Mechanical Behaviour and Deformation Mechanisms of Zn-Al-Cu-Mg Alloys

Mechanical behaviour and deformation mechanisms of Zn-Al-Cu-Mg alloys

Von der Fakultät für Georessourcen und Materialtechnik der Rheinisch-Westfälischen Technischen Hochschule Aachen

zur Erlangung des akademischen Grades eines Doktors der Ingenieurwissenschaften

genehmigte Dissertation

vorgelegt von

M.Sc. Zhicheng Wu

aus Quyuan, Provinz Hunan, China

Berichter: Univ.-Prof. Dr. Sandra Korte-Kerzel Univ.-Prof. Dr.-Ing. Sebastian Münstermann

Tag der mündlichen Prüfung: 05.September 2018

Diese Dissertation ist auf den Internetseiten der Universitätsbibliothek online verfügbar

To my family

Acknowledgements

In the beginning, I would like to express my appreciation to all people who have helped me with my doctoral thesis.

Foremost, I would like to express my deepest thanks and sincere appreciation to my advisor Prof. Dr. Sandra Korte-Kerzel, Institute of Physical Metallurgy and Metal Physics (IMM), RWTH Aachen University, for her expert, creative and comprehensive guidance, and encouragement throughout my PhD. My greatest gratitude also goes to Prof. Dr. Günter Gottstein, Institute of Physical Metallurgy and Metal Physics (IMM), RWTH Aachen University, for giving me the opportunity to work at IMM, for suggesting the topic of this dissertation, and for his kind supervision in the first two years of my PhD. My special thanks to my second examiner Prof. Dr. Sebastian Münstermann, Steel Institute (IEHK), RWTH Aachen University, and the chair of my examination committee Prof. Dr. Dieter Georg Senk, Steel Institute (IEHK), RWTH Aachen University, for their insightful comments and discussions.

I would like to thank Dr. Stefanie Sandlöbes and Dr. Weiping Hu for the inspiring discussions, insightful corrections, valuable comments and continuous encouragement. Very special thanks to Dr. Stefanie Sandlöbes, for her unswerving support not only in my research, but also in my personal life.

I would like to thank my colleagues at IMM for the stimulating discussions, for the introductions and support to experimental equipment, and for all the great time we have had during my PhD period. It was fantastic to work together with them at IMM. In particular, I am grateful to David Beckers for the support in metallography, Thomas Burlet for setting up the creep testing and in-situ testing devices, Jann-Erik Brandenburg for the introduction to SEM, Dr. James Gibson for the technical support and inspiring discussions in nanoindentation, Sebastian Schröders for the introduction to FIB and AFM, Fengxin Mao for the help with MATLAB scripts, and Arndt Ziemons for the alloys casting.

I am also grateful to Prof. Dr. Rainer Schmid-Fezter and Dr. Song-mao Liang at TU Clausthal for the cooperation and suggestions for alloy design in our joint projects, Prof. Dr. Christophe Tromas at CNRS-Université de Poitiers for the support with AFM experiments, Xiaoxiao Li at IEHK, RWTH Aachen University for the help with MATLAB scripts, Dr. Konda Gokuldoss Pradeep, Dr. Marcus Hans and Marshal Amalraj at Material Chemistry (MCh), RWTH Aachen University for the support with APT experiments.

I would like to extend my thanks to my student assistants Liang Wu, Jing Rao, Yufengnan Wang, An Zhang, Kuan Ding, Shuo Wang and Shanyu Chen for their great work in their theses and hiwi jobs.

i Acknowledgements

A very special gratitude goes to the German Research Foundation (DFG) for the financial support for my work within projects GO 335/47-1 and KO 4603/5-1. I am also grateful to Federation of German foundry industry e. V. (BDG) for the invitation to industry meet ups on Zn castings, and Dr. Didier Rollez at Grillo-Werke AG for his insightful comments from industry point of view.

Finally, I would like to explain my profound gratitude to my family and my friends, especially my parents, who have supported me along the way.

Thanks for all your encouragement and support!

ii Contents

Contents

Acknowledgements ...... i Contents ...... iii List of tables ...... vi List of figures ...... vii List of symbols...... xi List of abbreviations ...... xii 1. Introduction ...... 1 2. Theoretical background ...... 4 2.1. Zinc and Zinc alloy ...... 4 2.1.1. Alloying elements in Zn ...... 4 2.1.2. Application of Zn and Zn based alloys ...... 6 2.2. Deformation mechanisms ...... 7 2.2.1. Dislocation slip ...... 7 2.2.2. Kink bands ...... 10 2.2.3. Deformation twinning ...... 11 2.2.4. Grain and phase boundary sliding ...... 13 2.2.5. Creep ...... 14 2.2.6. Activation of deformation mechanisms ...... 21 2.2.6.1. Thermal activation of dislocations...... 21 2.2.6.2. Activation volume ...... 23 2.2.6.3. Activation energy ...... 24 2.2.7. Deformation mechanism maps ...... 25 2.3. Nanoindentation ...... 30 3. Experimental methods and materials ...... 34 3.1. Materials ...... 34 3.2. Sample preparation ...... 34 3.3. Macroscopic mechanical testing ...... 35 3.3.1. Constant strain rate tensile testing ...... 35 3.3.2. Strain rate jump test ...... 36 3.3.3. Creep test ...... 36 3.3.4. In-situ straining experiments ...... 37 3.4. Nano-indentation experiments ...... 38 3.4.1. Constant strain rate and strain rate jump tests ...... 38 3.4.2. Nanoindentation creep tests ...... 39 3.5. Microstructure characterisation ...... 40

iii Contents

3.6. Digital image correlation (DIC) measurements ...... 40 4. Macroscopic mechanical response and mechanisms ...... 42 4.1. Results and discussion ...... 42 4.1.1. As-cast microstructure ...... 42 4.1.2. Mechanical properties ...... 46 4.1.3. Fracture surfaces ...... 48 4.1.4. Deformation microstructure ...... 51 4.1.4.1. Room temperature and/or highest strain rate tested (5∙10−4 s-1) ...... 51 4.1.4.2. Elevated temperature and/or low strain rate ...... 55 4.2. Conclusions...... 59 5. Creep of ZnAl4Cu1 alloys ...... 60 5.1. Results ...... 60 5.1.1. Uniaxial tensile creep experiments ...... 60 5.1.2. Creep microstructure ...... 65 5.1.3. Creep behaviour of individual microstructural constituents ...... 68 5.2. Discussion ...... 70 5.2.1. Creep mechanisms ...... 70 5.2.1.1. Macroscopic uniaxial tensile creep ...... 70 5.2.1.2. Nanoindentation creep ...... 72 5.2.2. Linking micro- and macroscale deformation: geometrical constraints model....75 5.2.3. Influence of Mg on the creep properties of ZnAl4Cu1Mg alloys ...... 78 5.3. Conclusions...... 80 6. Micromechanical response and mechanisms ...... 82 6.1. Results ...... 82 6.1.1. Local mechanical properties ...... 82 6.1.1.1. Constant strain rate indentation ...... 82 6.1.1.2. Nanoindentation strain rate jump test ...... 83 6.1.2. Deformation microstructure ...... 86 6.1.2.1. Deformation microstructure in primary η-Zn phase ...... 86 6.1.2.2. Deformation microstructure in η-Zn + α-Al eutectoid structures ...... 88 6.1.2.3. Strain partitioning ...... 90 6.2. Discussion ...... 93 6.2.1. Local mechanical properties of individual microstructural constituents ...... 93 6.2.2. Deformation mechanisms and thermal activation ...... 94 6.2.2.1. Deformation of the primary η-Zn phase ...... 94 6.2.2.2. Deformatlion of η-Zn + α-Al eutectoid structures ...... 95

iv Contents

6.2.3. Local strain distribution and strain transfer ...... 96 6.2.4. Bulk deformation ...... 98 6.3. Conclusion ...... 99 7. Precipitation behaviour ...... 101 7.1. Results and discussion ...... 101 7.2. Conclusions...... 108 8. Summarising discussion and concluding remarks ...... 109 8.1. Temperature dependent activity of deformation mechanisms ...... 109 8.2. Micro- and macro-mechanical behaviour ...... 110 8.3. Fracture behaviour ...... 110 8.4. Creep behaviour ...... 111 8.5. Effect of Mg ...... 112 8.6. Precipitates and mechanical behaviour ...... 112 Bibliography ...... 113 Abstract ...... 126 Kurzzusammenfassung ...... 128

v List of tables

List of tables

Table 2.1. Important physical properties of zinc at room temperature [1, 33]...... 4 Table 2.2. Slip systems in zinc [72-74]...... 9 Table 2.3. Twinning elements in hexagonal structures [87]...... 11 Table 2.4: Creep parameters of Zn and Zn alloys reported in the literatures...... 18 Table 2.5. Activation energy of obstacles [83]...... 24 Table 2.6. Activation energies in Zn and Zn based alloys...... 24 Table 2.7. Deformation mechanisms in Zn and Zn alloys reported in literature. The corresponding references are included in the table...... 27 Table 3.1. Chemical composition of the zinc alloys investigated...... 35 Table 3.2. Applied parameters (temperature, stress) during uniaxial creep experiments*. ....37

Table 3.3. Detailed procedure of depositing SiO2 nanoparticles on the sample surface...... 41 Table 4.1. Grain size and phase fractions of investigated ZnAl4Cu1 alloys...... 45

Table 4.2. Mechanical properties of as-cast ZnAl4Cu1 alloys; 휎0.2 yield strength, 휀푓 elongation to fracture...... 48 Table 4.3. The area fraction (%) of ductile fracture under tensile deformation...... 51 Table 5.1. Time (in hours) to 1% creep strain of ZnAl4Cu1MgX (X = 0.04, 0.21, 0.31, in wt.%) alloys at 55°C, 85°C and 105°C and stresses of 0.6 – 0.8∙σ0.2...... 62 Table 5.2. The minimum creep rate 휀̇ (10-4h-1) of ZnAl4Cu1MgX (X = 0.04, 0.21, 0.31, in wt.%) alloys at 55 – 105°C and stresses of 0.6 – 0.8∙σ0.2...... 62 Table 5.3. Calculated shear moduli G of ZnAl4Cu1MgX (X = 0.04, 0.21, 0.31, in wt.%) alloys at different temperatures, [GPa]...... 63 Table 5.4. Creep stress exponents and activation energies of ZnAl4Cu1MgX (X = 0.04, 0.21, 0.31, in wt.%) alloys at 25 – 105°C determined from uniaxial tensile creep tests...... 64 Table 5.5. Nanoindentation creep stress exponents of different individual microstructural components at 25°C and 85°C at maximal loads of 15 – 35 mN...... 70 Table 5.6. Lattice diffusion coefficient of various elements in Zn-Al-Cu-Mg alloys [83]. The diffusion coefficients at 25°C and 85°C were calculated from the data in [83]...... 74 Table 5.7. Boundary diffusion coefficient of various elements in Zn-Al-Cu-Mg alloys [83]. The diffusion coefficients at 25°C and 85°C were calculated from the data in [83]...... 74 Table 6.1. Hardness, H, and elastic modulus, E, of individual microstructural constituents in ZnAl4Cu1Mg0.31 investigated at 25 ºC and 85°C, respectively. The relative values are given as a reduction of the mean value (e.g. 1.2 GPa for uncorrected Zn) with temperature to indicate the temperature dependence...... 83 Table 6.2. Nanoindentation strain rate sensitivity and activation volume of the primary η-Zn phase and η-Zn + α-Al eutectoid structures at RT and 85°C compared to bulk ZnAl4Cu1Mg0.31 at 85°C...... 86 Table 6.3. Average surface heights of Zn-Al phase boundaries in eutectoid structures around nano-indentation imprints. The indentations were performed at a maximum load of 25 mN at 25°C and 85 ºC at a strain rate of 0.1 s-1, and at 85°C at a strain rate of 0.01 s-1. For comparison the average height of Zn-Al phase boundaries in eutectoid structures in un-deformed material is also listed...... 90

vi List of figures

List of figures

Figure 1.1. Graphical abstract...... 3 Figure 2.1. Zn-Al equilibrium phase diagram [25, 38]...... 5 Figure 2.2. Atomic configuration of (a) an edge dislocation and (b) a screw dislocation [59]. . 8 Figure 2.3. Movement of an edge dislocation: the arrows indicate the applied shear stress [62]...... 8 Figure 2.4. Slip systems in the hexagonal structure of zinc: (a) basal slip; (b) prismatic slip; (c) pyramidal slip; and (d) pyramidal slip [69-71]...... 9 Figure 2.5. Schematic explanation of the formation of deformation kink bands in Zn. (a) Model of deformation kink band in Zn single crystal by [76] and [79]. (b) Schematic illustration on the crystal rotation angle in the deformation band [79]. (c) An example of kind bands in the Zn grains of ZnAlCuMg0.31 alloy after tension strain (current work)...... 10 Figure 2.6. Dependence of the twinning shear on the axial ratio in hexagonal metals [3]. ....12 Figure 2.7. Grain boundary sliding between two adjacent grains under tensile stress [93]. ...13 Figure 2.8. Schematic image of typical (a) creep curves and (b) creep rate curves [59]...... 15 Figure 2.9. Creep by diffusional transport through lattice or grain boundary [83]...... 16 Figure 2.10. Schematic illustration of the thermally activated glide process of a dislocation overcoming the Peierls barrier via the double kink mechanism. Reproduced from [61, 140]. 21 Figure 2.11. Deformation mechanism map of pure zinc with a grain size of 100 μm [83]...... 25 Figure 2.12. Deformation mechanism map of normalized grain size versus normalized stress for Zn–22% Al alloys tested at 200°C, reproduced from [160]...... 26 Figure 2.13. Geometries of commonly used indenters: (a) Spherical indenter; (b) Conical indenter; (c) Vickers indenter; (d) Berkovich indenter [165]...... 30 Figure 2.14. (a) Schematic illustration of the indenter and the specimen surface at full load and unloaded for a Berkovich indenter; (b) Schematic load versus displacement curve for elastic- plastic loading followed by elastic unloading. At the maximum load, Pmax, hmax is the depth from the original specimen surface to the tip of the indenter, hc is the contact depth of indenter and material and hs is the distance from the edge of the contact to the specimen surface. hf is the final depth of the residual impression after indentation. Upon elastic reloading, the eventual point of contact with the specimen surface moves through a distance hs [165, 167, 173]. ....31 Figure 3.1. Sketch of the samples for constant strain rate tensile tests...... 35 Figure 3.2. Geometry of specimens for tensile strain rate jump tests, in mm...... 36 Figure 3.3. Geometry of specimens for tensile creep tests, unit in mm...... 37 Figure 3.4. Geometry of specimens for in-situ straining experiments, unit in mm...... 38 Figure 3.5. Secondary electron (SE) micrographs of the region of interest prior to deformation. (a) In-lens SE image of the whole region of interest. (b) In-lens SE image of one individual image, enlarged from the red box in the top left corner in (a). (c) Enlarged in-lens SE micrograph of the region highlighted by the red box in (b), showing SiO2 particle speckle pattern on the sample surface...... 41 Figure 4.1. Microstructures of as-cast ZnAl4Cu1 alloys (BSE); (a) Z410, (b) ZnAl4Cu1Mg0.04, (c) ZnAl4Cu1Mg0.21, (d) ZnAl4Cu1Mg0.31...... 43 Figure 4.2. EDS analysis of alloys (a) ZnAl4Cu1Mg0.04 and (b) ZnAl4Cu1Mg0.21...... 44 Figure 4.3. True stress – true strain curves of as-cast laboratory alloys ZnAl4Cu1Mg0.04, ZnAl4Cu1Mg0.21, ZnAl4Cu1Mg0.31 and as-cast Z410 alloy (tested at lowest/highest rates and temperatures only). The legend given in (a) is valid for all diagrams, while testing conditions (strain rate, temperature) are given in the figures...... 47 Figure 4.4. Fracture surfaces of ZnAl4Cu1 alloys deformed at room temperature; (a) alloy

vii List of figures

viii List of figures at 85°C and 93 MPa. The corresponding primary η-Zn grains are marked by dashed rectangles in Figure 5.4 (e, f) are given with a higher magnification due to higher local misorientation angles which are not uniquely revealed at lower magnifications...... 67 Figure 5.6. Micrographs of individual microstructural constituents in ZnAl4Cu1Mg0.31 after indentation with a load of 25 mN at 25°C: (a) primary Zn, (b) η-Zn + α-Al eutectic structure, (c)

η-Zn + α-Al eutectoid structure, (d) Mg2Zn11 embedded in eutectoid structure ...... 68 Figure 5.7. Indentation depth increase (h) as a function of time (t) during the holding period of nanoindentation creep tests at different loads in primary η-Zn phase at (a) 25°C and (b) 85°C...... 68 Figure 5.8. Indentation depth as a function of time during the holding period of the individual microstructural constituents at 25 mN at (a) 25°C and (b) 85°C...... 69 Figure 5.9. Log-log plot of indentation strain rate as a function of the nominal contact pressure, pnom, for nanoindentation creep tests of the different individual microstructural components at 25 mN load at (a) 25°C and (b) 85°C...... 70 Figure 5.10. Typical microstructure of (a) ZnAl4Cu1Mg0.04; (d) ZnAl4Cu1Mg0.21; and (h) ZnAl4Cu1Mg0.31 with highlighted eutectic and eutectoid colony boundaries and corresponding schematic alignment of individual constituents in (b, c) ZnAl4Cu1Mg0.04; (e, f) ZnAl4Cu1Mg0.21; and (i, g) ZnAl4Cu1Mg0.31, where the black blocks represent primary η-Zn grains, blue lines represent interfaces in eutectic lamellar colonies and red lines represent interfaces in eutectoid colonies...... 77 Figure 5.11. Creep strain-time curves of ZnAl4Cu1MgX (X = 0.04, 0.21, 0.31, in wt.%) alloys at the same stress level at (a) 55°C, (b) 85°C and (c) 105°C...... 78 Figure 6.1. Dependence of the true stress on the strain rate of alloy ZnAl4Cu1Mg0.31 obtained by macroscopic tensile strain rate jump tests at 85°C (a). (b) shows how the corresponding stresses upon strain rate jumps from 5∙10-4 s-1 to 6∙10-6 s-1 were determined. The dependence of the hardness of the primary η-Zn phase and η-Zn + α-Al eutectoid structures on the strain rate obtained by indentation strain rate jump tests at RT (c) and 85°C (d)...... 85 Figure 6.2. Post-mortem micrographs of a typical indent in the primary η-Zn phase (RT, 0.1 s-1 strain rate, 25 mN peak load), (a) SE micrograph; (b) IPF map, twin boundaries are highlighted in yellow; (c) AFM micrograph, basal plane traces are highlighted in red; (d) height profile along line 1 in (c); (e) height profile along line 2 in (c) with corresponding AFM micrograph. The dashed white rectangle in (a) shows the region of the AFM map shown in (c)...... 87 Figure 6.3. Post-mortem micrographs of typical indents in the primary η-Zn phase (85°C, 0.1 s-1 strain rate, 25 mN peak load), (a) SE micrograph and (b) IPF map showing deformation twins highlighted by yellow boundaries, the IPF map has been rotated to guide the eyes; (c), (e) SE micrographs and (d), (f) IPF maps showing basal and non-basal slip traces; the insets in (c), (e) are enlarged micrographs of the slip traces...... 88 Figure 6.4. SE micrographs and AFM maps of typical indentation imprints in eutectoid structures and primary Zn phase, the indentations were performed at 25°C at a strain rate of 0.1 s-1 (a)-(c), 85 ºC at a strain rate of 0.1 s-1 (d)-(f), 85°C at a strain rate of 0.01 s-1 (g)-(h) and a maximum load of 25 mN. (a), (d), (g) are SEM images with imposed deformation traces; (b), (e), (h) are the corresponding AFM images with red arrows pointing to the surface traces; (c) and (f) are the corresponding IPF maps...... 89 Figure 6.5. SE micrographs of the region of interest (ROI) in alloy ZnAl4Cu1Mg0.31 at (a) 2% and (b) 5% tensile strain at 85°C; (c) BSE micrograph of the ROI at 0 strain, primary η-Zn grains are numbered by 1-15 (highlighted in red) and η-Zn + α-Al eutectic colonies labelled A- N (highlighted in blue); (d) IPF figure of the ROI at 0 strain...... 91 Figure 6.6. Local von Mises strain map of the region of interest in alloy ZnAl4Cu1Mg0.31 at (a, c, e) 2%, (b, d, f) 5% tensile strain at 85°C. (a, b) Strain maps calculated using the software

ix List of figures

GOM correlate (V8.1, GOM mbH).; (c, d) strain maps calculated using the non-rigid registration method; (e, f) DIC strain maps are overlaid with SE images of the microstructure...... 92 Figure 6.7. Enlarged DIC strain maps show strain transfer between primary η-Zn grains 1, 2 and eutectic colony A (a), and between grains 6, 7 and colonies O, K (b); (c) schematic displays the orientation of slip systems in Zn grain 9 (2nd order pyramidal plane traces observed) and eutectic colony O (basal traces observed) across the interface...... 98 Figure 7.1. True stress – true strain curves of ZnAl4Cu1Mg0.31 alloy at room temperature and at 85°C and at a strain rate of 5∙10−4 s-1(a) and the corresponding fracture surfaces of ZnAl4Cu1Mg0.31 alloy deformed at RT (b) and at 85°C (c) and strain rate of 5∙10-4 s-1...... 102 Figure 7.2. EDS analysis of alloy ZnAl4Cu1Mg0.31...... 103 Figure 7.3. TEM micrographs revealing precipitates in primary η-Zn grains of alloy ZnAl4Cu1Mg0.31 after deformation at RT (a, b) and 85°C (c, d); (a, c) TEM bright field micrographs show lenticular-shaped precipitates; (b, d) TEM dark field micrographs show the needle-shaped precipitates, the micrographs were taken from the spots marked by red circles; (e) diffraction patterns of the needle-shaped precipitates (b, d) and the primary η-Zn grain matrix with zone axis B=[011̅0], B=[0001], B=[112̅0]...... 104 Figure 7.4. 3-D elemental maps of alloy ZnAl4Cu1Mg0.31 deformed at RT showing the 3-D positions of (a, b) Al atoms, (h) Mg atoms and (i) Cu atoms in a primary η-Zn grain. The isoconcentration surfaces of 30 at.% Al (a, b), 0.7 at.% Mg (h) and 18 at.% Cu (i) are also shown. (c-g, j) Plots from different precipitates regions (marked by yellow) give the local chemical compositions...... 105 Figure 7.5. TEM bright field micrographs of alloy ZnAl4Cu1Mg0.31 deformed at RT: (a) area close to fracture surface; (b) enlarged images of the area marked by dashed yellow rectangle in (a), the insets show the diffraction patterns of the area marked by yellow circles...... 106 Figure 7.6. TEM dark field micrographs of alloy ZnAl4Cu1Mg0.31 after deformation at 85°C: (a) high strain sample (area close to fracture surface) with g=(112̅2); (b) low strain sample (area away from fracture surface) with g=(112̅2), (c) low strain sample with g=(0002), (d) low strain sample with g=(011̅0), (e) low strain sample with g=(011̅1); g: diffraction vector. Red arrows mark the dislocations on basal planes and orange arrows mark the dislocations on pyramidal planes...... 107

x List of symbols

List of symbols

1 Twinning shear direction Pmax Maximum indentation load

2 Conjugate twinning direction pnom Nominal contact pressure m' Luster-Morris parameter Q Activation energy R Universal gas constant UP Peierls potential S Stiffness, Plane of shear in twinning, A, ACO, ANH Material constant Ac Projected contact area Shear strain b Burgers vector T Temperature B Zone axis, Indentation pre-exponential T’ Al4Cu3Zn phase term Tm Melting point d Grain size, Lattice spacing V Activation volume * D Grain size V Apparent activation volume

DV Bulk self-diffusion coefficient νi Poisson's ratio of indenter material

D0 Pre-exponential factor for self-diffusion α Al-rich FCC phase

DGB Coefficient of grain boundary α' supersaturated α phase diffusion α1 Power law fitting constant E Young's modulus β Zn-rich FCC phase

Ei Young's modulus of indenter material β’ Zn-rich FCC phase

Er Residual modulus γ̇, ε̇ Strain rate g Geometrical factor δ Thickness of grain boundary G Shear modulus ε Strain, Intermetallic CuZn4 phase h Indentation depth εf Elongation to fracture H Hardness ϵ’ Geomtrical constant hc Contact depth ε’ Strain rate, creep rate hf Final indentation depth η Zn-rich hexagonal phase

HM Meyer hardness η’ supersaturated η phase hmax Depth from the original surface to the ν Poisson ratio

tip of the indenter ρm Mobile dislocation density hs Depth from the edge of the contact to σ, τ Stress the specimen surface τ0 Critical resolved shear stress k Boltzmann constant σ0.2 Yield strength K1 Twinning plane τP Perierls stress K2 Conjugate twinning plane φ Angle between two normal directions of l Dislocation length slip / twining planes m Strain rate sensitivity Ω Atomic volume 1 m Power law fitting constant ωA Frequency n Stress exponent ωD Debye frequency N Indentation stress exponent 휅 Angle between two slip / twinning P Load, Probability directions Ṗ Loading rate

xi List of abbreviations

List of abbreviations

AFM Atomic force microscopy APT Atom probe tomography BSE Backscattered electron microscopy CRSS Critical resolved shear stress DIC Digital image correlation DZM Electromechanical testing machine EBSD Electron backscatter diffraction EDS Energy dispersive X-ray spectroscopy FCC Face centered cubic structure HCP Hexagonal close packed structure IPF Inverse pole figure GBS Grain boundary sliding GROD Grain reference orientation deviation PB Phase boundary PBS Phase boundary sliding RT Room temperature ROI Region of interest SADP Slected area diffraction pattern SE Secondary electron SEM Scanning electron microscopy TD Tensile direction TEM Transmission electron microscopy UFG Ultra-fine grained

xii 1. Introduction

1. Introduction

Zinc is the most widely used metal in industrial production after steel, aluminium and copper and the 24th most abundant element in the earth’s crust [1]. However, Zn in its unalloyed form is relatively soft and brittle at room temperature. This is due to the low symmetry, high mechanical anisotropy of its hexagonal crystal structure and insufficient independent deformation systems for compatible polycrystalline deformation [2, 3]. Alloying with Al and Cu was shown to improve the mechanical properties [4-9] and Zn based alloys show attractive physical and mechanical properties such as high strength and wear resistance, excellent fluidity and castability, good surface quality as well as outstanding corrosion resistance [4, 10- 12]. Therefore, Zn-Al based alloys are widely used in the production of structural and decorative parts for automotive, architectural, electrical and electronic applications as well as machinery and equipment with a complex geometry or equipment needing a high manufacturing precision [9, 12-16]. In 2013 approximately 55,000 tons of zinc die castings products were produced in Germany, and about 453,000 tons were produced worldwide. However, wider application of Zn alloys is hindered by:

i. low creep resistance at moderately elevated temperatures [17, 18], ii. long-term mechanical and dimensional instability at ambient or slightly elevated temperatures [11, 19-21].

Efforts were therefore made to improve the creep resistance and mechanical stability of Zn alloys. To this end, an essential understanding of the mechanical properties and the corresponding plasticity mechanisms, creep mechanisms as well as precipitation behaviour in Zn alloys are necessary. Zn-Al based alloys solidify through a eutectic reaction followed by a eutectoid reaction upon cooling, forming a multi-phase microstructure composed of η-Zn and η-Zn + α-Al eutectic / eutectoid structures, where η-Zn is a Zn-rich hexagonal phase and α-Al is an Al-rich FCC phase. Therefore, to fully understand the mechanical behaviour and deformation mechanisms of Zn-Al alloys, it is also necessary to investigate both, the locally intrinsic properties of individual microstructural constituents as well as the mutual effect of these constituents as an aggregate.

Therefore, the thesis comprises the following four parts:

- Chapter 4 presents the macroscopic mechanical properties of three eutectic ZnAl4Cu1 alloys with different Mg content (0.04, 0.21 and 0.31 wt.% Mg) as well as the corresponding plastic deformation mechanisms and fracture mechanisms.

- Chapter 5 describes the creep properties of the three bulk ZnAl4Cu1 alloys as well as the local creep behaviour of individual microstructural constituents in alloy ZnAl4Cu1Mg0.31. A model bridging the macroscopic and microscopic creep

1 1. Introduction

properties is introduced.

- Chapter 6 focuses on the local mechanical response and corresponding deformation mechanisms of individual microstructural constituents as well as their specific role to the macroscopic deformation in alloy ZnAl4Cu1Mg0.31.

- Chapter 7 illustrates the local precipitation and decomposition behaviour in alloy ZnAl4Cu1Mg0.31 and discusses the influence of these precipitates on the mechanical properties of Zn based alloys.

Therefore, this dissertation bridges the gap between the mechanical properties as well as the plastic deformation mechanisms in ZnAl4Cu1 alloys from both, the macroscopic and microscale point of view.

2 1. Introduction

Figure 1.1. Graphical abstract1.

1 Part of this dissertation appeared as articles. The original citations are: Z. Wu, S. Sandlöbes, L. Wu, W. P. Hu, G. Gottstein and S. Korte-Kerzel (2016). "Mechanical behaviour of Zn-Al-Cu-Mg alloys: Deformation mechanisms of as-cast microstructures." Materials Science and Engineering: A 651: 675- 687., S. Sandlöbes, Z. Wu, K. Pradeep and S. Korte-Kerzel (2016). "Precipitation and decomposition phenomena in a Zn-Al-Cu-Mg alloy." Materials Letters 175: 27-31., Z. Wu, S. Sandlöbes, Y. Wang, J. S. K.-L. Gibson and S. Korte-Kerzel (2018). "Creep behaviour of eutectic Zn-Al-Cu-Mg alloys." Materials Science and Engineering: A 724: 80-94., Z. Wu, S. Sandlöbes, J. Rao, J. S. K.-L. Gibson, B. Berkels and S. Korte-Kerzel (2018). "Local mechanical properties and plasticity mechanisms in a Zn-Al eutectic alloy." Materials & Design 157: 337-350., and Z. Wu, S. Sandlöbes, J. Rao, J. S. K.-L. Gibson, B. Berkels and S. Korte-Kerzel (2018). "Data on measurement of the strain partitioning in a multiphase Zn-Al eutectic alloy." Data in Brief: https://doi.org/10.1016/j.dib.2018.09.010.

This page intentionally left blank.

2. Theoretical background

2.1. Zinc and Zinc alloy

Pure zinc is a bluish-white metal which crystallizes in a hexagonal crystal structure [22]. Table 2.1 lists the detailed physical properties of Zn at room temperature. Pure Zn is relatively soft and brittle at room temperature. To improve its mechanical properties, several methods have been applied, such as addition of alloying elements [12, 20, 23-28], severe plastic deformation [11, 19, 29, 30], proper heat treatment [12, 20, 31] and non-metallic compound reinforcement [32].

Table 2.1. Important physical properties of zinc at room temperature [1, 33].

Crystallographic structure Hexagonal

Lattice parameter a (nm) 0.2665

Lattice parameter c (nm) 0.4947

c/a ratio 1.856

Density (g/cm3) 7.134

Melting point (ºC) 419.53

Shear modulus G (GPa) 43.4

Young’s modulus E (GPa) 108.4

2.1.1. Alloying elements in Zn

Alloying with Al, Cu and Mg [12, 20, 23-28] has been proven to be effective to improve the mechanical properties of Zn. Al is a major alloying element to increase the fluidity, strength, wear resistance, corrosion resistance and decrease the grain size of zinc alloys [4, 11, 34- 37]. For example, El-khair et al. [35] have investigated the mechanical properties of Zn-xAl (x=8,12 and 27 wt.%) alloys and reported an increase in strength and a decrease in ductility with increasing Al concentration at room temperature to 100 ºC. Similarly, Yan et al. [37] have observed a hardness increase in Zn-Al alloys with increasing Al content from 27 to 48 wt.%. Türk et al. [36] have also reported the same strengthening effect of Al in Zn alloys with Al content from 5 to 11 wt.%.

According to the Zn-Al phase diagram, Figure 2.1, Zn rich Zn-Al alloys solidify through a eutectic reaction at 382°C upon cooling, forming a dual-phase microstructure composed of Zn-rich hexagonal η phase and Zn-rich FCC β phase. The Zn-rich β phase decomposes through a eutectoid transformation at 275°C, generating η + α eutectoid structures, where α

4 2. Theoretical background is an Al-rich FCC phase.

Figure 2.1. Zn-Al equilibrium phase diagram [25, 38].

It has been reported that rapid cooling of Zn-Al alloys results in super-saturation of Al atoms in η-Zn which causes various phase transformation and decomposition reactions including a number of metastable phases upon aging [1, 21, 39], which has been reported to proceed for months or even years during natural aging [1, 21, 39] and considered to cause a loss of strength and also dimensional changes [40]. Specifically, Zhu and Opitz [41, 42] observed strong volume changes occurring during the decomposition of supersaturated Zn-rich FCC β’ phase into eutectoid Al-rich FCC α and Zn-rich hexagonal η’, while they observed no significant changes in the mechanical properties [1]. The metastable Zn-rich hexagonal η’ phase has been reported to continuously decompose into the thermodynamically stable η and α phases accompanied by a volume contraction of about 0.06% causing a decrease in tensile strength and an increase in elongation [1, 43, 44]. LeHuy et al., Zhu et al. and Mykura et al. [43, 45, 46] observed volume changes during decomposition of supersaturated α' solid solution following α’ → α + η. However, while LeHuy et al. [45] associated this transformation with a volume contraction, Mykura et al. [46] reported a volume expansion upon this decomposition reaction.

Cu is another main alloying element in Zn alloys that has been shown to increase the solubility of Al in Zn and improve the creep resistance, corrosion behaviour, mechanical as well as the tribological properties of Zn-Al based alloys [7, 20, 26, 47, 48]. However, Cu contents higher than 2 wt.% result in the formation of a copper rich metastable ε (CuZn4) phase in the grain boundaries and interdendritic regions, which has been reported to deteriorate the strengthening effect and corrosion resistance of Zn-Al alloys [20, 25, 26]. The addition of ≥ 2 wt.% Cu has also been assumed to be responsible for the mechanical and dimensional instability of Zn alloys due to the precipitation of metastable intermetallic ε (CuZn4) and Al-rich

5 2. Theoretical background ternary T’ phases [6, 13, 31]. LeHuy et al. and da Costa et al. [25, 45] reported that the formation of these Cu-rich intermetallic compounds leads to a decrease of the Cu content in primary η grains, causing a decrease in strength due to reduced solid solution strengthening. Further, it has been reported that decomposition of the metastable ε phase following α + ε →T’ + η causes a volume expansion of up to 4.5% during prolonged aging of Zn-AI-Cu alloys [8, 21, 25, 43, 44, 49].

Mg is an important alloying element in zinc alloys primarily to counteract the intergranular corrosion effects of impurities such as Pb, Sn, Cd, In and Tl [1]. Mg is also effective to improve the hardness and strength of Zn-Al based alloys [23, 25, 50, 51], which has been attributed to a grain refinement effect [23, 52] and formation of Mg rich intermetallic phases such as

MgZn2 and Mg2Zn11 at grain boundaries [51]. In addition, the presence of Mg has been reported to inhibit the η transformation in the Zn-Al alloys [1] and suppress the formation of metastable ε phase precipitates in Zn-Al-Cu ternary alloys [25]. However, Mg contents higher than 0.1% has been reported to reduce the ductility and impact strength in Zn-Al alloys [1, 50]. High Mg contents also lead to hot shortness due to the formation of a ternary Zn-Al-Mg eutectic structure [1].

The presence of other alloying elements such as Zr, Sn and Ti also has beneficial effects on the Zn alloys. For instance, the addition of 0.1 wt.% Zr has been reported to improve the strength of ZnAl4 alloys due to grain refinement [28]. The addition of Sn has been reported to be beneficial to the strength and creep resistance of Zn alloys [53]. It has been reported that the addition of Ti improves the creep resistant of Zn alloys especially in conjunction with Cu [54].The strength of Zn-Al alloys has also been reported to increase by the addition of small amounts of Li [27] and Na [24].

2.1.2. Application of Zn and Zn based alloys

Every year around 13 million tons of Zn are consumed worldwide [55]. End uses of zinc products mainly include galvanizing to protect steel from corrosion, zinc based alloys, brass and bronze (Cu-Zn alloy and Cu-Sn alloy with a small amount of Zn), rolled zinc in architectural applications and production of coins, chemicals such as zinc as an activator oxide for the tire manufacturing industry and zinc sulphate as a micronutrient additive in animal feed and fertilizers [55, 56].

Specifically, the earliest application of zinc dates back to the Romans with the appearance of brass. Depending on the Zn contents, different kinds of brasses can be produced, which together with bronze makes up to 17% of the zinc application. Higher Zn content leads to a lower melting point of brasses [57]. The most commonly used brass is called Muntz metal,

6 2. Theoretical background which contains 40% Zn [57]. Besides, around 50% of zinc finds its application as coating and galvanizing materials due to its unique corrosion resistance properties [55]. Due to the formation of a unique patina, an inert layer of Zn at the surface in ambient atmospheres, further reaction could be avoided [1]. Moreover, the stronger electronegativity of Zn than Fe makes Zn coatings an effective method to protect iron and steel products from corrosion (cathodic protection [1]). Around 7% of Zn products is used as rolled zinc alloys in the forms of sheet, wire, plate and rod [55]. These products have mainly been used in dry cell battery manufacturing process and as roofing, wall cladding and downpipes in the building industry [58].

Due to the combination of several impressive mechanical and physical properties such as unique corrosion resistance properties and excellent castability [4, 10-12, 16], Zn also finds its application in Zn casting alloys, which are widely used for various engineering applications, such as production of both structural and decorative parts for architectural, automotive, electrical and electronic industries as well as general purpose machinery and equipment which need a high manufacturing precision [9, 12-16]. The zinc casting alloys mainly refer to two groups of Zn-Al alloys, namely ZAMAK and ZA alloys [1]. The ZAMAK alloys have been widely used as pressure die casting materials since the 1920s and contain around 4 wt.% Al and different contents of Cu and Mg in order to achieve optimum mechanical properties and castability [1, 12, 14]. ZA alloys are a new group of Zn-Al alloys with high Al contents which have been developed in the 1970s as high-performance gravity casting materials [1, 14]. However, wider application of both group of Zn alloys is restricted due to limited ductility, poor impact toughness and low strength at low temperature [5, 11, 19, 20, 28], low creep resistance at moderately elevated temperatures [17, 18] and long-term mechanical and geometrical instability at ambient or slightly elevated temperatures [11, 19-21]. In order to improve the creep resistance and long-term mechanical stability of Zn based alloys, a deeper and more quantitative knowledge of the creep mechanisms, deformation mechanisms and phase stabilities in Zn based alloys are essential.

2.2. Deformation mechanisms

2.2.1. Dislocation slip

As one-dimensional lattice defects within the crystal structure, dislocations are irregularities or perturbations of the perfect crystal along a line [59]. The spatial structure of dislocations can be visualized by imagining that an extra half-plane is inserted into a crystal, Figure 2.2 [59]. In the vicinity of the dislocations where the half plane ends, the crystal is locally distorted [60].

7 2. Theoretical background

Figure 2.2. Atomic configuration of (a) an edge dislocation and (b) a screw dislocation [59].

The movement of dislocations causes plastic deformation in crystalline solids [59, 61, 62]. Dislocations can move in two approaches, namely glide on the plane along which the dislocation line is displaced, i.e. slip plane [59], or leave the slip plane by absorption or emission of point defects, i.e. climb [59]. Dislocation slip results in glide steps or slip bands on the surfaces of crystals [63], it doesn’t change the volume of the crystalline, while dislocation climb results in a (very small) volume change of the crystal due to generation or emission of point defects [59, 61].

Due to the existence of dislocations, only small displacements of the adjacent atoms along dislocations rather than gliding of the whole slip plane is required during plastic deformation, Figure 2.3 [62]. In this case, the shape of the crystal is changed through the glide of dislocations, while the crystal structure remains unaffected according to Masing [64].

Figure 2.3. Movement of an edge dislocation: the arrows indicate the applied shear stress [62].

During movement, dislocations are also subjected to periodical lattice resistance, arising from the changes in energy from the atomic rearrangement and misfit of the atoms across the slip plane as dislocations move [61, 65]. The shear stress needed to overcome the periodical lattice resistance and move a single dislocation within a plane of atoms in a perfect lattice,

8 2. Theoretical background namely Peierls stress, or Peierls–Nabarro (P-N) stress, can be roughly estimated as [59, 61, 66-68]:

2퐺 2휋 푑 휏 = ∙ 푒푥푝⁡(− ∙ ) (2.1) 푃 1 − 휈 1 − 휈 푏 where 휏푃 is the Peierls stress, G the shear modulus, 휈 the Poisson ratio, d the lattice spacing and b the Burgers vector. The lowest shear resistance is therefore expected on slip systems where the ratio d/b is greatest, which generally corresponds to the most densely packed plane and closest-packed direction within the slip plane [59]. Figure 2.4 [69-71] and Table 2.2 [72- 74] list all possible slip systems in zinc. In Zn with an axis ratio c/a>1.63, the most densely packed planes and directions are the (0001) basal plane and the 〈12̅10〉 directions [59]. The basal slip system in Zn has the largest d/b ratio and therefore lowest Peierls stress 휏푃, and is conseuently easiest to be activated at room temperature.