Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.

)
_______________________________________________________________________________________________

A Study On Thermal and Dielectric Characteristics of Solid Glass

Microsphere filled Epoxy Composites
D.Mishra*,1and A. Satapathy2
1
Department of Mechanical Engineering, VSSUT, Burla, 768018
2
Department of Mechanical engineering, NIT, Rourkela, 769008

This research deals with the processing and characterization of solid glass micro-sphere filled epoxy composites. The first
part of the work includes the development of two theoretical correlations based on one dimensional heat conduction
models for estimation of effective thermal conductivity of polymer composites with single and multiple fillers. The second
part depicts the details of the test procedures and test results in regard to the physical, mechanical and micro-structural
characteristics of the epoxy composites filled with solid glass microspheres (SGM) and/or micro-sized boron nitride (BN).
The last part throws light on the thermal and dielectric characteristics of the composites with different filler type and
concentrations. The estimation of effective thermal conductivity of the composites using finite element method (FEM) and
using the proposed theoretical models is done and the results are validated by corresponding experimental results. The
effects of inclusion of SGM and/or BN on the effective thermal conductivity (k eff), glass transition temperature (Tg),
coefficient of thermal expansion (CTE), electrical resistivity (ρ) and dielectric constant (D k) of epoxy composites are also
studied.

Keywords: Solid Glass Microspheres, Boron Nitride, Insulation, Percolation

1. Introduction
With the rapid development of the electronic information industry, better properties are required for substrate and
packaging materials, such as high thermal conductivity, low coefficients of thermal expansion (CTE), low dielectric
constant and thermal stability. Polymers, such as polyethylene epoxy, and polyimide are ordinarily used as these
materials due to their high electrical resistivity, low dielectric constant, and excellent processability. However, these
polymers with low thermal conductivity cannot effectively dissipate the heat generated in the devices and their high
CTE results in thermal failure. In addition to the thermal and electrical properties, the mechanical properties of substrate
and packaging materials are also very important. Polymers usually exhibit low stiffness and strength. To offset these
deficiencies, adding inorganic particles to a polymer is a versatile method. This method synergistically integrates the
advantages of polymers and inorganic fillers; thus, the thermal, electrical and mechanical properties of the composites
can be improved by properly selecting the filler components, shapes, sizes and concentrations. Besides, epoxy resin
systems are used for encapsulating a variety of electronic components because of their high thermal stability, high
moisture resistance and low cost. But, unfortunately, the cured epoxy resin has poor thermal properties, which are the
high CTE and low thermal conductivity. Although the CTE of epoxy resin can be lowered with the addition of some
ceramic fillers, the improvement in thermal conductivity remains an increasingly pressing issue.
The heat transfer process of heterogenous materials is very complicated, especially for polymer composites. It is
quite important, therefore, to understand the mechanisms of heat transfer in polymer composites, which are potential
insulating materials. There are many situations where polymers are used for providing thermal insulation. Foamed
plastic is a polymeric material commonly used for thermal insulation. But its application is limited considerably due to
its poor mechanical properties. Therefore, there has been a focus on fabricating a kind of reinforced polymeric system
which is light but has better mechanical strength and good thermal insulation properties. In this context, rigid glass
micro-spheres (glass beads) have some advantages as fillers in polymers such as low thermal conductivity, CTE and
density. In addition, these micro-particles do not generate stress concentration in the interface between the fillers and
the matrix owing to their smooth spherical surface. This type of composites can be applied in building materials, space
flight and aviation industry. Glass beads are preferred as fillers especially when composite properties such as isotropy
or low melt viscosity are important. Moreover, the orientation effects associated to moulding are minimal.
In view of this, the present research is undertaken to study the thermal as well as dielectric properties of a class of epoxy
based polymer composites tailor-made for applications ranging from thermal insulation to encapsulation of electronic
components.

2. Review of Literature
The improved performance of polymers and their composites in industrial and structural applications by the addition of
solid filler materials has shown great promise and so has lately been a subject of considerable interest. Various kinds of
polymers, and polymer matrix composites reinforced with metal particles have a wide range of industrial applications

13
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

such as heaters, electrodes Jung-il et al [1] , composites with thermal durability at high temperature Nikkeshi et al [2]
etc. These composites are desired due to their low density, high corrosion resistance, ease of fabrication and low cost
(Tavman et al [3], Zhu [4], Rusu et al [5], Cheng et al [6]). Increasing use of polymer composites for various
applications emphasizes its importance/significance in the thermal property analysis of an engineering system.
Conductivity is one such important thermal property that needs to be evaluated for any new composite system.
Generally, measuring the thermal conductivity accurately is helpful to study the heat transfer process and mechanisms
in composite materials. Although it can be measured by experimental methods, analytical methods and equations are
often essential to predict thermal conductivities of composite materials.
For porous materials, several researchers have derived effective thermal conductivity equations based on the
Maxwell expression [7], or established a more accurate formula for calculating the effective thermal conductivity of
porous materials. The models proposed respectively by Nielsen and Cheng-Vochon [8] can better estimate the effective
thermal conductivity of filled composite materials, while the Agari–Nagai equation can predict the effective thermal
conductivity of composites with high-loading [9]. Liang [10] analyzed the thermal conductivity of a porous material
with closed spherical and cylindrical holes. Suvorov et al. [11] studied the thermal conductivity of hollow emery filled
composites. Considerable work has been reported on the subject of heat conductivity in polymers by Hansen and Ho
[12], Peng and Landel [13], Choy and Young [14], Tavman [15], etc. Most of these studies were confined to the thermal
behaviour of neat polymers only and not to their composites. Reports are available in the existing literature on
experimental as well as numerical and analytical studies on thermal conductivity of some filled polymer composites
[16-19]. The fillers most frequently used are aluminum particles, copper particles, brass particles, short carbon fibers,
carbon particles, graphite, aluminum nitrides and magnetite particles. Progelhof et al. [20] was the first to present an
exhaustive overview on models and methods for predicting the thermal conductivity of composite systems. Procter and
Solc [21] used Nielsen model as a prediction to investigate the thermal conductivity of several types of polymer
composites filled with different fillers and confirmed its applicability. Nagai and Lai [22] found that Bruggeman model
for Al2O3-epoxy system and a modified form of Bruggeman model for AlN-epoxy system are both good prediction
theories for thermal conductivity. Yu et al. [23] measured the thermal conductivity of polystyrene-aluminium nitride
composite, and found that the thermal conductivity of composites was higher for a polystyrene particle size of 2mm
than that for a particle size of 0.15 turn. The thermal conductivity of the composite was five times that of pure
polystyrene at about 20% volume fraction of aluminium nitride for the composite containing 2 mm polystyrene
particles. Griesinger et al. [24] reported that thermal conductivity of low-density PE increased from 0.35 for an isotropic
sample, to the value of 50 W/ m K for a sample with an orientation ratio of 50. The thermal and mechanical properties
of copper powder filled poly-ethylene composites are found by Tavman [25] while Sofian et al. [26] investigated
experimentally on thermal properties such as thermal conductivity, thermal diffusivity and specific heat of metal
(copper, zinc, iron, and bronze) powder filled HDPE composites in the range of filler content 0–24% by volume. They
observed a moderate increase in thermal conductivity up to 16% of metal powder filler content. Mamunya et al. [27]
also reported the improvement in electrical and thermal conductivity of polymers filled with metal powders. In a recent
research, Weidenfeller et al. [28] studied the effect of the interconnectivity of the filler particles and its important role in
the thermal conductivity of the composites. They prepared PP samples with different commercially available fillers by
extrusion and injection molding using various volume fractions of filler content to systematically vary density and
thermal transport properties of these composites. Surprisingly, they measured that the thermal conductivity of the PP
has increased from 0.27 up to 2.5 W/m K with 30 vol. % talc in the PP matrix, while the same matrix material
containing the same volume fraction of copper particles had a thermal conductivity of only 1.25 W/m K despite the fact
that copper particles have a thermal conductivity approximately 40 times greater than that of talc particles. Tekce et al.
[29] noticed the strong influence of the shape factor of fillers on thermal conductivity of the composite. While Kumlutas
and Tavman [30] carried out a numerical and experimental study on thermal conductivity of particle filled polymer
composites, Patnaik et al. [31] reported the existence of a possible correlation between thermal conductivity and wear
resistance of particulate filled composites. Recently, Hill and Supancic [32] proposed an indirect method to determine
this interfacial boundary resistance by preparing large-scale “macro-model” simulations of the polymer-ceramic
interface. They also investigated the effects of similar size and shape of platet shaped particles on the thermal
conductivity of polymer/ceramic composite materials. Yunsheng Xu, D.D.L.Chung and Cathleen Mroz [33] dealt with
thermally conducting aluminium nitride polymer-matrix composites and studied its dielectric behaviour. Shu-Hui Xie,
et.al. [34] explained the preparation and properties of polyamide/ aluminium nitride composites. Lee and Mock [35]
studied on enhancing thermal conductivity of polymer matrix composite of AlN in epoxy resins and analysed its
dielectric behaviour.
There are only a few published papers on evaluation of effective thermal conductivity of polymer composites filled
with glass beads [36-38]. Liang and Li [36] reported on measurement of thermal conductivity of hollow glass-bead-
filled polypropylene composites. Recently, they [37] also made two-dimensional and three-dimensional finite element
analysis on the heat transfer and simulated the variation of effective thermal conductivity of hollow glass microsphere
filled polymer composites. Liang and Li [38] further studied the heat transfer in polymer composites filled with hollow
glass micro-spheres and proposed a theoretical model to predict the thermal conductivity of such composite system.
Yung et al[39] reported on the preparation and properties of hollow glass microsphere-filled epoxy matrix composites.

14
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

3. Objectives of the present work

The objectives of this work are outlined as follows:
1. Development of a theoretical model for estimation of effective thermal conductivity (k eff) of micro-sphere
filled polymer composites.
2. Estimation of keff of SGM filled epoxy composites using proposed model.
3. Estimation of keff of SGM filled epoxy composites using finite element method (FEM) and other existing
models.
4. Fabrication of a set of SGM filled epoxy composites with an objective to improve the insulation properties of
neat polymer.
5. Measurement of keff of SGM filled epoxy composites by experimental method
6. Comparison of keff obtained from various correlations with the measured values.
7. Validation of the proposed model and FEM model by comparing with the keff values obtained experimentally.
8. Physical, Mechanical and Micro-structural Characterization of the composites
9. Fabrication of a series of composites filled with SGM and another conductive filler
10. Determination of keff, Tg and CTE of these hybrid composites and studying the effects of different filler type
and content on these thermal properties
11. Study of their dielectric behaviour at different operating frequencies and exploring the possibility of their use
in electronic devices

4. Theoretical Model for Effective Thermal Conductivity

The theoritical analysis of the heat transfer in this paper is based on the following suppositions: (a) the distribution or
dispersion of the solid micro-spheres in the polymer matrix is uniform; (b) the temperature distribution along the
direction of heat flow is linear. An element from the composite is selected for analysis which is a straight cube with
side length of H and there is a solid glass micro-sphere of radius (r) in the element. The element is divided into polymer
phase and micro-sphere phase.The heat quantity Q transfers from bottom to the top. The heat transport in solid glass
microsphere filled polymer composites has two mechanisms: (i) solid thermal conduction and (ii) heat radiation on the
surface between neighbouring particles. Polymer composite works usually under lower temperature conditions where
the proportion of the thermal radiation in the total heat transfer is very small, hence the thermal radiation is neglected.
Finally, the expression for effective thermal conductivity of the composite is deduced as:
6𝑣𝑓 1/3 4𝜋 2𝑣𝑓 1/3 𝜌𝑠
𝑘𝑒𝑓𝑓 = [1/𝑘𝑝 (1 − ) +2(𝑘𝑝 ( )1/3 +𝜋( ) × (𝑘𝑔 − 𝑘𝑝 ))−1 ]−1 (1)
𝜋 3𝑣𝑓 9𝜋 𝜌𝑔

Here, kp and kg are the respective heat conductivities of the polymer and the micro-sphere phase, 𝜌s and 𝜌g are the
effective densities of the polymer and the microsphere phase respectively, and vf is the volume fraction of the filler i.e.
the SGM in the composite.

5. Experimental Details

5.1 Composite fabrication

SGMs are manufactured from high grade Soda Lime Silica Glass containing SiO 2 around 70%. They are free from free
Lead, Iron and Silica. SGMs are produced by firing crushed glass with subsequent collection and cooling of spheroidal
product or by melting the formulated glass batch and subsequent breakup of the free falling molten stream to form small
droplets. SGM are basically those of their precursor silicate glasses.
Low temperature curing Epoxy LY 556 resin, used as the matrix material and the hardener (HY951) are mixed in a
ratio of 10:1 by weight as recommended. Epoxy is chosen primarily of its low density (1.1 gm/cc) and low thermal
conductivity (0.363 W/m K). Solid glass micro-spheres (SGM) of three different sizes (100, 200 and 300 micron
diameter) are reinforced in the resin to prepare the composites. The dough (epoxy filled with SGM) is then slowly
decanted into the glass molds, coated beforehand with wax and a uniform thin film of silicone-releasing agent. The
composites are cast in these molds so as to get disc type cylindrical specimens (dia 20 mm, thickness 5 mm).
Composites of 8 different compositions (0, 5, 10, 15, 20, 25, 30 and 35 vol % of SGMs respectively) of each size are
made. The castings are left to cure at room temperature for about 24 hours after which the glass molds are broken and
samples are released.

15
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

5.2 Experimental determination of thermal conductivity

Unitherm™ Model 2022 is used to measure thermal conductivity of the composites fabricated for this investigation.
This is an instrument based on guarded heat flow method and is used for a variety of materials. The test is carried out in
accordance with ASTM E-1530 standard.

5.3 Operating principle of Unitherm™ 2022

A sample of the material is held under a uniform compressive load between two polished surfaces, each controlled at a
different temperature. The lower surface is part of a calibrated heat flow transducer. The heat flows from the upper
surface, through the sample, to the lower surface, establishing an axial temperature gradient in the stack. After reaching
thermal equilibrium, the temperature difference across the sample is measured along with the output from the heat flow
transducer. These values and the sample thickness are then used to calculate the thermal conductivity. The temperature
drop through the sample is measured with temperature sensors in the highly conductive metal surface layers on either
side of the sample. By definition thermal conductivity means ‘‘The material property that describes the rate at which
heat flows within a body for a given temperature change.” For one-dimensional heat conduction the formula can be
given as Eq. (2).
𝑻 −𝑻
𝑄 = 𝐾𝐴 𝟏 𝟐 (2)
𝒙

where Q is the heat flux (W), K is the thermal conductivity (W/m K), A is the cross-sectional area (m2), T1 - T2 is the
difference in temperature (K), x is the thickness of the sample (m). The thermal resistance of a sample can be given as:

𝑻𝟏 −𝑻𝟐
𝑅= (3)
𝑸⁄𝑨

where R is the resistance of the sample between hot and cold surfaces (m2 K/W). From Eqs. (2) and (3) we can derive
that
𝒙
𝐾= (4)
𝑹

In Unitherm™ 2022 the heat flux transducer measures the Q value and the temperature difference can be obtained
between the upper plate and lower plate. Thus the thermal resistance can be calculated between the upper and lower
surfaces. Giving the input value of thickness and taking the known cross-sectional area, the thermal conductivity of the
samples can be calculated using Eq. (4).

6. Results and Discussion

The determination of effective properties of composites is of paramount importance for functional design and
application of composite materials. One of the important factors that influence the effective properties and can be
controlled to an appreciable extent is the microstructure of the composite. Here, microstructure means the shape, size
distribution, spatial distribution and orientation distribution of the reinforcing inclusion in the matrix. Although most
composite possess inclusion of random distributions, great insight of the effect of microstructure on the effective
properties can be gained from the investigation of composites with periodic structure. System with periodic structures
can be more easily analyzed because of the high degree of symmetry embedded in the system. A typical periodic
arrangement of solid glass beads within the epoxy body is schematically shown in Fig. 1. Fig. 2 clearly illustrates the
heat flow direction and the boundary conditions for the particulate-polymer composite body considered for the analysis
of this conduction problem. The temperature at the nodes along the surfaces ABCD is prescribed as T 1 (=1000C) and the
ambient convective heat transfer coefficient is assumed to be 25 W/m2-K at room temperature of 27°C. The other
surfaces parallel to the direction of the heat flow are all assumed adiabatic. The temperatures at the nodes in the interior
region and on the other boundaries are unknown. These temperatures are obtained with the help of the finite-element
program package ANSYS. In this analysis it is assumed that the composites are macroscopically homogeneous, locally
both the matrix and filler are homogeneous and isotropic, the thermal contact resistance between the filler and the
matrix is negligible and the composite lamina is free from voids. The problem is based on 3D physical model and the
filler arranged in a square periodic array are assumed to be uniformly distributed in the matrix.

16
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

F G

E
H
a
Kf

D
K T2 C
A a
y a B
x Q

Fig 1. 3D Sphere-in-cube model Fig 2. Boundary conditions

Using the theoretical model proposed in this work, the effective thermal conductivity k eff of the composites are
calculated. The variation of keff with the SGM (100 micron) content is shown in Fig.3, which also presents a comparison
of the calculated values with the experimentally measured ones.
0.4 0.4

Theoretical value SGM 100 micron size

Measured value SGM 200 micron size
SGM 300 micron size
0.3 0.3
K eff (W/m-K)

K eff (W/m-K )

0.2 0.2

0.1
0.1
0 5 10 15 20 25 30 35
0 5 10 15 20 25 30 35
SGM content (vol%) SGM content ( vol %)

Fig 3. Variation of keff with SGM content: Comparison of Fig 4. Variation of keff with SGM content: Effect of
theoretical and experimental values SGM size

0,45
Maxwell model
Effective thermal conductivity

Lewis-Neilson model Russell's

0,7
Effective thermal conductivity

0,4 Series Model

Russell model
Lewis Nielson
0,35 0,6 Model
(W/mK )

FEM
(W/m-K )

0,3 0,5

0,25 0,4
0,2
0,3
0 1,4 3,4 6,5 11,3 17,9 0 1,4 3,35 5,23 7,85 9,04 11,3
SGM content (vol %) BN content (vol %)

Fig 5. Variation of keff with SGM content: Comparison of values Fig 6. Variation of effective thermal conductivity with BN
obtained from various models with the measured values content: Comparison of values

17
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

It is seen that the results obtained from the proposed correlation are in good agreement with experimental results. Fig. 4
presents the variation of keff with filler content for different glass micro-sphere sizes and it is evident that the reduction
in heat conduction capability is maximum for the SGMs with 100 micron size. It is encouraging to note that the
incorporation of SGM results in significant drop in thermal conductivity of epoxy resin. With addition of 35 vol. % of
SGM (100 micron), the thermal conductivity decreases by about 66% as compared with neat epoxy resin. It is also
noted that, smaller the size of solid glass micro-spheres, higher is the drop in conductivity value of the composites.

6.1.Test Results: Thermal Conductivity of Epoxy Composites Filled with Boron Nitride Particles in Different
Proportions
In a similar manner, the temperature plots for Boron Nitride-Epoxy composites are numerically obtained using the
software ANSYS. It is noticed that the results obtained from the finite-element analysis taking sphere-in-cube
composite model are reasonably closer to the measured values of effective thermal conductivity for composites of
different filler content from 0% to 17.9% volume. The percentage errors associated with the FEM values with respect to
the experimental values is given in Table 1. It can be ascertained from the table that the errors associated with sphere-
in-cube model simulations for Boron Nitride filler lie in the range 0-9 %. On comparing, it is noticed that FEM
overestimates the value of thermal conductivity, with respect to the experimental ones unlike in the case of SGM as
filler where it shows a constant reduction in the value of thermal conductivity.

Table 1. Comparison of variation of thermal conductivity of existing models with FEM and experimental values
Sample BN ROM (Series Maxwell- Lewis- Theoretical FEM Experimental
(vol %) Model) model Nielson model
model

1 0 0.363 0.363 0.363 0.363 0.363 0.363

2 1.4 0.368 0.378 0.385 0.3525 0.368 0.372
3 3.35 0.376 0.400 0.419 0.3598 0.369 0.401
4 5.23 0.383 0.422 0.456 0.3654 0.4025 0.407
5 7.85 0.394 0.454 0.514 0.3738 0.424 0.432
6 9.04 0.398 0.469 0.545 0.3774 0.4335 0.442
7 11.3 0.410 0.499 0.611 0.3847 0.4545 0.464

18
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

Fig 7(a) Temperature profile for epoxy-SGM composite with Fig 7(b). Temperature profile for epoxy-SGM composite with
filler concentration of 1.4 vol % filler concentration of 3.35 vol %

Fig 7(c). Temperature profile for epoxy-BN composite with Fig 7(d). Temperature profile for epoxy-BN composite with
filler concentration of 9.04 vol % filler concentration of 11.3 vol %

Fig. 7(e)Temperature profile for epoxy-BN composite

The temperature plots obtained from FEM analysis are presented in Figs. 7(a), 7(b), 7(c), 7(d) and 7(e) with
particulate concentrations of 1.4 and 3.35 vol% of SGM and 9.04, 11.3 and 17.9vol% of BN respectively. The
numerical values are compared with the experimental as well as existing theoretical and empirical models.
Fig.6.gives a comparative picture of the thermal conductivity of the simulated results estimated from the FEM analysis
(spheres-in cube model) and the experimental values for BN as filler in the epoxy matrix. The variation of the effective
thermal conductivity with the change in volume concentration of the particulate filler in the composite is also presented.

19
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

It was noted that there is an anomaly in the conductivity of measured and the simulated values because some of the
premonitions are not real like the shape of BN is assumed to be spherical whereas the actual shape of BN particles are
somewhat elliptical in nature. With addition of 11.3vol% of BN (100 micron size), the thermal conductivity increases
by about 66 % as compared to neat epoxy resin. For any given content of BN filler, higher percentage of BN will
achieve higher thermal conductivity. At a volume fraction above 20vol% the BN particles show percolation behaviour
and at that point the thermal conductivity increases swiftly. This is the critical concentration at which BN particles start
contacting with each other and hence the actual size of the agglomerates becomes larger. However, it again leads to a
conclusion that for this kind of particulate filled composite, the finite element analysis serves very well as a predictive
mechanism for the feasibility of the effective thermal conductivity for a certain range of particle concentration of
particulate fillers. FEM values are pretty much closer to the experimental as well as the values obtained from the
proposed model and hence we can aptly justify FEM simulation technique as a very efficient, credible and holistic
approach for the validation of our thermal boundary value problem.
It is found that the thermal conductivity of BN-filled composite is higher than that of AlN-filled composites. This is
because the intrinsic conductivity of BN is greater than that of AlN and it is also due to the fact that BN fillers consist of
micro-sized platelet shaped and nano-sized spherical shaped BN which enhances the formation of the conductive
network in the composites.

6.2.Evaluation of Effective Thermal Conductivity Of Particulate Filled Epoxy Composites (With Hybrid
Fillers)
Miniaturization of devices due to their less space and higher portability is the demand of the customer of recent times.
Applications in electronic industry therefore require materials with high thermal conductivity and low dielectric
constant for encapsulation. Basic electronic components may be packaged discretely, as arrays or networks of like
components or integrated inside packages such as semiconductor integrated circuits, hybrid integrated circuits, or thick
film devices. A modern packaging application therefore embodies a host of materials and necessitates an integrated
design approach from the very outset. Last few decades have seen a rapid advancement in electronics technology as
there is a constant demand for miniaturization and hence the sizes are gradually shrinking whereas the number of
components and communication speed are increasing day-by-day which leads to a generation of high amount of heat
and gives rise to problems in heat dissipation. Epoxy is commonly used as material for heat sink applications but it
suffers from a drawback of low thermal conductivity and single filler might not satisfy both these requirements.
Therefore, in order to enhance the thermal conductivity and simultaneously keep a low dielectric constant, a thermally
conductive filler i.e. BN with suitable volume fraction is added onto these glass micro-sphere filled epoxy composites.
There are a plenty of theoretical as well as empirical models to predict the effective thermal conductivity of two-phase
mixtures. Materials are arranged in either parallel or series with respect to heat flow, which gives upper and lower
bounds of effective thermal conductivity which is otherwise called the Rule of mixture model but not much attention is
focused on the arrangement of three phase mixtures. An equation for the effective thermal conductivity can therefore be
deduced for a three component hybrid system with parallel heat flow as

𝑘𝑐 = (1 − ɸ1 − ɸ2 ). 𝑘𝑚 + ɸ1 . 𝑘𝑓1 + ɸ2 . 𝑘𝑓2 (5)

where km , kf1 and kf2 are the thermal conductivities of epoxy, SGM and of BN respectively and
ɸ1 = vol % of SGM fille , ɸ2 = vol% of BN filler

20
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

0,6
0.8
0 vol% SGM

Effective thermal conductivity (W/m-K )

Effective thermal conductivity
5 vol% SGM
0.7
10 vol% SGM
0,5 15 vol% SGM
0.6 20 vol% SGM
0,4
0.5
(W/m-K)
0,3
0.4

0,2
0.3

0.2
0,10 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5
BN content (vol%)

Fig.8. Variation of thermal conductivity with BN volume fraction for hybrid composites obtained from the proposed theoretical
model

The experimental values of thermal conductivities of composites with different SGM content for a filler
concentration of 0, 5, 10, 15 and 20 vol% are shown in Fig.8. There is a decrease in thermal conductivity with
increasing SGM content. The thermal conductivity of neat epoxy is 0.363 W/mK which is reduced to 0.2984 W/mK (by
8.22%) on addition of 20 vol% of SGM. Unitherm™ Model 2022 is used to measure thermal conductivity of these
samples. The tests are in accordance with ASTM E-1530 standard. Fig.13 shows the change in thermal conductivity in
accordance with the increase in BN content with SGM content of 20 vol%. When BN is added along with SGM fillers
in epoxy matrix, the effective thermal conductivity is ameliorated as BN is conductive in nature. For 10vol% of BN, the
thermal conductivity improves significantly by 31.01%.
2 2
Epoxy with SGM (5 vol%)
Epoxy with SGM (0 vol%) 1.8
Effective thermal conductivity (W/m-K)

1.8
Theoretical Model
Effective thermal conductivity (W/m-K)

Theoretical Model 1.6

1.6 Experimental
Experimental 1.4
1.4
1.2
1.2
1

1 0.8

0.8 0.6

0.6 0.4

0.2
0.4
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5

0.2 BN content (vol%)

0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5

BN content (vol%)

Fig. 9.Variation of thermal conductivity with BN volume Fig.10. Variation of thermal conductivity with BN volume
fraction for hybrid composites obtained from the proposed fraction for hybrid composites obtained from the proposed
theoretical model theoretical model

21
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

2 1.8
Epoxy with SGM (10 vol%) Epoxy with SGM (15 vol%)
1.8
Effective thermal coductivity (W/m-K )

Effective thermal conductivity (W/m-K)

1.6
1.6 Theoretical Model Theoretical Model
1.4
1.4 Experimental Experimental
1.2
1.2

1 1

0.8
0.8
0.6
0.6
0.4

0.2 0.4
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5
0.2
BN content (vol%)
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5
BN content (vol%)

Fig.12. Variation of thermal conductivity with BN volume

Fig.11. Variation of thermal conductivity with BN volume
fraction for hybrid composites obtained from the proposed
fraction for hybrid composites obtained from the proposed
theoretical model
theoretical model

2
Effective thermal conductivity (W/m-K)

1.6 0 vol% SGM

Epoxy with SGM ( 20vol% ) 1.8
5 vol% SGM
Effective thermal conductivity (W/m-K )

1.4 1.6 10 vol% SGM

Theoretical Model
Experimental 15 vol% SGM
1.4
1.2 20 vol% SGM
1.2
1
1

0.8 0.8

0.6
0.6
0.4
0.4
0.2
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5
0.2
0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5
BN content (vol%)

BN content (vol%)

Fig.14. Variation of thermal conductivity with BN volume Fig.13. Variation of thermal conductivity with BN volume
fraction for hybrid composites obtained from the proposed fraction for hybrid composites obtained from the proposed
theoretical model theoretical model

22
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

1,8 Theoretical Model

Effective Thermal Conductivity (W/mK )

Experimental
1,6

1,4 Percolation threshold

1,2

0,8

0,6

0,4

0,2

0
0 0,025 0,05 0,075 0,1 0,125 0,15 0,175 0,2 0,225 0,25 0,275 0,3

BN content (volume fraction)

Fig.15. Variation of thermal conductivity with BN volume fraction for hybrid composites obtained from the proposed theoretical
model

Fig. 8, Fig.9, Fig.10, Fig.11, Fig.12, Fig.13 and Fig.14 represent the variation of thermal conductivity at different
volume fractions of BN with the increase in SGM content for 0, 5, 10 , 15 and 20 vol% of SGM respectively. Fig.15.
clearly shows the percolation behaviour at 20 vol% of BN for different volume fractions of SGM. Immediately after the
percolation threshold, a slight increase in the concentration of conductive particles may greatly increase the bridges in
the conducting network which is shown in Fig.15. The insulating composite is thus transformed into a conducting
material in a jump wise fashion. Further increase in the concentration of the conductive particles, however, may only
cause the volume of the conducting domains to increase without any significant increase in the pathways for electrons,
leading to a monotonic increase in conductivity.

6.3.Glass Transition Temperature Of SGM and BN Filled Epoxy Composites

Glass transition temperature is the temperature at which the mechanical properties of amorphous polymer changes from
the brittle state of glass to a rubbery state. The most important property change at the glass transition temperature of the
PCB materials is the thermal expansion, which is swift from a relative low value to a very high value. This kind of
change is not desirable as it imposes stress on the PCBs when they experience high-temperature stress during
manufacturing, assembly or during their service-life. The glass transition temperature (Tg) of the composites are
measured with a Perkin Elmer DSC-7 thermal mechanical analyzer (TMA).

23
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

130 105

125 104

120 103
Tg (° C)

Tg ( ° C )
102
115

101
110

100
105

99
100
0 5 10 15 20 25 30 35 98
SGM content (vol%) 0 5 10 15 20 25

BN content (vol %)

Fig.16. Variation of thermal conductivity with SGM volume

fraction of SGM-Epoxy composites Fig.17. Variation of Tg with BN volume fraction for BN-
Epoxy composites

115
SGM 0 vol%
113 SGM 10 vol %
SGM 20 vol %
111
109
Tg ( ° C)

107
105
103
101
99
0 10 20
BN content (vol%)
Fig.18. Variation of Tg with BN volume fraction for hybrid composites varying SGM content

62
60
58
CTE ( ppm / ° C )

56
54
52
50
48
0 10 20 30 35
SGM content (vol%)
Fig.19. Variation of Tg with BN volume fraction for hybrid composites varying SGM content

24
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

Fig.16. shows the glass transition temperature (Tg) of the epoxy–SGM composite. It is observed that the T g of neat
epoxy is about 1040C and it gradually increases to 124 °C with increasing SGM content at 35vol%.The effect of the
filler in the polymer composites on the glass transition behaviour of the polymer matrix has been studied for different
polymer–filler composites. Adding micrometer-sized filler usually increases the T g of composites, which results from
the interaction between filler and polymer restricting the mobility of the polymer chain. In this study, the increase in T g
of the composites can be attributed to strong interaction between the SGMs and the epoxy-matrix, which can be
confirmed SEM analysis.
On the contrary, the addition of boron nitride reduces the T g by about 50C (Fig.17, Fig.18). Similar observation has
been reported by Couderc et. al for polymer filled with BN and this drop in T g has been attributed to possible
impediment of reticulation by boron nitride particles, acting as obstacles to the reaction. This results from a more
mobile molecular structure produced by less complete reticulation. But the surprising fact is that the composites
containing both boron nitride and solid glass microspheres exhibit improved T g confirming a more reticulated and so
less mobile structure. This means that the presence of only solid glass microspheres does not modify the reticulation
reaction, whereas in presence of boron nitride the reaction is more effective. Glass Transition Temperature (Tg) is a
computerized method of measuring the molecular change. It is the measure of the point where the cured resin goes
through a change in its molecular structure. At this point mechanical properties decrease at an increasing rate and the
coefficient of thermal expansion increases.

6.4.Coefficient of Thermal Expansion Of SGM and BN Filled Epoxy Composites

Coefficient of thermal expansion (CTE) is a measure of the rate of change of the thermal expansion of a dielectric. A
material with high coefficient of thermal expansion will induce stress on the barrel copper of the holes on PCBs.
Therefore a low CTE is preferable. Fig.19 shows the plot of CTE as a function of SGM content.
The CTE of the neat epoxy is about 60.83 ppm/°C, and it gradually decreases to 49.12 ppm/°C with an increase in
the filler content. A maximum decrease of about 11.71 ppm/°C in CTE is obtained at 35vol % of SGM-filled composite.
The low CTE of SGM and the constraint of deformation of the epoxy-matrix due to the interaction of SGM and epoxy
are responsible for the reduced CTE of the composites. For a copper cladded laminate used in printed circuit board
(PCB), the key to decreasing thermal stress is to decrease the CTE of the epoxy substrate due to the difference in CTE
between the epoxy substrate and Cu. In this study, the CTE of the epoxy–SGM composite reaches a value of about
49.12 ppm/°C when the SGM content is 35 vol.%, which is lower than that of the common commercial product
manufactured by Polyclad Laminates Inc. (e.g., 50 ppm/°C for PCL-FR-226 laminate/PCL-FRP-226 prepreg and PCL-
FR-240 laminate/PCL-FRP-240 prepreg).

6.5. Dielectric Characteristics Of Composites

An investigation into the effect of boron nitride content on the thermal conductivity as well as dielectric performance of
epoxy filled with solid glass micro-spheres is done. It is also recognized that the dielectric constant decreases with
increasing BN content in the glass-epoxy matrix at all frequencies. At the same time there is an apparent improvement
in the thermal conductivity of the hybrid composites in the presence of BN. It is therefore expected that these kinds of
hybrid composites with high thermal conduction and electrically insulating properties can surmount the obstacles for
fabrication of a better material for electronic packaging and printed circuit board substrate applications.

6.5.1.Dielectric Constant
Dielectric constant is a measure of the degree to which an electromagnetic wave slows down as it travels through the
insulating material. In general, a lower dielectric constant is preferred as it has lower impedance on signal propagation.
Dissipation factor is a measure of the loss-rate of the electromagnetic field travelling through a dielectric. Similar to
dielectric constant, a lower dissipation factor is preferred as lower energy is absorbed or lost. The complex permittivity
can be considered a fundamental parameter for the macroscopic description of a dielectric exposed to the alternating
fields. This concept can be used to describe media that are homogenous to the extent that scattering effects are
insignificant radio waves which penetrate into these materials. The dipole moments of simple discreet inclusions, like
spheres and ellipsoids, can be expressed in closed form and the mixing formula for a two phase mixture can be derived.
The effective permittivity is a quality attributable to heterogeneous media. To be able to introduce this concept, the
sizes of the inclusions have to be considerably smaller than the wavelength of the operating electromagnetic wave field.

6.5.2.Experimental Test results

Experimentally obtained dielectric properties data, dielectric constant of epoxy – E-glass composite material in 100 Hz
to 100 kHz frequency range and for 0 to 35 % volume fraction are presented in the following figures (Figs. 20 to 27).
Calculated dielectric values, based on developed analytical composite dielectric models are graphically presented using
solid line, whereas measured dielectric data are presented using solid line with error bars, for frequency range and
composite volume fractions of interest.

25
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

6
Rule-of-Mixture Model Litchtenecker Logarithmic Law of Mixing
5.5
Maxwell-Garnet Model Jayasundere-Smith Equation
5 Experimental Result

4.5

Dielectric constant
4

3.5

2.5

1.5

1
0 5 10 15 20 25 30 35

SGM content (vol%)

Fig.20. Comparision of Dk values of different empirical models and measured value with a varying range of SGM content

Fig.21 shows the room temperature dependence of dielectric constant with increase in BN content. It is found that
with increase in frequency, the value of D k decreases. The dielectric constant of a polymeric material depends on
interfacial, dipole, electronic and atomic polarization. The atomic and electronic polarizations are instantaneous
polarization components, the effect of which is seen of at higher frequencies. The interfacial polarization arises due to
heterogeneity, which is highest at lower frequency. Hence, the higher value of D k at low frequency is due to presence of
interfacial polarization. It is because at lower frequencies complete orientation of the molecule is possible but at
medium frequencies there is only a little time for orientation of the molecules. Orientation of molecules is not possible
at very high frequencies. It is similar to dipole polarization. Hence, the change of D k at lower frequency region is higher
than that of very high frequency region.
The electrical properties of the solid glass microspheres filled epoxy composites are also studied by measuring the
room temperature dielectric constant at different frequencies. The dielectric constant at 1 MHz frequency for cured
specimens with different SGM content is presented in Fig. 20.Fig. 21.shows the dielectric behaviour of BN-Epoxy
composites at different frequencies. The dielectric constant shows a decreasing and then increasing fashion when there
is increase in filler content for BN-Epoxy composite whereas it decreases with increase in SGM content. Park et al.
have reported similar observation in case of polymers filled with hollow glass microspheres. These results have been
attributed to the addition of glass microspheres into the epoxy resins and thus causing the charge migration, resulting in
decrease in the dielectric constant in the composites.

4
1kHz 4.5

10kHz 1 kHz 10 kHz

3,5 100kHz 100 kHz 1000 kHz

4
SGM Content: 5 vol%
Dielectric constant

Dielectric constant

3 %%
3.5

2,5
3

2
2.5
0 5 10 15 20 25 30 35
1,5 BN content (vol %)
5 10 15 20 25
BN content (vol%)
Fig.22. Dk values of hybrid composites at different
Fig.21. Dk values at different frequencies with a varying frequencies value with increase in BN content
range of BN content

26
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

The decrease in the dielectric constant with an increase in frequency is ascribed to the reason that the interfacial
dipoles have less time to orient themselves in the direction of the applied filed.

7. Conclusions
 Successful fabrication of epoxy based composites filled with solid glass micro-spheres by hand-lay-up
technique is possible.
 Determination of Keff of epoxy composite filled with SGM of different size and content is possible using FEM.
 The values of Keff estimated by the proposed mathematical model are in good agreement with the measured
values of Keff for SGM filled epoxy composites with different volume fractions of filler.
 Incorporation of SGMs results in reduction of thermal conductivity of epoxy resin and thereby improves its
thermal insulation capability. With addition of 35 vol% of SGM (100 micron size), the thermal conductivity
decreases by about 66 % as compared to neat epoxy resin.
 For 200 micron size with the addition of 35 vol% of SGM, the thermal conductivity decreases by about 63 %
whereas for 300 micron size the thermal conductivity decreases by about 55%.
 For same volume fraction of SGM in the composite, the improvement in composite insulation capability is
found to be more for smaller SGMs.
 With light weight and improved insulation capability SGM filled epoxy composite can be used for applications
such as electronic packages, insulation board, food container, thermo flasks, building materials, space flight
and aviation industry etc.

Acknowledgements The support by the Department of Mechanical Engineering, NIT, Rourkela, Institute of Minerals and Materials
Technology (IMMT), BBSR and CIPET (LARPM) BBSR is gratefully acknowledged.

References
[1] K.Jung-il, P.H.Kang, Y.C.Nho, J.Appl.Polym.Sci.92(2004) 394-401
[2] S.Nikkeshi, M.Kudo,T.Masuko,J.Appl.Polym.Sci.69(1998) 2593-2598.
[3] Ismail.H.Tavman, J.Powder Technol. 91(1997) 63-67.
[4] K.Zhu, S.Schmander, J.Comput. Mater. Sci. 28(2003) 743-748.
[5] M.Rusu, N.Sofian, D.Rusu, J.Polym.Test.20(2001) 409-417
[6] S.Cheng, R.Vachon.A technique for predicting the thermal conductivity of suspensions: emulsions and porous materials, Int. J
Heat & Mass Tr.13(1970) 537-542.
[7] J. Maxwell, Electricity and Magnetism, Oxford, Clarendon, 1873.
[8] Lewis T, Nielsen, Thermal conductivity of particulate- filled polymers. J. Appl. Polym. Sci.29 (1973) (3819-3825).
[9] Y.Agari, A. Ueda, M.Tanaka, S.Nagai, J.App.Polym.Sci40(5-6)(1990) 929-941.
[10] X.G.Liang, W.Qu, Effective thermal conductivity of gas-solid composite materials and the temperature difference effect at high
temperature,Int. J. Heat Mass Transfer 42 (1999) 1885-1890.
[11] S.A.Suvorov,V.N.Fishchev,S.N.Kapustina,SV.Kopylova,M.G.Nekhlopochina,Thermal Conductivity of composites with a filler
of hollow spherical corundum granules,Refractories 29(1989)689-693
[12] D.Hansen, C.Ho, J.Polym.Sci.Part A 3(2) (1965) 659-670
[13] S.Peng, R.Landel, J.Appl.Polym.Sci.19 (1) (1975) 49-68.
[14] C.L.Choy, K.Young, J.Polym.18(8) (1977) 769-776.
[15] Ismail H.Tavman, “Thermal Anisotropy of polymers as a Function of their Molecular orientation”, Experimental HeatTransfer,
Fluid Mechanics, and Thermodynamics,Elsevier,1991, pp. 1562-1568.
[16] D.Kumlutas et.al., “Thermal Conductivity of particle filled polyethylene composite materials”, J. Compos.Sci.Technol.63(1)
(2003) 113-117
[17] R.Nayak, A.Satapathy, T.Dora, “A computational and experimental investigation on thermal conductivity of particle reinforced
epoxy composites,” Comp Mater Sci., 2010,Vol 48,pp. 576-581.
[18] Y. Agari and T. Uno, Estimation on Thermal Conductivities of Filled Polymers, J. of App. Poly. Sci., 32(7) (1986) 5705–5712.
[19] Y. Agari, A. Ueda and S. Nagai, Thermal Conductivity of Polyethylene Filled with Disoriented Short-cut Carbon Fibers, J. of
Applied Polymer Composite Science, 43(6) (1991) 1117–1124.
[20] R.C.Progelhof, J.L.Throne, R.R. Ruetsch “Methods of Predicting the Thermal Conductivity of composite Systems”, Polym Eng
Sci,1976,Vol.16(9),pp.615-625.
[21] Procter & J.Solc, IEEE Trans. Comp. Hybrids Manuf. Technol. 14(4) (1991) 708-713
[22] Y. Nagai, G.C. Lai, Thermal conductivity of epoxy resin filled with particulate aluminum nitride powder, J. Ceram Soc Jpn,
105(3) (1997) 197–200.
[23] Yu SZ, Hing P, Hu X (2002) Compos Part A 33:289
[24] A. Griesinger, W. Hurler and M. Pietralla, A Photothermal Method with Step Heating for Measuring the Thermal Diffusivity of
Anisotropic Solids, Int. J. of Heat and Mass Transfer, 40(13) (1997) 3049–3058.
[25] I. Tavman, Thermal and Mechanical Properties of Copper Powder Filled Poly (ethylene) Composites, J. Powder Tech., 91(1)
(1997) 63–67.

27
 Polymer science: research advances, practical applications and educational aspects (A. Méndez-Vilas; A. Solano, Eds.)
_______________________________________________________________________________________________

[26] N.M. Sofian, M. Rusu, R. Neagu and E. Neagu, Metal Powder-filled Polyethylene Composites. V. Thermal Properties, J.
Thermoplastic Composite Materials, 14(1) (2001) 20–33.
[27] Y.P. Mamunya, V.V. Davydenko, P. Pissis and E.V. Lebedev, Electrical and Thermal Conductivity of Polymers Filled with
Metal Powders, J. European Polymer, 38(9) (2002) 1887–1897.
[28] B. Weidenfeller, M. Ho¨ fer and F.R. Schilling, Thermal Conductivity, Thermal Diffusivity, and Specific Heat Capacity of
Particle Filled Polypropylene, J. Composites Part A: Applied Science and Manufacturing, 35(4) (2004) 423–429.
[29] H.S. Tekce, D. Kumlutas, and I.H. Tavman, Determination of the Thermal Properties of Polyamide-6 (Nylon-6)/Copper
Composite by Hot Disk Method, In: Proceedings of the 10th Denizli Material Symposium, Denizli, (2004) 296–304.
[30] D. Kumlutas, and I.H. Tavman, A Numerical and Experimental Study on Thermal Conductivity of Particle Filled Polymer
Composites, J. of Thermoplastic Composite Materials, 19 (2006) 441.
[31] P.Amar, Md.Abdulla, S.Alok, B.Sandhyarani, K.S.Bhabani., Mat & Des. 31(2) (2010) 837-849
[32] R. F Hill and P. H. Supancic Thermal Conductivity of Platelet-Filled Polymer Composites. Journal of the American Ceramic
Society, 4 (2004) 851-857.
[33] Yunsheng Xu,D.D.L.Chung, Cathleen Mroz Composites: PartA 32(2001) 1749-1757.
[34] Shu-Hui Xie,Bao-Ku Zhu,Ju-Biao Li,Xiu-Zhen Wei,Zhi-Kang Xu Polymer Testing 23 (2004) 797-801
[35] Eun-Sung Lee and Sang Mock J.A.Ceram.Soc, 91[4] 1169-1174
[36] J.Z. Liang, F.H. Li, Measurement of thermal conductivity of hollow glass bead filled polypropylene composites, Polym. Test.
25(4) (2006) 527-531
[37] J.Z. Liang, F.H. Li, Simulation of heat transfer in hollow glass-bead- filled polypropylene composites by finite element method,
Polym.Test.26(3)(2007) 419-424.
[38] J.Z. Liang, F.H. Li, “Heat transfer in polymer composites filled with inorganic hollow micro-spheres:A theoretical
model”,Polym. Test. 26 (2007) 1025-1030
[39] K.C. Yung ,et al. , “Preparation and properties of hollow glass microsphere- filled epoxy- matrix composites” Comp. Sci.
Technol.(69) (2009) 260 -264.

28