Int. J. Emerg. Sci.

, 2(3), 433-454, September 2012

ISSN: 2222-4254
IJES

Experimental Study on Optimization of Thermal

Properties of Groundnut Shell Particle Reinforced
Polymer Composites
G. U. Raju1*, V. N. Gaitonde2, S. Kumarappa3
1

Department of Mechanical Engineering, B. V. B. College of Engineering and Technology,

Hubli-580 031, Karnataka, India
Tel: +91 836 2378275; Fax: +91 836 2374985
2
Department of Industrial and Production Engineering, B.V.B. College of Engineering and Technology,
Hubli-580 031, Karnataka, India, Tel: +91 836 2378275; Fax: +91 836 2374985;
3
Bapuji Institute of Engineering and Technology, Davangere-577 004, Karnataka, India
[email protected],[email protected], [email protected]

Abstract Recently, thermoplastic and thermoset polymers are combined with

natural fillers to produce the composites, which possess better strength and good
resistance to fracture. Due to an excellent property profile, these composites find
wide applications in packaging, building and civil engineering fields. The present
work aims to elucidate the optimization of thermal properties such as thermal
conductivity, linear thermal expansion and specific heat of groundnut shell
particles reinforced polymer composite materials. The composite specimens were
prepared with different weight percentages of randomly distributed groundnut
shell particles in polymer matrix. The experiments were planned as per Taguchi
L9orthogonal array. The analysis of means (ANOM) was performed to determine
the optimal parameter levels and analysis of variance (ANOVA) was employed to
identify the level of importance of the parameters on each of the properties. TGA
and DSC analyses were also carried out to ascertain the thermal stability of these
composites. The results revealed that using groundnut shell particles as
reinforcement for polymer matrix could successfully develop beneficial
composites and can be used for thermal applications.
Keywords: Groundnut shell particles, Polymer resin, Thermal properties,
Taguchi design, TGA, DSC

1 INTRODUCTION
Due to increased environmental consciousness throughout the world the application of
natural fibers has drawn much attention in different engineering fields. The make use of
natural fibers as reinforcing materials in thermoplastics and thermoset matrix
composites provides optimistic environmental profits with regard to ultimate
disposability and better use of raw materials. The natural fibers are now believed to be

433

G. U. Raju, V. N. Gaitonde, S. Kumarappa

as an option to synthetic fibers such as glass fiber, carbon fiber, etc. Presently,
lignocellulosic bio-fibers as reinforcing materials are being utilized widely for the
manufacture of cost effective eco-friendly bio-composites. Due to better strength
properties such as easy availability, light weight, high toughness, non-corrosive nature,
low density, low cost, good thermal properties, reduced tool wear, less dermal and
respiratory irritation, less abrasion to processing equipment and renewability the natural
fibers are preferred over synthetic fibers and hence find wide applications in different
industries. In recent years, major industries such as automotive, construction and
packaging industries have shown enormous interest in the development of new biocomposite materials and are currently engaged in searching for new and alternate
products to synthetic fiber reinforced composites.
Many authors have reviewed the latest developments in the application of natural
fibers [1-4]. The widespread investigations on the preparation and properties of
thermoset and thermoplastic composites with the application of natural fibers such as
kenaf [5-6], jute [7-8], sisal [9-10], bagasse [11], bamboo [12], pineapple [13], rice
husk [14] and groundnut shell[15] have also been carried out. The natural fibers are
used for variety of appliances such as packaging, low-cost housing and structures and
the use of agricultural crop residues could progress rural agriculture based economy.
The various thermal properties such as thermal conductivity, diffusivity and
specific heat of polyester/ natural fiber (banana/sisal) composites were investigated by
Idicula et al. [16] as the function of filler concentration and fibre surface treatments. It
was observed that the composite thermal contact resistance decreases with chemical
treatment of the fibres. The heat transport ability of the compositewas also improved
due to hybridization of natural fiber with glass. The banana/sisal fiber-polyester
composites with 20 and 40-volumepercentage of fibers have thermal conductivity of
0.153-0.140 W m1 K1 and specific heat of 1199-1246 J kg1 K1 respectively were
observed. The developed sodium hydroxide treated fiber composites showed 43%
higher thermal conductivity than the untreated fiber composites. It was also noticed in
their study that the variation of specific heat is not so significant.
Behzad and Sain [17] studied the transverse and in-plane thermal conductivities for
oriented and randomly oriented composites for several volume fractions of fibers in
hemp fiber reinforced composites. It was found that the orientation of fibers has a
noteworthy influence on thermal conductivity of composites. Li et al. [18] determined
various thermal properties, namely, thermal conductivity, thermal diffusivity and
specific heat of flax fiberHDPE biocomposites around 170200oC temperature range.
The thermal conductivity, thermal diffusivity and specific heat found to be decrease
with increased fiber content, however there is no appreciable change in thermal
conductivity as well as thermal diffusivity in the specified temperate range. Conversely,
the specific heat of flax fiberHDPE composites steadily increased with temperature.
Thermal properties like thermal conductivity and thermal diffusivity of oil-palmfiber-reinforced composites with and without alkali treatment at room temperature were
analyzed by Agrawal et al. [19] using transient plane source technique. The study
showed that the treatments employed, namely, saline alkali and acetylation of fibers

434

International Journal of Emerging Sciences 2(3), 433-454, September 2012

increased the proposed thermal properties of oil-palm-fiber-reinforced composites. It

was reported that the saline alkali treated fiber has superior polarity due to the
formation of silonal group on the surface and hence resulting in elevated thermal
conductivity of saline treated composites. The alkalization treatment eliminates
impurities and enhances the fiber surface adhesion characteristic with the resin and
gives to a superior thermal conductivity. On the other hand, the acetylation to some
extent increased the polarity of the fiber with minor increasing thermal conductivity of
the oil-palm-fiber-reinforced composites.
Agarwal et al. [20] assessed the thermal conductivity as well as thermal diffusivity
properties of banana-fiber reinforced polyester composites with the addition of glass
fiber. They found that, even though the thermal conductivity increased when compared
to polymer matrix, but the thermal conductivity decreases with increased percentage of
glass fiber compared to pure banana fiber composites. Alsina et al. [21] estimated the
thermal properties of jute-cotton, sisal-cotton and ramie-cotton hybrid fabric reinforced
unsaturated polyester composites. The volume fractions of ramie, sisal and jute fibers in
the fabrics were found to be 0.77, 0.69, and 0.64 respectively. The results showed that
sisal-cotton hybrid polyester composites have thermal conductivity 0.213-0.25 W/m-k
and specific heat of 1.065-1.236 J/cm3 oC; Jute-cotton hybrid polyester composites have
thermal conductivity 0.10-0.237 W/m-k and specific heat of 0.869-1.017 J/cm3 oC;
Ramie-cotton hybrid polyester composites have thermal conductivity 0.19-0.22 W/m-k
and specific heat of 0.839-0.894 J/cm3 oC.
A major deficiency in the natural fiber-plastic composites is the poor bonding
between the natural fiber and the plastic, mainly due to dissimilar chemical nature of
both the materials. The natural fiber surface is hydrophilic and that of the plastic is
hydrophobic. Hence, in order to have enhanced mechanical and thermal properties, it is
essential to apply hydropobicity to natural fibers by appropriate treatments to the
composites. Recent studies showed that surface modification techniques, namely,
chemical treatments, acetylation and graft co-polymerisation are used to overwhelm the
incompatible surface polarities between the natural fiber and the polymer matrix. The
chemical treatment permits a better make contact with fiber-matrix and diminishes the
thermal contact resistance significantly [16]. It has been reported that NaOH chemical
treatment of fiber allows major increase of thermal and mechanical properties of
composites [15, 22-24]. The treatment with NaOH eliminates almost all non-cellulose
components except waxes. By the dissolution of lignin by alkali, some pores are formed
on the surface of fiber, which progresses the contact area between the fiber and the
matrix.
The natural fiber-plastic composites found numerous applications in automotive
sector, building and construction, even though these materials have poor compatibility
between hydrophilic natural fibers and hydrophobic polymer matrices. As per authors
information, no investigation has been discussed in the literature on optimization of
thermal properties of groundnut shell particles reinforced polymer composite materials.
Hence, an attempt has been made in this paper to optimize the thermal properties of
groundnut shell particles reinforced polymer composite (GSPC) materials using

435

G. U. Raju, V. N. Gaitonde, S. Kumarappa

Taguchi technique. The novel bio-based composite materials were prepared from
groundnut shell particles in polymer matrix. Taguchi L9 orthogonal array was used to
conduct the experiments. The analysis of means (ANOM) was employed to identify the
optimal level of each parameters and the analysis of variance (ANOVA) was used to
find the relative importance among the parameters.

2 MATERIALS AND METHOD

2.1 Groundnut Shell

Groundnut botanically known as Arachishypogeae belongs to Leguminosae family. It is

the fourth largest oilseed produced in world and India is the second largest producer of
groundnut after China. In India, groundnut is the largest oilseed in terms of production
and accounted for about 7.5 million tons during 2009-10. A complete seed of groundnut
is called as pod and outer layer of groundnut is called shell. Groundnut shell chemical
composition is compared with some of the available natural fibers and is shown in
Table 1. The hemicelluloses content of the fiber is found to be 18.7%, cellulose 35.7%,
lignin 30.2% and ash content 5.9%. Lignin is often called the cementing agent that
binds individual fiber cells together. The lignin content of groundnut shell fiber is much
greater than that of banana, baggase, rice husk, jute, hemp, kenaf and sisal fibers. The
hemicellulose is accountable for substantial amount of moisture absorption. The hemi
cellulose content of groundnut shell is less than wood, banana, baggase, rice husk and
kenaf fibers. Pre-treated groundnut shell is used in this study to modify the surface
properties to ensure interfacial interactions between the particles and the resin.
Table 1.Chemical composition of natural resources.

Species

Cellulose
(wt%)

Hemicellulose
(wt%)

Lignin
(wt%)

Ash
(wt%)

Reference

Pine (softwood)
Maple (hardwood)
Banana
Coir
Sisal
Jute
Kenaf
Hemp
Bagasse
Groundnut shell
Rice husk
Pineapple

40-45
45-50
63-64
32-43
63-64
61-71.5
31-39
70.2-74.4
40-46
35.7
31.3
81

25-30
22-30
19
0.15-0.25
12
12-20.4
21.5
17.9-22.4
24.5-29
18.7
24.3
-

26-34
22-30
5
40-45
10-14
11.8-13
15-19
3.7-5.7
12.5-20
30.2
14.3
12.7

2
1.5-2.4
5.9
23.5
-

[15]
[15]
[16]
[25]
[25]
[26]
[26]
[26]
[27]
[15]
[28]
[16]

436

International Journal of Emerging Sciences 2(3), 433-454, September 2012

2.2 Polymer Resins

In the present study three different polymer resins, namely, epoxy, vinyl ester and
polyester were used as matrix materials. Epoxy resin is a polymer containing two or
more epoxy groups and has high mechanical properties due to its low shrinkage and
relatively unstressed structures. Epoxy resin system exhibits extremely high resistance
alkali, good acids and solvent resistance. It has good electrical properties over a range
of frequencies and temperature. The cured epoxy systems generally exhibit good
dimensional stability, thermal stability and exhibit resistance to most fungi. They are
self-excellent moisture barriers exhibiting low water absorption and moisture
transmission. The epoxy of grade LY554 and hardener HY951 was used with the
weight ratio of 10:1 to prepare the composite specimens. Vinyl ester resin exhibits a
polyester resin type of cross-linking molecules in the bonding process and is tougher
and more resilient than polyesters. A vinyl ester resin has excellent physical and
mechanical properties and is familiar for its versatility as a composite matrix. The
processability of vinyl ester resin at low temperatures has drawn substantial
responsiveness from the composite industry. The vinyl ester of grade GR 200-60 was
used with hardener, catalyst and accelerator with 1.5-wt % to prepare the composite
specimens. Polyester resin is usually used as matrix material in polymer composites, for
instance fiber-reinforced plastics and polyester concrete. Polyester resins have good
range of mechanical properties. Polyester resin is durable, comparatively inexpensive,
superior corrosion resistance and little weight. The polyester of grade PxGp 002 and the
catalyst benzoyl peroxide with prescribed proportion was used to develop the composite
specimens.
2.3 Taguchi Technique

Taguchi technique is a powerful methodology to contract with the response controlled

by the number of parameters. Taguchi design is applied to devise the experimental
layout, examine the effect of each parameter and to determine the optimal level of each
identified parameter. Taguchi design utilizes an orthogonal array to study the whole
space with minimum number of experiments [29, 30] and hence it is achievable to
condense time and cost of the experimental research. Taguchi method consists of plan
of experiments, in which the factors are situated at different rows in an intended
orthogonal array. An orthogonal array provides more consistent estimates of factor
effects with little number of experiments, when compared to conventional methods.
Depending on number of factors and identified levels of each factor, an appropriate
array is chosen [29, 30]. Each column of the orthogonal array delegates a parameter and
its setting levels in each experiment and each row entrusts an experiment with the level
of several parameters in that experiment.
After performing the experiments as per orthogonal array (OA), the results are
converted into signal to noise (S/N) ratio data. In Taguchi technique, the term signal

437

G. U. Raju, V. N. Gaitonde, S. Kumarappa

represents desirable value (mean) and noise represents undesirable value (standard
deviation) for the response. Therefore, S/N ratio is the mean to standard deviation,
which specifies the degree of predictable performance of a product or process in the
existence of noise factors [29, 30]. The S/N ratio is used to measure the performance
characteristics and to identify the important parameters through analysis of variance
(ANOVA). Taguchi classifies objective functions into three categories, namely, smaller
the better type, larger the better type and nominal the best type [29, 30]. The optimum
level for a factor is the level, which results in highest S/N ratio value in the
experimental space.

3 EXPERIMENTAL DETAILS
Clean and dried groundnut shells were initially washed with water to take away the
sand and other impurities. The washed shells were later chemically treated with 10%
NaOH solution for 2 hours and then washed with distilled water until all NaOH gets
eliminated. Subsequently, the shells were solar dried and ground. Then the particles
were sieved through 0.5, 1 and 1.5 mm BS sieves to get different size groundnut shell
particles. These particles are used as reinforcement material in polymer matrix.
3.1 Preparation of Composite Boards
A mould with the dimension of 130 mm in diameter and 10 mm thick was used to
prepare the composite specimen. A layer of wax was applied to the mould so that the
specimen can be easily taken out of the mould. Measured quantities of groundnut shell
particles and resin were taken in a plastic container and stirred thoroughly to get
homogeneous mixture. After adding the suitable quantity of hardener and catalyst, the
mixture was again stirred for 10 minutes and thoroughly mixed mixture was placed in
the mould and compressed uniformly. This set up allowed for curing and then the
composite specimen was taken out from the mould. Curing time was different for
different resins. Groundnut shell particles reinforced polymer composite (GSPC)
specimens were prepared by varying three parameters, namely, particle size, weight
percentage of reinforcement material and matrix material.
3.2 Orthogonal Array Selection
In the present study, particle size, weight percentage of reinforcement material and
matrix material are selected as the process parameters, which affect the thermal
properties, namely, thermal conductivity, linear thermal expansion and specific heat of
groundnut shell particles reinforced polymer composite materials. Each parameter was
examined at three levels to study the non-linearity effect of the process parameters. The

438

International Journal of Emerging Sciences 2(3), 433-454, September 2012

selected process parameters and their levels are given in Table 2. In the present
investigation, with three parameters at three levels each, L9 (34) orthogonal array (OA)
[29, 30] is used and accordingly nine GSPC specimens were prepared as per the
experimental layout plan (Table 3).
Table 2.Process parameters and their levels selected for the preparation of GSPC specimens.

Levels

Code

Parameters

A
B
C

Particle size (mm)

Groundnut shell particle (wt %)
Polymer resin

0.5
20
Epoxy resin

1
40
Vinyl ester

1.5
60
Polyester resin

Table 3.Experimental layout plan and thermal properties.

Levels of process parameter settings

Thermal properties
Trial Particle Groundnut
Polymer
Thermal
Linear thermal
Specific
no. size
shell particle resin
conductivity expansion 10-5 heat
(A)
(B)
(C)
(W/m-K)
(/ 0C)
(J/kg-K)
1
2
3
4
5
6
7
8
9

1
1
1
2
2
2
3
3
3

1
2
3
1
2
3
1
2
3

1
2
3
2
3
1
3
1
2

0.2545
0.1956
0.2061
0.2580
0.1703
0.1918
0.3140
0.1983
0.1262

10.43
5.33
4.62
5.29
4.19
3.69
3.46
3.36
3.25

1410.58
1043.02
1805.06
1448.36
2446.09
2188.08
1950.01
2099.78
1452.02

3.3 Thermal Conductivity Test

A circular disc shaped specimen of size 130 mm in diameter and 10 mm thick was
prepared for each of the trials as per the layout plan of OA. For measuring the thermal
conductivity of composites, the method presented in Behzad and Sain [17] was used.
The experimental setup consists of heating element, connected to a conducting material
of same size as that of the specimen. Three thermocouples were connected to the
specimen at different points. The entire setup was placed in a thermally insulated
evacuated chamber in order to prevent loss of heat from specimen. The heat supplied
was maintained constant until steady state is reached and then temperature at the
thermocouples was noted using digital temperature indicator. The thermal conductivity
was determined, measuring temperatures, heat supplied and area of the specimen by
using a discrete approximation of Fouriers law for one dimensional heat conduction,
given by:

439

G. U. Raju, V. N. Gaitonde, S. Kumarappa

Q KA

dT
dx

(1)

where, Q: Heat dissipated through the plate; K: thermal conductivity of the composite
plate; A: surface area of the specimen; dT : temperature difference (T1-T2); dx :
thickness of the disc. The validation of the test was done through measuring the thermal
conductivities of known materials such as neat epoxy, neat vinyl ester and neat
polyester. It was observed that the determined thermal conductivity values substantiate
the standard values with the greatest accuracy. The thermal conductivity for each trial
of an orthogonal array was obtained by averaging five measurements at various
positions of the prepared specimen and the mean values of the nine trials are presented
in Table 3.
3.4 Linear Thermal Expansion Test
Each specimen of size 150 40 10 mm3 was prepared as per the layout plan of OA
for carrying out the linear thermal expansion test. The experimental setup consists of
heater and conductive material such as aluminum plate. The specimen was kept over
the plate and dial gauges were placed at the ends of the specimen at different points to
measure the deflection. Thermocouples were placed in the specimen and are connected
to the temperature-measuring instrument. The initial temperature of the specimen was
recorded and then heated uniformly. For the increase in temperature, corresponding
deflection of the specimen was recorded at equal intervals. The coefficient of thermal
expansion of GSPC specimens was determined by averaging five readings and using the
following formula:
L L0 ( 1 t )
(2)
where, L0and L:initial and final length of the specimen at temperatures Ti and Tf; :
coefficient of thermal expansion; Ti and Tf: initial and intermediate temperatures of the
specimen; t = (Tf - Ti) temperature difference. The linear thermal expansion for each
trial of an orthogonal array was obtained by averaging five measurements at various
positions of the prepared specimen and the mean values of the nine trials are
summarized in Table 3.
3.5 Specific Heat Test
The spherical shaped composite specimen of size 30 mm diameter was prepared for
each of the trials of the experimental layout plan (Table 3).Three thermocouples were
placed at different depths and positions of the specimens and are connected to the
temperature indicator. The specimen was placed in a thermally insulated evacuated
chamber and the specific heat of each GSPC specimen was calculated by measuring
heat supplied, change in temperature and mass of the specimen using the formula:

440

International Journal of Emerging Sciences 2(3), 433-454, September 2012

Cp

Q
T * m

(3)

where, Q: heat supplied; T: change in temperature (Tf Ti); m: mass of the specimen.
The specific heat for each trial of an OA was obtained by averaging five measurements
at various positions of the prepared specimen and the mean values of the nine trials are
illustrated in Table 3.

4 ANALYSIS OF DATA, RESULTS AND DISCUSSION

Table 3 summarizes the experimental results of thermal conductivity, linear thermal
expansion and specific heat of groundnut shell particles reinforced polymer composite
materials. It is observed that the GSPC materials have thermal conductivities in the
range 0.1262 and 0.3140 W/m-K. The specimens having epoxy matrix with particle size
of 1 mm and highest weight (60%) of filler material decreases the thermal conductivity.
However, the specimens having vinyl ester matrix with 1.5 mm particle size and highest
weight (60%) of particle shows decreased thermal conductivity. On the other hand,
polyester matrix composites with medium ranges of particle size (1 mm) and 40% filler
material exhibit lower thermal conductivity. Decrease in thermal conductivity for
higher filler loading is due to the lower thermal conductivity of groundnut shell particle
filler material. Several researchers [16, 31] were also observed similar behaviour of
composites.
From Table 3, it is seen that the specimens have linear thermal expansion ranging
from 3.2489 10-5 to 10.4310 10-5 /C. It has been observed that decreased weight %
of particle increases the coefficient of linear thermal expansion for epoxy as well as
vinyl ester matrix composites. This behavior is possibly due to lower thermal expansion
of filler material than the epoxy and vinyl ester matrix. However, polyester matrix
composite exhibits increased linear thermal expansion with increase in weight
percentage of particles. It was also found that linear thermal expansion of composite
specimen is more for smaller the grain size of the groundnut shell particles for all the
polymer resins considered in the study.
The results of the specific heat capacity of different groundnut shell particle
reinforced polymer composite specimens (Table 3) revealed that the specific heat
capacity of GSPC specimen is high for 1 mm grain size for all the polymer resins
specified. The specific heat capacity is also high for 60-wt% of particles in case of
epoxy and vinyl ester resins; whereas higher heat capacity is observed for 40-wt% of
particles for polyester resin.
4.1 Analysis of Experimental Data Based on S/N Ratio

441

G. U. Raju, V. N. Gaitonde, S. Kumarappa

In Taguchi design, the S/N ratio analysis has been carried out to determine the optimal
parametric condition for each of the thermal properties considered. In our present
investigation, to obtain the optimal operating parameters, smaller the better type
category is used for thermal conductivity and linear thermal expansion; whereas larger
the better type is employed for specific heat.

1 n 2
yi ]
n i1

(4)

1 n 2
yi ]
n i1

(5)

S/N ratio for smaller-the-better type category is 10 log 10 [

S/N ratio for larger-the-better type category is 10 log 10 [

where, y is the response and n is the number of replications for each trial i. The
computed values of S/N ratio () for each trial of L9OA for each of the thermal
properties are demonstrated in Table 4.
Table 4.Computed values of S/N ratios for thermal properties.

S/N ratio (dB) for thermal properties

Trial
no.

Thermal conductivity

Linear thermal expansion

Specific heat

1
2
3
4
5
6
7
8
9

11.8862
14.1726
13.7184
1.7676
15.3757
14.3430
10.0614
14.0535
17.9788

-20.3657
-14.5345
-13.2928
-14.4691
-12.4443
-11.3405
-10.7815
-10.5268
-10.2377

62.9880
60.3659
65.1298
63.2175
67.7694
66.8013
65.8007
66.4435
63.2395

4.2 Analysis of Means and Analysis of Variance

The data analysis using Taguchi design entails analysis of means (ANOM) and analysis
of variance (ANOVA). The process of determining the direct effects of each variable is
the ANOM and the effect of a parameter level is the deviation it causes from the overall
mean response [29].The S/N ratio for each process parameter level is determined by
averaging the S/N ratios when the parameter is kept at that level. The ANOM helps in
identifying the optimal factor combinations. On the other hand, ANOVA ascertains the
comparative importance of the parameters in terms of % contribution to the response
[29, 30]. ANOVA is also required for determining the error variance for the effects and
variance of prediction error. This is to be accomplished by separating the total

442

International Journal of Emerging Sciences 2(3), 433-454, September 2012

variability of S/N ratio, which is measured by sum of squared deviations from the total
mean S/N ratio, into contributions by each of the design parameters and the error [29,
30]. The % contribution designates the relative power of a parameter to diminish
variation. For a parameter with a high % contribution with a small variation has a huge
control on the response.
15.5

S/N ratio for thermal conductivity (dB)

15
14.5
14
13.5
13
particle size
wt % of filler material

12.5

polymer resin
overall mean

12
11.5
11

2
Level

Figure 1.Response graph of S/N ratio for thermal conductivity.

443

G. U. Raju, V. N. Gaitonde, S. Kumarappa

-10

S/N ratio for linear thermal expansion (dB)

-11

-12

-13

-14
particle size
wt % of filler material
-15

polymer resin
overall mean

-16

-17

2
Level

Figure 2.Response graph of S/N ratio for linear thermal expansion.

The results of ANOM are represented in response graphs (Figures 1, 2, and 3). The
level of a process parameter with highest signal to noise (S/N) ratio value is the
optimum level. As seen in Figure 1,the optimal combination of process parameter
settings for minimizing the thermal conductivity of groundnut shell particles reinforced
polymer composite is A3, B3 and C2 i.e. the specimen having particle size of 1.5 mm
with 60-wt% of particles using vinyl ester as the matrix material. Figure 2presents the
optimal combination of process parameter settings for minimizing the linear thermal
expansion of groundnut shell particles reinforced polymer composite, which is given by
A3, B3 and C3 i.e. the specimen having particle size of 1.5 mm with 60-wt% of
particles using polyester resin as the matrix material. On other hand, as demonstrated in
Figure 3, it was found that the optimal combination process parameter settings for
maximizing the specific heat of groundnut shell particles reinforced polymer composite
is A2, B3 and C3 i.e. the specimen having particle size of 1.0 mm with 60-wt% of
particles using polyester resin as the matrix material. However, the comparative
magnitude among the process parameters has to be investigated through the ANOVA.

444

International Journal of Emerging Sciences 2(3), 433-454, September 2012

66.5
particle size
wt % of filler material

66

polymer resin
overall mean

S/N ratio for specific heat (dB)

65.5
65
64.5
64
63.5
63
62.5
62

2
Level

Figure 3.Response graph of S/N ratio for specific heat.

Tables 5-7 present the summary of ANOVA results for thermal conductivity, linear
thermal expansion and specific heat of groundnut shell particles reinforced polymer
composite materials. From Table 5, it can be seen that the % of filler material has major
influence (67.39%) on minimizing thermal conductivity and the vinyl ester matrix
material has less effect (9.81%), whereas the particle size does not have significant
effect in minimizing thermal conductivity. On the other hand, the particle size has more
contribution (58.12%) for minimizing thermal expansion followed by the % of filler
material (25.98%). However, the polyester resin matrix has the least effect for
minimizing thermal expansion as shown in Table 6. It is clear from Table 7 that, the
polyester resin matrix material is the most dominant significant parameter (59.13%) and
the particle size (35.35%) has the moderate effect to maximize the specific heat.
Table 5.Summary of ANOVA on S/N ratio for thermal conductivity.

Source

Degrees of
freedom

Sum of
squares

Mean
square

% contribution

Particle size (A)

Groundnut shell particle (B)
Polymer resin (C)

2
2
2

0.9617
28.3997
4.1315

0.4809
14.1998
2.0658

2.28
67.39
9.81

445

G. U. Raju, V. N. Gaitonde, S. Kumarappa

Error
Total

2
8

8.6483
42.1412

4.3242
5.2677

20.52
100

Table 6.Summary of ANOVA on S/N ratio for linear thermal expansion.

Source

Degrees of
freedom

Sum of
squares

Mean
square

% contribution

Particle size (A)

Groundnut shell particle (B)
Polymer resin (C)
Error
Total

2
2
2
2
8

46.7677
20.9095
5.4463
7.3507
80.4742

23.3838
10.4548
2.7232
3.6754
10.0593

58.12
25.98
6.76
9.14
100

Table 7.Summary of ANOVA on S/N ratio for specific heat.

Source

Degrees of
freedom

Sum of
squares

Mean
square

% contribution

Particle size (A)

Groundnut shell particle (B)
Polymer resin (C)
Error
Total

2
2
2
2
8

15.6541
1.8868
26.1890
0.5569
44.2868

7.8271
0.9434
13.0945
0.2785
5.5359

35.35
4.26
59.13
1.26
100

4.3 Verification of Experiments

Once the optimal levels of the process parameters have been identified for each of the
thermal properties, the next step is to predict and verify the performance characteristic
using the optimal level of design parameters [29, 30]. The predicted optimum value of
S/N ratio (opt) of the response is determined by the formula [29, 30]:
k1

opt m [( mi , j )max - m ]

(6)

j 1

where, m is the overall mean of S/N ratio; (mi,j)max is the S/N ratio of optimum level i of
factor j and k1 is the number of main design parameter that affect the response. In order
to see the closeness of observed value of S/N ratio (obs)with that of the predicted value
(opt), the confidence interval (CI) value of opt for the optimum parameter level
combination at 95% confidence level is calculate, given by [30]:

CI F( 1,e )Ve (

1
1

)
neff nver

446

(7)

International Journal of Emerging Sciences 2(3), 433-454, September 2012

where, F( 1 ,e ) is the F value for 95% confidence interval; e is the degrees of freedom

N
; N is the total trial number in
1
orthogonal array ; = degrees of freedom of k1 factors and nver is the confirmatory test
trial number. If the error of prediction i.e. (optobs) is within CI value, then the
optimum process parameter level combination and additive model for the variable
effects are valid. Here, three validation experiments were conducted at optimal levels of
process parameters for each of the thermal properties and the results of conformity tests
are presented in Table 8. It is observed that the calculated value of prediction error of
each of the thermal properties is within the confidence limit, thus clearly indicating the
adequacy of the additivity of thermal property models. The best combinations of
process parameters for achieving minimum thermal conductivity and thermal expansion
and for maximum specific heat capacity along with the corresponding optimal values of
thermal properties are exhibited in Table 9.
for error; Ve is the variance of error; neff

Table 8.Results of the verification tests.

Performance measure

Thermal
conductivity

Linear thermal
expansion

Specific heat

Levels (A, B, C)
S/N predicted( opt), dB
S/N observed(obs), dB
Prediction error, dB
Confidence interval (CI), dB

3, 3, 2
16.6049
17.9788
1.3739
7.8880

3, 3, 3
-8.0912
-9.7144
1.6232
7.2722

2, 3, 3
67.9406
68.7776
-0.837
2.0018

Table 9. Best combination values of the process parameters and the corresponding optimal values of
thermal properties.

Optimal process parameter settings

Thermal
Property
Thermal
conductivity
Linear thermal
expansion
Specific heat

Optimal value

Particle size
(mm)

Groundnut shell
particle (wt%)

Polymer resin

1.5

60

Vinyl ester

0.1262 W/m-K

1.5

60

Polyester resin

3.06

60

Polyester resin

2747.13 J/kg-K

4.4 Thermo-Gravimetric Analysis (TGA)

447

10-5 / 0C

G. U. Raju, V. N. Gaitonde, S. Kumarappa

Thermo-gravimetric analysis (TGA) is employed to analyze the thermal stability of

composites. Thermal stability is believed to be one of the limiting factors in the use of
natural fibers when compared to synthetic fibers. Hence, thermal stability of groundnut
shell particles reinforced polymer composite has been investigated in the current
investigation. The TGA tests were performed for all the nine developed GSPC
specimens as per ASTM E1131 standard. All the measurements were performed using
TG analyzer instrument of high Resolution SDT Q600 V20.9. The samples weighing
between 10 and 20mg were placed in a platinum pan and tests were performed within
the temperature range of 20700o C at a heating rate of 5o C/min under nitrogen
atmosphere at flow rate of 50 ml/min. TG and DTG curves were exercised to study the
high temperature degradation behavior of GSPC specimens. The combined thermogravimetric analysis (TGA) and differential thermal analysis (DTA) results of GSPC
composites are depicted in Figures 4 and 5respectively. S1 to S9 represent the GSPC
specimens prepared as per the trial numbers of an experimental layout plan of Table 3.
Table 10.Thermal degradation data of GSPC specimens.

Onset
End of
Weight
Speci degradation degradation
loss
mens temperature temperature
(%)
o
o
( C)
( C)

Weight loss
peak
temperature
(oC)

Melting point
temperature
(oC) #

S1
118.63
440.59
72.79
375.19
114.44
S2
157.97
509.24
82.80
402.05
159.72
S3
116.32
486.90
67.84
321.42
112.74
S4
149.49
471.28
69.49
396.38
148.73
S5
143.86
447.96
72.60
325.25
145.03
S6
122.97
467.09
79.98
382.19
115.34
S7
118.40
435.43
66.90
325.01
113.35
S8
155.77
458.87
73.20
379.25
158.32
S9
164.30
477.36
78.09
404.44
162.12
: Weight loss peak temperature determined from DTA thermograms.
# Melting point temperature determined from DSC curves.

448

Residual
char at
690oC
(wt. %)
20.97
13.82
28.54
25.52
20.94
15.05
24,80
22.02
16.73

International Journal of Emerging Sciences 2(3), 433-454, September 2012

S1
S2
S3
S4
S5
S6
S7
S8
S9

100

Weight loss (%)

80

60

40

20

0
0

100

200

300

400

500

600

700

Temperature ( C)

Figure 4.TGA thermograms of GSPC specimens.

The TGA curves of GSPC specimens provide three distinct temperature regions,
wherein the samples experience major weight loss. A small weight loss was observed
during first region attributed to the evaporation of moisture [32, 33]. Actual degradation
happens in second region attributed to the thermal degradation of hemicelluloses,
cellulose and lignin [34, 35] together with polymeric matrix and thereafter the rate of
decomposition was slow. These results are also confirmed by DTA curves wherein, it
can be appreciated that most decomposition occurs at the temperature of 321 to 405C
(second region). Quantitative data in second region including onset thermal degradation
temperature, end of degradation temperature, corresponding weight loss, weight loss
peak temperatures and the char yields at 690 oC are presented in Table 10. As can be
seen from TGA (Figure 4) and DTA (Figure 5) curves, the thermal decomposition
initiation temperature is the lowest for S3 composite specimen (particle size of 0.5 mm,
60-wt% of filler material with polyester resin) exhibiting 116.32 oC and the highest of
157.53 oC for S9 composite specimen (particle size of 1.5 mm, 60-wt% of filler material
with vinyl ester resin), indicating the higher thermal stability for S9 composite. On the
other hand, it was observed that, S2 composite specimen (particle size of 0.5 mm, 40wt% of filler material with vinyl ester resin) is having subsequent higher thermal
stability with 155.10oC onset degradation temperature. For S3 composite specimen, the

449

G. U. Raju, V. N. Gaitonde, S. Kumarappa

weight loss peak took place at a lower temperature of 321.42 oC, while it is the highest
for S9 composite specimen occurred at 404.44oC. Thermal degradation of S3 composite
specimen produced much higher charred residue of about 28.54 %.
1.0

S1
S2
S3
S4
S5
S6
S7
S8
S9

0.6

Derivative weight (%/ C)

0.8

0.4

0.2

0.0

-0.2
0

100

200

300

400

500

600

700

Temperature ( C)

Figure 5.DTA thermograms of GSPC specimens.

4.5 Differential Scanning Calorimetry (DSC) Test

Differential scanning calorimetry (DSC) was performed using a DSC Q20 V24.4
(Universal V4.5A TA) instrument to determine the melting point of composites. The
tests were performed from 27 to 400oC at a heating rate of 25oC/min in nitrogen
atmosphere of 100 mL/min flow rate. The melting point temperature of composite was
calculated from the main peak of endothermic curve plotted by DSC. The DSC
thermographs of the samples were shown in Figure 6. As seen from Table 10, melting
temperature of the GSPC materials is found to be around 112.74162.12C. It is also
observed that, the melting temperature is the lowest for S3 composite (particle size of
0.5 mm, 60-wt% of filler material with polyester resin) exhibitiing112.74 oC and the
highest temperature of 162.12 oC for S9 composite (particle size of 1.5 mm, 60-wt% of
filler material with vinyl ester resin).

450

International Journal of Emerging Sciences 2(3), 433-454, September 2012

0.0

S1
S2
S3
S4
S5
S6
S7
S8
S9

-0.4

Heat flow (mW)

-0.8
-1.2
-1.6
-2.0
-2.4
-2.8
-3.2
25

50

75

100

125

150

175

200

225

250

Temperature ( C)

Figure 6. DSC curves of GSPC specimens.

5 CONCLUSIONS
The present study highlights the application of Taguchi methodology to determine the
best combination of process parameters for optimizing the thermal properties such as
thermal conductivity, linear thermal expansion and specific heat of groundnut shell
particles reinforced polymer composite materials. The composite specimens were
prepared with different weight % of randomly distributed groundnut shell particles in
polymer matrix. The experiments were planned as per Taguchi L9orthogonal array
layout plan with particle size, % filler material and matrix material as the process
parameters. The optimal conditions were identified using ANOM and the contribution
of each process parameter in controlling the thermal properties was determined by
ANOVA. From the analysis of results using S/N ratio and ANOVA, the following
conclusions are drawn from the current investigation:
1. The ANOM results point out that the combination of higher particle size (1.5 mm)
with higher wt% of filler material (60%) with polyester resin as the matrix material
is beneficial for minimizing the thermal conductivity of groundnut shell particles
reinforced polymer composite materials. The best combination of process parameter
settings for minimal thermal expansion is particle size of 1.5 mm with 60-wt% of
particles using polyester resin as the matrix material. On other hand, the optimal
combination of process parameter settings for maximizing the specific heat of

451

G. U. Raju, V. N. Gaitonde, S. Kumarappa

2.

3.
4.

5.

groundnut shell particles reinforced polymer composite is particle size of 1mm with
60-wt% of particles using polyester resin as the matrix material.
The ANOVA results revealed that the wt % of filler material has major contribution
in minimizing both the thermal conductivity and thermal expansion. On the other
hand, polyester resin matrix is the dominant factor, which maximizes the specific
heat of groundnut shell particles reinforced polymer composite material. The particle
size has also the noticeable effect on controlling the thermal expansion and specific
heat.
The confirmation results indicate that the additive models are adequate for
determining the optimum thermal properties at 95% confidence interval.
The groundnut shell as reinforcing material is an agricultural product; eco-friendly,
non-toxic, low cost and easily available material as compared to conventional fibers
like glass, kevlar, asbestos etc. Hence, this composite can be used as good alternate
for wood in thermal applications such as thermal insulation and coatings.
Thermal stability is valuable information required to manufacture more thermally
stable composites, possibly with good fire resistance. The composite specimen of 1.5
mm particle, 60-wt% of filler material with vinyl ester resin has the higher thermal
stability.

REFERENCES
1.

Mwaikambo. LY, Ansell MP, Chemical Modification of Hemp, Sisal, Jute and Kapok
Fibers by Alkalization, Journal of Applied Polymer Science 2002; 84(12): 2222-2234.

2.

Mishra. S, Tripathy. SS, Misra. M, Mohanty. AK, Nayak. SK, Novel Eco-Friendly
Biocomposites: Biofiber Reinforced Biodegradable Polyester Amide Composites
Fabrication and Properties Evaluation,Journal of Reinforced Plastics and Composites
2002;21(1): 55-70.

3.

Satyanarayana. KG, Sukumaran. K, Mukherjee. RS, Pavithran. C, Pillai. SGK, Natural

Fibre-Polymer Composites,Journal of Cement and Concrete Composites 1990;12: 117-136.

4.

Bledzki. AK, Gassan. J, Composites reinforced with cellulose based fibres, Progress in
Polymer Science1999;24: 221-274.

5.

Hirao. K, Inagaki. H, Nakamae. K, Kotera. M, Nishino. TK, Kenaf Reinforced

Biodegradable Composite,Journal of Composites Science and Technology 2003;63: 12811286.

6.

Nor Azowa Ibrahim, KamarulArifinHadithon, KhalinaAbdan, Effect of Fiber Treatment

on Mechanical Properties of Kenaf Fiber Ecoflex Composites,Journal of Reinforced
Plastics and Composites2010;29: 2192-2198.

7.

Hinrichsen. G, Khan. MA, Mohanty. AK, Influence of Chemical Surface Modification on

the Properties of Biodegradable Jute Fabrics-Polyester Amide Composites,Journal of
Composites: Part A2000; 31(2): 143-150.

452

International Journal of Emerging Sciences 2(3), 433-454, September 2012

8.

AlokSatapathy, Alok Kumar Jha, SisirMantry, Singh. SK, AmarPatnaik, Processing and
Characterization of Jute-Epoxy Composites Reinforced with SiC Derived from Rice
Husk,Journal of Reinforced Plastics and Composites2010; 29 (18): 2869-2878.

9.

Joseph. PV, Kuruvilla. J, Thomas. S, Effect of Processing Variables on the Mechanical

Properties of Sisal-Fiber-Reinforced Polypropylene Composites, Journal ofComposites
Science and Technology 1999; 59(11): 1625-1640.

10. Bakare. IO, Okieimen. FE, Pavithran. C, Abdul Khalil. HPS, Brahmakumar. M,
Mechanical and thermal properties of sisal fiber-reinforced rubber seed oil-based
polyurethane composites, Materials and Design 2010; 31: 4274-4280.
11. Dominguez. VA, Kenny. JM, Vazquez. A, Bagasse Fiber-Polypropylene Based
Composites, Journal of Thermoplastic Composite Materials 1999; 12(6): 477-497.
12. SmitaMohanty, Sanjay K. Nayak, Short Bamboo Fiber-reinforced HDPE Composites:
Influence of Fiber Content and Modification on Strength of the Composite, Journal of
Reinforced Plastics and Composites 2010; 29 (14): 2199-2212.
13. Mukherjee. PS, Satyanarayana. KG, Structure and Properties of Some Vegetable Fibres Part 2 Pineapple Fibre, Journal of Materials Science 1986; 21: 51-56.
14. Kim. HS, Yang. HS, Kim. HJ, Park. HJ, Thermogravimetric Analysis of Rice Husk Flour
Filled Thermoplastic Polymer Composites,Journal of Thermal Analysis and Calorimetry
2004; 76: 395-404.
15. Raju. GU, Kumarappa. S, Experimental Study on Mechanical Properties of Groundnut
Shell Particle Reinforced Epoxy Composites, Journal of Reinforced Plastics and
Composites 2011; 30 (12): 1029-1037.
16. MariesIdicula, AbderrahimBoudenne. L, Umadevi, Laurent Ibos, Yves Candau, Sabu
Thomas, Thermophysical properties of natural fiber reinforced polyester composites,
Composites Science and Technology 2006; 66 (15): 2719-2725.
17. Behzad. T, Sain. M, Measurement and prediction of thermal conductivity for hemp fiber
reinforced composites, Journal of Polymer Engineering and Science 2007; 47(7):977-983.
18. Xue. Li, Lope. G Tabil, Ikechukwuka. N Oguocha, SatyanarayanPanigrahi, Thermal
diffusivity, thermal conductivity, and specific heat of flax fiberHDPE biocomposites at
processing temperatures, Journal of Composites Science and Technology2008; 68 (7):
1753-1758.
19. Agrawal. R, Saxena. NS, Sreekala. MS, Thomas. S, Effect of treatment on the thermal
conductivity and thermal diffusivity of oil-palm-fiber-reinforced phenol formaldehyde
composites, Journal of Polymer Science Part B: Polymer Physics 2000; 38(7): 916-921.
20. Agarwal. R, Saxena. NS, Sharma. KB, Thomas. S, Pothan. LA, Thermal conduction and
diffusion through glass-banana fiber polyester composites Indian Journal of Pure Applied
Physics 2003; 41(6): 448-452.
21. Alsina. OLS, de Carvalho. LH, Ramos Filho. FG, d Almeida. JRM, Thermal properties of
hybrid lignocellulosic fabric-reinforced polyester matrix composites, Journal of Polymer
Testing 2005;24: 81-85.

453

G. U. Raju, V. N. Gaitonde, S. Kumarappa

22. Mohanty. AK, Misra. M, Drzal. LT, Surface modifications of natural fibers and
performance of the resulting biocomposites: an overview, Composite Interfaces 2001; 313343.
23. Abdelmouleh. M, Boufi. S, Belgacem. MN, Dufresne. A, Short natural-fiber reinforced
polyethylene and natural rubber composites: Effect of silane coupling agents and fibers
loading, Journal of Composites Science and Technology 2007; 67: 1627-1639.
24. MominulHaque. Md, Saiful Islam. Md, Sakinul Islam. Md, Nazrul Islam. Md,
MonimulHuque. Md,
Physicomechanical properties of chemically treated palm
fiber reinforced polypropylene composites, Journal of Reinforced Plastics and Composites
2009; 29: 1734-1742.
25. Swamy. RP, Mohan kumar. GC, Study of areca-reinforced phenol formaldehyde
composites, Journal of Reinforced Plastics and Composites 2004;23(13):1373-1382.
26. Mohanty. AK, Misra. M, Hinrichsen. G, Biofibres, biodegradable polymers and
biocomposites: an overview, Journal of Macromolecular Materials and Engineering 2000;
276: 1-24.
27. John. Z Lu, Qinglin. Wu, Ioan. I, Negulescu. Yan Chen, The influences of fiber feature and
polymer melt index on mechanical properties of sugarcane fiber/polymer composites,
Journal of Applied Polymer Science 2006; 102: 5607-5619.
28. Raveendran. K, Anuradda Ganesh. Kartic. C, Khilart, Influence of mineral matter on
biomass pyrolysis characteristics, Journal of Fuel 1995; 74 (12): 1812-1822.
29. Phadke. MS, Quality Engineering using Robust Design,Prentice Hall, Englewood Cliffs,
New Jersey, 1989.
30. Ross. PJ, Taguchi Techniques for Quality Engineering, McGraw-Hill, New York, 1996.
31. Ramanaiah. K, Ratna Prasad. AV, Hema Chandra Reddy. K, Thermal and mechanical
properties of waste grass broom fiber-reinforced polyester composites, Materials and
Design 2012; 40: 103-108.
32. Sharifah. HA, Martin. PA, The effect of alkalization and fiber alignment on the mechanical
and thermal properties of kenaf and hemp bastfibre composites: part 2 - cashew nut shell
liquid matrix, CompositeScience and Technology 2004; 64: 1231-1238.
33. Igor Maria De Rosa, Carlo Santulli, FabrizioSarasini, Mechanical and thermal
characterization of epoxy composites reinforced with random and quasi-unidirectional
untreated Phormiumtenax leaf fibers, Materials and Design2010; 31: 2397-2405.
34. Kim. HS, Yang. HS, Kim. HJ, Park. HJ, Thermogravimetric Analysis of Husk Flour Filled
Thermoplastic Polymer Composites, Journal of Thermal Analysis and Calorimetry 2004;
76: 395-404.

35. Dobircau. L, Sreekumar. PA, Saiah. R, Leblanc. N, Terrie. C, Gattin. R, Saiter. JM, Wheat
flour thermoplastic matrix reinforced by waste cotton fibre: Agro-green-composites,
Journal of Composites: Part A 2009; 40: 329-334.

454