“Phase Transformation in Materials”
Eun Soo Park
Office: 33-313
Telephone: 880-7221 Email: espark@snu.ac.kr
Office hours: by an appointment
2021 Fall
09.29.2021
1
• Effect of Temperature on Solid Solubility
• Equilibrium Vacancy Concentration
• Influence of Interfaces on Equilibrium
• Gibbs-Duhem Equation:
: Be able to calculate the change in chemical potential that result from a change in alloy composition.
r G 2 γ V
m=
∆ Gibbs-Thomson effect
2 2
ln ln
1 1
ln ln
A B
A B
A B
d G d d
X X RT RT
dX d X d X
γ γ
= + = +
합금조성의 미소변화 (dX)로 인한 화학퍼텐셜의 미소변화
(dμ)를 계산
• Gibbs Phase Rule F = C − P + 1
(constant pressure) Gibbs' Phase Rule allows us to construct phase diagram to represent and interpret phase equilibria in heterogeneous geologic systems.2
A A B B
0
X d µ + X d µ =
Contents for previous class:
n V
m
G G per unit volume of
V β
∆ = ∆
0
V e
G X where X X X
∆ ∝ ∆ ∆ = −
Total Free Energy Decrease per Mole of Nuclei
Driving Force for Precipitate Nucleation
∆G
0=-V∆G
V+A γ + V∆G
sβ α
β
α
µ
µ
AX
A BX
BG = +
∆
1β β β
β
µ
µ
AX
A BX
BG = +
∆
21
2
G
G
G
n= ∆ − ∆
∆
For dilute solutions,
T X
G
V∝ ∆ ∝ ∆
∆
: 변태를 위한 전체 구동력/핵생성을 위한 구동력은 아님
: Decrease of total free E of system
by removing a small amount of material with the nucleus composition (XBβ) (P point)
: Increase of total free E of system
by forming β phase with composition XBβ (Q point)
: driving force for β precipitation
∝undercooling below Te (length PQ)
∆G
VChapter 5.1
3
• Ternary Phase Diagram
• Measurement of multi-component phase diagram
• Computation of Phase Diagram
Contents for today’s class
4
Q1: “Ternary Phase Diagram”?
5
What are ternary phase diagram?
www.sjsu.edu/faculty/selvaduray/page/phase/ternary_p_d.pdf
Diagrams that represent the equilibrium
between the various phases that are formed between three components, as a function of temperature.
Normally, pressure is not a viable variable in ternary phase diagram construction, and is therefore held constant at 1 atm.
6
Gibbs Phase Rule for 3-component Systems
7
Gibbs Triangle
An Equilateral triangle on which the pure components are represented by each corner.
Concentration can be expressed as either “wt. %” or “at.% = molar %”.
X
A+X
B+X
C= 1
Used to determine
the overall composition
8
Overall Composition
9
Overall Composition
10
Ternary Isomorphous System
Isomorphous System: A system (ternary in this case) that has only one solid phase. All components are totally soluble in the other components. The ternary system is therefore made up of three binaries that exhibit total solid solubility.
The Liquidus surface: A plot of the temperatures above which a homogeneous liquid forms for any given overall composition.
The Solidus Surface: A plot of the temperatures below which a (homogeneous) solid phase forms for any given overall composition.
11
Ternary Isomorphous System
TA
TB TC
TA
TA TA
TC
TC TB
TB
12
Ternary Isomorphous System
TB
TC
TA
13
Ternary Isomorphous System
TB
TC
TA
TB
TC
TA
14
Ternary Isomorphous System
TB
TC
TA
TB TC
TA
15
Ternary Isomorphous System
Isothermal section → F = C - P
TB
TC
TA
16
Ternary Isomorphous System
Isothermal section
TB
TC
TA
17
Ternary Isomorphous System
Isothermal section → F = C - P
Fig. 1.41 (a) Free energy surface of a liquid and three solid phases of a ternary system.
(b) A tangential plane construction to the free energy surfaces defined equilibrium between s and l in the ternary system
(c) Isothermal section through a ternary phase diagram
18
Ternary Isomorphous System
Locate overall composition using Gibbs triangle
19
20
Ternary Eutectic System
(No Solid Solubility)
TA
TB
TC TA
TA
TB
TC TA
TB
TC
21
Ternary Eutectic System
Liquidus projection
(No Solid Solubility)
TA
TB
TC
22
Ternary Eutectic System
(No Solid Solubility)
TA
TB
TC
23
Ternary Eutectic System
(No Solid Solubility)
TA
TB
TC
24
Ternary Eutectic System
(No Solid Solubility)
TA
TB
TC
25
Ternary Eutectic System
(No Solid Solubility)
TA
TB
TC
26
Ternary Eutectic System
(No Solid Solubility)
TA
TB
TC
27
Ternary Eutectic System
(No Solid Solubility)
TA
TB
TC
28
Ternary Eutectic System
(No Solid Solubility)
T= ternary eutectic temp.
A C
B L+A+C
L+A+B L+B+C
29
Ternary Eutectic System
(with Solid Solubility)
30
Ternary Eutectic System
(with Solid Solubility)
TA
TB TC
TC
TB TA
31
32
Ternary Eutectic System
(with Solid Solubility)
TA
TB TC
32
Ternary Eutectic System
(with Solid Solubility)
TA TB
TC
33
Ternary Eutectic System
(with Solid Solubility)
TA TB
TC
34
Ternary Eutectic System
(with Solid Solubility)
TA TB
TC
35
36
Ternary Eutectic System
(with Solid Solubility)
TA
TB TC
36
Ternary Eutectic System
(with Solid Solubility)
TA TB
TC
37
Ternary Eutectic System
(with Solid Solubility)
T= ternary eutectic temp.
C
L+ β+γ
L+ α+γ L+ α+β
38
Ternary Eutectic System
(with Solid Solubility)
정해솔 학생 제공 자료 참조: 실제 isothermal section의 온도에 따른 변화 http://www.youtube.com/watch?v=yzhVomAdetM
TA TB
TC
39
Ternary Eutectic System
3) Solidification Sequence: liquidus surface
40
Ternary Eutectic System
* The horizontal lines are not tie lines.
(no compositional information)
* Information for equilibrium phases at different tempeatures
* Vertical section
41
Vertical section Location of vertical section
10.1. THE EUTECTIC EQUILIBRIUM (l = α + β + γ)
42
Vertical section Location of vertical section
10.1. THE EUTECTIC EQUILIBRIUM (l = α + β + γ)
44
Vertical section Location of vertical section
10.1. THE EUTECTIC EQUILIBRIUM (l = α + β + γ)
45
< Quaternary phase Diagrams >
Four components: A, B, C, D
A difficulty of four-dimensional geometry
→ further restriction on the system Most common figure:
“ equilateral tetrahedron “ Assuming isobaric conditions, Four variables: XA, XB, XC and T
4 pure components 6 binary systems 4 ternary systems A quarternary system
a+b+c+d=100
%A=Pt=c,
%B=Pr=a,
%C=Pu=d,
%D=Ps=b
* Draw four small equilateral tetrahedron
→ formed with edge lengths of a, b, c, d
46
Q2: “Measurement of multi-component phase diagram”?
47
Construction of pseudo-binary phase diagram for multi-component alloy system
1MAR2017 / TMS 2017
0 10 20 30 40 50 60 70 80 90 100
600 700 800 900 1000 1100 1200 1300 1400 1500 1600
Temperature ( ℃ )
Cu composition (at. %)
α β
0 10 20 30 40 50 60 70 80 90 100
600 700 800 900 1000 1100 1200 1300 1400 1500 1600
Temperature ( ℃ )
Cu composition (at. %)
α β
0 10 20 30 40 50 60 70 80 90 100
600 700 800 900 1000 1100 1200 1300 1400 1500 1600
Temperature ( ℃ )
Cu composition (at. %)
α β
Phase diagram was expected to optimize composition and microstructure of phase separating HEA
Thermodynamic calculation X-ray diffraction
TGA/DSC FE-EPMA
• Expecting approximation of phase diagram
• Determination of phases
• Finding out temperatures of phase transformations
• Confirming invariant reaction points
• Investigation of equilibrium composition at each
temperature
48
Pseudo-binary phase diagram of PS-HEA
1MAR2017 / TMS 2017
Pseudo-binary system between FeCoCrNi and Cu shows monotectic reaction having liquid separation region.
Calculation courtesy of J.H.Kim et al.
Experiment Calculation
L
L 1 + L 2 FCC 1 + L 2
FCC 1 + FCC 2 FCC 2
Dendrite → Droplet → Dendrite
0 10 20 30 40 50 60 70 80 90 100
400 600 800 1000 1200 1400 1600
Te m pe ra tu re ( ℃ )
Cu content (at%)
CuFeCoCrNi
Posted on L-88 (HEA V: Poster session)
49
MoVNbTiZr: Construction of pseudo-ternary phase diagram Nb,Ti
Mo,V Zr
bcc IC+bc
c
bcc Mo, V, Nb, Ti : Complete S.S.
V-Zr, Mo-Zr Ti-Zr, Nb-Zr
TiNbMoVZr: Construction of pseudo-ternary phase diagram
1500K 1300K
1100K
Calculated pseudo-ternary isothermal sections of the MoNbTiVZr system
MoVNbTiZr: Construction of pseudo-ternary phase diagram
X-ray diffraction analysis of the as-cast samples
Construction of pseudo-ternary phase diagram
Find single phase region without intermetallic compounds BCC1+BCC2+IC
BCC1+IC / BCC2+IC BCC1+BCC2
Single BCC
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 0.0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
20
25 26
23 24
22
21
19 11
10
1213
14 15
16 17
18 9 7
8 5
6 4
2 3
MoV Zr
NbTi
Nb Ti Mo
V
Zr
1
Q3: “Computation of Phase Diagram”?
54
55
<Computation of Phase Diagram>
i ) Pure Stoichiometric Substances
Original Calculation of Enthalpy & Entropy
Debye Suggest Heat Capacity as a function of T!
θ : Debye Temperatrue
ω
D : Debye Frequency(Maximum Frequency of lattice vibration)
Enthalpy and Entropy isState Function!
Substract Initial state to final state, Integral of dH/dT, dS/dT
After Mayer Kelley, more accurate to experiment result of specific heat capacity at Cp
In the phase diagram, we found many differences between the predicted values and the experimental results! Therefore, it is necessary to calibrate the G value!
Here,
considering contribution of lattice & magnetic & pressure
<Magnetic Contribution>
T
c = Curie temperature(to lose magnetic property)
β
: magnetic moment per atomf
structure-dependent function ofT
<Pressure Contribution>
α(T) = thermal expansion of volume
n is pressure derivative of bulk
This polynomial Function fit into real graph! 56
ii) Calculating G for Solid Solution
57
Method of Redlich & Kister
Further multi-component system
59
Homework1 :
Please explain more detail how to calculate multi-component
phase diagram reflecting excess Gibbs energy (within 5 pages PPT)
60
* Incentive Homework 1
Please submit ternary phase diagram model which can clearly express 3D structure of ternary system by October 15 in Bldg. 33-313.
You can submit the model individually or with a small group under 2 persons.
* Homework 2 : Exercises 1 (pages 61-63) Good Luck!!
61
“Phase Transformation in Materials”
Eun Soo Park
Office: 33-313
Telephone: 880-7221 Email: espark@snu.ac.kr
Office hours: by an appointment
2021 Fall
2021.10.04.
1
2
• Ternary Equilibrium: Ternary Phase Diagram Contents for previous class
1) Gibbs Triangle
Used to determinethe overall composition
X
A+X
B+X
C= 1
2) Isothermal section → F = C - P
Tangential plane
2 phases region
Free energy surface
Fig. 1.41 (a) Free energy surface of a liquid and three solid phases of a ternary system.
(b) A tangential plane construction to the free energy surfaces defined equilibrium between s and l in the ternary system
(c) Isothermal section through a ternary phase diagram
3
Ternary Eutectic System
(with Solid Solubility)
3 phases region: tie triangle
4
Ternary Eutectic System
3) Solidification Sequence: liquidus surface
5
Ternary Eutectic System
* The horizontal lines are not tie lines.
(no compositional information)
* Information for equilibrium phases at different temperatures
* Vertical section
6
< Quaternary phase Diagrams >
Four components: A, B, C, D
A difficulty of four-dimensional geometry
→ further restriction on the system Most common figure:
“ equilateral tetrahedron “ Assuming isobaric conditions, Four variables: XA, XB, XC and T
4 pure components 6 binary systems 4 ternary systems A quarternary system
At constant T, D
A
B
C
7
a+b+c+d=100
%A=Pt=c,
%B=Pr=a,
%C=Pu=d,
%D=Ps=b
* Draw four small equilateral tetrahedron
→ formed with edge lengths of a, b, c, d
8
1.10 The kinetics of phase transformations
exp (-ΔGa/kT ) k: Boltzmann’s constant ( R/Na )
Kinetic theory:
probability of an atom reaching the activated stateActivation free energy barrier
Transformation rate ∝ exp (-ΔG
a/k T )
Putting ΔGa= ΔHa - TΔSa and changing from atomic to molar quantities
∝ exp (-ΔH
a/R T )
- First derived empirically from the observed temp.
dependence of the rate of chemical reactions - Apply to a wide range of processes and
transformation in metals and alloys
- “Simplest of these is the process of diffusion”
→ Arrhenius rate equation
9
Contents in Phase Transformation
(Ch1) Thermodynamics and Phase Diagrams (Ch2) Diffusion: Kinetics
(Ch3) Crystal Interface and Microstructure
(Ch4) Solidification: Liquid → Solid
(Ch5) Diffusional Transformations in Solid: Solid → Solid (Ch6) Diffusionless Transformations: Solid → Solid
Background to understand phase
transformation
Representative Phase
transformation
10
Contents for today’s class
• Diffusion
• Interstitial Diffusion – Fick’s First Law
- Effect of Temperature on Diffusivity
- Nonsteady-state diffusion – Fick’s Second Law
• Solutions to the diffusion equations
• Substitution Diffusion
1. Self diffusion in pure material 2. Vacancy diffusion
3. Diffusion in substitutional alloys
11
Q1. What is the driving force for diffusion?
⇒ a concentration gradient (x)
⇒ a chemical potential gradient (o)
Diffusion Movement of atoms to reduce its chemical potential μ.
Diffusion
1213
Diffusion: THE PHENOMENON
• Interdiffusion: in a solid with more than one type of element (an alloy), atoms tend to migrate from regions of large concentration.
Initially After some time
< Diffusion couple between Cu and Ni >
14
Mechanism of Solid-State Diffusion
R.E. Reed-Hill, Physical Metallurgy Principles
Substitutional diffusion Interstitial diffusion
15
Diffusion : Movement of atoms to reduce its chemical potential μ.
Driving force: Reduction of G Down-hill diffusion
movement of atoms from a high concentration region to low
concentration region.
Up-hill diffusion
movement of atoms from a low concentration region to high concentration region
2 1
2 1
B B
A
A
µ
µ
µ µ
>
<
1 21 2
A A
B B
µ µ
µ µ
>
<
‘down-hill’ diffusion
‘up-hill’ diffusion
16
Down-hill diffusion
movement of atoms from a high concentration region to low concentration region.
17
Diffusion
Movement of atoms to reduce its chemical potential μ.Driving force: Reduction of G Down-hill diffusion
movement of atoms from a high concentration region to low
concentration region.
Up-hill diffusion
movement of atoms from a low concentration region to high concentration region
2 1
2 1
B B
A
A
µ
µ
µ µ
>
<
1 21 2
A A
B B
µ µ
µ µ
>
<
‘down-hill’ diffusion
‘up-hill’ diffusion
18
5.5.5 Spinodal Decomposition
Spinodal mode of transformation has no barrier to nucleation
2) If the alloy lies outside the spinodal,
small variation in composition
leads to an increase in free energy and the alloy is therefore
metastable
.2
2
0
d G dX <
The free energy can only be decreased if nuclei are formed with a composition very different from the matrix.
→ nucleation and growth How does it differ between
inside and outside the inflection point of Gibbs free energy curve?
1) Within the spinodal
: phase separation by small fluctuations in composition/
“up-hill diffusion”
: “down-hill diffusion”
Fig. 5.38 Alloys between the spinodal points are unstable and can decompose into two coherent phasees α1and α2without overcoming an activation energy barrier. Alloys between the coherent miscibility gaps and the spinodal are metastable and can decompose only after nucleation of the other phase.
: describing the transformation of a system of two or more components in a metastable phase into two stable phases
At T2,
19
b) Normal down-hill diffusion outside the spinodal
a) Composition fluctuations within the spinodal
up-hill diffusion
interdiffusion coefficient
D<0
down-hill diffusion
Fig. 5.39 & 5.40 schematic composition profiles at increasing times in (a) an alloy quenched into the spinodal region (X0 in Figure 5.38) and (b) an alloy outside the spinodal points (X0’ in Figure 5.38)
20
Diffusion
Diffusion : Mechanism by which matter transported through matter What is the driving force for diffusion?
But this chapter will explain with
“concentration gradients for a convenience”.
(Down-hill diffusion)
⇒ a concentration gradient (x)
⇒ a chemical potential gradient (o)
21
Q2. Interstitial diffusion vs Substitutional diffusion
Fick’s First Law of Diffusion
22
Atomic mechanisms of diffusion
The mean vibrational energy is 3kT , therefore the amplitude of thermal oscillation of an atom increases markedly with temperature . If a neighboring atom site is vacant, then the oscillation atom can change site with a certain probability ( atom jumping ). The probability consists of creation of the vacant site and movement of a neighboring atom into the site.
Substitutional vs. Interstitial diffusion
Substitutional diffusion
FCC
# of vacant site
23
Interstitial diffusion
FCC (0.74) BCC (0.68)
Fig. 2.3 (a) Octahedral interstices (O) in an fcc crystal. (b) Octahedral interstices in a bcc crystal.
24
How interstitial diffusion differs from substitutional diffusion?
Interstitial diffusion
25
Assume that there is no lattice distortion and also that there are always six vacant sites around the diffusion atom.
JB : Net flux of B atom
ΓB : Average jump rate of B atoms n1 : # of atoms per unit area of plane 1 n2 : # of atoms per unit area of plane 2
2
1
6
1 6
1 n n
J
B= Γ
B− Γ
BInterstitial diffusion
Random jump of solute B atoms in a dilute solid solution of a simple cubic lattice
( )
α
− = − ∂ ∂
B B B
C (1) C (2) C x
Fig. 2.5 Interstitial diffusion by random jumps in a concentration gradient.
26
α α
= =
B B
C (1) n
1, C (2) n
2(atoms m-2 s-1)
( n
1− n
2) = α ( C
B(1) − C
B(2) )
DB: Intrinsic diffusivity or Diffusion coefficient of B
⇒ depends on microstructure of materials
Fick’s First Law of Diffusion
(
1 2)
2
1
6
1 6
1 6
1 n n n n
J
B= Γ
B− Γ
B= Γ
B−
( ) α ∂ ∂
= Γ − = − Γ ∂ = − ∂
2
1 2
1 1
6 6
B B
B B B B
C C
J n n D
x x
( )
α
− = − ∂ ∂
B B B
C (1) C (2) C x
α: Jump distance
Magnitude of D in various media Gas : D ≈ 10-1 cm2/s
Liquid : D ≈ 10-4 ~ 10-5 cm2/s
Solid : Materials near melting temp. D ≈ 10-8 cm2/s Elemental semiconductor (Si, Ge) D ≈ 10-12 cm2/s
Concentration varies with position.
D
B, in general, is concentration-dependent.
For example, in γ-Fe at 1000
oC,
D
c= 2.5 x 10
-11m
2s
-1, when C
c= 0.15 wt%
D
c= 7.7 x 10
-11m
2S
-1, when C
c= 1.4 wt%
2
6
1
Bα D
B= Γ
* Estimate the jump frequency of
a carbon atom in γ-Fe (FCC) at 1000
oC .
lattice parameter of γ-Fe : ~0.37 nm
nm .
. 37 2 0 26
0 =
α = D
C= 2 . 5 × 10
−11m
2s
−1Γ = × 2 10
9jumps s
−1the vibration frequency of carbon : ~ 1013
Only about one attempt in 104 results in a jump from one site to another.
↑
↑→
Cc
D
C
* If the crystal structure is not cubic, then the probability of jump is
anisotropic (different α)
. For example the probability of jumps along the basal direction and the axial direction of hcp crystal are different.∵C atoms stain the Fe lattice
thereby making diffusion easier as amount of strain increases.
28
Q3. What is the radial distance, r from the origin in random work?
( ) ( )
α
= Γ =
r t f D = 2.4 Dt
29
( ) ( )
α
= Γ =
r t f D
During random walk,
an atom will move in time (t)
a radial distance (r) from the origin For random walk in 3 dimensions, after n steps of length α
α
→ n
Dt D r
6 6
2
→ =
=
Γ α
In 1 s, each carbon atom will move a total distance of ~ 0.5 m
→ a net displacement :
= 2.4 Dt
~ 10 µm.
2
6 1 Bα DB = Γ Net distance from its original position
Very few of the atom jumps provide a useful contribution to the total diffusion distance.
The direction of each new jump is independent of the direction of the previous jump.
30
Q4. What is the effect of the temperature on diffusivity?
0
exp
IDB B
D D Q
RT
= −
31
EFFECT OF TEMPERATURE on Diffusivity
How D varies with T?
Thermal Activation
How varies with T? Γ
Fig. 2.6 Interstitial atom, (a) in equilibrium position, (b) at the position of maximum lattice distortion.
(c) Variation of the free energy of the lattice as a function of the position of interstitial.
32
0
exp
IDB B
D D Q
RT
= −
m m m
G H T S
∆ = ∆ − ∆ 1
2B
6
BD = Γ α
B
? jump frequency Γ
Thermally activated process
Ζ : nearest neighbor sites ν : vibration frequency
∆Gm : activation energy for moving
( ) ( )
ID m
m m
B
Q H
RT H
R S
D
≡
∆
∆
−
Ζ ∆
= exp / exp /
6
1 α
2ν
( G
mRT )
B
= Ζ exp − ∆ /
Γ ν
33
−
= . R T
D Q log D
log 1
3
0
2
Temperature Dependence of Diffusion
=
0exp −
IDB B
D D Q
RT
Therefore, from the slope of the D-curve in an log D vs 1/T coordinate, the activation energy may be found.
How to determine Q
IDexperimentally?
Fig. 2.7 The slope of log D v. 1/T gives the activation energy for diffusion Q.
34
Q5. Steady state diffusion vs Non-steady state diffusion?
Fick’s Second Law
∂ = ∂
∂ ∂
2 2
B B
B
C C
t D x
Fick’s first Law
Concentration varies with
“position” “Both position and time”
35
Steady-state diffusion
The simplest type of diffusion to deal with is when the concentration at every point does not change with time.
Apply Fick’s First Law:
the slope, dC/dx, is constant (does not vary with position)!
Steady State
Concentration, C, in the box
then
36
Steady-state diffusion
• Steel plate at 700℃
Q: How much carbon is transferring from the rich to deficient side?
The simplest type of diffusion to deal with is when the
concentration at every point does not change with time.
37
Nonsteady-state diffusion
In most practical situations steady-state conditions are not established, i.e. concentration varies with both distance and time, and Fick’s 1st law can no longer be used.
How do we know the variation of C
Bwith time? → Fick’s 2nd law
The number of interstitial B atoms that diffuse into the slice across plane (1) in a small time interval dt :
→ J
1A dt Likewise : J
2A dt
Sine J2 < J1, the concentration of B within the slice will have increased by
( ) δ
δ δ
=
1−
2B
J J A t
C A x
Due to mass conservation
( J
1− J
2) A t δ = = ? δ C A x
Bδ
38
Nonsteady-state diffusion
B B
C J
t x
∂ ∂
∂ = − ∂
Fick’s Second Law
substituting Fick's 1st law gives
∂
∂
∂
= ∂
∂
∂
x D C
x t
C
BB B
∂ = ∂
∂ ∂
2 2
B B
B
C C
t D x
x D C
J
B B∂
− ∂
=
If variation of D with concentration can be ignored, ( ) δ
δ δ
= 1− 2
B
J J A t
C A x
39
x ?
C
B=
∂
∂
2 2
∂ ∂
∂ = ∂
2 2
B B
B
C C
t D x
Fick’s Second Law
Concentration varies with time and position.
Note that is the curvature
2 of the CB versus x curve.2
x C
B∂
∂
2 0
2 >
∂
∂ x
CB
2 0
2 <
∂
∂ x
CB
>0
∂
∂ t CB
<0
∂
∂ t CB
All concentration increase with time All concentration decrease with time
40
Q6. How to solve the diffusion equations?
: Application of Fick’s 2
ndlaw
homogenization, carburization, decarburization, diffusion across a couple
41
Solutions to the diffusion equations (Application of Fick’s 2
ndlaw) Ex1. Homogenization
of sinusoidal varying composition
in the elimination of segregation in casting
l : half wavelength
l C x
C = + β
0sin π
: the mean composition C
0 : the amplitude of the initial concentration profile β
Initial or Boundary Cond.?
at t=0
Fig. 2.10 The effect of diffusion on a sinusoidal variation of composition.
∂ = ∂
∂ ∂
2 2
B B
B
C C
t D x
42
Rigorous solution of
for2 2
x D C t
C
∂
= ∂
∂
∂ ( )
l C x
x
C ,0 = +β0sinπ
( )
( )
D l
t n
n
n n
n n
Dt
l e C x
C
A B
C A
l C x
C t
x B
x A
A C
e x B
x A
C
2 2
2
/ 0
0 1
0
0 1
0
sin , 0 ,
;
sin 0
cos sin
sin cos
π
λ
β π
β β π
λ λ
λ λ
−
∞
=
−
+
≡
∴
=
=
=
+
≡
→
=
+ +
=
∴
+
=
∴
∑
τ : relaxation time
+ −
= τ
β π t
l C x
C
0sin exp
D l
2 2
τ = π
) 2 /
0 exp(
x l at
t =
−
=
β τ
β
2 2
2 2 2
1 1
Let
λ
−
≡
=
=
=
dx X d X dt
dT DT
dx X DT d
dt X dT
XT C
x B
x A
X dx X
X d
λ λ
λ
sin cos
2 0
2 2
+ ′
= ′
= +
Using a method of variable separation
l C β sin π x X(x,0) ≡ +
0e Dt
T T
dt D T d
dt D dT T
2
0
2 2
ln 1
λ
λ λ
= −
−
=
−
=
l λ =π
2 2
x D C t C
∂
= ∂
∂
∂
(An = 0 for all others)
43
Solutions to the diffusion equations
Ex1. Homogenization
of sinusoidal varying composition in the elimination of segregation in castingl C x
C = + β
0sin π
at t=0τ : relaxation time
+ −
= τ
β π t
l C x
C
0sin exp
D l
2 2
τ = π
) 2 /
0
exp(
x l at
t =
−
= β τ
β
The initial concentration profile will not usually be sinusoidal, but in general any con- centration profile can be considered as the sum of an infinite series of sine waves of varying wavelength and amplitude, and each wave decays at a rate determined by its
own “τ”. Thus, the short wavelength terms die away very rapidly and the homogenization will ultimately be determined by τ for the longest wavelength component.
decide homogenization rate
Fig. 2.10 The effect of diffusion on a sinusoidal variation of composition.
Amplitude of the concentration profile (β) decreases exponentially with time,
44
Solutions to the diffusion equations
Ex2. Carburization of Steel
The aim of carburization is to increase the carbon concentration in the surface layers of a steel product in order to achiever a harder wear-resistant surface.
1. Holding the steel in CH4 and/or Co at an austenitic temperature.
2. By controlling gases the concentration of C at the surface of the steel can be maintained at a suitable constant value.
3. At the same time carbon continually diffuses from the surface into the steel.
45
Carburizing of steel
The error function solution:
( ) ( ) 0 , , t C
0C
C t
C
B
s B
=
∞
=
( )
( )
Ζ =∫
Ζ −
−
−
=
0 0
2 2
2 dy e
erf
Dt erf x
C C
C C
y s
s
π
Depth of Carburization?
• Since erf(0.5)≈0.5, the depth at which the carbon concentration is midway between C
sand C
0is given
that is
( / 2 x Dt ) ≅ 0.5 x ≅ Dt
erf(0.5)≈0.5
=
−
−
Dt erf x
C C
C C
s s
0
2
0
2 C
sC
C +
=
→ Depth of Carburization
* Concentration profile : using boundary conditions
Solution of Fick’s 2nd law
1) Specimen ~ infinitely long.
2) Diffusion coefficient of carbon in austenite increases with increasing concentration, but
an approximate solution: Taking an average value
Fig. 2.11 Concentration profiles at successive times (t3>t2>t1) for diffusion into a semi-infinite bar when the surface concentration Cs is maintained constant.
∂ ∂
∂ = ∂
2 2
B B
B