Austenite Formation from Ferrite using MICRESS Examle

solid-solid phase transformations, influence of stresses and strains
Post Reply
billyzhangubc
Posts: 25
Joined: Fri Feb 01, 2019 8:53 pm
anti_bot: 333
Location: University of British Columbia

Austenite Formation from Ferrite using MICRESS Examle

Post by billyzhangubc » Wed Apr 03, 2019 6:12 pm

Hi all.
I'm trying to simulate austenite formation from ferrite during intercritical annealing and I started by manipulating one of the MICRESS standard examples.
I have a few confusion regarding how to edit the example drive file to reverse the transformation:
(1) How are the phase assigned? i.e. is phase 1 always ferrite and 2 always austenite, or is phase 1 always the parent phase and phase 2 the daughter phase?
(2) Can I simply exchange the phase interaction data of phase 1 and phase 2 in your example file? I think that includes changing data for phase interaction (interfacial energy and mobility) and diffusion data, as well as entropy of transformation. I wonder if I missed anything.
(3) My steel has a C content of 0.035 wt% which is above the solubility of C in ferrite. I wonder how I should input my linearized phase diagram. One of the lines should be the γ/α+γ line but I'm not sure about the other one. Should I use the α+γ/α+Fe3C line or should I use the γ/α+γ line? I guess if I use the former, I will have to extend the line and calculate the temperature difference to the α+γ/α+Fe3C line as the critical "undercooling" or minimum nucleation temperature.

I think my simulation is reasonable by examining the concentration profile of C (austenite has higher C content, see attached figure).
trial_4_conc1.png
trial_4_conc1.png (384.6 KiB) Viewed 2096 times
I attached a working example (modified drive file) in 2D. Component 1 is C and 2 is Mn.

Thanks in advance,
Billy
Attachments
trial_4.in
(18.74 KiB) Downloaded 238 times

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Austenite Formation from Ferrite using MICRESS Examle

Post by Bernd » Thu Apr 04, 2019 12:24 pm

Hi Billy,

I think the simplest way to change a simulation setup from γ->α to α->γ is to keep the phase identity, and instead just to alter the phase of the initial grain structure and the nucleation data. Then, all the rest can remain as it is.
If you change the phase identity, all phase-related parameter need to be changed, of course. These are the "phase data", diffusion data and the phase diagram data. In the phase diagram data, the c0 and slopes need to be interchanged, and dS changes sign. In case of database coupling, you would just change the phase relation with the database, and the diffusion and phase diagram data would change automatically.
However, I don't understand your question with Fe3C. Do you have this phase and want to include it as pearlite, e.g. in form of a pseudo-phase? Otherwise, you should take the lines and prolongations which correspond to γ and α only.

Bernd

Post Reply