MICRESS Forum

Hi,

I am trying to quench from a fully austenitic microstructure to a microstructure that contains both martensite (which I am modeling as acicular ferrite, taking some of the values/structure from here download/file.php?id=480) and austenite.

However, despite changing the undercooling, cooling rate, grid sizes ect, the austenite disappears completely towards the end of the simulation. Could you please tell me why that could be? I have included my driving file for reference

Thank you.

Hi,

Why do you expect the transformation to stop before austenite vanishes completely? There may be 2 reasons why this could happen: Either the transformation gets "frozen", i.e. it stops for kinetic reasons, or thermodynamics do not lead to complete transformation.

If you assume a "frozen" transformation, the interface mobility must decrease with temperature. I propose to use diffusion-limited kinetics with respect to carbon diffusion, while using nple or para-equilibrium for the substitutional elements (like it is done for the application example A004). Then, as diffusion of carbon gets slower with lower temperature, the numerical interface mobility automatically is decreased to maintain diffusion limited growth. Even if you do not assume a "frozen" transformation, I would recommend to use these assumptions because otherwise the transformation kinetics is not well-defined.

If you assume the transformation should stop thermodynamically, you could try to increase the carbon content of the dG-offset of the 1/2-interface.
Another option would be to include stress coupling. Stress accumulation in the remaining austenite would probably stop the transformation at some point (which I guess in reality is the main reason why the martensite transformation is not complete).

Bernd