Effect of high-intensity ultrasonic treatment on modification of primary Mg2Si in the hypereutectic Mg-Si alloys

Article abstract of DOI:10.1016/j.jallcom.2013.06.189

The effect of high intensity ultrasonic treatment (HIUST) on modification of primary Mg₂Si in the hypereutectic Mg-5 wt.%Si alloy have been studied. Various resulted microstructures were produced in this alloy by employing ultrasonic vibrations during solidification process at different pouring temperatures and for different application times. The results showed clearly that in the absence of HIUST, the dendrites of primary Mg₂Si were coarser and non-uniform in size. Upon HIUST of the alloy during solidification process, nearly uniform and polyhedral shape of primary Mg ₂Si with a network of Mg phase segregated along the grain boundaries were obtained. Interestingly, the average size of primary Mg₂Si decreased significantly with increasing the pouring temperature and the vibration time of HIUST reached a minimum at 800 °C and 90 s. After that the average size of primary Mg₂Si increased slightly with further increasing the pouring temperature and the ultrasonic vibration time. Modification mechanism resulting in the development of microstructure is also investigated.

Full text of DOI:10.1016/j.jallcom.2013.06.189

Journal of Alloys and Compounds 577 (2013) 693–700
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
Effect of high-intensity ultrasonic treatment on modification of primary
Mg2Si in the hypereutectic Mg–Si alloys
a
b
M.E. Moussa a, , M.A. Waly , A.M. El-Sheikh
⇑
a Department of Manufacturing Technology, Laboratory of Foundry, Central Metallurgical Research and Development Institute (CMRDI), P.O. 87, Halwan, Egypt
b Department of Mining, Petroleum and Metallurgical Engineering, Faculty of Engineering, Cairo University, P.O. 12613, Giza, Egypt
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 May 2013
Received in revised form 17 June 2013
Accepted 29 June 2013
Available online 8 July 2013
The effect of high intensity ultrasonic treatment (HIUST) on modification of primary Mg2Si in the hyper-
eutectic Mg–5 wt.%Si alloy have been studied. Various resulted microstructures were produced in this
alloy by employing ultrasonic vibrations during solidification process at different pouring temperatures
and for different application times. The results showed clearly that in the absence of HIUST, the dendrites
of primary Mg2Si were coarser and non-uniform in size. Upon HIUST of the alloy during solidification pro-
cess, nearly uniform and polyhedral shape of primary Mg2Si with a network of Mg phase segregated along
the grain boundaries were obtained. Interestingly, the average size of primary Mg2Si decreased signifi-
cantly with increasing the pouring temperature and the vibration time of HIUST reached a minimum
at 800 °C and 90 s. After that the average size of primary Mg2Si increased slightly with further increasing
the pouring temperature and the ultrasonic vibration time. Modification mechanism resulting in the
development of microstructure is also investigated.
Keywords:
Metals and alloys
Microstructure
Thermal analysis
Metallography
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
alloying [16]. However, compared with the above-mentioned tech-
niques, modification treatment is a more cost-effective processing
Magnesium alloys are widely applied in the fields which are
strongly driven towards weight-reduction, such as automotive,
aeronautic and astronautic industries [1–4]. Meanwhile, improving
the elevated temperature properties of magnesium alloys has be-
come a critical issue for the further application [5–7]. In recent
years, the fascinating properties and promising application of
hypereutectic Mg–Si alloys have attracted particular interest due
to the formation of thermally stable Mg2Si phase [7–9]. It is known
that the intermetallic compound of Mg2Si exhibits an excellent
combination of superior properties, such as high melting tempera-
ture (1085 °C), low density (1.99 Â 103 kg mÀ3), high hardness
technique and available for generally commercial application.
Much work has been focused on the modification effects of Ca
[4], P [6], Sr–Sb [7,8], Y [12], Sr [17], Sb [18,19], Ba [20], KBF4
[21,22], Bi [23], and Y2O3 [24] on the primary and eutectic Mg2Si
in magnesium alloys. However, certain internal defects could be
existed. For example, although KBF4 has a good effect on the mod-
ification of Mg2Si phase in Mg–5Si alloys, a large quantity of smoke
and sputter is produced, which is harmful to the surrounding envi-
ronment [21,22]. Y [12], Ba [20] and Bi [23] can modify and refine
the primary Mg2Si, but the Mg2Si phase becomes coarse again
when the adding amount exceeds a certain limit. Therefore, an
alternative method for the development of hypereutectic Mg–Si al-
loys with modification of Mg2Si phase and high mechanical
properties is desired.
HIUST is a simple and effective physical method for solidifica-
tion control. Investigations carried out between 1960 and 1990
[25,26], mainly in the former Soviet Union countries, clearly dem-
onstrated its grain-refinement effects on magnesium alloys and
significantly improved mechanical properties. The renewed inter-
est in magnesium materials in recent years has resulted in in-
creased attention in the grain-refinement potential of HIUST for
magnesium alloys [27–37], particularly for Mg–Al based alloys.
However, seldom investigation has been carried out with regard
to the modification effect of HIUST on morphology of primary Mg2-
Si in hypereutectic Mg–Si alloys.
(4.5 Â 109 N mÀ2),
low
thermal
expansion
coefficient
(7.5 Â 10À6 KÀ1) and reasonably high elastic modulus (120 GPa).
Furthermore, the Mg2Si phase is exceptionally stable and therefore
could effectively impede grain boundary sliding at elevated tem-
peratures [8–13]. However, the hypereutectic Mg–Si alloys pre-
pared by ordinary ingot metallurgy process showed very low
ductility and strength due to the large primary Mg2Si particle size
and the brittle eutectic phase [12,13].
Various researches are aimed at improving the mechanical
properties of Mg2Si reinforced Mg-alloys through processes, such
as hot extrusion [14], rapid solidification [15] and mechanical
⇑
Corresponding author. Tel./fax: +20 2 25010095.
E-mail address: eissa83@yahoo.com (M.E. Moussa).
0925-8388/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jallcom.2013.06.189

694
M.E. Moussa et al. / Journal of Alloys and Compounds 577 (2013) 693–700
Table 1
The chemical composition of prepared Mg–5 wt.%Si alloy (wt.%).
Si
5
Fe
Cu
Ni
Mg
0.022
0.015
0.0034
Bal.
2. Experimental procedures
2.1. Materials and processing
Commercial pure Mg (99.8 wt.% purity) and Si (99.9 wt.% purity) were used as
starting materials. Charges of about 15 kg with the nominal composition of hyper-
eutectic Mg–5 wt.%Si alloy were prepared as the base material in the present study.
The charge was melted in a graphite crucible by a 200 kW medium frequency
induction furnace under the mixed gas protection consists of tetrafluoroethane
(CF3CH2F, HFC-134a, 1 vol.%) and carbon dioxide (CO2, Bal.). Firstly commercial
Mg ingots were melted to above 650 °C, and then silicon was added into Mg melt.
After that the melt were heated to above 800 °C and kept about 30 min to ensure
Mg and Si fully reacted and formed Mg2Si. Finally the melt was poured into a cast
iron mold.
The solidification characteristic of the prepared alloy was confirmed using ther-
mal analysis in order to designate experimental conditions. The thermal analysis
test sample was obtained by pouring the quantity of the melt at 800 °C into stan-
dard Qiuk-Cup resin-bonded sand cup with dimensions described in Fig. 1. A high
sensitivity type K thermocouple (chromel–alumel) located horizontally at the cen-
ter of the cup, facilitated the capturing of the temperature during solidification. The
thermocouple within the mold was protected using silica glass tube. The data for
thermal analysis were collected using a data logger and transferred to a personal
computer for analysis. Thermal analysis trial was repeated three times to ensure
reproducibility of the results. From the thermal analysis data, the cooling curve
with its first derivative curve was plotted as shown in Fig. 2. The primary Mg2Si be-
gan to precipitate from the prepared alloy melt at 767 °C and the eutectic reaction
occurred at about 637.1 °C. The chemical compositions of the prepared alloy were
measured with inductive coupled plasma optical emission spectroscopy (ICP-OES)
(model Optima 2000, Germany), as shown in Table 1.
Fig. 1. Schematic of (a) experimental setup for thermal analysis and (b) resin-
bonded sand cup with type K thermocouple.
The ultrasonic treated samples were prepared as following. The hypereutectic
Mg–5 wt.%Si alloy of about 1 kg was melted at 800 °C in a mild steel crucible using
an electric resistance furnace under the mixed gas protection consists of tetrafluo-
roethane (CF3CH2F, HFC-134a, 1 vol.%) and carbon dioxide (CO2, Bal.). The melt was
manually stirred for 2 min using a stainless steel rod, and then was held for addi-
tional 10 min in order to get full homogenization. After that the slag was removed,
and then the melt was poured into a cylindrical resin-bonded sand mold with
dimensions of outer diameter (£ 100 mm), inner diameter (£ 42 mm) and length
(250 mm) which mounted on the ultrasonic sonotrode of diameter (£ 40 mm) as
shown in Fig. 3. The reason for using a cylindrical resin-bonded sand mold is to
reduce the cooling rate effect on the resulted microstructures of the prepared
The main purpose of the present work is to study the modifica-
tion effect of HIUST on morphology of primary Mg2Si in the hyper-
eutectic Mg–5 wt.%Si alloy. The experiments that involved
vibrating the molten alloy during solidification process are per-
formed under the application of HIUST at different pouring
temperatures and vibration times.
Fig. 2. The cooling curve with its first derivative curve of the investigated hypereutectic Mg–5 wt.%Si alloy.

M.E. Moussa et al. / Journal of Alloys and Compounds 577 (2013) 693–700
695
water, and 120 ml ethanol for 2–3 min. The microstructures of the specimens were
analyzed by optical microscope (OM) (model OPTIKA M-790, Italy). In the present
study, the average length of primary Mg2Si was measured as the size of Mg2Si.
Six OM micrographs were taken for each sample from the observed area at a low
magnification of 100Â. The average size of the primary Mg2Si was measured by Im-
ageJ1.44 software. All Mg2Si existed in one picture taken from the observed area
were measured.
Energy dispersion spectrum (EDS) (model INCA PENTAFET X3, England) affili-
ated to the scanning electron microscope (SEM) (model JSM-5410, Japan) was per-
formed to reveal the concentration of alloying elements in selected areas of the
microstructure. Phase constituents of samples were analyzed by X-ray diffraction
(XRD) (model X’PERT PRO, The Netherlands) using Cu Ka radiation in step scan of
2h from 20° to 80° with an increment of 0.02° and a scanning speed of 4°/min.
3. Results
3.1. The first group
According to the Mg–Si binary phase diagram [39], Mg–5 wt.%Si
alloy is a typical hypereutectic alloy. As-cast microstructures of
Mg–5 wt.%Si alloy processed without and with HIUST at different
pouring temperatures for 90 s are shown in Fig. 4a–f. The micro-
structures of hypereutectic base alloy reveal the presence of pri-
mary Mg2Si, Mg halos and eutectic Mg–Mg2Si. Moreover, all the
primary Mg2Si are surrounded by Mg halos, and then by eutectic
structure [8]. Without HIUST, the primary Mg2Si shows a coarse
dendritic morphology with the average size of about 200
shown in Fig. 4a.
lm, as
With the application of HIUST for 90 s at pouring temperature of
770 °C, most of primary Mg2Si phases are still dendrite, but their
average size are reduced significantly to about 80 lm, as shown in
Fig. 4b. As the pouring temperature increases from 780 to 800 °C,
most of primary Mg2Si phases become polyhedral shape and their
average size are reduced from about 43 to 33 lm respectively, as
shown in Fig. 4c–e. When the pouring temperature is further in-
creased to 810 °C, the average size of primary Mg2Si increases
slightly again to about 43 lm, as shown in Fig. 4f. Therefore, under
the application of HIUST at different pouring temperatures for
90 s, the average size of primary Mg2Si decreases significantly with
increasing pouring temperature, reaches a minimum at 800 °C, and
then increases slightly with further increasing pouring temperature
as compared without the application of HIUST, as shown in Fig. 5.
To ascertain phase constituents, composition and structure of
the based hypereutectic Mg–5 wt.%Si alloys, XRD and EDS were
conducted for both conditions without and with HIUST at pouring
temperature of 800 °C for 90 s respectively. XRD results reveal that
the constituents of the obtained microstructures for both condi-
tions without and with HIUST are only Mg2Si and Mg phases, as
shown in Fig. 6. Therefore, no change of the phase constituents ob-
tained due to HIUST.
The SEM image, EDS line scan and EDS elemental mapping of Si
and Mg for the based hypereutectic Mg–5 wt.%Si alloy without and
with HIUST at pouring temperature of 800 °C for 90 s are shown in
Figs. 7 and 8, respectively. It can be seen that the coarse primary
Mg2Si is formed by the preferred growth that occurs at the tips
of branches in the alloy without HIUST (Fig. 7a), resulting in com-
plex regular, sharp-angled and dendritic morphologies with a non-
uniform distribution of Mg in the interdendritic regions (Fig. 7b–d).
With the HIUST, the morphology of primary Mg2Si changes from
coarse dendritic structure to fine polyhedral shape (Fig. 8a) with
a network of Mg segregated along the grain boundaries (Fig. 8b–d).
Fig. 3. Schematic of the experimental setup used in this study.
samples. Therefore, the difference in the morphology and size of primary Mg2Si in
the microstructures of investigated samples were obtained mainly as a result of the
difference in the application of HIUST conditions.
Two types of experiments were carried out, namely the first group and the sec-
ond group. In the first group, HIUST which was generated by using ultrasonic gen-
erator (model TS6MD1, Russia) and magnetostrict transducer (model PMS-15-22,
Russia) with the maximum output power of 5 KW and the fixed frequency of
21.4 kHz was applied right before the melt was poured into cylindrical resin-
bonded sand mold. The HIUST effect was studied in this group at different desig-
nated pouring temperatures (770, 780, 790, 800, 810 °C) for 90 s above the liquidus
temperature of the investigated alloy of 767 °C (Fig. 2). At the end of planed HIUST
application time, the ultrasonic source was switched off and the melt was left to
room temperature. Pouring temperatures were controlled within an accuracy
of 2 °C. The ultrasonic waves emitted from the transducer and passed through
the acoustic sonotrode were propagated directly into the melt during solidification.
The poured melt became a part of acoustic sonotrode, so the action of ultrasonic en-
ergy on the melt was raised remarkably. For comparison, a sample without the
application of HIUST was prepared at pouring temperature of 800 °C. In the second
group, the application of HIUST was carried out for different vibration times 10, 50,
90, 130 s at optimum pouring temperature based on the results of the first group.
3.2. The second group
2.2. Materials characterization
The second group was carried out at optimum pouring temper-
ature of 800 °C based on the results of the first group for different
ultrasonic vibration times of 10, 50, 90 and 130 s. With the appli-
cation of HIUST for 10 s, the morphologies of primary Mg2Si are
All metallographic specimens were cut from the bottom of castings at the same
position of 10 mm above the sonotrode contact level of cylindrical samples. The
samples were prepared according to usual procedures developed for magnesium al-
loys [38] and etched by solution with 10 ml nitric acid, 30 ml acetic acid, 40 ml

696
M.E. Moussa et al. / Journal of Alloys and Compounds 577 (2013) 693–700
Fig. 4. OM micrographs of the investigated hypereutectic Mg–5 wt.%Si alloy processed (a) without HIUST at pouring temperature of 800 °C and with HIUST for 90 s at
different pouring temperatures of (b) 770 °C, (c) 780 °C, (d) 790 °C, (e) 800 °C and (f) 810 °C.
Fig. 5. Variation of average size of primary Mg2Si of the investigated hypereutectic Mg–5 wt.%Si alloys without and with HIUST at different pouring temperatures for 90 s.
mixture of dendritic and octahedron structure and their average
size are about 58 m, as shown in Fig. 9a. As vibration time in-
vibration times is shown in Fig. 10 for HIUST done at optimum
pouring temperature of 800 °C.
l
creases from 50 to 90 s, most of primary Mg2Si phases become
polyhedral shape and their average size are slightly reduced from
4. Discussion
about 35 (Fig. 9b) to 33
tion time is further increased to 130 s, the average size of primary
Mg2Si increases slightly again to about 45 m, as shown in Fig. 9c.
The variation of average size of primary Mg2Si with the ultrasonic
lm (Fig. 4e), respectively. When the vibra-
Under the present experimental conditions, the effect of the
cooling rate on the modification of primary Mg2Si in hypereutectic
Mg–5 wt.%Si could be negligible. Therefore, the difference in the
l

M.E. Moussa et al. / Journal of Alloys and Compounds 577 (2013) 693–700
697
enhanced heterogeneous nucleation and cavitation-induced den-
drite fragmentation. During the first stage of solidification between
pouring temperatures and the liquidus temperature of the investi-
gated alloy, HIUST can only be attributed to cavitation-enhanced
heterogeneous nucleation. This because dendrite fragmentation
will not be possible at those temperatures since solidification has
not started yet. The cavitation-enhanced heterogeneous nucleation
is further explained by three different mechanisms. The first is
based on the pressure pulse-melting point (Tm) mechanism
[27,34,40,41], where the pressure pulse arising from the collapse
of bubbles alters Tm according to the Clausius–Clapeyron equation
as dTm/dP = Tm(VL À VS)/
the pressure in MPa, VL and VS are the specific volume of the liquid
and the solid phase in cm3/g, respectively, and
H is the latent heat
DH, where Tm is the freezing point in K, P is
D
of freezing in J. An increase in Tm is equivalent to increasing the
undercooling so that an enhanced nucleation event is expected.
The second is based on cavitation-enhanced wetting [26], which
assumes that cavities and cracks on the substrate surfaces and
insoluble non-metallic inclusions that either pre-exist in the melt
or form on cooling during solidification, can be wetted by the melt
under the pressure pulse from the collapse of the bubbles. Conse-
quently, this enables these substrates to act as effective nucleation
sites. The third mechanism [30,40,42] assumes that rapid adiabatic
expansion of gas inside the bubbles created during cavitation und-
ercools the liquid at the bubble–liquid interfaces resulting in
nucleation on the bubble surfaces. Collapsing of these bubbles will
distribute the nuclei into the surrounding liquid producing a signif-
icant number of nuclei in the liquid.
Fig. 6. XRD patterns of the investigated hypereutectic Mg–5 wt.%Si alloys (a)
without and (b) with HIUST at optimum pouring temperature of 800 °C for 90 s.
morphology and size of primary Mg2Si were resulted almost exclu-
sively from the difference in the application of HIUST conditions.
Usually, the microstructure of materials depends on the nucleation
process and growth conditions [31]. For the Mg2Si, its structure
belongs to face centered cube (FCC) and its dendrite arm should
grow along the preferential [100] crystallographic directions
[22]. As a result, the morphologies of primary Mg2Si in the sample
without HIUST are mainly characterized by dendrites with
complex morphologies, as shown in Figs. 4a and 7a.
The refinement and modification of primary Mg2Si can be
mainly attributed to HIUST during the solidification process. This
can be explained as conjugation of two mechanisms: cavitation-
Below the liquidus temperature investigated, corresponding to
the formation of the first solid metal, the cavitation-induced
Fig. 7. (a) SEM micrograph, (b) EDS line scans of Mg and Si, (c) EDS elemental mapping of Mg and (d) EDS elemental mapping of Si for the investigated hypereutectic Mg–
5 wt.%Si alloy without HIUST at pouring temperature of 800 °C.

698
M.E. Moussa et al. / Journal of Alloys and Compounds 577 (2013) 693–700
Fig. 8. (a) SEM micrograph, (b) EDS line scans of Mg and Si, (c) EDS elemental mapping of Mg and (d) EDS elemental mapping of Si for the investigated hypereutectic Mg–
5 wt.%Si alloy with HIUST at pouring temperature of 800 °C for 90 s.
Fig. 9. OM micrographs of the investigated hypereutectic Mg–5 wt.%Si alloys processed with HIUST at optimum pouring temperature of 800 °C for (a) 10 s, (b) 50 s and (c)
130 s.
dendrite fragmentation is more effective and dominant. This mech-
anism assumes that the shock waves generated from the collapse
of bubbles lead to fragmentation of dendrites [29,34,41,43], which
are redistributed through acoustic streaming, thereby increasing
the number of crystals. The coexistence of these two mechanisms
seems to promote a high density of nuclei in the melt, thus leading
to the development of a large number of fine primary Mg2Si crys-
tals which exhibit an octahedral morphologies [8] with a variety of
polygonal outlines in the two-dimensional view, such as triangle,
square, trapezoid and hexagon because of the random cutting an-
gles polishing process of Metallographic samples.
The effect of HIUST at different pouring temperatures on mod-
ification of primary Mg2Si can be explained by the survival of the
ultrasonically induced embryos in the melt. Since these embryos

M.E. Moussa et al. / Journal of Alloys and Compounds 577 (2013) 693–700
699
(4) At optimum pouring temperature (800 °C) of HIUST, the
coarse primary Mg2Si is obtained at both lower and higher
vibration times of 10 s and 130 s, respectively. Thus, fine
polyhedral shape of primary Mg2Si can be obtained at opti-
mum vibration time of 90 s.
Acknowledgement
The authors would like to appreciate the financial support from
Academy of Scientific Research and Technology in the framework
of the scholarship for support of Ph.D study, Ministry of Scientific
Research, Egypt.
Fig. 10. Variation of average size of primary Mg2Si of the investigated hypereutectic
Mg–5 wt.%Si alloys with HIUST at optimum pouring temperature of 800 °C for
different vibration times.
References
[1] Z.Z. Shi, The crystallography of lath-shaped Mg2Sn precipitates in a Mg–Sn–
Zn–Mn alloy, J. Alloys Comp. 559 (2013) 158–161.
are not thermodynamically stable (they are formed only at the
interfaces of the cavitation bubble and the melt during the expan-
sion stage of the cavitation bubbles), they will be dissolved after
the cavitation bubbles are collapsed in the melt [26]. As a conse-
quence of such instability, the number of active nuclei in the melt
was quite low, as well as their thermal stability at pouring temper-
ature (810 °C) of the investigated alloy, resulting in coarse size of
primary Mg2Si, as shown in Fig. 4f. When the pouring temperature
(770 °C) is closer to the liquidus temperature (767 °C) of the inves-
tigated alloy, the melt will be more viscous, so the formation of
cavitation bubbles is more difficult, thus decreasing the quantity
of embryos in the melt and the number of polyhedral morphology
of primary Mg2Si in the final microstructure as shown in Fig. 4b. At
optimum pouring temperature (800 °C in this work), the embryos
life time was long enough to promote a high density of nuclei in
the melt, thus leading to the development of a fine primary Mg2Si
with polyhedral morphology, as shown in Fig. 4e.
The effect of HIUST at optimum pouring temperature of 800 °C
for different vibration times on modification of primary Mg2Si can
be explained as follows: shorter duration (10 s) of ultrasonic vibra-
tion results insufficiency nucleation in the period above the liqui-
dus temperature of investigated alloy as well as the vibration
effect by HIUST has not broken out the secondary branches and ter-
tiary branches. Therefore, mixture of dendritic and octahedron
structures of primary Mg2Si is obtained, as shown in Fig. 9a. On
the other hand, the greater exothermic heating with longer dura-
tion (130 s) of ultrasonic vibration leads to increase the melt tem-
perature in the field of ultrasonic vibration [44] so that some nuclei
remelt and the average size of primary Mg2Si is coarsened as
shown in Fig. 9c. So only if the optimum duration (90 s in this
work) of ultrasonic vibration is applied during the solidification
process of the investigated alloy, the finer polyhedral shape of pri-
mary Mg2Si can be obtained as shown in Fig. 4e.
[2] L. Li, Microstructure and texture evolution during super plastic deformation of
Mg–Re extruded alloy, J. Alloys Comp. 555 (2013) 255–262.
[3] J.S. Zhang, Y. Sun, W.L. Cheng, Z.P. Que, Y.M. Li, L. Liushan, The effect of Ca
addition on microstructures and mechanical properties of Mg–RE based alloys,
J. Alloys Comp. 554 (2013) 110–114.
[4] M. Cong, Z. Li, J. Liu, M. Yan, K. Chen, Y. Sun, M. Huang, C. Wang, B. Ding, S.
Wang, Effect of Ca on the microstructure and tensile properties of Mg–Zn–Si
alloys at ambient and elevated temperature, J. Alloys Comp. 539 (2012) 168–
173.
[5] Z. Huanga, S. Yu, Microstructure characterization on the formation of in situ
Mg2Si and MgO reinforcements in AZ91D/Flyash composites, J. Alloys Comp.
546 (2013) 28–40.
[6] J. Hou, C. Li, X. Liu, Nucleating role of an effective in situ Mg3P2 on Mg2Si in
Mg–Al–Si alloys, J. Alloys Comp. 509 (2011) 735–739.
[7] H.Y. Wang, M. Zha, B. Liu, D.M. Wang, Q.C. Jiang, Microstructural evolution
behavior of Mg–5Si–1Al alloy modified with Sr–Sb during isothermal heat
treatment, J. Alloys Comp. 480 (2009) L25–L28.
[8] H.Y. Wang, L. Chen, B. Liu, X.R. Li, J.G. Wang, Q.C. Jiang, Heterogeneous
nucleation of Mg2Si on Sr11Sb10 nucleus in MgÀx(3.5, 5 wt.%) SiÀ1Al alloys,
Mater. Chem. Phys. 135 (2012) 358–364.
[9] F. Mirshahi, M. Meratian, High temperature tensile properties of modified Mg/
Mg2Si in situ composite, Mater. Des. 33 (2012) 557–562.
[10] X. Zhang, J. Hu, L. Ye, Y. Deng, C. Tang, L. Yang, Z. Liu, Effects of Si addition on
microstructure and mechanical properties of Mg–8Gd–4Y–Nd–Zr alloy, Mater.
Des. 43 (2013) 74–79.
[11] J. Hu, X. Zhang, C. Tang, Y. Deng, Z. Liu, L. Yang, Microstructures and
mechanical properties of the Mg–8Gd–4Y–Nd–Zn–3Si (wt%) alloy, Mater. Sci.
Eng. A 571 (2013) 19–24.
[12] Q.C. Jiang, H.Y. Wang, Y. Wang, B.X. Ma, J.G. Wang, Modification of Mg2Si in
Mg–Si alloys with yttrium, Mater. Sci. Eng. A 392 (2005) 130–135.
[13] D. Jun, K. Iwai, L.W. Fang, P.J. Hua, Effects of alternating current imposition and
alkaline earth elements on modification of primary Mg2Si crystals in
hypereutectic MgÀSi alloy, Trans. Nonferrous Met. Soc. China 19 (2009)
1051–1056.
[14] K. Kondoh, H. Oginuma, R. Tuzuki, T. Aizawa, Magnesium matrix composite
with solid-state synthesized Mg2Si dispersoids, Mater. Trans. 44 (2003) 611–
618.
[15] L.X. Lin, C.Y. Bin, W. Xiang, M.G. Rui, Effect of cooling rates on as-cast
microstructures of Mg–9Al–xSi (x = 1, 3) alloys, Trans. Nonferrous Met. Soc.
China 20 (2010) s393–s396.
[16] L. Lu, K.K. Thong, M. Gupta, Mg-based composite reinforced by Mg2Si, Compos.
Sci. Technol. 63 (2003) 627–632.
[17] T.S. Giu, Z.J. Xue, T.C. Wen, Y.Y. Sheng, Morphology modification of Mg2Si by Sr
addition in Mg–4%Si alloy, Trans. Nonferrous Met. Soc. China 21 (2011) 1932–
1936.
5. Conclusions
[18] R. Alizadeh, R. Mahmudi, Effects of Sb addition on the modification of Mg2Si
particles and high-temperature mechanical properties of cast Mg–4Zn–2Si
alloy, J. Alloys Comp. 509 (2011) 9195–9199.
[19] Y. Hong, H. Yong, W.X. Gian, Influence of Sb modification on microstructures
and mechanical properties of Mg2Si/AM60 composites, Trans. Nonferrous Met.
Soc. China 20 (2010) s411–s415.
[20] K. Chen, Z.Q. Li, J.S. Liu, J.N. Yang, Y.D. Sun, S.G. Bian, The effect of Ba addition
on microstructure of in situ synthesized Mg2Si/Mg–Zn–Si composites, J. Alloys
Comp. 487 (2009) 293–297.
[21] N. Zheng, H.Y. Wang, Z.H. Gu, W. Wang, Q.C. Jiang, Development of an effective
modifier for hypereutectic Mg–Si alloys, J. Alloys Comp. 463 (2008) L1–L4.
[22] H.Y. Wang, W. Wang, M. Zha, N. Zheng, Z.H. Gu, D. Li, Q.C. Jiang, Influence of
the amount of KBF4 on the morphology of Mg2Si in Mg–5Si alloys, Mater.
Chem. Phys. 108 (2008) 353–358.
(1) The HIUST has a significant effect on the refinement and
modification of primary Mg2Si in the hypereutectic Mg–
5 wt.%Si alloy during the solidification process.
(2) The modification mechanism is mainly attributed to the
conjugation of two mechanisms: cavitation-enhanced
heterogeneous nucleation and cavitation-induced dendrite
fragmentation.
(3) For the ultrasonic vibration time of 90 s, the coarse primary
Mg2Si is obtained at both lower and higher pouring temper-
atures of 770 °C and 810 °C, respectively. Therefore, fine
polyhedral shape of primary Mg2Si can be obtained at opti-
mum pouring temperature of 800 °C.
[23] E.J. Guo, B.X. Ma, L.P. Wang, Modification of Mg2Si morphology in Mg-Si alloys
with Bi, J. Mater. Process. Technol. 206 (2008) 161–166.

700
M.E. Moussa et al. / Journal of Alloys and Compounds 577 (2013) 693–700
[24] Z. Na, W.H. Yuan, Z. Feng, G.Z. Hua, L. Dong, J.Q. Chuan, Modification of primary
Mg2Si in Mg-5Si alloys with Y2O3, Trans. Nonferrous Met. Soc. China 17 (2007)
s440–s443.
[35] J. Lan, Y. Yang, X. Li, Microstructure and microhardness of SiC nanoparticles
reinforced magnesium composites fabricated by ultrasonic method, Mater. Sci.
Eng. A 386 (2004) 284–290.
[25] O.V. Abramov, High-Intensity Ultrasonics: Theory And Industrial Applications,
First ed., Gordon and Breach Science Publishers, Amsterdam, 1998.
[26] G.I. Eskin, Ultrasonic Treatment of Light Alloy Melts, First ed., Gordon and
Breach Science Publishers, Amsterdam, 1998.
[27] P.P. Bhingole, G.P. Chaudhari, Synergy of nano carbon black inoculation and
high intensity ultrasonic processing in cast magnesium alloys, Mater. Sci. Eng.
A 556 (2012) 954–961.
[28] Z. Liang, W.G. Hua, W.S. Hua, D.W. Jiang, Effect of cooling condition on
microstructure of semi-solid AZ91 slurry produced via ultrasonic vibration
process, Trans. Nonferrous Met. Soc. China 22 (2012) 2357–2363.
[29] M.K. Aghayani, B. Niroumand, Effects of ultrasonic treatment on
microstructure and tensile strength of AZ91 magnesium alloy, J. Alloys
Comp. 509 (2011) 114–122.
[36] M. E. Moussa, M. A. Waly, A. A. Nofal, A. M. El-Sheikh, Influence of ultrasonic
vibration on the grain refinement and mechanical properties of AZ91
magnesium alloy, in: Proceedings of 1st InterQuadrennial Conference in
Middle East and Africa (IQCMEA-ICF): Processing, Performance and failure
analysis of Engineering Materials, Egypt 2011 pp. 103–113.
[37] M. E. Moussa, M. A. Waly, A. A. Nofal, A. M. El-Sheikh, Effect of ultrasonic
vibration and boron addition on microstructure and mechanical properties of
AZ91 magnesium alloy, in: Proceedings of 8th Arab Foundry symposium
(ARABCAST 2010), Egypt 2010 pp. 1–14.
[38] G.F.V. Voort, Metallography and Microstructures, first ed., ASM Handbook,
Metals Park, OH, USA, 2004.
[39] C. Malcolm, Binary Alloy Phase Diagrams, second ed., ASM International,
Materials Park, OH, USA, 1996.
[30] D. Gao, Z. Li, Q. Han, Q. Zhai, Effect of ultrasonic power on microstructure and
mechanical properties of AZ91 alloy, Mater. Sci. Eng. A 502 (2009) 2–5.
[31] X. Liu, Y. Osawa, S. Takamori, T. Mukai, Microstructure and mechanical
properties of AZ91 alloy produced with ultrasonic vibration, Mater. Sci. Eng. A
487 (2008) 120–123.
[40] J.D. Hunt, K.A. Jackson, Nucleation of solid in an undercooled liquid by
cavitation, J. Appl. Phys. 37 (1966) 254–257.
[41] B. Patel, G.P. Chaudhari, P.P. Bhingole, Microstructural evolution in
ultrasonicated AS41 magnesium alloy, Mater. Lett. 66 (2012) 335–
338.
[32] X. Liu, Y. Osawa, S. Takamori, T. Mukai, Grain refinement of AZ91 alloy by
introducing ultrasonic vibration during solidification, Mater. Lett. 62 (2008)
2872–2875.
[33] Z.Z. Qiang, L.Q. Chi, C.J. Zhong, Microstructures and mechanical properties of
AZ80 alloy treated by pulsed ultrasonic vibration, Trans. Nonferrous Met. Soc.
China 18 (2008) s113–s116.
[42] X. Jian, H. Xu, T.T. Meek, Q. Han, Effect of power ultrasound on solidification of
aluminum A356 alloy, Mater. Lett. 59 (2005) 190–193.
[43] M.C. Flemings, Solidification Processing, McGraw-Hill, NewYork, 1974.
[44] L.J. Wen, T. Momono, F. Ying, J. Zheng, Effect of ultrasonic stirring on
temperature distribution and grain refinement in Al–1.65%Si alloy melt, Trans.
Nonferrous Met. Soc. China 17 (2007) 691–697.
[34] A. Ramirez, M. Qian, B. Davis, T. Wilks, D.H. StJohn, Potency of high-intensity
ultrasonic treatment for grain refinement of magnesium alloys, Scr. Mater. 59
(2008) 19–22.