Progress in Natural Science: Materials International 2012;22(5):364–385

Chinese Materials Research Society

Progress in Natural Science: Materials International

www.elsevier.com/locate/pnsmi
www.sciencedirect.com

REVIEW

Effects of icosahedral phase formation on the

microstructure and mechanical improvement
of Mg alloys: A review
D.K. Xu, E.H. Hann

State Key Laboratory for Corrosion and Protection, Institute of Metal Research, Chinese Academy of Sciences, 62 Wencui Road,
Shenyang 110016, China

Received 4 July 2012; accepted 15 August 2012

Available online 23 October 2012

KEYWORDS Abstract Icosahedral phase (I-phase) is a relatively excellent strengthening phase in Mg alloys.
Icosahedral phase;
Depending on their volume fraction, the yield strength of Mg–Zn–Y–Zr alloys can vary from 150
Mg alloys; to 450 MPa at room temperature. Recently, the formation of I-phase has been considered as one of
Mechanical properties; the most effective methods for developing high strength lightweight Mg alloys for automotive and
Microstructural aerospace applications. In this review article, a series of research work about I-phase containing Mg
evolution alloys have been systematically investigated including I-phase formation mechanism and their
effects on mechanical properties of Mg alloys. Particular emphases have been given to: (1) Structure
of I-phase and its orientation relationship with the a-Mg matrix. (2) Influence of alloying elements
and solidification conditions on I-phase formation. (3) Effects of I-phase on microstructural
evolution and mechanical improvement of Mg–Zn–Y–(Zr) alloys. Moreover, the applications of
I-phase for the mechanical improvement of other Mg alloys such as AZ91 and super-lightweight
Mg–Li alloys are also reviewed.

& 2012 Chinese Materials Research Society. Production and hosting by Elsevier Ltd. All rights reserved.

n
Corresponding author.
E-mail address: ehhan@imr.ac.cn (E.H. Han).
Peer review under responsibility of Chinese Materials Research
Society.

1002-0071 & 2012 Chinese Materials Research Society. Production and hosting by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.pnsc.2012.09.005
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review 365

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
2. Structure of I-phase in Mg alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366
2.1. Structural investigation of I-phase in Mg–Zn–Y–(Zr) system alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366
2.2. Chemical composition of I-phase in Mg–Zn–Y–(Zr) system alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366
3. Design of I-phase containing Mg alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366
3.1. Relationship between Zn/Y ratio and the formation of I-phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366
3.2. Transformation reactions of I-phase during the solidification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368
3.3. Differential thermal analysis (DTA) and computer-aided cooling curve analysis (CA-CCA) for the formation
of I-phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
4. Severe plastic deformation of I-phase containing Mg alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
4.1. Influence of I-phase particles on the texture evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370
4.2. Effect of I-phase particles on the grain refinement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370
5. Mechanical behavior of I-phase containing Mg alloys at room temperature. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
5.1. Influence of I-phase on the tensile properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
5.2. Influence of I-phase on the impact toughness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
5.3. Influence of I-phase on the wear behavior. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
5.4. Influence of I-phase on the fatigue behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
6. Thermal stability and strengthening effects of the I-phase on Mg alloys at elevated temperature . . . . . . . . . . . . . . . 375
6.1. Thermally stability of I-phase in Mg–Zn–Y–(Zr) alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
7. Application of I-phase for the mechanical improvement of other Mg alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
7.1. Effect of I-phase on the mechanical improvement of AZ91 alloy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
7.2. Effect of I-phase on the mechanical improvement of as-extruded Mg–Li alloys . . . . . . . . . . . . . . . . . . . . . . 378
8. I-phase in other Mg alloys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
9. Future research about I-phase in Mg alloys. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381

1. Introduction
After the first discovery of a quasicrystalline phase (icosahedral
structure) by Shechtman and his collaborators in 1984 [1], much
research was performed on the formation of quasicrystals [2–23].
Among them, several icosahedral phases (I-phase) were also
found in Mg–Zn–Al and Mg–Cu–Al alloys [24–28]. However,
I-phase in Mg alloys is mostly formed by using the rapid
solidification technique. In 1993, Luo et al. found that a stable
icosahedral phase with chemical composition 76.2 at%Mg,
16.7 at%Y and 7.1 at%Zn can be formed in the as-cast
Mg–Zn–MM (misch metals containing Y, Nd, Gd, Dy, La, Pr,
Tb and Ce) alloy [29–33]. In the research of the properties of
I-phase, Pierce and Dubois reported [34,35] that quasicrystals
have many interesting properties such as high hardness, thermal
stability, high corrosion resistance, low coefficient of friction, and
low interfacial energy, etc. Following these findings, the strength-
ening effects of I-phase on mechanical properties of Mg alloys
(especially Mg–Zn–Y–(Zr) system alloys) at both room and
elevated temperatures have been widely researched [36–38].
In general, there are three kinds of ternary equilibrium
phases in Mg–Zn–Y system alloys, i.e., I-phase (Mg3Zn6Y),
W-phase (Mg3Zn3Y2, cubic structure) and Z-phase (Mg12ZnY)
[30,39]. Since the addition of zirconium has no effect on phase
composition [40], Mg–Zn–Y–Zr alloys also mainly contain
these three kinds of phases. Based on lots of reports about
Mg–Zn–Y and Mg–Zn–Y–Zr system alloys [36,37,41–47], it
suggested that element Y mainly existed in the form of I-phase
and cubic W-phase. According to the chemical formulas of
I-phase and W-phase, Zn(wt%)/Y(wt%) ratios for two phases
are 4.38 and 1.10, respectively. In the research of Mg–Zn–Y
system alloys, Lee et al. firstly found that the I-phase formation
is closely related to the Zn/Y ratio [44]. When Zn/Y ratio was
higher than 4.38 (considering a part of Zn dissolved in the Mg
matrix), the mainly formed phase was I-phase [43,44,48]. When
Zn/Y ratio was between 1.10 and 4.38, the mainly formed
phases were I-phase and W-phase [43,44,48].
Based on previous research results [43,49,50], the superior
mechanical properties of the I-phase containing Mg alloys
mainly rely on the volume fraction of I-phase and their
dispersive degree. By combining powder metallurgy and
hot extrusion processes, the fabricated I-phase containing
Mg–Zn–Y alloys can have a yield stress of 410 MPa and a
12% elongation [51]. Meanwhile, Singh et al. reported that by
using chill casting and direct extrusion processes, the yield
strength of Mg–93Zn–6Y alloy containing I-phase particles
can reach 400 MPa in both tension and compression tests,
with elongations ranging from 12% to 18% [52]. In addition,
Xu et al. indicated that the formation ability of I-phase is very
strong [53]. By introducing I-phase into Mg–Li alloys, the yield
strength and ultimate tensile strength (UTS) can be increased to
166 and 247 MPa, respectively [53]. Depending on the volume
fraction of I-phase, Mg–Zn–Y–Zr alloys can have yield strengths
of 150–450 MPa at room temperature [54]. Therefore, I-phase
strengthened Mg alloy materials can be considered as potential
candidates for wide applications in the automobile and aerospace
industries.
Generally, quasicrystals are isotropic and possess a specially
ordered lattice structure called the quasiperiodic lattice struc-
ture [39]. Due to their special structure, quasicrystals them-
selves are too brittle to be used as bulk structural materials at
ambient temperature, although quasicrystals exhibit limited
366
ductility at temperatures higher than 0.7Tm, where Tm is the
material’s absolute melting temperature [55]. Therefore, when
alloys contain quasicrystals as a second phase, they are stable
against coarsening at elevated temperatures due to their low
interfacial energy [35], providing improved bonding properties
at the I-phase/matrix interfaces.
However, in order to ensure I-phase strengthened Mg materi-
als to be used as load-bearing structural components such
as wheels, transmission housings and pedals components, the
evaluation about their fatigue properties becomes very necessary.
At present, only several studies about fatigue behavior of I-phase
containing wrought Mg alloys can be referred [53,56–58]. Since
the volume fraction of I-phase in these early investigated Mg
alloys was tiny and I-phase/a-Mg eutectic pockets can be easily
destroyed during severe plastic deformation (such as hot-extru-
sion, hot-rolling and forging), the influence of I-phase on the
fatigue behavior of Mg alloys (mainly Mg–Zn–Y–(Zr) system
alloys) is difficult to be characterized. Through investigating
fatigue behavior of an as-cast Mg–12%Zn–1.2%Y–0.4%Zr
alloy with high volume fraction of I-phase, Xu et al. successfully
disclosed the influence of I-phase/a-Mg eutectic pockets on the
fatigue crack initiation and fatigue crack propagation [37].
In this paper, the current understandings about the struc-
ture and formation mechanism of I-phase in Mg–Zn–Y–(Zr)
alloys, the effects of I-phase on microstructural evolution and
mechanical properties at both room and elevated temperatures
and applications of I-phase for the mechanical improvement
of other Mg alloys are reviewed.
2. Structure of I-phase in Mg alloys
Generally, icosahedral quasicrystals can be classified into two
types. All Al–(transition-metal) icosahedral phases have a quasi-
lattice constant a0E0.46 nm and a valence electron concentration
of about eE1.75, while the values for the Frank–Kasper-type
icosahedral phases are a0E0.52 nm and about eE2.1, respectively.
The ratios a0/d (d: average Goldschmidt atomic diameter) and e/d
in Zn–Mg–RE (RE¼Y, Tb, Dy, Ho, or Er) are calculated to be
about 1.75 (a0E0.52 nm; dE0.296 nm) and 2.08 (Zn, about 2;
Mg, about 2; RE about 3), respectively [59].
2.1. Structural investigation of I-phase in Mg–Zn–Y–(Zr)
system alloys
In the investigations on Mg–Zn–Y–(Zr) cast ingots, Luo and
Tang et al. firstly identified the Mg-rich [29] and Zn-rich [32]
I-phase by X-ray diffraction and transmission electron microscopy
(TEM). Fig. 1 shows the typical microstructure and the XRD
pattern of the as-cast Mg–5.0Zn–0.9Y–0.16Zr alloy specimen [60].
The as-cast alloy specimen exhibits dendritic solidification struc-
ture, in which the eutectic phases locate in the interdendritic
region, as shown in Fig. 1(a). To confirm the existence of I-phase,
TEM observations were performed. Fig. 1(b) and (c) show the
TEM morphology and its selected area diffraction patterns
(SADP). The SADPs taken from the eutectic lamellar phase show
the 2-, mirror, 3- and 5-fold symmetries, respectively, which are
distinctive characteristics of I-phase. The XRD pattern (Fig. 1(d))
also reveals that the as-cast alloy consists of two phases of I-phase
and the a-Mg matrix.
D.K. Xu, E.H. Han
Since the nature of the interface between the particle and
matrix is one of the factors in determining mechanical properties
of the alloy, a high resolution TEM investigation of the interface
region between an I-phase particle and the a-Mg matrix in a hot-
rolled Mg–Zn–Y alloy was conducted by Bae et al. [61]. The
results indicated that an interface layer of 3–5 nm thick a-Mg can
still preserve the orientation relationship with the I-phase, and
coherency between I-phase and the a-Mg can be achieved by
introducing steps and ledges periodically along the interfaces [46].
It also suggests that the atomic bonding between the I-phase and
the hexagonal structure is rigid enough to be retained during
severe plastic deformation [61]. Previous research indicated that
when the basal slip meets the I-phase/a-Mg matrix eutectic
pockets, it will be stopped and no cracks can form along the
I-phase/a-Mg matrix interfaces [42].
2.2. Chemical composition of I-phase in Mg–Zn–Y–(Zr)
system alloys
In previous research, Tang et al. has investigated the composi-
tion profiles of I-phase and surrounding matrix in an as-cast
Mg–Zn–Y–Zr alloy by using analytical transmission electron
microscopy [62]. It indicated that the I-phase existing along
the grain boundary appeared in a range of compositions
(67.0–76.3 wt%Zn, 5:1–9.8 wt%Mg, 9.9–13.4 wt%Y and
5.9–9.8 wt%Zr). The analytical solute concentrations in the
I-phase and surrounding matrix are variable.
Fig. 2(a)–(c) are the TEM morphology of I-phase and the
results of microanalysis from left to right across the I-phase,
respectively [62]. The composition of the I-phase has a maximum
deviation of 9 wt%Zn, 5 wt%Mg, 3 wt%Y and 3 wt%Zr.
We can see an obvious variation of solute contents in the I-phase.
Line A in Fig. 2(a) indicated the grain boundary direction along
which I-phase existed, while line B is the trace along which the
solute profiles of the I-phase were measured. From points 1 to 8
and 9, concentration profiles across the line B show an increase in
Mg as Zn is decreasing. There is a maximum in Mg and a
minimum in Zn at points 8 and 9, respectively. After that, the
content of Mg decreased as Zn increased rapidly. On the other
hand, elements Y and Zr did not change greatly even though there
is a minimum at point 7 and a maximum at point 9, respectively.
It is worth to be noted that the solute contents at point 14 are
greatly different compared with point 13, because the detected
position is very close to the matrix. In Mg alloys, element Zn in a
Zn-rich I-phase tends to diffuse into the matrix, and element Mg
tends to aggregate in the I-phase. This may suggest that the
I-phase was first nucleated along the grain boundary, and then
began to grow into the matrix while the solute diffused during or
after solidification. When growing, the I-phase also extended
along line C. Thus, it is reasonable that Mg content is maximum
and Zn content is minimum between points 8 and 9. The variation
of elements Y and Zr is very slight because they have lower
diffusion coefficient in Mg and small content in the I-phase.
3. Design of I-phase containing Mg alloys
3.1. Relationship between Zn/Y ratio and the formation
of I-phase
Based on lots of reports about Mg–Zn–Y and Mg–Zn–Y–Zr
system alloys [36,41–47,63], it indicated that the secondary
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review 367

Fig. 1 Typical microstructures and XRD pattern of the as-cast Mg–5.0Zn–0.9Y–0.16Zr alloy. (a) SEM image, (b) TEM micrograph of
eutectic I-phase (bright field image), (c) SADP from I-phase exhibiting the 2-fold, mirror fold, 3-fold and 5-fold symmetry and (d) XRD
pattern [60].

Fig. 2 TEM photograph of the I-phase (a) and the composition profiles across the I-phase along line B (b) and (c) [62].

phases mainly include I-phase (Mg3Zn6Y) and cubic W-phase alloys [44]. It indicated that the phases identified in the as-cast
(Mg3Zn3Y2). Lee et al. has systematically investigated the Mg–Zn–Y alloys were a-MgþMg7Zn3, when Zn/Y ratio
influence of Zn/Y ratio on the microstructure of Mg–Zn–Y was 10; a-MgþI-phase when Zn/Y ratio was 5–7; and
368 D.K. Xu, E.H. Han

a-MgþI-phaseþW-phase, when Zn/Y ratio was 1.5–2. The
summary of the relationship of phase formation and Zn/Y ratio
in Mg–Zn–Y–(Zr) alloys has been listed in Table 1 [42–44,47,49].
Therefore, the I-phase formation in Mg–Zn–Y and Mg–Zn–Y–
Zr system alloys is closely dependent on the Zn/Y ratio.
3.2. Transformation reactions of I-phase during the
solidification
Although I-phase existed in Mg–Zn–Y alloys with an especial
morphology and different growth tropism, its formation
process meets the nucleation and growth rules. The research
[64] showed that an eutectic reaction occurred firstly when the
temperature of the melt was reduced to 550 1C:
L-a-MgþMg3Zn3Y2 (1)
Here, Mg3Zn3Y2 is W phase. With the temperature decrease of
the melt, I-phase/a-Mg eutectic pockets will be formed by a
peritectic reaction at 448 1C:
LþW-a-MgþI-phase (2)
when Y content in the melt is low for Mg72Zn26.5Y1.5 and
Mg73Zn24.2Y2.8 alloys, I-phase and a-Mg eutectic pockets will
be generated by an eutectic reaction at 400 1C:
L-a-MgþI-phase (3)
With further temperature decrease of the melt, Mg7Zn3
phase will be generated by an eutectic reaction at 340 1C:
L-a-MgþMg7Zn3 (4)
The MgZn phase was eventually generated by an eutectoid
reaction at 330 1C:
Mg7Zn3-a-MgþMgZn (5)
Since the binary eutectic point between a-Mg and Mg7Zn3
(Eq. (4)) is very close to the Mg7Zn3 phase zone and the
eutectoid point (Eq. (5)), when Y content is lower (Mg72Zn26.5
Y1.5 alloy), the main phase solidified after the formation of
primary phase is almost single Mg7Zn3 phase. Although Mg7Zn3
phase is stable at more than 325 1C, Eq. (5) normally occurs for
the alloy with higher Y content (Mg73Zn24.2Y2.8 alloy). Accord-
ing to the classical nucleation theory, the occurrence of Eq. (2)
must overcome the activation energy DG [65]:
16s3sl
DGn ¼ f ðyÞ ð6Þ
3DG2
where DG is free energy difference and the driving force
of the phase transformation; ssl is interfacial energy of solid–
liquid two phases and the resistance of the phase transformation;
f(y) is contact factor. In an isotropy nucleation system, f(y)¼1.
According to Spaepen model [66]:
DSf
ssl ¼ as T ð7Þ
ðNA Vm2 Þ1=3
where DSf is melting entropy, NA is Avogadro’ number,
Vm is molar volume, T is temperature, and as is a parameter
correlating with crystal nucleus structure. The resistance of the
phase transformation depends on the type of crystal nucleus
structure in a specific temperature. The value of as is 0.36 in
an icosahedral structure and 0.86 in a face-centered cubic
structure [67].
The solid solubility of Zn and Y in the a-Mg matrix is very
low and reduces with the temperature decrease of the melt.
When Zn and Y enriches at the solidifies front of a-Mg and
the constitutional super-cooling generated, more a-Mg rami-
fication forms at the solidified front, and a layer of active
surface film containing Zn and Y forms between W phase and
the liquid, so that the interfacial energy of solid–liquid phase (ssl)
decreases. Consequently, the phase transformation resistance of
Eq. (2) decreases. The activation energy (DG) (Eq. (6)) that
must be overcome is lower. Therefore, I-phase is easy to be
nucleated during slow cooling [68].

Table 1 Summarization of the relationship between Zn/Y ratio and phase formation in Mg–Zn–Y–(Zr) system alloys
[42–44,47,49].

Alloy no. Chemical composition (in wt%) Zn/Y ratio Main phases

Mg Zn Y Zr

1 Bal 4 0.4 – 10 a-Mg and Mg7Zn3 [44]

2 – 8 0.8 – 10
3 – 3 0.6 – 5 a-Mg and I-phase [44]
4 – 8 1.6 – 5
5 – 4 2 – 2 a-Mg and W-phase [44]
6 – 4 2.2 – 1.8
7 – 5.53 1.08 0.83 5.12 a-Mg and I-phase [42,43,47]
8 – 5.69 0.92 0.80 6.12
9 – 5.43 1.17 0.68 4.64
10 – 5.54 0.74 0.42 7.48
11 – 5.64 1.97 0.73 2.86 a-Mg, I-phase and W-phase [42,43,47]
12 – 5.49 3.08 0.82 1.78
13 – 5.47 3.69 0.86 1.48
14 – 5.55 1.72 0.37 3.27
15 – 3.24 3.34 0.67 0.97 a-Mg and W-phase [49]
16 – 5.95 6.08 0.64 0.98
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review 369

3.3. Differential thermal analysis (DTA) and computer-aided

cooling curve analysis (CA-CCA) for the formation of I-phase

In the research of two Mg–Zn–Y system alloys, Bae et al. [61] has
investigated the phase composition and I-phase transformation
temperatures in Mg97.8Zn2Y0.2 and Mg95Zn4.3Y0.7 (in atomic
percentage) alloys, respectively. Differential thermal analysis
(DTA) indicated that during continuous heating at a heating rate
of 0.67 K/s, the first endothermic peak appeared with an onset
temperature of 440 1C for both alloys corresponds to the melting
of the I-phase/a-Mg eutectic pockets in the interdendritic region of
the alloys. Compared with other DTA results [38,45,47,61,63], it
proves that the transformation temperature of I-phase/a-Mg
eutectic pockets is about 440–450 1C.
Compared with DTA, computer-aided cooling curve analysis
(CA-CCA) can more precisely reveal characteristic temperature
points of phase transformations during practical solidifica-
tion processing. Therefore, the authors firstly carried out the
CA-CCA analysis for two Mg–Zn–Y–Zr alloys with Zn/Y
ratios of 5 and 10, respectively [69]. To effectively analyze and
compare the main phase transformation reactions of two alloys,
samples for thermal analysis were cut from the same position
of the cast ingots and machined into cylinders of 25 mm in
diameter and 50 mm in length. The samples were re-melted in a
steel crucible in an electrical resistance furnace, and then two
shielded K-type thermocouples were immersed from the top of
the crucible that was insulated from the top and the bottom.
One thermocouple was placed at the center of the crucible (Tc)
and the other close to the inner wall (Tw) located at exactly the
same depth in the melt (25 mm from the bottom). After holding Fig. 3 Thermal analysis results of the as-cast Mg–Zn–Y–Zr alloys
10 min at 720 1C, the crucible together with two thermocouples with Zn/Y ratio of: (a) 5 and (b) 10 [69].
were removed from the furnace and allowed cooling in air with
a cooling rate of about 1 K/s. The temperature changes were Therefore, for the alloy with Zn/Y ratio of 5, I-phase
continuously recorded during the solidification process by using formation should depend on W-phase transformation and
a high-speed data acquisition system linked to the computer. eutectic reaction of the residual melt. In the research of the
Fig. 3 shows the thermal analysis results of two I-phase ternary phase diagram of Mg–Zn–Y alloys, Langsdorf et al.
containing Mg–Zn–Y–Zr alloys with Zn/Y ratios of 5 and 10 [70] suggested that when Y content was lower than 4 at%,
[69]. The first derivative of the cooling curve (dTc/dt) was I-phase in Mg60–3yZn40þ2yYy and Mg50–2yZn50þyYy alloys
determined to enhance slope changes that are related to the can directly form from the melt. However, recent works
solidification reactions for different phases, and to facilitate [42–44,47,49] have firmly indicated that Zn/Y ratios (in wt%)
the determination of the critical solidification characteristics of were closely related with the phase formation in Mg–Zn–Y
two alloys. and Mg–Zn–Y–Zr system alloys, as listed in Table 1. Through
For the alloy with Zn/Y ratio of 5, four well-defined peaks calculating the Zn/Y ratios and analyzing the reactions for
are observed at 634.6, 513.5, 475.7 and 440.6 1C, as shown in I-phase formation of the Mg–Zn–Y alloys [60], it can further
Fig. 3(a). On the other hand, only three well-defined peaks can confirm that to make I-phase directly precipitate from the melt,
be observed at 613.8, 426.3 and 338.8 1C for the alloy with the Zn/Y ratio must be higher than 10. Therefore, for the alloy
Zn/Y ratio of 10, as shown in Fig. 3(b). Compared with the with Zn/Y ratio of 10, the formation of I-phase should solely
cooling curves of two alloys, it suggests that the reactions for rely on the eutectic reaction from the melt. To conveniently
phase formation are closely related with the Zn/Y ratio. describe and compare with the phase transformation of two
According to the differential thermal analysis (DTA), the alloys, the main phase reactions corresponding to the char-
melting temperature associated with W-phase is about 510 1C acteristic points (Fig. 3) have been listed in Table 2 [69]. It
[45,47], the melting temperature of I-phase eutectic pockets is firmly suggests that Zn/Y ratio is one of the key factors in
about 440–450 1C [36,45,47,61,63] and the end of formation controlling reactions for I-phase formation.
temperature of Mg7Zn3 phase is 325 1C [70]. From the cooling
curve, it indicates that there is another peak (475.7 1C) existing
between 440.6 and 513.5 1C for the alloy with Zn/Y ratio of 5.
Then, a transformation reaction should be existed between 4. Severe plastic deformation of I-phase containing Mg alloys
W-phase and I-phase at 475.7 1C. According to the Mg–Zn–Y
ternary phase diagram [71], W-phase can transform into Recently, significant interests have been focused on the develop-
I-phase by the following reaction: ment of wrought Mg alloys. However, due to hcp structure of
Mg, strong (0 0 0 2) basal plane texture is generally developed
LþW¼ MgþI; (8) during hot or cold deformation processing, significantly affecting
370
Table 2 Main reactions during solidification identified by thermal analysis for the Mg–Zn–Y–Zr alloys with Zn/Y ratio of 5
and 10 [69].
Mg–6%Zn–1.2%Y–0.4%Zr
ID no. Transformation reactions
1 L-a-Mg (liquidus)
2 Maximum growth of primary a-Mg phase
3 L-a-MgþW-phase (eutectic reaction)
4 LþW-phase-a-MgþI-phase (W-phase transformation reaction)
5 L(Residual)-a-MgþI-phase (eutectic reaction)
6 End of solidification (solidus)
on the mechanical properties, especially at low temperatures
[72–77]. It has been shown that by adopting processing techni-
ques such as equal channel angular processing and asymmetry
rolling, the texture softening occurs, resulting in the higher
ductility [78–81].
Compared with AZ and AM system alloys, Mg–Zn–Y–(Zr)
alloys reinforced with I-phase particles have been reported to
exhibit a better combination of elongation and strength
[41,44,61]. In the following parts, the effects of I-phase
particles on the texture evolution and grain refinement of
wrought Mg–Zn–Y–(Zr) alloys will be deeply reviewed.
4.1. Influence of I-phase particles on the texture evolution
To understand the effect of I-phase particles on the texture
development of Mg–Zn–Y alloys, Lee et al. has investigated and
compared the textures of hot-rolled Mg–3 wt%Zn–0.6 wt%Y
and Mg–6 wt%Zn–1.2 wt%Y alloys (with different volume
fraction of I-phase) by means of X-ray diffraction (XRD) and
electron back-scattered diffraction (EBSD) techniques [82].
Previous work [44] indicated that the microstructures of two
investigated alloys are composed of primary a-Mg dendrite and
interdendritic I-phase/a-Mg eutectic pockets. Moreover, the
volume fraction of I-phase increased with increasing solute
content, i.e., 1.2% in Mg–3 wt%Zn–0.6 wt%Y alloy and 2.9%
in Mg–6 wt%Zn–1.2 wt%Y alloy, respectively [44].
Fig. 4(a) shows the result of the macro-texture analysis by
XRD obtained from the Mg–3 wt%Zn–0.6 wt%Y alloy [82].
Egg-shaped distribution of (0 0 0 2) plane pole indicated that
the strong basal plane texture has been developed during
rolling and anneal treatment. Mg–6 wt%Zn–1.2 wt%Y alloy
exhibited the similar type of strong basal plane texture. To
compare the basal plane texture quantitatively, the basal plane
pole intensity was measured and shown in Fig. 4(b) and (c)
as a function of tilt angle (Phi) from the normal direction towards
the rolling direction (Fig. 4(b)) and towards the transverse
direction (Fig. 4(c)), respectively [82]. In Mg–3 wt%Zn–0.6 wt%Y
alloy the maximum intensity appeared at normal direction (01),
while in Mg–6 wt%Zn–1.2 wt%Y alloy the maximum intensity
appeared at 51 away from the normal direction particularly
when the intensity was measured from the normal direction to the
rolling direction. Moreover, the maximum intensity is much lower
in Mg–6 wt%Zn–1.2 wt%Y alloy (particularly in Fig. 4(c)),
indicating that the angular distribution of (0 0 0 2) basal plane
poles becomes broader by increasing the amount of I-phase
particles.
In general, the textures of hcp metals are determined by
their c/a ratio as well as by the active slip systems or twinning
D.K. Xu, E.H. Han
Mg–12%Zn–1.2%Y–0.4%Zr
ID no. Transformation reactions
1 L-a-Mg (liquidus)
2 Maximum growth of primary a-Mg phase
3 L-a-MgþI-phase (eutectic reaction)
4 L(Residual)-a-MgþMg7Zn3 (eutectic reaction)
5 End of solidification (solidus)
during deformation [83]. The basal plane textures are common
for hot-rolled Mg alloys. This is due to the activation of {10–
12} twinning, which reorients the c-axes perpendicular to the
rolling plane [84]. Subsequent recrytallization would lead to
the formation or strengthening of a basal fiber texture [85].
Moreover, it has been reported that the I-phase in Mg–Zn–Y
alloy can have various orientation relationships with a-Mg
grains [86,87], which may play a role in randomizing the
texture of Mg–Zn–Y alloy. Therefore, it suggested that the
relatively random texture in the Mg–6 wt%Zn–1.2 wt%Y
alloy was attributed to the larger fraction of I-phase particles.
It is well-known that on deformation by a shear strain, if the
particles does not deform, there is a local strain incompat-
ibility between particles and the matrix. This can be accom-
modated by the generation of dislocation at the particles–
matrix interfaces. It has been reported that in alloys with
larger strains and larger particles, more complex dislocation
structures are formed on deformation and are associated with
local lattice rotations close to the particles [88]. Generally, the
strain rate is strongly related with the dislocation density and
increases with dislocation velocity. For mobile dislocations,
they are a strong function of stress [89]. For this consideration,
it is thought that a more complex dislocation structure can
be developed in the region close to particles with increase of
rolling speed, resulting in an increment of maximum misor-
ientation of local lattice rotations. Since I-phase exhibits
attractive properties such as high strength, hardness, low
interfacial energy etc. [90,91], there may be extra effects from
I-phase particles on the texture development when compared
with other types of particles. As a result, it suggests that the
relatively random texture in Mg–Zn–Y alloys is attributed to
the effect of I-phase particles, i.e., its unique properties and
local lattice rotations around I-phase particles [82].
4.2. Effect of I-phase particles on the grain refinement
Generally, the mechanical properties of wrought Mg alloys are
superior to those of the as-cast Mg alloys, because the former
has a finer grain structure [92]. For wrought Mg alloys, finer
grains with the size of less than 10 mm can be easily obtained
just through primary processing such as hot rolling or
extrusion [93,94]. To understand the effect of I-phase on grain
refinement, the microstructure of as-extruded Mg–Zn–Y–Zr
alloys (with Y contents of 0, 1.08, 1.97 and 3.08 wt%) has been
investigated [46]. The chemical compositions of the alloys I to IV
were determined and listed in Table 3 [46].
Based on the previous results [43], the main phases in alloy
II are I-phase and a-Mg, whereas the main phases in alloys III
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review 371

Fig. 4 (a) Pole figure of (0 0 0 2) plane obtained from Mg–3 wt%Zn–0.6 wt%Y alloy: (b) and (c) comparison of (0 0 0 2) pole intensity
plotted as a function of tilt from the sheet normal direction toward; (b) rolling or (c) transverse direction [82].

is very slight. It firmly indicates that the grain-refinement

Table 3 Chemical composition of the as-extruded
effect of I-phase is superior to that of W-phase. A zone of
Mg–Zn–Y–Zr alloys [46].
higher stress concentration may form near a particle if the
Nominal alloy Composition (wt%) particle is harder compared to the matrix. Quasicrystalline
phases are known for their high hardness. Moreover, it has
Mg Zn Y Zr Zn/Y been reported [38,95,96] that I-phase could form strong
interfaces with the matrix in various orientations, which made
Alloy I Bulk 5.68 0 0.78 –
I-phase particles having a strong pinning effect on the
Alloy II Bulk 5.53 1.08 0.83 5.12
Alloy III Bulk 5.64 1.97 0.73 2.86 dislocation movement. On the other hand, it has been reported
Alloy IV Bulk 5.49 3.08 0.82 1.78 that W-phase is cubic structure [97,98]. Therefore, during the
hot deformation process, the zone of high dislocation density
near I-phase particles is greatly larger than that near W-phase.
and IV are I-phase, W-phase and a-Mg. Moreover, with the Humphreys et al. [99] indicated that recrystallization origi-
increase of Y content, the diffraction peak of W-phase will be nated within a zone of high dislocation density and large
gradually intensified. The microstructure observations of the lattice misorientation at the particles. Therefore, the ‘‘partial
as-extruded Mg–Zn–Y–Zr alloys with different Y contents are stimulated nucleation effect’’ of I-phase is stronger during
shown in Fig. 5 [46]. It shows that the grain size decreases with recrystallization, resulting in finer grains.
the addition of element Y. For alloy I (ZK60), the grain size is However, the result also indicates that with the increasing
about 15 mm, whereas the grain size of alloys II and III (with of Y content, the quantity of W-phase increases and then the
Y contents of 1.08 and 1.97 wt%, respectively) is about 8 mm. grain-refinement effect of W-phase is improved, which makes
When Y content reaches to 3.08 wt%, the grain size of alloy the change in grain size is not much for alloys II and IV [46].
IV is further refined, with the size about 2–4 mm. Although the Therefore, it suggests that element Y can effectively refine
grain size of as-cast alloys I to IV is gradually decreased with grains of the Mg alloys and the refinement effects of element Y
the increasing of Y content [42], the variation of grain size for is mainly ascribed to the following factors: (1) Rare earth
the as-extruded alloys II and III does not meet this trend. With element Y can change solution degree of Zn, which decreases
Y content increasing almost two times, the change in grain size the solidus curve and shortens the time for nucleation, and
372
Fig. 5 Optical micrographs observed on the L–T plane in the samples: (a) alloy I, (b) alloy II, (c) alloy III and (d) alloy IV [46].
then reduces the grain size [100]. (2) The formation of MgZnY
phases, especially I-phase can effectively restrain the grain
growth during the recrystallization process [47]. In addition,
Zr can refine the grains by acting as nucleus during solidifica-
tion [101]. However, compare with the refinement factors
mentioned above, severe plastic deformation and recrystalliza-
tion are the primary sources for grain refinement of wrought
Mg alloys. Therefore, although no Y content is contained in
alloy I, its grains are also refined greatly.
5. Mechanical behavior of I-phase containing Mg alloys at
room temperature
It has been reported that the interface layer of a-Mg with a
3–5 nm thickness still preserved the orientation relationship
with I-phase and the coherency between I-phase and a-Mg
could be achieved by introducing steps and ledges periodically
along the interface [61]. Therefore, the atomic bonding of
I-phase/a-Mg interface was rigid enough to be retained during
severe plastic deformation. It has been reported that depend-
ing on the volume fraction of I-phase, Mg–Zn–Y–Zr alloys
could have yield stress ranging from 150 to 450 MPa at room
temperature [54]. Moreover, in the research of microstructure
and compressive deformation behavior of EX-ECAP pro-
cessed (sequential application of extrusion and ECAP)
Mg–Zn–Y–Zr alloy, Zheng et al. indicated that I-phase particle
can effectively improve the compressive properties of I-phase
D.K. Xu, E.H. Han
containing Mg alloys [102]. Therefore, the mechanical improve-
ment of the as-extruded alloy with Zn/Y ratio of 10 can be
mainly ascribed to the larger amount of I-phase. In the following
sections, the influences of I-phase on tensile, impact toughness,
wear and fatigue behaviors will be reviewed.
5.1. Influence of I-phase on the tensile properties
In the previous research, Lee et al. has comprehensively
disclosed the strengthening effect of I-phase and the relationship
between I-phase volume fraction and mechanical improvement
of Mg–Zn––Y system alloys [44]. It firmly indicated the volume
fraction of I-phase on the mechanical properties of the alloys
with zinc content of 4 wt%. When I-phase volume fraction
increased slightly from 2.2% to 2.3%, the yield strength (YS)
and ultimate tensile strength (UTS) of Mg–4%Zn–0.6%Y and
Mg–4%Zn–0.8%Y increased slightly from 132 and 240 MPa to
142 and 245 MPa, respectively. On the other hand, the total
elongation to failure decreased slightly from 29.5% to 27.6%.
The results clearly showed that with increasing the volume
fraction of the I-phase, the YS and UTS of alloys increased
from 122 and 226 MPa (Mg–3%Zn–0.6%Y) to 169 and
270 MPa (Mg–8%Zn–1.6%Y), respectively. On the other hand,
the total elongation to failure only decreased from 30.2% to
26.9% [44].
In particular, the I-phase containing alloys exhibited very
high elongation to failure and its value can exceed 25%. The
results indicated that the alloys with a range of a combination
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review
of strength and ductility can be designed by changing the
volume fraction of the I-phase, i.e., by changing the total
solute content of Zn and Y. The strengthening mechanism
with increasing of the volume fraction of I-phase can be
explained by the dispersion hardening due to the presence of
dispersive I-phase particles [44]. Generally, elongation tends to
be low for alloys containing a large number of intermetallic
particles because dislocations are easily accumulated near hard
particles, resulting in the decohesion from the matrix [89].
However, no debonding or micro-scale defects at the I-phase/
a-Mg matrix interface are observed in the tested specimens,
which are mainly ascribed to the coherency structure of the
interfaces [61]. When quasicrystals exist in the a-Mg matrix
as a second phase in Mg–Zn–Y alloys, the quasicrystals are stable
against coarsening during thermomechanical process such as
hot rolling, due to the low interfacial energy of the quasicrystals
with the a-Mg matrix [35]. Therefore, the quasicrystal reinforced
Mg–Zn–Y alloys exhibit a good plasticity after thermomechanical
process.
Singh et al. reported that by using chill casting and direct
extrusion process [52], the tensile yield strengths Mg–6Zn–1Y
alloy extruded at temperatures of 390, 300 and 250 1C are
268.372, 340.573 and 386.372.8 MPa, respectively. The com-
pressive yield strengths for extrusions at temperatures of 390, 300
and 250 1C are 252.970.7, 342.379.0 and 354.678.0 MPa,
respectively. Fig. 6 shows tensile true strain-true stress curves
for both alloys extruded at 260, 300, 350 and 400 1C and tested
at room temperature. The yield stress in both alloys is always
higher for the lower extrusion temperature [51]. Particularly, for
the MgZn3.3Y0.43 alloy extruded at 260 1C, significant tensile
properties have been reached with a yield stress of 410 MPa
and an elongation of 12%. Furthermore, the MgZn3.3Y0.43 alloy
always exhibits the higher yield stress values. However, at
400 1C, both alloys present similar yield stress values.
5.2. Influence of I-phase on the impact toughness
For structural applications, their mechanical properties must
satisfy both reliability and safety requirements. One criterion of
these is to investigate their fracture toughness. Although several
literatures can be referred about the fracture toughness of Mg
and Mg alloys [103–105], the value of fracture toughness in Mg
alloys has generally been reported to be lower than that of Al
alloys. Thus, some works about the improved methods for the
fracture toughness of Mg and Mg alloys have been done. It has
Fig. 6 Tensile stress–strain curves of MgZn2.1Y0.26 and MgZn3.3
Y0.43 alloys extruded at 260, 300, 350 and 400 1C [51].
373
been reported that the fracture toughness of Mg [106] and
Mg alloys [107,108] can be improved by grain refinement. From
microstructure observations of fracture toughness tested
Mg–Al–Zn alloy samples, it is observed that grain refinement
helps (i) to prevent activation of deformation twins, which can
act as the origin of micro-cracks, and (ii) to activate dislocations
on basal as well as non-basal planes [109,110]. In addition,
dispersion of fine spherical-shaped precipitates can enhance the
fracture toughness by pinning dislocation movements [111].
Therefore, controlling the grain refinement and/or the presence
of precipitates in the matrix are currently effective methods for
the fracture toughness improvement of Mg and Mg alloys.
Generally, I-phase possesses five-fold symmetry and a
quasi-periodic structure, which is very different from crystal-
line phases. In addition, this kind of phase occurs in equili-
brium with the Mg matrix [112], showing a definite orientation
relationship and with a strong interface [113], and has been
studied as a very excellent strengthening phase in Mg alloys
[60,96,110,114]. I-phase can be dispersed as fine particles in the
matrix by severe plastic deformation such as rolling [61,63]
or extrusion [38,96]. Therefore, these alloys exhibited high
strength with good ductility at room temperature and a highly
stable microstructure at elevated temperatures. To indicate the
effect of I-phase particles on the fracture toughness of Mg
alloys, Somekawa et al. firstly carried out the fracture tough-
ness research of an I-phase particle dispersed extruded
Mg–Zn–Y alloy [115].
The relationship between the plane-strain fracture tough-
ness and specific strength (yield strength divided by density)
is shown in Fig. 7 [115]. This figure includes the data for
Fig. 7 Relationship between specific strength and fracture tough-
ness of Al alloys and Mg alloys [103,104,115–124].
374
conventional Al and Mg alloys [103,104,115–124]. The present
I-phase containing Mg–Zn–Y alloys have a good balance
of strength and toughness, and (i) shows a higher strength-
toughness combination than conventional Mg alloys, and
(ii) is comparable with conventional high strength Al alloys.
The main reason for the superior mechanical properties of
these Mg alloys is a combination of fine grain structure and
the existence of I-phase, which forms strong interfaces with the
matrix, acting as a second phase in the matrix and pinning
dislocation movements. Thus, controlling interfaces between the
matrix and the particles is an effective method for improving the
fracture toughness of Mg alloys.
5.3. Influence of I-phase on the wear behavior
Quasicrystals are plastically deformable at the temperatures
of about the half of the melting point [125], but at low
temperatures most quasicrystals show high hardness and
low friction coefficient [126]. This combination of properties,
i.e., high hardness/low friction coefficient, can be used to
improve the wear resistance of materials. Zhang et al. has
systematically investigated the friction and wear behavior of
I-phase containing Mg materials and discussed their wear
mechanisms [127].
The friction and wear behavior of Mg–Zn–Y alloys were
evaluated using a MM-200 model block-on-wheel friction and
wear testing machine under dry sliding condition and at
room temperature. During the tests, the specimens with 14 
10  10 mm3 in size were forced against a rotating GCr15 steel
wheel (HRC6575). Prior to the tests, the specimens were
polished to a surface roughness (Ra) of 4–5 mm and cleaned
with ultrasonic wave in alcohol. The sliding velocity was
0.42 m/s. The applied loads were 10, 30, 50 and 70 N,
respectively. The sliding time was 10, 20, 30 and 40 min,
respectively. The wear mass loss of the sample was measured
using an electronic balance of 0.01 mg accuracy. The friction
coefficient of the specimen, m, was calculated using the
following equation:
T considered as potential candidates for many applications in
m¼ ð9Þ
RP
where T is the friction torque, R the radius of the steel wheel, and
P is the applied load. Every test was carried out repeatedly three
times, and the average of the three data was acted as the reported
datum.
The variations of the friction coefficients of the Mg–Zn–Y
quasicrystal materials with the sliding time at different applied
loads are shown in Fig. 8 [127]. It revealed that the friction
coefficients decreased with increasing applied loads for all
alloys. Some studies [128–130] showed that two hard materials
in contact lead to the formation of small particles by fracture.
These particles were picked up again and gradually formed
a film on the wear surfaces. Thus, the contact of dissimilar
materials was transformed to a contact of similar materials,
which resulted in the increase of the friction coefficient. At a
low applied load, the contact temperature was low and the
hardness of alloys could be retained. Therefore, the higher
friction coefficients could be obtained when compared with
high applied loads.
The friction coefficients decreased initially with increasing Y
content. When Y content reached 2 at%, the friction coefficient
achieved the lowest value. Then, the friction coefficients increased
D.K. Xu, E.H. Han
with the further increase of Y content. Quasicrystal phases have
low friction coefficient [125,131] and high brittleness. With the
increase of Y content, the volume fraction of quasicrystal phase
increases. When Y content was less than 2 at%, the low friction
coefficient of quasicrystal phase played a key role in reducing the
friction coefficient of quasicrystal materials. So, the friction
coefficients reduced initially with increasing Y content. When Y
content was more than 2 at%, the brittleness of the quasicrystal
materials increased with the volume fraction and the size of I-
phase particles. Some small pieces of the specimen were easy to be
removed from the worn surface by the impact and shear stress
during the sliding. Therefore, the worn surface became coarse, and
the friction resistance increased. As a result, the friction coefficient
increased again.
5.4. Influence of I-phase on the fatigue behavior
Recently, majority of Mg alloy structural components are
produced by casting processes such as sand casting, gravity
die casting (permanent mold casting), squeeze casting and die
casting, etc. It has been reported that casting process covered
almost 90% of the total Mg alloys production [132]. In as-cast
Mg alloys, defects such as casting porosity and cavity are
usually present and the fatigue properties are significantly
affected by their shape and dimension [133,134]. In the
research of the fatigue properties of as-cast and as-extruded
AZ61 Mg alloy [132], it suggested that during the cyclic
loading, the casting defects mainly served as stress concentra-
tion sites for the fatigue crack initiation. In addition,
Horstemeyer et al. reported that fatigue cracks preferentially
initiated at pores and oxide inclusions for the die-casting
AZ91E-T4 [135]. However, previous researches mainly
focused on the fatigue behavior of Mg–Al–Zn [132,133,
136–139] and Mg–Al–Mn [140–142] system alloys. Compared
with these two system Mg alloys, Mg–Zn–Y–Zr alloys have
higher strength [45,143] and the highest tensile strength can
reach 380 MPa [144].
Therefore, I-phase strengthened Mg materials can be
the fields of automobile and aerospace. However, for the
applications to the load-bearing components, the reasonable
evaluation about the fatigue properties of the I-phase contain-
ing Mg–Zn–Y–Zr Mg alloys becomes very necessary. At
present, only several studies about the fatigue behavior of
wrought Mg–Zn–Y–Zr alloys can be referred [56,58,145]. In
this review, the influence of casting defects and I-phase on
the fatigue crack initiation and propagation for an I-phase
containing as-cast Mg–12%Zn–1.2%Y–0.4%Zr Mg alloy will
be described [37].
S–N curve of the alloy is shown in Fig. 9 [37]. Since the
fatigue limit does not exist in non-ferrous alloys [146], the
fatigue tests were continued until 1  107 cycles. When the
strength amplitude (s) is lower than 45 MPa, fatigue failure
did not occur until 1  107 cycles. In addition, due to the
existence of a pronounced horizontal asymptote between
105 and 107 cycles, the present alloy seems to possess a
definite fatigue limit. Similar fatigue behaviors have also been
observed in the as-extruded AZ61 and rolled AZ31 Mg alloys
[138,146]. Mayer et al. indicated that the non-ferrous alloys
such as Mg and Al alloys could have fatigue limits, which
was caused by the existence of relatively large porosity [134].
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review
Fig. 8 Variations of the friction coefficients vs. the sliding time at different applied loads: (a) Mg–25Zn–1Y, (b) Mg–25Zn–1.5Y,
(c) Mg–25Zn–2Y, and (d) Mg–25Zn–2.5Y [127].
For similar alloys without porosity, such as vacuum die-cast
AZ91 hp or defect free gravity die-cast AZ91 hp, no fatigue
limit was found [147]. The mean fatigue strength of the present
alloy is 4575 MPa. Since the tensile strength (sT) of the alloy
is 216 MPa, the fatigue ratio (s/sT) at 107 cycles is about 0.21,
indicating that the alloy has the considerably lower fatigue
strength. Ogarrevic and Stephens [148] have suggested that
fatigue ratio was between 0.25 and 0.5 for wrought Mg alloys
and higher ratios were corresponding to the higher strength
alloys.
To clearly understand the influence of I-phase on the
fatigue crack initiation and fatigue crack propagation (FCP)
behavior, the surfaces of the failed specimens and unfailed
specimens cycled until 107 cycles have been examined in detail
by means of SEM. For the specimen survived up to 107 cycles,
non-propagating fatigue crack can be observed on the speci-
men surface. Previous research suggested that the occurrence
of the non-propagating crack was related to the existence of
the fatigue limit [146]. Fig. 10 clearly shows two typical non-
propagating crack initiation modes [37]. Fig. 10(a) indicates
that the crack mainly initiates along the intense slip bands in
375
the grain interior. It has been suggested that this kind of crack
initiation was ascribed to the cyclic slip deformation [147]. On
the other hand, intense slip bands can be observed near the
interface of I-phase/a-Mg matrix, as shown in Fig. 10(b).
High-magnified image shows that non-propagating cracks
can also initiate along the slip bands close to the interface
of I-phase/a-Mg matrix, as shown in Fig. 10(c). It firmly indicates
that slip bands can be effectively retarded by I-phase and thus
stronger stress concentration can easily occur (especially at the slip
bands near the interfaces of I-phase/a-Mg matrix). Moreover, it
demonstrates that the quasi-periodic lattice structure of I-phase
can provide the stable I-phase/a-Mg matrix interfaces [61].
6. Thermal stability and strengthening effects of the I-phase
on Mg alloys at elevated temperature
Recently, it has been reported that Mg-rich Mg–Zn–Y alloys
containing thermally stable I-phase particles exhibit a signifi-
cantly high level of yield strength and elongation at ambient
temperature [54]. Furthermore, Singh [86,96,103], Bae [61,63],
376 D.K. Xu, E.H. Han

Crack initiation site

70
Strength amplitude (MPa)
External porosity
Internal porosity
60 Not break at 107 cyc Micro-cracks
S6
S7
S5 S3 S1
50 S4
S8 S2

40

104 105 106 107 108

5µm
Cycles to failure, Nf/cycle

Fig. 9 S–N curve of the as-cast Mg–12%Zn–1.2%Y–0.4%Zr

alloy [37].

Kim [149] and Zheng [150] have employed thermomechanical

processing (extrusion, hot-rolling and equal channel angular
extrusion-ECAE ect) to study the microstructure and properties
of Mg–Zn–Y alloys. It firmly indicated that I-phase particles
were stable against coarsening and could retain strength at
elevated temperatures. In this section, the thermal stability of
I-phase and their strengthening effects for Mg alloys at elevated
temperature will be reviewed.
5µm

6.1. Thermally stability of I-phase in Mg–Zn–Y–(Zr) alloys

In previous research, Bae et al. have deeply investigated the
flow behavior and microstructural evolution of I-phase con-
taining Mg alloys during deformation at elevated temperatures
[61]. Fig. 11 shows true stress vs. true strain curves at a strain
rate of 103 s1 in the temperature range of 220–400 1C for
Mg–2Zn–0.2Y and Mg–4.3Zn–0.7Y alloys (in at%), respec-
tively [61].
At 220 1C, the apparent flow stress of Mg–2Zn–0.2Y
reaches a peak at low strain and then decreases continuously
with increasing strain. This peak is a characteristic of strain
localization and activation of non-basal slips [151], together
with basal slip, in such a hexagonal metal, which produce a
gradual decrease in the apparent flow stress. For Mg–4.3Zn–
0.7Y, the constant level of flow stress is found to be main-
tained up to the strain of 0.4, due to an additional hardening
effect provided by the large number of particles that can
effectively block the propagation of dislocations. Near the
final fracture strain, apparent flow stress drops sharply due to
necking. The grains in both alloys are not refined but only
elongated along the applied load direction. At 300 1C, after
yielding flow stress continuously decreases with increasing
strain until the failure in both alloys due to the development of
finer grains via the concurrent DRX process, in which finer
grains can effectively contribute to the total deformation via
grain boundary sliding (GBS) [152], leading to a decrease
in the flow stress. For Mg–2Zn–0.2Y alloy, most particles,
initially distributed throughout grains, can be located at the
recrystallized grain boundaries, inferring that accumulated
dislocations at particles or at grain boundaries during defor-
mation stimulate the DRX process [153]. But for Mg–4.3Zn–
0.7Y alloy, the DRX process is not greatly activated when
compared to that of Mg–2Zn–0.2Y, which could be due to the
fact that most particles are closely spaced in some regions and
Slip bands
Micro-cracks
I-phase
2µm
Fig. 10 Non-propagating fatigue crack initiation at: (a) slip
bands in the grain interior, (b) slip bands near to the I-phase/a-Mg
matrix interfaces and (c) the magnified image of the squared area in
image (b) [37].
most grains are pinned by the particle strings. At 400 1C, flow
stress for both alloys is almost constant until a strain of
0.8 and then decreases with increasing strain due to diffuse
necking. To investigate the strain rate effect on the flow
behavior and microstructural evolution, true stress vs. true
strain curves at a test temperature of 400 1C are plotted in
Fig. 12 [61]. Typically, flow stress is relatively high at the lower
strain rates. At a strain rate of 102 s1, grains are observed to
be finer via the DRX process for both alloys, and the tendency
of DRX for Mg–2Zn–0.2Y and Mg–4.3Zn–0.7Y is almost
similar to that discussed above in the test condition of 300 1C
(103 s1). However, flow stress is almost constant in the low
strain range. In general, the stress level is lower for materials
with finer grains under the GBS controlled deformation
condition at a fixed strain rate [154]. Thus, the contribution
of GBS at 102 s1 could be small, leading to the flat shape in
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review 377

Fig. 11 True stress vs true strain curves at a strain rate of

103 s1 in the temperature range of 220–400 1C for (a) Mg–2Zn–
0.2Y and (b) Mg–4.3Zn–0.7Y [61]. Fig. 12 True stress vs. true strain curves at a test temperature of
400 1C in the strain rate range of 104 to 102 s1 for (a) Mg–2Zn–
0.2Y and (b) Mg–4.3Zn–0.7Y [61].
the stress–strain curves. At a strain rate of 104 s1, flow stress
slightly increases with increasing strain due to the grain
7. Application of I-phase for the mechanical improvement of
growth for both alloys. Overall, the flow stress of Mg–2Zn–
other Mg alloys
0.2Y is higher than that of Mg–4.3Zn–0.7Y at temperatures
higher than 220 1C, although the volume fraction of I-phase
Quasicrystals have unusual properties such as high hardness,
in Mg–4.3Zn–0.7Y is higher than that in Mg–2Zn–0.2Y. With
high Young’s modulus, high thermal-dynamical stability and
the consideration of the I-phase/a-Mg matrix eutectic tem-
low surface energy [96,156]. In the previous research, lots of
perature of the alloys 440 1C the softening of the eutectic
work has been done about the strengthening effect of I-phase
region is enhanced as the temperature increases, which may
particles on the mechanical properties of Mg–Zn–Y–(Zr)
provide the slightly lower level of flow stress in Mg–4.3Zn–
system alloys [36–38,44,45,54,61,63]. Moreover, superior wett-
0.7Y, in addition to the effect of the finer grain size on flow
ability of Mg-based quasicrystals with the Mg matrix makes it
stress.
possible be the desirable strengthening particles in Mg alloys.
Thermal stability of the I-phase particles in Mg–4.3Zn–0.7Y
Therefore, spherical morphology of reinforced particles and
was reported elsewhere [155]. In Mg–4.3Zn–0.7Y deformed to a
their dispersive distribution in the Mg matrix can effectively
strain of around 1.0 with a strain rate of 104 s1 at a test
decrease rending effects on the matrix and stress concentra-
temperature of 400 1C, the shape and size of I-phase particles
tion, resulting in stronger a-Mg/I-phase particle interfacial
were investigated using SEM and TEM, which were given in
bonding. In this section, the applications of I-phase particles
[155]. The I-phase particles initially distributed near the eutectic
to strengthen AZ91 and Mg–Li alloys will be reviewed.
pocket before testing were distributed more randomly in
the tested specimen and the particles did not coarsen, but only
the shape of the particle, initially faceted, was found to be 7.1. Effect of I-phase on the mechanical improvement
somewhat rounded. Also, some precipitated nano-scale I-phase of AZ91 alloy
particles were observed. Furthermore, any debonding or nano-
scale defects at the particle/matrix interface could not be seen in Due to better properties and good wettability of I-phase, the
the tested specimen. The existence of I-phase in this deformed addition of I-phase particles should be possible to make AZ91
specimen clearly indicates that I-phase is thermally stable in the alloy have a remarkable improvement in strength, toughness
a-Mg phase. Due to the reasonably rigid bonding with a and heat resistance. Zhang et al. firstly tried to introduce
relatively low interfacial energy at the I-phase particle/matrix I-phase (Mg45Zn47Y5Mn3) particles into AZ91 alloy with the
interface, the driving force for the coarsening of I-phase particles purpose to obtain a good method for dramatically improving
may be lower. mechanical properties of AZ91master alloy both at room and
378
Table 4 Chemical compositions of AZ91 alloys in experiments [50].
Alloys no. Chemical composition (wt%)
Mg Al Zn
1 Bal 9.000 0.800
2 Bal 8.900 1.548
3 Bal 8.900 2.296
4 Bal 8.890 3.088
5 Bal 8.870 3.792
Fig. 13 Relationship between additions of I-phase master alloy
and mechanical properties of AZ91 alloys [50].
elevated temperature [50]. The chemical compositions of
investigated alloys are listed in Table 4 [50].
Fig. 13 shows effect of the addition level of I-phase master
alloy on mechanical properties of AZ91 magnesium alloys
[50]. The macro-hardness of AZ91 magnesium alloys has the
tendency to increase when the addition level of I-phase master
alloy increases (shown in Fig. 13). This phenomenon may be
attributed to quasicrystal phase, which is distributed in the
a-Mg matrix or at grain boundaries after the introduction of
I-phase master alloy into AZ91 master alloy. Micro-hardness
value of quasicrystal phase (556.7HV), much higher than that
of b-Mg17Al12 phase (153HV), is positive to the improvement
of macro-hardness of AZ91 magnesium alloys. From Fig. 13,
it can be seen that impact toughness of AZ91 magnesium
alloys after the addition of I-phase master alloy is distinctively
higher than that of un-reinforced AZ91 master alloy [50]. With
the addition level of I-phase master alloy increases to 5.1%,
impact toughness reaches the peak, 15.3 J/cm2, which is nearly
252% as many as that of AZ91 master alloy. However, further
increasing the addition level to 6.8%, impact toughness
decreases slightly from the peak. This phenomenon is mainly
attributed to coarse-crystalline tendency resulting from impro-
perly high addition level of I-phase master alloy. The smaller
the grain size is, the more energy is consumed during crack
propagation (because the total interface area increases with
finer grains). Therefore, grain refinement can effectively
enhance toughness. Moreover, black granular compounds
distributed in the a-Mg matrix and at the grain boundaries
can restrain deformation of the matrix, impede dislocation, pin
the grain boundaries, and prevent grain-boundary slipping [50],
thereby strengthen AZ91 alloys. However, when the addition
level of the reinforced phase is too high, the precipitated phase
D.K. Xu, E.H. Han
I-phase addition
Mn Y
0.250 0.000 0
0.286 0.042 1.7
0.318 0.085 3.4
0.354 1.300 5.1
0.386 1.700 6.8
with large quantity at grain boundaries will agglomerate,
reducing bonding strength and increasing the stress concentra-
tion at grain boundaries. These grain boundaries are prone to
rupture during the plastic deformation. Therefore, excess
addition of I-phase master alloy will decrease the toughness
and increase the brittleness of AZ91 alloys. In addition, high
performance of I-phase at elevated temperature can guarantee
high strength at high temperature; and pinning effect of I-phase
on the grain boundaries can prevent grain-boundary sliding at
high temperature [38]. Consequently, the reinforced AZ91
alloys with improved heat resistance can be obtained.
To obtain Mg alloys with high strength and toughness, it is
indispensable to add the reinforced phase into Mg alloys.
Meanwhile, optimization and stability of interface structure
between the reinforced phase and the matrix is the key to
maximize properties of reinforced phase and the matrix and to
obtain optimal comprehensive performance. Due to the good
reactivity of Mg, the interface condition between the rein-
forced phase and the matrix become much more complicated.
As the reinforced phase, I-phase is distributed in the a-Mg
matrix or around b-Mg17Al12 phase, with smaller particle
size than that fabricated by low alloying or melt infiltration
[157,158]. Due to low surface energy on the I-phase particle
surface [61] as well as similarity between the I-phase and the
matrix, excellent wetting of the quasicrystal phase with the
matrix can be obtained. Therefore, the spherical I-phase has
less rending effect on the matrix than petal-like or rod-like
one; and the degree of stress concentration in the matrix is
much lower. The introduction of I-phase master alloy results
in fine grain size of a-Mg matrix and particle-like morphology
of b-Mg17Al12 phase. The grain refinement effect may be
attributed to heterogeneous nucleation of a-Mg primary phase
on spherical I-phase particles and suppressed grain growing
process.
These I-phase particles, which are still stable at 673 K, can
pin the grain boundary and effectively prevent the migration
of b-Mg17Al12 phase; I-phase particles with dispersive distri-
bution can prevent dislocation movement, exhibit their dis-
persion strengthening effects and improve heat resistance
of AZ91 alloys [96], ensuring the I-phase strengthened AZ91
alloys having excellent mechanical properties at elevated
temperatures.
7.2. Effect of I-phase on the mechanical improvement
of as-extruded Mg–Li alloys
Alloying Mg with Li of extremely low density, 0.534 g/cm3,
can further reduce the weight of Mg alloys. However, based
on the previous results listed in Table 5, it reveals that the
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review 379

Table 5 The summarization of the mechanical properties of different Mg–Li alloys (tested at room temperature) [159,160].

Condition As-extruded state Equal channel angular extrusion (ECAE)

o0.2 (MPa) UTS (MPa) Elongation (%) o0.2 (MPa) UTS (MPa) Elongation (%)

Mg–11%Li–1%Zn [159] 96 133 60 150 175 35

Mg–9%Li–1%Zn [159] 100 141 56 160 182 31
Mg–9%Li–1%Zn–0.2%Mn [159] 90 130 70 140 165 22
Mg–9%Li–1%Zn–1%Al–0.2%Mn [159] 105 150 60 145 180 24
Mg–9%Li–1%Zn–3%Al–0.2%Mn [159] 110 161 50 130 180 27
Mg–3.3%Li [160] 69 160 18 113 200 33

strength of Mg–Li alloys is very low [159,160]. Generally,

previous strengthening methods such as adding Zn or Al
alloying elements and severe plastic deformation (hot extru-
sion or ECAE) cannot make the UTS (ultimate tensile
strength) of Mg–Li alloys exceed 200 MPa.
On the other hand, it has been reported that Mg–Zn–Y
alloys containing I-phase (Mg3Zn6Y, icosahedral quasicrystal
structure) as a secondary phase exhibit good mechanical
properties both at room temperature and elevated temperature
[61]. Depending on the volume fraction of I-phase, Mg–Zn–
Y–Zr alloys can have yield strength from 150 to 450 MPa at
room temperature [54]. Previous studies [42–44,46,47] indi-
cated that the existence of I-phase in Mg–Zn–Y or Mg–Zn–Y–
Zr alloys was closely depended on Zn/Y weight ratio.
Literature [43] suggested that when Zn/Y ratio exceeded
4.38, it would make element Y almost fully existed in the
form of I-phase. When Zn/Y ratio was a certain value
(44.38), with the increase of Y content, more I-phase would
be formed in the Mg matrix.
Based on the analysis of these two alloy systems mentioned
above, two questions can be easily proposed: (1) Whether
I-phase can be introduce into Mg–Li alloys or not. (2) If
I-phase can be introduced, whether the mechanical properties
of Mg–Li alloys can be greatly improved or not. To answer
these two questions, Xu et al. has deeply investigated and
compared the microstructure and mechanical properties of
three kinds of new developed Mg–Li–Zn–Y alloys [53].
XRD analysis is shown in Fig. 14 [53]. It reveals that for
alloys I to III, the main phases are a-Mg, b-Li, LiMgZn and
I-phase. Meanwhile, with the increase of Zn and Y content,
the diffraction peak of W-phase will be gradually intensified.
In addition, it has been reported [36,44,61] that I-phase could
form interdendritic eutectic pockets with a-Mg. Therefore, it
was an easy way to determine I-phase by its morphology.
To describe and compare conveniently, the mechanical
properties of 0.2% proof yield stress (s0.2), ultimate tensile
strength (UTS), and elongation to failure for the alloys are
listed in Table 6 [53]. It shows that the quantity of I-phase can
effectively improve the yield strength and UTS of alloys.
Compared with alloys I and III, it indicates that with the
quantity of I-phase increasing approximately three times, the
yield strength and UTS increase from 148 and 222 MPa to 166
and 247 MPa, respectively. Meanwhile, the plasticity of alloy
III decreases greatly.
Previous research of Mg–Zn–Y–Zr alloys indicated [42,43,46]
that with the quantity of W-phase increasing, the strength of
alloys would be degraded. X-ray analysis indicates that for alloy
II, W-phase can hardly be detected. Therefore, it can effectively
avoid the influence of W-phase. To indicate the effect of I-phase
Fig. 14 X-ray diffraction patterns of as-extruded Mg–Li–Zn–Y
alloys. (Arrows in the figure indicate the intensifying tendency of
W-phase diffraction peak [53].)
on the mechanical properties of alloys, only the fracture of alloy
II has been chosen to be observed.
Based on Mg–Li–Zn ternary phase diagram [161], it
indicates that when Li content is between 6.0 and 9.5 wt%,
a-Mg and b-Li coexist and the Zn content in the solid
solutions cannot exceed 2 wt%. With the solidification tempera-
ture decreasing, the solid solubility of Zn decreases gradually.
Meanwhile, Mg–Li–Y ternary phase diagram reveals [162] that
Y content in the solid solutions is very tiny. In addition, due to
the interaction of elements Zn and Y, the solid solubility of Zn
and Y will be greatly decreased [163]. In this study, Li content of
alloys I to III is about 8 wt%. Therefore, as solidification
process goes on, the redundant Zn and Y (with Zn/Y ratio
higher than 4.38) will exist between a-Mg and b-Li phases and
preferentially form I-phase at the a-Mg/b-Li phase interfaces.
Certainly, I-phase can also form in the interior of a-Mg and
b-Li matrix. It has been reported that the melting temperature
of I-phase eutectic pockets is about 450 1C [45,47,61,63]. There-
fore, when temperature is lower than 450 1C, the forming of
I-phase will retard the further diffusion of Zn and Y. Especially
for alloys II and III, more I-phase can form during solidification
process, which will easily lead to the forming of areas with
higher and lower Zn/Y ratio in the liquid phase. Therefore, the
area (with lower Zn/Y ratio) cannot fully meet the requirement
of forming I-phase and W-phase will be formed, whereas the
area (with higher Zn/Y ratio) can successfully form I-phase and
the redundant Zn will form supersaturation solid solution in the
380
Table 6 Chemical composition and the mechanical properties of the as-extruded Mg–Li–Zn–Y alloys [53].
Normal alloys Chemical composition (wt%)
Mg Zn Y Li
Alloy I Bal 3.12 0.61 8.04
Alloy II Bal 6.47 1.26 7.86
Alloy III Bal 9.25 1.79 7.67
a-Mg matrix. Previous research reported [164] that after T6
temper treatment (solid solution for 2.5 h at 500 1C plus 15 h
artificial ageing at 180 1C), Mg–Zn–Y phases (I-phase and
W-phase) disappeared and rod-like MgZn0 precipitated from
the supersaturation solid solution. In addition, Mg–Li–Zn
ternary phase diagram [161] indicates that when the contents
of Mg and Zn are higher than about 40 at%, Li will preferen-
tially form LiMgZn phase with Zn and Mg. Therefore, the
formation of LiMgZn can be divided into two steps: (1) The
formation of rod-like MgZn0 causes an area of higher Zn
content. (2) The diffusion of Li and formation of LiMgZn.
Based on the discussion mentioned above, I-phase can be
successfully introduced into the matrix of Mg–Li alloys, which
clearly answers the first question. About whether I-phase can
effectively strengthen Mg–Li alloys or not, it will be discussed as
follows.
It has been reported [61] that the interface layer of a-Mg
with a 3–5 nm thickness still preserved the orientation rela-
tionship with I-phase and the coherency between I-phase
and a-Mg could be achieved by introducing steps and ledges
periodically along the interface. Therefore, the atomic bonding
between I-phase and the hexagonal structure was rigid enough
to be retained during severe plastic deformation. In the
research of as-cast Mg–Zn–Y–Zr alloys [42], it suggested that
the a-Mg/I-phase eutectic pockets could retard the basal slip
and no cracks can be observed at the a-Mg/I-phase interfaces.
Compared with the tensile properties listed in Tables 5 and 6,
it clearly suggests that by introducing I-phase into the Mg
matrix can effectively improve the strength of Mg–Li alloys.
During the tensile testing, the higher stress concentration
will occur around the big bulk I-phase, which degrades the
strength of alloys. At a certain stress level, the micro-cracks
will be formed in the interior of the big bulk I-phase to relieve
the deformation incompatibility between I-phase and a-Mg
matrix. Therefore, to fully exploit the potential strength of
Mg–Li alloys, the quantity of existing W-phase and the size of
I-phase must be strictly controlled.
8. I-phase in other Mg alloys
Besides the formation of I-phase in Mg–Zn–Y–(Zr) system alloys,
I-phase can also widely be found in the Mg–Zn–RE (RE¼Gd,
Er, Ho, Nd, Dy, La, Pr, Tb and Ce) rare-earth-containing alloys
[28–33,165–173]. Similarly, Li et al. reported that the I-phase
formation in the Mg–Zn–Er alloys is closely related to the Zn/Er
weight ratio. When Zn/Er is between 6 and 10, the mainly formed
secondary phase is I-phase [172]. Therefore, it can be predicted
that the I-phase formation in Mg–Zn–RE system alloys should
mainly depend on the Zn/RE ratio.
Moreover, some researchers reported that I-phase can also
form in Mg–Zn–Al and Mg–Cu–Al alloys [24–28]. However,
D.K. Xu, E.H. Han
Zn/Y ratio Mechanical properties
s0.2 (MPa) UTS (MPa) Elongation (%)
5.11 148 222 30.7
5.13 159 239 20.4
5.17 166 247 17.1
most icosahedral phases in these non-rare-earth-containing
Mg alloys are formed by using the rapid solidification
technique. In the research of quasicrystalline grain boundary
phase in ZA85 die-cast and re-melt Mg alloy [174], Vogel et al.
found that through controlling the cooling rate higher than
20 K/s during the die-cast process, meta-stable I-phase can be
formed in Mg–Zn–Al system alloys. According to Mg–Zn–Al
ternary diagram, the solidification of ZA85 alloy starts with
the crystallization of a-Mg with a dendritic grain structure.
Then, Mg17Al12 solidifies in a binary eutectic reaction. How-
ever, Mg17Al12 is not stable during further cooling and reacts
with the liquid in a transformation reaction to form ternary
inter-metallic quasi-crystal [174]. It is known from other
systems that quasicrystals are often formed by peritectic reactions
[39]. Additionally, Banerjee et al. reported that the I-phase formed
in a rapidly solidified Mg–Al–Zn alloy is metastable and can be
decomposed during heat treatment [175]. Similarly, Park et al.
further indicated that during solution treatment at 330 1C, the I-
phase in Mg–Zn–Mn–Al alloy can be rapidly replaced by
equilibrium f-phase (Mg21(Al,Zn)17), which gradually disappears
within a-Mg [176]. Based on the description mentioned above, it
can be seen that the conditions for the formation of I-phase in the
non-rare-earth-containing Mg alloys are quite severe. Thus, it is
very difficult to fabricate a bulk I-phase containing Mg–Al–Zn or
Mg–Cu–Al alloys for conventionally mechanical tests. Therefore,
only microstructural information about I-phase in these alloys can
be obtained from the previous literatures and basically no relevant
literatures about their effects on mechanical properties can be
referred.
9. Future research about I-phase in Mg alloys
In this review article, the characterization of I-phase, the
transformation reactions for I-phase formation during solidi-
fication, thermal stability of I-phase, grain refinement of
I-phase, effects of I-phase on mechanical properties at both
room and elevated temperatures and the applications of
I-phase in other Mg alloys have been described.
From the point of view for the mechanical improvement
of Mg alloys, it seems that the formation of I-phase is
very helpful. Depending on the volume fraction of I-phase,
Mg–Zn–Y–(Zr) alloys can have yield stress ranging from 150
to 450 MPa at room temperature. In addition, the presence of
I-phase particles in the matrix was the effective method of
improving the fracture toughness of Mg alloys. Due to high
hardness and low friction coefficient of I-phase, Mg alloys
reinforced with I-phase particles usually have superior wear
resistance. Moreover, the addition of I-phase can improve the
impact toughness of the reinforced AZ91 alloys to be 2.5 times
as high as that of un-reinforced AZ91 master alloy. Through
introducing I-phase to the Mg–Li alloys, the ultimate tensile
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review
strength of newly developed Mg–Li alloys can reach 250 MPa.
Generally, to fully use phases to improve the mechanical
properties of alloys, researchers always tried to make the
strengthening phase particles as fine as possible to avoid
strong stress concentration around them during mechanical
tests. However, so far, few research works about the refine-
ment of I-phase and their effects on the mechanical improve-
ment can be referred. Therefore, the potential strengthening
effect of I-phase has still not been fully exploited. Therefore, it
is definitely necessary to carry out further research work about
possible methods such as special casting, thermo-mechanical
deformation and heat treatment etc to ensure the dispersive
distribution of I-phase particles in the matrix of Mg alloys.
On the other hand, the applicable conditions for the I-phase
formation in the non-rare-earth-containing Mg alloys are still
not well understood. Moreover, whether the formation of
these I-phases with different chemical composition can have
the similar effect of the I-phase (Mg3Zn6Y) in Mg–Zn–Y–(Zr)
alloys for the mechanical improvement or not, so far no
relevant research work has been done. Therefore, the future
work about the research of I-phase in Mg alloys should also
focus on the practically applicable conditions for the forma-
tion of I-phase in non-rare-earth-containing Mg alloys and
their effects on the mechanical properties.
Acknowledgments
This work was supported by National Natural Science
Foundation of China projects under Grant No. 51171192
and No. 51271183, a National Basic Research Program of
China (973 Program) project under Grant No. 2012CB067425
and an Innovation Fund of Institute of Metal Research
(IMR), Chinese Academy of Sciences (CAS).
References
[1] D. Shechtman, I. Blech, D. Gratias, J.W. Cahn, Metallic phase
with long-range orientational order and no transitional sym-
metry, Physical Review Letters 53 (1984) 1951–1953.
[2] Z. Zhang, H.Q. Ye, K.H. Kuo, A new icosahedral phase with
M35 symmetry, Philosophical Magazine A 52 (1985) 49–52.
[3] K.H. Kuo, Salient features of quasi-crystals, Materials Chem-
istry and Physics 39 (1994) 1–11.
[4] D. Shi, H.Q. Ye, Q.B. Yang, A structural model of the
icosahedral Al6Li3Cu phase studied by high-resolution electron-
microscopy, Philosophical Magazine B 63 (1991) 1365–1374.
[5] K.H. Kuo, Quasi-crystals in rapidly solidified alloys of Al–Pt
group-metals. 1. An overview of quasi-crystals in aluminium-
transition metal-alloys, Journal of the Less-common Metals 163
(1990) 9–17.
[6] A.Q. He, H.Q. Ye, K.H. Kuo, Relationship between the
icosahedral quasicrystal and the related crystal phase in
a quenched Fe52Nb48 alloy, Scripta Metallurgica 23 (1989)
533–536.
[7] D.S. Zhou, D.X. Li, H.Q. Ye, Local translational order in the
icosahedral quasicrystalline phase of V41Ni36Si23, Philosophical
Magazine Letters 56 (1987) 209–215.
[8] Q.B. Yang, K.H. Kuo, A new description of pentagonal Frank–
Kasper phases and a possible structure model of the icosahedral
quasi-crystal, Acta Crystallographica Section A 43 (1987) 787–795.
[9] K.H. Kuo, The simplest of quasi-crystals, Recherche 18 (1987)
1406–1408.
381
[10] Z. Zhang, K.H. Kuo, Local translational order in the NiTi2
icosahedral quasi-crystal, Journal of Microscopy 146 (1987)
313–321.
[11] K.H. Kuo, D.S. Zhou, D.X. Li, Quasi-crystalline and Frank–
Kasper phases in a rapidly solidified V41Ni36Si23 alloy, Philo-
sophical Magazine Letters 55 (1987) 33–39.
[12] Z. Zhang, K.H. Kuo, Orientation relationship between the icosahe-
dral and crystalline phase in (Ti1-XVX)2Ni alloys, Philosophical
Magazine Part B 54 (1986) 83–87.
[13] K.H. Kuo, C. Dong, D.S. Zhou, A Friauf–Laves (Frank–Kasper)
phase related quasi-crystal in a rapidly solidified Mn3Ni2Si alloy,
Scripta Metallugica 20 (1986) 1695–1698.
[14] K.H. Kuo, From Frank–Kasper phases to the icosahedral
quasi-crystal, Journal de Physique 47 (1986) 425–436.
[15] Q.B. Yang, K.H. Kuo, A structural model of the icosahedral phase
derived from its close relationship to the pentagonal Frank–Kasper
phases, Philosophical Magazine B 53 (1986) 115–121.
[16] K.F. Kelton, Quasi-crystals—structure and stability, Interna-
tional Materials Reviews 38 (1993) 105–137.
[17] M.A. Marcus, V. Elser, T2-AlLiCu—a stable icosahedral phase,
Philosophical Magazine Part B 54 (1986) 101–104.
[18] A.P. Tsai, A. Inoue, T. Masumoto, A stable quasi-crystal in Al–
Cu–Fe system, Japanese Journal of Applied Physics 26 (1987)
1505–1507.
[19] L.X. He, Y.K. Wu, K.H. Kuo, Decagonal quasicrystals with
different periodicities along the tenfold axis in rapidly solidified
Al65Cu20Mn15, Al65Cu20Fe15, Al65Cu20Co15 or Al65Cu20Ni15,
Journal of Materials Science Letters 7 (1988) 1284–1286.
[20] S. Maamar, M. Harmelin, On the transitions of the icosahedral
and decagonal phases towards equilibrium phases in Al–Cu–Mn
alloys, Philosophical Magazine Letters 64 (1991) 343–348.
[21] C. Beeli, H.U. Nissen, J. Robadey, Stable Al–Mn–Pd quasi-
crystals, Philosophical Magazine Letters 63 (1991) 87–95.
[22] A.P. Tsai, A. Inoue, T. Masumoto, Stable decagonal quasi-
crystals with a periodicity of 1.6 nm in Al–Pd–(Fe, Ru or Os)
alloys, Philosophical Magazine Letters 64 (1991) 163–167.
[23] W. Ohashi, F. Spaepen, Stable Ga–Mg–Zn quasi-periodic
crystals with pentagonal dodecahedral solidification morphol-
ogy, Nature 330 (1987) 555–556.
[24] H. Yamane, K. Kimura, T. Shibuya, S. Takencji, Proceedings
of the International Workshop on Quasicrystals, Beijing, Ed.
K.H. Kuo (Trans. Tech., Aedermannsdorf, (1987) p. 185.
[25] P. Ramachandrarao, G.V. Sastry, A basis for the synthesis of
quasicrystals, Pramana 24 (1985) 225–230.
[26] N.K. Mukhopadhyay, G.N. Subbanna, S. Ranyanathan,
K. Chattopadhyay, An electron-microscopic study of quasi-
crystals in a quaternary alloy—Mg32(Al,Zn,Cu)49, Scripta Metal-
lurgica 20 (1986) 525–528.
[27] T. Rajasekharan, The quasi-crystalline phase in the Mg–Al–Zn
system, Nature 322 (1985) 528–530.
[28] G.V. Sastry, V.V. Rao, P. Ramachandrarao, T.R. Anantharaman,
A new quasi-crystalline phase in rapidly solidified Mg4CuAl6,
Scripta Metallurgica 20 (1986) 191–193.
[29] Z.P. Luo, S.Q. Zhang, Y.L. Tang, D.S. Zhao, Quasi-crystals in
as-cast Mg–Zn–RE alloys, Scripta Metallugica et Materialia 28
(1993) 1513–1518.
[30] Z.P. Luo, S.Q. Zhang, Comment on the so-called Z-phase in
magnesium alloys containing zinc rare-earth elements, Journal
of Materials Science Letters 12 (1993) 1490–1492.
[31] Y.L. Tang, D.S. Zhao, Z.P. Luo, N.F. Shen, S.Q. Zhang,
Morphology and the structure of quasi-crystal phase in as-cast
and melt-spun Mg–Zn–Y–Zr alloys, Scripta Metallugica et
Materialia 29 (1993) 955–958.
[32] Y.L. Tang, D.S. Zhao, Z.P. Luo, N.F. Shen, S.Q. Zhang, A new
quasi-crystalline phase in Mg–Zn–Y–Zr cast ingot, Materials
Letters 18 (1993) 148–150.
382
[33] Z.P. Luo, S.Q. Zhang, Y.L. Tang, D.S. Zhao, On the stable
quasi-crystals Zn slowly cooled Mg–Zn–Y alloys, Scripta
Metallugica et Materialia 32 (1995) 1411–1416.
[34] F.S. Pierce, S.J. Poon, Q. Guo, Electron localization in metallic
quasi-crystals, Science 261 (1993) 737–739.
[35] J.M. Dubois, P. Plaindoux, E. Berlin-Ferre, N. Tamura,
D.J. Sordelet, Proceedings of the Sixth International Conference
on Quasicrystals. Singapore: World Scientific; 1997.
[36] D.H. Bae, Y. Kim, I.J. Kim, Thermally stable quasicrystalline
phase in a superplastic Mg–Zn–Y–Zr alloy, Materials Letters 60
(2006) 2190–2193.
[37] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, The fatigue behavior of
I-phase containing as-cast Mg–Zn–Y–Zr alloy, Acta Materialia
56 (2008) 985–994.
[38] A. Singh, M. Nakamura, M. Watanabe, A. Kato, A.P. Tsai,
Quasicrystal strengthened Mg–Zn–Y alloys by extrusion, Scripta
Materialia 49 (2003) 417–422.
[39] C. Janot, Quasicrystals, Clarendon Press, Oxford, 1994.
[40] W.F. Smith (Ed.), McGraw-Hill, New York, 1993.
[41] J.Y. Lee, H.K. Lim, D.H. Kim, W.T. Kim, D.H. Kim, Effect of
volume fraction of quasicrystal on the mechanical properties of
quasicrystal-reinforced Mg–Zn–Y alloys, Materials Science and
Engineering 449-451 (2007) 987–990.
[42] D.K. Xu, W.T. Tang, L. Liu, Y.B. Xu, E.H. Han, Effect of Y
concentration on the microstructure and mechanical properties
of as-cast Mg–Zn–Y–Zr alloys, Journal of Alloys and Com-
pounds 432 (2007) 129–134.
[43] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, The influence of element Y
on the mechanical properties of the as-extruded Mg–Zn–Y–Zr
alloys, Journal of Alloys and Compounds 426 (2006) 155–161.
[44] J.Y. Lee, D.H. Kim, H.K. Lim, D.H. Kim, Effects of Zn/Y
ratio on microstructure and mechanical properties of Mg–Zn–Y
alloys, Materials Letters 59 (2005) 3801–3805.
[45] X.Q. Zeng, Y. Zhang, C. Lu, W.J. Ding, Y.X. Wang, Y.P. Zhu,
Precipitation behavior and mechanical properties of a Mg–Zn–Y–
Zr alloy processed by thermo-mechanical treatment, Journal of
Alloys and Compounds 395 (2005) 213–219.
[46] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, Effect of microstructure
and texture on the mechanical properties of the as-extruded
Mg–Zn–Y–Zr alloys, Materials Science and Engineering A 443
(2007) 248–256.
[47] Y. Zhang, X.Q. Zeng, L.F. Liu, C. Lu, H.T. Zhou, Q. Li, Y.P.
Zhu, Effects of yttrium on microstructure and mechanical
properties of the hot-extruded Mg–Zn–Y–Zr alloys, Materials
Science and Engineering A 373 (2004) 320–327.
[48] S.Q. Luo, A.T. Tang, F.S. Pan, K. Song, W.Q. Wang, Effect of
mole ratio of Y to Zn on phase constituent of Mg–Zn–Zr–Y
alloys, Transactions of the Nonferrous Metals Society of China
21 (2011) 795–800.
[49] D.K. Xu, W.T. Tang, L. Liu, Y.B. Xu, E.H. Han, Effect of W-
phase on the mechanical properties of as-cast Mg–Zn–Y–Zr
alloys, Journal of Alloys and Compounds 461 (2008)
248–252.
[50] J.S. Zhang, L.X. Pei, H.W. Du, W. Liang, C.X. Xu, B.F. Lu,
Effect of Mg-based spherical quasicrystals on microstructure
and mechanical properties of AZ91 alloys, Journal of Alloys
and Compounds 453 (2008) 309–315.
[51] E. Mora, G. Garce0 s, E. Onorbe, P. Pe0 rez, P. Adeva, High-
strength Mg–Zn–Y alloys produced by powder metallurgy,
Scripta Materialia 60 (2009) 776–779.
[52] A. Singh, Y. Osawa, H. Somekawa, T. Mukai, Ultra-fine grain
size and isotropic very high strength by direct extrusion of chill-
cast Mg–Zn–Y alloys containing quasicrystal phase, Scripta
Materialia 64 (2011) 661–664.
[53] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, The strengthening effect
of icosahedral phase on the as-extruded Mg–Li alloys, Scripta
Materialia 57 (2007) 285–288.
D.K. Xu, E.H. Han
[54] E.S. Park, S. Yi, J.B. Ok, D.H. Bae, W.T. Kim, Proceedings of
the MRS Fall Meeting, Boston, MA; 2001.
[55] S. Takeuchi. In: Dubois J.M., Thiel P.A., Tsai A.P., Urban K.,
Eds. Materials Research Society Symposium Proceedings: Qua-
sicrystals, vol. 553, 1999. p. 283.
[56] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, The fatigue crack
propagation behavior of the forged Mg–Zn–Y–Zr alloy, Journal
of Alloys and Compounds 431 (2007) 107–111.
[57] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, Super-long fatigue
behaviour of an as-extruded ZK60 Mg alloy, Acta Metallurgica
Sinica 43 (2007) 144–148.
[58] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, The micro-mechanism of
fatigue crack propagation for a forged Mg–Zn–Y–Zr alloy in
the gigacycle fatigue regime, Journal of Alloys and Compounds
454 (2008) 123–128.
[59] A. Niikura, A.P. Tsai, A. Inoue, T. Masumoto, New class of
amorphous and icosahedral phases in Zn–Mg–rare-earth metal
alloys, Japanese Journal of Applied Physics Part 1 33 (1994)
1538–1541.
[60] S.W. Xu, M.Y. Zheng, S. Kamado, K. Wu, G.J. Wang, X.Y. Lv,
Dynamic microstructural changes during hot extrusion and
mechanical properties of a Mg–5.0Zn–0.9Y–0.16Zr (wt%) alloy,
Materials Science and Engineering A 528 (2011) 4055–4067.
[61] D.H. Bae, S.H. Kim, D.H. Kim, W.T. Kim, Deformation
behavior of Mg–Zn–Y alloys reinforced by icosahedral
quasicrystalline particles, Acta Materialia 50 (2002) 2343–2356.
[62] Y.L. Tang, D.S. Zhao, Z.P. Luo, N.F. Shen, S.Q. Zhang, On
the chemical-composition analysis of the icosahedral phase in
an Mg–Zn–Y–Zr ingot, Journal of Alloys and Compounds 204
(1994) 17–19.
[63] D.H. Bae, M.H. Lee, K.T. Kim, W.T. Kim, D.H. Kim,
Application of quasicrystalline particles as a strengthening
phase in Mg–Zn–Y alloys, Journal of Alloys and Compounds
342 (2002) 445–450.
[64] J.B. Ok, I.J. Kim, S. Yi, T.W. Kim, D.H. Kim, Solidification
microstructure of as-cast Mg–Zn–Y alloys, Philosophical Maga-
zine A 83 (2003) 2359–2369.
[65] S. Sen, T.J. Mukerji, A generalized classical nucleation theory
for rough interfaces: application in the analysis of homogeneous
nucleation in silicate liquids, Journal of Non-Crystalline Solids
246 (1999) 229–239.
[66] F. Spaepen, Structural model for solid–liquid interface in
monatomic systems, Acta Metallurgica 23 (1975) 729–743.
[67] G.S. Song, M.J. Li, G.C. Yang, Y.H. Zhou, Primary solidifica-
tion of quasicrystalline phase in a bulk undercooled Al–Mn–
(Si,B) alloy, Chinese Journal of Materials Research 13 (1999)
261–266.
[68] Y.B. Zhang, S.R. Yu, X.Y. Zhu, Y.R. Luo, Study on as-cast
microstructures and solidification process of Mg–Zn–Y Alloys,
Journal of Non-Crystalline Solids 354 (2008) 1564–1568.
[69] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, Influence of higher Zn/Y
ratio on the microstructure and mechanical properties of Mg–
Zn–Y–Zr alloys, Metallurgical and Materials Transactions A:
Physical Metallurgy and Materials Science 40 (2009) 1727–1740.
[70] A. Langsdorf, F. Ritter, W. Assmus, Determination of the
primary solidification area of the icosahedral phase in the
ternary phase diagram of Zn–Mg–Y, Philosophical Magazine
Letters 75 (1997) 381–387.
[71] N. Lebrun, A. Stamou, C. Baetzner, J. Robinson, A. Pisch, in:
G. Effenberg, F. Aldinger, P. Rogl (Eds.), Ternary Alloys, vol. 18,
Materials Science International Services, Stuttgart, 2001, p. 702.
[72] S.R. Agnew, Magnesium Technology 2002, TMS, USA, 2002,
p. 169.
[73] F. Kaiser, D. Letzig, J. Bohlen, A. Styczynski, Ch. Hartig,
K.U. Kainer, Magnesium Alloys 2003, Trans. Tech. Pub.,
Switzerland, 2003, p. 315.
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review
[74] S.R. Agnew, O. Duyulu, Plastic anisotropy and the role of non-
basal slip in magnesium alloy AZ31B, International Journal of
Plastics 21 (2005) 1161–1193.
[75] A. Styczynski, Ch. Hartig, J. Bohlen, D. Letzig, Cold rolling
textures in AZ31 wrought magnesium alloy, Scripta Materialia
50 (2004) 943–947.
[76] R. Ohyama, J. Koike, M. Suzuki, K. Maruyama, Texture
dependence of elongation anisotropy in an AZ61 magnesium alloy
sheet, Journal of the Japan Institute of Metals 68 (2004) 27–33.
[77] M.R. Barnett, M.D. Nave, C.J. Bettles, Deformation micro-
structures and textures of some cold rolled Mg alloys, Materials
Science and Engineering A386 (2004) 205–211.
[78] W.J. Kim, S.I. Hong, Y.S. Kim, S.H. Min, H.T. Jeong, J.D. Lee,
Texture development and its effect on mechanical properties of an
AZ61 Mg alloy fabricated by equal channel angular pressing, Acta
Materialia 57 (2003) 3293–3307.
[79] S.H. Kim, B.S. You, C.D. Yim, Y.M. Seo, Texture and
microstructure changes in asymmetrically hot rolled AZ31
magnesium alloy sheets, Materials Letters 59 (2005) 3876–3880.
[80] T. Mukai, M. Yamanoi, H. Watanabe, K. Higashi, Ductility
enhancement in AZ31 magnesium alloy by controlling its grain
structure, Scripta Materialia 45 (2001) 89–94.
[81] S.R. Agnew, J.A. Horton, T.M. Lillo, D.W. Brown, Enhanced
ductility in strongly textured magnesium produced by equal
channel angular processing, Scripta Materialia 50 (2004) 377–381.
[82] J.Y. Lee, H.K. Lim, D.H. Kim, W.T. Kim, D.H. Kim, Effect of
icosahedral phase particles on the texture evolution in Mg–Zn–Y
alloys, Materials Science and Engineering A 491 (2008) 349–355.
[83] J.A. del Valle, M.T. Perez-Prado, O.A. Ruano, Deformation
mechanisms responsible for the high ductility in a MgAZ31 alloy
analyzed by electron backscattered diffraction, Metallurgical
and Materials Transactions 36A (2005) 1427–1438.
[84] S.R. Agnew, M.H. Yoo, C.N. Tome, Application of texture
simulation to understanding mechanical behavior of Mg and
solid solution alloys containing Li or Y, Acta Materialia 49
(2001) 4277–4289.
[85] M.J. Philippe, Texture formation in hexagonal materials,
Materials Science Forum 157-162 (1994) 1337–1350.
[86] A. Singh, M. Watanabe, A. Kato, A.P. Tsai, Twinning and the
orientation relationships of icosahedral phase with the magne-
sium matrix, Acta Materialia 53 (2005) 4733–4742.
[87] A. Singh, A.P. Tsai, A new orientation relationship OR4 of
icosahedral phase with magnesium matrix in Mg–Zn–Y alloys,
Scripta Materialia 53 (2005) 1083–1087.
[88] F.J. Humphreys, M. Hatherly, Recrystallization and Related
Annealing Phenomena, first ed. Great Britain by Galliard Ltd.,
UK, 1995 235-279.
[89] T.H. Courtney, Mechanical Behavior of Materials, second ed.
McGraw-Hill, New York, 2000, p. 135.
[90] W. Steurer, in: R.W. Cahn, P.H. Haasen (Eds.), The Structure
of Qusicystals, vol. 1, Physical Metallurgy, North-Holland,
Amsterdam, 1996.
[91] U. Koster, J. Meinhardt, S. Roos, H. Liebertz, Formation of
quasicrystals in bulk glass forming Zr–Cu–Ni–Al alloys,
Applied Physics Letters 69 (1996) 179–181.
[92] T. Mukai, H. Watababe, K. Higashi, Application of super-
plasticity in commercial magnesium alloy for fabrication of
structural components, Materials Science and Technology 16
(2000) 1314–1319.
[93] H. Somekawa, H. Watababe, T. Mukai, K. Higashi, Low
temperature diffusion bonding in a superplastic AZ31 magne-
sium alloy, Scripta Materialia 48 (2003) 1249–1254.
[94] H. Watanabe, T. Mukai, K. Ishikawa, K. Higashi, High-strain-
rate superplasticity in an AZ91 magnesium alloy processed by
ingot metallurgy route, Metallurgical and Materials 43 (2002)
78–80.
383
[95] A. Singh, M. Watanabe, A. Kato, A.P. Tsai, Strengthening
effects of icosahedral phase in magnesium alloys, Philosophical
Magazine 86 (2006) 951–956.
[96] A. Singh, M. Watanabe, A. Kato, A.P. Tsai, Microstructure
and strength of quasicrystal containing extruded Mg–Zn–Y
alloys for elevated temperature application, Materials Science
and Engineering A 385 (2004) 382–396.
[97] A. Singh, A.P. Tsai, On the cubic W phase and its relationship
to the icosahedral phase in Mg–Zn–Y alloys, Scripta Materialia
49 (2003) 143–148.
[98] A. Niikura, A.P. Tsai, A. Inoue, T. Masumoto, Stable Zn–Mg–
rare-earth face-centered icosahedral alloys with pentagonal
dodecathedral solidification morphology, Philosophical Maga-
zine Letters 69 (1994) 351–355.
[99] F.J. Humphreys, Local lattice rotations at 2nd phase particles in
deformed metals, Acta Metallurgica 27 (1979) 1801–1814.
[100] C.J. Ma, M.P. Liu, G.H. Wu, W.J. Ding, Y.P. Zhu, Tensile
properties of extruded ZK60-RE alloys, Materials Science and
Engineering 349 (2003) 207–212.
[101] G.L. Song, D. StJohn, The effect of zirconium grain refinement
on the corrosion behavior of magnesium–rare earth alloy MEZ,
Journal of Light Metals 2 (2002) 1–16.
[102] M.Y. Zheng, S.W. Xu, X.G. Qiao, K. Wu, S. Kamado,
Y. Kojima, Compressive deformation of Mg–Zn–Y–Zr alloy
processed by equal channel angular pressing, Journal of Materials
Science and Engineering A 483-484 (2008) 564–567.
[103] M.M. Avedesian, H. Baker, Magnesium and magnesium alloys,
ASM Specialty Handbook, ASM International, Materials Park,
OH, 1999.
[104] T. Sasaki, H. Somekawa, A. Takara, Y. Nishikawa, K. Higashi,
Plane-strain fracture toughness on thin AZ31 wrought
magnesium alloy sheets, Materials Transactions 44 (2003)
986–990.
[105] H. Somekawa, Y. Osawa, T. Mukai, Effect of solid-solution
strengthening on fracture toughness in extruded Mg–Zn alloys,
Scripta Materialia 55 (2006) 593–596.
[106] H. Somekawa, T. Mukai, Effect of grain refinement on fracture
toughness in extruded pure magnesium, Scripta Materialia 53
(2005) 1059–1064.
[107] H. Somekawa, T. Mukai, Fracture toughness in Mg–Al–Zn
alloy processed by equal-channel-angular extrusion, Scripta
Materialia 54 (2006) 633–638.
[108] H. Somekawa, T. Mukai, Fracture toughness in ultra fine-
grained magnesium alloy, Materials Science Forum 503-504
(2006) 155–160.
[109] J. Koike, T. Kobayashi, T. Mukai, H. Watanabe, M. Suzuki,
K. Maruyama, K. Higashi, The activity of non-basal slip systems
and dynamic recovery at room temperature in fine-grained AZ31B
magnesium alloys, Acta Materialia 51 (2003) 2055–2065.
[110] H. Somekawa, A. Singh, T. Mukai, Deformation structure after
fracture-toughness test of Mg–Al–Zn alloys processed by equal-
channel-angular extrusion, Philosophical Magazine Letters 86
(2006) 195–204.
[111] H. Somekawa, A. Singh, T. Mukai, Effect of precipitate shapes
on fracture toughness in extruded Mg–Zn–Zr magnesium alloys,
Journal of Materials Research 22 (2007) 965–973.
[112] A.P. Tsai, Y. Murakami, A. Niikura, The Zn–Mg–Y phase
diagram involving quasicrystals, Philosophical Magazine 80
(2000) 1043–1054.
[113] A. Singh, A.P. Tsai, M. Nakamura, M. Watanabe, A. Kato,
Nanoprecipitates of icosahedral phase in quasicrystal-
strengthened Mg–Zn–Y alloys, Philosophical Magazine Letters
83 (2003) 543–551.
[114] K.H. Kuo, Salient features of quasi-crystals, Materials Chem-
istry and Physics 39 (1994) 1–11.
[115] H. Somekawa, A. Singh, T. Mukai, High fracture toughness
of extruded Mg–Zn–Y alloy by the synergistic effect of grain
384
refinement and dispersion of quasicrystalline phase, Scripta
Materialia 56 (2007) 1091–1094.
[116] Y. Brechet, J.D. Embury, S. Tao, L. Luo, Damage initiation in
metal matrix composites, Acta Metallurgica et Materialia 39
(1991) 1781–1786.
[117] D.J. Lloyd, Aspects of fracture in particular reinforced metal matrix
composites, Acta Metallurgica et Materialia 39 (1991) 59–71.
[118] H. Somekawa, T. Mukai, Effect of grain refinement on fracture
toughness in extruded pure magnesium, Scripta Materialia 53
(2005) 541–545.
[119] J.R. Davis, Aluminum and Aluminum Alloys, ASM Specialty
Handbook, ASM International, Materials Park, OH, 1993.
[120] K. Purazrang, K.U. Kainer, B.L. Mordike, Fracture-toughness
behaviour of a magnesium alloy metal–matrix composite produced
by the infiltration technique, Composites 22 (1991) 456–462.
[121] K. Purazrang, P. Abachi, K.U. Kainer, Mechanical-behaviour
of magnesium alloy MMCs produced by squeeze casting and
powder metallurgical techniques, Composites Engineering 3
(1993) 489–505.
[122] S. Barbagallo, E. Cerri, Evaluation of the K-IC and J(IC)
fracture parameters in a sand cast AZ91 magnesium alloy,
Engineering Failure Analysis 11 (2004) 127–140.
[123] P. Abachi, K. Purazrang, Comparison of experimental and
analytical fracture toughness values of SiCP/QE22 Mg-alloy
composites, Materials & Design 27 (2006) 520–525.
[124] H. Somekawa, T. Mukai, Fracture toughness in an extruded ZK60
magnesium alloy, Materials Transactions 47 (2006) 995–998.
[125] L. Bresson, D. Gratias, Plastic-deformation in AlCuFe icosahedral
phase, Journal of Non-Crystalline Solids 153-154 (1993) 468–472.
[126] R. Wittmann, K. Urban, M. Schandl, E. Hornbogen, Mechanical-
properties of single-quasi-crystalline AlCuCoSi, Journal of Materi-
als Research 6 (1991) 1165–1168.
[127] Y.B. Zhang, S.R. Yu, Y.R. Luo, H.X. Hu, Friction and wear
behavior of as-cast Mg–Zn–Y quasicrystal materials, Materials
Science and Engineering A 472 (2008) 59–65.
[128] J.M. Dubois, New prospects from potential applications of
quasicrystalline materials, Materials Science and Engineering A
294-296 (2000) 4–9.
[129] P.P. Bandyopadhyay, S. Siegmann, Friction and wear behavior
of vacuum plasma-sprayed Ti–Zr–Ni quasicrystal coatings,
Surface and Coatings Technology 197 (2005) 1–9.
[130] E. Fleury, Y.C. Kim, J.S. Kim, D.H. Kim, W.T. Kim, H.S. Ahn,
S.M. Lee, Comparative study of the tribological behavior of
thermal sprayed quasicrystalline coating layers, Journal of
Alloys and Compounds 342 (2002) 321–325.
[131] J.M. Dubois, S.S. Kang, S. Von Stebut, Quasi-crystalline low-
friction coatings, Journal of Materials Science Letters 10 (9)
(1991) 537–541.
[132] Z.B. Sajuri, Y.M. Miyashita, Y. Hosokai, Y. Mutoh, Effects of
Mn content and texture on fatigue properties of as-cast and
extruded AZ61 magnesium alloys, International Journal of
Mechanical Sciences 48 (2006) 198–209.
[133] G. Eisenmeier, B. Holzwarth, H.W. Höppel, H. Mughrabi,
Cyclic deformation and fatigue behaviour of the magnesium
alloy AZ91, Materials Science and Engineering A 319-321 (2001)
578–582.
[134] H. Mayer, M. Papakyriacou, B. Zettl, S.E. Stanzl-Tschegg,
Influence of porosity on the fatigue limit of die cast magnesium
and aluminium alloys, International Journal of Fatigue 25
(2003) 245–256.
[135] M.F. Horstemeyer, N. Yang, D.L. Ken Gall, J. McDowell, P.M.
Fan, Gullett, High cycle fatigue of a die cast AZ91E-T4
magnesium alloy, Acta Materialia 52 (2004) 1327–1336.
[136] T.S. Shih, W.S. Liu, Y.J. Chen, Fatigue of as-extruded AZ61A
magnesium alloy, Materials Science and Engineering A 325
(2002) 152–162.
[137] M. Hilpert, L. Wagner, Magnesium Technology 2000, Warren-
dale, PA: TMS, 2000 375.
D.K. Xu, E.H. Han
[138] Z.B. Sajuri, Y. Miyashita, Y. Mutoh, Fatigue characteristics of
an extruded AZ61 magnesium alloy, Journal of Japan Institute
of Light Metals 52 (2002) 161–166.
[139] K. Tokaji, M. Kamakura, N. Hasegawa, T. Tsuboi, Fatigue
crack propagation in magnesium alloy AZ31 rolled plate,
Journal of the Society of Materials Science (Japan) 52 (2003)
821–826.
[140] K. Gall, G. Biallas, Hans J. Maier, Mark F. Horstemeyer,
David L. NcDowell, Environmentally influenced microstructu-
rally small fatigue crack growth in cast magnesium, Materials
Science and Engineering A 396 (2005) 143–154.
[141] E. Gariboldi, A. Lo Conte, Evaluation of toughness character-
istics of a die-cast Mg–Al–Mn alloy Am60B under different
microstructural conditions, Materials Science and Engineering A
378-379 (2004) 34–40.
[142] X.S. Wang, J.H. Fan, An evaluation on the growth rate of small
fatigue cracks in cast AM50 magnesium alloy at different
temperatures in vacuum conditions, International Journal of
Fatigue 28 (2006) 79–86.
[143] Z.P. Luo, D.Y. Song, S.Q. Zhang, Strengthening effects of rare-
earths on wrought Mg–Zn–Zr–RE alloys, Journal of Alloys and
Compounds 230 (1995) 109–114.
[144] Z.P. Luo, D.Y. Song, S.Q. Zhang, Microstructures of Mg–Zn–
Zr–RE alloys with high RE and low Zn contents, Journal of
Alloys and Compounds 209 (1994) 275–278.
[145] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, The crack initiation
mechanism of the forged Mg–Zn–Y–Zr alloy in the super-long
fatigue life regime, Scripta Materialia 56 (2007) 1–4.
[146] K. Tokaji, M. Kamakura, Y. Ishiizumi, N. Hasegawa, Fatigue
behaviour and fracture mechanism of a rolled AZ31 magnesium
alloy, International Journal of Fatigue 26 (2004) 1217–1224.
[147] H. Mayer, H. Lipowsky, M. Papakyriacou, R. Rösch, A. Stich,
S. Stanzl-Tschegg, Application of ultrasound for fatigue testing
of lightweight alloys, Fatigue and Fracture of Engineering
Materials and Structures 22 (1999) 591–599.
[148] V.V. Ogarevic, R.I. Stephens, Fatigue of magnesium alloys,
Annual Review of Materials Science 20 (1990) 141–177.
[149] I.J. Kim, K.H. Bae, D.H. Kim, Precipitates in a Mg–Zn–Y alloy
reinforced by an icosahedral quasicrystalline phase, Materials
Science and Engineering A 359 (2003) 313–318.
[150] M.Y. Zheng, X.G. Qiao, S.W. Xu, K. Wu, S. Kamado, Y. Kojima,
In-situ quasicrystal-reinforced magnesium matrix composite pro-
cessed by equal channel angular extrusion (ECAE), Journal of
Materials Science 40 (2005) 2587–2590.
[151] M. Regev, E. Aghion, S. Berger, M. Bamberger, A. Rosen,
Dislocation analysis of crept AZ91D ingot castings, Materials
Science and Engineering A 257 (1998) 349–352.
[152] T. Mohri, M. Mabuchi, M. Nakamura, T. Asahina, H. Iwasaki,
T. Aizawa, K. Higashi, Microstructural evolution and super-
plasticity of rolled Mg–9Al–1Zn, Materials Science and Engi-
neering A 290 (2000) 139–144.
[153] F.J. Humphreys, M. Hatherly, Recrystallization and Related
Annealing Phenomena, Pergamon, Oxford, 1995.
[154] D.H. Bae, A.K. Ghosh, Grain size and temperature dependence
of superplastic deformation in an Al–Mg alloy under isostruc-
tural condition, Acta Materialia 48 (2000) 1207–1224.
[155] D.H. Bae, S.H. Kim, W.T. Kim, D.H. Kim, Application of
quasicrystalline particles as a strengthening phase in Mg–Zn–Y
alloys, Materials Transactions JIM 42 (2001) 2144–2147.
[156] P.D. Bloom, K.G. Baikerikar, J.U. Otaigbe, Development of
novel polymer/quasicrystal composite materials, Materials Science
and Engineering A 294-296 (2000) 156–159.
[157] J.S. Zhang, L.X. Pei, D.H. Wei, Effect of Mg–Zn–Y quasicrys-
tals on microstructure and mechanical properties of AZ91 alloy,
Chinese Journal of Nonferrous Metals 16 (2006) 1884–1887.
[158] B.S. Murty, S.K. Thakur, B.K. Dhindaw, On the infiltration
behavior of Al, Al–Li, and Mg melts through SiCp bed,
Effects of icosahedral phase formation on the microstructure and mechanical improvement of Mg alloys: A review
Metallurgical and Materials Transactions A: Physical Metal-
lurgy and Materials Science 31 (2000) 319–325.
[159] T.C. Chang, J.Y. Wang, C.L. Chu, S.Y. Lee, Mechanical
properties and microstructures of various Mg–Li alloys, Materi-
als Letters 60 (2006) 3272–3276.
[160] T. Liu, Y.D. Wang, S.D. Wu, R. Lin Peng, C.X. Huang, C.B.
Jiang, S.X. Li, Textures and mechanical behavior of Mg–3.3%Li
alloy after ECAP, Scripta Materialia 51 (2004) 1057–1061.
[161] P.I. Kripyakevich, E.V. Mel’nik, in: P. Villars, A. Prince, H.
Okamoto (Eds.), Ternary Alloy Phase Diagrams, ASM Inter-
national, 1997, p. 12227.
[162] M.E. Drits, L.S. Guzei, M.L. Kharakterova, A.A. Burgyin,
in: P. Villars, A. Prince, H. Okamoto (Eds.), Ternary Alloy
Phase Diagrams, ASM International, 1997, p. 12224.
[163] E.M. Padezhnova, E.V. Mel’nik, R.A. Miliyevskiy, T.V. Dobatkina,
V.V. Kinzhibalo, in: P. Villars, A. Prince, H. Okamoto (Eds.),
Ternary Alloy Phase Diagrams, ASM International, 1997, p. 12369.
[164] D.K. Xu, L. Liu, Y.B. Xu, E.H. Han, The effect of precipitates
on the mechanical properties of ZK60-Y alloy, Materials
Science and Engineering A 420 (2006) 322–332.
[165] J.S. Zhang, J. Yan, F.Y. Han, W. Liang, C.X. Xu, C.L. Zhou,
Investigation on the icosahedral quasicrystal phase in Mg70.8
Zn28Nd1.2 alloy, Journal of Rare Earths 27 (2009) 264–269.
[166] G.Y. Yuan, Y. Liu, C. Lu, W.J. Ding, Effect of quasicrystal and
Laves phases on strength and ductility of as-extruded and heat
treated Mg–Zn–Gd-based alloys, Materials Science and Engi-
neering A 472 (2008) 75–82.
[167] G.Y. Yuan, Y. Liu, W.J. Ding, C. Lu, Effects of extrusion on
the microstructure and mechanical properties of Mg–Zn–Gd
alloy reinforced with quasicrystalline particles, Materials Science
and Engineering A 474 (2008) 348–354.
[168] J.H. Li, W.B. Du, S.B. Li, Z.H. Wang, Tensile and creep
behaviors of Mg–5Zn–2.5Er alloy improved by icosahedral
quasicrystal, Materials Science and Engineering A 527 (2010)
1255–1259.
[169] Y. Liu, G.Y. Yuan, C. Lu, W.J. Ding, Stable icosahedral phase
in Mg–Zn–Gd alloy, Scripta Materialia 55 (2006) 919–922.
[170] J.H. Li, W.B. Du, S.B. Li, Z.H. Wang, Effect of aging on
microstructure of Mg–Zn–Er alloys, Journal of Rare Earths 27
(2009) 1042–1045.
385
[171] M.R. Li, D.W. Deng, K.H. Kuo, Crystal structure of the
hexagonal (Zn,Mg)4Ho and (Zn,Mg)4Er, Journal of Alloys and
Compounds 414 (2006) 66–72.
[172] H. Li, W.B. Du, S.B. Li, Z.H. Wang, Effect of Zn/Er weight
ratio on phase formation and mechanical properties of as-cast
Mg–Zn–Er alloys, Materials and Design 35 (2012) 259–265.
[173] Q.F. Wang, H. Li, S.B. Li, Z.H. Wang, W.B. Du, Microstruc-
ture evolution and mechanical properties of extruded Mg–12Zn–
1.5Er alloy, Transactions of Nonferrous Metals Society of China
21 (2011) 874–879.
[174] M. Vogel, O. Krsft, G. Delm, E. Arzt, Quasi-crystalline grain-
boundary phase in the magnesium die-cast alloy ZA85, Scripta
Materialia 45 (2001) 517–524.
[175] R. Banerjee, R.T. Savalia, D. Prakash, U.D. Kulkami, G.K.
Dey, Decomposition of the quasicrystalline phase in rapidly
solidified Mg32Al17Zn32 alloy, Scripta Metallurgica et Materialia
32 (1995) 1607–1610.
[176] S.S. Park, Y.S. Oh, D.H. Kang, N.J. Kim, Microstructural
evolution in twin-roll strip cast Mg–Zn–Mn–Al alloy, Materials
Science and Engineering A 449-451 (2007) 352–355.
En-Hou Han is the director of Corrosion
Center in the Institute of Metal Research,
Chinese Academy of Sciences. He received
Fellow of NACE International in 2008. He
has published more than 200 peer reviewed
scientific papers, 38 plenary and invited lec-
ture, and hold 42 patents. His papers was
cited more than 2000 times. His research
interests include (1) Development of
advanced magnesium alloy and its corrosion
prevention techniques. (2) Interaction of mechanics and chemistry,
including stress corrosion cracking, corrosion fatigue and hydrogen
embrittlement for various materials. (3) R & D of various coatings for
corrosion protection. (4) Corrosion in severe corrosive industrial
environments, e.g., high temperature and high pressure, H2S, acidic
or alkaline environments. (5) Service lifetime prediction and failure
analysis for engineering structure and components.