Full text of DOI:10.1557/mrs2001.93

www.mrs.org/publications/bulletin
dominance of interfacial regions resulting
from the nanoscopic phase dimensions
implies that the behavior of PNCs cannot
be understood by simple scaling argu-
ments that begin with the behavior of tra-
ditional polymer composites.
Three major characteristics define and
form the basis of PNC performance: a nano-
scopically confined matrix polymer, nano-
scale inorganic constituents, and nanoscale
arrangement of these constituents.
Polymer
Nanocomposites:
Status and
The proliferation of internal inorganic–
polymer interfaces means the majority of
polymer chains reside near an inorganic
surface. Since an interface limits the num-
ber of conformations that polymer mole-
cules can adopt, the free energy of the
polymer in this interfacial region is funda-
mentally different from that of a polymer
far removed from the interface (i.e., bulk).
The influence of an interface is related to a
fundamental length scale of the adjacent
matrix, which for polymers is of the order
of the radius of gyration of a chain, Rg
(5–10 nm).7 Thus, in PNCs with only a few
volume percent of dispersed nanoelements,
the entire matrix polymer may be consid-
ered to be a nanoscopically confined inter-
facial polymer. The restrictions in chain
conformations will alter molecular mobil-
ity, relaxation behavior, free volume, and
thermal transitions such as the glass-
transition temperature. More complicated
is the picture for polymers capable of
crystallization (semicrystalline polymers)
or long-range ordering (liquid-crystalline
polymers and block copolymers), where
the interface will alter the degree of order-
ing and packing perfection and thus crys-
tallite and domain growth, structure, and
organization.
The second major characteristic of PNCs
is the dimensions of the added nano-
elements. As with the matrix polymers,
when the dimensions of the cluster or
particle approach the fundamental length
scale of a physical property (the so-called
confinement effect), new mechanical, opti-
cal, and electrical properties arise that are
not present in the macroscopic counter-
part. Examples include superplastic form-
ing of nano-grained ceramics, plasmon
absorption in metal nanoparticles, quan-
tum confinement in semiconductor nano-
particles, and superparamagnetic response
in nanoparticle magnets. Dispersions of
nanoelements exhibiting these unique
properties create bulk materials dominated
by solid-state physics of the nanoscale.
A short list of potential nanoelements
includes layered chalcogenides; metal
nanoparticles; graphitic layers; carbon
nanotubes; metal oxide, nitride, and car-
bide clusters; quantum dots; and biologi-
cal components.
Opportunities
Richard A.Vaia and Emmanuel P. Giannelis
Introduction
Reinforcement of polymers with a sec-
a general understanding has yet to emerge.
For example, the combination of enhanced
modulus, strength, and toughness is a
unique feature of only a fraction of the
PNCs fabricated to date. A major chal-
lenge in developing nanocomposites for
systems ranging from high-performance
to commodity polymers is the lack of even
simple structure–property models. With-
out such models, progress in nanocom-
posites has remained largely empirical.
Similarly, predicting the ultimate material
limits or maximum performances for dif-
ferent classes of nanocomposites is almost
impossible at present.
In this article, we briefly discuss
PNCs and illustrate some successes and
remaining challenges from the perspec-
tive of recent efforts, mainly in our groups,
in layered silicate (clay) polymer nano-
composites.
ond phase, whether inorganic or organic,
to produce a polymer composite is com-
mon in the production of modern plastics.
Polymer nanocomposites (PNCs) represent
a radical alternative to these conventional
polymer composites.1–5 Efforts within the
last 10 years have demonstrated a dou-
bling in tensile modulus and strength
without sacrificing impact resistance for
nylon-layer silicate nanocomposites con-
taining as little as 2 vol% inorganics. In
addition, the heat-distortion temperature
of the nanocomposites can be increased by
up to 100ꢀC, extending the use of the
composite to higher temperature environ-
ments, such as under-the-hood parts in
automobiles.
Besides their improved properties, PNCs
are also easily extruded or molded to near-
final shape, simplifying the manufactur-
ing process. Since high degrees of stiffness
and strength can be realized using far less
high-density inorganic material, PNCs are
much lighter compared to conventional
polymer composites. This weight advan-
tage could have a significant impact on
environmental concerns. For example, it
has been reported that widespread use
of PNCs by U.S. vehicle manufacturers
could save 1.5 billion liters of gasoline
over one year of vehicle production and
reduce related carbon dioxide emissions
by more than 10 billion pounds.6 In addi-
tion, the outstanding combination of bar-
rier and mechanical properties of PNCs
may eliminate the need for a multipoly-
mer layer design in packaging materials,
enabling greater recycling of food and
beverage packaging.
The Nanocomposite Concept
Uniform dispersion of nanoscopically
sized particles (nanoelements) can lead to
an ultralarge interfacial area between the
constituents per volume of material, ap-
proaching 700 m2/cm3 in dispersions of
layered silicates in polymers. This is com-
parable to a football field within a rain-
drop. Figure 1 summarizes the effect and
implication of decreasing the thickness of
a plate—and thus increasing the number
of plates per volume for a given volume
fraction of plates—on the mean distance
between plates. For a 1-nm-thick plate, the
distance between plates approaches 10 nm
at only 7 vol% of plates.
This immense internal interfacial area
and the nanoscopic dimensions between
nanoelements differentiate PNCs from tra-
ditional composites and filled plastics. The
Even though significant progress has
been made in developing different PNCs,
394
MRS BULLETIN/MAY 2001

Polymer Nanocomposites: Status and Opportunities
their own manifold of structure–property
relationships, only indirectly related to their
components and their micro- and macro-
scale composite counterparts. Though nano-
composites with inorganic components of
different dimensionality and chemistry
are possible, efforts have only begun to
uncover the wealth of possibilities for
these new materials.
Synthesis of Polymer Layered
Silicate Nanocomposites
Layered aluminosilicates (alternatively
referred to as 2ꢀ1 layered silicates, phyl-
losilicates, clay minerals, and smectites)
are the most commonly used inorganic
nanoelements in polymer layered silicate
nanocomposites (PLSNs). These layered
silicates possess the same structural char-
acteristics as the well-known minerals talc
and mica and are composed of hydrated
aluminosilicates.8 Their crystal structures
are summarized in Figure 2.
To enhance compatibility with organo-
philic polymers, the hydrophilic alkali
metal and alkaline-earth interlayer cations
are replaced with organic ammonium and
phosphonium cations, functionalizing the
aluminosilicate surface (Figure 3). Depend-
ing on the functionality, packing density,
and length of these modifiers, the organi-
cally modified layered silicate (OLS) may
be engineered to optimize compatibility
with a given polymer. The large majority
of OLSs used for PLSNs contain alkyl am-
monium surfactants. The alkyl chains in
the interlayer adopt conformations rang-
ing from liquid-like to solid-like, depend-
ing on the chain length and exchange
capacity.10,11 The cations modify interlayer
interactions by effectively lowering the
interfacial free energy of the silicate. Fur-
thermore, they can serve to catalyze inter-
layer interactions, initiate polymerizations,
or serve as anchoring points for the matrix
and thereby improve the strength of the
interface between the polymer and in-
organic. Note that in addition to nano-
composites, OLSs are extensively used as
gelling agents, thickeners, and fillers.12
The objective of PLSN fabrication is to
uniformly disperse and distribute the in-
organic, initially composed of aggregates
of stacks of parallel layers, within the
polymer. The final structure, resulting
from the transformation of an initially mi-
croscopically heterogeneous system to a
nanoscopically homogenous system, will
depend on the techniques utilized and the
processing history. If no miscibility exists
between the polymer and the layered sili-
cate, an unintercalated or traditional filled
polymer results. On the other hand,
polymer–OLS compatibility results in sys-
tems exhibiting intercalated or exfoliated
Figure 1. (a) Equilibrium distance between uniformly aligned and dispersed plates of
thickness h at various volume fractions of plates, ꢁ. Increasing the extent of disruption of
the crystallite (h ꢀ 50 nm is ꢀ50 layers/crystallite; h ꢀ 1 nm is individual layers) greatly
decreases the spacing between plates, d, for a given volume fraction of plates. (b) For a
comparable volume fraction of high-aspect-ratio inorganic (l/h ꢂꢂ 1, large dimension, l), the
interfacial region, which is an imperceptible volume fraction of the matrix in a traditional filled
polymer, becomes the dominant bulk phase in a polymer nanocomposite.
Finally, as with any composite, the
arrangement of constituents critically de-
termines the material’s behavior. Con-
ceptually, spatial ordering of spherical,
rodlike, or platelike nanoelements into po-
sitional (one-, two-, or three- dimensional)
arrays with varying degrees of orienta-
tional order will result in a large variety of
systems. The possibilities are further ex-
panded by varying degrees of particle–
particle association, clustering, percolation
(formation of an interconnected network),
and heterogeneous distribution of particles.
The final property of the PNC system will
depend as much on the individual prop-
erties of the constituents as on the relative
arrangement and subsequent synergy be-
tween the constituents.
Ultimately, PNCs offer the possibility of
developing a new class of materials with
MRS BULLETIN/MAY 2001
395

Polymer Nanocomposites: Status and Opportunities
ciated with Brownian motion and shear
alignment of the layers as well as proc-
essing history (melt processing and in situ
polymerization) will contribute to, and
perhaps dominate, the development of a
hierarchical morphology exhibiting nano-
(1–100 nm), meso- (100–500 nm), and
micro- (500–10,000 nm) scale features.
This is especially critical because of the
relatively high viscosity of polymer melts
and particle–particle impingement that
minimizes thermal diffusion of large-aspect-
ratio plates into a spatially homogenous,
equilibrium configuration. Figure 5 shows
representative transmission micrographs
of the various morphologies.
In general, four approaches have been
developed to fabricate PLSNs. In solution
processing, a common solvent or solvent
mixture is used to disperse the OLS as
well as the polymer.16–18 Upon combina-
tion of the two solutions and removal of
the solvent, uniform mixing of polymer
and layered silicate is achieved. The final
structure of the PLSN will depend on the
thermodynamics of the three-or-more-
component system and the rate and extent
of polymer–particle adsorption and chain
bridging between layers during solvent
removal. Recent variations of solution
processing, for example, in mesophase me-
diated processing, utilize a two-phase re-
action medium, such as emulsion or
suspension polymerization, in which the
silicate layers are suspended in an aque-
ous phase, and a monomer is polymerized
in a second phase within the suspen-
sion.19,20 Note that the silicate need not be
organically modified in this approach. The
first commercially successful approach,
pioneered by researchers at Toyota, is in situ
polymerization, whereby the monomer is
used directly as the solubilizing agent for
the OLS.21,22 An interesting variation en-
tails anchoring a living free-radical poly-
merization initiator inside the host galleries,
then polymerizing to prepare polystyrene
nanocomposites.23 Functionalization of
the OLS is chosen to control polymeriza-
tion rates and initiation points so that
layer separation occurs before or during
polymerization. This is especially critical
for thermoset systems, in which the extent
of cross-linking reactions determines the
gel point of the matrix and ultimately
the extent of layer dispersion and final
PLSN morphology.24,25 Finally, in melt proc-
essing, OLS powder and a polymer melt
are mixed statically or under shear.26–28
Polymer–OLS interactions determine the
extent of layer separation and inclusion of
polymer. Shear flows encountered in tradi-
tional polymer melt-processing techniques
facilitate homogenization of the PLSN
morphology as well as global alignment
Figure 2. Crystal structure of 2ꢀ1 layered silicates (smectites).8 Van der Waals interlayer or
gallery containing charge-compensating cations (Mꢃ) separates covalently bonded oxide
layers, 0.96 nm thick, formed by fusing two silica tetrahedral sheets with an edge-shared
octahedral sheet of either alumina or magnesia.8 The charge per unit cell (generally
between 0.5 and 1.3) originates from isomorphous substitution within the silicate layer
(e.g., tetrahedral Si4ꢃ by Al3ꢃ or octahedral Al3ꢃ by Mg2ꢃ). The number of exchangeable
interlayer cations is also referred to as the cation exchange capacity (CEC). This is
generally expressed as milliequivalents/100 g, and ranges between 60 and 120 for relevant
smectites. Variation in the amount, type, and crystallographic origin of the excess layer
charge results in a large family of natural (e.g., montmorillonite, hectorite, saponite) and
synthetic (e.g., laponite, fluorohectorite) layered silicates exhibiting different physical and
chemical characteristics, such as layer size, stacking perfection, reactivity, and Lewis acidity.
For example, the lateral dimension of the layers may range from 20 to 1000 nm.
Figure 3. Cross section of a molecular-dynamics simulation of a dioctadecyl ammonium
exchanged montmorillonite (CEC ꢀ 100 mequiv/100 g, d001 ꢀ 2.4 nm).9 Gray-and-white
chains correspond to molecules associated with the upper and lower surfaces.
morphology. Figure 4 summarizes these
idealized structures. The morphologies are
thought to be thermodynamically stable
and have been predicted by mean field13
and self-consistent field models.14 Further-
more, consideration of positional and ori-
entational correlation between individual
plates leads to higher-order structures,
and the construction of phase diagrams
for polymer–platelet mixtures in the pres-
ence and absence of organic surfactants.15
In practice, however, many PLSN systems
fall between, or exhibit mixtures of, these
idealized structures. Kinetic aspects asso-
396
MRS BULLETIN/MAY 2001

Polymer Nanocomposites: Status and Opportunities
of the layers. This approach is commer-
cially advantageous, especially with regard
to modification of commodity plastics.
ethylene propylene diene monomer,
mers at unexpectedly small inorganic
contents, sometimes as low as 1 wt%. In
general, enhancements of the glassy modu-
lus are 1–2 times for matrices with Tg’s
greater than room temperature. A 5–20-fold
increase in rubbery modulus above Tg
is common and thus an increase in the
heat-distortion temperature is reported
for many systems. The enhancement of
mechanical properties at low volume frac-
tions of OLS reinforces suppositions that
the presence of the layers has a substantial
effect on cooperative chain relaxation and
that the degree of connectivity between
layers determines the stress distribution
throughout the material. However, many
enhancements are system-specific, such as
fracture toughness and strength, and thus
no systematic predictive methodologies
for mechanical properties have been de-
veloped. Furthermore, realization that
global and local orientation of the layered
silicate and meso- and microscale mor-
phologies contribute substantially to these
properties is growing, necessitating more
precise attention to processing history
and morphological characterization on
numerous length scales in the future.
EPDM; and acrylonitrile butadiene rub-
ber, NBR], polyurethanes, poly(ethylene
oxide), poly(vinyl alcohol), polyaniline,
epoxies, and phenolics. Alexandre and
Dubois provide a summary of some of the
polymers and fabrication routes available
in the literature.5 Additionally, a vast re-
source of information is available in the
patent literature.3,4
Polymers and Properties
To date, PLSNs have been fabricated
using an incredibly wide range of poly-
meric resins. An abridged list includes
polystyrene, various polyamides (6, 6-6,
11, 12, aromatic), polyimides, polypropyl-
ene, ethylene vinyl acetate copolymers,
poly(styrene-b-butadiene) copolymers, elas-
tomers [poly(dimethylsiloxane), PDMS;
Dispersion of the layered silicates has
resulted in reproducible enhancements of
mechanical properties of the matrix poly-
Recent processing and rheology studies
have reinforced suspicions that processing
conditions and shear rates are critical to
optimal inorganic morphology develop-
ment and thus the composite properties.
Linear viscoelastic experiments indicate
that pseudo-solid-like behavior in the
melt occurs for intercalated and exfoliated
PLSNs. In addition to stretching of chains
physisorbed and tethered to the silicate
Figure 4. Schematic of the idealized polymer layered silicate nanocomposite (PLSN)
morphologies: (a) intercalated and (b) exfoliated. Intercalated structures are analogous
to traditional inclusion or guest–host compounds and result from polymer penetration
into the interlayer and subsequent expansion to a thermodynamically defined equilibrium
spacing.13,14 The expansion of the interlayer is finite, of the order of 1–4 nm, and is
nominally associated with incorporation of individual polymer chains. In contrast, idealized
exfoliated morphologies consist of individual nanometer-thick silicate layers suspended in
a polymer matrix and result from extensive penetration of the polymer chains within and
delamination of the crystallites. For idealized exfoliated morphologies, the mean layer–layer
separation distance is concentration-dependent.
Figure 5. High-resolution bright-field transmission electron micrographs of (a) an intercalated crystallite of organically modified layered
silicate (OLS) (polystyrene alkyl ammonium montmorillonite, melt-processed), (b) exfoliated system [poly(caprolactam) with 4 wt% OLS,
melt-processed]32 and (c) system with a mixed morphology (Jeffamine D2000/EPON 828 epoxy with 10 wt% Closite 30A). Dark lines are
individual layers of aluminosilicate 1 nm thick, oriented perpendicular to the sample surface.
MRS BULLETIN/MAY 2001
397

Polymer Nanocomposites: Status and Opportunities
surface, long-time behavior is attributed
to the presence of anisotropic silicate par-
ticles randomly oriented and forming a
percolated network structure incapable of
complete relaxation.29,30 Nonlinear visco-
elastic measurements have also shown
that the layers tend to align during shear,
leading to dramatic reductions in viscosity
and elasticity. Recovery and reorientation
upon cessation of shear are not always
complete, and the layers can be thought of
as behaving as non-Brownian attractive
colloidal particles. Complementary stud-
ies have shown that extruder type (single-
versus twin-screw) and screw design
drastically influence the extent of layer
dispersion and thus the properties of melt
processed nylon 6 nanocomposites.27 These
initial studies imply the existence of an
optimal combination of shear rate and
residence time to nanoscopically homoge-
nize layer dispersion. Additionally, these
studies have necessitated further develop-
ment of small-angle scattering models of
nanocomposites to enable real-time exami-
nation of morphology development under
various processing conditions.3
The influence of processing and silicate
dispersion on the polymer structure and
organization also contributes critically to
PLSN mechanical response. Vaia and co-
workers have extensively examined the de-
velopment of polymer crystalline regions
in nylon 6 nanocomposites.31 The presence
of the layered silicate was found not only
to alter the crystal form (from ꢄ to ꢅ) of
polyamide 6, but also to stabilize the re-
sulting ꢄ phase to substantially higher
temperatures than the equilibrium ꢅ phase.
Ito et al. observed that the mechanical
properties, as well as their temperature
dependence, were different for the two
polymer crystal phases.32 The ꢅ phase
exhibited a higher modulus below Tg, but
a more rapid decrease above Tg, than the
ꢄ phase, implying that the ꢄ phase has a
higher heat-distortion temperature. Thus,
an increased fraction of ꢄ phase will lead
to a higher heat-distortion temperature,
irrespective of the contribution of the in-
organic layers to the mechanical response.
Additionally, the layer distribution, size,
and interactions with the polymer were
found to substantially alter the nucleation
rate and growth kinetics of the crystal
phase and thus the lamellae organization
and crystal habit. Optical microscopy in-
dicated that the spherulitic texture of pure
nylon 6 is disrupted by the presence of the
dispersed silicate layers, producing an axi-
alitic texture. Since the fracture behavior
of semicrystalline polymers is intimately
related to the spherulitic texture, the ex-
tent to which the lamella organization is
altered should impact the mechanism of
crack propagation and thus the fracture
properties of these nanocomposites. These
observations imply that the extent to
which mechanical properties are modified
by the OLS may be polymer-system-
specific, and those observed in semicrys-
talline polymers may not be directly
transferable to amorphous polymers.
From a commercial perspective, how-
ever, the value of PLSN technology is
based less on pure mechanical enhance-
ments than on so-called value-added
properties not present in the neat resin but
arising from low volume additions of the
OLS. For example, in contrast to conven-
tional filled resins, which are opaque due
to the high loading and associated scat-
tering from micron-sized particles, low
volume-fraction additions and nanoscopic
dimensions of the dispersed OLS result in
PLSNs with optical transparency, compa-
rable to the base resin. Thermal stability
and fire retardancy through char forma-
tion have motivated investigation of
PLSNs as a component to antiflammabil-
ity additives for commodity polymers.33
Superior barrier properties against gas
and vapor transmission have resulted in
applications for food and beverage pack-
aging34,35 and for barrier liners in storage
tanks and fuel lines for cryogenic fuels in
aerospace systems. Also, depending on the
type of polymeric host, they can display
interesting ionic conductivity for solid-
state electrolytes in batteries36,37 or thermal-
expansion control.38 Recent efforts have
begun to explore use of the nanoelements
as ceramic precursors that react in situ
with aggressive environments to form a
tough ceramic passivation layer on the
polymer surface. Examples of such self-
passivation/self-healing during exposure
to solid-rocket motor exhaust39 and plasma
environments are shown in Figure 6.
Figure 6. Demonstration of PLSN
self-passivation behavior. (a) Scanning
electron micrograph of the surface of
a 5 wt% OLS nylon 6 nanocomposite
after exposure to simulated solid-rocket
motor exhaust.39 (b) Optical micrograph
demonstrating the durability of a nylon
6 nanocomposite to 10 min of oxygen
plasma exposure [concentration ꢀ1018
ions (or radical) per liter], relative to that
of pure nylon 6 (shown on the right).49
an understanding of the characteristics of
this interphase region, its dependence on
nanoelement surface chemistry, the rela-
tive arrangement of constituents, and
ultimately its relationship to the PNC
properties.40 Equally important is the de-
velopment of a general understanding of
the morphology–property relationships
for mechanical, barrier, and thermal re-
sponses of these composite systems. This
necessitates the determination of the criti-
cal length and temporal scales at which
continuum description of a physical proc-
ess must give way to a mesoscopic and
atomistic view of these nanoscale systems.
In general, most investigations to date
have focused on the former issues associ-
ated with chain behavior in the nanoscop-
ically confined interfacial region, which
are briefly discussed below.
Fundamental Investigations
Establishing relationships between the
nanostructures and the ultimate physical
properties of PNCs has just begun. De-
spite the large number of matrices and po-
tential reinforcing materials with different
chemistries, sizes, shapes, and properties,
PNCs share a common feature: the pres-
ence of an interphase polymer layer near
the inorganic surface with dramatically
different properties from the bulk poly-
mer. Because of this interphase layer, the
composite properties are more than the
sum of the constituent bulk properties.
Furthermore, since nanocomposite fabri-
cation entails de-aggregation of nanoscop-
ically associated constituents, elucidation
of polymer-chain mobility and transport
in these environments is critical. Current
pioneering research includes developing
Chain Mobility and PLSN Formation
The ability of the polymer chains to
undergo center-of-mass transport during
398
MRS BULLETIN/MAY 2001

Polymer Nanocomposites: Status and Opportunities
intercalation in a lamellar solid is rather
surprising because the unperturbed ra-
dius of gyration of the polymer is roughly
an order of magnitude greater than the in-
terlayer distance between the silicate lay-
ers (ꢀ2 nm). The kinetics of intercalation
of polystyrene into organically modified
silicates using x-ray diffraction and neu-
tron scattering were examined by Manias
et al.41 Surprisingly, the filling of the sili-
cate galleries with polymer occurs on a
time scale that is faster than in self-
diffusion of the polymer in the bulk, and it
depends on the extent of polymer–silicate
interaction, decreasing as more silicate
surface is exposed or a small fraction of
strong polar groups are incorporated into
the polymer chain. Furthermore, the rate
scales as the inverse of polymer molecular
weight, Nꢆ1. Overall, these studies imply
that the intercalation kinetics are driven
by a strong chemical potential gradient.
Following the description by Valignat
et al.,42 of spreading of ultrathin polymer
films over an inorganic substrate, the ef-
fective diffusion coefficient can be expressed
as the ratio of the free-energy difference of
polymer outside and inside the silicate
galleries, ꢇGi, to the friction coefficient of
polymer motion within the galleries, ꢈ.
When ꢇGi ꢂ kT (an inequality valid for
very moderate polymer/silicate inter-
actions), the intercalation rate can be
greater than in self-diffusion. However, ꢈ
exhibits a stronger dependency on the ex-
tent of interactions between the polymer
and silicate than ꢇGi, eventually leading
to a decrease in the intercalation rate for
the strongest polymer-silicate interactions.
The Nꢆ1 dependence of the effective diffu-
sion coefficient can be rationalized on the
basis of long-lived polymer “bridges” that
act as obstacles to the diffusion of other
chains, much as entanglements act as con-
straints to chain motion.
importance of cooperativity between inter-
layers within a stack.45 They found that a
realistic description of the initial stages of
intercalation requires consideration of a
stack of layers and not an individual slit.
Layer rigidity, layer edge roughness, co-
operativity of polymer flow between ad-
jacent interlayers, bridging of polymer
chains between adjacent and nonadjacent
interlayers, and mass transport of the poly-
mer away from the lateral faces of the
swelling layered stack all contribute to the
rate of chain intercalation and may influ-
ence the molecular-weight dependence.
Figure 7 shows the effect of layer rigidity
on the process of intercalation. These ad-
ditional features are critical to developing
an understanding of how layers exfoliate
and morphologies develop in these systems.
Figure 8. 2H quadrupole spin-echo
experiments for bulk and intercalated 3d
and 5d polystyrene.46 Intensities of less
than one correspond to 2H sites mobile
on the time scale of the spin-echo
experiment. In bulk, ring modes are
enabled somewhat below Tg, while
backbone modes become significant
only near Tg. In intercalated samples,
the two modes are coupled and grow
in over a broad temperature range from
well below Tg.
Confined and Interfacial
Segment Dynamics in PLSNs
In a PLSN, the confined interphase and
intragallery regions will alter the equilib-
rium dynamics of the polymer chains and
their thermal and mechanical responses.
Figure 8 summarizes the results of spin-
echo nuclear magnetic resonance (NMR)
experiments for polystyrene and poly-
styrene nanocomposites46 using poly-
styrene deuterated at the backbone, 3d,
and the ring, 5d. For bulk 3d polystyrene
(i.e., before it has been intercalated), the
intensity of the NMR signal remains con-
stant in the glassy regime, then greatly
decreases above Tg, since backbone dy-
namics, typically absent below Tg, com-
mence at Tg. In the bulk, 180ꢀ flips of the
pendant ring occur below Tg, as indicated
by the slight decrease in intensity below Tg
for the ring-labeled polystyrene, 5d. In
Coarse-grained molecular-dynamics
simulations provide additional insight to
the intercalation processes. Loring and co-
workers43 studied the flow of a polymer
melt (homo- and block copolymers), rep-
resented by the bead-spring model of
Kremer and Grest (KG),44 into a slit, using
coarse-grained molecular-dynamics simu-
lations. The effective diffusion coefficient
(Deff) for the intercalation process was
found to be ꢀ2 times the value for the
self-diffusion coefficient (Dself) in the bulk.
Increasing the affinity of the polymer
chains for the wall resulted in a decrease
in Deff, in accord with experimental re-
sults. However, Deff was nearly independ-
ent of molecular weight, in contrast to the
experiments.
Figure 7. Snapshots of coarse-grained, unified-atom, molecular-dynamics simulations of
intercalation of polymer chains (polymer monomer units depicted as ꢃ; connectivity
between units not shown for clarity) into flexible and stiff sheets. The relative plate stiffness,
size, and intermolecular interactions (plate–plate and plate–polymer) determine the
mechanism of structural evolution [simultaneous multi-gallery diffusion (not shown),
sequential layer-by-layer intercalation (flexible), or layer sliding (stiff)]. For clarity, only a
fraction of the chains interacting with the layers are shown. Simulation details may be found
in Bharadwaj et al.45
Complementing these studies, Bharadwaj
et al. examined, through simulations, the
MRS BULLETIN/MAY 2001
399

Polymer Nanocomposites: Status and Opportunities
contrast, the intercalated polymer shows
significant mobility, evidenced by the loss
of intensity, even at temperatures well
below Tg. Additionally, there is no distinct
change from solidlike to liquidlike behav-
ior, as in the bulk. The broad distribution
of relaxation times, however, suggests that
multiple environments (with liquidlike and
solidlike characteristics) exist for the inter-
calated chains, with a fraction of segments
exhibiting solidlike characteristics even at
temperatures above the bulk polymer Tg.
Complementary dielectric spectroscopy
studies on poly(methylphenyl siloxane)
nanocomposites47 and NMR, and ther-
mally stimulated current investigations of
poly(ethylene oxide) nanocomposites48 in-
dicate that the complex relaxation behav-
ior of intercalated polymer and polymer
near the silicate surface is general and not
polymer-system-specific.
The presence of higher mobility in the
nanocomposites compared with the bulk
is certainly counterintuitive, as “confine-
ment” of the polymer chains within ꢀ2 nm
is expected to increase their solidlike char-
acter and decrease their mobility. Recent
molecular-dynamics simulations provide
insight into this complex dynamic behav-
ior.46,49 When confined to a nanoscale gap
or near a surface, the polymer chains
order into discrete subnanometer layers
(Figure 9), imparting a strong density in-
homogeneity in the direction normal to
the surface. Fast dynamics occur in the
lower-density regions, whereas slower
dynamics occur close to the surface, where
the segment density is high.
around Tg, in the nanocomposites it may
gratefully acknowledges the warm hospi-
tality of E. Economou, C. Fotakis, G. Fytas,
and the entire Polymer Group.
persist well below the bulk Tg. The hetero-
geneity in mobility (faster and slower seg-
ments compared with the bulk polymer)
and the persistence of mobility below the
bulk Tg might be the key to the new set of
properties (i.e., stiffness without loss of
toughness) in nanocomposites. Further-
more, the heterogeneous mobility has
implications with regard to mechanical
optimization of the PLSNs, where the inter-
facial shear strength between layered sili-
cate and polymer matrix is key.
References
1. E.P. Giannelis, Adv. Mater. 8 (1) (1996) p. 29.
2. T.J. Pinnavaia and G.W. Beal, eds., Polymer-
Clay Nanocomposites (John Wiley & Sons, New
York, 2001).
3. R.A. Vaia and R. Krishnamoorti, eds., Polymer
Nanocomposites (American Chemical Society,
Washington, DC, 2001).
4. Nanocomposites 1999: Polymer Technology for
the Next Century (Principia Partners, Exton, PA,
1999).
5. M. Alexandre and P. Dubois, Mater. Sci. Eng.
28 (2000) p. 1.
Conclusions
Polymer nanocomposites, as exempli-
fied by polymer layered silicate nanocom-
posites, offer the possibility of developing
new classes of materials with a unique
manifold of structure–property relation-
ships. Unraveling these (nano)structure–
property relationships is the new frontier
in nanocomposites.
6. Nanocomposites New Low-Cost, High-Strength
Materials for Automotive Parts, ATP Project 97-
02-0047 (National Institute of Standards and
Technology, Gaithersburg, MD, 1997).
7. Y.S. Lipatov, Polymer Reinforcement (ChemTec
Publishing, Toronto, 1995).
8. R.E. Grim, Clay Mineralogy, 2nd ed. (McGraw
Hill, New York, 1968) p. 362.
9. E. Hackett, E. Manias, and E.P. Giannelis,
J. Chem. Phys. 108 (1998) p. 7410.
10. G. Lagaly, Solid State Ionics 22 (1986) p. 43.
11. R.A. Vaia, R.K. Teukolsky, and E.P. Giannelis,
Chem. Mater. 6 (1994) p. 1017; L.-Q. Wang, J. Liu,
G.J. Exarhos, K.Y. Flanigan, and R. Bordia, J. Phys.
Chem. B 104 (2000) p. 2810.
12. S.J. Kemnetz, A.L. Still, C.A. Cody, and R.
Schwindt, J. Coating Technol. 61 (1989) p. 47.
13. R.A. Vaia and E.P. Giannelis, Macromolecules
30 (1997) p. 7990, p. 8000.
14. Y. Lyatskaya and A.C. Balazs, Macromolecules
31 (1998) p. 6676; A.C. Balazs, C. Singh, and E.
Zhulina, Macromolecules 31 (1998) p. 8370; A.C.
Balazs, C. Singh, E. Zhulina, and Y. Lyatskaya,
Acc. Chem. Res. 32 (1999) p. 651.
15. V.V. Ginzburg and A.C. Balazs, Macromole-
cules 32 (1999) p. 5681; V.V. Ginzburg, C. Singh,
and A.C. Balazs, Macromolecules 33 (2000) p. 1089.
16. R. Schöllhorn, Physica B 99 (1980) p. 89.
17. B.K.G. Theng, Formation and Properties of Clay
Polymer Complexes (Elsevier, New York, 1979).
18. K. Yano, A. Usuki, and A. Okada, J. Polym.
Sci., Part A: Polym. Chem. 31 (1993) p. 2493.
19. D.C. Lee and L.W. Jang, J. Appl. Polym. Sci.
61 (1996) p. 1117.
Acknowledgments
Efforts in nanocomposites within our
groups are supported through programs
at the U.S. Air Force Office of Scientific
Research, the Office of Naval Research,
the Army Research Laboratory, the Cor-
nell Center for Materials Research (an
NSF-funded MRSEC), and the Air Force
Research Laboratory’s Materials and
Manufacturing Directorate. We especially
thank our colleagues S. Anastasiadis,
E.J. Kramer, R. Krishnamoorti, E. Manias,
D. Zax, and B. Farmer for the use of fig-
ures. This article was partially written
while E.P. Giannelis was on sabbatical at
IESL/FORTH, Heraklion, Greece, and he
Thus, in contrast to the bulk polymers
where chain mobility slows precipitously
20. S. Bandyopadhyay, A. Hsieh, and E.P.
Giannelis, in Polymer Nanocomposites, edited by
R.A. Vaia and R. Krishnamoorti (American
Chemical Society, Washington, DC, 2001).
21. A. Usuki, M. Kawasumi, Y. Kojima, A.
Okada, T. Kurauchi, and O. Kamigaito, J. Mater.
Res. 8 (1993) p. 1174.
22. Y. Kojima, A. Usuki, M. Kawasumi, A. Okada,
T. Kurauchi, and O. Kamigaito, J. Polym. Sci.,
Part A: Polym. Chem. 31 (1993) p. 983.
23. M.W. Weimer, H. Chen, E.P. Giannelis, and
D.Y. Sogah, J. Am. Chem. Soc. 121 (1999) p. 1615.
24. T. Lan, P.D. Kaviratna, and T.J. Pinnavaia,
Chem. Mater. 7 (1995) p. 2144; T. Lan and T.J.
Pinnavaia, Chem. Mater. 6 (1994) p. 2216.
25. P.B. Messersmith and E.P. Giannelis, Chem.
Mater. 6 (1994) p. 1719.
26. R.A. Vaia, H. Ishi, and E.P. Giannelis, Chem.
Mater. 5 (1993) p. 1694.
27. J.W. Cho and D.R. Paul, Polymer 42 (2001)
p. 1083.
Figure 9. (a) Molecular-dynamics simulation “snapshot” of a silicate-surfactant-polystyrene
nanocomposite, giving an atomistic picture of the various system components; (b) the
corresponding ensemble-averaged, number density of carbon atoms as a function of
distance, offset from the lower confining silicate surface.46,49
400
MRS BULLETIN/MAY 2001

Polymer Nanocomposites: Status and Opportunities
28. S.D. Burnside and E.P. Giannelis, Chem.
Mater. 7 (1995) p. 1597.
29. R. Krishnamoorti and E.P. Giannelis,
Macromolecules 30 (1997) p. 4097.
30. J. Ren, A.S. Silva, and R. Krishnamoorti,
Macromolecules 33 (2000) p. 3739.
31. D.B. Lincoln, R.A. Vaia, Z.-G. Wang, and
B.S. Hsiao, Polymer 42 (2001) p. 1621; D.B.
Lincoln, R.A. Vaia, Z.-G. Wang, B.S. Hsiao, and
R. Krishnamoorti, Polymer (2001) submitted for
publication.
46. D.B. Zax, D.-K. Yang, R.A. Santos, H.
Hegemann, E.P. Giannelis, and E. Manias,
J. Chem. Phys. 112 (2000) p. 2945.
47. S.H. Anastasiadis, K. Karatasos, G. Vlachos,
E. Manias, and E.P. Giannelis, Phys. Rev. Lett. 84
(2000) p. 915.
48. R.A. Vaia, B.B. Sauer, O. Tse, and E.P.
Giannelis, J. Polym. Sci., Part B: Polym. Phys. 35
(1997) p. 59; S. Wong, R.A. Vaia, E.P. Giannelis,
and D.B. Zax, Solid State Ionics 86–88 (1) (1996)
p. 547.
49. E. Hackett, E. Manias, and E.P. Giannelis,
Chem. Mater. 12 (2000) p. 2161; M.R. Nyden and
J. Gilman, Comp. Theor. Polym. Sci. 7 (1997)
p. 191.
ꢀ
32. M. Ito, K. Mizuochi, and T. Kanamoto, Poly-
mer 39 (1998) p. 4593.
online Web access in 2001 with all print subscriptions!
33. J.W. Gilman, Appl. Clay Sci. 15 (1999) p. 31.
34. G.W. Beall, in Polymer-Clay Nanocomposites,
edited by T.J. Pinnavaia and G.W. Beall (John
Wiley & Sons, New York, 2001) p. 267.
35. K. Yano, A. Usuki, and A. Okada, J. Polym.
Sci., Part A: Polym. Chem. 35 (1997) p. 2289.
36. E. Ruiz-Hitzky, P. Aranda, B. Casal, and J.C.
Galvan, Adv. Mater. 7 (1995) p. 180.
37. R.A. Vaia, S. Vasudevan, W. Krawiec, L.G.
Scanlon, and E.P. Giannelis, Adv. Mater. 7 (1995)
p. 154 .
38. Y. Yang, Z.-K. Zhu, J. Yin, X.-Y. Wang, and
Z.-E. Qi, Polymer 40 (1999) p. 4407.
39. R.A. Vaia, G. Price, P.N. Ruth, H.T. Nguyen,
and J. Lichtenhan, Appl. Clay Sci. 15 (1999) p. 67.
40. E.P. Giannelis, R. Krishnamoorti, and E.
Manias, Adv. Prog. Polym. Sci. 138 (1999) p. 107.
41. E. Manias, H. Chen, R. Krishnamoorti,
J. Genzer, E.J. Kramer, and E.P. Giannelis,
Macromolecules 33 (2000) p. 7955.
42. M.P. Valignat, G. Oshanin, S. Villette, A.M.
Cazabat, and M. Morceau, Phys. Rev. Lett. 80
(1998) p. 5337.
43. J.Y. Lee, A.R.C. Baljon, R.F. Loring, and A.Z.
Panagiotopoulos, J. Chem. Phys. 109 (1998)
p. 10321; J.Y. Lee, A.R.C. Baljon, and R.F. Loring,
J. Chem. Phys. 111 (1999) p. 9068, p. 9754.
44. K. Kremer and G.S. Grest, J. Chem. Phys. 97
(1990) p. 5057.
45. R. Bharadwaj, B.F. Farmer, and R.A. Vaia, in
Polymer Nanocomposites, edited by R.A. Vaia
and R. Krishnamoorti (American Chemical So-
ciety, Washington, DC, 2001); R. Bharadwaj, B.F.
Farmer, and R.A. Vaia, J. Chem. Phys. (2001) sub-
mitted for publication.
FREE
®
impact rating
of159
19
As the Joumal of Materials Research (JMR) celebrates 16 years, it has become one
of the world’s premier archival publications on advanced materials research.
Published monthly (over 4000 technical pages annually), JMR contains archival
papers, rapid communications and reviews. It is comprehensive in nature, and
over the past 16 years has addressed more than 150 different topics including:
metals; semiconductors; superconductors; ceramics; dielectrics; electronic and
magnetic materials; polymers; fullerenes; diamonds; adhesives; thin films;
composites; nanostructures; and materials synthesis, growth and characteriza-
tion, as well as their chemical, physical and mechanical properties.
All print subscriptions to the 2001 edition of JMR include FREE online
Web access—full text of all JMR articles from January 1996 to the
current issue.
Journal of Materials Research—more than ever, the archival front-runner in
international materials research.
Advertisers in This Issue
Print ISSN: 0884-2914
Coden: JMREE
Page No.
2001 Subscription Rates:
Nonmembers
Chemat Technology, Inc.
EDAX Inc.
358
353
408
MRS Members
$785 USA
$ 90 USA
Gelest Inc.
$815 Non-US (surface)
$110 Non-US (surface)
High Voltage Engineering Inside front cover
$870 Non-US (air freight) $180 Non-US (air freight)
Huntington Mechanical
Laboratories, Inc.
Outside back cover
393
Nissei Sangyo Co., Ltd.
Thermionics Vacuum Products
357
Materials Research Society
506 Keystone Drive
Warrendale, PA 15086-7573 USA
Tel 724-779-3003
Fax 724-779-8313
info@mrs.org • www.mrs.org
For free information about the products and services
offered in this issue, check www.mrs.org/publications/
bulletin/advertisers or fill out and fax the inside back
cover to 724-779-4397.
www.mrs.org/publications/bulletin
MRS BULLETIN/MAY 2001
401