Journal of Minerals and Materials Characterization and Engineering, 2012, 11, 1063-1068
Published Online November 2012 (http://www.SciRP.org/journal/jmmce)
Identifica ti on o f Growth Promoter to Fabrication
SiCp/Al2O3 Ceramic Matrix Composites Prepared by
Directed Metal Oxidation of an Al Alloy
Malkapuram Devaiah1,2, Thodeti Srihari3, Thankappan Pillai Rajasekharan1
1Ceramics & Composites Group, Defence Metallurgical Research Laboratory, Hyderabad, India.
2Department of Mechanical Engineering, Geethanjali College of Engineering & Technology,
Hyderabad, India
3Department of Mechanical Engineering, Sha dan College of Engineering & Technology, Hyderabad, India
Email: devaiah.malkapuram@gmail.com,
Received July 3, 2012; revised August 7, 2012; accepted August 14, 2012
ABSTRACT
SiC particulates reinforced alumina matrix composites were fabricated using Directed Metal Oxidation (DIMOX) pro-
cess. Continuous oxidation of an Al-Si-Mg-Zn alloy with different interlayers (dopents) as growth promoters, will en-
compasses the early heating of the alloy ingot, melting and continued heating to temperature in the narrow range of
950˚C to 980˚C in an atmosphere of oxygen. Varying interlayers (dopents) are incorporated to examine the growth con-
ditions of the composite materials and to identification of suitable growth promoter. The process is extremely difficult
because molten aluminum does not oxidize after prolonged duration at high temperatures due to the formation of a pas-
sivating oxide layer. It is known that the Lanxide Corporation had used a combination of dopents to cause the growth of
alumina from molten metal. This growth was directed, i.e. the growth is allowed only in the required direction and re-
stricted in the other directions. The react nature of the dopants was a trade secret. Though it is roughly known that Mg
and Si in the Al melt can aid growth, additional dopents used, the temperatures at which the process was carried out, the
experimental configurations that aided directed growth were not precisely known. In this paper we have evaluated the
conditions in which co mposites can be grown in large enough sizes for evaluation applicatio n and have arriv ed at a pro-
cedure that enables the fabrication of large composite samples by determining the suitable growth promoter (dopant).
Scanning electron microscopic, EDS analysis of the composite was found to contain a continuous network of Al2O3,
which was predominantly free of grain-boundary phases, a continuous network of Al alloy. Fabrication of large enough
samples was done only by the inventor company and the property measurements by the company were confirmed to
those needed to enable immediate applications. Since there are a large number of variable affecting robust grow th of the
composite, fabrication large sized samples for measurements is a difficult task. In the present work, to identify a suit-
able window of parameters that enables robust growth of the composite has been attempted.
Keywords: Ceramic-Matrix Composites; Scanning Electron Microscopy; Liquid Metal Infiltration; Al2O3, SiC
1. Introduction
The directed oxidation of molten metal has been utilized
to produce ceramic composites. This process offers the
ability to produce unreinforced [1 ], particulate-reinforced
[2] and fiber-reinforced [3,4] composites with a wide
range of composition and microstructures. The composite
formation process involves the rapid reaction of a molten
metal with a gaseous oxidant: for example, the reactions
of a molten aluminum alloy containing a growth dopant
such as a few wt% of Mg and a group IVA elements (Si,
Ge, Sn, or Pb) with air to form alpha aluminum oxide [2].
The reaction is sustained by the wicking of liquid me-
tal—along interconnected microscope channels in the re-
action pro duc t .
Numerous studies now exist of the oxidative grow th of
Al2O3/Al composites from Al alloys into free space [5-7],
experiments on the infiltration of preforms, which are
fewer, including the work of Manor et al. [8] into SiC
and of Nagelberg [9] into Al2O3. As an example, the re-
action of an aluminum alloy with oxygen produces an
Al2O3/metal composite containing both ceramic and me-
tal as three dimensionally interconnected networks [4].
The directed oxidation produced composite materials is
quite flexible and offers the ability to make near-net-
shape low-porosity in situ composites with a wide range
of composition and microstructures [10]. These compos-
ites can, in principle, be designated to have good tough-
Copyright © 2012 SciRes. JMMCE
M. DEVAIAH ET AL.
1064
ness and thermal shock resistances, as well as high stiff-
ness, wear resistance, and high temperature stability. The
reaction product may be formed by itself as a ceramic/
metal composite or as the matrix of a reinforced compos-
ite. In either case, the composite formation reaction is
sustained by the wicking of liquid metal along intercon-
nected microscopic channels within the alumina structure.
The microstructure of the α-Al2O3/metal composite in
the absence of a reinforcing preform has been described
by Breval et al. [11]. The composite in the absence of
preform was characterized by preferred orientation of in-
terconnected Al2O3 perpendicular to the plane of the
original alloy surface and a bond of MgAl2O4 at the in-
terface between the composite and original alloy surface.
In a previous paper, the microstructure development of
Al2O3/metal ceramic bodies grown from an Al-10 wt%
Si-3 wt% Mg alloy was discussed in detail [12]. Nagel-
berg et al. found the growth rate of Al-10 w t% Si-3 wt%
Mg alloy to exhibit activation energy of ~400 kj/mol.
Growth rates varied as the oxygen partial pressure to the
one quarter power. These authors proposed that the over-
all composite growth process involves the transport of
oxygen through the external MgO layer to the molten Al
alloy interface where it dissolves in the molten metal.
The dissolved oxygen diffuses through the molten metal
and then reacts with Al at the metal/Al2O3 interface to
form additional Al2O3. It was proposed that the growth
rate was controlled by the electronic conductivity of the
external MgO layer [13].
Characterizing the initial fabrication of the composite
microstructure in the absence of a reinforcing preform is
an essential component for the fabrication of a detailed
understanding of the growth dopants or promoter. This
paper describes the conversion of molten Al alloy into
Al2O3 by using different growth dopants with help of the
microstructure morphology (scanning electron micros-
copy) and fabrication of large enough sizes of ceramic
matrix composites with SiC particulates as reinforcing
preform.
2. Experimental Procedure
For growth of molten Al-9Zn-8.5Si-1.5Mg alloy in O2
atmosphere was studied by two experiments such as: 1)
The alloy is exposed in O2 atmosphere with different do-
pants to know exact growth promoter (dopant) without
preform. 2) Further experiments were carried with SiC
particulates as reinforcement (preform) to fabricate large
enough sizes SiC/Al2O3 ceramic matrix composite ma-
terials.
In initial experiments to know growth promoter studies,
a rectangular Al alloy ingot was machined into 15 mm ×
15 mm × 15 mm size piece (Total 9 Nos). The surface of
the Al alloy piece was evenly coated with a thin inter-
layer (dopants), namely SnO2, Bi2O3, CaCO3, MgO, ZnO,
TiO2, Y2O3, (SnO2 + Bi2O3), and (MgO + ZnO) and other
sides are coated with gypsum to prevent growth. The
coated Al alloy piece was preheated in an oven for 4 hrs.
The preheated alloy pieces are then placed in zircon sand
mold as shown in Figure 1. Then the samples were hea-
ted to a temperature in the narrow range of 950˚C to
980˚C in an atmosphere of oxygen for a dwell time of 62
hours followed by furnace cooling to room temperature.
Grown sample bodies were sectioned and metallo-
graphically polished. Microscopic observations were
made to determine the formatio n of Al2O3 at the top sur-
face of the sample with different interlayer (dopant) and
growth condition. Where appropriate, SEM and EDS
analysis was performed. Energy Dispersive Spectroscopy
(EDS) was especially helpful in identifying the chemical
composition and sequence of layers in the surface oxide
at the interface between the growing composite and the
air atmosphere. In the case of second experiment, to fab-
ricate large enough size of SiCp/Al2O3 composites with
different volume fractions were prepared by directed metal
oxidation process. This was comprised of two steps
namely preparation of SiC preforms with different vo-
lume fraction and appropriate heat treatment schedule to
aid formation of Al2O3 matrix. The volume fraction of
SiC was carried by using SiC particulates of different grit
sizes namely #100, #120 and #220. The corresponding
particle size, tapped packing density and volume fraction,
as calculated using density of α-SiC (3197 Kg/m3) are
displayed in Table 1. SiC powders were subjected to
artificial oxidation by heat treatment in air at 1100˚C for
4 hours. This would ensure that an adherent coating of
SiO2 would develop on the surface of SiC particles. Sub-
sequently, loose powder preforms (measuring 70 × 70 ×
20, in mm) of oxidized SiC with corresponding tapped
packing densities were contained in refractory crucibles.
Al alloy (Al-9Zn-8.5Si-1.5Mg) blocks were used as
source for formation of Al2O3. The Al alloy block was
ground and polished to good surface finish and cleaned
Figure 1. Zircon sand mold used to grow Al alloy pieces
with different interlayer in O2 at m o sphere.
Copyright © 2012 SciRes. JMMCE
M. DEVAIAH ET AL. 1065
Table 1. Details of volume fractions of composites studied in
this work.
Label Grit size
Average particle
size (µm) Tapped packing
density (Kg/m3) Volume
Fraction
B1 #100 125 1118 0.35
B2 #120 106 1286 0.40
B3 #220 53 1382 0.43
Figure 2. Schematic view of set-up used in DIMOX process
for fabrication of SiCp/Al2O3.
with acetone. Gypsum was coated on five sides of the
block to prevent growth into sides and the top, the bot-
tom side of the Al block is coated with growth at the in-
terface. A schematic of the experimental set-up is shown
in Figure 2. The samples were heated to a temperature in
the narrow range of 950˚C to 980˚C in an atmosphere of
oxygen for a dwell time of 62 hours followed by furnace
cooling to room temperature. Subsequent to cooling, the
composite block shall be subjected to machining in order
to obtain specimens with required dimensions necessary
for various measurements.
3. Results and Discussion
Experiments are performed by Al-9Zn-8.5Si-1.5Mg alloy
with different interlayer (dopants), namely SnO2, Bi2O3,
CaCO3, MgO, ZnO, TiO2, Y2O3, SnO2+ Bi2O3, and MgO+
ZnO to know the Al2O3 formation on the top surface [1].
Al alloy ingot without filler material was exposed into
oxygen; the exposed surface of the alloy ingot was
evenly coated with a thin layer of the selected dopant
(SnO2, Bi2O3, CaCO3, MgO, ZnO, TiO2, Y2O3, (SnO2+
Bi2O3), and (MgO+ ZnO) to promote uniform growth
initiation at temperatures varying from 950˚C to 980˚C in
O2 atmosphere. The details results of the as grown sam-
ples are shown in Table 2. As grown samples in oxygen
atmosphere are taken out from home made zircon sand
mold for examination. A small piece is cut from each
sample by low speed diamond saw and metallographi-
cally polished for scanning electron microscopy observa-
tions. Figure 3(a) shows the experimental sample of Al
alloy without filler material and exposed surface of the
alloy ingot was evenly coated with a thin layer of (SnO2
+ Bi2O3) interlayer (dopant) into air.
From the Figure 3(a) we can observe that the length
of the alloy as been changed and increase longitudinally.
The longitudinal cross section is shown in Figure 3(b). It
can be observed from Figure 3(a) the growth of the alloy
at the top surface of the sample and a hollow shape at the
top in cross section view (Figure 3(b)). Visual examina-
tion of the (SnO2+ Bi2O3) specimens showed that oxide
growth occurred exclusively on the exposed surface of
the metal, the microstructural features if the reaction pro-
duct were used to guide the selection of a growth pro-
moter. Figure 4 shows the macroscopic features of reac-
tion products grown from Al alloy at a process temperature
Table 2. Details of grown and non grown samples.
Label Interlayer Results
B1 SnO2 No growth
B2 Bi2O3 No growth
B3 CaCO3 No growth
B4 MgO No growth
B5 ZnO No growth
B6 TiO2 No growth
B7 Y2O3 No growth
B8 (SnO2 + Bi2O3) Composite grows Vigorously
B9 (MgO + ZnO) No growth
(a)
(b)
Figure 3. A schematic representation of as grown Al alloy
SnO2 + Bi2O3 as interlayer (a) Experimental sample (b)
cross section.
Copyright © 2012 SciRes. JMMCE
M. DEVAIAH ET AL.
1066
of 950˚C to 980˚C in O2 atmosphere using (SnO2+ Bi2O3)
as interlayer (growth promoter) system. Differences in
surface reflectivity reveal that the material is of columnar
character, consistent with the directed nature of the
growth process. Results from SEM observations show a
Al2O3 layer, below MgO and Al2O3 layer is formed and
followed by a metal. The Al2O3 columns in material pro-
cessed at 950˚C to 980˚C contain interpenetrant but fu lly
interconnected networks of Al2O3 and Al.
Typically; they are several millimeters wide and have
indistinct bound aries. Figure 5 show s the EDS line map-
ping observation shows the some microns of O and other
elements available in the parent metal and the dopants
(SnO2+ Bi2O3) also. Similar experiments were conducted
with different interlayer (dopants) which are listed in
Table 2. Figure 6 shows experimental sample and lon-
gitudinal cross section using CaCO3 as interlayer under
similar conditions. From the Figure the experiment sam-
ple clearly show that there in no change in the dimen-
sions of the sample and it retained its original dimension s
of the sample. The microscopic observations with CaCO3
as interlayer is not found any oxidation layer on the top
surface and any Al2O3 content at the top surface of the
grown sample as shown in Figure 7 and 8. Similarly
results are found in other experiments which are done by
other interlayer (dopants) which are listed in Table 2.
Figure 4. Representative grain boundaries between Al2O3
crystallites for as Al alloy with SnO2 + Bi2O3 as interlayer.
Figure 5. EDS line mapping representation of as grown Al
alloy SnO2 + Bi2O3 as interlayer.
Figure 9 shows as grown sample and longitudinal
cross section using Bi2O3 as interlayer under similar con-
ditions. The Figure 10 shows the scanning electron mi-
croscopic observations will not found any Al2O3 content
in the case Bi2O3 as interlayer same results can be ob-
served in EDS reorientation in Figure 11. The above ex-
periments reveal that the vigorous growth was found in
the case of (SnO2 + Bi2O3) as interlayer. Other dopents
which are listed in Table 2 will not found any growth.
From the observed results to fabricated SiC/Al2O3 ce-
ramic matrix composites (SnO2 + Bi2O3) can be used as
interlayer or growth promoter for the fabrication of the
large size SiC/Al2O3 ceramic matrix composites. Then
the experiments are done under similar conditions as a
function of SiC particulates as reinforcement with vol-
ume fractions in the range of 0.35 to 0.43 are used to pre-
pare the preforms to infiltrate into Al alloy to fabriccate
(a)
(b)
Figure 6. A schematic representation of as grown Al alloy
CaCO3 as interlayer (a) Experimental sample (b) cross sec-
tion.
Figure 7. SEM representation of as grown Al alloy CaCO3
as interlayer.
Copyright © 2012 SciRes. JMMCE
M. DEVAIAH ET AL. 1067
Figure 8. EDS of as grown Al alloy with CaCO3 as Inter-
layer.
(a)
(b)
Figure 9. A schematic representation of as grown Al alloy
Bi2O3 as interlayer (a) Experimental sample (b) cross sec-
tion.
Figure 10. SEM representation of as grown Al alloy with
Bi2O3 as interlayer.
Figure 11. EDS representation of as grown Al alloy with
Bi2O3 as interlayer.
large size SiCp/Al2O3 ceramic matrix composites. Figure
12 shows the different dimension of SiCp/Al2O3 ceramic
matrix composites are fabricated in the present work with
dimensions. Figure 12(b) with dimensions measuring 70
× 70 × 20, in mm.
(a)
(b)
Com
p
osite
Hollow
re
g
ion
Residua
l metal
(c)
Figure 12. Examples of composite growth obtained in the
present work (a) the aluminum alloy is allowed to oxidize
and grow into SiC particulate preform in a conical refrac-
tory mold. We note that the shape of the mould could be
replicated in the composite. (b) We also note the large size
(70 mm × 70 mm × 20 mm) of the composite samples. An
example of growth of composite from a rectangular mold. (c)
a example showing the way the composite grows-the growth
process was stopped midway, the outer surface was cleaned
and suitably machined to reveal the salient features. The
aluminum left after the process can be seen at the bottom.
The hollow region is the region from where the Al was used
up in the oxidation growth. The top is the composite that
grow. The composite is machined to reveal the absence of
macroscopic porosity. The edges of the alloy retain a box
shape because of the suppression of melting due to the
presence of gypsum coating.
Copyright © 2012 SciRes. JMMCE
M. DEVAIAH ET AL.
Copyright © 2012 SciRes. JMMCE
1068
4. Conclusion
The process details are reported for the specific example
of oxidation of molten Al in oxygen atmosphere to form
Al2O3. In the present work, the experiments are per-
formed with and without reinforcement to fabricate large
size SiCp/Al2O3 ceramic matrix composites. (SnO2 + Bi2O3)
was identified as suitable growth promoter for fabrication
SiCp/Al2O3 ceramic matrix composites. The scanning
electron observations r evealed that the conversion of me-
tal to oxide was po ssible on ly with (SnO2+ Bi2O3) dopant
as growth promoter. Other did not cause any growth of
oxide from molten metal. So, the (SnO2+ Bi2O3) dopant
was used as growth promoter for the fabrication of SiCp/
Al2O3 ceramic matrix composites. SiCp/Al2O3 ceramic
matrix composites are successfully fabricated with SiC
particulates as reinforcement with volume fractions in th e
range of 0.35 to 0.43 with dimensions measuring 70 × 70
× 20, in mm. the sample are used for different mechani-
cal and physical property measurements.
REFERENCES
[1] M. S. Newkirk, A. W. Urquhart and H. R. Zwicker,
“Formation of Lanxide Ceramic Composite,” Journal of
Materials Research, Vol. 1, No. 1, 1986, pp. 81-89.
doi:10.1557/JMR.1986.0081
[2] M. S. Newkirk, H. D. Lesher and D. R. White, “Forma-
tion of Lanxide Ceramic Matrix Composites: Matrix
Formation by the Directed Oxidation of Molten Metals,”
Ceramic Engineering & Science Proceedings, Vol. 8, No.
7-8, 1987, pp. 879-885.
doi:10.1002/9780470320402.ch46
[3] P. Barron-Antolin, G. H. Schiroky, and C. A. Anderson,
“Properties of Fiber-Reinforced Alumina Matrix Com-
posites,” Ceramic Engineering & Science Proceedings,
Vol. 9, No. 7-8, 1988, pp. 759-766.
doi:10.1002/9780470310496.ch27
[4] K. Aghajanian, N. H. MacMillan, C. R. Kennedy, S. J.
Luszcz and R. Roy, “Properties and Microstructures of
Lanxide Al2O3-Al Ceramic Composite Materials,” Jour-
nal of Materials Science, Vol. 24, No. 2, 1989, pp. 658-
670. doi:10.1007/BF01107457
[5] A. S. Nagelberg, “Growth Kinetics of Al2O3/Metal Com-
posites from a Complex Aluminum Alloy,” Solid State
Ionics, Vol. 32-33, 1989, pp. 783-788.
doi:10.1016/0167-2738(89)90358-5
[6] A. S. Nagelberg, S. Antolin and A. W. Urquhart, “Forma-
tion of Al2O3/Metal Composites by the Directed Oxida-
tion of Molten Aluminum-Magnesium-Silicon Alloys:
Part II, Growth Kinetics,” Journal of the American Ce-
ramic Society, Vol. 75, No. 2, 1992, pp. 455-462.
doi:10.1111/j.1151-2916.1992.tb08201.x
[7] O. Salas, H. Ni, V. Jaya ram, K.C. Vlach, C.Q. Levi and R.
Mehrabian, “Nucleation and Growth of Al2O3/Metal
Composites by Oxidation of Aluminum Alloys,” Journal
of Materials Research, Vol. 6, No. 9, 1991, pp. 1964-
1981. doi:10.1557/JMR.1991.1964
[8] E. Monar, H. Ni, C.G. Levi. and R. Meharabin, “Forma-
tion of Al2O3/Metal Composites by the Directed Oxida-
tion of Molten Aluminum-Magnesium-Silicon Alloys: Part
II, Growth Kinetics,” Journal of the American Ceramic
Society, Vol. 75, No. 2, 1992, pp. 455-462.
doi:10.1111/j.1151-2916.1992.tb08201.x
[9] A. S. Nagelberg, “The Effect of Processing Parameters on
the Growth Rate and Microstructure of Al2O3/Metal Ma-
trix Composites,” Materials Research Society Symposium
Proceedings, Vol. 155, 1989, pp. 155-275.
[10] A. W. Urquhart. “Molten Metals Sire MMC’s, CMC’s,”
Advanced Materials and Processes, Vol. 140, No. 1, 1991,
pp. 25-29.
[11] E. Breval, M. Aghajanian and S. Luszcz, “Microstruc-
ture and Composition of Alumina/Aluminum Composites
Made by Directed Oxidation of Aluminum,” Journal of
the American Ceramic Society, Vol. 73, No. 9, 1990, pp.
2610-2614. doi:10.1111/j.1151-2916.1990.tb06735.x
[12] S. Antolin and A. S. Nagelberg, “Formation of Al2O3/
Metal Composites by the Directed Oxidation of Molten
Aluminum-Magnesium-Silicon Alloy: Part I, Micro-
structural Development,” Journal of the American Ce-
ramic Society, Vol. 75, No. 2, 1992, pp. 447-454.
doi:10.1111/j.1151-2916.1992.tb08200.x
[13] A. S. Nagelberg and S. Antolin, “Formation of Al2O3/
Metal Composites by the Directed Oxidation of Molten
Aluminum-Magnesium-Silicon Alloy: Part II, Growth
Kinetics,” Journal of the American Ceramic Society, Vol.
75, No. 2, 1992, pp. 455-462.
doi:10.1111/j.1151-2916.1992.tb08201.x