Experimental investigation of a composite phase change material: Thermal-energy storage and release

Abstract

In recent times, composites made out of polymers and paraffin waxes were thought to be good thermal energy storage materials, in which the heat is stored as latent heat of fusion in the paraffin wax. In this study, phase change composite material with spherical shape calibrated based paraffin wax (RT27) was produced. The properties of the prepared composite phase change material have been characterized. The objective of this article was to study the energy storage and the energy recovery by using a phase change composite material. An experimental set-up consisting of fluxmetric measurement has been constructed to provide the thermal performance of the composite. In addition, a differential scanning calorimetry analysis was carried out. The experimental apparatus allows providing heat storage capacities and “apparent” thermal conductivities of the composite at the solid and liquid states, and also a measurement of the latent heat of fusion. The proposed test provides temperature and heat flux measurements at the material borders. The amount of energy exchanged during the variation of the thermodynamic state samples could be calculated when the boundary temperatures vary. In this article, one shows how it can allow the study of complex composite material with PCM. In particular, heat flux measurements make it possible to highlight very specific behaviors of these products and are thus a very interesting experimental source of data which comes to complete the traditional measurement methods like calorimetric device (differential scanning calorimetry).

Full text

http://jcm.sagepub.com/
Journal of Composite Materials
http://jcm.sagepub.com/content/early/2012/12/06/0021998312468185
The online version of this article can be found at:
DOI: 10.1177/0021998312468185
published online 12 December 2012Journal of Composite Materials
Abdelwaheb Trigui, Mustapha Karkri, Chokri Boudaya, Yves Candau, Laurent Ibos and Magali Fois
Experimental investigation of a composite phase change material: Thermal-energy storage and release
Published by:
http://www.sagepublications.com
On behalf of:
American Society for Composites
can be found at:Journal of Composite MaterialsAdditional services and information for
http://jcm.sagepub.com/cgi/alertsEmail Alerts:
http://jcm.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
What is This?
- Dec 12, 2012OnlineFirst Version of Record >>
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

JOURNAL OF
COMPOSITE
MATERIALS
Article
Experimental investigation of a composite
phase change material: Thermal-energy
storage and release
Abdelwaheb Trigui
1
, Mustapha Karkri
1
, Chokri Boudaya
2
,
Yves Candau
1
, Laurent Ibos
1
and Magali Fois
1
Abstract
In recent times, composites made out of polymers and paraffin waxes were thought to be good thermal energy storage
materials, in which the heat is stored as latent heat of fusion in the paraffin wax. In this study, phase change composite
material with spherical shape calibrated based paraffin wax (RT27) was produced. The properties of the prepared
composite phase change material have been characterized. The objective of this article was to study the energy storage
and the energy recovery by using a phase change composite material. An experimental set-up consisting of fluxmetric
measurement has been constructed to provide the thermal performance of the composite. In addition, a differential
scanning calorimetry analysis was carried out. The experimental apparatus allows providing heat storage capacities and
‘‘apparent’’ thermal conductivities of the composite at the solid and liquid states, and also a measurement of the latent
heat of fusion. The proposed test provides temperature and heat flux measurements at the material borders. The
amount of energy exchanged during the variation of the thermodynamic state samples could be calculated when the
boundary temperatures vary. In this article, one shows how it can allow the study of complex composite material with
PCM. In particular, heat flux measurements make it possible to highlight very specific behaviors of these products and are
thus a very interesting experimental source of data which comes to complete the traditional measurement methods like
calorimetric device (differential scanning calorimetry).
Keywords
Epoxy resin/spherical paraffin wax, differential scanning calorimetry analysis, thermophysical properties, fluxmetric
measurement, thermal energy storage and release
Introduction
Latent heat thermal energy storage (LHTES) systems
utilize the enthalpy of melting/solidification of phase-
change materials. Phase change materials (PCM) that
are used as storage media in latent thermal energy stor-
age have applications in diverse areas, such as building
heating/cooling systems, solar energy collectors, power
and industrial waste heat recovery.
1
Among several
thermal energy storage techniques, latent thermal
energy storage is a particularly attractive technique
that provides a high storage capacity per unit mass
(and also per unit volume generally) and has the prop-
erty of storing heat as the latent heat of fusion at a
constant temperature, i.e. the phase change
temperature.
The appeal of phase change materials lies in the fact
that their use may bridge the time lag between
availability and use of energy, thus making the use of
solar energy more economical, and by storing electrical
energy as thermal energy in periods outside the peaks,
the load of electrical power stations may be made more
uniform.
The work done with phase change materials during
the past decade consisted mainly of fundamental
research and pilot plant experiments. Several models
have been developed for analyzing the thermal charac-
teristics of LHTES systems.
2
Pilot plant experiments
have been focused on solar heat pump systems
3
and
1
Universite
´Paris-Est, CERTES, Cre
´teil Cedex, France
2
Universite
´de Sfax, De
´partement de Physique, Route Sokra, Sfax, Tunisie
Corresponding author:
Abdelwaheb Trigui, Universite
´Paris-Est, CERTES, 61 avenue du Ge
´ne
´ral
de Gaulle, 94010 Cre
´teil Cedex, France.
Email: abdelwaheb.trigui@etu.u-pec.fr
Journal of Composite Materials
0(0) 1–14
!The Author(s) 2012
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0021998312468185
jcm.sagepub.com
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

concrete wall and floor systems for building energy
storage.
4,5
The repeated melting and crystallization during the
heat storage process may result in a decrease of the heat
storage capacity. Several accelerated thermal cycle tests
of latent heat-storage materials have been suggested to
predict the thermal characteristics in case of longer
use.
6
The well-known PCMs are as follows:
CaCl
2
.6H
2
O, Na
2
SO
4
. 10H
2
O, CH
3
COONa.3H
2
O,
paraffins, stearic acid.
3
The properties of normal paraf-
fin wax are very suitable for latent heat storage.
Paraffin waxes are composed mainly of alkanes,
approximately 75%. Alkanes and paraffin waxes are
both organic compounds. They have a large heat of
fusion per unit weight, non-corrosive, non-toxic, chem-
ically inert and stable below 500C. On melting, they
have a low volume change and a low vapor pressure.
Mixing different molecular weight paraffin waxes
together can easily vary melting temperature. Since
they are commercially available, the cost is reasonable.
Prime candidates for passive applications are tetrade-
cane, hexadecane, octadecane and eicosane. However,
paraffin waxes suffer from a low thermal conductivity
(0.24 W.m
1
.K
1
) and liquid leakage when they
undergo the solid–liquid phase change.
13,15
A composite
material is a material that is composed of several differ-
ent materials, usually to improve a property of a mater-
ial or to combine properties of different materials. In the
case of PCM, a PCM composite material is produced to
improve at least one of the PCM properties or to
improve the heat storage capacity of another material.
There are different ways to form a composite: by embed-
ding PCM in a matrix of another material, or by embed-
ding another material into the PCM. To maintain
material properties, the order of magnitude of the struc-
tures in the composite should be microscopic, or at least
below the scale of mm. Otherwise, the properties of the
composite will depend on the sample size and the com-
posite therefore cannot be called a material anymore.
When working with new PCMs, reliable methods for
PCM materials testing and long-term performance
evaluation are very important. Presently, there is a
lack of standard methods for a long-term stability test-
ing of large PCM specimens. In order to provide further
knowledge about the steady periodic heat transfer
during cyclic processes of melting and freezing, in the
present paper the problem has been analyzed experimen-
tally. The following paper describes the first store built
with the composite material (resin epoxy/ spherical par-
affin wax) developed to improve different properties of
PCM and open a wide field of applications to latent heat
storage systems.
Experimental set-up
Sample preparation
The first step is to produce a calibrated spherical shape
based on a commercial paraffin wax (RT27) using a
specific mold, see Figure 1(a). This mold is in the
form of two aluminum plates of 160 * 120 mm
2
and
15 mm thickness with spherical shells manufactured
within each plate.
The diameter of the sphere is 11 mm. Paraffin is
injected in liquid state under constant pressure and
the center of each shell is using a specific syringe.
After filling, the cooling and shrinkage compensation
(10%) of the paraffin are performed. The second step is
to place these spherical capsules of paraffin in a mold
(200 200 mm2) with a controlled distribution keeping
a distance of 2 mm between each spherical paraffin wax,
see Figure 1(b). The resin is then injected in the center
of one of the walls of the mold at room temperature
and under atmospheric pressure to impregnate these
spherical paraffin wax. Finally, the sample is pressur-
ized in the mold and the injection of an excess of mater-
ial will be introduced to compensate the shrinkage,
see Figure 1(c). The sample to be characterized
Figure 1. (a) Mold for the manufacture of paraffin spherical shape, (b) Positions of spherical paraffin shape into a mold, (c) epoxy
resin þparaffin spherical shape.
2Journal of Composite Materials 0(0)
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

(Figure 1(c)) has a mass of 667 g; 200200 16 mm
3
of
dimensions and 1042.18 kg/m
3
of density. The amount
of paraffin (RT 27) introduced in the matrix is 98 g.
Experimental apparatus
To determine the specific heat and latent heat of mater-
ials, a number of methods such as differential thermal
analysis (DTA) or differential scanning calorimetry
(DSC) are commonly used.
7
These methods are very
well developed, but they use only small micro-samples
of the test material. When evaluating the performance
of these materials for commercial scale use a test
method that can accommodate a larger sample may
provide additional information on the long-term stabil-
ity of the product. It is believed that a very small
sample, taken out of the large testing specimen and
out of testing system, might not reflect properly the
bulk material characteristics. During long-term per-
formance testing, when undergoing huge number for
freezing/melting cycles, large PCM specimens often
are subject to settling or stratification and are typically
not perfectly homogeneous. For large size samples, test-
ing using a noninvasive method for heat of fusion and
specific heat determination is necessary. It was found
that the determination of the overall thermophysical
properties of PCMs over several cycles (solidification
and fusion) requires the design of a genuine experimen-
tal device (Figure 2). The proposed test bench for the
parallelepiped-shape of composite provides tempera-
ture and heat flux measurements at the material bor-
ders. The sample is located between two horizontal
exchanger aluminium plates. Thermo-regulated baths,
supplying the plates, allow a fine regulation of the
injected huile H10 temperature with a precision of
about 0.1C. Heat flux sensors and thermocouples
(type T) are placed on each side of the composite.
The whole thing is maintained in place by use of a
slightly tightened pneumatic jack. The thickness of
flux-meters is about 0.2 mm and their sensitivity is
about 202 V=W=m2for a sensor having an active sur-
face area of 400 cm
2
.
14
The various sensors are con-
nected to a LabviewÕprogram adapted to measure
temperature fluctuations and heat flux exchanged
during fusion and solidification processes.
Experimental data are recorded with regular and
adjustable time steps (6 s). The lateral side faces are
insulated by 11 mm thickness of polyethylene expanded
foam (PE) which reduces multidimensional heat trans-
fer to a 1D problem. In this work, the temperature was
varied between þ15C and þ50C.
Results and discussion
Thermal analysis of paraffin wax
Generally, PCMs are characterized by calorimetric
methods, like differential scanning calorimetry (DSC),
Figure 2. Experimental set-up.
Trigui et al. 3
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

applied to very small amounts of product. This tech-
nique is derived from the differential thermal analysis
(DTA). Over the years, DSC became the reference
method for thermal characterization of phase change
material.
8
This equipment provides heat fluxes by
Joule effect, which corresponds to the necessary
power difference to maintain at the same temperature
a ‘‘cell’’ containing material to be characterized and a
reference cell which is generally empty. The paraffin
waxes (RT27) have a melting point provided by the
manufacturer RubbithermÕequal to þ27C.
9
In this
study, the thermophysical properties of the PCM were
measured by a DSC instrument (Perkin–Elmer Pyris
calorimeter). All DSC measurements were repeated
for each sample at a different heating rate to ensure
the reproducibility. Finally, a 10C/min heating rate
was employed and considered sufficient for the experi-
ments in the case of fusion. Samples were conditioned
in aluminium pans.
10
The DSC thermal analyses were performed in the
temperature range of 10C/þ50C. The DSC curves
for the heating and cooling of the paraffin are shown in
Figure 3. It is seen that there exist two peaks, the larger
peak is due to solid-liquid phase change and the smaller
peak is due to solid–solid phase transition. The melting
temperature of the PCM corresponds to the onset
Figure 3. Stored heat (Cp) as a function of temperature measured by differential scanning calorimetry (scanning rate, 10C/min).
4Journal of Composite Materials 0(0)
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

temperature (25.53C) obtained by the conventional
procedure as the intersecting point between the tangent
line at the point of the maximum slope of the DSC peak
and the non-linear base line taking into account the
change of heat capacity between the solid and the
liquid phases.
11
Otherwise, the temperature 25.69C
corresponds to the intersecting point between the tan-
gent line at the point of the maximum slope of the DSC
peak and the sigmoı
¨de. The latent heat of the paraffin is
obtained as the total area under the peaks of the solid–
solid and solid–liquid transitions by numerical integra-
tion. The latent heat of melting corresponds to
Lm¼155.960 J/g obtained by the conventional proced-
ure as the area under the peak by numerical integration.
Otherwise, it is 140.581 J/g corresponds to numerical
integration of the area under the peak having as basis
the sigmoı
¨de. The difference of 15 J/g between the two
procedures shows the impact of this choice. In the case
of cooling, the latent heats of the solid–solid and solid–
liquid transition are 22.594 J/g and 162.165 J/g, respect-
ively, obtained by the conventional procedure. As
shown in Figure 3, the onset temperature of crystalliza-
tion is 26.2C.Table 1 shows the thermophysical prop-
erties for the present investigation.
To clarify the effects on the heat transfer perform-
ance, an experimental device was developed, based on
the measurement of temperatures and heat fluxes
exchanged between the two lateral sides of the PCM
samples, providing the total heat stored during the
phase change process.
Apparent thermal conductivity
To determine solid and liquid thermal apparent conduc-
tivities of the composite (resin epoxy/ spherical paraffin
wax), a temperature difference was imposed between the
two lateral sides of the composite until observing a zero
heat flux (equilibrium state). The apparent thermal con-
ductivities are calculated by the following expression
12
:
s,l ¼e:s,l
2:Ts,l
ð1Þ
where eis the dimension of the material and s,l is the
sum of the measured heat fluxes.
In Figures (4 and 5), the temperature variations on
each side of the composite (external surface temperature
T2internal surface temperature T1) are represented.
Figure 4 shows the results of the solid-phase state. The
problem to be considered was that of a composite that
initially held a solid paraffin at T1¼T2¼Tinit ¼15C,
lower than the melting temperature (Tm). This tempera-
ture was maintained until thermal equilibrium. The
whole material was entirely solid. At time t>0, the com-
posite was heated by modifying the temperature on a
single face only (T1¼Tend ¼20C), such that
Tend <Tm(fusion process).
A similar experiment was carried out for the liquid
phase (Figure 5); the material is subjected to a tempera-
ture difference. The temperature levels on both sides
(40C and 50C) are higher than the melting tempera-
ture. Several tests were carried out on the material to
check the reproducibility of the measurement.
The results were found to be satisfactory and pro-
vided values of apparent thermal conductivities are
listed in Table 2.
Experimental procedure
The different tests performed on this sample correspond
to different boundary conditions following:
– A temperature ramp on both sides of the sample
imposed by two exchangers plates.
– A temperature variation sinusoidal on the lower
exchanger (which simulates daily changes in out-
door temperature), constant temperature on the
other (to simulate the indoor temperature).
– A temperature ramp of the lower exchanger, con-
stant temperature on the other.
Temperature ramp on both sides of the sample
a. Energy storage. Initially, the material is isother-
mal. Then it is heated by modifying the temperature
set point of the thermo-regulated bath. The material
will thus evolve from Tinit toTend. Between these two
permanent steady states, the material stores energy.
The flux-meters make it possible to measure the heat
fluxes exchanged at the borders of the sample. The total
Table 1. Thermophysical properties of paraffin wax (RT27)
Properties Values measured
Heating state
Solid–solid phase change temperature Tpm 5.08C
Melting point, Tm25.53C–25,69C
a
Latent heat of solid-solid change, Lpm 23.692 J/g
Latent heat of melting, Lm155.960 J/g–140,581 J/g
a
Cooling state
Solid-solid phase change temperature, Tpc 3.95C
Crystallization point, Tc26.2C
Latent heat of solid-solid change, Lpc 22.594 J/g
Latent heat of crystallization, Lc162.165 J/g 154.848 J/g
a
Specific heat of solid, Cp,s (15C) 3.25 0.35 J/g.C
Specific heat of liquid, Cp,l (40C) 2.23 0.005 J/g.C
a
Value measured by sigmoı¨de.
Trigui et al. 5
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

amount of energy per mass stored can then be obtained
from the following expression:
Q¼1
:eZtend
tinit
:dt ¼Cp:ðTend TinitÞðkJ=kgÞð2Þ
represents the cumulated heat rate entering the
sample; Cp: apparent specific heat capacity composite
(kJ/kg.C). This quantity can also be expressed by:
Q¼Qsens þLm¼ðCps:TsþCpl:TlÞþLmkJ=kgðÞ
ð3Þ
Cpsand Cplare the average solid state and the liquid
state specific heat of the material, Tsand Tlare the
temperature variations for the material in solid phase
and in liquid phase and Lmis the latent heat of melting.
The experience consists first in imposing on the sample
a superficial temperature of 15C on each one of its
faces, until obtaining a thermal steady state corres-
ponding to an isothermal material. The heat fluxes is
then zero at the initial time t¼0. It is also confirmed
that the thermal losses are negligible at the isolated side
faces. At a particular moment (tinit), a sharp huile H10
temperature variation is imposed in the bath. This
induces a thermal evolution of the system (storage)
until another state of equilibrium is obtained.
Figures 6 and 7 present the variation of the heat storage
capacity of the sample in solid phase and liquid phase,
respectively.
For the liquid and solid phases, the temperatures
evolve in an asymptotic way to the set point. It is
also noticed that the heat fluxes evolves very quickly
at the beginning of recording and then to a zero value
which corresponds in a new state of balance obtained at
the end of the test. The thermal evolution from 15Cto
50C (Figure 8) allowed us to follow the complete melt-
ing process, from the solid state to the liquid state,
during which a great quantity of heat has been stored
by the material. The selected temperatures are suffi-
ciently far away from the zone of melting point to con-
sider that indeed the material is strictly in one or the
other state. With regard to the variation between 15C
and 50C during which there is phase change, a new
thermal balance is reached in a little more than 1 hour.
Figure 8 shows that the heat stored is much more
important than sensible heat transfer when a phase
change occurs. This confirms the interest of latent
heat storage. The thermal behavior of the sample is
very particular during the fusion process. One clearly
observes on this figure the different phases of the mater-
ial evolution between a solid state at start and the liquid
state at the end. For each test, the integration of the
heat flux over time determines the amount of heat
stored during the process (stationary and isothermal
states (’¼0). Several tests were carried out to ensure
reproducibility of the experiments. To demonstrate the
importance of thermal storage by latent heat, evolution
0
20
40
60
80
100
(Τ2)
(Τ1)
(φ2)
Time (h)
Heat Flux (W/m2)
(φ1)
5
10
15
20
25
Temperature (°C)
0,0 0,1 0,2 0,3 0,4 0,5 0,6
Figure 4. Measurements with thermal variation one side of the solid sample (15Cto20
C).
6Journal of Composite Materials 0(0)
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9
0
50
100
150
200
(Τ2)
(Τ1)
(φ2)
Time (h)
Heat Flux (W/m 2)
(φ1)
0
10
20
30
40
50
Temperature (°C)
Figure 5. Measurements with thermal variation one side of the solid sample (40Cto50
C).
Table 2. Apparent thermal conductivities of composites (resin epoxy/ spherical paraffin wax)
Thermal conductivity (W.m
1
.K
1
) Test 1 Test 2 Test 3 Average Errors (%)
Solid state 0.184 0.193 0.203 0.193 6
Liquid state 0.209 0.211 0.213 0.211 6
–80
–60
–40
–20
0
20
40
60
80
(Τ1)
(Τ2)
(φ2)
Time (h)
Heat Flux (W/m
2)
(φ1)
0
5
10
15
20
25
Temperature (°C)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
Figure 6. Heat flux and temperatures evolution of the solid phase (15Cto20
C).
Trigui et al. 7
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

of heat flux and temperatures are presented in Figures 9
and 10. Figure 9 corresponds to a variation of tempera-
ture of 30C (initial state: 20C – final state: 50C) and
Figure 10 to a variation of temperature of 20C (initial
state: 20C – final state: 40C).
The results were found to be satisfactory and pro-
vided values of thermophysical properties of compos-
ites are listed in Tables 3 and 4.
b. Energy release. In the cooling case, where the tem-
perature evolves from 50Cto15
C, solidification of
the material occurs. For our parallelepipedic sample,
one observes (Figure 11) symmetrical evolution on the
measured heat fluxes on the two faces of the sample.
Initially, it can be observed a normal evolution of mea-
sured heat flux corresponding to the cooling of the
liquid phase. At the end of this phase (t¼24 min),
when the temperatures of surfaces are in the vicinity
of 27C, the heat flux evolution is reversed. From this
critical moment, the cooling of the sample continues,
the material is solidified slowly and cools until it
reaches the prescribed 15C: after more than one
hour and half, the sample reaches an equilibrium state.
Heat restitution is a very long process. At the
beginning of solidification a layer of solid PCM
starts to be formed on the plane surface in contact
with the exchanging plate, a layer which ‘‘isolates’’
the liquid phase from the cooling source.
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0
–150
–100
–50
0
50
100
150 (Τ2)
(Τ1)
(φ2)
Time (h)
Heat Flux (W/m2)
(φ1)
0
5
10
15
20
25
30
35
40
45
50
Temperature (°C)
Figure 7. Heat flux and temperatures evolution of the liquid phase (40Cto50
C).
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0
–600
–400
–200
0
200
400
600
(Τ
2)
(Τ1)
(φ2)
Time (h)
Heat Flux (W/m2)
(φ1)
0
10
20
30
40
50
60
Temperature (°C)
Figure 8. Heat flux and temperatures evolution from solid to liquid (15Cto50
C).
8Journal of Composite Materials 0(0)
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

Solidification continues slowly because of the low
thermal conductivity of the solid PCM (is estimated
at ¼0.193 W.m
1
.K
1
). The temperature of the sample
surface reaches 15C only at the end of the test.
However, if the amount of released heat is calculated,
it is seen this value is 67.42 kJ/kg. It is almost the
same for storage heat.
Thermal cycle with temperature sinusoidal variation. The
simulation chosen here imposed a sinusoidal variation
of the outdoor temperature which represents the daily
variation of the atmosphere temperature (equation 4).
In this case, the temperature of the upper exchanger is
maintained constant (Texch 2 ¼23C).
Texch 1 ¼23 þ8 sin 2t


:¼86400s ð4Þ
In Figure 12, the temperature variations on each side
of the composite (external surface temperature T2,
–500
–400
–300
–200
–100
0
100
200
300
400
500
600
(Τ2)
(Τ1)
(φ2)
Time (h)
Heat Flux (W/m2)
(φ1)
0
10
20
30
40
50
60
Temperature (°C)
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 1,1
Figure 9. Heat flux and temperatures evolution from solid to liquid state (20Cto50
).
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2
–300
–200
–100
0
100
200
300
(Τ2)
(Τ1)
(φ2)
Time (h)
Heat Flux (W/m 2)
(φ1)
0
10
20
30
40
50
Temperature (°C)
Figure 10. Heat flux and temperatures evolution from solid to liquid state (20Cto40
C).
Trigui et al. 9
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

internal surface temperature T1) and the temperature of
exchangers variation (Texch 1,Texch 2 ) are represented. It
can be observed two inflection points, the first at t&5h
34 min and the second at t&17 h 34 min. These two
inflection points correspond to a composite wall tem-
perature of about 26C, reflecting the change of phase
liquid/solid paraffin state. The different temperature
curves show that they are in phase with each other.
Figure 13 shows a significant effect of PCM on the
flux densities.
There is a slight lag between the flux density of each
composite side and a narrow landing on the rising part
of curves, characteristic of energy storage. The same
figure shows the difference of the storage and release
heat flux densities. Indeed, the periodic integration of
the heat flux evolution (12) is equal to zero, see
equation 5.
Z
0
12
ðÞdt ¼0being the period ð5Þ
However, if the amount of the stored and released
heat is calculated during a period (areas above or under
the time axis), it is seen that their values are 26.13 kJ/kg
more than the storage capacity of the material
(24.96 kJ/kg). The estimated error on energy measure-
ments is about 2%.
Thermal cycle with temperature ramp variation. In the first
test, a ramp was performed by varying the temperature of
the lower exchanger on the cycle 15C, 50C and back to
15C. The temperature on the higher exchanger is 15C.
In Figure 14, it is observed that the time of temperature
rise of the underside of the sample T1is relatively shorter
than the cooling time at the end of the cycle. Indeed, the
heating time (against 24 min) and the cooling time (1 h
30 min) would explain a slower releasing heat when
changing from a liquid / solid. The difference in fluxes
densities (12) is shown in Figure 15.
The phases of storage and release are clearly high-
lighted. The energy stored and released from heat is
calculated in the same way as before. The energies
stored by composite are about 43.50 kJ/kg and
45.84 kJ/kg of releasing. The energy measurements
stored and released from heat are given with a precision
about 2.3% of the full scale.
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 1,1 1,2 1,3 1,4 1,5 1,6
–350
–300
–250
–200
–150
–100
–50
0
50
100
150
200
250
300
(Τ1)
(Τ2)
(φ1)
Time (h)
Heat Flux (W/m 2)
(φ2)
0
5
10
15
20
25
30
35
40
45
50
Temperature (°C)
Figure 11. Heat flux and temperatures evolutions (50Cto15
C).
Table 4. Quantity of heat stored and heat capacity
Thermophysical properties 15–50C 20–40C 20–50C
Total stored heat (kJ /kg) 68 44.23 58.01
Heat Capacity (kJ /kg.C) 1.95 2.21 1.93
Table 3. Thermophysical properties of composites (resin
epoxy / spherical paraffin wax)
Thermophysical properties Values Errors (%)
Total stored heat (15C–50C) (kJ /kg) 68 3.8
Sensible heat (solid state) (kJ /kg) 7.59 3.8
Sensible heat (liquid state) (kJ /kg) 14.10 3.8
Heat capacity (solid state) (kJ /kg.C) 1.52 0.5
Heat capacity (liquid state) (kJ /kg.C) 1.41 0.8
Latent heat (kJ /kg) 46.31 2.3
10 Journal of Composite Materials 0(0)
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

The second test was elaborated with the following
boundary conditions:
.Temperature ramp of the lower exchanger; by vary-
ing Texch 1 on the cycle 15C, 40C and back to 15C.
.Temperature constant set at 15C on the upper
exchanger (Texch 2 ¼15C).
By the use of the same analysis done before, the
integral of the curve difference fluxes densities
(12), the energy stored during this ramp is about
27.57 kJ/kg for 1 h 30 min (Figure 16). The energy
released is 29.61 kJ/kg. These energies are almost
identical. The estimated error on energy measurements
is 2%.
0 5 10 15 20 25
–80
–60
–40
–20
0
20
40
60
80
100
Heat Flux (W/m2)
Time (h)
φ2φ1 (φ1– φ2)
Figure 13. Heat flux variation of sinusoidal cycle.
10
12
14
16
18
20
22
24
26
28
30
32
34
Temperature (°C)
Time (h)
T2 T1 Texch1 Texch2
0 5 10 15 20 25
t ≈ 5h34mn t ≈ 17h34mn
Figure 12. Temperature variation of sinusoidal cycle.
Trigui et al. 11
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

Conclusion
Thermal energy storage is a very important topic in
present times. In this article were presented the first
results of settling an experimental method for thermal
material characterization in case of phase change. The
determination of the thermophysical properties could
be carried out for composite (resin epoxy/ spherical
paraffin wax). Commercial paraffin wax RT 27 was
used with a melting point close to normal comfort tem-
peratures. The obtained results were very satisfactory.
In this article, one shows how it can allow to study the
energy storage–release of complex composite with
PCM. In particular, heat flux measurements make it
0
10
20
30
40
50
60
Temperature (°C)
Time (h)
T2 T1 Texch1 Texch2
0,0 0,5 1,0 1,5 2,0 2,5 3,0
Figure 14. Temperature evolution versus ramp variation.
0,0 0,5 1,0 1,5 2,0 2,5 3,0
–400
–200
0
200
400
600
Heat Flux (W/m2)
Time (h)
φ
2φ
1 (φ1
– φ
2
)
Figure 15. Heat flux evolution versus ramp variation.
12 Journal of Composite Materials 0(0)
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

possible to highlight very specific behaviors of these
products and are thus a very interesting experimental
source of data which comes to complete the traditional
measurement methods like calorimetric device (DSC).
Further comparison needs to be performed between
reliable experimental data and numerical modeling,
by using a phase change material. Work in progress
will be in particular of great importance to the influence
of matrix like (bio-concrete, thermoplastic...) and geo-
metrical properties in order to maximize the stored
energy for a given set of boundary conditions.
Funding
This research received no specific grant from any funding
agency in the public, commercial, or not-for-profit sectors.
Conflict of Interest
None declared.
Nomenclature
e Composite width

,
t Times,s
Density of composite, kg:m3
T Temperature, C
CpSpecific heat capacity, kJ=kg:C
Thermal conductivity, W:m1:K1
Q Energy per mass stored, J=gor kJ=kg
Density of heat flux, W=m2
Subscripts
1,2 Lower and higher face of the composite
init Initial thermal steady state
end Final thermal steady state
exch Exchanger
p Solid-solid change
sens Sensible
s Solid state
l Liquid state
m Melting
cCrystallization
References
1. Zalba B, Marin JM, Cabeza LF, et al. Review on thermal
energy storage with phase change: Material, heat transfer
analysis and applications. Appl Thermal Eng 2003; 23:
251–283.
2. Bele
´n Zalba, Jose
´Ma Marina, F Cabeza Luisa, et al.
Review on thermal energy storage with phase change:
materials, heat transfer analysis and applications. Appl
Thermal Eng 2003; 23: 251–283.
3. Ayhan T and Kaygusuz K. Experimental and theoretical
investigation of combined solar heat pump system for resi-
dential heating in Turkey. Energy Edu Sci Technol 1998;
2(1): 43–61.
4. Hawes D, Feldman D and Banu D. Latent heat storage in
building materials. Energy Build 1993; 20(1): 77–86.
5. Khudhair AM and Farid MM. A review on energy con-
servation in building applications with thermal storage by
latent heat using phase change materials. Energy Convers
Mgmt 2004; 45(2): 263–275.
0.0 0.5 1.0 1.5 2.0 2.5
–400
–300
–200
–100
0
100
400
(φ1–
φ2) (W/m2)
Time (h)
heat stored
heat discharged
Figure 16. The difference of heat flux densities (12).
Trigui et al. 13
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from

6. Maha Ahmad, Andre
´Bontemps, He
´bert Salle
´e, et al.
Experimental investigation and computer simulation of
thermal behaviour of wallboards containing a phase
change material. Energy Build 2006; 38: 357–366.
7. Li J, Liu Z and Ma C. An experimental study on the
stability and reliability of the thermal properties of
barium hydroxide octahydrate as a phase change mate-
rial. The Key Laboratory of Enhanced Heat Transfer and
Energy Conservation of the Education Ministry, China,
1985.
8. Jamil A, Kousksou T, Zeraouli Y, et al. Simulation of the
thermal transfer during an eutectic melting of a binary
solution. Thermochimica Acta 2006; 44: 30–34.
9. Rubitherm. http://www.rubitherm.de/english/index.htm.
10. Yinping Z and Yi J. A simple method, the T-history
method, of determining the heat of fusion, specific heat
and thermal conductivity of phase change materials.
Measur Sci Technol 1999; 10(3): 201–205.
11. Hatakeyama T and Quinn FX. Thermal analysis, 2nd ed.
Chichester: Wiley, 1999.
12. Joulin A, Younsi Z, Zalewski L, et al. Experimental and
numerical investigation of a phase change material:
Thermal-energy storage and release. Appl Energy 2011;
88: 2454–2462.
13. He B and Setterwall F. Technical grade paraffin waxes as
phase change materials for cool thermal storage and cool
storage systems capital cost estimation. Energy
Conversion and Management 2002; 43: 1709–1723.
14. Lassue S, Guths S, Leclercq D and Duthoit B. Natural
convection by heat flux measurement and anemometry
using thermoelectric effects. Third World Conference
on Experimental Heat Transfer, Fluid Mechanics, and
Thermodynamics, Honolulu, Hawaii, USA, 31 October–
5 November 1993, Vol. 1, pp. 831–838.
15. Karkri M, Mousseau P and Yvon J. Thermal state of an
incompressible pseudo-plastic fluid and Nusselt number
at the interface fluid–die wall. Int J Therm Sci 2008;
47(10): 1284–1293.
14 Journal of Composite Materials 0(0)
at UPEC on November 16, 2013jcm.sagepub.comDownloaded from