Composites Science and Technology 69 (2009) 1298–1302
Contents lists available at ScienceDirect
Composites Science and Technology
journal homepage: www.elsevier.com/locate/compscitech
Three-phase polymer–ceramic–metal composite for embedded
capacitor applications
Sumesh George, Mailadil Thomas Sebastian *
Materials and Minerals Division, National Institute for Interdisciplinary Science and Technology, Thiruvananthapuram 695 019, India
a r t i c l e
i n f o
Article history:
Received 18 January 2009
Received in revised form 20 February 2009
Accepted 4 March 2009
Available online 13 March 2009
Keywords:
A. Polymer–matrix composites
B. Electrical properties
C. Optical microscopy
High permittivity
Embedded passives
a b s t r a c t
This paper addresses the materials and processes for printed wiring board compatible embedded capacitor using ceramic, polymer and metal. The Ca[(Li1/3Nb2/3)0.8Ti0.2]O3d (CLNT)–epoxy–silver, three-phase
composites were prepared by two step mixing and thermosetting technique. The dielectric properties of
the three-phase composites were investigated in terms of volume fraction of silver, temperature and frequency. The dielectric properties of epoxy–CLNT composites were compared with theoretical predictions.
The relative permittivity of the three-phase composites increased with silver loading. Addition of 0.28
volume fraction of silver increases the relative permittivity of epoxy–CLNT composites from 8 to 142
at 1 MHz. This composite is flexible and can be fabricated into various shapes with low processing
temperature.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Recent years have witnessed constant search for high permittivity materials that have wide range of technologically important
applications such as microelectronic, embedded passive, and electrostrictive devices [1,2]. Majority of the electronic component in
microelectronic circuits are passive and occupy more than 80% of
the printed wired surface area. Integration of embedded passive
components into printed circuit board offers a significant reduction
in size, better electrical performance, reliability, lower cost and improved design options. Among embedded passive components, the
embedded capacitor is particularly favorable because they are used
in large number of various functions such as decoupling, bypassing, energy storage and filtering capacitors. One major challenge
for implementing the embedded capacitor technology is the development of new dielectric material that possesses good dielectric
and mechanical properties and processabilities. Monolithic polymers [3] and dielectrics [4] have low dielectric loss but low permittivity limits their application in high permittivity based devices.
Therefore the concepts of two component and three component
composite which contain randomly distributed conductive fillers
in an insulating matrix have been developed [5–11]. In the context
of high permittivity composites, several ceramic–metal, polymer–
metal and polymer–ceramic composites have been extensively
studied [12–21]. The metallic powders used in ceramic metal composites are Pt, Pd, Ag, Al, Cu, Ni, etc. Among them Pt and Pd are too
* Corresponding author. Tel.: +91 471 2515294; fax: +91 471 2491712.
E-mail address: [email protected] (M.T. Sebastian).
0266-3538/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compscitech.2009.03.003
expensive and Al, Cu, Ni have the problem of oxidization while sintering at higher temperatures and require inert atmosphere. In
addition high processing temperature of cermet system limits its
compatibility with printed wired board. Polymer based composites
are advantages in terms of flexibility, compatibility, low processing
temperature and can be easily fabricated into various shapes. The
lower relative permittivity of the commercially available polymers
can be greatly enhanced by incorporating high permittivity ceramics. Commonly used high permittivity ceramics in polymer–ceramic composites are ferroelectrics which have high dielectric loss
[22,23]. In addition, lead based ceramics are not environmental
friendly and are less stable in terms of temperature and frequency.
Polymer–metal and ceramic–metal composites show an abrupt
increase in the permittivity near percolation threshold and have a
narrow smearing region [8–11,18]. In these composites, it is very
difficult to control the filler loading very close to the percolation
threshold to achieve high relative permittivity. Recently, Kim and
co-workers [24] reported that the smearing region can be expanded by introducing a third component in the two-phase composite. In this context, polymer–ceramic–metal three-phase
composites have been proposed [24–27]. This polymer based composite is advantages in terms of its broad smearing region, good
dielectric properties, low processing temperature, easily fabricate
into various shapes and the compatibility with PWB.
In the present paper we report the synthesis and dielectric
properties of three-phase composite containing epoxy as the polymer base and silver and Ca[(Li1/3Nb2/3)0.8Ti0.2]O3d (CLNT) [28]
ceramic as the fillers. In this three-phase composite, temperature
stable low loss CLNT ceramics was used as the dielectric filler
S. George, M.T. Sebastian / Composites Science and Technology 69 (2009) 1298–1302
and silver as the conducting phase to improve the dielectric properties of the composites. Epoxy was taken as the polymer base because it is thermosetting and can be easily fabricated into various
shapes.
2. Experimental
The epoxy–CLNT–Ag three-phase composites were prepared by
two step mixing and thermosetting method. The low loss CLNT
ceramics were prepared by solid state ceramic route. Different volume fractions of fine powders of sintered CLNT ceramics, epoxy
and hardener were mechanically mixed for 30 min to have uniform
distribution of ceramic powder in the matrix. The mixture was
then poured into a cylindrical mold of 10 mm diameter and
2 mm thickness. These samples were cured at 70 °C for 2 h. A series
of CLNT/epoxy composite with CLNT volume varying from 0.0 to
0.40 volume fractions were fabricated. Epoxy with 0.30 volume
fractions of CLNT was considered for further studies to visualize
the effect of metallic inclusions on the dielectric properties of
three-phase composites. The epoxy–CLNT–Ag composites in which
the volume fraction of Ag varying from 0.0 to 0.32 were prepared.
Copper electrodes were connected to both sides of the polished
compacts using silver paste and these samples were used for measuring the dielectric properties. The dielectric properties (1 kHz to
1 MHz) were measured using LCR meter (HIOKI 3532-50 LCR Hi
TESTER, Japan). The surface morphology of the samples was studied using an optical microscope (Leica, DMRX, Germany). The variation in relative permittivity with temperature was also
investigated for the composites in the temperature range 50 to
+100 °C. The theoretical densities of epoxy–CLNT–Ag three-phase
composites were calculated using the equation [18].
D¼
W1 þ W2 þ W3
W 1 =D1 þ W 2 =D2 þ W 3 =D3
ð1Þ
where W1, W2 and W3 are the wt% of CLNT, epoxy and silver in the
mixture with densities D1, D2 and D3, respectively.
3. Results and discussion
Fig. 1 shows the variation of dielectric properties of epoxy–
CLNT two-phase composites with different volume fraction of the
CLNT loading. A gradual increase in relative permittivity with CLNT
loading is observed in the epoxy–CLNT polymer ceramic composite. The overall properties of the composites mainly depend on
the properties of constituent components. The increase in the
Fig. 1. Variation of relative permittivity and dielectric loss of epoxy–CLNT
composites at 1 MHz and the comparison of experimental relative permittivity
with theoretical model.
1299
relative permittivity of the epoxy–CLNT composite is due to the
relatively high permittivity of CLNT ceramic (41 at 1 MHz) compared to epoxy. In addition, the increase of CLNT filler content in
the composite increases the interface between CLNT and polymer
and hence the influence of interfacial polarization on the relative
permittivity and dielectric loss tangent is significant. Hence the relative permittivity increases with ceramic loading. Generally, the
dielectric loss of almost all the polymer ceramic composites increases with ceramic filler loading. However, in the present
epoxy–CLNT composite the dielectric loss decreases with ceramic
loading. This could be due to the low dielectric loss of CLNT ceramic (tan d = 104) compared to that of epoxy (tan d = 102). For
0.3 volume fraction of CLNT ceramic loading, the epoxy–CLNT composite shows er = 8.0 with tan d = 0.009 at 1 MHz. As the volume
fraction increases the agglomeration of ceramic particle also increases which results in higher porosity and thereby decreases
the densification. This could be the reason for the slight increase
in the dielectric loss with higher volume fraction (Vf = 0.40) of
CLNT loading in epoxy–CLNT composite.
The composites are statistical mixtures of two or more components. The relative permittivity of these composites should lie between the relative permittivity of the constituents. Hence the
prediction of effective dielectric behavior of the composite is
important as far as its application in microelectronics is concerned.
A number of models for the description of effective properties of
randomly distributed composites have been developed. Following
equations are used to calculate the effective relative permittivity
of the polymer–ceramic composite [29–31]:
ln eeff ¼ V f ln ei þ ð1 V f Þ ln em Lichtenecker equation
eeff em
ei em
¼ Vf
Maxwell Garnet equation
eeff þ 2em
ei þ 2em
V f ðei em Þ
EMT model
eeff ¼ em 1 þ
em þ nð1 V f Þðei em Þ
ð2Þ
ð3Þ
ð4Þ
where eeff, ei, and em are the relative permittivity of the composites,
CLNT and the polymer matrix, respectively, Vf is the volume fraction
of the CLNT ceramics and n is the correction factor to compensate
for the shape of the fillers used in polymer–ceramic composites.
Fig. 1 also shows the comparison of experimentally observed
effective relative permittivity with the values predicted using the
above equations. The variation of effective permittivity of both
the polymer–ceramic composites obtained using numerical equations shows the same trend as experimentally observed results. Recently, Rao et al. [32] proposed an EMT model to predict the
effective relative permittivity of the composite. In this model the
composite can be treated in terms of an effective medium whose
effective er can be obtained by averaging over the relative permittivities of the constituents. The ceramic morphology factor ‘n’ of
the EMT model is subjected to ceramic particle and can be determined empirically. The morphology factor is found to be 0.33 for
epoxy–CLNT composite. It is found that the EMT and Maxwell Garnet models are found to match with experimental values, where as
the Litchtenecker equation is valid for low volume fraction of filler
loading. Generally, the theoretical predictions are valid for lower
volume fraction of filler loading [33,34]. This is due the imperfect
dispersion of ceramic particle in polymer matrix. Many theoretical
models suggest that the filler particles in a material should ideally
be separated, non-interacting and roughly spherical [35]. However,
many composites of interest in practice deviate markedly from this
ideal configuration. Recent studies show that, the interphase regions of polymer–ceramic composite have a relative permittivity
significantly different from that of polymer or ceramic phases
[36]. Hence the effective permittivity of the composite depends
on the permittivity of each phase in the mixture, their volume fractions, shape, size, porosity, interphase polarizabiliy and interphase
1300
S. George, M.T. Sebastian / Composites Science and Technology 69 (2009) 1298–1302
volume fractions. As a result, the effective permittivity has a nonlinear nature and consideration of all these parameters make the
calculation tedious. These could be the reason for the deviation
of experimental value of relative permittivity from the theoretical
value for higher filler loading. It is observed that the addition of
0.40 volume fraction of CLNT ceramics in the epoxy increases the
relative permittivity from 4.3 to 10.2. Report shows that higher
ceramic loading is necessary to enhance the relative permittivity
in the polymer ceramic composite for embedded capacitor applications [37]. However, higher ceramic loading dramatically decreases
the adhesion of the composite, which will reduce the processability
and reliability of embedded capacitors. Therefore a limitation of
ceramic loading is existing for polymer–ceramic composite to improve its relative permittivity. The relative permittivity of the polymer–ceramic composites can be enhanced further by incorporating
conducting phase in the two-phase composites. Hence we have
added metallic silver as a conducting component in epoxy with
0.30 volume fraction of CLNT ceramics.
Table 1 shows the densification, dielectric properties and conductivity of epoxy–CLNT–Ag composite with different volume fraction of silver loading at 1 kHz and 1 MHz. It is found that the
relative density decreases with silver loading. This could be due
to the increased porosity with silver loading. It is also noted that
silver has higher density than both ceramic and polymer. During
polishing silver may get detached from the surface and the relative
density decreased considerably. Table 1 also shows the variation of
relative permittivity, dielectric loss and conductivity of epoxy–
CLNT–Ag composites as a function of silver loading. The increase
in relative permittivity with silver loading is a result of interfacial
polarization, which occurs at the interface of dissimilar materials.
The charge carriers in the different phases of the composites are
trapped at the interface within the composite. These charges are
unable to discharge freely and give rise to an overall field distortion. This dielectric field developed around the conducting phase
can result an increase of relative permittivity [38]. The existence
of large number of conducting particles in parallel and blocked
by thin insulating layers can also increase the relative permittivity
[9–12]. Each conducting phase in the composite acts as an internal
electrode of tiny capacitors. These tiny super capacitor networks
with large electrode area and small dielectric thickness macroscopically add up to result giant permittivity. The addition of silver increases not only the permittivity but also the dielectric loss. The
increase in dielectric loss in the present three-phase composite
could be due to the increased defects and induced space charge
with silver loading. These composites are useful for a number of
applications such as flash lamp and heart actuators working at
low frequency for which large capacitance is more important than
dielectric loss [39]. As the volume fraction of silver is increased, the
clusters of silver begin to merge and above percolation limit, there
is a continuous path of adjacent allowed sites across the system.
This leads to a sudden increase in conductivity of the composite
above percolation threshold. Generally, percolation threshold of
both the ceramic–metal and polymer–metal composites is around
0.15 volume fraction of metal and have a narrow smearing region
[8–11,18]. However, in the present three-phase composite, the
ceramic particle effectively isolates the silver particles and extends
the smearing region of present high permittivity composite. Addition of 0.28 volume fraction of silver in epoxy–CLNT composite increases the relative permittivity from 8.4 to 550 at 1 kHz and 8 to
142 at 1 MHz, respectively, with low dielectric loss and low conductivity. The increase in the capacitance and relative permittivity
is not particular to a given composition of a particular dielectric
material, but depends only on the microstructure of the aggregate
and on the ratio between the conducting and insulating phase. The
broad smearing region of this composite improves the reproducibility and workability.
The present epoxy–CLNT–Ag three-phase composite can be
considered as the loading of conducting silver particle in the insulating epoxy–CLNT phase. Fig. 2 shows both experimental and theoretical values of relative permittivity of epoxy–CLNT–Ag
composite with different volume fraction of silver at 1 MHz. It is
observed that as the silver content increases from 0 to 0.26 volume
fraction, the relative permittivity of the epoxy–CLNT composite increases from 8 to 73. The concentration dependence of relative per-
Fig. 2. Variation of relative permittivity of epoxy–CLNT–Ag composites with
different volume fraction of silver below percolation threshold and the comparison
with power law.
Table 1
The relative density, dielectric properties and conductivity of epoxy–CLNT–Ag composite.
Volume fraction of silver
Relative density (%)
er (1 kHz)
Tan d (1 kHz)
r (S/cm) (1 kHz)
er (1 MHz)
Tan d (1 MHz)
r (S/cm) (1 MHz)
0.00
0.05
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0.22
0.24
0.26
0.28
0.30
0.32
99.4
99.2
99.1
98.7
97.4
97.0
96.2
94.2
94.1
91.1
87.1
85.2
83.2
81.7
80.1
8.4
11.5
14.6
16.8
18.4
22.1
23.0
24.1
30.4
34.8
65.0
95.0
550
9.73 108
4.46 106
0.013
0.011
0.030
0.019
0.026
0.030
0.026
0.025
0.025
0.031
0.065
0.090
0.870
4.400
9.900
6.26 1011
6.45 1011
2.67 1010
6.15 1010
8.75 1010
7.53 1010
9.31 1010
1.06 109
1.44 109
1.93 109
2.41 109
6.7 109
5.9 107
2.9 102
1.29
8.00
10.8
13.1
15.4
16.6
19.1
20.2
22.1
26.4
29.3
51.1
72.3
142.3
818000
510000
0.009
0.020
0.023
0.024
0.027
0.032
0.032
0.039
0.042
0.060
0.069
0.065
0.150
2.600
9.900
4.30 108
1.30 107
1.79 107
2.08 107
2.54 107
3.47 107
4.24 107
6.21 107
7.09 107
9.54 107
1.87 106
2.51 106
2.05 105
2.80 102
1.31
S. George, M.T. Sebastian / Composites Science and Technology 69 (2009) 1298–1302
1301
Fig. 3. Optical micrograph of epoxy–CLNT–Ag composite having (a) 0.10 and (b) 0.20 volume fraction of silver.
mittivity is given by the following power law on the basis of percolation theory [40]:
eðmÞ ¼ e0
q
Vc V
Vc
ð5Þ
where eo is the relative permittivity of silver free epoxy–CLNT composite, Vc is the volume fraction of silver at percolation, V is the volume fraction of silver and q is a critical exponent. The experimental
values of effective relative permittivity are in agreement with
Eq. (5), with q = 0.86 and Vc = 0.28 as can be seen from Fig. 2.
Fig. 3 shows the optical micrograph of epoxy–CLNT–Ag composite with (a) 0.10 and (b) 0.20 volume fraction of Ag. The silver
particles and CLNT ceramics are distributed randomly in the epoxy
Fig. 4. Variation of (a) relative permittivity and (b) conductivity of epoxy–CLNT–Ag
composites as a function of frequency.
matrix and there is no apparent interconnection among silver particles is observed.
Fig. 4 shows the frequency dependence of both relative permittivity and conductivity of epoxy–CLNT–Ag composite. A gradual
decrease in the relative permittivity with frequency was observed
for all the composites. This is expected since different polarization
mechanisms are frequency dependent. It has been reported that
the gradual decrease of permittivity with frequency was due to
an interfacial relaxation [17,25–27]. It can be observed that the
conductivity increases with increase in frequency for low volume
fraction of silver loading. According to percolation theory below
the percolation threshold, the conductivity of the composite is proportional to the applied frequency (r = fx, where f is the frequency
and x is the critical exponent) [41]. The frequency dependence of
relative permittivity and conductivity of disordered solids generally results from the polarization between the clusters and anomalous diffusion within each cluster. The capacitive effect between
clusters on the conductivity and relative permittivity becomes significant at high frequency. As a result, the conductivity increases
and relative permittivity decreases with frequency. Such a frequency dependence behavior of the present three-phase composite
is in agreement with previous reports [7–13].
The drift in relative permittivity with temperature should be
small for practical applications. Fig. 5 shows the variation of relative permittivity in epoxy–CLNT–Ag composite. The relative permittivity of the composite is found to be nearly independent of
temperature and it satisfies X7R EIA specifications [16]. In ceramic–metal and polymer–metal composites, above percolation
threshold the conductivity is also increased. This may lead to undesirable leakage current in the composite. However, the composite
Fig. 5. Variation of relative permittivity of epoxy–CLNT–Ag composites with
temperature and the inset shows the variation of relative permittivity for 0.30
volume fraction of silver loaded epoxy–CLNT–Ag composite.
1302
S. George, M.T. Sebastian / Composites Science and Technology 69 (2009) 1298–1302
for 0.28 volume fraction of silver loading in epoxy–CLNT composite
shows excellent dielectric properties such as high permittivity and
low conductivity and this composition is suitable for embedded
capacitor applications.
4. Conclusions
The epoxy–CLNT–Ag three-phase composites are prepared by
two step mixing and thermosetting method. Near the percolation
threshold (0.28 volume fraction) the composite shows a relative
permittivity of 142 with low dielectric loss (0.15) and low conductivity (105 S/cm) at 1 MHz and is suitable for embedded capacitor
applications. The composite under study has low processing temperature and shows a broad smearing region which improves the
reproducibility and workability. The excellent dielectric properties,
low processing temperature and broad smearing region of epoxy–
CLNT–Ag three-phase composite can cover an important technological gap in the microelectronic industry.
Acknowledgements
The authors are grateful to the Defense Research and Development Organization and CSIR, New Delhi, India for the financial
assistance.
References
[1] Chen F, Chu C, He J, Yang Y, Lin J. Organic thin-film transistors with
nanocomposite dielectric gate insulator. Appl Phys Lett 2004;85(15):3295–7.
[2] 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 Polym Sci 2002;83(5):1084–90.
[3] Bur JA. Dielectric properties of polymer at microwave frequencies. Polymer
1985;26:963–97.
[4] Sebastian MT. Dielectric materials for microwave communications. UK:
Elseiver Publishers; 2008.
[5] Dang Z-M, Lin Y-H, Nan C-W. Novel ferroelectric polymer composites with
high dielectric constant. Adv Mater 2003;15(19):1625–9.
[6] Huang C, Zhang Q. Enhanced dielectric and electromechanical response in high
dielectric constant all-polymer percolative composites. Adv Funct Mater
2004;14(5):501–6.
[7] Li Y-J, Xu M, Feng J-Q, Dang Z-M. Dielectric behavior of a metal–polymer
composite with low percolation threshold. Appl Phys Lett 2006;89:072902–3.
[8] Li JY, Huang C, Zhang Q. Enhanced electromechanical properties in all-polymer
percolative composites. Appl Phys Lett 2004;84(16):3124–6.
[9] Pecharroman C, Moya JS. Experimental evidence of a giant capacitance in
insulator–conductor composites at the percolation threshold. Adv Mater
2000;12(4):294–7.
[10] Pecharroman C, Betegon FE, Bartolome JF, Esteban SL, Moya JS. New
percolative BaTiO3–Ni composites with a high and frequency-independent
dielectric constant (er 80000). Adv Mater 2001;13(20):1541–4.
[11] George S, James J, Sebastian MT. Giant permittivity of bismuth zinc niobate–
silver composite. J Am Ceram Soc 2007;90(11):3522–8.
[12] Deepa KS, Sebastian MT, James J. Effect of interparticle distance and interfacial
area on the properties of insulator–conductor composites. Appl Phys Lett
2007;91(20):202904.
[13] Choi H-W, Heo Y-W, Lee J-H, Kim J-J, Lee H-Y, Park E-T, et al. Effects of BaTiO3
on dielectric behavior of BaTiO3–Ni–polymethyl methacrylate composites.
Appl Phys Lett 2006;89:132910–3.
[14] Xiang P-H, Dong X-L, Feng C-D, Liang R-H, Wang Y-L. Dielectric behavior of
lead zirconate titanate/silver composites. Mater Chem Phys 2006;97:410–4.
[15] Xu J, Wong CP. Low-loss percolative dielectric composite. Appl Phys Lett
2005;87:082907–9.
[16] Chen RZ, Wang X, Gui ZL, Li LT. Effect of silver addition on the dielectric
properties of barium titanate-based X7R ceramics. J Am Ceram Soc
2003;86(6):1022–4.
[17] Panteny S, Bowen CR, Stevens R. Characterisation of barium titanate–silver
composites part II: electrical properties. J Mater Sci 2006;41:3845–51.
[18] George S, Sebastian MT. Effect of sintering temperature on the percolation
threshold and the dielectric properties Ca[(Li1/3Nb2/3)1xTix]O3d–Ag
composite. Compos Sci Technol 2008;68:2461–7.
[19] Duan N, Elshof JE, Verweij H, Greuel G, Dannapple O. Enhancement of
dielectric and ferroelectric properties by addition of Pt particles to a lead
zirconate titanate matrix. Appl Phys Lett 2000;77(20):3263–5.
[20] Kirkpatrick S. Percolation and conduction. Rev Mod Phys 1973;45:574–88.
[21] Grannan DM, Garland JC, Tanner DB. Critical behavior of the dielectric constant
of a random composite near the percolation threshold. Phys Rev Lett
1981;46(5):375–8.
[22] Bai Y, Cheng Z-Y, Bharti V, Xu HS, Zhang QM. High dielectric constant ceramic
powder polymer composite. Appl Phys Lett 2000;76:3804–6.
[23] Hwang HJ, Yasunoka M, Sando M, Toriyama M, Niihara K. Low-temperature
sintering and high-strength Pb(Zr,Ti)O3d matrix composites incorporating
silver particles. J Am Ceram Soc 1997;80(3):791–3.
[24] Lim ES, Lee JC, Kim JJ, Park ET, Chung YK, Lee HY. Dielectric characteristics of
polymer–ceramic–metal composites for the application of embedded passive
devices. Integr Ferroelectr 2005;74:53–60.
[25] Dang Z-M, Lin Y-H, Nan C-W. Dielectric behavior of three phase percolative
fluoride
composites.
Appl
Phys
Lett
Ni–BaTiO3/polyvinylidene
2002;81(25):4814–6.
[26] George S, Sebastian MT. Enhanced permittivity by the adhesion of conducting
and low loss insulating ceramics in polystyrene. J Appl Polym Sci, in press.
[27] Prakash BS, Varma KBR. Dielectric behavior of CCTO/epoxy and Al–CCTO/
epoxy composites. Compos Sci Technol 2007;67:2363–8.
[28] Choi J-W, Kang C-Y, Yoon S-J, Kim H-J, Jung H-J. Microwave dielectric
properties of Ca[(Li1/3Nb2/3)1xMx]O3d (M = Sn, Ti) Ceramics. J Mater Res
1999;14:3567–70.
[29] Sihvola AH. Effective permittivity of dielectric mixture. IEEE Trans Geosci
Remote Sens 1988;26(4):420–9.
[30] Sareni B, Krahenubuh L, Beroul A, Brosseau C. Effective dielectric constant of
random composite material. J Appl Phys 1997;81(5):2375–83.
[31] Jylhai L, Sihvola AH. Numerical modeling of disordered mixture using pseudo
random simulations. IEEE Transc Geosci Remote Sens 2005;43(1):59–64.
[32] Rao Y, Qu J, Marinis T, Wong CP. A precise numerical prediction of effective
dielectric constant for polymer ceramic composite based on effective medium
theory. IEE Trans Compon Pack Technol 2000;23:680–3.
[33] 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
2007;68:106–11.
[34] Subodh G, Joseph M, Mohanan P, Sebastian MT. Low dielectric loss
polytetrafluoroethylene/TeO2 polymer ceramic composites. J Am Ceram Soc
2007;90(11):3507–11.
[35] Stroud D. Percolation effects and sum rules in the optical properties of
composites. Phys Rev B 1979;19:1783–91.
[36] Todd MG, Shi FG. Characterizing the interphase dielectric constant of polymer
composite materials: effect of chemical coupling agents. J App Phys
2003;94(7):4551–7.
[37] Bhattacharaya SK, Tummala RR. Next generation integral passives: materials,
processes, and integration of resistors and capacitors on PWB substrates. J
Mate Sci Mate Electron 2000;11:253–68.
[38] Chylek P, Srivastava V. Effective dielectric constant of a metal-dielectric
composite. Phys Rev B 1984;30(2):1008–9.
[39] Lai Q, Lee BI, Samuels WD, Exarhos GJ, Parler SG. Three-phase percolative
silver–BaTiO3–epoxy nanocomposites with high dielectric constants. J Appl
Poly Sci 2006;102:967–71.
[40] Efors EL, Shktovskii BI. Critical behaviour of conductivity and dielectric
constant near the metal–nonmetal transition threshold. Phys Status Solidi B
1976;76(2):475–85.
[41] Nan CW. Physics of inhomogeneous inorganic materials. Prog Mater Sci
1993;37:1–116.