Abraham Et Al. 2013

7/18/2019 Abraham Et Al. 2013

http://slidepdf.com/reader/full/abraham-et-al-2013 1/11

O R I G I N A L P A P E R

Physicomechanical properties of nanocomposites based

on cellulose nanofibre and natural rubber latex

Eldho Abraham • B. Deepa • L. A. Pothan •

Maya John • S. S. Narine • S. Thomas •

R. Anandjiwala

Received: 5 July 2012 / Accepted: 12 November 2012 / Published online: 22 November 2012

Springer Science+Business Media Dordrecht 2012

Abstract Cellulose nanofibres (CNF) with diameter

10–60 nm were isolated from raw banana fibres by

steam explosion process. These CNF were used as

reinforcing elements in natural rubber (NR) latex

along with cross linking agents to prepare nanocom-

posite films. The effect of CNF loading on the

mechanical and dynamic mechanical (DMA) proper-

ties of NR/CNF nanocomposite was studied. The

morphological, crystallographic and spectroscopic

changes were also analyzed. Significant improvement

of Young’s modulus and tensile strength was observedas a result of addition of CNF to the rubber matrix

especially at higher CNF loading. DMA showed a

change in the storage modulus of the rubber matrix

upon addition of CNF which proves the reinforcing

effect of CNF in the NR latex. A mechanism is

suggested for the introduction of the Zn–cellulose

complex and its three dimensional network as a result

of the reaction between the cellulose and the Zinc

metal which is originated during the composite

formation.

Keywords Bionanocomposites Fibre/matrix bond

Thermomechanical properties Natural rubber

Nanocellulose Zn–cellulose complex

Introduction

There is currently a considerable interest in processing

polymeric composite materials filled with rigid parti-

cles having at least one dimension in the nanometer

range. Because of the nanometric size effect, these

composites display some unique properties with

respect to their conventional microcomposite coun-

terparts (Dufresne et al. 1996). In the context of both

biomass valorization and nanocomposite materialsdevelopment, cellulose nanofibres used as a filler in a

polymeric matrix appear to be an interesting reinforc-

ing agent (Angles and Dufresne 2000; Angellier et al.

2005). Cellulose is the most abundant and renewable

biomaterial on earth and is totally biodegradable.

Furthermore, compared to platelets like morphology

of nanocrystals extracted from other sources like

starch and potato, CNFs present the originality to have

a rod like morphology having very high aspect ratio

E. Abraham S. S. Narine

Trent Biomaterials Research Program, Trent University,

Peterborough, ON, Canada

E. Abraham B. Deepa L. A. Pothan (&)

Department of Chemistry, Bishop Moore College,Mavelikkara 690 101, Kerala, India

e-mail: [email protected]

E. Abraham (&) S. Thomas

School of Chemical Sciences, Mahatma Gandhi

University, Kottayam 686 560, Kerala, India

e-mail: [email protected]

M. John R. Anandjiwala

Polymers and Composites Competence Area, CSIR,

Materials Science and Manufacturing, Port Elizabeth,

South Africa

1 3

Cellulose (2013) 20:417–427

DOI 10.1007/s10570-012-9830-1

7/18/2019 Abraham Et Al. 2013


which gives exceptional characteristics to the nanofi-

bres (Putaux et al. 2003; Youssef et al. 2010). We have

prepared CNFs by a patented steam explosion tech-

nique in order to extend their use in nanocomposite

applications (Abraham et al. 2011). The preparation of

CNF by this technique will give additional advantages

over earlier ones by the high yield (*60 %) and theisolation of the nanofibre as embedded in the raw fibre

entity. They consist of crystalline nanofibres about

10–40 nm diameter with a length of few micrometers.

Cellulose nanofibres have a density of 1.5 g/cm3 and it

has high value of tensile strength (*10 GPa) and

Young’s modulus (*134 GPa) (Schurz 1999). In the

past decades, research has been focused on the devel-

opment of other reinforcing agents to replace carbon

black in rubber compounds since it is potentially toxic

and it gives to the rubber a black color. Recently kaolin

and silica were commonly used as reinforcing agents,but their reinforcing properties are lower than those

obtained with carbon black. A variety of clays (Kim

et al. 2004) have been used to obtain unusual nanocom-

posites by exploiting theability of the clay silicatelayers

to disperse into polymer matrices. The use of clay

minerals such as montmorillonite (Vu et al. 2001) and

organoclays (Bala et al. 2004) has been extended to

natural rubber, and they seem to be a potential substitute

to carbon black. When compared with glass fibers, silica

and carbon black, CNF as reinforcing filler in compos-

ites has many advantages: low cost, low density, easyprocessability, and little abrasion to equipment, renew-

ability, and biodegradability.

Nanocomposites of natural rubber filled with cel-

lulose fibre were reported by earlier researchers

(Bendahou et al. 2010; Siqueira et al. 2011; Visakh

et al. 2012). But we have specifically used CNF which

is obtained by steam pre-treatment of the banana bast

fibre as the reinforcing filler for NR latex. More over

the crosslinking agents like Zinc dithiocarbomate

(ZDC), Zinc mercapto benzothiozole (ZMBT), Zinc

oxide (ZnO) and sulfur were used during the process-

ing stage to get a product with maximum mechanical

and barrier properties.

Materials

Natural rubber latex, the matrix

Centrifuged latex of natural rubber was kindly

supplied by Rubber Board, Kottayam, Kerala, India.

It contained spherical particles with an average

diameter around 1 lm, with a dry rubber content of

60 DRC and it contains more than 98 % of cis-1,4-

polyisoprene.

Cellulose nanofibre, the reinforcement

A homogenous CNF (with a diameter of 10–60 nm)

dispersion which is obtained by the steam explosion of

the banana fibre is used as the reinforcing material. As

reported in our previous work, the isolation of the

nanofibe includes the alkali treatment coupled with

steam explosion, bleaching and mild acid hydrolysis

with oxalic acid are the main preparation steps

(Abraham et al. 2011).

Film processing

The CNF is mixed with the NR matrix along with other

cross linking components as described in the Table 1

formulation. The composite films were prepared from

prevulcanised latex (Stephen et al. 2006) by casting on

Table 1 Formulation of the cross linked NR/CNF composite

Weight % in the nanocomposite

Neat NR 2.5 % 5 % 7.5 % 10 %

Centrifuged natural rubber latex 96.5 94 91.5 89 86.5

Potassium hydroxide solution 0.35 0.35 0.35 0.35 0.35

Sulphur dispersion 1.5 1.5 1.5 1.5 1.5

ZDC dispersion 0.75 0.75 0.75 0.75 0.75

ZMBT dispersion 0.6 0.6 0.6 0.6 0.6

Zinc oxide dispersion 0.3 0.3 0.3 0.3 0.3

CNF dispersion 0 2.5 5 7.5 10

418 Cellulose (2013) 20:417–427

1 3

7/18/2019 Abraham Et Al. 2013


a glass plate followed by drying. The prevulcanization

of the compounded latex was conducted at 70 C for

2 h using water bath with constant gentle stirring. The

sample numbers 0, 2.5, 5, 7.5 and 10 indicate the

weight percentage of fillers (CNF) used.

The aqueous suspension of the various proportions

of CNF, latex and the crosslinking agents were mixedby ball milling followed by ultra sonication. The ball

milling process was conducted for 2 h in water

medium with ceramic balls (1.5 cm diameter) in

stainless steel container (1 l). The speed of the mill

was 300 rotations per minute. It is then subjected to

ultra-sonication for 10 min in 50 % amplitude at room

temperature. The mixed aqueous suspension of the

nanocomposites poured into glass plates and dried for

24 h at oven with a temperature of 50–60 C in order

to obtain dry films between 1 and 2 mm thick

depending on the test and with weight fractions of CNF within the NR matrix ranging from 0 to 10 wt%.

Resulting films were conditioned at room temperature

in desiccators containing P2O5 until tested.

Methods

Scanning electron microscopic (SEM) analysis

SEM analysis of the samples were done by an

Analytical Scanning Electron Microscope (A-SEM),ZEISS EVO 60. The Microscope works with tungsten

filament and maximum acceleration voltage of 20 kV.

The samples were mounted on aluminium stabs and

gold-coated with a sputter coater.

Atomic force microscopy (AFM)

The morphological study of the CNF and composites

were performed by atomic force microscopy (AFM).

The AFM images were made with a Multimode AFM

(Agilent Inc. Santa Barbara, USA) with a NanoscopeIV controller in tapping mode at room temperature. A

dilute solution of CNF dispersion which was sonicated

just before the experiment was used. A drop of CNF

suspension was deposited onto a freshly cleaved mica

surface and air-dried. For the evaluation of the

dispersion of CNF on the composite, a suitable knife

was used to cut the inner part of the sample and placed

on the mica surface. The silicon nitride cantilever with

a spring constant of 40 Nm-1 was used. The scan rate

of 1.0 Hz and 512 lines per 5 lm was used to optimum

contrast.

Fourier transform infrared spectroscopy (FTIR)

Fourier Transform-Infra Red spectroscopy (FTIR)

spectra of the fibers were recorded by a Shimadzu IR-470 IR spectrophotometer.

X-ray diffraction analysis

X-ray diffraction was performed using a Bruker AXS

X-ray diffractometer equipped with a filtered Cu–K a

radiation source (k = 0.1542 nm) at the operating

voltage and current of 45 kV and 40 mA, respectively

and a 2D detector.

Mechanical testing

Tests for measuring the tensile strength were carried

out according to ASTM designation D 412-98 using

dumb-bell specimens. Rate of separation of power-

actuated grip was 500 mm/min, which was maintained

throughout the experiment.

DMA analysis

The dynamic mechanical thermal analysis was con-

ducted using rectangular samples with dimensions35 9 10 9 2 mm on the DMTA machine of the TA

instrument. The experiment was conducted at the

frequency of 1 Hz in the temperature range of -80 to

?100 C.

Results and discussion

Morphological analysis of CNF

and nanocomposite films

Morphological analysis of the CNF which is isolated

from banana fibre by steam explosion process was

conducted by AFM and is shown in Fig. 1. The

diameter of the individual nanofibres is ranging from 5

to 20 nm. The morphology of NR/CNF composite

materials and the distribution level of the filler within

the matrix were evaluated by SEM and AFM and are

shown in Figs. 2 and 3, respectively. Fractured surface

of nanometric reinforced rubbers were observed by

Cellulose (2013) 20:417–427 419

1 3

7/18/2019 Abraham Et Al. 2013


SEM in order to know how particles were oriented

inside the composites together with the possible

examination of rubber-nanofibre bonding after the

mechanical evaluation. SEM of the tensile fracture

surface of 2.5 % CNF nanocomposite (Fig. 2) reveals

that the reinforced CNFs are coated with NR latex.

CNF appear like white tubes, whose concentration is a

direct function of the cellulose content in the com-

posite. We can observe that the distribution of the fillerin the matrix is almost homogeneous but some

agglomeration of the nanofillers which is evident from

SEM analysis. The fibre pull out is not observed in the

SEM picture but the CNF is embedded in the rubber

matrix. On the other hand, the microphotographs of

the fracture surface of the samples indicate that CNF

are well dispersed in the rubber matrix and fractured

surfaces did not show a high concentration of defects

like broken particles or detached particles as in the

microcomposites of earlier reports (Ismail et al. 2002),most likely as a consequence of nanomeric size of the

celluloses and the strong interaction between the

matrix and the filler. Furthermore, no particular

sedimentation phenomenon of CNF within the thick-

ness of the films was observed during the composite

preparation.

AFM analysis of the inner part of the nanocom-

posite enabled more detailed information about the

dispersion CNF in the rubber latex matrix. The Fig. 3

shows the presence of a homogeneous dispersion of

CNF along with some agglomeration of nanofibres inthe composite. Most of the CNF were dispersed in the

10–40 nm range as evaluated by SEM data. But some

parts show some agglomeration with a range of

200–400 nm range because AFM is more sensitive

in its resolution at nanometer dimension than SEM.

The agglomeration is more pronounced with the

increasing of CNF percentage. NR has a dielectric

constant value of 250 and CNF has *3,000 at

1,000 Hz frequency (Carter et al. 1946). The differ-

ence in the polarity of the matrix and filler makes this

agglomeration which will definitely affect the physicalinteraction between the filler and the matrix. The CNF

appeared significantly broader having a rounded shape

(Deepa et al. 2011). It was therefore difficult to judge

whether the structures observed in the AFM were

individual nanofibre or several nanofibres, agglomer-

ated side-by-side. The latter is reflected in the

measured length of the nanofibres. An estimate of

the width of the CNF by AFM was not obtained due to

the broadening effect and possibly agglomeration of

the nanofibres in the natural rubber matrix. Comparing

with the neat matrix (Fig. 3a), the darkened portionsare the reinforced nanofillers. The AFM of the

nanocomposite (Fig. 3b) has some white portions

which represent the 400–500 nm dimensions and the

agglomerated nanofibres might have some contribu-

tion to this area. The findings from the agglomerated

nanofibres as marked in the SEM (Fig. 2) and the

evaluation of the mechanical properties leading to the

conclusion for the agglomeration of the nanofibres in

the reinforced nanocomposites.

Fig. 1 AFM image of the cellulose nanofibre

Fig. 2 SEM of the tensile fracture surface of 2.5 % CNF

nanocomposite

420 Cellulose (2013) 20:417–427

1 3

7/18/2019 Abraham Et Al. 2013


Deprotonation of cellulose and formation

Zn–cellulose complex

It is reported that the cellulose can be dissolved in

aqueous solvent systems containing metals such as

copper and zinc in Cuam and Cuen solutions (Klemm

et al. 1998). The dissolution mechanism involves

deprotonation of C2–OH and C3–OH during the metal

complex formation. The deprotonation of cellulose is

taking place at alkaline medium with the presence of

copper and zinc (Ning-Jun et al. 1994). Centrifuged

latex, the matrix, whose pH is 11, will make the

favorable atmosphere for the deprotonation of cellu-lose. It is reported that the formation of a zinc–

cellulose complex during the pre-treatment of cellu-

lose improves the yield of cellulose in both the

enzymatic and acid hydrolysis of cellulose (Cao et al.

1995). The spectral characteristics of aqueous zinc

chloride solutions containing cellulose were also

reported (Xu and Chen 1999). The results suggest

that zinc ion forms loose complexes with the C2 or C3

hydroxyl groups of glucopyranose. The high solubility

of cellulose in amine containing copper (II) and Zinc

(II) is well known (Arthur et al. 1971) but the exactrole of Cu (II) and Zinc (II) in cellulose dissolution

process and the nature of the Cu (II) and Zinc (II)

coordination are ill-defined. The formation of a

transient cellulose-zinc is reported by researchers

(Calvin 2000) and the deprotonation of the cellulose

taking place with the formation of a complex ion

with metal. The accelerators and crosslinking

agents like Zinc dithiocarbamate (ZDC), zinc merca-

ptobenzothiozole (ZMBT) and zinc oxide (ZnO) were

used in the nanocomposite preparation stage to

generate active sites in the NR backbone for thecrosslinking of the matrix. These accelerators not only

make active sites in the NR backbone by breaking the

C = C double bond but also have a major role in the

deprotonation of the CNF and make them participate

in the crosslinking network.

The ball milling process is applied for the mixing of

the CNF, crosslinking agents and NR latex. After

mixing the correct formulations of components, the

mixture is subjected to ball milling for 2 h. It is then

subjected to sonication for 15 min at 50 % amplitude

at room temperature. The mechanism of ball-millingfollowed by sonication caused an increase of amor-

phous phase of cellulose. The deprotonation of

cellulose is difficult by mechanical agitation but ball

milling followed by ultra-sonication will make suit-

able atmosphere to induce reaction sites in the

cellulose molecule (Zhang et al. 2008) for de-proton-

ation in the presence of alkaline zinc solution. The ball

milling followed by ultra-sonication of the composite

mixture will activate the cellulose for deprotonation

and weakens inter- and intramolecular hydrogen

bonding (Biswas et al. 2005). It is reported thatcellulose can be preactivated via ball-milling and they

described a two-step mechanism for the synthesis of

cellulose-MAPP by a mechano-chemical process (Qiu

et al. 2004). In this study, a similar mechanism might

be assumed. Under ball-milling, cellulose fibers

dispersed into nano particles whose specific surface

area increased remarkably which favoured chemical

reactions with other reagents. Besides, ball-milling

followed by ultra-sonication caused an increase of

Fig. 3 AFM image of the (a) neat matrix and (b) 10 % CNF nanocomposite

Cellulose (2013) 20:417–427 421

1 3

7/18/2019 Abraham Et Al. 2013


amorphous phase of cellulose, making favourable

atmosphere for the breakage of inter and intramolec-

ular hydrogen bonds which could produce a large

number of free OH groups at the surface of CNF. The

resultant OH groups at the cellulose surface were

activated, and could form complexes with Zinc (II)

during pan-milling. The incorporation of fillers causesinterruption in the alignment process of chains. When

filler loading is increased, weak interfacial regions

between filler surface and rubber matrix are formed.

The resultant Zn–cellulose complex diminishes the

crystalline nature of the composite and dispersed with

metal ion (Abraham et al. 2012). A reaction mecha-

nism is suggested and is shown in Fig. 4.

XRD analysis of the NR/CNF composites

X-ray patterns were collected for different composi-tions of NR/CNF and are displayed in Fig. 5. The

diffractograms of unfilled natural rubber and pure

CNF were added as references. The diffraction pattern

recorded for a film of pure CNF obtained by pressing

freeze-dried nanofibres displays typical peaks of

A-type amylose allomorph (Xu and Chen 1999). It is

characterized by a strong peak at 2h = 17.9, a very

strong peak at 25.07. The natural rubber film displays

a typical behavior of a fully amorphous polymer. It is

characterized by a broad hump located around

2h = 18. By adding CNF, the peaks corresponding

to A-type amylose allomorph completely disappear

and this is unexpected. This shows that during the

composite formation with CNF, the natural rubber andcross linking agents makes some additional chemical

interactions with the CNF which leads to the loss of

crystalline nature of the cellulose as well as the entire

composite. The proposed reaction mechanism

between the matrix, filler and crosslinking agents are

shown in Fig. 4. The crystalline peak of the cellulose

is completely absent for all the composite composi-

tion. Under the experimental conditions, there is a

perfect agreement between the parallel chain arrange-

ments in the crystals of CNF. The rubber makes

chemical bonds with CNF through Zn–cellulosecomplex which will result in the loss of its crystallin-

ity. The filler and the cross linking agents enter in

between the layers of natural rubber making new

bonds with each other. The breaking of one C = C

bond in the natural rubber back bone makes two active

sites and a cross linking network between sulfur and

natural rubber. In addition to that cross linking

Fig. 4 Proposed

mechanism for interactionof the cross linked NR/CNF

nanocomposite

422 Cellulose (2013) 20:417–427

1 3

7/18/2019 Abraham Et Al. 2013


network, cellulose units makes additional network of bonding with zinc to form a zinc–cellulose complex.

This makes the clear dispersion of the CNF in the

layers of rubber latex which will leads to the loss of

crystallinity of cellulose in the composite entity.

Fourier transform infrared (FTIR) analysis

The FTIR transmittance spectra of NR and its compos-

ites with CNF are shown in Fig. 6. The transmittance

spectra, 3,340 and 3,400 cm-1arethe peaks attributed to

the stretching vibrations of the hydroxyl group of CNFand 10 % CNF composite respectively. The adsorbed

water has also some contribution to the corresponding

peak and its intensity is attributed to the three hydroxyl

group present in each cellulosic unit. There is a gradual

increase in the intensity of this peak on the nanocom-

posites with the increase in percentage of CNF which is

expecting. As already discussed, during the composite

formation, in addition to the physical interactions, there

are some weak chemical reactions taking place between

zinc metal and CNF. The corresponding spectrum of CNF is blue-shifted to 3,400 cm-1 in thespectrum of the

composite and the hydroxyl groups of CNF might have

some important role in this shift. These phenomena

confirm that CNF is successfully activated by crosslink-

ing agents to react with accelerator, zinc metal to form

zinc–cellulose complex.

The stretching vibration of methane (2,932 cm-1)

which is attached to the isoprene backbone is pro-

nounced in the spectra of natural rubber. The vibration

of the methyl group is one of the reasons for the

amorphous nature of natural rubber. The intensity of the CH3 and C–H vibration whichgives a strong peak at

2,932 and 2,896 cm-1 gradually increases with the

percentage of the reinforcing filler, CNF. The nano-

composites of NR/CNF give more freedom to the side

chain CH3- for vibration because of the presence of

Zn–cellulose complex in between the molecular chains

of NR. The breaking up of double bond in the

polyisoprene units also gives some relief to the side

chain CH3- molecule for vibration. It is observed that

there is an increase in the intensity of the CH2

stretching vibration at 2,949 cm-

1 and C–H stretchingvibration at 1,700 cm-1 for the composite. This

suggests the breaking up of C = C bond of the NR

backbone and a comparative relaxation in the adjacent

CH2 and C–H bonds. The strong peak at 2,896 cm-1

which is present in CNF is completely absent in all the

nanocomposites. The absence of this C–H stretching

vibration peak characterizing the hydrogen bonds

between cellulose chains, suggests the breakage of

inter-molecular hydrogen bonds during reinforcing

CNF. The free hydroxyl groups of CNF are expected to

establish new hydrogen bonds with the other neigh-boring CNF molecules during the compounding pro-

cess, leading to a more robust three dimensional

network of reinforcing filler in between the rubber

phase. The rubber particles are already interconnected

by the cross linking agents and resulting a composite

with improved mechanical and dynamic mechanical

properties. In addition to hydroxyl density which is at

3,400 cm-1, the intensity of transmittance bands at

832, 1,247 and 1,725 cm-1 is also pronounced in the

Fig. 5 XRD analysis of the nanocomposites

Fig. 6 FTIR spectra of the nanocomposites

Cellulose (2013) 20:417–427 423

1 3

7/18/2019 Abraham Et Al. 2013


IR of natural rubber (Freire et al. 2011). The vibrations

of C = O stretching peak at 1,671 cm-1 is almost

absent in natural rubber but its intensity shows a

gradual increase with increase in the percentage of

CNF which reveals the formation of the complex.

Mechanical properties of the composite

The importance of CNF reinforced composites of

polymeric materials comes from the substantial

improvement of strength and modulus which gives

the suitable application of the composites in various

fields. The mechanical properties of the natural rubber

and its nanocomposites at different filler loading are

shown in Table 2. At room temperature, unfilled NR

shows a highly elastic non-linear behavior typically of

elastomeric materials. Its elongation at break and the

tensile modulus was higher than 900 % and lower than2 MPa, respectively. A decrease of the elongation at

break and significant increase of both the Young’s

modulus and strength were observed upon adding

cellulose nanoparticles. The presence of hydroxyl

groups in the CNF is responsible for the inherent

hydrophilic nature of the nanocomposite. Since the

matrix, NR latex is in its liquid aqueous form with its

highest tendency of hydrophilicity, the compatibility

between the hydrophobic natural rubber and hydro-

philic CNF is compromised to some extent during the

compounding stage. As a result, it becomes very easyto compound CNF dispersion with natural rubber latex

along with other cross linking agents which are in

aqueous medium. This results in an efficient nano-

composite with good interfacial interaction.

As the NR is reinforced with CNF, some interactions

are formed between the filler and the crosslinking agents

and thereby matrix, (Fig. 4) which consequently

increase the mechanical properties (Jamil et al. 2006;

Siqueira et al. 2011). The mechanism suggests a three

dimensional network of NR/Zn/CNF which increases

the mechanical properties of the composite. Of the three

hydroxyl groups present in cellulose, a hydroglucose

unit, one is primary hydroxyl group at C 6, while the

other two are secondary hydroxyl groups at C 2 and C 3

positions. A three dimensional network of CNF isformed within the composite entity and this network is

formed by reacting with these hydroxyl groups. Due to

this three dimensional network between the CNF, the

interfacial bonding and mechanical interlocking

between the nanofibre through crosslinking agents to

matrix is increased in the resultant composite. This in

turn increases the tensile strength of the composites at all

mixing ratios compared to that of neat NR. The

decreased crystalline nature of the composite which is

evident from the XRD leading to the deterioration of the

mechanical performance of the composites. But theenhancement of these properties upon addition of CNF

makes clear that the proposed mechanism with the three

dimensional network of CNF/Zn/NR dominate over the

crystalline nature of the composite when mechanical

stress is applied. More over an increase in the CNF

content increases the interaction sites between filler–

matrix. The specific surface area of the filler in contact

with thematrix is also increased. As a result, thevalues of

tensile strength show an increasing trend with increasing

nanofibre content in the composite. The enhanced

interfacial bonding will increasestress transferefficiencyfrom the matrix to the filler with a consequent improve-

ment in the mechanical properties of the composites.

Of the studied nanocomposites, a maximum tensile

strength is shown by composites with loading 10 %

CNF. Beyond 10 % CNF concentration, the composite

preparation is difficult since the matrix is not sufficient

to reinforce the nano fibre. The addition of the filler is

expected to increase the modulus of composites that

results from the inclusion of rigid filler particles into the

rubber matrix as reported by Swedish group (Visakh

et al. 2012). It is clear from the Table 1 that with theincrease in the CNF content in the matrix results a

tremendous increase in the modulus of the composites.

The Young’s modulus of the 10 % nanocomposite is six

times higher than the modulus of neat matrix. Usually

crystallites possess higher modulus compared to amor-

phous substances (Karmakar et Al. 2007). Furthermore,

incorporation of fiber into the polymer matrix reduces

the matrix mobility, resulting in stiffness of the com-

posite. As a result, Young’s modulus increase with

Table 2 Mechanical properties of NR and its composite withCNF: Young modulus (E), tensile strength (r R) and strain at

break (e R)

Material E (MPa) rR (MPa) eR (%)

Natural rubber 1.3 ± 0.15 1.6 ± 0.20 912 ± 19

2.5 % nanocomposite 4.2 ± 0.25 5.2 ± 0.15 576 ± 23

5 % nanocomposite 6.3 ± 0.22 6.8 ± 0.18 413 ± 22

7.5 % nanocomposite 8.1 ± 0.35 9.8 ± 0.24 275 ± 12

10 % nanocomposite 9.6 ± 0.31 12.2 ± 0.36 144 ± 5

424 Cellulose (2013) 20:417–427

1 3

7/18/2019 Abraham Et Al. 2013


increasing the filler content of the composites and the

percentage of the elongation is regularly decreased with

increasing the cellulose content. The gum rubber shows

a 900 % elongation but the 10 % nanocomposite shows

*140 % of elongation.

In conclusion, the interactions between the CNF and

the Zn–cellulose complex have a major contribution inthe increment of the stress level of the composite. As

described above, CNF are highly strong materials

compared to other types of fillers like Kevlar and are

good candidate to replace conventional fibres (Liu

et al. 2011). Under load, the matrix distributes the force

to the CNF which carry most of the applied load. The

synergic effect of rubber–rubber, rubber-cellulose,

Zn–cellulose and cellulose–cellulose interactions give

excellent mechanical properties to nanocomposites.

Dynamic mechanical analysis of NR/CNFcomposites

Dynamic mechanical analysis (DMA) is carried out to

study the composite structure and performance under

various temperatures. Figure 7a shows the storage

modulus (E0) and Fig. 7b shows the tangent of the loss

angle (tand) at 1 Hz as a function of temperature for the

NR matrix and nanocomposites reinforced with CNF.

Earlier studies did not observed important changes

in the Tg of nanocomposites reinforced with cellulose

nanofillers. This result was surprising due to the highspecific area of such nanofillers, (Angles and Dufresne

2000; Samir et al. 2004; Morandi et al. 2009). The

increase in values of E0 with fiber loading indicates

that mobility of the polymer chain is decreased in the

presence of fibers and here also the change in Tg is

comparatively negligible. The curve corresponding to

the unfilled NR matrix is typical for fully amorphous

high molecular weight thermoplastic polymer. The

sharp decrease observed around -75 C corresponds

to the main relaxation phenomenon and it is therefore

associated to the inelastic manifestation of the glasstransition. In this temperature range, the loss angle

passes through a maximum (Fig. 7b). It is interesting

to note that at the low temperature, i.e., below Tg, the

reinforcing effect of cellulosic nanoparticle is high.

The exact determination of the glassy modulus

depends on the precise knowledge of the samples

dimensions and at room temperature, the films were

relatively soft and it was difficult to obtain a constant

and precise thickness of these samples. Above Tg, a

much more significant reinforcing effect of the

nanoparticles with the filler percentage was observed

up to -55 C. This reinforcing effect could be

assigned as already reported in the literature to a

mechanical percolation phenomenon of cellulose

nanofibres. Above the percolation threshold, these

nanofibres connect and form a stiff continuous three

dimensional network linked through hydrogen bond-

ing (Dufresne 2008). This is another evidence for the

existence of the three dimensional CNF in the

proposed mechanism in Fig. 4. This effect was wellpredicted from the adaptation of the percolation

concept to the classical series–parallel model. In this

model and at sufficiently high temperature, i.e., when

the modulus of the matrix is much lower than the one

of the percolating network, the elastic modulus of the

composite is simply the product of the volume fraction

and modulus of the rigid percolating network.

It has been observed that the incorporation of solid

crystalline filler into a polymer matrix may increase or

Fig. 7 a Effect of fibre loading on storage modulus of the

nanocomposites. b Effect of fibre loading on the tan d of the

nanocomposites

Cellulose (2013) 20:417–427 425

1 3

7/18/2019 Abraham Et Al. 2013


decrease the mechanical damping of the polymer

(Otaigbe 1991). In the present study also, Fig. 7a, it

was seen that, E’ of all the nanocomposites are greater

than that of gum rubber. Fiber incorporation increases

the storage and loss modulus, which indicates the

higher heat dissipation in the CNF reinforced NR

nanocomposites compared to that of gum rubber. Thestorage and loss modulus of nanocomposites exhibits a

significant drop in the transition zone and the intensity

of the peaks varies with fiber loading. It was observed

that in conjunction with the E’ values, tan d values of

the gum natural rubber is much higher than those of the

composites with 2.5, 5, 7.5 and 10 % CNF at the

experimental frequency. The Fig. 7b shows the lower

peak in tan d curves of composite with 10 % CNF

loading. The nanocomposites exhibit an enhancement

of the modulus at lower temperature range, indicating

the elastic responses of pure NR towards deformationand are strongly influenced by the presence of

cellulose dispersion. This behavior also reflects the

strong confinement of CNF dispersion on the rubber

molecules and the existence of a Zn–cellulose com-

plex. Increasing the amount of CNF successively

increases the values of E 0

, and the composite with

10 % CNF shows highest modulus.

The storage modulus below the transition region

shows a difference based on nanofibre content where as

above the transition there is no improvement is observed

contrary to other researchers (Dufresne et al. 1996)findings. But the mechanical properties like tensile

strength and modulus have a positive improvement

above the transition region. In the mechanical analysis,

the strain was macroscopically homogeneous and uni-

form along the sample, until its break. The samples

exhibit an elastic nonlinear behavior typical of amor-

phous polymer at T[Tg. The stress continuously

increases with the strain. The polymeric matrix is in the

rubbery state and its elasticity from entropic origin is

ascribed to the presence of numerous entanglements due

to high molecular weight chainsand the proposed doublenetwork (Zn–cellulose complex and rubber/rubber).

Contrary to DMA experiments, tensile results which

is done at room temperature show a higher reinforcing

effect for CNF filled NR matrix. This discrepancy

could originate from the fact that dynamic mechanical

measurements involve weak stresses. The possible

interactions of Zn/cellulose network and the rubber/

rubber network are not damaged and/or recorded under

these weak stresses. Under the higher stress level, as

applied in tensile tests, these interactions seem to be

partially destroyed or highly disturbed. More over

Dufresne et al. prepared composite without cross

linking agents and hence the possibility of these

networks are absent. This is another evidence for the

strength of the rubber/rubber and Zn/cellulose network

in the cross-linked NR matrix which is higher in tensiletension than that in the unvulcanized sample. The

chemical cross-linking of the matrix most probably

interferes and strengthen the formation of the

Zn/cellulose network which is more pronounced at

higher (room) temperature analysis.

Conclusion

Nanocomposite materials were obtained by casting

and evaporating a mixture of NR latex and aqueous

suspension of cellulose nanofibrils which is obtained

from banana fibre by steam explosion. The filler was

evenly distributed in the composite structure which is

evident from SEM and AFM analysis. The increase of

CNF content in the NR matrix causes a drastic

influence in the mechanical properties of the compos-

ite, increasing the Young’s modulus and tensile

strength of materials, but decreases the characteristic

rubber elongation. XRD analysis reveals that the

nanocomposites of NR with CNF as filler along with

the cross linking agents will lead to the loss of

crystallinity of the whole composite because of the

nanolevel dispersion of the latter. The formation of a

Zn–cellulose complex and the three dimensional

network of the filler is anticipated as a result of the

deprotonation of the cellulose, during the composite

formation which makes the nanocomposites with good

mechanical and dynamic mechanical properties. The

enhancement in modulus of the composites even

below the glass transition temperature of NR proves

the strong reinforcing tendency of CNF in the NR latex

matrix. The strong stress level involved in DMA

experiments is sufficient to explain the extent of

reinforcing effect of cellulose in NR latex matrix.

References

Abraham E, Deepa B, Pothan LA, Jacob M, Thomas S, Cvelbar

U, Anandjiwala R (2011) Extraction of nanocellulose

426 Cellulose (2013) 20:417–427

1 3

7/18/2019 Abraham Et Al. 2013


fibrils from lignocellulosic fibres: a novel approach. Car-

bohydr Polym 86(4):1468–1475

Abraham E, Elbi PA, Deepa B, Jyotishkumar P, Pothen LA,

Narine SS, Thomas S (2012) X-ray diffraction and bio-

degradation analysis of green composites of natural rubber/

nanocellulose. Polym Degrad Stab 97(11):2378–2387

Angellier H, Molina-Boisseau S, Dufresne A (2005) Mechanical

properties of waxy maize starch nanocrystal reinforced

natural rubber. Macromolecules 38:9161–9170

Angles MN, Dufresne A (2000) Plasticized starch/tunicin

whiskers nanocomposites. 1. Structural analysis. Macro-

molecules 33(22):8344–8353

Arthur JC, Bikales NM, Segal L (1971) Cellulose and cellulose

derivatives. Wiley, New York Part V eds

Bala P, Samantaray BK, Srivastava SK, Nando GB (2004) Or-

ganomodified montmorillonite as filler in natural and

synthetic rubber. J Appl Polym Sci 92(6):3583–3592

Bendahou A, Kaddami H, Dufresne A (2010) Investigation on

the effect of cellulosic nanoparticles’ morphology on the

properties of natural rubber based nanocomposites. Eur

Polym J 46(4):609–620

Biswas A, Shogren RL, Willett JL (2005) Solvent-free processto esterify polysaccharides. Biomacromolecules 6(4):

1843–1845

Calvin W (2000) Regenerated cellulose fibres. Woodhead

publishers Ltd, CRC press LLC

Cao NJ, Xu Q, Chen LF (1995) Acid hydrolysis of cellulose in

zinc chloride solution. Appl Biochem Biotechnol

51–52(1):21–28

Carter WC, Magat M, Schneider WC, Smyth CP (1946)

Dielectric dispersion and absorption in natural rubber,

Neoprene, Butaprene NM and Butaprene S, gum, and tread

stocks. Trans Faraday Soc 42:213–220

Deepa B, Abraham E, Cherian BM, Bismarck A, Blaker JJ,

Pothan LA, Leao AL, Kottaisamy M (2011) Structure,

morphology and thermal characteristics of banana nanofibers obtained by steam explosion. Bioresour Technol

102(2):1988–1997

Dufresne A (2008) Polysaccharide nano crystal reinforced

nanocomposites. Can J Chem 86(6):484–494

Dufresne A, Cavaille JY, Helbert W (1996) Newnanocomposite

materials: microcrystalline starch reinforced thermoplas-

tic. Macromolecules 29(23):7624–7626

Freire MG, Teles ARR, Ferreira RAS, Carlos LD, Lopes-da-

Silva JA, Coutinho JAP (2011) Electrospun nanosized

cellulose fibers using ionic liquids at room temperature.

Green Chem 13:3173–3180

IsmailH, Edyhan M, Wirjosentono B (2002) Bamboo fiber filled

natural rubber composites: the effects of filler loading and

bonding agent. Polym Test 21(2):139–144

Jamil MS, Ahmed I, Abdullah I (2006) Effects of rice husk filler

on the mechanical and thermal properties of liquid natural

rubber compatibilized high-densitypolyethylene/natural

rubber blends. J Polym Res 13(4):315–321

Karmakar A, Chauhan SS, Modak JM, Chanda M (2007)

Mechanical properties of wood–fiber reinforced polypro-

pylene composites: effect of a novel compatibilizer with

isocyanate functional group. Composites A 38(2):227–233

Kim JT, Oh TS, Lee DH (2004) Curing and barrier properties of

NBR/organo-clay nanocomposites. Polym Int 53(4):406–

411

Klemm D, Philipp B, Heinze T, Heinze U, Wagenknecht W

(1998) Comprehensive cellulose chemistry, volume 1,

fundamentals and analytical methods. Wiley-VCH,

Weinheim

Liu A, Walther A, IkkalaO, Belova L, Berglund LA (2011) Clay

nanopaper with tough cellulose nanofiber matrix for fire

retardancy and gas barrier functions. Biomacromolecules

12(3):633–641

Morandi G, Heath L, Thielemans W (2009) Cellulose nano-

crystals grafted with polystyrene chains through surface-

initiated atom transfer radical polymerization (SI-ATRP).

Langmuir 25(14):8280–8286

Ning-Jun C, Xu Q, Chee-Shan C, Gong CS, Chen LF (1994)

Cellulose hydrolysis using zinc chloride as a solvent and

catalyst. Appl Biochem Biotech 45–46:521–530

Otaigbe JU (1991) Dynamic mechanical response of a thermo-

plastic sheet molding compound-glass fiber composite.

Polym Eng Sci 31(2):104–109

Putaux JL, Molina-Boisseau S, Momaur T, Dufresne A (2003)

Platelet nanocrystals resulting from the disruption of waxy

maize starch granules by acid hydrolysis. Biomacromole-

cules 4(5):1198–1202Qiu WL, Zhang FR, Endo T, Hirotsu T (2004) Milling-induced

esterification between cellulose and mateated polypropyl-

ene. J Appl Polym Sci 91(3):1703–1709

Samir AMAS, Alloin F, Sanchez JY, Dufresne A (2004) Cel-

lulose nanocrystals reinforced poly(oxyethylene). Polymer

45(12):4149–4157

Schurz J (1999) A bright future for cellulose. Prog Polym Sci

24(4):481–483

Siqueira G, Tapin-Lingua S, Bras J, da Silva PD, Dufresne A

(2011) Mechanical properties of natural rubber nanocom-

posites reinforced with cellulosic nanoparticles obtained

from combined mechanical shearing, and enzymatic and

acid hydrolysis of sisal fibers. Cellulose 18(1):57–65

Stephen R, Varghese S, Joseph K, Oommen Z, Thomas S (2006)Diffusion and transport through nanocomposites of natural

rubber (NR), carboxylated styrene butadiene rubber

(XSBR) and their blends. J Membr Sci 282(1–2):162–170

Visakh PM, Thomas S, Oksman K, Mathew AP (2012) Cross-

linked natural rubber nanocomposites reinforced with

cellulose whiskers isolated from bamboo waste: processing

and mechanical/thermal properties. Compos A Appl Sci

Manuf 43(4):735–741

Vu YT, Mark JE, Pham LH, Engelhardt MJ (2001) Clay nano-

layer reinforcement of cis-1,4-polyisoprene and epoxi-

dized natural rubber. J Appl Polym Sci 82(6):1391–1403

Xu Q, Chen LF (1999) Ultraviolet spectra and structure of zinc–

cellulose complexes in zinc chloride solution. J Appl

Polym Sci 71(9):1441–1446

Youssef H, Lucian AL, Orlando JR (2010) Cellulose nano-

crystals: chemistry, self-assembly, and applications. Chem

Rev 110:3479–3500

Zhang W, Zhang X, Liang M, Lu C (2008) Mechanochemical

preparation of surface-acetylated cellulose powder to

enhance mechanical properties of cellulose-filler-rein-

forced NR vulcanizates. Compos Sci Technol 68(12):

2479–2484

Cellulose (2013) 20:417–427 427

1 3

Documents

Abraham Et Al. 2013