Is it possible to simulate both recrystallization and precipitation process at the same time?

ripening phenomena, dislocations, grainboundary topology
Junzhen
Posts: 11
Joined: Mon Jun 03, 2019 9:50 pm
anti_bot: 333

Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Junzhen » Fri Jun 07, 2019 1:39 pm

Hi, I am new here.
I am trying to simulate the microstructure evolution during hot-rolling process. And it is possible for my steel to recrystallize and precipate at the same rolling process. I wonder if it is possible to simulate both recrystallization and precipitation at the same time?

Any ideas and comments are welcome.

Bernd
Posts: 1148
Joined: Mon Jun 23, 2008 9:29 pm

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Bernd » Fri Jun 07, 2019 10:15 pm

Dear Junzhen,

Welcome to the MICRESS Forum!

Yes, recrystallisation can be combined with precipitation or other types of microstructure simulation in MICRESS, although this is not the typical application reflected in the MICRESS examples. The recrystallisation model just adds an extra driving force for grain or phase boundaries. The important question would be whether recrystallisation and precipitation happen at the same time scale in your application.

Bernd

Junzhen
Posts: 11
Joined: Mon Jun 03, 2019 9:50 pm
anti_bot: 333

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Junzhen » Tue Jun 11, 2019 12:31 pm

Hi,Bernd

Thanks for the response. That is very helpful.
I have another question. In the recrystallisation simulation, is the energy shreshold for recrystallisation model equal to the stored energy?
I know literally that the driving force control the growth of grain. However,except for temperature, which parameters control the growth in Micress?
Thank you for your time

Junzhen

Bernd
Posts: 1148
Joined: Mon Jun 23, 2008 9:29 pm

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Bernd » Tue Jun 11, 2019 10:19 pm

Dear Junzhen,

Yes, the driving force controls the movement of interfaces, and thus the growth of grains. Depending on the type of coupling, there are different contributions which sum up as driving force and can be monitored (directly or indirectly) with the .driv output file in DP_MICRESS.

In case of pure phase-field coupling ("phase"), e.g. for grain growth in a pure substance, curvature is the only source of driving force. Depending on the interface mobility, this leads to a slower or faster growth of bigger grains at the expense of smaller ones (grain growth).

If phase-field is coupled to a temperature field ("temperature"), there is a second source of driving force ΔG=ΔS (T-T°) which adds up to the curvature.

If phase-field is coupled to concentration ("concentration"), then the chemical contribution ΔG=ΔS [(T-T°) +Σmk(ck-c°k] is added to the curvature contribution. T is treated here as boundary condition and not as field variable.

Furthermore, an elastic driving force can be added in case of stress coupling ("stress_coupled") as well as a driving force for recrystallisation if this model is activated for a given phase. In the latter case, the extra contribution is ΔG=E1-E2, i.e. the difference in stored energy (recrystallisation energy). However, for recrystallisation we mostly assume that only new grains without stored energy are allowed to grow, i.e. those with E1=0. More precisely, this is described by the energy threshold for recrystallisation which is typically chosen as a very small value. This is the input parameter which you referred to.

Bernd

Junzhen
Posts: 11
Joined: Mon Jun 03, 2019 9:50 pm
anti_bot: 333

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Junzhen » Wed Jun 12, 2019 4:42 pm

Hi, Bernd, thank you so much for the reply. I learned much from your reply.
When I am trying to simulate recrystallisation, minimum and maximum of recrystallisation energy are required. When the values of them are smaller than energy threshold of recrystallisation energy (stored energy), no recrystallisation takes place. so my question is what minimum and maximum of recrystallisation energy mean? And how can I get the parameters?

Thank you

Junzhen

Bernd
Posts: 1148
Joined: Mon Jun 23, 2008 9:29 pm

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Bernd » Wed Jun 12, 2019 5:15 pm

Hi Junzhen,

If a polycrystalline material has been deformed, typically the stored deformation energy will somehow depend on the grain orientation, i.e. it will not be constant. The simplest assumption is that it is constant per grain and there is a random distribution over the grains. This is what you input then as random in Grain Input by using minimum and maximum of recrystallisation energy. Typically, the threshold will be close to 0, so the whole range of deformation energies is above the threshold value.
You can get the deformation energy either from experiments (measurement of dislocation densities), from simulations of dislocation dynamics, or just by guessing. Starting from a guess, you also can calibrate the values by comparison between simulation and experiment. However, as the interface mobility is also typically unknown, you probably can only calibrate them together...

Bernd

Junzhen
Posts: 11
Joined: Mon Jun 03, 2019 9:50 pm
anti_bot: 333

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Junzhen » Thu Jun 13, 2019 2:34 pm

Hi, Bernd,
thank you for your timely reply.
I have another question about precipitation caculation.
My steel contains more than ten Elements, like Fe, C, Mn, si etc. However, only two or three elements are involved in precipatation.

Should I input all of the elements in my steel when i am trying to simulate precipitation? or just input Fe as well as the elements involed in precipatation?
Thank you.

Junzhen

Junzhen
Posts: 11
Joined: Mon Jun 03, 2019 9:50 pm
anti_bot: 333

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Junzhen » Thu Jun 13, 2019 5:56 pm

Hi, Bernd,
As said in the Vol3 manual, we can get the .vtk file by the post-peocessing using DP_Micress.
Then I tried to use the .vtk file I got through DP_Micress as the Grain Input. But it didn't work.
Could you explain it a little bit ?
Thank you very much.

Junzhen

Bernd
Posts: 1148
Joined: Mon Jun 23, 2008 9:29 pm

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Bernd » Mon Jun 17, 2019 5:02 pm

Hi Junzhen,

Sorry for the late reply! Let me first answer the first question:

If you have a complex steel, the question is always how many and which elements should I use in MICRESS. Generally, those elements which have a very low content and which are not taking part in precipitation should be omitted. If you are unsure, it makes sense to use a Thermo-Calc equilibrium calculation (or Scheil in a solidification case) to see whether the corresponding element has a considerable influence on the thermodynamic properties (equilibrium temperatures and compositions) or not. Of course, you can only include elements which are part to the thermodynamic database you have. Practically, it is viable to add up to 10-12 elements. However, simulation time is increased considerably, and there is a high risk to introduce numerical difficulties. Please note that introducing elements having low solubility in the matrix phase can easily cause problems if the corresponding intermetallic phases are not considered. E.g., a solidification simulation where carbide formation is intentionally suppressed, should never include the element carbon!

Your second question cannot be answered in general. It would be interesting to know what was your intention to do. Of course, it is possible to read in the .korn output of MICRESS as initial grain structure in ASCII or vtk format, which you can create using DP_MICRESS. But without further information I cannot tell you what might have been going wrong. Which error did you get? Can you show your input file?
However, it may also be possible that (depending on what you want to achieve) using the restart (.rest) file instead of .korn would be the better option...

Bernd

Junzhen
Posts: 11
Joined: Mon Jun 03, 2019 9:50 pm
anti_bot: 333

Re: Is it possible to simulate both recrystallization and precipitation process at the same time?

Post by Junzhen » Tue Jun 18, 2019 2:37 pm

Hi, Bernd,
Thanks for your Reply.
I generated 1_87.vtk file from a prior simulation using DP_MICRESS. And I changed the Header according to examples. Then I tried to Input it as a initial microstructure of the second Simulation. It aborted with an error in the input stage.
I asked one of my colleagues, and we found that Micress failed to derermine the correct number of grains in the .vtk file generated from a prior simulation using DP_MICRESS and asked for Input data of nonexistent grains.
Then it aborted with an error in the input stage.

We rearrange the IDs of remained grains, then it works.

PS:My colleague has reported this problem to Ralph.

Post Reply