- Email: [email protected]
Ultra-severe plastic deformation: Evolution of microstructure, phase transformation and hardness in immiscible magnesium-based systems
Materials Science & Engineering A 701 (2017) 158–166
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Ultra-severe plastic deformation: Evolution of microstructure, phase transformation and hardness in immiscible magnesium-based systems
MARK
⁎
Kaveh Edalatia,b, , Ryoko Uehirob, Keisuke Fujiwarab, Yuji Ikedac, Hai-Wen Lia,d, Xavier Sauvagee, Ruslan Z. Valievf,g, Etsuo Akibaa,d,h, Isao Tanakac,i, Zenji Horitaa,b a
WPI, International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, Fukuoka 819-0395, Japan Department of Materials Science and Engineering, Faculty of Engineering, Kyushu University, Fukuoka 819-0395, Japan c Center for Elements Strategy Initiative for Structure Materials (ESISM), Kyoto University, Sakyo, Kyoto 606-8501, Japan d International Research Center for Hydrogen Energy, Kyushu University, Fukuoka 819-0395, Japan e Normandie Université, UNIROUEN, INSA Rouen, CNRS, Groupe de Physique des Matériaux, 76000 Rouen, France f Institute of Physics of Advanced Materials, Ufa State Aviation Technical University, Ufa, Russia g Laboratory for Mechanics of Bulk Nanomaterials, Saint Petersburg State University, Saint Petersburg, Russia h Department of Mechanical Engineering, Faculty of Engineering, Kyushu University, Fukuoka 819-0395, Japan i Department of Materials Science and Engineering, Kyoto University, Kyoto 606-8501, Japan b
A R T I C L E I N F O
A B S T R A C T
Keywords: Severe plastic deformation (SPD) Ultrafine-grained (UFG) materials Nanostructured materials DFT calculations Magnesium alloys Phase transition
Although severe plastic deformation (SPD) alters the microstructure and phase transformation at the early stages of straining, the microstructural features finally saturate to the steady states at large shear strains. However, from the atomic point of view, to achieve the steady state in immiscible systems with positive heat of mixing, the minimum shear strain should be so high that the thickness of sheared phases becomes comparable to one atomic distance. In this study, ultrahigh shear strains up to ~70,000 are introduced in different Mg-based immiscible systems by high-pressure torsion (HPT) method for up to 1500 turns. New metastable phases are formed in most of the selected magnesium alloys by ultra-SPD, in good agreement with the first-principles calculations. However, the microstructural/structural saturation hardly occurs in many alloys even at ultrahigh strains. The materials processed by ultra-SPD exhibit unique hardness-strain and tensile behaviors which cannot be observed after conventional SPD.
1. Introduction Severe plastic deformation (SPD) techniques [1,2] are currently used to achieve two main microstructural and structural features: (i) refine grains to enhance the mechanical and functional properties [3], and (ii) control phase transformation and mechanical alloying [4]. Numerous studies showed that despite the microstructure and phase evolutions at the early stages of straining, the microstructural and structural features finally become saturated at large strain, where a steady state is achieved [5]. Although the occurrence of steady states at large strains is expected in single-phase metals [6] because of the contribution of dynamic recovery [7], dynamic recrystallization [8], grain-boundary rotation [9] and/or grain-boundary migration [5], it is not still well understood why and when a steady state is reached in multiple-phase materials with immiscible phases [10,11]. When pure shear strain is introduced in a multiple-phase system with immiscible phases, the phases are elongated in the shear direction
⁎
and their thicknesses are continuously reduced. Therefore, to achieve the saturation in an immiscible system under pure shear deformation mode, the strain should be increased so that the thicknesses of sheared phases are reduced to the sub-nanometer level or ideally to one atomic distance. For example, the shear strain should be at least 10,000 to achieve the saturation in an immiscible system with initial phase sizes of 10 µm. However, since the fragmentation, rotation and/or sliding of phases occur practically during the SPD process and the co-deformation of phases does not follow an ideal and homogenous pattern, the shear strain should be even higher than the one calculated above. Among numerous SPD methods developed in recent decades [1–3], the high-pressure torsion (HPT) provides a unique opportunity for the fundamental study of the behavior of different materials under very large shear strain [12,13]. Since the material in HPT processing is constrained between two rotating anvils under high pressure, the shear strain (γ = 2πrN / h , γ: shear strain, r: distance from disc center, N: number of HPT turns, h: thickness of disc [2]) can be significantly
Corresponding author at: WPI, International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, Fukuoka 819-0395, Japan. E-mail address: [email protected] (K. Edalati).
http://dx.doi.org/10.1016/j.msea.2017.06.076 Received 27 February 2017; Received in revised form 18 June 2017; Accepted 19 June 2017 Available online 22 June 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
500 and 1500 turns. For the Mg-Li alloy, the HPT was conducted on 10 mm diameter and ~0.8 mm thick discs for N = 0 (as-cast condition), 5, 50, 200 and 1000 turns. The processing speed was ω = 1 rpm, and the processing temperature was room temperature for all selected compositions. The temperature rise was less than 100 K even after processing for 1500 turns because of slow rotation speed and due to the small size of sample (as the heat source) compared to the size of anvils and other large metallic parts of the HPT facility (as the heat sinks). Details concerning the temperature rise during HPT were discussed in Refs. [20,21]. Following the HPT processing, the discs were first polished and the Vickers microhardness was measured in four radial directions from the disc center to periphery. The hardness could not be measured for the Mg-V-based systems because of the presence of many cracks. Second, the chemical homogeneity was examined at the middle of discs using scanning electron microscopy (SEM) equipped with energy dispersive X-ray spectroscopy (EDS). Third, the phase transformations were studied by X-ray diffraction (XRD) analyses using the Cu Kα radiation. Fourth, for examination of microstructure, transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM) were conducted using the accelerating voltages of 300 kV and 200 kV, respectively. Thin foils were prepared from the middle or edge of the discs using a focused ion beam system. For making the TEM foils from the Mg-Li alloy, a twin-jet electro-chemical polishing system with a solution of 5 vol% HClO4, 25 vol% C3H5(OH)3 and 70 vol% C2H5OH was used at 263 K. Fifth, for examination of tensile properties, tensile specimens with gauge dimensions of 1.5×0.3×0.7 mm3 were cut from the discs at 2–2.5 mm away from the disc center and pulled to failure using an initial strain rate of ε ̇ = 4×10−3 s−1. The only material that could be examined by tensile test was Mg-Li because of the absence of cracks.
increased with increasing the numbers of HPT turns without any limitation. Moreover, the extent of strain is controllable and the contamination of sample is minor during HPT, and thus it is basically more appropriate than ball milling for investigation of immiscible systems [14]. Although the HPT method has been widely used in the past to control the phase transformations in different kinds of metallic and nonmetallic materials [15], the significance of ultra-high shear strains (N = 1000 turns or higher) on phase transformations, mechanical alloying and properties has not received appreciable attention, except in a recent publication on the immiscible Mg-Zr system [16]. In this study, thus, different immiscible Mg-based systems, which are of interest for hydrogen storage or structural applications, are processed by HPT method for up to 1500 turns, and ultrahigh shear strains (up to 70,000) are introduced in the materials. Mechanical mixing of elements at the atomic scale and formation of new phases are reviewed together with the evolution of microstructure and mechanical properties. 2. Experimental materials and procedures Several immiscible Mg-based systems were selected for this study, as shown in Table 1: (1) Mg − 25% V − 25% Sn, (2) Mg − 25% V − 25% Pd, (3) Mg − 25% V − 25% Ni, (4) Mg − 50% Zr, (5) Mg − 17% Ni − 17% Sn, (6) Mg − 17% Ni − 17% Pd, and (7) Mg − 25% Li (all compositions are in atom%). These compositions were originally selected for possible applications as hydrogen storage materials (based on some first-principles calculations in Ref. [17]), although their performance for hydrogen storage is still under investigation. It should be noted that since the evolution of microstructure in Mg-Zr system were studied in Refs. [16] and [18], respectively, only their mechanical properties were reported in this study. The mechanical properties of Mg-Li system after processing by HPT for 5 turns were investigated in Ref. [19], but the material was investigated after processing by HPT for much larger number of turns in this study. For the first three compositions (Mg-V-based systems), MgH2 powders were mixed with the powders of V (99.5%), Sn (99.99%), Pd (99.9%) and Ni (99.99%) with particle sizes smaller than ~44 µm (325 mesh). MgH2 was intentionally used as a source of Mg because it is harder than metallic Mg and can be mixed easier with other selected elements. For the fourth composition (Mg-Zr alloy), Mg and Zr powder mixtures with particle sizes smaller than ~44 µm were used. For the fifth and sixth compositions (Mg-Ni-based systems), the elements were melted under controlled atmosphere and casted into the form of ingots (plates with 200×100×13 mm3 for Mg-Ni-Sn and a rod with 10 mm diameter and 10 mm length for Mg-Ni-Pd). The last composition (MgLi) was prepared by casting under controlled atmosphere followed by extrusion at 373 K with an extrusion speed of 1 mm/s and an extrusion ratio of 25:1 (final diameter: 10 mm). Almost 0.5 g of powder mixtures for Mg-V-based and Mg-Zr systems were processed by HPT under a pressure of 3 GPa for N = 0 (pure compression), 10, 100, 1000 or 1200 turns to produce discs with 14 mm diameter and ~0.8 mm thickness. For the Mg-Ni-based systems, discs with 10 mm diameter and ~0.8 mm thickness were processed by HPT under a pressure of 6 GPa for N = 0 (as-cast condition), 20, 100, 300,
3. Calculation methods In this study, the crystal structure of severely deformed Mg − 17% Ni – 17% Sn and Mg − 17% Ni − 17% Pd alloys were examined by first-principles calculations. The B2-type structure was selected as the initial structure for the structure optimization because this structure was experimentally detected for Mg-Ni-Pd. The initial structure was modeled by the special quasirandom structure (SQS) [22] based on the 3×3×3 (54 atoms) supercell of the unit cell of the B2-type structure. In the SQS model, body-center sites in the B2 unit cells were fully occupied by Mg atoms, while the corner sites were occupied by the same numbers of Mg, Ni, and Sn or Pd as randomly as possible under the periodic boundary conditions of the supercells. The SQS used in this study was obtained by simulated annealing [23,24] as implemented in the CLUPAN code [25,26]. To analyze the optimized atomic positions of the alloys, the radial distribution function (RDF) was calculated using the following equation as linear combinations of the Dirac delta functions, which are broadened by a normal Gaussian function with the standard deviation of 0.05 Å.
g (r ) =
Table 1 Composition of selected systems and their initial form before HPT processing. Composition (in atom%) Mg Mg Mg Mg Mg Mg Mg
− − − − − − −
25% 25% 25% 50% 17% 17% 25%
V − 25% Sn V − 25% Pd V − 25% Ni Zr Ni − 17% Sn Ni − 17% Pd Li
1 1 1 N ρ 4πr 2
N
N
∑ ∑ δ (r − i=1 j≠i
rj − ri ) (1)
In this equation, g(r) is the RDF, r is the distance, N is the number of atoms in a simulated cell (54 in this study), ρ is the density of atoms of the optimized structure, i and j are the indices for atoms in the cell, and ri and ri are the position of the ith and jth atoms, respectively. For calculations, the plane-wave basis projector augmented wave (PAW) method [27] was employed in the framework of density functional theory (DFT) within the generalized gradient approximation of the Perdew-Burke-Ernzerhof form [28] as implemented in the VASP (Vienna Ab initio simulation package) code [29–31]. A plane-wave energy cutoff of 350 eV was used. The 3s electrons for Mg, the 3d and 4s
Initial Form MgH2+V+Sn powder mixture MgH2+V+Pd powder mixture MgH2+V+Ni powder mixture Elemental powder mixture As-cast ingot As-cast ingot Extruded bar
159
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
optimized until the residual forces became less than 1×10−3 eV/Å under zero external stress. The VESTA code was employed for 3D visualization of the crystal structure [33]. 4. Results and discussion 4.1. Microstructure and phase evolution XRD profiles are shown in Fig. 1 for (a) Mg-V-Sn, (b) Mg-V-Pd and (c) Mg-V-Ni systems after processing by HPT for N = 0, 10, 100 and 1200 turns. An examination of Fig. 1 shows that while the peaks for starting powders are clearly visible after N = 0, their intensity gradually decreases and peaks corresponding to stable and new metastable phases gradually appear. Thermodynamically stable Mg2Sn and Mg2Ni intermetallics are formed at lower shear strains (N = 10 or 100) when compared to the metastable phases because of high thermodynamic driving force and negative heat of mixing for the formation of these stable phases [17]. Close examination of the XRD profiles suggests that several new metastable phases are formed after HPT processing for N = 1200: B2-type structure with a lattice parameter of 0.360 nm in Mg-VSn, B2-type structure with a lattice parameter of 0.316 nm in Mg-V-Pd, bcc with a lattice parameter of 0.291 nm in Mg-V-Pd, and a single-phase bcc with a lattice parameter of 0.325 nm in Mg-V-Ni. The XRD profiles in Fig. 1 clearly shows that even immiscible systems with positive heat of mixing such as Mg-V-based systems can be mixed at the atomic scale by ultra-SPD and new phases can be formed. The formation of these metastable phases should be due to the formation of high density of lattice defects (point defects and grain boundaries) and their effect on the dynamic stability of phases. Earlier theoretical and experimental studies confirmed that point defects [34] and grain boundaries [35] can change the effective internal energy and influence the stability of phases. The differences between the XRD profiles after N = 100 and after N = 1200 clearly confirms that the structural features can be totally different after ultra-SPD (N = 1200 which corresponds to shear strains up to ~70,000) when compared to those after conventional SPD. It should be noted that an earlier attempt to synthesize one single-phase bcc structure from the Mg-25%V-25%Ni alloy using high-energy ball milling was not successful [36], while pure bcc phase could be synthesized in this study because of introduction of ultrahigh shear strain. SEM-EDS mappings are shown in Fig. 2 for (a) Mg-V-Sn, (b) Mg-VPd and (c) Mg-V-Ni systems after processing by HPT for N = 0, 10, 100 and 1200 turns. It is apparent that the starting elemental phases are well distinguishable after N = 0. The phases are fragmented and/or elongated in the shearing direction after N = 10 (vertical axis is the shearing direction). The elements are partially dissolved in each other after N = 100 and the dissolution becomes more significant with increasing the numbers of turns to N = 1200. Except for the Mg-V-Ni system, in which a complete mixing of elements and structural saturation occur, the elemental heterogeneity is still visible in the two other systems even after N = 1200. The mixing in the Mg-V-Sn evolves slower than the other two alloys because of the co-existence of several phases with different hardness levels. The current SEM-EDS results, which are well consistent with the XRD profiles, suggest that a real steady state and structural/microstructural homogeneity in the immiscible systems is achievable only at extremely large shear strains (a few orders of magnitude larger than those reported for single-phase metals [37–39] and alloys [40]). Structural and microstructural analyses of Mg-Zr, Mg-Ni-Sn and MgNi-Pd also confirm that the initial phases dissolve in each other by HPT processing and new phases are formed. Details concerning the phase transitions in Mg-Zr were published in Ref. [16], indicating that the hcp-Mg and hcp-Zr phases transform to a supersaturated hcp phase with the lattice parameters of a = 0.321 nm and c = 0.516 nm after HPT processing for N = 1000. Moreover, small amounts of new bcc and fcc phases are formed in the Mg-Zr system after HPT processing [16].
Fig. 1. XRD profiles of (a) Mg-V-Sn, (b) Mg-V-Pd and (c) Mg-V-Ni processed by HPT for various turns.
electrons for Ni, the 5s and 5p electrons for Sn, and the 4d and 5s electrons for Pd were treated as valence. The Brillouin zones were sampled by the Γ-centered 15×15×15 k-point mesh per bcc unit cell, and the Methfessel-Paxton scheme [32] with a smearing width of 0.4 eV was employed. The total energies were minimized until the energy convergences were less than 10−8 eV. Internal atomic positions were 160
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
Fig. 3. Microstructure of Mg-Ni-Sn processed by HPT for various turns, where (a-c) are SEM-SE micrographs, (d) is TEM bright-field image and corresponding SAED pattern and (e) is high-resolution TEM image and corresponding FFT difractogram.
transform to a B2-type structure with a lattice parameter of a = 0.317 nm after HPT processing for N = 1500 turns. Close examination of the as-cast Mg-Ni-Sn alloy by SEM analysis in the secondary electron (SE) mode shows that an eutectic-like lamellar structure, as shown by arrows in Fig. 3(a), forms after casting. However, as shown in Fig. 3(b) and (c), the eutectic structure is destroyed after N = 20 and a homogenous structure is formed after N = 300. Although the eutectic structure should first transform to a network of broken particles at the early stages of straining, these particles could not be detected in this study because the minimum applied strain after N = 20 is too large to save the broken eutectic network. Examination of homogenous structure after HPT processing for N = 1500 by TEM indicates that the material has mainly an amorphous structure. The hollow ring pattern in the selected-area electron diffraction (SAED) pattern of Fig. 3(d) clearly confirm the formation of amorphous structure after N = 1500. Moreover, the random distribution of atoms in an amorphous form is clearly visible in the lattice image of Fig. 3(e). Examination of Fig. 3(e) by fast Fourier transform (FFT) analysis also confirms that the material has mainly an amorphous structure (see the difractogram with a hollow ring pattern). SEM-EDS elemental mapping and XRD profiles for the Mg-Ni-Sn system are shown in Figs. 4 and 5, respectively. Three intermetallics (Mg2Sn, Mg11.92Ni2.32Sn1.75 and Mg1.6NiSn0.3) are visible in the as-cast material, but these intermetallics gradually dissolve in each other and an amorphous-like structure, composed of mainly amorphous structure and partly nanograins, is formed after N = 1500 (see the XRD pattern after N = 1500 with a significant peak broadening). Mixing of Mg, Ni and Sn at the nanoscale after HPT processing for N = 1500 is evident in
Fig. 2. SEM-EDS elemental mapping of (a) Mg-V-Sn, (b) Mg-V-Pd and (c) Mg-V-Ni processed by HPT for various turns. Vertical axis is shearing direction.
Similar metastable bcc, fcc and hcp phases are also formed in the Mg-Ti system after processing by HPT [41]. Detailed analyses of the Mg-Ni-Pd system, which will appear in Ref. [18], confirm that the as-cast materials contains three intermetallic phases, but theses intermetallic phases 161
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
Fig. 4. XRD profiles of Mg-Ni-Sn processed by HPT for various turns.
Fig. 6. (a) High-angle annular dark-field image and corresponding EDS elemental mapping with (b) Mg, (c) Ni and (d) Sn for Mg-Ni-Sn processed by HPT for N = 1500 turns.
Fig. 5. SEM-EDS elemental mapping of Mg-Ni-Sn processed by HPT for various turns. Vertical axis is shearing direction. Fig. 7. SEM-BSED micrographs of Mg-Li processed by HPT for various turns. Vertical axis is shearing direction.
the STEM-EDS mapping of Fig. 6. It should be noted that good atomicscale mixing of elements was also confirmed in the Mg-Ni-Pd system after 1500 turns of HPT by using the atom probe tomography technique [18]. For the Mg-Li alloy, the initial structure contains two phases: Li-rich bcc structure and Mg-rich hcp structure [42]. The evolution of the two phases with increasing the number of HPT turns is shown in the SEM images of Fig. 7 in the back-scatter electron diffraction (BSED) mode. The bright regions in Fig. 7 correspond to the Mg-rich hcp phase and the dark regions correspond to the Li-rich bcc phase. Fig. 7 show that although these two phases cannot be mixed at the atomic scale even after HPT processing for N = 1000, a significant fragmentation of two phases and formation of nanosized phases occur. The variation of grain size with respect to the number of HPT turns is shown in Fig. 8. Note that the average grain sizes were measured using the TEM dark-field images for both phases. The grain size continuously reduces with increasing the number of HPT turns from N = 1 to N = 1000 and a real saturation of grain size (i.e. no change in the grain size with further
increase in the strain) cannot be achieved even after HPT processing for N = 1000, as shown in Fig. 8. It is concluded that unlike the singlephase metals, in which a saturation of grain size at large strains is achieved by dynamic recrystallization or grain boundary migration [5–9], the co-presence of two immiscible phases diminishes the occurrence of dynamic recrystallization or grain boundary migration even for the Mg-Li system with a low melting temperature. 4.2. First-principles calculations Both TEM and XRD analyses confirm that the amorphization after ultra-SPD processing occurs only in the Mg-Ni-Sn system, while the other selected systems can save their crystallinity. To explore the possible reasons for the formation of amorphous phase in this system, firstprinciples calculations were conducted on Mg − 17% Ni − 17% Sn alloy and the results were compared with those for Mg − 17% Ni − 162
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
Fig. 8. Average grain size, measured from TEM dark-field images, against number of HPT turns for Mg-Li.
Fig. 9. (a) Initial and (b) calculated optimized lattice structure of Mg-Ni-Sn and (c) calculated RDFs for ideal bcc structure and optimized structure of Mg-Ni-Sn and Mg-Ni-Pd. Effective lattice parameter (aeff) was calculated as (2 volume per atom)1/3.
17% Pd alloy. Fig. 9 shows (a) the initial model structure and (b) the optimized structure of the Mg-Ni-Sn alloy. The optimized atomic positions are largely deviated from the ideal bcc lattice sites in Fig. 9(b). Fig. 9(c) shows the RDF for the supercell model of Mg-Ni-Sn. Here, for comparison, we also show the RDF for the ideal bcc structure with the same volume as the optimized structure of Mg-Ni-Sn. Moreover, the RDF is given for the Mg-Ni-Pd system, which has a similar composition to the Mg-Ni-Sn alloy, but it does not transform to an amorphous structure after Ultra-SPD processing. The RDFs show clear peaks for MgNi-Pd but no clear and sharp peak for Mg-Ni-Sn when r ≳ aeff (aeff: effective lattice parameter). This indicates that the Mg-Ni-Pd alloy is able to save its B2-type crystal structure while the Mg-Ni-Sn is significantly distorted to a non-crystalline form. The current calculations are in excellent agreement with the experimental data, confirming that if the Mg, Ni, Pd and Sn atoms are mixed at the atomic scale by a process of ultra-SPD, the Mg-Ni-Pd and Mg-Ni-Sn systems should have
Fig. 10. Vickers microhardness against shear strain for (a) Mg-Zr, (b) Mg-Ni-Sn, (c) MgNi-Pd and (d) Mg-Li processed by HPT for various turns.
the bcc-based and amorphous structures, respectively. Earlier calculations also confirmed that if Mg and Zr are mixed at the atomic scale, the hcp, fcc and partially ordered bcc phases can become dynamically 163
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
stable (see Ref. [16] for details). 4.3. Evolution of mechanical properties One standard procedure to examine the occurrence of steady state in HPT processing is to examine the evolution of hardness with respect to shear strain. Two issues should be considered regarding this procedure, which has been widely used by different groups in recent years [37–40]. First, when the hardness does not reach the steady state, the microstructural features are not at the steady state. Nevertheless, the saturation of hardness does not necessarily correspond to the saturation of microstructure. Second, although the occurrence of steady state has been reported in many different kinds of materials since 1935 [12], there have been no attempts to confirm the occurrence of saturation at ultra-high shear strains. The hardness values are plotted against the shear strain in Fig. 10 for (a) Mg-Zr, (b) Mg-Ni-Sn, (c) Mg-Ni-Pd and (d) Mg-Li. Despite some scatterings of hardness data at the early stages of straining, the data points fall reasonably on single curves at large strains in good agreement with numerous earlier publications [37–40]. Four different unpredictable hardness-strain behaviors are seen in Fig. 10.
Fig. 11. Nominal tensile stress again strain for Mg-Li processed by HPT for N = 1000 turns. Inset for appearance of tensile specimens before and after deformation.
of the significant grain refinement, as shown in Figs. 7 and 8. However, the decrease in hardness at large shear strains despite further grain refinement should be due to the contribution of other softening mechanisms. Since the Mg-Li alloy alloys can exhibit grain boundary sliding at relatively low temperatures [52], it is likely that the softening at very large strains is due to the enhancement of grain boundary sliding. To examine the occurrence of grain boundary sliding, tensile tests were conducted on the sample processed by HPT for N = 1000. As shown in Fig. 11, the sample processed by ultra-SPD exhibits a tensile plasticity as high as 310% at room temperature, indicating that significant grain boundary sliding should have occurred even at room temperature. It should be noted that such a high plasticity cannot be achieved in the Mg-Li alloys after extrusion or after conventional SPD processing [53]. The occurrence of grain boundary sliding at room temperature may be the main reason that the Mg-rich and Li-rich phases could not be mixed at the atomic scale even after N = 1000. The occurrence of room-temperature grain boundary sliding with enhanced strain-rate sensitivity in ultrafine-grained Mg-Li alloy was discussed in details in Ref. [54].
(i) The hardness of the Mg-Zr alloy initially decreases and after reaching a minimum increases with a further increase in the shear strain. However, no saturation of hardness to the steady state occurs even with increasing the shear strain to 50,000. The initial decrease in the hardness should be due to the fragmentation of Mg and Zr particles and easy movement of hard Zr particles in the soft Mg phase. However, the hardness increases at larger strains because of the dissolution of Mg and Zr in each other and the formation of new nanograined metastable phases (see Ref. [16] for the evolution of microstructure and phase transformation in the Mg-Zr system at different levels of strain). (ii) The hardness of the Mg-Ni-Sn alloy, initially decreases to a minimum, increases with a further increase in the strain and finally reaches a steady state. The initial decrease in the hardness should be due to the destruction of eutectic structure, as shown in Fig. 3. The decrease in hardness by destruction of eutectic structure was reported in several earlier publications [43,44]. The latter increase in hardness up to the steady state should be due to the formation of amorphous phase, as shown in Figs. 3 and 4. Although the hardness saturates to the steady state at large shear strains, the small difference between the XRD profiles after N = 300 and 1500 (see Fig. 4) suggests that the microstructural and structural features may not be really at the steady state even after N = 1500. (iii) The hardness of the Mg-Ni-Pd alloy decreases after HPT processing, although no eutectic-like structure was detected in the as-cast MgNi-Pd alloy. The decrease of hardness in this material despite the significant grain refinement to ~10 nm [18] should be due to the lower hardness of new B2-type structure when compared to the initial hardness of hard intermetallic phases. This kind of hardnessstrain behavior was already reported during HPT processing of some other materials such as nano-grained Ni [45], low-melting temperature Pb [46], Sn [46], In [46] and Zn [47] metals, ultrahigh pure (99.9999%) Al [48] as well as an Al-Zn alloy [49] and some eutectic alloys [43,44]. However, the mechanisms of softening in these reported materials are different from that for the Mg-Ni-Pd alloy, in which a kind of phase-transition-induced softening occurs. (iv) The hardness of the Mg-Li alloy initially increases to a maximum but decreases with further increase in the strain. Such a hardnessstrain behavior was already reported in pure Al [38,48] and Mg [50,51] after HPT processing. However, the strain in which a hardness peak appears in pure Mg and Al is 2–3 orders of magnitudes lower than that in the Mg-Li alloy (γ = 1000 in Fig. 10(d)). The initial hardening of the Mg-Li alloy is quite reasonable because
In summary, although it has been established that the structural and microstructural features in single-phase materials after SPD processing saturates to the steady states at large shear strains (γ > 10–100) [5–9,37–40], the occurrence of real steady state in the immiscible multiple-phase systems should be investigated at much larger shear strains. The immiscible systems processed by ultra-SPD can transform to new metastable phases, which cannot be formed by conventional melting techniques or even by normal SPD processing. The occurrence of unusual hardening/softening and high plasticity are some other features that can be observed after ultra-SPD. Ultra-SPD can be considered as a potential processing route to develop advanced materials for different applications such as superplasticity or hydrogen storage. 5. Conclusions Ultra-severe plastic deformation (ultra-SPD) through the highpressure torsion (HPT) method for up to 1500 turns (shear strains up to 70,000) was applied to different kinds of immiscible Mg-based systems (Mg-V-Sn, Mg-V-Pd, Mg-V-Ni, Mg-Zr, Mg-Ni-Sn, Mg-Ni-Pd, Mg-Li). The following conclusions were obtained. 1. New metastable bcc-based and amorphous phases are formed in most of the selected systems. The first-principles calculations confirm that these phase transformations should occur, provided that the elements can be mixed at the atomic scale. 2. A real saturation of structural and microstructural features does not occur in most of the selected systems even at extremely large shear 164
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
[20] K. Edalati, R. Miresmaeili, Z. Horita, H. Kanayama, R. Pippan, Significance of temperature increase in processing by high-pressure torsion, Mater. Sci. Eng. A 528 (2011) 7301–7305. [21] P.H.R. Pereira, R.B. Figueiredo, Y. Huang, P.R. Cetlin, T.G. Langdon, Modeling the temperature rise in high-pressure torsion, Mater. Sci. Eng. A 593 (2014) 185–188. [22] A. Zunger, S.H. Wei, L.G. Ferreira, J.E. Bernard, Special quasirandom structures, Phys. Rev. Lett. 65 (1990) 353–356. [23] S. Kirkpatrick, C.D. Gelatt, M.P. Vecchi, Optimization by simulated annealing, Science 220 (1983) 671–680. [24] S. Kirkpatrick, Optimization by simulated annealing: quantitative studies, J. Stat. Phys. 34 (1984) 975–986. [25] A. Seko, K. Yuge, F. Oba, A. Kuwabara, I. Tanaka, T. Yamamoto, First-principles study of cation disordering in MgAl2O4 spinel with cluster expansion and Monte Carlo simulation, Phys. Rev. B 73 (2006) 094116. [26] A. Seko, Y. Koyama, I. Tanaka, Cluster expansion method for multicomponent systems based on optimal selection of structures for density-functional theory calculations, Phys. Rev. B 80 (2009) 165122. [27] P.E. Blochl, Projector augmented-wave method, Phys. Rev. B 50 (1994) 17953–17979. [28] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett. 77 (1996) 3865–3868. [29] G. Kresse, Ab initio molecular dynamics for liquid metals, J. Non-Cryst. Solids 192 (1995) 222–229. [30] G. Kresse, J. Furthmuller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set, Comput. Mater. Sci. 6 (1996) 15–50. [31] G. Kresse, D. Joubert, From ultrasoft pseudopotentials to the projector augmentedwave method, Phys. Rev. B 59 (1999) 1758–1775. [32] M. Methfessel, A.T. Paxton, High-precision sampling for Brillouin-zone integration in metals, Phys. Rev. B 40 (1989) 3616–3621. [33] K. Momma, F. Izumi, VESTA 3 for three-dimensional visualization of crystal volumetric and morphology data, J. Appl. Crystallogr. 44 (2011) 1272–1276. [34] V.L. Levitas, M. Javanbakht, Phase transformations in nanograin materials under high pressure and plastic shear: nanoscale mechanisms, Nanoscale 6 (2014) 162–166. [35] A. Badillo, P. Bellon, R.S. Averback, A phase field model for segregation and precipitation induced by irradiation in alloys, Model. Simul. Mater. Sci. Eng. 23 (2015) 035008. [36] T. Kuji, S. Nakayama, N. Hanzawa, Y. Tabira, Synthesis of nano-structured b.c.c. Mg-Tm-V (Tm = Ni, Co, Cu) alloys and their hydrogen solubility, J. Alloy. Compd. 356–357 (2003) 456–460. [37] A.P. Zhilyaev, S. Lee, G.V. Nurislamova, R.Z. Valiev, T.G. Langdon, Microhardness and microstructural evolution in pure nickel during high-pressure torsion, Acta Mater. 44 (2001) 2753–2758. [38] M. Kawasaki, R.B. Figueiredo, T.G. Langdon, An investigation of hardness homogeneity throughout disks processed by high-pressure torsion, Acta Mater. 59 (2011) 308–316. [39] D.J. Lee, E.Y. Yoon, D.H. Ahn, B.H. Park, H.W. Park, L.J. Park, Y. Estrin, H.S. Kim, Dislocation density-based finite element analysis of large strain deformation behavior of copper under high-pressure torsion, Acta Mater. 76 (2014) 281–293. [40] K. Edalati, D. Akama, A. Nishio, S. Lee, Y. Yonenaga, J.M. Cubero-sesin, Z. Horita, Influence of dislocation-solute atom interactions and stacking fault energy on grain size of single-phase alloys after severe plastic deformation using high-pressure torsion, Acta Mater. 69 (2014) 68–77. [41] K. Edalati, H. Emami, A. Staykov, D.J. Smith, E. Akiba, Z. Horita, Formation of metastable phases in magnesium-titanium system by high-pressure torsion and their hydrogen storage performance, Acta Mater. 99 (2015) 150–156. [42] A.A. Nayeb-Hashemi, J.B. Clark, A.D. Pelton, The Li-Mg (lithium-magnesium) system, Bull. Alloy. Phase Diag. 5 (1984) 365–374. [43] N.X. Zhang, M. Kawasaki, Y. Huang, T.G. Langdon, Microstructural evolution in two-phase alloys processed by high-pressure torsion, J. Mater. Sci. 48 (2013) 4582–4591. [44] N.X. Zhang, N.Q. Chinh, M. Kawasaki, Y. Huang, T.G. Langdon, Self-annealing in a two-phase Pb-Sn alloy after processing by high-pressure torsion, Mater. Sci. Eng. A 666 (2016) 350–359. [45] X.Z. Liao, A.R. Kilmametov, R.Z. Valiev, H. Gao, X. Li, A.K. Mukherjee, J.F. Bingert, Y.T. Zhu, High-pressure torsion-induced grain growth in electrodeposited nanocrystalline Ni, Appl. Phys. Lett. 88 (2006) 021909. [46] K. Edalati, Z. Horita, Significance of homologous temperature in softening behavior and grain size of pure metals processed by high-pressure torsion, Mater. Sci. Eng. A 528 (2011) 7514–7523. [47] B. Srinivasarao, A.P. Zhilyaev, T.G. Langdon, M.T. Perez-Prado, On the relation between the microstructure and the mechanical behavior of pure Zn processed by high pressure torsion, Mater. Sci. Eng. A 562 (2013) 196–202. [48] Y. Ito, K. Edalati, Z. Horita, High-pressure torsion of aluminum with ultrahigh purity (99.9999%) and occurrence of inverse Hall-Petch relationship, Mater. Sci. Eng. A 679 (2017) 428–434. [49] A.A. Mazilkin, B.B. Straumal, E. Rabkin, B. Baretzky, S. Enders, S.G. Protasova, O.A. Kogtenkova, R.Z. Valiev, Softening of nanostructured Al–Zn and Al–Mg alloys after severe plastic deformation, Acta Mater. 54 (2006) 3933–3939. [50] K. Edalati, A. Yamamoto, Z. Horita, T. Ishihara, High-pressure torsion of pure magnesium: evolution of mechanical properties, microstructures and hydrogen storage capacity with equivalent strain, Scr. Mater. 64 (2011) 880–883. [51] S. Panda, L.S. Toth, J.J. Fundenberger, O. Perroud, J. Guyon, J. Zou, T. Grosdidier, Analysis of heterogeneities in strain and microstructure in aluminum alloy and magnesium processed by high-pressure torsion, Mater. Charact. 123 (2017)
strains. This observation suggests that the steady states reported in different immiscible system within a few decades may not correspond to the real microstructural and structural saturation. 3. The selected materials exhibit unpredictable hardness-strain and tensile behaviors at ultrahigh strain when compared to those at relatively large strains. These differences suggest the microstructural features and mechanical behavior of ultra-SPD-processed materials opens for further investigation in future. Acknowledgments The authors thank Prof. Mitsuaki Furui of Tokyo University of Technology for providing the Mg-Li ingot and Dr. Hoda Emami of Kyushu University and Prof. Yaroslav Filinchuk of Université Catholique de Louvain for their precious comments on structural analysis. One of the authors (K.E.) thanks Kyushu University for the QdaiJump Research grant (No. 28325) and the MEXT, Japan, for a Grant-inAid for Scientific Research (B) (No. 16H04539). This study was supported in part by the Light Metals Educational Foundation of Japan, in part by a Grant-in-Aid for Scientific Research (S) from the MEXT, Japan (No. 26220909), and in part by the Russian Federal Ministry for Education and Science (No. 14.B25.31.0017). The HPT process was carried out in the International Research Center on Giant Straining for Advanced Materials (IRC-GSAM) at Kyushu University. References [1] R.Z. Valiev, R.K. Islamgaliev, I.V. Alexandrov, Bulk nanostructured materials from severe plastic deformation, Prog. Mater. Sci. 45 (2000) 103–189. [2] R.Z. Valiev, Y. Estrin, Z. Horita, T.G. Langdon, M.J. Zehetbauer, Y.T. Zhu, Producing bulk ultrafine-grained materials by severe plastic deformation, JOM 58 (4) (2006) 33–39. [3] Y. Estrin, A. Vinogradov, Extreme grain refinement by severe plastic deformation: a wealth of challenging science, Acta Mater. 61 (2013) 782–817. [4] X. Sauvage, A. Chbihi, X. Quelennec, Severe plastic deformation and phase transformations, J. Phys.: Conf. Ser. 240 (2010) 012003. [5] R. Pippan, S. Scheriau, A. Taylor, M. Hafok, A. Hohenwarter, A. Bachmaier, Saturation of fragmentation during severe plastic deformation, Annu. Rev. Mater. Res. 40 (2010) 319–343. [6] K. Edalati, Z. Horita, Correlations between hardness and atomic bond parameters of pure metals and semi-metals after processing by high-pressure torsion, Scr. Mater. 64 (2011) 161–164. [7] M.J. Starink, X.C. Cheng, S. Yang, Hardening of pure metals by high-pressure torsion: a physically based model employing volume-averaged defect evolutions, Acta Mater. 61 (2013) 183–192. [8] T. Sakai, A. Belyakov, R. Kaibyshev, H. Miura, J.J. Jonas, Dynamic and post-dynamic recrystallization under hot, cold and severe plastic deformation conditions, Prog. Mater. Sci. 60 (2014) 130–207. [9] A. Mishra, B.K. Kad, F. Gregori, M.A. Meyers, Microstructural evolution in copper subjected to severe plastic deformation: experiments and analysis, Acta Mater. 55 (2007) 13–28. [10] X. Sauvage, F. Wetscher, P. Pareige, Mechanical alloying of Cu and Fe induced by severe plastic deformation of a Cu-Fe composite, Acta Mater. 53 (2005) 2127–2135. [11] Y. Ivanisenko, I. MacLaren, X. Sauvage, R.Z. Valiev, H.J. Fecht, Shear-induced α→γ transformation in nanoscale Fe-C composite, Acta Mater. 54 (2006) 1659–1669. [12] P.W. Bridgman, Effects of high shearing stress combined with high hydrostatic pressure, Phys. Rev. 48 (1935) 825–847. [13] A.P. Zhilyaev, T.G. Langdon, Using high-pressure torsion for metal processing: fundamentals and applications, Prog. Mater. Sci. 53 (2008) 893–979. [14] C. Suryanarayana, Mechanical alloying and milling, Prog. Mater. Sci. 46 (2001) 1–184. [15] K. Edalati, Z. Horita, A review on high-pressure torsion (HPT) from 1935 to 1988, Mater. Sci. Eng. A 652 (2016) 325–352. [16] K. Edalati, H. Emami, Y. Ikeda, H. Iwaoka, I. Tanaka, E. Akiba, Z. Horita, New nanostructured phases with reversible hydrogen storage capability in immiscible magnesium-zirconium system produced by high-pressure torsion, Acta Mater. 108 (2016) 293–303. [17] H. Emami, K. Edalati, A. Staykov, T. Hongo, H. Iwaoka, Z. Horita, E. Akiba, Solidstate reactions and hydrogen storage in magnesium mixed with various elements by high-pressure torsion: experiments and first-principles calculations, RCS Adv. 6 (2016) 11665–11674. [18] K. Edalati, R. Uehiro, Y. Ikeda, H.W. Li, H. Emami, Y. Filinchuck, M. Arita, X. Sauvage, I. Tanaka, E. Akiba, Z. Horita, Development of new hydrogen storage materials by binding energy engineering, under submission, 2017. [19] H. Matsunoshita, K. Edalati, M. Furui, Z. Horita, Ultrafine-Grained magnesium-lithium alloy processed by high-pressure torsion: low-temperature superplasticity and potential for hydroforming, Mater. Sci. Eng. A 640 (2015) 443–448.
165
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
Acta Mater. 55 (2007) 1083–1091. [54] K. Edalati, T. Masuda, M. Arita, M. Furui, X. Sauvage, Z. Horita, R.Z. Valiev, Roomtemperature superplasticity in an ultrafine-grained magnesium alloy, Sci. Rep. 7 (2017) 2662.
159–165. [52] R. Wu, Y. Yan, G. Wang, L.E. Murr, W. Han, Z. Zhang, M. Zhang, Recent progress in magnesium-lithium alloys, Int. Mater. Rev. 60 (2015) 65–100. [53] M. Furui, H. Kitamura, H. Anada, T.G. Langdon, Influence of preliminary extrusion conditions on the superplastic properties of a magnesium alloy processed by ECAP,
166
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Ultra-severe plastic deformation: Evolution of microstructure, phase transformation and hardness in immiscible magnesium-based systems
MARK
⁎
Kaveh Edalatia,b, , Ryoko Uehirob, Keisuke Fujiwarab, Yuji Ikedac, Hai-Wen Lia,d, Xavier Sauvagee, Ruslan Z. Valievf,g, Etsuo Akibaa,d,h, Isao Tanakac,i, Zenji Horitaa,b a
WPI, International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, Fukuoka 819-0395, Japan Department of Materials Science and Engineering, Faculty of Engineering, Kyushu University, Fukuoka 819-0395, Japan c Center for Elements Strategy Initiative for Structure Materials (ESISM), Kyoto University, Sakyo, Kyoto 606-8501, Japan d International Research Center for Hydrogen Energy, Kyushu University, Fukuoka 819-0395, Japan e Normandie Université, UNIROUEN, INSA Rouen, CNRS, Groupe de Physique des Matériaux, 76000 Rouen, France f Institute of Physics of Advanced Materials, Ufa State Aviation Technical University, Ufa, Russia g Laboratory for Mechanics of Bulk Nanomaterials, Saint Petersburg State University, Saint Petersburg, Russia h Department of Mechanical Engineering, Faculty of Engineering, Kyushu University, Fukuoka 819-0395, Japan i Department of Materials Science and Engineering, Kyoto University, Kyoto 606-8501, Japan b
A R T I C L E I N F O
A B S T R A C T
Keywords: Severe plastic deformation (SPD) Ultrafine-grained (UFG) materials Nanostructured materials DFT calculations Magnesium alloys Phase transition
Although severe plastic deformation (SPD) alters the microstructure and phase transformation at the early stages of straining, the microstructural features finally saturate to the steady states at large shear strains. However, from the atomic point of view, to achieve the steady state in immiscible systems with positive heat of mixing, the minimum shear strain should be so high that the thickness of sheared phases becomes comparable to one atomic distance. In this study, ultrahigh shear strains up to ~70,000 are introduced in different Mg-based immiscible systems by high-pressure torsion (HPT) method for up to 1500 turns. New metastable phases are formed in most of the selected magnesium alloys by ultra-SPD, in good agreement with the first-principles calculations. However, the microstructural/structural saturation hardly occurs in many alloys even at ultrahigh strains. The materials processed by ultra-SPD exhibit unique hardness-strain and tensile behaviors which cannot be observed after conventional SPD.
1. Introduction Severe plastic deformation (SPD) techniques [1,2] are currently used to achieve two main microstructural and structural features: (i) refine grains to enhance the mechanical and functional properties [3], and (ii) control phase transformation and mechanical alloying [4]. Numerous studies showed that despite the microstructure and phase evolutions at the early stages of straining, the microstructural and structural features finally become saturated at large strain, where a steady state is achieved [5]. Although the occurrence of steady states at large strains is expected in single-phase metals [6] because of the contribution of dynamic recovery [7], dynamic recrystallization [8], grain-boundary rotation [9] and/or grain-boundary migration [5], it is not still well understood why and when a steady state is reached in multiple-phase materials with immiscible phases [10,11]. When pure shear strain is introduced in a multiple-phase system with immiscible phases, the phases are elongated in the shear direction
⁎
and their thicknesses are continuously reduced. Therefore, to achieve the saturation in an immiscible system under pure shear deformation mode, the strain should be increased so that the thicknesses of sheared phases are reduced to the sub-nanometer level or ideally to one atomic distance. For example, the shear strain should be at least 10,000 to achieve the saturation in an immiscible system with initial phase sizes of 10 µm. However, since the fragmentation, rotation and/or sliding of phases occur practically during the SPD process and the co-deformation of phases does not follow an ideal and homogenous pattern, the shear strain should be even higher than the one calculated above. Among numerous SPD methods developed in recent decades [1–3], the high-pressure torsion (HPT) provides a unique opportunity for the fundamental study of the behavior of different materials under very large shear strain [12,13]. Since the material in HPT processing is constrained between two rotating anvils under high pressure, the shear strain (γ = 2πrN / h , γ: shear strain, r: distance from disc center, N: number of HPT turns, h: thickness of disc [2]) can be significantly
Corresponding author at: WPI, International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, Fukuoka 819-0395, Japan. E-mail address: [email protected] (K. Edalati).
http://dx.doi.org/10.1016/j.msea.2017.06.076 Received 27 February 2017; Received in revised form 18 June 2017; Accepted 19 June 2017 Available online 22 June 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
500 and 1500 turns. For the Mg-Li alloy, the HPT was conducted on 10 mm diameter and ~0.8 mm thick discs for N = 0 (as-cast condition), 5, 50, 200 and 1000 turns. The processing speed was ω = 1 rpm, and the processing temperature was room temperature for all selected compositions. The temperature rise was less than 100 K even after processing for 1500 turns because of slow rotation speed and due to the small size of sample (as the heat source) compared to the size of anvils and other large metallic parts of the HPT facility (as the heat sinks). Details concerning the temperature rise during HPT were discussed in Refs. [20,21]. Following the HPT processing, the discs were first polished and the Vickers microhardness was measured in four radial directions from the disc center to periphery. The hardness could not be measured for the Mg-V-based systems because of the presence of many cracks. Second, the chemical homogeneity was examined at the middle of discs using scanning electron microscopy (SEM) equipped with energy dispersive X-ray spectroscopy (EDS). Third, the phase transformations were studied by X-ray diffraction (XRD) analyses using the Cu Kα radiation. Fourth, for examination of microstructure, transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM) were conducted using the accelerating voltages of 300 kV and 200 kV, respectively. Thin foils were prepared from the middle or edge of the discs using a focused ion beam system. For making the TEM foils from the Mg-Li alloy, a twin-jet electro-chemical polishing system with a solution of 5 vol% HClO4, 25 vol% C3H5(OH)3 and 70 vol% C2H5OH was used at 263 K. Fifth, for examination of tensile properties, tensile specimens with gauge dimensions of 1.5×0.3×0.7 mm3 were cut from the discs at 2–2.5 mm away from the disc center and pulled to failure using an initial strain rate of ε ̇ = 4×10−3 s−1. The only material that could be examined by tensile test was Mg-Li because of the absence of cracks.
increased with increasing the numbers of HPT turns without any limitation. Moreover, the extent of strain is controllable and the contamination of sample is minor during HPT, and thus it is basically more appropriate than ball milling for investigation of immiscible systems [14]. Although the HPT method has been widely used in the past to control the phase transformations in different kinds of metallic and nonmetallic materials [15], the significance of ultra-high shear strains (N = 1000 turns or higher) on phase transformations, mechanical alloying and properties has not received appreciable attention, except in a recent publication on the immiscible Mg-Zr system [16]. In this study, thus, different immiscible Mg-based systems, which are of interest for hydrogen storage or structural applications, are processed by HPT method for up to 1500 turns, and ultrahigh shear strains (up to 70,000) are introduced in the materials. Mechanical mixing of elements at the atomic scale and formation of new phases are reviewed together with the evolution of microstructure and mechanical properties. 2. Experimental materials and procedures Several immiscible Mg-based systems were selected for this study, as shown in Table 1: (1) Mg − 25% V − 25% Sn, (2) Mg − 25% V − 25% Pd, (3) Mg − 25% V − 25% Ni, (4) Mg − 50% Zr, (5) Mg − 17% Ni − 17% Sn, (6) Mg − 17% Ni − 17% Pd, and (7) Mg − 25% Li (all compositions are in atom%). These compositions were originally selected for possible applications as hydrogen storage materials (based on some first-principles calculations in Ref. [17]), although their performance for hydrogen storage is still under investigation. It should be noted that since the evolution of microstructure in Mg-Zr system were studied in Refs. [16] and [18], respectively, only their mechanical properties were reported in this study. The mechanical properties of Mg-Li system after processing by HPT for 5 turns were investigated in Ref. [19], but the material was investigated after processing by HPT for much larger number of turns in this study. For the first three compositions (Mg-V-based systems), MgH2 powders were mixed with the powders of V (99.5%), Sn (99.99%), Pd (99.9%) and Ni (99.99%) with particle sizes smaller than ~44 µm (325 mesh). MgH2 was intentionally used as a source of Mg because it is harder than metallic Mg and can be mixed easier with other selected elements. For the fourth composition (Mg-Zr alloy), Mg and Zr powder mixtures with particle sizes smaller than ~44 µm were used. For the fifth and sixth compositions (Mg-Ni-based systems), the elements were melted under controlled atmosphere and casted into the form of ingots (plates with 200×100×13 mm3 for Mg-Ni-Sn and a rod with 10 mm diameter and 10 mm length for Mg-Ni-Pd). The last composition (MgLi) was prepared by casting under controlled atmosphere followed by extrusion at 373 K with an extrusion speed of 1 mm/s and an extrusion ratio of 25:1 (final diameter: 10 mm). Almost 0.5 g of powder mixtures for Mg-V-based and Mg-Zr systems were processed by HPT under a pressure of 3 GPa for N = 0 (pure compression), 10, 100, 1000 or 1200 turns to produce discs with 14 mm diameter and ~0.8 mm thickness. For the Mg-Ni-based systems, discs with 10 mm diameter and ~0.8 mm thickness were processed by HPT under a pressure of 6 GPa for N = 0 (as-cast condition), 20, 100, 300,
3. Calculation methods In this study, the crystal structure of severely deformed Mg − 17% Ni – 17% Sn and Mg − 17% Ni − 17% Pd alloys were examined by first-principles calculations. The B2-type structure was selected as the initial structure for the structure optimization because this structure was experimentally detected for Mg-Ni-Pd. The initial structure was modeled by the special quasirandom structure (SQS) [22] based on the 3×3×3 (54 atoms) supercell of the unit cell of the B2-type structure. In the SQS model, body-center sites in the B2 unit cells were fully occupied by Mg atoms, while the corner sites were occupied by the same numbers of Mg, Ni, and Sn or Pd as randomly as possible under the periodic boundary conditions of the supercells. The SQS used in this study was obtained by simulated annealing [23,24] as implemented in the CLUPAN code [25,26]. To analyze the optimized atomic positions of the alloys, the radial distribution function (RDF) was calculated using the following equation as linear combinations of the Dirac delta functions, which are broadened by a normal Gaussian function with the standard deviation of 0.05 Å.
g (r ) =
Table 1 Composition of selected systems and their initial form before HPT processing. Composition (in atom%) Mg Mg Mg Mg Mg Mg Mg
− − − − − − −
25% 25% 25% 50% 17% 17% 25%
V − 25% Sn V − 25% Pd V − 25% Ni Zr Ni − 17% Sn Ni − 17% Pd Li
1 1 1 N ρ 4πr 2
N
N
∑ ∑ δ (r − i=1 j≠i
rj − ri ) (1)
In this equation, g(r) is the RDF, r is the distance, N is the number of atoms in a simulated cell (54 in this study), ρ is the density of atoms of the optimized structure, i and j are the indices for atoms in the cell, and ri and ri are the position of the ith and jth atoms, respectively. For calculations, the plane-wave basis projector augmented wave (PAW) method [27] was employed in the framework of density functional theory (DFT) within the generalized gradient approximation of the Perdew-Burke-Ernzerhof form [28] as implemented in the VASP (Vienna Ab initio simulation package) code [29–31]. A plane-wave energy cutoff of 350 eV was used. The 3s electrons for Mg, the 3d and 4s
Initial Form MgH2+V+Sn powder mixture MgH2+V+Pd powder mixture MgH2+V+Ni powder mixture Elemental powder mixture As-cast ingot As-cast ingot Extruded bar
159
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
optimized until the residual forces became less than 1×10−3 eV/Å under zero external stress. The VESTA code was employed for 3D visualization of the crystal structure [33]. 4. Results and discussion 4.1. Microstructure and phase evolution XRD profiles are shown in Fig. 1 for (a) Mg-V-Sn, (b) Mg-V-Pd and (c) Mg-V-Ni systems after processing by HPT for N = 0, 10, 100 and 1200 turns. An examination of Fig. 1 shows that while the peaks for starting powders are clearly visible after N = 0, their intensity gradually decreases and peaks corresponding to stable and new metastable phases gradually appear. Thermodynamically stable Mg2Sn and Mg2Ni intermetallics are formed at lower shear strains (N = 10 or 100) when compared to the metastable phases because of high thermodynamic driving force and negative heat of mixing for the formation of these stable phases [17]. Close examination of the XRD profiles suggests that several new metastable phases are formed after HPT processing for N = 1200: B2-type structure with a lattice parameter of 0.360 nm in Mg-VSn, B2-type structure with a lattice parameter of 0.316 nm in Mg-V-Pd, bcc with a lattice parameter of 0.291 nm in Mg-V-Pd, and a single-phase bcc with a lattice parameter of 0.325 nm in Mg-V-Ni. The XRD profiles in Fig. 1 clearly shows that even immiscible systems with positive heat of mixing such as Mg-V-based systems can be mixed at the atomic scale by ultra-SPD and new phases can be formed. The formation of these metastable phases should be due to the formation of high density of lattice defects (point defects and grain boundaries) and their effect on the dynamic stability of phases. Earlier theoretical and experimental studies confirmed that point defects [34] and grain boundaries [35] can change the effective internal energy and influence the stability of phases. The differences between the XRD profiles after N = 100 and after N = 1200 clearly confirms that the structural features can be totally different after ultra-SPD (N = 1200 which corresponds to shear strains up to ~70,000) when compared to those after conventional SPD. It should be noted that an earlier attempt to synthesize one single-phase bcc structure from the Mg-25%V-25%Ni alloy using high-energy ball milling was not successful [36], while pure bcc phase could be synthesized in this study because of introduction of ultrahigh shear strain. SEM-EDS mappings are shown in Fig. 2 for (a) Mg-V-Sn, (b) Mg-VPd and (c) Mg-V-Ni systems after processing by HPT for N = 0, 10, 100 and 1200 turns. It is apparent that the starting elemental phases are well distinguishable after N = 0. The phases are fragmented and/or elongated in the shearing direction after N = 10 (vertical axis is the shearing direction). The elements are partially dissolved in each other after N = 100 and the dissolution becomes more significant with increasing the numbers of turns to N = 1200. Except for the Mg-V-Ni system, in which a complete mixing of elements and structural saturation occur, the elemental heterogeneity is still visible in the two other systems even after N = 1200. The mixing in the Mg-V-Sn evolves slower than the other two alloys because of the co-existence of several phases with different hardness levels. The current SEM-EDS results, which are well consistent with the XRD profiles, suggest that a real steady state and structural/microstructural homogeneity in the immiscible systems is achievable only at extremely large shear strains (a few orders of magnitude larger than those reported for single-phase metals [37–39] and alloys [40]). Structural and microstructural analyses of Mg-Zr, Mg-Ni-Sn and MgNi-Pd also confirm that the initial phases dissolve in each other by HPT processing and new phases are formed. Details concerning the phase transitions in Mg-Zr were published in Ref. [16], indicating that the hcp-Mg and hcp-Zr phases transform to a supersaturated hcp phase with the lattice parameters of a = 0.321 nm and c = 0.516 nm after HPT processing for N = 1000. Moreover, small amounts of new bcc and fcc phases are formed in the Mg-Zr system after HPT processing [16].
Fig. 1. XRD profiles of (a) Mg-V-Sn, (b) Mg-V-Pd and (c) Mg-V-Ni processed by HPT for various turns.
electrons for Ni, the 5s and 5p electrons for Sn, and the 4d and 5s electrons for Pd were treated as valence. The Brillouin zones were sampled by the Γ-centered 15×15×15 k-point mesh per bcc unit cell, and the Methfessel-Paxton scheme [32] with a smearing width of 0.4 eV was employed. The total energies were minimized until the energy convergences were less than 10−8 eV. Internal atomic positions were 160
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
Fig. 3. Microstructure of Mg-Ni-Sn processed by HPT for various turns, where (a-c) are SEM-SE micrographs, (d) is TEM bright-field image and corresponding SAED pattern and (e) is high-resolution TEM image and corresponding FFT difractogram.
transform to a B2-type structure with a lattice parameter of a = 0.317 nm after HPT processing for N = 1500 turns. Close examination of the as-cast Mg-Ni-Sn alloy by SEM analysis in the secondary electron (SE) mode shows that an eutectic-like lamellar structure, as shown by arrows in Fig. 3(a), forms after casting. However, as shown in Fig. 3(b) and (c), the eutectic structure is destroyed after N = 20 and a homogenous structure is formed after N = 300. Although the eutectic structure should first transform to a network of broken particles at the early stages of straining, these particles could not be detected in this study because the minimum applied strain after N = 20 is too large to save the broken eutectic network. Examination of homogenous structure after HPT processing for N = 1500 by TEM indicates that the material has mainly an amorphous structure. The hollow ring pattern in the selected-area electron diffraction (SAED) pattern of Fig. 3(d) clearly confirm the formation of amorphous structure after N = 1500. Moreover, the random distribution of atoms in an amorphous form is clearly visible in the lattice image of Fig. 3(e). Examination of Fig. 3(e) by fast Fourier transform (FFT) analysis also confirms that the material has mainly an amorphous structure (see the difractogram with a hollow ring pattern). SEM-EDS elemental mapping and XRD profiles for the Mg-Ni-Sn system are shown in Figs. 4 and 5, respectively. Three intermetallics (Mg2Sn, Mg11.92Ni2.32Sn1.75 and Mg1.6NiSn0.3) are visible in the as-cast material, but these intermetallics gradually dissolve in each other and an amorphous-like structure, composed of mainly amorphous structure and partly nanograins, is formed after N = 1500 (see the XRD pattern after N = 1500 with a significant peak broadening). Mixing of Mg, Ni and Sn at the nanoscale after HPT processing for N = 1500 is evident in
Fig. 2. SEM-EDS elemental mapping of (a) Mg-V-Sn, (b) Mg-V-Pd and (c) Mg-V-Ni processed by HPT for various turns. Vertical axis is shearing direction.
Similar metastable bcc, fcc and hcp phases are also formed in the Mg-Ti system after processing by HPT [41]. Detailed analyses of the Mg-Ni-Pd system, which will appear in Ref. [18], confirm that the as-cast materials contains three intermetallic phases, but theses intermetallic phases 161
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
Fig. 4. XRD profiles of Mg-Ni-Sn processed by HPT for various turns.
Fig. 6. (a) High-angle annular dark-field image and corresponding EDS elemental mapping with (b) Mg, (c) Ni and (d) Sn for Mg-Ni-Sn processed by HPT for N = 1500 turns.
Fig. 5. SEM-EDS elemental mapping of Mg-Ni-Sn processed by HPT for various turns. Vertical axis is shearing direction. Fig. 7. SEM-BSED micrographs of Mg-Li processed by HPT for various turns. Vertical axis is shearing direction.
the STEM-EDS mapping of Fig. 6. It should be noted that good atomicscale mixing of elements was also confirmed in the Mg-Ni-Pd system after 1500 turns of HPT by using the atom probe tomography technique [18]. For the Mg-Li alloy, the initial structure contains two phases: Li-rich bcc structure and Mg-rich hcp structure [42]. The evolution of the two phases with increasing the number of HPT turns is shown in the SEM images of Fig. 7 in the back-scatter electron diffraction (BSED) mode. The bright regions in Fig. 7 correspond to the Mg-rich hcp phase and the dark regions correspond to the Li-rich bcc phase. Fig. 7 show that although these two phases cannot be mixed at the atomic scale even after HPT processing for N = 1000, a significant fragmentation of two phases and formation of nanosized phases occur. The variation of grain size with respect to the number of HPT turns is shown in Fig. 8. Note that the average grain sizes were measured using the TEM dark-field images for both phases. The grain size continuously reduces with increasing the number of HPT turns from N = 1 to N = 1000 and a real saturation of grain size (i.e. no change in the grain size with further
increase in the strain) cannot be achieved even after HPT processing for N = 1000, as shown in Fig. 8. It is concluded that unlike the singlephase metals, in which a saturation of grain size at large strains is achieved by dynamic recrystallization or grain boundary migration [5–9], the co-presence of two immiscible phases diminishes the occurrence of dynamic recrystallization or grain boundary migration even for the Mg-Li system with a low melting temperature. 4.2. First-principles calculations Both TEM and XRD analyses confirm that the amorphization after ultra-SPD processing occurs only in the Mg-Ni-Sn system, while the other selected systems can save their crystallinity. To explore the possible reasons for the formation of amorphous phase in this system, firstprinciples calculations were conducted on Mg − 17% Ni − 17% Sn alloy and the results were compared with those for Mg − 17% Ni − 162
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
Fig. 8. Average grain size, measured from TEM dark-field images, against number of HPT turns for Mg-Li.
Fig. 9. (a) Initial and (b) calculated optimized lattice structure of Mg-Ni-Sn and (c) calculated RDFs for ideal bcc structure and optimized structure of Mg-Ni-Sn and Mg-Ni-Pd. Effective lattice parameter (aeff) was calculated as (2 volume per atom)1/3.
17% Pd alloy. Fig. 9 shows (a) the initial model structure and (b) the optimized structure of the Mg-Ni-Sn alloy. The optimized atomic positions are largely deviated from the ideal bcc lattice sites in Fig. 9(b). Fig. 9(c) shows the RDF for the supercell model of Mg-Ni-Sn. Here, for comparison, we also show the RDF for the ideal bcc structure with the same volume as the optimized structure of Mg-Ni-Sn. Moreover, the RDF is given for the Mg-Ni-Pd system, which has a similar composition to the Mg-Ni-Sn alloy, but it does not transform to an amorphous structure after Ultra-SPD processing. The RDFs show clear peaks for MgNi-Pd but no clear and sharp peak for Mg-Ni-Sn when r ≳ aeff (aeff: effective lattice parameter). This indicates that the Mg-Ni-Pd alloy is able to save its B2-type crystal structure while the Mg-Ni-Sn is significantly distorted to a non-crystalline form. The current calculations are in excellent agreement with the experimental data, confirming that if the Mg, Ni, Pd and Sn atoms are mixed at the atomic scale by a process of ultra-SPD, the Mg-Ni-Pd and Mg-Ni-Sn systems should have
Fig. 10. Vickers microhardness against shear strain for (a) Mg-Zr, (b) Mg-Ni-Sn, (c) MgNi-Pd and (d) Mg-Li processed by HPT for various turns.
the bcc-based and amorphous structures, respectively. Earlier calculations also confirmed that if Mg and Zr are mixed at the atomic scale, the hcp, fcc and partially ordered bcc phases can become dynamically 163
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
stable (see Ref. [16] for details). 4.3. Evolution of mechanical properties One standard procedure to examine the occurrence of steady state in HPT processing is to examine the evolution of hardness with respect to shear strain. Two issues should be considered regarding this procedure, which has been widely used by different groups in recent years [37–40]. First, when the hardness does not reach the steady state, the microstructural features are not at the steady state. Nevertheless, the saturation of hardness does not necessarily correspond to the saturation of microstructure. Second, although the occurrence of steady state has been reported in many different kinds of materials since 1935 [12], there have been no attempts to confirm the occurrence of saturation at ultra-high shear strains. The hardness values are plotted against the shear strain in Fig. 10 for (a) Mg-Zr, (b) Mg-Ni-Sn, (c) Mg-Ni-Pd and (d) Mg-Li. Despite some scatterings of hardness data at the early stages of straining, the data points fall reasonably on single curves at large strains in good agreement with numerous earlier publications [37–40]. Four different unpredictable hardness-strain behaviors are seen in Fig. 10.
Fig. 11. Nominal tensile stress again strain for Mg-Li processed by HPT for N = 1000 turns. Inset for appearance of tensile specimens before and after deformation.
of the significant grain refinement, as shown in Figs. 7 and 8. However, the decrease in hardness at large shear strains despite further grain refinement should be due to the contribution of other softening mechanisms. Since the Mg-Li alloy alloys can exhibit grain boundary sliding at relatively low temperatures [52], it is likely that the softening at very large strains is due to the enhancement of grain boundary sliding. To examine the occurrence of grain boundary sliding, tensile tests were conducted on the sample processed by HPT for N = 1000. As shown in Fig. 11, the sample processed by ultra-SPD exhibits a tensile plasticity as high as 310% at room temperature, indicating that significant grain boundary sliding should have occurred even at room temperature. It should be noted that such a high plasticity cannot be achieved in the Mg-Li alloys after extrusion or after conventional SPD processing [53]. The occurrence of grain boundary sliding at room temperature may be the main reason that the Mg-rich and Li-rich phases could not be mixed at the atomic scale even after N = 1000. The occurrence of room-temperature grain boundary sliding with enhanced strain-rate sensitivity in ultrafine-grained Mg-Li alloy was discussed in details in Ref. [54].
(i) The hardness of the Mg-Zr alloy initially decreases and after reaching a minimum increases with a further increase in the shear strain. However, no saturation of hardness to the steady state occurs even with increasing the shear strain to 50,000. The initial decrease in the hardness should be due to the fragmentation of Mg and Zr particles and easy movement of hard Zr particles in the soft Mg phase. However, the hardness increases at larger strains because of the dissolution of Mg and Zr in each other and the formation of new nanograined metastable phases (see Ref. [16] for the evolution of microstructure and phase transformation in the Mg-Zr system at different levels of strain). (ii) The hardness of the Mg-Ni-Sn alloy, initially decreases to a minimum, increases with a further increase in the strain and finally reaches a steady state. The initial decrease in the hardness should be due to the destruction of eutectic structure, as shown in Fig. 3. The decrease in hardness by destruction of eutectic structure was reported in several earlier publications [43,44]. The latter increase in hardness up to the steady state should be due to the formation of amorphous phase, as shown in Figs. 3 and 4. Although the hardness saturates to the steady state at large shear strains, the small difference between the XRD profiles after N = 300 and 1500 (see Fig. 4) suggests that the microstructural and structural features may not be really at the steady state even after N = 1500. (iii) The hardness of the Mg-Ni-Pd alloy decreases after HPT processing, although no eutectic-like structure was detected in the as-cast MgNi-Pd alloy. The decrease of hardness in this material despite the significant grain refinement to ~10 nm [18] should be due to the lower hardness of new B2-type structure when compared to the initial hardness of hard intermetallic phases. This kind of hardnessstrain behavior was already reported during HPT processing of some other materials such as nano-grained Ni [45], low-melting temperature Pb [46], Sn [46], In [46] and Zn [47] metals, ultrahigh pure (99.9999%) Al [48] as well as an Al-Zn alloy [49] and some eutectic alloys [43,44]. However, the mechanisms of softening in these reported materials are different from that for the Mg-Ni-Pd alloy, in which a kind of phase-transition-induced softening occurs. (iv) The hardness of the Mg-Li alloy initially increases to a maximum but decreases with further increase in the strain. Such a hardnessstrain behavior was already reported in pure Al [38,48] and Mg [50,51] after HPT processing. However, the strain in which a hardness peak appears in pure Mg and Al is 2–3 orders of magnitudes lower than that in the Mg-Li alloy (γ = 1000 in Fig. 10(d)). The initial hardening of the Mg-Li alloy is quite reasonable because
In summary, although it has been established that the structural and microstructural features in single-phase materials after SPD processing saturates to the steady states at large shear strains (γ > 10–100) [5–9,37–40], the occurrence of real steady state in the immiscible multiple-phase systems should be investigated at much larger shear strains. The immiscible systems processed by ultra-SPD can transform to new metastable phases, which cannot be formed by conventional melting techniques or even by normal SPD processing. The occurrence of unusual hardening/softening and high plasticity are some other features that can be observed after ultra-SPD. Ultra-SPD can be considered as a potential processing route to develop advanced materials for different applications such as superplasticity or hydrogen storage. 5. Conclusions Ultra-severe plastic deformation (ultra-SPD) through the highpressure torsion (HPT) method for up to 1500 turns (shear strains up to 70,000) was applied to different kinds of immiscible Mg-based systems (Mg-V-Sn, Mg-V-Pd, Mg-V-Ni, Mg-Zr, Mg-Ni-Sn, Mg-Ni-Pd, Mg-Li). The following conclusions were obtained. 1. New metastable bcc-based and amorphous phases are formed in most of the selected systems. The first-principles calculations confirm that these phase transformations should occur, provided that the elements can be mixed at the atomic scale. 2. A real saturation of structural and microstructural features does not occur in most of the selected systems even at extremely large shear 164
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
[20] K. Edalati, R. Miresmaeili, Z. Horita, H. Kanayama, R. Pippan, Significance of temperature increase in processing by high-pressure torsion, Mater. Sci. Eng. A 528 (2011) 7301–7305. [21] P.H.R. Pereira, R.B. Figueiredo, Y. Huang, P.R. Cetlin, T.G. Langdon, Modeling the temperature rise in high-pressure torsion, Mater. Sci. Eng. A 593 (2014) 185–188. [22] A. Zunger, S.H. Wei, L.G. Ferreira, J.E. Bernard, Special quasirandom structures, Phys. Rev. Lett. 65 (1990) 353–356. [23] S. Kirkpatrick, C.D. Gelatt, M.P. Vecchi, Optimization by simulated annealing, Science 220 (1983) 671–680. [24] S. Kirkpatrick, Optimization by simulated annealing: quantitative studies, J. Stat. Phys. 34 (1984) 975–986. [25] A. Seko, K. Yuge, F. Oba, A. Kuwabara, I. Tanaka, T. Yamamoto, First-principles study of cation disordering in MgAl2O4 spinel with cluster expansion and Monte Carlo simulation, Phys. Rev. B 73 (2006) 094116. [26] A. Seko, Y. Koyama, I. Tanaka, Cluster expansion method for multicomponent systems based on optimal selection of structures for density-functional theory calculations, Phys. Rev. B 80 (2009) 165122. [27] P.E. Blochl, Projector augmented-wave method, Phys. Rev. B 50 (1994) 17953–17979. [28] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett. 77 (1996) 3865–3868. [29] G. Kresse, Ab initio molecular dynamics for liquid metals, J. Non-Cryst. Solids 192 (1995) 222–229. [30] G. Kresse, J. Furthmuller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set, Comput. Mater. Sci. 6 (1996) 15–50. [31] G. Kresse, D. Joubert, From ultrasoft pseudopotentials to the projector augmentedwave method, Phys. Rev. B 59 (1999) 1758–1775. [32] M. Methfessel, A.T. Paxton, High-precision sampling for Brillouin-zone integration in metals, Phys. Rev. B 40 (1989) 3616–3621. [33] K. Momma, F. Izumi, VESTA 3 for three-dimensional visualization of crystal volumetric and morphology data, J. Appl. Crystallogr. 44 (2011) 1272–1276. [34] V.L. Levitas, M. Javanbakht, Phase transformations in nanograin materials under high pressure and plastic shear: nanoscale mechanisms, Nanoscale 6 (2014) 162–166. [35] A. Badillo, P. Bellon, R.S. Averback, A phase field model for segregation and precipitation induced by irradiation in alloys, Model. Simul. Mater. Sci. Eng. 23 (2015) 035008. [36] T. Kuji, S. Nakayama, N. Hanzawa, Y. Tabira, Synthesis of nano-structured b.c.c. Mg-Tm-V (Tm = Ni, Co, Cu) alloys and their hydrogen solubility, J. Alloy. Compd. 356–357 (2003) 456–460. [37] A.P. Zhilyaev, S. Lee, G.V. Nurislamova, R.Z. Valiev, T.G. Langdon, Microhardness and microstructural evolution in pure nickel during high-pressure torsion, Acta Mater. 44 (2001) 2753–2758. [38] M. Kawasaki, R.B. Figueiredo, T.G. Langdon, An investigation of hardness homogeneity throughout disks processed by high-pressure torsion, Acta Mater. 59 (2011) 308–316. [39] D.J. Lee, E.Y. Yoon, D.H. Ahn, B.H. Park, H.W. Park, L.J. Park, Y. Estrin, H.S. Kim, Dislocation density-based finite element analysis of large strain deformation behavior of copper under high-pressure torsion, Acta Mater. 76 (2014) 281–293. [40] K. Edalati, D. Akama, A. Nishio, S. Lee, Y. Yonenaga, J.M. Cubero-sesin, Z. Horita, Influence of dislocation-solute atom interactions and stacking fault energy on grain size of single-phase alloys after severe plastic deformation using high-pressure torsion, Acta Mater. 69 (2014) 68–77. [41] K. Edalati, H. Emami, A. Staykov, D.J. Smith, E. Akiba, Z. Horita, Formation of metastable phases in magnesium-titanium system by high-pressure torsion and their hydrogen storage performance, Acta Mater. 99 (2015) 150–156. [42] A.A. Nayeb-Hashemi, J.B. Clark, A.D. Pelton, The Li-Mg (lithium-magnesium) system, Bull. Alloy. Phase Diag. 5 (1984) 365–374. [43] N.X. Zhang, M. Kawasaki, Y. Huang, T.G. Langdon, Microstructural evolution in two-phase alloys processed by high-pressure torsion, J. Mater. Sci. 48 (2013) 4582–4591. [44] N.X. Zhang, N.Q. Chinh, M. Kawasaki, Y. Huang, T.G. Langdon, Self-annealing in a two-phase Pb-Sn alloy after processing by high-pressure torsion, Mater. Sci. Eng. A 666 (2016) 350–359. [45] X.Z. Liao, A.R. Kilmametov, R.Z. Valiev, H. Gao, X. Li, A.K. Mukherjee, J.F. Bingert, Y.T. Zhu, High-pressure torsion-induced grain growth in electrodeposited nanocrystalline Ni, Appl. Phys. Lett. 88 (2006) 021909. [46] K. Edalati, Z. Horita, Significance of homologous temperature in softening behavior and grain size of pure metals processed by high-pressure torsion, Mater. Sci. Eng. A 528 (2011) 7514–7523. [47] B. Srinivasarao, A.P. Zhilyaev, T.G. Langdon, M.T. Perez-Prado, On the relation between the microstructure and the mechanical behavior of pure Zn processed by high pressure torsion, Mater. Sci. Eng. A 562 (2013) 196–202. [48] Y. Ito, K. Edalati, Z. Horita, High-pressure torsion of aluminum with ultrahigh purity (99.9999%) and occurrence of inverse Hall-Petch relationship, Mater. Sci. Eng. A 679 (2017) 428–434. [49] A.A. Mazilkin, B.B. Straumal, E. Rabkin, B. Baretzky, S. Enders, S.G. Protasova, O.A. Kogtenkova, R.Z. Valiev, Softening of nanostructured Al–Zn and Al–Mg alloys after severe plastic deformation, Acta Mater. 54 (2006) 3933–3939. [50] K. Edalati, A. Yamamoto, Z. Horita, T. Ishihara, High-pressure torsion of pure magnesium: evolution of mechanical properties, microstructures and hydrogen storage capacity with equivalent strain, Scr. Mater. 64 (2011) 880–883. [51] S. Panda, L.S. Toth, J.J. Fundenberger, O. Perroud, J. Guyon, J. Zou, T. Grosdidier, Analysis of heterogeneities in strain and microstructure in aluminum alloy and magnesium processed by high-pressure torsion, Mater. Charact. 123 (2017)
strains. This observation suggests that the steady states reported in different immiscible system within a few decades may not correspond to the real microstructural and structural saturation. 3. The selected materials exhibit unpredictable hardness-strain and tensile behaviors at ultrahigh strain when compared to those at relatively large strains. These differences suggest the microstructural features and mechanical behavior of ultra-SPD-processed materials opens for further investigation in future. Acknowledgments The authors thank Prof. Mitsuaki Furui of Tokyo University of Technology for providing the Mg-Li ingot and Dr. Hoda Emami of Kyushu University and Prof. Yaroslav Filinchuk of Université Catholique de Louvain for their precious comments on structural analysis. One of the authors (K.E.) thanks Kyushu University for the QdaiJump Research grant (No. 28325) and the MEXT, Japan, for a Grant-inAid for Scientific Research (B) (No. 16H04539). This study was supported in part by the Light Metals Educational Foundation of Japan, in part by a Grant-in-Aid for Scientific Research (S) from the MEXT, Japan (No. 26220909), and in part by the Russian Federal Ministry for Education and Science (No. 14.B25.31.0017). The HPT process was carried out in the International Research Center on Giant Straining for Advanced Materials (IRC-GSAM) at Kyushu University. References [1] R.Z. Valiev, R.K. Islamgaliev, I.V. Alexandrov, Bulk nanostructured materials from severe plastic deformation, Prog. Mater. Sci. 45 (2000) 103–189. [2] R.Z. Valiev, Y. Estrin, Z. Horita, T.G. Langdon, M.J. Zehetbauer, Y.T. Zhu, Producing bulk ultrafine-grained materials by severe plastic deformation, JOM 58 (4) (2006) 33–39. [3] Y. Estrin, A. Vinogradov, Extreme grain refinement by severe plastic deformation: a wealth of challenging science, Acta Mater. 61 (2013) 782–817. [4] X. Sauvage, A. Chbihi, X. Quelennec, Severe plastic deformation and phase transformations, J. Phys.: Conf. Ser. 240 (2010) 012003. [5] R. Pippan, S. Scheriau, A. Taylor, M. Hafok, A. Hohenwarter, A. Bachmaier, Saturation of fragmentation during severe plastic deformation, Annu. Rev. Mater. Res. 40 (2010) 319–343. [6] K. Edalati, Z. Horita, Correlations between hardness and atomic bond parameters of pure metals and semi-metals after processing by high-pressure torsion, Scr. Mater. 64 (2011) 161–164. [7] M.J. Starink, X.C. Cheng, S. Yang, Hardening of pure metals by high-pressure torsion: a physically based model employing volume-averaged defect evolutions, Acta Mater. 61 (2013) 183–192. [8] T. Sakai, A. Belyakov, R. Kaibyshev, H. Miura, J.J. Jonas, Dynamic and post-dynamic recrystallization under hot, cold and severe plastic deformation conditions, Prog. Mater. Sci. 60 (2014) 130–207. [9] A. Mishra, B.K. Kad, F. Gregori, M.A. Meyers, Microstructural evolution in copper subjected to severe plastic deformation: experiments and analysis, Acta Mater. 55 (2007) 13–28. [10] X. Sauvage, F. Wetscher, P. Pareige, Mechanical alloying of Cu and Fe induced by severe plastic deformation of a Cu-Fe composite, Acta Mater. 53 (2005) 2127–2135. [11] Y. Ivanisenko, I. MacLaren, X. Sauvage, R.Z. Valiev, H.J. Fecht, Shear-induced α→γ transformation in nanoscale Fe-C composite, Acta Mater. 54 (2006) 1659–1669. [12] P.W. Bridgman, Effects of high shearing stress combined with high hydrostatic pressure, Phys. Rev. 48 (1935) 825–847. [13] A.P. Zhilyaev, T.G. Langdon, Using high-pressure torsion for metal processing: fundamentals and applications, Prog. Mater. Sci. 53 (2008) 893–979. [14] C. Suryanarayana, Mechanical alloying and milling, Prog. Mater. Sci. 46 (2001) 1–184. [15] K. Edalati, Z. Horita, A review on high-pressure torsion (HPT) from 1935 to 1988, Mater. Sci. Eng. A 652 (2016) 325–352. [16] K. Edalati, H. Emami, Y. Ikeda, H. Iwaoka, I. Tanaka, E. Akiba, Z. Horita, New nanostructured phases with reversible hydrogen storage capability in immiscible magnesium-zirconium system produced by high-pressure torsion, Acta Mater. 108 (2016) 293–303. [17] H. Emami, K. Edalati, A. Staykov, T. Hongo, H. Iwaoka, Z. Horita, E. Akiba, Solidstate reactions and hydrogen storage in magnesium mixed with various elements by high-pressure torsion: experiments and first-principles calculations, RCS Adv. 6 (2016) 11665–11674. [18] K. Edalati, R. Uehiro, Y. Ikeda, H.W. Li, H. Emami, Y. Filinchuck, M. Arita, X. Sauvage, I. Tanaka, E. Akiba, Z. Horita, Development of new hydrogen storage materials by binding energy engineering, under submission, 2017. [19] H. Matsunoshita, K. Edalati, M. Furui, Z. Horita, Ultrafine-Grained magnesium-lithium alloy processed by high-pressure torsion: low-temperature superplasticity and potential for hydroforming, Mater. Sci. Eng. A 640 (2015) 443–448.
165
Materials Science & Engineering A 701 (2017) 158–166
K. Edalati et al.
Acta Mater. 55 (2007) 1083–1091. [54] K. Edalati, T. Masuda, M. Arita, M. Furui, X. Sauvage, Z. Horita, R.Z. Valiev, Roomtemperature superplasticity in an ultrafine-grained magnesium alloy, Sci. Rep. 7 (2017) 2662.
159–165. [52] R. Wu, Y. Yan, G. Wang, L.E. Murr, W. Han, Z. Zhang, M. Zhang, Recent progress in magnesium-lithium alloys, Int. Mater. Rev. 60 (2015) 65–100. [53] M. Furui, H. Kitamura, H. Anada, T.G. Langdon, Influence of preliminary extrusion conditions on the superplastic properties of a magnesium alloy processed by ECAP,
166