Hindawi Publishing Corporation

Journal of Composites
Volume 2013, Article ID 274527, 9 pages
http://dx.doi.org/10.1155/2013/274527

Research Article
Tribological Performance Optimization of Al-7.5%
SiCp Composites Using the Taguchi Method and Grey
Relational Analysis

Shouvik Ghosh, Prasanta Sahoo, and Goutam Sutradhar

Department of Mechanical Engineering, Jadavpur University, Kolkata 700032, India

Correspondence should be addressed to Prasanta Sahoo; [email protected]

Received 11 May 2013; Accepted 7 August 2013

Academic Editor: Masamichi Kawai

Copyright © 2013 Shouvik Ghosh et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The present study considers an experimental study of tribological performance of Al-7.5% SiCp metal matrix composite and
optimization of tribological testing parameters based on the Taguchi method coupled with grey relational analysis. A grey relational
grade obtained from grey relational analysis is used as a performance index to study the behaviour of Al-7.5% SiCp MMC with
respect to friction and wear characteristics. The tribological experiments are carried out by utilizing the combinations of tribological
test parameters based on the L27 Taguchi orthogonal design with three test parameters, namely, load, speed, and time. The material
Al-7.5% SiCp metal matrix composite is developed by reinforcing LM6 aluminium alloy with 7.5% (by weight) SiC particle of 400
mesh size (∼37 𝜇m) in an electric melting furnace. It is observed that sliding time has a significant contribution in controlling
the friction and wear behaviour of Al-7.5% SiCp MMC. Furthermore, all the interactions between the parameters have significant
influence on tribological performance. A confirmation test is also carried out to verify the accuracy of the results obtained through
the optimization problem. In addition, a scanning electron microscopy (SEM) test is performed on the wear tracks to study the
wear mechanism.

1. Introduction Ma et al. [1] conducted wear test with block-on-ring con-

figuration for Al-20% SiCp (particulate) and Al-50% SiCp.
In recent years, metal matrix composites have gained pop- From the tests, they reported that wear increases with the
ularity for use in industrial applications. The metal matrix increase in load and sliding speed, while wear decreases with
composites have become increasingly desirable to engineers, the increase in SiC incorporation. It was also observed that at
largely due to improved toughness, strength, and stiffness
loads of above 100 N and at sliding speed of 6 m/s severe wear
properties offered by these materials relative to the unrein-
took place. Yalcin and Akbulut [2] observed that wear rate
forced base metals. Mostly light metals such as aluminium,
varies inversely with reinforcement volume fraction. Sharma
magnesium, and their alloys have received the commercial
recognition due to their relatively low cost and easy process- et al. [3] conducted the dry wear test for SiC reinforcement 0–
ing. Arguably, aluminium based composites have shown per- 5 volume fraction range and observed that wear rates decrease
formance improvements in a variety of applications requiring with increase in volume fraction of the reinforcement. Ghosh
high strength, low mass, and tailored thermal properties. In et al. [4] studied the friction behaviour of Al-5% SiC MMC
the past years, many researchers have studied the friction and and concluded that applied load and sliding speed are the
wear behaviour of Al based composites. Mostly, researchers most significant parameters affecting the friction behaviour
have studied the friction and wear behaviour for sliding of the material. Another similar study was conducted by
contact [1–6]. Some of the researchers have studied the Ghosh et al. [5] on the friction behaviour of Al-10% SiC
abrasive wear behaviour for the same [7–12]. The effect of heat MMC. Onat [6] studied the dry sliding wear properties of
treatment [13, 14] and temperature change [15] is also studied SiCp reinforced Al-Cu composite and concluded that with
by some researchers. increase in applied load and sliding speed, the wear rates
2 Journal of Composites

increased. Ahlatci et al. [7] carried out abrasive test of Al- performance, and computational cost. This method achieves
8% SiC MMC by abrasive wear tester. Sahin [8] studied the the integration of design of experiments (DOE) [20] with the
abrasive wear behaviour of Al-SiC composite and optimized parametric optimization of the process yielding the desired
the testing parameters using Taguchi method. In another results. The traditional experimental design procedures focus
study, Şahin [9] performed abrasive wear test of Al-15% SiC at on the average process performance characteristics, but the
different loads, sliding distances, and reinforcement particles Taguchi method concentrates on the effect of variation on
size. The researcher observed that reinforcement particle size the process quality characteristics rather than on its averages.
has the greatest effect on wear. Al-Rubaie et al. [10] performed That is, the Taguchi approach makes the process performance
wear test on two body abrasion wear models. The tests were insensitive (robust) to variation in uncontrolled or noise fac-
conducted with SiC reinforcement having mean size of 10, 27, tors. Taguchi recommends that this can be done by the proper
and 43 𝜇m. The reinforcement is mixed with Al in the propor- design of parameters during the “parameter design” phase of
tion of 5, 10, and 20 vol.%. From the results, it was observed off-line quality control. He designed certain standard OAs
that wear resistance of the MMC increases with increase in by which simultaneous and independent valuation of two or
particle size of the SiC reinforcement. In another similar more parameters for their ability to affect the variability of a
study, Al-Rubaie et al. [11] studied three body abrasion of Al- particular product or process characteristic can be done in a
SiC using a thin oil film as lubricant between the sample and minimum number of tests. Using OA, the Taguchi method
the rotating disk. Howell and Ball [12] performed dry wear explores the entire design space through a small number of
tests using Al-20% SiC against brake pads. From the results, experiments in order to determine all of the parameter effects
it was concluded that inclusion of reinforcement enhanced and several of the interactions. These data are then used to
the wear resistance of the material alloy. The effect of matrix predict the optimum combination of the design parameters
alloy on the wear behaviour of the composite was studied by
that will minimize the objective function and satisfy all
Rao and Das [13]. The experimental study was conducted
the constraints. In addition to locating a near optimum
using AA7010, AA7009, and AA2024 aluminium alloys, and
objective function, the Taguchi method provides information
reinforcement volume fraction was varied from 10 to 25%.
The test showed that the increase in volume fraction of on parameter trends and noise sensitivities thereby enabling
reinforcement decreased the wear rate. In another study by a robust design. The parameter design phase of the Taguchi
Rao and Das [14], it was concluded that heat treatment of the method generally includes the following steps: (1) identify the
composites improved the friction and wear property of the objective of the experiment; (2) identify the quality character-
composite. Martı́n et al. [15] studied wear behaviour in the istic (performance measure) and its measurement systems;
temperature range of 20–200∘ C and noted that heat treatment (3) identify the factors that may influence the quality char-
did not modify substantially the wear resistance of the com- acteristic, their levels, and possible interactions; (4) select
posites. Another study on the influence of temperature on the appropriate OA and assign the factors at their levels to
the wear behaviour of Al base composites was conducted by the OA; (5) conduct the test described by the trials in the
Straffelini et al. [16]. The researcher concluded that external OA; (6) analysis of the experimental data using the signal-to-
heating decreased both friction and wear for the composite noise (𝑆/𝑁) ratio, factor effects, and the analysis of variance
material. (ANOVA) to see which factors are statistically significant
For the present tribological study, aluminium alloy LM6 and to find the optimum levels of factors; (7) verification of
is used as base metal, and silicon carbide (SiCp) is used the optimal design parameters through confirmation exper-
as reinforcement. The MMC is synthesized by stir casting iment. The OA requires a set of well-balanced (minimum
process in an electric melting furnace where silicon carbide experimental runs) experiments. The Taguchi method uses a
(7.5% by wt.) is introduced into the aluminium alloy. The statistical measure of performance called (𝑆/𝑁) ratios, which
tribological tests are conducted on the material to study the are logarithmic functions of desired output to serve as
friction and wear properties of the material. The result data objective functions for optimization. The 𝑆/𝑁 ratio takes both
are generalized to grey relational grade and analyzed by the the mean and the variability into account and is defined as the
Taguchi Method. A statistical analysis of variance (ANOVA)
ratio of the mean (signal) to the standard deviation (noise).
is performed. Finally, confirmation test is carried out to
The ratio depends on the quality characteristics of the
verify the optimal process parameters combination. The
product/process to be optimized. The three categories of 𝑆/𝑁
microstructure study is carried out with the help of SEM to
judge the wear mode of the material. ratios are used: lower the better (LB), higher the better (HB),
and nominal the best (NB). The parameter level combination
that maximizes the appropriate 𝑆/𝑁 ratio is the optimal
setting. For the case of minimization of wear, LB character-
2. The Taguchi Method istic needs to be used. Furthermore, a statistical analysis of
The Taguchi method [17–19] is a powerful tool for design- variance (ANOVA) [21] is performed to find which process
ing high quality systems based on orthogonal array (OA) parameters are statistically significant. With the 𝑆/𝑁 ratio and
experiments that provide much reduced variance for the ANOVA analyses, the optimal combination of the process
experiments with an optimum setting of process control parameters can be predicted. Finally, a confirmation exper-
parameters. It introduces an integrated approach that is sim- iment is conducted to verify the optimal process parameters
ple and efficient to find the best range of designs for quality, obtained from the parameter design.
Journal of Composites 3

Table 1: Chemical composition of LM6.

Elements (%) Si (10–13) Cu (0.1) Fe (0.6) Mn (0.5) Zn (0.1) Pb (0.1) Ni (0.1) Mg (0.1) Ti (0.2)

3. Grey Relational Analysis Table 2: Design factors used in the experiment.

The aim of the present paper is to optimize the tribological Levels

Design factors Unit
test parameters with the objective to minimize both friction 1 2 3
and wear of Al-7.5% SiCp. Thus, it is a case of multiresponse Load (L) N 50 75i 100
optimization, which is different from that of a single perfor- Speed (S) RPM 180 200i 220
mance characteristic. The higher 𝑆/𝑁 ratio for one perfor- Time (T) MIN 20 30i 40
mance characteristic may correspond to a lower 𝑆/𝑁 ratio for i
Initial testing condition.
another. Therefore, the overall evaluation of the 𝑆/𝑁 ratio is
required for the optimization of multiple performance char-
acteristics. Grey relational analysis [22] is an efficient tool for 4.2. Design of Experiment. Design factors or control factors
such multiresponse analysis. In grey relational analysis, the are those which are varied during the experimental tests.
first step is to perform the grey relational generation in which There are a number of control factors that can affect the
the results of the experiments are normalized in the range tribological behaviour of Al-7.5% SiCp. For this experimental
between 0 and 1. Then, the second step is to calculate the grey purpose, the control factors chosen are load, speed, and time
relational coefficient from the normalized data to represent since these are the significant ones. Table 2 shows the design
the correlation between the desired and actual experimental factors with their levels. The present study considers the fric-
data. The overall grey relational grade is then computed by tion and wear characteristics of Al-7.5% SiCp as the response
averaging the grey relational coefficient corresponding to variable. It may be noted here that the levels chosen for the
each performance characteristic. Overall evaluation of the design factors (load, speed, and time) are not very similar to
multiple performance characteristics is based on the cal- real life applications. Typically, all these parameters have been
culated grey relational grade. As a result, optimization of selected at lower levels as compared to real life applications.
the complicated multiple performance characteristics is con- This is due to the limitation of the tribotester used in the
verted into optimization of a single grey relational grade. The experimentation.
optimal level of the process parameters is the level with the Based on the Taguchi method, an orthogonal array (OA)
highest grey relational grade. is considered to reduce the number of experiments required
Furthermore, analysis of variance (ANOVA) is performed to determine the optimal friction and wear for Al-7.5% SiCp
to predict the importance and significance of each process metal matrix composite. To choose an orthogonal array, the
parameter and their interactions on the tribological proper- total number of degrees of freedom is to be chosen. For this
ties of Al-7.5% SiCp metal matrix composite. Finally, confir- experimental purpose, L27 orthogonal array is chosen. This
mation test is performed to verify the optimal combination L27 OA has 27 rows corresponding to the number of tests, and
of process parameter obtained from the analysis. the degree of freedom is 26. The degree of freedom of each
design factor is 2, and for two-way interaction of the factors,
4. Experimental Details the degree of freedom is 4. So, the total degree of freedom for
the conducted experiment is (2 × 3 + 4 × 3 = 18).
4.1. Fabrication Process. The metal matrix composite is fabri- So, the L27 OA is chosen for the experimental purpose.
cated using LM6 as matrix metal and silicon carbide particle The 1st column is assigned to load (𝐿), 2nd column is assigned
(size ∼37 𝜇m) as reinforcement. The chemical composition of to speed (𝑆), and the 3rd column is assigned to time (𝑇). The
the matrix metal (LM6) is given in Table 1. The fabricating rest of the columns are assigned to the two-way interactions
process used for this purpose is liquid stir casting technique of the factors and error terms. Table 3 shows the orthogonal
which is simple and less expensive. The matrix metal LM6 is array along with the experimental results for friction and
melted in clay graphite crucible using an electric resistance wear.
furnace. To increase the fluidity of the liquid metal by 3%
weight magnesium is added. This addition of magnesium will 4.3. Tribological Test. The tribological tests are carried out in
also increase the bonding power by decreasing the surface a multitribotester TR25 (Ducom, India) (Figure 1) under dry
energy (wetting angle) between the matrix alloy and the nonlubricated condition and at ambient temperature (28∘ C).
reinforcement particles. The silicon carbide particle is pre- It uses a block-on-roller configuration where EN32 steel
heated at 850–900∘ C for 2-3 hrs before mixing in the molten roller is used. The Al-SiCp samples (size 20 mm × 20 mm ×
metal. The melt is then mechanically stirred using a mild steel 8 mm) are pressed against a rotating steel roller (diameter
impeller and then the preheated SiC particles are added to 50 mm, thickness 50 mm) of hardness 65 HRc. The setup is
the stirred liquid metal. The mixing process is carried out placed in such a way that the rotating roller serves as the
at 750∘ C with a stirring speed of 400–500 rpm. Then, the counter face material and stationary plate serves as the test
molten mixture is poured into green silica sand mould. And specimen. Figure 2 shows the close-up view of specimen in
after cooling, the samples required for tribological testing are multitribotester. A 1 : 5 ratio loading lever is used to apply
prepared by different machining processes. normal load on top specimen. The loading lever is pivoted
4 Journal of Composites

Table 3: L27 orthogonal array with design factors and interactions assigned.

Trial no. 1 (L) 2 (S) 3 (L × S) 4 (L × S) 5 (T) 6 (L × T) 7 (L × T) 8 (S × T) 9- 10- 11 (S × T) 12- 13- COF Wear (𝜇m)
1 1 1 1 1 1 1 1 1 1 1 1 1 1 0.341 60.07
2 1 1 1 1 2 2 2 2 2 2 2 2 2 0.321 71.84
3 1 1 1 1 3 3 3 3 3 3 3 3 3 0.387 85.24
4 1 2 2 2 1 1 1 2 2 2 3 3 3 0.323 79.44
5 1 2 2 2 2 2 2 3 3 3 1 1 1 0.373 94.89
6 1 2 2 2 3 3 3 1 1 1 2 2 2 0.303 107.9
7 1 3 3 3 1 1 1 3 3 3 2 2 2 0.316 50.14
8 1 3 3 3 2 2 2 1 1 1 3 3 3 0.348 69.63
9 1 3 3 3 3 3 3 2 2 2 1 1 1 0.312 86.03
10 2 1 2 3 1 2 3 1 2 3 1 2 3 0.396 50.63
11 2 1 2 3 2 3 1 2 3 1 2 3 1 0.469 61.39
12 2 1 2 3 3 1 2 3 1 2 3 1 2 0.462 77.16
13 2 2 3 1 1 2 3 2 3 1 3 1 2 0.397 64.45
14 2 2 3 1 2 3 1 3 1 2 1 2 3 0.439 82.24
15 2 2 3 1 3 1 2 1 2 3 2 3 1 0.456 96.24
16 2 3 1 2 1 2 3 3 1 2 2 3 1 0.299 71.07
17 2 3 1 2 2 3 1 1 2 3 3 1 2 0.308 89.68
18 2 3 1 2 3 1 2 2 3 1 1 2 3 0.336 106.2
19 3 1 3 2 1 3 2 1 3 2 1 3 2 0.383 57.24
20 3 1 3 2 2 1 3 2 1 3 2 1 3 0.409 67.93
21 3 1 3 2 3 2 1 3 2 1 3 2 1 0.419 83.21
22 3 2 1 3 1 3 2 2 1 3 3 2 1 0.387 59.24
23 3 2 1 3 2 1 3 3 2 1 1 3 2 0.452 67.93
24 3 2 1 3 3 2 1 1 3 2 2 1 3 0.409 83.21
25 3 3 2 1 1 3 2 3 2 1 2 1 3 0.363 65.58
26 3 3 2 1 2 1 3 1 3 2 3 2 1 0.395 89.09
27 3 3 2 1 3 2 1 2 1 3 1 3 2 0.334 113.77

near the normal load sensor and carries a counter weight at performed at different magnifications ranging from 250 to
one end, while at the other end, a loading pan is suspended for 1000 at constant electron beam energy of 20 Kv. From the
placing the dead weights. The load is applied by placing dead SEM images, it is determined that whether the wear tracks
weight in a loading pan which is connected by a lever. The are adhesive or abrasive in nature.
experimental data is recorded by a computer attached with
the apparatus. The wear is measured in terms of wear depth 5. Result and Discussion
(microns) and friction in terms of coefficient of friction. It
may be noted here that wear behaviour is normally expressed The objective of the present study is to minimize friction
as wear volume or weight loss, while in the present experi- and wear for Al-7.5% SiCp metal matrix composite using the
mental set up, wear is measured in terms of displacement. Taguchi method. The tribological testing is carried out using
Thus, to access the accuracy of wear measurement, the dis- three control parameters, namely, applied load, sliding speed,
placement results for wear are compared with weight loss and and time. Coefficients of friction and wear depth are taken
it shows almost linear relationship for the range of test param- as system responses. The experimental results for friction
eters considered in the present study. Moreover, since in the and wear tests are shown in Table 3. Grey relational analysis
present study time is taken as a design parameter to study tri- is carried out following the steps mentioned earlier. Table 4
bological performance, it is not possible to go for wear mea- shows the calculated values for each steps of grey relational
surement through weight loss as that requires off-line mea- analysis.
surement. Hence, wear has been measured in terms of wear
depth in situ. The wear tests are carried out as per L27 ortho- 5.1. Analysis of 𝑆/𝑁 Ratio for Grey Relational Grade. The
gonal array in Table 3. normal method of calculating the desirable factors levels is
to look at simple averages of the results. But the variability
4.4. Microstructure Study. After wear tests, scanning elec- of results within a trial condition cannot be judged by this
tron microscopy (JEOL, JSM 6360) is done to evaluate method. Thus, signal-to-noise ratio analysis is done consid-
the microstructure of the specimen. The SEM evaluation is ering wear as the performance index. The analysis is carried
Journal of Composites 5

Friction Wear
Load RPM PC monitor
Friction wear
measuring unit
Wear sensor
Counter weight
Normal load sensor Friction force Loading lever
Pivot sensor

Specimen holder
Plate specimen
Steel roller

Loading pan
Drive unit

Figure 1: Layout of multitribotester.

Table 4: Computation of Grey Relational Grade.

Exp. Experimental data Normalized data Value of Δ Grey relational coefficient Grey relational
no. Normalized Normalized Δ value COF Δ value WEAR 𝜉 value COF 𝜉 value WEAR grade
COF Wear (𝜇m)
COF WEAR
1 0.341 60.07 0.7529 0.8439 0.2471 0.1561 0.6693 0.7621 0.7157
2 0.321 71.84 0.8706 0.6590 0.1294 0.3410 0.7944 0.5945 0.6944
3 0.387 85.24 0.4824 0.4484 0.5176 0.5516 0.4913 0.4755 0.4834
4 0.323 79.44 0.8588 0.5395 0.1412 0.4605 0.7798 0.5206 0.6502
5 0.373 94.89 0.5647 0.2967 0.4353 0.7033 0.5346 0.4155 0.4751
6 0.303 107.9 0.9765 0.0923 0.0235 0.9077 0.9551 0.3552 0.6551
7 0.316 50.14 0.9000 1.0000 0.1000 0.0000 0.8333 1.0000 0.9167
8 0.348 69.63 0.7118 0.6937 0.2882 0.3063 0.6343 0.6201 0.6272
9 0.312 86.03 0.9235 0.4360 0.0765 0.5640 0.8673 0.4699 0.6686
10 0.396 50.63 0.4294 0.9923 0.5706 0.0077 0.4670 0.9848 0.7259
11 0.469 61.39 0.0000 0.8232 1.0000 0.1768 0.3333 0.7388 0.5361
12 0.462 77.16 0.0412 0.5754 0.9588 0.4246 0.3427 0.5407 0.4417
13 0.397 64.45 0.4235 0.7751 0.5765 0.2249 0.4645 0.6898 0.5771
14 0.439 82.24 0.1765 0.4955 0.8235 0.5045 0.3778 0.4978 0.4378
15 0.456 96.24 0.0765 0.2755 0.9235 0.7245 0.3512 0.4083 0.3798
16 0.299 71.07 1.0000 0.6711 0.0000 0.3289 1.0000 0.6032 0.8016
17 0.308 89.68 0.9471 0.3786 0.0529 0.6214 0.9043 0.4459 0.6751
18 0.336 106.2 0.7824 0.1190 0.2176 0.8810 0.6967 0.3620 0.5294
19 0.383 57.24 0.5059 0.8884 0.4941 0.1116 0.5030 0.8176 0.6603
20 0.409 67.93 0.3529 0.7204 0.6471 0.2796 0.4359 0.6414 0.5386
21 0.419 83.21 0.2941 0.4803 0.7059 0.5197 0.4146 0.4903 0.4525
22 0.387 59.24 0.4824 0.8570 0.5176 0.1430 0.4913 0.7776 0.6345
23 0.452 67.93 0.1000 0.7204 0.9000 0.2796 0.3571 0.6414 0.4993
24 0.409 83.21 0.3529 0.4803 0.6471 0.5197 0.4359 0.4903 0.4631
25 0.363 65.58 0.6235 0.7573 0.3765 0.2427 0.5705 0.6733 0.6219
26 0.395 89.09 0.4353 0.3879 0.5647 0.6121 0.4696 0.4496 0.4596
27 0.334 113.77 0.7941 0.0000 0.2059 1.0000 0.7083 0.3333 0.5208
6 Journal of Composites

Load Speed Time

0.7
0.65

Mean
0.6
0.55
0.5
1 2 3 1 2 3 1 2 3

Figure 3: Main effects plot for grey relational grade.

0.75
Figure 2: Close-up view of specimen loading in multitribotester.
0.7
Table 5: Response table for grey relational grade.
0.65
Level Load Speed Time

Mean
1 0.654 0.583 0.7004 0.6
2 0.5672 0.530 0.5492
3 0.5389 0.647 0.5105 0.55
Rank 3 2 1
Delta 0.1151 0.117 0.1899 0.5
Total mean grey relational grade = 0.5867.
0.45

1 2 3
out using lower-the-better criterion and the same is expressed
Speed
as
Load
𝑆 𝑦2 1
= −10 log (∑ ) . (1) 2
𝑁 𝑛 3
Here, 𝑦 is the experimental data and 𝑛 is the number Figure 4: Interaction plot between load and speed.
of experiments. Table 3 shows the experimental results for
tribological tests. The experimental design being orthogonal,
it is possible to separate out the effect of each control factor at
different levels. As an example, the mean 𝑆/𝑁 ratio for factor parameters in controlling the tribological behaviour of the
𝐿 (load) at levels 1, 2, and 3 can be calculated by averaging the MMC. The interaction plots are studied on the basis of non-
𝑆/𝑁 ratios for the experiments 1–9, 10–18, and 19–27, respec- parallelism of the parameter effects. If the lines of an interac-
tively. The mean 𝑆/𝑁 ratio for the other factors at different tion plots are not parallel or intersecting, then there is strong
levels can be calculated in a similar manner. In the response interactions between the parameters. And if the lines are
table (Table 5), the mean 𝑆/𝑁 ratio for each level of the con- parallel to each other, then there is nominal or no interaction
trolling factors is shown. In addition, the total mean 𝑆/𝑁 ratio between them. From the interaction plots in Figures 4, 5, and
for the 27 experiments is also calculated and listed in the same 6, it can be seen that the lines are intersecting each other.
table. All the calculations are performed using Minitab soft- Thus, there is good interaction between the parameters. From
ware [23]. The response table includes ranks based on delta the present analysis, it is observed that time (𝑇) is the most
value (the highest average of each factor minus the lowest influencing parameter for tribological characteristics of Al-
average of the same); rank 1 is assigned to the parameter SiCp composites followed by sliding speed and applied load,
with highest delta value, rank 2 to second highest delta value respectively. The optimal process parameter combination is
and so on. In this case, time has the highest delta value thus the one that yields maximum mean value, and thus, it is found
rank 1 is assigned to time (𝑇). The corresponding main effects to be L1S3T1, that is, the lowest level of applied load along
plot for mean is shown in Figure 3. The interaction plots for with the highest level of sliding speed and lowest level of
parameters applied load, sliding speed, and time are given sliding time within the experimental domain considered in
in Figures 4, 5, and 6. In main effects plot, the significance the present study.
of each parameter can be judged by the inclination of
plot. The parameter with highest inclination line has greater 5.2. Analysis of Variance for Grey Relational Grade. ANOVA
significance than the rest on the tribological behaviour of is a statistical technique which can infer some important
the material. From the main effects plot, it is seen that the conclusions based on analysis of the experimental data. This
parameter time 𝑇 is the most significant parameter, while method is rather useful for revealing the level of significance
other parameters 𝐿 (load) and 𝑆 (speed) are also significant of the influence of factor(s) or their interaction on a particular
Journal of Composites 7

Table 6: Results of ANOVA for grey relational grade.

Source DF SS MS 𝐹 Contribution %
L 2 0.064785 0.032392 7.08# 15.6
S 2 0.061299 0.03065 6.7# 14.76
T 2 0.181288 0.090644 19.8∗ 43.65
𝐿∗𝑆 4 0.035076 0.008769 1.92 8.45
𝐿∗𝑇 4 0.01168 0.00292 0.64 2.81
𝑆∗𝑇 4 0.024569 0.006142 1.34 5.91
Error 8 0.036623 0.004578 8.82
Total 26 0.415319
Significant parameters and interactions – (∗ F 0.01, 2, 8 = 8.65, # F 0.05,2,8 = 4.46).

0.8 0.8

0.75 0.75

0.7 0.7

0.65 0.65
Mean
Mean

0.6 0.6

0.55 0.55

0.5 0.5

0.45 0.45

1 2 3 1 2 3
Time Time
Load Speed
1 1
2 2
3 3

Figure 5: Interaction plot between load and time. Figure 6: Interaction plot between speed and time.

response. It separates the total variability of the response tribological behaviour at the confidence level of 99%, while
into contributions of each of the factors and the error. Using parameters 𝐿 (applied load) and 𝑆 (sliding speed) are sig-
Minitab [23], ANOVA is performed to determine which nificant at the confidence level of 95%. The interactions of
parameter and interaction significantly affect the perfor- parameters have almost no influence on friction and wear
mance characteristics. Table 6 shows the ANOVA result for property of the composite. The literature review shows that
tribological behaviour of Al-SiCp metal matrix composites. the load was more effective than others. However, to the best
ANOVA calculates the 𝐹-ratio, which is the ratio between of the authors’ knowledge, no study is available where time
the regression mean square and the mean square error. The has been considered as a parameter. Typically, researchers so
𝐹-ratio, also called the variance ratio, is the ratio of variance far have considered load (contact pressure) and speed (sliding
due to the effect of a factor and variance due to the error term. velocity) as the parameters since wear characteristics mainly
This ratio is used to measure the significance of the parame- depend on so-called “PV (pressure-velocity)” factor. In the
ters under investigation with respect to the variance of all the present study, tribological performance of the composite
terms included in the error term at the desired significance has been evaluated considering both friction and wear as
level, 𝛼. If the calculated value of the 𝐹-ratio is higher than response variables in a flat-on-roller configuration. The flat
the tabulated value of the 𝐹-ratio, then the factor is significant specimen is held against rotating roller counterface. Thus,
at a desired 𝛼 level. In general, when the 𝐹 value increases to consider the effect of sliding distance, both speed of the
the significance of the parameter also increases. The ANOVA roller and time of contact have been included as design
table shows the percentage contribution of each parameter. factors along with load. For the combined response of friction
From the ANOVA table, it is seen that parameter 𝑇, that and wear, finally, time appears as the most dominant factor
is, time, is the most significant parameter influencing the controlling the tribological behavior.
8 Journal of Composites

Table 7: Results of confirmation test of multiple performance.

Initial parameter Optimal parameter Exp.

Level L2S2T2 L1S3T1 L1S3T1
Wear (𝜇m) 82.24 50.14
COF 0.439 0.316
Grade 0.4378 0.3456 0.9167
Improvement of grey relational grade = 0.0922.

Figure 7: SEM of Al-7.5% SiCp worn surface at 500x magnification. Figure 8: SEM of Al-7.5% SiCp worn surface at 750x magnification.

5.3. Confirmation Tests. After the optimal level of testing wear. Also, presence of pits and prows can be observed in the
parameters has been found, it is necessary that verification micrographs; thus, occurrence of adhesive wear is also visible.
tests are carried out in order to evaluate the accuracy of the So, from overall microstructure study, it can be concluded
analysis and to validate the experimental results. The esti- that mostly abrasive wear has taken place with some traces
mated 𝑆/𝑁 ratio 𝜂̂, using the optimal level of the testing of adhesive wear. Typical wear surface of such materials under
parameters, can be calculated as mild wear conditions [1] consists of three different layers:
𝑜
plastic deformation layer characterized by deformed or
𝜂̂ = 𝜂𝑚 + ∑ (𝜂𝑖 − 𝜂𝑚 ) , (2) inclined grains, oxide layer composed of iron oxide, alu-
𝑖=1 minium matrix and small fractured SiC particles, and the
transfer layer containing numerous microcracks as a result of
where 𝜂𝑚 is the total mean 𝑆/𝑁 ratio, 𝜂𝑖 is the mean 𝑆/𝑁 ratio microcutting by fractured SiC particles. The present obser-
at the optimal testing parameter level, and 𝑜 is the number of vations on wear mechanism match well with other studies.
main design process parameters that significantly affect the However, there is ample scope for future study to consider
friction and wear performance of Al-7.5% SiCp MMC. The the wear mechanism under different test conditions.
confirmation tests for all the experiments are presented in the In the present investigation, the effect of process param-
following section. eters applied load, sliding speed, and time on the tribological
Table 7 shows the comparison of the estimated grey behaviour of Al-SiCp composite is studied. Apart from these,
relational grade with the actual grey relational grade using other factors like volume fraction of reinforcement, heat
the optimal parameters. The improvement of grey relational treatment, temperature variation, and particle size of rein-
grade from initial to optimal condition is 0.0922. Thus, there forcement are assumed constant during this experimental
is an improvement of 21% in grey relational grade. study. In the future, studies related to effects of these other
factors on the tribological behaviour of Al-SiCp can be
5.4. Microstructure Analysis. Microstructure study of the carried out.
wear tracks is carried out to analyze the wear mechanism that
the composites undergo during tribological testing. Figures 6. Conclusion
7 and 8 show wear tracks of Al-7.5% SiCp samples having
undergone tribological testing. From the SEM micrographs, The tribological behaviour of Al-7.5% SiCp metal matrix
it can be observed that the worn surface mainly consists of composite is studied for varying applied load, sliding speed,
longitudinal grooves and partially irregular pits. The presence and time using the Taguchi orthogonal array design and grey
of grooves indicates microcutting and microploughing effect. relational grade. The multiple responses friction and wear are
Thus, wear mechanism is found to be dominated by abrasive simplified to grey relational grade for the analysis process.
Journal of Composites 9

It is observed that parameter 𝑇, that is, time, is the most [10] K. S. Al-Rubaie, H. N. Yoshimura, and J. D. Biasoli De Mello,
significant parameter influencing the tribological behaviour “Two-body abrasive wear of Al-SiC composites,” Wear, vol. 233–
at the confidence level of 99%, while parameters 𝐿 (applied 235, pp. 444–454, 1999.
load) and 𝑆 (sliding speed) are also significant at 95% con- [11] K. S. Al-Rubaie, H. Goldenstein, and J. D. Biasoli De Mello,
fidence level. The interactions between the parameters have “Three-body abrasion of Al-SiC composites,” Wear, vol. 225–
almost no influence on friction and wear property of the com- 229, pp. 163–173, 1999.
posite. From the Taguchi analysis, the optimal combination [12] G. J. Howell and A. Ball, “Dry sliding wear of particulate-
of process parameter for minimum friction and wear is found reinforced aluminium alloys against automobile friction mate-
to be L1S3T1, that is, the highest level of sliding speed along rials,” Wear, vol. 181–183, no. 1, pp. 379–390, 1995.
with the lowest levels of applied load and time. Wear depth is [13] R. N. Rao and S. Das, “Effect of matrix alloy and influence of SiC
reduced by nearly 39% from initial to optimal process param- particle on the sliding wear characteristics of aluminium alloy
eter condition, and friction is reduced by 28%. From the composites,” Materials and Design, vol. 31, no. 3, pp. 1200–1207,
2010.
present study, it is revealed that a proper control of process
parameters can result in improved design of the Al-SiCp com- [14] R. N. Rao and S. Das, “Effect of sliding distance on the wear
and friction behavior of as cast and heat-treated Al-SiCp com-
posite for tribological applications. From the microstructure
posites,” Materials and Design, vol. 32, no. 5, pp. 3051–3058, 2011.
study of worn surfaces, it is observed that mostly abrasive
[15] A. Martı́n, M. A. Martı́nez, and J. Llorca, “Wear of SiC-rein-
wear mechanism has occurred on the wear tracks with some
forced Al-matrix composites in the temperature range 20–200∘ ,”
traces of adhesive wear mechanism.
Wear, vol. 193, no. 2, pp. 169–179, 1996.
[16] G. Straffelini, M. Pellizzari, and A. Molinari, “Influence of load
Conflict of Interests and temperature on the dry sliding behaviour of Al-based
metal-matrix-composites against friction material,” Wear, vol.
The authors of the paper do not have a direct financial relation 256, no. 7-8, pp. 754–763, 2004.
with the commercial identity mentioned in their paper that [17] G. Taguchi, Introduction to Quality Engineering, Asian Produc-
might lead to a conflict of interests for any of the authors. tivity Organization, Tokyo, Japan, 1990.
[18] R. K. Roy, A Primer on Taguchi Method, Van Nostraid Reinhold,
References New York, NY, USA, 1990.
[19] P. J. Ross, Taguchi Techniques for Quality Engineering, McGraw-
[1] T. Ma, H. Yamaura, D. A. Koss, and R. C. Voigt, “Dry sliding
Hill, New York, NY, USA, 2nd edition, 1996.
wear behavior of cast SiC-reinforced Al MMCs,” Materials
Science and Engineering A, vol. 360, no. 1-2, pp. 116–125, 2003. [20] D. C. Montgomery, Design and Analysis of Experiments, John
Wiley & Sons, New York, NY, USA, 2001.
[2] Y. Yalcin and H. Akbulut, “Dry wear properties of A356-SiC
particle reinforced MMCs produced by two melting routes,” [21] R. A. Fisher, Design of Experiments, Oliver & Boyd, Edinburgh,
Materials and Design, vol. 27, no. 10, pp. 872–881, 2006. UK, 1951.
[3] S. C. Sharma, B. M. Girish, R. Kamath, and B. M. Satish, “Effect [22] J. Deng, “Introduction to grey system,” Journal of Grey System,
of SiC particle reinforcement on the unlubricated sliding wear vol. 1, no. 1, pp. 1–24, 1989.
behaviour of ZA-27 alloy composites,” Wear, vol. 213, no. 1-2, pp. [23] Minitab User Manual (Release 13.2), Making Data Analysis
33–40, 1997. Easier, MINITAB Inc, State College, Pa, USA, 2001.
[4] S. Ghosh, R. Behera, G. Sutradhar, and P. Sahoo, “Optimization
of friction performance of Al-5%SiC metal matrix composite
using Taguchi method,” Journal of Tribology Research, vol. 1, no.
2, pp. 83–89, 2010.
[5] S. Ghosh, P. Sahoo, and G. Sutradhar, “Friction performance of
Al-10% SiCp reinforced metal matrix composite using Taguchi
Method,” ISRN Tribology, vol. 2013, Article ID 386861, 9 pages,
2013.
[6] A. Onat, “Mechanical and dry sliding wear properties of
silicon carbide particulate reinforced aluminium-copper alloy
matrix composites produced by direct squeeze casting method,”
Journal of Alloys and Compounds, vol. 489, no. 1, pp. 119–124,
2010.
[7] H. Ahlatci, E. Candan, and H. Cimenoglu, “Abrasive wear
behavior and mechanical properties of Al-Si/SiC composites,”
Wear, vol. 257, no. 5-6, pp. 625–632, 2004.
[8] Y. Sahin, “Optimization of testing parameters on the wear
behaviour of metal matrix composites based on the Taguchi
method,” Materials Science and Engineering A, vol. 408, no. 1-
2, pp. 1–8, 2005.
[9] Y. Şahin, “Abrasive wear behaviour of SiC/2014 aluminium
composite,” Tribology International, vol. 43, no. 5-6, pp. 939–
943, 2010.
Journal of International Journal of International Journal of Smart Materials Journal of
Nanotechnology
Hindawi Publishing Corporation
Corrosion
Hindawi Publishing Corporation
Polymer Science
Hindawi Publishing Corporation
Research
Hindawi Publishing Corporation
Composites
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of
Metallurgy

BioMed
Research International
Hindawi Publishing Corporation Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Nanomaterials
Submit your manuscripts at
http://www.hindawi.com

Journal of Journal of
Materials
Hindawi Publishing Corporation
Nanoparticles
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Nanomaterials
Journal of

Advances in The Scientific International Journal of

Materials Science and Engineering
Hindawi Publishing Corporation
Scientifica
Hindawi Publishing Corporation Hindawi Publishing Corporation
World Journal
Hindawi Publishing Corporation
Biomaterials
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of Journal of Journal of Journal of Journal of

Nanoscience
Hindawi Publishing Corporation
Coatings
Hindawi Publishing Corporation
Crystallography
Hindawi Publishing Corporation
Ceramics
Hindawi Publishing Corporation
Textiles
Hindawi Publishing Corporation
http://www.hindawi.com
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Volume 2014