STRUCTURE-PROPERTY CORRELATIONS OF POLYHEDRAL
OLIGOMERIC SILSESQUIOXANE IN MODEL POLYSTYRENE
By
Madhu Namani

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

ABSTRACT
STRUCTURE - PROPERTY CORRELATIONS OF POLYHEDRAL
OLIGOMERIC SILSESQUIOXANE IN MODEL POLYSTYRENE
By
Madhu Namani
In recent years the blending of 0-, 1-, and 2- dimensional inorganic fillers
such as POSS, carbon nanotubes, and nano-clay platelets into pol ymeric
matrices

enabled

exploration

mechanical

properties.

connections

amongst

of

However,
the

various
a

filler/matrix

fundamental

microstructure

and

combinations

understanding

macroscopic

of

properties

on
the
of

pol ymeric based nanocomposites are yet to be full y explored to optimize the
trul y multifunctional propert y potential of pol ymer nanocomposites. Pol yhedral
oligomeric silsesquioxane (POSS) offers a unique approach to examine the
effect of molecularly dispersed nanoscopic fillers on rheological properties of
entangled pol ymer melts.

Experiments were performed using a nearl y-

monodisperse molecular weight pol yst yrene (PS) blended with varying amounts
of two full y condensed POSS molecules surrounded with phenethyl and st yren yl
groups.

Due to the chemical similarit y between these organic moieties

surrounding the silicon-ox ygen framework (SiO 1 . 5 ) of POSS and PS, we were
able to obtain pol ymer blends with molecular dispersed nanoscopic fillers
needed

to

stud y

the

effect

of

intermolecular

nanoparticle-nanoparticle

interactions and the associated intramolecular interactions on the dynamics of
the pol ymer chains. Differential scanning calorimetry (DSC), wide-angle X-ra y

diffraction (WAXD) and transmission electron microscopy (TEM) were used to
characterize the thermal properties and morphologies of the POSS/PS blends.
Oscillatory Shear both small and large amplitude and/or tensile method was used
to probe the d ynamics of pol ymer chains as influenced by the addition of
different chemical moiet y of POSS at the glass transition, rubbery state and the
terminal-flow transition regions. Further studies were done with varying
molecular weight of PS for understanding of the chain length effects on the
phase behavior of the blends, results obtained from the thermal and mechanical
characterization methods were compared with morphological observations to
better understand the structure-property relationship of pol ymers containing
molecularl y dispersed nanoscopic fillers.

TABLE OF CONTENTS

LIST OF TABLES …………….……………………………………vi
LIST OF FIGURES…………….………………………………..…vii
CHAPTER 1: REVIEW OF POLYMER NANOCOMPOSITES….1
1.1 INTRODUCTION………………………………….…………..1
1.2 SYNTHETIC STRATEGIES TOWARDS HYBRID
MATERIALS………………………………………………...... ...... 12
1.2.1 Sol–Gel Process……………………………………… .. 15
1.2.2 Hybrid Materials by the Sol–Gel Process…………. .. 17
1.2.3 Building Block Approach………………………..….. . 22
1.2.4 Insitu Polymerization Approach…………................26
1.3 TYPES OF HYBRID MATERIALS……………..…………… 30
1.4 PROPERTIES AND APPLICATIONS………………………..34
1.5 SUMMARY…..………………………………………………… 37
RESEARCH OBJECTIVES………………………………….…….38
CHAPTER 2: EFFECTS OF CHEMICAL SUBSTITUENTS ON
PHASE BEHAVIOR OF POSS - PS BLENDS………………….. 40
2.1 Introduction……………………………………………….…… 40
2.2 Experimental……………………………………………….…...44
2.2.1 Materials………………………………....……………. 44
2.2.2 Sample Preparations……………………………….… . 47
2.2.2.1
Samples
Preparations
for
Differential
Scanning Calorimeter (DSC)………… ………………47
2.2.2.2 Samples Preparations for X-Ray Diffraction.47
2.2.2.3 Samples Preparations for Transmission
Electron Microscopy (TEM)……………………...….. 47
2.2.3 Characterization Techniques……………….... .......... 48
2.2.3.1 Transmission Electron Microscopy (TEM)… . 48
2.2.3.2 X-ray Diffraction………………………..…. ...48
2.2.3.3 Determination of Transition Temperatures of
POSS macromers……………………………………….49
2.2.4 Determination of Transition Temperatures of POSS
macromers…………………………………………………… . 49
iv

2.3 Results and Discussion…………………………………….…..49
2.3.1 Effects of POSS Corner Groups……………………………. 49
2.3.2 Miscibility in Polystyrene Based on Different POSS…… . 52
2.4 Summary……….…………………………………………….….64
CHAPTER 3: EFFECTS OF CONCENTRATION ON PHASE
BEHAVIOR OF POSS - PS BLENDS…………………. …………67
3.1 Introduction…………………………………………………..…67
3.2 Experimental…………………………………………………....68
3.2.1 Materials………………………………………..……... 68
3.2.2 Sample Preparation……………………………………. 70
3.2.2.1 Melt Rheological Characterization……………… .. 70
3.3 Results and Discussion……………………………………….. . 71
3.3.1 Rheology during the Rubbery State and Flow Transition. . 71
3.4 Summary………………………………………………………. ..82
CHAPTER 4: EFFECTS OF PS MOLECULAR WEIGHT ON
PHASE BEHAVIOR OF POSS - PS BLENDS…………………. .86
4.1 Introduction………………………………………………….… . 86
4.2 Experimental…………………………………………………... . 90
4.2.1 Materials………………………………………..……. .. 90
4.2.2 Sample Preparation………………………………….. .. 90
4.2.2.1 Rheological Characterization……………………… 90
4.3 Results and Discussion……………………………………….. . 91
4.4 Summary……………………………………………………… . 108
CHAPTER 5: FLOW TRANSITIONS OF POSS - PS BLENDS 110
5.1 Introduction………………………………………………….. . 110
5.1.1 Large Amplitude Oscillatory Shear (LAOS)……………. 111
5.2 Experimental……………………………………………….... . 115
5.2.1 Thermo-Mechanical Analysis……………………….. ....... 115
5.2.2 Rheological Characterization…………………. ............... 116
5.3 Results and Discussion………………………………..……. . 117
5.3.1 Glass Transition Region………… ................................. 117
5.3.2 Non-linear oscillation measurements……………………. 122
5.4 Summary……………………………………………………... . 128
CHAPTER 6: CONCLUSION…………………………………….130
REFERENCES……………………………………………………..139
v

LIST OF TABLES

Table 2.1: Abbreviations, Chemical Formula and Molecular Weight of
POSS Macromers..…………………………………………………………..… 46
Table 2.2: Transition temperatures of POSS macromers (DSC Results) .52
Table 3.1: The crossover frequency versus POSS weight fraction and the
zero-shear viscosit y based on an Ellis model fit……………………………80

vi

LIST OF FIGURES

Figure 2.1: POSS Macromers with Different Corner Groups…………………………...45
Figure 2.2: DSC Curve of Cp8POSS Macromer……………………………………… ...50
Figure 2.3: DSC Curve of Cy8POSS Macromer……………………………………….. .50
Figure 2.4: DSC Curve of Styrenyl8POSS Macromer…………………………………. .51
Figure 2.5: DSC Curve of Phenthyl8POSS Macromer………………………………… .51
Figure 2.6: TEM Image of POSS PS blends at 50 wt% loadings……………………… .54
Figure 2.7 X-Ray plot of Phenethyl POSS (Ph8T8) and PS (MW= 290K Da) / Phenethyl
POSS 20%wt blend…………………………………………………………………… …57
Figure 2.8 X-Ray plot of Styrenyl POSS (St8T8) and PS (MW= 290K Da) /Styrenyl
POSS 20%wt blend………………………………………………………………….. .….58
Figure 2.9: TEM Image of Styrenyl8POSS/PS290K Blend (20 wt% POSS)………… ...61
Figure 2.10: DSC plot of Phenethyl POSS (Ph8T8) and PS (MW= 290K Da) / Phenethyl
POSS 30%wt blend…………………………………………..……………………… .….62
Figure 2.11: DSC plot of Styrenyl POSS (St8T8) and PS (MW= 290K Da) /Styrenyl
POSS 30%wt blend………………………………………………………………..… .….63
Figure 3.1 Time temperature superposed plots of PS (MW= 290K Da) and Phenethyl
POSS (Phe8T8) blends with varying weight fractions of POSS at 1500C a) Storage
Modulus (G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT) versus
Reduced frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)……….… ....73
Figure 3.2 Time temperature superposed plots of PS (MW= 290K Da) and Styrenyl
POSS (St8T8) blends with varying weight fractions of POSS at 1500C a) Storage Modulus
(G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT) versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)………………… ..…..76
Figure 3.3 Log-Log plot of Normalized Plateau Modulus (GN (φ)/GN (1)) versus Volume
fraction (φ))………………………………………………………………………… ..…..79
Figure 4.1 Time temperature superposed plots of PS (MW= 1.6M Da) and Styrenyl
POSS (St8T8) blends with varying weight fractions of POSS at 1500C a) Storage Modulus
vii

(G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT) versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)……………… ..……..96
Figure 4.2 Time temperature superposed plots of PS (MW= 650K Da) and Styrenyl
POSS (St8T8) blends with varying weight fractions of POSS at 1500C a) Storage Modulus
(G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT) versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)……………... ……….99
Figure 4.3 Time temperature superposed plots of PS (MW= 130K Da) and Styrenyl
POSS (St8T8) blends with varying weight fractions of POSS at 1500C a) Storage Modulus
(G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT) versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)…………………… ..102
Figure 4.4 Time temperature superposed plots of PS (MW= 65K Da) and Styrenyl
POSS (St8T8) blends with varying weight fractions of POSS at 1500C a) Storage Modulus
(G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT) versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)…………………… ..105
Figure 4.5: Normalized viscosity (η∗/ aT) versus volume fraction of Styrenyl
POSS………………………………………………………………………………........ 108
Figure 5.1: Stress, Strain and Strain rate waves during an oscillation
measurement. The stress can be decomposed in an elastic and a viscous
stress wave. The Lissajous figure results from plotting the stress versus
the strain……………………………………………… ………………….……112
Figure 5.2 Complex Modulus (E * ) and tanδ versus Temperature plot for
PS (M W = 290K Da) and Phenethyl POSS (Ph 8 T 8 ) blends with varying
weight fractions of POSS………………………….………………………… 118
Figure 5.3 Complex Modulus (E * ) and tanδ versus Temperature plot for
PS (M W = 290K Da) and St yrenyl POSS (St 8 T 8 ) blends with varying weight
fractions of POSS……………………………………………………….…….120
Figure 5.4: Strain sweep for the viscoelastic material 290K PS, showing
the transition from the linear into the non-linear regime……………..… 121
Figure 5.5: Strain sweep for the viscoelastic material 6% Phe 8 T 8 PS 290K
blend, showing the transition from the linear into the non-linear
regim…………………………………………………………………………. ..123
Figure 5.6: Strain sweep for the viscoelastic material 6% St 8 T 8 PS 290K
blend, showing the transition from the linear into the non-linear
regime……………………………………………………………………... …..124

viii

Figure 5.7: Strain sweep for the viscoelastic material 50% Phe 8 T 8 /PS
blend, showing the transition from the linear into the non-linear
regime……………………………………………………………………... …..125
Figure 5.8: Strain sweep for the viscoelastic material 20% Phe 8 T 8 /PS
blend, showing the transition from the linear into the non-linear
regime……………………………………………………………...……..… …126
Figure 5.9: Strain sweep for the viscoelastic material 20% St 8 T 8 /PS
blend, showing the transition from the linear into the non-linear
regime……………………………………………………………………… ….127

ix

CHAPTER 1
REVIEW OF POLYMER NANOCOMPOSITES
1.1 INTRODUCTION
Recent

technological

breakthroughs

and

the

desire

for

new

functions generate an enormous demand for novel materials. Many of the
well-established materials, such as metals, ceramics or plastics cannot
fulfill

all

technological

desires

for

the

various

new

applications.

Scientists and engineers realized earl y on that mixtures of materials can
show superior properties compared with their pure counterparts. One of
the most successful examples is the group of composites which are formed
b y the incorporation of a basic structural material into a second substance,
the matrix. Usuall y the s ystems incorporated are in the form of particles,
whiskers, fibers, lamellae, or a mesh. Most of the resulting materials show
improved mechanical properties and a well-known example is inorganic
fiber-reinforced

polymers.

Nowadays

they

are

regularl y

used

for

lightweight materials with advanced mechanical properties, for example in
the construction of vehicles of all types or sports equipment. The
structural building blocks in these materials which are incorporated into
the matrix are predominantl y inorganic in nature and show a size range
from the lower micrometer to the millimeter range and therefore their
heterogeneous composition is quite often visible to the eye. Soon it
became evident that decreasing the size of the inorganic units to the same
level as the organic building blocks could lead to more homogeneous
1

materials that allow a further fine tuning of materials’ properties on the
molecular and nanoscale level, generating novel materials that either show
characteristics in between the two original phases or even new properties.
Although we do not know the original birth of hybrid materials
exactl y it is clear that the mixing of organic and inorganic components
was carried out in ancient world. At that time the production of bright and
colorful paints was the driving force to consistentl y try novel mixtures of
dyes or inorganic pigments and other inorganic and organic components to
form paints that were used thousands of years ago. Therefore, hybrid
materials or even nanotechnology is not an invention of the last decade
but was developed a long time ago. However, it was onl y at the end of the
20th and the beginning of the 21st century that it was realized b y
scientists, in particular because of the availabilit y of novel physico–
chemical characterization methods, the field of nanoscience opened man y
perspectives for approaches to new materials. The combination of
different anal ytical techniques gives rise to novel insights into hybrid
materials and makes it clear that bottom up strategies from the molecular
level towards materials’ design will lead to novel properties in this class
of materials.
Apart from the use of inorganic materials as fillers for organic
pol ymers, such as rubber, it was a long time before much scientific
activit y was devoted to mixtures of inorganic and organic materials. One
process changed this situation: the sol–gel process 1 - 3 . This process, which

2

will be discussed in more detail later on, was developed in the 1930s
using silicon alkoxides as precursors from which silica was produced. In
fact this process is similar to an organic pol ymerization starting from
molecular precursors resulting in a bulk material. Contrary to many other
procedures used in the production of inorganic materials this is one of the
first processes where ambient conditions were applied to produce
ceramics. The control over the preparation of multicomponent s ystems b y
a mild reaction method also led to industrial interest in that process. In
particular the silicon based sol–gel process was one of the major driving
forces what has become the broad field of inorganic–organic hybrid
materials. The reason for the special role of silicon was its good
processabilit y and the stabilit y of the Si—C bond during the formation of
a silica network which allowed the production of organic-modified
inorganic networks in one step.
Inorganic–organic hybrids can be applied in many branches of
materials chemistry because they are simple to process and are amenable
to design on the molecular scale. Currently there are four major topics in
the

s ynthesis

of

inorganic–organic

materials:

(a)

their

molecular

engineering, (b) their nanometer and micrometer-sized organization, (c)
the transition from functional to multifunctional hybrids, and (d) their
combination with bioactive components. Some similarities to sol–gel
chemistry are shown b y the stable metal sols and colloids, such as gold
colloids, developed hundreds of years ago. In fact sols prepared b y the

3

sol–gel process, i.e. the state of matter before gelation, and the gold
colloids have in common that their building blocks are nanosized particles
surrounded b y a solvent matrix 4 . Such metal colloids have been used for
optical applications in nanocomposites for centuries. Glass, for example,
was alread y colored with such colloids centuries ago. In particular man y
reports of the scientific examination of gold colloids, often prepared b y
reduction of gold salts, are known from the end of the 18th century.
Probabl y the first nanocomposites were produced in the middle of the 19th
century when gold salts were reduced in the presence of gum. Currentl y
many of the colloidal systems already known are being reinvestigated b y
modern instrumental techniques to get new insights into the origin of the
specific chemistry and physics behind these materials.
The term h ybrid material is used for many different s ystems
spanning a wide area of different materials, such as crystalline highl y
ordered coordination pol ymers, amorphous sol–gel compounds, materials
with and without interactions between the inorganic and organic units.
Before the discussion of s ynthesis and properties of such materials we tr y
to delimit this broadly-used term by taking into account various concepts
of composition and structure. The most wide-ranging definition is the
following: a h ybrid material is a material that includes two moieties
blended on the molecular scale. The matrix could be either crystalline or
amorphous but commonl y one of these compounds is inorganic and the
other one organic in nature 5 .

4

A more detailed definition distinguishes between the possible
interactions connecting the inorganic and organic species. Class I h ybrid
materials are those that show weak interactions between the two phases,
such

as

van

der

Waals,

hydrogen

bonding

or

weak

electrostatic

interactions. Class II hybrid materials are those that show strong chemical
interactions between the components. Because of the gradual change in
the strength of chemical interactions it becomes clear that there is a
stead y transition between weak and strong interactions. For example there
are h ydrogen bonds that are definitel y stronger than for example weak
coordinative bonds. In addition to the bonding characteristics structural
properties can also be used to distinguish between various hybrid
materials. An organic moiet y containing a functional group that allows the
attachment to an inorganic network, e.g. a trialkoxysilane group, can act
as a network modifying compound because in the final structure the
inorganic

network

is

onl y

modified

by

the

organic

group.

Phen yltrialkox ysilanes are an example for such compounds; they modify
the silica network in the sol–gel process via the reaction of the
trialkox ysilane group without suppl ying additional functional groups
intended to undergo further chemical reactions to the material formed. If a
reactive functional group is incorporated the s ystem is called a network
functionalizer. The situation is different if two or three of such anchor
groups modify an organic segment; this leads to materials in which the
inorganic group is afterwards an integral part of the h ybrid network 6 - 8 .

5

Blends are formed if no strong chemical interactions exist between
the inorganic and organic building blocks. One example for such a
material is the combination of inorganic clusters or particles with organic
pol ymers lacking a strong (e.g.

covalent) interaction

between the

components. In this case a material is formed that consists for example of
an organic pol ymer with entrapped discrete inorganic moieties in which,
depending on the functionalities of the components, for example weak
crosslinking occurs by the entrapped inorganic units through physical
interactions or the inorganic components are entrapped in a crosslinked
pol ymer matrix. If an inorganic and an organic network interpenetrate
each other without strong chemical interactions, so called interpenetrating
networks (IPNs) are formed, which is for example the case if a sol–gel
material is formed in presence of an organic pol ymer or vice versa. Both
materials described belong to class I hybrids. Class II hybrids are formed
when the discrete inorganic building blocks, e.g. clusters, are covalentl y
bonded to the organic pol ymers or inorganic and organic pol ymers are
covalentl y connected with each other 9 - 1 0 .
After having discussed the above examples one question arises:
what is the difference between inorganic–organic hybrid materials and
inorganic–organic nanocomposites? In fact there is no clear borderline
between these materials. The term nanocomposite is used if one of the
structural units, either the organic or the inorganic, is in a defined size
range of 1–100 nm. Therefore there is a gradual transition between hybrid

6

materials and nanocomposites, because large molecular building blocks
for h ybrid materials, such as large inorganic clusters, can already be of
the nanometer length scale. Commonl y the term nanocomposites is used if
discrete structural units in the respective size regime are used and the
term h ybrid materials is more often used if the inorganic units are formed
in situ b y molecular precursors, for example appl ying sol–gel reactions.
Examples

of

discrete

inorganic

units

for

nanocomposites

are

nanoparticles, nanorods, carbon nanotubes and galleries of clay minerals.
Usuall y a nanocomposite is formed from these building blocks by their
incorporation in organic pol ymers 1 1 .
The most obvious advantage of inorganic–organic hybrids is that
they can favorabl y combine the often dissimilar properties of organic and
inorganic components in one material. Because of the many possible
combinations of components this field is very creative, since it provides
the opportunit y to invent an almost unlimited set of new materials with a
large spectrum of known and as yet unknown properties. Another driving
force in the area of h ybrid materials is the possibilit y to create
multifunctional materials. Probabl y the most intriguing property of hybrid
materials that makes this material class interesting for many applications
is their processing. Contrary to pure solid state inorganic materials that
often require a high temperature treatment for their processing, hybrid
materials show a more pol ymer-like handling, either because of their large
organic content or because of the formation of crosslinked inorganic

7

networks from small molecular precursors just like in polymerization
reactions 1 2 . Hence, these materials can be shaped in any form in bulk and
in films. Although from an economical point of view bulk hybrid
materials can currentl y onl y compete in very special areas with classical
inorganic or organic materials, e.g. in the biomaterials sector, the
possibilit y of their processing as thin films can lead to propert y
improvements of cheaper materials by a simple surface treatment, e.g.
scratch

resistant

coatings.

Based

on

the

molecular

or

nanoscale

dimensions of the building blocks, light scattering in homogeneous hybrid
material can be avoided and therefore the optical transparency of the
resulting h ybrid materials and nanocomposites is, dependent on the
composition used, relativel y high. Furthermore, the materials’ building
blocks can also deliver an internal structure to the material which can be
regularl y ordered. While in most cases phase separation is avoided, phase
separation of organic and inorganic components is used for the formation
of porous materials.
Material properties of hybrid materials are usuall y changed b y
modifications of the composition on the molecular scale. If, for example,
more h ydrophobicit y of a material is desired, the amount of hydrophobic
molecular components is increased. In sol–gel materials this is usuall y
achieved if alk yl- or aryl-substituted trialkoxysilanes are introduced in
the formulation. Hydrophobic and lipophobic materials are composed if
partiall y

or

full y

fluorinated

molecules

8

are

included.

Mechanical

properties, such as toughness or scratch resistance, are tailored if hard
inorganic nanoparticles are included into the pol ymer matrix. Because the
compositional variations are carried out on the molecular scale a gradual
fine tuning of the material properties is possible. One important subject in
materials chemistry is the formation of smart materials, such as materials
that react to environmental changes or switchable s ystems, because they
open routes to novel technologies, for example electroactive materials,
electrochromic materials, sensors and membranes, biohybrid materials,
etc. The desired function can be delivered from the organic or inorganic
or from both components. One of the advantages of hybrid materials in
this context is that functional organic molecules as well as biomolecules
often show better stabilit y and performance if introduced in an inorganic
matrix 1 3 - 1 7 .
The

transition

from

the

macroscopic

world

to

microscopic,

nanoscopic and molecular objects leads, beside the change of physical
properties of the material itself, i.e. the so called quantum size effects, to
the change of the surface area of the objects. While in macroscopic
materials the majority of the atoms are hidden in the bulk of the material
it becomes vice versa in very small objects. In small nanoparticles
(<10nm) nearl y every atom is a surface atom that can interact with the
environment.

One

predominant

feature

of

hybrid

materials

or

nanocomposites is their inner interface, which has a direct impact on the
properties of the different building blocks and therefore on the materials’

9

properties 1 8 . If the two phases have opposite properties, such as different
polarit y, the s ystem would thermodynamicall y phase separate. The same
can happen on the molecular or nanometer level, leading to microphase
separation. Usuall y, such a s ystem would thermodynamically equilibrate
over time. However in many cases in hybrid materials the s ystem is
kineticall y stabilized b y network forming reactions such as the sol–gel
process leading to a spatial fixation of the structure. The materials formed
can be macroscopicall y homogeneous and opticall y clear, because the
phase

segregation

is

of

small

length

scale

and

therefore

limited

interaction with visible light occurs. However, the composition on the
molecular or nanometer length scale can be heterogeneous. If the phase
segregation reaches the several hundred nanometer length scale or the
refractive index of the formed domains is very different, materials often
turn opaque. Effects like this are avoided if the reaction parameters are
controlled in such a way that the speed of network formation is kept faster
than the phase separation reactions.
The high surface area of nanobuilding blocks can lead to additional
effects; for example if surface atoms strongl y interact with molecules of
the matrix b y chemical bonding, reactions like surface reorganization,
electron transfer, etc. can occur which can have a large influence on the
ph ysical properties of the nano-building blocks and thus the overall
performance of the material formed 1 9 - 2 1 . It is, for example, known that
conjugated π-electron s ystems coordinated to the surface of titania

10

nanoparticles can lead to charge transfer reactions that influence the
color, and therefore the surface electronic properties of the particles.
Nanosized objects, such as inorganic nanoparticles, in addition show a
very high surface energy. Usuall y if the surface energy is not reduced b y
surface active agents (e.g. surfactants), such particles tend to agglomerate
in an organic medium. Thus, the physical properties of the nanoparticles
(e.g. quantum size effects) diminish, and/or the resulting materials are no
longer homogeneous. Both facts have effects on the final material
properties. For example the desired optical properties of nanocomposites
fade away, or mechanical properties are weakened. However, sometimes a
controlled

aggregation

conducting

particles

can
in

a

also

be

pol ymer

required,
matrix

e.g.

percolation

increases

the

of

overall

conductivit y of the material 2 2 .
Earlier

the

interaction

mechanism

between

the

organic

and

inorganic species was used to categorize the different t ypes of hybrid
materials, furthermore of course the interaction also has an impact on the
material properties. Weak chemical interactions between the inorganic and
organic entities leave some potential for dynamic phenomena in the final
materials, meaning that over longer periods of time changes in the
material, such as aggregation, phase separation or leaching of one of the
components, can occur. These phenomena can be avoided if strong
interactions are employed such as covalent bonds, as in nanoparticle
crosslinked pol ymers. Depending on the desired materials’ properties the

11

interactions can be graduall y tuned. Weak interactions are, for example,
preferred where a mobilit y of one component in the other is required for
the target properties. This is for example the case for ion conducting
pol ymers, where the inorganic ion (often Li+) has to migrate through the
pol ymer matrix. In man y examples the interactions between the inorganic
and organic species are maximized by appl ying covalent attachment of
one to the other species. But there are also cases where small changes in
the composition, which on the first sight should not result in large effects,
can make considerable differences. It was, for example, shown that
interpenetrating networks between polyst yrene and sol–gel materials
modified with phen yl groups show less microphase segregation than sol–
gel materials with pure alkyl groups, which was interpreted to be an effect
of π-π-interactions between the two materials 2 3 . In addition the interaction
of the two components can have an influence on other properties, such as
electronic properties if coordination complexes are formed or electron
transfer processes are enabled by the interaction.
1.2 SYNTHETIC STRATEGIES TOWARDS HYBRID MATERIALS
In principle two different approaches can be used for the formation
of hybrid materials: Either well-defined preformed building blocks are
applied that react with each other to form the final hybrid material in
which the precursors still at least partially keep their original integrit y or
one or both structural units are formed from the precursors that are
transformed into a novel (network) structure. Both methodologies hav e

12

their advantages and disadvantages and will be described here in more
detail.
Building block approach As mentioned above building blocks at least
partiall y keep their molecular integrit y throughout the material formation,
which means that structural units that are present in these sources for
materials formation can also be found in the final material. At the sam e
time t ypical properties of these building blocks usuall y survive the matrix
formation, which is not the case if material precursors are transferred into
novel materials. Representative examples of such well-defined building
blocks are modified inorganic clusters or nanoparticles with attached
reactive organic groups. Cluster compounds often consist of at least one
functional group that allows an interaction with an organic matrix, for
example b y copol ymerization. Depending on the number of groups that
can interact, these building blocks are able to modify an organic matrix
(one functional group) or form partially or full y crosslinked materials
(more than one group). For instance, two reactive groups can lead to the
formation of chain structures. If the building blocks contain at least three
reactive groups they can be used without additional molecules for the
formation of a crosslinked material. Beside the molecular building blocks
mentioned, nanosized building blocks, such as particles or nanorods could
be used to form nanocomposites. The building block approach has one
large advantage compared with the in situ formation of the inorganic or
organic entities: because at least one structural unit (the building block)

13

is well-defined and usuall y does not undergo significant structural
changes during the matrix formation, better structure–property predictions
are possible. Furthermore, the building blocks can be designed in such a
way to give the best performance in the materials’ formation, for example
good solubilit y of inorganic compounds in organic monomers by surface
groups showing a similar polarit y as the monomers.
In situ formation of the components Contrary to the building block
approach the in situ formation of the hybrid materials is based on the
chemical transformation of the precursors used throughout materials’
preparation. Typically this is the case if organic pol ymers are formed but
also if the sol–gel process is applied to produce the inorganic component.
In these cases well-defined discrete molecules are transformed to
multidimensional structures, which often show totall y different properties
from the original precursors. Generall y simple, commercially available
molecules are applied and the internal structure of the final material is
determined b y the composition of these precursors but also by the reaction
conditions. Therefore control over the latter is a crucial step in this
process. Changing one parameter can often lead to two very different
materials. If, for example, the inorganic species is a silica derivative
formed b y the sol–gel process, the change from base to acid catal ysis
makes a large difference because base catal ysis leads to a more particlelike

microstructure

while

acid

catalysis

14

leads

to

a

pol ymer-like

microstructure. Hence, the final performance of the derived materials is
strongl y dependent on their processing and its optimization.
Man y of the classical inorganic solid state materials are formed
using solid precursors and high temperature processes, which are often
not compatible with the presence of organic groups because they are
decomposed at elevated temperatures. Hence, these high temperature
processes are not suitable for the in situ formation of hybrid materials.
Reactions that are emplo yed should have more the character of classical
covalent bond formation in solutions. One of the most prominent
processes which fulfill these demands is the sol–gel process. However,
such

rather

low

temperature

processes

often

do

not

lead

to

the

thermod ynamicall y most stable structure but to kinetic products, which
has some implications for the structures obtained. For example low
temperature

derived

inorganic

materials

are

often

amorphous

or

crystallinit y is onl y observed on a very small length scale, i.e. the
nanometer range. An example of the latter is the formation of metal
nanoparticles in organic or inorganic matrices by reduction of metal salts
or organometallic precursors.

1.2.1 Sol–Gel Process
This process is chemicall y related to an organic pol ycondensation reaction
in which small molecules form pol ym eric structures by the loss of
substituents. Usually the reaction results in a three-dimensional (3-D)

15

crosslinked network. The fact that small molecules are used as precursors
for the formation of the crosslinked materials implies several advantages,
for example a high control of the purity and composition of the final
materials and the use of a solvent based chemistry which offers many
advantages for the processing of the materials formed. The silicon-based
sol–gel process is probabl y the one that has been most investigated;
therefore the fundamental reaction principles are discussed using this
process as a model s ystem. One important fact also makes the siliconbased sol–gel processes a predominant process in the formation of hybrid
materials, which is the simple incorporation of organic groups using
organicall y modified silanes. Si—C bonds have enhanced stabilit y against
h ydrol ysis in the aqueous media usuall y used, which is not the case for
man y metal–carbon bonds, so it is possible to easil y incorporate a large
variet y of organic groups in the network formed. Principally R4–nSiXn
compounds (n = 1–4, X = OR’, halogen) are used as molecular precursors,
in which the Si—X bond is labile towards hydrol ysis reactions forming
unstable silanols (Si—OH) that condensate leading to Si—O—Si bonds. In
the first steps of this reaction oligo- and pol ymers as well as cyclics are
formed subsequently resulting in colloids that define the sol. Solid
particles in the sol afterwards undergo crosslinking reactions and form the
gel. The reaction could be acid and base catal yzed leading differing
architecture based on the pH. More detailed discussions of the sol–gel
process can be found in the cited literature 1 - 3 .

16

1.2.2 Hybrid Materials by the Sol–Gel Process
Compared with other inorganic network forming reactions, the sol–
gel processes show mild reaction conditions and a broad solvent
compatibilit y. These two characteristics offer the possibilit y to carry out
the inorganic network forming process in presence of a preformed organic
pol ymer or to carry out the organic pol ymerization before, during or after
the sol–gel process. The properties of the final hybrid materials are not
onl y

determined

by

the

properties

of

the

inorganic

and

organic

component, but also by the phase morphology and the interfacial region
between the two components. The often dissimilar reaction mechanisms of
the sol–gel process and t ypical organic pol ymerizations, such as free
radical pol ymerizations, allow the temporal separation of the two
pol ymerization reactions which offers many advantages in the material
formation. One major parameter in the synthesis of these materials is the
identification of a solvent in which the organic macromolecules are
soluble and which is compatible with either the monomers or preformed
inorganic oligomers derived by the sol–gel approach. Many commonl y
applied organic pol ymers, such as pol ystyrene or pol ymethacrylates, are
immiscible with alcohols that are released during the sol–gel process and
which are also used as solvents, therefore phase separation is enforced in
these cases. This can be avoided if the solvent is switched from the
typicall y used alcohols to, for example, THF in which many organic
pol ymers are soluble and which is compatible with many sol–gel

17

reactions. Phase separation can also be avoided if the pol ymers contain
functional groups that are more compatible with the reaction conditions of
the sol–gel process or even undergo an interaction with the inorganic
material formed. In addition, the functional group incorporated changes
the properties of the final material, for example fluoro-substituted
compounds can create h ydrophobic and lipophobic materials, additional
reactive functional groups can be introduced to allow further reactions
such as amino, epox y or vinyl groups. Beside molecules with a single
trialkox ysilane group also multifunctional organic molecules can be used.
b y the incorporation of OH-groups that interact with, for example,
h ydrox yl

groups

formed

during

the

sol–gel

process

or

by

ionic

modifications of the organic pol ymer. Covalent linkages can be formed if
functional groups that undergo hydrol ysis and condensation reactions are
covalentl y attached to the organic monomers.
If the organic pol ym erization occurs in the presence of an inorganic
material to form the h ybrid material one has to distinguish between
several possibilities to overcome the incompatibilt y of the two species.
The inorganic material can either have no surface functionalization but
the bare material surface; it can be modified with nonreactive organic
groups (e.g. alk yl chains); or it can contain reactive surface groups such
as pol ymerizable functionalities. Depending on these prerequisites the
material can be pretreated, for example a pure inorganic surface can be
treated with surfactants or silane coupling agents to make it compatible

18

with the organic monomers, or functional monomers can be added that
react with the surface of the inorganic material. If the inorganic
component has nonreactive organic groups attached to its surface and it
can be dissolved in a monomer which is subsequentl y pol ymerized, the
resulting material after the organic pol ymerization is a blend. In this case
the inorganic component interacts onl y weakl y or not at all with the
organic pol ymer; hence, a Class I material formed. Homogeneous
materials are onl y obtained in this case if agglomeration of the inorganic
components in the organic environment is prevented. This can be achieved
if the interactions between the inorganic components and the monomers
are better or at least the same as between the inorganic components.
However, if no strong chemical interactions are formed, the long-term
stabilit y of a once homogeneous material is questionable because of
diffusion effects in the resulting hybrid material. Examples of such
materials are alk yl chain functionalized silica nanoparticles that can be
introduced into many h ydrophobic pol ymers, the use of block copol ymers
containing a pol y(vin yl pyridine) segment that can attach to many metal
nanoparticles,

or

the

use

of

hydroxyethyl

methacrylates

in

the

pol ymerization mixture together with metal oxide nanoparticles. In the
latter example h ydrogen bridges are formed between the polymer matrix
and the particle surface. The stronger the respective interaction between
the components, the more stable is the final material. The strongest
interaction is achieved if class II materials are formed, for example with

19

covalent interactions. Examples for such strong interactions are the use of
surface-attached

polymerizable

groups

that

are

copol ymerized

with

organic monomers.
If a porous 3-D inorganic network is used as the inorganic
component for the formation of the hybrid material a different approach
has

to

be

employed

functionalization

of

depending

the

pores

on

and

the

the

pore

stiffness

size,
of

the
the

surface
inorganic

framework. In man y cases intercalation of organic components into the
cavities is difficult because of diffusion limits. Several porous or layered
inorganic materials have already been used to prepare hybrid materials
and nanocomposites 2 4 . Probabl y the most studied materials, class in this
respect is that of two-dimensional (2-D) layered inorganic materials that
can intercalate organic molecules and if pol ymerization between the
layers occurs even exfoliate, producing nanocomposites. Contrary to
intercalated s ystems the exfoliated hybrids onl y contain a small weight
percentage of host layers with no structural order. Principall y three
methods for the formation of pol ymer–clay nanocomposites can be used:
1. Intercalation of monomers followed by in situ pol ymerization
2. Direct intercalation of pol ymer chains from solution
3. Pol ymer melt intercalation
The method applied depends on the inorganic component and on the
pol ymerization

technique

used

and

introductory chapter.

20

will

not

be

discussed

in

this

Contrary to the layered materials, which are able to completel y
delaminate if the forces produced by the intercalated pol ymers overcome
the attracting energy of the single layers, this is not possible in the case
of the stable 3-D framework structures, such as zeolites, molecular sieves
and M41S-materials. The composites obtained can be viewed as host–
guest h ybrid materials. There are two possible routes towards this kind of
h ybrid material; (a) direct threading of preformed pol ymer through the
host channels (soluble and melting pol ymers) which is usually limited b y
the size, conformation, and diffusion behavior of the pol ym ers and, (b)
the in situ pol ymerization in the pores and channels of the hosts. The
latter is the most widel y used method for the s ynthesis of such s ystems.
Of course, diffusion of the monomers in the pores is a function of the pore
size, therefore the pores in zeolites with pore sizes of several hundred
picometers are much more difficult to use in such reactions than
mesoporous materials with pore diameters of several nanometers. Two
methods proved to be very valuable for the filling of the porous structures
with monomers: one is the soaking of the materials in liquid monomers
and the other one is the filling of the pores in the gas phase. A better
uptake of the monomers by the inorganic porous materials is achieved if
the pores are pre-functionalized with organic groups increasing the
absorption of monomers on the concave surface. In principle this
technique is similar to the increase of monomer absorption on the surface
of silica nanoparticles by the surface functionalization with silane

21

coupling agents. Beside of well-defined 3-D porous structures, sol–gel
networks are also inherentl y porous materials. Uniform homogeneous
materials can be obtained if the solvent of the sol–gel process is a
monomer for a pol ymerization. This can be pol ymerized in a second step
after the sol–gel process has occurred.
1.2.3 Building Block Approach
In recent years many building blocks have been s ynthesized and
used for the preparation of hybrid materials. Chemists can design these
compounds on a molecular scale with highl y sophisticated methods and
the resulting s ystems are used for the formation of functional hybrid
materials. Man y future applications, in particular in nanotechnology,
focus

on

a

bottom-up

approach

in

which

complex

structures

are

hierarchicall y formed b y these small building blocks. This idea is also one
of the driving forces of the building block approach in hybrid materials.
Another point which was also already mentioned is the predictabilit y of
the final material properties if well-defined building blocks are used. A
typical building block should consist of a well-defined molecular or
nanosized structure and of a well-defined size and shape, with a tailored
surface structure and composition. In regard of the preparation of
functional h ybrid materials the building block should also deliver
interesting chemical or physical properties, in areas like conductivit y,
magnetic behavior, thermal properties, switching possibilities, etc. All
these characteristics should be kept during the material formation, for

22

example the embedment into a different phase. Building blocks can be
inorganic or organic in nature, but because they are incorporated into
another phase they should be somehow compatible with the second phase.
Most of the times the compatibilit y is achieved by surface groups that
allow some kind of interaction with a second component.
Prime examples of inorganic building blocks that can keep their
molecular integrit y are cluster compounds of various compositions.
Usuall y clusters are defined as agglomerates of elements that either
exclusivel y contain pure metals or metals in mixture with other elements.
Although the classical chemical understanding of a cluster includes the
existence of metal–metal bonds, the term cluster should be used in the
context of this book in its meaning of an agglomerate of atoms in a given
shape. Regularl y pure metal clusters are not stable without surface
functionalization with groups that decrease surface energy and thus avoid
coalescence to larger particles. Both coalescence and surface reactivit y of
clusters are closely related to that of nanoparticles

of the same

composition. Because of this similarit y and the fact that the transition
from large clusters to small nanoparticles is fluent, we will not clearl y
distinguish between them. While in commonl y applied metal clusters the
main role of the coordinating ligands is the stabilization, they also can
serve for a better compatibilization or interaction with an organic matrix.
Similar mechanisms are valid for binary s ystems like metal chalcogenide
or multicomponent clusters. Hence, the goal in the chemical design of

23

these s ystems is the preparation of clusters carrying organic surface
functionalizations that tailor the interface to an organic matrix by making
the inorganic core compatible and by the addition of functional groups
available for certain interactions with the matrix.
Surface-functionalized metal clusters are one prominent model
s ystem for well defined inorganic building blocks that can be used in the
s ynthesis of hybrid materials. However, as with many other nanoscaled
materials it is not possible to s ynthesize such pure clusters and to handle
them without a specific surface coverage that limits the reactivit y of the
surface atoms towards agglomeration. From the aspect of the s ynthesis of
hybrid materials this is no problem as long as the surface coverage of the
cluster or

nanoparticle

contains

the

desired

functionalities

for

an

interaction with an organic matrix. Beside pure metal clusters and
nanoparticles an interesting class of materials is metal oxides, because
they have interesting magnetic and electronic properties often paired with
low toxicit y. Simple eas y-to-s ynthesize oxidic compounds are siliconbased s ystems such as silica particles or spherosilicate clusters, therefore
these s ystems are often used as model compounds for the class of metal
oxides, although they do not reall y represent the class of transition metal
oxides that are probabl y more often used in technological relevant areas.
Silica particles or spherosilicate clusters both have in common that the
surface contains reactive oxygen groups that can be used for further
functionalization.

Mono-functional

polyhedral

24

silsesquioxane

(POSS)

derivatives of the t ype R ′R7Si8O12 (R′ = functional group, R =
nonfunctional group) are prepared by reacting the incompletely condensed
molecule R7Si7O9(OH)3 with R′SiCl3. The eighth corner of the cubic
closed structure is inserted by this reaction, and a variet y of functional
organic groups R ′ can be introduced, such as vinyl, all yl, st yryl,
norbornadien yl, 3-prop yl methacrylate, etc. The incompletely condensed
compounds R7Si7O9(OH)3 are obtained when certain bulky groups R (e.g.
cyclopent yl, cyclohex yl, tert.-but yl) prevent the formation of the closed
structures from RSiX3 precursors and lead to the precipitation of openframework POSS. These bulky substituents not onl y lead to open
framework structures but also increase the solubilit y of the inorganic
units in organic solvents. The closed cubic s ystems still show the high
solubilit y which is necessary if the inorganic building blocks are to be
incorporated in an organic environment for the functionalization of
organic materials. The simple handling of these s ystems caused their
boom

in

the

preparation

of

hybrid

materials.

Other

popular

silsesquioxanes that have been prepared are the octahydrido- or the
octavin yl-substituted molecules, which offer eight reactive sites. The
preparation of these s ystems is still very costl y not least because of the
low yields of the targeted products. After functionalization usuall y eight
reaction sites are attached to these silica cores. Some t ypical reactions
lead to the attachment of initiating or polymerizable groups at the corners

25

and therefore the resulting clusters can be used as multifunctional
initiators for pol ymerizations or as crosslinking monomers.
Post s ynthetic modification means that the cluster or nanoparticle is
formed in a first step appl ying well-established procedures and the
functionalization with organic groups is applied in a second step. Reactive
surface functionalizations are required that allow a chemical reaction with
the surface decorating molecules, for example nucleophilic substitution
reactions.
1.2.4 Insitu Polymerization Approach
In the post-s ynthesis modification, functionalized building blocks
are formed in two steps which are distinctl y separate from each other: the
inorganic core is formed first, and the functional organic groups are
introduced later in a different reaction. An alternative method is the
formation of the inorganic building blocks in the presence of functional
organic molecules (i.e. the functionalization of the clusters/particles
occurs in situ). Silsesquioxanes with only one substitution pattern at each
silicon

atom

are

typical

examples

for

the

in

situ

formation

of

functionalized building blocks. As mentioned above they are prepared b y
the h ydrol ysis and condensation of trialkoxy- or trichlorosilanes, thus
they contain inherentl y one functional group that is also present in the
final material. Depending on the reaction procedure either ladderlike
pol ymers

or

pol yhedral

silsesquioxanes

(POSS)

are

obtained.

The

pol yhedral compounds [RSiO3/2]n can be considered silicon oxide

26

clusters capped b y the organic groups R. Pol yhedral Silsesquioxanes
[RSiO3/2]n are obtained by controlled hydrol ytic condensation of RSiX3
(X = Cl, OR ′) in an organic solvent. There is a high driving force for the
formation of pol yhedral rather than pol ymeric compounds, particularl y if
the precursor concentration in the employed solvent is low (when the
concentration is increased, increasing portions of network polymers may
be formed). Which oligomers are produced and at which rate, depends on
the reaction conditions, such as solvent, concentration of the monomer,
temperature, pH and the nature of the organic group R. Chlorosilanes have
a higher reactivit y than the corresponding alkox ysilanes. In most cases,
intractable mixtures of products are obtained except for those species that
precipitate from the solution. The best investigated silsesquioxane cages
are the cubic octamers, R8Si8O12. The compound [HSiO3/2]8 is, for
example, prepared by h ydrol ysis of HSiCl3 in a benzene / H2SO4 mixture
or by using partiall y hydrated FeCl3 as the water source. It is a valuable
starting compound for organicall y substituted derivatives as the Si—H
functions of the silsesquioxane can be converted into organofunctional
groups either b y h ydrosilation reactions (in the equation: X = Cl, OR′,
CN, etc.). Onl y a few POSS with functional organic groups are insoluble
enough to be obtained directl y from the corresponding RSiCl3 precursor
by

precipitation

reactions.

Examples

are

(CH2==CH)8Si8O12,

(p-

ClCH2C6H4)8Si8O12, (R2NCH2CH2)8Si8O12 or (ClCH2CH2)8Si8O12.
These

compounds

can

also

be

transformed

27

to

other

functional

octa(silsesquioxanes). Examples include epoxidation of vinyl groups or
nucleophilic substitution of the chloro group. Nonfunctional organic
groups may be converted to functional organic by standard organic
reactions. For example, the phen yl groups of the cubic silsesquiox ane
Ph8Si8O12

were

first

nitrated

and

then

reduced

to

give

(H2NC6H4)8Si8O12. The s ystems prepared in this way can be used as
building blocks for materials depending on their functional groups. The
pol yhedral compounds [RSiO3/2]n (POSS) or [RO—SiO3/2]n discussed so
far, formed b y h ydrol ysis and condensation of a single precursor, are
models for larger silica particles covered by organic groups and prepared
from RSi(OR′)3 / Si(OR′)4 mixtures. The main parameter that controls the
mutual arrangement of the [SiO4] and [RSiO3/2] building blocks is the
pH. It was shown that upon sol–gel processing of RSi(OR′)3 / Si(OR ′)4
mixtures (with nonbasic groups R) under basic conditions Si(OR′)4 reacts
first and forms a gel network of agglomerated spherical nanoparticles.
The RSi(OR′)3 precursor reacts in a later stage and condenses to th e
surface of the pre-formed silica nanoparticulate network. This kineticall y
controlled arrangement of the two building blocks from RSi(OR ′)3 /
Si(OR ′)4 mixtures is another method to obtain surface-modified spherical
Stöber particles.
Interactions between metals or metal oxides cores to molecules that
act as surface functionalizations are similar on a molecular scale and
therefore do not usuall y change with the size of the core. Thus, the

28

chemistry developed for isolated and structurall y characterized metal and
metal oxide clusters can also be applied for the functionalization of larger
nanoparticles. This interface analogy offers the chance to study the
chemistry on the molecular scale, which can be anal yzed by conventional
spectroscopic techniques much more easil y, and transfer the obtained
conclusions to the larger scale. There are many examples which show
these similarities. Generall y, as already mentioned above, the organic
groups present in the reaction mixture and attached to the surface after th e
particle formation were mainl y used to limit the growth of the derived
particles b y blocking reactive surface sites and guarantee a stable
suspension in a specific solvent. Onl y recentl y have these groups been
used to introduce a different surface characteristic to the surface of these
building blocks or to add chemical functionalities to the surface of the
particles and on their preparation route. The head groups of these capping
agents can also be other functional molecules or pol ymers to graft them
on the surface of the inorganic nanobuilding blocks.
Beside

the

surface-functionalized

inorganic

building

blocks

described, of course organic building blocks can also be used for the
formation of h ybrid materials. Typical examples are oligo- and pol ymers
as well as biological active molecules like enz ymes. In principal, similar
methods have to be applied as in the case of inorganic s ystems to increase
the compatibilit y and the bonding between the two organic phases.

29

1.3 TYPES OF HYBRID MATERIALS:
Small molecules The modification of inorganic networks with small
organic molecules can be defined as the origin of hybrid materials. This is
particularl y true for sol–gel derived silicon-based materials. The mild
conditions of the sol–gel process permit the introduction into the
inorganic network of an y organic molecule that consists of groups which
do not interfere with these conditions, e.g. an aqueous and an acid or
alkaline environment; it can then either be physicall y trapped in the
cavities or covalently connected to the inorganic backbone. The latter is
achieved b y the modification of the organic molecules with h ydrol ysable
alkox ysilane or chlorosilane groups. Phase separation is usuall y avoided
b y matching the polarit y of the often hydrophobic molecules to that of the
h ydrophilic environment. If such a match can be obtained nearl y ever y
organic molecule can be applied to create hybrid materials. In recent
years functional h ybrids have been the particular focus of interest.
Organic d yes, nonlinear optical groups, or switchable groups are onl y a
small selection of molecules which have already been used to prepare
hybrid materials and nanocomposites.
Macromolecules

Oligo-

and

pol ymers

as

well

as

other

organic

macromolecules often show different solubilities in specific solvents
compared with their monomers; most often the solubilit y of the pol ymers
is much lower than that of the monomers. However, many formation
mechanisms for h ybrid materials and nanocomposites are based on solvent

30

chemistry, for example the sol–gel process or the wet chemistry formation
of nanoparticles. 2 5 - 2 6 Therefore, if homogeneous materials are targeted, an
appropriate solvent for both the inorganic and the organic macromolecules
is of great benefit. For example many macromolecules are soluble in THF
which is also a reasonable good reaction environment for the sol–gel
process. Additional compatibilization is obtained if the pol ymers contain
groups

that

can

interact

with

the

inorganic

components.

Similar

mechanisms of interactions can be employed as already mentioned above,
i.e. groups that interact via the formation of covalent bonds or others that
compatibilize

between

the

organic

and

inorganic

components.

A

particularl y interesting group of macromolecules are block copol ymers,
consisting of a h yd rophilic and a hydrophobic segment. They can be
tailored in such a way that they can react with two phases that reveal
totall y different chemical characters and therefore, they are known as
good compatibilizers between two components. These surfactants were for
example often used for the modification of nanoparticles, where one
segment interacts with the surface of the particle, and the other segment
sticks away from this surface. In technology such s ystems are often used
to overcome interfaces, for example when inorganic fillers are used for
the modification of organic pol ymers. Novel controlled polymerization
methods of late have allowed the preparation of block copol ymers with a
plethora of functional groups and therefore novel applications will soon
be available.

31

Particles and particle-like structures Organic colloids formed from
ph ysicall y or chemicall y crosslinked pol ymers can also be used as
building

blocks

for

inorganic–

organic

hybrid

materials

and

nanocomposites. The good control over their properties, such as their size,
the broad size range in which they can be produced, from several
nanometers to micrometers, accompanied by their narrow size distribution
makes them ideal building blocks for many applications. Similarl y to
dendrimers, special interest in these s ystems is achieved after their
surface modification, because of their self-assembling or simpl y by their
heterophase dispersion. As already mentioned above these latex colloids
are

formed

in

aqueous

dispersions

which,

in

addition

to

being

environmentall y more acceptable or even a mandatory choice for an y
future

development

of

large

output

applications,

can

provide

the

thermod ynamic drive for self-assembling of amphiphilics, adsorption onto
colloidal particles or partitioning of the hybrid’s precursors between
dispersed nanosized reaction loci, as in emulsion or mini-emulsion freeradical pol ymerizations. For the use as precursors in inorganic–organic
h ybrid materials styrene or acrylate homo- and copol ymer core latex
particles are usually modified with a reactive comonomer, such as
trimethox ysil ylprop yl

methacrylate,

to

achieve

efficient

interfacial

coupling with silica environment during the sol–gel process. 3 2 - 3 4 Organic
colloidal building blocks were in particular used for the preparation of 3D colloidal crystals that were subsequently applied as templates in whose

32

voids inorganic precursors were infiltrated and reacted to inorganic
materials followed by removal of the colloids. Furthermore, discrete coreshell particles can also be produced consisting of an organic core and an
inorganic shell. After removal of the organic core, for example b y
calcination, hollow inorganic spheres are obtained. Another t ype of
organic macromolecular building block is the hyperbranched molecules,
so-called dendrimers. Dendrimers are highl y branched
monodisperse

macromolecules

with

a

tree-like

regular 3-D

structure.

These

macromolecules offer a wide range of unusual physical and chemical
properties mainl y because they have a well-defined number of peripheral
functional groups that are introduced during their s ynthesis as well as
internal cavities (guest–host s ystems). In particular the deliberate control
of their size and functionalit y makes these compounds also interesting
candidates as nanoscopic building blocks for hybrid materials. Sphericall y
shaped dendrimers are, for example, ideal templates for porous structures
with porosities that are determined by the radius of the dendrimeric
building block. Generall y approaches for surface functionalizations of
these molecules are the modification with charged end-groups or the use
of reactive organic groups. The end groups of dendrimers can also be used
for an interaction with metal clusters or particles and thus nanocomposites
are

formed

often

by

simpl y

mixing

the

two

components.

Multifunctionalized inorganic molecules can also act as the core of
dendrimers. POSS and spherosilicate cages were, for example, used as the

33

core units from which dendrons were either grown divergentl y or to which
they were appended convergentl y. Appl ying both silsesquioxanes and endgroup

functionalized

dendrimers,

both

as

multifuntional

molecules

crosslinked h ybrid materials were obtained where welldefined inorganic
molecules acted as crosslinking components. 2 7 - 3 1
Beside their role as crosslinking building blocks dendrimers can
form h ybrid materials by themselves. For example in their outer or inner
shell precursors for nanoparticle s ynthesis can be attached to functional
groups introduced during the s ynthesis and afterwards nanoparticles are
grown within the branches of the dendrimers.
1.4 PROPERTIES AND APPLICATIONS
There is almost no limit to the combinations of inorganic and
organic components in the formation of hybrid materials. Therefore
materials with novel composition– property relationships can be generated
that have not yet been possible. Because of the plethora of possible
combinations this introductory chapter can onl y present some selected
examples. Man y of the properties and applications are dependent on the
properties of the precursors and the reader is therefore referred to the
following chapters. Based on the increased importance of optical data
transmission and storage, optical properties of materials play a major role
in man y high-tech applications. The materials used can reveal passive
optical properties, which do not change by environmental excitation, or
active optical properties such as photochromic (change of color during

34

light exposure) or electrochromic (change of color if electrical current is
applied) materials. Both properties can be incorporated by building blocks
with the specific properties, in many cases organic compounds, which are
incorporated in a matrix. Hybrid materials based on silicates prepared b y
the sol–gel process and such building blocks reveal many advantages
compared with other types of materials because silica is transparent and if
the building blocks are small enough, does not scatter light, and on the
other hand organic materials are often more stable in an inorganic matrix.
One of the most prominent passive features of hybrid materials alread y
used in industry are decorative coatings obtained by the embedment of
organic d yes in h ybrid coatings. Another advantage of hybrid materials is
the increased mechanical strength based on the inorganic structures.
Scratch-resistant coatings for plastic glasses are based on this principle.
One of the major advantages of hybrid materials is that it is possible to
include more than one function into a material by simpl y incorporating a
second component with another propert y into the material formulation. In
the case of scratch resistant coatings, for example, additional hydrophobic
or antifogging properties can be introduced. However, in many cases the
precursors for h ybrid materials and nanocomposites are quite expensive
and therefore the preparation of bulk materials is economicall y not
feasible. One of the advantages of hybrid materials, namel y, their quite
simple processing into coatings and thin films, can be one solution to this
disadvantage.

Applying

such

coatings

35

to

cheaper supports

can

be

advantageous. Silica is preferred as the inorganic component in such
applications because of its low optical loss. Other inorganic components,
for example zirconia, can incorporate high refractive index properties, or
titania in its rutile phase can be applied for UV absorbers. Functional
organic molecules can add third order nonlinear optical (NLO) properties
and conjugated polymers, conductive pol ymers can add interesting
electrical properties. Nanocomposite based devices for electronic and
optoelectronic applications include light-emitting diodes, photodiodes,
solar cells, gas sensors and field effect transistors. 3 5 While most of these
devices can also be produced as full y organic pol ymer-based systems, the
composites with inorganic moieties have important advantages such as the
improvement of long term stabilit y, the improvement of electronic
properties b y doping with functionalized particles and the tailoring of the
band gap b y changing the size of the particles. The enhancement of
mechanical and thermal properties of pol ymers by the inclusion of
inorganic moieties, especiall y in the form of nanocomposites, offers the
possibilit y for these materials to substitute classical compounds based on
metals or on traditional composites in the transportation industry or as
fire retardant materials for construction industry. Medical materials are
also one t ypical application area of hybrid materials, as their mechanical
properties can be tailored in combination with their biocompatibilit y, for
example nanocomposites for dental filling materials. A high content of
inorganic particles in these materials provides the necessary toughness

36

and low shrinkage, while the organic components provide the curin g
properties combined with the paste-like behavior. Additional organic
groups can improve the adhesion properties between the nanocomposites
and the dentine. Composite electrol yte materials for applications such as
solid-state lithium batteries or super capacitors are produced using
organic–inorganic pol ymeric s ystems formed by the mixture of organic
pol ymers and inorganic moieties prepared by sol–gel techniques. In these
s ystems at least one of the network-forming species should contain
components that allow an interaction with the conducting ions. This is
often realized using organic pol ymers which allow an interaction with the
ions, for example via coordinative or by electrostatic interactions. One
typical example is proton conducting membranes which are important for
the production of fuel cells. The application of hybrid composites is
interesting for these s ystems because this membrane is stable at high
temperatures compared with pure organic s ystems. These are onl y some
applications for h ybrid materials and there is a plethora of s ystems under
development for future applications in various fields.
1.5 SUMMARY
Hybrid

materials

represent

one

of

the

most

fascinating

developments in materials chemistry in recent years. The tremendous
possibilities of combination of different properties in one material
initiated an explosion of ideas about potential materials and applications.
However, the basic science is sometimes still not understood, therefore

37

investigations in this field in particular to understand the structure–
propert y relationships are crucial. This introduction has shown the
importance of the interface between the inorganic and organic materials
which has to be tailored to overcome serious problems in the preparation
of hybrid materials. Different building blocks and approaches can be used
for their preparation and these have to be adapted to bridge the
differences of inorganic and organic materials. Beside the preparation of
hybrid materials, their nano- and microstructure formation, processing and
anal ysis is important.

RESEARCH OBJECTIVES
Why are these POSS/pol ymer studies significant? Polymers have
properties which make them very useful industriall y and commerciall y.
Most pol ymers in common use such as polyethylene and pol ypropylene are
organic in nature and have limited thermal stabilit y. M y goal was to stud y
and therefore be in a position to guide in the development of even more
useful materials (i.e. plastics) which have properties intermediate between
these traditional organic s ystems and inorganic ceramics - hybrid
materials, in other words. POSS macromers, having the general formula
(RSiO 1 . 5 ) n , can be incorporated into the traditional linear, thermoplastic
s ystems either as pendants or as part of the pol ymer backbone to impart
thermal stabilit y, resistance to oxidation, flame and heat resistance, while
avoiding the formation of intractable, crosslinked networks. While a
number of these s ystems have been s ynthesized, it was important to

38

achieve an understanding of the structure-properties relationships in these
organic-inorganic hybrid pol ymers, in particular, the origin of the
reinforcing and enhancing effects of the POSS structure and the nature of
the POSS-pol ymer interactions.
Even though POSS can be attached onto the backbone of the
pol ymers, in this stud y we use onl y full y condensed POSS s ystems where
b y altering the organic functional groups appended to Si-O core where
different extent of compatibilit y can be controlled. The focus of this stud y
is to investigate the influence of the various organic groups on the
interactions

between

POSS

and

pol ymer

which

would

govern

the

properties of the POSS-pol ymer blends.
1) Dispersibility of POSS Moieties. Miscibility effects on Polymer
Microstructure.
2) Understanding the morphology/rheology properties of various POSSpolymer blends.
3) Nanoparticulate loading effects in polymer blends with respect their
Viscoelastic properties
4) Polymer Chain length effects on the linear rheological properties of
the nanoparticle/polymer blends.
5) Effect of large deformations on the dynamics of the polymer chain
segments and their segmental motions.

39

CHAPTER 2
EFFECTS OF CHEMICAL SUBSTITUENTS ON
PHASE BEHAVIOR OF POSS – PS BLENDS
2.1 INTRODUCTION
Literatures regarding the s ynthesis, structure and properties have
their origins in the late 1940’s. The first oligomeric organo silsequioxanes
were first s ynthesized b y Scott in 1946 4 1 and this work in later years was
done b y Brown 4 2 . The research on POSS based s ystems has been stagnant
until the earl y 90’s when researchers like Feher 4 3 - 4 4 and Lichtenhan 4 5
focused on exploiting the advantages of this unique silica cage like
frameworks

containing

various

functionalities.

A

variety

of

POSS

nanostructured chemicals have been prepared which can either be
chemicall y inert or containing reactive chemical functionalities which can
be

used

to

grafting,

pol ymerization,

surface

bonding

or

other

transformations 4 6 - 4 7 .
The s ynthesis of POSS and its derivatives can be formed through
various reactions which can be categorized into two main t ypes. The first
being the formation of new Si-O-Si bonds which eventuall y leads to the
cage framework of POSS containing different organic functionalities, the
second being the variation of the group on the Si through different
techniques. There are two excellent reviews by Voronkov 1 2 and Feher 1 3

40

which extensivel y talk about the various reactions available in literature
for the s ynthesis of POSS macromers and its derivatives.
POSS macromers can be incorporated into pol ymeric systems
through

two

techniques.

First

through

producing

hybrid

pol ymers

containing POSS, where POSS macromers which contain functional
organic reactive sites are pol ymerized or copol ymerized with other
organic

species

homopol ymers

to
or

produce

a

copol ymers.

variety
The

of

organic-inorganic

second

being

hybrid

POSS/pol ymer

nanoscopic blends where POSS macromers are used as nano-reinforcing
agents, are blended with pol ymers to produce inorganic/organic nanoreinforced materials.
POSS macromers can be pol ymerized using standard pol ymerization
protocols like radical pol ymerizations, condensation pol ymerizations, ring
opening pol ymerizations etc., to provide pol ymers with a variet y of
architectures. Depending on the t ype of functionalit y contained on the
POSS

macromers

and

on

the

desired

pol ymer

architecture,

POSS

macromers can be introduced into macromolecular s ystems as either a
main chain, side chain, or as chain terminus groups 3 9 .
Several POSS homopol ymers and copol ymers have been s ynthesized
and characterized, for example: POSS-Styryl based homopol ymers and
copol ymers 4 8 - 4 9 ,

methacrylates-POSS

pol ymers 5 0 ,

norboryl-

POSS

copol ymers 5 1 , POSS-siloxane copol ymers 5 2 , POSS-epox y polymers 5 3 - 5 4 ,
POSS-pol yurethane copol ymers 5 5 - 5 6 . The propert y studies of these POSS

41

containing pol ymers showed that because the massive inorganic groups
(i.e. POSS cages) are attached to polymer chains, the POSS-pol ymer
chains act like nanoscale reinforcing fibers or like a hard block phase
separated from a soft block, producing enhanced heat resistance and
mechanical properties.
Currentl y there

are

relativel y few

literatures

concerning

the

investigation of POSS/pol ymer nano-materials, where POSS macromers
are incorporated into pol ymer by blending. Though comparativel y a lot of
POSS copol ymer work has been published which talk about the POSS
incorporation benefits but most of them have no s ystematic chemistr y
involved. Lichtenhan 4 6 was the first to work on POSS/pol ymers blends in
the earl y and mid 90’s where most of the work was done on PMMA and
POSS blends and talks mostl y with reference to thin films rather than the
bulk propert y influence.
Later Fu, Yang, Somani, etc 5 8 . studied the aggregation behavior of
isotactic pol yprop ylene containing nanostructured POSS at quiescent and
shear states. The results showed that the addition of POSS significantl y
increased the crystallization rate during shear. The authors postulate that
POSS molecules behave as weak cross linkers in pol ymer melts and
increase the relaxation time of iPP chains after shear. Therefore, the
overall orientation of the pol ymer chains is improved and a faster
crystallization rate is obtained with the addition of POSS. Most of the

42

literature published and the current progress in POSS/pol ymer systems are
covered in a recent review articles by Pittman 5 9 and Shawn Phillips 6 0 .
Molecular d ynamic simulations on POSS filled pol ymers have been
done b y Starr et al 6 1 and Glotzer et al 6 2 which talk about the influence on
the pol ymer chains near the surface of the nanoscopic filler and hence the
influence on the glass transition temperature and the agglomeration
behavior of POSS respectivel y. Most recentl y a McKinley’s group from
MIT has published their work which talks about the influence of POSS on
PMMA both when tethered and untethered onto PMMA chains 6 3 . The
authors

talk

about

that

POSS

acting

as

plasticizers

at

very low

concentrations (<5wt %) and at higher concentrations POSS agglomerates
in the matrix.
The detailed s ynthesis of the POSS used in this work is presented in
a pol ymer preprint paper by Blanski et al 6 4 . The MALDI of the POSS used
in this stud y were done by Bowers et al 6 5 . The combining of POSS
macromers with polymer by blending is a new approach to achieve nanomaterials, there are still a lot of unknown domains that need to be
investigated for this technique. The key issue in this field is to develop
new design principles that allow to control POSS macromers at nanometer
level and to achieve effective POSS-pol ymer interface.
development

of

POSS

macromers

affords

an

The recent

opportunity

for

the

preparation of new pol ymers (thermoset, thermoplastics, and elastomers)
and new pol ymeric blends. As part of our ongoing effort to understand,

43

and develop these materials an investigation into the morphology and
thermal properties of these POSS macromers was initiated. This stud y is
essential to optimizing the processing of the POSS/Pol ymer blends, and
for understanding the structure-propert y relationships of the blends, since
any given propert y of a multi-component s ystem is some (more or less
complex) function of the properties of the constituents and of the
interactions between them.
In the following chapter, the morphology and thermal properties of
the

POSS

macromers

were

examined,

using

X-ray

diffraction,

Transmission Electron Microscopy (TEM) and Differential Scanning
Calorimetry (DSC), with special emphasis given to the effects of the
organic corner groups on the POSS cages. We expect that the chemistry of
the corner groups affect the degree of miscibilit y inside the POSS
Pol ymer blends.
2.2 EXPERIMENTAL
2.2.1 Materials:
POSS Macromers used in this study are all T 8 cages bearing
different corner groups. These POSS macromers were obtained from
Hybrid

Plastics

Corporation.

Their

chemical

formulae,

structures,

abbreviations and molecular weight (M.W.) are shown in Table 2.2 and
Figure 2.2. Monodispersed PS was obtained from Pressure Chemicals.
PS2M: Mw=2316000, (pol ydispersit y less than 1.6) and PS 290K: polystyrene
with Mw= 290K Daltons (polydispersity less than 1.1).
44

Vinyl POSS

Isobutyl POSS

Cyclopentyl POSS

Cyclohexyl POSS

Phenethyl POSS

Styrenyl POSS

Figure 2.1: POSS Macromers with Different Corner Groups

45

Full Chemical Name

Abbreviations
for POSS

Chemical
Formula

Molecular
Weight

Octa-C yclopent yl-POSS

(Cp) 8 POSS

C 4 0 H 7 2 Si 8 O 1 2

969.7

Octa-C yclohex yl-POSS

(C y) 8 POSS

C 4 8 H 8 8 Si 8 O 1 2

1081.9

Octa-St yren yl-POSS

(St) 8 POSS

C 6 4 H 5 6 Si 8 O 1 2

1241.8

Octa-Phenethyl-POSS

(Phe) 8 POSS

C 6 4 H 7 2 Si 8 O 1 2

1257.9

Octa-Isobut yl-POSS

(iBu) 8 POSS

C 3 2 H 7 2 Si 8 O 1 2

873.6

Octa-Vin yl-POSS

(Vi) 8 POSS

C 1 6 H 2 4 Si 8 O 1 2

633

Table 2.1: Abbreviations, Chemical Formula and Molecular Weight of POSS Macromers

46

2.2.2 Sample Preparations:
2.2.2.1 Samples Preparations for Differential Scanning Calorimeter
(DSC)
POSS powders were used directl y for DSC tests. Preparation of
POSS/PS blends were performed by dissolving PS and POSS macromers in
chloroform and stirring the solution for 12 hours using a magnetic stirrer
and crashing the sample out by using a bad solvent methanol. The
precipitated sample on crashing out of the solvent is filtered and then
dried under vacuum for more than 12 hours at 60 o C.
2.2.2.2 Samples Preparations for X-Ray Diffraction:
POSS macromers were used as they were, POSS/PS blends were
prepared b y dissolving both POSS and PS chloroform and stirred for 12
hours with a magnetic stirrer. The mixture was poured on a glass surface
and covered with a slight gap to allow slow evaporation of solvent to form
a thin pol ymer film. The as-cast film was dried in a vacuum oven for 24
hours at 60 o C.
2.2.2.3 Samples Preparations for Transmission Electron Microscopy
(TEM):
TEM images were obtained using a JEOL 100CX TEM using an
acceleration voltage of 120 KV. PS and POSS were dissolved in THF at a
concentration of 5 mg per ml of THF and stirred for more than 3 hours.
The solution was then dropped onto a glass slide and let to dry in air. The
film was removed from the glass slide by slowl y immersing the slide into

47

a vessel of water at an angle of 45 o to the surface.

TEM grids were

dropped onto the film and the test assembl y was lifted out of the water
using a section of paper. The samples were dried on filter paper and
subsequentl y carbon coated to increase their beam stabilit y.
2.2.3 Characterization Techniques:
2.2.3.1 Transmission Electron Microscopy (TEM)
TEM was utilized to characterize the morphologies of the POSS/PS
blends.

2.2.3.2 X-ray Diffraction:
X-ray diffraction was employed to identify the morphological
changes of POSS macromers after they were blended with PS. X-ray
diffraction measurements were performed using a Scintag 2000 XRD with
Cu-K α target (λ=1.5406Å) radiation generated at 45 kV and 200 mA. The
diffraction angle was ranged from 5 o to 30 o , with step size and scan rate
of 0.05 o and 2 o per minute, respectivel y.
The X-ray diffraction pattern obtained from a diffractometer
records the X-ray intensit y as a function of diffraction angle. The interatomic spacing is determined by Bragg’s law:
d = n λ / (2sinθ)
Where d is the inter-atomic spacing; λ is the wavelength of the xray (λ =1.5406Å for Cu target); θ is the diffraction angle.

48

2.2.3.3

Determination

of

Transition

Temperatures

of

POSS

macromers:
Transition temperatures were determined using a Mettler-Toledo
821e/400 Differential Scanning Calorimeter (DSC) under a flow of
nitrogen and with a heating rate of 10 o C per minute. The transition
temperature is taken as the inflection point of the transition region.
2.3 RESULTS AND DISCUSSION
2.3.1 Effects of POSS Corner Groups:
Differential Scanning Calorimetry (DSC) was used to test the
transition temperatures of POSS macromers with different corner groups.
The DSC results are showed in Figures 2.2 to 2.5, and the transition
temperature data are listed in Table 2.2. Some of these transitions shown
in the DSC figures are the fusions of POSS crystalline structures, and
some of them are the destruction of POSS weak associations. The heats of
these fusions or disassociations of POSS macromers are very weak
(between 2~40J/g).
We can see that the Cp 8 POSS (Figures 2.2), St yren yl 8 POSS (Figure 2.4),
have

two

transition

peaks:

12.5 o C

(disassociation)

and

29.9 o C

(disassociation) for Cp 8 POSS; 189.1 o C (disassociation) and 274.3 o C
(melting) for St 8 POSS; and 49.6 o C (disassociation) and 272.1 o C (melting)
for Isobu 8 POSS. The disassociation temperature of the C y 8 POSS is around
23.7C. Ph 8 POSS, have a melting temperature at 76.5C.

49

1
Cp8POSS

Heat of Flow (J/g)

0.5
0
-0.5
29.9oC

-1
-1.5

12.5oC

-2
0

50

100

150
200
250
Temperature (oC)

300

350

Figure 2.2: DSC Curve of Cp8POSS Macromer

0.9

Cy8POSS

Heat of Flow (J/g)

0.4
-0.1
-0.6
23.7oC

-1.1
-1.6
0

100

200
Temperature (oC)

300

Figure 2.3: DSC Curve of Cy8POSS Macromer

50

400

400

1
Styrenyl8POSS

Heat of Flow (J/g)

0.5
0
189.1oC

-0.5
-1

274.3oC

-1.5
-2
-2.5
0

100

200
Temperature (oC)

300

400

Figure 2.4: DSC Curve of Styrenyl8POSS Macromer

1
Ph8POSS

0.5

Heat of Flow (J/g)

0
-0.5
-1
-1.5
-2

76.5oC

-2.5
-3
-3.5
-4
0

100

200
Temperature (oC)

300

Figure 2.5: DSC Curve of Phe8POSS Macromer

51

400

Transition
Temperature
( o C)

POSS
Macromers

Heat

of Dissociation
(J/g)

1

2

1

2

Cp 8 POSS

12.5

29.9

4.4

6.2

Cy 8 POSS

23.7

2.5

~

Styrenyl 8 POSS

189.1

274.3

4.5

29.6

Ph 8 POSS

76.5

~

37.8

~

~

Table 2.2: Transition temperatures of POSS macromers (DSC Results)
The melting transition of St 8 POSS is observed in the DSC
experiments having higher heat of fusion, relativel y Cp 8 POSS, C y 8 POSS
and Ph 8 T 8 macromers are observed to have very low heat of fusion in their
DSC curves, and we assume that it is because of the weak interactions
between the POSS macromers. The small disassociation peaks found in
Cp 8 POSS, C y 8 POSS and St yren yl 8 POSS macromers are also indicative of
destruction of weak aggregations in these POSS macromers.

2.3.2 Miscibility in Polystyrene Based on Different POSS:
The morphology of the PS/POSS blends was observed by using a
Transmission Electron Microscope. These experiments were done to stud y
the effects of different chemical moieties of POSS used under similar
POSS loadings on the morphological structures of the PS/POSS blends.
From previous studies done in our group by haipeng different phase
characteristics have been observed in these blends. A higher molecular
52

weight PS was used, 1.6M Da to facilitate better mechanical properties to
the film especiall y since very dilute solutions were employed to cast the
sample films.
Figure 2.6 depict the TEM photographs of PS2M (Molecular weight
2M Daltons) blended with Cp 8 POSS, St 8 POSS, Phe 8 POSS, Cy 8 POSS.

A

high POSS loading (50wt %) is used in this study to facilitate better
contrast between the dispersed and the continuous phases; if at all phase
separation is observed.
As s hown i n Fi gure 2. 6 , t he C p 8 P OS S /PS 2M bl end has t wo ph as es
wi t h

PS

as

t he

cont i nuous

phas e

and

s nowfl ake

like

C p 8 P OSS

m acrom ers as t h e di s pers e ph as e. Th e di m ens i ons of t h e C p 8 P OS S
ag gre gat es ar e i n order of a m i cr on. The substitution of all the
cyclopent yl groups in the Cp 8 POSS macromer with st yrenyl groups
(St yrenyl 8 POSS) renders a two phase s ystem in the St yren yl 8 POSS/PS2M
blend: PS rich phase (white round particles) becomes the disperse phase,
and the continuous phase is the POSS rich Phase, is a mixture of
Styrenyl 8 POSS and PS. The replacement of all the cyclopentyl groups in
Cp 8 POSS macromer with phenethyl groups dramaticall y improves the
compatibilit y between Ph 8 POSS and PS. The Ph 8 POSS macromer is
homogeneousl y dispersed in the PS matrix. There is no phase separation
in the Ph 8 POSS/PS2M blend. A phase separation is also observed in
Cy 8 POSS/PS2M,

with

POSS

being

continuous phase.

53

the

disperse

phase,

PS

as

the

POSS-Molecular Silica Blends
Blended into 1.6M Da MW Polystyrene

R
R = cyclopentyl
O
O

R

Si

O

Si

O

Si

R

R = cyclopentyl

R
R

Si

O

R
Si
O
Si

R

O

Si

O

Si

R O
O
Si
O

R

R

R

O
O

O

Si
O

Si

O
RR=Sipenethyl
Si

Si

Si

O
O

partial compatibility

O

O

R O

R = cyclopentyl

domain formation

Si
R = styrenyl
O O

O

R

Si
O

Si

R
Si
O

R

O

Si
O

Si O

O

Si

O
Si

O

R

R O
O

R
Si
O

R

O

Si

O

Si

RR= Cyclohexyl
O

R

Si
O

Si
R

O

Si

R
Si
O

R

R

O

O

O

R O
O

O
Si

Si
O
R

R = Phenethyl

R = styrenyl

transparent

phase inversion

Figure 2.6: TEM Image of POSS PS blends at 50 wt% loadings
54

R

From the morphology studies it clearl y suggests that the organic
pendent group on the POSS macromers play a crucial role in determining
the dispersibilit y of the nanoparticulates, self similar s ystems similar to
the pol ymer matrix tend to show better dispersibilit y than other s ystems.
For better understanding of the blend characteristics we have
considered the two cases of Ph 8 T 8 POSS/PS and the St 8 T 8 POSS/PS blends
due to the similarit y of the organic pendent group with the polymer matrix
and also based on the TEM images showed the better miscibilit y among
the blend s ystems. Even though in the case of St 8 T 8 POSS/PS2M blends
phase separation takes place into a two phase s ystem where both POSS
and PS seem to exist in both the continuous and the dispersed phases lack
of well defined crys tallite domains like in the case of Cp 8 T 8 and C y 8 T 8
suggests that something different was driving the phase behavior of these
blends.
TEM

images

showing

phase

separation

in

the

case

of

St 8 T 8 POSS/PS1.6M blends suggests either the high molecular weight of
PS used and higher loadings of POSS used might have shifted the Phase
diagram to show a two phase s ystem or the higher molecular weight might
have hindered the crystallization process of the POSS macromers. To
further elucidate on this issue lower molecular weight pol ystyrene s ystem
has been used with lower POSS loadings since concentration is a driving
force for phase behavior, to study the miscibilit y of St 8 T 8 POSS in PS. To
further examine the microstructures of POSS macromers, TEM, DSC and

55

X-Ray diffraction anal ysis was performed. The study below is mainl y
focused on the impacts of st yren yl and phenethyl on the morphology of
the POSS/PS blend systems.
The effects of the PS on the crystalline structures of POSS were
studied b y comparing the X-ray diffraction patterns of the POSS
macromers in the POSS/PS blends

with those of the neat POSS

macromers. The X-ray diffraction pattern obtained from a diffractometer
records the X-ray intensit y as a function of the diffraction angle 2θ, and it
gives the information about the orderly packing of the molecules in
crystals. Figure 2.7 shows the X-ray diffraction curves of Ph 8 T 8 POSS
macromers with and without PS; Figure 2.8 shows the X-ray diffraction
curves of St 8 T 8 POSS macromers with and without PS. Table 2.5 has their
corresponding diffraction data (peak positions, inter-atomic spacing, and
the 2θ width of the diffraction peaks). Onl y some weak and broad peaks
are found in Ph 8 POSS macromer, and Styrenyl 8 POSS macromer has onl y
one weak peak in its diffraction pattern.
From Figures 2.8 and 2.9, it could be seen that the crystalline POSS
peak in the case of PS blends containing 20 wt% of POSS disappear for
Phe 8 T 8 and St 8 T 8 respectivel y compared to the pure POSS macromers
results, suggesting the absence of crystallite domains and indicating good
compatibilit y between the POSS macromers and PS. The disappearance of
the crystalline POSS peak suggests that the POSS macromers are
completel y dispersed in the PS matrix.

56

Figure 2.7: X-Ray plot of Phenethyl POSS (Ph8T8) and PS (MW= 290K Da) / Phenethyl POSS 20%wt blend

57

2500

Styrenenyl POSS-Pure
20wt% Styrenenyl POSS/PS

Absorbance

2000

1500

1000

500

0

5

10

15

2Θ

20

25

30

Figure 2.8 X-Ray plot of Styrenyl POSS (St8T8) and PS (MW= 290K Da) /Styrenyl
POSS 20%wt blend

58

The X Ray diffraction of the pol ymer blend s ystems suggest
miscibilit y contrary what we have observed in the TEM images showing
phase separation in the case of St 8 POSS/PS blends, this suggests the high
molecular weight of PS used and higher loadings of POSS used might
have shifted the Phase diagram to show a two phase s ystem and also the
pol ymer chains might have hindered the crystallization process of the
POSS

macromers,

to

further

verify

the

above

observations

DSC

experiments have been performed along with TEM to get further insight
on the phase characteristics of these blends. Figure 2.10 shows the TEM
images of St 8 POSS macromers with 290K PS, Figure 2.10 suggests the
existence of two phase s ystem behavior with both the pol ymer rich phase
and the POSS rich phase. The results reflect the degree of compatibilit y
between the POSS macromers and PS.
Figures 2.10 and 2.11 depicted DSC curves of Ph 8 T 8 POSS, its
blend with PS, St 8 T 8 POSS and its blend with PS, respectivel y.

As

observed from these DSC curves, the value of heat of fusion for Ph 8 T 8 and
St 8 T 8 POSS is nearl y identical (~ 37.7 kJ/mole). In addition, for the case
of 30wt% POSS/PS blends, we did not observe any crystalline melting
peaks associated with the pure POSS macromers, which suggests that
POSS-PS blends form a single phase mixture.
The melting transition temperature from the DSC measurement can
be used to estimate the strength of associations within the POSS
crystallites. As determined from the DSC, the onset of melting transition

59

peak for St 8 T 8 is around 274 o C and Ph 8 T 8 was at 77 o C.

The higher

transition temperature indicates a stronger POSS-POSS association in
St 8 T 8 as compared to Ph 8 T 8 .
DSC results were also used to examine the effect of POSS addition
on the T g of blends. For the case of PS blended with Ph 8 T 8 POSS, the
value of T g decreases on the addition of POSS macromers from 106.07C
for the pure PS to around 91.88C with POSS, in the case of Styrenyl POSS
blend the T g decreases to 103.66. We believe the suppression of T g in the
case of Ph 8 T 8 blends was caused by the well-dispersed POSS clusters in
PS and the excellent mutual solubilit y between Ph 8 T 8 and PS thereb y
causing a swelling in the pol ymer network, similar to that of solvent
plasticization

effect,

which

was

further

measurements in the next chapter.

60

evaluated

by

rheological

200nm
Figure 2.9: TEM Image of Styrenyl8POSS/PS290K Blend (20 wt% POSS)

61

Ph8T8
PS/30% Ph8T8 POSS blend

DSC
0

20

40

60

80

100

120

140

-2
-4

Heat Flow

-6
-8
-10
-12
-14
-16
-18

Temperature (C)

Figure 2.10: DSC plot of Phenethyl POSS (Ph8T8) and PS (MW= 290K Da) /
Phenethyl POSS 30%wt blend

62

160

DSC
0

St8T8
PS/ 30% St8T8 POSS blend
50

100

150

200

250

300

350

-1
-2

Heat Flow

-3
-4
-5
-6
-7
-8
-9
-10

Temperature (C)

Figure 2.11: DSC plot of Styrenyl POSS (St8T8) and PS (MW= 290K Da) /Styrenyl
POSS 30%wt blend

63

As for St 8 T 8 /PS blends, although POSS may be well dispersed, but
there was a lack of interaction between POSS and PS due to the two phase
s ystem obtained from the TEM Images. The reason for us to believe it
being dispersed is from the lack of any melting peaks as seen from the
DSC theromogram. The lack of thermal transitions indicates the domains
present if present are too small for the DSC to differentiate through a
change in heat flow signal. The presence of two phases as indicated b y
TEM images may have occurred because of a shift in the phase diagram
due to the solvent evaporation technique employed in preparing the
samples. The evaporating solvent may have driven the POSS blend to
form a two phase s ystem. To further investigate various blends with
different

POSS

loadings

have

been

used

and

their

corresponding

rheological characterization done to evaluate the phase characteristics in
the next chapter.
2.4 SUMMARY
The chemistry of POSS macromers plays an important role in
determining the morphologies of the POSS/PS blends. Depending on the
attached chemical groups on the POSS macromer, the morphologies of
POSS/PS blends ranged from a complete phase separation between POSS
and PS to a homogeneous dispersion of POSS in the PS matrix in a
nanoscopic scale. Among the five POSS macromers used, Ph 8 POSS is the
most compatible one with PS and can be homogeneousl y dispersed in PS

64

matrix. All other POSS/PS blends display a certain amount of phase
separation to a various degrees.
The morphology of a blend depends on the compatibilit y between
the two components, the t ype of molecular interaction and the resultant
interface

of

the

components

and

the

processing

history.

For

POSS/Pol ymer blends, the following factors influence the compatibilit y
between the two components, and hence the morphologies of the blends.
The chemistry of POSS, the chemical structures of the corner groups on
the POSS cage greatl y influence the compatibilit y between POSS and
pol ymers. The above studies showed that the compatibilities between
POSS and PS varied with different POSS chemical structures. Notabl y,
among the five POSS macromers studied, the Ph 8 POSS macromer is the
most compatible one with PS and, hence, it is the one which can be
homogenousl y dispersed in the PS matrix. The degree of the crystallinit y
of POSS macromers also affects the morphologies of the POSS/PS blends.
The potential for achieving miscible blends in which one or both
components are crystalline is low because of the heat of fusion which
would have to be overcome to achieve the necessary thermodynamic
criteria for mixing. High crystallinit y POSS does not favor the formation
of homogeneous dispersion of the POSS. POSS macromers with a strong
tendency to crystallize are inclined to aggregate and phase separate from
the pol ymer host.

65

The chemistry of the corner groups on the POSS cage affects the
morphological structures

of the POSS macromers.

The higher the

s ymmetry and regularit y of the POSS macromers, and the smaller the size
of the corner groups, the more ordered the POSS macromers.

66

CHAPTER 3
EFFECTS OF CONCENTRATION ON PHASE
BEHAVIOR OF POSS – PS BLENDS
3.1 INTRODUCTION
The effect of concentration on the viscometric properties of the
nanocomposite blends has been investigated. Rheological measurements
well above T g have been used to determine the time-temperature
superposition behavior of these thermorheologicall y simple materials.
The size of POSS macromers on similar length scales when compared to
the entanglement spacing of the pol ymer chains makes it interesting to
study the role these nano-structured constituents would play in controlling
the pol ymer chain dynamics. However as we have seen Chapter 2 various
organic groups on the POSS macromers led to the complex morphologies
of various POSS/PS blends, some miscible some showing two phase
behavior with both phases containing both the components and some
showing complete segregation as filler systems. The rheological response
of the filler s ystems has been widel y studied and the effect a filler has on
the viscoelastic properties widel y understood, the part which is still
sketch y is the understanding of the presence of miscible or partl y miscible
nanoparticulates in the pol ymer matrix, and the part to be yet answered is
do these changes in microstructure cause profound changes to the
relaxation of the polymer backbones. To address the above issue we took

67

two POSS macromers having the organic pendant group as the same or
nearl y similar

as

the

monomer

unit

as

the

pol ymer.

Rheological

characterization was done for understanding the chain dynamics due to the
presence of POSS or due to the domains formed due to the presence of
POSS.
3.2 EXPERIMENTAL DETAILS
3.2.1 Materials
Nearl y monodisperse molecular weight distribution of pol ys tyrene
with M w = 290K Daltons (pol ydispersit y less than 1.1) were purchased
from Pressure Chemicals.

Two full y condensed POSS, St yrenyl 8 -POSS

(St 8 T 8 , 1) and Pheneth yl 8 -POSS (Ph 8 T 8 , 2) and were used in this study.
Synthesis of POSS macromers used in the study 6 4
Degassed toluene (Fisher) was purified by elution through an
alumina column.

Grubbs Catal yst Cl 2 Ru(=CHPh)(PC y 3 ) 2 (Strem) was

used as received.

Styrene (Aldrich) was purified by distillation under

diminished pressure onto ~10 ppm of catechol. Hydrogen (Scott Specialt y
Gases) and 10% Palladium on Carbon (Aldrich) were used as received.
NMR Spectra were taken on a Bruker Avance 300 spectrometer
Synthesis of Styrenyl 8 T 8 (1): A 500 mL round bottom flask was charged
with 15 grams (26.7 mmol, 189.5 mmol vinyl) of Vinyl 8 T 8 , 35mL (305
mmol) of St yrene, 50 mL of toluene and a stir bar under nitrogen. 200 mg
(0.24 mmol) of the Cl 2 Ru(=CHPh)(PC y 3 ) 2 (dissolved in 10 mL of toluene)
was added to the reaction via s yringe. A slight vacuum was applied to the

68

flask to assist in the volatilization of ethylene that was generated in the
process and the mixture was allowed to stir overnight. The volatiles were
removed under vacuum and redissolved in toluene. To this mixture was
added 30 grams of montmorillonite clay and the mixture stirred for 45
minutes.

The reaction was filtered through Celite and the colorless

filtrate is evaporated to give a colorless solid. This solid is heated under
vacuum at 150 ºC for two hours and then 100 ºC overnight to remove an y
stilbene formed. 27 grams of 1 is obtained as a free flowing solid (92%
yield).
Synthesis of Phenethyl 8 T 8 (2): A 300 mL Parr reactor was charged with
20 grams of St yren yl 8 T 8 , 50 mg of 10% palladium on carbon, and 30 mL
of toluene.

The reactor was pressurized to 500 psi of hydrogen and

allowed to stir overnight at 50 °C.

The reaction was cooled to ambient

temperature and the solution filtered through Celite. The volatiles were
removed and the remaining liquid was heated to 150 °C under vacuum for
two hours. 19.5 grams of a colorless solid was obtained (97% yield.)

1

H

NMR (300.1 MHz, CDCl 3 ) δ 7.381 (m, 5H, C 6 H 5 ), δ 2.904 (m, 2H,
CH 2 Ph), δ 1.152 (m, 2H, Si-CH 2 ).

13

C NMR (75.5 MHz, CDCl 3 ) δ 143.97,

128.32, 127.84, 125.74, 28.93, 13.73.
67.2.

69

29

Si NMR (59.6 MHz, CDCl3) δ -

3.2.2 Sample Preparation
3.2.2.1 Melt Rheological Characterization
Preparations of POSS/PS blends were performed by dissolving PS
and POSS macromers in chloroform and stirring the solution for 12 hours
using a magnetic stirrer and crashing the sample out by using a bad
solvent methanol. The precipitated sample on crashing out of the solvent
is filtered and then dried under vacuum for more than 12 hours at 60 o C.
Rheological measurements were carried out using a Rheometric
Scientific ARES rheometer equipped with a force convection oven in the
parallel plate geometry.

Parallel plate geometry with diameter of 8 mm

and gap of 0.5 mm was used for all measurements obtained in this stud y.
Rheological behavior of all samples was investigated using a series of
isothermal, small-strain oscillatory shear with oscillatory frequency
ranging from 100 to 0.1 radian/second and oscillatory shear strain
amplitude of 5%.

To examine the validit y of the time-temperature

superposition principle and the effect of POSS addition on the values of
time-temperature shift factor, sample was tested at temperatures ranging
from T g +10 o C to T g +100 o C with 10 o C intervals.
modulus,

G’(ω),

loss

modulus,

G”(ω),

and

Values of storage
damping

factor,

tanδ( ω) = G”(ω)/G’(ω) , were determined using software provided b y
Rheometric Scientific.
conditions

were

reproducibilit y.

Samples tested under the same experimental

repeated

two

to

three

times

to

confirm

their

Master curves for G’, G” and tanδ were also obtained

70

with software package provided by Rheometric Scientific and a reference
temperature of T r e f =150 o C for pure pol ymers and blends was used.
3.3 RESULTS AND DISCUSSION
3.3.1 Rheology during the Rubbery State and Flow Transition
The effect of POSS loading and distribution of POSS macromers in
PS

is

understood

through

the

thermal

and

morphological

studies.

Rheology experiments were performed to address the effect of POSS
macromers on the chain dynamics of PS. Since rheological measurements
and their correlation to pol ymer chain dynamics using narrow molecular
weight PS properties were well documented, our measurements of
POSS/PS blends will provide insight into how the POSS macromers affect
the pol ymer chain d ynamics.
The melt rheological experiments were carried out using smallstrain amplitude oscillatory shear so to minimize deformation on sample
morphology.

The dynamic properties of the POSS/PS blends were

investigated as a function of frequency at different temperatures. Changes
in GC (ω ), ω C , G 0N and lim ω →0 η * (ω ) were compared for all studied s ystems.
In addition, the validit y of time-temperature superposition was also
verified for all the samples used in this study.
Ph 8 T 8 POSS / PS blends:
Figure 3.1(a) is the time-temperature master curves shifted to 150 o C
for the PS (Molecular weight 290K Daltons) blended with varying
amounts of Ph 8 T 8 POSS. The corresponding plot of reduced complex

71

viscosit y versus reduced frequency is shown in Figure 3.1(b).

The

temperature shift factor versus temperature is plotted in Figure 3.1(c).
From the various reduced storage modulus curves shown in Figure 3.1(a),
we observed that these curves shift downwards and towards the right with
increasing amount of Ph 8 T 8 POSS blended. In addition, the values of the
plateau modulus, G 0N , and the cross-over frequency, ω C , as shown in Table
3.1, decrease with increasing amount of Ph 8 T 8 POSS added.

These

observations correspond to PS chains becoming less entangled when
blended with Ph 8 T 8 POSS, which suggests that the Ph 8 T 8 POSS molecules
are completel y dissolved into PS and form a homogeneous solution. The
concept of Ph 8 T 8 POSS as a good solvent for pol yst yrene is verified b y
plotting the normalized plateau modulus, G 0N (c) / G 0N (1), versus the
concentration of solution as depicted in Figure 3.3. For a good solvent,
we expect a power law dependence whose concentration dependant
exponent is approximatel y 2.3 irrespective of the t ype of solvent in which
the pol ymer chains are present 6 6 .
concentration

dependant

exponent

As shown in Figure 3.3, the
is

slightl y

less

than

2.3,

thus

confirming that the observed viscosit y reduction in Ph 8 T 8 POSS/PS blends
is due to the solvent effect.

72

Figure 3.1(a)
Figure 3.1 Time temperature superposed plots of PS (MW= 290K Da) and Phenethyl POSS (Ph8T8) blends with varying weight fractions
of POSS at 1500C a) Storage Modulus (G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT)versus Reduced frequency (aTω)
c) Shift factors (aT) versus Temperature (Temp)
73

Figure 3.1 (cont’d)

107
PS290K-Pure
PS290K-2%
PS290K-6%
PS290K-13.3%
PS290K-20%
PS290K-30%

6

η*/aT (Pa-s)

10

5

10

104

3

10
10-4

-3

10

-2

10

-1

10

aT ω (rad/s)

Figure 3.1(b)

74

0

10

1

10

102

Figure 3.1 (cont’d)

10

0

PS290K-Pure
PS290K-2%
PS290K-6%
PS290K-13.3%
PS290K-20%
PS290K-30%

-1

aT (

[]

)

10

-2

10

10

-3

140.0

150.0

160.0

170.0

180.0

Temp [°C]

Figure 3.1(c)

75

190.0

200.0

210.0

Figure 3.2(a)
Figure 3.2 Time temperature superposed plots of PS (MW= 290K Da) and Styrenyl
POSS (St8T8) blends with varying weight fractions
of POSS at 1500C a) Storage Modulus (G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT)versus Reduced frequency (aTω)
c) Shift factors (aT) versus Temperature (Temp)
76

Figure 3.2 (cont’d)

107
PS290K-Pure
PS290K-2%
PS290K-6%
PS290K-13.3%
PS290K-20%
PS290K-30%

η∗/aT (Pa-s)

106

5

10

104

103
-4
10

10-3

-2

10

0

10-1

10

aT ω (rad/s)

Figure 3.2(b)

77

101

10

2

Figure 3.2 (cont’d)

100

PS290K-Pure
PS290K-2%
PS290K-6%
PS290K-13.3%
PS290K-20%
PS290K-30%

aT (

[]

)

10-1

10-2

10-3
140.0

150.0

160.0

170.0

180.0

Temp [°C]

Figure 3.2(c)

78

190.0

200.0

210.0

0

10

GN(φ)/GN(1)

10-1

10-2

PS/Ph8T8 blends
PS/ST8T8 blends
2.3 Dependence
10-3

10-1

100

Volume Fraction φ

Figure 3.3 Log-Log plot of Normalized Plateau Modulus (GN (φ)/GN (1)) versus Volume
fraction (φ)

79

Tg
o
C

ωC

GC (ω )

lim ω →0 η * (ω )

G 0N

r ad /s

Pa

P a- s

Pa
2. 3 52 E + 0 5

10 6 .0 7

0. 0 18 2

4. 1 82 E + 0 4

6. 0 2E + 0 6

2%
6%
13 . 3 %
20 %
30 %
50 %
70 %
90 %

10 2 .2 3
10 0 .7 5
10 1 .4 4
98 . 57
91 . 88
N/ A
N/ A
N/ A

0. 0 26 0 9
0. 0 29 4 7
0. 0 27 5 7
0. 0 70 8 9
0. 1 82 7
N/ A
N/ A
N/ A

4. 6 8E + 0 4
3. 4 32 E + 0 4
3. 5 06 E + 0 4
3. 2 7E + 0 4
2. 7 1E + 0 4
N/ A
N/ A
N/ A

4. 5 1E + 0 6
2. 9 8E + 0 6
3. 5 8E + 0 6
1. 1 9E + 0 6
3. 7 7E + 0 5
N/ A
N/ A
N/ A

2. 2 31 E + 0 5
1. 5 43 E + 0 5
1. 6 86 E + 0 5
1. 4 16 E + 0 5
1. 0 45 E + 0 5
3. 6 80 E + 0 4
1. 8 70 E + 0 4
3. 0 05 E + 0 3

2%
6%
13 . 3 %
20 %
30 %

10 5 .3 9
10 4 .6 7
10 4 .9 1
10 5 .7 5
10 3 .6 6

0. 0 22 7 1
0. 0 17 0 3
0. 0 20 0 3
0. 0 20 3 7
0. 0 28 5 4

4. 2 7E + 0 4
3. 3 25 E + 0 4
3. 7 24 E + 0 4
3. 9 79 E + 0 4
2. 9 12 E + 0 4

4. 2 7E + 0 6
3. 3 25 E + 0 6
3. 7 24 E + 0 6
3. 9 79 E + 0 6
2. 9 12 E + 0 6

1. 6 67 E + 0 5
1. 6 44 E + 0 5
1. 9 04 E + 0 5
1. 7 42 E + 0 5
1. 5 66 E + 0 5

Φ
P S 2 9 0K
Pu r e
Ph 8T 8/ P S

St 8T 8/ P S

Table 3.1 The crossover frequency versus POSS weight fraction and the
zero-shear viscosit y based on an Ellis model fit
Φ=POSS weight fraction is ≈ volume fraction since densities of both
POSS and PS are approximatel y 1 .

G 0N is the plateau modulus and is determined as the value of G’ at
minimum tanδ position.
T g is determined as the midpoint of onset and end set of transition from
the DSC curve.

ω C is the crossover frequency and GC (ω ) is the modulus at the crossover
frequency
lim ω →0 η * (ω ) is the zero shear viscosit y from the master curve at 150 o C

based on Ellis model.
N/A Not Available
80

St 8 T 8 POSS / PS blends:
St 8 T 8 POSS is used to study the effect of POSS-POSS interactions on the
pol ymer

chain

d yn amics,

since

it

is

known

that

the

POSS-POSS

interaction in the case of St 8 T 8 is greater compared to Ph 8 T 8 POSS from
the DSC studies. Figure 3.2(a) is the time-temperature master curves
shifted to 150 o C for the PS (Molecular weight 290K Daltons) blended
with different amount of St 8 T 8 POSS.

The crossover frequency versus

POSS weight fraction and the zero-shear viscosit y based on an Ellis model
fit were tabulated and shown in Table 3.1. The reduced-complex viscosit y
versus frequency curves for various St 8 T 8 POSS fractions are shown in
Figure 3.2(b).

Data were shown for the 2%, 6%, 13.3%, 20%, and 30%

St 8 T 8 POSS blends with 290K PS on the same plot.
In Figure 3.2(a) the storage modulus curve and the cross over
frequency, ω C , from Table 1 shifts downwards with the addition of St 8 T 8
POSS. Although there are some similarities to the effect observed in the
case of Ph 8 T 8 /PS blends, however, no s ys temic change on ω c and G 0N were
observed. This is in agreement with the DSC results, where no significant
T g depression is observed.

We believe this observation can be explain

based on the effect of miscibilit y of St 8 T 8 POSS in PS as influenced b y a
stronger POSS-POSS interaction and the thermorheological behavior of
well-dispersed, albeit inert, nano-particulate filled pol ymer. As discussed
previousl y, the POSS-POSS association is stronger for St 8 T 8 POSS as

81

compared to Ph 8 T 8 POSS.

Consequentl y, the stronger POSS-POSS

interaction affect the miscibilit y curve of St 8 T 8 in PS in such a manner
that concentration fluctuations might occur thus leading to a two phase
behavior, one POSS-rich and the other polymer-rich, as shown in the TEM
images in Chapter 2.

Furthermore, the two-phase morphology remains

even with increases in the total POSS concentration.

The formation of

white drop-like domains can be seen as the dispersed PS domains with
lower POSS concentration, where the difference in free energies in both
the phases leads to the formation of drops in the homogeneous blend
where spinodal decomposition occurs.

We believe the droplet domain

which is rich in PS can reinforce the pol ymer matrix thus the value of
plateau modulus, G 0N , and ω c are not strongly affected by the addition of
St 8 T 8 POSS. Although the value of G 0N and ω c are not strongl y affected b y
the addition of St 8 T 8 POSS, however we observed a s ystemic decrease in
the value of zero-shear viscosit y,

lim ω →0 η * (ω ) , and the cross-over

modulus, G c (ω), as amounts of St 8 T 8 POSS increase.

This is consistent

with some other investigations of nano-particle filled pol ym ers 6 7 , where
the presence of nanoparticles decreases the viscosit y of the pol ymer
blend.

3.4 SUMMARY
Rheological

experiments

were

performed

to

investigate

the

influence of POSS addition on the melt dynamics of the polymer chains.
82

We are particularl y interested in the influence of well-dispersed, nanoparticulate additions on the zero-shear viscosit y, lim ω →0 η * (ω ) , plateau
modulus, G 0N and the cross-over frequency, ω C of pol ymer melts.
In the Ph 8 T 8 POSS/PS blends, because of the low POSS-POSS
interaction, an y POSS interactions are screen-out by the presence of PS.
Ph 8 T 8 POSS/PS blend forms a homogeneous solution.

The rheological

responses of these blends are the same as if as a good solvent was added
to PS.

We observed a s ystemic decrease in the zero-shear viscosit y,

plateau modulus, cross-over frequency as the concentration of Ph 8 T 8
POSS increases, as shown in Table 1.
For

St 8 T 8

POSS/PS

blends,

due

to

a

stronger

POSS-POSS

association, the blend forms a two-phase morphology. Regardless to the
concentration of St 8 T 8 POSS in PS, the POSS-rich form a continuous
phase and PS-rich domain as the discontinuous phase. As we increase the
concentration of St 8 T 8 POSS, onl y the fraction of PS-rich discontinue
phase is increased. Since St 8 T 8 POSS is molecularl y dispersed in PS, the
zero shear viscosit y of the blends decreases with the addition of St 8 T 8
POSS which is similar to that observed in the case of Ph 8 T 8 /PS blends
(shown in Table 1).

However due to persistence of the two-phase

morphology, no s ystematic changes in ω c and G 0N was observed with
increasing concentration of POSS since phase behavior controls the
rheological behavior of these blends.

83

The effect of concentration on the viscometric properties of the
nanocomposite blends has been investigated. Rheological measurements
well above Tg have been used to determine the time-temperature
superposition behavior of these thermorheologicall y simple materials.
The size of POSS macromers on similar length scales when compared to
the entanglement spacing of the pol ymer chains makes it interesting to
study the role these nano-structured constituents would play in controlling
the pol ymer chain dynamics. However due to the complex morphologies
of various POSS/PS blends, similar rheological behavior may be obtained
due to two different mechanisms. DSC, X-ray, TEM were used to
understand the morphology of the blends. Rheological characterization
was done for understanding the chain dynamics due to the presence of
POSS or due to the domains formed due to the presence of POSS.
The shape of the POSS macromers plays an important role in
governing the microstructure in the blend. The changes in microstructure
cause profound changes to the relaxation of the pol ymer backbones. Our
findings clearl y suggest that POSS nanoparticles which are compatible
with the pol ymer chains could be used to lower the free energy of the
pol ymer mixture and also reduce the viscosit y of a pol ymer mixture as a
plasticizer.

Though

miscibilit y

and

aggregation

kinetics

could

be

controlled in these types of nanocomposites since the chemistry of POSS
particles can be modified depending on the s ystem to be used, it should be
noted that restrictions on the organic pendant group on the silicone cage

84

plays an important role in governing the dynamics of s ystems involving
POSS. Interestingl y in the case of Octa styrenyl POSS PS Blends, at small
volume fractions, free (untethered) POSS nano-dispersed in the melt
appears to act as a plasticizer, reducing the magnitude of selected material
properties such as plateau modulus or zero-shear rate viscosity at a given
temperature, while simultaneousl y causing a decrease in free volume
(estimated from W LF coefficient anal ysis). As the volume of dispersed
POSS increases, steric effects of the filler dominate and the rheological
properties of the melt (modulus, relaxation time, viscosit y) are all
enhanced.

85

CHAPTER 4
EFFECTS OF PS MOLECULAR WEIGHT ON PHASE
BEHAVIOR OF POSS – PS BLENDS
4.1 INTRODUCTION
Molecular weight is the main structural parameter of pol ymer flow
behavior at temperatures above the glass transition temperature (for an
amorphous material) or the melting point (for a semi-crystalline pol ymer).
Melt viscosit y is a constant at low shear rates or frequencies. The
viscosit y in this region is known as the zero shear, or Newtonian,
viscosit y ηo. For low molecular weight pol ymers in which chain
entanglement is not a factor, the zero shear viscosit y is proportional to the
pol ymers molecular weight. However, above a critical molecular weight,
chains begin to entangle and the zero shear viscosit y depends much
stronger on molecular weight, proportional now to about the 3.4 power of
the molecular weight. Rheological measurements are therefore ideal for
stud ying the effects of molecular weight differences in resins as small
differences in molecular weight are manifested in large changes in
viscosit y. Whether a blend is compatible or not, also depends on
temperature; in this case the blend is considered partiall y miscible. If
blends are incompatible, mechanical energy is needed to disperse the
minor phase (mixing) and coalescence occurs if the blend morphology is
not stabilized. Interfacial forces such as the interfacial tension become

86

important

and

can

change the rheological

signature of

the

blend

significantl y. Moreover the elastic properties of non-compatible blends
depend on energy storage mechanisms at the interphase. The relaxation of
the dispersed phase itself is often much longer than the relaxation of the
pol ymer chains of the individual components. As a result of this fact, the
morphology can be modified to change the final product specific
properties.
The reptation model describes the motion of a pol ymer along its
contour as it passes the physical constraints imposed by the surrounding
pol ymers

and

predicts

the

molecular

weight

dependence

of

the

macroscopic diffusion coefficient. Pol ym er diffusion studies first focused
on simple linear chains in matrices of equall y simple polymers while
subsequent

studies

explored

more

complex

architectures

including

branched molecules, star molecules, and loops. Here we investigate the
diffusion

of

linear

pol ymers

in

the

presence

of

nanoparticles.

Nanoparticles with desirable properties are now widel y available, and
there

has

been

an

explosion

of

activities

focused

on

combining

nanoparticles with pol ymers to create pol ymer nanocomposites with
unique and valuable properties. Nanoparticles also provide access to a
new range of size differences between particles and pol ymers, wherein the
radius of gyration (Rg) of the pol ymer is considerabl y larger than the
nanoparticle. For example, POSS have diameters of 1.2 nm and a
pol yst yrene (PS) of 290 000 g/mol molecular weight has 2Rg = 30 nm.

87

The addition of nanoparticles to pol ymers provides a new challenge to
building a fundamental understanding of pol ymer dynamics in complex
environments. In the example of POSS and PS, one might predict that the
addition of the nanoparticles to a pol ymer would act as a diluent and
thereb y smoothl y increase both the free volume of the s ystem and the
diffusion coefficient as the POSS concentration increases as observed in
the previous chapter in the case of Phenethyl POSS PS blends. Also in the
same stud y the St yren yl POSS PS blends reports that the additional POSS
produces a minimum in zero shear viscosit y with increasing nanoparticle
concentration. Thus, the addition of nanoparticles to polymers has
unforeseen consequences on pol ymer diffusion, which in turn has direct
impact on practical problems, such as the stabilit y of nanoparticle
dispersions. Since the 1980s, elastic recoil detection (ERD) has been used
to stud y pol ymer diffusion and thereby played a key role in testing
leading pol ymer diffusion theories including reptation, constraint release,
and mutual diffusion. ERD is an ion beam technique that directl y provides
a concentration profile of diffusing species, t ypicall y a deuterated
pol ymer, as a function of depth into a matrix. But a more practical route
is the stud y of the pol ymer dynamics using rheology, the limitation being
its use in low nanoparticle concentrations to avoid solid like behavior.
Using rheology, we follow the crossover frequency along with the plateau
modulus and the zero shear viscosit y in pol yst yrene (PS) nanocomposites
to understand the influence of POSS on pol ymer dynamics. This new

88

understanding will enable better manipulation of pol ymer nanocomposite
properties and melt processing, and could impact our understanding of
macromolecular movement in h ybrid materials.
Based on our results in the previous chapter which indicate the
abilit y for POSS to very locall y interact with the matrix molecules
altering their mobility, and thereby able to either stiffen or plasticize the
material. Apart from understanding the concentration dependence on the
phase behavior the pol ymer molecular weight plays a crucial role in the
phase characteristics of the hybrid materials. Not only the phase
characteristics the corresponding dynamics involving the polymer chains
blended with the POSS spheres effects the dynamics of the blended
s ystem. In a confined s ystem like an entangled pol ymer melt blended with
POSS, the motion of either the POSS macromer or the pol ym er chain or
both should in some way be affected. Though this problem is studied
previousl y, several h ypothesizes have arised and several model theories
like the reptation theory along with constraint release mechanisms have
been used to explain what exact mechanisms control the dynamics of the
pol ymer chains due to the presence of huge (on a molecular length scale)
cage like particles. There are however, many questions left due to the
absence of a s ystematic study. Will the higher molecular weight pol ymer
chains take a longer time to relax due to the interference by POSS
macromers? Or the d ynamics of POSS macromers within the entangled
pol ymer chain play a role in governing the rheological characteristics of

89

these blends. We intend to address this problem through the study of
POSS-PS blends b y varying the molecular weight of the linear nearl y
monodisperse PS ranging from around 30KDa to around 1.6MDa.
4.2 EXPERIMENTAL
4.2.1 Materials
Nearl y monodisperse molecular weight distribution of pol ys tyrene
with M w = 1.6M, 650K, 290K, 130K and 65K Daltons (pol ydispersit y less
than 1.1) were purchased from Scientific Pol ymers. Two fully condensed
POSS, St yren yl 8 -POSS (St 8 T 8 , 1) and Phenethyl 8 -POSS (Ph 8 T 8 , 2) and
were used in this stud y.
4.2.2 Sample Preparation
4.2.2.1 Rheological Characterization: Preparations of POSS/PS blends
were performed b y dissolving PS and POSS macromers in chloroform and
stirring the solution for 12 hours using a magnetic stirrer and crashing the
sample out b y using a bad solvent methanol. The precipitated sample on
crashing out of the solvent is filtered and then dried under vacuum for
more than 12 hours at 60 o C.
Rheological measurements were carried out using a Rheometric
Scientific ARES rheometer equipped with a force convection oven in the
parallel plate geometry.

Parallel plate geometry with diameter of 8 mm

and gap of 0.5 mm was used for all measurements obtained in this stud y.
Rheological behavior of all samples was investigated using a series of
isothermal, small-strain oscillatory shear with oscillatory frequency

90

ranging from 100 to 0.1 radian/second and oscillatory shear strain
amplitude of 5%.

To examine the validit y of the time-temperature

superposition principle and the effect of POSS addition on the values of
time-temperature shift factor, sample was tested at temperatures ranging
from T g +10 o C to T g +100 o C with 10 o C intervals.
modulus,

G’(ω),

loss

modulus,

G”(ω),

and

Values of storage
damping

factor,

tanδ( ω) = G”(ω)/G’(ω) , were determined using software provided b y
Rheometric Scientific.
conditions

were

reproducibilit y.

Samples tested under the same experimental

repeated

two

to

three

times

to

confirm

their

Master curves for G’, G” and tanδ were also obtained

with software package provided by Rheometric Scientific and a reference
temperature of T r e f =150 o C for pure pol ymers and blends was used. These
data was also used to see the differences in the pol ymer relaxation times,
which is the inverse of the frequency corresponding to the intersection of
the storage modulus and the loss modulus. Furthermore, the plateau
storage modulus at high frequencies was observed for all of the
nanocomposites, demonstrating the effect of POSS on the entanglement
molecular weight (Me) of the PS.
4.3 RESULTS & DISCUSSION
The effect of POSS loading and distribution of POSS macromers in
varying molecular weights of PS is understood through the rheological
studies. Rheology experiments were performed to address the effect of
POSS macromers on the chain dynamics of PS.

91

Since rheological

measurements and their correlation to pol ymer chain dynamics using
narrow molecular weight PS properties were well documented, our
measurements of POSS/PS blends will provide insight into how the POSS
macromers affect the pol ymer chain dynamics.
The melt rheological experiments were carried out using smallstrain amplitude oscillatory shear so to minimize deformation on sample
morphology.

The dynamic properties of the POSS/PS blends were

investigated as a function of frequency at different temperatures. Changes
in GC (ω ), ω C , G 0N and lim ω →0 η * (ω ) were compared for all studied s ystems.
In addition, the validit y of time-temperature superposition was also
verified for all the samples used in this study.
Figure 4.1(a) is the time-temperature master curves shifted to 150 o C
for the PS (Molecular weight 1.6M Daltons) blended with varying
amounts of St 8 T 8 POSS. The corresponding plot of reduced complex
viscosit y versus reduced frequency is shown in Figure 4.1(b).

The

temperature shift factor versus temperature is plotted in Figure 4.1(c).
From the various reduced storage modulus curves shown in Figure 4.1(a),
we observed that these curves shift downwards and towards the right with
increasing amount of St 8 T 8 POSS blended. In addition, the values of the
plateau modulus, G 0N , and the cross-over frequency, ω C , as shown in Fig
4.1, decrease with increasing amount of St 8 T 8 POSS added expect for very
high loading of 30wt%.

These observations correspond to PS chains

becoming less entangled when blended with St 8 T 8 POSS. There are two
92

ways of looking at the individual pol ym er blends from a volumetric point
of view to look at effect on free volume of the blends and secondl y the
viscometric effect arising from the phase behavior in these blends. For the
reall y high molecular weight 1.6M PS POSS blend s ystem, observed is a
difference in the shifting parameter with increasing POSS loadings based
on the shift factor curve 4.1(c), the higher the concentration the lower the
shift factor, which is indicative of increased free volume, at the same time
based on the zero shear viscosit y there seems to a drop in zero shear
viscosities at low volume loadings indicative of the free volume changes
but with weight loadings the zero shear viscosit y tends to go up along
with an increase in the plateau modulus indicative of filler like behavior,
which suggests a two phase behavior.
By decreasing the molecular weight to around 650K as observed in
Fig 4.2, there is an increased effect on the shift factor plot Fig 4.2 (c),
which suggests the free volume effect dominating the volumetric effects
hence a reduction in the zero shear viscosit y at higher weight loadings.
Similar behavior has been observed in the previous chapter with blend
s ystems with 290K PS.
Figures 4.1 to 4.4 (a) is the time-temperature master curves shifted
to 150 o C for the PS, Molecular weight 1.6M, 650K, 130K and 65K
Daltons respectivel y blended with different amount of St 8 T 8 POSS. The
crossover frequency versus POSS weight fraction has not changed but the
the zero-shear viscosit y has had a minimum before filler type phase

93

separation behavior dominates the viscoelastic response of these blends.
The reduced-complex viscosit y versus frequency curves for various St 8 T 8
POSS fractions are shown in Figure 4.1 to Figures 4.4(b).

Data were

shown for the 2%, 6%, 20%, and 30% St 8 T 8 POSS blends with varying
molecular weight of PS on the same plot.
Although there are some similarities to the effect observed in the
case of Ph 8 T 8 /PS blends, however, no s ys temic change on ω c and G 0N were
observed. We believe this observation can be explain based on the effect
of miscibilit y of St 8 T 8 POSS in PS as influenced by a stronger POSSPOSS interaction and the thermo rheological behavior of well-dispersed,
albeit inert, nano-particulate filled pol ymer. As discussed previousl y, the
POSS-POSS association is stronger for St 8 T 8 POSS as compared to Ph 8 T 8
POSS.

Consequently, the stronger POSS-POSS interaction affect the

miscibilit y curve of St 8 T 8 in PS in such a manner that concentration
fluctuations might occur thus leading to a two phase behavior, one POSSrich and the other pol ymer-rich, as shown in the TEM images in Chapter
2. Furthermore, the two-phase morphology remains but shifts horizontall y
with varying pol ymer molecular weight as observed in Fig 4.5.
Based on the Figure we report a minimum in the zero shear
viscosit y

as

the

nanoparticle

concentration

increases

in

pol ymer

composites with spherical nanoparticles. POSS particles give rise to
pol ymer nanocomposites with much larger specific interfacial areas, such
that even a modest perturbation in the vicinit y of the nanoparticles can

94

have significant ramifications on the observed pol ymer properties. The
minimum in the zero shear viscosit y diffusion is inconsistent with a
simple reptation model; for example, the significant changes in D are not
coupled with changes in Me. A possible ph ysical interpretation of the
anisotropic diffusion could involve a disruption of pol ymer reptation near
nanoparticles. Reptation describes the motion of a pol ymer chain along a
confining tube that is defined by constraints or entanglements within a
high molecular weight pol ymer melt. With sufficient thermal energy and
time the entanglements that define the confining tube relax because these
entangled pol ymers have diffused out of their own tubes. Perhaps, in the
vicinit y of nanoparticles, the confining tube for pol ymer reptation is
perturbed such that diffusion becomes locall y anisotropic.. We have
several studies underway that are designed to determine the range of
materials

showing

a

minimum

in

the

diffusion

coefficient

with

nanoparticle concentration and to ascertain the underl ying pol ymer
ph ysics.

95

1

6

10

10

G' (
[Pa]

5

0

10

10

tan_delta (
[]

)

1.6M-Polystyrene - 150C - MasterCurve
1.6M-2%St8T8-Master-150C - MasterCurve
1.6M-10%St8T8-Master-150C - MasterCurve
1.6M-20%St8T8-Master-150C - MasterCurve
1.6M-30%St8T8-Master-150C - MasterCurve

)

104

-1

-4

10

-3

10

-2

10

-1

10

0

10

1

10

2

10

10
3

10

aTw [rad/s]

Fig 4.1 (a)
Figure 4.1 Time temperature superposed plots of PS (MW= 1.6M Da) and Styrenyl POSS (St8T8) blends with varying weight
fractions of POSS at 1500C a) Storage Modulus (G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT)versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)

96

Figure 4.1 (cont’d)

108
1.6M-Polystyrene - 150C - MasterCurve
1.6M-2%St8T8-Master-150C - MasterCurve
1.6M-10%St8T8-Master-150C - MasterCurve
1.6M-20%St8T8-Master-150C - MasterCurve
1.6M-30%St8T8-Master-150C - MasterCurve

Eta* (
[Pa-s]

)

107

106

105

104

103
10-4

10-3

10-2

10-1

100

aTw [rad/s]

Figure 4.1 (b)

97

101

102

3

10

Figure 4.1 (cont’d)

100

aT (

[]

)

10-1

10-2

[1.6M] : Pure--TTS Shift Factors
[1.6M] : 2% --TTS Shift Factors
[1.6M] : 10%--TTS Shift Factors
[1.6M] : 20%--TTS Shift Factors
[1.6M] : 30%--TTS Shift Factors
-3

10

140.0

150.0

160.0

170.0

180.0

Temp [°C]

Figure 4.1 (c)

98

190.0

200.0

210.0

1

106

10

5

10

)
G' (
[Pa]

4

10

0

10

tan(δ
w) (
[]

PS650K-Pure - MasterCurve
PS650k-2%-Octa Styrenyl POSS - MasterCurve
PS650k-10%-Octa Styrenyl POSS - MasterCurve
PS650k-20%-Octa Styrenyl POSS - MasterCurve
PS650k-30%-Octa Styrenyl POSS - MasterCurve

)

3

10

102
10-4

-3

10

-2

10

-1

0

10

10

1

10

2

10

10-1
103

w [rad/s]
ω

Figure 4.2 (a)
Figure 4.2 Time temperature superposed plots of PS (MW= 650K Da) and Styrenyl
POSS (St8T8) blends with varying weight
0
/
fractions of POSS at 150 C a) Storage Modulus (G ) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT)versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)

99

Figure 4.2 (cont’d)

108
PS650K-Pure - MasterCurve
PS650k-2%-Octa Styrenyl POSS - MasterCurve
PS650k-10%-Octa Styrenyl POSS - MasterCurve
PS650k-20%-Octa Styrenyl POSS - MasterCurve
PS650k-30%-Octa Styrenyl POSS - MasterCurve

7

10

h* (
[Pa-s]

)

6

10

5

10

104

103
10-4

10-3

10-2

10-1

100

w [rad/s]

Figure 4.2 (b)

100

101

102

103

Figure 4.2 (cont’d)

100
[Pg8]:PS650K-Pure--TTS Shift Factors
[Pg8]:PS650k-30%-Octa Styrenyl POSS--TTS Shift Factors
[Pg8]:PS650k-2%-Octa Styrenyl POSS--TTS Shift Factors
[Pg8]:PS650k-10%-Octa Styrenyl POSS--TTS Shift Factors
[Pg8]:PS650k-20%-Octa Styrenyl POSS--TTS Shift Factors

-1

aT (

[]

)

10

-2

10

10-3
150.0

160.0

170.0

180.0

190.0

Temp [°C]

Figure 4.2 (c)

101

200.0

210.0

220.0

107

6

10

106
105
5

104

)

G' (
[Pa]

104

w* (
η
[Pa-s]

)

10

103
103
PS130k-Pure - MasterCurve
PS130k-2%-Octa Styrenyl POSS - MasterCurve
PS130k-6%-Octa Styrenyl POSS - MasterCurve
PS130k-20%-Octa Styrenyl POSS - MasterCurve
PS130k-30%-Octa Styrenyl POSS - MasterCurve

102

1

10

-3

10

-2

10

-1

10

0

1

10

10

2

10

3

10

102
4

10

w [rad/s]
ω

Figure 4.3(a)
Figure 4.3 Time temperature superposed plots of PS (MW= 130K Da) and Styrenyl
POSS (St8T8) blends with varying weight
0
/
fractions of POSS at 150 C a) Storage Modulus (G ) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT) versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)

102

Figure 4.3 (cont’d)

6

10

PS130k-Pure - MasterCurve
PS130k-2%-Octa Styrenyl POSS - MasterCurve
PS130k-6%-Octa Styrenyl POSS - MasterCurve
PS130k-20%-Octa Styrenyl POSS - MasterCurve
PS130k-30%-Octa Styrenyl POSS - MasterCurve

h* (
[Pa-s]

)

105

104

103

102
-3

10

10-2

10-1

100

101

w [rad/s]

Figure 4.3 (b)

103

102

103

4

10

Figure 4.3 (cont’d)

102

[Pg8]:PS130k-Pure--TTS Shift Factors
[Pg8]:PS130k-2%-Octa Styrenyl POSS--TTS Shift Factors
[Pg8]:PS130k-6%-Octa Styrenyl POSS--TTS Shift Factors
[Pg8]:PS130k-20%-Octa Styrenyl POSS--TTS Shift Factors
[Pg8]:PS130k-30%-Octa Styrenyl POSS--TTS Shift Factors

1

0

10

aT (

[]

)

10

-1

10

-2

10

120.0

130.0

140.0

150.0

160.0

Temp [°C]

Figure 4.3 (c)

104

170.0

180.0

190.0

107

102

106

105

G' (
[Pa]

tan(δw) (
[]

)

101
104

103

)
100
2

10

PS 65k-Pure - MasterCurve
PS65k-2%-Octa Styrenyl POSS - MasterCurve
PS65k-6%-Octa Styrenyl POSS - MasterCurve
PS65k-10%-Octa Styrenyl POSS - MasterCurve
PS65k-20%-Octa Styrenyl POSS - MasterCurve
PS65k-30%-Octa Styrenyl POSS - MasterCurve

101

100
10-3

-2

10

-1

10

0

1

10

10

2

10

3

10

10-1
4

10

w [rad/s]
ω

Figure 4.4 (a)
Figure 4.4 Time temperature superposed plots of PS (MW= 65K Da) and Styrenyl
POSS (St8T8) blends with varying weight
fractions of POSS at 1500C a) Storage Modulus (G/) versus Reduced frequency (aTω) b) Reduced viscosity (η∗/ aT) versus Reduced
frequency (aTω) c) Shift factors (aT) versus Temperature (Temp)

105

Figure 4.4 (cont’d)

5

10

4

h* (
[Pa-s]

)

10

103

PS 65k-Pure - MasterCurve
PS65k-2%-Octa Styrenyl POSS - MasterCurve
PS65k-6%-Octa Styrenyl POSS - MasterCurve
PS65k-10%-Octa Styrenyl POSS - MasterCurve
PS65k-20%-Octa Styrenyl POSS - MasterCurve
PS65k-30%-Octa Styrenyl POSS - MasterCurve

2

10

-3

10

-2

10

10-1

1

100

10

w [rad/s]

Figure 4.4 (b)

106

102

3

10

104

Figure 4.4 (cont’d)

2

10

[Pg7]:PS 65k-Pure--TTS Shift Factors
[Pg7]:PS65k-2%-Octa Styrenyl POSS--TTS Shift Factors
[Pg7]:PS65k-6%-Octa Styrenyl POSS--TTS Shift Factors
[Pg7]:PS65k-10%-Octa Styrenyl POSS--TTS Shift Factors
[Pg7]:PS65k-20%-Octa Styrenyl POSS--TTS Shift Factors
[Pg7]:PS65k-30%-Octa Styrenyl POSS--TTS Shift Factors

1

100

aT (

[]

)

10

10-1

10-2
120.0

130.0

140.0

150.0

Temp [°C]

Figure 4.4 (c)

107

160.0

170.0

180.0

4

Normalized Zero Shear Viscosity

3.5

3

2.5

2

65k
129k
290k
650k

1.5

1

0.5

0
0

0.2

0.4

0.6

0.8

1

Vol Fraction of POSS

Figure 4.5 Normalized viscosity (η∗/ aT) versus volume fraction of Styrenyl POSS

4.4. SUMMARY
Based on our results in the previous chapters which indicate the
abilit y for POSS to very locall y interact with the matrix molecules
altering their mobility, and thereby able to either stiffen or plasticize the
material. The focus of this chapter has been to understand the exact
mechanisms which control the dynamics of the pol ymer chains due to the
presence of huge (on a molecular length scale) cage like particles. There
are however, man y questions left due to the absence of a s ystematic stud y.
108

But with higher molecular weight pol ym er chains take a longer time to
relax due to the interference by POSS macromers, also the dynamics of
POSS macromers within the entangled pol ymer chain play a role in
governing the rheological characteristics of these blends. Some of the
viscometric and volumetric properties of POSS-PS blends have been
studied b y varying the molecular weight of the linear nearl y monodisperse
PS ranging from around 30KDa to around 1.6MDa. Since we already knew
from our work in the previous chapter phenethyl POSS acts most likel y as
a plasticizer focus has been primaril y in understanding how st yren yl
POSS behaves with varying chain length. Styrenyl POSS blends to act as a
plasticizer at low weight fractions and also tend to form two phase
s ystems, one pol ymer rich and the other POSS rich, the molecular weight
drives the phase behavior higher or lower depending on the chain length
following classic Flory Huggins relationships. As observed in the case of
st yrenyl blends the presence of the dispersed pol ymer rich phase gives a
unique opportunit y in understanding rheological properties of domains
filled with soft cores, especiall y in nonlinear regimes. This led us in a
direction of exploring the effects of POSS on the non linear mechanical
behavior of POSS polymer blends which is discussed in the next chapter

109

CHAPTER 5
FLOW TRANSITIONS OF POSS – PS BLENDS
5.1 INTRODUCTION
Pol ymer blends

are ubiquitous

in the pol ymer industry; the

characteristic morphology of these blends plays a crucial role in its
applications, since it strongl y affects the mechanical response of the
blend. Typicall y either d ynamic small amplitude oscillatory shear (SAOS)
flow is usuall y adopted to understand their viscoelastic response. Indeed
the d ynamic moduli or in general the master curve can be used to
correlate the blend morphology for a t ypical dispersed filler based s ystem.
But in the case of nanoparticulate blends shear flows are not strong
enough to effectively showcase the differences which originate because of
the size of the dispersed phase. Hence the use of large deformations for
understanding the characteristic response in nanoparticulate blends has
been studied in this chapter, also Dynamic Mechanical Analysis studies
have been performed to get an insight into the effects of POSS on the
d ynamics especiall y at the transition temperatures where one would see an
increased mobilit y of the backbone pol ymer network, previous studies b y
Mulliken and Soong from Boyce and Cohen’s group have seen POSS
effecting the secondary motions of the pol ymeric matrix. Their findings
suggest that POSS clearl y enhanced the mobilit y of the secondary beta
motions, significantly reducing the resistance to high rate elastic and
plastic deformation and also POSS was found to plasticize PVC and hence
110

to dramaticall y influence its glass transition and corresponding non linear
mechanical behavior.
In this chapter the study was based on understanding the effect of
the presence of small nanoparticulate domains under large deformations in
shear. We discuss the effects of the POSS macromers on PS chain under
large deformations.
5.1.1 Large Amplitude Oscillatory Shear (LAOS)
During the oscillation experiment on a rheometer, a sinusoidal
deformation is applied to the material. Within the boundaries of the linear
viscoelasticit y the measured stress at steady state oscillation, preceding
the strain remains sinusoidal. The strain rate, the time derivative of the
strain has a phase offset of 90 o in reference to the strain. This means that
the strain is zero, when the strain rate is at a maximum and vice versa.
In order to characterize the time dependent periodic material response, the
stress can be decomposed into two waves, one in phase with the strain,
one in phase with the strain rate. In analogy to a solid material (strain and
stress are in phase) and a simple fluid (strain rate and stress are in phase),
the part in phase with the strain is referred to as the elastic (solid)
component, the part in phase with the strain rate (out off phase with the
strain) is the viscous (fluid) component. Thus the stress can be expressed
b y the sum of the two components as follows:

(t ) σ '(t ) + σ "(t )
σ=
111

(5.1)

The amplitude of the measured stress is σˆ

and for the two wave

components σˆ ' and σˆ " in figure 5.1.

γ(t

Strain
Strain
rate

.

γ(t
σˆ

Stress
In phase
Out if
phase

Elastic
component σ ’
Viscous
component σ ”
Stress

σ(t)
σ“(t)
σˆ "

σ‘(t)

σˆ '

Strain

Stres

In

Figure 5.1: Stress, Strain and Strain rate waves during an oscillation
measurement. The stress can be decomposed in an elastic and a viscous
stress wave. The Lissajous figure results from plotting the stress versus
the strain.
Instead of representing the stress and strain as sinusoidal waves
versus time, they can be plotted as a function of each other. This
representation is referred to as Lissajous figure. As long as the measured
stress wave is sinusoidal, the Lissajous figure is an ellipsoid. When the
stress is in-phase with the strain, the ellipsoid collapses to a straight line

112

through the origin, in case of the stress being out-off phase with the strain
(phase offset 90 o ), the ellipsoid changes to a circle.
Although the Lissajous figures represent an excellent visual aid for
qualitative anal ysis, they are more difficult to use when a quantitative
evaluation is required. In this case it is more convenient to convert the
time dependent periodic signals and evaluate the data in the Fourier
space. The Fourier transformation decomposes the stress signal in a series
of sinusoidal wave functions with increasing frequency. The results are
represented as a magnitude and phase versus frequency. When the stress
signal is sinusoidal, the Fourier spectrum shows a single peak at the
fundamental frequency. In the linear region (sinusoidal stress response) at
a particular strain deformation, onl y one peak at the excitation frequency
(fundamental) is seen. In the non-linear region the peak at the excitation
frequency (ω= ωo) remains and in addition higher odd harmonics at 3, 5, 7
time the excitation frequency ωo appear. That means, that in order to full y
describe the material behavior in the non-linear region, fundamental and
harmonics magnitude and phase need to be accounted for.
Until recentl y, d ynamic mechanical data were anal yzed onl y when
the conditions of linearit y were fulfilled. Manfred Wilhelm was the first
to s ystematicall y investigate experiments in the non-linear viscoelastic
region. He measured the stress response in the time domain during large
strain

oscillation

experiments

(LAOS)

and

evaluated

frequency spectrum using discrete Fourier transformation.

113

the

complete

At a first glance it might be surprising, that onl y odd harmonics
(3 r d , 5 t h , 7 t h , etc.) show in the Fourier spectrum. This can be easil y shown
b y developing the expression for the non-linear response of the viscosit y
to a sinusoidal shear rate input. In order to describe the non-linearit y, the
viscosit y is approximated by a Taylor expansion with even functions in
the shear rate.

σ (t ) = ηγ
η =ηo + aγ 2 + bγ 4 ...
γ (t) ∝ eiω t
1

(5.2)

With these assumptions the shear stress in the time domain can be
expressed as a sum of odd exponentials, their magnitude and phase as
follows:

σ (t ) =(ηo + aγ 2 + bγ 4 ...)γ
= ηoγ + aγ 3 + bγ 5 ....
=

∞

∑ae

n = odd

− i ( nω1t +ϕn )

n

=
a1 ηo=
a3 a=
a5 b) (5.3)
(n 1:=
; n 3 :=
; n 5 :=

At the onset of the non-linear viscoelasticit y all higher terms become zero
and a1 reduces to the zero shear viscosit y.
The most useful test procedure to anal yze non-linear material
behavior is the oscillation strain sweep and permits to monitor the

114

transition from the linear into the non-linear regime. In the linear region
the material behavior, t ypicall y represented by the storage and loss
modulus G’ and G” are independent of strain. Higher order harmonic
contributions are zero. Beyond the onset of non-linear behavior, along
with changes of the storage and loss modulus, the third harmonic stress
contribution becomes significant.

5.2 EXPERIMENTAL
5.2.1 Thermo-Mechanical Analysis: POSS/PS solutions of varying POSS
concentrations (0%, 6%, 13.3%, 20% and 30%) were made by dissolving
PS (M w = 290K Daltons) and POSS macromers in chloroform and stirring
the solution for 12 hours using a magnetic stirrer.

From this solution

films were cast on a glass surface allowing the solvent to evaporate out
slowl y at room temperature. The films were dried and annealed at 105 o C
overnight in a vacuum oven to remove trace amounts of solvent and
residual stress due to the solvent casting method.
Thermo-mechanical properties of PS/POSS films were anal yzed
using a Rheometric Scientific Solid Anal yzer RSA III operated in a
rectangular compression-tension mode. Samples used were in a shape of
rectangular film strips with dimensions of 12× 3 × 0.2 mm 3 .

The

experiments were performed using an oscillatory frequency of 6.26
radians/second

and

oscillatory

strain

115

amplitude

of

0.1%.

All

measurements were carried out from room temperature to 120 o C for all
PS/POSS blends at a heating rate of 2 o C per minute.
5.2.2 Rheological Characterization: Preparations of POSS/PS blends
were performed b y dissolving PS and POSS macromers in chloroform and
stirring the solution for 12 hours using a magnetic stirrer and crashing the
sample out b y using a bad solvent methanol. The precipitated sample on
crashing out of the solvent is filtered and then dried under vacuum for
more than 12 hours at 60 o C.
Rheological measurements were carried out using a TA Instruments
ARES G2 rheometer equipped with a force convection oven.

Parallel

plate geometry with diameter of 8 mm and varying gap of 0.3 to 0.5 mm
was used for all measurements obtained in this study.

Rheological

behavior of all samples was investigated using a series of isothermal,
large-strain amplitude oscillatory shear with the oscillatory frequency of
6.28radian/second at a temperature of 220C.

Strain sweep experiments

were performed from 0.1% to 100% strain deformation.
Values of storage modulus, G’(ω), loss modulus, G”(ω), and
damping factor, tanδ(ω) = G”(ω)/ G’(ω) , were determined using software
provided b y TA In struments.

Also the harmonic components of the

response sinusoidal wave have been obtained using the software which
was provided with the rheometer. Samples tested under the same
experimental conditions were repeated two to three time to check
reproducibilit y

116

5.3 RESULTS AND DISCUSSION
5.3.1 Glass Transition Region
Thermomechanical propert y of POSS/PS blends in the glassy state
was investigated through dynamic thermo-mechanical techniques.

Small

oscillatory tensile strain at a fixed oscillatory frequency coupled with a
simple temperature ramp was applied on films of PS/POSS blends from
glass y state through glass transition.

The results of the complex elastic

modulus, E * , and the damping factor, tan δ, for pol yst yrene blended with
varying amount of Ph 8 T 8 POSS and St 8 T 8 POSS are depicted in Figures 11
and 12, respectivel y. In Figure 11, the Ph 8 T 8 POSS acting with increasing
amount of POSS added in PS. In addition, we also observed a s ystematic
decrease of the glass y state complex elastic modulus with increasing
weight fractions of POSS as nanoscopic solvent molecule to PS is clearl y
seen as the tanδ peak shifts to lower temperatures due the plasticization
effect.
As depicted in Figure 12, the addition of St 8 T 8 POSS has a ver y
interesting effect on the glass transition of PS.

Although the peak

position of the tanδ curve shifts very little with the addition of St 8 T 8
POSS, the width of glass transition becomes much sharper with increasing
amount St 8 T 8 POSS.

Moreover, the onset of glass transition as

determined b y the complex elastic modulus curve shows a s ystematic
increase as the weight fraction of St 8 T 8 POSS increase.

117

1010

3.5

3.0
9

10

2.5

8

10

tanδ

E* (Pa)

2.0

1.5

7

10

106

PS290K
PS290K
PS290K
PS290K
PS290K

Pure
Ph8T8-6%
Ph8T8-13.3%
Ph8T8-20%
Ph8T8-30%

1.0

0.5

105
20.0

30.0

40.0

50.0

60.0

70.0

80.0

Temperature (

90.0

100.0

110.0

0.0
120.0

o
C)

Figure 5.2 Complex Modulus (E * ) and tanδ versus Temperature plot for PS (M W = 290K Da) and Pheneth yl
POSS (Ph 8 T 8 ) blends with varying weight fractions of POSS

118

The glass transition in amorphous pol ym ers is very complex; there
are man y papers published in the past 25 years 6 8 - 6 9 .

In general, the

d ynamics of glass transition is controlled by two molecular properties:
order and inter-molecular potentials.

The influence of inter-molecular

potentials on the value of glass transition temperature, T g , is obvious.
Higher inter-molecular binding potentials require more energy input to the
material as it undergoes transition from the glass y state to the liquid state,
thus leading to higher value of T g . Similarl y the higher the order in the
material, the higher the energy input are needed to achieve a complete
“disorder” when material undergoes transition from glass-like to liquidlike behavior.

We propose that to accommodate molecularly dispersed,

nanoscopic rigid fillers such as St 8 T 8 POSS, the local pol ymeric segment
may become more “ordered”, which enhances the onset of glass transition
temperature. With this increase in the local order, the relaxation spectrum
of the pol ymer chain in the glass transition region becomes narrower,
which produces a sharper transition. This local order may persist to the
rubbery state, which can facilitate the disentanglement process and leads
to the observed reduction in values of lim ω →0 η * (ω ) with St 8 T 8 POSS
addition.

119

1010

4.0

109
3.0

2.0

tanδ

E* (Pa)

108

107
PS290K
PS290K
PS290K
PS290K
PS290K

6

10

105
20.0

30.0

40.0

Pure
St8T8-6%
St8T8-13.3%
St8T8-20%
St8T8-30%

50.0

1.0

60.0

70.0

80.0

90.0

100.0

110.0

0.0
120.0

Temperature (oC)

Figure 5.3 Complex Modulus (E * ) and tanδ versus Temperature plot for PS (M W = 290K Da) and St yrenyl POSS
(St 8 T 8 ) blends with varying weight fractions of POSS

120

290K PS Strain Sweep at 220C
105

0.03

OscTorque3Ratio (
[]

0.035

G" (
[Pa]

)

0.025

0.02

)

4

OscTorque5Ratio (
[]

)

10

G' (
[Pa]

0.015

0.01

0.005

3

0.0

1

0

102

10

10

)

10

Strain [%]

Stress

Stress
Stress

Strain

Strain
Strain

γ=1%

γ=10%

γ=100%

Figure 5.4: Strain sweep for the viscoelastic material 290K PS, showing
the transition from the linear into the non-linear regime.
121

5.3.2 Non-linear oscillation measurements
For viscoelastic materials such as 290K PS in figure 5.4, both G’
and G” decrease at the onset of non-linear behavior with increasing strain.
The 3 r d harmonic contribution starts to be significant at the same strain,
G’ deviates from linear behavior. The 5 t h harmonic follows the trend of
the 3 r d , onl y less pronounced and at a higher strain. The higher harmonic
contributions are t yp icall y presented as the ratio of the magnitude of the
n t h harmonic and the fundamental stress I n / 1 =I(ω n )/I( ω 1 ), referred to as
the “harmonic ratio” in the following. The harmonic ratio increases
steadil y with increasing strain.
In the non-linear region the s ymmetry of the stress response can be
lost or conserved. For viscoelastic materials such as the reference PS
290K, the s ymmetry is mostl y conserved. The peak of the stress signal in
figure 5.4 flattens; the reduction of the peak maximum translates into
shear thinning behavior (decrease of G’, G”) of viscoelastic materials. It
is worthwhile to compare the shape of the stress wave at different strain
amplitudes in the non-linear region.

122

290K PS 6% Phenethyl POSS Strain Sweep at 220C
105

OscTorque3Ratio (
[]

0.03

0.02

)

G" (
[Pa]

)

0.025

4

0.015

G' (
[Pa]

0.01

OscTorque5Ratio (
[]

)

10

0.005

0.0

1

0

2

10

10

)

103

10

Strain [%]

Stress

Stress

Strain

Strain

Stress
Strain

γ=1%

γ=10%

γ=100%

Figure 5.5: Strain sweep for the viscoelastic material 6% Phe 8 T 8 PS 290K
blend, showing the transition from the linear into the non-linear regime.
123

290K PS 6% Styrenyl POSS Strain Sweep at 220C
5

0.05

0.03

)

104

G' (
[Pa]

0.02

0.01

0.0

101

100

)

103

OscTorque5Ratio (
[]

)

G" (
[Pa]

)

0.04

OscTorque3Ratio (
[]

10

102

Strain [%]

Stress

Stress

Strain

Stress

Strain

Strain

γ=1%

γ=10%

γ=100%

Figure 5.6: Strain sweep for the viscoelastic material 6% St 8 T 8 PS 290K
blend, showing the transition from the linear into the non-linear regime.

124

290K PS 50% Phenethyl POSS Strain Sweep at 220C
4

0.03

OscTorque3Ratio (
[]

10

0.02

)

103

0.015

G' (
[Pa]

0.01

OscTorque5Ratio (
[]

)

G" (
[Pa]

)

0.025

0.005

0.0

1

0

2

10

10

)

102

10

Strain [%]

Stress
Strain

γ=1%

γ=10%

Stress

Stress

Strain

Strain

γ=100%

Figure 5.7: Strain sweep for the viscoelastic material 50% Phe 8 T 8 /PS
blend, showing the transition from the linear into the non-linear regime.

125

290K PS 20% Phenethyl POSS Strain Sweep at 220C
105

OscTorque3Ratio (
[]

0.018
0.016

0.012

0.01

)

104

OscTorque5Ratio (
[]

)

G" (
[Pa]

)

0.014

G' (
[Pa]

0.008

0.006

0.004

0.002

0.0

101

100

)

103

102

Strain [%]

γ=1%

Stress

Stress

Stress

Strain

Strain

Strain

γ=10%

γ=100%

Figure 5.8: Strain sweep for the viscoelastic material 20% Phe 8 T 8 /PS
blend, showing the transition from the linear into the non-linear regime.

126

290K PS 20% Styrenyl POSS Strain Sweep at 220C
5

10

OscTorque3Ratio (
[]

0.016
0.014

G" (
[Pa]

)

0.012

0.01

)

4

10

0.008

)

OscTorque5Ratio (
[]

G' (
[Pa]

0.006

0.004

0.002

3
1

100

10

0.0

)

10

2

10

Strain [%]

Stress

Stress

Stress

Strain

Strain

Strain

Figure 5.9: Strain sweep for the viscoelastic material 20% St 8 T 8 /PS
blend, showing the transition from the linear into the non-linear regime.

127

5.4 SUMMARY
In this chapter we discuss the effects of the POSS macromers on PS
chain under large deformations under flow conditions, also Dynamic
Mechanical Anal ysis studies have been performed to get an insight into
the effects of POSS on the dynamics especiall y at the transition
temperatures, previous studies by Mulliken and Soong from Boyce and
Cohen’s group have seen POSS effecting the secondary motions of the
pol ymeric matrix. Their findings suggest that POSS clearl y enhanced the
mobilit y of the secondary beta motions, significantl y reducing the
resistance to high rate elastic and plastic deformation and also POSS was
found to plasticize PVC and hence to dramaticall y influence its glass
transition and corresponding non linear mechanical behavior.
Thermomechanical properties of POSS/PS blends in the glassy state
were investigated through dynamic thermo-mechanical techniques. Small
oscillatory tensile tests on the Phe 8 T 8 /PS blend films show a gradual
decrease in

glass

transition temperature and elastic modulus with

increasing concentration which further confirm the plasticization theory.
In the case of St 8 T 8 POSS/PS blends not much variation in glass transition
temperature is observed which supports the DSC data, however the width
of glass transition becomes much sharper with increasing amount of St 8 T 8
POSS.

Moreover, the onset of glass transition as determined by the

complex elastic modulus curve shows a systematic increase as the weight
fraction of St 8 T 8 POSS increases. Due to the higher order in the St 8 T 8

128

POSS as observed from DSC, when material undergoes transition from
glass-like to liquid-like behavior we propose to accommodate molecularl y
dispersed, nanoscopic rigid fillers such as St 8 T 8 POSS, the local
pol ymeric segment may become more “ordered”, which enhances the onset
of glass transition temperature.
The strain sweep data especiall y in the non linear regimes on the
various POSS PS blends show a delay in the progression from the linear to
non linear regimes due to the presence of POSS irrespective of the phase
behavior of the blend s ystem, also with increasing weight fractions the
effect is magnified suggesting POSS acting as friction points in between
the free flowing pol ymer chains.
POSS nanoparticles which are compatible with the pol ymer chains
could be used to lower the free energy of the pol ymer mixture and also
reduce the viscosity of a pol ymer mixture as a plasticizer. Though
miscibilit y and aggregation kinetics could be controlled in these t ypes of
nanocomposites since the chemistry of POSS particles can be modified
depending on the s ys tem to be used, it should be noted that restrictions on
the organic pendant group on the silicone cage plays an important role in
governing the d ynamics of s ystems involving POSS.

129

CHAPTER 6
CONCLUSIONS
In recent years the blending of 0-, 1-, and 2- dimensional inorganic
fillers such as POSS, carbon nanotubes, and nano-clay platelets into
pol ymeric

matrices

enabled

exploration
properties.

combinations

on

mechanical

understanding

of

the

connections

of

various

However,

amongst

the

a

filler/matrix
fundamental

microstructure

and

macroscopic properties of pol ymeric based nanocomposites are yet to be
full y explored to optimize the trul y multifunctional propert y potential of
pol ymer nanocomposites.
Several key unifying issues/themes for determining, understanding
and optimizing properties for all classes of pol ymer nanocomposites exist.
A key feature of primary importance in understanding the nanocomposite
is the actual identification and definition of the structure and the
properties of the nanoparticle itself; whether the nanoparticle be OD, ID
or 2D, it is critical to provide a well defined description of the particl e
geometry, structure and properties and its surrounding zone of influence
on the matrix morphology and properties and its interface with the matrix
in order to assess the impact of the nanoparticle on the nanocomposite
propert y. Similar developments addressed POSS "particles" which are
observed to have very different mechanical properties if acting as isolated
entities or as crystallized (or near crystallized) rafts - a basic , but
somewhat different, observation also seen in nanoclay particles when
130

comparing intercalated particle structure/properties with that of the
exfoliated condition. Furthermore, POSS particles were found to have the
abilit y to act as either stiffening agents or as plasticizing agents
depending upon the interactions between the functional groups and the
matrix pol ymer. Another key issue was the development of synthesis and
processing routes to (I) properl y disperse the particles within the matrix
in order to full y make use of their nanoscopic potential and (2) to provide
a good interaction with the matrix in order to facilitate load transfer from
the matrix to the particle in order to exploit the stiffening and
strengthening potential of the nanoparticle. A third key issue is proper
identification of the effect of the nanoparticle on the surrounding matrix
where use of microscop y and mechanical spectroscopy prove enlightenin g
on identifying the effects on the structure and the local dynamics of the
pol ymer molecules. Details specific to these key issues and others for
each material s ystem are presented below where we note that each effect
manifests itself differentl y in each of the nanocomposite material s ystems.
In Chapter 2, the focus was primaril y to investigate the influence of
the various organic groups on the interactions between POSS and pol ymer
which would govern the properties of the POSS-pol ymer blends. The
degree of dispersion was investigated using DSC, WAXD and TEM. Even
though POSS can be attached onto the backbone of the pol ymers, in this
stud y we use onl y full y condensed POSS s ystems where by altering the
organic functional groups appended to Si-O core the different extent of

131

compatibilit y can be controlled. The focus of this study is to investigate
the influence of the various organic groups on the interactions between
POSS and pol ymer which would govern the properties of the POSSpol ymer blends. Samples were prepared using a common solvent and then
crashing in a bad solvent and the resulting blend properties characterized
b y rheology and d ynamical thermal anal ysis.
The architectures of POSS such as size of the inorganic core,
pendant organic group and the shape of the POSS can all play some role
in

governing

the

microstructure

of

the

blend.

Morphological

investigations of the POSS blends indicate that there is a significant
amount of aggregation (crystallization) of POSS into large particles. In
some cases what we found was by controlling the compatibilit y between
the organic group surrounding inorganic silica-like core of POSS and
pol ymer matrix is that the molecular dispersion can be achieved.
In Chapter 3 the effect of concentration on the viscometric
properties

of

the

nanocomposite

blends

have

been

investigated.

Rheological measurements well above Tg have been used to determine the
time-temperature superposition behavior of these thermorheologicall y
simple materials.

The size of POSS macromers on similar length scales

when compared to the entanglement spacing of the pol ymer chains makes
it interesting to stud y the role these nano-structured constituents would
play in controlling the pol ymer chain dynamics. However due to the
complex morphologies of various POSS/PS blends, similar rheological

132

behavior may be obtained due to two different mechanisms. To address
the above issue we took a POSS macromer having the organic pendant
group as the same or nearl y similar as the monomer unit as the pol ymer.
DSC, X-ray, TEM were used to understand the morphology of the blends.
Rheological characterization was done for understanding the chain
d ynamics due to the presence of POSS or due to the domains formed due
to the presence of POSS.
The shape of the POSS macromers may play an important role in
governing

the

microstructure

in

the

blend.

Do

these

changes

in

microstructure cause profound changes to the relaxation of the pol ymer
backbones? Our findings clearl y suggest that POSS nanoparticles which
are compatible with the pol ymer chains could be used to lower the free
energy of the pol ymer mixture and also reduce the viscosit y of a pol ymer
mixture as a plasticizer. Though miscibilit y and aggregation kinetics
could be controlled in these t ypes of nanocomposites since the chemistry
of POSS particles can be modified depending on the s ystem to be used, it
should be noted that restrictions on the organic pendant group on the
silicone cage

plays an important role in governing the dynamics of

s ystems involving POSS. Interestingl y in the case of Octa styrenyl POSS
PS Blends, at small volume fractions, free (untethered) POSS nanodispersed in the melt appears to act as a plasticizer, reducing the
magnitude of selected material properties such as plateau modulus or
zero-shear rate viscosit y at a given temperature, while simultaneousl y

133

causing a decrease in free volume (estimated from WLF coefficient
anal ysis). As the volume of dispersed POSS increases, steric effects of the
filler dominate and the rheological properties of the melt (modulus,
relaxation time, viscosit y) are all enhanced.
In Chapter 4, based on our results in the previous chapters which
indicate the abilit y for POSS to very locall y interact with the matrix
molecules altering their mobilit y, and thereby able to either stiffen or
plasticize the material. In a confined s ystem like an entangled pol ymer
melt blended with POSS, the motion of either the POSS macromer or the
pol ymer chain or both should in some way be affected. Though this
problem is studied previousl y, several hypothesizes have arised but none
conclusive enough to explain what exact mechanisms control the dynamics
of the pol ymer chains due to the presence of huge (on a molecular length
scale) cage like particles. There are however, many questions left due to
the absence of a systematic study. Will the higher molecular weight
pol ymer chains take a longer time to relax due to the interference b y
POSS macromers? Or the dynamics of POSS macromers within the
entangled pol ymer chain play a role in governing the rheological
characteristics of these blends. We intend to address this problem through
the stud y of POSS-PS blends by varyi ng the molecular weight of the
linear nearl y monodisperse PS ranging from around 30KDa to around
1.6MDa. As observed in the case of st yrenyl blends the presence of the
dispersed pol ymer rich phase gives a unique opportunit y in understanding

134

rheological properties of domains filled with soft cores, especiall y in
nonlinear regimes. This led us in a direction of exploring the effects of
POSS on the non linear mechanical behavior of POSS pol ymer blends.
In Chapter 5 we discuss the effects of the POSS macromers on PS
chain previous under large deformations, also Dynamic Mechanical
Anal ysis studies have been performed to get an insight into the effects of
POSS on the d ynamics especiall y at the transition temperatures, previous
studies b y Mulliken and Soong from Boyce and Cohen’s group have seen
POSS effecting the secondary motions of the pol ymeric matrix. Their
findings suggest that POSS clearl y enhanced the mobilit y of the secondar y
beta motions, significantl y reducing the resistance to high rate elastic and
plastic deformation and also POSS was found to plasticize PVC and hence
to dramaticall y influence its glass transition and corresponding non linear
mechanical behavior.
In this stud y we have investigated experimentall y the effect of
POSS macromers when physicall y blended in entangled PS matrix.

The

effect on pol ymer chain dynamics and the phase characteristics by the
presence of POSS macromers in PS blends were discussed.
Using DSC, we were able to determine the melting temperature and
the heat of fusion for the two POSS molecules, St yrenyl 8 -POSS (St 8 T 8 )
and Pheneth yl 8 -POSS (Ph 8 T 8 ) used in this study.

The hydrogenation of

st yren yl to pheneth yl reduces the melting temperature from 274 o C (St 8 T 8
POSS) to 77 o C (Ph 8 T 8 POSS), while the heat of fusion remain unchanged

135

(37.7 kJ/mole). In addition, T g of all POSS/PS blends used in this stud y
was also determined using DSC. No POSS melting peak was observed in
all POSS/PS blends used, which suggests the high miscibilit y between
pol yst yrene and the two POSS molecules used here. The morphologies of
the blends were studied by using wide angle X-ray diffraction techniques
and TEM.

Similar to the DSC observation, X-ray results show the

disappearance of the crystalline nature of POSS when blended with PS
indicating good compatibilit y.

From the TEM images it was clearl y

observed that dispersion occurs for all weight fractions of POSS in PS.
Different phase behavior was achieved depending on the t ype of POSS
used, for Ph 8 T 8 POSS a homogeneous blend is formed because of the
lower POSS-POSS interaction, but phase inversion occurs when solid
St 8 T 8 POSS macromers were used which have a high binding energy.
Rheological

experiments

were

performed

to

investigate

the

influence of POSS addition on the melt dynamics of the polymer chains.
We are particularl y interested in the influence of well-dispersed, nanoparticulate additions on the zero-shear viscosit y, lim ω →0 η * (ω ) , plateau
modulus, G 0N and the cross-over frequency, ω C of pol ymer melts.
In the Ph 8 T 8 POSS/PS blends, because of the low POSS-POSS
interaction, an y POSS interactions are screen-out by the presence of PS.
Ph 8 T 8 POSS/PS blend forms a homogeneous solution.

The rheological

responses of these blends are the same as if as a good solvent was added
to PS.

We observed a s ystemic decrease in the zero-shear viscosit y,

136

plateau modulus, cross-over frequency as the concentration of Ph 8 T 8
POSS increases, as shown in Table 1.
For

St 8 T 8

POSS/PS

blends,

due

to

a

stronger

POSS-POSS

association, the blend forms a two-phase morphology. Regardless to the
concentration of St 8 T 8 POSS in PS, the POSS-rich form a continuous
phase and PS-rich domain as the discontinuous phase. As we increase the
concentration of St 8 T 8 POSS, onl y the fraction of PS-rich discontinue
phase is increased. Since St 8 T 8 POSS is molecularl y dispersed in PS, the
zero shear viscosit y of the blends decreases with the addition of St 8 T 8
POSS which is similar to that observed in the case of Ph 8 T 8 /PS blends
(shown in Table 1).

However due to persistence of the two-phase

morphology, no s ystematic changes in ω c and G 0N was observed with
increasing concentration of POSS since phase behavior controls the
rheological behavior of these blends.
Thermomechanical properties of POSS/PS blends in the glassy state
were investigated through dynamic thermo-mechanical techniques. Small
oscillatory tensile tests on the Ph 8 T 8 /PS blend films show a gradual
decrease in

glass

transition temperature and elastic modulus with

increasing concentration which further confirm the plasticization theory.
In the case of St 8 T 8 POSS/PS blends not much variation in glass transition
temperature is observed which supports the DSC data, however the width
of glass transition becomes much sharper with increasing amount of St 8 T 8
POSS.

Moreover, the onset of glass transition as determined by the

137

complex elastic modulus curve shows a systematic increase as the weight
fraction of St 8 T 8 POSS increases. Due to the higher order in the St 8 T 8
POSS as observed from DSC, when material undergoes transition from
glass-like to liquid-like behavior we propose to accommodate molecularl y
dispersed, nanoscopic rigid fillers such as St 8 T 8 POSS, the local
pol ymeric segment may become more “ordered”, which enhances the onset
of glass transition temperature.

138

REFERENCES

139

REFERENCES
1. J. D. Wright, N. Sommerdijk, Sol-Gel Materials: Their Chemistry and
Biological Properties, Taylor & Francis Group, London,2000.
2. U. Schubert, N. Hüsing, A. Lorenz, Hybrid Inorganic-Organic
Materials b y Sol-Gel Processing of Organofunctional Metal Alkoxides.
Chem. Mater. 1995, 7, 2010–2027.
3. C. J. Brinker, G. W. Scherer, Sol-Gel Science: The Physics and
Chemistry of Sol-Gel Processing, Academic Press, London,1990.
4. Hybrid Materials, Special issue of J. Mater. Chem. 2005, 15, 3543–
3986.
5. P. Gómez-Romero, C. Sanchez (Eds.) Functional Hybrid Materials,
Wiley-VCH, Weinheim, 2004.
6. G. Schottner, Hybrid Sol-Gel-Derived Pol ymers: Applications of
Multifunctional Materials. Chem. Mater. 2001, 13, 3422–3435.
7. G. Kickelbick, U. Schubert, Inorganic Clusters in Organic Pol ymers
and the Use of Pol yfunctional Inorganic Compounds as Pol ymerization
Initiators, Monatsh. Chem.2001, 132, 13.
8. K.-H. Haas, Hybrid inorganic-organic pol ymers based on organicall y
modified Si-alkoxides.Adv. Eng. Mater. 2000, 2, 571–582.
9. C. Sanchez, F. Ribot, B. Lebeau, Molecular design of hybrid organicinorganic nanocomposites s ynthesized via sol-gel chemistry. J. Mater.
Chem. 1999, 9, 35–44.
10.K. G. Sharp, Inorganic/organic hybrid materials. Adv. Mater. 1998, 10,
1243–1248.
11.Nanocomposites Special issue of J. Nanosci. and Nanotech. 2006, 6,
265–572.
12.R. M. Laine, Nanobuilding blocks based on the [OSiO1.5]x (x = 6, 8,
10) octasilsesquioxanes. J. Mater. Chem. 2005, 15, 3725–3744.
13.A. Usuki, N. Hasegawa, M. Kato, Pol ymer-clay nanocomposites. Adv.
Pol ym. Sci. 2005, 179,135–195
14.B. C. Sih, M. O. Wolf, Metal nanoparticle conjugated pol ymer
nanocomposites. Chem.Commun. 2005, 3375–3384.

140

15.C. Sanchez, B. Julian, P. Belleville, M. Popall, Applications of hybrid
organic-inorganic nanocomposites. J. Mater. Chem. 2005, 15, 3559–
3592.
16.P. Gómez-Romero, K. Cuentas-Gallegos, M. Lira-Cantu, N. CasanPastor, Hybrid nanocomposite materials for energy storage and
conversion applications. J. Mater. Sci.2005, 40, 1423–1428.
17.G. Kickelbick, Concepts for Incorporation of Inorganic Building
Blocks into Organic Pol ymers on a Nanoscale. Progr. Pol ym. Sci.2003,
28, 83–114.
18.A. Sayari, S. Hamoudi, Periodic Mesoporous Silica-Based OrganicInorganic Nanocomposite Materials. Chem. Mater. 2001, 13, 3151–
3168.
19.F. Hoffmann, M. Cornelius, J.
Nanotechnol. 2006, 6, 265–288.

Morell,

M.

Fröba,

J.

Nanosci.

20.B. Hatton, K. Landskron, W. Whitnall, D. Perovic, G. A. Ozin, Past,
Present, and Future of Periodic Mesoporous Organosilicas – The
PMOs. Acc. Chem. Res. 2005, 38, 305–312
21.J-L. Shi, Z.-L. Hua, L.-X. Zhang, Nanocomposites from ordered
mesoporous materials. J. Mater. Chem. 2004, 14, 795–806.
22.R. Gangopadh ya y, A. De, Conducting Pol ymer Nanocomposites: A
Brief Overview. Chem. Mater. 2000, 12, 608–622.
23.C. J. Brinker, Y. Lu , A. Sellinger, H. Fan, Evaporation-induced selfassembl y. Nanostructures made eas y. Adv. Mater. 1999, 11, 579–585.
24.N. Hüsing, U. Schubert, Aerogels – airy materials: chemistry,
structure, and properties. Angew. Chem. Int. Ed. 1998, 37, 22–45.
25.D. Y. Godovsk y, Device applications of pol ymer nanocomposites. Adv.
Pol ym. Sci. 2000, 153, 163–205.
26.J. Wen, G. L. Wilkes, Garth L. Organic/Inorganic Hybrid Network
Materials by the Sol-Gel Approach. Chem. Mater. 1996, 8, 1667–1681.
27.G. J. de A. A. Soler-Illia, L. Rozes, M. K. Boggiano, C. Sanchez, C.-O.
Turrin, A.-M. Caminade, J.-P. Majoral, New mesotextured h ybrid
materials made from assemblies of dendrimers and titanium(IV)-oxoorgano clusters. Angew. Chem. Int. Ed. 2000, 39, 4249–4254.
28.J. L. Hedrick, T. Magbitang, E. F. Connor, T. Glauser, W. Volksen, C.
J. Hawker, V. Y. Lee, R. D. Miller, Application of complex
141

macromolecular architectures for advanced microelectronic materials.
Chem. – Europ. J. 2002, 8, 3308–3319.
29.J. W. Kriesel, T. D. Tilley, Carbosilane dendrimers as nanoscopic
building blocks for hybrid organic-inorganic materials and catal yst
supports. Adv. Mater. 2001, 13, 1645–1648.
30.P. R. Dvornic, C. Hartmann-Thompson, S. E. Keinath, E. J. Hill,
Organic-Inorganic Pol yamidoamine (PAMAM) Dendrimer- Pol yhedral
Oligosilsesquioxane (POSS) Nanohybrids. Macromolecules 2004, 37,
7818–7831.
31.R. W. J. Scott, O. M. Wilson, R. M. Crooks, S ynthesis,
Characterization, and Applications of Dendrimer-Encapsulated
Nanoparticles. J. Phys. Chem. B 2005, 109, 692–704.
32.V. Castelvetro, C. De Vita, Cinzia. Nanostructured hybrid materials
from aqueous pol ym er dispersions. Adv. Coll. Interf. Sci. 2004, 108–
109, 167–185.
33.E. Bourgeat-Lami, Organic/inorganic nanocomposite colloids in
Encyclopedia of Nanoscience and Nanotechnology (Ed. H. S. Nalwa),
American Scientific Publishers, Stevenson Ranch 2004.
34.G. Kickelbick, L. M. Liz-Marzán, Core-Shell Nanoparticles in
Encyclopedia of Nanoscience and Nanotechnology (Ed. H. S. Nalwa),
American Scientific Publishers, Stevenson Ranch, 2004.
35.G. A. Ozin, A. C. Arsenault, Nanochemistry – A Chemical Approach to
Nanomaterials, RSC Publishing, London, 2005.
36.Ning, Y.P.; Tang, M.Y.; Jiang, C.Y.; Mark, J.E. and Roth, W.C.; J
Appl. Pol ym. Sci., 1984, 29, 3209
37.Kojima, Y.; Usuki, A.; Kawasumi, M.; Okada, A.; Kurauchi, T. and
Kamigaito, J.; J. Polym. Sci: Part A: Polym. Chem.,1993, 31, 983.
38.Wang, M.S. and Pinnavaia, T.J.; Chem. Mat., 1994, 6, 468.
39.Lichtenhan, J.D.; Comments Inorg. Chem., 1995, 17, 115.
40.Geng H. P.; PhD Thesis, Michigan State Universit y, 2001
41.Scott D.W.; J. Am. Chem. Soc., 1946, 68, 356.
42.Brown J.F.; Vogt L.H.; J. Am. Chem. Soc., 1965, 87, 4313.

142

43. Feher F.J.; Newman D.A.; Walzer J.F.; J. Am. Chem. Soc.,1989, 111,
1741.
44.Feher F.J.; Newman D.A.; J. Am. Chem. Soc.,1989, 112,1931.
45.Lichtenhan J. D.; Schwab J.J.; Feher F.J.; Soulivong D.; U.S. Patent
5942638, 1999.
46.Lichtenhan J. D.; Schwab J.J.; Reinerth W. A. Sr.; Chem. Innovat.,
2001, 1, 3.
47.Voronkov M. G.; Lavrent’yev V. I.; Topics Curr. Chem.,1982, 102,
199.
48.Feher F. J.; Terroba R.; Jin R.; W yndham K. O.; Lucke S.; Brutchey
R.; Nguyen F.; Pol ym. Mater. Sci. Eng, 2001, 82, 301,.
49.Haddad T. S.; Lichtenhan J.D.; Macromolecules, 1996, 29, 7302.
50.Romo-Uribe A.; Mather P. T.; Haddad T. S.; Lichtenhan J.D.; J. of
Pol ym. Sci. Part B: Pol ym. Phys., 1998, 36, 1857.
51.Lichtenhan JD, Otonari Y, Carr MJ. Macromolecules, 1995, 28, 8435.
52.Mather P. T.; Jeon H. G.; Romo-Uribe A.; Haddad T. S.; Lichtenhan
J.D.; Macromolecules, 1999, 32, 1194.
53.Gilman J. W.; Schlitzer D. S.; Lichtenhan J. D.; J. of Appl .Pol ym.
Sci., 1996, 60, 591.
54.Lee A. Y.; Lichtenhan J. D.; Macromolecules, 1998, 31, 4970.
55.Lee A. Y.; Lichtenhan J. D.; J. Appl. Pol ym. Sci., 1999, 73, 1993
56.Hsiao B. S.; Fu B. X.; Mather P. T.; Chaffee K. P.; Jeon H.; White H.;
Rafailovich M.; Lichtenhan J. D.; Schwab J. J.; Pol ym. Mater. Sci.
Eng., 1998, 79, 389.
57.Schwab J. J.; Lichtenhan J.D.; Carr M. J.; Chaffee K. P.; Mather P. T.;
Romo-Uribe A.; Pol ym. Mater. Sci. Eng., 1997, 77, 549.
58.Fu, B. X.; Yang, L.; Somani, R. H.; Zong, S. X.; Hsiao, B. S.; Phillips,
S.; Blanski, R.; Ruth, P.; J. Poly. Sci. Part B: Pol y. Phy., 2001, 39,
2727.

143

59.Li, G.Z., Wang, L.C., Ni, H.L.,
Organomet. Pol ym., 2001, 11, 123.

and Pittman, C.U.;

J.

Inorg.

60.Phillips S. H.; Haddad T. S.; Tomczak S. J.; Curr. Opi. Solid State &
Mat. Sci. 2004, 8, 21.
61.Starr, F.W.; Schroeder, T.B. and Glotzer, S.C.; Phys. Rev. E, 2001, 64,
021802.
62.Starr, F.W.; Schroeder, T.B. and Glotzer, S.C.; Macromolecules, 2002,
35, 4481.
63.Kopesky E. T.; Haddad, T. S.; Cohen, R. E.; McKinley, G. H.;
Macromolecules, 2004, 37, 8992.
64.Blanski, R.L.; Lee, A.; Geng, H.P.; Phillips, S.H.; Chaffee, K. and
Lichtenhan, J.L.; Polym. Preprint, 2001, 41, 585.
65.Baker E. S. Gidden J; Fee D. P.; Kemper P. R.; Anderson S. E.;
Bowers M. T.; Intl. J. Mass Spec, 2003, 227, 205.
66.Rubinstein M. and Colby R.H; “Pol ymer Physics”, Oxford Universit y
Press, 2003.
67.Mackay M. E.; Dao T. T.; Tuteja A.; Ho D. L.; Van Horn B.; Kim
H.C.; Hawker C. J.; Nat. Mat.,2003, 2, 762.
68.Donth R. J.; “Relaxation and Thermodynamics in Pol ym ers-Glass
Transition”, Akademie Verlag, Berlin, 1992.
69.Larson R.G.; Chapter 4 of “The Structure and Rheology of Complex
Fluids”, Oxford Universit y Press, 1999.

144