Chapter 9: Principles of Solidification

9-5 What is the difference between homogenous nucleation and heterogeneous

nucleation?
Solution:
Homogenous nucleation occurs when nuclei of the resulting phase have the same
probability of forming at every location. Heterogeneous nucleation of the resulting
phase occurs at certain preferred locations in the original phase. In practice,
homogenous nucleation does not occur. Heterogeneous nucleation commonly occurs at
the surfaces for molds of castings, surfaces of inclusions and/or induced by inoculation.

9-6 From the thermodynamic point of view, what two things must occur for solidification
from liquid to solid to proceed?
Solution:
Thermodynamically speaking, the first thing that must occur is the formation of a
critical-sized nucleus. At this critical size, there is an activation energy that must be
overcome. At the melting or freezing point, this activation energy will approach infinity.
Thus the second thing that must occur is an undercooling in the liquid that is to solidify.
A greater undercooling will lower the activation energy resulting in a greater tendency
to solidify.

9-9 Derive the formula for the change in total energy change (i.e. the derivative of ΔG with
respect to r) as a function of the radius r of the solid nucleus. Begin with Equation 9-1.
Numerically, how is the critical point at which solidification proceeds defined?
Solution:
This is a calculus problem. Starting with the equation:
4
∆ =   ∆ + 4   
3
Differentiating:
∆ 4    
= ∆   + 4   
 3  
∆ 4
= ∆ 3   + 4  2
 3
∆
= 4  ∆ + 8 

The critical point for solidification is when the radius reaches the critical radius. This is
also when ΔG reaches an inflection point, so:
∆
= 0;  > 0


9-10 Using Equation 9-2, prove that interfacial energy (σsl) has units of J/m2 in SI.
Solution:
Equation 9-2 is

197
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 2  
∗ =
∆ ∆
Solving for the interfacial energy:
 ∗ ∆ ∆
  =
2
In SI, the units are as follows:
Name SI units
Critical radius m
Freezing temperature K
Heat of fusion J/m3
Temperature difference K

Inserting these into the previous equation:

m J K 1
  = "  $ $
1 m 1 K
J
  = 
m

9-11 Calculate the total interfacial surface energy for 1016 spheres of copper, each with a
radius of r*.
Solution:
Example 9-1 found the critical radius of copper to be 12.51 × 10-8 centimeters.
Calculating the total surface area of a single sphere:
& = 4  = 412.51 × 10*+ cm = 1.9110*/ cm
The total surface area for all spheres:
0 & = 10/1 1.9210*/ cm = 2310 cm
Using the interfacial energy from Table 9-1:
2410 cm 177710*8 J
5 = 35 mJ
1 1 cm

9-12 Of the ferrous elements, which has the lowest undercooling required for homogeneous
nucleation? Does this have any practical significance?
Solution:
Between iron and nickel, the lowest ΔT in Table 9-1 belongs to iron at 420 °C. This does
not have a great practical significance in itself because homogeneous nucleation is not
seen in industry or commerce. Heterogeneous nucleation is seen in production.

*With apologies, as written, this problem statement does not make sense. Perhaps it
should have been of the “ferromagnetic elements.”

9-13 If the total change in free energy of a molten metal being cooled is 5.34 × 10-17 J upon
formation of the first stable solid and the free energy per unit volume is -17.7 J/cm3,
approximate the radius of the spherical solid. The surface free energy of the solid-liquid
interface is 100 × 10-7 J/cm2.

198
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Solution:
This is a straightforward use of Equation 9-1, however making it explicit in r is difficult
algebraically. Instead a numerical solution using a spreadsheet’s goal seek function or
calculator’s SOLVE application is advised. The result will be 1.16 × 10-6 cm.

9-14 Using the densities in Appendix A, convert the heats of fusion in Table 9-1 from units
J/cm3 to kJ/kg.
Solution:
While doing the conversions, note that [kJ/kg] are the same as [J/g]
Element Heat of fusion (kJ/kg)
Ga 83
Bi 55
Pb 21
Ag 92
Cu 183
Ni 310
Fe 195

9-15 Assume that instead of a spherical nucleus, we have a nucleus in the form of a cube of
length x. Calculate the critical dimension x* of the cube necessary for nucleation. Write
an equation similar to Equation 9-1 for a cubical nucleus, and derive an expression for x*
similar to Equation 9-2.
Solution:

∆G = x 3∆Gv + 6x 2σ sl

∂G
= 3x 2 ∆Gv +12xσ sl = 0
∂x
∂G
= 3 ( x * ) ∆Gv +12x * σ sl = 0; x > 0
2

∂x x=x*

12 σ 4σ
x* = − = − sl
3∆Gv ∆Gv

Note that ∆Gv is negative such that x * is positive.

A similar treatment for Equation 9-1 for a spherical nucleus produces

∂G
= 4π ( r * ) ∆Gv + 8π r* σ sl = 0; r > 0
2

∂r r=r*

2 σ sl
r* = −
∆Gv

199
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 Comparison to Equation 9-2 shows that

∆H f ∆T
∆Gv = −
Tm
Thus,

4 σ sl 4 σ sl Tm
x* = − =
∆Gv ∆H f ∆T

9-16 Why is undercooling required for solidification? Derive an equation showing the total
free energy change as a function of undercooling when the nucleating solid has the
critical nucleus radius r*.

Solution: Undercooling is required for solidification because the energy to

create a new solid-liquid interface needs to be overcome for a solid
phase to begin forming. The undercooling below the melting
temperature is the driving force that overcomes this barrier.
From Equation 9–1,
∆G = (4/3) πr3 ∆Gv + 4πr2σsl,
where ∆G is the total change in free energy upon solidification, r is
the radius of spherical solid, ∆Gv is the free energy change per unit
volume for solidification, and σsl is the surface energy per unit area of
the solid–liquid interface. From Equation 9–2, the critical radius for
solidification r* is given by
r* = 2σslTm/(∆Hf∆T),
where Tm is the melting temperature, ∆Hf is the latent heat of fusion,
and ∆T is the undercooling. At r = r*,
∆G = (4/3)π [2σ slTm /(∆H f ∆T )]3 ∆Gv + 4π [2σ slTm /(∆H f ∆T )]2 σ sl
∆G = (32π /3)(σ sl3 Tm3 ∆Gv /∆H 3f )(1/∆T 3 ) + 16π (σ sl3 Tm2 /∆H 2f )(1/∆T 2 )

9-17 Why is it that nuclei seen experimentally are often sphere-like but faceted? Why are
they sphere-like and not like cubes or other shapes?
Solution:
Nuclei are sphere-like because spheres have the largest volume to surface area ratio.
Nuclei are faceted to favor crystallographic planes with low surface energies.

9-18 Explain the meaning of each term in Equation 9-2.

Solution:
∆H f is the latent heat of fusion per unit volume, Tm is the equilibrium solidification
temperature in kelvin, and ∆T is the undercooling when the liquid temperature is T.
The latent heat of fusion represents the heat given up during the liquid-to-solid
transformation. σ sl Is the surface energy per area of the solid–liquid interface.

200
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
9-19 Suppose that liquid nickel is undercooled until homogeneous nucleation occurs.
Calculate (a) the critical radius of the nucleus required and (b) the number of nickel
atoms in the nucleus. Assume that the lattice parameter of the solid FCC nickel is 0.356
nm.
Solution: From Table 9–1, ∆Tmax = 480°C
(2)(255 ×10−7 J/cm 2 )(1453 + 273)
r* = = 6.65 ×10−8 cm
(2756 J/cm3 )(480)
ao = 3.56 Å V = 45.118 × 10–24 cm3
Vnucleus = (4π/3)(6.65 × 10–8 cm)3 = 1232 × 10–24 cm3
number of unit cells = 1232/45.118 = 27.3
atoms per nucleus = (4 atoms/cell)(27.3 cells) = 109 atoms

9-20 Suppose that liquid iron is undercooled until homogeneous nucleation occurs. Calculate
(a) the critical radius of the nucleus required and (b) the number of iron atoms in the
nucleus. Assume that the lattice parameter of the solid BCC iron is 2.92 Å.
Solution: (2)(204 ×10−7 J/cm 2 )(1538 + 273)
r* = 3
= 10.128 ×10−8 cm
(1737 J/cm )(420)
V = (4π/3)(10.128)3 = 4352 Å3 = 4352 × 10–24 cm3
Vuc = (2.92 Å)3 = 24.897 Å3 = 24.897 × 10–24 cm3
number of unit cells = 4352/24.897 = 175
atoms per nucleus = (175 cells)(2 atoms/cell) = 350 atoms

9-21 Suppose that solid nickel was able to nucleate homogeneously with an undercooling of
only 22°C. How many atoms would have to group together spontaneously for this to
occur? Assume that the lattice parameter of the solid FCC nickel is 0.356 nm.
Solution: (2)(255 ×10−7 J/cm 2 )(1453 + 273)
r* = 3
= 145.18 ×10−8 cm
(2756 J/cm )(22)
Vuc = 45.118 × 10–24 cm3 (see Problem 9–11)
Vnucleus = (4π/3)(145.18 × 10–8 cm)3 = 1.282 × 10–17 cm3
number of unit cells = 1.282 × 10–17 / (45.118 × 10–24) =
2.84 × 105
atoms per nucleus = (4 atoms/cells)(2.84 × 105 cells) =
1.136 × 106

9-22 Suppose that solid iron was able to nucleate homogeneously with an undercooling of
only 15°C. How many atoms would have to group together spontaneously for this to
occur? Assume that the lattice parameter of the solid BCC iron is 2.92 Å.
Solution: (2)(204 ×10−7 J/cm 2 )(1538 + 273)
r* = 3
= 283.6 ×10−8 cm
(1737 J/cm )(15)
Vuc = 24.897 × 10–24 cm3 (see Problem 9–12)
Vnucleus = (4π/3)(283.6 × 10–8 cm)3 = 95,544,850 × 10–24 cm3

201
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 number of unit cells = 95,544,850/24.897 = 3.838 × 106
atoms per nucleus = (2 atoms/cells)(3.838 × 106 cells) =
7.676 × 106

9-23 Explain the term inoculation.

Solution:

Inoculation is the addition of particles to the liquid to induce the nucleation of equiaxed
grains in a casting. The common inoculants are refractory elements with melting points
much higher than the liquid melt. The inoculants provide surfaces on which
heterogeneous nucleation of solid crystals may occur.

9-35 What is a dendrite and why do dendrites form during solidification?

Solution:
A dendrite is a “tree-like” formation of solid crystals from liquid. The primary dendrite
(main trunk) is due to a preferred growth of crystals in specific crystallographic
directions which occurs opposite the direction of heat extraction. From the main trunk,
branches (secondary dendrites) or dendritic arms form and grow also along preferred
directions into the liquid. For cubic crystals, the easiest growth path are along the <100>
directions.

9-36 Use the data in Table 9–1 and the specific heat data given below to calculate the
undercooling required to keep the dendritic fraction at 0.5 for each metal.

Metal Specific heat [J/(cm3–K)]

Bi 1.27
Pb 1.47
Cu 3.48
Ni 4.75

Solution: The dendritic fraction f is given by Equation 9–3:

f = c∆T/∆Hf,
where c is the specific heat, T is the temperature, and ∆Hf is the latent
heat of fusion.
For f = 0.5, the undercooling required is
∆T = 0.5 ∆Hf/c.
The results for the various metals are shown in the table.

Metal Heat of Fusion ∆Hf Specific heat c ∆T for f = 0.5 (K)

(J/cm3) [J/(cm3–K)]
Bi 543 1.27 213.8
Pb 237 1.47 80.6
Cu 1628 3.48 233.9
Ni 2756 4.75 290.1

202
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
9-37 Calculate the fraction of solidification that occurs dendritically when silver nucleates (a)
at 10°C undercooling; (b) at 100°C undercooling; and (c) homogeneously. The specific
heat of silver is 3.25 J/(cm3·°C).
Solution: c∆T [3.25 J/cm3 ⋅°C](10°C)
(a) f = = = 0.0337
∆Hf 965 J/cm
c∆T [3.25 J/(cm 3 ⋅°C)](100°C)
(b) = = 0.337
∆H f 965 J/cm3
c∆T [3.25 J/(cm3 ⋅°C)](250°C)
(c) = = 0.842
∆H f 965 J/cm3

9-38 Calculate the fraction of solidification that occurs dendritically when iron nucleates (a)
at 10°C undercooling; (b) at 100°C undercooling; and (c) homogeneously. The specific
heat of iron is 5.78 J/(cm3 · °C).
Solution: c∆T [5.78 J/(cm3 ⋅ °C)](10°C)
(a) f = = = 0.0333
∆H f 1737 J/cm3
c∆T [5.78 J/(cm 3 ⋅°C)](100°C)
(b) = = 0.333
∆H f 1737 J/cm3
c∆T [5.78 J/(cm 3 ⋅°C)](420°C)
(c) = = 1.40
∆H f 1737 J/cm3
Therefore, essentially all of the solidification occurs dendritically.

9-39 Analysis of a nickel casting suggests that 28% of the solidification process occurred in a
dendritic manner. Calculate the temperature at which nucleation occurred. The specific
heat of nickel is 4.1 J/(cm3 · °C).
Solution: c∆T [4.1 J/(cm3 ⋅°C)]( ∆T )
f = = = 0.28
∆H f 2756 J/cm3
∆T = 188°C or Tn = 1453 − 188 = 1265°C

9-40 It is desired to increase the dendritic fraction from 0.012 to 0.025. Assuming that the
specific heat of the metal being solidified is constant, determine the increase in the
amount of undercooling required.
Solution:
Equation 9-3 is linear, so we can simply divide the two fractions to find the ratios of ΔT.
:∆
9=
∆
:∆
9 ∆
=
9/ :∆/
∆

203
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 9 ∆
=
9/ ∆/
0.025 ∆
= 2.1 =
0.012 ∆/
So, the undercooling is to be increased by a factor of 2.1. Note that this means the
temperature difference, not the actual temperature the metal is being undercooled to.

9-42 Find the mold constant B and exponent n in Chvorinov’s rule using the following data
and a log–log plot.

Solution: Chvorinov’s rule gives the solidification time ts as

n
V 
ts = B   ,
 A
where B is the mold constant, V is the volume of the casting, A is the
surface area of the casting in contact with the mold, and n is a constant.
Taking the natural logarithm of both sides,
ln (ts) = ln (B) + n ln (V/A)
The values of n and B can be found by plotting ln (ts) versus ln (V / A).
The volume and surface area of each shape is given in the table.

Shape Dimensions (cm) Volume (cm3) Area (cm2) Volume / Area (cm)
Cylinder Radius = 10, 9425 2513 3.75
Length = 30
Sphere Radius = 9 3054 1018 3
Cube Length = 6 216 216 1
Plate Length = 30, 600 1300 0.46
Width = 20,
Height = 1

204
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 From the graph, the slope n = 2.21 and ln (B) = 5.3407 such that B = 209 s/cm2.

9-43 A 2-in. cube solidifies in 4.6 min. Assume that n = 2. Calculate (a) the mold constant in
Chvorinov’s rule; and (b) the solidification time for a 0.5 in. × 0.5 in. × 6 in. bar cast
under the same conditions.
Solution: (a) We can find the volume and surface area of the cube:
V = (2)3 = 8 in.3 A = 6(2)2 = 24 in.2 t = 4.6 = B(8/24)2
B = 4.6/(0.333)2 = 41.48 min/in.2
(b) For the bar, assuming that B = 41.48 min/in.2:
V = (0.5)(0.5)(6) = 1.5 in.2
A = 2(0.5)(0.5) + 4(0.5)(6) = 12.5 in.2
t = (41.48)(1.5/12.5)2 = 0.60 min

9-44 A 5-cm diameter sphere solidifies in 1050 s. Calculate the solidification time for a 0.3 cm
× 10 cm × 20 cm plate cast under the same conditions. Assume that n = 2.
Solution:  (4π /3)(2.5)3 
2

t = 1050 s = B   = B[2.5/3]2 or B = 1512 s/cm 2

 4π (2.5) 
2

(1512)(0.3 ×10 × 20) 2

t= = 1512[60/418]2 = 31.15 s
[2(0.3)(10) + 2(0.3)(20) + 2(10)(20)] 2

9-45 Find the constants B and n in Chvorinov’s rule by plotting the following data on a log-log
plot:

205
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Solution: V (in.3) A (in.2) V/A (in.)
48 212 0.226
60 112 0.536
15.6 37.5 0.416
36 98 0.367
From the graph, we find that
B = 48 min/in.2 and n = 1.72

9-46 Find the constants B and n in Chvorinov’s rule by plotting the following data on a log-log
plot:

Solution: V (cm3) A (cm2) V/A (cm)

6 26 0.23
32 64 0.5
64 96 0.67
240 236 1.02
From the graph, we find that
B = 305 s/cm2 and n = 1.58

206
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
9-47 A 3-in.-diameter casting was produced. The times required for the solid-liquid interface
to reach different distances beneath the casting surface were measured and are shown
in the following table:

Distance from surface d (in.) Time t (s) t

0.1 32.6 5.71
0.3 73.5 8.57
0.5 130.6 11.43
0.75 225.0 15.00
1.0 334.9 18.22
Determine (a) the time at which solidification begins at the surface and (b) the time at
which the entire casting is expected to be solid. (c) Suppose the center of the casting
actually solidified in 720 s. Explain why this time might differ from the time calculated in
part (b).
Solution: We could plot d versus t , as shown, finding tsurface from where the
plot intersects the x-axis and tcenter where the plot intersects d = 1.5
in. Or we could take two of the data points and solve for c and k.
d = k t −c
0.1 = k 32.6 − c
0.5 = k 130.6 − c
−0.4 = k[ 32.6 − 130.6] = −5.718 k
k = 0.070
c = 0.070 32.6 − 0.1 = 0.30

207
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 d = 0 = 0.070 t − 0.30
tsurface = (0.3/0.07)2 = 18.4 s
1.5 = 0.070 t − 0.3
tcenter = (1.8/0.07)2 = 661 s
The mold gets hot during the solidification process, and consequently
heat is extracted from the casting more slowly. This in turn changes the
constants in the equation and increases the time required for complete
solidification.

9-48 What solidification time is required for a casting of 9.5 cm3 and a mold contact area of
10.0 cm2? The mold constant is 838 s/cm2.01.
Solution:
This is an easy application of Chvorinov’s rule. The only curveball is that the exponential
constant must be read from the units of the mold constant.
> @
; = < = ?
&
.B/
s 9.5 cm
; = 838 .B/  
cm 10.0 cm
; = 756 s

9-49 An aluminum alloy plate with dimensions 20 cm × 10 cm × 2 cm needs to be cast with a

secondary dendrite arm spacing of 10–2 cm (refer to Figure 9–6). What mold constant B
is required (assume n = 2)?

208
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Solution: From Figure 9–6, for a secondary dendrite arm spacing of 0.01 cm, the
solidification time is 500 s. Chvorinov’s rule for solidification time
(Equation 9–4) is
ts = B(V/A)n,
where B is the mold constant, V is the volume of the casting, A is the
surface area of the casting in contact with the mold, and n is a
constant. In this case, the volume of the casting is
V = 20 × 10 × 2 = 400 cm3,
and the surface area of the casting is
A = 2 × 20 × 10 + 2 × 2 × 10 + 2 × 2 × 20 = 520 cm2.
Substituting into the equation above and taking n = 2,
ts = B(V/A)n
500 = B(400/520)2
B = 845 s/cm2.

9-50 Figure 9–5(b) shows a micrograph of an aluminum alloy. Estimate (a) the secondary
dendrite arm spacing and (b) the local solidification time for that area of the casting.

209
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Solution: The distance between adjacent dendrite arms can be measured.
Although most people doing these measurements will arrive at
slightly different numbers, the author’s calculations obtained from
four different primary arms are
16 mm / 6 arms = 2.67 mm
9 mm / 5 arms = 1.80 mm
13 mm / 7 arms = 1.85 mm
18 mm / 9 rms = 2.00 mm
average = 2.08 mm = 0.208 cm
Dividing by the magnification of ×50:
SDAS = 0.208 cm / 50 = 4.16 × 10–3 cm
From Figure 9–6, we find that the local solidification time (LST) =
90 s.

9-51 Find the constants k and m relating the secondary dendrite arm spacing to the local
solidification time by plotting the following data on a log-log plot:

Solution: The secondary dendrite arm spacing (SDAS) is given by

SDAS = ktsm ,
where k and m are constants and ts is the solidification time. Taking the
natural logarithm of both sides,
ln (SDAS) = ln (k) + m ln (ts).
The values of k and m can be found by plotting ln (SDAS) versus ln (ts).

210
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 From the graph, the slope m = 0.35 and ln (k) = –5.8013 such that k =
0.003 cm / s.

9-52 Figure 9–25 shows dendrites in a titanium powder particle that has been rapidly
solidified. Assuming that the size of the titanium dendrites is related to solidification
time by the same relationship as in aluminum, estimate the solidification time of the
powder particle.
Solution: The secondary dendrite arm spacing can be estimated from the
photomicrograph at several locations. The author’s calculations,
derived from measurements at three locations, are
11 mm / 8 arms = 1.375 mm
13 mm / 8 arms = 1.625 mm
13 mm / 8 arms = 1.625 mm
average = 1.540 mm
Dividing by the magnification of 2200:
SDAS = (1.540 mm)(0.1 cm/mm) / 2200 = 7 × 10–5 cm
The relationship between SDAS and solidification time for
aluminum is
SDAS = 8 × 10–4 t0.42 = 7 × 10–5
t = (0.0875)1/0.42 = 0.003 s

9-53 The secondary dendrite arm spacing in an electron-beam weld of copper is 9.5 × 10–4
cm. Estimate the solidification time of the weld.
Solution: From Figure 9–6, we can determine the equation relating SDAS and
solidification time for copper:
m = 20/50 = 0.4 k = 4 × 10–3 cm
Then for the copper weld:
9.5 × 10–4 = 4 × 10–3(LST)0.4
(Note: LST is local solidification time)
0.2375 = (LST)0.4 or –1.4376 = 0.4 ln LST
ln LST = –3.5940 or LST = 0.03 s

211
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
9-54 A zinc alloy has a SDAS of 0.01 cm and m = 0.40. What is the solidification time?
Solution:
The m value is a distraction here. By referring to Figure 9-6, we directly read the
solidification time as being about 120 seconds.

9-55 In Figure 9-7, what is the equation (slope intercept form) of the tensile strength line,
and what SDAS would appear to give a tensile strength of zero if the line is
extrapolated? Does extrapolating the trend like this make sense?

Solution:
Since the graph is purely linear, we can easily take two points and find the equation of
the line. Points that land on a gridline can be more accurately read, such as (0.005 cm,
44 ksi) and (0.010 cm, 39.5 ksi) that both fall on vertical gridlines.
Taking the difference to find the slope:
∆D 39.5 ksi − 44 ksi ksi
3= = = −900
∆E 0.010 cm − 0.005 cm cm
Applying the slope to find the x-intercept:
∆D 39.5 ksi − E ksi
3= = = −900
∆E 0.010 cm − 0 cm
39.5 ksi − E ksi
= −900
0.010 cm cm

212
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 ksi E ksi
3950 − = −900
cm 0.010 cm cm
ksi E
4850 =
cm 0.010 cm
E = 48.5 ksi
Together:
ksi
D = =−900 ? E + 48.5 ksi
cm
Where x is in cm and y is in ksi.
To find out what SDAS (x) results in a strength of zero, we set y to zero and solve:
ksi
0 ksi = =−900 ? E + 48.5 ksi
cm
ksi
48.5 ksi = =900 ? E
cm
E = SDAS = 0.0539 cm
The tensile strength for the aluminum alloy will never be zero regardless of how large
the SDAS spacing is. Thus we see that extrapolation can be dangerous since trends
change. The curve will plateau at a value for which the SDAS becomes sufficiently large
that the presence of dendrites no longer causes strength to be increased relative to a
casting that solidified without dendrites.

9-60 In Figure 9-8, what do the slopes of the lines A-B and E-onward represent? Do these
quantities represent physical properties?

213
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Solution:
No. The slopes of these lines represent how fast the temperature is falling in the solid
and liquid phases respectively. They can be changed at will by increasing or decreasing
the rate of heat removal from the samples.

9-61 A cooling curve is shown in Figure 9–26. Determine (a) the pouring temperature; (b) the
solidification temperature; (c) the superheat; (d) the cooling rate, just before
solidification begins; (e) the total solidification time; (f) the local solidification time; and
(g) the probable identity of the metal. (h) If the cooling curve was obtained at the center
of the casting sketched in the figure, determine the mold constant, assuming that n = 2.

Solution: (a) Tpour = 475°C (e) ts = 470 s

(b) Tsol = 320°C (f) LST = 470 – 160 = 310 s
(c) ∆Ts = 475 – 320 = 155°C (g) Cadmium (Cd)
475 − 320 (h) ts = 470 = B[38.4/121.6]2
(d) ∆T /∆t = = 1.0°C/s B = 4713 s/cm2
160 − 0

9-62 A cooling curve is shown in Figure 9–27. Determine (a) the pouring temperature; (b) the
solidification temperature; (c) the superheat; (d) the cooling rate, just before
solidification begins; (e) the total solidification time; (f) the local solidification time; (g)
the undercooling; and (h) the probable identity of the metal. (i) If the cooling curve was
obtained at the center of the casting sketched in the figure, determine the mold
constant, assuming n = 2.

214
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Solution: (a) Tpour = 900°C (e) ts = 9.7 min
(b) Tsol = 420°C (f) LST = 9.7 – 2.5 = 7.2 min
(c) ∆Ts = 900 – 420 = 480°C (g) 420 – 360 = 60°C
900 − 400 (h) Zn
(d) ∆T /∆t = = 250°C/min
2−0

(i) ts = 9.7 = B[8/24]2 or B = 87.3 min/in.2

9-63 Figure 9–28 shows the cooling curves obtained from several locations within a
cylindrical aluminum casting. Determine the local solidification times and the SDAS at
each location, then plot the tensile strength versus distance from the casting surface.
Would you recommend that the casting be designed so that a large or small amount of
material must be machined from the surface during finishing? Explain.

Solution: The local solidification times can be found from the cooling curves and
can be used to find the expected SDAS values from Figure 9–6. The
SDAS values can then be used to find the tensile strength, using Figure

215
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 9–7.
Surface: LST = 10 s ⇒ SDAS = 1.5 × 10–3 cm ⇒ TS = 47 ksi
Mid-radius: LST = 100 s ⇒ SDAS = 5 × 10–3 cm ⇒ TS = 44 ksi
Center: LST = 500 s ⇒ SDAS = 10 × 10–3 cm ⇒ TS = 39.5 ksi
If high strength is desirable, you prefer to machine as little material off
the surface of the casting as possible; the surface material has the
finest structure and highest strength; any excessive machining simply
removes the “best” material.

9-70 Calculate the volume, diameter, and height of the cylindrical riser required to prevent
shrinkage in a 1 in. × 6 in. × 6 in. casting if the H/D of the riser is 1.0.

Solution: The riser should take longer to solidify than the casting:
tr > tc,
or
V  V 
B  > B  ,
 A r  A c
where tr is the solidification time for the riser, tc is the solidification
time for the casting, and ( VA )r and ( VA )c are the volume (V) to surface
area (A) ratios of the riser (r) and casting (c), respectively. The mold
constant B is the same for the riser and casting. ( VA )c is given by
V  1× 6 × 6 36
  = = ,
 A c 2 ×1× 6 + 2 ×1× 6 + 2 × 6 × 6 − π D 2 96 − π D 2
4 4
V 
and   is given by
 A r
π
D2 H
 
V 4
  =π ,
 A r D + π DH
2

4
where D is the diameter of the riser and H is its height. Note that the
riser area in contact with the casting is not included in either the riser
or casting surface area; no heat is lost across this interface. Since H /
D is 1.0,

216
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 V  D
  = .
 A r 5
Thus,
 
D  36 
> ,
5  96 − π D 2 
 4 
and solving for D,
D > 1.93 in.
For this minimum diameter, the minimum height is 1.93 in., and the
minimum volume is
π
Vr = (1.93 in.) 2 × (1.93 in.) = 5.7 in.3
4

9-71 Calculate the volume, diameter, and height of the cylindrical riser required to prevent
shrinkage in a 4 in. × 10 in. × 20 in. casting if the H/D of the riser is 1.5.
Solution: The riser should take longer to solidify than the casting:
tr > tc,
or
V  V 
B  > B  ,
 A r  A c
where tr is the solidification time for the riser, tc is the solidification time
for the casting, and ( VA )r and ( VA )c are the volume (V) to surface area
(A) ratios of the riser (r) and casting (c), respectively. The mold constant
B is the same for the riser and casting. ( VA )c is given by
V  4 ×10 × 20 800
  = =
 A c 2 × 4 ×10 + 2 × 4 × 20 + 2 ×10 × 20 − π D 2 640 − π D 2
4 4
V 
and   is given by
 A r
π
D2 H
V  4
 = π ,
 
A D + π DH
2

4
where D is the diameter of the riser and H is its height. Note that the
riser area in contact with the casting is not included in either the riser or
casting surface area; no heat is lost across this interface. Since H / D is
1.5,
 V  3D
  = .
 A r 14
Thus,

217
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 3D 800
> .
14 640 − π D 2
4
Therefore, solving for D,
D > 6.11 in.
For this minimum diameter, the minimum height is 9.17 in., and the
minimum volume is
π
Vr = (6.11 in.) 2 × (9.17 in.) = 269 in.3
4

9-72 Figure 9–29 shows a cylindrical riser attached to a casting. Compare the solidification
times for each casting section and the riser and determine whether the riser will be
effective.

Solution: (8)(6)(3)
(V /A) thin = = 0.889
(3)(6) + 2(3)(8) + 2(6)(8)
(6)(6)(6)
(V /A) thick = = 1.13
(6)(3) + 5(6)(6) − (π /4)(3) 2
(π /4)(3)2 (7)
(V /A) riser = = 0.68
π (3)(7) + (π /4)(3)2
Note that the riser area in contact with the casting is not included in
either the riser or casting surface area; no heat is lost across this
interface. In a like manner, the area of contact between the thick and
thin portions of the casting are not included in the calculation of the
casting area.
The riser will not be effective; the thick section of the casting has the
largest V/A ratio and therefore requires the longest solidification
time. Consequently the riser will be completely solid before the thick
section is solidified; no liquid metal will be available to compensate
for the solidification shrinkage.

218
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
9-73 Figure 9–30 shows a cylindrical riser attached to a casting. Compare the solidification
times for each casting section and the riser and determine whether the riser will be
effective.

Solution: (4)(4)(4)
(V /A) thick = = 0.73
5(4)(4) + 1(2)(4)
(2)(2)(4)
(V /A) thin = = 0.50
3(2)(4) + 2(2)(2)
(π /4)(42 )(8)
(V /A) riser = = 0.8
π (4)(8) + 2(π /4)4 2
The area between the thick and thin sections of the casting are not
included in calculating casting area; no heat is lost across this
interface.
The riser will not be effective; the thin section has the smallest V/A
ratio and therefore freezes first. Even though the riser has the longest
solidification time, the thin section isolates the thick section from the
riser, preventing liquid metal from feeding from the riser to the thick
section. Shrinkage will occur in the thick section.

9-74 A hollow cylindrical mold for casting aluminum ingots has a 300 mm inside diameter and
is 2 m high. If the mold is filled with liquid aluminum at 935 K, what is the largest
spherical cavity that may form in the ingot?
Solution:
Since aluminum has 7% volume shrinkage, if the top of the ingot freezes first before the
entire ingot solidifies, there will be a cavity within the ingot. The largest cavity forms
when the entire shrinkage volume is converted to one cavity, which we shall assume to
be spherical.
L L
Volume of initial liquid = M D2H = M (0.3)2(2) = 0.1414 m3
Shrinkage volume = 0.1414 (0.07) = 0.009896 m3
Assuming a spherical cavity:
M
0.009896 m3 =  πr3 ; we find that r = 0.1332 m or 13.32 cm, so the diameter of the
cavity would be 26.64 cm.
If we assume that the solidification process proceeds from the bottom up with no
cavities formed within the ingot, the top of the ingot will be lower than the inside height
of the ingot mold. Assuming the outside diameter of the ingot to be the same as the

219
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 inside diameter of the ingot mold, the height of the ingot produced would be 93% (100
– 7% from shrinkage) of the mold height.
Ingot height = 0.93 (2 m mold height) = 1.86 m Diameter of the ingot = 0.3 m

9-75 A 4-in.-diameter sphere of liquid copper is allowed to solidify, producing a spherical

shrinkage cavity at the center of the casting. Compare the volume and diameter of the
shrinkage cavity in the copper casting to that obtained when a 4-in. sphere of liquid iron
is allowed to solidify.
Solution: Cu: 5.1% Fe: 3.4% rsphere = 4/2 = 2 in.
Cu: Vshrinkage = (4π/3)(2)3 (0.051) = 1.709 in.3
(4π/3)r3 = 1.709 in.3 or r = 0.742 in. dcavity = 1.48 in.
Fe: Vshrinkage = (4π/3)(2)3 (0.034) = 1.139 in.3
(4π/3)r3 = 1.139 in.3 or r = 0.648 in.
dcavity = 1.30 in.

9-76 A 4-in. cube of a liquid metal is allowed to solidify. A spherical shrinkage cavity with a
diameter of 1.49 in. is observed in the solid casting. Determine the percent volume
change that must have occurred during solidification.
Solution: Vliquid = (4 in.)3 = 64 in.3
Vshrinkage = (4π/3)(1.49/2)3 = 1.732 in.3
Vsolid = 64 – 1.732 = 62.268 in.3
64 − 62.268
% volume change = × 100 = 2.7%
64

9-77 A 2 cm × 4 cm × 6 cm magnesium casting is produced. After cooling to room

temperature, the casting is found to weigh 80 g. Determine (a) the volume of the
shrinkage cavity at the center of the casting and (b) the percent shrinkage that must
have occurred during solidification.
Solution: The density of magnesium is 1.738 g/cm3.
(a) Vinitial = (2)(4)(6) = 48 cm3
Vfinal = 80 g/(1.738 g/cm3) = 46.03 cm3
48 − 46.03
(b) % shrinkage = × 100% = 4.1%
48

9-78 A 2 in. × 8 in. × 10 in. iron casting is produced and, after cooling to room temperature, is
found to weigh 43.9 lb. Determine (a) the percent shrinkage that must have occurred
during solidification and (b) the number of shrinkage pores in the casting if all of the
shrinkage occurs as pores with a diameter of 0.05 in.
Solution: The density of the iron is 7.87 g/cm3
(43.9 lb)(454 g/lb)
Vactual = 3
= 2532.5 cm3
7.87 g/cm
Vintended = (2)(8)(10) = 160 in.3 × (2.54 cm/in)3 = 2621.9 cm3
2621.9 − 2532.5
shrinkage = × 100% = 3.4%
2621.9
Vpores = 2621.9 – 2532.5 = 89.4 cm3

220
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 rpores = (0.05 in./2)(2.54 cm/in.) = 0.0635 cm
89.4 cm3
# pores = = 83,354 pores
(4π /3)(0.0635 cm)3

9-79 If you cool an open vessel of liquid gallium until a thin solid “cake” forms, will the cake
form on the top or the bottom? Assume the air above the liquid gallium is stagnant and
does not cool the surface at all.
Solution:
The cake will float on top. Table 9-2 shows that gallium expands when it solidifies. This
means that a given volume of solid gallium is lighter than the same volume of liquid
gallium. In other words, it is less dense.

9-80 Give examples of materials that expand upon solidification.

Solution:
From Table 9-2, gallium, gray cast iron and water.

9-81 What is Sievert’s Law? How can gas porosity in molten alloys be removed or minimized?
Solution:
Sievert’s Law expresses the solubility of gases in liquid metals as a function of the partial
pressure of the gas over the liquid. In vacuum degassing of the melt, the partial pressure
of the dissolved gas over the melt is reduced and therefore the concentration of the gas
in the liquid is reduced.

9-83 Liquid magnesium is poured into a 2 cm × 2 cm × 24 cm mold and, as a result of

directional solidification, all of the solidification shrinkage occurs along the 24-cm length
of the casting. Determine the length of the casting immediately after solidification is
completed.

Solution: Vinitial = (2)(2)(24) = 96 cm3

% contraction = 4 or 0.04 × 96 = 3.84 cm3
Vfinal = 96 – 3.84 = 92.16 cm3 = (2)(2)(L)
Length (L) = 23.04 cm

9-84 A liquid cast iron has a density of 7.65 g/cm3. Immediately after solidification, the
density of the solid cast iron is found to be 7.71 g/cm3. Determine the percent volume
change that occurs during solidification. Does the cast iron expand or contract during
solidification?
Solution:
1/7.71−1/7.65 0.1297 cm 3 − 0.1307 cm 3
×100% = ×100% = −0.78%
1/7.65 0.1307 cm3
The casting contracts.

221
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
9-85 Molten copper at atmospheric pressure contains 0.01 wt% oxygen. The molten copper is
placed in a chamber that is pumped down to 1 Pa to remove gas from the melt prior to
pouring into the mold. Calculate the oxygen content of the copper melt after it is
subjected to this degassing treatment.
Solution: According to Sievert’s law, the amount of gas that can be dissolved in
a molten metal is given by
Percent of gas = K pgas ,
where K is a constant for a particular metal-gas system at constant
temperature and pgas is the partial pressure of the gas in contact with
the metal. Taking the ratio of the oxygen concentrations,
%O atm K patm patm
= = .
%O1 Pa K p1 Pa p1 Pa
Solving for the oxygen content at 1 Pa,
p1 Pa
%O1 Pa = %Oatm .
patm
Recalling that 1 atm = 1.013 × 105 Pa,
1
%O1 Pa = 0.01 = 3.14 ×10−5 wt% oxygen .
1.013 ×105

9-86 From Figure 9–14, find the solubility of hydrogen in liquid aluminum just before
solidification begins when the partial pressure of hydrogen is 1 atm. Determine the
solubility of hydrogen (in cm3/100 g Al) at the same temperature if the partial pressure
were reduced to 0.01 atm.

222
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Solution: 0.46 cm3 H2/100 g aluminum
1
0.46/x =
0.01
x = 0.46 0.01 = 0.046 cm 3 /100 g Al

9-87 The solubility of hydrogen in liquid aluminum at 715°C is found to be 1 cm3/(100 g Al). If
all of this hydrogen precipitated as gas bubbles during solidification and remained
trapped in the casting, calculate the volume percent gas in the solid aluminum.
Solution: (1 cm3 H2/100 g Al)(2.699 g/cm3) = 0.02699 cm3 H2/cm3 Al
= 2.699%

9-88 Review Example 9-7. The usual method of measuring a vacuum in the United States
vacuum pump industry is in inches of mercury. For example, atmospheric pressure is
29.92 inches of mercury. Covert the pressure found in the example to in. Hg.
Solution:
The pressure required is 10-6 atm. An atmosphere conversion factor is given, so:
10*1 atm 29.92 in. Hg
N= 5 = 0.00002992 in. Hg
1 1 atm
Note that the method usually used in industry is more confusing: A perfect vacuum is
29.92 in. Hg and atmospheric pressure is zero. This is the absolute value of the reading
of a vacuum gage.

9-98 Why has continuous casting of steels and other alloys assumed increased importance?
Solution:
The advantages of the continuous casting process are 1) more uniform composition and
less segregation across the section of the ingot, 2) produces higher yield (less cropping)
in primary shapes, 3) the shapes may be immediately deformed after casting with
minimal reheating of the slabs, blooms or billets to form other desired shapes and 4),
saves energy cost and increases productivity.

223
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
 224
© 2016 Cengage Learning. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.