Materials Transactions, Vol. 47, No. 4 (2006) pp. 977 to 982
Special Issue on Platform Science and Technology for Advanced Magnesium Alloys, III
#2006 The Japan Institute of Metals
The Effect of Aluminium Content on the Mechanical Properties
and Microstructure of Die Cast Binary Magnesium-Aluminium Alloys
Matthew S. Dargusch1;2 , Ketil Pettersen3 , Kazuhiro Nogita1; * ,
Mark D. Nave4 and Gordon L. Dunlop1;2
1
Division of Materials Engineering, The University of Queensland, St. Lucia, Brisbane, QLD 4072, Australia
CRC for Cast Metals Manufacturing (CAST), The University of Queensland, St. Lucia, Brisbane, QLD 4072, Australia
3
Norsk Hydro ASA, Magnesium Materials Technology, P.O. Box 2560, N-3907 Porsgrunn, Norway
4
School of Engineering and Technology, Deakin University, Geelong, VIC, 3217 Australia
2
This paper investigates the relationship between mechanical properties and microstructure in high pressure die cast binary Mg-Al alloys.
As-cast test bars produced using high pressure die casting have been tested in tension in order to determine the properties for castings produced
using this technique. It has been shown that increasing aluminium levels results in increases in yield strength and a decrease in ductility for these
alloys. Higher aluminium levels also result in a decrease in creep rate at 150 C. It has also been shown that an increase in aluminium levels
results in an increase in the volume fraction of eutectic Mg17 Al12 in the microstructure.
(Received November 4, 2005; Accepted February 15, 2006; Published April 15, 2006)
Keywords: magnesium, die casting, mechanical properties
1.
Introduction
The growth in magnesium alloy consumption is driven by
an increasing use in automotive applications.1) The automotive industry accounts for 90% of the casting demand.
Magnesium applications demand different critical properties,
such as creep resistance for automotive drive train components, ductility and energy absorption in safety parts, and
high yield strength in structural parts. In applications where
ductility is important, such as safety parts, magnesium has
become important. In other applications with significant
weight saving potential, for instance drive-train components,
magnesium has so far had limited use, partly due to limited
availability of die cast alloys with the appropriate property
profile. This is the main area of research within alloy
development today.
Consequently, alloy design must be targeted to achieve
specific microstructural features for the various application
areas. The properties of an alloy are closely linked to
composition and processing parameters. Properties for high
pressure die castings are different from those produced by
alternative casting techniques where solidification occurs
under very different conditions. This requires testing on ascast high pressure die cast test bars in order to reliably
determine correct property–microstructure relationships.
It has been widely reported that the level of aluminium
plays a significant role in determining the properties of
magnesium alloys. In most cases however, mechanical
property data for die castings has been reported for
commercial alloys such as AZ91 and AM60 which contain
other alloying elements in addition to aluminium such as
manganese which is added to improve the corrosion
resistance. Mechanical property information for high pressure die cast binary magnesium-aluminium alloys has not
been reported. The effect of increasing aluminium content on
the morphology of the eutectic in high pressure die cast alloys
*Corresponding
author. E-mail: k.nogita@minmet.uq.edu.au
Table 1 Composition of die cast Mg-Al alloys (only compositional
elements present at levels above 0.002 mass% are shown).
Alloy
Al (mass%)
Mn (mass%)
Fe (mass%)
Mg-2%Al
1.92
0.016
0.02
Mg-5%Al
4.88
0.016
0.02
Mg-9%Al
9.44
0.015
0.02
Mg-14%Al
14.1
0.015
0.02
Mg-18%Al
18.2
0.014
0.02
has also not been reported previously. In this paper the effect
of aluminium on the mechanical properties of binary die cast
Mg-Al alloys are investigated and related to the microstructure of these alloys.
2.
Experimental
The compositions of the alloys that were die cast and tested
are given in Table 1. Cylindrical specimens for tensile and
creep testing were die cast in a Frech 200 tonne clamping
force cold chamber die casting machine. 0.3% SF6 in dried
air was used as a protective gas cover. A 3-cavity die that
produced three different test bars was used (Fig. 1.)
The molten metal was hand-ladled into the shot sleeve
requiring a melt temperature 20 K higher than would be
necessary if an automated dosing system with heated transfer
tube was used. The die was equipped with an oil heating/
cooling system set at 513 K. The small shot weight of the
castings (250 g), combined with the low specific heat of the
Mg alloys, necessitated a net input of heat from the oil
heating system. The melt temperature prior to casting was
680 C. The phase 1 piston speed was 0.2 ms 1 , phase 2 was
2.7 ms 1 and the phase 3 piston pressure was 750 bar. The
corresponding gate speed was 75 ms 1 .
Each alloy was tensile tested at room temperature and
creep tested at 150 C. The room temperature tensile tests
were conducted in accordance with ASTM E8-04. Creep
testing was carried out on un-machined die cast specimens in
978
M. S. Dargusch, K. Pettersen, K. Nogita, M. D. Nave and G. L. Dunlop
Table 2 Room temperature (25 C) tensile properties of die cast binary MgAl alloys. Data presented is the average value standard deviation for 10
as-cast tensile test bars.
Alloy (mass%)
3.
Results
Mg-2%Al
86:6 1:6
199:6 2:1
19 2:7
Mg-5%Al
112:2 2:4
236:5 10
16:2 3:6
Elongation (%)
6:3 0:8
147:6 3:0
244:5 8:1
191:2 3:8
255:6 9:7
1:8 0:4
Mg-18%Al
243:9 2:9
253:9 5:5
0:7 0:1
Creep Strain (%)
Mg-9%Al
Mg-14%Al
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
2% Al
5% Al
15% Al
20% Al
0
20
40
60
80
100
120
Time, t/h
Fig. 2 Typical constant load creep curves for high pressure die cast binary
Mg-Al alloys tested at 150 C with an applied stress of 50 MPa.
4
3.5
Creep Strain (%)
accordance with ASTM E139. Temperatures were maintained at 150 2 C during testing and the initial applied
stress was maintained at 50 0:3 MPa throughout each test.
Strain measurements were made using an LVDT arrangement attached directly to the creep specimens.
Microstructural observations were carried out using
optical microscopy, scanning electron microscopy. For
SEM and optical microscopy, the specimens were moulded
into resin and then ground with 240–4000 grit SiC paper.
After grinding the specimens were polished with 6 mm and
then 1 mm diamond paste. A final polish with 0.05 mm silica
suspension was used to eliminate fine scratches. Specimens
were then etched with a solution of 2 vol% nitric acid in
ethanol for 5 seconds at room temperature. Specimens for
investigation using scanning electron microscopy were
coated with either carbon or palladium before inserting into
the microscope.
Optical investigations were performed with a ReichertJung Polyvar-Met optical microscope. SEM investigations
were undertaken using JEOL 6400 and JEOL 820 scanning
electron microscopes operated at 10 kV and a Leo 1530 SEM
operated at 15 kV. Qualitative energy dispersive X-ray
spectroscopy (EDS) analysis was carried out using a Link
Ge EDS spectrometer coupled to a Moran Scientific
Analyser.
Nuclear magnetic resonance (NMR) spectroscopy2–4) was
used to determine the volume fraction of the -Mg17 Al12
phase precipitates in the as-cast binary Magnesium-aluminium alloys.
UTS (MPa)
3
2.5
2
1.5
1
0.5
0
0
5
10
15
20
Aluminium Content (mass%)
Fig. 3 Creep strain at 100 hours vs. Al content for high pressure die cast
binary Mg-Al alloys tested at 150 C with an applied stress of 50 MPa.
0.035
Creep Rate (% / h)
Fig. 1 Complete casting from the test bar die, with an impact test bar (top),
tensile test bar (middle) and creep test bar (bottom).
0.2% YS (MPa)
0.03
0.025
0.02
0.015
0.01
0.005
0
0
5
10
15
20
Aluminium Content (mass%)
3.1 Room temperature mechanical properties
The effect of increasing aluminium content on the room
temperature tensile properties of the die cast binary alloys is
presented in Table 2. Yield Strength increases with increasing aluminium content from 87 MPa at an aluminium content
of 1.9 mass% up to 244 MPa at an aluminium content of
17.8 mass%. Ductility decreased with increasing aluminium
content from a value of 19% at an aluminium content of
1.9 mass% down to 0.7% with 17.8 mass% aluminium.
tested at 150 C are shown in Fig. 2. The results presented in
Figs. 3 and 4 show a clear decrease in both creep strain and
creep rate with increasing aluminium content during creep
testing at 150 C.
3.2 Elevated temperature creep properties
Typical creep curves for die cast binary Mg-Al alloys
3.3 Microstructure
The microstructures of the binary die cast Mg-Al alloys
Fig. 4 Creep rate at 100 hours vs. Al content for high pressure die cast
binary Mg-Al alloys tested at 150 C with an applied stress of 50 MPa.
The Effect of Aluminium Content on the Mechanical Properties and Microstructure of Die Cast Binary Magnesium-Aluminium Alloys
(a)
(b)
(c)
(d)
979
(e)
Fig. 5 Effect of aluminium content on the microstructure of binary Mg-Al alloys: (a) Mg-2 mass%Al, (b) Mg-5 mass%Al, (c) Mg-9
mass%Al, (d) Mg-14 mass%Al, (e) Mg-18 mass%Al.
consist of -Mg solid solution and particles of the Mg17 Al12 phase (Fig. 5). The high cooling rates experienced
in high pressure die casting resulted in the presence of
eutectic -Mg17 Al12 at lower aluminium contents than would
be expected under equilibrium conditions (13 mass% Al). For
the alloys with aluminium contents less than 9 mass%
(Fig. 5(a), (b)), the eutectic is divorced. In these alloys, the
-Mg17 Al12 appears as discrete particles surrounded by
regions of -Mg that are richer in Al than the primary -Mg.
As the aluminium content increases, the eutectic morphology
changes from fully divorced (parts of Fig. 5(b)) to partially
divorced (parts of Fig. 5(b), (c)) to fibrous (Fig. 5(d), (e)).
In the partially divorced structure, there are some islands of
-Mg within the particles of -Mg17 Al12 , while in the fibrous
eutectic structure the two phases are finely intermixed.
With increasing aluminium content, the amount of eutectic
in the binary Mg-Al alloys increases, leading to higher
volume fractions of -Mg17 Al12 and supersaturated Mg
matrix (Fig. 6) shows the volume fraction of -Mg17 Al12
versus aluminium content as measured using NMR techniques. While only very small amounts are present in the
2 mass% Al alloy, the -Mg17 Al12 phase occupies a quarter
of the microstructure in the alloy containing 18 mass% Al.
As observed in other Mg die castings5–7) grain size varies
across the specimen with a generally finer grain size towards
the edge of the cylindrical specimen compared to the centre.
In addition grains that may have nucleated and started to
grow in the shot sleeve may also be present in the microstructure. Measurements of the grain size at the center of the
cylindrical creep specimens showed that the Mg-2 mass%
aluminium alloy castings had an average grain size ranging
from 50–100 mm. A decrease in average grain size at the
center of the die castings was observed with increasing
aluminium content such that the Mg-9 mass% Al alloy had an
average grain size ranging from 10–50 mm. Because Mg has a
large Taylor factor its strength at room temperature is
980
M. S. Dargusch, K. Pettersen, K. Nogita, M. D. Nave and G. L. Dunlop
sensitive to grain size. The observed increase in yield
strength between the 2 mass% Al alloy and the 9 mass% Al
alloy was 180%. This increase in room temperature strength
can be attributed to both the mild refinement of the grain size
along with the substantial increase in the volume fraction of
second phase particles in the higher aluminium content
alloys.
0.3
Vf (β)
0.2
0.1
0
0
4
8
12
16
20
Aluminium Content (mass%)
Fig. 6 Volume fraction of -Mg17 Al12 versus aluminium content in the
binary alloys, measured using NMR.
3.4 Microstructure after creep testing
The microstructures of all alloys tested were very unstable
during creep testing at 150 C. Precipitation of -Mg17 Al12
from the supersaturated solid solution matrix was observed in
the microstructure in the alloys that were tested (Figs. 7, 8).
Along with the other alloys the microstructure of the Mg1.9 mass% alloy changed during the creep test. The as-cast
microstructure of the Mg-1.9 mass% alloy was predominantly constituted by -Mg (Figs. 5, 8) but it also contained very
small amounts of eutectic -Mg17 Al12 and localised areas of
-Mg that were supersaturated with aluminium. During creep
testing at 150 C initially these regions of supersaturated
-Mg decompose resulting in the presence of additional
precipitates in the areas immediately surrounding the -
(b)
(a)
Fig. 7 Microstructures of the Mg-9 mass% Al alloy (a) before and (b) after creep testing.
(b)
(a)
Fig. 8
Microstructures of the Mg-1.9 mass% Al alloy (a) before and (b) after creep testing.
The Effect of Aluminium Content on the Mechanical Properties and Microstructure of Die Cast Binary Magnesium-Aluminium Alloys
Mg17 Al12 . The binary Mg-Al phase diagram shows that
2 mass% Al is able to dissolve into the matrix at 150 C if
given enough time. However after 100 hours of creep testing
at 150 C small amounts of the -Mg17 Al12 still remained in
the microstructure [Fig. 8(b)]. The amount of precipitation
increased with increasing aluminium content of the alloys.
Figure 7(b) shows the substantial amount of precipitation
that has occurred in an alloy containing 9 mass% Al after
creep testing. This concurrent precipitation occurring during
creep testing of the higher aluminium content alloys
continued to have a positive influence on the creep resistance
with decreasing creep rates in the high aluminium content
alloys compared to the lower aluminium content alloys.
4.
Discussion
Pettersen et al.,16) have reported that the higher aluminium
contents in commercial high pressure die cast magnesium
alloys such as AZ91 and AM60 give rise to a decrease in the
ductility of the alloys compared to AM20 as a result of the
much higher volume fractions of eutectic in these alloys
compared to AM20. No detailed study of the effect of
aluminium content on the mechanical properties of highpressure die cast binary Mg-Al alloys has previously been
performed. In the present investigation, no other alloying
elements have been added to the melt before casting in order
to focus exclusively on the role of aluminium in determining
the mechanical properties. The results presented in this paper
show a clear decrease in ductility with increasing aluminium
levels (Table 2) and consequently, higher volume fractions
of eutectic (Fig. 6).
It has been suggested in the literature8–11) that the main
reason for the relatively poor creep properties of AZ91 and
other high aluminium content magnesium alloys such as
AM60, is that the intergranular -Mg17 Al12 phase softens at
elevated temperatures and this results in excessive deformation in the grain boundary regions. Results obtained in the
present work are in disagreement with this hypothesis. It was
shown in Figs. 3 and 4 that substantial increases in the
volume fraction of -Mg17 Al12 , resulted in significantly
improved creep strength of die cast Mg-Al alloys. If the Mg17 Al12 phase softens in the temperature range of interest
and promotes grain boundary deformation then, contrary to
the present results, considerably lower creep strength would
be expected in the higher aluminium content specimens. In
fact, the opposite seems to occur: the creep strength of the
material increases with increasing -phase content. Thus it is
concluded that the presence of -Mg17 Al12 does not decrease
the creep strength of Mg-Al castings at temperatures in the
vicinity of 150 C.
These results are supported by the work of Fukuchi and
Watanabe12,13) who have shown with experiments on bulk
polycrystalline -Mg17 Al12 that this phase only softens
appreciably at temperatures of 260 C or above. This is
explained by the strong Al -Al covalent bonding within the
crystal structure of the phase.
It is likely that solute aluminium may have a beneficial
influence on creep resistance. Sato and Oikawa14) have
proposed that the minimum creep rate ("_) of Mg-Al solid
solution alloys decreases with increasing solute aluminium
981
concentration. In addition to solute strengthening effects it is
most probable that continuous precipitation throughout the
matrix will assist in the reduction of creep rate. Since the
solid solubility of aluminium (at 150 C) was exceeded in
most of the magnesium alloys tested, as cast second phase
particles of -Mg17 Al12 and precipitate -Mg17 Al12 particles
were present in the alloys during and after creep testing. The
volume fraction of these hard particles has been shown to
increase with increasing aluminium content. This increase in
volume fraction corresponded with an increase in room
temperature strength and a decrease in creep rate.
The supersaturated -Mg decomposes via both discontinuous and continuous precipitation during creep testing. It
would appear reasonable to suggest that the presence of the
continuous precipitates could inhibit dislocation creep in the
matrix phase although the strengthening effect of these
precipitates is believed to be quite low as they form generally
as thin plates parallel to the basal plane and therefore do not
present a large cross section to dislocations slipping on this
plane. In addition, these equilibrium phase precipitates
coarsen rapidly and would continue to coarsen as the creep
tests progressed, reducing resistance to dislocation motion.
Nevertheless it seems feasible that the presence of an
increased volume fraction of these types of precipitates, as
the aluminium content of the alloy is increased, will
strengthen the matrix against creep. Alloys with higher
aluminium contents also contain increasing amounts of
interdendritic eutectic Mg17 Al12 . The increase in the volume
fraction of this brittle phase results in a decrease in ductility
at room temperature and does not appear to have any
detrimental effect on the elevated temperature creep properties.
5.
Conclusions
The ductility of binary high pressure die cast Mg-Al alloys
has been shown to decrease with increasing aluminium
content from 2 up to 18 mass%. The yield strength of high
pressure die cast Mg-Al alloys increases with increasing
aluminium content, while the ultimate tensile strength
increases with aluminium content up to 14 mass%.
The creep rate of these alloys when tested at 150 C and an
initial applied stress of 50 MPa decreases with increasing
aluminium content.
Increasing aluminium levels result in an increase in the
volume fraction of eutectic Mg17 Al12 in the microstructure of
these alloys. Thus it is concluded that both the increasing
volume fraction of supersaturated -Mg and the increasing
volume fractions of Mg17 Al12 phase improve both the room
temperature yield strength and creep strength at 150 C in
these alloys despite having a detrimental effect on the
ductility.
REFERENCES
1) M. Easton, T. Abbott and C. Caceres: Materials Science Forum 419–
422 (2003) 147–152.
2) S. Celotto, T. J. Bastow, P. Humble and C. J. Bettles: Proc. of the 3rd
Int. Magnesium Conf., (Institute of Materials, London, 1997) 391.
3) S. Celotto and T. J. Bastow: Acta Mater. 49 (2001) 41–51.
4) T. J. Bastow and M. E. Smith: J. Phys.-Condens. Mat. 7 (1995) 4929–
982
M. S. Dargusch, K. Pettersen, K. Nogita, M. D. Nave and G. L. Dunlop
4937.
5) W. P. Sequeira, G. L. Dunlop and M. T. Murray: Proceedings of the 3rd
International Magnesium Conference, (Institute of Materials, 1996)
63–73.
6) B. E. Carlson: Journal of Materials and Manufacturing 104 (1995) 343–
351.
7) D. Rodrigo, M. Murray, M. Mao, J. Brevick, C. Mobley and R. Esdaile:
Proceedings of the 20th International Die Casting Congress, World of
Die Casting, (North American Die Casting Association, Cleveland,
USA, 1999) 219–225.
8) A. Luo and M. O. Pekguleryuz: J. Mat. Sci. 29 (1994) 5259.
9) J. S. Waltrip: Proc. 47th Ann. World of Magnesium Conf., (International Magnesium Association, Cannes, 1990) 214.
10) F. Hollrigl-Rosta, E. Just, J. Kohler and H. J. Meltzer: Proceedings 37th
World of Magnesium Conference, (IMA, Dayton, USA, 1980).
11) G. Raynor: The Physical Metallurgy of Magnesium and its Alloys
(Pergamon Press, London, 1959).
12) M. Fukuchi and F. Watanabe: J. Japan Inst. Met. 30 (1975) 493.
13) M. Fukuchi and F. Watanabe: J. Japan Inst. Light Met. 30 (1980) 253.
14) H. Sato and H. Oikawa: 9th Int. Conf. on Strength of Metals and Alloys,
(Haifa, Israel, 1991) 463.
15) J. E. Harris and R. B. Jones: J. Nuclear Materials 10 (1963) 360.
16) K. Pettersen, P. Bakke and D. Albright: Magnesium Technology 2002,
(TMS, Seattle, USA) p. 240–246.
17) ASTM E8-04 Standard Test Methods for Tension Testing of Metallic
Materials ASTM International, 01-Apr-2004.
18) ASTM E139-00E1, Standard Test Methods for Conducting Creep,
Creep-Rupture, and Stress-Rupture Tests of Metallic Materials ASTM
International, 10-May-2000.