Materials Transactions, Vol. 44, No. 4 (2003) pp.
468 to 475
Special Issue on Platform Science and Technology for Advanced Magnesium Alloys, II
#2003 The Japan Institute of Light Metals
Effect of Microstructural Factors on Tensile Properties of
an ECAE-Processed AZ31 Magnesium Alloy*1
Yu Yoshida*2 , Lawrence Cisar*2 , Shigeharu Kamado and Yo Kojima
Department of Mechanical Engineering, Nagaoka University of Technology, Nagaoka 940-2188, Japan
Mg–3%Al–1%Zn (AZ31) alloy was subjected to ECAE (Equal Channel Angular Extrusion) processing under various processing
conditions. Then tensile tests were carried out at room temperature to investigate the relationship between tensile properties and microstructural
parameters that include grain size and the texture generated by ECAE processing. In 4-pass ECAE specimens processed at 523 K, tensile
ductility is improved as a result of easy basal slip during tensile test along the extrusion direction, because such specimens have textures in which
the basal plane is inclined at 45 to the extrusion direction. On the other hand, in the specimens processed at 573 K, 0.2% proof stress is higher
than those of specimens processed at lower temperatures, but elongation is smaller. This is because of difficult basal slip caused by the textures in
which the basal plane is oriented parallel to the extrusion direction. However, 8-pass specimens processed at 473 K and subsequently annealed,
which have similar textures but different grain sizes (d), exhibit clear grain size dependencies of 0.2% proof stress (0:2 ) according to Hall-Petch
relationship; 0:2 ¼ 30 þ 0:17d 1=2 . Therefore, crystallographic orientation has a profound effect on the tensile properties of AZ31 alloy, and
grain size has a little effect.
(Received October 23, 2002; Accepted February 7, 2003)
Keywords: equal channel angular extrusion process, magnesium alloy, tensile properties, texture, grain size dependence
1. Introduction is nearly 0. Therefore, their workability will also be small
when secondary working is carried out at room temperature
Although magnesium alloys are expected to be useful using the rolled or extruded materials.
materials for minimizing global environmental problems, In a previous study,18) ECAE processing19,20) was applied
wider applications of the alloys have not been achieved due to Mg–Al–Zn alloys as a method not only for grain
to their inferior cold workability. Generally, it has been said refinement but also for texture control. As a result, a peculiar
that the less cold workability of magnesium results from its texture in which the basal plane is inclined at 45 to the
limited slip system at room temperature. Recently, however, extrusion direction was formed, and therefore, the ductility
there are some instances1–4) where the ductility is improved was improved significantly. However, the relationship
in grain-refined magnesium alloys even at room temperature. between ECAE processing conditions and the formed
This suggests that the poor ductility of magnesium relates to microstructures including texture, as well as the effect of
microstructural conditions, and not necessarily to intrinsic the microstructure on tensile properties of the ECAE-
nature of the material. processed specimen have not been clarified yet.
There is a large number of reports4–12) about the correla- In the present study, extruded AZ31 magnesium alloy was
tion between grain size and mechanical properties of subjected to ECAE processing. Then the effect of ECAE
magnesium alloys based on Hall-Petch relation. In magne- processing conditions, such as the process temperature and
sium, the critical resolved shear stress (CRSS) for non-basal repetitive number of ECAE process, on grain size and texture
slip is about one hundred times as large as that for basal formation was examined. Subsequently, the ECAE-proces-
slip.13) This difference of the CRSS gives rise to anisotropy sed specimens were subjected to tensile tests. Then the
of mechanical properties resulting from the slip system. correlation between tensile properties and microstructural
Consequently, when the mechanical properties of wrought parameters, such as grain size and texture are investigated.
magnesium alloys that have strong crystallographic texture in Furthermore, the effect of microstuructural changes with
their structures are investigated, considering the crystal- annealing on tensile properties of ECAE-processed speci-
lographic orientation in addition to the grain size is mens is also investigated.
important.6,7)
In cold rolled magnesium alloys, the basal plane is oriented 2. Experimental Procedure
parallel to the rolling direction, that is, the so-called basal
texture is formed.14,15) The basal texture is also generated in Commercial AZ31 extruded rods with 19 mm diameter
hot rolled magnesium alloys.7,15,16) Furthermore, in extruded were obtained for the investigations carried out in the present
magnesium alloys, the texture in which the basal plane is study. The chemical composition of the alloy is shown in
oriented parallel to the extrusion direction is formed.6,12,16,17) Table 1. Figure 1 shows a schematic illustration of the ECAE
If these materials were deformed along the working direction,
the ductility is expected to be small because shear force does
not operate on the basal plane due to small Schmid factor that Table 1 Chemical composition of AZ31 alloy used for present study
(mass%).
*1This Paper was Originally Published in the Journal of Japan Institute of Alloy Al Zn Mn Fe Si Cu Ni Mg
Light Metals 52 (2002) 559–565.
*2Graduate Student, Nagaoka University of Technology. AZ31 3.26 0.92 0.43 0.004 0.001 0.000 0.001 bal.
Tensile Properties of ECAE-Processed AZ31 Magnesium Alloy 469
central axes remaining the same as before ECAE processing.
After tensile test, the microstructures near the fractured
surfaces were observed using optical microscope.
In order to analyze the textures generated by plastic
working, (0002) pole figures were drawn by Schulz reflection
method22) using X-ray diffractometer. The measurements
were performed using CuK (wave length ¼ 0:15406 nm)
radiation at 40 kV and 200 mA with the sample tilt angle
ranging from 0–70 .
3. Results and Discussion
3.1 Microstructures and tensile properties of as-ECAE
specimens
3.1.1 Microstructures of as-ECAE specimens
Figure 2 shows microstructures of as-received specimen, 2
and 4-pass ECAE processed specimens (hereafter referred to
as 2-pass and 4-pass specimens, respectively). The micro-
Fig. 1 Schematic illustration of ECAE process. structures of all ECAE specimens consist of many unde-
formed equiaxed grains. This suggests that dynamic
process used for the present study. During the ECAE recrystallization occurs during ECAE processing. Average
processing, the specimen is inserted into the upper part of grain sizes increase with increasing processing temperature
the die and pressed. Then the specimen is subjected to shear in both 2 and 4-pass specimens. However, the average grain
force at the intersecting point of two equal channels. The sizes decrease with increasing number of processing at each
ECAE die used for the present study has two equal channels temperature. In the specimen processed at 473 K, although
of 15 mm diameter. The intersecting angle between the two very fine grains of about 1 mm are observed, some coarse
channels is 90 and the angle of the outer arc at the
intersection is 60 . Using this ECAE die, an equivalent strain
of 94% can be introduced.21)
The extruded bars were machined to cylindrical specimens
with a diameter of 15 mm and a length of 80 mm for ECAE
processing. Table 2 shows the ECAE processing conditions.
The ECAE processing was carried out at 473, 523 and 573 K,
respectively, with constant extrusion rate of 20 mm/min. The
processing was repeated 2, 4 and 8 times with the specimens
rotated at the angles shown in Table 2 around the specimens
longitudinal axis. Furthermore, 8-pass ECAE specimens
processed at 473 K were subjected to isochronal annealing at
448–673 K for 1 h and at 673–773 K for 4 h in order to
investigate the correlation between grain size and tensile
properties.
Optical microscope observations of all specimens were
performed, and the grain size was simultaneously evaluated
by image processing analysis. Furthermore, in the annealed
specimens, Vickers hardness test was carried out and the
change in hardness with annealing was investigated. Tensile
tests were carried out at room temperature under a strain rate
of 8:33 104 s1 . The ECAE-processed samples were
machined to JIS14A tensile specimens having a gage
dimension of 4 mm diameter and 20 mm length, with their
Table 2 ECAE processing conditions.
Number of processing 2, 4 8
Specimen rotation (deg) 2nd: 180 3rd: 90 4th: 180 90
Working temperature (K) 473 523 573 473
Extrusion rate (mm/min) 20
448–673 K-1 h
Annealing condition — Fig. 2 Microstructures of as-received and ECAE specimens processed at
673–773 K-4 h
(a) 473 K, (b) 523 K and (c) 573 K. Grain sizes are indicated in the figure.
470 Y. Yoshida, L. Cisar, S. Kamado and Y. Kojima
unrecrystallized grains still remain. This is because the low conditions. Both the tensile strength and the 0.2% proof stress
temperature processing makes both dynamic recrystallization of the 2-pass specimens increase with increasing processing
and subsequent grain growth difficult. Furthermore, the temperature. The stress-strain curve of the 573 K-2-pass
homogeneity of the grain sizes is improved by repeating the specimen is similar to that of the as-received specimen.
processing. On the other hand, in the specimens processed at However, the 523 K-2-pass specimen exhibits an elongation
523 K and 573 K, the variation in grain sizes becomes small. of 25%, which is higher than that of the as-received specimen
This is as a result of increase in the volume fraction of (19%), while the elongation of the 473 K-2-pass specimen is
recrystallized grains generated by dynamic recrystallization merely 8%. This value of 8% is much lower than that of the
during ECAE processing and subsequent grain growth due to as-received specimen. On the other hand, the elongation of
relatively high temperature processing. the 4-pass specimens processed at 473 K and 523 K are
From the above results, it can be concluded that the improved significantly compared to that of the 2-pass
processing temperature and the number of repetitive proces- specimens processed at those temperatures. Comparing
sing have profound effect on the recrystallized grain size and 473 K-4-pass specimen with 523 K-4-pass specimen, the
homogeneity of the microstructure, respectively. Further- tensile strength and 0.2% proof stress are almost same.
more, in high temperature processing, although the grain However, the elongation of the latter specimen is higher than
sizes are large, homogeneous microstructure can be obtained that of the former specimen. Furthermore, the tensile proper-
with a small number of processing. For the achievement of ties of 573 K-4-pass specimen are almost the same as those of
fine and homogeneous microstructure, a large number of the 573 K-2-pass specimen. However, the 0.2% proof stress
processing at a relatively low temperature will be required. of the former is slightly lower than that of the latter.
3.1.2 Tensile properties As described in section 3.1.1, the grain size decreases with
Figure 3 shows stress-strain curves of the as-received decreasing ECAE processing temperature. Generally, in
specimen and as-ECAE specimens processed under various discussing the strength of a material, Hall-Petch relation is
often used. However, such grain size dependency of the 0.2%
proof stress could not be found from the present results.
Therefore, it is inferred that the tensile properties, especially
0.2% proof stress, must have been influenced by the
crystallographic texture due to strong anisotropy of the slip
systems at room temperature. The relationship between
tensile properties and textures will be discussed in details in
section 3.1.4.
3.1.3 Microstructures of tensile fractured specimens
Figure 4 shows microstructures near the fractured surfaces
of the tensile tested specimens. Some twins are observed in
the coarse grains in the specimens processed at 473 K. The
existence of twins indicates that the deformability of the
coarse grains is lower than that of the fine grains. Conse-
quently, the coarse grains may be responsible for the low
elongation of these specimens. In fact, the elongation is
enhanced in the 473 K-4-pass specimen that has less coarse
grains than the 473 K-2-pass specimen. On the other hand,
the microstructures of the specimens processed at 523 K are
homogeneous without coarse grains. It is noted that the grains
of 523 K-4-pass specimen that exhibits high elongation of
35% are elongated towards the tensile direction. This
elongation of the grains suggests that dislocation slip
operates actively in the grains and that contributes to the
ductility of the specimen. Furthermore, some twins are
observed in the 523 K-4-pass specimen, but the twins are
uniformly distributed in the microstructure. This result
indicates that the deformation is uniform in each grain.
Thus, the uniform deformation of the grains may cause
improvement of the ductility. A large number of twins are
observed in the specimens processed at 573 K. In these
specimens, it is considered that twinning also contributes to
the whole deformation in addition to dislocation slip. This
deformation mechanism is different from that of the speci-
mens processed at 523 K because the grains of the specimens
processed at 573 K remain equiaxed.
Fig. 3 Stress-strain curves of 2-pass (top) and 4-pass (bottom) ECAE
specimens processed at various temperatures.
Tensile Properties of ECAE-Processed AZ31 Magnesium Alloy 471
Fig. 4 Microstructures near the fractured surfaces of ECAE specimens
processed at (a) 473 K, (b) 523 K and (c) 573 K.
3.1.4 Texture
Figure 5 shows (0002) pole figures of as-received and as-
ECAE specimens. The measurements were carried out on the
planes parallel to the extrusion direction.
The basal plane of the as-received specimen is oriented
parallel to the extrusion direction. This is the typical texture
observed in extruded rods of magnesium alloys.6,12)
In the 473 K-2-pass specimen, the texture consists of high
intensity basal planes inclined at 45 to the extrusion
direction and lower intensity ones parallel to the extrusion
direction. The intensity of the former decreases with Fig. 5 (0002) pole figures of as-received and ECAE specimens processed
at (a) 473 K, (b) 523 K and (c) 573 K.
increasing processing temperature, while that of the latter
increases. In the 573 K-2-pass specimen, the basal plane is
only oriented parallel to the extrusion direction. The 473 K the basal plane is inclined at 45 to the extrusion direction
and 523 K-4-pass specimens have similar textures. In these along the shear plane during the ECAE processing. On the
specimens, the intensity of the basal planes parallel to the other hand, at a relatively high temperature of about 573 K,
extrusion direction decreases compared to 2-pass specimens the basal plane is oriented parallel to the extrusion direction.
processed at same temperature. However, the intensity of the Furthermore, in both texture (a) and (b), two types of
inclined basal planes increases. Furthermore, the intensity of orientations that are misaligned at 30 around the c-axis can
the basal plane parallel to the extrusion direction of the be expected according to a previous study.18) If the specimen
573 K-4-pass specimen is higher than that of the 573 K-2- having texture (a) is tensioned along the extrusion direction,
pass specimen. Thus, repetitive ECAE processing makes the easy basal slip would occur because maximum shear force
obtained texture strongly exhibit preferred orientation ac- occurs on the plane inclined at 45 to the tensile direction. As
cording to the processing temperatures. a result, both the tensile strength and 0.2% proof stress would
The textures generated by ECAE processing can be be remarkably reduced. It is expected that the easy basal slip
classified into two types according to the orientations of the will also be related to the enhancement of the elongation. In
basal plane. These orientations are represented schematically fact, the elongation reaches an extraordinarily high value of
in Fig. 6. At a moderately low temperature of about 473 K, 35% in the 523 K-4-pass specimen that has a homogeneous
472 Y. Yoshida, L. Cisar, S. Kamado and Y. Kojima
Fig. 6 Schematic illustration of crystallographic orientations formed by ECAE processing.
microstructure. However, it is difficult for the grains to be
significantly elongated, as shown in Fig. 4, by only basal slip
because the independent slip modes of basal slip are only
two.23) Therefore, in the specimens that exhibit high ductility
like the 523 K-4-pass specimen, cross-slip from basal plane
to non-basal plane as established by Koike et al.24,25) might
be in operation in addition to basal slip. On the other hand, in
the specimens having texture (b) in which the basal plane is
oriented parallel to the extrusion direction, the Schmid
factors for basal slip is nearly 0 when the specimens are
tensioned along the extrusion direction. Therefore, almost no
shear force operates on the basal plane. As a result, it would
be difficult for basal slip to occur. Consequently, the tensile
strength and 0.2% proof stress would increase simultaneously
with decrease of the elongation compared to the specimens
having texture (a). This tendency is similar to the as-received
specimen. Furthermore, in the 523 K-2-pass specimen that
has mixed texture of (a) and (b), the 0.2% proof stress
exhibits intermediate value between that of the specimens
having textures (a) and (b).
Thus, the tensile strength and 0.2% proof stress of all
specimens, as well as the elongation of the specimens having
homogeneous microstructures, correspond well to their
crystallographic orientations. Fig. 7 Microstructures of (a) 8-pass ECAE-processed specimen and
specimens subsequently annealed at (b) 523 K-1 h, (c) 723 K-4 h and
3.2 Changes in microstructures and tensile properties (d) 773 K-4 h. Grain sizes are indicated in the figure.
with annealing
The 8-pass as-ECAE specimens were annealed in order to
investigate the changes in microstructures and mechanical
properties with annealing.
3.2.1 Changes in hardness and microstructures
Figure 7 shows optical micrographs of the 8-pass as-
ECAE and the annealed specimens, while the changes in
Vickers hardness with annealing are shown in Fig. 8. Fine
and equiaxed grains are homogeneously generated in the
microstructure of the 8-pass as-ECAE specimen. This is as a
result of the multiple pass ECAE processing at 473 K. The
hardness decreases suddenly after annealing at 448 K for 1 h.
The specimen annealed at 523 K for 1 h has fully softened,
and the grains have coarsened up to average grain size of
4.3 mm with homogeneous microstructure as shown in Fig. 7.
At this point, it is inferred that recrystallization driven by
residual strain is complete. Therefore, the subsequent
decrease of the hardness becomes gentle. Isochronal anneal-
ing for 4 h was carried out at higher temperatures in order to Fig. 8 Change in hardness with isochronal annealing.
Tensile Properties of ECAE-Processed AZ31 Magnesium Alloy 473
Fig. 9 Stress-strain curves of 8-pass as-ECAE and annealed specimens.
obtain further coarse grain size and microstructures. The
microstructures and the changes in hardness are also shown
in Figs. 7 and 8, respectively. The grain size of 723 K-4 h
annealed specimen is coarser than that of the 523 K-1 h
annealed specimen but the microstructure remains homo-
geneous. This result suggests that normal grain growth occurs
up to relatively high temperatures in the investigated alloy.
However, the grains of the 773 K-4 h annealed specimen
coarsen up to average grain size of 25.5 mm with a large Fig. 10 (0002) pole figures of (a) 8-pass ECAE-processed specimen and
reduction of hardness as a result of abnormal grain growth. specimens subsequently annealed at (b) 523 K-1 h, (c) 723 K-4 h and
3.2.2 Tensile properties (d) 773 K-4 h.
Figure 9 shows the stress-strain curves of 8-pass as-ECAE
and annealed specimens. Both the tensile strength and 0.2%
proof stress of all specimens decrease with increasing where 0 is yield strength of single crystalline material, k is
annealing temperature. That is, the strength of the specimens material constant, and d is grain size. Armstrong5) proposed a
exhibits grain size dependencies. The specimens exhibit ‘‘pile-up’’ model in which 0 and k of hcp materials relate to
almost the same elongation, except the 773 K-4 h annealed both CRSS for the easiest slip system and the CRSS for the
specimen, which is lower. According to Koike et al.,24) the slip system required to maintain the continuity of crystals at
cross-slip from basal to non-basal plane, as mentioned the grain boundaries when dislocations are accumulated at
earlier, can be operative within the grain size range up to the grain boundaries. In addition, Rao et al.7) supposed that
10 mm. Therefore, in the 773 K-4 h annealed specimen having 0 and k relate to CRSS for basal a slip and for prismatic a
extraordinary coarse grains due to abnormal grain growth, the slip, respectively, and concluded that their orientation factors
elongation would decrease. influence 0 and k. Thus, in magnesium alloys, the crystal-
3.2.3 Texture lographic orientation is extremely important during investi-
Figure 10 shows the (0002) pole figures of the 8-pass as- gation of the grain size dependency of 0.2% proof stress.
ECAE and annealed specimens. Annealing reduces the However, the grain size dependency dose not exist in the
intensities of preferred orientation. This tendency is similar specimens having different textures as discussed in section
to that of annealed AZ31 hot-rolled samples.15) In the pole 3.1.
figure of 773 K-4 h annealed specimen, some high peaks, Figure 11 shows the correlation between the grain size and
which may correspond to extraordinarily coarse grains, are the 0.2% proof stress of both the 8-pass as-ECAE and
observed inclined at about 10 to the extrusion direction. annealed specimens. In the annealed specimens having
However, significant textural changes are not observed similar textures, the clear grain size dependency of 0.2%
between the annealed specimens. proof stress is exhibited according to Hall-Petch relation.
3.2.4 Grain size dependency Then 0 is 30 MPa and k is 0.17 MPa/m1=2 . Koike et al.26)
As expressed in section 3.1.2, the yield stress (y ) of have shown that 0 is approximately 60 MPa when the grain
polycrystalline metals usually depends on the grain size, size is the only effective factor. That is, when the effects of
according to the Hall-Petch relation as described below, precipitates and/or plastic strain are negligible. However, the
y ¼ 0 þ kd 1=2 ð1Þ 0 value obtained in the present study is half of the value
proposed by Koike et al. It is expected that the basal a slip of
474 Y. Yoshida, L. Cisar, S. Kamado and Y. Kojima
the 0.2% proof stress is small because the microstructures of
the specimens are fully recrystallized. Consequently, it can
be concluded the 0.2% proof stress of the annealed specimens
depend essentially on grain size.
4. Conclusions
The correlation between ECAE processing conditions,
microstructures including texture and tensile properties have
been investigated in order to reveal the effect of the
microstructural parameters on the strength and ductility of
AZ31 magnesium alloy. The results are summarized as
follows:
(1) The grain sizes of ECAE-processed specimens decrease
with decreasing processing temperature. However, the
Fig. 11 Grain size dependencies of 0.2% proof stress of 8-pass ECAE- microstructures are heterogeneous in the case of a small
processed and annealed specimens. number of repetitive processing. A large number of
repetitive processing at a low temperature is required to
obtain fine and homogeneous grains.
the annealed specimens operates easily because most of the
(2) The tensile strength and 0.2% proof stress of as-ECAE
basal planes are inclined to the tensile axis. Thus, the easy
processed specimens are largely influenced by crystal-
basal slip brings about the reduction of the 0 value. On the
lographic orientation rather than grain size. In 523 K-4-
other hand, the k value of 0.17 MPa/m1=2 corresponds to that
pass specimen, the 0.2% proof stress is reduced but
obtained by Koike et al.25) Wilson6) and Rao et al.7) reported
elongation is significantly enhanced compared to those
that the k value depends on crystallographic orientation and
of as-received sample. This is because the specimen has
decreases when magnesium alloys have strong preferred
a texture in which the basal plane is inclined at 45 to
orientation in their texture. Thus, further researches are
the tensile direction with a homogeneous microstruc-
required to clarify the correlation between texture and grain
ture. On the other hand, in the 4-pass specimen
size dependencies of 0.2% proof stress.
processed at 573 K, the 0.2% proof stress is higher
The 0.2% proof stress of the 8-pass as-ECAE specimen is
and the elongation is lower because the basal plane is
lower than the value expected from extrapolating the data for
oriented parallel to the tensile direction.
the annealed specimens. In this specimen, basal slip occurs
(3) Intensified degrees of texture of 8-pass ECAE-proces-
easily because the intensity of the basal planes inclined to the
sed specimens are weakened by annealing. The 0.2%
tensile direction is higher than that of the annealed speci-
proof stress of the annealed specimens decreases with
mens. Therefore, cross-slip to non-basal plane may occur
increasing grain sizes, exhibiting clear grain size
easily. This mechanism weakens the ability to obstruct
dependency according to Hall-Petch relation, that is,
dislocation motion. As a result, the grain size dependency
0:2 ¼ 30 þ 0:17d1=2 .
might be reduced. Furthermore, Mabuchi et al.12) reported
that the 0.2% proof stress of ECAE processed AZ91 alloy
Acknowledgements
having average grain size of 0.5 mm is lower than the value
obtained from Hall-Petch relation using other specimens. The
The present study is supported by Grant-in-Aids for
specimen of the ECAE processed AZ91 alloy contains many
Scientific Research (A) (No. 11305049) and the Priority Area
low-angle grain boundaries due to low processing tempera-
(B) ‘‘Platform Science and Technology for Advanced
ture of 448 K. The authors concluded that absorption of
Magnesium Alloys (No. 11225101)’’ from the Ministry of
dislocations at the low-angle grain boundaries causes the
Education, Culture, Sports, Science and Technology of
reduction of the 0.2% proof stress. It is also deduced that the
Japan.
above mechanism reduces the 0.2% proof stress of the
present 8-pass as-ECAE specimen because low-angle grain
REFERENCES
boundaries might be formed as a result of 8-pass ECAE
processing at a relatively low temperature of 473 K. 1) T. Mukai, T. Mohri, M. Mabuchi, M. Nakamura, K. Ishikawa and K.
As mentioned earlier, if the grain size dependency of Higashi: Scr. Mater. 39 (1998) 1249–1253.
tensile properties of magnesium alloys having strong crystal 2) K. Kubota, M, Mabuchi and K. Higashi: J. Mater. Sci. 34 (1999) 2255–
anisotropy is discussed, the effect of crystallographic 2262.
3) T. Mukai, M. Yamanoi and K. Higashi: Mater. Sci. Forum 350–351
orientation should be considered. In addition, other micro-
(2000) 97–102.
structural factors, for instance, dislocation density, amount of 4) A. Yamashita, Z. Horita and T. G. Langdon: Mater. Sci. Eng. A300
precipitates, grain boundary characteristics and so on, should (2001) 142–147.
also be considered. In the present study, the textures of the 5) R. W. Armstrong: Acta Metall. 16 (1968) 347–355.
annealed specimens that exhibit clear grain size dependency 6) D. V. Wilson: J. Inst. Metals 98 (1970) 133–143.
7) G. Sambasiva Rao and Y. V. R. K. Prasad: Metall. Trans. A 13A (1982)
are similar. Furthermore, it is inferred that the effect of
2219–2226.
dislocation density and the grain boundary characteristics on 8) T. Mohri, T. Nishiwaki, T. Kinoshita, H. Iwasaki, M. Mabuchi, M.
Tensile Properties of ECAE-Processed AZ31 Magnesium Alloy 475
Nakamura, T. Asahina, T. Aizawa and K. Higashi: Mater. Trans., JIM 18) Y. Yoshida, H. Yamada, K. Kamado and Y. Kojima: J. JILM 51 (2001)
41 (2000) 1154–1156. 556–562.
9) M. Mabuchi and K. Higashi: Acta Mater. 44 (1996) 4611–4618. 19) V. M. Segal: Mater. Sci. Eng. A197 (1995) 157–164.
10) Y. Chino, M. Mabuchi, K. Shimojima, Y. Yamada, C. Wen, K. Miwa, 20) V. M. Segal: Mater. Sci. Eng. A271 (1999) 322–333.
M. Nakamura, T. Asahina, K. Higashi and T. Aizawa: Mater. Trans. 42 21) Y. Iwahashi, J. Wang, Z. Horita, M. Nemoto and T. G. Langdon: Scr.
(2001) 414–417. Mater. 35 (1996) 143–146.
11) T. Mukai, M. Yamanoi, H. Watanabe, K. Ishikawa and K. Higashi: 22) L. G. Schulz: J. Appl. Phys. 20 (1949) 1030–1033.
Mater. Trans. 42 (2001) 1177–1181. 23) M. H. Yoo: Metall. Trans. A 12A (1981) 409–418.
12) M. Mabuchi, Y. Chino, H. Iwasaki, T. Aizawa and K. Higashi: Mater. 24) J. Koike, T. Kobayashi, T. Mukai, H. Watanabe, M. Suzuki, K.
Trans. 42 (2001) 1182–1189. Maruyama and K. Higashi: Acta Mater. in press.
13) H. Yoshinaga and R. Horiuchi: Trans. JIM 4 (1963) 1–8. 25) K. Kobayashi, J. Koike, T. Mukai, M. Suzuki, K. Higashi and K.
14) I. Gokyu: J. JSME 67 (1964) 488–493. Maruyama: Collected Abstracts of the 2002 Spring Meeting of the
15) K. Ohtoshi and M. Katsuta: J. JILM 51 (2001) 534–538. Japan Inst. Metals (2002) p. 88.
16) J. Kaneko, M. Sugamata, M. Numa, Y. Nishikawa and H. Takada: J. 26) J. Koike and K. Kobayashi: Proceedings of a subcommittee symposium
Japan Inst. Metals 64 (2000) 141–147. of the Japan Inst. Metals ‘‘Fundamentals of magnesium and industrial
17) M. T. Pérez-Prado and O. A. Ruano: Scr. Mater. 46 (2002) 149–155. applications’’, (The Japan Inst. Metals, 2001) pp. 1–3.