ICSAM Course Pissis

National Technical
University of Athens
Seminar on Nanomaterials
(Short Course in Polymer Nanocomposites)
Polymer nanocomposites
1. Introduction
2. Synthesis-Processing
3. Properties-Applications
(a) polymer/silica nanocomposites
(b) polymer/clay nanocomposites
(c) polymer/CNT nanocomposites
We do not have the ambition to cover with this seminar the whole area of
nanomaterials, we focus from the beginning on polymer nanocomposites.
After an extended introduction into the topic, we discuss in Part 2 synthesis and
processing of polymer nanocomposites. Processing has been here included in
the title to stress its significance for determining and tuning the final properties of
polymer nanocomposites.
In Part 3 we discuss three families of polymer nanocomposites selected on the
basis of their significance for the field in terms of applications. It is interesting to
note that in these three families the filler is nano in 3, 1 and 2 dimensions,
respectively. With nano we mean here and in the following between 1 and 100
nm.
1
National Technical
Introduction
1. Introduction
2. Properties improvement
3. Effects of interfaces
In this extended introduction we will focus on two points: properties improvement

(with respect to the pure matrix) and the significance of interfacial effects for that
improvement.
2
National Technical
Polymer nanocomposites University of Athens
Introduction
First report on improvement of polymer properties by incorporation of nanofillers

A. Okada, M. Kawasumi, T. Kurauchi, O. Kamigaito, Polym.Prepr. (Am. Chem. Soc.
Div. Polym. Chem.) 28 (1987) 447
(from the Toyota research group)
Dependence of tensile modulus E at 120°C on clay content for organo-

modified montmorillonite and saponite-based nanocomposites
Y. Kojima et al., J. Mater. Res. 6 (1993) 1185-9
The figure shown here is not from the original report but from a later publication
by the same research group.
The striking result in that report and subsequent reports (and in the figure shown
here) is the significant improvement of polymer properties at very low filler
contents, much lower than those required to achieve the same improvement with
conventional (macro- or microscale) composites.
3
National Technical
Introduction
Searching in SCOPUS for polym nanocompo OR polymer inorganic hybrids yields

more than 8.000 hits from journals
Number of published journal articles and issued and pending patents

on nanotubes and nanotube/polymer composites as a function of year
M. Moniruzzaman, K. I. Winey, Macromolecules 39 (2006) 5194-5205
As you know (and this is reflected also in the scientific programme of this
conference – ICSAM 2007) research in the field of polymer nanocomposites is
rapidly growing all over the world.
4
Polymer nanocomposites National Technical
Introduction
Properties Improvement in Nanocomposites (1)

Mechanical properties
silica
Guth & Gold equation:
G = G0(1 + 2.5φ + 14.1φ2)

rubber
stress
G = G0 . X . Y
strain
L. Bokobza, J. P. Chauvin, Polymer 46, 4144-4151, 2005.
What kind of properties we typically seek to improve?

First, mechanical properties.
The figure shows the results of stress-strain measurements in natural
rubber/silica nanocomposites. The filled symbols show results predicted on the
basis of the experimental data for natural rubber and the Guth and Gold equation,
a well established equation for conventional equations (where Go is the modulus
of the pure matrix and fi the filler factor). This equation fails here, it
underestimates the modulus. It is generally accepted that the additional
improvement comes from interfacial effects (to be discussed later). An alternative
to the Guth and Gold equation is then to consider two factors of improvement, X
being the expression in brackets in the “classical” Guth and Gold equation and Y
representing the additional improvement coming from interfacial effects.
5
National Technical
Introduction
Barrier properties
Water permeability in poly(ε-caprolactone)/clay nanocomposites

P. B. Messersmith, E. P. Giannelis, J. Polym. Sci. Part A Polym. Chem. 33 (1995) 1047-57
A second property we would like to improve is related to packaging applications,

namely barrier to diffusion of small molecules. Please note the low filler factors,
permeability is reduced by a factor of 5 on addition of 5 vol% clays.
6
Introduction

Thermal stability
Predicted thermal longevity in polyimide/silica nanocomposites

calculated from thermogravimetric data
V. A. Bershtein et al., J. Polym. Sci. Part B Polym. Phys. 40, 1056-69 (2002)
A third property we would like to improve is thermal stability. Here we show

results for the predicted thermal longevity of polyimide/silica nanocomposites,
calculated on the basis of thermogravimetric data and a model for thermal
degradation. We observe improvement of thermal stability up to about 200 C with
respect to the pure matrix.
7
National Technical
Introduction
Electrical properties
Here we make use of the good properties of the filler
MWCNTs in PC
Evaluation of dc
conductivity in terms
of percolation
σdc(p) ~ (p- pc)t
P. Poetschke et al., Polymer 44 (2003) 5023-30
So far we discussed examples of improved properties of the polymer matrix. In

this example we utilize the good properties of the filler.
In the figure we observe a transition from an insulating material to a conductive
one at a low filler content, when a conducting network of carbon nanotubes
(CNTs) (here multiple walled carbon nanotubes – MWCNTs) is formed.
In the equation coming from percolation theory p is the filler content (actually in
vol%), pc the percolation threshold and t an exponent related with dimensionality.
From percolation theory we know that the percolation threshold for a
homogeneous distribution of spherical inclusions in three dimensions is about 16
vol%. The low percolation threshold of only 1.5 wt% (which means less than 1.5
vol%) here is related to the large aspect ratio of CNTs.
Other properties of nanofillers which are often utilized in polymer nanocomposites
include magnetic and optical ones.
8
National Technical
Introduction

Applications
Timing belt covered with injection molded nylon-6/clay nanocomposites

M. kawasumi, J. Polym. Sci. Part A Polym. Chem. 42 (2004) 819-24
There are already several commercial applications of polymer nanocomposites,

two examples being given here and in the following slide.
9
National Technical
Introduction
Applications
Thermoplastic polyolefine nanocomposites, applications for automotive part

F. Hussain et al., J. Compos. Mater. 40 (2006) 1511-75
The second example, applications for automotive parts.
10
National Technical
Introduction
Size effects (classical)
(a) Size dependence of the melting

points of Sn particles
(b) Size dependence of the normalized
heat of fusion of Sn particles
Classical size effect
Lai et al., Phys. Rev. Lett. 77 (1996) 99
How to calculate the properties of a polymer nanocomposite from those of the

matrix and the filler? We often use classical theories we know from conventional
composites. We have to take, however, some additional points into account. We
will consider only two here: size effects and interfacial effects.
An example of size effect is given here. The reduction in melting point and heat
of fusion with decreasing particle size below about 50 nm is obviously due to the
fact that an increasing number of atoms is on the surface, i.e. more free. This is a
classical size effect.
11
National Technical
Introduction
Synergy
Photoluminescence blue shift versus polymer dielectric constant at a frequency of

60 Hz. PSi-polymer nanocomposites were produced (74% porosity) by diffusion of
the polymers for a period of 2-3 days. Initial peak PL energy before infiltration was
1.63 eV.
H. A. Lopez et al., J. Luminescence 80 (1999) 115-9
An example of quantum size effect is photoluminescence of Si nanocrystals: a

shift to higher energy of luminescence (blue shift) with decreasing crystal size.
In the figure, by filling the pores of nanoporous silica with different polymers we
achieve a fine tuning of blue shift depending on the dielectric constant of the
polymer (which modifies the applied electromagnetic field). This is a good
example of synergy in a composite material.
12
National Technical
Introduction
Effects of interfaces (1)
Polymeric chains close (a few nm) to a solid surface: changes in

z structure (density, chain configuration)
z phase transitions
z molecular mobility
The properties of the interface affect

or dominate over the bulk properties
z new properties or combination of properties interesting for applications

z (fundamental physics) effects close to interfaces become bulk properties
G. C. Papanicolau et al., Coll. Polym. Sci., The concept of the boundary interface
in composite mechanics, Coll. Polym. Sci. 256 (1978) 625-30
G. C. Papanicolau et al., Compos. Interf. 14 (2007) 131-52
Now to interfacial effects, a point we have to consider if we would like to predict

the properties of polymer nanocomposites (see slide 5).
Interfacial effects, arising from the presence of an interfacial layer around the
filler particles with different properties than the bulk (interphase, mesophase), are
significant also for conventional composites. They become more significant for
nanocomposites, due to the larger surface to volume area, as we discuss in the
following.
13
National Technical
Introduction
Schematic drawings of microstructural appearance of typical-particulate vs. fine-

particulate vs. nano-particulate composites based on electronic microscopic
observations: (a) 3 vol% of particles with 10 μm diameter (2.86 particles within a
volume of 50 000 cubic μm); (b) 3 vol% of particles with 1 μm diameter (2860
particles within a volume of 50 000 cubic μm); and (c) 3 vol% of particles with 0.1 μm
(100 nm) diameter (2.86 million particles within a volume of 50 000 cubic μm)
M. Z. Rong et al., Polymer 42 (2001) 3301-4
We consider a volume of a sample of 50.000 cubic microns with 3 vol% of

spherical filler. We have then in that volume about 3 particles if their mean
diameter is 10 microns, about 3.000 particles if their mean diameter is 1 micron,
and about 3.000.000 particles if their mean diameter is 0.1 micron. Is that not
impressiv?
14
Introduction
Calculated interfacial
area per volume of
particles (in 1/nm),
assuming a right –
circular cylindrical
particle shape,
for different particle
diameters and aspect
ratios
M. R. Bockstaller et al.,
Adv. Mater. 17 (2005)
1331-49
Now we allow for the aspect ratio, which is 1 for spherical fillers, to change, i.e.
we consider clays and nanotubes. The calculated interfacial area per volume of
particles becomes very high for clays.
15
National Technical
Introduction
Papakonstantopoulos et al., Phys. Rev. E 72 (2005) 031801

Simulation of local mechanical properties reveals that a glassy layer is formed in the
vicinity of an attractive filler, contributing to the increased strength of the composite
material
Local shear modulus as a function of the distance from the center of the nanoparticle
for the three types of interaction in (a) and for the attractive particle in the melt and
the glassy regime in (b)
No additional comments are necessary here.
16
National Technical
Introduction
CNTs in PVA, Polymer crystallinity (left) and rate of increase of modulus with
volume fraction against rate of increase of crystallinity with volume fraction (right)
Correlation between polymer ordering and reinforcement – the mayor role played
by the CNTs in improving the mechanical properties of composites is to nucleate
an ordered polymer coating
Coleman et al., Polymer 47 (2006) 8556
17
National Technical
Introduction
Computer simulations
Systems investigated: polymer nanocomposites, liquids confined in pores,
liquids close to a surface -> similar results
rough surface smooth surface
Reduction or increase of mobility depending on

z strength of interaction with the surface
z roughness of the surface
Gradual variation of mobility, effect of temperature
Computer simulations are an attractive tool to calculate molecular mobility of a

polymer or a liquid near a solid surface.
Here calculations of the relaxation time as a function of the distance from a rough
and a smooth solid surface. Please note the gradual variation of tau and the
effect of temperature.
18
National Technical
Synthesis-processing
1. Introduction
2. Mixing with preformed particles
3. Sol-gel techniques
4. Using nanobuilding blocks
5. Nanoengineering of composite materials
How to synthesize polymer nanocomposites?

Typically polymer and inorganic filler do not like each other – they are
incompatible. So, the main task is how to render them compatible, i.e. how to
distribute the nanoparticles in the matrix. Typically (however, there are
exceptions!) we would like to have a homogeneous distribution of nanoparticles.
A big effort is made in the field of polymer nanocomposites to optimize and
control the dispersion of nanoparticles – processing is very significant in that
respect.
19
National Technical
Synthesis – processing
Introduction
Two approaches for device fabrication where nanoparticle properties

can be exploited
1. “top-down” approach
photolithography, electron-beam lithography
(microelectronics)
2. “bottom-up” approach
self-assembly processes
control of structural arrangement of nanoparticles and of morphology
Going a step back, there are in general the two above approaches.
The second one is going to be our approach.
20
National Technical
Synthesis-processing University of Athens
Introduction
For a given composition, properties are determined to a large extent by the
final morphology
Different kinds of inorganic-organic composite materials

a) Embedding of the inorganic moiety into the organic polymer
b) Interpenetrating networks (IPNs)
c) Incorporation of inorganic groups by bonding to the polymer backbone
d) Dual inorganic-organic hybrid polymer
G. Kickelbick, Prog. Polym. Sci. 28 (2003) 83-114
Here a rough classification of morphologies in terms of interpenetrating networks.

We use the term hybrid in the case of chemical bonds between the filler and the
polymer. So, C is a hybrid nanocomposite and d a hybrid dual inorganic-organic
polymer.
21
National Technical
Introduction
Three routes for preparing polymer nanocomposites
1. Mixing with preformed particles

2. Sol-gel techniques to generate the particles in-situ
3. Incorporation of nanobuilding blocks (clusters) into the polymer
matrix
In route 1 the main difficulty is to control dispersion of nanoparticles and

morphology. Sol-gel techniques are very good in that respect, however reactions
are often difficult to control and there are problems also with scaling up in
industry. Route 3 is then a compromise.
22
National Technical
Mixing with preformed particles
Mixing in solution (dispersion), i.e. use of a common solvent, which

will be then evaporated, and melt-mixing
Mixing with the monomer followed by in-situ polymerization
Typically monomer and filler do not like each other, so there is a need
for Functionalization (chemical treatment) prior to mixing
Conditions of mixing and processing very significant for the final

properties
23
National Technical
Mixing with inorganic layered 2D materials (natural or synthetic)

This is an example of mixing with preformed particles (here natural or synthetic).

Depending on the type of the layered inorganic material and the polymer and on
the conditions of mixing, different morphologies are obtained. We will come back
to these morphologies later with respect to applications, here it is just an
example.
24
National Technical
Sol-gel techniques
Formation of metal oxide frameworks by sol-gel techniques

Sol-gel techniques were known before for preparing ceramic particles and
powders of high purity and with less pollution to the environment than solid state
chemistry.
The two steps, hydrolysis of metal alkoxides and polycondensation, are
schematically shown here.
25
National Technical
Sol-gel techniques
Simultaneous formation of interpenetrating networks
Three processes in competition: (a) the kinetics of formation of the inorganic
phase, (b) the kinetics of polymerization of the organic phase, (c) the
thermodynamics of phase separation between the two phases
Silicon alkoxides possessing polymerizable alkoxy groups were used and

the alkoxy groups were liberated during the sol-gel processing as the
corresponding alcohol. Aqueous ring-opening metathesis polymerization
(ROMP) as well as free radical polymerization were used for the organic
polymerization reaction.
It is possible to generate the inorganic particles and the polymer simultaneously.

For that we combine the two precursors in one, two examples are given here.
26
National Technical
Sol-gel techniques
Dual network precursors

The network forming functionalities can also be covalently connected to
each other. One way to realize that is with the functional groups for the
organic and inorganic polymerization/polycondensation process being
incorporated in one molecule (commercially available molecules GLYMO
or MEMO, Scheme below). Another possibility is that functional groups for
the second polymer are incorporated in a preformed polymer of the other
type
Here we would like to develop the inorganic and the organic network with
chemical bonds between them.
27
National Technical
Sol-gel techniques
Dual network precursors
Functionalization of a preformed network or polymer with pending

functional groups capable of forming the second network
and use of end-capped polymers
Examples for that are given here.
28
National Technical
Using nanobuilding blocks
Different paths for obtaining hybrid materials. Conventional sol-gel route (path
A), use of templates capable of self-assembly (paths B, D), assembly of
nanobuilding blocks (paths C, D)
C. Sanchez et al., Chem. Mater. 13 (2001) 3061-83
Here different paths for obtaining hybrid materials.

A sol-gel techniques.
B, D using of organic templates (structure directing agent), like block copolymers,
to control morphology, not further discussed here (we will come back to that
later).
C using nanobuilding blocks.
In the following we focus on nanobuilding blocks.
29
National Technical
Surface modification with polymerizable groups
Examples include modification of commonly used inorganic fillers or

particles but also commercially available polyhedral oligomeric
silsequioxanes (POSS) nanoclusters
Here an example of a nanobuilding block: commercially available POSS, which

are compact structures of nm size, typically with organic substituents. We may
replace a part of the organic substituents R by reactive ones, R’, and do
chemistry.
30
National Technical
Surface modification with polymerizable groups
Depending on the number of reactive functional groups available on the

surface linear or crosslikned systems can be prepared
Other examples of surface functionalization are schematically given here.
31
National Technical
Surface modification with initiating groups
Initiating groups attached to the inorganic core allow the grafting of polymers
from the surface of the inorganic moiety. This approach has some advantages
over the more often used “grafting to” technique, where end-functionalized
polymers are grafted to the surface of inorganic particles, e.g. leading to
restrictions in the surface coverage.
Here another possibility of surface modification with functional groups.

The concept of grafting from the surface (by using initiating groups) against the
concept of grafting to the surface. We have a better arrangement of polymer
around the particle in grafting from the surface.
32
National Technical
Schematic representation of the various possible strategies that can be used to

assemble functionalized tin-12 clusters
C. Sanchez et al., Chem. Mater. 13 (2001) 3061-83
Here a summary of different functionalization strategies at the example of tin

clusters.
33
National Technical
Nanoengineering of composite materials

3D control of aggregate shape and size
Aggregate size control using recognition-functionalized block copolymers. a)

Schematic representation and TEM images demonstrating the control of
aggregate core and corona size using different diblock lengths. b)
Comparison between aggregate core size (from TEM) and hydrodynamic
radius (from DLS).
R. Shenhar et al., Adv. Mater. 17 (2005) 657-69
Two slides to nanoengineering. The term refers to tailored-made synthesis using

block copolymers as templates. The particles prefer one of the blocks. The size
of the blocks determines the size of particle aggregates.
34
National Technical
Nanoengineering of composite materials
Schematic representation of the formation of structured nano- or

mesoporous materials by a self-assembly process of block copolymers
Here the second example, starting again with diblock copolymers.

The template can be removed in the last step by heating or by using a solvent.
35
National Technical
Properties-Applications
(a) polymer/silica nanocomposites
1. rubber/silica – mechanical properties
2. polyimide/silica – low-k materials
3. PHEA/silica hydrogels - biomaterials
Now to the first of the three families of polymer nanocomposites which will be
discussed in some detail.
In some of these and in other applications silica can be replaced by other oxides,
such as alumina, titania etc. However, silica is much more used than any other
oxide.
36
National Technical
PDMS/silica nanocomposites
Polydimethylsiloxane (PDMS)
OH OH
| |
– O – Si – O – Si – O -
| |
Silica nanoparticles prepared by sol-gel techniques in the presence of crosslinked PDMS,

diameter ~10 nm
Very good dispersion
Hydrogen bonding between polymer and filler
L. Dewimille, B. Bresson, L. Bokobza, Polymer 46 (2005) 4135-43
Silica particles are here generated by sol-gel techniques in the presence of

crosslinked PDMS. The dispersion is by far much better than in the case of
mixing preformed silica with the monomer and then polymerization and
crosslinking or mixing preformed silica with the polymer and then crosslinking.
37
National Technical
Natural rubber (NR) / silica

nanocomposites
Natural rubber = cis-1,4-polyisoprene
Vulcanization with sulfur
Two series of samples

Series A Series B
after crosslinking before crosslinking
Good dispersion Aggregation
No strong interaction between matrix and filler
L. Bokobza & J. P. Chauvin, Polymer 46, 4144-4151, 2005.
Here a second example, where PDMS has been replaced by natural rubber.
We compare two series of samples with each other. Dispersion of silica
nanoparticles in the polymer matrix is less good if we generate the nanoparticles
before crosslinking. The reason is that in this case there is more space for the
nanoparticles to grow and aggregate.
38
National Technical
Natural rubber (NR) / silica nanocomposites
Stress–strain curves for the unfilled NR and NR films filled with

silica after the cross-linking process.
L. Bokobza & J. P. Chauvin, Polymer 46, 4144-4151, 2005
Other mechanical properties are also improved.

An example is shown here, these materials being investigated with respect to
applications in the tires industry, where carbon black is being replaced by silica.
39
National Technical
Polyimide/silica nanocomposites
• Increasing demand in microelectronics for low-k materials, i.e. materials

with low ε′, to replace SiO2 as inter-metal and inter-layer dielectrics (IMD,
ILD).
• Candidate materials should combine low ε′ values (below 3.0, possible even
below 2.5) with several other good properties.
• Aromatic polyimides (PIs) are interesting materials for such applications.
• However ε′ of the starting material (in the range 3.2-3.5) should be further
reduced (by keeping the other good properties).
• Two possible ways to achieve that is by introducing porosity and by
reinforcing with inorganic nanoparticles.
• Here we combined these two ways: we prepared hybrids of polyimide (PI)
and porous organosilicon nanophase (ON) and investigated their dielectric
properties
D. Fragiadakis et al., J. Phys. Conf. Series 10 (2005) 139-42

V. Yu. Kramarenko et al., Polym. Adv. Techn. 15 (2004) 144-8
No comments are necessary here.
40
National Technical
Preparation – structure Anticipated structure
• From polyamic acid of molar mass

5.000 (series 5) or 10.000 (series
10) or 15.000 (series 15) with
ethoxysilane end groups (PAAS)
and methyl triethoxysilane (MTS).
• Sol-gel techniques
• The PAAS/MTS mass ratio was
systematically varied from 100/0
to 100/120 corresponding to
PI/ON mass fractions varying from
100/0 to 64.4/35.6
The structure has been confirmed by IR spectroscopy.
41
National Technical
Dielectric relaxation spectroscopy (DRS)
3,4 100/8 - 10
3,2 100/60 - 10
ε'
100/70 - 10
3,0
100/80 - 10
-1 1 3 5 7
Frequency dependencies of the r.t.
10 10 10 10 10 dielectric permittivities for the PNs with
frequency (Hz) PAAS/MTS ratios from 100/0
Frequency dependencies of the r.t. dielectric (uppermost curve) to 100/120
permittivities for the samples indicated on (lowermost curve) for series 5.
the plot for series 10.
D. Fragiadakis et al., J. Phys. Conf. Series 10 (2005) 139-42
Measurements were performed at room temperature.

At high frequencies we observe a drop of the dielectric constant, it is due to the
secondary gamma relaxation , arising from the imide cycles.
42
National Technical
polyimide/silica nanocomposites
effective medium theory (EMT) analysis

•Series 5: three EM models to calculate the dielectric permittivity of the ON
•Looyenga model ε′1/3= εm′ 1/3 (1-φ) + εi′ 1/3φ , (1a)
•Lichtenecker model lnε′= lnεm′(1-φ) + lnεi′ 1/3φ , (1b)
•Stoltze et al.′s model ε′a = εm′a (1-φ) + εi′a φ (1c)
Table. Experimental and calculated dielectric

permittivities of nanocomposites
εtot= εtot(εi, εm,...)
EMT has been used to calculate the dielectric constant of the inclusion (silica)
from the measured dielectric constant of the composite (effective dielectric
constant) and the measured dielectric constant of the matrix. This is a way to
calculate, under some assumptions, the dielectric constant of a dispersed phase.
We have chosen a middle frequency (1 kHz) for that calculation, i.e. we used
values measured for the composite and the matrix at 1 kHz. The result is that we
get values for the dielectric constant of ON around 2, different for the different
models (this is obvious) but with the same trend. This is much less than the
dielectric constant of bulk silica, which is around 4. The conclusion is that our ON
generated by sol-gel in the nanocomposites is porous, i.e. a mixture of compact
silica (dielectric constant 4) and air (dielectric constant 1). This result was
confirmed by density measurements not shown here.
43
National Technical
PHEA/silica hydrogels
Introduction: PHEA matrix
► PHEA Hydrogels
PHEA – poly(hydroxyl ethyl) acrylate

.........
Bulk polymerization of 2-hydroxyethyl acrylate

and ethylenglycol dimethacrylate (as crosslinking agent)
9 Good water sorption/diffusion properties

9 PHEA accommodates about 1/4 to 1/3 of its dry mass
as bound (non-freezable) water, it becomes however very soft
Kyritsis et al. J Polym Sci Part B Polym Phys 33 (1995) 1737

Kyritsis et al. Polym Gels Networks 3 (1995) 445
The next example is a hydrogel.

PHEA is an interesting material for applications in medicine and biotechnology:
materials for implantation, scaffolds for tissue engineering, drug delivery systems
etc. For some of these applications (e.g. the first two here) mechanical stability
has to be improved.
44
National Technical
Reinforcement: Silica nanoparticles
► Inclusion of silica nanoparticles
9 The mechanical properties of polymer nanocomposites may be

significantly improved with respect to those of the pure matrix. It
is assumed that the large surface to volume ratio of the
nanoparticles plays a significant role
9 The properties are to a large extent determine by the
morphology (degree of dispersion of the nanoparticles)
Techniques: SEM, TGA, DMA, water sorption-desorption, DRS, TSDC
J. A. Gomez Tejedor et al., J. Polym. Sci. Part B Polym. Phys. 46 (2007) 43-54
J. G. Rodriguez Hernandez et al., Eur. Polym. J. 43 (2007) 2775-83
PHEA/silica hydrogels were studied by several techniques listed here. Some of

the results obtained are also listed here.
45
National Technical
PHEA/silica hydrogels - Preparation
silica tunneled nanoparticles
1. Polymerisation of PHEA
OE t
2. Hydrolysis of TEOS
OH
HO OH
organic polymer chains
3nm
≡ Si − O − C2 H 5 + H 2O → Si − OH + C2 H 5OH domains 30-50 nm

domains < 400 nm
3. Condensation
≡ Si − OH + HO − Si ≡→≡ Si − O − Si ≡ + H 2O
and/or
≡ Si − OH + C2 H 5 − O − Si ≡→≡ Si − O − Si ≡ +C2 H 5OH
f= 0% to 30% in silica
9 SEM: excellent dispersion

9 TGA: Nanoparticles form a continuous network
=> An organic and an inorganic network combined
Polymerization of PHEA and hydrolysis/condensation occur simultaneously.

TEOS is the precursor of silica.
From the results of scanning electron microscopy (SEM) and atomic force
microscopy (AFM) measurements we conclude on the structure shown here.
46
National Technical
Dynamic Mechanical Analysis
10
10
9
10
E´ at 1 Hz
8
10
7 Silica Content
10 0%
15%
6
30%
10
-150 -100 -50 0 50 100 150 200
T (C)
At 1 Hz: glass-rubbery transition at 10-20 C

• mechanical enhancement by 3 o.o.m. compared to pure PHEA!
Dynamic mechanical analysis (DMA) measurements.

In the rubbery phase we observe a significant increase of E’ (the rubbery
plateau). This result confirms the formation of a network of silica nanoparticles,
observed on the same samples by thermogravimetric analysis (TGA).
47
3
-5
2
.1
0
H
d
ilc
S
4
y
r
The results here suggest that the hydrophilicity of the matrix is retained, which can be explained
randomly distributed in the nanocomposite. The steep increase of water content at higher values
48
National Technical
Properties - Applications
(b) polymer/clay nanocomposites
1. Morphology
2. Synthesis
3. Mechanical properties
4. Barrier properties
We move now to the second family of polymer nanocomposites, polymer/clay

nanocomposites.
49
National Technical
polymer/clay University of Athens
morphology
Structure of 2:1 phyllosilicates
S. Sinha Ray, M. Okamoto, Prog. Polym. Sci. 28 (2003) 1539-1641
A typical structure of a layered silicate is shown here: platelets of a thickness of

about 1 nm separated by galleries, where cations are located. They can be
exchanged, e.g. with organic cations, so that the clay, which is originally
hydrophilic, becomes organophilic.
50
National Technical
polymer/clay
morphology
M. Alexandre, P. Dubois, Mater. Sci. Eng. 28 (2000) 1-63
Depending on the method and the conditions of preparation we may have one of
the three distinct morphologies shown here. In practice we have often mixtures of
these morphologies.
51
National Technical
morphology
XRD patterns of: (a) phase separated microcomposite (organo-modified

fluorohectorite in a HDPE matrix); (b) intercalated nanocomposite (same
organomodified fluorohectorite in a PS matrix) and (c) exfoliated
nanocomposite (the same organo-modified fluorohectorite in a silicone
rubber matrix)
M. Alexandre, P. Dubois, Mater. Sci. Eng. 28 (2000) 1-63
How to find out which is our morphology?

Typically with X-ray measurements taking advantage of the crystallinity of the
clays. In the case of immiscibility (phase separation) no change in scattering
angles (distance between the platelets) is observed. In case of intercalation the
peaks shift to lower angles as the distance between the platelets increases.
Finally, in the case of exfoliation the peaks disappear, as no crystal order is
present.
Scanning electron microscopy (SEM) and, in particular transmission electron
microscopy (TEM), provide also information on morphology.
52
National Technical
polymer/clay
Synthesis (solution intercalation)
Schematic drawing of the preparation of polyurethane/clay nanocomposites. The

black circles represent solvent molecules; the lines, PUU chains; and the blocks
a single silicate layer. The relative size of the blocks are for illustrative purposes
only and the relative dimensions do not represent those of a single silicate layer
( ~1 nm thick and having an aspect ratio of ~200-500).
R. Xu et al., J. Biomed. Mater. Res. 64A (2003) 114-9
This is the method of preparation used in the first years after the first report on
polymer/clay nanocomposites appeared (1987).
53
National Technical
polymer/clay
Synthesis (melt intercalation/processing)
Schematic illustration of nanocomposite synthesis (PDMS/MMT)
S. D. Burnside, E. P. Giannelis, Chem. Mater. 7 (1995) 1597-1600
The next milestone was the observation in 1993 that it is possible to melt-mix
polymers and layered silicates.
Here an example of synthesis of crosslinked
polydimethylsiloxane/montmorillonite (PDMS/MMT) nanocomposites by melt
intercalation.
54
National Technical
Synthesis (in situ polymerization)
Nylon-6 nanocomposite formed through in situ polymerization

with 12-aminododecanoic acid (ADA)–MMT
http://www.nanocor.com
An alternative method is polymerization in situ.

The first step consists of rendering the clay organophilic, i. e. the preparation of
the organically modified clay. Surfactants are used to that aim.
The second step consists of intercalation of the monomer, assisted by the
presence of the surfactant. During the third step the monomer is polymerized.
A problem with this method is that the optimal conditions of polymerization are
usually different than for bulk polymerization and have to be found out before.
55
National Technical
polymer/clay
Effect of clay content on tensile modulus, measured at room

temperature, of organomodified montmorillonite/nylon-6-based
nanocomposite obtained by melt intercalation
L. M. Liu et al., J. Appl. Polym. Sci. 71 (1999) 1133-38
In clay nanocomposites we have a large surface to volume ratio, so we expect a

significant improvement of mechanical properties already at low filler factors
(exfoliated morphology). The results confirm that expectation. At higher filler
factors a saturation is observed, probably due to a change of morphology (no
more exfoliation possible).
56
National Technical
Trend of the storage modulus E′ at 25°C for SBS-based

nanocomposites (□□) and microcomposites (▪▪) as a function of the
filler level
(SBS = symmetric styrene-butadiene-styrene block copolymer)
M. Laus et al., J. Mater. Res. 12 (1997) 3134-9
Improvement of mechanical properties depends strongly on on morphology.

Nanocomposite here means that we have intercalation and/or exfoliation.
Microcomposite means that we have a phase separated morphology, see slide
51.
57
National Technical
polymer/clay
Dependence of tensile modulus E at 120°C on clay content for

organo-modified montmorillonite and saponite-based
nanocomposites
Y. Kojima et al., J. Mater. Res. 6 (1993) 1185-9
Improvement depends also on the surface to volume ratio of the clay.
58
National Technical
Variation of elastic modulus as a function of particle content for

extended graphite (EG) and for clay
A. Yasmin et al., Comp. Sci. Technol. 66 (2006) 1182-9
In addition to clays, other layered materials, natural or synthetic, can be used as

fillers with similar or even better results, e.g. extended graphite.
59
National Technical
Barrier properties
Water permeability in poly(ε-caprolactone)/clay nanocomposites
P. B. Messersmith, E. P. Giannelis, J. Polym. Sci. Part A Polym. Chem.

33 (1995) 1047-57
Good barrier properties to small molecules are typical for clay nanocomposites
and form the basis for applications in food packaging.
60
National Technical
polymer/clay
Barrier properties
Oxygen gas permeability of neat PLA and various nanocomposites as a

function of OMLS content measured at 20 °C and 90% relative humidity.
The filled circles represent the experimental data. Theoretical fits based
on Nielsen tortuousity model (J. Macromol. Sci. Chem. A1 (1967) 929-42).
OMLS means organically modified layer silicate.

The fit is based on a model by Nielsen and the fitting parameter is the aspect
ratio of the platelet.
61
National Technical
polymer/clay
Barrier properties
Formation of tortuous path in PLS nanocomposites
PLS means polymer/layered silicate.

The model s schematically presented here. The diffusing molecules can not go
through the platelets, so their path through the nanocomposite becomes longer
and diffusivity and permeability decrease.
62
polymer/clay National Technical
Barrier properties
Relative H2O vapor permeability for PU/clay nanocomposites.

The solid lines represent the theoretical value for aspect
ratios =300 and 1000 (α is the aspect ratio, φ the volume
fraction and μ a geometrical factor.
R. Hu et al., Macromolecules 34 (2001) 337-9
The experimental data indicate a decrease of the aspect ratio with increasing
filler content, which is compatible with a gradual change of morphology from
exfoliated at low filler content to intercalated and phase separated at higher filler
content. Such a change can be confirmed by morphological characterization (X-
rays and/or TEM).
63
National Technical
Barrier properties
Clay length dependence of the relative permeability coefficient for

water in poly(imide)/clay nanocomposites
K. Yano et al., J. Polym. Sci. Part A Polym. Chem. 35 (1997) 2289-94
In agreement with the model we get an improvement of barrier properties with

increasing length of clay.
64
National Technical
Properties - Applications
(c) polymer/CNT nanocomposites
1. Introduction
2. Preparation/processing
3. Electrical properties (percolation)
Now to the last of the three families of polymer nanocomposites. The filler here is
carbon nanotubes (CNT), which is nano in two dimensions (compare slide 1).
65
National Technical
Introduction University of Athens
Number of published journal articles and issued and pending

patents on nanotubes and nanotube/polymer composites as a
function of year
Please note that CNT are intensively investigated also in microelectronics and
other fields of research.
66
National Technical
Introduction
Schematic of a: (a) single-wall and (b) multiwall nanotube
F. Hussain et al., J. Comp. Mater. 40 (2006) 1511-75
No comments.
67
National Technical
Introduction
Schematic diagram showing how a hexagonal sheet of

graphene is “rolled” to form a carbon nanotube. The rolling shown in
the diagram will form a (3,2) nanotube.
No comments.
68
National Technical
Introduction
Tensile strength comparison of common engineering materials
F. Hussain et al., J. Comp. Mater. 40 (2006) 1511-75
CNT possess excellent mechanical properties, the question is how to make use
of these good properties in composite materials and in devices.
69
Preparation/processing National Technical
Preparation/processing methods for the production of polymer/CNT
nanocomposites:
Suspensions of nanotubes in polymer solutions, preparation as thin films
In-situ polymerization in presence of nanotubes
Melt mixing of nanotubes with polymers (advantages: speed, simplicity, availability
in the plastic industry, free of solvents and contaminants)
Main interesting property of CNT for composite application:

→ Electrical conductivity
In order to transfer this property in an insulating matrix (polymer)
formation of a nanotube network required
CNT network should be:

→ free of CNT agglomerates (agglomerates reduce amount
of CNT which can contribute to the network formation,
act as defect in mechanical behavior)
→ good dispersion of single tubes (or small bundles in case
of MWCNT)
→ 3D network with contacts between CNT (within hopping
or tunneling distance of ~ 3-10 nm)
By courtesy of Dr. Petra Poetschke, IPF Dresden (Germany)
From the three methods for preparation of polymer/CNT nanocomposites

given here melt mixing is that most widely used.
70
National Technical
Preparation/processing (Incorporation of CNTs into a University of Athens
polymer by melt mixing)
Starting from a masterbatch Direct incorporation
Highly concentrated batch of polymer Starting from premixtures of
with 15-20 wt% CNTs (MWCNTs) polymer and nanotube powders,
Commercially available , i.e. Hyperion functionalization
Catalysis Intern. Cambridge, USA
Tasks Wetting of CNT by polymer

• surface characteristics
Distribution Dispersion • interfacial tension polymer -CNT
• melt viscosity of polymer
masterbatch CNT
nanocomposite
pure polymer pure polymer
Two different routes are mainly used for preparing polymer/CNT

nanocomposites by melt mixing: (1) starting from a masterbatch and (2)
by direct incorporation. Masterbatches (with typically 15-20 wt% CNT) are
commercially available. They can be diluted with the same or a different
polymer.
71
National Technical
Preparation/processing
Rheological measurements – MA PP/MWCNT series
6
10
5
10
storage modulus G´ [Pa]
4
10
3
10
PP OREVAC PPC
2
10
PP OREVAC PPC, 1,2 wt% MWNT
PP OREVAC PPC, 2,7 wt% MWNT
PP OREVAC PPC, 4 wt% MWNT
1
PP OREVAC PPC, 8 wt% MWNT
10
-2 -1 0 1 2 3
10 10 10 10 10 10
-1
frequency [s ]
Storage modulus (Fig.) and viscosity increase with CNT addition. At

190°C first clear rheological indication of percolation is seen starting at 8
wt% MWNT. At 4 wt% already a small increase in η* and G’ at low
frequencies can be observed. This concentration is much higher than
detected in electrical measurements (already percolated at 2.7 wt%)
Rheological measurements in the melt allow to follow in-situ the morphology, in

particular the dispersion of CNT (and nanoparticles in general). The percolation
here refers to the formation of a continuous network of CNT, which affects
significantly the mechanical properties. In addition it affects also the electrical
properties (see slide 73), there are however differences between the two
percolations. The in-situ monitoring of the changing morphology is essential for
optimizing the processing conditions, see also slide 78.
72
National Technical
Electrical properties University of Athens
PA6/MWCNT nanocomposites
σac(f)=2πfε0ε′′(f)
σ′(ω) = σ(0) + σac(ω) = σdc + Aωs (Universal Dynamic Response)
A. K. Jonscher, Dielectric Relaxation in Solids, Chelsea Dielectric Press, London, 1983
In contrast to the previous slide, the measurements here refer to the final
nanocomposite, e.g. prepared by compression molding. Ac conductivity is
obtained from dielectric measurements, see the equation on the slide. We
observe a transition from non-conductive (insulating) to conductive state. This
transition is related with connectivity of the conducting filler (percolation). The
transition is manifested by the appearance of dc conductivity, i.e. ac conductivity
becomes independent of frequency. The plateau values give the dc conductivity.
The transition occurs here between 2.5 and 5.0 wt% CNTs. With increasing CNT
content dc conductivity increases and the plateau is extended to higher
frequencies. At each composition, the frequency dependence of ac conductivity is
described by the equation given on the slide.
73
National Technical
Percolation
in 3D
R. Zallen, The Physics of Amorphous Solids, Wiley, New York, 1983
This is an experiment to follow percolation in three dimensions. We mix

conducting (metal, Al) and insulating (glass) spheres in a glass cylinder, apply
electrodes (metal, Al) and a voltage and measure the current through the system.
If we do that by systematically increasing the volume fraction of conducting
spheres and mixing well each time, we will get the result that the system
becomes conducting at about 16 vol% conducting spheres. 16 vol% is also the
result we get by applying probability theory. This means that (statistically) the first
conducting way is formed at about this composition. This is the percolation
threshold.
74
National Technical
Percolation
in 2D
R. Zallen, The Physics of Amorphous Solids, Wiley, New York, 1983
This is an experiment in two dimensions. The result we get is that the percolation
threshold is about 50%. We could get the same result with the experiment of the
previous slide by replacing the glass cylinder with a table surface (two-
dimensional distribution of conducting and insulating spheres). We can easily
understand why the percolation threshold is higher in two dimensions: less
chance for the charge carriers to escape if they are confined in two dimensions.
We can also easily predict what is going to be the percolation threshold in one
dimension (a line of spheres), obviously 100%.
75
National Technical
Electrical properties
Percolation
Calculations (site
percolation)
D. Stauffer and A. Aharony, Introduction to Percolation Theory, Taylor & Francis, London, 1992
We can do calculations on a two-dimensional lattice. We have occupied and

empty places (we use then the term site percolation, there is also the alternative
of bond percolation). P is the fraction of occupied sites and s is the number of
neighbor occupied sites connected in a cluster. We have percolation if we get an
infinite cluster.
76
National Technical
Evaluation of dc conductivity in terms of percolation (MWCNTs in PC)
σdc(p) ~ (p- pc)t
P. Poetschke et al., Polymer 44 (2003) 5023-30
Percolation theory results for a nanocomposite with conducting inclusions (like

here CNT) in an insulating matrix (polycarbonate here) in the equation given here
for dc conductivity as a function of the volume fraction p of conducting inclusions.
Pc is the percolation threshold and t a fractional power law exponent depending
on the dimensionality. The figure shows experimental results (for convenience
the weight fraction of CNT is used here to describe the composition). The line is
the fit of the percolation equation to the data. The inset shows the same results in
a different presentation. The percolation threshold is here much lower than the
prediction for a three-dimensional statistical distribution of conducting spheres
(about 16 vol%, slide 74), obviously because of the shape of CNT (needles).
77
National Technical
DRS Results – PP/CNT series

Effects (optimization) of processing
PP+ 5.3 wt% CNT

0
-4
140rpm-210 C-15min
10 0
140rpm-190 C-15min
0
-5
10 70rpm-210 C-15min
0
70rpm-210 C-5min
σ(S/cm)
-6
10 0
70rpm-190 C-15min
0
-7 140rpm-190 C-5min
10
0
-8
70rpm-190 C-5min
10
-2 -1 0 1 2 3 4 5 6
10 10 10 10 10 10 10 10 10
f(Hz)
Here results for ac conductivity for a fixed composition (5.3 wt% CNT in a
polypropylene matrix) and different mixing conditions: speed, temperature and
time of mixing. We use here conductivity as a measure of the quality of
dispersion of CNT to optimize the processing (mixing) conditions.
78
National Technical
Electrical conductivity – effects of CNT dimensions University of Athens
and purity (incorporation of different CNTs into PC)
10
18 MWNT2 (Nanocyl, very thin crude)
MWNT3 (Nanocyl,very thin purified)
Volume resistivity (Ohm cm)
MWNT4 (TsNaMWCnt1, C-nano, China)

15 Nanocyl N 7000
10 (industrial product)
DWNT (Nanocyl, purified)
12 SWNT (CNI, HiPCo)
10
9
10
6
10
3
10
DACA Microcompounder
0
10
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
CNT content in PC (wt%)
• Percolation between 0.35 wt% (SWNT) and 2.0 wt% (MWNT)

PC Iupilon E2000, DACA Microcompounder, mixing under comparable conditions
(Pötschke et al. Fullerenes, Nanotubes, and Carbon Nanostructures 13 (2005), Suppl. 1, 211-224, Pötschke et al. GAK
58 (2005) 1, 45-51)
This slide and the next two, as well as slides 70 and 71, were kindly given
to me by Dr. Poetschke, an expert in the field. We follow here the effects
of CNT dimensions and purity on percolation behavior.
79
Electrical conductivity – effects of CNT/polymer National Technical
interactions (incorporation of the same CNT into
different polymers)
(Pötschke et al. VDI Berichte 1920 (2005) 209)
18
10 • PC = PC Iupilon E 2000
• PP = Moplen 520H
Volume resistivity (Ohm cm)
15
10 • TPU = Elastollan 1185A
12
10 TPU
PP
9
10
6
10
3
10 PC
0
10
0 1 2 3 4 5 6 7 8 9 10 11
MWNT content (wt%)
• Percolation concentration in PC < TPU < PP

Nanocyl thin (10-15 nm) crude MWNT, DACA Microcompounder, 50 rpm, 15 min, T adapted to polymers
Here we follow the effects of CNT/polymer interactions on percolation

behavior. We use three different matrices: polypropylene (PP),
thermoplastic polyurethane (TPU) and polycarbonate (PC). The
differences in percolation behavior reflect differences in the degree
(quality) of dispersion of CNT.
80
National Technical
Electrical conductivity – effects of polymer University of Athens
viscosity (direct incorporation of Nanocyl N 7000
(10 nm) in different polyamide 6) 16
10
(Pegel, Pötschke et al., AP18, CD-ROM,
Volume resistivity (Ohm cm) Technomer 2005, ISBN 3-00-017458-3)
13
10
10
10
7
10
4
10
Capron 8202 low viscous/Nanocyl 7000
Capron B135WP high viscous/Nanocyl 7000
Capron 8202/Nanocyl 3150 (short CNT)
1
10
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
MWNT content (wt%)

• Percolation of MWNT in PA between 3 und 7 wt%,
lower viscous matrix → lower percolation concentration
Here effects of polymer viscosity on percolation behavior. The results

confirm the expectation that the quality of dispersion increases with
decreasing viscosity of the matrix.
Here we are at the end of our excursion in the world of polymer

nanocomposites, many thanks for your attention!
81

ICSAM Course Pissis

Încărcat de

Informații document

Descriere originală:

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

ICSAM Course Pissis

Încărcat de

Drepturi de autor:

Formate disponibile

National Technical

In this extended introduction we will focus on two points: properties improvement

First report on improvement of polymer properties by incorporation of nanofillers

Dependence of tensile modulus E at 120°C on clay content for organo-

Searching in SCOPUS for polym nanocompo OR polymer inorganic hybrids yields

Number of published journal articles and issued and pending patents

Properties Improvement in Nanocomposites (1)

Guth & Gold equation:

G = G0(1 + 2.5φ + 14.1φ2)

L. Bokobza, J. P. Chauvin, Polymer 46, 4144-4151, 2005.

What kind of properties we typically seek to improve?

Water permeability in poly(ε-caprolactone)/clay nanocomposites

A second property we would like to improve is related to packaging applications,

Properties Improvement in Nanocomposites (3)

Predicted thermal longevity in polyimide/silica nanocomposites

A third property we would like to improve is thermal stability. Here we show

P. Poetschke et al., Polymer 44 (2003) 5023-30

So far we discussed examples of improved properties of the polymer matrix. In

Properties Improvement in Nanocomposites (5)

Timing belt covered with injection molded nylon-6/clay nanocomposites

There are already several commercial applications of polymer nanocomposites,

Thermoplastic polyolefine nanocomposites, applications for automotive part

The second example, applications for automotive parts.

(a) Size dependence of the melting

Lai et al., Phys. Rev. Lett. 77 (1996) 99

How to calculate the properties of a polymer nanocomposite from those of the

Photoluminescence blue shift versus polymer dielectric constant at a frequency of

An example of quantum size effect is photoluminescence of Si nanocrystals: a

Polymeric chains close (a few nm) to a solid surface: changes in

The properties of the interface affect

z new properties or combination of properties interesting for applications

Now to interfacial effects, a point we have to consider if we would like to predict

Schematic drawings of microstructural appearance of typical-particulate vs. fine-

We consider a volume of a sample of 50.000 cubic microns with 3 vol% of

Effects of interfaces (4)

Papakonstantopoulos et al., Phys. Rev. E 72 (2005) 031801

No additional comments are necessary here.

No additional comments are necessary here.

rough surface smooth surface

Reduction or increase of mobility depending on

Gradual variation of mobility, effect of temperature

Computer simulations are an attractive tool to calculate molecular mobility of a

How to synthesize polymer nanocomposites?

Two approaches for device fabrication where nanoparticle properties

Different kinds of inorganic-organic composite materials

Here a rough classification of morphologies in terms of interpenetrating networks.

Three routes for preparing polymer nanocomposites

1. Mixing with preformed particles

In route 1 the main difficulty is to control dispersion of nanoparticles and

Mixing with preformed particles

Mixing in solution (dispersion), i.e. use of a common solvent, which

Mixing with the monomer followed by in-situ polymerization

Conditions of mixing and processing very significant for the final

No additional comments are necessary here.

Mixing with inorganic layered 2D materials (natural or synthetic)

This is an example of mixing with preformed particles (here natural or synthetic).

Formation of metal oxide frameworks by sol-gel techniques

Silicon alkoxides possessing polymerizable alkoxy groups were used and

It is possible to generate the inorganic particles and the polymer simultaneously.

Dual network precursors

G. Kickelbick, Prog. Polym. Sci. 28 (2003) 83-114

Dual network precursors

Functionalization of a preformed network or polymer with pending