grain boundary diffusion

solid-solid phase transformations, influence of stresses and strains
Post Reply
yzn12
Posts: 11
Joined: Wed Apr 09, 2014 8:33 am
anti_bot: 333

grain boundary diffusion

Post by yzn12 » Wed Apr 09, 2014 9:04 am

In the concentration section of the Vol2_manuals, the following words concerning grain boundary diffusion can be found on page 41:
A grain boundary diffusion model is available to consider a increase...
...by using the keyword "+b"
However, I can not find an example concerning grain boundary diffusion.
I failed after several times of trail and error.
So could anyone help me in solving this problem? Or just showing me how to do with it.
My version is MICRess 6.001
Thank you!

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

Re: grain boundary diffusion

Post by Bernd » Thu Apr 17, 2014 3:01 am

Hi yzn12,

I am sorry for not answering before, but I am on a trip to the US and for some technical reasons could not access the MICRESS forum for some time :oops:

Unfortunately, we do not have a standard example for grain boundary diffusion, and this option has not yet been used such often. But there is a thread in this forum about it which hopefully can help you further - please tell me if you already have read it or if it does not answer all your questions!

By the way, you post was exactly number 1000, congratulations!

Best wishes

Bernd

yzn12
Posts: 11
Joined: Wed Apr 09, 2014 8:33 am
anti_bot: 333

Re: grain boundary diffusion

Post by yzn12 » Wed Apr 23, 2014 8:17 am

Dear Bernd,

Thank you for your reply.

I have read it and tried it in MICRess. There is no problem with the code now. However, I do not quite understand how to deal with my simulation. The following is what I find on the thread:
# How shall diffusion of component 1 in phase 1 be solved?
1 1 multi gg
# How shall diffusion of component 2 in phase 1 be solved?
2 1 multi gg
# How shall diffusion of component 1 in phase 1 be solved?
1 1 +b nnbn
# Grain boundary diffusion settings for component
# 1 at boundary of phases 1 and 2:
Correction for activation energy? [J/mol]
1.8E3
Physical width of the interface? [nm]
2.5
...
end_diffusion_data
My simulation is about the process of pearlite cooperative growth. I want to check the partition of Alloying element in the moving interface, so grain boundary diffusion between ferrite and austenite(this can be done) and between cementite and austenite(this is not easy) should be considered. Since cementite is a stoichiometric phase, diffusion in cementite is not expected, does this necessarily mean that I can only consider Grain boundary diffusion at boundary of phases fcc and cem while not at boundary of phases cem and fcc?

Also, will the selection of the width of the interface greatly influence the simulation result? As the grid is rescaled, I think the width is not that important, 1-2nm is a good approximation, am I right?

Best wishes!

Yang Zenan

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

Re: grain boundary diffusion

Post by Bernd » Thu Apr 24, 2014 9:39 pm

Hi Yang,

in principle you are right. Grain boundary diffusion in MICRESS is treated as an increased flux inside one or both phases which form the interface. Considering a stoichiometric phase (with no solubility range for a given element) there are no concentration gradients in this phase, and thus it would not make any effect defining diffusion or grain boundary diffusion in this phase. Therefore, for a fcc-cementite interface, only the fcc "side" can be used to provide an increased diffusivity at the grain boundary, i.e. the "+b" keyword should be applied for fcc. But in general, as the local quasi-equilibrium condition at the interface links the diffusion fluxes in both phases, it does not matter which of them is used to implement the additional flux!
As the additional diffusion flux in the grain boundary region is rescaled to a physical interface thickness, there should be no effect on the integral diffusion fluxes when the numerical interface thickness or resolution is changed. This integral flux is determined by the correction of the activation energy and the physical interface thickness.
Are there other reasons why you think considering grain boundary diffusion in fcc-cementite interfaces could be difficult?

Bernd

yzn12
Posts: 11
Joined: Wed Apr 09, 2014 8:33 am
anti_bot: 333

Re: grain boundary diffusion

Post by yzn12 » Sun Apr 27, 2014 1:23 pm

Dear Bernd:
Thank you for your suggestion. I will deal with the simulation and if there is any further problem, I would post here. It is very kind of you~ :D
Yang Zenan

yzn12
Posts: 11
Joined: Wed Apr 09, 2014 8:33 am
anti_bot: 333

Re: grain boundary diffusion

Post by yzn12 » Fri Nov 28, 2014 12:11 pm

Hi Bernd,

I have another question concerning boundary diffusion.
How do you set the prefactor of diffusion coeffcient?
The physical width of interface δ= 0.5 nm or 1nm
However, according to a handbook, the boundary diffusivity is

Dδ= 5.4x10^-8 exp(-Q/RT) cm2/s

Do I need to input the prefactor (5.4x10^-8 divided by δ in the
DRV file.
How is the prefactor considered in MICRess?

Best wishes!
Yang

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

Re: grain boundary diffusion

Post by Bernd » Fri Nov 28, 2014 5:05 pm

Dear Yang,

If I understand correctly, the extra diffusion due to the grain boundaries is given in the handbook by frequency factor and temperature dependency. But this specification is not compatible to the way we are treating grain boundary diffusion in MICRESS: We cannot specify boundary diffusion independent from bulk diffusion, but we define it as an increased bulk diffusion! The advantage is that we need only one extra parameter which is the change of the activation energy!
This means that if you want to implement the value from the handbook, you can do it correctly only for one temperature! I would try to plot the T-dependent values from the handbook on one side, and the difference of the bulk term and the bulk term with modified activation energy on the other side in order to find the activation energy modification which best fits in the relevant temperature region.

Bernd

yzn12
Posts: 11
Joined: Wed Apr 09, 2014 8:33 am
anti_bot: 333

Re: grain boundary diffusion

Post by yzn12 » Tue Dec 02, 2014 2:19 pm

Dear Bernd:

Thanks a lot!
I will carefully consider your advice!
BTW, could you please provide more information about the MICRess conference the next year?

Best wishes!
Yang

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

Re: grain boundary diffusion

Post by Bernd » Tue Dec 02, 2014 4:05 pm

Hi Yang,

What do you mean with MICRESS conference? You mean our up-coming MICRESS training course here in Aachen?

If yes, please see the attached official announcement!

Bernd
Attachments
20150303_basic_training_schedule.pdf
(77.28 KiB) Downloaded 333 times

Post Reply