The distribution of solute element

dendritic solidification, eutectics, peritectics,....
Post Reply
Shenyz
Posts: 23
Joined: Mon Jul 25, 2016 4:33 pm
anti_bot: 333

The distribution of solute element

Post by Shenyz » Sun Aug 16, 2020 7:58 pm

Hi Bernd,
Recently I am studying the solute distribution during the solidification process using MICRESS and find some phenomena that confused me.
1) When I simulated the distribution of Al in steel, I found the Al content decrease to a small value (much smaller than the experiment result) during the solidification process, even negative in the cooling process after the solidification, as the picture shows. I don't know what's wrong.
1.png
1.png (772.37 KiB) Viewed 761 times
2) I found the C content will fluctuate in the cooling process after the solidification instead of slowly change or keep constant. I‘m wondering is this the correct change?
2.jpg
2.jpg (147.49 KiB) Viewed 761 times

Bernd
Posts: 1155
Joined: Mon Jun 23, 2008 9:29 pm

Re: The distribution of solute element

Post by Bernd » Mon Aug 17, 2020 11:43 am

Hi Shenyz,

It is difficult to say what causes the problems without having closer insight into your input file. What I could imagine as reason for the negative Al composition is an insufficient updating rate of the thermodynamic data. This can lead to a too far extrapolation, ending up in the negative compositions.

The second problem is not very typical, so it is hard to guess. If you use the full diffusion matrix for the solid matrix, it could be due to a stability problem, which sometimes happens with high-alloyed systems. It should vanish if the diffusion time stepping is reduced. This is automatically done as long as there is other fast diffusion present (like in the liquid), and can be done manually by reducing the general prefactor ("Factor for diffusion time stepping") in Numerical Parameters. It can also be done individually for each diffusion term using "cushion" in the diffusion data input.

Please send me more information on your application case if this does not solve your problem.

Bernd

Shenyz
Posts: 23
Joined: Mon Jul 25, 2016 4:33 pm
anti_bot: 333

Re: The distribution of solute element

Post by Shenyz » Mon Aug 17, 2020 2:20 pm

Hi Bernd,

Ok, I'll try it. Thanks very much!

Kind regards,
Yaozu

Shenyz
Posts: 23
Joined: Mon Jul 25, 2016 4:33 pm
anti_bot: 333

Re: The distribution of solute element

Post by Shenyz » Wed Aug 26, 2020 5:27 pm

Hi Bernd,

I use the same project to simulate the solidification process of steel with different initial temperatures (different undercooling). I found that when the initial temperature is 17K below the solidus temperature, the solidification frequency is liquid→ liquid+ferrite→ liquid+ferrite+austenite. When the initial temperature is 20K below the solidus temperature, the solidification frequency is liquid→ liquid+austenite and the ferrite phase disappears during the whole solidification process. However, when the undercooling increases to 24K (the initial temperature is 24K below the solidus temperature), the ferrite phase reappears and the solidification frequency is liquid→ liquid+ferrite→ liquid+ferrite+austenite, as the picture shows.
a.png
a.png (183.13 KiB) Viewed 733 times
I don't know why this happens.

Kind regards,
Yaozu

Bernd
Posts: 1155
Joined: Mon Jun 23, 2008 9:29 pm

Re: The distribution of solute element

Post by Bernd » Wed Aug 26, 2020 7:43 pm

Dear Yaozu,

I am a bit astonished about your description, because if your initial temperature is below solidus temperature (and you don't consider release of latent heat), then you would get a massive transformation without any segregation - perhaps you mean below "liqudus" temperature instead?

It is difficult to know the reason for the behaviour you describe. However, there seems to be no reason for that, and you definitively end up in an unstable situation for -20K, just due to a different initial temperature. So it seems pretty clear that the problem is purely numerical, e.g. that nucleation was not successful for whichever reason. I think you should re-check your simulation setup from this point of view (if you could paste or attach your input file here, I could give you more concrete advice).

Is your last question linked to the problem which you described before (negative Al composition etc.)? Otherwise I would like to detach it and create a new thread instead.

Bernd

Shenyz
Posts: 23
Joined: Mon Jul 25, 2016 4:33 pm
anti_bot: 333

Re: The distribution of solute element

Post by Shenyz » Thu Aug 27, 2020 8:48 am

Dear Bernd:

I'm sorry for my mistake. Yes, I mean the 'liquidus' temperature.

The related nucleation parts of the input file are as follow:
# Data for further nucleation
# ===========================
# Enable further nucleation?
# Options: nucleation nucleation_symm no_nucleation [verbose|no_verbose]
nucleation
# Additional output for nucleation?
# Options: out_nucleation no_out_nucleation
out_nucleation
#
# Number of types of seeds?
1
#
# Input for seed type 1:
# ----------------------
# Type of 'position' of the seeds?
# Options: bulk region interface triple quadruple front [restrictive]
interface
# Phase of new grains (integer) [unresolved|add_to_grain]?
2
# Reference phase (integer) [min. and max. fraction (real)]?
0
# Substrate phase [2nd phase in interface]?
# (set to 0 to disable the effect of substrate curvature)
1
# maximum number of new nuclei 1?
# (set negative for unlimited number)
5
# Grain radius [micrometers]?
0.00000
# Choice of growth mode:
# Options: stabilisation analytical_curvature
stabilisation
# min. undercooling [K] (>0)?
1.0000
# Determination of nuclei orientations?
# Options: random randomZ fix range parent_relation
parent_relation
# Minimal value of rotation angle? [Degree]
-5.000
# Maximal value of rotation angle? [Degree]
+5.000
# Shield effect:
# Shield time [s] [shield phase or group number] ?
10
# Shield distance [micrometers] [ nucleation distance [micrometers] ]?
25
# Nucleation range
# min. nucleation temperature for seed type 1 [K]
1600.000
# max. nucleation temperature for seed type 1 [K]
1740.000
# Time between checks for nucleation? [s]
# Options: constant from_file
constant
# Time interval [s]
1.0000
# Shall random noise be applied?
# Options: nucleation_noise no_nucleation_noise
no_nucleation_noise
#
# Max. number of simultaneous nucleations?
# ----------------------------------------
# (set to 0 for automatic)
0
#
# Shall metastable small seeds be killed?
# ---------------------------------------
# Options: kill_metastable no_kill_metastable
no_kill_metastable
#

This is the same simulation as the previous one. I'm wondering if there are some connections.
Thanks very much!

Greetings,
Yaozu

Bernd
Posts: 1155
Joined: Mon Jun 23, 2008 9:29 pm

Re: The distribution of solute element

Post by Bernd » Thu Aug 27, 2020 12:15 pm

Hi Yaozu,

using "maximum number of new nuclei" is dangerous: If MICRESS happen to choose 5 unfavourable places, it will not check again, and you will not get any seed. I generally advise to use "-1" (unrestricted) here to avoid getting erratic results, and to adjust the number of seeds by using shield data, nucleation distance and checking interval instead.
Did you get any check for nucleation in the -20K case?

Bernd

Post Reply