Journal of Minerals & Materials Characterization & Engineering, Vol. 11, No.5, pp.493-507, 2012 Printed in the USA. All rights reserved
Influence of Microst r uc tu re and Experimental Parameters on M ec hanical and
Wear Prop erties of Al-TiC Su rfac e C omposite by FSP Route
S. Jerome*, S. Govind Bhalchandra, S.P. Kumaresh Babu, B. Ravisankar
Department of Metallurgical and Materials Engineering, National Institute of Technology,
Tiruchirappalli – 620 015, Tamilnadu, India
*Corresponding Author:,
Surface properties decide the fatigue, wear and corrosion behavior of a material. Hence their
performance can be improved by surface modifications. Friction Stir Processing (FSP) is a
promising technique to develop surface composite. The aim of the present study is to develop
defect free surface composite of Al 5083 alloy reinforced with TiC particles and investigate the
particle distribution in the matrix, mechanical properties and wear beha vior of the composites.
Microstructural observations were carried out by using optical and scanning electron
microscopy (SEM). The microstructural studies revealed that distribution of particles were more
uniform in samples subjected to double pass than the single pass FSP. The microhardness
profiles along top surface and across the cross section of the processed samples were evaluated.
The average hardness along the top surface was found to increase by 27.27%, as compared to
that of the base metal (88Hv). The particles were incorporated maximum average depth about
250µm in the surface composite. The slurry erosion tests revealed that the wear rate was highly
reduced in case of double pass FSP samples as compared to base metal and single pass FSPed
Keywords: Al- TiC surface composite, Friction stir processing, Microhardness, Wear resistance.
Properties like high strength, elastic modulus, and improved resi s tan ce to wear, fat igue and cre ep
of metal-matrix composites reinforced with hard ceramic phases, make them promising structural
materials for aerospace and automobile industries compared to unreinforced metals. However,
494 S. Jerome, S. Govind Bhalchandra Vol.11, No.5
due to the incorporation of non-deformable ceramic phases, these metal-matrix composites also
suffer a great loss in ductility and toughness, which limits their wide range of applications to a
certain extent. For wear resistance applications, life of the components can be improved by
surface modification with ceramic phases. This concept results in the development of surface
composite, where the surface layer of the alloys alone is reinforced with ceramic phases while
the bulk of components retain their original composition and structure with higher toughness.
The various surface modification techniques to fabricate surface metal-matrix composites
include high-energy laser melt treatment, high-energy electron beam irradiation, plasma
spraying, powder metallurgy and casting. But in all these existing techniques, it is hard to avoid
interfacial reactions and the formation of detrimental phases. Moreover, as these techniques
involves liquid phase processing at elevated temperatures, critical control over the processing
parameters is essential to get ideal solidified microstructure in the surface layer, which makes
their use further difficult. To avoid the above mentioned problems, the process should carry out
at a temperature below the melting point of the substrate as did in solid phase surface
modification techniques. There are several solid phase surface modification techniques are
available which include diffusion bonding, co- extrusion, Friction Stir Processing etc. In this
study, Friction Stir Processing (FSP) developed by R.S Mishra [1], was attempted to
incorporate ceramic particles into the surface of aluminum alloy to fabricate surface composite.
R.S. Mishra et. al. [1] had developed the first Al- SiC surface composite synthesized by FSP
route and obtained a composite layer ranged from 50 to 200 mm. The SiC particles are
distributed on the aluminum matrix and particles have excellent bonding with the aluminum
alloy substrate. The microhardness of the surface composite was increased almost two times
greater than the substrate. Morisada et. al. [2-4] had developed magnesium based surface
composites by incorporating different particles such as SiC, fullerene and multi walled carbon
nano tubes and reported that the properties of surface composites were better than monolithic
alloys. Wei Wang [5] had extended the same concept to fabricate bulk composite. Al-
Al2O3 nano surface composite was developed by Shafiei - Zarghani [6] and A.P. Gerlich et. al.
[7] had developed multi walled CNT reinforced composite by the same route. Based on the
limited available literature, the research work had been mainly focused on the development of
the composite and few studies were focused to understand the effects of process parameters on
the mechanical properties of the composites. In th e present work, microstructural studies for the
top and cross- sections had been carried out to analyze the particle distribution in a surface
composite developed by FSP. Mechanical properties mainly on hardness and wear studies in
corrosive environment also evaluated.
Vol.11, No.5 Influence of Microstructure and Experimental Parameters 495
Friction Stir Processing was used for the development of surface composite of aluminium 5083
alloy reinforced with TiC particles. A cold rolled and stabilized plate of aluminium 5083 alloy
was used as the substrate. The chemical composition of the alloy is as shown in Table 1.
Table.1. Chemical Composition of the Aluminum 5083 Alloy
Wt (%)
Samples with the dimensions of 150mm x 100mm x 6mm thick plates were used for processin g.
The average particle size of TiC particle used was 4 to 10 µm. The samples were subjected to
single and double pass of FSP with and without TiC particles under different conditions. A
cylindrical FSP tool made of EN 31 steel was used for FSP (hardened to 64 HRc) and its
dimensions are listed in the Table 2. The processing parameters used in this study are listed in
the Table 3.
Table.2. Tool details
EN 31, quenched & tempered condition
Ø 18 mm, Concave shoulder, 3o included angle
Ø 6 mm, pin length: 2 mm, left handed thread
Table 3. Process parameters used for the FSP with TiC powder
Traverse Speed
(m m/mi n )
Visual Observations
1 1400 16 No defect, smooth surface
No defect, smooth surface
3 900 16 No defect, smooth surface
4 710 16 No defect, smooth surface
Particle distributions were observed at the cross, traverse and top section of the processed
sampl e. Hitachi s canning electron microscop y was used for pa rticle distribu tion observation. The
Zwick Vickers hardness tester was used for the hardness survey of the processed sample using
300gm of load with15 seconds of dwell time. Phase analysis was carried out by XRD (Rigaku
Corporation, Japan, D/MAX ULTIMA II, 3kW, Cu Kα). Wear resistance of the base plate and
the composites were studied by performing pot erosion wear tests. These tests were performed
for base metal, single pass and double pass FSP with TiC particles. The siz e of the sam ples used
was 20mm x20 mm and the sand particle size was 80µm. The slurry concentration of 25% and
496 S. Jerome, S. Govind Bhalchandra Vol.11, No.5
50% was prepared and erosion rates were performed at 500 and 900 r.p.m. The weight loss for
each specimen measured after 1 hr of the test. The total test duration for each sample was
conducted for 4 hrs. The marine environment was simulated by using 3.5 % brine solution during
the wear test.
3.1 Microstructural Observations
FSP was carried out using the parameters shown in the Table 3. Visibly defect free composites
were fabricated in all passes. Meager flash and a smooth surface over the FSPed region were
obtained. Metallographs of the cross sections of the samples taken perpendicular to the direction
of tool travel revealed that the particle distribution improved with the higher tool rotational
From the figure 1a it was observed that for the tool rotation of 710 r.p.m. TiC particles could not
be mixed well with the Al matrix and were found either in the form of small lumps below a thin
layer of Al or in the form of clusters. This can be attributed to the low heat input associated with
lower rotational speed that induces insufficient material flow. With the rotational speed increased
to 900 r.p.m. and 1200 r.p.m. the thin layer on the top of the TiC particles was sheared by the
trailing shoulder over the undisturbed TiC particles present in the grooves (Figure 1b and 1c
respectively). Compaction was ensured by the plunge force which was accentuated by the
trailing edge due to rake angle effect. It can be observed in the figure 1d that as the rotational
speed was increased to 1400 the particle distribution was improved. The metallographs revealed
that the TiC particle tend to originate from the groove (that acts as a source) and then spreads
towards the surface where they are mixed with Al by the shearing action of the shoulder. The
layer of Al above the TiC particles reduced considerably with increasing rotational speed and
was replaced by the surface composites when the rotational speed of 1400 r.p.m. was used.
The etched specimen shown in Figure 2 helped to understand how the shoulder-driven flow has
resulted in the formation of surface composites along with grain refinement due to dynamic
recrystalisation of the plasticized material. On getting the surface composites on the base plate an
attempt was made to homogenize the TiC particles in the surface composite by performing
multipass FSP. A second FSP pass was taken over the first pass in two different methods i) the
tool traverse was in the same direction as those of the first pass and ii) in a direction opposite to
the first pass. 100% overlap was made over the first pass in both the cases as shown in
Figure 3.
Vol.11, No.5 Influence of Microstructure and Experimental Parameters 497
Figure 1. Micros tructu res of unetched specimens showing particle distribution after
single pass FSP along the cross section for rotational speed of a) 710, b)
900, c)1120 and d)1400 r.p.m.
Figure 2 Microstructure of etched specimen of Al-TiC composites
The pa ramete rs fo r the s econd pas s were t he s am e as thos e of the firs t pas s. Al l th e trials yielded
defect free composites with nominal flash. Microstructural characterization revealed that the
particle distribution was improved invariably in both the cases of double pass. However, double
pass FSP made in the direction opposite to the first pass showed more uniform distribut ion. SEM
498 S. Jerome, S. Govind Bhalchandra Vol.11, No.5
images shown in figure 3 revealed that in the case of double pass in the same direction, particles
were not uniformly distributed throughout the region of the surface composite layer. In the latter
case, uniform layer of surface composite was s een over major portion of t he FSPed region. This
can be attributed to the fact that AS of the second pass overlapped on the RS of the first pass
resulting in homogenization of the FSP region and the surface composites contained in it. The
SEM images of the cross section of these specimens proved this fact.
Figure 3 SEM images of cross section of double pass FSPed specimen a & b) Opposite
direction, c &d ) same direction
3.2 Microhardness Details
The base metal microhardness was observed to be 88Hv and it is shown in figure 4. Variation in
maximum hardness values with the parameters like rotary speed for a feed rate of 16 mm/min
have been displayed in the same figure. From the figure 5 it was observed that the hardness
values of the single pass FSP samples shown the maximum hardness values were achieved at
Vol.11, No.5 Influence of Microstructure and Experimental Parameters 499
1400 r.p.m. These results are in consensus with the results obtained through microstructural
changes obtained after FSP and addition of TiC. Also, M. R aaft et al [9] revealed that, in cre asi n g
the tool rotational speed increases the hardness of the composite layers. It is because of the fact
that Al 5083 is strain hardenable alloy. Thus, owing to increased flowability at higher heat input
imparted by higher rotational speeds, strain hardening effect contributes to the increased
hardness. It is clear that, in FSP, as the rotational speed increases, the heat input also increases.
However, it has been reported that the calculated max imum temperatures are nearl y the same in
all the rotational speeds. This phenomenon can be ex plained by the following two r easons: first,
the coefficient of friction decreases when a local melt occurs; secondly, the latent heat absorbs
some heat input.
Figure. 4 Variation in Average hardness values with the rotary speed and given feed rate of
the tool during FSP using concave shouldered tool
In these c ases, th e hardn ess was meas ured and plot ted at equal interv als from the advanci ng side
(AS). From the figure 5 it was observed that, the hardness would be more at the advancing side
as compared to the retreating side (RS). Also the hardness within the stir zone would be more or
less uniform. As many of the researchers reported, grain refinement is due to dynamic re-
crystallization and mechanical stirring. Also it was reported that the grain size is even finer at the
AS than that at the SZ and RS. Hence, for samples without TiC particles, the Hall-pitch
hardening takes and an increase in hardness takes place in the processed zone. Also it was
expected that the hardness in the AS would be more, as the material is tran sferred from the RS t o
the AS, and the same is forged against the advancing side base material. Although the increase
in hardness is mainly due to the grain refinement (Hall-Petch Strengthening) in case of Al 5083,
dispersion strengthening is reported to contribute to a larger extent. FSPed samples without TiC
powder, whereas the presence of TiC powder contributes much to the hardness in case of the
500 S. Jerome, S. Govind Bhalchandra Vol.11, No.5
surface composite. The grain size would be even finer as TiC particles prevent the grains to grow
by providing a large number of nucleation sites.
Figure. 5 Variation in the hardness values with the rotatioanl speed and effect of FSP
passes at 1400 rpm
Rotational speed appears to be the most significant process variable since it also tends to
influence the translational velocity. Higher tool rotational speed resulted in a higher temperature
and slower cooling rate in the FSP zone. Lower heat input condition due to lower rotational
speed resulted in lack of stirring resulting in less improvement in hardness. It was observed that
the hardness values were invariably higher for surface composites than those of the base metal.
This increase can be attributed to the presence of hard TiC particles in the Al matrix along with
the grain refinement effect of FSP.
In the present study, in case of surface composite developed by single pass FSP with groove
design, the average hardness along the top surface was found to increase by 22.72% as compared
to that of th e base m etal where as the i n cas e of s urf ace composite develop ed by double pass FSP
in same and opposite direction, the average hardness along the top surface was found to increase
by 25% and 27.27% respectively, as compared to that of the base metal.
It is already known that the presence of external particles reduces the flowability of the base
metal in the FSP zone. This is the major hurdle in formation of d efect free surface composites. It
is very well known that it is the frictional heat produced by the shoulder that results in
plasticising the material below shoulder and around the pin. In the presence of extended pin the
effect of stirring action of the shoulder at the pin end fades as the length of the pin increases.
Vol.11, No.5 Influence of Microstructure and Experimental Parameters 501
This is due to the fact that the pin rotation is insufficient to drive the material from the RS to the
AS of the tool pin. It finall y results in what is called as channel/worm/groove defect. This defect
increases as pin length increases for given set of parameters. Thus it can be inferred that the
shoulder driven flow influences the pin driven flow when the pin is short and hence results in
defect free passes. Tool geometry also plays a vital role in defect formation. In the present case
the concave shoulder helps to contain the plastic material within it and thus prevents to some
extent the loss of material from the stir zone due to flash formation. So also as the length of the
pin increases the amount of metal extruded from the plate also increases. In a case when the
shoulder is not able to contain this material excessive flash formation c an be seen . In the pres ent
study, the tool design successfully averted problems associated with reduced flowability of TiC
particles, flash formation and defect formation and the use of rake angle helped inclusion of
particles in the stir zone to form surface composites.
3.3 XRD Analysis
Figure 6. XRD of a) base, b) single pass FSP and c) double pass FSP
Figures 6 a-c showed the XRD pattern of the base metal and the surface composites. The
presence of TiC peaks confirmed that the particles were successfully incorporated in the Al
matrix but there were no detrimental phases or other intermetallics revealed by the XRD.
Because, during FSP, the plate temperature could be reached above the recrstallisation
temperature of the matrix and not exceeding the melting point of both the matrix and the
reinforcement particles [10] and the retention time at this particular temperature also less hence,
no interfacial reaction could be happened between the particles with the matrix. From these
502 S. Jerome, S. Govind Bhalchandra Vol.11, No.5
results, it is confirmed that the surface composite developed b y the FSP m ethod is proved to be a
clean and better than the other processing routes but the uniform distribution of the particles
could be obtained only by increasing the number of FSP passes.
3.4 Wear Studies
The FSPed specimens were subjected to Pot Erosion Test shown in Figure 7 with four different
set of conditions as tabulated in Table 4. Sl u rr y w as prep ared at neutral (pH~7) brine solution of
3.5%. The percentage weight loss measured after every one hour has been plotted for every
specimen for the experiments mentioned above.
Table 4. Experimental conditions for Pot Erosion Test
Experiment no.
R.P.M .
Slurry Concentration (%)
Figure 7. Pot erosion Test setup
From the results shown in figure 8 it can be concluded that the weight loss due to erosive wear
was more when the r.p.m. and the slurry concentration were higher. Interestingly for each
experimental condition the wear resistance is increased from base metal to single pass FSP to
doubl e pass FSPed sam ple. Thes e results con firm the incr ease in hardnes s in surface composit es
improves its wear resistance as compared to base metal. Experimental data and corresponding
graphs have been shown in Figure 8.
It can be seen that initial wear rate of SP is lesser and it increases with time. This is due to the
fact that the initial wear resistance is provided by the surface composites. With time as this layers
Vol.11, No.5 Influence of Microstructure and Experimental Parameters 503
wears out the wear resistance is still provided by the grain refined FSPed material. The double
pass FSPed composite sample shows persistent resistance to wear.
Figure 8. Wear rate of base metal and Al-TiC composites by single and double pass FSP
3.4.1 Effect of speed
Erosive/abrasive attack is a function of the energy of the impinging particle and the severity of
attack increases with energy. The energy of the impinging particle is directly proportional to the
speed in line with the kinetic energy of the impinging particles (1/ 2mv2). Thus, the wear rates
increased as the speed was increased from 500 to 900 rpm irrespective of t he material. However,
at higher speeds, say 900 rpm, the mobility of sand particles is decreased and hence they do not
have enough time to make an effective impact and they simply slide off instead of making any
sensible impact. Further at such high speeds the chances of intercollisions amongst the erodent
particles are more. Additionally the chances of rebounding of erodent particles are also high at
such speeds. All these factors cumulatively reduced the wear rates at 900 rpm.
3.4.2 Effect of concentration
Increasing the sand content simply increases the severity of erosive/abrasive attack because a
greater number of particles are impinging on the surface. On the other hand, the severity of
504 S. Jerome, S. Govind Bhalchandra Vol.11, No.5
corrosive attack may decrease because the effective volume of the corrodant decreases. It is
observed that the pH of the solution normally approached towards neutral value, when the sand
content was varied indicating decrease in the basicity of the corrodant. In the present study the
wear rates increased on increasing the sand content in marine solutions irrespective of the
material in spite of decrease in the corrosivity of the medium. Further, at normal impact, material
Losses due to abrasion are minimal and thus erosion is the most dominant mode of material
removal. So, by increasing sand concentration the wear rat e is increased irrespective of material
and it is shown in Figure 9.
Figure 9. Effects of experimental conditions on wear rate
3.4.3 Effect of reinforcement
From the Figures 8 and 9 it is observed that the composites have shown higher wear resistance
than the base material because, these reinforced TiC particles protect the matrix by providing a
kind of shadowing effect. Further, the presence of TiC particles reduced the effective metallic
area exposed to the corrosion solution and thus reduced the extent of corrosive attack on the
matrix. These factors cumulatively improved the wear resistance of composite. Along with TiC
particles, after every FSP passes the refined grain structure were obtained and hardness also
increased due to strain hardening effect which is also improves the wear resistance of the
composites. Further co r r os iv e att ack i s r educed becau se t he ef fe ct iv e met al li c area ex p os ed t o t he
corrosive media is reduced to a greater extent [ 1 1] . Thus, the Al-TiC surface composite exhibited
greatest wear resistance than the base material t ested.
Figure 9 compares the wear rate of the materials considered during the four different experiments
and shows the comparative wear rates during these experiments. From the figure the
experimental conditions can be arranged in the ascending order of wear rate as follows:
Expt 1 < Expt 3 < Expt 2 << Expt 4
Vol.11, No.5 Influence of Microstructure and Experimental Parameters 505
Figure 10. Wear specimens of double pass FSPed samples for a) Expt. 1, b) Expt. 2,
c) Expt. 3 and d) Expt. 4
Figure 10 shows the effect of speed and sand concentration of the wear of the Al-TiC surface
composites fabricated by double pass FSP. In general, the wear rate was quite low for all the
cases owing to small particle size. Aluminum is highly ductile material and in the present case
wear is due to both particle impacts at high velocity as well as abrasion. In particular, high
velocity impact resulted in deformation wear when higher r.p.m. (900) was used for both the
values of slurry concentration. Thus r.p.m had more influence than slurry concentration.
Noticeably, high values of wear were seen in Expt 4 due to the higher collision efficiency
resulting from higher slurry concentration.
From the present study the following conclusions were drawn:
i. In the present study particle distribution during development of surface composites by
FSP of Al 5083 was studied. Defect free composites were achieved in all FSP passes.
Based on the microstructural observations it was found that the distribution of particles
506 S. Jerome, S. Govind Bhalchandra Vol.11, No.5
was noticeable in AS and SZ. Double pass in opposite direction resulted in more uniform
distribution of particles.
ii. Surface composite developed by single pass FSP with groove design, the average
hardness along the top surface was found to increase by 22.72% as compared to that of
the base m et al where as the in c ase of surface co m p os it e dev elo ped by doubl e pas s FS P in
same and opposite direction, the average hardness along the top surface was found to
increase by 25% and 27.27% respectively, as compared to that of the base metal. The
maximum average depth of surface composite was found to be 250µm in hole-design.
iii. The weight loss due to erosive wear was more when the r.p.m. and the slurry
concentration were higher. For each experimental condition the wear resistance is
increased from base metal to single pass FSP to double pass FSPed s ample. These results
confirm the inc rease in hardness in surface composites improves its wear resistance than
the base metal.
The authors wish to acknowledge Naval Research Board, Government of India
(DNRD/05/4003/NRB/172) for supporting this research work.
[1] R.S. Mishra , Z.Y. Ma, I. Charit, "Friction stir processing: a novel t echnique for fabrication
of surface Composite", J. Materials Science and Engineering A, 341, pp 307-310, (2003)
[2] Y. Morisada , H. Fujii , T. Nagaoka , M. Fukusumi, "Effect of friction stir processing with
SiC particles on microstructure and hardness of AZ31", J. Materials Science and
Engineering A, 433, pp 50–54, (2006)
[3] Y. Morisada, H. Fujii, T. Nagaoka , K. Nogi, M. Fukusumi, "Fullerene/A5083 composites
fabricated by material flow during friction stir processing", J. Composites: Part A, 38, pp
[4] Yoshiaki Morisada, Hidetoshi Fujii, Tadashi Mizuno, Genryu Abe, Toru Nagaoka, Masao
Fukusumi, "Modification of thermally sprayed cemented carbide layer by friction stir
processing", J. Surface & Coatings Technology, 204, pp 2459–2464, (2010).
[5] Wei Wang, Qing-yu Shi, Peng Liu, Hong-ke Li, Ting Li, "A novel way to produce bulk
SiCp reinforced aluminum metal matrix composites by friction stir processing", J.
Materials processing technology, 209, pp 2099–2103, (2009).
[6] A. Shafiei-Zarghani, S.F. Kashani-Bozorg, A. Zarei-Hanzaki, "Microstructures and
mechanical properties of Al/Al2O3 surface nano-composite layer produced by friction stir
processing", J. Materials Science and Engineering A 500, pp 84–91, (2009).
Vol.11, No.5 Influence of Microstructure and Experimental Parameters 507
[7] D.K. Lima, T. Shibayanagib, A.P. Gerlich, "Synthesis of multi-walled CNT reinforced
aluminium alloy composite via friction stir processing", J. Materials Science and
Engineering A, 507, pp 194–199, (2009).
[8] M. Raaft, T.S. Mahmoud, H.M. Zakaria, T.A. Khalifa, "Microstructural, Mechanical and
wear behaviour of A390/graphite and A390/Al2O3 surface composites fabricated using
FSP", J. Materials Science and Engineering A,(2009.)
[9] H.N.B. Schmidt , T.L. Dickerson, J.H. Hattel, "Material flow in butt friction stir welds in
AA2024-T3", J. Acta Materialia, 54, pp1199–1209, (2006).
[10] Y.L. Saraswathi, S. Das, D.P. Mondal, ''Influence of microstructure and experimental
parameters on the erosion–corrosion behavior of Al alloy composites'', J.Materials Science
and Engineering A 425, pp 244–254, (2006).