Metals and Materials International

, Volume 24, Issue 2, pp 283–289 | Cite as

Effects of {10–12} Twins on Dynamic Torsional Properties of Extruded AZ31 Magnesium Alloy

  • Jong Un Lee
  • Seok Weon Song
  • Yongjin Kim
  • Sang-Hoon Kim
  • Ye Jin Kim
  • Sung Hyuk Park


Effects of initial twins on dynamic torsional properties of extruded AZ31 alloy were investigated by introducing {10–12} twins into it through precompression to 3 and 6% strains along the extrusion direction and performing torsional testing at a strain rate of 1.4 × 103 s−1 using a torsional Kolsky bar system. The as-extruded sample without twins showed higher dynamic torsional properties than the precompressed samples with many initial twins; the maximum shear strength and fracture shear strain decreased with increasing amount of initial twins. In the as-extruded sample, twinning occurred vigorously throughout the gage section of the tubular specimen during high-strain-rate torsional tests, resulting in heavily deformed morphology, many macrocracks, and rough fractured surfaces. The increased amount of initial twins suppressed the twinning behavior and localized the applied torsional deformation; this resulted in an almost unchanged sample shape, no secondary cracks, and a flat fracture plane, thereby deteriorating the dynamic torsional properties of the extruded alloy.


Metal Extrusion Twinning Optical microscopy Dynamic torsion 

1 Introduction

In view of the increasing demand for weight reduction of automobiles due to the ever-tightening regulations on fuel efficiency and carbon dioxide emissions, Mg alloys—which are the lightest among commercially available metal materials—have attracted great attention in the transportation industry. Wrought Mg alloys subjected to hot working processes such as rolling, extrusion, and forging generally have a basal texture in which the basal plane of the hexagonal close-packed (HCP) lattice aligns perpendicular to the direction of compression applied during the process [1, 2, 3]. In Mg, only basal slip—which has merely two independent slip systems—is activated during deformation at room temperature; this is because non-basal slip systems such as prismatic and pyramidal slips have a high critical resolved shear stress [4, 5]. Because of the lack of active slip systems in Mg alloys, twinning occurs easily during plastic deformation, especially during room-temperature deformation [6, 7]. Among the various types of twins activated in Mg, {10–12}<10–11> twins form most frequently and play a crucial role in the plastic deformation behavior of Mg [8, 9, 10, 11, 12]. {10–12} Twinning is activated under loading conditions wherein the c-axis of the HCP unit cell is extended, i.e., under tension parallel to the c-axis or under compression perpendicular to the c-axis [13].

When {10–12} twinning is induced under favorable loading modes, the applied strain can be easily accommodated by the formation and propagation of twin bands; this results in a low yield strength and low strain-hardening rate of the material [14, 15, 16]. In addition, Hall–Petch hardening can be induced by the formation of twin boundaries that act as barriers to dislocation motion, and dislocation slips can be promoted or suppressed by lattice reorientation in the twinned region [17, 18, 19, 20]. In addition to the {10–12} twins formed during deformation, the initial {10–12} twins formed in wrought Mg alloys by predeformation also have a significant influence on the subsequent deformation behaviors and mechanical properties. For instance, initial {10–12} twins can improve the rollability [21], strength [22, 23], stretch formability [24], and fatigue resistance [25, 26] of rolled Mg alloys. They can also cause in-plane anisotropy in the monotonic and cyclic deformation behaviors of rolled Mg alloys [27, 28] and promote dynamic recrystallization behavior of extruded and rolled Mg alloys [29, 30]. Although the effects of initial twins on the quasi-static deformation behaviors and properties of wrought Mg alloys have been actively studied, their influence on these alloys’ dynamic properties under high-strain-rate deformation has not yet been investigated. Some studies investigated the compressive properties of rolled and extruded Mg alloys at high strain rates of above 1.6 × 103 s−1 [31, 32]; however, they focused on the effects of texture or sample direction on the dynamic deformation behavior. In addition, many automotive components such as drive shafts, crankshafts, and steering columns are subjected to high-speed torsional forces during their operation. In order to simulate real-life service conditions, verify product quality, and ensure use of appropriate manufacturing techniques, the torsional properties of such components should be suitably evaluated before use. Studies have been conducted on the torsional deformation behavior of extruded Mg alloys at quasi-static strain rates [33, 34, 35], but research on their torsional properties at high strain rates of above 103 s−1 is lacking. Therefore, in the present study, {10–12} twins are introduced into an extruded AZ31 alloy having a typical basal texture through precompression along the extrusion direction (ED), and the effects of initial twins on the dynamic torsional behavior of the alloy are investigated via high-strain-rate torsional testing of samples with and without {10–12} twins by using a torsional Kolsky bar system.

2 Experimental Procedure

A commercial Mg-3.0 wt% Al-1.0 wt% Zn (AZ31) alloy was used in this study, and a cast billet was prepared according to a previously reported procedure [29]. The as-cast billet was homogenized at 400 °C for 24 h and then water-quenched. After the billet was preheated to 350 °C for 1 h, it was directly extruded at a temperature of 350 °C and ram speed of 1 mm/s, with an extrusion ratio of 15. In order to introduce {10–12} twins into the extruded bar, cylindrical samples (Ø20 mm × 30 mm) were machined from the extruded bar. {10–12} Twinning is activated under loading conditions in which the c-axis is extended [8, 22, 23, 31]. Since the c-axes of most grains in extruded Mg alloys are generally aligned perpendicular to the ED, {10–12} twins form easily when compressive loading is applied along the ED (i.e., compression perpendicular to the c-axis). Accordingly, in this study, the machined cylindrical samples were compressed to strains of 3 and 6% in order to induce formation of different amounts of {10–12} twins; these samples are hereafter denoted as 3PC and 6PC, respectively. The axes of these samples corresponded to the ED.

To identify variations in the microstructure and texture with precompression, cross-sectional areas of the as-extruded and precompressed samples were analyzed using an optical microscope (OM) and an X-ray diffraction (XRD) spectrometer. For the dynamic torsional tests, thin-walled tubular specimens with a gage length of 2.5 mm and gage thickness of 280 μm were machined from the as-extruded and precompressed samples (Fig. 1a). The torsional Kolsky bar was composed of a pair of 2024-T6 aluminum bars 2 m in length and 25.4 mm in diameter (Fig. 1b). In the dynamic torsional test, a considerable amount of torque is stored between a dynamic loading pulley and a clamp. When the clamp fractures, the elastic shear wave is momentarily transmitted into the specimen; this leads to dynamic deformation and fracture of the specimen. The measured shear strain rate of the dynamic torsional tests performed at room temperature in this study was 1.4 × 103 s−1. The principle of the torsional Kolsky bar system and a detailed description of the dynamic torsional tests can be found elsewhere [36, 37, 38, 39, 40, 41].
Fig. 1

a Shape and dimensions of thin-walled tubular specimen used for dynamic torsional test and b schematic illustration of torsional Kolsky bar [36]

3 Results and Discussion

Figure 2 shows the compressive stress–strain curves obtained from the precompression tests. Compressive yielding occurs at 287 MPa, which is followed by nearly constant stress up to 0.9%; thereafter, the strain-hardening rate gradually increases with increasing strain. Consequently, compressive curves with a sigmoidal shape are generated, which is a typical feature of {10–12}-twinning-dominant deformation. Optical micrographs and (0002) pole figures of the as-extruded and precompressed samples are shown in Fig. 3. The as-extruded sample has a twin-free equiaxed grain structure with an average grain size of 29.5 μm and a typical fiber basal texture, with the basal plane of most grains oriented almost parallel to the ED (Fig. 3a). The 3PC sample contains many twin bands with an area fraction of 21.3% (white region in the microstructure in Fig. 3b). Since the {10–12} twinning induces a crystallographic lattice reorientation of 86.3° [42], the c-axis of the formed {10–12} twin bands of the precompressed samples is aligned almost parallel to the ED (Fig. 3b, c). As shown in the pole figure in Fig. 3b, in the 3PC sample, the initial fiber basal texture weakens and a new twin texture wherein the c-axis is oriented along the ED is generated by the formation of twin bands. The 6PC sample has a large amount of twin bands (area fraction of 63.8%) owing to the growth of the formed twin bands and the additional formation of new twin bands (Fig. 3c). The initial fiber texture weakens to a pole intensity lower than 2.0, whereas the twin texture strengthens and its pole intensity increases from 3.0 for the 3PC sample to 5.0 for the 6PC sample. This indicates that precompression causes not only a microstructural change (i.e., formation of twin bands) but also an abrupt variation in texture.
Fig. 2

Compressive stress–strain curves during precompression along extrusion direction

Fig. 3

Optical micrographs and XRD (0002) pole figures of a as-extruded, b 3% precompressed, and c 6% precompressed samples. davg and ftwin denote the average grain size and the area fractions of twin bands formed by precompression, respectively

Figure 4 shows the shear stress–strain curves obtained from the dynamic torsional tests of the as-extruded and precompressed samples. The as-extruded sample exhibits the highest dynamic torsional properties, and the developed shear stress and shear strain decrease with increasing amount of precompression. The maximum shear stresses of the as-extruded, 3PC, and 6PC samples are 212 MPa, 186 MPa, and 163 MPa, respectively (Table 1); each time that 3% precompression is applied, the maximum shear stress decreases by ~ 12%. Moreover, the fracture shear strains of the as-extruded, 3PC, and 6PC samples are 42.8, 31.3, and 22.4%, respectively (Table 1); here, each time that 3% precompression is applied, the fracture shear strain decreases by ~ 28%. In general, when a metal material is predeformed, its strength increases and elongation decreases because of an increase in the dislocation density [43, 44]. Although precompression of the as-extruded sample along the ED causes {10–12}-twinning-dominant deformation, twinning alone cannot accommodate the imposed strain and dislocation slips occur together with twinning. The plastic strain accommodated by twinning (i.e., twinning strain) can be calculated by multiplying the volume fraction of twins, the average Schmid factor (SF), and the characteristic twinning shear [42]. When 3% precompression is applied along the ED, the plastic strain excluding elastic strain is 2.67% (Fig. 2). In the 3% precompressed sample, the twinning strain calculated by multiplication of the twin volume fraction (21.3%), average SF (0.437), and twinning shear (0.13) is 1.21%; information on the average SF value and {10–12} twinning shear in Mg alloys can be found elsewhere [17, 42]. Accordingly, only 45.3% of the imposed plastic strain is accommodated by {10–12} twinning, and the remaining plastic strain of 54.7% is caused by dislocation slip. Similar results related to the co-occurrence of twinning and slip have been reported for rolled AZ31 alloy subjected to compression along the rolling direction [17]. This means that twinning and slip occur simultaneously even under plastic deformation conditions, which are favorable for twinning. Therefore, the dislocation density decreases in the following order: 6PC sample > 3PC sample > as-extruded sample. However, under dynamic torsional deformation at a high strain rate of 1.4 × 103 s−1, the shear strength is inversely related to the dislocation density; that is, the shear strength is highest in the as-extruded sample and lowest in the 6PC sample (Fig. 4). In addition, it is known that twin bands reduce the effective grain size by dividing the crystal grains and that twin boundaries impede dislocation motion during deformation; these effects result in an increase in material strength [45]. Although the 3PC and 6PC samples have finer microstructures than the as-extruded sample owing to the presence of numerous twin bands, they exhibit a lower shear stress than the as-extruded sample. Namely, the sample containing more dislocations and twin boundaries—both of which inhibit the movement of mobile dislocations during deformation—has a lower strength than the sample containing fewer dislocations and no twins. This implies that dislocation slips are not the main deformation mechanism under dynamic torsional deformation.
Fig. 4

Shear stress–shear strain curves obtained from dynamic torsional test

Table 1

Dynamic torsional properties of as-extruded and compressed samples



3% precompressed

6% precompressed

Maximum shear stress (MPa)




Shear strain at maximum shear stress (%)




Fracture shear strain (%)




Images of the specimens fractured in the dynamic torsional tests, shown in Fig. 5, reveal that the fracture features of the as-extruded and precompressed samples are completely different. In the as-extruded sample, the gage section is severely distorted and the fracture plane is extremely rough. Moreover, many macrocracks are present in the gage section, which are almost perpendicular to the fracture line. On the other hand, in the precompressed samples, the fracture plane is parallel to the shear strain direction and no secondary cracks are present in the gage section. In addition, the 3PC sample includes some steps along the fracture line whereas the 6PC sample has a flat fracture plane, indicating that torsional deformation occurred more locally in the 6PC sample than in the 3PC sample. The as-extruded sample accommodates the imposed shear strain through overall deformation of the gage section, which results in a high fracture shear strain through effective resistance to the applied dynamic loading. In contrast, shear deformation in the precompressed samples is concentrated in the center of the gage section and the specimens fracture prematurely; such localized deformation and premature rupture are more pronounced in the 6PC sample, which has more twins, than in the 3PC sample.
Fig. 5

Images of fractured specimens after dynamic torsional tests of a as-extruded, b 3% precompressed, and c 6% precompressed samples

Figure 6 shows optical micrographs of the longitudinal and cross-sectional areas near the fractured surface of the failed specimens. Since the shear loading applied to a tubular specimen during torsional deformation does not produce tensile stress in the axial and radial directions [46], it is expected that deformation twinning will not be induced. However, Lou et al. [10] reported that pure shear could cause deformation twinning in extruded Mg alloys at large shear strains. Zhang et al. [46] also performed a cyclic torsional test at quasi-static strain rates by using thin-walled tubular specimens of extruded AZ61 alloy and reported the formation of a small amount of twins under cyclic torsion with large shear strain amplitudes. Furthermore, Song et al. [34] performed torsion tests at a strain rate of 0.9 × 10−2 s−1 by using dogbone-shaped rod specimens of extruded AZ91 alloy and reported that torsional deformation could generate some {10–12} twins and a large number of dislocations. It can be seen from the microstructure of the outer surface of the gage section (Fig. 6a) that many twins are formed by the dynamic torsional deformation of the as-extruded sample, which did not have any twins before the dynamic torsional test. Zhao et al. [47] and Marian et al. [48] demonstrated through molecular dynamics simulations and experimental results that at high strain rates, the spontaneous self-pinning of dislocations hinders their motion and deformation twinning occurs more actively. Indeed, Chen et al. [35] performed free-end torsion tests at strain rates ranging from 7 × 10−6 to 7 × 10−3 s−1 by using dogbone-shaped round specimens of extruded AZ31 alloy and reported that the nucleation of {10–12} twins during the torsion test accelerated as the strain rate increased. Therefore, under dynamic torsional deformation induced at a much higher strain rate of 1.4 × 103 s−1, dislocation slips will be suppressed to a greater extent and twinning will be a predominant deformation mechanism, as is evident from the formation of a considerable amount of twins in the failed specimen of the as-extruded sample (Fig. 6a). Furthermore, as shown in Fig. 6d, more twins are formed on the outer side of the tubular gage section than inside it, which can be attributed to the plastic strain in the outer region being larger than that in the inner region.
Fig. 6

ac Side-view and df top-view optical micrographs near fractured surface of dynamically torsioned specimens of a, d as-extruded, b, e 3% precompressed, and c, f 6% precompressed samples

Similar to the case of the as-extruded sample, numerous twins are observed in the longitudinal microstructures of the precompressed samples (Fig. 6b, c). However, unlike in the as-extruded sample, in the precompressed samples, the difference in twin fraction according to the thickness of the tubular specimen is extremely small, as can be seen in the cross-sectional microstructures (Fig. 6e, f). Accordingly, most of the twins observed in the fractured specimens of the precompressed samples are considered as twins that were initially formed by precompression. As the twin bands formed by precompression reduce the area of the untwinned matrix region, the amount of twins that can be formed during dynamic torsional deformation decreases with increasing amount of precompression. In addition, since the twinning stress increases rapidly with decreasing grain size [18, 49, 50], the reduction in the effective grain size caused by the initial twins will render the formation of new twins during the dynamic torsional test difficult. Consequently, in extruded Mg alloys without any twins, twinning occurs throughout the specimen during the dynamic torsional test and the imposed shear strain is easily accommodated by the formation of numerous twins; this consequently results in excellent dynamic torsional properties of the material. On the other hand, as the amount of initial twins increases, the twinning behavior is suppressed during the dynamic torsional test and the applied deformation is localized to a limited region of the specimen; this eventually leads to degradation of the dynamic torsional properties of the material. Deformation twins can form unintentionally during manufacturing processes for the production of final products, such as bending, forging, cutting, and shaping. Accordingly, when wrought Mg alloys are used as component materials that can undergo torsional deformation at high strain rates during service, considerable care must be taken to prevent the formation of twins during the manufacturing processes. In addition, the twinning behavior can vary with microstructural characteristics such as grain size, second phase, texture, and dislocation density; it can also vary with deformation parameters such as temperature, imposed strain, and strain rate [18, 49, 51, 52, 53, 54]. Therefore, further research is required to investigate the effects of initial twins on the dynamic torsional behavior of various Mg alloy systems and under various deformation conditions in order to understand the correlation among alloying elements, process variables, twinning behavior, and torsional properties.

4 Conclusion

The effects of initial twins on the dynamic torsional properties of extruded AZ31 alloy were investigated by introducing {10–12} twins into the alloy through precompression along the ED. The as-extruded sample without initial twins showed superior dynamic torsional properties to the precompressed samples with many twins, and the maximum shear strength and fracture shear strain decreased gradually as the amount of initial twins increased. The fractured surface of the as-extruded sample was extremely rough and many macrocracks were observed in the gage section of its specimen. On the other hand, the precompressed samples had a flat fracture plane and did not show any cracks around the fractured surface. In the as-extruded sample, shear deformation applied during the dynamic torsional test was accommodated by twinning throughout the gage section, which resulted in heavily deformed specimen morphology and excellent torsional properties. An increase in the amount of initial twins caused suppression of the twinning behavior and localization of the applied deformation; this resulted in an almost unchanged tubular shape of the failed specimens and caused gradual degradation of the dynamic torsional properties of the extruded alloy.



This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIP, South Korea) (No. 2016R1C1B2012140 and No. 2017R1A4A1015628).


  1. 1.
    G.S. Rao, Y.V.R.K. Prasad, Mater. Sci. Eng. A 13, 2219 (1982)Google Scholar
  2. 2.
    S. Kleiner, P.J. Uggowitzer, Mater. Sci. Eng. A 379, 258 (2004)CrossRefGoogle Scholar
  3. 3.
    L. Tang, C. Liu, Z. Chen, D. Ji, H. Xiao, Mater. Des. 50, 587 (2013)CrossRefGoogle Scholar
  4. 4.
    J. Koike, T. Kobayashi, T. Mukai, H. Watanabe, M. Suzuki, K. Maruyama, K. Higashi, Acta Mater. 51, 2055 (2003)CrossRefGoogle Scholar
  5. 5.
    S. Sandlöbes, S. Zaefferer, I. Schestakow, S. Yi, R.G. Martinez, Acta Mater. 59, 429 (2011)CrossRefGoogle Scholar
  6. 6.
    A. Jain, S.R. Agnew, Mater. Sci. Eng. A 462, 29 (2007)CrossRefGoogle Scholar
  7. 7.
    Y.N. Wang, J.C. Huang, Acta Mater. 55, 897 (2007)CrossRefGoogle Scholar
  8. 8.
    M.R. Barnett, Mater. Sci. Eng. A 464, 1 (2007)CrossRefGoogle Scholar
  9. 9.
    L. Wu, S.R. Agnew, Y. Ren, D.W. Brown, B. Clausen, G.M. Stoica, H.R. Wenk, P.K. Liaw, Mater. Sci. Eng. A 527, 7057 (2010)CrossRefGoogle Scholar
  10. 10.
    X.Y. Lou, M. Li, R.K. Boger, S.R. Agnew, R.H. Wagoner, Int. J. Plast. 23, 44 (2007)CrossRefGoogle Scholar
  11. 11.
    L. Wu, A. Jain, D.W. Brown, G.M. Stoica, S.R. Agnew, B. Clausen, D.E. Fielden, P.K. Liaw, Acta Mater. 56, 688 (2008)CrossRefGoogle Scholar
  12. 12.
    G. Proust, C.N. Tomé, A. Jain, S.R. Agnew, Int. J. Plast. 25, 861 (2009)CrossRefGoogle Scholar
  13. 13.
    R.E.R. Hill, R. Abbaschian, Physical Metallurgy Principles, 3rd edn. (PWS-Kent, Boston, 1994)Google Scholar
  14. 14.
    H.E. Kadiri, J. Kapil, A.L. Oppedal, L.G. Hector Jr., S.R. Agnew, M. Cherkaoui, S.C. Vogel, Acta Mater. 61, 3549 (2013)CrossRefGoogle Scholar
  15. 15.
    L. Jiang, J.J. Jonas, A.A. Luo, A.K. Sachdev, S. Godet, Scr. Mater. 54, 771 (2006)CrossRefGoogle Scholar
  16. 16.
    L. Jiang, J.J. Jonas, R.K. Mishra, A.A. Luo, A.K. Sachdev, S. Godet, Acta Mater. 55, 3899 (2007)CrossRefGoogle Scholar
  17. 17.
    S.-G. Hong, S.H. Park, C.S. Lee, Acta Mater. 58, 5873 (2010)CrossRefGoogle Scholar
  18. 18.
    M.R. Barnett, Z. Keshavarz, A.G. Beer, D. Atwell, Acta Mater. 52, 5093 (2004)CrossRefGoogle Scholar
  19. 19.
    I. Karaman, H. Sehitoglu, A.J. Beaudoin, Y.I. Chumlyakov, H.J. Maier, C.N. Tome, Acta Mater. 48, 2031 (2000)CrossRefGoogle Scholar
  20. 20.
    B. Song, R. Xin, G. Chen, X. Zhang, Q. Liu, Scr. Mater. 66, 1061 (2012)CrossRefGoogle Scholar
  21. 21.
    Q. Dai, W. Lan, X. Chen, J. Eng. Mater. Technol. 136, 011005-1 (2013)CrossRefGoogle Scholar
  22. 22.
    L. Jiang, J.J. Jonas, A.A. Luo, A.K. Sachdev, S. Godet, Mater. Sci. Eng. A 445–446, 302 (2007)CrossRefGoogle Scholar
  23. 23.
    M. Knezevic, A. Levinson, R. Harris, R.K. Mishra, R.D. Doherty, S.R. Kalidindi, Acta Mater. 58, 6230 (2010)CrossRefGoogle Scholar
  24. 24.
    S.H. Park, S.-G. Hong, C.S. Lee, Mater. Sci. Eng. A 578, 271 (2013)CrossRefGoogle Scholar
  25. 25.
    S.-G. Hong, S.H. Park, C.S. Lee, J. Mater. Res. 25, 784 (2010)CrossRefGoogle Scholar
  26. 26.
    S.H. Park, Mater. Sci. Eng. A 680, 214 (2017)CrossRefGoogle Scholar
  27. 27.
    S.H. Park, S.-G. Hong, C.S. Lee, Mater. Sci. Eng. A 570, 149 (2013)CrossRefGoogle Scholar
  28. 28.
    S.-H. Kim, S.-G. Hong, J.-H. Lee, C.S. Lee, J. Yoon, H. Yu, S.H. Park, Mater. Sci. Eng. A 700, 191 (2017)CrossRefGoogle Scholar
  29. 29.
    S.H. Park, H.S. Kim, J.H. Bae, C.D. Yim, B.S. You, Scr. Mater. 69, 250 (2013)CrossRefGoogle Scholar
  30. 30.
    C.H. Park, C.-S. Oh, S. Kim, Mater. Sci. Eng. A 542, 127 (2012)CrossRefGoogle Scholar
  31. 31.
    G. Wan, B.L. Wu, Y.D. Zhang, G.Y. Sha, C. Esling, Mater. Sci. Eng. A 527, 2915 (2010)CrossRefGoogle Scholar
  32. 32.
    F. Kabirian, A.S. Khan, T.G. Herlod, Int. J. Plast. 68, 1 (2015)CrossRefGoogle Scholar
  33. 33.
    N. Guo, B. Song, C. Guo, R. Xin, Q. Liu, Mater. Des. 83, 270 (2015)CrossRefGoogle Scholar
  34. 34.
    B. Song, C. Wang, N. Guo, H. Pan, R. Xin, Mater. 10, 280 (2017)CrossRefGoogle Scholar
  35. 35.
    H. Chen, T. Liu, H. Yu, B. Song, D. Hou, N. Guo, J. He, Adv. Eng. Mater. 18, 1683 (2016)CrossRefGoogle Scholar
  36. 36.
    B. Hwang, S. Lee, Y.C. Kim, N.J. Kim, D.H. Shin, Mater. Sci. Eng. A 441, 308 (2006)CrossRefGoogle Scholar
  37. 37.
    H. Kolsky, Proc. Phys. Soc. 62B, 676 (1949)CrossRefGoogle Scholar
  38. 38.
    A. Gilat, in ASM Handbook, vol. 8: Mechanical Testing and Evaluation, ed. by H. Kuhn, D. Medlin (ASM Int, Metals Park, 2000)Google Scholar
  39. 39.
    Y. Shiota, Y. Tomota, A. Moriai, T. Kamiyama, Met. Mater. Int. 11, 371 (2005)CrossRefGoogle Scholar
  40. 40.
    H.S. Lee, B. Hwang, S. Lee, C.G. Lee, S.J. Kim, Metall. Mater. Trans. 35A, 2373 (2004)Google Scholar
  41. 41.
    D.G. Lee, S. Lee, C.S. Lee, Mater. Sci. Eng. A 366, 25 (2004)CrossRefGoogle Scholar
  42. 42.
    C.S. Roberts, Magnesium and Its Alloys (Wiley, New York, 1960)Google Scholar
  43. 43.
    G.E. Dieter, Mechanical Metallurgy, SI Metric edn. (McGraw-Hill Book Co., Singapore, 1988)Google Scholar
  44. 44.
    P. Franciosi, Acta Metall. 33, 1601 (1985)CrossRefGoogle Scholar
  45. 45.
    C.S. Pande, B.B. Rath, M.A. Imam, Mater. Sci. Eng. A 367, 171 (2004)CrossRefGoogle Scholar
  46. 46.
    J. Zhang, Q. Yu, Y. Jiang, Q. Li, Int. J. Plast. 27, 768 (2011)CrossRefGoogle Scholar
  47. 47.
    Y.-H. Zhao, J.F. Bingert, X.-Z. Liao, B.-Z. Cui, K. Han, A.V. Sergueeva, A.K. Mukherjee, R.Z. Valiev, T.G. Langdon, Y.T. Zhu, Adv. Mater. 18, 2949 (2006)CrossRefGoogle Scholar
  48. 48.
    J. Marian, W. Cai, V.V. Bulatov, Nat. Mater. 3, 158 (2004)CrossRefGoogle Scholar
  49. 49.
    M.A. Meyers, O. Vöhringer, V.A. Lubarda, Acta Mater. 49, 4025 (2001)CrossRefGoogle Scholar
  50. 50.
    M.R. Barnett, Scr. Mater. 59, 696 (2008)CrossRefGoogle Scholar
  51. 51.
    J.W. Christian, S. Mahajan, Prog. Mater Sci. 39, 1 (1995)CrossRefGoogle Scholar
  52. 52.
    A. Ghaderi, M.R. Barnett, Acta Mater. 59, 7824 (2011)CrossRefGoogle Scholar
  53. 53.
    N. Stanford, J. Geng, Y.B. Chun, C.H.J. Davies, J.F. Nie, M.R. Barnett, Acta Mater. 60, 218 (2012)CrossRefGoogle Scholar
  54. 54.
    I. Ulacia, C.P. Salisbury, I. Hurtado, M.J. Worswick, J. Mater. Process. Technol. 211, 830 (2011)CrossRefGoogle Scholar

Copyright information

© The Korean Institute of Metals and Materials 2018

Authors and Affiliations

  • Jong Un Lee
    • 1
  • Seok Weon Song
    • 2
  • Yongjin Kim
    • 3
  • Sang-Hoon Kim
    • 1
  • Ye Jin Kim
    • 1
  • Sung Hyuk Park
    • 1
  1. 1.School of Materials Science and EngineeringKyungpook National UniversityDaeguRepublic of Korea
  2. 2.Technical Research LaboratoriesPOSCOPohangRepublic of Korea
  3. 3.Gwangyang Steel Products Research Group, Technical Research LaboratoriesPOSCOGwangyangRepublic of Korea

Personalised recommendations