Sei sulla pagina 1di 33

DIFFUSION IN SOLIDS

ISSUES TO ADDRESS

How does diffusion occur?

• Why is it an important part of processing?

• How can the rate of diffusion be predicted for some simple cases?

• How does diffusion depend on structure and temperature?

• How to control diffusion process?

How to control carburizing process?

1
1

DIFFUSION DEMO

• Glass tube filled with water.

• At time t = 0, add some drops of ink to one end of the tube.

Measure the diffusion distance, x, over some time.

• The concentration of ink is a function of time and distance x.

t x (mm) o t 1 t 2 t 3 time (s) x o x
t
x (mm)
o
t
1
t
2
t
3
time (s)
x
o x 1
x 2
x 3
2

DIFFUSION: THE PHENOMENA (1)

Interdiffusion: In an alloy, atoms tend to migrate from regions of high concentration to low concentration.

Initially

After some time

Adapted from Figs. 5.1 and 5.2, Callister 6e. C u Ni 100% 100% 0 0
Adapted
from Figs.
5.1 and 5.2,
Callister 6e.
C
u
Ni
100%
100%
0
0
Concentration Profiles
C
oncen ra on
t
ti
P
ro
fil
es
t=0
t > 0
3

DIFFUSION: THE PHENOMENA (2)

Self-diffusion: In an elemental solid, atoms also migrate.

Label some atoms C A D B
Label some atoms
C
A
D
B
After some time C D A B
After some time
C
D
A
B

Diffusion: The movement of atoms or molecules from an area of higher concentration to an area of lower concentration. (Interdiffusion or impurity diffusion)

4
4

DIFFUSION MECHANISMS

Substitutional Diffusion:

• applies to substitutional impurities

• atoms exchange with vacancies

• rate depends on:

--number of vacancies --activation energy to exchange.

Both self- diffusion and inter-diffusion occur

i ncreas ng e apse i l d ti me 5
i
ncreas ng e apse
i
l
d ti
me
5

INTERSTITIAL SIMULATION

Applies to interstitial impurities.

• More rapid than vacancy diffusion.

• Simulation:

--shows the jumping of a smaller atom (gray) from one interstitial site to another in a BCC structure. The i considered here are at midpoints along the unit cell edges.

t

n ers

titi

a

l

s

it

es

along the unit cell ed g es. t n ers titi a l s it es

(Courtesy P.M. Anderson)

Interdiffusion of impurities such as H, C, N and O; which have atom small enough to fit into the interstitial position

Atomic radius

Iron

0.124 nm

C

0.071 nm

6
6
©2003 Brooks/Cole, a division of Thoms on Learning, Inc. Thomson Learning ™ is a trademark

©2003 Brooks/Cole, a division of Thomson Learning, Inc. Thomson Learning is a trademark used herein under license.

Diffusion mechanisms in material: (a) vacancy or substitutional atom diffusion and (b) interstitial diffusion

 
  7
  7
  7
7
7
 

Background - Steel

• Pure iron is relatively soft and would not last very long if used as a tool

 

• Luckily, when a small amount of carbon (up to about 1.5%) is added to the iron it is called steel and can be made much harder by a heat treatment called quite simply, hardening.

 

• If some other metals, such as chromium, nickel and manganese are added to the steel it can be made much stronger and tougher and is called alloy steel

• In simple terms:

– The amount of carbon in the steel determines how hard it will be after hardening

 

– The various metals with which it is alloyed determine how strong or tough it will be, after hardening

 
  8
  8
  8
8
8

MECHANICAL PROP: Fe-C SYSTEM (1)

• Effect of wt%C Pearlite (med) Pearlite (med) ferrite (soft) Cementite (hard) C o <0.77wt%C
• Effect of wt%C
Pearlite (med)
Pearlite (med)
ferrite (soft)
Cementite
(hard)
C o <0.77wt%C
C o >0.77wt%C
Adapted from Fig. 9.27,Callister
6e. (Fig. 9.27 courtesy Republic
ee
Adapted from Fig. 9.30,Callister
6e. (Fig. 9.30 copyright 1971 by
Hypoeutectoid
Hypereutectoid
St
l C
orpora on.)
ti
U
n
it
e
d St
a es
t
St
ee
l C
orpora
ti
on.)
Hypo
Hyper
Hypo
Hyper
TS(MPa)
%EL
1100
80
YS(MPa)
100
Adapted from Fig.
10.20, Callister 6e.
900
(Fig. 10.20 based on
hardness
data from Metals
40
700
Handbook: Heat
Treating Vol 4 9th
,
,
50
500
0
.
ed., V. Masseria
(Managing Ed.),
American Society for
Metals, 1981, p. 9.)
300
0
0
0.5
1
0
0.5
1
wt%C
wt%C
• More wt%C: TS and YS increase, %EL decreases.
9
0.77
0.77
Im
pact ene rgy (Izod, ft-lb)

PROCESSING USING DIFFUSION (1)

pact ene rgy (Izod, ft-lb) PROCESSING USING DIFFUSION (1) • Case Hardening : -- Diffuse carbon

Case Hardening:

--Diffuse carbon atoms into the host iron atoms at the surface. --Example of interstitial diffusion is a case hardened gear.

Fi

g.

.

5 0

,

Callister 6e.

(Fig. 5.0 is courtesy of Surface Division,

Midland-

Ross.)

• Result: The "Case" is

--hard to deform: C atoms "lock" planes from shearing. --hard to crack: C atoms put the surface in compression.

Improve: Wear resistance Wear r=kw/H, w-load, H-hardness Fatigue life – compress stress

10
10

Carburizing

The process:

• Pack carburizing

• Gas carburizing

Liquid carburizing

• Vacuum carburizing

• Plasma carburizing

Depth of Hardening:

Case depths from as light as 0.08 mm (0.003") to as deep as 6.4 mm (0.250") may be specified, depending on the service requirements of the product.

 
  11
  11
  11
11
11

Application

Depth of Case

 

High wear resistance, low to moderate loading-Small and delicate machine parts subject to wear

Cases to 0.51 mm

 

(0.020")

High wear resistance, moderate to heavy loading-Light industrial gearing

0.51 mm to 1.02 mm (0.020" to 0.040")

 

High wear resistance, heavy loading, crushing loads or high magnitude alternating bending stresses-Heavy duty industrial gearing

1.02mm to 1.52mm (0.040" to 0.060")

High wear resistance, shock resistance, high crushing loads-Bearing surfaces, mill gearing, rollers

1.52 mm to 6.4mm (0.060" to 0.250")

Carburizing Time: 4 to 72 hours

Carburizing Temperature: 850-950 °C (1550-1750 °F) (i.e., above the upper critical temperatures - austenite area)

 

Quenching: After carburizing, the part is either slow cooled for later quench hardening, or quenched directly into various liquid quenches. The part is then tempered to the desired hardness to achieve the optimum properties with acceptable levels of dimensional change.

 
  12
  12
  12
12
12

Materials

Most steels specified for carburizing contain less than 0.25% carbon, with sufficient alloys to improve case and core hardenability. Depending on the application, any of numerous grades may be used. In general, steels that are applicable to carburizing are the following:

1. Low-carbon steels

2. Resulfurized low-carbon steels

3. Low-carbon alloy steels

4. Low-carbon powder metal (P/M) compacts

13
13

Resulfurized low-carbon steels

Sulfur - is usually an undesirable impurity in steel rather than an alloying element. In amounts exceeding 0.05% it tends to cause brittleness and reduce weldability. Alloying additions of sulfur in amounts from 0.10% to 0.30% will tend to improve the machinability of a steel. Such types may be referred to as "resulfurized" or "free-machining". Free-machining alloys are not intended for use where welding is required.

14
14

Pack carburizing

The part is packed in a steel container so that it is completely surrounded

by granules of charcoal. The charcoal is treated with an activating chemical (a catalyst) such as barium carbonate (BaCO 3 ) that promotes the formation of carbon dioxide (CO 2 ). This gas in turn reacts with the excess carbon in the charcoal to produce carbon monoxide, CO. Carbon monoxide reacts with the low-carbon steel surface to form atomic carbon which diffuses into the steel. Carbon monoxide supplies the carbon gradient that is necessary for diffusion. The carburizing process does not harden the steel. It only increases the carbon content to some predetermined depth below the surface to a sufficient level to allow subsequent quench hardening.

Steel

container

Heat

Part to be carburized C CO CO Charcoal
Part to be
carburized
C
CO
CO
Charcoal

F gure. Pac

carburizing process

i

k

 

Heat

  Heat 15
  Heat 15
  Heat 15
15
15
 

Pack carburizing-continued

Quenching Process:

 

It is difficult to quench the part immediately, the sealed pack has to be opened and the part must be removed from the pack. One technique that is used often is to slow cool the entire pack and subsequently harden and temper the part after it is removed from the sealed pack. Depth of Hardening:

 

There is no technical limit to the depth of hardening with carburizing techniques, but it is not common to carburize to depths in excess of 1.3

mm

(0.050").

Carburizing Time: 4 to 10 hours

 

The degree of carburizing depends on

- Substrate (its carbon content and alloy content)

- Carburizing temperature, and

- Time that the part is soaked at the carburizing temperature.

16
16
16
16
16

Advantages of Pack Carburizing

1. It can be done in almost any type of furnace

2. The equipment requirement is minimal (furnace, box, compound)

3. A wide variety of parts can be accommodated (as many as could be fitted and separated in a box, or as large as the box that will fit in the available furnace)

4. Requires lower operator skills than other processes.

Disadvantages of Pack Carburizing

1. Carburizing times are longer than for some of the other processes

2. Not suitable for continuous production

3. Labour intensive (pack loading, box maintenance, sealing, pack handling etc)

4. Unsuitable for thin, carefully controlled case depths.

At carburizing temperatures, say 900 °C, the following reactions occur:

(1)

(2)

(3)

C + O

Fe +

(initial air in charcoal) CO CO Fe(C in solution)+ CO

2

2

2

2

2

CO + C CO CO + C CO

2

2

.

.

.

BaCO BaO + CO

3

CO + C CO

2

 
  17
  17
  17
17
17
 

Gas Carburizing

The parts are heated above the upper critical temperature in a furnace with an atmosphere of carbon-containing gas such as methane, ethane, propane, natural gas, acetylene, manufactured gas or mixed hydrocarbon gases. Most carburizing gases are flammable and controls are needed to keep carburizing gas at 927 °C (1700 °F) from contacting air (oxygen). The carburizing gases are often diluted with an endothermic carrier gas, mainly nitrogen (N 2 ) and CO along with smaller amounts of CO 2 , H 2 and H 2 O. Of all the gases, N 2 is inert and acts only as a dilutent. The carrier gas serves to control the amount of carbon supplied to the steel surface and prevents the formation of soot residue.

 

Mechanism

 

(1)

Transport of gas molecules containing carbon to the surface of the steel part Reaction of the molecules at the surface to raise carbon content of steel, and

 

(2)

(3)

Diffusion of the carbon into the steel.

 
  18
  18
  18
18
18

First, methane reacts with CO 2 and H 2 O to generate CO and H 2 .

CH 4

+

CO 2

2CO + 2H 2

.

.

.

.

.

. (4)

CH 4 + H 2 O

CO + 3H 2

.

.

.

.

.

. (5)

CO

and H 2 . Carbon monoxide is the primary gas responsible for raising the carbon content at surface of the steel. Second, the CO decomposes to allow carbon to diffuse into the steel surface by the

These reactions decrease the amounts of CO

2 and H 2 O but

i

h

f

ncrease t e amounts o

following reversible reactions

2CO C (in Fe) + CO 2 C

CO + H

i

F

+ H O

2

( n

e)

2

 

.

.

.

.

.

. (6)

 

7

.

.

.

.

.

.

(

)

Thus, the carbon content on the surface of the steel may be controlled by either a constant CO 2 content or a constant water vapor content determined by the dew point of the gas.

If we combine Equations (4) and (6) or Equations (5) and (7), we have

CH

4

C (in Fe) + 2H

2

.

.

.

.

.

.

(8)

19
19

Advantages of Gas Carburizing (over pack carburizing)

(1)

More accurate control of the composition and depth of the hardened case

(2)

Suitable for continuous production and high-volume production surface hardening

Disadvantages

(1)

(2)

(3)

(4)

High equipment requirements Soaking time required is longer than for pack carburizing High safety demands. The gases used for gas carburizing can be explosive. Requires experienced and skilled personnel and very reliable gas control systems.

Requires experienced and skilled personnel and very reliable gas control systems . Fi gure. G as

Figure. Gas

carburizing

20
20

PROCESSING USING DIFFUSION (2)

Doping Silicon with P for n-type semiconductors:

• Process:

Silicon with P for n-type semiconductors: • P rocess: 0.5mm light regions: Si atoms li g

0.5mm

with P for n-type semiconductors: • P rocess: 0.5mm light regions: Si atoms li g ht
light regions: Si atoms li g ht reg ons: i Al a oms t
light regions: Si atoms
li
g
ht
reg ons:
i
Al
a oms
t

magnified image of a computer chip

1. Deposit P rich la ers on surface

y

of a computer chip 1. Deposit P rich la ers on surface y . silicon Fig.
of a computer chip 1. Deposit P rich la ers on surface y . silicon Fig.
of a computer chip 1. Deposit P rich la ers on surface y . silicon Fig.
of a computer chip 1. Deposit P rich la ers on surface y . silicon Fig.

.

silicon
silicon
silicon
silicon
silicon
silicon
silicon

silicon

silicon

Fig. 18.0,

Callister 6e.

2. Heat it.

3. Result: Doped semiconductor regions.

6e . 2. Heat it. 3. Result: Doped semiconductor regions. silicon 21 MODELING DIFFUSION: FLUX •
6e . 2. Heat it. 3. Result: Doped semiconductor regions. silicon 21 MODELING DIFFUSION: FLUX •
6e . 2. Heat it. 3. Result: Doped semiconductor regions. silicon 21 MODELING DIFFUSION: FLUX •
6e . 2. Heat it. 3. Result: Doped semiconductor regions. silicon 21 MODELING DIFFUSION: FLUX •
silicon
silicon
silicon
silicon
silicon
silicon
silicon
silicon
silicon
silicon
silicon
silicon
silicon

silicon

silicon
silicon
silicon
silicon
21
21

MODELING DIFFUSION: FLUX

Flux:

J = 1

dM

A dt

kg

m 2 s

or

atoms

m 2 s

• Directional Quantity

y J y J x J z x z
y
J
y
J x
J
z
x
z

• Flux can be measured for:

--vacancies --host (A) atoms --impurity (B) atoms

⎤ ⎥ ⎦ x-direction Unit area A through which atoms move. 22
x-direction
Unit area A
through
which
atoms
move.
22

CONCENTRATION PROFILES & FLUX

Concentration Profile, C(x): [kg/m 3 ]

Cu flux Ni flux Concentration of Cu [kg/m 3 ] Concentration of Ni [kg/m 3
Cu flux
Ni flux
Concentration
of Cu [kg/m 3 ]
Concentration
of Ni [kg/m 3 ]
Position, x
• Fick's First Law:
Diff
us on coe
i
ffi
c en
i
t
[m
2 /
s]
flux in x-dir.
[kg/m 2 -s]
dC
J x = −D
concentration
dx
4
gradient [kg/m ]

Ad

t

ap e

d

from Fig.

5.2(c),

Callister 6e.

• The steeper the concentration profile, the greater the flux!

23 Illustration of the concentration gradient 24 03 Brooks/Cole, a division o©20 f Thomson Learning,
23
Illustration of the
concentration
gradient
24
03 Brooks/Cole, a division o©20
f Thomson Learning, Inc. T
used herein under lichomson
Learning ™ is a trad
ense.emark

1.2kg/m 3

g/m 3

STEADY STATE DIFFUSION

Steady State: the concentration profile doesn't g

chan e with time

J

l

x( e

C

oncen ra on,

.

Steady State:

J

= J

x(left)

x(right)

/ti

me.

) t ti C i th b d
)
t
ti
C
i
th
b
d

ft

,

n

e

ox

J

i

x(r g

x

oesn

’t

h

ht

)

c ange w

• Apply Fick's First Law:

• If J x ) left = J x ) right , then

= −D J x ⎛ ⎞ dC ⎜ ⎟ ⎝ d x ⎠ left
= −D
J x
dC
d
x
left

• stop 12/03= −D J x ⎛ ⎞ dC ⎜ ⎟ ⎝ d x ⎠ left ⎛ ⎞

⎛ ⎞ dC = ⎜ ⎟ ⎝ d x ⎠
dC
= ⎜
d
x

right

• Result: the slope, dC/dx, must be constant

(i.e., slope doesn't vary with position)!

25
25

EX: STEADY STATE DIFFUSION

or 0.015% of C

• S

l

l

tee p ate at 700C with geometry shown:

C 1 = C 2 = 0.8k
C 1
=
C 2 = 0.8k
p ate at 700C with g eometr y shown: C 1 = C 2 = 0.8k

or 0.010% of C

t

ap e

d

Carbon

rich

gas

Steady State = straight line!

Carbon

deficient

gas

Ad

from Fig.

5.4,

Callister 6e.

D=3x10 -11 m 2 /s

• Q: How much carbon transfers from the rich to the deficient side?

10mm 5mm 0 x 1 x 2 C − C kg J = −D 2
10mm
5mm
0 x 1 x
2
C
− C
kg
J = −D
2
1
=
2.4 × 10 −9
x
2
x 1
m 2 s
0.8
− 1.2
− 11
J = −
3
×
10
26
0.01
− 0.005
27
27

NON STEADY STATE DIFFUSION

• Concentration profile,

C(x) chan es w/ time.

,

g

To conserve matter:

J (left)

dx
dx
es w/ time. , g • T o conserve ma tt er: J (left) dx J
es w/ time. , g • T o conserve ma tt er: J (left) dx J

J (right)

Concentration, C, in the box

• Fick's First Law:

J (ri ht) J (left)

g

dx

dJ
dJ

=

dC

dt

dC

= − dx
=
dx

dt

equate

dC

J = − D dx d 2 C

D dx 2

dJ = dx
dJ
=
dx

• Governing Eqn.:

dC

d 2 C

dt = D dx 2

dt = D dx 2

or

(if D does

not vary with x)

28
28

EX: NON STEADY STATE DIFFUSION

• Copper diffuses into a bar of aluminum.

Surface conc., C s of Cu atoms bar pre-existing conc., C o of copper atoms
Surface conc.,
C s of Cu atoms
bar
pre-existing conc., C o of copper atoms
C(x,t)
C
2
z
2
s
− t
erf
(
z
) =
e
dt
π
0
t 3
Adapted from
Fig. 5.5,
t 2
t 1
Callister 6e.
t o
C
o
position, x
General solution:
C(x, t ) − C o = 1 −
x
(1)
er f
Assuming: C s and D
are constants
2
Dt
C s − C o

Co: Constant when t=0; Cs: Concentration at surface it is independent to time.

Co: Constant when t=0; Cs: Concentration at surface it is independent to time. 29
Co: Constant when t=0; Cs: Concentration at surface it is independent to time. 29
Co: Constant when t=0; Cs: Concentration at surface it is independent to time. 29
29
29
 

Error Function Value

 
 

Tabulation of Error Function Values

       

z

erf(z)

z

 

erf(z)

z

erf(z)

 

0

0

0.55

 

0.5633

1.3

0.934

 

0.025

0.0282

0.6

 

0.6039

1.4

0.9523

 

0.05

0.0564

0.65

 

0.642

1.5

0.9661

 

0.1

0.1125

0.7

 

0.6778

1.6

0.9763

 

0.15

0.168

0.75

 

0.7112

1.7

0.9838

 

0.2

0.2227

0.8

 

0.7421

1.8

0.9891

 

0.25

0.2763

0.85

 

0.7707

1.9

0.9928

 

0.3

0.3286

0.9

 

0.797

2

0.9953

 

0.35

0.3794

0.95

 

0.8209

2.2

0.9981

 

0.4

0.4284

1

 

0.8427

2.4

0.9993

 

0.45

0.4755

1.1

 

0.8802

2.6

0.9998

 

0.5

0.5205

1.2

 

0.9103

2.8

0.9999

 

x

 

z =

2 Dt
2
Dt

When z<0 6 erf(z) z

 
   

.

,

 
  30
  30
  30
30
30

Factors affect the diffusion

C(x, t ) C o = 1

C s C o

er f

⎝ ⎜

x ⎞ ⎟ ⎠ 2 Dt
x ⎞
2
Dt

1)

Diffusion coefficient

2)

C o original carbon content in the steel

3) C s Surface concentration

4)

X distance from Surface

5)

t: Time

4) X distance from Surface 5) t: Time Adapted from Fig 5.6 Callister 6e 31 DIFFUSION

Adapted from Fig 5.6 Callister 6e

31
31

DIFFUSION AND TEMPERATURE

• Diffusivity increases with T.

diffusivity

TEMPERATURE • Diffusivity increases with T. diffusivity D = D o exp ⎜ − Q d

D= D o exp ⎜ − Q d RT

pre-exponential [m 2 /s] (see Table 5.2, Callister 6e) activation energy

⎞

[J/mol],[eV/mol] (see Table 5.2, Callister 6e)

• Experimental Data:

C a l l i s t e r 6 e ) • Experimental Data: gas

gas constant [8.31J/mol-K]

Al in Al e C i n α-Fe C in γ-F Cu Zn in T(C)
Al
in Al
e
C i n α-Fe
C in γ-F
Cu
Zn in
T(C)
Cu in Cu
Fe
Fe in α-Fe
in γ-Fe
10 -8
D (m 2 /s)
-14
10
10
-20
0.5
1.0
1.5
2.0
15
10
00
00
60
30
0
0

D has exp. dependence on T Recall: Vacancy does also!

D interstitial >> D substitutional

C in α-Fe C in

γ -Fe

Cu in Cu Al in Al

Fe in α-Fe Fe in γ-Fe Zn in Cu 32
Fe in α-Fe
Fe in γ-Fe
Zn in Cu
32

1000K/T

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.)

©2003 Brooks/Co le, a division of Thomson L earning, Inc. Thomson Lear ™ is a
©2003 Brooks/Co le, a division of Thomson L earning, Inc. Thomson Lear
™ is a trademark used h erein under license.ning

. Arrhenius plot of in (rate) versus 1/T can be used to determine the activation energy required for a

Fi

5 8

Th

e

gure

the activation energy required for a Fi 5 8 Th e gure 33 33 reaction 148000
the activation energy required for a Fi 5 8 Th e gure 33 33 reaction 148000
the activation energy required for a Fi 5 8 Th e gure 33 33 reaction 148000
33
33

33 reaction

148000 Diffusion coefficient − 5 D = 2.3 × 10 exp( − ) 8.31×1173 Q:
148000
Diffusion coefficient
− 5
D =
2.3
×
10
exp(
)
8.31×1173
Q: Using D o and T to calculate
34

Diffusion coefficient

35
35

Carburizing

Three steps:

• Transfer C from the gas to steel surface

• C diffusion from surface to interior of steel section

• Quench/tempering treatment to achieve hard case with a tough interior

Aim

To obtain specified carbon profile and hardness distribution through section thickness of the component

36
36

MECHANICAL PROP: Fe-C SYSTEM (1)

• Effect of wt%C

MECHANICAL PROP: Fe - C SYSTEM (1) • Effect of wt%C C o <0.77wt%C Hypo eutectoid

C o <0.77wt%C

Hypoeutectoid

Pearlite (med) ferrite (soft)

ti

orpora on.)

eutectoid Pearlite (med) ferrite (soft) ti orpora on .) Adapted from Fig. 9.27, Callister 6 e

Adapted from Fig. 9.27,Callister 6e. (Fig. 9.27 courtesy Republic ee

St

l C

6 e . (Fig. 9.27 courtesy Republic ee St l C Pearlite (med) Cementite (hard) C
6 e . (Fig. 9.27 courtesy Republic ee St l C Pearlite (med) Cementite (hard) C

Pearlite (med) Cementite (hard)

C o >0.77wt%C

Hypereutectoid

Adapted from Fig. 9.30,Callister 6e. (Fig. 9.30 copyright 1971 by

U

it

d St

t

St

l C

ti

 

n

e

a es

ee

orpora

on.)

Hypo Hyper Hypo Hyper TS(MPa) %EL 1100 80 YS(MPa) 100 900 hardness 40 700 50
Hypo
Hyper
Hypo
Hyper
TS(MPa)
%EL
1100
80
YS(MPa)
100
900
hardness
40
700
50
500
0
300
0
0
0.5
1
0
0.5
1
wt%C
wt%C
• More wt%C: TS and YS increase, %EL decreases.
0.77
0.77
Im
pact ene rgy (Izod, ft-lb)

Adapted from Fig.

10.20, Callister 6e.

(Fig. 10.20 based on

data from Metals

Handbook: Heat Treating Vol 4 9th

,

.

,

ed., V. Masseria (Managing Ed.), American Society for Metals, 1981, p. 9.)

 
  37
  37
  37
37
37
 

MECHANICAL PROP: Fe-C SYSTEM (2)

 

• Fine vs coarse pearlite vs spheroidite

 
 

320

Hypo

Hyper

 

90

Hypo Hyper h

Hypo

Hyper

h

 
 
fine pearlite
fine
pearlite

0

0.5

1

 

R)

Brinel l hardnes s

   

idit

240

coarse

Du ctility (%A

60

sp ero

e

 

pearlite

   

spheroidite

160

30

coarse

pearlite

 
 

fine

 

80

pearlite

 
   

0

0

0.5

1

wt%C

 

wt%C

• Hardness: fine > coarse > spheroidite

Adapted from Fig. 10.21, Callister 6e. (Fig. 10.21 based on data from

 
 

Metals Handbook: Heat Treating,

• %AR:

fine < coarse < spheroidite

Vol. 4, 9th ed., V. Masseria (Managing Ed.), American Society for Metals, 1981, pp. 9 and 17.)

 
  38
  38
  38
38
38

MECHANICAL PROP: Fe-C SYSTEM (3)

• Fine Pearlite vs Martensite:

Hypo Hyper 600 martensite 400 200 fine pearlite 0 0 0.5 1 wt%C Brinell hardness
Hypo
Hyper
600
martensite
400
200
fine pearlite
0
0
0.5
1
wt%C
Brinell hardness

Adapted from Fig. 10.23,

Callister 6e. (Fig. 10.23

 

d

t

d f

Ed

C

B

i

a

ap e

rom

gar

.

a n,

Functions of the Alloying Elements in Steel, American

Society for Metals, 1939, p. 36;

and R.A. Grange, C.R. Hribal, and L.F. Porter, Metall. Trans. A, Vol. 8A, p. 1776.)

• Hardness: fine pearlite << martensite.

39
39
Martensite
Martensite
Hardness: fine pearlite << martensite. 39 Martensite (b) (a) H ar d ness o f mar

(b)

fine pearlite << martensite. 39 Martensite (b) (a) H ar d ness o f mar t

(a) Hardness of martensite, as a function of carbon content. (b) Micrograph of martensite containing 0.8% carbon. The gray platelike regions are martensite; they have the same composition as the original austenite (white regions). Magnification: 1000X. Source: Courtesy of USX Corporation.

40
40

Quenching test

Quenching test   41   WHY Quench is required • Fast cooling is required to form
 
  41
  41
  41
41
41
 

WHY Quench is required

Fast cooling is required to form Martensite

60 40 20 0 1 2 3 A → P Hardness, H RC
60
40
20
0
1
2
3
A → P
Hardness, H RC

0%

distance from quenched end (in)

100%

T(°C)

600

400

Adapted from

60 40 20 0 1 2 3 A → P Hardness, H RC 0% distance from
 
 

Fig. 11.12,

Callister 6e.

 

H.

n

n

Ma Ma Pearlit

Fine

rtensite

e

rtensite

Pearlite

(Fig. 11.12

M(start)

200

adapted from

Boyer (Ed.)

Atlas of

Isothermal

Transformatio

and Cooling

0

A → M
A → M

M(finish)

Transformatio

Diagrams,

American

Society for

Metals, 1977,

Only the position with ΔT/t > critical cooling rate will change to Martensite

 

p.

+ Pe

376.)

0.1

1

10

100

1000

arlite

Time (s)

arlite Time (s)
arlite Time (s) 42
arlite Time (s) 42
42
42

Properties of Oil-Quenched Steel

Figure 4.25 Mechanical properties of oil-quenched 4340 steel, as a function of tempering temperature. Source:

Courtesy of LTV Steel Company

43
43

Surface carbon concentration

C s : during process, it is the maximum solubility of carbon or nitrogen in iron at the carburizing or nitriding temperature.

It shows the approximate limits of carbon solubility in austenite for 8 common steels
It shows the
approximate
limits of
carbon
solubility in
austenite for 8
common steels

Q: Determine Cs values for 4820, 1020 and 3115 steels at 800 and 900 C.

44
44

Effective case depth

While we can specify the total case depth, a more meaningful specification in carburized parts is to require a certain hardness at a specific depth, x, from the surface, which is called the effective case depth.

In steels, the hardness specification is equivalent to carbon content. Thus, an effective case depth is defined as that depth at which a 0.40 wt% C concentration is attained. In both case-depth definitions, a particular value of x can be achieved by varying the product, Dt, to give the particular z value for the particular C. Thus, for a desired case depth, x, the parameters to be controlled are D and t.

However, D depends on temperature.

The actual process parameters are the temperature and time of carburization.

While there is an infinite number of combinations of these two variables, the desired properties of the carburized case are obtained when the temperature of carburization is limited to about 900-950°C.

C(x t) C

,

o =

C s C o

⎛ x ⎞ ⎜ ⎟ 1 − erf ⎝ ⎠ 2 Dt 45
x ⎞
1 − erf
2
Dt
45

Limitation of sample size

The case depth discussed up to this point is for a concentration gradient beneath a plane surface on a solid of infinite magnitude. For solid slabs with finite dimensions and diffusion from both surfaces, the equations are excellent

approximations for case depth as long as

where 2L is the thickness of the slab. For the equation (1) to be valid, the finite thickness of the slab must be greater than twice the total case depth.

Dt
Dt

2 L

< 0.2

− 12 5.9 × 10 × 2 × 3600 = 0.00103m ≈ 1mm 0.2
− 12
5.9
× 10
×
2
×
3600
=
0.00103m
1mm
0.2

Ex. At 900 C, 2 hour, 2L >

(use the data in Table 5.2)

The curvature of the surface being carburized also influences the case depth when the radius of curvature is comparable in magnitude with the case depth.

For convex surfaces, the case depth obtained is greater than that expected on plane surfaces.

For concave surfaces, the case depth is lesser than that expected from plan surfaces.

o

d on plane surfaces. For concave surfaces , the case depth is lesser than that exp
d on plane surfaces. For concave surfaces , the case depth is lesser than that exp
46
46

PROCESSING QUESTION-1

• 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?

it take to g et the same C (x) if we processed at 500C? Constant Key

Constant

Key point 1: C(x,t 500C ) = C(x,t 600C ). Key point 2: Both cases have the same C o and C s .

• Result: Dt should be held constant

.

C(x,t) − C o = 1 − erf ⎛ ⎜ ⎞ x ⎟ (Dt) 500ºC
C(x,t) − C o = 1 − erf
x
(Dt) 500ºC =(Dt) 600ºC
2
Dt
C s − C o
Constant
Note: values
-13
5.3x10
m
2
/s
10hrs
• Answer:
of D are
provided here.
t 500 = (Dt) 600 = 110hr
D
500
4.8x10 -14 m 2 /s
47

PROCESSING QUESTION-Carburizing C-γFe

• 10 hour required at 900 C / Cost $1000/hour(500 parts).

• To get same C(x)at 1000 C, cost $1500/hour (500 parts).

• Is it economical to operate at 1000 C?

• What other factor must be considered? Key point 1: C(x,t 900C ) = C(x,t 1000C ). Key point 2: Both cases have the same C o and C s .

• Result: Dt should be held constant .


C(x,t) C o = 1 erf

C s C o

x ⎞ ⎟ ⎠ 2 Dt
x
2
Dt

(Dt) 900ºC =(Dt) 1000ºC

• Answer:

t

1273

=

0 exp

137800

8.31 1173

×

137800

D

0 exp(

)

8.31×1273

= 3.3 hours

D

e

(

D e ( ) × 1 0

)

×

10

1173 × − 137800 D 0 exp( ) 8.31 × 1273 = 3.3 hours D e
48
48

QUESTION-continued

At 900°C, the cost per part is ($1000/h) (10 h)/500 parts = $20/part

At 1000°C, the cost per part is ($1500/h) (3.3 h)/500 parts = $9.90/part

Considering only the cost of operating the furnace, increasing the temperature reduces the heat-treating cost of the gears and increases the production rate.

Another factor to consider is if the heat treatment at 1000°C could cause microstructural or some other changes? For example, would increased temperature cause grains to grow significantly? If this is the case, we will be weakening the bulk of the material. How does the increased temperature affect the life of the other equipment such as the furnace itself and any accessories? How long would the cooling take? Will cooling from a higher temperature cause residual stresses? Would the product still meet all other specifications? These and other questions should be considered. The point is, as engineers, we need to ensure that the solution we propose is not only technically sound and economically sensible, it should recognize and make sense for the system as a whole (i.e., bigger picture). A good solution is

f

i

l

l

o ten s mp e, so ves pro

bl

f

h

ems or t e system, an

d d

oes not create new pro

bl

ems.

49 50
49
50

1

⎛ ⎞ C(x t) − C , x o ⎜ ⎟ = 1 − erf
C(x t) − C
,
x
o
= 1 − erf
C s − C o
2
Dt ⎠
3
0
.
5
× 10 −
− 11
2
1.6
× 10
t
0.8 − 0.25

1.2

0.25

0.4 z

0.4

z = 0.392

0.35

=

0.4284

0.421

0.4284

0.3794

= 0.00755

51
51

Worked example

A 25-mm diameter 8620 steel bar was carburized at 900°C for eight hours. The diffusion coefficient of carbon in austenite is

D =

16.2 exp(

137800

8.314 T

)

Determine:

(1)

the location in the carburized case where a quenched hardness of 54 HRC may be obtained with a minimum of 95 percent martensite;

whether the desired hardness can be obtained by quenching in agitated water or in agitated oil?

(2)

52
52

Solution-1

The as-quenched hardness is 54 HRC.

Then, we use Fig. 1 to convert the asquenched hardness to carbon content because the hardness of as-quenched martensite is only a function of carbon content. The carbon content is found to be 0.45 percent C = C(x,t) in the carburizing equation

to be 0.45 percent C = C(x,t) in the carburizing equation 1 53 Solution-2 C (
1
1
53 Solution-2 C ( x , t ) − C x 0 = 1 −
53
Solution-2
C
(
x , t
) −
C
x
0
=
1
− erf
(
)
C
C
s −
2
Dt
0
from which we can solve for x, the location of C(x,t) = 0.45% of C. At 900°C
(1173 K) Cs = 1.24 percent C for 8620 steel from Fig. A and the diffusion
coefficient is
,
137800
−5
2
D
=
16.2 exp(
)
=
1.183
×
10
mm
/ s
8.314
×
(900
+
273)
0.45
− 0.2
x
Then
=
1
− erf
(
)
1.24
− 0.2
2
Dt
x
and
erf (
)
=
1
0.24
=
0.76
2
Dt
x
And from erf table
= 0.84
2
Dt
5
and
x =
1.68 1.183
×
10
×
8 × 3600
=
0.9806 mm
54

Solution-3

(2) We need to know now whether we can obtain the desired as-quenched hardness of 54 HRC at 0.98 mm from the surface by quenching in water or oil.

We need to know the Jominy equivalent cooling rate at this location and then use a hardenability curve. Rounding the location to 1 mm, this location from the centre of the 25-mm diameter bar is (11.5/12.5 = 0.92R).

For agitated water quenching, the Jominy equivalent rate at 0.92R (very close to surface) is 1mm.

To get the hardness at this location we look for the hardenability curve 8645 not 8620 since C = 0.45% at this location not 0.2% . The minimum at l mm for 8645 from Fig. 9-51 it is 57 HRC.

Doing the same for the agitation in oil and using Fig. 9-53, the Jominy equivalent cooling rate at 0.92R location is found to be 2.5 mm, and using 0.45% C, is 56 HRC.

We see that 57 HRC for water quenching and 56 HRC for oil quenching both
We see that 57 HRC for water quenching and 56 HRC for oil quenching both
exceed the 54 HRC. Therefore, we can use either quench, because oil
quenching is less drastic, we should use oil quenching.
55
Water quenched
oil quenched
56
57 58
57 58
57
58
59
59

DIFFUSION DEMO: ANALYSIS

• The experiment: we recorded combinations of t and x that kept C constant.

t o t 1 t 2 t 3 x o x 1 x 2 x
t
o
t
1
t
2
t
3
x o
x 1
x 2
x 3
C(x i ,t i ) − C o = 1− erf
C s − C o
⎜ x i
= (constant here)
⎝ ⎜ 2
Dt i ⎠

• Diffusion depth given by:

x i

Dt i 60
Dt i
60

DATA FROM DIFFUSION DEMO

4

3.5

3

2.5

2

1.5

1

0.5

0

B B B B B B B B B B B B B B Linear
B
B
B B
B
B
B
B
B
B B
B
B
B
Linear regression
fit to data:
ln[x(mm)] = 0.58ln[t(min)] + 2.2
=
2
R
0.999
0
0.5
1
1.5
2
2.5
3

ln[t(min)]

• Experimental result: x ~ t 0.58

• Theory predicts x ~ t 0.50

• Reasonable agreement!

61
61

Other Applications

Processing of microelectronic circuits

Applications Processing of microelectronic ci rcu i ts Schemati cs o f the lith ograp hic

Schematics of the lithographic methods to create metallic pattens and selected areas for infusion of dopent atoms

The ability to produce a large number of circuits on such a small surface arises from the techniques of masking and then patterning by lithography. The procedure is illustrated here. The "mask" used is the oxide of silicon that is grown by thermal oxidation, referred to as thermox in the industry. The thicknessof this oxide can be carefully controlled from previous experience. A layer of an organic material called photoresist is applied over the oxide layer on which lithography is done. Webster's Dictionary defines lithography as the process of printing from a plane surface (smooth stone or metal plate) on which the image to be printed is ink-receptive and the blank area ink- repellant. In microelectronics processing, a masking pattern is placed over the photoresist and ultraviolet light is passed through. Depending on whether the photoresist is positive (or negative), the area exposed (or unexposed) is washed away by a suitable developer to provide a window over the oxide. The oxide is etched away by ydrofluoric acid to expose the silicon surface onto which dopants are predeposited and driven-in or where metallic interconnect is deposited.

62
62

der license.s

a trademark used herein unInc.

Thomson Learning iision

of Thomson Learning,©2003

Brooks/Cole, a div

Example: Silicon Device Fabrication

Devices such as transistors are made by doping semiconductors with different dopants to generate regions that have p- or n-type semiconductivity.[1] The

diffusion coefficient of phosphorus (P) in Si is D = 65 × 10 -13 cm 2 /s at a temperature of 1100 o C. Assume the source provides a surface concentration of

20

3

10 atoms/cm and the diffusion time is one hour. Assume that the silicon

wafer contains no P to begin with.

(a) Calculate the depth at which the concentration of P will be 10 18 atoms/cm 3 .

State any assumptions you have made while solving this problem.

(b) What will happen to the concentration pro.le as we cool the Si wafer

containing P?

(c) What will happen if now the wafer has to be heated again for boron (B)

diffusion for creating a p-type region?

63
63
for boron (B) diffusion for creatin g a p-type region? 63 Schematic of a n-p-n transistor.

Schematic of a n-p-n transistor. Diffusion plays a critical role in formation of the different regions created in the semiconductor substrates. The creation of millions of such transistors is at the l

h

t

ear

o

f

i

l

t

i

t

h

m croe ec ron cs ec no ogy

64
64

SOLUTION

SOLUTION   65 SUMMARY:   STRUCTURE & DIFFUSION   Diffusion FASTER f or Diffusion
SOLUTION   65 SUMMARY:   STRUCTURE & DIFFUSION   Diffusion FASTER f or Diffusion
 
  65
  65
  65
65
65

SUMMARY:

 

STRUCTURE & DIFFUSION

 

Diffusion FASTER for

Diffusion SLOWER for

 

• open crystal structures

• close-packed structures

• lower melting T materials

• higher melting T materials

 

• materials w/secondary bonding

materials w/covalent bonding

 

• smaller diffusing atoms

• larger diffusing atoms

 

• cations

• anions

lower density materials

higher density materials

 
 
  66
  66
  66
66
66