctd) 3d3 W(mctd ctdo) 1d3 W(fctd mnctd) 3d3 W(fctd ctdo) 5d3 W(mnctd ctdo) 4d3 end_excess_function 2 1 site entropy model 3 1 2 species on M1b, site multiplicity = 1. z(Fe) = 1 fctd z(Mg) = 1 mctd + 1 ctdo | added by DEK 02-Jun-2021 (put all Mg-bearing EMs here) 2 .5 2 species on M1a, site multiplicity = 0.5 z(Fe3) = 1 ctdo begin_dqf_corrections dqf(mnctd) 660 | added by DEK 02-JUN-2021 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model 688-type Ti-Fe-Fe3-Mg-Mn Staurolite, White et al., JMG 32:261-286, 2014. This model requires the make definitions: mnst = 1 mnst | added DEK 2nd June 2021 -0.19d3 msto = 1 andr -1 gr + 1 mst 9d3 mstt = 1 mst - 1 cor +3/2 ru 13d3 JADC, 4/14 Missing DQF for mst added by Mark Caddick. JADC, 1/26/2014 mnst added, vacancy on Al site disassociated from Ti, by Felix Gervais. JADC, 1/26/2014 DEK 17th Sept 2021: reconfigured to 688 model type so as to allow calculation of structural formulae with activity output in meemum. St(W) abbreviation St full_name staurolite 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 5 | 5 independent end-members (or number of vertices on each simplex) mstt msto mst mnst fst | end-member names | NAME, XMIN, XMAX, XINC, IMOD (IMOD = 0 is cartesian scheme; IMOD = 1 is non-linear scheme, See https://www.perplex.ethz.ch/perplex/datafiles/solution_model.dat) | If only one simplex no need to put NAME out front. 0. .3 .1 1 | first simplex: range and resolution for X(Ti), imod = 1 -> asymmetric transform subdivision 0. .3 .1 1 | range and resolution for X(Fe3), imod = 1 -> asymmetric transform subdivision 0. 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision 0. 6 .1 0 | range and resolution for X(Mn), imod = 0 -> cartesian subdivision begin_excess_function W(mst fst) 16d3 W(mst mnst) 12d3 W(mst msto) 2d3 W(mst mstt) 20d3 W(fst msto) 18d3 W(fst mnst) 8d3 W(fst mstt) 36d3 W(mnst msto) 14d3 W(mnst mstt) 32d3 W(msto mstt) 30d3 end_excess_function 2 2 site entropy model X 3 4 4 3 species, effective site multiplicity of 4, true multiplicity z(Mg,X) = 1 mst + 1 mstt + 1 msto | corrected from just 1 mst by Felix Gervais, 4/9/2015. JADC x(Mn,X) = 1 mnst z(Fe,X) = 1 fst Y 4 2 2 4 species, effective site multiplicity of 2, true multiplicity (not clear if the vacancy is associated with Ti) z(Ti,Y) = 3/4 mstt z(Fe3,Y) = 1 msto z(v,Y) = 1/4 mstt z(Al,Y) = 1 mst + 1 fst + 1 mnst [Al16Si7.5O44(OH)4] | formula suffix, enter "none" for no suffix begin_dqf_corrections dqf(mst) -8000 dqf(mnst) -190 | corrected from 0.19 by Felix Gervais, 4/9/2015. JADC end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Opx(W) - Fe2+-Fe3+-Mg-Al-Ca Orthopyroxene, White et al., JMG 32:261-286, 2014. This model requires the make definitions: mots1 = 1 mgts + 1/2 andr - 1/2 gr 2d3 0 0 mnopx = 2 pxmn 6.68d3 0 0 JADC, 4/14 Al site fraction of T "corrected" to (mots1 + mgts)/8 by Mark Caddick. JADC, 1/26/2015 Mn added by Felix Gervais, 1/28/2015. This model is as implemented in THERMOCALC. The implementation is irrational because Mn is assumed to remain disordered across M1 and M2, while Al, Fe, Fe3+ and Ca order. The model should only be used with the site_check set to true in perplex_option.dat JADC, 1/28/2015 The correction on 1/26/15 was NOT what is done in thermocalc, apparently the correct form is to compute the true disordered tetrahedral site fraction (mots1 + mgts)/2, but apply a fake multiplicity of 1/4. JADC, 6/10/2016 odi replaced by a internal dqf on di JADC, 30/11/2016 reformulated as a prismatic + orphan vertex model. JADC, 10/5/2018 Site: M1 M2 T ____________________ Mutliplicity: 1 1 1/4 <- fake ____________________ 1 en Mg Mg Si Species: 2 fs Fe2+ Fe2+ Si 3 mnopx Mn Mn Si independent orphan 4 mgts Al Mg AlSi 5 fets Al Fe2+ AlSi dependent 7 mots1 Fe3+ Mg AlSi 8 feots Fe3+ Fe2+ AlSi dependent 10 odi_q Mg Ca Si 11 fdi Fe Ca Si dependent ___________________ Internal: 13 opx Mg Fe2+ Si Dependent: fets = mgts + opx - en feots = mots1 + opx - en fdi = odi_q + opx - en T = Al, Fe3; M1 C = Ca, M2 M = Mg, Fe 2 polytope model: M1, M2 site populations are indicated, remaining are dependent Polytope 1 [Mn][Mn] => 1 Polytope 2 [M,T][M,C] => 2*5 ---------------------------------------------------- Opx(W) | solution name. abbreviation Opx full_name orthopyroxene 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [Mn][Mn] 0 .7 .1 0 [M,T][M,C] by difference | ---------------------------- | Polytope 1 1 | number of simplices, [Mn][Mn] 1 | number of vertices on each simplex mnopx | endmembers on the vertices | ---------------------------- | Polytope 2 2 | number of simplices 2 4 | number of vertices on each simplex | endmembers on the vertices mots1 feots_d odi_q fdi_d mgts fets_d en fs | Simplex 1 X_Mg 0 .8 .1 0 | X(1,1) => bulk Mg/M X_Fe by difference | Simplex 2 X_FeTs 0 .1 .1 0 | X(2,1) - M-Fe3+ X_CaPx 0 .2 .1 0 | X(2,2) - M-Ca X_AlTs 0 .1 .1 0 | X(2,3) - M-Ts X_MPx by difference begin_ordered_endmembers opx = 1/2 en + 1/2 fs Delta(enthalpy) = -6.6d3 | changed from -6.95 DEK 02-JUN-2021 end_ordered_endmembers begin_dependent_endmembers fets_d = 1 mgts + 1 opx - 1 en feots_d = 1 mots1 + 1 opx - 1 en fdi_d = 1 odi_q + 1 fs - 1 opx end_dependent_endmembers begin_excess_function W(en fs) 7d3 W(en opx) 4d3 W(en mgts) 13d3 -0.15 * P_bar W(en mots1) 11d3 -0.15 * P_bar W(en mnopx) 5d3 W(en odi_q) 32.2d3 0.12 * P_bar W(fs opx) 4d3 W(fs mgts) 13d3 -0.15 * P_bar W(fs mots1) 11.6d3 -0.15 * P_bar W(fs mnopx) 4.2d3 W(fs odi_q) 25.54d3 0.084 * P_bar W(opx mgts) 17d3 -0.15 * P_bar W(opx mots1) 15d3 -0.15 * P_bar W(opx mnopx) 5.1d3 W(opx odi_q) 22.54d3 0.084 * P_bar W(mgts mots1) 1d3 W(mgts mnopx) 12d3 -0.15 * P_bar W(mgts odi_q) 75.4d3 -0.94 * P_bar W(mots1 mnopx) 10.6d3 -0.15 * P_bar W(mots1 odi_q) 73.4d3 -0.94 * P_bar W(odi_q mnopx) 24.54d3 0.084 * P_bar end_excess_function 3 | 3 site (M1, M2, T) configurational entropy model M1 | site name 5 1 1 | number of species, effective multiplicity, true multiplicity z(Fe) = 1 fs z(Al) = 1 mgts z(Fe3+) = 1 mots1 z(Mn) = 1 mnopx z(Mg) = 1 en + 1 opx + 1 odi_q M2 | site name 4 1 1 | number of species, effective multiplicity, true multiplicity z(Fe) = 1 fs + 1 opx z(Ca) = 1 odi_q z(Mn) = 1 mnopx z(Mg) = 1 en + 1 mgts + 1 mots1 T | site name | effective site multiplicity increased from .25 to .5, DEK 22/9/2021 | the ".25" value is apparently a typo in the TC commentary. 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 mgts + 1/2 mots1 z(Si,T) = 1/2 mgts + 1/2 mots1 + 1 en + 1 fs + 1 opx + 1 odi_q + 1 mnopx [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(en) 1 alpha(fs) 1 alpha(opx) 1 alpha(mgts) 1 alpha(mots1) 1 alpha(odi_q) 1.2 alpha(mnopx) 1 end_van_laar_sizes begin_dqf_corrections dqf(odi_q) = -100 + 0.211 *T 0.005 *P end_dqf_corrections end_of_model -------------------------------------------------------- begin_model BIOTITE, White et al., JMG 32:261-286, 2014. Modified model entered by Tim Johnson, Apr 7, 2014. enthalpy of ordering for obi corrected, Mark Caddick, JADC 1/26/2015. Mn added and Ti occupancy corrected by Felix Gervais, 1/28/2015. Enthalpy of ordering corrected from -2 to -1 kJ/mol. JADC 5/19 Enthalpy of ordering corrected back to -2 kJ/mol. JADC 6/21 Holland et al. 2018 modify the parameters of this model, the modified version is present as Bi(HGP). M1 M2 T1 H ____________________________ Mutliplicity 1 2 2 2 ____________________________ Dependent: ffbi Fe3+ Fe AlAl OH fbi Fe3+ Mg AlAl OH Dependent: ftbi Ti Fe AlSi O tbi_q Ti Mg AlSi O Dependent: Sdph Al Fe AlAl OH East Al Mg AlAl OH Ann Fe Fe AlSi OH Phl Mg Mg AlSi OH mnbi Mn Mn AlSi OH independent orphan __________________________ Ordered: Obi Fe Mg AlSi OH Bi(W) | solution name. abbreviation Bio full_name biotite 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [Mn][Mn] 0 .5 .1 0 [M,T][M] by difference | ---------------------------- | Polytope 1 - 1 simplex 1 | number of simplices, [Mn][Mn] 1 | number of vertices on each simplex mnbi | endmembers on the vertices | ---------------------------- | Polytope 2 - 2x4 simplices 2 | number of simplices 2 4 | number of vertices on each simplex | endmembers on the vertices ffbi_d fbi_q sdph_d east ftbi_d tbi_q ann phl | First 2-simplex X_Mg 0 1 .1 0 | X(1,1) is bulk Fe/M X_Fe by difference | Second 4-simplex X_FeTs 0 .3 .1 0 | X(2,1) is Fe3/[T+M] on M1, M on M2 X_AlTs 0 .3 .1 0 | X(2,2) is Al/[T+M] on M1, M on M2 X_TiTs 0 .3 .1 0 | X(2,3) is Ti/[M+T] on M2, M on M1 X_MBio by difference begin_ordered_endmembers | enthalpy corrected to -1d3 from -2d3, May 1, 2019. JADC | uncorrected to -2d3 to be consistent with 12-12-2019 TC files. | -2d3 for older (WhiteEtAl14), -3d3 for newer/tc350 | changed from -1d3 DEK 02-JUN-2021. Different tc a-x files have ann vs annm as end-members. | Depending on which a-x file version will determine what DQF_obi is. DEK obi = 2/3 phl + 1/3 ann delta_g_of_ordering = -2d3 end_ordered_endmembers begin_dependent_endmembers sdph_d = 1 east + 1 ann - 1 obi ffbi_d = 1 fbi_q + 1 ann - 1 obi ftbi_d = 1 tbi_q + 1 ann - 1 obi end_dependent_endmembers begin_excess_function W(phl ann) 12d3 W(phl obi) 4d3 W(phl east) 10d3 W(phl tbi_q) 30d3 W(phl fbi_q) 8d3 W(phl mnbi) 9d3 W(ann obi) 8d3 W(ann east) 15e3 W(ann tbi_q) 32d3 W(ann fbi_q) 136d2 W(ann mnbi) 63d2 W(obi east) 7d3 W(obi tbi_q) 24d3 W(obi fbi_q) 5.6d3 W(obi mnbi) 8.1d3 W(east tbi_q) 40d3 W(east fbi_q) 1d3 W(east mnbi) 13d3 W(tbi_q fbi_q) 40d3 W(tbi_q mnbi) 30d3 W(fbi_q mnbi) 116d2 end_excess_function 4 | Configurational entropy: 4 sites, M1, M2, T1, OH. M1 | site name 6 1 1 | 6 species on M1, 1 site per formula unit. z(Mg) = 1 phl z(Fe3+) = 1 fbi_q z(Mn) = 1 mnbi z(Ti) = 1 tbi_q z(Al) = 1 east z(Fe) = 1 ann + 1 obi M2 | site name 3 2 2 | 3 species on M2, 2 sites per formula unit. z(Fe) = 1 ann z(Mn) = 1 mnbi z(Mg) = 1 phl + 1 tbi_q + 1 obi + 1 fbi_q + 1 east T1 | site name 2 2 2 | 2 species on T1, 2 site per formula unit. z(Si) = 1/2 phl + 1/2 ann + 1/2 tbi_q + 1/2 obi + 1/2 mnbi z(Al) = 1/2 phl + 1/2 ann + 1/2 tbi_q + 1/2 obi + 1/2 mnbi + 1 east + 1 fbi_q OH | site name 2 2 2 | 2 species on H, 2 site per formula unit. z(O,OH) = 1 tbi_q z(OH,OH) = 1 phl + 1 obi + 1 fbi_q + 1 east + 1 ann + 1 mnbi [Si2O10] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(ann) = -3000 dqf(mnbi) = -7890 dqf(tbi_q) = 55e3 dqf(fbi_q) = -3e3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model BIOTITE, "HGP" version of model after White et al., JMG 32:261-286, 2014. The Holland et al (2018) TC files (3/19) indicate changes to Bi(W) that are not discussed in that publication. Change noted by Debaditya Bandyopadhyay 5/1/19. M1 M2 T1 H ____________________________ Mutliplicity 1 2 2 2 ____________________________ Dependent: ffbi Fe3+ Fe AlAl OH fbi Fe3+ Mg AlAl OH Dependent: ftbi Ti Fe AlSi O tbi_q Ti Mg AlSi O Dependent: Sdph Al Fe AlAl OH East Al Mg AlAl OH Species: Ann Fe Fe AlSi OH Phl Mg Mg AlSi OH mnbi Mn Mn AlSi OH independent orphan __________________________ Ordered: Obi Fe Mg AlSi OH Bi(HGP) | solution name. abbreviation Bio full_name biotite 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [Mn][Mn] 0 .5 .1 0 [M,T][M] by difference | ---------------------------- | Polytope 1 - 1 simplex 1 | number of simplices, [Mn][Mn] 1 | number of vertices on each simplex mnbi | endmembers on the vertices | ---------------------------- | Polytope 2 - 2x4 simplices 2 | number of simplices 2 4 | number of vertices on each simplex | endmembers on the vertices ffbi_i fbi_q sdph_i east ftbi_i tbi_q ann phl | First 2-simplex X_Mg 0 1 .1 0 | X(1,1) is bulk Fe/M X_Fe by difference | Second 4-simplex X_FeTs 0 1 .1 0 | X(2,1) is Fe3/[T+M] on M1, M on M2 X_AlTs 0 1 .1 0 | X(2,2) is Al/[T+M] on M1, M on M2 X_TiTs 0 1 .1 0 | X(2,3) is Ti/[M+T] on M2, M on M1 X_MBio by difference begin_ordered_endmembers | enthalpy corrected to -1d3 from -2d3, May 1, 2019. JADC | uncorrected to -4d3 to be consistent with 31-10-20 TC files. JADC, June 20, 2021. obi = 2/3 phl + 1/3 ann delta_g_of_ordering = -4d3 end_ordered_endmembers begin_dependent_endmembers sdph_i = 1 east + 1 ann - 1 obi ffbi_i = 1 fbi_q + 1 ann - 1 obi ftbi_i = 1 tbi_q + 1 ann - 1 obi end_dependent_endmembers begin_excess_function W(phl ann) 12d3 W(phl obi) 4d3 W(phl east) 10d3 W(phl tbi_q) 30d3 W(phl fbi_q) 8d3 | W24 14000 W(phl mnbi) 9d3 W(ann obi) 8d3 W(ann east) 5e3 | changed from 15d3 by Holland et al 2018 TC file. Debaditya Bandyopadhyay 5/1/19. W(ann tbi_q) 32d3 W(ann fbi_q) 136d2 W(ann mnbi) 63d2 W(obi east) 7d3 W(obi tbi_q) 24d3 W(obi fbi_q) 5.6d3 W(obi mnbi) 8.1d3 W(east tbi_q) 40d3 W(east fbi_q) 1d3 W(east mnbi) 13d3 W(tbi_q fbi_q) 40d3 W(tbi_q mnbi) 30d3 W(fbi_q mnbi) 116d2 end_excess_function 4 | Configurational entropy: 4 sites, M1, M2, T1, OH. M1 | site name 6 1 1 | 6 species on M1, 1 site per formula unit. z(Mg) = 1 phl z(Fe3+) = 1 fbi_q z(Mn) = 1 mnbi z(Ti) = 1 tbi_q z(Al) = 1 east z(Fe) = 1 ann + 1 obi M2 | site name 3 2 2 | 3 species on M2, 2 sites per formula unit. z(Fe) = 1 ann z(Mn) = 1 mnbi z(Mg) = 1 phl + 1 tbi_q + 1 obi + 1 fbi_q + 1 east T1 | site name 2 2 2 | 2 species on T1, 2 site per formula unit. z(Si) = 1/2 phl + 1/2 ann + 1/2 tbi_q + 1/2 obi + 1/2 mnbi z(Al) = 1/2 phl + 1/2 ann + 1/2 tbi_q + 1/2 obi + 1/2 mnbi + 1 east + 1 fbi_q OH | site name 2 2 2 | 2 species on H, 2 site per formula unit. z(O,OH) = 1 tbi_q z(OH,OH) = 1 phl + 1 obi + 1 fbi_q + 1 east + 1 ann + 1 mnbi [Si2O10] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(ann) = -6000 | changed from -3d3 in Holland et al. 2018 TC file. Debaditya Bandyopadhyay 5/1/19. dqf(mnbi) = -7890 dqf(tbi_q) = 55e3 dqf(fbi_q) = -3e3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Biotite - Bi(G25) - after White et al., JMG 32:261-286, 2014. Revised by Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Revised by Green et al, J Pet, 2025 These revisions are all minor and neither considered Mn, both Bi(HGP) and this model retain the Mn formulation of the original model. JADC, Jan 2026 -------------------------------------------------------- M1 M2 T1 H ____________________________ Mutliplicity 1 2 2 2 ____________________________ Dependent: ffbi Fe3+ Fe AlAl OH fbi1_q Fe3+ Mg AlAl OH Dependent: ftbi Ti Fe AlSi O tbi_q Ti Mg AlSi O Dependent: Sdph Al Fe AlAl OH East Al Mg AlAl OH Species: Ann Fe Fe AlSi OH Phl Mg Mg AlSi OH mnbi Mn Mn AlSi OH independent orphan __________________________ Ordered: Obi Fe Mg AlSi OH -------------------------------------------------------- Bi(G25) | solution name. abbreviation Bio full_name biotite 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [Mn][Mn] 0 .5 .1 0 [M,T][M] by difference | ---------------------------- | Polytope 1 - 1 simplex 1 | number of simplices, [Mn][Mn] 1 | number of vertices on each simplex mnbi | endmembers on the vertices | ---------------------------- | Polytope 2 - 2x4 simplices 2 | number of simplices 2 4 | number of vertices on each simplex | endmembers on the vertices ffbi_i fbi1_q sdph_i east ftbi_i tbi_q ann phl | First 2-simplex X_Mg 0 1 .1 0 | X(1,1) is bulk Fe/M X_Fe by difference | Second 4-simplex X_FeTs 0 1 .1 0 | X(2,1) is Fe3/[T+M] on M1, M on M2 X_AlTs 0 1 .1 0 | X(2,2) is Al/[T+M] on M1, M on M2 X_TiTs 0 1 .1 0 | X(2,3) is Ti/[M+T] on M2, M on M1 X_MBio by difference begin_ordered_endmembers | dqf(obi) = -6d3 | -1/3 dqf(ann) = 3d3 => -7d3 obi = 2/3 phl + 1/3 ann delta_g_of_ordering = -7d3 end_ordered_endmembers begin_dependent_endmembers sdph_i = 1 east + 1 ann - 1 obi ffbi_i = 1 fbi1_q + 1 ann - 1 obi ftbi_i = 1 tbi_q + 1 ann - 1 obi end_dependent_endmembers begin_excess_function w(phl ann) +12000 w(phl obi) +4000 w(phl east) +10000 w(phl tbi_q) +30000 w(phl fbi1_q) +14000 w(ann obi) +8000 w(ann tbi_q) +32000 w(ann fbi1_q) +4000 w(obi east) +7000 w(obi tbi_q) +24000 w(obi fbi1_q) +7000 w(east tbi_q) +40000 w(east fbi1_q) +1000 w(tbi_q fbi1_q) +40000 W(phl mnbi) 9d3 W(ann mnbi) 63d2 W(obi mnbi) 8.1d3 W(east mnbi) 13d3 W(tbi_q mnbi) 30d3 W(fbi1_q mnbi) 116d2 end_excess_function 4 | Configurational entropy: 4 sites, M1, M2, T1, OH. M1 | site name 6 1 1 | 6 species on M1, 1 site per formula unit. z(Mg) = 1 phl z(Fe3+) = 1 fbi1_q z(Mn) = 1 mnbi z(Ti) = 1 tbi_q z(Al) = 1 east z(Fe) = 1 ann + 1 obi M2 | site name 3 2 2 | 3 species on M2, 2 sites per formula unit. z(Fe) = 1 ann z(Mn) = 1 mnbi z(Mg) = 1 phl + 1 tbi_q + 1 obi + 1 fbi1_q + 1 east T1 | site name 2 2 2 | 2 species on T1, 2 site per formula unit. z(Si) = 1/2 phl + 1/2 ann + 1/2 tbi_q + 1/2 obi + 1/2 mnbi z(Al) = 1/2 phl + 1/2 ann + 1/2 tbi_q + 1/2 obi + 1/2 mnbi + 1 east + 1 fbi1_q OH | site name 2 2 2 | 2 species on H, 2 site per formula unit. z(O,OH) = 1 tbi_q z(OH,OH) = 1 phl + 1 obi + 1 fbi1_q + 1 east + 1 ann + 1 mnbi [Si2O10] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(ann) = -3000 dqf(east) = 2200 | apparently this compensates for the elimination of w(ann east) dqf(mnbi) = -7890 dqf(tbi_q) = 55e3 dqf(fbi1_q) = -7200 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model non-ideal CORDIERITE White et al., JMG 32:261-286, 2014 with Mn added M H ________________ Mutliplicity 2 1 ________________ 1 mncrd Mn Vac Species: 2 fcrd Fe Vac 3 crd Mg Vac 4 hmncrd Mn H2O dependent 5 hfcrd Fe H2O dependent 6 hcrd Mg H2O _______________ Dependent: hfcrd = hcrd + (fcrd - crd) Modified model entered by Tim Johnson, Apr 7 2014. w(fcrd hcrd) changed from w(crd hcrd), May 16, 2014. Felix Gervais. w(mncrd ...)'s and mncrd DQF added. Felix Gervais, Jan 28, 2015. Crd(W) abbreviation Crd full_name cordierite 7 model type: reciprocal 2 2 reciprocal sites 3 2 3 species on site 1, 2 on site 2 mncrd fcrd crd hmncrd_i hfcrd_i hcrd 2 2 dependent endmembers hfcrd_i = 1 hcrd + 1 fcrd - 1 crd hmncrd_i = 1 hcrd + 1 mncrd - 1 crd 1 0 0 1 0 0 | endmember flags | Note restricted range on X(Mn) 0 .2 .1 0 | range and resolution for X(Mn) on M site, imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | range and resolution for X(Fe) on M site, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for 1-X(H2O) on H site, imod = 0 -> cartesian subdivision begin_excess_function W(crd fcrd) 8d3 W(fcrd hcrd) 9d3 W(crd mncrd) 6d3 W(fcrd mncrd) 4d3 W(hcrd mncrd) 6d3 end_excess_function 2 2 site entropy model. 3 2. 3 species on M, 2 sites per formula unit. z(m,mg) = 1 crd + 1 hcrd z(m,fe) = 1 fcrd 2 1 2 species on H, 1 sites per formula unit. z(H,H2O) = 1 hcrd begin_dqf_corrections dqf(mncrd) -4210 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model non-ideal CORDIERITE model, HGP version of White et al., JMG 32:261-286, 2014 with Mn added M H ________________ Mutliplicity 2 1 ________________ 1 mncrd Mn Vac Species: 2 fcrd Fe Vac 3 crd Mg Vac 4 hmncrd Mn H2O dependent 5 hfcrd Fe H2O dependent 6 hcrd Mg H2O _______________ Dependent: hfcrd = hcrd + (fcrd - crd) Modified model entered by Tim Johnson, Apr 7 2014. w(fcrd hcrd) changed from w(crd hcrd), May 16, 2014. Felix Gervais. w(mncrd ...)'s and mncrd DQF added. Felix Gervais, Jan 28, 2015. Crd(HGP) abbreviation Crd full_name cordierite 7 model type: reciprocal 2 2 reciprocal sites 3 2 3 species on site 1, 2 on site 2 mncrd fcrd crd hmncrd_i hfcrd_i hcrd 2 2 dependent endmembers hfcrd_i = 1 hcrd + 1 fcrd - 1 crd hmncrd_i = 1 hcrd + 1 mncrd - 1 crd 1 0 0 1 0 0 | endmember flags | Note restricted range on X(Mn) 0 .2 .1 0 | range and resolution for X(Mn) on M site, imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | range and resolution for X(Fe) on M site, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for 1-X(H2O) on H site, imod = 0 -> cartesian subdivision begin_excess_function W(crd fcrd) 6d3 W(crd mncrd) 6d3 W(fcrd mncrd) 4d3 W(hcrd mncrd) 6d3 end_excess_function 2 2 site entropy model. 3 2. 3 species on M, 2 sites per formula unit. z(m,mg) = 1 crd + 1 hcrd z(m,fe) = 1 fcrd 2 1 2 species on H, 1 sites per formula unit. z(H,H2O) = 1 hcrd begin_dqf_corrections dqf(mncrd) -4210 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Sapphirine, O/D non-ideal, Wheller & Powell (JMG, 32, 287 299, 2014) ---------------------------------------------------------- refer to the commentary of Sapp(KWP) for a resume of Sapphirine models in solution_model.dat ---------------------------------------------------------- Modified JADC model entered by Tim Johnson, 7 April 2014. 1 2 3 M3 M456 T _________________ Mutliplicity 1 3 1 _________________ 1 spr4 Mg Mg Si Species: 2 fspr Fe Fe Si 3 spr5 Al Mg Al 4 fsp5_d Al Fe Al dependent 5 ospr Fe3+ Mg Al 6 fospr_d Fe3+ Fe Al dependent 7 spro Fe Mg Si ordered Dependent endmember: FeMg-1_M456 = (fspr-spro)/3 fsp5_d = spr5 + fspr - spro fospro_d = ospr + fspr - spro Sa(WP) abbreviation Sap full_name sapphirine 8 | model type: order-disorder, prismatic composition space 2 | 2 independent mixing sites 2 3 | 2 components {Fe2+, Mg} for composition 1, 3 components {Al, Si, Fe3+} for composition 2. | endmember names spr4 fspr spr5 fsp5_d ospr fospr_d 1 | 1 ordered species: spro = 3/4 spr4 + 1/4 fspr delta_g_of_ordering = -3.5d3 2 | 2 dependent endmembers fsp5_d = 1 spr5 + 1 fspr - 1 spro fospr_d = 1 spr5 + 1 fspr - 1 spro 0 0 0 0 0 0 | endmember flags, indicate if endmember is part of the solution (i.e. iend = 0). | subdivision model for (binary) site 1 (M2): 0 1 .1 0 | range and resolution of X(Mg), subdivision scheme for site 1: imod = 0 -> cartesian | subdivision model for (ternary) site 2 0 1 .1 0 | range and resolution of X(Ts), subdivision scheme for site 3, species 1: imod = 0 -> cartesian 0 1 .1 0 | range and resolution of X(un-Ts), subdivision scheme for site 3, species 2: imod = 0 -> cartesian begin_excess_function W(spr4 spr5) 10d3 -0.02 * P_bar W(spr4 fspr) 16d3 W(spr4 spro) 12d3 W(spr4 ospr) 8d3 -0.02 * P_bar W(spr5 fspr) 19d3 -0.02 * P_bar W(spr5 spro) 22d3 -0.02 * P_bar W(spr5 ospr) 1d3 W(fspr spro) 4d3 W(fspr ospr) 17600 -0.02 * P_bar W(spro ospr) 20d3 -0.02 * P_bar end_excess_function 3 | 3 site (M3, M456, T) configurational entropy model 2 3. | 2 species on M456, 3 sites per formula unit z(m456,fe) = 1 fspr 4 1 | 4 species on M3, 1 sites per formula unit. z(m3,Al) = 1 spr5 z(m3,Fe3) = 1 ospr z(m3,mg) = 1 spr4 2 1 | 2 species on T, 1 sites per formula unit. z(T,si) = 1 spr4 + 1 fspr + 1 spro begin_dqf_corrections dqf(fspr) -2d3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Orthopyroxene with compound formation, PH '99 Am Min. JADC 3/03 site population limits added Feb 20, 2011, JADC. added ferric iron, Jul 24, 2012. JADC. NOTES: * This model will only function for the FASH subsystem if MGO is present as a component in VERTEX. 1 2 M1 M2 _____________ Mutliplicity 1 1 _____________ 1 en Mg Mg Species: 2 fs Fe2+ Fe2+ 3 mgts Al Mg 4 fets Al Fe2+ dependent 5 mots Fe3+ Mg 6 feots Fe3+ Fe2+ dependent _____________ Internal: 7 opx Mg Fe2+ Dependent: fets = mgts + opx - en feots = mots + opx - en Opx(HP) | solution name. abbreviation Opx full_name orthopyroxene 8 | model type: order-disorder, prismatic composition space 2 | 2 independent composition spaces 2 3 | 2 dimensions on first space, 3 on second | endmembers: mgts fets_d en fs mots feots 1 | ordered species definition opx = 1/2 en + 1/2 fs Delta(enthalpy) = -6.95d3 2 | 2 dependent endmembers fets_d = 1 mgts + 1 opx - 1 en feots = 1 mots + 1 opx - 1 en 0 0 0 0 0 0 | endmember flags, indicate if endmember is part of the solution (i.e. iend = 0). | subdivision model for (binary) site 1 (M2): 0 1 .1 0 | range and resolution of X(Mg) | subdivision model for (ternary) site 2 0 1 .1 0 | range and resolution of X(Ts), subdivision scheme for site 3, species 1: imod = 0 -> cartesian 0 1 .1 0 | range and resolution of X(un-Ts), subdivision scheme for site 3, species 2: imod = 0 -> cartesian begin_excess_function w(en fs) 68d2 w(fs mgts) -1d3 w(en opx) 45d2 w(fs opx) 45d2 w(mgts opx) 12d2 w(en mots) -14d3 w(fs mots) 6d3 w(opx mots) 6d3 end_excess_function 2 | 2 site (M1, M2) configurational entropy model 4 1 | 3 species on M1, 1 site per formula unit. z(m1,fe) = 1 fs z(m1,al) = 1 mgts z(m1,fe3+) = 1 mots 2 1 | 2 species on M2, 1 site per formula unit. z(m2,mg) = 1 en + 1 mgts + 1 mots end_of_model -------------------------------------------------------- begin_model Ti-Fe3+ Biotite from Tajcmanova et al., JMG 2009; extended for Mn solution after Tinkham et al. 2001. Model entered by Lucie Tajcmanova, December, 2008. site population limits added Feb 20, 2011, JADC. reformulated as a prismatic + orphan vertex model. JADC, 10/5/2018 688 format, JADC, 9/9/2019. Site: M1 M2 T1 H ____________________________ Mutliplicity 1 2 2 2 endmember ____________________________ type _________ _________ ffbit Fe3+ FeFe AlAl OH dependent f3b_q Fe3+ MgMg AlAl OH ftbit Fe TiFe AlSi O dependent tbi_q Mg TiMg AlSi O sdph Al FeFe AlAl OH dependent east Al MgMg AlAl OH mnbi Mn MnMn AlSi OH independent orphan ann Fe FeFe AlSi OH phl Mg MgMg AlSi OH obi Fe MgMg AlSi OH ordered Bio(TCC) | model name abbreviation Bio full_name biotite 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [Mn] 0 .5 .1 0 [M,T][M] by difference | ---------------------------- | Polytope 1 - 1 simplex 1 | number of simplices, [Mn][Mn] 1 | number of vertices on each simplex mnbi | endmembers on the vertices | ---------------------------- | Polytope 2 - 2x4 simplices 2 | number of simplices 2 4 | number of vertices on each simplex | endmembers on the vertices ffbit f3b_q sdph east ftbit tbi_q ann phl | First 2-simplex X_Mg 0 1 .1 0 | X(1,1) is bulk Mg/M X_Fe by difference | Second 4-simplex X_FeTs 0 1 .1 0 | X(2,1) is Fe3/[T+M] on M1, M on M2 X_AlTs 0 1 .1 0 | X(2,2) is Al/[T+M] on M1, M on M2 X_TiTs 0 1 .1 0 | X(2,3) is MTi/[M+MTi] on M2, M on M1 X_MBio by difference begin_ordered_endmembers obi = 2/3 phl + 1/3 ann delta_g_of_ordering = -6.8d3 end_ordered_endmembers begin_dependent_endmembers sdph = 1 east + 1 ann - 1 obi ffbit = 1 f3b_q + 1 ann - 1 obi ftbit = 1 tbi_q + 1/2 ann + 1/2 obi - 1 phl end_dependent_endmembers begin_excess_function W(phl ann) 12000 | excess parameters from Holland & Powell, JMG, 2006 W(phl east) 10000 W(phl obi) 4000 W(ann east) 3000 W(ann obi) 8000 W(obi east) 7000 end_excess_function 4 | number of identisites in configurational entropy model M1 | site name 5 1 1 | number of species, effective multiplicity, true multiplicity z(Al,M1) = 1 east z(Mg,M1) = 1 phl + 1 tbi_q z(Mn,M1) = 1 mnbi z(Fe3+,M1) = 1 f3b_q z(Fe,M1) = 1 ann + 1 obi M2 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Fe,M2) = 1 ann z(Ti,M2) = 1/2 tbi_q z(Mn,M2) = 1 mnbi z(Mg,M2) = 1 phl + 1 obi + 1 east + 1/2 tbi_q + 1 f3b_q T1 | site name 2 2 2 | number of species, effective multiplicity, true multiplicity z(Si,T1) = 1/2 phl + 1/2 ann + 1/2 tbi_q + 1/2 obi + 1/2 mnbi z(Al,T1) = 1 east + 1/2 phl + 1/2 ann + 1/2 tbi_q + 1/2 obi + 1/2 mnbi + 1 f3b_q OH | site name 2 2 2 | 2 species on H, 2 site per formula unit. z(O,OH) = 1 tbi_q z(OH,OH) = 1 east + 1 phl + 1 ann + 1 obi + 1 mnbi + 1 f3b_q [Si2O10(OH)2] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(tbi_q) = 84e3 -11.5 *T dqf(f3b_q) = 6000 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model | Edgar Dachs and Artur Benisek (2024): "A new activity model for biotite and its application". | Contributions to Mineralogy and Petrology (2024) 179:93 doi.org/10.1007/s00410-024-02173-6 Fe-Mg-Al(Ts)-Al(ex)-Ti-Fe3+-Mn Biotite: Dachs and Benisek (2024) with Fe-Mg ordering as in literature, without Mg-Al ordering. 688 format and testing, E. Dachs Dez/2023. Notation and ranges modified, JADC, 12/20/24 Polytope 1 = [Mn,A] Polytope 2 = [M,T][M] where T = Al,Ti,Fe3 on M1, A = Al on M2 M = Mg, Fe A M1 M2 T1 H ____________________________________________ Mutliplicity 1 1 2 2 2 ____________________________________________ endmember type _________ ________ polytope 1: 1: pyp V V AlAl SiSi OH orphan mnbi K Mn MnMn AlSi OH orphan polytope 2: 2: ffbio K Fe3 FeFe AlAl OH dependent 3: fbio K Fe3 MgMg AlAl OH 4: sid K Al FeFe AlAl OH dependent 5: east K Al MgMg AlAl OH 6: ftbio K Fe FeTi AlSi O dependent 7: tbio K Mg MgTi AlSi O 8: ann K Fe FeFe AlSi OH 9: phl K Mg MgMg AlSi OH ordered species: 10: obi K Fe MgMg AlSi OH ordered Bio(D) | model name abbreviation Bio full_name biotite 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [Mn,A] 0 0.5 .1 0 [M,T][M] by difference | ---------------------------- | Polytope 1 - 1 simplex 1 | number of simplices 2 | number of vertices on each simplex prl mnbi | endmembers on the vertices X_Mn 0 1 .1 0 | X(1,1) is Mn/(Mn+A) on M2 X_Prl by difference | ---------------------------- | Polytope 2 - 2x4 simplices 2 | number of simplices 2 4 | number of vertices on each simplex | endmembers on the vertices ffbio fbioD sid eastD ftbio tbioD annD phlD | First 2-simplex X_Mg 0 1 .1 0 | X(1,1) is bulk Mg/M X_Fe by difference | Second 2-simplex X_FeTs 0 1 .1 0 | X(2,1) is Fe3/[T+M] on M1, M on M2 X_AlTs 0 1 .1 0 | X(2,2) is Al/[T+M] on M1, M on M2 X_TiTs 0 1 .1 0 | X(2,3) is MTi/[M+MTi] on M2, M on M1 X_MBio by difference begin_ordered_endmembers obi = 2/3 phlD + 1/3 annD enthalpy_of_ordering = -2d3 end_ordered_endmembers begin_dependent_endmembers sid = 1 annD + 1 eastD - 1 obi ftbio = 1 tbioD + 1/2 annD + 1/2 obi - 1 phlD ffbio = 1 fbioD + 1 annD - 1 obi end_dependent_endmembers begin_excess_function W(phlD annD annD) -8800 | excess parameters from Dachs and Benisek (2021, 2024) W(phlD phlD annD) 14300 W(phlD eastD) 19000 W(phlD obi) -100 W(phlD prl) 116800 W(annD eastD) -5000 W(annD obi) -400 W(annD prl) 108200 W(annD tbioD) -30000 W(eastD obi) -5000 W(eastD prl) 120000 W(obi prl) 120000 W(prl tbioD) 120000 W(prl fbioD) 120000 end_excess_function 5 | number of identisites in configurational entropy model A 2 1 1 z(K,A) = 1 phlD + 1 eastD + 1 annD + 1 obi + 1 tbioD + 1 fbioD + 1 mnbi z(Vac,A) = 1 prl M1 | site name 6 1 1 | number of species, effective multiplicity, true multiplicity z(Al,M1) = 1 eastD z(Mg,M1) = 1 phlD + 1 tbioD z(Fe3,M1) = 1 fbioD z(Mn,M1) = 1 mnbi z(Fe,M1) = 1 annD + 1 obi z(Vac,M1) = 1 prl M2 | site name 5 2 2 | number of species, effective multiplicity, true multiplicity z(Fe,M2) = 1 annD z(Al,M2) = 1 prl z(Ti,M2) = 1/2 tbioD z(Mn,M2) = 1 mnbi z(Mg,M2) = 1 eastD + 1 obi + 1 phlD + 1/2 tbioD + 1 fbioD T1 | site name 2 2 2 | number of species, effective multiplicity, true multiplicity z(Si,T1) = 1/2 phlD + 1/2 annD + 1/2 obi + 1 prl + 1/2 tbioD + 1/2 mnbi z(Al,T1) = 1 eastD + 1/2 phlD + 1/2 annD + 1/2 obi + 1/2 tbioD + 1 fbioD + 1/2 mnbi OH | site name 2 2 2 | 2 species on H, 2 site per formula unit. z(O,OH) = 1 tbioD z(OH,OH) = 1 eastD + 1 phlD + 1 annD + 1 obi + 1 prl + 1 fbioD + 1 mnbi [Si2O10(OH)2] | formula suffix, enter "none" for no suffix. end_of_model -------------------------------------------------------- begin_model Ti-Biotite model after White, Powell & Holland (JMG, 2007) Model entered by Lucie Tajcmanova, May 11, 2007. DQF corrections to annite added, Mark Caddick, Nov, 2007. NOTE: this model requires make defintions for fbi and tbi in the thermodynamic data file. M1 M2 T1 H ____________________________ Mutliplicity 1 2 2 2 ____________________________ Dependent: 1 ffbi Fe3+ Fe AlAl OH 2 fbi Fe3+ Mg AlAl OH Dependent: 3 ftbi Ti Fe AlSi O 4 tbi1 Ti Mg AlSi O Dependent: 5 Sdph Al Fe AlAl OH 6 East Al Mg AlAl OH Species: 7 Ann Fe Fe AlSi OH 8 Phl Mg Mg AlSi OH __________________________ Ordered: 9 Obi Fe Mg AlSi OH Bio(WPH) | solution name. abbreviation Bio full_name biotite 8 | model type: order-disorder, prismatic composition space 2 | 2 4 | 2 species on site 1, 4 species on site 2. | M2 and M1 can be identified as sites 1 and 2, respectively. the | species that mix on site 1 are Mg-Fe and the species that mix on | site 2 are M2+, Al, Ti. Fe3+. The identity of M2+ on site 2 is determined by | the identity of the M2+ cation on site 1 ffbi_i fbi ftbi_i tbi1 sdph_i east ann phl 1 | ordered species: obi = 2/3 phl + 1/3 ann delta_g_of_ordering = -10.73d3 3 | 3 dependent endmembers sdph_i = 1 east + 1 ann - 1 obi ffbi_i = 1 fbi + 1 ann - 1 obi ftbi_i = 1 tbi1 + 1 ann - 1 obi 0 0 0 0 0 0 0 0 0 | endmember flags: if 0 the endmember is considered to be part of the solution. | subdivision model for (binary) site 1 (M2): 0 1 .1 0 | range and resolution of X(Fe) | subdivision model for (quinary) site 2 (M1) 0. .2 .1 0 | range and resolution of X(Fe3+,M1) 0. .2 .1 0 | range and resolution of X(Ti,M1) 0 1 .1 0 | range and resolution of X(Al,M1) begin_excess_function | current preferred thermocalc values, Caddick, Nov '07 W(phl ann) 9000. W(phl east) 10000. W(phl obi) 3000. W(ann east) -1000. W(ann obi) 6000. W(ann fbi) 8000 W(ann tbi1) 10000 W(obi east) 10000. | values from White et al paper and earlier Perple_X verions. | W(phl ann) 12000 | W(phl east) 10000 | W(phl obi) 4000 | W(ann east) 3000 | W(ann obi) 8000 | W(ann fbi) 8000 | W(ann tbi1) 10000 | W(obi east) 7000 end_excess_function 4 | Configurational entropy: 4 sites, M1, M2, T1 H. 5 1 | 5 species on M1, 1 site per formula unit. z(m1,fe) = 1 ann + 1 obi z(m1,mg) = 1 phl z(m1,Fe3+) = 1 fbi z(m1,Ti) = 1 tbi1 2 2. | 2 species on M2, 2 sites per formula unit. z(m2,fe) = 1 ann 2 2. | 2 species on T1, 2 site per formula unit. z(Al,T1) = 1/2 tbi1 + 1/2 ann + 1/2 phl + 1/2 obi + 1 east + 1 fbi 2 2. | 2 species on H, 2 site per formula unit. z(h,o) = 1 tbi1 begin_dqf_corrections dqf(ann) -3000 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Orthopyroxene with compound formation, PH '99 Am Min. JADC 3/03 Modified for ideal Cr, additionally a temperature dependence of -16 J/K has been assigned adhoc to W(mgts-en) and W(mgts-opx) in order to increase the Al-content of opx for upper mantle compositions and conditions. A better way of accomplishing the same result would be to increase the entropy of mgts. PGP Workshop 4/12/06. (folk.uio.no/ninasim/Cr_results.html) site population limits added Feb 20, 2011, JADC. NOTES: * This model should not be used for crustal rocks/conditions! * This model will only function for the FASH subsystem if MGO is present as a component in VERTEX. * Added ferric iron. JADC, 3/29/13. 1 2 M1 M2 _____________ Mutliplicity 1 1 _____________ 1 en Mg Mg Species: 2 fs Fe2+ Fe2+ 3 mgts Al Mg 5 mots Fe3+ Mg 7 crts Cr Mg _____________ Internal: 9 opx Mg Fe2+ Dependent: fets = mgts + opx - en feots = mots + opx - en fcrts = crts + opx - en CrOpx(HP) | solution name. abbreviation Opx full_name orthopyroxene 8 | model type: Reciprocal with speciation 2 | 2 independent composition spaces 2 4 | 2 dimensions on first space, 3 on second | endmembers: crts fcrts_d mgts fets_d en fs mots feots 1 | ordered species definition opx = 1/2 en + 1/2 fs Delta(enthalpy) = -6.95d3 3 | 3 dependent endmembers fets_d = 1 mgts + 1 opx - 1 en feots = 1 mots + 1 opx - 1 en fcrts_d = 1 crts - 1/2 en + 1/2 fs 0 0 0 0 0 0 0 0 | endmember flags, indicate if endmember is part of the solution (i.e. iend = 0). | subdivision model for (binary) site 1 (M2): 0 1 .1 0 | range and resolution of X(Mg),subdivision scheme for site 1: imod = 0 -> cartesian | subdivision model for (ternary) site 2 0 1 .1 0 | range and resolution of X(Cr), subdivision scheme for site 3, species 1: imod = 0 -> cartesian 0 1 .1 0 | range and resolution of X(Ts), subdivision scheme for site 3, species 2: imod = 0 -> cartesian 0 1 .1 0 | range and resolution of X(un-Ts), subdivision scheme for site 3, species 3: imod = 0 -> cartesian begin_excess_function w(en fs) 68d2 w(fs mgts) -1d3 -14. 0 w(en opx) 45d2 w(fs opx) 45d2 w(mgts opx) 12d2 -14. 0 w(en mots) -14d3 w(fs mots) 6d3 w(opx mots) 6d3 end_excess_function 2 | 2 site (M1, M2) configurational entropy model 5 1 | 5 species on M1, 1 site per formula unit. z(m1,fe) = 1 fs z(m1,al) = 1 mgts z(m1,fe3+) = 1 mots z(m1,cr) = 1 crts 2 1 | 2 species on M2, 1 site per formula unit. z(m2,mg) = 1 en + 1 mgts + 1 mots + 1 crts site_check_override end_of_model -------------------------------------------------------- begin_model Mg-Fe-Ca-Al-Cr Garnet Hybrid Holland & Powell + Simon/PGP Cr Workshop (folk.uio.no/ninasim/Cr_results.html) CrGt abbreviation Gt full_name garnet 7 | model type: reciprocal 2 | 2 chemical site model 3 2 | 3 endmembers mix on site 1, 2 endmembers mix on site 2 uv_d fuv_d knor gr alm py 2 | 2 dependent endmembers uv_d = 1 gr + 1 knor - 1 py fuv_d = 1 alm + 1 knor - 1 py 0 0 0 0 0 0 | endmember flags 0 include it 1 drop it 0 1 .1 0 | range and resolution for XCa on A 0 1 .1 0 | range and resolution for XFe on A 0 1 .1 0 | range and resolution for XCr on B begin_excess_function w(py gr) 33d3 w(py knor) 3d3 | corrected as in erratum to ziberna et al 2013 (2014) | w(py py gr) 59304. -10.5 .036 Ganguly excess paramters | w(py gr gr) 25860. -10.5 .174 end_excess_function 2 2 site entropy model 3 3. 3 species, site multiplicity of 3 z(Ca) = 1 gr z(Mg) = 1 py + 1 knor 2 2. 2 species, site multiplicity of 2 z(Al) = 1 gr + 1 alm + 1 py end_of_model -------------------------------------------------------- begin_model Garnet model of Malaspina et al. 2009 the andradite endmember should be excluded from calculations when this model is use. JADC, 6/2010 Cr added as in Simon/PGP Cr Workshop, folk.uio.no/ninasim/Cr_results.html. JADC, 3/29/13 For Cr garnet it's probably better to use CrGt. JADC, 23/1/15. M1 M2 _____________ Mutliplicity 3 2 _____________ skiag Fe Fe3+ Dependent: kho Mg Fe3+ Dependent: andr Ca Fe3+ Dependent: fuv_d Fe Cr knor Mg Cr Dependent: uv_d Ca Cr alm Fe Al py Mg Al gr Ca Al Gt(MPF) abbreviation Gt full_name garnet 7 | model type: macroscopic with dependent endmembers 2 | number of independent mixing sites, reciprocal solution 3 3 | 3 species on site M1, 2 species on site M2. alm py gr | endmember names skiag kho_d andr fuv_d knor uv_d 4 | number of dependent endmembers andr = 1 gr + 1 skiag - 1 alm kho_d = 1 py + 1 skiag - 1 alm uv_d = 1 gr + 1 knor - 1 py fuv_d = 1 alm + 1 knor - 1 py 0 0 0 0 0 0 0 0 0 | endmember flags 0. 1 .1 0 | imod = 0 -> cartesian subdivision (xfe) on M1 0. 1 .1 0 | imod = 0 -> cartesian subdivision (xmg) on M1 0. 1 .1 0 | imod = 0 -> cartesian subdivision x(al) on M2 0. 1 .1 0 | imod = 0 -> cartesian subdivision x(fe3+) on M2 begin_excess_function w(py knor) 30000. w(alm gr) 15000. w(py gr) 80000. | HP (33000.) w(alm py) 2500. | HP w(alm skiag) -37053.18 | Skiag end_excess_function 2 |2 site entropy model 3 3. |3 species, site multiplicity 3 z(x,fe) = 1 alm + 1 skiag z(x,Mg) = 1 py + 1 knor 3 2. |2 species, site multiplicity 2 z(y,al) = 1 alm + 1 py + 1 gr z(y,cr) = 1 knor end_of_model -------------------------------------------------------- begin_model ad hoc pumpellyite, assumes fe-mg occupy two octahedral (M1) sites and al-fe3 occupy 5 (M2). JADC, Sep 17, 2017 Pu abbreviation Pu full_name pumpellyite 7 | model type: macroscopic with dependent endmembers 2 | number of independent mixing sites, reciprocal solution 2 2 | 2 species on site M1, 2 species on site M2. mpm fpm | endmembers mjgd jgd 1 | number of dependent endmembers mjgd = 1 jgd + 1 mpm - 1 fpm 0 0 0 0 | endmember flags 0. 1 .1 0 | imod = 0 -> cartesian subdivision (xfe) on M1 0. 1 .1 0 | imod = 0 -> cartesian subdivision (xmg) on M1 ideal 2 |2 site entropy model 2 1 |2 species, M1 site multiplicity 1 z(M1,Mg) = 1 mpm 2 5. |2 species, M2 site multiplicity 5 z(M2,Fe3) = 1 jgd end_of_model -------------------------------------------------------- begin_model Sack & Ghiorso (1989 CMP 102:41-68) for Fe-Mg olivine. [an astoundingly complex presentation] JADC 7/03 O(SG) abbreviation Ol full_name olivine 2 model type macroscopic 2 2 endmembers fo fa 0 0 | endmember flags 0 1 .1 0 | range and resolution of X(mg), imod = 0 -> cartesian subdivision begin_excess_function w(fo fa) 20314. 0 3e-2 end_excess_function 1 1 site entropy model (m) 2 2. 2 species on m2, mutiplicity = 2 z(m,mg) = 1 fo end_of_model -------------------------------------------------------- begin_model Ghiorso et al (2002) G3 v. 3 (n. 5) model for mantle melting, 1-3 GPa. Read cautionary notes in Ghirso for applications beyond this pressure range. Calculations with this model can be sped up significantly by restricting the subdivsion ranges specified below. This is a reduced version of the pMELTS model that excludes the Cr-, Ni-, Co-, and P-bearing melt components. The melt endmember names have been changed from those used in the pMELTS paper. This model should be applied with solid phase data and solution models from the MELTS program. The melt endmember data converted to PERPLE_X format is in the file pMELTSver.dat. The PERPLE_X solid phase data files b92ver.dat and hp98ver.dat also include this data and therefore can be used with the pMELTS model, however using these files will almost certainly result in inconsistencies with published pMELTS results. JADC 7/03 WARNING 3: the subdivision ranges below may not span the entire range of validity for the pMELT model. Check these ranges, and adjust them as necessary before using this model. ==================================================== Modified to use the pure water (H2O) endmember instead of the Ghiorso et al. (2003) h2oGL endmember. My understanding is H2O and h2oGL are essentially equivalent. Ghiorso et al. base h2oGL on the Sterner and Pitzer EoS. JADC May 23, 2004. Added TiO2 endmember. JADC Feb 7, 2012. converted to an ordinary temkin model. JADC, 12/18. pMELTS(G) abbreviation Melt full_name liquid 2 model type: ordinary simplicial 9 number of endmembers H2O foGL faGL woGL kalGL nasGL coGL tiGL qGL 1 0 0 0 0 0 0 0 0 | endmember flags | NOTE restricted compositional ranges! 0 1 .1 0 | range and resolution of X(h2o), 0 => cartesian subdivision 0 .5 .1 0 | range and resolution of X(fo), 0 => cartesian subdivision 0 .2 .1 0 | range and resolution of X(fa), 0 => cartesian subdivision 0 .2 .1 0 | range and resolution of X(wo), 0 => cartesian subdivision 0 .6 .1 0 | range and resolution of X(kal), 0 => cartesian subdivision 0 .6 .1 0 | range and resolution of X(nas), 0 => cartesian subdivision 0 .2 .1 0 | range and resolution of X(co), 0 => cartesian subdivision 0 .2 .1 0 | range and resolution of X(Ti), 0 => cartesian subdivision begin_excess_function w( coGL qGL ) -296975.2 w( faGL qGL ) -18841.4 w( foGL qGL ) -33833.5 w( woGL qGL ) -34232.9 w( nasGL qGL ) -59822.7 w( kalGL qGL ) -102706.5 w( H2O qGL ) -45181.6 w( faGL coGL ) -200788.1 w( foGL coGL ) -192709.0 w( woGL coGL ) -270700.8 w( nasGL coGL ) -205068.6 w( kalGL coGL ) -114506.5 w( H2O coGL ) -161944.4 w( foGL faGL ) -28736.4 w( woGL faGL ) -28573.8 w( nasGL faGL ) -4723.9 w( kalGL faGL ) 22245 w( H2O faGL ) 9769.4 w( woGL foGL ) 574.1 w( nasGL foGL ) 9272.3 w( kalGL foGL ) 36512.7 w( H2O foGL ) 24630.1 w( nasGL woGL ) 7430.3 w( kalGL woGL ) 19927.4 w( H2O woGL ) -1583.7 w( kalGL nasGL) -1102.3 w( H2O nasGL) 13043.1 w( H2O kalGL) 35572.8 w(coGL tiGL) 144804.9 w(faGL tiGL) 9324.2 w(foGL tiGL) 16355.6 w(woGL tiGL) 9471.5 w(nasGL tiGL) 22194.2 w(kalGL tiGL) 3744 end_excess_function | the configurational entropy for this model is unclear | eq 1 of ghiorso et al. (2002) is almost certainly wrong | the model below follows ghioro and sack 1995 & Nicholls 1980 CMP 74:211 2 | 2 sites 9 0 | melt site, 9 species z(h2o) = 1 H2O z(foGL) = 1 foGL z(faGL) = 1 faGL z(woGL) = 1 woGL z(kalGL) = 1 kalGL z(nasGL) = 1 nasGL z(coGL) = 1 coGL z(tiGL) = 1 tiGL z(qGL) = 1 qGL 2 1 | water vacancy site z(h2o) = 1 H2O end_of_model -------------------------------------------------------- begin_model mMELTS(G) is a chopped down version of pMELTS(G) with Cr added to model lunar melting without changing dimensioning. Ghiorso et al (2002) G3 v. 3 (n. 5) model for mantle melting, 1-3 GPa. Read cautionary notes in Ghirso for applications beyond this pressure range. Calculations with this model can be sped up significantly by restricting the subdivsion ranges specified below. This is a reduced version of the pMELTS model that excludes the Cr-, Ni-, Co-, and P-bearing melt components. The melt endmember names have been changed from those used in the pMELTS paper. This model should be applied with solid phase data and solution models from the MELTS program. The melt endmember data converted to PERPLE_X format is in the file pMELTSver.dat. The PERPLE_X solid phase data files b92ver.dat and hp98ver.dat also include this data and therefore can be used with the pMELTS model, however using these files will almost certainly result in inconsistencies with published pMELTS results. JADC 7/03 WARNING 1: This model can only be used for hydrous systems if H2O is specified as a thermodynamic component (i.e., if H2O is specified as a saturated component, VERTEX will reject the h2oG(L) endmember and the model will be applicable only to dry melts). WARNING 2: this model uses an internal routine to compute the entropy of the melt that assumes h2oG(l) is the first endmember. WARNING 3: the subdivision ranges below may not span the entire range of validity for the pMELT model. Check these ranges, and adjust them as necessary before using this model. ==================================================== Modified to use pure water at P and T standard state (the H2O endmember) as opposed to the h2oGL endmember of Ghiorso et al. (2003), which so far as I understand should be equivalent. NOTE: Use of H2O as an endmember requires the user specify an equation of state for H2O when running BUILD. To be strictly consistent with Ghiorso et al the Sterner and Pitzer EoS for water should be used for this purpose. JADC May 23, 2004. Added TiO2 endmember. JADC Feb 7, 2012 reformulated as ordinary solution model. JADC 12/18. mMELTS(G) abbreviation Melt full_name liquid 2 model type: internal entropy routine 8 number of endmembers H2O foGL faGL woGL crGL coGL tiGL qGL 1 0 0 0 0 0 0 0 0 | endmember flags | NOTE restricted compositional ranges! 0 1 .1 0 | range and resolution of X(h2o), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(fo), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(fa), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(wo), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(kal), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(co), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(Ti), 0 => cartesian subdivision begin_excess_function w( coGL qGL ) -296975.2 w( faGL qGL ) -18841.4 w( foGL qGL ) -33833.5 w( woGL qGL ) -34232.9 w( H2O qGL ) -45181.6 w( faGL coGL ) -200788.1 w( foGL coGL ) -192709 w( woGL coGL ) -270700.8 w( H2O coGL ) -161944.4 w( foGL faGL ) -28736.4 w( woGL faGL ) -28573.8 w( H2O faGL ) 9769.4 w( woGL foGL ) 574.1 w( H2O foGL ) 24630.1 w( H2O woGL ) -1583.7 w(coGL tiGL) 144804.9 w(faGL tiGL) 9324.2 w(foGL tiGL) 16355.6 w(woGL tiGL) 9471.5 w(coGL crGL) -269339.7 w(faGL crGL) -74759 w(foGL crGL) -3638.5 w(woGL crGL) 48337.5 w(tiGL crGL) -22455.8 end_excess_function | the configurational entropy for this model is unclear | eq 1 of ghiorso et al. (2002) is almost certainly wrong | the model below follows ghioro and sack 1995 & Nicholls 1980 CMP 74:211 2 | 2 sites 8 0 | melt site, 8 species multiplicity 1 z(h2o) = 1 H2O z(foGL) = 1 foGL z(faGL) = 1 faGL z(crGL) = 1 crGL z(woGL) = 1 woGL z(coGL) = 1 coGL z(tiGL) = 1 tiGL z(qGL) = 1 qGL 2 1 | water vacancy site z(h2o) = 1 H2O end_of_model -------------------------------------------------------- begin_model Ghiorso & Sack (1995) CMP 119:197-212 (the MELTS model) Read cautionary notes in Ghirso. Calculations with this model can be sped up significantly by restricting the subdivsion ranges specified below. This is a reduced version of the MELTS model that excludes the Cr- and P-bearing melt components. The melt endmember names have been changed from those used in the MELTS paper. This model should be applied with solid phase data and solution models from the MELTS program. The melt endmember data converted to PERPLE_X format is in the file pMELTSver.dat. The PERPLE_X solid phase data files b92ver.dat and hp98ver.dat also include this data and therefore can be used with the MELTS model, however using these files will almost certainly result in inconsistencies with published MELTS results. JADC 7/03 WARNING 3: the subdivision ranges below may not span the entire range of validity for the MELT model. Check these ranges, and adjust them as necessary before using this model. MELTS(GS) abbreviation Melt full_name liquid 2 model type: simplicial 8 number of endmembers h2oGM foGM faGM woGM kalGM nasGM coGM qGM 0 0 0 0 0 0 0 0 | endmember flags 0 0.3 0.04 0 0.3 0.7 0.04 0 0 0.3 0.04 0 0 .2 0.04 0 0 0.4 0.04 0 0 .2 0.04 0 0 0.3 0.04 0 | (restricted) subdivision ranges and model begin_excess_function w( coGM qGM ) -39120. w( faGM qGM ) 23661. w( foGM qGM ) 3421. w( woGM qGM ) -864. w( nasGM qGM ) -99039. w( kalGM qGM ) -33922. w( h2oGM qGM ) 30967. w( faGM coGM ) -30509. w( foGM coGM ) -32880. w( woGM coGM ) -57918. w( nasGM coGM ) -130785. w( kalGM coGM ) -25859. w( h2oGM coGM ) -16098. w( foGM faGM ) -37257. w( woGM faGM ) -12971. w( nasGM faGM ) -90534. w( kalGM faGM ) 23649. w( h2oGM faGM ) 28874. w( woGM foGM ) -31732. w( nasGM foGM ) -41877. w( kalGM foGM ) 22323. w( h2oGM foGM ) 35634. w( nasGM woGM ) -13247. w( kalGM woGM ) 17111. w( h2oGM woGM ) 20375. w( kalGM nasGM) 6523. w( h2oGM nasGM) -96938. w( h2oGM kalGM) 10374. end_excess_function 2 8 0 | melt site, 8 species, temkin multiplicity 1 z(h2o) = 1 h2oGM z(fo) = 1 foGM z(fa) = 1 faGM z(wo) = 1 woGM z(kal) = 1 kalGM z(nas) = 1 nasGM z(co) = 1 coGM z(q) = 1 qGM 2 1 | water vacancy site z(h2o) = 1 h2oGM end_of_model -------------------------------------------------------- begin_model holland and powell '98 non-ideal cz-fep solution 1 2 M1 M3 _____________ Mutliplicity 1 1 _____________ 1 cz Al Al Species: 2 fep Fe Fe _____________ Ordered Cpd: 3 ep Al Fe Ep(HP) abbreviation Ep full_name epidote 6 | model type: speciation 2 | 2 endmembers cz fep | endmember names 1 | ordered species definition ep = 1/2 fep + 1/2 cz Delta(enthalpy) = -13.05d3 0 0 0 | endmember flags 0. 1 .1 0 | range and resolution for X(cz), imod = 0 -> cartesian subdivision begin_excess_function w(cz fep) 15400 w(fep ep) 3000 end_excess_function 2 | 2 site (M1, M3) configurational entropy model 2 1 | 2 species on M1, 1 site per formula unit. z(fe,m1) = 1 fep 2 1 | 2 species on M3, 1 site per formula unit. z(al,m3) = 1 cz end_of_model -------------------------------------------------------- begin_model Phengite as http://www.esc.cam.ac.uk/astaff/holland/ds5/muscovites/mu.html This model assumes M2 (multiplicity 2) is split into 1 M2a site on which tri- and di-valent cations mix, and an M2b site occupied solely by Al. JADC 2/03 config entropy corrected, D Tinkham, 5/6/03 A M2a T1 _________________________ Mutliplicity 1 1 2 _________________________ 1 mu K Al Al_Si Species: 2 cel K Mg Si_Si 3 fcel K Fe Si_Si 4 pa Na Al Al_Si this model makes a dqf correction for paragonite (meaning that this model is not valid for Na-rich compositions). Pheng(HP) | solution name abbreviation Mica full_name white-mica 2 | model type: endmember fractions 4 | 4 species pa | endmember names cel fcel mu | endmember names 1 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for x(pa), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for x(cel), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for x(fcel), imod = 0 -> cartesian subdivision begin_excess_function w(mu pa) 12d3 0.4 * P_bar w(cel pa) 14d3 0.2 * P_bar w(fcel pa) 14d3 0.2 * P_bar end_excess_function 3 | 3 sites (A, M2, T1) configurational entrpoy model 2 1 | 2 species on A, 1 site per formula unit. z(A,Na) = 1 pa 3 1 | 3 species on M2a, 1 sites per formula unit. z(m2,Mg) = 1 cel z(m2,Fe) = 1 fcel 2 2. | 2 species on T1, 2 sites per formula unit. z(T1,Si) = 1/2 mu + 1/2 pa begin_dqf_corrections | for endmember "name" the dqf correction is | entered as | dqf(name) num1 num2 num3 | where the dqf correction to the endmembers | Gibbs energy is computed as | Gdqf(J/mol) = num1 + T[K]*num2 + P[bar]*num3 dqf(pa) 1420 0.4 * P_bar end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Osumilite, ideal, Holland et al Contrib Mineral Petrol (1996) 124: 383-394 Osm(HP) uses the TC ds5 osumulite endmember names, use Osm(HP11) for calculations with ds6 versions of the TC data base. Internal DQF corrections added for all endmembers. If the true endmembers interfere with the Osm(HP), then the DQF "corrections" must be removed below and added to the thermodynamic data file as make definitions. JADC, 6/7/2019 1 2 3 M1 T1 T2 _________________ Mutliplicity 2 3 2 _________________ Species: osm1 Mg Al Al osm2 Mg MgAl2 AlSi fosm Fe Al Al Osm(HP) abbreviation Osm full_name osumilite 2 model type: endmember fractions. 3 number of endmembers osm1 osm2 fosm endmember names 0 0 0 endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision ideal 3 | 3 site (M1, T1, T2) configurational entropy model 2 2 | 2 species on M1, 2 sites per formula unit z(m1,fe) = 1 fosm 2 3 | 2 species on T1, 3 sites per formula unit. z(t1,mg) = 1/3 osm2 2 2 | 2 species on T2, 2 sites per formula unit. z(t2,si) = 1/2 osm2 begin_dqf_corrections dqf(osm1) = 22.6d3 - 0.02d3 * T_K dqf(osm2) = 20.7d3 - 0.02d3 * T_K dqf(fosm) = 24.8d3 - 0.02d3 * T_K end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Osumilite, ideal, Holland et al Contrib Mineral Petrol (1996) 124: 383-394 1 2 3 M1 T1 T2 _________________ Mutliplicity 2 3 2 _________________ Species: osm1 Mg Al Al => osma osm2 Mg MgAl2 AlSi => osmb fosm Fe Al Al => osfa In DS6 the endmembers have been renamed osma, ossm, and osfa respectively. Korhonen et al 2013 indicate DQF's for the DS5 endmember names, these are applied below. If the true endmembers interfere with the Osm(HP), then the DQF "corrections" must be removed below and added to the thermodynamic data file as make definitions. JADC, 6/7/2019. Osm(HP11) abbreviation Osm full_name osumilite 2 model type: endmember fractions. 3 number of endmembers osma osmm osfa endmember names 0 0 0 endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision ideal 3 | 3 site (M1, T1, T2) configurational entropy model 2 2 | 2 species on M1, 2 sites per formula unit z(m1,fe) = 1 osfa 2 3 | 2 species on T1, 3 sites per formula unit. z(t1,mg) = 1/3 osmm 2 2 | 2 species on T2, 2 sites per formula unit. z(t2,si) = 1/2 osmm begin_dqf_corrections dqf(osma) = 22.6d3 - 0.02d3 * T_K dqf(osmm) = 20.7d3 - 0.02d3 * T_K dqf(osfa) = 24.8d3 - 0.02d3 * T_K end_dqf_corrections end_of_model -------------------------------------------------------- begin_model primitive non-inverse spinel model with gahnite end-member. Interaction parameters after Nichols et al.(1992), CMP 111, 362-377. Jiri Konopasek If this model is used with Holland & Powell data it automatically has T-tempendent disorder that is independent of composition. JADC, 5/19. GaHcSp abbreviation Sp full_name spinel 2 model type macroscopic 3 3 endmembers gah herc sp endmember names - this order implies: x(1) = x(Zn),x(2) = x(Fe), x(3) = x(Mg) 0 0 0 endmember flags 0 .3 .1 0 | range and resolution for X(Zn), imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | range and resolution for X(Fe), imod = 0 -> cartesian subdivision begin_excess_function w(gah herc) -3800. w(herc sp) 1960. w(gah sp) -2600. end_excess_function 1 one site configurational entropy model 3 1 3 species on M site with multiplicity 1 z(M,mg) = 1 sp z(M,fe) = 1 herc end_of_model -------------------------------------------------------- begin_model | Fluid, Connolly & Trommsdorff CMP 1991. | The EoS used for this model is chosen by the user | when BUILD is run (the value of IFUG in the problem definition file). | If this choice is identified as a hybrid EoS, then the EoS used | for the individual species are controlled by the hybrid_EoS_XXX option. F abbreviation F full_name fluid 0 | solution model type: internal EoS. See explanation in the header of | this file for a list of model types 2 | number of endmembers | NOTE: for model type 0, the first endmember listed here, must correspond | to the second special component in the thermodynamic | file, i.e., the endmembers CANNOT be reordered. CO2 H2O | see commentary in the header of this file for explanation of | endmember flags and subdivision schemes. 0 0 | endmember flags, 1 -> the endmember composition is not considered part of the solution, otherwise 0 0 1 .1 0 | subdivision scheme for CO2 ideal | the ideal tag indicates excess properties are computed internally 0 | zero indicates a molecular configurational entropy model end_of_model -------------------------------------------------------- begin_model | Aranovich et al. (2010) H2O-CO2-NaCl fluid model. | This model calls the HCNEOS subroutine and uses | CORK for H2O and CO2. HCNEOS was programmed in | 2004 with incorrect pressure dependence on mixing | parameters w3, w4, and w5 (the pressure dependence | was off by a factor of 100). The parameters were | corrected courtesy of Alexander Koltsoz, RAS, 10/24. | The corrected version of HCNEOS is implemented in | Perple_X release 7.1.9+. JADC, 10/24. | Because the use of CORK is hardwired into HCNEOS | it is possible that conflicts may arise between | the HCNEOS and GFSM EoS assignments. To avoid | such conflicts set hybrid_EoS_H2O 2 | [4] 0-2, 4-5 => 0 - MRK, 1 HSMRK, 2 - CORK ... hybrid_EoS_CO2 2 | [4] 0-4 => 0 - MRK, 1 HSMRK, 2 - CORK ... | in perplex_option.dat. F(salt) abbreviation F full_name fluid 26 | model type internal EoS. 3 | three endmembers hltL H2O CO2 0 0 0 0 1 .1 0 | range and increment for x(hlt), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and increment for x(H2O), imod = 0 -> cartesian subdivision ideal | internal excess function 0 | internal config entropy model end_of_model -------------------------------------------------------- begin_model Scapolite. Presumably from Barbara Kuhn's thesis. M T2a T2b ______________________ Mutliplicity 4 1 2 ______________________ 1 me Ca Al Al independent Species: 2 coma Na3Ca Si Si independent 3 koma K3Ca Si Si _____________________________________ 3 mizz NaCa3 Si Al ordered enthalpy of ordering changed to match thermocalc phase eq, rather than the incorrect enthalpy that follows from table 6.4. JADC, Dec 30, 2013. Scap(RP) | solution name abbreviation Scp full_name scapolite 6 | model type: compound formation 3 | 2 endmembers me coma koma 1 | ordered species definition mizz = 2/3 me + 1/3 coma delta_g_of_ordering = -29000 0 0 0| endmember flags 0 1 .1 0 0 1 .1 0 begin_excess_function w(me coma) 20000. w(me mizz) 20000. w(coma mizz) 30000. w(me koma) 20000. w(koma mizz) 30000. end_excess_function 3 | 3 site (M1, T2a, T2b) conigurational entropy model 3 4. | 2 species on M, 4 sites per formula unit. z(m,na) = 1/4 mizz + 3/4 coma z(m,k) = 3/4 koma 2 1 | 2 species on T2a, 1 site per formula unit. z(t2a,al) = 1 me 2 2. | 2 species on T2b, 1 site per formula unit. z(t1a,si) = 1 coma + 1 koma end_of_model -------------------------------------------------------- begin_model Scapolite. Presumably from Barbara Kuhn's thesis. M T2a T2b ______________________ Mutliplicity 4 2 1 ______________________ 1 me Ca Al Al independent 3 miz NaCa3 Si Al ordered 3 kmiz KCa3 Si Al ordered enthalpy of ordering changed to match thermocalc phase eq, rather than the incorrect enthalpy that follows from table 6.4. JADC, Dec 30, 2013. Scap(KA) | solution name abbreviation Scp full_name scapolite 2 | model type: compound formation 3 | 2 endmembers me miz kmiz 0 0 0 | endmember flags 0 1 .1 0 0 1 .1 0 begin_excess_function end_excess_function 2 | 3 site (M1, T2a, T2b) conigurational entropy model 3 4. | 2 species on M, 4 sites per formula unit. z(m,na) = 1/4 miz z(m,k) = 1/4 kmiz 2 1 | 2 species on T2a, 2 sites per formula unit. z(t2a,al) = 1 me end_of_model -------------------------------------------------------- begin_model St(HP) | Mn-Fe-Mg Staurolite abbreviation St full_name staurolite 2 model type: Ideal or macroscopic 3 3 endmembers mnst fst mst 1 0 0 | endmember flags 0. 1 .1 0 | range and resolution for X(Mn), imod = 1 -> asymmetric transform subdivision 0. 1 .1 0 | range and resolution for X(Fe), imod = 0 -> cartesian subdivision begin_excess_function w(mst fst) -8d3 end_excess_function 1 1 site entropy model 3 4. 3 species, site multiplicity of 4 z(Fe) = 1 fst z(Mg) = 1 mst end_of_model -------------------------------------------------------- begin_model Mn-Fe-Mg Ctd Ctd(HP) abbreviation Ctd full_name chloritoid 2 | model type: simplicial composition space 3 | 3 endmembers mnctd fctd mctd 1 0 0 | endmember flags | Note restricted range on X(Mn) 0. .2 .1 0 | range and resolution for X(Mn), imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | range and resolution for X(Fe), imod = 0 -> cartesian subdivision begin_excess_function w(mctd fctd) 1000. end_excess_function 1 1 site entropy model 3 1 3 species, site multiplicity = 1. z(Fe) = 1 fctd z(Mg) = 1 mctd end_of_model -------------------------------------------------------- begin_model Carp | Carpholite abbreviation Crp full_name carpholite 2 | model type: simplicial composition space. 2 | 2 endmembers mcar fcar 0 0 | endmember flags 0. 1 .1 0 | imod = 0 -> cartesian subdivision ideal 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(Mg) = 1 mcar end_of_model -------------------------------------------------------- begin_model ideal anhydrous/hydrous mg/fe HP cordierite model. hfcrd_i stoichiometry corrected, L. Baumgartner, 5/6/03 M H ________________ Mutliplicity 2 1 ________________ 1 mncrd Mn Vac Species: 2 fcrd Fe Vac 3 crd Mg Vac 4 hmncrd Mn H2O dependent 5 hfcrd Fe H2O dependent 6 hcrd Mg H2O _______________ Dependent: hfcrd = hcrd + (fcrd - crd) hCrd abbreviation Crd full_name cordierite 7 model type: reciprocal 2 2 reciprocal sites 3 2 3 species on site 1, 2 on site 2 mncrd fcrd crd hmncrd_i hfcrd_i hcrd 2 2 dependent endmembers hfcrd_i = 1 hcrd + 1 fcrd - 1 crd hmncrd_i = 1 hcrd + 1 mncrd - 1 crd 1 0 0 1 0 0 | endmember flags | Note restricted range on X(Mn) 0. .2 .1 0 | range and resolution for X(Mn) on M site, imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | range and resolution for X(Fe) on M site, imod = 0 -> cartesian subdivision 0. 1 .1 0 | range and resolution for 1-X(H2O) on H site, imod = 0 -> cartesian subdivision ideal 2 2 site entropy model. 3 2. 3 species on M, 2 sites per formula unit. z(m,mg) = 1 crd + 1 hcrd z(m,fe) = 1 fcrd 2 1 2 species on H, 1 sites per formula unit. z(H,H2O) = 1 hcrd end_of_model -------------------------------------------------------- begin_model | ideal model for mg-fe sudoite assuming | mg fe and al are distributed over | 4 m1 sites. Sud(Livi) abbreviation Sud full_name sudoite 2 | Macroscopic 2 | 2 endmembers fsud sud 0 0 | endmember flags 0. 1 .1 0 | imod = 0 -> cartesian subdivision ideal 1 | 1 independent mixing site, M1. 3 4. | 3 species on M1, 4 sites per formula unit. z(Mg) = 1/2 sud z(Fe) = 1/2 fsud end_of_model -------------------------------------------------------- begin_model | ideal model for mg-fe sudoite assuming | mg and fe are distributed over | 2 sites. Sud abbreviation Sud full_name sudoite 2 | model type: simplicial composition space 2 | 2 endmembers fsud sud 0 0 | endmember flags 0. 1 .1 0 | imod = 0 -> cartesian subdivision ideal 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(Mg) = 1 sud end_of_model -------------------------------------------------------- begin_model HP '98 Non-ideal amphibole Cumm abbreviation Cumm full_name clinoamphibole 2 | model type: macroscopic 2 cumm grun 0 0 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function W(cumm grun) 17500 end_excess_function 1 1 site entropy model 2 7. 2 species, site multiplicity = 7. z(mg) = 1 cumm end_of_model -------------------------------------------------------- begin_model anthophyllite Anth abbreviation oAmph full_name orthoamphibole 2 | model type: macroscopic 2 isp fanth anth 0 0 endmember flags 0. 1 .1 0 subdivision ranges and model ideal 1 1 site entropy model 2 7. 2 species, site multiplicity = 7. z(mg) = 1 anth end_of_model -------------------------------------------------------- begin_model "anthophyllite" a compromise model using the clinoamphibole Fe-endmember, cumm and fap should be excluded. A abbreviation oAmph full_name orthoamphibole 2 | model type: Macroscopic 2 grun ap 0 0 0. 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal 1 1 site entropy model 2 7. 2 species, site multiplicity = 7. z(mg) = 1 ap end_of_model -------------------------------------------------------- begin_model Gl abbreviation Gl full_name clinoamphibole 2 | model type: Macroscopic 2 gl fgl 0 0 endmember flags 0. 1 .1 0 | subdivision ranges and model, imod = 0 -> cartesian subdivision ideal 1 1 site entropy model 2 3. 2 species, site multiplicity = 3 z(gl) = 1 gl end_of_model -------------------------------------------------------- begin_model Tr | Tremolite abbreviation Tr full_name clinoamphibole 2 | model type: Macroscopic 2 ftr tr 0 0 endmember flags 0. 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal 1 1 site entropy model 2 5 2 species, site multiplicity = 5. z(mg) = 1 tr end_of_model -------------------------------------------------------- begin_model | ternary feldspar (Benisek et al, CMP 160:327-337, 2010) | for T > 973 K. | Entered by Vratislav Hurai, May 10, 2011. feldspar_B abbreviation Fsp full_name ternary-feldspar 2 | model type: simplicial composition space 3 | 3 endmembers abh an san 0 0 0 | endmember flags = 0 if the endmember is part of the solution. 0 1 .1 0 | range and resolution for albite, cartesian subdivision 0 1 .1 0 | range and resolution for anorthite, cartesian subdivision begin_excess_function w(abh abh san) 17711. -10.3 .461 * P_bar w(abh san san) 22945. -10.3 .327 * P_bar w(an an san) 90600. -.257 * P_bar w(an san san) 60300. -.21 * P_bar w(an an abh) 40000. -16.4 .069 * P_bar w(an abh abh) 14000. -4.7 -.049 * P_bar w(an abh san) 210078. -114.75 -.2965 * P_bar end_excess_function 2 | 2 site (O-site and T-site) entropy model 3 1 | 3 species on O-site, 1 site per formula unit. z(Na) = 0 + 1 abh z(Ca) = 0 + 1 an 2 2 | 2 species on T-site, 2 sites per formula (al-avoidance model) z(Al) = 1/2 + 1/2 an end_of_model -------------------------------------------------------- begin_model Pl(h) | Newton et al 1981 abbreviation Pl full_name binary-feldspar 2 | model type: simplicial composition space 2 | 2 endmembers abh an 1 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function w(abh abh an) 8477 w(an an abh) 28246 end_excess_function 2 | 2 sites (O, T) kerrick and darkens Al-avoidance model: 2 1 | 2 species on O site, multiplicity = 1. z(Na) = 1 abh 2 2. | 2 species on T, mutiplicity = 2. z(Al) = 1/2 + 1/2 an end_of_model -------------------------------------------------------- begin_model | Waldbaum and Thompson 1968. This model is just the San | model with the low structural state endmembers. Kf abbreviation Mic full_name binary-feldspar 2 | model type: Macroscopic 2 | 2 endmembers mic ab 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(mic ab ab) 32098 -16.1356 0.469020 * P_bar w(mic mic ab) 26470 -19.3810 0.387020 * P_bar end_excess_function 1 | 1 site mixing model 2 1 | 2 species on O-site, 1 site per formula unit. z(Na) = 1 ab end_of_model -------------------------------------------------------- begin_model | San | Waldbaum and Thompson 1968. abbreviation San full_name binary-feldspar 2 | model type: Macroscopic 2 | 2 endmembers san abh 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(san abh abh) 32098 -16.1356 0.469020 * P_bar w(san san abh) 26470 -19.3810 0.387020 * P_bar end_excess_function 1 | 1 site mixing model 2 1 | 2 species on O-site, 1 site per formula unit. z(Na) = 1 abh end_of_model -------------------------------------------------------- begin_model San(TH) | Thompson and Hovis 1979. abbreviation San full_name binary-feldspar 2 | model type: Macroscopic 2 | 2 endmembers san abh 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(san abh abh) 17062.4 0.360661 * P_bar w(san san abh) 30978.3 -21.7568 0.360661 * P_bar end_excess_function 1 | 1 site mixing model 2 1 | 2 species on O-site, 1 site per formula unit. z(Na) = 1 abh end_of_model -------------------------------------------------------- begin_model Connolly and Cesare C-O-H Fluid this model is for X(O) = 0-1 | The EoS used for this model is chosen by the user | when BUILD is run (the value of IFUG in the problem definition file). | If this choice is identified as a hybrid EoS, then the EoS used | for the individual species are controlled by the hybrid_EoS_XXX option. GCOHF abbreviation F full_name fluid 0 | model type: Internal EoS 2 | NOTE: for model type 0, the first endmember listed here, must correspond | to the second special component in the thermodynamic | file, i.e., the endmembers CANNOT be reordered. O2 H2 0 0 endmember flags 1d-5 0.999999 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal 0 end_of_model -------------------------------------------------------- begin_model | Ternary C-O-H Fluid | see: perplex.ethz.ch/perplex/faq/calculations_with_unbuffered_COH_fluids.txt COHF abbreviation F full_name fluid 41 | model type: Internal EoS 3 gph O2 H2 1 0 0 | endmember flags 1d-5 0.67 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 1d-5 0.999999 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal 0 end_of_model -------------------------------------------------------- begin_model Non-ideal margarite-paragonite to fit field data of Bucher-Nurminen et al (1983) and Frank (1983), critical T = 972 K, X(Ma) = 33%. JADC, 4/08. MaPa | margarite-paragonite abbreviation Ma full_name white-mica 2 | macroscopic 2 pa ma 0 0 endmember flags 0 1 .1 0 subdivision ranges and model begin_excess_function w(pa pa ma) 18201 w(ma ma pa) 9101 end_excess_function 1 | 1 site mixing model 2 1 | 2 species on O-site, 1 site per formula unit. z(Na) = 1 pa end_of_model -------------------------------------------------------- begin_model | non-ideal hybrid model for K-Na phengitic mica | mixes Chaterjee and Froese (1975) with ideal phengite model | this configurational entropy is that of HP '98, see Phen(HP) config entropy corrected, D Tinkham, 5/6/03 A M2a T _________________________ Mutliplicity 1 1 2 _________________________ Species: 1 cel K Mg Si_Si 2 fcel K Fe Si_Si 3 mu K Al Al_Si 4 pa Na Al Al_Si Mica(CF) | solution name abbreviation Mica full_name white-mica 2 | model type: simplicial composition space 4 | endmembers cel fcel mu pa 0 0 0 0 | endmember flags | subdivision schemes for cel, fcel, and pa, | see commentary in the header of this file for further explanation 0. 1 .1 0 | cel subdivision range 0. 1 .1 0 | fcel subdivision range 0. 1 .1 0 | pa subdivision range begin_excess_function W(mu pa pa) 19456 1.65440 -.4561 * P_bar W(mu mu pa) 12230 0.710440 0.6653 * P_bar end_excess_function 3 | 3 site (M2a, T2, A) entropy model 3 1 | 3 species on M2a, 1 sites per formula unit. z(m,mg) = 1 cel z(m,fe) = 1 fcel 2 2. | 2 species on T2, 2 sites per formula unit. z(t,al) = 1/2 mu + 1/2 pa 2 1 | 2 species on A, 1 site per formula unit. z(a,k) = 1 mu + 1 cel + 1 fcel end_of_model -------------------------------------------------------- begin_model Mica(CHA1): Reciprocal version of white mica model Mica(CHA) after: Coggon & Holland (JMG, 2002, 20:683-696) Auzanneau et al. (CMP 159:1-24, 2010) Coggon & Holland model orginally entered by Mark Caddick, Aug 30, 2005. This model allows Tschermaks and Ti substitution in both the Na and Ca mica subsytems. When these substitutions are insignificant, use of the reduced version of the model used in Mica(CHA) is much more efficient. This model requires the make definition: tip = 1 fcel + 1 geik - 1/2 fs -482876. -14.694 .84 in the thermodynamic data file (e.g., hp02ver.dat) for Ti-phengite (tip). MODIFICATION/CORRECTION HISTORY: 1) van laar size terms added Nov 25, 2005. JADC. 2) ma t12 site occupancy corrected from AlSi to AlAl, D. Dolejs. Mar 23, 2006 3) endmember order corrected from: mu pa ma cel npa nfpa fcel nfma nma to: mu pa ma cel npa nma fcel nfpa nfma JADC, May 4, 2006. 4) van Laar size terms for potassic endmembers corrected from 0.67 to 0.63. L. Tajcmanova, Jan 6, 2010. 5) extended from Mica(CH1) to include Ti-substitution afer Auzanneau et al (2010). N.B., the Coggon and Holland mica model predicts a significantly narrower mu-pa solvus than Chatterjee and Froese's model (1975), so far as I know both models are based on Chartterjee and Froese's experimental determination of the solvus. JADC, 5/17/13 A M2a M2b T12 M1 ___________________________________ Mutliplicity 1 1 1 2 1 ___________________________________ 1 mu K Al Al AlSi _ 2 pa Na Al Al AlSi _ 3 ma Ca Al Al AlAl _ 4 cel K Mg Al SiSi _ 5 npa Na Mg Al SiSi _ dependent 6 nma Ca Mg Al AlSi _ dependent 7 fcel K Fe Al SiSi _ 8 nfpa Na Fe Al SiSi _ dependent 9 nfma Ca Fe Al SiSi _ dependent 10 tip K Mg Ti AlSi _ 11 ntip Na Mg Ti AlSi _ dependent 12 ctip Ca Mg Ti AlAl _ dependent 13 ftip K Fe Ti AlSi _ dependent 14 nftip Na Fe Ti AlSi _ dependent 15 cftip Ca Fe Ti AlAl _ dependent ___________________________________ Mica(CHA1) abbreviation Mica full_name white-mica 7 | model type: Macroscopic 2 | number of independent mixing sites 3 5 | 3 species on site 1, 5 species on site 2. cel npa nma fcel nfpa nfma tip ntip ctip ftip nftip cftip mu pa ma 9 | dependent end-members nma = 1 cel + 1 ma - 1 mu npa = 1 cel + 1 pa - 1 mu nfpa = 1 fcel + 1 pa - 1 mu nfma = 1 fcel + 1 ma - 1 mu ntip = 1 tip + 1 pa - 1 mu ctip = 1 tip + 1 ma - 1 mu ftip = 1 tip + 1 fcel - 1 cel nftip = 1 tip + 1 fcel - 1 cel + 1 pa - 1 mu cftip = 1 tip + 1 fcel - 1 cel + 1 ma - 1 mu 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 | endmember flags | subdivision model, site 1 (A): 0 1 .1 0 | range and resolution of X(K) 0 1 .1 0 | range and resolution of X(Na) | subdivision model, site 2 (M2) 0. .8 .1 0 | range and resolution of X(MgAl,M2) 0. .5 .1 0 | range and resolution of X(FeAl,M2) 0. .5 .1 0 | range and resolution of X(MgTi,M2) 0. .3 .1 0 | range and resolution of X(FeTi,M2) begin_excess_function W(mu pa) 10120. 3.4 0.353 * P_bar W(mu ma) 30000. W(mu cel) 0 0.2 * P_bar W(mu fcel) 0 0.2 * P_bar W(pa ma) 14500. W(pa cel) 52000. W(pa fcel) 52000. W(ma cel) 30000. 0.2 * P_bar W(ma fcel) 30000. 0.2 * P_bar W(tip cel) 10000. W(tip fcel) 10000. W(tip pa) 80000. | W(prl mu) 20000. 0 0.2 | W(prl pa) 20000. 0 0.2 | W(prl ma) 30000. 0 0.2 | W(prl cel) 25000. 0 0.2 | W(prl fcel) 25000. 0 0.2 | W(prl tip) 40000. end_excess function 4 | Configurational entropy: 4 sites, A, M2a, M2b, T1. 3 1 | 3 species on A, 1 site per formula unit. z(a,k) = 1 mu + 1 cel + 1 fcel + 1 tip z(a,na) = 1 pa 3 1 | 3 species on M2a, 1 site per formula unit. z(m2a,al) = 1 mu + 1 pa + 1 ma z(m2a,mg) = 1 cel + 1 tip 2 1 | 2 species on M2b, 1 site per formula unit. z(m2b,ti) = 1 tip 2 2. | 2 species on T1, 2 sites per formula unit. z(t,al) = 1/2 mu + 1/2 pa + 1/2 tip + 1 ma begin_van_laar_sizes alpha(mu) 0.63 alpha(pa) 0.37 alpha(ma) 0.37 alpha(cel) 0.63 alpha(tip) 0.63 alpha(fcel) 0.63 | alpha(prl) .5 | alpha(phl) 0.63 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model Mica(CHA): Non-reciprocal version of white mica model Mica(CHA1) after: Coggon & Holland (JMG, 2002, 20:683-696) Auzanneau et al. (CMP 159:1-24, 2010) The non-reciprocal version does not allow tschermaks or Ti substutions in the Ca- and Na-subsystems. If these substitutions are important use the more costly model Mica(CHA1) This model requires the make definition: tip = 1 fcel + 1 geik - 1/2 fs -482876. -14.694 .84 in the thermodynamic data file (e.g., hp02ver.dat) for Ti-phengite (tip). This version does not include the prl and phl endmembers considered by both Coggon & Holland (2002) and Auzanneau et al. (2010) (the data is commented below) as solution of these endmembers is usually unimportant. JADC, 1/18/10 N.B., the Coggon and Holland mica model predicts a significantly narrower mu-pa solvus than Chatterjee and Froese's model (1975), so far as I know both models are based on Chartterjee and Froese's experimental determination of the solvus. JADC, 5/17/13 A M2a M2b T1 M1 ___________________________________ Mutliplicity 1 1 1 2 1 ___________________________________ 1 mu K Al Al AlSi _ 2 pa Na Al Al AlSi _ 3 ma Ca Al Al AlAl _ 4 cel K Mg Al SiSi _ 5 fcel K Fe Al SiSi _ 6 tip K Mg Ti AlSi _ not included: ___________________________________ 7 prl _ Al Al SiSi _ 8 phl K Mg Mg AlSi Mg ___________________________________ Mica(CHA) abbreviation Mica full_name white-mica 2 | model type: macroscopic. 6 | 6 endmembers mu pa ma cel tip fcel | prl phl 0 0 0 0 0 0 0 0 | endmember flags | subdivision model 0 1 .1 0 | range and resolution of X(mu), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(pa), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(ma), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(cel), imod = 0 -> cartesian subdivision 0. 0.3 .1 0 | range and resolution of X(tip), imod = 1 -> asymmetric transform subdivision begin_excess_function W(mu pa) 10120. 3.4 0.353 * P_bar W(mu ma) 30000 W(mu cel) 0 0.2 * P_bar W(mu fcel) 0 0.2 * P_bar W(pa ma) 14500. W(pa cel) 52000. W(pa fcel) 52000. W(ma cel) 30000 0.2 * P_bar W(ma fcel) 30000. .2 * P_bar W(tip cel) 10000. W(tip fcel) 10000. W(tip pa) 80000. | W(prl mu) 20000. 0 .2 | W(prl pa) 20000. 0 .2 | W(prl ma) 30000. 0 .2 | W(prl cel) 25000. 0 .2 | W(prl fcel) 25000. 0 .2 | W(prl tip) 40000. end_excess function 4 | Configurational entropy: 4 sites, A, M2a, M2b, T1. 3 1 | 3 species on A, 1 site per formula unit. z(a,k) = 1 mu + 1 cel + 1 fcel + 1 tip z(a,na) = 1 pa 3 1 | 3 species on M2a, 1 site per formula unit. z(m2a,al) = 1 mu + 1 pa + 1 ma z(m2a,mg) = 1 cel + 1 tip 2 1 | 2 species on M2b, 1 site per formula unit. z(m2b,ti) = 1 tip 2 2. | 2 species on T1, 2 sites per formula unit. z(t,al) = 1/2 mu + 1/2 pa + 1/2 tip + 1 ma begin_van_laar_sizes alpha(mu) .63 alpha(pa) .37 alpha(ma) .37 alpha(cel) .63 alpha(tip) .63 alpha(fcel) .63 | alpha(prl) .5 | alpha(phl) .63 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model Mica+(CHA): Non-reciprocal version of white mica model Mica(CHA1) after: Coggon & Holland (JMG, 2002, 20:683-696) Auzanneau et al. (CMP 159:1-24, 2010) extended from Mica(CHA) to allow the pyrophyllite substitution, renamed to prevent unintentional use. The non-reciprocal version does not allow tschermaks or Ti substutions in the Ca- and Na-subsystems. If these substitutions are important use the more costly model Mica(CHA1) This model requires the make definition: tip = 1 fcel + 1 geik - 1/2 fs -482876. -14.694 .84 in the thermodynamic data file (e.g., hp02ver.dat) for Ti-phengite (tip). This version does not include the prl and phl endmembers considered by both Coggon & Holland (2002) and Auzanneau et al. (2010) (the data is commented below) as solution of these endmembers is usually unimportant. JADC, 1/18/10 N.B., the Coggon and Holland mica model predicts a significantly narrower mu-pa solvus than Chatterjee and Froese's model (1975), so far as I know both models are based on Chartterjee and Froese's experimental determination of the solvus. JADC, 5/17/13 A M2a M2b T1 M1 ___________________________________ Mutliplicity 1 1 1 2 1 ___________________________________ 1 mu K Al Al AlSi _ 2 pa Na Al Al AlSi _ 3 ma Ca Al Al AlAl _ 4 cel K Mg Al SiSi _ 5 fcel K Fe Al SiSi _ 6 tip K Mg Ti AlSi _ 7 prl _ Al Al SiSi _ not included: ___________________________________ 8 phl K Mg Mg AlSi Mg ___________________________________ Mica+(CHA) abbreviation Mica full_name white-mica 2 | model type: macroscopic. 7 | # of endmembers mu pa ma cel tip prl fcel | phl 0 0 0 0 0 0 0 0 | endmember flags | subdivision model 0 1 .1 0 | range and resolution of X(mu), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(pa), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(ma), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(cel), imod = 0 -> cartesian subdivision 0. .3 .1 0 | range and resolution of X(tip), imod = 1 -> asymmetric transform subdivision 0 .3 .1 0 | range and resolution of X(prl), imod = 1 -> asymmetric transform subdivision begin_excess_function W(mu pa) 10120 3.4 .353 * P_bar W(mu ma) 30000. W(mu cel) 0 .2 * P_bar W(mu fcel) 0 .2 * P_bar W(pa ma) 14500. W(pa cel) 52000. W(pa fcel) 52000. W(ma cel) 30000 .2 * P_bar W(ma fcel) 30000 .2 * P_bar W(tip cel) 10000. W(tip fcel) 10000. W(tip pa) 80000. W(prl mu) 20000 .2 * P_bar W(prl pa) 20000 .2 * P_bar W(prl ma) 30000 .2 * P_bar W(prl cel) 25000 .2 * P_bar W(prl fcel) 25000 .2 * P_bar W(prl tip) 40000. end_excess function 4 | Configurational entropy: 4 sites, A, M2a, M2b, T1. 4 1 | 3 species on A, 1 site per formula unit. z(a,k) = 1 mu + 1 cel + 1 fcel z(a,na) = 1 pa z(a,V) = 1 prl 3 1 | 3 species on M2a, 1 site per formula unit. z(m2a,al) = 1 mu + 1 pa + 1 ma + 1 prl z(m2a,mg) = 1 cel + 1 tip 2 1 | 2 species on M2b, 1 site per formula unit. z(m2b,ti) = 1 tip 2 2. | 2 species on T1, 2 sites per formula unit. z(t,al) = 1/2 mu + 1/2 pa + 1/2 tip + 1 ma begin_van_laar_sizes alpha(mu) .63 alpha(pa) .37 alpha(ma) .37 alpha(cel) .63 alpha(tip) .63 alpha(fcel) .63 alpha(prl) .5 | alpha(phl) 0.63 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model HP '98 olivine solution. O(HP) abbreviation Ol full_name olivine 2 model type: simplicial composition space 4 endmembers teph fran fo fa 0 0 0 0 | endmember flags | NOTE restricted compositional range for Mn 0 1 .1 0 | range and resolution for X(Mn), imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | range and resolution for X(Ni), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision begin_excess_function W(fo fa) 8400 | corrected from 4.2 kJ Nov 15, 2004. end_excess_function 1 1 site entropy model 4 2. 3 species, site multiplicity = 2. z(mg) = 1 fo z(ni) = 1 fran z(fe) = 1 fa end_of_model -------------------------------------------------------- begin_model ad hoc stilpnomelane Stlp abbreviation Stlp full_name stilpnomelane 2 model type: simplicial composition space 2 endmembers mstp fstp 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision ideal 1 1 site entropy model 2 5. 3 species, site multiplicity = 2. z(mg) = 1 mstp end_of_model -------------------------------------------------------- begin_model HP '98 olivine solution with an approximation to the T Kawasaki (J Min Pet Sci, 96:54-66, 2001) symetric fo-monticellite excess function, this is probably adequate a first order model for Ca solution in olivine. To reproduce the solvus or evaluate pressure effects refer to Kawasaki or Warner and Luth 1973 or find more recent work. JADC, Jul 10, 2016. O(HPK) abbreviation Ol full_name olivine 2 model type: simplicial composition space 4 3 endmembers mont teph fo fa 0 0 0 0 | endmember flags | NOTE restricted compositional range for Mn 0 1 .1 0 | range and resolution for X(Ca), imod = 0 0 1 .1 0 | range and resolution for X(Mn), imod = 0 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision begin_excess_function W(fo fa) 8400 | corrected from 4.2 kJ Nov 15, 2004. W(fo mont) 34050. end_excess_function | this model is not incorrect because Ca partitions onto M1, i.e., | the model should be an o/d model, such a model has been published | (probably by Ghiorso or coworkers). 1 1 site entropy model 4 2. 4 species, site multiplicity = 2 z(mn) = 1 teph z(mg) = 1 fo z(fe) = 1 fa end_of_model -------------------------------------------------------- begin_model "Ordered" Jadeite-Diopside-Hedenbergite-CaTs, as: 1) Gasparik '85 (GCA) in the jd/di limit. 2) HP'98 in the di/hed limit 3) Assuming nonideality in the jd/hed limit is the same as for jd/di. 4) No ternary interactions. 5) Gasparik '85 (CMP) in the jd/cats limit. This should be Gaspariks preferred model. JADC Apr. 99. the configurational entropy model has been constructed to take into account that Gasparik uses an X^2 molecular model for CaTs-Di and an X molecular model for Jd-Di. This implies that there is no disorder associated with placing Na on M2 (i.e., it is associated with Al on M1), whereas Al on M1 is not assocated with Al on T. To get Gasparik's molecular formulation it is necessary to specify that Al mixes on only one of the two T-sites. Cpx(l) abbreviation Cpx full_name clinopyroxene 2 | model type: simplicial composition space 4 | 4 endmembers di hed cats jd 0 0 0 0 | endmember flags 0 1 .1 0 0 1 .1 0 0 1 .1 0 | subdivision ranges and model begin_excess_function w(cats jd) 14810 -7.15 0 w(cats cats jd) -5070 w(cats jd jd) 5070 w(cats cats cats jd) -3350 w(cats cats jd jd) 6700 w(cats jd jd jd) -3350 w(di jd jd) 12600 -7.6 0 w(di di jd) -12600 7.6 0 w(di jd jd jd) -21400 16.2 0 w(di di jd jd) 42800 -32.4 0 w(di di di jd) 21400 -16.2 0 w(di jd) 12600 -9.45 0. w(di hed) 2500 w(hed hed jd) -12600 7.6 0 w(hed jd jd jd) -21400 16.2 0 w(hed jd) 12600 -9.45 0 w(hed jd jd) 12600 -7.6 0 w(hed hed jd jd) 42800 -32.4 0 w(hed hed hed jd) -21400 16.2 0. end_excess_function 2 wierd entropy model, see above (maybe it's right). 3 1 M1, Al-Mg-Fe, 1 site z(fe,m1) = 1 hed z(mg,m1) = 1 di 2 1 T, Al-Si, this is fake to get gasparik's model. z(al,t) = 1 cats end_of_model -------------------------------------------------------- begin_model "disordered" Jadeite-Diopside-Hedenbergite-CaTs, as: 1) Gasparik '85 (GCA) in the jd/di limit. 2) HP'98 in the di/hed limit 3) Assuming nonideality in the jd/hed limit is the same as for jd/di. 4) No ternary interactions. 5) Gasparik '85 (CMP) in the jd/cats limit. JADC Apr. 99. the configurational entropy model has been constructed to take into account that Gasparik uses an X^2 molecular model for CaTs-Di and Jd-Di. See comments for Cpx(l) above. Cpx(h) abbreviation Cpx full_name clinopyroxene 2 4 di hed cats jd 0 0 0 0 | endmember flags 0 1 .1 0 0 1 .1 0 0 1 .1 0 | subdivision ranges and model begin_excess_function w(cats jd) 14810 -7.15 0 w(cats cats jd) -5070 w(cats jd jd) 5070 w(cats cats cats jd) -3350 w(cats cats jd jd) 6700 w(cats jd jd jd) -3350 w(di jd jd) 12430 -6.21 0 w(di di jd) -12430 6.21 0 w(di jd jd jd) -22290 23.19 0 w(di di jd jd) 44580 -46.38 0 w(di di di jd) -22290 23.19 0 w(di jd) 12540 12.63 0 w(di hed) 2500 w(hed hed jd) -12430 6.21 0 w(hed jd jd jd) -22290 23.19 0 w(hed jd) 12540 12.63 0 w(hed jd jd) 12430 -6.21 0 w(hed hed jd jd) 44580 -46.38 0 w(hed hed hed jd) -22290 23.19 0 end_excess_function 3 msite 3 1 M1, Al-Mg-Fe, 1 site z(Fe,m1) = 1 hed z(Mg,m1) = 1 di 2 1 M2, Ca-Na, 1 site z(na,m2) = 1 jd 2 1 T, Al-Si, this is fake to get gasparik's model. z(al,t) = 1 cats end_of_model -------------------------------------------------------- begin_model HP '98 dolomite-ankerite solution Do(HP) abbreviation Do full_name carbonate 2 | macroscopic 2 | 2 endmembers dol ank 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(dol ank) 3000 end_excess_function 1 | 1 site entropy model 2 1 z(Mg) = 1 dol end_of_model -------------------------------------------------------- begin_model HP '98 Magnesite/siderite modified by DMH to include rhc M(HP) abbreviation Mag full_name carbonate 2 model type: simplicial composition space 3 number of endmembers rhc mag sid endmember names 0 0 0 | endmember flags 0 1 .1 0 | range X(Mn), imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | range X(Fe), imod = 0 -> cartesian subdivision begin_excess_function w(mag sid) 4000 | hp '98 give 4 kJ end_excess_function 1 1 site entropy model 3 1 3 species, site multiplicity 1 z(Fe) = 1 sid z(Mg) = 1 mag end_of_model -------------------------------------------------------- begin_model A solution model for Dolomite from Anovitz & Essene 1987 J Pet 28:389-414; this model requires fictive do-structure endmembers that have a standard state G 20920 j > than the cc-structure endmember, these are made here by a "DQF" correction. Do(AE) abbreviation Do full_name carbonate 2 | model type: simplicial composition space 3 | 3 endmembers cc mag sid 1 1 1 | endmember flags 0 1 .1 0 | subdivision range X(cc), imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range X(mag), imod = 0 -> cartesian subdivision begin_excess_function w(mag mag cc) -96850 36.23 0 w(mag cc cc) -55480 -22.85 0 w(sid sid cc) -155523 141.5 0 | this is a linearization of -1040 -T*(212.4 - 0.2027*T) at 873 K, JADC 5/08. w(sid cc cc) -16746 -89.2 0 | this is a linearization of -86740-T*(-71.18+.9184e-1*T) at 873 K, JADC 5/08. w(sid cc mag) -293520 -121.6 0 | this is a linearization of -185450-T*(369.2-.1418*T) at 873 K, JADC 5/08. end_excess_function 1 | 1 site entropy model 3 1 | 2 species, site multiplicity of 1? should check against source z(Mg) = 1 mag z(Fe) = 1 sid begin_dqf_corrections dqf(cc) 20920 dqf(mag) 20920 dqf(sid) 20920 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model A solution model for Magnesite from Anovitz & Essene 1987 J Pet 28:389-414. Cc(AE) abbreviation Cc full_name carbonate 2 | model type: simplicial composition space 3 | 2 endmembers mag cc sid 0 0 0 endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(cc cc mag) 24300 -7.743 0 w(cc mag mag) 23240 0 0 w(sid sid cc) 28404 -2.5 0 | this is a linearization of 27313.2-5.9651e-6/T-1.43147e-3*T^2 at T = 873. JADC, 5/08. w(sid cc cc) 20624 -7.533 0 | this is a linearization of -70751.4 + 91.8848*T + 2.79244e7/T - 3.59552e-2 * T^2 at T = 873. JADC, 5/08. | missing the ternary interaction parameter estimated by Anovitz & Essene. end_excess_function 1 3 1. z(mg) = 1 mag z(fe) = 1 sid end_of_model -------------------------------------------------------- begin_model Magnesioferrite/magnetite (inverse) MF abbreviation Mt full_name spinel 2 | model type: simplicial composition space 2 1 isp(1), ist(1) mt mft 0 0 endmember flags 0 1 .1 0 subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(mt mft) 8368 | Sack & Ghiorso, W_FeMg, CMP, 1991. end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity of 2 z(Fe) = 1 mt end_of_model -------------------------------------------------------- begin_model Sp(JR) Jamieson and Roeder '85 (iron + ol,1300 C) abbreviation Sp full_name spinel 2 | model type: simplicial composition space 2 | 2 endmembers sp herc 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(sp herc) -3102 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity of 2 z(Fe) = 1 herc end_of_model -------------------------------------------------------- begin_model Ghiorso 1991, gives similar W=8638 Sp(GS) Ganguly and Saxena '87 (ol, 1200-1300 C, 1-5 kb) abbreviation Sp full_name spinel 2 | model type: simplicial composition space 2 | 2 endmembers sp herc 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(sp herc) 7703 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity of 2 z(Fe) = 1 herc end_of_model -------------------------------------------------------- begin_model Sp(HP) HP '98: abbreviation Sp full_name spinel 2 | model type: simplicial composition space 2 | 2 endmembers sp herc 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(sp herc) 7d2 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity of 1 z(Fe) = 1 herc end_of_model -------------------------------------------------------- begin_model valid for T>800C<1300C Mt(W) Wood et al 1991 abbreviation Mt full_name spinel 2 | model type: simplicial composition space 2 | 2 endmembers usp mt 0 0 endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(usp mt mt) 42110 w(usp usp mt) 10580 end_excess_function 2 | 2 site model 3 2. | 3 species on O, 2 sites per formula unit. z(Ti,O) = 1/2 usp z(Fe3+,O) = 1/2 mt 2 1 | 2 species on T, 1 site per formula unit. z(Fe3+,T) = 1 mt end_of_model -------------------------------------------------------- begin_model The Andersen and Lindsley models (Am Min v 73, p 714, 1988) are for ilmenite coexisiting with magnetite, its performance at high T (ca 1200) has been criticized by Ghiorso, but this is probably the best model for T<800 C Jury-rigged for geik, JADC, 2/14/13 Ti only on B, Fe3+ disordered IlHm(A) abbreviation Ilm full_name ilmenite 2 | macroscopic 3 | 3 endmembers ilm hem geik 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(ilm hem hem) 126342.5 -100.6 0 term 1 w(ilm ilm hem) 44204.8 -12.274 0 term 2 w(geik hem hem) 126342.5 -100.6 0 term 1 w(geik geik hem) 44204.8 -12.274 0 term 2 end_excess_function 2 2 1 | B-site, Fe3+ and Ti z(Ti,B) = 1 ilm + 1 geik 3 1 | A-site, Fe2+, Fe3+ and Mg z(Mg,A) = 1 geik z(Fe2+,A) = 1 ilm end_of_model -------------------------------------------------------- begin_model Ideal ilmenite-geikielite-pyrophanite solution IlGkPy abbreviation Ilm full_name ilmenite 2 | model type: simplicial composition space 3 | 3 endmembers pnt geik ilm 0 0 0 | endmember flags | restricted mn range! 0 .2 .1 0 | imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision ideal 1 | 1 site entropy model 3 1 | 3 species, site multiplicity of 1 z(Mn) = 1 pnt z(Mg) = 1 geik end_of_model -------------------------------------------------------- begin_model MtUl(A) | Andersen and Lindsley 1988, Akimoto model abbreviation Mt full_name spinel | assumes Ti is only octahedral, Mg and Fe2+ | jury rigged for magnesio-ferrite, JADC 2/14/13 2 | model type: simplicial composition space 3 | 3 species usp mt mft 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(usp mt mt) 46175. -23.077 0 w(usp usp mt) 15748. w(usp mft mft) 46175. -23.077 0 w(usp usp mft) 15748. end_excess_function 2 | 2 site model, assumes Fe3+ is disorderd 4 2. | 4 species on O, 2 sites per formula unit. z(Ti,O) = 1/2 usp z(Fe3+,O) = 1/2 mt + 1/2 mft z(Mg,O) = 1/2 mft 3 1 | 2 species on T, 1 site per formula unit. z(Fe3+,T) = 1 mt + 1 mft z(Fe2+,T) = 1 usp end_of_model -------------------------------------------------------- begin_model Neph(FB) Ferry and Blencoe '78 abbreviation Neph full_name feldspathoid 2 | macroscopic 2 | 2 endmembers ne kls 0 0 endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(ne kls kls) 85057 -20.0500 -.550000 w(ne ne kls) 35945 23.7800 0.69 end_excess_function 1 | 1 site model 2 1 | 2 species, 1 sites per formula unit. z(na) = 1 ne end_of_model -------------------------------------------------------- begin_model Gt(B) Grossular-pyrope-almandine-spessartine, Berman '90, abbreviation Gt full_name garnet 2 | macroscopic 4 | 4 endmembers gr py alm spss 0 0 0 0 endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function w(gr gr py) 21560 -18.79 0.1 w(gr py py) 69200 -18.79 0.1 w(gr gr alm) 20320 -5.08 0.17 w(gr alm alm) 2620 -5.08 0.09 w(py py alm) 230 0 0.01 w(py alm alm) 3720 0 0.06 w(gr py alm) 58825. -23.87 0.265 w(gr py spss) 45424. -18.7900 0.1 w(gr alm spss) 11470 -18.7900 0.13 w(py alm spss) 1975 0 0.035 end_excess_function 1 1 site entropy model 4 3. 4 species, site multiplicity 3 z(Fe) = 1 alm z(Mg) = 1 py z(Ca) = 1 gr end_of_model -------------------------------------------------------- begin_model Grossular-pyrope-almandine-spessartine garnet Ganguly, Cheng & Tirrone (1996) Contrib Mineral Petrol 126:137-151 The expansion of the Cheng & Ganguly (1994) GCA 58:3763-3767 model to the Perple_X excess function is in Maple file garnet_jiba_96.mws JADC Nov 23, 2010. Gt(GCT) abbreviation Gt full_name garnet 2 | model type: simplicial composition space 4 | endmembers gr py alm spss 0 0 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function w(gr py ) 47191.395 -17.34 .105 w(gr alm) 11468.955 -5.07 .045 w(alm spss) 1616.925 0 .075 w(gr spss) 1616.925 0 .075 w(py alm) 4217.895 0 .105 w(py spss) 36248.895 -23.01 .105 w(gr gr py ) -17689.569 0 .069 w(py py gr) 17689.569 0 -.069 w(gr gr alm) 8849.955 0 .045 w(alm alm gr) -8849.955 0 -.045 w(py py alm) -2132.895 0 -.105 w(alm alm py) 2132.895 0 .105 | these could be deleted: w(py py spss) .15e-1 0 -.015 w(spss spss py) -.15e-1 0 .015 w(alm alm spss) .45e-1 0 -.045 w(spss spss alm) -.45e-1 0 .045 w(gr gr spss) .45e-1 0 -.045 w(spss spss gr) -.45e-1 0 .045 end_excess_function 1 1 site entropy model 4 3. 4 species, site multiplicity 3 z(Fe) = 1 alm z(Mg) = 1 py z(Ca) = 1 gr end_of_model -------------------------------------------------------- begin_model hp '98 quaternary garnet model Gt(HP) abbreviation Gt full_name garnet 2 model type: simplicial composition space 4 number of endmembers spss alm py gr endmember names 1 0 0 0 | endmember flags 0 1 .1 0 | imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision | NOTE restricted subdivision range on Mn (Species 1)! begin_excess_function w(py gr) 33000 w(alm py) 2500 | hp '98 give 2.4 kJ w(py spss) 4500 w(alm spss) 240 end_excess_function 1 1 site entropy model 4 3. 4 species, site multiplicity 3 z(Fe) = 1 alm z(Mg) = 1 py z(Ca) = 1 gr end_of_model -------------------------------------------------------- begin_model Mg-Fe-Ca-Al-Fe3 Garnet Hybrid Holland & Powell + Engi & Wersin The Engi & Wersin terms for the excess energy are suspect and should probably be set to zero. Gt(EWHP) abbreviation Gt full_name garnet 7 | model type: reciprocal 2 | 2 chemical site model 3 2 | 3 endmembers mix on site 1, 2 endmembers mix on site 2 gr alm py andr FA_d MA_d 2 | 2 dependent endmembers MA_d = -1 gr + 1 py + 1 andr FA_d = -1 gr + 1 alm + 1 andr 0 0 0 0 0 0 | endmember flags 0 include it 1 drop it 0 1 .1 0 | range and resolution for XCa on A 0 1 .1 0 | range and resolution for XFe on A 0 1 .1 0 | range and resolution for XAl on B, imod = 0 -> cartesian begin_excess_function w(py gr) 33d3 w(gr andr andr) 25812 0 -0.52 | eliminate this terms to make Fe3+ - Al exchange ideal w(andr gr gr) -93820 0 -0.11 | eliminate this terms to make Fe3+ - Al exchange ideal end_excess_function 2 2 site entropy model 3 3. 3 species, site multiplicity of 3 z(Ca) = 1 gr + 1 andr z(Mg) = 1 py 2 2. 2 species, site multiplicity of 2 z(Al) = 1 gr + 1 alm + 1 py end_of_model -------------------------------------------------------- begin_model Ca-Fe2+-Mg-Al-Fe3+ Garnet model after White, Powell & Holland (JMG, 2007, 25:511-527) Model entered by Lucie Tajcmanova, May 11, 2007. w(alm py) value and reference corrected, Thomas Wagner, June 22, 2007. Interaction parameters (W terms) updated to the current THERMOCALC "preferred" values, Van Laar size parameters, added spessartine, Mark Caddick, Nov, 07. Deleted the unused kho_i interaction term, JADC, Nov 07. alphas and w's updated to current TC values, Lucie Tajcmanova, April 14, 2010. NOTE: the more recent Smye et al. (2011) models [Mica(SGH), Carp(SGH), and Ctd(SGH)]are calibrated in terms of the Gt(WPPH) model for garnet. JADC, Oct 27, 2011. X Y _____________ Mutliplicity 3 2 _____________ Dependent: fkho_i Fe Fe3+ Dependent: kho_i Mg Fe3+ Dependent: fmn_i Mn Fe3+ andr Ca Fe3+ spss Mn Al alm Fe Al py Mg Al gr Ca Al ____________ Gt(WPH) abbreviation Gt full_name garnet 7 | model type: simplicial composition space 2 | the number of independent subcompositions, reciprocal solution if > 1. 4 2 | 4 species on site 1, 2 species on site 2. | M2 and M1 can be identified as sites 1 and 2, respectively. the | species that mix on site 1 are Mn-Mg-Fe-Ca and the species that mix on | site 2 are Al-Fe3+. spss alm py gr | endmember names fmn_i fkho_i kho_i andr 3 | number of dependent endmembers fmn_i = 1 andr + 1 spss -1 gr fkho_i = 1 andr + 1 alm -1 gr kho_i = 1 andr + 1 py -1 gr 0 0 0 0 0 0 0 0 | endmember flags 0 .2 .1 0 | imod = 0 -> cartesian subdivision (xmn) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision (xfe) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision (xmg) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision x(fe3+) on Y begin_excess_function | commented values are from White et al. (2007): w(alm gr) 10000 | w(alm gr) 15000 w(py gr) 45000 | w(py gr) 33000 w(alm py) 2500 | w(alm py) 2500 w(py andr) 90000 | w(py andr) 160000 w(alm andr) 75000 | w(alm andr) 135000 end_excess_function 2 |2 site entropy model 4 3. |4 species, site multiplicity 3 z(x,mn) = 1 spss z(x,fe) = 1 alm z(x,Mg) = 1 py 2 2. |2 species, site multiplicity 2 z(y,al) = 1 spss + 1 alm + 1 py + 1 gr begin_van_laar_sizes | commented values are from White et al. (2007): alpha(py) 1 alpha(alm) 1 alpha(spss) 1 alpha(gr) 3 | 9 alpha(andr) 3 | 9 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model berman and brown 1984, cao-al2o3-sio2 melt model. reformulated as macroscopic Aug 21, 2007. casmelt abbreviation Melt full_name liquid 2 | model type: macroscopic 3 | 3 species SiO2 Al2O3 CaO | endmember names 0 0 0 | endmember flags 0 1 .1 0 | cartesian 0 1 .1 0 begin_excess_function w(SiO2 Al2O3 Al2O3 Al2O3) 63617.2 -23.7400 0 term 1 w(SiO2 SiO2 Al2O3 Al2O3) 0.164266d7 -763.870 0 term 2 w(SiO2 SiO2 SiO2 Al2O3) -106635. 28.1300 0 term 3 w(SiO2 CaO CaO CaO) -898693. 240.770 0 term 4 w(SiO2 SiO2 CaO CaO) -350208. -48.6200 0 term 5 w(SiO2 SiO2 SiO2 CaO) -14081.8 -35.4900 0 term 6 w(Al2O3 CaO CaO CaO) -455634. 2.47000 0 term 7 w(Al2O3 Al2O3 CaO CaO) -725166. 255.390 0 term 8 w(Al2O3 Al2O3 Al2O3 CaO) -240215. 26.7000 0 term 9 w(SiO2 SiO2 Al2O3 CaO) -.284791d7 1046.35 0 term 10 w(SiO2 Al2O3 Al2O3 CaO) -.214904d7 641.840 0 term 11 w(SiO2 Al2O3 CaO CaO) 209109. -313.360 0 term 12 end_excess_function 1 1 site molecular entropy model 3 1 3 species, site multiplicity 1 z(SiO2) = 1 SiO2 z(Al2O3) = 1 Al2O3 end_of_model -------------------------------------------------------- begin_model Phase A (Frueh-Green et al., 2004, AGU Geophysical Monograph 144:119-136) originally from Connolly, 1994, Phase diagram applications in earth material sciences, VII Summer Schoool Proceedings, University of Siena). -------------------------------------------------------- A-phase(F) abbreviation phA full_name alphabet-phase 2 | macroscopic 2 phA fphA_q 0 0 endmember flags 0 1 .1 0 | subdivision range, imod = 1 -> asymmetric transform subdivision ideal 1 | 1 site entropy model 2 7. | 2 species, 7 sites pfu z(mg) = 1 phA begin_dqf_corrections dqf(fphA_q) = 0 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Clinohumite (Frueh-Green et al., 2004, AGU Geophysical Monograph 144:119-136) or (Connolly, 1994, Phase diagram applications in earth material sciences, VII Summer Schoool Proceedings, University of Siena). -------------------------------------------------------- Chum(F) abbreviation Chu full_name clinohumite 2 | macroscopic 2 chum fchum_q 0 0 endmember flags 0 1 .1 0 | subdivision range, imod = 1 -> asymmetric transform subdivision ideal 1 | 1 site entropy model 2 9. | 2 species, 9 sites pfu z(mg) = 1 chum begin_dqf_corrections dqf(fchum_q) = 0 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Brucite (Frueh-Green et al., 2004, AGU Geophysical Monograph 144:119-136) or (Connolly, 1994, Phase diagram applications in earth material sciences, VII Summer Schoool Proceedings, University of Siena). -------------------------------------------------------- B(F) abbreviation Br full_name brucite 2 | macroscopic 2 isp(1) br fbr_q 0 0 endmember flags 0 1 .1 0 | subdivision range, imod = 1 -> asymmetric transform subdivision ideal 1 | 1 site entropy model 2 1 | 2 species, 1 site pfu z(mg) = 1 br begin_dqf_corrections dqf(fbr_q) = 0 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Ternary feldspars (Holland and Powell, 2003, CMP, p.492-501) Van Laar Versions. USE THIS MODEL WITH EXTREME CAUTION (AND DON'T USE THE MODEL IF YOU DON'T UNDERSTAND THE IMPLICATIONS OF THIS WARNING): The model has the following difficulties for the Ab-An binary and they undoubtedly cause problems in the ternary: 1) The I1/C1 models are not (and cannot be used to) predict the stable structural state, the I1 model is always more stable than the C1; thus the "stable" state is supposed to be chosen by assuming the C1 model is valid at X(Ab) > 1 - (0.12+0.00038*TK). 2) The model assigns a large positive interaction term for I1 plagioclase, as a result the I1 model gives a solvus that closes at ~ 910 K. At temperatures below the critical temperature plagioclases in vicinity of the prescribed I1/C1 transition (as defined in 1, above), that in principle is supposed to be stable, is metastable with respect to a mixture of I1 and C1 plagioclase. JADC 1/04 MORE NOTES: * These models use the asymmetric formalism as outlined by H & P. * They are complicated by the C1/I1 transition across the plag. join. * Model parameters for the C1 field (Ab-rich plag. component) Wabsan = 25100 - 10.8*T + 0.343*P Wansan = 40000 Wanab = 3100 a-san = 1.0 a-ab = 0.643 a-an = 1.0 Ian = 7030 - 4.66*T * Model parameters for the I1 field (An-rich plag. component) Wabsan = 25100 - 10.8*T + 0.343*P Wansan = 40000 Wanab = 15000 a-san = 1.0 a-ab = 0.643 a-an = 1.0 Iab = 570 - 4.12*T D.Tinkham, 12-18-2003 reformatted and compositional ranges chosen given the following considerations (JADC 12/03): * On the Ab-An join C1 is not stable below ~ 800 K, and has X(An) < 0.4 and I1 is not stable for X(An) < 0.6 * On the Ab-Or join C1 is always stable, with critical composition X(Ab) ~ 0.34 * The above considerations suggest the I1 model could be effectively represented by a binary An-Ab (with X(Ab)>0.2) model; and the C1 model should be split into a model with X(ab) < 0.33 with extensive ternary solution; and a model with X(Ab) > 0.33 with very limited ternary solution. Assuming these relationships are valid the ternary feldspar phase relations are well represented by three solutions models. When these conditions are not met, the HP feldpsar models should not be used. Pl(I1,HP) | solution name. abbreviation Pl full_name ternary-feldspar 2 | model type 3 | number of endmembers san abh an 0 0 0 | endmember flags 0 1 .1 0 | compositional range and resolution of san 0 1 .1 0 | compositional range and resolution of abh begin_excess_function w(san abh) 25100 -10.8 0.338 | pressure coefficient changed from 0.343 to match current TC value, JADC, May 2, 2019. w(san an) 40000 w(abh an) 15000 end_excess_function 1 | 1 site entropy model 3 1 | 3 species, site multiplicity of 1 z(Ca) = 1 an z(K) = 1 san begin_van_laar_sizes alpha(san) 1 alpha(abh) .643 alpha(an) 1 end_van_laar_sizes begin_dqf_corrections dqf(abh) 570 -4.12*T end_dqf_corrections end_of_model -------------------------------------------------------- begin_model | See WARNING above for HP ASF Ternary Feldspar Fsp(C1) abbreviation Fsp full_name ternary-feldspar 2 | model type 3 | number of endmembers an san abh 0 0 0 | endmember flags 0 1 .1 0 | compositional range and resolution of an 0 1 .1 0 | compositional range and resolution of san begin_excess_function w(san abh) = 25100 -10.8 * TK + 0.338 * Pb | pressure coefficient changed from 0.343 to match current TC value, JADC, May 2, 2019. w(san an) = 40000 w(abh an) = 3100 end_excess_function 1 | 1 site entropy model 3 1 | 3 species, site multiplicity of 1 z(Ca) = 1 an z(K) = 1 san begin_van_laar_sizes alpha(san) = 1 alpha(abh) = 0.643 alpha(an) = 1 end_van_laar_sizes begin_dqf_corrections dqf(an) = 7030 - 4.66 * T_K end_dqf_corrections end_of_model -------------------------------------------------------- begin_model HP '03 CMP van Laar Calcite-Magnesite with Dolomite compound formation. oCcM(HP) abbreviation Do full_name carbonate 2 model type 3 number of endmembers cc mag odo 0 0 0 endmember flags 0 1 .1 0 | range and increments on X(cc) 0 1 .1 0 | range and increments on X(mag) begin_excess_function w(cc mag) 35000 w(cc odo) 10250 w(mag odo) 14950 end_excess_function 2 2 site (m1 m2) entropy model 2 .5 2 species on m2, mutiplicity = 1/2 z(m2,Ca) = 1 cc + 1 odo 2 .5 2 species on m1, mult. = 1/2 z(m1,Ca) = 1 cc begin_van_laar_sizes alpha(cc) .5 + 0.000546 *T alpha(mag) 1 alpha(odo) .7 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model Tourmaline model with no Li formulated as a hexary solution. NOTE: ffoit, mfoit, drav and shrl are linearly dependent 1 2 3 x y z _________________________ Mutliplicity 1 3 6 _________________________ Species: 1 drav Na Mg3 Al6 2 uvit Ca Mg3 MgAl5 3 mfoit vac Mg2Al Al6 4 shrl Na Fe3 Al6 5 olen Na Al3 Al6 6 ffoit vac Fe2Al Al6 ________________________ Vincent van Hinsberg Tour(V) | solution name. abbreviation Tour full_name tourmaline 2 | model type: simplicial composition space 6 | 6 species mix on the chemical mixing site drav uvit mfoit | endmember names shrl olen ffoit | 0 0 0 0 0 0 | endmember flags, indicate if the endmember is part of the solution. 0 1 .1 0 | range drav 0 .2 .1 0 | restricted range uvit 0 1 .1 0 | range mfoit 0 1 .1 0 | range shrl 0 1 .1 0 | range olen ideal 3 | configurational entropy: 3 sites 3 1 | 3 species on X, 1 site per formula unit. z(na,x) = 1 drav + 1 shrl + 1 olen z(ca,x) = 1 uvit 3 3. | 3 species on Y, 3 sites per formula unit. z(mg,y) = 1 drav + 1 uvit + 2/3 mfoit z(al,y) = 1/3 mfoit + 1 olen + 1/3 ffoit 2 6. | 2 species on Z, 6 site per formula unit. z(al,z) = 1 drav + 5/6 uvit + 1 mfoit + 1 shrl + 1 olen + 1 ffoit end_of_model -------------------------------------------------------- begin_model dolomite order disorder model Site: 1 2 M1 M2 ____________ Mutliplicity 1 1 ____________ 1 adol Ca Mg Species: 2 bdol CaMg CaMg ___________ DoDo abbreviation Do full_name carbonate 2 model type simplicial composition space 2 number of endmembers adol bdol 0 0 endmember flags 0 1 .1 0 | range and increments on X(cc) ideal 2 2 site (m1 m2) entropy model 2 .5 2 species on m2, mutiplicity = 1/2 z(m2,Ca) = 1/2 bdol 2 .5 2 species on m1, mult. = 1/2 z(m1,Mg) = 1/2 bdol end_of_model begin_model magnesio-wuestite solution, after fabrichnaya '99 Wus(fab) abbreviation Wus full_name wuestite 2 model type: simplicial composition space 2 2 endmembers per wus 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(per wus) 24d3 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(mg) = 1 per end_of_model -------------------------------------------------------- begin_model akimotoite (ilmenite-structure) solution, after fabrichnaya '99 Aki(fab) abbreviation Aki full_name ilmenite 2 model type: simplicial composition space 3 number of endmembers cor aki faki 1 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(aki cor) 60000 W(faki cor) 60000 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 aki z(fe) = 1 faki 2 1 2 species on T site multiplicity = 1. z(al) = 1 cor end_of_model -------------------------------------------------------- begin_model | perovskite solution, after fabrichnaya '99 Pv(fab) abbreviation Pv full_name perovskite 2 model type: simplicial composition space 3 3 endmembers aperov perov fperov 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(perov aperov) 49000 W(fperov aperov) 49000 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 perov z(fe) = 1 fperov 2 1 2 species on T site multiplicity = 1. z(al) = 1 aperov end_of_model -------------------------------------------------------- begin_model | perovskite solution, after oganov. the fppv and | appv endmembers are henry's law ss. Ppv(og) abbreviation Ppv full_name postperovskite 2 model type: simplicial composition space 3 number of endmembers appv ppv fppv 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision ideal 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 ppv z(fe) = 1 fppv 2 1 2 species on T site multiplicity = 1. z(al) = 1 appv end_of_model -------------------------------------------------------- begin_model olivine solution O(stx) abbreviation Ol full_name olivine 2 model type: simplicial composition space 2 2 endmembers fo fa 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(fo fa) 14400 | was 7200 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 fo end_of_model -------------------------------------------------------- begin_model Wad(stx) abbreviation Wad full_name wadleysite 2 | model type: simplicial composition space 2 | 2 endmembers wad fwad 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(wad fwad) 3000 | was 1500. end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 wad end_of_model -------------------------------------------------------- begin_model Ring(stx) abbreviation Ring full_name ringwoodite 2 model type: simplicial composition space 2 2 endmembers ring fring 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(ring fring) 7800 | was 3900 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 ring end_of_model -------------------------------------------------------- begin_model Spinel solution, fixed order! Sp(stx) abbreviation Sp full_name spinel 2 model type: simplicial composition space 2 2 endmembers sp herc 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(sp herc) 28800 | was 7200 end_excess_function 2 2 site entropy model 3 8. 3 species, site multiplicity = 8. z(B,mg) = 1/8 sp z(B,fe) = 1/8 herc 3 4. 3 species, site multiplicity = 4. z(B,mg) = 3/4 sp z(B,fe) = 3/4 herc end_of_model -------------------------------------------------------- begin_model From Stixrude's endmember notation (parenthesis used to indicate disordered site populations), it appears the B site should be split into two M-sites for the 05 paper. Gt(stx) abbreviation Gt full_name garnet 2 model type: simplicial composition space 4 4 endmembers gr alm maj py 0 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(gr), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(alm), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(py), imod = 0 -> cartesian subdivision ideal 3 3 site entropy model 3 3. 3 species, A site multiplicity = 3. z(A,ca) = 1 gr z(A,fe) = 1 alm 2 1 2 species, M1 site multiplicity = 1. z(M1,Mg) = 1 maj 2 1 2 species, M2 site multiplicity = 1. z(M1,Si) = 1 maj end_of_model -------------------------------------------------------- begin_model C2/c(stx) abbreviation C2/c full_name C2/c-pyroxene 2 model type: simplicial composition space 2 2 endmembers c2/c fc2/c 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision ideal 1 1 site entropy model 2 4. 2 species, site multiplicity = 4. z(mg) = 1 c2/c end_of_model -------------------------------------------------------- begin_model From Stixrude's endmember notation (parenthesis used to indicate disordered site populations), it appears disorder is assumed across all 4 M sites, for the 05-07 papers. Opx(stx) abbreviation Opx full_name orthopyroxene 2 model type: simplicial composition space 3 3 endmembers en fs ts 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(en) 0 1 .1 0 | range and resolution for X(fs) ideal 1 1 site entropy model 3 4. 3 species, M site multiplicity = 4. z(M,al) = 1/2 ts z(M,fe) = 1 fs end_of_model -------------------------------------------------------- begin_model Cpx solution, entropy model not specified by stixrude, additionally there is some ambiguity about his excess term, here i assume its Gex = X(M1,Ca)*X(M1,M)*W. Cpx(stx) abbreviation Cpx full_name clinopyroxene 2 model type: simplicial composition space 3 3 endmembers di hed mdi 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(di) 0 1 .1 0 | range and resolution for X(hed) begin_excess_function W(mdi di) 52000 W(mdi hed) 52000 end_excess_function 2 2 site entropy model 2 2. 2 species, M1 site multiplicity = 2. z(M1,Mg) = 1 mdi 2 2. 2 species, M2 site multiplicity = 2. z(M2,Fe) = 1 hed end_of_model -------------------------------------------------------- begin_model Stixrude pers com (10/07) indicates Gex = X(M1,Ca)*X(M1,Mg)*W. JADC 12/07 Corrected excess function to include W(mdi hed), 9/08, JADC. Cpx(stx7) abbreviation Cpx full_name clinopyroxene 2 model type: simplicial composition space 3 3 endmembers di hed mdi 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(di) 0 1 .1 0 | range and resolution for X(hed) begin_excess_function W(mdi di) 52600 W(mdi hed) 52600 end_excess_function 2 2 site entropy model 2 2. 2 species, M1 site multiplicity = 2. z(M1,Mg) = 1 mdi 2 2. 2 species, M2 site multiplicity = 2. z(M2,Fe) = 1 hed end_of_model -------------------------------------------------------- begin_model HP '96 Am Min, Non-ideal quasi ordered omphacite, i.e., compound formation only occurs for omph. The value of wdh appears discrepant with the value in HP '98. The interaction parameters here are from the Omphacite model distributed with the '00 Thermocalc program. JADC 2/03 Added ideal acm+cats, JADC, 1/06. CORRECTIONS: enthalpy of ordering corrected from -16 kJ to -3.5 kJ. JADC, Aug 20, 2003. Modified for non-ideal cats after Zeh et al. (2005) JMG, v. 23, p. 1-17. T. Wagner 2/18/06. Site: 1 2 3 4 5 M2a M2b M1a M1b T1 ____________________________________ Mutliplicity 1/2 1/2 1/2 1/2 1 ____________________________________ 1 Diopside Ca Ca Mg Mg Si Species: 2 Jadeite Na Na Al Al Si 3 Hedenbergit Ca Ca Fe2+ Fe2+ Si 4 Ca-Tschermaks Ca Ca Al Al Al 5 Acmite Na Na Fe3+ Fe3+ Si ___________________________________ Ordered Cpd: 6 Omphacite Na Ca Al Mg Si Omph(HP) abbreviation Cpx full_name clinopyroxene 6 model type: o/d, simplicial composition space 5 number of disordered endmembers di jd cats acm hed 1 | ordered species definition omph = 1/2 jd + 1/2 di delta_g_of_ordering = -35d2 0 0 0 0 0 | endmember flags 0 1 .1 0 | range and resolution of X(di) 0 1 .1 0 | range and resolution of X(jd) 0 1 .1 0 | range and resolution of X(cats) 0 1 .1 0 | range and resolution of X(acm) begin_excess_function w(di jd) 26000 w(omph jd) 16000 w(omph di) 16000 w(omph hed) 17000 w(hed jd) 24000 w(di hed) 4000 | hp 98 give 2.5 kJ w(cats di) 7d3 w(cats hed) 4d3 end_excess_function 5 | 4 site entropy model (m1a, m1b, m2b, m2a) 2 .5 | 2 species on m2a, mutiplicity = 1/2 | WARNING! fractions can only be used in the site | fraction definitions, do not use fractions to specify | site multiplicities in the above line. z(m2a,ca) = 1 di + 1 hed + 1 cats 2 .5 2 species on m2b, mult. = 1/2 z(m2a,na) = 1 jd + 1 acm 4 .5 4 species on m1a, mult = 1/2 z(m1a,mg) = 1 di z(m1a,fe2+) = 1 hed z(m1a,fe3+) = 1 acm 4 .5 4 species on m1b, mult = 1/2 z(m1b,al) = 1 jd + 1 cats z(m1a,fe2+) = 1 hed z(m1a,fe3+) = 1 acm 2 1 2 species on T1 (perhaps Al should be disordered over T1-T2?) z(t1,al) = 1 cats end_of_model -------------------------------------------------------- begin_model HP '96 Am Min, Non-ideal disordered cpx Note HP '98 give Wdh = 2500 j/mol. Configurational entropy model changed (corrected) from 1 site two 2 site model and model reformatted. D. Tinkham, 1/04. Added ideal acm+cats, JADC, 1/06. Modified for non-ideal cats after Zeh et al. (2005) JMG, v. 23, p. 1-17. T. Wagner 2/18/06. Added ideal Cr, PGP Workshop 4/12/06. (folk.uio.no/ninasim/Cr_results.html) Corrected z(m1,al) = 1 jd to z(m1,al) = 1 jd + 1 cats M. Caddick, 5/8/08. converted to 688 format. JADC, 12/19. Cpx(HP) abbreviation Cpx full_name clinopyroxene 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 7 | number of vertices on each simplex esn ccrts cats jd acm hed di 0 1 .1 0 | range and resolution of X(esn): imod = 1 -> assymmetric stretching 0 1 .1 0 | range and resolution of X(ccrts): imod = 1 -> assymmetric stretching 0 1 .1 0 | range and resolution of X(cats): imod = 1 -> assymmetric stretching 0 1 .1 0 | range and resolution of X(jd) 0 1 .1 0 | range and resolution of X(acm): imod = 1 -> assymmetric stretching 0 1 .1 0 | range and resolution of X(hed) begin_excess_function w(jd di) 26d3 w(jd hed) 24d3 w(di hed) 4d3 w(cats di) 7d3 w(cats hed) 4d3 end_excess_function 3 | number of sites in the configurational entropy model M1 | site name 5 1 1 | number of species, effective multiplicity, true multiplicity. z(Fe,M1) = 1 hed z(Al,M1) = 1 jd + 1 cats z(Fe3+,M1) = 1 acm + 1 esn z(Cr,M1) = 1 ccrts z(Mg,M1) = 1 di M2 | site name 2 1 1 | number of species, effective multiplicity, true multiplicity. z(Na,M2) = 1 jd + 1 acm z(Ca,M2) = 1 ccrts + 1 esn + 1 cats + 1 di + 1 hed T1 | site name 2 1 1 | number of species, effective multiplicity, true multiplicity. | ccrts not counted in Al intentionally; | this is nonsense, but keeps consistency with Klemme et al 2011. z(Al,T1) = 1 cats + 1 esn z(Si,T1) = 1 ccrts + 1 di + 1 hed + 1 jd + 1 acm [SiO6] | formula suffix, enter "none" for no suffix. end_of_model -------------------------------------------------------- begin_model Chromite/Spinel, Klemme et al. 2010. To use this model, the fcrm endmember must be excluded from thermodynamic stability calculations. A B _____________ Mutliplicity 1 2 _____________ 1 sp Mg Al Species: 2 herc Fe Al 3 mcrm Mg Cr 4 fcrm_d Fe Cr 5 mft Fe3+ MgFe3+ 6 mt Fe3+ FeFe3+ Ad-hoc incorporation of Fe3+ made here by assuming Mt remains inverse and the associated divalent Mg and Fe is disordered on B. An alternative model would be to assume the divalent B cation is associated with Fe3+. Model lacks excess function for mt-crm, which could be estimated if the solvus is known. JADC, 3/29/13 CrSp | solution name. abbreviation Sp full_name spinel 7 | model type: Reciprocal 2 | 2 independent subcompositions 2 3 | 2 dimensions on each site mcrm fcrm_d sp herc mft mt 1 | 1 dependent endmember fcrm_d = 1 herc + 1 mcrm - 1 sp 0 0 0 0 0 0 | endmember flags, indicate if the endmember is part of the solution. | subdivision model for (binary) site 1 (A): 0 1 .1 0 | range and resolution of X(Mg): imod = 0 -> cartesian | subdivision model for (binary) site 2 (B) 0 1 .1 0 | range and resolution of X(Cr): imod = 0 -> cartesion 0 1 .1 0 begin_excess_function w(sp herc) 7d2 0 0 w(mcrm sp sp) 4208. 1.501 .321e-1 | from Oka et al CMP '84 w(mcrm sp) 19686 0.463 0.0183 | 19686 + 0.0183*P + 0.463*T; w(mcrm herc) 20d3 | guessed solvus end_excess_function 2 | 2 site (a, b) configurational entropy model 3 1 | 3 species on a, 1 site per formula unit. z(a,fe) = 1 herc z(a,mg) = 1 sp + 1 mcrm 5 2. | 5 species on b, 2 site per formula unit. z(b,cr) = 1 mcrm z(b,al) = 1 sp + 1 herc z(b,fe3+) = 1/2 mt + 1/2 mft z(b,mg) = 1/2 mft end_of_model -------------------------------------------------------- begin_model Eskolaite from Chaterjee et al '82 Am Min. added at PGP Workshop on 4/12/06 (folk.uio.no/ninasim/Cr_results.html). notes: w(esk cor cor) corrected to w(esk esk cor). this error was noted by J. A. Padron-Navarta, 3/30/12, but left uncorrected until 11/24/2015; the thermodynamic data in cr_hp02ver.dat as used by Klemme et al. 2009 and Ziberna et al. 2013; to reproduce the calculations in those papers the error must be reinstated. added ideal hem. JADC, 3/29/13 Eskol(C) abbreviation Esk full_name ilmenite 2 | model type: simplicial composition space 3 | number of endmembers esk cor hem | endmember names 0 0 0 | endmember flags 0 1 .1 0 | X(Cr), cartesian 0 1 .1 0 | X(Al), cartesian begin_excess_function w(esk esk cor) -5755. .385 -.38e-1 | Chattejee et al '82 Am Min w(esk cor) 37484 4.334 0.0386 | Chattejee et al '82 Am Min end_excess_function 1 1 site entropy model 3 2. 2 species, site multiplicity 2 z(Al) = 1 cor z(Cr) = 1 esk end_of_model -------------------------------------------------------- begin_model magnesio-wuestite solution, stixrude EPSL 07 also xu et al EPSL 08 Wus(stx7) abbreviation Wus full_name wuestite 2 model type: simplicial composition space 2 2 endmembers per wus 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(per wus) 13d3 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(mg) = 1 per end_of_model -------------------------------------------------------- begin_model akimotoite (ilmenite-structure) solution, stixrude EPSL 07 Aki(stx7) abbreviation Aki full_name ilmenite 2 model type: simplicial composition space 3 3 endmembers cor aki faki 1 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function | Stixrude pers com (10/07) states | Gex = X(One_Site,Al)*X(One_Site,Mg)*W | JADC 12/07 W(aki cor) 66000 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 aki z(fe) = 1 faki 2 1 2 species on T site multiplicity = 1. z(al) = 1 cor end_of_model -------------------------------------------------------- begin_model | perovskite solution, stixrude epsl 07 also xu et al epsl 08 Pv(stx7) abbreviation Pv full_name perovskite 2 model type: simplicial composition space 3 3 endmembers aperov perov fperov 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function | Stixrude pers com (10/07) | Gex = X(Big_Site,Al)*X(Big_Site,Mg)*W | JADC 12/07 W(perov aperov) 12000 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 perov z(fe) = 1 fperov 2 1 2 species on T site multiplicity = 1. z(al) = 1 aperov end_of_model -------------------------------------------------------- begin_model O(stx7) abbreviation Ol full_name olivine 2 model type: simplicial composition space 2 2 endmembers fo fa 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(fo fa) 10.6d3 | was 5.3 kJ/molar site end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 fo end_of_model -------------------------------------------------------- begin_model Wad(stx7) abbreviation Wad full_name wadleysite 2 | model type: simplicial composition space 2 | 2 endmembers wad fwad 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(wad fwad) 12.2d3 | 6.1 kJ/site end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 wad end_of_model -------------------------------------------------------- begin_model Ring(stx7) abbreviation Ring full_name ringwoodite 2 model type: simplicial composition space, macroscopic 2 2 endmembers ring fring 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(ring fring) 7d3 | 3.5 kJ/site end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 ring end_of_model -------------------------------------------------------- begin_model Spinel solution, fixed order! Sp(stx7) abbreviation Sp full_name spinel 2 model type: simplicial composition space, macroscopic 2 2 endmembers sp herc 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(sp herc) 20d3 | was 5 kJ/site end_excess_function 2 2 site entropy model 3 8. 3 species, site multiplicity = 8. z(B,mg) = 1/8 sp z(B,fe) = 1/8 herc 3 4. 3 species, site multiplicity = 4. z(B,mg) = 3/4 sp z(B,fe) = 3/4 herc end_of_model -------------------------------------------------------- begin_model Spinel after White, RW, Powell, R & Clarke, GL (JMG, 2002) DO NOT USE THIS MODEL WITH ds6.33+. USE Sp(HGP) FOR ds6.33+ The THERMOCALC version of Sp(WPC) invokes both imaginary crystallographic sites and the equipartition assumption, this combination is impossible to implement with complete fidelity in Perple_X. Two approximations of this model are available in Perple_X Sp(WPC) and Sp_II(WPC). Sp(WPC) yields results closer to those generated by THERMOCALC. However in 6.9.1+ the stable compositions of Sp(WPC) drift outside the simplicial composition space of the model, this drift has no noticeable consequences other than a number of warnings. Sp_II(WPC) eliminates these warning be defining the a compositional prism that spans the entire composition space allowed by the imaginary sites. This model has two fake sites, each with a mulitplicity of one. Fe3+ Al and Ti mix on the fake A site, and Mg-Fe2+ mix on the fake B site Replaces Sp(WPH), JADC Mar 5, 2009. A B _____________ Mutliplicity 1 1 _____________ 1 sp Al Mg Species: 2 herc Al Fe2+ 3 usp Ti Fe2+ 4 mt Fe3+ Fe2+ Sp(WPC) abbreviation Sp full_name spinel 2 | model type: simplicial composition space 4 | 4 endmembers sp herc usp mt 0 0 0 0 | endmember flags 0 1 .1 0 | X(sp) subdivision range, imod = 0 -> cartesian 0 1 .1 0 | X(herc) subdivision range, imod = 1 -> assymmetric stretching 0 1 .1 0 | X(usp) subdivision range, imod = 1 -> assymmetric stretching begin_excess_function W(herc mt) 18.5d3 W(herc usp) 27d3 W(sp mt) 40d3 W(sp usp) 30d3 end_excess_function 2 2 site entropy model, see comments above 3 1 3 species, site multiplicity of 1 z(a,Fe3+) = 1 mt z(a,Ti) = 1 usp 2 1 2 species, site multiplicity of 1 z(b,Mg) = 1 sp end_of_model -------------------------------------------------------- begin_model Spinel_II after White, RW, Powell, R & Clarke, GL (JMG 20:41-55 2002) The THERMOCALC version of Sp(WPC) invokes both imaginary crystallographic sites and the equipartition assumption, this combination is impossible to implement with complete fidelity in Perple_X. Two approximations of this model are available in Perple_X Sp(WPC) and Sp_II(WPC). Sp(WPC) yields results closer to those generated by THERMOCALC. However in 6.9.1+ the stable compositions of Sp(WPC) drift outside the simplicial composition space of the model, this drift has no noticeable consequences other than a number of warnings. Sp_II(WPC) eliminates these warning be defining the a compositional prism that spans the entire composition space allowed by the imaginary sites. * to prevent Perple_X from rejecting the Ti endmember (usp) of this model because of invalid site populations the musp endmember be specified by the make definition musp_d = 1 usp + 2 sp - 2 herc DQF = 0 in the header of the thermodynamic data file to be used for the calculation, for more information on make definitions see: www.perplex.ethz.ch/perplex_thermodynamic_data_file_body.html#Make_definitions * Converted to reciprocal 688 format solution model by adding mft_d and musp_d, JADC Aug 1, 2020. * Replaces Sp(WPH), JADC Mar 5, 2009. This model has two fake sites, each with a mulitplicity of one. Fe3+ Al and Ti mix on the fake A site, and Mg-Fe2+ mix on the fake B site, a mind boggling contortion. There is no way to reproduce this model exactly in Perple_X because it has no rational expression. In fact, if you wanted to make something worse than equipartition, this is it. Two semi-rational implementations are: 1) assume a distinct subsite B' and reduce the effective multiplicity of B and B' to 1/2 so that the total B' + B site configurational entropy is comparable to the WPC single B site. B B' A _____________________ Mutliplicity 1 1 1 Effective Multiplicity 1/2 1/2 1 _____________________ 1 sp Al Al Mg Species: 2 herc Al Al Fe 3 usp Ti Fe Fe musp Ti Mg Mg 4 mt Fe3+ Fe3+ Fe mft_d Fe3+ Fe3+ Mg B B' A _____________________ Mutliplicity 1 1 1 Effective Multiplicity 1/2 1/2 1 _____________________ 1 sp Al Al Mg Species: 2 herc Al Al Fe 3 usp Ti Fe Fe musp Ti Mg Mg 4 mt Fe3+ Fe3+ Fe mft_d Fe3+ Fe3+ Mg 2) Make usp inverse. This is unsatisfactory in the mt IS inverse, but treated as normal, who knows what usp, etc are. B A ____________________________ Effective Multiplicity 1 1 True Multiplicity 2 1 ____________________________ sp [Al]2 Mg Species: herc [Al]2 Fe usp TiFe Fe musp_d TiMg Mg mt [Fe3+]2 Fe mft_d [Fe3+]2 Mg Scheme 1 is adopted below. The implementation underestimates the site fraction of Ti compared to the WPC formulation and therefore is likely to destabilize Ti-solution compared to the THERMOCALC implementation, the effect seems slight (cf the bl478 benchmark). Sp_II(WPC) abbreviation Sp full_name spinel 688 | model type: 688 format standard model 1 | number of polytopes 2 | number of simplices 2 3 | number of vertices on each simplex musp usp mft_d mt sp herc X_Mg,A 0. 1 .1 0 | imod = 0 -> cartesian subdivision X_Fe,A by difference X_MT,B 0. 1 .1 0 | imod = 0 -> cartesian subdivision X_Fe,B 0. 1 .1 0 | imod = 0 -> cartesian subdivision X_Al,B by difference begin_dependent_endmembers mft_d = 1 mt + 1 sp - 1 herc end_dependent_endmembers begin_excess_function W(herc mt) 18.5d3 W(herc usp) 27d3 W(sp mt) 40d3 W(sp usp) 30d3 end_excess_function 2 | number of sites in configurational entropy model A | site name 2 1 1 | number of species, effective multiplicity, true multiplicity z(Mg,A) = 1 sp + 1 musp z(Fe,A) = 1 herc + 1 usp + 1 mt B | site name 5 1 2 | number of species, effective multiplicity, true multiplicity z(Ti,B) = 1/2 usp + 1/2 musp z(Fe3,B) = 1 mt z(Al,B) = 1 herc + 1 sp z(Mg,B) = 1/2 musp z(Fe,B) = 1/2 usp [cf model comment!]O4 | formula suffix, enter "none" for no suffix. end_of_model -------------------------------------------------------- begin_model CPX in CMASCH marbles (di-en-cats), hence enstatite endmember added. Entered by A Proyer, Aug 27, 2008. based on: HP '96 Am Min, Non-ideal disordered cpx Note HP '98 give Wdh = 2500 j/mol. Configurational entropy model changed (corrected) from 1 site two 2 site model and model reformatted. D. Tinkham, 1/04. Modified for non-ideal cats after Zeh et al. (2005) JMG, v. 23, p. 1-17. T. Wagner 2/18/06. Cpx(m) abbreviation Cpx full_name clinopyroxene 2 | model type sf. 3 | number of endmembers en cats di 0 0 0 | endmember flags | NOTE RESTRICTED RANGES 0 1 .1 0 | range and resolution of X(en): imod = 1 -> assymmetric stretching 0 1 .1 0 | range and resolution of X(cats): imod = 1 -> assymmetric stretching begin_excess_function w(en cats) 24d3 w(en di) 24d3 w(cats di) 7d3 end_excess_function 2 | 2 site (M1, M2) configurational entropy model 2 1 | 2 species on M1, 1 site per formula unit. z(m1,al) = 1 cats 2 1 | 2 species on M2, 1 site per formula unit. z(m2,mg) = 1 en begin_dqf_corrections dqf(en) 8100 -4.5*T end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Olivine in marble; from TC-solution model: Ca only on M2 Entered by A Proyer, Aug 27, 2008. Ol(m) abbreviation Ol full_name olivine 2 | model type: simplicial composition space 2 | 2 endmembers fo mont 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function w(fo mont) 24000 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity of 1 z(Ca) = 1 mont end_of_model -------------------------------------------------------- begin_model Pl(stx8) | Xu et al. EPSL 08, to be used with the stx08ver.dat data generated from that paper. abbreviation Pl full_name binary-feldspar 2 | model type: simplicial composition space 2 | # of endmembers ab an 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function w(an ab) 26d3 end_excess_function 1 | 1 site molecular model: 2 1 z(Na) = 1 ab end_of_model -------------------------------------------------------- begin_model Spinel solution, fixed order! Sp(stx8) abbreviation Sp full_name spinel 2 model type: simplicial composition space 2 2 endmembers sp herc 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(sp herc) 29.6d3 end_excess_function 2 2 site configurational entropy model 3 8. 3 species, site multiplicity = 8. z(B,mg) = 1/8 sp z(B,fe) = 1/8 herc 3 4. 3 species, site multiplicity = 4. z(B,mg) = 3/4 sp z(B,fe) = 3/4 herc end_of_model begin_model -------------------------------------------------------- HP 2011 model, "tro" must be excluded to stabilize this solution model. Po(HP) abbreviation Po full_name pyrrhotite 2 model type: simplicial composition space 2 endmembers trov trot 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(trov trot) = -3190 end_excess_function 1 1 site configurational entropy model 2 1 2 species (Fe, V), site multiplicity = 1. z(M2,V) = 1/8 trov end_of_model -------------------------------------------------------- begin_model O(stx8) abbreviation Ol full_name olivine 2 model type: simplicial composition space 2 2 endmembers fo fa 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(fo fa) 9d3 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 fo end_of_model -------------------------------------------------------- begin_model Wad(stx8) abbreviation Wad full_name wadleysite 2 | model type: simplicial composition space 2 | 2 endmembers wad fwad 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(wad fwad) -8.6d3 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 wad end_of_model -------------------------------------------------------- begin_model Ring(stx8) abbreviation Ring full_name ringwoodite 2 model type: simplicial composition space 2 2 endmembers ring fring 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(ring fring) 8d3 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 ring end_of_model -------------------------------------------------------- begin_model From Stixrude's endmember notation (parenthesis used to indicate disordered site populations), it appears disorder that in the '08 papers the M sites are now split. Opx(stx8) abbreviation Opx full_name orthopyroxene 2 model type: simplicial composition space 4 4 endmembers odi en fs ts 0 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(odi) 0 1 .1 0 | range and resolution for X(en) 0 1 .1 0 | range and resolution for X(fs) begin_excess_function W(odi ts) 43.8d3 W(odi en) 43.8d3 end_excess_function 2 |2 site entropy model, Al on M2 is associated with Mg on M1 and |Ca on M1 is associated with Mg on M2, so reduce M2 site model to |a two species model? 3 2. 3 species, M1 site multiplicity = 2. z(M1,Ca) = 1 odi z(M1,Fe) = 1 fs 3 2. 3 species, M2 site multiplicity = 2. z(M2,Al) = 1 ts z(M2,Fe) = 1 fs end_of_model -------------------------------------------------------- begin_model Cpx(stx8) abbreviation Cpx full_name clinopyroxene 2 model type: simplicial composition space 5 # of endmembers jd di hed mdi cts 0 0 0 0 0 endmember flags 0 1 .1 0 | range and resolution for X(jd) 0 1 .1 0 | range and resolution for X(di) 0 1 .1 0 | range and resolution for X(hed) 0 1 .1 0 | range and resolution for X(mdi) begin_excess_function W(mdi di) 49d3 | Mg-Ca W(mdi hed) 49d3 W(mdi cts) 49d3 W(jd di) 486d2 | Na-Ca W(jd hed) 486d2 W(jd cts) 486d2 end_excess_function 2 2 site entropy model 3 2. 3 species, M1 site multiplicity = 2. z(M1,Mg) = 1 mdi z(M1,Na) = 1 jd 3 2. 3 species, M2 site multiplicity = 2. z(M2,Fe) = 1 hed z(M2,Al) = 1 jd + 1 cts end_of_model -------------------------------------------------------- begin_model akimotoite (ilmenite-structure) solution Aki(stx8) abbreviation Aki full_name ilmenite 2 model type: simplicial composition space 3 3 endmembers cor aki faki 1 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(aki cor) 42000 W(faki cor) 52000 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 aki z(fe) = 1 faki 2 1 2 species on T site multiplicity = 1. z(al) = 1 cor end_of_model -------------------------------------------------------- begin_model Garnet solution, from Xu et al '08. Gt(stx8) abbreviation Gt full_name garnet 2 model type: simplicial composition space 5 # of endmembers gr alm maj py jmaj 0 0 0 0 0 endmember flags 0 1 .1 0 | range and resolution for X(gr), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(alm), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(maj), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(py), imod = 0 -> cartesian subdivision begin_excess_function W(gr maj) 45000 W(gr py) 45000 end_excess_function 2 # of sites for configurational entropy model 4 0 4 independent species, A site multiplicity = 0 => variable multiplicity, means need molar amounts of all species n(A,ca) = 3 gr n(A,fe) = 3 alm n(A,na) = 2 jmaj n(A,mg) = 3 py + 3 maj 3 2. 3 species, B site multiplicity = 2, this is a peculiarity in stixrude's model because he sets r[Al,jmj,B]=2 z(B,Mg) = 1/2 maj z(B,Si) = 1/2 maj end_of_model -------------------------------------------------------- begin_model Ppv(stx8) abbreviation Ppv full_name postperovskite 2 model type: simplicial composition space 3 3 endmembers appv ppv fppv 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(ppv appv) 21000 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 ppv z(fe) = 1 fppv 2 1 2 species on T site multiplicity = 1. z(al) = 1 appv end_of_model -------------------------------------------------------- begin_model Ca-Ferrite solution. CF(stx8) abbreviation CF full_name calcium-ferrite 2 model type: simplicial composition space 3 3 endmembers mfer ffer nfer 0 0 0 0 1 .1 0 0 1 .1 0 ideal 2 number of sites for the entropy model 3 1 3 species, A site multiplicity = 1. z(A,fe) = 1 ffer z(A,mg) = 1 mfer 2 1 al-si mixing on only one "T" site. z(M,Si) = 1 nfer end_of_model -------------------------------------------------------- begin_model Fe2+-Fe3+-Mg pumpellyite Massonne & Willner (EJM, 2008) Ideal Pu(M) abbreviation Pu full_name pumpellyite 2 | model type: simplicial composition space 3 | 3 endmembers pump fpum ffpu 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mg) 0 1 .1 0 | range and resolution for X(Fe2), 0 -> cartesian ideal 1 1 site entropy model 3 1 3 species, site multiplicity = 1. z(Fe) = 1 fpum z(Mg) = 1 pump end_of_model -------------------------------------------------------- begin_model Low Temperature Amphibole from Massonne & Willner (EJM, 2008) Act(M) abbreviation Amph full_name clinoamphibole 2 | model type: simplicial composition space 4 | 4 endmembers tr acti gl mrie 0 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mn) 0 1 .1 0 | range and resolution for X(Fe) 0 1 .1 0 | range and resolution for X(Al) begin_excess_function w(tr gl) 27000 w(acti gl) 27000 w(tr mrie) 15000 w(gl mrie) 15000 end_excess_function 1 1 site entropy model 4 2. 4 species, site multiplicity = 2. z(Fe) = 0 + 1 acti z(Mg) = 0 + 1 tr z(Al) = 0 + 1 gl end_of_model -------------------------------------------------------- begin_model Fe2-Mg-Mn Stilpnomelane from Massonne & Willner (EJM, 2008) Stlp(M) abbreviation Stlp full_name stilpnomelane 2 | model type: simplicial composition space 3 | 3 endmembers stlp mstl mnsp 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mg) 0 1 .1 0 | range and resolution for X(Fe2) ideal 1 1 site entropy model 3 48. 3 species, site multiplicity = 48. z(Fe) = 0 + 1 stlp z(Mg) = 0 + 1 mstl end_of_model -------------------------------------------------------- begin_model Margarite-Muscovite-Paragonite from Massonne & Willner (EJM, 2008) Mica(M) abbreviation Mica full_name white-mica 2 | model type: simplicial composition space 3 | 3 endmembers ma mu pa 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Ca) 0 1 .1 0 | range and resolution for X(K) begin_excess_function w(ma ma pa) 18200 w(pa pa ma) 10000 W(mu pa pa) 19456 1.6544*T -.456100*P W(mu mu pa) 12230 .71044*T .665300*P w(ma mu) 35000 end_excess_function 1 1 site entropy model 3 1 2 species, site multiplicity = 1. z(Ca) = 0 + 1 ma z(K) = 0 + 1 mu end_of_model -------------------------------------------------------- begin_model Mn-Fe-Mg Carpholite from Massonne & Willner (EJM, 2008) Carp(M) abbreviation Carp full_name carpholite 2 | model type: simplicial composition space 3 | 3 endmembers mnca fcar mcar 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mn) 0 1 .1 0 | range and resolution for X(Fe) ideal 1 1 site entropy model 3 1 3 species, site multiplicity = 1. z(Fe) = 0 + 1 fcar z(Mg) = 0 + 1 mcar end_of_model -------------------------------------------------------- begin_model fe-mg sudoite from Massonne & Willner (EJM, 2008) Sud(M) abbreviation Sud full_name sudoite 2 | model type: simplicial composition space 2 | 2 endmembers fsud sud 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Fe) ideal 1 1 site entropy model 2 3. 2 species, site multiplicity = 3. z(Fe) = 0 + 1 fsud end_of_model -------------------------------------------------------- begin_model Magnesite-Siderite-Calcite-Rhodochrosite Carb(M) abbreviation Cc full_name carbonate 2 | model type: simplicial composition space 4 | 4 endmembers cc mag sid rhc 0 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Ca) 0 1 .1 0 | range and resolution for X(Mg) 0 1 .1 0 | range and resolution for X(Fe) begin_excess_function w(cc mag) 35000 w(cc cc sid) 13500 w(cc sid sid) 21000 w(mag sid) 4000 | hp '98 give 4 kJ w(mag rhc) 20000 w(sid rhc) 4000 end_excess_function 1 1 site entropy model 4 1 4 species, site multiplicity = 1. z(Ca) = 0 + 1 cc z(Mg) = 0 + 1 mag z(Fe) = 0 + 1 sid end_of_model -------------------------------------------------------- begin_model Ti-Fe-Mg-Mn-Biotite with compound formation, Powell and Holland '99 Am Min, extended for Mn-solution. reformulated as dual polytope model. JADC, 10/9/2019 NOTES: * Limits added 5/6/2011. JADC. 1 2 3 M1 M2 T2 _________________________ Mutliplicity 1 2 2 _________________________ Dependent: 2 ftbi Ti FeV AlSi 3 tbi Ti MgV AlSi Dependent: 5 Sdph Al Fe AlAl 6 East Al Mg AlAl 7 MnBi Mn Mn AlSi Species: 8 Ann Fe Fe AlSi 9 Phl Mg Mg AlSi ________________________ Ordered Cpd: 10 Obi Fe Mg AlSi | Comments can be placed before character data within a solution | model as long as they are preceded by the comment marker "|", | in general comments should not be placed before numerica data, | but they can be written following numeric data on the same line. Bio(HP) | solution name. abbreviation Bio full_name biotite 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [Mn][Mn] 0 .5 .1 0 [M,T][M] by difference | ---------------------------- | Polytope 1 - 1 simplex 1 | number of simplices 1 | number of vertices on each simplex mnbi | endmembers on the vertices | ---------------------------- | Polytope 2 - 2x3 simplices 2 | number of simplices 2 3 | number of vertices on each simplex | endmembers on the vertices ftbi_i tbi sdph_i east ann phl X_Mg 0 1 .1 0 | X(1,1) is bulk Fe/M X_Fe by difference | Second 3-simplex X_TiTs 0 1 .1 0 | X(2,1) is Ti/[M+T] on M1 X_AlTs 0 1 .1 0 | X(2,2) is Al/[T+M] on M1 X_MBio by difference begin_ordered_endmembers | model types 6 and 8 require data defining the | properties of an ordered "species". this species | is defined as a stoichiometric combination of | two independent endmembers and the enthalpy of | formation of the ordered species from the | these independent endmbers. the format for this | data is | name = num_1 * name_1 + num_2 * name_2 text = enthalpy | where name is the arbitrary name of the ordered | species, num_j is a number or fraction (i.e., two | numbers separated by a '/') and name_j is the | name of a valid endmember. text is arbitrary and | enthalpy is the enthalpy of formation of the ordered species. obi = 2/3 phl + 1/3 ann delta_g_of_ordering = -10.73d3 end_ordered_endmembers begin_dependent_endmembers | model types 7 and 8 (reciprocal solutions) use | internal endmembers that are defined as a | stoichiometric combination of the other endmembers. | the names of these endmembers are arbitrary, but | here dependent endmembers are highlighted by the | suffix "_i", this also serves to distinguish the | endmembers from real equivalents that may be | present in the thermodynamic data file. | the format of this data is | name = num_1 * name_1 + num_2 * name_2 | where num_j is a number or fraction (i.e., two | numbers separated by a '/') and name_j is the | name of a valid endmember. sdph_i = 1 east + 1 ann - 1 obi ftbi_i = 1 tbi + 1/2 ann - 1/2 obi end_dependent_endmembers begin_excess_function | format is W(e1 e1 e2 ...) num1 num2 num3 | where the excess parameter = num1 + num2*T + num3*P | and is multiplied by y(e1)*y(e1)*y(e2)... W(phl ann) 9000 W(phl east) 10000 W(phl obi) 3000 W(phl tbi) -10000 W(ann east) -1000 W(ann obi) 6000. W(ann tbi) 12000 W(obi east) 10000 end_excess_function 3 | Configurational entropy: 3 sites, M1, M2, T1 M2 | site name 5 1 1 | 5 species on M1, 1 site per formula unit. z(Mg) = 1 phl z(Mn) = 1 mnbi z(Al) = 1 east z(Ti) = 1 tbi z(Fe) = 1 ann + 1 obi M2 | site name 4 2 2 | 4 species on M2, 2 sites per formula unit. z(Fe) = 1 ann z(Mn) = 1 mnbi z(Vac) = 1/2 tbi z(Mg) = 1/2 tbi + 1 phl + 1 obi + 1 east T1 | site name 2 2 2 | number of species, effective multiplicity, true multiplicity z(Al,T1) = 1/2 tbi + 1/2 mnbi + 1/2 ann + 1/2 phl +1/2 obi + 1 east z(Si,T1) = 1/2 tbi + 1/2 mnbi + 1/2 ann + 1/2 phl +1/2 obi [Si2O10(OH)2] | formula suffix, enter "none" for no suffix. end_of_model -------------------------------------------------------- begin_model Mica(SGH): Non-reciprocal version of white mica model Mica(SGH1) after: Smye et al (JMG, 2011, 28:753-768) This model requires the make definition: fmu = 1 mu + 1/2 hem - 1/2 cor -30d3 made internal Jan 31, 2026 ma_dqf = 1 ma 3d3 in the thermodynamic data file (e.g., hp02ver.dat), additionally the endmember "ma" must be exlcuded from any calculations that employ this model. WARNING! The computed phase relations in Fig 1 of Smye et al (2011) were computed by imposing the tetrahedral silica content. Thus it is unlikely that the model accurately predicts silica content. JADC, 19/09/11 A M2a M2b T1 M1 ___________________________________ Mutliplicity 1 1 1 2 1 ___________________________________ 1 mu K Al Al AlSi _ 2 pa Na Al Al AlSi _ 3 ma Ca Al Al AlAl _ 4 cel K Mg Al SiSi _ 5 fcel K Fe Al SiSi _ 6 fmu K Al Fe3+ AlSi _ Mica(SGH) abbreviation Mica full_name white-mica 2 | model type: simplex 6 | 6 endmembers mu pa ma cel fmu fcel 0 0 0 0 0 0 0 0 | endmember flags | subdivision model 0 1 .1 0 | range and resolution of X(mu), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(pa), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(ma ), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(cel), imod = 0 -> cartesian subdivision 0 0.3 .1 0 | range and resolution of X(fmu) begin_excess_function W(mu pa) 10120 3.4*T 0.353*P W(mu ma ) 35000 W(mu cel) 0 0.2*P W(mu fcel) 0 0.2*P W(pa cel) 45000 0.25*P W(ma cel) 40000 W(pa fcel) 45000 0.25*P W(ma fcel) 40000 W(pa ma ) 15000 W(pa fmu) 30000 W(ma fmu) 35000 end_excess function 4 | Configurational entropy: 4 sites, A, M2a, M2b, T1. 3 1 | 3 species on A, 1 site per formula unit. z(a,k) = 1 mu + 1 cel + 1 fcel + 1 fmu z(a,na) = 1 pa 3 1 | 3 species on M2a, 1 site per formula unit. z(m2a,al) = 1 mu + 1 pa + 1 ma + 1 fmu z(m2a,mg) = 1 cel 2 1 | 2 species on M2b, 1 site per formula unit. z(m2b,fe) = 1 fmu 2 2. | 2 species on T1, 2 sites per formula unit. z(t,al) = 1/2 mu + 1/2 pa + 1/2 fmu + 1 ma begin_van_laar_sizes alpha(mu) 0.63 alpha(pa) 0.37 alpha(ma ) 0.63 alpha(cel) 0.63 alpha(fmu) 0.63 alpha(fcel) 0.63 end_van_laar_sizes begin_dqf_corrections dqf(ma) 3000 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Mn-Fe-Mg-Fe3+ Chloritoid after: Smye et al (JMG, 2011, 28:753-768) This model requires the make definition: octd = 1 fctd + 1/4 hem - 1/4 cor 125d2 0 0 in the thermodynamic data file (e.g., hp02ver.dat). NOTE: this version is formulated with 5 Oxygen formula unit, NOT the 10 oxygen formula unit used by Smye et al (2011). JADC, 19/09/11 M1a M1b _______________ Mutliplicity 1/2 1 _______________ mctd Al Mg fctd Al Fe mnctd Al Mn octd Fe3+ Fe Ctd(SGH) abbreviation Ctd full_name chloritoid 2 | model type 4 | endmembers octd mnctd fctd mctd 0 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Fe3+), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(Mn), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(Fe), imod = 0 -> cartesian subdivision begin_excess_function w(mctd fctd ) 500 w(mctd mnctd) 500 w(fctd mnctd) 500 end_excess_function 2 2 site entropy model 2 .5 2 species on M1a, site multiplicity = 1/2 z(Fe) = 1 octd 3 1 3 species on M1b, site multiplicity = 1 z(Mn) = 1 mnctd z(Mg) = 1 mctd end_of_model -------------------------------------------------------- begin_model Mn-Fe-Mg-Fe3+ Carpholite after: Smye et al (JMG, 2011, 28:753-768) This model requires the make definitions: ocar = 1 fcar + 1/2 hem - 1/2 cor 45d3 0 0 mncar = 1 mcar + 1 mang - 1/2 cor 30d3 0 0 in the thermodynamic data file (e.g., hp02ver.dat). JADC, 19/09/11 M2 M1 _______________ Mutliplicity 1 1 _______________ mcar Al Mg fcar Al Fe mncar Al Mn ocar Fe3+ Fe Carp(SGH) abbreviation Crp full_name carpholite 2 | model type 4 | endmembers mncar ocar fcar mcar 0 0 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function w(mcar fcar ) 1d3 w(mcar mncar) 1d3 w(fcar mncar) 1d3 end_excess_function 2 2 site entropy model 2 .5 2 species on M2, site multiplicity = 1 z(Fe) = 1 ocar 3 1 3 species on M1, site multiplicity = 1 z(Mn) = 1 mncar z(Mg) = 1 mcar end_of_model -------------------------------------------------------- begin_model Ca-Fe2+-Mg-Al-Fe3+ Garnet model after White, Pomroy, Powell & Holland (JMG, 2005) In calculations that use this model, the andradite endmember ("andr") in the Holland and Powell data base must be excluded. This model also requires the following make definition for khoharite in the thermodynamic data file: 1 2 X Y _____________ Mutliplicity 3 2 _____________ Dependent: fkho_i Fe Fe3+ Dependent: kho_q Mg Fe3+ Dependent: fmn_i Mn Fe3+ andr_i Ca Fe3+ spss Mn Al alm Fe Al py Mg Al gr Ca Al ____________ When originally entered 9/20/11, this model was formulated (incorrectly) in terms of andradite, it was corrected to use the khoharite endmember 10/24/11. JADC definition of andr_i corrected from 1 kho_q + 1 py - 1 gr to 1 kho_q - 1 py + 1 gr on 1/4/13. JADC Gt(WPPH) abbreviation Gt full_name garnet 7 | model type 2 | the number of independent subcompositions, reciprocal solution if > 1. 4 2 | 4 species on site 1, 2 species on site 2. | M2 and M1 can be identified as sites 1 and 2, respectively. the | species that mix on site 1 are Mn-Mg-Fe-Ca and the species that mix on | site 2 are Al-Fe3+. spss alm py gr | endmember names fmn_i fkho_i kho_q andr_i 3 | number of dependent endmembers andr_i = 1 kho_q - 1 py +1 gr fkho_i = 1 kho_q + 1 alm -1 py fmn_i = 1 kho_q + 1 spss -1 py 0 0 0 0 0 0 0 0 | endmember flags 0 .2 .1 0 | imod = 0 -> cartesian subdivision (xmn) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision (xfe) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision (xmg) on X 0 1 .1 0 | imod = 0 -> cartesian subdivision x(fe3+) on Y begin_excess_function w(alm py) 2.5d3 w(alm kho_q) 22.5d3 w(py gr) 33d3 w(gr kho_q) -7d3 w(spss kho_q) 20d3 end_excess_function 2 |2 site entropy model 4 3. |4 species, site multiplicity 3 z(x,mn) = 1 spss z(x,fe) = 1 alm z(x,ca) = 1 gr 2 2. |2 species, site multiplicity 2 z(y,al) = 1 spss + 1 alm + 1 py + 1 gr begin_dqf_corrections dqf(kho_q) = 40d3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model CLINOAMPHIBOLE: Diener et al, JMG 2011 25:631-656, modified from Diener et al, JMG 2008. --------------------------------------------------------- Reformulated as a 2-polytope 688 format model. JADC, 11/19 The composition space of the complete (published) model corresponds to: [A][M][M,T,MT][M,C,N] where the brackets indicate the A, M1, M2, and M4 sites and A = Na, K, V T = Al, Fe3 MT = MAl, MFe3 C = Ca M = Mg, Fe N = Na The composition space of the model here has been simplified to [A][M][M,T,MT][M,C,N] where the brackets indicate the A, M1, M2, and M4 sites and A = Na, K, V T = Al, Fe3 MT = MAl C = Ca M = Mg, Fe N = Na Refer to to cAmph(G) for an (commented out) example with MT = MAl, MFe3. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = Glaucophane = [M][T][N] => 2*2 Polytope 2 = ACMT-Amph = [M][M,T,MT][M,C] => 2*2*4 T is generalized Tschermaks [M][T][M,C] MT is generalized pargasite [M][MT][N] N is generalized glaucophane [M][T][N] M on M2 is generalized amphibole [M][M][M,C] -------------------------------------------------------- earlier corrections/revisions 10/07 z(m2a,al) corrected. Y Y. Podladchikov 3/08 z(m1a,mg),z(m4,ca),z(m4,mg), and fged corrected, T1 multiplicity reduced to 1 Enthalpies of ordering corrected. JADC. 6/08 z(m1,mg) corrected to include mrb, second limit equations for cammo1 and cammo2 corrected accordingly. corrected enthalpies for cammo1 and cammo2. 8/31/11 2011 DP model revisions entered by MJC. 1/24/17 fgrk-grk exchange added 3/ 8/18 dqfs for cumm and grun moved from model to data file 11/2/18 reformulated as 2x2x5 prism, adding Ca-free parg and Ts, and Ca-bearing grk exchanges. JADC. -------------------------------------------------------- independent endmembers and site populations: A M1 M2 M4 T1 _________________________________________ Mutliplicity 1 3 2 2 1(4)* _________________________________________ tr Vac Mg Mg Ca Si_Si ts_q Vac Mg Al Ca Al_Si parg Na Mg Mg_Al Ca Al_Si gl Vac Mg Al Na Si_Si cumm Vac Mg Mg Mg Si_Si grun Vac Fe Fe Fe Si_Si mrb_q Vac Mg Fe3+ Na Si_Si cammo1 Vac Mg Fe Fe Si_Si ordered cammo2 Vac Fe Mg Fe Si_Si ordered *T1 has a true multiplicity of 4, H&P previously used an effective multiplicity of 2; however in Diener et al. '07 the multiplicity has been reduced to 1 JADC 9/07. 2 NaCa-1(M4) = (Gl-Ts) 2 CaMg-1(M4) = (Tr-Cumm) 2 Fe3Al-1(M2) = (Rb-Gl) 3 FeMg-1(M1) = (grun - a) 2 FeMg-1(M2) = (grun - b) 2 FeMg-1(M4) = (-grun - cumm + a + b) ---------------------------------------------------------------------------------------- cAmph(DP) abbreviation Amph full_name clinoamphibole 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes GlRb-Amph 0 .2 .1 0 | X([ ][M][Fe3,Al][N]), Mg-Fe-Al-Fe3+ "glaucophane" ACMT-Amph by difference | X([A][M][M,T,MT][M,C]), K-Na-Mg-Fe-Al-Ti-Fe3+ "pargasite" | ------------------------------------------- | Polytope 1 [ ][M][T ][N] (generalized glaucophane) 2 | number of simplices 2 2 | number of vertices on each simplex | Simplex 1 is M | Simplex 2 is T | endmembers on the vertices, index on the first simplex | changes fastest, last slowest | => endmember 11 is [ ][Mg][Fe3][Na] | => endmember 21 is [ ][Fe][Fe3][Na] | => endmember 12 is [ ][Mg][Al ][Na] | => endmember 22 is [ ][Fe][Al ][Na] mrb_q frb_d | Fe3-glaucophane, aka riebeckite gl fgl_d | Al-glaucophane | First (1d) simplex X_Mg1 0 1 .1 0 | X(1,1) = bulk Mg/M X_Fe1 by difference | Second (3d) simplex X_rb 0 1 .1 0 | X(2,1) = Fe3/T' on M2, T' = Al + Fe3+ X_gl by difference | ------------------------------------------- | Polytope 2 [A][M][M,T,MT][M,C] 3 | number of simplices 2 2 4 | number of vertices on each simplex | Simplex 1 is bulk Mg/M | Simplex 2 is M/[C + M] on M4 | Simplex 3 is M, T, MT on M2 | endmembers on the vertices, index on the first simplex | changes fastest, last slowest ts_q fcts_d | M-Ca-Al_tschermaks mts_d fts_d parg fcparg_d | M-Ca-Na-Al_parg mparg_d fparg_d mcgrk_d fcgrk_d | M-Ca-Fe3+_tschermaks (Na-free riebeckite) mgrk_d fgrk_d cumm grun | M-Ca divalent amphibole tr ftr_d | First (1d) simplex X_Mg 0 1 .1 0 | X(1,1) = bulk Mg/M X_Fe by difference | Second (1d) simplex X_M_M2 0 1 .1 0 | X(2,1) = M/[Ca + M] on M4 X_C_M2 by difference | Third simplex X_AlTs 0 1 .1 0 | X(3,1) = X(Al-Ts) X_NaPg 0 1 .1 0 | X(3,2) = X(Na-Al-Parg) X_FeTs 0 1 .1 0 | X(3,9) = X(Fe3-Ts) X_MCamph by difference begin_ordered_endmembers cammo1 = 3/7 cumm + 4/7 grun delta_g_of_ordering = -9.5d3 cammo2 = 2/7 cumm + 5/7 grun delta_g_of_ordering = -11.7d3 | these enthalpies differ from the published values | becuase Thermocalc reactions are specified in terms of the | endmember without "DQF" corrections, while in Perple_X | endmember include any DQF corrections. end_ordered_endmembers begin_dependent_endmembers ftr_d = 1 tr + 2 grun - 1 cammo1 - 1 cammo2 fcparg_d = 1 parg + 3/2 grun - 1 cammo1 - 1/2 cammo2 fparg_d = 1 parg + 1/2 grun + 1/2 cammo2 - 1 tr mparg_d = 1 parg - 1 tr + 1 cumm fcts_d = 1 ts_q + 1 grun - 1 cammo1 fts_d = 1 ts_q - 1 tr + 1 cammo2 mts_d = 1 ts_q - 1 tr + 1 cumm fgl_d = 1 gl + 1 grun - 1 cammo1 frb_d = 1 mrb_q + 1 grun - 1 cammo1 mcgrk_d = 1 mrb_q + 1 ts_q - 1 gl mgrk_d = 1 mrb_q + 1 ts_q - 1 gl - 1 tr + 1 cumm fgrk_d = 1 mrb_q + 1 ts_q - 1 gl - 1 tr + 1 cammo2 fcgrk_d = 1 mrb_q + 1 ts_q - 1 gl - 1 cammo1 + 1 grun end_dependent_endmembers begin_excess_function w(mrb_q tr) 52e3 w(mrb_q ts_q) 20e3 w(mrb_q parg) 40e3 w(mrb_q cumm) 80e3 w(mrb_q grun) 91e3 w(mrb_q cammo1) 80e3 w(mrb_q cammo2) 90e3 w(cammo2 tr) 63e3 w(cammo2 ts_q) 72.5e3 w(cammo2 parg) 94.8e3 w(cammo2 gl) 111.2e3 w(cammo2 cumm) 23e3 w(cammo2 grun) 8e3 w(cammo2 cammo1) 20e3 w(cammo1 tr) 57e3 w(cammo1 ts_q) 70e3 w(cammo1 parg) 94.8e3 w(cammo1 gl) 100e3 w(cammo1 cumm) 18e3 w(cammo1 grun) 12e3 w(grun tr) 75e3 w(grun ts_q) 80e3 w(grun parg) 106.7e3 w(grun gl) 113.5e3 w(grun cumm) 33e3 w(cumm tr) 45e3 w(cumm ts_q) 70e3 w(cumm parg) 90e3 w(cumm gl) 100e3 w(gl tr) 65e3 w(gl ts_q) 25e3 w(gl parg) 50e3 w(parg tr) 25e3 w(parg ts_q) -40e3 w(ts_q tr) 20e3 end_excess_function 5 | number of sites in configurational entropy model (A, M1, M2, M4, T1) A | site name 2 1 1 | number of species, effective multiplicity, true multiplicity z(Na,A) = 1 parg z(Vac,A) = 1 cumm + 1 grun + 1 tr + 1 ts_q + 1 gl + 1 cammo1 + 1 cammo2 + 1 mrb_q M1 | site name 2 3 3 | number of species, effective multiplicity, true multiplicity z(Fe,M1) = 1 grun + 1 cammo2 z(Mg,M1) = 1 cumm + 1 ts_q + 1 parg + 1 mrb_q + 1 gl + 1 cammo1 + 1 tr M2 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Fe,M2) = 1 grun + 1 cammo1 z(Fe3,M2) = 1 mrb_q z(Al,M2) = 1 ts_q + 1/2 parg + 1 gl z(Mg,M2) = 1/2 parg + 1 tr + 1 cumm + 1 cammo2 M4 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Mg,M4) = 1 cumm z(Na,M4) = 1 gl + 1 mrb_q z(Fe,M4) = 1 grun + 1 cammo1 + 1 cammo2 z(Ca,M4) = 1 ts_q + 1 parg + 1 tr T1 | site name 2 1 4 | number of species, effective multiplicity, true multiplicity z(Al,T1) = 1/2 ts_q + 1/2 parg z(Si,T1) = 1/2 ts_q + 1/2 parg + 1 gl + 1 mrb_q + 1 tr + 1 grun + 1 cammo1 + 1 cammo2 + 1 cumm [Si4O22(OH)2] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(tr) 1 alpha(parg) 1.7 alpha(ts_q) 1.5 alpha(gl) 0.8 alpha(cumm) 1 alpha(grun) 1 alpha(cammo1) 1 alpha(cammo2) 1 alpha(mrb_q) 0.8 end_van_laar_sizes begin_dqf_corrections DQF(ts_q) = 10000 | ds5/ds6 DQF(gl) = -3000 | ds6 DQF(parg) = -10000 | ds6 DQF(grun) = -3000 | ds6 DQF(mrb_q) = 0 | ds6 | DQF(cumm) = 0 | ds6 | DQF(mrb_q) = 8000 | ds5 | DQF(gl) = 3000 | ds5 | DQF(parg) = 15000 | ds5 | DQF(grun) = -5000 | ds5 | DQF(cumm) = -6400 | ds5 end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model CLINOAMPHIBOLE, cAmph_I(DP): Diener et al, JMG 2011 25:631-656, modified from Diener et al, JMG 2008. --------------------------------------------------------- Reformulated as a 4-subcomposition 688 format model. JADC, 3/20 The composition space of the complete (published) model corresponds to: [A][M][M,T,MT][M,C,N,MN,CN] where the brackets indicate the A, M1, M2, and M4 sites (for brevity the dependent T1 and OH site chemistry has been dropped) and A = Na, V T = Al, Fe3 MT = MAl, MFe3 C = Ca M = Mg, Fe N = Na MN = MNa CN = CNa NOTE: the model composition space below has been simplified by commenting out some of the possible endmembers. It is probable that VERTEX/MEEMUM will run out of dynamic memory if the full model is restored, if this occurs uncomment the low_reach option keyword at the end of this model. The low_reach option dramatically reduces the memory and time required for calculations, but also causes a significant deterioration in phase diagram quality. The composition space of the model is formed as a mixture of four independent subcompositions: tschermakitic amphibole: {Mg,Fe}*[Al,Fe3]*[M,C,N] = 12 endmembers pargasitic amphibole: {Ng,Fe}*[N]*[MAl,MFe3]*[M,C] = 8 endmembers piglitic amphibole: {Mg,Fe}*[N,CN,MN]*[N]*[Al,Fe3] = 12 endmembers divalent amphibole: {Ng,Fe}*[M,C] = 4 endmembers => 36 endmembers + 2 ordered endmembers where T1 = f(N,M4) -------------------------------------------------------- earlier corrections/revisions 10/07 z(m2a,al) corrected. Y Y. Podladchikov 3/08 z(m1a,mg),z(m4,ca),z(m4,mg), and fged corrected, T1 multiplicity reduced to 1 Enthalpies of ordering corrected. JADC. 6/08 z(m1,mg) corrected to include mrb, second limit equations for cammo1 and cammo2 corrected accordingly. corrected enthalpies for cammo1 and cammo2. 8/31/11 2011 DP model revisions entered by MJC. 1/24/17 fgrk-grk exchange added 3/ 8/18 dqfs for cumm and grun moved from model to data file 11/2/18 reformulated as 2x2x5 prism, adding Ca-free parg and Ts, and Ca-bearing grk exchanges. JADC. -------------------------------------------------------- NOTE to use this model the following endmembers must be specified with make definitions in the thermodynamic data file: -------------------------------------------------------- independent endmembers and site populations: A M1 M2 M4 T1 _________________________________________ Mutliplicity 1 3 2 2 1(4)* _________________________________________ tr Vac Mg Mg Ca Si_Si ts_q Vac Mg Al Ca Al_Si parg Na Mg Mg_Al Ca Al_Si gl Vac Mg Al Na Si_Si cumm Vac Mg Mg Mg Si_Si grun Vac Fe Fe Fe Si_Si mrb_q Vac Mg Fe3+ Na Si_Si cammo1 Vac Mg Fe Fe Si_Si ordered cammo2 Vac Fe Mg Fe Si_Si ordered *T1 has a true multiplicity of 4, H&P previously used an effective multiplicity of 2; however in Diener et al. '07 the multiplicity has been reduced to 1 JADC 9/07. 2 NaCa-1(M4) = (Gl-Ts) 2 CaMg-1(M4) = (Tr-Cumm) 2 Fe3Al-1(M2) = (Rb-Gl) 3 FeMg-1(M1) = (grun - a) 2 FeMg-1(M2) = (grun - b) 2 FeMg-1(M4) = (-grun + cumm + a + b) ---------------------------------------------------------------------------------------- cAmph_I(DP) abbreviation Amph full_name clinoamphibole 688 | model type: 688 format standard model 4 | number of subcompositions | polyhedron names and composite composition space subdivision schemes Ts-Amph 0 .5 .1 0 | X([ ][M][T][M,C,N]) Pg-Amph 0 .5 .1 0 | X([A][M][MT][M,C,N]) Pgl-Amph 0 .1 .1 0 | X([A][M][T][MN,CN,N]) Trem-Amph by difference | X([ ][M][M][MC]) | ------------------------------------------- | subcomposition 1 [ ][M][T][M,C,N] 3 | number of simplices 2 2 3 | number of vertices on each non-zero-d simplex mrb_q frb_d | [ ][M][Fe][N] Fe3-glaucophane, aka riebeckite gl fgl_d | [ ][M][Al][N] Al-glaucophane mgrk_d fgrk_d | [ ][M][Fe][M] Fe3-Ts mts_d fts_d | [ ][M][Al][M] Al-Ts mcgrk_d fcgrk_d | [ ][M][Fe][C] M-Ca-Fe3+_tschermaks (Na-free riebeckite) ts_q fcts_d | [ ][M][Al][C] Al-cTs | First simplex X_Mg1 0 1 .1 0 | X(1,1) = bulk Mg/M X_Fe1 by difference | 2nd simplex z_Fe3 0 1 .1 0 | T on M2 z_Al by difference | 3rd simplex z_Na 0 .2 .1 0 | M,C,N on M4 z_M 0 .7 .1 0 | z_Ca by difference | ------------------------------------------- | subcomposition 2 [A][M][MT][M,C] 3 | number of simplices 2 2 2 | number of vertices on each simplex | Simplex 1 {Mg,Fe} pseudo-site that defines bulk Mg/Fe | Simplex 2 {Fe3,Al} on M2 | Simplex 3 {M,C} on M4 mparg_d fparg_d | [N][M][MAl][M] Na-Al-M parg parg fcparg_d | [N][M][MAl][C] Na-Al-Ca parg ompg_d ofpg_d | [N][M][MFe][M] Na-Fe3+-M parg* ocmpg_d ocfpg_d | [N][M][MFe][C] Na-Fe3-Ca parg* | 1st simplex x_Mg 0 1 .1 0 | bulk Mg/M x_Fe by difference | 2nd simplex z_Fe3 0 .7 .1 0 | T on M2 z_Al by difference | 3rd simplex z_M 0 .1 .1 0 | M,C on M4 z_C by difference | ------------------------------------------- | Polyhedron 3 [Mg,Fe]*[N,CN,MN]*[K,N]*[Al,Ti,Fe3] 3 | number of simplices 2 3 2 | number of vertices on each simplex [Mg,Fe]*[N,CN,MN]*[N]*[Al,Fe3] nmfn_d nffn_d | [N][M][MFe][N] N-Fe-N-Pgl* nmfmn_d nfffn_d | [N][M][Fe][MN] N-Fe-MN-Pgl* nmfcn_d nffcn_d | [N][M][Fe][CN] N-Fe-CN-Pgl* nman_d nfan_d | [N][M][MAl][N] N-Al-N-Pgl* nmamn_d nfafn_d | [N][M][Al][MN] N-Al-MN-Pgl* nmacn_d nfacn_d | [N][M][Al][CN] N-Al-CN-Pgl* | 1st simplex x_Mg 0 1 .1 0 | bulk Mg/M x_Fe by difference | 2nd simplex z_MN 0 1 .1 0 | MN,CN,N on M4 z_CN 0 1 .1 0 z_N by difference | 3rd simplex z_Fe3 0 .7 .1 0 | T on M2 z_Al by difference | ------------------------------------------- | Polyhedron 4 [Mg,Fe]*[ ][M][M][M,C] 2 | number of simplices 2 2 | number of vertices on each non-zero-d simplex cumm grun | M-Ca divalent amphibole tr ftr_d | First simplex x_Mg,2 0 1 .1 0 | X(1,1) = bulk Mg/M x_Fe,2 by difference | Second simplex z_M 0 1 .1 0 | [ ][M][M][M] z_C by difference | [ ][M][M][C] begin_ordered_endmembers cammo1 = 3/7 cumm + 4/7 grun delta_g_of_ordering = -9.5d3 cammo2 = 2/7 cumm + 5/7 grun delta_g_of_ordering = -11.7d3 | these enthalpies differ from the published values | becuase Thermocalc reactions are specified in terms of the | endmember without "DQF" corrections, while in Perple_X | endmember include any DQF corrections. end_ordered_endmembers begin_dependent_endmembers ftr_d = 1 tr + 2 grun - 1 cammo1 - 1 cammo2 fcparg_d = 1 parg + 3/2 grun - 1 cammo1 - 1/2 cammo2 fparg_d = 1 parg + 1/2 grun + 1/2 cammo2 - 1 tr mparg_d = 1 parg - 1 tr + 1 cumm fcts_d = 1 ts_q + 1 grun - 1 cammo1 fts_d = 1 ts_q - 1 tr + 1 cammo2 mts_d = 1 ts_q - 1 tr + 1 cumm fgl_d = 1 gl + 1 grun - 1 cammo1 frb_d = 1 mrb_q + 1 grun - 1 cammo1 mcgrk_d = 1 mrb_q + 1 ts_q - 1 gl mgrk_d = 1 mrb_q + 1 ts_q - 1 gl - 1 tr + 1 cumm fgrk_d = 1 mrb_q + 1 ts_q - 1 gl - 1 tr + 1 cammo2 fcgrk_d = 1 mrb_q + 1 ts_q - 1 gl - 1 cammo1 + 1 grun | M or Ca on M4, Fe3 on M2 ocmpg_d = 1 parg + 1/2 mrb_q - 1/2 gl ocfpg_d = 1 parg + 3/2 grun - 1 cammo1 - 1/2 cammo2 + 1/2 mrb_q - 1/2 gl ompg_d = 1 parg - 1 tr + 1 cumm + 1/2 mrb_q - 1/2 gl ofpg_d = 1 parg + 1/2 grun + 1/2 cammo2 - 1 tr + 1/2 mrb_q - 1/2 gl | piglitic endmembers nfan_d = 1 parg + 3/2 grun - 1 cammo1 - 1/2 cammo2 + 1 gl - 1 ts_q nman_d = 1 parg + 1 gl - 1 ts_q nmacn_d = 1 parg + 1/2 gl - 1/2 ts_q nfafn_d = 1 parg + 1/2 grun + 1/2 cammo2 - 1 tr + 1/2 gl - 1/2 cammo1 nmamn_d = 1 parg - 1 tr + 1/2 cumm + 1/2 gl nfacn_d = 1 parg + 3/2 grun - 1 cammo1 - 1/2 cammo2 + 1/2 gl - 1/2 ts_q nmfn_d = 1 parg + 1/2 mrb_q + 1/2 gl - 1 ts_q nffn_d = 1 parg + 3/2 grun - 1 cammo1 - 1/2 cammo2 + 1/2 mrb_q + 1/2 gl - 1 ts_q nmfmn_d = 1 parg - 1 tr + 1/2 cumm + 1/2 mrb_q nfffn_d = 1 parg + 1/2 grun + 1/2 cammo2 - 1 tr + 1/2 mrb_q - 1/2 cammo1 nmfcn_d = 1 parg + 3/2 grun - 1 cammo1 - 1/2 cammo2 + 1/2 mrb_q - 1/2 ts_q nffcn_d = 1 parg + 3/2 grun - 1 cammo1 - 1/2 cammo2 + 1/2 mrb_q - 1/2 ts_q end_dependent_endmembers begin_excess_function w(mrb_q tr) 52e3 w(mrb_q ts_q) 20e3 w(mrb_q parg) 40e3 w(mrb_q cumm ) 80e3 w(mrb_q grun) 91e3 w(mrb_q cammo1) 80e3 w(mrb_q cammo2) 90e3 w(cammo2 tr) 63e3 w(cammo2 ts_q) 72.5e3 w(cammo2 parg) 94.8e3 w(cammo2 gl) 111.2e3 w(cammo2 cumm ) 23e3 w(cammo2 grun) 8e3 w(cammo2 cammo1) 20e3 w(cammo1 tr) 57e3 w(cammo1 ts_q) 70e3 w(cammo1 parg) 94.8e3 w(cammo1 gl) 100e3 w(cammo1 cumm ) 18e3 w(cammo1 grun) 12e3 w(grun tr) 75e3 w(grun ts_q) 80e3 w(grun parg) 106.7e3 w(grun gl) 113.5e3 w(grun cumm ) 33e3 w(cumm tr) 45e3 w(cumm ts_q) 70e3 w(cumm parg) 90e3 w(cumm gl) 100e3 w(gl tr) 65e3 w(gl ts_q) 25e3 w(gl parg) 50e3 w(parg tr) 25e3 w(parg ts_q) -40e3 w(ts_q tr) 20e3 end_excess_function 5 | number of sites in configurational entropy model (A, M1, M2, M4, T1) A | site name 2 1 1 | number of species, effective multiplicity, true multiplicity z(Na,A) = 1 parg z(Vac,A) = 1 cumm + 1 grun + 1 tr + 1 ts_q + 1 gl + 1 cammo1 + 1 cammo2 + 1 mrb_q M1 | site name 2 3 3 | number of species, effective multiplicity, true multiplicity z(Fe,M1) = 1 grun + 1 cammo2 z(Mg,M1) = 1 cumm + 1 ts_q + 1 parg + 1 mrb_q + 1 gl + 1 cammo1 + 1 tr M2 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Fe,M2) = 1 grun + 1 cammo1 z(Fe3,M2) = 1 mrb_q z(Al,M2) = 1 ts_q + 1/2 parg + 1 gl z(Mg,M2) = 1/2 parg + 1 tr + 1 cumm + 1 cammo2 M4 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Mg,M4) = 1 cumm z(Na,M4) = 1 gl + 1 mrb_q z(Fe,M4) = 1 grun + 1 cammo1 + 1 cammo2 z(Ca,M4) = 1 ts_q + 1 parg + 1 tr T1 | site name 2 1 4 | number of species, effective multiplicity, true multiplicity z(Al,T1) = 1/2 ts_q + 1/2 parg z(Si,T1) = 1/2 ts_q + 1/2 parg + 1 gl + 1 mrb_q + 1 tr + 1 grun + 1 cammo1 + 1 cammo2 + 1 cumm [Si4O22(OH)2] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(tr) 1 alpha(parg) 1.7 alpha(ts_q) 1.5 alpha(gl) 0.8 alpha(cumm) 1 alpha(grun) 1 alpha(cammo1) 1 alpha(cammo2) 1 alpha(mrb_q) 0.8 end_van_laar_sizes begin_dqf_corrections DQF(ts_q) = 10000 | ds5/ds6 DQF(gl) = -3000 | ds6 DQF(parg) = -10000 | ds6 DQF(grun) = -3000 | ds6 DQF(mrb_q) = 0 | ds6 | DQF(cumm) = 0 | ds6 | DQF(mrb_q) = 8000 | ds5 | DQF(gl) = 3000 | ds5 | DQF(parg) = 15000 | ds5 | DQF(grun) = -5000 | ds5 | DQF(cumm) = -6400 | ds5 end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model -------------------------------------------------------- begin_model ORTHOAMPHIBOLE: Diener et al, JMG 2011 25:631-656, modified from Diener et al, JMG 2008 NOTE!! to use this model with ds5 data sets comment the ds6 DQF corrections listed at the end of the model and uncomment the corresponding ds5 DQF corrections. --------------------------------------------------------- Reformulated as a 2-polytope 688 format model. JADC, 11/19 The composition space of the complete (published) model corresponds to: [A][M][M,T,MT][M,C,N] where the brackets indicate the A, M1, M2, and M4 sites and A = Na, K, V T = Al, Fe3 MT = MAl, MFe3 C = Ca M = Mg, Fe N = Na The composition space of the model here has been simplified to [A][M][M,T,MAl][M,C,N] where the brackets indicate the A, M1, M2, and M4 sites and A = Na, K, V T = Al, Fe3 MT = MAl C = Ca M = Mg, Fe N = Na Refer to to cAmph(G) for an (commented out) example with MT = MAl, MFe3. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = Glaucophane = [M][T][N] => 2*2 Polytope 2 = ACMT-Amph = [M][M,T,MT][M,C] => 2*2*4 T is generalized Tschermaks [M][T][M,C] MT is generalized pargasite [M][MT][N] N is generalized glaucophane [M][T][N] M on M2 is generalized amphibole [M][M][M,C] -------------------------------------------------------- earlier corrections/revisions 10/07 z(m2a,al) corrected. Y. Podladchikov. 3/08 z(m1a,mg),z(m4,ca),z(m4,mg), and fged corrected, T1 multiplicity reduced to 1 Enthalpies of ordering corrected. JADC. 8/08 Enthalpy of ordering corrected for HP dqfs. 10/11 2011 DP revisions entered by MJC. 11/18 reformulated as a 2x2x6 prism adding Na-free riebeckite (grk) -------------------------------------------------------- independent endmembers and site populations: A M1 M2 M4 T1 _________________________________________ Mutliplicity 1 3 2 2 1(4)* _________________________________________ tr Vac Mg Mg Ca Si_Si ged Vac Mg Al Mg Al_Si mpa_q Na Mg Mg_Al Mg Al_Si gl Vac Mg Al Na Si_Si anth Vac Mg Mg Mg Si_Si fanth Vac Fe Fe Fe Si_Si mrb_q Vac Mg Fe3+ Na Si_Si ammo1 Vac Mg Fe Fe Si_Si ordered ammo2 Vac Fe Mg Fe Si_Si ordered *T1 has a true multiplicity of 4, H&P previously used an effective multiplicity of 2; in Diener et al '07 the multiplicity has been reduced to 1 JADC 9/07. (AlMg-1,M2) = ged - anth (CaNa-1,M4) = tr + (ged - anth) - gl (CaMg-1,M4) = tr - anth (MgNa-1,M4) = ged - gl (Fe3Al-1,M2) = mrb_q - gl FeMg-1(M4) = (-fanth - anth + ammo1 + ammo2) --------------------------------------------------------- oAmph(DP) abbreviation oAmph full_name orthoamphibole 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes ACMT-Amph 0 1 .1 0 | X([A][M][M,T,MT][M,C]), K-Na-Mg-Fe-Al-Ti-Fe3+ "pargasite" GlRb-Amph by difference | X([ ][M][Fe3,Al][N]), Mg-Fe-Al-Fe3+ "glaucophane" | ------------------------------------------- | Polytope 2 [A][M][M,T,MT][M,C] 3 | number of simplices, then for each simplex 2 2 4 | number of vertices on each simplex | Simplex 1 is bulk Mg/M | Simplex 2 is C/[C + M] on M4 | Simplex 3 is M, T, MT on M2 | endmembers on the vertices, index on the first simplex | changes fastest, last slowest cmged_d cfged_d ged_q fged_d | M-Ca-Al_tschermaks cmpa_d cfpa_d mpa_q fpa_d | M-Ca-Na-Al_parg mcgrk_d fcgrk_d mgrk_d fgrk_d | M-Ca-Fe3+_tschermaks (Na-free riebeckite) tr ftr_d anth fanth | M-Ca divalent amphibole | First (1d) simplex X_Mg 0 1 .1 0 | X(1,1) = bulk Mg/M X_Fe by difference | Second (1d) simplex X_C_M2 0 1 .1 0 | X(2,1) = Ca/[Ca + M] on M4 X_M_M2 by difference | Third simplex X_AlTs 0 1 .1 0 | X(3,1) = X(Al-Ts) X_NaPg 0 1 .1 0 | X(3,2) = X(Na-Al-Parg) X_FeTs 0 1 .1 0 | X(3,9) = X(Fe3-Ts) X_MCamph by difference | ------------------------------------------- | Polytope 1 [ ][M][T ][N] (Glaucophane) 2 | number of simplices 2 2 | number of vertices on each simplex | Simplex 1 is M | Simplex 2 is T | endmembers on the vertices, index on the first simplex | changes fastest, last slowest | => endmember 11 is [ ][Mg][Fe3][Na] | => endmember 21 is [ ][Fe][Fe3][Na] | => endmember 12 is [ ][Mg][Al ][Na] | => endmember 22 is [ ][Fe][Al ][Na] mrb_q frb_d | Fe3-glaucophane, aka riebeckite gl_q fgl_d | Al-glaucophane | First (1d) simplex X_Mg1 0 1 .1 0 | X(1,1) = bulk Mg/M X_Fe1 by difference | Second (3d) simplex X_rb 0 1 .1 0 | X(2,1) = Fe3/T' on M2, T' = Al + Fe3+ X_gl by difference | ------------------------------------------- begin_ordered_endmembers ammo1 = 3/7 anth + 4/7 fanth delta_g_of_ordering = -9.5d3 ammo2 = 2/7 anth + 5/7 fanth delta_g_of_ordering = -11.7d3 | these enthalpies differ from the published values | becuase Thermocalc reactions are specified in terms of the | endmember without "DQF" corrections, while in Perple_X | endmember include any DQF corrections. end_ordered_endmembers | ------------------------------------------- begin_dependent_endmembers ftr_d = 1 tr + 2 fanth - 1 ammo1 - 1 ammo2 fpa_d = 1 mpa_q + 1/2 fanth + 1/2 ammo2 - 1 anth cfpa_d = 1 mpa_q + 1 tr + 3/2 fanth - 1 ammo1 - 1/2 ammo2 - 1 anth cmpa_d = 1 mpa_q + 1 tr - 1 anth cmged_d = 1 ged_q - 1 anth + 1 tr cfged_d = 1 ged_q - 1 anth + 1 tr + 1 fanth - 1 ammo1 mgrk_d = 1 mrb_q + 1 ged_q - 1 gl_q mcgrk_d = 1 mrb_q + 1 ged_q - 1 anth - 1 gl_q + 1 tr fcgrk_d = 1 mrb_q + 1 ged_q - 1 gl_q + 1 fanth - 1 ammo1 fgrk_d = 1 mrb_q + 1 ged_q - 1 gl_q - 1 anth + 1 ammo2 fged_d = 1 ged_q - 1 anth + 1 ammo2 fgl_d = 1 gl_q + 1 fanth - 1 ammo1 frb_d = 1 mrb_q + 1 fanth - 1 ammo1 end_dependent_endmembers | ------------------------------------------- begin_excess_function W(anth ged_q ) 25d3 W(anth mpa_q ) 25d3 W(anth gl_q ) 65d3 W(anth tr) 45d3 W(anth fanth ) 33d3 W(anth mrb_q ) 52d3 W(anth ammo1) 18d3 W(anth ammo2) 23d3 W(ged_q mpa_q ) -40d3 W(ged_q gl_q ) 25d3 W(ged_q tr) 70d3 W(ged_q fanth ) 38.5d3 W(ged_q mrb_q ) 20d3 W(ged_q ammo1) 29d3 W(ged_q ammo2) 34.6d3 W(mpa_q gl_q ) 50d3 W(mpa_q tr) 90d3 W(mpa_q fanth ) 45d3 W(mpa_q mrb_q ) 40d3 W(mpa_q ammo1) 33.2d3 W(mpa_q ammo2) 36d3 W(gl_q tr) 65d3 W(gl_q fanth ) 81.2d3 W(gl_q ammo1) 65.5d3 W(gl_q ammo2) 78.4d3 W(tr fanth ) 75d3 W(tr mrb_q ) 52d3 W(tr ammo1) 57d3 W(tr ammo2) 63d3 W(fanth mrb_q ) 65d3 W(fanth ammo1) 12d3 W(fanth ammo2) 8d3 W(mrb_q ammo1) 52d3 W(mrb_q ammo2) 63d3 W(ammo1 ammo2) 20d3 end_excess_function | ------------------------------------------- 5 | number of sites in configurational entropy model (A, M1, M2, M4, T1) A | site name 2 1 1 | number of species, effective multiplicity, true multiplicity z(Na,A) = 1 mpa_q z(Vac,A) = 1 anth + 1 fanth + 1 tr + 1 ged_q + 1 gl_q + 1 ammo1 + 1 ammo2 + 1 mrb_q M1 | site name 2 3 3 | number of species, effective multiplicity, true multiplicity z(Fe,M1) = 1 fanth + 1 ammo2 z(Mg,M1) = 1 anth + 1 ged_q + 1 mpa_q + 1 mrb_q + 1 gl_q + 1 ammo1 + 1 tr M2 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Fe,M2) = 1 fanth + 1 ammo1 z(Fe3,M2) = 1 mrb_q z(Al,M2) = 1 ged_q + 1/2 mpa_q + 1 gl_q z(Mg,M2) = 1/2 mpa_q + 1 tr + 1 anth + 1 ammo2 M4 | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Ca,M4) = 1 tr z(Mg,M4) = 1 anth + 1 ged_q + 1 mpa_q z(Na,M4) = 1 gl_q + 1 mrb_q z(Fe,M4) = 1 fanth + 1 ammo1 + 1 ammo2 T1 | site name 2 1 4 | number of species, effective multiplicity, true multiplicity z(Al,T1) = 1/2 ged_q + 1/2 mpa_q z(Si,T1) = 1/2 ged_q + 1/2 mpa_q + 1 anth + 1 gl_q + 1 mrb_q + 1 tr + 1 fanth + 1 ammo1 + 1 ammo2 [Si4O22(OH)2] | formula suffix, enter "none" for no suffix. | ------------------------------------------- begin_van_laar_sizes alpha(tr) 1 alpha(ged_q) 1.5 alpha(mpa_q) 1.7 alpha(gl_q) 0.8 alpha(anth) 1 alpha(fanth) 1 alpha(ammo1) 1 alpha(ammo2) 1 alpha(mrb_q) 0.8 end_van_laar_sizes | ------------------------------------------- begin_dqf_corrections dqf(fanth) 7000 | ds6/ds5 dqf(mrb_q) 33000 | ds6/ds5 = omrb dqf(gl_q) 15000 | ds6/ds5 dqf(ged_q) 20000 | ds6 dqf(mpa_q) 25000 | ds6 | dqf(ged_q) 22000 | ds5 | dqf(mpa_q) 27000 | ds5 end_dqf_corrections | ------------------------------------------- begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Green, ECR, Holland, TJB & Powell, R (2007) An order-disorder model for omphacitic pyroxenes in the system jadeite-diopside-hedenbergite-acmite, with applications to eclogite rocks. American Mineralogist, 92, 1181-1189. Originally this model had no DQF corrections. To use this model with ds5 versions (pre-2011) of the THERMOCALC data base, the DQF on acmite should be -4000 J/mol as specified by Diener & Powell (2011), entered by MJC, Oct 31, 2011. To use this model with ds6 versions (2011+) of the THERMOCALC data base, the DQF [specified at the end of this model] on acmite should be -7000 J/mol as specified by Green et al (2016) and noted by Felix Gervais. Converted to 688 format. Aug, 2020. JADC. --------------------------------------------- WARNING: The choice of independent ordered species (here, cfm, om, jac) has the conseqence that this model CANNOT be used for the following subcompositions: jd-hed, hed-acm, di-acm, hed-acm-jd, hed-acm-di to work in these joins either fom, hac, or dac must replace either om or/and cfm. Site: 1 2 3 4 M2a M2b M1a M1b ____________________________________ Mutliplicity 1/2 1/2 1/2 1/2 ____________________________________ 1 Diopside Ca Ca Mg Mg Species: 2 Jadeite Na Na Al Al 3 Hedenbergite Ca Ca Fe2+ Fe2+ 4 Acmite Na Na Fe3+ Fe3+ ___________________________________ Ordered Cpd: 5 om Na Ca Al Mg 6 cfm Ca Ca Mg Fe2+ 7 jac Na Na Fe3+ Al Omph(GHP) abbreviation Cpx full_name clinopyroxene 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 4 | number of vertices (endmembers) on each simplex di jd acm hed 0 1 .1 0 | range and resolution of X(di) 0 1 .1 0 | range and resolution of X(jd) 0 1 .1 0 | range and resolution of X(acm) begin_ordered_endmembers om = 1/2 jd + 1/2 di delta_g_of_ordering = -2.9d3 cfm = 1/2 di + 1/2 hed delta_g_of_ordering = -1.5d3 jac = 1/2 jd + 1/2 acm delta_g_of_ordering = -1d3 | this differs from the published value of -3d3 J/mol | because Thermocalc reactions are specified in terms of the | endmember without "DQF" corrections, while in Perple_X | endmember include any DQF corrections. end_ordered_endmembers begin_excess_function W(jd di) 26d3 W(jd hed) 24d3 W(jd acm ) 5d3 W(jd om) 15.5d3 W(jd cfm) 25.2d3 W(jd jac) 3d3 W(di hed) 4d3 W(di acm ) 21d3 W(di om) 15.75d3 W(di cfm) 2d3 W(di jac) 24.65d3 W(hed acm ) 20.8d3 W(hed om) 17.2d3 W(hed cfm) 2d3 W(hed jac) 24.6d3 W(acm om) 16.4d3 W(acm cfm) 22.2d3 W(acm jac) 3d3 W(om cfm) 18.45d3 W(om jac) 19.5d3 W(cfm jac) 24.55d3 end_excess_function 4 | 4 site entropy model (m1a, m1b, m2b, m2a) M1a 4 .5 .5 | 4 species on m1a, mult = 1/2 z(Mg,M1a) = 1 di + 1 cfm z(Fe2+,M1a) = 1 hed z(Fe3+,M1a) = 1 acm + 1 jac z(Al,M1a) = 1 jd + 1 om M1b 4 .5 .5 | 4 species on m1b, mult = 1/2 z(Al,m1b) = 1 jd + 1 jac z(Fe2+,m1b) = 1 hed + 1 cfm z(Fe3+,m1b) = 1 acm z(Mg,m1b) = 1 di + 1 om M2a 2 .5 .5 | 2 species on m2a, mutiplicity = 1/2 z(Ca,m2a) = 1 di + 1 hed + 1 cfm z(Na,m2a) = 1 jd + 1 om + 1 jac + 1 acm M2b 2 .5 .5 | 2 species on m2b, mult. = 1/2 z(Na,M2b) = 1 jd + 1 acm + 1 jac z(Ca,M2b) = 1 di + 1 hed + 1 om + 1 cfm [Si2O6] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(acm) -7000 | ds6 |dqf(acm) -4000 | ds5 (originally zero) end_dqf_corrections end_of_model -------------------------------------------------------- begin_model | preliminary BCC iron solution model FeSi(BCC) abbreviation BCC full_name alloy 2 | model type: simplicial composition space. 2 | 2 endmembers iron Si 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function W(iron Si) -100d3 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(Fe) = 1 iron end_of_model -------------------------------------------------------- begin_model | FCC Fe-Si alloy after Lacaze and Sundman (1990) Metal. Trans 22A:1991-2211 FeSi(fcc) abbreviation FCC full_name alloy 2 | model type: simplicial composition space. 2 | 2 endmembers Fe-FCC Si-FCC 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function W(Fe-FCC Si-FCC) -125247.7 41.116 0 W(Fe-FCC Fe-FCC Si-FCC) -142707.6 W(Fe-FCC Si-FCC Si-FCC) 142707.6 W(Fe-FCC Fe-FCC Fe-FCC Si-FCC) 89907.3 W(Fe-FCC Fe-FCC Si-FCC Si-FCC) -179814.6 W(Fe-FCC Si-FCC Si-FCC Si-FCC) 89907.3 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(Fe) = 1 Fe-FCC end_of_model -------------------------------------------------------- begin_model | BCC Fe-Si alloy after Lacaze and Sundman (1990) Metal. Trans 22A:1991-2211 | because the order/disorder formulation implemented in this model is thermodynamically | inconsistent, the model is implemented internally in Perple_X (function gbccfesi). | The information provided below is a stub for the internal implementation and should | not be modified (the compositional limits and subdivision scheme are an exception in | this regard). oFeSi(bcc) abbreviation BCC full_name alloy 29 2 Fe-BCC Si-BCC 0 0 0 1 .1 0 | compositional limits and subdivision scheme ideal 0 end_of_model -------------------------------------------------------- begin_model FeSi_liq abbreviation Liq full_name liquid 2 2 Fe_LIQ Si_LIQ 0 0 0 1 .1 0 begin_excess_function w(Fe_LIQ Si_LIQ) -164434.600 41.9773 0. w(Fe_LIQ Fe_LIQ Si_LIQ) 0.000 -21.523 0. w(Fe_LIQ Si_LIQ Si_LIQ) 0.000 21.523 0. w(Fe_LIQ Fe_LIQ Fe_LIQ Si_LIQ) -18821.542 22.070 0. w(Fe_LIQ Fe_LIQ Si_LIQ Si_LIQ) 37643.080 -44.14 0 w(Fe_LIQ Si_LIQ Si_LIQ Si_LIQ) -18821.542 22.07 0. w(Fe_LIQ Fe_LIQ Fe_LIQ Fe_LIQ Si_LIQ) 9695.8 | 9695.8 w(Fe_LIQ Fe_LIQ Fe_LIQ Si_LIQ Si_LIQ) -29087.4 w(Fe_LIQ Fe_LIQ Si_LIQ Si_LIQ Si_LIQ) 29087.4 w(Fe_LIQ Si_LIQ Si_LIQ Si_LIQ Si_LIQ) -9695.8 | 9695.8 end_excess_function 1 2 1. z(Fe) = 1 Fe_LIQ end_of_model -------------------------------------------------------- begin_model | Entered by Jeff Marsh, Jan 10, 2012. ZrRu | zr in rutile after Tomkins et al., 07 abbreviation Ru full_name ilmenite 2 | model type 2 2 | number of endmembers zrru ru 0 0 | endmember flags 1e-5 1e-2 .1 1 | non-linear resolution of zrru begin_excess_function w(ru zrru) 10000 end_excess_function 1 | 1 site configurational entropy model 2 1 | 2 species, site multiplicity = 1. z(Ti) = 1 ru end_of_model -------------------------------------------------------- begin_model Silica Fluid this is a hybrid EoS, the EoS used for the individual species are controlled by the hybrid_EoS_XXX option. Si-Fluid abbreviation F full_name fluid 0 | model type: Internal EoS 2 | NOTE: for model type 0, the first endmember listed here, must correspond | to the second special component in the thermodynamic | file, i.e., the endmembers CANNOT be reordered. O SIO 0 0 endmember flags 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal 0 end_of_model -------------------------------------------------------- begin_model COH-Fluid Generic Hybrid Fluid EoS with linear Subdivision, see COH-Fluid+ for the non-linear subdivision version of this model. See perplex.ethz.ch/Perple_X_generic_hyrbid_fluid_EoS.html for explanation of this type of fluid model. -------------------------------------------------------- COH-Fluid abbreviation F full_name fluid 39 | model type: Generic Hybrid EoS 12 CO2 CH4 H2S SO2 H2 CO N2 NH3 HF C2H6 HCl H2O 0 0 0 0 0 0 0 0 0 0 0 0 | endmember flags 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal | the fluid is non-ideal, this tag means only no excess function 0 refine_endmembers end_of_model -------------------------------------------------------- begin_model COH-Fluid+ Generic Hybrid Fluid EoS with non-linear Subdivision. see COH-Fluid for the linear subdivision version of this model. See the header of this file for an explanation of non-linear subdivision parameters. See perplex.ethz.ch/Perple_X_generic_hyrbid_fluid_EoS.html for explanation of this type of fluid model. -------------------------------------------------------- COH-Fluid+ abbreviation F full_name fluid 39 | model type: Generic Hybrid EoS 9 CO2 CH4 H2S SO2 H2 CO HF C2H6 H2O 0 0 0 0 0 0 0 0 0 | endmember flags 1e-5 5e-1 .1 1 | subdivision ranges, imod = 1 -> non-linear subdivision 1e-5 1e-2 .1 1 | subdivision ranges, imod = 1 -> non-linear subdivision 1e-5 1e-3 .1 1 | H2S subdivision ranges, imod = 1 -> non-linear subdivision 1e-5 2.5e-5 .1 1 | SO2 subdivision ranges, imod = 1 -> non-linear subdivision 1e-5 1e-2 .1 1 | subdivision ranges, imod = 1 -> non-linear subdivision 1e-5 1e-2 .1 1 | subdivision ranges, imod = 1 -> non-linear subdivision 1e-5 1e-2 .1 1 | subdivision ranges, imod = 1 -> non-linear subdivision 1e-5 1e-2 .1 1 | subdivision ranges, imod = 1 -> non-linear subdivision ideal 0 refine_endmembers end_of_model -------------------------------------------------------- begin_model See perplex.ethz.ch/Perple_X_generic_hyrbid_fluid_EoS.html for explanation of this type of fluid model. WADDAH abbreviation F full_name fluid 39 | model type: Generic Hybrid EoS 1 H2O 0 | endmember flags ideal 0 | ideal configurational entropy refine_endmembers end_of_model -------------------------------------------------------- begin_model See perplex.ethz.ch/Perple_X_generic_hyrbid_fluid_EoS.html for explanation of this type of fluid model. C-H-Fluid abbreviation F full_name fluid 39 | model type: Generic Hybrid EoS 3 C2H6 CH4 H2 0 0 0 endmember flags 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal 0 | ideal configurational entropy end_of_model -------------------------------------------------------- begin_model See perplex.ethz.ch/Perple_X_generic_hyrbid_fluid_EoS.html for explanation of this type of fluid model. HOS-Fluid abbreviation F full_name fluid 39 | model type: Generic Hybrid EoS 5 H2S SO2 H2 O2 H2O 0 0 0 0 0 | endmember flags 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal 0 | ideal configurational entropy end_of_model -------------------------------------------------------- begin_model See perplex.ethz.ch/Perple_X_generic_hyrbid_fluid_EoS.html for explanation of this type of fluid model. HO-Fluid abbreviation F full_name fluid 39 | model type: Generic Hybrid EoS 3 O2 H2 H2O 0 0 0 0 | endmember flags 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision ranges, imod = 0 -> cartesian subdivision ideal 0 | ideal configurational entropy end_of_model -------------------------------------------------------- begin_model majoritic garnet from stixrude '11 Maj abbreviation Gt full_name garnet 2 model type: simplicial composition space 4 number of endmembers maj alm py gr endmember names 1 0 0 0 | endmember flags 0 .2 .1 0 | imod = 1 -> asymmetric transform subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision | NOTE restricted subdivision range on Mn (Species 1)! begin_excess_function w(py gr) 33000 w(alm py) 2500 | hp '98 give 2.4 kJ W(gr maj) 58d3 W(py maj) 21.3d3 end_excess_function 2 2 site entropy model 3 3. 3 species, A site multiplicity 3 z(Fe) = 1 alm z(Ca) = 1 gr 3 2. 2 species, B site multiplicity 2 z(Mg) = 1/2 maj z(Si) = 1/2 maj end_of_model -------------------------------------------------------- begin_model Wad abbreviation Wad full_name wadleysite 2 | model type: simplicial composition space 2 | 2 endmembers mwd fwd 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(mwd fwd) 16.5d3 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 mwd end_of_model -------------------------------------------------------- begin_model Ring abbreviation Ring full_name ringwoodite 2 | model type: simplicial composition space 2 | 2 endmembers mrw frw 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(mrw frw) 9.1d3 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(mg) = 1 mrw end_of_model -------------------------------------------------------- begin_model Wus abbreviation Wus full_name wuestite 2 model type: simplicial composition space 2 2 endmembers per fper 0 0 0 1 .1 0 begin_excess_function W(per fper) 13d3 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(mg) = 1 per end_of_model -------------------------------------------------------- begin_model | akimotoite (ilmenite-structure) solution Aki abbreviation Aki full_name ilmenite 2 | model type: simplicial composition space 3 | 3 endmembers cor mak fak 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function W(mak cor) 66d3 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 mak z(fe) = 1 fak 2 1 2 species on T site multiplicity = 1. z(al) = 1 cor end_of_model -------------------------------------------------------- begin_model Pv abbreviation Pv full_name perovskite 2 | model type: simplicial composition space 3 | 3 endmembers apv mpv fpv 0 0 0 0 1 .1 0 0 1 .1 0 begin_excess_function W(mpv apv) 116d3 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 mpv z(fe) = 1 fpv 2 1 2 species on T site multiplicity = 1. z(al) = 1 apv begin_van_laar_sizes alpha(mpv) 1 alpha(apv) 0.39 alpha(fpv) 1 end_van_laar_sizes end_of_model ------------------------------------------------ begin_model FeSiC_liq abbreviation Liq full_name liquid 2 3 Fe_LIQ Si_LIQ C_LIQ 0 0 0 0 1 .1 0 0 1 .1 0 begin_excess function w(Fe_LIQ Si_LIQ C_LIQ) 995155.5667 -495.31 0. w(Fe_LIQ C_LIQ) -124320 28.5 0. w(Fe_LIQ Fe_LIQ C_LIQ) -19300 w(Fe_LIQ C_LIQ C_LIQ) 19300 w(Fe_LIQ Fe_LIQ Fe_LIQ C_LIQ) 49260 -19. 0. w(Fe_LIQ Fe_LIQ C_LIQ C_LIQ) -98520 38. 0. w(Fe_LIQ C_LIQ C_LIQ C_LIQ) 49260 -19. 0. w(Fe_LIQ Fe_LIQ Fe_LIQ Si_LIQ) -18821.542 22.07 0. w(Fe_LIQ Fe_LIQ Si_LIQ Si_LIQ) 37643.084 -44.14 0. w(Fe_LIQ Si_LIQ Si_LIQ Si_LIQ) -18821.542 22.07 0. w(Fe_LIQ Fe_LIQ Fe_LIQ Fe_LIQ Si_LIQ) 9695.8 w(Fe_LIQ Fe_LIQ Fe_LIQ Si_LIQ Si_LIQ) -29087.4 w(Fe_LIQ Fe_LIQ Si_LIQ Si_LIQ Si_LIQ) 29087.4 w(Fe_LIQ Si_LIQ Si_LIQ Si_LIQ Si_LIQ) -9695.8 w(Si_LIQ C_LIQ) -133000 30.97 0. w(Fe_LIQ Si_LIQ) -164434.6 41.9773 0. w(Fe_LIQ Fe_LIQ Si_LIQ) 0 -21.523 0. w(Fe_LIQ Si_LIQ Si_LIQ) 0 21.523 0. w(Fe_LIQ Si_LIQ C_LIQ C_LIQ) -549415.5667 495.31 0. w(Fe_LIQ Fe_LIQ Si_LIQ C_LIQ) -0.1001220867d7 495.98 0. w(Fe_LIQ Si_LIQ Si_LIQ C_LIQ) 0.1550636433d7 -955.29 0. end_excess_function 1 3 1. z(Fe) = 1 Fe_LIQ z(Si) = 1 Si_LIQ end_of_model -------------------------------------------------------- begin_model Sapphirine, O/D non-ideal, Taylor-Jones & Powell (JMG, 2010, 28, 615-633) * requires the kel04ver.dat thermodynamic data file (for endmember spr5) Entered June 26, 2013. JADC. ---------------------------------------------------------- refer to the commentary of Sapp(KWP) for a resume of Sapphirine models in solution_model.dat ---------------------------------------------------------- 1 2 3 M3 M456 T _________________ Mutliplicity 1 3 1 _________________ 1 spr4 Mg Mg Si Species: 2 fspr Fe Fe Si 3 spr5 Al Mg Al 4 fsp5_d Al Fe Al dependent 5 ospr Fe3+ Mg Al 6 fospr_d Fe3+ Fe Al dependent 7 spro Fe Mg Si ordered Dependent endmember: FeMg-1_M3456 = (fspr-spro)/3 fsp5_d = spr5 + fspr - spro fospro_d = spr5 + fspr - spro Sapp(TP) abbreviation Sap full_name sapphirine 8 | model type, should be 9 2 | 2 independent mixing sites, reciprocal solution 2 3 | 2 components {Fe2+, Mg} for composition 1, 3 components {Al, Si, Fe3+} for composition 2. | endmember names spr4 fspr spr5 fsp5_d ospr fospr_d 1 | 1 ordered species: spro = 3/4 spr4 + 1/4 fspr delta_g_of_ordering = -10d3 2 | 2 dependent endmembers fsp5_d = 1 spr5 + 1 fspr - 1 spro fospr_d = 1 spr5 + 1 fspr - 1 spro 0 0 0 0 0 0 | endmember flags, indicate if endmember is part of the solution (i.e. iend = 0). | subdivision model for (binary) site 1 (M2): 0 1 .1 0 | range and resolution of X(Mg), subdivision scheme for site 1: imod = 0 -> cartesian | subdivision model for (ternary) site 2 0 1 .1 0 | range and resolution of X(Ts), subdivision scheme for site 3, species 1: imod = 0 -> cartesian 0 1 .1 0 | range and resolution of X(un-Ts), subdivision scheme for site 3, species 2: imod = 0 -> cartesian begin_excess_function W(spr4 spr5) 10d3 W(spr4 fspr) 16d3 W(spr4 spro) 4d3 W(spr4 ospr) 10d3 W(spr5 fspr) 22d3 W(spr5 spro) 10d3 W(fspr spro) 12d3 W(fspr ospr) 22d3 W(spro ospr) 10d3 end_excess_function 3 | 3 site (M3, M456, T) configurational entropy model 2 3. | 2 species on M46, 3 sites per formula unit z(m456,fe) = 1 fspr 4 1 | 4 species on M3, 1 sites per formula unit. z(m3,Al) = 1 spr5 z(m3,Fe3) = 1 ospr z(m3,mg) = 1 spr4 2 1 | 2 species on T, 1 sites per formula unit. z(T,si) = 1 spr4 + 1 fspr + 1 spro begin_dqf_corrections dqf(fspr) -3d3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model |Calcite-Magnesite with Dolomite/Ankerite compound formation. |Franzolin, Schmidt and Poli (2011) CMP 161:213-227, DOI: 10.1007/s00410-010-0527-x NOTES/WARNINGS: This model requires the following make definitions in the thermodynamic data file for the odo and oank endmembers. odo = 1/2 mag + 1/2 cc DQF(J/mol) = -1000 oank = 1/4 sid + 1/2 cc + 1/4 mag DQF(J/mol) = -750 JADC, 3/26/2012. oCcM(EF) abbreviation Do full_name carbonate 2 |model type 5 |number of endmembers mag cc odo sid oank 0 0 0 0 0 |endmember flags 0 1 .1 0 |range and increments on X(cc) 0 1 .1 0 0 1 .1 0 0 1 .1 0 |range and increments on X(mag) begin_excess_function w(mag cc) 28000 w(cc odo) 11200 w(mag odo) 14000 W(cc sid) 20503 w(oank sid) 73650 -50*T w(oank cc) 12730 -10*T w(mag sid) 10000 w(sid odo) 51190 -30*T w(oank odo) -5000 w(oank mag) 30000 end_excess_function 3 |3 site (m1 m2a m2b) entropy model 3 .5 |3 species on m1, mutiplicity = 1/2 z(m1,ca) = 1 cc + 1 odo + 1 oank z(m1,mg) = 1 mag 3 0.25 |3 species on m2a, mult. = 1/4 z(m2a,ca) = 1 cc z(m2a,mg) = 1 mag + 1 odo 3 0.25 |3 species on m2b, mult. = 1/4 z(m2b,ca) = 1 cc z(m2b,mg) = 1 mag + 1 odo + 1 oank begin_van_laar_sizes alpha(cc) 0.25 + 0.000929*T alpha(mag) 1 alpha(odo) 0.95 alpha(sid) 0.01 + 0.000666*T alpha(oank) 0.929 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model |Magnesite-Siderite melt, nathan kang, 4/2015. LIQ(NK) abbreviation Liq full_name liquid 2 |model type 2 |number of endmembers MGCO3L FECO3L 0 0 |endmember flags 0 1 .1 0 |range and increments on X(Fe) begin_excess_function W(FECO3L MGCO3L MGCO3L) -7600 |W_sid-mag = W12 W(MGCO3L FECO3L FECO3L) -7600 |W_mag-sid = W21 end_excess_function 1 |1 site entropy model 2 1 |2 species, multiplicity = 1 z(fe) = 1 FECO3L end_of_model -------------------------------------------------------- begin_model LIQ(EF) abbreviation Liq full_name liquid 2 |model type 3 |number of endmembers CACO3L MGCO3L FECO3L 0 0 0 |endmember flags 0 1 .1 0 0 1 .1 0 | range and increments on X(cc) begin_excess_function W(CACO3L CACO3L FECO3L) -65000 W(CACO3L FECO3L FECO3L) 2000 W(MGCO3L MGCO3L CACO3L) 491900 - 300*T W(MGCO3L CACO3L CACO3L) -142000 + 1.2*P W(MGCO3L MGCO3L FECO3L) -5594.2 W(MGCO3L FECO3L FECO3L) -5594.2 end_excess_function 1 3 1 z(ca) = 1 CACO3L z(fe) = 1 FECO3L end_of_model -------------------------------------------------------- begin_model |EF disordered ternary carb dis(EF) abbreviation Cc full_name carbonate 2 |model type 3 |number of endmembers mag cc sid 0 0 0 | endmember flags 0 1 .1 0 | range and increments on X(cc) 0 1 .1 0 begin_excess_function w(mag cc) 28000 W(cc sid) 20503 w(mag sid) 10000 end_excess_function 1 |1 site entropy model 3 1 |3 species on m, mutiplicity = 1 z(ca) = 1 cc z(mg) = 1 mag begin_van_laar_sizes alpha(cc) 0.25 0.000929*T alpha(mag) 1 alpha(sid) 0.01 0.000666*T end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model Silicate gas, Lewis and Randal (Ideal) Mixing, the EoS used for the pure species is determined by the EoS flag specified for the species in the thermodynamic data file Si-vapor abbreviation gas full_name gas 2 9 number of endmembers | ENDMEMBER NAMES: SiO2 SiO Si O2 O MgO Mg FeO Fe 0 0 0 0 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution of X(SiO2), 0 => cartesian subdivision/1 => asymmetric subdivision 0 .5 .1 0 | range and resolution of X(SiO) 0 .5 .1 0 | range and resolution of X(Si) 0 .5 .1 0 | range and resolution of X(O2) 0 .5 .1 0 | range and resolution of X(O) 0 .5 .1 0 | range and resolution of X(MgO) 0 .5 .1 0 | range and resolution of X(Mg) 0 .5 .1 0 | range and resolution of X(FeO) ideal 1 | molecular configurational entropy model 9 1 | 6 species, multiplicity 1 z(SiO2) = 1 SiO2 z(SiO) = 1 SiO z(O2) = 1 O2 z(O) = 1 O z(Mg) = 1 Mg z(Fe) = 1 Fe z(FeO) = 1 FeO z(MgO) = 1 MgO end_of_model -------------------------------------------------------- begin_model Silicate gas, Lewis and Randal (Ideal) Mixing, the EoS used for the pure species is determined by the EoS flag specified for the species in the thermodynamic data file CCO-vapor abbreviation gas full_name gas 2 7 number of endmembers | ENDMEMBER NAMES: CO2 CO O2 | xO(g) xMgO(g) xMg(g) xFeO(g) xFe(g) 0 0 0 0 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution of X(CO2), 0 => cartesian subdivision/1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(CO) 0 .5 .1 1 | range and resolution of X(O2) |0 .5 .1 1 | range and resolution of X(O) 0 .5 .1 1 | range and resolution of X(MgO) 0 .5 .1 1 | range and resolution of X(Mg) 0 .5 .1 1 | range and resolution of X(FeO) ideal 1 | molecular configurational entropy model 7 1 | 6 species, multiplicity 1 z(CO2) = 1 CO2 z(CO) = 1 CO z(O2) = 1 O2 | z(O) = 1 xO(g) z(Mg) = 1 xMg(g) z(Fe) = 1 xFe(g) z(FeO) = 1 xFeO(g) end_of_model -------------------------------------------------------- begin_model Formally this need not be an ideal gas, but it assumes Lewis and Randall (Ideal) mixing, the EoS (e.g., ideal gas) used for the pure species is determined by the EoS flag specified for the species in the thermodynamic data file ideal_gas abbreviation gas full_name gas 2 | model type 9 | number of species | species O2(g) O(g) CO CO2 SiO(g) Mg(g) Fe(g) FeO(g) MgO(g) 0 0 0 0 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution of X(O2), 0 => cartesian subdivision/1 0 1 .1 0 | range and resolution of X(O) 0 1 .1 0 | range and resolution of X(Mg) 0 1 .1 0 | range and resolution of X(Fe) 0 1 .1 0 | range and resolution of X(MgO) 0 1 .1 0 | range and resolution of X(Mg) 0 1 .1 0 | range and resolution of X(Fe) 0 1 .1 0 | range and resolution of X(MgO) ideal 1 | molecular configurational entropy model 9 1 | 6 species, multiplicity 1 z(O2) = 1 O2(g) z(O) = 1 O(g) z(Mg) = 1 Mg(g) z(Fe) = 1 Fe(g) z(FeO) = 1 FeO(g) z(CO) = 1 CO z(CO2) = 1 CO2 z(SiO) = 1 SiO(g) end_of_model -------------------------------------------------------- begin_model FeSiC-BCC abbreviation BCC full_name alloy 30 | model type: special prismatic (Lacaze & Sundman, 1991) 2 | number of independent mixing sites, reciprocal solution 2 2 | Fe-Si on site 1, Vacancy-C on site 2 Fe-BCC Si-BCC | endmember names FeC-BCC SiC-BCC 0 | number of dependent endmembers 0 0 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> Fe fraction on site 1 0 1 .1 0 | imod = 0 -> vacancy fraction on site 2 ideal | internal excess model 0 | 0 site entropy model -> internal model end_of_model -------------------------------------------------------- begin_model FeSiC-FCC abbreviation FCC full_name alloy 31 | model type: special prismatic (Lacaze & Sundman, 1991) 2 | number of independent mixing sites, reciprocal solution 2 2 | Fe-Si on site 1, Vacancy-C on site 2 Fe-FCC Si-FCC | endmember names FeC-FCC SiC-FCC 0 | number of dependent endmembers 0 0 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> Fe fraction on site 1 0 1 .1 0 | imod = 0 -> vacancy fraction on site 2 ideal | internal excess model 0 | 0 site entropy model -> internal model end_of_model -------------------------------------------------------- begin_model |FeCr-liquid solution model after Xiong et al., 2011 |G. Helffrich, April 2016 FeCr(liq) abbreviation Liq full_name liquid 2 |model type: simplicial composition space. 2 |2 endmembers Fe_LIQ Cr_LIQ 0 0 |endmember flags 0 1 .1 0 |imod = 0 -> cartesian subdivision begin_excess_function Wk(Cr_LIQ Fe_LIQ) w0 = 0 wT = -5.983 w0 = -384.41 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(Fe) = 1 Fe_LIQ end_of_model -------------------------------------------------------- begin_model |FeCr-FCC solution model after Xiong et al., 2011 |G. Helffrich, April 2016 FeCr-FCC abbreviation FCC full_name alloy 2 |model type: simplicial composition space. 2 |2 endmembers Fe-FCC Cr-FCC 0 0 |endmember flags 0 1 .1 0 |imod = 0 -> cartesian subdivision begin_excess_function Wk(Cr-FCC Fe-FCC) w0 = 28871.89 wT = -22.318 w0 = 32711.42 wT = -18.180 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(Fe) = 1 Fe-FCC end_of_model -------------------------------------------------------- begin_model |BCC Fe-Cr alloy after Xiong et al., 2011, with the contribution from magnetic ordering FeCr-BCC abbreviation BCC full_name alloy 32 |special solution model 2 |2 end-members Fe-BCC Cr-BCC 0 0 0 1 .1 0 |compositional limits and subdivision scheme ideal 0 end_of_model -------------------------------------------------------- begin_model |FeCr-sigma solution model after Andersson & Sundman, |1987, which is based on 30 sites and mixing on 18. |This source defines no excess mixing for the soln., |but it is needed to stabilize the sigma phase. One |cause might be use of the Xiong et al. (2011) for |BCC Fe-Cr. As it is, this frankenstein-like mix |of parts from Andersson & Sundman '87 and Xiong |et al. '11 works concisely to reproduce the |subsolidus and melting relations. | G. Helffrich, ELSI, 8 Apr. 2016. FeCr-s abbreviation Sigma full_name alloy 2 |solution model type 2 |2 endmembers FeCr_sig CrFe_sig 0 0 |endmember flags 0 1 .1 0 |imod = 0 -> cartesian subdivision begin_excess_function W(FeCr_sig CrFe_sig) -1936.273 -0.5671667 0 | original from Nastia: -11617.64 -3.403 0 end_excess_function 1 |1 site entropy model | Nastia's version: 2 0.5333 |2 species, 1 site multiplicity = 16/30 2 0.6 |2 species, 1 site, multiplicity = 18/30 z(Fe) = 1 FeCr_sig end_of_model -------------------------------------------------------- begin_model |solution model parameters after Brosh, 2013 FeCliq abbreviation Liq full_name liquid 2 |solution model type: simplicial composition space 2 |2 end-members Feliq Cliq 0 0 |end-member flags 0 1 .1 0 |imod = 0 -> cartesian subdivision scheme begin_excess_function Wk(Feliq Cliq) w0 = -124320 wT = 28.5 wP = -1.53d-1 wP1 = 6d4 wP2 = 2.3d-1 w0 = 19300 w0 = 49260 wT = -19. end_excess_function 1 |one site entropy model 2 1 |two end-members mixing on one site and site multiplicity z(Fe) = 1 Feliq |site fraction of Fe on this site in terms of end-members end_of_model -------------------------------------------------------- begin_model | solution model parameters after Brosh, 2013 FeCfcc abbreviation FCC full_name alloy 2 2 FeA1 FeCA1 0 0 0 1 .1 0 begin_excess_function Wk(FeA1 FeCA1) w0 = -34671. end_excess_function 1 2 1. z(C) = 1 FeCA1 end_of_model -------------------------------------------------------- begin_model |solution model parameters after Brosh, 2013 FeCbcc abbreviation BCC full_name alloy 2 2 FeA2 FeCA2 0 0 0 1 .1 0 begin_excess_function Wk(FeA2 FeCA2) w0 = 0 wT = -190 end_excess_function 1 2 3. z(c) = 1 FeCA2 end_of_model -------------------------------------------------------- begin_model Ilmenite after White, RW, Powell, R, Holland, TJB & Worley, BA (2000, JMG), modified as described by White et al (2014). NOTES: DO NOT USE THIS MODEL WITH ds6.34+. USE Ilm(W24) FOR ds6.34+ 7.2.0 - This model has been Perple_X's bete noir for more than 2 decades. The reason for this is that the model is referenced to the disordered thermodynamic base function for the ilmenite endmember that is generated internally by THERMOCALC. Reproducing this base function is challenging because round-off errors and the error in the ds5 version of the landau model allow ambiguity. In Perple_X it is assumed that the TC Landau error does not propagate into the base function. H&P give the smax of the ilm landau model variously as 11 or 12 J/mol/K, whereas the true value is 11.526 J/mol/K (R ln 4). It is probable that TC uses the rounded values. The good news is that in ds6.36, TC has stopped using the Landau model for solution model endmembers. JADC 2/26. Green et al (2016) remark that the White et al (2014) modification "may be suspect" and use the original 2000 model, i.e., that obtained by excluding geikielite. A B _____________________ Mutliplicity 1 1 _____________________ 1 oilm Fe Ti Species: 2 dilm TiFe TiFe 3 dhem Fe3+ Fe3+ 4 pnt Mn Ti 5 geik Mg Ti Previously present as a brute force model which did not allow the anti-ordered ilmenite configuration. Re-formulated as standard o/d model to remove this limitation; however, the model has been parameterized so that the ordered and antiordered states are energetucally equivalent. JADC Jan 14, 2017. NOTES: * For Mg and Mn solution this model differs from the Thermocalc version in that Mg and Mn are confined to the A-site, in contrast in Thermocalc the "equipartition" assumption is used to allocate Mg and Mn to the B-site in proportion to the Ferrous iron intersite partitioning. Consequently, this model predicts lower Mn and Mg solubility in ilmenite than the Thermocalc version of the model. JADC, Oct 29, 2011. * Added W(pnt oilm). Felix Gervais, Oct 29, 2011. * Corrected to count Ti in A-site configurational entropy. JADC, Jun 16, 2013. * Make definitions corrected from the raw thermocalc versions. JADC, Jun 19, 2013. * The geikielite and most of the pyrhophanite excess terms were accidentally deleted (by me) ca Feb 6, 2017 at the same time that the Ilm(W) and Ilm(WPH0) equivalents of Ilm(WPH) were deleted from this file. Correction courtesy of Bob Myhill and the perplex discussion group (groups.io/g/PerpleX/topic/74420499). JADC, May 27, 2020. -------------------------------------------------------- Ilm(WPH) abbreviation Ilm full_name ilmenite 6 | model type 4 | 4 endmembers pnt geik hem ilm 1 | 1 ordered species: oilm = 1 ilm delta_g_of_ordering = -15600.3 + 11.526*T | dqf(oilm) - dqf(ilm) 0 0 0 0 | endmember flags 0 1 .1 0 | subdivision range, X(pnt) 0 1 .1 0 | subdivision range, X(geik) 0 1 .1 0 | subdivision range, X(hem) begin_excess_function W(oilm ilm) 15600 W(oilm hem) 26600 W(oilm geik) 4000 W(oilm pnt) 2000 W(ilm hem) 11000 W(ilm geik) 4000 W(ilm pnt) 2000 W(hem geik) 36000 W(hem pnt) 25000 W(geik pnt) 4000 end_excess_function 2 | 2 site model 5 1 | A - Fe2+ Mn Fe3+ Mg TiF z(A,Mg) = 1 geik z(A,mn) = 1 pnt z(A,fe3+) = 1 hem z(A,ti) = 1/2 ilm 3 1 | B - Fe3+ Ti4+ Fe2+ z(b,fe3+) = 1 hem z(b,fe2+) = 1/2 ilm begin_special_operations spop(ilm) disorder dqf(J/mol) = 1992.8 - 2.1 * T_K spop(hem) disorder dqf(J/mol) = 0 end_special_operations refine_endmembers end_of_model -------------------------------------------------------- begin_model Ilmenite - Ilm(W24) - Weller, OM, Holland, TJB, Soderman, CR, Green, ECR, Powell, R, Beard, CD & Riel, N (2024). Journal of Petrology, 65,doi: 10.1093/petrology/egae098 DO NOT USE THIS MODEL WITH ds6.33-. USE Ilm(W24) FOR ds6.34+. JADC, Jan 2026 rewrote as 688 format model (required for multiple OD). JADC, 7.2.2 A B _____________________ Mutliplicity 1 1 _____________________ 1 oilm Fe Ti Species: 2 ilm TiFe TiFe 3 hem Fe3+ Fe3+ 4 ogk Mg Ti 5 geik MgTi MgTi ----------------------------------------------------------- Ilm(W24) abbreviation Ilm full_name ilmenite 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 3 | number of vertices on each simplex | endmembers on the vertices ilm hem geik |------------------------------------------- 0 1 .1 0 | subdivision range, X(geik) 0 1 .1 0 | subdivision range, X(hem) |------------------------------------------- begin_ordered_endmembers | 0 - dqf(ilm) oilm = 1 ilm delta_g_of_ordering = -7050 + 11.526 * T_K - 0.13 * P_bar | 0 - dqf(geik) ogk = 1 geik delta_g_of_ordering = -8310 + 11.526 * T_K end_ordered_endmembers |------------------------------------------- begin_excess_function w(oilm ilm) +7050 +0.13 * P_bar w(oilm hem) +14300 w(oilm ogk) -7600 w(oilm geik) +600 w(ilm hem) +7250 -0.13 * P_bar w(ilm ogk) -5500 w(ilm geik) -2200 w(hem ogk) +12500 w(hem geik) +2700 w(ogk geik) +8300 end_excess_function 2 | 2 site model A | site name 4 0.5 1 | number of species, effective multiplicity, true multiplicity z(Mg,A) = 1 ogk + 1/2 geik z(Fe3+,A) = 1 hem z(Fe2+,A) = 1 oilm + 1/2 ilm z(Ti,A) = 1/2 ilm + 1/2 geik B | site name 4 0.5 1 | number of species, effective multiplicity, true multiplicity z(Mg,B) = 1/2 geik z(Fe3+,B) = 1 hem z(Fe2+,B) = 1/2 ilm z(Ti,B) = 1/2 ilm + 1/2 geik + 1 oilm + 1 ogk refine_endmembers begin_special_operations | these DQFs create the disordered endmember from the ordered base function spop(ilm) disorder dqf(J/mol) = 7050 - 11.526 * T_K + 0.13 * P_bar spop(geik) disorder dqf(J/mol) = 8310 - 11.526 * T_K end_special_operations solvus_tolerance 0.2 end_of_model -------------------------------------------------------- begin_model Majoritic garnet, Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 * originally entered by Bob Myhill, 2018/02/13 * reformulated as a prismatic + orphan (nagt) vertex model. JADC, May 10, 2018 * added cnagt and fnagt and changed the orphan to maj. JADC, Nov 2, 2018. * Reformulated as a 2-polytope 688 format model. JADC, 9/19 ___________________ Prism I 5 maj Mg MgSi independent 7 cmmaj_d Ca MgSi dependent 6 fmaj_d Fe FeSi dependent 11 cfmaj_d Ca FeSi dependent ___________________ Prism II 3 gr Ca AlAl independent 1 py Mg AlAl independent 2 alm Fe AlAl independent 4 cnagt Ca2/3Na1/3 AlSi dependent 4 nagt Mg2/3Na1/3 AlSi independent 4 fnagt Fe2/3Na1/3 AlSi dependent ___________________ 13 gfm Mg FeSi ordered (same as gfm_d, above) Gt(H) abbreviation Gt full_name garnet 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [maj] 0 1 .1 0 [~maj] by difference | ------------------------------------------- | Polytope 1 2 | number of simplices 2 2 | number of vertices on each simplex maj fmaj_d cmmaj_d cfmaj_d | First 2-simplex X_Mg 0 1 .1 0 | X(1,1) is bulk Fe/M X_Fe by difference | Second 2-simplex X_Ca 0 1 .1 0 | X(2,1) is Ca/[M+Ca] on M1 X_M by difference | ------------------------------------------- | Polytope 2 2 | number of simplices 3 2 | number of vertices on each simplex cnagt_d fnagt_d nagt | ternary 1 gr alm py | ternary 2 | First 3-simplex X_Ca 0 1 .1 0 | range and resolution of X(1,1) => X(Ca) in the ternaries X_Fe 0 1 .1 0 | range and resolution of X(1,2) => X(Fe) in the ternaries X_Mg by difference | second 2-simplex X_NaGt 0 .3 .1 0 | range and resolution of X(2,1) => X(M-nagt) X_MGt by difference begin_ordered_endmembers gfm = 1 maj - 1/3 py + 1/3 alm delta_g_of_ordering = -10d3 end_ordered_endmembers begin_dependent_endmembers fnagt_d = 1 nagt + 2/3 alm - 2/3 py cnagt_d = 1 nagt + 2/3 gr - 2/3 py fmaj_d = 1 gfm + 1 alm - 1 py cmmaj_d = 1 maj + 1 gr - 1 py cfmaj_d = 1 gfm + 1 gr - 1 py end_dependent_endmembers begin_excess_function w(py alm) 3d3 w(py gr) 33d3 w(py maj) 15d3 w(py gfm) 13.5d3 w(py nagt) 14d3 w(alm gr) 5d3 w(alm maj) 18d3 w(alm gfm) 16.5d3 w(alm nagt) 11.2d3 w(gr maj) 48d3 w(gr gfm) 46.5d3 w(gr nagt) 30d3 w(maj gfm) 0.5d3 w(maj nagt) 8.5d3 w(gfm nagt) 7.0d3 end_excess_function 2 2 site entropy model X | site name 4 3 3 | number of species, effective multiplicity, true multiplicity z(Fe) = 1 alm z(Ca) = 1 gr z(Na) = 1/3 nagt z(Mg) = 1 py + 1 maj + 1 gfm + 2/3 nagt | added maj + gfm 7/3/20, JADC Y | site name 4 2 2 | number of species, effective multiplicity, true multiplicity z(Mg) = 1/2 maj z(Fe) = 1/2 gfm z(Si) = 1/2 maj + 1/2 gfm + 1/2 nagt z(Al) = 1 gr + 1 alm + 1 py + 1/2 nagt [Si3O12] | formula suffix, enter "none" for no suffix. end_of_model -------------------------------------------------------- begin_model Ferropericlase, Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 [Bob Myhill, 2018/02/13] Notes: This model consists of symmetric mixing of Mg-Fe on 1 site with a multiplicity of 1 Fper(H) abbreviation Fper full_name ferropericlase 2 model type: Margules, macroscopic 2 2 endmembers per fper 0 0 0 1 .1 0 begin_excess_function W(per fper) 18d3 end_excess_function 1 1 site entropy model 2 1 2 species, site multiplicity = 1. z(mg) = 1 per end_of_model -------------------------------------------------------- begin_model MgSi perovskite, Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 [Bob Myhill, 2018/02/13] Notes: Identical to CaSi perovskite, but with different dqfs for all the endmembers This model consists of symmetric mixing on 2 sites, each with a multiplicity of 1: 1 2 M1 M2 ________________ Multiplicity 1 1 ________________ 1 mpv Mg Si Species: 2 mfpv_q Fe2+ Si 3 mcpv_q Ca Si 4 apv Al Al 5 mnpv_q Na1/2Al1/2 Si Mpv(H) abbreviation Mpv full_name MgSi Perovskite 2 | model type: Margules, macroscopic 5 | 5 endmembers mpv mfpv_q mcpv_q apv mnpv_q 0 0 0 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function w(mpv mfpv_q) 12d3 w(mpv mcpv_q) 15d3 W(mpv apv) 20d3 W(mpv mnpv_q) 22d3 w(mfpv_q mcpv_q) 10.5d3 W(mfpv_q apv) 14d3 W(mfpv_q mnpv_q) 15.4d3 W(mcpv_q apv) 5d3 W(mcpv_q mnpv_q) 7.5d3 W(apv mnpv_q) 2.5d3 end_excess_function 2 2 site entropy model 5 1 5 species, M1 site multiplicity 1 z(Mg) = 1 mpv z(Fe) = 1 mfpv_q z(Ca) = 1 mcpv_q z(Na) = 1/2 mnpv_q 2 1 2 species, M2 site multiplicity 1 z(Al) = 1 apv begin_dqf_corrections DQF(mfpv_q) = -9500 DQF(mcpv_q) = 60000 DQF(mnpv_q) = 16000 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model CaSi perovskite Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 [Bob Myhill, 2018/02/13] Notes: Identical to MgSi perovskite, but with different dqfs for all the endmembers 1 2 M1 M2 ________________ Multiplicity 1 1 ________________ 1 cmpv Mg Si Species: 2 cfpv Fe2+ Si 3 cpv Ca Si 4 capv Al Al 5 cnpv Na1/2Al1/2 Si Cpv(H) abbreviation Cpv full_name CaSi Perovskite 2 | model type: Margules, macroscopic 5 | 5 endmembers cmpv_q cfpv_q cpv capv_q cnpv_q 0 0 0 0 0 | endmember flags 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision begin_excess_function W(cmpv_q cfpv_q) 12d3 W(cmpv_q cpv) 15d3 W(cmpv_q capv_q) 20d3 W(cmpv_q cnpv_q) 22d3 W(cfpv_q cpv) 10.5d3 W(cfpv_q capv_q) 14d3 W(cfpv_q cnpv_q) 15.4d3 W(cpv capv_q) 5d3 W(cpv cnpv_q) 7.5d3 W(capv_q cnpv_q) 2.5d3 end_excess_function 2 2 site entropy model 5 1 5 species, M1 site multiplicity 1 z(Mg) = 1 cmpv_q z(Fe) = 1 cfpv_q z(Ca) = 1 cpv z(Na) = 1/2 cnpv_q 2 1 2 species, M2 site multiplicity 1 z(Al) = 1 capv_q begin_dqf_corrections DQF(cmpv_q) = 35e3 DQF(cfpv_q) = 24e3 DQF(capv_q) = 45e3 DQF(cnpv_q) = 25e3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Corundum Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 [Bob Myhill, 2018/02/13] 1 2 M T _____________ Multiplicity 1 1 _____________ 1 cor Al Al Species: 2 mcor Mg Si 3 fcor Fe2+ Si Cor(H) abbreviation Cor full_name Corundum 2 | model type: Margules, macroscopic 3 | 3 endmembers cor mcor fcor_q 0 0 0 0 1 .1 0 0 1 .1 0 begin_excess_function W(cor mcor) 12d3 W(cor fcor_q) 10d3 W(mcor fcor_q) 4d3 end_excess_function 2 2 site entropy model 3 1 3 species on M site multiplicity = 1. z(mg) = 1 mcor z(fe) = 1 fcor_q 2 1 2 species on T site multiplicity = 1. z(al) = 1 cor begin_dqf_corrections dqf(fcor_q) = -15d3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model High pressure calcium-ferrite structure phase, Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 Notes: * originally entered by Bob Myhill, 2018/02/13. * reformulated as a prismatic + orphan vertex model. JADC, 10/5/2018 * eliminated ordered dependent endmembers. JADC, 10/2018 * Suspect: nacfb = 1 nacf DQF(J/mol) = -9000 This dqf was presumably added by Holland et al. as a free parameter during inversion, and not reintegrated into their data table. * The M2 site has a fake multiplicity of 1 (true multiplicity would be 2). M3 M2 ____________ Multiplicity 1 1(2) ____________ Prism I: 1 macf Mg AlAl independent 5 mscf Mg MgSi independent 7 cscf_d Ca MgSi dependent 2 facf_d Fe AlAl dependent 10 fscf Fe FeSi independent 11 cfscf_d Ca FeSi dependent ____________ Orphans: 3 cacf Ca AlAl independent 4 nacfb_q Na AlSi independent (orphan) ____________ Ordered: 13 oscf Fe2+ MgSi (same as oscf_d above) CFer(H) abbreviation CFer full_name Ca-ferrite type 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [~M] 0 1 .1 0 | subdivision range for X(1) = M-free [M] by difference | ------------------------------------------- | Polytope 1 [~M] 1 | number of simplices 2 | number of vertices on the simplex nacfb_q cacf X_Na 0 1 .1 0 | range and resolution of X(nacfb/[nacfb+cacf]) X_Ca by difference | ------------------------------------------- | Polytope 2 [M] 2 | number of simplices 3 2 | number of vertices on each simplex macf cscf_d mscf facf_d cfscf_d fscf | First 3-simplex X_Al 0 1 .1 0 | range and resolution of X(AlAl) X_Ca 0 1 .1 0 | range and resolution of X(CaM) X_M by difference | second 2-simplex X_Mg 0 1 .1 0 | range and resolution of X(Mg) X_Fe by difference begin_ordered_endmembers oscf = 1/2 mscf + 1/2 fscf delta_g_of_ordering = -3.5d3 end_ordered_endmembers begin_dependent_endmembers facf_d = 1 macf + 1 oscf - 1 mscf cscf_d = 1 cacf + 1 mscf - 1 macf cfscf_d = 1 cacf + 1 fscf - 1 macf - 1 oscf + 1 mscf end_dependent_endmembers begin_excess_function w(macf cacf) 11d3 w(macf mscf) 7.5d3 w(macf fscf) 10.75d3 w(macf oscf) 11.5d3 w(macf nacfb_q) 24.5d3 w(cacf mscf) 18.5d3 w(cacf fscf) 14.45d3 w(cacf oscf) 15.2d3 w(cacf nacfb_q) 18.5d3 w(mscf fscf) 5d3 w(mscf oscf) 4d3 w(mscf nacfb_q) 15.5d3 w(fscf oscf) 1d3 w(fscf nacfb_q) 9.95d3 w(oscf nacfb_q) 10.7d3 end_excess_function 2 | 2 site entropy model (M1, M2) M1 | site name 4 1 1 | 4 species on M1, site multiplicity of 1 z(Ca) = 1 cacf z(Fe) = 1 fscf + 1 oscf z(Na) = 1 nacfb_q z(Mg) = 1 macf + 1 mscf M2 | site name 4 1 2 | 4 species on M2, (fake) site multiplicity of 1 z(Al) = 1 macf + 1 cacf + 1/2 nacfb_q z,Mg) = 1/2 mscf + 1/2 oscf z(Fe) = 1/2 fscf z(Si) = 1/2 mscf + 1/2 oscf + 1/2 fscf + 1/2 nacfb_q [O4] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(nacfb_q) = -9000 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model NAL phase, Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 Notes: Originally entered by Bob Myhill, 2018/02/13. The M1 site has a fake multiplicity of 3 (6 apfu) Eliminated ordered dependent endmembers. JADC, 10/2018. Converted to 688 format. JADC, 3/7/2020. M3 M2 M1 ___________________ Multiplicity 1 2 3*(6) ___________________ 1 nanal Na Mg Al5/6Si1/6 independent 3 canal Ca Mg Al independent 5 manal Mg Mg Al independent 9 msnal Mg Mg Mg1/2Si1/2 independent 2 nfnal_d Na Fe2+ Al5/6Si1/6 dependent 4 cfnal_d Ca Fe2+ Al dependent 8 fanal_d Fe2+ Fe2+ Al dependent 16 fsnal Fe2+ Fe2+ Fe1/2Si1/2 independent ___________________ Ordered: 9 o1nal Fe2+ Mg Mg1/2Si1/2 ordered 10 o2nal Fe2+ Fe2+ Mg1/2Si1/2 ordered NAl(H) abbreviation NAl full_name New aluminous phase 688 | model type: 688 format standard model 1 | number of prisms | ---------------------------- 2 | number of simplices in the prism 2 4 | number of vertices on each simplex nanal nfnal_d canal cfnal_d manal fanal_d msnal fsnal X_MgM2 0 1 .1 0 | range and resolution for X(Mg) on site 1, imod = 0 -> cartesian subdivision X_FeM2 by difference X_NAM13 0 1 .1 0 | range and resolution for X(nanal) on site 2, imod = 0 -> cartesian subdivision X_CAM13 0 1 .1 0 | range and resolution for X(canal) on site 2, imod = 0 -> cartesian subdivision X_MAM13 0 1 .1 0 | range and resolution for X(manal) on site 2, imod = 0 -> cartesian subdivision X_MSM13 by difference begin_ordered_endmembers o1nal = 5/6 msnal + 1/6 fsnal delta_g_of_ordering = 2d3 o2nal = 1/2 msnal + 1/2 fsnal delta_g_of_ordering = 6.5d3 end_ordered_endmembers begin_dependent_endmembers nfnal_d = 1 nanal - 1 o1nal + 1 o2nal | NM - FM + FF = NF cfnal_d = 1 canal - 1 o1nal + 1 o2nal | CM - FM + FF = CF fanal_d = 1 manal - 1 msnal + 1 o2nal | MM - MM + FF = FF end_dependent_endmembers begin_excess_function W(nanal canal) 14.2d3 W(nanal manal) 20.2d3 W(nanal msnal) 20.2d3 W(nanal fsnal) 21.2d3 W(nanal o1nal) 15.4d3 W(nanal o2nal) 23.4d3 W(canal manal) 11d3 W(canal msnal) 33.5d3 W(canal fsnal) 36d3 W(canal o1nal) 30.2d3 W(canal o2nal) 38.2d3 W(manal msnal) 22.5d3 W(manal fsnal) 32.2d3 W(manal o1nal) 26.5d3 W(manal o2nal) 34.5d3 W(msnal fsnal) 15d3 W(msnal o1nal) 4d3 W(msnal o2nal) 12d3 W(fsnal o1nal) 11d3 W(fsnal o2nal) 3d3 W(o1nal o2nal) 8d3 end_excess_function 3 | 3 site (M3, M2, M1) entropy model | (note: occupancies should not include dependent endmembers, | but do include ordered endmembers) M3 4 1 1 | 4 species on M3, effective site multiplicity, true multiplicity z(Na,M3) = 1 nanal z(Ca,M3) = 1 canal z(Mg,M3) = 1 manal + 1 msnal z(Fe,M3) = 1 fsnal + 1 o2nal + 1 o1nal M2 2 2 2 | 2 species on M2, effective site multiplicity, true multiplicity z(Fe,M2) = 1 fsnal + 1 o2nal z(Mg,M2) = 1 nanal + 1 canal + 1 manal + 1 msnal + 1 o1nal M1 4 3 6 | 4 species on M1, effective site multiplicity, true multiplicity z(Mg,M1) = 1/2 msnal + 1/2 o1nal + 1/2 o2nal z(Fe,M1) = 1/2 fsnal z(Al,M1) = 5/6 nanal + 1 canal + 1 manal z(Si,M1) = 1/6 nanal + 1/2 fsnal + 1/2 msnal + 1/2 o1nal + 1/2 o2nal [look-it-up] | formula suffix, enter "none" for no suffix. end_of_model -------------------------------------------------------- begin_model Akimotoite, Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 [Bob Myhill, 2018/02/13] mixing on 2 sites, each with a fake site multiplicity of 1/2 (Al-avoidance). Note that this is different to the corundum model, which is assumed to mix on two independent sites. The properties of the aak endmember are assumed identical to corundum (cor): 1 2 M T _____________ Multiplicity 1/2*(1) 1/2*(1) _____________ 1 cor Al Al Species: 2 mak Mg Si 3 fak Fe2+ Si Aki(H) abbreviation Aki full_name Akimotoite 2 | model type: Margules, macroscopic 3 | 3 endmembers cor mak fak 0 0 0 0 1 .1 0 0 1 .1 0 begin_excess_function W(cor mak) 8d3 W(cor fak) 6d3 W(cor fak) 4d3 end_excess_function 2 2 site entropy model 3 .5 3 species on M site with fake multiplicity = 1/2 z(mg) = 1 mak z(fe) = 1 fak 2 .5 2 species on T site with fake multiplicity = 1/2 z(al) = 1 cor end_of_model -------------------------------------------------------- begin_model Wadsleyite Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 [Bob Myhill, 2018/02/13] Notes: mixing of Mg-Fe on 1 site with a multiplicity of 2 Wad(H) abbreviation Wad full_name wadsleyite 2 model type: Margules, macroscopic 2 2 endmembers mwd fwd 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision begin_excess_function W(mwd fwd) 13000 end_excess_function 1 1 site entropy model 2 2 2 species, site multiplicity = 2 z(fe) = 1 fwd end_of_model -------------------------------------------------------- begin_model Ringwoodite Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 [Bob Myhill, 2018/02/13] Notes: mixing of Mg-Fe on 1 site with a multiplicity of 2 Ring(H) abbreviation Ring full_name ringwoodite 2 model type: Margules, macroscopic 2 2 endmembers mrw frw 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision begin_excess_function W(mrw frw) 4000 end_excess_function 1 1 site entropy model 2 2. 2 species, site multiplicity = 2. z(fe) = 1 frw end_of_model -------------------------------------------------------- begin_model High pressure C2/c clinopyroxene, Holland et al., 2013; dx.doi.org/10.1093/petrology/egt035 Notes: Originally entered by Bob Myhill, 2018/02/13 Eliminated ordered dependent endmembers. JADC, 10/2018. odi has been replaced by an internal make 1/2026, JADC di = 1 di DQF = -100 + 0.211 * T_K + 0.005 * P_bar M1 M2 T _____________________ Multiplicity 1 1 1/2 _____________________ hen Mg Mg SiSi independent Species: di Mg Ca SiSi independent hmgts Al Mg SiAl independent Species: hfs Fe Fe SiSi independent tsfs_d Al Fe SiAl dependent ofdi_d Fe Ca SiSi dependent _____________________ Internal: hfm Mg Fe SiSi Hpx(H) abbreviation Hpx full_name HP_clinopyroxene 8 | model type: Reciprocal with speciation 2 | 2 independent composition spaces 3 2 | 3 dimensions on first space, 2 on second | endmembers: hen di hmgts_q hfs tsfs_d ofdi_d 1 | ordered species definition hfm = 1/2 hen + 1/2 hfs Delta(enthalpy) = -6d3 2 | dependent endmember definitions ofdi_d = 1 di + 1 hfs - 1 hfm tsfs_d = 1 hmgts_q + 1 hfm - 1 hen 0 0 0 0 0 0 0 0 0 | endmember flags, indicate if endmember is part of the solution (i.e. iend = 0). 0 1 .1 0 | subdivision model for (ternary) site 1 (X_en): 0 1 .1 0 | subdivision model for (ternary) site 1 (X_di): 0 1 .1 0 | subdivision model for (binary) site 2 (XMg): begin_excess_function W(hen hfs) 5.2d3 W(hen hfm) 4d3 W(hen di ) 32.2d3 0 0.12 W(hen hmgts_q) 13d3 0 -0.15 W(hfs hfm) 4d3 W(hfs di ) 24d3 W(hfs hmgts_q) 7d3 0 -0.15 W(hfm di ) 18d3 W(hfm hmgts_q) 2d3 0 -0.15 W(di hmgts_q) 75.4d3 0 -0.94 end_excess_function 3 | 3 site (M1, M2, T) configurational entropy model 3 1 | 3 species on M1, 1 site per formula unit. z(m1,fe) = 1 hfs z(m1,al) = 1 hmgts_q 3 1 | 3 species on M2, 1 site per formula unit. z(m2,ca) = 1 di z(m2,fe) = 1 hfs + 1 hfm 2 .5 | 2 species on .5 T sites (true multiplicity = 2) z(t,al) = 1/2 hmgts_q begin_van_laar_sizes alpha(hen) 1 alpha(hfs) 1 alpha(hfm) 1 alpha(di ) 1.2 alpha(hmgts_q) 1 end_van_laar_sizes begin_dqf_corrections dqf(di) = -100 + 0.211 *T 0.005 *P | from White et al. 2014. DQF(hmgts_q) = -1d3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Pyrrhotite, Saxena & Eriksson 2015 ECRG, 5/2018 Po(SE) | solution name. abbreviation Po full_name hT-pyrrhotite 2 | model type: Compound energy formalism 2 | number of endmembers APyrr SVa 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision begin_excess_function Wk(APyrr SVa) w0 = -225830.67 wT = 26.357836 wP0 = -0.400228 end_excess_function | wP0 is a coefft of P 1 | 1 site entropy model 2 1 | 2 species, site multiplicity of 1 z(SVa) = 1 SVa end_of_model -------------------------------------------------------- begin_model FeS-S fluid, Saxena & Eriksson 2015 ECRG, 5/2018 FeS_liq abbreviation Melt full_name liquid 42 | model type: modified QC with composition-dependent coordination, binary only. 2 | no. independent end-members FeLiq SLiq | endmember names 0 0 | endmember flags 0 1 .1 0 | subdivision range, imod = 0 -> cartesian subdivision | model has an excess function and configurational entropy ideal 0 | entropy function is hard-wired end_of_model -------------------------------------------------------- begin_model Talc (Frueh-Green et al., 2004, AGU Geophysical Monograph 144:119-136) originally from Connolly, 1994, Phase diagram applications in earth material sciences, VII Summer Schoool Proceedings, University of Siena). 1 2 M1 M2 T2 ______________________ Mutliplicity 2 1 2 ______________________ 1 en Mg Mg SiSi Species: 2 fs Fe Fe SiSi 3 mgts Mg Al AlSi ______________________ reformulated from relict equipartion (model type 7) to simplicial composition space (model type 2). JADC 5/10/2018 T(F) | solution name abbreviation Tlc full_name talc 2 | model type: simplicial, equipartition relict 3 | number of independent endmembers ta fta tats | endmember names 0 0 0 | endmember flags, indicate if the endmember is part of the solution. 0 1 .1 0 | range and resolution for ta, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for fta, imod = 0 -> cartesian subdivision ideal 3 | 3 site (M1, M2, T2) conigurational entropy model 2 2. | 2 species on M1, 2 sites per formula unit. z(m1,mg) = 1 ta + 1 tats 2 2. | 2 species on T2, 2 sites per formula unit. z(t2,al) = 1/2 tats 3 1 | 3 species on M2, 1 site per formula unit. z(m2,mg) = 1 ta z(fe,m2) = 1 fta end_of_model -------------------------------------------------------- begin_model Antigorite with Tschermak's substitution (Padron-Navarta et al., 2013, Lithos) ferro-atg enemdmeber from Frueh-Green et al., AGU Geophysical Monograph (144:119-136) reformulated from relict equipartion (model type 7) to simplicial composition space (model type 2). JADC 5/10/2018 M0 M1 T1 ______________________ Mutliplicity 44 4 8 ______________________ 1 atg Mg Mg SiSi Species: 2 fatg_q Fe Fe SiSi 3 atgts_q Mg Al AlSi ______________________ --------------------------------------------------------- Atg(PN) | solution name abbreviation Atg full_name serpentine 2 | model type: simplicial composition space, equipartition relict 3 | number of endmembers atg fatg_q atgts_q | endmember names, this order implies: 0 0 0 | endmember flags, indicate if the endmember is part of the solution. 0 1 .1 0 | range and resolution for atg, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for ftag, imod = 0 -> cartesian subdivision ideal 3 | 3 site (M0, M1, T1) configurational entropy model 2 44. | 2 species on M0, 2 sites per formula unit. z(m1,mg) = 1 atg + 1 atgts_q 2 8. | 2 species on T1, 2 sites per formula unit. z(t2,al) = 1/2 atgts_q 3 4. | 3 species on M1, 1 site per formula unit. z(m2,mg) = 1 atg z(fe,m2) = 1 fatg_q begin_dqf_corrections dqf(fatg_q) = 0 dqf(atgts_q) = -2000 + 46.1 * T end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Full ferri-tschermak solid solution model for antigorite from Eberhard et al. 2023, JPet, doi.org/10.1093/petrology/egad069 All endmembers reformulated to match antigorite stoichiometry of 17(M0)-31(M1)-17(T0)-17(T1). DQF terms adjusted to match experiments and natural samples NOTE: the names of the new atigorite endmembers in Eberhard et al (2023) have been changed to prevent conflicts with the earlier solution models. Converted to 688 format, JADC, 9/27/2023 Eliminated useless ordered oatg endmember, JADC 10/1/2023 M1 M0 T1 ___________________ Mutliplicity 17 31 17 ___________________ 1 atg Mg Mg Si Species: 2 fatg_q Fe Fe Si 3 atgts_q Al Mg Al 4 f3atg_q Fe3+ Mg Al ____________________ NB! to use Chl(W) with Atg(LE) and Liz(LE) models require that the internal dqf of ferri-clinchlore be modified from its original value: DQF(f3cl_q) = 2000 to DQF(f3cl_q) = 40000 In essence the f3cl_q "modification" is bullpucky, it simply eliminates ferric iron from Chl and has no experimental basis. Admittedly the White version is excessively stable. -------------------------------------------------------- Atg(LE) | solution name abbreviation Atg full_name serpentine 688 | model type: 688 format 1 | number of polytopes 1 | number of simplices 4 | number of vertices on each simplex atg fatg_q atgAl_q f3atg_q | endmembers on the vertices 0 1 .1 0 | range and resolution for atg, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for fatg_q, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for atgAl_q, imod = 0 -> cartesian subdivision ideal 3 | 3 site (M0, M1, T1) configurational entropy model M0 | site name 2 31 31 | number of species, effective multiplicity, true multiplicity z(Mg,M1) = 1 atg + 1 f3atg_q + 1 atgAl_q z(Fe,M1) = 1 fatg_q M1 | site name 4 17 17 | number of species, effective multiplicity, true multiplicity z(Mg,M1) = 1 atg z(Fe,M1) = 1 fatg_q z(Al,M1) = 1 atgAl_q z(Fe3,M1) = 1 f3atg_q T1 | site name 2 17 17 | number of species, effective multiplicity, true multiplicity z(Al,T1) = 1 atgAl_q + 1 f3atg_q z(Si,T1) = 1 atg + 1 fatg_q [O85(OH)62] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(fatg_q) = -350e3 dqf(atgAl_q) = -450e3 dqf(f3atg_q) = -127e3 -3.4 * P_bar end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Full ferri-tschermak solid solution model for lizardite from Eberhard et al. 2023, JPet, doi.org/10.1093/petrology/egad069 All endmembers reformulated to match stoichiometry of 1(M0)-2(M1)-1(T1). DQF terms adjusted to match experiments and natural samples Eliminated useless ordered oliz endmember, JADC, 10/1/2023 M1 M0 T1 _____________________ Mutliplicity 1 2 1 _____________________ 1 liz Mg Mg Si Species: 2 fliz Fe Fe Si 3 lizts Al Mg Al 4 f3liz Fe3+ Mg Al ____________________ This model requires the following make definition in the thermodynamic data: fliz = 1 liz - 1 ta + 1 fta, DQF(J/mol) = -21e3 lizts = 1 liz - 1 ta + 1 tats, DQF = -37e3 +0.1 * P_bar f3liz = 1 liz - 1 ta + 1 tats - 0.5 gr + 0.5 andr, DQF = -18e3 -0.1 * P_bar NB! to use Chl(W) with Atg(LE) and Liz(LE) models require that the internal dqf of ferri-clinchlore be modified from its original value: DQF(f3cl_q) = 2000 to DQF(f3cl_q) = 40000 In essence the f3cl "modification" is bullpucky, it simply eliminates ferric iron from Chl and has no experimental basis. Admittedly the White version is excessively stable. -------------------------------------------------------- Liz(LE) | solution name abbreviation Liz full_name serpentine 2 | model type: simplicial composition space 4 | number of endmembers liz fliz lizts f3liz | endmember names, this order implies: 0 0 0 0 0 | endmember flags, indicate if the endmember is part of the solution. 0 1 .1 0 | range and resolution for liz, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for fliz, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for lizts, imod = 0 -> cartesian subdivision ideal 3 | 3 site (M0, M1, T1) configurational entropy model 4 1. | 4 species on M1, 1 sites per formula unit. z(m1,mg) = 1 liz z(m1,fe) = 1 fliz z(m1,al) = 1 lizts 2 1. | 2 species on T1, 1 sites per formula unit. z(t1,al) = 1 lizts + 1 f3liz 2 2. | 2 species on M0, 1 site per formula unit. z(m2,mg) = 1 liz + 1 lizts + 1 f3liz end_of_model -------------------------------------------------------- begin_model Amphibole from Massonne & Willner (EJM, 2008) See notes for TrTsPg (above). GlTrTsMr abbreviation Amph full_name clinoamphibole 7 | model type: reciprocal, margules 2 | 2 site reciprocal solution 2 4 | 1 binary and 1 quaternary tr ftr mrie fmrie_i ts_q fts_i gl fgl_i 3 | number of dependent endmembers fmrie_i = 1 mrie + 3/5 ftr - 3/5 tr fts_i = 1 ts_q + 3/5 ftr - 3/5 tr fgl_i = 1 gl + 3/5 ftr - 3/5 tr 0 0 0 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution for X(Mg) on site 1 0 1 .1 0 | range and resolution for X(tr) on site 2 0 1 .1 0 | range and resolution for X(pg) on site 2 0 1 .1 0 | range and resolution for X(ts) on site 2 begin_excess_function W(gl tr) 77d3 W(gl ftr) 83d3 W(ts_q tr) 20d3 W(ts_q ftr) -38d3 W(tr ftr) 10d3 end_excess_function 4 | 4 site (M1, M2, M4, T1) entropu model 2 2. | 2 species on T1, fake site multiplicity of 2. z(T1,Al) = 0 + 1 ts_q 2 3. | 2 species on M1, 3 sites per formula unit z(m1,mg) = 0 + 1 tr + 1 ts_q + 1 mrie + 1 gl 3 2. | 3 species on M2, 2 sites pfu z(m2,mg) = 0 + 1 tr z(m2,fe) = 0 + 1 ftr 2 2. | 2 species on M4, 2 sites pfu z(m4,na) = 0 + 1 gl begin_dqf_corrections dqf(ts_q) 10000 end_dqf_corrections site_check_override end_of_model -------------------------------------------------------- begin_model Dale et al, CMP 2000 140:353-362 amphibole model without Na, K, Ti or Mn solution. See Amph(DHP) or bAmph(DHP) for Na-Ca amphibole JADC 5/5/06. A M1 M2 M4 T1 _________________________________________ Mutliplicity 1 3 2 2 2 _________________________________________ 1 tr Vac Mg Mg Ca Si_Si 2 ftr Vac Fe Fe Ca Si_Si 3 ts Vac Mg Al Ca Al_Si 4 fts Vac Fe Al Ca Al_Si 5 parg Na Mg Mg_Al Ca Al_Si 6 fparg Na Fe Fe_Al Ca Al_Si 9 mfets Vac Mg Fe3+ Ca Al_Si 10 ffets Vac Fe Fe3+ Ca Al_Si fparg = parg + 4/5 (ftr - tr) fts = ts + 3/5 (ftr - tr) Ca-Amph(D) | solution name abbreviation Amph full_name clinoamphibole 7 | model type: reciprocal, margules 2 | 2 site reciprocal solution 2 4 | 1 binary and 1 quinary tr ftr parg fparg_i ts_q fts_i mfets ffets_i 3 | number of dependent endmembers fparg_i = 1 parg + 4/5 ftr - 4/5 tr fts_i = 1 ts_q + 3/5 ftr - 3/5 tr ffets_i = 1 mfets + 3/5 ftr - 3/5 tr 0 0 0 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution for X(Mg) on site 1 0 1 .1 0 | range and resolution for X(tr) on site 2 0 1 .1 0 | range and resolution for X(pg) on site 2 0 1 .1 0 | range and resolution for X(ts) on site 2 begin_excess_function W(parg tr) 29.3d3 W(parg ts_q) 18.2d3 W(parg ftr) 11.4d3 W(ts_q tr) 20.8d3 W(tr ftr) 11.4d3 end_excess_function 4 | 4 site (A, M1, M2, T1) entropy model 2 1 | 2 species on A (V, Na), 1 site per formula unit. z(A,Na) = 1 parg 2 2. | 2 species on T1, fake site multiplicity of 2. z(T1,Al) = 1/2 ts_q + 1/2 parg + 1/2 mfets 2 3. | 2 species on M1, 3 sites per formula unit z(m1,mg) = 1 tr + 1 ts_q + 1 parg + 1 mfets 4 2. | 4 species on M2, 2 sites pfu z(m2,mg) = 1 tr + 1/2 parg z(m2,fe) = 1 ftr z(m2,fe3+) = 1 mfets begin_dqf_corrections dqf(ts_q) 10000 end_dqf_corrections site_check_override end_of_model -------------------------------------------------------- begin_model Dale et al, CMP 2000 140:353-362 amphibole model without Ca, K, Ti or Mn solution. This model requires a fgl endmember, created as decribed by Powell's mdep paper. See Amph(DHP) for Na-Ca amphibole JADC 5/5/06. A M1 M2 M4 T1 _________________________________________ Mutliplicity 1 3 2 2 2 _________________________________________ 7 gl Vac Mg Al Na Si_Si 8 fgl Vac Fe Al Na Si_Si 12 mrieb Vac Mg Fe3+ Na Si_Si 11 rieb Vac Fe Fe3+ Na Si_Si mrieb_i = 1 rieb + 3/5 tr - 3/5 ftr Na-Amph(D) | solution name abbreviation Amph full_name clinoamphibole 7 | model type: reciprocal, margules 2 | 2 site reciprocal solution 2 2 | 2 binaries gl fgl mrieb_i rieb 1 | number of dependent endmembers mrieb_i = 1 rieb + 1 gl - 1 fgl 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution for X(Mg) on site 1 0 1 .1 0 | range and resolution for X(Al) on site 2 ideal 2 | 2 site (M1, M2) entropy model 2 3. | 2 species on M1, 3 sites per formula unit z(m1,mg) = 1 gl 2 2. | 2 species on M2, 2 sites pfu z(m2,fe3+) = 1 rieb site_check_override end_of_model begin_model -------------------------------------------------------- begin_model Ideal orthoamphibole, this model assumes Al is present on only two tetrahedral sites and all five M2 sites. I have no idea if this is correct! fgedr endmember stoichiometry corrected, T. Wagner 2/18/06. M1 M2 T ______________________ Mutliplicity 2 5 2 ______________________ 1 anth Mg Mg SiSi Species: 2 fanth Fe Fe SiSi 3 ged Mg Mg3Al2 AlAl 4 fged Fe Fe3Al2 AlAl ______________________ Dependent: fged = ged + 5/7*(fanth - anth) o-Amph | solution name abbreviation oAmph full_name orthoamphibole 7 | model type: reciprocal, macroscopic 2 | 2 site reciprocal solution 2 2 | 2 species on each site anth fanth | endmember names, this order implies: ged fged_i | x(11)=x(mg); x(12) = x(fe); x(21) = x(si,t); x(22) = x(Al,t) 1 | 1 dependent endmember: fged_i = 1 ged + 5/7 fanth - 5/7 anth 0 0 0 0 | endmember flags, indicate if the endmember is part of the solution. 0 1 .1 | range and resolution for X(Mg) on site 1 0 | subdivision scheme on site 2: imod = 0 -> cartesian 0 1 .1 | range and resolution for 1-X(Tschermaks) on site 2 0 | subdivision scheme on site 2: imod = 0 -> cartesian ideal 3 | 3 site (M1, M2, T) conigurational entropy model 2 2. | 2 species on M1, 2 sites per formula unit. z(m1,mg) = 1 anth + 1 ged 2 2. | 2 species on T, 2 sites per formula unit. z(t,al) = 1 ged 3 1 | 3 species on M2, 1 site per formula unit. z(m2,mg) = 1 anth + 3/5 ged z(m2,fe) = 1 fanth site_check_override end_of_model -------------------------------------------------------- begin_model tr-ts-parg non-ideal model for holland and powell. assumes 2 M2 sites are coupled to 4 T1 sites. site multiplicity of the T1 site is reduced to 2, this is suggested by HP98 to account for charge balance constraints. but doesn't make a lot of sense for the tr-parg mixing. assume Na on the A-site is coupled to Al on M2. JADC Nov, 98. HP Am Min 99, 84:1-14 Oli Jagoutz revised april 9, 2002 in contrast to the earlier version of TrTsPg this version assumes the A site is decoupled from M2 JADC 4/03. fparg = parg + 4/5 (ftr - tr) fgl = gl + 3/5 (ftr - tr) fts = ts + 3/5 (ftr - tr) GlTrTsPg | solution name abbreviation Amph full_name clinoamphibole 7 | model type: reciprocal, margules 2 | 2 site reciprocal solution 2 4 | 1 binary and 1 quaternary tr ftr parg fparg_i ts_q fts_i gl fgl_i 3 | number of dependent endmembers fparg_i = 1 parg + 4/5 ftr - 4/5 tr fts_i = 1 ts_q + 3/5 ftr - 3/5 tr fgl_i = 1 gl + 3/5 ftr - 3/5 tr 0 0 0 0 1 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution for X(Mg) on site 1, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(tr) on site 2, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(pg) on site 2, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(ts) on site 2, imod = 0 -> cartesian subdivision begin_excess_function | interaction parameters from | White, Powell & Phillips (2003, JMG) | and Wei, Powell, & Zhang (2003, JMG) | compiled by D. Tinkham. JADC 11/03 W(parg gl) 80d3 W(parg tr) 30d3 W(parg ftr) 38d3 W(gl tr) 77d3 W(gl ftr) 83d3 W(ts_q tr) 20d3 W(ts_q ftr) -38d3 W(tr ftr) 10d3 | earlier versions used (provenance unknown) | W(ts parg) -25000. | W(tr parg) 20000. | W(tr ts) 38000 end_excess_function 5 | 5 site (A, M1, M2, M4, T1) entropy model 2 1 | 2 species on A (V, Na), 1 site per formula unit. z(A,Na) = 1 parg 2 2. | 2 species on T1, fake site multiplicity of 2. z(T1,Al) = 1/2 ts_q + 1/2 parg 2 3. | 2 species on M1, 3 sites per formula unit z(m1,mg) = 1 tr + 1 ts_q + 1 parg + 1 gl 3 2. | 3 species on M2, 2 sites pfu z(m2,mg) = 1 tr + 1/2 parg z(m2,fe) = 1 ftr 2 2. | 2 species on M4, 2 sites pfu z(m4,na) = 1 gl begin_dqf_corrections dqf(ts_q) 10000 end_dqf_corrections site_check_override end_of_model -------------------------------------------------------- begin_model A M1 M2 M4 T1 _________________________________________ Mutliplicity 1 3 2 2 2 _________________________________________ 1 tr Vac Mg Mg Ca Si_Si 2 ftr Vac Fe Fe Ca Si_Si 3 ts Vac Mg Al Ca Al_Si 4 fts Vac Fe Al Ca Al_Si 5 parg Na Mg Mg_Al Ca Al_Si 6 fparg Na Fe Fe_Al Ca Al_Si 7 gl Vac Mg Al Na Si_Si 8 fgl Vac Fe Al Na Si_Si 9 mfets Vac Mg Fe3+ Ca Al_Si 10 ffets Vac Fe Fe3+ Ca Al_Si Dale et al, CMP 2000 140:353-362 amphibole model without K, Ti or Mn solution. JADC 9/05. fparg = parg + 4/5 (ftr - tr) fgl = gl + 3/5 (ftr - tr) fts = ts + 3/5 (ftr - tr) Amph(DHP) | solution name abbreviation Amph full_name clinoamphibole 7 | model type: reciprocal, margules 2 | 2 site reciprocal solution 2 5 | 1 binary and 1 quinary tr ftr parg fparg_i ts_q fts_i gl fgl_i mfets ffets_i 4 | number of dependent endmembers fparg_i = 1 parg + 4/5 ftr - 4/5 tr fts_i = 1 ts_q + 3/5 ftr - 3/5 tr fgl_i = 1 gl + 3/5 ftr - 3/5 tr ffets_i = 1 mfets + 3/5 ftr - 3/5 tr 0 0 0 0 1 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution for X(Mg) on site 1, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(tr) on site 2, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(pg) on site 2, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(ts) on site 2, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(gl) on site 2, imod = 0 -> cartesian subdivision begin_excess_function W(parg gl) 84.5d3 W(parg tr) 29.3d3 W(parg ts_q) 18.2d3 W(parg ftr) 11.4d3 W(gl tr) 35.3d3 W(ts_q tr) 20.8d3 W(tr ftr) 11.4d3 W(gl ftr) 15d3 W(gl ts_q) 15d3 end_excess_function 5 | 5 site (A, M1, M2, M4, T1) entropy model 2 1 | 2 species on A (V, Na), 1 site per formula unit. z(A,Na) = 1 parg 2 2. | 2 species on T1, fake site multiplicity of 2. z(T1,Al) = 1/2 ts_q + 1/2 parg + 1/2 mfets 2 3. | 2 species on M1, 3 sites per formula unit z(m1,mg) = 1 tr + 1 ts_q + 1 parg + 1 gl + 1 mfets 4 2. | 4 species on M2, 2 sites pfu z(m2,mg) = 1 tr + 1/2 parg z(m2,fe) = 1 ftr z(m2,fe3+) = 1 mfets 2 2. | 2 species on M4, 2 sites pfu z(m4,na) = 1 gl begin_dqf_corrections dqf(ts_q) 10000 end_dqf_corrections site_check_override end_of_model -------------------------------------------------------- begin_model Dale et al, JMG 2005 23:771-791 amphibole model. JADC, 11/05. Margules parameters corrected from W(gl ftr) 393d3 W(gl mfets) 459d3 W(ftr mfets) 125d3 to current values. M. Racek, 2/10/06. A M13 M2 M4 T1* _________________________________________ Mutliplicity 1 3 2 2 4(1) _________________________________________ 1 tr Vac Mg Mg Ca Si_Si 2 ftr Vac Fe Fe Ca Si_Si 3 ts Vac Mg Al Ca Al_Si 4 fts_d Vac Fe Al Ca Al_Si 5 parg Na Mg Mg_Al Ca Al_Si 6 fparg_d Na Fe Fe_Al Ca Al_Si 7 gl Vac Mg Al Na Si_Si 8 fgl_d Vac Fe Al Na Si_Si 9 mfets Vac Mg Fe3+ Ca Al_Si 10 ffets_d Vac Fe Fe3+ Ca Al_Si fparg_d = 1 parg + 4/5 ftr - 4/5 tr fts_d = 1 ts + 3/5 ftr - 3/5 tr fgl_d = 1 gl + 3/5 ftr - 3/5 tr ffets_d = 1 mfets + 3/5 ftr - 3/5 tr *Dale et al compute amphibole T1 site fractions assuming a site multiplicity of 4, but compute activities for a T1 site multiplicity of 1 In previous models H&P computed activities for a T1 site multiplicity of 2. Amph(DPW) | solution name abbreviation Amph full_name clinoamphibole 7 | model type: reciprocal, margules 2 | 2 site reciprocal solution 2 5 | 1 binary and 1 quinary tr ftr parg fparg_i ts_q fts_i gl fgl_i mfets ffets_i 4 | number of dependent endmembers fparg_i = 1 parg + 4/5 ftr - 4/5 tr fts_i = 1 ts_q + 3/5 ftr - 3/5 tr fgl_i = 1 gl + 3/5 ftr - 3/5 tr ffets_i = 1 mfets + 3/5 ftr - 3/5 tr 0 0 0 0 1 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution for X(Mg) on site 1, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(tr) on site 2, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(pg) on site 2, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(ts) on site 2, imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for X(gl) on site 2, imod = 0 -> cartesian subdivision begin_excess_function W(tr ts_q) 20d3 W(tr parg) 33d3 W(tr gl) 65d3 W(tr ftr) 10d3 W(tr mfets) 20d3 W(ts_q parg) -385d2 W(ts_q gl) 25d3 W(ts_q ftr) 125d2 W(parg gl) 50d3 W(parg ftr) -19d2 W(parg mfets) -385d2 W(gl ftr) 393d2 W(gl mfets) 459d2 W(ftr mfets) 125d2 end_excess_function 5 | 5 site (A, M13, M2, M4, T1) entropy model 2 1 | 2 species on A (V, Na), 1 site per formula unit. z(A,Na) = 1 parg 2 1 | 2 species on T1, fake site multiplicity of 1. z(T1,Al) = 1/2 ts_q + 1/2 parg + 1/2 mfets 2 3. | 2 species on M1, 3 sites per formula unit z(m1,mg) = 1 tr + 1 ts_q + 1 parg + 1 gl + 1 mfets 4 2. | 4 species on M2, 2 sites pfu z(m2,mg) = 1 tr + 1/2 parg z(m2,fe) = 1 ftr z(m2,fe3+) = 1 mfets 2 2. | 2 species on M4, 2 sites pfu z(m4,na) = 1 gl begin_van_laar_sizes alpha(tr) 1 alpha(ts_q) 1.5 alpha(parg) 1.7 alpha(gl) 0.8 alpha(ftr) 1 alpha(mfets) 1.5 end_van_laar_sizes begin_dqf_corrections dqf(gl) 5d3 dqf(ts_q) 1d4 dqf(parg) 15d3 end_dqf_corrections site_check_override end_of_model -------------------------------------------------------- begin_model Sapphirine, ideal, holland and powell '98 ---------------------------------------------------------- refer to the commentary of Sapp(KWP) for a resume of Sapphirine models in solution_model.dat ---------------------------------------------------------- config entropy corrected, P Goncalves/JADC, 10/1/03 the corrected model assumes (after the text on TJBH's saphhirine web page www.esc.cam.ac.uk/astaff/holland/ds5/sapphirines/spr.html) that: 1) a 14 cation unit formula 2) Si occupies T2, Al occupies T5, Si and Al mix on remaining 4 sites T1 T3 T4 T6 (the T site below) 3) Al occupies M7; only Fe and Mg may occupy sites M4, M5, M6 (Site MB below); Al, Mg, and Fe may occupy sites M1, M2, M3 and M8 (Site MA below) N.B. This model seems to differ from the Thermocalc format version on TJBH's web page in that it accounts for the configurational entropy arising from mixing on MB (i.e., the Thermocalc model looks like it was written for Fe-free sapphirine). Eliminate site MB to reproduce the TJBH web page Thermocalc model. 1 2 3 MA MB T _________________ Mutliplicity 4 3 4 _________________ 1 spr7 MgAl7 Mg SiAl7 Species: 2 fspr FeAl7 Fe SiAl7 3 spr4 MgAl3 Mg SiAl3 4 fsp4_i FeAl3 Fe SiAl3 Dependent endmember: fsp4_i = spr4 + 8/7 * (fspr - spr7) Sapp(HP) abbreviation Sap full_name sapphirine 7 model type 2 reciprocal solution 2 2 2 species on each site spr7 fspr spr4 fsp4_i 1 1 dependent endmember fsp4_i = 1 spr4 + 8/7 fspr - 8/7 spr7 0 0 0 0 endmember flags 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for 1-X(Tschermaks), imod = 0 -> cartesian subdivision ideal 3 | 3 site (MA, MB, T) configurational entropy model 2 3. | 2 species on MB, 3 sites per formula unit z(mb,mg) = 1 spr4 + 1 spr7 3 4. | 3 species on MA, 4 sites per formula unit. z(ma,Al) = 3/4 + 1/8 spr7 + 1/8 fspr z(ma,fe) = 1/8 fspr 2 4. | 2 species on T, 4 sites per formula unit. z(T,Al) = 3/4 + 1/8 spr7 + 1/8 fspr site_check_override end_of_model -------------------------------------------------------- begin_model Sapphirine, non-ideal, Kelsey et al. (J. metamorphic Geol., 2004, 22, 559-578) NOTE: This model should be used in conjunction with a special high temperature version of the HP data base (kel04ver.dat). ---------------------------------------------------------- A resume of the HP sapphirine solution models in solution model.dat: in the h&p world sapphirine has 14 cation sites per 20 oxygen and, taking the mg-al-si endmembers as an example, 3 limiting compositions (endmembers): spr4 - Mg(4)Si(2)Al(8)O(20) spr5 - Mg(3)Si(1)Al(10)O(20) spr7 - Mg(3.5)Si(1.5)Al(9)O(20) In terms of divalent (Mg,Fe2+) and trivalent (Al,Fe3+) ratio, again simplifying by only considering Mg and Al, there are 3 sapphirine models in perplex: Sa(KW) - 2014, between spr4 and spr5 - has the widest Al-range. In this model, sapphirine has 4 octahedral (M) sites and 8 tetrahedral sites. Probably, this model should be used with a ds6 database, as Akhila did. Sapp(TP) - 2010, as Sa(KW), but, supposedly, should be used with kel04ver.dat. Sapp(HP) - 1998-2003 - between spr4 and spr7 - the narrowest Al-range. in contrast to the later models, this model has 8 octahedral sites and 6 tetrahedral sites. It was meant to be used with a ds5 database (<=2011), but since it's ideal it can be used with any database, except, perhaps, kel04ver.dat. Sapp(KWP) 2004 - between spr4 and spr5 - The comments in solution_model.dat imply this model, as in the later models, 4 octahedral sites and 10 tetrahedral sites. this model ostensibly should be used with kel04ver.dat and models of the HP generation. ---------------------------------------------------------- Model originally entered by Pulak Sengupta, 7/16/05. 1) Site populations corrected to correspond to those of Kelsey et al by P Goncalves, 10/12/2010. 1 2 3 M3 M46 T _________________ Mutliplicity 1 3 1 _________________ 1 spr4 Mg Mg Si Species: 2 fspr Fe Fe Si 3 spr5 Al Mg Al 4 fsp5_i Al Fe Al Dependent endmember: fsp5_i = spr5 + 3/4 * (fspr - spr4) ---------------------------------------------------------- Sapp(KWP) abbreviation Sap full_name sapphirine 7 model type 2 reciprocal solution 2 2 2 species on each site spr4 fspr spr5 fsp5_i 1 1 dependent endmember fsp5_i = 1 spr5 + 3/4 fspr - 3/4 spr4 0 0 0 0 | endmember flags 0 1 .1 0 | range and resolution for X(Mg), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution for 1-X(Tschermaks), imod = 0 -> cartesian subdivision begin_excess_function w(spr5 spr4) 10000 w(spr5 fspr) 12000 w(spr4 fspr) 8000 end_excess_function 3 | 3 site (M3, M46, T) configurational entropy model 2 3. | 2 species on M46, 3 sites per formula unit z(m46,mg) = 1 spr5 + 1 spr4 3 1 | 3 species on M3, 1 sites per formula unit. z(ma,Al) = 1 spr5 z(ma,fe) = 1 fspr 2 1 | 2 species on T, 1 sites per formula unit. z(T,si) = 1 spr4 + 1 fspr site_check_override end_of_model -------------------------------------------------------- begin_model | CHL(HP) - CHLORITE: extended from holland et al. 1998, EJM. | This model is a relict equipartition model and has been superceded | by the non-equipartition model Chl(W), which should be used in | preference to Chl(HP). JADC, 12/20. NOTES: * This is an oddball model because Al disorder is implicit The model should be reformulated so that clin is an explicit ordered species. The current formulaton precludes the anti-ordered state. * This model will only function for the FASH subsystem if MGO is also used as a component in VERTEX. * This model was tested with the maple script complete_chl.mws * For normal aluminous chlorites there is little to be gained by considering the afchl endmember becuase the endmember has negligible contribution to the total energy of the solution (see fig 4 of holland et al). Exclude this endmember to save computational resources. For Al-poor systems exclude ames and retain afchl. JADC 4/03 | Despite this models complexity it can be | written as a simple ternary with one independent | ordering parameter, however perplex does not | yet have speciation models implemented for | solutions in which multiple endmembers are | characterized by a single ordering parameter. | hence the model is formulated here explicitly | in terms of the dependent endmembers with the | site occupancy table: 1 2 3 4 M1 M2+M3 M4 T2 ________________________________ Mutliplicity 1 4 1 2 ________________________________ 1 mame Al Mn Al Al_Al dependent 2 mafchl Mn Mn Mn Si_Si dependent 3 mnchl Mn Mn Al Al_Si 4 fames Al Fe Al Al_Al dependent 5 fafchl Fe Fe Fe Si_Si dependent 6 daph Fe Fe Al Al_Si 7 ames Al Mg Al Al_Al 8 afchl Mg Mg Mg Si_Si 9 clin Mg Mg Al Al_Si Dependent endmembers: fame = ames + 4/5 * (daph - clin) fafchl = afchl + 6/5 * (daph - clin) mame = ames + 4/5 * (mnchl - clin) mafchl = afchl + 6/5 * (mnchl - clin) | For normal aluminous chlorites there is little to be gained | by considering the afchl and fafchl endmembers | becuase the endmember has negligible | contribution to the total energy of the solution | (see fig 4 of holland et al) Chl(HP) abbreviation Chl full_name chlorite 7 | model type reciprocal, macroscopic formulation 2 3 3 mame_i mafchl_i mnchl fame_i fafchl_i daph ames afchl clin 4 | 4 dependent endmembers fame_i = 1 ames + 4/5 daph - 4/5 clin fafchl_i = 1 afchl + 6/5 daph - 6/5 clin mame_i = 1 ames + 4/5 mnchl - 4/5 clin mafchl_i = 1 afchl + 6/5 mnchl - 6/5 clin 0 0 0 0 0 0 0 0 0 |endmember flags | subdivision model for (ternary) site 1 (T2): 0 1 .1 0 | range and resolution of X(Al_Si), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(Al_Al), imod = 0 -> cartesian subdivision | subdivision model for (ternary) site 2 (M2) 0 .2 .1 0 | range and resolution of X(Mn), imod = 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(Fe), imod = 0 -> cartesian subdivision begin_excess_function w(clin ames) 18000. w(clin afchl) 18000. w(ames afchl) 20000. w(clin daph) 2500 w(daph ames) 13500. w(daph afchl) 14500. end_excess_function 4 |4 site configurational entropy model: 4 1 |4 species on 1 M1 site z(al,M1) = 1 ames z(mn,M1) = 1 mnchl z(fe,M1) = 1 daph 3 4. |3 species on 4 M2+M3 sites z(mn,m2+m3)= 1 mnchl z(fe,m2+m3)= 1 daph 2 1 |3 species on 1 M4 site z(mg,m4) = 1 afchl 2 2. |2 species on 2 T2 sites z(al,T2)= 1 ames + 1/2 clin + 1/2 daph + 1/2 mnchl site_check_override | reject_bad_compositions | see the May 15, 2020 Perple_X update for explanation of the | reject_bad_compositions option. The option was turned off | by commenting Dec 2, 2020. JADC. end_of_model -------------------------------------------------------- begin_model | CHLORITE: extended from Holland et al. (1998) for sud substitution | Entered by Thomas Wagner, 5/12. LWV 7/12 | The reference for this model is Lanari, Wagner and Vidal, | CMP 2014 167:968 | the model is formulated here explicitly | in terms of the dependent endmembers with the | site occupancy table: 1 2 3 4 M1 M2+M3 M4 T2 ________________________________ Mutliplicity 1 4 1 2 ________________________________ 1 fames Al Fe Al Al_Al dependent 2 ames Al Mg Al Al_Al 3 clin Mg Mg Al Al_Si 4 daph Fe Fe Al Al_Si 5 fsud Va Al2_Fe2 Al Al_Si dependent 6 sud Va Al2_Mg2 Al Al_Si Dependent endmembers: fames = ames + 4/5 * (daph - clin) fsud = sud + 2/5 * (daph - clin) Chl(LWV) abbreviation Chl full_name chlorite 7 | model type reciprocal, macroscopic formulation 2 3 2 fames_i fsud_i daph ames sud_dqf clin 2 | 2 dependent endmembers fames_i = 1 ames + 4/5 daph - 4/5 clin fsud_i = 1 sud_dqf + 2/5 daph - 2/5 clin 0 0 0 0 0 0 |endmember flags | subdivision model for (ternary) site 1 (T2): 0 1 .1 0 | range and resolution of X(Al_Si), imod = 0 -> cartesian subdivision 0 1 .1 0 | range and resolution of X(Al_Al), imod = 0 -> cartesian subdivision | subdivision model for (ternary) site 2 (M2) 0 1 .1 0 | range and resolution of X(Fe), imod = 0 -> cartesian subdivision begin_excess_function w(clin ames) 18000. w(clin daph) 2500. w(clin sud_dqf) 49100. w(daph ames) 13500. w(daph sud_dqf) 43400. w(ames sud_dqf) 43300. end_excess_function 3 |3 site configurational entropy model: 4 1 |4 species on 1 M1 site z(al,M1) = 1 ames z(mg,M1) = 1 clin z(fe,M1) = 1 daph 3 4. |3 species on 4 M2+M3 sites z(fe,m2+m3) = 1 daph z(mg,m2+m3) = 1 clin + 1 ames + 1/2 sud_dqf 2 2. |2 species on 2 T2 sites z(al,T2) = 1 ames + 1/2 clin + 1/2 daph + 1/2 sud_dqf site_check_override end_of_model -------------------------------------------------------- begin_model Olivine - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus JADC, April 2019. M1 M2 _____________ Mutliplicity 1 1 _____________ fa Fe Fe fmont Fe Ca dependent fo Mg Mg mont Mg Ca _____________ mf Mg Fe ordered O(HGP) abbreviation Ol full_name olivine 8 | model type: o/d + reciprocal quadrilateral 2 | two simplicies 2x2 2 2 | endmembers on the prism vertices fmont mont fa fo 1 | ordered species: mf = 1/2 fo + 1/2 fa delta_g_of_ordering = 0 1 | dependent endmember fmont = 1 mont + 1 fa - 1 mf | see commentary in the header of this file for explanation of | endmember flags and subdivision schemes. 0 0 0 0 | endmember flags 0 1 .1 0 | subdivision scheme for Fe/(Fe+Mg) 0 .1 .1 0 | subdivision scheme for Ca/(Fe+Mg+Ca) begin_excess_function W(mont fa) 24d3 W(mont fo) 38d3 W(mont mf) 24d3 W(fa fo) 9d3 W(fa mf) 4.5d3 W(fo mf) 4.5d3 end_excess_function 2 | 2 site entropy model 2 1 | M1 2 species, site multiplicity = 1 z(M1,fe) = 1 fa 3 1 | M2 3 species, site multiplicity = 1 z(M2,mg) = 1 fo z(M2,ca) = 1 mont end_of_model -------------------------------------------------------- begin_model Orthopyroxene - Opx(HGP) - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus -------------------------------------------------------- W(en odi) and the odi DQF revised to destabilize Ca-rich Opx at high pressure. Eleanor Green. July 13, 2020. -------------------------------------------------------- Reformulated as a 2-polytope 688 format model. JADC, 9/19 The composition space of the published model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3, Cr, MTi on M1 C = Ca on M2 M = Mg, Fe A = Na on M2* *A is partially coupled to T by charge balance. The composition space of the model here has been simplified by eliminating the [FeTi][A,C] endmembers. The retained [FeTi][Fe] endmember is probably a waste of resources. NOTE: the definitions for the FeTi exchanges allowed in the full model are provided but have been commented out below. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = [T'][A,C] Polytope 2 = [T,M][M]+[M][C] where T' = Al, Fe3, MgTi on M1 -------------------------------------------------------- First coded for Perple_X by ECR Green and JRDC Aug 2018. Rearranged Apr 2019, JADC. NOTE: to use this the following endmembers must be specified with make definitions in the thermodynamic data file -------------------------------------------------------- M1 M2 T ____________________________________ Multiplicity 1 1 1/2 <- fake T multiplicity ____________________________________ Polytope 1: crjd_d Cr Na Si dependent crdi_d Cr Ca AlSi dependent ___ ness_d Fe3 Na Si dependent cess_d (acm) Fe3 Ca AlSi dependent ___ ojd_q Al Na Si ocats_d Al Ca AlSi dependent ___ mnbuf_d MgTi Na Si dependent mcbuf_d MgTi Ca AlSi dependent fnbuf_d FeTi Na Si dependent fcbuf_d FeTi Ca AlSi dependent ____________________________________ Polytope 2: cren_q Cr Mg AlSi crfs_d Cr Fe AlSi dependent _________ macm_q (mess) Fe3 Mg AlSi facm_d (fess) Fe3 Fe AlSi dependent _________ mgts Al Mg AlSi fts_d Al Fe AlSi dependent _________ obuf_q MgTi Mg AlSi ffbuf_d FeTi Fe AlSi dependent _________ odi_q Mg Ca Si hed_d Fe Ca Si dependent ___ en Mg Mg Si fs Fe Fe Si ___________________________________ Ordered species: fm Mg Fe Si independent femg-1(M2) = fm - en femg-1(M1) = fs - fm mgca-1(M2) = en - odi_q feca-1(M2) = fm - odi_q -------------------------------------------------------- Opx(HGP) abbreviation Opx full_name orthopyroxene 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and subdivision schemes for the composite composition space [T'][A,C] 0 .2 .1 0 | subdivision range for X(1) = X([T][A,C]) TM+MC by difference | ------------------------------------------- | Polytope 1 [T'][A,C] 2 | number of simplices 2 3 | change to 3 5 for full model crjd_d crdi_d ness_d cess_d |mnbuf_d mcbuf_d |fnbuf_d fcbuf_d ojd_q ocats_d | Simplex 1 X_K 0 1 .1 0 | X(1,1) = Na/[Na + Ca] on M2 X_Ca1 by difference | X(1,N) = Ca/[Na + Ca] on M2 by difference | Simplex 2 X_CrTs 0 1 .1 0 | X(2,1) = Cr/T on M1 X_FeTs 0 1 .1 0 | X(2,2) = Fe3+/T on M1 |0 1 .1 0 | X(2,3) = MgTi/T on M1 |0 1 .1 0 |X(2,4) = FeTi/T on M1 X_AlTs by difference | X(2,N) = Al/T on M1 by difference | ------------------------------------------- | Polytope 2 TM+MC 2 | number of simplices 2 6 | number of vertices on each simplex ffbuf_d obuf_q crfs_d cren_q facm_d macm_q fts_d mgts hed_d odi_q fs en | Simplex 1 X_Fe 0 1 .1 0 | X(1,1) => Fe/M X_Mg by difference | X(1,N) = Mg/M by difference | Simplex 2 X_MTiTs 0 .2 .1 0 | X(2,1) - M-buf(MgTi) X_MCrTs 0 .1 .1 0 | X(2,2) - M-Cr X_MFeTs 0 .1 .1 0 | X(2,3) - M-Fe3+ X_MAlTs 0 .2 .1 0 | X(2,4) - M-Ts X_MCPx 0 1 .1 0 | X(2,5) - M-Ca X_MMPx by difference begin_ordered_endmembers fm = 1/2 en + 1/2 fs delta_g_of_ordering = -6600 end_ordered_endmembers begin_dependent_endmembers fts_d = 1 mgts - 1 en + 1 fm ocats_d = 1 mgts - 1 en + 1 odi_q crjd_d = 1 cren_q - 1 mgts + 1 ojd_q crfs_d = 1 cren_q - 1 en + 1 fm crdi_d = 1 cren_q - 1 en + 1 odi_q facm_d = 1 macm_q - 1 en + 1 fm cess_d = 1 macm_q - 1 en + 1 odi_q ness_d = 1 macm_q - 1 mgts + 1 ojd_q |mcbuf_d = 1 obuf_q + 1 odi_q - 1 en |fcbuf_d = 1 obuf_q + 1/2 fs - 1/2 fm + 1 odi_q - 1 en ffbuf_d = 1 obuf_q - 1 en + 1/2 fs + 1/2 fm hed_d = -1 fm + 1 fs + 1 odi_q end_dependent_endmembers begin_excess_function W(en fs) 7d3 W(en fm) 4d3 W(en mgts) 12.5d3 - 0.04 * p | W(en odi_q ) revised from 32.2d3 + 0.12 * p | by ECRG, 13/7/20 W(en odi_q ) 29.4d3 | changed from W(en odi_q ) 29d3 + 0.15 * p by DB to match the ECRG 10/20 revision. W(en cren_q) 8d3 W(en obuf_q) 6d3 | added by DB to match the ECRG 10/20 revision. W(en macm_q) 8d3 W(en ojd_q ) 35d3 W(fs fm) 4d3 W(fs odi_q ) 21.5d3 + 0.08 * p |changed from W(fs odi_q ) 25.54d3 + 0.084 * p by DB to match the ECRG 10/20 revision. W(fs mgts) 11d3 - 0.15 * p W(fs cren_q) 10d3 W(fs obuf_q) 7d3 | added by DB to match the ECRG 10/20 revision. W(fs macm_q) 10d3 W(fs ojd_q ) 35d3 W(fm odi_q ) 18d3 + 0.08 * p |changed from W(fm odi_q ) 22.54d3 + 0.084 * p by DB to match the ECRG 10/20 revision. W(fm mgts) 15d3 - 0.15 * p W(fm cren_q) 12d3 W(fm obuf_q) 8d3 | added by DB to match the ECRG 10/20 revision. W(fm macm_q) 12d3 W(fm ojd_q ) 35d3 W(odi_q mgts) 75.5d3 - 0.84 * p W(odi_q cren_q) 20d3 W(odi_q obuf_q) 40d3 | added by DB to match the ECRG 10/20 revision. W(odi_q macm_q) 20d3 W(odi_q ojd_q ) 35d3 W(mgts cren_q) 2d3 W(mgts obuf_q) 10d3 | added by DB to match the ECRG 10/20 revision. W(mgts macm_q) 2d3 W(mgts ojd_q ) 7d3 W(cren_q obuf_q) 6d3 | added by DB to match the ECRG 10/20 revision. W(cren_q macm_q) 2d3 W(cren_q ojd_q ) -11d3 W(obuf_q macm_q) 6d3 | added by DB to match the ECRG 10/20 revision. W(obuf_q ojd_q ) 20d3 | changed from W(obuf_q ojd_q ) 4d3 by DB to match the ECRG 10/20 revision. W(macm_q ojd_q ) -11d3 end_excess_function 3 | number of sites in entropy model (M1, M2, T) M1 | site name 6 1 1 | number of species, effective multiplicity, true multiplicity z(fe) = 1 fs z(al) = 1 mgts + 1 ojd_q z(fe3+) = 1 macm_q z(cr) = 1 cren_q z(ti) = 1/2 obuf_q z(Mg) = 1/2 obuf_q + 1 odi_q + 1 en + 1 fm M2 | site name 4 1 1 | number of species, effective multiplicity, true multiplicity z(fe) = 1 fs + 1 fm z(ca) = 1 odi_q z(na) = 1 ojd_q z(Mg) = 1 obuf_q + 1 en + 1 cren_q + 1 macm_q + 1 mgts T | site name 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 mgts + 1/2 cren_q + 1/2 obuf_q + 1/2 macm_q z(Si,T) = 1/2 mgts + 1/2 cren_q + 1/2 obuf_q + 1/2 macm_q + 1 ojd_q + 1 odi_q + 1 en + 1 fs + 1 fm [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(en) 1 alpha(fs) 1 alpha(fm) 1 alpha(odi_q ) 1.2 alpha(mgts) 1 alpha(cren_q) 1 alpha(obuf_q) 1 alpha(macm_q) 1 alpha(ojd_q ) 1.2 end_van_laar_sizes begin_dqf_corrections dqf(cren_q) -25900 + 15.5 * T + 0.05 * P | crenh dqf(ojd_q) 18800 | dqf(odi_q) -100 + 0.211 * T + 0.005 * P | = odi prior to 13/7/20: | dqf(odi_q) 1900 + 0.005 * P | = odi update by ECRG 13/7/20 dqf(odi_q) = 2800 + 0.005 * P | DB/ECRG update 10/20 dqf(obuf_q) = -5000 - 5.1 * T - 0.0061 * PP | DB/ECRG update 10/20 dqf(macm_q) = 4800 - 0.089 * P end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Spinel - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus DO NOT USE THIS MODEL WITH ds6.22-. USE Sp(WPC) FOR ds6.22-. Initially coded for Perple_X by JRDC Oct 2018; reworked JADC Apr 2019. NOTE: to use this model: 1) the picr endmember here is assumed to be ordered as is the case in ds6.33, but was NOT the case in earlier versions of ds6, therefore if this model is to be used with earlier versions of ds6, the o/d model must also be stripped from picr. 3) the qnd endmember DQF is added in the dqf_correction section at the end of this model. 4) if both TiO2 nor O2 are absent from the Sp(HGP), Perple_X reformulate this model as the sp-picr binary, rather than the sp-herc-picr-fpcr quadrilateral. To prevent this specify TiO2 or O2 as components present with zero amount. JADC, 12/20 M T _________ Multiplicity 1 2 -> fake T-site multiplicity of 1 _________ isp Al MgAl imgmt Fe3 MgFe3 dependent picr Mg Cr qnd Mg MgTi iherc Al FeAl imt Fe3 FeFe3 fpcr Fe 2Cr dependent fqnd Fe FeTi dependent Ordered species _________ nsp Mg Al nherc Fe Al nmt Fe Fe3+ Sp(HGP) abbreviation Sp full_name spinel 688 | model type: prism , order-disorder 1 | number of prisms 2 | two simplicies 2 4 | number of cations on independent sites forming simplices: | fold prism (Al - Fe3+ - Cr - Ti) + binary simplices (Mg Fe on T) | endmembers on the prism vertices picr fpcr qnd fqnd imgmt mt sp herc X_Fe2+ 0 1 .1 0 | subdivision scheme for Fe2/(Fe2+Mg) X_Mg by difference Z_Cr 0 1 .1 0 | subdivision scheme for X(Cr) Z_MTi 0 1 .1 0 | subdivision scheme for X(Ti) Z_Fe3+ 0 1 .1 0 | subdivision scheme for X(Fe3+) Z_MAl by difference begin_ordered_endmembers | dqf(nmt) - dqf(imt) = nmt = 1 mt delta_g_of_ordering = -0.3d3 + 5.76303 * T |changed from 5.76303 * T by DB to match the ECRG 10/20 revision. | 0 - dqf(isp) = nsp = 1 sp delta_g_of_ordering = -23.6d3 + 5.76303 * T | 0 - dqf(iherc) = nherc = 1 herc delta_g_of_ordering = -23.6d3 + 5.76303 * T end_ordered_endmembers begin_dependent_endmembers imgmt = 1 sp - 1 herc + 1 mt fpcr = -1 nsp + 1 nherc + 1 picr fqnd = -1 nsp - 1 sp + 1 nherc + 1 herc + 1 qnd end_dependent_endmembers begin_excess_function W(nsp sp) -8.2d3 W(nsp nherc) 3.5d3 W(nsp herc) -13d3 W(nsp nmt) 43.2d3 W(nsp mt) 49.1d3 W(nsp picr) -5d3 W(nsp qnd) 22.5d3 W(sp nherc) 4.4d3 W(sp herc) -6d3 W(sp nmt) 36.8d3 W(sp mt) 20d3 W(sp picr) 14d3 W(sp qnd) 21.5d3 W(nherc herc) -8.2d3 W(nherc nmt) 18.1d3 W(nherc mt) 49d3 W(nherc picr) -19d3 W(nherc qnd) 35.1d3 W(herc nmt) -4d3 W(herc mt) 7.6d3 W(herc picr) -11d3 W(herc qnd) 9d3 W(nmt mt) 18.1d3 W(nmt picr) 11.9d3 W(nmt qnd) 62.2d3 W(mt picr) -6.4d3 W(mt qnd) 24.3d3 W(picr qnd) 60d3 end_excess_function 2 | 2 site entropy model (M , T) M 4 1 1 | 4 species on M, mult. = 1 z(Fe,M) = 1 nherc + 1 nmt z(Fe3,M) = 1 mt z(Al,M) = 1 sp + 1 herc z(Mg,M) = 1 nsp + 1 picr + 1 qnd T 6 1 2 | 6 species on T, effective mult. = 1, true mult = 2 z(Mg,T) = 1/2 sp + 1/2 qnd z(Fe,T) = 1/2 herc + 1/2 mt z(Ti,T) = 1/2 qnd z(Fe3,T) = 1 nmt + 1/2 mt z(Cr,T) = 1 picr z(Al,T) = 1 nsp + 1/2 sp + 1 nherc + 1/2 herc [O4] | formula suffix, enter "none" for no suffix. begin_dqf_corrections dqf(qnd) = -30000 end_dqf_corrections begin_special_operations spop(sp) inverse dqf(J/mol) = 23600 - 5.763 * T_K spop(herc) inverse dqf(J/mol) = 23600 - 5.763 * T_K spop(mt) eq dqf(J/mol) = 300 end_special_operations refine_endmembers | solvus_tolerance 0.02 end_of_model -------------------------------------------------------- begin_model Garnet - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus JADC, April 2019 X Y _____________ Mutliplicity 3 2 _____________ Dependent: fkho_d Fe Fe3+ Dependent: kho_d Mg Fe3+ andr Ca Fe3+ Dependent: fkno_d Fe Cr knor Mg Cr Dependent: ckno_d Ca Cr Dependent: ftig_d Fe Al(MgTi)_0.5 tig_q Mg Al(MgTi)_0.5 Dependent: ctig_d Ca Al(MgTi)_0.5 alm Fe Al py Mg Al gr Ca Al Gt(HGP) abbreviation Gt full_name garnet 7 2 3 4 ftig_d tig_q ctig_d fkno_d knor ckno_d fkho_d kho_d andr alm py gr 6 fkno_d = 1 alm + 1 knor - 1 py ckno_d = 1 gr + 1 knor - 1 py fkho_d = 1 alm + 1 andr - 1 gr kho_d = 1 py + 1 andr - 1 gr ftig_d = 1 alm + 1 tig_q - 1 py ctig_d = 1 gr + 1 tig_q - 1 py | see commentary in the header of this file for explanation of | endmember flags and subdivision schemes. 0 0 0 0 0 0 0 0 0 0 0 0 | endmember flags 0 1 .1 0 | subdivision scheme for Fe on A 0 1 .1 0 | subdivision scheme for Mg on A 0 1 .1 0 | subdivision scheme for Al(MgTi)_0.5 on B 0 1 .1 0 | subdivision scheme for Cr on B 0 1 .1 0 | subdivision scheme for Fe3+ on B begin_excess_function W(py alm) 4d3 + .1 * P W(py gr) 45.4d3 - 10 * T + 0.04 * P W(py andr) 107d3 - 10 * T - 0.036 * P W(py knor) 2d3 W(alm gr) 17d3 - 10 * T + .1 * P W(alm andr) 65d3 - 10 * T + 0.039 * P W(alm knor) 6d3 + 0.01 * P W(gr andr) 2d3 W(gr knor) 1d3 - 10 * T + 0.180 * P W(andr knor) 63d3 - 10 * T + 0.10 * P end_excess_function 2 3 3. z(m1,fe) = 1 alm z(m1,mg) = 1 py + 1 knor + 1 tig_q 5 2. z(m2,cr) = 1 knor z(m2,fe3) = 1 andr z(m2,ti) = 1/4 tig_q z(m2,mg) = 1/4 tig_q begin_van_laar_sizes alpha(py) 1 alpha(alm) 1 alpha(gr) 2.5 alpha(andr) 2.5 alpha(knor) 1 alpha(tig_q) 1 end_van_laar_sizes begin_dqf_corrections dqf(knor) = 18200 dqf(tig_q) = 46.7d3 - 17.3 * T | tig HGP end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Holland and Powell (2001) J.Petrol 426, 673-683 (KNCASH) + White et al (2001) JMG 19: 139-153 (FM+KNCASH). JADC 3/03 * Changed DQF terms to account for modifications of faL and foL (White et al 2007). Mark Caddick, Nov, 07. * Zr8L added after Kelsey & Powell (2010). Jeff Marsh, Jan 10, 2012. * reformulated as ordinary configurational entropy model. JADC 12/18. * Zn - Wi8L (= 2 WiL [ZnSi2O4]) endmember estimated using HSC caloric data and foL volumetric properties, Wi8L interaction terms replicate foL_q. Zn is assumed to mix on the Temkin "Olivine" site. JADC, May 26, 2019. WARNING 2: the endmembers Wi8L, qL_q, faL_q, foL_q, and siL_q must be "made" in the thermodynamic data file. WARNING 3: the (stabilizing) dqf corrections made to the foL_q, faL_q, and siL_q enedmembers make the haplogranite melt model inapplicable to melts where these endmembers are present in high concentrations, to model such situations or to reproduce the published fo-fa-q or sil-q melting phase relations, the dqf corrections (below) should be set to zero. WARNING 4: different versions of the HP data base may result in significant variations in the predicted position of the wet-solidus (Powell, pers comm.). WARNING 5: The melt model incorrectly predicts a high pressure-low temperature stability field for water-silica rich melts. at 10 kb this field extends to ca 750 K and at 3 kb to ca 550 K. To eliminate this artifact set T_melt melt(HP) abbreviation Melt full_name liquid 2 model type: simplicial composition space 10 number of endmembers foL_q faL_q zr8L abL siL_q anL kspL wiL_q qL_q wL_q 0 0 0 0 0 0 0 0 0 0 | endmember flags. | RESTRICTED(!) subdivision ranges and model | the ranges must be extended if calculations | are to be done in subsystems like fo-fa-h2o, | Ab-H2O etc etc; these numbers seem about right | for pelite melting. 0 1 .1 0 | range and resolution of X(fo), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(fa), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(zrL), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(ab), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(sil), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(an), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(ksp), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(wiL), 1 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(q), 0 => cartesian subdivision begin_excess_function | the excess parameters used for the haplo- | granite melt model vary between papers (notably w(an-sil)), | the values here are from the thermocalc | model file 'thdmloss.txt' and appear largely | consistent with the white et al paper. w(qL_q anL) -10d3 w(qL_q siL_q) 12d3 w(qL_q abL) 12d3 -0.4 * P_bar w(qL_q kspL) -2d3 -0.5 * P_bar w(qL_q wL_q) 15d3 w(qL_q foL_q) 12d3 -0.4 * P_bar w(qL_q faL_q) 14d3 w(abL siL_q) 12d3 w(abL kspL) -6d3 3. * P_bar w(abL foL_q) 10d3 w(abL faL_q) 2d3 w(abL wL_q) 1d3 -0.2 * P_bar w(kspL anL) 0 -1. * P_bar w(kspL siL_q) 12d3 w(kspL wL_q) 11d3 -0.45 * P_bar w(kspL foL_q) 12d3 w(kspL faL_q) 12d3 | w(anL sil8) is 12 kJ in the Holland & Powell J Pet paper | but appears to have accidentally been set to zero | for the calculations published in the White et al. JMG | paper. To restore this parameter delete the comment | marker "|" on the following line: | w(anL sil8) 12d3 w(anL wL_q) 9d3 -0.85 * P_bar w(siL_q foL_q) 12d3 w(siL_q faL_q) 12d3 w(siL_q wL_q) 16d3 w(foL_q faL_q) 18d3 w(foL_q wL_q) 11d3 -0.5 * P_bar w(faL_q wL_q) 12d3 w(qL_q wiL_q) 14d3 w(abL wiL_q) 10d3 w(kspL wiL_q) 12d3 w(wiL_q foL_q) 18d3 w(wiL_q faL_q) 18d3 w(wiL_q wL_q) 11d3 -0.5 * P_bar w(qL_q zr8L) 14d3 w(abL zr8L) 12d3 w(kspL zr8L) 12d3 w(zr8L foL_q) 12d3 w(zr8L faL_q) 12d3 w(zr8L wL_q) 25d3 end_excess_function 3 | Configurational entropy: two non-temkin sites (Water, Melt) | and one temkin site (olvine). hp assume a fsp = ab + or "molecule"" | with mixing on a temkin M site, but the math works out the same as | the endmembers are treated as separate endmembers and the M site | dropped. 2 1 | water-vacancy site z(H) = 1 wL_q 3 0 | temkin olivine site n(Mg) = 4 foL_q n(Fe) = 4 faL_q n(Zn) = 4 wiL_q 8 0 | silicate species site z(w) = 1 wL_q z(q) = 1 qL_q z(ksp) = 1 kspL z(ab) = 1 abL z(sil) = 1 siL_q z(an) = 1 anL z(zr) = 1 zr8L z(ol) = 1 foL_q + 1 faL_q + 1 wiL_q begin_dqf_corrections dqf(wL_q) = 0 dqf(wiL_q) = 0 dqf(qL_q) = 0 dqf(foL_q) = -15d3 dqf(faL_q) = -15d3 dqf(siL_q) = -10d3 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model HP '98 quaternary garnet model Zr-endmember (zrg, Ca3Al2[Si2Zr]O12) after Kelsey & Powell '10 - Jeff Marsh, Jan 10, 2012. Zn-endmember (zngt,Zn3Al3Si3O12) estimated by octahedral metasilicate (HSC) Zn-Mn exchange from spss, Zn excess terms replicate Mn. - JADC, 5/26/2019. ZrGt(KP) abbreviation Gt full_name garnet 2 model type: simplicial composition space 6 number of endmembers zrg zngt spss py gr alm endmember names 1 1 1 0 0 0 | endmember flags 1e-5 1e-4 .1 1 | imod = 1 -> non-linear subdivision 1d-4 0.2 .1 1 | zngt 0 .2 .1 0 | 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision | NOTE restricted subdivision range on Mn (Species 1)! begin_excess_function w(py gr) 33000 w(alm py) 2500 | hp '98 give 2.4 kJ w(py spss) 4500 w(alm spss) 240 w(py zngt) 4500 w(alm zngt) 240 end_excess_function 2 | 1 site entropy model 5 3 | 5 species, site multiplicity 3 z(Fe) = 1 alm z(Mg) = 1 py z(Zn) = 1 zngt z(Zr) = 1 zrg 2 3 | 2 species, site multiplicity 3 | was 2 2, zr = 1 zrg, Zr goes on the | the Si site as Si2Zr | corrected(?) May 5, 2019, JADC. z(Zr) = 1/3 zrg end_of_model -------------------------------------------------------- begin_model Ideal hydrous wadleysite, extracted from Komabayashi & Omori 2006 who indicate the H4SiO4 substituion mechanism, the h2o_wd endmember estimated as in komabayashi_water_endmembers.mws JADC, July 2019. M _____________ Mutliplicity 2 _____________ wd Mg h2o_wd H2 hWd abbreviation Wd full_name wadleysite 2 | model type: simplicial 2 | endmembers h2o_wd wd | see commentary in the header of this file for explanation of | endmember flags and subdivision schemes. 0 0 | endmember flags 0 1 .1 0 | H2O/(H2O+MgO) ideal 1 | 1 site entropy model 2 2 | M 2 species, site multiplicity = 1 z(M,mg) = 1 wd end_of_model -------------------------------------------------------- begin_model Ideal hydrous ringwoodite, extracted from Komabayashi & Omori 2006 who infer that both the H4SiO4 and MgH4O4 substituion mechanisms, but summarize these in a single hydrous species (Mg(1.91)Si(0.99)O(3.78)[OH](0.22). This choice obviates the extraction of endmembers for both exchanges from their data. Here the two exchanges are coupled by the the partially disordered endmember h2o_rg H(4)Si(0.7777.)H(0.8888.)O(4) estimated as in komabayashi_water_endmembers.mws JADC, July 2019. M T _____________ Mutliplicity 2 1 _____________ wd Mg Si h2o_rg H2 H4 hRg abbreviation Rg full_name ringwoodite 2 | model type: simplicial 2 | endmembers h2o_rg rg | see commentary in the header of this file for explanation of | endmember flags and subdivision schemes. 0 0 | endmember flags 0 1 .1 0 | glug ideal 2 | 2 site entropy model 2 2 | M, 2 species, site multiplicity = 2 z(M,mg) = 1 rg 2 1 | T, 2 species, site multiplicity = 1 z(T,H4) = 2/9 h2o_rg end_of_model -------------------------------------------------------- begin_model Peridotite melting - Tomlinson and Holland; Journal of Petrology, 2021, 10.1093/petrology/egab012 Entered by Debaditya Bandyopadhyay, Sept 2022. modified JADC, Oct 2022 This model requires the following make definitions in the thermodynamic data file make_definitions section: q3L_q = 3 qL foL_q = 2 foL faL_q = 2 faL neL_q = 1 abL - 2 qL hmL_q = 1/2 hemL ekL_q = 1/2 eskL tiL_q = 1 ruL kjL_q = 1 kspL - 1 qL to avoid cluttering thermodynamic data files with the new plague of TC DQF "corrections", the following simple DQF's are specified in the make_definition section at the end of this model: dqf(silL_q) = 6.35d3 -0.320*Pbar updated to DQF 7.86d3 - 0.337*Pbar dqf(woL_q) = -0.18d3 -0.118*Pbar updated to DQF -0.09d3 - 0.102*Pbar melt(TH) abbreviation Melt full_name liquid 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 12 | number of vertices (endmembers) on each simplex foL_q faL_q abL kspL neL_q silL_q kjL_q woL_q ekL_q hmL_q tiL_q q3L_q 0 1 .1 0 | range and resolution of X(fo) 0 1 .1 0 | range and resolution of X(fa) 0 1 .1 0 | range and resolution of X(ab) 0 1 .1 0 | range and resolution of X(ksp) 0 1 .1 0 | range and resolution of X(ne) 0 1 .1 0 | range and resolution of X(sil) 0 1 .1 0 | range and resolution of X(kj) 0 1 .1 0 | range and resolution of X(wo) 0 1 .1 0 | range and resolution of X(ek) 0 1 .1 0 | range and resolution of X(hm) 0 1 .1 0 | range and resolution of X(ti) begin_ordered_endmembers | H(anTHL) = -102.8d3 + 55*TK + 0.031 *Pbar - dqfwoL_q - dqfsilL_q | dqf(silL_q) = 7.86d3 - 0.337*Pbar | dqf(woL_q) = -0.09d3 - 0.102*Pbar anTHL = 1 woL_q + 1 silL_q Delta(enthalpy) = -110.57d3 + 55*TK + 0.47 *Pbar | H(enTHL) = -19.80d3 - 0.371*Pbar - 1/2 dqffoL_q - 1/3 dqfq3L_q | dqf(q3L_q) = 0.47d3 - 0.078*Pbar | dqf(foL_q) = 9.73d3 - 0.176*Pbar enTHL = 1/2 foL_q + 1/3 q3L_q Delta(enthalpy) = -24822 - 0.257*Pbar end_ordered_endmembers begin_excess_function W(q3L_q silL_q) 10.7d3 -0.1 * Pbar W(q3L_q woL_q) 14.7d3 W(q3L_q foL_q) 45.6d3 -0.54 * Pbar W(q3L_q faL_q) 4.9d3 -0.56 * Pbar W(q3L_q neL_q) 12.4d3 W(q3L_q hmL_q) 20d3 W(q3L_q ekL_q) -4d3 -0.02 * Pbar W(q3L_q tiL_q) 31.6d3 W(q3L_q kjL_q) 2.3d3 +0.07 * Pbar W(q3L_q anTHL) -14.1d3 +0.02 * Pbar W(q3L_q abL) -0.1d3 W(q3L_q enTHL) 33.3d3 -0.41 * Pbar W(q3L_q kspL) -3.4d3 -0.44 * Pbar W(silL_q woL_q) -24.1d3 +0.84 * Pbar W(silL_q foL_q) 6.2d3 -0.15 * Pbar W(silL_q faL_q) 2.7d3 W(silL_q neL_q) 16.8d3 W(silL_q hmL_q) -5d3 W(silL_q ekL_q) -3d3 W(silL_q tiL_q) 14.2d3 -0.03 * Pbar W(silL_q kjL_q) 6.9d3 +0.02 * Pbar W(silL_q anTHL) 4.6d3 W(silL_q enTHL) 2.5d3 +0.01 * Pbar W(silL_q kspL) 4d3 W(woL_q foL_q) 33d3 +0.04 * Pbar W(woL_q faL_q) 26d3 W(woL_q neL_q) -2.1d3 -0.02 * Pbar W(woL_q ekL_q) -10d3 W(woL_q tiL_q) 7.8d3 +0.03 * Pbar W(woL_q kjL_q) -1.1d3 +0.03 * Pbar W(woL_q anTHL) 9.8d3 W(woL_q abL) 4.1d3 W(woL_q enTHL) 13.3d3 +0.09 * Pbar W(woL_q kspL) 9.8d3 W(foL_q faL_q) 17.7d3 -0.21 * Pbar W(foL_q neL_q) 1.2d3 +0.02 * Pbar W(foL_q ekL_q) -3d3 W(foL_q tiL_q) 2.2d3 -0.16 * Pbar W(foL_q kjL_q) 2.3d3 W(foL_q anTHL) -5.2d3 -0.02 * Pbar W(foL_q abL) 1.9d3 W(foL_q enTHL) 1.4d3 +0.36 * Pbar W(foL_q kspL) -5.6d3 W(faL_q neL_q) 7.7d3 -0.05 * Pbar W(faL_q hmL_q) -30d3 W(faL_q tiL_q) -9.5d3 +0.02 * Pbar W(faL_q kjL_q) 8.9d3 W(faL_q anTHL) -6.5d3 W(faL_q abL) 0.6d3 W(faL_q enTHL) 2.9d3 W(faL_q kspL) -6.5d3 W(neL_q hmL_q) 10d3 W(neL_q tiL_q) 10.4d3 +0.14 * Pbar W(neL_q kjL_q) -5.2d3 W(neL_q anTHL) 7.6d3 W(neL_q abL) 0.5d3 +0.12 * Pbar W(neL_q enTHL) 2.6d3 W(neL_q kspL) -2.2d3 +0.09 * Pbar W(hmL_q tiL_q) -2.8d3 W(hmL_q kjL_q) 10d3 W(ekL_q tiL_q) -2.1d3 W(tiL_q kjL_q) 3d3 W(tiL_q anTHL) -4.8d3 W(tiL_q abL) -2.3d3 W(tiL_q enTHL) 17.2d3 W(tiL_q kspL) -8.1d3 W(kjL_q anTHL) -5.6d3 W(kjL_q abL) -2.9d3 W(kjL_q enTHL) -3d3 W(kjL_q kspL) 15.3d3 +0.32 * Pbar W(anTHL abL) 0.2d3 W(abL kspL) 10d3 end_excess_function 2 M 4 0 0 | Temkin M-site n(Mg) = 4 foL_q n(Fe) = 4 faL_q n(Ca) = 1 woL_q n(Al) = 1 silL_q F 13 0 0 | Temkin F-site n(ne) = 1 neL_q n(kj) = 1 kjL_q n(ksp) = 1 kspL n(ab) = 1 abL n(an) = 1 anTHL n(en) = 1 enTHL n(alsi) = 1 silL_q n(si0) = 1 woL_q n(ol) = 1 faL_q + 1 foL_q n(q) = 1 q3L_q n(ek) = 1 ekL_q n(ti) = 1 tiL_q n(hm) = 1 hmL_q none | formula suffix, enter "none" for no suffix. begin_dqf_corrections begin_dqf_corrections dqf(foL_q) = 9.73d3 - 0.176*Pbar dqf(faL_q) = 16.52d3 - 0.034*Pbar dqf(hmL_q) = 3.76d3 - 0.027*Pbar dqf(kjL_q) = 22.93d3 + 0.292*Pbar dqf(ekL_q) = 24.98d3 + 0.170*Pbar dqf(tiL_q) = 0.42d3 - 0.257*Pbar dqf(q3L_q) = 0.47d3 - 0.078*Pbar dqf(neL_q) = 33.11d3 + 0.306*Pbar dqf(silL_q) = 7.86d3 - 0.337 *Pbar dqf(woL_q) = -0.09d3 - 0.102 *Pbar dqf(abL) = 5.02d3 - 0.082 *Pbar dqf(kspL) = 5.22d3 - 0.086 *Pbar end_dqf_corrections begin_van_laar_sizes alpha(q3L_q) = 1 alpha(silL_q) = 1.45 alpha(woL_q) = 1.45 alpha(foL_q) = 2 alpha(faL_q) = 1 alpha(neL_q) = 1 alpha(hmL_q) = 1 alpha(ekL_q) = 1 alpha(tiL_q) = 1 alpha(kjL_q) = 1 alpha(anTHL) = 1 alpha(abL) = 1 alpha(kspL) = 1 alpha(enTHL) = 1 end_van_laar_sizes begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. This | may be useful for testing purposes for systems in which the | the endmember compositions are not expected to be stabLe, i.e., | the stability of pure wL_q would alert the user to unexpected | behaviour. if/when endmember compositions of the solution are | expected, those endmembers should be removed from the list of | flagged endmembers. foL_q faL_q neL_q silL_q kjL_q woL_q ekL_q hmL_q tiL_q q3L_q abL kspL anTHL enTHL end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Spinel - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus as revised by Tomlinson and Holland; Journal of Petrology, 2021, 10.1093/petrology/egab012 Initially coded for Perple_X by JRDC Oct 2018; reworked JADC Apr 2019. Modified parameters entered by Debaditya Bandyopadhyay, Sept 2022. NOTE: to use this model: 1) the endmembers nsp, nherc must be created in the thermodynamic data file from sp and herc by stripping off the bragg o/d model. the picr endmember here is assumed to be ordered as is the case in ds6.33, but was NOT the case in earlier versions of ds6, therefore if this model is to be used with earlier versions of ds6, the o/d model must also be stripped from picr. 2) the following endmembers must be specified with make definitions in the thermodynamic data file isp = 1 nsp DQF(J/mol) = 23.6d3 - 5.76303 * T iherc = 1 nherc DQF(J/mol) = 23.6d3 - 5.76303 * T |HGP values imt = 1 mt DQF(J/mol) = 300 |HGP values 3) if both TiO2 nor O2 are absent from Sp(HGP), Perple_X reformulate this model as the sp-picr binary, rather than the sp-herc-picr-fpcr quadrilateral. To prevent this specify TiO2 or O2 as components present with zero amount. JADC, 12/20 M T _________ Multiplicity 1 2 -> fake T-site multiplicity of 1 _________ isp Al MgAl imgmt Fe3 MgFe3 dependent picr Mg Cr qndTH Mg MgTi dependent iherc Al FeAl imt Fe3 FeFe3 fpcr Fe 2Cr dependent usp Fe FeTi Ordered species _________ nsp Mg Al nherc Fe Al nmt Fe Fe3+ Sp(TH) abbreviation Sp full_name spinel 688 | model type: prism , order-disorder 1 | number of prisms 2 | two simplicies 2 4 | number of cations on independent sites forming simplices: | fold prism (Al - Fe3+ - Cr - Ti) + binary simplices (Mg Fe on T) | endmembers on the prism vertices picr fpcr qndTH usp imgmt imt isp iherc X_Fe2+ 0 1 .1 0 | subdivision scheme for Fe2/(Fe2+Mg) X_Mg by difference Z_Cr 0 1 .1 0 | subdivision scheme for X(Cr) Z_MTi 0 1 .1 0 | subdivision scheme for X(Ti) Z_Fe3+ 0 1 .1 0 | subdivision scheme for X(Fe3+) Z_MAl by difference begin_ordered_endmembers nmt = 1 imt delta_g_of_ordering = -0.3d3 + 5.76303 * T |changed from 5.76303 * T by DB to match the ECRG 10/20 revision. nsp = 1 isp delta_g_of_ordering = -23.5d3 + 5.76303 * T nherc = 1 iherc delta_g_of_ordering = -23.6d3 + 5.76303 * T end_ordered_endmembers begin_dependent_endmembers imgmt = 1 isp - 1 iherc + 1 imt fpcr = -1 nsp + 1 nherc + 1 picr qndTH = 1 nsp + 1 isp - 1 nherc - 1 iherc + 1 usp end_dependent_endmembers begin_excess_function W(nsp isp) -6.7d3 W(nsp nherc) 3.6d3 W(nsp iherc) -9.8d3 W(nsp nmt) 43.2d3 W(nsp imt) 49.5d3 W(nsp picr) -38.4d3 - 0.08*Pbar W(nsp usp) 40d3 W(isp nherc) 2.7d3 W(isp iherc) -3.5d3 W(isp nmt) 36.8d3 W(isp imt) 20.7d3 W(isp picr) -21.6d3 - 0.08*Pbar W(isp usp) 38.2d3 W(nherc iherc) -6.0d3 W(nherc nmt) 17.5d3 W(nherc imt) 51.6d3 W(nherc picr) -53.8d3 W(nherc usp) 25.7d3 W(iherc nmt) -4.1d3 W(iherc imt) 10d3 W(iherc picr) -38.8d3 W(iherc usp) 21d3 W(nmt imt) 18.1d3 W(nmt picr) 12.1d3 W(nmt usp) 5.2d3 W(imt picr) -8.7d3 W(imt usp) 21.5d3 W(picr usp) 15d3 end_excess_function 2 | 2 site entropy model (M , T) M 4 1 1 | 4 species on M, mult. = 1 z(Fe,M) = 1 nherc + 1 nmt + 1 usp z(Fe3,M) = 1 imt z(Al,M) = 1 isp + 1 iherc z(Mg,M) = 1 nsp + 1 picr T 6 1 2 | 6 species on T, effective mult. = 1, true mult = 2 z(Mg,T) = 1/2 isp z(Fe,T) = 1/2 iherc + 1/2 imt + 1/2 usp z(Ti,T) = 1/2 usp z(Fe3,T) = 1 nmt + 1/2 imt z(Cr,T) = 1 picr z(Al,T) = 1 nsp + 1/2 isp + 1 nherc + 1/2 iherc [O4] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(nsp) = 1 alpha(isp) = 1 alpha(nherc) = 1 alpha(iherc) = 1 alpha(nmt) = 1 alpha(imt) = 1 alpha(picr) = 1 alpha(usp) = 0.9 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model Garnet - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus Revised by Tomlinson and Holland; Journal of Petrology, 2021, 10.1093/petrology/egab012 Revised parameters entered by Debaditya Bandyopadhyay, Sept 2022. JADC, Oct 2022 X Y _____________ Mutliplicity 3 2 _____________ Dependent: fkho_d Fe Fe3+ Dependent: kho_d Mg Fe3+ andr Ca Fe3+ Dependent: fkno_d Fe Cr knor Mg Cr Dependent: ckno_d Ca Cr Dependent: ftig_d Fe Al(MgTi)_0.5 tig_q Mg Al(MgTi)_0.5 Dependent: ctig_d Ca Al(MgTi)_0.5 alm Fe Al py Mg Al gr Ca Al Gt(TH) abbreviation Gt full_name garnet 7 2 3 4 ftig_d tig_q ctig_d fkno_d knor ckno_d fkho_d kho_d andr alm py gr 6 fkno_d = 1 alm + 1 knor - 1 py ckno_d = 1 gr + 1 knor - 1 py fkho_d = 1 alm + 1 andr - 1 gr kho_d = 1 py + 1 andr - 1 gr ftig_d = 1 alm + 1 tig_q - 1 py ctig_d = 1 gr + 1 tig_q - 1 py | see commentary in the header of this file for explanation of | endmember flags and subdivision schemes. 0 0 0 0 0 0 0 0 0 0 0 0 | endmember flags 0 1 .1 0 | subdivision scheme for Fe on A 0 1 .1 0 | subdivision scheme for Mg on A 0 1 .1 0 | subdivision scheme for Al(MgTi)_0.5 on B 0 1 .1 0 | subdivision scheme for Cr on B 0 1 .1 0 | subdivision scheme for Fe3+ on B begin_excess_function W(py alm) 4d3 + 0.1 * P W(py gr) 45.4d3 - 10 * T + 0.04 * P W(py andr) 107d3 - 10 * T - 0.036 * P W(py knor) 2d3 W(alm gr) 17d3 - 10 * T + 0.1 * P W(alm andr) 65d3 - 10 * T + 0.039 * P W(alm knor) 8.2d3 + 0.01 * P W(gr andr) 2d3 W(gr knor) 5d3 - 10 * T + 0.180 * P W(andr knor) 63d3 - 10 * T + 0.10 * P end_excess_function 2 3 3. z(m1,fe) = 1 alm z(m1,mg) = 1 py + 1 knor + 1 tig_q 5 2. z(m2,cr) = 1 knor z(m2,fe3) = 1 andr z(m2,ti) = 1/4 tig_q z(m2,mg) = 1/4 tig_q begin_van_laar_sizes alpha(py) 1 alpha(alm) 1 alpha(gr) 2.5 alpha(andr) 2.5 alpha(knor) 1 alpha(tig_q) 1 end_van_laar_sizes begin_dqf_corrections | what about knor?? JADC, 2/2026 dqf(tig_q) -31850 + 15.5 * T + 0.05 * P | tigTH end_dqf_corrections end_of_model -------------------------------------------------------- begin_model Orthopyroxene - Opx(TH) - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus Revised by Tomlinson and Holland; Journal of Petrology, 2021, 10.1093/petrology/egab012 Revised parameters entered by Debaditya Bandyopadhyay, Sept 2022. JADC, Oct 2022 Reformulated as a 2-polytope 688 format model. JADC, 9/19 The composition space of the published model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3, Cr, MTi on M1 C = Ca on M2 M = Mg, Fe A = Na on M2* *A is partially coupled to T by charge balance. The composition space of the model here has been simplified by eliminating the [FeTi][A,C] endmembers. The retained [FeTi][Fe] endmember is probably a waste of resources. NOTE: the definitions for the FeTi exchanges allowed in the full model are provided but have been commented out below. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = [T'][A,C] Polytope 2 = [T,M][M]+[M][C] where T' = Al, Fe3, MgTi on M1 -------------------------------------------------------- First coded for Perple_X by ECR Green and JRDC Aug 2018. Rearranged Apr 2019, JADC. NOTE: to use this the following endmembers must be specified with make definitions in the thermodynamic data file: -------------------------------------------------------- M1 M2 T ____________________________________ Multiplicity 1 1 1/2 <- fake T multiplicity ____________________________________ Polytope 1: crjd_d Cr Na SiSi dependent crdi_d Cr Ca AlSi dependent ___ ness_d Fe3 Na SiSi dependent cess_d (acm) Fe3 Ca AlSi dependent ___ ojd_q Al Na SiSi ocats_d Al Ca AlSi dependent ___ mnbuf_d MgTi Na SiSi dependent mcbuf_d MgTi Ca AlSi dependent fnbuf_d FeTi Na SiSi dependent fcbuf_d FeTi Ca AlSi dependent ____________________________________ Polytope 2: cren_q Cr Mg AlSi crfs_d Cr Fe AlSi dependent _________ macm_q (mess) Fe3 Mg AlSi facm_d (fess) Fe3 Fe AlSi dependent _________ mgts Al Mg AlSi fts_d Al Fe AlSi dependent _________ obuf_q MgTi Mg AlSi ffbuf_d FeTi Fe AlSi dependent _________ odi_q Mg Ca SiSi hed_d Fe Ca SiSi dependent ___ en Mg Mg SiSi fs Fe Fe SiSi ___________________________________ Ordered species: fm Mg Fe SiSi independent femg-1(M2) = fm - en femg-1(M1) = fs - fm mgca-1(M2) = en - odi_q feca-1(M2) = fm - odi_q -------------------------------------------------------- Opx(TH) abbreviation Opx full_name orthopyroxene 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and subdivision schemes for the composite composition space [T'][A,C] 0 .2 .1 0 | subdivision range for X(1) = X([T][A,C]) TM+MC by difference | ------------------------------------------- | Polytope 1 [T'][A,C] 2 | number of simplices 2 3 | change to 3 5 for full model crjd_d crdi_d ness_d cess_d |mnbuf_d mcbuf_d |fnbuf_d fcbuf_d ojd_q ocats_d | Simplex 1 X_K 0 1 .1 0 | X(1,1) = Na/[Na + Ca] on M2 X_Ca1 by difference | X(1,N) = Ca/[Na + Ca] on M2 by difference | Simplex 2 X_CrTs 0 1 .1 0 | X(2,1) = Cr/T on M1 X_FeTs 0 1 .1 0 | X(2,2) = Fe3+/T on M1 |0 1 .1 0 | X(2,3) = MgTi/T on M1 |0 1 .1 0 |X(2,4) = FeTi/T on M1 X_AlTs by difference | X(2,N) = Al/T on M1 by difference | ------------------------------------------- | Polytope 2 TM+MC 2 | number of simplices 2 6 | number of vertices on each simplex ffbuf_d obuf_q crfs_d cren_q facm_d macm_q fts_d mgts hed_d odi_q fs en | Simplex 1 X_Fe 0 .2 .1 0 | X(1,1) => Fe/M X_Mg by difference | X(1,N) = Mg/M by difference | Simplex 2 X_MTiTs 0 .2 .1 0 | X(2,1) - M-buf(MgTi) X_MCrTs 0 .1 .1 0 | X(2,2) - M-Cr X_MFeTs 0 .1 .1 0 | X(2,3) - M-Fe3+ X_MAlTs 0 .2 .1 0 | X(2,4) - M-Ts X_MCPx 0 1 .1 0 | X(2,5) - M-Ca X_MMPx by difference begin_ordered_endmembers fm = 1/2 en + 1/2 fs delta_g_of_ordering = -6600 end_ordered_endmembers begin_dependent_endmembers fts_d = 1 mgts - 1 en + 1 fm ocats_d = 1 mgts - 1 en + 1 odi_q crjd_d = 1 cren_q - 1 mgts + 1 ojd_q crfs_d = 1 cren_q - 1 en + 1 fm crdi_d = 1 cren_q - 1 en + 1 odi_q facm_d = 1 macm_q - 1 en + 1 fm cess_d = 1 macm_q - 1 en + 1 odi_q ness_d = 1 macm_q - 1 mgts + 1 ojd_q |mcbuf_d = 1 obuf_q + 1 odi_q - 1 en |fcbuf_d = 1 obuf_q + 1/2 fs - 1/2 fm + 1 odi_q - 1 en ffbuf_d = 1 obuf_q - 1 en + 1/2 fs + 1/2 fm hed_d = -1 fm + 1 fs + 1 odi_q end_dependent_endmembers begin_excess_function W(en fs) 5.2d3 W(en fm) 1.7d3 W(en odi_q ) 29d3 + 0.15 * p W(en mgts) 12.5d3 - 0.04 * p W(en cren_q) 8d3 W(en macm_q ) 8d3 W(en ojd_q ) 35d3 W(fs fm) 5d3 W(fs odi_q ) 25.54d3 + 0.084 * p W(fs mgts) 11d3 - 0.15 * p W(fs cren_q) 8d3 W(fs macm_q ) 10d3 W(fs ojd_q ) 35d3 W(fm odi_q ) 22.54d3 + 0.084 * p W(fm mgts) 15d3 - 0.15 * p W(fm cren_q) 10d3 W(fm macm_q ) 12d3 W(fm ojd_q ) 35d3 W(odi_q mgts) 75.5d3 - 0.84 * p W(odi_q cren_q) 20d3 W(odi_q macm_q ) 20d3 W(odi_q ojd_q ) 35d3 W(mgts cren_q) 2d3 W(mgts macm_q ) 2d3 W(mgts ojd_q ) 7d3 W(cren_q macm_q ) 2d3 W(cren_q ojd_q ) -11d3 W(obuf_q ojd_q ) 4d3 W(macm_q ojd_q ) -11d3 end_excess_function 3 | number of sites in entropy model (M1, M2, T) M1 | site name 6 1 1 | number of species, effective multiplicity, true multiplicity z(Fe,M1) = 1 fs z(Al,M1) = 1 mgts + 1 ojd_q z(Fe3+,M1) = 1 macm_q z(Cr,M1) = 1 cren_q z(Ti,M1) = 1/2 obuf_q z(Mg,M1) = 1/2 obuf_q + 1 odi_q + 1 en + 1 fm M2 | site name 4 1 1 | number of species, effective multiplicity, true multiplicity z(Fe,M2) = 1 fs + 1 fm z(Ca,M2) = 1 odi_q z(Na,M2) = 1 ojd_q z(Mg,M2) = 1 obuf_q + 1 en + 1 cren_q + 1 macm_q + 1 mgts T | site name 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 mgts + 1/2 cren_q + 1/2 obuf_q + 1/2 macm_q z(Si,T) = 1/2 mgts + 1/2 cren_q + 1/2 obuf_q + 1/2 macm_q + 1 ojd_q + 1 odi_q + 1 en + 1 fs + 1 fm [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(en) 1 alpha(fs) 1 alpha(fm) 1 alpha(odi_q ) 1.2 alpha(mgts) 1 alpha(cren_q) 1 alpha(obuf_q ) 1 alpha(macm_q ) 1 alpha(ojd_q ) 1.2 end_van_laar_sizes begin_dqf_corrections dqf(cren_q) -31850 + 15.5 * T + 0.05 * P | crenTH dqf(ojd_q) 18840 - 0.043 * P | = ojd dqf(odi_q) 1900 + 0.005 * P | = odi dqf(obuf_q) 4350 - 5.1 * T - 0.0061 * P | = obuf dqf(macm_q) 2630 - 0.089 * P | = mess end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Clinopyroxene - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus Revised by Tomlinson and Holland; Journal of Petrology, 2021, 10.1093/petrology/egab012 Revised parameters entered by Debaditya Bandyopadhyay, Sept 2022. JADC, Oct 2022 Reformulated as a 2-polytope 688 format model. JADC, 9/19 The composition space of the published model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3, Cr, MTi on M1 C = Ca on M2 M = Mg, Fe A = Na, K on M2* *A is partially coupled to T by charge balance. The composition space of the model here has been simplified by eliminating the [FeTi][A,C] endmembers. The retained [FeTi][Fe] endmember is probably a waste of resources. NOTE: the definitions for the FeTi exchanges allowed in the full model are provided but have been commented out below. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = [T'][A,C] Polytope 2 = [T,M][M]+[M][C] where T' = Al, Fe3, MgTi on M1 --------------------------------------------------------- First coded for Perple_X by Julien Cornet Sept 2018. Re-formulated as a triple simplex prism. JADC, Dec 11, 2018. Corrected cfm enthalpy and site limit expression for z(Mg,M1). JADC, April 29, 2019. NOTES: this model requires that the following endmembers are specified by make definitions in the thermodynamic data file: --------------------------------------------------------- M1 M2 T ____________________________________ Multiplicity 1 1 2 -> fake T multiplicity = 1/2 ____________________________________ Polytope 1: mkbuf_d MgTi K Si dependent mnbuf_d MgTi Na Si dependent mcbuf_q MgTi Ca AlSi _________ fkbuf_d FeTi K Si dependent fnbuf_d FeTi Na Si dependent fcbuf_d FeTi Ca AlSi dependent _________ crkjd_d Cr K Si dependent crjd_d Cr Na Si dependent crdi_q Cr Ca AlSi _________ kess_d Fe3+ K Si dependent ness_d Fe3+ Na Si dependent cess_q Fe3 Ca AlSi _________ kjd_q Al K Si jd Al Na Si cats Al Ca AlSi ____________________________________ Polytope 2: mmbuf_d MgTi Mg AlSi dependent ffbuf_d FeTi Fe AlSi dependent cren_d Cr Mg AlSi dependent crfs_d Cr Fe AlSi dependent mess_d Fe3 Mg AlSi dependent fess_d Fe3 Fe AlSi dependent mats_d Al Mg AlSi dependent fats_d Al Fe AlSi dependent di Mg Ca Si hed_d Fe Ca Si dependent cen_q Mg Mg Si cfs_q Fe Fe Si _______________________________________________ Ordered species: cfm Mg Fe Si independent femg-1(M2) = cfm - cen_q femg-1(M1) = cfs_q - cfm mgca-1(M2) = cen_q - di feca-1(M2) = cfm - di --------------------------------------------------------- Cpx(TH) abbreviation Cpx full_name clinopyroxene 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [T'][A,C] 0 .25 .1 0 TM+MC by difference | ------------------------------------------- | Polytope 1 [T'][A,C] 2 | number of simplices 3 4 | number of vertices on each simplex, change to 3 5 for full model mkbuf_d mnbuf_d mcbuf_q crkjd_d crjd_d crdi_q kess_d ness_d cess_q |fkbuf_d fnbuf_d fcbuf_d kjd_q jd cats | Simplex 1 X_K 0 1 .1 0 | X(1,1) = Na/[A + Ca] on M2 X_Na 0 1 .1 0 | X(1,2) = K /[A + Ca] on M2 X_Ca by difference | X(1,N) = Ca/[A + Ca] on M2 | Simplex 2 X_TiTs 0 1 .1 0 | X(2,1) = MgTi/T on M1 X_CrTs 0 1 .1 0 | X(2,2) = Cr/T on M1 X_FeTs 0 1 .1 0 | X(2,3) = Fe3+/T on M1 |0 1 .1 0 |X(2,4) = FeTi/T on M1 X_AlTs by difference | X(2,N) = Al/T on M1 | ------------------------------------------- | Polytope 2 TM+MC 2 | number of simplices 2 6 | number of vertices on each simplex ffbuf_d mmbuf_d crfs_d cren_d fess_d mess_d fats_d mats_d cfs_q cen_q hed_d di | Simplex 1 X_Fe 0 .2 .1 0 | X(1,1) => Fe/M X_Mg by difference | X(1,N) = Mg/M | Simplex 2 X_MTiTs 0 .1 .1 0 | X(2,1) - M-buf(MgTi) X_MCrTs 0 .1 .1 0 | X(2,2) - M-Cr X_MFeTs 0 .1 .1 0 | X(2,3) - M-Fe3+ X_MAlTs 0 .2 .1 0 | X(2,4) - M-Ts X_MMPx 0 1 .1 0 | X(2,5) - M-M X_MCPx by difference | X(2,6) - M-Ca begin_ordered_endmembers cfm = 1/2 cen_q + 1/2 cfs_q delta_g_of_ordering = -4400 | DQF(cfm) - (DQF(cen_q) + DQF(cfs_q ))/2 end_ordered_endmembers begin_dependent_endmembers mkbuf_d = -1 cats + 1 mcbuf_q + 1 kjd_q mnbuf_d = -1 cats + 1 mcbuf_q + 1 jd |fcbuf_d = 1/2 cfs_q + 1 mcbuf_q - 1/2 cfm |fkbuf_d = 1/2 cfs_q - cats + mcbuf_q - 1/2 cfm + kjd_q |fnbuf_d = 1/2 cfs_q - 1 cats + 1 mcbuf_q + jd - 1/2 cfm ffbuf_d = -1 di + 1/2 cfs_q + 1 mcbuf_q + 1/2 cfm mmbuf_d = -1 di + 1 mcbuf_q + 1 cen_q crkjd_d = -1 cats + 1 crdi_q + 1 kjd_q crjd_d = -1 cats + 1 crdi_q + 1 jd cren_d = -1 di + 1 crdi_q + 1 cen_q crfs_d = -1 di + 1 crdi_q + 1 cfm kess_d = -1 cats + 1 cess_q + 1 kjd_q ness_d = -1 cats + 1 cess_q + 1 jd mess_d = -1 di + 1 cess_q + 1 cen_q fess_d = -1 di + 1 cess_q + 1 cfm mats_d = -1 di + 1 cats + 1 cen_q fats_d = -1 di + 1 cats + 1 cfm hed_d = 1 di + 1 cfs_q - 1 cfm end_dependent_endmembers begin_excess_function W(di cfs_q ) 25.8d3 - 0.03 * p W(di cats) 13d3 - 0.06 * p W(di crdi_q) 8d3 W(di cess_q) 8d3 W(di jd) 26d3 W(di cen_q) 29.8d3 - 0.03 * p W(di cfm) 20.6d3 - 0.03 * p W(di kjd_q) 26d3 W(cfs_q cats) 25d3 - .1 * p W(cfs_q crdi_q) 38.3d3 W(cfs_q cess_q) 43.3d3 W(cfs_q jd) 24d3 W(cfs_q cen_q) 2.3d3 W(cfs_q cfm) 3.5d3 W(cfs_q kjd_q) 24d3 W(cats crdi_q) 2d3 W(cats cess_q) 2d3 W(cats jd) 6d3 W(cats cen_q) 45.2d3 - 0.35 * p W(cats cfm) 27d3 - .1 * p W(cats kjd_q) 6d3 W(crdi_q cess_q) 2d3 W(crdi_q jd) 3d3 W(crdi_q cen_q) 52.3d3 W(crdi_q cfm) 40.3d3 W(crdi_q kjd_q) 3d3 W(cess_q jd) 3d3 W(cess_q cen_q) 57.3d3 W(cess_q cfm) 45.3d3 W(cess_q kjd_q) 3d3 W(jd cen_q) 40d3 W(jd cfm) 40d3 W(jd kjd_q) 10d3 W(cen_q cfm) 4d3 W(cen_q kjd_q) 40d3 W(cfm kjd_q) 40d3 end_excess_function 3 | number of sites in configurational entropy model (M1, M2, T) M1 | site name 6 1 1 | number of species, effective multiplicity, true multiplicity z(Fe) = 1 cfs_q z(Mg) = 1 di + 1 cen_q + 1/2 mcbuf_q + 1 cfm z(Fe3+) = 1 cess_q z(Ti) = 1/2 mcbuf_q z(Cr) = 1 crdi_q z(Al) = 1 cats + 1 jd + 1 kjd_q M2 | site name 5 1 1 | number of species, effective multiplicity, true multiplicity z(Na) = 1 jd z(K) = 1 kjd_q z(Mg) = 1 cen_q z(Fe) = 1 cfm + 1 cfs_q z(Ca) = 1 di + 1 cats + 1 cess_q + 1 crdi_q + 1 mcbuf_q T | site name 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 cats + 1/2 mcbuf_q + 1/2 crdi_q + 1/2 cess_q z(Si,T) = 1/2 cats + 1/2 mcbuf_q + 1/2 crdi_q + 1/2 cess_q + 1 cfs_q + 1 cen_q + 1 di + 1 cfm + 1 jd + 1 kjd_q [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(di) 1.2 alpha(cen_q) 1 alpha(cfs_q ) 1 alpha(jd) 1.2 alpha(kjd_q) 1.2 alpha(cats) 1.9 alpha(mcbuf_q) 1.9 alpha(cess_q) 1.9 alpha(crdi_q) 1.9 alpha(cfm) 1 end_van_laar_sizes begin_dqf_corrections dqf(cen_q) 3500 - 2 * T + 0.048 * P | cenjh dqf(cfs_q) 2100 - 2 * T + 0.045 * P | cfs_q dqf(cess_q) -5810 | cessTH dqf(mcbuf_q) -6.47d3 - 1.2 * T + 0.078 * P | mcbufTH dqf(crdi_q) -5660 | crdiTH dqf(kjd_q) -4d3 + 1.35 * P | kdjTH end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Ternary feldspar 4TR model Holland, TJB, Green, ECR & Powell, R (2021) * entered by Bob Myhill, 2023/07/08 * swapped abh for ab? converted to 688 format, JADC, 13/7/23 * reverted to ab, though how this is compatible with san is beyond me. JADC, 30/8/23 * if this is the "plagioclase parameterization" then is there supposed to be another model for alkali fsp? * Thermocalc input: % ternary feldspar, 4TR model, with plagioclase-friendly % parameterisation. % % Holland, TJB, Green, ECR & Powell, R (2021). A thermodynamic model % for feldspars in KAlSi3O8-NaAlSi3O8-CaAl2Si2O8 for mineral % equilibrium calculations. Journal of Metamorphic Geology, 1-14. % Published online as DOI 10.1111/jmg.12639 % % E-m Formula Mixing sites % A TB* % Na Ca K Al Si % ab NaAlSi3O8 1 0 0 1 3 % san KAlSi3O8 0 0 1 1 3 % an CaAl2Si2O8 0 1 0 2 2 % *use 1/4 entropy of mixing from TB-sites asf W(ab,an) 14.6 -0.00935 -0.04 W(ab,san) 24.1 -0.00957 0.338 W(an,san) 48.5 0 -0.13 ab 0.674 0 0 an 0.550 0 0 san 1.000 0 0 --------------------------------------------------------- Fsp(HGP) | solution name. abbreviation Fsp full_name ternary-feldspar 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 3 | number of vertices on each simplex ab | endmembers on the vertices an san 0. 1 .1 0 | range and resolution for albite, imod = 0 -> cartesian subdivision 0. 1 .1 0 | range and resolution for anorthite, imod = 0 -> cartesian subdivision begin_excess_function w(ab an) 14600 -9.35 * T_K - 0.04 * P_bar w(ab san) 24100 -9.57 * T_K + 0.338 * P_bar w(an san) 48500 -0.13 * P_bar end_excess_function 2 | number of sites in configurational entropy model M | site name 3 1 1 | number of species, effective multiplicity, true multiplicity z(Na,M) = 1 ab z(Ca,M) = 1 an z(K,M) = 1 san T | site name 2 1 4 | number of species, effective multiplicity, true multiplicity z(Al,T) = 0.25 ab + 0.25 san + 0.5 an z(Si,T) = 0.75 ab + 0.75 san + 0.5 an [O8] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(ab) 0.674 alpha(an) 0.555 alpha(san) 1 end_van_laar_sizes end_of_model -------------------------------------------------------- begin_model Model for the Na-rich side of the Peristerite gap to be used together with Fsp(HGP) from: Holland, TJB, Green, ECR & Powell, R (2021) Entered by Maria M. Ariza, 2024/01/17 --------------------------------------------------------- lAb(HGP) | solution name. abbreviation lAb full_name binary-feldspar 688 | model type: 688 format standard model 1 | number of polytopes 1 | number of simplices 2 | number of vertices on each simplex ab | endmembers on the vertices an 0. 1 .1 0 | range and resolution for albite, imod = 0 -> cartesian subdivision begin_excess_function w(an ab) 3400 end_excess_function 1 | number of sites in configurational entropy model M | site name 2 1 1 | number of species, effective multiplicity, true multiplicity z(Na,M) = 1 ab z(Ca,M) = 1 an [O8] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(ab) 0.64 alpha(an) 1 end_van_laar_sizes begin_dqf_corrections dqf(ab) = -1746 + 2 * T_K dqf(an) = 10000 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model melt(SP) abbreviation Liq full_name liquid 2 model type: simplicial composition space 13 number of endmembers foL_q faL_q zr8L abL siL_q anL kspL wiL_q qL_q alp8L cep8L yp8L wL_q 0 0 0 0 0 0 0 0 0 0 0 0 0 | endmember flags. 0 1 .1 0 | range and resolution of X(fo), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(fa), 0 => cartesian subdivision 0 .1 .1 0 | range and resolution of X(zrL), 0 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(ab), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(sil), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(an), 0 => cartesian subdivision 0 1 .1 0 | range and resolution of X(ksp), 0 => cartesian subdivision 0 .1 .1 0 | range and resolution of X(wiL), 0 => asymmetric subdivision 0 1 .1 0 | range and resolution of X(q), 0 => cartesian subdivision 0 .1 .1 0 | range and resolution of X(apL), 0 => asymmetric subdivision 0 .1 .1 0 | range and resolution of X(cepL), 0 => asymmetric subdivision 0 .1 .1 0 | range and resolution of X(ypL), 0 => asymmetric subdivision begin_excess_function | the excess parameters used for the haplo- | granite melt model vary between papers (notably w(an-sil)), | the values here are from the thermocalc | model file 'thdmloss.txt' and appear largely | consistent with the white et al paper. w(qL_q anL) -10d3 w(qL_q siL_q) 12d3 w(qL_q abL) 12d3 -0.4 * P_bar w(qL_q kspL) -2d3 -0.5 * P_bar w(qL_q wL_q) 15d3 w(qL_q foL_q) 12d3 -0.4 * P_bar w(qL_q faL_q) 14d3 w(abL siL_q) 12d3 w(abL kspL) -6d3 3. * P_bar w(abL foL_q) 10d3 w(abL faL_q) 2d3 w(abL wL_q) 1d3 -0.2 * P_bar w(kspL anL) 0 -1. * P_bar w(kspL siL_q) 12d3 w(kspL wL_q) 11d3 -0.45 * P_bar w(kspL foL_q) 12d3 w(kspL faL_q) 12d3 | w(anL sil8) is 12 kJ in the Holland & Powell J Pet paper | but appears to have accidentally been set to zero | for the calculations published in the White et al. JMG | paper. To restore this parameter delete the comment | marker "|" on the following line: | w(anL sil8) 12d3 w(anL wL_q) 9d3 -0.85 * P_bar w(siL_q foL_q) 12d3 w(siL_q faL_q) 12d3 w(siL_q wL_q) 16d3 w(foL_q faL_q) 18d3 w(foL_q wL_q) 11d3 -0.5 * P_bar w(faL_q wL_q) 12d3 w(qL_q wiL_q) 14d3 w(abL wiL_q) 10d3 w(kspL wiL_q) 12d3 w(wiL_q foL_q) 18d3 w(wiL_q faL_q) 18d3 w(wiL_q wL_q) 11d3 -0.5 * P_bar w(qL_q cep8L) 14d3 w(abL cep8L) 12d3 w(kspL cep8L) 12d3 w(cep8L foL_q) 12d3 w(cep8L faL_q) 12d3 w(cep8L wL_q) 25d3 w(qL_q yp8L) 14d3 w(abL yp8L) 12d3 w(kspL yp8L) 12d3 w(yp8L foL_q) 12d3 w(yp8L faL_q) 12d3 w(yp8L wL_q) 25d3 w(qL_q alp8L) 14d3 w(abL alp8L) 12d3 w(kspL alp8L) 12d3 w(alp8L foL_q) 12d3 w(alp8L faL_q) 12d3 w(alp8L wL_q) 25d3 end_excess_function 4 | Configurational entropy: two non-temkin sites (Water, Melt) | and one temkin site (olvine). hp assume a fsp = ab + or "molecule"" | with mixing on a temkin M site, but the math works out the same as | the endmembers are treated as separate endmembers and the M site | dropped. 2 1 | water-vacancy site z(H) = 1 wL_q 3 0 | temkin olivine site n(Mg) = 4 foL_q n(Fe) = 4 faL_q n(Zn) = 4 wiL_q 9 0 | silicate species site z(w) = 1 wL_q z(q) = 1 qL_q z(ksp) = 1 kspL z(ab) = 1 abL z(sil) = 1 siL_q z(an) = 1 anL z(zr) = 1 zr8L z(ol) = 1 foL_q + 1 faL_q + 1 wiL_q z(phos) = 1 alp8L + 1 cep8L + 1 yp8L 3 2 z(P) = 1 alp8L z(CE) = 1 cep8L begin_dqf_corrections dqf(wL_q) = 0 dqf(wiL_q) = 0 dqf(qL_q) = 0 dqf(foL_q) = -10d3 dqf(faL_q) = -9d3 0 -1.3 dqf(siL_q) = -10d3 | pre-white et al '07 value: dqf(foL_q) -15d3 0 0 | pre-white et al '07 value: dqf(faL_q) -15d3 0 0 | pre-white et al '07 value: dqf(siL_q) -10d3 0 0 end_dqf_corrections end_of_model -------------------------------------------------------- begin_model HP '98 quaternary garnet + Y model Spear and Pyle, 2010 converted to HP by Shrestha et al.,2019. F. Gervais COVID time 2020 YGt(SP) abbreviation Gt full_name garnet 2 model type: simplicial composition space 5 number of endmembers yag spss py gr alm endmember names 1 1 0 0 0 | endmember flags 1e-5 1e-4 .1 1 | imod = 1 -> non-linear subdivision 0 1 .1 0 | 0 1 .1 0 | imod = 0 -> cartesian subdivision 0 1 .1 0 | imod = 0 -> cartesian subdivision | NOTE restricted subdivision range on Mn (Species 1)! begin_excess_function w(py gr) 33000 w(alm py) 2500 | hp '98 give 2.4 kJ w(py spss) 4500 w(alm spss) 240 end_excess_function 1 | 1 site entropy model 5 3 | 5 species, site multiplicity 3 z(Fe) = 1 alm z(Mg) = 1 py z(Ca) = 1 gr z(Mn) = 1 spss end_of_model -------------------------------------------------------- begin_model Ideal monazite from Spear and Pyle, 2010 converted to HP by Shrestha et al.,2019. F. Gervais COVID time 2020 M _____________ Mutliplicity 2 _____________ cemnz ce ymnz y Mnz(SP) abbreviation mz full_name monazite 2 | model type: simplicial 2 | endmembers cemnz ymnz | see commentary in the header of this file for explanation of | endmember flags and subdivision schemes. 0 0 | endmember flags 0 1 .1 0 | cemnz ideal 1 | 1 site entropy model 2 2 | M 2 species, site multiplicity = 1 z(ce) = 1 cemnz end_of_model -------------------------------------------------------- begin_model Orthopyroxene - Opx(W24) - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for pressure to 7 GPa, 923 K < T < peridotite liquidus Revised by Tomlinson and Holland; Journal of Petrology, 2021, 10.1093/petrology/egab012 Revised by Weller, OM, Holland, TJB, Soderman, CR, Green, ECR, Powell, R, Beard, CD & Riel, N (2024). Journal of Petrology, 65,doi: 10.1093/petrology/egae098 To reduce the zillions of TC make definitions odi, ojd are replaced by di and jd with internal dqfs. JADC, Jan 2026 -------------------------------------------------------- The composition space of the published model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3, Cr, MTi on M1 C = Ca on M2 M = Mg, Fe A = Na on M2* *A is partially coupled to T by charge balance. The composition space of the model here has been simplified by eliminating the [FeTi][A,C] endmembers. The retained [FeTi][Fe] endmember is probably a waste of resources. NOTE: the definitions for the FeTi exchanges allowed in the full model are provided but have been commented out below. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = [T'][A,C] Polytope 2 = [T,M][M]+[M][C] where T' = Al, Fe3, MgTi on M1 -------------------------------------------------------- M1 M2 T ____________________________________ Multiplicity 1 1 1/2 <- fake T multiplicity ____________________________________ Polytope 1: crjd_d Cr Na Si dependent crdi_d Cr Ca AlSi dependent ___ ness_d Fe3 Na Si dependent cess_d (acm) Fe3 Ca AlSi dependent ___ ojd_q Al Na Si ocats_d Al Ca AlSi dependent ___ mnbuf_d MgTi Na Si dependent mcbuf_d MgTi Ca AlSi dependent fnbuf_d FeTi Na Si dependent fcbuf_d FeTi Ca AlSi dependent ____________________________________ Polytope 2: cren1_q Cr Mg AlSi crfs_d Cr Fe AlSi dependent _________ macm_q (mess) Fe3 Mg AlSi facm_d (fess) Fe3 Fe AlSi dependent _________ mgts Al Mg AlSi fts_d Al Fe AlSi dependent _________ obuf_q MgTi Mg AlSi ffbuf_d FeTi Fe AlSi dependent _________ odi_q Mg Ca Si hed_d Fe Ca Si dependent ___ en Mg Mg Si fs Fe Fe Si ___________________________________ Ordered species: fm Mg Fe Si independent femg-1(M2) = fm - en femg-1(M1) = fs - fm mgca-1(M2) = en - odi_q feca-1(M2) = fm - odi_q -------------------------------------------------------- Opx(W24) abbreviation Opx full_name orthopyroxene 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and subdivision schemes for the composite composition space [T'][A,C] 0 1 .1 0 | subdivision range for X(1) = X([T][A,C]) TM+MC by difference | ------------------------------------------- | Polytope 1 [T'][A,C] 2 | number of simplices 2 3 | change to 3 5 for full model crjd_d crdi_d ness_d cess_d |mnbuf_d mcbuf_d |fnbuf_d fcbuf_d ojd_q ocats_d | Simplex 1 X_K 0 1 .1 0 | X(1,1) = Na/[Na + Ca] on M2 X_Ca1 by difference | X(1,N) = Ca/[Na + Ca] on M2 by difference | Simplex 2 X_CrTs 0 1 .1 0 | X(2,1) = Cr/T on M1 X_FeTs 0 1 .1 0 | X(2,2) = Fe3+/T on M1 |0 1 .1 0 | X(2,3) = MgTi/T on M1 |0 1 .1 0 |X(2,4) = FeTi/T on M1 X_AlTs by difference | X(2,N) = Al/T on M1 by difference | ------------------------------------------- | Polytope 2 TM+MC 2 | number of simplices 2 6 | number of vertices on each simplex ffbuf_d obuf_q crfs_d cren1_q facm_d macm_q fts_d mgts hed_d odi_q fs en | Simplex 1 X_Fe 0 1 .1 0 | X(1,1) => Fe/M X_Mg by difference | X(1,N) = Mg/M by difference | Simplex 2 X_MTiTs 0 1 .1 0 | X(2,1) - M-buf(MgTi) X_MCrTs 0 1 .1 0 | X(2,2) - M-Cr X_MFeTs 0 1 .1 0 | X(2,3) - M-Fe3+ X_MAlTs 0 1 .1 0 | X(2,4) - M-Ts X_MCPx 0 1 .1 0 | X(2,5) - M-Ca X_MMPx by difference begin_ordered_endmembers fm = 1/2 en + 1/2 fs delta_g_of_ordering = -6600 end_ordered_endmembers begin_dependent_endmembers fts_d = 1 mgts - 1 en + 1 fm ocats_d = 1 mgts - 1 en + 1 odi_q crjd_d = 1 cren1_q - 1 mgts + 1 ojd_q crfs_d = 1 cren1_q - 1 en + 1 fm crdi_d = 1 cren1_q - 1 en + 1 odi_q facm_d = 1 macm_q - 1 en + 1 fm cess_d = 1 macm_q - 1 en + 1 odi_q ness_d = 1 macm_q - 1 mgts + 1 ojd_q |mcbuf_d = 1 obuf_q + 1 odi_q - 1 en |fcbuf_d = 1 obuf_q + 1/2 fs - 1/2 fm + 1 di - 1 en ffbuf_d = 1 obuf_q - 1 en + 1/2 fs + 1/2 fm hed_d = -1 fm + 1 fs + 1 odi_q end_dependent_endmembers begin_excess_function w(en fs) +7000 w(en fm) +3500 w(en odi_q) +29000 +0.15 * P_bar w(en mgts) +12500 -0.04 * P_bar w(en cren1_q) +8000 w(en obuf_q) +6000 w(en macm_q) +8000 w(en ojd_q) +35000 w(fs fm) +4500 w(fs odi_q) +23000 +0.08 * P_bar w(fs mgts) +11000 -0.15 * P_bar w(fs cren1_q) +10000 w(fs obuf_q) +7000 w(fs macm_q) +10000 w(fs ojd_q) +35000 w(fm odi_q) +19000 +0.08 * P_bar w(fm mgts) +15000 -0.15 * P_bar w(fm cren1_q) +12000 w(fm obuf_q) +8000 w(fm macm_q) +12000 w(fm ojd_q) +35000 w(odi_q mgts) +75500 -0.84 * P_bar w(odi_q cren1_q) +20000 w(odi_q obuf_q) +40000 w(odi_q macm_q) +20000 w(odi_q ojd_q) +35000 w(mgts cren1_q) +2000 w(mgts obuf_q) +10000 w(mgts macm_q) +2000 w(mgts ojd_q) +7000 w(cren1_q obuf_q) +6000 w(cren1_q macm_q) +2000 w(cren1_q ojd_q) -11000 w(obuf_q macm_q) +6000 w(obuf_q ojd_q) +20000 w(macm_q ojd_q) -11000 end_excess_function 3 | number of sites in entropy model (M1, M2, T) M1 | site name 6 1 1 | number of species, effective multiplicity, true multiplicity z(fe) = 1 fs z(al) = 1 mgts + 1 ojd_q z(fe3+) = 1 macm_q z(cr) = 1 cren1_q z(ti) = 1/2 obuf_q z(Mg) = 1/2 obuf_q + 1 odi_q + 1 en + 1 fm M2 | site name 4 1 1 | number of species, effective multiplicity, true multiplicity z(fe) = 1 fs + 1 fm z(ca) = 1 odi_q z(na) = 1 ojd_q z(Mg) = 1 obuf_q + 1 en + 1 cren1_q + 1 macm_q + 1 mgts T | site name 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 mgts + 1/2 cren1_q + 1/2 obuf_q + 1/2 macm_q z(Si,T) = 1/2 mgts + 1/2 cren1_q + 1/2 obuf_q + 1/2 macm_q + 1 ojd_q + 1 odi_q + 1 en + 1 fs + 1 fm [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(en) 1 alpha(fs) 1 alpha(fm) 1 alpha(odi_q) 1.2 alpha(mgts) 1 alpha(cren1_q) 1 alpha(obuf_q) 1 alpha(macm_q) 1 alpha(ojd_q) 1.2 end_van_laar_sizes begin_dqf_corrections dqf(ojd_q) = 18200 | = ojd dqf(odi_q) = 2.8d3 + 0.005 P | = odi dqf(obuf_q) =-3.91d3 - 5.1 T + 0.37 P | = obuf dqf(macm_q) = 3000 | = mess dqf(cren1_q) = -6d3 + 0.14 P end_dqf_corrections | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers end_of_model -------------------------------------------------------- begin_model Clinopyroxene - Cpx(W24) - Holland, Green, and Powell; Journal of Petrology, 2018, Vol. 59, 881 900. Intended for low Ca-px at pressure to 7 GPa, 923 K < T < peridotite liquidus Revised by Tomlinson and Holland; Journal of Petrology, 2021, 10.1093/petrology/egab012 Revised by Weller, OM, Holland, TJB, Soderman, CR, Green, ECR, Powell, R, Beard, CD & Riel, N (2024). Journal of Petrology, 65,doi: 10.1093/petrology/egae098 JADC, Jan 2026 The composition space of the published model corresponds to (M1, M2 site populations are indicated, remaining are dependent): [M,T][A,C,M] T = Al, Fe3, Cr, MTi on M1 C = Ca on M2 M = Mg, Fe A = Na, K on M2* *A is partially coupled to T by charge balance. The composition space of the model here has been simplified by eliminating the [FeTi][A,C] endmembers. The retained [FeTi][Fe] endmember is probably a waste of resources. NOTE: the definitions for the FeTi exchanges allowed in the full model are provided but have been commented out below. The composition space of the model is formed as a mixture of two polytopes (dropping the dependent A site notation): Polytope 1 = [T'][A,C] Polytope 2 = [T,M][M]+[M][C] where T' = Al, Fe3, MgTi on M1 M1 M2 T ____________________________________ Multiplicity 1 1 2 -> fake T multiplicity = 1/2 ____________________________________ Polytope 1: mkbuf_d MgTi K Si dependent mnbuf_d MgTi Na Si dependent mcbuf_q MgTi Ca AlSi _________ fkbuf_d FeTi K Si dependent fnbuf_d FeTi Na Si dependent fcbuf_d FeTi Ca AlSi dependent _________ crkjd_d Cr K Si dependent crjd_d Cr Na Si dependent crdi_q Cr Ca AlSi _________ kess_d Fe3+ K Si dependent ness_d Fe3+ Na Si dependent cess_q Fe3 Ca AlSi _________ kjd_q Al K Si jd Al Na Si cats Al Ca AlSi ____________________________________ Polytope 2: mmbuf_d MgTi Mg AlSi dependent ffbuf_d FeTi Fe AlSi dependent cren_d Cr Mg AlSi dependent crfs_d Cr Fe AlSi dependent mess_d Fe3 Mg AlSi dependent fess_d Fe3 Fe AlSi dependent mats_d Al Mg AlSi dependent fats_d Al Fe AlSi dependent di Mg Ca Si hed_d Fe Ca Si dependent cen_q Mg Mg Si cfs_q Fe Fe Si _______________________________________________ Ordered species: cfm Mg Fe Si independent femg-1(M2) = cfm - cen_q femg-1(M1) = cfs_q - cfm mgca-1(M2) = cen_q - di feca-1(M2) = cfm - di --------------------------------------------------------- Cpx(W24) abbreviation Cpx full_name clinopyroxene 688 | model type: 688 format standard model 2 | number of polytopes | polytope names and composite composition space subdivision schemes [T'][A,C] 0 1 .1 0 TM+MC by difference | ------------------------------------------- | Polytope 1 [T'][A,C] 2 | number of simplices 3 4 | number of vertices on each simplex, change to 3 5 for full model mkbuf_d mnbuf_d mcbuf_q crkjd_d crjd_d crdi_q kess_d ness_d cess_q |fkbuf_d fnbuf_d fcbuf_d kjd_q jd cats | Simplex 1 X_K 0 1 .1 0 | X(1,1) = Na/[A + Ca] on M2 X_Na 0 1 .1 0 | X(1,2) = K /[A + Ca] on M2 X_Ca by difference | X(1,N) = Ca/[A + Ca] on M2 | Simplex 2 X_TiTs 0 1 .1 0 | X(2,1) = MgTi/T on M1 X_CrTs 0 1 .1 0 | X(2,2) = Cr/T on M1 X_FeTs 0 1 .1 0 | X(2,3) = Fe3+/T on M1 |0 1 .1 0 |X(2,4) = FeTi/T on M1 X_AlTs by difference | X(2,N) = Al/T on M1 | ------------------------------------------- | Polytope 2 TM+MC 2 | number of simplices 2 6 | number of vertices on each simplex ffbuf_d mmbuf_d crfs_d cren_d fess_d mess_d fats_d mats_d cfs_q cen_q hed_d di | Simplex 1 X_Fe 0 1 .1 0 | X(1,1) => Fe/M X_Mg by difference | X(1,N) = Mg/M | Simplex 2 X_MTiTs 0 1 .1 0 | X(2,1) - M-buf(MgTi) X_MCrTs 0 1 .1 0 | X(2,2) - M-Cr X_MFeTs 0 1 .1 0 | X(2,3) - M-Fe3+ X_MAlTs 0 1 .1 0 | X(2,4) - M-Ts X_MMPx 0 1 .1 0 | X(2,5) - M-M X_MCPx by difference | X(2,6) - M-Ca begin_ordered_endmembers cfm = 1/2 cen_q + 1/2 cfs_q delta_g_of_ordering = -5100 | DQF(cfm) - (DQF(cen_q) + DQF(cfs_q))/2 | cfm -1.6 -0.002 0.0465 | cen 3.5 -0.002 0.048 | cfs 3.5 -0.002 0.045 end_ordered_endmembers begin_dependent_endmembers mkbuf_d = -1 cats + 1 mcbuf_q + 1 kjd_q mnbuf_d = -1 cats + 1 mcbuf_q + 1 jd |fcbuf_d = 1/2 cfs_q + 1 mcbuf_q - 1/2 cfm |fkbuf_d = 1/2 cfs_q - 1 cats + 1 mcbuf_q - 1/2 cfm + 1 kjd_q |fnbuf_d = 1/2 cfs_q - 1 cats + 1 mcbuf_q + jd - 1/2 cfm ffbuf_d = -1 di + 1/2 cfs_q + 1 mcbuf_q + 1/2 cfm mmbuf_d = -1 di + 1 mcbuf_q + 1 cen_q crkjd_d = -1 cats + 1 crdi_q + 1 kjd_q crjd_d = -1 cats + 1 crdi_q + 1 jd cren_d = -1 di + 1 crdi_q + 1 cen_q crfs_d = -1 di + 1 crdi_q + 1 cfm kess_d = -1 cats + 1 cess_q + 1 kjd_q ness_d = -1 cats + 1 cess_q + 1 jd mess_d = -1 di + 1 cess_q + 1 cen_q fess_d = -1 di + 1 cess_q + 1 cfm mats_d = -1 di + 1 cats + 1 cen_q fats_d = -1 di + 1 cats + 1 cfm hed_d = 1 di + 1 cfs_q - 1 cfm end_dependent_endmembers begin_excess_function w(di cfs_q) +25800 -0.03 * P_bar w(di cats) +13000 -0.06 * P_bar w(di crdi_q) +8000 w(di cess_q) +8000 w(di mcbuf_q) +8000 w(di jd) +26000 w(di cen_q) +29800 -0.03 * P_bar w(di cfm) +20600 -0.03 * P_bar w(di kjd_q) +26000 w(cfs_q cats) +25000 -0.1 * P_bar w(cfs_q crdi_q) +38300 w(cfs_q cess_q) +43300 w(cfs_q mcbuf_q) +24000 w(cfs_q jd) +24000 w(cfs_q cen_q) +2300 w(cfs_q cfm) +3500 w(cfs_q kjd_q) +24000 w(cats crdi_q) +2000 w(cats cess_q) +2000 w(cats mcbuf_q) +6000 w(cats jd) +6000 w(cats cen_q) +45200 -0.35 * P_bar w(cats cfm) +27000 -0.1 * P_bar w(cats kjd_q) +6000 w(crdi_q cess_q) +2000 w(crdi_q mcbuf_q) +6000 w(crdi_q jd) +3000 w(crdi_q cen_q) +52300 w(crdi_q cfm) +40300 w(crdi_q kjd_q) +3000 w(cess_q mcbuf_q) +6000 w(cess_q jd) +3000 w(cess_q cen_q) +57300 w(cess_q cfm) +45300 w(cess_q kjd_q) +3000 w(mcbuf_q jd) +16000 w(mcbuf_q cen_q) +24000 w(mcbuf_q cfm) +22000 w(mcbuf_q kjd_q) +16000 w(jd cen_q) +40000 w(jd cfm) +26000 w(jd kjd_q) +28000 w(cen_q cfm) +4000 w(cen_q kjd_q) +40000 w(cfm kjd_q) +40000 end_excess_function 3 | number of sites in configurational entropy model (M1, M2, T) M1 | site name 6 1 1 | number of species, effective multiplicity, true multiplicity z(Fe) = 1 cfs_q z(Mg) = 1 di + 1 cen_q + 1/2 mcbuf_q + 1 cfm z(Fe3+) = 1 cess_q z(Ti) = 1/2 mcbuf_q z(Cr) = 1 crdi_q z(Al) = 1 cats + 1 jd + 1 kjd_q M2 | site name 5 1 1 | number of species, effective multiplicity, true multiplicity z(Na) = 1 jd z(K) = 1 kjd_q z(Mg) = 1 cen_q z(Fe) = 1 cfm + 1 cfs_q z(Ca) = 1 di + 1 cats + 1 cess_q + 1 crdi_q + 1 mcbuf_q T | site name 2 .5 2 | number of species, effective multiplicity, true multiplicity z(Al,T) = 1/2 cats + 1/2 mcbuf_q + 1/2 crdi_q + 1/2 cess_q z(Si,T) = 1/2 cats + 1/2 mcbuf_q + 1/2 crdi_q + 1/2 cess_q + 1 cfs_q + 1 cen_q + 1 di + 1 cfm + 1 jd + 1 kjd_q [O6] | formula suffix, enter "none" for no suffix. begin_van_laar_sizes alpha(di) 1.2 alpha(cen_q) 1 alpha(cfs_q ) 1 alpha(jd) 1.2 alpha(kjd_q) 1.2 alpha(cats) 1.9 alpha(mcbuf_q) 1.9 alpha(cess_q) 1.9 alpha(crdi_q) 1.9 alpha(cfm) 1 end_van_laar_sizes | solution model specific computational options: begin_flagged_endmembers | endmembers listed in this section are not associated with the | solution on output and are identified by endmember name. end_flagged_endmembers begin_dqf_corrections dqf(cen_q) = 3500 - 2 * T + 0.048 * P | cenjh dqf(cfs_q) = 2100 - 2 * T + 0.045 * P | cfsg dqf(cess_q) = -3460 | cesW dqf(mcbuf_q) = -20.89d3 -1.2 * T + 0.248 * P | cbuW dqf(crdi_q) = -4850 | crdW dqf(kjd_q) = 10.82d3 + 0.60 * P_bar | kjW end_dqf_corrections end_of_model -------------------------------------------------------- begin_model | Ti-Fe-Mg-F Clinohumite, with non-ideal F-OH interaction after Duffy & Greenwood, Am Min 64:1156-1174, 1979. | Ad-hoc Ti and Fe, assumes, somewhat improbably, that Fe and Mg are disordered across | MO and MB and that Ti is disordered on MB. this model is an implicit o/d model because | the tichu_q endmember is specified as an independent endmember. JADC/PB, 5/20. | exclude HP endmembers: chum and chdr | Published as Herviou et al. (2026), doi.org/10.1093/petrology/egag045 Site: MB MO OH ____________________________ Mutliplicity 1 8 2 endmember ____________________________ type _________ _________ chu Mg Mg OH independent fchu Mg Mg F independent tchu_q Mg TiMg OOH independent ftchu_q Mg TiMg OF dependent ichu_q Fe Fe OH independent fichu_q Fe Fe F dependent tichu_q Fe TiFe OOH independent ftichu_q Fe TiFe OF dependent This model requires data (or a make definition) for tichdr_q, my first guess would be to use whatever you use for tchdr_q + 4*(fa-fo) + (fta-ta). with any luck that will result in "realisitic" Fe-content. TiCh(JB) abbreviation TiChu full_name clinohumite 688 | model type: 688 format standard model 1 | number of polytopes 3 | number of simplices 2 2 2 | number of vertices on each simplex chu fchu tchu_q ftchu_d ichu_q fichu_d tichu_q ftichu_d x_F 0. 1 .1 0 x_H by difference x_Ti 0. 1 .1 0 x_M by difference x_Fe 0. 1 .1 0 | X(Fe), imod = 0 -> cartesian subdivision x_Mg by difference begin_dependent_endmembers ftchu_d = 1 tchu_q + 1/2 fchu - 1/2 chu fichu_d = 1 ichu_q + 1 fchu - 1 chu ftichu_d = 1 tichu_q + 1/2 fchu - 1/2 chu end_dependent_endmembers begin_excess_function W(ichu_q ichu_q fchu) -21572 W(chu tichu_q fchu) -21572 W(tichu_q tichu_q fchu) -5393 W(ichu_q fchu fchu) -59120 W(fchu fchu tichu_q) -29560 W(fchu chu ichu_q) -43144.2 W(fchu ichu_q tchu_q) -21572.1 W(fchu ichu_q tichu_q) -21572.1 W(fchu tchu_q tichu_q) -10786.05 W(chu chu fchu) -21572 W(chu tchu_q fchu) -21572 W(tchu_q tchu_q fchu) -5393 W(chu fchu fchu) -59120 W(fchu fchu tchu_q) -29560 end_excess_function 3 | number of sites in configurational entropy model MO | site name 2 8 8 | number of species, effective multiplicity, true multiplicity z(Mg,M) = 1 fchu + 1 chu + 1 tchu_q z(Fe,M) = 1 ichu_q + 1 tichu_q MB | site name 3 1 1 | number of species, effective multiplicity, true multiplicity z(Ti,M) = 1/2 tchu_q + 1/2 tichu_q z(Mg,M) = 1 fchu + 1 chu + 1/2 tchu_q z(Fe,M) = 1 ichu_q + 1/2 tichu_q OH | site name 3 2 2 | number of species, effective multiplicity, true multiplicity z(OH,OH) = 1 chu + 1/2 tchu_q + 1 ichu_q + 1/2 tichu_q z(F,OH) = 1 fchu z(O,OH) = 1/2 tchu_q + 1/2 tichu_q [Si4O16] | formula suffix, enter "none" for no suffix. begin_dqf_corrections DQF(ichu_q) = 0 DQF(tichu_q) = 0 DQF(tchu_q) = - 33581 + 32 * T_K | DQF(tchu_q) = 33581 -32*T_K | as for Ch-H2O-p4-2: - 33581 | why + 32 in the final version? JADC 5/26 end_dqf_corrections end_of_model -------------------------------------------------------------- begin_model | Ti-Fe-Mg-F Chondrodite, with non-ideal F-OH interaction after Duffy & Greenwood, Am Min 64:1156-1174, 1979. | Ad-hoc Ti and Fe, assumes, somewhat improbably, that Fe and Mg are disordered across | MO and MB and that Ti is disordered on MB. this model is an implicit o/d model because | the tichdr_q endmember is specified as an independent endmember. JADC/PB, 5/20. | exclude HP endmembers: chum and chdr | Published as Herviou et al. (2026), doi.org/10.1093/petrology/egag045 Site: MB MO OH ____________________________ Mutliplicity 1 8 2 endmember ____________________________ type _________ _________ chdrt Mg Mg OH independent fchdr Mg Mg F independent tchdr_q Mg TiMg OOH independent ftchdr_d Mg TiMg OF dependent ichdr_q Fe Fe OH independent fichdr_d Fe Fe F dependent tichdr_q Fe TiFe OOH independent ftichdrd Fe TiFe OF dependent This model requires data (or a make definition) for tichdr_q, my first guess would be to use whatever you use for tchdr_q + 4*(fa-fo) + (fta-ta). with any luck that will result in "realisitic" Fe-content. TiChdr(JB) abbreviation TiChr full_name chondrodite 688 | model type: 688 format standard model 1 | number of polytopes 3 | number of simplices 2 2 2 | number of vertices on each simplex chdrt fchdr tchdr_q ftchdr_d ichdr_q fichdr_d tichdr_q ftichdrd x_F 0. 1 .1 0 x_H by difference x_Ti 0. 1 .1 0 x_M by difference x_Fe 0. 1 .1 0 | X(Fe), imod = 0 -> cartesian subdivision x_Mg by difference begin_dependent_endmembers ftchdr_d = 1 tchdr_q + 1/2 fchdr - 1/2 chdrt fichdr_d = 1 ichdr_q + 1 fchdr - 1 chdrt ftichdrd = 1 tichdr_q + 1/2 fchdr - 1/2 chdrt end_dependent_endmembers begin_excess_function W(ichdr_q ichdr_q fchdr) -21572 W(chdrt tichdr_q fchdr) -21572 W(tichdr_q tichdr_q fchdr) -5393 W(ichdr_q fchdr fchdr) -59120 W(fchdr fchdr tichdr_q) -29560 W(fchdr chdrt ichdr_q) -43144.2 W(fchdr ichdr_q tchdr_q) -21572.1 W(fchdr ichdr_q tichdr_q) -21572.1 W(fchdr tchdr_q tichdr_q) -10786.05 W(chdrt chdrt fchdr) -21572 W(chdrt tchdr_q fchdr) -21572 W(tchdr_q tchdr_q fchdr) -5393 W(chdrt fchdr fchdr) -59120 W(fchdr fchdr tchdr_q) -29560 end_excess_function 3 | number of sites in configurational entropy model MO | site name 2 8 8 | number of species, effective multiplicity, true multiplicity z(Mg,M) = 1 fchdr + 1 chdrt + 1 tchdr_q z(Fe,M) = 1 ichdr_q + 1 tichdr_q MB | site name 3 1 1 | number of species, effective multiplicity, true multiplicity z(Ti,M) = 1/2 tchdr_q + 1/2 tichdr_q z(Mg,M) = 1 fchdr + 1 chdrt + 1/2 tchdr_q z(Fe,M) = 1 ichdr_q + 1/2 tichdr_q OH | site name 3 2 2 | number of species, effective multiplicity, true multiplicity z(OH,OH) = 1 chdrt + 1/2 tchdr_q + 1 ichdr_q + 1/2 tichdr_q z(F,OH) = 1 fchdr z(O,OH) = 1/2 tchdr_q + 1/2 tichdr_q [Si4O16] | formula suffix, enter "none" for no suffix. begin_dqf_corrections DQF(ichdr_q) = 0 DQF(tichdr_q) = 0 DQF(tchdr_q) = 0 - 5.76 * T_K end_dqf_corrections end_of_model