Materials and Design 70 (2015) 1–9

Contents lists available at ScienceDirect

Materials and Design

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

Technical Report

Tribological analysis of nano clay/epoxy/glass fiber by using Taguchi’s

technique
M.S. Senthil Kumar a,⇑, N. Mohana Sundara Raju b, P.S. Sampath a, U. Vivek a
a
Department of Mechanical Engineering, K.S. Rangasamy College of Technology, Tiruchengode, Tamil Nadu, India
b
Department of Mechanical Engineering, Mahendra Institute of Technology, Tiruchengode, Tamil Nadu, India

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

Article history: In this work, a detailed analysis was performed to profoundly study the tribological property of various
Received 13 October 2014 nano clay (Cloisite 25A) loaded epoxy, with and without inclusion of E-glass fiber using Taguchi’s tech-
Accepted 20 December 2014 nique. For this purpose, the test samples were prepared according to the ASTM standard, and the test was
Available online 31 December 2014
carried out with the assistance of pin-on-disk machine. To proceed further, L25 orthogonal array was con-
structed to evaluate the tribological property with four control variables such as filler content, normal
Keywords: load, sliding velocity and sliding distance at each level. The results indicated that the combination of fac-
Nano clay
tors greatly influenced the process to achieve the minimum wear and coefficient of friction. Overall, the
Wear
Friction coefficient
experiment results depicted least wear and friction coefficient for fiber reinforced laminates. In the same
Taguchi way, appreciable wear and friction coefficient was noted for without fiber laminates. Additionally, the SN
ANOVA ratio results too exhibited the similar trend. Moreover, ANOVA analysis revealed that the fiber inclusion
Surface morphology on laminates has lesser contribution on coefficient of friction and wear when compared to without fiber
laminates. At last, the microstructure behavior of the test samples was investigated with an assistance of
Scanning Electron Microscope (SEM) to analyze the surface morphology.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction powders (tungsten carbide and tantalum niobium carbide) filled

glass fabric added with epoxy [4], woven E-glass fibers reinforced
Based on the in depth review, Kato [1] recognized that wear and with plastic material (polypropylene) [5], diamond-like carbon
coefficient of friction are two primary factors that control the tribo- (DLC) flakes mixed with epoxy resin [6], high-density polyethylene
logical properties of coatings, composites, metallic alloys and (HDPE) reinforced with hydroxyapatite nanorods (nHA) [7], micro
ceramics. Furthermore, Mimaroglu et al. [2] illustrated that low and of nano scale SiO2 mixed with epoxy [8,9], aramid (Twaron)
density and tailoring capability of polymer composite and fiber fabric added with polytetrafluoroethylene [10], natural fiber (net-
reinforced polymer composite offer the attractive mechanical and tle, grewia optiva and sisal fiber) reinforced with Polylactic acid
tribological characteristics compared to the traditional metallic (PLA) [11], nano rubber particle added with epoxy resin [12], and
materials. In the past few decades or so, polymer composites are mica-filled polyamide 6 and short glass fiber reinforced with poly-
widely influenced in the application areas like gears, bearings sulfone polymer composites [13]. Recently, among these various
and brakes. Renukappa et al. [3] conducted a detailed study on nano fillers, major researches were carried out by using nano clay.
Organo-Modified Montmorillonite (OMMT) nano clay/epoxy resin, Besides, nano clay was first described by Montmorillon in the
and identified that in addition to mechanical properties, the tribo- department of Vienne, France in 1847 [14].
logical properties of the polymer composites could be enhanced Premachandran Nair et al. [15] studied the pin-on-disk machine
significantly by the inclusion of various nano fillers like nano tube, for testing three properties: wear properties, frictional properties
nano particle, nano clay, nano whiskers and nano fibers with the and contact resistance. The study further revealed that load, track
matrix. Besides that, various other types of nano fillers added with diameter, time and temperature were used as parameters to con-
polymer to improve the tribology property, for instance, hard trol the test environment. The technical potential of pin-on-disk
tribo machine could be attained by working under optimal param-
eters [16]. Okafor et al. [17] have identified that Taguchi technique
⇑ Corresponding author at: Department of Mechanical Engineering, K.S. Rangas-
amy College of Technology, Tiruchengode, Tamil Nadu 637 215, India. Tel.: +91
was one among the various optimization techniques and the opti-
04288 274741 43, mobile: +91 80121 84454; fax: +91 04288 274860/274745. mal parameters for manufacturing process could be arrived by
E-mail address: senthilkumarms@ksrct.ac.in (M.S. Senthil Kumar). constructing the orthogonal array. Another study conducted by

http://dx.doi.org/10.1016/j.matdes.2014.12.033
0261-3069/Ó 2014 Elsevier Ltd. All rights reserved.
2 M.S. Senthil Kumar et al. / Materials and Design 70 (2015) 1–9

persion of clay. The 2 mm chopped E-glass fiber supplied by Saint

Gobain was used to manufacture the test specimens.
To begin with, the epoxy (DGEBA) was heated up to 70 °C
(based on trial and error method) so as to scale back its viscous-
ness. Nano clay (Cloisite 25A) in wt% (2, 4, 6, 8, and 10) was dis-
persed in solvent acetone, individually. Then the dispersed clay
Fig. 1. Cloisite 25A structure. was mixed up with epoxy with an aid of mechanical stirrer. During
the entire process, the temperature of the mixture was maintained
at 70 °C with an assistance of heating mandrel. Using sonicator, the
Narasimha Murthy et al. [18] on twin screw extrusion for dispers-
nano clay/epoxy mixture was further stirred up for 30 min in order
ing montmorillonite (MMT) in vinylester by using orthogonal array
to evade the agglomeration. Also, degasification was carried out for
technique and gray relational analysis. Their study revealed that
30 min to facilitate the removal of bubbles from the clay/epoxy
4 wt% loading of MMT yielded highest gray relational grade for
blend. In conjunction, the hardener (HY 951) was supplemented
the mechanical properties of MMT/vinylester/glass composite.
into the clay/epoxy mixture once the temperature of mixture
Moreover, Biswas and Satapathy [19] identified that the perfor-
was cooled down to the room temperature.
mance characteristics of tribology property was identified by
The test specimens (10 mm diameter each) were prepared with
selecting the lower S/N ratio value. Owing to the superior features
the assistance of syringe, and chopped E-glass fiber was reinforced
of Taguchi method, Kumar Padhi [20] extended the application of
with the prepared matrix. Likewise, specimens were also fabri-
Taguchi method through an orthogonal array for friction and wear
cated for the same wt% (2, 4, 6, 8, and 10) by using syringe without
calculation of polymer composite material.
fiber. Furthermore, the specimens were allowed to 24 h at room
In general, MMT has a great surface area supporting enough
temperature for curing. Then the prepared specimens were cut to
interfacial regions in the nano composite; permitting improved tri-
ASTM:G99 for testing. Overall, among 2 wt%, 4 wt%, 6 wt%, 8 wt%,
bo mechanical properties at lower percentage loading with poly-
and 10 wt% of the fabricated laminates with and without fiber
mer. In recent years, various kinds of MMT nano clay such as
specimens, the optimum wt% of Cloisite 25A was obtained by test-
Cloisite Na+, Cloisite 30B, Cloisite 25A, Cloisite 20A, Cloisite 93A,
ing their individual sliding wear properties by using Taguchi
Cloisite 15A and Nanofil were used to modify the matrix so as to
technique.
improve the tribo mechanical properties, for instance, cotton lint
and cotton ginning waste [21], KO Buss Kneader [22], poly (lactic
acid) [23], bismaleimide [24], poly(trimethylene terephthalate) 2.2. Test apparatus
[25], epoxide monomer and DGEBA [26], polyamide 6 [27], 3-ami-
nopropyltriethoxysilane, 3-(glycidoxypropyl)trimethoxysilane, A pin-on-disk test rig (ASTM:G99, DUCOM – India) was used to
and 3-isocyanate propyltriethoxysilane [28] and polyurethane study the wear properties of the specimen. The experiment was
maleic anhydride grafted polypropylene (MAPP) [29]. conducted for the calculated speed, time and sliding distance in
In this study, Cloisite 25A nano clay, Araldite LY 556 epoxy, HY En-32 steel disk with 80 mm track diameter. The 10 mm diameter
951 hardener were used to prepare the matrix at various blends specimen was held within the specimen holder perpendicular to
0%, 2%, 4%, 6%, 8%, and 10%. In addition to, tribology test specimens the steel disk. Three trials were conducted for each specimen indi-
were prepared with and without reinforcement of E-glass fiber. In vidually in the pin-on-disk test rig and the average values were
conjunction, the wear properties and friction coefficient are deeply used. Before conducting the test, the disk was cleaned with ace-
discussed using Taguchi method (L25 orthogonal array), S/N ratio, tone. The experimental test set up can be witnessed in Fig. 2.
and Analysis of Variance (ANOVA).

2. Experimental details

2.1. Materials and specimen preparation

In this study, natural montmorillonite (MMT) modified with a

quaternary ammonium salt (Fig. 1) was used as an inclusion for
epoxy. Specifically, Cloisite 25A (hydrogenated tallow alkyl
methyl-2-ethylhexyl ammonium salts with bentonite). The clay
was provided by Southern Clay Products, Inc. The properties of
the Cloisite 25A is portrayed in Table 1. Generally, the base matrix
consists of two parts: Epoxy Part A-LY 556 based on diglycidyl
ether of Bisphenol A (DGEBA), which was supplied by Huntsman
Advanced Materials and hardener Part B-HY 951 (aliphatic primary
amine) supplied by Seenu & Co, Coimbatore, India. The hardener
was added to the epoxy in the ratio of 10:100 by weight. Then
again, laboratory grade acetone was used as a solvent for the dis- Fig. 2. Pin-on-disk test rig.

Table 1
Properties of nanoclay.

Nanoclay Organic modifier Modifier % % Weight loss on Density g/

concentration Moisture ignition cc
Cloisite 2MHTL8 (dimethyl, dehydrogenated tallow, 2-ethylhexyl quaternary 95 meq/100 g clay <2% 34% 1.87
25A ammonium)
 M.S. Senthil Kumar et al. / Materials and Design 70 (2015) 1–9 3

Table 2 machine and involves a sequence for the experiment to yield opti-
Levels of the variables used in the experiment. mum parameters for minimizing tribological properties [19]. Tagu-
Control factors Levels chi technique uses a range of the orthogonal arrays based on the
I II III IV V Units condition control factors and levels that helps to study the entire
parameter with only a minimum number of experiments. Further,
A: Filler content 2 4 6 8 10 %
B: Normal load 15 20 25 30 35 N
the signal-to-noise ratio (S/N) was calculated which is the logarith-
C: Sliding velocity 1 2 3 4 5 m/s mic function used to measure the performance of desired output.
D: Sliding distance 500 1000 1500 2000 2500 m In General, the S/N ratio is chosen based on three characteristics;
lower-the better, nominal-the better, higher-the better. Also, the
process parameter with the highest signal-to-noise ratio continu-
Table 3
ally yields the optimum quality with lowest variance. Besides,
Orthogonal array for L25 (54) Taguchi design.
the current study is focused on selecting better testing characteris-
Expt. no. A B C D tics by choosing lower-the better (LB) performance characteristics
1 1 1 1 1 for coefficient of friction (COF), frictions force (FF) and wear (w)
2 1 2 2 2 which can be calculated as logarithmic transformation of loss func-
3 1 3 3 3
tion as given in Eq. (1) [30–32].
4 1 4 4 4
5 1 5 5 5
S 1 X 2 
6 2 1 2 3 ¼ 10 log y ð1Þ
7 2 2 3 4 N n
8 2 3 4 5
9 2 4 5 1 where n = number of observations and y = observed data, S/N = sig-
10 2 5 1 2 nal to noise ratio.
11 3 1 3 5 The sliding wear investigation was carried out with four control
12 3 2 4 1
variables such as Filler content (A), Normal load (B), Sliding veloc-
13 3 3 5 2
14 3 4 1 3 ity (C) and Sliding distance (D) each at five levels as listed in Table 2
15 3 5 2 4 in connection with L25 orthogonal array. In usual factorial experi-
16 4 1 4 2 mental design, it would require 54 = 625 runs to review four
17 4 2 5 3 parameters each at five levels, whereas Taguchi’s approach reduces
18 4 3 1 4
19 4 4 2 5
it to exclusively 25 runs providing an excellent advantage in terms
20 4 5 3 1 of experimental time cost with noticeable accuracy. Furthermore,
21 5 1 5 4 the tests results were subjected to the Analysis of Variance
22 5 2 1 5 (ANOVA). The tests were conducted keeping in mind the experi-
23 5 3 2 1
mental design which is given in Table 3.
24 5 4 3 2
25 5 5 4 3
2.4. Microstructure

2.3. Experimental design The worn out surface of the specimens was investigated by
using scanning electron microscope (TESCAN – VEGA 3 – SBU)
Taguchi technique is one among the important statistical tech- underneath 20 kV accelerating voltage. In order to create the con-
niques which is in turn used in the present study to address the ducting layer on the worn out specimens, they were coated with
impact of multiple variables at the same time in pin-on-disk sputtering for a few nanometer thicknesses.

Table 4
Experimental design using L25 orthogonal array (without fiber).

Runs A (%) B (N) C (m/s) D (m) COF S/N ratio (db) WR (lm) S/N ratio (db)
1 2 15 1 500 0.52 5.76 58.50 35.34
2 2 20 2 1000 0.42 7.55 3.86 11.74
3 2 25 3 1500 0.05 26.24 5.35 14.56
4 2 30 4 2000 0.03 31.04 1.25 1.94
5 2 35 5 2500 0.05 25.68 2.54 8.10
6 4 15 2 1500 0.25 11.98 19.55 25.82
7 4 20 3 2000 0.15 16.42 34.75 30.82
8 4 25 4 2500 0.17 15.51 31.20 29.88
9 4 30 5 500 0.41 7.82 62.70 35.95
10 4 35 1 1000 0.23 12.69 132.70 42.46
11 6 15 3 2500 0.06 24.19 14.47 23.21
12 6 20 4 500 0.52 5.68 101.20 40.10
13 6 25 5 1000 0.15 16.67 76.70 37.70
14 6 30 1 1500 0.42 7.53 137.40 42.76
15 6 35 2 2000 0.30 10.39 102.07 40.18
16 8 15 4 1000 0.17 15.53 15.40 23.75
17 8 20 5 1500 0.17 15.49 41.47 32.35
18 8 25 1 2000 0.31 10.20 56.20 34.99
19 8 30 2 2500 0.40 7.96 91.70 39.25
20 8 35 3 500 0.43 7.43 241.70 47.67
21 10 15 5 2000 0.05 25.68 39.53 31.94
22 10 20 1 2500 0.22 13.20 39.29 31.88
23 10 25 2 500 0.82 1.68 98.23 39.85
24 10 30 3 1000 0.58 4.70 85.20 38.61
25 10 35 4 1500 0.44 7.07 94.30 39.49
4 M.S. Senthil Kumar et al. / Materials and Design 70 (2015) 1–9

Table 5
Experimental design using L25 orthogonal array (with fiber).

Runs A (%) B (N) C (m/s) D (m) COF S/N ratio (db) WR (lm) S/N ratio (db)
1 2 15 1 500 0.273 11.266 9.14 19.22
2 2 20 2 1000 0.260 11.683 11.43 21.17
3 2 25 3 1500 0.376 8.496 57.26 35.16
4 2 30 4 2000 0.347 9.187 128.67 42.19
5 2 35 5 2500 0.407 7.811 141.43 43.01
6 4 15 2 1500 0.645 3.802 27.43 28.77
7 4 20 3 2000 0.400 7.930 51.50 34.24
8 4 25 4 2500 0.247 12.133 11.70 21.36
9 4 30 5 500 0.050 26.016 10.18 20.15
10 4 35 1 1000 0.225 12.969 3.27 10.30
11 6 15 3 2500 0.741 2.599 36.53 31.25
12 6 20 4 500 0.199 14.000 8.60 18.69
13 6 25 5 1000 0.619 4.165 88.47 38.94
14 6 30 1 1500 0.271 9.982 267.73 48.55
15 6 35 2 2000 0.467 6.620 245.37 47.80
16 8 15 4 1000 0.588 4.612 16.30 24.24
17 8 20 5 1500 0.230 12.762 43.37 32.74
18 8 25 1 2000 0.196 14.140 40.67 32.18
19 8 30 2 2500 0.222 13.073 30.23 29.61
20 8 35 3 500 0.512 5.815 127.30 42.10
21 10 15 5 2000 0.433 7.263 23.73 27.51
22 10 20 1 2500 0.331 9.610 32.47 30.23
23 10 25 2 500 0.804 1.891 64.03 37.41
24 10 30 3 1000 0.528 5.552 43.50 32.77
25 10 35 4 1500 0.618 4.185 56.70 35.07

Fig. 3. (a) and (b) Effect of control factors on friction coefficient and wear (without fiber).
 M.S. Senthil Kumar et al. / Materials and Design 70 (2015) 1–9 5

Fig. 4. (a) and (b) Effect of control factors on friction coefficient and wear (with fiber).

Table 6 Table 7
S/N ratio response table for coefficient of friction (without fiber). S/N ratio response table for wear (without fiber).

Level Filler content Normal load Sliding velocity Sliding velocity Level Filler content Normal load Sliding velocity Sliding velocity
(%) (N) (m/s) (m) (%) (N) (m/s) (m)
1 19.255 16.629 9.877 5.673 1 14.34 28.01 37.49 39.78
2 12.884 11.669 7.912 11.427 2 32.99 29.38 31.37 30.85
3 12.892 14.061 15.797 13.663 3 36.79 31.40 30.97 31.00
4 11.321 11.809 14.965 18.745 4 35.60 31.70 27.03 27.97
5 10.465 12.649 18.266 17.309 5 36.35 35.58 29.21 26.46
Delta 8.789 4.960 10.354 13.072 Delta 22.45 7.56 10.45 13.32
Rank 3 4 2 1 Rank 1 4 3 2

3. Results and discussion Table 8

S/N ratio response table for coefficient of friction (with fiber).
3.1. Taguchi analysis for tribological properties Level Filler content Normal load Sliding velocity Sliding velocity
(%) (N) (m/s) (m)
In general, the wear resistance and friction coefficient consider- 1 9.689 5.909 11.593 11.797
ably improve for nano clay loading than the neat epoxy [3]. In this 2 12.57 11.197 7.414 7.796
regard, the intention of this study is to find the important factors 3 7.473 8.165 6.078 7.846
and combination of factors influencing the process to achieve the 4 10.08 12.762 8.823 9.028
5 5.7 7.48 11.604 9.045
minimum wear and coefficient of friction. By the way, the experi-
Delta 6.87 6.853 5.525 4.001
ments were developed based on an orthogonal array, with the aim Rank 1 2 3 4
of relating the influence of filler content, normal load, sliding veloc-
ity and sliding distance. These design parameters are distinct and
intrinsic feature of the process that influence and determine the the response tables for signal to noise ratios demonstrate the aver-
composite performance. Indeed, Taguchi recommends analyzing age of selected characteristics for each level of the factor.
the S/N ratio using conceptual approach that involves graphing The measured results were analyzed using the commercial soft-
the effects and visually identifying the significant factors. Also, ware MINITAB 14. In fact, the results for various combinations of
6 M.S. Senthil Kumar et al. / Materials and Design 70 (2015) 1–9

parameters were obtained by conducting the experiments as per the 3.2. ANOVA and the effects of factors
orthogonal array as shown in Tables 4 and 5. Also, Figs. 3 and 4 show
the effect of the four control factors on coefficient of friction and In order to observe statistical significance of various factors like
wear for both with and without fiber laminates. In addition to, the Filler content (%), Normal load (B), Sliding velocity (m/s), and Slid-
S/N ratio responses for coefficient of friction and wear are shown ing distance (m) on coefficient of friction and wear for both fiber
in Tables 6–9. Overall, the results direct to the conclusion that com- reinforced and without fiber laminates, Analysis of Variance
bination of factors A1B1C5D5 and A1B1C4D5 offer minimum friction (ANOVA) was performed on experimental data using MINITAB.
coefficient and wear for without fiber laminates. Conversely, Tables 10–13 show the results of the ANOVA with the coefficient
A2B4C5D1 and A2B1C4D2 offer minimum friction coefficient and wear of friction and wear. This analysis is undertaken for a level of con-
for fiber reinforced laminates. fidence of significance of 5%. The last columns of the tables (p-val-

Table 9
S/N ratio response table for wear (with fiber).

Level Filler content (%) Normal load (N) Sliding velocity (m/s) Sliding velocity (m)
1 35.15 26.20 28.10 27.52
2 22.96 27.41 32.95 25.48
3 37.05 33.01 35.10 36.06
4 32.18 34.66 28.31 36.78
5 32.60 35.65 32.47 31.09
Delta 14.08 9.46 7.00 11.30
Rank 1 3 4 2

Table 10
Analysis of variance for COF with fiber, using adjusted SS for tests.

Source DF Seq SS Adj SS Adj MS F P

A: Filler content (%) 4 0.19251 0.19251 0.04813 1.82 0.218
B: Normal load (N) 4 0.24970 0.24970 0.06243 2.36 0.140
C: Sliding velocity (m/s) 4 0.20647 0.20647 0.05162 1.95 0.195
D: Sliding distance (m) 4 0.02431 0.02431 0.00608 0.23 0.914
Error 8 0.21155 0.21155 0.02644
Total 24 0.88455

Table 11
Analysis of variance for COF without fiber, using adjusted SS for tests.

Source DF Seq SS Adj SS Adj MS F P

A: Filler content (%) 4 0.12750 0.12750 0.03188 2.35 0.142
B: Normal load (N) 4 0.06290 0.06290 0.01573 1.16 0.397
C: Sliding velocity (m/s) 4 0.20806 0.20806 0.05202 3.83 0.050
D: Sliding distance (m) 4 0.45210 0.45210 0.11303 8.32 0.006
Error 8 0.10873 0.10873 0.01359
Total 24 0.95930

Table 12
Analysis of variance for wear with fiber, using adjusted SS for tests.

Source DF Seq SS Adj SS Adj MS F P

A: Filler content (%) 4 33,561 33,561 8390 2.21 0.157
B: Normal load (N) 4 33,218 33,218 8305 2.19 0.160
C: Sliding velocity (m/s) 4 2842 2842 711 0.19 0.938
D: Sliding distance (m) 4 17,142 17,142 4286 1.13 0.407
Error 8 30,331 30,331 3791
Total 24 117,094

Table 13
Analysis of variance for wear without fiber, using adjusted SS for tests.

Source DF Seq SS Adj SS Adj MS F P

A: Filler content (%) 4 18,617 18,617 4654 3.96 0.046
B: Normal load (N) 4 21,985 21,985 5496 4.67 0.031
C: Sliding velocity (m/s) 4 5980 5980 1495 1.27 0.357
D: Sliding distance (m) 4 17,293 17,293 4323 3.68 0.055
Error 8 9408 9408 1176
Total 24 73,283
 M.S. Senthil Kumar et al. / Materials and Design 70 (2015) 1–9 7

Fig. 5. (a)–(f) Scanning electron microscope of 0 wt%, 2 wt%, 4 wt%, 6 wt%, 8 wt% and 10 wt% MMT filled epoxy.

ues) signify that the main effects are more and or less significant greater influence of fiber inclusion on epoxy laminates which
(i.e. the effects are statistically more significant only when the p- might have primarily lead to the lesser contribution on coefficient
values are less than 5%). From Table 10 (for fiber reinforced lami- of friction and wear when compared to without fiber laminates.
nates), one can observe that filler content, normal load, sliding
velocity, and sliding distance have relatively less significant contri- 3.3. Surface morphology
bution on coefficient of friction when compared to without fiber
laminates (see Table 11). This analogous trend could also be wit- Fig. 5a–f shows the images of the worn out surfaces of Scan-
nessed in the case of wear results for both with and without fiber ning Electron Microscope (SEM) for 0 wt%, 2 wt%, 4 wt%, 6 wt%,
laminates (see Tables 12 and 13). These trends clearly show the 8 wt% and 10 wt% MMT filled epoxy without fiber. Likewise,
8 M.S. Senthil Kumar et al. / Materials and Design 70 (2015) 1–9

Fig. 6. (a)–(f) Scanning electron microscope of 0 wt%, 2 wt%, 4 wt%, 6 wt%, 8 wt% and 10 wt% MMT filled epoxy with glass fiber.

Fig. 6a–f shows the images of the worn out surfaces of Scanning loading and 10 wt% loading too exhibited appreciable adhereness
Electron Microscope (SEM) for 0 wt%, 2 wt%, 4 wt%, 6 wt%, 8 wt% and similarities in their surface morphology. This adhereness
and 10 wt% MMT filled epoxy with E glass fiber. These images might be due to the better attraction between the MMT layers
obviously depict that neat epoxy composite is brittle in nature and the epoxy. Additionally, this trend might be due to the filling
and the outer surface of the composite is rough. Furthermore, it up of the micro-cracks gap by nano clay. This enhancement might
was observed that 2 wt% loading and 4 wt% loading disclosed bet- also be due to the superior matrix interface with glass fiber which
ter adhereness with epoxy than 6 wt% loading. Similarly, 8 wt% would have yielded reduced impact strength.
 M.S. Senthil Kumar et al. / Materials and Design 70 (2015) 1–9 9

4. Conclusions [11] Bajpai PK, Singh I, Madaan J. Tribological behavior of natural fiber reinforced
PLA composites. Wear 2013;297:829–40.
[12] Yu S, Hu H, Mab J, Yin J. Tribological properties of epoxy/rubber
After a thorough and critical analysis of the test results, three nanocomposites. Tribol Int 2008;41:1205–11.
major conclusions are drawn: [13] Demir Z. Tribological performance of polymer composites used in electrical
engineering applications. Bull Mater Sci 2013;36:341–4.
[14] http://www.mindat.org/min-2821.html.
 The first major conclusion (Taguchi analysis) exposes that the [15] Premachandran Nair R, Griffin D, Randall NX. The use of the pin-on-disk
results direct to the conclusion that combination of factors tribology test method to study three unique industrial applications. Wear
2009;267:823–7.
A1B1C5D5 and A1B1C4D5 offer minimum friction coefficient and
[16] Muhammed OS, Saleh HR, Alwan HL. Using of Taguchi method to optimize the
wear for without fiber laminates. On the other hand, combina- casting of Al–Si/Al2O3 composites. Eng Technol J 2009;27:1143–50.
tion of factors A2B4C5D1 and A2B1C4D2 offer minimum friction [17] Okafor EC, Ihueze CC, Nwigbo SC. Optimization of hardness strengths response
coefficient and wear for fiber reinforced laminates. of Plantain Fibers Reinforced Polyester Matrix Composites (PFRP) applying
Taguchi robust design. IJE Trans A: Basics 2013;26:1–11.
 The second major conclusion (ANOVA results) reveals that fiber [18] Sridhar R, Narasimha Murthy HN, Pattar N, Vishnu Mahesh KR, Krishna M.
inclusion on epoxy laminates had greater influence on coeffi- Parametric study of twin screw extrusion for dispersing MMT in vinylester
cient of friction and wear. Indeed, this had lead to the lesser using orthogonal array technique and gray relational analysis. Compos B
2012;43:599–608.
contribution on coefficient of friction and wear when compared [19] Biswas S, Satapathy A. Tribo-performance analysis of red mud filled glass-
to without fiber laminates. epoxy composites using Taguchi experimental design. Mater Des
 The third and the last major conclusion (SEM results) disclosed 2009;30:2841–53.
[20] Kumar Padhi P, Satapathy A, Kumar Mishra S, Mantry S. Erosive wear behavior
better adhereness between the MMT layers and the epoxy. In of glass-epoxy hybrid composites reinforced with blast furnace slag. In:
addition to, inclusion of E glass fiber within epoxy laminates International conference on ‘‘advancements in polymeric materials’’; 2012.
in various compositions viz 2 wt%, 4 wt%, 6 wt%, 8 wt% and [21] Delhom CD, White-Ghoorahoo LA, Pang SS. Development and characterization
of cellulose/clay nanocomposites. Compos B 2010;41:475–81.
10 wt% exhibited better morphological properties for 2 wt%
[22] Merinska D, Kubisova H, Kalendova A, Svoboda P, Hromadkova J. Processing
and 4 wt%. and properties of polyethylene/montmorillonite nanocomposites. J
Thermoplast Compos Mater 2012;25:115–31.
[23] Krishnamachari P, Zhang J, Lou J, Yan J, Uitenham L. Biodegradable poly(lactic
acid)/clay nanocomposites by melt intercalation: a study of morphological,
thermal, and mechanical properties. Int J Polym Anal Charact 2009;14:336–50.
References [24] Upadhyaya P, Roy S, Haque MH, Lu H. Influence of nano-clay compounding on
thermo-oxidative stability and mechanical properties of a thermoset polymer
[1] Kato K. Wear in relation to friction – a review. Wear 2000;241:151–7. system. Compos Sci Technol 2013;84:8–14.
[2] Unal H, Sen U, Mimaroglu A. Abrasive wear behavior of polymeric materials. [25] Xue M-L, Yu Y-L, Li P. Preparation, dispersion, and crystallization of the
Mater Des 2005;26:705–10. Poly(trimethylene terephthalate)/Organically Modified Montmorillonite (PTT/
[3] Rashmi, Renukappa NM, Suresha B, Devarajaiah RM, Shivakumar KN. Dry MMT) nanocomposites. J Macromol Sci Part B Phys 2010;49:1105–16.
sliding wear behavior of organo-modified montmorillonite filled epoxy [26] Kint DPR, Seeley G, Gio-Batta M, Burgess AN. Structure and properties of epoxy
nanocomposites using Taguchi’s techniques. Mater Des 2011;32:4528–36. based layered silicate nanocomposites. J Macromol Sci Part B Phys
[4] Mohan N, Natarajan S, KumareshBabu SP. Abrasive wear behavior of hard 2005;4:1021–40.
powders filled glass fabric–epoxy hybrid composites. Mater Des [27] Isik-Gulsac I, Yilmazer U, Bayram G. Mechanical and rheological properties,
2011;32:1704–9. and morphology of polyamide-6/organoclay/elastomer nanocomposites. J
[5] Hufenbach WA, Stelmakh A, Kunze K, Böhm R, Kupfer R. Tribo-mechanical Appl Polym Sci 2012;125:4060–73.
properties of glass fiber reinforced polypropylene composites. Tribol Int [28] Kim ES, Shim JH, Woo JY, Yoo KS, Yoon JS. Effect of the silane modification of
2012;49:8–16. clay on the tensile properties of nylon 6/clay nanocomposites. J Appl Polym Sci
[6] Ohana T, Nakamura T, Tanaka A. Tribological properties of polymer composites 2010;117:809–16.
with diamond-like carbon flakes. Diam Relat Mater 2010;19:894–8. [29] Chun BC, Cho TK, Chong MH, Chung Y-C, Chen J, Martin D, et al. Mechanical
[7] Li K, Chin Tjong S. Preparation and mechanical and tribological properties of properties of polyurethane/montmorillonite nanocomposite prepared by melt
high-density polyethylene/hydroxyapatite nanocomposites. J Macromol Sci mixing. J Appl Polym Sci 2007;106:712–21.
Part B Phys 2011;50:1325–37. [30] Kök M. Computational investigation of testing parameter effects on abrasive
[8] Zhi Wang Z, Gu P, Zhang Z, Gu L, Zhong Xu Y. Mechanical and tribological wear behavior of Al2O3 particle-reinforced MMCS using statistical analysis. Int
behavior of epoxy/silica nanocomposites at the micro/nano scale. Tribol Lett J Adv Manuf Technol 2011;52:207–15.
2011;42:185–91. [31] Okafor EC, Ilhueze CC, Nwigbo SC. Optimization of hardness strengths
[9] Kang Y, Chen X, Song S, Yu L, Zhang P. Friction and wear behavior of nanosilica- response of Plantain Fibers Reinforced Polyester Matrix Composites (PFRP)
filled epoxy resin composite coatings. Appl Surf Sci 2012;258:6384–90. applying Taguchi robust design. Int J Eng Trans A 2013;26:1–11.
[10] Liu P, Lu R, Wang H, Huang T, Li T. The mechanical and tribological properties [32] Cho MH, Bahadur S, Pogosian AK. Friction and wear studies using Taguchi
of aramid (Twaron) fabric/polytetrafluoroethylene composites. J Macromol Sci method on polyphenylene sulfide filled with a complex mixture of MoS2,
Part B Phys 2012;51:1693–704. Al2O3, and other compounds. Wear 2005;258:1825–35.