scieee Science in your language
[en] (orig)
Study of the microstructure and texture evolution during extrusion
and their effect on the mechanical properties of Mg-Zn based alloys
modified with Ca or Nd
vorgelegt von
Guadalupe Cano-Castillo, MSc
von der Fakultät III Prozesswissenschaften
der Technischen Universität Berlin
zur Erlangung des akademischen Grades
Doktor der Ingenieurwissenschaften
- Dr.-Ing.
genehmigte Dissertation
Promotionsauschuss:
Vorsitzender: PD Dr. Sören Müller
Gutachterin: Prof. Dr.-Ing. Claudia Fleck
Gutachter: Prof. Dr.-Ing. Karl Ulrich Kainer
Tag der wissenschaftlichen Aussprache: 5. Januar 2022
Berlin, 2022
I
Abstract
Magnesium based alloys are attractive materials to be used in the fabrication of structural and
non-structural components of automobiles and casings for electronics. They are also becoming
increasingly attractive for the aerospace industry. This is due to their combined low density
with excellent specific properties. However, wrought magnesium alloys must be improved due
to their often-strong crystallographic textures during processing which can result in low
mechanical properties. This is mainly caused by the hexagonal closed packed crystal (HCP)
structure and the lack of sufficient independent slip systems. The improvement of the
mechanical properties can be achieved by the activation of different deformation modes, for
this reason, special attention must be given to the understanding of their activity.
The present work has been carried out to study the extrusion processing of the Mg-Zn, Mg-Zn-
Ca and Mg-Zn-Nd alloys. A focus was given to the influence of the Neodymium (Nd) and
Calcium (Ca) additions on the microstructural evolution, texture development and
recrystallization behavior of the Mg-Zn based alloy. In that regard, the crystallographic texture
formed in the wrought Mg alloys depends on the active deformation mechanisms during the
thermomechanical processing but also during recrystallization. Mostly, two different
recrystallization processes influence the microstructure and crystallographic texture, i.e.
dynamic (DRX) and static recrystallization (SRX). DRX process takes place during the
thermomechanical processing while SRX after processing. Both recrystallization processes
were investigated in this work. Furthermore, a detailed analysis on deformation mechanisms
has been carried out by applying a crystal plasticity model, i.e. the viscoplastic self-consistent
(VPSC) model and by following the slip trace methodology combined with electron backscatter
diffraction (EBSD) measurements in deformed tensile/compression samples.
During processing, the binary Mg-Zn alloy showed a homogenous, coarse and completely
recrystallized microstructure, whereas the ternary Ca containing alloy also exhibited a
completely recrystallized but finer microstructure. Contrarily, the ternary Nd containing alloy
displayed a partly recrystallized microstructure at some extrusion conditions. At the completion
of the recrystallization process, a fine microstructure was formed which was finer than that
observed in the Ca containing alloy. This result indicates that the Ca and Nd additions not only
induce a delay in the DRX process but also in grain coarsening. In addition, the Mg-Zn alloy
exhibited in general a basal crystallographic texture <1120 >-<1010 >, whereas in the Ca
containing alloy the main texture component is the <1011 > pole. Finally, in the Nd containing
alloy, the <1121 > pole dominates the texture. In contrast, if partly recrystallized
microstructures are annealed, it was found that this distinctive behavior could be changed. The
non-basal texture observed after the DRX process evolved to a basal texture after the SRX
process. This leads to an increase of the importance of SRX and a different resulting texture
while maintaining a similar grain structure.
Uni-axial tensile and compression tests of the extruded alloys showed that the addition of Nd
and Ca results in a significant increase ductility at room temperature. Additionally, there is a
remarkable reduction of the tension-compression yield asymmetry. These properties are due to
non-basal crystallographic textures and the activity of non-basal deformation modes.
The VPSC simulations demonstrated that basal <a> slip and prismatic <a> dominate the
tension deformation of the Mg-Zn alloy, whereas extension twining and basal <a> slip control
the deformation in compression. In Mg-Zn-Ca and Mg-Zn-Nd alloys, additionally to basal <a>
and prismatic <a> modes, evidence of pyramidal <a> was observed in tension. Although basal
<a> slip and extension twinning accommodated the deformation in compression, pyramidal
<a>, pyramidal I <c+a> and pyramidal II <c+a> played an important role during the
deformation. The deformation modes activity observed during the slip trace analysis is
comparable with the obtained during the VPSC simulations.
II
Kurzfassung
Magnesiumlegierungen sind attraktive Werkstoffe für die Herstellung von nicht-strukturellen
Automobilbauteilen und Gehäusen für Elektronik. Auch für die Luft- und Raumfahrtindustrie
werden sie immer interessanter. Dies liegt an der Kombination aus geringer Dichte und
hervorragenden spezifischen Eigenschaften. Knetlegierungen aus Magnesium müssen jedoch
verbessert werden, da sie bei der Herstellung oft starke kristallografische Texturen aufweisen,
die zu schlechten mechanischen Eigenschaften führen können. Dies ist vor allem auf die
hexagonale Kristallstruktur (HCP) und das Fehlen von ausreichend unabhängigen
Gleitsystemen zurückzuführen. Die Verbesserung der mechanischen Eigenschaften kann durch
die Aktivierung verschiedener Verformungsmechanismen erreicht werden, deshalb muss dem
Verständnis ihrer Aktivität besondere Aufmerksamkeit geschenkt werden.
Die vorliegende Arbeit wurde durchgeführt, um die Strangpressbarkeit der Legierungen Mg-
Zn, Mg-Zn-Ca und Mg-Zn-Nd zu untersuchen. Ein Schwerpunkt wurde auf den Einfluss von
Neodym- (Nd) und Calcium- (Ca) auf die mikrostrukturelle Entwicklung, die
Texturentwicklung und dass Rekristallisationsverhalten der Mg-Zn-Basislegierung gesetzt.
Dabei hängt die kristallographische Textur, die sich in den Mg-Knetlegierungen ausbildet, von
den aktiven Verformungsmechanismen während der thermomechanischen Verarbeitung, aber
auch während der Rekristallisation ab. Meistens beeinflussen zwei verschiedene
Rekristallisationsprozesse das Mikrostruktur und die kristallographische Textur, nämlich die
dynamische (DRX) und die statische Rekristallisation (SRX). Der DRX-Prozess findet
während der thermomechanischen Verarbeitung statt, während SRX nach dem Prozess erfolgt.
Beide Rekristallisationsprozesse wurden in dieser Arbeit untersucht. Darüber hinaus wurde
eine detaillierte Analyse der Verformungsmechanismen durch Anwendung eines
Kristallplastizitätsmodells, d. h. des viscoplastic self-consistent (VPSC) model, und durch die
Anwendung der slip trace Methode in Kombination mit electron backscatter diffraction
(EBSD) Messungen an verformten Zug-/Druckproben durchgeführt.
Während der Verarbeitung zeigte die Mg-Zn-Legierung ein homogenes, grobes und
vollständig rekristallisiertes Mikrostruktur, während die Ca-haltige Legierung ebenfalls ein
vollständig rekristallisiertes, aber feineres Mikrostruktur aufwies. Im Gegensatz dazu zeigte
die Nd-haltige Legierung bei einigen Strangpressbedingungen ein teilweise rekristallisierte
Mikrostruktur. Nach Abschluss des Rekristallisationsprozesses bildete sich ein feinkörniges
Gege, das feiner war als dass in der Ca-haltigen Legierung beobachtete. Dieses Ergebnis
zeigt, dass die Ca- und Nd-Zusätze nicht nur eine Verzögerung des DRX-Prozesses, sondern
auch der Kornvergröberung bewirken. Darüber hinaus wies die Mg-Zn-Legierung im
Allgemeinen eine basale kristallographische Textur <1120 >-<1010 > auf, während in der
Ca-haltigen Legierung die Haupttexturkomponente der <1011 > Pole ist. In der Nd-haltigen
Legierung dominiert der <1121 > Pole die Textur. Werden dagegen teilweise rekristallisierte
Mikrostruktur geglüht, so konnte dieses ausgeprägte Texturverhalten verändert werden. Die
nach dem DRX-Prozess beobachtete nicht-basale Textur entwickelte sich nach dem SRX-
Prozess zu einer basalen Textur. Dies führt zu einer Erhöhung der Bedeutung von SRX und
einer anderen resultierenden Textur bei gleichbleibender Kornstruktur.
Einachsige Zug- und Druckversuche an den stranggepressten Legierungen zeigten, dass die
Zugabe von Nd und Ca zu einer signifikanten Erhöhung der Duktilität bei Raumtemperatur
führt. Zusätzlich kommt es zu einer bemerkenswerten Verringerung der Zug-Druck-
Asymmetrie. Diese Eigenschaften sind auf nicht-basale kristallographische Texturen und die
Aktivität nicht-basaler Verformungsmechanismen zurückzuführen.
III
Die VPSC-Simulationen zeigten, dass basales <a> Gleiten und prismatisches <a> die
Verformung im Zugversuch der Mg-Zn-Legierung dominieren, während Dehnungszwillinge
und basales <a> Gleiten die Verformung im Druckversuch kontrollieren. In Mg-Zn-Ca- und
Mg-Zn-Nd-Legierungen wurden zusätzlich zu den basalen <a>- und prismatischen <a>-Modi
Anzeichen von pyramidalem <a> in der Spannung beobachtet. Obwohl basales <a> Gleiten
und Zugzwillinge die Verformung in der Kompression aufnahmen, spielten pyramidales <a>,
pyramidales I <c+a> und pyramidales II <c+a> eine wichtige Rolle während der Verformung.
Die Aktivität der Verformungsmoden, die während der Analyse der Gleitspuren beobachtet
wurde, ist vergleichbar mit den Ergebnissen der VPSC-Simulationen.
IV
Acknowledgements
The present thesis was written during my work as a PhD student at the Helmholtz-Zentrum
Hereon. I would like to take this opportunity to thank all those who supported me and
contributed to the success of this work.
In first place, I would like to thank to my academic supervisor Prof. Dr.-Ing. Karl Ulrich Kainer
and to my group leader Dr. Dietmar Letzig, who have given me the opportunity to do this work
and have always encouraged and supported me in its completion. I am also grateful to Prof.
Dr.-Ing. Claudia Fleck for giving me the opportunity to carry out my doctoral research under
her supervision at the Technical University of Berlin.
My deep gratitude to Dr. Jose Victoria-Hernandez who supported, advised, and motivated me
tirelessly in all situations and had the time and pleasure in correcting this work. I would like to
thank to Dr. Jan Bohlen for the interesting discussions as well as suggestions and motivating
guidance since the start of this work. I would like to thank Dr. Gerrit Kurz and Dr. Sangbong
Yi for their help with lab facilities as well as experiments.
I value the colleagues help with whom I have been working during my stay at Helmholtz-
Zentrum Hereon: Dr. Sumi Jo, Dr. Sangkyu Woo, Dr. Rosario Silva, Mr. Xun Zeng, Mrs. Maria
Nienaber and Mr. Changwan Ha. Thank all of you for such a nice and friendly atmosphere
during working.
I would like to thank to Mrs. Petra Fischer, Mrs. Yukyung Shin and Mr. Gerd Wiese for their
support in the metallographic preparation and microscope observation. I want to thank to Mr.
nter Meister for assisting me in the foundry, Mr. Alexander Reichert and Mr. Stefan Koch
for helping me in the sample preparation for mechanical testing.
I want to thank the financial support from the Mexican Council of Science and Technology
(CONACyT) and the German Academic Exchange Service (DAAD) who providing me the
opportunity to do this project.
I am deeply and forever indebted to my family, my mother Felipa Castillo, my brothers and
sisters, thank you very much for your love throughout my life. I want to thank to my friend
Maria Teresa Rojas Sánchez for encouraging me to continue with my professional preparation.
Thanks to my friends, Alan Quispe and Melina Ortega for making me feel like part of their
family. I also want to thank to Daniel Tamara, Sara Huamanzana, Martha Vinueza, Alex
Almeida, Mario Gomez, Zully Flores, Ana Bayas, Edelina Gentile, Kathia Gerdts and Urban
Röpke for their friendship and invaluable advices.
V
Table of Contents
1. Introduction ............................................................................................................................ 1
1.1 Objectives ........................................................................................................................ 2
1.1.1 Specific objectives .................................................................................................... 2
2. Background ............................................................................................................................ 3
2.1 Magnesium ....................................................................................................................... 3
2.2 Influence of alloying additions on Magnesium ............................................................... 4
2.2.1 Aluminium ................................................................................................................ 4
2.2.2 Zinc ........................................................................................................................... 4
2.2.3 Manganese ................................................................................................................ 4
2.2.4 Calcium ..................................................................................................................... 5
2.2.5 Neodymium............................................................................................................... 5
2.2.6 Cerium....................................................................................................................... 5
2.2.7 Zirconium .................................................................................................................. 5
2.3 Deformation mechanisms in magnesium ......................................................................... 6
2.3.1 Slip systems .............................................................................................................. 6
2.3.2 Deformation by twinning .......................................................................................... 9
2.4 Extrusion process ........................................................................................................... 11
2.5 Extrusion of magnesium alloys ...................................................................................... 11
2.6 Recrystallization in magnesium alloys .......................................................................... 13
2.7 Texture of magnesium alloys ......................................................................................... 15
2.8 Mechanical properties on magnesium alloys ................................................................. 17
2.9 The Visco-Plastic Self-Consistent model ...................................................................... 19
2.10 Summary and motivation of the work ......................................................................... 22
3. Experimental procedure ....................................................................................................... 23
3.1 Studied and cast alloys ................................................................................................... 23
3.2 Heat treatments before extrusion ................................................................................... 23
3.3 Extrusion process ........................................................................................................... 24
3.4 Indirect extrusion ........................................................................................................... 24
3.5 Post-extrusion heat treatments ....................................................................................... 24
3.6 Microstructure characterization ..................................................................................... 25
3.7 Mechanical characterization .......................................................................................... 25
3.8 Slip trace analysis .......................................................................................................... 26
3.9 Crystal plasticity simulations applying the VPSC model .............................................. 27
4. Results .................................................................................................................................. 28
4.1 Extrusion processing ...................................................................................................... 28
4.2 Microstructure evolution ................................................................................................ 32
4.2.1 Z1 alloy ................................................................................................................... 32
VI
4.2.2 ZX10 alloy .............................................................................................................. 33
4.2.3 ZNd10 alloy ............................................................................................................ 34
4.3 Crystallographic texture evolution ................................................................................. 36
4.3.1 Z1 alloy ................................................................................................................... 36
4.3.2 ZX10 alloy .............................................................................................................. 37
4.3.3 ZNd10 alloy ............................................................................................................ 38
4.4 Mechanical properties .................................................................................................... 39
4.4.1 Z1 alloy ................................................................................................................... 39
4.4.2 ZX10 alloy .............................................................................................................. 42
4.4.3 ZNd10 alloy ............................................................................................................ 45
4.5 Yield asymmetry on the extruded bars .......................................................................... 48
5. Discussion ............................................................................................................................ 50
5.1 The importance of the alloy composition on the extrusion processing ......................... 50
5.2 Effect of extrusion conditions on the microstructure and texture evolution. ................. 52
5.3 Recrystallization on the extruded alloys ........................................................................ 54
5.3.1 Effect of dynamic recrystallization on the texture .................................................. 56
5.3.2 Effect of static recrystallization on the texture ....................................................... 58
5.4 Correlation between microstructure and mechanical properties .................................... 59
5.5 Plastic deformation of the extruded alloys ..................................................................... 62
5.5.1 Microstructure evolution during tension and compression ..................................... 63
5.5.2 Experimental and VPSC modelling ........................................................................ 69
5.5.3 Effect of Ca and Nd additions on the CRSS ratio of deformation modes .............. 73
5.5.4 Crystallographic texture evolution: Experimental and simulated ........................... 74
5.5.5 Ex-situ deformation ................................................................................................ 80
6. Conclusions and summary ................................................................................................... 85
7. Directions for future work ................................................................................................... 88
8. References ............................................................................................................................ 89
Appendix .................................................................................................................................. 98
1
1. Introduction
Nowadays, the use of light materials for diverse applications has become inevitable in the
world. The growing demand for improved fuel economy has created a huge attention in
lightweight materials to create new structures for the transportation industry [1]. This can be
achieved through the development of innovative strategies directed to saving weight by using
lightweight materials e.g. magnesium. Magnesium and its alloys have the potential to be used
for diverse components in automotive and aerospace industries owing to its low density (1.74
g/cm), high specific strength and stiffness. Furthermore, the use of magnesium for
manufacturing diverse devices applicable to biomedical industry is also possible; this is due to
the excellent biocompatibility with the human body.
Differently from the magnesium castings alloys, which are more commonly used, the wrought
magnesium alloys still are not often used for industrial applications [2]. To take advantage of
the full weight saving capacity of parts made of magnesium, wrought alloys are required,
because they usually exhibit better mechanical properties such as higher strength and ductility
in comparison with cast alloys. Conventional thermomechanical processes such as extrusion
and rolling as well as non-conventional processes like equal channel angular extrusion (ECAE)
and friction stir processing (FSP) are quite important to alter the grain size, crystallographic
texture and the distribution of secondary phases [3]. Despite such microstructure modifications,
issues such as low strength and high anisotropy still limit the extensive use of the wrought Mg
alloys. This is due to the strong crystallographic textures formed during processing where the
basal planes of the hexagonal close packed (HCP) structure of magnesium align parallel to the
deformation direction. The use of different alloying elements combined with
thermomechanical treatments have been applied to overcome the limitations mentioned
previously.
It is well accepted that the mechanical properties of Mg alloys are impacted by the
microstructural evolution and the crystallographic texture development [4]. The
crystallographic texture promotes or hinders the activation of individual deformation modes
due to the changing of critical resolved shear stress (CRSS) or Schmid factor (SF). To alter the
grain size and tailor the crystallographic texture, the extrusion process is one of the most
commonly applied thermomechanical processes. Processing conditions such as extrusion ratio,
temperature and speed can influence the recrystallization behavior and the development of
microstructure and therefore have a strong effect, first on the crystallographic texture and
finally, modifying the mechanical response of the extruded materials.
Consequently, a profound understanding of the relationship between the processing parameters
and the resulting microstructure as well as between the microstructural development and the
resultant mechanical properties is required. Therefore, the present work analyzes the effect of
the extrusion temperature and speed and of post-extrusion heat treatments of Mg-Zn based
alloys modified with single additions of Ca and Nd. This enables the analysis and the
comparison of the effects of the Ca and Nd on the microstructure and crystallographic texture
development of the extruded Mg-Zn based alloys.
As a processed magnesium product, the microstructure and crystallographic texture developed
during the processing has a great impact on several features of the mechanical properties such
as yield stress on tension or compression (TYS or CYS), ultimate stress on tension or
compression (UTS or UCS) and fracture strain. From a general point of view, the
crystallographic texture and microstructure affect the plastic flow of the extruded products. It
is well known that the plastic behavior is susceptible to different deformation mechanisms.
2
Even though the relationship between plastic behavior and different deformation mechanisms
is known, the general relationship remains largely unclear. Hence, this research work seeks to
determine the effect of recrystallization evolution during extrusion and its impact on the
microstructure and crystallographic texture development of Mg-Zn based alloys with separate
additions of Ca and the rare earth Nd. An emphasis to the development of different
crystallographic textures based on the dynamic recrystallization (DRX) and grain growth due
to the modification of extrusion conditions is given. Furthermore, the effect of static
recrystallization (SRX) on the grain growth to modify the crystallographic texture by using
partially dynamic recrystallized microstructures is also investigated.
Finally, the mechanical properties of the extruded materials are correlated with the resultant
microstructure and crystallographic texture during the extrusion processing. A detailed analysis
of the possible deformation mechanisms dominating the plastic behavior of the extruded
magnesium alloys due to the different microstructure is carried out.
1.1 Objectives
The present work aims to provide a comprehensive analysis of the relationship between single
additions of Ca and Nd and extrusion parameters on the microstructure evolution of extruded
Mg-Zn based alloys. This work also aims to elucidate the role of the initial microstructure and
texture on the activity of possible deformation mechanisms and the correlated mechanical
properties of the extruded alloys.
1.1.1 Specific objectives
Assessing the extrusion process parameters for each extruded alloy.
o The extrusion process is analyzed in terms of the alloy composition as well as
the final surface quality of the extruded bars. Also analyzed is the effect of Ca
or Nd on the microstructure and crystallographic texture evolution of Mg-Zn
based alloys. Adding Ca or Nd combined with the extrusion parameters will
alter differently the microstructure and crystallographic texture of the Mg-Zn
based alloy.
Studying the microstructure and crystallographic texture evolution of each alloy due to
the recrystallization process.
o The crystallographic texture evolution is analyzed in terms of different fractions
of deformed and dynamically recrystallized grains. Furthermore, the importance
of static recrystallization (SRX) on the crystallographic texture modification is
also studied. The analysis of grain growth with specific orientations, following
DRX or SRX nucleation, is essential for understanding the recrystallization
texture evolution.
Correlating the resultant microstructure and crystallographic texture with the
mechanical properties of the extruded alloys.
o The importance of the initial microstructure and texture on the mechanical
behavior and the activation of the possible deformation modes during the plastic
behavior of extruded bars is analyzed by coupling experimental work, e.g. slip
trace analysis with crystal plasticity simulations using the viscoplastic self-
consistent (VPSC) model. This to provide a sound understanding of the possible
activation of additional non-basal slip systems according not only to chemical
composition, but to also initial crystallographic texture.
3
2. Background
2.1 Magnesium
Magnesium (Mg) is the lightest engineering metal with a lower density than aluminum and
steel, i.e., 1.74 g/cm3, 2.7 g/cm3 and 7.8 g/cm3, respectively. Useful characteristics of Mg
include its high stiffness, which is higher than for example Al, Fe, and plastics, i.e., 2.0 GPa1/3/
(Mg/m3), 1.5 GPa1/3/ (Mg/m3), 0.75 GPa1/3/ (Mg/m3), 1.1-1.5 GPa1/3/ (Mg/m3), respectively [5,
6]. In addition, its Young’s modulus is about 45 GPa at room temperature, it has a specific heat
of roughly 1.05 kJ/kg°C and a low melting temperature (650°C) [7].
Magnesium as well as its alloys are readily produced following the near-net shape casting
process. Already in 1909, an Mg crankcase was presented at the International Air Transport
Fair in Frankfurt and in 1924, Mg alloys were used in the automobile industry for the first time
to produce pistons by die casting [8].
Even though the Mg alloys have been used in the industry since the early 20th century, the
spread of their applications remain limited, predominantly due to the cost-effective commercial
Mg alloys do not have the adequate properties such as yield strength (YS), ductility, formability
and corrosion resistance [9, 10]. In recent years, the renewed interest in Mg and its alloys has
been motivated due to the requirement of weight-saving in the automotive, aerospace and
portable electronics industries.
In the transport industry, the magnesium alloys are considered as a material important to
achieve lightweight and therefore in the improvement of the energy efficiency and the
greenhouse gas reduction on the world. Notwithstanding the emerging importance of
magnesium alloys as a class of engineering material, up to now the magnesium alloys have had
a minimal usage in the industry, the same can also be said for the extruded alloys.
The extruded magnesium products represent less than 1.5 % of the annual production of
magnesium in 2004 and remained at less than 3% in 2013 [1, 3]. In contrast, roughly 25% of
the Al alloy products are produced by extrusion and have been adopted in different applications
[11]. Three major issues have restricted the wider use of magnesium alloy extrusions.
The first issue corresponds to the properties of the extruded profiles, in that regard, the Mg
extrusions has lower strength than the Al extrusions, limited formability as well as a high
tension-compression yield asymmetry. The next issue is the commercial viability, being some
of the Mg alloys more expensive, which depends on the alloying elements. In addition, only
few magnesium alloys are capable of being extruded at high enough speeds to be viable. The
last issue is related with the poor corrosion resistance of the magnesium extrusion alloys. In
general, the corrosion rate of magnesium alloys is dominantly given by the electrochemical
potential and secondly by the alloy chemistry, it is determined by the specific elements with
which the magnesium is alloyed [12].
Nevertheless, the corrosion topic is beyond the scope of this work. Therefore, it is covered the
effect of the extrusion processing parameters on the subsequent microstructure development as
well as the mechanical performance of magnesium alloys.
Through the last years, improvements on the magnesium alloys have been done. These
improvements have been reported on the extrudability and the mechanical properties of
magnesium alloys. With respect to the extrudability, the most investigated has been principally
the extrusion speed. It has been reported that the maximum extrusion speed of some dilute Mg
alloys can be as high as that of Al extrusion alloys [13]. In terms of the mechanical
performance, the tensile yield strength of the extruded magnesium alloys has shown an
4
increment from less than 300 MPa to a yield strength of over 400 MPa for rare earth element
free alloys, for instance, the Mg-Al-Ca-Mn alloy [14] and roughly 500 MPa for a Mg-Gd-Y-
Zn-Zr alloy reported by N. Kunito et al [15]. The ductility has also been improved, from less
than 20% to bigger than 40% [16].
The improvements are all of them correlated with the microstructure development due to the
combination of processing and alloy design. To date, different alloying elements have been
used to produce different magnesium alloys. Each alloying element has a specific effect on
magnesium. In the next section is described the effect of some alloying elements used to alloy
it.
2.2 Influence of alloying additions on Magnesium
An effective way to improve the properties of magnesium is by using alloying elements. The
presence of alloying elements improves the properties of magnesium, this is due to the grain
size refining as well as the formation of intermetallic particles. Significant advances have been
achieved in the alloy development using alloying elements such as Aluminum (Al), Zinc (Zn),
Manganese (Mn), Zirconium (Zr), Calcium (Ca) or rare earths (RE). In this section is described
the effect of some alloying elements on magnesium.
2.2.1 Aluminium
Aluminium (Al) has a large solubility in magnesium, which is about 12.9 wt.% at the eutectic
temperature of 437 °C, and it decreases to about 3.6 wt.% at 200 °C [17]. It is the most used
alloying element in magnesium. Most of the commercial extruded alloys are based on the Mg-
Al alloy system. Some of these alloys are AZ31, AZ61 and AZ80. The addition of certain
amount of Al can increase not only strength but also ductility. The low density of Al is an
advantage when it is added to magnesium, such advantage does not compromise the density of
the alloy when Al is added in large additions [12]. The formation of the intermetallic phase
Mg17Al12 contribute to increasing the tensile strength.
2.2.2 Zinc
Zinc (Zn) is one of the most frequently used alloying elements in magnesium. Its maximum
solubility in magnesium is 6.2 wt.% [17]. Combined with aluminum, Zn is often used to
produce improvements in the strength at room temperature [18]. Zn can improve the
mechanical properties by solid solution hardening and age hardening. Nevertheless, with
amounts higher than 1 wt.% leads a precipitation sequence which is associated with hardness
decrease [19, 20]. When Zn is added in quantities bigger than 1 wt.% to magnesium alloys
containing aluminum from 7 to 10 wt. %, the hot-shortness increases [18]. In terms of cast-
ability, Zinc produces a similar effect as Al. Besides, Zinc is an important alloying element that
contributes to the grain refinement [21, 22]. Such a grain refinement can be related with the
increase of strength because of Hall-Petch effect [23, 24].
2.2.3 Manganese
Manganese (Mn) as alloying element helps to increase the strength of magnesium alloys but
also improves their corrosion resistance by reducing the harmful effects of Fe [25]. This
alloying element has a solubility limit of 2.2 wt.% on magnesium at high temperature [26]. At
lower temperatures, the precipitation of α-Mn is expected [27]. It is usually added in
combination with Al. Mn as alloying element helps to refine the microstructure or forming
5
special structures. When Mn is added to magnesium alloys, such alloying element tend to form
weak textures after extrusion [4].
2.2.4 Calcium
Calcium (Ca) as alloying element has a maximum solubility of 1.35 wt.% in Magnesium [17].
It also works as a grain refiner when is added to Mg. Thermally stable intermetallic compounds,
such as the Mg2Ca would precipitate along grain boundaries of α-Magnesium matrix during
solidification [28]. A size misfit in the magnesium lattice structure or a decrease of the stacking
fault energies are some of the features, which change the balance of deformation mechanisms
in alloys modified by Ca [29, 30]. Therefore, Calcium also modifies the texture during
processing. The addition of this alloying element, for example to Mg-Zn based alloys leads to
the development of some characteristic texture components also resulting in a certain tilt of
basal planes out of the extrusion direction [16, 31, 32]. The changes in the grain structure and
texture lead to an increase of the ductile behavior at room temperature.
2.2.5 Neodymium
Recent investigations pointed out the role of rare earths (RE) elements on Magnesium [4, 33,
34]. The RE alloying elements can be divided into two groups with large and limited solubility
in Mg. In the specific case of Neodymium (Nd), this alloying element belongs to the group
with a low solubility in Magnesium, roughly 0.59 wt.% [35]. Different authors have found that
still this element leads to reduce the grain size and weakening the texture even in low quantities
into Magnesium [36-38]. The retardation of the recrystallization process and the formation of
specific texture components after thermomechanical treatments are the effect of the addition of
rare earth elements to magnesium [39]. The particular feature associated with the texture
weakening are some special texture components named as ‘‘RE-textures. A common example
of RE textures are the <11-21> formed during extrusion where the basal planes are tilted out
of the extrusion direction [40]. The formation of that kind of texture contributes with an
enhancement of ductility and the reduction of the asymmetry of mechanical properties at room
temperature [37, 39].
2.2.6 Cerium
Cerium (Ce) belongs to the rare earths allying elements commonly used in magnesium. This
rare earth element has a solubility of 0.74 wt.% in magnesium [26]. It is one of the strongest
potential texture modifiers for the wrought magnesium alloys [41]. Ce weakens the
deformation and annealing texture at concentrations as small as 0.057 wt.%, however this effect
is quickly leveled at concentrations above 0.17 wt.% [39]. Magnesium alloys containing Ce
develop relative softer texture away from the basal texture. It has been suggested in [42] that
the superior efficiency of Ce has to do with its low solubility limit in comparison with other
RE elements.
2.2.7 Zirconium
Zirconium (Zr) is added to magnesium as a potent grain refiner to improve the mechanical
properties. The maximum solubility of Zr in magnesium is 3.8 wt.% [26]. This alloying element
helps to enhance the extrudability of magnesium alloys, furthermore its addition leads to a
much denser distribution of precipitates [12]. The Zr addition also includes the corrosion
resistance improvement of the magnesium alloys [43].
6
From the previously described alloying elements, the relevant alloying elements for this study
are Zn, Ca and the rare earth Nd.
2.3 Deformation mechanisms in magnesium
The crystal structure of magnesium corresponds to HCP, which is schematically shown in
Figure 2.1. In the HCP lattice, the planes and directions are described by using the Miller-
Bravais indices which relates to a coordinate system of three basal vectors (ai) and a
longitudinal axis called c-axis [44]. The hexagonal unit cell has axes a1=a2=a3 c, and angles
α=β=90°, γ=120°. In case of magnesium, the c/a ratio is 1.623, which is quite close to the lattice
parameters (c/a) for the ideal packing arrangement in the HCP unit cell, 1.633.
In contrast to FCC and BCC materials, the HCP structure has limited slip systems. At room
temperature, the deformation of Mg and its alloys is made by slip, which means that the
dislocations glide takes place along different crystallographic planes known as slip planes. The
basal, prismatic and pyramidal are the principal slip systems on magnesium. Besides, twining
is other possible deformation mechanism that can be activated at room temperature.
Figure 2.1. The hexagonal unit cell [45].
2.3.1 Slip systems
A slip plane together with the slip direction defines a slip system. The crystallographic slip
normally occurs due to the sliding of a particular crystallographic plane over a neighbor plane
in a specific crystallographic direction. Slip begins when the shearing stress reaches a threshold
value called as the critical resolved shear stress (CRSS), which for uniaxial loading, the CRSS
is calculated as follow in equation 1 [46] and schematically shown in figure 2.2.
7
Figure 2.2. Schematic representation of the resolved shear stress determination [46].
τR=𝑃𝑃cosλ
𝐴𝐴/cosϕ=𝑃𝑃
𝐴𝐴cos ϕcos λ (Eq. 1)
Where:
A is the cross-sectional area, ϕ corresponds to the angle between the normal to the slip plane
and the tensile axis, λ is the angle between the slip direction and the stress axis, 𝐴𝐴/cos ϕ
corresponds to the area of the slip plane inclined at the angle ϕ and 𝑃𝑃cos λ is the component
of the axial load acting in the slip plane in the slip direction.
Then, the deformation is influenced by the anisotropy that results from the low symmetry of
the HCP structure. According to Von Mises criterion, a material requires five independent slip
systems to deform in an arbitrary way [47]. The low symmetry of the HCP structure of
magnesium strongly influences the deformation. Such a low symmetry restricts the availability
of slip modes, which can be activated simultaneously. In magnesium, the possible slip modes
are schematically illustrated in Figure 2.3. Each slip mode has a number of systems that can
operate during deformation. It is generally accepted that basal slip is the softest mechanism;
and the activation of the other slip modes normally depends on the temperature and the stress
level [48]. For magnesium and its alloys, the principal slip mode is the basal slip [49]. However,
the basal slip provides only three independent slip systems [8], therefore it becomes necessary
the activation of the non-basal slip modes [8, 50]. Nevertheless, the activation of the non-basal
slips is hard, in that case, the high temperature and the high stresses are the responsible for the
activation of such slip modes [51].
In figure 2.4, the effect of temperature on the CRSS of different slip systems of the magnesium
single crystal is illustrated. A transition occurs after 230 °C when the values of CRSS of
prismatic and pyramidal slip decrease considerably.
8
Figure 2.3. Slip systems in magnesium [52].
Figure 2.4. Effect of temperature on the CRSS of slip systems and twinning in pure Mg [53].
9
The use of alloying elements may also affect the balance of slip systems through different ways,
for example by altering the c/a ratio and changing the CRSS of the slip systems, changing the
crystal structure of magnesium, modifying the stacking fault energy (SFE) and changing the
grain size [12, 39].
2.3.2 Deformation by twinning
The other important mechanism by which is possible to deform a metal is the process known
as twinning. The twinning results when a portion of the crystal takes up an orientation that is
related to the orientation of the rest of the untwined lattice in a definite, symmetrical way. The
twined portion of the crystal is a mirror image of the parent crystal. The plane of symmetry
between the two portions is called the twinning plane [46]. Figure 2.5 illustrates a schematic
representation of twinning. Twinning takes place when the slip systems are restricted, then, the
occurrence of twining at low temperatures or high strain rates in hcp metals is due to the
unfavorable orientation for basal slip [54]. The importance of twinning in plastic deformation
does not come from the strain produced by the twinning process but from the fact that
orientation changes resulting from twinning may place new slip systems in a favorable
orientation with respect to the stress axis so that additional slip can take place [46].
Figure 2.5. Schematic presentation of how twinning results from shear stress τ [46].
Contrasting to slip, due to the polar nature of twinning, the shear can take place only in one
direction. According to the amount of twinning shear as a function of the c/a ratio, the twinning
modes may be either tensile or compressive [55]. A twin mode showing a positive slope in
Figure 2.6 results in contraction along the c-axis meanwhile a twin mode showing a negative
slope results in extension along c-axis.
In magnesium alloys, three types of twins have been reported: tensile twins, compressive twins
and double twins. Extension alongside the c-axis favors {10-12} twinning whereas a
compression along this axis favors {10-11} <10-12> twinning [56]. Therefore, the most
common twinning mode in magnesium, i.e. {10-12} twinning is recognized as tension or
extension twin since it provides extension along the c-axis [57, 58]. Thus, the {10-11} twining
is known as compression or contraction twin.
10
Figure 2.6. Variation of twinning shear with the c/a ratio for various twinning modes in HCP
metals [55].
Twinning can alter the orientation of the original grains and is accepted that reorients the basal
planes. The reorientation by 86 ° is associated to the tensile twinning, the reorientation by 56 °
is related to compression twinning while the double twinning is correlated with a reorientation
by 38°. A schematic representation of the tensile and compression twinning is shown in figure
2.7.
Figure 2.7. Schematic representation of twinning, a) tension twinning 1012<101 1 > and
b) compression twinning 1011<101 2 > [59].
11
2.4 Extrusion process
Extrusion is an important processing method for producing different kind of profiles in a single
forming step. In this forming process, a billet, usually round, is pressed by a stem at high
pressure through a tool of the desired shape, the die [11] .
The extrusion processing is usually subdivided according to the billet temperature, i.e., cold or
hot extrusion and the relative movement of the extruded material and the ram. Concerning to
the relative movement of the material and the ram, the two basic types of extrusion are the
direct and indirect extrusion [60]. In figure 2.8 is depicted a schematic representation of both
extrusion process.
In direct extrusion, a ram, usually with a pressure pad in front, pushes the billet in a stationary
container through a tool of the desired shape, the die. Relative movement takes place between
the billet and the container [11].
In contrast, in indirect extrusion, the die is located in front of a hollow ram and pushed against
the billet by the forward movement on the container closed to at the back. There is, therefore,
no relative movement between the billet and the container [11].
The properties of the extruded bars are affected greatly by the way in which the metal flows
during extrusion. The metal flow is influenced by parameters like the extrusion ratio, the billet
temperature, the ram or extrusion speed and the frictional conditions at the die and container
wall [11, 60]. The first three parameters are the most usually varied in order to control the
extrudability of the material and the resulting conditions of the produced bars.
Figure 2.8. Schematic representation of the extrusion process; a) Direct extrusion, and b)
Indirect extrusion [11].
2.5 Extrusion of magnesium alloys
Magnesium alloys are usually extruded in the rage of 250 °C- 450 °C [61]. Diverse factors
affect its extrudability. The property of extrudability refers to the multivariable processing
window of an alloy during extrusion [12]. The press limit (left lines) and the limit of hot
cracking (right lines) determine the extrusion window as is shown in figure 2.9. In comparison
with the aluminum extrusion alloys, magnesium alloys have lower extrudability, i.e. they must
be extruded at lower speeds and within a narrower range of extrusion temperatures [62].
Therefore, this low extrudability of magnesium alloys leads to a lower production efficiency
and higher cost than aluminum extrusion alloys.
Magnesium alloys such as AZ61, ZK60 and alloys with high amount of rare earth elements
have a low extrudability owing to these alloys are hard and the extrusion press machine cannot
12
press them at low extrusion temperature. Thus, they must be extruded at high temperature with
low extrusion speed. Attention should be paid to the control of the extrusion speed; this is
because when the extrusion speed is beyond the right limit of the adequate processing window
some cracks are formed in the surface of the extruded profiles. These cracks formed during
extrusion are known as hot shortness [63]. The hot shortness phenomenon is formed when the
local temperature in the die exceeds the solidus temperature of the alloy or the melting
temperature of the second phases [64].
Figure 2.9. a) Extrusion limits diagrams of some commercial Mg alloys and the Al-alloy
AA6063 [62], and b) surface quality of AZ31 extruded at 375 °C and different extrusion speeds
[63].
The alloy design has been applied to enhance the extrudability. The low alloying concentration
on the magnesium alloys is related to a softer deformation during extrusion and thus a lower
extrusion temperature and higher speeds can be tolerated before the press capacity is reached
[12]. Then, a lower alloy content can result in better extrudability.
In recent years, diverse research have been done with the aim to obtain a high performance on
the extruded magnesium alloys. In that regard, the development on the extruded magnesium
alloys has been focused mostly in the evolution of the microstructure as well as the
crystallographic texture. Of these two alloy characteristics, the investigation of the basal texture
weakening has received much more attention. Such microstructure or crystallographic texture
evolution has been associated to dynamic or static recrystallization. Diverse recrystallization
or deformation mechanisms have been considered to explain the formation of different texture
after processing.
13
It is well accepted that in the evolution of microstructure and crystallographic texture, the
temperature plays an important role to a great extent. During the deformation, cross slip and
climb are the main plastic deformation mechanisms in the extruded magnesium alloys, besides
at high temperatures usually dynamic recrystallization occurs. The combination of both,
recrystallization and deformation mechanisms are implied in the crystallographic texture
formation.
In the selection of the appropriate process parameters, it is important to consider the effect of
the selected alloying elements on the extrudability of magnesium. A high extrudability can be
reached by using alloying elements that can increase the solidus temperature of the alloy, such
as Ca and RE elements.
In case of Zn, this element decreases the solidus temperature of the alloy or induce the
formation of second-phase particles with low melting temperature [12], then its presence in
magnesium should be low. Murai et al [65] found that during direct extrusion, Zn additions
greater than 1 wt. % promoted the occurrence of hot-cracking. Yan and Chen [66] reported that
Ca additions promotes an increase in the solidus temperature, they found that the excellent
extrudability of the dilute Mg-Zn-Ca-Mn alloys can be mainly ascribed to the thermally stable
Mg2Ca phase and high solidus temperatures of 620 °C. In the case of the rare earth alloying
elements, the high content of it produces a low extrudability of magnesium alloys [12]. The RE
elements such as Ce, Gd, Y, La and Nd are commonly added to improve the mechanical
properties of magnesium via microstructure and texture modification.
2.6 Recrystallization in magnesium alloys
The recrystallization process is defined as the replacement of a deformed microstructure by a
new set of strain-free grains, this process involves the nucleation and subsequent growth of the
new grains at the expense of the surrounding deformed grains [67].
There are two types of recrystallization, i.e. dynamic (DRX) and static recrystallization (SRX).
During the extrusion processing the microstructure of magnesium changes significantly, which
includes both, the dynamic and static recrystallization. In figure 2.10 is depicted a
representation of the possible softening process during extrusion.
14
Figure 2.10. Schematic representation of possible softening process during extrusion [68].
Considering the tree most important extrusion parameters, i.e., temperature, speed, and ratio,
three different mechanisms of DRX have been observed to operate in magnesium during
processing [69]. The continuous DRX (CDRX), discontinuous DRX (DDRX) and twin DRX
(TDRX) are included on those three types of DRX.
The continuous DRX (CDRX) consists of the formation of stable three-dimensional arrays of
deformation low angle boundaries (LAGBs) followed by their gradual transformation into high
angle grain boundaries (HAGBs) upon straining [67, 70, 71]. During the CDRX, a rotation and
migration of low-angle grain boundaries take place homogeneously in the microstructure,
therefore this type of recrystallization could be considered as an extended recovery
phenomenon [39].
The discontinuous DRX (DDRX) involves the development of HGABs via the nucleation and
growth of new grains. The nuclei develop on original HGABs due to the operation of a bulging
mechanism [72, 73]. Such recrystallization mechanisms usually occur in materials with
relatively low stacking fault energies [67, 73]. The local migration of the grain boundaries, i.e.
bulging, leads to the formation of nuclei, which then grow out and consume a deformed matrix,
resulting in decreased dislocation density, and providing strain softening. Thus, this mechanism
involves the development of high-angle grain boundaries via the nucleation and growth of new
grains. It is closely related to strain induced migration of initial boundaries [69, 74].
Finally, the DRX mechanism associated with twinning (TDRX); in this case, the twinning leads
to the formation of coarse lamellae surrounded by special grain boundaries [75]. This
mechanism can occur by at least three processes. Such processes are (1) mutual intersection of
primary twins, (2) the occurrence of secondary twinning within the coarse lamella and (3) the
subdivision of the coarse twin lamellae [72, 76].
15
The recrystallization process inside the twins is governed by local recovery, which can be
possible by slip assisted or thermally activated, which gives to the formation of sub-grains
similar to twin orientations [77, 78]. Studies indicate that the nucleation inside the compression
and double twins can give rise to soft orientations, which gives an important role in the
recrystallization texture modification [79, 80]. A higher thermal activation can trigger growth
of such orientations owing to the associated high driving pressures [81].
The deformation during processing leads to a buildup of dislocations and this increases the
amount of strain energy stored in the material [82]. Then, during the subsequent annealing of
the hot deformed material, there is a reduction of this stored energy, which drives the nucleation
and growth of new, strain-free grains, i.e., the static recrystallization (SRX) occurs [67]. During
the static recrystallization (SRX) of Mg alloys, the formation of twins plays an important role.
It has been shown that under certain conditions, the Mg and Mg alloys have a tendency to form
twins during the deformation processing [81, 83-85]. Studies have revealed that the twins and
the twin boundaries serve as dominant nucleation sites during recrystallization of Mg alloys
[81, 83-87].
2.7 Texture of magnesium alloys
Due to of the effects of deformation and recrystallization during the processing, the magnesium
alloys develop a crystallographic texture or crystallographic orientation. The crystallographic
orientation, or in this context simply texture, refers to how the atomic planes in a volume of
crystal are positioned relative to a fixed reference [88].
Due to the anisotropy of magnesium single crystals combined with the uniaxial character of
deformation twinning [89], the crystallographic texture of a polycrystalline material plays an
important role on the impact on the deformation behavior of magnesium wrought products.
The global behavior of the polycrystalline material is somewhat similar to that of a single
crystal if the texture is strongly developed; therefore, the anisotropy of the HCP structure of
magnesium is also carried over to the thermo-mechanically processed polycrystalline material.
During the thermomechanical processing, some specific planes or directions can be strongly
oriented with respect to a specific direction in the material, for example, the extrusion direction
of the extruded bar. Figure 2.11 shows a schematic representation of the alignment of basal
planes during extrusion.
Figure 2.11. Schematic representation of the distribution of magnesium crystals after extrusion
[90].
The X-Ray diffraction technique is normally applied to do texture analysis of the processed
materials. The quantitative determination of textures can be done using a pole-figure
goniometer, which uses a reflection geometry and a monochromatic X-rays Cu radiation.
16
In the goniometer, the source and counter are arranged in a fixed geometry, depending only on
the Bragg angle of the investigated crystallographic plane. The Bragg angle refers to a
reflection condition if Bragg’s law is obeyed [91]:
𝑛𝑛𝜆𝜆 = 2𝑑𝑑sin 𝜃𝜃 (Eq. 2)
Where:
λ is the wavelength, d is the spacing of the reflecting planes, θ is the angle of incidence and
reflection and n is the order of diffraction
The sample is mounted on a holder, which can be rotated around two perpendicular axes to
orient the specimen in any position with respect to the incident X-ray beam. The goniometer
moves the detector with respect to the X-ray beam (rotation 2θ); the sample is positioned
relative to the X-ray beam by two rotations, Φ and χ. Figure 2.12 displays the definition of such
rotations. The χ circle is generally symmetrical between incoming and diffracted beam
(positioned at an angle θ). The 2θ and ω axis coincide. The nomenclature Φ, χ, and θ is standard
in single crystal diffractometry and marked on most instruments. In a pole figure goniometer,
the crystallographic ‘goniometer head’ is replaced by a texture attachment on which the sample
can be mounted and oscillated. Stepper motors, controlled by a personal computer, enable one
to obtain any arbitrary angular position on the three axes 2θ, χ and Φ (within a certain range to
avoid mechanical collisions). The axe ω sets only the detector to the proper Bragg angle, 2θ,
of the diffraction peak of interest [92].
Figure 2.12. Ray path and sample rotation in an X-ray texture goniometer and definition of the
instrument angles [92].
The movement of the specimen unveils the spatial orientation of the respective poles {hkl}. In
a stereographic projection, the measured intensity distribution generates the {hkl} pole figure
(P.F). A pole figure (P.F) is a two-dimensional stereographic projection in which the positions
and intensities of specific crystallographic orientations are plotted in relation to the specimen
geometry. In figure 2.13a is displayed a pole figure that shows the distribution of a selected
crystallographic direction relative to certain directions in the specimen. In the case of
magnesium, typical pole figures of interest are (0001), (1120), (1010), (1011), (1012) and
(1013).
Drawing the conventional pole figures is the most common method to represent textural data
for materials. Nevertheless, the satisfactory description of texture, sometimes, can also be given
17
in terms of an inverse pole figure (IPF). In such inverse pole figure, the distribution of a selected
direction in the sample is depicted in relation to the crystal axes. The representation in the I.P.F
is exactly inverse of the pole figure. In the I.P.F, the sample directions are projected into the
crystal frame as opposed to pole figures, which are essentially the projection of crystallographic
directions in the sample frame of reference. For an I.P.F, the projection plane is a standard
projection of the crystal, of which only the unit stereographic triangle needs be shown. Figure
2.13b shows the stereographic projection of a standard representation of an inverse pole figure
in the HCP lattice structure. The texture component <0001 > is associated to the basal planes
of the HCP structure, while the <1010 > and the <1120 > are related to the prismatic planes of
the first and the second order, respectively.
Figure 2.13. a) Standard (0001) stereographic projection of an HCP crystal structure [91], and
b) inverse pole figure (I.P.F) used in this study for the representation of textures in extruded
bars [91].
2.8 Mechanical properties on magnesium alloys
Nowadays, the main applications of magnesium are as cast products. These cast products can
be complex, but in many cases, they lack the desired mechanical properties. It is well known
that the mechanical properties of magnesium and its alloys are influenced by the microstructure
and texture. In general, the mechanical properties shown by the wrought magnesium alloys are
better compared to the as-cast alloys; this is mainly because of the small grain size, the
crystallographic texture and the absence of porosity [93]. However, still the wrought
magnesium alloys have a low strength, ductility and tend to exhibit a tension-compression yield
asymmetry.
The yield asymmetry refers to the difference between the yield stress (YS) in tension and in
compression; the compressive YS is typically lower than the tensile YS when the material is
18
tested [77]. An example of texture developed in Z1 alloy as well as its mechanical properties
are depicted in figure 2.14. The tension-compression asymmetry is observed (figure 2.14b).
This behavior is normally associated to the texture developed during the thermomechanical
processing. In these textures, the basal planes of the HCP structure of magnesium are aligned
parallel to the main forming direction, [94] e.g., extrusion direction.
Figure 2.14. a) Texture in the form of I.P.F, and b) mechanical properties of extruded
magnesium alloy Z1 [95].
Studies have been reported that the yield asymmetry is related to the twining activity [57, 96,
97]. In the textures where the basal planes are predominately oriented parallel to the extrusion
direction, the twinning activity is more pronounced in compression rather than in tension.
Therefore, if twinning activity could be suppressed, the asymmetry may be reduced. The
reduction of twining activity can be achieved by crystallographic texture modification. Besides,
it has been also reported that the grain refinement leads to a reduction in the yield asymmetry
[98]. This observation is also related to twinning activity since twinning becomes less favorable
if the grain size is reduced. Lately, different studies have reported that the presence of certain
alloying elements on magnesium can lead to the suppression of the development of a strong
fiber texture during thermomechanical processing. This idea is related with the development
of weaker textures and more randomly oriented basal planes, which increases the ductility and
reduces the yield asymmetry, as is shown in figure 2.15.
Figure 2.15. a) Texture in the form of I.P.F, and b) mechanical properties of wrought
magnesium alloy ZNd10 [95].
19
2.9 The Visco-Plastic Self-Consistent model
The use of semi-finished magnesium products is still limited due to its low mechanical
properties, which is normally associated to the lack of adequate activation of slip modes. The
differences in the texture of such semi-finished products developed during processing can have
significant effects on its mechanical properties. For this reason, it is of fundamental interest to
gain a more detailed insight during the deformation of magnesium alloys. In that regard, the
modelling of deformation behavior plays an important role on a better understanding of the
deformation of magnesium alloys. The crystal plasticity simulations is a powerful tool that
have been used for simulating the deformation. Different plasticity models have been suggested
to simulate the deformation. For the understanding of the deformation modes in the extruded
magnesium alloys of this research work, the visco-plastic self-consistent model (VPSC) has
been applied.
The VPSC model is a computer code written mostly in Fortran 77. The VPSC model stands for
Visco Plastic Self Consistent and refers to the particular mechanical regime addressed (VP)
and to the approach used (SC). It allows the simulation of the plastic deformation of
polycrystalline aggregates. Such model it is been developed for application to low-symmetry
materials like hexagonal, trigonal, orthorhombic, monoclinic, triclinic, but also performs well
when simulating plasticity of cubic materials. It simulates the plastic deformation of aggregates
subjected to external strain, external stress, or a combination of both. It is based on the physical
shear mechanisms of slip and twinning, and accounts for grain interaction effects. The model
considers the deformation based on the crystal plasticity mechanisms activated by a Resolved
Shear Stress [99]. In addition to the provided macroscopic stress-strain response, it accounts
for hardening, reorientation and shape change of individual grains. Therefore, it predicts the
evolution of hardening and texture associated with plastic forming.
As it is well known, the deformation is accommodated by crystallographic slip and twin shear
rates inside the grains. In that sense, it is typically assumed that in the HCP metals, the shear is
accommodated by {0001}<1120 > basal slip, 1010<1120 > prismatic slip, 1010<
1123 > pyramidal slip and 1012<1011 > tensile twinning. Their threshold stresses for
activation are very different depending on the material, temperature and deformation rate. The
threshold stress describes the resistance for activation that the deformation mechanisms
experience, and it usually increases with deformation. Then, the VPSC model has the capability
of using a reference hardening function for each system, described by an extended Voce law
[100].
The Voce hardening is characterized by an evolution of the threshold stress with accumulated
shear strain in each grain of the form:
𝜏𝜏𝑠𝑠= 𝜏𝜏0
𝑠𝑠+ (𝜏𝜏1
𝑠𝑠+𝜃𝜃1
𝑠𝑠Г)(1 exp �−Г𝜃𝜃0
𝑠𝑠
𝜏𝜏1
𝑠𝑠��) (Eq.3)
Where
Г=∆𝛾𝛾𝑠𝑠
𝑠𝑠 is the accumulated shear in the grain, 𝜏𝜏0 is the initial CRSS, 𝜃𝜃0 is the initial
hardening rate, 𝜃𝜃1 is the asymptotic hardening rate and (𝜏𝜏0+𝜏𝜏1) corresponds to the back
extrapolated CRSS.
Furthermore, the code also incorporates the ‘self’ and ‘latent’ hardening defining the
coefficient 𝑠𝑠𝑠𝑠, which account for the obstacles that new dislocations associated with 𝑠𝑠
activity represent for the propagation of system ‘s’. The increase of the threshold stress of a
system due to the shear activity ∆𝛾𝛾𝑠𝑠 in the grain is given by:
20
∆𝜏𝜏𝑠𝑠=d𝜏𝜏
𝑠𝑠
𝑠𝑠𝑠𝑠
𝑠𝑠′ ∆𝛾𝛾𝑠𝑠′ (Eq. 4)
Where
d𝜏𝜏
𝑠𝑠
=�𝜃𝜃1+��𝜃𝜃0
𝜏𝜏1+𝜏𝜏1𝜃𝜃1𝑒𝑒𝑒𝑒𝑒𝑒�−Г𝜃𝜃0
𝜏𝜏1��+𝜃𝜃0
𝜏𝜏1𝜃𝜃1Г 𝑒𝑒𝑒𝑒𝑒𝑒�−Г𝜃𝜃0
𝜏𝜏1��� (Eq. 5)
The code uses the self-hardening as a reference and set 𝑠𝑠𝑠𝑠 = 1. At that time, when ‘self’ and
‘latent’ hardening are indistinguishable, then 𝑠𝑠𝑠𝑠= 1, the evolution of threshold stress is given
by only the Voce hardening function:
∆𝜏𝜏𝑠𝑠=d𝜏𝜏
𝑠𝑠
∆Г (Eq. 6)
The hardening law labeled in the equations previously showed, let to describe the hardening
rate observed at the beginning of plasticity and its decrease towards constant hardening rate at
large strains. The condition 𝜃𝜃0𝜃𝜃10, 𝜏𝜏10 corresponds to an increment of the yield stress
and a reduction of the hardening rate with a tendency to linear saturation. The linear hardening
corresponds to a limit case of the law corresponding to 𝜏𝜏1
𝑠𝑠= 0 while the case of rigid-perfectly
plastic hardening corresponds to 𝜃𝜃0=𝜃𝜃1=𝜏𝜏1= 0. This is schematically exemplified in
figure 2.15a.
The incremental expression referred in Eq. 6 represents a forward extrapolation, which tends
to overestimate the hardening and make it dependent on the step size, more so when the
derivative is large. Consequently, in the code is implemented an analytic integration of Eq. 6.
Normally, the evolution of the threshold stress represented by Eq. 3 is monotonically increasing
and the hardening rate Eq. 4 is monotonically decreasing. This is achieved using a ‘kosher’ set
of parameters 𝜏𝜏0> 0, 𝜏𝜏1> 0, 𝜃𝜃0>𝜃𝜃1> 0. Nevertheless, for some empirical cases, it is
possible to use parameters giving monotonic decrease 𝜏𝜏0> 0, 𝜏𝜏1< 0, 𝜃𝜃0<𝜃𝜃1< 0 or
increased the hardening rate 𝜃𝜃1>𝜃𝜃0> 0. The VPSC model accepts parameters describing
negative hardening. The figure 2.15b shows possible configurations of Voce parameters
leading to non-classic hardening.
Figure 2.16. Schematic representation of a) the dependence of the hardening curve on the
modified Voce model parameters, and b) possible configurations of Voce parameters leading
to non-classic hardening [99].
21
In the VPSC model is assumed that twinning has associated a critical resolved shear of
activation in the twinning plane and along the twinning direction, like a slip system. However,
it differs from slip in its directionality, which is modeled by allowing activation only if the
resolved shear stress is positive along the Burgers vector of the twin. Here, the hardening
induced by twinning is empirically enforced by assigning high values to the latent hardening
coefficients 𝑠𝑠𝑠𝑠 describing slip-twin and twin-twin interactions.
As for the effect on texture of the twinned fractions, here it is used the Predominant Twin
Reorientation Scheme (PTR) proposed by Tomé et al [101].
Such PTR scheme works as follows: within each grain g it is keep a track of the shear strain
γt,g contributed by each twin system t, and of the associated volume fraction Vt,g= γt,g/St as
well. In this case, St refers to twin shear. The sum over all the twin systems associated with a
given twin mode, and over all grains, represents the accumulated twin fraction (Vacc, mode) in
the aggregate for the particular twin mode.
Vacc,mode =��γt,gSt
tg
(Eq. 7)
Each twinned fraction is not numerically feasible to be considered as a new orientation, then;
in the PTR scheme is adopted a statistical approach. At each incremental step, some grains are
fully reoriented by twinning, provided certain conditions are fulfilled.
This is call as 'effective twinned fraction' (Veff, mode) the volume associated with the fully
reoriented grains for that mode, and define a threshold volume fraction as
Vth, mode = A th1 + A th2 Veff, mode
Vacc, mode (Eq. 8)
After each deformation increment, a grain is picked at random and is identified the twin system
with the highest accumulated volume fraction.
If the latter is larger than the threshold (Vth, mode) then the grain is allowed to reorient and
(Veff, mode) and Vth, mode are updated. The process is repeated either until all grains are
randomly checked or until the effective twin volume exceeds the accumulated twin volume. In
the latter case, it is stopped the reorientation by twinning and proceed to the next deformation
step. In this process, two things are achieved: a) only the historically most active twin system
in each grain is considered for reorienting the whole grain by twinning; b) the twinned fraction
is consistent with the shear activity that the twins contribute to deformation. The algorithm
given in Eq. 8 prevents the grain reorientation by twinning until a threshold value A th1 is
accumulated in any given system and rapidly raises the threshold to a value around A th1 +
A th2.
The input to the code consists in the initial crystallographic texture (grain orientations and
weights), the single crystal properties (active slip and twinning systems, their critical resolved
shear stresses, and the associated hardening parameters), the boundary conditions (overall
velocity gradient components, or overall stress components) and the parameters controlling
convergence, precision and type of run. The output of the code is: the final crystallographic
and morphologic textures after deformation (additionally is also possible to obtain the
intermediate textures), the evolution of the stress and strain components during deformation,
the statistics of slip and twinning systems activity during deformation, the statistic over grain
stress and strain-rate components and their standard deviations.
22
2.10 Summary and motivation of the work
In summary, Magnesium and its alloys have a great potential to be used in the replacement of
diverse components in automotive, aerospace and biomedical industries. Nevertheless, the low
strength, limited ductility and high anisotropy still limit the use of wrought Mg alloys. Such
issues are directly associated to the limited number of active slip systems in directions other
than those contained in the basal planes of the HCP structure. Usually, the activation of
determined slip systems are connected to the texture of the Mg alloys. In agreement to that,
great progress has been made in the last decade to produce alloys and design thermomechanical
treatments for modifying the texture to overcome the issues previously mentioned. However,
still significant efforts are required to optimize the microstructure and mechanical properties.
It is known that thermomechanical processes like rolling or extrusion have been employed to
enhance the microstructure of Mg alloys. Concerning to extrusion process, it has a high
potential to be employed in the manufacturing of components for industrial applications. For
years, many studies have been realized while focusing in the classical wrought Mg alloys such
as AZ31. More recently, Mg alloys with additions of Rare earth elements (RE) like Nd or
Calcium (Ca) show important improvements on the mechanical behavior. The combination of
Zn with Ca or Nd can create a synergetic effect to modify the microstructure and texture of
Magnesium alloys. The processing routes combined with alloying elements offer the prospect
to modify the microstructure and the strong textures of wrought Mg alloys and hence their
mechanical behavior. Therefore, a profound understanding of the correlation between
processing parameters and the microstructure as well as the microstructure-mechanical
properties correlation of the processed alloys is necessary for tailoring materials for specific
applications or for processes post-extrusion.
During deformation, in addition to the activation of basal <a> mode, the Ca or Nd additions
into Mg-Zn alloy system, activate non-basal deformation modes, for instance, prismatic <a>
and pyramidal <c+a> at low temperatures. Such deformation modes mitigate the need of
twinning and thus improve ductility and reduce the tension-compression yield asymmetry.
However, although Ca or Nd additions offer the prospect to improve mechanical properties,
understanding of the mechanisms by which they can enhance it is still lagging. Consequently,
the study of the effect of initial microstructure and texture is fundamental to understand which
slip modes are activated to satisfy the material behavior during deformation and texture
development.
23
3. Experimental procedure
3.1 Studied and cast alloys
In this work, three different alloys were investigated. The base alloy Z1 (Mg-Zn) was modified
with separate additions of Calcium and the rare earth element Neodymium to obtain the ZX10
(Mg-Zn-Ca) and ZNd10 (Mg-Zn-Nd) alloys, respectively.
The billets for extrusion processing were produced by permanent mold casting. The alloying
element additions were added to the molten magnesium in the furnace at temperatures between
750 and 770 °C. The molten melt was stirred for around 15 minutes to make a homogeneous
material before pouring. Table 3.1 shows the chemical composition of each alloy. The values
are in weight percent. The figure 3.1a shows an example of the billets after casting. In the
preparation of the material for extrusion, the cylindrical billets were prepared with a length of
150 mm and a diameter of 49 mm. An example of the billets prepared with the dimensions for
extrusion processing is depicted in figure 3.1b.
Table 3.1. Chemical composition of each alloy.
Chemical composition (wt%)
Alloy
Zn
Ca
Nd
Mg
Mg-Zn
0.91
----------
----------
Bal.
Mg-Zn-Ca
0.94
0.15
----------
Bal.
Mg-Zn-Nd
0.98
----------
0.57
Bal.
Figure 3.1. a) Billet after casting process, and b) prepared billet for extrusion processing.
3.2 Heat treatments before extrusion
Solution heat treatments were applied to the billets before extrusion. For the Z1 and ZX10
alloys, the heat treatments were done at 400 °C. In case of the ZNd10 alloy, the applied
temperature was 450 °C.
A higher temperature was required for the ZNd10 alloy because of complexity to achieve a
solid solution. In each alloy, the heat treatment was applied for 16 hours. In the figure A.1 of
appendix, the zones that were reached during the solution treatments are depicted.
24
3.3 Extrusion process
The thermomechanical process chosen for the processing of the selected alloys was indirect
extrusion. The billets were preheated before extrusion at the extrusion temperature for 60
minutes. The selection of the first parameters like extrusion speed and extrusion temperature
was done according to the literature review. A constant extrusion ratio was used during the
extrusion modifying the temperature and speed. The modification of extrusion speed and
temperature led to obtain different materials in terms of microstructure and texture in each
alloy. The obtained results allowed analyzing the effect of both parameters on the
recrystallization behavior as well as show a processing window in each alloy. The extrusion
experiments were executed at the Helmholtz-Zentrum Hereon. It was carried out using a 2.5
MN automatic extrusion press of Müller Engineering (Müller Engineering GmbH & Co. KG,
Todtenweis/Sand, Germany), see Figure 3.2.
Figure 3.2. Extrusion press machine of the Magnesium Innovation Center at the Helmholtz-
Zentrum Hereon.
3.4 Indirect extrusion
The applied parameters during the indirect process are shown in Table 3.2. It was used as an
extrusion die with a round hole of 10 mm in diameter. The combination of the extrusion
parameters and the die used allowed the production of extruded round bars with a diameter of
10 mm.
Table 3.2. Extrusion parameters during processing.
Indirect extrusion
Extrusion temperature (°C)
Extrusion speed (mm/s)
Extrusion ratio
250
5.0
7.5
1:25
300
5.0
7.5
1:25
400
5.0
7.5
1:25
3.5 Post-extrusion heat treatments
To understand the effect of static recrystallization on the microstructure and texture
development, partly recrystallized microstructures were annealed to fully recrystallized
microstructures. The partly recrystallized samples from ZX10 and ZNd10 were obtained at 0.1
mm/s of extrusion speed and 300 °C of extrusion temperature. The Z1 alloy does not presented
changes in the texture after SRX because of the high fraction of recrystallized microstructure
Specifications
Maximum ram speed
8.0 mm/s
Diameter of the container
50 mm
Length of the billet
150 mm
Press force
2.5 MN
25
found even in samples extruded at low speed. Therefore, it analysis is not presented. The
microstructures of samples from the ZX10 and ZNd10 alloys were tailored so the grain size of
the SRX samples could be compared to grain structures of DRXed samples, i.e. samples from
ZX10 alloy (extruded at 300 °C and 2.0 mm/s) and ZNd10 alloy (extruded at 300 °C and 5.0
mm/s). The partly recrystallized samples were annealed at 400 °C for 3.0 minutes in case of
ZX10 alloy and 30 minutes for the ZNd10 alloy.
3.6 Microstructure characterization
To study the microstructure development of each alloy, samples from the extruded bars were
obtained, and then those samples were prepared following the standard methodology for
metallographic observations. The samples were taken from the middle of each extruded bar.
Then, the samples were ground to obtain a surface perpendicular to the extrusion direction. The
grinding procedure started with a grinding paper of the number 500 finishing such process with
a grinding paper of the number 2500. Then, the polishing process using a colloidal suspension
(Struers OPS). Following this, an etchant liquid based on picric acid was used to reveal the
microstructure.
The microstructure of all the alloys were observed by light optical microscopy. Pictures of each
etched sample were taken at different magnifications using a Leica optical microscope (Leica
DM15000M). The average grain size of each material was determined on pictures at 100X by
means of the computer software AnalySIS ProTM that uses the linear intercept method. The
preparation of the samples for texture analysis followed exactly the methodology previously
explained with the difference that the samples were not etched. Pole figures (1120), (0001),
(1010), (1011), (1012), (1013) were measured up to a tilt angle of 70° in a Panalytical texture
goniometer (PANalytical X'Pert PRO MRD). A beam size of 1.0 mm2 and Cu radiation
were employed during the operation of the texture goniometer. The inverse pole figures were
recalculated using an open-source code MTEX [102].
The electron back scattered diffraction (EBSD) technique was employed to do a more detailed
microstructure and grain orientation analysis. The sample preparation consisted on the
implementation of the standard metallographic techniques followed by electro polishing using
an AC2 solution (Struers ™) at 30 V for 25 s at -20 °C. The EBSD analyses were performed
to measure the local orientation patterns using a field emission scanning microscope (Zeiss,
Ultra 55, EDAX/TSL) on the longitudinal sections of the samples. The operation voltage was
15 kV and the used scanning step size was 0.45 μm.
3.7 Mechanical characterization
The mechanical properties were characterized applying uniaxial tensile and compression
testing at room temperature. Samples for tensile tests were prepared with a diameter of 5.0 mm
and a gauge length of 30.0 mm. Samples for compression tests were prepared with a length of
13.5 mm and a diameter of 9.0 mm. Mechanical testing was carried out along the extrusion
direction using a 50 KN testing machine (Zwick Z050). All tests were performed at a constant
strain rate of 10-3 s-1. In figure A.2 of appendix, the schematic representation of the samples
used for the mechanical characterization are shown.
26
3.8 Slip trace analysis
Ex-situ tension and compression tests were performed on samples with comparable grain sizes
but different crystallographic texture. One sample per alloy was used either in tension or in
compression. The tests were performed at room temperature at a strain rate of 1.5 x 10-4 s-1 to
observed slip lines of distinct deformation modes. In tension, the tests were paused at strains
of 0.03 and 0.12 in the three alloys. However, in compression the Z1 alloy was paused first at
0.03 and then at 0.08 strain. On the other hand, the ZX10 and ZNd10 alloys were tested as in
tension. The tests were paused at such strains to allow the EBSD data acquisition for
determining the grains orientation after each strain. Slip lines on polished and electro polished
samples have been observed using pictures from a Laser Scanning Microscope (LSM),
Keyence VK-X100/X200 series.
The slip trace methodology applied to determine the slip activity is as follows. In first place,
on (LSM) pictures, straight lines on the grains were defined. The slip traces were pointed out
with a black dashed line like the one displayed in figure 3.3a. The orientation of the grains
showing slip lines was obtained from the inverse pole figure (IPF) maps of EBSD, i.e. its Euler
angles (figure 3.3b). The Euler angles were used as the input of a MATLAB code [103] that
gives as a result the possible slip system (Figure 3.3c). In this work, the theoretical slip trace
directions for 36 slip systems (SS) were computed using the Euler angles of each grain. The
observed slip trace was compared to each of the possible slip systems.
Finally, the one that provides the best match was considered as the activated slip system in
such a grain. An example of a basal <a> slip trace is denoted by the black dashed line; see
Figure 3.3c.
In table 3.3 are depicted the slip systems considered in the slip trace analysis. The previously
explained procedure was applied to each grain that showed slip traces.
Figure 3.3. Example of slip trace analysis; a) LSM picture showing the appearance of slip trace
(black dashed line), b) IPF map corresponding to the LSM picture, and c) estimation of the 36
possible slip traces and determination of the active slip system according to the best match.
The black dashed line points out the corresponding slip trace.
27
Table 3.3. Slip systems taken into account during the slip trace analysis.
Slip
system
number
Slip
system
Slip system
type
Identification
color
Slip
system
number
Slip
system
Slip system
type
Identification
color
1
<0001 > (2110)
Basal
<a> Black
19
< 1011 > (2113)
Paramidal I
<c+a> Green
2
<0001 > (1210)
20
< 1011 > (1123)
3
<0001 > (1120)
21
<0111 > (1123)
4
<1010 > (1210)
Prismatic
<a> Red
22
<0111 > (1213)
5
< 0110 > (2110)
23
< 1101 > (2113)
6
< 1100 > (1120)
24
< 1101 > (1213)
7
<1011 > (1210)
Pyramidal
<a> Blue
25
<1122 > (1123)
Pyramidal II
<c+a> Wine
8
< 0111 > (2110)
26
<1122 > (1123)
9
< 1101 > (1120)
27
< 2112 > (2113)
10
< 1011 > (1210)
28
< 2112 > (2113)
11
<0111 > (2110)
29
< 1212 > (1213)
12
< 1101 > (1120)
30
< 1212 > (1213)
13
<1011 > (1123)
Paramidal I
<c+a> Green
31
< 1122 > (1123)
14
<1011 > (2113)
32
< 1122 > (1123)
15
< 0111 > (1123)
33
< 2112 > (2113)
16
< 0111 > (1213)
34
< 2112 > (2113)
17
< 1101 > (2113)
35
< 1212 > (1213)
18
< 1101 > (1213)
36
< 1212 > (1213)
3.9 Crystal plasticity simulations applying the VPSC model
Depending on the crystallographic texture, different combinations of deformations systems can
be activated. In this work, the Visco Plastic Self Consistent (VPSC) model has been applied to
simulate the plastic deformation during of the extruded bars with a comparable grain size and
different crystallographic texture. The simulations with the VPSC model was carried out using
the PC- based software selfcon [104].
The crystallographic texture was used in the form of discrete orientations with their weights.
The texture discretization was generated using an MTEX code [102]. In order to have a better
fit with the experimental texture and stress-strain curve, first a number of crystal orientations
were selected.
In the development of this work, a number of 3000 crystal orientations were used. In the
simulation, the considered slip modes were the basal <a>, prismatic <a> the pyramidal <a>,
pyramidal I <c+a> and pyramidal II <c+a> slip as well as tensile twinning 1012<101 1 >
and compressive twinning 1011<101 2 >. The input parameters used at the beginning of
the simulation were initially collected from different references [105-108]. Then, these
parameters were modified during the course of the simulations to obtain the best fit between
the experimental and simulated results, i.e., crystallographic texture (measured at the different
strains) and the stress-strain curves.
28
4. Results
4.1 Extrusion processing
The extrusion experiments have been made at different extrusion conditions. The measured
data during the processing extrusion are graphically represented in this section. In figures 4.1
to 4.3 the behavior of each alloy during the indirect extrusion processing is shown. The
evolution of extrusion force during the process in each alloy is observed.
Figures 4.1 to 4.3 show the evolution of the extrusion force as a function of the ram
displacement during processing. Except for the beginning of the extrusion of ZX10 alloy at
250 °C and 2.0 mm/s (Figure 4.1a), the rest of the experiments show a quite stable behavior.
Once the force rises to its peak value, the force starts to decrease and a steady region during
extrusion occurs, i.e., there is a displacement of the ram, but the force is constant. It is visible
that in all the extrusion experiments, the addition of Ca and Nd into the binary Z1 alloy show
an increase of the extrusion peak force. From the ternary alloys extruded at 250 °C and 300 °C,
the highest-force peak is observed in the Ca containing alloy. At 400 °C of extrusion
temperature, the force peak in the ternary alloys is similar. In all the alloys, there is a reduction
of the extrusion force with the increase of extrusion temperature.
Figure 4.1. Extrusion force evolution as a function of ram displacement at 250 °C of processing
temperature, a) 2.0 mm/s, b) 5.0 mm/s, and c) 7.5 mm/s.
29
Figure 4.2. Extrusion force evolution as a function of ram displacement at 300 °C of processing
temperature, a) 2.0 mm/s, b) 5.0 mm/s, and c) 7.5 mm/s.
Figure 4.3. Extrusion force evolution as a function of ram displacement at 400 °C of processing
temperature, a) 2.0 mm/s, b) 5.0 mm/s, and c) 7.5 mm/s.
30
Based on the applied extrusion parameters, figures 4.4 to 4.6 show the evolution of the surface
appearance in each alloy during the extrusion processing.
Figure 4.4 displays the surface appearance on the Z1 alloy. From the figure, some lines or
striations perpendicular to the extrusion direction (ED) are slightly visible. These perpendicular
lines are more visible at the extrusion temperature of 400 °C at all the extrusion speeds, figure
4.4c. The extruded surfaces also show the presence of lines/scratches parallel to the ED.
Clearly, these lines are visible at all the extrusion conditions.
Figure 4.4. Pictures showing the surface of as-extruded alloy Z1 processed at different
extrusion speeds and temperatures; a) 250 °C, b) 300 °C, and c) 400 °C. The extrusion speed
increases from top to bottom. Yellow arrow indicates the extrusion direction (ED).
Figure 4.5 provides the surface appearance on the ZX10 alloy. As can be seen from the figure,
the lines/striations perpendicular to the ED are also visible. As opposed to previous alloy, it
seems that these perpendicular lines are present at all the extrusion conditions.
31
Similarly to previous alloy, the extruded surfaces also show the presence of lines/scratches
parallel to the ED.
Figure 4.5. Pictures showing the surface of as-extruded alloy ZX10 processed at different
extrusion speeds and temperatures; a) 250 °C, b) 300 °C, and c) 400 °C. The extrusion speed
increases from top to bottom. Yellow arrow indicates the extrusion direction (ED).
Figure 4.6 presents the surface appearance on the ZNd10 alloy. As can be seen from the figure,
there is no clear presence of lines/striations perpendicular to the ED. On the other hand, as in
the previous alloys, the extruded surfaces of this ZNd10 alloy also show the presence of
lines/scratches parallel to the ED.
32
Figure 4.6. Pictures showing the surface of as-extruded alloy ZNd10 processed at different
extrusion speeds and temperatures; a) 250 °C, b) 300 °C, and c) 400 °C. The extrusion speed
increases from top to bottom. Yellow arrow indicates the extrusion direction (ED).
4.2 Microstructure evolution
4.2.1 Z1 alloy
Figure 4.7 shows the micrographs from longitudinal sections of the extruded bars, parallel to
the extrusion direction. The average grain size (d) is shown as an inset in each picture. The
microstructure evolution is organized as a function of processing temperature in such a way
that the extrusion temperature increases from left to the right. It also shows the microstructure
evolution in terms of the processing speed so that, the extrusion speed increases from up to
down. The resulted microstructure exhibited completely recrystallized grains. It consists in
homogeneous equiaxial coarse grains. It is observed that at each constant temperature, with the
increase of extrusion speed there is a clear increase of the average grain size. It increases from
21.0 µm to 37.1 µm, as can be seen in figure 4.7a. On the other side, the figure 4.7b shows that
33
the average grain size increases from 21.3 µm to 44.5 µm. Furthermore, an increase of the
average grain size from 35.2 µm to 51.4 µm at the extrusion temperature of 400 °C, see figure
4.7c. As shown also in figure 4.7, increasing the extrusion temperature while maintaining
constant the extrusion speed, the same basic behavior is found in the average grain size
evolution. It increases with the increase of extrusion temperature. The average grain size
increases from 21.0 µm to 35.2 µm, from 34.7 µm to 42.9 µm and from 37.1 µm to 51.4 µm at
the extrusion speeds of 2.0 mm/s, 5.0 mm/s and 7.5 mm/s, respectively.
Overall, the microstructure results shown in figure 4.7 indicate that the extrusion parameters
applied to the alloy Z1 do not affect its recrystallization process. Clearly, the applied extrusion
parameters only influence the average grain size.
Figure 4.7. Microstructure evolution of the Z1 alloy extruded at a) 250 °C, b) 300 °C, and c)
400 °C. The inset (d) in each picture represents the average grain size. The extrusion speed
increases from top to bottom. The black arrow indicates the extrusion direction (ED).
4.2.2 ZX10 alloy
Figure 4.8 provides the results on the microstructure evolution of the ZX10 alloy. The
microstructure evolution is organized as in the previous alloy. The microstructure consists of
homogeneous equiaxial grains. Clearly, the microstructure of this alloy is fully recrystallized
at all the extrusion conditions. Then, the figure 4.8 suggests that this alloy developed a
complete recrystallization process at the applied extrusion conditions in this work. It is
observed that maintaining constant the extrusion temperature while modifying the extrusion
speed, there is an increase of the grain size. There is a grain size increase from 9.5 µm to 18.5
µm at the extrusion temperature of 250 °C as can be seen in figure 4.8a. Meanwhile, at the
34
extrusion temperature of 300 °C (figure 4.8b), the grain size grows from 10.0 µm to 25.7 µm.
At the highest extrusion temperature, i.e., 400 °C, the grain size increases from 25.1 µm to 34.0
µm, see figure 4.8c.
A similar finding is observed when the extrusion speed is constant, and the extrusion
temperature varies. From the upper section of the figure 4.8, at 2.0 mm/s, the average grain
size increases from 9.5 µm to 25.1 µm when increasing the extrusion temperature. An increase
from 14.0 µm to 31.3 µm is observed when the alloy is extruded at 5.0 mm/s modifying the
extrusion temperature, see the middle section of figure 4.8. At the bottom section of table 4.8,
it is observed that when the alloy is extruded at constant speed of 7.5 mm/s, the average grain
size increases from 18.5 µm to 34.0 µm.
In summary, the findings in this alloy relating to the grain size evolution is comparable to the
one observed in the Z1 alloy, it increases either at the increment of extrusion speed or
temperature. However, in comparison with the behavior of the Z1 alloy, the addition of Ca to
the Mg-Zn system reduced the average grain size during the processing. That means, the Ca
delayed the grain growth of the Mg-Zn alloy system.
Figure 4.8. Microstructure evolution of the ZX10 alloy extruded at a) 250 °C, b) 300 °C, and
c) 400 °C. The inset (d) in each picture represents the average grain size. The extrusion speed
increases from top to bottom. The black arrow indicates the extrusion direction (ED).
4.2.3 ZNd10 alloy
Figure 4.9 displays the microstructure evolution in the ZNd10 alloy. The microstructure
evolution in this alloy is different in comparison with the alloys previously observed.
35
This alloy shows a partly recrystallized microstructure which consists of large deform
elongated grains parallel to the extrusion direction surrounded by small recrystallized grains,
see figure 4.9a at 2.0 mm/s. Figure 4.9a also shows that the microstructure completely
recrystallizes increasing the extrusion speed. The average grain size increases from 3.4 µm to
10.5 µm. Figure 4.9b depicts the microstructure evolution at 300 °C of extrusion temperature.
However, even this extrusion temperature, the microstructure is not completely recrystallized.
There are still some traces of non-recrystallized grains, see figure 4.9b at 2.0 mm/s. With
increasing the extrusion speed, the microstructure is completely recrystallized. An increase of
the average grain size from 4.5 µm to 16.1 µm is observed. After extrusion at 400 °C, the alloy
shows a fully recrystallized microstructure, see figure 4.9c. At these extrusion conditions the
average grain size increases from 13.8 µm to 26.5 µm.
The upper section of figure 4.9 shows the microstructure evolution of the alloy extruded at 2.0
mm/s increasing the extrusion temperature. As can be seen, the average grain size increases
from 3.4 µm to 13.8 µm. The middle section of the figure 4.9 displays that at constant extrusion
speed of 5.0 mm/s, the grain size increases from 8.4 µm to 22.4 µm. The bottom section of
figure 4.9 shows the microstructure evolution at 7.5 mm/s of extrusion speed varying the
extrusion temperature. The average grain size increases from 10.5 µm to 26.5 µm.
In summary, the microstructure evolution of the ZNd10 alloy shows a distinct behavior if
compare with the alloys previously observed. The addition of Nd into the Mg-Zn system also
delayed the recrystallization process. However, such a delay is more pronounced than with the
Ca addition. This is revealed due to the presence of non-recrystallized grains as well as higher
retardation of the grain growth.
Figure 4.9. Microstructure evolution of the ZNd10 alloy extruded at a) 250 °C, b) 300 °C, and
c) 400 °C. The inset (d) in each picture represents the average grain size. The extrusion speed
increases from top to bottom. The black arrow indicates the extrusion direction (ED).
36
4.3 Crystallographic texture evolution
4.3.1 Z1 alloy
Figure 4.10 shows the crystallographic texture development of the Z1 alloy as a function of the
applied extrusion processing parameters. The crystallographic texture is shown in the form of
inverse pole figures in the extrusion direction. The texture intensity is in the unit of multiple of
a random distribution (m.r.d).
As in traditional Mg alloys, the Z1 alloy shows a typical texture where the basal planes are
aligned parallel to the extrusion direction. The preferential orientation of the grains is mainly
distributed between the <1010 > and the <1120 > poles. As can be observed, there is a
certain tendency of the grains to be preferably located at the <1120 > pole. This texture
development is observed at all the extrusion conditions.
In summary, the results in figure 4.10 show that there is no significant influence of the applied
extrusion parameters in the crystallographic texture evolution of this alloy.
Figure 4.10. Inverse pole figures showing the crystallographic texture evolution of Z1 alloy
extruded at a) 250°C, b) 300°C, and c) 400°C. Texture intensity is in m.r.d. The extrusion speed
increases from top to bottom.
37
4.3.2 ZX10 alloy
Figure 4.11 illustrates the crystallographic texture evolution in the ZX10 alloy. As opposed to
Z1 alloy, in this ZX10 alloy, a different crystallographic texture development is observed.
In figure 4.11a, the crystallographic texture evolution at a constant extrusion temperature of
250 °C varying the extrusion speed is shown. As can be seen, the texture is dominated mainly
by the <1011 > pole. This crystallographic texture component remains with increasing the
extrusion speed.
Furthermore to <1011 > pole, an additional component appears between the <0001 > and
the <1120 > poles, i.e. the <1121 > pole at the extrusion of 300 °C, see figure 4.11b at 2.0
mm/s. With the following increase of extrusion speed, additionally to the <1011 > pole some
orientations appear at the <1120 > pole.
In figure 4.11c, the crystallographic texture at the highest extrusion temperature, i.e., 400 °C,
is shown. The most surprising aspect at this extrusion temperature is that the crystallographic
texture evolved to a basal one at all the extrusion speeds. Clearly, the intensities are distributed
between the <1010 > and the <1120 > poles.
Figure 4.11. Inverse pole figures showing the crystallographic texture evolution of the ZX10
alloy extruded at a) 25C, b) 300°C, and c) 400°C. Texture intensity is in m.r.d. The extrusion
speed increases from top to bottom.
38
4.3.3 ZNd10 alloy
Figure 4.12 shows the crystallographic texture development on the ZNd10 alloy. In figure
4.12a the crystallographic texture evolution at the extrusion temperature of 250 °C in
combination with the extrusion speed variation is shown. As can be seen, a fibre-type texture,
which corresponds to the prismatic <1010 > pole is formed, here the prismatic planes are
being mainly oriented perpendicular to the extrusion direction, figure 4.12a at 2.0 mm/s. As it
is observed, the crystallographic texture consists of a strong prismatic texture. The <1010 >
component is related with the presence of long elongated grains on the microstructure. Figure
4.12a also depicts that, in addition to the <1010 > component, the <1121 > pole is slightly
visible. Increasing the extrusion speed, the crystallographic texture is weaker. The intensity of
the single <1010 > component totally vanishes and the <1121 > pole dominates the texture.
Figure 4.12b, shows the crystallographic texture development at 300 °C of extrusion
temperature. The <1010 > component is still visible but its intensity decreases, figure 4.12b
at 2.0 mm/s. Moreover, the <1121 > pole gains intensity and a further pole, i.e., the <2023 >
pole develops between the <0001 > pole and the <1010 > pole. With a further increase of
extrusion speed, the texture intensity decreases more. Then, an arc is formed between the poles
<1121 > and <2023 > and dominates the crystallographic texture.
At the extrusion temperature of 400 °C, the <1121 > pole is established and dominates the
texture at 2.0 mm/s and 5.0 mm/s, see figure 4.12c. With the further increase of extrusion speed,
the arc is formed between the poles <1121 > and <2023 >.
Figure 4.12. Inverse pole figures showing the crystallographic texture evolution of the ZNd10
alloy extruded at a) 25C, b) 300°C, and c) 400°C. Texture intensity is in m.r.d. The extrusion
speed increases from top to bottom.
39
4.4 Mechanical properties
4.4.1 Z1 alloy
Figure 4.13 shows the stress-strain curves of the alloy tested in tension and in compression.
The graphs on the left side correspond to the tension tests while the graphs on the right side
correspond to the compression tests.
At the extrusion temperature of 250 °C and varying the extrusion speed, the figure 4.13a
displays that the tensile yield stress (TYS) of the alloy decreases with the increment of
extrusion speed. Furthermore, the ultimate tensile stress (UTS) is higher at 2.0 mm/s, while it
is quite comparable at 5.0 and 7.5 mm/s. A reduction of the fracture strain is observed with the
increment of the extrusion speed. On the other hand, in compression (right side in figure 4.13a)
the characteristic sigmoidal hardening (S-shape) is visible when a material is tested in
compression. The stress-strain curves show a reduction of the compressive yield stress (CYS)
with the increment of extrusion speed. As can be seen, the ultimate compressive stress (UCS)
and the fraction strain are comparable at all the extrusion speeds.
Figure 4.13b shows the stress-strain curves of the alloy at the extrusion temperature of 300 °C.
The TYS is higher at 2.0 mm/s but comparable at 5.0 and 7.5 mm/s. The UTS and fracture
strain both decrease with the increment of the extrusion speed. A similar trend is observed in
the material tested in compression. Clearly, there is a reduction of the CYS and UCS as the
extrusion speed increases. The lowest fracture strain is observed in the material extruded at 2.0
mm/s.
In figure 4.13c the stress-strain curves of the alloy extruded at 400 °C are presented. The lowest
TYS and UTS are observed at 7.5 mm/s. The fracture strain is slightly reduced with the
increment of extrusion speed from 2.0 mm/s to 5.0 mm/s. The highest reduction is observed at
the highest extrusion speed. Finally, in compression, with the increase of extrusion speed from
2.0 mm/s to 5.0 mm/s, the CYS is reduced, then, it increases with the further increase of the
extrusion speed. The UCS shows a slight increase. The fracture strain is reduced as the
extrusion speed increases.
40
Figure 4.13. Tension and compression stress-strain curves of Z1 alloy extruded at a) 250 °C,
b) 300 °C, and c) 400 °C. Tested at room temperature.
41
In table 4.1, the mechanical properties of the Z1 alloy are summarized.
Table 4.1. Tension and compression properties of the Z1 alloy measured parallel to the
extrusion direction.
Extrusion
temperature
(°C)
Extrusion
speed
(mm/s)
TYS
(MPa)
UTS
(MPa)
Fracture
strain
CYS
(MPa)
UCS
(MPa)
Fracture
strain
250
2.0
141.0±1.1
254.0±0.1
0.22±0.01
84.1±1.1
312.0±2.1
0.12±0.01
5.0
130.0±0.6
245.0±0.2
0.19±0.01
71.0±0.1
316.0±3.0
0.11±0.01
7.5
127.0±0.5
241.0±1.0
0.18±0.01
67.0±0.5
318.0±1.0
0.11±0.01
300
2.0
151.0±1.0
261.1±1.0
0.17±0.01
88.0±0.2
322.3±1.0
0.11±0.01
5.0
141.0±2.0
251.0±0.1
0.15±0.01
74.0±0.3
311.0±1.0
0.13±0.01
7.5
140.0±1.0
246.0±4.0
0.17±0.01
68.0±0.1
307.0±2.0
0.13±0.01
400
2.0
145.0±1.0
244.0±1.0
0.16±0.01
71.0±0.3
330.0±1.0
0.12±0.01
5.0
142.0±0.5
241.0±1.0
0.15±0.01
58.0±0.1
316.0±0.5
0.12±0.01
7.5
135.5±0.5
238.0±1.5
0.16±0.01
66.0±0.5
323.0±2.0
0.12±0.01
Tensile and compressive properties of the extruded alloy Z1 as a function of extrusion speed
are shown in figure 4.14. Figure 4.14a suggests that after extrusion at 250 °C, the TYS reduces
approximately 11.0 MPa from 2.0 mm/s to 5.0 mm/s of extrusion. With the further increase of
extrusion speed, from 5.0 to 7.5 mm/s, there is still a reduction of 3.0 MPa. The UTS also
decreases; 9.0 MPa from 2.0 mm/s to 5.0 mm/s and 4.0 MPa from 5.0 mm/s to 7.5 mm/s. The
fracture strain decreases a total amount of 0.04, where the maximum reduction (0.03) is
observed from 2.0 mm/s to 5.0 mm/s. The CYS decreases 13.0 MPa and 4.0 MPa when the
alloy is extruded from 2.0 mm/s to 5.0 mm/s and from 5.0 mm/s to 7.5 mm/s, respectively.
However, the UCS increases a total amount of 6.0 MPa, where the highest increase, i.e. 4.0
MPa is observed from 2.0 mm/s to 5.0 mm/s. The fracture strain does not show significant
changes.
From figure 4.14b, at extrusion temperature of 300 °C, the TYS reduces 10.0 MPa (2.0 mm/s
to 5.0 mm/s). With the further increase of extrusion speed to 7.5 mm/s, there is only a reduction
of 1.0 MPa. The UTS decreases 10.0 MPa (2.0 mm/s to 5.0 mm/s) and 5.0 MPa (5.0 mm/s to
7.5 mm/s). The fracture strain first decreases 0.02 (2.0 mm/s to 5.0 mm/s), then increases. On
the other hand, the CYS decreases 14.0 MPa (2.0 mm/s to 5.0 mm/s) and 6.0 MPa (5.0 mm/s
to 7.5 mm/s). The UCS decreases a total amount of 15.0 MPa, i.e., 11.0 MPa from 2.0 mm/s to
5.0 mm/s and 4.0 MPa from 5.0 mm/s to 7.5 mm/s. The fracture strain increases 0.02.
Figure 4.14c shows the mechanical properties at extrusion temperature of 400 °C. First, the
TYS reduces 3.0 MPa (2.0 mm/s to 5.0 mm/s) and 7.0 MPa (5.0 mm/s to 7.5). Secondly, the
UTS is reduced 3.0 MPa at each extrusion speed increment. The fracture strain is quite
comparable. The CYS decreases 13.0 MPa (2.0 mm/s to 5.0 mm/s) but increases 8.0 MPa (5.0
mm/s to 7.5 mm/s). A similar tendency is observed in the UCS; it decreases 14.0 MPa (2.0
mm/s to 5.0 mm/s) but increases 7.0 MPa (5.0 mm/s to 7.5). The fracture strain is practically
the same.
42
Figure 4.14. Mechanical properties of Z1 alloy at different extrusion speeds and temperature;
a) 250 °C, b) 300 °C, c) 400 °C.
4.4.2 ZX10 alloy
The stress-strain curves obtained from the tension and compression tests of ZX10 alloy are
shown in figure 4.15.
The stress-strain curves of the alloy at the extrusion temperature of 250 °C are shown in figure
4.15a. The TYS of ZX10 alloy decreases with the increment of extrusion speed. Furthermore,
the highest UTS is observed at 2.0 mm/s, while at 5.0 and 7.5 mm/s it is comparable. The
fracture strain is comparable at each extrusion speed.
In compression, the CYS is highest at 2.0 mm/s. It seems that at 5.0 mm/s and 7.5 mm/s the
CYS is similar. The UCS follows a similar tendency with the increment of extrusion speed. As
can be seen, the fraction strain is quite comparable at all the extrusion speeds.
In figure 4.15b the stress-strain curves at the extrusion temperature of 300 °C is shown. As can
be seen, the TYS decreases with the increment of extrusion speed. The highest TYS
corresponds to the material extruded at 2.0 mm/s; at 5.0 mm/s and 7.5 mm/s such feature is
comparable. The UTS behaves similarly to TYS with the increment of extrusion speed. It seems
that the fracture strain is quite comparable at all the extrusion speeds. On the other hand, the
highest UCS corresponds to the material extruded at 2.0 mm/s. This property is reduced at the
extrusion speed of 5.0 mm/s and is maintained at 7.5 mm/s. In reference to the fracture strain,
it seems that is kept constant with the increment of extrusion speed.
The stress-strain curves at the extrusion temperature of 400 °C are displayed in figure 4.15c. It
seems that increasing the extrusion speed, the TYS and the UTS are quite comparable. The
lowest fracture strain is observed on the alloy extruded at 2.0 mm/s. Then, it slightly increased
with increasing the extrusion speed. In compression, it seems that the CYS is slightly higher at
2.0 mm/s of extrusion speed. At 5.0 mm/s and 7.5 mm/s the CYS is quite comparable. Clearly,
the highest UCS corresponds to the alloy extruded at 2.0 mm/s. This property is reduced with
the further increment of the extrusion speed at 5.0 mm/s. Then, it is conserved at 7.5 mm/s.
The lowest fracture strain is observed at 2.0 mm/s of extrusion speed and increases with the
increment of extrusion speed.
43
Figure 4.15. Tension and compression stress-strain curves of ZX10 alloy extruded at a) 250
°C, b) 300 °C, and c) 400 °C. Tested at room temperature.
44
The mechanical properties of this ZX10 alloy are summarized in table 4.2.
Table 4.2. Tension and compression properties of the ZX10 alloy measured parallel to the
extrusion direction.
Extrusion
temperature
(°C)
Extrusion
speed
(mm/s)
TYS
(MPa)
UTS
(MPa)
Fracture
strain
CYS
(MPa)
UCS
(MPa)
Fracture
strain
250
2.0
124.0±1.0
276.0±3.0
0.37±0.01
121.0±0.3
284.0±3.0
0.27±0.01
5.0
94.0±1.0
265.5±0.2
0.36±0.01
92.0±0.5
262.0±1.0
0.27±0.01
7.5
87.0±1.0
258.0±1.0
0.36±0.01
86.0±0.1
269.0±1.0
0.26±0.03
300
2.0
109.0±1.0
272.0±0.5
0.32±0.02
106.0±1.0
277.0±4.0
0.24±0.01
5.0
89.0±0.2
256.5±0.2
0.32±0.01
88.0±0.3
266.0±0.5
0.23±0.01
7.5
83.0±2.0
256.0±0.5
0.31±0.01
84.0±1.0
277.0±3.5
0.25±0.01
400
2.0
140.1±1.0
254.0±1.0
0.23±0.01
89.0±0.2
308.0±2.0
0.17±0.01
5.0
136.5±1.0
264.0±1.5
0.23±0.01
87.0±0.5
292.0±1.0
0.21±0.01
7.5
138.0±1.0
262.0±0.5
0.23±0.01
87.0±0.2
292.0±2.0
0.21±0.01
Tensile and compressive properties of the extruded ZX10 alloy are presented in figure 4.16.
Figure 4.16a depicts that after extrusion at 250 °C while the extrusion speed varies, the TYS
reduces 30.0 MPa (2.0 mm/s to 5.0 mm/s). With the further increase of extrusion speed, the
TYS decreases 7.0 MPa (5.0 to 7.5 mm/s). The UTS also decreases; 10.5 MPa (2.0 mm/s to
5.0 mm/s) and 7.5 MPa (5.0 mm/s to 7.5 mm/s). The fracture strain decreases 0.01 (2.0 mm/s
to 5.0 mm/s) and it is maintained with the further increases of extrusion speed. A reduction is
also observed, approximately 29.0 MPa and 6.0 MPa when the alloy is extruded from 2.0 mm/s
to 5.0 mm/s and from 5.0 mm/s to 7.5 mm/s, respectively. The UCS decreases 22.0 MPa (2.0
mm/s to 5.0 mm/s) but increases 7.0 MPa (5.0 mm/s to 7.5 mm/s). The fracture strain is similar
at 2.0 mm/s and 5.0 mm/s but decreases 0.01 from 5.0 mm/s to 7.5 mm/s.
Figure 4.16b displays that at the extrusion temperature of 300 °C; the TYS reduces 26.0 MPa
in total, i.e. 20.0 MPa (2.0 mm/s to 5.0 mm/s) and 6.0 MPa (5.0 mm/s to 7.5 mm/s). The UTS
decreases 15.5 MPa (2.0 mm/s to 5.0 mm/s) and only 0.5 MPa (5.0 mm/s to 7.5 mm/s). The
fracture strain decreases only 0.01 at the highest extrusion speed. On the other hand, the CYS
decreases 18.0 MPa (2.0 mm/s to 5.0 mm/s) and 4.0 MPa (5.0 mm/s to 7.5 mm/s). The UCS
decreases 11.0 MPa (2.0 mm/s to 5.0 mm/s) but increases also 11.0 MPa (5.0 mm/s to 7.5
mm/s). The fracture strain decreases 0.01 (2.0 mm/s to 5.0 mm/s) but increases 0.02 (5.0 mm/s
to 7.5 mm/s).
Figure 4.16c shows the mechanical properties at the constant extrusion temperature of 400 °C.
Clearly, the TYS is higher at these extrusion conditions if compared with the previously
observed ones. However, it is also modified as an effect of the extrusion speed variation. First,
the TYS reduces 3.6 MPa (2.0 mm/s to 5.0 mm/s), then increases 1.5 MPa (5.0 mm/s to 7.5).
The UTS increases 10.0 MPa (2.0 mm/s to 5.0 mm/s) but decreases 2.0 MPa (5.0 mm/s to 7.5).
The fracture strain is practically the same at all extrusion speeds. The CYS decreases 2.0 MPa
(2.0 mm/s to 5.0 mm/s), then it is maintained from 5.0 mm/s to 7.5 mm/s. A similar trend is
observed in the UCS; it decreases 16.0 MPa (2.0 mm/s to 5.0 mm/s), then is maintained from
5.0 mm/s to 7.5 mm/s. The fracture strain increases 0.04 from 2.0 mm/s to 5.0 mm/s and
preserved at the highest extrusion speed.
45
Figure 4.16. Mechanical properties of ZX10 alloy at different extrusion speeds and
temperature; a) 250 °C, b) 300 °C, c) 400 °C.
4.4.3 ZNd10 alloy
Figure 4.17 displays the stress strain curves in tension and in compression of the alloy ZNd10.
Figure 4.17a shows the stress-strain curves at the extrusion temperature of 250 °C. As it is
observed, the TYS shows a significant reduction from 2.0 mm/s to 5.0 mm/s of extrusion speed.
The TYS still decreases with the increase of extrusion speed at 7.5 mm/s. The UTS follows a
similar trend as the extrusion speed increases. Clearly, the fracture strain increases as the
extrusion speed increases. The higher fracture strain increment is observed when the alloy is
extruded at 5.0 mm/s. In compression, the alloy has the characteristic sigmoidal shape when is
tested. Such a characteristic is maintained at the three extrusion speeds. The increment of the
extrusion speed represented a reduction of the CYS as well as the UCS. The compression
fracture strain is increased as the extrusion speed increases.
The stress-strain curves of the alloy extruded at 300 °C are illustrated in figure 4.17b. As can
be seen from the graph, the TYS decreases as the extrusion speed increases. The highest
reduction is observed from 2.0 mm/s to 5.0 mm/s of extrusion speed and still some reduction
is observed at 7.5 mm/s. The fracture strain is increased with the increment of the extrusion
speed. The highest corresponds to 5.0 mm/s and 7.5 mm/s extrusion speeds. In compression,
the CYS also shows a reduction as the extrusion speed increases. The UCS follows a different
trend, first decreases (2.0 mm/s to 5.0 mm/s) and then increases with the further extrusion speed
increment. The fracture strain is quite comparable at all the extrusion speeds.
Figure 4.17c presents the stress-strain curves of the alloy extruded at 400 °C. A reduction in
the TYS is observed while the UTS first decreases and then increases with the increase of the
extrusion speed. It is observed that the highest fracture strain corresponds to the alloy extruded
at 7.5 mm/s. As can be seen, the increment of the extrusion speed reduces the CYS. In reference
to the UCS, it shows first a decrease and then an increase as an effect of the extrusion speed
increment. It seems that the fracture strain decreases as the extrusion speed increases.
46
Figure 4.17. Tension and compression stress-strain curves of ZNd10 alloy extruded at a) 250
°C, b) 300 °C, and c) 400 °C. Tested at room temperature.
47
Table 4.3 summarizes the mechanical properties of the ZNd10 alloy.
Table 4.3. Tension and compression properties of the ZNd10 alloy measured parallel to the
extrusion direction.
Extrusion
temperature
(°C)
Extrusion
speed
(mm/s)
TYS
(MPa)
UTS
(MPa)
Fracture
strain
CYS
(MPa)
UCS
(MPa)
Fracture
strain
250 °C
2.0
286.0±1.0
333.0±1.0
0.27±0.01
196.0±0.5
395.0±5.0
0.13±0.01
5.0
137.0±1.0
273.0±1.0
0.38±0.01
133.0±0.1
296.0±2.5
0.24±0.01
7.5
124.0±0.5
267.0±1.0
0.39±0.01
116.0±1.0
283.0±2.0
0.28±0.01
300 °C
2.0
183.0±7.0
282.0±3.0
0.28±0.03
157.0±0.5
317.0±1.0
0.23±0.01
5.0
101.0±2.0
262.0±2.0
0.40±0.02
101.0±1.0
268.0±1.0
0.23±0.01
7.5
94.5±0.5
261.0±0.5
0.37±0.01
95.0±0.1
299.0±1.0
0.22±0.01
400 °C
2.0
117.5±0.1
256.0±1.0
0.34±0.01
98.0±0.5
277.0±2.0
0.27±0.01
5.0
100.0±0.1
249.0±1.0
0.34±0.01
83.0±0.1
264.0±0.5
0.22±0.01
7.5
91.5±1.0
253.5±1.0
0.34±0.02
86.5±0.2
282.0±1.0
0.18±0.01
Figure 4.18 displays the tension and compression properties of ZNd10 alloy. Figure 4.18a
shows the mechanical properties after extrusion at 250 °C. It can be observed that the TYS
reduces by 149.0 MPa (2.0 mm/s to 5.0 mm/s). With the further increase of extrusion speed,
the TYS decreases by 13.0 MPa (5.0 to 7.5 mm/s). The UTS also decreases; 60.0 MPa (2.0
mm/s to 5.0 mm/s) and 6.0 MPa (5.0 mm/s to 7.5 mm/s). The fracture strain increases, first
0.11 (2.0 mm/s to 5.0 mm/s) and then 0.01 (5.0 mm/s to 7.5 mm/s). The CYS decreases 63.0
MPa (2.0 mm/s to 5.0 mm/s) and then 17.0 MPa (5.0 mm/s to 7.5 mm/s). In reference to the
UCS, it decreases 99. MPa (2.0 mm/s to 5.0 mm/s) and 13.0 MPa (5.0 mm/s to 7.5 mm/s). The
fracture strain increases 0.11 (2.0 mm/s to 5.0 mm/s) and then 0.04 (5.0 mm/s to 7.5 mm/s).
Figure 4.18b shows that at the extrusion temperature of 300 °C, the TYS reduces 82.0 MPa
(2.0 mm/s to 5.0 mm/s) and 6.5 MPa (5.0 mm/s to 7.5 mm/s). As can be seen, the UTS decreases
20.0 MPa (2.0 mm/s to 5.0 mm/s) and only 1.0 MPa (5.0 mm/s to 7.5 mm/s). The fracture strain
increases 0.12 (2.0 mm/s to 5.0 mm/s) and decreases 0.03 (5.0 mm/s to 7.5 mm/s). On the other
hand, the CYS decreases 56.0 MPa (2.0 mm/s to 5.0 mm/s) and 6.0 MPa (5.0 mm/s to 7.5
mm/s). The UCS decreases 49.0 MPa (2.0 mm/s to 5.0 mm/s) but increases 31.0 MPa (5.0
mm/s to 7.5 mm/s). At the highest extrusion speed, the fracture strain decreases 0.01.
Figure 4.18c exhibits the mechanical properties at extrusion temperature of 400 °C. Clearly,
the TYS reduces 17.5 MPa (2.0 mm/s to 5.0 mm/s), then 8.5 MPa (5.0 mm/s to 7.5). The UTS
decreases 7.0 MPa (2.0 mm/s to 5.0 mm/s) but increases 4.5 MPa (5.0 mm/s to 7.5). The
fracture strain is practically the same at all extrusion speeds. The CYS decreases 15.0 MPa (2.0
mm/s to 5.0 mm/s), then increases 3.5 MPa from 5.0 mm/s to 7.5 mm/s. As can be seen, the
UCS decreases 13.0 MPa (2.0 mm/s to 5.0 mm/s), then is increases 18.0 MPa from 5.0 mm/s
to 7.5 mm/s. The fracture strain decreases 0.05 from 2.0 mm/s to 5.0 mm/s and 0.04 from 5.0
mm/s to 7.5 mm/s.
48
Figure 4.18. Mechanical properties of ZNd10 alloy at different extrusion speeds and
temperature; a) 250 °C, b) 300 °C and c) 400 °C.
4.5 Yield asymmetry on the extruded bars
In this section, the tension-compression yield asymmetry of the extruded bars is analyzed.
Such yield asymmetry is presented as a function of the extrusion speed, see figure 4.19. The
extrusion temperature is constant.
Figure 4.19a displays the yield asymmetry in the Z1 alloy. The alloy shows the well know yield
asymmetry where the TYS is significantly higher compared to the CYS. At the extrusion
temperature of 250 °C (black dotted line) and 300 °C (blue dotted line), the yield asymmetry
shows a similar trend. The alloy displays an increment of the yield asymmetry as the extrusion
speed increases. At extrusion temperature of 400 °C (gray dotted line), the modification of the
extrusion speed shows an increment of the yield asymmetry (2.0 mm/s to 5.0 mm/s). Then, at
7.5 mm/s, it reduces.
Figure 4.19b shows the yield asymmetry in the ZX10 alloy. From the graph, it can be seen that
at the extrusion temperature of 250 °C (black dotted line) and 300 °C (blue dotted line) the
yield asymmetry decreases. Clearly, the highest yield asymmetry is observed at the extrusion
temperature of 400 °C (gray dotted line). At this extrusion temperature, the yield asymmetry is
quite comparable at all extrusion speeds.
Figure 4.19c provides the yield asymmetry in the ZNd10 alloy. At 250 °C of extrusion
temperature, the yield asymmetry reduces when the material is extruded from 2.0 mm/s to 5.0
mm/s (black dotted line). There is observed a slight increment with the further increase of
extrusion speed. A similar tendency is observed when the material is extruded at 300 °C (blue
dotted line). The yield asymmetry is quite similar at the extrusion speed of 2.0 mm/s and 5.0
mm/s, while the lowest is observed at 7.5 mm/s. At the extrusion temperature of 400 °C (gray
dotted line), the yield asymmetry is quite similar at the extrusion speed of 2.0 mm/s and 5.0
mm/s, while the lowest is observed at 7.5 mm/s.
49
Figure 4.19. Variation of the tension-compression yield asymmetry as a function of extrusion
speed for a) Z1 alloy, b) ZX10 alloy, and c) ZNd10 alloy. The extrusion temperature is
constant; 250 °C (black dotted line), 300 °C (blue dotted line) and 400 °C (gray dotted line).
50
5. Discussion
5.1 The importance of the alloy composition on the extrusion processing
The alloy composition plays an important role in the flow metal during the extrusion
processing. The force resulted during processing determines the characteristics of the resulting
profiles.
The peak force reached of each alloy is quite different. Taken from the diagram of extrusion
force vs ram displacement (figures 4.1 to 4.3 of section results); the figure 5.1 displays the
progressive evolution of peak force during processing of each alloy. Figure 5.1a displays the
peak force of the alloy Z1. The peak force follows a similar trend at all constant extrusion
temperature. It increases slightly as the extrusion speed increases. The lowest peak forces on
the Z1 alloy are at the extrusion temperature of 400 °C (gray dotted line).
Figure 5.1b exhibits the peak force in the ZX10 alloy. The drop in the peak force of the ZX10
alloy at the extrusion temperature of 250 °C can be correlated with the activation of a softening
mechanism (black dotted line). At the extrusion temperature of 300 °C (blue dotted line), the
peak force is maintained constant. However, at 400 °C (gray dotted line) a reversed effect
occurs, the peak force increases with the increment of extrusion speed.
Figure 5.1c shows the peak force in the ZNd10 alloy. In general, in the Nd containing alloy the
peak force is quite constant at all the extrusion temperature modifying the extrusion speed. The
lowest peak forces are observed at the extrusion temperature of 400 °C (gray dotted line).
In summary, the highest peak force reached varies in each alloy, as can be seen in figure 5.1.
It is quite clear that the maximum peak force achieved during processing is higher in the alloys
with Ca and Nd additions. In contrast, the Z1 alloy, which is the base alloy, i.e. with the lower
alloying concentration, is softer during extrusion. It has been reported that the increment of the
alloying content increases the hot working flow stress and thus the extrusion load [1, 109]. In
this regard, the findings in this work are consistent with results previously reported.
51
Figure 5.1. Variation of peak force reached during extrusion processing as a function of
extrusion speed for a) Z1 alloy, b) ZX10 alloy, and c) ZNd10 alloy. In each plot, the black
dotted line represents the temperature of 250 °C; the blue dotted line denotes the temperature
300 °C while the gray dotted line stands for the temperature of 400 °C.
The surface of the extruded bars was visually examined to determine the effect of the applied
extrusion parameters, figures 4.4 to 4.6. At some extrusion parameters in the alloys was reached
an eventual condition, in such a way that produced the visibility of some striations
perpendicular to the ED. Those striations could be considered as micro cracks. Such micro
cracks can be consistent with the hot cracking phenomenon as observed previously [62].
However, in the alloys of the present work, such a phenomenon is not so severe. Then, the
surface quality of the bars is not so affected. According to the extrusion limit diagrams, the
extrusion of the alloys in this work have been done near to the press limit (left side of figure
2.9). In the case of a material with good extrudability, it can be extruded with low extrusion
load, having a good surface finish [11]. Then, it can be considered that the alloys of this work
show a good extrudability. Through the observation of the barssurface, some lines parallel to
the extrusion direction are also perceived. The extrusion dies used could have caused the
appearance of these lines on the surface of the extruded bars.
52
5.2 Effect of extrusion conditions on the microstructure and texture evolution.
Due to the applied extrusion parameters, each alloy developed different microstructures and
crystallographic textures. Thus, in this section, the microstructure and texture evolution are
discussed as a function of the extrusion temperature or extrusion speed.
From the micrographs of the longitudinal sections after extrusion with varied extrusion speed
or temperature, it is possible to analyze the recrystallization process on the three alloys. The
recrystallization process on the binary alloy Z1 is completed at any extrusion condition. Then,
this binary alloy shows a completely recrystallized microstructure. In comparison to the Z1
alloy, single additions of Ca or Nd into the Mg-Zn based alloy have a significant effect on the
microstructure development during extrusion. The recrystallization process is delayed in the
presence of Ca or Nd. The retardation of the recrystallization process is more pronounced in
the Nd containing alloy rather than the Ca containing alloy. This is observable due to the
presence of large elongated grains, especially at the lowest extrusion temperature combined
with the lowest extrusion speed (figure 4.7 on the section results). Once the recrystallized
process is finalized, the grain growth starts. Figures 5.2 and 5.3 depicted the evolution of the
average grain size as a function of the extrusion temperature and extrusion speed, respectively.
Figure 5.2. Variation of the average grain size (d) as a function of extrusion temperature; a) Z1
alloy, b) ZX10 alloy, and c) ZNd10 alloy. The letter (m), represents the slope of each curve,
and is differentiated by black, blue and gray colors. The extrusion speed is constant.
53
Figure 5.3. Variation of the average grain size (d) as a function of extrusion speed; a) Z1 alloy,
b) ZX10 alloy, and c) ZNd10 alloy. The letter (m), represents the slope of each curve, and is
differentiated by black, blue and gray colors. The extrusion temperature is constant.
As is depicted, the grain coarsening in the three alloys is affected in a similar way (figures 5.2
and 5.3). There is an increment of the average grain size (d) as the extrusion temperature or
speed is increased. However, the grain growth in the binary Z1 alloy is more pronounced. The
grain growth kinetics decreases substantially in the Ca and Nd containing alloys compared with
the Z1 alloy. Furthermore, from the tendency displayed in the previous pictures, the grain
growth is more retarded in the presence of Nd rather than Ca. This is visible even at the highest
extrusion temperature or extrusion speed.
Regarding the evolution of the grain growth, investigations have proved that by increasing the
extrusion temperature [110, 111] or the extrusion speed [4, 31, 95] there is an increase on the
average size of dynamically recrystallized grains. The observed grain growth during the
increment of extrusion speed is associated with the deformation related heating. This can lead
to a temperature difference between the initial billet temperature and processing temperature
[112]. Furthermore, it has been suggested that the net result in temperature increase is a linear
function of the logarithmic ram speed [113]. Consequently, the deformation modes, DRX and
the subsequent grain growth are affected by the temperature, thus influencing the grain size
[112].
54
According to the findings, the grain coarsening velocity is much more pronounced when the
extrusion speed is modified rather than when the extrusion temperature is modified.
The generated extrusion crystallographic textures during the application of the different
extrusion parameters are shown in the section results, figures 4.10 to 4.12. The developed
texture in the Z1 alloy is a classical texture where the basal planes are located between the
prismatic planes. This result in such an alloy is present at any modification of extrusion speed
or temperature tested. Although there is a spread between the prismatic planes, the highest
intensities are found at the <1120 > pole. This kind of texture orients basal planes mostly
parallel to the extrusion direction and has often been reported for fully recrystallized extruded
bars of magnesium alloys such as AZ31 [114].
Different features in the texture evolution are observed in the ZX10 and ZNd10 alloys. The
intensities observed at the <1010 > component in the ZNd10 alloy, also orients the basal
planes parallel to the extrusion direction. Such a fibre texture has been associated with the non-
recrystallized fraction of the microstructure [4]. In both ternary alloys, a component is
established with intensity in the vicinity of the <1121 > pole. Such <1121 > component is
often observed in rare earth or calcium containing alloys and it is normally known as “rare
earth component” [40]. This component tilts the basal planes out of the ED. The
crystallographic texture in ZX10 and ZNd10 alloys is rather weak and the main texture
component changes. In the ZX10 alloy, an intensity seems to concentrate close to the <
1011 > pole, thus maintaining the same tilt of basal planes out of the ED compared to the <
1121 > “rare earth” component, but with a rotation up to 30° around the c-axis. Figure 5.4
displays a schematic representation of such rotation.
Figure 5.4. Schematic representation of the rotation around the c-axis.
5.3 Recrystallization on the extruded alloys
The recrystallization is a process that has a direct influence on the microstructure-texture and
therefore on the mechanical properties of the extruded products. It has been observed in the
results section that depending on the extrusion parameters it is possible to obtain a variety of
microstructures. In that regard, the alloys were extruded at very low extrusion speed to achieve
partly recrystallized microstructures to study the impact of specific fractions of the
microstructure on the crystallographic texture. In figure 5.5, the EBSD-IPF maps of the
partially recrystallized microstructure of the three alloys extruded at 300 °C and 0.1 mm/s are
depicted. The microstructure consists of large elongated, un-recrystallized grains surrounded
by newly formed recrystallized grains. In the Z1 alloy the classical texture where the basal
planes are between the <1120 > and <1010 > prismatic poles were observed.
55
On the other side, the texture of the ternary alloys was dominated by the <1010 > and <
1121 > components. A separation of the recrystallized and un-recrystallized fractions of grains
is assisted based on the consideration of the internal orientation spread of grains, the grain
orientation spread (GOS) [4, 74, 115]. It is assumed that recrystallized grains exhibit low in-
grain misorientations because of a lattice free of dislocations, which are an important source of
such misorientations. On the other hand, grains that underwent plastic deformation and
experienced active slip modes would result in higher GOS. These recrystallized and un-
recrystallized fractions were separated to observe possible hints of the origin of the observed
texture components. Arbitrarily, a GOS value of 1° as a separator is considered in this work.
This analysis revealed that the fraction of the microstructure corresponding to the recrystallized
grains (GOS<1°) in the Z1 alloy have a clear orientation tendency to develop the <1120 >
pole while the ZX10 and ZNd10 alloys show a relatively weak texture, although they preserve
their orientations mainly at the prismatic <1010 > pole (figures 5.5 a-c). According to this
analysis, the component <1010 > is associated to the un-recrystallized grains in all alloys
(GOS>1°), besides, this component is significantly more pronounced.
One interesting finding here is that in the ZX10 alloy, the rare earth texture component at the
<1121 > pole is revealed in the un-recrystallized fraction (GOS >1°) of the microstructure but
not in the fraction of recrystallized grains (GOS <1°), see Figure 5.5b. On the other side, in
ZNd10, both fractions include grains with such orientation, i.e.,<1121 >, (Figure 5.5c). It is
believed that the element segregation modifies the grain growth kinetics, which changes the
components in the texture. This is the case for the Nd containing alloy, where the component
<1121 > in the deformed condition with small DRX grains remain and preserve a certain grain
growth uniformity. Such grain growth restrictions have been also linked with the appearance
of the “rare-earth texture” in conventional AZ31 Mg alloys [116]. In AZ31 the grains with the
basal planes parallel to ED normally grow unless a mechanism is active that changes the growth
kinetics e.g., particles that restrict the grain growth. The grain growth tendency can be changed
by particles in a way, that in recrystallized microstructures of AZ31 the <1121 > pole
component can also be observed. On the contrary, in the Ca containing alloy there is not such
a strong restriction to recrystallization and grain growth.
56
Figure 5.5. IPF maps corresponding to partly recrystallized microstructure in a) Z1 alloy, b)
ZX10 alloy and c) ZNd10 alloy extruded at 0.1 mm/s and 300 °C [95].
The grain growth plays a key role in the determination of the final texture of a thermo-
mechanically processed product. Grain growth is regarded as the migration of grain boundaries
taking place during the recrystallization process [117]. Two types of recrystallization are
considered in the grain growth of a processed material, i.e. dynamic and static recrystallization.
Therefore, in the following two sections, is discussed the effect of dynamic (DRX) and static
recrystallization (SRX) processes on the texture of the alloys.
5.3.1 Effect of dynamic recrystallization on the texture
In order to differentiate the effect of different fractions and grain dependence of dynamically
recrystallized microstructures on the texture, the fractions of dynamically recrystallized grains
developed at low and intermediate extrusion speeds were evaluated. The average grain size of
dynamically recrystallized grains (drec) has been considered as a limit to determine the effect
of grain growth on the orientation development of such grains, thus separating small and large
recrystallized grains.
In figures 5.6a and 5.6b the recrystallized fraction of the microstructures of Z1 extruded at 0.1
mm/s and 0.6 mm/s, is shown respectively. The picture that emerges from the analysis at these
extrusion conditions is that in this alloy both fraction of the grains show a spread between the
<1010 > and the dominant <1120 > components. However, there is a subtle trend to strengthen
the <1120 > component as grain coarsening takes place. Figures 5.6c and 5.6e exhibit the
recrystallized fraction of microstructures and respective textures after slow extrusion (0.1
mm/s) of the ZX10 and ZNd10 alloys.
57
According to the same criterion, i.e., the consideration of average grain size, it revealed a slight
change compared to the deformation component in figure 5.6 for ZNd10 and ZX10 alloys. In
ZX10 alloy, the main orientation at the < 1010 > pole is maintained; however it seems to be
more pronounced in the smaller grains, this is observed in the texture in figure 5.6c, d <3.0 µm.
In contrast, in the d >3.0 µm, potentially those grains with a growth advantage during
recrystallization, an angular spread towards the <0001 > pole (and therefore towards the <
1011 > pole) is found. Different to the previous alloy, ZNd10 was shown not to have such a
grain size dependence. Both intensities at the < 1010 > pole, and the < 1121 > pole, remain
comparable, see Figure 5.6e. It has been observed that increasing the extrusion temperature or
speed there is an increment of the grain growth (section results); therefore, a detailed analysis
of the effect of grain coarsening on the texture is done. For that analysis, the effect of grain
coarsening as a function of increasing the extrusion speed for ZX10 (2.0 mm/s) and ZNd10
(5.0 mm/s) alloy was used which is depicted in figures 5.6d and 5.6f, respectively.
With grain coarsening by increasing the extrusion speed, the separation of grains applying the
average grain size criterion, both alloys tend to develop intensity along the arc between the <
1011 > and the <1121 > poles, which establishes a rotation of 30° around the c-axis of the
hcp structure. For ZX10, in the small grain size, there is no clear dominance between the
intensities at <1011 > and <1121 > poles, see texture in figure 5.6d, d <11.0 µm. However,
for the d > 11.0 µm (Figure 5.6d) the <1011 > pole is stronger. This indicates the importance
of grain coarsening in the formation of the <1011 > component in the ZX10 alloy. For ZNd10,
in the d < 9.8 µm (figure 5.6f), it seems that the intensity at the <1011 > pole is somewhat
higher, while with the d >9.8 µm the <1121 > pole is more pronounced. Thus, the preference
of grain orientations changes with the extrusion speed and the concomitant grain coarsening of
recrystallized grains.
In each alloy, the texture evolution due to the grain growth during the DRX recrystallization
process is different. In Z1 alloy, the grains are principally spread between the prismatic poles
<1010 > and <1120 >. The latter one is strengthened as grain growth of recrystallized grains
takes place. Such behavior is modified if Ca or Nd is added to the MgZn alloy systems. The
<1011 > becomes more dominant in the Ca containing alloy as grain growth takes place, while
in the Nd containing alloy the coarse grains are mostly oriented towards the <1121 >
component.
58
Figure 5.6. Recrystallized grains: a) Z1, c) ZX10, e) ZNd10 alloy extruded at 0.1 mm/s and b)
Z1 extruded at 0.6 mm/s, d) ZX10 extruded at 2.0 mm/s and f) ZNd10 extruded at 5.0 mm/s
using 300 °C of extrusion temperature [95].
5.3.2 Effect of static recrystallization on the texture
After extrusion, the remaining heat during cooling down of the extruded bars is affecting the
microstructure and texture, i.e. SRX is already activated but in an uncontrolled way. Such a
change can be better explained using samples with partially recrystallized microstructures
subjected to subsequent annealing. Then, to systematically determine the role of SRX on the
texture development, partly recrystallized of samples from ZX10 and ZNd10 extruded bars at
0.1 mm/s were annealed. It is important to note that in all the applied extrusion parameters on
Z1 alloy of this study, the microstructure is completely recrystallized. Therefore, the analysis
of Z1 is not presented because no changes in the texture were found after SRX because of the
high fraction of recrystallized.
On the other hand, the microstructures of samples from the ZX10 and ZNd10 alloys were
tailored so the grain structure of the SRX samples could be compared to grain structures of
DRXed samples, i.e., samples extruded at 2.0 mm/s for ZX10 alloy and 5.0 mm/s for ZNd10
alloy. The annealed microstructures of the ternary alloys revealed visible texture changes.
Figure 5.7 shows that after static recrystallization at 400 °C for 3.0 min in the case of ZX10
(figure 5.7a) and 30 min for ZNd10 alloy (figure 5.7b) the texture is remarkably different to
the counterparts dynamically recrystallized microstructures, ZX10 alloy (figure 4.11) and
ZNd10 alloy (figure 4.12) extruded at 300 °C. During SRX of the two alloys, intensity along
the arc between <1010 > and <1120 > poles is maintained. In ZX10, there are orientations
with tilt out of the <1010 > pole, e.g., to the <2021 > pole, which also leaves basal planes
with a tilt out of the extrusion direction, figure 5.7a.
59
A stronger alignment of basal planes along the arc between <1010 > and <1120 > poles after
static recrystallization is also visible on the ZNd10 alloy, figure 5.7b. Besides, there are
orientations with varied tilt angle (e.g., at <2023 > pole).
The static recrystallization weakens the texture of the partly recrystallized samples but in
essence, the grains maintain their orientation in the surrounding of the <1010 > pole in Ca
and along the arc between the <1010 > - <1120 > poles in the Nd containing alloy. This
means that they inherit the orientation of the dominant deformation texture component with
the concomitant stronger recrystallization texture compared to the DRXed textures. At the end
of static recrystallization, the basal planes remain aligned parallel to the ED. Although the <
1121 > RE component is present in the partly recrystallized microstructure (initial condition,
figures 5.5b and 5.5c), in both alloys such a component is suppressed after SRX.
These results reflect those of Imandoust et al. [118] who also found that the RE grains disappear
after the SRX process of an Mg-Zn-Al-Mn-Y alloy. It was explained that such behavior could
be attributed to an isotropic grain growth process due to a grain boundary co-segregation effect
that decreases the anisotropy in grain boundary energy and mobility. It is hypothesized in the
context of the modification of extrusion speed that the change of the balance of active
deformation mechanisms as the temperature is increased as the extrusion speed increases. The
importance of certain deformation mechanisms along with rapid coarsening of recrystallized
grains could also influence the dominance of the RE-texture component.
Figure 5.7. Microstructure and texture after static recrystallization (SRX) at 400 °C; a) ZX10
alloy for 3.0 min and b) ZNd10 alloy for 30 min. The inset (d) represents the average grain size
[95].
5.4 Correlation between microstructure and mechanical properties
The mechanical properties of magnesium alloys, such as yield strength, yield asymmetry and
strain to fracture are strongly influenced by grain size and texture. It is well known that the
strength increases based on the grain refinement.
60
In that sense, the dependence of the yield strength with respect to the grain size is principally
governed by the Hall-Petch relationship:
𝜎𝜎𝑦𝑦=𝜎𝜎0 + 𝑘𝑘𝑦𝑦 𝑑𝑑−1/2 (Eq. 9)
Where σy corresponds to the yield stress, σ0 is the friction stress when the dislocations move
on the slip plane, d refers to the average grain size and ky is the stress concentration factor. The
micro mechanism could be understood in terms of pile-up of dislocations in the vicinity of the
grain boundary [119].
In this section, the effect of grain size on the yield stress on tension and compression is
analyzed. The considered grain size variation is from the extruded materials at constant
temperature with different extrusion speed. Considering the previous extrusion parameters, the
average grain size (d) varies in the range from 21 to 51 µm for the Z1 alloy, while in the ZX10
alloy; (d) has a range from 9.5 to 35 µm and from 3.5 to 26.5 µm for the ZNd10 alloy. Figures
5.8 to 5.10 and table 5.1 displays the dependence of the CYS and TYS on the grain size. It was
evaluated according to the Hall-Petch relationship. The results are shown separately for each
alloy. The slope of the curves in each plot refers to the k value, which represents the grain
boundary as an obstacle to slip across the grain boundaries.
In the present analysis of the Z1 alloy (figure 5.8), the slope of the curves (k value) of the
material extruded at the different extrusion temperature and tested in compression is slightly
bigger than in tension, see table 5.1a. This is observed in the alloy extruded at 250 and 300 °C,
while at 400 °C there is an inverse relation. The bigger k value in compression rather than in
tension agrees with previous studies [2, 120, 121]; this can be considered as an indicator that
during deformation, the tensile twinning is more sensitive to grain refinement than
crystallographic slip.
On the other hand, the ZX10 alloy (figure 5.9 and table 5.1b) and ZNd10 alloy (figure 5.10 and
table 5.1c) alloys behave in a different way. The slope of the Hall-Petch plots, i.e., the k value,
is greater in tension than in compression. This observed in all the analyzed cases. Such behavior
can be attributed to the texture effect. As it is observed in the section of crystallographic texture
evolution, these alloys developed textures where the main components are tilted out of the
extrusion direction. Then, it can be suggested that in these two alloys the k value is not only
affected by the grain size but also by the crystallographic texture. Here, there is observed a
certain effect in the k values where such effect can be attributed to the crystallographic texture
that modifies the balance of deformation modes. It is considered in [122] that when the texture
development due to the mechanical processing makes the activation of basal slip during
deformation easier; the k value would be reduced.
61
Figure 5.8. Variation of the tensile and compressive yield strength against d -1/2 for the Z1 alloy
extruded at a) 250 °C, b) 300 °C and c) 400 °C.
Figure 5.9. Variation of the tensile and compressive yield strength against d -1/2 for the ZX10
alloy extruded at a) 250 °C, b) 300 °C and c) 400 °C.
62
Figure 5.10. Variation of the tensile and compressive yield strength against d -1/2 for the ZNd10
alloy extruded at a) 250 °C, b) 300 °C and c) 400 °C.
Table 5.1. Macroscopic Hall-Petch parameters for a) Z1 alloy, b) ZX10 alloy, c) ZNd10 alloy
tested under tension and compression.
Extrusion
temperature (°C)
a) Z1 alloy
b) ZX10 alloy
c) ZNd10 alloy
k (MPa µm-1/2)
σ
0
k (MPa µm-1/2)
σ
0
k (MPa µm-1/2)
σ
0
250 °C
TYS
245
88
415
-12
740
-110
CYS
297
20
393
-8
347
11
300 °C
TYS
159
116
213
42
408
-10
CYS
261
31
183
48
283
23
400 °C
TYS
320
92
94
121
344
26
CYS
194
35
76
74
185
48
5.5 Plastic deformation of the extruded alloys
As was previously shown, the addition of Ca or Nd modifies the microstructure and
crystallographic texture and subsequently the mechanical properties of the Mg-Zn alloys.
Therefore, in the following section, is analyzed the role of Ca and Nd additions on the activation
of deformation modes in the Mg-Zn alloys and, furthermore, the development of deformation
texture. To do so, experimental and simulation results are analyzed at different deformation
strains. In figure 5.11 the optical micrographs and the crystallographic textures of the
considered extruded bars are shown. The stress-strain curves obtained at room temperature in
tension and in compression have been simulated using the VPSC model. During the simulation,
the basal <a>, prismatic <a>, pyramidal <a>, pyramidal I <c+a> and pyramidal II <c+a> as
63
well as tensile twinning 1012<101 1 > and compressive twinning 1011<101 2 > were
considered. The results from the VPSC simulations (crystallographic texture and hardening
behavior) are then compared with experimental results.
As observed, materials with similar grain structure but different texture were employed. The
average grain size (d) of each material for this study is observed in figure 5.11. A typical basal
texture in Z1 alloy is observed, figure 5.11a. On the ZX10 alloy, the texture is dominated by
the <1011 > pole, which is tilted out of the extrusion direction (figure 5.11b). The <1121 >
pole, i.e., the rare earth texture component, dominates the texture in the ZNd10 alloy (figure
5.11c). Since the average grain size is comparable; then, its effect on the deformation behavior
can be discarded. Therefore, the observed deformation behavior is entirely considered as an
effect of the crystallographic texture. The different crystallographic texture in each alloy allows
studying the activation of different deformation modes due the modification of its CRSS owing
to the addition of alloying elements (Ca or Nd).
Figure 5.11. Optical micrographs and crystallographic texture of the a) Z1 alloy extruded at
300 °C and 2.0 mm/s, b) ZX10 alloy extruded at 300 and 5.0 mm/s and c) ZNd10 alloy extruded
at 400 °C and 5.0 mm/s.
5.5.1 Microstructure evolution during tension and compression
In the following section, the evolution of the microstructure at the different analyzed
deformation including the IPF maps together with and the IPF textures and the Image Quality
(IQ) maps are shown. The IPF maps (colorful figures) show the microstructure evolution, while
the IQ maps (gray figures) display the twinning structures formed during the deformation. At
the bottom section of figures 5.12, 5.13 and 5.14 is displayed an inserted table, which shows
the cumulative fraction of twinning.
64
During the deformation, in tension or compression, there is usually observed a characteristic
grain rotation which is directly correlated with the activation of different deformation
mechanisms. Then, applying the technique of EBSD the microstructure was analized. The
microstrucutre does not experience a significant change during the tension of the Z1 alloy, see
IPF maps of figure 5.12a. As the deformation proceeds from 0 to 0.12, the microstructure grain
orientation is quite comparable. The IQ maps of this Z1 alloy under tension show that few
twins nucleate at strain of 0.03, and the twin fraction is still low at larger tension strains. With
the increment of strain, there is observed an increment of cumulative fraction of twinning (see
inserted table at the bottom of figure 5.12a and b).
On the other hand, in compression the observed grain rotation (IPF maps in figure 5.12b) is
attribuited to the high twining activation. This is confirmed in the IQ maps (figure 5.12b). It is
clear that in compression the fraction twinning is much more pronounced than in tension. The
major fraction of twinning in compression rather than in tension is due to the basal texture
which is formed during the extrusion process.
Figure 5.12. IPF maps (colorful figures) and IQ maps (gray figures) of Z1 alloy deformed in a)
tension and b) compression. The twins are labeled in the IQ maps. The twinning fraction at
each deformed microstructure is differentiated by red, blue and yellow colors. The black arrow
indicates the load direction (LD).
In figure 5.13a the deformed microstructure of ZX10 alloy tested in tension is shown. The
captured microstructure at each deformation stage shows an increment of twinning fraction.
65
In tension, traces of compresion and double twins are observed at 0.12. During the deformation
in compression, twins first apear at 0.03 of strain and such twinning structures are also observed
with the further deformation at 0.12 of strain, figure 5.13b. Altough the major fraction
corresponds to the tensile twinning, the compression and double twins are also observed at
such deformation stage.
Figure 5.13. IPF maps (colorful figures) and IQ maps (gray figures) of ZX10 alloy deformed
in a) tension and b) compression. The twins are labeled in the IQ maps. The twinning fraction
at each deformed microstructure is differentiated by red, blue and yellow colors. The black
arrow indicates the load direction (LD).
The twinning evolution in the ZNd10 alloy in tension and compression is observed in figure
5.14. In tension, the activation of twinning is clearly observed at 0.03, figure 5.14a. From the
activated fraction, the most activated corresponds to the tensile twin. According to the EBSD
analysis, the cumulative fraction increases at 0.12 of deformation. On the other hand, when the
alloy is tested in compression, the activation of twinning is observed since a low deformation,
i.e., 0.03 (figure 5.14b). With the increment of strain up to 0.12, an increment is observed
according to the cumulative fraction. A very low fraction of compression and double twins are
also observed.
In the preceding figures, it is observed that the twinning mechanism is activated in the three
alloys either in tension or in compression. However, the twinning activity is more pronounced
in compression than in tension.
66
Moreover, from the considered three types of twinning, i.e., tensile twin, compression twin and
double twin, the most activated consists on the tensile twinning.
Figures 5.12, 5.13 and 5.14 display the twin evolution in the three alloys as deformation
proceeds. Fewer twins were seemingly detected at high deformation strains compared with low
deformation strains (IQ maps). This could be explained due to the coalescence of twins or the
vanishing of twin boundaries. As the deformation proceeds, there is a merging of pre-existing
extension twins or the disappearance of twin boundaries [123, 124].
Figure 5.14. IPF maps (colorful figures) and IQ maps (gray figures) of ZNd10 alloy deformed
in a) tension and b) compression. The twins are labeled in the IQ maps. The twinning fraction
at each deformed microstructure is labeled by red, blue and yellow colors. The black arrow
indicates the load direction (LD).
As with the previously observed evolution of the microstructure, an evolution of the
crystallographic texture is also observed. Such texture evolution is shown in the subsequent
figures.
Tested in tension, the Z1 alloy does not show a significant change in the crystallographic
texture, this can be observed in figure 5.15a. Such a texture maintains its orientations located
between <1120 > and <1010 > poles. Although there is a slight activation of twinning, that
deformation mechanism did not produce big changes even at the deformation of 0.12.
67
Nevertheless, there is a slight tendency of some of the deformed grains to show orientations
towards the <1010 > pole. M.T Perez Prado et al. [125] also observed the alignment of the
crystal orientations towards the <1010 > pole during the tension of an AZ31 alloy.
On the other hand, during the uniaxial compression there is a major effect on the texture
evolution, figure 5.15b. Here, a rotation of the HCP structure towards the <0001 > pole
occurs. With the further deformation at 0.08, the basal planes are accommodated perpendicular
to the direction of the applied force. This is because of the high activation of tensile twin since
0.03 of deformation (figure 5.12b). Amit Pandey et al. [126] reports similar findings in an
AZ31 alloy tested in compression.
Thus, the observed difference in the crystallographic texture evolution of the Z1 alloy in this
work, tested in tension or compression, is correlated to the easy activation of twinning in
compression rather than in tension.
Figure 5.15. Inverse pole figures showing the crystallographic texture evolution in Z1 alloy
tested in a) tension and b) compression. The texture intensity in the inverse pole figures is in
the unit of m.r.d.
Twinning is also activated in the ZX10 alloy tested in tension, (as was displayed in figure
5.13a). Compared with the Z1 alloy, the fraction of twinning is slightly higher in the ZX10
alloy. Although there is observed the activation of twinning, it did not produce a marked effect
on the crystallographic texture at low deformation of 0.03 (figure 5.16a). However, as the
deformation increases to 0.12, there is a rotation towards the <1010 > pole (figure 5.16a).
Then, there is a high tendency of the basal planes to be aligned parallel to the direction of the
applied load. In compression, the activation of twinning is highly activated since 0.03 of
deformation; however, the texture is quite comparable to the one without deformation, see
68
figure 5.16b. The compression up to 0.12 changes the orientation of the grains and most of
them are aligned perpendicular to the applied load.
Figure 5.16. Inverse pole figures showing the crystallographic texture evolution in ZX10 alloy
tested in a) tension and b) compression. The texture intensity in the inverse pole figures is in
the unit of m.r.d.
Although at low fraction, during the deformation of ZNd10 alloy in tension, the twining
mechanism is activated, (as was depicted in figure 5.14a). Nevertheless, the texture at 0.03 is
similar to the initial one. Such twinning activation is not enough to modify the texture, i.e., the
main texture component <1121 > remains at 0.03, figure 5.17a. The increment of applied load
up to 0.12 produces a more visible texture modification. As in the ZX10, in the ZNd10 alloy,
the grains are mainly orientated at the <1010 > pole. This can be correlated with the observed
twinning activity at 0.12 of strain. Changing the direction of the applied load, i.e., in
compression, although there is a high twinning activity, the main texture component <1121 >
remains at 0.03, see figure 5.17b. The texture is highly modified at a strain of 0.12. There is a
rotation towards the basal <0001 > pole.
In summary, tested in tension, both ternary alloys (ZX10 and ZNd10) have a similar behavior
with the tendency to develop the prismatic <1010 > pole as the deformation proceeds. This
also is in good agreement with earlier observations [127, 128], which showed the strengthening
of <1010 > pole during the deformation under tension. This differs from the Z1 alloy where
such a component is not clearly developed. In compression, the three alloys show a comparable
crystallographic texture development, there is a trend for the basal <0001 > pole to be formed.
In accordance with the present results, previous studies [123, 126, 128] have also shown the
rotation towards the <0001 > pole.
69
Figure 5.17. Inverse pole figures showing the crystallographic texture evolution in ZNd10 alloy
tested in a) tension and b) compression. The texture intensity in the inverse pole figures is in
the unit of m.r.d.
5.5.2 Experimental and VPSC modelling
In this section, experimental and simulated results are analyzed. The use of the VPSC model
gives the possibility to associate the plastic behavior and the texture evolution during the
deformation. Using the initial crystallographic texture of each alloy and applying the CRSS of
each slip system in combination with the Voce parameters shown in tables 5.2 to 5.4 it is
possible to simulate the hardening behavior in tension and compression as well as the
corresponding crystallographic texture evolution. The experimental results and the VPSC
simulations are compared in the following figures. The relative activities obtained during the
simulation are also shown.
Table 5.2. CRSSs and strain hardening parameters of slip and twin systems considered in the
simulation of Z1 alloy under tension and compression loading.
Slip mode
CRSS
τsat
θ0
θ1
Basal <a>
5.00
1.75
50.00
6.23
Prismatic <a>
13.05
5.11
200.00
6.53
Pyramidal <a>
75.22
47.28
495.65
0.43
Pyramidal I <c+a>
67.07
39.13
353.26
0.35
Pyramidal II <c+a>
33.04
39.24
3.37
0.05
Tensile twin 1012<101 1>
5.43
0.00
0.00
0.00
Compressive twin 1011<101 2 >
115.0
0.00
0.00
0.00
70
Table 5.3. CRSSs and strain hardening parameters of slip and twin systems considered in the
simulation of ZX10 alloy under tension and compression loading.
Slip mode
CRSS
τsat
θ0
θ1
Basal <a>
5.00
6.85
12.28
50.10
Prismatic <a>
13.37
4.46
10.33
3.59
Pyramidal <a>
54.35
64.13
3.72
1.68
Pyramidal I <c+a>
22.28
53.26
2.61
0.22
Pyramidal II <c+a>
22.61
53.70
2.93
0.43
Tensile twin 1012<101 1>
11.40
0.00
0.00
0.00
Compressive twin 1011<101 2 >
112.0
0.00
0.00
0.00
Table 5.4. CRSSs and strain hardening parameters of slip and twin systems considered in the
simulation of ZNd10 alloy under tension and compression loading.
Slip mode
CRSS
τsat
θ0
θ1
Basal <a>
5.00
9.78
14.30
55.87
Prismatic <a>
11.96
4.89
10.76
3.80
Pyramidal <a>
47.17
65.22
3.39
1.36
Pyramidal I <c+a>
21.20
52.72
2.35
0.23
Pyramidal II <c+a>
22.28
53.26
2.72
0.45
Tensile twin 1012<101 1>
12.59
0.00
0.00
0.00
Compressive twin 1011<101 2 >
109.0
0.00
0.00
0.00
The experiments and simulations carried out give an understanding of the active deformation
modes that take place during tension and compression. According to the VPSC simulation in
tension, at the onset of deformation of the Z1 alloy (figure 5.18c), the deformation is dominated
by basal slip up to roughly 0.027 of strain. As the deformation proceeds, basal <a> slip
decreases and the prismatic <a> slip begins to dominate the deformation up to 0.12. Tensile
twinning is also activated right at the beginning. This is also observed in the IQ map in figure
5.12a. The findings regarding the relative activities in tension of this Z1 alloy agree with the
reported results by [129] in which a rolled sheet of Z1 was tested in tension along the rolling
direction.
In compression, the tensile twinning mode dominates at the onset of the deformation; figure
5.18d. In this case, the twinning takes much more relevance during this kind of deformation;
this is confirmed with the observed on the IPF map (figure 5.12b). At higher strains, the
twinning activity decreases and basal <a> slip becomes dominant. In reference to the non-basal
deformation modes, there is a low activation of the prismatic <a> slip and pyramidal II <c+a>
at the end of the deformation. In a extruded AZ31 alloy, with a basal texture comparable to the
observed in the Z1 alloy, Ebeling et al. [105] reported a similar behavior with respect to tensile
twining and basal slip like the one reported in this work. However, the observed activity of
pyramidal mode in this work, contrasts with the observed inactivity reported by Ebeling’s
work.
71
Figure 5.18. Experimental and simulated uniaxial a) tension and b) compression curves for Z1
alloy. Estimated relative activities of the seven deformation modes considered during c) tension
and d) compression simulations.
The activity of deformation modes observed in the ZX10 alloy differs from tension to
compression. For this alloy, both, basal <a> and prismatic <a> slip dominate the deformation
in tension (figure 5.19c). A slight tensile twining activity is also observed. This agrees with the
low twinning fraction determined by the EBSD analysis (figure 5.13a). In this ZX10 alloy, the
lower twinning mode activity could be attributed to the non-basal texture that promotes more
activation of non-basal modes. The simulation findings shown in this work demonstrate the
activation of non-basal slip, e.g., prismatic <a> slip even at room temperature, this has been
also reported previously [105, 129, 130]. Furthermore, the simulation in this work suggests the
activity of pyramidal modes, where there is a higher activity of pyramidal <a> if compare with
pyramidal <c+a> slip. Such deformation mode, i.e., pyramidal <a> is observed since a strain
of 0.095 and slightly increases at the end of deformation. However, the pyramidal <a> slip
activity in this work is lower than the reported for a binary Mg-Ca alloy [131].
On the other hand, in compression, the deformation is practically dominated by basal <a> slip,
see figure 5.19d. In this case, the tensile twinning is activated since the beginning of
deformation as well as its activation is pronounced. This is also observed on the EBSD analysis
(figure 5.13b). The activation of prismatic slip is quite low, and it is slightly visible at the
highest deformation. It was found that all the considered pyramidal modes are similarly
activated. This was found particularly at high deformation. The inactivity of prismatic <a> slip
at the onset of deformation in compression has been also observed in [105]. However, a
difference arises in the present work, where such a deformation mode is activated especially at
the end of deformation.
72
Figure 5.19. Experimental and simulated uniaxial a) tension and b) compression curves for
ZX10 alloy. Estimated relative activities of the seven deformation modes considered during c)
tension and d) compression simulations.
As in the previous alloy, in the ZNd10 alloy, the deformation under tension is dominated by
basal <a> and prismatic <a> slip modes; see figure 5.20c. Although the basal <a> deformation
mode is highly important at the start of deformation, it becomes less activated at around 0.080
of strain since the prismatic <a> mode starts to dominate the deformation. In this case, the
considered tensile twinning is slightly more activated than in the ZX10 alloy. The pyramidal
<a> slip is activated. Its activation is slightly higher than in the previous ZX10 alloy. The
activation of basal <a> and prismatic <a> deformation modes in this study corroborates earlier
findings published by Zhou et al. [129] for a similar alloy. As well in [129], the pyramidal II
<c+a> mode is proposed as the one involved in the deformation along the c-axis, nevertheless
in the present work, it is suggested that there is a higher activation of pyramidal <a> than of
pyramidal II <c+a> slip. A higher activity of pyramidal <a> in comparison to pyramidal <c+a>
mode was also found by Zeng et al. [131] for a binary Mg-Nd alloy.
In compression, the activity of the different deformation modes is quite different, especially
the non-basal modes and the tensile twinning, figure 5.20d. Firstly, the basal <a> slip is
activated from the start of deformation, which decreases as the deformation proceeds. The
activation of tensile twinning is more pronounced in compression rather than in tension. Based
on the relative activity of deformation modes, figure 5.20d, there is a clear activation of
pyramidal <a> slip from roughly 0.070 of strain. At a strain of 0.12, the activation of pyramidal
I and pyramidal II slip is quite comparable. In relation to the compression of a Mg-Li-Al alloy,
Lentz et al. [132] suggest a deformation dominated by prismatic <a> slip followed by tensile
twinning and to a smaller extent by basal <a> slip, this contrasts with the finding reported here
(figure 5.20d). This discrepancy can be correlated with the non-basal texture observed this Mg-
Zn-Nd alloy.
73
Figure 5.20. Experimental and simulated uniaxial a) tension and b) compression curves for
ZNd10 alloy. Estimated relative activities of the seven deformation modes considered during
c) tension and d) compression simulations.
5.5.3 Effect of Ca and Nd additions on the CRSS ratio of deformation modes
Figure 5.21 displays the ratio between the CRSS of each deformation mode considering the
VPSC simulations. The addition of Ca or Nd into the Mg-Zn systems modifies the relation
between the critical resolved shear stresses of the slip modes. Based on the VPSC simulation,
the ratio between the CRSSprismatic<a> slip and the CRSSbasal<a> slip is 2.61 in the Z1 alloy.
Regarding to the ZX10 alloy, such ratio corresponds to 2.68. In reference to the ZNd10 alloy,
the ratio between the CRSSprismatic<a> slip and the CRSSbasal<a> slip is 2.40. The previous
observed CRSS ratios show the relevance of the alloying elements on magnesium, where these
values are lower than the reported for pure magnesium [133, 134].
The balance among the basal <a> and pyramidal modes is also modified. Then, the ratio
between CRSSpyramidal <a> and CRSSbasal<a> slip is 15.04, 10.90 and 9.43 for the Z1, ZX10 and
ZNd10 alloy, respectively. The estimated CRSSpyramidal <a>/CRSSbasal<a> for the alloys in this
work contrasts with Zeng et al. [131], where it was estimated a CRSSpyramidal <a>/CRSSbasal<a> of
18.0 for a AZ31 alloy, while 4.33 in an Mg-Ca alloy and 3.75 for a Mg-Nd alloy. Moreover, in
the present work the ratio between CRSSpyramidal I <c+a> and CRSSbasal<a> slip is evaluated in
13.41, 4.46, 4.24 for the Z1, ZX10 and ZNd10 alloy, respectively. A CRSSpyramidal I<c+a>
/CRSSbasal<a> of 24.0, 5.33 and 4.75 were estimated for the AZ31, Mg-Ca and Mg-Nd alloys,
respectively in [131].
The CRSSpyramial II <c+a> slip /CRSSbasal<a> slip corresponds to 6.61, 4.52 and 4.46 for Z1, ZX10
and ZNd10 alloy, respectively. For the binary Z1 alloy, the CRSS ratio is slightly higher in
comparison with the estimated from the work by Zhou et al. [129]. In the specific case of the
Mg-Zn-Ca alloy, the CRSS ratio between the CRSSpyramdial II <c+a> and the CRSSbasal<a> has not
been assessed. The alloy that can be comparable corresponds to the Mg-Zn-Nd. Perez-Prado et
al. [135] determined for a Mg-Zn-Nd alloy, a smaller CRSS ratio when compared with the
74
reported in this work. Furthermore, from the considered deformation twinning, the CRSStensile-
twinning is the most affected with respect to the CRSSbasal<a>. Then, the CRSS tensile-twinning/CRSS
basal<a> in the Z1 alloy corresponds to 1.10, such ratio is changed to 2.28 in the ZX10 alloy and
to 2.52 in the ZNd10 alloy. The findings showed a similar trend in comparison with [131].
In figure 5.21 a summary of the ratio between the CRSS of non-basal slip and the CRSS of the
extension twinning both with respect to CRSS of the basal is shown. The
CRSSprismatic/CRSSbasal<a> increases with the addition of Ca into the Mg-Zn system but
decreases with the addition of Nd. In the case of the pyramidal modes, all of them follow a
similar behavior. Then, the CRSSpyramidal <a>, I and II <c+a>/CRSSbasal<a> reduces with the addition
of Ca and is even lower with the addition of Nd. Regarding the CRSSextension-
twinning/CRSSbasal<a>, it follows an opposite trend, that means, the addition of Ca to the Z1 alloy
increases the ratio and with the addition of Nd, the ratio is further increased.
Figure 5.21. Ratio between the CRSSnon-basal slip and CRSSbasal<a> slip as well as the ratio
between the CRSSextension twinnig and CRSSbasal<a> slip based of the VPSC simulations.
5.5.4 Crystallographic texture evolution: Experimental and simulated
Figure 5.22 presents the experimental and simulated inverse pole figures for the Z1 alloy during
the deformation under tension and compression. The compared experimental and simulated
crystallographic texture are in good agreement either in tension or in compression (see figure
5.22). The simulation results in tension presented particularly the dominance of two
deformation modes, that is, the decreasing activity of basal <a> slip combined with the
increasing activity of prismatic <a> slip as the strain proceeds. The activity of these two
deformation modes generates a suggested rotation around the c-axis of the HCP structure,
which aligns the basal planes between the <1120 > and <1010 > pole with a high intensity
in the surroundings of the <1010 > pole, see figure 5.22a. In a previous study, evaluating the
texture evolution during plastic deformation by simulation in tension of AZ31 alloy [105]
reported consistent results with the findings in this work. In addition, the higher texture
intensity in comparison with the simulated texture has been also reported in Z1 alloy in [129].
On the other hand, in compression, most of the c-axes align perpendicular to the applied load.
Such alignment favors the formation of basal <0001 > pole, figure 5.22b.
75
This texture component can be correlated with the dominance of basal <a> slip and tensile
twinning, where the latter is highly activated since the onset of deformation. The extension
twinning is an important deformation mechanism that leads to fast and quite clear texture
developments. This finding broadly agrees the simulation work for an extruded AZ31 alloy
[105], linking the activity, principally of extension twin and basal <a> deformation modes in
compression with the texture development.
Figure 5.22. Inverse pole figures comparing the experimental and simulated crystallographic
texture of Z1 alloy tested under a) tension and b) compression. The texture intensity is in the
unit of m.r.d.
76
Figure 5.23 shows the crystallographic texture evolution of ZX10 observed experimentally and
in simulation. Both, the experimental and simulated textures are in good agreement. As in the
Z1 alloy, the basal <a> and prismatic <a> slip dominate the deformation when the alloy is
tested in tension. However, the initial texture observed in the ZX10 alloy promotes the
dominance of the basal <a> slip until a higher deformation strain (at roughly 0.10) compared
with the Z1 alloy. Nevertheless, the prismatic <a> slip dominates the deformation particularly
at the end of the deformation.
Additionally, the pyramidal <a> mode, it was slightly activated for this ZX10 alloy at the final
strain stage. Therefore, in tension, the rotation of the component <1011 > towards the
development of the prismatic <1010 > pole can be attributed to the prismatic <a> slip, that
starts to dominate the deformation from 0.10 of strain, see figure 5.23a. Lentz et al. [132] also
observed an enhancement of the prismatic texture component in an Mg-Zn-Al alloy during the
deformation in tension, as a consequence of the predominant activation of prismatic <a> mode.
The texture evolution in compression (figure 5.23b) is determined by the distinct balance of
deformation mechanisms, where such a balance is quite different if compare with tension.
Basal <a> slip and tensile twinning accomplish the deformation in compression predominantly.
Therefore, the dominance of basal <a> slip in combination with a high activation of tensile
twining mode leads to the rotation of the <1011 > pole towards the <0001 > pole.
Furthermore, regarding the non-basal deformation modes, it seems that the prismatic <a> slip
is not activated while the pyramidal <a> slip is observed from approximately 0.10 of strain.
The activation of the pyramidal <a> generates a rotation of the <c> axis towards the load
compression [136]. Then, such rotation also contributes to the formation of the <0001 > pole.
The rotation towards the formation of basal <0001 > pole in compression agrees with that
determined for an AZ31 alloy [105], where the plastic deformation is predominantly
accomplished by tensile twinning and basal <a> mode. However, in the present work, in
addition to the activation of such deformation modes, the pyramidal mode family also
influences to some extent the texture development.
77
Figure 5.23. Inverse pole figures comparing the experimental and simulated crystallographic
texture of ZX10 alloy tested under a) tension and b) compression. The texture intensity is in
the unit of m.r.d.
There is a good agreement between the experimental and simulated crystallographic texture for
the ZNd10 alloy as can be seen in figure 5.24. Figure 5.24a shows the rotation of the rare earth
component <1121 > towards the prismatic <1010 > pole when the alloy is tested under
tension. This is associated to the dominance of basal <a> and prismatic <a> modes.
78
Where the prismatic <a> slip starts to play a predominant role from approximately 0.08 until
the end of the deformation. The initial texture as the one observed in the ZNd10 alloy, promotes
the activation of tensile twinning (which is slightly higher than in ZX10 alloy) and pyramidal
<a> slip as well. The activation of these two last deformation modes is also contributing to the
formation of <1010 > pole.
According to the present results, in the ZNd10 alloy, the formation of the <1010 > pole
requires a higher activity of prismatic <a> slip as well as a slightly higher activity of pyramidal
<a> slip if compare with ZX10 alloy. This finding is also in agreement with earlier observations
[137, 138], which showed the development of the prismatic component after tension, due to a
high activity of prismatic <a> mode followed by basal <a> mode.
Figure 5.24b displays the experimental and simulated crystallographic texture of the alloy
tested under compression. In compression, the basal <a> and prismatic <a> slip as well as
extension twin are the most activated deformation modes. However, the activity of basal <a>
slip drops down fast. It is well known that twinning produces a reorientation of the matrix, and
such reorientation activates different deformation modes. Then, the observed activation of
prismatic <a> slip and the relative activity of pyramidal <a> slip is associated with the high
activation of extension twin, which is visible from the onset of deformation.
In addition, a small relative activity of pyramidal I <c+a> and pyramidal II <c+a> is predicted.
Therefore, the formation of the basal <0001 > pole (figure 5.24b) which comes from the
rotation of the <1121 > pole, it can be correlated to the activity of basal <a> slip in
combination with prismatic <a> slip as well as a certain contribution of pyramidal <a> slip.
The observed texture development in this work was also reported by Lentz et al. [137]. In that
work, the texture evolution is associated predominantly to the activity of basal <a> mode,
followed by tensile twin and prismatic <a> mode. As opposed to the results reported in [137],
in the present work it is also suggested that a small activation of pyramidal I <c+a> and
pyramidal II <c+a> modes also have a certain role in the texture formation.
79
Figure 5.24. Inverse pole figures comparing the experimental and simulated crystallographic
texture of ZNd10 alloy tested under a) tension and b) compression. The texture intensity is in
the unit of m.r.d.
80
5.5.5 Ex-situ deformation
In this section, a detailed analysis of the possible deformation modes is given using ex-situ
tension and compression tests. To develop this goal, slip trace analyses were done on ex-situ
measurements. The slip traces are the result of high-localized slip activity. During the
deformation either in tension or in compression, owing to slip on specific crystallographic
planes, steps can be formed that appear as lines on the sample surface when it is observed in
the SEM or LSM. In this work, pictures from LSM are used.
The estimation of the slip activity through this methodology sometimes disregards the
observation of some deformation systems. Some facts complicate the visibility of the
deformation modes. Those deformation systems with Burgers vector parallel to the sample
surface will not develop visible traces [139]. Due to the nature of the constraints differing, the
activity of deformation systems at the surface likely does not reflect the activity inside, and
consequently, the activation of the different deformation modes is affected. Additionally, at
small magnitudes of slip, the slip bands will not be developed. Therefore, it is probable that not
all the slip traces will be visible, and consequently, not all the deformation modes can be
detected. Even so, an effort has been made to distinguish the appearance of slip traces in as
many grains as possible. The dislocation behavior in magnesium alloys has been studied
previously by surface slip trace analysis [140-144].
In this work, the slip trace analysis has been performed under tension at 0.03 and 0.12 of strain
in the three alloys. Under compression, the analysis was done at 0.03 and 0.08 of strain on the
Z1 alloy, and at 0.03 and 0.12 on the ZX10 and ZNd10 alloys.
Figure A3 and A4 from appendix show examples of grains where slip lines are developed in
the alloy Z1. In the aforementioned figures, example of each considered deformation mode
during the tests in tension and in compression are shown. As an important parameter, the
Schmid factor is frequently estimated to analyze the possibility of the activation of deformation
modes. Table 5.5 displays the estimated Schmid factors of the considered slip systems for the
exemplified grains that showed slip traces in Z1 alloy (figures A3 and A4 from appendix).
According to the table below, among the three basal slip systems the closest one to the slip line
developed in tension is the (0001)< 2110 > while in compression is the (0001)< 1120 >.
From the prismatic slip systems, the 1010< 1210 > is activated in tension while the
0110< 2110 > is activated in compression. Regarding the pyramidal <a>, the 0111<
2110 > is activated in tension and the 0111< 2110 > in compression. In reference to the
pyramidal I <c+a>, the 1011< 2113 > is activated in tension while the 0111< 1213 >
is activated in compression. Finally, from the pyramidal II <c+a>, the 2112< 2113 > is
activated in tension and in compression the 1212< 1213 >.
Table 5.5. Estimated Schmid factors of the slip systems in exemplified grains showing slip
traces in Z1 alloy. The Schmid factor of the activated slip systems are in bold for each grain.
Slip
system
number
Grains showing slip lines in tension
Slip
system
number
Grains showing slip lines in compression
Basal
<a>
Prismatic
<a>
Pyramidal
<a>
Pyramidal
I <c+a>
Pyramidal
II <c+a>
Basal
<a>
Prismatic
<a>
Pyramidal
<a>
Pyramidal
I <c+a>
Pyramidal
II <c+a>
1
0.4833
0.0667
0.1071
0.0559
0.0028
3
0.4045
0.3097
0.1030
0.0437
0.4441
4
0.2261
0.4305
0.3594
0.4215
0.4555
5
0.1245
0.0652
0.3510
0.0029
0.2433
11
0.1678
0.4009
0.4551
0.0091
0.0475
8
0.2727
0.2032
0.4669
0.0571
0.0171
14
0.3205
0.1124
0.1274
0.4260
0.0092
16
0.4744
0.4751
0.0318
0.3971
0.0168
27
0.1163
0.1266
0.1813
0.4189
0.0012
35
0.3035
0.3796
0.3511
0.4734
0.2193
81
Table 5.6 displays the estimated Schmid factors of the considered slip systems for the
exemplified grains that showed slip traces in ZX10 alloy (figures A5 and A6 from appendix).
Among the basal slip systems, the slip line developed in tension and in compression is
correlated to the activation of the slip system(0001)< 1120 >. In reference to the prismatic
slip systems, the 1100< 1120 > is activated both in tension and compression. The pyramidal
<a>, i.e., the 0111< 2110 > is activated in tension and in compression. In reference to the
pyramidal I <c+a>, the 0111< 1213 > is activated in tension while the 1011< 2113 > is
activated in compression. Finally, from the pyramidal II <c+a>, the 2112< 2113 > is
activated in tension and in compression the 1122< 1123 >.
Table 5.6. Estimated Schmid factors of the slip systems in exemplified grains showing slip
traces in ZX10 alloy. The Schmid factor of the activated slip systems are in bold for each grain.
Slip
system
number
Grains showing slip lines tension
Slip
system
number
Grains showing slip lines compression
Basal
<a>
Prismatic
<a>
Pyramidal
<a>
Pyramidal
I <c+a>
Pyramidal
II <c+a>
Basal
<a>
Prismatic
<a>
Pyramidal
<a>
Pyramidal
I <c+a>
Pyramidal
II <c+a>
3
0.2935
0.1005
0.3899
0.4382
0.1387
3
0.4678
0.0904
0.0711
0.2510
0.2258
6
0.0104
0.1782
0.2090
0.0425
0.2805
6
0.1736
0.0590
0.1378
0.3871
0.4317
11
0.4205
0.0090
0.2115
0.1534
0.0696
11
0.4121
0.2974
0.4030
0.0866
0.0608
22
0.0153
0.0079
0.0504
0.4610
0.0506
14
0.2739
0.2188
0.2424
0.4393
0.0438
28
0.1297
0.3953
0.0275
0.2404
0.4508
31
0.2106
0.2641
0.1124
0.0669
0.1354
Table 5.7 displays the estimated Schmid factors of the considered slip systems for the
exemplified grains that showed slip traces in ZNd10 alloy (figures A7 and A8 from appendix).
In this alloy, among the three basal slip systems, the slip line developed in tension is associated
to the slip system (0001)< 1210 > while in compression is the (0001)< 1120 >. From the
prismatic slip systems, the 0110< 2110 > is activated in tension while the 1100< 1120 >
is activated in compression. In relation to the pyramidal <a>, the 1101< 1120 > is activated
in tension and the 0111< 2110 > in compression. In reference to the pyramidal I <c+a>, the
0111<1123 > is activated in tension while the 1101< 1213 > is activated in
compression. Finally, from the pyramidal II <c+a>, the 1122<1123 > is activated in tension
and in compression the 1212< 1213 >.
Table 5.7. Estimated Schmid factors of the slip systems in exemplified grains showing slip
traces in ZNd10 alloy. The Schmid factor of the activated slip systems are in bold for each
grain.
Slip
system
number
Grains showing slip lines tension
Slip
system
number
Grains showing slip lines compression
Basal
<a>
Prismatic
<a>
Pyramidal
<a>
Pyramidal
I <c+a>
Pyramidal
II <c+a>
Basal
<a>
Prismatic
<a>
Pyramidal
<a>
Pyramidal
I <c+a>
Pyramidal
II <c+a>
2
0.4431
0.1738
0.0474
0.3686
0.2760
3
0.4740
0.1889
0.0507
0.0395
0.4628
5
0.3498
0.4832
0.3654
0.0024
0.2753
6
0.1082
0.4510
0.0537
0.0025
0.0403
9
0.1450
0.2301
0.4306
0.1194
0.0962
8
0.0161
0.2860
0.3896
0.0349
0.1856
15
0.2327
0.1499
0.0058
0.0170
0.1923
24
0.3784
0.2045
0.2621
0.4366
0.2357
25
0.0006
0.0078
0.3563
0.3140
0.0843
29
0.1965
0.0430
0.2349
0.4596
0.3780
After tension and compression tests, the statistics of the identified grains with slip lines is
reported for each alloy in the pictures 5.25 to 5.27 as well as in the tables 5.8 and 5.9.
The results on the slip trace analysis show a dominance of the basal <a> on the Z1 alloy since
the beginning of the deformation either in tension or in compression see figure 5.25. The
prismatic <a> slip is also important during the deformation. During tension, as the deformation
proceeds, the amount of prismatic <a> slip increases, see figure 5.25a. In contrast, during the
deformation in compression the amount of such slip mode is low, see figure 5.25b. Regarding
to the activation of pyramidal <a>, pyramidal I <c+a> and pyramidal <c+a> II modes, it was
82
quite hard to be observed in this Z1 alloy during the deformation. Nevertheless, some grains
were observed showing slip lines corresponding to the pyramidal slip family. The amount of
this kind of deformation modes is rather low when the alloy is tested in tension, see figure
5.25a. In contrast, more grains are visible when is tested in compression, figure 5.25b.
Thus, under tension, the texture evolution previously observed (figure 5.15a) is attributed
mainly to the activation of basal <a> and prismatic <a> slip (figure 5.31a), with a rather low
activation of extension twinning (figure 5.12a). On the contrary, in compression, the basal
texture (figure 5.15b) developed during the deformation is attributed to the high activation of
extension twinning (figure 5.12b) combined with basal slip (figure 5.25b). With a comparable
texture to the Z1 alloy, AZ31 alloy during tension [145] slip traces on grains were also
observed, which were associated with basal <a>, prismatic <a> as well as pyramidal <a> slip.
As opposed to that research, in the present work, in addition to the previously mentioned slip
traces, a low number of grains showing slip lines associated to pyramidal I <c+a> and
pyramidal II <c+a> were also present. This discrepancy could be attributed to the fact that the
slip trace analysis in the present work was performed at higher deformation strain.
The summary of the slip trace analysis performed during the deformation in tension and in
compression of the ZX10 alloy is illustrated in figure 5.26. In tension, there is a high activation
of basal <a> followed by prismatic <a> slip, figure 5.26a. Both slip modes play an important
role since the start of the deformation. At 0.12 of strain, the activation of pyramidal slip mode
family is shown. From these pyramidal modes, the most activated correspond to the pyramidal
<a>. In reference to the pyramidal I <c+a> and pyramidal II <c+a> its activation is quite
comparable, figure 5.26a.
In compression, the deformation is dominated principally by basal <a> slip, figure 5.26b. In
this case, a low activation of prismatic <a> slip is remarkable. Grains showing slip lines of
pyramidal <a>, pyramidal I <c+a> and pyramidal II <c+a> were observed at 0.12 of
deformation. From these deformation modes, the high activation corresponds to the pyramidal
<a>. Therefore, from the previously observed, the crystallographic texture evolution in tension
(figure 5.16a) is mainly attributed to the activation of basal <a> and prismatic <a> slip (figure
5.26a) as well as the extension twin activation since 0.03 of strain (figure 5.13a).
On the other hand, the texture in compression (figure 5.16b) is developed due to the dominance
of basal <a> slip (figure 5.26b) with a pronounced activation of extension twinning (figure
5.13b). Basal <a>, prismatic <a>, pyramidal <a> and pyramidal II <c+a> slip traces reported
in this work were also observed for a binary Mg-Ca alloy tested in tension [145]. In contrast,
some slip lines associated to pyramidal I <c+a> were detected in the present research work
(figure. 5.26a), but not in [145]. These results show the importance of Ca addition to the
magnesium alloys to promote the activity of non-basal modes, especially, those ones belonging
to the pyramidal mode family which are harder to be activated at room temperature.
The results on the estimation of slip traces in the ZNd10 alloy are shown in figure 5.27. Like
in the previous alloys, the basal <a> and prismatic <a> slips take relevance in the early stage
of deformation. As the deformation proceeds in tension, the activity of prismatic <a> slip
increases, see figure 5.27a. At 0.12 of strain, the activation of pyramidal <a> is higher in
comparison with pyramidal I and II <c+a>. On the other hand, in compression, basal <a> and
prismatic <a> slips play an important role from the beginning of deformation, see figure 5.27b.
The activation of the pyramidal slip family is observed at the deformation of 0.12. Again, the
pyramidal <a> has a higher activity if compared to the pyramidal I and II <c+a>, figure 5.27b.
Thus, the deformation in tension was dominated by basal <a> and prismatic <a> slip (figure
5.27a) with a pronounced activation of extension twins (figure 5.14a) that can be consider the
responsible in the texture evolution in tension (figure 5.17a).
83
However, in compression the activity of basal <a> slip dominates the deformation (figure
5.27b), in combination with a profuse activation of extension twin (figure 5.14b) which
contribute to the texture development (figure 5.17b).
It is known that the addition of rare earth elements to Mg and its alloys increases the activity
of non-basal slip. At a deformation of 4% in tension of a Mg-Nd alloy [145], basal <a> and
prismatic <a> slip traces were detected; this in in agreement with the findings in the present
work, at least until 0.03 of strain (figure 5.27a). With the further increase of strain, grains with
slip traces associated with the pyramidal <a>, pyramidal I <c+a> and pyramidal II <c+a>
modes were also detected.
Figure 5.25. Statistics of the identified slip activity at different strains for the Z1 alloy tested in
a) tension and b) compression.
Figure 5.26. Statistics of the identified slip activity at different strains for the ZX10 alloy tested
in a) tension and b) compression.
84
Figure 5.27. Statistics of the identified slip activity at different strains for the ZNd10 alloy
tested in a) tension and b) compression.
Table 5.8. Identified slip activities in each alloy tested in tension.
Alloy
Strain
Slip traces
Basal <a>
Prismatic <a>
Pyramidal <a>
Pyramidal I <c+a>
Pyramidal II <c+a>
Number
Fraction
Number
Fraction
Number
Fraction
Number
Fraction
Number
Fraction
Z1
0.03
27
93.10 %
2
6.90 %
0
0.00
0
0.00
0
0.00
0.12
16
37.21 %
23
53.48 %
2
4.65 %
1
2.33 %
1
2.33 %
ZX10
0.03
38
92.68 %
3
7.32 %
0
0.00
0
0.00
0
0.00
0.12
18
36.70 %
11
22.46 %
14
28.60 %
3
6.12 %
3
6.12 %
ZNd10
0.03
38
95.00 %
2
5.00 %
0
0.00 %
0
0.00
0
0.00
0.12
30
45.45 %
19
28.80 %
10
15.15 %
3
4.54 %
4
6.06 %
Table 5.9. Identified slip activities in each alloy tested in compression.
Alloy Strain
Slip traces
Basal <a>
Prismatic <a>
Pyramidal <a>
Pyramidal I <c+a>
Pyramidal II <c+a>
Number
Fraction
Number
Fraction
Number
Fraction
Number
Fraction
Number
Fraction
Z1
0.03
36
97.30 %
1
2.70 %
0
0.00
0
0.00
0
0.00
0.08
35
60.36 %
4
6.89 %
1
1.72 %
5
8.62 %
13
22.41%
ZX10
0.03
37
97.37 %
1
2.63 %
0
0.00
0
0.00
0
0.00
0.12
25
53.19 %
3
6.38 %
8
17.03%
6
12.76 %
5
10.64%
ZNd10
0.03
32
86.49 %
5
13.51 %
0
0.00
0
0.00
0
0.00
0.12
25
49.03 %
10
19.60 %
7
13.73 %
5
9.80 %
4
7.84 %
85
6. Conclusions and summary
Conclusions
Throughout this work a systematic study of the effect of alloying elements, extrusion
parameters, microstructure and crystallographic texture development, mechanical properties as
well as the deformation behavior of Mg-Zn based alloys was performed. The applied different
extrusion parameters gave several microstructures that allowed the investigation of the grain
size and crystallographic texture as well as its effects on the mechanical properties and the
plastic deformation of the alloys. The effect of single additions of Ca and Nd on the activation
of deformation mechanisms was experimentally studied and these results were compared with
simulation results, which involved the application of the crystal plasticity model, i.e., the VPSC
model. In a deeper analysis, ex-situ tests to observe and determine the activation of the different
deformation modes were accomplished to corroborate the simulation results.
In the following, the conclusions are stated.
Z1 alloy
The Z1 alloy shows a characteristic extrusion behavior with low extrusion pressures.
These reached pressures are even lower with increasing the extrusion temperature.
In this binary alloy, the recrystallization process is not greatly affected at any applied
extrusion condition tested. The most altered feature is the average grain size, which
increases at each increment of extrusion temperature or extrusion speed.
The different extrusion parameters applied to the Z1 alloy, give rise to the formation of
basal textures, where the basal planes of the HCP structure are parallel to the extrusion
direction. That means there is a conventional development of the classical <1120 > -
<1010 > texture independent of modification of extrusion speed or extrusion
temperature.
The DRX process, which takes place during the extrusion process of the Z1 alloy
allowed the observation that the dynamically recrystallized grains tend to orient to the
<1120 > pole. This texture component is strengthened as grain growth takes place.
Due to the high fraction of dynamically recrystallized grains founded in the alloy at any
extrusion condition, not changes in the texture were observed after SRX process.
The mechanical properties are influenced by the initial texture and grain size. The basal
texture like the one that is observed in this alloy Z1 results in the well-known yield
asymmetry. This is attributed to a higher activation of twinning in compression.
The VPSC model successfully predicted the flow curves as well as the texture
evolution, either in tension or in compression. The crystal plasticity simulations using
the VPSC model suggest that the prismatic <a> and basal <a> slip dominate the
deformation in tension. On the other hand, the basal <a> slip and extension twinning
dominate the deformation in compression.
The slip trace analysis at room temperature confirms the activation of basal <a> and
prismatic <a> slip during the deformation in tension. Scarce activation of pyramidal
<a>, pyramidal I <c+a> and pyramidal II <c+a> are observed at large strains. In
86
contrast, slip traces of basal <a> slip is highly observed in compression. Among the
pyramidal slip systems, the pyramidal <a> is the most active but also at large
deformation.
ZX10 alloy
The pressure reached during extrusion processing of ZX10 alloy is higher than that
observed in the Z1 alloy owing to a higher alloy content, i.e., the addition of Ca to the
Mg-Zn base alloy. Some micro cracks are visible perpendicular to the extrusion
direction. However, such micro cracks do not affect severely the surface quality.
In this ZX10 alloy, the different processing conditions give rise to a distinct
microstructure and crystallographic texture. The DRX process is affected due to the
addition of Ca into the binary system. The grain growth is delayed when Ca is added.
This is manifested with a smaller average grain size if compare with the average grain
size revealed in the Z1 alloy.
The crystallographic texture is modified in such a way that the main component
becomes the <1011 > pole when the alloy is extruded at 250 and 300 °C. With the
grain growth, the orientations are spread between the <1120 > and <1010 > pole.
The formation of a basal crystallographic texture is suggested.
Post extrusion heat treatments in partly recrystallized microstructures revealed the
importance of SRX process. The strong prismatic texture of the partly recrystallized
microstructure is weakened with the applied heat treatment, but in essence, the grains
maintain their orientation in the surroundings of the <1010 > as well as along the arc
between the <1120 > and <1010 > pole.
The crystallographic textures dominated by the <1011 > pole observed during the
processing of ZX10 alloy resulted in the reduction of the tension-compression yield
asymmetry. However, in those samples where the <1120 >-<1010 > texture is
observed, the yield asymmetry increased. In this Zx10 alloy, the obtained ductility is
higher if compared with the results from the Z1 alloy.
The VPSC simulation results and the slip trace analysis show the dominance of basal
and prismatic slip during the deformation in tension. In addition, such results suggest
that Ca enhances the activity of the pyramidal <a> slip system. On the other hand,
although basal slip and extension twinning dominate the deformation in compression,
there is a clear enhancement of the pyramidal slip systems. The slip trace analysis
reveals a higher activity of pyramidal <a> slip among the pyramidal modes.
ZNd10 alloy
A higher pressure reached during the extrusion processing due to a higher alloying
content is confirmed, which in this case is owing to the addition of Nd into the Mg-Zn
base alloy. On the other hand, there is no clear evidence of cracks on the bars surface
of ZNd10 alloy.
The addition of Nd into the Mg-Zn based alloy system delayed the recrystallization
process, as some non-recrystallized grains are present at the extrusion of lowest
87
extrusion temperature combined with the lowest extrusion speed, i.e 250 °C and 2.0
mm/s. Some traces of un-recrystallized grains are also perceptible when is extruded at
300 °C and 2.0 mm/s.
The presence of Nd into the Mg-Zn base alloy also delays the grain growth. The grain
growth retardation during the DRX is more pronounced if compared with the addition
of Ca. High extrusion speeds or temperatures are needed to attain completely
recrystallized microstructures in this ZNd10 alloy. The resulted average grain size is
the smallest in this Nd containing alloy. Therefore, from the two alloying elements, Nd
is more effective in the grain refining compared with Ca.
The strong prismatic <1010 > texture component results at the lowest extrusion speed
combined with the lowest extrusion temperature. This texture component is associated
with the presence of non-recrystallized grains in the microstructure. With the fully
recrystallized microstructure, the texture intensity reduces and the non-basal
component, <1121 > pole dominates the texture in this ZNd10 alloy.
Regarding to annealing heat treatments to trigger SRX in partially recrystallized
samples, the results showed that the texture intensity reduces and the development of
the <1121 > component was inhibited. The orientation of the grains is mainly
distributed in between the <1120 > and <1010 > poles.
The ductility increases as well as the reduction of the yield asymmetry observed during
the characterization of the mechanical properties is associated with the increase of
recrystallized fraction of the microstructure and the non-basal texture.
According to the plasticity simulations with the VPSC model, basal <a> and prismatic
<a> slip dominate the deformation in tension. The activity of pyramidal <a> slip was
also suggested with the implementation of the VPSC model. In this ZNd10 alloy, the
activation of pyramidal <a> is slightly higher than its activation in the Ca containing
alloy. On the other hand, in compression, the VPSC simulation suggests that basal <a>
slip dominate the deformation with a high activation of extension twin and prismatic
<a> slip. Additionally, until 0.12 strain, the activity of the three pyramidal slip is quite
comparable.
The slip trace analysis in tension confirms the dominance of basal <a> and prismatic
<a> slip as well as the high activation of pyramidal <a> slip among all the pyramidal
slips. During the deformation under compression, basal <a> and prismatic <a> slip are
also confirmed. In addition, the activation of pyramidal <a>, pyramidal I <c+a> and
pyramidal II <c+a> slip is quite similar.
In this dissertation, a systematic study of the effect of Ca or Nd, the extrusion processing
parameters on the microstructure, texture, mechanical properties and the deformation
behaviour of Mg-Zn based alloys was performed. It was shown that the microstructure and
texture evolution of the Mg-Zn alloys are a result of Ca or Nd additions and is also affected
by extrusion parameters. The addition of alloying elements modifies the recrystallization
kinetics, which is different in each alloy. This work indicates that the preference of grain
orientations undoubtedly changes with the extrusion condition and the concomitant grain
growth of dynamically recrystallized grains.
88
This work suggests that during processing, the SRX plays a very important role in the texture
modification. Via static recrystallization of partially recrystallized samples, it is possible to
achieve grain structures that are very comparable to those obtained by dynamic
recrystallization however with a different texture. During the SRX process, the development
of the <11–21> component was inhibited in the ZX10 and ZNd10 alloys. Furthermore, both
alloys inherit the prismatic character of the deformed microstructure. Due to a change in the
balance of deformation mechanisms as well as the increased importance of grain growth of the
recrystallized grains, the SRXed texture is very different compared to DRXed texture.
The ZX10 and ZNd10 alloys show significantly reduction of tension-compression yield
asymmetry and enhanced room temperature ductility compared to Z1 alloy. Such
enhancements are associated to the activation of additional non-basal slip systems due to not
only to chemical composition, but also to the initial non-basal crystallographic texture. This
dissertation indicates that adding Ca or Nd to the Mg-Zn system, there is a reduction of the
ratio between the CRSS of non-basal deformation modes and the CRSS of basal mode.
Furthermore, this work shows evidence of the activation of pyramidal <a> mode. Such a
deformation mode is not normally reported. However, the pyramidal <a> mode in this work
takes relevance in explaining both, the hardening behavior of the materials as well as the texture
development.
7. Directions for future work
This thesis covered diverse aspects of the extrusion behavior of magnesium alloys, but still
many research opportunities remain that could be explored. A possible study could be the
variation in the profile shape, e.g., the extrusion of flat bars, which could have the possibility
to study not only its microstructure-texture evolution but also its mechanical properties as for
example formability in addition to the common measured tensile properties.
To gain a deeper understanding of the texture effect on the activity of deformation modes,
future work can focus on TEM experiments to determine the type of dislocations that are
activated. This will be useful to support the slip trace observations made in this study.
In the present work, applying the VPSC model an effort was made to estimate the CRSS of the
different slip modes involved in the deformation. Therefore, it would be interesting the
experimental and systematic determination of those CRSS. The experimental determination of
the CRSS could facilitate the simulation of deformation modes applying the VPSC model, not
only considering that the material is deformed at room temperature but also at high temperature.
This can be particularly beneficial for the accurate estimation of the CRSS ratios for different
deformation modes.
89
8. References
[1] A.G. Beer, 8 - Enhancing the extrudability of wrought magnesium alloys, in: C. Bettles, M.
Barnett (Eds.), Advances in Wrought Magnesium Alloys, Woodhead Publishing2012, pp. 304-
322.
[2] J. Bohlen, D. Letzig, K.U. Kainer, New perspectives for wrought magnesium alloys,
Materials Science Forum, Trans Tech Publ, 2007, pp. 1-10.
[3] I. Polmear, D. StJohn, J.-F. Nie, M. Qian, 6 - Magnesium Alloys, in: I. Polmear, D. StJohn,
J.-F. Nie, M. Qian (Eds.), Light Alloys (Fifth Edition), Butterworth-Heinemann, Boston, 2017,
pp. 287-367.
[4] J. Bohlen, S. Yi, D. Letzig, K.U. Kainer, Effect of rare earth elements on the microstructure
and texture development in magnesium–manganese alloys during extrusion, Materials Science
and Engineering: A 527(26) (2010) 7092-7098.
[5] M.K. Kulekci, Magnesium and its alloys applications in automotive industry, The
International Journal of Advanced Manufacturing Technology 39(9) (2008) 851-865.
[6] W.D. Callister, D.G. Rethwisch, Materials Science and Engineering: An Introduction, 8th
Edition, Wiley2009.
[7] K.H. Matucha, Materials Science and Technology, Volume 8, Structure and Properties of
Nonferrous Alloys, Masters Thesis 8 (1996) 837.
[8] H.E. Friedrich, B.L. Mordike, Technology of magnesium and magnesium alloys,
Magnesium Technology: Metallurgy, Design Data, Applications (2006) 219-430.
[9] T.B. Abbott, Magnesium: Industrial and Research Developments Over the Last 15 Years,
CORROSION 71(2) (2015) 120-127.
[10] K. Gusieva, C.H.J. Davies, J.R. Scully, N. Birbilis, Corrosion of magnesium alloys: the
role of alloying, International Materials Reviews 60(3) (2015) 169-194.
[11] M. Bauser, G. Sauer, K. Siegert, Extrusion, ASM International, Materials Park, OH, 2006.
[12] Z. Zhuoran, N. Stanford, C. Davies, J.F. Nie, N. Birbilis, Magnesium extrusion alloys: a
review of developments and prospects, International Materials Reviews (2018) 1-36.
[13] T. Nakata, T. Mezaki, R. Ajima, C. Xu, K. Oh-ishi, K. Shimizu, S. Hanaki, T.T. Sasaki,
K. Hono, S. Kamado, High-speed extrusion of heat-treatable MgAlCaMn dilute alloy,
Scripta Materialia 101 (2015) 28-31.
[14] S.W. Xu, K. Oh-ishi, S. Kamado, F. Uchida, T. Homma, K. Hono, High-strength extruded
MgAlCa–Mn alloy, Scripta Materialia 65(3) (2011) 269-272.
[15] T. Homma, N. Kunito, S. Kamado, Fabrication of extraordinary high-strength magnesium
alloy by hot extrusion, Scripta Materialia 61(6) (2009) 644-647.
[16] B. Zhang, Y. Wang, L. Geng, C. Lu, Effects of calcium on texture and mechanical
properties of hot-extruded MgZnCa alloys, Materials Science and Engineering: A 539
(2012) 56-60.
[17] A.A. Nayeb-Hashemi, J.B. Clark, A.S.M. International, Phase diagrams of binary
magnesium alloys, ASM International, Metals Park, Ohio, 1988.
[18] A.S. Handbook, Magnesium and magnesium alloys, ASM international (1999) 106-118.
90
[19] D. Wu, R.S. Chen, W. Ke, Microstructure and mechanical properties of a sand-cast Mg
NdZn alloy, Materials & Design 58 (2014) 324-331.
[20] J. Yang, J. Wang, L. Wang, Y. Wu, L. Wang, H. Zhang, Microstructure and mechanical
properties of Mg–4.5Zn–xNd (x=0, 1 and 2, wt%) alloys, Materials Science and Engineering:
A 479(1) (2008) 339-344.
[21] E.F. Emley, Principles of magnesium technology, Pergamon Press, Oxford; New York,
1966.
[22] A.K.D. Y.C. Lee, D.H. St.John, Grain refinement of magnesium in Magnesium
Technology 2000, 211-218.
[23] A.H.B. C.H. Cáceres, The Strength of Concentrated Mg-Zn Solid Solutions, Physica
Status Solidi Applied Research 194(1) (2002) 147-158.
[24] G. Mann, J.R. Griffiths, C.H. Cáceres, Hall-Petch parameters in tension and compression
in cast Mg–2Zn alloys, Journal of Alloys and Compounds 378(1) (2004) 188-191.
[25] D. Persaud-Sharma, A. McGoron, Biodegradable magnesium alloys: a review of material
development and applications, Journal of Biomimetics, Biomaterials and Tissue Engineering,
Trans Tech Publ, 2011, pp. 25-39.
[26] Y. Chen, Z. Xu, C. Smith, J. Sankar, Recent advances on the development of magnesium
alloys for biodegradable implants, Acta Biomaterialia 10(11) (2014) 4561-4573.
[27] M. Mezbahul-Islam, A. Mostafa, M. Medraj, Essential magnesium alloys binary phase
diagrams and their thermochemical data, Journal of Materials 2014 (2014).
[28] J.D. Robson, D.T. Henry, B. Davis, Particle effects on recrystallization in magnesium–
manganese alloys: Particle-stimulated nucleation, Acta Materialia 57(9) (2009) 2739-2747.
[29] S. Ganeshan, S.L. Shang, Y. Wang, Z.K. Liu, Effect of alloying elements on the elastic
properties of Mg from first-principles calculations, Acta Materialia 57(13) (2009) 3876-3884.
[30] J. Zhang, Y. Dou, G. Liu, Z. Guo, First-principles study of stacking fault energies in Mg-
based binary alloys, Computational Materials Science 79 (2013) 564-569.
[31] L.B. Tong, M.Y. Zheng, L.R. Cheng, D.P. Zhang, S. Kamado, J. Meng, H.J. Zhang,
Influence of deformation rate on microstructure, texture and mechanical properties of indirect-
extruded Mg–ZnCa alloy, Materials Characterization 104 (2015) 66-72.
[32] M. Nienaber, K.U. Kainer, D. Letzig, J. Bohlen, Processing Effects on the Formability of
Extruded Flat Products of Magnesium Alloys, Frontiers in Materials 6(253) (2019).
[33] J.P. Hadorn, K. Hantzsche, S. Yi, J. Bohlen, D. Letzig, J.A. Wollmershauser, S.R. Agnew,
Role of Solute in the Texture Modification During Hot Deformation of Mg-Rare Earth Alloys,
Metallurgical and Materials Transactions A 43(4) (2012) 1347-1362.
[34] J.P. Hadorn, K. Hantzsche, S. Yi, J. Bohlen, D. Letzig, S.R. Agnew, Effects of Solute and
Second-Phase Particles on the Texture of Nd-Containing Mg Alloys, Metallurgical and
Materials Transactions A 43(4) (2012) 1363-1375.
[35] L.L. Rokhlin, Magnesium alloys containing rare earth metals: structure and properties,
Crc Press2003.
91
[36] L. Ma, R.K. Mishra, L. Peng, A.A. Luo, W. Ding, A.K. Sachdev, Texture and mechanical
behavior evolution of age-hardenable MgNdZn extrusions during aging treatment, Materials
Science and Engineering: A 529 (2011) 151-155.
[37] P. Hidalgo-Manrique, S.B. Yi, J. Bohlen, D. Letzig, M.T. Pérez-Prado, Effect of Nd
Additions on Extrusion Texture Development and on Slip Activity in a Mg-Mn Alloy,
Metallurgical and Materials Transactions A 44(10) (2013) 4819-4829.
[38] B. Lv, J. Peng, Y. Peng, A. Tang, The effect of addition of Nd and Ce on the microstructure
and mechanical properties of ZM21 Mg alloy, Journal of Magnesium and Alloys 1(1) (2013)
94-100.
[39] A. Imandoust, C.D. Barrett, T. Al-Samman, K.A. Inal, H. El Kadiri, A review on the effect
of rare-earth elements on texture evolution during processing of magnesium alloys, Journal of
Materials Science 52(1) (2017) 1-29.
[40] N. Stanford, M.R. Barnett, The origin of “rare earth” texture development in extruded Mg-
based alloys and its effect on tensile ductility, Materials Science and Engineering: A 496(1-2)
(2008) 399-408.
[41] N. Stanford, Micro-alloying Mg with Y, Ce, Gd and La for texture modificationA
comparative study, Materials Science and Engineering: A 527(10) (2010) 2669-2677.
[42] J.P. Hadorn, R.P. Mulay, K. Hantzsche, S. Yi, J. Bohlen, D. Letzig, S.R. Agnew, Texture
Weakening Effects in Ce-Containing Mg Alloys, Metallurgical and Materials Transactions A
44(3) (2013) 1566-1576.
[43] G. Ben-Hamu, D. Eliezer, K.S. Shin, S. Cohen, The relation between microstructure and
corrosion behavior of Mg–Y–REZr alloys, Journal of Alloys and Compounds 431(1) (2007)
269-276.
[44] S. Graff, Micromechanical modeling of the deformation of HCP Metals, GKSS-
Forschungszentrum Geesthacht, Bibliothek, 2007.
[45] D.R. Askeland, P.P. Fulay, W.J. Wright, The Science and Engineering of Materials,
Cengage Learning2010.
[46] G.E. Dieter, D. Bacon, Mechanical metallurgy, McGraw-hill New York1986.
[47] Y.N. Berdovsky, Intermetallics research progress, Nova Publishers2008.
[48] J. Pelleg, Mechanical properties of materials, Springer Science & Business Media2012.
[49] K.U. Kainer, Magnesium: proceedings of the 7th International Conference on Magnesium
Alloys and their Applications, John Wiley & Sons2007.
[50] M.H. Yoo, S.R. Agnew, J.R. Morris, K.M. Ho, Non-basal slip systems in HCP metals and
alloys: source mechanisms, Materials Science and Engineering: A 319-321 (2001) 87-92.
[51] F. Czerwinski, Magnesium Injection Molding, Springer US2007.
[52] Z. Zheng, D.S. Balint, F.P.E. Dunne, Rate sensitivity in discrete dislocation plasticity in
hexagonal close-packed crystals, Acta Materialia 107 (2016) 17-26.
[53] M.R. Barnett, A taylor model based description of the proof stress of magnesium AZ31
during hot working, Metallurgical and Materials Transactions A 34(9) (2003) 1799-1806.
92
[54] J.W. Christian, S. Mahajan, Deformation twinning, Progress in Materials Science 39(1)
(1995) 1-157.
[55] M.H. Yoo, Slip, twinning, and fracture in hexagonal close-packed metals, Metallurgical
Transactions A 12(3) (1981) 409-418.
[56] E. Tenckhoff, Deformation mechanisms, texture, and anisotropy in zirconium and
zircaloy, ASTM International1988.
[57] M.R. Barnett, Twinning and the ductility of magnesium alloys: Part I: “Tension” twins,
Materials Science and Engineering: A 464(1) (2007) 1-7.
[58] R. Abbaschian, R.E. Reed-Hill, Physical metallurgy principles, Cengage Learning2008.
[59] H.A. Padilla, C.D. Smith, J. Lambros, A.J. Beaudoin, I.M. Robertson, Effects of
Deformation Twinning on Energy Dissipation in High Rate Deformed Zirconium,
Metallurgical and Materials Transactions A 38(12) (2007) 2916-2927.
[60] P.K. Saha, Aluminum extrusion technology, Asm International2000.
[61] J. Swiostek, J. Göken, D. Letzig, K.U. Kainer, Hydrostatic extrusion of commercial
magnesium alloys at 100°C and its influence on grain refinement and mechanical properties,
Materials Science and Engineering: A 424(1) (2006) 223-229.
[62] D.L. Atwell, M.R. Barnett, Extrusion Limits of Magnesium Alloys, Metallurgical and
Materials Transactions A 38(12) (2007) 3032-3041.
[63] W.H. Sillekens, M.H.F.M. Hout, F. Pravdic, Extrusion technology for magnesium:
Avenues for improving performance, 2005.
[64] A.A. Luo, C. Zhang, A.K. Sachdev, Effect of eutectic temperature on the extrudability of
magnesium–aluminum alloys, Scripta Materialia 66(7) (2012) 491-494.
[65] T. Murai, S.-i. Matsuoka, S. Miyamoto, Y. Oki, S. Nagao, H. Sano, Effects of zinc and
manganese contents on extrudability of Mg&ndash;Al&ndash;Zn alloys, Journal of Japan
Institute of Light Metals 53(1) (2003) 27-31.
[66] M.G. Jiang, C. Xu, T. Nakata, H. Yan, R.S. Chen, S. Kamado, High-speed extrusion of
dilute Mg-Zn-Ca-Mn alloys and its effect on microstructure, texture and mechanical properties,
Materials Science and Engineering: A 678 (2016) 329-338.
[67] F.J. Humphreys, M. Hatherly, Chapter 13 - Hot Deformation and Dynamic Restoration,
in: F.J. Humphreys, M. Hatherly (Eds.), Recrystallization and Related Annealing Phenomena
(Second Edition), Elsevier, Oxford, 2004, pp. 415-V.
[68] H.J.M. J.J. Jonas, Recovery and Recrystallisation during High Temperature Deformation,
CNRS.“Mise en forme des metaux et alliages”, Villars-sur-Ollon, 1975.
[69] R. Kaibyshev, 5 - Dynamic recrystallization in magnesium alloys, in: C. Bettles, M.
Barnett (Eds.), Advances in Wrought Magnesium Alloys, Woodhead Publishing2012, pp. 186-
225.
[70] O. Sivakesavam, I.S. Rao, Y.V.R.K. Prasad, Processing map for hot working of as cast
magnesium, Materials Science and Technology 9 (1993) 805-810.
[71] S. Ion, F. Humphreys, S. White, Dynamic recrystallisation and the development of
microstructure during the high temperature deformation of magnesium, Acta Metallurgica
30(10) (1982) 1909-1919.
93
[72] O. Sitdikov, R. Kaibyshev, Dynamic Recrystallization in Pure Magnesium, MATERIALS
TRANSACTIONS 42(9) (2001) 1928-1937.
[73] A.G. Beer, M.R. Barnett, Microstructural Development during Hot Working of Mg-3Al-
1Zn, Metallurgical and Materials Transactions A 38(8) (2007) 1856-1867.
[74] C. Barrett, A. Imandoust, A. Oppedal, K. Inal, M. Tschopp, H. Kadiri, Effect of Grain
Boundaries on Texture Formation during Dynamic Recrystallization of Magnesium Alloys,
Acta Materialia 128 (2017).
[75] O. Muránsky, D.G. Carr, M.R. Barnett, E.C. Oliver, P. Šittner, Investigation of
deformation mechanisms involved in the plasticity of AZ31 Mg alloy: In situ neutron
diffraction and EPSC modelling, Materials Science and Engineering: A 496(1) (2008) 14-24.
[76] R. Kaibyshev, O. Sitdikov, On the role of twinning in dynamic recrystallization, The
Physics of Metals and Metallography 89 (2000) 384-390.
[77] E.A. Ball, P.B. Prangnell, Tensile-compressive yield asymmetries in high strength
wrought magnesium alloys, Scripta Metallurgica et Materialia 31(2) (1994) 111-116.
[78] K.D. Molodov, T. Al-Samman, D.A. Molodov, G. Gottstein, Mechanisms of exceptional
ductility of magnesium single crystal during deformation at room temperature: Multiple
twinning and dynamic recrystallization, Acta Materialia 76 (2014) 314-330.
[79] X. Li, P. Yang, L.N. Wang, L. Meng, F. Cui, Orientational analysis of static
recrystallization at compression twins in a magnesium alloy AZ31, Materials Science and
Engineering: A 517(1) (2009) 160-169.
[80] É. Martin, R.K. Mishra, J.J. Jonas, Effect of twinning on recrystallisation textures in
deformed magnesium alloy AZ31, Philosophical Magazine 91(27) (2011) 3613-3626.
[81] C. Drouven, I. Basu, T. Al-Samman, S. Korte-Kerzel, Twinning effects in deformed and
annealed magnesiumneodymium alloys, Materials Science and Engineering: A 647 (2015)
91-104.
[82] A.D. Murphy, J.E. Allison, The Recrystallization Behavior of Unalloyed Mg and a Mg-
Al Alloy, Metallurgical and Materials Transactions A 49(5) (2018) 1492-1508.
[83] C.W. Su, L. Lu, M.O. Lai, Recrystallization and grain growth of deformed magnesium
alloy, Philosophical Magazine 88(2) (2008) 181-200.
[84] D. Guan, W.M. Rainforth, L. Ma, B. Wynne, J. Gao, Twin recrystallization mechanisms
and exceptional contribution to texture evolution during annealing in a magnesium alloy, Acta
Materialia 126 (2017) 132-144.
[85] J. Victoria-Hernández, S. Yi, D. Klaumünzer, D. Letzig, Recrystallization behavior and
its relationship with deformation mechanisms of a hot rolled Mg-Zn-Ca-Zr alloy, Materials
Science and Engineering: A 761 (2019) 138054.
[86] J. Hirsch, T. Al-Samman, Superior light metals by texture engineering: Optimized
aluminum and magnesium alloys for automotive applications, Acta Materialia 61(3) (2013)
818-843.
[87] T. Al-Samman, K.D. Molodov, D.A. Molodov, G. Gottstein, S. Suwas, Softening and
dynamic recrystallization in magnesium single crystals during c-axis compression, Acta
Materialia 60(2) (2012) 537-545.
94
[88] O. Engler, V. Randle, Introduction to texture analysis: macrotexture, microtexture, and
orientation mapping, CRC press2009.
[89] E.W. Kelley, W.F. Hosford, Part I January 1968 - Papers - Plane-Strain Compression of
Magnesium and Magnesium Alloy Crystals, The American Institute of Mining, Metallurgical,
and Petroleum Engineers, 1969.
[90] L. Nascimento Silva Ferri, High cycle fatigue behaviour of extruded magnesium alloys
containing neodymium, (2014).
[91] B.D. Cullity, S.R. Stock, Elements of x-ray diffraction, Prentice Hall, Upper Saddle River,
NJ, 2001.
[92] U.F. Kocks, C.N. Tomé, H.-R. Wenk, Texture and anisotropy: preferred orientations in
polycrystals and their effect on materials properties, Cambridge university press1998.
[93] F. Kaiser, K. Kainer, Magnesium alloys and technology, John Wiley & Sons2003.
[94] J. Bohlen, S. Yi, J. Swiostek, D. Letzig, H. Brokmeier, K. Kainer, Microstructure and
texture development during hydrostatic extrusion of magnesium alloy AZ31, Scripta Materialia
53(2) (2005) 259-264.
[95] G. Cano-Castillo, J. Victoria-Hernández, J. Bohlen, D. Letzig, K.U. Kainer, Effect of Ca
and Nd on the microstructural development during dynamic and static recrystallization of
indirectly extruded MgZn based alloys, Materials Science and Engineering: A 793 (2020)
139527.
[96] M.R. Barnett, Twinning and the ductility of magnesium alloys: Part II. “Contraction”
twins, Materials Science and Engineering: A 464(1) (2007) 8-16.
[97] M. Lugo, M. Tschopp, J. Jordon, M. Horstemeyer, Microstructure and damage evolution
during tensile loading in a wrought magnesium alloy, Scripta Materialia 64(9) (2011) 912-915.
[98] S.S. Park, W.N. Tang, B.S. You, Microstructure and mechanical properties of an indirect-
extruded Mg–8Sn–1Al–1Zn alloy, Materials Letters 64(1) (2010) 31-34.
[99] C. Tomé, R. Lebensohn, VISCO-PLASTIC SELF-CONSISTENT (VPSC).
[100] C. Tome, G.R. Canova, U.F. Kocks, N. Christodoulou, J.J. Jonas, The relation between
macroscopic and microscopic strain hardening in F.C.C. polycrystals, Acta Metallurgica
32(10) (1984) 1637-1653.
[101] R. Lebensohn, C. Tome, Modelling twinning in texture development codes, Textures and
Microstructures 14 (1991).
[102] F. Bachmann, R. Hielscher, H. Schaeben, Texture analysis with MTEX–free and open
source software toolbox, Solid State Phenomena, Trans Tech Publ, 2010, pp. 63-68.
[103] A. Chakkedath, A study of the effects of rare-earth elements on the microstructural
evolution and deformation behavior of magnesium alloys at temperatures up to 523K, 2016.
[104] J. Signorelli, S. Freschi, P.A. Turner, R.E. Bolmaro, SELF-CON: A Self Consistent
Software Package for Micromechanical Simulations, Materials Science Forum 408-412 (2002)
341-346.
[105] T. Ebeling, C. Hartig, T. Laser, R. Bormann, Material law parameter determination of
magnesium alloys, Materials Science and Engineering: A 527(1) (2009) 272-280.
95
[106] S.B. Yi, C.H.J. Davies, H.G. Brokmeier, R.E. Bolmaro, K.U. Kainer, J. Homeyer,
Deformation and texture evolution in AZ31 magnesium alloy during uniaxial loading, Acta
Materialia 54(2) (2006) 549-562.
[107] C. Tomé, S.R. Agnew, M.A. Bourke, W. Blumenthal, D.W. Brown, G.C. Kaschner, P.
Rangaswamy, The relation between texture, twinning and mechanical properties in hexagonal
aggregates, Materials Science Forum, Trans Tech Publ, 2002, pp. 263-268.
[108] S.R. Agnew, M.H. Yoo, C.N. Tomé, Application of texture simulation to understanding
mechanical behavior of Mg and solid solution alloys containing Li or Y, Acta Materialia 49(20)
(2001) 4277-4289.
[109] C. Davies, M. Barnett, Expanding the extrusion limits of wrought magnesium alloys,
JOM 56(5) (2004) 22-24.
[110] C.-j. Li, H.-f. Sun, W.-b. Fang, Effect of Extrusion Temperatures on Microstructures and
Mechanical Properties of Mg-3Zn-0.2Ca-0.5Y Alloy, Procedia Engineering 81 (2014) 610-
615.
[111] N.G. Ross, M.R. Barnett, A.G. Beer, Effect of alloying and extrusion temperature on the
microstructure and mechanical properties of MgZn and MgZnRE alloys, Materials Science
and Engineering: A 619 (2014) 238-246.
[112] M. Shahzad, L. Wagner, Influence of extrusion parameters on microstructure and texture
developments, and their effects on mechanical properties of the magnesium alloy AZ80,
Materials Science and Engineering: A 506(1) (2009) 141-147.
[113] G. Liu, J. Zhou, J. Duszczyk, Finite Element Simulation of Magnesium Extrusion to
Manufacture a Cross-Shaped Profile, Journal of Manufacturing Science and Engineering
129(3) (2007) 607-614.
[114] S. Yi, H.-G. Brokmeier, D. Letzig, Microstructural evolution during the annealing of an
extruded AZ31 magnesium alloy, Journal of Alloys and Compounds 506(1) (2010) 364-371.
[115] J. Victoria-Hernandez, S. Yi, J. Bohlen, G. Kurz, D. Letzig, The influence of the
recrystallization mechanisms and grain growth on the texture of a hot rolled AZ31 sheet during
subsequent isochronal annealing, Journal of Alloys and Compounds 616 (2014) 189-197.
[116] J. Victoria-Hernandez, S. Yi, D. Letzig, D. Hernandez-Silva, J. Bohlen, Microstructure
and texture development in hydrostatically extruded Mg–AlZn alloys during tensile testing at
intermediate temperatures, Acta Materialia 61(6) (2013) 2179-2193.
[117] I. Baker, Recovery, recrystallization and grain growth in ordered alloys, Intermetallics
8(9) (2000) 1183-1196.
[118] A. Imandoust, C.D. Barrett, T. Al-Samman, M.A. Tschopp, E. Essadiqi, N. Hort, H. El
Kadiri, Unraveling Recrystallization Mechanisms Governing Texture Development from Rare-
Earth Element Additions to Magnesium, Metallurgical and Materials Transactions A 49(5)
(2018) 1809-1829.
[119] W. Yuan, S.K. Panigrahi, J.Q. Su, R.S. Mishra, Influence of grain size and texture on
Hall–Petch relationship for a magnesium alloy, Scripta Materialia 65(11) (2011) 994-997.
[120] M.R. Barnett, 6 - Forming of magnesium and its alloys, in: M.O. Pekguleryuz, K.U.
Kainer, A. Arslan Kaya (Eds.), Fundamentals of Magnesium Alloy Metallurgy, Woodhead
Publishing2013, pp. 197-231.
96
[121] K. Illkova, P. Dobroň, F. Chmelík, K.U. Kainer, J. Balík, S. Yi, D. Letzig, J. Bohlen,
Effect of aluminium and calcium on the microstructure, texture, plastic deformation and related
acoustic emission of extruded magnesium–manganese alloys, Journal of Alloys and
Compounds 617 (2014) 253-264.
[122] H. Yu, Y. Xin, M. Wang, Q. Liu, Hall-Petch relationship in Mg alloys: A review, Journal
of Materials Science & Technology 34(2) (2018) 248-256.
[123] Q. Ma, H. El Kadiri, A.L. Oppedal, J.C. Baird, B. Li, M.F. Horstemeyer, S.C. Vogel,
Twinning effects in a rod-textured AM30 Magnesium alloy, International Journal of Plasticity
29 (2012) 60-76.
[124] F. Mokdad, D.L. Chen, D.Y. Li, Single and double twin nucleation, growth, and
interaction in an extruded magnesium alloy, Materials & Design 119 (2017) 376-396.
[125] I. Ulacia, N.V. Dudamell, F. Gálvez, S. Yi, M.T. Pérez-Prado, I. Hurtado, Mechanical
behavior and microstructural evolution of a Mg AZ31 sheet at dynamic strain rates, Acta
Materialia 58(8) (2010) 2988-2998.
[126] A.S. Khan, A. Pandey, T. Gnäupel-Herold, R.K. Mishra, Mechanical response and
texture evolution of AZ31 alloy at large strains for different strain rates and temperatures,
International Journal of Plasticity 27(5) (2011) 688-706.
[127] C. Ha, J. Bohlen, S. Yi, X. Zhou, H.-G. Brokmeier, N. Schell, D. Letzig, K.U. Kainer,
Influence of Nd or Ca addition on the dislocation activity and texture changes of MgZn alloy
sheets under uniaxial tensile loading, Materials Science and Engineering: A 761 (2019)
138053.
[128] S.A. Habib, A.S. Khan, T. Gnäupel-Herold, J.T. Lloyd, S.E. Schoenfeld, Anisotropy,
tension-compression asymmetry and texture evolution of a rare-earth-containing magnesium
alloy sheet, ZEK100, at different strain rates and temperatures: Experiments and modeling,
International Journal of Plasticity 95 (2017) 163-190.
[129] Zhou, C. Ha, Yi, J. Bohlen, Schell, Chi, X. Zheng, H.-G. Brokmeier, Texture and Lattice
Strain Evolution during Tensile Loading of MgZn Alloys Measured by Synchrotron
Diffraction, Metals 10 (2020) 124.
[130] C. Ma, A. Chapuis, X. Guo, L. Zhao, P. Wu, Q. Liu, X. Mao, Modeling the deformation
behavior of a rolled Mg alloy with the EVPSC-TDT model, Materials Science and Engineering:
A 682 (2017) 332-340.
[131] A. Maldar, L. Wang, G. Zhu, X. Zeng, Investigation of the alloying effect on deformation
behavior in Mg by Visco-Plastic Self-Consistent modeling, Journal of Magnesium and Alloys
8(1) (2020) 210-218.
[132] M. Lentz, M. Klaus, I.J. Beyerlein, M. Zecevic, W. Reimers, M. Knezevic, In situ X-ray
diffraction and crystal plasticity modeling of the deformation behavior of extruded MgLi
(Al) alloys: An uncommon tension–compression asymmetry, Acta Materialia 86 (2015) 254-
268.
[133] W.B. Hutchinson, M.R. Barnett, Effective values of critical resolved shear stress for slip
in polycrystalline magnesium and other hcp metals, Scripta Materialia 63(7) (2010) 737-740.
[134] N. Stanford, M.R. Barnett, Solute strengthening of prismatic slip, basal slip and {101¯2}
twinning in Mg and Mg–Zn binary alloys, International Journal of Plasticity 47 (2013) 165-
181.
97
[135] R. Sánchez-Martín, M.T. Pérez-Prado, J. Segurado, J. Bohlen, I. Gutiérrez-Urrutia, J.
Llorca, J.M. Molina-Aldareguia, Measuring the critical resolved shear stresses in Mg alloys by
instrumented nanoindentation, Acta Materialia 71 (2014) 283-292.
[136] A. Chapuis, Q. Liu, Simulations of texture evolution for HCP metals: Influence of the
main slip systems, Computational Materials Science 97 (2015) 121-126.
[137] M. Lentz, M. Klaus, R.S. Coelho, N. Schaefer, F. Schmack, W. Reimers, B. Clausen,
Analysis of the Deformation Behavior of Magnesium-Rare Earth Alloys Mg-2 pct Mn-1 pct
Rare Earth and Mg-5 pct Y-4 pct Rare Earth by In Situ Energy-Dispersive X-ray Synchrotron
Diffraction and Elasto-Plastic Self-Consistent Modeling, Metallurgical and Materials
Transactions A 45(12) (2014) 5721-5735.
[138] M. Lentz, M. Klaus, M. Wagner, C. Fahrenson, I.J. Beyerlein, M. Zecevic, W. Reimers,
M. Knezevic, Effect of age hardening on the deformation behavior of an Mg–Y–Nd alloy: In-
situ X-ray diffraction and crystal plasticity modeling, Materials Science and Engineering: A
628 (2015) 396-409.
[139] A. Chakkedath, J. Bohlen, S. Yi, D. Letzig, Z. Chen, C.J. Boehlert, The Effect of Nd on
the Tension and Compression Deformation Behavior of Extruded Mg-1Mn (wt pct) at
Temperatures Between 298 K and 523 K (25 °C and 250 °C), Metallurgical and Materials
Transactions A 45(8) (2014) 3254-3274.
[140] C.M. Cepeda-Jiménez, J.M. Molina-Aldareguia, M.T. Pérez-Prado, EBSD-Assisted Slip
Trace Analysis During In Situ SEM Mechanical Testing: Application to Unravel Grain Size
Effects on Plasticity of Pure Mg Polycrystals, JOM 68(1) (2016) 116-126.
[141] C.J. Boehlert, Z. Chen, A. Chakkedath, I. Gutiérrez-Urrutia, J. Llorca, J. Bohlen, S. Yi,
D. Letzig, M.T. Pérez-Prado, In situ analysis of the tensile deformation mechanisms in
extruded Mg–1Mn–1Nd (wt%), Philosophical Magazine 93(6) (2013) 598-617.
[142] C.J. Boehlert, Z. Chen, I. Gutiérrez-Urrutia, J. Llorca, M.T. Pérez-Prado, In situ analysis
of the tensile and tensile-creep deformation mechanisms in rolled AZ31, Acta Materialia 60(4)
(2012) 1889-1904.
[143] D.D. Yin, Q.D. Wang, C.J. Boehlert, Z. Chen, H.M. Li, R.K. Mishra, A. Chakkedath, In-
Situ Study of the Tensile Deformation and Fracture Modes in Peak-Aged Cast Mg-11Y-5Gd-
2Zn-0.5Zr (Weight Percent), Metallurgical and Materials Transactions A 47(12) (2016) 6438-
6452.
[144] A. Chakkedath, C.J. Boehlert, In Situ Scanning Electron Microscopy Observations of
Contraction Twinning and Double Twinning in Extruded Mg-1Mn (wt.%), JOM 67(8) (2015)
1748-1760.
[145] G. Zhu, L. Wang, H. Zhou, J. Wang, Y. Shen, P. Tu, H. Zhu, W. Liu, P. Jin, X. Zeng,
Improving ductility of a Mg alloy via non-basal <a> slip induced by Ca addition, International
Journal of Plasticity 120 (2019) 164-179.
98
Appendix
99
Figure A.1. Phase diagrams showing where the solubilization heat treatments were done; a) Z1
alloy, b) ZX10 alloy and c) ZNd10 alloy.
Figure A.2. Schematic representation of the samples for mechanical characterization, a) tension
tests from round bars, b) compression tests from round bars.
100
Figure A3. Grains where slip lines are developed: a) basal <a> slip, b) prismatic <a> slip, c)
pyramidal <a>, d) pyramidal I <c+a> and e) pyramidal II <c+a>. Alloy Z1 tested in tension.
The black dashed line points out the corresponding slip trace. The violet dot indicates the grain
showing the slip trace.
101
Figure A4. Grains where slip lines are developed: a) basal <a> slip, b) prismatic <a> slip, c)
pyramidal <a>, d) pyramidal I <c+a> and e) pyramidal II <c+a>. Alloy Z1 tested in
compression. The black dashed line points out the corresponding slip trace. The violet dot
indicates the grain showing the slip trace.
102
Figure A5. Grains where slip lines are developed: a) basal <a> slip, b) prismatic <a> slip, c)
pyramidal <a>, d) pyramidal I <c+a> and e) pyramidal II <c+a>. Alloy ZX10 tested in tension.
The black dashed line points out the corresponding slip trace. The violet dot indicates the grain
showing the slip trace.
103
Figure A6. Grains where slip lines are developed: a) basal <a> slip, b) prismatic <a> slip, c)
pyramidal <a>, d) pyramidal I <c+a> and e) pyramidal II <c+a>. Alloy ZX10 tested in
compression. The black dashed line points out the corresponding slip trace. The violet dot
indicates the grain showing the slip trace.
104
Figure A7. Grains where slip lines are developed: a) basal <a> slip, b) prismatic <a> slip, c)
pyramidal <a>, d) pyramidal I <c+a> and e) pyramidal II <c+a>. Alloy ZNd10 tested in
tension. The black dashed line points out the corresponding slip trace. The violet dot indicates
the grain showing the slip trace.
105
Figure A8. Grains where slip lines are developed: a) basal <a> slip, b) prismatic <a> slip, c)
pyramidal <a>, d) pyramidal I <c+a> and e) pyramidal II <c+a>. Alloy ZNd10 tested in
compression. The black dashed line points out the corresponding slip trace. The violet dot
indicates the grain showing the slip trace.