Diffusion Callister
Diffusion Callister
Diffusion Callister
kg
m
2
s
(
(
or
atoms
m
2
s
(
(
Directional Quantity
Flux can be measured for:
--vacancies
--host (A) atoms
--impurity (B) atoms
J
x
J
y
J
z
x
y
z
x-direction
Unit area A
through
which
atoms
move.
MODELING DIFFUSION: FLUX
Concentration Profile, C(x): [kg/m
3
]
Fick's First Law:
Concentration
of Cu [kg/m
3
]
Concentration
of Ni [kg/m
3
]
Position, x
Cu flux Ni flux
The steeper the concentration profile,
the greater the flux!
Adapted from
Fig. 5.2(c),
Callister 6e.
J
x
=
D
dC
dx
Diffusion coefficient [m
2
/s]
concentration
gradient [kg/m
4
]
flux in x-dir.
[kg/m
2
-s]
CONCENTRATION PROFILES & FLUX
Steady State: Steady rate of diffusion from one end to the other.
Implies that the concentration profile doesn't change with time. Why?
Apply Fick's First Law:
Result: the slope, dC/dx, must be constant
(i.e., slope doesn't vary with position)!
J
x(left)
= J
x(right)
Steady State:
Concentration, C, in the box doesnt change w/time.
J
x(right)
J
x(left)
x
J
x
= D
dC
dx
dC
dx
|
\
|
.
|
left
=
dC
dx
|
\
|
.
|
right
If J
x
)
left
= J
x
)
right
, then
STEADY STATE DIFFUSION
Steel plate at
700C with
geometry
shown:
Q: How much
carbon transfers
from the rich to
the deficient side?
J
=
D
C
2
C
1
x
2
x
1
=
2.4
10
9
kg
m
2
s
Adapted from
Fig. 5.4,
Callister 6e.
C
1
=
1
.
2
k
g
/
m
3
C
2
=
0
.
8
k
g
/
m
3
Carbon
rich
gas
1
0
m
m
Carbon
deficient
gas
x
1
x
2
0
5
m
m
D=3x10
-11
m
2
/s
Steady State =
straight line!
EX: STEADY STATE DIFFUSION
Note: Steady state does not set in instantaneously.
STEADY STATE DIFFUSION:
ANOTHER PERSPECTIVE
Hose connected to tap; tap turned on
At the instant tap is turned on, pressure is high at the tap
end, and 1 atmosphere at the other end
After steady state is reached, pressure linearly drops
from tap to other end, and will not change anymore
Tap end Flow end
Pressure
Increasing time
Steady state
Concentration profile,
C(x), changes
w/ time.
To conserve matter: Fick's First Law:
Governing Eqn.:
Concentration,
C, in the box
J
(right)
J
(left)
dx
dC
dt
=
D
d
2
C
dx
2
dx
=
dC
dt
J
(left)
J
(right)
NON STEADY STATE DIFFUSION
Ficks second law
dJ
dx
=
dC
dt
dJ
dx
=
D
d
2
C
dx
2
(if D does
not vary
with x)
equate
J
=
D
dC
dx
Copper diffuses into a bar of aluminum.
Boundary conditions:
For t = 0, C = C
0
at x > 0
For t > 0, C = C
s
at x = 0
C = C
0
at x =
pre-existing conc., C
o
of copper atoms
Surface conc.,
C
s
of Cu atoms
bar
C
o
C
s
position, x
C(x,t)
t
o
t
1
t
2
t
3
Adapted from
Fig. 5.5,
Callister 6e.
EX: NON STEADY STATE DIFFUSION
dC
dt
=
D
d
2
C
dx
2
Copper diffuses into a bar of aluminum.
General solution:
"error function"
Values calibrated in Table 5.1, Callister 6e.
C(x, t)
C
o
C
s
C
o
=
1
erf
x
2 Dt
|
\
|
.
|
pre-existing conc., C
o
of copper atoms
Surface conc.,
C
s
of Cu atoms
bar
C
o
C
s
position, x
C(x,t)
t
o
t
1
t
2
t
3
Adapted from
Fig. 5.5,
Callister 6e.
EX: NON STEADY STATE DIFFUSION
Suppose we desire to achieve a specific concentration C1
at a certain point in the sample at a certain time
PROCESS DESIGN EXAMPLE
|
.
|
\
|
=
Dt
x
erf
C C
C t x C
s
2
1
) , (
0
0
|
.
|
\
|
= =
Dt
x
erf
C C
C C
s
2
1 constant
0
0 1
becomes
constant
2
=
Dt
x
The experiment: record combinations of
t and x that kept C constant.
t
o
t
1
t
2
t
3
x
o
x
1
x
2
x
3
Diffusion depth given by:
x
i
Dt
i
C(x
i
, t
i
) C
o
C
s
C
o
= 1 erf
x
i
2 Dt
i
|
\
|
.
|
|
= (constant here)
DIFFUSION DEMO: ANALYSIS
Experimental result: x ~ t
0.58
Theory predicts x ~ t
0.50
Reasonable agreement!
B
B
B
B
B
B
B
B
B
B
B
B
B
B
0
0.5
1
1.5
2
2.5
3
3.5
4
0 0.5 1 1.5 2 2.5 3
( )
ln[t(min)]
Linear regression fit to data:
ln[x(mm)]
=
0.58ln[t(min)]
+
2.2
R
2
=
0.999
DATA FROM DIFFUSION DEMO
Copper diffuses into a bar of aluminum.
10 hours at 600C gives desired C(x).
How many hours would it take to get the same C(x)
if we processed at 500C, given D
500
and D
600
?
(Dt)
500C
=(Dt)
600C
s
C(x,t)
C
o
C C
o
=
1
erf
x
2
Dt
|
\
|
.
|
Result: Dt should be held constant.
Answer:
Note: values
of D are
provided here.
Key point 1: C(x,t
500C
) = C(x,t
600C
).
Key point 2: Both cases have the same C
o
and C
s
.
t
500
=
(Dt)
600
D
500
=
110
hr
4.8x10
-
14
m
2
/s
5.3x10
-
13
m
2
/s
10hrs
PROCESSING QUESTION
Diffusivity increases with T.
pre-exponential [m
2
/s] (see Table 5.2, Callister 6e)
activation energy
gas constant [8.31J/mol-K]
D
=
D
o
exp
Q
d
RT
|
\
|
.
|
diffusivity
[J/mol],[eV/mol]
(see Table 5.2, Callister 6e)
DIFFUSION AND TEMPERATURE
Remember vacancy concentration: N
V
= N exp(-QV/kT)
QV is vacancy formation energy (larger this energy,
smaller the number of vacancies)
Qd is the activation energy (larger this energy, smaller
the diffusivity and lower the probability of atomic diffusion)
ACTIVATION ENERGY FOR
DIFFUSION
Also called energy barrier for diffusion
Initial state Final state Intermediate state
Energy Activation energy
Experimental Data:
1000K/T
D (m
2
/s)
C
i
n
-
F
e
C
i
n
-
F
e
A
l
i
n
A
l
C
u
i
n
C
u
Z
n
i
n
C
u
F
e
i
n
-
F
e
F
e
i
n
-
F
e
0.5 1.0 1.5 2.0
10
-20
10
-14
10
-8
T(C)
1
5
0
0
1
0
0
0
6
0
0
3
0
0
D has exp. dependence on T
Recall: Vacancy does also!
pre-exponential [m
2
/s] (see Table 5.2, Callister 6e)
activation energy
gas constant [8.31J/mol-K]
D
=
D
o
exp
Q
d
RT
|
\
|
.
|
diffusivity
[J/mol],[eV/mol]
(see Table 5.2, Callister 6e)
D
interstitial
>> D
substitutional
C in -Fe
C in -Fe
Al in Al
Cu in Cu
Zn in Cu
Fe in -Fe
Fe in -Fe
Adapted from Fig. 5.7, Callister 6e. (Date for Fig. 5.7 taken from E.A.
Brandes and G.B. Brook (Ed.) Smithells Metals Reference Book, 7th
ed., Butterworth-Heinemann, Oxford, 1992.)
DIFFUSION AND TEMPERATURE
NOTE: log(D) = log(D0) Qd/(RT)
Diffusion FASTER for...
open crystal structures
lower melting T materials
materials w/secondary
bonding
smaller diffusing atoms
lower density materials
Diffusion SLOWER for...
close-packed structures
higher melting T materials
materials w/covalent
bonding
larger diffusing atoms
higher density materials
SUMMARY:
STRUCTURE & DIFFUSION