1. Studies on the Mechanical, Thermal, Morphological and Barrier Properties Based on PVA

8/13/2019 1. Studies on the Mechanical, Thermal, Morphological and Barrier Properties Based on PVA

1/12

Studies on the mechanical, thermal, morphological and barrier

properties of nanocomposites based on poly(vinyl alcohol) and

nanocellulose from sugarcane bagasse

Arup Mandal, Debabrata Chakrabarty*

Department of Polymer Science and Technology, Calcutta University, 92 Acharya Prafulla Chandra Road, Kolkata 700 009, India

1. Introduction

Polymer nanocomposites are made up of nanometric particles

(nanofillers) dispersed in a polymer matrix. The incorporation of a

small amount of nanometer-sized filler can yield composites with

enhanced properties earnestly required for many industrial and

technological applications [1]. The conventional polymerinor-

ganic filler nanocomposites can have improved stiffness, strength,

hardness and high temperature creep resistance compared to the

unfilled polymers [26]. These nanocomposites have recently

become an issue of great concern from environmental, economic

and performance point of view. This can be alleviated by the

replacement of inorganic fillers with natural ones [7].

Cellulose, synthesized mainly in biomass by photosynthesis, is

the most abundant natural biopolymer in the world. Natural

cellulosic fibers, particles, fibrils (micro and nano scale), and

crystals/whiskers are used as reinforcement while making

environmental friendly products. These cellulosic materials have

many advantages including, renewability, low cost, low density,

low energy consumption, high specific strength, modulus,

biodegradability and biocompatibility with less susceptibility to

fracture during processing due to their high aspect ratios in

composites [8,9]. In addition, the waste disposal becomes easier by

combustion for lignocellulosic filled composites (that can be

completely converted into water and CO2) [10]. That is why, the

possibility of using lignocellulosic fillers in the plastic industry

have received considerable attention. Automotive applications

display strong promise for natural fiber reinforcements as well

[1114]. Potential applications of lignocellulosic fiber based

composites in railways, aircraft, irrigation systems, furniture

industries, and sports and leisure items are currently being

researched [15].

Cellulose fibers modified at nanometer size induce much higher

mechanical properties to polymer matrices as regards to common

cellulose fibers because of their higher crystallinity and mechani-

cal properties combined with higher surface area and active

interfaces [16]. Crystalline cellulose nanofibers often referred to as

nanowhiskers display an elastic modulus of 120150 GPa [17]. Due

to their strongly interacting surface hydroxyl groups [18], cellulose

nanowhiskers have a significant tendency for self-association,

which is advantageous for the formation of load-bearing percolat-

ing architectures within the host polymer matrix [19]. The

spectacular reinforcement of polymers observed for this class of

materials is attributed to the formation of rigid nanowhisker

networks in which stress transfer is facilitated by hydrogen-

bonding between the nanowhiskers [20]; Van der Waals interac-

tions also have been shown to play a significant role [18]. However,

these same nanowhiskernanowhisker interactions can also lead

to aggregation during the nanocomposite fabrication [21], which

significantly reduces the mechanical properties of the resulting

Journal of Industrial and Engineering Chemistry xxx (2013) xxxxxx

A R T I C L E I N F O

Article history:Received 22 November 2012

Accepted 6 May 2013

Available online xxx

Keywords:

PVA

Nanocellulose

Morphology

Crystallography

Barrier property

Thermal stability

A B S T R A C T

Nanocomposites from poly(vinyl alcohol) [PVA] in linear and crosslinked state were synthesized usingvaryingproportions of bagasse extractednanocellulose.Thesewerecharacterizedby tensile, thermal, X-

ray diffraction(XRD),moisture vapor transmission rate (MVTR), andmorphological studies. Crosslinked

PVA and linear PVA nanocomposite exhibited highest tensile strength at 5 wt.% and 7.5 wt.% of

nanocellulose respectively. Thermogravimetric analysis (TGA) studies showed higher thermal stability

of nanocomposite made of crosslinked PVA and nanocellulose with respect to linear PVA and

nanocellulose. TEM and AFM studies confirm the formation of nanocomposites while the SEM images

show the dispersion of nanocellulose particles in them.

2013 TheKorean Society of Industrial andEngineering Chemistry. Publishedby Elsevier B.V. All rights

reserved.

* Corresponding author. Tel.: +91 9830773792.

E-mail address: [email protected] (D. Chakrabarty).

G Model

JIEC-1351; No. of Pages 12

Please cite this article in press as: A. Mandal, D. Chakrabarty, J. Ind. Eng. Chem. (2013), http://dx.doi.org/10.1016/j.jiec.2013.05.003

Contents lists available at SciVerse ScienceDirect

Journal of Industrial and Engineering Chemistry

journ al homepage: www.elsev ier .co m/ locate / j iec

1226-086X/$ see front matter 2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.jiec.2013.05.003
http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.jiec.2013.05.003http://www.sciencedirect.com/science/journal/1226086Xhttp://www.sciencedirect.com/science/journal/1226086Xhttp://www.sciencedirect.com/science/journal/1226086Xhttp://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003http://www.sciencedirect.com/science/journal/1226086Xhttp://dx.doi.org/10.1016/j.jiec.2013.05.003mailto:[email protected]://dx.doi.org/10.1016/j.jiec.2013.05.003


2/12

materials compared to predicted values [20]. The traditional

approach to solve this problem is surface functionalization, which

mediates particleparticle and particlepolymer interactions and

significantly influences nanoparticle dispersion [2225].

A drawback of the nanocellulosic filler is their high moisture

absorption and the consequent swelling leading to decrease in

mechanical properties. Moisture absorption and corresponding

dimensional changes can be largely prevented by removing the

reactivity of the surface hydroxyl groups of the nanocellulose by

way of either intramolecular interactions or by intermolecular

interactions with another hydrophilic biocompatible polymer.

Polyvinyl alcohol (PVA) has excellent film forming and

emulsifying properties. It has also high tensile strength and

flexibility. Roohani et al. [26] have accounted for this excellent film

forming ability of PVA due to the hydrogen bonding in PVA-

cellulose whiskers or nanofibers. PVA is also a biodegradable

polymer [27] making it suitable in combination with cellulosic

materials to produce green nanocomposites.

In this paper, we have made an attempt to modify the properties

of PVA [28] by way of incorporating highly reactive nanocellulose

isolated from waste sugarcane bagasse, the synthesis of which is

described in detail in our previous work [29]. The objective of the

present work thus is to evaluate the effect of incorporation of

nanocellulose on the thermomechanical properties of linear andcrosslinked PVA, in relation to theirmorphologies accrued from the

system ofblending used. In this context, theapplicability ofsuchfilm

in packaging fields has been adjudged with respect to theirmoisture

vapor transmission rate (MVTR) performance.

2. Experimental

2.1. Materials

Polyvinyl alcohol [degree of polymerization: 17001800; M.W.:

75,00080,000; and degree of hydrolysis between 98% and 99% from

poly(vinyl acetate)] and glyoxal (40% content in water) were

supplied by Loba Chemie Pvt. Ltd., India. Other reagents used were:

sodium hydroxide (Merck, India), sulfuric acid (Merck, India) andhydrochloric acid (Merck, India). All chemical reagents were used

without any further purification processes. Nanocellulose used was

synthesized from waste sugarcane bagasse in our laboratory.

2.2. Methods

2.2.1. Isolation of nanocellulose

The nanocellulose suspensions were obtained by acid hydroly-

sis of cellulose isolated from sugarcane bagasse according to a

method described in our previous work [29]. Briefly, the delignified

and hemicellulose free cellulose was hydrolyzed with 60 wt.%

sulfuric acid at 50 8C for 5 h under strong agitation. The resulting

suspension was cooled to room temperature and washed with

distilled

water

by

successive

centrifugations

until

pH

7

wasachieved. Finally, the suspension was sonicated (UP-500 Ultrasonic

Processor with Probe) for 5 min in an ice bath to avoid overheating.

The suspension was kept refrigerated until use. The concentration

of nanocelluloses in the final dispersion was determined gravi-

metrically. TEM studies on the resulting particles revealed that the

majority of them have the dimensions 170 nm 35 nm which

were further confirmed by AFM studies [29].

2.2.2. Preparation of PVA nanocomposite films reinforced with

nanocellulose

The PVA solution was first prepared by dissolving free-flowing

granules of PVA in distilled water to a concentration of 5 wt.%, and

stirred at 80 8C for 3 h in a round bottom flask equipped with a

condenser.

Varying

proportions

of

nanocellulose

suspension

with

known solid content of 1 wt.% were added to the prepared PVA

solution to adjust the nanocellulose concentration to 2.5, 5, 7.5 and

10 wt.% (of the weight of the solid PVA content) respectively. The

mixtures were further mechanically stirred for another 2 h and

sonicated for 2 min. The final suspensions were then cast in a

polypropylene petridish and dried at the ambient temperature for 2

days. The resulting composite films were then placed in a vacuum

oven at 60 8C to ensure complete removal of water. The films thus

obtained were kept in the desiccator to remove any remaining water

and also to equilibrate for 24 h before characterization.

2.2.3. Preparation of crosslinked PVA composite films containing

nanocellulose

In order to obtain crosslinked PVA films reinforced with

nanocellulose, a 5 wt.% aqueous solution of PVA was first prepared

and crosslinked with glyoxal (10 wt.% of solid PVA) in the following

process. To a 30 ml reaction vial containing magnetic stirrer, a

known amount of glyoxal solution was combined with 10 ml of

distilled water, followed by PVA solution as prepared above. The

pH of the solution was adjusted to 4 with 1 M HCl solution. The

reaction mixture was stirred at 80 8C for an hour and allowed to

cool down to room temperature. The product mixture was then

neutralized to pH = 7.0 with 1 M NaOH solution [30]. The

nanocellulose suspension as prepared earlier was added to thecrosslinked PVA solution at 2.5, 5, 7.5 and 10 wt.% (of the weight of

the solid PVA content) loadings. The crosslinked PVAnanocellu-

lose suspension was further mechanically stirred, sonicated and

then cast on a polypropylene petridish. The nanocomposite films

were obtained by the evaporation of water followed by the drying

method.

3. Characterization methods

3.1. Fourier transform infrared (FTIR) spectroscopy

FTIR spectra of the various films were recorded with a

spectrophotometer (Jasco FTIR 6300, UK) equipped with an

attenuated total reflectance (ATR) device using a tri-glycenesulfate (TGS) detector. The spectrum for each sample was recorded

in the region of 5004500 cm1 at a resolution of 4 cm1. The

resulting FTIR spectra were compared to evaluate the effects of

nanocellulose filling in the PVA films, based on the intensity and

shift of vibrational bands.

3.2. Mechanical properties

The mechanical behavior [tensile strength (TS), % elongation at

break (% Eb), yield force (YF)] and pictorial representation of

stressstrain behavior of specimen undergoing tensile deforma-

tion of the various PVA composite films with varying proportions

of nanocellulose was determined using an Instron H50KT (Tinius

Olsen

Ltd.,

UK),

tensile

testing

equipment.

The

maximum

force

ofthe cell used in the tensile tests on Instron Machine is 100 N.

Tensile deformation was determined at a crosshead speed of

50 mm/min. The tests were carried out at room temperature, 25 8C.

The dimensions of the test samples according to the standard test

method ASTM D638 were as follows: length 50 mm, width

25.4 mm and thickness 0.05 mm. TS, % Eb, YF were calculated

on the basis of initial sample dimensions, and the results were

presented as the average of five measurements.

3.3. Thermal properties

3.3.1. Thermogravimetric analysis (TGA)

TGA of the films was carried out using a Netzsch TG 209 F1

instrument.

Approximately

6

mg

of

each

sample

was

heated

from

A. Mandal, D. Chakrabarty/Journal of Industrial and Engineering Chemistry xxx (2013) xxxxxx2

G Model


http://dx.doi.org/10.1016/j.jiec.2013.05.003http://dx.doi.org/10.1016/j.jiec.2013.05.003


3/12

30 8C to 700 8C at the heating rate of 10 8C/min. All of the

measurements were performed under a nitrogen atmosphere with

gas flow of 20 cm3/min.

3.3.2. Differential scanning calorimetry (DSC)

DSC was performed with a TA instruments DSC Q1000. The

scanning temperature was from 30 8C to 300 8C using a heating

rate of 10 8C/min under nitrogen atmosphere. The scanning

process comprised an initial heating followed by cooling, and

finally a second temperature scanning was performed.

3.4. X-ray diffraction (XRD)

Wide-angle X-ray diffraction patterns from the nanocomposite

film samples were studied with an XPert Pro Panalytical X-ray

diffractometer (Panalytical Ltd., Cambridge, UK). The generator

operated at 40 kV and 30 mA. The samples were scanned between

2u= 108 and 508 at the rate of 38/min with a Ni-filtered Cu Ka beam

(wavelength 1.5406 A).

3.5. Moisture vapor transmission rate (MVTR)

MVTR tests were conducted gravimetrically using an ASTM

procedure by PATRA method [31]. Films were mechanically sealedonto Patradish containing 5 g of anhydrous calcium chloride. The

Patradishes were initially weighed and placed in a controlled

humidity chamber maintained at 35 8C and 100% RH for 24 h. The

amount of moisture vapor transferred through the film and

absorbed by the desiccant was determined from the weight gain of

the Patradish. The assemblies were weighed initially and after

every 24 h for all samples and continued till a constant weight was

reached. Changes in weight of the Patradish were recorded. The

test was continued until an equilibrium was reached and there was

no further change in weight.

3.6. Morphological studies

3.6.1. Scanning electron microscopy (SEM)SEM was used to investigate the morphology of the PVA

composite films with varying proportions of nanocellulose by

using a Hitachi SEM S3400N (Japan) with an accelerating voltage of

10 kV. In this case, the samples were sputter-coated with a thin

layer of gold to prevent the build up of an electrostatic charge.

3.6.2. Transmission electron microscopy (TEM)

TEM images of the film samples were obtained using a Tecnai

G12 transmission electron microscope (Germany) with an

accelerating voltage of 120 kV in the bright field of transmitted

mode. The film samples were prepared by cutting small pieces

using scalpel knife and then microtoming 100 nm thin sections

from it using a microtome and a diamond knife at 140 8C.

Samples were then taken for TEM observation without staining.

3.6.3. Atomic force microscopy (AFM)

AFM was used to characterize the morphology of the film

samples using a scanning probe microscope and Veeco Nanoscope

IIIa controller (USA). The images were acquired in Tapping mode

etched silicone probes (RTESP) in air using a phosphorus doped

silicon tip (radius 10 nm and cantilever length 125mm) with

frequency of 150 kHz at ambient temperature.

4. Results and discussion

4.1. FTIR spectroscopy analysis

FTIR spectra of nanocellulose, neat linear PVA (spectrum d and a

respectively) and its composite containing 5 wt.% nanocellulose

(spectrum b) and composite with 10 wt.% nanocellulose content

(spectrum c) are presented in Fig. 1. The corresponding spectra of

crosslinked PVA (spectrum a1) and its composites with 5 wt.% and

10 wt.% of nanocellulose (spectrum b1 and c1 respectively) are

presented in Fig. 1. There is practically little difference in the

spectral pattern with respect to its linear counterpart, and almost

all the characteristic peaks of linear PVA composites are present

[30].

For linear neat PVA and its composites with varying proportions

of nanocellulose (5 wt.% and 10 wt.% respectively), a band around

3250 cm1 is observed. The nanocellulose however shows a muchbroader band at 3400 cm1. This peak has been assigned to the free

OH stretching vibration of the OH groups in nanocellulose

Fig. 1. FTIR spectra of the neat linear PVA film (a) and its composites with 5 wt.% NC (b) and 10 wt.% NC (c) & the crosslinked PVA film (a1) and its composites containing 5 wt.%

NC (b1) and 10 wt.% NC (c1). The FTIR spectrum for NC is depicted as a reference (d).

A. Mandal, D. Chakrabarty/Journal of Industrial and Engineering Chemistry xxx (2013) xxxxxx 3

G Model




4/12

molecules [32]. The band at 3250 cm1 is attributed to the typical

OH stretching vibration from the intermolecular and intramolec-

ular hydrogen bonds between the hydroxyl groups of PVA and

nanocellulose and also within the PVA itself. The peak due to the

aliphatic CH stretching vibrations from alkyl groups is observed at

around 2917 cm1 in all cases of linear PVA and its various

composites including nanocellulose. The peak at 1710 cm1 is

assigned to the C55O and CO stretching from the residual acetate

groups in the PVA matrix [33]. The absorbance peak observed at

1645 cm1 in almost all spectra (intensity reduced to a great extent

in neat PVA and its composites) is attributed to the OH bending of

the adsorbed water. The vibration peaks detected at 1420 cm1

and 1375 cm1 have been related to the bending vibration of the C

H bonds. The intensity of the peak in the region 1060 cm1

increased (sharpened) with the addition of nanocellulose (a

characteristic peak of nanocellulose) to the PVA matrix because

of the contribution of COC pyranose ring stretching from the

cellulosic component which in case of nanocellulose alone shows a

broad band width. This stands for a possible interaction of PVA

with nanocellulose. The rocking vibration of CH at around

900 cm1, quite prominent with nanocellulose is also reduced in

case of composites.

The addition of nanocellulose to the PVA matrix is found to have

a mild effect on the intensity of OH stretching. This may be due tothe fact that although the hydroxyl groups on the surface of

nanocelluloses interact with adjacent hydroxyl groups in the PVA

matrix through secondary valence bond formation, its extent is too

low as the nanocellulose content in the composite is not too

marked.

On close inspection it can be observed that the absorption band

of PVA at 850 cm1 which does not overlap with any band for

cellulose was gradually reduced with increasing proportion of

nanocellulose a phenomenon which supports the possible

interaction between nanocellulose and PVA [34].

4.2. Mechanical properties

4.2.1. Tensile strengthAn increase in tensile strength with increases in filler loading

was observed for both linear and crosslinked PVAnanocellulose

composite systems, the degree of enhancement being much more

in case of the crosslinked PVA system and this follows our

expectation as the crosslinked network of neat PVA itself has

higher mechanicals over the corresponding linear ones. The steady

and progressive increases in tensile strength and yield forces of the

uncrosslinked reinforced PVA nanocomposite system with in-

creasing proportions of nanocellulose as shown in Fig. 2(a) and (c)

may be attributed to the inherent chain stiffness and rigidity in

nanocellulose (because of extensive inter and intra molecular

hydrogen bonding within itself), homogeneous distribution of the

nanofillers in the polymer (solution mixing) and high level of

compatibility

between

the

fiber

and

matrix

which

was

furtheraided by high interfacial surface area. The hydrogen bonding

between the nanocellulose and PVA matrix resulted in improve-

ment in mechanicals. Similar observations were made by

Bhatnagar and Sain [35]. Toward the higher ranges of nanocellu-

lose incorporation within the range of study, (beyond 7.5% of

nanocellulose) there was a leveling effect and increases in

mechanical parameters under investigation were little possibly

because of either dilution effect or agglomerating tendency of the

highly active nanosized particles. The tensile strength of neat

linear PVA and crosslinked PVA films were 41.3 kPa and 57.7 kPa

respectively. There was an approximately 48% improvement in the

tensile strength of the linear PVA nanocomposite films with the

addition of nanocellulose at only a 7.5 wt.% concentration in the

linear

PVA

matrix.

The similar trend was displayed for the crosslinked PVA system

nanocomposite films which possessed intrinsically more strength

because of crosslinks. The maximum strength was achieved at

5 wt.% nanocellulose content in this case. This was higher by about

44% with respect to the corresponding linear PVA composite. The

addition of more than 5 wt.% of nanocellulose to the crosslinked

PVA,

caused

a

gradual

and

slow

decrease

in

strength

properties.

Fig. 2. Effect of nanocellulose content on the mechanical property of PVA-based

nanocomposite films: (a) tensile strength, (b) % elongation at break and (c) yield

force.


G Model




5/12

The tensile strengths of crosslinked PVA films with 7.5 wt.% and

10 wt.% nanocellulose were 32% and 37.7% lower, respectively,

compared to that with 5 wt.% nanocellulose.

The tensile strengths of crosslinked PVA films with 7.5 wt.% and

10 wt.% nanocellulose were gradually decreased, compared to

those with 5 wt.% nanocellulose. Thus the maximum strength for

the crosslinked PVA system was achieved at a lower loading of

nanocellulose. In this case beyond 5 wt.% nanocellulose there

might be some problem in homogeneous distribution of nano-

cellulose with high interfacial area because of the presence of

crosslinks. Moreover as the hydroxyl groups were consumed in

crosslinking the possibility of hydrogen bonding with excess of

nanocellulose was also remote.

4.2.2. Percent elongation at break

Contrary to our normal observation of inverse relationship

between tensile strength and % elongation at break (Fig. 2(b)) we

can find here that the % elongation at break of linear PVA

nanocellulose composite films increases with increasing propor-

tions of nanocellulose. Similar observation was made by Qua [36]

and Ibrahim [37]. The latter however reported a decrease in %

elongation at break beyond about 20% of nanocellulose incorpo-

ration for nanospherical cellulose particle from cotton linters. Two

opposing criteria may be assumed to be operative in determiningthe mechanical properties of such composites. Firstly, with the

incorporation of nanomaterials, PVA might be failing in forming

hydrogen bonds (either intramolecularly or intermolecularly) to

the extent desired and thus looses in tensile strength but gains in

ductility as during drying after casting, the formation of hydrogen

bonds amongst PVA chains is disturbed due to the presence of

nanocellulose particles. It may be assumed that the nanocellulose

being a very good carrier of water (as it is hydrophilic in nature)

plasticizes the matrix somewhat both in linear and crosslinked

PVA leading to an increase in ductility and hence in % elongation at

break.

Crosslinked PVA is more brittle than the linear one due to the

fact that the chemical covalent crosslinks reduce the extensibility

on tensile deformation. With the introduction of more and morenanocellulose particles, the % elongation at break remains almost

the same and exhibit very little tendency to increase beyond 5 wt.%

of nanocellulose incorporation. The PVA chains may be assumed to

pass over a large number of rigid inclusions which allow them to

slip past over the particles at the expenses of reasonably higher

energy (stress transfer to the nanocellulose particles) and thus

accounting for higher elongation and hence higher toughness also,

as has been reflected in the forcedisplacement curve, Fig. 3. The

incorporation of more and more nanocellulose into the crosslinked

network of PVA conveys increasing proportions of moisture in it.

The plasticizing influences of this moisture might be counter

balance by the presence of crosslinked in the network of PVA.

Therefore the % elongation at break is much lower compared to the

linear

composites.

The

increases

in

percent

elongation

at

break

andtensile toughness although not too remarkable make the

nanocomposite film strong and tough and we observe a simulta-

neous development in tensile strength and % elongation at break.

This however occurs up to 5 wt.% of nanocellulose incorporation

beyond which the nanosized reinforcements become ineffective

because of the predominance of aggregation.

4.2.3. Yield force

The pattern of changes in yield force with increasing

nanocellulose content (Fig. 2(c)) resembles that of changes in

tensile strength. With variation in nanocellulose incorporation

both linear and crosslinked PVA nanocomposites behave similarly

to changes in tensile strength exhibited by the analogous

composites.

4.2.4. Stressstrain behavior

The force vs displacement curves for both sets of linear neat

PVA and its composites with nanocellulose (2.5wt.% to 10wt.%

respectively) and those of the crosslinked neat PVA and its

corresponding composites like linear ones have been shown in

Fig. 3(a)(e) and (a1)(e1) respectively. These curves can be

considered as stressstrain diagrams and the course of mechani-

cal failure of PVA and its various composites can be traced from

them.

In case of neat linear PVA composites it is apparent from

Fig.

3(a)(e)

that

on

incorporation

of

nanocellulose,

both

theintramolecular and intermolecular hydrogen bonding of neat PVA

undergo breakage and ductility sets in. The toughness which is

measured from the area under the stress vs strain plot also thus

increases. However beyond 7.5 wt.% of nanocellulose incorpo-

ration the properties like % elongation at break and toughness fall

off possibly because of the dilution effect and probably because of

agglomeration of nanocellulose particles.

All the samples of linear PVA and its composites exhibit strain

hardening. It is quite apparent that all these nanocomposites have

hardly any differences in their moduli values either within

themselves or from uncrosslinked neat PVA. They however greatly

differ in their tensile strength values and that is why no graphical

presentation of variation of moduli with variation in nanocellulose

content

has

been

made.

Fig. 3. Forcedisplacement curves of linear PVA (a) and its composites with 2.5 wt.%

NC (b), 5 wt.% NC (c), 7.5 wt.% NC (d) and 10 wt.% NC (e) & crosslinked PVA (a1) and

its composites containing 2.5 wt.% NC (b1), 5 wt.% NC (c1), 7.5 wt.% NC (d1) and

10 wt.% NC (e1).


G Model




6/12


7/12

respectively are more resistant to degradation as indicated by the

extent of residues left (curve 4b1 and 4c1 respectively).

In case of crosslinked PVA besides the interlocking of the

hydroxyl groups of linear PVA through hydrogen bonding, the

remaining hydroxyl groups appear to be involved in the process of

crosslinking through glyoxal and as a result the thermal energy

needed for a cleavage of such a crosslinked network is much higher

with respect to the linear one. That is why the residues left for

crosslinked PVA (57%) is much less compared to that in the

linear PVA.

4.3.2. Differential scanning calorimetry (DSC) study

The DSC thermogram of the linear PVA and its various

composites with nanocellulose in varying proportions have been

shown in Fig. 5(a)(c). In all the cases the studies were undertaken

by a heatingcoolingheating cycle, so-called the HCH proce-

dure. The purpose of the first heating cycle was to remove any

thermal history of the PVA composite systems. It is only the second

heating curve that has been considered in the present case so as to

avoid the influence of moisture and other thermal history (if any).

It is quite apparent that all the heating curves exhibit an

endotherm corresponding to the melting temperatures of virgin

PVA and its composites. We can find here a gradual and marginal

decrease in the melting temperatures of the composites in relationto the neat linear PVA in which case the melting temperature

stands at around 212 8C. It was also distinct from the heating

curves of the composites that the linear PVA almost retains its glass

transition temperature (tg) of 75 8C in all its composites studied

here. On composite formation with nanocellulose we find a drop in

melting temperature of the composites containing 5 wt.% of

nanocellulose to be around 205 8C and that of the composite with

10 wt.% of nanocellulose at 204 8C. On further investigation of the

cooling curves it can be seen that the linear PVA has a very sharp

crystallization temperature at 181 8C while its nanocomposite

containing 5 wt.% of nanocellulose has a slight earlier onset of

crystallization at 183 8C and subsequently the composite contain-

ing 10 wt.% of nanocellulose has the same at 180 8C. It is also worth

mentioning here that while the crystallization process is reason-ably sharp in case of linear PVA but the same for the

nanocomposites gets a little diffused over a range of temperature

although very narrow.

The behavior of the crystals present in neat linear PVA and in its

composites is quite different. In case of linear neat PVA the crystals

present may be assumed to be of more uniform in size and shape

such that the crystalline melting takes place very sharply over a

narrow range. During cooling the homogeneous phase nucleation

dictates the formation of crystals which appear to be of uniform in

size and shape. In the presence of nanocellulose which has an

infinitely large surface area the process of crystallization is

governed by heterogeneous phase nucleation which leads to a

large number of small crystallites varying in their sizes such that

the melting takes place over a range of temperature. The influence

of increasing the proportion of nanocellulose in the various

composites is not so marked. On the other hand the variation in

crystallization temperature is not so significant as the crystal

formation is dependent on the phase condition which in the case of

composite is a heterogeneous one. This leads to large number of

small crystallites where the impurities of nanocellulose act as

nuclei for crystal initiation.

In case of crosslinked PVA (a1in Fig. 5) and its composites with

varying proportions of nanocellulose (5 wt.% and 10 wt.% respec-

tively 5b1and 5c1) the second heating curve exhibits a clear glass

transition temperature of 109 8C for composite containing 5 wt.%nanocellulose and 99 8C for composite with 10 wt.% of nanocellu-

lose content. It is quite expected that once the mobility of the chain

segments of PVA is restricted both by the presence of crosslinks

and by the adsorptive forces exerted by the nanocellulose particles.

The tg should be shifted right i.e., toward higher temperature. In

the present case we find that the crosslinked PVA has a relatively

higher glass transition temperature than the linear one and this

trend has been continued in its composites also. Thus the

composites with varying proportion of nanocellulose (considered

under study) have higher glass transition temperatures compared

to the composites of linear PVA with identical proportions of

nanocellulose in it.

A salient feature of the DSC curve for the crosslinked PVA

composite system is the complete absence of any sharp endothermduring the heating cycle, representing the crystalline melting or

any sharp exotherm during the cooling cycle which could have

been interpreted as the temperature of crystallization. However

the corresponding heating and cooling curves exhibit transition

over wide ranges of temperature in relation to the sharp ones as

obtained for linear PVA composites.

A close inspection of the TGA and the DSC thermograms

indicates that the melting endotherm of different samples of linear

PVA and its composites has a melting point of 210213 8C (from

the DSC curve) and about 20% of weight loss has occurred till this

temperature is reached. This can possibly be ascribed to the loss of

about 20% of moisture which was tightly bound with the different

samples of PVA and its composites. Thus the weight loss is possibly

not

due

to

degradation

but

due

to

loss

of

moisture.

4.4. X-ray diffraction studies

XRD patterns of neat linear PVA and its composites with

nanocellulose (5 wt.% and 10 wt.% respectively) have been

compared in Fig. 6(a)(c), while the same of crosslinked neat

PVA and its corresponding composites (5 wt.% and 10 wt.%

respectively) have been shown in Fig. 6(a1)(c1).

The linear PVA and its composites with nanocellulose are all

characterized by a sharp peak at 2u value of 19.38, the single

scattering peak characteristic of PVA [42]. The striking feature in

this diffractogram is that the major peaks of nanocellulose (which

has been used in the present study) as were detected in our earlier

work

[29]

were

at

2u=

12.58

(for

1

1

0

plane)

and

2u=

22.58

(for

Fig. 5. DSC thermograms for linear PVA (a) and its composites containing 5 wt.% NC

(b) and 10 wt.% NC (c) & crosslinked PVA (a1) and its composites containing 5 wt.%

NC

(b1)

and

10

wt.%

NC

(c1).


G Model




8/12

2 0 0 plane). This can be attributed to the presence of very

insignificant

amount

of

nanocellulose

in

the

composite

(wt./wt.).The composite containing 10 wt.% of nanocellulose is however

found to display feebly the characteristic peak of cellulose at

2u= 21.68which appears to be shifted (original nanocellulose peak

at 2u= 22.58 accountable for 2 0 0 plane) toward left.

In case of the crosslinked PVA and its corresponding composites

with 5 wt.% and 10 wt.% nanocellulose respectively, hardly any

difference is observed in the XRD diffractogram when compared to

the linear ones. The sharp peaks of crosslinked PVA and its

composites at 2u= 19.38 exactly resemble the peaks of the linear

ones as observed earlier. The incorporation of crosslinks has thus

very little influence on the crystal structure of PVA. The composites

of the crosslinked PVA and nanocellulose behave in a similar

manner as those of their linear counter parts. It can however be

predicted

from

the

XRD

that

the

composites

of

the

crosslinked

PVA

have got more amorphous character than those of the linear ones.

Both the composites of crosslinked PVA display small humps in the

neighborhood of 2u= 228. The peak for the composite with 10 wt.%

of nanocellulose is more prominent than that of the composite

containing 5 wt.%. This behavior is quite expected.

In this connection it may be mention that the intrinsic

characteristics of neat PVA is destroyed in solution and during

its recrystallization in the presence of 5 wt.% nanocellulose the

original crystallinity is never achieved because of the heteroge-

neous phase nucleation from its solution. However it may be

assumed that in the presence of higher nanocellulose content

(10 wt.%) the overall crystallinity of the composite is improved as

at a higher concentration of nanocellulose it can impart its own

crystallinity which is much higher than PVA and furthermore an

increase in the number of nucleating agents a large number of

small crystallites are bundled together.

4.5. Moisture vapor transmission rate (MVTR) studies

The MVTR study was carried out using PATRA method (ASTM

E96). Fig. 7 represents the % increase in weight due to absorption of

moisture (here shown in the ordinate) over 24 days (as shown in

the abscissa). The behavior of PVA (linear) nanocomposite film

containing varying proportions of nanocellulose has been followedin the present study. From the figure it is apparent that the

nanocomposites containing 2.5 wt.%, 5 wt.%, and 7.5 wt.% of

nanocellulose absorbed moisture at a rate much slower than that

of the linear PVA film. It is important to mention here that the rate

of moisture absorption and diffusion through the film of different

composite (identical thickness) increases with increasing propor-

tions of nanocellulose content beyond a certain minimum. In the

present case the linear PVA nanocomposite with 5 wt.% of

nanocellulose showed the lowest rate of moisture uptake and

beyond this the rate of moisture absorption increases as the %

nanocellulose content in the composite increases. If the level of

nanocellulose is raised further to 10 wt.% the composite film

appears to be inferior in terms of moisture absorption with respect

to the PVA film itself. Nanocellulose is hygroscopic in nature byvirtue of its structure. It is expected that the incorporation of

nanocellulose into a PVA film which itself is hygroscopic in nature

would definitely increase the affinity of moisture absorption.

However here we can find that at the lower % of nanocellulose

incorporation (2.57.5 wt.%) the overall tendency of the nano-

composite films to absorb moisture is lower than PVA. The highly

Fig. 6. XRD patterns of linear PVA (a) and its composites with 5 wt.% NC (b) and

10 wt.% NC (c) & crosslinked PVA film (a1) and its composites containing 5 wt.% NC

(b1) and 10 wt.% NC (c1). The diffractogram of the NC is depicted as a reference (d).

Fig. 7. MVTR analysis of linear PVA composite films with different proportions of

nanocelluloses

and

crosslinked

PVA

film.


G Model




9/12

active nanocellulose when incorporated in PVA undergoes

intermolecular hydrogen bonding with PVA and thus renders

the film tighter and free of sites of moisture absorption. Thus

immediately at the lower level of nanocellulose incorporation, the

rate of moisture absorption of the composite film decreases.

Possibly after a certain concentration of nanocellulose, most of the

OH groups of PVA get blocked making the film stiffer and resistant

to diffusion (possibly occurring in the present case at 5 wt.% of

nanocellulose incorporation). Beyond this concentration the free

nanocellulose present in the nanocomposite starts absorbing

moisture and this tendency goes on increasing with further

increase in proportion of nanocellulose.

Fig. 8. SEM images of linear PVA (a) and its composites with 5 wt.% NC (b) and 10 wt.% NC (c) & crosslinked PVA film (a1) and its composites containing 5 wt.% NC (b1) and

10

wt.%

NC

(c1).

SEM

image

of

NC

is

presented

in

(d).


G Model




10/12

As is apparent from the figure the crosslinked PVA offers much

more resistance to the ingression of moisture compared to the

linear one. This is as per our expectation. The tighter network of the

crosslinked three dimensional network PVA possess much higher

degree of resistance to the transmission of moisture. Although the

linear PVA contains a number of hydrogen bonds the chemical

covalent crosslinks of the crosslinked PVA allow much little

absorption of moisture. A combined action of hydrogen bond and

crosslinks as well, appears to be prevalent in case of crosslinked

PVA.

4.6. Morphological properties

4.6.1. Scanning electron microscopy (SEM) analysis

Fig. 8(a)(c) shows the scanning electron micrographs of neat

linear PVA, nanocomposite of linear PVA containing 5% nanocel-

lulose and the composite of linear PVA with 10% nanocellulose

respectively. The dark homogeneous phase of the linear PVA

appears to accommodate the light phase of nanocellulose from its

suspension and in the process of dispersion in PVA its shape and

size undergo dramatic changes. It is interesting to note here that

the rod shape of nanocellulose particles as obtained in our earlier

work present in its suspension [29] has been converted into a

statistical distribution of somewhat spherical particles of different

dimensions. This may possibly arise from the mode of fabrication

of the composite films. The linear PVA chains seem to roll down the

rod shaped nanocellulose particles during the process of drying

and thus exert a force of winding and shrinkage as well, after

casting the composite blend of nanocellulose. The shrinkage force

appears to be much more predominant because of intermolecular

hydrogen bond formation between the linear PVA chains and free

hydroxyl of nanocellulose. In case of the crosslinked PVA this

possibility appears to be remote as the OH groups of PVA have

already been engaged in crosslink formation and are therefore not

available for hydrogen bond formation. In the ultimate composite,

of linear PVA and nanocellulose the reinforcing particles appear to

remain dissolved in the matrix of PVA as they look diffused and no

sharp boundary of interface between the matrix and the dispersed

phase is observed.

On increasing the concentration of reinforcing nanocellulose

particles in the PVA matrix substantial fragments of particles tend

to agglomerate and cohere. They tend to phase out and pervade the

surface in the form of very bright spherical droplets. Thus these

phase separated particles do not contribute much to the

mechanical properties of the ultimate composite. Thus the

nanocomposite with 10 wt.% of nanocellulose incorporation does

not exhibit much more improved mechanical properties compared

to the one containing 5 wt.% of nanocellulose.

Fig.

9.

TEM

images

of

linear

PVA

(a)

and

its

composites

with

2.5

wt.%

NC

(b)

and

7.5

wt.%

NC

(c).

TEM

image

of

NC

is

presented

in

(d).


G Model




11/12

Fig. 8(a1)(c1) shows the scanning electron micrographs of

crosslinked PVA and its nanocomposites with varying proportions

of reinforcing nanocellulose. The scanning micrograph of the

crosslinked PVA nanocomposites exhibit quite different morphol-

ogy from the linear ones particularly with respect to its dispersion

and size and shape of the dispersed cellulose particles present in

them. The rod shaped morphology of the nanocellulose suspension

[29] appears to be modified in case of nanocomposite. It may be

presumed that the glyoxal motivated crosslinks present in the PVA

network appear to resist the force of shrinkage exerted by the PVA

chains during the fabrication of the composite from a blend of

nanocellulose suspension and PVA solution. The possibility of

hydrogen bond formation with nanocellulose is also not possible as

mentioned above. The rod shaped reinforcement of nanocellulose

particles are dispersed within the crosslinked network of PVA

either in the diffused solubilized (in the matrix) form or in phase

separated sharp rods non uniformly distributed over the composite

surface. It can naturally be expected that mechanical properties of

the crosslinked PVA composites are much higher compared to the

linear ones and this can be ascribed to the aspect ratio and the

consequent more surface attachment between the reinforcement

and the matrix.

On increasing the proportions nanocellulose (10 wt.%) in the

crosslinked network of PVA, the rods of nanocellulose appear to besomewhat broken and thus have lower aspect ratio with respect to

the PVA composite containing 5 wt.% nanocellulose. Furthermore,

because of the high surface energy of the nanocellular rods a

tendency to cluster the rods into thicker brunch is observed. Only

the fragments of nanocellulose which get encapsulated with

matrix help to impart the mechanical properties as was evident in

tensile strength values.

It is important to mention here in the SEM micrographs an idea

of the length of the nanocellulose particles has only been given

which may lie within a micrometer range. However its width has

not been considered which may lie within the range of

nanodimension. The nanocellulose particles may be subjected to

lot of processing strain during the composite formation and its

dimension is expected to undergo a change. Moreover, thenanocellulose particles have a natural tendency to undergo

agglomeration because of high surface energy which may

sometimes lead to an increase in length and width also. In the

present case an attempt has been made to determine the state of

dispersion of such particles by carrying out SEM analysis.

It is possible to result fibers with a wide range of variation in

thickness and shape from the process of fiber isolation. Another

possibility is the further destruction of fibers during the prepara-

tion of nanocomposite.

4.6.2. Transmission electron microscopy (TEM)

Fig. 9(a)(d) represents the transmission electron micrograph

of the neat linear PVA film and nanocellulose reinforced PVA

composites.

The

image

9b

stands

for

composite

containing

2.5

wt.%nanocellulose and 9c represents the composite with 7.5 wt.%

nanocellulose and 9d stands for nanocellulose. In all the cases the

extent of magnification has been kept identical at 97k in order to

investigate whether there has been a nanometric distribution of

the reinforcing nanoparticles in the matrix of PVA. The nanocellu-

lose incorporated in the PVA matrix has average particle dimension

of the order of 175 nm 35 nm (Fig. 9(d)) [29]. In these two linear

composites we find a relatively haphazard non uniform distribu-

tion of the nanocellulose in the matrix. In conformity with the SEM

image of linear PVA nanocomposite we find chains of irregularly

shaped nanocellulose spherical particles because of the reason as

explained earlier. This chain distribution of the nanocellulose

particles enabled the matrix to achieve much higher mechanical

properties

with

respect

to

the

non

reinforced

linear

PVA

matrix

through the surface adsorption of the PVA chains over the

nanocellulose particles. This chain length and its density increased

with increasing nanocellulose content within the range under

study and the effect was reflected in the mechanical properties.

Sometimes the chains superimposed on each other along with the

inclusion of some matrix PVA within themselves.

4.6.3. Atomic force microscopy (AFM)

In order to investigate the sizes of the nanocellulose reinforce-

ments (present in nanocomposite) in the crosslinked network of

PVA, atomic force micrographs were taken of the crosslinked

composite containing 5 and 10 wt.% of nanocellulose. Fig. 10(a)

represents the amplitude image of crosslinked PVA nanocomposite

containing 5 wt.% of nanocellulose. The light regions of the

amplitude image indicate the regions which can be considered

the regions of nanocellulose disposition. It is quite evident that the

marked nanocellulose particle has its width in the nano range,

while the same reinforcement is exhibited in its longitudinal form

in the phase image. The dimension of nanocellulose in the

composite was 63 nm in width.

Fig. 10(b) represents the amplitude image of crosslinked PVA

nanocomposite containing 10 wt.% of nanocellulose. In both of its

images we can find a relatively higher concentration of light

Fig. 10. AFM images of crosslinked PVA nanocomposite films: (a) crosslinked PVA/

5

wt.%

nanocellulose

and

(b)

crosslinked

PVA/10

wt.%

nanocellulose.


G Model




12/12

regions particularly indicating the higher proportions of reinforce-

ment. In spite of the possibility of the coalescence of the

nanocellulose particles we can find the reinforcements retaining

its nanodimension in the matrix. The dimension of nanocellulose

in the composite was 71 nm in width. Thus it can be concluded

from AFM studies that the nanocellulose is distributed in the

crosslinked network of PVA in nanodimension. This substantiates

our claim of nanocomposite synthesis.

5. Conclusions

Nanocellulose synthesized in-house has been used as rein-

forcement of PVA. There was remarkable improvement in tensile

strength, % elongation at break, yield force and toughness, in both

cases of linear PVA and crosslinked PVA. All the mechanical

parameters were found to be optimized at 5 wt.% of nanocellulose

incorporation. The thermal stability of the reinforced PVA

composite was reasonably higher than the neat PVA with a

simultaneous decrease in crystalline melting temperature. The

glass transition temperature (tg) of the crosslinked PVA composite

was however higher than that of the neat PVA. The FTIR spectral

analysis displays gradual narrowing of the 1060 cm1 peak

attributed to the COC pyranose ring of nanocellulose when

incorporated in composites with PVA. The XRD data indicate thepresence of 2 0 0 plane corresponding to nanocellulose (appearing

at 2u = 22.58) in the various composites with PVA. The barrier

property in terms of MVTR was found to be much improved in case

of PVA composite in relation to linear PVA. The AFM image shows

the presence of exclusive nano and micro particles of cellulose in

its composites with PVA. The SEM micrograph of the crosslinked

PVA composites differs from those of the linear PVA. The

crosslinked PVA composite exhibits rod shaped nanocellulose

particles while the linear once are of irregularly spherical droplets.

The TEM analysis corroborates our earlier findings from AFM

analysis.

Acknowledgement

We hereby acknowledge the assistance from The Calcutta

University for its help by the way of research fellowship from the

project Nanoscience and Nanotechnology and various other

infrastructures required for the research work.

References

[1] G.M. Whitesides, Small 1 (2005) 172.[2] F.N. Cogswell, International Polymer Processing 1 (1987) 157.[3] H.X. Nguyen, H. Ishida, Polymer Composites 8 (1987) 57.[4] H. Menendez, J.L. White, Polymer Engineering & Science 24 (1984) 1051.[5] I. Crivelli-Visconti, Polymer Engineering & Science 15 (1975) 167.

[6] T.M. Malik, P.J. Carreau, M. Grmela, A. Dufresne, Polymer Composites 9 (1988)412.

[7] S. Spoljaric, A. Genovese, R.A. Shanks, Composites Part A: Applied Science andManufacturing 40 (2009) 791.

[8] K. Kalaitzidou, H. Fukushima, L.T. Drzal, Composites Part A: Applied Science andManufacturing 38 (2007) 1675.

[9] N. Ljungberg, J.Y. Cavaille, L. Heux, Polymer 47 (2006) 6285.[10] M.A.S. Azizi Samir, F. Alloin, A. Dufresne, Biomacromolecules 6 (2005) 612.[11] S. Hill, New Scientist 153 (1997) 36.[12] R. Kozlowski, B.Mieleniak,in: L.Mattoso, A. Leao, E.Frollini (Eds.), Proceedings of

the 3rd International Symposiumon Natural Polymers and Composites (ISNaPol

2000),

Sao Pedro, Brazil, (2000), p. 504.[13] A.L. Leao, R. Rowell, N. Tavares, in: P.N. Prasad, J.E. Mark, S. Kandil, Z.H. Kafafi(Eds.), Science and Technology of Polymers and Advanced Materials, PlenumPress, New York, 1998, p. 755.

[14] B. Dahlke, H. Larbig, H.D. Scherzer, R. Poltrock, Journal of Cellular Plastics 34(1998) 361.

[15] A.S. Herrmann, J. Nickel, U. Riedel, Polymer Degradation and Stability 59 (1998)251.

[16] D.M. Panaitescu, D.M. Vuluga, H. Paven, M.D. Iorga, M. Ghiurea, I. Matasaru, P.Nechita, Molecular Crystals and Liquid Crystals 484 (2008) 86.

[17] A. Sturcova, J.R. Davies, S.J. Eichhorn, Biomacromolecules 6 (2005) 1055.[18] O. van den Berg, M. Schroeter, J.R. Capadona, C. Weder, Journal of Materials

Chemistry 17 (2007) 2746.[19] S.J. Chun,S.Y. Lee, G.H. Doh, S. Lee, J.H. Kim, Journal of Industrial andEngineering

Chemistry 17 (2011) 521.[20] J.R. Capadona, K. Shanmuganathan, D.J. Tyler, S.J. Rowan, C. Weder, Science 319

(2008) 1370.[21] M. Schroers, A. Kokil, C. Weder, Journal of Applied Polymer Science 93 (2004)

2883.

[22]

C. Bonini, L. Heux, J.Y. Cavaille, P. Lindner, C. Dewhurst, P. Terech, Langmuir 18(2002) 3311.

[23] C. Gousse, H. Chanzy, G. Excoffier, L. Soubeyrand, E. Fleury, Polymer 43 (2002)2645.

[24] J.M. Araki, M. Wada, S. Kuga, Langmuir 17 (2001) 21.[25] H. Yuan, Y. Nishiyama, M. Wada, S. Kuga, Biomacromolecules 7 (2006) 696.[26] M. Roohani, Y. Habibi, N.M. Belgacem, G. Ebrahim, A.N. Karimi, A. Dufresne,

European Polymer Journal 44 (2008) 2489.[27] C.A. Finch, Polyvinyl AlcoholDevelopments, 2nd Edition, Wiley, Chichester,

1992, p. 769.[28] M.J. Cho, B.D. Park, Journal of Industrial andEngineeringChemistry 17(2011) 36.[29] A. Mandal, D. Chakrabarty, Carbohydrate Polymers 86 (2011) 1291.[30] Y. Zhang, P.C. Zhu, D. Edgren, Journal of Polymer Research 17 (2010) 725.[31] ASTM, E96Standard Test Method forWaterVaporTransmission ofMaterials, vol.

04.06, American Society for Testing andMaterials, 1983.[32] H.P.A. Khalil, H. Ismail, H.D. Rozman, M.N. Ahmad, European Polymer Journal 37

(5) (2001) 1037.[33] P.S. Thomas, J.P. Guerbois, G.F. Russell, B.J. Briscoe, Journal of Thermal Analysis

and Calorimetry 64 (2001) 501.

[34] M.S. Peresin, Y. Habibi, J.O. Zoppe, J.J. Pawlak, O.J. Rojas, Biomacromolecules 11(2010) 674.

[35] A. Bhatnagar, M. Sain, Journal of Reinforced Plastics and Composites 24 (2005)1259.

[36] E.H. Qua, P.R. Hornsby, H.S.S. Sharma, G. Lyons, R.D. McCall, Journal of AppliedPolymer Science 113 (2009) 2238.

[37] M.M. Ibrahim, W.K. El-Zawawy, M.A. Nassar, Carbohydrate Polymers 79 (2010)694.

[38] S.K. Sen, Canadian Journal of Chemistry 41 (1963) 2346.[39] Y. Tsuchiya, K. Sumi, Journal of Polymer Science Part A: Polymer Chemistry 7

(1969) 3151.[40] R.K. Tubbs, T.K. Wu, PolyvinylAlcohol,Wiley,NewYork, 1973p. 167 (Chapter 8).[41] X. Jia, Y. Li, Q. Cheng, S. Zhang, B. Zhang, European Polymer Journal 43 (2007)

1123.[42] J. Sriupayo, P. Supaphol, J. Blackwell, R. Rujiravanit, Polymer 46 (2005) 5637.


G Model

http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0005http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0010http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0015http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0020http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0025http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0030http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0030http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0035http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0035http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0040http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0040http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0045http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0050http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0055http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0070http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0070http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0070http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0070http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0075http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0075http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0080http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0080http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0080http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0080http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0085http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0090http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0090http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0090http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0090http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0095http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0095http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0100http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0100http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0100http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0100http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0105http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0105http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0110http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0110http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0110http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0110http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0115http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0115http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0120http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0125http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0130http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0130http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0140http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0145http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0150http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0155http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0155http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0155http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0155http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0160http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0160http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0160http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0160http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0165http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0165http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0165http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0165http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0170http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0170http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0170http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0170http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0175http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0175http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0180http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0180http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0185http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0185http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0190http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0195http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0195http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0195http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0195http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0200http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0205http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0205http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0210http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0210http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0205http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0205http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0200http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0200http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0195http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0195http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0190http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0185http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0185http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0180http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0180http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0175http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0175http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0170http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0170http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0165http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0165http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0160http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0160http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0155http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0155http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0150http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0145http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0140http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0130http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0130http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0125http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0120http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0115http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0115http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0110http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0110http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0105http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0105http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0100http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0100http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0095http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0095http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0090http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0090http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0085http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0080http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0080http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0075http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0075http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0070http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0070http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0065http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0060http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0055http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0050http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0045http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0040http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0040http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0035http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0035http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0030http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0030http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0025http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0020http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0015http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0010http://refhub.elsevier.com/S1226-086X(13)00207-4/sbref0005

1. Studies on the Mechanical, Thermal, Morphological and Barrier Properties Based on PVA

Documents

Transcript of 1. Studies on the Mechanical, Thermal, Morphological and Barrier Properties Based on PVA