Materials Science & Engineering A 607 (2014) 512–520

Contents lists available at ScienceDirect

Materials Science & Engineering A

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

Work hardening behavior of the extruded and equal-channel angularly

pressed Mg–Li–Zn alloys under tensile and shear deformation modes
M. Karami, R. Mahmudi n
School of Metallurgical and Materials Engineering, College of Engineering, University of Tehran, Tehran, Iran

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

Article history: The flow and work hardening behaviors of extruded and equal-channel angularly pressed (ECAPed)
Received 13 February 2014 Mg–6Li–1Zn (LZ61) and Mg–12Li–1Zn (LZ121) alloys were studied by tensile and shear punch testing
Received in revised form methods. It was shown that the Kocks–Mecking type plots for tensile and shear deformation of both
8 April 2014
alloys, exhibited similar work hardening (WH) stages in both extruded and ECAPed conditions. WH rates
Accepted 10 April 2014
Available online 18 April 2014
were found to be lower for the ECAPed materials, due to a reasonably uniform and well-refined
microstructure. In the case of hcp LZ61 alloy, textural studies showed that the extruded fiber-type
Keywords: texture was replaced by a typical ECAP texture, in which basal planes rotated about 451 to the extrusion
Magnesium alloys axis. This was found to be responsible for the lower tensile strength and higher shear strength in the
Work hardening
ECAPed material, as compared to the extruded condition. For the bcc LZ121 alloy, it was observed that
ECAP
the grain refinement achieved after ECAP increases the strength and ductility in both tensile and shear
Shear punch test
Plastic deformation deformation, compared with those of extruded condition. Stage II of the Kocks–Mecking plot in both
shear and tensile deformation of LZ121 was eliminated most likely due to stacking fault energy
improvement caused by higher Li content of the Mg lattice structure. The shear punch testing (SPT)
method was found to yield the flow and WH curves similar to those obtained in tensile testing.
& 2014 Elsevier B.V. All rights reserved.

1. Introduction deformation mechanisms and enhance the formability of the

material, which is accompanied by a decrease in the strength [9].
Magnesium alloys have attracted increasing attention from auto- Work hardening (WH) behavior, which influences strength and
motive and aerospace industries because of their low density, high ductility, is one of the important considerations in evaluating the
specific strength and excellent damping capacity. Despite these plastic deformation of materials [10]. Despite the great interest in the
advantages, the poor room temperature formability of Mg alloys, a WH behavior of Mg alloys [11–13], only limited studies have been
consequence of the limited slip systems in the hexagonal close- performed on Mg–Li alloys [14]. Wu and co-workers have investigated
packed (hcp) structure, restricts their applications [1,2]. Some of the the plastic anisotropy and work hardening behavior of the cold-rolled
possible means of addressing this deficiency are microstructural dual-phase Mg–Li–Zn alloys at room [14] and high temperatures [15].
refinement [3,4], and use of alloying elements [5]. Among many They used Kocks–Mecking type plots to illustrate different stages
possibilities, the promising severe plastic deformation technique of of work hardening and concluded that the anisotropic tensile behavior
equal-channel angular pressing (ECAP) is capable of producing fine- of the alloys may be related to the development of texture and
grained microstructures [6]. The principles of the ECAP process have microstructure during cold deformation.
been reviewed by Valiev and Langdon [7]. Enhancement of cold Most of the WH studies have mainly focused on the investigation
formability of Mg alloys could also be achieved using Li as an of tensile deformation behavior determined by the conventional
alloying element. Mg–Li alloys, with Li contents between 5 and tensile tests. There are, however, cases in which the material is only
11 wt%, exhibit a two-phase microstructure consisting of the hcp available as small thin test pieces such as those usually obtained by
Mg-rich α and the body-centered cubic (bcc) Li-rich β phases. The severe plastic deformation processes. In such circumstances, the
single β-phase structure can exist for Li contents greater than 11 wt miniature shear punch test (SPT) that is an easy-to-perform method
% [8]. A mixture of bcc and hcp phases is expected to change the [16] capable of evaluating stress–strain behavior of both cast [17,18]
and wrought [19,20] magnesium alloys, can be advantageous. In this
test, a flat-ended cylindrical punch is driven through a securely
clamped sheet sample, punching a circular disc from it. By plotting
n
Corresponding author. Tel.: þ 98 21 8208 4137; fax: þ 98 21 8800 6076. shear stress against normalized displacement, mechanical properties
E-mail address: mahmudi@ut.ac.ir (R. Mahmudi). such as shear yield stress (SYS), ultimate shear strength (USS) and

http://dx.doi.org/10.1016/j.msea.2014.04.040
0921-5093/& 2014 Elsevier B.V. All rights reserved.
 M. Karami, R. Mahmudi / Materials Science & Engineering A 607 (2014) 512–520 513

shear elongation values can be obtained from the SPT data. To the
best of authors' knowledge, SPT has not previously been used for Punch
evaluating WH behavior of any alloy through determination of WH
rates. It is therefore the aim of the current study to characterize and
compare the tensile and shear WH behaviors of the fine-grained Mg–
6Li–1Zn and Mg–12Li–1Zn alloys at room temperature. These were
accomplished by measuring tensile and shear WH rates for alloys Main body
having various textures and grain sizes, obtained by conventional
Holders
extrusion and by ECAP.

2. Experimental procedures

2.1. Materials and processing

Specimen
This investigation involved two alloys with the nominal chemical 10 mm
compositions of Mg–6 wt% Li–1 wt% Zn and Mg–12 wt% Li–1 wt% Zn
that are designated as LZ61 and LZ121, respectively. High-purity Mg Fig. 1. Schematic representation of shear-punch die assembly.
(99.8 wt%), Zn (99.9 wt%), and an Mg–30 wt% Li master alloy were
used to prepare the alloys. Melting was carried out in a graphite Shear punch tests were performed using a SANTAM universal
crucible placed in an electrical furnace under the protection of a testing machine at an initial shear strain rate of 1.2  10  3 s  1.
covering flux. The melt was held at 750 1C for 20 min and mechani- One-millimeter thick slices were cut from the extruded and
cally stirred for 2 min, using a stainless steel rod before Li-containing ECAPed bars perpendicular to the pressing direction. These slices
master alloy was added successively every 10 min to it. An additional were mechanically ground to a thickness of about 0.7 mm and
20 min was allowed to ensure a homogeneous composition and to located in a fixture with a 3.175 mm diameter flat cylindrical
settle the oxides, before pouring the melt into a steel die preheated to punch and 3.225 mm diameter receiving-hole, the schematic of
150 1C. Pouring was accomplished by a tilt-casting technique in order which is presented in Fig. 1. No lubricant was used between the
to minimize casting defects and the turbulence of the melt. specimens and dies. The applied load P was measured as a
Extrusion was performed on the as-cast billets with an extrusion function of punch displacement and the shear stress was calcu-
ratio of 11:1 at 300 1C. The ECAP process was conducted at 200 1C lated in MPa using the relationship:
through route BC, in which each sample was rotated 901 around its
P
longitudinal axis between the passages. This configuration leads to τ¼ ð1Þ
an imposed strain of about 1 on each passage through the die. πdt
Samples were sprayed with MoS2 lubricant and pressed at a speed where P is the punch load in N, t is the specimen thickness in mm
of 1 mm/s for four passes using a solid die with channel angles of and d is the average of the punch and die diameters in mm. The
ϕ¼901 and ψ¼ 201. The microstructure of the cross sections shear strain (γ) was calculated from γ¼h/W, where h is the punch-
perpendicular to the pressing direction of the extruded and ECAPed displacement and W is the die–punch clearance. Similar to tensile
billets was examined by optical microscopy. The metallographic tests, three different samples were tested for each condition, and it
samples of LZ61 and LZ121 alloys were etched with a solution of 5 g was observed that the variation in the measured strength and
picric acid, 10 mL acetic acid, and 8% nitric acid and 92% ethanol, elongation values was less than 2%.
respectively. The intensity distributions of the (0002) and (1010)
pole figures were measured by the Schultz reflection reference
method from the plane perpendicular to the pressing direction for 3. Results and discussion
both extruded and ECAPed materials. The measurement was
performed using Cu Kα radiation at 50 kV with the sample tilt 3.1. Microstructural observations
angle ranging from 01 to 901.
The microstructural evolution of the LZ61 alloy in the as-cast,
2.2. Mechanical testing extruded, and ECAPed conditions is depicted in Fig. 2a–c. As shown
in Fig. 2a, the relatively large α grains with some β phase in the grain
Miniature dog-bone tensile specimens, 4 mm long, 3 mm wide, boundary areas can be observed in the as-cast microstructure. The
and 2 mm thick, were prepared by electro-discharge wire-cut extruded microstructure, shown in Fig. 2b, consists of the α-Mg
machining along the longitudinal direction of the processed matrix with some coarse deformed grains (15–30 mm) and fine
materials. Tensile tests were carried out at room temperature with recrystallized grains (2–8 mm). Some elongated β-Li constituents
an initial strain rate of 1  10  3 s  1. Load–extension curves were located at the α grain boundaries are also evident in the micro-
obtained, from which the stress–strain curves, 0.2% yield stress, structure. Processing by ECAP leads to a more refined microstructure,
and tensile strength were automatically calculated by the machine so that a nearly uniform fine-grained structure with an average grain
where the data were acquired by a computer. Using this config- size of 6.3 mm forms, as exhibited in Fig. 2c. The grain refinement
uration, it was possible to control the load with the accuracy of mechanism occurring during the ECAP process can be described as
1/30,000 of the load cell nominal capacity of 20 kN, which was continuous dynamic recovery and recrystallization (CDRR), in which
70.66 N, and to record the extension with the resolution of a combination of mechanical shearing and subsequent dynamic
70.001 mm. Three different samples were tested for each condi- recovery, recrystallization and growth of grains and subgrains
tion and the measured data were averaged. WH rates were produce refined and equiaxed grains [21].
obtained by differentiating the true stress–true strain data. This The micrographs showing the microstructure of the LZ121 alloy
was accomplished by using a computer program, in which the at different conditions are illustrated in Fig. 3a–c. The as-cast
derivatives were obtained by a five-point cubic spline numerical microstructure of LZ121 in Fig. 3a is indicative of coarse single phase
differentiation method. β grains. In the authors' previous work [6], the X-ray diffraction
514 M. Karami, R. Mahmudi / Materials Science & Engineering A 607 (2014) 512–520

50 μm 100 μm

β-Li

α-Mg

50 μm 50 μm

50 μm
50 μm
Fig. 3. Micrographs of the LZ121 alloy for (a) as-cast, (b) extruded, and (b) ECAPed
Fig. 2. Micrographs of the LZ61 alloy for (a) as-cast, (b) extruded, and (c) ECAPed conditions with grain sizes of 1200, 30.3 and 6.1 mm, respectively.
conditions with grain sizes of 82, 9.6 and 6.3 mm, respectively.

analysis of the investigated alloys in the as-cast condition showed the LZ121 alloy by decreasing the average grain size from 30.3 to
that with increasing the Li content from 6% to 12%, the α-phase 6.1 mm. It is also evident that, after ECAP, a reasonably homogeneous
peaks nearly vanish and that of the β-Mg phase increases, so that the fine-grained microstructure is obtained and the coarse-grained
XRD pattern of LZ121 mainly consists of the β-phase peaks. As can be structure in the extruded condition entirely vanishes. As already
seen in Fig. 3b, the inhomogeneous microstructure of the extruded pointed out, the microstructural evolution during the ECAP process
condition is formed by a small fraction of large grains (115 mm) could be the result of CDRR mechanism, in which the gradual
embedded in a matrix of relatively finer grains (21 mm). It seems that transformation of low angle subgrain boundaries to high angle grain
discontinuous dynamic recrystallization has completely occurred in boundaries occurs by absorbing dislocations generated during suc-
the microstructure due to the high extrusion temperature ( 0.66 cessive passes of ECAP [23].
Tm). The formation of some excessively coarse grains surrounded
by dynamically recrystallized finer grains can be ascribed to the 3.2. Textural evolution
abnormal grain growth. This process, which is originated from the
preferential growth of a few grains having some special growth To examine the textural evolution of the material during the
advantage over their neighbors, is usually promoted in the strongly extrusion and ECAP processes, the (0002) and (1010) pole figures
textured structures caused by rolling and extrusion [22]. Fig. 3c of the extruded and ECAPed LZ61 are shown in Fig. 4. It is evident
clearly shows the role of the ECAP process in the grain refinement of that the texture developed after extrusion, shown in Fig. 4a, has a
 M. Karami, R. Mahmudi / Materials Science & Engineering A 607 (2014) 512–520 515

(0002) (0002)

ND
Max = 3.1 Max = 3.5
TD
ED

(1010) (1010)

Max = 24.9 Max = 5.9

Fig. 4. (0002) and (1010) pole figures of the LZ61 alloy after (a) extrusion and (b) ECAP.

fiber-type character with the basal planes being parallel to the 137 MPa, though the corresponding grain size decreased from
extrusion direction. In this state, the basal planes tend to align 9.6 to 6.3 mm. A ductility enhancement of about 48% can also be
along the 〈1010〉 poles parallel to the extrusion axis, which is observed, where the elongation increases from 28.4% in the
verified by the ð1010Þ pole figure observed in Fig. 4a. After ECAP, extruded condition to 42% in the ECAPed material. Such a decrease
however, most of basal planes depart from the extrusion direction, in strength together with the ductility enhancement in the ECAPed
being located at about 451 to the extrusion axis, mainly due to the condition is attributed to the texture modifications occurring in
shearing parallel to the basal planes. Similar orientation of basal the hcp crystal structure of Mg during ECAP. Primary slip occurs on
planes with respect to the pressing axis has been observed in the (0001) basal planes in the Mg alloys at room temperature due
different ECAPed Mg alloys [24,25]. to their low critical resolved shear stress (CRSS), as compared to
Fig. 5 shows the (110) and (211) pole figures of the LZ121 alloy after those for non-basal slip systems [27]. For the extruded material,
extrusion and ECAP. In the extruded condition (Fig. 5a), the aggrega- the basal planes tend to lie parallel to the extrusion axis (Fig. 4a),
tion of the 〈110〉 poles at the center of pole figure, forming a strong implying that primary slip would be difficult, and therefore, the
texture after extrusion, indicates that the (110) planes are mostly yield stress increases. The rotation of basal planes ( 451) during
located perpendicular to the extrusion axis. As discussed earlier, the ECAP process (Fig. 4b), however, enhances the Schmid factor
development of such a highly textured structure after extrusion could for the (0001) basal planes, leading to a decrease in stress required
lead to the abnormal grain growth. It is also evident that the 〈211〉 for yielding of the ECAPed materials. The texture modification
poles are mainly inclined at about 251 to the extrusion direction, along with the grain refinement can also be the main reasons for
which develops a fiber-type texture after extrusion. Processing by the observed ductility enhancement [3]. However, the basal slip
ECAP encourages the formation of a new weakened texture with a due to the texture modification cannot individually provide the
characteristic three-point intensity and a randomized distribution in necessary five independent slip systems for homogeneous defor-
the (110) and (211) pole figures, respectively (Fig. 5b). Similar trend in mation, according to the Von-Mises criterion. Accordingly, to
the orientation of (110) poles after ECAP has been observed in the bcc- sustain the large increase of tensile ductility, some non-basal slip
structured IF-steel [26]. planes have been activated due to a rotation of about 451 from the
extrusion direction by ECAP. This argument is in agreement with
3.3. Tensile flow behavior the view of Agnew et al. in solid solution Mg–Li alloys [28].
Fig. 6b shows a Kocks–Mecking type plot of tensile WH rate
The tensile flow behavior of the extruded and ECAPed LZ61 (θt ¼ds/dε) against net flow stress (s  sy) for the LZ61 alloy in the
samples is plotted in terms of true stress–strain curves in Fig. 6a. extruded and ECAPed conditions. A nearly linear initial stage of
Concerning the yield stresses, the extruded material had a yield WH is evident for both conditions, which may be related to the
stress of 144 MPa and the ECAPed material had a yield stress of elastic–plastic transition, rather than dislocation mediated WH, as
516 M. Karami, R. Mahmudi / Materials Science & Engineering A 607 (2014) 512–520

(110) (110)

ND
Max = 50.3 Max = 20.3

TD
ED

(211) (211)

Max = 12.3 Max = 5.1

Fig. 5. (110) and (211) pole figures of the LZ121 alloy after (a) extrusion and (b) ECAP.

400 proposed by Jain et al. [29]. No stage I hardening or ‘easy glide’,

Tensile which depends strongly on the orientation of the crystal, can be
observed in any of the tested conditions. The extruded condition
300 1
exhibits an almost constant hardening behavior (θII ¼ ds/dε ¼
True stress (MPa)

2
const.) associated with stage II of WH, with θΙΙ E880 MPa. The
200 chief hardening mechanism has been related to the evolution of
long-range stresses caused by dislocation pile ups at the grain
boundaries [30]. It is has also been suggested that the presence of
100 stage II of WH might be due to the interactions of the dislocations
1: Extruded in the primary slip system with those in an intersecting slip
2: ECAPed
system [31]. Further increasing of stress is accompanied by a
0
0 0.2 0.4 0.6 non-linear behavior that conforms to stage III of the Kocks–
True strain Mecking plot [32]. The WH plot of the ECAPed material, however,
shows no stage II of WH, and the behavior is very similar to the
well-known linear stage III of WH in fcc polycrystals. According to
1500
Tensile Rollett and Kocks [33], stage III of WH is characterized by a
hardening rate which decreases monotonically with increasing
flow stress arising from the parabolic hardening on the stress–
1000 Stage II strain curve.
θt (MPa)

II It is worth describing the stress–strain curves and WH of the

III
0 1
1: Extruded
alloys by mathematical expressions as a widely used approach.
2: ECAPed Among many empirical equations, Voce equation [34], which is an
500
2 alternative exponential stress–strain law compared to the com-
Stage III mon power law relation, has the advantage of showing an
asymptotic saturation stress. Therefore, the WH rate in stage III
0 can be well described by a differential form of the Voce equation in
0 50 100 150 200
terms of net flow stress:
σ-σy (MPa)


s  sy
Fig. 6. (a) True stress–strain curves, and (b) corresponding Kocks–Mecking plot of θt ¼ θΙΙΙ 1 ð2Þ
the extruded and ECAPed LZ61 alloy.
0
ss
 M. Karami, R. Mahmudi / Materials Science & Engineering A 607 (2014) 512–520 517

where θΙΙΙ0 is the WH limit extrapolated to s ¼ sy, and ss is the Thus, if a relatively large stress concentration takes place in some
saturation stress extrapolated to θt ¼0. The respective values of θΙΙΙ
0 areas during room temperature testing, the deformation can be
and ss obtained in this study are about 780 and 220 MPa accommodated by the β phase [38]. This results in the lower
respectively. The observed differences in WH behavior of the strength and higher ductility of the LZ121 alloy as compared to
extruded and ECAPed conditions can be attributed to the grain those of LZ61. A summary of the tensile and shear strength values
size effects. A significant drop in WH rate of the ECAPed condition, of the investigated alloys in the extruded and ECAPed conditions is
compared to the extruded material, might be due to the grain given in Table 1.
refinement of about 34%, which leads to a higher contribution of The Kocks–Mecking plots of the tensile WH rate against net flow
grain boundary sliding (fGBS) to the total deformation even at room stress for the LZ121 alloy in the extruded and ECAPed conditions are
temperature. This is because WH rate is proportional to the factor depicted in Fig. 7b. In both conditions, a linear elastic–plastic
(1  fGBS) [35]. The refined grain structure, in the ECAPed sample, is transition takes place at the beginning of plastic deformation, which
also responsible for the suppression of stage II and development of moves toward a linear stage III as the deformation proceeds. Similar
a linear stage III at the beginning of deformation via enforcement to the tensile WH behavior of LZ61 (Fig. 6b), the observed lower WH
of multiple slips. Considering the cross-slip as a significant rates of the ECAPed condition in comparison with the extruded
recovery process in stage III of WH, it should be mentioned that sample is due to the finer grain sizes achieved after ECAP. Since Li
the activity of non-basal 〈a〉 slip through the cross-slip of 〈a〉 increasing tends to increase stacking-fault energy (SFE) in the Mg
dislocations from basal to non-basal planes could be enhanced in alloys [39], it is not surprising that only stage III of WH is discernible
more refined grain structures during room-temperature deforma- in the tensile WH plots of LZ121. It is reported that for materials
tion. The grain-size dependent compatibility stress at grain with high SFE, stage III may limit the extent of stage II and even
boundaries is considered to be the main reason for this enhance- eliminate it as a separate stage [32]. For the LZ121 alloy, the values
ment [36]. 0 and ss in the modeled equation (Eq. (2)) are summarized in
of θΙΙΙ
To characterize the influence of crystal structure on the Table 2. It can be inferred that the above-mentioned effects are
mechanical properties, tensile flow behavior of the extruded and reflected in lower values of tensile saturation stress (ss), decreasing
ECAPed bcc-structured LZ121 alloys is plotted in terms of true from 185 to 65 MPa for extruded alloys, and from 220 to 41 MPa for
stress–strain curves in Fig. 7a. It can be observed that the yield the ECAPed materials.
stress increases from 84 MPa in the extruded condition to 111 MPa Tensile WH capacity (H Tc ) of a material can be defined in terms
after ECAP. This is simply a consequence of the significant grain of ultimate tensile strength (UTS) and yield stress (YS) as [11]
refinement from 30.3 to 6.1 mm that occurs after ECAP. It is
UTS  YS UTS
believed that, as the Li content of the Mg–Li alloy increases, the H Tc ¼ ¼ 1 ð3Þ
YS YS
crystal lattice axes ratio (c/a) of the hcp α phase decreases so that
slip between crystal planes become easier [37]. For Li contents Table 1 depicts the tensile WH capacity of the LZ61 and LZ121
greater than 11 wt%, a ductile bcc β-phase structure develops, alloys in both extruded and ECAPed conditions. It is evident that
which has more independent slip systems than the hcp α phase. processing by ECAP leads to a decrease in the tensile WH capacity
of both alloys. H Tc drops from 0.41 to 0.37 for LZ61 and from 0.34 to
0.07 for LZ121. It is believed that the WH capacity of an alloy is
associated with its yield stress, which is further related to grain
200 size in accordance with the Hall–Petch relationship [30,40]. Grain
Tensile refinement would increase YS and decrease WH capacity. It also
2 reduces the difference between flow resistance of the grain
150
True stress (MPa)

boundary and grain interior, which decreases WH capacity [41].

1 Therefore, the grain refinement caused by ECAP can be the main
100 reason for decreasing WH capacity in both alloys. A comparison of
tensile WH capacity of the LZ61 and LZ121 alloys at the same
processing condition shows that it decreases with increasing Li
50 content. This might be due to the fact that with increasing the
1: Extruded amount of Li, the CRSS for non-basal slip diminishes in the hcp α
2: ECAPed
phase [42], activating the non-basal slip and facilitating the cross-
0
0 0.2 0.4 0.6 slip between crystal planes, and thereby, decreasing WH rate [43].
True strain Besides, the presence of the softer bcc-structured β phase with
more independent slip systems, as compared to the hcp α phase,
could contribute to the lowering of WH rate. Accordingly, the WH
1500
Tensile capacities in the LZ121 alloy for both extruded and ECAPed
III conditions have lower values, compared to the LZ61 alloy.
0

1000 3.4. Shear flow behavior

θt (MPa)

Stage III
To study the shear flow behavior of the LZ61 alloy, the shear
1
500 stress (τ) was plotted against shear strain (γ) for both extruded and
2 ECAPed conditions, as illustrated in Fig. 8a. It is evident that,
1: Extruded analogous to the tensile stress–strain curves, after a linear elastic
2: ECAPed
behavior the SPT curves deviate from linearity before they reach a
0 maximum stress. The deviation point, obtained by plotting a
0 25 50 75 100
tangent to the elastic part of the curve, is taken as the shear yield
σ-σy (MPa) stress (SYS or τy) and the stress corresponding to the maximum
Fig. 7. (a) True stress–strain curves, and (b) corresponding Kocks–Mecking plot of point is referred to as the ultimate shear strength (USS). It can be
the extruded and ECAPed LZ121 alloys. seen in Table 1 that, in contrast to the tensile deformation, the SPT
518 M. Karami, R. Mahmudi / Materials Science & Engineering A 607 (2014) 512–520

Table 1
Tensile and shear strengths (MPa) and WH capacity for the tensile (HTc ) and shear (H SPT
c ) deformations.

Materials Extruded ECAPed

YS UTS SYS USS HTc H SPT

c
YS UTS SYS USS H Tc H SPT
c

LZ61 1447 2.5 203 7 3.6 947 1.7 1227 2.2 0.41 70.05 0.30 7 0.04 1377 2.5 188 7 3.6 1157 2.1 1457 2.6 0.377 0.05 0.26 70.05
LZ121 847 1.4 1137 2.1 797 1.5 1077 1.9 0.34 70.05 0.357 0.06 1117 1.9 1197 2.1 977 1.7 1227 2.2 0.077 0.04 0.25 70.06

Table 2
ΙΙΙ
Extrapolated WH limit of stage III (MPa) for the tensile (θIII
0 ) and shear (Θ0 ) deformations, and corresponding saturation stress (MPa) for both materials.

Materials Extruded ECAPed

θIII
0 ss ΘΙΙΙ
0 τs θIII
0 ss ΘΙΙΙ
0 τs

LZ61 11507 19 1857 3.2 357 0.6 527 0.9 780 711 220 7 3.8 407 0.6 65 7 1.1
LZ121 1090 7 18 657 1.1 337 0.5 407 0.8 900 716 417 0.6 307 0.4 30 7 0.5

Fig. 8b exhibits the shear WH rate (θs ¼dτ/dγ) plotted against

160
SPT shear net flow stress (τ τy) for the LZ61 alloy in the extruded and
1 ECAPed conditions. It can be seen that the shear WH curves in
120 both conditions show the same patterns as those obtained in
Shear stress (MPa)

2 tensile deformation (Fig. 6b). However, contrary to the tensile WH

curves, the WH rate for the ECAPed sample is greater than that of
80 the extruded sample. It seems that unfavorable orientation of
basal planes respect to the shearing direction is responsible for the
observed increase in the WH rate of the ECAPed material, the
40
1: ECAPed
effect which offsets the influence of grain refinement. Similar to
2: Extruded the tensile WH behavior, the shear stress dependence of the shear
0 WH rate in stage III can be well modeled in the following form:
0 2 4 6

τ  τy
Shear strain θs ¼ ΘΙΙΙ
0 1  ð4Þ
τs
50
where ΘΙΙΙ0 is the WH limit extrapolated to τ ¼τy and τs is the
SPT
III saturation stress extrapolated to θs ¼0. The respective values of ΘΙΙΙ
40 Θ 0
0
and τs are about 40 and 65 MPa respectively, as tabulated in
Stage III
Table 2. It is to be noted that the low values of θs observed in
30 Θ II 1
θs (MPa)

Fig. 8b are consistent with those of Les et al. [44], who investigated
the shear WH rates of Al alloy using simple shear test.
2
20 The SPT curves of the LZ121 alloy for the extruded and ECAPed
Stage II
conditions are presented in Fig. 9a. It is clear that processing by
10 ECAP leads to higher values of SYS and USS as well as shear
1: ECAPed
ductility. The texture-dependence of mechanical properties in the
2: Extruded
0 bcc-structured β phase is less pronounced compared to a hex-
0 10 20 30 40 50 agonally close-packed α phase, because of the higher lattice
τ-τy (MPa) symmetry and more activated slip systems in cubic structures.
Therefore, the significant grain refinement achieved after ECAP
Fig. 8. (a) Shear stress–strain curves, and (b) corresponding curves of shear WH could be the main reason for the observed strength and ductility
rate vs. shear net flow stress for the extruded and ECAPed LZ61 alloy.
enhancements of LZ121.
The Kocks–Mecking plot of shear WH rate against net flow
stress for the LZ121 alloy in the extruded and ECAPed conditions is
curve in ECAPed condition has higher values of SYS and USS (115 shown in Fig. 9b. As expected, similar patterns to tensile deforma-
and 145 MPa), as compared to the extruded condition (94 and tion have been achieved in SPT of LZ121 in both conditions, as
122 MPa). This can be ascribed to the finer grain sizes achieved shown in Fig. 7b. This confirms that the SPT is a promising
after ECAP. Moreover, textural strengthening can partly enhance technique for evaluating WH behavior using small test pieces
strength levels in shear deformation, the effect which was acting usually produced by severe plastic deformation processes such as
in opposite way in tensile deformation. In the extruded material, ECAP. Considering the linear stage III of WH in the both conditions,
the basal planes are mostly aligned in the extrusion direction, and the values of ΘΙΙΙ
0 and τs in the modified Eq. (4) are given in Table 2.
thus, the shearing during SPT most likely occurred on the basal Tensile WH capacity equation (Eq. (3)) can be simply modified
planes under a lower yielding stress. In the case of ECAPed mate- for evaluating shear WH capacity (H SPT c ) in the SPT method by
rial, however, a decreased value of shear stress on the basal planes replacing UTS with USS and YS with SYS as follows:
caused by the rotation of basal planes (  451 respect to the
shearing direction) makes deformation more difficult, and thereby, USS  SYS USS
H SPT
c ¼ ¼ 1 ð5Þ
the required stress for deformation increases. SYS SYS
 M. Karami, R. Mahmudi / Materials Science & Engineering A 607 (2014) 512–520 519

deformations, as compared to those of the extruded condition.

160
SPT
Considering a weaker texture-dependence of mechanical prop-
erties in the bcc-structured LZ121 alloy, the significant grain
2
refinement achieved after ECAP could be the main reason for
Shear stress (MPa)
120
the observed strength and ductility enhancements.
1
3. The study of work hardening showed that both tensile and
80 shear WH rates of the hcp LZ61 correspond well to stages II and
III of classical Kocks–Mecking plot in the extruded condition.
However, the work hardening plot of the ECAPed samples
40
exhibits only stage III work hardening, due to the refined
1: Extruded
2: ECAPed microstructure caused by severe plastic deformation. For the
0 bcc LZ121, no stage II was observed in both extruded and
0 2 4 6 8 10 ECAPed conditions. Work hardening capacities of both alloys
Shear strain decreased after ECAP, mainly due to the higher density of grain
boundaries provided by smaller grains.
4. It is demonstrated that shear punch testing (SPT) is capable of
50
SPT evaluating flow and WH behavior of the fine-grained Mg–Li–Zn
III
alloys by measuring the WH rates. This can be particularly
40 Θ 0 advantageous when the material is only available as small test
pieces such as those usually obtained by severe plastic defor-
30
θs (MPa)

Stage III mation processes. The validity of the SPT method was con-
1 firmed by obtaining the flow and WH curves similar to those
20 obtained in tensile testing.
2 5. The relationships between different microstructures, textures,
10 and deformation conditions are based on the initial micro-
Stage II 1: Extruded
2: ECAPed structures and textures of the materials. Although the micro-
0 structural changes seem to be negligible, the textural evolution
0 10 20 30 40 50 of the samples during both tests is expected to affect the WH
τ-τy (MPa) behaviors. Unfortunately, textural studies of the shear-punched
Fig. 9. (a) Shear stress–strain curves, and (b) corresponding curves of shear WH specimens in their very small shear zones are not feasible, and
rate vs. shear net flow stress for the extruded and ECAPed LZ121 alloys. thus, it is not possible to elucidate textural evolution during
SPT, to be compared with that in the tensile tests. Moreover, the
tensile and SPT samples were selected from different positions
This is based on the Von-Mises criterion for a state of pure in the processed materials and this might have an effect on
shear of kinematically hardening materials, in which theprelation-
ffiffiffi some of the different trends observed in the tensile and SPT
ship between tensile and shear stresses obeys s ¼ 3τ [30]. measurements.
Employing this approach, the shear WH capacity of the alloys in
both extruded and ECAPed samples is calculated and tabulated in
Table 1. It is observed that, similar to tensile WH capacity, the
grain refinement imposed by ECAP leads to the lowering of shear Acknowledgments
WH capacities in both LZ61 and LZ121 alloys, the effect which is
more pronounced in the LZ121 alloy. It seems that the formation of The authors thank the Iran National Science Foundation (INSF)
abnormally grown large grains surrounded by dynamically recrys- for support of this work under Grant no. 91041956.
tallized finer grains achieved after extrusion in LZ121 (Fig. 3b) is
responsible for the significant increase in WH capacity of extruded References
samples. The larger grain sizes provide more space to accommo-
date dislocations and result in the increased dislocation storage [1] I.J. Polmear, Light Alloys, 2nd ed., Chapman and Hall, Inc., New York, NY (1989)
capacity, leading to enhanced WH capacities. 170–188.
[2] B.L. Mordike, T. Ebert, Mater. Sci. Eng. A 302 (2001) 37–45.
[3] S.M. Masoudpanah, R. Mahmudi, Mater. Sci. Eng. A 526 (2009) 22–30.
[4] M. Karami, R. Mahmudi, Mater. Sci. Eng. A 576 (2013) 156–159.
4. Summary and conclusions [5] Y.N. Lin, H.Y. Wu, G.Z. Zhou, C.H. Chiu, S. Lee, Mater. Des. 29 (2008) 2061–2065.
[6] M. Karami, R. Mahmudi, Metall. Mater. Trans. A 44A (2013) 3934–3946.
[7] R.Z. Valiev, T.G. Langdon, Prog. Mater. Sci. 51 (2006) 881–981.
The flow and work hardening behavior of the extruded and [8] A.A. Nayeb-Hashemi, J.B. Clark, A.D. Pelton, Phases Diagrams of Binary
ECAPed Mg–6Li–1Zn (LZ61) and Mg–12Li–1Zn (LZ121) alloys were Magnesium Alloys, ASM International, Materials Park, OH (1998) 184.
[9] J.M. Song, Y.H. Lin, C.W. Su, J.Y. Wang, Metall. Mater. Trans. A 40A (2009)
studied and the following conclusions are made: 1026–1030.
[10] X.H. Chen, L. Lu, Scr. Mater. 57 (2007) 133–136.
1. The LZ61 alloy with an hcp lattice structure, showed higher [11] N. Afrin, D.L. Chen, X. Cao, M. Jahazi, Scr. Mater. 57 (2007) 1004–1007.
[12] G.H. Hasani, R. Mahmudi, Mater. Des. 32 (2011) 3736–3741.
tensile strength and lower ductility in the extruded condition, [13] M. Habibnejad-Korayem, R. Mahmudi, W.J. Poole, Mater. Sci. Eng. A 567 (2013)
as compared to the ECAP condition. After ECAP, however, this 89–94.
trend was reversed mainly due to textural modification. The [14] H.Y. Wu, J.C. Yan, H.H. Tsai, C.H. Chiu, G.Z. Zhou, C.F. Lin, Mater. Sci. Eng. A 527
(2010) 7197–7203.
observed textural effect acts in an opposite way in shear
[15] H.Y. Wu, G.Z. Zhou, J. Mater. Sci. 44 (2009) 6182–6186.
deformation, due to the shearing nature of the SPT method, [16] R.K. Guduru, K.A. Darling, R. Kishore, R.O. Scattergood, C.C. Koch, K.L. Murty,
leading to the higher values of shear strength in ECAPed Mater. Sci. Eng. A 395 (2005) 307–314.
condition. [17] R. Alizadeh, R. Mahmudi, Mater. Sci. Eng. A 527 (2010) 3975–3983.
[18] G. Nayyeri, R. Mahmudi, F. Salehi, Mater. Sci. Eng. A 527 (2010) 5353–5359.
2. For the LZ121 alloy, it was observed that processing by ECAP [19] M. Karami, R. Mahmudi, Mater. Lett. 81 (2012) 235–238.
increases the strength and ductility in both tensile and shear [20] M. Karami, R. Mahmudi, Mater. Des. 53 (2014) 534–539.
520 M. Karami, R. Mahmudi / Materials Science & Engineering A 607 (2014) 512–520

[21] C.W. Su, L. Lu, M.O. Lai, Mater. Sci. Eng. A 434 (2006) 227–236. [33] A.D. Rollett, U.F. Kocks, Solid State Phenom.35/36 (1994) 1–18.
[22] F.J. Humphreys, M. Hatherly, Recrystallization and Related Annealing Phe- [34] E. Voce, J. Inst. Metals 74 (1948) 537–562.
nomena, 2nd ed., Elsevier, Oxford, 1996. [35] J.A. del Valle, F. Carreno, O.A. Ruano, Acta Mater. 54 (2006) 4247–4259.
[23] H.K. Kim, W.J. Kim, Mater. Sci. Eng. A 385 (2004) 300–308. [36] J. Koike, T. Kobayashi, T. Mukai, H. Watanabe, M. Suzuki, K. Maruyama,
[24] S.R. Agnew, J.A. Horton, T.M. Lillo, D.W. Brown, Scr. Mater. 50 (2004) 377–381. K. Higashi, Acta Mater. 51 (2003) 2055–2065.
[25] T. Liu, Y.D. Wang, S.D. Wu, R.L. Peng, C.X. Huang, C.B. Jiang, S.X. Li, Scr. Mater. [37] F.H. Herbstein, B.L. Averbach, Acta Metall. 4 (1956) 407–413.
51 (2004) 1057–1061. [38] T. Liu, W. Zhang, S.D. Wu, C.B. Jiang, S.X. Li, Y.B. Xu, Mater. Sci. Eng. A 360
[26] A.A. Gazder, F.D. Torre, C.F. Gu, C.H.J. Davies, E.V. Pereloma, Mater. Sci. Eng. A (2003) 345–349.
415 (2006) 126–139. [39] J. Han, X.M. Su, Z.H. Jin, Y.T. Zhu, Scr. Mater. 64 (2011) 693–696.
[27] T. Obara, H. Yoshinga, S. Morozumi, Acta Metall. 21 (1973) 845–853. [40] W.F. Hosford, Mechanical Behavior of Materials, Cambridge University Press,
[28] S.R. Agnew, J.A. Horton, M.H. Yoo, Metall. Mater. Trans. A 33A (2002) 851–858. New York, 2005.
[29] J. Jain, W.J. Poole, C.W. Sinclair, Mater. Sci. Eng. A 547 (2012) 128–137. [41] J. Luo, Z. Mei, W. Tian, Z. Wang, Mater. Sci. Eng. A 441 (2006) 282–290.
[30] G.E. Dieter, Mechanical Metallurgy, third ed., McGraw-Hill, Boston, MA, 1986. [42] H. Yoshinaga, R. Horiuchi, Trans. JIM 4 (1963) 134–141.
[31] M. Srinivasan, T.G. Stoebe, J. Mater. Sci. 9 (1974) 121–128. [43] Z. Drozd, Z. Trojanová, S. Kúdela, J. Alloys Compd. 378 (2004) 192–195.
[32] U.F. Kocks, H. Mecking, Prog. Mater. Sci. 48 (2003) 171–273. [44] P. Les, M. Zehetbauer, E.F. Rauch, I. Kopacz, Scr. Mater. 41 (1999) 523–528.