Hello dear users,
I am a new user. I am trying to define two identical phases (austenite here) with different chemistry. I tried no phase interaction that was partly suitable for my purpose since it didn't ask for a phase diagram between the two phases. However, the boundary between the two phases is missing. Then, I tried the identical phases nb.. The problem is that I don't know what this command exactly does. At the same time, it did not work too. The error "bad integer for item 1 in list input" shows up too.
Any suggestion?
Thank you,
Mehran
Phase interaction
Re: Phase interaction
Dear mehran,
welcome to the MICRESS forum!
From your post it is not completely clear what you intend to do. As I understand, you want to start a simulation with an initial microstructure which contains of two grains of the same phase (austenite), but with different composition. This sounds like if you want to simulate interdiffusion in a diffusion couple or something similar. Is this true?
In MICRESS, there is no direct way to define different grains with different compostion because this is a quite specific condition. Newertheless, there are (at least) two ways how to achieve that, a clean one which takes some more time, and a "dirty" one:
1.) The "clean" way is to define only one phase (austenite) in MICRESS as phase no. 1, and to create the initial microstructure by setting various grains of phase 1 (grain input). Then, in the "phase interaction data" part of the input file, you need to define only the 1/1 phase interaction, if you expect movement of the grain interface due to curvature. Otherwise, you could also select "no_phase_interaction" here (or even use only one grain so that you do not have any interfaces if you do not need them!).
In "phase_diagram data", you do not need to specify anything as you have only one phase (if we suppose that you covered all the domain with grains of phase 1, and no part of the background phase 0 is left...).
In "initial concentrations" you now would select "from_file" and read in the initial concentration field(s) from ASCII file(s). By this way you can create whichever composition distribution at the beginning! (More information you can find here.)
For your application, you can create the ASCII files for the initial concentration fields in the following way: First make a short MICRESS simulation run using two different phases without interaction ("no_phase_interaction") with constant concentrations for each component in each phase ("input" keyword in "inital concentrations). Then, you open the concentration outputs *.conc* for output time 0 created by MICRESS is (one by one) using DP_MICRESS, and convert them by writing ASCII outputs ("Output"-"ASCII output from current"). Alternatively, you can convert all composition files (in multicomponent case) in a single run using Transfer_MICRESS, which is part of the DP_MICRESS tool.
2.) The "dirty" way is just to use two different phases with different compositions, and to define a "dummy" phase diagram to make the phases thermodynamically identical. This would mean, in the phase diagram input, to use the option "linear" and define an arbitrary T0 and
, identical equilibrium compositions in both phases (to switch off any redistribution) and slopes with value 0 (to avoid any driving forces). By this way, it is possible to specify different compostions for the different grains, as they are formally of different phase, by using "input" in "initial concentrations".
Both methods should work. Please tell us if you have still problems or I misinterpreted your question, probably by appending your input file, which helps to identify problems.
Bernd
welcome to the MICRESS forum!
From your post it is not completely clear what you intend to do. As I understand, you want to start a simulation with an initial microstructure which contains of two grains of the same phase (austenite), but with different composition. This sounds like if you want to simulate interdiffusion in a diffusion couple or something similar. Is this true?
In MICRESS, there is no direct way to define different grains with different compostion because this is a quite specific condition. Newertheless, there are (at least) two ways how to achieve that, a clean one which takes some more time, and a "dirty" one:
1.) The "clean" way is to define only one phase (austenite) in MICRESS as phase no. 1, and to create the initial microstructure by setting various grains of phase 1 (grain input). Then, in the "phase interaction data" part of the input file, you need to define only the 1/1 phase interaction, if you expect movement of the grain interface due to curvature. Otherwise, you could also select "no_phase_interaction" here (or even use only one grain so that you do not have any interfaces if you do not need them!).
In "phase_diagram data", you do not need to specify anything as you have only one phase (if we suppose that you covered all the domain with grains of phase 1, and no part of the background phase 0 is left...).
In "initial concentrations" you now would select "from_file" and read in the initial concentration field(s) from ASCII file(s). By this way you can create whichever composition distribution at the beginning! (More information you can find here.)
For your application, you can create the ASCII files for the initial concentration fields in the following way: First make a short MICRESS simulation run using two different phases without interaction ("no_phase_interaction") with constant concentrations for each component in each phase ("input" keyword in "inital concentrations). Then, you open the concentration outputs *.conc* for output time 0 created by MICRESS is (one by one) using DP_MICRESS, and convert them by writing ASCII outputs ("Output"-"ASCII output from current"). Alternatively, you can convert all composition files (in multicomponent case) in a single run using Transfer_MICRESS, which is part of the DP_MICRESS tool.
2.) The "dirty" way is just to use two different phases with different compositions, and to define a "dummy" phase diagram to make the phases thermodynamically identical. This would mean, in the phase diagram input, to use the option "linear" and define an arbitrary T0 and
Both methods should work. Please tell us if you have still problems or I misinterpreted your question, probably by appending your input file, which helps to identify problems.
Bernd