See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/272310509

Influence of solidification rate on material properties of cast aluminium alloys

based on Al–Si–Cu and Al–Si–Mg

Article in International Journal of Cast Metals Research · April 2015

DOI: 10.1179/1743133614Y.0000000128

CITATIONS READS

8 3,883

2 authors, including:

Mitsuhiro Okayasu
Okayama University
176 PUBLICATIONS 1,452 CITATIONS

SEE PROFILE

All content following this page was uploaded by Mitsuhiro Okayasu on 19 August 2015.

The user has requested enhancement of the downloaded file.

Influence of solidification rate on material
properties of cast aluminium alloys based on
Al–Si–Cu and Al–Si–Mg
M. Okayasu* and S. Yoshida
Material properties of two different aluminium alloys (ADC12 and AC4C) were investigated using
cast samples produced by conventional gravity and pressure casting at various solidification
rates. The microstructural characteristics and defect density vary depending on the cast sample.
Those material characteristics affect directly the mechanical properties. The mechanical
properties, including tensile strength and fatigue strength, are improved with increasing
solidification rate, due to the fine grains, very small eutectic structures and low dislocation
density. Tensile strength and ductility for the AC4C alloy are overall higher than those for the
ADC12 alloy. The reason for this is attributed to mainly the lower volume fraction of brittle Si- and
Fe-based eutectic structures. The mechanical properties are more sensitive to the eutectic
structures than to the grain size, although both factors affect the properties significantly. Dimple
based ductile fracture is the main feature for the AC4C alloy, while cleavage based brittle failure is
the dominant feature for the ADC12 alloy. Severe lattice strain occurs around the cracks in the
AC4C alloy because of the high ductility, whereas weak lattice strain occurs in the brittle ADC12
alloy. On the basis of the microstructural and defect characteristics of the cast aluminium alloys,
their tensile properties could be approximated using a proposed formula.
Keywords: Aluminium alloy, Microstructure, Dislocation, Mechanical property, Fatigue property

Introduction direct chill casting, the microstructure is refined markedly

and cracking is eliminated.5 It is generally considered that
In recent years, many cast aluminium alloys have received an electromagnetic field leads to an increase in melt flow
special attention especially in automotive industries due to velocity in the injection process, thus giving good casta-
their low weight. The amount of cast aluminium alloys bility. Similar research was carried out by Dong et al.6 in
produced has increased significantly. However, there are which low frequency electromagnetic casting was devel-
problems with these cast Al alloys because some of their oped to allow grain refinement and to improve the surface
mechanical properties are poor, e.g. low strength and low quality.
ductility. In order to improve their mechanical properties, A good combination of high strength and high ductility
various casting technologies have been proposed. Ultra- at room temperature is achieved by a-Al grain refinement,
sonic melt treatment allows refining of the microstructure i.e. a small grain size of less than 1 mm.7 To obtain good
in cast Al alloys1–3 where acoustic waves with a frequency mechanical properties, a grain size at the sub-micrometre
higher than 17 kHz are induced into liquid metal. The level is a significant advantage, which can be explained
high ultrasonic frequency and high amplitude oscillations through the Hall–Petch relationship8
lead to cavitation of the melt and also promote intense :
mixing through agitation.1–3 The influence of ultrasonic sy ~s0 zky d {0 5
processing on microstructure and porosity was investi- where sy is the yield stress, s0 is the friction stress, ky is a
gated by Alba-Baena et al.,4 with the amount of porosity positive constant and d is the grain size.9 Effect of cooling
decreasing in A380 alloy after ultrasonic melt treatment rate on solidified microstructure (dendrite arm spacing)
due to a reduction of hydrogen in the melt. Furthermore, and mechanical properties of cast aluminium alloys has
grain refinement of the A380 alloy occurs to some extent. been investigated by several researchers.10,11 In general, it
With an air blade and simultaneously applying an is reported that as the dendrite arm spacing decreases, the
electromagnetic field for cast aluminium alloys during microhardness and strength correspondingly increase.10
To obtain the different grain size, various moulds made of
different materials were employed, such as copper, cast
Department of Materials Science and Engineering, Ehime University, 3 iron, graphite and sand, in which the cast Al–Si–Mg alloys
Bunkyo-cho, Matsuyama, Ehime, 790-8577, Japan with finer grains, made by the graphite moulds, have the
*Corresponding author, email mitsuhiro.okayasu@utoronto.ca best mechanical properties.11

ß 2015 W. S. Maney & Son Ltd.

Received 17 January 2014; accepted 14 July 2014
DOI 10.1179/1743133614Y.0000000128 International Journal of Cast Metals Research 2015 VOL 28 NO 2 105
Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

It is considered from the above literature surveys that the metal mould (cavity) for RC was 50612065 mm,
the mechanical properties of aluminium alloys are directly and this mould was made of a hot work tool steel SKD61.
affected by the grain size, although other material charac- During the casting process, the mould temperature was
teristics, e.g. size and shape of eutectic phases (Si- and Fe- set at about 293 K for ACM and at about 423 K for RC
based phases), are also significantly important in the respectively. Because of different mould, the sample
interpretation of mechanical properties.12 Recently, Wang thickness is different: the thickness for FC and ACG is
reported that the tensile properties and fracture behaviour 50 mm, while those for ACM and RC are 40 and 5 mm
of cast aluminium alloys depend strongly on the secon- respectively. The cooling rates obtained are indicated in
dary dendrite arm spacing (SDAS) of the a-Al matrix Table 1. The cooling rates were approximated using
and on the size and shape of the eutectic Si particles and temperature profiles measured directly using a K-type
Fe rich intermetallic, in which the cooling rate during thermocouple. In this case, the cooling rate was defined
the solidification refines both the dendrites and eutectic by the slope of the temperature profiles between liquidus
particles, resulting in high ductility.13 Wang et al.14 further and solidus. On the other hand, the cooling rate for RC
studied the effect of size and shape of eutectic structure on was approximated by the following: cooling rate526
the fatigue strength. Large and elongated eutectic particles 1046SDAS22?67.15 The reason for this approximation for
in cast aluminium alloys result in shorter fatigue lives, RC is due to the technical difficulty of directly measuring
where Mg- and Fe-based eutectic structures reduce the the cooling rate. Table 1 gives the different cooling rates.
fatigue strength. Even though some researchers have in- The cooling rates for ACM are higher than those for
vestigated the effect of eutectic structures on mechanical ACG and FC because of the high thermal conductivity
properties, there is apparently a lack of related experi- of the metal mould compared to the graphite one.
mental data. Moreover, material characteristics of a-grains, Furthermore, the cooling rate for AC4C is lower than
eutectic particles and defect density have not been con- that for ADC12 even if the same solidification process
sidered simultaneously to interpret the mechanical pro- was carried out, e.g. furnace or air cooling. This is
perties. In the work presented here, cast Al alloys were because of the temperature difference of solid–liquid
produced at various cooling rates to obtain different micro- coexistence.
structural and defect characteristics in advance, and their
mechanical properties were investigated systematically. Microstructural analysis
Microstructural characteristics for all the samples were
Experimental procedures investigated at room temperature using various methods,
including X-ray diffraction (XRD), electron backscatter
Materials diffraction (EBSD), transmission electron microscopy
In this study, two cast aluminium alloys, ADC12 and (TEM) and energy dispersive X-ray spectroscopy (EDX).
AC4C, were selected, which have been employed for XRD analysis was carried out with Cu Ka incident radia-
various automotive parts, e.g. transmission cases, con- tion using a X’Pert Power (PANalytical), where a soft-
verter housings, engine blocks and wheels. Chemical ware X’Pert High Score was employed. For EBSD
compositions of ADC12 and AC4C alloys measured analysis, a JSM-7000F SEM (JEOL Ltd.) with HKL
were (in mass-%) Al–Si10?6–Cu2?3–Zn0?9–Fe0?8–Mg0?2–Mn0?3 Channel 5 software was used. In this analysis, the samples
and Al–Si6?8–Mg0?32–Fe0?25–Cu0?11 respectively. The cast were prepared for less than 5 mm thick, and their surfaces
samples were solidified at different cooling rates using air for the observation were polished to mirror flatness in a
and furnace cooling. Moreover, moulds made of different vibropolisher using colloidal silica. The prepared sample
materials with different shape were used to alter the was then set on a sample holder, which tilted to an angle
cooling rate: graphite with many very small pores and of 70u to the electron beam with an acceleration voltage
hardened steel. Therefore, four different cast samples of 15 kV and beam current 5 nA. TEM analysis was
were prepared: air cooling with the graphite mold (ACG); conducted with an acceleration voltage of 100 kV to
furnace cooling with the graphite mold (FC); air cooling investigate the dislocation density in the cast Al alloys. In
with the metal mould (ACM); rapid cooling with the this analysis, a JEM-100CXII microscope (JEOL Ltd.)
metal mould (RC). Note that the RC process is related to was used. The samples for the TEM analysis were pre-
the die-casting process with high pressure (P) and high pared by conventional methods, e.g., mechanical thinning
casting speed (V): P540 MPa and V540 m s21 at the followed by electrolytic polishing to less than 300 nm
gate. Graphite molds with DO560 mm (outer diameter) thickness at low temperature (about 243 K). EDX analy-
and DI550 mm (inner diameter)670 mm in depth were sis was conducted with an acceleration voltage of 15 kV
employed for the ACG and the FC process. In both cases, using a JSM-6510 scanning electron microscope (JEOL
the ingot putting into the graphite moulds was heated to Ltd.).
and at the temperature above melting points of the cast Al
alloys for 20 min before the cooling in air or furnace to Mechanical tests
room temperature. Metal mould for ACM was made of Mechanical properties (tensile and fatigue properties)
a cast iron, which is for creating a metal ingot with were investigated at room temperature using an electro-
dimension (cavity): 600690640 mm. The dimension of servo-hydraulic system with 50 kN capacity (Shimadzu

Table 1 Cooling rate for ADC12 and AC4C samples

Al alloys ADC12 AC4C

Cast samples FC ACG ACM RC FC ACG ACM

Cooling rate/K s 0.08 0.44 4.83 101.2 0.04 0.32 2.86

106 International Journal of Cast Metals Research 2015 VOL 28 NO 2

 Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

1 Optical micrographs of ADC12 and AC4C samples

International Journal of Cast Metals Research 2015 VOL 28 NO 2 107

Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

2 EDX and XRD analyses for ADC12 and AC4C samples

108 International Journal of Cast Metals Research 2015 VOL 28 NO 2

 Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

3 Secondary dendrite arm spacing (SDAS) for a ADC12 4 Size of Si and Fe based eutectic phases for a ADC12
and b AC4C samples and b AC4C samples

Corp.). The tensile strength and elongation were examined structures are similarly obtained in related papers.16,17,18
via tensile stress versus strain curves. The tensile stress and From this analysis, needle shaped largely grown eutectic
strain were measured by a commercial load cell and strain structures (Si- and Fe-based) are present in the FC
gauge (2 mm gauge length) respectively. The tensile stress samples, while fine structures are present in the RC
was applied with a loading speed of 1 mm min21 to sample. It should be pointed out that the hardness of the
failure. For the fatigue test, tensile–tensile cyclic loading Si- and Fe-based eutectic phases, examined using a dyna-
was conducted with a sinusoidal waveform at a frequency mic ultra-micro-hardness tester (DUH-211, Shimadzu),
of 30 Hz and a stress ratio of 0?1 up to 107 cycles. The are found to be about 4000 MPa, which is more than five
fatigue strength was evaluated via stress amplitude (sa) times higher than that for the a-Al matrix and CuAl2.12
versus cyclic number to fracture (Nf). The maximum The SDAS values of all samples were measured, and
tensile stresses (smax) for the cyclic loading were the results obtained are shown in Fig. 3. Note that the
determined on the basis of the tensile strength, e.g. smax number of repetition for this measurement (nm) is more
is less than sUTS. than 100. Although the SDAS values are widely scattered,
the mean SDAS increases with decreasing cooling rate for
Results and discussion both ADC12 and AC4C. The SDAS for RC-ADC12 is as
small as 7?2 mm, which is about 12 times smaller than that
Microstructural characteristics for FC-ADC12 (84?6 mm). Figure 4 displays the size of
Figure 1 shows optical micrographs of the four samples Si- and Fe-based eutectic structures. Like the SDAS
of the ADC12 alloy and three of the AC4C alloy. results, the size of the eutectic phases varies (nm.100),
Observations reveal the formation of a-Al grains and where the higher the cooling rate, the smaller the eutectic
several eutectic structures in all samples. However, the size phase. The mean size of the eutectic structures for RC-
of the microstructures is different due to the different ADC12 is about 3?8 mm, which is about 16 times smaller
cooling rates. From the X-ray diffraction and energy than that for FC-ADC12.
dispersive X-ray analysis shown in Fig. 2, the eutectic Figure 5 presents transmission electron microscopy
structures appear to consist of Si, CuAl2 and Al8Fe2Si for images of the ADC12 samples. It is clear that dislocations
ADC12 and Si and Al8FeMg3Si6 for AC4C. Their eutectic of cell boundaries are distributed with low density across

International Journal of Cast Metals Research 2015 VOL 28 NO 2 109

Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

5 TEM images of FC-, ACG-, ACM- and RC-ADC12 samples, showing dislocation

all the cast samples, but are of higher density in the RC HV50?1896W/dI2, where W is the load value for the
sample. The variation in dislocation density is due to the hardness test and dI is the average diagonal length of the
casting conditions: for example, the higher cooling rate indentation. It is seen that the hardness decreases with
and the higher casting pressure for the RC sample lead to increasing SDAS. Such high hardness is caused by the fine
the high dislocation density. Based upon several transmis- microstructures. Similar approaches have been carried out
sion electron micrographs, the dislocation density was by several investigators13,19 using heat treated aluminium
estimated by measuring the total length L (m) of the alloys, and the material strength increases with decreasing
dislocation line in a cube of side l (m) and then dividing this the SDAS. The hardness for RC- and ACM-ADC12 is
by the volume of the cube l3 (m3), to give the dislocation found to be about 1?0 GPa, which is about 20% higher
density as L/l3 (m22). The dislocation density values than that for ACG- and FC-ADC12. Although a similar
are indicated in Table 2. Low dislocation densities are trend is obtained for AC4C samples, overall the hardness
obtained for the FC, ACG and ACM samples (between for AC4C is lower than that for ADC12. The reason
0?1861014 and 0?4161014 m22) and a high dislocation behind this is not clear at the moment, but this could be
density of 2?6661014 m22 is obtained for the RC sample. because of the different volume fraction of the hard Si
Such different dislocation densities could make a difference element as described above, i.e., the larger the amount of
to their mechanical properties, which is discussed in the eutectic Si, the higher the hardness. It should be pointed
following section. out that the hardness data for ADC12 are slightly scattered
(e.g. standard deviation of 0?01 for FC-ADC12) compared
Mechanical properties
to AC4C (e.g. standard deviation of 0?002 for FC-AC4C),
Figure 6 shows the Vickers hardness (HV) versus SDAS which could be due to the randomly distributed large
for the Al alloy samples (nm.10). In this case, the amount of the hard eutectic structures in the ADC12 alloy.
hardness values were obtained by the following formulas: Figure 7 shows the representative tensile stress versus
tensile strain curves for the cast ADC12 and AC4C alloys.
Table 2 Dislocation density for FC-, ACG-, ACM-, and
RC-ADC12 samples
As can be seen, different tensile properties are evident,
with the stress versus strain relationship for the samples
Al alloys ADC12 with high cooling rate indicating better tensile properties
compared to those with low cooling rate: the higher the
Cast samples FC ACG ACM RC tensile strength, the higher the material ductility. Similar
result was also reported in the study by Li et al.,20 in which
Dislocation density (61014) m22 0.18 0.41 0.29 2.66
there is an almost linear correlation between the strength

110 International Journal of Cast Metals Research 2015 VOL 28 NO 2

 Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

7 Relationship between tensile stress and strain for a

ADC12 and b AC4C samples

reason for the low ductility of ADC12 is attributed to the

high volume fraction of the eutectic Si embedded in the
a-Al grains.
It is reported that the ultimate tensile strength of
aluminium alloys is directly attributed to not only the grain
6 Vickers hardness for a ADC12 and b AC4C samples
size, but also the size and aspect ratio of eutectic Si. Those
and elongation for the related aluminium alloys. The microstructural factors are directly related to their UTS.21
average values of tensile properties (tensile strength sUTS In addition, the effects of cast defect and SDAS on the
and fracture strain ef) for RC-ADC12 samples (nm53), tensile properties of cast aluminium alloy were examined.22
sUTS5246 MPa and ef51?7%, are higher than those for In our previous work,23 the tensile strength of ADC12
the other ADC12 samples. Similar results are obtained for alloy was estimated by a multiple regression analysis, using
the AC4C samples, where the tensile strength and fracture five independent factors: SDAS, defect density of micro-
strain for ACM-AC4C are 214 MPa and ef55?2% respec- porosity (DR), size of eutectic structures (SE), aspect ratio
tively. It should be noted that the strain levels for ACM- of eutectic structures (AE) and dislocation density (DD).
AC4C are overall higher than those for ACM-ADC12, To interpret clearly the tensile properties of our samples,
although their tensile strengths are almost the same. The the influences of these five factors on the mechanical
Table 3 Microstructural and defect properties for ADC12 and AC4C samples

Al alloys ADC12 AC4C

Cast samples FC ACG ACM RC FC ACG ACM

SDAS/mm 84.6 59.0 22.6 7.2 131.9 78.5 27.6

Size of eutectic structures (SE)/mm 61.9 36.9 11.9 3.8 35.8 18.4 9.2
Aspect ratio of eutectic phase (AE) 8.8 13.5 9.5 10.7 6.2 8.3 10.4
Defect density (DR)/% 2.03 1.69 1.33 0.04 1.82 0.92 0.07

Table 4 Predictions of ultimate tensile strength using equation (1)

Al alloys ADC12

Cast samples FC ACG ACM RC

Ultimate tensile strength (sUTS)/MPa 94.1 113.0 202.9 245.7

Prediction by equation (1)/MPa 235.4 18.5 175.0 230.8

International Journal of Cast Metals Research 2015 VOL 28 NO 2 111

Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

The calculated and experimentally obtained values of

sUTS are indicated in Table 4. A comparison of the
numerical and experimental sUTS values shows close
levels for RC- and ACM-ADC12, whereas different
values are found for ACG- and FC-ADC12. The reason
for the poor prediction for the latter samples could be a
result of the influence of the large grown microstructure
for FC and ACG. In fact, the above formula proposed in
previous work was determined through statistical analysis
using cast ADC12 alloys with finer microstructures, e.g.
SDAS of 3?7–33?3 mm and SE of 0?3–15?8 mm,23 these
values being much smaller than those for the ACG- and
FC-ADC12 samples of the present study (see Table 3).
Similar approach was carried out by Ceshini et al.,21
where the ultimate tensile strength of the aluminium alloy
is approximated by the microstructural characteristics
(grain size, area of Si particle and aspect ratio of Si particle)
and defect rate. Using their proposed equations, an
attempt was made to estimate the tensile strength of our
samples. It appears that our tensile strength is relatively
approximated accurately by the grain size (d) and area of
eutectic phase (ASi): sUTS520?167 dz371?7 and sUTS5
21?252 ASiz337?12.
Figure 8 shows the relationship between stress ampli-
tude (sa) and fatigue life (S–N curves) for ADC12 and
AC4C. The arrows indicate the test specimens which are
not completely fractured by 107 cycles, i.e. the endur-
ance limit (sen). As is seen, a high endurance limit is
obtained for the samples with high cooling rate: the
mean sen is 45 MPa for RC-ADC12 and 60 MPa for
ACM-AC4C, which are much higher than those for the
samples with low cooling rates: sen519?6 MPa (FC-
ADC12) and 27?0 MPa (FC-AC4C). The fatigue prop-
erty was quantified using a power-law dependence to
obtain the fatigue strength coefficient sf and the fatigue
exponent b, sa5sfNfb, where two independent factors
are employed: the number of cycles to fracture Nf and
the stress amplitude sa. In this approximation, increased
8 S–N curves for a ADC12 and b AC4C samples
fatigue life is expected for an increasing sf and decreasing
b. The sf and b values for all samples are indicated
properties were investigated. The material properties (five together with the sen values in Table 5. There is a high
factors) for our investigated ADC12 alloy are indicated in correlation rate between sf and sen of about R250?95,
Tables 2 and 3. Note first that the measurement data in while the coefficient of determination between b and sen is
the tables (SDAS, SE and AE) are mean values from more less than R250?2.
than a hundred measurements. Because the associated
eutectic phases are formed with needle shape, no clear Failure analysis
change of aspect ratio is obtained. On the other hand, the The fracture surfaces of all samples after being fractured
defect density seems to have clear trend with changing by monotonic and cyclic loading were analysed. Figure 9
the cooling rate or sample thickness, where the thinner the shows scanning electron microscopy images of the
thickness, the lower the defect rate. With these material fracture surfaces for FC-ADC12, ACM-ADC12 and
parameters, ultimate tensile strength for the ADC12 ACM-AC4C after tensile testing. From Fig. 9, cleavage-
samples was estimated using the following equation23 like brittle fracture is the dominant feature for both
sUTS ~330:2{(3:35|SDAS){(4:71|DR){ ADC12 samples, while dimple-based ductile fracture is
(1) observed for the AC4C sample. Such different fracture
(0:1|SE){(7:66|AE)z(2:88|DD) modes are related to the different ductility, as indicated in

Table 5 Fatigue strength coefficient and endurance limit for ADC12 and AC4C samples

Al alloys ADC12 AC4C

Cast samples FC ACG ACM RC FC ACG ACM

Fatigue strength coefficient (sf)/MPa 69.0 108.3 136.0 155.7 51.4 61.7 116.8
Fatigue exponent (b) 20.070 20.080 20.072 20.072 20.041 20.031 20.046
Endurance limit (sen)/MPa 19.6 26.1 38.4 44.3 27.0 37.3 54.0

112 International Journal of Cast Metals Research 2015 VOL 28 NO 2

 Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

9 Images (SEM) of fracture surfaces for ADC12 and AC4C samples after fracture by monotonic loading

International Journal of Cast Metals Research 2015 VOL 28 NO 2 113

Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

10 Images (SEM) of fracture surfaces for ADC12 and AC4C samples after fracture by cyclic loading

114 International Journal of Cast Metals Research 2015 VOL 28 NO 2

 Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

11 Inverse pole figure (IPF) maps and misorientation (MO) maps for ACM-ADC12, ACM-AC4C and FC-AC4C samples
around fatigue cracks created by monotonic loading

Fig. 7. The roughness of the fracture surface is also 1. The sizes of a-Al grains and eutectic phases depend
different among the samples, where a rougher surface is on the cooling rate. In addition, the defect density
observed in the FC sample than in the ACM ones. This (porosity and dislocation density) varies: the lower
could be attributed to the large grown eutectic phases cooling rate makes the lower dislocation density and the
between the a-Al grains in the FC sample. Cast defects higher defect density. Such different material character-
of shrinkage porosity can also be seen in the FC sam- istics affect directly the mechanical properties and failure
ple, which accelerate the failure. Figure 10 displays the characteristics.
fracture surfaces of the FC and ACM samples after 2. The mechanical properties, including tensile strength
fatigue testing. Note that these samples were fractured and fatigue strength, are improved with increasing cooling
after cyclic loading of more than 100 000 cycles. Like the rate, due to the fine grains and very small eutectic
tensile test samples, relatively rough fracture surfaces are structures. The good mechanical properties are also
seen in the FC-ADC12 sample, where defects and cleavage affected by the low defect density and the high dislocation
facet are clearly observed. On the other hand, a step density. The material ductility for AC4C is higher than
morphology and rugged fracture surfaces are observed in that for ADC12, which is because of the low volume
the AC4C sample, in which striation-like formation is fraction of brittle eutectic structures.
detected. 3. With five independent variables (secondary dendrite
To understand the failure characteristics in detail, the arm spacing, porosity density, size of eutectic structures,
strain characteristics were examined by electron back- aspect ratio of eutectic structures and dislocation density),
scatter diffraction analysis. Figure 11 shows the inverse- ultimate tensile strength for the ADC12 samples can be
pole figure (IPF) maps and misorientation (MO) maps estimated accurately, especially for ACM- and RC-
for ACM and FC samples around cracks, after being ADC12. On the other hand, the poor estimations for
fractured by monotonic loading. The colour levels of ACG- and FC-ADC12 are a result of the influence of the
the pixels in the IPF maps represent the deviation of largely grown microstructure in the FC and ACG samples.
measured face centred cubic crystal orientation. The red 4. Ductile fracture (dimple and striation) is a domi-
solid lines in the MO maps are related to MO angles of nant feature for AC4C, although brittle fracture (cleavage
more than 2u. The MO maps show the extent of the facet) is detected for ADC12. Internal strain is evaluated
internal strain (lattice strain), accumulated mainly by the by the misorientation angle of crystal orientation, where
stress during the tensile tests. From this analysis, overall severe internal strain (lattice strain) is obtained for the
lattice strain (density of reddish zone) for ACM-AC4C is AC4C samples, because of the high plastic deformation.
high compared to ACM-ADC12. Moreover, the internal
strain for ACM-AC4C is higher than that for FC-AC4C. Acknowledgements
Such a different strain level is influenced by the different
severity of plastic deformation, i.e. the higher the This work was supported by a grant (Grant-in-Aid for
deformation, the severer the lattice strain level. Scientific Research (C), 2014) from the Japanese Govern-
ment (Ministry of Education, Science, Sports and Culture).

Conclusions
References
The effects of microstructural and defect characteristics
on mechanical properties of two different cast alumi- 1. G. I. Eskin: ‘Ultrasonic treatment of light alloy melts’, 155; 1998,
Amsterdam, Gordon and Breach Science Publishers.
nium alloys (ADC12 and AC4C) have been investigated. 2. L. Zhang, D. G. Eskin, A. Miroux and L. Katgerman: ‘Light
On the basis of the results obtained, the following metals’, The Minerals, Metals and Materials Society, 2012, 2012,
conclusions can be drawn. 999–1004.

International Journal of Cast Metals Research 2015 VOL 28 NO 2 115

Okayasu and Yoshida Influence of solidification rate on cast aluminium alloys

3. L. Zhang, D. G. Eskin and L. Katgerman: J. Mater. Sci., 2011, 46, 14. Q. C. Wang, D. Apelian and D. A. Lados: J. Light Met., 2002, 1,
5252–5259. 85–97.
4. N. Alba-Baena, T. Pabel, N. Villa-Sierra and D. Eskin: Mater. Sci. 15. M. Okayasu, Y. Ohkura, S. Takeuchi, S. Takasu, H. Ohfuji and T.
Forum, 2013, 765, 271–275. Shiraishi: Mater. Sci. Eng. A, 2012, A543, 185–192.
5. J. Cui, H. Zhang and Y. Zuo: Mater. Sci. Forum, 2013, 765, 165–169. 16. H. Xin and Y. Hong: J. Wuhan Univ. Tech.-Mater. Sci. Ed., 2013,
6. J. Dong, J. Cui, F. Yu, C. Ban and Z. Zhao: Metall. Mater. Trans. 28, 202–205.
A, 2004, 35A, 2487–2494. 17. C.-L. Chen and R. C. Thomson: Intermetallics, 2010, 18, 1750–
7. K. Kubota, M. Mabuchi and K. Higashi: J. Mater. Sci., 1999, 34, 1757.
2255–2262. 18. M. Abduiwahab, I. A. Madugu, S. A. Yaro and A. P. I. Popoola:
8. Z. Horita, T. Fujinami, M. Nemoto and T. G. Langdon: Metall. Slicon, 2012, 4, 137–143.
Mater. Trans. A, 2000, 31A, 691–701. 19. G. Ran, J. Zhou and Q. C. Wang: J. Alloys Compos., 2006, 421, 80–
9. Y. S. Sato, M. Urata, H. Kokawa and K. Ikeda: Mater. Sci. Eng. 86.
A, 2003, A354, 298–305. 20. Z. Li, A. M. Samuel, F. H. Samuel, C. Ravindran, S. Valtierra and
10. L. Y. Zhang, Y. H. Jiang, Z. Ma, S. F. Shan, Y. Z. Jia, C. Z. Fan H. W. Doty: Mater. Sci. Eng. A, 2004, A367, 96–110.
and W. K. Wang: J. Mater. Process. Technol., 2008, 207, 107–111. 21. L. Ceshini, A. Morri, A. Morri, A. Gamberini and S. Messieri:
11. S. G. Shabestari and H. Moemeni: J. Mater. Process. Technol., J. Mater. Process. Technol., 2009, 30, 4525–4531.
2004, 153–154, 193–198. 22. L. Ceschini, I. Boromei, A. Morri, S. Seifeddine and I. Svensson:
12. M. Okayasu, S. Takasu and M. Mizuno: J. Mater. Sci., 2012, 47, J. Mater. Process. Technol., 2009, 209, 5669–5679.
241–250. 23. M. Okayasu, K. Ota, S. Takeuchi, H. Ohfuji and T. Shiraishi:
13. Q. C. Wang: Metall. Mater. Trans. A, 2003, 34A, 2887–2899. Mater. Sci. Eng. A, 2014, A592, 189–200.

116 International Journal of Cast Metals Research 2015 VOL 28 NO 2

View publication stats