Nanocomposites based on silica-reinforced ethylene–propylene–diene–monomer/acrylonitrile–butadiene rubber blends

Abstract

Rheometric characteristics, curing kinetics, mechanical properties before and after thermal aging and morphology of nanocomposites based on various network precursors (i.e., acrylonitrile–butadiene rubber (NBR), ethylene–propilene–diene monomer (EPDM) and its blend (NBR/EPDM) reinforced of nanosilica) is presented here. The ratios of EPDM and NBR as binary blend system vary significantly. The rheometric characteristics and curing kinetics of nanocomposites were determined using a rheometer with an oscillating disk, in which the network formation process was registered by the varying torque durations. The mechanical properties of the elastomeric composites were determined before and after thermal aging using an air-circulating oven. The specific interactions between the rubber and filler were characterized by Fourier transform infrared spectroscopy (FTIR). Scanning electron microscopy (SEM) was employed to study the surface morphology of fractured rubber.The obtained results demonstrated a correlation between the calculated activation energies of cross linking (Eac) and reversion (Ear) and mechanical properties, which can be seen in sample EPDM/NBR=20/80. For this blend, maximum tensile strength values and synergism were observed. Additionally, this blend exhibited a relatively co-continuous morphology, which was investigated by SEM. The differential scanning calorimetry (DSC) curves reported that the silica-reinforced EPDM/NBR rubber blends were immiscible. FTIR studies showed a strong interaction between the polymer matrix and filler, which was reflected in the peak shifts at 1441.9 and 1462.5cm−1 to higher wave numbers.

Full text

Nanocomposites based on silica-reinforced
ethylene–propylene–diene–monomer/acrylonitrile–butadiene rubber blends
Suzana Samarz
ˇija-Jovanovic
´
a,
⇑
, Vojislav Jovanovic
´
a
, Gordana Markovic
´
b
, Sandra Konstantinovic
´
c
,
Milena Marinovic
´-Cincovic
´
d
a
Faculty of Natural Science and Mathematics, University of Priština, Lole Ribara 29, 38220 Kosovska Mitrovica, Serbia
b
Tigar, Nikole Pašic
´a 213, 18300 Pirot, Serbia
c
Faculty of Technology, University of Niš, Bulevar oslobo -
denja 124, 16000 Leskovac, Serbia
d
Institute of Nuclear Science Vinc
ˇa, University of Belgrade, Mike Petrovic
´a Alasa 12-14, 11000 Belgrade, Serbia
article info
Article history:
Received 5 December 2010
Received in revised form 17 February 2011
Accepted 27 February 2011
Available online 5 March 2011
Keywords:
A. Nano-structures
B. Mechanical properties
D. Electron microscopy
Rheometric characteristics
abstract
Rheometric characteristics, curing kinetics, mechanical properties before and after thermal aging and
morphology of nanocomposites based on various network precursors (i.e., acrylonitrile–butadiene rubber
(NBR), ethylene–propilene–diene monomer (EPDM) and its blend (NBR/EPDM) reinforced of nanosilica) is
presented here. The ratios of EPDM and NBR as binary blend system vary significantly. The rheometric
characteristics and curing kinetics of nanocomposites were determined using a rheometer with an oscil-
lating disk, in which the network formation process was registered by the varying torque durations. The
mechanical properties of the elastomeric composites were determined before and after thermal aging
using an air-circulating oven. The specific interactions between the rubber and filler were characterized
by Fourier transform infrared spectroscopy (FTIR). Scanning electron microscopy (SEM) was employed to
study the surface morphology of fractured rubber.
The obtained results demonstrated a correlation between the calculated activation energies of cross
linking (Eac) and reversion (Ear) and mechanical properties, which can be seen in sample EPDM/
NBR = 20/80. For this blend, maximum tensile strength values and synergism were observed. Addition-
ally, this blend exhibited a relatively co-continuous morphology, which was investigated by SEM. The
differential scanning calorimetry (DSC) curves reported that the silica-reinforced EPDM/NBR rubber
blends were immiscible. FTIR studies showed a strong interaction between the polymer matrix and filler,
which was reflected in the peak shifts at 1441.9 and 1462.5 cm
1
to higher wave numbers.
Ó2011 Elsevier Ltd. All rights reserved.
1. Introduction
In recent years, the production of nanocomposites based on
polymer blends has markedly increased due to their well-balanced
physical and mechanical properties, facile processability and rela-
tively low production cost [1].
Ethylene–propylene–diene monomer (EPDM) is a saturated,
nonpolar rubber (i.e., very low AC@CAcontent). EPDM exhibits
several properties, including balanced heat stability, aging resis-
tance, elasticity (especially at very low temperatures) and water
resistance; therefore, EPDM is widely applied in many rubber
products (e.g., in styrene–butadiene and butadiene rubbers and
as a substitute for natural rubber). Unfortunately, the application
of EPDM is restricted due to its poor solvent resistance and
adhesion properties. However, blending EPDM with acrylonitrile–
butadiene rubber (NBR) can improve the aforementioned
disadvantages of EPDM because polar NBR exhibits excellent sol-
vent resistance and adhesion properties. A number of reports dem-
onstrating the utility of EPDM/NBR blends have been previously
published [2–5]. Additionally, the morphology of the polymer/
polymer blend is dependent on the blend ratio, viscosity, surface
property of each component and mixing process [6,7].
Investigators have studied the solubility properties of water in
elastomers based on EPDM/NBR for use as packing materials, bio-
medical devices and materials for marine applications [8]. Further-
more, EPDM, halobutyl and NBR rubbers have been compounded
for a variety of automotive applications [9]. The effect of vulcaniz-
ing systems on EPDM and NBR properties has also been studied
[10,11]. Elastomers that do not contain a filler compound are prac-
tically insignificant in any application; therefore, in practice, an
elastomer is commonly combined with fillers to form a mixture,
whose filler content is usually approximately 30–50%.
The fillers generate stronger elastomers and are of great impor-
tance from a practical point of view. Organic and inorganic fillers,
such as carbon black and silica, are often added to the elastomers.
1359-8368/$ - see front matter Ó2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compositesb.2011.02.008
⇑
Corresponding author. Tel.: +381 28 425 396.
E-mail address: vojani@sbb.rs (S. Samarz
ˇija-Jovanovic
´).
Composites: Part B 42 (2011) 1244–1250
Contents lists available at ScienceDirect
Composites: Part B
journal homepage: www.elsevier.com/locate/compositesb

Usually, fillers are mechanically introduced into a polymer via
milling [12–14]. This milling is a slow, energy-intensive process
that may cause chemical degradation of the polymer chains. More-
over, undesirable particulate agglomerates often remain even after
aggressive milling. Small, separated particles are ideal for reinforc-
ing a polymer matrix. Therefore, it is desirable to explore alterna-
tive methods of incorporating inorganic fillers into elastomer
mixtures.
Herein, different nanocomposites based on silica-reinforced
EPDM, NBR and their EPDM/NBR rubber blends are described. The
rheometric characteristics of these materials, such as scorch safety
and cure rate were analyzed. The mechanical properties were also
investigated as a function of blend ratio before and after thermal
aging. Scanning electron microscopy (SEM) and differential
scanning calorimetry (DSC) were conducted to detect the polymer
compatibility within the rubber blends. Additionally, interactions
between the polymer matrix and silica fillers were investigated by
Fourier transform infrared-attenuated total reflection (FTIR–ATR).
2. Experimental
2.1. Materials
Ethylene–propylene–diene monomer (EPDM, Vistalon 2504)
was obtained from Exxon Mobil (Notre Dame, France). Acryloni-
trile–butadiene rubber (NBR, Krynac 3550) was obtained from
Lanxes, France. The properties and specifications of the rubbers
and filler are reported in Table 1. Commercial grade sulfur, zinc
oxide, stearic acid, accelerator (CBS, TMTD) and antioxidant were
used in this study. Silica-reinforced nanocomposites based on
EPDM and NBR rubbers and EPDM/NBR rubber blends were pre-
pared in seven different combinations. Sulfur was used as a vulca-
nizing agent. The sample designations of the silica-reinforced
nanocomposites are given in Table 2.
2.2. Techniques
2.2.1. Mixing, vulcanization and testing of the nanocomposites
Formulation of the nanocomposites is given in Table 2. The mix-
tures were sheeted out in a laboratory-sized, two-roll mixing mill
that had a friction ratio of 1:1.4. The rheometric characteristics
were assessed by a Monsanto Oscillating Disc Rheometer R-100,
according to the ASTM D2084-95 standard testing method. The
optimum curing time (t
c90
) was determined at 160 °C. The com-
pounded blends were molded using an electrically heated hydrau-
lic press (Indexpell, Kerala, India) under a pressure of 5 MPa and
their optimum curing time. These cured sheets were conditioned
before testing (24 h maturation at room temperature).
2.2.2. Curing kinetics
The kinetic parameters for the crosslinking process, such as
apparent activation energies of crosslinking (Eac) and reversion
(Ear) were calculated from the torque-time curves. The torque
and time experiments were performed using an accelerated sulfur
curing system with an oscillating disk rheometer (Monsanto Rhe-
ometer model 100C) at two temperatures: 180 and 190 °C.
2.2.3. Mechanical properties
Mechanical properties, such as tensile strength, modulus (%)
and elongation at break were measured with a Zwick-1425 tensile
tester according to the ASTM D412-98 standard testing method
using a crosshead speed of 500 mm/min and at 25 ± 2 °C. The ten-
sile properties of the blends were examined according to the ASTM
D-412 standard testing method. The tear test was conducted, as
per the standard ASTM D-624 testing method using a 908 angle
test pieces. The experimental conditions and equipment for the
tear testing were the same as that of the tensile testing. Five sam-
ples from each formulation were tested. The hardness of the sam-
ples was measured, as per the standard ASTM D-2240 testing
method. For hardness measurements, the sheets having an effec-
tive thickness of 6 mm were used. At least five measurements were
recorded, and the average values were reported. The rebound resil-
ience measurements were performed using a Schob pendulum, fol-
lowing the standard ASTM D7121-05 testing method. The abrasion
resistance was determined as the relative volume loss from the
wear test using a Zwick DIN Abrader, as per the ASTM D-2228 stan-
dard testing method, and the rolling sliding test was employed. In
this study, a 0.1 kN load was applied to the rolling sliding cylinder
on SiC paper.
2.2.4. Thermal aging
The thermal aging of the silica-reinforced nanocomposites
based on EPDM and NBR rubbers and EPDM/NBR rubber blends
were performed in an air-circulating oven operated at 100 °C for
Table 1
Properties and specifications of EPDM, NBR and SIL-1.
Product name Vistalon 2504 (EPDM)
Mooney viscosity, ML1 + 4/125 °C 26.00
E/P
a
, weight ratio 50/45
Molecular distribution Broad
Specific gravity (g/cm
3
) 0.86
Volatile matter, weight (%) 0.5
Ash, weight (%) 0.3
Color Light gray to light amber
Product name Krynac 3550 (NBR)
Moooney viscosity, ML1 + 4/100 °C 50.00
Bound ACN
b
, weight (%) 35.00
Volatile matter, weight (%) 0.50
Ash, weight (%) 0.50
Specific gravity (g/cm
3
) 0.98
Color Light tan
Product name SIL-1 (SiO
2
)
SiO
2
content (mass%) >90
Al
2
O
3
content (mass%) >0.5
(Cu) content (ppm) <10.0
(Mn) content (ppm) <10.0
Na
2
O content (mass%) <1.5
Specific gravity (g/cm
3
) 1.7
BET (m
2
g
1
) 200
pH (5% paste) 6–8
Primary particle size (nm) 9–16
DBP factor (cm
3
/100 g) 200
The rest (mass%) <0.5
Bulk density (kg/m
3
) 170
Appear Amorphous white dust
a
E/P, ethylene/propylene.
b
ACN, acrylonitrile.
Table 2
Formulation of the silica-reinforced nanocomposites based on EPDM and NBR rubbers
and EPDM/NBR rubber blends.
Components (phr
a
)
EPDM/NBR ZnO Stearic acid SIL-1 4010 NA
b
CBS
c
TMTD
c
Sulphur
100/0 5 1 50 1 1.5 0.5 0.8
80/20 5 1 50 1 1.5 0.5 0.8
60/40 5 1 50 1 1.5 0.5 0.8
50/50 5 1 50 1 1.5 0.5 0.8
40/60 5 1 50 1 1.5 0.5 0.8
20/80 5 1 50 1 1.5 0.5 0.8
0/100 5 1 50 1 1.5 0.5 0.8
a
Phr, parts per hundred.
b
Antioxidans (N-isopropyl-N-phenyl-p-phenylendiamine).
c
Accelerators (N-cyclohexyl-2-benzothiazole sulphenamide-CBS; tetramethyl-
thiuram disulfide-TMTD).
S. Samarz
ˇija-Jovanovic
´et al./ Composites: Part B 42 (2011) 1244–1250 1245

50 h duration. The retained percentage values of tensile strength,
elongation at break and hardness were calculated. The tensile
properties (tensile strength and elongation at break) and hardness
were measured before and after the aging studies.
2.2.5. DSC analysis
A DuPont 990 Thermal Analyzer and DuPont Differential Scan-
ning Calorimeter, model 910 were used for testing the sample
materials. DSC analysis of the samples was performed under a
nitrogen atmosphere using a flow of 50 cm
3
per min in tempera-
tures ranging from 110 °C to 110 °C (heating rate, 20 °C per
min). The examined sample weights ranged from 10 to 15 mg.
The glass transition temperature (T
g
) was determined from the pri-
mary point that intersected the tangent discontinuity in the DSC
data.
2.2.6. Morphology studies
The fractured surfaces of the blended materials were imaged by
scanning electron microscopy (SEM) using a JEOL JSM-5400 model
SEM. The samples were sputter coated with gold for 3 min under
high vacuum with image magnifications of 2000. From our preli-
minary experiments, we found that the fractured surface did not
change even after storing the specimen for 72 h (without gold
coating) prior to the SEM observations.
2.2.7. FTIR–ATR spectroscopy
The FTIR spectra were recorded using a Perkin Elmer spectrom-
eter equipped with an ATR attachment. A minimum of 500 scans
were signal-averaged at a resolution of 4 cm
1
. For FTIR–ATR mea-
surements, the spectrometer was equipped with a liquid nitrogen
cooled mercury cadmium telluride (MCT) detector. The selected
internal reflection element (IRE) was a 45°KRS-5. The samples un-
der investigation were approximately 0.3 mm thick sheets, which
were prepared by compression molding the sample between two
Teflon films at 163 °C.
3. Results and discussion
3.1. Rheometric characteristics
Fig. 1 presents the torque–cure time plots generated from the
silica-reinforced nanocomposites based on EPDM and NBR rubbers
and their EPDM/NBR rubber blends. The initial decrease in torque
is due to the softening of the matrix. The torque then increased
due to the crosslinking between the macromolecular chains. It
can be seen from Fig. 1 that as the percentage of NBR increased
in the blend systems, the torque also increased.
Table 3 also shows the rheometric characteristics of the EPDM/
NBR rubber blends under investigation. Regular variations in the
maximum torque (M
h
) and optimum curing time (t
c90
) were ob-
served for the reinforced blends. The EPDM/NBR (100/0) system
demonstrated an optimal curing time. This curing time also de-
creased with increasing NBR content in the blends. The t
c90
and
rate of vulcanization (CRI) values were lowest for the blend with
a higher NBR percentage. The scorch time (t
s2
) is the time taken
to reach the minimum torque (M
l
) value. T
s2
increased by two units
and measured the premature vulcanization of the material. The
maximum scorch time and scorch safety was observed for the
EPDM/NBR (0/100) blend system.
Generally, acidic compounds retard the vulcanization of elasto-
mers. For this reason, precipitated silica, which contains a large
number of acidic silanol (Si–OH) groups, is not generally incorpo-
rated without an activator in EPDM. The surface of silica is acidic
and therefore, a strong hydrogen bond with polymer groups is
formed. NBR rubber possesses nitrile groups (ACN), which are
basic, resulting in the hydrogen bonding of the silanol group of sil-
ica and a strong silica/NBR interaction. Silica particles also adsorb
polar curative molecules on their surfaces, rendering the deactiva-
tion of vulcanization and thus, lower the curing state.
The maximum torque increased proportional to the NBR rubber
content, indicating crosslink density enhancement. The calculated
values for the curing rate also increased with the percentage of
NBR content. This indicates that NBR rubber is a curative activating
component for the EPDM/NBR rubber blends. This curative feature
is due to the presence of unsaturation in the butadiene component
of NBR. Crosslinks (sulfuric linkages) are formed between the
unsaturated sites and macromolecules during vulcanization. As
the number of NBR molecules (weight percentage) increases, the
number of active sites for vulcanization also increases.
3.2. Curing kinetics
The method for calculating the Eac and Ear is described in our
earlier works [15–17].
During the vulcanization process, sulfur crosslinks are formed
between the rubber polymer chains (crosslinking), whereas some
of the links decay (reversed). Reversion tendency was only
observed at higher temperatures. At 160 °C, only the blends formu-
lated with a higher amount of NBR rubber (>40 phr) demonstrated
a reversion tendency. For the stability of the system, the Ear must
be greater than the Eac, which is determined by the rheometric
curve. That is, the curing process, which has lower activation
energy, occurs more readily and rapidly because the energy barrier
is lower.
The thresholds for these two reactions, Eac and Ear are
characteristic parameters of the curing properties of the rubber
compounds and can be used as criteria for the energy compatibility
Fig. 1. The relationship between torque and time during the curing of silica-
reinforced nanocomposites based on EPDM and NBR rubbers and EPDM/NBR rubber
blends.
Table 3
Rheometric characteristics of the silica-reinforced nanocomposites based on EPDM
and NBR rubbers and EPDM/NBR rubber blends.
EPDM/NBR (phr) M
h
(Nm) M
l
(Nm) t
S2
(s) t
C90
(s) CRI (s
1
)
100/0 6.65 2.88 96 1212 0.1
80/20 6.30 4.40 96 456 0.3
60/40 6.30 4.40 126 172 2.2
50/50 7.50 4.70 132 174 2.4
40/60 7.00 3.70 144 181 2.7
20/80 8.90 4.70 156 204 2.1
0/100 7.80 2.48 240 288 2.1
1246 S. Samarz
ˇija-Jovanovic
´et al. / Composites: Part B 42 (2011) 1244–1250

of several rubber compounds that comprise the product. The pre-
diction of the curing state is usually determined by rheometer, in
which the kinetics is described by the torque variation during vul-
canization [18,19].Table 4 lists the Eac,Ear and Ear/Eac ratio values
of the silica-reinforced EPDM and NBR rubbers and the EPDM/NBR
rubber blends.
The nanocomposites based on NBR rubber exhibit Eac values,
which are 2 to 3-fold less than that of the nanocomposites based
on EPDM/NBR rubber blends. This lower Eac may be explained by
the different nature of the rubbers. NBR is more polar than EPDM
due to the present nitrile groups (ACN). Thus, more crosslinking
reactions and faster curing processes occur with NBR [20]. The
driving force of the curative and filler diffusion from EPDM to the
more polar diene rubber phase that contains a high amount of
unsaturation is the saturated backbone of EPDM (less polar) [21].
Due to a slow curing nature of the EPDM elastomer (low unsatura-
tion content), curatives react and deplete more rapidly at the
unsaturation sites of the other diene rubber phases present in
the blend. A difference in elastomer polarity occurs in a concentra-
tion gradient for the fillers and curatives, which further leads to
the diffusion of the curatives from the less polar EPDM phase
to the more polar elastomer phase. Evidently, the EPDM phase in
the blend remains severely undercured, even after prolonged
vulcanization.
The Eac values of the EPDM/NBR rubber blends decrease with
increasing NBR rubber content that contain 40 phr (maximum
value of Eac, which is contrary to our expectations). EPDM rubber
exhibits good resistance to high temperatures and does not under-
go the reversion process. A small addition of NBR rubber (20 phr)
to the EPDM rubber matrix does not affect its reversion; Ear does
not occur (i.e., reversion does not register with the rheometric
curves). However, the Eac values are proportionally reduced.
When the content of NBR rubber is further increased, a greater
degree of unsaturation occurs, which contributes to reversion.
Simultaneously, a migration process of the crosslinking system,
as well as of the filler from the EPDM phase to the NBR phase oc-
curs. Therefore, the curing process proceeds slowly with higher Eac
values. Increasing the NBR content in the EPDM/NBR rubber blends
produces decreased Eac values; however, when the sample ratio
reaches 20/80 of EPDM/NBR, the Eac value again increase. When
this maximum Eac is observed, a minimum value for the Ear is re-
corded, which indicates that the reversion process can easily occur.
As such, the relationship between Eac, Ear and the precursors of the
network formation is evident.
The Eac and Ear serve as the criteria for the energy compatibility
of the examined compounds. A compound with the smallest possi-
ble Eac and higher values for Ear and their Ear/Eac has a tendency
to retain the basic physical and mechanical characteristics.
The lowest Eac value (35.4 kJ mol
1
) with a maximum Ear/Eac
ratio (8.40) was found for the EPDM/NBR rubber blends with a
20/80 ratio.
3.3. Mechanical properties
The nature of rubber deformation under an applied load can be
understood from its mechanical properties. The tensile strength
varied with the rubber blend ratio variation. When NBR rubber is
dispersed within the EPDM matrix, the dispersion domain becomes
larger and has a very wide size distribution. A weak interphase
rubber adhesion results in poor mechanical properties. Thus, the
EPDM and NBR rubbers exhibited different vulcanization rates.
Tensile strength is a complex function consisting of the nature
and type of crosslinks, crosslink densities and chemical structure
of the used rubber. It is well known that if rubber is deformed by
an external force, part of the input energy is stored elastically in
the chains and is available (released upon crack growth) as a driv-
ing force for fracturing. The remaining energy is dissipated through
molecular motions by heat; and as such, is made unavailable to
break the chains. At higher crosslinking levels, chain motions be-
come restricted, and the dense network is incapable of dissipating
as much energy. This results in a relatively facile brittle fracture at
low elongation.
As the NBR rubber content increases, the tensile strength in-
creases to its optimal value and then decreases (Table 5). The ten-
sile strength showed a maximum with the EPDM/NBR = 20/80
rubber blend. When the EPDM content is above 20 phr, the tensile
strength decreases. The elongation at break maximum was ob-
served with the same blend. A maximum modulus value was found
with the EPDM/NBR = 50/50 blend system. When the NBR rubber
content was increased, the hardness values decreased. The abra-
sion resistance increased with increasing NBR rubber content.
The percentage of rebound resilience for the different EPDM/NBR
blends afforded average values with low variance between the sil-
ica-reinforced nanocomposites based on EPDM and NBR rubbers.
The mechanical properties of the silica-reinforced EPDM/
NBR = 20/80 rubber blend were improved, and chemical bonds be-
tween phases formed during the networking process. In addition to
the networking process, crosslinking between the EPDM and NBR
rubber via sulfide bonds was possible during the co-crosslinking
process (Scheme 1).
3.4. Thermal aging
Elastomers that are based on more network precursors are
widely used in rubber products for a number of advantages, includ-
ing improved physical properties and service life, easier processing
and reduced product costs [22–24]. An important characteristic of
any elastomer for its potential application is thermal stability. The
thermal stability of a material is defined by the specific tempera-
ture or temperature–time limit in which the material can be used
without excessive property loss.
To investigate the influence of thermal aging on the mechanical
properties of the rubbers, the crosslinking reactions were per-
formed in an air-circulating oven operated at 100 °C for 50 h. The
retained tensile strength percentage and elongation at break values
were then calculated before and after aging. Based on the results
(Table 6), the tensile strength increased. The sulfur vulcanization
of the unsaturated rubbers occurs through complicated radical
substitution, forming mono-, di- or poly-sulfide bridges and sul-
fur-induced intracyclization of the polymer molecules. However,
at higher temperatures, the sulfur crosslinks are less effective and
the physical properties are weaker due to the dissociation of sulfur
bonds and rubber chains. A change in the hardness values increased
with increasing NBR rubber loading, which can be attributed to the
increased crosslinking density after thermal aging. This is can be
explained by the sulfur networking addition process of the rubbers
and the polysulfide crosslink density reduction process. The poly-
sulfide reacts further to form mono-, di- and cyclic-sulfide bonds
Table 4
Activation energies of crosslinking (Eac) and reversion (Ear) and Ear/Eac ratios of the
silica-reinforced nanocomposites based on EPDM and NBR rubbers and EPDM/NBR
rubber blends.
EPDM/NBR (phr) Energy activation (kJ mol
1
)Ear/Eac
Eac Ear
100/0 57.4 – –
80/20
a
45.1 – –
60/40 85.1 267.5 3.14
50/50 39.3 206.8 5.26
40/60 35.4 297.5 2.73
20/80 68.0 185.6 8.40
0/100 26.9 411.3 15.29
a
Sample EPDM/NBR = 80/20 do not shown of reversion.
S. Samarz
ˇija-Jovanovic
´et al. / Composites: Part B 42 (2011) 1244–1250 1247

during vulcanization via the dissociation, recombination and rear-
rangement of the sulfur linkages. The elongation at break changes
decreases. After thermal aging, minimal changes of the mechanical
properties of the nanocomposites were observed.
3.5. DSC studies
The T
g
of the polymer depends on the structure and cooperative
mobility of its segments. This behavior is reflected in a single T
g
for
miscible blends. In the case of partially miscible blends, the T
g
of
the blend components remain separated but are shifted toward
each other, as compared with pure components. In completely
immiscible polymer blends, the T
g
values of all components remain
largely unaltered. Therefore, DSC can be considered as a practical
method for investigating the miscibility of polymer blends.
DSC curves of the silica-reinforced nanocomposites based on
EPDM and NBR rubbers and their EPDM/NBR rubber blends are
shown in Fig. 2. Two separate T
g
values of the silica-reinforced
nanocomposites based on EPDM/NBR rubber blends can be ob-
served (Fig. 3). The first T
g
(T
g1
) results from the EPDM rubber with
a temperature range of 60 °Cto52 °C. The second T
g
(T
g2
) is pro-
duced by the NBR rubber with a temperature range of 30 °Cto
23 °C. Therefore, as indicated by the DSC curves of the silica-rein-
forced EPDM/NBR rubber blends, the EPDM/NBR polymer blends
are immiscible.
3.6. Morphology analysis by SEM
Morphology is a major factor of rubber blends, which can deter-
mine the extent to which the blends are compatible. It is well
known that the phase structure of the blend is influenced by sev-
eral factors, including the surface characteristics, blend ratio, vis-
cosity of each component and compounding process. The
primary factor that determines the final morphology of the mixes
is their composition. NBR, with strong molecular polarity has high-
er surface tension than that of EPDM, resulting in their incompat-
ibility. In the EPDM/NBR = 80/20 rubber blend, NBR, with higher
viscosity is inclined to form the dispersion phase so that it can
be dispersed as round particles with maximum surface energy.
As the NBR content increases, it becomes more difficult for NBR
to be dispersed in the EPDM rubber [25]. Particle surfaces of silica
possess hydrophilic silanol (or hydroxyl) groups, which results in a
strong filler/filler interaction via hydrogen bonding. Silica forms a
strong hydrogen bond with basic materials because of its acidic
surface. NBR possesses a nitrile group (ACN), and because the ni-
trile group is basic, it will hydrogen bond with the silanol group
of silica, resulting in a strong silica/NBR interaction.
These types of rubbers were chosen partly because of their dif-
ferences in relative polarity and solubility, as well as the ease of
distinguishing them under an electron microscope due to their
electron density contrast. Fig. 4a–c presents SEM micrographs of
EPDM, NBR, and the EPDM/NBR = 20/80 rubber composites. As
shown, the component dispersion nature in the blends was not
uniform, which is indicative of their heterogeneous nature. The
changes in the domain sizes of the dispersed phase are due to
the volume fraction and viscosity differences of the components.
These results further demonstrate that the interphase adhesion
between NBR and EPDM is very weak. Aside from their poor
Table 5
Mechanical properties of the silica-reinforced nanocomposites based on EPDM and NBR rubbers and EPDM/NBR rubber blends.
EPDM/NBR (phr) Modulus at 200%
elongation (MPa)
Modulus at 300%
elongation (MPa)
Tensile strength
(MPa)
Elongation at
break (%)
Hardness
(Sh
o
A)
Rebound
resilience (%)
Abrasion (mm
3
)
100/0 1.86 1.96 3.04 850 74 26 265
80/20 2.84 2.84 2.84 150 75 41 235
60/40 3.34 3.83 4.02 305 75 37 367
50/50 2.94 3.92 5.7 510 74 34 328
40/60 2.65 3.53 7.88 680 73 31 226
20/80 2.11 3.43 11.1 850 71 24 184
0/100 1.96 2.35 16.2 940 69 29 193
Scheme 1. Schematic presentation the crosslinking of the polymer chains of EPDM
and NBR rubbers.
Table 6
Mechanical properties of the silica-reinforced nanocomposites based on EPDM and
NBR rubbers and EPDM/NBR rubber blends after thermal aging.
EPDM/NBR (phr) Tensile
strength (%)
Elongation
at break (%)
Hardness (Sh
o
A)
100/0 +66.8 30.6 0
80/20 +45 26.7 1
60/40 +17 47.5 2
50/50 +3.2 53 3
40/60 +5.7 42.6 4
20/80 +23.7 18.8 6
0/100 +9 12 4
Fig. 2. DSC curves of the silica-reinforced nanocomposites based on EPDM and NBR
rubbers and EPDM/NBR rubber blends.
1248 S. Samarz
ˇija-Jovanovic
´et al. / Composites: Part B 42 (2011) 1244–1250

compatibility, NBR and EPDM could not make good use of the large
surface area-to-volume ratio of nanoparticles to improve their
interphase adhesion.
Architectural homogeneity and the appearance of strong ridge-
lines indicate a better dispersion state with efficient interfacial
crosslinking. The compact nature of blend systems, as revealed
from the study of blend morphology is manifested in the signifi-
cant enhancement of the mechanical properties. Vertical steps
and stairs are more frequent and more prevalent in crosslinked
rubber blends with higher tensile strength values [26]. SEM was
employed to study the fractured nanocomposites based on the
EPDM/NBR rubber blends and confirmed that certain features of
the fracture can be correlated with the mechanical properties.
3.7. FTIR–ATR spectroscopy analysis
In the FTIR–ATR spectra of EPDM (Fig. 5), there are bands at
1456 and 1373 cm
1
, which originated from the deformation
vibrations of the ACH
2
group and the valention vibrations of the
ACH
3
in the propylene unit, respectively. The bands at 751 and
722 cm
1
present the deformation vibrations of (ACH
2
)nAgroups,
where nP5, in the ethylene sequences of the EPDM backbone
[27,28]. At 876 cm
1
, there is an absorption band, which originated
from the deformation vibration of the C@C groups in ethylidene
norbornene-ENB (EPDM rubber). The two intense absorption bands
at 2915 and 2845 cm
1
result from asymmetric and symmetric
CAH valention vibrations (aliphatic CH
2
group).
In the FTIR–ATR spectra of NBR rubber, there are two intense
bands at 2915 and 2843 cm
1
, which originated from the asym-
metric and symmetric CAH valention vibrations (CH
2
group). A
medium intensity absorption band at 2238 cm
1
originated from
the ACNvalention vibration. At 1433 cm
1
, the absorption band
originated from the C-C deformation vibration (ACH
2
ACH@). Three
absorption bands at 1114, 1240 and 1267 cm
1
were also observed
and assigned to the accelerator reaction product (CBS, TMTD) and
sulfur. These bands disappear in the spectra of the silica-reinforced
nanocomposites based on EPDM/NBR rubber blend [29]. Addition-
ally, a very strong absorption band at 966 cm
1
originated from the
trans-1,4-disupstitutes AHC@CHAstretching vibrations [30,31].
In the FTIR–ATR spectra of the nanocomposites based on the sil-
ica-reinforced EPDM/NBR rubber blends, a broad, low intensity
band was assigned at 3303 cm
1
, and this stretch contributed to
the hydrogen-related vibrations of water. At 2915 and 2843 cm
1
,
two absorption bands that originated from the CAH asymmetric
and symmetric valention vibrations were seen (CH
2
aliphatic
group). Two absorption bands at 1648 cm
1
and 1542 cm
1
from
the asymmetric and symmetric C@N stretching vibration were also
assigned (during crosslinking).
In the FTIR–ATR spectra of the nanocomposites based on the sil-
ica-reinforced EPDM/NBR rubber blends, an absorption band that
originated from the ACNvalention vibration shifted to a higher
value (2349 cm
1
). This shift results in the hydrogen bonding be-
tween the silanol groups. Absorption bands at 1441.9 and
1462.5 cm
1
(ACH
2
Adeformation vibrations) and 1048.9 and
1058.8 cm
1
(three forms of silicon dioxide) also shifted to higher
Fig. 3. Glass transition temperature variations of the silica-reinforced nanocom-
posites based on EPDM and NBR rubbers and EPDM/NBR rubber blends, measured
by DSC.
Fig. 4. SEM micrographs of (a) EPDM and (b) NBR rubbers without filler and (c)
EPDM/NBR rubber blend with 50 phr SIL-1.
S. Samarz
ˇija-Jovanovic
´et al. / Composites: Part B 42 (2011) 1244–1250 1249

values as a result of the strong interaction between the polymer
matrix and the SiO
2
fillers.
4. Conclusion
In this study, different nanocomposites based on the silica-
reinforced ethylene–propylene–diene monomer (EPDM), acryloni-
trile–butadiene rubber (NBR) and their EPDM/NBR rubber blends
are reported. The effect of the rubber content on the curing charac-
teristics, mechanical properties before and after thermal aging,
glass transition temperature and morphology of the silica-rein-
forced nanocomposites were studied. The rheometric characteris-
tics indicated that the optimal curing time decreased with
increasing NBR rubber content in the EPDM/NBR rubber blend.
The scorching time and maximum torque also increased, and a
maximum scorch safety was produced by the EPDM/NBR (20/80)
rubber blend. The activation energies of crosslinking (Eac) and
reversion (Ear) can be used as criteria for the curing and mechan-
ical properties, compatibility and processability of the silica-
reinforced EPDM/NBR rubber blends. SEM analysis showed that
the nanocomposites based on the EPDM/NBR rubber blends exhib-
ited a heterogeneous nature. A tensile strength maximum was
observed with the EPDM/NBR = 20/80 rubber blend. The percent-
age of rebound resilience decreased with increasing NBR rubber
content. DSC analysis concluded that the components in the sil-
ica-reinforced EPDM/NBR rubber blends were immiscible due to
the two separate glass transition temperatures. The absorption
bands of the FTIR–ATR spectra of the silica-reinforced EPDM/NBR
rubber blends at 1441.9 and at 1462.5 cm
1
(ACH
2
Adeformation
vibrations) and 1048.9 and at 1058.8 cm
1
(three forms of silica)
shifted to higher values as a result of the strong interaction
between the polymer matrix and the silica fillers.
Acknowledgement
The authors acknowledge the support from the Ministry of
Science of the Republic of Serbia (Project Numbers 45022 and
45020).
References
[1] Findik F, Yilmaz R, Koksal T. Investigation of mechanical and physical
properties of several industrial rubbers. Mater Des 2004;25:269–76.
[2] Sau K, Chaki TK, Khastgir D. Carbon fibre filled conductive composites based on
nitrile rubber (NBR), ethylene propylene diene rubber (EPDM) and their blend.
Polymer 1998;39:6461–71.
[3] Grigoryeva OP, Karger-Kocsis J. Melt grafting of maleic anhydride onto an
ethylene–propylene–diene terpolymer (EPDM). Eur Polym J 2000;36:1419–29.
[4] Papke N, Karger-Kocsis J. Thermoplastic elastomers based on compatibilized
poly(ethylene terephthalate) blends: effect of rubber type and dynamic curing.
Polymer 2001;42:1109–20.
[5] Wu D, Wang X, Jin R. Toughening of poly(2,6-dimethyl-1,4-phenylene oxide)/
nylon 6 alloys with functionalized elastomers via reactive compatibilization:
morphology, mechanical properties, and rheology. Eur Polym J
2004;40:1223–32.
[6] Kang TK, Kim Y, Lee WK, Park HD, Cho WJ, Ha CS. Properties of
uncompatibilized and compatibilized poly(butylene terephthalate)–LLDPE
blends. J Appl Polym Sci 1999;72:989–97.
[7] Pukanszky B, Tudos F, Kolarik J, Lednicky F. Ternary composites of
polypropylene, elastomer, and filler – analysis of phase-structure formation.
Polym Compos 1990;11:98–104.
[8] Aminabhavi TM, Manjeshwar LS, Cassidy PE. Water permeation through
elastomer laminates. IV. NBR/EPDM. J Appl Polym Sci 1986;32(2):3719–23.
[9] Vara R, Dunn JR. Developments in fuel hoses to meet changing environmental
needs. Rubber World 1991;209:24–31.
[10] Namboodiri CSS, Tripathy DK. Strain-dependent isothermal damping
behaviour of filled EPDM rubber: effect of vulcanizing system. Plast Rubber
Compos Process Appl 1992;17(3):171–8.
[11] Maity SK, Chakraborty KK. Studies on curing characteristics of natural rubber-,
nitrile rubber- and silicone rubber-based gum vulcanizates in the presence of
boron compounds. J Elastomers Plast 1993;25(4):358–80.
[12] Bhuvaneswari CM, Kakade SD, Deuskar VD, Dange AB, Gupta Manoj. Filled
ethylene–propylene diene terpolymer elastomer as thermal insulator for case-
bonded solid rocket motors. Defence Sci J 2008;58:94–102.
[13] Subramanian V, Ganapathy S. Aging of vulcanizates of formulations for rubber
seals. J Appl Polym Sci 1998;70(5):985–94.
[14] Hassander H, Tornell B. Methods for testing morphology–property
relationships in rubber blends. Polym Test 1985;5(1):11–25.
[15] Samarz
ˇija-Jovanovic
´S, Jovanovic
´V, Markovic
´G. Thermal and vulcanization
kinetic behaviour of acrylonitrile butadiene rubber reinforced by carbon black.
J Therm Anal Calorim 2008;94:797–803.
[16] Samarz
ˇija-Jovanovic
´S, Jovanovic
´V, Markovic
´G, Marinovic
´-Cincovic
´M. The
effect of different types of carbon blacks on the rheological and thermal
properties of acrylonitrile butadiene rubber. J Therm Anal Calorim
2009;98:275–83.
[17] Jovanovic
´V, Budinski-Simendic
´J, Samarz
ˇija-Jovanovic
´S, Markovic
´G,
Marinovic
´-Cincovic
´M. The influence of carbon black on curing kinetics and
thermal aging of acrylonitrile–butadiene rubber. CI&CEQ 2009;15:283–9.
[18] Sirqueira AS, Bluma SG. The effect of mercapto- and thioacetate-modified
EPDM on the curing parameters and mechanical properties of natural rubber/
EPDM blends. Eur Polym J 2003;39:2283–90.
[19] Lukomskaya AI, Badenkov PF, Kepersha LM. Teplovie osnovi vulkanizacii
rezinovih izdeliya, Moskva; 1972.
[20] Choi SS, Kim JC, Woo CS. Accelerated thermal aging behaviors of EPDM and
NBR vulcanizates. Bull Korean Chem Soc 2006;27(6):936–8.
[21] Ghosh AK, Basu DK. CO-vulcanization of acrylonitrile–butadiene rubber and
ethylene–propylene–diene rubber blends. KGK 2003;56(3):101–9.
[22] Agullo N, Borros S. Qualitative and quantitative determination of the polymer
content in rubber formulations. J Therm Anal Calorim 2002;67:513–22.
[23] Essawy H, El-Nashar D. The use of montmorillonite as a reinforcing and
compatibilizing filler for NBR/SBR rubber blend. Polym Test 2004;23(7):803–7.
[24] Zaharescu T, Meltzer V, Vîlcu R. Thermal properties of EPDM/NR blends. Polym
Degrad Stab 2000;70(3):341–5.
[25] Zhang LQ, Li T, Lu YL, Tang YW, Qiao JL, Tian M. The morphology and property
of ultra-fine full-vulcanized acrylonitrile butadiene rubber particles/EPDM
blends. J Appl Polym Sci 2006;100:3673–9.
[26] Gent AN, Pulford CRT. Micromechanics of fracture in elastomers. J Mater Sci
1984;19:3612–9.
[27] Gunasekaran S, Natarajan RK, Kala A. FTIR spectra and mechanical strength
analysis of some selected rubber derivatives. Spectrochim Acta Part A
2007;68:323–30.
[28] Moldovan Z, Ionescu F, Litescu S, Vasilescu I, Lucian Radu G. EPDM–HDPE
blends with different cure systems/mechanical and infra-red spectrometric
properties. J Appl Polym Sci 2008;8(1):86–94.
[29] Hwang W, Wei K. Mechanical, thermal, and barrier properties of NBR/
organosilicate nanocomposites. Polym Eng Sci 2004;44(11):2117–24.
[30] Suzuki N, Ito M, Ono S. Effects of rubber/filler interactions on the structural
development and mechanical properties of NBR/silica composites. J Appl
Polym Sci 2005;95:74–81.
[31] Oliveira M, Soares B. Mercapto-modified copolymers in polymer blends. III.
The effect of functionalized ethylene–propylene–diene rubber (EPDM) on
curing and mechanical properties of NBR/EPDM blends. J Appl Polym Sci
2001;82:38–52.
Fig. 5. The FTIR spectra of EPDM and NBR rubbers without filler and EPDM/NBR
rubber blend with 50 phr SIL-1.
1250 S. Samarz
ˇija-Jovanovic
´et al. / Composites: Part B 42 (2011) 1244–1250