- Email: [email protected]
Improving the mechanical properties of pure magnesium through cold hydrostatic extrusion and low-temperature annealing
Materials Science & Engineering A 627 (2015) 56–60
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Improving the mechanical properties of pure magnesium through cold hydrostatic extrusion and low-temperature annealing A.Yu. Volkov n, I.V. Kliukin Institute of Metal Physics, Ural Branch, Russian Academy of Sciences, 18 S. Kovalevskaya St., Ekaterinburg 620990, Russia
art ic l e i nf o Article history: Received 19 September 2014 Received in revised form 17 December 2014 Accepted 24 December 2014 Available online 7 January 2015 Keywords: Mechanical characterization Magnesium Bulk deformation Thermal anomaly
a b s t r a c t A way to produce ultrafine grained structure with d 3 μm and high plasticity in pure magnesium through the original technique of hydrostatic extrusion at room temperature has been demonstrated. The microstructure and the mechanical properties of the extruded magnesium rods have been investigated. The evolution of mechanical properties of the samples after treatments up to 450 1C has been analyzed. It has been shown that low-temperature treatments lead to some increase in strength and plasticity of pre-deformed magnesium. The annealing hardening phenomenon is explained by the thermally activated processes of rearrangement of the dislocation structure and blocking of dislocations. & 2015 Elsevier B.V. All rights reserved.
1. Introduction It is well known that at temperatures below 225 1C the workability of pure magnesium and magnesium alloys is limited or even not possible [1]. Poor room-temperature formability of magnesium is connected with its hexagonal close packed (hcp) crystalline structure in which the basal slip system dominates room-temperature deformation [2]. At high temperature formation, the prismatic and pyramidal slip systems are activated and thus the plasticity of magnesium increases. Grain refinement and texture modification are both considered to contribute to the improvement of plasticity of magnesium. As a rule, different severe plastic deformation (SPD) techniques are used for this purpose [3,4]. A detailed review of studies, which used the well-known SPD method of equal channel angular extrusion (ECAE) for deformation of magnesium, was given in [4]. It was shown that the deformation of magnesium by the ECAE method is much impeded at lower temperatures. Nevertheless, over and over again, researchers have attempted to deform magnesium at room temperature in order to obtain ultrafine-grain structure (UFG) with high strength and ductility. A UFG material is stronger than one in a coarse grained state because of grain size hardening according to the Hall–Petch relationship [2,5]. In addition, UFG magnesium has a better ductility as well as a low ductile to brittle transition temperature; thus, their formability at room temperature can be improved [6].
n
Corresponding author. E-mail address: [email protected] (A.Yu. Volkov).
http://dx.doi.org/10.1016/j.msea.2014.12.104 0921-5093/& 2015 Elsevier B.V. All rights reserved.
Undesirable cracking of magnesium during deformation may be avoided even at room temperature if processing by highpressure torsion is used [7]. In this case, formation of the magnesium sample occurs within the field of compression stresses. As a result, pores, defects, and cracks existing in the material cannot develop. Therefore, it is necessary to provide all-around compression stresses for the successful deformation of materials with poor plasticity. Of course, it was well known before [8]. Recently, we used this method for deformation of bulk magnesium workpiece at room temperature [9]. For this purpose a magnesium cylinder of 20 mm diameter and 30 mm height was placed in a thickwalled copper container. The deformation was carried out by squeezing this “assembly” at room temperature. Such a method of deformation generates high compression stresses on the side surfaces of the magnesium workpiece. They are caused by the response of the container walls to the extension of the sample during pressing. At the final stage of the experiment the remains of the copper container were cut and removed. After processing by squeezing into the container the magnesium sample was significantly shorter and bigger in diameter, having the form of a puck. The degree of true (or logarithmic) deformation (ε) of magnesium reached ε 0.85. No sign of destruction of the magnesium sample was observed. As the structure became virtually isotropic and the grain size decreased to 5 mm, the plasticity of magnesium increased. A very interesting phenomenon was revealed during these experiments. The yield strength of the pre-deformed magnesium samples does not drop but increases as a result of low-temperature annealing. Obviously, this phenomenon is of both fundamental and practical interest. In addition, obviously, this result needs verifying. These experiments showed that the use of a container increases the formability of magnesium at room temperature.
A.Yu. Volkov, I.V. Kliukin / Materials Science & Engineering A 627 (2015) 56–60
This is caused by the formation of compression stresses around the sample during deformation. It seems interesting to use a container to increase the formability of magnesium with other methods of SPD. For example, magnesium placed in a shell can be deformed by hydroextrusion. A great variety of sources did not produce any description of successful experiments of hydrostatic extrusion of pure magnesium at room temperature. Moreover, magnesium alloys are very difficult to be cold extruded, although they have better formability than pure magnesium. As shown in [10], hydrostatic extrusion of magnesium alloys at temperatures below 100 1С leads to the destruction of workpieces. The purpose of this study was to develop a hydrostatic extrusion technique for magnesium at room temperature, followed by an investigation of the structure and mechanical properties of the samples in the obtained structural state.
2. Material and method As a material for experiment, an ingot of commercially pure magnesium (99.98%) was chosen. The investigation has shown that the ingot consists of very large grains (Fig. 1). The largest grains are disposed in the centre of the ingot and they are more than 30 mm in length and near 5 mm in width. Such coarse grained structure is typical for magnesium ingot (see, for example, [11]). Cylindrical workpieces of 10 mm in diameter and 70 mm in length for hydrostatic extrusion experiments were cut out of this ingot. The result of our experiment using a conventional technique is shown in Fig. 2a. We are far from being the first who observed magnesium workpieces cracking after SPD at room temperature (see, for example, [10–13]). To prevent cracking of the magnesium during its flowing out from the die, we developed an original method (Fig. 2b). We placed a magnesium workpiece (1) of 10 mm in diameter in a shell (2) of 20 mm in diameter. Hydrostatic
Fig. 1. Macrostructure of initial as-cast magnesium.
57
extrusion was made through a die (3) of 10 mm at room temperature. The distinction of this method consists in the material of the shell. As was described above, we carried out successful squeeze of magnesium workpiece placed in a copper container before [9]. Nevertheless, this method had a disadvantage. When the deformation process was finished, we had to cut the container to take the sample out of it. In our experiments on hydroextrusion, we replaced copper by another material. It is clearly seen in Fig. 2b that the shell (2) breaks down once hydrostatic extrusion is completed. It provides us an easy way to extract the sample out of the shell. The material of the shell is specialized knowledge (“know-how” of the Institute of Metal Physics of the Russian Academy of Sciences). As the result of the employment of this technique, a long magnesium rod of 5 mm in diameter was formed without any flaws (Fig. 2b). Microstructure observations were realized by means of a scanning electron microscopy (SEM) instrument QUANTA 200 FEI. Tensile mechanical tests were carried out at room temperature using an Instron test instrument. The length of the samples was 50 mm and the strain rate was 2 10 3 s 1. Five samples were tested per each experimental point. All of the heat treatments were carried out in evacuated glass ampoules. Samples were subjected to annealing in the temperature interval from 100 to 450 1C for 6 h followed by cooling in water.
3. Results 3.1. Microstructure of the extruded rod After the first step of hydroextrusion processing we observed grains of dE 3–5 μm in the microstructure of Ø5 mm rod (Fig. 3a). The largest grains attain 15 μm in size; they are noted in the central part of this rod. The magnesium rod in the obtained structural state has high plasticity. For example, samples for mechanical tests of 2 mm in diameter were made by hydroextrusion processing of magnesium rod Ø5 mm through two passes at room temperature without any shell. First, rods of 3 mm diameter were obtained. Then, they underwent hydrostatic extrusion again through a die of Ø2 mm. So, the total degree of true deformation (ε) of this magnesium sample was ε 3.2. After deformation to Ø2 mm, the structure becomes homogenous with grain size dE2–3 μm (Fig. 3b). As it can be seen, in this case the refinement of the initial cast structure by more than three orders of magnitude takes place (we may compare Fig. 1 with Fig. 3b). Small grain formation in highly deformed magnesium is caused by dynamic recrystallization [1,13,14]. The values of grain sizes obtained can be compared to the minimal values cited in the literature [2–5]. An average grain size of 6–8 mm was achieved by ECAE of commercially pure magnesium at 250 1С after four passes [6]. As was found in [4], the minimum average grain size obtained after ECAE processing was not less than 2 mm. 3.2. Mechanical properties
Fig. 2. Appearance of magnesium after hydrostatic extrusion at room temperature: (a) a destroyed workpiece after experiment using conventional technique and (b) a successful experiment with workpiece (1) which was placed in a self-destroying shell (2) and passed through the die (3).
True stress (σ) versus elongation (δ) responses of magnesium samples with 2 mm in diameter are plotted in Fig. 4a. It is seen that magnesium samples after hydrostatic extrusion at ε 3.2 following the developed technique have a very high plasticity. For example, the elongation-to-failure of such samples is δ ¼19% (curve 1 in Fig. 4a). The mechanical properties of samples that we obtained by our technique can be compared to the mechanical properties of those reported in the literature. For example, in the rods of 2 mm in
A.Yu. Volkov, I.V. Kliukin / Materials Science & Engineering A 627 (2015) 56–60
True stress, σ(MPa)
58
2
200 180 160
1
140 120 100 80 0
5
10
15
20
Elongation,δ (%)
(100)
(002)
Intensity
(101)
(110) (200)(112) (201)
(102)
(103)
2 1
20
30
40
50
60
70
80
2θ, degree Fig. 4. Tensile diagrams (a) and θ–2θ XRD scans obtained from the cross section (b) of magnesium samples Ø2 mm with true deformation ε 3.2. Sample (1) was deformed by three passes through cold hydroextrusion and sample (2) was drawn at room temperature after one pass of cold extrusion.
Fig. 3. Microstructure of magnesium samples after hydroextrusion at room temperature: (a) a rod with 5 mm diameter extruded into self-destroying shell and (b) a rod with 2 mm diameter obtained by passes through dies with 5 mm, 3 mm and 2 mm diameters.
diameter the grain size is 2–3 mm, which virtually coincides with the value of grain size (dE 5 mm) obtained on samples of pure magnesium as the result of ECAE-processing in the work [5]. The strain rates in our work and in [5] are also close to each other. As can be seen from Fig. 4a (curve 1), the yield strength of our samples is σYS ¼ 98 MPa and the ultimate tensile strength is σUTS ¼182 MPa. In accordance with the results of [5], the strength properties of their samples were σYS ¼141 MPa, σUTS ¼ 167 MPa. The elongation-to-failure of the rods with Ø2 mm is δ ¼19%, and for samples from [5], δ ¼ 16%. Thus, though grain sizes in the samples are nearly equal, their yield strengths differ from each other by more than 40%. It should be noted that σYS ¼102 MPa was obtained in [5] on the samples with the grain size d E55 mm. Thus, if yield strengths of the extruded rods and the samples from the paper [5] are nearly equal, their grain sizes differ from each other by one order of magnitude. On the one hand, such behavior does not match the well-known Hall–Petch relationship according to which σYS E σ0 þ kd 0.5. On the other hand, different researches [2,15,16] have already registered some deviations from the correlation between the yield strength (σYS) and the grain size (d) for a polycrystalline magnesium. Moreover, we can find some manuscripts that describe magnesium samples with different grain sizes demonstrating very close tensile strength values (for example, results from [5] can be compared to the ones from [16]). One more example was given in [16]: σYS of magnesium with UFG structure
(d E200 nm) was exceptionally low, but magnesium samples with grain sizes larger than 1 μm had higher yield strength. To understand the deformation behavior of magnesium, the texture of the sample must be taken into account [17]. The θ–2θ XRD-scan for the sample extruded at room temperature is shown in Fig. 4b (the lower diffractogram). The X-ray diffraction pattern of the cross-section actually reveals the absence of the (002) peak. In contradiction, in this X-ray diffraction pattern, the (100) and (101) peaks have sharp form and high intensity. Such a picture of X-ray reflections is explained by the parallel alignment of the basal planes in the all crystallites along the direction of extrusion. Therefore, a sharp radial texture is formed in the sample under extrusion at room temperature. We tested the effect of texture on the mechanical properties of magnesium. For this aim, the rod 5 mm in diameter obtained by hydrostatic extrusion of a workpiece within the shell was cold drawn to a diameter of 2 mm. The total degree of true deformation of the samples after one pass through the hydroextrusion followed by cold drawing was ε 3.2. This value is equal to true deformation of the extruded sample Ø2 mm. The grain size in the sample is also near 3 μm. Therefore, the grain sizes in both samples are almost the same. The X-ray diffraction pattern 2 in Fig. 4b clearly shows that the texture of the sample after the extrusion followed by drawing is not so sharp. For example, when shot from the crosssection area, the reflections from the basal planes give a visible peak (002). The strength properties of these samples are σYS ¼ 127 MPa, σUTS ¼195 MPa and the elongation-to-failure δ ¼ 10% (curve 2 in Fig. 4b). Therefore, the crystallographic texture might contribute to the dissimilarity of the mechanical behavior of magnesium samples with close grain sizes. A very thorough study on this subject has been published recently in [18]. We have also studied the thermal stability of the mechanical properties of the samples after their annealing in the temperature interval from 100 to 450 1C for 6 h, followed by cooling in water. It is well seen that annealing at temperature below 200 1С leads to
A.Yu. Volkov, I.V. Kliukin / Materials Science & Engineering A 627 (2015) 56–60
Fig. 5. The dependence of the yield strength (σYS), ultimate tensile strength (σUTS) and the elongation-to-failure (δ) on the holding temperature for 6 h of predeformed magnesium rods with 2 mm in diameter. (a) Samples were subjected to cold hydrostatic extrusion by three passes at room temperature and (b) samples were obtained by one pass of hydroextrusion followed by cold drawing.
some increase in strength and plasticity (Fig. 5). Small increase in plasticity can be explained by the operation of recovery processes when, in the course of a low-temperature annealing, the level of elastic stresses is decreased. However, the increase in strength resulting from the annealing of highly deformed material is an anomaly, which needs explaining.
4. Discussion It should be noted that the increase in yield strength (σYS) along with the increase in test temperature (T) is a characteristic feature of many intermetallics: Ni3Al, TiAl etc. (see Refs. [19] and references herein). It has been assumed to be an established fact that the anomalous temperature dependence of σYS(T) is due to the blocking of dislocations. As far as pure metals are concerned, the thermal anomaly of the yield strength was observed before only in hcp metals (for example, in magnesium [20] and titanium [21]). Such a behavior of these metals is explained by a limited number of slip systems in crystals with hcp structure and different slip systems are subject to different amounts of Peierls barrier [16]. For example, there is a well-marked difference between the light basic slip and the much more difficult pyramidal and prismatic slip in magnesium [1,22]. In the papers [20,23], thermal mechanical tests of Mg single crystals were carried out. It was found that, for a [0001] orientation of the crystal, the yield strength in the interval between 100 and 150 1С is higher than at room temperature. For a [112̄ 0] orientation of the magnesium crystal, the maximum yield strength was observed at a temperature of about 100 1С. The anomalous temperature trend of the σYS(T) is explained by the thermally activated processes of rearrangement of the dislocation
59
structure and blocking of dislocations. The blocked (cþa) dislocation configurations located on second-order pyramidal planes were observed in the region of this temperature anomaly. There is no such blocked configuration after deformation at a higher temperature. It should be noted once more that, in works [20,23], mechanical tests were carried out on samples heated up to the test temperature. Our experiments are different because preliminarily deformed samples of magnesium after annealing were cooled and mechanical tests were carried out at room temperature. Nevertheless, the increase in strength after low-temperature annealing is observed in Fig. 5. Earlier, such effect was predicted theoretically after the investigation of dislocation structure of magnesium single crystals [24]. The possibility of the rise of the strengths was explained in this work by the processes of self-blocking of dislocations upon a low-temperature annealing of preliminarily deformed magnesium single crystals. It is shown that Peierls relief of the edge (с þ а) dislocation has two valleys at the second-order pyramidal plane. When thermal activation is present, the dislocation drops into the deepest valley of the potential relief. This position of the edge (cþa) dislocation has the lowest energy and therefore such configuration is blocked. The model presented explains well the dislocation extension along a selected direction, which is observed after heating of pre-deformed magnesium samples without external stress. According to this investigation [24], the self-blocking of dislocations in magnesium can be observed in a temperature interval lower than 150 1С. The thermal dependences of strength properties in Fig. 5 correspond to the results presented in the work [24]. As a result of the above discussion and taking into account the Hall–Petch relationship we can conclude from Fig. 5 that the grain size in the magnesium samples does not change in the course of low-temperature anneals for 6 h. Consequently, the structure formed by the deformation to ε1 3.2 is quite stable. The decrease in yield strength of samples starts after anneals at temperatures higher than 250 1С. Consequently, it is in this temperature range the coarsening of the grain structure begins. A sharp drop of elongation after annealing at temperatures higher than 300 1С confirms this conclusion. Indeed, a rapid exhaustion of feasible slip systems takes place in coarse grains of the magnesium, which leads to the failure of the sample after moderate deformation. It is known that the temperature anomaly of the yield strength is observed only in single crystals or coarse-grain materials [19,20,23,24]. The critical size of grains (d), below which the σYS(T) anomaly is not observed, is different in different materials. For example, in Ni3Al intermatallic it is dE 8 μm and d E50 μm in TiAl [19]. It is worth noting that in magnesium the thermal anomaly of yield stress was discovered in single crystals only [20,23,24]. The grain size in our samples is a very few microns. The temperature anomaly of the yield strength observed on such samples can be explained by their sharp texture. In fact, in Fig. 4b, it is seen that some peaks of diffraction patterns are small or absent. As it follows from [25], the predominance of a specific slip system is dependent on the deformation temperature, as well as on the crystallographic texture of magnesium. Previously, in paper [26], it was reported that an annealing hardening occurs in pre-strained magnesium alloy AZ31 but such a phenomenon was not observed in pure Mg. Sample annealing temperatures in this work were 170 1С and 200 1С. It is proposed that pinning of twin boundary by segregated solute atoms results in an increased activation stress for detwinning deformation of AZ31 alloy. In our work, we observe the annealing hardening phenomenon at the same temperatures in pure magnesium despite the fact that segregations of impure atoms are almost absent. It should be noted that the degree of deformation of the samples in paper [26] was very small: the compressing was not
60
A.Yu. Volkov, I.V. Kliukin / Materials Science & Engineering A 627 (2015) 56–60
more than 3.5%. It can be supposed that the increase in the degree of pre-straining would allow the authors of paper [26] to find the increase in strength resulting from the annealing of pure magnesium. It follows from our recent results [9], the present investigation and paper [26] that annealing hardening is a characteristic feature of pre-deformed magnesium and magnesium alloys. 5. Conclusion Thus, we propose in our paper an original technique for the successful hydrostatic extrusion of magnesium at room temperature. The distinction of our method consists in using the shell to increase the formability of magnesium. The shell forms a field of compression stresses around the sample during deformation. This method allows decreasing in temperature of magnesium hydroextrusion and increasing the degree of its deformation by a single pass. After cold extrusion with true deformation to ε1 3.2 the magnesium rod has grain size near 3 mm and sharp radial texture. The plasticity of magnesium in the obtained structural state is very high and attains 19% elongation to failure. Moreover, it was experimentally shown in the present work that annealing of pre-deformed magnesium at 150 1С leads to increase in strength and plasticity. This result is explained from the viewpoint of self-blocking of dislocations, which dropped into the deepest valley of the potential relief by the thermal activated processes. The annealing hardening phenomenon can be of interest for practical application as the method for improving of mechanical properties of pre-deformed magnesium alloys. Acknowledgements The authors thank engineer Alexandrov A.V. whose knowledge and technical experience were indispensable in carrying out this work. The work was carried out as a part of the theme "Deformation" No.
01201463327 and partially supported by Ural Branch of the RAS (Project no.12-U-2-1007). References [1] E.F. Emley, Principles of Magnesium Technology, Pergamon Press, Oxford, 1966. [2] N. Ono, R. Nowak, S. Miura, Mater. Lett. 58 (2003) 39–43. [3] A. Yamashita, Z. Horita, T.G. Langdon, Mater. Sci. Eng. A 300 (2001) 142–147. [4] S. Biswas, S.S. Dhinwal, S. Suwas, Acta Mater. 58 (2010) 3247–3261. [5] H. Somekawa, T. Mukai, Scr. Mater. 53 (2005) 1059–1064. [6] S. Suwas, G. Gottstein, R. Kumar, Mater. Sci. Eng. A 471 (2007) 1–14. [7] I.C. Choi, D.H. Lee, B. Ahn, K. Durst, M. Kawasaki, T.G. Langdon, J. Jang, Scr. Mater. 94 (2015) 44–47. [8] H.L.D. Pugh, Mechanical Behavior of Materials under Pressure, Applied Science Publishers, London, 1971. [9] B.I. Kamenetskii, Yu.N. Loginov, A.Yu. Volkov, Zagotovitelnyie proizvodstva v mashinostroenii/Заготовительные производства в машиностроении (Prod. Technol. Mach. Eng.) 9 (2013) 17–23 (in Russian). [10] J. Swiostek, J. Goken, D. Letzig, K.U. Kainer, Mater. Sci. Eng. A 424 (2006) 223–229. [11] S. Yi, I. Schestakow, S. Zaefferer, Mater. Sci. Eng. A 516 (2009) 58–63. [12] N.B. Tork, N. Pardis, R. Ebrahimi, Mater. Sci. Eng. A 560 (2013) 34–39. [13] H. Kitahara, F. Maruno, M. Tsushida, S. Ando, Mater. Sci. Eng. A 590 (2014) 274–280. [14] T.G. Langdon, Acta Mater 61 (2013) 7035–7059. [15] H.J. Choi, Y. Kim, J.H. Shin, D.H. Bae, Mater. Sci. Eng. A 527 (2010) 1565–1570. [16] J.A. Sharon, Y. Zhang, F. Mompiou, M. Legros, K.J. Hemker, Scr. Mater. 75 (2014) 10–13. [17] R. Gehrmann, M.M. Frommert, G. Gottstein, Mater. Sci. Eng. A 395 (2005) 338–349. [18] Y. Wang, H. Choo, Acta Mater. 81 (2014) 83–97. [19] O.V. Antonova, B.A. Greenberg, A.Yu. Volkov, Gold Bull. 41 (2008) 326–335. [20] S. Ando, N. Harada, M. Tsushida, H. Kitahara, H. Tonda, Key Eng. Mater. 345–346 (2007) 101–107. [21] R.Z. Valiev, A.V. Sergueeva, A.K. Mukherjee, Scr. Mater. 49 (2003) 669–674. [22] B. Bhattacharya, M. Niewczas, Philos. Mag. 91 (2011) 2227–2247. [23] J.F. Stohr, J.P. Poirier, Philos. Mag. 25 (1972) 1311–1318. [24] B.A. Greenberg, O.V. Antonova, A.M. Vlasova, M.A. Ivanov, Crystallogr. Rep. 57 (2012) 541–549. [25] N.V. Dudamell, I. Ulacia, F. Galvez, S. Yi, J. Bohlen, D. Letzig, I. Hurtado, M.T. Perez-Prado, Acta Mater. 59 (2011) 6949–6962. [26] Y. Xin, X. Zhou, H. Chen, J.F. Nie, H. Zhang, Y. Zhang, Q Liu, Mater. Sci. Eng A 594 (2014) 287–291.
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Improving the mechanical properties of pure magnesium through cold hydrostatic extrusion and low-temperature annealing A.Yu. Volkov n, I.V. Kliukin Institute of Metal Physics, Ural Branch, Russian Academy of Sciences, 18 S. Kovalevskaya St., Ekaterinburg 620990, Russia
art ic l e i nf o Article history: Received 19 September 2014 Received in revised form 17 December 2014 Accepted 24 December 2014 Available online 7 January 2015 Keywords: Mechanical characterization Magnesium Bulk deformation Thermal anomaly
a b s t r a c t A way to produce ultrafine grained structure with d 3 μm and high plasticity in pure magnesium through the original technique of hydrostatic extrusion at room temperature has been demonstrated. The microstructure and the mechanical properties of the extruded magnesium rods have been investigated. The evolution of mechanical properties of the samples after treatments up to 450 1C has been analyzed. It has been shown that low-temperature treatments lead to some increase in strength and plasticity of pre-deformed magnesium. The annealing hardening phenomenon is explained by the thermally activated processes of rearrangement of the dislocation structure and blocking of dislocations. & 2015 Elsevier B.V. All rights reserved.
1. Introduction It is well known that at temperatures below 225 1C the workability of pure magnesium and magnesium alloys is limited or even not possible [1]. Poor room-temperature formability of magnesium is connected with its hexagonal close packed (hcp) crystalline structure in which the basal slip system dominates room-temperature deformation [2]. At high temperature formation, the prismatic and pyramidal slip systems are activated and thus the plasticity of magnesium increases. Grain refinement and texture modification are both considered to contribute to the improvement of plasticity of magnesium. As a rule, different severe plastic deformation (SPD) techniques are used for this purpose [3,4]. A detailed review of studies, which used the well-known SPD method of equal channel angular extrusion (ECAE) for deformation of magnesium, was given in [4]. It was shown that the deformation of magnesium by the ECAE method is much impeded at lower temperatures. Nevertheless, over and over again, researchers have attempted to deform magnesium at room temperature in order to obtain ultrafine-grain structure (UFG) with high strength and ductility. A UFG material is stronger than one in a coarse grained state because of grain size hardening according to the Hall–Petch relationship [2,5]. In addition, UFG magnesium has a better ductility as well as a low ductile to brittle transition temperature; thus, their formability at room temperature can be improved [6].
n
Corresponding author. E-mail address: [email protected] (A.Yu. Volkov).
http://dx.doi.org/10.1016/j.msea.2014.12.104 0921-5093/& 2015 Elsevier B.V. All rights reserved.
Undesirable cracking of magnesium during deformation may be avoided even at room temperature if processing by highpressure torsion is used [7]. In this case, formation of the magnesium sample occurs within the field of compression stresses. As a result, pores, defects, and cracks existing in the material cannot develop. Therefore, it is necessary to provide all-around compression stresses for the successful deformation of materials with poor plasticity. Of course, it was well known before [8]. Recently, we used this method for deformation of bulk magnesium workpiece at room temperature [9]. For this purpose a magnesium cylinder of 20 mm diameter and 30 mm height was placed in a thickwalled copper container. The deformation was carried out by squeezing this “assembly” at room temperature. Such a method of deformation generates high compression stresses on the side surfaces of the magnesium workpiece. They are caused by the response of the container walls to the extension of the sample during pressing. At the final stage of the experiment the remains of the copper container were cut and removed. After processing by squeezing into the container the magnesium sample was significantly shorter and bigger in diameter, having the form of a puck. The degree of true (or logarithmic) deformation (ε) of magnesium reached ε 0.85. No sign of destruction of the magnesium sample was observed. As the structure became virtually isotropic and the grain size decreased to 5 mm, the plasticity of magnesium increased. A very interesting phenomenon was revealed during these experiments. The yield strength of the pre-deformed magnesium samples does not drop but increases as a result of low-temperature annealing. Obviously, this phenomenon is of both fundamental and practical interest. In addition, obviously, this result needs verifying. These experiments showed that the use of a container increases the formability of magnesium at room temperature.
A.Yu. Volkov, I.V. Kliukin / Materials Science & Engineering A 627 (2015) 56–60
This is caused by the formation of compression stresses around the sample during deformation. It seems interesting to use a container to increase the formability of magnesium with other methods of SPD. For example, magnesium placed in a shell can be deformed by hydroextrusion. A great variety of sources did not produce any description of successful experiments of hydrostatic extrusion of pure magnesium at room temperature. Moreover, magnesium alloys are very difficult to be cold extruded, although they have better formability than pure magnesium. As shown in [10], hydrostatic extrusion of magnesium alloys at temperatures below 100 1С leads to the destruction of workpieces. The purpose of this study was to develop a hydrostatic extrusion technique for magnesium at room temperature, followed by an investigation of the structure and mechanical properties of the samples in the obtained structural state.
2. Material and method As a material for experiment, an ingot of commercially pure magnesium (99.98%) was chosen. The investigation has shown that the ingot consists of very large grains (Fig. 1). The largest grains are disposed in the centre of the ingot and they are more than 30 mm in length and near 5 mm in width. Such coarse grained structure is typical for magnesium ingot (see, for example, [11]). Cylindrical workpieces of 10 mm in diameter and 70 mm in length for hydrostatic extrusion experiments were cut out of this ingot. The result of our experiment using a conventional technique is shown in Fig. 2a. We are far from being the first who observed magnesium workpieces cracking after SPD at room temperature (see, for example, [10–13]). To prevent cracking of the magnesium during its flowing out from the die, we developed an original method (Fig. 2b). We placed a magnesium workpiece (1) of 10 mm in diameter in a shell (2) of 20 mm in diameter. Hydrostatic
Fig. 1. Macrostructure of initial as-cast magnesium.
57
extrusion was made through a die (3) of 10 mm at room temperature. The distinction of this method consists in the material of the shell. As was described above, we carried out successful squeeze of magnesium workpiece placed in a copper container before [9]. Nevertheless, this method had a disadvantage. When the deformation process was finished, we had to cut the container to take the sample out of it. In our experiments on hydroextrusion, we replaced copper by another material. It is clearly seen in Fig. 2b that the shell (2) breaks down once hydrostatic extrusion is completed. It provides us an easy way to extract the sample out of the shell. The material of the shell is specialized knowledge (“know-how” of the Institute of Metal Physics of the Russian Academy of Sciences). As the result of the employment of this technique, a long magnesium rod of 5 mm in diameter was formed without any flaws (Fig. 2b). Microstructure observations were realized by means of a scanning electron microscopy (SEM) instrument QUANTA 200 FEI. Tensile mechanical tests were carried out at room temperature using an Instron test instrument. The length of the samples was 50 mm and the strain rate was 2 10 3 s 1. Five samples were tested per each experimental point. All of the heat treatments were carried out in evacuated glass ampoules. Samples were subjected to annealing in the temperature interval from 100 to 450 1C for 6 h followed by cooling in water.
3. Results 3.1. Microstructure of the extruded rod After the first step of hydroextrusion processing we observed grains of dE 3–5 μm in the microstructure of Ø5 mm rod (Fig. 3a). The largest grains attain 15 μm in size; they are noted in the central part of this rod. The magnesium rod in the obtained structural state has high plasticity. For example, samples for mechanical tests of 2 mm in diameter were made by hydroextrusion processing of magnesium rod Ø5 mm through two passes at room temperature without any shell. First, rods of 3 mm diameter were obtained. Then, they underwent hydrostatic extrusion again through a die of Ø2 mm. So, the total degree of true deformation (ε) of this magnesium sample was ε 3.2. After deformation to Ø2 mm, the structure becomes homogenous with grain size dE2–3 μm (Fig. 3b). As it can be seen, in this case the refinement of the initial cast structure by more than three orders of magnitude takes place (we may compare Fig. 1 with Fig. 3b). Small grain formation in highly deformed magnesium is caused by dynamic recrystallization [1,13,14]. The values of grain sizes obtained can be compared to the minimal values cited in the literature [2–5]. An average grain size of 6–8 mm was achieved by ECAE of commercially pure magnesium at 250 1С after four passes [6]. As was found in [4], the minimum average grain size obtained after ECAE processing was not less than 2 mm. 3.2. Mechanical properties
Fig. 2. Appearance of magnesium after hydrostatic extrusion at room temperature: (a) a destroyed workpiece after experiment using conventional technique and (b) a successful experiment with workpiece (1) which was placed in a self-destroying shell (2) and passed through the die (3).
True stress (σ) versus elongation (δ) responses of magnesium samples with 2 mm in diameter are plotted in Fig. 4a. It is seen that magnesium samples after hydrostatic extrusion at ε 3.2 following the developed technique have a very high plasticity. For example, the elongation-to-failure of such samples is δ ¼19% (curve 1 in Fig. 4a). The mechanical properties of samples that we obtained by our technique can be compared to the mechanical properties of those reported in the literature. For example, in the rods of 2 mm in
A.Yu. Volkov, I.V. Kliukin / Materials Science & Engineering A 627 (2015) 56–60
True stress, σ(MPa)
58
2
200 180 160
1
140 120 100 80 0
5
10
15
20
Elongation,δ (%)
(100)
(002)
Intensity
(101)
(110) (200)(112) (201)
(102)
(103)
2 1
20
30
40
50
60
70
80
2θ, degree Fig. 4. Tensile diagrams (a) and θ–2θ XRD scans obtained from the cross section (b) of magnesium samples Ø2 mm with true deformation ε 3.2. Sample (1) was deformed by three passes through cold hydroextrusion and sample (2) was drawn at room temperature after one pass of cold extrusion.
Fig. 3. Microstructure of magnesium samples after hydroextrusion at room temperature: (a) a rod with 5 mm diameter extruded into self-destroying shell and (b) a rod with 2 mm diameter obtained by passes through dies with 5 mm, 3 mm and 2 mm diameters.
diameter the grain size is 2–3 mm, which virtually coincides with the value of grain size (dE 5 mm) obtained on samples of pure magnesium as the result of ECAE-processing in the work [5]. The strain rates in our work and in [5] are also close to each other. As can be seen from Fig. 4a (curve 1), the yield strength of our samples is σYS ¼ 98 MPa and the ultimate tensile strength is σUTS ¼182 MPa. In accordance with the results of [5], the strength properties of their samples were σYS ¼141 MPa, σUTS ¼ 167 MPa. The elongation-to-failure of the rods with Ø2 mm is δ ¼19%, and for samples from [5], δ ¼ 16%. Thus, though grain sizes in the samples are nearly equal, their yield strengths differ from each other by more than 40%. It should be noted that σYS ¼102 MPa was obtained in [5] on the samples with the grain size d E55 mm. Thus, if yield strengths of the extruded rods and the samples from the paper [5] are nearly equal, their grain sizes differ from each other by one order of magnitude. On the one hand, such behavior does not match the well-known Hall–Petch relationship according to which σYS E σ0 þ kd 0.5. On the other hand, different researches [2,15,16] have already registered some deviations from the correlation between the yield strength (σYS) and the grain size (d) for a polycrystalline magnesium. Moreover, we can find some manuscripts that describe magnesium samples with different grain sizes demonstrating very close tensile strength values (for example, results from [5] can be compared to the ones from [16]). One more example was given in [16]: σYS of magnesium with UFG structure
(d E200 nm) was exceptionally low, but magnesium samples with grain sizes larger than 1 μm had higher yield strength. To understand the deformation behavior of magnesium, the texture of the sample must be taken into account [17]. The θ–2θ XRD-scan for the sample extruded at room temperature is shown in Fig. 4b (the lower diffractogram). The X-ray diffraction pattern of the cross-section actually reveals the absence of the (002) peak. In contradiction, in this X-ray diffraction pattern, the (100) and (101) peaks have sharp form and high intensity. Such a picture of X-ray reflections is explained by the parallel alignment of the basal planes in the all crystallites along the direction of extrusion. Therefore, a sharp radial texture is formed in the sample under extrusion at room temperature. We tested the effect of texture on the mechanical properties of magnesium. For this aim, the rod 5 mm in diameter obtained by hydrostatic extrusion of a workpiece within the shell was cold drawn to a diameter of 2 mm. The total degree of true deformation of the samples after one pass through the hydroextrusion followed by cold drawing was ε 3.2. This value is equal to true deformation of the extruded sample Ø2 mm. The grain size in the sample is also near 3 μm. Therefore, the grain sizes in both samples are almost the same. The X-ray diffraction pattern 2 in Fig. 4b clearly shows that the texture of the sample after the extrusion followed by drawing is not so sharp. For example, when shot from the crosssection area, the reflections from the basal planes give a visible peak (002). The strength properties of these samples are σYS ¼ 127 MPa, σUTS ¼195 MPa and the elongation-to-failure δ ¼ 10% (curve 2 in Fig. 4b). Therefore, the crystallographic texture might contribute to the dissimilarity of the mechanical behavior of magnesium samples with close grain sizes. A very thorough study on this subject has been published recently in [18]. We have also studied the thermal stability of the mechanical properties of the samples after their annealing in the temperature interval from 100 to 450 1C for 6 h, followed by cooling in water. It is well seen that annealing at temperature below 200 1С leads to
A.Yu. Volkov, I.V. Kliukin / Materials Science & Engineering A 627 (2015) 56–60
Fig. 5. The dependence of the yield strength (σYS), ultimate tensile strength (σUTS) and the elongation-to-failure (δ) on the holding temperature for 6 h of predeformed magnesium rods with 2 mm in diameter. (a) Samples were subjected to cold hydrostatic extrusion by three passes at room temperature and (b) samples were obtained by one pass of hydroextrusion followed by cold drawing.
some increase in strength and plasticity (Fig. 5). Small increase in plasticity can be explained by the operation of recovery processes when, in the course of a low-temperature annealing, the level of elastic stresses is decreased. However, the increase in strength resulting from the annealing of highly deformed material is an anomaly, which needs explaining.
4. Discussion It should be noted that the increase in yield strength (σYS) along with the increase in test temperature (T) is a characteristic feature of many intermetallics: Ni3Al, TiAl etc. (see Refs. [19] and references herein). It has been assumed to be an established fact that the anomalous temperature dependence of σYS(T) is due to the blocking of dislocations. As far as pure metals are concerned, the thermal anomaly of the yield strength was observed before only in hcp metals (for example, in magnesium [20] and titanium [21]). Such a behavior of these metals is explained by a limited number of slip systems in crystals with hcp structure and different slip systems are subject to different amounts of Peierls barrier [16]. For example, there is a well-marked difference between the light basic slip and the much more difficult pyramidal and prismatic slip in magnesium [1,22]. In the papers [20,23], thermal mechanical tests of Mg single crystals were carried out. It was found that, for a [0001] orientation of the crystal, the yield strength in the interval between 100 and 150 1С is higher than at room temperature. For a [112̄ 0] orientation of the magnesium crystal, the maximum yield strength was observed at a temperature of about 100 1С. The anomalous temperature trend of the σYS(T) is explained by the thermally activated processes of rearrangement of the dislocation
59
structure and blocking of dislocations. The blocked (cþa) dislocation configurations located on second-order pyramidal planes were observed in the region of this temperature anomaly. There is no such blocked configuration after deformation at a higher temperature. It should be noted once more that, in works [20,23], mechanical tests were carried out on samples heated up to the test temperature. Our experiments are different because preliminarily deformed samples of magnesium after annealing were cooled and mechanical tests were carried out at room temperature. Nevertheless, the increase in strength after low-temperature annealing is observed in Fig. 5. Earlier, such effect was predicted theoretically after the investigation of dislocation structure of magnesium single crystals [24]. The possibility of the rise of the strengths was explained in this work by the processes of self-blocking of dislocations upon a low-temperature annealing of preliminarily deformed magnesium single crystals. It is shown that Peierls relief of the edge (с þ а) dislocation has two valleys at the second-order pyramidal plane. When thermal activation is present, the dislocation drops into the deepest valley of the potential relief. This position of the edge (cþa) dislocation has the lowest energy and therefore such configuration is blocked. The model presented explains well the dislocation extension along a selected direction, which is observed after heating of pre-deformed magnesium samples without external stress. According to this investigation [24], the self-blocking of dislocations in magnesium can be observed in a temperature interval lower than 150 1С. The thermal dependences of strength properties in Fig. 5 correspond to the results presented in the work [24]. As a result of the above discussion and taking into account the Hall–Petch relationship we can conclude from Fig. 5 that the grain size in the magnesium samples does not change in the course of low-temperature anneals for 6 h. Consequently, the structure formed by the deformation to ε1 3.2 is quite stable. The decrease in yield strength of samples starts after anneals at temperatures higher than 250 1С. Consequently, it is in this temperature range the coarsening of the grain structure begins. A sharp drop of elongation after annealing at temperatures higher than 300 1С confirms this conclusion. Indeed, a rapid exhaustion of feasible slip systems takes place in coarse grains of the magnesium, which leads to the failure of the sample after moderate deformation. It is known that the temperature anomaly of the yield strength is observed only in single crystals or coarse-grain materials [19,20,23,24]. The critical size of grains (d), below which the σYS(T) anomaly is not observed, is different in different materials. For example, in Ni3Al intermatallic it is dE 8 μm and d E50 μm in TiAl [19]. It is worth noting that in magnesium the thermal anomaly of yield stress was discovered in single crystals only [20,23,24]. The grain size in our samples is a very few microns. The temperature anomaly of the yield strength observed on such samples can be explained by their sharp texture. In fact, in Fig. 4b, it is seen that some peaks of diffraction patterns are small or absent. As it follows from [25], the predominance of a specific slip system is dependent on the deformation temperature, as well as on the crystallographic texture of magnesium. Previously, in paper [26], it was reported that an annealing hardening occurs in pre-strained magnesium alloy AZ31 but such a phenomenon was not observed in pure Mg. Sample annealing temperatures in this work were 170 1С and 200 1С. It is proposed that pinning of twin boundary by segregated solute atoms results in an increased activation stress for detwinning deformation of AZ31 alloy. In our work, we observe the annealing hardening phenomenon at the same temperatures in pure magnesium despite the fact that segregations of impure atoms are almost absent. It should be noted that the degree of deformation of the samples in paper [26] was very small: the compressing was not
60
A.Yu. Volkov, I.V. Kliukin / Materials Science & Engineering A 627 (2015) 56–60
more than 3.5%. It can be supposed that the increase in the degree of pre-straining would allow the authors of paper [26] to find the increase in strength resulting from the annealing of pure magnesium. It follows from our recent results [9], the present investigation and paper [26] that annealing hardening is a characteristic feature of pre-deformed magnesium and magnesium alloys. 5. Conclusion Thus, we propose in our paper an original technique for the successful hydrostatic extrusion of magnesium at room temperature. The distinction of our method consists in using the shell to increase the formability of magnesium. The shell forms a field of compression stresses around the sample during deformation. This method allows decreasing in temperature of magnesium hydroextrusion and increasing the degree of its deformation by a single pass. After cold extrusion with true deformation to ε1 3.2 the magnesium rod has grain size near 3 mm and sharp radial texture. The plasticity of magnesium in the obtained structural state is very high and attains 19% elongation to failure. Moreover, it was experimentally shown in the present work that annealing of pre-deformed magnesium at 150 1С leads to increase in strength and plasticity. This result is explained from the viewpoint of self-blocking of dislocations, which dropped into the deepest valley of the potential relief by the thermal activated processes. The annealing hardening phenomenon can be of interest for practical application as the method for improving of mechanical properties of pre-deformed magnesium alloys. Acknowledgements The authors thank engineer Alexandrov A.V. whose knowledge and technical experience were indispensable in carrying out this work. The work was carried out as a part of the theme "Deformation" No.
01201463327 and partially supported by Ural Branch of the RAS (Project no.12-U-2-1007). References [1] E.F. Emley, Principles of Magnesium Technology, Pergamon Press, Oxford, 1966. [2] N. Ono, R. Nowak, S. Miura, Mater. Lett. 58 (2003) 39–43. [3] A. Yamashita, Z. Horita, T.G. Langdon, Mater. Sci. Eng. A 300 (2001) 142–147. [4] S. Biswas, S.S. Dhinwal, S. Suwas, Acta Mater. 58 (2010) 3247–3261. [5] H. Somekawa, T. Mukai, Scr. Mater. 53 (2005) 1059–1064. [6] S. Suwas, G. Gottstein, R. Kumar, Mater. Sci. Eng. A 471 (2007) 1–14. [7] I.C. Choi, D.H. Lee, B. Ahn, K. Durst, M. Kawasaki, T.G. Langdon, J. Jang, Scr. Mater. 94 (2015) 44–47. [8] H.L.D. Pugh, Mechanical Behavior of Materials under Pressure, Applied Science Publishers, London, 1971. [9] B.I. Kamenetskii, Yu.N. Loginov, A.Yu. Volkov, Zagotovitelnyie proizvodstva v mashinostroenii/Заготовительные производства в машиностроении (Prod. Technol. Mach. Eng.) 9 (2013) 17–23 (in Russian). [10] J. Swiostek, J. Goken, D. Letzig, K.U. Kainer, Mater. Sci. Eng. A 424 (2006) 223–229. [11] S. Yi, I. Schestakow, S. Zaefferer, Mater. Sci. Eng. A 516 (2009) 58–63. [12] N.B. Tork, N. Pardis, R. Ebrahimi, Mater. Sci. Eng. A 560 (2013) 34–39. [13] H. Kitahara, F. Maruno, M. Tsushida, S. Ando, Mater. Sci. Eng. A 590 (2014) 274–280. [14] T.G. Langdon, Acta Mater 61 (2013) 7035–7059. [15] H.J. Choi, Y. Kim, J.H. Shin, D.H. Bae, Mater. Sci. Eng. A 527 (2010) 1565–1570. [16] J.A. Sharon, Y. Zhang, F. Mompiou, M. Legros, K.J. Hemker, Scr. Mater. 75 (2014) 10–13. [17] R. Gehrmann, M.M. Frommert, G. Gottstein, Mater. Sci. Eng. A 395 (2005) 338–349. [18] Y. Wang, H. Choo, Acta Mater. 81 (2014) 83–97. [19] O.V. Antonova, B.A. Greenberg, A.Yu. Volkov, Gold Bull. 41 (2008) 326–335. [20] S. Ando, N. Harada, M. Tsushida, H. Kitahara, H. Tonda, Key Eng. Mater. 345–346 (2007) 101–107. [21] R.Z. Valiev, A.V. Sergueeva, A.K. Mukherjee, Scr. Mater. 49 (2003) 669–674. [22] B. Bhattacharya, M. Niewczas, Philos. Mag. 91 (2011) 2227–2247. [23] J.F. Stohr, J.P. Poirier, Philos. Mag. 25 (1972) 1311–1318. [24] B.A. Greenberg, O.V. Antonova, A.M. Vlasova, M.A. Ivanov, Crystallogr. Rep. 57 (2012) 541–549. [25] N.V. Dudamell, I. Ulacia, F. Galvez, S. Yi, J. Bohlen, D. Letzig, I. Hurtado, M.T. Perez-Prado, Acta Mater. 59 (2011) 6949–6962. [26] Y. Xin, X. Zhou, H. Chen, J.F. Nie, H. Zhang, Y. Zhang, Q Liu, Mater. Sci. Eng A 594 (2014) 287–291.