Surface Oxidation

technical aspects of .dri file generation (e.g. debug mode ) etc...
Post Reply
SteelOx
Posts: 11
Joined: Mon Dec 07, 2020 10:17 am
anti_bot: 333

Surface Oxidation

Post by SteelOx » Mon Jan 11, 2021 11:46 am

Hi all,

I am looking to model the surface oxides on the surface of steel and had a few questions about adapting a training example.'

I am currently using Example T009 Cast Iron Dendrite nodules and it says adding additional elements it relatively easy, how do i add more elements and change the chemical compositions for this steel/ cast iron.

How do i change the seed points from one in the bottom left corner to a random number across the surface.

And finally how to i model the oxidation growth from these seed nucleation points.

Any help would be greatly appreciated.

Regards SteelOX :D

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

Re: Surface Oxidation

Post by Bernd » Mon Jan 11, 2021 12:08 pm

Hi SteelOX:

My first question in this regard is: Why do you use the example T009_CastIronDendriteNodules_dri as starting point? This example is quite specific for cast iron with formation of graphite nodules and uses the volume_change option because graphite is creating new volume due to the interstitial diffusion mechanism of carbon in steels.

For finding a suitable example to start from, one would need to know more about the process conditions (isothermal, T-gradient, cooling rate, relevance of latent heat), the geometry of the setup (planar surface, dendrite, spherical particles, ...), and the type of alloy (which elements are involved and which phases are expected).

In principle, one can start from any example, however the better it fits to the simulation you want to do, the less is the effort of adapting it.

Bernd

SteelOx
Posts: 11
Joined: Mon Dec 07, 2020 10:17 am
anti_bot: 333

Re: Surface Oxidation

Post by SteelOx » Mon Jan 11, 2021 12:16 pm

Bernd wrote:
Mon Jan 11, 2021 12:08 pm
Hi SteelOX:

My first question in this regard is: Why do you use the example T009_CastIronDendriteNodules_dri as starting point? This example is quite specific for cast iron with formation of graphite nodules and uses the volume_change option because graphite is creating new volume due to the interstitial diffusion mechanism of carbon in steels.

For finding a suitable example to start from, one would need to know more about the process conditions (isothermal, T-gradient, cooling rate, relevance of latent heat), the geometry of the setup (planar surface, dendrite, spherical particles, ...), and the type of alloy (which elements are involved and which phases are expected).

In principle, one can start from any example, however the better it fits to the simulation you want to do, the less is the effort of adapting it.

Bernd
Perfect, i only starting with that example as my supervisor has mentioned Dendrite growth, but as you mention it spherical oxidation particles would be more accurate. My samples are mainly FeSiB, and we expects oxides for all of these elements, which from Thermocalc are Halite, Quartz and BBC_A2/FCC_A1. Cooling rates and heating rates aren't largely important but the isothermal temperature is between 700-800C .

Thank you for the help :)

Regards SteelOX

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

Re: Surface Oxidation

Post by Bernd » Mon Jan 11, 2021 1:00 pm

Hi SteelOX,

ok, so essentially you could start with the quaternary system Fe-Si-B-O. Probably you want to simulate the spherical particles in contact to some atmosphere which contains oxygen, right? Then you need the phases GAS, FCC_A1, BCC_A2, HALITE, and QUARTZ. Which database do you want to use? Probably you can the TCFE database.

With this information you can create a corresponding .GES file. It needs to contain exactly these for elements, and at least the five phases which I propose. The simplest way of creating such a .GES5 file is by editing the .TCM file which corresponds to the .GES file of the MICRESS example you start with. You will need to change the name of the database (if you want to use a different one), the list of elements and the list of phases. There is probably only diffusion data available for fcc and bcc, so you need to list only these two phases in the second part where the mobility database is appended.

There is no MICRESS example which fits directly to your setup. I would recommend using T011_Gamma_Alpha_TQ_dri because it is already a steel, has no temperature gradient or latent heat, and because you can simply adapt the initial microstructure by changing the "voronoi" to "no_voronoi" and adjust the grain radii for getting circular initial particles. In this example, phase 0 is unused and should be used for the GAS phase. Nevertheless, you will need to adjust the number of elements and add more phases, which in practice needs some trial and error.

From this point on, you can adjust further parameters like initial temperature, gas pressure, nucleation conditions for the oxide phases, etc.

Bernd

SteelOx
Posts: 11
Joined: Mon Dec 07, 2020 10:17 am
anti_bot: 333

Re: Surface Oxidation

Post by SteelOx » Mon Jan 11, 2021 1:28 pm

Bernd wrote:
Mon Jan 11, 2021 1:00 pm
Hi SteelOX,

ok, so essentially you could start with the quaternary system Fe-Si-B-O. Probably you want to simulate the spherical particles in contact to some atmosphere which contains oxygen, right? Then you need the phases GAS, FCC_A1, BCC_A2, HALITE, and QUARTZ. Which database do you want to use? Probably you can the TCFE database.

With this information you can create a corresponding .GES file. It needs to contain exactly these for elements, and at least the five phases which I propose. The simplest way of creating such a .GES5 file is by editing the .TCM file which corresponds to the .GES file of the MICRESS example you start with. You will need to change the name of the database (if you want to use a different one), the list of elements and the list of phases. There is probably only diffusion data available for fcc and bcc, so you need to list only these two phases in the second part where the mobility database is appended.

There is no MICRESS example which fits directly to your setup. I would recommend using T011_Gamma_Alpha_TQ_dri because it is already a steel, has no temperature gradient or latent heat, and because you can simply adapt the initial microstructure by changing the "voronoi" to "no_voronoi" and adjust the grain radii for getting circular initial particles. In this example, phase 0 is unused and should be used for the GAS phase. Nevertheless, you will need to adjust the number of elements and add more phases, which in practice needs some trial and error.

From this point on, you can adjust further parameters like initial temperature, gas pressure, nucleation conditions for the oxide phases, etc.

Bernd
Thank you so much.

Where should i start? Im trying to edit a GES5 file but cant manage it. And how to i chose a database for my simulation

Regards David

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

Re: Surface Oxidation

Post by Bernd » Mon Jan 11, 2021 3:39 pm

Hi David,

You said before that you already have Thermo-Calc calculations. Just use the same database for MICRESS which you have used with Thermo-Calc. I think your newest version of the TCFE (e.g. TCFE9) should be fine.

You cannot edit the .GES-file because it is a binary file. But in the GES_FILES directory of you MICRESS installations, besides the .GES-files themselves, there are Thermo-Calc macro-files (.TCM). Those you can edit and adapt to your system. When you run them with Thermo-Calc, they create the .GES-file you need.

Bernd

SteelOx
Posts: 11
Joined: Mon Dec 07, 2020 10:17 am
anti_bot: 333

Re: Surface Oxidation

Post by SteelOx » Tue Jan 12, 2021 10:20 am

Bernd wrote:
Mon Jan 11, 2021 3:39 pm
Hi David,

You said before that you already have Thermo-Calc calculations. Just use the same database for MICRESS which you have used with Thermo-Calc. I think your newest version of the TCFE (e.g. TCFE9) should be fine.

You cannot edit the .GES-file because it is a binary file. But in the GES_FILES directory of you MICRESS installations, besides the .GES-files themselves, there are Thermo-Calc macro-files (.TCM). Those you can edit and adapt to your system. When you run them with Thermo-Calc, they create the .GES-file you need.

Bernd
Thanks for the help!

I have created a thermocalc of my system with the relevant elements and phases, but i dont now know how to use this in my micress simulations, i cant save or export this script. So i dont know how micress will be able to use this.

I also dont know where the GES files are for that.

Any help is welcomed

Kind Regards SteelOX

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

Re: Surface Oxidation

Post by Bernd » Tue Jan 12, 2021 11:27 am

Hi SteelOX,

In your MICRESS installation directory, there is a directory "Examples" which contains different types of examples. Under subdirectory "Training" you find the file T011_Gamma_Alpha_TQ_dri. This is the example I proposed to start with.
At the same place there is a subdirectory called "GES_FILES" which contains the file

Examples/Training/GES_FILES/TC2019a_FeCMn.TCM

This is the Macro-file which has been used for creating TC2019a_FeCMn.GES5 which is also located there and which is used by the MICRESS example T011_Gamma_Alpha_TQ_dri. In case you have an older MICRESS version, the year number in the name may differ.
If you want to modify the example T011_Gamma_Alpha_TQ_dri for use of different elements and phases, you need first to modify TC2019a_FeCMn.TCM. First open it with any text editor. The content looks like:

@@ LOGFILE GENERATED ON UNIX / KTH DATE 2013- 9-11
@@
@@ GES5 file generation for Gamma Alpha (FeCMn)

go dat
sw
tcfe9
d-e fe c mn
rej phases *
rest phases liquid bcc_a2 cementite fcc_a1
get-data
app
mobfe4
d-e fe c mn
rej phases *
rest phases liquid fcc_a1 cementite bcc_a2
get data
go gibbs
save TC2019a_FeCMn
exit



The script essentially opens the thermodynamic database, defines elements and phases, then appends a mobility database for which it again defines elements and phases, and then saves the workspace as .GES5 file. I marked as bold what you need to adapt:
  • name of thermodynamic database. The tcfe steel database is probably what you need, you should adjust the number to the most recent version available in your place
  • list of elements
  • list of phases
  • name of mobility database which is appended: This is the database where you take diffusion data from. Again, you should adjust the number to the most recent version available in your place
  • list of elements for mobility data (should be identical as for the thermodynamic database)
  • list of phases: same list as above - phases for which no mobility data is available are automatically ignored
  • name of .GES-file to be created: give it a new name which you then refer to in the modified MICRESS example dri-file
After modification, you can run the .TCM file with Thermo-Calc. There are different options how you can do it, depending on your installation. The simplest is to just double-click the .TCM-file. Another option is to start Thermo-Calc in the console mode, and to past the content of the .TCM-file into the shell. I propose to paste the text line by line so that you can see whether there were error messages in case of wrong input syntax.

After that, you have your modified .GES5 files which you can use with your modified MICRESS example.

Bernd

Post Reply