5
$\begingroup$

I primarily use Thermo-Calc so I will use that to describe my problem but I am interested to hear general CALPHAD insight as well.

Using Thermo-Calc I can quickly get an idea of the phase regions present (isothermally) by running a phase diagram calculation for a ternary. What appeals to me most about this is how fast the calculation is (as opposed to having to run ~5000 single eq calcs to get an idea of the phase regions). I have a vague understanding that behind the scenes Thermo-Calc is zeroing on each of the phase boundaries and so only has to do a limited number of calculations. In Thermo-Calc and TC-Python it only seems possible to do this for binary and ternary systems (e.g., using "with_phase_diagram_calculation()" in TC-Python it seems to only let me specify "with_first_axis" and "with_second_axis"). I guess they don't have this implemented because you can't easily visualize quaternaries, etc., but I am interested in still extracting the data relating to the phase regions/boundaries. So, my question is how can I extract this phase data using TC-Python for an arbitrary system of elements?

A secondary question I would be interested to hear an answer to is how exactly Thermo-Calc (or other CALPHAD implementations) calculate the phase region/boundary information so quickly.

Improve this question
$\endgroup$
1
  • $\begingroup$ You could attempt fit analytical models to the free energy curves and then solve for an exact solution $\endgroup$
    – STP
    Commented Aug 21, 2023 at 21:22

1 Answer 1

Reset to default
3
+25
$\begingroup$

I assume someone at TCSAB (which market Thermo-Calc) can answer this better but as I participated in developing TC a long time ago and now I work on a similar free software called OpenCalphad (I assume you use the version of TC which is free but limited to 3 elements). You can use OC for as many elements as you want PROVIDED you have a database with those elements. A good database is expensive because it takes a lot of efforts and many months to collect all the experimental data and fit these data to models even for a binary system. I worked with the Al-Fe system for more than a year before it could be published and a 10 component system has 45 binaries ... With OC there is also a python interface if you prefer that, you can find links to OpenCalphad easily on the web. But the limitation of the Calphad method is not the software, it is developing the databases which requires a lot of efforts. You may find some free databases of varying quality and accuracy on the web also but you are most welcome to participate in the development if you find them useful.

Improve this answer
$\endgroup$
2

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .