MICRESS Forum

Hello,

I have problems to simulate a multiphase solidification of an aluminum bronze (CuAl9Ni5Fe5). In the quasi-binary phase diagram (http://www.kupfer-institut.de/front_fra ... I_i006.pdf - page 3) you can see the solidification sequence and phase changing of the interesting composition. I know that the alpha-phase is FCC_A1, the beta-phase is BCC_A2 and the kappa-phase can be AL13FE4 or BCC_B2 (References: Chapter 6.1.5 in 'Metallografie' by Schumann and Oettel, ISBN 352730679X and ThermoCalc
support). I can simulate the solidification of the first phase (beta, BBC_A2) but I do not get the second (alpha, FCC_A1) in all my simulations. I found a few things in this forum (e.g. demixing) and with this entry I could set a grain in FCC_A1 but either it disappeared or I got an error for larger grains.

I hope you can give me tips how I come forth and thank you for your efforts.

Best regards,
Nico

Hi NicoSim

Welcome to the MICRESS forum! As the moderator of this Forum, I spit off your reply from the former thread and created a new thread in MICRESS APPLICATIONS/Solidification, where the topic fits much better!


When starting a MICRESS simulation for a new and quite complex system like your Aluminium bronze, a lot of problems may occur, and it is not so easy to figure out where they come from. It is wise to start with considering thermodynamics:
One should go on step by step, i.e. after having managed to get liquid-bcc stable, to include the next one (liquid-fcc), but to switch off fcc-bcc first. Then, you should have a look on the initial linearisation data for these interfaces in the .log output, which is created either during initialisation or when nucleation of the corresponding phases is checked for the first time. Please show us these data (by copying from the .log file into the forum), because they show whether the system is behaving well and whether initialisation was successfull. If this is not the case, it makes no sense to look at other parameters which could cause such problems. Maybe, we can already see some problems...

Bernd

Hi Bernd,

thank you for the quick reply.

Here is the content of one of my *.log-files with the 'best' result until now:

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

MICRESS binary
**************
version number: 5.501 (Linux)
compiled: 09/03/2010
('double precision' binary)
license expires in 279 days

Miscellanea
***********
Date and time: 27/04/2011 (14:08)
User: steinern
Machine: frodo
Driving file: /home/steinern/22


Flags and settings
******************

Geometry
--------
AnzX, AnzY, AnzZ = 100, 1, 100
deltaX, deltaY, deltaZ = 1.0000E-04, 1.0000E-04, 1.0000E-04

Flags
-----
Simulation with concentration coupling
'Double-obstacle' potential
averaging
no_1d_far_field
no_1d_temp
IFaceDim coeff. : 0.100 (0.75, 0.50)
nTupelDim coeff.: 0.100 (0.75, 0.50)


The data compression is done with 'zlib' (cf. http://www.zlib.net/).
The routine for appending is loosely derived from the 'gzappend.c' example.


Time input data
***************
Number of intermediate outputs = 30
Intermediate output at t = 1.0000000E-01 s
Intermediate output at t = 2.0000000E-01 s
Intermediate output at t = 3.0000000E-01 s
Intermediate output at t = 4.0000000E-01 s
Intermediate output at t = 5.0000000E-01 s
Intermediate output at t = 6.0000000E-01 s
Intermediate output at t = 7.0000000E-01 s
Intermediate output at t = 8.0000000E-01 s
Intermediate output at t = 9.0000000E-01 s
Intermediate output at t = 1.0000000E+00 s
Intermediate output at t = 1.1000000E+00 s
Intermediate output at t = 1.2000000E+00 s
Intermediate output at t = 1.3000000E+00 s
Intermediate output at t = 1.4000000E+00 s
Intermediate output at t = 1.5000000E+00 s
Intermediate output at t = 1.6000000E+00 s
Intermediate output at t = 1.7000000E+00 s
Intermediate output at t = 1.8000000E+00 s
Intermediate output at t = 1.9000000E+00 s
Intermediate output at t = 2.0000000E+00 s
Intermediate output at t = 2.1000000E+00 s
Intermediate output at t = 2.2000000E+00 s
Intermediate output at t = 2.3000000E+00 s
Intermediate output at t = 2.4000000E+00 s
Intermediate output at t = 2.5000000E+00 s
Intermediate output at t = 2.6000000E+00 s
Intermediate output at t = 2.7000000E+00 s
Intermediate output at t = 2.8000000E+00 s
Intermediate output at t = 2.9000000E+00 s
Intermediate output at t = 3.0000000E+00 s
Coefficient for phase-field criterion 0.900
Coefficient for segregation criterion 0.900
Upper limit for time step: 1.000E-02
Lower limit for time step: 1.000E-06
Number of iterations for initialisation: 10


Phase data
**********
Number of distinct solid phases = 2

Data for phase 1:
-----------------
in phase 1, recrystallisation will not be considered.
Phase 1 is anisotropic
cubic crystal symmetry will be considered.
No categorization is allowed for phase 1

Data for phase 2:
-----------------
in phase 2, recrystallisation will not be considered.
Phase 2 is anisotropic
cubic crystal symmetry will be considered.
No categorization is allowed for phase 2
Grain orientations will be defined by 2D angles.


Grain input
***********
Grains will be positioned deterministically
Number of grains = 6
Input data for grain number 1:
'Round' grain.
x,z coordinates : 75.0000, 75.0000 micrometers
Grain radius : 5.00000 micrometers
Grain 1 is stabilized
Grain set without Voronoi construction.
Phase number : 1
Rotation angle : +0.0000 degree
Input data for grain number 2:
'Round' grain.
x,z coordinates : 75.0000, 25.0000 micrometers
Grain radius : 5.00000 micrometers
Grain 2 is stabilized
Grain set without Voronoi construction.
Phase number : 1
Rotation angle : +0.0000 degree
Input data for grain number 3:
'Round' grain.
x,z coordinates : 25.0000, 50.0000 micrometers
Grain radius : 5.00000 micrometers
Grain 3 is stabilized
Grain set without Voronoi construction.
Phase number : 1
Rotation angle : +0.0000 degree
Input data for grain number 4:
'Round' grain.
x,z coordinates : 75.0000, 50.0000 micrometers
Grain radius : 2.00000 micrometers
Grain 4 is stabilized
Grain set without Voronoi construction.
Phase number : 2
Rotation angle : +0.0000 degree
Input data for grain number 5:
'Round' grain.
x,z coordinates : 25.0000, 75.0000 micrometers
Grain radius : 2.00000 micrometers
Grain 5 is stabilized
Grain set without Voronoi construction.
Phase number : 2
Rotation angle : +0.0000 degree
Input data for grain number 6:
'Round' grain.
x,z coordinates : 25.0000, 25.0000 micrometers
Grain radius : 2.00000 micrometers
Grain 6 is stabilized
Grain set without Voronoi construction.
Phase number : 2
Rotation angle : +0.0000 degree


Data for further nucleation
***************************
Run-time nucleation disabled


Phase interaction data
**********************

Data for phase interaction 0 / 1:
---------------------------------
Interaction between 0 and 1 will be simulated.
Interaction parameters between phases LIQUID and 1:
Averaging coefficient Av = +0.55
Maximal driving force dGMax = 100.00 [J/cm**3]
'Smoothing angle' smooth = 45.00
Surface energy sigma = 1.70000E-04 [J/cm**2]
Kinetic coefficient mu = 5.00000E-02 [cm**4/(Js)]
static anisotropy coefficient: 0.5000000
static anisotropy coefficient: 0.2000000

Data for phase interaction 0 / 2:
---------------------------------
Interaction between 0 and 2 will be simulated.
Interaction parameters between phases LIQUID and 2:
Averaging coefficient Av = +0.55
Maximal driving force dGMax = 100.00 [J/cm**3]
'Smoothing angle' smooth = 45.00
Surface energy sigma = 2.00000E-06 [J/cm**2]
Kinetic coefficient mu = 5.00000E-02 [cm**4/(Js)]
static anisotropy coefficient: 0.5000000
static anisotropy coefficient: 0.2000000

Data for phase interaction 1 / 1:
---------------------------------
Interaction between 1 and 1 will not be simulated.

Data for phase interaction 1 / 2:
---------------------------------
Interaction between 1 and 2 will not be simulated.

Data for phase interaction 2 / 2:
---------------------------------
Interaction between 2 and 2 will not be simulated.


Concentration data
******************
Number of dissolved constituents = 3
Concentration in weight percent wt%
Diffusion of component 1 in phase 0 will be solved.
Diff.-coefficient:
Prefactor: 2.00000000000000010E-004 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]
Diffusion of component 1 in phase 1 will be solved.
Diff.-coefficient:
Prefactor: 1.00000000000000002E-008 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]
Diffusion of component 1 in phase 2 will be solved.
Diff.-coefficient:
Prefactor: 1.00000000000000002E-008 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]
Diffusion of component 2 in phase 0 will be solved.
Diff.-coefficient:
Prefactor: 2.00000000000000010E-004 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]
Diffusion of component 2 in phase 1 will be solved.
Diff.-coefficient:
Prefactor: 1.00000000000000002E-008 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]
Diffusion of component 2 in phase 2 will be solved.
Diff.-coefficient:
Prefactor: 1.00000000000000002E-008 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]
Diffusion of component 3 in phase 0 will be solved.
Diff.-coefficient:
Prefactor: 2.00000000000000010E-004 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]
Diffusion of component 3 in phase 1 will be solved.
Diff.-coefficient:
Prefactor: 1.00000000000000002E-008 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]
Diffusion of component 3 in phase 2 will be solved.
Diff.-coefficient:
Prefactor: 1.00000000000000002E-008 [cm**2/s]
Activation energy: 0.0000000000000000 [J/mol]


Phase diagram - input data
**************************
In phase 1 components 1 and 2 are enhanced stoichiometric.
In phase 2 components 1 and 2 are enhanced stoichiometric.
Name of Thermo-Calc *.GES5 file: /home/steinern/thermocalc/albr1
Interval for updating thermodynamic data [s] = 100.00

Input of the phase diagram of phase 0 and phase 1:
--------------------------------------------------
TC-Coupling activated
Maximal temperature deviation -1.0000000000000000 K

Input of the phase diagram of phase 0 and phase 2:
--------------------------------------------------
TC-Coupling activated
Maximal temperature deviation -1.0000000000000000 K
The database contains the following components:
1: AL
2: CU
3: FE
4: NI
Thermo-Calc index of (MICRESS) component 0 = 2
Thermo-Calc index of (MICRESS) component 1 = 1
Thermo-Calc index of (MICRESS) component 2 = 4
Thermo-Calc index of (MICRESS) component 3 = 3
0 -> CU
1 -> AL
2 -> NI
3 -> FE
The database contains 56 phases:
1: LIQUID
2: AL11MN4
3: AL12MN
4: AL13FE4
5: AL2FE
6: AL3NB
7: AL3NI
8: AL3NI2
9: AL3NI5
10: AL4MN
11: AL5FE2
12: AL5FE4
13: AL6MN
14: ALCE_AMORPHO
15: ALCUZN_GAMMA
16: ALCUZN_T
17: ALCU_DELTA
18: ALCU_EPSILON
19: ALCU_ETA
20: ALCU_GAMMA_D
21: ALCU_PRIME
22: ALCU_THETA
23: ALCU_ZETA
24: ALLI
25: ALMO
26: ALM_D019
27: ALNB3
28: ALTI
29: BCC_A2
30: BCT_A5
31: CBCC_A12
32: CU4TI
33: CU6Y
34: CUB_A13
35: CUSB_FCC_BET
36: CUSN_GAMMA_D
37: CUTI
38: DIAMOND_A4
39: FCC_A1
40: FE2U
41: FE4N
42: FESB
43: FEU6
44: FEUZR_DELTA
45: FEZR2
46: FEZR3
47: HCP_A3
48: HCP_ZN
49: LAVES_C14
50: LAVES_C15
51: LAVES_C36
52: NI3NB
53: NI3TI
54: NI3V
55: ORTHORHOMBIC
56: TETRAGONAL_U
Thermo-Calc index of the (MICRESS) phase 0 = 1
Thermo-Calc index of the (MICRESS) phase 1 = 29
Thermo-Calc index of the (MICRESS) phase 2 = 39
0 -> LIQUID
1 -> BCC_A2
2 -> FCC_A1
Molar volume of (MICRESS) phase 0 (LIQUID): 10.000 [cm**3/mol]
Molar volume of (MICRESS) phase 1 (BCC_A2): 10.000 [cm**3/mol]
Molar volume of (MICRESS) phase 2 (FCC_A1): 10.000 [cm**3/mol]
initial equilibrium at T = 1319.000


Initial concentrations
**********************
Concentrations will be automatically set, matrix phase: 0
Initial concentration of component 1 (AL) in phase 0 (LIQUID) = 9.000000 wt%
Initial concentration of component 2 (NI) in phase 0 (LIQUID) = 5.000000 wt%
Initial concentration of component 3 (FE) in phase 0 (LIQUID) = 5.000000 wt%


Parameters for latent heat and 1D temperature field
***************************************************
Simulation with release of latent heat
Simulation with release of pseudo-3D latent heat of phase 1 (BCC_A2), fracKrit= 0.0000000000000000 !
Simulation with release of pseudo-3D latent heat of phase 2 (FCC_A1), fracKrit= 0.0000000000000000 !


Boundary conditions
*******************
Initial temperature at the bottom: 1319.000 K
Temperature gradient at beginning: 0.0000 K/cm
Heat flow = -50.000 [J/s*cm^3]

In W-direction periodic/wrap-around boundary condition for phase field
In E-direction periodic/wrap-around boundary condition for phase field
In B-direction periodic/wrap-around boundary condition for phase field
In T-direction periodic/wrap-around boundary condition for phase field

In W-direction periodic/wrap-around boundary condition for concentration field
In E-direction periodic/wrap-around boundary condition for concentration field
In B-direction periodic/wrap-around boundary condition for concentration field
In T-direction periodic/wrap-around boundary condition for concentration field

Unit-cell model symmetric with respect to the x/y diagonal plane not activated


Other numerical parameters
**************************
Phase minimum phMin = 1.00E-04
Interface thickness etaZ = 4.0000000000000000


Beginning of initialisation
***************************
Grain number 1 set
Grain number 2 set
Grain number 3 set
Grain number 4 set
Grain number 5 set
Grain number 6 set

# The linearisation parameters of the phases LIQUID/BCC_A2 are:
# -------------------------------------------------------------
1319.0000 ! T0 [K]
2.8933337 ! dG [J/cm**3]
2.5303643 ! dSf+ [J/cm**3K]
-1.1582007 ! dSf- [J/cm**3K]
688.35424 ! dH [J/cm3]
8.9999056 ! c0(AL)/LIQUID
10.042321 ! c0(AL)/BCC_A2
5.0003640 ! c0(NI)/LIQUID
0.98164914 ! c0(NI)/BCC_A2
4.9929700 ! c0(FE)/LIQUID
82.601157 ! c0(FE)/BCC_A2
-5.5524510 ! m(AL)/LIQUID
-147.87317 ! m(AL)/BCC_A2
-22.863330 ! m(NI)/LIQUID
-36.188007 ! m(NI)/BCC_A2
39.137665 ! m(FE)/LIQUID
-75.958609 ! m(FE)/BCC_A2
1.27593558E-02 ! dcdT(AL)/LIQUID
1.83758671E-03 ! dcdT(AL)/BCC_A2
3.07636772E-02 ! dcdT(NI)/LIQUID
3.88883900E-03 ! dcdT(NI)/BCC_A2
5.70275871E-02 ! dcdT(FE)/LIQUID
-4.76579066E-02 ! dcdT(FE)/BCC_A2

# The linearisation parameters of the phases LIQUID/FCC_A1 are:
# -------------------------------------------------------------
1319.0000 ! T0 [K]
82.891475 ! dG [J/cm**3]
1.0314681 ! dSf+ [J/cm**3K]
0.55563576 ! dSf- [J/cm**3K]
628.13457 ! dH [J/cm3]
9.0000279 ! c0(AL)/LIQUID
8.7134614 ! c0(AL)/FCC_A1
4.9989028 ! c0(NI)/LIQUID
16.274923 ! c0(NI)/FCC_A1
4.9981667 ! c0(FE)/LIQUID
23.839136 ! c0(FE)/FCC_A1
-27.048716 ! m(AL)/LIQUID
-36.394903 ! m(AL)/FCC_A1
8.4231318 ! m(NI)/LIQUID
-1.2431158 ! m(NI)/FCC_A1
16.912946 ! m(FE)/LIQUID
-12.420206 ! m(FE)/FCC_A1
4.17435337E-03 ! dcdT(AL)/LIQUID
2.35611553E-03 ! dcdT(AL)/FCC_A1
2.01057007E-02 ! dcdT(NI)/LIQUID
-5.75860983E-03 ! dcdT(NI)/FCC_A1
2.39386454E-02 ! dcdT(FE)/LIQUID
6.12515168E-02 ! dcdT(FE)/FCC_A1
Initial concentration in the phases:
LIQUID, CU: 81.00000 wt%
LIQUID, AL: 9.000000 wt%
LIQUID, NI: 5.000000 wt%
LIQUID, FE: 5.000000 wt%
BCC_A2, CU: 6.374872 wt%
BCC_A2, AL: 10.04232 wt%
BCC_A2, NI: 0.9816491 wt%
BCC_A2, FE: 82.60116 wt%
FCC_A1, CU: 51.17248 wt%
FCC_A1, AL: 8.713461 wt%
FCC_A1, NI: 16.27492 wt%
FCC_A1, FE: 23.83914 wt%
Overall concentration = 9.020283 wt%
Overall concentration = 4.949247 wt%
Overall concentration = 6.618467 wt%
tWidth_max( 0 : 1 ) = 2.3284313E-04 s
tWidth_max( 0 : 2 ) = 1.9791666E-02 s
Maximal value for tWidth = 2.3284313E-04 s for phase-field solver
Maximal value for tWidth = 1.1875000E-05 s for conc-field solver
Automatic time stepping (phase-field solver): decreased value for tWidth = 1.12812E-05 s
Initial value for tWidth = 1.12812E-05 s for automatic time stepping (phase-field solver)
Critical grain radius:
of phase 1 in phase 0 = 0.67184 / dT_unt [micrometers]
of phase 2 in phase 0 = 1.93898E-02 / dT_unt [micrometers]

==================================================

Time t = 0.0000 s
CPU-time: 1 s
Current phase-field solver time step = 1.13E-05 s
Temperature at the bottom = 1319.0 K
Temperature gradient = 0.00000 K/cm
Fraction of phase 0: 0.99653
Fraction of phase 1: 0.00332
Fraction of phase 2: 0.00015

**********************************************
* Begining of simulation *
**********************************************

Intermediate output for t = 0.10000 s
CPU-time: 679 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1317.5 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 0.20000 s
CPU-time: 701 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1315.9 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 0.30000 s
CPU-time: 735 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1314.4 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 0.40000 s
CPU-time: 909 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1312.9 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 0.50000 s
CPU-time: 859 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1311.3 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 0.60000 s
CPU-time: 1114 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1309.8 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 0.70000 s
CPU-time: 1392 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1308.3 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 0.80000 s
CPU-time: 1389 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1306.7 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 0.90000 s
CPU-time: 1391 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1305.2 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.0000 s
CPU-time: 1395 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1303.7 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.1000 s
CPU-time: 1394 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1302.1 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.2000 s
CPU-time: 1424 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1300.6 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.3000 s
CPU-time: 1430 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1299.1 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.4000 s
CPU-time: 1443 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1297.5 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.5000 s
CPU-time: 1475 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1296.0 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.6000 s
CPU-time: 1502 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1294.5 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.7000 s
CPU-time: 1495 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1292.9 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.8000 s
CPU-time: 1486 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1291.4 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 1.9000 s
CPU-time: 1500 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1289.8 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.0000 s
CPU-time: 1519 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1288.3 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.1000 s
CPU-time: 1575 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1286.8 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.2000 s
CPU-time: 1604 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1285.2 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.3000 s
CPU-time: 1630 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1283.7 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.4000 s
CPU-time: 1633 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1282.1 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.5000 s
CPU-time: 1684 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1280.6 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.6000 s
CPU-time: 1614 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1279.0 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.7000 s
CPU-time: 1823 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1277.5 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.8000 s
CPU-time: 2038 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1276.0 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 2.9000 s
CPU-time: 1708 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1274.4 K
Temperature gradient = 0.00000 K/cm

Intermediate output for t = 3.0000 s
CPU-time: 1769 s
Current phase-field solver time step = 1.00E-06 s
Average conc. of comp. 1 = 9.0202830, Variation = -0.0000000 wt%
Average conc. of comp. 2 = 4.9492471, Variation = -0.0000000 wt%
Average conc. of comp. 3 = 6.6184669, Variation = -0.0000000 wt%
Temperature at the bottom = 1272.9 K
Temperature gradient = 0.00000 K/cm

==================================================

Simulation run on machine frodo
CPU-time in seconds: 42327
End
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

How can I identify the optimal grain size or it is not important?

Nico

Dear NicoSim,

from what I see in the pasted log-file, in phase 2, component 2 and 3 should be set to stoichiometric instead of 1 and 2 (for phase 1, setting of 1 and 2 is perfect). Maybe, this makes the problems, please check and tell me!
Finding the correct stoichiometric conditions is one of the first steps and essential in case of complex systems. Sometimes, "demixing" tendencies appear at a later stage, then one has to consult the .TabLin output.

As I see, you never perform an actualisation of the thermodynamic data (all 100 seconds means never in a simulation time of 3 seconds)! Especially with stoichiometric phases, you should invest the time to recalculate thermodynamic data from time to time, around 100 times during the solidification interval as a crude rule of thumb.

By the way, I would reduce the number of phase descriptions in the .ges5 file - it is not only confusing, but also increases the amount of memory and simulation time. I always include only those phases which I plan to use actually or in the near future.

I do not understand what you mean by "optimal" grain size...

Bernd

Dear Bernd,

the simulations are more stable if I set component 2 and 3 in phase 2 to stoichiometric. But instead demixing occurs for component 1 (aluminum). In a further simulation I have set all components in phase 2 to stoichiometric and the result is better than before.

Furthermore I reduced the kinetic coefficient of phase 2 and the distance between the different grains. I achieved a better grain grow with it.

With 'optimal' I mean a ideal grain size. I set the grain radii arbitrarily until now and I want know if it is the right way? In my simulations, for example, I set r(BCC_A2) = 5 um and r(FCC_A1) = 2 um.

In addition, I have a few new questions:

1) Why can I choose a round, elliptic or rectangular shape for the grains?

2) In the TabLin of my simulations I get negative values of component's mass. Why it is possible?

Best regards,
Nico

Dear NicoSim,

if it is like you say, the system seems to be quite complex, or the database is not not valid for the composition of your alloy. Which database are you using?
Anyway, you should define the stoichiometric components such that no "demixing" occurs, in worst case you have to set all compontents stoichiometric. But take into account that the behaviour can change with temperature, and whether demixing occurs also at lower temperatures can be seen from warnings in the screeen output or in the .TabLin file.

With respect to the "optimal" grain size: You are simulating solidification, and it would be logical to use nucleation for making the phases appear. The usual way would be to use "grain input" for setting initial seeds of bcc with radius 0 and "stabilisation", i.e. to use "small grains" as initial microstructure. You have to decrease the initial temperature slightly to get a negative driving force value, so that they can grow.
fcc is expected to appear at lower temperature, therefor it should be set via nucleation during runtime. You can decide to nucleate it from the bulk or on the bcc-liquid interface if a certain critical undercooling is reached.

Small seeds are always round, so it does not make sense to give them another geometry. Elliptical or quadratic grains are options for other purposes like solid-solid reactions.
"m" in the TabLin file means the liquidus or solidius slope, the output has the same meaning as in the initial linearisation parameters in the .log file. There you should check whether "demixing" also occurs at a later stage. If you nucleate fcc at a lower temperature like I propose, "demixing" behaviour could be different...

Bernd

Dear Bernd,

I use the SSOL4-database of ThermoCalc. I know that it gives a problem with the relationship between copper and iron. I found it as I had tried to check the thermodynamic data. But the ThermoCalc service wrote back that it should not be a problem if I only use FCC and BCC .

I set the grain to a radius of 0 and decreased the temperature to the point where dG is negative. But the initial temperature is lower than my reference now. So the set BCC-grain grows only a little bit and then the FCC-phase starts to grow. Does it give an other possibility to control the dG as decreasing the initial temperature?

Regards,
Nico

Dear NicoSim

You cannot change the order of appearance of the phases by numerical measures, because it is a physical property of the system (as long as we trust in the database and do not change composition). From the phase diagram to which you send us the link, it seems that fcc and bcc should appear at similar temperature, so for me it sound reasonable...
Of course, the exact temperatures and the order in which they appear should be quite sensitive to the exact alloy composition!
What is your "reference" temperature and how is it defined? Do you have any experiments to compare with?

Bernd

Dear Bernd,

I assume that the phase diagram was plotted in a correct relation of axis. I drew additional lines for my composition and then I got the following temperatures:

Liquid --> Liquid+Beta: about 1355K
Liquid+Beta --> Liquid+Beta+Alpha: about 1345K
Liquid+Beta+Alpha --> Beta+Alpha: about 1327K
Beta+Alpha --> Beta+Alpha+Kappa: about 1207K
Beta+Alpha+Kappa --> Alpha+Kappa: about 1141K

The first problem is that the solidification temperature in the simulation is too low. The solidification of the beta-phase starts at 1317.5K. At 1315.9K the alpha-phase already solidifies and so the difference is much smaller how I expect.
An other problem is if I increase the weight percent of aluminum to 10% and 11% the initial temperatures decrease to 1310.5K and 1303K with rising aluminum fraction. But the reference shows that the initial temperature should increase slightly.

The phase diagram is my sole reference until now. The experimental results which I have are only the micrograph by different cast technics of the composition. With the simulation I want to imitate the micrographs.

Regards,
Nico

Dear Nico,

this sounds as if the phase diagram you have could also be rather schematic than quantitatively correct...
But at least simulation seems to give something which qualitatively is comparable to experimental evidence, the phase sequence is correct, isn't it? Of course it would be nice to perform some DSC experiments to clarify the correct liquidus temperature (are you able to perform own experiments?) - perhaps it would turn out that the database is even more correct than the phase diagram. The liquidus temperature seems to be quite sensitive to the Ni and Fe content, so there could easily arise experimental inaccuracies...


For simulation, a next check could be to compare microstructures, especially the phase fractions which could be compared to the experimental micrographs.

Bernd