Materials and Design 30 (2009) 2732–2736

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

Short Communication

Al matrix syntactic foam fabricated with bimodal ceramic microspheres

X.F. Tao, L.P. Zhang, Y.Y. Zhao*
Department of Engineering, University of Liverpool, Brown Hill, Liverpool, L69 3GH, UK

a r t i c l e i n f o a b s t r a c t

Article history: The energy absorption capability of cellular solids is determined by their plateau strength and onset
Received 11 September 2008 strain of densification, which in turn are dependent upon their porosity. Metal matrix syntactic foams
Accepted 7 November 2008 fabricated with ceramic microspheres of a single size range have a nearly fixed porosity and thus have
Available online 17 November 2008
a limited variability in energy absorption. This paper fabricates Al matrix syntactic foams with monomo-
dal or bimodal ceramic microspheres and compares their mechanical properties. The syntactic foams
with bimodal ceramic microsphere have up to 10% higher porosity, which leads to 8% higher onset strain
of densification. The bimodal foams have the advantages of a flat deformation regime, high plateau stress
and good ductility. They are potentially excellent choice for energy absorption applications.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction In this study, Al matrix syntactic foams embedded with CMs of

different size ranges are fabricated. The compressive behavior,
Metal matrix syntactic foams have recently attracted interest of including the onset strain of densification and plateau strength,
many researchers because of their high specific strength and stiff- of the syntactic foams with CMs of a single size range (monomo-
ness and good capability of energy absorption [1–4]. Metal matrix dal) and of dual size ranges (bimodal) are compared. The perfor-
syntactic foams can provide higher compressive plateau strength mance of these materials for energy absorption applications is
than resin matrix syntactic foams [5] because of a stronger matrix discussed.
or closed cell monolithic metal foams [6] because of reinforcement
by embedded hollow or porous ceramic microspheres (CMs). As a
consequence, a large amount of energy can be absorbed due to 2. Experimental procedure
extensive strain accumulation at relatively high plateau stresses
until final densification, where the porous or hollow CMs are fully The raw materials used for fabricating the Al matrix syntactic
crushed. However, metal matrix syntactic foams have a higher foam samples were 6082 Al alloy and CM powder supplied by
density and lower porosity than closed cell metal foams. Such Pty Ltd Australia. The CM powder has a composition of 60%
foams can be manufactured by infiltrating liquid metal into a stack SiO2, 40% Al2O3 and 0.4–0.5% Fe2O3 by weight, and has an effec-
of hollow or porous CMs, where metals, such as aluminum or mag- tive density of 0.6 g/cm3, which is the mass of the powder divided
nesium are used as the matrix and the porosity is provided by the by the volume the particles occupy without the air void between
embedded CMs. It is difficult to increase the porosity and decrease them. Two particle size ranges of CMs, fine (75–125 lm) and
the density of such foams because the volume fraction of the CMs coarse (250–500 lm), as shown in Fig. 1, were used in this study.
is largely fixed. Under the condition of random packing of CMs They have a similar porosity of about 80% but different inner struc-
with a similar size, the volume percentage of CMs is approximately tures. Most of the fine CMs have a hollow inner structure, while the
63% [2]. Since the porosity of the syntactic foam is determined by inner structure of the coarse CMs is dominated by porous type, as
the porosity of the CMs, it can be increased by decreasing the rel- shown in Fig. 2 [7]. In fabricating the syntactic foams, either the
ative wall thickness (the ratio between the wall thickness and the fine, coarse or mixtures of both powders (30%, 50% and 70% fine
sphere radius) of the hollow CMs [1,2]. However, the increase is powder) were used.
limited and there is a disadvantage that the compressive strength Al matrix syntactic foams were fabricated by the melt infiltra-
of the CMs decreases with decreasing relative wall thickness [2], tion casting process. The detailed fabrication process was de-
which leads to decreased compressive strength and thus decreased scribed in [4] and a brief introduction is given here. A block of Al
energy absorption of the syntactic foams. 6082 alloy was placed at the top of a predetermined amount of
CM powder contained in a steel tube and was heated in an electric
furnace at 700 °C for 30 min. The assembly was removed from the
* Corresponding author. furnace and the molten Al alloy was pressed into the CM powder.
E-mail address: y.y.zhao@liv.ac.uk (Y.Y. Zhao). After complete solidification, the syntactic foam sample was

0261-3069/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2008.11.005
 X.F. Tao et al. / Materials and Design 30 (2009) 2732–2736 2733

Fig. 1. Micrographs of the: (a) fine (75–125 lm) and (b) coarse (250–500 lm) CM powders.

Fig. 2. Optical micrographs of the polished cross sections of two syntactic foam samples showing: (a) hollow structure for fine CMs and (b) porous structure for coarse CMs, as
indicated by the arrows [7].

removed from the tube, machined to the desired dimensions and CMs have little effect on the distribution of the latter. When the
polished by sand papers. Standard T6 heat treatment was then per- volume percentage of the fine CMs increases to 50% (Fig. 3d) and
formed on the sample. Specifically, the sample was homogenized 70% (Fig. 3e), however, they not only distribute in the gaps of the
in air at 540 °C for 100 min and then quenched in water, followed coarse CMs, but also disperse in large areas in their own.
by aging at 180 °C for 10 h. The volume percentages of the CMs in the syntactic foams
The densities of the samples were measured by the Archimedes embedded with 100% fine, 100% coarse, 30% fine + 70% coarse,
method. The microstructure was observed by a Nikon optical 50% fine + 50% coarse and 70% fine + 30% coarse CMs are measured
microscope. Quasi-static compression tests were carried out on to be 61.4%, 60.9%, 73.8%, 69.5% and 68.1%, respectively. It is found
cylindrical syntactic foam samples with a diameter about 10 mm that syntactic foams made with monomodal CMs have a similar
and a length about 10 mm. The tests were performed on an Instron volume percentage of CMs. By mixing CMs of two size ranges,
4505 machine and with a cross-head speed of 1mm/min. Three the volume percentage of CMs in the Al matrix syntactic foam
samples of each type of foam were tested to verify the can be increased by up to 13%. From density point of view, it seems
repeatability. the optimal composition of the bimodal CMs lies near 30%
fine + 70% coarse, at which the syntactic foam has the highest vol-
3. Results and discussion ume percentage of CMs and thus the highest porosity or lowest
density.
Fig. 3 shows the microstructure of the five different types of The effect of bimodal packing on the volume percentage of
syntactic foams. The CMs are randomly distributed in the Al CMs in the syntactic foam can be explained as illustrated in
6082 matrix in all samples, resulting in a homogeneous macro- Fig. 4. In a syntactic foam embedded with monomodal CMs, the
scopic structure. The syntactic foams with monomodal CMs, i.e. CMs are randomly packed as shown in Fig. 4a. The volume per-
either fine or coarse, have a similar microstructure except having centage of the CMs in the syntactic foams is lower than that of
different scales, as shown in Fig. 3a and b. Fine CMs have small the close packing of monosized spheres (0.74) but is more or less
interparticle spaces and thus thin Al matrix network (Fig. 3a); fixed. For a stack of monosized coarse CMs, adding fine CMs can
coarse CMs have big interparticle spaces and thus thick Al matrix increase the overall volume percentage of CMs. When the fine
network (Fig. 3b). The syntactic foams with bimodal CMs, i.e. a CMs are fully accommodated in the interstices between the coarse
mixture of fine and coarse, can have different microstructures. In particles, as illustrated in Fig. 4b, the overall volume percentage of
the sample made with 30% fine and 70% coarse CMs (Fig. 3c), the CMs increases with increasing amount of fine CMs. When the
coarse CMs are nearly close-packed and the fine CMs distribute amount of fine CMs is increased further, however, the coarse
in the area between the coarse CMs, replacing pure Al matrix as CMs are pushed apart and areas of randomly packed fine CMs
shown in Fig. 3b. The fine CMs distribute in the gaps of the coarse are formed (Fig. 4c), which is equivalent to monomodal packing.
2734 X.F. Tao et al. / Materials and Design 30 (2009) 2732–2736

Fig. 3. Micrographs of cross sections of the five types of syntactic foams with different CM powders: (a) fine, (b) coarse, (c) 30% fine and 70% coarse, (d) 50% fine and 50%
coarse, and (e) 70% fine and 30% coarse.

As a consequence, the overall volume percentage of CMs in the strengths, between those of the foams with monomodal CMs.
syntactic foam starts to decrease. Their strength increases with increasing volume percentage of fine
The measured density and the calculated porosity of the five CMs.
types of the fabricated foams are presented in Fig. 5. The foams The foams have quite different plateau regimes. The syntactic
with monomodal CMs have a similar density and porosity due to foam with fine CMs shows a brittle plastic deformation where sev-
the similar volume percentage of CMs in the foams. The foams with eral stress drops are observed in the plateau regime. The decreas-
bimodal CMs have lower densities and higher porosities due to the ing stress in the plateau regime is because of the catastrophic
increases of the volume percentage of CMs in the foams. The syn- fracture of the foam, which is associated with the brittle nature
tactic foam with 30% fine + 70% coarse CMs has the highest poros- of the embedded hollow CMs. In contrast, the foam with coarse
ity and the lowest density, because it is close to the ideal packing CMs has a hardening plateau regime, where the plateau stress in-
arrangement as illustrated in Fig. 4b. Compared with foams with creases gradually with increasing strain. This ductile deformation
monomodal CMs, the porosity of this foam is increased by 10% is due to the gradual crush of the embedded porous CMs. By com-
from 49% to 59% and the density decreased by 25%, from 1.41 to bining the fine and coarse CMs, all bimodal foams show a nearly
1.14 g/cm3. perfect plastic plateau regime, where a flat plateau stress is ob-
The representative compressive behavior of the as-fabricated served before entering the densification regime [9].
foams is displayed in Fig. 6. The compressive stress–strain curves The plateau regime is characterised by yield strain, onset strain
of all the samples exhibit the classic regimes for cellular solids, of densification and plateau strength, which determine the energy
which are the linear, plateau, and densification regimes [8]. The absorbing capability of cellular materials. The onset strain of den-
average yield strengths for the syntactic foams with 100% fine, sification and plateau strength of the syntactic foams were deter-
100% coarse, 30% fine, 50% fine and 70% fine CMs are 115.7, 53.3, mined by the energy efficiency method developed by Avalle et al.
57.3, 67.5 and 78.5 MPa, respectively. The foam with fine CMs [10] and modified by Li et al. [9]. The energy absorption capacity
has a much higher yield strength than that of the foam with coarse of the foams is taken as the amount of energy absorbed in the pla-
CMs because the fine CMs are much stronger than the coarse CMs. teau regime before onset of densification. The yield strength, onset
The difference in particle strength is a result of the different inner strain of densification, plateau strength and energy absorption per
structures. The foams with bimodal CMs have medium yield unit weight of the syntactic foams are presented in Table 1. The
 X.F. Tao et al. / Materials and Design 30 (2009) 2732–2736 2735

Fig. 6. Representative compressive stress–strain curves of the as-fabricated foams

with: (a) monomodal CMs and (b) bimodal CMs.

Table 1
Characteristic properties of the syntactic foams in compression.

Foam Porosity Onset Yield Plateau Specific energy

type (%) strain of strength strength absorption
densification (MPa) (MPa) (kJ/kg)
Fine CMs 49.0 0.42 115.7 77.4 25.0
Coarse 48.7 0.43 53.3 63.7 18.6
CMs
30 vol.% 59.0 0.50 57.3 44.4 20.0
fine
Fig. 4. Schematic representative packing of CMs: (a) monomodal, (b) bimodal with CMs
fine particles completely contained within the interstices between coarse particles, 50 vol.% 56.0 0.47 67.5 60.6 22.7
and (c) bimodal with more fine particles than in (b). fine
CMs
70 vol.% 55.6 0.46 78.5 62.9 23.8
fine
CMs

porosity of the foams is also included in Table 1 for comparison

purposes.
The foam with fine CMs has the highest energy absorption per
unit weight due to the highest plateau strength. However, it cannot
be used in energy absorption applications because of its brittle
plastic deformation. It breaks into pieces at a relatively low strain
under compression, whereas the other foams remain intact at the
strain of 0.5. The foam with coarse CMs has the best ductility but
lowest energy absorption per unit weight. Compared with mono-
modal foams, the foams with bimodal CMs have higher onset
strain of densification, a very flat plateau regime and reasonable
ductility. The energy per unit weight absorbed is 25% more than
the foam with coarse CMs and only 10% less than the foam with
fine CMs. Overall, bimodal syntactic foams may be the best choice
Fig. 5. Relationship between density and porosity of the syntactic foams. for optimum performance in energy absorption.
2736 X.F. Tao et al. / Materials and Design 30 (2009) 2732–2736

4. Conclusions [3] Balch DK, Dwyer JGO, Davis GR, Cady CM, Gray III GT, Dunand DC. Plasticity
and damage in aluminum syntactic foams deformed under dynamic and quasi-
static conditions. Mater Sci Eng A 2005;391:408–17.
The mechanical properties of syntactic foams with monomodal [4] Zhang LP, Zhao YY. Mechanical response of Al matrix syntactic foams produced
and bimodal of CMs have been compared. By combining fine and by pressure infiltration casting. J Compos Mater 2007;41:2105–16.
[5] Gupta N, Woldesenbet E, Mensah P. Compression properties of syntactic
coarse CMs, the density of bimodal syntactic foams can be de-
foams: effect of cenosphere radius ratio and specimen aspect ratio. Composites
creased by up to 25%. The bimodal foams have the advantages of Part A 2004;35:103–11.
a flat plateau regime, high plateau stress and good ductility. They [6] Sugimura Y, Meyer J, He MY, Smith HB, Grenstedt J, Evans AG. On the
mechanical performance of closed cell Al alloy foams. Acta Mater
are potentially excellent choice for energy absorption applications.
1997;45:5245–59.
[7] Tao XF, Schleyer GK, Zhao YY. In: Proceedings of the IUTAM symposium on
References mechanical properties of cellular materials, 17–21 September 2007, Paris,
France; in press.
[1] Kiser M, He MY, Zok FW. The mechanical response of ceramic microballoon [8] Gibson LJ, Ashby MF. Cellular solids, structure and properties. UK, Cambridge;
reinforced aluminum matrix composites under compressive loading. Acta 1997.
Mater 1999;47:2685–94. [9] Li QM, Magkiriadis I, Harrigan JJ. Compressive strain at the onset of
[2] Hartmann M, Cröbmann I, Reindel K, Singer RF. Microstructure and mechanical densification of cellular solids. J Cell Plast 2006;42:371–92.
properties of cellular magnesium matrix composites. In: Banhart J, Ashby MF, [10] Avalle M, Belingardi G, Montanini R. Characterisation of energy-absorption
Fleck NA, editors. Metal foams and porous metal structures. Bremen Verlag diagram. Int J Impact Eng 2001;25:455–72.
MIT Publishing; 1999. p. 331–6.