- Email: [email protected]
Producing ultrafine-grained aluminum rods by cyclic forward-backward extrusion: Study the microstructures and mechanical properties
Materials Letters 74 (2012) 147–150
Contents lists available at SciVerse ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
Producing ultrafine-grained aluminum rods by cyclic forward-backward extrusion: Study the microstructures and mechanical properties H. Alihosseini a, M. Asle Zaeem b,⁎, K. Dehghani a, H.A. Shivaee c a b c
Materials Science and Engineering Department, Engineering School, Amirkabir University, Tehran, Iran Center for Advanced Vehicular Systems, Mississippi State University, Starkville, MS 39759, USA Institute of Nanoscience and Nanotechnology, Sharif University of Technology, Tehran, Iran
a r t i c l e
i n f o
Article history: Received 28 November 2011 Accepted 23 January 2012 Available online 30 January 2012 Keywords: Severe plastic deformation Aluminum alloy Ultrafine grained rods Mechanical properties TEM EBSD
a b s t r a c t A cyclic forward-backward extrusion (CFBE) process was used as a severe plastic deformation (SPD) technique to produce ultrafine-grained aluminum rods. Yield strength and tensile strength of the specimens increased by increasing the number of CFBE cycles, while elongation to break decreased due to an increase in the grain refinement and microhardness. According to transmission electron microscopy (TEM) and electron backscatter diffraction (EBSD) results, the average grain size was reduced from 120 μm to 315 nm after only 3 cycles of CFBE. © 2012 Elsevier B.V. All rights reserved.
1. Introduction
2. Experimental procedures
Recently, severe plastic deformation (SPD) has been widely used to produce ultrafine-grained (UFG) and nano-grained (NG) materials [1]. Among the SPD processes, equal channel angular pressing (ECAP) [2], high-pressure torsion (HPT) [3] and accumulated roll bonding (ARB) [4] are the most commonly used methods. In these processes, several passes should be employed in order to achieve the large strains that produce UFG and NG structures. However, extrusionbased SPD techniques have recently attracted more attention because they achieve even more extreme grain size reductions. Consequently, new extrusion-based SPD methods, such as twist extrusion [5], cyclic extrusion and compression [6], simple shear extrusion [7] and torsion extrusion [8], have been developed. However, not all of these methods are efficient enough for large scale production. In a recent work, we presented a novel cyclic forward-backward extrusion (CFBE) method to produce UFG materials [9]. The initial grain size of 47 μm of specimen made of an aluminum alloy was refined to ~1 μm after only 1 cycle of CFBE. We only studied 1 cycle of CFBE process and the resulting microstructures, without investigating the subsequent mechanical properties. In the current work, different cycles of the CFBE process are applied to produce ultrafine-grained aluminum rods. We study microstructures and resultant mechanical properties of specimens after different cycles of CFBE.
CFBE consists of forward-backward extrusion, followed by constrained back-pressing (Fig. 1), and completed with conventional extrusion to produce the UFG rods. These steps are performed using a twin punch setup. The outer hollow punch and the forward punch can slide parallel but in opposite directions. In the first step of CFBE process, the forward-backward process allows the sample to be extruded into the gap between the forward punch and the die. In the second step, the extruded material is pressed back by the backward punch and outer hollow punch. While outer and backward punches push the deformed sample, the forward punch is loosely lifted up by material flow, which causes the deformation to continue without any reduction in sample cross section. Moreover, the forward punch remains inside the cup during the second step, preventing the cup from collapsing or buckling inward. Consequently, the initial shape of the sample was reproduced at the end of each cycle. Finally, after n-cycles of CFBE, the process is ended by conventional forward or backward extrusion, resulting in the product's final shape, such as the circular rods in this work. Cylindrical specimens with diameters of 30 mm and lengths of 50 mm were machined from as-received AA1050 billets. They were then annealed at 600 °C for 2 h and cooled down at a rate of 25 °C/h, which resulted in an average grain size of about 120 μm. The CFBE die was made of H13 steel with a hardness of 55 RC. The CFBE process was carried out using a press with the cross-head speed of 5 mm/min at 25 °C. The produced rods had diameter of 10 mm and length of 20 cm. Tensile specimens were cut from the CFBEed rods in the
⁎ Corresponding author. Tel.: + 1 662 325 0126. E-mail address: [email protected] (M. Asle Zaeem). 0167-577X/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2012.01.102
148
H. Alihosseini et al. / Materials Letters 74 (2012) 147–150
Fig. 1. Schematic of the experimental cyclic forward-backward extrusion (CFBE) setup.
extrusion direction according to ASTM B557. The gauge length and cross-section diameter of all tensile test specimens were 26 mm and 6.5 mm, respectively. Tensile tests were carried out by an Instron machine at the strain rate of 2.6× 10− 3 s− 1. Vickers microhardness tests were carried out under a 1 kg load applied for 15 s. The microstructures of CFBEed rods were characterized by transmission electron microscopy (TEM) using a Philips CM200 model operated at 200 kV. The specimens for TEM analysis were prepared with a twin-jet polisher, and a solution of 66% methanol and 33% nitric acid was used for etching. The electron back-scatter diffraction (EBSD) technique was used to study the microstructure in the areas of approximately 600 μm2 [10]. 3. Results and discussion Fig. 2 shows TEM micrographs and the selected area diffraction (SAD) patterns taken from the cross sectional area perpendicular to
the extrusion direction after conventional extrusion (CE) and after different cycles of CFBE. A high extrusion ratio of 9 was applied during CE, leading to a decrease in average grain size from 120 μm to 8 μm. However, for the same initial average grain size of 120 μm, the specimen subjected to 1 cycle of CFBE reached an average grain size of about 980 nm, and the resulting large number of subgrain boundaries can be observed in Fig. 2b. Further considering Fig. 2b, it is apparent that dislocation walls are formed although the dislocation density is low. From the corresponding SAD pattern, it can be observed that the misorientations between adjacent subgrains are negligible. These results are in agreement with those reported by Lee et al. [11] for ECAP of AA1050. Fig. 2c illustrates the micrograph and SAD pattern of the CFBEed specimen after 2 cycles, which produced UFG structures of about 600 nm in size. Fig. 2d shows the microstructure of specimen after 3 cycles of CFBE for which the average grain size observed is about 315 nm. The SAD for this case indicates a sharper
Fig. 2. TEM micrographs and corresponding SAD patterns of AA1050: (a) after conventional extrusion (CE), (b) after 1 cycle of CFBE, (c) after 2 cycles of CFBE, and (d) after 3 cycles of CFBE.
H. Alihosseini et al. / Materials Letters 74 (2012) 147–150
149
pattern compared to the specimens subjected to 1 and 2 cycles of CFBE. Thus, as the number of CFBE cycles increases, the microstructure of AA1050 is characterized by more distinct dislocation cells [12]. It is likely that the occurrence of dynamic recovery after 3 cycles of CFBE results in finer grains with less dislocation density (Fig. 2d). Again, these results are comparable with those obtained for processing of AA1050 by other SPD methods such as ECAP [12] and friction stir process [13]. The distribution of crystallographic orientations was evaluated through inverse pole figure (IPF) by using EBSD. In Fig. 3(a), colors correspond to different orientations in the IPF of AA1050 after 3 cycles of CFBE (extrusion direction (ED) and normal direction (ND)). The texture has more equiaxed grains distribution after 3 cycles of CFBE, and grains are broken down to ultrafine grains. In addition, the orientation of grains became more miscellaneous. This map also
illustrates the random texture since various grains with wide color spectrum from the three angles and even from the central regions of IPF can be observed. Fig. 3(b) and (c) illustrates the grain-distribution histogram and the corresponding grain misorientation of AA1050 processed after 3 cycles of CFBE. From the EBSD results, the parameters, such as the fHAGb (volume fraction of High Angle Grain boundaries) and θm, (misorientation mean), were calculated. For low-angle grain boundaries, the misorientation between grains is in the range of 2° ≤ θ b 15°, whereas for high-angle grain boundaries, the misorientation exceeds 15°. According to Fig. 3, the microstructure is very homogeneous after 3 cycles of CFBE. The average grain size after 3 cycles of CFBE is about 315 nm. In Fig. 3c, which shows the grain-boundary misorientation distribution, the fHAGb and θm are 73% and 32.2°, respectively. It is evident from this chart that increasing the number of CFBE cycles to three considerably increased fHAGb, resulting in significant grain refinement [14]. Microhardness profiles taken from the cross section of rods after conventional extrusion and after different numbers of CFBE cycles are compared in Fig. 4a. The dashed line shows a microhardness of Hv ~ 30 representing the microhardness of the specimen material in its initial annealed condition. Two different trends for the CFBEed rods are evident from these microhardness profiles. First, the
Fig. 3. A specimen after 3 cycles of CFBE: (a) EBSD inverse-pole-figure map obtained from the center of the rod, (b) histograms of distribution of the grain sizes for AA1050, and (c) distribution of boundary misorientation angles.
Fig. 4. (a) Distribution of microhardness from the center toward the edge of the rodshape specimens (b) tensile stress–strain curves for AA1050 specimens as-received, after CE, and after different cycles of CFBE.
150
H. Alihosseini et al. / Materials Letters 74 (2012) 147–150
hardness increases as the number of CFBE cycles increases. Second, the changes in hardness become more uniform by increasing the number of CFBE cycles. The microhardness after 3 cycles is Hv ~ 70, about 2.3 times greater than the initial hardness. Fig. 4b compares the stress–strain curves of the starting material, a conventional-extruded sample, and CFBEed specimens. The yield strength and tensile strength of the initial sample material were approximately 45 MPa and 65 MPa, respectively, and its ductility was 18%. After three CFBE cycles, both yield strength and tensile strength increased significantly, to 156 MPa and 200 MPa, respectively, which are about 3.5 and 3 times greater than those of the as-received samples. According to Fig. 4b, the difference between the yield and the tensile strength is decreased by increasing the CFBE cycles. Similar to other SPD methods, the increase in strength during CFBE can be attributed to strain hardening (σs ∝ ρ 1/2, where ρ is dislocation density) and to grain refinement (σb ∝ d − 1/2, where d is the grain size). The same hardening mechanisms were reported by other researchers [15], where the dominant hardening mechanism is dependent on the amount of strain. Notably, the decrease in elongation to break after CFBE, however, does not follow the same trend as the increase in the tensile strength. Instead, there is a noticeable drop in uniform elongation to break with an increase in the number of CFBE cycles, which can be explained in terms of plastic instability (necking in the tensile test) [16]. Plastic instability can be explained by this relation: σ ≥ dσ/dε, where σ and ε are true stress and true strain, respectively [17]. For an UFG material, plastic instability occurs at the very early stage of the tensile test, resulting in limited elongation. Nevertheless, the present AA1050 specimens subjected to CFBE did not exhibit remarkable decrease in ductility. The same mentioned mechanisms accountable for increase in the yield stress and tensile strength after CFBE cycles can be responsible for a two-fold increase in hardness, Fig. 4a. It was also reported that the rapid increase of microhardness can be attributed to the strain hardening as a result of subgrain boundaries/cell wall formation rather than grain refinement [18]. Accordingly, due to the large applied strains, the CFBEed specimen exhibits saturation in strain hardening.
This phenomenon has been reported for the cases of UFGed samples produced by other SPD methods such as ARB [16] and ECAP [19]. This saturation occurs because the SPDed materials reach a steady-state in dislocation density. 4. Conclusion CFBE was used to produce ultrafine-grained aluminum rods. The strength and hardness of AA1050 specimens subjected to CFBE were significantly improved as the initial coarse grain of 120 μm was reduced to ultra-fine grains of 315 nm only after 3 cycles of CFBE. Compared to conventional SPD techniques, the structure produced by CFBE is more homogenous, containing nearly equiaxed grains. After three CFBE cycles, both yield strength and tensile strength increased about 3.5 and 3 times greater than those of the as-received samples. Results showed that the CFBE process is a promising SPD method to produce bulk nanostructured products for industrial applications. References [1] Wang M, Shan A. J Mater Process Technol 2008;202:549. [2] Segal VM. Mater Sci Eng A 1999;271:322. [3] Languillaume J, Chmelik F, Kapelski G, Bordeaux F, Nazarov AA, Canova G, et al. Acta Metall Mater 1993;41:2953. [4] Saito Y, Utsunomiya H, Tsuji N. Sakai T. Acta Mater 1999;47:579. [5] Beygelzimer Y, Varyukhin V, Synkov S, Orlov D. Mater Sci Eng A 2009;503:14. [6] Korbel A, Richert M, Richert J. Proc. Sec. RISO Int. Symp. Metall. Mater. Sci; 1981. p. 445. September. [7] Pardis N, Ebrahimi R. Mater Sci Eng A 2009;527:355. [8] Shin DH, Park JJ, Kim YS, Park KT. Mater Sci Eng A 2002;328:98. [9] Alihosseini H, Asle Zaeem M, Dehghani K. Mater Lett 2012;68:204. [10] Humphreys FJ. J Mater Sci 2001;36:3833. [11] Lee JC, Suh JY, Ahn JP. Metall Mater Trans A 2003;34:625. [12] Lee JC, Seok HK, Suh JY. Acta Mater 2002;50:4005. [13] Kwon YJ, Shigematsu I, Saito N. Scr Mater 2003;49:785. [14] Chang CP, Sun PL, Kao PW. Acta Mater 2000;48:3377. [15] Hansen N, Huang X, Ueji R, Tsuji N. Mater Sci Eng A 2004;387–389:191. [16] Tsuji N, Ito Y, Saito Y, Minamino Y. Scr Mater 2002;47:893. [17] Dieter GE. Mechanical Metallurgy. McGraw-Hill; 1986. p. 289–92. [18] Shaarbaf M, Toroghinejad MR. Mater Sci Eng A 2008;473:28. [19] El-Danaf EA, Soliman MS, Almajid AA, El-Rayes MM. Mater Sci Eng A 2007;458: 226.
Contents lists available at SciVerse ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
Producing ultrafine-grained aluminum rods by cyclic forward-backward extrusion: Study the microstructures and mechanical properties H. Alihosseini a, M. Asle Zaeem b,⁎, K. Dehghani a, H.A. Shivaee c a b c
Materials Science and Engineering Department, Engineering School, Amirkabir University, Tehran, Iran Center for Advanced Vehicular Systems, Mississippi State University, Starkville, MS 39759, USA Institute of Nanoscience and Nanotechnology, Sharif University of Technology, Tehran, Iran
a r t i c l e
i n f o
Article history: Received 28 November 2011 Accepted 23 January 2012 Available online 30 January 2012 Keywords: Severe plastic deformation Aluminum alloy Ultrafine grained rods Mechanical properties TEM EBSD
a b s t r a c t A cyclic forward-backward extrusion (CFBE) process was used as a severe plastic deformation (SPD) technique to produce ultrafine-grained aluminum rods. Yield strength and tensile strength of the specimens increased by increasing the number of CFBE cycles, while elongation to break decreased due to an increase in the grain refinement and microhardness. According to transmission electron microscopy (TEM) and electron backscatter diffraction (EBSD) results, the average grain size was reduced from 120 μm to 315 nm after only 3 cycles of CFBE. © 2012 Elsevier B.V. All rights reserved.
1. Introduction
2. Experimental procedures
Recently, severe plastic deformation (SPD) has been widely used to produce ultrafine-grained (UFG) and nano-grained (NG) materials [1]. Among the SPD processes, equal channel angular pressing (ECAP) [2], high-pressure torsion (HPT) [3] and accumulated roll bonding (ARB) [4] are the most commonly used methods. In these processes, several passes should be employed in order to achieve the large strains that produce UFG and NG structures. However, extrusionbased SPD techniques have recently attracted more attention because they achieve even more extreme grain size reductions. Consequently, new extrusion-based SPD methods, such as twist extrusion [5], cyclic extrusion and compression [6], simple shear extrusion [7] and torsion extrusion [8], have been developed. However, not all of these methods are efficient enough for large scale production. In a recent work, we presented a novel cyclic forward-backward extrusion (CFBE) method to produce UFG materials [9]. The initial grain size of 47 μm of specimen made of an aluminum alloy was refined to ~1 μm after only 1 cycle of CFBE. We only studied 1 cycle of CFBE process and the resulting microstructures, without investigating the subsequent mechanical properties. In the current work, different cycles of the CFBE process are applied to produce ultrafine-grained aluminum rods. We study microstructures and resultant mechanical properties of specimens after different cycles of CFBE.
CFBE consists of forward-backward extrusion, followed by constrained back-pressing (Fig. 1), and completed with conventional extrusion to produce the UFG rods. These steps are performed using a twin punch setup. The outer hollow punch and the forward punch can slide parallel but in opposite directions. In the first step of CFBE process, the forward-backward process allows the sample to be extruded into the gap between the forward punch and the die. In the second step, the extruded material is pressed back by the backward punch and outer hollow punch. While outer and backward punches push the deformed sample, the forward punch is loosely lifted up by material flow, which causes the deformation to continue without any reduction in sample cross section. Moreover, the forward punch remains inside the cup during the second step, preventing the cup from collapsing or buckling inward. Consequently, the initial shape of the sample was reproduced at the end of each cycle. Finally, after n-cycles of CFBE, the process is ended by conventional forward or backward extrusion, resulting in the product's final shape, such as the circular rods in this work. Cylindrical specimens with diameters of 30 mm and lengths of 50 mm were machined from as-received AA1050 billets. They were then annealed at 600 °C for 2 h and cooled down at a rate of 25 °C/h, which resulted in an average grain size of about 120 μm. The CFBE die was made of H13 steel with a hardness of 55 RC. The CFBE process was carried out using a press with the cross-head speed of 5 mm/min at 25 °C. The produced rods had diameter of 10 mm and length of 20 cm. Tensile specimens were cut from the CFBEed rods in the
⁎ Corresponding author. Tel.: + 1 662 325 0126. E-mail address: [email protected] (M. Asle Zaeem). 0167-577X/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2012.01.102
148
H. Alihosseini et al. / Materials Letters 74 (2012) 147–150
Fig. 1. Schematic of the experimental cyclic forward-backward extrusion (CFBE) setup.
extrusion direction according to ASTM B557. The gauge length and cross-section diameter of all tensile test specimens were 26 mm and 6.5 mm, respectively. Tensile tests were carried out by an Instron machine at the strain rate of 2.6× 10− 3 s− 1. Vickers microhardness tests were carried out under a 1 kg load applied for 15 s. The microstructures of CFBEed rods were characterized by transmission electron microscopy (TEM) using a Philips CM200 model operated at 200 kV. The specimens for TEM analysis were prepared with a twin-jet polisher, and a solution of 66% methanol and 33% nitric acid was used for etching. The electron back-scatter diffraction (EBSD) technique was used to study the microstructure in the areas of approximately 600 μm2 [10]. 3. Results and discussion Fig. 2 shows TEM micrographs and the selected area diffraction (SAD) patterns taken from the cross sectional area perpendicular to
the extrusion direction after conventional extrusion (CE) and after different cycles of CFBE. A high extrusion ratio of 9 was applied during CE, leading to a decrease in average grain size from 120 μm to 8 μm. However, for the same initial average grain size of 120 μm, the specimen subjected to 1 cycle of CFBE reached an average grain size of about 980 nm, and the resulting large number of subgrain boundaries can be observed in Fig. 2b. Further considering Fig. 2b, it is apparent that dislocation walls are formed although the dislocation density is low. From the corresponding SAD pattern, it can be observed that the misorientations between adjacent subgrains are negligible. These results are in agreement with those reported by Lee et al. [11] for ECAP of AA1050. Fig. 2c illustrates the micrograph and SAD pattern of the CFBEed specimen after 2 cycles, which produced UFG structures of about 600 nm in size. Fig. 2d shows the microstructure of specimen after 3 cycles of CFBE for which the average grain size observed is about 315 nm. The SAD for this case indicates a sharper
Fig. 2. TEM micrographs and corresponding SAD patterns of AA1050: (a) after conventional extrusion (CE), (b) after 1 cycle of CFBE, (c) after 2 cycles of CFBE, and (d) after 3 cycles of CFBE.
H. Alihosseini et al. / Materials Letters 74 (2012) 147–150
149
pattern compared to the specimens subjected to 1 and 2 cycles of CFBE. Thus, as the number of CFBE cycles increases, the microstructure of AA1050 is characterized by more distinct dislocation cells [12]. It is likely that the occurrence of dynamic recovery after 3 cycles of CFBE results in finer grains with less dislocation density (Fig. 2d). Again, these results are comparable with those obtained for processing of AA1050 by other SPD methods such as ECAP [12] and friction stir process [13]. The distribution of crystallographic orientations was evaluated through inverse pole figure (IPF) by using EBSD. In Fig. 3(a), colors correspond to different orientations in the IPF of AA1050 after 3 cycles of CFBE (extrusion direction (ED) and normal direction (ND)). The texture has more equiaxed grains distribution after 3 cycles of CFBE, and grains are broken down to ultrafine grains. In addition, the orientation of grains became more miscellaneous. This map also
illustrates the random texture since various grains with wide color spectrum from the three angles and even from the central regions of IPF can be observed. Fig. 3(b) and (c) illustrates the grain-distribution histogram and the corresponding grain misorientation of AA1050 processed after 3 cycles of CFBE. From the EBSD results, the parameters, such as the fHAGb (volume fraction of High Angle Grain boundaries) and θm, (misorientation mean), were calculated. For low-angle grain boundaries, the misorientation between grains is in the range of 2° ≤ θ b 15°, whereas for high-angle grain boundaries, the misorientation exceeds 15°. According to Fig. 3, the microstructure is very homogeneous after 3 cycles of CFBE. The average grain size after 3 cycles of CFBE is about 315 nm. In Fig. 3c, which shows the grain-boundary misorientation distribution, the fHAGb and θm are 73% and 32.2°, respectively. It is evident from this chart that increasing the number of CFBE cycles to three considerably increased fHAGb, resulting in significant grain refinement [14]. Microhardness profiles taken from the cross section of rods after conventional extrusion and after different numbers of CFBE cycles are compared in Fig. 4a. The dashed line shows a microhardness of Hv ~ 30 representing the microhardness of the specimen material in its initial annealed condition. Two different trends for the CFBEed rods are evident from these microhardness profiles. First, the
Fig. 3. A specimen after 3 cycles of CFBE: (a) EBSD inverse-pole-figure map obtained from the center of the rod, (b) histograms of distribution of the grain sizes for AA1050, and (c) distribution of boundary misorientation angles.
Fig. 4. (a) Distribution of microhardness from the center toward the edge of the rodshape specimens (b) tensile stress–strain curves for AA1050 specimens as-received, after CE, and after different cycles of CFBE.
150
H. Alihosseini et al. / Materials Letters 74 (2012) 147–150
hardness increases as the number of CFBE cycles increases. Second, the changes in hardness become more uniform by increasing the number of CFBE cycles. The microhardness after 3 cycles is Hv ~ 70, about 2.3 times greater than the initial hardness. Fig. 4b compares the stress–strain curves of the starting material, a conventional-extruded sample, and CFBEed specimens. The yield strength and tensile strength of the initial sample material were approximately 45 MPa and 65 MPa, respectively, and its ductility was 18%. After three CFBE cycles, both yield strength and tensile strength increased significantly, to 156 MPa and 200 MPa, respectively, which are about 3.5 and 3 times greater than those of the as-received samples. According to Fig. 4b, the difference between the yield and the tensile strength is decreased by increasing the CFBE cycles. Similar to other SPD methods, the increase in strength during CFBE can be attributed to strain hardening (σs ∝ ρ 1/2, where ρ is dislocation density) and to grain refinement (σb ∝ d − 1/2, where d is the grain size). The same hardening mechanisms were reported by other researchers [15], where the dominant hardening mechanism is dependent on the amount of strain. Notably, the decrease in elongation to break after CFBE, however, does not follow the same trend as the increase in the tensile strength. Instead, there is a noticeable drop in uniform elongation to break with an increase in the number of CFBE cycles, which can be explained in terms of plastic instability (necking in the tensile test) [16]. Plastic instability can be explained by this relation: σ ≥ dσ/dε, where σ and ε are true stress and true strain, respectively [17]. For an UFG material, plastic instability occurs at the very early stage of the tensile test, resulting in limited elongation. Nevertheless, the present AA1050 specimens subjected to CFBE did not exhibit remarkable decrease in ductility. The same mentioned mechanisms accountable for increase in the yield stress and tensile strength after CFBE cycles can be responsible for a two-fold increase in hardness, Fig. 4a. It was also reported that the rapid increase of microhardness can be attributed to the strain hardening as a result of subgrain boundaries/cell wall formation rather than grain refinement [18]. Accordingly, due to the large applied strains, the CFBEed specimen exhibits saturation in strain hardening.
This phenomenon has been reported for the cases of UFGed samples produced by other SPD methods such as ARB [16] and ECAP [19]. This saturation occurs because the SPDed materials reach a steady-state in dislocation density. 4. Conclusion CFBE was used to produce ultrafine-grained aluminum rods. The strength and hardness of AA1050 specimens subjected to CFBE were significantly improved as the initial coarse grain of 120 μm was reduced to ultra-fine grains of 315 nm only after 3 cycles of CFBE. Compared to conventional SPD techniques, the structure produced by CFBE is more homogenous, containing nearly equiaxed grains. After three CFBE cycles, both yield strength and tensile strength increased about 3.5 and 3 times greater than those of the as-received samples. Results showed that the CFBE process is a promising SPD method to produce bulk nanostructured products for industrial applications. References [1] Wang M, Shan A. J Mater Process Technol 2008;202:549. [2] Segal VM. Mater Sci Eng A 1999;271:322. [3] Languillaume J, Chmelik F, Kapelski G, Bordeaux F, Nazarov AA, Canova G, et al. Acta Metall Mater 1993;41:2953. [4] Saito Y, Utsunomiya H, Tsuji N. Sakai T. Acta Mater 1999;47:579. [5] Beygelzimer Y, Varyukhin V, Synkov S, Orlov D. Mater Sci Eng A 2009;503:14. [6] Korbel A, Richert M, Richert J. Proc. Sec. RISO Int. Symp. Metall. Mater. Sci; 1981. p. 445. September. [7] Pardis N, Ebrahimi R. Mater Sci Eng A 2009;527:355. [8] Shin DH, Park JJ, Kim YS, Park KT. Mater Sci Eng A 2002;328:98. [9] Alihosseini H, Asle Zaeem M, Dehghani K. Mater Lett 2012;68:204. [10] Humphreys FJ. J Mater Sci 2001;36:3833. [11] Lee JC, Suh JY, Ahn JP. Metall Mater Trans A 2003;34:625. [12] Lee JC, Seok HK, Suh JY. Acta Mater 2002;50:4005. [13] Kwon YJ, Shigematsu I, Saito N. Scr Mater 2003;49:785. [14] Chang CP, Sun PL, Kao PW. Acta Mater 2000;48:3377. [15] Hansen N, Huang X, Ueji R, Tsuji N. Mater Sci Eng A 2004;387–389:191. [16] Tsuji N, Ito Y, Saito Y, Minamino Y. Scr Mater 2002;47:893. [17] Dieter GE. Mechanical Metallurgy. McGraw-Hill; 1986. p. 289–92. [18] Shaarbaf M, Toroghinejad MR. Mater Sci Eng A 2008;473:28. [19] El-Danaf EA, Soliman MS, Almajid AA, El-Rayes MM. Mater Sci Eng A 2007;458: 226.