Essentials of Materials Science and

Engineering SI Edition 3rd Edition

Full download at link: https://testbankpack.com/p/solution-
manual-for-essentials-of-materials-science-and-engineering-si-
edition-3rd-edition-by-askeland-and-wright-isbn-1111576866-
9781111576868/

Chapter 5
Atom and Ion Movements in Materials
5–4 Atoms are found to move from one lattice position to another at the rate of
5 × 105 jumps/s at 400 ˚C when the activation energy for their movement is
30,000 cal/mol. Calculate the jump rate at 750 ˚C.

Solution:
5 ×10 5 c0exp[ −30, 000/ (1.987)/(673)]
Rate = = = exp( − 22.434 + 14.759)
x c0exp[ − 30,000/(1.987)/(1023)]

5 ×105
= exp( − 7.675) = 4.64 ×10−4
x
5 ×105
x= −4
= 1.08×109 jumps/s
4.64 ×10

5–5 The number of vacancies in a material is related to temperature by an Arrhenius

equation. If the fraction of lattice points containing vacancies is 8 × 10–5 at
600 ˚C, determine the fraction of lattice points containing vacancies at 1000 ˚C.

Solution: 8 × 10–5 = exp[–Q/(1.987)/(873)] Q = 16,364 cal/mol

f = nv/n = exp[–16,364/(1.987)/(1273)] = 0.00155

5–6 The diffusion coefficient for Cr+3 in Cr2O3 is 6 × 10–15 cm2/s at 727 ˚C and
1 × 10–9 cm2/s at 1400 ˚C. Calculate
(a) the activation energy; and
(b) the constant D0.
Solution: 6×10−15 D exp[−Q/(1.987)/(1000)]
61
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
 (a) = 0

1×10−9 D0 exp[−Q/(1.987)/(1673)]
6 × 10–6 =

exp[–Q(0.000503271 – 0.00030082)] = exp[–0.000202452 Q]

−12.0238 = −0.000202452Q or Q = 59, 390 cal/mol
(b) 1 × 10–9 = D0 exp[–59,390/(1.987)/(1673)] = D0 exp(–17.8657)
1×10 –9 = 1.828×10 –8 D0 or D0 = 0.055 cm 2 /s

1 × 10 −9 = 1.7412 × 10 −8 D 0 or D0 = 0.057 cm 2 / s

62
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–7 The diffusion coefficient for O–2 in Cr2O3 is 4 × 10–15 cm2/s at 1150 ˚C and
6 × 10–11 cm2/s at 1715 ˚C. Calculate
(a) the activation energy; and
(b) the constant D0.
Solution: 4×10−15 D exp[−Q/(1.987)/(1423)]
= 0
6×10−11 D0 exp[−Q/(1.987)/(1988)]

6.67 × 10–5 = exp[–0.0001005 Q]

–9.615 = –0.0001005 or Q = 95, 700 cal/mol
Q

4 × 10–15 = D0 exp[–95,700/(1.987)/(1423)] = D0(2.00 × 10–15)

D0 = 2.00 cm2/s

5–11 A 1-mm-thick BCC iron foil is used to separate a region of high nitrogen gas
concentration of 0.1 atomic percent from a region of low nitrogen gas concentration
at 650 ˚C. If the flux of nitrogen through the foil is
1012 atoms/(cm2 · s), what is the nitrogen concentration in the low concentration
region?
Solution: The flux J is given by
Δc
J = −D
Δx
where D is the diffusion coefficient, c is concentration, and x is position.
The term ∆c/∆x is the concentration gradient. The diffusion coefficient
D is given by
⎛ Q ⎞
D = D0 exp ⎜ − ⎟
⎝ RT ⎠
where D0 is the pre–exponential term, Q is the activation energy, R is
the gas constant, and T is the absolute temperature. From Table 5–1,
the activation energy is 18,300 cal/mol, and D0 is 0.0047 cm2/s. Thus
Q ⎞ Δc
J = −D0 exp ⎛⎜ − ⎟
⎝ RT ⎠ Δx
The concentration is given in atomic percent and must be converted to
a per unit volume basis. For BCC iron, there are two atoms per unit cell
with an axial length of 2.866 × 10–8 cm. Substituting,
⎡ 18, 300 ⎤⎛ c −0.1⎞ 2
1012 = −0.0047 exp ⎢ − ⎥ ⎜ low
⎢ 1.987 923 ( )( ) ⎟ −8 3
⎥⎦ ⎝ 0.1 ⎠ (2.866 × 10 )

and solving for the low concentration of atoms clow:

clow = 0.099995 atomic percent.
63
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–12 A 0.2-mm-thick wafer of silicon is treated so that a uniform concentration
gradient of antimony is produced. One surface contains 1 Sb atom per 108 Si
atoms and the other surface contains 500 Sb atoms per 108 Si atoms. The lattice
parameter for Si is 5.4307 Å (Appendix A). Calculate the concentration gradient
in
(a) atomic percent Sb per cm; and
(b) Sb atoms/(cm3 · cm).
Solution: (1/108 − 500/108 )
Δc/Δx = ×100% = −0.02495 at %/cm
0.02 cm
−22
a0 = 5.4307 Vunit cell = 1.6016 × 10 cm
3

(8 Si atoms/unit cell)(1 Sb/108Si)

c1 = = 0.004995 × 1017 Sb atoms/cm 3
(5.4307 × 10 )
3
−8
cm /unit cell
(8 Si atoms/unit cell)(500 Sb/10 8Si)
c2 = = 2.4974 × 1017 Sb atoms/cm 3
(5.4307 × 10 )
3
−8
cm /unit cell

(0.004995 −2.4974) ×1017

Δc/Δx = = −1.246 × 1019 Sb atoms/(cm 3 ⋅
cm)
0.02 cm

5–13 When a Cu-Zn alloy solidifies, one portion of the structure contains 25 atomic
percent zinc and another portion 0.025 mm away contains 20 atomic percent zinc.
The lattice parameter for the FCC alloy is about 3.63 × 10–8 cm. Determine the
concentration gradient in
(a) atomic percent Zn per cm;
(b) weight percent Zn per cm; and
(c) Zn atoms/(cm3 · cm).
Solution: 20% −25%
(a) Δc/Δx = = −2000 at % Zn/cm
(0.025 mm)(0.1 cm/mm)

(b) We now need to determine the wt% of zinc in each portion:

(20)(65.38 g/mol)
wt% Zn = ×100 = 20.46
(20)(65.38) + (80)(63.54)
(25)(65.38 g/mol)
wt% Zn = ×100 =
(25)(65.38) + (75)(63.54)
25.54
20.46% −25.54%
Δc/Δx = = −2032 wt% Zn/cm
0.0025 cm

64
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
 (c) Now find the number of atoms per cm3:
(4 atoms/cell)(0.2 Zn fraction)
c1 = = 1.67 × 10 22 Zn atoms/cm 3
(3.63 × 10−8cm)3
(4 atoms/cell)(0.25 Zn fraction)
c2 = = 2.09 × 1022 Zn atoms/cm 3
(3.63 × 10 −8cm)3

0.0167 ×10 24 −0.0209 ×10 24

Δc/Δx =
0.0025 cm
Δc/Δx = −1.67 × 1024 Zn atoms/(cm 3 ⋅ cm)

5–14 A 0.0025 cm BCC iron foil is used to separate a high hydrogen content gas from a
low hydrogen gas at 650 ˚C. 5 × 108 H atoms/cm3 are in equilibrium on one side
of the foil, and 2 × 103 H atoms/cm3 are in equilibrium with the other side.
Determine
(a) the concentration gradient of hydrogen; and
(b) the flux of hydrogen through the foil.
Solution:
2 ×103 −5 ×108
(a) Δc / Δx = = −1.969 × 1011 H atoms/(cm 3 .cm)
0.0025
J = − D(Δc/Δx) = −0.0012 exp[ − 3600/(1.987)/(923)]( − 1.969 × 1011 )
(b)
J = 3.3 × 107 H atoms/(cm 2 ⋅ s)

5–15 A 1-mm-thick sheet of FCC iron is used to contain nitrogen in a heat exchanger at
1200 ˚C. The concentration of N at one surface is 0.04 atomic percent and the
concentration at the second surface is 0.005 atomic percent. Determine the flux of
nitrogen through the foil in N atoms/(cm2 · s).
Solution: −8 3
(0.00005 −0.0004) (4 atoms per cell)/(3.589 ×10 cm)
Δc/Δx =
(a) (1 mm)(0.1 cm/mm)
= −3.03 × 10 20 N atoms/(cm 3 ⋅ cm)
(b)

J = −D(Δc/Δx) = −0.0034 exp[ − 34,600/(1.987)/(1473)]( − 3.03 × 1020 )

= 7.57 × 1012 N atom/(cm 2 ⋅ s)

5–16 A 4-cm-diameter, 0.5-mm-thick spherical container made from BCC iron holds
nitrogen at 700 ˚C. The concentration at the inner surface is 0.05 atomic percent and
at the outer surface is 0.002 atomic percent. Calculate the number of grams of
nitrogen that are lost from the container per hour.

65
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
 −8 3
Solution:
[0.00002 −0.0005](2 atoms/cell)/(2.866 ×10 cm)
Δc/Δx =
(0.5 mm)(0.1 cm/mm)
= −8.16 × 1020 N/(cm 3 ⋅ cm)

J = −0.0047 exp[ − 18,300/(1.987)/(973)][ − 8.16 × 10 20 ]

= 2.97 × 1014 N atoms/(cm 2 ⋅ s)
Asphere = 4π r 2 = 4π (2 cm)2 = 50.27 cm 2

t = 3600 s/h
N atoms/h = (2.97 × 1014) (50.27) (3600) = 5.37 × 1019 N atoms/h
(5.37 ×1019atoms) (14.007 g/mol)
N loss = = 1.25×10−3g/h
(6.022 ×10 atoms/mol)
23

5–17 A BCC iron structure is to be manufactured that will allow no more than 50 g of
hydrogen to be lost per year through each square centimeter of the iron at 400 ˚C.
If the concentration of hydrogen at one surface is 0.05 H atom per unit cell and
0.001 H atom per unit cell at the second surface, determine the minimum
thickness of the iron.

Solution: c1 = 0.05 H/(2.866 × 10–8 cm)3 = 212.4 × 1019 H atoms/cm3

c2 = 0.001 H/(2.866 × 10–8 cm)3 = 4.25 × 1019 H atoms/cm3
4.25 ×10 19 −212.4 ×10 19 −2.08 ×10 21
Δc/Δx = =
Δx Δx
2 23
(50 g/cm /yr) (6.022 ×10 atoms/mol)
J= = 9.47 × 1017 H atoms/(cm 2 ⋅
(1.00797 g/mol)(31.536 × 10 s/yr)
6
s)
J = 9.47 × 1017 H atoms/(cm 2 ⋅ s)
= −(0.0012)exp[−3600/(1.987)/(673)](−2.08 × 10 21 /Δx)
Δx = 0.179 cm

5–18 Determine the maximum allowable temperature that will produce a flux of less
than 2000 H atoms/(cm2·s) through a BCC iron foil when the concentration gradient
is –5 × 1016 atoms/(cm3 · cm). (Note the negative sign for the flux.)

Solution:
2000 H atoms/(cm2·s) = –0.0012 exp[–3600/(1.987T)][–5 × 1016 atoms/(cm3·cm)]
ln(3.33 × 10–11) = –3600/(1.987T)
T = –3600/[(–24.12)(1.987)] = 75 ˚K = –198 ˚C

66
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–19 Certain ceramic materials such as those based on oxides of yttrium, barium, and
copper have been shown to be superconductors near liquid nitrogen temperature
(~77 to 100 ˚K). Since ceramics are brittle, it has been proposed to make long wires
of these materials by encasing them in a silver tube. In this work, researchers
investigated the diffusion of oxygen in a compound YBa2Cu3O7. The data in the
temperature range 500 to 650 ˚C are shown below for undoped (i.e., silver free)
samples.
Diffusion
Temperature T ( ˚C)
Coefficient D (cm2/s)
500 2.77 × 10–6
600 5.2 × 10–6
650 9.24 × 10–6
(Source: D.K. Aswal, S.K. Gupta, P.K. Mishra, V.C. Sahni, Superconductor
Science and Technology, 1998, 11 [7], pp. 631-36.)
Assume that these data are sufficient to make a straight line fit for the relationship
between ln(D) and 1/T and calculate the values of the activation energy for diffusion
of oxygen in YBa2Cu3O7 containing no silver.

Solution: The data for ln(D) vs. 1/T are shown below.
Temperature Temperature 1/T D ln D
(˚C) (˚K) (˚K–1) (cm2/s)
500 773 0.001294 0.00000277 –12.7967
600 873 0.001145 0.0000052 –12.1669
650 923 0.001083 0.00000924 –11.5920

When fitted to a straight line, the data for ln(D) vs. 1/T give a slope of
–5460.3, corresponding to an activation energy (Q) of 10.8 kcal/mol
or 45.4 kJ/mol.

67
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–20 Diffusion of oxygen in YBa2Cu3O7 doped with silver was measured. It was seen
that the diffusion of oxygen was slowed down by silver doping, as shown in the
data here.
Diffusion
Temperature T (˚C)
Coefficient D (cm2/s)
650 2.89 × 10–7
700 8.03 × 10–7
750 3.07 × 10–6
(Source: D.K. Aswal, S.K. Gupta, P.K. Mishra, V.C. Sahni, Superconductor
Science and Technology, 1998, 11 [7], pp. 631-36.)
Assume that these data are sufficient to make a straight line fit for the relationship
between ln(D) and 1/T and calculate the values of the activation energy for diffusion
of oxygen in YBa2Cu3O7 containing silver.

Solution: The data for ln D vs. 1/T are shown below.

Temperature Temperature 1/T D ln D
–1 2
(˚C) (˚K) (˚K ) (cm /s)

650 923 0.001083 2.89 × 10–7 –15.0568

700 973 0.001028 8.03 × 10–7 –14.0349
750 1023 0.000978 3.07 × 10–6 –12.6938

When fitted to a straight line, the data for ln(D) vs. 1/T give a slope of
–22,241 corresponding to an activation energy (Q) of 44.1 kcal/mol
or 185 kJ/mol.

68
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–21 Zinc oxide (ZnO) ceramics are used in a variety of applications, such as surge-
protection devices. The diffusion of oxygen in single crystals of ZnO was studied
by Tomlins and co-workers. These data are shown in the table here.
Diffusion
Temperature T (˚C)
Coefficient D (cm2/s)
850 2.73 × 10–17
925 8.20 × 10–17
995 2.62 × 10–15
1000 2.21 × 10–15
1040 5.48 × 10–15
1095 4.20 × 10–15
1100 6.16 × 10–15
1150 1.31 × 10–14
1175 1.97 × 10–14
1200 3.50 × 10–14
(Source-. G.W. Tomlins, J.L. Routbort, and T.O. Mason, Journal of the American
Ceramic Society, 1998, 81[4], pp. 869-76.)
Using these data, calculate the activation energy for the diffusion of oxygen in
ZnO. What is the value of D0 in cm2/s?

Solution: The data for ln(D) vs. 1/T are shown below.
Temperature Temperature 1/T D ln D
(˚C) (˚K) (˚K–1) (cm2/s)
850 1123 0.000890472 2.73 × 10–17 –38.1396
925 1198 0.000834725 8.2 × 10–17 –37.0398
995 1268 0.000788644 2.62× 10–15 –33.5756
1000 1273 0.000785546 2.21 × 10–15 –33.7458
1040 1313 0.000761615 5.48 × 10–15 –32.8377
1095 1368 0.000730994 4.2 × 10–15 –33.1037
1100 1373 0.000728332 6.16 × 10–15 –32.7207
1150 1423 0.000702741 1.31 × 10–14 –31.9662
1175 1448 0.000690608 1.97 × 10–14 –31.5582
1200 1473 0.000678887 3.5 × 10–14 –30.9834

The slope of a straight line fitted to ln(D) vs. 1/T corresponds to an

activation energy (Q) of 65.6 kcal/mol or 275 kJ/mol. The value of D0
is 2.05 × 10–4 cm2/s since ln(D0) = –8.493.

69
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–22 Use the diffusion data in the table below for atoms in iron to answer the questions
that follow. Assume metastable equilibrium conditions and trace amounts of C in
Fe. The gas constant in SI units is 8.314 J/(mol · ˚K).

Diffusion Couple Diffusion Mechanism Q (J/mol) D0 (m2/s)

C in FCC Iron Interstitial 1.38 × 105 2.3 × 10–5
C in BCC Iron Interstitial 8.74 × 104 1.1 × 10–6
Fe in FCC Iron Vacancy 2.79 × 105 6.5 × 10–5
Fe in BCC Iron Vacancy 2.46 × 105 4.1 × 10–4
(a) Plot the diffusion coefficient as a function of inverse temperature (1/T)
showing all four diffusion couples in the table.
(b) Recall the temperatures for phase transitions in iron, and for each case,
indicate on the graph the temperature range over which the diffusion data is
valid.
(c) Why is the activation energy for Fe diffusion higher than that for C diffusion
in iron?
(d) Why is the activation energy for diffusion higher in FCC iron when compared
to BCC iron?
(e) Does C diffuse faster in FCC Fe than in BCC Fe? Support your answer with a
numerical calculation and state any assumptions made.
Solution:
(a) The diffusion coefficient D is plotted as a function of 1/T
Q⎞
according to D = D exp ⎛ − ⎟ , where D0 is the pre–exponential
0 ⎜
⎝ RT ⎠

70
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
 term, Q is the activation energy, R is the gas constant, and T is the
absolute temperature.

(b) The temperature at which BCC iron transforms to FCC iron is

912 ˚C (1185 ˚K), and the temperature at which FCC iron transforms
to iron is 1394 ˚C (1667 ˚K). The graph in (a) shows the lines for
each diffusion couple over the temperature ranges for which the
diffusion data is valid.
(c) C diffuses by interstitial diffusion and Fe by vacancy diffusion.
The C atom is smaller than the Fe atom, and there are more
interstitial sites available than vacancies, making interstitial
diffusion easier than vacancy diffusion. This is reflected in the lower
activation energy for C diffusion in Fe when compared to Fe
diffusion in Fe.
(d) FCC Fe has a higher packing fraction than BCC Fe; hence, the
activation energy is expected to be higher for diffusion in FCC Fe.
(e) We can calculate the value of D at 912 ˚C for interstitial diffusion
of C in both FCC and BCC Fe according to

D = D0 exp⎛⎜ −
Q ⎞
⎟
⎝ RT ⎠
We know the activation energies Q and diffusion coefficients D0
from the table.
DFCC = 2.3 × 10–5 exp[–1.38 × 105/(8.314 × 1183)] = 1.85 × 10–11 m2/s
DBCC = 1.1 × 10–6 exp[–8.74 × 104/(8.314 × 1183)] = 1.52 × 10–10 m2/s
At the transition temperature from BCC to FCC iron (912 ˚C), C
diffuses faster in BCC Fe than in FCC Fe as evidenced by the higher
diffusivity for C in BCC Fe compared to FCC Fe.

71
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–23 The plot has three lines representing grain boundary, surface, and volume self-
diffusion in a metal. Match the lines labeled A, B, and C with the type of
diffusion. Justify your answer by calculating the activation energy for diffusion
for each case.

Solution: The expression for the diffusion coefficient D is given by

D = D0 exp⎛⎜ −
Q ⎞
⎟
⎝ RT ⎠
where D0 is the pre–exponential term, Q is the activation energy, R is
the gas constant, and T is the absolute temperature. Taking the natural
logarithm of both sides,
ln(D) = ln(D0) – (Q/R)(1/T).
The activation energy can be determined from the slope of a ln(D)
versus (1/T) graph.
Slope = –(Q/R) = [ln(D2) – ln(D1)] / [(1/T2) – (1/T1)]
From the graph, the values for the diffusion coefficient and (1000/T)
can be determined. The table below shows these values for the end
points of each line in the graph and calculates the slopes and
activation energies for each line.
Diffusivity 1000/T ln(D) 1/T Q/R Q
Line (m2/s) (˚K–1) (m2/s) (˚K–1) (˚K) (J/mol)
–12
A 10 0.8 –27.631 0.0008
A 10–15 1.0 –34.539 0.0010 34,539 2.87 × 105
B 10–10 1.1 –23.026 0.0011
B 5 × 10–12 1.4 –26.022 0.0014 9,986 8.30 × 104
C 10–8 1.5 –18.421 0.0015
C 10–9 1.8 –20.723 0.0018 7,675 6.38 × 104

72
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
 We know that surface diffusion has the least constraints and hence is
expected to have the lowest activation energy. In this case, line C has
the lowest activation energy and represents surface diffusion. Line A
has the highest activation energy and hence represents volume
diffusion, and line B with the intermediate value of activation energy
represents grain boundary diffusion.

5–24 The diffusion of yttrium ions along grain boundaries in chromium oxide (Cr2CO3)
has been studied by Lesage and co-workers. Their data for grain-boundary
diffusivities are shown here.
Grain-Boundary
Temperature T (˚C) Diffusion
Coefficient D (cm2/s)
800 1.2 × 10–13
850 5.4 × 10–13
900 6.7 × 10–13
950 1.8 × 10–12
1000 4.6 × 10–12
(Source: J. Li, M.K. Loudjani, B. Lesage, A.M. Huntz, Philosophical Magazine A,
1997, 76[4], pp. 857-69.)
(a) From these data, show that the activation energy for grain-boundary diffusion
of yttrium in Cr2CO3 is 190 kJ/cal.
(b) What is the value of the pre-exponential term D0 in cm2/s?
(c) What is the relationship between D and 1/T for the grain-boundary diffusivity
in this temperature range?
(d) At any given temperature, the diffusivity of chromium along grain boundaries
is several orders of magnitude higher than that within the bulk. Is this to be
expected? Explain.

Solution: The data for ln D vs. 1/T are shown below.

Temperature Temperature 1/T D ln D
(˚C) (˚K) (˚K–1) (cm2/s)
800 1073 0.000932 1.2 × 10–13 –29.7513
850 1123 0.000890 5.4 × 10–13 –28.2472
900 1173 0.000853 6.7 × 10–13 –28.0315
950 1223 0.000818 1.8 × 10–12 –27.0432
–12
1000 1273 0.000786 4.6 × 10 –26.105

The slope of a straight line fitted to ln(D) vs. 1/T corresponds to an

activation energy (Q) of 46.15 kcal/mol or 193 kJ/mol. The value of
D0 is 3.49 × 10–4 cm2/s since ln(D0) = –7.9618.

73
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
 One way to write this relationship is
⎛ Q ⎞ −4 ⎛ −46,150 ⎞
D = D0 exp ⎜ − ⎟ = 3.49 × 10 exp
⎝ RT ⎠ 1.987T

Yes, in general, there is more disorder at grain boundaries, and hence,

diffusion of atoms or ions is faster compared to that in the bulk.

5–25 Explain why a rubber balloon filled with helium gas deflates over time.

Solution: A balloon deflates due to the ease with which the small helium atoms
diffuse between the loosely-packed chains.

5–26 Consider a 2-mm-thick silicon wafer to be doped (infused with impurities) using
antimony. Assume that the dopant source (gas mixture of antimony chloride and
other gases) provides a constant concentration of 1022 atoms/m3. We need a
dopant profile such that the concentration of Sb at a depth of 1 micrometer is
5 × 1021 atoms/m3. What will be the time for the diffusion heat treatment?
Assume that initially the silicon water contains no impurities or dopants. Assume
that the activation energy for diffusion of Sb in silicon is 380 kJ/mol and D0 for
Sb diffusion in Si is 1.3 × 10–3 m2/s. Assume T = 1250 ˚C.
Solution:
10 22 − 5 × 10 21 ⎧ −6 ⎫
1 ×10
= erf ⎨ ⎬
10 22 − 0
⎪ ⎪

⎩⎪ 2 1.3 × 10 exp −380,000 / (8.314 ) / (1523)⎤⎦ t ⎪⎭

−3
Solving for t:
t = 9130 s = 2.5 h

74
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–27 Compare the diffusion coefficients of carbon in BCC and FCC iron at the
allotropic transformation temperature of 912 ˚C and explain the difference.

Solution: DBCC = 0.011 exp[–20,900/(1.987)/(1185)] = 1.54 × 10–6 cm2/s

DFCC = 0.23 exp [–32,900/(1.987)/(1185)] = 1.97 × 10–7 cm2/s
The packing factor of the BCC lattice (0.68) is less than that of the FCC
lattice (0.74); consequently, atoms are expected to be able to diffuse
more rapidly in the BCC iron.

5–29 A carburizing process is carried out on a 0.10% C steel by introducing 1.0% C at

the surface at 980 ˚C, where the iron is FCC. Calculate the carbon content at
0.01 cm, 0.05 cm, and 0.10 cm beneath the surface after 1 h.

Solution: D = 0.23 exp[–32,900/(1.987)/(1253)] = 4.2 × 10–7 cm2/s

1 −cx ⎧⎪ x ⎫⎪
= erf ⎨ ⎬
1 − 0.1 ⎪⎩ 2 0.23exp [−32, 900 / (1.987) / (1253)](3600 ) ⎪⎭

⎧⎪
1 −cx 0.01 ⎫⎪
x = 0.01 : = erf
⎨ ⎬
1 − 0.1 ⎩⎪ 2 0.23exp [−32, 900 / (1.987) / (1253)](3600 ) ⎭⎪
cx = 0.87% C

1 −cx ⎧⎪ 0.05 ⎫⎪
x = 0.05 : = erf ⎨ ⎬
1 − 0.1 ⎩⎪ 2 0.23exp [−32, 900 / (1.987) / (1253)](3600 ) ⎭⎪
cx = 0.43% C
1 −cx ⎧⎪ 0.10 ⎫⎪
x = 0.10 : = erf ⎨ ⎬
1 − 0.1 ⎪⎩ 2 0.23exp [−32, 900 / (1.987) / (1253)](3600 ) ⎪⎭
cx = 0.16% C

75
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–30 Iron containing 0.05% C is heated to 912 ˚C in an atmosphere that produces
1.20% C at the surface and is held for 24 h. Calculate the carbon content at
0.05 cm beneath the surface if
(a) the iron is BCC; and
(b) the iron is FCC. Explain the difference.

Solution: t = (24 h)(3600 s/h) = 86,400 s

DBCC = 0.011 exp[–20,900/(1.987)/(1185)] = 1.54 × 10–6 cm2/s
DFCC = 0.23 exp[–32,900/(1.987)/(1185)] = 1.97 × 10–7 cm2/s
1.2 −cx
BCC : = erf [0.05/(2 (1.54×10−6 )(86, 400))] = erf [0.0685] = 0.077
1.2 − 0.05
cx = 1.11% C
1.2 −cx
FCC : = erf [0.05/(2 (1.97 ×10−7 )(86, 400))] = erf [0.192] = 0.2139
1.2 − 0.05
cx = 0.95% C
Faster diffusion occurs in the more loosely packed BCC structure,
leading to the higher carbon content at point “x”.

5–31 What temperature is required to obtain 0.50% C at a distance of 0.5 mm beneath

the surface of a 0.20% C steel in 2 h, when 1.10% C is present at the surface?
Assume that the iron is FCC.
Solution: 1.1−0.5
= 0.667 = erf [0.05 / (2 Dt )]
1.1− 0.2

0.05 / (2 Dt ) = 0.685 or Dt = 0.0365 or Dt = 0.00133

t = (2h)(3600 s / h) = 7200 s
D = 0.00133/7200 =1.85 ×10 −7 = 0.23exp[ − 32,900 / (1.987T )]
exp ( −16,558/T) = 8.043×10−7
T =1180 °K= 907 °C

5–32 A 0.15% C steel is to be carburized at 1100o C, giving 0.35% C at a distance of

1 mm beneath the surface. If the surface composition is maintained at 0.90% C,
what time is required?
Solution: 0.9 −0.35
= 0.733 = erf[0.1/(2 Dt )]
0.9 − 0.15

0.1/(2 Dt ) = 0.786 or Dt = 0.0636 or Dt = 0.00405

D = 0.23 exp[–32,900/(1.987)/(1373)] = 1.332 × 10–6 cm2/s

76
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
 t = 0.00405/(1.332 × 10–6) = 3040 s = 51 min

77
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–33 A 0.02% C steel is to be carburized at 1200 ˚C in 4 h, giving 0.45% C at a
distance of 0.6 mm beneath the surface. Calculate the carbon content required at the
surface of the steel.
Solution: cs −0.45
= erf [0.06/(2 Dt )]
cs − 0.02

D = 0.23 exp[–32,900/(1.987)/(1473)] = 3.019 × 10–6 cm2/s

t = (4 h)(3600) = 14,400 s
Dt = (3.019×10 −6 )(14, 400) = 0.2085

erf{0.06/[(2)(0.2085)]} = erf(0.144) = 0.161

cs −0.45
= 0.161 or cs = 0.53% C
cs − 0.02

5–34 A 1.2% C tool steel held at 1150 ˚C is exposed to oxygen for 48 h. The carbon
content at the steel surface is zero. To what depth will the steel be decarburized to
less than 0.20% C?
Solution: 0 −0.2
= 0.1667 ∴ x/(2 Dt ) = 0.149
0 −1.2

D = 0.23 exp[–32,900/(1.987)/(1423)] = 2.034 × 10–6 cm2/s

t = (48 h)(3600 s/h) = 1.728 × 105 s
Dt = 0.5929
Then from above, x = (0.149)(2)(0.5929) = 0.177 cm

5–35 A 0.80% C steel must operate at 950 ˚C in an oxidizing environment for which
the carbon content at the steel surface is zero. Only the outermost 0.02 cm of the
steel part can fall below 0.75% C. What is the maximum time that the steel part
can operate?
Solution: 0 −0.75
= 0.9375 = erf [x/(2 Dt )] ∴ x/(2 Dt ) = 1.317
0 − 0.8

0.02/(2 Dt ) = 1.317 or Dt = 0.00759 or Dt = 5.76×10−5

D = 0.23 exp[–32,900/(1.987)/(1223)] = 3.03 × 10–7 cm2/s

t = 5.76 × 10–5 / (3.03 × 10–7) = 190 s

78
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–36 A steel with the BCC crystal structure containing 0.001% N is nitrided at 550 ˚C
for 5 h. If the nitrogen content at the steel surface is 0.08%, determine the
nitrogen content at 0.25 mm from the surface.
Solution: 0.08 −cx
= erf [0.025/(2 Dt )] t = (5 h)(3600 s/h) = 1.8×104 s
0.08 − 0.001

D = 0.0047 exp[–18,300/(1.987)/(823)]
= 6.488×10 –8 cm 2 /s
Dt = 0.0342
erf{0.025/[(2)(0.0342)]}= erf(0.3655) = 0.394
0.08 −cx
= 0.394 or cx = 0.049% N
0.079

5–37 What time is required to nitride a 0.002% N steel to obtain 0.12% N at a distance
of 0.005 cm beneath the surface at 625 ˚C? The nitrogen content at the surface is
0.15%.
Solution: 0.15 −0.12
= 0.2027 = erf [x/(2 Dt )] ∴ x/(2 Dt ) = 0.1816
0.15 − 0.002

D = 0.0047 exp[–18,300/(1.987)/(898)] = 1.652 × 10–7 cm2/s

x = 0.005 cm
0.005
= 0.1816
2 (1.652×10−7 )t
Dt = 1.956 × 10–4 or t = 1.956 × 10–4/(1.652 × 10–7) = 1184 s

5–38 We can successfully perform a carburizing heat treatment at 1200 ˚C in 1 h. In an

effort to reduce the cost of the brick lining in our furnace, we propose to reduce
the carburizing temperature to 950 ˚C. What time will be required to give us a
similar carburizing treatment?

Solution: D1200 = 0.23 exp[–32,900/(1.987)/(1473)] = 3.019 × 10–6 cm2/s

D950 = 0.23 exp[–32,900/(1.987)/(1223)] = 3.034 × 10–7 cm2/s
t1200 = 1 h
(3.019 ×10 −6 )(1)
t950 = D1200t1200 /D950 = = 9.95 h
3.034 ×10−7

79
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–39 Arrange the following materials in increasing order of self-diffusion coefficient:
Ar gas, water, single crystal aluminum, and liquid aluminum at 700 ˚C.

Solution: Single crystal aluminum, liquid aluminum at 700 ˚C, water, and Ar
gas.

5–40 Calculate the constant surface concentrations of phosphorus and boron needed to
achieve a concentration of 1018 atoms/cm3 at a depth of 0.1 µm from the surface
of single crystal silicon using a thermal treatment at 1100 ˚C for 1 hour. Consider
phosphorus and boron separately. The diffusion coefficients of P and B in single
crystal silicon at 1100 ˚C are 6.5 × 10–13 cm2/s and 6.1 × 10–13 cm2/s, respectively.

Solution: For P,
⎡ ⎤
c −1018 cm −3 0.1×10−4 cm
s
= erf ⎢ ⎥
cs − 0 ⎢2
( 6.5×10 -13
cm / s ) ( 3600 s ) ⎥
2

⎣ ⎦
1018
1− = 0.1162
cs

cs = 1.1 × 1018 cm −3
For B,

⎡ ⎤
c −1018 cm −3 0.1×10 −4 cm
s
= erf ⎢ ⎥
cs − 0 ⎢2
( 6.1×10 -13
cm / s ) ( 3600 s ) ⎥
2

⎣ ⎦
1018
1− = 0.1199
cs

cs = 1.1 × 1018 cm −3

80
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
5–41 When processing silicon single crystals for microelectronics applications, precise
quantities of impurities are often introduced at relatively shallow depths by ion
implantation and diffused into the silicon substrate in a subsequent thermal
treatment. This can be approximated as a finite source diffusion problem.
Applying the appropriate boundary conditions, the solution to Fick’s second law
under these conditions is
Q ⎛ x2 ⎞
c (x, t ) = exp −
π Dt 4 Dt

where Q is the initial surface concentration with units of atoms/cm2.

Assume that we implant 1014 atoms/cm2 of phosphorus at the surface of a silicon
wafer with a background boron concentration of 1016 atoms/cm3 and this wafer is
subsequently annealed at 1100 ˚C. The diffusion coefficient (D) of phosphorus in
silicon at 1100 ˚C is 6.5 × 10–13 cm2/s
(a) Plot a graph of the concentration c (atoms/cm3) versus x (cm) for anneal times
of 5 minutes, 10 minutes, and 15 minutes.
(b) What is the anneal time required for the phosphorus concentration to equal the
boron concentration at a depth of 1 μm?
Solution:
(a) Using Q = 1014 atoms/cm2 and D = 6.5 × 10–13 cm2/s in the equation
above, the concentration c for t = 5 min, 10 min, and 15 min can be
calculated using a series of values for “x”. A graph of c versus x can
be made with most standard spreadsheet software.

(b) From the graph, the time for the P concentration to match the
background B concentration is slightly more than 10 minutes.
Software packages are available that can solve this equation, but we
can also find the solution by substituting t values close to

81
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.
 10 minutes. Rearranging the equation above and taking the natural
logarithm of each side
ln(c) = ln[Q/√(πDt)] – x2/(4Dt)
c = 1016 atoms/cm3, and the values of Q and D are known.
For t = 650 s, x = 0.974 μm, which is too small.
For t = 700 s, x = 1.008 μm, which is too large.
For t = 689 s, x = 1 μm.

82
© 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied, or duplicated, or posted to a publicly available website, in whole or in part.