Page 1 of 1

Nucleation model in MICRESS

Posted: Wed Feb 10, 2021 10:49 am
by kamalnath
Dear all,

I am trying to understand the nucleation model implemented in MICRESS. I found a very old thread https://board.micress.de/viewtopic.php?f=10&t=10 discussing about it.

In this thread, it was mentioned that
In general, these models are designed for simulation at microscale, not at nano-scale, i.e., there is no model for prediction of homogeneous nucleation based on fluctuations the formation of a critical seed.
Was there any upgrades made recently to capture the nucleation process at nano-scale ? For example, to simulate the microstructure of Ni-Al binary alloy to gamma/gamma' microstructure. In superalloys, we can get extremely small precipitate sizes (gamma') sizes (20nm~50nm). I found this article where they used classical nucleation theory to implant the gamma' nuclei.
Simmons, J. P., Wen, Y., Shen, C., & Wang, Y. Z. (2004). Microstructural development involving nucleation and growth phenomena simulated with the phase field method. Materials Science and Engineering: A, 365(1-2), 136-143.
https://www.sciencedirect.com/science/a ... 07809#SEC6

In MICRESS, is there a way to use classical nucleation theory to implant the nucleus ?

Also, for solid-state precipitation processes, if I were to use Seed-undercooling or Seed-density model, how can I know the seed-density distribution or particle size ?

Re: Nucleation model in MICRESS

Posted: Thu Feb 11, 2021 5:09 pm
by Bernd
Dear kamalnath,

No, there has not been any new development of nucleation models on the atomistic scale, and we also do not have a specific homogeneous nucleation model in MICRESS.
Unfortunately, I currently have no access to the article which you linked. However, I assume that they use a different type of phase-field model which allows treating nucleation in NiAl as some sort of spinodal decomposition.
In MICRESS nucleation in such cases has to be treated in a phenomenological way, that means by calibration...

Bernd

Re: Nucleation model in MICRESS

Posted: Fri Apr 16, 2021 10:45 am
by CharMIC
Hi Bernd,

Is there a way to plot the number of nucleation events in MICRESS with time.

Re: Nucleation model in MICRESS

Posted: Fri Apr 16, 2021 11:09 am
by Bernd
Hi Chamara,

Nucleation events in MICRESS are displayed in the Screen/.log-output including (among other information) the consecutive number of the nucleation event and the time. So, using text processing, you could create a table of nucleation events and use that for plotting. However, numbering is over all seed types, and there is no way to see from the output whether nucleation was successful (i.e. the grain does not vanish shortly after nucleation).
Thus, in principle, it is possible to create a short script which creates such a plot and which also could distinguish different seed types or phases. We did not do that yet, and I did not hear about such a script, so you would have to build it yourself. There is no other output in MICRESS which contains data on nucleation events.
If you are interested in successful nucleation events only, then you simply could plot the .TabK file, which contains information on the number of existing grains over time. This of course would not work with "add_to_grain" which does not create new grains.

Bernd