engineering thermodynamics mime212in-ho-group.snu.ac.kr/wp-content/uploads/calphad... ·...

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

http://in-ho-group.snu.ac.kr/

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


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 = -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


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


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


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


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)


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


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


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


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


(--)(--)

(~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)


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


47


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


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



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


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



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


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


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


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)


62Casting process

Submerged Entry Nozzle (SEN)

- Reoxidation of liquid steel

Mold

- Solidification range of steel

- Mold flux design

- Surface quality (AlN)


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


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


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


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)


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)


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


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


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


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


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)


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


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


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


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.”


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


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

engineering thermodynamics mime212in-ho-group.snu.ac.kr/wp-content/uploads/calphad... ·...

Documents