Hi Shawn,
To answer your question I need to first provide a little background. There are multiple software packages, both open-source and proprietary, that implement the CALPHAD method. Some of these packages use different input file formats for the thermodynamic models,
but the most common is the TDB file format. The compatibility of a particular TDB between Thermo-Calc, Pandat, OpenCalphad, and pycalphad is nearly seamless, in most cases.
For historical reasons, the majority of the work that has been published using the TDB file format is in support of structural alloys, e.g., steel, nickel, aluminum. There is a CALPHAD-based commercial software package called FactSage that (again, for historical
reasons) has collected a lot of work in molten salt mixtures. Unfortunately, pycalphad does not currently have the ability to read the FactSage database file format.
Lucky for you, there is a little bit of work in the open literature on CALPHAD and salts! Here is a 2014 paper on modeling of KCl - NaCl - ZnCl2 ternary system: https://doi.org/10.1016/j.calphad.2014.04.004 -
The TDB is linked at the very bottom, under Supplementary Material. (You don't need a journal subscription to download the TDB.) I haven't tested it in pycalphad myself, but I gave a quick glance at the file and everything in there looks supported by the latest
pycalphad (0.7).
For MgCl2-CaCl2, I only see FactSage publications, and no TDBs.
The best central place to find TDBs is the TDBDB, which is maintained by Axel van de Walle at Brown University: https://avdwgroup.engin.brown.edu/ -
you type in the elements you want, and it searches its index of the CALPHAD literature. In general, for reasons stated above, you're going to find a lot less on salts compared to alloys.
Regards,
Richard Otis