Materials Science & Engineering A 637 (2015) 243–250

Contents lists available at ScienceDirect

Materials Science & Engineering A

journal homepage: www.elsevier.com/locate/msea

Improved mechanical properties of AZ31 magnesium alloy sheets

by repeated cold rolling and annealing using a small pass reduction
Hua Zhang a,b,c,n, Weili Cheng a,b,c,n, Jianfeng Fan a,b,c, Bingshe Xu a,b,c, Hongbiao Dong d
a
Key Laboratory of Interface Science and Engineering in Advanced Materials, Ministry of Education, Taiyuan University of Technology, Taiyuan 030024, China
b
Shanxi Research Center of Advanced Materials Science and Technology, Taiyuan 030024, China
c
College of Materials Science and Engineering, Taiyuan University of Technology, Taiyuan 030024, China
d
Department of Engineering, University of Leicester, Leicester LE1 7RH, UK

art ic l e i nf o a b s t r a c t

Article history: Much less attention has been paid to improve the mechanical properties of magnesium alloy by cold
Received 4 March 2015 rolling due to its poor room temperature workability. So the present work attempts to improve the
Received in revised form mechanical properties of AZ31 magnesium alloy sheets processed by repeated cold rolling and annealing
8 April 2015
using a small pass reduction in correlation with the concurrent microstructure evolution. The obtained
Accepted 17 April 2015
Available online 28 April 2015
results indicated that after 87% total repeated cold rolling reduction and annealing, the grain micro-
structure of AZ31 alloy sheet was successfully refined from 22.3 μm to 7.4 μm. The alternating change
Keywords: induced by repeated cold rolling and annealing between deformed microstructure and recrystallized
Magnesium alloys microstructure resulted in the continued refinement of grains. After repeated cold rolling and annealing,
Microstructure
the mechanical properties of AZ31 alloy sheets were improved, especially the Erichsen value (IE) of the
Mechanical properties
87% total rolled sample in annealed condition was significantly increased by more than 36% compared to
Repeated cold rolling
Thin sheets that of the as-received sample under the same thickness. Improving mechanical properties of AZ31 alloy
sheets processed by repeated cold rolling and annealing are ascribed to the grain refinement and
weakened basal texture. Finally, 0.40 mm thick thin sheets with good comprehensive mechanical
properties were obtained by repeated cold rolling and annealing using a small pass reduction.
& 2015 Elsevier B.V. All rights reserved.

1. Introduction compression (CEC) [11] and equal channel angular pressing (ECAP)
[12,13]. In addition, it has been found that aluminum as an alloying
Magnesium alloys, the lightest metal structural material, are element resulted in a decrease in the tension/compression asym-
increasingly being applied for electronic communications, aerospace metry in magnesium alloys, which is mainly due to a change in the
and automotive industries due to their high specific strength, high activity of basal and non-basal slip [14]. Chino et al. [15] demon-
specific stiffness, good heat radiation, high electromagnetic shielding strated that the split basal texture at 735° along the transverse
and so on [1,2]. However, magnesium alloys exhibit poor room direction in magnesium alloys was obtained by adding RE elements,
temperature ductility due to their hexagonal close packed (hcp) which resulted in the high formability.
crystal structure, where there are limited operative slip systems [3]. As one of the important metal plastic processing technologies,
Their processing technologies like rolling, extrusion, drawing, etc. are rolling has also been widely used in magnesium alloys. There have
usually carried out at elevated temperature [4–6], thus their wide been many reports on the study and application of hot rolling in
applications are limited. Therefore, extensive efforts have been magnesium alloys [16–18]. After the AZ31 alloy sheets were sym-
devoted to improving the mechanical properties of magnesium metrically rolled at a high temperature of 823 K followed by a warm
alloys using different processing technologies, e.g. friction stir pro- rolling at 498 K, the basal texture was inclined and thus the superior
cess (FSP) [7,8], different speed rolling (DSR) [9,10], cyclic extrusion formability was obtained [19]. Xu et al. [20] reported that large
thickness reduction in the final hot rolling pass could increase the
n
Corresponding author at: Key Laboratory of Interface Science and Engineering volume fraction of the dynamic recrystallization and weaken the
in Advanced Materials, Ministry of Education, Taiyuan University of Technology, basal texture of Mg–RE alloys, which resulted in the improvement of
Taiyuan 030024, China. Tel.: þ 86 351 6014852; fax: þ86 351 6010311.
E-mail addresses: [email protected] (H. Zhang),
the yield anisotropy and the ductility. Liu et al. [21] found that sig-
[email protected] (W. Cheng). nificant grain refinement of magnesium alloys was achieved by

http://dx.doi.org/10.1016/j.msea.2015.04.057
0921-5093/& 2015 Elsevier B.V. All rights reserved.
244 H. Zhang et al. / Materials Science & Engineering A 637 (2015) 243–250

multi-pass hot rolling, where the grain size was reduced from thickness of magnesium alloy sheets was reduced to 1.72 mm. When
250 μm in ingots to less than 30 μm in as-rolled sheets. the total cold rolling reduction accumulated to 87%, magnesium alloy
Similarly, many research works have also been conducted on thin sheets with a thickness of 0.40 mm were obtained. After repe-
the cold rolling of magnesium alloy sheets [22–24]. Styczynski ated cold rolling, the rolled sheets were subjected to final annealing
et al. [22] reported that the cold rolling texture of AZ31 magne- at 250 °C for 30 min.
sium alloy sheets consisted of double peaks tilting 715° from the A standard metallographic technique was used to examine the
normal direction to the rolling direction in the basal pole. The optical microstructure. For electron backscatter diffraction (EBSD)
severity, frequency and lifetime of shear bands formed during cold observation, samples preparation consisted of mechanical grinding,
rolling affect the cold rollability of magnesium alloy sheets [23]. In mechanical polishing and electropolishing. After mechanical grind-
a word, most studies focused on the microstructure evolution of ing, mechanical polishing was conducted with a diluted OPS solution
magnesium alloys during cold rolling [22–24]. However, much less (10%), which contained 10% colloidal silica suspension. Then elec-
attention has been paid to improve the mechanical properties of tropolishing was performed for 10–20 s in an AC2 solution under
magnesium alloys by cold rolling due to their poor workability.  30 °C using a voltage of 15 V. Finally, the samples were cleaned
Therefore, the present work aims to produce thin sheets with good with methanol and acetone. The EBSD operations were carried out
comprehensive mechanical properties by repeated cold rolling and on a Zeiss Supra 55VP FEG-SEM using a voltage of 20 kV, 20 mm
annealing with a small pass reduction. Moreover, the concurrent working distance, a tilt angle of 70° and a scan step of 1 μm.
microstructure evolution was also investigated. Tensile samples with nominal gage dimensions of 10 mm length,
6 mm width and sheet thickness were prepared at the angles of 0°,
45° and 90° to the rolling direction by wire-cutting. Then, uniaxial
2. Experiment procedures tensile tests were performed at room temperature under a strain rate
of 1  10  3 s  1 on a CMT6305-300KN electronic universal testing
As-received AZ31 alloy sheets (Mg–3 wt% Al–1 wt% Zn) with a machine to determine the mechanical properties, e.g. yield strength,
thickness of 3 mm, machined into 100 mm  80 mm (length  ultimate strength and fracture elongation. The blanks with a square
width) samples along the rolling direction, were used in repeated of 55 mm  55 mm were used for the Erichsen tests in order to
cold rolling and annealing. A small rolling reduction of 10–15% was investigate the stretch formability. Standard Erichsen tests were
applied to the cold rolling. The cold rolled samples were sent into the carried out at room temperature using a hemispherical punch having
furnace and annealed at 250 °C for 30 min after each cold rolling a diameter of 20 mm. Here, the punch speed was set at 2.5 mm/min
pass. Being cooled in the air, the rolled samples were subjected to the and the blank-holder force was set at 10 kN. Agraphite grease as a
next cold rolling. Finally, after repeated cold rolling and annealing, lubricant was pasted on only the center part of the samples. The
the samples were cold rolled to different total thickness reductions Erichsen value (IE) which was the punch stroke at fracture initiation
of 43% and 87%, respectively. After 43% total rolled reduction, the was measured. In the present work, RD, ND and TD represent the

Fig. 1. EBSD maps and distributions of grain size of various samples in annealed condition: (a) as-received, (b) 43% total rolled and (c) 87% total rolled.
 H. Zhang et al. / Materials Science & Engineering A 637 (2015) 243–250 245

rolling, the normal and the transverse direction of the as-received

sheets, respectively. The fracture surfaces of tensile samples were
examined by Zeiss Supra 55VP FEG-SEM, using an accelerating vol-
tage of 20 kV.

3. Results and discussion

3.1. Microstructure evolution

Fig. 1 shows EBSD maps and distributions of grain size of the

as-received, 43% total rolled and 87% total rolled samples in
annealed condition. The orientation imaging microscopy (OIM)
software was applied to determine the distribution of grain size to Fig. 3. Optical microstructure of 87% total rolled samples, where twins and shear
analyze its change. It is seen from Fig. 1a that the as-received bands are labeled by green arrows and red arrows, respectively. (For interpretation
sample possesses an equiaxed grain microstructure with a statis- of the references to color in this figure legend, the reader is referred to the web
version of this article.)
tical average grain size of 22.3 μm. Compared with the as-received
sample, 43% total rolled and 87% total rolled samples in annealed
condition show finer recrystallized grain microstructure. Fig. 2 refine the grain microstructure of magnesium alloys. It is important
shows the misorientation distribution of 87% total rolled sample in to note that the fine grain bands with a tilting angle of 25–50° to the
annealed condition, where there is a peak distribution around 30° RD are found in repeated cold rolled and annealed sheets, especially
which corresponds to DRX grain boundaries [25] and a much in Fig. 1c. The formation of fine grain bands is mainly due to the
lower fraction of low angle boundaries (LABs). A low peak around formed share bands during cold rolling (see Fig. 3), where there is
86° resulting from {10  12} twinning boundaries is also observed. high distortion energy. Therefore, fine grains easily form in shear
These indicate that the microstructure of repeated cold rolled bands due to static recrystallization during annealing. In addition,
sample has been nearly recrystallized completely after annealing the different colors represent different grain orientations in EBSD
at 250 °C for 30 min. maps. It is seen from Fig. 1 that the grains with various different
After cold rolling, the dominant features of the microstructure are colors gradually increase with the increasing rolling reduction, which
twins and shear bands, as shown in Fig. 3. It was suggested that means some grain orientations appear to have changed after repe-
these twins were mainly {10–12} twins and occasional {10–11}–{10– ated cold rolling and annealing. To verify this, Fig. 4a and b provides
12} double twins [23]. Double twinning, which involves the forma- the enlarged EBSD images of local areas in Fig. 1a and c, respectively.
tion of secondary {10–12} twins within primary {10–11} twins [26], The distributions of some grains are also shown in (0002) pole fig-
greatly affects the formation of the cold rolled microstructure ures. As shown in Fig. 4a, almost all of grains (G1–G12) distribute in
[27,28]. For magnesium alloys with hcp crystal structure, the lack of or near the center of (0002) pole figure in the as-received sample. In
sufficient independent slip systems cannot contribute to the total contrast, some grains (G7–G10, G12) rotate about 60–85° away from
strain during cold rolling, and accordingly a large number of twins the ND in the 87% total rolled sample in annealed condition, as
accompanied shear bands are easily formed [23,24]. In addition, the indicated in Fig. 4b.
formed shear bands go through the thickness direction of sheet with
3.2. Texture evolution and Schmid factor
a tilting angle of 25–50° to the RD, as shown in Fig. 3.
Recrystallization preferentially develops within shear bands,
Fig. 5 shows the (0002) pole figures of the as-received, 43% total
twins, intersections between grain boundaries and twins, which are
rolled and 87% total rolled samples in annealed condition. It can be
the potential nucleation sites for recrystallization [29,30]. This will
seen from the (0002) pole figure in Fig. 5a that the as-received
result in the development of fine recrystallized grains during sub-
sample possesses a typical (0001) basal texture and the maximum
sequent annealing. The alternating change induced by repeated cold
texture intensity is 14.5, where the c-axis of most grains are parallel
rolling and annealing between deformed microstructure and
to the ND. After repeated cold rolling and annealing, the rolled
recrystallized microstructure will lead to the continued refinement
samples in annealed condition still exhibit typical basal textures but
of grains. As shown in Fig. 1c, after 87% total repeated cold rolling
the c-axis of some grains are tilted away from the ND towards the RD
reduction and annealing, the grain microstructure of AZ31 alloy
and lay on the ND–RD plane, which leads to the ellipsoidal dis-
sheet was successfully refined from 22.3 μm to 7.4 μm. The results
tributions of basal texture intensity in the (0002) pole figures, as
show that the repeated cold rolling and annealing can effectively
shown in Fig. 5b and c. This is mainly due to fact that the combi-
nation of non-basal slip and twinning results in the formation of
such textures [31]. Compared with the basal texture intensity of the
as-received sample, those of the repeated cold rolled samples in
annealed condition gradually reduce with the increasing total rolling
reduction, which is ascribed to the increasingly accumulated strain
during repeated cold rolling.
Fig. 6 shows distributions of Schmid factors in the case that the
loading axis is parallel to the RD of the as-received, 43% total rolled
and 87% total rolled samples in annealed condition. It was sug-
gested that for magnesium alloys, the critical resolved shear stress
(CRSS) of a basal slip system at room temperature was far below
those of non-basal slip systems [32], thus Fig. 6 only indicates the
Schmid factor of the basal 〈a〉 slip system, obtained by EBSD ana-
lysis of various samples. It is seen from Fig. 6a that for the as-
Fig. 2. Misorientation distribution of 87% total rolled sample in annealed condition. received sample, the Schmid factor of basal {0001}〈11–20〉 slip is
246 H. Zhang et al. / Materials Science & Engineering A 637 (2015) 243–250

Fig. 4. (a) and (b) are the enlarged EBSD images of local areas in Fig. 1(a) and (c), respectively. Distributions of some grains are obtained in (0002) pole figures. Grains 1–12
are marked by G1–G12, respectively.

relatively small, only 0.19, which means most of the grains have 3.3. Mechanical properties
hard-orientation and it is difficult to activate the basal 〈a〉 slip
system during subsequent deformation, indicating that the as- True stress–strain curves of various samples in annealed condi-
received sample possesses the poor plasticity. However, after tion in different tensile direction are presented in Fig. 8 and the
repeated cold rolling and annealing, the soft-oriented grains of corresponding mechanical properties, e.g. 0.2% yield strength (YS),
basal 〈a〉 slip system are greatly increased. For the 43% total rolled ultimate strength (UTS) and failure elongation (FE) are summarized
and 87% total rolled samples in annealed condition, the average in Table 1, also included are the average grain size, basal texture
Schmid factors of basal 〈a〉 slip system increase from 0.19 to 0.23 intensity and Schmid factor. For various samples in annealed con-
and 0.26, respectively. dition, the average YS, UTS and FE in different tensile direction are
Fig. 7 shows optical microstructure in the top part of different summarized in Fig. 9. It is can be seen from the true stress–stain
samples stretch-formed to failure. It has been reported that dis- curves in Fig. 8 that various samples exhibit yield anisotropy in three
locations in {10–12} boundaries were wide, had a low energy and different tensile direction, which is a typical characteristic of the
were very glissile, so {10–12} twins had a convex configuration, conventional rolled AZ31 alloy sheets [36] and relates to the texture
whereas those in {10–11} boundaries had the opposite character- [37]. In addition, the increasing total repeated rolling reduction
istics and accordingly {10–11} twins exhibited a flat configuration causes larger yield anisotropy, as shown in Fig. 8b and c. This is
[33]. It is seen from Fig. 7a that many twins with a convex con- mainly due to the fact that after repeated cold rolling and annealing,
figuration are found in the as-received sample stretch-formed to the c-axis of some grains become tilted away from the ND towards
failure and can be identified to be {10–12} twins. In addition to the the RD, as indicated in Fig. 5. It is seen from Table 1 that compared
twins with a convex configuration, many twins with a flat con- with the as-received sample, the 43% total rolled and 87% total rolled
figuration are also observed in the 87% total rolled samples and samples in annealed condition exhibit higher strengths especially
can be judged to be {10–11} twins, as shown in Fig. 7b. The {10– the yield strength regardless of the tensile direction. As shown in
12} extention twins generally happen at the early stage of defor- Fig. 9, the average YS and UTS of the 87% total rolled sample in
mation, while the appearance of {10–11} contraction twins in a annealed condition increase from 147 MPa and 274 MPa of the as-
significant number usually occurs at strain beyond 10% [34,35]. received sample to 169 MPa and 283 MPa, respectively. According to
Therefore, it can be concluded that the 87% total rolled sample in Hall–Petch relationship, the increasing strength is mainly due to the
annealed condition has experienced much larger strain than the grain refinement strengthening resulting from repeated cold rolling
as-received sample during Erichsen testing. and annealing. The decrease in grain size will increase the resistance
 H. Zhang et al. / Materials Science & Engineering A 637 (2015) 243–250 247

Fig. 5. (0002) pole figures of various samples in annealed condition: (a) as-received, (b) 43% total rolled and (c) 87% total rolled.

Fig. 6. Distributions of Schmid factors in the case that the loading axis is parallel to the RD of various samples in annealed condition: (a) as-received, (b) 43% total rolled and
(c) 87% total rolled.

to dislocation motion, and thus resulting the increasing yield sample in different tensile direction. Especially in the RD, the
strength. increasing FE of the 43% total rolled and 87% total rolled samples in
As shown in Table 1, the 43% total rolled and 87% total rolled annealed condition is more evident, which is due to the fact that the
samples in annealed condition exhibit larger FE than the as-received basal texture becomes more diverse along the RD. As shown in Fig. 9,
248 H. Zhang et al. / Materials Science & Engineering A 637 (2015) 243–250

Fig. 7. Optical microstructure in the top part of different samples stretch-formed to failure: (a) as-received and (b) 87% total rolled samples in annealed condition with a
thickness of 0.40 mm.

Fig. 8. True stress–strain curves of various samples in annealed condition in different tensile direction: (a) as-received, (b) 43% total rolled and (c) 87% total rolled.

Table 1
Mechanical properties of various samples in annealed condition.

Sample YS (MPa) UTS (MPa) FE (%) Grain size (μm) Basal texture intensity Schmid factor

RD 45° TD RD 45° TD RD 45° TD

As-received 141 145 157 273 273 275 15.2 16.1 15.1 22.3 14.5 0.19
43% Total rolled 153 160 169 273 274 279 17.9 17.6 16.7 14.7 11.8 0.23
87% Total rolled 158 167 183 280 281 291 20.3 19.0 18.6 7.4 8.3 0.26

the average FE of the 43% total rolled and 87% total rolled samples in weakened basal texture intensity. For magnesium alloys with a small
annealed condition increase from 15.6% of the as-received sample to Schmid factor, it is difficult for basal slip system activation during
17.5% and 19.2%, respectively. For the 43% total rolled and 87% total deformation, and accordingly the strain localization and premature
rolled samples in annealed condition, the increase in the FE is ascribed shear failure are easily produced [38,39]. Furthermore, the grain
to the larger Schmid factor of basal 〈a〉 slip resulting from the refinement can also contribute to the increasing fracture elongation.
 H. Zhang et al. / Materials Science & Engineering A 637 (2015) 243–250 249

It can be seen from Fig. 10a that the Erichsen value (IE) is only polished to 0.40 mm in order to exclude the effect of thickness. The
2.17 mm for the as-received sample with a thickness of 3 mm. As results show that the IE of the as-received samples with a thickness
shown in Fig. 10c, compared with the as-received sample, the IE of of 0.40 mm is still low, with 3.41 mm, as shown in Fig. 10b. There-
the 87% total rolled sample in annealed condition with a thickness of fore, under the same thickness, the IE of the 87% total rolled sample
0.40 mm is significantly increased to 4.64 mm, with 114% improved. in annealed condition is still increased by more than 36% compared
Considering the different thickness affecting the formability of to that of the as-received sample. The improved formability of
samples, the as-received samples with a thickness of 3 mm were also repeated cold rolled and annealed sheet can be attributed to the
reduction in (0002) texture intensity (8.3 vs. 14.5), as shown in Fig. 5.
The strong basal texture results in a small Schmid factor of basal
{0001}〈11–20〉 slip in the as-received sample, as shown in Fig. 6a. The
small Schmid factor indicates that most grains are in hard orienta-
tion and it is difficult to activate the basal 〈a〉 slip system, which leads
to stress localization and early failure [38,39]. Furthermore, the
strong texture of magnesium alloys will induce a poor deformation
capability of sheet thinning [40,41], which lead to a low ductility at
room temperature. The grain refinement can also contribute to the
increasing formability. Thus, the repeated cold rolling and annealing
effectively improved the stretch formability of AZ31 alloy sheets by
refining grain and weakening basal texture. Above knowable,
0.40 mm thick thin sheets with good comprehensive mechanical
Fig. 9. Average YS, UTS and FE of various samples in annealed condition in different properties were obtained by repeated cold rolling and annealing
tensile direction.
using a small pass reduction.

3.4. Fractography

Fig. 11 shows the SEM tensile fracture images of the as-received,

43% total rolled and 87% total rolled samples in annealed condition
along the RD. The dominant feature of the fracture surface is quasi-
cleavage fracture in the as-received sample and a small amount of
local shallower dimples are formed inside the cleavage planes,
indicating that the as-received sample exhibits a poor plasticity.
However, it can be seen clearly from Fig. 11b that the 43% total rolled
sample in annealed condition shows more dimples than the as-
received sample. And the 87% total rolled sample in annealed con-
dition exhibits some tearing edges and a large number of dimples, as
shown in Fig. 11c. These features belong to the ductile fracture and
Fig. 10. Erichsen values of (a) as-received sample, (b) as-received sample with a
thickness of 0.40 mm and (c) 87% total rolled sample in annealed condition with a they are believed to be indications for the improved plasticity. These
thickness of 0.40 mm. results obviously indicate that the plasticity of AZ31 magnesium

Fig. 11. SEM tensile fracture images of various samples in annealed condition along RD: (a) as-received, (b) 43% total rolled and (c) 87% total rolled.
250 H. Zhang et al. / Materials Science & Engineering A 637 (2015) 243–250

alloy sheets can be enhanced by repeated cold rolling and annealing [5] L.L. Chang, Y.N. Wang, X. Zhao, J.C. Huang, Mater. Sci. Eng. A 496 (2008)
using a small pass reduction. 512–516.
[6] S. Gall, R.S. Coelho, S. Müller, W. Reimers, Mater. Sci. Eng. A 579 (2013)
180–187.
[7] F.Y. Zheng, Y.J. Wu, L.M. Peng, X.W. Li, P.H. Fu, W.J. Ding, J. Magnes. Alloy. 1
4. Conclusions (2013) 122–127.
[8] G. Venkateswarlu, M.J. Davidson, G.R.N. Tagore, Mater. Des. 40 (2012) 1–6.
[9] W.J. Kim, H.W. Lee, S.J. Yoo, Y.B. Park, Mater. Sci. Eng. A 528 (2011) 874–879.
In summary, microstructure evolution and mechanical prop- [10] X. Huang, K. Suzuki, A. Watazu, I. Shigematsu, N. Saito, J. Alloy. Compd. 457
erties of AZ31 magnesium alloy sheets processed by repeated cold (2008) 408–412.
[11] Y.J. Chen, Q.D. Wang, H.J. Roven, M. Karlsen, Y.D. Yu, M.P. Liu, J. Hjelen., J. Alloy.
rolling and annealing were investigated. The results indicated that Compd. 462 (2008) 192–200.
after repeated cold rolling and annealing, the grain microstructure [12] Y. Miyahara, Z. Horita, T.G. Langdon, Mater. Sci. Eng. A 420 (2006) 240–244.
of AZ31 magnesium alloys was effectively refined, which was [13] M. Janeček, M. Popov, M.G. Krieger, R.J. Hellmig, Y. Estrin, Mater. Sci. Eng. A
462 (2007) 116–120.
ascribed to the alternating change between deformed micro-
[14] J. Bohlen, P. Dobroň, J. Swiostek, D. Letzig, F. Chmelíkb, P. Lukáčb, K.U. Kainer,
structure and recrystallized microstructure. Repeated cold rolling Mater. Sci. Eng. A 462 (2007) 302–306.
and annealing also produced the weakened basal texture, which [15] Y. Chino, K. Sassa, M. Mabuchi, Mater. Trans. 49 (2008) 1710–1712.
led to a larger Schmid factor of 〈a〉 basal slip system. The repeated [16] C. Xu, S.W. Xu, M.Y. Zheng, K. Wu, E.D. Wang, S. Kamado, G.J. Wang, X.Y. Lv,
J. Alloy. Compd. 524 (2012) 46–52.
cold rolled and annealed samples exhibited higher strength and [17] S.M. Fatemi-Varzaneh, A. Zarei-Hanzaki, M. Haghshenas, J. Alloy. Compd. 475
larger fracture elongation than the as-received sample, especially (2009) 126–130.
the IE of the 87% total rolled sample in annealed condition was [18] M.T. Pérez-Prado, J.A. del Valle, J.M. Contreras, O.A. Ruano, Scr. Mater. 50
(2004) 661–665.
significantly increased by more than 36% compared to that of the [19] X. Huang, K. Suzuki, Y. Chino, Scr. Mater. 63 (2010) 395–398.
as-received sample under the same thickness. The improved [20] C. Xu, M.Y. Zheng, K. Wu, E.D. Wang, G.H. Fan, S.W. Xu, S. Kamado, X.D. Liu,
mechanical properties were ascribed to the grain refinement and G.J. Wang, X.Y. Lv, M.J. Li, Y.T. Liu, Mater. Sci. Eng. A 559 (2013) 232–240.
[21] D. Liu, Z. Liu, E. Wang, Mater. Sci. Eng. A 612 (2014) 208–213.
weakened basal texture. Finally, 0.40 mm thick thin sheets with [22] A. Styczynski, Ch Hartig, J. Bohlen, D. Letzig, Scr. Mater. 50 (2004) 943–947.
good comprehensive mechanical properties were produced by [23] M.R. Barnett, M.D. Nave, C.J. Bettles, Mater. Sci. Eng. A 386 (2004) 205–211.
repeated cold rolling annealing using a small pass reduction. [24] L.L. Chang, E.F. Shang, Y.N. Wang, X. Zhao, M. Qi, Mater. Charact. 60 (2009)
487–491.
[25] Y. Xin, M. Wang, Z. Zeng, G. Huang, Q. Liu, Scr. Mater. 64 (2011) 986–989.
[26] R.E. Reed-Hill, Trans. Metall. Soc. AIME 218 (1960) 554–558.
Acknowledgments [27] B.C. Wonsiewicz, W.A. Backofen, Trans. Metall. Soc. AIME 239 (1967)
1422–1431.
[28] S.L. Couling, J.F. Pashak, L. Sturkey, Trans. ASM 51 (1959) 94–107.
This work is supported by Fund Program for the Scientific [29] X. Li, P. Yang, L.N. Wang, L. Meng, F. Cui, Mater. Sci. Eng. A 517 (2009) 160–169.
Activities of Selected Returned Overseas Professionals in Shanxi [30] M.M. Myshlyaev, H.J. McQueen, A. Mwembela, E. Konopleva, Mater. Sci. Eng. A
337 (2002) 121–133.
Province, Shanxi Province Science Foundation for Youths, Scientific [31] Y. Chino, K. Kimura, M. Mabuchi, Mater. Sci. Eng. A 483 (2008) 481–488.
and Technological Innovation Programs of Higher Education Insti- [32] H. Yoshinaga, R. Horiuchï, Trans. Jpn. Inst. Met. 4 (1963) 14–21.
tutions in Shanxi (2014118), Program for New Century Excellent [33] A. Serra, D.J. Bacon, Philos. Mag. A 63 (1991) 1001–1012.
[34] Z. Keshavarz, M.R. Barnett, Scr. Mater. 55 (2006) 915–918.
Talents in University (NCET-12-1040) and National Natural Science
[35] A. Levinson, R.K. Mishra, R.D. Doherty, S.R. Kalidindi, Acta Mater. 61 (2013)
Foundation of China (50901048 and 51174143). 5966–5978.
[36] D.W. Brown, S.R. Agnew, M.A.M. Bourke, T.M. Holden, S.C. Vogel, C.N. Tomé,
Mater. Sci. Eng. A 399 (2005) 1–12.
[37] Q. Yang, B. Jiang, G. Zhou, J. Dai, F. Pan, Mater. Sci. Eng. A 590 (2014) 440–447.
References [38] S.R. Agnew, O. Duygulu, Mater. Sci. Forum 419–422 (2003) 177–188.
[39] S.R. Agnew, O. Duygulu, Int. J. Plast. 21 (2005) 1161–1193.
[1] E. Aghion, B. Bronfin, Mater. Sci. Forum 350–351 (2000) 19–30. [40] H. Zhang, G.S. Huang, L.F. Wang, J.H. Li, Scr. Mater. 67 (2012) 495–498.
[2] B.L. Mordike, T. Ebert, Mater. Sci. Eng. A 302 (2001) 37–45. [41] J. Koike, T. Kobayashi, T. Mukai, H. Watanabe, M. Suzuki, K. Maruyama,
[3] R.E. Reed-Hill, W.D. Robertson, Acta Metall 5 (1957) 728–737. K. Higashi, Acta Mater. 51 (2003) 2055–2065.
[4] S. Liang, H. Sun, Z. Liu, E. Wang, J. Alloy. Compd. 472 (2009) 127–132.