Austenite to Pearlite Transformation

solid-solid phase transformations, influence of stresses and strains
SuperX
Posts: 15
Joined: Thu Apr 20, 2017 1:32 pm
anti_bot: 333

Austenite to Pearlite Transformation

Post by SuperX » Thu Jun 08, 2017 4:45 pm

Hello Bernd:

Thank you very much for you kindly help! After the solidification I need to simulate the solid state transformation. It is an eutectiod reaction, from austenite to pearlite.
According to the information I get, the transformation process may be like the picture. I really do not know how to set the parameters in order to let the grains grow in that way, but I think maybe you know a reasonable model to describe pearlite
formation of pearlite.PNG
formation of pearlite.PNG (925.89 KiB) Viewed 6036 times
There is a standard driving file in micress examples, but the grain shape in that file, like the picture shows, is circular rather than acicular.
dfgdf.PNG
dfgdf.PNG (27.8 KiB) Viewed 6036 times
I will be very glad if you can give me some guide.

SuperX

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

Re: Austenite to Pearlite Transformation

Post by Bernd » Thu Jun 08, 2017 6:04 pm

Dear SuperX,

I don't really understand what you mean by "acicular". What you show in the lower image is the gamma/alpha microstructure in the standard GammaAlphaPearlite_TQ_dri example before pearlite formation.

Do you mean the alpha phase should already be acicular? Then, Gamma_Alpha_FeC_Acicular_dri would be an example which shows how to do that. Or do you mean the cementite lamellae? Then you should go to a finer scale and simulate pearlite microstucture as a resolved alpha/cementite eutectoid.

The different approaches how to simulate pearlite have been discussed in this thread.

Bernd

PS: I shifted your question to a new topic in the corresponding board and with a meaningful subject.

SuperX
Posts: 15
Joined: Thu Apr 20, 2017 1:32 pm
anti_bot: 333

Re: Austenite to Pearlite Transformation

Post by SuperX » Fri Jun 09, 2017 1:51 pm

Dear Bernd:
Firstly thank you for your help!
Sorry I did't express clearly. What I want to simulate is the cementite lamellae. As you told, I refered to the standard "gamma_alpha_pearlite TQ_dri" file. Like the standard file, I set austenite as substrate phase of ferrite and set ferrite as substrate phase of cementite. Both are interface nucleation. But in the simulation result, nucleation of ferrite appears at the austenite-graphite interface rather than interface of austenite grains.
sdfsdf.PNG
sdfsdf.PNG (28.68 KiB) Viewed 6024 times
(colour yellow represents graphite, red represents austenite and white represent ferrite)
Also a warning about demixing of both C and Si in interface of ferrite and austenite occurs...
I want to use the "unresolved model" to describe the pearlite phase. If I want to get a cementite lamellae, I think the nucleation orientation as well as the anisotrophy of phase interaction must be the important points, am I right? Can you please give me some guide about the orientation setting?

SuperX

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

Re: Austenite to Pearlite Transformation

Post by Bernd » Fri Jun 09, 2017 2:59 pm

Dear SuperX,

I am still confused whether you want to simulate cementite lamellae, or whether you want to simulate pearlite as an effective phase ("unresolved"). In the latter case, the shape of pearlite (phase mixture of ferrite and cementite) has nothing to do with the anisotropy of cementite...

Bernd

SuperX
Posts: 15
Joined: Thu Apr 20, 2017 1:32 pm
anti_bot: 333

Re: Austenite to Pearlite Transformation

Post by SuperX » Fri Jun 09, 2017 3:52 pm

Dear Bernd:

Sorry I wasn't so clear about this issue. Now I begin to understand: The "unsolved" model treat pearlite as one single phase, but the cementite lamellae need to simulate both cementite and ferrite separately. If the "unsolved" model is adopted, we can not see the lamellar structure of pearlite. Is that right?
In that case I think it would be easier to adopt the "unsolved" model. Since the manual only mentions this model only in one sentence, I do not have much information about the setting.
Thankyou for your help.

SuperX

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

Re: Austenite to Pearlite Transformation

Post by Bernd » Fri Jun 09, 2017 6:05 pm

Dear SuperX,

Apart from the manual, there is some basic explanation as well as some discussion about the effective phase model ("unresolved") here.

Bernd

SuperX
Posts: 15
Joined: Thu Apr 20, 2017 1:32 pm
anti_bot: 333

Re: Austenite to Pearlite Transformation

Post by SuperX » Tue Jun 13, 2017 10:29 am

Hello Bernd:

Thank you for your help. I searched the discussion before about the effective phase model ("unresolved") and then set the solid-solid transformation from austenite to pearlite as follows:(phase 3 is ferrite, phase 1 is austenite and phase 4 is cementite)

# Input for seed type 3 (Ferrite):
# ----------------------
# Type of 'position' of the seeds?
# Options: bulk region interface triple quadruple [restrictive]
interface
# Phase of new grains (integer) [unresolved|add_to_grain]?
3
# Reference phase (integer) [min. and max. fraction (real)]?
1
# Substrate phase [2nd phase in interface]?
# (set to 1 to disable the effect of substrate curvature)
4

# Input for seed type 4 (Cementite):
# ----------------------
# Type of 'position' of the seeds?
# Options: bulk region interface triple quadruple [restrictive]
interface
# Phase of new grains (integer) [unresolved|add_to_grain]?
4 unresolved
# Microstructure size (radius) [micrometers]?
0.1
# Reference phase (integer) [min. and max. fraction (real)]?
1
# Substrate phase [2nd phase in interface]?
# (set to 1 to disable the effect of substrate curvature)
3

I set cementite as substrate phase of ferrite nucleation and ferrite as substrate phase of cementite nucleation.I want to ask if the setting is reasonable.
In my simulation no change takes place to austenite under such a setting. I hope you can give me some guide.

SuperX

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

Re: Austenite to Pearlite Transformation

Post by Bernd » Tue Jun 13, 2017 12:37 pm

Dear SuperX,

the way you set it up, you have created a "chicken or the egg" dilemma: Cementite can nucleate only on ferrite, and ferrite can nucleate only on cementite. If you have none of both, nothing will happen...

Depending on the composition of your alloy, one of the two phases will come first on cooling. This phase (typically ferrite) should nucleate on austenite without the necessity that cementite is present. Afterwards, cementite grains can nucleate on ferrite, forming a pair with the locally existing ferrite grain, giving foundation for a "pearlite grain".
Nucleation of a further ferrite grain on such a pearlitic cementite would not constitute a new pearlite grain, it would be a "normal" type of nucleation, and the already "pearlitic" cementite grain is not allowed to grow into other ferrite grains than its partner grain which it got in the moment of nucleation.

Thus (if you only have the two nucleation types you showed) changing the substrate of ferrite nucleation from cementite to austenite should do the job.

Bernd

SuperX
Posts: 15
Joined: Thu Apr 20, 2017 1:32 pm
anti_bot: 333

Re: Austenite to Pearlite Transformation

Post by SuperX » Wed Jun 14, 2017 3:19 pm

Hello Bernd:

Thank you for your help. A running problem occured in my simulation of the transformation from austenite to pearlite. It is a problem of "trying hard phases 3 1 level: 3 zp= 12362 errror=3". (3 is ferrite and 1 is austenite)

I know that is really a frequently-occured problem. I have tried lots of methods that is mentioned in posts before, like stoichio setting, using the "interaction" keyword...I also changed the the interfacial energie, kinetic coefficient, but nothing helps. That error just keeps occuring in the whole simulation process, like the picture shows.
fgdfg.PNG
fgdfg.PNG (24.56 KiB) Viewed 6000 times
In my simulation all of austenite transforms into pearlite and the effective phase model is to be used. I really have spent lots of time changing various parameters but made no progress. Information I know about this error is quite limited. I think I am not able to solve this problem on my own.Can you please give me some help?

SuperX
Attachments
kuai.txt
(37.53 KiB) Downloaded 322 times

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

Re: Austenite to Pearlite Transformation

Post by Bernd » Wed Jun 14, 2017 6:13 pm

Dear SuperX,

as we already talked personally, this type of error seems not to be a very bad one, if it does not occur too often. Without digging deeper into the problem I cannot tell where it may come from.
If you send the .ges5 file which is also needed for running the simulation, I could have a closer look.

Bernd

Post Reply