1. Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.4, No.3, 2014
1
Influence of Low Temperature on Heat Transfer in Epoxide
NanoComposites
Dr.Abbas Alwi Sakhir Abed
University of Al-Qadissiya-College of Engineering-IRAQ
abbasabed59@gmail.com
Abstract
The process of heat transfer in polymer systems and composites due to the variety of factors determining their
properties is extremely complex and has been poorly studied practically. Heat transfer in unfilled polymers
largely depends on polymer structure in general, on the degree of order of structural elements within one.
Infusion of nanofillers into a polymer system leads to significant changes of mechanical, thermal, electrical, and
other properties. At infusion of fillers, the heterogeneity of a polymer system at a macroscopic level also
increases.
Keywords:-nanofillers . polymer . heterogeneity . intramolecular . nanocomposites .
Introduction
Nowadays a lot of materials about research and compilation of mechanical and thermal properties of
conventional filled polymers have been accumulated. For polymers filled by nanoparticles, there is only
fragmentary information, often contradictory, mainly investigating mechanical properties of materials.
Mechanisms of the interaction between polymer and filler for crystallizing amorphous polymers have a different
character. Filled to crystalline polymer solid particles can be placed in the middle of supramolecular structures
and serve as a basis for growth of conglomerates or to be displaced in the regions between structural elements.
At filling of amorphous polymers, with the surface of filler both individual macromolecules and supramolecular
structures interact.
It has been found that the main feature in the influential mechanism of fillers on properties of polymers is the
nature of the processes at supramolecular levels. The order of supramolecular structure of polymer covers more
outlying from filler surface areas than it leads from thermodynamic concepts. Therefore, as a result of the
interaction between polymer and filler, rigid structural elements form, permeating the entire volume of polymer
and having a significant effect on its mechanical properties. In nanocomposites the entire polymer matrix
transforms to the state of an interfacial layer with different from polymer in volume properties. These processes
are illustrated by figures (figure 1), made by scanning electron microscope Carl Zeiss SupraTM
55 for epoxide
nanocomposites.
Thus, in composites heat transfer processes are caused both by conductivity of individual components and
significantly by a boundary layer filler – fixant, its structure and physical –mechanical changes depending on
external influences are still poorly researched.
Therefore, experimental data is of particular value for the creation and operation of devices from composite
materials.
From a wide variety of polymers, thermoset fixants have been chosen – namely epoxide – as the most widely
used in various structural and electroinsulating materials in aeronautics, automotive and electronics industries.
Thermosetting epoxides are widely used due to the fact that they combine high strength properties, excellent
dimensional stability, heat resistance and resistance to various external environmental influences. However,
these highly crosslinked thermoset polymers have inherent fragility, which increases with increasing of
crosslinking density. So they are filled by different mineral fillers to eliminate this disadvantage.
Nanocomposites based on polymeric fixants are promising materials from the viewpoint of reinforcement of
polymers and imparting them improved properties by infusion of nanoparticles of different materials into a
polymer matrix.
With decreasing of particle size, specific surface area of filler, length of the interfaces and boundary layer
proportion increase. Boundary layer quantity at the particle size of 0.5 – 5 microns increases up to 50%, and its
influence on properties of composite material increases. At the particle size of 10 – 100 nm, composite material
can be called nanocomposite. In such materials even with the proportion of filler on the order of parts of a
percent, practically the entire polymer transforms into the boundary layer state [1].
Many of promising directions in material sciences, nanotechnology, nanoelectronics and applied chemistry are
associated recently with nanotubes and other similar nanostructures which can be called by a general term –
carbon skeletal structures. Due to different properties of nanoparticles, such as extended surface, dimensional
and quantum effects of nanoepoxide composites exhibit improved mechanical and thermal properties.
But, despite the fact that carbon nanotubes can be considered as heat superconductors (the highest values of
thermal conductivity of carbon nanotubes are more than 3000 W/(m·K) [2]), it does not mean that they can have

2. Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.4, No.3, 2014
2
the same high level of conductivity, being integrated into other materials. As oscillation frequency of atoms
forming carbon nanotubes is much higher than of surrounding material atoms, it results large thermal resistance
at the interface between nanotubes and other material components. Therefore thermal conductivity of polymers
with nanotubes as filler is many times lower than it was expected. However, there is still the belief that it is
possible to increase thermal conductivity of polymers in 2 – 3 times by a small amount of added nanofiller (up to
1%). The problem is in the difficulty of dispergation, dispersion of nanoparticles in a polymer matrix without
adhesion, without formation of conglomerates, because special properties of carbon nanotubes are better
observed when they are in the isolated state. In this case, a polymeric matrix turns into the state of a boundary
layer with properties which are different from volume polymer samples.
Physical properties of carbon nanotubes and polymer composites have been widely studied in recent years.
Numerous works are devoted to mathematic modeling processes, occurring in carbon nanotubes and to the
calculation of transfer coefficients, while there is limitation of experimental works. Many articles are devoted to
the study of mechanical and electrical properties of carbon nanotubes. As for thermal physical properties such as
specific heat capacity, heat and thermal conductivity, such works are not enough. Moreover, the published
results show a large scatter of characteristics.
The authors [3] by three different methods were measured thermal conductivity and specific heat capacity of
single – walled carbon nanotubes in a temperature range from – 20o
C up to + 70o
C. Clean single – walled carbon
nanotubes modified by thionyl chloride were investigated as well as graphite powder (for comparison). The
calculation of heat conductivity coefficient based on the measured values of thermal conductivity coefficient and
specific heat capacity showed that clean nanotubes have heat conductivity 0.16 W/(m·K), modified – 0.3
W/(m·K), that is much less than the values shown by other authors [4-6].
Only a few papers are devoted to heat conductivity of epoxide nanocomposites. The authors [7] studied heat
conductivity of composite based on epoxide, where randomly oriented single – walled carbon nanotubes are
uniformly dispersed on the micrometer level. Even a small quantity of nanotubes introduced into an epoxide
matrix significantly improved its thermal properties. So, addition of 1 mass% of uncleaned nanotubes led to the
increase of heat condactivity by 70 % at 40 K and by 125% at room temperature. The influence of the same
additives of carbon fibers having much larger diameter (~ 200 nm) is three times lower. The measured values
were lower than it leads from the estimated values [8].
The authors [9] investigated rheological, mechanical, electric and thermal properties of epoxide nanocomposites
containing carbon nanomaterials, depending on filler concentration. In particular, for heat conductivity it was
shown that the maximum value 0.25 W/(m·K) was observed at 1.5 mass% of carbon nanotubes. However, the
value of heat conductivity of pure epoxide 0.12 W/(m·K) is surprising due to it is usually from 0.18 up to 0.21
W/(m·K).
Experimental part
1. Materials and preparation of samples of nanocomposites
As the studied objects, polymeric composite materials based on epoxide, filled by nanostructured powders were
chosen. As nanofillers carbon nanotubes and aerosil were selected with sufficiently small concentrations from
0.05 mass% up to 3.0 mass%, due to definitely small concentrations of nanofillers produce the most significant
effect at filling of polymer composites [11-12]. With the purpose of checking the influence of low concentrations
of microfillers on heat conductivity of composite materials, epoxide composites with concentration of talc and
activated carbon 0.1 mass% were studied.
Сarbon nanopowders used as fillers in polymer nanocomposites were prepared at LIHM NAS in plasma of a
high – voltage discharge at atmospheric pressure [13]. As raw material, a mixture of methane and air was used.
These nanomaterials include besides carbon nanotubes amorphous carbon additives, graphite nanoparticles and
metal catalyst nanoparticles. So, their cleaning from additives to get nanotubes with maximum purity is a very
important step in the production of carbon nanotubes. All nanofillers were introduced as powders which had
been cleaned before (heating in the air, treatment with hydrochloric or nitric acid with following heat treatment).
However, density, and therefore porosity, varied. So, multi – walled carbon nanotubes had density 260 kg/m3
,
and nanopowders containing carbon nanotubes – 100 kg/m3
. The differences in density provide evidence of
different porosity as well as significant differences in specific surface of the obtained nanomaterials.
The main problem during creating of nanocomposites is the process of uniformly dispersing of nanofillers in a
polymer matrix due to ultradisperse particles are apt to aggregate. A review of existing methods of epoxide
nanocomposites manufacturing showed that there are some general principles of forming of such materials.
Firstly, epoxide at room temperature has high viscosity. Therefore, usually to reduce its viscosity it is heated up
to 50o
C or is diluted by various diluents, for example, by acetone. Secondly, hardener is not injected into epoxide
immediately, but firstly nanoparticles are added and mechanical mixing of components is produced. Besides
mechanical mixing, different external power influences are used such as ultrasound, magnetic and electric fields
for more uniform dispersion of nanoparticles in a polymer matrix. Thirdly, hardening is carried out in stages:

3. Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.4, No.3, 2014
3
first at room temperature and then at elevated temperatures.
As a base epoxide was used, which is a part of universal epoxide glue. Acetone was a diluent. Mechanical and
ultrasonic mixing procedures of the components were used. Hardener was polyethylenepolyamine (PEPA).
Temperatures and hardening times were varied. However, the best results were obtained with the following
scheme of thermal treatment: 16 hours at room temperature, then 3 hours at 60°C and 3 hours at 130o
C.
Density of all the investigated nanocomposites was about the same (1110 kg/m3
), due to the concentration of
filler is too small to have a significant effect on density.
2. Heat conductivity measurement
Heat conductivity measurements were carried out using device IT – λ – 400 by the monotonous heating method
[14] in a range of temperatures from – 150o
C up to 150o
C with an error 5%. Firstly a sample was cooled until
liquid nitrogen temperature and then was heated adiabatically with velocity 6 K/min up to 150o
C.
Results and discussion
On figure 2 temperature dependences of heat conductivity coefficient of powders of carbon materials are shown,
used as filler in epoxide nanocomposites.
In spite of nanosize character of the powders, heat conductivity of the powders on macroscopic scale in the state
of free filling is many orders less than calculated and expected heat conductivity of a separate nanotube. It is
connected with the fact that nanotubes in the powder locate chaotically and their package is quite permeable. The
samples are obtained with high porosity that decreases heat conductivity coefficient. Heat conductivity
coefficient of the powder of single – walled carbon nanotubes is less than of the powder of multi – walled carbon
nanotubes due to its density is less.
On figure 3 temperature differences of heat conductivity coefficient of epoxide nanocomposites with different
fillers at the same concentration are shown.
The most significant increase of heat conductivity coefficient is observed during filling of epoxide by
nanomaterials, containing multi – walled and single – walled tubes. Usage of activated coal as filler looses the
structure that leads to decrease of heat conductivity of composite in comparison with clean epoxide at elevated
temperatures. Small quantities of aerosil also increase heat conductivity of epoxide, although not with the same
rate as carbon nanomaterials. Since the average size of aerosil particles is from 7 – 25 nm and aerosil has
developed surface (specific surface is from 115 up to 380 m2
/g) [15] so it can be considered as nanofiller. Up to
the temperature of 50o
C heat conductivity coefficient can be approximated by linear dependence. The
temperature at the bend zone point characterizes the transition to the glass formation zone.
To compare temperature dependences of all samples, their values can be normalized using their heat
conductivity coefficient at room temperature (25o
C). Such normalized temperature dependencies are shown on
figure 4.
All the materials have almost linear dependence, although slope of these dependences is different. It is especially
significant at low temperatures. So it is possible to estimate heat conductivity of any nanocomposite at any
temperature in a range from – 150o
C up to 25o
C, assessing its heat conductivity just at room temperature. The
temperature dependence of nanocomposites can be described by quasilinear dependence:
and temperature dependence of epoxide and composite with filling by activated coal:
where λ(T) – heat conductivity coefficient at the temperature of to
C, λ(T) – heat conductivity coefficient at the
temperature of 25o
C, T – temperature in Celsius degrees.
The research of the influence of the concentration of carbon filler on heat conductivity of epoxide
nanocomposite was carried out using the material of epoxide + multi – walled carbon nanotubes at the
concentrations of 0.0%, 0.05 mass%, 0.1 mass%, 0.3 mass%, 0.5 mass%, 0.7 mass%, 1.0 mass%, 3.0 mass%. On
figure 5 the concentration dependence of heat conductivity coefficient of this material at room temperature is
shown.
It is observed that even such small concentration of multi – walled carbon nanotubes as 0.05 mass% increases
heat conductivity coefficient. The maximum increase is observed not at a certain concentration, but in a range of
concentrations from 0.1 mass% up to 1.0 mass%. The decrease of heat conductivity coefficient at the increase of
concentration is due to the technological features.
Low temperatures influence on the structure of composites. Due to the differences in heat expansion coefficients

4. Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.4, No.3, 2014
4
of filler and fixant during cooling, the process of cracking with formation of micro and macroporosity can occur.
As a result heat conductivity of the material decreases. On figure 6 temperature characteristics of heat
conductivity coefficient of nanocomposite «epoxide + micro – walled carbon nanotubes» with different
concentration of filler are shown.
The most significant increase of heat conductivity in the whole temperature range is referred to nanocomposite
with 0.1 mass% of multi – walled carbon nanotubes. It seems as this concentration is optimal for this material.
For more complete realization of properties of different nanotubes it is necessary to disperse them carefully in a
polymer matrix with the absence of agglomerates from the strands of nanotubes due to exactly the interfaces of
nanotubes contribute to thermal resistance of heat transmission in the material. During the increase of filling rate,
the increase of contact surface of nanotubes and, correspondingly, of thermal resistance, occurs.
Therefore it is necessary to gain different dispergation methods, including ultrasound, and also to use electric
and magnetic fields for lining of nanotubes and their uniform dispersion in a polymer matrix. In recent times
close attention is paid to functioning, to activating of nanotubes’ surface to separate them better and to increasing
of adhesion, due to majority of researchers refer less values of properties of polymer nanocomposites with
carbon filler in comparison with the predicted theoretical values to inert nature of graphene structure and, as a
result, to diminished interface interaction [16, 8].
FIG. 2. Temperature dependencies of heat conductivity coefficient of carbon nanofillers.
( b )

5. Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.4, No.3, 2014
5
FIG. 3. Temperature dependencies of heat conductivity coefficient of epoxide composites with the concentration
of filler 0.1 mass%.
FIG. 4. Heat conductivity coefficient normalized to the value at 25o
C.
FIG. 5. Dependence of heat conductivity coefficient from the concentration of multi – walled carbon nanotubes.
1 – heat conductivity coefficient, W/(m·K); c – the concentration of filler, mass%.

6. Journal of Energy Technologies and Policy www.iiste.org
ISSN 2224-3232 (Paper) ISSN 2225-0573 (Online)
Vol.4, No.3, 2014
6
FIG. 6. Temperature dependences of heat conductivity coefficient of epoxide composite with different
concentrations of multi – walled carbon nanotubes.
References
1. Vaia R.A., Wagner H.D. Framework for nanocomposites. Materials today, 2004, November, pp. 32 – 37.
2.Iijima S. Helical microtubules of graphitic carbon. Nature. 1991. Vol. 354, pp. 56 – 58.
3.Vretenar V., Skakalova V., Kubicar L., Roth S. Thermophysical properties of single wall carbon nanotubes.
EuroMat – 2003, Lausanne, Switzerland, Sept. 1 – 5, 2003. U2 – 1560.
4.Hone J., Whitney M., Piskoti C., Zettl A. Thermal conductivity of single – walled carbon nanotubes. Physical
review B. 1999. Vol. 59, No. 4, R2514 – 2516.
5. Hone J., Llaguno M.C., Nemes N.M., Johnson A.T., Fischer J.E., Walters D.A., Casavant M.J., Schmidt J.,
Smalley R.E. Electrical and thermal transport properties of magnetically aligned single wall carbon nanotube
films. Applied Physics Letters. 2000. Vol. 77, No. 5, pp. 666 – 668.
6.Fischer J.E., Zhow W., Vavro J., Llaguno M.C., Guthy C., Haggenmueller R., Casavant M.J., Walters D.E. and
Smalley R.E. Magnetically aligned single wall carbon nanotube films: Preferred orientation and anisotropic
transport properties. Journal of Applied Physics. 2003. Vol. 93, No. 4, pp. 2157
7. Biercuk M.J., Llaguno M.C., Radosavljevic, Hyun K.J., Johnson A.T. and Fischer J.E. Carbon nanotube
composites for thermal management. Appl. Phys. Letters. 2002. Vol. 80, pp. 2767 – 2769.
8.Nan C. – W., Shi Z., Lin Y. A simple model for thermal conductivity of carbon nanotube – based composites.
Chem. Phys. Letters. 2003. Vol. 375, p. 666.
9.Song Y.S., Youn J.R., Properties of epoxy nanocomposites filled with carbon nanomaterials. 12th
POLYCHAR
World Forum on Advanced Materials. Portugal, Jan. 6 – 9, 2004.
10.Schadler L.S., Giannaris S.C., Ajayan P.M. Single – walled carbon nanotube – polymer composites: strength
and weakness. Advanced Matter. 2000. Vol. 12, No. 10, p. 750 – 753.
11.Allaoui A., Bai S., Cheng H.M., Bai J.B. Mechanical and electrical properties of a MWNT/epoxy composite.
Compos. Sci. Tech. 2002. Vol. 62, p. 1993 – 1998.
12.Lau K.T., Shi S.Q., Cheng H.M. Micro – mechanical properties and morphological observation on fracture
surface of carbon nanotube composites pre – treated at different temperatures. Comp. Sci. Tech. 2003, Vol. 63, p.
1161 – 1164.
13.Zhdanok S.A., Buyakov I.F., Chernukho A.P., Krauklis A.V., Solntsev A.P., Shashkov A.E. Carbon
nanotubes synthesis under nonequilibrium conditions. Nanotechnology in Physics, Chemistry and Biotechnology,
St – Petersburg. 2002.
14.Platunov E.S. Thermophysical analysis under monotonous mode. M.: Energiya, 1972.
15.Maligin A.A. Surface chemistry and nanotechnology: interaction and prospects. Sorosov scientific journal.
2004. Vol. 8, No. 1, pp. 32 – 38.
16.Weisenberger M.C., Grulke E.A., Jasques D., Rantell T., Andrews R. Enhanced mechanical properties of
polyacrylonitrile. J. Nanosci. Nanotech. 2003. Vol. 3, pp. 535 – 539.