engineering thermodynamics mime212in-ho-group.snu.ac.kr/wp-content/uploads/calphad... ·...
TRANSCRIPT
The University of Michigan
Structural Dynamics Laboratory
열역학계산의기초및응용
CALPHAD Thermodynamic Database and Applications to
Steelmaking process and Alloy design
Department of Materials Science and Engineering
Professor In-Ho Jung
Tel : 82-2-880-7077
E-mail: [email protected]
http://in-ho-group.snu.ac.kr/
In-Ho Jung
Seoul National University, Seoul, South Korea
Contents
2
Review of Engineering Thermodynamics
Computational thermodynamic database
– Development of CALPHAD thermodynamic database
Applications to Steelmaking process
3
REVIEW OF
ENGINEERING THERMODYNAMICS
G = H – TS; G: Gibbs Energy, H: Enthalpy, S: Entropy
1. For pure element or pure compound (Al, O2, Al2O3, etc.)
o
T
o
T
o
T TSHG
T
K
po
K
o
T dTT
CSS
298
298
Enthalpy for compound at 298 K with
reference of pure stable elemental species
at 298 K and 1 atm ( , unknown)
: Cp = a + bT + cT-2
+ dTlnT + · · ·
is known (measurable)
Standard reference state for H :
Fe(bcc), Fe(fcc), Fe(l), H2O(l), H2O(g), H2(g),
O2(g), O(g), CaO, FeO, C(s), CO2, CO,.
0ΔH o
298K
Standard entropy at 298 K
( )00 oKS00 o
KH
* In FactSage compound database,
, , are stored to
calculate Gibbs energy of solid, liquid
and gas species
o
298KΔHo
KS298 pC
Gibbs Energy
4
T
K
p
o
K
o
T dTCHH298
298
5
2. Chemical reaction between pure compounds (No solution)
nA + mB = AnBm
o
rxn
o
rxn
o
B
o
A
o
BArxn
STH
mGnGGGmn
)(
In many thermo books, these , are given. These values are
not absolute values, but dependent on each chemical reaction.
In the FactSage, absolute Gibbs energy of each species (relative to
elemental species) is stored. Then, any reaction Gibbs energy can be
automatically calculated from the Gibbs energy of each species.
o
rxnH o
rxnS
Gibbs Energy
∆𝐺𝑟𝑥𝑛𝑜
6
3. Chemical reaction involving gas
nA + mO2(g) = AnO2m
At Equilibrium state
)(22 O
oA
oOArxn mGnGGG
n
ioii PRTGG ln
for gas species i
2ln O
o PmRTG
0 rxnG
)ln(2
1m
OP
o RTG
Gibbs Energy
= ∆𝐺𝑟𝑥𝑛𝑜
∆𝐺𝑟𝑥𝑛𝑜
7
3. Chemical reaction involving gas (continue)
In general, for aA + bB(g) = cC + dD(g)
At Equilibrium
)ln( bB
dD
P
Po RTG
Keq: Equilibrium constant
Gibbs Energy
∆𝐺𝑟𝑥𝑛𝑜
∆𝐺𝑟𝑥𝑛𝑜 = -RT ln Keq
8
4. Chemical reaction involving solid or liquid solution
)ln()(ln)( io
pureisoini aRTGG a: activity
change of Gibbs energy of i in solution
by interacting with surrounding species
Definition of activity
o
Ap
Pure Liquid A
Pure Gas A
A in solution
ApGas Mixture
AAo
A
AA x
P
Pa
∴ activity is movement of species in solution
Gibbs Energy
9
Definition of activity
ii
o
iii xppa )/(
0 x i1
ia
(+) deviation
(-) deviation
ideal
1
4. Chemical reaction involving solid or liquid solution
(+) deviation: repulsion between i and other species
: more active chemical reaction of i
(-) deviation: attraction between i and other species
: less active chemical reaction of i
ii xa
ii xa
In general, for aA + bB(g) = cC + dD(g)
At Equilibrium
)ln( bB
aA
dD
cC
Pa
Pao RTG
reactantsproductsrxn GGG
* FactSage solution database contains
the model and model parameters to
calculate and eventually get ia
iG
Gibbs Energy
∆𝐺𝑟𝑥𝑛𝑜
10
In most of thermodynamic book, we always calculate equilibrium condition
0 rxnG eq
o KRTG ln
But in reality, we want to first know the direction of reaction
mA + nB
many possible outputs
A2B
Inputs (initial condition)
Tfinal, Pfinal
(m-2)A
(n-1)B
AB2
(m-1)A
(n-2)B
A2B (m-3)A
(n-3)BAB2
(m-x)A
(n-y)B
(xA-yB)soln
Final equilibrium state?
Gibbs Energy
∆𝐺𝑟𝑥𝑛𝑜
11
(continue)
We have to find out which phase assemblage is the most stable at given Tf
and Pf with respect to the mass balance.
Gibbs energy minimization routine: ChemSage, Solgas-mix, etc.
The most stable phase assemblage has the lowest Gibbs energy.
In FactSage
i) Put inputs amount
ii) Select all possible phases (solid compounds, solid solutions,
liquid solutions, gases)
iii) Set Tfinal and Pfinal
iv) Calculation (Gibbs energy minimization routine)
v) Equilibrium phase assemblage
Gibbs Energy Minimization
Ellingham Diagram
12
Ellingham Diagram
13
ΔG
o(k
J /
mole
of O
2)
T (K)
(A)
(B)
- Collection of ΔGo for oxidation reaction
mA + O2 = AmO2 (reference: 1 mol of O2)
- Only consider for pure species.
(No solutions are considered.)
TpRG
pRTG
pa
aRTGG
O
o
O
o
OA
AOo
)ln(
ln
)()(
)(ln
2
2
2
2 ):0(, mEquilibriuG
A + O2 = AO2
Solution thermodynamics
14
A-B solution, (Solid or Liquid solution)
iisolution GxG
i
o
ii aRTGG ln Gi: partial Gibbs energy of i in solution
)lnln()( BBAA
o
BB
o
AA axaxRTGxGx
XB
mixgΔ
solutiong
o
Ag
o
Bg
AAg
BBg A
o
AA aRTgg ln
AaRT ln
: Chemical potential of ii
Solution thermodynamics
15
A-B solution (Solid or Liquid solution)
1. Ideal solution:
)lnln()( BBAA
o
BB
o
AAsoln axaxRTGxGxG
1,1 BA
)lnln()( BBAA
o
BB
o
AAsoln xxxxRTGxGxG
2. Regular solution:2ln BABA xRT
BAABBBAA
o
BB
o
AAsoln xxxxxxRTGxGxG )lnln()(
Ω: Regular solution parameter
Solution thermodynamics
16
3. General solution: ),( TxfA
1, ji
j
B
i
A
ij
AB
ex xxG
A-B solution, (Solid or Liquid solution)
)lnln()( BBAA
o
BB
o
AAsoln axaxRTGxGxG
ex
BBAA
o
BB
o
AAsoln GxxxxRTGxGxG )lnln()(
* FactSage supports many complex solution models.
Solution database (FToxid, FTSalt, ....) contains optimized
model parameters reproducing Gibbs energy of solution.
17
Gibbs Energy vs. Phase Diagram
Phase diagram is the collection of minimum Gibbs energy
assemblage of given system with temperature.
T2
Porter, D.A., and Easterling, K.E., Phase Transformation in Metals and Alloys, 2nd Ed. CHAMAN & HALL (1992)
18
10
20
30
40
50
60
70
80
90
102030405060708090
10
20
30
40
50
60
70
80
90
SiO2
CaO MgOmass percent
2018-10-22 7 secs
CaO - MgO - SiO2
1600oC, 1 atm
Ternary phase diagram: isothermal phase diagram
Liquid
Mg2SiO4
Ca2SiO4
60%SiO2-10%CaO-30%MgO
19
Ternary phase diagram: Liquidus projection
CaO
1507
1438 1429
1438
1516
1382
1331
1366
(1454)
1576
1369
1464
2400
2300
22002100
2000
1900
1800170016001500
1600
MgOCaO
P-Woll
2 liquids
Diop ppx
Aker
MerwMont
Wo
ll
2000
1923
opxPig
2500
2500
2600
MgO
2700
2400
Crist
1409
SiO2
Fors
1361
1395
Trid
Ca3Si2O7
14001366
a -Ca2SiO4
Ca3SiO5
1410
1336
mole fraction
CaSiO3
Mg2SiO4
18631962
2017
139213911461
1411
1453
(1723)
(2825)(2572)
1463
(1888)
1558
1548
1687
13631370
(2154)
1370
1388
(1392)
1370
2374
(1799)
a
(1540)
14371465
1689
2100
MgSiO3
• Primary crystalline phase
• Univariant line
• Pseudo-binary phase diagram
• Solidification pass
20
T(max)2704.66
T(min)1124.73
T(inc)100
Four-Phase Intersection Points with LIQUID
CEMENTITE / C_graphite(s) / M5C2CEMENTITE / C_graphite(s) / FCC_A1#1
1:2:
X = 100Mn/(Fe+Mn+Al+C) (g/g)Y = 100C/(Fe+Mn+Al+C) (g/g)
X Y oC
30.92494 4.85288 1147.2322.24020 4.56979 1137.02
1:2:
1147
1137
FCC
C
CEMENTITEM5C2
BCC
1100
1350
1600
1850
2100
2350
2600
2800T oC
T(max)2704.66
T(min)1124.73
1400 o
C
1300 o
C
1200 o
C
1500 o
C
Fe - Mn - Al - C
1.5 wt.% Al
wt% Mn
wt%
C
0 10 20 30 40 50
0
2
4
6
8
10
Quaternary phase diagram: Liquidus projection
2
2ln
0,
22
OA
AOo
pa
aRTG
GmEquilibriuat
AOOA
Advantage of thermodynamic database
Ellingham Diagram FactSage calculations
• Ellingham diagram : Reaction between pure stoichiometric oxide phases.
• FactSage calc.: Multicomponent phase equilibria including many solid/liquid/gas solutions.
- for example, Spinel/Slag/Monoxide/Corundum/Fe-steel/Gas/etc….
21
FactSage 7.3
(gram) 40 CaO + 30 SiO2 + 10 Al2O3 + 20
MgO + (gram) 5 Cr2O3 =
93.310 gram ASlag-liq
(93.310 gram, 1.6394 mol)
(1600 C, 1 atm, a=1.0000)
( 9.6948 wt.% Al2O3
+ 32.151 wt.% SiO2
+ 42.862 wt.% CaO
+ 14.359 wt.% MgO
+ 4.7413E-04 wt.% CrO
+ 0.93357 wt.% Cr2O3
+ 5.9255 gram ASpinel
(5.9255 gram, 3.3221E-02 mol)
(1600 C, 1 atm, a=1.0000)
( 2.4413 wt.% Al3O4[1+]
+ 8.7136E-07 wt.% Al1O4[5-]
+ 13.954 wt.% Mg1Al2O4
+ 0.84280 wt.% Al1Mg2O4[1-]
+ 4.8139 wt.% Mg3O4[2-]
+ 4.9251E-06 wt.% Mg1O4[6-]
+ 66.331 wt.% Mg1Cr2O4
+ 5.2307E-02 wt.% Cr1Cr2O4[1+]
+ 3.9894E-03 wt.% Cr1Mg2O4[1-]
+ 11.549 wt.% Al1Cr2O4[1+]
+ 5.7638 gram AMonoxide#1
(5.7638 gram, 0.13398 mol)
(1600 C, 1 atm, a=1.0000)
( 0.10016 wt.% CaO
+ 91.310 wt.% MgO
+ 0.18976 wt.% Al2O3
+ 8.3998 wt.% Cr2O3
22
CALPHAD THERMODYNAMIC DATABASE
• Thermodynamic models
• Database development
Development of Thermodynamic Database
A BGibbs energy between A-B = f (X, T, P)
in multicomponent system
Thermodynamic Database
Phase diagram data• Phase diagram
• S/L/G phase equilibria
Crystal Structural data
Thermodynamic data• Calorimetric data: Heat capacity,
H of mixing, H of melting, etc.
• Vapour pressures
• Chemical Potentials: activity
23
CALPHAD
CALPHAD type thermodynamic modeling
𝑔𝐵 𝑠𝑜 =ℎ𝑇
𝑜 -T𝑠𝑇𝑜
XB BA
T
XB
L
L+BA+L
A+B
XBi
XBi
𝑔𝐵 𝐿 =𝑔𝐵 𝐿𝑜 + 𝑅𝑇𝑙𝑛𝑎𝐵(𝐿)
𝑔𝐵 𝐿𝑜
Gib
bs e
nerg
y (
J/m
ol)
𝑔𝐴 𝐿𝑜
𝑔𝐴 𝑆𝑜
gex = f(X, T)
Commercial software and databases
1970’ 1990’1980’
Slag/Refractory/
Inclusions/Flux/
Steel
Slag:Viscosity,
Molar volume,
Thermal
Conductivity, etc.
Thermodynamic
database
Physical Property
database
Kinetic Process
Simulation models
(EERZ Concept)
- Secondary Refining Units
- Continuous Casting Process
Overall Goal of FactSage Steelmaking Consortium Project
(2009~2020)
Combining Thermodynamics
& Mass transfer based on
numerical analysis and plant
sampling data26
2017-2
020
Steel database
High alloy steels with all common elements
(steelmaking, alloy design, metallic coating)R
E o
xy-s
ulfid
es,
Hig
h V
an
d T
io
xid
e s
lag
s
Addition of C, H into oxy-fluoride systems
SF
CA
s
Steelmaking Consortium project (2009 ~ 2020)
27
FactSage DatabaseOxide database (FToxid): CaO-
MgO-FeO-Fe2O3-Al2O3-SiO2-…
FSStel and FeLq databasae
P2O5
(de-P process)
Na2O, K2O, Li2O, F, S
Oxy Fluoride system
(mold flux & refining slag)
Na
2 O-F
et O
, Mn
O,…
,
(refin
ing
, refra
cto
ry)
Ti oxides, ZrO2, B2O3
Tra
nsie
nt m
eta
l oxid
es
(Mn
, Cr, V
, Ti
oxid
es)
2014-2
017
2009-2
014
Steel database
(high Mn, Al, Si alloys)
Development of Thermodynamic Database
28
Pure compound
Calorimetry
emf
Knudsen cell
Vapor pressure
o
T
o
T
o
T TSHG
T
K
p
o
K
o
T dTCHH298
298
T
K
po
K
o
T dTT
CSS
298
298
K
K
po
K dTT
CS
298
0
298
𝐶𝑝 = 𝑎 + 𝑏𝑇 + 𝐶𝑇2 +⋯
29
Solution
emf (activity)
Knudsen cell (activity)
Vapor pressure (activity)
Solution calorimetry (enthalpy)
Phase diagram
1, ji
j
B
i
A
ij
AB
ex xxG
)lnln()( BBAA
o
BB
o
AAsoln axaxRTGxGxG
𝐺𝑠𝑜𝑙𝑛 =𝑥𝑖𝐺𝑖𝐺𝑖 = 𝐺𝑖
𝑜 + 𝑅𝑇𝑙𝑛𝑎𝑖
Binary solution
=(𝑥𝐴𝐺𝐴𝑜 + 𝑥𝐵𝐺𝐵
𝑜) − 𝑇∆𝑆𝑐𝑜𝑛𝑓 + 𝐺𝑒𝑥
Expression of Sconf and Gex is dependent on
the thermodynamic (solution) model
Development of Thermodynamic Database
Solution model (1): Bragg Williams Random Mixing model
30
ex
BBAA
Lo
BB
Lo
AA
L GXXXXRTGXGXG )lnln(,,
n
B
m
A
mn
AB
ex XXqG
Bragg Williams Random Mixing Model (BWRMM)
• Bad description of entropy of solution
Strong temperature dependence terms are required
• Many excess parameters in high order systems
• No proper interpolation technique:
Only using Muggianu interpolation technique
Solution model (2): Quasi-chemical Model
31
Modified Quasichemical Model (MQM)
)2/()( ABAB
configo
BB
o
AA gnSTGnGnG
)]2/ln(
)/ln()/ln([
)lnln(
22
BAABAB
BBBBBAAAAA
BBAA
config
YYXn
YXnYXnR
XnXnRS
)(2)()( BABBAA ABg:
1
0
1
0 )()(j
j
BB
j
AB
i
i
AA
i
ABABAB XgXggg
• Good description of entropy less T dependent parameters
• Can use proper interpolation technique
• Liquid Slags and Salts (oxide, fluoride, ….), Liquid alloys
CaO-SiO2
T = 1600oC
broken oxygen bridges
bridg
ed o
xygen
Park and Rhee, 2001free oxygenbroken oxygen bridges
bridged oxygen
free o
xygen
CaO mole fraction SiO2
Bo
nd
fra
ctio
n
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Calculated bonding in liquid silicates
O0 + O2- = 2 O-
32
Fe-Si
Fe-FeSi-Si
T=1600C
Fe mole fraction Si
Bon
d F
ract
ion
s
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Liquid Fe-Si solution
(Fe-Fe)+(Si-Si) =2(Fe-Si)
Solution model (2): Quasi-chemical Model
Bond fraction Structure Physical property models
Solution model (3): Wagner Interaction parameter formalism
33
)lnlnln(
)lnlnln(
)(
OOMMFeFe
OOMMFeFe
oOO
oMM
oFeFe
fnfnfnRT
XnXnXnRT
gngngnG
Unified (Wagner) Interaction parameter model
iiHi xfa )( ...ln k
k
ij
j
ii
i
ii xxxf
iiwti xfa .%)( ...]%[]%[]%[log kwtejwteiwtef ki
ji
iii
• Limited to dilute solution
• Many interaction parameters with increasing elements
• No interpolation technique implemented
Gibbs energy of Fe-M-O system
Solid solution (4): Compound Energy Formalism (CEF)
34
MgAl2O4 spinel: (Mg2+,Al3+)T[Mg2+,Al3+]O2O4
Al3
+in
Tetr
ahedra
l site
G =
i
j
YiTYj
OGi j – TSconfig + Gexcess
Sconfig = – R
i
YiTlnYi
T + 2
j
YjOlnYj
O
where Yi is site fraction of i in given site.
Many solid solutions are
described using CEF.
35
Solid solution (4): Compound Energy Formalism (CEF)
Many solid solutions in Steel database and Oxide database
are described using the CEF in consideration of the structure
of each solution phase.
FCC: (Fe, Mn, Ni, Cr, ….)[Va, C, N, …]
BCC: (Fe, Mn, Ni, Cr, ….)[Va, C, N, …]3…
Spinel: (Mg2+,Fe2+,Fe3+,Al3+)T[Mg2+,Fe2+,Fe3+,Al3+]2OO4
Olivine: (Mg2+,Fe2+,Mn2+)M2(Mg2+,Fe2+,Mn2+)M1O4
…
Interpolation technique for the prediction of Gibbs Energy
A = B = C A = B / C
A ~ B / CA ~ B ~ C
First approximation of ternary or higher order Gibbs energy from binary parameters
A
B C
36
Theoretically,
flexible interpolation technique can be applied to both MQM and BWRMM
Toop
Kohler
Muggianu
Interpolation technique: Dilute solute elements
37
Fe X (Al, Mn,…)
O (N, P, … )
Database development for ternary system
38
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.10.20.30.40.50.60.70.80.9
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Mg
Si Snmole fraction
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
563
758
857
200
Si
(Mg2Si-rich) (Mg2Sn-rich)
Mg2X Mg2X
Mg s.s.
1000
900
800
700
1300
1200
1100
1000
700
1400
600
Sn
Mg
Si Sn
783
563
1086
637
945
232
200
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Ternary Mg-Al-Si system
Thermodynamic Modeling: Mg-Si-Sn system
39
Vogel, 1909Raynor, 1940Gefficen and Miller, 1968Schurmann and FischerRao and Belton, 1981
Mg
2S
i
Liquid
L + Si
L + Mg2Si
637
945
1086
Mg2Si + SiMg + Mg2Si
Mg mole fraction Si
Tem
pea
ture
, oC
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
100
300
500
700
900
1100
1300
1500
Rao and Belton, 1981: emf
Eldridge, 1967: isopiestic
T =1350K
Gefficen and Miller, 1968: revision of Eldridge, 1967
Sryvalin, 1964: emf
Mg mole fraction Si
act
ivit
y
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Mannchen and Jacobi, 1965
Feufel et al., 1997: DSC
Gerstein et al., 1967
Temperature, K
Hea
t ca
paci
ty o
f M
g2S
i, J
/mol-
K
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300
0
10
20
30
40
50
60
70
80
90
100
Lukashenko and Eremenko, 1964:emf
Grjotheim et al, 1962: vapor pressure
Rao and Belton, 1981: emf
dG
dH
Kubaschewski and Villa,1949
Schneider and Stobbe, 1946
Schchukarev and Vasil'kova, 1953
Feufel et al., 1997
Temperature, K
dG
an
d d
H, k
J/m
ol:
2M
g(s
) +
Si(
s) =
Mg
2S
i(s)
200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500
-100
-90
-80
-70
-60
-50
-40
-30
T = 1350K
Gefficen and Miller, 1968
Feufel et al., 1997
SiMg
Mg mole fraction Si
pa
rtia
l en
tha
lpy
of
mix
ing
, k
J/m
ol
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Eldridge et al., 1967: isopiestic
Gefficen and Miller, 1968: revision of Eldridge et al., 1967
T = 1350K
Mg mole fraction Si
enth
alp
y o
f m
ixin
g, k
J/m
ol
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
-25
-20
-15
-10
-5
0
Consistency in data
40
Grube, 1905
Kurnakow and Stepanow, 1905
Hume-Rothery, 1926
Navak and Oelsen, 1968
Raynor, 1940
Steiner et al., 1964
Heycock and Neville, 1890
Ellmer et al., 1973
Vosskuhler, 1941
Grube and Vosskuhler, 1934
α-MgM
g2S
n
L + Sn
Liquid
Willey, 1948
200
563
783
Mg mole fraction Sn
Tem
per
atu
re,
oC
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
100
200
300
400
500
600
700
800
900
1000
Thermodynamic Modeling: Mg-Si-Sn system
41
Kawakami, 1930: claorimetry, 800oC
Eremenko and Lukashenko, 1963: emf, 650oC
Steiner et al, 1964: vapor, 770oC
Eldridge et al, 1967: vapor, 770oC
Sharma, 1970: vapor, 800oC
Nayak and Oelsen, 1971: calorimetry, 800oC
Sommer et al, 1980: calorimetry, 800oC
800oC
650oC
Mg mole fraction Sn
En
tha
lpy
of
mix
ing
, k
J/m
ol
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
-20
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
2
Kubaschewski, 1939
Kubaschewski and Evans, 1956
Beardmore et al., 1966
Nayak and Oelsen, 1971
Borsese et al., 1975
Sommer et al, 1980
Temperature, K
dH
, 2M
g(s
) +
Sn
(s)
-> M
g2S
n(s
), k
J/m
ol
200 300 400 500 600 700 800 900 1000 1100 1200
-90
-80
-70
-60
-50
-40
-30
Sn(l)Mg(l)
Eldridge et al., 1966: Isopiestic
Moser et al., 1990: EMF
Eckert et al., 1982: EMF
Sharma, 1970: EMF
Egan, 1974, EMF
T = 1073K
Mg mole fraction Sn
act
ivit
y
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Sommer et al., 1980
Temperature, K
HM
g2S
n (
T) -
HM
g2S
n (
98
3K
), k
J/m
ol
800 900 1000 1100 1200 1300
-20
0
20
40
60
80
100
120
dH-Liquid dH-Mg2Sn
activity-Liquid dHL-Mg2Sn
Thermodynamic Modeling: Mg-Si-Sn system
42
Mg
Vogel, 1909
Raynor, 1940
Gefficen and M
iller, 1968
Schurm
ann and Fischer
Rao and B
elton, 1981
Mg2
Si
Liquid
L + Si
L + Mg2S
i
637
945
1086
Mg2S
i + Si
Mg + M
g2S
iM
g mole fraction Si
Tempeatu
re, oC
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
100
300
500
700
900
1100
1300
1500
Grube, 1905
Kurnakow
and Stepanow
, 1905
Hum
e-Rothery, 1926
Navak and O
elsen, 1968
Raynor, 1940
Steiner et al., 1964
Heycock and N
eville, 1890
Ellm
er et al., 1973
Vosskuhler, 1941
Grube and V
osskuhler, 1934
α-Mg
Mg 2
Sn
L + S
nLiq
uid
Willey, 1948
200563 783
Mg m
ole fraction Sn
Temper
ature
, o C
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
100
200
300
400
500
600
700
800
900
1000
232
Liquid
L + Si
Si + Sn
Si mole fraction Sn
Temperature, o C
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
100
300
500
700
900
1100
1300
1500
Si Sn
Toop
Thermodynamic Modeling: Mg-Si-Sn system
(--)(--)
(~0)
43
Nikitin et al., 1968
Muntyanu et al., 1966
Mg2Sn-rich
Mg2Si-rich
Liquid
L + Mg2Si-rich
Mg2Sn-rich + Mg2Si-rich
857
Mg2Sn mole fraction Mg2Si
Tem
pera
ture,
oC
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
100
200
300
400
500
600
700
800
900
1000
1100
1200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.10.20.30.40.50.60.70.80.9
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Mg
Si Snmole fraction
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
563
758
857
200
Si
(Mg2Si-rich) (Mg2Sn-rich)
Mg2X Mg2X
Mg s.s.
1000
900
800
700
1300
1200
1100
1000
700
1400
600
Sn
Mg
Si Sn
783
563
1086
637
945
232
200
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
Mg
Si Sn
• Liquid solution: Two ternary
excess terms
• Solid solution: Mg2(Sn,Si)
Thermodynamic Modeling: Mg-Si-Sn system
44
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.10.20.30.40.50.60.70.80.9
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Ag
Cu Zrmole fraction
P1
P2 P4P5
P6
M2
M2
P7
P8
E1
P9
E2
E3
(Ag)
m
2 Liquids
AgZr
(Ag,Cu)Zr2
(Zr)CuZr
Cu51Zr14
(Ag)
(Cu)
P3
M1
M1
T in oC
P2: 949.5
P3: 948.5
P5: 909.9
P6: 905.8
P4: 921.2
P7: 894.9
P8: 885.8
P9: 863.8
M1: 905.8
M2: 898.0
E1: 882.2
E2: 855.6
E3: 779.6
P1: 966.7
1000
1100
1600
14001
200
900
1000
1100
1083
(--)
(+) (-)
Stable liquid miscibility gap (2 liquids)
Amorphous phase – Ag-rich segregation
Thermodynamic Modeling: Ag-Cu-Zr system
45
Thermodynamic modeling of Zn-Al-Fe system
Fe
Al ZnRepulsion
?
FactSage FSstel database
Nakano et al. (2007)
Luo et al. (2012)
Present work
46
Thermodynamic modeling of Zn-Al-Fe system
47
Thermodynamic modeling of Zn-Al-Fe system
Liquid
Liquid + BCC_B2
Liquid + BCC_B2 + Fe2Al5(Zn,Va)3
Liquid + Fe2Al5(Zn,Va)3 + FeAl2
FEZN_DELTA + Fe2Al5(Zn,Va)3 + FeAl2
HCP_Zn + FEZN_DELTA
+ Fe2Al5(Zn,Va)3Al13Fe4 + Fe2Al5(Zn,Va)3 + FeAl2
Liquid + BCC_B2 + FeAl2
HCP_Zn + Al13Fe4
+ Fe2Al5(Zn,Va)3
Liquid + FEZN_DELTA + Fe2Al5(Zn,Va)3
wt. %
Fe
Al Zn
Al - Zn - FeNew (MQM)
Tem
per
atu
re (
oC
)
0
200
400
600
800
1000
1200
1400
1600
1800
2000
wt. fraction of Zn0 0.2 0.4 0.6 0.8 1
48
Predictive ability of Thermodynamic model
Thermodynamic model should
have good predictive ability !!
-3
-2
-1
0
-2 2-4
T = 1600°C
0 1-1
L+MnAl2O4(s)
Al oxidation by MnO
Al deoxidation with CaO
Al oxidation by Fe2O3
L
log [pct Al]
log [
pct
O]
L+Al2O3(s)
15.1<[pct Mn]<21.9Al deoxidation Associate model
Present study
WIPF
[pct Mn]=20
*Associate model:
FACTSAGE FeLQ database
-3
-2
-1
-4 -3 -1 0 1 2-2-4
0
L+Al2O3(s)
L
log
[%
O]
log [%Al]
[pp
m O
]
10000
1
2
5
10
20
50
100
200
500
1000
T = 1600°C
Rohde et al. (1971)Fruehan (1970)Janke & Fischer (1976)Dimitrov et al. (1995)
Seo et al. (1998)Swisher, 1580°C (1967)
Suito et al. (1991)Paek et al. (2015)
Kang et al. (2009)
Shevtsov (1981)
log [
%O
]
log [%Mn]
-2
-1
0
-3 -2 -1 0 1 2-3
L+(FexO-MnO)(s.s.)
1600°C, Shevtsov et al. (1987)1600°C, Janke & Fischer (1976)
1550°C
1650°C1600°C Takahashi & Hino (2000)
1550°C
1650°C1600°C
Dimitrov et al. (1995)1600°C1550°C
Hino et al. (1990)
Chipman et al. (1950)
1587-1620°C
1680-1723°C1630-1655°C
1550°C
1650°C1600°C Hilty & Crafts (1950)
Linchevskii & Samarin (1954)1605°C1565°C
L+slag(l)
L
1700°C1650°C1600°C
1550°C
Fe-Mn-O
Present study: MQM
Predicted from Al- and Mn-deoxidation curve
49
APPLICATIONS TO STEELMAKING
PROCESS AND ALLOY DESIGN
• Simple calculations
• Advanced applications
Current thermodynamic databases for steel and Oxides
50
• Steel database for alloy development: Physical metallurgy
Thermo-Calc: TCFE database
FactSage: FSStel database (up to 72 version)
Bragg Williams Random Mixing Model (sub-regular solution models): Liquid solution
Compound Energy Formalism: Solid solutions
• Dedicated database for Molten steel for refining process
FactSage: FTmisc FeLq database
Unified Interaction parameter model (Modified Wagner model)
• New steel database – MQM for liquid solution
FactSage: FSStel database for steel (73 version)
FTlite database for Al and Mg alloys
• FTOxid database (FactSage):
Oxide database containing
CaO-MgO-Al2O3-SiO2-FeOx-MnO-TiOx-CrOx-P2O5…S…F…
51
Accuracy of new database: Fe-Mn-Si-Al-C system
Persoly et al., IOC conf. Series, vol. 119, 2016, 012013.
High Temperature Thermochemistry Laboratory
52Accuracy of new database: Fe-Mn-Si-Al-C system
2019-02-12 5 secs
LIQUID
BCC_A2 + LIQUID
FCC_A1
FCC_A1 + LIQUIDBCC_A2
Fe - Mn - Si - C100Mn/Z (g/g)=0.37, 100Si/Z (g/g)=0.07,
Z=(Fe+Mn+Si+C), 1 atm
100C/(Fe+Mn+Si+C) (g/g)
T(K
)
0 0.05 0.1 0.15 0.2 0.25 0.3
1600
1650
1700
1750
1800
1850
1900
2019-02-12 6 secs
LIQUID
FCC_A1
FCC_A1 + LIQUID
BCC_A2
Fe - Mn - Si - C
100Mn/Z (g/g) = 2, 100Si/Z (g/g) = 2, Z=(Fe+Mn+Si+C), 1 atm
100C/(Fe+Mn+Si+C) (g/g)
T(K
)
0 0.05 0.1 0.15 0.2 0.25 0.3
1600
1650
1700
1750
1800
1850
1900
Fe-C-0.37Mn-0.07Si Fe-C-2Mn-2Si
Red-line is new FS73 database;
I believe this assessment work by Chinese group (labled as this work) is more or less taken by
ThermoCalc-TCFE9
High Temperature Thermochemistry Laboratory
53Accuracy of new database: Fe-Mn-Si-Al-C system
2019-02-12 6 secs
Fe - Mn - Si - C
100Mn/Z (g/g) = 2, 100Si/Z (g/g) = 2, Z=(Fe+Mn+Si+C), 1 atm
100C/(Fe+Mn+Si+C) (g/g)
T(C
)
0 0.05 0.1 0.15 0.2 0.25 0.3
1350
1370
1390
1410
1430
1450
1470
1490
1510 Thick red line: FS73 database
Data from Presoly et al. (DTA; 2016~2017); Our group (Minkyu Paek now at Aalto Univ.) performed the same DSC
experiments for a couple of samples and confirmed these data
High Temperature Thermochemistry Laboratory
54Accuracy of new database: Fe-Mn-Si-Al-C system
Previously database cannot reproduce the liquidus and solidus & BCC/FCC transition
Data from Presoly et al. (DTA; 2016~2017); Our group (Minkyu Paek now at Aalto Univ.) performed the same DSC
experiments for a couple of samples and confirmed these data
2019-03-13 6 secs2019-03-09 9 secs
Fe - Mn - Si - C
100Si/Z (g/g)=3, 100C/Z (g/g)=0.01, Z=(Fe+Mn+Si+C), 1 atm
100Mn/(Fe+Mn+Si+C) (g/g)
T(C
)
0 5 10 15 20 25
1200
1250
1300
1350
1400
1450
1500
Secondary Steelmaking Process
- Ladle – Deoxidation
- LF – Mn alloying process
- RH – Degassing (gas removal – N, H, …)
- Refractory of ladle
Casting
- Solidification: Liquidus projection, Scheil cooling,…
- Defect control: AlN formation
Applications: TWIP Steel Production
Al deoxidation equilibria in Fe-Mn-Al-O melt
56
-3
-2
-1
0
-2 2-4
T = 1600°C
0 1-1
L+MnAl2O4(s)
Al oxidation by MnO
Al deoxidation with CaO
Al oxidation by Fe2O3
L
log [pct Al]
log [
pct
O]
L+Al2O3(s)
15.1<[pct Mn]<21.9Al deoxidation Associate model
Present study
WIPF
[pct Mn]=20
Al deoxidation equilibria in Fe-Mn-Al-O melt
57
Stability diagram (MQM)
-5 2
T=1600°C
-4 -1 0 1-3
MnAl2O4
Al2O3
Dimitrov et al. (1995)
MnO
5102050200 2 [mass ppm O] 3 5100 2
-2
20 400 3 10
10
156 160
75
37
30
15
9
7
7 4 3113
3 34
4 45
81076110102
32815
57
43
4
11 344 4
5
4
3 4
5
Park et al. (2012)
Paek et al. (2015)
0
1
2
-2
-1
200
200
Ogasawara et al. (2012)
log
[%
Mn
]
log [%Al]
New steel database
LF – addition of FeMn alloy
58
Liquid Fe at 1600C + FeMn (75%Mn-24%Fe-1%C) at 25 oC
High Temperature Thermochemistry Laboratory
59Vacuum treatment: RH degassor / Tank Degassor
Al Al Al Al
Fe Fe Fe Fe
Si Si Si Si
MnMn Mn Mn
wt.% Mn
log
10(a
ctiv
ity)
0 5 10 15 20 25 30
-10.0
-8.0
-6.0
-4.0
-2.0
0.0
High vapor pressure of Mn Degassing limit ??
Lo
g P
(atm
)
Fe-Mn-1.5%Al-0.5%C
Fe-Mn-1.5%Si-0.5%C
T = 1600oC
1torr
High Temperature Thermochemistry Laboratory
60Vacuum treatment: RH degassor / Tank Degassor
Mn(g)
Temperature, oC
log
P(a
tm)
1450 1500 1550 1600
-5
-4
-3
-2
-1
0
1torr
Fe-Mn-1.5%Al-0.5%C
High Temperature Thermochemistry Laboratory
61Refractory: Ladle refractory / liquid steel rxn
Steel
Al2O3 weight percent MgO
Co
mp
osit
ion
, g
ram
0 10 20 30 40 50 60 70 80 90 1000
20
40
60
80
100
Spinel
MgAl2O4-MnAl2O4
Al2O3
TWIP
Steel
Al2O3-MgO
Refractory
Stable spinel layer
Negligible change in
steel composition
Enhancement of spinel matrix
MgAl2O4 MgAl2O4-MnAl2O4
Thermodynamic Calculation
High alumina spinel castable
(90%Al2O3-10%MgO)
High Temperature Thermochemistry Laboratory
62Casting process
Submerged Entry Nozzle (SEN)
- Reoxidation of liquid steel
Mold
- Solidification range of steel
- Mold flux design
- Surface quality (AlN)
High Temperature Thermochemistry Laboratory
63Liquidus projection of Fe-Mn-1.5%Al-C
T(max)2704.66
T(min)1124.73
T(inc)100
Four-Phase Intersection Points with LIQUID
CEMENTITE / C_graphite(s) / M5C2CEMENTITE / C_graphite(s) / FCC_A1#1
1:2:
X = 100Mn/(Fe+Mn+Al+C) (g/g)Y = 100C/(Fe+Mn+Al+C) (g/g)
X Y oC
30.92494 4.85288 1147.2322.24020 4.56979 1137.02
1:2:
1147
1137
FCC
C
CEMENTITEM5C2
BCC
1100
1350
1600
1850
2100
2350
2600
2800T oC
T(max)2704.66
T(min)1124.73
1400 o
C
1300 o
C
1200 o
C
1500 o
C
Fe - Mn - Al - C
1.5 wt.% Al
wt% Mn
wt%
C
0 10 20 30 40 50
0
2
4
6
8
10
High Temperature Thermochemistry Laboratory
64Solidification range of high Mn steels
LIQUID
FCC_A1 + LIQUID
FCC_A1
BCC_A2 + LIQUID
wt% Mn
Tem
per
atu
re,
oC
0 5 10 15 20 25 30
1200
1250
1300
1350
1400
1450
1500
1550
1600
Fe-Mn-1.5%Si-0.5%C
LIQUID
FCC_A1 + LIQUID
FCC_A1
BCC_A2 + LIQUID
wt% Mn
Tem
per
atu
re,
oC
0 5 10 15 20 25 30
1200
1250
1300
1350
1400
1450
1500
1550
1600
Fe-Mn-1.5%Al-0.5%C
Fe-Mn-1.5%Si-0.5%C
High Temperature Thermochemistry Laboratory
65Scheil solidification
Liquid
Cementite
FCC
Temperature, oC
ph
ase
dis
trib
uti
on
, w
t%
1100 1150 1200 1250 1300 1350 1400 1450 1500
0
10
20
30
40
50
60
70
80
90
100
Fe-Mn-1.5%Al-0.5%C
Scheil cooling: complete diffusion in liquid and no diffusion in solids
High Temperature Thermochemistry Laboratory
66Nozzle Clogging - SEN
Ar+CO
Al2O3+CAr gas
Liquid steel
(Al-killed)
2[Al] + 3CO(g) =
Al2O3 + 3[C]
2Al2O3+9C = Al4C3+6CO(g)
Al2O3 + C
C:\FactSage73\Equi0.res 10Jul19
Temperature, oC
log
PC
O(a
tm)
1200 1300 1400 1500 1600
-6
-5
-4
-3
-2
-1
0
1550oC1500oC
1450oC
Formation of Al2O3 in Nozzle: Al killed steel
67
Ar+CO
Al2O3+C
Ar gas
Liquid steel
(Al-killed)
2[Al] + 3CO(g) =
Al2O3 + 3[C]
Fe-liq + Al2O3
Fe-liq
Fe - Al - CO
wt% Al
log
10 p
(CO
)/atm
0 0.02 0.04 0.06 0.08 0.1
-5
-4
-3
-2
-1
0
Al2O3 formation
2Al2O3+9C = Al4C3+6CO(g)
Nozzle clogging in Al-Ti killed steel
68
Fe-liq + Al2O3
Fe-liq + Al2O3 + Slag
Fe-liq + Slag
Fe-liq + Ti3O5
Fe
-liq
+ T
i 3O
5
Fe-liq
+ Slag
Fe - Al - Ti - CO
Ti = 1000 ppm, 1550oC
wt% Al
log
10 p
(CO
)/atm
0 0.02 0.04 0.06 0.08 0.1
-5
-4
-3
-2
-1
0
Reoxidation of steel by CO gas through ceramic nozzle to form slag(Al-Ti-O) and Al2O3
Nozzle clogging in high Mn Steel
69
LIQUID
L + Corundum (Al2O3)
L + Spinel (MnAl2O4)
L + FeO + Spinel
Fe - Mn - Al - C - CO100Mn/Z (g/g)=20, 100C/Z (g/g)=0.5,
Z=(Fe+Mn+Al+C), 1500oC, 1 atm
wt% Al
log
10 p
(CO
)/a
tm
0 0.5 1 1.5
-4
-3
-2
-1
0
LIQUID
L + Corundum (Al2O3)
L + Spinel (MnAl2O4)
L + FeO + Spinel
1500oC
1450oC
Fe - Mn - Al - C - CO
wt% Al
log
10 p
(CO
)/a
tm
0.0 0.5 1.0 1.5
-4
-3
-2
-1
0
Fe-20wt%Mn-1.5%Al-0.5%C
AlN solubility product in TWIP steel
70
[%N
]
0.04
0.08
0
0.06
0.02
0 0.5 2.5
[%Al]31.51 2
1550°C1500°C1450°C
L
L+AlN(s)
Jang et al. (2015)
1500°C1450°C
0.58<[%C]<0.59
AlN saturation
1550°C
23.5<[%Mn]<24.0
0.32<[%Si]<0.34
Fe-24%Mn-0.3%Si-0.6%C-Al-N
AlN: bad for surface quality
4AlN + 3SiO2(mold flux) = 2Al2O3(mold flux) + 2N2(g)
High Temperature Thermochemistry Laboratory
71AlN formation
200 ppm [N] (Liquid)
Liquid
CEME
541 ppm
473 ppm
Solid lines: 200 ppm [N]
77.58 Fe + 20 Mn + 0.6 C + 1.5 Al + 0.3 Si + (0.02) N
T(oC)
gra
m
1100 1200 1300 1400 1500 1600 1700 1800
0.00
0.02
0.04
0.06
0.08
FCC
249 ppm AlN (secondary)
182 ppm
Dash lines: 100 ppm [N]
T(oC)1100 1200 1300 1400 1500 1600 1700 1800
0.00
0.02
0.04
0.06
0.08
100 ppm [N] (Liquid)
(0.01) N
40 ppm [N] (Liquid)
74 ppm AlN (secondary)
Dotted lines: 40 ppm [N]
T(oC)1100 1200 1300 1400 1500 1600 1700 1800
0.00
0.02
0.04
0.06
0.08
(0.004) N
Mold flux design
Liquid
L +
SiO
2L
+ S
iO2 +
Ca
F2
L + CaSiO3
L + CaSiO3(s2)
L + Ca4Si2F2O7
L +
1:2
:3 +
Ca
4S
i 2F
2O
7
L + Ca2SiO4
L + CaO
70
60
50
CaO - SiO2 - Na2O - CaF2
Na2O/Z (g/g) = 0.1, CaF
2/Z (g/g) = 0.2,
Z=(CaO+SiO2+Na
2O+CaF
2)
SiO2/(CaO+SiO2+Na2O+CaF2) (g/g)
Tem
per
atu
re,
oC
0 0.1 0.2 0.3 0.4 0.5 0.6
900
950
1000
1050
1100
1150
1200
1250
1300
0.7
5
C/S
= 1
.33
1.0
Thermodynamic Modeling of oxy-fluoride system 72
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.10.20.30.40.50.60.70.80.9
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
SiO2
CaO CaF2mass fraction
T(min) = 1088.53 oC, T(max) = 2571.99
oC
Four-Phase Intersection Points w ith Slag-liq
Ca3SiO5_hatrurite(s) / CaO_lime(s) / a-Ca2SiO4SiO2_cristobalite(h)(s6) / SiO2_tridymite(h)(s4) / Slag-liq#2Ca3Si2O7_rankinite(s) / a'Ca2SiO4 / a-Ca2SiO4CaO_lime(s) / a'Ca2SiO4 / a-Ca2SiO4Ca3Si2O7_rankinite(s) / Ca4Si2F2O7_-1237159.7(s) / a'Ca2SiO4Ca3Si2O7_rankinite(s) / Ca4Si2F2O7_-1237159.7(s) / CaSiO3_ps-w ollastoni(s2)CaF21 / SiO2_tridymite(h)(s4) / Slag-liq#2CaF2 / CaF21 / CaO_lime(s)
CaF2 / CaF21 / SiO2_tridymite(h)(s4)Ca4Si2F2O7_-1237159.7(s) / CaF2 / CaF21Ca4Si2F2O7_-1237159.7(s) / CaF2 / CaF21Ca4Si2F2O7_-1237159.7(s) / CaF2 / a'Ca2SiO4CaF2 / CaO_lime(s) / a'Ca2SiO4CaSiO3_ps-w ollastoni(s2) / CaSiO3_w ollastonite(s) / SiO2_tridymite(h)(s4)Ca4Si2F2O7_-1237159.7(s) / CaSiO3_ps-w ollastoni(s2) / CaSiO3_w ollastonite(s)Ca4Si2F2O7_-1237159.7(s) / CaF2 / CaSiO3_w ollastonite(s)CaF2 / CaSiO3_w ollastonite(s) / SiO2_tridymite(h)(s4)
1:2:3:4:5:6:7:8:
9:10:11:12:13:14:15:16:17:
A = SiO2, B = CaO, C = CaF2W(A) W(B) W(C)
oC
0.25828 0.54115 0.20058 1512.880.53748 0.12647 0.33605 1465.350.42414 0.55347 0.02239 1436.410.24802 0.51256 0.23943 1423.740.37886 0.52450 0.09664 1350.630.39702 0.50527 0.09771 1342.790.26278 0.00902 0.72820 1254.49
0.17191 0.34981 0.47827 1150.850.33325 0.14396 0.52280 1150.850.28120 0.21987 0.49893 1150.850.18313 0.33487 0.48200 1150.840.18985 0.35002 0.46013 1135.830.18493 0.37029 0.44478 1126.350.41021 0.23067 0.35912 1125.310.33547 0.23246 0.43207 1125.310.32609 0.22375 0.45016 1113.790.36804 0.20631 0.42565 1088.53
1:2:3:4:5:6:7:
8:9:
10:11:12:13:14:15:16:17:
1
2
3
4
56
7
8
910
1112
13
14
15 16
17
Ca3SiO5(s)
a-Ca2SiO4
Ca3Si2O7(s)
CaSiO3(s2)SiO2(s4)
SiO2(s6)
SiO2(s6)
CaF21
CaO(s)
Ca4Si2F2O7(s)
CaF2
a'Ca2SiO4
Slag-liq#2
CaSiO3(s)
High Temperature Thermochemistry Laboratory
Mold flux: Equilibrium solidification
Ca4Si2F2O7
Liquid
Nepheline
Experimental dataLiquid measuredCuspidine measuredNepheline measured
22.3 CaF2 + 9.2 Na2O + 4.4 MgO + 3.8 Al2O3 + 33.6 SiO2 + 26.2 CaO
Temperature, oC
ph
ase
dis
trib
uti
on
(w
t. %
)
900 1000 1100 1200 1300 1400
0
10
20
30
40
50
60
70
80
90
100
High Temperature Thermochemistry Laboratory
74Mold flux design: Na2O-based flux
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.10.20.30.40.50.60.70.80.9
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
SiO2
CaO CaF2mass fractions /(SiO2+CaO+CaF2)
T(min) = 829.44 oC, T(max) = 2292.83
oCFour-Phase Intersection Points with Slag-liq
Combeite / Na2Ca2Si2O7_NC2S2(s) / a'(Ca,Ba)2SiO4Combeite / Na2Ca2Si2O7_NC2S2(s) / Na2CaSiO4_NCS(s)
Ca4Si2F2O7_cuspidine(s) / CaF2-LT / CombeiteCa4Si2F2O7_cuspidine(s) / Combeite / Na2CaSiO4_NCS(s)Ca4Si2F2O7_cuspidine(s) / Ca5Si2F2O8_s1(s) / Na2CaSiO4_NCS(s)
Ca5Si2F2O8_s1(s) / Monoxide#1 / Na2CaSiO4_NCS(s)CaF2-LT / Combeite / Na2Ca3Si6O16_NC3S6(s)
Ca4Si2F2O7_cuspidine(s) / Ca5Si2F2O8_s1(s) / CaF2-LT
Ca5Si2F2O8_s1(s) / CaF2-LT / Monoxide#1Na2Ca3Si6O16_NC3S6(s) / SiO2_Quartz(h)(s2) / SiO2_Tridymite(h)(s4)CaF2-LT / Na2Ca3Si6O16_NC3S6(s) / SiO2_Quartz(h)(s2)
1:2:
3:4:5:
6:7:
8:
9:10:11:
A = SiO2, B = CaO, C = CaF2
W(A) W(B) W(C) oC
0.50682 0.49153 0.00165 1164.780.42784 0.46037 0.11179 1051.060.34975 0.04060 0.60965 1022.21
0.37368 0.39304 0.23328 1004.51
0.28351 0.29134 0.42514 947.290.26937 0.28675 0.44387 944.140.68893 0.07611 0.23495 931.750.17335 0.09411 0.73253 914.360.17075 0.10345 0.72580 912.450.90487 0.06291 0.03222 866.94
0.82308 0.03029 0.14662 829.44
1:2:3:
4:
5:6:7:8:9:
10:
11:
1
2
3
4
56
7
89
10
11
1
11
0
500
1000
1500
2000
2500
3000
3500
T oC
Na2CaSiO4(s)
Na2Ca2Si2O7(s)
Combeite
Na2Ca3Si6O16(s)
SiO2(s4)
SiO2(s2)
CaF2-LT
Ca4Si2F2O7(s)
CaF2-LTMonoxide
Ca5Si2F2O8(s)
SiO2 - CaO - CaF2 - Na2OProjection (Slag-liq), Na
2O/(SiO
2+CaO+CaF
2) (g/g) = 0.25,
1 atm
20 wt % Na2O
cuspidine
High Temperature Thermochemistry Laboratory
75Mold flux design: Li2O-based flux
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.10.20.30.40.50.60.70.80.9
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
SiO2
CaO CaF2mass fractions /(SiO2+CaO+CaF2)
T(min) = 902.52 oC, T(max) = 2299.98
oCFour-Phase Intersection Points with Slag-liq
Ca3Si2O7_Rankinite(s) / Ca4Si2F2O7_cuspidine(s) / CaSiO3_Wollastonite(s)Ca2SiO4_Alpha-prime(s2) / Ca3Si2O7_Rankinite(s) / Li2CaSiO4_S1(s)
Ca4Si2F2O7_cuspidine(s) / CaF2-LT / CaSiO3_Wollastonite(s)Ca3Si2O7_Rankinite(s) / Ca4Si2F2O7_cuspidine(s) / Li2CaSiO4_S1(s)Ca4Si2F2O7_cuspidine(s) / Ca5Si2F2O8_s1(s) / Li2CaSiO4_S1(s)
CaF2-LT / CaSiO3_Wollastonite(s) / SiO2_Tridymite(h)(s4)Ca5Si2F2O8_s1(s) / CaO_Lime(s) / Li2CaSiO4_S1(s)
Ca4Si2F2O7_cuspidine(s) / Ca5Si2F2O8_s1(s) / CaF2-LTCa5Si2F2O8_s1(s) / CaF2-LT / CaO_Lime(s)
1:2:
3:4:5:
6:7:
8:9:
A = SiO2, B = CaO, C = CaF2W(A) W(B) W(C)
oC
0.49357 0.39369 0.11275 1058.98
0.43194 0.55557 0.01249 969.02
0.35925 0.07607 0.56467 952.610.39746 0.50342 0.09912 923.890.37757 0.48352 0.13891 910.580.50398 0.05019 0.44583 908.65
0.36549 0.49592 0.13858 905.590.21909 0.20092 0.57999 904.75
0.20941 0.22198 0.56861 902.52
1:
2:
3:4:5:6:
7:8:
9:
1
2
3
45
6
7
89
1
9
0
500
1000
1500
2000
2500
3000
3500
T oC
Li2CaSiO4(s)
Ca2SiO4(s2)
Ca3Si2O7(s)
CaSiO3(s)
CaSiO3(s2)
CaSiO3(s) SiO2(s4)
CaF2-LT
CaO(s)
Ca4Si2F2O7(s)
Ca5Si2F2O8(s)
SiO2 - CaO - CaF2 - Li2OProjection (Slag-liq), Li
2O/(SiO
2+CaO+CaF
2) (g/g) = 0.111111111,
1 atm
10 wt % Li2O
cuspidine
High Temperature Thermochemistry Laboratory
76Reaction between Mold flux and Liquid steel
Steel
Mold flux Typical mold flux:
CaO-SiO2-Na2O-CaF2
Fe-Mn-Si-Al-C alloy
4[Al] + 3(SiO2) = 2(Al2O3) + 3[Si]
Addition of MnO2 to minimize the change of mold flux chemistry
Typical mass ratio between mold flux and liquid steel = 1:20 ~ 1:10
High Temperature Thermochemistry Laboratory
77Reaction between Mold flux and Liquid steel
[wt.% Si]
100 x [wt.% Al]
[wt%.Mn]
Al content, wt.% (original in liquid steel)
wt.
% s
olu
te (
aft
er
rea
ctio
n w
ith
mo
ld f
lux)
0.0 0.2 0.4 0.6 0.8 1.0
0
1
2
3
4
Fe
Mn
Ca
Si
Al
Mg
Na
F
Al content, wt% (originally in liquid steel)m
old
flu
x e
lem
en
t w
t.%
aft
er
rea
ctio
n w
ith
ste
el
0 0.20 0.40 0.60 0.80 1.00
0
05
10
15
20
25
30
100 gram of liquid steel (Fe-3%Mn-xAl) + 5 gram of mold flux + 0.5 gram of MnO2
Before MnO2 (MnO) is completely consumed by Al, reduction of SiO2 begins
Simply due to the equilibrium of:
2(MnO) + Si = 2Mn + (SiO2)
3(MnO) + 2Al = 3Mn + (Al2O3)
High Temperature Thermochemistry Laboratory
78Phase diagram: Fe-Mn-1.5Al-0.6C steel (wt.%)
LIQUID
FCC_A1 + LIQUID
FCC_A1
CEMENTITE + FCC_A1
BCC_A2 + CEMENTITE + FCC_A1
BCC_A2 + FCC_A1
BCC_A2 + CEMENTITE
BCC_A2 + LIQUID
wt.% Mn
Tem
per
atu
re,
oC
0 5 10 15 20 25 30
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
High Temperature Thermochemistry Laboratory
79Phase diagram: Development of Medium Mn steel
Fe-X%Mn-3Al-3Si-0.2C (X= 7.5 & 10)
FCC
BCC
FCCBCC
FCC
BCC
1500
1600
700
1400
1300
1200
1000
900
1100
800
Tem
per
atu
re (°C
)
208
[%Mn]0 4 12 16
BCC+FCC
L
FCC
BCC+C
BCC+FCC
Sun & Yue (2015)Present studyFSStel
High Temperature Thermochemistry Laboratory
80Phase diagram: Development of Medium Mn steel
Fe-X%Mn-3Al-3Si-0.2C (X= 7.5 & 10)
FCC
BCC
FCCBCC
FCC
BCC
20
100
Ph
ase
fra
ctio
n o
f F
CC
(%
)
Temperature (°C)1200800 900 1000 1100
80
60
40
7%Mn
Sun & Yue (2015)
10%MnPresent studyFSStel
10%Mn
7%Mn
81
Metastable phase transformation at rolling process
Jonas et al. Materials Science Forum, 2017, vol. 879, p. 29-35
Static condition: Equilibrium Dynamic condition: Non-equilibrium
Deformation of -austanite
Shear accommodation + Dilatation energy
to -austanite
de-stabilization of -austanite
82
Galvanizing process
Zn bath temperature: ~460 oC
typically 463-465 oC
Line speed: up to 200 m min-1
Immersion time: 2-8 seconds
Galvannealing process
Induction or
gas heating furnace
to raise strip
temperature to 500-
565 oC
Zn bath (~460 oC)
Air knives
Holding furnace
500-565 oC
10 seconds
(Zn + 0.15-0.20% Al)
(Zn + 0.11-0.14% Al)
Coating layer:
(Zn +
10%
Fe
)
Continuous galvanizing and galvannealing line
83
xMn + ySi +(x+2y)O MnxSiyOx+2y
Mn Si
Metal
Gas
H2O H2 + 1/2O2
(a) (b)
(c)
(d)
In the early stage of the internal/external oxidation :
• Mn and Si contents are enough to form oxides on the surface of steel
Oxidation reaction can be controlled by
(1) supply of O2 gas (c)
(2) formation of MnxSiyOx+2y oxides on the steel surface (d)
Reducing gas atmosphere using an N2-H2 mixture
Annealing furnace: Oxidation of steel in reducing condition
84
B + F F
F + S + MS
B + F + MS + S
B + F
+ M
S
B +
Cem
entite
+
MS
+ S F +
S
B + F + S
Temperature, oC
log
10(p
(O2))
(b
ar)
700 750 800 850 900 950 1000-32
-31
-30
-29
-28
-27
-26
-25
-24
-23
-22
Fe-C-1.5%Mn-1.5%Si-0.08%C
B, F, MS and S stand for BCC, FCC, MnSiO3 and SiO2
Oxidation of steel: PO2-Temperature diagram
85
Sludge
Dross refractory
Zn-Al-Fe….
Annealing: coating/steel rxnRolls
Dross and Sludge formation
86
New coating system: ZincMagnesium coating Zn-Mg-Al
ZM Stroncoat® in Salzgitter
Liquid
HCP-Zn
Mg2Zn11FCC-Al
Mg2Zn11
HCP-Zn
97 Zn + 1.5 Al + 1.5 Mg
T(C)
ph
ase
dis
trib
utio
n (
wt%
)
330 340 350 360 370 380 390 400
0
10
20
30
40
50
60
70
80
90
100
Stroncoat
Eu
tectic r
xn
T. Koll: 18th Galvanising & Coil Coating Conf., 2013
MgZn2
FCC-Al
Mg2Zn11
Liquid
93 Zn + 4 Al + 3 Mg
T(C)p
ha
se
dis
trib
utio
n (
wt%
)330 340 350 360 370 380 390 400
0
10
20
30
40
50
60
70
80
90
100
Zagnelis™
As cast microstructure
Scheil cooling calculation
High Temperature Thermochemistry Laboratory
87Database toward Super Steel: order/disorder FCC & BCC
“Here we show that an FeAl-type brittle but hard
intermetallic compound (B2) can be effectively
used as a strengthening second phase in high-
aluminium low-density steel, while alleviating its
harmful effect on ductility by controlling its
morphology and dispersion.”
High Temperature Thermochemistry Laboratory
88Fe-16.3Mn-8.2Al-4.8Ni-C steel
BCC_B2 + FCC_A1 + Kappa-Carbide
BCC_B2 + FCC_A1
FCC_A1
LIQUID
BCC_A2 + LIQUID
BCC_A2 + FCC_A1
BCC_B2 + C + FCC_A1 + Kappa-Carbide
FCC_A1 + LIQUID
BCC_A2 + BCC_B2 + FCC_A1
BCC_A2
Fe - Mn - Al - Ni - C100Mn/Z (g/g)=16.3, 100Al/Z (g/g)=8.2, 100Ni/Z
(g/g)=4.8, Z=(Fe+Mn+Al+Ni+C), 1 atm
100C/(Fe+Mn+Al+Ni+C) (g/g)
T(C
)
0 0.5 1 1.5 2 2.5
700
800
900
1000
1100
1200
1300
1400
1500Ni-Al BCC-B2 stabilizer
High Temperature Thermochemistry Laboratory
89BCC-B2 containing steel
Liquid
FCC
BCCBCC-B2
Kappa-carbide
68.14 Fe + 16 Mn + 5 Ni + 10 Al + 0.86C in wt.%
Temperature, oC
ph
ase
dis
trib
utio
n,
wt.
%
700 800 900 1000 1100 1200 1300 1400
0
10
20
30
40
50
60
70
80
90
100
(1) Original BCC-B2 phase
(2) BCC-B2 precipitates from FCC
(1) Original BCC-B2
FCC
BCC-B2
Temperature, oC
ph
ase
dis
trib
utio
n,
wt.
%
700 750 800 850 900 950 1000
0
10
20
30
40
50
60
70
80
90
100
Supercooled-FCC from 1000oC
BCC-B2 precipitation
(2) Original BCC-B2
90
Advanced applications
Process simulationEERZ model: mass transfer + thermodymamics
- EAF, BOF, RH degassor, LF, Powder injection,
Tundish, mold flux
SolidificationSolidification model: Back diffusion in solid steel
Solid state transformationSolid state phase transition and precipitation: Diffusion and
precipitation kinetics
- Dictra, Prisma
Summary
91
Thermodynamics Kinetics
Industrial
Experience
(Ideas) $$
Direction Speed
Steelmaking consortium project (2009~2020) – 2019 Annual meeting, July 9-11, Seoul, Korea
Acknowledgement
93
Acknowledgement
Steelmaking consortium project (2009~2020) –2019 Annual meeting, Seoul, Korea, July 9-11, 2019