MASTER THESIS

MASTER THESIS

Tomáš Vávra

Superplastic deformation of ultrafine-grained magnesium alloys containing rare earth metals and zinc

Department of Physics of Materials

Supervisor of the master thesis: RNDr. Peter Minárik, Ph.D.

Study programme: Physics

Specialization: Physics of Condensed Matter and Materials

Prague 2019

I declare that I carried out this master thesis independently, and only with the cited sources, literature and other professional sources.

I understand that my work relates to the rights and obligations under the Act No.

121/2000 Coll., the Copyright Act, as amended, in particular the fact that the Charles University has the right to conclude a license agreement on the use of this work as a school work pursuant to Section 60 paragraph 1 of the Copyright Act.

In…... date... signature

Title: Superplastic deformation of ultrafine-grained magnesium alloys containing rare earth metals and zinc

Author: Tomáš Vávra

Department: Department of Physics of Materials

Supervisor: RNDr. Peter Minárik, Ph.D., Department of Physics of Materials Abstract:

Superplastic behavior of two ultrafine-grained (UFG) magnesium alloys was investigated in this thesis. Commercial Mg-4Y-3RE (wt.%) alloy was prepared by equal channel angular pressing (ECAP) and new experimental alloy Mg-23Zn-2Y (wt.%) was prepared by extrusion and ECAP. Eight passes of Mg-4Y-3RE through ECAP resulted in grain refinement down to ~340 nm and formation of a high volume fraction of fine secondary phase particles. UFG microstructure with an average grain size of 3.2 µm after extrusion and 1.6 µm after Ex-ECAP was achieved in Mg-23Zn- 2Y alloy. The microstructure of Mg-23Zn-2Y was observed by scanning and

transmission electron microscopy. The thermal stability of both alloys was measured by microhardness tests. Superplastic behavior was investigated in the temperature range of 250-450 °C and strain rate range of 5x10^-4 s^-1 - 10^-1 s^-1. The results revealed a high strain rate superplasticity in Mg-4Y-3RE alloy. Deformation to fracture

exceeded 1000% for several deformation conditions, even at the strain rate of 10^-1 s^-1. The highest elongation of 656 % in Mg-23Zn-2Y alloy was achieved in extruded state at the strain rate of 10^-3 s^-1.

Keywords: ultrafine-grained materials, magnesium alloys, superplasticity

I would like to thank my supervisor Peter Minárik for his guidance and worthful advice. My thanks also go to Dr. Jan Bohlen for casting the alloy, Robert Král and Pavel Beran for preparation of ultrafine grained material through ECAP.

Furthermore, I would like to thank Dr. František Lukáč for X-Ray measurements.

Finally, I would like to thank Jozef Veselý for transmission electron microscopy and Jana Kálalová for help with samples preparation.

Contents

1. Introduction 1

1.1. Magnesium alloys 1

1.2. Mechanical properties 2

1.2. Plastic deformation 2

1.3. Superplasticity 3

1.4. Material processing 4

2. Aims and motivation 7

3. Experimental methods 8

3.1. Material and processing 8

3.2. SEM and EBSD 9

3.3. TEM 10

3.4. XRF 11

3.6. Tensile deformation tests 13

4. Experimental results 14

4.1 Microstructure of Mg-23Zn-2Y 14

4.2 Microstructure of WE43 21

4.3 Tensile tests 23

4.3.1 m-parameter measurement 24

4.3.2 Superplastic tensile test 26

4.4 Microstructure after deformation 28

5. Discussion 32

5.1 Microstructure of investigated alloys 32

5.3 Thermal stability 34

5.4 Tensile tests 35

5.4.1 m-parameter 35

5.4.2 Superplastic tensile tests 36

5.5. Microstructure after deformation 38

6. Conclusion 39

Bibliography 41

List of Tables 47

List of Abbreviations 48

1. Introduction

1.1. Magnesium alloys

Magnesium is a lightweight metal with a density of 1.738 g/cm³ at room temperature. It´s melting point is 650 °C at atmospheric pressure. Mg has atomic number 12 and crystalizes in hexagonal closed packed structure (HCP). The lattice parameters are a=0.32 nm, c=0.52 nm with the c/a ratio of 1.62, which slightly lower than ideal HCP.

Magnesium as an element is relatively common. Its concentration is about 2 % in Earth´s crust and 1.3 g/l in sea water. Metal magnesium is mostly produced from magnesium chloride (MgCl2) by electrolysis. Other possibilities are a thermal reduction of dolomite (CaMg(CO3)2) and the extraction of magnesium oxide (MgO) from sea water.

Magnesium alloys offer some unique futures such as the lowest density of all structural metals, good castability or high specific strength. For these qualities, magnesium alloys are used as construction materials in automotive aerospace and electronics industry. Furthermore, it is used as a material for energy absorption and vibration damping. For their nontoxicity, Mg alloys have potential in medicine as biodegradable implants [1].

In spite of many qualities, Magnesium and its alloys have some limitations.

The main disadvantage is a low melting point, causing the magnesium to be inappropriate for high temperature application. Another drawback is insufficient ductility at room temperature and related bad cold working capabilities. This is caused by its HCP structure, which offers a limited number of slip systems. Although magnesium is not as reactive as some other alkaline earth metals, the corrosion resistance is low [2] and magnesium powder is highly flammable.

1.2. Mechanical properties

The mechanical properties of a given alloy depend on microstructure. One of the most important features is grain size. Smaller grain size means higher grain boundaries density. Grain boundaries are barriers for dislocation motion. The

increase of yield stress σy (material strength) with a decrease in the average grain size d is described by Hall-Petch relation:

𝝈_𝒚 = 𝝈_𝟎+ ^𝒌

√𝒅 (1)

where σ0 is the resistance of the lattice to dislocation movement and k is strengthening constant specific to the material [3].

Another important mechanism, that influences mechanical properties is precipitation hardening because secondary phase particles create obstacles for dislocation motion.

1.3. Plastic deformation

Plastic deformation causes a permanent change of sample shape in response to the applied force. The main mechanisms in Mg are twinning and slip of atomic planes in crystals. Slip is realized by dislocation motion in closed packed planes in the direction with the highest atom density. Slip plane together with slip direction creates the slip system. According to von Mises yield criterion, at least five independent slip systems are needed for homogeneous plastic deformation [5]. As magnesium crystalizes in the HCP structure, mostly the basal slip system and little of prismatic slip system is active at room temperature (Figure 1). The criterion is not fulfilled and another plastic deformation mechanism such as twinning is needed.

Thus, the ductility of magnesium is low at room temperature. At higher temperature also pyramidal slip system is active and the ductility increases.

Figure 1, slip systems of HCP structure [6].

1.4. Superplasticity

Superplasticity is defined as plastic deformation greater than 400 % of tensile strain [7]. Another feature is high strain rate sensitivity of the stress with strain rate sensitivity parameter m>0,3 [8] or m>0,5 [7]. The main mechanisms of superplastic deformation are grain boundary sliding and diffusion creep. Grain boundary sliding leads to rearrangement of nearby grains or groups of grains. Grain boundary sliding can´t appear independently. Diffusion creep is essential to prevent voids formation at ledges and triple points.

Superplastic deformation is dependent on three main parameters: the grain size, temperature and strain rate. Grain size under 10 µm with a high fraction of high angle grain boundaries is necessary for grain boundary sliding to occur. Because both mechanisms of superplasticity are thermally activated, the suitable temperature range is above 0,4 Tm (melting temperature). Higher temperature means better grain

boundary sliding, but grains growth can appear. Strain rate range depends on the material. With higher temperature, higher strain rate can be used. Hence thermally stable microstructure is necessary for high strain rate superplasticity.

Superplastic forming offers an attractive option for manufacturing of products from magnesium alloys. This method provides the ability to produce a unique shapes with the minimum of waste compared to conventional methods as milling, turning, etc. Thus, superplastic forming significantly reduces production costs. However,

superplastic deformation in Mg is time consuming, which is the main limitation for practical applications. Most of the magnesium alloy proved superplastic behavior only at slow strain rates [10–25]. Such deformation takes several hours at elevated temperature to perform. High strain rate superplasticity is key for practical

applications. The ability to achieve superplastic deformation can be improved by microstructure optimization

1.5. Material processing

The ultrafine grained microstructure is essential for superplasticity. There are many processing methods to decrease average grain size down to micron level based on thermomechanical treatment, rapid solidification or severe plastic deformation (SPD).

One of the commonly used methods of thermomechanical treatment is extrusion. Extrusion is a forming process used to produce rod-like objects of a fixed cross-sectional profile. The material is pushed through a die of a smaller cross- section. The scheme of extrusion is shown in Figure 2. Important extrusion

parameters are the speed of the ram, extrusion ratio and temperature. Extrusion ratio ER is defined as:

𝐸𝑅 =^𝐴_𝐴⁰

1 (2)

where A0 is the initial cross-section and A1 is the final cross-section after extrusion.

Figure 2, scheme of extrusion [26].

For further grain refinement, SPD method can be used. Equal channel angular pressing (ECAP) is one of the mostly used SPD techniques. Material is pushed through a bent channel with the same input and output cross-section. The process can be repeated many times for finer and more homogeneous microstructure. The sample can be rotated between passes. That distinguish four different routes, which are shown in Figure 3 together with the scheme of the ECAP.

Figure 3, the scheme of ECAP and processing routes [27].

Parameter influencing final microstructure are processing route, temperature, processing speed and geometry of ECAP. The route BC is considered to be the most effective for a homogeneous ultra-fine grained microstructure with a high

fractiongraat of high angle grain boundaries [28]. The main mechanism of grain refinement is intensive plastic deformation caused by sheer stress. Different shear planes occur in distinct routes as displayed in Figure 4. ECAP with right angle bent channel (Φ=90°, Ψ=0°) is considered to be the most effective for grain refinement. In this case, deformation introduced to the material during one pass is ε=1.15 [27].

Temperature and processing speed are the key parameters and depends on the material. The higher temperature is necessary to prevent fragmentation, especially for magnesium, which is not very ductile at room temperature. However, too high temperature can lead to recrystallization and grain growth. Beside uniform ultra-fine grained microstructure, processing by ECAP can fragment secondary phase particles.

Furthermore, these small particles can spread out over the matrix and further increase mechanical properties of the material such as strength and thermal stability [29].

Figure 4, shear planes of ECAP routes [27].

2. Aims and motivation

Two magnesium alloys were investigated in this thesis: commercial alloy WE43 and experimental alloy Mg-23Zn-2Y (wt.%).

WE43 (Mg-4Y-3RE (rare-earth elements) (wt.%)) is a well-known

commercial alloy. This alloy has exceptional qualities such as high strength, creep and good corrosion resistance [30]. It is suitable for use even at elevated

temperatures up to 300°C. Therefore, it is used especially in aviation and automotive industry. This alloy has also potential to be used in medicine for biodegradable implants [31–33]. It was shown that WE43 exhibits good superplastic behavior after ECAP processing [20, 34]. Lately, it was achieved significantly better grain

refinement compared to previous studies resulting in an average grain size about 340 nm thanks to ECAP parameters optimization [35]. Moreover, microstructure with a high density of fine secondary phase particles Mg5RE was achieved. This improved the thermal stability of ultra-fine grained microstructure, which is stable up to 280 °C for 1 hour of annealing [36]. Since thermally stable microstructure with remarkably small grain size is key factor for superplasticity, high superplastic

deformation is expected to be achieved in the WE43 alloy with optimized microstructure.

MgZnY based alloys have the potential to achieve high superplastic

deformation. It was shown, that this system may contain icosahedral quasicrystalline Mg3YZn6 phase [37]. Secondary phase particles of this phase help grain boundary sliding during superplastic deformation [38]. only few studies about superplasticity of this system have been published so far, but only with composition not exceeding 13 wt% of Zn and 1,5 wt% of Y[23, 39–42]. Mg-23Zn-2Y is close to the eutectic point, therefore it should contain a much higher fraction of quasicrystalline phase.

Together with ECAP processing, ultra-fine grained microstructure with a high density of quasicrystalline particles is expected. Achieving of such microstructure should improve superplastic behavior.

The ultra-fine grained WE43 alloy was prepared by ECAP and two states of Mg-23Zn-2Y alloy was prepared by extrusion and extrusion + ECAP (Ex-ECAP).

The objective of this study was to characterize microstructure of both alloys and possibilities of superplastic deformation, more specifically:

Characterization of microstructure of Mg-23Zn-2Y alloy after extrusion and ECAP.

1) Finding suitable conditions of superplastic deformation for both alloys, especially deformation temperature and strain rate.

2) Study the microstructure changes of both alloys after superplastic deformation.

3. Experimental methods

3.1. Material and processing

Two states of Mg-23Zn-2Y alloy prepared by extrusion and by extrusion + ECAP were investigated. As cast Mg-23Zn-2Y alloy was homogenization annealed for 10 hours at 400 °C and the chemical composition was roughly checked by spark emission spectroscopy. Then extrusion of Mg-23Zn-2Y alloy was processed at

250 °C and 18 mm/min with an extrusion ratio of 16 and afterward processed by ECAP.

As cast WE43 (Mg –3,8 wt% Y –2.6 wt% RE –0.45 wt% Zr –0.01 wt% Mn) alloy was homogenization annealed to the T4 temper (525 °C for 16 h) and afterward processed by ECAP [35].

Right angle ECAP (Φ=90°, Ψ=0°) was used with route BC. The processing parameters for Mg-23Zn-2Y alloy are shown in Table 1 and for WE43 alloy inTable 2.

Table 1, ECAP processing parameters for Mg-23Zn-2Y alloy.

Table 2, ECAP processing parameters for WE43 alloy

3.2. SEM and EBSD

Scanning electron microscopy uses the focused electron beam with energy about 20 keV to scan the surface of the sample. Electron signal intensity is collected above the surface from each point of the scanned area to create an image. Two main electron signals are detected: signal from secondary electrons (SE) and backscattered electrons (BSE). BSE are electrons from a primary beam which are elastically

scattered by atoms of the sample. Because BSE are sensitive to proton number Z of the specimen atoms, this signal helps to determine the distribution of different phases (Z contrast). BSE signal is also influenced by the crystallographic orientation of the

passes 1P 2P 3P-8P

temperature (°C) 210 190 180

rate (mm/min) 5 7 10

passes 1P 2P 3P 4P 5P 6P-8P

temperature (°C) 335 315 300 295 290 285

rate (mm/min) 5 7 10 10 10 10

grains and can distinguish individual grains (channeling contrast). BSE are also used in Electron backscatter diffraction (EBSD). This method provides local information about the crystallographic orientation of the measured point; therefore EBSD mapping and analysis give information about a shape of individual grains, texture, grain size and misorientation of grain boundaries. In EBSD orientation maps (inverse pole figure (IPF) maps), the orientation of each point is usually represented by color, which corresponds to the orientation triangle shown in Figure 5. SE are ejected from an electron shell of the sample´s atoms by inelastic scattering interactions with beam electrons. SE signal reports about the topography of the specimen’s surface.

Figure 5, orientation triangle for HCP structure

The sample preparation for SEM is one of the key factors of image quality.

The samples were mechanically ground and polished with decreasing grinding particles size down to 1/4 µm. Samples for EBSD were in addition ion-polished by Leica EM RES102. FEI Quanta 200F scanning electron microscope was used for all method previously described. EBSD mapping was performed on 200x200 µm area with step size of 0.2 µm. The data were analyzed by OIM TSL Analysis program.

Raw data were partially cleaned by one step of confidence index standardization and one step of grain dilatation.

3.3. TEM

Transition electron microscope detects high energy electrons (200 keV) transmitted through a thin specimen. The image is captured from fluorescent screen or CCD camera. TEM can operate in two modes. In image mode TEM project

crystallographic defects of the sample as dislocation, secondary phase particles and grains boundaries. In diffraction mode, crystallographic structure and orientation can be obtained from the diffraction pattern.

TEM project crystallographic defects of the sample as dislocation, secondary phase particles and grains boundaries. In diffraction mode, crystallographic structure and orientation can be obtained from the diffraction pattern. TEM is also able to measure chemical composition by Energy-dispersive X-ray spectroscopy (EDX).

Some transmission electron microscopes are able to operate in scanning mode (STEM). In this mode, chemical composition mapping by EDX can be used.

The samples for TEM were machined from the billet and mechanically ground to the thickness of approximately 150 µm. Then 3 mm discs were cut out.

Eventually, the samples were ion-milled by Leica EM RES102. Because the resolution is better in the comparison with SEM, Jeol 2200FS TEM was used for small secondary phase particles characterization.

3.4. XRF

X-ray fluorescence is a spectroscopy method, which uses characteristic X-ray radiation for element identification. The sample’s atoms are excited by irradiation by high energy X-ray or gamma rays. Afterward, the emitted characteristic X-ray is detected.

Composition measurement by XRF is usually more accurate then EDX in SEM. Thus, it was used for the analysis of chemical composition of Mg-23Zn-2Y.

The samples were prepared similarly as for SEM.

3.5. Microhardness tests

Microhardness tests are often used to study microstructure changes, because microhardness depends strongly on the microstructure and it is a very fast method.

Vickers method is one of the most frequently used one. A diamond in the shape of a square-based pyramid with a 136° top angle is indented to the sample surface. The scheme of the method is shown in Figure 6. After unload, the diagonals of square shaped indentation are measured. Then microhardness is calculated as:

𝐻𝑉 = 0,189 ^𝐹

𝑢² (6)

where F is force applied to the indentor [N] and u is the average length of

diagonals[mm]. Microhardness usually decreases with an increasing grain size and decreasing density of dislocations and precipitates, due to their strengthening effects.

Figure 6, the scheme of Vickers microhardness test [43].

Microhardness tests were measured on samples prepared similarly as for SEM by Qness Q10A. 500g load (HV 0.5) for 10 s was used. Average value of microhardness was calculated from at least 36 indents for each sample.

3.6. Tensile deformation tests

During the tensile deformation test, the sample is mounted to the device and strained. The crosshead speed is calculated from the initial dimensions of sample in order to achieve desired strain rate. Usually, the crosshead speed is being kept constant during the test. Meanwhile, the force F and extension ΔL are measured. Tensile strain

ε

𝜺_𝑬𝑵𝑮 =^∆𝑳

𝑳𝟎 (3)

where L0 stands for the initial length. Tensile stress σENG is defined as:

𝝈_𝑬𝑵𝑮 =_𝑨^𝑭

𝟎 (4)

where A0 is the initial cross-section of the sample. Tensile strain and tensile stress are associated with initial dimensions of the sample and don´t consider dimension change during the tensile test. True values of strain ε and stress σ do consider a change of the sample dimensions and are defined as:

𝜺 = 𝒍𝒏 (^𝑳𝟎_𝑳^+∆𝑳

𝟎 ) (5)

𝝈 = ^𝑭_𝑨=_𝑨^𝑭

𝟎∙^𝑳𝟎_𝑳^+∆𝑳

𝟎 (6)

Strain rate ε̇ is specified as the time derivation of tensile strain. Strain rate sensitivity m parameter is defined:

𝒎 =^{𝜹 𝒍𝒏(𝝈)}_{𝜹 𝒍𝒏(𝜺 )} (7)

Tensile tests were measured by Instron 5882 deformation machine equipped with the furnace. The dog-bone shaped samples were machined from the investigated materials with the dimensions of the active part 6x4x1 mm³. Different values of the initial strain rate and temperature were used and are precisely specified through the manuscript.

4. Experimental results

4.1 Microstructure of Mg-23Zn-2Y

Firstly, the chemical composition of Mg-23Zn-2Y was measured in different parts of the extruded rod by XRF. The results are shown in Table 3, where l means part of the rod, where the sample was taken from. The gradient of composition occurred with slightly higher concentration of the alloying elements at the end of the extruded rod in comparison with the beginning.

Table 3, the chemical composition of the extruded Mg-23Zn-2Y rod in different parts of its length l.

Secondly, the microstructure of the extruded rod was observed by SEM and EBSD. The microstructure of different parts of the extruded rod is shown in Figure 7.

We can see, that the microstructure isn´t homogeneous through the rod. The secondary phase particles are getting finer to the end of the extruded rod. The

particles are arranged in the stripes aligned along in the extrusion direction, as can be seen in the transverse scan.

l 0 1/3 2/3 1

Mg 76.9% 75.8% 75.3% 74.3%

Zn 21.2% 22.1% 22.6% 23.6%

Y 1.8% 2.0% 2.0% 2.1%

(a) (b)

(c) (d)

Figure 7, the microstructure of Mg-23Zn-2Y at the beginning (l=0) (a), in the middle (l=2/3) (b), in the end (l=1) (c,d) of the extruded rod. Cross-section (a-c),

longitudinal section (d).

ECAP was made both from the beginning and from the end of the extruded rod. The SEM images of microstructure are shown in Figure 8. ECAP wasn’t able to crush the particles completely and the microstructure depends on the initial state. For that reason, only last third of the extruded rod was used for further experiments and as the initial material for further ECAP processing.

(a) (b)

Figure 8, the microstructure of Mg-23Zn-2Y after ECAP from the beginning (a) and from the end (b) of the rod.

EBSD maps with the dimensions 200x200 µm² were made from both states of Mg-23Zn-2Y. Cropped areas are shown in Figure 9 and grain size distribution in Figure 10. and pole figures in Figure 11. Grains are considered to be an area bounded by high angle grain boundaries. The average grain size was calculated as area

fraction. Grain refinement was achieved with an average grain size of Mg-23Zn-2Y after extrusion is 3.2 µm and 1.6 µm after ECAP. Both extruded (83%) and ECAP (82%) state exhibit a high fraction of high angle grain boundaries (Figure 10).

(a) (b)

Figure 9, EBSD of Mg-23Zn-2Y after extrusion (a) and ECAP (b)

(a) (b)

Figure 10, grain size distribution of Mg-23Zn-2Y after extrusion (a) and ECAP (b)

(a) (b)

Figure 11, pole figures of Mg-23Zn-2Y after extrusion (a) and ECAP (b)

Secondary phase particles in Mg-23Zn-2Y were observed by TEM.

Precipitates were characterized by electron diffraction, STEM and EDX. Three types of secondary phase particles were found. The largest ones are icosahedral

quasicrystalline I (Mg3YZn6) phase particles with lattice parameter ar=0.519 [37].

Smaller rod-shaped coherent particles are β1’(Mg2Zn3) phase with lattice parameters a=2.596, b=1.428, c=0.524, γ=102.5°. The small plate-shaped coherent particles could be β2’(MgZn2) phase with lattice parameters a=0.525, c=0.857, γ=120° [37, 44]. Chemical composition maps of a given area are shown in Figure 12, where (a) is STEM image of the area, (b) shows distribution of Mg atoms, (c) Zn atoms, (d) Y atoms. Chemical composition maps show, that all Yttrium is located in

quasicrystalline phase particles. The TEM image and diffraction pattern of quasicrystalline phase particle are shown in Figure 13. The rod and plate-shaped particles are shown in Figure 14. The rod particles were observed only in the extruded sample.

(a) (b)

(c) (d)

Figure 12, TEM EDX analysis of particles in Mg-23Zn-2Y with a hole in the bottom

(a) (b)

Figure 13, TEM image (a) and diffractogram (b) of quasicrystalline phase particles in Mg-23Zn-2Y

(a) (b)

Figure 14, TEM images of rod and plate secondary phase particles in Mg-23Zn-2Y (a) and detail image of plate particle (b)

4.2 Microstructure of WE43

The microstructure and ECAP processing of WE43 alloy is well described in [35, 36]. TEM images of WE43 alloy prepared by ECAP are shown in Figure 15.

Microstructure with an average grain size of 340 nm with a high density of fine secondary phase particles was achieved. The precipitates were determined as Mg5RE with an average size of 150 nm. These precipitates predominately segregate at grain boundaries and thus help to stabilize the microstructure. Annealing for 1 hour showed that ultrafine microstructure is stable up to 280 °C [36]. Furthermore, microstructure with very week texture was accomplished by ECAP processing.

Figure 15, TEM image (a) and ACOM-TEM map (b) of WE43 and diffraction pattern of Mg5RE particles [35]

4.2 Thermal stability of WE43 and Mg-23Zn-2Y

The aim of the thermal stability measurements was to find a suitable range of temperature for superplastic deformation. It was studied by Vickers microhardness tests performed at annealed samples. Samples were annealed at different

temperatures for 30 minutes and then water quenched. The microhardness tests were subsequently performed, using 500 g load for 10 s. The results are shown in Figure 16. The microstructure of WE43 is stable up to 300 °C. The microhardness of Mg- 23Zn-2Y_8P decreases at 200 °C which is approximately the temperature of ECAP.

The microhardness of Mg-23Zn-2Y_Ex decreases at 250 °C which is approximately the temperature of extrusion. The picture from the indentation machine of the sample

annealed at 450 °C is shown in Figure 17. There are large grains with secondary phase particles at grain boundaries.

Figure 16, microhardness tests of annealed samples

Figure 17, Mg-23Zn-2Y_8P sample annealed at 450 °C for 30 min

4.3 Tensile tests

Tensile tests were performed at elevated temperatures where high strain rate superplasticity was expected. Tensile test at a strain rate of 10^-1 s^-1 would only last between one and two minutes. Thus, the time necessary to heat the sample to test temperature was an important parameter. Too short heating time in the furnace before the test beginning would lead to a lower temperature of samples than required.

On the other hand, too long heating time could cause unnecessary grains growth. For these reasons, the warming up of the sample was measured by the attached

thermocouple. The furnace was pre-heated to 200 °C before the sample was inserted.

The dependence of the sample temperature on heating time is shown in Figure 18 (a).

Furthermore, microhardness tests were performed at different samples

annealed for 5 to 20 min at 350 °C and 450 °C. The dependence of microhardness on annealing time is shown in Figure 18 (b). We can see that the microstructure nearly doesn’t change in the first 10 minutes in the sample annealed at 350 °C. The biggest change in the microstructure of sample annealed at 450 °C happens in the first five minutes.

It is clear, that the 10 minutes of heating time is optimal for heating the sample before the tensile test.

(a) (b)

Figure 18, the dependence of the sample temperature on heating time (a), the dependence of microhardness on annealing time (b)

4.3.1 m-parameter measurement

The strain rate sensitivity m-parameter indicates the ability of a material to achieve superplastic deformation. We are looking for a suitable temperature and the highest strain rate where m>0.3 because high strain rate superplasticity is less time consuming and thus has advantages for practical applications. The m-parameter was measured by tensile tests. According to thermal stability measurement, the

temperatures from 300 °C to 500 °C were chosen for WE43 and temperatures from 200 °C to 350 °C for Mg-23Zn-2Y. The strain rate was stepwise changed in the range from 10^-5 s^-1 to 10⁰ s^-1 during each tensile test. The changes in true stress were observed and the m-parameter was calculated according to (7). The dependence of m-parameter on strain rate is shown in Figure 19 for WE43 and in Figure 20 in the case of Mg-23Zn-2Y. There is a line showing m>0.3 in both graphs. The m-

parameter measurement at 350 °C of Mg-23Zn-2Y is not shown, because the sample broke too early to measure the m-parameter in all attempts.

For the subsequent deformation tests of WE43, the strain rates of 10^-2 s^-1 and 10^-1 s^-1 were chosen, because the aim of this study was the investigation of the high strain rate superplasticity and the m-parameter was high in this strain rete range. The highest m-parameter in this strain rate range in question was measured for the

temperature of 400 °C, therefore, the best superplastic behavior was expected for this temperature.

For superplastic deformation of Mg-23Zn-2Y slow strain rate of 5x10^-4 s^-1 and 10^-3 s^-1 were chosen, because m-parameter is high in this region. Furthermore, the strain rate of 10^-2 s^-1 was chosen for high strain rate superplasticity because it is the highest strain rate where m>0,3. The measurement of m-parameter indicates that suitable temperature for superplastic deformation of Mg-23Zn-2Y could be around 300 °C.

Figure 19, m-parameter of WE43

(a) (b)

Figure 20, m-parameter of Mg-23Zn-2Y after extrusion (a) and ECAP (b)

4.3.2 Superplastic tensile test

Deformation to failure tensile tests were performed at the temperature range from 350 °C to 450 °C for WE43 and from 250 °C to 350 °C for Mg-23Zn-2Y.

Constant strain rates were used in the range from 5x10^-4 s^-1 to 10^-1 s^-1. The true stress true strain curves are shown in Figure 21, Figure 22 and Figure 23. Maximal

elongation (and true strain rate) achieved during superplastic deformation is shown in Table 4.

Figure 21, tensile tests of WE43

Figure 22, tensile test of Mg-23Zn-2Y after extrusion

Figure 23, tensile test of Mg-23Zn-2Y after Ex-ECAP

Table 4, maximum elongation reached during superplastic deformation

4.4 Microstructure after deformation

The samples for microstructure observation after superplastic deformation were taken from the middle of the active part of tensile samples. The microstructure after all tensile tests performed at strain rate 10^-2 s^-1 was observed with SEM by secondary electron detection and EBSD. Cavities in the direction of deformation occurred in all studied samples of Mg-23Zn-2Y alloy (10^-2 s^-1 250 °C and 300 °C). The cavities appeared in the stripes of secondary phase particles. These stripes were formed during extrusion (Figure 7) and remained even after subsequent processing by ECAP and superplastic deformation. Some of the SEM pictures of Mg-23Zn-2Y are shown in Figure 24

In WE43 cavities only occurred in a deformed sample at 450 °C. In this sample, secondary phase particles joined together as can be seen in Figure 25 (b) compares with the uniform distribution of precipitates in sample strained at 350 °C Figure 25 (a). The cavities are also caused by grains growth.

alloy temperature (°C) strain rate (s^-1) elongation (%) true strain rate

10^-1 799 2.20

10^-2 1233 2.59

10^-1 1013 2.41

10^-2 1232 2.59

10^-1 475 1.75

10^-2 542 1.86

10^-2 322 1.44

10^-3 514 1.81

5x10^-4 545 1.86

10^-2 533 1.85

10^-3 656 2.02

350 10^-2 63 0.49

10^-2 391 1.59

10^-3 556 1.88

10^-2 513 1.81

10^-3 520 1.82

WE43_8P

Mg-23Zn-2Y_Ex

Mg-23Zn-2Y_8P

350 400 450

300

250 300 250

(a) (b)

Figure 24, SEM images of extruded (a) and ECAP (b) Mg-23Zn-2Y after superplastic deformation at 300 °C and 10^-2 s^-1

(a) (b)

Figure 25, SEM images of WE43_8P after superplastic deformation at 350 °C (a) and 450 °C (b) at 10^-2 s^-1

The EBSD maps of Mg-23Zn-2Y deformed at 250 °C are shown in Figure 26 and at 300 °C in Figure 27. In Figure 28 are shown EBSD maps of WE43 deformed at temperatures from 350 °C to 450 °C. Average grain size (area fraction) of

deformed samples is shown in Table 5. We can see that grain size is growing with a

temperature of tensile tests. Grains elongation appeared in WE43 samples strained at 350 °C and 400 °C.

(a) (b)

Figure 26, EBSD maps of extruded (a) and ECAP (b) Mg-23Zn-2Y after superplastic deformation at 250 °C and 10^-2 s^-1

(a) (b)

Figure 27, EBSD maps of extruded (a) and ECAP (b) Mg-23Zn-2Y after superplastic deformation at 300 °C and 10^-2 s^-1

(a) (b)

(c)

Figure 28, EBSD maps of WE43_8P after superplastic deformation at 350 °C (a), 400 °C (a), 450 °C (a) and 10^-2 s^-1

Table 5, average grain size (area fraction) after deformation at different temperatures

5. Discussion

5.1 Microstructure of investigated alloys

The chemical composition of Mg-23Zn-2Y changed slightly alongside the extruded rod (Table 3). The gradient of composition probably results from casting the alloy. The end of the rod contains the highest concentration of alloying elements which was the closest to the ideal composition of Mg-23Zn-2Y. This chemical composition leads to eutectic microstructure with a maximum amount of quasicrystalline I-phase in α-Mg [45].

The microstructure of Mg-23Zn-2Y changed alongside the extruded rod as well. It is common, that the microstructure is different at the beginning of the rod because the extrusion process has not been stable yet. But the microstructure also varies in the second half of the rod, which is unusual. The secondary phase particles are large at the beginning and getting smaller to the end of the rod. There is the most convenient microstructure for superplastic deformation at the last third of the

extruded rod. This microstructure contains a high density of fine precipitates, which helps to achieve superplastic deformation by grain boundary sliding [38]. The precipitates form stripes in the extrusion direction (Figure 7), which is typical in extruded material.

It was thought, that 8 passes through ECAP introduced so large plastic deformation to the material, that the microstructure after ECAP would be

independent on the initial microstructure [46]. To prove this, two samples from the alloy temperature (°C) strain rate (s^-1) grain size (µm)

350 2,6

400 5,0

450 10,4

250 2,7

300 4,8

250 2,8

300 4,4

WE43_8P

Mg-23Zn-2Y_Ex Mg-23Zn-2Y_8P

10^-2

10-2 10-2

beginning and the end of the extruded rod were processed by ECAP. Figure 8 shows, that the microstructure of these two ECAP samples is different. The large particles from the beginning of the extruded rod were able to resist ECAP processing. For these reasons, only the last third of the extruded rod was used for all further experiments as well as ECAP processing.

EBSD maps both from the extruded material and the material after Ex-ECAP were measured (Figure 9). Both states exhibit a high fraction of high angle grain boundaries, which is necessary for superplastic deformation. Pole figures (Figure 11) shows, that the texture of Mg-23Zn-2Y is a typical texture of magnesium alloys after extrusion and ECAP [47, 48]. Even though a small average grain size was achieved in extruded material, ECAP successfully decreased it from 3,2 µm to 1,6 µm. Despite grain refinement, ECAP wasn’t able to effectively fragment secondary phase

particles as effectively as described in [49].

The next step was to identify these secondary phase particles by TEM. EDX analysis in TEM revealed, that all Zn and Y is located in secondary phases particles (Figure 12). Moreover, Y was found only in large particles.

Three types of particles were found. Large particles, also visible in SEM micrograph, are quasicrystalline I-phase particles (Figure 13). These are

homogenously distributed in material and their size varies in the extruded rod as was described. The small plate-shaped particles were identified as β2’-phase [44]. They are coherent and relatively rare in the material. The last type is small coherent rod- shaped β1’-phase particles [44]. They are rare and occur only in the extruded sample (Figure 14). TEM proved, that microstructure contains a high fraction of

quasicrystalline I-phase particles. Other β2’-phase and β1’-phase particles are small and rare.

The microstructure of WE43 after ECAP with an average grain size of 340 nm with a high density of fine secondary phase Mg5RE particles is well described in [35, 36].

The particles have a diameter of approximately 150 nm were predominately found at grain boundaries and triple points, and thus help to stabilize the microstructure (Figure 15).

5.3 Thermal stability

The thermal stability of the microstructure of Mg-23Zn-2Y and WE43 was measured by microhardness tests (Figure 16). The microhardness of WE43 was nearly constant to 250 °C. Then an increase of the microhardness was detected around 300 °C. Significant decrease of microhardness was observed in higher temperatures. The decrease is caused by grain growth. According to [36], the increase of microhardness is caused by precipitation of Y and RE enriched particles in the grain boundaries. In this study similar curve was measured and the

microstructure of annealed samples was investigated in detail.

The microstructure of Mg-23Zn-2Y alloy was less thermally stable compares to WE43. The microhardness of the initial state (annealed at 25 °C) processed by Ex- ECAP was slightly higher than the extruded sample. It is probably caused by a smaller grain size after ECAP, which was described earlier. The microhardness was constant up to 250 °C in the extruded sample and 200 °C in the sample after Ex- ECAP (Figure 16). These are approximately the temperatures of extrusion and ECAP, respectively. The microhardness starts to decrease at temperatures of 300 °C and 250 °C due to the grain growth.

Surprisingly, the microhardness increases in both states at the temperatures above 300 °C. Although grains have grown significantly, the microhardness doesn’t decrease at high temperatures. It could be caused by secondary phase precipitation at grain boundaries as can be seen in Figure 17. LPSO phases precipitated at grain boundaries were reported in many MgZnY systems annealed above 300 °C [37, 50].

It was proved, that LPSO phases has a significant impact on the strength of the material [51]. Even though phase formation in annealed material would be an interesting topic, it wasn’t the subject of this study and therefore it wasn’t further investigated.

The thermal stability measurement suggested, that suitable temperatures for superplastic deformation could be around 350 °C for WE43 and around 250 °C for Mg-23Zn-2Y.

5.4 Tensile tests

Tensile tests were used for m-parameter measurement to find the suitable condition of superplasticity. These conditions were used in deformation to failure tensile test to prove superplastic behavior.

It was found, that the suitable time for the sample to heat up in the furnace before the tensile test is 10 minutes (Figure 18). Moreover, Figure 18 (b) shows, that most of the microstructure change happens in the first 5 minutes in the furnace.

5.4.1 m-parameter

Then the aim of the m-parameter measurement was to find suitable conditions for superplastic deformation. High m-parameter indicates good superplastic

behavior, where the limit for superplasticity is considered m>0,3.

The m-parameter of WE43 was sufficiently high in the whole investigated range (Figure 19). The results showed that the m-parameter was much higher than 0,3 even at the strain rate of 10^-1 s^-1. The complicated evolution of m-parameter is caused by a combination of two main mechanisms of superplasticity: diffusion creep and grain boundary sliding. The highest values of m-parameter at the highest strain rate was found at the temperature of 400 °C. For lower temperatures (from 300 °C to 400 °C) the m-parameter maximum is lower and shifted to the lower strain rates. It is expected that the average grain size remained relatively small and superplastic deformation can proceed by grain boundary sliding and diffusion creep. However, these thermally activated processes are not as fast as in the case of 400 °C. On the other hand, the increase in the deformation temperature to 450 °C and 500 °C led to a decrease of m-parameter as well. This deterioration can be explained by a grain growth, which probably occurred during the temperature stabilization before the execution of the test (Figure 18). Therefore, the atoms have to travel a longer distance in larger grains during diffusion creep, which is slower than in the sample measured at 400 °C.

The evolution of m-parameter in Mg-23Zn-2Y is less complicated than in WE43 alloy. The m-parameter is high at slow strain rate and gradually decreases at a

higher strain rate at all measured temperatures. Mg-23Zn-2Y has the highest m- parameter between strain rates 10^-4 s^-1 and 10^-3 s^-1. The highest strain rate, where m>0,3 is 10^-2 s^-1 for both states. The highest m-parameter of both states at a high strain rate was measured 300 °C and the lowest m at 200 °C. Higher temperature appears to be more convenient for high strain rate superplasticity. The biggest difference between extruded material and material after Ex-ECAP is at low strain rates. The m-parameter of extruded state measured at 200 °C is significantly lower than after Ex-ECAP. It could be caused by the finer microstructure of material prepared by Ex-ECAP, which remains thermally stable up to 200 °C. On the other hand, the extruded state reached higher m-parameter measured at 250 °C than material prepared by Ex-ECAP, because extruded state proved better thermal stability at 250 °C (Figure 16).

5.4.2 Superplastic tensile tests

Deformation to failure tensile tests of WE43 were measured at the

temperature range from 350 °C to 450 °C and the strain rate range from 10^-2 s^-1 to 10^-

1 s^-1 (Figure 21). The maximal elongation achieved during superplastic deformation is shown in Table 4. For the strain rate of 10-1 s-1, the maximal elongation of 1013 % was achieved at the temperature of 400 °C. For the strain rate of 10-2 s-1, the maximal elongation of 1232 % was achieved for two temperatures 350 °C and 400 °C. These results correspond with the m-parameter measurement, where the highest m-parameter for these strain rates was achieved at 400 °C followed by 350

°C (Figure 19). Grain boundary sliding and diffusion creep at 350 °C were not as active as at 400 °C, because both processes are thermally activated. Lower

deformation to failure 799 % was achieved and high yield stress occurred. Because both deformation processes are slower at lower temperatures, WE43 alloy was able to reach high elongation of 1233 % only at a slower strain rate of 10^-2 s^-1 at 350 °C, but the yield stress was still higher than measured at 400 °C. On the other hand, the results showed that the temperature of 450 °C was too high for superplastic

deformation. According to [36] and thermal stability of microstructure (Figure 16), significant grain growth occurs at this temperature. Increase in the grain size during temperature stabilization (prior to the begin of the test (Figure 18)) resulted in higher yield stress due to the decreased activity of grain boundary sliding, as depicts Figure

21. Consequently, maximal elongation was lower than the one measured at 400 °C for both strain rates. Moreover, at a strain rate of 10-1 s-1 sharp yield point appeared, which suggest that the dislocation deformation mechanism was the most active.

Upon the literature review, such excellent high strain rate superplasticity hasn’t been reported in WE43 alloy yet [20–22, 34, 52]. It was accomplished thanks to exceptional ultrafine-grained microstructure prepared by ECAP [35, 36]. The best elongation achieved so far has been 860% and 960% at 10⁻¹ s⁻¹ and 2×10⁻² s⁻¹, respectively [20].

Tensile tests of Mg-23Zn-2Y alloy prepared by extrusion are shown in Figure 22 and Ex-ECAP in Figure 23. The maximal elongation of 656 % was achieved in Mg-23Zn-2Y_Ex at a strain rate of 10⁻³ s⁻¹ at 300 °C (Table 4). The extruded state reached better elongation at 300 °C than material prepared by Ex-ECAP. On the other hand, alloy after Ex-ECAP proved better superplasticity at 250 °C at both strain rates. It could be caused by finer initial microstructure after Ex-ECAP. At 300 °C, grains grow in both microstructures (Figure 16) and the advantage of a smaller initial grain size disappears. Tensile tests at 350 °C were also performed, but the

temperature was well above the microstructure stability and very little elongation was achieved. Better elongation was achieved at a slower strain rate of 10⁻³ s⁻¹. For that reason, even slower strain rate tensile test was performed at 5x10⁻⁴ s⁻¹. Despite slower strain rate of this time-consuming tensile test, only a few percent better elongation was obtained. That was the main reason why more tensile tests at 5x10⁻⁴ s⁻¹ weren’t performed at different conditions. As was observed in WE43 alloy, tensile tests at lower temperature have higher yield stress. Also, yield stress was higher at higher strain rates because more stress is needed to deform the material.

Both states of Mg-23Zn-2Y alloy showed similar superplastic behavior, moreover alloy after extrusion achieved better superplastic deformation. The ECAP processing after extrusion wasn’t effective enough and ECAP parameters would have to be optimized to make processing by ECAP more rewarding. On the other hand, 656 % of elongation achieved at a strain rate of 10⁻³ s⁻¹ in just extruded material is quite a good result among other MgZnY studies [23–25, 39–42, 53]. Extrusion is

simple material processing and can be implemented right after casting, which has advantages for practical applications.

Superplasticity of Mg-23Zn-2Y could be improved by ECAP optimization, as was mentioned, or by changing the chemical composition of the alloy. Chemical composition obtained is ideal for a maximum amount of quasicrystalline I-phase, but the secondary phase particles were quite large and ECAP wasn’t sufficient in the fragmentation of these particles. Alloying by Zirconium could help with decreasing the grain size as well as the particle size. Microstructure with an average grain size of 0,6 µm and secondary I-phase particles of 0.3-0,8 µm was achieved by Ex-ECAP processing in Mg–5.00Zn–0.92Y–0.16Zr (wt.%) [42]. In that study, elongation 865%

was obtained at 200 °C and a strain rate of 1.67x10⁻³ s⁻¹. The best superplasticity among MgZnY(Zr) alloys was observed in Mg–7.12Zn–1.2Y–0.84Zr (wt.%)

processed by friction stir processing. The authors contribute their success to achieve the maximal elongation of 1110% at 10⁻² s⁻¹ and 450 °C to the microstructure with fine, uniformly distributed W-phase particles [53]. The alloy with the closest chemical composition to the investigated one was Mg-13Zn-1.55Y. The maximum elongation of 1021 % at 10⁻³ s⁻¹ was obtained thanks to fragmentation of I-phase particles to a typical size of 0.2–0.5 μm by high-ratio differential speed rolling [23].

5.5. Microstructure after deformation

The microstructure after tensile tests at 10^-2 s^-1 was investigated by SEM and EBSD. Cavities, formed in the stripes of particles, appeared in all observed Mg- 23Zn-2Y samples (Figure 24). These stripes were created during extrusion (Figure 7) and stand not only ECAP processing but also superplastic deformation.

EBSD maps of microstructure after deformation are shown in Figure 26 and Figure 27. The microstructure of strained samples prepared by extrusion and Ex- ECAP is similar. The stripes of the grains with similar size appeared in the direction of the deformation. These stripes are separated by the stripes of secondary phase particles. The average grain size after deformation is also comparable in material prepared by extrusion and Ex-ECAP (Table 5). The average grain size is increasing with the temperature of performed tensile test, as was expected.

In WE43 cavities appeared only after deformation at 450 °C, which is the highest temperature investigated (Figure 25). The cavities formation could be caused by secondary phase particles aggregation [36] and grain growth at high temperature.

EBSD maps of WE43 showed microstructure with elongated grains in the direction of deformation in samples strained at 350 °C and 400 °C (Figure 28). At 450 °C, the microstructure is recrystallized completely. The average grain size increase with the temperature of deformation tests (Table 5).

6. Conclusion

Superplastic behavior of magnesium alloys Mg-4Y-3RE (wt.%) prepared by ECAP and Mg-23Zn-2Y (wt.%) prepared by extrusion and ECAP were investigated in this thesis. The microstructure of Mg-23Zn-2Y alloy was investigated by scanning and transmission electron microscopy. An average grain size of 3.2 µm and 1.6 µm was achieved by extrusion and Ex-ECAP, respectively. The microstructure with high fraction of secondary I-phase particles was observed. Processing of Mg-4Y-3RE through ECAP resulted in grain refinement down to ~340 nm and formation of a high volume fraction of fine secondary phase particles located at the grain

boundaries and triple-points. The microstructure of extruded Mg-23Zn-2Y alloy was thermally stable up to 250 °C and 200 °C in material prepared by Ex-ECAP. The microstructure of Mg-4Y-3RE was thermally stable to 250 °C.

m-parameter was measured in order to find the suitable condition for

superplastic deformation. Mg-4Y-3RE alloy had high m-parameter even at strain rate 10^-1 s^-1. The highest m-parameter was measured at 400 °C. Mg-23Zn-2Y alloy had the highest m-parameter at strain rate of 10^-4 s^-1 - 10^-3 s^-1. The highest strain rate where m>0.3, was 10^-2 s^-1. At this strain rate the highest m-parameter was measured at 300 °C. High strain rate superplasticity in WE43 was achieved at 400 °C with 1013 % of elongation at strain rate 10^-1 s^-1 and 1132 % at 10^-2 s^-1. The extraordinary superplasticity was attributed to the exceptional ultrafine-grained thermally stable microstructure. The highest elongating of 656 % was measured in Mg-23Zn-2Y at

10^-3 s^-1 and 300 °C prepared by extrusion. The highest elongation of 533 % at strain rate of 10^-2 s^-1 was observed is extruded Mg-23Zn-2Y at 300 °C.

Bibliography

[1] ZHENG, Y.F., X.N. GU and F. WITTE. Biodegradable metals. Materials Science and Engineering: R: Reports. 2014, vol. 77, pp. 1–34.

10.1016/j.mser.2014.01.001

[2] MORDIKE, B.L. and T. EBERT. Magnesium Properties — applications — potential. Materials Science and Engineering A. 2001, p. 9.

[3] JR, W.D.C. and D.G. RETHWISCH. Fundamentals of Materials Science and Engineering: An Integrated Approach. 4 edition. Hoboken, N.J: Wiley, 2011.

ISBN 978-1-118-06160-2.

[4] GLADMAN, T. Precipitation hardening in metals. Materials Science and Technology. 1999, vol. 15, no. 1, pp. 30–36. 10.1179/026708399773002782 [5] JONES, R.M. Deformation Theory of Plasticity. B.m.: Bull Ridge

Corporation, 2009. ISBN 978-0-9787223-1-9.

[6] DAWSON, P.R., D.E. BOYCE, J.-S. PARK, E. WIELEWSKI and M.P.

MILLER. Determining the strengths of HCP slip systems using harmonic analyses of lattice strain distributions. Acta Materialia. 2018, vol. 144, pp. 92–

106. 10.1016/j.actamat.2017.10.032

[7] LANGDON, T.G. Seventy-five years of superplasticity: historic developments and new opportunities. Journal of Materials Science. 2009, vol. 44, no. 22, pp.

5998–6010. 10.1007/s10853-009-3780-5

[8] (PDF) Superplasticita a její praktické užití | Pavel Lukac - Academia.edu.

[accessed. 5. April 2019]. Retrieved from:

https://www.academia.edu/32506105/Superplasticita_a_jej%C3%AD_praktic k%C3%A9_u%C5%BEit%C3%AD

[9] NIEH, T.G., J. WADSWORTH and O.D. SHERBY. Superplasticity in Metals and Ceramics. B.m.: Cambridge University Press, 2005. ISBN 978-0-521- 02034-3.

[10] KIM, B., J.G. LEE and S.S. PARK. Superplasticity and load relaxation

behavior of extruded Mg–8Sn–3Al–1Zn alloy at 250°C. Materials Science and Engineering: A. 2016, vol. 656, pp. 234–240. 10.1016/j.msea.2016.01.014 [11] AL-ZUBAYDI, A.S.J., A.P. ZHILYAEV, S.C. WANG and P.A.S. REED.

Superplastic behaviour of AZ91 magnesium alloy processed by high-pressure torsion. Materials Science and Engineering: A. 2015, vol. 637, pp. 1–11.

10.1016/j.msea.2015.04.004

[12] JEONG, H.G., Y.G. JEONG and W.J. KIM. Microstructure and

superplasticity of AZ31 sheet fabricated by differential speed rolling. Journal of Alloys and Compounds. 2009, vol. 483, no. 1, 14th International

Symposium on Metastable and Nano-Materials (ISMANAM-2007), pp. 279–

282. 10.1016/j.jallcom.2008.08.130

[13] PARK, S.S., G.T. BAE, D.H. KANG, B.S. YOU and N.J. KIM. Superplastic deformation behavior of twin-roll cast Mg–6Zn–1Mn–1Al alloy. Scripta Materialia. 2009, vol. 61, no. 2, pp. 223–226.

10.1016/j.scriptamat.2009.03.057

[14] LIN, H.K., J.C. HUANG and T.G. LANGDON. Relationship between texture and low temperature superplasticity in an extruded AZ31 Mg alloy processed by ECAP. Materials Science and Engineering: A. 2005, vol. 402, no. 1, pp.

250–257. 10.1016/j.msea.2005.04.018

[15] YIN, D.L., K.F. ZHANG, G.F. WANG and W.B. HAN. Superplasticity and cavitation in AZ31 Mg alloy at elevated temperatures. Materials Letters. 2005, vol. 59, no. 14, pp. 1714–1718. 10.1016/j.matlet.2005.01.053

[16] CHUVIL’DEEV, V.N., T.G. NIEH, M.Yu. GRYAZNOV, V.I. KOPYLOV and A.N. SYSOEV. Superplasticity and internal friction in microcrystalline AZ91 and ZK60 magnesium alloys processed by equal-channel angular pressing. Journal of Alloys and Compounds. 2004, vol. 378, no. 1, Proceedings of the 9th International Symposium on Physics of Materials (ISPMA 9), pp. 253–257. 10.1016/j.jallcom.2003.10.062

[17] MATSUBARA, K., Y. MIYAHARA, Z. HORITA and T.G. LANGDON.

Developing superplasticity in a magnesium alloy through a combination of extrusion and ECAP. Acta Materialia. 2003, vol. 51, no. 11, pp. 3073–3084.

10.1016/S1359-6454(03)00118-6

[18] TORBATI-SARRAF, S.A., R. ALIZADEH, R. MAHMUDI and T.G.

LANGDON. Evaluating the flow properties of a magnesium ZK60 alloy processed by high-pressure torsion: A comparison of two different miniature testing techniques. Materials Science and Engineering: A. 2017, vol. 708, pp.

432–439. 10.1016/j.msea.2017.10.009

[19] KIM, Y.S. and W.J. KIM. Microstructure and superplasticity of the as-cast Mg–9Al–1Zn magnesium alloy after high-ratio differential speed rolling.

Materials Science and Engineering: A. 2016, vol. 677, pp. 332–339.

10.1016/j.msea.2016.09.063

[20] KANG, Z., L. ZHOU and J. ZHANG. Achieving high strain rate

superplasticity in Mg–Y–Nd–Zr alloy processed by homogenization treatment and equal channel angular pressing. Materials Science and Engineering: A.

2015, vol. 633, pp. 59–62. 10.1016/j.msea.2015.03.011

[21] WATANABE, H., T. MUKAI, K. ISHIKAWA, T. MOHRI, M. MABUCHI and K. HIGASHI. Superplasticity of a Particle-Strengthened WE43

Magnesium Alloy. Materials Transactions - MATER TRANS. 2001, vol. 42, pp. 157–162. 10.2320/matertrans.42.157

[22] CAO, G., D. ZHANG, F. CHAI, W. ZHANG and C. QIU. Superplastic behavior and microstructure evolution of a fine-grained Mg–Y–Nd alloy processed by submerged friction stir processing. Materials Science and Engineering: A. 2015, vol. 642, pp. 157–166. 10.1016/j.msea.2015.06.086

[23] KWAK, T.Y. and W.J. KIM. Superplastic behavior of an ultrafine-grained Mg-13Zn-1.55Y alloy with a high volume fraction of icosahedral phases prepared by high-ratio differential speed rolling. Journal of Materials Science

& Technology. 2017, vol. 33, no. 9, pp. 919–925. 10.1016/j.jmst.2017.05.003 [24] XIE, G.M., Z.Y. MA, L. GENG and R.S. CHEN. Microstructural evolution

and enhanced superplasticity in friction stir processed Mg-Zn-Y-Zr alloy.

Journal of Materials Research. 2008, vol. 23, no. 5, pp. 1207–1213.

10.1557/jmr.2008.0164

[25] YAN, H., R. CHEN and E. HAN. Superplasticity of Mg-5.8Zn-1Y-Zr alloy sheet fabricated by combination of extrusion and hot-rolling processes.

Science in China, Series E: Technological Sciences. 2009, vol. 52, no. 1, pp.

166–171. 10.1007/s11431-008-0293-4

[26] Figure 1 Direct extrusion [29]. ResearchGate. [accessed. 5. April 2019].

Retrieved from: https://www.researchgate.net/figure/Direct-extrusion- 29_fig1_326210297

[27] VALIEV, R.Z. and T.G. LANGDON. Principles of equal-channel angular pressing as a processing tool for grain refinement. Progress in Materials Science. 2006, vol. 51, no. 7, pp. 881–981. 10.1016/j.pmatsci.2006.02.003 [28] Influence of incremental ECAP on the microstructure and tensile behaviour of

commercial purity titanium - ScienceDirect. [accessed. 6. April 2019].

Retrieved from:

https://www.sciencedirect.com/science/article/pii/S187770581735703X [29] MINÁRIK, P., M. LANDA, I. LESNÁ, M. ZEMKOVÁ, E. JABLONSKÁ, B.

HADZIMA, M. JANEČEK and R. KRÁL. Interrelation of Microstructure and Corrosion Resistance in Biodegradable Magnesium Alloys with Aluminum, Lithium and Rare Earth Additions. Acta Physica Polonica A. 2015, vol. 128, no. 4, pp. 491–496. 10.12693/APhysPolA.128.491

[30] MORDIKE, B.L. Creep-resistant magnesium alloys. Materials Science and Engineering: A. 2002, vol. 324, no. 1, pp. 103–112. 10.1016/S0921-

5093(01)01290-4

[31] DRYNDA, A., N. DEINET, N. BRAUN and M. PEUSTER. Rare earth metals used in biodegradable magnesium-based stents do not interfere with

proliferation of smooth muscle cells but do induce the upregulation of

inflammatory genes. Journal of Biomedical Materials Research. Part A. 2009, vol. 91, no. 2, pp. 360–369. 10.1002/jbm.a.32235

[32] FEYERABEND, F., J. FISCHER, J. HOLTZ, F. WITTE, R. WILLUMEIT, H.

DRÜCKER, C. VOGT and N. HORT. Evaluation of short-term effects of rare earth and other elements used in magnesium alloys on primary cells and cell lines. Acta Biomaterialia. 2010, vol. 6, no. 5, The THERMEC’2009

Biodegradable Metals, pp. 1834–1842. 10.1016/j.actbio.2009.09.024

[33] DING, Y., C. WEN, P. HODGSON and Y. LI. Effects of alloying elements on the corrosion behavior and biocompatibility of biodegradable magnesium

alloys: a review. Journal of Materials Chemistry B. 2014, vol. 2, no. 14, pp.

1912–1933. 10.1039/C3TB21746A

[34] KANDALAM, S., R.K. SABAT, N. BIBHANSHU, G.S. AVADHANI, S.

KUMAR and S. SUWAS. Superplasticity in high temperature magnesium alloy WE43. Materials Science and Engineering: A. 2017, vol. 687, pp. 85–

92. 10.1016/j.msea.2016.12.129

[35] MINÁRIK, P., J. VESELÝ, R. KRÁL, J. BOHLEN, J. KUBÁSEK, M.

JANEČEK and J. STRÁSKÁ. Exceptional mechanical properties of ultra-fine grain Mg-4Y-3RE alloy processed by ECAP. Materials Science and

Engineering: A. 2017, vol. 708, pp. 193–198. 10.1016/j.msea.2017.09.106 [36] MINÁRIK, P., J. VESELÝ, J. ČÍŽEK, M. ZEMKOVÁ, T. VLASÁK, T.

KRAJŇÁK, J. KUBÁSEK, R. KRÁL, D. HOFMAN and J. STRÁSKÁ. Effect of secondary phase particles on thermal stability of ultra-fine grained Mg-4Y- 3RE alloy prepared by equal channel angular pressing. Materials

Characterization. 2018, vol. 140, pp. 207–216. 10.1016/j.matchar.2018.04.006 [37] GRÖBNER, J., A. KOZLOV, X.Y. FANG, J. GENG, J.F. NIE and R.

SCHMID-FETZER. Phase equilibria and transformations in ternary Mg-rich Mg–Y–Zn alloys. Acta Materialia. 2012, vol. 60, no. 17, pp. 5948–5962.

10.1016/j.actamat.2012.05.035

[38] BAE, D.H., S.H. KIM, D.H. KIM and W.T. KIM. Deformation behavior of Mg–Zn–Y alloys reinforced by icosahedral quasicrystalline particles. Acta Materialia. 2002, vol. 50, no. 9, Computational Thermodynamics and Materials Design, pp. 2343–2356. 10.1016/S1359-6454(02)00067-8

[39] BAE, D.H., Y. KIM and I.J. KIM. Thermally stable quasicrystalline phase in a superplastic Mg–Zn–Y–Zr alloy. Materials Letters. 2006, vol. 60, no. 17, pp.

2190–2193. 10.1016/j.matlet.2005.12.096

[40] ZHENG, M.Y., S.W. XU, K. WU, S. KAMADO and Y. KOJIMA.

Superplasticity of Mg–Zn–Y alloy containing quasicrystal phase processed by equal channel angular pressing. Materials Letters. 2007, vol. 61, no. 22, pp.

4406–4408. 10.1016/j.matlet.2007.02.013

[41] TANG, W.N., R.S. CHEN and E.H. HAN. Superplastic behaviors of a Mg–

Zn–Y–Zr alloy processed by extrusion and equal channel angular extrusion.

Journal of Alloys and Compounds. 2009, vol. 477, no. 1, pp. 636–643.

10.1016/j.jallcom.2008.10.089

[42] XU, S.W., M.Y. ZHENG, S. KAMADO and K. WU. The microstructural evolution and superplastic behavior at low temperatures of Mg–5.00Zn–

0.92Y–0.16Zr (wt.%) alloys after hot extrusion and ECAP process. Materials Science and Engineering: A. 2012, vol. 549, pp. 60–68.

10.1016/j.msea.2012.03.116

[43] Vickers Hardness Test. [accessed. 5. April 2019]. Retrieved from:

https://www.gordonengland.co.uk/hardness/vickers.htm