Materials and Design 30 (2009) 35543559

Contents lists available at ScienceDirect

Materials and Design

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

Effect of equal channel angular pressing (ECAP) on microstructure

and properties of AlSiCp composites
G. Ramu a,b, Ranjit Bauri a,*
a
b

Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600 036, India
Research Centre Imarat (RCI), Vignayanakancha, Hyderabad 500 069, India

a r t i c l e

i n f o

Article history:
Received 13 January 2009
Accepted 2 March 2009
Available online 9 March 2009
Keywords:
Metal matrix composites (A)
Stir casting (C)
Mechanical properties (E)

a b s t r a c t
Aluminium metal matrix composites were processed by stir casting technique with commercially pure
aluminium as the matrix material and 5, 10 vol.% of SiC (30 lm) particulates as the reinforcement.
AlSiCp composites were subjected to severe plastic deformation through equal channel angular pressing
(ECAP). As cast AlSiCp composites shows uniform distribution of particles and there is no change in the
particle distribution after ECAP. It was found that ECAP renes the grain size of matrix material. In
Al5 vol.% SiCp composite grain size was reduced to 8 lm from 45 lm after two ECAP passes whereas
in Al10 vol.% SiCp composite grain size was rened to 16 lm from 45 lm after rst ECAP pass. Hardness
and compression test were conducted at room temperature to evaluate the mechanical properties of the
composites. Both the composites exhibit improvement in mechanical properties after ECAP.
2009 Elsevier Ltd. All rights reserved.

1. Introduction
In recent years ceramic particle-reinforced metal matrix composites (MMCs) have gained wide acceptance because of their
attractive properties. Amongst MMCs aluminium matrix composites (AMMCs) have received particular attention in the past three
decades due to their high specic strength and stiffness and superior wear resistance [15].
Number of processing routes has been developed for the manufacture of particle/whisker/short ber-reinforced composites [2,6
9]. Melt stirring or stir casting technique is currently one of the
simplest and most economical fabrication routes for manufacturing particle-reinforced metal matrix composites. A careful control
of process variables yields high degree of microstructural integrity
and uniformity in this technique [10,11].
Severe plastic deformation is a useful processing tool to rene
the grain size to the submicron or even nanometer size [1214].
Although several severe plastic deformation techniques are available, equal channel angular pressing is an attractive process because it has potential to produce large samples [1214]. Most
investigations on ECAP have concentrated on pure metals and
metallic alloys. Some limited reports are available on application
of ECAP to metal matrix composites [1518]. However, reinforcements in the composites used in these studies are primarily
Al2O3 and no reports are found on Al metal matrix composites reinforced with SiC particles. Further, the ECAP process in these studies

* Corresponding author. Tel.: +91 44 22574778; fax: +91 44 22574752.

E-mail address: [email protected] (R. Bauri).
0261-3069/$ - see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2009.03.001

was carried out at elevated temperature. The aim of the present

work is to fabricate AlSiC composites by equal channel angular
pressing at room temperature and evaluate the effect of SiC particles on the deformation behavior during ECAP. The effect of ECAP
on the microstructure and mechanical properties of AlSiCp composites will also be evaluated.

2. Experimental details
2.1. Material selection and composite processing
The composites used here were processed by stir casting. Commercial pure Al was used as a matrix material in the present study.
Silicon carbide particles of average particle size 30 lm were used
as reinforcement. Fig. 1 shows the schematic of the stir casting
set up used in this study. Weighed quantity of aluminium was
taken for melting and corresponding SiC powder was weighed
according to the required volume fraction. SiC powder was preheated at 850 C for 4 h. This treatment was given to ensure good
wettability between SiCp and liquid aluminium [9]. Once Al was
melted, temperature was measured with the help of a thermocouple and degassing was carried out with argon gas for 5 min. After
degassing, a preheated graphite impeller was lowered in to the
molten metal and rotated by a shaft connected to a motor
(Fig. 1). Preheated SiC particles were added by an addition chute
through the periphery of the vortex, which was created by the stirring action of the rotating mechanical impeller. During stirring
small pieces of magnesium was added to the melt to improve the
wettability of SiC particles with Al melt [9]. Once all of the powder

3555

G. Ramu, R. Bauri / Materials and Design 30 (2009) 35543559

different portions. Weight of composite pieces was measured using

an electronic balance having an accuracy of 0.1 mg. Diluted hydrochloric acid was used for dissolution. Aluminium was dissolved in
HCl and SiC particles were settled at the bottom of the beaker as a
residue. Undissolved SiC powder was carefully ltered and dried in
an oven at 120 C. Weight of dried powder was measured using
electronic balance and the corresponding volume fractions were
measured using the following relation.

Vp

mp

qp

!,

mp

qp

mm

qm

where Vp volume fraction of particles. mp, qp mass and density of

the particles, mm, qm mass and density of matrix.

Fig. 1. Schematic of the stir casting set-up used for the fabrication of Al/SiCp
composite.

is added, the motor is stopped and the impeller is taken out from
the melt. Finally, the composite melt was poured in preheated
cylindrical cast iron molds. Two composites having 5 vol.% and
10 vol.% of SiCp were processed.
2.2. Equal channel angular pressing of AlSiCp composites

2.2.1. Sample preparation for equal channel angular pressing (ECAP)

Samples of required diameter were machined from the cast
composite billets for ECAP. The composite billets were encapsulated in pure aluminium hollow cylinder with a wall thickness of
2 mm to avoid die damage by the hard composite surface during
ECAP and also to prevent instabilities at the surface of the composite that may occur during ECAP [19]. The size of the nal sample
was 19.3 mm in diameter and 90 mm in length. Press-t encapsulated samples were annealed at 400 C for 4 h to homogenize
microstructure and relieve stresses introduced due to press tting.
2.2.2. Equal channel angular pressing (ECAP)
Annealed composite billets were subjected to ECAP using a
channel angle of 120. MoS2 was used as lubricant to minimize friction between die wall and the sample. Bc route, where the sample
is rotated by 90 in one direction either clockwise or anti clockwise
after each successive pass, was adopted in the present work. The
ECAP was carried out at room temperature. Samples were taken
after each pass for characterization. The ECAP experiments were
repeated once to verify the reproducibility of the results. The pure
Al layer was removed by machining for making samples for microstructural and mechanical behavior studies.
2.3. Characterization
2.3.1. Density and porosity
Density of AlSiCp composite was measured using Archimedes
principle. Distilled water was used as immersion uid. Theoretical
density was calculated by rule of mixture and compared with measured densities. Percent porosity was calculated from the difference between measured and theoretical densities.
2.3.2. Volume fraction
Volume fraction of as cast composites was measured by chemical dissolution method. Small pieces of composite were cut from

2.3.3. Microstructural characterization

Sliced samples of composites were polished with emery paper
up to 400 grit size, followed by polishing with Al2O3 suspension
on velvet cloth. Finally samples were polished with 0.5 l diamond
paste. Microstructural characterization was done by scanning electron microscope (SEM) and optical microscope. Particle distribution, interface features, particle orientation were observed in as
cast, as annealed and as ECAPed condition. Fracture surface of
cracks formed after second ECAP pass in Al/10SiCp was also observed under SEM.
2.3.4. Microhardness
Microhardness of the composites was measured in as cast, as
annealed conditions and after each ECAP pass using a Vickers
micohardness tester (WILSON WOLPERT). A load of 1 kg and a
dwell time of 9 s was used. Hardness was measured on the matrix
of the composites. The values reported are average of at least ve
readings.
2.3.5. Compressive test
Room temperature compressive tests were carried out using
standard ASTM samples on an Instron machine in as cast and as annealed condition and after each ECAP pass. Cylindrical samples
with an aspect ratio of 1.2 were used for the compressive tests. A
strain rate of 1.8  103 s1 was used. At least three samples were
tested for each material.
3. Results and discussion
3.1. Density and porosity
Table 1 shows the results of density and porosity measurements. Almost 100% densication was achieved in the unreinforced metal. However, some amount of porosity was observed
in the composites in as cast condition. Porosity in Al/5SiCp was
found less compared to the Al/10SiCp composite. The porosity is
mainly due to gas entrapment during stirring of melt in open
atmosphere. Longer stirring time and higher SiC content in Al/
10SiCp composite can be attributed to comparatively higher porosity in this composite.
3.2. Microstructural characterization
SEM micrographs in Fig. 2a and b shows the distribution of SiC
particles in as cast Al/5SiCp and Al/10SiCp composites, respectively.
The distribution is fairly uniform and there was no segregation or
clustering of the particles. The interfacial bonding between SiC particle and matrix was also observed. The bonding between particle
and matrix is good and there was no indication of any interfacial
reactions. Fig. 3a shows the particle matrix interface in the Al/
10SiCp composite. The corresponding EDAX (Fig. 3b) taken on the

3556

G. Ramu, R. Bauri / Materials and Design 30 (2009) 35543559

Table 1
Densities and porosity of pure Al and its composites.
Material

Measured density
(gm/cm3)

Theoretical density
(gm/cm3)

Porosity
(%)

As cast aluminium
As cast Al/5SiCp
As cast Al/10SiCp

2.690
2.683
2.685

2.70
2.72
2.75

0.37
1.36
2.63

interface conrms this to be interface between the Al matrix and

SiC particles. SEM micrograph in Fig. 2c shows particle distribution
after ECAP. No change in particle distribution or orientation was
observed after ECAP. There was no change in particle size which
means particles were not fractured during ECAP. Optical micrographs of the etched samples before and after ECAP are shown in
Fig. 4. Fig. 4a and b shows the grain size of Al/5SiCp composite in
as annealed and ECAPed (2nd pass) conditions, respectively. A considerable amount of grain renement was observed after ECAP.
Similar observations were made for Al/10SiCp composites as can
be seen from optical micrographs in Fig. 4c and d (as annealed
and after 1st ECAP pass, respectively). The grain size of both the
composites in as annealed condition was observed to be 45 lm.
It is interesting to note that the grain size of both the composite
after annealing is equal. After second ECAP pass grain size of Al/
5SiCp is reduced to 8 lm. In Al/10SiCp grain size was reduced to
16 lm after rst ECAP pass. In case of the Al/10SiCp composite surface cracks starts appearing after the second ECAP pass (inset picture in Fig. 5a). Semiatin et al. [20] investigated the deformation
behavior during ECAP for materials having markedly different

strain rate sensitivity (m) values. Metal matrix composites (MMCs)

generally exhibits higher strain rate sensitivity than the base metal
[2124]. The authors of Ref. [20] report that deformation during
ECAP process in strain-hardenable materials with higher m value
is non-uniform. They have shown that for such a material the
deformation takes place in a relatively diffuse zone with higher
strain rates near the front leg radius and the bottom corner of
the die. They have also shown that the stress distribution across
the billet is also non-uniform. The central portion of the billet
experiences compressive stress while top surface is subjected to
tensile stress. This leads to formation of cracks propagating vertically from the top surface of the billet. Since similar kind of surface
cracks were observed (inset picture in Fig. 5a) on the Al/10SiCp
composite, formation of these cracks can be attributed to non-uniform stress distribution and deformation in the billet from top surface to center of the billet. The fracture surface was observed using
SEM. Fig. 5 shows the SEM micrographs of fracture surface of Al/
10SiCp after second ECAP pass. Particulate reinforced composites
fail by a mixed mode [25] and gives rise to a bimodal distribution
of dimples. As can be seen from Fig. 5a larger dimples are associated with SiC particles whereas smaller dimples are associated
with ductile failure of the matrix. Fig. 5b shows a magnied image
of Fig. 5a. The micrograph in Fig. 5b clearly shows that particle
debonding has taken place at particlematrix interface. In a composite the load is transferred from the matrix to the stronger reinforcement via the reinforcementmatrix interface [26]. The high
shear strains generated in the ECAP process might have led to
the separation at the interface. Using Cockcroft-and-Latham [27]
approach Semiatin et al. [20] have also shown that the number

Fig. 2. SEM micrograph of composites (a) Al/SiCp as cast, (b) Al/10SiCp as cast, and (c) Al/10SiCp after 1st ECAP pass.

G. Ramu, R. Bauri / Materials and Design 30 (2009) 35543559

3557

Fig. 3. (a) Particlematrix interface and (b) the corresponding EDAX taken at the interface in Al/10SiCp composite.

of ECAP passes without fracture is proportional to the tensile fracture strain (ef) of the material. Since the fracture strain decreases
with increasing reinforcement content in MMCs [28,29], number
of passes without fracture is one in the composites containing
higher volume fraction of SiC (Al/10SiC) compared to two for Al/
5SiC composite. The strain rate sensitivity increases with increasing SiC content [24] and this may lead to more non-uniform deformation and early failure of Al/10SiC composite during ECAP. Higher
volume fraction of SiCp also causes more strain hardening during
plastic deformation [30,31] and as a result the ductility comes
down after rst pass in the Al/10SiCp composite. This may also increase the chance of cracking of the billet during the 2nd ECAP
pass. In the composite containing lower volume fraction of SiCp

i.e. Al/5SiCp these effects were less prominent and therefore two
ECAP passes without fracture was possible in this composite.
3.3. Hardness
Table 2 shows the results of the Vickers hardness measurements. The microhardness of both the composites are higher than
the unreinforced pure metal in as cast condition. The hardness increases with increasing SiC content. A considerable enhancement
in the hardness is observed after ECAP. It can also be observed that
hardness has decreased after annealing compared to as cast condition. This is attributed to annihilation of dislocations due to recovery process that takes place during annealing.

Fig. 4. Optical micrographs of (a) Al/5SiCp as annealed, (b) Al/5SiCp after second ECAP pass, (c) Al/10SiCp as annealed, and (d) Al/10SiCp after rst ECAP pass.

3558

G. Ramu, R. Bauri / Materials and Design 30 (2009) 35543559

10SiCp shows higher strength than Al/5SiCp composite due to presence of higher amount of SiC particles. Dislocations are generated
due to mismatch in thermal expansion coefcient between the matrix and the reinforcement [32,33]. As a result the matrix of composites contains higher dislocation. Higher the volume fraction of
reinforcement higher will be the dislocation density. This leads
to higher hardness and strength with increase in SiC content. The
0.2% proof stress increases considerably after each ECAP pass. An
increment of 42% and 29% is noticed in Al/5SiCp and Al/10SiCp composite, respectively, after rst ECAP pass. After second ECAP pass,
0.2% proof stress was increased by 75% in Al/5SiCp composite. As
described above ECAP leads to signicant grain renement in the
composites. This in turn leads to increment in the strength according to the HallPetch relation. Grain renement also leads to increased hardness after ECAP as described above. The measured
properties (hardness and proof stress) of the composites were
compared with those reported in the literature [3438] for ECAPed
pure Al in order to asses the effect of SiC particles. It was found that
hardness and 0.2% proof stress of both the composites after ECAP
are higher than ECAPed pure Al. e.g. ow stress of pure Al after
eight ECAP passes is reported to be 132 MPa [35] and 126 MPa
after two ECAP passes [36]. As can be seen from Table 3, 0.2% proof
stress of Al/5SiCp after 2nd ECAP pass is considerably higher than
these values. Similarly the hardness values (Table 2) are also significantly higher than that reported in the literature [34] for pure Al.
Although the Al/10SiCp composite cracks after the second pass, it
should be noted that there is a considerable improvement in hardness and strength even after one pass. It should also be noted that
all the ECAP experiments were carried out at room temperature.
This opens up possibilities of further studies on high temperature
ECAP of AlSiC composites as has been done previously for some
other Al alloy composite containing a relatively softer reinforcement [1518].
Fig. 5. Fracture surface of Al/10SiCp after 2nd pass. Inset picture in (a) shows the
cracks on the billet. High magnication micrograph in (b) shows the particle
matrix debonding at the interface.

3.4. Compressive test

Table 3 shows the 0.2% proof stress of the composites in as cast
and annealed conditions and after each ECAP pass. As expected Al/

Table 2
Vickers hardness values (HV1)a.
Material

As cast

Annealed

After ECAP
(1st pass)

After ECAP
(2nd pass)

Pure aluminium
Al/5SiCp
Al/10SiCp

24.5
29.4
44.3

26.8
38.0

48.5
50.2

61.2

a
Corresponding values for pure Al as reported in the literature [34] are 18.6, 40
and 44 in as annealed condition, after 1st ECAP pass and after 2nd pass,
respectively.

4. Conclusions
 AlSiCp composites with 5 and 10 volume fractions of SiCp were
successfully produced by stir casting route. Castings were free
from porosity and segregation or clustering with fairly uniform
distribution of SiC particles in aluminium matrix.
 ECAP of the AlSiCp composites were done in Bc route with die
angle of 120. ECAP of Al/5SiCp composite was successfully carried upto two passes. Whereas in case of Al/10SiCp composite
only one pass was possible. In the second pass cracks appears
at the surface of Al/10SiCp composite. Further studies are needed
to see if the number of passes can be increased by high temperature ECAP.
 Hardness and 0.2% proof stress of AlSiCp composites with 5 and
10 vol.% SiCp were increased signicantly by EACP.
 Considerable grain size renement was observed in Al/5SiCp and
Al/10SiCp composites after ECAP.
 No particle breakage was observed after ECAP. There was no
change in SiC particle orientation or distribution due to ECAP.
Acknowledgements
The authors would like to thank Dr. Uday Chakkingal, Associate
Professor, Dept. of Metallurgical and Materials Engineering, IIT Madras, India, for providing the ECAP facility used in this study.

Table 3
0.2% proof stress of Al/SiCp composites before and after ECAPa.
Material

Al/5SiCp
Al/10SiCp
a

0.2% proof stress (MPa)

As cast

Annealed

After
(1st ECAP pass)

After
(2nd ECAP pass)

89
98

85
90

121
125

149

Flow stress of pure Al after eight ECAP passes: 132 MPa [35]; / = 90, w = 20.
126 MPa after 2nd pass, / = 90, w = 90 [36].

References
[1] Miller WS, Humphreys FJ. Strengthening mechanisms in particulate metal
matrix composites. Scr Metall Mater 1991;25:338.
[2] Lloyd DJ. Particle reinforced aluminium and magnesium matrix composites. Int
Mater Rev 1994;39:123.

G. Ramu, R. Bauri / Materials and Design 30 (2009) 35543559

[3] Arsenault RJ, Wang L, Feng CR. Strengthening of composites due to
microstructural
changes
in
the
matrix.
Acta
Metall
Mater
1991;39:4757.
[4] Pramila Bai BN, Ramashesh BS, Surappa MK. Dry sliding wear of A356AlSiCp
composites. Wear 1992;157:295304.
[5] Kwok JKM, Lim SC. High-speed tribological properties of some Al/SiCp
composites: I. Frictional and wear-rate characteristics. Compos Sci Technol
1999;59:5563.
[6] Surappa MK, Rohatgi PK. Preparation and properties of cast aluminium
ceramic particle composites. J Mater Sci 1981;16:98393.
[7] Singer ARE. Metal matrix composites made by spray forming. Mater Sci Eng A
1991;135:137.
[8] Srivatsan TS, Ibrahim IA, Mohamed FA, Lavernia EJ. Processing techniques for
particulate-reinforced metal aluminium matrix composites. J Mater Sci
1991;26:596578.
[9] Hashim J, Looney L, Hashmi MSJ. The enhancement of wettability of SiC
particles in cast aluminium matrix composites. J Mater Process Technol
2001;119:32435.
[10] Hashim J, Looney L, Hashmi MSJ. Particle distribution in cast metal matrix
composites Part I. J Mater Process Technol 2002;123:2517.
[11] Hashim J, Looney L, Hashmi MSJ. Particle distribution in cast metal matrix
composites Part II. J Mater Process Technol 2002;123:25863.
[12] Berbon PB, Langdon TG, Tsenev NK, Valiev RZ, M Furukawa, Horita Z, et al.
Fabrication of bulk ultrane-grained materials through intense plastic
straining. Metall Mater Trans A 1998;29:223743.
[13] Valiev RZ, Islamgaliev RK, Alexandrov IV. Bulk nanostructured materials from
severe plastic deformation. Prog Mater Sci 2000;45:10389.
[14] Nakashima K, Horita Z, Nemoto M. Development of a multipass facility for
equal-channel angular pressing to high total strains. Mater Sci Eng A
2000;281:827.
[15] Valiev RZ, Islamgaliev RK, Kuzmina NF, Li Y, Langdon TG. Strengthening and
grain renement in an Al-6061 metal matrix composite through intense
plastic straining. Scr Mater 1999;40:11722.
[16] Li Y, Langdon TG. Equal-channel angular pressing of an Al-6061 metal matrix
composite. J Mater Sci 2000;35:12014.
[17] Kawasaki M, Huang Y, Xu C, Furukawa M, Horita Z, Langdon TG. A quantitative
study of cavity development in the tensile testing of an aluminum metal
matrix composite processed by equal-channel angular pressing. Mater Sci Eng
A 2005;410411:4027.
[18] Sabirov I, Kolednik O, Valiev RZ, Pippan R. Equal channel angular pressing of
metal matrix composites: effect on particle distribution and fracture
toughness. Acta Mater 2005;53:491930.
[19] Bauri R, Surappa MK. Processing and compressive strength of AlLiSiCp
composites fabricated by a compound billet technique. J Mater Process
Technol 2009;209:207784.
[20] Semiatin SL, Delo DP, Shell EB. The effect of material properties and tooling
design on deformation and fracture during equal channel angular extrusion.
Acta Mater 2000;48:184151.

3559

[21] Perng CC, Hwang JR, Doong JL. High strain rate tensile properties of an (Al2 O3
particles)(Al alloy 6061-T6) metal matrix composite. Mater Sci Eng A
1993;171:21321.
[22] Hong SI, Gray GT. Dynamic mechanical response of a 1060 Al/Al2 O3
composite. J Mater Sci 1994;29:298792.
[23] Yadav S, Chichili DR, Ramesh KT. The mechanical response of a 6061-T6 A1/
A12 O3 metal matrix composite at high rates of deformation. Acta Metal Mater
1995;43:445364.
[24] Tirtom I, Mustafa G, Hasan Y. Simulation of the strain rate sensitive ow
behavior of SiC-particulate reinforced aluminum metal matrix composites.
Comput Mater Sci 2008;42:5708.
[25] Bauri R, Surappa MK. Processing and properties of AlLiSiCp composites. Sci.
Technol. Adv. Mater. 2007;8:494502.
[26] Hull D, Clyne TW. An introduction to composite materials. 2nd
ed. Cambridge: Cambridge University Press; 1996.
[27] Cockcroft MG, Latham DJ. NEL report no. 240, National Engineering Laboratory,
Glasgow, UK; 1966.
[28] Mummery P, Derby B. The inuence of microstructure on the fracture
behaviour of particulate metal matrix composites. Mater Sci Eng A
1991;135:2214.
[29] Skolianos S. Mechanical behavior of cast SiCp-reinforced A14.5%Cu1.5%Mg
alloy. Mater Sci Eng A 1996;210:7682.
[30] Manoharan M, Gupta M. Effect of silicon carbide volume fraction on the work
hardening behavior of thermo mechanically processed aluminium-based
metalmatrix composites. Composites Part B 1999;30:10712.
[31] Kamat SV, Manoharan M. Work hardening behaviour of alumina particulate
reinforced 2024 aluminium alloy matrix composites. J Compos Mater
1993;27:171421.
[32] Arsenault RJ, Shi N. Dislocation generation due to differences between the
coefcients of thermal expansion. Mater Sci Eng 1986;81:17587.
[33] Vogelsang M, Arsenault RJ, Fisher RM. An in situ HVEM study of dislocation
generation at Al/SiC interfaces in metal matrix composites. Metall Trans A
1986;17:37989.
[34] Xu C, Furukawa M, Horita Z, Langdon TG. The evolution of homogeneity and
grain renement during equal-channel angular pressing: a model for grain
renement in ECAP. Mater Sci Eng A 2005;398:6676.
[35] Inoue T, Horita Z, Somekawa H, Ogawa K. Effect of initial grain sizes on
hardness variation and strain distribution of pure aluminum severely
deformed by compression tests. Acta Mater 2008;56:6291303.
[36] Wanga JW, Duana QQ, Huanga CX, Wua SD, Zhang ZF. Tensile and compressive
deformation behaviors of commercially pure Al processed by equal-channel
angular pressing with different dies. Mat Sci Eng A 2008;496:40916.
[37] Chinh NQ, Szommer P, Csanadi T, Langdon TG. Flow processes at low
temperatures in ultrane-grained aluminum. Mater Sci Eng A
2006;434:32634.
[38] Sivaraman A, Chakkingal U. Flow properties of commercial purity aluminum
processed by equal channel angular pressing. Mater Sci Eng A
2008;487:26470.