REVIEW

WWW.POLYMERPHYSICS.ORG

Polymer Nanocomposites for Electrical Energy Storage

Qing Wang,1 Lei Zhu2
1

Department of Materials Science and Engineering, The Pennsylvania State University, University Park, Philadelphia 16802

Department of Macromolecular Science and Engineering, Case Western Reserve University, Cleveland, Ohio 44106
Correspondence to: Q. Wang (E-mail: [email protected]) or L. Zhu (E-mail: [email protected])
Received 3 July 2011; revised 26 July 2011; accepted 27 July 2011; published online 17 August 2011
DOI: 10.1002/polb.22337

ABSTRACT: This review highlights the frontier scientific

research in the development of polymer nanocomposites for
electrical energy storage applications. Considerable progress
has been made over the past several years in the enhancement
of the energy densities of the polymer nanocomposites via tuning the chemical structures of ceramic fillers and polymer matrix and engineering the polymerceramic interfaces. This
article summarizes a range of current approaches to dielectric
polymer nanocomposites, including the ferroelectric polymer
matrix, increase of the dielectric permittivity using high-permittivity ceramic fillers and conductive dopants, preparation of
uniform composite films based on surface-functionalized fillers,
and utilization of the interfacial coupling effect. Primary atten-

tions have been paid to the dielectric properties at different

electric fields and their correlation with film morphology,
chemical structure, and filler concentration. This article concludes with a discussion of scientific issues that remain to
be addressed as well as recommendations for future research.
C 2011 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys
V
49: 14211429, 2011

INTRODUCTION Advances in portable electronic devices, stationary power systems, and hybrid electric vehicles create
demand for low-cost, compact, and high-performance electrical energy storage devices.1,2 Among various energy storage
technologies including batteries, fuel cells, capacitors, and
supercapacitors, capacitors possess the advantage of high
power density due to the fast electrical energy storage and
discharge capability. However, the energy density of capacitors, which is governed by the dielectric materials that separate the opposite static charges, is relatively low and falls
significantly short of rising demand in advanced applications.

materials. However, they generally suffer from relatively low

breakdown strength. While polymers exhibit high breakdown
strength, their low dielectric permittivity (e.g., <10) always
limits the energy density. The composite approach capitalizes
on the idea that the combination of inorganic materials of
large permittivity with polymers of high breakdown strength
may lead to a large energy storage capacity. Moreover, large
interfacial areas in the composites containing nanometer
scale fillers promote the exchange coupling effect through a
dipolar interface layer and result in high polarization levels
and dielectric responses.3 Compared to conventional ceramic
materials, polymer-based dielectric materials also offer processing advantages including mechanical flexibility and ability
to be molded into intricate configurations for electronic and
electric devices with reduced volume and weight.

The introduction of inorganic nanoparticles into polymer

matrices to form dielectric polymer nanocomposites represents one of the most promising and exciting avenues for the
development of dielectric materials with high energy density.
In general, the energy density of a dielectric material can be
derived from
Z
Ue EdD;
where E is applied electric field and D is electric displacement. For linear dielectric materials with a relative dielectric
permittivity er,
Ue 1=2 DE 1=2 er e0 E 2 ;
where e0 is the vacuum permittivity (8.85
1012 F/m).
Large dielectric permittivity can be obtained from ceramic

KEYWORDS: capacitors; dielectric properties; energy density;

ferroelectric polymers; ferroelectricity; interfaces; nanocomposites; nanoparticles; organicinorganic interfaces; polymer

nanocomposites; poly(vinylidene fluoride); structureproperty
relations

COMPOSITES BASED ON HYDROCARBON POLYMER MATRIX

One straightforward strategy to improve the dielectric permittivity of a polymer is the addition of inorganic fillers with
high dielectric permittivity. Normally, ferroelectric ceramics,
such as BaTiO3 (BT), Pb(Mg1/3Nb2/3)O3-PbTiO3 (PMN-PT), or
other ferroelectrics or relaxor ferroelectrics, possess a very
large dielectric permittivity.4 It has been demonstrated that
high dielectric permittivity ceramic fillers could greatly
increase the dielectric properties of composites. For instance,
a high dielectric permittivity of 49 was obtained at 100 Hz

C 2011 Wiley Periodicals, Inc.

V

WWW.MATERIALSVIEWS.COM

JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

1421

REVIEW

WWW.POLYMERPHYSICS.ORG

Qing Wang is an Associate Professor of Materials Science and Engineering at the

Pennsylvania State University. He received his Ph.D. in Chemistry from the University of
Chicago under the supervision of Professor Luping Yu in 2000. Prior to joining the faculty at
Penn State in 2002, he was a postdoctoral researcher with Prof. Christopher Ober at Cornell
University. Among other awards, he has received the National Science Foundation CAREER
Award, Rustum and Della Roy Innovation in Materials Research Award, and Virginia S. and
Philip L. Walker Faculty Fellow. His research interests include the development of
ferroelectric polymers, polymer nanocomposites, conjugated polymers, and ion-containing
polymers for energy storage and conversion applications.

Lei Zhu is currently an Associate Professor of Macromolecular Science and Engineering at

Case Western Reserve University. He received his B.S. degree in Materials Chemistry in 1993
and M.S. degree in Polymer Chemistry and Physics in 1996 from Fudan University. He
obtained his Ph.D. degree in Polymer Science in 2000 from the University of Akron. After 2year postdoctoral experience at University of Akron, he joined Institute of Materials Science
and Department of Chemical, Materials and Biomolecular Engineering at University of
Connecticut, as an Assistant Professor. In 2009, he moved to Case Western Reserve
University. He is recipient of NSF Career Award, 3M Non-tenured Faculty Award, and DuPont
Young Professor Award. His research interests include polymer and polymer-inorganic
hybrid materials for electric energy storage, development of artificial antibody as
nanomedicines, and supramolecular self-assembly of discotic liquid crystals.

in the calcium copper titanate (CCTO)/polyimide hybrid

films, which is 14 times that of the pure polyimide matrix.5
The loss tangent of the hybrid film is lower than 0.2 when
the concentration of the CCTO filler reaches 40 vol %. The
BT powder with a particle size of 1 lm and crystallite size
of 57.7 nm was uniformly dispersed in an epoxy resin matrix, leading to the composites with a dielectric constant of
13 and loss of 0.18 at a volume filler concentration of 25%.6
The breakdown strength was 124.3 MV/m at a volume filler
content of 7% and decreased as the concentration of BT was
increased.
The effective dielectric permittivity of the nanocomposite has
been estimated using various theoretical approaches. The
Lichtenecker logarithmic rule is commonly used in a twophase composite system to predict the effective dielectric
permittivity based on the volume-fraction average:
log e /1 log e1 /2 log e2 ;
where /1 and /2 denote the volume fractions of ceramic fillers and polymer matrix, which have dielectric permittivities
of e1 and e2, respectively.7 The effective dielectric permittivity
of the binary composite given by the Bruggeman model is:






e1  e
e2  e
/2
0;
/1
e1 2e
e2 2e
which is based on a mean field approximation of spherical
inclusions surrounded by polymer matrix.8 It has been
shown that the effective dielectric constant predicted by the
Bruggeman equation increases sharply for filler volume fractions above 20% and can be very high for ceramic particle
loadings higher than 50 vol %. Additionally, the dielectric
permittivity of the composites could be calculated according
to the effective-medium theory (EMT) by averaging over the
1422

JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

dielectric permittivity of the two constituents.9 An arbitrary

fitting parameter was introduced in the model to account for
the irregular morphology of the ceramic fillers. The experimental results of lead magnesium niobate-lead titanate/epoxy composites have been successfully fitted using the EMT
model with less than 10% error.
A variety of chemical approaches have been developed to
improve the dispersion of ceramic filler in polymer matrix.
Polystyrene was directly grafted onto TiO2 surface via the
atom transfer radical polymerization (ATRP), which exhibited
a dielectric constant enhancement of over three times that of
bulk polystyrene.10 The high permittivity core lends high capacitance, while the flexible polystyrene shell endows the
materials with good dispersability and film-form properties.
As shown in Scheme 1, BT-PMMA (polymethyl methacrylate)
nanocomposites were prepared via in situ ATRP of MMA from
the nanoparticle surface.11 It was found that the nanocomposites showed an increased dielectric constant (14 at 1 kHz),
and the nanoparticles had no influence on the relaxation activation energy of PMMA matrix. Guo et al. prepared isotactic
polypropylene-TiO2 and BT nanocomposites via in situ-supported metallocene olefin polymerization process.12 The composites exhibited respectable permittivities (6.1), high breakdown strength (400 MV/m), and an estimated energy
density of 9.4 J/cm3. It was speculated that, in a well-dispersed 0-3 composite, interfaces between the ceramic nanoparticle and polymer phases create effective electron scatters
and transport centers, thus reducing the breakdown probability. Moreover, well-dispersed ceramic nanoparticles should
block degradation tree growth and thus increase the longterm breakdown strength. The effect of shape of the nanoparticles on the effective permittivity has been further investigated and simulated.13 As illustrated in Scheme 2, aluminum
oxide encapsulated high-permittivity BT, and ZrO2 core-shell

REVIEW

WWW.POLYMERPHYSICS.ORG

SCHEME 1 Schematic diagram illustrating the process of ATRP from the surface of BaTiO3. From Xie et al., J. Mater. Chem., 2011,
C The Royal Society of Chemistry, reproduced by permission.
21, 5897-5906, V

nanoparticles were prepared via a layer-by-layer methyaluminoxane coating process and utilized as fillers in in situ-prepared polypropylenes.14 The effective permittivity of the nanocomposites increases with the increase of the filler
concentration, affording e values as high as 6.2. It was found
that the effective permittivity of the nanocomposites can be
predicted by the Maxwell-Garnett formalism using the EMT
for volume fraction of nanoparticles below 0.06. The incorporated moderate permittivity of the Al2O3 layer greatly suppressed leakage currents and decreased the dielectric loss of
the composite. Recently, diblock poly(styrene-co-vinylbenzylchloride) was utilized as shielding layers on BT particles to
enhance compatibility between the filler and polystyrene matrix.15 A large breakdown strength of up to 222 V/lm was
achieved in the composites, leading to a noticeable energy
density value of 9.7 J/cm3.
It is known that the introduction of specific functional
groups at the particle/matrix interface can have profound
effects on the electrical properties of nanocomposites. The
influence of surface modification of TiO2 nanoparticles on
the short-term breakdown strength and space charge distribution of low-density polyethylene (LDPE) has been examined.16 As shown in Figure 1, it was found that the existence
of polar groups such as N-(2-aminoethyl)-3-aminopropyl-tri-

methoxysilane (AEAPS) onto the nanoparticle surface

improved both the dielectric breakdown strength and space
charge distribution as compared to LDPE filled with
untreated nanoparticles, which were mainly due to enhanced
electron scattering from the polar groups in AEAPS.
COMPOSITES BASED ON FERROELECTRIC
FLUOROPOLYMERS

The hydrocarbon polymers used as matrices in the dielectric

nanocomposites, including polyethylenes, poly(methyl methacrylate)s, epoxy resins, and polyimides, usually possess dielectric permittivities of 25 that are significantly lower than
their inorganic counterparts, thus severely limiting the energy
density obtained in the polymer matrix and consequently, in
the resulting nanocomposites. Generally, the energy density Ue
of a diphasic composite is represented by the equation below:
Ue /1 Ue1 /2 Ue2 gU 3 ;
where /1 and /2 are volume fractions of the constituent
dielectric materials in a composite and Ue(1) and Ue(2) are
their corresponding energy densities, U(3) is the energy density associated with interface effects, and g is proportional to
the interfacial areaeither a positive or negative

C
SCHEME 2 Synthesis of isotactic polypropylene-metal oxide nanocomposites. From Li et al., Chem. Mater., 2010, 22, 5154-5164, V
American Chemistry Society, reproduced by permission.

WWW.MATERIALSVIEWS.COM

JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

1423

REVIEW

WWW.POLYMERPHYSICS.ORG

phases.17,18 Superior energy density (e.g., 17 J/cm3 at 600

MV) that far exceeds those of conventional dielectric materials, such as biaxially oriented polypropylene, has been
obtained in a defect-modified PVDF-based copolymer.19
PVDF and its copolymers, such as poly(vinylidene fluorideco-trifluoroethylene) [P(VDF-TrFE)] and poly(vinylidene fluoride-co-hexafluoropropylene) [P(VDF-HFP)], have thus been
utilized in the formation of the dielectric nanocomposites.

FIGURE 1 Weibull plots of dielectric breakdown strength of neat

LDPE, LDPE filled with as-received TiO2 nanoparticles (AR-TiO2/
LDPE), and LDPE filled with AEAPS-coated TiO2 nanoparticles
(AEAPS-TiO2/LDPE). From Ma et al., Nanotechnology, 2005, 16,
C IOP Publishing Ltd, reproduced by permission.
724-731, V

contribution to energy density. Therefore, it is important to

have high energy densities from both phases in order for the
nanocomposite to exhibit considerable energy density.
Poly(vinylidene fluoride) (PVDF)-based ferroelectric polymers exhibit large spontaneous polarization and high dielectric constants (10 at 1 kHz) because of the presence of
highly electronegative fluorine on the polymer chains and
the spontaneous alignment of C-F dipoles in the crystalline

Ferroelectric Polymer Composites Containing

Conductive Fillers
It is known that the dielectric constant (k) of composites can be
greatly improved by adding conductive or semiconductive fillers
in the insulator matrix as the volume fraction f of the fillers
increases to the vicinity of the percolation threshold fc. k of the
composites can be described by the well-known power law
k
/ jfc  f js ;
km
where km is the dielectric constant of the polymer matrix
and s is an exponent of about 1. Using this strategy, the
dielectric constant of composites can be increased to tens or
even hundreds times the polymer matrix.20 For instance, tetrameric Cu-phthalocyanine was shown to possess a very
high apparent dielectric constant (er 104105) due to a nomadic polarization mechanism.4 Its polymer blends or
grafted polymers with P(VDF-TrFE) exhibited large er value

FIGURE 2 Effective dielectric constant of PVDF/xGnP nanocomposites as a function of the xGnP volume fraction, measured at
1000 Hz and room temperature. Inset (a) shows the best fits of the conductivity to the power law. Inset (b) shows the loss tangent
C John Wiley
of PVDF/xGnP nanocomposites as a function of xGnP volume fraction. From He et al., Adv. Mater., 2009, 21, 710-715, V
& Sons, Inc., reproduced by permission.

1424

JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

WWW.POLYMERPHYSICS.ORG

REVIEW

plates (xGnP) fabricated by a solution-cast and hot-pressing

method, a high dielectric constant of more than 200 and
2,700 could be obtained near the percolation threshold (1.01
vol %) at 100 and 1,000 Hz, respectively, which are 20 and
270 times higher than that of the PVDF matrix (Fig. 2).27
The giant increment in the dielectric permittivity was
explained by the combination of the microcapacitance-structure model and the Maxwell-Wagner-Sillars (MWS) effect.
One issue of these composites is the increase of dielectric
loss simultaneously with the dielectric constant. When the
composition of composites approaches the percolation
threshold, the dielectric loss is normally very high and the
breakdown strength is also significantly reduced. Since the
dielectric strength of this class of composites remains
unclear, the usefulness of this approach for high energy density capacitors requires further study.

FIGURE 3 Dielectric spectroscopy of each PFBPA-BT:P(VDFHFP) nanocomposite from 20 Hz to 1 MHz: dielectric constant
(a) and the loss tangent (b). The dielectric loss gradually
increases with increasing frequency and it is mainly due to a
resonance of the polymer host. From Kim et al., ACS Nano.,
C American Chemistry Society, reproduced
2009, 3, 2581-2592, V
by permission.

Ferroelectric Polymer Composites Containing

Surface-Functionalized Fillers
Dispersion of inorganic fillers in the ferroelectric polymers is
always problematical because of the low surface energy of
the fluoropolymers. The agglomeration of the ceramic dopants gives rise to electron conduction for a high dielectric
loss and undesirable porosity for dielectric failure at much
lower fields. It has been shown by Perry et al. that phosphonic acid surface-modified BT nanoparticles yielded welldispersed P(VDF-HFP) composites.28 The effective permittivity of nanocomposites with low volume fractions (<50%)
was in good agreement with standard theoretical models,
with a maximum relative permittivity of 35. However, as
shown in Figure 3, for nanoparticle volume fractions of
greater than 50%, the effective permittivity was observed to
decrease with increasing nanoparticle volume fraction, and
this was correlated with an increase in porosity of the spincoated nanocomposite films. The dielectric breakdown
strength was also found to decrease with increasing volume

of 3550 with 2540 wt % loadings at 10 kHz.21 The dielectric constant of the Ni-PVDF composites with fNi ! fc can
reach about 400 and possess a weak frequency and temperature depenence.22 The incorporation of CCTO into P(VDFTrFE) led to 0-3 composites with a dielectric constant of 610
at room temperature and 100 Hz, which was attributed to
the heterogeneous relaxation in addition to the semiconductive nature of the fillers.23 Dang et al. fabricated the PVDF/
carbon nanotube composites, with a percolation threshold of
8 vol %, possessing a dielectric constant of 600 and a dielectric loss (tan d) of 2 at 1 kHz.24 For poly(vinylidene fluorideter-trifluoroethylene-ter-chlorofluoroethylene) P(VDF-TrFECFE)/carbon nanotube composites, the dielectric constant
increased from 57 to 102 (tan d  0.36) at 100 Hz by inclusion of only 2 wt % (1.2 vol %) carbon nanotube.25 A dielectric constant as high as 56 was observed in PVDF/acetyleneblack composites, when the acetylene-black concentration
was in the neighborhood of the percolation threshold (about
1.3 vol %).26 More recently, Fan and coworkers reported
that, for the composites of PVDF/exfoliated graphite nano-

FIGURE 4 The breakdown strengths (failure probabilities:

63.2%) at each volume fraction as determined from the Weibull
analysis. From Kim et al., ACS Nano., 2009, 3, 2581-2592,
C American Chemistry Society, reproduced by permission.
V

WWW.MATERIALSVIEWS.COM

JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

1425

REVIEW

WWW.POLYMERPHYSICS.ORG

particle sizes of 80100 nm and decreased again when the

particle size is further reduced under 50 nm.

FIGURE 5 The dependence of the energy density of ferroelectric

P(VDF-TrFE-CTFE) composites on the content of BaTiO3 nanoparticles. Inset: cross-sectional FE-SEM image of the nanocomposite
thin films with 23 vol % surface-modified BaTiO3 nanoparticles.
C American
From Li et al., Chem. Mater., 2008, 20, 6304-6306, V
Chemistry Society, reproduced by permission.

fraction of the BT nanoparticles, with an abrupt decrease

observed around 10% and a gradual decrease for volume
fractions of 2050% (Fig. 4). The measured energy density
at a field strength of 164 V/lm increased to a value of 3.2 J/
cm3 as the nanoparticle volume fraction is increased to 50%,
roughly in line with the trend of the permittivity.
The chemical functionalization of the BT nanoparticles with
ethylene diamine moieties on the surface also rendered the
particles a homogeneous distribution in the poly(vinylidene
fluoride-co-chlorotrifluoroethylene) [P(VDF-CTFE)] and poly(vinylidene fluoride-ter-trifluoroethylene-ter-chlorotrifluoroethylene) [P(VDF-TrFE-CTFE)] matrix.29 The dielectric permittivity increased steadily with the increase of modified BT
content. The electrical energy density of the nanocomposites
as a function of filler concentration is summarized in Figure 5.
Clearly, the addition of modified BT nanoparticles into the
polymers greatly increased the energy density from 1.9 J/
cm3 of P(VDF-CTFE) to 3.7 J/cm3 of the composites containing 23 vol % BT nanoparticles at 150 MV/m. The dominant
role of dielectric permittivity of the polymer matrix in determining the energy density of the nanocomposite has also
been demonstrated. Titanate was used to modify the surface
of BT nanoparticles to establish the crosslinking between BT
particles and polymers.30 Microstructural analysis revealed
that coated BT nanoparticles were uniformly dispersed in
PVDF, which led to a large enhancement of the dielectric
breakdown strength (up to 250 kV/mm) at the BT volume
content of 7%. More recently, the dependences of the dielectric and ferroelectric properties of the PVDF/BT composites
on the nanoparticle sizes were investigated systematically.31
The remanent polarization was found to increase as the size
of BT particles increases from 25 to 500 nm. The dielectric
constant of the composites exhibited a complex variation: it
reached a maximum value in the composites with BT nano1426

JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

Ferroelectric Polymer Composites with the

Interfacial Effect
From the energy storage point of view, inclusion of the nanoparticles with a permittivity of about hundreds and even
thousands into the polymers, which generally possess a permittivity less than 10, might not be desirable for an appreciable increase in energy density. As the filler has a much
greater permittivity than polymer matrix, most of the
increase in effective dielectric permittivity comes through an
increase in the average field in the polymer matrix with very
little of the energy being stored in the high permittivity filler
phase. Furthermore, the presence of a large contrast in permittivity between two phases gives rise to a highly inhomogeneous electric field and thus a significantly reduced effective breakdown strength of the composite.
It has been proposed recently that the presence of a large
amount of interfaces in nanocomposites could be utilized to
benefit the electrical and dielectric properties of the polymers. It is argued that the properties of the interfaces
between the particles and the matrix, which will themselves
be of nanometer dimensions, will have an increasing dominant role in determining dielectric performance as the particle size decreases.3 Tanaka et al. proposed a multicore model
for the interfaces in the nanocomposites, in which the interfacial area consists of the transitional layer tightly bonded
with polymer and nanoparticles, a bound layer about 29
nm, which is formed by a polymer chain strongly bound to
the first layer, and a third loose layers, which is a region
loosely coupling the second layer.32 In addition, the diffuse
Couy-Chapman layer is superimposed over the three layers.
It was found that interface layers in the nanocomposites
might be more conductive than the polymer matrix, which
mitigated the space charge accumulation and field concentration by fast charge dissipation.
Zhang and coworkers demonstrated large enhancement in
the electric energy density and electric displacement level in
the nanocomposites of P(VDF-TrFE-CFE) terpolymer/ZrO2
nanoparticles.33 Through the interface effect, the presence of
1.6 vol % of ZrO2 nanoparticles raised the maximum electric
displacement D from 0.085 C/m2 under 400 MV/m in the
neat terpolymer to more than 0.11 C/m2 under 300 MV/m
in the nanocomposites.
The dielectric nanocomposites composed of P(VDF-TrFECTFE)34 and surface-functionalized TiO2 nanoparticles with
comparable dielectric permittivities (40) and homogeneous
nanoparticle dispersions were prepared.35 It was found that
the presence of the nanoscale filler favors the formation of
smaller crystalline domains and a higher degree of crystallinity in the polymer. In drastic contrast to their weak-field
dielectric behavior, substantial enhancements in electric displacement and energy density at high electric fields have
been demonstrated in the nanocomposites. As shown in Figure 6(a), for the composites with 10 vol % TiO2, the energy
density is 6.9 J/cm3 at 200 MV/m, which represents a

REVIEW

WWW.POLYMERPHYSICS.ORG

for directly coupling with oxide nanoparticles to yield the

covalent-bonded nanocomposites. It was demonstrated that
covalent assembly of polymer matrix and inorganic filler not
only led to highly dispersed nanofillers without additional
surface modification but also provided great stability and
offered enhanced interfacial interactions for high polarization
responses under the applied fields. More interestingly, the
nanocomposites exhibit remarkable dielectric strengths without the usual penalty of sacrificing electric strength and
greatly improved energy densities than the matrix. The
energy density is 11.2 J/cm3 for P(VDF-CTFE) with 9.1 wt %
ZrO2 at 270 MV/m, which corresponds to a 60% increase
in comparison to the polymer matrix [Fig. 7(b)]. The
improvement in the energy storage capability of the nanocomposites has been rationalized based on the changes of
polymer microstructures and the rise of the electric displacement induced by the incorporated nanofillers.
SUMMARY AND OUTLOOK

FIGURE 6 (a) The stored energy density of the polymer and

nanocomposites as a function of the applied field. (b) Temperature dependence of the dielectric constant of the polymer and
nanocomposite measured at 1 kHz. From Li et al., Adv. Mater.,
C John Wiley & Sons, Inc., reproduced by
2009, 21, 217-221, V
permission.

45% increase in comparison to the polymer matrix with an

energy density of 4.7 J/cm3 at the same field. The existence
of the nanoparticle/polymer interfaces in the nanocomposites was evidenced in the dependence of the dielectric properties on temperature shown in Figure 6(b). A shift of the
dielectric relaxation peak toward a lower temperature and a
reduced dielectric loss tangent are presumably indicative of
interface polarization interaction and an increased trap density in the nanocomposites. The large interface area in the
nanocomposites would produce the MWS interfacial polarization at low frequencies and/or leads to an interaction zone
with Gouy-Chapman diffuse layer, thereby greatly affecting
polarization and dielectric responses of polymer matrix.
More recently, a new approach for the preparation of dielectric polymer nanocomposites was reported.36 As shown in
Figure 7(a), the strategy included the synthesis of the ferroelectric polymers terminated with phosphonic acids37 and
subsequent utilization of reactive end-groups of the polymer

WWW.MATERIALSVIEWS.COM

The field of dielectric polymer nanocomposites has witnessed many exciting progress over the past several years,
and the pace of progress has continued to accelerate. By
judiciously selecting a combination of polymer matrix and
nanoparticles, the dielectric properties can be tuned and the
energy density has been greatly improved in the polymer
nanocomposites compared to those of polymer matrix. Various methods have been developed to incorporate the ceramic fillers into polymer matrix in a controlled fashion. The
optimization organicinorganic interface via surface functionalization of ceramic nanoparticle affords excellent compatibility between the fillers and the polymer matrix and ensures
uniform composite films even at higher filler concentrations,
leading to high breakdown strength of the composites. High
dielectric performance in the nanocomposites has been realized via the large enhancement in polarization response at
high electric fields and changes in polymer microstructure
induced by the nanofillers.
There is plenty room for further improvement as several
issues remain to be addressed. Enhancing dielectric constant
but retaining high dielectric breakdown and achieving low
dielectric loss are the major challenges in this field. Currently, a high volume percentage of ceramic loading is
required to significantly improve the dielectric constants of
polymer nanocomposites. However, at such high volume contents, the mechanical properties of the composites will deteriorate and thus affect the processability of the thin films.
The breakdown strength of the composites will also be
reduced at high filler concentrations. Since the energy density scales quadratically with the applied field, the reduction
in breakdown strength may negate any potential increase in
energy density. The contribution from the polymerceramic
interfaces can be exploited to improve the energy density of
the composites by improvement of the breakdown strength
or enhancement of polarization response, or both. However,
the nature of the interfaces is still not well understood.
Many questions, such as how the interfaces are related to
the structures of the components and how the interfacial
JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

1427

REVIEW

WWW.POLYMERPHYSICS.ORG

FIGURE 7 (a) Synthesis of covalent-bonded ferroelectric polymer nanocomposites. (b) The energy density of the polymer and the
C American Chemistry Society, reproduced by
nanocomposites at varied fields. From Li et al., Chem. Mater., 2010, 22, 5350-5357, V
permission.

properties change with the filler size and the applied electric
field, need to be addressed by a focused effort extending all
the way from theoretical modeling, to rational design and
synthesis of hybrid structures with engineered interfaces, to
in-depth dielectric characterization. An improved understanding of the role of ceramic fillers and interfaces on
charge trapping, electron scattering and packing of polymer
chains on the breakdown strength,38,39 and the high-field
dielectric responses is central to development of next-generation dielectric polymer nanocomposites. A systematic study
of the chemical structures of polymers and fillers, and size
and morphology of fillers versus dielectric performance at
high electric field has not yet been carried out. Manufacturing issues such as film processing, deposition of electrodes,
and integration into module assemblies must also be
researched. As a highly interdisciplinary field, the progress
in polymer composites for capacitive energy storage is critically dependent on successful interactions across the boundaries of traditional disciplines and collaborative efforts
from chemists, physicists, and engineers.
1428

JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

The authors gratefully acknowledge support from the Office of

Naval Research and the National Science Foundation.

REFERENCES AND NOTES

1 Office of Basic Energy Science, US Department of Energy.
Basic Research Needs for Electrical Energy Storage; Office of
Basic Energy Science, US Department of Energy: Washington,
DC, 2007.
2 Whittingham, M. S. MRS Bull. 2008, 33, 411419.
3 Lewis, T. J. J. Phys. D Appl. Phys. 2005, 38, 202212.
4 Handbook of Low and High Dielectric Constant Materials and Their
Applications; Nalwa, H. S., Ed.; Academic Press: London, 1999.
5 Dang, Z. M.; Zhou, T.; Yao, S. H.; Yuan, J. K.; Zha, J. W.;
Song, H. T.; Li, J. Y.; Chen, Q.; Yang, W. T.; Bai, J. Adv. Mater.
2009, 21, 20772082.
6 Gilbert, L. J.; Schuman, T. P.; Dogan, F. Ceram. Trans. 2006,
179, 1726.
7 Rao, Y.; Qu, J. M.; Marinis, T.; Wong, C. P. IEEE Trans. Compon. Packag. Technol. 2000, 23, 680683.

WWW.POLYMERPHYSICS.ORG

REVIEW

8 Yoon, D. H.; Zhang, J.; Lee, B. I. Mater. Res. Bull. 2003, 38,
765772.

24 Dang, Z. M.; Wang, L.; Yin, Y.; Zhang, Q.; Lei, Q. Q. Adv.
Mater. 2007, 19, 852857.

9 Todd, M. G.; Shi, F. G. IEEE Trans. Dielectr. Electr. Insul.

2005, 12, 601611.

25 Zhang, S.; Zhang, N.; Huang, C.; Ren, K.; Zhang, Q. M. Adv.
Mater. 2005, 17, 18971901.

10 Maliakal, A.; Katz, H.; Cotts, P. M.; Subramoney, S.; Mirau,

P. J. Am. Chem. Soc. 2005, 127, 1465514662.

26 Chen, Q.; Du, P.; Jin, L.; Weng, W.; Han, G. Appl. Phys. Lett.
2007, 91, 022912.

11 Xie, L.; Huang, X.; Wu, C.; Jiang, P. J. Mater. Chem. 2011,
21, 58975906.

27 He, F.; Lau, S.; Chan, H. L.; Fan, J. Adv. Mater. 2009, 21,
710715.

12 Guo, N.; DiBenedetto, S. A.; Kwon, D.-K.; Wang, L.; Russell,

M. T.; Lanagan, M. T.; Facchetti, A.; Marks, T. J. J. Am. Chem.
Soc. 2007, 129, 766767.
13 Guo, N.; DiBenedetto, S. A.; Tewari, P.; Lanagan, M. T.;
Ratner, M. A.; Marks, T. J. Chem. Mater. 2010, 22, 15671578.
14 Li, Z.; Fredin, L. A.; Tewari, P.; DiBenedetto, S. A.; Lanagan,
M. T.; Ratner, M. A.; Marks, T. J. Chem. Mater. 2010, 22, 51545164.
15 Jung, H. M.; Kang, J. H.; Yang, S. Y.; Won, J. C.; Kim, Y. S.
Chem. Mater. 2010, 22, 450456.

28 Kim, P.; Doss, N. M.; Tillotson, J. P.; Hotchkiss, P. J.; Pan,

M. J.; Marder, S. R.; Li, J. Y.; Calame, J. P.; Perry, J. W. ACS
Nano. 2009, 3, 25812592.
29 Li, J.; Claude, J.; Norena-Franco, L. E.; Seok, S. I.; Wang, Q.
Chem. Mater. 2008, 20, 63046306.
30 Dou, X.; Liu, X.; Zhang, Y.; Feng, H.; Chen, J. F.; Du, S.
Appl. Phys. Lett. 2009, 95, 132904.
31 Mao, Y. P.; Mao, S. Y.; Ye, Z. G.; Xie, Z. X.; Zheng, L. S. J.
Appl. Phys. 2010, 108, 014102.

16 Ma, D. L.; Hugener, T. A.; Siegel, R. W.; Christerson, A.;

nneby, C.; Schadler, L. S. Nanotechnology
Martensson, E.; O
2005, 16, 724731.

32 Tanaka, T.; Kozako, M.; Fuse, N.; Ohki, Y. IEEE Trans.

Dielectr. Electr. Insul. 2005, 12, 669681.

17 Lovinger, A. J. Science 1983, 220, 11151121.

33 Chu, B.; Lin, M.; Neese, B.; Zhou, X.; Chen, Q.; Zhang, Q. M.
Appl. Phys. Lett. 2007, 91, 122909.

18 Lu, Y.; Claude, J.; Neese, B.; Zhang, Q. M.; Wang, Q. J. Am.
Chem. Soc. 2006, 128, 81208121.

34 Lu, Y.; Claude, J.; Zhang, Q. M.; Wang, Q. Macromolecules

2006, 39, 69626968.

19 Chu, B.; Zhou, X.; Ren, K.; Neese, B.; Lin, M.; Wang, Q.;
Bauer, F.; Zhang, Q. M. Science 2006, 313, 334336.

35 Li, J.; Seok, S. I.; Chu, B.; Dogan, F.; Zhang, Q. M.; Wang, Q.
Adv. Mater. 2009, 21, 217221.

20 Liang, S.; Claude, J.; Xu, K.; Wang, Q. Macromolecules

2008, 41, 62656268.

36 Li, J.; Khanchaitit, P.; Han, K.; Wang, Q. Chem. Mater. 2010,
22, 53505357.

21 Zhang, Q. M.; Li, H.; Poh, M.; Xu, H.; Cheng, Z. Y.; Xia, F.;
Huang, C. Nature 2002, 419, 284287.

37 Li, K.; Liang, S.; Lu, Y.; Wang, Q. Macromolecules 2007, 40,
41214123.

22 Dang, Z. M.; Lin, Y. H.; Nan, C. W. Adv. Mater. 2003, 15,

16251629.

38 Claude, J.; Lu, Y.; Li, K.; Wang, Q. Chem. Mater. 2008, 20,
20782080.

23 Arbatti, M.; Shan, X. B.; Cheng, Z. Y. Adv. Mater. 2007, 19,

13691372.

39 Claude, J.; Lu, Y.; Wang, Q. Appl. Phys. Lett. 2007, 91,
212904.

WWW.MATERIALSVIEWS.COM

JOURNAL OF POLYMER SCIENCE PART B: POLYMER PHYSICS 2011, 49, 14211429

1429