Is it possible to introduce new mineral phases and solution models into the datafiles? And if so, what files and parameters will I have to change in order to get Perplex to include these in its calculations? (I assume that just adding them to an existing datafile is not enough). At the moment, I'm mainly interested in including a solution model within the chemical components defined in Perplex, but I've also been thinking about trying to include Zn as a component in staurolite. I guess adding a chemical component is a lot more work, or can that also be done relatively easily in Perplex? --------------------------------------------------------------------------------------------- you need to modify two files, the thermodynamic data file (e.g., hp02ver.dat) and the solution model file (e.g., solution_model.dat). refer to the www.perplex.ethz.ch/perplex_thermodynamic_data_file.html page for information on formatting data for the thermodynamic data file (this can also be done automatically by the FRENDLY program). the solution model modifications are also relatively straightforward, unfortunately the documentation is mostly in the file so you have to work through the file to figure out how to do things, but taking your staurolite as an example, the current model is ( the St model in newest_format_solut.dat is not well documented, so it's probably better to look at other models to figure out what all the numbers mean): begin_model | keyword indicating beginning of a solution model Mn-Fe-Mg St St(HP) 2 model type: Ideal or Margules 1 non-reciprocal solution 3 3 endmembers mnst fst mst 1 0 0 | endmember flags | Note restricted range on X(Mn) 0. .2 0.01 | range and resolution for X(Mn) 0. 1. 0.02 | range and resolution for X(Fe) 0 | subdivision scheme : imod = 0 -> cartesian begin_excess_function w(mst fst) -8d3 0. 0. end_excess_function 1 1 site entropy model 3 4. 3 species, site multiplicity of 4 z(Fe) = 0 + 1 fst z(Mg) = 0 + 1 mst end_of_model | end of model keyword assuming you name the zn endmember (which you would have to add to the thermodynamic data base) of staurolite zst, then the new model would be: begin_model | keyword indicating beginning of a solution model Zn-Mn-Fe-Mg St St(HP) 2 model type: Ideal or Margules 1 non-reciprocal solution 4 4 endmembers zst mnst fst mst 1 1 0 0 | endmember flags 0. .2 0.01 | range and resolution for X(Zn) 0. .2 0.01 | range and resolution for X(Mn) 0. 1. 0.02 | range and resolution for X(Fe) 0 | subdivision scheme : imod = 0 -> cartesian begin_excess_function w(mst fst) -8d3 0. 0. end_excess_function 1 1 site entropy model 4 4. 4 species, site multiplicity of 4 z(Fe) = 0 + 1 fst z(Mg) = 0 + 1 mst z(Zn) = 0 + 1 zst end_of_model | end of model keyword