NON LINEAR VISCOELASTICITY AND EXTENSIONAL VISCOSITY OF LINEAR
MALEATED POLYPROPYLENE-CLAY NANOCOMPOSITES
By
Tanmay Janardan Pathak

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Chemical Engineering - Doctor of Philosophy
2014

ABSTRACT
NON LINEAR VISCOELASTICITY AND EXTENSIONAL VISCOSITY OF LINEAR
MALEATED POLYPROPYLENE-CLAY NANCOMPOSITES
By
Tanmay Janardan Pathak

Strain hardening of polymer melts in elongational flow is important for a variety of
polymer processing operations such as thermoforming, foaming and blow molding. As such the
measurement of transient extensional viscosity becomes important to understand the elongational
behavior of the polymeric materials. It is well known that linear polypropylene; a widely used
commodity polymer; does not strain harden in melt extensional flow and there arises a need to
structurally modify this polymer so as to achieve strain hardening.
In the present work, the effect of dispersing two different organoclays in linear maleated
polypropylene on the transient uniaxial extensional viscosity and shear viscoelasticity (linear and
nonlinear) has been investigated.

The matrix polymer was a linear random copolymer of

propylene and ethylene grafted with maleic anhydride (PP-g-MA). Two different grades of
nanoclay organically modified with different ionic surfactants were used as fillers. The
extensional rheology was carried out at different strain rates, on an Extensional Viscosity Fixture
(EVF) mounted on a TA-ARES rheometer. The neat PP-g-MA matrix did not show any strain
hardening at different extensional strain rates whereas the nanocomposites with different
organoclays showed different extents of strain hardening. Strain hardening was quantified by
calculating the ratio of transient extensional viscosity to the viscosity obtained from linear
viscoelastic envelope, called as the strain hardening parameter (χ) which was around 3 at lower
strain rates for one of the nanocomposite with 5wt% clay loading. Linear and non-linear shear

rheology of these nanocomposites was also investigated to understand the reinforcement effects
upon filler additions in the polymer matrix and to get an insight on the nanocomposite structure.
From the combined observations in linear and nonlinear shear rheology along with nonlinear
extensional rheology we were able to conclude that the polymer matrix chains physically
coupled (mainly by hydrogen bonding) to the nanoclays and these particle attached polymer
chains formed hindered chain entanglements with the free matrix chains. The formation of such
entangled structure led to improvements in the reinforcement levels as well as strain hardening
behavior in uniaxial extensional flow. The second part of this study is focused on understanding
the effects of filler loadings on the strain hardening and non-linear viscoelasticity of these
nanocomposites. This study showed that there lies an optimum level of nanoclay loading (2
vol%) with an effective aspect ratio and specific surface area that are necessary to form hindered
chain entanglements necessary for strain hardening. For composites with higher nanoclay
loadings, the strain hardening diminishes with decreasing effective aspect ratios.

Dedicated to my parents:
Dr. Janardan M. Pathak
&
Mrs. Bindu J. Pathak

iv

ACKNOWLEDGMENTS

I take this opportunity to sincerely thank Dr K. Jayaraman who is not only my doctoral
guide but also has been a mentor throughout my Ph.D. work and stay at MSU. He has provided
continuous support through these years and has always been available to clarify my queries. I
have been able to build a strong understanding of the fundamental concepts in the research work
with his help. I have been very fortunate in having a guide and a mentor like Dr. Jay. I also take
this opportunity to thank my committee members Dr. Baker, Dr. Lee, Dr. Drzal and Dr. Hawley
for serving on my committee and providing valuable guidance.
I thank my parents and take pride in this moment to acknowledge their continuous and
immense support and encouragement throughout my Ph.D. I have always sought inspiration and
patience in my doctoral research from my father Dr. Janardan Pathak and would like to thank my
mother Mrs. Bindu Pathak for her courage, sacrifice, and unconditional love. Without their
support, this task would not have been possible. I also thank my whole family without whose
support this would have been a daunting task.
I would like to express my deepest gratitude to my fellow colleagues, who helped me
learn the research techniques and skills in my Ph.D. I would especially like to thank Sharad, Giri,
John, Amit, Katie, Rahul, Susan, Madhu, Hemant, Devesh and Shreya. Finally yet significantly, I
would also like to acknowledge my closest buddies Nikunj, Malini, Madhuri, Akshay, Hemant,
Ritesh, Keyur, Dhruvi, Devang Arpi, Ashutosh, Priya, and a special thanks to Saurabh and
Anamika for being my extended family at MSU.

v

TABLE OF CONTENTS

LIST OF TABLES…………………………………………………………..............viii
LIST OF FIGURES……………………………………………………….……...….ix
CHAPTER 1……………………………………………………………………….…1
INTRODUCTION…………………………………………………………………....1
1.1
BACKGROUND………………………………………………………......1
1.2
RESEARCH OBJECTIVES……………………………………………...11
CHAPTER 2…………………………………………………………………………13
EXPERIMETAL TECHNIQUES FOR PREPARATION AND CHARACTERIZATION
OF NANOCOMPOSITES………………………………………………………….13
2.1
2.2
2.3

MATERIALS………………………………………………………….....13
COMPOUNDING OF NANOCOMPOSITES………………………..…13
COMPRESSION MOLDING………………………………………........15
2.3.1 Shear Rheology and XRD Samples…………………………………..15
2.3.2 Elongational Rheology Samples……………………………………..16
2.4
X RAY DIFFRACTION………………………………………………....17
2.5
THERMOGRAVIMETRIC ANALYSIS……………………………......19
2.6
RHEOLOGY…………………………………………………………….20
2.6.1 Shear Rheology…………………………………………….………...21
2.6.2 Extensional Rheology……………………………………………......24
CHAPTER 3……………………………………………………………………….29
RHEOLOGY OF MALEATED POLYPROPYLENE-CLAY NANOCOMPOSITES
VARYING THE CLAY ASPECT RATIO AND ORGANIC CONTENT……..29
3.1
INTRODUCTION……………………………………………………....29
3.2
EXPERIMENTAL METHODS.……………………………………......33
3.2.1 Materials……………………………………………………………33
3.2.2 Compounding.……………………………………………………....34
3.3
CHARACTERIZATION TECHNIQUES……………………………...34
3.3.1 DSC and TGA Measurements…………............................................34
3.3.2 X ray Diffraction Measurement…………………………………….35
3.3.3 Extensional Viscosity Measurements…………………………….....35
3.3.4 Shear Rheology…………………………………………...………...35
3.4
RESULTS AND DISCUSSION………………………………….…….36
3.4.1 Effect of nanoclay dispersion and stability in nanocomposites..…...36
3.4.2 Oscillatory Shear rheology………………………………………....40
3.4.2.1 Linear Viscoelastic Measurements………………………….….40
3.4.2.2 Non-Linear Viscoelastic Measurements…………………..……44
3.4.3 Uniaxial Extensional Rheology…………………………………......52
3.4.3.1 Effect of filler addition……………………………………..…...52
vi

3.5

3.4.3.2 Effect of surfactant head group and effective aspect ratio..…...…54
CONCLUSIONS…………………………………………………..…......57

CHAPTER 4……………………………………………………………………...…59
EEFECT OF LOADING AND TEMPERATURE ON NON LINEAR RHEOLOGY OF
MALEATED LINEAR POLYPROPYLENE-CLAY NANOCOMPOSITES.....59
4.1
INTRODUCTION AND BACKGROUND……………………………..59
4.2
EXPERIMENTAL TECHNIQUES………………………………..….…62
4.2.1 Materials……………………………………………………………..62
4.2.2 Compounding………………………………………………………...62
4.3
CHARACTERIZATION TECHNIQUES…………………………….....63
4.3.1 DSC and TGA Measurements………………………………………..63
4.3.2 X ray Diffraction Measurement………………………………….…..63
4.3.3 Extensional Viscosity Measurement…………………………………63
4.3.4 Shear Rheology……………………………………………………....64
4.4
RESULTS AND DISCUSSIONS………………………………………..65
4.4.1 Dispersion of nanoclay and thermal stability of nanocomposites…...65
4.4.2 Oscillatory Shear Rheology………………………………………….68
4.4.2.1 Linear Viscoelastic Measurements……………………………....68
4.4.2.2 Non-Linear Viscoelastic Measurements…………………...….....72
4.4.2.3 Modulus recovery after strain sweep………………………….....77
4.4.3 Uniaxial Extensional Rheology……………………………………...79
4.4.3.1 Effect of filler addition and test temperature………………….....79
4.4.3.2 Effect of clay loading on the extensional viscosity of
nanocomposites ……………………………………………….....82
4.4.3.3 Estimate of particle dimensions and inter-particle spacing….…...85
4.5 CONCLUSIONS..……………………………………….………………..89
CHAPTER 5………………………………………………………………….....91
CONCLUSIONS AND RECOMMENDATIONS………………………….....91
5.1
CONCLUSIONS……………………………………………………........91
5.1.1 Addition of nanoclay and its effect on viscoelastic properties
and strain hardening…………………………………………….…...91
5.1.2 Effect of surfactant head group and effective aspect ratio…..…….....92
5.1.3 Effect of test temperature and nanoclay loading………………..……94
5.2 RECOMMENDATIONS………………………………………………..……95
BIBLIOGRAPHY…………………………………………………………..…...98

vii

LIST OF TABLES

Table 4.1:

Details of Aspect ratio Estimation for Nanoparticles in Various Composites….71

Table 4.2:

Estimates of Particle Dimensions and Packing..………………………………..86

viii

LIST OF FIGURES

Figure 2.1.

Representative DSC scans for melting transitions
of maleated polypropylene nanocomposites:
(▬) Nanocomposite with 5 wt% of I.30P clay,
() Nanocomposite with 5 wt% of I.44P clay………….…………..15

Figure 2.2.

Schematic of a parallel plate rheometer...............................................21

Figure 2.3.

Schematic of a Elongational Viscosity Fixture (EVF).........................26

Figure 3.1.

X-ray diffraction of organoclay: a) I.30 P clay (clay A),
b) I.44 P clay (clay B)………..............................................................37

Figure 3.2.

X-ray diffraction of composites: a) A5 nanocomposite,
b) B5 nanocomposite……………………………………………........38

Figure 3.3.

Thermo gravimetric Analysis (TGA) curves obtained for
neat matrix, A5 and B5 nanocomposites at
10°C / min up to 600°C.………...........................................................39

Figure 3.4.

Time Sweep test: Change in normalized Storage modulus
curves wrt time for the neat matrix (●),
A5 nanocomposite (*) and B5 nanocomposite (∆) .............................40

Figure 3.5.

Dynamic viscosity curves for the neat matrix (●),
A5 nanocomposite (*) and B5 nanocomposite (∆)...........……………41

Figure 3.6.

Dynamic Storage Modulus curves for the neat matrix (●),
A5 nanocomposite (*) and B5 nanocomposite (∆)........…………..….43

Figure 3.7.

Subsequent strain sweep tests. Variation of storage modulus
with strain amplitude for A5 nanocomposite at 155°C and 1 Hz…...45

Figure 3.8.

Subsequent strain sweep tests. Variation of storage modulus
with strain amplitude for B5 nanocomposite at 155°C and 1 Hz……46

Figure 3.9.

Subsequent strain sweep tests. Variation of loss modulus
with strain amplitude for A5 nanocomposite at 155°C and 1 Hz…...47

Figure 3.10. Subsequent strain sweep tests. Variation of loss modulus
with strain amplitude for B5 nanocomposite at 155°C and 1 Hz……48
Figure 3.11. Storage modulus recovery versus time (z and γ = 0.5%).
ix

A5 and B5 nanocomposites at 155°C. G’ (t) is the storage
modulus value as obtained from time sweep test which
is done immediately after one strain sweep..........…………..………49
Figure 3.12. Loss modulus recovery versus time (z and γ = 0.5%).
A5 and B5 nanocomposites at 155°C. G” (t) is the Loss
modulus value as obtained from time sweep test which
is done immediately after one strain sweep.........................................50
Figure 3.13. Rate of Recovery of the loss factor after initial strain sweep
for PPNC A5 and PPNC B5. The Loss Factor LF (t) is
that obtained from the time sweep test and LF0 is the plateau
value of loss factor obtained from initial strain sweep test…………..51
Figure 3.14. Transient extensional viscosity curve for maleated
polypropylene matrix………………………………….……………..52
Figure 3.15. Transient extensional viscosity curve for nanocomposites
A5 and B5 at 155° C and four extensional
strain rates: 0.1s-1 (∆), 0.5s-1(□), 1s-1(◊) and 2s-1(○)......…………..….53
Figure 3.16. Reduced Strain hardening parameter v/s strain rate:
Comparison at Hencky strain of 2.25 for PPNC –A5 and B5
at 155° C………………………………………………………….…..56
Figure 3.17. Schematic of the interaction between the PP-g-MA chains
and organoclay…………………………………………………….….57
Figure 4.1:

X-ray diffraction of organoclay: a) I.44 P clay, b) B5 nanocomposite
(5wt% clay), c) B8 nanocomposite (8 wt% clay), d) B12 nanocomposite
(12 wt% clay)…………………….……….........………………….….65

Figure 4.2:

Thermo gravimetric Analysis (TGA) curves obtained for pure clay,
neat matrix, and nanocomposites at 10°C / min up to 600°C..............67

Figure 4.3:

Comparison of the increase in storage modulus for the matrix and
B5 nanocomposite at 160° C for 1 hour...............................................68

Figure 4.4:

Complex viscosity curves for the neat matrix (●) and nanocomposites:
PPNC B5 (∆), PPNC B8 (□), and PPNC B12(×)………………….…69

Figure 4.5:

Storage Modulus (G’) viscosity curves for the neat matrix (●) and
nanocomposites: PPNC B5 (∆), PPNC B8 (□), and
PPNC B12(×)…………………………………….………………..…70

Figure 4.6:

Strain sweep tests. Variation of storage modulus with strain amplitude
x

for PPNC B5 (●), PPNC B8 (∆), and PPNC B12(□) at 155°C
and 1 Hz………………………………………………………………73
Figure 4.7:

Strain sweep tests. Reduced Storage Modulus versus strain amplitude
for PPNC B5 (●), PPNC B8 (∆), and PPNC B12(□) at 155°C
and 1 Hz………………………………………………………….…..74

Figure 4.8:

Strain sweep tests. Variation of loss modulus with strain amplitude
for PPNC B5 (●), PPNC B8 (∆), and PPNC B12(□) at 155°C
and 1 Hz………………………………………………………………75

Figure 4.9:

Critical Strain Plot. Variation of critical strain with respect to Specific
Surface Area (Asp) per gram of clay....................................................76

Figure 4.10: Storage modulus recovery versus time (z and γ = 0.5%).
PPNC B5, PPNC B8 and PPNC B12 nanocomposites at 155°C.
G’ (t) is the storage modulus value as obtained from time
sweep test which is done immediately after one strain sweep……....78
Figure 4.11: Rate of Recovery of the loss factor after initial strain sweep
for PPNC B5, PPNC B8 and PPNC B12. The Loss Factor LF (t)
is that obtained from the time sweep test and LF0 is the plateau
value of loss factor obtained from initial strain sweep test…………..79
Figure 4.12: Transient extensional viscosity curve for nanocomposite B5
at 155° C and 160° C at four extensional strain rates:
0.1s-1 (∆), 0.5s-1(□), 1s-1(◊) and 2s-1(○)......…………………….……..80
Figure 4.13: Reduced Strain hardening parameter v/s strain rate:
Comparison at Hencky strain of 2.25, for
PPNC B5 at 155 ° C (∆) and at 160° C (♦)………................….…….81
Figure 4.14: Transient extensional viscosity curve for nanocomposite PPNC B5,
PPNC B8 and PPNC B12 at 155° C and four extensional strain rates:
0.1s-1 (∆), 0.5s-1(□), 1s-1(◊) and 2s-1(○)......……………………….......83
Figure 4.15: Reduced Strain hardening parameter v/s strain rate:
Comparison at Hencky strain of 2.25, for PPNC B5 (∆) and
PPNC B8 (□) at 155 ° C…………………………................………...84

xi

INTRODUCTION
________________________________________________________________________
CHAPTER 1
1.1

BACKGROUND
Recently, the polymer layered silicate nanocomposites have been widely developed for

various applications in automotive industries, for food packaging films, etc. In the packaging
industry polymer composites have gained importance because of their improved barrier
properties as well as flame retardancy [1-6]. In the automotive industry polymer nanocomposites
are preferred widely to make automotive bumpers, car interiors such as dash board panels, hard
and soft instrument panels, trims etc. because of their low density, light weight and mold ability
as compared to traditional micro composite systems and metal parts. For such applications the
nanocomposites are processed through various processes such as extrusion foaming,
thermoforming, blow molding, etc. where high melt strength of the polymer nanocomposite
system is preferred. The study of the extensional flow behavior of these polymer nanocomposite
melts is therefore very important for such processes.
Thermoplastic polymer-layered silicate nanocomposites first came into light in the mid
1990’s after the initial research work done by Toyota researchers on nylon/clay nanocomposites
[7,8]. Since then there has been considerable research interest in making these nanocomposites
more viable for the processing industry. The addition of small amount of high aspect ratio (of the
order of ~200) nanoclay platelets to polymer matrix, through melt mixing, leads to significant
improvements in the polymer/clay nanocomposite properties such as stiffness, tensile strength,
barrier properties as well as melt strength [9-12]. Polypropylene (PP) which is a commodity
1

polymer has been widely investigated to make its nanocomposites with clay. PP being non polar
needs to be functionalized with maleic anhydride to form maleated PP so as to wet the clay
surface more readily and mix properly for making the nanocomposites . Okada and co-workers
were the first to make nanocomposites out of PP, maleated PP and clay [13-15]. They dissolved
the functionalized polyolefin oligomer with clay in toluene such that the polymer intercalates in
the clay galleries. The toluene was then evaporated and the intercalated mix was melt blended
with the bulk PP. The functionalized polymers that facilitate the interactions between the clay
particles and the bulk PP matrix are generally called as compatibilizers. Literature also reports
the mixing of all three, the compatibilizer, the clay as well as the bulk polymer by directly melt
mixing them [15, 16]. The presence of a compatibilizer with a polar functional group facilitates
the delamination of the clay platelets and eases the dispersion of the non polar polymer chains in
the clay galleries has been shown for a wide range of polymers [17-22]. Hence in the present
study maleated polypropylene is used as the entire polymer matrix to which montmorillonite clay
is added. The primary objective in using the maleated PP alone was to understand the kind of
interactions between the polymer and clay particles and to avoid the complexity of adding a third
phase which would have been non functionalized bulk polypropylene. It has been established in
literature that functionalization of polypropylene with maleic anhydride leads to its chain
scissions [23] which reduces the molecular weight. A maleated polypropylene {Exxelor PO
1015} was therefore chosen in this study which has a fairly decent amount of maleic anyhride
functionality (0.4 wt%) and a molecular weight (Mw) of around 180,000 and to this a
montmorillonite clay was added.
Montmorillonite is a widely used layered silicate for enhancing the properties of polymer
composites. The clay in its natural form has layered platelets which form stacks. The clay
2

platelet has a thickness of around 1nm and a length dimension of around 100-300 nm thereby
giving a high aspect ratios particle and also an interlayer spacing of around 1nm. They provide a
2

surface area of around 700 m /g [24, 25]. The layered sheets in montmorillonite clay have an
+

excess negative charge which is neutralized by highly charged inorganic cations such as Na and
2+

Ca

present in a hydrated form in the interlayers of untreated clay [1, 26]. The hydration

spheres of these inorganic cations form unfavorable organophobic conditions for adsorption of
organic polymer molecules between the interlayer. Also, the removal of water of hydration of
these cations leads to collapsing of the interlayer spacing thereby preventing the entrance of the
organic polymer chains between the interlayer spacing and exfoliation of clay in polymer matrix.
The clay surface is therefore treated so as to make the clay interlayer region organophilic and
make the adsorption of organic polymer molecules between the clay platelets easier [11, 27-31].
The polymer chains can then enter the clay galleries between two subsequent clay layers and
separate them causing exfoliation of clay in the polymer. The cation exchange capacity of clay is
very important in determining the effectiveness with which it can be treated with organophilic
surfactants and also the level of exfoliation that can be achieved when mixed with a polymer.
There has been a lot of work done and reported in literature on achieving exfoliation of fillers in
polymer matrices by either using different minerals, varying the cation exchange capacity,
surfactant chemistry as well as reacting the polymer precursor first with the clay and then
polymerizing [27-37].
It is well reported in literature so as to make the clay interlayer region organophilic, it is
first intercalated with surfactants which have long chain alkyl ammonium ions, wherein the ion
is tethered to the clay surface and the alkyl tails swell the clay improving its dispersion in the
3

polymer. This surface treatment of clay is done by exchanging the hydrated cations with
quaternary organo-ammonium salts or with primary amines having hydrocarbon groups attached
to them, thereby creating organophilic condition in the clay interlayer region. Heinz and coworkers [38] have investigated the structure and dynamics of alkyl-ammonium modified clays
having different cation exchange capacity, different head groups (quaternary : NMe3

+

-R and

+

primary: NH3 -R), as well as different alkyl chain lengths (C4 to C22). They have also studied
the effect of these variables on the cleavage energy required to delaminate the clay platelets to
achieve exfoliation [39].
In the present study, organoclays with two different surfactant chemistries are used for
making the PP – clay nanocomposites. One grade of montmorillonite clay has a primary amine
ion surfactant with a single alkyl (C18) tail and another one has a quaternary ammonium ion
surfactant with two alkyl (C18) tails in the clay interlayer galleries. Heinz et al. [38] have
commented on the structure and conformation of these surfactants in the interlayer galleries.
According to them, both the surfactant head groups are attached to the clay face ionically,
however the primary amine ion surfactant head can have three additional hydrogen bonds (N-H--O-Si). As such this surfactant with a C-18 tail is hydrogen bonded more tightly to the clay
surface and tends to orient itself perpendicular to the clay surface. The quaternary ammonium
ion surfactant with two C-18 tails has larger head group size, flexibly bonded and is more mobile
in the interlayer galleries; it also packs to a greater extent in the interlayer galleries.
These organoclays were melt mixed at different loadings with linear maleated
polypropylene and different nanocomposites were made, in the current study. These

4

nanocomposites were tested for nonlinear, uniaxial extensional melt viscosity transients to check
if the addition of the organically modified nanoclay imparted any melt strength to the non strain
hardening linear polypropylene matrix. Also, the effect of increasing the test temperature and
clay loadings on the melt strength of these nanocomposites has been investigated. To further
probe into the non-linear extensional behavior of the nanocomposites, linear as well as non-linear
shear rheology was used as an investigative tool.
As mentioned earlier, polymeric materials, blends and composites are used in a wide
range of daily applications and products and this demand is ever increasing. For the
manufacturing of these daily commodities, the polymers need to be processed through different
processes such as thermoforming, blow molding, foaming, etc. The polymeric materials while
undergoing through these processes experience mainly extensional and shear flow. Also these
processes require the material to be processed at higher temperatures and show good melt
strength so that the final product shows good properties. For this reason, it becomes imperative
to study the transient extensional behavior and melt strength of the polymers, polymer blends
and polymer composites. There is an abundance of literature on the extensional flow behavior of
different polymeric blends, crystalline polymers, amorphous polymers, food items, polymer
composites etc [40-67]. As early as 1977, Agrawal et.al [40] worked with several different
polymer melts and measured their shear and extensional flow. Wagner [41] measured the
uniaxial, biaxial and ellipsoidal extensions on a polyisobutylene melt and compared the data to
predictions from Doi-Edwards Model. He found out that the model needed more modifications to
accurately predict the transient extensional data. Wilson and Baird [43] worked with
thermotropic liquid crystalline polymer melts and measured their transient extensional viscosity.
They found out that the extensional data followed linear viscoelastic behavior at low strains and
5

showed mild strain hardening with increasing strains. Elongational flow behavior has also been
investigated for daily use items such as viscoelastic liquid foods [45], ceramics [51], and
polymeric blends such as LLDPE/LDPE [47, 55] as well polymeric composites [48,54,61,67,68].
It however is necessary to understand the relevance between the transient elongation
measurements and the actual industrial processes in which the materials experience such
elongational flows. Micic and Bhattacharya [47, 55] have shown in their work on LLDPE/LDPE
blends, the relevance between the shear and elongational measurements and film blowing
process. Foaming is another process where in the transient extensional measurements can give
insight on how the material will foam. Bhattacharya and co-workers [67] as well as Jayaraman.K
and Chaudhury.A [68] have shown that polypropylene nanocomposites that show better strain
hardening also foam well. These foams in general show closed foam cells as well as a higher cell
density and lower bulk density. Foaming studies and their relation to strain hardening behavior in
polypropylenes has also been shown by Spitael and Macosko [69] as well as Munstedt et.al [70].
Previously, measurements of transient uniaxial extensional viscosity [69-72] and of melt strength
[73, 74] of pure polymer melts and their blends have been done to understand their strain
hardening behavior. In the current research study, we focus on the “strain hardening” behavior
of linear polypropylene and polypropylene nanocomposites that is observed in uniaxial
extensional flow. In particular, a rise in polymer melt viscosity (above the linear viscoelastic
limit) with increasing Hencky strain is called as “strain hardening” and it is desirable to a certain
extent for processing operations. Linear polypropylene, however, does not display any strain
hardening and therefore it becomes necessary to modify the molecular structure or incorporate
components in the linear polypropylene so as to achieve strain hardening in melt extensional
flow.
6

Sugimoto and coworkers have shown that linear polypropylene on its own does not
display any strain hardening. However they were able to show strain hardening in polypropylene
which was modified either by introducing branched molecular structure through irradiation or by
adding di-2-ethylhexyl peroxy dicarbonate (EHPC) treated polymer [75]. By irradiation or by
adding the EHPC polymer they were able to obtain high melt strength PP with long chain
branches which showed improved strain hardening behavior. Similar observations of good strain
hardening by long chain branches have been reported elsewhere [77, 78]. In another research
study Sugimoto and co-workers were able to achieve good strain hardening in linear PP by
adding a small amount of uhmw–PE to linear PP [76]. In that study, they used a Meissner type
rheometer to measure the uniaxial and biaxial extension of the modified PP blend and found out
that this modified PP having long chain PE showed enhanced strain hardening beyond a critical
strain. Similar extensional flow studies on blends of polymers have been done by Yamaguchi
et.al [79-81]. They modified linear polypropylene by adding a small amount of a crosslinked
terpolymer {gEHDM} characterized as gel just beyond the sol-gel transition point and the blend
showed improved strain hardening behavior. They suggested that with appropriate cross link
density the stretching of trapped chains between the crosslinking points were responsible for the
strain hardening and an increase in the crosslinking density leads to suppression of the strain
hardening behavior. Strain hardening of crosslinked polymers such as poly methyl
(methacryalate) has also been reported in literature by Takahashi.M and Ogura [82] and for
sparsely crosslinked PDMS by Takahasi.H and Watanabe [83]. They both have reported that
with appropriate cross link density, good amount of strain hardening can be achieved in
polymeric systems and that excessive cross link density can hinder the strain hardening property.

7

A similar analogy has been drawn in the current work involving nanocomposites having physical
crosslinks between polymer chains and organically modified nanoclay particles.
The extensional flow of composites with micron and nano sized fillers has also been
investigated and reported in literature. Micro composites with different sizes, shapes and aspect
ratios were made and tested for extensional flow by Takahashi et al. [84] and Le Meins et al [85].
Takahashi used microscale fillers such as mica, talc, glass beads and glass flakes in a low density
polyethylene matrix, whereas Le Meins et al. suspended monodisperse polystyrene spheres in
polyisobutylene liquid. They found that the strain hardening goes down with filler addition and
even more so with high aspect ratio particles such as fibers. They also observed that the
formation of aggregates led to a decrease in strain hardening behavior. Similar observations of
decreasing strain hardening have also been reported by Bhattacharya and coworkers [66, 86]. In
one of his studies he has mixed talc with polypropylene and found out that the melt strength and
drawability of these composites goes down with filler loadings. In his other work, he mixed
nanoclay to ethylene vinyl acetate copolymer (EVA) and found that the strain hardening of the
nanocomposites matches with that of the unfilled EVA beyond a certain strain. The drop in the
strain hardening for filled EVA systems beyond a certain strain was attributed to the
restructuring of the clay platelets in a way that slows down the strain hardening. There have
been couple reports, however, with nylon 6 [87] and maleated polypropylene [88] composites
wherein the strain hardening improved slightly upon addition of nano size filler particles having
high aspect ratio. Seong and coworkers [87] dispersed Closite nanoclay in Polyamide 6 matrix
and observed that with well dispersed systems there was an improvement in the storage modulus,
viscosity as well as the strain hardening behavior in extension. Okamoto et al. [88, 89] dispersed
treated montmorillonite clay in maleated polypropylene and investigated the strain hardening as
8

well rheopexy features of the nanocomposites. They observed an improvement in the strain
hardening behavior of these nanocomposites and attributed it to the perpendicular alignment of
the clay platelets to flow direction, forming a “house of cards” structure. They made similar
observations in polylactide-layered silicate nanocomposites [90].
There have been several attempts and many theories that are used in literature to explain
the possible causes of strain hardening behavior in different systems such as polymer melts,
polymer blends and polymer composites. Sugimoto et al. [76] and Wagner et al. [91] have
observed that larger relaxation times lead to greater extents of strain hardening. Larger relaxation
times would be associated with stronger entanglements. In general the presence of entanglement
structure within the polymer causes a slow relaxation of the polymer chains upon deformation
and longer the relaxation time scale greater is the lift off in extensional viscosity which is
nothing but strain hardening in melt extension. Several authors have made attempts to predict the
extensional behavior and strain hardening based on shear viscoelastic data. Measurement of
shear viscoelastic data both in linear and non-linear shear rheology is very important to
understand the flow behavior and the molecular structure of the polymeric system. Especially in
polymeric composite systems, non-linear shear rheology along with shear viscoelastic data can
give useful insight of the reinforcements brought about by fillers.
It is well know that addition of filler particles in a polymer leads to greater modulus
values and more elastic behavior in shear thereby showing greater reinforcement. It is reported in
literature that at higher filler loadings, improvement in the reinforcement levels is a result of
filler network formation and filler agglomeration [92-94]. Also in non-linear shear, with
increasing strain the network formation breaks down causing a drop in the measured modulus
values which is called as the “Payne effect”. However the mechanisms behind these
9

reinforcements and nonlinearity with increasing strain, at lower filler loadings, are still an area of
considerable research. Sternstein and co-workers have tried to understand these mechanisms by
which the nano-scale fillers promote both reinforcement and nonlinear viscoelastic behavior in
polymer melts [95, 96]. They have worked with fumed silica fillers, at various loadings, having
different surface treatments and poly (vinyl acetate) matrix with different molecular weights and
explored the nonlinear vsicoelasticity of these nanocomposites. They have proposed that the
reinforcement and nonlinear viscoelastic behavior of the nano filled polymers is a result of
“trapped entanglements” which occur because of interactions between the filler and the polymer
chains at the filler-polymer interface. Also, when large strain amplitudes are applied, these
trapped entanglements are lost because of the debonding of polymers chains from the filler
surface which results in overall relaxation of the matrix entanglement structure. This loss of
entanglement structure causes non linearity in the viscoelastic behavior of the nanocomposites
[95].
Cassagnau and Melis [97] have worked with fumed silica composites with different
polymer systems such as PP, PS and EVA and carried subsequent strain sweep tests and time
sweep tests to understand the non-linear viscoelastic behavior of these composites and modulus
recovery kinetics. Contrary to Sternstein’s work, they proposed that the non-linear behavior of
the filled polymer composites can be associated with both mechanisms of chain
disentanglements and breakdown of filler network depending on filler concentration and
amplitude deformation. With the actual mechanisms, that cause improved reinforcements, nonlinearity in shear and improved strain hardening behavior, still being obscure or unclear for
polymeric composites, it becomes necessary to explore the correlation if any between these three
measurements for the polypropylene-layered silicate nanocomposites made in this study.
10

1.2

RESEARCH OBJECTIVES
The present study revolves around several objectives that need to be achieved in order to

obtain better melt strength and strain hardening in linear polypropylene system and understand
the possible mechanism behind it. One of the primary objectives, therefore, is to achieve the
strain hardening by incorporating inorganic filler in the linear polypropylene system and testing
these nanocomposites for nonlinear, uniaxial extensional melt viscosity transients. This research
is also motivated by the need for a better understanding of the interactions between the polymer
matrix and the nano-filler which is montmorillonite clay with different surfactant treatment in the
interlayer galleries. Also for this purpose maleated polypropylene alone was used as the entire
matrix because in bulk polypropylene composites, the maleated PP acts as a functionalized
compatibilizer and serves to provide favorable interactions at the clay faces. In addition to
measuring transient extensional behavior, linear and non-linear shear viscoelasticity is also
investigated for these nanocomposites to better understand the reinforcement mechanisms and
get an insight of the possible polymer – filler structure. While investigating the structure –
property relationships for the nanocomposites, there are several factors that are varied to
understand their effects on the morphology as well as the melt extensional and shear rheology.
The different research objectives thus aimed to be explored though this study includes:
1) Preparation and characterization of polypropylene-layered silicate nanocomposites with
different organoclays. These nanocomposites will be tested for their strain hardening property in
melt extension as well viscoelasticity.
2) Investigation of the effects of varying surfactant treatment in clay interlayer galleries and
aspect ratios on the nanoclay dispersion, thermal stability, linear and non-linear shear rheology
11

and uniaxial elongational flow of nanocomposites. This is mainly done in the first part of the
research wherein two different organically modified nanoclays are incorporated in the linear
maleated polypropylene matrix. The final motive behind this study is to understand the possible
mechanisms and molecular interactions that would help improve the measured reinforcement and
strain hardening properties as compared to the neat polymer matrix.
3) Investigation of the effects of increasing the clay loading on the level of reinforcement and
strain hardening in these nanocomposites. This is an important part of the second phase of this
research. It aims at understanding the effects of varying the filler volume fractions dispersed in
the matrix. A similar set of experiments as done in the first part are done for the nanocomposites
with different nanoclay loadings. Once again this study focuses on combining the observations
made in shear and elongational flows to get an idea about the possible structural arrangements of
the layered silicates and polymer chains and the interactions between them.
4) Investigation of the effects of increasing the test temperature on the level of strain
hardening in these nanocomposites. This is also a part of the second phase of the research work
where in the strain hardening parameters will be calculated for the nanocomposite and compared
at different extensional test temperatures.

12

EXPERIMETAL TECHNIQUES FOR PREPARATION AND CHARACTERIZATION OF
NANOCOMPOSITES
________________________________________________________________________
CHAPTER 2
2.1

MATERIALS
The matrix material used in the present study was a functionalized polypropylene having

maleic anhydride as the functional group and was obtained from ExxonMobil under the trade
name of Exxelor PO 1015. The amount of maleic anhydride grafted on to the linear
polypropylene was about 0.4 wt%. This maleated polypropylene (PP-g-MA) had a molecular
weight of about 70,000. Maleated polypropylene is generally used as a compatibilizer in making
bulk polypropylene clay nanocomposites so as to effect the exfoliation of clay. The choice of
using a maleated polypropylene as the whole matrix was specifically made to understand the
polymer filler interactions while observing the melt strength properties of these nanocomposites.
Two different clays were used in preparing the nanocomposites: A) Nanomer I.30P which
had a primary ammonium ion surfactant with a single alkyl tail in the galleries and; B) Nanomer
I.44P which had quaternary ammonium ion surfactant with two C-18 tails. These clays were
obtained from Nanocor.
2.2

COMPOUNDING OF NANOCOMPOSITES
The nanocomposites for this study are all compounded using a batch melt mixing

technique. This technique was used to provide enough shear and residence time for good mixing,
to have lesser contamination and also to minimize the degradation of maleated polypropylene. A
13

mixing bowl with 75cc volume and two Banbury impellers from CW Brabender Instruments, Inc
was used for this purpose. Previous studies involving preparation of polymer nanocomposites
have used small scale batch mixers like the Banbury [22, 98-100] and Gelimat [101] with shear
rates around 200s

-1

and mixing time between 5 to 10mins. The mixing bowl was purged with

nitrogen during mixing so as to minimize polymer degradation. Park [102] and coworkers have
also used nitrogen purge during mixing of polypropylene nanocomposites.
The first step before melt mixing maleated polypropylene pellets and clay powder was to
dry them overnight in a vacuum oven at 80°C and 20 mm Hg vacuum. This was done to ensure
that moisture does not degrade the polymer nanocomposites while mixing. The dried materials
were then shaken vigorously in a bag for about 2 minutes to have a homogenized powder-pellet
mixture. This mixture was then melt blended in a pre-heated at a temperature of 165°C at 150
rpm for 10 minutes. The mixing speed was determined by the maximum speed on the mixers
drive such that maximum possible shear is imparted into the material and 10 minutes of mixing
time is sufficient to ensure the complete melting and mixing of polymer pellets with the nanoclay
powder. The mixing speed and time on Banbury mixer for compounding polypropylene clay
nanocomposites have been optimized through previous work in our lab. The mixing temperature
of 165°C was chosen to be well above the melting point of the maleated polypropylene as has
been observed through DSC scans, run on the pellets. Figure 2.1 shows representative DSC scans
run on the nanocomposites which show that the melting transition for the materials is around
145°C. A high mixing temperature also provides enough polymer melt viscosity necessary to
generate shear stresses that are sufficient to separate the clay platelets and obtain complete
mixing. One batch of the mixer amounted to 47 grams of the compound. The compounded
material was then removed from the mixing bowl and impeller screws with the help of a brass
14

scraper to form small clumps. These clumps were then directly used to make samples through
compression molding, for rheological analysis as well as X-ray diffraction measurements.

0

-1

NC with 5 wt% I.30P clay
NC with 5 wt% I.44P clay

Heat flow (mW)

-2

-3

-4

-5

-6
50

80

110

140

170

200

Temperature (degree Celcius)

Figure 2.1.

Representative DSC scans for melting transitions of maleated polypropylene

nanocomposites: (▬) Nanocomposite with 5 wt% of I.30P clay, () Nanocomposite with 5 wt%
of I.44P clay.
2.3

COMPRESSION MOLDING

2.3.1 Shear Rheology and XRD samples
The compression molding of shear rheology and XRD samples was done in Wabash
Compression Molder at a temperature of 175°C, which is well above the melting transition of the

15

nanocomposites. A mold with 4 square cavities was used for the compression molding
procedure. Small clumps of compounded nanocomposite material were directly placed in these
cavities, the weight of which was calculated on the basis of volume of one mold cavity times the
density of the material taken as 0.9 g/cc. The thickness of the cavity was about 1 mm. The mold
filled with the material was covered on both sides with polyimide release films and this was then
put in between 2 thick steel plates. This whole mold assembly was then placed in between the
pre heated compression molder platens and the clumps of nanocomposites were allowed to melt
completely such that there was no gap left between the mold and the steel plates. After this the
platens were closed and initially a low pressure of about 2 tons was applied for 5minutes. This
was followed by the next step wherein the tonnage pressure was increased to 10 tons and kept
there for another 5 minutes. At the end of this, valves of cooling water to the platens were
opened so as to cool down the whole assembly to room temperature at the 10 ton pressure. Once
the room temperature was reached the pressure was released and the mold assembly was taken
out of the compression molder. Samples obtained from the 4 cavities were then used for XRD
and shear rheology characterization. For shear rheology, 40 mm diameter disks were cut out
from the square samples and used for characterization.
2.3.2 Elongational Rheology samples
The procedure used for compression molding samples for elongational rheology is
different from that used for shear rheology samples. A 20-specimen mold was used for
compression molding samples that were 18mm long, 10mm wide and 0.7mm thick for the
elongational theology tests. Once again required weight of the nanocomposite was calculated on
the basis of volume of a single cavity times total number of the cavities times the density of the
material (0.9 g/cc). The mold was covered with polyimide release films and then placed between
16

two thick steel plates. This mold assembly was then transferred between the compression molder
platens preheated to 175°C. Initially, the clumps of nanocomposite were allowed to melt down
completely into the mold. The tonnage pressure used for compression molding elongational
rheology samples was different than that for shear rheology samples. The first step required
closing the platens such that a 0 ton pressure was applied to the mold assembly for 5 minutes.
After this the tonnage pressure was increased to 5 tons for additional 5 minutes. The platens were
then cooled to room temperature using cooling water and the samples were gently removed from
the mold.
2.4

X RAY DIFFRACTION
All the nanoclays were investigated for their interlayer spacing using X-ray diffraction

(XRD) patters. XRD was also used to investigate the extent of intercalation or exfoliation of clay
in the nanocomposites depending on the shoulder observed in their XRD pattern. X ray
diffraction is a method of determining the arrangement of atoms in crystals by diffracting an Xray beam through the crystal. The atoms that are spaced in a regular order within a crystal act as
scattering sites for the X-ray beam which has a wavelength equal to the inter atomic distance in a
crystal. In 1912, W.H. Bragg and W.L. Bragg expressed the necessary conditions for diffraction
as the mathematical form given in Equation 2.1.

n  2d sin ………………………. (2.1)
Equation 2.1 is the expression for the Bragg Law of diffraction where ‘n’ is the order of
reflection, ‘λ’ is the wavelength of the beam, ‘d’ represents the spacing between atoms or planes
and ‘’is the diffraction angle. The first order reflection was considered for the work done here
[103]. Montmorillonite clay, in its natural form has an interlayer basal spacing of around 0.95 nm

17

[104]. The addition of octadecyl ammonium surfactants with primary and quaternary onium ion
surfactant heads increases this interlayer basal spacing. The basal spacing for the organoclay
I.30P having a primary onium ion surfactant increases to about 2.2 nm whereas the spacing for
I.44P organoclay with quarternary onium ion surfactant increases to 2.59 nm. For maleated
polypropylene, the X-ray peak is obtained at a 2 value of about 14 degrees.
The X-ray diffractometer used for the present study was a Rigaku Rotaflex Ru-200BH Xray diffractometer, which is equipped with a Ni-filtered Cu K radiation source and operated at
45 kV and 100 mA. The sample is scanned over a 2 range of 0.5o to 15o at a rate of 0.5o/min
and measurements are recorded at equal increments of 0.01o. The low end of the scanning range
is defined by the smallest angle the wide-angle instrument can scan whereas the high end of the
scanning range is chosen so that the first diffraction peak for the maleated polypropylene matrix
occurring at about 14º 2 is observed. The powder x-ray diffraction of the clays is carried out by
compacting the clay in a 1"-1" cavity on a glass slide. For all the nanocomposites, compressionmolded disks were used which were first pasted onto glass slides to provide enough rigidity to
the samples. This would allow us to obtain enough diffraction intensities with the x-ray beam.
The scanning speed of 0.5 degrees/minute was chosen to obtain high resolution
diffraction. . The speed of the detector is important because a high scanning speed may not
register small variations in the diffraction intensity, which may be essential for testing in the
lower range of 2 values. For the same reasons, a smaller sampling interval of 0.01o was chosen
over the entire range of 2 values. Also, the X ray diffraction machine has four diffraction slits;
one between the x-ray source (diffraction slit) and the sample; two after the sample on the
goniometer and the last slit is on the detector. To obtain low angle diffraction data, a small
diffraction slit of 1/6o was chosen for the first 2 slits which conventionally have the same width.
18

The other 2 slits do not have to be of the same width and on the basis of previous work done in
our lab on polypropylene –clay nanocomposites, a slit size of 0.3o was chosen for them.
2.5

THERMOGRAVIMETRIC ANALYSIS
Thermo gravimetric analysis was carried out on the pure organoclays, neat matrix as well

as the nanocomposites to determine the amount of surfactant in the organoclay and also the total
clay content in the nanocomposites. The TGA measurements also gave an idea about the thermal
stability of the nanocomposites. TGA is a technique used to determine changes in weight with
respect to changes in temperature. TGA is commonly used to determine the material
characteristics of different polymers, their decomposition temperatures and also the amount of
organic and inorganic components in a system.
The Thermo gravimetric analyzer used in the present study was a TA instrument TGA
Q500 with a low mass furnace, ultra-sensitive thermo balance and an efficient purge gas system.
A high precision platinum pan was used to measure the percentage weight change with respect to
change in temperature for the different materials. The platinum pan was first tarred to remove
any offset with respect to weight. The pan was then loaded with approximately 10 mg of the
different materials each time to measure their weight changes. A high resolution temperature
ramp was used at a rate of 10°C / min from 4°C to 600°C on the different materials. The high
resolution TGA procedure allows greater accuracy since the temperature changes slowly when
the weight loss increases. An inert atmosphere was created by purging nitrogen gas in the
electrically heated oven equipped with a high precision thermocouple to measure the accurate
temperature.

19

The TGA measurements on the pure organoclays revealed that the I.44P clay had about
30% of the quaternary ammonium ion surfactant whereas the I.30P clay had 23% of the primary
amine surfactant.

The TGA measurements also indicated that the thermal stability of the

nanocomposites increased by around 50°C to 55°C as compared to neat PP-g-MA matrix at
about 10% of the weight loss.
2.6

RHEOLOGY
Rheology, the study of flow and deformation [105], is an important tool for investigating

the micro structural changes occurring in polymer nanocomposites upon deformation. Based on
the applied deformation, rheological measurements help to determine the linear as well as nonlinear viscoelastic properties such as viscosity, modulus etc. by using rheometers. Rheometers
can be classified into different categories depending on what material functions they are used to
determine. The first basic division is by kinematics: Shear versus Extension [105]. They can be
also classified based on the type of strain applied viz., small strains, large strains, steady strains.
The present work focuses on both the shear as well as extensional rheology on the
polypropylene-layered silicate nanocomposites. At smaller strains or strain rates both shear and
extensional measurements give the same material function which is the linear viscoealstic limit
of time dependent viscosity.

lim  (t ,  )  lim 1 (t , )   (t )
 0

 0

…………….. (2.2)

+

However at long times or very large strains, η (t,  ) becomes a function of shear rate only.
Between these two limits extensional behavior for a material maybe different form the shear
behavior and it depends on time and strain or strain rate. The following sections describe in
20

detail the various viscoelastic properties measured on these nanocomposites through shear and
extensional rheology.
2.6.1

Shear Rheology
Shear rheology was used here to investigate the linear viscoelastic response of the

polypropylene clay nanocomposites. Shear rheometers can be classified based on the flow that is
used to generate the shear on the material, viz., “drag flow” and “pressure driven flow”. In drag
flows shear is generated between a moving and a fixed surface whereas in pressure driven flows
the shear is generated by a pressure difference across a channel. For the present work a parallel
plate geometry was used in which shear is generated on the material between a moving plate and
a parallel fixed plate. The parallel disk geometry is shown in figure 2.2.

Ω, θ
R

h

M
Fz
Figure 2.2.

Schematic of a parallel plate rheometer.
21

Based on the assumptions of [105]:
1. Steady, laminar, isothermal flow
2. (r,z) only, r = z = 0
3. Negligible body forces
4. Cylindrical edge
The equations of motion for parallel plate geometry with radial co-ordinates become:

 z

:

z
 z z

z:

z

 0.......... .......... .......... ......( 2.3)
 0.......... .......... .......... .......( 2.4)

 
2
1
r rr  
 
.......... ..( 2.5)
r r
r
r

r:

The above equations can be solved by assuming a condition of no slip at plate surfaces and
the shear stress can be related to the torque observed on the plate by:

12 R  


d ln M 
3 
.......... .......... ..( 2.6)
d ln  R 
2R 
M

3

This shear stress data can be then used to calculate the linear viscoelastic properties such as
the viscoelastic modulus. By applying small amplitude oscillatory strains one can obtain
oscillatory stress data which can be mathematically expressed as:

 

0

sin  t …………………………… (2.7)

22

   sin ( t   ) ……………………….(2.8)
0

This oscillatory stress data can be decomposed into an in phase elastic component with
strain wave and 90° out of phase viscous component. Based on these two components two
different dynamic moduli can be obtained: the elastic component called as the “storage modulus”
(G’) and the viscous component called as the “loss modulus” (G”). The equations for the storage,
loss and complex modulus are given below:


G '  0 .......... .......... .......... .......... ..( 2.9)
'

0

 0"
G" 
.......... .......... .......... .......... . (2.10)
0
G*  G 'iG" .......... .......... .......... ... (2.11)
A complex viscosity is defined in a similar manner and its magnitude is given as:

  G   2  G   2
 *  (  2   2 ) 1 / 2       
  



1/ 2
G*







…… (2.12)

Based on these equations, the different material functions for the various nanocomposites
were determined which were useful in gaining an insight into the material micro structure.
Parallel plate geometry on a TA Instruments AR2000 rheometer was used to perform the
oscillatory shear measurements. The linear viscoelastic regime was determined by first running a
strain sweep test on the nanocomposites. In a strain sweep test the strains are varied on the
material by keeping a fixed frequency and observing the various viscoelastic properties (G', ')
as a function of strain. The strain at which these properties show a change indicates that beyond
23

this strain value the non-linear region starts. A strain value of 0.5% was well within the
viscoelastic regime for these nanocomposites and all the linear viscoelastic properties were
obtained at this strain. The strain sweep tests were also used for probing into the non-linear
viscoelasticity of the nanocomposites. To obtain torque values much greater than the lower
bound of torque values for the AR 2000 rheometer, 40 mm diameter disk geometry was used
instead of 25mm diameter disk geometry. To probe into the structure of the nanocomposites,
lower strains are used because it has been shown previously that large amplitude oscillatory
strains align the nanoparticles [106, 107]. A frequency range of 0.01 to 100  (rad/s) at 0.5%
strain amplitude was chosen to perform the frequency sweep tests and the linear viscoelastic data
was measured over this entire range. A time sweep was also carried out on these nanocomposites
on the 40 mm parallel plate geometry for about 1 hour to check for the stability of modulus for
these nanocomposites with respect to time.
For the non-linear shear rheology three subsequent strain sweep tests were run to observe
the drop in storage and loss modulus values. The strain amplitude varied from 0.1% to 100% and
the consequent storage and loss moduli were recorded. To understand the recovery kinetics of
the modulus values of the nanocomposites, a time sweep was run at low strain amplitude and
fixed frequency immediately after running a strain sweep test. The delay between the first point
of the time sweep test and the last strain amplitude point in strain sweep was 10 sec. A
continuous purge of nitrogen was supplied during all these tests to avoid thermal degradation.
2.6.2

Extensional Rheology
In the early 1900’s, Trouton carefully tested different viscous sample at slow extensional

deformation rates and found that the viscosity measured in uniaxial extension is three time the
24

shear viscosity for Newtonian fluid. This was the start of the extensional rheometry field;
however it significantly advanced only after the 1970’s. Extensional rheology is very sensitive to
structural changes and plays a significant role in many processing operations such as
thermoforming, blow molding, fiber spinning and foaming. This is because the polymer chains
and asymmetric particles get oriented along the extensional flow which bears a significant effect
on the final product properties [105]. However, it is very difficult to generate reliable extensional
flow data because of the shear stresses generated at stationary boundaries. Therefore utmost care
needs to be taken while measuring extensional flow properties of a material.
In the present work, simple uniaxial extension has been used to obtain the extensional
rheology data. The test samples are pulled between rotating clamps, a design developed by
Meissner [108] and further improved by him and his coworkers [109-111]. In this design one end
of the sample is fixed and the free end being wrapped around a rotating rod [112-118]. The
apparatus or the fixture used in this work is called as the “Elongational Viscosity Fixture” (EVF)
that can be mounted onto a TA ARES rheometer. The design of EVF is based on the original
Meissner concept, however, instead of the rotary clamps, two cylinders are used to wind up the
sample; one cylinder is rotating while the other is stationary and measuring the force. The
schematic of the elongational viscosity fixture is shown in figure 2.3. The rotating cylinder
moves on a circular orbit around the force- measuring cylinder while rotating around its own axis
at the same time and this helps in winding up the sample equally on both sides. The rotation of
the cylinder is made by the ARES actuator. The advantage with this design is that the stationary
cylinder measuring the force is fixed in space and as such can be decoupled with the torque
transducer of the ARES instrument. Since the force measurement is decoupled from all the

25

moving parts, factor such as friction and inertia do not contribute to the force signal giving pure
extensional force.

Rotating drum

X-ducer shaft

Test sample

Pair of gears
Fixed outer axis

Motor axis

Figure 2.3.

Schematic of an Elongational Viscosity Fixture (EVF).

Based on the torque measured the following working equations help in calculating the
extensional viscosity of the material tested on EVF. The extensional strain rate is given as

 

(t ).Dd
d
H 
dt
Ls

26

………………… (2.13)

Where,



is the extensional strain rate,  H is the Hencky strain, Ω (t) is the motor

speed, Dd is the drum diameter of cylinder and Ls is the sample length. The time required for
performing uniaxial extension on a sample can be calculated on the basis of the Hencky strain to
which the sample has to be stretched and the extensional strain rate.

t

H


………………………. (2.14)

The extensional stress is calculated from the extensional force measured on EVF divided by
the area of the sample As.

 E (t ) 

Fs (t )
……………………. (2.15)
As (t )

Based on this extensional stress and the extensional strain rate, the uniaxial extensional
viscosity of the samples can be given by the following equation:

 E (t ) 

 E (t )
d
H
dt

………………… (2.16)

In the EVF the motor and transducer axis are aligned and the actual center to center
distance between the rotating and stationary drum on the EVF is 12.7 mm. This means that the
initial sample length is taken as 12.7 mm. A fixed hollow shaft having one end attached as a spur
gear on top of the rotating drum and the other end mounted on the motor shaft imparts spin to the
rotating drum. The diameters of the two drums are 10.3 mm and the clearing distance is 2.4 mm.
27

The whole EVF assembly fits into the conventionally heated ARES oven. Once the desired test
temperature is reached the sample is clipped between the two cylinders. The transient uniaxial
extensional viscosity of the nanocomposites as well as the pure maleated polypropylene matrix
-1

was measured at strain rates ranging from 0.1s

-1

to 2s up to a Hencky strain of 3. The test

temperature for the matrix was taken as 150°C which was well above its melting point. However
the different polypropylene clay nanocomposites were measured at two different test
temperatures of 155° C and 160° C. These temperatures were well beyond the melting transition
for these nanocomposites. Due to the small sample size and the fast heating rate of the ARES
oven, the waiting time after loading the sample could be reduced to less than three minutes, thus
preventing creep in the sample. During the measurement itself, the gravitational force does not
contribute significantly. The test samples were examined after completion of the test and if any
necking was noted in the stretched portion in the middle, the test was rejected.

28

RHEOLOGY OF MALEATED LINEAR POLYPROPYLENE-CLAY
NANOCOMPOSITES VARYING THE CLAY ASPECT RATIO AND ORGANIC CONTENT
________________________________________________________________________
CHAPTER 3
3.1

INTRODUCTION
Polymer-layered silicate nanocomposites have gained considerable research interest in

recent years, after the initial research work done by Toyota researchers on nylon/clay
nanocomposites [7,8]. The addition of small amount of high aspect ratio (of the order of ~200)
nanoclay platelets to polymer matrix, through melt mixing, leads to significant improvements in
the polymer/clay nanocomposite properties such as stiffness, tensile strength, barrier properties,
reduction in flammability, etc. [9-12]. Polypropylene (PP) which is a commodity polymer has
been widely investigated to make its nanocomposites with clay. PP being non polar needs to be
functionalized with maleic anhydride to form maleated PP so as to wet the clay surface more
readily and mix properly for making the nanocomposites . Okada and coworkers were the first to
make nanocomposites out of PP, maleated PP and clay [13-15]. These polymer clay
nanocomposites are characterized by structural as well rheological techniques depending on the
kind of industrial process they need to be developed for.
Recently, the polymer- layered silicate nanocomposites have been widely
developed for various applications in automotive industries, for food packaging films, etc. For
such applications the nanocomposites are processed through various processes such as extrusion
foaming, thermoforming, blow molding, compression molding etc. The study of the extensional
flow behavior of these polymer nanocomposite melts is very important for such processes and as
29

such the measurements of transient uniaxial extensional viscosity for polymer melts become
imperative. Several researchers have measured the transient uniaxial extensional viscosity [6972] and melt strength [73, 74] of pure polymer melts and their blends to understand their stain
hardening behavior. In particular, a rise in polymer melt viscosity (above the linear viscoelastic
limit) with increasing Hencky strain is called as “strain hardening” and is desirable to a certain
extent for above mentioned processes.
Sugimoto and coworkers have shown that linear polypropylene on its own does not
display any strain hardening. However they were able to show strain hardening in PP which was
modified either by introducing branched molecular structure through irradiation [75] or by
adding a small amount of uhmw–PE to linear PP [76]. The extensional flow of composites with
micron size fillers has also been investigated with different sizes, shapes and aspect ratios by
Takahashi et al. [84] and Le Meins et al [85]. Takahashi used microscale fillers such as mica,
talc, glass beads and glass flakes in a low density polyethylene matrix, whereas Le Meins et al.
suspended monodisperse polystyrene spheres in polyisobutylene liquid. They found that the
strain hardening goes down with filler addition and even more so with high aspect ratio particles
such as fibers. They also observed that the formation of aggregates led to a decrease in strain
hardening behavior. Although similar observations of decreasing strain hardening have also been
reported on nanocomposites by several authors including Bhattacharya and coworkers [66, 86],
there have been two reports with nylon 6 [87] and maleated polypropylene [88] nanocomposites
wherein the strain hardening improved slightly. Seong and coworkers [87] dispersed Closite
nanoclay in Polyamide 6 matrix and observed that with well dispersed systems there was an
improvement in the storage modulus, viscosity as well as the strain hardening behavior in
extension. Okamoto et al. [88] dispersed treated montmorillonite clay in maleated polypropylene
30

and investigated the strain hardening as well rheopexy features of the nanocomposites. However,
there have been no reports which show a thorough investigation of the strain hardening behavior
of linear maleated polypropylene and layered silicate nanocomposite melts with respect to
varying organic surfactant content in the nanoclays as well different effective aspect ratios of the
nanoclays. Also, finding a possible correlation between the effects of varying aspect ratio and
organic content in the filler and the linear as well as nonlinear shear viscoelastic behavior of the
nanocomposites leaves us with a scope of research work that is done in the present study.
The system of interest here is linear polypropylene grafted with maleic anhydride, also
called as linear maleated polypropylene (PP-g-MA), used as the matrix. The primary objective in
choosing just the functionalized (maleated) polypropylene as the complete matrix was to
understand the kind of interactions taking place between the maleic anhydride group on the
polymer chains and the -OH groups on the clay edges, that may aid in achieving strain hardening
in the nanocomposites. The addition of linear bulk polypropylene would add a third phase
making the system more complex. Hence just the linear maleated polypropylene copolymer
(with ethylene) was chosen as the matrix to which layered silicates (Montmorillonite clay)
having varying aspect ratios and different surfactant treatments were added as the filler.
The layered sheets in montmorillonite clay have an excess negative charge which is
+

2+

neutralized by highly charged inorganic cations such as Na and Ca

present in a hydrated form

in the interlayers of untreated clay [1, 26]. The hydration spheres of these inorganic cations form
unfavorable organophobic conditions for adsorption of organic polymer molecules between the
interlayer. Also, the removal of water of hydration of these cations leads to collapsing of the
interlayer spacing thereby preventing the entrance of the organic polymer chains between the

31

interlayer spacing and exfoliation of clay in polymer matrix. The clay surface is therefore treated
so as to make the clay interlayer region organophilic and make the adsorption of organic polymer
molecules between the clay platelets easier. The polymer chains can then enter the clay galleries
between two subsequent clay layers and separate them causing exfoliation of clay in the
polymer. The surface treatment of clay is done by exchanging the hydrated cations with
quaternary organo-ammonium salts or with primary amines having hydrocarbon groups attached
to them, thereby creating organophilic condition in the clay interlayer region. In the present
study, two different montmorillonite clays, organically modified by incorporating primary amine
surfactant in one and quaternary ammonium ion surfactant in other are used. These surfactants
are present in the clay interlayer galleries. The organoclays were melt-mixed with linear
polypropylene and different nanocomposites were made. These nanocomposites were tested for
nonlinear, uniaxial extensional melt viscosity transients to check if the addition of the organically
modified nanoclay imparted any strain hardening to the non-strain hardening PP matrix. Also,
linear and nonlinear shear rheology was used to further probe into the structure of the
nanocomposites and understand the shear and extensional behavior.
In general, the addition of nanofiller to polymeric systems leads to improvements in the
reinforcement and various physical properties of the polymer. Sternstein et.al, [95, 96] proposed
different mechanisms to explain the reinforcements observed in silica-filled poly (vinyl acetate)
[PVAc] matrix at low filler loadings in linear and nonlinear viscoelastic properties. They
proposed that the possible reinforcements in the nanofilled polymer melt may be due to the
trapped or hindered chain entanglements formed between the bulk PVAc chains and the chains
that are physically attached to the filler surfaces. A similar observation of increase in
reinforcement as well as strain hardening was previously made by Yamaguchi and coworkers
32

[79-81] with blends of linear and cross linked polymers. They observed that the addition of a
small amount of cross linked terpolymer (gEHDM) characterized as gel just beyond the sol-gel
transition point to a linear polypropylene improved the strain hardening behavior of the blend
greatly. They suggested that with appropriate cross link density the stretching of trapped chains
between the crosslinking points were responsible for the strain hardening and an increase in the
crosslinking density leads to suppression of the strain hardening behavior. Cassagnau and Melis
[97] studied the nonlinear viscoelastic behavior and modulus recovery in silica filled polymers
and showed that in physically crosslinked composites the complex shear modulus is perfectly
recoverable as shown through subsequent strain sweeps and time sweeps.
In the current study, both nonlinear extension and linear and nonlinear shear rheology are
used as investigative tools to understand the effects of nanoclay addition in linear maleated
polypropylene. An increase in reinforcement and strain hardening in the nanocomposites has
been explained by the possible formation of hindered chain entanglements between particle
attached polymer chains and free polymer chains.
3.2

EXPERIMENTAL METHODS

3.2.1 Materials
The matrix material is maleated polypropylene (PP-g-MA) (Exxelor PO1015) with 0.4
wt% maleic anhydride and a molecular weight of 180,000. (ζ =0.9 g/cc). It is important to
choose a PP-g-MA having a high molecular weight and substantial polymer chain length
necessary for the formation of hindered chain entanglements. Exxelor PO1015 was chosen
amongst several other PP-g-MA’s (Honeywell AC 950 and AC 1325) to fit the criteria for this
study. The melt mass flow rate (MFR) of Exxelor PO1015 is 150g /10 min at 230°C under a load
33

of 2.16 kg and it melts at 145°C. It is a random copolymer of propylene and ethylene which is
the reason for its low melting point. The organoclay used for preparing the nanocomposites was
(a) Nanocor I.30P which had a primary amine surfactant with single alkyl tail and (b) Nanocor
I.44P which had quaternary ammonium ion surfactant with two C-18 tails.
3.2.2 Compounding
PP-g-MA and clay were dried overnight at 100°C and 20 mm Hg vacuum in a vacuum
oven and then compounded with maleated polypropylene at 165° C and at 150 rpm for 10
minutes in a Banbury mixer; this was done under nitrogen to avoid thermal degradation. One
batch of the mixer amounted to 47 grams of compound. The compounded material was removed
from the mixer with a brass scraper and ground up with a mini granulator. The granules were
further compression molded to desired thicknesses for different shear and extensional tests.
3.3

CHARACTERIZATION TECHNIQUES

3.3.1 DSC and TGA Measurements
The melting temperature of the nanocomposite pellets as well as the neat polymer was
obtained using DSC runs on a TA DSC instrument. The melting point of neat PP-g-MA was
145C while the nanocomposites melted around 146.5C. Hence compounding was carried out at
165C and compression molding was carried out at 175C which is well above the melting points
of these nanocomposites. Thermo gravimetric Analysis was carried out on the pure clay, neat
matrix as well the nanocomposites on a TA Q500 TGA instrument from 4°C to 600°C through a
high resolution temperature ramp of 10°C / min. The weight loss from TGA revealed that the
pure I.30P clay had lesser amount of organic surfactant than the I.44P clay.

34

3.3.2 X ray Diffraction Measurement
All the clays and nanocomposites were analyzed using X-ray diffraction patterns. The
characterization of the structure by XRD was performed on the compression-molded disks with a
Rigaku Rotaflex Ru-200BH X-ray diffractometer, which is equipped with a Ni-filtered Cu K α
radiation source and operated at 45 kV and 100 mA. The sample was scanned over a 2 range of
0.5° to 10° at a rate of 0.5°/min and measurements were recorded at equal increments of 0.01°.
3.3.3 Extensional Viscosity Measurements
An extensional viscosity fixture (EVF) mounted on a TA-ARES rheometer was used to
-1

-1

record the transient uniaxial extensional viscosity at strain rates ranging from 0.1s to 2s up to
a Hencky strain of 3. A 20-specimen mold was used for compression molding samples 18mm
long, 10mm wide and 0.7mm thick for the EVF tests. Test specimens were molded at 175° C and
the mold was cooled slowly using cooling water under a pressure of 5 tons. This ensured stress
free and relaxed homogenous samples with minimum dimensional variation. The molded
specimens were dried overnight before performing the extensional tests at 155° C. The test
samples were examined after completion of the test and if any necking was noted in the stretched
portion in the middle, the test was rejected.
3.3.4 Shear Rheology
Parallel plate geometry on a TA Instruments AR2000 rheometer was used to perform
frequency sweep tests on these nanocomposites in the linear viscoelastic region. The diameter of
the parallel plates was 40 mm and a frequency range of 0.01 to 100  (rad/s) at 0.5% strain

35

amplitude was chosen to perform these tests. The corresponding storage modulus and complex
shear viscosity were noted over the entire frequency range. The strain amplitude was chosen
such that it was in the linear viscoelastic region and this was found out by running strain sweep
tests at 1 rad/s on the nanocomposites. A time sweep was also carried out on these
nanocomposites on the 40 mm parallel plate geometry for about 1 hour to check for the stability
of modulus for these nanocomposites with respect to time. Subsequent strain sweep tests were
used for probing into the nonlinear viscoelasticity of the nanocomposites. Also, a time sweep
was run immediately after a strain sweep to get an estimate of the recovery time of complex
shear modulus. A continuous purge of nitrogen was supplied during these tests to avoid thermal
degradation.
3.4

RESULTS AND DISCUSSIONS

3.4.1 Effect of nanoclay dispersion and stability in nanocomposites
Organoclays with different surfactant chemistries are used for making the PP – clay
nanocomposites. One grade of montmorillonite clay has a primary onium ion surfactant (NH3

+

+

head group with single C-18 tail) and another one has a quaternary onium ion surfactant (N

(CH3)2 with two C-18 tails) in the clay interlayer galleries. Heinz et al. [29] have commented on
the structure and conformation of these surfactants in the interlayer galleries. The primary onium
ion surfactant with a C-18 tail is hydrogen bonded more tightly to the clay surface and tends to
orient itself perpendicular to the clay surface. The quaternary onium ion surfactant with two C18 tails is more mobile in the interlayer galleries; it also packs to a greater extent in the interlayer
galleries. The extent of surfactant packing can be judged from the X-ray diffraction patterns of
the different organoclays, presented in Figure 3.1.
36

Figure 3.1. X-ray diffraction of organoclay: a) I.30 P clay (clay A), b) I.44 P clay (clay B)
It can be seen that the basal spacing for I.30P clay (2.2 nm) is smaller as compared to that of
I.44P clay (2.59 nm). We infer that the greater basal spacing of the I.44P clay is mainly because
of the greater extent of surfactant packing in its interlayer galleries.
XRD patterns for the different nanocomposites made with 5 wt% of the clays are
presented in Figure 3.2. It can be seen that the composites with I.30 P clay (PPNC-A5) and with
I.44 P clay (PPNC-B5) do not show distinct peaks indicating a good level of dispersion of clay
platelets in the polymer matrix. The presence of a broader shoulder, instead of sharp peak in
XRD suggests that there must be some intercalated clay structure present in these
nanocomposites. This is more so evident with PPNC–B5 as compared to PPNC-A5.

37

Intensity

PPNC d-spacing
A5 = 3.46 nm
B5 = 3.52 nm

PPNC-B5

PPNC-A5

1

2

3

4

5

6

7

8

9

10

2deg)

Figure 3.2. X-ray diffraction of composites: a) A5 nanocomposite, b) B5 nanocomposite.
Thermo gravimetric Analysis on the pure clay further confirmed the XRD observations
and proved from the ash content calculations that I.30P clay had lesser organic content in the
interlayer galleries than the I.44P clay. Figure 3.3 shows the weight percentage lost with increase
in temperature for the neat matrix and nanocomposites. The dispersion of clay in the matrix
increased the thermal stability of the composites by almost 50°C-60°C. The onset of
decomposition for neat matrix begins around 330°C whereas for the nanocomposites it is 390°C.
The amount of clay dispersed in the nanocomposites was calculated from the char yield at 600°C
and was about 5.1% for PPNC A5 nanocomposite and 4.7% for PPNC B5 nanocomposite. The
shear and extensional rheology tests were carried out at a temperature which is well within the
stability limits of the clay and the matrix polymer.
38

100
90

80
70

% wt

60
50
40
30

Pure Matrix

20

PPNC A5

10

PPNC B5

0
300

320

340

360

380

400

temperature (° C)

Figure 3.3. Thermo gravimetric Analysis (TGA) curves obtained for neat matrix, A5 and B5
nanocomposites at 10°C / min up to 600°C.
The stability of the modulus of maleated polypropylene matrix and the nanocomposites
with the 5wt% of the organoclays was also checked with the help of a time sweep. The time
sweep was carried out for about one hour and at 1 rad/s frequency and 0.5% strain amplitude. It
can be seen from Figure 3.4, that the modulus of the matrix and the nanocomposite was stable
and did not fluctuate over the time period of the test. The strain amplitude chosen was obtained
by running strain sweep tests on the nanocomposites at 1rad/s frequency and was found to be
well within the linear viscoelastic regime.

39

1.5

160˚C, 0.5% strain, =1rad/s

1.4
1.3
1.2

G' / G'0

1.1
1
0.9
0.8
Matrix
0.7
PPNC A5
0.6

PPNC B5

0.5
0

500

1000

1500

2000

2500

3000

3500

4000

t (sec)

Figure 3.4. Time Sweep test: Change in normalized Storage modulus curves with respect to time
for the neat matrix (●), A5 nanocomposite (*) and B5 nanocomposite (∆)
3.4.2 Oscillatory Shear rheology
3.4.2.1 Linear Viscoelastic Measurements
The frequency sweep measurements done by parallel plate oscillatory shear rheology
yielded results of the dynamic modulus and dynamic viscosity of the nanocomposites. A
comparison of the dynamic viscosities for the two composites as well as the base polymer matrix
is shown in Figure 3.5.

40

1.E+04

155°C
Matrix
PPNC A5

' (Pa.s)

PPNC B5

1.E+03

1.E+02
0.01

0.1

1

10

100

 (rad/s)
Figure 3.5. Dynamic viscosity curves for the neat matrix (●), A5 nanocomposite (*) and B5
nanocomposite (∆).
The dynamic viscosity for the pure matrix is fairly constant at lower values of the
frequency and much lower than that of the nanocomposites exhibiting a pseudo-Newtonian
behavior. However, the dynamic viscosity increases significantly upon addition of a small
amount of nanoclay to the pure matrix. The comparison between the dynamic viscosities of the
two composites shows that PPNC B5 (5wt% I.44P clay) does not show a complete leveling off at
lower frequencies, in the given frequency range, indicating a more elastic response in this
nanocomposite.
In order to gain more physical insight, the nanocomposite melt viscosity relative to that
of the matrix was correlated with the help of the Pierce-Maron empirical model, equation (3.1)
41

which applies to concentrated suspensions. Carreau and co-workers [125] used the same
equation and correlated the complex viscosity data of PET nanocomposites at higher frequencies
to volume fraction of the clay platelets in order to determine the extent of polymer matrix
degradation upon compounding the nanocomposites. In this work, the dynamic viscosity data in
the low frequency region were used instead of the complex viscosity. In the low frequency
region, the dynamic storage modulus is quite sensitive to the particle-particle interactions which
are not accounted for in the Pierce-Maron equation given below. Hence, we have used dynamic
viscosity data for our calculations. .

(

)

……………….. (3.1)

’

’

where, η is the dynamic viscosity of the nanocomposite and η matrix is that for the linear
maleated polypropylene matrix, ϕ is the volume fraction of nanoclay (including surfactant) and

ϕM is the maximum packing volume fraction. The maximum packing volume fraction can be
related to the average ratio a of diameter D to thickness t for the nanolayers as done by Sun et.al
[126] and Wan et.al [127] in their work on nanoclay filled composites and shown in Equation
3.2;

………………………………(3.2)

In this work, Equation (3.1) was used first estimate the maximum packing fraction ϕM, of the
nanoclay in the composites with the two different clays: I.30 P and I.44P at 5wt% loading.
Equation (3.2) was used next to determine the effective aspect ratio of the different nanoclays.

42

These calculations showed that the effective aspect ratio for the I.30P clay was around 56 and
that of I.44P clay was 68, which is significantly higher.

1.E+05

155°C
1.E+04

G' (Pa)

1.E+03
1.E+02

1.E+01
1.E+00

Matrix
PPNC A5

1.E-01

PPNC B5
1.E-02
0.01

0.1

1

10

100

(rad/s)
Figure 3.6. Dynamic Storage Modulus curves for the neat matrix (●), A5 nanocomposite (*) and
B5 nanocomposite (∆).
The dynamic modulus curves for the two nanocomposites are presented along with that
for the polymer matrix in Figure 3.6, The dynamic modulus curve for the B5 nanocomposite
(with I.44P) is greater than that of the A5 nanocomposite (with I.30P) and also shows an
approach to a plateau like region at low frequencies. The low frequency storage modulus of
nanocomposites is raised above the matrix values because of the filler network formed by
particle-particle interactions [97]. Higher aspect ratio particles present more surface area per unit
volume for such interactions and strengthen the filler network leading to a greater increase in low
43

frequency storage modulus of the B5 nanocomposite. So the comparison of the storage modulus
curves is consistent with estimated trends in aspect ratio from dynamic viscosity data. The low
frequency “plateau” or increase in storage modulus may arise in general from two contributions:
(1) a filler particle network formed by van der Waals forces and, (2) entanglements between free
polymer chains and particle attached polymer chains. The second factor involves the volume
fraction of chains attached to particles which can also be increased by increasing aspect ratio of
the nanolayers.
3.4.2.3 Non-Linear Viscoelastic Measurements
It is widely understood that higher levels of reinforcement observed upon addition of
filler particles in a polymer are a result of filler network formation and filler agglomeration [9294]. However the mechanisms behind these reinforcements and nonlinearity with increasing
strain, at lower filler loadings, are still an area of considerable research. Sternstein and coworkers
have tried to understand these mechanisms by which the nano-scale fillers promote both
reinforcement and nonlinear viscoelastic behavior in polymer melts [95, 96]. They have worked
with fumed silica fillers, at various loadings, having different surface treatments and poly (vinyl
acetate) matrix with different molecular weights and explored the nonlinear viscoelasticity of
these nanocomposites. They also proposed that the reinforcement and nonlinear viscoelastic
behavior of the nano filled polymers is a result of “trapped entanglements” which occur because
of interactions between the filler and the polymer chains at the filler-polymer interface. The
polymer chains attach to the filler surface through polymer filler interactions thereby causing
physical cross links. These polymer chains in turn entangle with the surrounding polymer chains
causing trapped or hindered chain entanglements. Thus trapping of polymer chain loops at the
filler surface lead to an increase in the entanglement density and this increase leads to overall
44

stiffening of the matrix. Also, when large strain amplitudes are applied, these entanglements are
lost because of the debonding of polymers chains from the filler surface which results in overall
relaxation of the matrix entanglement structure. This loss of entanglement structure causes non
linearity in the viscoelastic behavior of the nanocomposites [95]. Cassagnau and Melis [97] have
also worked with fumed silica composites with different polymer systems such as PP, PS and
EVA and carried subsequent strain sweep tests and time sweep tests to understand the nonlinear
viscoelastic behavior of these composites and modulus recovery kinetics.

2500

155°C
2000

G' (Pa)

1500

1000

500

Strain Sweep 1
Strain Sweep 2
Strain Sweep 3

0
0.0001

0.001

0.01

0.1

1

10

Strain Amplitude
Figure 3.7. Subsequent strain sweep tests. Variation of storage modulus with strain amplitude
for A5 nanocomposite at 155°C and 1 Hz.

45

In the present work, to understand the possible causes of the reinforcements as well as
recovery kinetics of storage modulus from nonlinearity of the polypropylene-layered silicate
nanocomposites, the nanocomposites were tested for their nonlinear viscoelastic behavior. This
was achieved by performing subsequent strain sweep tests on the nanocomposites at a test
temperature of 155°C and a frequency of 1 Hz. The strain amplitude varied from 0.1% to 100%
and the consequent storage and loss moduli were recorded. Figures 3.7 and 3.8 show the storage
modulus over three subsequent strain sweep measurements for PPNC A5 and PPNC B5
nanocomposites respectively, over the entire strain amplitude range. It can be seen for both
PPNC A5 and PPNC B5 that mostly all the dynamic stress softening happens in the first strain
sweep and very less in the subsequent strain sweep

2500

155°C
2000

G' (Pa)

1500

1000

Strain Sweep 1

500

Strain Sweep 2
Strain Sweep 3
0
0.0001

0.001

0.01

0.1

1

10

Strain Amplitude
Figure 3. 8.Subsequent strain sweep tests. Variation of storage modulus with strain amplitude
for B5 nanocomposite at 155°C and 1 Hz.
46

In addition to this, PPNC B5 nanocomposite shows a two stage drop in the storage modulus with
increasing strain which highlights the breakup of two structures—the particle network followed
by the hindered chain entanglement network involving attached chains.

6000

155°C

Strain Sweep 1
Strain Sweep 2
Strain Sweep 3

5500

G" (Pa)

5000

4500

4000

3500

3000
0.0001

0.001

0.01

0.1

1

10

Strain Amplitude
Figure 3.9. Subsequent strain sweep tests. Variation of loss modulus with strain amplitude for
A5 nanocomposite at 155°C and 1 Hz.
A similar behavior is observed in the loss modulus of the two composites over the subsequent
strain sweeps as shown in Figures 3.9 and 3.10. The two stage drop is more prominent in the loss
modulus curve for PPNC B5 nanocomposite, further confirming the existence and breakup of the
particle network followed by that of hindered chain entanglement network.

47

6000

155°C
5500

G" (Pa)

5000

4500

4000
Strain Sweep 1

3500

Strain Sweep 2
Strain Sweep 3

3000
0.0001

0.001

0.01

0.1

1

10

Strain Amplitude
Figure 3. 10. Subsequent strain sweep tests. Variation of loss modulus with strain amplitude for
B5 nanocomposite at 155°C and 1 Hz.
The modulus recovery between subsequent strain sweeps happens quicker in the PPNC
A5 nanocomposite since the actual drop in modulus is not as great as the PPNC B5 composite.
To get a more clear idea about the complex modulus recovery kinetics from nonlinearity, a time
sweep experiment was run at lower strain amplitude (in linear viscoelastic region) immediately
after a strain sweep experiment. Figure 3.11 shows the recovery kinetics of the storage modulus
for both the nanocomposites. It can be clearly seen here that PPNC A5 composite shows a faster
recovery as compared to PPNC B5. Also it can be seen from the time sweep experiment that the
storage modulus keeps increasing even beyond the initial modulus that was measured before the
strain sweep experiment.

48

2500

155°C
2000

G'(t)

1500

1000

PPNC A5
PPNC B5

500
trecover = 234.4 s

trecover = 130.5 s
0
10

100

1000

t (s)

10000

Figure 3.11. Storage modulus recovery versus time (z and γ = 0.5%). A5 and B5
nanocomposites at 155°C. G’ (t) is the storage modulus value as obtained from time
sweep test which is done immediately after one strain sweep.

This could imply that the strain sweep experiment acts as a trigger in deforming the
entanglement structure such that it requires much more time to reach an equilibrium state and
have a leveled out modulus value. A similar recovery behavior is seen for the loss modulus of
these composites (Figure 3.12), although the rate of recovery for loss modulus (G”) is lesser than
that for storage modulus (G’).

49

7000

155°C
6000
5000

G"(t)

4000
3000
2000
1000

PPNC A5
PPNC B5

0
10

100

1000

t (s)

10000

Figure 3.12. Loss modulus recovery versus time (z and γ = 0.5%). A5 and B5
nanocomposites at 155°C. G” (t) is the Loss modulus value as obtained from time
sweep test which is done immediately after one strain sweep.

This is further evident from the loss factor values when plotted against the recovery time.
Sternstein et.al [96] have shown trends in loss factor recovery for composites having fillers with
different surface treatment by plotting the ratio of loss factor observed during the recovery time
sweep to that observed in the linear viscoelastic region of the strain sweep experiment. Such a
plot is shown in Figure 3.13. It can be seen through this plot that the loss factor values for PPNC
B5 are greater than that of PPNC A5 indicating that the rate at which storage modulus is
recovered is greater in PPNC B5 case. However, since the drop in the storage modulus value

50

after first time sweep was greater in case of PPNC B5, it takes more time to reach that initial
value.

1.1

155°C

LF(t) / LF0

1.05

1

0.95
PPNC A5
PPNC B5
0.9
10

100

1000

t (s)

10000

Figure 3.13. Rate of Recovery of the loss factor after initial strain sweep for PPNC A5 and
PPNC B5. The Loss Factor LF (t) is that obtained from the time sweep test and LF 0
is the plateau value of loss factor obtained from initial strain sweep test.

This effect of non-linearity is further investigated in uniaxial extension by performing transient
extensional viscosity measurements on the nanocomposites at different test temperatures.

51

3.4.3 Uniaxial Extensional Rheology
3.4.3.3 Effect of filler addition
+

The transient extensional viscosity curve E (t) for the neat matrix was obtained by
stretching it in an uniaxial direction at a temperature well above its melting point, as shown in
Figure 3.14.

1.0E+04

150°C

PPgMA Matrix

E (Pa.s)

1.0E+03

1.0E+02
0.5 s-1
2.0 s-1

1.0E+01
0.01

0.1

1

10

t (sec)

Figure 3.14. Transient extensional viscosity curve for maleated polypropylene matrix.
It can be clearly seen that the neat matrix doesn’t show any strain hardening characteristic
for the different extensional strain rates. However by the addition of just 5 wt% of nanoclay in
the polymer matrix significant amount of strain hardening is observed for the nanocomposites
with different clays as seen in Figure 3.15.
52

1.E+07

155°C

X 20

1.E+06

PPNC B5

E+ (Pa.s)

E0+
1.E+05

PPNC A5
1.E+04

E0+
0.1 s-1
0.5 s-1
1.0 s-1
2.0 s-1

1.E+03

1.E+02
0.05

0.5

5

50

t (sec)

Figure 3.15. Transient extensional viscosity curve for nanocomposites A5 and B5 at 155° C
-1

-1

-1

-1

and four extensional strain rates: 0.1s (∆), 0.5s (□), 1s (◊) and 2s (○).
In general, with any additive that causes strain hardening, Sugimoto et al. [76] and
Wagner et al. [91] have observed that larger relaxation times lead to greater extents of strain
hardening. Larger relaxation times would be associated with stronger entanglements. The
maleated polypropylene chains that are physically attached at the ends to the clay edges would
entangle with the bulk polymer chains thereby forming hindered chain entanglements. The
increase in entanglements would in turn cause the polymer chains to relax slowly upon stretching
thereby increasing the relaxation times. This kind of strain hardening behavior observed in
maleated polypropylene – layered silicate nanocomposites is similar to that reported for polymer
blends containing small amount of chemically cross linked polymer chains. Yamaguchi and
53

Miyata [79, 80] blended small amounts of cross linked terpolymer (gEHDM) characterized as gel
just beyond the sol-gel transition point to a linear polypropylene. They observed that for systems
with appropriate cross link density the stretching of trapped or hindered chains between the
crosslinking points were responsible for the improved strain hardening behavior. Also, an
increase in the crosslinking density leads to suppression of the strain hardening behavior.
Yamaguchi and Suzuki [81] have similar reports wherein they blended small amounts of cross
linked linear low density polyethylene to bulk linear low density polyethylene and the blends
showed enhanced strain hardening behavior. It was reported that the cross linked polymer chains
formed trapped or hindered chain entanglements with the bulk chain segments between the cross
links which lead to improvement in the strain hardening of these blends.
3.4.3.2 Effect of surfactant head group and effective aspect ratio
Figure 3.15 shows a comparison between the transient extensional viscosity curves
+

E (t) for the nanocomposites PPNC A5 and PPNC B5 at four different strain rates and at a test
temperature of 155˚C. The test temperature chosen is well above the melting point of the
nanocomposites. It has been well know through literature that transient extensional viscosity
measurements are difficult to obtain and therefore care should be taken while performing the
EVF tests so that the test sample does not sag or neck during the test. Also, a relatively lower
molecular weight, a low melting point and a high melt flow rate (150 g/10min @ 230°C/2.16 kg)
of the maleated polypropylene matrix did not allow us to go to higher test temperatures on the
extensional viscosity fixture. The comparison of the extensional viscosity transients show that
the viscosity curve rose more sharply from the base viscoelastic envelope 3η0
nanocomposite PPNC B5 at each strain rate than the nanocomposite PPNC A5.
54

+

(t) for

Quantitative comparison of the extents of strain hardening may be obtained by evaluating
a strain hardening ratio  which increases with increasing strain.

 E (, t )

 E 0
+

…………………. (3.2)

+

Where, E is the transient extensional viscosity and E0 is the base viscoelastic envelope.
There are in general couple different methods to obtain the linear viscoelastic envelope: (a) by
measuring or calculating from the relaxation spectrum a linear zero shear viscosity growth curve,
or (b) by extrapolating the linear part of the extensional viscosity curve. Unlike homogenous
polymer melts, for nanocomposite systems it is observed that there is an offset between the base
viscoelastic envelope and the linear region of the extensional viscosity curves. Such observations
have been previously made by Okamoto and coworkers for polypropylene clay nanocomposites
[88]. To take into account such constant offsets, here method (b) was chosen to obtain the linear
viscoelastic envelope and calculate the reduced strain hardening parameter /0. A plot of /0
versus the strain rate at a Hencky strain of 2.25 is shown in Figure 3.16. It is further evident from
this plot that the level of strain hardening in PPNC B5 nanocomposite is greater than that
observed for PPNC A5 nanocomposite at all given extensional strain rates. Hence a greater
extent of strain hardening in the nanocomposite was obtained with the quaternary ammonium
ion, 2-alkyl tail surfactant clay than with the more tightly bound primary amine ion, single alkyl
tail surfactant clay. The higher amount of strain hardening in the I.44P clay nanocomposite can
be explained on the basis of a higher effective aspect ratio of these nanoclay particles in the
composite.

55

4

155°C, 5wt% clay

PPNC A5
PPNC B5

/0

3

2

1
0

0.5

1

1.5

2

Strain rate
Figure 3.16. Reduced Strain hardening parameter v/s strain rate: Comparison at Hencky strain
of 2.25 for PPNC –A5 and B5 at 155° C.
With higher effective aspect ratio, the number of interfacial interaction sites on the clay platelet
increases thereby increasing the chance of having more physical attachments between the filler
and the polymer chain as shown in figure 3.17. With an increase in the number of particle
attached polymer chains the possibility of these chains forming entanglements with bulk matrix
free polymer chains increases, thereby increasing the overall entanglement density. The increase
in such entanglements in turn causes the overall polymer-particle entanglement network to relax
slowly and have longer relaxation times.

56

Bu lk PP-g-MA chain
Hindered chain
Clay platelet
Entanglement
Direction of Stretch

Figure 3.17. Schematic of the interaction between the PP-g-MA chains and organoclay.
This increase in the entanglement density is reflected not only in the strain hardening
behavior in extension but also in the nonlinear and linear shear viscoelastic behavior for PPNC
B5. Also the strain hardening observed is highest for the lowest strain rate for both
nanocomposites and decreases with increasing strain rates.
3.5

CONCLUSIONS
Two different layered silicates were dispersed equally well (at about 2 vol%) in a high

molecular weight (Mw=180,000) linear maleated polypropylene copolymer (with ethylene)
which melts at 145°C; no other component was used. The aspect ratios (side / thickness) of the
two different nanoparticles dispersed were determined to be 56 and 68 from measurements of
57

relative dynamic viscosity on the nanocomposite melts. The low frequency storage modulus of
the nanocomposite with the higher aspect ratio nanoparticles was greater indicating a stronger
filler network and/or a greater fraction of polymer chains attached to the particle surface with
higher aspect ratio. Large amplitude oscillatory shear measurements on the nanocomposite melt
with the higher aspect ratio particles showed a two stage drop in the storage modulus with
increasing strain which highlights the breakup of two structures—the particle network followed
by the hindered chain entanglement network involving attached chains. The buildup of structure
during recovery is also slower for the nanocomposite with the higher aspect ratio nanoparticle
system. Both nanocomposite melts display strain hardening in uniaxial extensional flow over
several strain rates, unlike the matrix polymer which shows no strain hardening: the melt with
the higher aspect ratio layered silicates displayed a higher extent of strain hardening at all rates.
This last phenomenon may be attributed to the greater incidence of particle attached chains and
entanglements involving these chains when the particle aspect ratio is greater.

58

EEFECT OF LOADING AND TEMPERATURE ON NON LINEAR RHEOLOGY OF
MALEATED LINEAR POLYPROPYLENE-CLAY NANOCOMPOSITES
________________________________________________________________________
CHAPTER 4
4.1

INTRODUCTION AND BACKGROUND
Strain hardening of polymer melts in elongational flow, which in particular is a rise in

polymer melt viscosity (above the linear viscoelastic limit) with increasing Hencky strain, is
desirable for variety of polymer processing operations such as thermoforming, foaming and blow
molding. Recently, structurally modified polymers, cross linked polymer blends and polymer
composites, that show strain hardening property, have been widely developed for various
applications in automotive industries, for food packaging films, etc. The study of the extensional
flow behavior of these polymer nanocomposite melts is very important for the above mentioned
processes and as such the measurements of transient uniaxial extensional viscosity for such
polymer melts become important. Transient uniaxial extensional viscosity [69-71] and melt
strength [72-74] of pure polymer melts and their blends have been measured previously to
understand their stain hardening behavior.
Polypropylene (PP) is a widely used commodity polymer and linear PP does not strain
harden upon melt extension. Researchers like Sugimoto have shown that strain hardening can be
achieved in linear PP by introducing branched molecular structure through irradiation [75] or by
adding a small amount of uhmw–PE to linear PP [76]. Extensional viscosity and strain hardening
behavior of polymers with micro fillers have been investigated with different sizes, shapes and
aspect ratios by Takahashi et al. [84], Le Meins et al [85] and Bhattacharya and coworkers [66,

59

86] and it has been shown that the strain hardening goes down with filler addition and even more
so with high aspect ratio particles such as fibers. They observed that the formation of aggregates
led to a decrease in strain hardening behavior. However, Seong and coworkers [87] dispersed
Closite nanoclay in Polyamide 6 matrix and observed that with well dispersed systems there was
an improvement in the storage modulus, viscosity as well as the strain hardening behavior in
extension. Okamoto et al. [88, 89] dispersed treated montmorillonite clay in maleated
polypropylene and investigated the strain hardening as well as rheopexy features of the
nanocomposites.
In our previous work, we have shown that the addition of a small amount of nanoclay (5
wt%) to linear maleated polypropylene improves its strain hardening behavior significantly. We
have further shown that the effective aspect ratio of dispersed nanoparticles and the nanoclay
surfactant chemistry also affect the strain hardening behavior as well as the linear and nonlinear
viscoelasticity of these nanocomposites. The improvements in the strain hardening behavior as
well as the modulus recovery in nonlinear shear rheology were explained by considering the
nanocomposite system to be similar to a blend of particle attached polymer chains and free
matrix polymer chains. The polymer chains form physical cross links (-OH bonds) with the
layered silicates and these chains in turn form hindered chain entanglements with other bulk
polymer chains.
Yamaguchi and coworkers [79-81] made a similar observation of increase in reinforcement
as well as strain hardening while working with blends of linear PP and chemically cross linked
terpolymer (gEHDM) characterized as gel just beyond the sol-gel transition. They observed that
the addition of a small amount of cross linked terpolymer (gEHDM) characterized as gel just
beyond the sol-gel transition point to a linear polypropylene improved the strain hardening
60

behavior of the blend greatly. They suggested that with appropriate cross link density, the
stretching of hindered chains between the crosslinking points were responsible for the strain
hardening. However, an increase in the crosslinking density leads to suppression of the strain
hardening behavior. Similar reports on decrease in strain hardening with increasing degree of
chemical cross linking in the gel phase of the blend have been given with PMMA systems by
Takahashi and Ogura [82]. These kinds of observations made with linear and chemically cross
linked polymer blends leave us with a question in our work, as to whether a change in the
loading of nanoclay particles to the linear maleated polypropylene affect its strain hardening
behavior or not. It would be of great interest to investigate the strain hardening behavior of
nanocomposites having a higher loading of nanoclay and check for the mechanisms that may or
may not affect the strain hardening and reinforcement in the nanocomposites. Therefore, to
answer all these questions, we varied the nanoclay loadings in the linear polypropylene matrix in
this chapter and studied the linear shear, nonlinear shear and extensional rheology of the
nanocomposites.
In this study, a polypropylene copolymer (with ethylene) grafted with maleic anhydride,
also called as linear maleated polypropylene (PP-g-MA), is used as the matrix to which a layered
silicate (Montmorillonite) was added as the filler. The montmorillonite clay is organically
modified by incorporating quaternary ammonium ions in the interlayer galleries. From our
previous work, this organoclay (I.44P Clay – B) shows a higher effective aspect ratio upon
dispersion in the linear PP matrix as well as a better strain hardening behavior in extensional
flow as compared to another clay with primary amine surfactant chemistry (I.30P Clay – A). The
organoclay used in this study was melt mixed at different loadings with linear maleated
polypropylene and several different nanocomposites were made. These nanocomposites were
61

tested for nonlinear, uniaxial extensional melt viscosity as well as low amplitude and high
amplitude oscillatory shear rheology. Also, the effect of increasing the test temperature on the
strain hardening behavior of these nanocomposites has been investigated.
4.2

EXPERIMENTAL TECHNIQUES

4.2.1 Materials
The matrix material is linear maleated polypropylene (PP-g-MA) (Exxelor PO1015) with
0.4 wt% maleic anhydride and a molecular weight of 180,000. (ζ =0.9 g/cc). It is important to
choose a PP-g-MA having a high molecular weight and substantial polymer chain length
necessary for the formation of hindered chain entanglements. Exxelor PO1015 was chosen
amongst several other PP-g-MA’s (Honeywell AC 950 and AC 1325) to fit the criteria for this
study. The melt mass flow rate (MFR) of the neat PP-g-MA matrix is 150g /10 min at 230°C
under a load of 2.16 kg and it melts at 145°C. The organoclay used for preparing the
nanocomposites was Nanocor I.44P which had quaternary ammonium ion surfactant with two C18 tails (Clay – B).
4.2.2 Compounding
PP-g-MA and the clay were dried overnight at 100°C and 20 mm Hg vacuum in a
vacuum oven and then compounded with maleated polypropylene at 165° C and at 150 rpm for
10 minutes in a Banbury mixer; this was done under nitrogen to avoid thermal degradation. One
batch of the mixer amounted to 47 grams of compound. The compounded material was removed
from the mixer with a brass scraper and ground up with a mini granulator. The different
nanocomposites and their formulation composition are given in Table 4.1. Based on the amount

62

of maleation on PP-g-MA and the weight of clay added to the polymer a ratio of g-mol of MA /
kg of clay is also given for the different nanocomposite formulations in the table.
4.3

CHARACTERIZATION TECHNIQUES

4.3.1 DSC and TGA Measurements
The melting temperature of the nanocomposite pellets as well as the neat polymer was
obtained using DSC runs on a TA DSC instrument. The melting point of neat PP-g-MA was
145C while the nanocomposites melted around 146.5C. Hence, the melt compounding was
carried out at 165C and compression molding was carried out at 175C which is well above the
melting points of these nanocomposites. Thermo gravimetric Analysis was carried out on the
pure clay, neat matrix as well the nanocomposites on a TA Q500 TGA instrument from 4°C to
600°C through a high resolution temperature ramp of 10°C / min.
4.3.2 X ray Diffraction Measurement
All the clays and nanocomposites were analyzed using X-ray diffraction patterns. The
characterization of the structure by XRD was performed on the compression-molded disks with a
Rigaku Rotaflex Ru-200BH X-ray diffractometer, which is equipped with a Ni-filtered Cu K α
radiation source and operated at 45 kV and 100 mA. The sample was scanned over a 2 range of
0.5° to 10° at a rate of 0.5°/min and measurements were recorded at equal increments of 0.01°.
4.3.3 Extensional Viscosity Measurement
An extensional viscosity fixture (EVF) mounted on a TA-ARES rheometer was used to
-1

-1

record the transient uniaxial extensional viscosity at strain rates ranging from 0.1s to 2s up to

63

a Hencky strain of 3. A 20-specimen mold was used for compression molding samples 18mm
long, 10mm wide and 0.7mm thick for the EVF tests. Test specimens were molded at 175° C and
the mold was cooled slowly using cooling water under a pressure of 5 tons. This ensured stress
free and relaxed homogenous samples with minimum dimensional variation. The molded
specimens were dried overnight before performing the extensional tests at two different test
temperatures --155° C and 160° C. The test samples were examined after completion of the test
and if any necking was noted in the stretched portion in the middle, the test was rejected.
4.3.4 Shear Rheology
Parallel plate geometry on a TA Instruments AR2000 rheometer was used to perform
frequency sweep tests on these nanocomposites. The diameter of the parallel plates was 40 mm
and a frequency range of 0.01 to 100  (rad/s) at 0.5% strain amplitude was chosen to perform
these tests. The corresponding storage modulus and complex shear viscosity were noted over the
entire frequency range. The strain amplitude was chosen such that it was in the linear viscoelastic
region and this was found out by running strain sweep tests at 1 rad/s on the nanocomposites. A
time sweep was also carried out on these nanocomposites on the 40 mm parallel plate geometry
for about 1 hour to check for the stability of modulus for these nanocomposites with respect to
time. Strain sweep tests were performed for probing into the non-linear viscoelasticity of the
nanocomposites by varying the strain amplitude % from 0.1 to 100. A continuous purge of
nitrogen was supplied during these tests to avoid thermal degradation.

64

4.4

RESULTS AND DISCUSSION

4.4.1 Dispersion of nanoclay and thermal stability of nanocomposites
The I.44P organoclay melt mixed with linear polypropylene has a quaternary onium ion
surfactant with two alkyl tails in the clay interlayer galleries. Heinz et al. [29] have commented
on the structure and conformation of the surfactants in the interlayer galleries. They have shown
through computations that the quaternary onium ion surfactant with two C-18 tails is more
mobile in the interlayer galleries and it also packs to a greater extent in these galleries. The extent
of surfactant packing can be judged from the X-ray diffraction pattern of the I.44P organoclay,
presented in Figure 4.1. It can be seen that the first X-ray diffraction peak for I.44P clay occurs

d-spacing:
I.44P (B) clay = 2.59 nm
B5 = 3.52 nm
B8 = 3.60 nm
B12 = 3.33 nm

Intensity

I.44P clay

PPNC B8

PPNC B12

PPNC B5

1

2

3

4

5

6

7

8

9

10

2deg)
Figure 4.1. X-ray diffraction of organoclay: a) I.44 P clay, b) B5 nanocomposite (5wt% clay),
c) B8 nanocomposite (8 wt% clay), d) B12 nanocomposite (12 wt% clay).
65

at a 2θ value of 3.4 degrees which corresponds to a basal spacing of 2.59 nm. We infer that the
greater basal spacing of the I.44P clay is mainly because of the greater extent of surfactant
packing in its interlayer galleries. Figure 4.1 also shows the XRD patterns for the different
nanocomposites made with 5, 8 and 12 wt% of the I.44P nanoclay. The nanocomposites with 5
and 8 wt% clay do not show a prominent peak in the XRD, but a presence of a slight shoulder in
the diffraction pattern. The d-spacing calculated for the nanoclay particles in these composites is
around 3.5 nm (greater than that of base clay) indicating that the interlayer galleries in clay have
been swelled by the polymer chains. The presence of this shoulder in XRD suggests that there
must be some intercalated clay structure present in these nanocomposites. Also, the shoulder
becomes more prominent and sharp in the 12 wt% filled nanocomposites with a decrease in the
d-spacing to 3.33 nm. This indicates that at higher loading, the dispersion of clay is not as
efficient and that there is an increase in the intercalated structure with more filler addition
creating a higher probability of particle-particle interaction than that of polymer –particle
interaction.
Figure 4.2 shows the %weight lost with increase in temperature for the clay, neat matrix
and nanocomposites. The dispersion of clay in the matrix has increased the thermal stability of
the composites by 50°C-55°C. The neat matrix decomposes completely around 400°C. The
amount of clay dispersed in the nanocomposites was calculated from the char yield at 600°C and
was about 5% for B5 nanocomposite and 7.9% for B8 nanocomposite.

66

100
90
80

% Weight

70
60
50
40

I.44P clay

30

Pure Matrix

20

PPNC B5

10

PPNC B8

0
0

100

200

300

400

500

600

Temperature (degree C)

Figure 4.2. Thermo Gravimetric Analysis (TGA) curves obtained for pure clay, neat matrix, and
nanocomposites at 10°C / min up to 600°C.

The stability of the modulus of maleated polypropylene matrix and the nanocomposites
with the 5wt% of the I.44P organoclay was also checked with the help of a time sweep. The time
sweep was carried out for about one hour and at 1 rad/s frequency and 0.5% strain amplitude. It
can be seen from Figure 4.3, that the modulus of the matrix and the nanocomposite was stable
and did not fluctuate over

67

1.5

160˚C, 0.5% strain, =1rad/s

1.4
1.3
1.2

G' / G'0

1.1
1

0.9
0.8
0.7

Matrix

0.6

PPNC B5

0.5
0

500

1000

1500

2000

2500

3000

3500

4000

t (sec)

Figure 4.3. Comparison of the increase in storage modulus for the matrix and B5 nanocomposite
at 160° C for 1 hour.
the time period of the test. The strain amplitude chosen was obtained by running strain sweep
tests on the nanocomposites at 1 rad/s angular frequency and was found to be well within the
linear viscoelastic regime.
4.4.2 Oscillatory Shear rheology
4.4.2.1 Linear Viscoelastic Measurements
The frequency sweep measurements done by parallel plate oscillatory shear rheology
yielded results of the dynamic modulus and dynamic viscosity of the nanocomposites, in the
linear viscoelastic regime. A comparison of the dynamic viscosities for all three composites as
68

well as the base polymer matrix is shown in Figure 4.4. The dynamic viscosity increases with an
increase in the nanoclay loading, it being highest for composite PPNC B12 over entire frequency
range.

1.E+05
155°C

' (Pa)

1.E+04

1.E+03

1.E+02

PPNC B5
PPNC B8

1.E+01

PPNC B12
Matrix

1.E+00
0.01

0.1

1

10

100

(rad/s)

Figure 4.4. Complex viscosity curves for the neat matrix (●) and nanocomposites: PPNC B5 (∆),
PPNC B8 (□), and PPNC B12(×)
The dynamic viscosity for the pure matrix is fairly constant at lower values of the
frequency. However, the dynamic viscosity for the nanocomposites does not show a complete
leveling off at lower frequencies, indicating a more elastic response in these nanocomposites.
The elastic or solid like response increases gradually from PPNC B5 to PPNC B8 composite and
is most prominent for PPNC B12 in the low amplitude frequency region. This is further evident

69

from Figure 4.5, which shows a comparison of the dynamic storage moduli of these
nanocomposites with the matrix.

1.E+05
155°C
1.E+04

G' (Pa)

1.E+03
1.E+02
1.E+01

PPNC B5
PPNC B8

1.E+00

PPNC B12
Matrix

1.E-01
1.E-02
0.01

0.1

1

10

100

(rad/s)

Figure 4.5. Storage Modulus (G’) curves versus frequency (for the neat matrix (●) and
nanocomposites: PPNC B5 (∆), PPNC B8 (□), and PPNC B12(×)
The nanocomposites show a higher modulus than the polymer matrix over the entire
frequency range and a plateau like region in the low frequency range, especially for the
nanocomposite with the greatest filler content. The increase in the low frequency storage
modulus was about hundred fold from the melt with 5wt% to that with 12wt% loading. As
discussed in chapter 3, the low frequency “plateau” in the storage modulus can be attributed to
two different contributions: (1) filler particle network formed by van der Waals interactions or
bridging and (2) entanglements between free chains and particle attached polymer chains. For
70

composites with higher filler loadings, it is seen in literature that an increase in the low
frequency modulus is mainly because of the filler network formed by particle-particle
interactions [97]. To understand and narrow down the actual contribution responsible for the
increase in viscosity and modulus values of the nanocomposites, we further investigated the
dispersion levels of the clay platelets at different nanoparticle loadings by calculating their
effective aspect ratio in the nanocomposites. This was estimated with the Pierce-Maron empirical
model that correlates the relative shear viscosity with the volume fraction of nanoparticles in a
concentrated suspension as shown in equation 3.1 and 3.2. The relative dynamic viscosities were
taken in the low frequency range. These viscosity ratios were correlated with that of volume
fractions for the nanoparticles and maximum packing fraction at each weight loading and the
effective aspect ratios were calculated as explained in chapter 3. The results of the effective
aspect ratios are tabulated in Table 4.1.
Table 4.1.
PPNC

Details of Aspect Ratio Estimation for Nanoparticles in Various Composites.
Clay

ϕ

loading

g-mol

ϕM

a=D/t

ϕ a

MA / kg
clay

wt %
B5

5

0.027

0.78

4.38

0.052

68

1.84

B8

8

0.044

0.47

13.7

0.060

59

2.55

B12

12

0.067

0.38

117

0.074

48

3.15

71

Note: Composition of Polypropylene Nanocomposites, amount of bound maleic anhydride / kg
of clay, relative dynamic viscosity (

), particle volume fraction (ϕ), Maximum packing

fraction (ϕM ) and effective aspect ratio (a) of nanocomposite are at 155°C

It can be seen that with an increase in the weight loading of the nanoparticles the
maximum packing fraction of the particles goes up and the effective aspect ratio decreases. This
can indicate that with more nanoparticles the dispersion level goes down progressively and there
is presence of more stacked particles in the nanocomposite. This was further evident in the XRD
patterns of the 5wt%, 8wt% and 12 wt% I.44P filled nanocomposites as discussed earlier. The
evidence from the XRD patterns and the decrease in effective aspect ratio with increase in filler
loading does suggest that the actual contribution responsible for the higher viscosity and
modulus values, especially in B12 nanocomposite, are mainly due to the filler network formed
by particle-particle interactions.
4.4.2.2 Non-Linear Viscoelastic Measurements
Non-Linear viscoelasticity in composites can be observed both in large amplitude
oscillatory shear and in extensional flow. In dynamic shear rheology, higher levels of strain
amplitude lead to strain softening of the dynamic moduli and the critical strain at which this
begins, scales with the loading level of filler particles. It has been discussed in chapter 3, that
higher levels of reinforcement for the linear PP and layered silicate nanocomposites is associated
with the formation of hindered chain entanglements between the free PP matrix chains and the
particle attached chains. It was also shown that at large strain amplitudes, the dynamic moduli
begin to drop at a critical strain. This drop was attributed to the breaking up of two structures—
the particle network followed by the hindered chain entanglement network involving attached
72

chains. Cassagnau [97] also has showed the effect of non-linearity in silica filled composites of
different polymeric materials. He investigated the effect of filler particles and their loadings on
the subsequent strain sweeps in nonlinear shear and pointed out that the degree of non-linearity
increases with filler concentrations. He established this occurrence as a result of the “Payne
effect” and also developed a correlation between the critical strain and the volume fraction of the
fumed silica particles.
In the present work, the nanocomposites having different nanoclay loading were tested
for their nonlinear viscoelastic behavior by performing strain sweep tests at a test temperature of
155°C and a frequency of 1 Hz.

16000

155°C

14000
PPNC B5

12000
PPNC B8

G' (Pa)

10000
PPNC B12

8000
6000
4000
2000
0
0.0001

0.001

0.01
0.1
Strain Amplitude

1

10

Figure 4.6. Strain sweep tests. Variation of storage modulus with strain amplitude for
PPNC B5 (●), PPNC B8 (∆), and PPNC B12(□) at 155°C and 1 Hz
73

The strain amplitude varied from 0.1% to 100% and the consequent storage and loss moduli were
recorded. Figure 4.6 shows the strain sweep measurements on PPNC B5, PPNC B8 and PPNC
B12. It was observed that the storage moduli increased with an increase in the filler loading for
all strain amplitudes. The modulus, however showed a sharp drop as the strain increased and this
was most evident in the highest filled composite – PPNC B12.

1.2
155°C
1

G'/G'0

0.8

0.6

0.4
PPNC B5
PPNC B8

0.2

PPNC B12
0
0.0001

0.001

0.01
0.1
Strain Amplitude

1

10

Figure 4.7. Strain sweep tests. Reduced Storage Modulus versus strain amplitude for
PPNC B5 (●), PPNC B8 (∆), and PPNC B12(□) at 155°C and 1 Hz.
Such a drop in the moduli values with increasing strain is a result of the strain softening
mechanism which is generally observed as the” Payne effect”. The causes of a drop in the
modulus or strain softening are, in order of importance (i) Breaking of particle-particle network,
(ii) Breaking of hindered chain entanglement network. Furthermore, it can be seen that the
74

“critical strain” at which the storage modulus begins to drop is lower for the higher filled
loadings. It is the lowest for PPNC B12, indicating that the non-linearity sets earliest in this
composite. This is more evident from Figure 4.7, which is a plot of the reduced shear modulus
v/s the strain. Also a maximum is observed in the storage as well as loss modulus for PPNC B12
nanocomposite just before the critical strain. This could suggest that with increasing strain
perturbations the filler network gets rearranged in the B12 composite such that it increases the
moduli values just before the critical strain. At the critical strain, the perturbation is large enough
to break down this filler network and causes a sudden drop in the moduli values. The loss moduli
of these nanocomposites, as shown in figure 4.8, also show a trend similar to that of the storage
moduli.

G" (Pa)

18000

155°C

16000

PPNC B5

14000

PPNC B8

12000

PPNC B12

10000
8000
6000
4000
2000
0
0.0001

0.001

0.01
0.1
Strain Amplitude

1

10

Figure 4.8. Strain sweep tests. Variation of loss modulus with strain amplitude for
PPNC B5 (●), PPNC B8 (∆), and PPNC B12(□) at 155°C and 1 Hz
75

To understand the effect of aspect ratio on non-linearity of these nanocomposites, we
calculated the specific surface area (Asp) of particles / g clay as shown in Table 4.2. It can be
seen that as the aspect ratio goes down with increasing filler loading, the overall specific surface
area available for interaction with the polymer chains also goes down. This area was then
correlated to the critical strain at which the storage modulus shows a sharp drop in strain sweep
experiments for each nanocomposite. Figure 4.9 shows such a correlation and it can be seen that
the critical strain decreases along with a decrease in the specific surface area (Asp).

1

Critical Strain (CR)

155°C

0.1

0.01
100

Asp (m2/g clay)

Figure 4.9. Critical Strain Plot: Variation of critical strain with respect to Specific Surface Area
(Asp) per gram of clay.

76

In the case of PPNC B12 it is seen from the linear viscoelastic data and the XRD patterns
that there is a higher probability of particle-particle interaction which is consistent with a lower
effective aspect ratio and a lower critical strain. As compared to that, PPNC B5 and PPNC B8
have much higher effective aspect ratios and greater specific surface area per gram of clay that is
available for interaction with polymer chains which lead to more chances of particle-polymer
interactions than particle-particle interactions. In order to break down the combination of these
two interactions, it requires slightly higher strain amplitude and as such the critical strain is
higher in the case of PPNC B8 and highest in the case of PPNC B5.
4.4.2.3 Modulus recovery after strain sweep
Another significant aspect of understanding the mechanisms of non-linear viscoelasticity
in filled polymer systems is the recovery of modulus following the breakdown of the filler
network caused by a large strain perturbation. Hence, similar to the work shown in previous
chapter between two different clays, we investigated the complex modulus recovery kinetics
from nonlinearity for the nanocomposites with different loadings. For this purpose, a time sweep
experiment was run at lower strain amplitude (in linear viscoelastic region) immediately after a
strain sweep experiment. Figure 4.10 shows the recovery kinetics of the storage modulus for all
three nanocomposites. It can be clearly seen here that PPNC B5 shows a much faster recovery as
compared to PPNC B8, whereas PPNC B12 does not show complete recovery in the given time
range of the experiment. Also it can be seen from the time sweep experiment that the storage
modulus keeps increasing even beyond the initial modulus that was measured before the strain
sweep experiment for PPNC B5 and PPNC B8. This could imply that the strain sweep
experiment acts as a trigger in rearranging the entanglement structure such that it requires much
more time to reach an equilibrium state and have a leveled out modulus value.
77

16000
155°C
14000
12000

G'(t)

10000
8000
315 sec
6000
PPNC B5
PPNC B8
PPNC B12

4000
234 sec
2000
0
10

100

1000

10000

t (s)

Figure 4.10. Storage modulus recovery versus time (z and γ = 0.5%). PPNC B5, PPNC
B8 and PPNC B12 nanocomposites at 155°C. G’ (t) is the storage modulus value as
obtained from time sweep test which is done immediately after one strain sweep.
The rate of recovery can also be compared between the nanocomposites by plotting the
ratio of loss factor observed during the recovery time sweep to that observed in the linear
viscoelastic region of the strain sweep experiment. Such a plot is shown in Figure 4.11. It can be
seen through this plot that the loss factor value for PPNC B12 is much greater than that of PPNC
B8 and PPNC B5. Also, since the drop in the storage modulus value after first time sweep was
greater in case of PPNC B8, it takes slightly more time to reach that initial value as compared to
PPNC B5. In case of PPNC B12 the time range of experiment is not sufficient to recover the
drop in modulus completely.

78

1.2
155°C
1.15

LF(t) / LF0

1.1
1.05
1
0.95
PPNC B5

0.9

PPNC B8
0.85

PPNC B12

0.8
10

100

1000

t (s)

10000

Figure 4.11. Rate of Recovery of the loss factor after initial strain sweep for PPNC B5, PPNC
B8 and PPNC B12. The Loss Factor LF (t) is that obtained from the time sweep test
and LF0 is the plateau value of loss factor obtained from initial strain sweep test.
This could indicate that in PPNC B12 the non-linearity causes a much greater breakdown
of the filler network and possibly disentanglement of the polymer chains, which hinders the
recovery of the composite modulus.
4.4.3 Uniaxial Extensional rheology
4.4.3.1 Effect of filler addition and test temperature
It was established in the previous chapter, that neat matrix doesn’t show any strain
hardening for the different extensional strain rates measured in the uniaxial extensional flow

79

(Figure3.15). Also, by the addition of just 5 wt% of nanoclay in the polymer matrix significant
amount of strain hardening is observed for the nanocomposites as seen in Figure 4.12.
1.E+07
5wt% I.44P clay

X 20

1.E+06

155°C

E0+
E+ (Pa.s)

1.E+05
160°C

E0+

1.E+04

0.1 s-1
0.5 s-1
1.0 s-1
2.0 s-1

1.E+03

1.E+02
0.05

0.5

5

50

t (sec)

Figure 4.12. Transient extensional viscosity curve for nanocomposite B5 at 155° C and 160° C
-1

-1

-1

-1

at four extensional strain rates: 0.1s (∆), 0.5s (□), 1s (◊) and 2s (○).
The addition of nanoclay provides the necessary interaction sites or cross link points on
the edges and faces of the clay for the functional groups on the maleated polypropylene chains.
The polymer chains form physical cross links with clay via hydrogen bonds and in turn entangle
with the surrounding free matrix chains. These kind of entanglements that are formed between
the particle attached polymer chains and free polymer chains are termed as hindered chain
80

entanglements and the increase in entanglements would in turn cause the polymer chains to relax
slowly upon stretching thereby increasing the relaxation times. Figure 4.12 shows the transient
extensional viscosity curves E

+

(t) for the nanocomposite with 5wt% I.44P clay at four

different strain rates and at two different test temperatures: 155˚C and 160° C.

The test

temperatures chosen were well above the melting point of the nanocomposites. While doing the
EVF tests care was taken such that the test sample does not sag or neck during the test. The
samples that sagged were discarded. Also, a relatively lower molecular weight, a low melting
point and a high melt flow rate (150 g/10min @ 230°C/2.16 kg) of the maleated polypropylene
matrix did not allow us to go to higher test temperatures on the extensional viscosity fixture.

4
5wt% I.44P clay

155C
160C

/0

3

2

1
0

0.5

1

1.5

2

Strain rate

Figure 4.13. Reduced Strain hardening parameter v/s strain rate: Comparison at Hencky strain
of 2.25, for PPNC B5 at 155° C (∆) and 160° C (♦)
81

The comparison of the extensional viscosity transients show that the viscosity curve rose
+

sharply from the base viscoelastic envelope 3η0 (t) for nanocomposite B5 at each strain rate for
both the test temperatures. It can also be seen that temperature does affect the strain hardening
behavior slightly, especially at higher strain rates. This is evident from a plot of /0 versus the
strain rate at a Hencky strain of 2.25 (Figure 4.13). A possible explanation for the decrease in
the strain hardening parameter is that the convective constraint release of the polymer chains
becomes more dominant at higher test temperatures as well as higher strain rates. The convective
constraint release leads to loss of the trapped entanglements and as such relaxes the overall
matrix from the applied extensional force. This reflects in a steeper drop of the strain hardening
parameter at 160°C and at higher strain rates. As explained in the previous chapter the reduced
strain hardening parameter was calculated by taking a ratio of the transient extensional viscosity
+

+

E to the viscosity obtained from linear viscoelastic measurements E0 . The base linear
viscoelastic envelope was obtained by multiplying the viscosity obtained from linear viscoelastic
measurements with the offset value between the base curve and linear portion of the transient
extensional viscosity.
4.4.3.2 Effect of clay loading on the extensional viscosity of nanocomposites
The effect of filler loading on the strain hardening behavior of the nanocomposites was
studied by comparing the extensional viscosities of nanocomposites with three different loadings
of I.44 P clay. As mentioned earlier, nanocomposites PPNC B5, PPNC B8 and PPNC B12 were
made with 5, 8 and 12 wt% of nanoclay loading respectively. A lower loading of nanoclay (3%)
was used in formulating a nanocomposite, however due to lower viscosity of the sample at
155°C; it was extremely difficult to measure its extensional viscosity transient because of its
82

sagging. Figure 4.14 shows a comparison of the extensional viscosities for the nanocomposites
with different loadings at 155° C.
1.E+09
155°C

PPNC B12
1.E+08

X 100

PPNC B8

E+ (Pa.s)

1.E+07

X 50

1.E+06

PPNC B5
X 10

1.E+05

0.1 s-1
0.5 s-1
1.0 s-1
2.0 s-1

1.E+04

1.E+03
0.05

0.5

5

50

t (sec)

Figure 4.14. Transient extensional viscosity curve for nanocomposite PPNC B5, PPNC B8,
-1

-1

and PPNC B12 at 155° C and four extensional strain rates: 0.1s (∆), 0.5s (□),
-1

-1

1s (◊) and 2s (○).

It is observed that PPNC B8 shows a slight reduction in the lift off of the extensional
viscosity curve, at higher rates as compared to the PPNC B5 nanocomposite. In case of PPNC
B12, the strain hardening is completely lost as no lift off in the extensional viscosity is observed
at all possible extensional strain rates. To quantify this effect, the strain hardening parameter was
calculated and compared at a Hencky strain of 2.25 at the different strain rates for only the
83

composites - PPNC B5 and PPNC B8. It was evident from Figure 4.14 that the strain hardening
parameter for PPNC B12 at all measured rates would be 1, as there is no strain hardening. The
strain hardening parameter comparison for PPNC B5 and PPNC B8 is shown in Figure 4.15 at
155°C. The reduced strain hardening parameter is clearly higher for PPNC B5 as compared to
PPNC B8 nanocomposite, especially at higher extensional strain rates.

3.5
155°C
3

2.5

/0

2

1.5

1

PPNC B5

0.5

PPNC B8
0
0

0.5

1
Strain rate

1.5

2

Figure 4.15. Reduced Strain hardening parameter v/s strain rate: Comparison at Hencky strain
of 2.25, for PPNC B5 (∆) and PPNC B8 (□) at 155° C.
The progressive decrease in strain hardening observed for the nanocomposites with
increasing loading is mainly be due to the decrease in the overall hindered chain entanglements.
With a decrease in the number of hindered chain entanglements, the overall relaxation of the
melt happens much faster and as such the strain hardening behavior starts diminishing. In the
84

nanocomposite melts, the hindered chain entanglements are mainly formed between the particle
attached polymer chains and the free polymer chains. With a decrease in the effective aspect
ratio of the particles with increasing filler loading, even the specific surface area of the particles
that is available for interaction with the polymer chains decreases. This in turn lowers the
probability of forming an entanglement network between the particle attached polymer chains
and the free matrix polymer chains, thereby reducing the strain hardening behavior in extensional
rheology. In order to understand the strain hardening trends with increasing filler loading, we
quantitatively estimated the particle statistics and the particle dimensions; discussed in the
following section.
4.4.3.3 Estimate of particle dimensions and inter-particle spacing
In the previous sections, we discussed that an increase in filler loading leads to greater
reinforcement levels in the linear viscoelastic regime. However in the non – linear viscoelastic
regime, an increase in filler loading causes strain softening or drop in the modulus as well as
suppression of the strain hardening behavior in extensional flow. This can be attributed to a
decrease in the effective aspect ratio of the particles as well as the specific surface area available
for polymer particle interactions which in turn reduces the probability of forming hindered chain
entanglements. In order to understand this hypothesis better, we evaluated the particle statistics
to obtain quantitative comparisons between the different nanocomposites. The effective aspect
ratio (a) has already been calculated earlier from the linear viscoelastic data (Table 4.1). In order
to obtain the particle thickness, we assumed an average particle length (Lp) in nm based on some
TEM analysis done in our research group on polypropylene –I.44P nanoclay composites. Taking
in account the aspect ratio and the average lengths of I.44P clay particles observed in TEM, the

85

average particle length was assumed to be 306 nm. The average particle length was then
multiplied with the effective aspect ratio for respective nanocomposites and the average particle
thickness (tp) was found. It can be seen from Table 4.2 that as the filler loading increases, the
effective aspect ratio goes down and the particle thickness increases.
Table 4.2.
PPNCs

Estimates of Particle Dimensions and Packing.
a=D/t

(Lp)

(tp)

(Asp)

(Vp)

nm

nm

m2/g clay

micron^3

(ϕ)

(n/Vp)

(δ)

nm

B5

68

306

4.5

327

4.21E-04

0.027

64

1329

B8

59

306

5.2

285

4.86E-04

0.044

91

1086

B12

48

306

6.4

233

5.97E-04

0.067

113

944

Note: Effective aspect ratio (a), Length of particle (Lp), Thickness of particle (tp), Specific
surface area (Asp), Particle volume (Vp), Number of particles (n) and Average lateral spacing
between particles (δ) of nanocomposite are at 155°C.

For a single platelet thickness of 1 nm and an average d-spacing of 3.5 nm for the
composites, we can conclude that with an increase in filler loading and particle thickness, the
number of clay platelets in a particle stack also increases. This indicates that for the higher
loading composites like PPNC B12, there are more clay platelets aggregated in a stack and as

86

such there is more particle-particle kind of interaction. This is further verified by calculating the
specific area (Asp) of the particles that is available for interaction. The specific area for the
2

particles was calculated in m /g of clay using the following equation:

(

)

Where, ρclay is the density of I.44P clay in this case. For a particle with length (Lp) and
thickness (tp), equation 4.1 becomes

(

)(

)

Equation 4.2 gives the specific surface area of the particles at different loadings and It can be
seen in Table 4.2, that as the filler loading increases from 5wt% to 12wt%, the specific area of
the particles available for interactions reduces by almost 30%. As such the chances of polymerparticle interaction goes down in the case of PPNC B12 composite which in necessary for
creating hindered entanglements between the particle attached polymer chains and the free
polymer chains. Besides this, if the particles are spaced very close to each other, there are more
chances of particle-particle kind of interactions which is observed from the linear and non-linear
viscoelastic data trends in the case of composites with higher filler loadings. In order to verify
this and get an estimate of the particle spacing, we first calculated the particle volume (Vp) in
3

micron using the average particle length and thickness as shown in Table 4.2. The total number
of particles per unit volume were then calculated by taking the ratio of particle volume (Vp) to
87

that of the volume fraction of the filler particles (ϕ) inclusive of the surfactant between them. It
can be seen in Table 4.2 that the number of particle per unit volume increases as the filler
loading increases from 5 wt% to 12 wt%. With an increase in the number of particles in a given
volume, the lateral spacing between them decreases. To calculate the average lateral spacing
between the particles, we used the mean chord length formula as used in the work done by Olson
and co-workers [123-124]. They have shown that in a binary medium the discs can be arranged
either in a tightly packed manner with a coordination number of 4 or in a hexagonal pattern with
a coordination number of 6. In our case we first assumed the filler particles to be disc shaped and
used the mean chord length formula for discs in a tightly packed pattern to calculate the average
lateral spacing between the nanoparticles. According to Olson, the area or volume fraction of the
discs also gives the probability (p1) that a given point is inside a disk or sphere as shown in
equation (4.3). For a tightly packed pattern
2

p1= π (r/δ) …………. (4.3),

Where, r is the radii of the disc (i.e. L/2*√ ) and δ is the lateral spacing between the
adjacent discs. In order to obtain δ, we first calculated the area fraction values from the
respective volume fraction of the fillers in the different composites. These area fraction values
are the same as p1, as stated in the work done by Olson. We then used equation 4.3 and
calculated the lateral spacing (δ) for each composite shown in Table 4.2. It can be seen that with
an increase in filler loading the lateral spacing between the particles decreases which is also
consistent with an increase in the number of particles in a given volume. The lateral spacing
between particle stacks for all nanocomposites are however too large to cause any kind of
bridging effect between them and as such it can be concluded that bridging doesn’t contribute to
88

any kind of filler network formation; it is mainly formed due to direct particle-particle
interaction. The estimation of particle statistics gives a quantitative analysis of the level of
dispersion of the filler particles in the different nanocomposites as well as an insight of the
possible mechanisms responsible for their shear and extensional flow behavior.
4.5

CONCLUSIONS
The quality of dispersion is lowered with increasing volume fraction (from 2.7% to 6.7%

vol.) of the same layered silicates in a high molecular weight (Mw=180,000) linear maleated
polypropylene copolymer (with ethylene) which melts at 145°C; no other component was used.
This was confirmed with both XRD and estimates of effective particle aspect ratio from relative
dynamic viscosity. The low frequency storage modulus was increased a hundred fold from the
melt with lowest to that with the highest loading and this may be attributed largely to the
stronger filler network with the highest loading.

Observations with a strain sweep (large

amplitude oscillatory shear) on the nanocomposite with the highest loading revealed a maximum
in both dynamic moduli with increasing strain for the nanocomposite melt before they dropped in
magnitude. The strain sweep on the nanocomposite with the best dispersion (and the lowest
loading) showed a two-stage drop. The critical strain decreased with increasing loading and
decreasing effective aspect ratio.
Two nanocomposite melts (B5 and B8) with lower loading of nanoparticles and better
dispersion showed significant strain hardening in uniaxial extensional flow tests over a range of
strain rates while the matrix polymer melt did not show any strain hardening. In the case of B12,
with highest filler loading, the strain hardening was completely lost. The specific surface area
(Asp) of particles that is available for interaction with the polymer chains decreases by almost
89

30% for the highest filled composite as compared to the lowest filled one. With a progressive
decrease in specific surface area, there is a progressively lower probability of forming a network
of hindered chain entanglements between the particle-attached polymer chains and free polymer
chains that is necessary to achieve strain hardening.

90

CONCLUSIONS AND RECOMMENDATIONS
________________________________________________________________________
CHAPTER 5
5.1

CONCLUSIONS
The research work in the present study was focused on achieving strain hardening in

linear polypropylene matrix that was functionalized at ends with maleic anhydride. Although,
there have been several methods of achieving strain hardening in linear polymeric systems by
either modifying the polymer molecular structure or by incorporating other components like high
molecular weight fractions, cross-linked polymers etc., here , we have taken the approach of
adding a very small amount of an inorganic nano-sized filler to a linear maleated polyproypelene
copolymer. In doing this we have addressed several issues that are necessary to understand the
effect of nano filler addition into linear polymers based on shear and extensional rheology:
5.1.1 Addition of nanoclay and its effect on viscoelastic properties and strain hardening
Linear polypropylene, which is a widely used commodity polymer, does not exhibit any
strain hardening upon stretching it in uniaxial extensional melt flow. In order to verify if the
addition of nanoclay provided the necessary strengthening of PP melt upon stretching, we added
different nanoclays to the linear maleated polypropylene matrix. The primary objective in
choosing just the functionalized (maleated) polypropylene as the complete matrix was to
understand the kind of interactions taking place between the maleic anhydride group on the
polymer chains and the -OH groups on the clay edges, which would lead to any kind of strain
hardening in the nanocomposites. The addition of linear bulk polypropylene would add a third
phase making the system more complex. Hence a linear maleated polypropylene (PP-g-MA)
91

(Mw = 180,000, 0.4% maleic anhydride) was chosen as the matrix to which layered silicates
(Montmorillonite) having different surfactant treatments were added as the filler. The X-ray
diffraction patterns suggested that the nanoclays dispersed well in the polypropylene matrix at 2
vol%. The additions of nanoclay lead to a substantial increase in the dynamic moduli and
viscosity values as compared to the base matrix in the linear viscoelastic region. The
nanocomposites also showed greater degree of non-linearity in large amplitude oscillatory shear
measurements as compared to the base polymer with a steep drop in the moduli values beyond a
critical strain. But the most striking result was the marked improvement in strain hardening
behavior in melt extensional flow for the nanocomposites. The study showed that by addition of
just 5 wt% of nanoclay to linear polypropylene matrix, the extensional viscosity as well as the
strain hardening behavior of the linear matrix improves as seen for the nanocomposites. The
improvement in strain hardening behavior is mainly because of the formation of hindered chain
entanglements between the free matrix chains and the PP chains that are physically cross linked
to the clay sites. Upon stretching, the polymer chains try to relax to regain their original state,
however, the formation of the hindered chain entanglement structure slows down this polymer
chain relaxation and this reflects in a lift off of the extensional viscosity curve from the base
linear viscoelastic curve for the composites. This lift off is the observed strain hardening in
nanoclay filled polypropylene composites.
5.1.2 Effect of surfactant head group and effective aspect ratio
To further investigate the effect of nanoclay addition, two different grades of
nanoclay were added in the linear polymer matrix to check for their effects on the viscoelastic
and extensional flow properties. One of the clay was organically modified by incorporating

92

primary amine surfactant (I.30P clay) and other had a quaternary ammonium ion surfactant
(I.44P clay) in the interlayer galleries. The amount of surfactant packing in the interlayer
galleries of the clays was verified by X-ray diffraction and Thermo gravimetric Analysis and was
concluded that the I.44P organoclay having quaternary ammonium surfactant was packed more
with the surfactant and had a higher d-spacing. The dispersion of these clays in the matrix PP
was fairly intercalated as seen from a shoulder in the XRD pattern of the nanocomposites. The
addition of both these clays improved the dynamic modulus and viscosity values in linear shear,
with I.44P clay nanocomposite (PPNC B5) showing a greater value than the I.30P clay
nanocomposite (PPNC A5).

The aspect ratios (side / thickness) of the two different

nanoparticles dispersed were determined to be 56 and 68 from measurements of relative dynamic
viscosity on the nanocomposite melts. The low frequency storage modulus of the nanocomposite
with the higher aspect ratio nanoparticles was greater indicating a stronger filler network and/or a
greater fraction of polymer chains attached to the particle surface with higher aspect ratio. Large
amplitude oscillatory shear measurements on the nanocomposite melt with the higher aspect
ratio particles showed a two stage drop in the storage modulus with increasing strain which
highlights the breakup of two structures—the particle network followed by the hindered chain
entanglement network involving attached chains. The buildup of structure is also slower for the
nanocomposite with the higher aspect ratio nanoparticle system. Both nanocomposite melts
display strain hardening in uniaxial extensional flow over several strain rates, unlike the matrix
polymer which shows no strain hardening: the melt with the higher aspect ratio layered silicates
displayed a higher extent of strain hardening at all rates. This last phenomenon may be attributed
to the greater incidence of particle attached chains and entanglements involving these chains
when the particle aspect ratio is greater.

93

5.1.3 Effect of test temperature and nanoclay loading
The second part of the research work was focused on investigating the effects of different
loadings of the nanoclay and the test temperature on the strain hardening behavior of these
nanocomposites. From the first part of the research work, it was concluded that the organoclay
with higher effective aspect ratio and quaternary ammonium surfactant with 2 alkyl tails showed
better rheological properties. The same clay was used at different loadings in the linear PP matrix
in second part of the research and the linear shear, non-linear shear and uniaxial extensional flow
properties were investigated. With respect to different test temperatures in uniaxial extensional
flow, the nanocomposite with a loading of 5wt% (PPNC B5) was used. It was found that the
strain hardening of the nanocomposites decreased with an increase in temperature from 155°C to
160°C especially at higher strain rates which is due to the dominant convective constraint release
at those higher test temperatures as well as higher strain rates.
The loading of filler particles also significantly affects the linear and non-linear
viscoelastic properties as well as the strain hardening behavior of the nanocomposites. A lower
loading of nanoclay (3%) was used in formulating a nanocomposite, however due to lower
viscosity of the sample at 155°C; it was extremely difficult to measure its extensional viscosity
transient because of its sagging. The other loadings were at 5wt% (2.7 vol%), 8wt% (4.4 vol%)
and 12wt% (6.7 vol%) of the nanoclay. It was seen that the quality of dispersion is lowered with
increasing volume fraction (from 2.7% to 6.7% vol.) of the same layered silicates in a high
molecular weight maleated linear polypropylene copolymer (with ethylene). This was confirmed
with both XRD and estimates of effective particle aspect ratio from relative dynamic viscosity.
The low frequency storage modulus was increased a hundred fold from the melt with lowest to
that with the highest loading and this may be attributed largely to the stronger filler network with
94

the highest loading. Observations with a strain sweep (large amplitude oscillatory shear) on the
nanocomposite with the highest loading revealed a maximum in both dynamic moduli with
increasing strain for the nanocomposite melt before they dropped in magnitude. The strain sweep
on the nanocomposite with the best dispersion (and the lowest loading) showed a two-stage drop.
The critical strain decreased with increasing loading and decreasing effective aspect ratio.
Two nanocomposite melts (B5 and B8) with lower loading of nanoparticles and better
dispersion showed significant strain hardening in uniaxial extensional flow tests over a range of
strain rates while the matrix polymer melt did not show any strain hardening. In the case of B12,
with highest filler loading, the strain hardening was completely lost. The specific surface area
(Asp) of particles that is available for interaction with the polymer chains decreases by almost
30% for the highest filled composite as compared to the lowest filled one. With a progressive
decrease in specific surface area, there is a progressively lower probability of forming a network
of hindered chain entanglements between the particle-attached polymer chains and free polymer
chains that is necessary to achieve strain hardening.
5.2

RECOMMENDATIONS
The research done in this study still has a wide scope and provides further opportunities

to explore other aspects that were not covered in here. The following recommendations are
therefore made:
1.

One of the main goals of this study was to study the effect of nanoclay addition on the
strain hardening behavior of linear polypropylene nanocomposites and understand the
interaction between the nanoclay platelets and the functional groups on maleated
polypropylene. It was conclusive that a higher effective aspect ratio and better interaction
via hydrogen bonding, led to greater loose reversible attachments between the polymer
95

chains and clay particles. This in turn helped in increasing the entanglement density of
the nanocomposites and also the strain hardening improved, however it still remains to
find out the exact nature of these interactions and also if possible to quantify the
hydrogen bonding interactions between different composites. This can also be related to
the entanglement density of the nanocomposites and the strain hardening parameters.
2.

Also in this study we had prepared nanocomposites using two different types of
organoclays. One of the clay had a primary amine surfactant with a single alkyl tail and
other a quaternary ammonium surfactant with 2 alkyl tails. In future a study on the
extensional behavior of the nanocomposites can be made wherein the clay surfactant head
group is same and the numbers of alkyl tails vary. Also, one can have different surfactant
head groups with the same number of alkyl tails dangling. Heinz and coworkers [38, 39]
have worked and simulated the conformations of the different surfactants in interlayer
galleries with increasing number of C (C4 to C22) atoms in the alkyl tails as well with
increasing number of alkyl tails. Such organoclays can be incorporated into the polymer
matrix and investigated for their effects on the dispersion levels, linear and non-linear
viscoelastic properties as well the extensional flow properties for the nanocomposites.

3.

The non-linearity in shear and extension can be further explored using a combination of
different molecular weight polymer matrices as well as different clays. Also the modulus
recovery kinetics can be studied with respect to differences in the molecular weight of the
polymer matrix and different clay loadings.

4.

The effects of using additional coupling agents such as silanes on nanoclay can be
investigated for the non-linear shear and extensional behavior. Silane functional groups in

96

general provide additional interaction sites as has been shown by research work done
previously in our group [100] and significantly improve the linear viscoelastic properties.
5.

The non-linear viscoelastic and extensional behavior of these nanocomposites can be
modeled by identifying the shear and extensional damping functions as was done by
Wagner and Fulchiron for polymer melts. As has been shown by Fulchiron [121] for PP
melts the transient shear data can be used to identify the elongation behavior using
Wagner’s time integral constitutive model. A similar modeling can be done with shear
viscoelastic data of the nanocomposites and the transient extensional viscosity may be
predicted.

97

BIBLIOGRAPHY

98

BIBLIOGRAPHY
1.

Manias, E., Touny, A., Wu, L., Strawhecker, K., Lu, B., and Chung, T.,
“Polypropylene/Montmorillonite nanocomposites. Review of the synthetic routes and
materials properties.” Chem. Mater., 13, (10), (2001), 3516-3523.

2.

Qian, G., Cho, J.W. and Lan, T., “Preparation and Properties of Polyolefin
Nanocomposites.” Polyolefins 2001, Houston, TX, (2001).

3.

Bartholmai, M. and Schartel, B., “Layered silicate polymer nanocomposites: new
approach or illusion for fire retardancy? Investigations of the potentials and the tasks
using a model system.” Polym. Advan Technol., 15, (7), (2004), 355-364.

4.

Gilman, J., Jackson, C., Morgan, A., Harris, R., Manias, E., Giannelis, E., Wuthenow,
M., Hilton, D., and Phillips, S., “Flammability properties of polymer - Layered-silicate
nanocomposites. Polypropylene and polystyrene nanocomposites.” Chem. Mater., 12, (7),
(2000), 1866-1873.

5.

Tang, Y., Hu, Y., Li, B., Liu, L., Wang, Z., Chen, Z., and Fan, W.,
“Polypropylene/montmorillonite nanocomposites and intumescent, flame-retardant
montmorillonite synergism in polypropylene nanocomposites.” J. Poly. Sci. Poly. Chem.,
42 (23), (2004), 6163-6173.

6.

Tang, Y., Hu, Y., Wang, S., Gui, Z., Chen, Z., and Fan, W., “Intumescent flame
retardant-montmorillonite synergism in polypropylene-layered silicate nanocomposites.”
Poly. Int., 52 (8), (2003), 1396-1400.

7.

Usuki,A., Kojima,Y., Kawasumi,M., Okada,A., Fukushima,Y., Kurauchi,T., and
Kamigaito,O., Synthesis of Nylon 6 - clay Hybrid, J. Mater. Res, 8, (1993), 1179-1184.

8.

Kojima, Y., Usuki, A., Kawasumi, M., Okada, A., Kurauchi, T., and Kamigaito, O.,
Synthesis of Nylon-6-Clay Hybrid by Montmorillonite Intercalated with EpsilonCaprolactam, J. Polym. Sci., Part A: Polym. Chem., 31, (1993), 983-986.

9.

Giannelis, E.P., Polymer Layered Silicate Nanocomposites, Adv. Mater., 8, (1996), 2935.

99

10.

Giannelis, E.P., Polymer Layered Silicate Nanocomposites: Synthesis, Properties, and
Applications, Appl. Organomet. Chem., 12, (1998), 675-680.

11.

Ray, S. and Okamoto, M., Polymer/layered silicate nanocomposites: a review from
preparation to processing, Prog. Polym. Sci., 28, (2003), 1539-1641.

12.

Vaia, R. and Giannelis, E., Polymer melt intercalation in organically-modified layered
silicates: Model predictions and experiment, Macromolecules, 30, (25), (1997), 80008009.

13.

Kawasumi, M., Hasegawa, N., Kato, M., Usuki, A. and Okada, A. Preparation and
mechanical properties of polypropylene-clay hybrids, Macromolecules, 30, (1997), 63336338.

14.

Kato, M., Usuki, A. and Okada, A., Synthesis of polypropylene oligomer-clay
intercalation compounds, J. Appl. Poly. Sci., 66, (1997), 1781-1785.

15.

Hasegawa, N., Kawasumi, M., Kato, M., Usuki, A., and Okada, A., Preparation and
mechanical properties of polypropylene-clay hybrids using a maleic anhydride-modified
polypropylene oligomer, J. Appl. Poly. Sci., 67, (1998), 87-92.

16.

Usuki, A., Kato, M., Okada, A., and Kurauchi, T., “Synthesis of polypropylene-clay
hybrid.” J. Appl. Polym. Sci., 63, (1), (1997), 137-139.

17.

Ginzburg, V. and Balazs, A., “Calculating phase diagrams of polymer-platelet mixtures
using density functional theory: Implications for polymer/clay composites.”
Macromolecules, 32, (17), (1999), 5681-5688.

18.

Hasegawa, N., Okamoto, H., Kawasumi, M., Kato, M., Tsukigase, A., and Usuki, A.,
“Polyolefin-clay hybrids based on modified polyolefins and organophilic clay.”
Macromol. Mater. Eng., 280, (7-8), (2000), 76-79.

19.

Hasegawa, N. and Usuki, A., “Silicate layer exfoliation in polvolefin/clay
nanocomposites based on maleic anhydride modified polyolefins and organophilic clay.”
J. Appl. Polym. Sci., 93, (1), (2004), 464-470.

100

20.

Gopakumar, T., Lee, J., Kontopoulou, M., and Parent, J., “Influence of clay exfoliation
on the physical properties of montmorillonite/polyethylene composites.” Polymer, 43,
(20), (2002), 5483-5491.

21.

Ishida, H., Campbell, S., and Blackwell, J., “General approach to nanocomposite
preparation.” Chem. Mater., 12, (5), (2000), 1260-1267.

22.

Marchant, D. and Jayaraman, K., “Strategies for optimizing polypropylene-clay
nanocomposite structure.” Ind. Eng. Chem. Res., 41, (25), (2002), 6402-6408.

23.

Minoura, Y., Ueda, M., Mizunuma, S., and Oba, M., “The Reaction of Polypropylene
with Maleic Anhydride.” J. Appl. Polym. Sci., 13, (1969),1625.

24.

Van Olphen, H., An Introduction to Clay Colloid Chemistry; Interscience Publishers:
New York, (1962).

25.

Theng, B.K.G., Soils with Variable Charge., New Zealand: Offset Publications (1980).

26.

Greenwell, H.C., Harvey, M.J., Boulet, P., Bowden, A., Coveney, P.V., Whiting, A.,
Interlayer structure and bonding in nonswelling primary amine intercalated clays,
Macromolecules, 38, (2005), 6189-6200.

27.

Osman, MA; Seyfang, G; Suter, UW.. Two-dimensional melting of alkane monolayers
ionically bonded to mica, Journal of Physical Chemistry B, 104 (18), (2000), 4433-4439.

28.

Osman, MA; Ernst, M; Meier, BH; Suter, UW., Structure and molecular dynamics of
alkane monolayers self-assembled on mica platelets, Journal of Physical Chemistry B,
106 (3), (2002), 653-662.

29.

Heinz, H; Castelijns, HJ; Suter, UW, Structure and phase transitions of alkyl chains on
mica. Journal of American Chemical Society, 125 (31), (2003), 9500-9510.

30.

Osman, MA; Ploetze, M; Skrabal, P, Structure and properties of alkylammonium
monolayers self-assembled on montmorillonite platelets, Journal of Physical Chemistry
B, 108 (8), (2004), 2580-2588.

101

31.

Zhu, JX; He, HP; Zhu, LZ; Wen, XY; Deng, F. Characterization of organic phases in the
interlayer of montmorillonite using FTIR and C-13 NMR, Journal of Colloid and
Interface Science, 286 (1), (2005), 239-244.

32.

Vaia, RA; Teukolsky, RK; Giannelis, EP., Interlayer Structure and Molecular
Environment of Alkylammonium Layered Silicates, Chemistry of Materials, 6 (7),
(1994), 1017-1022.

33.

Vaia, R. and Giannelis, E., “Lattice model of polymer melt intercalation in organicallymodified layered silicates.” Macromolecules, 30, (25), (1997), 7990-7999.

34.

Vaia, R. and Giannelis, E., “Polymer melt intercalation in organically-modified layered
silicates: Model predictions and experiment.” Macromolecules, 30, (25), (1997), 80008009.
Osman, MA; Rupp, JEP; Suter, UW, Gas permeation properties of polyethylene-layered
silicate nanocomposites, Journal of Material Chemistry, 15 (12), (2005), 1298-1304.

35.

36.

Yalcin, B; Cakmak, M., Superstructural hierarchy developed in coupled high shear/high
thermal gradient conditions of injection molding in nylon 6 nanocomposites, Polymer, 45
(8), (2004), 2691-2710.

37.

Jacobs, JD; Koerner, H; Heinz, H; Farmer, BL; Mirau, P; Garrett, PH; Vaia, RA.,
Dynamics of alkyl ammonium intercalants within organically modified montmorillonite:
Dielectric relaxation and ionic conductivity, Journal of Physical Chemistry:B, 110 (41),
(2006), 20143-20157.

38.

Heinz, H., Vaia, R.A. Krishnamoorti, R. and Farmer, B.L., Self-assembly of
alkylammonium chains on montmorillonite: Effect of chain length, head group structure,
and cation exchange capacity, Chem. Mater., 19, (2007), 59-68.

39.

Fu, YT; Heinz, H., Cleavage Energy of Alkylammonium-Modified Montmorillonite and
Relation to Exfoliation in Nanocomposites: Influence of Cation Density, Head Group
Structure, and Chain Length, Chemistry of Materials, 22 (4), (2010), 1595-1605.

40.

Agrawal, PK; Lee, WK; Lornston, JM; RIichardson, CI; Wissburn, KF; Metzner, AB.,
Rheological behavior of molten polymer in shear and in extensional flows, Transactions
of the Society of Rheology, 21 (3), (1977), 355-379.

102

41.

Wagner, MH, The Nonlinear Strain Measure of Polyisobutylene Melt in General Biaxial
Flow & its Comparison to the Doi-Edwards Model., Rheologica Acta, 29 (6), (1990),
594-603.

42.

Sridhar, T; Gupta, RK., Material Properties of Viscoelastic Liquids in Uniaxial
Extension, Journal of Rheology, 35 (3), (1991), 363-377.

43.

Wilson, TS, Baird, DG, Transient Elongational Flow Behavior of Thermotropic Liquid
Crystalline Polymer, Journal of Non-Newtonian Fluid Mechanics, 44,( 1992), 85-112.

44.

Kim.HC, Pendse.A., Collier JR., Polymer Melt Lubricated Elongational Flow, Journal of
Rheology, 38 (4), (1994), 831-845.

45.

PADMANABHAN, R.. Measurement of Extensional Viscosity of Viscoelastic Liquid
Foods, Journal of Food Eng., 25 (3), (1995), 311-327.

46.

Kelly, J; Graessley, WW.. Some tests of constitutive equations for entangled polymers
based on planar-extension flow histories in a periodically constricted channel,
Rheologica Acta, 35 (1), (1996), 24-38.

47.

Micic, P; Bhattacharya, SN; Field, G., Melt strength and elastic behaviour of
LLDPE/LDPE blends. International Polymer Processing, 11 (1), (1996), 14-20.

48.

Takahashi, T; Wu, WG; Toda, H; Takimoto, J; Akatsuka, T; Koyama, K., Elongational
viscosity of ABS polymer melts with soft or hard butadiene particles.Journal of NonNewtonian Fluid Mechanics, 68 (2-3), (1997), 259-269.

49.

Levitt, L; Macosko, CW; Schweizer, T; Meissner, J, Extensional rheometry of polymer
multilayers: A sensitive probe of interfaces, Journal of Rheology, 41 (3), (1997), 671685.

50.

Lin, GG; Hu, MC, Measurement of elongation viscosity for polymer melts by fiber
spinning. Advances in Polymer Technology, 16 (3), (1997), 199-207.

51.

Greener, J; Evans, JRG, Measurement of elongational flows in ceramic processing –
Review, Journal of the European Ceramic Society, 17 (10), (1997), 1173-1183.
103

52.

Kasehagen, LJ; Macosko, CW, Nonlinear shear and extensional rheology of long-chain
randomly branched polybutadiene. Journal of Rheology, 42 (6), (1998), 1303-1327.

53.

Kapoor, B; Bhattacharya, M., Transient shear and extensional properties of biodegradable
polycaprolactone, Polymer Eng. And Science, 39 (4), (1999), 676-687.

54.

Haworth, B; Jumpa, S; Miller, NA, Melt-state elongational rheometry of mineral-filled
polyethylene, Polymer Testing, 19 (4), (2000), 459-477.

55.

Micic, P; Bhattacharya, SN, Rheology of LLDPE, LDPE and LLDPE/LDPE blends and
its relevance to the film blowing process, Polymer International, 49 (12), (2000), 15801589.

56.

Minegishi, A; Nishioka, A; Takahashi, T; Masubuchi, Y; Takimoto, J; Koyama, K.,
Uniaxial elongational viscosity of PS/a small amount of UHMW-PS blends, Rheologica
Acta, 40 (4), (2001), 329-338.

57.

Sugimoto, M; Masubuchi, Y; Takimoto, J; Koyama, K., Melt rheology of polypropylene
containing small amounts of high-molecular-weight chain. 2. Uniaxial and biaxial
extensional flow, Macromolecules, 34 (17), (2001), 6056-6063.

58.

Muke, S; Ivanov, I; Kao, N; Bhattacharya, SN, Extensional rheology of polypropylene
melts from the Rheotens test, Journal of Non-Newtonian Fluid Mechanics, 101 (1-3),
(2001), 77-93.

59.

Malmberg, A; Gabriel, C; Steffl, T; Munstedt, H; Lofgren, B., Long-chain branching in
metallocene-catalyzed polyethylenes investigated by low oscillatory shear and uniaxial
extensional rheometry, Macromolecules, 35 (3), (2002), 1038-1048.

60.

Huang, JC; Leong, KS, Shear viscosity, extensional viscosity, and die swell of
polypropylene in capillary flow with pressure dependency, Journal of Applied Polymer
Science, 84 (6), (2002), 1269-1276.

61.

Kao, N; Chandra, A; Bhattacharya, S, Melt strength of calcium carbonate filled
polypropylene melts, Polymer International, 51 (12), (2002), 1385-1389.

104

62.

Hachmann, P; Meissner, J.. Rheometer for equibiaxial and planar elongations of polymer
melts. Journal of Rheology, 47 (4), (2003), 989-1010.

63.

Sentmanat, ML, Miniature universal testing platform: from extensional melt rheology to
solid-state deformation behavior, Rheologica Acta, 43 (6), (2004), 657-669.

64.

Hong, JS; Ahn, KH; Lee, SJ., Strain hardening behavior of linear polymer melts, KoreaAustralia Rheology Journal, 16 (4), (2004), 213-218.

65.

Sentmanat, M; Muliawan, EB; Hatzikiriakos, SG, Fingerprinting the processing behavior
of polyethylenes from transient extensional flow and peel experiments in the melt state.
Rheologica Acta, 44 (1), (2004), 1-15.

66.

Gupta, RK; Pasanovic-Zujo, V; Bhattacharya, SN, Shear and extensional rheology of
EVA/layered silicate-nanocomposites. Journal of Non Newtonian Mechanics, 128 (2-3),
(2005), 116-125.

67.

Bhattacharya, S; Gupta, RK; Jollands, M; Battacharya, SN., Foaming Behavior of HighMelt Strength Polypropylene/Clay Nanocomposites, Polymer Eng. And Sci., 49 (10),
(2009), 2070-2084.

68.

Chaudhary, A.K.; Jayaraman, K., "Extrusion of
Nanocomposite Foams," Polym. Eng. Sci., 51, (2011).

69.

Spitael, P. and C.W. Macosko, Strain hardening in polypropylenes and its role in
extrusion foaming, Polym. Eng. Sci., 44, (2004), 2090–2100.

70.

Stange, J. and H. Munstedt, Rheological properties and foaming behavior of
polypropylenes with different molecular structures, J. Rheol., 50, (2006), 907-923.

71.

Kurzbeck, S., F. Oster, H. Münstedt, T. Q. Ngyuen, and R. Gensler, Rheological
properties of two polypropylenes with different molecular structure, J. Rheol., 43, (1999),
359–374.

72.

Gotsis AD, BLF Zeevenhoven, AH Hogt, The effect of long chain branching on the
processability of polypropylene in thermoforming, Polym. Eng. Sci., 44, (2004), 973-982.

105

Linear

Polypropylene-Clay

73.

Wollecke F, Altstadt V, Sandler J, Rakutt D, The influence of melt elongational
properties on the morphology of PES cellular materials, Cellular Polymers, 21, (2002),
445-453.

74.

McInerney, L. N. Kao and S.N. Bhattacharya, Melt strength and extensibility of talcfilled polypropylene, Polym. Eng. Sci., 43, (2003), 1821-1829.

75.

Sugimoto, M., Tanaka, T., Y. Masubuchi, J. Takimoto and K. Koyama, Effect of chain
structure on the melt rheology of modified polypropylene, Journal of Applied Polymer
Science, 73, (1999), 1493-1500.

76.

Sugimoto, M., Y. Masubuchi, J. Takimoto and K. Koyama, Melt rheology of
polypropylene containing small amounts of high-molecular-weight chain. 2. Uniaxial and
biaxial extensional flow, Macromolecules, 34, (2001), 6056-6063.

77.

Minegishi, A; Naka, Y; Takahashi, T; Masubuchi, Y; Takimoto, J; Koyama, K, The
effect of ultrahigh molecular weight polymers on the nonlinear response in uniaxial
elongational viscosity, Nihon Reorji Gakkaishi ,25 (4), (1997), 215-216.

78.

Berger.L, Meissner, J., Linear Viscoelasticity, Simple and Planar Melt Extension of
Linear Polybutadienes with Bimodal Molar Mass Distributions, Rheologica Acta ,31 (1),
(1992), 63-74.

79.

Yamaguchi.M and Miyata.H., Strain hardening behavior in elongational viscosity for
binary blends of linear polymer and crosslinked polymer, Polymer Journal, 32, (2000),
164-170.

80.

Yamaguchi.M, Rheological Properties of linear and crosslinked polymer blends: Relation
between crosslink density and enhancement of elongational viscosity, Journal of Polymer
Science: Part B: Polymer Physics, 39, (2001), 228-235.

81.

Yamaguchi.M, Suzuki.K, Rheological Properties and foam process ability for blends of
linear and crosslinked polyethylenes, Journal of Polymer Science: Part B: Polymer
Physics, 39, (2001), 2159-2167.

82.

Ogura.K, Takahashi.M., Uniaxial extension behavior of cross-linked Poly(methyl
methacryalte)s with various degrees of cross linking, J. Soc. Rheol., 31, (2003), 85-89.
106

83.

Takahashi, H; Ishimuro, Y; Watanabe, H. 2008. Nonlinear mechanical Behavior of
scarcely crosslinked poly(dimethyl siloxane) gel: Effect of strand length polydispersity.
Polymer, 40 (5): 465-474

84.

Takahashi.T., Takimoto.J, and Koyama.K, Uniaxial elongational viscosity of various
molten polymer composites, Polymer Composites, 20, (1999), 357-¬366.

85.

Le Meins, J-F., P. Moldenaers and J. Mewis, Suspensions of monodisperse spheres in
polymer melts: particle size effects in extensional flow, Rheol. Acta, 42, (2002), 184-190.

86.

Prasad, R., Pasanovic-Zujo, V., Gupta, R.K., Cser, F. and Bhattacharya, S.N.,
Morphology of EVA based nanocomposites under shear and extensional flow, Polymer
Engineering and Science, 44, (2004), 1220-1230.

87.

Seong. D., Kang. T., Young. J., Rheological characterization of polymer-based
nanocomposites with different nanoscale dispersions, e-Polymers, (2005).

88.

Okamoto.M, Nam.P, Maiti.P, Kotaka.T, Hasegawa.N, Usuki.A, A house of card
structures in polypropylene/clay nanocomposites under elongational flow, Nano Letters,
1, (2001), 295-298.

89.

Okamoto, M; Nam, PH; Maiti, P; Kotaka, T; Nakayama, T; Takada, M; Ohshima, M;
Usuki, A; Hasegawa, N; Okamoto, H. 2001. Biaxial flow-induced alignment of silicate
layers in polypropylene/clay nanocomposite foam. Nano Letters, 1 (9): 503-505.

90.

Ray, SS; Okamoto, M.., New polylactide/layered silicate nanocomposites- Melt rheology
and foam processing, Macromolecular Materials and Engineering, 288 (12), (2003), 936944.

91.

Wagner, M.H., S. Kheirandish, J. Stange and H. Munstedt, Modeling elongational
viscosity of linear and long-chain branched polypropylenes, Rheol. Acta, 46, (2006), 211221.

92.

Payne.A.R, J. Appl. Polym. Sci., 6, (1962), 57.

107

93.

Kraus.G.J., Mechanical losses in carbon-black-filled rubbers, J. Appl. Polym. Sci.: Appl.
Polym. Symp., 39, (1984), 75.

94.

Heinrich.G, Kluppel.M., Recent advances in the theory of filler networking in elastomers,
Adv. Polym. Sci., 160, (2002), 1-44.

95.

Sternstein.S.S and Zhu.A.J., Reinforcement mechanism of nanofilled polymer melts as
elucidated by non-linear viscoelastic behavior, Macromolecules, 35, (2002), 7262-7232.

96.

Sternstein.S.S and Zhu.A.J., Nonlinear viscoelasticity of nanofilled polymers: interfaces,
chain statistics and properties recovery kinetics, Composites Science and Technology,
63, (2003), 1113-1126.

97.

Cassagnau.Ph, Melis.F, Non-linear viscoelastic behavior and modulus recovery in silica
filled polymers, Polymer, 44, (2003), 6607-6615.

98.

Shariatpanahi.H., Sarabi.F., Mirrali.M., Hemmati.M., and Mahadavi.F., “Polypropylene
Organoclay Nanocomposite: Preparation, Microstructure, Mechanical Properties.”
Journal of Applied Polymer Science, 113, (2009), 922-926.

99.

Gianelli.W., Camino.G., Dintcheva.N.T., Verso.S.L., Mantia. F.P., “EVAMontmorillonite Nanocomposites: Effect Of Processing Conditions.” Macromol. Mater.
Eng., 289, (2004), 238-244.

100.

Kumar.S., “Effects of grafting from edges of nanoclayers on flow induced structure in
Polypropylene/Clay nanocomposites” PhD Dissertation, Michigan State University,
(2005).

101.

Gopakumar, T. and Page, D., "Compounding of nanocomposites by thermokinetic
mixing." J Appl Polym Sci, 96, (5), (2005), 1557-1563.

102.

Lee, J., Lim, Y., and Park, O., "Thermal characteristics of organoclay and their effects
upon the formation of polypropylene/organoclay nanocomposites." Polymer Bullet, 45,
(2), (2000), 191-198.

108

103.

Cullity, B.D., Elements of X-ray Diffraction. 2nd ed., Massachusetts: Addison-Wesley,
(1978).

104.

Theng, B.K.G., Formation and Properties of Clay-Polymer Complexes, Amsterdam:
Elsevier, (1979).

105.

Macosko, C.W. Rheology: Principles, Measurements and Applications, New York:
Wiley, (1994).

106.

Krishnamoorti, R. and Giannelis, E., "Rheology of end-tethered polymer layered silicate
nanocomposites." Macromolecules, 30, (14), 4097-4102 (1997).

107.

Krishnamoorti, R. and Giannelis, E., "Strain hardening in model polymer brushes under
shear." LANGMUIR, 17, (5), 1448-1452 (2001).

108.

Meissner.J., Rheologica Acta, 8, 78, (1969).

109.

Meissner.J., Rheologica Acta, 10, 230, (1971).

110.

Meissner.J., Raible.T., Stephenson.S.E., “Addendum: Rotary Clamp in Uniaxial and
Biaxial Extensional Rheometry of Polymer Melts.” J.Rheol., 25, 1, 673, (1981).

111.

Li.L, Masuda.T., Takahasi.M., “Elongational Flow Behavior of ABS Polymer Melts.”, J.
Rheol., 34, 103, (1990)

112.

Macosko.C.W., Lornston.J.M., SPE Tech. Pap., 19, 461, (1973).

113.

Everage.A.E., ballma.R.L., “Extensional Viscosity of Amorphous Polypstyrene.”, J.
Appl. Polym. Sci., 20, 1137, (1976)

114.

Ide.Y., White.J.L., “The Spinnability of Polymer Fluid Filaments”, J. Appl. Polym. Sci.,
20, 2511, (1976)

115.

Ide.Y., White.J.L., “Experimental Study of Elongational Flow and Failure of Polymer
Melts”, J. Appl. Polym. Sci., 22, 1061, (1978)
109

116.

Munstedt.H., “New Universal Extensional Rheometer for Polymer Melts. Measurements
on a Polystyrene Sample.” J. Rheol., 23, 412, (1979)

117.

Connelly.R.W., Garfield.L.J., Pearson.G.H., “Notes: Local Stretch history of Fixed-EndConstant Length-Polymer-Melt Stretching Experiment.” J.Rheol., 23, 651, (1979)

118.

Ishizuka.O., Koyama.K., “Elongational Viscosity at a constant Elongational strain rate of
Polypropylene Melt.” Polymer, 21, 164, (1980).

119.

Aranguren MI, Mora EM, DeGroot JV, Macosko CW, Effect of reinforcing fillers on the
rheology of polymer melts, Journal of Rheology, 36, (1992), 1165-1182.

120.

Lee.K., Han.C., Linear dynamic viscoelastic properties of functionalized block
copolymer/organoclay nanocomposites, Maromolecules, 36, (2003) , 804-815.

121.

Fulchiron. R, Verney.V, Determination of the elongational behavior of polypropylene
melts from transient shear experiments using Wagners model, Jounral of Non- Newtonian
Fluid Mechanics, 48, (1992), 49-61.

122.

Lertwimolnum.W, Vergnes.B, Influence of compatibilizer and processing conditions on
the dispersion of nanoclay in a polypropylene matrix, Polymer, 46, (2008), 3462-3471.

123.

Olson.G.L, Miller.D.S, Larsen.E.W, Morel.J.E, Chord length distributions in binary
stochastic media in two and three dimensions, Journal Of Quantitative Spectroscopy and
Radiative Transfer, 101, (2006), 269-283.

124.

Olson.G.L, Chord length distributions between hard disks and spheres in regular, semiregular, and quasi-random structures, Annals of Nuclear Energy, 35, (2008), 2150-2155.

125.

Carreau.P.J, Ghanbari.A, Heuzey.M.C, Ton-That.M, Morphological and rheological
properties of PET/clay nanocomposites, Rheol Acta, 52, (2013), 59-74.

126.

Sun.L, Boo.W., etal., Effect of Nanoplatelets on the Rheological Behavior of Epoxy
Monomers, Macromol Mater Eng., 294, (2009), 103-113.
110

127.

Wan.T, Wang.B., Liao.S, Clifford.M, Rheological investigation on the interaction of
polyamide 6 with clay., J Appl Polym Sci., 125, (2012), E27-E33.

111