MQMQA support: B-type excess mixing parameters

[bocklund]

MQMQA models in the DAT format can have mixing code Q, G, or B that correspond to different excess equations. Right now Q and G are implemented in ModelMQMQA.excess_energy, but mixing code B is not yet supported.

[maxposchmann]

Note that R ('reciprocal') is also a possible mixing code.

[bocklund]

Note that R ('reciprocal') is also a possible mixing code.

Do you know the mathematical form of B and R parameters or a reference where I can find them? Our current implementation was heavily influenced by your recent paper and the friendly notation within 😄

[maxposchmann]

R specifies two constituents from each sublattice (i.e. two cations A,B and two anions X,Y). I've only ever seen it all 0 exponents, so Δg_ab/xy = L_AB/XY X_AB/XY. Probably the exponents act on the constituent fractions (i.e X_A), but as I said I've never seen them used. Not surprising, it's hard enough to find and fit data for reciprocal systems.

[maxposchmann]

B is Bragg-Williams mixing. Pelton's textbook is probably the best reference for you.

While looking at Thermochimica to see what pointers I could give, I was reminded that H is also a possible code, and also refers to Bragg-Williams mixing.

From FactSage, here's a B term:

and here's an H term:

Take these with a grain of salt, as the solution module that displays these often glosses over important details, and is occasionally outright incorrect.