Microstructures and mechanical properties of pure

Materials Science and Engineering A 477 (2008) 366–371

Microstructures and mechanical properties of pure copper deformed severely by equal-channel angular pressing and high pressure torsion N. Lugo a,∗ , N. Llorca b , J.M. Cabrera a,c , Z. Horita d a

Department of Materials Science and Metallurgical Engineering, ETSEIB, Polytechnic University of Catalonia, Av. Diagonal 647, 08028 Barcelona, Spain b Department of Materials Science and Metallurgical Engineering, University of Barcelona, C/Mart´ı i Franqu´ es 1-11, 08028 Barcelona, Spain c CTM Technical Center, Av. Bases de Manresa 1, 08242 Manresa, Spain d Department of Materials Science and Engineering, Faculty of Engineering, Kyushu University, Fukuoka 819-0395, Japan Received 21 February 2007; received in revised form 10 May 2007; accepted 21 May 2007

Abstract Pure Cu of 99.98 wt.% purity has been processed at room temperature by diverse techniques of severe plastic deformation, namely equalchannel angular pressing (ECAP), high pressure torsion (HPT) and a combination of both in order to find out the evolution on the microstructural homogeneity for each of the processes and their combination. Starting with a grain size of ∼60 ␮m, severe plastic deformation has been introduced to the material while maintaining the sample dimensions unchanged through the processes of ECAP and HPT. A significant decrease in grain size was observed by transmission electronic microscopy (TEM). Microtensile and microhardness tests were carried out on the deformed material in the three processing conditions. A significant improvement of the tensile strength was promoted with admissible penalization on ductility. © 2007 Elsevier B.V. All rights reserved. Keywords: Equal-channel angular pressing; High pressure torsion; Severe plastic deformation

1. Introduction The processes of severe plastic deformation (SDP) have acquired a great importance in the last decade, not only because of their great potential to inflict significant deformations in alloys and metallic materials but also due to their respective potential in the production of ultrafine-grained microstructures, within the submicrometer and nanometer ranges. Two processes have taken the leadership in the SPD field, namely, equal-channel angular pressing (ECAP, see Fig. 1) [1] and high pressure torsion (HPT, see Fig. 2) [2,3]. For the first process, a considerable number of samples having cross-sections of a square, rectangular or circular shape can be deformed without suffering significant changes in the dimensions, enabling large deformations through an unlimited number of passes within a die. The ECAP die contains two channels with an equal cross-section intersecting at an angle that can range typ-

∗

Corresponding author. Tel.: +34 934016706. E-mail addresses: [email protected] (N. Lugo), [email protected] (N. Llorca), [email protected] (J.M. Cabrera), [email protected] (Z. Horita). 0921-5093/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2007.05.083

ically between 90◦ and 135◦ [1,4–7]. For the second process, a disc of a given material is located between the upper and lower anvils and a compression force is applied on the disc [3]. Once the material is upset, one of the anvils is turned at a given rotation speed. Significant reduction of grain sizes is attributed to this high pressure torsion process. This process leads to the presence of a high fraction of high angle boundaries (>15◦ ) which contributes to a significant improvement of the mechanical properties, offering great potential for the advent of the superplastic behaviour in a given material [3]. The aim of the present study is to compare the microstructural features produced by ECAP and HPT processes including those obtained by a combination of both (i.e., ECAP followed by HPT). Scarce attention has been paid in literature to the combination of these two techniques. An exception can be found in the study of Stolyarov et al. [8]. These authors studied the effect of deforming Ti-alloys under ECAP combined with HPT but at high temperature conditions. However, the present research is focused on pure copper deformed at room temperature. Also the aim of the present research is extended to evaluate whether it is possible to obtain smaller grain sizes and, consequently, to improve the mechanical properties of the material strained by shearing (ECAP) if it is subjected to ulterior higher plastic deformation process by a different strain-

N. Lugo et al. / Materials Science and Engineering A 477 (2008) 366–371

Fig. 1. Test system of the equal-channel angular pressing (ECAP) process.

ing path (torsion by HPT). The possible advantages granted by such SPD process combination will be evaluated. 2. Experimental procedure High purity copper (99.98 mass%) was selected for the present study. This copper was supplied in a cold-drawn state. An annealing treatment was performed to the as-received material (600 ◦ C for 2 h), giving a grain size of ∼60 ␮m and a hardness of 65 HV. Bars of the as-received material were then machined to obtain cylindrical samples with 10 mm in diameter and 60 mm in length for ECAP. Samples for HPT were 0.8 mm in thickness with the same diameter as the ECAP samples. ECAP was conducted at room temperature using a die having a channel angle of 90◦ and an angle of 20◦ for the outer arc of

367

curvature. The samples were pressed for eight passes through route BC where the sample was rotated about the longitudinal axis by 90◦ in the same sense [7]. The total equivalent strain introduced in this ECAP process was ∼8 [9]. For HPT, five turns were imposed at room temperature on the discs with an applied pressure of 6 GPa. Some discs were sliced with a thickness of 0.8 mm from ECAP samples in the direction perpendicular to the longitudinal axis and then subjected to additional deformation by HPT. Microstructures were observed using optical microscopy and transmission electron microscopy (TEM) using specimens taken from the discs after HPT or the sliced discs from the ECAP samples but before HPT (dotted line in Fig. 3). TEM samples were prepared using a Precision Ion Polishing System (Model 691 PIPS, GATAN Inc.) These samples were observed in a High Resolution TEM (300 kV HRTEM Phillips). Mechanical properties at room temperature were evaluated in every specimen by microindentation (hardness) at positions marked X as shown in Fig. 3 with an applied load of 50 g for 15 s. Tensile specimens were extracted from the disc as shown in Fig. 3. They were pulled with an initial strain rate of 3.3 × 10−4 s−1 at room temperature using a Microtest DEBEN machine. 3. Experimental results 3.1. Microstructure The microstructure of the material after annealing but before SPD processing showed an average equiaxed grains size of 60 ␮m. Several twins were present within the grains. TEM observations confirmed the presence of grains with highly deformed morphology after eight ECAP passes (Fig. 4). The presence of many dislocations as well as the formation of subgrains is apparent (see arrows in Fig. 4). These micrographs

Fig. 2. Test system for the high pressure torsion (HPT) process.

368


Fig. 3. Sketch of samples used in the HPT process. TEM and microtensile samples were extracted from this specimen as illustrated in the figure.

also show a significant decrease in the grain size with regard to the undeformed material. The HPT sample exhibited a well-defined bimodal grain structure (Fig. 5). This bimodal distribution consists of many fine grains and a few large grains. Although the small grains are very similar in size to those produced by ECAP, their dislocation density seems to be lower. It is worth mentioning that large grains are free of dislocations suggesting that partial recrystallization has occurred during the HPT process, possibly due to the heat generated during the deformation and due to inherent nature of low stacking fault energy materials. Twins are visible within a large grain as indicated by an arrow. This

observation can be understood because relatively large strain and strain rates were attained during HPT which in turn can heat the sample high enough to promote partial recrystallization. It was recently pointed out that dynamic recrystallization could also be occurring under SPD conditions [10,11]. Wang et al. have argued that a bimodal distribution of ultrafine grains and dislocation-free coarse grains can improve the mechanical properties in terms of both the strength and ductility [12]. It is probable that ultrafine grains would provide good mechanical resistance and coarse recrystallized grains would promote ductility. For the samples subjected to the combination of both SPD processes (ECAP eight passes + HPT five turns), a more

Fig. 4. TEM image of a sample deformed by ECAP after eight passes.

Fig. 5. TEM image of a sample deformed by HPT after five turns.


369

Fig. 6. TEM image of a sample deformed by ECAP eight passes + HPT five turns.

homogeneous microstructure is developed (see Fig. 6) with a noticeable presence of recrystallized grains. These grains showed low dislocations density with well-defined grain boundaries, suggesting the presence of a large fraction of high angle grain boundaries. Some grains (see arrows in Fig. 6) contain annealing twins, providing further evidence that recrystallization has occurred. Several micrographs were taken in the TEM study in order to determine the grain size distribution and an average value for each SPD process. Grain/cells size determination of the ECAPed samples was carried out on TEM images for which the grain size distribution is shown in Fig. 7a. One can notice that the grain size ranges between 150 and 300 nm. However, when subgrain sizes are considered, the grain size distribution changes to that shown in Fig. 7b. It is worth noting that most of the subgrains range from 60 to 100 nm. According to the TEM image in ECAPed sample (Fig. 4) the distribution is rather homogeneous and no evidence of recrystallization is found. A similar analysis was applied to the sample processed by HPT (see Fig. 8). It is apparent that the grain size distribution is grouped into two modes: one for ultrafine grains with grain sizes lying between 100 and 300 nm while the other for coarse grains having sizes between 800 and 1200 nm. These coarse grains are rarely present although they occupy a large volume fraction of the material. The grain size distribution of samples subjected to ECAP + HPT is shown in Fig. 9. It is important to mention that grain boundaries are now clearly observed due to recrystallization and grain growth both taken place while processing the samples. The distribution does not show a Gaussian shape, being closer to a bimodal one. Two populations of grain sizes can be distinguished, one centred in grain sizes between 400 and 700 nm and the other in 1000–1500 nm. This distribution is somehow different from the correspond-

Fig. 7. (a) Grain size distribution of a sample deformed by ECAP after eight passes. (b) Grain and subgrain size distribution of a sample deformed by ECAP after eight passes.

ing to single processes. Indeed is closer to the distribution obtained when deforming by HPT alone: The smallest grain sizes obtained in ECAP or single HPT samples have disappeared. However, the grain size distribution of the ECAP + HPT samples is more homogeneous a dislocation-free microstructure is present. It can be concluded that this is due to the progress of static or dynamic recrystallization promoted by the large strain introduced in the material and the temperatures reached during deformation. One must keep in mind that commercially pure copper can recrystallize at temperatures as low as 200 ◦ C.

Fig. 8. Grain size distribution of a sample deformed by HPT five turns.

370

N. Lugo et al. / Materials Science and Engineering A 477 (2008) 366–371 Table 1 Results of mechanical properties of tensile tests

Fig. 9. Grain size distribution of a sample deformed by ECAP eight passes + HPT five turns.

3.2. Microhardness determination Vickers microhardness measurements were carried out along the diameter as illustrated in Fig. 3. After eight passes of ECAP, a significant increase in the hardness over the corresponding annealed state is observed as in Fig. 10. The hardness for the HPT sample is higher than for the ECAP sample. The sample for the combined process of ECAP + HPT exhibits the highest hardness invariably along the diameter. A close inspection of Fig. 10 shows that the hardness increases gradually from the centre to the edge for the HPT samples with and without ECAP. For the ECAP samples, such a difference in hardness is not observed because the straining process is rather homogenous with ECAP. 3.3. Microtensile tests Additional evaluation of mechanical properties of the SPDprocessed samples was undertaken by microtensile tests. Yield strength and ultimate tensile strength (listed in Table 1) are in agreement with the hardness values reported in the previ-

Fig. 10. Vickers microhardness along the diameter of the discs for the different samples processed by SPD.

Sample

YS (MPa)

UTS (MPa)

% RA

Annealed ECAP eight passes HPT five turns ECAP eight passes + HPT five turns

144 322 414 406

220 378 470 452

94 87 89 90

ous paragraph. Flow curves are plotted in Fig. 11. One can notice that the largest resistance was obtained for the sample deformed by HPT with five turns, followed by the sample where both processes have been combined (Fig. 11). This also agrees with the larger grain sizes found in the ECAP + HPT sample than in the HPT alone. In terms of ductility, all three samples show similar values (ε ∼ 0.3) which are rather the half of the annealed material. However, the elongation to failure can give a false impression of high ductility in samples with very short gauge length due to large post-necking strain [13]. Instead, uniform elongation can be used as a measure of the ductility. Another alternative is the evaluation of the transversal fractured area (%RA in Table 1) because most deformation after necking is located in this area. For this purpose, the fractured area of present samples was measured using a scanning electron microscope, an illustrative example is showing in Fig. 12. High %RA values were found in all the samples confirming the large ductility already shown in the flow curve of Fig. 11. It is worth noting that, beside the fact that an important improvement in the material resistance is produced, no heavy penalization on ductility is found. Even though the value of mechanical resistance obtained for the eight passes ECAP sample is smaller, this process still introduces an important increase in the mechanical properties, which makes them attractive for industrial application, giving potential to create a higher quantity of material in comparison with the HPT process.

Fig. 11. Flow curves of a sample in the annealed state compared with samples processed by different SPD processes.


371

2. A process of recrystallization appears after subjecting the samples to a high deformation by HPT as well as for the combination of ECAP + HPT, possibly due to the heat generated during the deformation and also due to inherent nature of low stacking fault energy materials. This fact is also associated with a bimodal grain size distribution. 3. Rather homogeneous microstructure is obtained by the combined ECAE + HPT process due to the recrystallization. Acknowledgements Thanks are given to the Ministry of Science and Technology of Spain to support the project DPI2005-09324-C02-01. NL thanks the scholarship (BES-2003-2754) granted by the Ministry of Science and Technology of Spain. ZH is grateful for a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. D. Pati˜no is also acknowledged for his support in the sample preparation. References

Fig. 12. Transversal fracture area. (a) Annealed sample and (b) ECAP eight passes + HPT five turns.

4. Conclusions 1. The processes of severe plastic deformation ECAP and HPT are effective for the production of ultrafine-grained microstructures. A significant increase in the mechanical strength has been observed after subjecting the material to both ECAP and HPT.

[1] V.M. Segal, V.I. Reznikov, A.E. Drobyshevskiy, V.I. Kopylov, Russ. Metall. 1 (1981) 99–105. [2] N.A. Smirnova, V.I. Levit, V.I. Pilyugin, R.I. Kuznetsov, L.S. Davydova, V.A. Sazonova, Fiz. Met. Metalloved. 61 (1986) 1170–1177. [3] G. Sakai, Z. Horita, T.G. Langdon, Mater. Sci. Eng. A 393 (2005) 344– 351. [4] R.Z. Valiev, N.A. Krasilinikov, N.K. Tsenev, Mater. Sci. Eng. A 137 (1991) 35–40. [5] K. Nakashima, Z. Horita, M. Nemoto, T.G. Langdon, Acta Mater. 46 (1998) 1589–1599. [6] Y. Iwahashi, Z. Horita, M. Nemoto, T.G. Langdon, Acta Mater. 46 (1998) 3317–3331. [7] M. Furukawa, Z. Horita, N. Nemoto, T.G. Langdon, J. Mater. Sci. 36 (2001) 2835–2843. [8] V.V. Stolyarov, Y.T. Zhu, T.C. Lowe, R.K. Islamgaliev, R.Z. Valiev, NanoStruct. Mater. 11 (1999) 947–954. [9] Y. Iwahashi, J. Wang, Z. Horita, M. Nemoto, T.G. Langdon, Scripta Mater. 35 (1996) 143–146. [10] B.L. Li, N. Shigeiri, N. Tsuji, Y. Minamino, Mater. Sci. Forum 503/504 (2006) 615–620. [11] Y.M. Wang, E. Ma, Acta Mater. 52 (2004) 1699–1709. [12] Y. Wang, M. Chen, F. Zhou, E. Ma, Nature 419 (2002) 912–915. [13] Y. Zhao, X. Liao, S. Cheng, E. Ma, Y.T. Zhu, Adv. Mater. 18 (2006) 2280–2283.