Fundamentals, Processes and Applications

Progress in Materials Science 57 (2012) 660–723
Contents lists available at SciVerse ScienceDirect
Progress in Materials Science

journal homepage: www.elsevier.com/locate/pmatsci
Fundamentals, processes and applications

of high-permittivity polymer–matrix composites
Zhi-Min Dang a,b,c,⇑, Jin-Kai Yuan b, Jun-Wei Zha a, Tao Zhou b,
Sheng-Tao Li c, Guo-Hua Hu d,e,⇑
a
Department of Polymer Science and Engineering, School of Chemical and Biological Engineering,
University of Science & Technology Beijing, Beijing 100083, China
b
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, China
c
State Key Laboratory of Electrical Insulation and Power Equipment, Xi’an Jiaotong University, Xi’an 710049, China
d
Laboratory of Reactions and Process Engineering, Nancy University, CNRS-ENSIC-INPL, 1 rue Grandville, B.P. 451,
54001 Nancy Cedex, France
e
Institut Universitaire de France, Maison des Universités, 103 Boulevard Saint-Michel, 75005 Paris, France
a r t i c l e i n f o a b s t r a c t
Article history: There is an increasing need for high-permittivity (high-k) materials

Received 21 April 2011 due to rapid development of electrical/electronic industry. It is
Received in revised form 27 May 2011 well-known that single composition materials cannot meet the
Accepted 14 August 2011
high-k need. The combination of dissimilar materials is expected
Available online 25 August 2011
to be an effective way to fabricate composites with high-k, especial
Abbreviations: Al2O3, alumina; BaTiO3, barium titanate; CB, carbon black; CCTO, calcium copper titanate; CF, carbon fiber;
CH2Cl2, dichlodo methylene chloride; CNF, carbon nanofiber; CNT, carbon nanotubes; CuPc, copper-phthalocyanine; DBSA,
dodecylbenzene sulfonic acid; DMF, dimethyl formamide; HA, hydroxyapatite; HDPE, high-density polyethylene; LDPE, low-
density polyethylene; LTNO, Li and Ti codoped NiO; MNCB, modified nanoscale carbon black; MWNT, multi-wall carbon
nanotubes; NMP, N-methyl-pyrrolidone; ODA, 4,40 -oxydianiline; PA, polyamide; PAA, poly(amic acid); PANI, polyaniline;
PbTiO3, lead titanate; PC, polycarbonate; PCMS, poly(p-chloromethyl styrene); PE, polyethylene; PEEK, polyetheretherketone;
PFSA, perfluorosulfonic acid; PHAE, polyhydroxyaminoether; PI, polyimide; PLZT, lead lanthanum zirconium titanate; PMDA,
pyromellitic dianhydride; PMeT, poly(3-methylthiophene); PMMA, polymethylmethacrylate; PMN–PT, lead magnesium
niobate–lead titanate; POM, polyoxymethylene or polyformaldehyde; PP, polypropylene; PPY, polypyrrole; PS, polystyrene;
PU, polyurethane; PVA, polyvinyl alcohol; PVDF, polyvinylidene fluoride; P(VDF–TrFE), poly(vinylidene fluoride–trifluoroeth-
ylene); P(VDF–TrFE–CFE), poly (vinylidene fluoride–trifluoroethylene–chlorofluoroethylene); P(VDF–TrFE–CTFE), poly(vinyli-
dene fluoride–trifluoroethylene–chlorotrifluoroethylene); PVP, polyvinyl pyrrolidone; PZT, lead zirconium titanate; SiO2,
silicon dioxide; SPAI, siloxanemodified polyamideimide; SrTiO3, zirconium titanate; SWNT, single-wall carbon nanotubes; TFBB,
3,4,5-trifluorobromobenzene; TFP-MWNT, trifluorophenyl (TFP)-functionalized MWNTs; THF, tetrahydrofurane; TiO2, titanium
dioxide; TMPTA, trimethylolpropane triacrylate; UHMWPE, ultrahigh molecular weight polyethylene; xGnP, exfoliated graphite
nanoplates.
⇑ Corresponding authors. Addresses: Department of Polymer Science and Engineering, School of Chemical and Biological
Engineering, University of Science & Technology Beijing, Beijing 100083, China. Tel./Fax: +86 10 6233 2599 (Z.-M. Dang),
Laboratory of Reactions and Process Engineering, Nancy University, CNRS-ENSIC-INPL, 1 rue Grandville, B.P. 451, 54001 Nancy
Cedex, France. Tel.: +33 383 17 53 39; fax: +33 383 32 29 75 (G.-H. Hu).
E-mail addresses: [email protected] (Z.-M. Dang), [email protected] (G.-H. Hu).
0079-6425/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.pmatsci.2011.08.001
Z.-M. Dang et al. / Progress in Materials Science 57 (2012) 660–723 661
for high-k polymer–matrix composites (PMC). This review paper

focuses on the important role and challenges of high-k PMC in
new technologies. The use of different materials in the PMC creates
interfaces which have a crucial effect on final dielectric properties.
Therefore it is necessary to understand dielectric properties and
processing need before the high-k PMC can be made and applied
commercially. Theoretical models for increasing dielectric permit-
tivity are summarized and are used to explain the behavior of
dielectric properties. The effects of fillers, fabrication processes
and the nature of the interfaces between fillers and polymers are
discussed. Potential applications of high-k PMC are also discussed.
Ó 2011 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 662
2. Fundamental aspects of high-k composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 665
2.1. Definition of high permittivity (high-k) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 665
2.2. Capacitance and electric energy storage of materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 666
2.3. Polarization and relaxation of dielectric materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 666
2.4. Dielectric strength of random polymer–matrix composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . 668
2.5. Theoretical models for dielectric properties of polymer–matrix composites . . . . . . . . . . . . . . 669
2.5.1. Maxwell–Garnett equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 669
2.5.2. Bruggeman self-consistent effective medium approximation . . . . . . . . . . . . . . . . . . . . 670
2.5.3. Jaysundere–Smith equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 670
2.5.4. Lichtenker rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 671
2.5.5. Percolation model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 671
2.6. Connection type of PMC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674
3. Processes for the fabrication of high-k PMC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674
3.1. Solid phase processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674
3.1.1. Direct compounding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674
3.1.2. Melt compounding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 675
3.2. Liquid phase processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 675
3.2.1. Liquid-phase assisted dispersion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 675
3.2.2. Solution route . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 676
3.3. In situ polymerization processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 676
4. Microstructure and interfaces of high-k PMC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 677
4.1. Two-phase high-k PMC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 677
4.2. Three-phase high-k PMC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 681
5. Effect of fillers on dielectric properties of high-k PMC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 683
5.1. Concentration effect of fillers on dielectric properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 684
5.1.1. Effect of ceramic fillers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 684
5.1.2. Effect of electrical conducting fillers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 687
5.2. Size effect of fillers on dielectric properties of PMC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689
5.2.1. Change in physical properties of fillers with size reduction . . . . . . . . . . . . . . . . . . . . . 689
5.2.2. Dependence of dielectric properties of PMC on size of fillers . . . . . . . . . . . . . . . . . . . . 690
5.2.3. Effect of micro–nanosize cofillers on dielectric properties of PMC . . . . . . . . . . . . . . . . 692
5.3. Shape effect of fillers on dielectric properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694
5.3.1. Shape of fillers and filler–polymer connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694
5.3.2. Effect of 1-dimensional fiber-shape fillers on dielectric properties. . . . . . . . . . . . . . . . 695
5.3.3. Effect of 2-demension plate-shape fillers on dielectric properties . . . . . . . . . . . . . . . . 698
5.3.4. Effect of core–shell fillers on dielectric properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 698
6. Effects of measurement conditions on dielectric properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 700
6.1. Temperature dependence of dielectric properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 700
6.1.1. Organic fillers/polymer composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 700
6.1.2. Insulating ceramic fillers/polymer composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 701
662 Z.-M. Dang et al. / Progress in Materials Science 57 (2012) 660–723
6.1.3. Semiconducting ceramic fillers/polymer composites . . . . . . . . . . . . . . . . . . . . . . . . . . . 702

6.1.4. Conducting fillers/polymer composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 703
6.2. Frequency dependence of dielectric properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 703
6.2.1. Organic fillers/polymer composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 703
6.2.2. Ceramic fillers/polymer composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 705
6.2.3. Conducting fillers/polymer composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 707
6.3. Electrical field dependence of dielectric properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 707
7. Applications for high-k PMCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 709
7.1. Applications in microelectronic field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 709
7.2. Applications in electrical engineering field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712
7.3. Applications in biomedical fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 713
8. Concluding remarks and future perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 714
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715
1. Introduction
As predicted by Moore’s law, the efficiency of electronic products is improving in an exponential

manner [1,2]. This rapid improvement in efficiency is concomitant with the creation of new materials
with high permittivity (called high-k dielectric materials). A higher-k dielectric material can store
more electric energy than a lower one. As a result, its use in electronic devices allows improving their
efficiency. Electronic systems are often composed of active components such as integrate circuits (ICs)
and passive components. Passive components have become of an increasing interest because they are
steadily growing in number as the electronics industry is progressing toward higher functionality [1].
For example, the ratio of the passive to active components in a mobile cellular phone is over 20 [2]. In
the next generation packaging technology, the passive components such as capacitors (C), resistors (R)
and inductors (L) will be integrated into the substrate as a thin film layer instead of being surface
mounted on the top of the substrate as discrete components. The passive components buried inside
the substrate are called integral passives or embedded ones. Embedded passives provide many advan-
tages over discrete components and play an important role in microelectronic field, as shown in Fig. 1
[3].
Among all passive components, capacitors call for special attention. Fig. 2 shows the market shares
of capacitors, resistors, and inductors in the United States. Because of the large amount of capacitors
employed in electronic systems, integration of capacitors is of much importance. Capacitors have
many applications, as shown in Table 1. They can be used for filtering, timing, alternating/direct (A/
D) current conversion, termination, decoupling, and energy storage. Particularly, the development
of microelectronics requires decoupling capacitors with higher capacitance and shorter distance from
its serving devices [4].
Apart from electronic industry, high-k materials are widely used in many civilian and military
applications including active vibration control, aerospace, underwater navigation and surveillance,
hydrophones, biomedical imaging, non-destructive testing and air imaging microphones [5,6]. For
example, high-k elastic rubber–matrix composites could be used as potential functional materials
for cable accessories in electrical engineering because they could balance the distribution of electric
field of cable terminal to prevent the cable from failure. Fig. 3 shows a clear field distribution of cable
terminal before and after the use of high-k rubber–matrix materials.
In fact, ceramics and metals with high stiffness and excellent thermal stability have also a high-k.
However, their high density, brittleness and challenging processing conditions impede their use as
high-k materials. On the other hand, polymers have the advantage of easy processing and mechanical
flexibility and low cost. Moreover, integration of resistors and capacitors into the internal structure of
printed wiring boards (PWB), or, directly into integrated circuits packaging requires materials compat-
ible with polymers used as support of electronic circuits [7–10]. Mechanical flexibility and tunable
properties of polymer matrix composites (PMC) make them attractive. In addition, polymer materials
Fig. 1. (a) Technology trend and market for organic substrates, (b) a schematic image of embedded passive substrate, the arrow
showing position of passives [3].
Fig. 2. Market share of capacitors, resistors, and inductors [4].
Table 1
Applications and properties of capacitors.
Application Value range Stability required Tolerance required

Filtering, timing 1 pF to 100 nF Moderate Moderate
A/D Conversion 1 pF to 10 nF Very high Very high
Termination 50–200 pF Low Low
Decoupling 1–100 nF Low Low
Energy storage >1 lF Low Low
have found many applications for electromechanical devices to perform energy conversion between
the electric and mechanical forms. These devices could be served as artificial muscles, smart skins
for drag reduction, actuators for active noise and vibration controls, and microfluidic systems for drug
delivery and micro-reactors [11–14]. However compared with inorganic materials, organic polymer
materials have often low dielectric permittivity, in the range of 2–5 [15]. In exceptional cases the
dielectric permittivity of a pure polymer can go beyond 10, but is still very low, impeding their use
for high-k applications, despite their excellent physical properties. Thus a key issue is to substantially
raise the dielectric permittivity of polymers while retaining their excellent mechanical properties.
Very recently, polymer composites with high-k and low dielectric loss and good process compatibility
Fig. 3. Electric field distribution of cable terminal (a) before and (b) after the use of high-k material. The values from 10% to 90%
show an electric field distribution at this site, respectively.
with printed circuit boards (PCBs) have been recognized as promising candidate dielectric materials
for embedded capacitors. This is because high-k polymer composites provide an ideal solution to com-
bine the dielectric and electrical properties of the ceramic or metal fillers and the low-temperature
processability and mechanical properties of the polymer matrix [16–45]. Furthermore, their good
adhesive properties are additional advantage for their use in embedded capacitor technology, which
pure ceramics or other dielectric materials lack.
Recently, a few strategies, including random composites of polymers, field structured composites
and synthesis of new polymers, have been developed to improve the dielectric permittivity of PMC.
The most common one is addition of fillers in polymers to make composites. These fillers includes
metals [16–25], ceramics [26–45], carbon based materials [46–62] and organic fillers such as semi-
conductive olligomer [63–71] and conducting polymers [72–76]. In general, ceramic–polymer com-
posites are prepared by adding high-k ceramic fillers in polymer matrices [26–45]. The advantages
of these composites include predictable dielectric properties, relatively low dielectric loss and easy
fabrication [20,28,29,43]. However, remarkable issues for use of the ceramic–polymer composites
are related to the deterioration of mechanical and processing properties due to the high concentration
of rigid ceramic particles in the flexible polymer matrix. By replacing ceramic particles with conduc-
tive particles, percolative polymer composites can be made with a drastic increase in dielectric per-
mittivity in the vicinity of the percolation threshold of the conductive particles. For example, high-k
dielectric composites with highly conducting fillers such as metals and carbon nanotubes have been
developed on the basis of percolation [16–25,63–76]. A permittivity value of 2000 has been reported
for silver flakes-doped epoxy systems [77]. Dang et al. reported a value of 400 in Ni/PVDF composites
[19,20]. These remarkable increases in dielectric permittivity, however, are always concomitant with
significant increases in electrical conductivity and dielectric loss due to the ‘insulator–conductor’ tran-
sition occurring at the percolation threshold. This transition also leads to extreme sensitivity of the
dielectric permittivity to the content of the conductive fillers. In other words, a small deviation from
the percolation threshold could result in serious drop of the dielectric permittivity, making it rather
difficult to control the parameters of the preparation process. The issues mentioned above are consid-
ered as drawbacks for conducting fillers/polymer composite systems.
As a milestone, superior electrical properties of nanotubes offer exciting opportunities for new high-
k polymer composites. Apart from electrical properties, nanotubes also impart better mechanical prop-
erties to composites at relatively low filler content [52]. Higher surface area and larger aspect ratios are
responsible for superior dielectric properties of nanotube composites [50,51]. At the same time,
agglomeration of nanotubes and their compatibility to polymer matrices are prime concern of
researchers in this field. Although different strategies including modifications of nanotubes are em-
ployed, it is yet difficult to fully overcome these challenges [78]. All-organic composites based on con-
ducting organic fillers such as conducting polymers are a new and interesting field in high-k dielectric
composites. This type of filler has the advantage of increasing the dielectric permittivity of host poly-
mers without much deterioration of their mechanical properties. Zhang et al. have developed a number
of high-k polymer composites using organic semiconducting fillers [64]. Other types of fillers can also
be used in these polymer composite systems to achieve required electrical and mechanical properties.
These types of composites are termed as three-phase polymer composite systems
[20,44,54,57,59,61,65]. Apart from fillers, different types of polymer matrices are also used. Choice
of polymer matrix depends upon the type of application of the final product. Thermoplastic,
thermoset [23,33,38,43,45] and elastomer [56,61] are all good candidates as hosts to high-k polymer
composites.
Dielectric permittivity of composites depends on the filler concentration. In the case of dielectric
fillers with low conductivity, such as ferroelectric ceramics and organic fillers, an increase in dielectric
permittivity of composite with respect to the filler concentration can be explained by well established
laws such as Bruggeman self-consistent effective medium approximation [79,80], Maxwell–Garnett
equation [80,81], Lichtenker rule [82], Jaysundere and Smith (J–S equation) [83] models. Compared
to ceramic–polymer composite systems, a percolative behavior is observed in the case of high con-
ducting fillers. In these percolative composites, high dielectric permittivity up to 1000 can be achieved
in the vicinity of the percolation threshold. An increase in dielectric permittivity of composites in the
vicinity of the percolation threshold can be explained by the well-known power law. It is widely be-
lieved that in the case of a percolative system, ultra-high dielectric permittivity of composites is not a
direct consequence of intrinsic dielectric permittivity value of fillers and host polymer as in the case of
ceramic–polymer composites. It is often considered as a consequence of interfacial polarization and an
effective increase of electrode surface area due to interconnection of conductive particle clusters near
the percolation threshold [80,84]. In other words, interfaces between polymer and filler are very
important in these percolative systems. They can be tailored by a number of variables. They include
the type and level of reinforcement, size and shape of filler, the type of matrix, and the composite
preparation process. All these factors are inter-related and should be considered systemically when
developing a new material with high-k. Other parameters like temperature and electric field also have
an effect on dielectric properties of polymer composites.
This paper attempts to review the development in the field of high-k polymer matrix
composites and provide a systematic and comprehensive analysis of the fundamental aspects
of high-k flexible materials and their potential applications. It is organized as follows. In Section
2, fundamentals relative to dielectric permittivity are introduced. They help to understand the
characteristics of dielectric properties of composite materials. Section 3 discusses fabrication
processes for high-k PMC. In Section 4, the effects of microstructure and interfaces of high-k
PMC on dielectric properties are addressed. Other factors affecting the dielectric properties of
high-k PMC are discussed in Sections 5 and 6. Section 7 presents potential applications and
unsolved issues of high-k PMC in different fields. This paper ends up with concluding remarks
and future perspectives.
It should be mentioned that this review paper does not intend to cover all aspects of high-k PMC
and related topics, due mainly to the rapid growth of this field and space limitations. The concluding
remarks and perspectives only reveal the authors’ own viewpoint. We apologize in advance for any
glaring omissions of pertinent works related to the topic of this paper. Of course, any technical defi-
ciencies in this paper are our own.
2. Fundamental aspects of high-k composites
2.1. Definition of high permittivity (high-k)
The term high-k dielectric refers to a material with a high dielectric permittivity (k) as compared to
SiO2 used in semiconductor manufacturing processes. The use of high-k materials has extended be-
yond electronics and has triggered the development of other high-k materials including PMC.
Although high-k PMC cannot be considered as gate dielectrics due to some well-known disadvantages,
they could be applied in other electronic devices. Moreover, their mechanical flexibility and tunable
properties make them appealing for high-k materials. It should be pointed out that the symbol ‘k’
for dielectric permittivity is mostly used in the microelectronic field. In electrical engineering and
other fields, the symbol ‘e’ is commonly employed. In this review paper, the symbol ‘k’ is used
throughout.
2.2. Capacitance and electric energy storage of materials
Two conducting electrodes separated by a dielectric constitute a capacitor (formerly called a con-
denser). If a battery of potential difference (V) is connected across the electrodes then the capacitor
can store electric charge (Q), expressed in coulombs, that is directly proportional to the applied volt-
age, expressed in volts (V), according to the following equation:
Q ¼ CV ð1Þ
where C is the capacitance expressed in farad (F). Hence the capacitance value is defined as 1 F when
the electric potential difference across the capacitor is 1 V, and a charging current of 1 A flowing for
1 s. The farad is a very large unit and is not encountered in practice. Therefore submultiples of the far-
ad are commonly encountered. In decreasing order of use, they are the picofarad (pF), the nanofarad
(nF), and the microfarad (mF). The capacitance of a capacitor with parallel electrodes is directly pro-
portional to the active electrode area and inversely proportional to the dielectric thickness as de-
scribed by the following equation:
A
C ¼ k0 kr ð2Þ
d
where k0 is the dielectric permittivity in vacuum and is 8.854 pF/m, kr is the relative permittivity. The
latter is a dimensionless physical quantity equal to the ratio of the permittivity of the medium to that
in vacuum. If the medium is a macroscopical material, it is often called the dielectric constant of the
material. Hence the dielectric permittivity (k) is defined by the equation k = k0kr. Eq. (1) shows when
an electric field is applied to a capacitive material; electric energy can be stored in it.
2.3. Polarization and relaxation of dielectric materials
In general, the permittivity of an insulating material depends on the frequency (v) in Hertz (Hz) of
the applied electric field and can be described as a complex physical quantity, where the imaginary
part is related to dielectric loss. The frequency dependence reflects the possible existence of dispersive
behaviors somewhere in the electromagnetic spectrum. The relative dielectric permittivity is a fre-
quency dependent complex quantity.
0 00
k ðxÞ ¼ k ðxÞ þ jk ðxÞ ð3Þ
where x is angular frequency and x = 2pm, k0 (x) denotes the real part of dielectric permittivity and
k00 (x), the imaginary part. Usually the dielectric permittivities of materials listed in databases are mea-
sured at a frequency of 1 MHz unless otherwise specified. The real part k0 is always different from zero
and represents the contribution to the polarization responsible for the energy storage in the material.
In order to address its substantial independence from the electric field, it is often referred to as the
relative dielectric permittivity of a material. The imaginary part k00 , usually called loss factor, shows
possible dissipative effects and its frequency spectrum differs from zero only in dispersive regions.
Sometimes, the dissipative behavior is characterized by means of the so-called loss tangent, which
is defined as tan d(x) = k00 (x)/k0 (x).
In a homogeneous material, the polarization and thus k00 (x) result from various contributions. The

latter add up to give an effective relative complex dielectric permittivity keff , which may be defined
in a very general form as follows:
X
rDC
keff ¼ k ðxÞ þ kMW;i ðxÞ þ j
i
xk 0
! !
X X r
¼
0
k ðxÞ þ
0
xÞ þ j k ðxÞ þ
kMW;i ð
00 00
kMW;i ð xÞ þ DC ð4Þ
i i
xk0
0
where kMW;i accounts for the ith interfacial contribution and rDC is direct current (DC) electrical con-
ductivity. It is worth noting that all the dispersive phenomena associated with the intrinsic polariza-
tion are accounted for by k(x) in Eq. (4). In the presence of multiple intrinsic relaxation, k(x) can be
expressed as a superposition of individual contributions. Indeed, each relaxation process, either orien-
tational or interfacial, can be analytically described by means of a proper relaxation function. In the
simplest case of a material consisting of identical and non-interacting dipoles and whose relaxation
is characterized by a unique time constant s, Debye derived a relaxation function for the complex per-
mittivity in the associated dispersion region [79,80]. In the frequency domain, the Dyebe model gives
the well-known formula:
ks k1
k ðxÞ ¼ k1 þ ð5Þ
1 þ jxs
where ks and k1 represent, respectively, the static (relaxed) and the high frequency (unrelaxed) values
of the permittivity with respect to the considered process. For a material showing a unique dipolar
relaxation in its whole spectrum, k1 in turn coincides with the aforementioned relaxed value of the
deformational permittivity. Eq. (5) can be spitted in its real and imaginary parts and then equivalently
expressed by the following pair of equations:
0 ks k1
k ðxÞ ¼ k1 þ ð6aÞ
1 þ x2 s 2
00 ðks k1 Þxs
k ðxÞ ¼ ð6bÞ
1 þ x2 s 2
Due to the very simplistic assumptions above, the Debye model fails in describing relaxation phe-
nomena in complex systems. Thus, Havriliak et al. proposed a more general formula using a phenom-
enological approach [79,80]. It has the merit of including the possibility of a distribution of different
time constants through the introduction in Eq. (6a) of two shape parameters a and b:
ðks k1 Þ
k ðxÞ ¼ k1 þ ð0 < a < 1; 0 < b < 1Þ ð7Þ
½1 þ ðjxsÞ1a b
Actually, the occurrence of a dispersive process of any nature in the frequency region of interest
represents a drawback. In order to advantageously exploit the application of a dielectric material,
the effective loss factor should be kept as low as possible. In fact, losses not only waste part of the in-
put energy, but also worsen the insulation properties of the materials. In particular, for any dielectric
material it is possible to identify a threshold electric field, which generates irreversible modifications
in the medium accompanied by the onset of an intense and disruptive flow of charges. Such a sudden
loss of insulation due to a very high electric field is called dielectric breakdown of the material. The
minimum field responsible for such an effect is named breakdown field or breakdown strength
(Ebreak). For most polymers, Ebreak is in the range 106–108 V/m. Up to now, it is still difficult to make
a reliable prediction of Ebreak according to the present physical models. In fact, there are several
possible causes and processes related to dielectric breakdown. However, it is generally accepted that
mechanisms responsible for dielectric discharges have both thermal and intrinsic (bulk) origins. In the
first case, both dielectric polarization and conduction losses determine a temperature increase of
the material. Indeed, the power density dissipated into the dielectric medium, at the expense of the
00
electric field, is proportional to keff ðxÞ:
! 00
W / xj E j2 keff ðxÞ ð8Þ
Such a heating, in turn, enhances the conductivity in a self-amplified process with catastrophic
consequences. In fact, the amount of heat that the material is not able to dissipate drives it to the
breakdown. As a second possibility, breakdown may be related to an avalanche discharge process that
begins with the promotion of few valence electrons to the conduction band. These electrons, being
accelerated by the applied electric field, strike against other valence electrons, driving them to the
conduction band by a kinetic energy transfer; as this process of charge carrier multiplication goes
on, the current flow grows rapidly in the dielectric and the material can locally melt.
2.4. Dielectric strength of random polymer–matrix composites
High-k polymer composites are often used to store electric energy. The maximum energy storage
density (wmax: J m3) of a material can be acquired via the follow equation:
1
wmax ¼ k0 kr E2break ð9Þ
2
In order to possess the maximum energy storage density, we must endow the material should have
both high dielectric permittivity and high breakdown field at same time. Nevertheless, in most prac-
tical applications, the electric field employed is less than the breakdown field. It is well-known that
one of the typical drawbacks of the random composite approach lies in a substantial decrease in
the material dielectric strength Ebreak. Higher values of k0 typically correspond to lower values of Ebreak.
This experimental evidence is confirmed by an abundant literature [43] and might be reasonably
interpreted as mainly due to interfacial (Maxwell–Wagner) polarization phenomenon [41–
43,85,86]. Although a general expression capable of reliably relating the dielectric strength to the
dielectric permittivity of composites is currently not available, it is worth reporting at least a couple
of simple models that can help predict results to a certain extent.
The first model deals with a very rough schematization: consider the polymer as a homogeneous
material and a purely elastic body at low strains. Under such conditions and in the case of Max-
well-stress actuation, the nominal breakdown field of the material is given by the following
expression:
sffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffi
V break Y Y
Ebreak ¼ ¼ e1=2 0 ffi 0:6 0 ð10Þ
d0 2k0 k 2k0 k
where k0 is the dielectric permittivity of the material, Y its elastic modulus, Vbreak the breakdown volt-
age of the sample and d0 its thickness at rest. This relation suggests that by increasing
pffiffiffiffi0 the dielectric
permittivity of the material, its dielectric strength is expected to decrease as 1= k . Interestingly, such
a trend is often found to provide a satisfying fit to experimental data [87].
A delicate model is offered by the local field theory of dielectric media [88]. Consider a dielectric
material subjected to an applied electric field E. Although the space-averaged electric field can be uni-
form, the actual internal electric field can locally vary from point to point, depending on interactions of
local fields generated by dipoles. Accordingly, the local electric field Elocal is defined as the field that
actually acts on an individual polarizable unit (such as a molecule or an atom); it is also known as
the Lorentz local field and is given by the following expression [88]:
0
k þ2
Elocal ¼ E ð11Þ
3
where k0 is the dielectric permittivity of the material surrounding the local polarizable unit (excluding
the unit itself). Eq. (11) can be used to
obtain anexpression that relates the dielectric permittivity
0 and
0
the dielectric strength of the matrix km ; Ebreak;m with those of the resulting composite kc ; Ebreak;c . In
fact, if it is assumed that the value of the local breakdown field of the matrix is the same when the
matrix is in its pure and its composite form by equating the right-hand member of Eq. (11) evaluated
for these two situations, the following expression can be obtained:
0 0
km þ 2 k þ2
Ebreak;m ¼ c Ebreak;c ð12Þ
3 3
0
Eq. (12) can be further developed by using kc , one of the mixing rules described below, according to
different models. It provides a useful tool to estimate the overall electrical behavior of the composite
system.
2.5. Theoretical models for dielectric properties of polymer–matrix composites
In a high-k PMC, fillers impart mainly dielectric properties while the matrix phase imparts mechan-
ical properties. Properties of composite generally show different trends in electrical properties com-
pared to pure constituent phases. Several mixing rules have been proposed to account for the
effective permittivity of a system consisting of two immiscible phases or multiphase.
2.5.1. Maxwell–Garnett equation

A number of models have been proposed and used for predicting the effects of second phases on
the dielectric properties of the composites. Among these models, those pertaining to continuous med-
ia filled with spherical particles are considered as a starting point for discussion. To start with mixing
rules, consider an isotropic medium (matrix) of dielectric permittivity km filled with spheroids of per-
mittivity kf. The volume fraction of filler particles is designated as /f and the resulting volume faction
of the matrix /m ¼ 1 /f . It is assumed that both the filler and matrix components have no dielectric
loss in frequency regions of interest.
For any two-phase composite one can write the lower permittivity, which denotes a series model
as shown in Fig. 4a,
km kf
kc;min ¼ ð13aÞ
km /f þ kf /m
and the upper dielectric permittivity, which corresponds to a parallel model as shown in Fig. 4b,
kc;max ¼ km /m þ kf /f ð13bÞ
This means that for a given physical system, the dielectric function, kc, must lie between these
bounds. Namely, the permittivity of a two-phase composite is between the kc,min and kc,max
(kc,min 6 kc 6 kc,max), which can be described by mixing models with series and parallel in consistent
with practical composites as shown in Fig. 4c.
Beyond such rough limits, further steps in modeling the dielectric properties of binary mixtures
have been accomplished in the framework of the so-called Wagner theoretical schemes [89,90]. By
equating these two alternative expressions, the so-called Sillars [91] or Landzu–Lifshitz [92] rules
can be obtained as follows:

3/f ðkf km Þ
kc ¼ km 1 þ ð14Þ
2km þ kf
However, these rules hold only for low volume fractions of filler and are restricted by electrical con-
ductivity values of filler and matrix. More accurate equation for predicting dielectric properties can be
achieved by the following equation known as Maxwell–Garnett equation [93–96].
" #
3/f ðkf km Þ
kc ¼ km 1 þ ð15Þ
ð1 /f Þðkf km Þ þ 3km
Fig. 4. Ideal connection ways of two-phase composites, (a) series model, (b) parallel model, and (c) a mixing model, pink lines
stand for electrodes.
This equation considers the dielectric permittivity arising from spherical fillers dispersed in a med-
ium. It should be noted that Maxwell–Garnett rule does not consider the resistivity of medium or filler
particles and hence gives advantage over Landu–Lifshitz mixing rule, which is limited only to compos-
ites where fillers should have higher electrical resistivity than matrix.
In the literature, Maxwell–Garnett equation is expressed in many forms and is also referred to sev-
eral other names such as Maxwell–Wagner [97] or Rayleigh [98] or Lorentz–Lorenz [99] or Kernner-
Bottcher [83] equation. One of such Maxwell–Garnett equation forms is given below,
kc km kf km
¼ /f ð16Þ
kc þ 2km kf þ 2km
This equation is applicable only to spherical particles. However, when dispersed particles are not
spherical in shape, it needs to be modified in order to take into account the geometry of dispersed par-
ticles. A common way to include the information about the geometry of dispersed particles is to intro-
duce a depolarization factor, which is related to their deviation from sphericity. Thus, Maxwell–
Garnett could be changed into this form to make it more general [90].
" #
/f ðkf km Þ
kc ¼ km 1 þ for /f < 0:1 ð17Þ
Að1 /f Þðkf km Þ þ km
where the parameter A is the depolarization. When A = 1/3 equation turns back to Eq. (16). The value
of A can be calculated or can be found in the literature [80].
2.5.2. Bruggeman self-consistent effective medium approximation

Bruggeman proposed an improvement of mixing rules as an extension of the Wagner scheme, in
which the initially low volume fraction of the filler is gradually increased by infinitesimal additions.
Exact solutions of the Wagner’s theory for very common cases can also be shown as Bruggeman equa-
tion when the samples are lamellae or disks and the spherical fillers are used.
3km þ 2/f ðkf km Þ

kc ¼ kf ð18Þ
3kf /f ðkf km Þ
The Bruggeman integration method leads to a distinct mixing rule, which permits to assess the
overall electrical response at much higher content of spherical filler. This is particularly attractive
for a disordered system, where constituent particles may become very close to each other and even
agglomerate, so that deviations from an ideally uniform and dilute system may be substantial even
at low filler concentrations ð/f > 0:1Þ. Thus the solution of a differential equation obtained from either
Eqs. (15) or (16) leads to the final Bruggeman’s formula [100,98]:
kf kc ð1 /f Þðkf km Þ
1=3
¼ 1=3
ð19Þ
kc km
This equation is expected to hold for /f values up to 0.5, with the constraint that the dispersed
spheres do not form a percolative path throughout the medium.
2.5.3. Jaysundere–Smith equation

Maxwell–Garnett equation is valid only for lower concentration of filler as shown in Fig. 5a. Since
the interaction between filler particles is relatively weak, it is often neglected in the course of theoret-
ical prediction for dielectric properties due to a large distance between filler particles. At higher con-
centration of fillers, the interaction between fillers is becoming significant because the distance
between fillers is extremely close, especial for nanosized fillers as shown in Fig. 5b. Moreover, electri-
cal field arising from the neat induced distribution of dipole moment is no more negligible when cal-
culating overall field locally experienced in the matrix. Based on this assumption, Jaysundere and
Smith proposed a more realistic mixing rule. They calculated the electric field with a dielectric sphere
embedded in a continuous dielectric medium by taking into account polarization of adjacent particles
and arrived at the following equation [83]:
Fig. 5. A schematic showing the distribution of filler particles in a polymer matrix at (a) low concentration of fillers and (b) high
concentration of fillers.
h i
3km ðkf km Þ
km /m þ kf /f ð2km þkf Þ
1 þ 3/f 2km þkf
kc ¼ h i ð20Þ
3km ðkf km Þ
/m þ /f ð2km þkf Þ
1 þ 3/f 2km þkf
2.5.4. Lichtenker rule

Lichtenker’s idea was based on the Wiener theory for bounds of effective dielectric function or
effective conductivity of a composite [101]. According to Weiner’s theory, the lower and upper bounds
are represented in Eqs. (13a) and (13b). The upper bound for this effective dielectric function is
reached in a system consisting of plane-parallel layers disposed along the electric field. The lower
bound is reached in a similar system, but with the layers perpendicular to the field. Similar results
were obtained using the method of cross-sections. Starting from Eqs. (13a) and (13b), Lichtenker
assumed that the effective dielectric function of the considered composite satisfies the equation:
a a a
kL ¼ /m km þ /f kf ð21Þ
where the parameter a varies from 1 to 1. Thus, the extreme values of this a index correspond to the
Wiener boundary values. The parameter a may be considered as describing a transition from anisot-
ropy at a = 1 to isotropy at a = 1. Each a value describes a specific microgeometrical topology of a
composite. One can therefore expect that the applicability of Eq. (21) is wider than that of the well-
known Maxwell–Garnett and Bruggeman equations that, in general, have no free parameters related
to the topology. For Bruggeman’s theory, the filling factor may be considered as a parameter describ-
ing the topology; in particular, it specifies the percolation threshold, as will be discussed in Section 2.6.
Several examples of applications of Lichtenecker’s equation can be found, especially in Ref. [101]. At
the same time, it is not very clear for what systems this Eq. (21) is valid, and what is the range of
its universality.
2.5.5. Percolation model

To increase the dielectric permittivity of polymers, the inorganic fillers mentioned in Section 2.5.1–
2.5.4 are electrical insulators. The dielectric permittivity of inorganic fillers, such as ceramics powders,
BaTiO3, PbTiO3, Pb(Zr,Ti)O3 (PZT), and Pb(Mg1/3Nb2/3)O3–PbTiO3 (PMN–PT) is high. However, the
dielectric permittivity of their polymer–matrix composites is still not high enough, even at high con-
centration of fillers [28,102–107]. Perhaps weak interfacial interaction and pores in composites are
responsible for the low dielectric permittivity of the composites with ceramic fillers at high content
[28,108–111]. In addition, the flexibility of the polymer–matrix composites is poor if the concentra-
tion of fillers employed is high [102–111]. Therefore, one must seek a new way to improve the dielec-
tric permittivity of composites. Most recently, based on percolation theory, the permittivity of
polymer–matrix composites is improved significantly using conducting fillers [19,20,50,51,55–
60,68,72–75].
Although conductive fillers can result in high dielectric loss, a need for high-k materials has pushed
researchers to test the effect of loading of these fillers in polymers. The use of conducting fillers is
motivated by the fact that they not only increase conductivity but also induce Maxwell–Wagner polar-
ization. As a result, the polarization leads to a high dielectric permittivity in conductive filler–polymer
composites. In general, the electrical conductivity of conductive filler–polymer composites exhibits a
non-linear increase in dielectric value when the concentration of fillers is above a percolation thresh-
old. This phenomenon cannot be explained by classical mixing rules but by percolation theories. At
low concentration of filler, conductive particles are separated from each other (see Fig. 6a and b)
and the electrical properties of the composites are dominated by the matrix. With increasing filler
concentration, local clusters of particles are formed, which is shown in Fig. 6c. At the percolation
threshold, these clusters form a connected three-dimensional network through the component (see
Fig. 6d), resulting in a jump in the electrical conductivity.
The microstructures shown in Fig. 6 are isotropic, which are formed by spheres or spheroids with
random orientation. When the spheroid grains of the minor phase are oriented, the microstructrures
become anisotropic, as shown in Fig. 7. Furthermore, if conducting fillers have a high aspect (length/
diameter) ratio, the anisotropic microstructure of composites can more easily be formed. In this case,
the percolation threshold in the oriented direction can be very low and the composite can exhibit
excellent flexibility.
In fact, when the concentration of conducting fillers is close to the percolation threshold, the com-
posites often display a metal–insulator transition in electrical property. Namely, the percolation
threshold represents the filler concentration at which conducting paths come into contact with each
other throughout the medium. The percolation theory allows estimating fc with power laws describing
the conductivity and dielectric permittivity of the composite near the metal–insulator transition by
fitting the experimental r data for ffiller > fc and ffiller < fc. They often abide the following relations
[19,20,50,80]:
Fig. 6. Schemes showing the formation of a percolation-like clusters structure with an increase in the concentration of
conducting fillers in a composite. (a) randomly dispersed grain structure; (b) uniformly dispersed grain structure; (c)
aggregated grain structure; (d) percolation-like clusters structure [80].
Fig. 7. Schematic microstructure of a two-phase composite with oriented minor-phase spheroids [80].
For electrical conductivity,
rc / ðffiller fc Þt for f filter > fc ð22aÞ

rc / ðfc ffiller Þq for f filter < fc ð22bÞ
For dielectric permittivity,

0
kc / ðfc ffiller Þq for f filler < fc ð23Þ
where fc is the percolation threshold, t the critical exponent in the conducting region, q the critical
exponent in the insulating region, and q dielectric critical exponent. Two-phase random media nor-
mally have a value of about 0.16 for the percolation threshold. The universal values of the critical
exponents q and t are commonly 0.8–1 and 1.6–2, respectively. The universality of percolation theory
also suggests that the dielectric permittivity should exhibit the same power-law dependence on the
volume fraction as the conductivity below fc, i.e., q 1. However, this is not always observed in prac-
tical continuum systems [112].
The percolation threshold also depends upon the size and shape of conducting fillers. Common con-
ductive fillers are metallic or graphitic and are in different shapes (spherical, platelet-like or fibrous)
Fig. 8. Ten connectivity patterns for two-phase composites [80,114].

and sizes. However, the incorporation of CNT, because of their large aspect ratio, allows for a low per-
colation threshold, as illustrated in Fig. 7 [50–56,113]. In addition, the dielectric values in these perc-
olative systems are very sensitive to the concentration of fillers. A slight change in concentration near
the percolation threshold can bring about a drastic change in dielectric properties. Parallel increases of
dielectric permittivity and loss are frequently observed.
2.6. Connection type of PMC
Properties of composite materials depend not only on the type of filler and that of the matrix, but
also on how they are coupled. This coupling concept is called connectivity and was first introduced by
Newnham [114]. Consider the connectivity of a two-phase system. If a phase is self-connected in three
directions, its connectivity is termed as 3. If self-connected in two directions, its connectivity is termed
as 2, and so on. A binary composite connectivity is defined by combination of terms m–n. Here m rep-
resents the connectivity of the active phase (filler) while n represents the connectivity of the inactive
one (matrix). In general, there are 10 different patterns for binary composites on the basis of connec-
tivity: 0–0, 0–1, 0–2, 0–3, 1–1, 1–2, 1–3, 2–2, 2–3, and 3–3, which are illustrated in Fig. 8. Herein, the
0–0, 0–1, 0–2, and 0–3 connectivity patterns in Fig. 8 are the dispersed grain structures, in which the
0–0 pattern is equivalent to the dispersed grain structure in Fig. 6b, and the 0–1, 0–2, and 0–3 patterns
correspond to the dispersed grain structure in Figs. 6a and 7a. The 1–2, 2–3, and 1–3 connectivity pat-
terns in Fig. 8 correspond to the aggregated grain structure in Fig. 6c, in which the minor phase grains
in the 1–3 pattern are aggregated in the form of single chain-like clusters, and those grains in the 1–2
and 2–3 patterns are aggregated in the form of close-packed clusters. Similarly to Fig. 7b, the 1–1 and
2–2 connectivity patterns are specific cases of aggregated grain structures, in which the grains are
aggregated along a definite direction as laminated structures. The 3–3 connectivity pattern in Fig. 8
stands for the percolation-like cluster structure in Figs. 6d and 7c in which the two phases form inter-
penetrating three-dimensional networks (percolation cluster). In current works, the 0–3 [115–120]
and 1–3 [121–133] connectivity patterns are often designed to fabricate filler/polymer matrix
composites.
3. Processes for the fabrication of high-k PMC
Homogeneous dispersion of fillers into polymer matrices is crucial to the properties of PMC, espe-
cially in the field of functional applications. All the processes used normally aim at distributing and
dispersing the fillers in the polymer matrices in a desired manner. There are mainly two types of dis-
persion processes: solid phase processes and liquid ones. The former often call up mechanical ap-
proaches. Usually they are very simple and straightforward and rely directly on mixing the filler
with the aid of mechanical force and can be used widely in practical production due to their conve-
nience, low cost and mass production. Mechanical mixing processes can be divided into categories:
direct mixing and melt mixing. A big disadvantage of mechanical mixing processes lies in relatively
poor dispersion and weak interfacial interaction between fillers and polymer matrices in comparison
to chemical ones (also named as liquid phase processes), which often ensure better dispersion of fillers
and stronger interfacial interaction.
3.1. Solid phase processes
3.1.1. Direct compounding

Direct mixing involves the mixture of filler and polymer without any pre-treatment. Such a process
was attempted by Haggenmuellera et al. [134]. They cut into small pieces a MWNT/PMMA composite
sample in which the MWNT were visibly not well dispersed. They mixed those pieces and hot pressed
the mixture to create a new sample. They repeated this cutting and mixing procedure 20 times and
found that the MWNT dispersion quality was improved. Their comment about the quality was based
on sections of the sample observed under an optical microscope [26]. Wong et al. [135] prepared
graphite nanostructured composites by ball milling, which reduced particle sizes to smaller platelets.
Not other critical advantages were noted. A bimodal distribution of 100 and 400 nm particles was ob-
served using a particle analyzer. The ultimate objective was to develop an alternative cost-effective
nano-scale carbon material with properties comparable to those of carbon nanotube based compos-
ites. Azhdar et al. investigated the compaction of polymer powders and PMMA based nanocomposites
by uniaxial high-velocity cold compaction (HVC), high-energy ball milling (HEBM) and a relaxation as-
sisted process [136]. It was found that a longer mixing time yielded a higher degree of dispersion of
the NiFe2O4 nanopowder on the PMMA particle surfaces. Kim et al. [137] fabricated Al2O3-based poly-
imide composite thick films for integrated substrates by using benefits of aerosol deposition method.
The relative permittivity and loss tangent of 3 wt%-Al2O3 composite film was 7.6 and 0.008 at 1 MHz,
respectively. By employing a direct mixing process, Dang et al. fabricated PVDF composites based on
MWNT [51], CF [58], Ni [19], BaTiO3 [40], (Ni–BaTiO3) [20], (MWNT–BaTiO3) [57], PANI [75], and LTNO
[35–37], respectively. Those composites showed different dielectric properties, depending on the type
of the filler. A big disadvantage of the direct mixing process lies in that it is not efficient at dispersing
fillers at the nanoscale if fillers tend to agglomerate. A pre-treatment for fillers is usually necessary to
achieve reasonable dispersion. However, in some cases pretreatment still does not yield commendable
results.
3.1.2. Melt compounding

To mix CNT with highly viscous fluids, especially thermoplastic melts such as PE and PS, one could
rely on high shear forces to break down CNT aggregates and to improve CNT dispersion. Twin screw
extruders are the most popular equipment for dispersing and processing CNT based composites owing
to their good dispersive and distributive mixing capability [138–141]. Mechanical stir or magnetic stir
could also be used to disperse CNT in lower viscosity solutions, usually thermosets or thermoplastics
containing solvents. Moniruzzaman et al. mixed SWNT in epoxy at 100 rpm for 1 h in a twin–screw
batch mixer (MicroCompounderw, DACA Instruments) and claimed that a uniform dispersion was
achieved [142]. However, Sandler et al. used a stir to mix MWNT in epoxy at 2000 rpm for 5 min
and still found MWNT aggregates dispersed in the resin [143]. The discrepancy shows the fact that
the high speed mechanical stir or magnetic stir cannot generate sufficiently high shear stress to break
down CNT aggregates as an extruder does. Safadi et al. [144] found that ultrasonic agitation improved
the CNT dispersion by decreasing the CNT aggregates in size and in some cases even separated the
CNT. However, the usefulness of ultrasonic agitation is limited to low viscosity media, such as water,
acetone and dimethylformamide (DMF). For example, Vaccarini et al. obtained uniform SWNT disper-
sion in epoxy by sonicating SWNT and polymer with CH2Cl2 and then evaporating CH2Cl2 [145]. The
use of ultrasonic agitation alone usually does not lead to uniform CNT suspension. Usually solution
sonication is an important step combined with other methods to disperse CNT in polymer melts
and solutions. It should be noticed that extrusion and ultrasonic agitation, the two most frequently
used processes for dispersion, may cause damage to CNT surface and may also lead to attrition in
CNT length which may be partially responsible for deterioration of mechanical properties of CNT
based polymer composites. However, the viscosity of polymer melts usually increases after addition
of fillers, especially at high concentrations of fillers. In such cases, processing of polymer melts be-
comes more and more difficult.
3.2. Liquid phase processes
3.2.1. Liquid-phase assisted dispersion

Dang et al. [29] reported a liquid-phase assisted dispersion method by utilizing ethanol as a disper-
sion medium to prepare uniformly dispersed BaTiO3/PVDF nanocomposites. When the BaTiO3 nano-
particles and PVDF particles were dispersed in ethanol at the same time, the former were inclined
to absorb on the surface of the PVDF particles to form a special structure. This is because an interstitial
hydrogen ion may exist in the BaTiO3 lattice when BaTiO3 nanoparticles are synthesized from a water
solution [146]. In fact, BaTiO3nanoparticles also lean towards mutual attraction to reduce the surface
energy. The mixture of BaTiO3 nanoparticles and PVDF particles in ethanol reaches optimum disper-
sion by virtue of these two simultaneous adsorption actions. The result showed that the mixture
with a suitable BaTiO3/PVDF volume fraction ratio promoted the absorption of nanosized BaTiO3
nanoparticles on the surfaces of the PVDF particles. Therefore, homogenous BaTiO3/PVDF nanocom-
posites could be prepared by the natural adsorption effects occurring between the BaTiO3 and PVDF
particles when the BaTiO3/PVDF volume ratio in the composite is appropriate. This powerful principle
could be applied to acquire inorganic filler/polymer nanocomposites.
3.2.2. Solution route

At high filler concentration, melt mixing is not necessarily the best choice for composite processing
because of higher viscosity of the mixture. By contrast, high concentrations of fillers can be incorpo-
rated into polymer matrices by using solution processing. This process involves dispersion of both
polymer and filler in a solvent. However, most systems require the use of large volumes of solvents
to fully dissolve the polymer and disperse the filler. Since common solvents such as toluene, chloro-
form, THF, or DMF are of high toxicity, a number of aqueous systems have been explored. Solution-
casting has been used to manufacture MWNT-containing PS [144,147,148], PHAE [149,150], PVA
[151], UHMWPE [152], and PP [153] composite films with homogeneous nanotube dispersions. Sim-
ilarly, SWNT-containing PP [154], PVA [155,156], and PVA/PVP [157] composite films have been pre-
pared. In many cases, ultrasonication processes are used to aid nanotube dispersion in liquid media.
Nevertheless, prolonged high-energy sonication has the potential to introduce defects to nanotubes
[158]. Ultrasonic treatment may also stabilize the dispersion by grafting polymers onto the CNT sur-
face through trapping of radicals generated as a result of chain scission [159]. In addition, surfactants
[157,160], polymer-functionalized nanotubes [151,155], and other chemical treatments of the constit-
uents [161] are often employed. DMF dissolves PVDF well. As a result, it is often used to prepare filler/
PVDF composites with high dielectric permittivity. After the PVDF is dissolved in DMF, filler particles
such as ceramic particles and MWNT are mixed with PVDF solution, respectively [44,50,52–55]. In
short, solution processes can be very efficient at dispersing fillers in PVDF solution.
3.3. In situ polymerization processes
In situ polymerization processes for fabricating polymer composites are usually associated with
better filler to matrix interactions leading to improved electrical and mechanical properties. A number
of works have used such processes to prepare nanotube–polymer composites [162–167]. Popielraz
and Chiang [166] fabricated BaTiO3/TMPTA and CdO/TMPTA composites by photo polymerization of
TMPTA monomer in the presence of BaTiO3 and CdO, respectively. Polymer composites obtained by
this process showed dielectric permittivity comparable with that of BaTiO3. The CdO/TMPTA compos-
ites showed relatively high dielectric permittivity and low dielectric loss as compared to those of
BaTiO3/TMPTA composites. Cochet et al. [167] prepared PANI/MWNT composites by an in situ
polymerization process in the presence of MWNT. Their results revealed site-selective interaction be-
tween the quinoid ring of the PANI and MWNT, thus opening the way to charge transfer processes, and
improving the electrical properties of the PANI/MWNT composites. Xiao and Zhou deposited polypyr-
role or PMeT on the surfaces of the MWNT by in situ polymerization. The Faraday effect of the con-
ducting polymer enhances the performance of super-capacitors with MWNT deposited with the
conducting polymer. Xiao and Zhou [168] prepared CNT/PMMA composites by in situ polymerization
of MMA in the presence of CNT.
To acquire materials with high dielectric permittivity and good thermal stability, Dang et al. called
upon in situ polymerization processes to prepare BaTiO3/PI composites [43] and CCTO/PI composites
[45] with high dielectric permittivity, high electrical breakdown strength and high thermal stability. In
situ polymerization processes allow for homogenous dispersion of functional fillers and therefore en-
sure strong interfacial interaction between the fillers and the polymer. As a result, the composites dis-
play excellent electrical properties.
In summary, the preparation process is very important for polymer composites because it
has a crucial effect on their dielectric properties. The type of process and its operation condi-
tions should be defined based on the physical, chemical, rheological and thermal characteristics
of the system.
4. Microstructure and interfaces of high-k PMC
Often composites compose of more than two kinds of constituents. Composition, microstructure
and interface of the composites have a crucial effect on the ultimate properties of materials
[169,170]. To acquire composite materials with desired properties, it is very important to properly de-
sign their composition, microstructure and interface.
4.1. Two-phase high-k PMC
The behavior and performance of polymer composites cannot be understood solely on the basis of
the inherent properties of its principal components. The interphase that exists between the filler and
polymer is essential part of the composite. It is unanimously acknowledged by the scientific commu-
nity that improvement in electrical and mechanical properties of composites is possible by full under-
standing of the interfacial region of composites. In traditional composites, the interfacial region is
defined as the volume in which the properties deviate from those of the bulk matrix or filler [171].
However, it is simpler, in the case of uniformly sized spherical particles, to consider a straightforward
calculation [172] of the interparticle distance l:
" 1 #
p 3
l¼d 1 ð24Þ
6/v
where d is the diameter and /v is the volume fraction of the particles. For example, a 15 vol% loading
of 10 nm diameter particles leads to an interparticle distance of only 5 nm, which is close to the radius
of gyration of a typical polymer molecule. This shows that in such a case, the polymer matrix can be-
have as if it all belonged to an ‘interphase. In addition, the strong influence of interfacial interactions
during processing can alter the matrix microstructure which can significantly affect the mechanical
behavior of the nanocomposite independently of direct load-bearing by the filler [173]. This is espe-
cially so when the polymer is semicrystalline. Therefore, the matrix microstructure must be critically
assessed when evaluating the performance of polymer composites. An important first stage to deter-
mine the microstructure of composites is to study the interaction between compounds participating in
composite formation. Interaction between filler and polymer matrix can be chemical or physical or
both. It is assumed that initial interaction way between polymer and filler can be a decisive factor
in determining overall dielectric properties of composites.
Ceramic particles such as BaTiO3 [29,34,146,174,175], PMN–PT [28], PZT [107], LTNO [35–37] and
CCTO [45,117] are often chosen as fillers to obtain two-phase high permittivity polymer composites.
The latter composites may display different microstructures and interfacial interactions because of
differences in size and surface characteristics of ceramic particles. Dang et al. explained strong attrac-
tion between BaTiO3 and PVDF on the basis of the microstructure of BaTiO3/PVDF composites [29].
BaTiO3 nanoparticles would be inclined to absorb on the surface of the PVDF particles to form a
half-baked core–shell structure. Kobayashi et al. [174] further clarified the effect of polarity of the
polymer matrix on the final microstructure of polymer composites. Xie et al. [175] highlighted the
importance of interactions between BaTiO3 and polyimide. To improve the interaction between filler
and polymer, appropriate surface modification on fillers is often employed [34]. For example, Dang
et al. reported that treatment of BaTiO3 surface with a coupling agent significantly improved the
dielectric permittivity. This was attributed to stronger bonding between BaTiO3 and PVDF. The
amount of the coupling agent also had a remarkable effect on the microstructure and interaction,
as shown in Fig. 9. BaTiO3/polyimide composites with high permittivity and good thermal stability
were also prepared by an in situ polymerization process [43].
To prepare polymer matrix composites with even higher dielectric permittivity, LTNO and CCTO,
two high permittivity ceramic fillers are often used [35–37,45,117]. The high permittivity of ceramics
mainly originates from an internal boundary layer capacitance (IBLC) structure as illustrated in
Fig. 10a [176–179].
Dang et al. reported CCTO/PI functional hybrid films with high dielectric permittivity prepared
using the in situ polymerization [45]. A surface morphological image of the hybrid film with
Fig. 9. Morphologies of fractured surfaces of BaTiO3/PVDF composites (40.0 vol% BaTiO3), in which BaTiO3 was modified with
different concentrations of KH550, (a) 0 wt%, (b) 1.0 wt%, (c) 2.0 wt%, and (d) 5.0 wt%, respectively [34].
40 vol% CCTO filler is shown in Fig. 10b. It can be seen that the CCTO grains are dispersed in the PI
matrix without any agglomeration. As such, the CCTO is coated by the PI and uniformly dispersed
in the hybrid film after imidization, as shown on the fractured cross-surface of the CCTO/PI hybrid film
in Fig. 10c. Moreover, the distribution of CCTO filler did not depend on the thickness of the film, indi-
cating that the in situ polymerization process is effective for preparing CCTO/PI hybrid films. Fig. 10d
shows that the grain particles of CCTO are not destroyed during the fabrication of the hybrid films and
that the boundary layer surrounding the grains which was of the IBLC structure can be clearly seen.
The choice of the grain particles size of a few micrometers was based on a balance between mechan-
ical and dielectric properties of the hybrid films. The CCTO was also dispersed into P(VDF–TrFE) 55/
45 mol% copolymer to prepare ceramic/polymer composites [117]. Although the solution-cast
CCTO/P(VDF–TrFE) composite sample exhibits more pores, the six-layer hot-press sample after ther-
mal annealing yields a dense microstructure, which is advantageous to the final high permittivity. The
interaction between the polymer matrix and filler also has an important effect on the dispersibility of
the filler in the polymer matrix. The strong interaction between PI and BaTiO3 particles led to the dis-
persion of BaTiO3 particles at the nanometer scale). Similar results were obtained for (Bi0.5Na0.5)0.94-
Ba0.06TiO3–P(VDF–TrFE) 70/30 composites with 0.1 and 0.3 ceramic volume fractions [180].
Change of the microstructure of two-phase polymer composites with highly conducting filler is of-
ten from the change of the formed percolation structure when the concentration of conducting filler is
close to critical value. Therefore, the microstructure as well as the percolation structure of the two-
phase polymer composites depends on several factors such as the size, the shape and the spatial dis-
tribution of the particles within the polymer matrix, and the adhesion and the interactions between
the two phases. These in turn depend on the preparation process. The influence of these factors on
the percolation threshold is summarized in Refs. [181–184]. Zois et al. [185] explained the percolative
Fig. 10. (a) Sketch of the IBLC structure: the light-gray areas and the gray lines represent the semiconductive grains and the
insulative grain boundaries, respectively. SEM morphology of (b) a fresh surface and (c) the fractured cross-surface of the CCTO/
PI hybrid film with 40 vol% CCTO filler. (d) High-magnification image of a CCTO grain embedded in the PI polymer, which is also
seen in a white-circle mark in image (c) [45].
behavior of several metal/polymer composites based on the above viewpoints. Xu et al. reported the
CB-filled polymer composites [184,186,187] with a double-percolation microstructure, which can be
formed in binary-polymer–matrix composites [188]. In this case, an extremely low percolation thresh-
old can be observed. Xu et al. reported. As shown in Fig. 11, Xu explained a typical composite consist-
ing of CB particles, HDPE and PVDF with the double-percolation microstructure [189].
Compared to spherical fillers, the percolation microstructure can be formed at quite low critical va-
lue for the CNTs/polymer composites because of their large aspect ratio and small dimensions. A crit-
ical obstacle in using CNTs filler is that it is difficult to disperse them well in most polymer matrices
because of the presence of strong interaction among nanotubes. The latter mainly originates from
physical attraction via van-der walls forces, hydrogen bonding and chemical bonding. A great effort
has been put to find suitable processes to disperse nanotubes into polymer matrices. One of the more
effective ways is to create interfacial bonding between nanotubes and polymer matrices. For this pur-
pose, surfaces of nanotubes are modified with certain organic groups such as carboxylic, fluoro, amino
[50,190]. In fact, this method is not only limited to nanotubes but can also be used for other types of
fillers. This increase in interfacial bonding promotes the dispersion of nanotubes, which in turn im-
proves dielectric properties. Several studies have shown improved dielectric properties of composite
after surface treatment of fillers like CNTs [191]. Dang et al. reported on a very high dielectric permit-
tivity (8000) in functionalized carbon nanotube/electroactive polymer nanocomposites [50]. The very
high permittivity was from the Maxwell–Wagner–Sillars (MWS) effect at the percolation and was
caused by a thin insulating layer of PVDF that was part of the functionalized MWNT. However, it is
very difficult to gauge interfacial bonding because no standard methods are available.
Researchers have also used atomic force microscopy to measure the force required to separate an
embedded nanotube from a polymer matrix, which can be used to estimate the interfacial bonding.
They could also align the nanotubes using mechanical stretching [192,193], extrusion [139], magnetic
Fig. 11. Schematic microstructure changes of the binary-polymer composites with increasing volume ratio of HDPE/PVDF:
(a) volume ratio of binary polymer is 1/4, indicating the formation of electrically conductive networks in the HDPE phase
solely, (b) volume ratio of binary polymer is 1/1, indicating the formation of a double-percolation structure in the
composite and (c) volume ratio of binary polymer is 4/1, indicating no electrically conductive networks in the composite
[189].
field [194], and electric field [195]. This alignment of nanotubes further lowers percolation thresholds
of CNT/polymer composites [196]. Fig. 12 shows the alignment of CNTs in PE under influence of mag-
netic filed. Fig. 13 gives the orientation of MWNT in PVDF parallel and perpendicular to tensile direc-
tion. In fact, the properties of CNT reinforced composites depend on orientation of CNT in addition to
dispersion, aspect ratio and interfacial bonding. Wang et al. reported on the enhancement of electrical
properties of ferroelectric polymers by PANI nanofibers with controllable conductivities [74]. Unlike 0-
demension fillers with sphere shape and 1-demension fillers with fiber or tube shape, a high dielectric
permittivity and low percolation threshold in nanocomposites based on PVDF and 2-demension exfo-
liated graphite nanoplates was reported [49]. The exfoliated graphite nanoplates can effectively in-
crease the dielectric permittivity at percolation structure.
Fig. 12. TEM image of MWNT aligned by high magnetic field (10T) in PE [194].
Fig. 13. Morphologies of samples ||a| perpendicular and ||b| parallel to the tensile-strain direction in the MWNT/PVDF
composite with 2.0 wt% MWNT [52].
4.2. Three-phase high-k PMC
Though the volume concentration of ceramic fillers is up to 50–60%, the dielectric permittivity of
the two-phase polymer composites consisting of ceramic particles with high dielectric permittivity
and polymers is often low due to the composition and structure limitation of this kind of composites
[28–34]. If the concentration of ceramic fillers is further increased, pores will appear so that the dielec-
tric permittivity and the breakdown field decrease [28]. If the fillers in the polymer composites have a
high electrical conductivity, according to the percolation theory the dielectric permittivity of the poly-
mer composites increases rapidly [19,21–25]. In many papers, the dielectric permittivity up to several
hundreds or several thousands has been reported [19,21,49,50]. However, a crucial issue is that the
dielectric loss is very high so that the high-k PMC cannot be used in many fields because a great deal
of heat will be produced under electrical field, which may not be acceptable. In order to overcome this
problem, it is suggested that the three-phase polymer matrix composites consisting of ceramic phase,
electrical conductive phase and polymer phase may be a good choice to require the high-k and low-
dielectric-loss materials [20,27,197,198].
Dang et al. [20] reported a three-phase high-k PMC material consisting of BaTiO3 and Ni as inor-
ganic fillers and PVDF as polymer matrix. When the volume fraction of BaTiO3 is 0.40, the dielectric
permittivity of the (Ni–BaTiO3)/PCDF composites increases from 40 to 800 with increasing volume
fraction of Ni from 0 to 0.23 (herein the percolation threshold is close to 0.23) and the dielectric loss
is less than 0.5 at percolation. Additionally, the dielectric permittivity of this kind of three-phase poly-
mer composite displays remarkable frequency dependence. Similarly, Devaraju and Lee [27] fabricated
a three-phase PI composite by dispersing a prescribed amount of BaTiO3 in PAA solution. That solution
was added to silver nanoparticles and was homogenized by stirring BaTiO3–PI mixture with 0.3 vol-
ume fraction of BaTiO3. The dielectric permittivity of 500 and dielectric loss of 0.23 at 100 kHz were
observed at percolation. Shen et al. [197] proposed a three-phase system consisting of PVDF, metallic
and magnetic particle Ni and ferrite particles. Ni particles were incorporated in ferrite-filled polymer
matrix. Ni particles preferentially existed in the gaps between large ferrite particles as shown in the
schematic Fig. 14. In such a three-phase system, large ferrite particles not only acted as a magnetic
phase, but also as a high volume fraction, discrete phase confining polymers and metallic particle into
continuous double-percolating structure. The addition of the secondary filler can bring about different
morphologies owing to the interaction with the primary filler. As mentioned above, the presence of
large ferrite particles in ferrite–polymer matrix forces Ni particles to reside in between them gives rise
to clusters-like structure of Ni with a large aspect ratio, which otherwise might not be possible in a
two phase Ni–polymer composite. A better understanding of morphology of three-phase composites
can only be achieved by studying the interaction of fillers in polymer matrices.
Fig. 14. (a) A cross section micrograph of a Ni/Ni–Zn-ferrite/PVDF composite with 60 vol% Ni–Zn-ferrite and 11 vol% nickel. (b)
Schematic of composite structure, oval marks nickel cluster with large aspect ratios [197].
Inorganic conductive particles can be used to improve the dielectric permittivity of polymer com-
posites. Organic conductive particles may even be a better choice because they might have better
compatibility with polymer matrices, which is helpful for improving the dispersion of organic particles
in the polymers. Patil et al. [198] explained the possible interaction of fillers and polymer matrix in the
three-phase composites where the secondary filler was PANI, an organic and conducting polymer. This
unique microstructure of three-phase composites resulted in unique dielectric properties. It is very
clear from examples above that the difference in nature of fillers can result in exceptional microstruc-
ture, which in turn will yield exceptional dielectric properties of these composites. By exploiting the
unique microstructure, the dielectric properties as well as mechanical properties of three-phase poly-
mer composites can be tailored to meet prescribed needs, which otherwise may not be possible for
two-phase polymer composites. Huang et al. [68,70] reported on a fully functionalized high-dielec-
tric-constant nanophase polymer with high electromechanical response by employing the copper
phthalocyanine oligomer (PolyCuPc) as high dielectric-permittivity filler, PANI as conductive phase
and PU as polymer matrix to form three-phase polymer composites. For these multicomponent nano-
composites, the high dielectric permittivity of the PolyCuPc enhances the dielectric permittivity of the
PU matrix, while the conductive PANI raises the dielectric permittivity of the composite via the per-
colation structure. This combined two-phase dielectric matrix serves as the high-dielectric-constant
host, and a dielectric permittivity near 1000 is achieved at 20 Hz. It is also explained that the enhanced
dielectric response in the PolyCuPc graft PU could be attributed to the much-reduced filler size, which
enhances the exchange coupling (an interface effect) as theoretically predicted by Li [199,200]. It is
known that the CuPc and PolyCuPc have strong tendency to form stacked assemblies due to their pla-
nar shape and aromatic nature [201]. Since the pendant oligomer grafts are located separately along
the polymer backbone and especially in the hard segments of the PU, the size of the aggregates is re-
stricted by the accessibility of adjacent oligomer molecules in the polymer network [202,203]. During
the deformation of the polymers under external fields, immobilization of the nanophases and polymer
backbones via chemical bonding greatly reduces the relative motion and further aggregation of the
nanophases within the polymer matrix, which improves the material reliability and breakdown field.
According to percolation, spherical conductive particles can be dispersed into polymers to improve
the dielectric permittivity. However, their volume concentration at percolation is often higher than
10% [19–25,72]. Such a high loading of conductive particles results in a decrease in flexibility of the
polymer composites. This is because spherical shape conductive particles have a low aspect ratio,
which is not helpful to form the percolation structure. Thus, conductive fillers with high aspect ratio,
such as CNT and CF, should be used for three-phase polymer composites [54,57,59]. Results of Dang
et al. [54] show when the MWNT with a diameter of 20–40 nm and a length of 1–2 lm (aspect
ratio = 25–100) is used as the conductive phase dispersed in the BaTiO3/polymer mixtures, the
three-phase (MWNT–BaTiO3)/PVDF nanocomposites display significant enhancement in dielectric
permittivity. Furthermore, when the volume fraction of MWNT is 0.01 and 0.02 respectively, the
increase in dielectric permittivity of the three-phase (MWNT–BaTiO3)/PVDF nanocomposites is
different, depending on the volume fraction of nanosize BaTiO3 particles. The differences in dielectric
properties can be attributed to the detailed microstructure of the three-phase (MWNT–BaTiO3)/PVDF
nanocomposites due to the concentration variation of MWNT and BaTiO3. Fig. 15 shows TEM images of
the fractured two-phase and three-phase composites with a fixed fMWNT = 0.02. In the absence of
nanosize BaTiO3, MWNTs trend to be poorly dispersed and a network of aggregated MWNT bundles
is observed (Fig. 15a). Moreover, a more effective three-dimensional conductive network has been
formed because of the additional connection of the isolated MWNT bundles (as indicated by arrows
in Fig. 15b) when the volume fraction of BaTiO3 is increased to 0.05.
In contrast, the image of nanocomposites with the 0.20 volume fraction of BaTiO3 shows significant
improvement in MWNT dispersion. MWNT particles (as pointed by arrows in Fig. 15c) are well sepa-
rated by BaTiO3 particles rich areas. The aggregated MWNT bundles mentioned above cannot be seen
any more. Fig. 16 shows schematic images of the microstructure of the (MWNT–BaTiO3)/PVDF nano-
composites with a fixed MWNT content (fMWNT = 0.02) and different concentrations of BaTiO3. The
threads and solid circles represent the MWNT and BaTiO3 particles, respectively, and the region in dash
circles shows the interfaces and microcapacitor structure. From the microstructure evolution pro-
cesses, a critical concentration of BaTiO3 particles (0.05) is found. Below 0.05, additional conductive
network of MWNT bundles are induced (see Fig. 16b). However, above that critical volume fraction
the conductive networks are interrupted rapidly with increasing concentration of BaTiO3 (see Fig. 16c).
Generally speaking, by controlling the microstructure and interface of two-phase and three-phase
polymer composites, the dielectric properties of the composites can be tailored. The composition of
the multi-phase polymer composites and preparation process are important parameters for that
purpose.
5. Effect of fillers on dielectric properties of high-k PMC
Many factors would have an influence on dielectric properties of the polymer composites. Among
them are the concentration, size and shape of fillers [7,20,28,36,37,43,45,80,117,205,206,191,207–
209].
Fig. 15. TEM images of the fractured (MWNT–BaTiO3)/PVDF composites with the volume fraction of BaTiO3 at (a) 0, (b) 0.05,
and (c) 0.20 [204].
Fig. 16. Schematic images of the microstructure of the (MWNT–BaTiO3)/PVDF composites with the volume fraction of BaTiO3 at
(a) 0, (b) 0.05, and (c) 0.20 [204].
80
70 100 Hz
Dielectric permittivity
60
50
40
30
20
Maxwell-Garnett
10 Bruggeman
Experimental
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6
fBaTiO3
Fig. 17. Variation of the dielectric permittivity of the BaTiO3/PVDF composites with the volume fraction of BaTiO3 particles. For
comparison, the calculations by using Maxwell–Garrett and Bruggeman equations are also shown [20].
5.1. Concentration effect of fillers on dielectric properties
5.1.1. Effect of ceramic fillers

The increase in dielectric permittivity of the polymer composites containing inorganic fillers with
low conductivity, such as ceramics, is due to the relatively high permittivity of ceramics compared to
polymers. In addition, an increase in the content of ceramic fillers in the composites increases the
interfacial area between the ceramic phase and polymer phase. As a result, the influence of interfacial
polarization on the dielectric permittivity and dielectric loss can be significant. Accordingly the rela-
tive permittivity increases with ceramic loading. Generally, the dielectric loss of almost all ceramic/
polymer composites increases with increasing ceramic filler loading. There are some exceptions to this
rule. Namely, at higher filler loading, the particles trend to agglomerate resulting in an increase in
porosity and a decrease in densification of composites. An increase in porosity leads to a decrease
in dielectric permittivity because the permittivity of air is low. Generally, theoretical predictions
are valid for lower volume fraction of filler loading [205,206]. This is attributed to imperfect dispersion
of ceramic particles in polymer matrices. Many theoretical models suggest that fillers in a material
should be ideally separated, non-interactive and roughly spherical shape. However, in practice they
are not in such case. Many composites of interest in practice deviate markedly from this ideal config-
uration. Recent studies show that the interphase regions of ceramic/polymer composites have a rela-
tive permittivity significantly different from that of the polymer or ceramic phases [191]. Hence the
effective permittivity of the composites depends on the permittivity of each phase in the mixture,
their volume fractions, shape, size, porosity, interphase polarizability and interphase volume fractions.
As a result, the effective permittivity has a nonlinear nature and consideration of all these parameters
make the calculations tedious. These could be the reason for the deviation of experimental values of
relative permittivity from theoretical ones at higher filler loading. In ceramic/polymer composites
high filler loadings are generally necessary to achieve high dielectric permittivity. However, high cera-
mic loadings dramatically decrease the adhesion of the composites, which will reduce their process-
ability and reliability of embedded capacitors. Therefore there is an upper limit for the ceramic loading
in ceramic/polymer composites [210]. Bai et al. [28] reported when the volume percent of the ceramic
powder was up to 60%, the measured dielectric permittivity of the composite became much lower
than that predicted from the following equation:

nqðkc kp Þ
k ¼ kp 1 þ ð25Þ
nkp þ ðkc kp Þð1 þ aÞ
where k is the dielectric permittivity of the composite, kp and kc are the dielectric permittivities of the
polymer marix and ceramic, respectively. a is the volume fraction of ceramic and n is a parameter re-
lated to the geometry of ceramic particles [207]. According to different models, Dang et al. [20]
showed the permittivity differences between experimental data of the BaTiO3/PVDF composites and
calculated ones by employing Maxwell–Garrett and Bruggeman equations mentioned in Chapter 2
(see Fig. 17).
Due to the effects of pores, interface interactions between ceramic and polymer, and the dielectric
permittivity of ceramic particles, the improvement in dielectric permittivity of the ceramic particle/
polymer composites is not very significant, whatever the types of ceramic particles, polymers and/
or preparation processes used. For example, Chahal et al. [7] reported a value of dielectric permittivity
of 39 for the 55 vol% ceramic/polyacrylonitrile composite and and 22 for the 50% ceramic/polynor-
bornene composite. Bhattacharya et al. [208] reported values of dielectric permittivity of 9 and 34
for two kinds of composites for filler loadings of 21% and 46%, respectively. Dang et al. [29] reported
that the dielectric permittivity of the BaTiO3/PVDF nanocomposites is about 40 and 30 at 103 and
107 Hz, respectively. This indicates strong frequency dependence of dielectric permittivity. To acquire
high-k PMC with excellent thermal stability, Dang et al. [43] reported on a BaTiO3/PI nanocomposite
prepared by in situ polymerization process. The dielectric permittivity is only 18 when the volume
concentration of BaTiO3 is 40%. However, it shows weak frequency dependence in the in situ prepared
BaTiO3/PVDF nanocomposite as shown in Fig. 18. [211] showed a direct assembly of BaTiO3 (50 nm)/
PMMA nanocomposite films by using electrophoretic deposition in the acetone/isopropyl alcohol (1:1)
mixture solution with PMMA and BaTiO3. As shown in Fig. 19, the nanosize BaTiO3 particles can be
dispersed in the PMMA homogeneously. The maximum dielectric permittivity of the BaTiO3/PMMA
nanocomposites reaches about 30 at 100 Hz when the volume content of BaTiO3 is 60%, while a low
dielectric loss (0.1) as shown in Fig. 19c. These results prove although the dielectric permittivity of
conventional ceramic particles is about several thousands, the dielectric permittivity of the polymer
composites is still low due to the effect of interface structure of the composites.
In order to further improve the dielectric permittivity of the polymer composites for a given cera-
mic concentration, two main methods can be employed [43]. One is to design and optimize the fab-
rication process so as to minimize pores in polymer composites [43,117]. Another one is to develop
novel ceramic fillers with very high dielectric permittivity by doping the metal oxide to form a bound-
ary-layer capacitor structure [36,37,45,108,117,179]. Dang et al. [43] reported that the dielectric per-
mittivity of BaTiO3/PI nanocomposites can be improved by a degassing process during in situ
polymerization, which may reduce the pores in the final composites. Arbatti et al. [117] reported
hot-pressed solution-cast films repeatedly at 200 °C in order to improve the uniformity and reduce
the pore. During the hot-press process, a multilayer stack of solution-cast films was packed in a ‘‘sand-
wich’’ figuration. That is, the top of a solution cast film faced the top of another solution-cast film in
the stack, whereas the bottom of the cast film faced the bottom of another cast film. The number of the
solution-cast films in the stack was two, four, and six, respectively. Based on this stack process, the
20 24
(a) (b) 40vol% BaTiO3
Dielectric Permittivity
15
22
10
10 3 Hz
10 4 Hz 20
5
10 5 Hz
10 6 Hz
0
0 10 20 30 40 18
3 4 5 6
10 10 10 10
Volume concentration of BaTiO3 (%)
Frequency (Hz)
Fig. 18. a) Dependence of dielectric permittivity on the volume concentration of BaTiO3 of the BaTiO3/PI composite films
obtained by in situ polymerization. (b) Frequency dependence of dielectric permittivity of the BaTiO3/PI composite film with
BaTiO3 loading at 40 vol%. Thermal imidization conditions were: 100, 200, and 300 °C for 1 h, respectively [43].
Fig. 19. FESEM micrographs of (a) surface and (b) cross-section for the BaTiO3/PMMA nanocomposite film with 45 vol% of
BaTiO3. Frequency dependence of (c) dielectric permittivity and (d) dielectric loss for the BaTiO3/PMMA nanocomposites as a
function of BaTiO3 [208].
400 1000
Two Layers
Four Layers (b)
Six Layers
Dielectric Constant
Dielectric Constant
300
200 100 Six Layers

Four Layers
TwoLayers
100 P(VDF-TrFE)
(a)
0 10
100 1k 10k 100k 1M 100 1k 10k 100k 1M
Frequency (Hz) Frequency (Hz)
Fig. 20. (a) Dielectric constant vs. frequency from 100 Hz to 1 MHz for the 50 vol% CCTO/P(VDF–TrFE) composites at room
temperature. (b) Dielectric constant vs. frequency from 100 Hz to 1 MHz for the composites after thermal annealing. For
comparison, the dielectric constant of the P(VDF–TrFE) 55/45 mol% copolymer is included [117].
dielectric permittivity of the CCTO/polymer composites can be increased with more layers hot-pressed
together as shown in Fig. 20a. After further annealing of the composites, the dielectric permittivity sig-
nificantly increases (Fig. 20b). The detailed change in dielectric properties of the CCTO/polymer com-
posites is also in Table 2. Dang et al. [37] reported LTNO/PVDF composites with high dielectric
permittivity of about 600 at 40 vol% LTNO loading. Compared to the undoped ceramic fillers at the
same concentration, the increase in permittivity with the doped ones is surprisingly high. The main
reasons are related to the very high dielectric effect and semiconducting characteristics of the LTNO
[177]. Compared to the LTNO/PVDF composites, the CCTO/PI composite films containing 40 vol% CCTO
prepared by in situ polymerization had a maximum value of dielectric permittivity of about 50 at
100 Hz, which was 16 times larger than thatof PI matrix [45]. The magnitude of increase in dielectric
permittivity of the CCTO/PI composites is less significant than that in the LTNO/PVDF composites,
which is about 60 times that of the PVDF matrix. However, the magnitude of increase in permittivity
of the CCTO/PI composites is higher than that in the BaTiO3/PI nanocomposites [43], which can be
attributed to the IBLC microstructure of CCTO fillers as shown in Fig. 12.
Table 2
Dielectric properties of hot-press CCTO/polymer composites at 22 °C [117].
Hot-press composites Dielectric constant/loss (100 Hz) Dielectric constant/loss (1 kHz)

Room temperature 70 °C Room temperature 70 °C
Two layers 245/0.32 483/0.21 174/0.22 394/0.12
Four layers 392/0.60 859/0.31 244/0.32 651/0.16
Six layers 362/0.43 838/0.41 243/0.26 595/0.21
50 30
(a) (b)
Relative dielectric constant
Relative dielectric constant

40
25
30
20
20
15
10 o
1100 oC 1300 C
o o o
900 C 1200 C 1400 C
0 10
10 20 30 40 20 30 40 50 60 70 80 90 100
o
Ceramic vol% Temprature( C)
Fig. 21. (a) Relative dielectric constants of the ceramic/polymer composite thick films, measured at 25 °C and 100 kHz, as a
function of the volume fraction of multi-doped ceramic particles. (b) Variation of relative dielectric constant of the 30 vol%
ceramic/polymer composite thick films with test temperatures for multi-doped ceramic fillers heat treated at 900–1400 °C for
1 h followed by 4 h milling, measured at 100 kHz [108].
Kuo et al. [108] also reported on a multi-doped (La, Mg and Sr) BaTiO3 ceramic particles filled epoxy
resin composites. After doping, the dielectric permittivity of the multi-doped BaTiO3 ceramic reaches
25,000, which is 10-times that of a classical undoped BaTiO3 ceramic. As a result, compared to the
common BaTiO3/polymer composites at same concentration of fillers, the dielectric permittivity of
the multi-doped BaTiO3/epoxy composites is extremely high. In Fig. 21a, the dielectric permittivity
of the multi-doped BaTiO3/epoxy composites is as high as 45 at 40 vol% loading. Additionally, after
doping and heat treatment at 900–1400 °C for 1 h, as shown in Fig. 21b, temperature dependence
of dielectric permittivity of the multi-doped BaTiO3/epoxy composites is so weak that the materials
might be used within a wide temperature range. It is obvious that the dielectric permittivity of the
polymer composites can be improved markedly by chemical doping to ceramic fillers so that the IBLC
structure can be formed, which is charge of the high dielectric permittivity of ceramic fillers.
5.1.2. Effect of electrical conducting fillers

The dielectric permittivity of current polymer composites is still low and cannot satisfy the need of
microelectronic industry. Flexible polymer composites with dielectric permittivity larger than 100 can
be used in embedded capacitors applied at low working electrical field. This is a challenging issue. For-
tunately, as discussed in Chapter 5.2, conducting fillers can markedly increase dielectric permittivity
of polymer composites by taking the advantage of their percolative behavior. Polymer–metal compos-
ites show an abrupt increase in the permittivity near the percolation threshold and have a narrow
smearing region [19,20,50,51,55–60,68,72–75,212–214]. A small change in filler concentration near
the percolation threshold can drastically alter the dielectric properties of the composites. This may
make the dielectric properties of polymer unpredictable and may render their applications difficult.
Dang et al. [23] reported on high dielectric permittivity Ag/PI composite films with excellent thermal
stability prepared by an in situ polymerization process as shown in Fig. 22. The effective dielectric
permittivity of the film with 12.5 vol% of Ag fillers achieved 400 at 103 Hz. The relative tolerance of the
dielectric permittivity of the film with 12.0 vol% of Ag fillers in a wide temperature range was less than
4.0%, which would satisfy the need in practical applications. Qi et al. [21] also reported on an Ag/epoxy
nanocomposite with 22 vol% of Ag having a high dielectric permittivity of 308 and a relatively low
dielectric loss of 0.05 at 103 Hz as shown in Fig. 23. In this composite material, 40 nm Ag particles
coated with a thin layer of mercaptosuccinic acid were randomly distributed in the polymer matrix.
Because of the low performance of inorganic fillers such as ceramic particles and metal particles for
flexible polymer composites, very recently organic fillers with high electrical conductivity have been
used to improve the dielectric properties of polymer composites [72,74–76]. Huang et al. [72] recently
reported on a high-dielectric-constant all-polymer percolative composite consisting of surface coated
PANI (c-PANI) particles (less than 1 lm) and P(VDF–TrFE–CTFE) terpolymer with relatively high room-
temperature dielectric permittivity (>50). In this c-PANI/P(VDF–TrFE–CTFE) composite system, upon
changing the concentration of c-PANI, the dielectric permittivity close to 1000 at 103 Hz is observed
when the volume fraction of c-PANI is 23%, which is about 50 times higher than that of the terpolymer.
Dang et al. [76] reported on a remarkable increase of dielectric permittivity in the PANI/PVDF nano-
hybrid films at low volume fraction of PANI (4.2%) shown in Fig. 24, owing to the use of PANI with
about 50 nm average diameter and special preparation process.
3 2
10 10 16
(a) 0 (b) 103 Hz
10 15
Conductivity (S m-1)
104 Hz
2 -2
10 10 14 105 Hz
-4
10 106 Hz
13
1 -6
10 10
12
-8
10
11
0 -10
10 10
0.00 0.05 0.10 0.15 0.20 0.25 10
-50 0 50 100 150
Volume fraction of Ag
Temperature (oC)
Fig. 22. (a) Dielectric permittivity (left side) and conductivity (right side) of the Ag/PI composite films as a function of the
volume fraction of Ag at 103 Hz and room temperature. (b) Temperature dependence of permittivity of the Ag/PI composite film
(12 vol%) at different frequencies [23].
300
(a)
300 22.3 vol% Ag (b)
200
1 kHz 200
10 kHz
100 kHz
1 MHz 14.2 vol% Ag
100 100
8.7 vol% Ag
0
0 Epoxy
0 5 10 15 20 25 20 40 60 80 100 120 140

O
vol% of Ag Temperature ( C)
Fig. 23. (a) Evolution of dielectric permittivity in the Ag/epoxy composites as a function of silver volume fraction at room
temperature. (b) Dielectric permittivity of the Ag/epoxy composites with different concentrations of silver as a function of
temperature at 103 Hz [21].
Fig. 24. Dependence of (a) dielectric permittivity measured at room temperature and 103 Hz, DC breakdown strength (Eb) and
(b) energy density (ED) of the PANI/PVDF films on the volume fraction of PANI [76].
Compared to the use of ceramic fillers, the use of conductive fillers can improve the dielectric per-
mittivity of the polymer composites to a greater extent and at low loading of conductive fillers. A low
loading of filler allows retaining the flexibility of polymers as much as possible.
5.2. Size effect of fillers on dielectric properties of PMC
5.2.1. Change in physical properties of fillers with size reduction

Size is an important property of fillers. Understanding the effects of the size and size distribution of
fillers help understand the dielectric properties of the filler/polymer composites. Since the physical
and chemical properties of fillers depend very much on size, especially when the sizes are smaller than
100 nm. In this size range, the bulk properties of the materials gradually disappear and approach
molecular behavior when the sizes approach about 1 nm. The band gap opening increases with
decreasing particle size. Some measured and calculated data for the effect of particle size on the band
gap of nanoparticles is displayed [215]. In addition, the dielectric permittivity also depends on particle
size. If the size of particles dispersed in the polymer matrix is in the order of nanometer, the dielectric
permittivity of the particles is dependent on size. Many expressions for obtaining the size-dependent
dielectric permittivity have been proposed. For example, the size-dependent dielectric permittivity on
a nanocrystal is approximated in the following equation [171]:

1 1 1 1
¼ 1 bðaÞ 1 1 ð26Þ
knc ðaÞ knc ðaÞ knc ðaÞ knc ðaÞ þ 3:5
And the electronic contribution to the total polarizability is shown in the following equation:
1
1 kbulk 1
knc ðaÞ ¼ 1 þ ð27Þ
1 þ ð3:75=aÞ1:2
where a is in Angstrom units, and b(a) indicates how much the ions participate in the screening.
Small size of fillers leads to an exceptionally large interfacial area in the composites. Fig. 25a shows
the surface area per unit volume as a function of particle size for spherical particles that are ideally
dispersed. Fig. 25b shows interparticle spacing as a function of particle size for an ideally dispersed
nanoparticle composite: at low volume fractions the entire matrix is essentially part of the interfacial
0.08
Surface Area/Unit Volume (cm )

-1
2 volume %
0.07 (a) 5 volume %
0.06 20 volume %
0.05
0.04
0.03
0.02
0.01
0.00
20 30 40 50 60
Partical Diameter (nm)
200
(b) 15 nm particles
Interparticle distance (nm)
180
160 30 nm particles
60 nm particles
140
120
100
80
60
40
20
0
0.00 0.05 0.10 0.15 0.20
Volume fraction Nanopaticles
Fig. 25. (a) Surface area per unit volume vs. particle size for spherical particles that are ideally dispersed, and (b) interparticle
distance for spherical particles that are ideally dispersed [213].
region. For example, for 15 nm particles at 10 vol% loading of filler, the interparticle spacing is only
10 nm. Even if the interfacial region is only a few nanometers, very quickly the entire polymer matrix
has a different behavior than the bulk. Therefore, by controlling the degree of interaction between the
polymer and the nanofiller, the properties of the entire composites can be controlled. To achieve novel
properties of polymer composites, processing methods that lead to controlled particle size distribu-
tion, dispersion, and interfacial interactions are critical. Processing technologies for nanocomposites
are different from those for composites with micrometer-scale fillers, and new developments in nano-
composite processing have contributed to their recent success.
5.2.2. Dependence of dielectric properties of PMC on size of fillers

Fillers with different particle sizes are found to affect dielectric properties of polymer composites.
Kobayashi et al. [172] prepared BaTiO3/PVDF composites with different BaTiO3 particle sizes in nano-
meter scale. The dielectric permittivity and dissipation factor of composites increased with increasing
crystal size of BaTiO3 particle as shown in Fig. 26. Similar results were also obtained [216]. However,
Dang et al. used BaTiO3 particles having different sizes in micrometer range and obtained different re-
sults [40]. These BaTiO3/PVDF composites with different sizes and concentrations of BaTiO3 particles
were fabricated via a simple physical mixing and subsequently hot-press processing. In order to study
the effect of particle size on the dielectric properties of the SrTiO3/PEEK composite system, a compar-
ative study has been performed with nanosize and microsize SrTiO3 fillers with 27 wt% loading [217].
Table 3 compares the permittivity, loss tangent and temperature coefficient of permittivity of 27 wt%
micron- and nano-SrTiO3 filled PEEK composites. Interestingly, the permittivity, loss tangent as well as
the temperature coefficient of permittivity (Tkr) of the nanosized SrTiO3 filled PEEK composites are
higher than those of the microsized SrTiO3 filled composites. Because of the high surface area, polar-
izability also increases. This is because more charge can be accumulated on the grain boundary of the
nanosized powder. Hence, the increase in permittivity of the PEEK nanocomposite can be attributed to
the high polarizability of the nanosized SrTiO3 filler. However, the increase in loss tangent of the PEEK/
40 0.20
Dissipation Factor (-)

Dielectric Constant (-)
30 0.15
20 0.10
10 0.05
0 0.00
10 20 30 40
Crystal Size (nm)
Fig. 26. (a) Dielectric constant and dissipation factor of the BaTiO3/PVDF composite films measured at 104 Hz as a function of
BaTiO3 crystal size. BaTiO3 volume fraction: 30%, spin speed: 3000 rpm and drying temperature: 150 °C [172].
Table 3
Effect of particle size on the dielectric properties of 27 wt% SrTiO3 filled PEEK composites [217].
Powder characteristics k at 1 MHz tan d at 1 MHz Tkr (ppm/°C)

Micronsize powder 5.27 0.0037 12
Nanosize powder 5.9 0.0278 444
nano SrTiO3 filler at 27 wt% loading is believed to be caused by the lattice strain associated with the
reduction in particle size.
Presently, the dielectric properties of materials are often measured in a relatively high frequency
range. Therefore, it is difficult to reveal their dielectric properties of polymer composites with fillers
of different sizes at ultra-low frequencies. By employing a broad frequency high resolution dielectric
analyzer (Novocontrol), Dang et al. [41] found highly enhanced low-frequency dielectric permittivity
in BaTiO3/PVDF nanocomposites with 100, 200, 300, 400 and 500 nm BaTiO3 particles, respectively.
Fig. 27 shows the frequency dependence of the dielectric properties of the BaTiO3/PVDF nanocompos-
ites with fillers of different sizes at 60 vol% loading. It is clear that the composite with 100 nm particles
exhibits the maximal permittivity from 101 Hz to 102 Hz and the appearance of the loss tangent peak
is at about 102 Hz. Fig. 28 shows the effect of the filler size on the dielectric permittivity in the BaTiO3/
7000 400
1.6
(a) 350
60 vol% BT dBT= 100 nm
(b) 100 nm
dBT= 100 nm
6000 200 nm
100 nm 300 nm 200 nm

300
5000
400 nm
1.2 500 nm
300 nm
250 500 nm
Loss tangent
400 nm
4000 200 500 nm
150
3000 0.8
100
2000 50 2 3 4 5 6 7
10 10 10 10 10 10
1000 0.4
0 500 nm
-1 0 1 2 3 4 5 6 7 0.0 -1
10 10 10 10 10 10 10 10 10 0 1 2 3 4 5 6 7
10 10 10 10 10 10 10 10 10
Frequency (Hz) Frequency (Hz)
Fig. 27. Frequency dependences of (a) dielectric permittivity and (b) loss tangent of all BaTiO3/PVDF composites at room
temperature. The inset in (a) is the dependence of dielectric permittivity on frequency ranging from 102 to 107 Hz [41].
4
10
Freq.
-1
10 Hz
0
10 Hz
1
3
10 Hz
10 2
10 Hz
4
10 Hz
6
10 Hz
2
10
100 200 300 400 500 600

dBT (nm)
Fig. 28. Dependence of dielectric permittivity of all BaTiO3/PVDF composites on the BaTiO3 particle size at different frequencies
from top to bottom, 101, 100, 101, 102, 104, and 106 Hz [41].
PVDF nanocomposites. At low frequency, the dielectric permittivity decreases with increasing BaTiO3
particle size. The result could be explained by a marked increase in interfacial area and dipole polar-
izations at low frequency.
It can be seen that in the above mentioned examples, the dielectric permittivity of composites is
clearly dependent upon the size of fillers. Nevertheless, trend of the dependence is not always the
same. In some cases dielectric permittivity increased with increasing filler size and in some others
cases it follows an opposite trend. One of the reasons is that change in dielectric properties with size
in micrometer scale is not necessarily the same as that in nanometer scale. Another reason is the effect
of the matrix on the effective crystal size of particles inside the composite and the interactions be-
tween the particles and the polymer matrix. The concept of effective filler is very important. Most filler
size measurements are made on the powder samples before use. However, composite properties de-
pend on the filler morphology in the final composite which may be significantly different from the ini-
tial filler morphology because of the effects of compounding and molding. The extent to which fillers
are modified depends upon inherent filler properties and processing.
Percolation probability is closely related to the size of the conducting fillers in the composites. In
classic models [218] for bulk samples, the insulating matrix is considered to be infinitely large in three
dimensions compared with the size of filler. Therefore the influence of the filler size should be negli-
gible. However for practical application such as embedded capacitor, a film with a thickness in
micrometer is required to increase packaging density. In this case, classic models may not be valid,
especially when the filler size is comparable to film thickness. Thus for dielectric composites used
in embedded capacitor applications, the size of filler should be in nanometer to achieve high packag-
ing density. For conducting fillers, the effect of size and size distribution of fillers on dielectric prop-
erties has been studied [22]. For example, Lu et al. [22] reported that the size, size distribution and
loading level of in situ formed Ag nanoparticles in the Ag/CB/epoxy composites were found to have
significant influences on the dielectric properties of the nanocomposite system. Fig. 29 indicates that
the dielectric permittivity and dielectric loss of Ag/CB/epoxy composites with Ag nanoparticles of dif-
ferent sizes and size distributions varied considerably. A plausible explanation on those results may be
that the interfacial loss due to the newly induced interface is higher than the suppressed conduction
loss by incorporation of metal nanoparticles. Therefore, the overall dissipation factor is still increased
for samples with a higher loading. Table 4 summarizes some examples of the effect of filler size on
dielectric properties in various composition composites.
5.2.3. Effect of micro–nanosize cofillers on dielectric properties of PMC

To further improve the dielectric permittivity, the dense packing-density of ceramic/polymer com-
posites is important. Particles with different sizes were filled into the polymer at same time so that the
dielectric permittivity of the cofilled composites is higher than that of single-size filled composites at
2500
@ 10 kHz 1.4
(a1)
2000
1.2
Dissipation factor
1500
1.0
(b1)
1000
0.8
(b2)
500 0.6
(a2)
0 0.4
control sample 7 7.4 36.3 37
average size of large nanoparticles (nm)
Fig. 29. Effect of size of Ag nanoparticles on (a1) dielectric permittivity and (a2) dissipation loss of 4.2Ag/19.6CB/epoxy
composites (solid curves) and (b1) dielectric permittivity and (b2) dissipation loss of 6.2Ag/14.6CB/epoxy composites (dashed
curves) [22].
Table 4
Effect of filler size on dielectric properties in the different composition composites.
Materials k tan d Filler size Filler loading

BaTiO3/epoxy 40 (1 Hz) 0.035 (1 Hz) 100–200 nm 60 vol%
Pb(Mg1/3Nb2/3)O3–Pb TiO3/P(VDF–TrFE) 200 (10 kHz) 0.1 (10 kHz) 0.5 lm 50 vol%
Bimodal BaTiO3/epoxy 90 (100 kHz) 0.03 (100 kHz) 916 nm + 60 75 vol%
PMN–PT + BaTiO3/high-k epoxy 150 (10 kHz) N/A 900 nm/50 nm 85 vol%
BaTiO3/P(VDF–HFP) 37 (1 kHz) <0.07 (1 MHz) 30–50 nm 50 vol%
CB/epoxy 13,000 (10 kHz) 3.5 (10 kHz) 30 nm 15 vol%
Ag/CB/epoxy 2260 (10 kHz) 0.45 (10 kHz) Ag: 13 nm Ag: 3.7 wt%
CB: 20 wt%
Al/Ag–epoxy 160 (10 kHz) 0.045 (10 kHz) Ag: <20 nm Al:80 wt%
Ag/epoxy 300 (1 kHz) 0.05 (1 kHz) 40 nm 22 vol%
Ag@C/epoxy >300 (1 kHz) <0.05 (1 kHz) 80–90 nm core 25–30 vol%
the same concentration of fillers [33,40,42,219]. Cho et al. [33] reported the maximum dielectric con-
stant of embedded capacitor films using 0.916 lm BaTiO3 particle (P1) was only about 57. After finer
particles (0.06 lm in diameter (P2)) were added to 60 vol% P1, the powder packing density was further
90
P1 unimodal
80 P2 unimodal
Dielectric constant
Bimodal
70
P1(60vol%)+P2(5,10,15,20vol%)
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80
Powder contents (vol%)
(a) (b)
Fig. 30. (a) Dielectric constant changes with BaTiO3 powder loading. (b) Cross section image of ECFs containing bimodal
BaTiO3powders at 70 vol% (P1 60 vol% + P2 10 vol%) [33].
140 0.30 10
-1
(a) Double model composites (b)
vBT= 0.60 Single model composites
130 0.25 10
-2
Double model composites
Conductivity (S m )
-1
-3
120 10 Single model composites
0.20
-4
110 10
Tanδ
0.15
-5
100 10
0.10 -6
90 10 vBT=0.60
80 0.05 10
-7
-8
70 0.00 10
3 4 5 6 3 4 5 6 7
10 10 10 10 107 10 10 10 10 10
f (Hz) f (Hz)
Fig. 31. Dependence of (a) dielectric permittivity and loss and (b) conductivity on frequency for double model and single model
composites at 60 vol% loading [42].
improved because the finer particles could fill interstitial sites of coarser particles [220]. As a result, at
75 vol% bimodal powder loading, a dielectric constant of 90 was achieved as shown in Fig. 30a. Further
powder addition over 75 vol% at bimodal condition decreased dielectric constants of embedded capac-
itor films. Fig. 30b shows a cross section image of the embedded capacitor film containing 70 vol%
(60 vol% P1 + 10 vol% P2) bimodal BaTiO3 powders. Dang et al. [42] reported on theoretical prediction
and experimental study of dielectric properties in PVDF-matrix composites with micro–nanosize Ba-
TiO3 filler. The dielectric permittivity of composites with nanosize BaTiO3 is close to the predicted val-
ues at high concentrations of fillers. The enhanced dielectric permittivity as shown in Fig. 31 mainly
originates from a noticeably increasing phase interfaces in the composites due to micro–nanosize Ba-
TiO3 cofillers, which brings about a strong interfacial polarization effect at low frequencies. Compared
to the morphology of the single model BaTiO3/PVDF composite shown in Fig. 32, the stacking density
of the double model composite is higher than that of the single model composite because the nanosize
BaTiO3 occupies the spaces among the large BaTiO3 particles. The detailed change in dielectric prop-
erties of the BaTiO3/PVDF composites with different volume ratios of BaTiO3 fillers is also studied with
fillers of different sizes (1000, 700 and 100 nm) [219].
5.3. Shape effect of fillers on dielectric properties
5.3.1. Shape of fillers and filler–polymer connectivity

A general classification of fillers on the base of particle shape has been presented [221]. This clas-
sification is somewhat arbitrary and is based on the primary properties of particle size and surface
area, both of which being directly measurable and serve as systemizing filler functions. Shape of filler
can be divided on the basis of whether they are one, two or three dimensional. Along with size of filler,
shape of filler is decisive in determining dielectric properties of final composites. Filler with different
shapes has different surface areas resulting in different interfacial areas in the composites, which can
lead to different interfacial polarization and hence different dielectric properties. Moreover, different
shapes of fillers can lead to different types of connectivity in composites. A schematic of different
types of connectivities in high-k PMC based on filler shape are presented in Fig. 33. In general, the
0–3 connectivity composites could be formed when the fillers are sphere particles and 1–3 connectiv-
ity composites when the fillers are fiber or tube by means of blending technique.
Different types of fillers including spherical, cubic and fiber and flakes have been used in high-k
PMC. For both of percolative and non-percolative composite systems, spherical fillers are amongst
the most used fillers. Typical examples are ceramic and metal particles. For non-percolative systems
containing spherical particles, common mixing rules like those mentioned in Chapter 2 can be used to
explain the dielectric behavior of composites. However, when the dispersed particles are not spherical
in shape, the mixing formulas must be revised, in order to take into account their geometry. A typical
case is particles with a spheroidal or cylindrical geometry, characterized by the ratio of their axes. The
Fig. 32. Morphology of the fractured surfaces of (a) the double model composite and (b) the single model composite at 50 vol%
loading, respectively [42].
common way to include the information about such kinds of geometries of fillers is to properly intro-
duce a depolarization factor, which contains the information about the shape of particles in terms of
their deviation from sphericity. Thus, for example, the widely used Maxwell–Garnett equation is mod-
ified into the other forms mentioned in Chapter 2.
5.3.2. Effect of 1-dimensional fiber-shape fillers on dielectric properties

For percolative systems, the more the shape of filler deviates from spherical particles, the lower is
the percolation threshold. In other words, the high aspect ratio of fillers would induce a low percola-
tion threshold. The most common high aspect ratio fillers incorporated in composites are CNT and
CNF. CNT has typical diameter in the range of about 1–50 nm and length of many microns (even cen-
timeters in some cases). They can consist of one or more concentric graphitic cylinders. In contrast,
commercial (PAN and pitch) CF are typically in the 7–20 lm diameter range, while vapor-grown car-
bon fibers (VGCFs) have intermediate diameters ranging from a few hundred nanometers up to around
a millimeter. Crucially, conventional carbon fibers do not have the same potential for structural per-
fection that can be observed in CNTs. Indeed, there is a general question as to whether the smallest
CNTs should be regarded as very small fibers or heavy molecules, since the diameters of the smallest
nanotubes are similar to those of common polymer molecules. This ambiguity is characteristic of
nanomaterials, and it is not yet clear to what extent conventional fiber-composite understanding
can be extended to CNT-composites. SWNTs are usually obtained in the form of so-called ropes or
bundles, containing between 20 and 100 individual tubes packed in a hexagonal array [222–225].
Rope formation is energetically favorable due to the Van der Waals attractions or other interactions
between isolated nanotubes [226]. MWNTs provide an alternative route to stabilization. They consist
of two or more coaxial cylinders, each rolled out of single sheets, separated by the interlayer spacing in
graphite. The outer diameter of such MWNTs can vary between 2 and a somewhat arbitrary upper lim-
it of about 50 nm; the inner hollow core is often (though not necessarily) quite large with a diameter
commonly about half of that of the whole tube.
As mentioned in Section 5.1, when spherical conducting fillers are used to improve the dielectric
properties such as dielectric permittivity and conductivity, the percolation threshold is often high.
This is a big disadvantage for flexibility of polymer composites. Therefore, conducting fillers with high
aspect ratio are often used in order to have a low percolation threshold [49–56,74,121,123]. The mor-
phology of CNT has a remarkable effect on the dielectric properties of nanotube containing compos-
ites. For example, entangled nanotubes usually need higher concentrations to achieve the
percolation threshold while lower percolation threshold can be achieved with aligned nanotubes.
Higher aspect ratio of nanotube is an obvious advantage for their use in high-k composite. However,
in some cases higher concentrations tend to give rise to the entanglement of nanotubes destroying
their effectiveness. Yao et al. [55] studied the effect of aspect ratio of MWNT on the percolation thresh-
old as shown in Fig. 34. An increase in dielectric permittivity with increasing aspect ratio (AR) of
Fig. 33. Schematics illustrating five types of composites, their spatial arrangements and their representative volume elements.
(a) particulate (0–3) composites with particulates arranged at the vertices of a cube in space; (b) short-fiber (0–3) composites
with short-fibers arranged at the vertices of a cube in space; (c) long-fiber (1–3) composites with long circular fibers arranged in
a square array; (d) laminate (2–2) composites with the interface perpendicular to the 1-axis; and (e) networked (3–3)
composites with non-intersecting fibers arranged along three mutually perpendicular directions with fibers arranged in a
square array on each face [221].
MWNT was observed in the composites. Meanwhile, the change in percolation threshold was complex
and at the first glance surprising. The percolation threshold first increased with increasing AR, reached
a maximum at a certain AR and then decreased with further increasing AR. Entanglement of the nano-
tubes would explain the results. An increase in the entanglement of nanotubes results in an increase in
the percolation threshold. It is likely that the entanglement reduced the effective aspect ratio of the
nanotubes. As MWNT become more flexible with increasing AR, they may have strong tendency to
form interlocked inherent structures, like group of threads, and would not exhibit rod like rigidity any-
more. In spite of nanotube entanglement at higher concentration, MWNT provide enough conductive
paths to eclipse the effect of entanglement and would results in lower percolation threshold in prac-
tice. The same results were also obtained with SWNT composites [227].
0.021 2.0
3
(b) fc 10 Hz
s
Percolation Threshold (fc)

0.018
Critical Exponent (s)

1.5
0.015
1.0
0.012
0.5
0.009
0.006 0.0
0 200 400 600 800
Aspect Ratio(AR)
Fig. 34. Dependence of percolation threshold and critical exponent of MWNT/PVDF composites on the average AR value of
MWNT [55].
The morphology of CNTs in polymers depends very much on processing conditions, so as dielectric
properties. A transmission electron microscopy (TEM) image for PVDF composite with 2.0 wt% MWNT
shows that the MWNT were of uniform spherical particles with a diameter of several micrometers, in-
stead of rope-like structure. However, no distinct MWNTs were observed on the surface, indicating
that the MWNTs are wrapped by PVDF [53]. A similar phenomenon has been observed by Jin [228]
and Zhang [229]. For the PVDF prepared under the same conditions, however, no spherical particles
are seen. This corroborates that the MWNTs are responsible for the formation of the spherical parti-
cles. So far, the exact origin of the formation of the spherical particles is unknown. One possibility
is that when ultrasonic waves pass through the solution, a large number of microbubbles form, grow
and collapse during very short periods of time in a process called ultrasonic cavitations. Theoretical
calculations and corresponding experiments suggest that ultrasonic cavitations can generate a local
temperature of as high as 5000 K, local pressure of as high as 50.6 MPa, and heating and cooling rates
greater than 109 K/s [230]. Under such circumstances, it might be that there is interaction between
PVDF molecules and MWNTs. This interaction results in the formation of composites with a spherical
substructure by the combination of MWNTs with PVDF. The fact that MWNTs are wrapped by PVDF is
indicative of the interaction between PVDF molecules and MWNTs. In order to improve the interaction
between PVDF and MWNT and consequently high dielectric permittivity, Dang et al. [50] reported
very high dielectric permittivities for functionalized TFP-MWNT/electroactive-polymer nanocompos-
ites using MWNT that was chemically modified with 3,4,5-trifluorobromobenzene (TFBB). The dielec-
tric permittivity can reach 4000 when the volume fraction of TFP-MWNT is 0.15. It is also noted that
there is weak temperature dependence of permittivity from 0 to 100 °C for the composite containing
8.0 vol% MWNT. The very high dielectric permittivity originated from the Maxwell–Wagner–Sillars
(MWS) effect at percolation [231], and was attributed to a thin insulating layer of PVDF that was part
of the TFP-MWNT. The nomadic charge carriers were blocked at the internal interfaces between the
TFP-MWNT and the PVDF, and the large p-orbital of the MWNT provided the nomadic electrons with
large domains. Strongly electrophilic F groups on the TFP-MWNTs further reinforced the MWS effect.
The improved dielectric permittivity was also obtained using other types of conducting fillers with
high AR [74,123]. For example, Sui et al. [123] reported that PP nanocomposite containing 5.0 wt% car-
bon fibers (CNF) exhibited a surprisingly high dielectric permittivity under wide sweep frequencies
and low dielectric loss as well. Its dielectric permittivity was larger than 600 under low frequency,
and remained >200 at a frequency of 4000 Hz. Wang et al. [74] reported that the conducting PANI
nanofibers doped by protonic acids showed a high dispersion stability in P(VDF–TrFE) and led to perc-
olative nanocomposites with enhanced dielectric responses. About a 50-fold rise in the dielectric per-
mittivity of the ferroelectric polymer matrix has been achieved. Percolation thresholds of the
nanocomposites are related to the doping level of PANI nanofibers and can be as low as 2.9 wt% for
fully doped nanofibers. The interface between the conductive nanofiber and the polymer matrix plays
a crucial role in the dielectric enhancement of the nanocomposites in the vicinity of the percolation
threshold. Compared with other dopants, perfluorosulfonic acid resin is best at improving the perfor-
mance of the nanofibers in that it serves as a surface passivation layer for the conductive fillers and
suppresses leakage current at low frequency.
5.3.3. Effect of 2-demension plate-shape fillers on dielectric properties

Plate-like fillers such as graphite nanoplates are also associated with high dielectric permittivity.
PVDF composites of exfoliated graphite nanoplates (xGnP) having dielectric constant up to
4.5 107, have been reported at about 2.4 vol% xGnP loading. In fact, the percolation threshold for
the xGnP/PVDF composite system is very low (only about 1.0 vol%), which is of the same order of mag-
nitude as for MWNT/PVDF composites [51]. During the fabrication process graphite nanoplates tend to
lie down inside the PVDF matrix. As a result, the faces of the exfoliated graphite nanoplates were par-
allel to the nanocomposite-film plate. During a hot-press process, several solution-cast films were
stacked layer by layer in a ‘‘sandwich’’ form, and then pressed at 200 °C [117]. This promoted the for-
mation of a larger number of microcapacitors in the nanocomposites. As such, the graphite nanoplates
were parallel to each other, isolated by a polymer layer as a medium between the nanoplates. It is be-
lieved that the excellent dispersion of exfoliated graphite nanoplates and the existence of many mic-
rocapacitors in the PVDF matrix contribute to high dielectric permittivity.
5.3.4. Effect of core–shell fillers on dielectric properties

Fillers with core–shell microstructure are unique and are reported for many years. They could be
used in many fields such as catalyzers in chemical industry and novel drugs in biological engineering
[232–244]. As discussed in the above sections, the behavior of the dielectric properties of inorganic
filler/polymer composites is very complex. Very often, a remarkable improvement in the permittivity
is accompanied by an unacceptable dielectric loss due to the use of conducting fillers. To overcome the
issue, very recently fillers with core–shell structure have been dispersed into polymers. By tailoring
the thickness and characteristics of the core–shell fillers, the integrative dielectric properties of the
filler/polymer composites were improved [115,245–247]. Shen et al. [245] reported on a new ap-
proach, leading to both extremely stable high-permittivity and low dielectric loss of the polymer
dielectrics as shown in Fig. 35. Instead of directly adding dielectric ceramic or conductive fillers, they
used core/shell hybrid particles (denoted as Ag@C) with metal Ag cores coated by organic dielectric
shells as fillers. The organic dielectric shells not only act as interparticle barriers to prevent Ag parti-
cles from direct connection but also produce excellent compatibility between the fillers and the poly-
mer matrix and ensure the dispersion of fillers in the polymer matrix. Furthermore, Shen et al. [246]
studied the effect of interfacial layers on the dielectric properties of polymer nanocomposites filled
with core–shell structured Ag@C particles. By controlling the process conditions, four kinds of core/
Fig. 35. Variations of the dielectric permittivity, conductivity, and dielectric loss (see the inset) of the composite films
containing Ag@C particles with thicker shells or thinner shells with the volume fraction of particles at 103 Hz [245].
Fig. 36. Variations of (a) dielectric permittivity, (b) conductivity, and (c) dielectric loss with the volume fraction f of the Ag@C
nanoparticles at 1 kHz. The microscopy images in (a) indicate the representation of the symbols [246].
shell Ag@C nanoparticles were employed. The diameters of the silver cores were distributed over 60–
110 nm with an average diameter of 80–90 nm. Fig. 36 shows strong dependence of dielectric permit-
tivity, conductivity and dielectric loss of the Ag@C/epoxy composites on the thickness of C-shell layer
and the concentration of Ag@C fillers. The use of fillers with thin C-shell layer would result in the high
permittivity, low conductivity and low dielectric loss, indicating the significant role of the C-shell layer
in improving the dielectric properties of the he Ag@C/epoxy composites. Dang et al. [247] also pointed
out the effect of shell-layer thickness on the dielectric properties in Ag@TiO2 core@shell nanoparticles
filled ferroelectric PVDF composites. Therefore, it is an important way to adjust the dielectric proper-
ties of polymer composites by employing suitable core–shell structure particles. Core/shell nanopar-
ticles with metal cores and dielectric shells are ideal candidates to be used as fillers in percolative
composites as illustrated most recently. They provide not only a barrier layer between the conductive
fillers but also the possibility of tuning the dielectric properties. Easy processibility of core/shell nano-
particles has paved the way for the application of core/shell nanoparticles in percolative composites
[248,249].
Fillers are very important factors to affect the dielectric properties of the polymer composites.
In order to acquire the required dielectric properties, it is necessary to choose suitable fillers with
different physical–chemical characteristics. Dielectric properties can also be tuned by changing

the concentration of fillers, developing unique filling way (micro–nanosize cofilled) or employing
core–shell structure fillers.
6. Effects of measurement conditions on dielectric properties
Dielectric materials are often used in different electromagnetic environments [1,2,4,6]. It is found
that many factors have a decisive effect on the dielectric properties of polymer–matrix composites.
Within these factors, researchers often play more attention to environmental temperature, measure-
ment frequency and strength of applied electrical field. For new dielectric materials, it is important to
develop laws that describe the the variation of dielectric properties on the basis of three crucial factors
mentioned above.
6.1. Temperature dependence of dielectric properties
6.1.1. Organic fillers/polymer composites

Polymers are made of a large number of monomer units that are chemically bound together.
Macroscopical properties of polymers may change when temperature changes. Therefore, we may ex-
plore the effect of temperature on the dielectric properties of polymers and polymer–matrix compos-
ites. As a very important dielectric polymer, the change in the dielectric behavior of PVDF, its
copolymers and terpolymers is often studied due to their high-permittivity and remarkable electro-
strictive strain [19,20,24,28,29,34–37,40–42,64–73,250–253].
For example, P(VDF–TrFE–CFE/CTFE) terpolymers exhibit a very high electrostrictive strain of 4–5%
[251,252]. This has been ascribed to their relaxor-like structure [253], similar to electron-irradiated
P(VDF–TrFE) copolymers. A typical relaxor linear as well as nonlinear dielectric response has indeed
been detected in these polymer systems [254]. Fig. 37 shows a typical relaxor broad dispersive dielec-
tric maximum in the temperature dependence of the real part of the complex dielectric constant (k0 ) in
the P(VDF–TrFE–CFE) terpolymer. Contrary to ferroelectrics, relaxors do not undergo a symmetry-
breaking transition into a long-range ordered state on cooling in zero bias electric field. Rather, only
polar nanoregions persist down to the lowest temperatures. Therefore, the dispersive maximum in
k0 (T) does not denote a phase transition but the result of the fact that k0 , at a certain temperature that
depends on the experimental time scale, starts to deviate from its static value [254,255]. In spite of a
relatively large ionic conductivity, it can clearly be seen that values of the dielectric constant in the
P(VDF–TrFE–CFE) terpolymer does not exceed 50, and are much lower than those in inorganic piezo-
electric systems. However, after conducting PANI (c-PANI) particles at different concentrations are dis-
persed into the terpolymer matrix by a solution casting process, temperature dependences of k0 in the
Fig. 37. Temperature dependence of k0 measured at different frequencies in P(VDF–TrFE–CFE) terpolymer. The temperature
dependence of the static dielectric constant (ks) measured via charge accumulation technique, is also shown [250].
c-PANI/P(VDF–TrFE–CTFE) is shown in Fig. 38. It can clearly be seen that the addition of cPANI in-
creases the dielectric constant, particularly when the amount of c-PANI approaches the percolation
threshold. These two tends are related to the fact that the percolation is approaching [256]. However,
o-CuPc/P(VDF–TrFE) composites show a complicate temperature dependence of the dielectric proper-
ties, which might be associated with the microstructure of composite and physical characteristics of
o-CuPc fillers [257]. For the pure (61/29/10 mol%) P(VDF–TrFE–CFE) terpolymer, the dielectric permit-
tivity and dielectric loss display very small difference during heating and cooling processes. The peak
value (near room temperature) of the dielectric permittivity is greater than 50 at 100 Hz [258].
6.1.2. Insulating ceramic fillers/polymer composites

When PLZT [(Pb0.005La0.095)(Zr0.65Ti0.35)O3] ceramic particles are mixed with the P(VDF–TrFE)
copolymer, the paraelectric-to-ferroelectric phase transition and b-relaxation were detected [254].
The thermal hysteresis detected during cooling and subsequent heating runs and the typical Curie–
Weiss behavior of the dielectric strength. The dielectric response of the PLZT–P(VDF–TrFE) composite
is therefore very similar to that observed in the P(VDF–TrFE) copolymer. Although k0 and k00 values in
composite are due to the high values of the dielectric constant of PLZT ceramics which are higher than
those in the copolymer, the dielectric relaxation processes of the ferroelectric copolymer dictates the
dielectric response of the composite. An increase in the dielectric constant due to the PLZT admixture
Fig. 38. Temperature dependence of k0 , measured at several frequencies in cPANI/P(VDF–TrFE–CTFE) composite samples of two
different cPANI vol%, (a) x = 25.7 and (b) x = 12.2 [250].
800 Four Layers 800

Six Layers
P(VDF-TrFE)
Dielectric Constant
600 600
400 100 Hz 400

1 KHz
200 200
0 0
30 60 90 30 60 90
o
Temperature ( C)
Fig. 39. Dielectric constant at 100 Hz and 1 kHz vs. temperature for hot-press composites. For comparison, the dielectric
constant of the P(VDF–TrFE) 55/45 mol% copolymer prepared using the same process is also shown in the Fig. [117].
suggests that this composite might present an example of a piezoactive system whose dielectric prop-
erties could be relatively easily tuned by changing the PLZT/copolymer ratio [254]. Arbatti et al. [117]
reported on novel ceramic–polymer composites with high dielectric permittivity by dispersing CCTO
ceramic particles into a P(VDF–TrFE) copolymer. In order to acquire a homogeneous mixture, a mul-
tilayer stack of solution-cast films was packed in a ‘‘sandwich’’ figuration during the hot-press process.
Temperature dependence of the dielectric behavior of the multilayer composite samples is shown in
Fig. 39. Clearly, the dielectric constant reaches its maximum at ca. 70 °C, which is the transformation
temperature between the ferroelectric and paraelectric phases for the P(VDF–TrFE) 55/45 mol% poly-
mer matrix. It is well-known that the dielectric constant of PVDF-based ferroelectric polymers at the
ferroelectric-to-paraelectric phase-transition temperature is much higher than that at room temper-
ature [257]. Bai et al. [28] reported the temperature dependence of the dielectric permittivity of the
50 vol% PMN–PT/P(VDF–TrFE) composites can be tuned by changing the irradiated dose. Namely,
the temperature dependence of the dielectric permittivity of the composites becomes weak with
increasing irradiated dose, which is very useful for tuning the dielectric properties of ferroelectric
polymer–matrix composites.
In fact, the variation of the dielectric properties of the polymer–matrix composites is very complex
due to the effects of fillers with different physical and chemical characteristics. When the insulating
fillers are dispersed in polymers, the dielectric permittivity of the resulting composites does not
change much with temperature [43,260]. It is likely that the temperature had a weak effect on the
physical and chemical characteristics of fillers. When the temperature is changed from 50 to
150 °C, the dielectric permittivity of the BaTiO3/PI composite films prepared via in situ polymerization
exhibits weak temperature dependence [43]. For the sample with 40 vol% BaTiO3, the maximum
dielectric permittivity is 19.7, 18.8, 18.5, and 18.4 at 103, 104, 105, and 106 Hz, respectively. According
to the X7R EIA specification (EIA = Electronic Industries Association), the permittivity of a capacitor
should vary less than ±15% of that at 25 °C in the temperature interval from 55 to 125 °C. In this case,
the maximum dielectric permittivity between 20 and 30 °C was regarded as the 25 °C value of the
specification. As a result, the calculated relative tolerance of the dielectric permittivity in the temper-
ature range of 50 to 130 °C is 12.3%, 9.5%, 9.9% and 9.5% at 103, 104, 105 and 106 Hz, respectively.
However, the variation of the dielectric permittivity of a three-phase (BaTiO3–HA)/PVDF nanocompos-
ites does not support the above explanation [44]. Namely, the dielectric permittivity of the (BaTiO3-
HA)/PVDF nanocomposites has a remarkable improvement at about 100 °C. It is likely that the HA
phase plays a crucial effect on the dielectric properties of the composite system.
6.1.3. Semiconducting ceramic fillers/polymer composites

When semiconducting particles are employed as fillers, the dielectric permittivity of the polymer
composites displays often a remarkable increase [35,45,247,261], which can be attributed to the
Fig. 40. Temperature dependence of the dielectric permittivity at 1 kHz (a) the volume fraction is 0.21, 0.24, 0.19, and 0.18 for
the shell_8, shell_13, shell_60, and shell_120 cases [246]. (b) Temperature dependence of the dielectric permittivity at 1 kHz of
the composite films containing Ag@C particles with the thicker shells at different concentrations [245].
conductivity increase of semiconducting fillers with temperature. For example, Dang et al. [35]
reported a rescaled temperature dependence of the dielectric behavior of the LTNO/PVDF ferroelectric
polymer composites. The PVDF–LTNO composite with high conductivity LTNO ceramic particles also
exhibits a high dielectric constant. However, the dielectric loss tangent of this kind of composite is
always low. This result could be explained by complicate polarization such as interface polarization
interaction appearing between the PVDF and the LTNO phases. It also shows that the dielectric
properties of the LTNO/PVDF composites could be tailored by employing LTNO ceramic fillers with
different chemical stoichiometric amounts. The results from Dang’s group support this explanation
[45,247].
6.1.4. Conducting fillers/polymer composites

Unlike the fillers mentioned above, when metal particles are filled into polymers, the high dielec-
tric permittivity at the percolation concentration often decreases dramatically with an increase in
temperature [19], which might be attributed to the inevitable thermal expansion of polymer compos-
ites. It is possible that the thermal expansion results in a breakdown of conducting network structure.
The decrease in conductivity in the Ni/PVDF composites with temperature supports the effect of ther-
mal expansion. However, when conducting fillers with large aspect ratios, such as CNT and CF, are
mixed with polymers, the dielectric permittivity of the fiber-shape fillers/polymer composites in-
creases with increasing temperature. The reason might be that there is chemical bond interaction be-
tween fiber-shape fillers and the polymer matrix that blocks the thermal expansion of composites. For
practice applications of high dielectric permittivity materials, a weak temperature dependence of
dielectric permittivity of the polymer composites is very important. Therefore, how to decrease the
temperature dependent instability of dielectric permittivity is a very important challenge. Progresses
have been made in this direction. For example, Qi et al. [21] reported that when nanosize Ag particles
were dispersed in an epoxy resin, the Ag/epoxy nanocomposites exhibited a high dielectric permittiv-
ity (300) near the percolation threshold (23 vol% Ag). Furthermore, the dielectric permittivity had
higher temperature stability from 25 to 135 °C. The reason would be that the nanosize Ag particles
were prepared in the epoxy resin direct (in situ method). As such, the microstructure of the Ag/epoxy
resin composites became more resistant to temperature induced structure change. Nan’s group re-
ported on a different method to acquire the temperature stability of dielectric permittivity in metal
particles/polymer composites [245,246]. The use of Ag@C core/shell structure particles with a thin
(nanosize) organic shell layer, as shown in Fig. 40, resulted in remarkable temperature stability of
dielectric permittivity in the Ag@C/epoxy composites. Here the organic nanoshells served as electrical
barriers between the Ag cores to form a continuous interparticle-barrier-layer network and to retain a
stable high dielectric permittivity and low loss. Obviously, the use of nanosize metal particles and
their core–shell structure is efficient at tunning the dielectric properties of the conducting fillers/poly-
mer composites.
6.2. Frequency dependence of dielectric properties
As described in Chapter 2, the variation of dielectric properties of the materials always depends on
frequency because different molecule groups and structures have different response characteristics in
different frequency ranges. Therefore, the dielectric behavior of the composites consisting of different
components may display different frequency dependence.
6.2.1. Organic fillers/polymer composites

Two kinds of organic fillers, CuPc and PANI, are often used to prepare high-k polymer composites
[64–73,75,76,122,250–254,257–259]. CuPc and PANI are organic materials, their composites show
excellent flexibility and can be employed as electrostriction materials [64,68,73]. Temperature depen-
dence of the dielectric permittivity in these organic fillers/polymer composites is also strong. For
example, Zhang et al. [64] reported that although the dielectric permittivity of the 40 wt% CuPc/
P(VDF–TrFE) composites increased significantly compared to polymer matrix, as shown in Fig. 41a,
it also depended very strongly on frequency. A simple solution method would be effective at creating
strong interaction between CuPc and P(VDF–TrFE).
After the CuPc/P(VDF–TrFE) composite was irradiated, its permittivity was increased, especially at
50 °C as shown in Fig. 41b. However the dielectric permittivity decreased dramatically with temper-
ature. Based on this idea, surface modification of the CuPc particles and chemical graft methods were
employed to realize excellent composites. Wang et al. [262] studied a nanocomposite consisting of
PVDF as matrix and PCMS grafted with high dielectric constant CuPc (PCMS-g-CuPc) as filler. Com-
pared to the PVDF and CuPc/PVDF composites, the (PCMS-g-CuPc)/PVDF composite displayed a
marked increase in dielectric permittivity and a relatively weak frequency dependence of dielectric
permittivity. The significant enhancement of the dielectric response can be attributed to an enhanced
exchange coupling effect as well as the Maxwell–Wagner–Sillars polarization mechanism. Wang et al.
[65,66] also employed P(VDF–TrFE) and P(VDF–TrFE–CFE) as matrices, respectively, to study the effect
of graft and blending processes of the CuPc/polymer composites on their dielectric properties. They
found that compared with simple blend of the P(VDF–TrFE–CFE) and CuPc, a lower dielectric loss
and a higher breakdown field were achieved in the grafted nanocomposite. This could be attributed
to a more uniform distribution of o-CuPc particles in the polymer matrix as well as a much smaller
inclusion size. Moreover, the dielectric constant of the grafted nanocomposite at frequencies above
1 kHz is much higher than that of the simple blend, indicating an increased interface effect such as
the exchange coupling that can enhance the dielectric response.
By employing PU as polymer matrix, Huang et al. [70] reported on fully functionalized high-dielec-
tric-constant nanophase polymers with high electromechanical response. Three-component multi-
functional PANI-g-PolyCuPc-g-PU composites exhibit relatively weak frequency dependence. Such a
nanophase polymer has a strong interface effect which through exchange coupling markedly increases
the dielectric response to an extent that could exceed what would be expected form simple mixing
rules for dielectric composites, confirming a recent theoretical prediction [199]. Consequently, these
self-organized nanophase polymers exhibited high dielectric permittivities, reduced dielectric losses,
and improved breakdown fields and reliability, all of which are highly desirable for high-dielectric-
permittivity materials for actuator and dielectric applications.
When the organic conducting particles (PANI) are employed as fillers, the frequency dependence of
dielectric properties is often strong [74–76,250]. Bobnar et al. [250] reported on an enhanced dielec-
tric response in all-organic PANI/P(VDF–TrFE–CFE) composite. Although the conductivity at the lowest
frequencies approaches the direct current conductivity, the dielectric permittivity and conductivity
data are strongly remindful of the dielectric relaxation with the characteristic frequency of
104 Hz. Maxwell–Wagner effection – the charging of interfaces that occurs in electrically heteroge-
neous materials by itself leads to dielectric response with the frequency character being identical with
that of the Debye dipolar absorption. It is obvious that the pure terpolymer exhibits dielectric prop-
erties typical of inorganic relaxors, yet its dielectric constant is much lower. On the other hand, an
Fig. 41. (a) The weak-field (100 V/cm) dielectric permittivity as a function of frequency for the composite (squares) and the
polymer matrix (crosses). Data points are shown and solid curves are a guide to the eye [64]. (b) Weak-field dielectric
permittivity as a function of the frequency of the irradiated composite film with 40 wt% oligomer CuPc measured at 25 and
50 °C [257].
admixture of PANI, whose conductivity follows the Mott’s variable range hopping behavior, strongly
increases the dielectric constant of the terpolymer, especially when approaching the percolation
threshold. The frequency dependence of the composite’s dielectric response is tentatively described
in terms of the extended Maxwell–Wagner mixture approach. Wang et al. [74] reported on an
enhancement of the electrical properties of ferroelectric polymers by adding polyaniline nanofibers
with controllable conductivities. As shown in Fig. 42, the dielectric constant of the PANI–HCl nanofiber
(2.9 wt%)/P(VDF–TrFE) composites is 2–4 times higher than that of the polymer matrix at frequencies
above 0.1 MHz. The exchange coupling between the nanosized filler and P(VDF–TrFE) with a large dif-
ference in dielectric response or inhomogeneous field distribution in the matrix is likely responsible
for the observed dielectric enhancement [263].
It is common sense that the difference in the electron work function of the semiconductor and the
conductor results in a thin layer in the semiconductor close to the junction which is depleted of mobile
electron carriers. When the concentration of conductive PANI nanofibers randomly distributed in the
insulating ferroelectric polymer is close to the percolation threshold, the composites are semiconduc-
tive and the depletion regions may be located at the interface between the composites and metallic
electrodes. The occurrence of the depletion layer is also possible because of the partial overlap of
the nanofiber grain boundaries near the percolation threshold. When the PANI was doped with dode-
cylbenzene sulfonic acid (DBSA) and perfluorosulfonic acid (PFSA) resin, respectively, the Debye-like
relaxation of the PANI–PFSA nanofiber/P(VDF–TrFE) composite takes place at a lower frequency than
that of the PANI–HCl/P(VDF–TrFE) composite and effectively restrains the leakage current of the for-
mer. However, in comparison to the PANI–PFSA/PANI composite, the PANI–DBSA/PANI composite pre-
sents weak frequency dependence of dielectric permittivity.
6.2.2. Ceramic fillers/polymer composites

All kinds of ceramic fillers were used in order to acquire high-dielectric-permittivity polymer–ma-
trix composites [28–44,108–111]. Frequency dependence of the dielectric permittivity in the insulat-
ing ceramic fillers/polymer composites is relatively weak. The high-k of the composites is mainly from
Fig. 42. The dielectric constant, loss and conductivity as a function of frequency measured for the composites with 2.9 wt% of
PANI–HCl nanofibers. Inset: an equivalent circuit of the nanocomposites (R is resistor. C represents capacitor. Subscript c and b
denote the contact region and the bulk matrix, respectively) [74].
interface polarization and the high dielectric permittivity of ceramic fillers themselves, which have
weak frequency dependence. The slow decrease in dielectric permittivity might be related to the
high-frequency dielectric relaxation. For example, Kuo et al. [108] reported on the frequency depen-
dence of the dielectric properties for the BaTiO3/epoxy composites in the bulk form with different heat
treated ferroelectric ceramic fillers. The dielectric permittivity slightly decreases and the dielectric loss
slightly increases as the test frequency increases. It is has been reported that the loss tangent values
are high at low frequencies up to 1 kHz and are very low at frequencies from 10 kHz to 1 MHz. The
dielectric losses increase again at frequencies high than 1 MHz. The behavior of the dielectric loss
at low frequencies below 1 kHz is attributed to the domain motion and inhomogeneous conductivity,
and that at frequencies above 1 MHz to the domain-wall motion. Dang et al. [29,43] reported that the
dielectric permittivity of the BaTiO3/PVDF composites was weakly frequency dependent. If the BaTiO3/
PI nanocomposite was prepared by in situ polymerization, the frequency dependence of the dielectric
permittivity could be further decreased. This is because the in situ polymerization process can ensure
the strong interfacial interaction between nanosize BaTiO3 particles and polymer chains.
When semiconducting ceramic fillers are used to prepare polymer composites, their dielectric per-
mittivity improves markedly due to very high dielectric permittivity and relatively high conductivity
of the ceramics [37,45,264]. However, in most cases a remarkable frequency dependence of dielectric
properties can be observed. For example, Dang et al. [37] reported that the dielectric permittivity
reaches about 150 when the volume fraction of LTNO is 0.30 in the LTNO/PVDF composite. However,
the dielectric permittivity and dielectric loss depends strongly on temperature. For the CCTO/PI hybrid
films, when the concentration of CCTO is lower than 20 vol%, the permittivity of the hybrid film does
not change with an increase in frequency. However, when the concentration of CCTO is 30 vol%, the
permittivity has a weak dependence on frequency [45]. An obvious frequency dependence of the
dielectric permittivity can be observed when the concentration of CCTO is more than 40 vol%. Namely,
the permittivity decreases dramatically with an increase in frequency. This phenomenon could be also
ascribed to the semiconductive characteristic of the CCTO filler. Arbatti et al. [117] reported the dielec-
tric spectra of the hot-press multi-layer composite samples at room temperature, as shown in Fig. 43.
The dielectric constant increases with the initial number of layers used in the hot press and then levels
off with a further increase in that number. For example, the difference in the dielectric behavior be-
tween four-layer and six-layer composites is very small. The Cole–Cole plot for the dielectric behavior
of the composites at room temperature is shown in Fig. 43b. In addition to the major dielectric relax-
ation process, a low-frequency process was observed at the low-frequency end (left-up end of the
curve), which is responsible for the dielectric loss obtained at low frequency. This low-frequency pro-
cess can be caused by the conductive behavior of the composites or the relaxation process caused by
the heterogeneous nature of the composites (heterogeneous relaxation).
Fig. 43. (a) Dielectric constant vs. frequency from 100 Hz to 1 MHz for the composites at room temperature. (b) Cole–Cole plot
of the dielectric behavior for the hot-press composites at room temperature. The dotted lines are the fitted results using the
Cole–Cole equation [117].
6.2.3. Conducting fillers/polymer composites

Many research works have proven strong frequency dependence of dielectric permittivity in con-
ducting fillers/polymer composites [19–25,58,61,72–77,265,266]. Namely, the low-frequency dielec-
tric permittivity and dielectric loss are high in the composites, which would be attributed to the
crucial electrode polarization and low-frequency leakage current. For example, Panda et al. [265] re-
ported on weak frequency dependence of nanocrystalline Ni/PVDF composites over the whole fre-
quency range when the volume fraction of Ni was 0.25. When its concentration approaches the
percolation threshold, the variation of permittivity with frequency is more pronounced. When con-
ducting xGnP were dispersed in PVDF to form the xGnP/PVDF nanocomposites, the obvious frequency
dependence of permittivity was also reported [49], which followed the same law as for the use of CNF
[123] and CNT [50,51] as conducting fillers. In order to reduce the frequency dependence of the con-
ducting fillers/polymer composites, three-component (ceramic filler-conducting filler)/polymer com-
posites are designed [20,54,57,59,61,68,27,204,209]. For example, Dang et al. [20] reported that the
frequency dependence of dielectric permittivity was weakened when insulative ceramic BaTiO3 and
conducting Ni particles were mixed with PVDF together. Moreover, the (Ni–BaTiO3)/PVDF composite
had a low conductivity even if at low frequency (100 Hz). Another way to decrease the frequency
dependence of permittivity is to use core–shell structure particles, which has been proven to be effec-
tive method [245,246]. For example, Nan’s group studied percolative nanocomposites using core/
shell-structured particles as fillers. The Ag cores were coated with organic carbonaceous shells (de-
noted as Ag@C). The matrix is an epoxy resin. The thickness of the organic carbonaceous shell has a
very significant effect on the frequency dependence of the permittivity of the Ag@C/epoxy composite
The results of Dang’s group also confirm that the Ag@TiO2 core–shell particles with thin TiO2 layer re-
sults in the weak frequency dependence of dielectric permittivity and dielectric loss in the Ag@TiO2/
PVDF composites [247]. These results suggest that the strong frequency dependence of the dielectric
properties in conducting fillers/polymer composites can be lowered by changing the composition of
composites and employing the core–shell structure particles.
6.3. Electrical field dependence of dielectric properties
As introduced in Section 2.4, high-k polymer composites are often used as electric energy dielectric
materials. As shown in Eq. (9), the maximum energy storage density (wmax) of materials is directly
proportional to the dielectric permittivity and the square of the breakdown field. It is very desirable
if the polymer composites possess high dielectric permittivity and high breakdown field. It is well-
known that the dielectric permittivity of a material is relative to its electrical polarization. This drives
us to employ a strong electrical field to induce strong electrical polarization. However, the electrical
field is not stronger than the breakdown field of the material. As dielectric polymer composites, we
should investigate the electrical field dependence of the dielectric properties of inorganic/polymer
composites. Very recently, numerous studies have addressed this issue [12,64,67–
70,73,245,246,251,254,257]. Additionally, the high dielectric permittivity also helps to induce large
electrostriction of the polymer composites [64,67–70,73,252,254,257].
Zhang’s group reported on an electrical field induced high dielectric permittivity in the 40 wt%
CuPc/P(VDF–TrFE) composite prepared via a simple solution process [64]. The maximal dielectric per-
mittivity value reaches 450 at 10 kV/mm while only 220 without electrical field. The high dielectric
permittivity, therefore, contributes greatly to the electrical strain as shown in Fig. 44 [64]. Nan
et al. [245] shows that the applied electrical field does not significantly improve the dielectric permit-
tivity in the core–shell Ag@C/epoxy composites (Fig. 45a). However, the I–V characteristic of this com-
posite displays obvious electrical field dependence (Fig. 45b). A strongly nonlinear I–V relationship is
observed only in the composites where the volume concentration of fillers exceeds the percolation
threshold. Wang et al. [74] showed that the applied field had a significant effect on electrical polari-
zation of the PANI/P(VDF–TrFE) composites. Moreover, the polarization depends very much on the
doping composition in PANI. Hysteresis loops are typical characteristics of ferroelectric materials.
P(VDF–TrFE) nanocomposites with 2 wt% of PANI–HCl nanofibers display polarization hysteresis,
which gradually expands with increasing field. For the current composite, however, the breakdown
field still falls in the range to afford significant electromechanical response. Huang et al. [72] confirm
Fig. 44. (a) Dielectric properties of the composites as a function of the applied-field amplitude for the composite with 40 wt%
CuPc. (b) The strain amplitude as a function of the applied-field amplitude measured at room temperature. The applied-field
frequency is 1 Hz. Crosses are the data and the solid curve is a guide to the eye. For comparison, the strain from the
electrostrictive P(VDF–TrFE) copolymer at the same field range is also shown (the dashed curve) [64].
that PANI/P(VDF–TrFE–CTFE) composites with high dielectric permittivity near the percolation con-
centration of PANI fillers exhibit a large electrostriction strain under low applied electrical field. They
also prepared fully functionalized high-dielectric-permittivity nanophase polymers with high electro-
mechanical responses [70], which is very high compared with other types of electroactive polymers
[267,268]. The longitudinal or thickness strain is a 13% strain induced under 27 kV/mm, which is
currently is a very high strain value. However, it is lower that what is expected from the result
[86]. In fact, recent works show that the improvement of dielectric permittivity is not the sole method
to acquire a large electromechanical strain [86] and high electric energy storage density [269]. Carpi
and Rossi reported that after TiO2 powder was dispersed into a silicone dielectric elastomer, the
dielectric permittivity increased very slowly with increasing concentration in TiO2. However, the in-
crease in the applied electric field can induce a surprisingly high electromechanical strain. Zhang
et al. also reported that nanocomposites of ferroelectric polymers with TiO2 nanoparticles exhibited
significantly enhanced electrical energy density [269]. When the volume concentration of TiO2 nano-
particles increases, the dielectric permittivity of the TiO2/P(VDF–TrFE–CTFE) composites and the TiO2/
P(VDF–CTFE) composites shows very weak field dependences as shown in Fig. 46a. On the other hand,
Fig. 45. (a) Bias dependence of dielectric permittivity at 103 Hz of the composite films containing Ag@C particles with the
thicker or thinner shells. (b) I–V characteristics of these composite films, where the inset shows repeated measurements to the
I–V behavior of the 24 vol% Ag@C/epoxy composite with the thicker shells [245].
the electric energy density of the composites displays remarkable field dependence (see Fig. 46b).
Namely, the electric energy density of the composites increases with the applied field.
7. Applications for high-k PMCs
High-k polymer composites have a promising wide range of applications in microelectronics, elec-
trical engineering, biomedical engineering, etc., due to their electromechanical response. In what fol-
lows, we give outlines of these applications and expect that new applied fields will be developed in the
future.
7.1. Applications in microelectronic field
Recently high-k materials have received increasing interest due to various potential applications
such as gate dielectrics, high charge-storage capacitor and electroactive materials [17–33,270–274].
Fig. 47 schematically shows an example of realization of embedded passive technology by integrating
resistor and capacitor films into laminate substrates [275]. By removing these discrete passive com-
ponents from the substrate surface and embedding them into the inner layers of substrate board,
embedded passives cannot only provide the advantage of size and weight reduction, but also have
many other benefits such as increased reliability, improved performance and reduced cost, which have
driven a significant amount of effort during the past decade for this technology. However, embedded
passive technology has not been commercialized for electronic packages yet due to materials and pro-
cess issues. Therefore, it is necessary to develop materials that satisfy the requirements of fabrication
50 0.3
(a) 8 P(VDF-TrFE-CTFE)
2.5 vol% TiO2
7
Energy Density (J/cm2)
40 5 vol% TiO2
10 vol% TiO2
6
0.2 20 vol% TiO2
Loss Tangent
30 5 P(VDF-CTFE)-10 vol% TiO2
4
20
0.1 3
2
10
1
(b)
0 0.0 0
0 5 10 15 20 25 30 40 60 80 100 120 140 160 180 200 220 240
Volume Fraction of TiO2(%) Electric Field (MV/m)
Fig. 46. Dielectric permittivity and loss of the TiO2/P(VDF–TrFE–CTFE) (circle) and TiO2/P(VDF–CTFE) (square) nanocomposites.
(b) The stored energy density of the polymer and nanocomposites as a function of the applied field [270].
Fig. 47. Schematic illustration of embedded passives integrated into the laminate substrate [275].
Table 5
Summary of high-k ferroelectric ceramic/polymer composites candidates for embedded capacitor application.
Materials Dielectric Dissipation fractors Filler size Filler loading (vol%) Ref
permittivity
BaTiO3/epoxy 40 (1 Hz) 0.035 100–200 nm 60 [279]
PZT/PVDF 50 20 lm 50 [116]
PMN–PT/P(VDF–TrFE) 200 (10 kHz) 0.1 (10 kHz) 0.5 lm 50 [28]
Bimodal BaTiO3/epoxy 90 (100 kHz) 0.03 (100 kHz) 916 nm + 60 nm 75 [253]
PMN–PT + BaTiO3/epoxy 150 (10 kHz) 900 nm/50 nm 85 [280]
BaTiO3/P(VDF–HFP) 37 (1 kHz) <0.07 (1 MHz) 30–50 nm 50 [281]
CCTO/P(VDF–TrFE) 243 (1 kHz) 0.26 (1 kHz) 50 [117]
CCTO/PI 49.1 (1 kHz) 0.2 (1 kHz) 1–4 lm 40 [45]
LTNO/PVDF 90 (1 kHz) 40 [37]
Table 6
Summary of high-k conductive filler/polymer composites candidates for embedded capacitor application.
Materials Dielectric constant Dissipation factors Filler size Filler loading Ref
Ag flake/epoxy 1000 (10 kHz) 0.02 (10 kHz) 1.5 mm 11.23 vol%
Ag@C/epoxy >300(1 kHz) <0.05 (1 kHz) 80–90 nm 25–30 vol% [245]
Al/epoxy 109 (10 kHz) 0.02 (10 kHz) 3 mm 80 wt% [282]
Ni–BaTiO3/PVDF 300 (10 kHz) 0.5 (10 kHz) Ni: 0.2 mm Ni:23 vol% [20]
BaTiO3: 1 lm BaTiO3: 20 vol%
Ni–BaTiO3/PMMA 150 (1 MHz) Ni: 4 mm Ni: 12 vol% [283]
BaTiO3: 1 mm BaTiO3: 20 vol%
CB/epoxy 13,000 (10 kHz) 3.5 (10 kHz) 30 nm 15 vol% [284]
Al–Ag/epoxy 160 (10 kHz) 0.045(10 kHz) Al: 3 mm Al: 80 wt% [285]
Ag: <20 nm
Ag/epoxy 300 (1 kHz) 0.05 (10 kHz) 40 nm 22 vol% [21]
Ag-CB/epoxy 2260 (10 kHz) 0.45 (10 kHz) Ag: 13 nm Ag: 3.7 vol% [22]
CB: 30 wt%
as well as electrical and mechanical performances to enable embedded passive technology to be com-
mercialized. Materials challenges include dielectrics with k above 1000 and low dielectric loss. A very
low dissipation factor is desired for radio frequency applications to avoid signal losses. However, much
higher values can be tolerated for decoupling applications.
Ceramics are traditionally used high-k materials for capacitor applications. However, they also
have many prominent disadvantages. They either require expensive tools or cannot be easily imple-
mented over large PCB substrate areas. Moreover, high temperature annealing or sintering
(>500 °C) is often needed to improve the quality of deposited ceramic dielectric films. Such a high
temperature annealing process will destroy the organic PCB board, as the organic board needs to be
Fig. 48. Metal–insulator–metal (MIM) capacitor fabrication processes using ECFs on PCBs. (a) ECF lamination on a PCB
substrate, (b) releasing film removal, and (c) top electrodes deposition by a sputtering method [33].
processed under 250 °C. On the other hand, polymers have low cost and can be easily and compatible
PCB process whereas their dielectric permittivity typically is low. Polymer–ceramic nanocomposites
may have the potential to introduce high dielectric constant with low temperature thick-film fabrica-
tion process, because they can combine the advantages of high dielectric constant from the ceramic
part and good processibility from the polymer part. Because of the combined advantages of poly-
mer–ceramic composites, they have been subjected to extensive research and have been selected as
a major material candidate for the development of embedded capacitors inside organic substrates
[276,277].
In most cases even at very high filler concentrations the dielectric constant of polymers does not
exceed above 100. For embedded capacitor technology polymer composites should have a dielectric
constant of 50–200 to make layout area small enough for embedded capacitors in many cases such
as for decoupling. To achieve dielectric constants above 50, high filler loadings around 60 vol% have
to be used, which drastically deteriorates the peel adhesion strength and thermal stress reliability
of the high-k composites, inhibiting their real application as an embedded capacitor dielectric. To
make the high-k composites useful, a lower volume filler loading (650 vol%) has to be used, but this
in turn will reduce the dielectric constant of the composites. Therefore, fundamental understanding of
the adhesion and thermal stress reliability of dielectric films needs to be addressed in order to obtain
high dielectric permittivity (k > 50) embedded capacitor dielectrics that can successfully pass the
adhesion and thermal stress reliability tests.
While ceramic/polymer composites have shown some successful commercial applications, the
dielectric permittivity of ceramic/polymer composites is not sufficiently high (k < 100) for many appli-
cations in the next generation electronics [1–4]. To address this issue, disruptive innovation on poly-
mer high-k composites has also been explored. An approach showing good potential is the high-k
polymer-conductive filler composites based on percolation mechanism. According to scaling theory,
high dielectric permittivity can be obtained when the actual filler loading is very close to the perco-
lation threshold [18–23,50–60,68,72–75,188,212,278]. Because the filler loading used in the percola-
tive composites is lower than ceramic/polymer composites, better material processibility and
adhesion strength can be expected. However, due to its percolation nature, the electric and dielectric
properties of a percolative composite are very sensitive to the composition. A slight change in the
composition will lead to a significant change of materials properties, which impose serious challenge
of producing materials with reproducible properties. In addition, the dielectric loss (tan d) of a perco-
lative composite usually is high, which is not desirable for high frequency applications. In order to ap-
ply percoaltive systems for manufacturing of capacitors, fundamental understanding of the dielectric
loss of the polymer-conductive filler composites needs to be obtained, and a viable method to obtain a
percolative composite with reproducible high dielectric constant and low dielectric loss needs to be
developed. Here high-k ferroelectric ceramic/polymer composites and high-k conductive filler/poly-
mer composites candidates for embedded capacitor application are summaried in Tables 5 and 6.
Now let us to show how to prepare embedded capacitors with examples. Cho et al. [33] reported on
the fabrication of embedded capacitor films (ECF) using a comma roll coater. Compared with a spin
coating process for the deposition of BaTiO3/epoxy composite films, the comma roll coating method
has advantages such as no waste of materials, good coating thickness uniformity, and high productiv-
ity. The ECF coated on the releasing film was dried sequentially before rewinding. As an example,
Fig. 49. Principle of dielectric elastomer generators: (a) dielectric elastomer stretched (charge at low voltage) and (b) dielectric
elastomer contracted (charge at high voltage).
using BaTiO3/epoxy composite ECFs, metal–insulator–metal (MIM) capacitors were fabricated on PCB.
The capacitor fabrication process is shown in Fig. 48. Kakimoto et al. [3] also reported on ceramic/
polymer nanocomposites based on new concepts, which could be developed for embedded capacitor
applications. The dielectric constant was above 80 at 1 MHz and the specific capacitance was as high
as 8 nF/cm2. By use of this nanocomposite, multilayer printed wiring boards with embedded passive
components were fabricated for prototypes.
7.2. Applications in electrical engineering field
High-k polymer composites as one of novel functional materials have a great potential application
in electrical engineering. For example, high elastic rubber–matrix composites with high dielectric per-
mittivity are attracting great attention due to their easy, low-temperature processing and flexibility,
especially in electrical engineering application such as potential cable accessories [85,263,286–291].
This is because they could balance the distribution of electric field of cable termination to prevent
the cable from failure, as shown in Fig. 3. In this field, one has to explore flexible rubber–matrix com-
posites with high dielectric permittivity (>100) and low dielectric loss (<0.05). Up to now, it is difficult
to obtain flexible two-component rubber–matrix composites that meet all properties mentioned
above when single functional filler is employed. For example, one has to increase the concentration
of ceramics in order to improve the dielectric permittivity of two-component rubber–matrix compos-
ites. This results in a concomitant decreases in their elasticity [287]. However, conducting fillers, such
as metal, carbon black, carbon fiber, and carbon nanotube are used, the dielectric permittivity of the
composites increases remarkably when the concentration of conducting fillers is close to the percola-
tion threshold. In this case, the dielectric loss of the composites always increases dramatically and
their resistivity decreases sharply. Therefore, the resolution of the problems above is essential if flex-
ible rubber–matrix composites are applied as electric cable accessories. Based on this viewpoint, Dang
et al. [61] reported high-dielectric-permittivity three-component high-elasticity nanocomposites with
low percolation threshold and low dielectric loss.
Electrical energy storage plays also a key role in mobile electronic devices, stationary power sys-
tems, hybrid electric vehicles, and pulse power applications [292,293]. In particular, there is a growing
need for capacitors that can accumulate a large amount of energy and then deliver it nearly instanta-
neously. This kind of ‘‘pulse power’’ is needed for a variety of military and commercial applications.
Over time, these applications demand ever higher energy and power densities as well as higher rate
capability [294]. Barber et al. [294] showed a complication of reported and calculated energy densities
from the literature for the ceramic/polymer composites. However, the results cannot give enough
guidance for practical application.
A new application for dielectric elastomers with high dielectric permittivity is to produce power
energy because dielectric elastomers can be run in reverse of their usual actuator operation to imple-
ment mechanical-to-electrical transition. In the generator mode of operation, charges are placed on
the elastomer film in the stretched state. Upon contraction, the increase in thickness of the elastomer
increases the separation of opposite charges, and the decrease in planar area compresses like charges.
Both changes increase the voltage of the charge, and hence increase the stored electrical energy.
Dielectric elastomer generator films have shown very good performance with a generator energy den-
sity of up to 0.4 J/g. More importantly, dielectric elastomers have unique advantages compared with
competitive electromagnetic generators. In particular, dielectric elastomers have excellent low and
variable frequency response, and together with their high energy density are well suited for direct
drive designs that eliminate the need for costly and complicated transmissions. Fig. 49 illustrates
the elementary generator mode of dielectric elastomers.
Dielectric elastomers (DEs) can be used for traditional point generator applications as a direct
replacement for electromagnetic generators. A dielectric elastomer generator can be coupled to an
internal combustion engine to make a fuel-powered generator, for example. For these traditional point
generator applications, dielectric elastomers may offer advantages such as lower cost, lighter weight,
or smaller size. However, for traditional high frequency, point generator applications, dielectric elas-
tomer generators are a new technology competing against a mature one (electromagnetics) that is
well suited for the task. Hence, the competitiveness of the dielectric elastomers in these traditional
high frequency applications may depend a great deal on the specifics of the application. It is worthy
noting that dielectric elastomers can be made in large area sheets with high performance. Even when
not packaged in a sheet form, dielectric elastomers use potentially low-cost materials. These capabil-
ities suggest that dielectric elastomers may be well suited for harvesting energy from large area, dis-
tributed power sources such as wind and wave energy.
7.3. Applications in biomedical fields
High-k PMC can act as electromechanical transducers that convert or transduce electrical energy to
mechanical energy. In an actuator mode, these transducers convert electrical to mechanical energy,
whereas in a generator mode they perform the reverse function and convert mechanical to electrical
energy. Flexible electroactive polymers (EAPs) can behave as actuators, changing their shape in re-
sponse to electrical stimulation. EAPs that are controlled by external electric fields—referred to here
as field type EAPs—include ferroelectric polymers, electrostrictive polymers, dielectric elastomers
and liquid crystal polymers. Field-type EAPs can exhibit fast response speeds, low hysteresis and strain
levels far above those of traditional piezoelectric materials [252,253,268,295,296] with elastic energy
densities even higher than those of piezoceramics. However, these polymers also require a high field
(>70 V mm) to generate such high elastic energy densities (>0.1 J/cm3). Zhang et al. [64] developed a
new class of all-organic field-type EAP composites, which can exhibit high elastic energy densities in-
duced by an electric field of only 13 V/lm. The composites can exhibit high net dielectric constants
while retaining the flexibility of the matrix. These all-organic actuators could find applications as arti-
ficial muscles, ‘smart skins’ for drag reduction, and in microfluidic systems for drug delivery [297].
Colloquially referred to as ‘artificial muscles’, actuators based on dielectric elastomers are uniquely
suited to orthotic and prosthetic development in biomechanics due to their rough similarity in func-
tion to natural muscle [298].
Based on the electrostriction of the high-k polymer materials, Xu et al. [299] studied the perfor-
mance of micromachined unimorph actuators (polymer micromachined actuator PMAT) based on an
electrostrictive P(VDF–TrFE) copolymer. Because of the high electrostrictive strain and high elastic
energy density (1 J/cm3) of the active polymer, the PMAT exhibits a very high stroke level with
high load capability and high displacement voltage ratio. At the same time, they also reported on
an electroactive ceramic/polymer hybrid actuation system for enhanced electromechanical perfor-
mance [300]. Xia et al. [301] reported an electroactive polymer based microfluidic pump, which
was realized by integrating a nozzle/diffuser type fluidic mechanical-diode structure with the poly-
mer microactuator. Additionally, Ren et al. [302] developed a compact electroactive polymer actua-
tor suitable for refreshable Braille display. The compact polymer actuator was developed utilizing
the electrostrictive terpolymer, which is suitable for full page Braille display and graphic display.
Key issues related to the reliability of electroactive polymers used in the compact actuators and
for the mass fabrication of these polymer actuators were investigated. The recent results demon-
strate that the EAP Braille actuator meets all the functional requirements of actuators for refreshable
full Braille display, which offers compact size, reduced cost and weight. However in 2000, a wheel-
based Braille display was reported which can significantly reduce the number of actuators to move
Fig. 50. Outlook for applications of the high-k polymer composites.

the dots in a refreshable display and hence, can reduce the cost of the Braille display markedly
[303].
Pelrine et al. [304] also reported on high-strain actuator materials based on dielectric elastomer.
Upon applying a voltage, the active portion of the elastomer expands and the strain can easily be mea-
sured optically. The active portion expands when voltage is applied and the strain is easy to measure
optically. Kofod et al. [305] presented energy minimization for self-organized structure formation and
actuation. Here we must stress if the suitable structure to actuators is designed, the flexible actuators
can be made into microfluidic pump behaving as ‘artificial heart’, which can be used in biomedical
field.
Generally speaking, the high-k polymer composites have a resplendent outlook from the work re-
ported in this review. The significant strides have been made in the development of polymer compos-
ites with greater energy storage capacity, higher energy density and other excellent characteristics. Of
course, due to the limitation of this review, we cannot show full applications of the high-k polymer
composites. A simply outlook for applications of the high-k polymer composites is shown in Fig. 50.
8. Concluding remarks and future perspective
The present paper provides a review on the fundamentals, processes and applications of high-k
polymer–matrix composites. The concluding remarks are as follows:
(1) Existing theoretical models for predicting the dielectric properties of polymer–matrix compos-
ites are presented. None of them explains, describes and/or predicts reasonably well the current
results due to complexities in microstructure and composition of such composites. For a given
composition, the dielectric properties can be significantly different, depending on the prepara-
tion process employed.
(2) Interfaces between inorganic fillers and polymers are crucial. Appropriate surface modification
of inorganic fillers can improve their compatibility with polymers. Inclusion of core–shell struc-
ture fillers is also an important way to promote interfacial interactions between filler and poly-
mer and dispersion of the fillers in the polymers. This improves the dielectric permittivity of the
composites. Better filler–polymer compatibility also improves the stability of the micro-struc-
ture of the composite and increases the specific surface area of the filler. It may also minimize
defects or voids in the composite that can deteriorate the breakdown strength and thus the
overall energy density [294].
(3) Ceramic/polymer composites often have the advantage of exhibiting weak frequency and tem-
perature dependence of dielectric permittivity. However, their dielectric permittivity is often
low even if a high loading of ceramic particles is incorporated. Moreover, the breakdown
strength is weak if ceramic particles are not surface treated.
(4) Conducting fillers/polymer composites display high dielectric permittivity near the percola-
tion threshold. The variation in dielectric permittivity of such composites is very sensitive
to the concentration of conducting fillers. It is also sensitive to temperature due to thermal
expansion. Three-phase composites consisting of ceramic particles, conducting filler and poly-
mer can be good candidates for acquiring good dielectric properties as well as other
characteristics.
(5) The size, shape and concentration of fillers have important effects on dielectric properties. How-
ever, the effects of the size and shape of fillers have yet been explored to a sufficient degree.
High-k polymer composites with weak frequency/temperature dependence are crucial for prac-
tice applications. Good solutions are still not available for effectively tuning the frequency/tem-
perature dependence of the dielectric permittivity of polymer–matrix composites.
(6) Many studies have been carried out on the use of high-k polymer composites as dielectric layers
in embedded capacitors and pulse power energy storage capacitors. Flexible high-k polymer
composites can also be used as biomedical devices owing to their great electrical mechanical
response.
Based on the state of the art, future perspectives are as follows:
(1) It is necessary to refine existing models or develop new ones for dielectric composites.
(2) Much more attention should be paid to nanosize fillers. They may create much higher
amounts of polymer–filler interfaces and consequently generates much more significant
polarization than microsize ones. Controlled dispersion of nanofillers is a very challenging
issue [306].
(3) It is worthywhile to explore the synergy between different types of fillers and the potential
of core–shell fillers in order to attain high-k polymer composites with other excellent prop-
erties [307,308]. Surface design and modification of fillers are a very important issue.
(4) It is important to develop high-k materials with weak frequency/temperature dependence
of dielectric permittivity, especially when they are used in special environment [309].
(5) Fabrication of electronic devices with high-k polymer composites can be very interesting as
it can significantly reduce their weights and volumes. High-k polymer composites may also
find applications in biomedical field.
Acknowledgements
This work was financially supported by NSF of China (No. 50977001, 51073015), the Ministry of
Sciences and Technology of China through 863-Project (No. 2008AA03Z307), the Ministry of Sciences
and Technology of China through China-Europe International Incorporation Project (No.
2010DFA51490), State Key Laboratory of Power System (No. SKLD09KZ03), State Key Laboratory of
Electrical Insulation and Power Equipments (No. 09201), Program for New Century Excellent Talents
in University (NCET), and the Fundamental Research Funds for the Central Universities (No.06103012,
06103011).
References
[1] Tummala RR. Electronic packaging for high reliability, low cost electronics. Kluwer Academic Publishers; 1999.
[2] Lau J. Chip on board technologies for multichip modules. Kluwer Academic Publishers; 1994.
[3] Kakimoto M, Takahashi A, Tsurumi T, Hao J, Li L, Kikuchi R, et al. Polymer–ceramic nanocomposites based on new
concepts for embedded capacitor. Mater Sci Eng B 2006;132:74–8.
[4] Kapadia H, Cole H, Saia R, Durocher K. Evaluating the need for integrated passive substrates. Adv Microelectron
1999;26:12–6.
[5] Bar-Cohen Y. Electroactive polymers as artificial muscles: a review. J Spacecraft Rockets 2002;39:822–7.
[6] Uchino K. Piezoelectric actuators and ultrasonic motors. Kluwer Academic Publisher; 1997.
[7] Chahal P, Tummala RR, Allen MG, Swaminathan M. A novel integrated decoupling capacitor for MCM-L technology. IEEE
Trans Compon Packag Manufact Technol B 1998;21:184–93.
[8] Bhattacharya SK, Tummala RR. Next generation integral passives: materials, processes, and integration of resistors and
capacitors on PWB substrates. J Mater Sci 2000;11:253–68.
[9] Prymark J, Bhattacharya S, Paik K, Tummala RR. Fundamentals of microsystems packaging. New York: McGraw-Hill; 2001.
[10] Ulrich RK, Schaper LW. Integrated passive component technology. Piscataway (NJ): IEEE Press; 2003.
[11] Bar-Cohen Y. Electroactive polymer actuators as artificial muscles. Bellingham (WA): SPIE; 2001.
[12] Zhang QM, Furukawa T, Bar-Cohen Y, Scheinbeim J. Electroactive polymers. In: MRS Symp Proc, Warrendale, PA, 2000. p.
600.
[13] Nalwa H. Ferroelectric polymers. NY: Marcel Dekker, Inc.; 1995.
[14] Dario P, Carrozza M, Benvenuto A, Menciassi A. Microsystems in biomedical applications. J Micromech Microeng
2000;10:235–44.
[15] Immergut EH, Brandrup J, McDowell W. Polymer handbook. New York: Wiley; 2008.
[16] Popielarz R, Chiang CK, Nozaki R. Obrzut dielectric properties of polymer/ferroelectric ceramic composites from 100 Hz to
10 GHz. Macromolecules 2001;34:5910–5.
[17] Rao Y, Wong CP. Material characterization of high dielectric constant polymer–ceramic composite for embedded
capacitor to RF application. J Appl Polym Sci 2004;92:2228–31.
[18] Pothukuchi S, Li Y, Wong CP. Development of a novel polymer–metal nanocomposite obtained through the route of in situ
reduction for integral capacitor application. J. Appl Polym Sci 2004;93:1531–5.
[19] Dang ZM, Lin YH, Nan CW. Novel ferroelectric polymer composites with high dielectric constants. Adv Mater
2003;15:1525–9.
[20] Dang ZM, Shen Y, Nan CW. Dielectric behavior of three-phase percolative Ni–BaTiO3/polyvinylidene fluoride composites.
Appl Phys Lett 2002;81:4814–6.
[21] Qi L, Lee BI, Chen S, Samuels WD, Exarhos GJ. High-dielectric-constant silver–epoxy composites as embedded dielectrics.
Adv Mater 2005;17:1777–82.
[22] Lu J, Moon KS, Xu J, Wong CP. Synthesis and dielectric properties of novel high-k polymer composites containing in-situ
formed silver nanoparticles for embedded capacitor applications. J Mater Chem 2006;16:1543–8.
[23] Dang ZM, Peng B, Xie D, Yao SH, Jiang MJ, Bai J. High dielectric permittivity silver/polyimide composite films with
excellent thermal stability. Appl Phys Lett 2008;92:112910.
[24] Dang ZM, Zhang YH, TJong SC. Dependence of dielectric behavior on the physical property of fillers in the polymer–matrix
composites. Synthetic Met 2004;146:79–84.
[25] Li YJ, Xu M, Feng JQ, Dang ZM. Dielectric behavior of a metal–polymer composite with low percolation threshold. Appl
Phys Lett 2006;89:072902.
[26] Ang C, Yu Z, Guo R, Bhalla AS. Calculation of dielectric constant and loss of two-phase composites. J Appl Phys
2003;93:3475–80.
[27] DevaraJu NG, Lee BI. Dielectric behavior of three phase polyimide percolative nanocomposites. J Appl Polym Sci
2006;99:3018–22.
[28] Bai Y, Cheng ZY, Bharti V, Xu HS, Zhang QM. High-dielectric-constant ceramic-powder polymer composites. Appl Phys
Lett 2000;76:3804–6.
[29] Dang ZM, Wang HY, Zhang YH, Qi JQ. Morphology and dielectric property of homogenous BaTiO3/PVDF nanocomposites
prepared via the natural adsorption action of nanosized BaTiO3. Macro Rap Commun 2005;26:1185–90.
[30] Dias CJ, Das-Gupta KK. Inorganic ceramic/polymer ferroelectric composite electrets. IEEE Trans Dielect Electr Insul
1996;3:706–10.
[31] Rao Y, Ogitani S, Kohl P, Wong CP. Novel polymer–ceramic nanocomposite based on high dielectric constant epoxy
formula for embedded capacitor application. J Appl Poly Sci 2002;83:1084–9.
[32] Bidstrup-Allen SA, Tanikella RV, Kohl P. Novel low-temperature processing of polymer dielectrics on organic substrates by
variable frequency microwave processing. IEEE Trans Adv Pack 2002;23:313–7.
[33] Cho SD, Lee JY, Hyun JG, Paik KW. Study on epoxy/BaTiO3 composite embedded capacitor films (ECFs) for organic
substrate applications. Mater Sci Eng B 2004;110:233–7.
[34] Dang ZM, Wang HY, Xu HP. Influence of silane coupling agent on morphology and dielectric property in BaTiO3/
polyvinylidene fluoride composites. Appl Phys Lett 2006;89:112902.
[35] Dang ZM, Wang L, Wang HY, Yin Y, Tjong SC, Xie D, et al. Rescaled temperature dependence of dielectric behavior in
ferroelectric PVDF–LTNO composites. Appl Phys Lett 2005;86:179205.
[36] Dang ZM, Nan CW. Dielectric properties of LTNO ceramics and LTNO/PVDF composites. Ceram Int 2005;31:349–51.
[37] Dang ZM, Wu JB, Fan LZ, Nan CW. Dielectric behavior of Li and Ti co-doped NiO/PVDF composites. Chem Phys Lett
2003;376:389–93.
[38] Dang ZM, Yu YF, Xu HP, Bai J. Study on microstructure and dielectric property of the BaTiO3/epoxy resin composites.
Compos Sci Technol 2008;68:171–7.
[39] Dang ZM, Zheng Y, Xu HP. Effect of ceramic particle sizes on morphology and dielectric properties in the BaTiO3/
polystyrene composites. J Appl Polym Sci 2008;111:3473–9.
[40] Dang ZM, Wang HY, Peng B, Nan CW. Effect of BaTiO3 size on dielectric property of BaTiO3/PVDF composites. J
Electroceram 2008;21:381–4.
[41] Dang ZM, Xu HP, Wang HY. Significantly enhanced low-frequency dielectric constant in BaTiO3/poly(vinylidene fluoride)
nanocomposite. Appl Phys Lett 2007;90:012901.
[42] Dang ZM, Xie D, Shi CY. Theoretical prediction and experimental study of dielectric properties in poly(vinylidene fluoride)
matrix composites with micro–nanosize BaTiO3 filler. Appl Phys Lett 2007;91:222902.
[43] Dang ZM, Lin YQ, Xu HP, Shi CY, Li ST, Bai J. Fabrication and dielectric characterization of advanced BaTiO3/polyimide
nanocomposite films. Adv Funct Mater 2008;18:1509–17.
[44] Dang ZM, Tian CY, Zha JW, Yao SH, Xia YJ, Shi CY, et al. Potential bioelectroactive bone-recovery polymer nanocomposites
with high dielectric permittivity. Adv Eng Mater 2009;11:B144–7.
[45] Dang ZM, Zhou T, Yao SH, Yuan JK, Zha JW, Song HT, et al. Advanced calcium copper titanate/polyimide hybrid films with
high dielectric permittivity. Adv Mater 2009;21:2077–82.
[46] Landi BJ, Raffaelle RP, Heben MJ, Alleman JL, VanDerveer W, Gennett T. Single wall carbon nanotube–nafion composite
actuators. Nano Lett 2002;2:1329–33.
[47] Fan Z, Luo G, Zhang Z, Zhou LI, Wei FEI. Electromagnetic and microwave absorbing properties of multi-walled carbon
nanotubes/polymer composites. Mater Sci Eng B 2006;132:85–9.
[48] Moniruzzaman M, Winey KI. Polymer nanocomposites containing carbon nanotubes. Macromolecules
2006;39:5194–205.
[49] He F, Lau S, Chan HL, Fan J. High dielectric permittivity and low percolation threshold in nanocomposites based on
poly(vinylidene fluoride) and exfoliated graphite nanoplates. Adv Mater 2009;21:710–5.
[50] Dang ZM, Wang L, Yin Y, Zhang Q, Lei QQ. Giant dielectric permittivities in functionalized carbon–nanotube/electroactive-
polymer nanocomposites. Adv Mater 2007;19:852–7.
[51] Wang L, Dang ZM. Carbon nanotube composites with high dielectric constant at low percolation threshold. Appl Phys Lett
2005;87:042903.
[52] Dang ZM, Yao SH, Xu HP. Effect of tensile strain on morphology and dielectric property in nanotubes/polymer composites.
Appl Phys Lett 2007;90:012907.
[53] Yao SH, Dang ZM, Xu HP, Jiang MJ, Bai J. Exploration of dielectric constant dependence on evolution of microstructure in
nanotube/ferroelectric polymer nanocomposites. Appl Phys Lett 2008;92:082902.
[54] Yao SH, Dang ZM, Jiang MJ, Bai J. BaTiO3-nanotube/polyvinylidene fluoride three-phase composites with high dielectric
constant and low dielectric loss. Appl Phys Lett 2008;93:182905.
[55] Yao SH, Dang ZM, Jiang MJ, Xu HP, Bai J. Influence of aspect ratio of carbon nanotube on percolation threshold in
ferroelectric polymer nanocomposite. Appl Phys Lett 2007;91:212901.
[56] Jiang MJ, Dang ZM, Xu HP. Giant dielectric constant and resistance-pressure sensitivity in carbon nanotubes/rubber
nanocomposites with low percolation threshold. Appl Phys Lett 2007;90:042914.
[57] Dang ZM, Fan LZ, Nan CW. Dielectric behavior of novel three-phase carbon nanotubes/BaTiO3/PVDF composites. Mater Sci
Eng B 2003;103:140–4.
[58] Dang ZM, Wu JP, Xu HP, Yao SH, Jiang MJ, Bai J. Dielectric properties of upright carbon fiber filled poly(vinylidene fluoride)
composite with low percolation threshold and weak temperature dependence. Appl Phys Lett 2007;91:072912.
[59] Dang ZM, Fan LZ, Shen Y, Nan CW. Study on dielectric behavior of a three-phase CF/(PVDF+BaTiO3) composite. Chem Phys
Lett 2003;369:95–100.
[60] Dang ZM, Nan CW. Dielectric properties of carbon fiber filled low-density polyethylene. J Appl Phys 2003;93:5543–5.
[61] Dang ZM, Xia B, Yao SH, Jiang MJ, Song HT, Zhang LQ, et al. High-dielectric-permittivity three-component high-elasticity
nanocomposites with low percolation threshold and low dielectric loss. Appl Phys Lett 2009;94:042902.
[62] Xu HP, Dang ZM. Electrical property and microstructure analysis of poly(vinylidene fluoride)-based composites with
different conductive fillers. Chem Phys Lett 2007;438:196–202.
[63] Achar BN, Fohlen GG, Parker JA. Phthalocyanine polymer II. Synthesis and characterization of some metal phthalocyanine
sheet oligomers. J Polym Sci Polym Chem 1982;20:2480–4.
[64] Zhang QM, Li H, Poh M, Xu H, Cheng ZY, Xia F, et al. An all-organic composite actuator material with a high dielectric
constant. Nature 2002;419:284–7.
[65] Wang JW, Shen QD, Yang CZ, Zhang QM. High dielectric constant composite of P(VDF–TrFE) with grafted copper
phthalocyanine oligomer. Macromolecules 2004;37:2294–8.
[66] Wang JW, Shen QD, Bao HM, Yang CZ, Zhang QM. Microstructure and dielectric properties of P(VDF–TrFE–CFE) with
partially grafted copper phthalocyanine oligomer. Macromolecules 2005;38:2247–52.
[67] Huang C, Zhang QM, Li JJ. Colossal dielectric and electromechanical responses in self-assembled polymeric
nanocomposites. Appl Phys Lett 2005;87:182901.
[68] Huang C, Zhang QM, deBotton G, Bhattacharya K. All-organic dielectric-percolative three-component composite materials
with high electromechanical response. Appl Phys Lett 2004;84:4391–3.
[69] Bobnar V, Levstik A, Huang C, Zhang QM. Dielectric properties and charge transport in all-organic relaxor like CuPc–
P(VDF–TrFE–CFE) composites and its comstituents. Ferroelectrics 2006;338:107–16.
[70] Huang C, Zhang QM. Fully functionalized high-dielectric-constant nanophase polymers with high electromechanical
response. Adv Mater 2005;17:1153–8.
[71] Bobnar V, Levstik A, Huang C, Zhang QM. Intrinsic dielectric properties and charge transport in oligomers of organic
semiconductor copper phthalocyanine. Phys Rev B 2005;71:041202.
[72] Huang C, Zhang QM, Su J. High-dielectric-constant all-polymer percolative composites. Appl Phys Lett 2003;82:3502–4.
[73] Li JY, Huang C, Zhang QM. Enhanced electromechanical properties in all-polymer percolative composites. Appl Phys Lett
2004;84:3124–6.
[74] Wang CC, Song JF, Bao HM, Shen QD, Yang CZ. Enhancement of electrical properties of ferroelectric polymers by
polyaniline nanofibers with controllable conductivities. Adv Funct Mater 2008;18:1299–306.
[75] Yuan JK, Dang ZM, Bai J. Unique dielectric properties of ferroelectric polyaniline/poly(vinylidene fluoride) composites
induced by temperature variation. Phys Status Solidi: Rap Res Lett 2008;2:233–5.
[76] Yuan JK, Dang ZM, Yao SH, Zha JW, Zhou T, Li ST, et al. Fabrication and dielectric properties of advanced high permittivity
polyaniline/poly(vinylidene fluoride) nanohybrid films with high energy storage density. J Mater Chem 2010;20:2441–7.
[77] Rao Y, Wong CP. Ultra high dielectric constant polymer metal composite for embedded capacitor application. US Patent
6864306, 2005.
[78] Balasubramanian K, Burghard M. Asymmetric end-functionalization of carbon nanotubes. Small 2005;1:1148–50.
[79] Bruggeman DAG. The calculation of various physical constants of heterogeneous substances. I. The dielectric constants
and conductivities of mixtures composed of isotropic substances. Ann Phys 1935;24:636–42.
[80] Nan CW. Physics of inhomogeneous inorganic materials. Prog Mater Sci 1993;37:1–126.
[81] Nan CW. Comment on ‘Effective dielectric function of a random medium’. Phys Rev B 2001;63:176201-1.
[82] Sihvola AH, Pekonen OPM. Effective medium formula for bi-anisotropic mixtures. J Phys D: Appl Phys 1996;29:514–21.
[83] Jayasundere N, Smith BV. Dielectric constant for binary piezoelectric 0–3 composites. J Appl Phys 1993;73:2462–6.
[84] Kilbride BE, Coleman JN, Fraysse J, Fournet P, Cadek M, Drury, et al. Experimental observation of scaling laws for
alternating current and direct current conductivity in polymer–carbon nanotube composite thin films. Appl Phys Lett
2002;92:4024–6.
[85] Gallone G, Carpi F, De Rossi D, Rossi DD, Levita G, Marchetti A. Dielectric constant enhancement in a silicone elastomer
filled with lead magnesium niobate–lead titanate. Mater Sci Eng C 2007;27:110–6.
[86] Carpi F, De Rossi D. Improvement of electromechanical actuating performances of a silicone dielectric elastomer by
dispersion of titanium dioxide powder. IEEE Trans Dielect Electr Insul 2005;12:835–43.
[87] Mc Pherson J, Kim JK, Shanware A, Mogul H. Thermochemical description of dielectric breakdown in high dielectric
constant materials. Appl Phys Lett 2003;13:2121–3.
[88] Blythe AR. Dielectric breakdown. Electrical properties of polymers. New York: Cambridge University Press; 1979.
[89] Bottcher CJF, Van Belle OC, Bordewijk P, Rip A, Yue DD. Theory of electric polarization. J Electrochem Soc 1974;121:211C.
[90] Van Beek LKH. Dielectric behaviour of heterogeneous systems. Prog Dielect 1967;7:69–114.
[91] Sillars RW. The properties of a dielectric containing semi-conducting particles of various shapes. J Inst Elect Eng
1937:378–94.
[92] Landauer R. The electrical resistance of binary metallic mixtures. J Appl Phys 1952;23:779–84.
[93] Maxwell-Garnett JC. Colours in metal glasses and in metallic films. Philos Trans Roy Soc Lond 1904;203:385–9.
[94] Smith GB. Dielectric constant for mixed media. J Phys D Appl Phys 1977;10:39–43.
[95] Tuncer E, Gubanske SM, Nettelblad B. Dielectric relaxation in dielectric mixtures: application of the finite element method
and its comparison with mixture formulas. Appl Phys 2001;89:8092–100.
[96] Shalev VM. Electromagnetic properties of small-particle composites. Phys Rep 1996;272:61–137.
[97] Wagner KW. The after effect in dielectrics. Arch Electrotech 1914;2:378–80.
[98] Nelson SO, You TS. Relationships between microwave permittivities of solid and pulverized plastics. J Phys D Appl Phys
1990;23:346–53.
[99] Shen LC, Savret WC, Price JM, Athavale K. Dielectric properties of reservoir rocks at ultra-frequencies. Geophysics
1985;50:692.
[100] Wilson SA, Maistros GM, Whatmore RW. Structure modification of 0–3 piezoelectric ceramic/polymer composites
through dielectrophoresis. J Phys D: Appl Phys 2005;38:175–82.
[101] Zakri T, Laurent JP, Vauclin M. Theoretical evidence for Lichtenecker’s mixture formulae’ based on the effective medium
theory. J Phys D Appl Phys 1998;31:1589–94.
[102] Banno H, Ogura K. Dielectric and piezoelectric properties of a flexible composite consisting of polymer and mixed ceramic
powder of PZT and PbTiO3. Ferroelectrics 1989;95:171–4.
[103] Daben Y. Composite piezoelectric film made from PVDF polymer and PCM-PZT ferroelectric ceramics. Ferroelectrics
1990;101:291–6.
[104] Ngoma JB, Cavaille JY, Paletto J, Perez J, Macchi F. Dielectric and piezoelectric properties of copolymer-ferroelectric
composite. Ferroelectrics 1990;109:205–10.
[105] Das-Gupta DK. Pyroelectricity in polymers. Ferroelectrics 1991;118:165–85.
[106] Wei B, Daben Y. Dielectric and piezoelectric properties of 0–3 composite film in PCM/PVDF and PZT/PVDF. Ferroelectrics
1994;157:427–30.
[107] Chan HLW, Cheung MC, Choy CL. Study on BaTiO3/P(VDF–TREE) 0–3 composite. Ferroelectrics 1999;224:541–8.
[108] Kuo DH, Chang CC, Su TY, Wang WK, Lin BY. Dielectric behaviors of multi-doped BaTiO3/epoxy composites. J Eur Ceram
Soc 2001;21:1171–7.
[109] Gregorio R, Cestari JM, Bernardino FE. Dielectric behaviour of thin films of b-PVDF/PZT and b-PVDF/BaTiO3 composites. J
Mater Sci 1996;31:2925–30.
[110] Chan HLW, Chan WK, Zhang Y, Choy CL. Pyroelectric and piezoelectric properties of lead titanate/polyvinylidene fluoride–
trifluoroethylene 0–3 composites. IEEE Trans Electr Insul 1998;5:505–12.
[111] Dias CJ, Das-Gupta DK. Inorganic ceramic/polymer ferroelectric composite electrets. IEEE Trans Electr Insul
1996;3:706–34.
[112] Chiteme C, Mclachlan DS. Measurements of universal and non-universal percolation exponents in macroscopically
similar systems. Physica B 2000;279:69–71.
[113] Thostenson ET, Chou TW. Carbon nanotube networks: sensing of distributed strain and damage for life prediction and self
healing. Adv Mater 2006;18:2837.
[114] Newnham RE, Skinner DP, Cross LE. Connectivity and piezoelectric–pyroelectric composites. Mater Res Bull
1978;13:525–36.
[115] Wei T, Jin CQ, Zhang W, Liu JM. High permittivity polymer embedded with Co/ZnO core/shell nanoparticles modified by
organophosphorus. Appl Phys Lett 2007;91:222907.
[116] Dasgupta DK, Doughty K. Polymer–ceramic composite materials with high dielectric constants. Thin Solid Films
1988;158:93–105.
[117] Arbatti M, Shan X, Cheng Z. Ceramic–polymer composites with high dielectric constant. Adv Mater 2007;19:1369–72.
[118] Zheng W, Wong SC. Electrical conductivity and dielectric properties of PMMA/expanded graphite composites. Compos Sci
Technol 2003;63:225–35.
[119] Garner CM, Kloster G, Atwood G, Mosley L, Palanduz AC. Challenges for dielectric materials in future integrated circuit
technologies. Microelectr Reliab 2005;45:919–24.
[120] McLachlan DS, Blaszkiewics M, Newnham RE. Electrical resistivity of composites. J Am Ceram Soc 1990;73:2187–203.
[121] Wang CC, Shen QD, Tang SC, Wu Q, Bao HM, Yang CZ, et al. Ferroelectric polymer nanotubes with large dielectric
constants for potential all-organic electronic devices. Macromol Rap Commun 2008;29:724–8.
[122] Zhang S, Zhang N, Huang C, Ren K, Zhang Q. Microstructure and electromechanical properties of carbon nanotube/
poly(vinylidene fluoride–trifluoroethylene–chlorofluoroethylene) composites. Adv Mater 2005;17:1897–901.
[123] Sui G, Jana S, Zhong WH, Fuqua MA, Ulven CA. Dielectric properties and conductivity of carbon nanofiber/semi-crystalline
polymer composites. Acta Mater 2008;56:2381–8.
[124] McLachlan DS, Chiteme C, Park C, Wise KE, Lowther SE, Lillehei PT, et al. AC and DC percolative conductivity of single wall
carbon nanotube polymer composites. J Polym Sci B 2005;43:3273–87.
[125] Wu J, McLachlan DS. Scaling behavior of the complex conductivity of graphite–boron nitride percolation systems. Phys
Rev B 1998;58:14880–7.
[126] Valentini L, Puglia D, Frulloni E, Armentano I, Kenny JM, Santucci S. Dielectric behavior of epoxy matrix/single-walled
carbon nanotube composites. Compos Sci Technol 2004;64:23–33.
[127] Jiang MJ, Dang ZM, Bozlar M, Miomandre F, Bai J. Remarkable broad-frequency dielectric behaviors in multiwalled carbon
nanotubes/rubber nanocomposites. J Appl Phys 2009;106:084902.
[128] Dang ZM, Wei-Kang Li, Hai-Ping Xu. Origin of remarkable positive temperature coefficient effect in the modified carbon
black and carbon fiber cofilled polymer composites. J Appl Phys 2009;106:024913.
[129] Jiang MJ, Dang ZM, Yao SH, Bai J. Effect of surface modification of carbon nanotubes on microstructure and electrical
Property in the nanotubes/rubber–matrix nanocomposites. Chem Phys Lett 2008;457:352–6.
[130] Jiang MJ, Dang ZM, Xu HP, Yao SH, Bai J. Effect of aspect ratio of multiwall carbon nanotubes on resistance-pressure
sensitivity of rubber nanocomposites. Appl Phys Lett 2007;91:072907.
[131] Jiang MJ, Dang ZM, Xu HP. Significant enhanced electrical conductivity in chemically modified carbon nanotube/
methylvinyl silicone rubber nanocomposite. Eur Polym J 2007;43:4924–30.
[132] Jiang MJ, Dang ZM, Xu HP. Significant temperature and pressure sensitivities of electrical properties in multi-walled
carbon nanotubes/methylvinyl silicone rubber nanocomposite. Appl Phys Lett 2006;89:182902.
[133] Dang ZM, Jiang MJ, Xie D, Yao SH, Zhang LQ, Bai J. Supersensitive linear piezoresistive property in carbon nanotubes/
silicone rubber nanocomposites. J Appl Phys 2008;104:024114.
[134] Haggenmueller R, Gommans HH, Rinzler AG, Fischer JE, Winey KI. Aligned single-wall carbon nanotubes in composites by
melt processing methods. Chem Phys Lett 2000;330:219–25.
[135] Wong SC, Sutherland E, Uhl F. Materials processes of graphite nanostructured composites using ball milling. Mater
Manufact Process 2006;21:159–66.
[136] Azhdar B, Stenberg B, Kari L. Polymer–nanofiller prepared by high-energy ball milling and high velocity cold compaction.
Polym Compos 2008;29:252–61.
[137] Kim HJ, Yoon YJ, Kim JH, Nam SM. Application of Al2O3-based polyimide composite thick films to integrated substrates
using aerosol deposition method. Mater Sci Eng B 2009;161:104–8.
[138] Baker W, Scott C, Hu GH. Reactive polymer blending. Munich: Hanser publisher; 2001. 290pp.
[139] Hu GH. Reactive polymer processing: fundamentals of REX. In: Buschow KHJ, Cahn RW, Flemings MC, Ilschner B, Kramer
EJ, Mahajan S, editors. Encyclopedia of materials: science and technology. The Netherlands: Elsevier Science, Amsterdam;
2001. p. 8049–57.
[140] Hu GH, Hoppe S, Feng L, Fonteix C. Reactive compounding: mixing and compounding of polymers: theory and practice.
2nd ed. Hanser (Germany): Ica Manas-Zloczower; 2009. p. 1019–80 [chapter 27].
[141] Potshke P, Bhattacharyya AR, Janke A. Melt mixing of polycarbonate with multiwalled carbon nanotubes: microscopic
studies on the state of dispersion. Eur Polym J 2004;40:137–48.
[142] Moniruzzaman M, Du F, Romero N, Winey KI. Increased flexural modulus and strength in SWNT/epoxy composites by a
new fabrication method. Polymer 2006;47:293–8.
[143] Sandler J, Shaffer MSP, Prasse T, Bauhofer W, Schulte K, Windle AH. Development of a dispersion process for carbon
nanotubes in an epoxy matrix and the resulting electrical properties. Polymer 1999;40:5967–71.
[144] Safadi B, Andrews R, Grulke EA. Multiwalled carbon nanotube polymer composites: synthesis and characterization of thin
films. J Appl Polym Sci 2002;84:2660–9.
[145] Vaccarini L, Desarmot G, Almairac R, Tahir S, Goze C, Bernier P. Reinforcement of an epoxy resin by single walled
nanotubes. AIP Conf Proc 2000;544:521–5.
[146] Qi JQ, Li LT, Wang YL, Gui ZL. Preparation of nanoscaled BaTiO3 powders by DSS method near room temperature under
normal pressure. J Cryst Growth 2004;260:551–6.
[147] Qian D, Dickey EC, Andrews R, Rantell T. Load transfer and deformation mechanisms in carbon nanotube–polystyrene
composites. Appl Phys Lett 2000;76:2868–70.
[148] Watts PCP, Hsu WK, Chen GZ, Fray DJ, Kroto HW, Walton DRM. A low resistance boron-doped carbon nanotube–
polystyrene composite. J Mater Chem 2001;11:2482–8.
[149] Bower C, Rosen R, Jin L, Han J, Zhou O. Deformation of carbon nanotubes in nanotube–polymer composites. Appl Phys Lett
1999;74:3317–9.
[150] Lin Y, Zhou B, Fernando KAS, Liu P, Allard LF, Sun YP. Polymeric carbon nanocomposites from carbon nanotubes
functionalized with matrix polymer. Macromolecules 2003;36:7199–204.
[151] Cadek M, Coleman JN, Barron V, Hedicke K, Blau WJ. Morphological and mechanical properties of carbon–nanotube-
reinforced semicrystalline and amorphous polymer composites. Appl Phys Lett 2002;81:5123–5.
[152] Ruan SL, Gao P, Yang XG, Yu TX. Toughening high performance ultrahigh molecular weight polyethylene using
multiwalled carbon nanotubes. Polymer 2003;44:5643–54.
[153] Assouline E, Lustiger A, Barber AH, Copper CA, Klein E, Wachtel E, et al. Nucleation ability of multiwall carbon nanotubes
in polypropylene composites. J Polym Sci B 2003;41:520–7.
[154] Grady BP, Pompeo F, Shambaugh RL, Resasco DE. Nucleation of polypropylene crystallization by single-walled carbon
nanotubes. J Phys Chem B 2002;106:5852–8.
[155] Paiva MC, Zhou B, Fernando KAS, Lin Y, Kennedy JM, Sun YP. Mechanical and morphological characterization of polymer–
carbon nanocomposites from functionalized carbon nanotubes. Carbon 2004;42:2849–54.
[156] Probst O, Moore EM, Resasco DE, Grady BP. Nucleation of polyvinyl alcohol crystallization by single-walled carbon
nanotubes. Polymer 2004;45:4437–43.
[157] Zhang X, Liu F, Sreekumar TV, Kumar S, Moore VC, Hauge RH, et al. Poly (vinyl alcohol)/SWNT composite film. Nano Lett
2003;3:1285–8.
[158] Lu KL, Lago RM, Chen YK, Green MLH, Harris PJF, Tsang SC. Mechanical damage of carbon nanotubes by ultrasound.
Carbon 1996;34:814–6.
[159] Koshio A, Yudasaka M, Zhang M, IiJima S. A simple way to chemically react single-wall carbon nanotubes with organic
materials using ultrasonication. Nano Lett 2001;1:361–3.
[160] Dufresne A, Paillet M, Putaux JL, Canet R, Carmona F, Delhaes P, et al. Processing and characterization of carbon nanotube/
poly (styrene-co-butyl acrylate) nanocomposites. J Mater Sci 2002;37:3915–23.
[161] Mitchell CA, Bahr JL, Arepalli S, Tour JM, Krishnamoorti R. Dispersion of functionalized carbon nanotubes in polystyrene.
Macromolecules 2002;35:8825–30.
[162] Barraza HJ, Pompeo F, Orear EA, Resasco DE. SWNT-filled thermoplastic and elastomeric composites prepared by
miniemulsion polymerization. Nano Lett 2002;2:797–802.
[163] Penu C, Hu GH, Fonteix C, Marchal P, Choplin L. Effects of carbon nanotubes and their state of dispersion on the anionic
polymerization of e-caprolactam: 2. Calorimetry. Polym Eng sci 2010;50:2287–97.
[164] Ounaies Z, Park C, Wise KE, Siochi EJ, Harrision JS. Electrical properties of single wall carbon nanotube reinforced
polyimide composites. Compos Sci Technol 2003;63:1637–46.
[165] Roslaniec Z, Broza G, Schulte K. Nanocomposites based on multiblock polyester elastomers (PEE) and carbon nanotubes
(CNT). Compos Interface 2003;10:95–102.
[166] Popielarz R, Chiang CK. Polymer composites with the dielectric constant comparable to that of barium titanate ceramics.
Mater Sci Eng B 2007;139:48–54.
[167] Cochet M, Maser WK, Benito AM, CalleJas MA, Martinez MT, Benoit JM, et al. Synthesis of a new polyaniline/nanotube
composite: ‘‘in-situ’’ polymerisation and charge transfer through site-selective interaction. Chem Commun
2001;16:1450–1.
[168] Xiao Q, Zhou X. The study of multiwalled carbon nanotube deposited with conducting polymer for supercapacitor.
Electrochem Acta 2003;48:575–80.
[169] Lewis TJ. Interfaces are the dominant feature of dielectrics at the nanometric level. IEEE Trans Dielect Electr Insul
2004;11:739–53.
[170] Lewis TJ. Interfaces: nanometric dielectrics. J Phys D: Appl Phys 2005;38:202–12.
[171] Drzal LT, Rich MJ, Lloyd PF. Adhesion of graphite fibers to epoxy matrices: I. The role of fiber surface treatment. J Adhes
1983;16:1–30.
[172] Wu SH. Phase-structure and adhesion in polymer blends: criteria for rubber toughening. Polymer 1985;26:1855–63.
[173] Galeski A. Strength and toughness of crystalline polymer systems. Prog Polym Sci 2003;28:1643–99.
[174] Kobayashi Y, Tanase T, Tabata T, Miwa T, Konno M. Fabrication and dielectric properties of the BaTiO3–polymer nano-
composite thin films. J Eur Ceram Soc 2008;28:117–22.
[175] Xie SH, Zhu BK, Wei XZ, Xu ZK, Xu YY. Polyimide/BaTiO3 composites with controllable dielectric properties. Composites A
2005;36:1152–7.
[176] Ramirez AP, Subramanian MA, Gardel M, Blumberg G, Li D, Vogt T, et al. Giant dielectric constant response in a copper-
titanate. Solid State Commun 2000;115:217–20.
[177] Subramanian MA, Li D, Duan N, Reisner BA, Sleight AW. High dielectric constant in ACu3Ti4O12 and ACu3Ti3FeO12 phases. J
Solid State Chem 2000;151:323–5.
[178] Homes CC, Vogt T, Shapiro SM, Wakimoto S, Ramirez AP. Optical response of high-dielectric-constant perovskite-related
oxide. Science 2001;293:673–6.
[179] Wu JB, Nan CW, Lin YH, Deng Y. Giant dielectric permittivity observed in Li and Ti doped NiO. Phys Rev Lett
2002;89:217601.
[180] Wang XX, Lam KH, Tang XG, Chan HLW. Dielectric characteristics and polarization response of lead-free ferroelectric
(Bi0.5Na0.5)0.94Ba0.06TiO3–P(VDF–TrFE) 0–3 composites. Solid State Commun 2004;130:695–9.
[181] Bhattacharya SK. Metal-filled polymers: properties and applications. Marcel Dekker; 1986.
[182] Mamunya EP, Davidenko VV, Lebedev EV. Percolation conductivity of polymer composites filled with dispersed
conductive filler. Polym Compos 1995;16:319–24.
[183] Bigg DM, Stutz DE. Plastic composites for electromagnetic interference shielding applications. Polym Compos
1983;4:40–6.
[184] Mamunya Y. Polymer blends filled with carbon black: structure and electrical properties. Macromol Symp
2001;170:257–64.
[185] Zois H, Apekis L, Mamunya YP. Dielectric properties and morphology of polymer composites filled with dispersed iron. J
Appl Polym Sci 2003;88:3013–20.
[186] Mamunya YP. Morphology and percolation conductivity of polymer blends containing carbon black. J Macromol Sci Phys
1999;38:615–22.
[187] Miyasaka K et al. Electrical conductivity of carbon–polymer composites as a function of carbon content. J Mater Sci
1982;17:1610–6.
[188] Sumita M et al. Double percolation effect on the electrical conductivity of conductive particles filled polymer blends.
Colloid Polym Sci 1992;270:134–9.
[189] Xu HP, Dang ZM, Shi DH, Bai J. Remarkable selective localization of modified nanoscaled carbon black and positive
temperature coefficient resistance effect in binary-polymer matrix nanocomposites. J Mater Chem 2008;18:2685–90.
[190] Tasis D et al. Chemistry of carbon nanotubes. Chem Rev 2006;106:1105–36.
[191] Todd MG, Shi FG. Characterizing the interphase dielectric constant of polymer composite materials: effect of chemical
coupling agents. J Appl Phys 2003;94:4551–7.
[192] Yeh MK, Tai NH, Liu JH. Mechanical behavior of phenolic-based composites reinforced with multi-walled carbon
nanotubes. Carbon 2006;44:1–9.
[193] Lourie O, Wagner HD. Evidence of stress transfer and formation of fracture clusters in carbon nanotube-based
composites. Compos Sci Technol 1999;59:975–7.
[194] Kimura T, Ago H, Tobita M, Ohshima S, Kyotani M, Yumura M. Polymer composites of carbon nanotubes aligned by a
magnetic field. Adv Mater 2002;14:1380–3.
[195] Park C, Wilkinson J, Banda S, Ounaies Z, Wise KE, Sauti G, et al. Aligned single-wall carbon nanotube polymer composites
using an electric field. J Polym Sci B 2006;44:1751–62.
[196] Kuriger RJ, Alam MK, Anderson DP, Jacobsen RL. Processing and characterization of aligned vapor grown carbon fiber
reinforced polypropylene. Composites A 2002;33:53–62.
[197] Shen Y, Yue Z, Li M, Nan CW. Enhanced initial permeability and dielectric constant in a double-percolating Ni
0.3Zn0.7Fe1.95O4–Ni–polymer composite. Adv Funct Mater 2005;15:1100–3.
[198] Patil R, Ashwin A, Radhakrishnan S. Novel polyaniline/PVDF/BaTiO3 hybrid composites with high piezo-sensitivity.
Sensors Actuat A: Phys 2007;138:361–5.
[199] Li JY. Exchange coupling in P(VDF–TrFE) copolymer based all-organic composites with giant electrostriction. Phys Rev
Lett 2003;90:217601.
[200] Leu C, Chang Y, Wei K. Polyimide-side-chain tethered polyhedral oligomeric silsesquioxane nanocomposites for low-
dielectric film applications. Chem Mater 2003;15:3721–7.
[201] Carr SH. Polymer preparation. Am Chem Soc Div Polym Chem 1991:374.
[202] Mezzenga R, Ruokolainen J, Fredrickson GH, Kramer EJ, Moses D, Heeger J, et al. Templating organic semiconductors via
self-assembly of polymer colloids. Science 2003;299:1872–4.
[203] Hawker CJ, Hedrick JL, Miller RD, Volksen W. Supramolecular approaches to nanoscale dielectric foams for advanced
microelectronic devices. MRS Bull 2000;4:54–8.
[204] Dang ZM, Yao SH, Yuan JK, Bai J. Tailored Dielectric Properties based on Microstructure Change in BaTiO3-Carbon
Nanotube/Polyvinylidene Fluoride Three-Phase Nanocomposites. J. Phys. Chem. C 2010;114:13204–9.
[205] Nisa VS, Rajesh S, Murali KP, Priyadarsini V, Potty SN, Ratheesh R. Preparation, characterization and dielectric properties
of temperature stable SrTiO3/PEEK composites for microwave substrate applications. Compos Sci Technol
2008;68:106–12.
[206] Subodh G, Joseph M, Mohanan P, Sebastian MT. Low dielectric loss polytetrafluoroethylene/TeO2 polymer ceramic
composites. J Am Ceram Soc 2007;90:3507–11.
[207] Yamada T, Ueda T, Kitayama T. Piezoelectricity of a high-content lead zirconate titanate/polymer composite. J Appl Phys
1982;53:4328–32.
[208] Bhattacharya S, Tummaka RR, Chahal P, White G. Integration of polymer/ceramic thin film capacitor on PWB. Int Symp
Adv Pack Mater 1997:68–70.
[209] Kim DW, Lee DH, Kim BK, Je HJ, Park JG. Direct assembly of BaTiO3-polymethylmethacrylate nanocomposite films. Macro
Rap Commun 2006;27:1821–5.
[210] Yoon DH, Zhang J, Lee BI. Dielectric constant and mixing model of barium titanate composite thick films. Mater Res Bull
2003;38:765–72.
[211] Pecharroman C, Moya JS. Experimental evidence of a giant capacitance in insulator–conductor composites at the
percolation threshold. Adv Mater 2000;12:294–7.
[212] Brus LE. Electron–electron and electron–hole interactions in small semiconductor crystallites: the size dependence of the
lowest excited electronic state. J Chem Phys 1984;80:4403–9.
[213] Brus LE. A simple model for the ionization potential, electron affinity, and aqueous redox potentials of small
semiconductor crystallites. J Chem Phys 1983;79:5566–71.
[214] Lee WH, Lee C, Jang J. Quantum size effects on the conductivity in porous silicon. J Non-Cryst solids 1996;198200:911–4.
[215] Rabani E, Hetenyi B, Berne BJ, Brus LE. Electronic properties of CdSe nanocrystals in the absence and presence of a
dielectric medium. J Chem Phys 1999;110:5355–69.
[216] Nisa VS, RaJesh S, Murali KP, Priyadarsini V, Potty SN, Ratheesh R. Preparation, characterization and dielectric properties
of temperature stable SrTiO3/PEEK composites for microwave substrate applications. Compos Sci Technol
2008;68:106–12.
[217] Kirkpatrick S. Percolation and conduction. Rev Mod Phys 1973;45:574–88.
[218] Yu Y, Dang ZM, Zha JW. Micro-nanosize cofilled high dielectric permittivity composites. In: Proceedings of the 9th
international conference on properties and applications of dielectric materials. ICPADM 2009;2009. p.769–72.
[219] Agarwal V, Chahal P, Tummala RR, Allen MG. Improvements and recent advances in nanocomposite capacitors using a
colloidal technique. In: Proceedings of the 48th electronic components and technology conference, Seatle, WA, May 1998.
p. 165–70.
[220] Kar-Gupta R, Venkatesh TA. Electromechanical response of piezoelectric composites: effects of geometric connectivity
and grain size. Acta Mater 2008;56:3810–23.
[221] Bethune DS, Kiang CH, De Vries MS, Gorman G, Savoy R, Vazquez J, et al. Cobalt-catalysed growth of carbon nanotubes
with single-atomic-layer walls. Nature 1993;363:605–7.
[222] Journet C, Maser WK, Bernier P, Loiseau A, de La Chapelle ML, Lefrant S, et al. Large-scale production of single-walled
carbon nanotubes by the electric-arc technique. Nature 1997;388:756–7.
[223] Thess A, Lee R, Nikolaev P, Dai H, Petit P, Robert J, et al. Crystalline ropes of metallic carbon nanotubes. Science
1996;273:483–7.
[224] Cheng HM, Li F, Su G, Pan HY, He LL, Sun X, et al. Large-scale and low-cost synthesis of single-walled carbon nanotubes by
the catalytic pyrolysis of hydrocarbons. Appl Phys Lett 1998;72:3282–4.
[225] Tersoff J, Ruoff RS. Structural properties of a carbon–nanotube crystal. Phys Rev Lett 1994;73:676–9.
[226] Bryning MB, Islam MF, Kikkawa JM, Yodh AG. Very low conductivity threshold in bulk isotropic single-walled carbon
nanotube–epoxy composites. Adv Mater 2005;17:1186–91.
[227] Jin Z, Pramoda KP, Goh SH, Xu G. Poly (vinylidene fluoride)-assisted melt-blending of multi-walled carbon nanotube/
poly(methyl methacrylate) composites. Mater Res Bull 2002;37:271–8.
[228] Zhang S, Kumar S. Shaping polymer particles by carbon nanotubes. Macromol Rap Commun 2008;29:557–61.
[229] Doktycz SJ, Suslick KS. Interparticle collisions driven by ultrasound. Science 1990;247:1067–9.
[230] Tamura R, Lim E, Manaka T, Iwamotoa M. Analysis of pentacene field effect transistor as a Maxwell–Wagner effect
element. J Appl Phys 2006;100:114515.
[231] Oyama HT, Pyreha RS, Xie Y, Partclz RE, MaJJievie E. Coating of uniform inorganic particles with polymers. J Colloid
Interface Sci 1993;160:298–303.
[232] Sprycha R, Oyama HT, Zelenev A, Matijević E. Characterization of polymer-coated silica particles by microelectrophoresis.
Colloid Polym Sci 1995;273:693–700.
[233] Partch R, Gangolli SG, MatJievie E, Cai W, AraJis S. Conducting polymer composites I. Surface-induced polymerization of
pyrrole on iron (II) and cerium (IV)oxide particles. J Colloid Interface Sci 1991;144:27–35.
[234] Huang CL, MatJievie E. Coating of uniform inorganic particles with polymers 3-polypyrrole on different metal-oxides. J
Mater Res 1995;10:1327–36.
[235] Marinakos SM, Shultz DA, Feldheim DL. Gold nanoparticles as templates for the synthesis of hollow nanometer-sized
conductive polymer capsules. Chem Mater 1999;11:34–7.
[236] Quaroni L, Chumanov G. Preparation of polymer-coated functionalized silver nanoparticles. J Am Chem Soc
1999;121:10642–3.
[237] Liu Q, Xu Z, Fineh JA, Egerton R. A novel two-step silica-coating process for engineering magnetic nanocomposites. Chem
Mater 1998;10:3936–40.
[238] Ung T, Liz-Marzan LM, Mulvaney P. Redox catalysis using Ag@SiO2 colloids. J Phys Chem B 1999;103:6770–3.
[239] Bruggen MPB. Preparation and properties of colloidal core–shell rods with adjustable aspect ratios. Langmuir
1998;14:2245–55.
[240] Hall SR, Dvais SA, Mann S. Condensation of organosilica hybrid shells on nanoparticle templates: a direct synthetic route
to functionalized core–shell colloids. Langmuir 2000;16:1454–6.
[241] Pastoriza-Santos I, Kokysh DS, Mamedov AA, Giesig M, Kotov NA, Liz-Marzan LM. One-pot synthesis of Ag@TiO2 core–
shell nanoparticles and their layer-by-layer assembly. Langmuir 2000;16:2731–5.
[242] Partch R, Xie Y, Oyama ST, MatJievie E. Preparation and properties of uniform coated colloidal particles: VIII. Titanium
nitride on silica. J Mater Res 1993;8:2014–20.
[243] Liz-Marzan LM, Giersig M, Mulvaney P. Synthesis of nanosized gold–silica core–shell particles. Langmuir
1996;12:4329–35.
[244] Shen Y, Lin YH, Li M, Nan CW. High dielectric performance of polymer composite films induced by a percolating
interparticle barrier layer. Adv Mater 2007;19:1418–22.
[245] Shen Y, Lin YH, Nan CW. Interfacial effect on dielectric properties of polymer nanocomposites filled with core–shell-
structured particles. Adv Funct Mater 2007;17:2405–10.
[246] Dang ZM, You SS, Zha JW, Song HT, Li ST. Effect of shell-layer thickness on dielectric permittivity in Ag@TiO2 core@shell
nanoparticles polymer composites. Phys Status Solidi A 2010;207(3):739–42.
[247] Lu Y, Yin Y, Mayers BT, Xia Y. Modifying the surface properties of superparamagnetic iron oxide nanoparticles through a
sol–gel approach. Nano Lett 2002;2:183–6.
[248] Caruso F. Nanoengineering of particle surfaces. Adv Mater 2001;13:11–22.
[249] Bobnar V, Levstik A, Huang C, Zhang QM. Enhanced dielectric response in all-organic polyaniline–poly(vinylidene
fluoride–trifluoroethylene–chlorotrifluoroethylene) composite. J Non-Cryst Solids 2007;353:205–9.
[250] Xia F, Cheng ZY, Xu H, Li H, Zhang QM, Kavarnos GJ, et al. High electromechanical response in a poly (vinylidene fluoride–
trifluoroethylene–chlorofluoroethylene) terpolymer. Adv Mater 2002;14:1574–7.
[251] Xu H, Cheng ZY, Olson D, Mai T, Zhang QM, Kavarnos GJ. Ferroelectric and electromechanical properties of poly(VDF–
TrFE–CTFE) terpolymer. Appl Phys Lett 2001;78:2360–2.
[252] Zhang QM, Bharti V, Zhao X. Giant electrostriction and relaxor ferroelectric behavior in electron-irradiated
poly(vinylidene fluoride–trifluoroethylene) copolymer. Science 1998;208:2101–4.
[253] Bobnar V, Vodopivec B, Levstik A, Kosec M, Hilczer B, Zhang QM. Dielectric properties of relaxor-like vinylidene fluoride–
trifluoroethylene-based electroactive polymers. Macromolecules 2003;36:4436–42.
[254] Kutnjak Z, Filipic C, Pirc R, Levstik A, Farhi R, Marssi ME. Slow dynamics and periodicity breaking in a lanthanum-modified
lead zirconate titanate relaxor system. Phys Rev B 1999;59:294–301.
[255] Efros AL, Shklovskii BI. Critical behaviour of conductivity and dielectric constant near the metal–non-metal transition
threshold. Phys Status Solidi B 1976;76:475–85.
[256] Huang C, Klein T, Xia F, Li HF, Zhang QM. Poly(vinylidene fluoride–trifluoroethylene) based high performance
electroactive polymers. IEEE Trans Dielect Electr Insul 2004;11:299–310.
[257] Ma D, Hugener TA, Siegel RW, Christerson A, Martensson E, Onneby C, et al. Influence of nanoparticle surface modification
on the electrical behavior of polyethylene nanocomposites. Nanotechnology 2005;16:724–31.
[258] Bauer F, Fousson E, Zhang QM, Lee LM. Ferroelectric copolymer and terpolymers for electrostrictors: synthesis and
properties. IEEE Trans Dielect Electr Insul 2004;11:293–8.
[259] Cheng ZY, Zhang QM, Bateman FB. Dielectric relaxation behavior and its relation to microstructure in relaxor ferroelectric
polymers: high-energy electron irradiated poly(vinylidene fluoride–trifluoroethylene) copolymers. J Appl Phys
2002;92:6749–55.
[260] Dang ZM, Yan WT, Xu HP. Novel high-dielectric-permittivity poly(vinylidene fluoride)/polypropylene blend composites:
the Influence of the poly(vinylidene fluoride) concentration and compatibilizer. J Appl Polym Sci 2007;105:3649–55.
[261] An L, Boggs SA, Calame JP. Energy storage in polymer films with high dielectric constant fillers. IEEE Trans Electr Insul
Mag 2008;24:5–10.
[262] Wang JW, Wang Y, Wang F, Li SQ, Xiao J, Shen QD. A large enhancement in dielectric properties of poly(vinylidene
fluoride) based all-organic nanocomposite. Polymer 2009;50:679–84.
[263] Lunkenheimer P, Bobnar V, Pronin AV, Ritus AI, Volkov AA, Loidl A. Origin of apparent colossal dielectric constants. Phys
Rev B 2002;66:052105(1)–5(4).
[264] Amaral F, Rubinger CPL, Henry F, Costa LC, Valente MA, Barros-Timmons A. Dielectric properties of polystyrene–CCTO
composite. J Non-Cryst Solids 2008;354:5321–2.
[265] Panda M, Srinivas V, Thakur AK. On the question of percolation threshold in polyvinylidene fluoride/nanocrystalline
nickel composites. Appl Phys Lett 2008;92:132905.
[266] Sarasqueta G, Choudhury KR, Kim DY, So F. Organic/inorganic nanocomposites for high-dielectric-constant materials.
[267] Pelrine R, Kombluh R, Joseph J. Electrostriction of polymer dielectrics with compliant electrodes as a means of actuation.
Sensor Actuat A 1998;64:77–85.
[268] Pelrine R, Kombluh R, Pei Q, Joseph J. Highspeed electrically actuated elastomers with strain greater than 100%. Science
2000;287:836–9.
[269] Li J, Il Seok S, Chu B, Dogan F, Zhang Q, Wang Q. Nanocomposites of ferroelectric polymers with TiO2 nanoparticles
exhibiting significantly enhanced electrical energy density. Adv Mater 2009;21:217–21.
[270] Facchetti A, Yoon MH, Marks TJ. Gate dielectrics for organic field-effect transistors: new opportunities for organic
electronics. Adv Mater 2005;17:1705–25.
[271] Lu J, Wong CP. Recent advances in high-k nanocomposite materials for embedded capacitor applications. IEEE Trans
Dielect Electr Insul 2008;15:1322–8.
[272] Xu HP, Dang ZM, Jiang MJ, Yao SH, Bai J. Enhanced dielectric properties and positive temperature coefficient resistance
effect in the binary-polymer nanocomposites with surface modified carbon black. J Mater Chem 2008;18:229–34.
[273] Liang S, Chong S, Giannelis E. Barium titanate/epoxy composite dielectric materials for integrated thin film capacitors.
Electron Compon Technol Conf 1998:171–5.
[274] Rao Y, Yue J, Wong C. Material characterization of high dielectric constant polymer–ceramic composite for embedded
capacitor to RF application. Active Passive Electron Compon 2002;25:123–6.
[275] Kim P, Jones S, Hotchkiss P, Haddock J, Kippelen B, Marder S, et al. Phosphonic acid-modified barium titanate polymer
nanocomposites with high permittivity and dielectric strength. Adv Mater 2007;19:1001–5.
[276] Rao Y, Wong CP. Material characterization of a high-dielectric constant polymer–ceramic composite for embedded
capacitor for RF application. J Appl Polym Sci 2004;92:2228–31.
[277] Murugaraj P, Mainwaring D, Mora-Huertas N. Dielectric enhancement of polymer–nanoparticle through interphase
polarizability. J Appl Phys 2005;98:054304.
[278] Xu J, Wong CP. Low-loss percolative dielectric composite. Appl Phys Lett 2005;87:082907.
[279] Choi H, Heo Y, Lee J, Kim J, Lee H, Park E, et al. Effects of particle size on dielectric properties of PMMA–Ni–BaTiO3
composites. Integr Ferroelectr 2007;87:85–90.
[280] Xu J, Wong M, Wong CP. Super high dielectric constant carbon black-filled polymer composites as integral capacitor
dielectrics. Electron Compon Technol Conf 2004;1:536–41.
[281] Lu J, Moon K, Wong CP. Development of novel silver nanoparticles/polymer composites as high k polymer matrix by
in situ photochemical method. Electron Compon Technol Conf 2006:1841–6.
[282] Rao Y, Qu J, Marinis T, Wong CP. A precise numerical prediction of the effective dielectric constant for polymer–ceramic
composites based on effective-medium theory. IEEE Trans Comp Pack Tech 2000;23:680–3.
[283] Smith RC, Liang C, Landry M, Nelson JK, Schadler LS. The mechanisms leading to the useful electrical properties of
polymer nanodielectrics. IEEE Trans Dielect Electr Insul 2008;15:187–96.
[284] Ramesh S, Shutzberg BA, Huang C, Gao J, Giannelis EP. Dielectric nanocomposites for integral thin films capacitors:
materials design, fabrication, and integration issues. IEEE Trans Adv Pack 2003;26:17–24.
[285] Prakash BS, Varma KBR. Dielectric behavior of CCTO/epoxy and Al-CCTO/epoxy composites. Compos Sci Technol
2007;67:2363–8.
[286] Li J, Claude J, Norena-Franco LE, Seok SI, Wang Q. Electrical energy storage in ferroelectric polymer nanocomposites
containing surface-functionalized BaTiO3 nanoparticles. Chem Mater 2008;20:6304–6.
[287] Gallone G, Carpi F, Rossi DD, Levita G, Marchetti A. Dielectric constant enhancement in a silicone elastomer filled with
lead magnesium niobate–lead titanate. Mater Sci Eng C 2007;27:110–6.
[288] Park IS, Kim KJ, Nam JD, Lee J, Yim W. Mechanical, dielectric, and magnetic properties of the silicone elastomer with
multi-walled carbon nanotubes as a nanofiller. Polym Eng Sci 2007;47:1396–405.
[289] Mathew G, Rhee JM, Nah C, Leo DJ, Nah C. Effects of silicone rubber on properties of dielectric acrylate elastomer actuator.
Polym Eng Sci 2006;46:1455–60.
[290] Carpi F, Rossi DD. Dielectric elastomer cylindrical actuators: electromechanical modelling and experimental evaluation.
Mater Sci Eng C 2004;24:555–62.
[291] Zhang X, Löwe C, Wissler M, Jähne B, Kovacs G. Dielectric elastomers in actuator technology. Adv Eng Mater
2005;7:361–7.
[292] Nalwa H. Handbook of low and high dielectric constant materials and their applications. London (UK): Academic Press;
1999.
[293] Osaka T, Datta M. Energy storage systems for electronics, Gordon and Breach. The Netherlands: Amsterdam; 2001.
[294] Barber P, Balasubramanian S, Anguchamy Y, Gong S, Wibowo A, Gao H, et al. Polymer composite and nanocomposite
dielectric materials for pulse power energy storage. Materials 2009;2:1697–733.
[295] Lehmann W, Skupin H, Tolksdorf C, Gebhard E, Zentel R, Krüger P, et al. Giant lateral electrostriction in ferroelectric
liquid-crystalline elastomers. Nature 2001;410:447–50.
[296] Cheng ZY et al. Electrostrictive poly(vinylidene fluoride–trifluoroethylene) copolymers. Sensor Actuat A 2001;90:138–47.
[297] Dario P, Carrozza M, Benvenuto A, Menciassi A. Micro-systems in biomedical applications. J Micromech Microeng
2000;10:235–44.
[298] Herr H, Kornbluh R. New horizons for orthotic and prosthetic technology: artificial muscle for ambulation. Smart Struct
Mater: Electroactive Polymer Actuat Dev 2004;5385:1–9.
[299] Xu TB, Cheng ZY, Zhang QM. High-performance micromachined unimorph actuators based on electrostrictive
poly(vinylidene fluoride–trifluoroethylene) copolymer. Appl Phys Lett 2002;80:1082–4.
[300] Su J, Xu TB, Zhang S, Shrout TR, Zhang QM. An electroactive ceramic/polymer hybrid actuation system for enhanced
electromechanical performance. Appl Phys Lett 2004;85:1045–7.
[301] Xia F, Tadigadapa S, Zhang QM. Electroactive polymer based microfluidic pump. Sensor Actuat A 2006;125:346–52.
[302] Ren K, Liu S, Lin M, Wang Y, Zhang QM. A compact electroactive polymer actuator suitable for refreshable Braille display.
Sensor Actuat A 2008;143:335–42.
[303] Roberts J, Slattery O, Swope B. New technology enables many-fold reduction in the cost of refreshable Braille display, in:
Proc. ASSETS2000, Arlington (VA); 2000.
[304] Pelrine R, Kornbluh R, Kofod G. High-strain actuator materials based on dielectric elastomers. Adv Mater
2000;12:1223–5.
[305] Kofod G, Wirges W, Paajanen M, Bauerb S. Energy minimization for self-organized structure formation and actuation.
[306] Tanaka T, Montanari GC, Mulhaupt R. Polymer nanocomposites as dielectrics and electrical insulation-perspectives for
processing technologies, material characterization, and future applications. IEEE Dielect Electr Insul 2004;11:763–84.
[307] Khalil MS. The role of barium titanate in modifying the dc breakdown strength of LDPE. IEEE Trans Dielect Electr Insul
2000;7:261–8.
[308] Tuncer E, Sauers I, James DR, Ellis AR, Paranthaman MP, Goyal A, et al. Enhancement of dielectric strength in
nanocomposites. Nanotechnology 2007;18:325704.
[309] Tuncer E, Sauers I, James DR, Ellis AR, Duckworth RC. Nanodielectric system for cryogenic applications: barium titanate
filled polyvinyl alcohol. IEEE Trans Dielect Electr Insul 2008;15:237–42.

Fundamentals, Processes and Applications

Uploaded by

Copyright:

Available Formats

Fundamentals, Processes and Applications

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fundamentals, Processes and Applications

Uploaded by

Copyright:

Available Formats

Progress in Materials Science 57 (2012) 660–723

Contents lists available at SciVerse ScienceDirect

Progress in Materials Science

Fundamentals, processes and applications

Article history: There is an increasing need for high-permittivity (high-k) materials

for high-k polymer–matrix composites (PMC). This review paper

6.1.3. Semiconducting ceramic fillers/polymer composites . . . . . . . . . . . . . . . . . . . . . . . . . . . 702

As predicted by Moore’s law, the efﬁciency of electronic products is improving in an exponential

Fig. 2. Market share of capacitors, resistors, and inductors [4].

Application Value range Stability required Tolerance required

2. Fundamental aspects of high-k composites

2.1. Deﬁnition of high permittivity (high-k)

2.2. Capacitance and electric energy storage of materials

2.3. Polarization and relaxation of dielectric materials

2.4. Dielectric strength of random polymer–matrix composites

2.5. Theoretical models for dielectric properties of polymer–matrix composites

2.5.1. Maxwell–Garnett equation

2.5.2. Bruggeman self-consistent effective medium approximation

3km þ 2/f ðkf km Þ

2.5.3. Jaysundere–Smith equation

2.5.4. Lichtenker rule

2.5.5. Percolation model

For electrical conductivity,

rc / ðffiller fc Þt for f filter > fc ð22aÞ

For dielectric permittivity,

Fig. 8. Ten connectivity patterns for two-phase composites [80,114].

2.6. Connection type of PMC

3. Processes for the fabrication of high-k PMC

3.1. Solid phase processes

3.1.1. Direct compounding

3.1.2. Melt compounding

3.2. Liquid phase processes

3.2.1. Liquid-phase assisted dispersion

3.2.2. Solution route

3.3. In situ polymerization processes

4. Microstructure and interfaces of high-k PMC

4.1. Two-phase high-k PMC

4.2. Three-phase high-k PMC

5. Effect of ﬁllers on dielectric properties of high-k PMC

5.1. Concentration effect of ﬁllers on dielectric properties

5.1.1. Effect of ceramic ﬁllers

200 100 Six Layers

Hot-press composites Dielectric constant/loss (100 Hz) Dielectric constant/loss (1 kHz)

Relative dielectric constant

5.1.2. Effect of electrical conducting ﬁllers

0 5 10 15 20 25 20 40 60 80 100 120 140

5.2. Size effect of ﬁllers on dielectric properties of PMC

5.2.1. Change in physical properties of ﬁllers with size reduction

Surface Area/Unit Volume (cm )

5.2.2. Dependence of dielectric properties of PMC on size of ﬁllers

Dissipation Factor (-)

Powder characteristics k at 1 MHz tan d at 1 MHz Tkr (ppm/°C)

100 nm 300 nm 200 nm

100 200 300 400 500 600

5.2.3. Effect of micro–nanosize coﬁllers on dielectric properties of PMC

Materials k tan d Filler size Filler loading

Double model composites

5.3. Shape effect of ﬁllers on dielectric properties

5.3.1. Shape of ﬁllers and ﬁller–polymer connectivity