Applied Energy 135 (2014) 339–349

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Preparation of nitrogen-doped graphene/palmitic acid shape stabilized

composite phase change material with remarkable thermal properties
for thermal energy storage
Mohammad Mehrali a,⇑, Sara Tahan Latibari a, Mehdi Mehrali a, Teuku Meurah Indra Mahlia b,
Emad Sadeghinezhad a, Hendrik Simon Cornelis Metselaar a,⇑
a
Advanced Material Research Center, Department of Mechanical Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia
b
Department of Mechanical Engineering, Universiti Tenaga Nasional, 43009 Kajang, Selangor, Malaysia

h i g h l i g h t s

 Nitrogen-doped graphene (NDG)/palmitic acid (PA) composite shape stabilized PCMs were prepared.
 The loading of PA in the composite PCMs with good form-stability could attain 97 wt%.
 The phase change enthalpy of the composite PCMs could attain 199.48 J/g.
 The thermal conductivity of the PCM containing 3 wt% NDG is 3.5 times higher than that of PA.

a r t i c l e i n f o a b s t r a c t

Article history: Palmitic acid (PA) is one of the main phase change materials (PCMs) for medium temperature thermal
Received 22 March 2014 energy storage systems. In order to stabilize the shape and enhance the thermal conductivity of PA,
Received in revised form 8 August 2014 the effects of adding nitrogen-doped graphene (NDG) as a carbon nanofiller were examined experimen-
Accepted 26 August 2014
tally. NDG was dispersed in liquid PA at various mass fractions (1–5 wt%) using high power ultrasonica-
Available online 16 September 2014
tion. The dropping point test shows that there was clearly no liquid leakage through the phase change
process at the operating temperature range of the composite PCMs. The thermal stability and thermal
Keywords:
properties of composite PCM were investigated with a thermogravimetric analyzer (TGA) and differential
Graphene
Phase change materials
scanning calorimeter (DSC), respectively. The thermal conductivity of the PA/NDG composite was deter-
Thermal energy storage mined by the laser flash method. The thermal conductivity at 35 °C increased by more than 500% for the
Thermal conductivity highest loading of NDG (5 wt%). The electrical conductivity of composite PCMs was increased signifi-
Latent heat cantly by using NDG. The thermal cycling test proved that the PA/NDG composites PCMs had good ther-
mal reliability and chemical durability after 1000 cycles of melting and freezing. The thermal effusivity of
the PA/NDG composite PCMs was larger than that of pure PA, which is advantageous for latent heat ther-
mal energy storage (LHTES).
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction
The energy crisis and global warming have become serious con-
cerns, thus, outstanding attempts have been made for the efficient
utilization of alternative energy, like solar energy. Nevertheless,
the fluctuation of solar radiation makes latent heat thermal energy
storage (LHTES) indispensable within the solar thermal energy
applications. Phase change materials (PCMs) are recognized to
⇑ Corresponding authors. Tel.: +60 3 79674451; fax: +60 3 79675317.
E-mail addresses: mohamad.mehrali@siswa.um.edu.my, mehrali.gary@gmail.
com (M. Mehrali), h.metselaar@um.edu.my (H.S.C. Metselaar).
become critical for LHTES since they can store and release consid-
erable amounts of latent heat during their phase transition for effi-
cient utilization of thermal energy [1]. PCMs are usually
categorized into organic and inorganic materials based on their
components. Organic PCMs have obtained extensive attention
because of their higher latent heat density, suitable phase-transi-
tion temperature and stable physical and chemical properties [2–
5]. The pure organic PCMs possess some shortcomings that hinder
their application in practice, including low thermal conductivity,
high volume change and liquid seepage during phase transition.
There are several strategies to eliminate these particular difficul-
ties such as encapsulation and shape stabilization of PCMs [6,7].

http://dx.doi.org/10.1016/j.apenergy.2014.08.100
0306-2619/Ó 2014 Elsevier Ltd. All rights reserved.
340 M. Mehrali et al. / Applied Energy 135 (2014) 339–349
Nowadays, form-stable PCM composites with good thermal con-
ductive supporting materials which could maintain the solid shape
even if the temperature is more than the melting point of PCMs
have become the main focus of scientific studies [8–12]. This
method is among the most effective ways to resolve the leakage
and low thermal conductivity problems of PCMs. There have been
a lots of studies on the preparation and characterization of the
shape stabilized PCMs using porous materials, including expanded
perlite [13], graphite [14], expanded graphite [15–17], carbon
nanofibers [18,19], carbon nanotubes [20], graphite nanoplatelets
[19], graphene [10,21,22], graphene oxide [23,12,24], graphene
nanoplatelets [8,25]. The accessible research outcomes reveal that
the shape stabilized composite PCMs based on porous materials
possess a great application possibility within the thermal storage
field.
Graphene is a two-dimensional all-sp2-hybridized low density
carbon with fascinating thermal, electrical and mechanical proper-
ties and has generated a rapidly growing research interest. Graph-
ene sheets have high thermal conductivity, high specific surface
area, and good electrical properties compared to other carbon
based materials [26]. The potential of using graphene-based prod-
ucts for PCMs has drawn much interest recently because it can be
simply obtained by basic chemical processing of graphite [27]. The
thermal conductivity of composite PCMs depends not merely on
the innate thermal conductivity of the additives, but it is also
strongly related to the compatibility between additives and pure
PCMs [28]. In our earlier works, graphene nanoplatelets (GNPs)
and graphene oxide (GO) were used as supporting materials for
shape stabilization to enhance the thermal conductivity of the
composite PCMs. Since graphene properties are directly linked to
the nano structure of the graphene sheets, quite a few attempts
have been carried out to improve the electrical and thermal prop-
erties of graphene flakes by modifying their structures, as well as
using novel synthesis methods, chemical functionalization and
chemical doping with external atoms [29]. On the whole, chemical
doping with sulfur(S), boron (B) or nitrogen (N) is regarded as a
highly effective technique to modify the inherent properties of
graphene-based materials. Particularly N-doping performs an
important role in regulating the electronic, thermal and chemical
properties of carbon materials because of the similar atomic size
and the availability of five valence electrons to form strong valence
bonds with carbon atoms [30]. There is a fundamental interest to
examine how nitrogen doping influences the structure of graphene
and to see whether the chemical properties can be modified
accordingly [31]. A few works have been carried out to investigate
the structural difference between graphene and N-doped graphene
(NDG) but there have been no reports related to the effect of N-
doping on the thermal conductivity of graphene. The NDG exhibits
an excellent adsorption capacity for various classes of organic liq-
uids including organic solvents. In particular, NDG can adsorb
amounts of the liquids up to 200–600 times its own weight, which
is much higher than other typical carbonaceous sorbents such as
graphene foam.
In this work, we report the preparation of PA/NDG composite
PCM as a novel form-stable composite PCM by using NDG as
supporting material. The great compatibility between NDG and
PA leads to a large improvement in the thermal conductivity at
low filler loading ratio along with a small reduction in latent
heat. Furthermore, capillary forces and surface tensions were
improved due to the larger specific surface area (793 m2/g) of
NDG that is effective for shape-stabilization over the phase
change process of PCMs mainly because it helps to hold the mol-
ten PCM. This high thermal conductivity as well as the minimal
loss in the phase transition enthalpy makes the PA/NDG compos-
ite a promising candidate for thermal energy storage
applications.
2. Experimental description
2.1. Preparation of NDG
Graphene oxide was synthesized from natural graphite powder
(99.99%, <45 lm, Sigma Aldrich) employing a simplified Hummers’
method [32]. NDG was synthesized by a hydrothermal process
with GO as raw material. In a typical method for preparing NDG,
the pH of a dispersion of 50 mg GO in 100 ml H2O after 1 h ultra-
sonication was adjusted to 11 by using ammonia. The dispersion
was treated hydrothermally in a Teflon-lined autoclave at a tem-
perature of 160 °C for 12 h. A black precipitate was washed with
deionized water and dried at 50 °C.
2.2. Preparation of PA/NDG composite PCMs
The preparation procedure was started by dissolving 2 g of pal-
mitic acid (PA) in 50 ml toluene at 80 °C, after which NDG with dif-
ferent mass percentages (1–5 wt%) was added to the solution. The
solution was sonicated for 30 min at power of 150 W to break
down the NDG aggregates and obtain a homogeneous dispersion
of the nano particles. The mixture was then left inside a fume hood
at 130 °C to evaporate the toluene and then poured into a mold.
Finally, the PCM samples have been dried inside a vacuum oven
overnight at 120 °C to eliminate PA which had not been absorbed
by NDGs.
The PA/NDG samples that were prepared by this process are
seems to be naturally isotropic by taking into consideration that
the NDG nano sheets had no preferential direction and haphaz-
ardly oriented. The prepared composite PCMs with 1–5 wt% NDG
were named S1–S5, respectively.
2.3. Analysis methods
Field emission scanning electron microscopy (FESEM- CARL
ZEISS- AURIGA 60) was used to visually characterize the morphol-
ogy of the NDG and composite PCMs. The Brunauer–Emmett–
Teller method (BET-Autosorb-iQ2) was used to measure specific
surface area and pore distribution of the NDG sample. Transmis-
sion electron microscopy (TEM) measurements were conducted
on a CARL ZEISS-LIBRA120 microscope. An X-Ray photoemission
spectrometer (XPS-PHI 5400 ESCA) with an Al Ka (hm = 1486.69 eV)
X-ray source was used to identify the elements’ states in the NDG.
X-ray diffraction (XRD) patterns were measured on an Empyrean
PANALYTICAL diffractometer. The melting and freezing tempera-
tures, latent heats and specific heat capacity of composite PCMs
were obtained by differential scanning calorimeter (METTLER
TOLEDO 820C-Error ± 0.25–1 °C) at a heating rate of 5 °C/min.
The weight loss and thermal stability of PCMs are obtained by ther-
mogravimetric analysis (METTLER TOLEDO SDTA 851-Error ± 5 lg)
at a heating rate of 10 °C/min and a temperature of 50–500 °C in
purified nitrogen atmosphere. The temperature distribution photos
were taken by an infrared camera (FLIR i5) with a 0.1 °C thermal
sensitivity. The shape stability of composite PCMs was analyzed
by using Mettler Toledo FP83HT Dropping Point Cell at a heating
rate of 2 °C/min.
2.4. Thermal cycling test
The thermal reliability of PA/NDG composite PCMs was investi-
gated after 1000 thermal cycles by using an accelerated thermal
cycling system (Fig. 1). The heat storage/retrieval performance of
the samples was investigated by recording temperatures during
the whole cycling process.
 M. Mehrali et al. / Applied Energy 135 (2014) 339–349 341

Fig. 1. Accelerated thermal cycler [12].

2.5. Thermal conductivity and electrical resistivity measurement
The laser flash technique (Netzsch LFA 447 NanoFlash) was
used to measure the thermal diffusivity of prepared samples at
35 °C. The thermal conductivity may also be derived from the ther-
mal diffusivity when the specific heat and the sample bulk density
are known. The electrical resistance of PA/NDG composites was
measured with a LCR HiTESTER 3532 meter. The resistance of the
sample was measured in the frequency range from 0.6 Hz to
100 kHz. Electrical resistivity was calculated from the resistance
values at room temperature.
3. Results and discussion
3.1. Characterization of NDG
Fig. 2(a) and (b) shows the FESEM images of the sample, dis-
playing a uniform structure like crumpled silk veil waves with a
porous architecture while NDG nanosheets are randomly stacked
together. The FESEM images of the NDG sample reveal that the
two dimensional graphene structures with high specific surface
area and volume ratio are well retained. Three major peaks located
at 284.4, 399.5 and 532 eV was determined from the XPS spectrum
of the NDG sample (Fig. 2(c)) that was linked to C1s, N1s and O1s,
respectively, showing that nitrogen was incorporated within the
graphene structure [33]. The nitrogen was doped by atomic per-
centage of 2.6% within the graphene network by pyridinic and pyr-
rolic nitrogen species, respectively. They refer to the nitrogen
atoms which might be positioned in a p conjugated system and
contribute to the p system with one or two p-electrons, respec-
tively [34]. The specific surface area of the NDG sample was mea-
sured and the graph is shown in Fig. 2(d). The unique mesoporous
structure of NDG contributes to the high specific surface area
(793 m2/g) which is higher than our prepared GO (684 m2/g) with
a uniform pore size distribution around 3–5 nm. Over the synthesis
of NDG sheets, beside the carbon atoms that was replaced by nitro-
gen atoms (most likely located on the reactive edge) ammonia can
also react with graphene to form hydrogen cyanide and hydrogen
(C + NH3 = HCN + H2) [35]. This reaction takes some carbon that
will make the NDG sheets more porous, which in turn resulted in
the higher specific surface area, pore volume, and pore size. The
large surface area can enhance both the heat transfer rate and
the adsorption quantity of the PA.

Fig. 2. (a and b) FESEM images of NDG (5 K, 60 K); (c) XPS spectra of NDG and graphene; (d) nitrogen adsorption/desorption isotherms of NDG. Inset in (d) is the BJH pore
size distribution.
342 M. Mehrali et al. / Applied Energy 135 (2014) 339–349

3.2. Morphology and microstructure of the form-stable PCMs the NDG. That is attributable to the PA adsorbed by the network
structure of NDG. The outcomes from TEM images are similar to
The FESEM images of the form-stable PCMs with varying NDG the results from the SEM images.
content are shown in Fig. 3. The NDG particles are generally
embedded in the PA as indicated by the white parts protruding
3.3. Shape-stabilization tests of PA/NDG composite PCMs
from the background or lying on the surface and suggested by their
irregular but sharp edges in contrast with the smooth and soft PA.
Almost certainly the majority of the PCMs which have been
At the lower nanofiller loading, the NDG particles in the nanocom-
employed in applications have numerous advantages such as high
posites are easier to recognize because of their particular structure
latent heat and minimal costs however they might need storage
as shown in Fig. 3(a) and (b). The NDG particles in PA composites
containers. The thermal efficiency of the systems will be affected
preserved their structure while the PA was adsorbed completely
by utilizing storage containers while the leakage issues also will
by the surface and pores of NDG. Fig. 3(c–e) shows nearly only
appear in energy storage systems. Shape stabilization of PCMs will
white areas, suggesting that the PA was adsorbed well into the
resolve these issues and may even change the thermal conductivity
NDG network. It can be clearly observed in Fig. 3(f) that NDG layers
of the PCMs based on the additives used. Fig. 5 shows different
possess the substantial capability to absorb the melted PA once the
structures of PA/NDG composite PCMs after drying and it can be
temperature is higher than the melting point of PA. Besides, the
clearly seen that the PA was totally absorbed by NDG particles
NDG particles develop a network which could avoid the composite
for higher mass percentage of NDG. The continuous solid shape
PCM from falling apart even when the PA is melted. Therefore, the
of the PA was observed in mass percentage of 1% and 2%, while
composite PCM still demonstrates the exact property of form-sta-
in other samples this shape was totally changed due to the absorp-
bility. In fact, in our experiments, the composite PCM has not been
tion of the PA by NDG particles.
collapsed even when the PA loading was 97%.
The results of the dropping tests are presented in Table 1.
Fig. 4 displays the TEM images of NDG along with the PA/NDG
The dropping point of pure PA was 65.03 °C and was increased
composite PCM. The TEM image shows that NDG sheets are ran-
significantly with the addition of NDG particles. The capillary and
domly compact and stacked together, demonstrating uniform lam-
surface tension forces between PA and NDG may result in a great
inar morphology like crumpled silk veil waves. The partial lining
enhancement in the shape stabilization of composite PCMs. It
level of the PA/NDG composite (Fig. 4(b)) is deeper compared to
can be mentioned that even at a NDG content of 2 wt%, the shape

Fig. 3. FESEM images of (a) S1, (b) S2, (c) S3, (d) S4, (e) S5, and (f) S5 (25 K).
 M. Mehrali et al. / Applied Energy 135 (2014) 339–349 343

Fig. 4. TEM images of (a) NDG, and (b) PA/NDG composite PCM.

Fig. 5. Dried PA/NDG composite PCMs.

Table 1 It can be seen that there was some leakage from the surface of
Dropping points of PA and PA/NDG composite PCMs. the composites that was reduced by increasing the NDG mass per-
Sample Dropping point (°C) centage however for S1 and S2 major leakage was seen while the
PA 65.03 shape of pellets was remained for all samples.
S1 77.40
S2 142.20
S3 195.60 3.4. XRD characterization of PA/NDG composites PCM
S4 >375.00a
S5 >375.00a Fig. 7 displays the XRD patterns of pure PA, NDG and PA/NDG
a
The upper limit of the measuring range is at 375 °C. composite PCM samples. A wide peak centered at around 24.2° is
observed for the NDG sample, confirming the recovery of a gra-
phitic crystal structure [28]. The XRD pattern of the PA/NDG com-
posite PCM mostly indicates the features of the PA due to the
of the composite can be preserved up to 142.2 °C, considerably strong diffraction intensity of crystalline PA. The diffraction peaks
more than the normal operating temperature range of energy stor- of NDG at 2h = 24.2° was covered by PA as a result of the higher
age systems. The dropping test was corroborated by putting the loading of the PA. The XRD patterns from the composite PCMs gen-
prepared PA/NDG pellets into the oven at 150 °C for 1 h. Fig. 6 erally show that the crystal formation of the PA within the com-
shows the effect of temperature on the shape of composite PCMs. posite PCMs is not different from pure PA.

Fig. 6. The images of PA/NDG composite PCMs (a) 30 °C, and (b) 150 °C.
344 M. Mehrali et al. / Applied Energy 135 (2014) 339–349

Fig. 8. DSC curves of the pure PA and PA/NDG composite PCMs.

Fig. 7. XRD patterns of NDG, pure PA and PA/NDG composite PCM.

storage while the total heat stored in a PCM can be calculated as

3.5. Thermal energy storage properties
follows:
Thermal energy storage properties of pure PA and prepared PA/ Z Tm Z Tf

NDG composite PCMs were studied with DSC and results are sum-
marized in Table 2. For each sample, four different parts of the pel-
let were taken to achieve accurate results and the relative standard
deviation (RSD) also was calculated. All DSC curves in Fig. 8 exhibit
one big endothermic peak with Tpeak located in the temperature
range of 64–67 °C, which is ascribed to the melting of PA. In other
words, it is a PA that performs the role of storing thermal energy.
The DSC curves shown in Fig. 8 and the data listed in Table 2 indi-
cate that the thermal energy storage properties of the prepared
form-stable PCMs are not affected by using NDG as a supporting
material.
The Tm of composite PCMs increased slightly with increasing
NDG content. The NDG can play a role in promoting the heteroge-
neously nucleated phase transition from melting to a triclinic
phase, and consequently enhances its crystallization temperature.
In addition, the relative error between theoretical and experimen-
tal latent heat values of composites decreased with increasing the
mass percentage of NDG in the prepared composites. This was indi-
cated that the molten PA was adsorbed properly by the NDG nano
sheets network. In addition, the NDG nano sheets with high and
uniform porous structure was capable to minimize seepage of
the molten PA that will improve the latent heat of the PA/NDG
composite PCMs. Here the S3 sample, which contained 3 wt% of
NDG with the relative error of less than 1%, gave the highest latent
heat value without any leakage, which was 96.9% of that of pure
PA.
Latent heat is particularly attractive due to its ability to provide
high-energy storage density in quasi-isothermal process. The
energy storage density could be increased using PCM having a high
phase change (latent heat) within the temperature range of the
Qt ¼ mC p  dT þ mam DHm þ mC p  dT ð1Þ
Ti Tm
where Q (J) – quantity of heat stored, Ti (°C) – initial temperature, Tf
(°C) – final temperature, m (kg) – mass of heat storage medium, Cp
(J/kg K) – specific heat, am – fraction of melted PCM and DHm is heat
of melting per unit mass (J/kg). The Specific heat of PCMs and its
change with temperature are not given in most of the articles.
Researchers ignore this thermal property because of the lower ther-
mal storage density of sensible heat storage. However, it has an
additional side effect on total thermal energy storage capacity of
the materials and it is better to investigate the change of specific
heat values with temperature to obtain more thermal data for engi-
neering approaches. The specific heat of the pure PA and PA/NDG
composite PCM was measured with the multiple curves method
in temperatures between 35 and 50 °C and results are shown in
Fig. 9.
The results show that the specific heats of composite PCMs
were increased by adding NDG, with the increment more pro-
nounced for higher NDG loadings. The equations are generated
according to the measured data between 35 and 50 °C and they
represent the temperature interval where no phase change occurs.
The quadratic function was chosen according to the literature [36].
The A, B and C coefficients of the solid phase are given in Table 3.
3.6. Thermal stability of the PA/NDG composite PCMs
Thermal stability is considered as an important parameter in
assessing the performance of newly designed composite PCMs
when used for thermal energy storage or thermal regulation.
Fig. 10 shows the TG curves of PA and the prepared form-stable

Table 2
Thermal energy storage properties of the PA and PA/NDG composite PCMs.

Samples Melting temperature Tm Temperature of the melting peak Phase change latent heat (DHm) Relative standard
(°C) Tp (°C) * deviation (%)
Experimental value Calculated value Relative error
(kJ/kg) (kJ/kg) (%)
Pure PA 62.40 64.12 206.32 – – 0.63
S1 63.27 65.13 199.65 204.25 2.25 0.58
S2 63.36 65.89 198.84 202.19 1.65 0.67
S3 63.78 65.99 199.48 200.13 0.32 0.73
S4 63.85 66.44 197.53 198.06 0.26 0.76
S5 63.89 66.52 195.54 196.00 0.23 0.85
*
Calculated by: DHcomposite = (1-NDG (wt%)) * DHPA.
 M. Mehrali et al. / Applied Energy 135 (2014) 339–349 345

indicates that the thicker NDG layers are more effective in enhanc-
ing the thermal stability of composite PCMs. It is also observed
from Fig. 10(a) that the composite samples exhibit different char
yields depending on the remained NDG. There is no doubt that
the composite PCMs achieved higher stability by adding NDGs as
a supporting material.

3.7. Thermal conductivity of PA/NDG composite PCMs

The PA has a high latent heat, but a low thermal conductivity

Fig. 9. Specific heat curves of PA and composite PCMs.
that may delay the thermal response to the storage and release
of latent heat. Therefore, thermal conductivity enhancement is
absolutely required when designing composite PCMs. It was
expected that adding inorganic materials with a higher thermal
conductivity can enhance the thermal conductivity of the resulting
composite PCMs. The thermal diffusivity was measured by the
laser flash method and the thermal conductivity can be calculated
as follows:

k ¼ a  q  Cp ð2Þ
Table 3
Coefficients of the second order polynomials in solid State, Cp (kJ/kg °C) = AT2 (°C) + BT where k is the thermal conductivity (W/(m k)), a is thermal diffusiv-
(°C) + C. ity (m2/s), q is density (kg/m3) and Cp is specific heat capacity (J/
Sample A B C R2 (kg K)). The thermal conductivity of PA and PA/NDG composite
PCMs are measured at 35 °C and results are shown in Fig. 11.
Pure PA 0.0024 0.1862 5.4358 0.94
S1 0.0012 0.0757 3.1324 0.99 From the data presented in Fig. 11, it can be seen that the ther-
S2 0.0015 0.1012 3.6725 0.99 mal conductivities of the PA/NDG composites (S1–S5) clearly
S3 0.0016 0.103 3.6636 0.99 improved compared to the pure PA. The maximum enhancement
S4 0.0012 0.0663 2.9727 0.99 was 517% by adding only 5 wt% of NDG and it shows that NDG
S5 0.0018 0.1078 3.8882 0.99
has a significant effect on the thermal conductivity of the PA. How-
ever, the figure also shows that the thermal conductivity does not
increase linearly with increasing NDG loading. It was suggested
PCMs. The onset and maximum weight loss temperatures are given that the arrangement of the layer structures in NDGs influences
in Table 4. PA exhibits a one-step weight loss starting at 201.48 °C the interaction between PA and NDGs. By comparing S2 with S3,
and the maximum weight loss temperatures at 275.16 °C, which is it can be seen that the enhancement was significant due to the vir-
caused by the evaporation of PA. The PA/NDG samples also show a tual heat transfer network made by NDG layers. At the high NDG
one-stage weight loss behavior while the temperatures at maxi- loading, as the content of NDG is further increased, the increasing
mum weight loss rate are significantly increased up to 349.21 °C. of thermal conductivity slows down and exhibits a tendency to
It is reasonable to believe that the PA must first break through approach a limit. As a result, the PA/NDG composite PCMs achieved
the NDG layers during the heating process and then can evaporate a much higher thermal conductivity than pure PA.
out of the composite. The NDG layers are compact and rigid Temperature versus time graphs for the PA and PA/NDG com-
enough to protect the absorbed PA from diffusing at the normal posite PCMs during heating and cooling are shown in Fig. 12. These
boiling point, thus improving the degradation temperature of com- results was taken from thermal cycler setup and indicate that the
posite PCMs. The derivative TG thermograms in Fig. 10(b) also con- total charging and discharging time for the pure PA was around
firm that the maximum weight loss occurs at a higher temperature 1100 s and was decreased to 790 s for S5 due to the higher thermal
for the composite PCMs with a higher content of NDGs. This conductivity of the S5. By comparing the melting and freezing time

Fig. 10. TG (a) and DTG (b) graphs of the PA and composite PCMs.
346 M. Mehrali et al. / Applied Energy 135 (2014) 339–349

Table 4 results also show that the thermal conductivity of composite sam-
Decomposition temperatures of the PA and composite PCMs. ples was improved by using a higher mass fraction of the NDG.
Sample Onset decomposition Maximum weight loss Moreover, the comparison of thermal properties of the prepared
temperature (°C) temperature (°C) form-stable PCM with those of different composite PCMs in the lit-
Pure 201.48 275.16 erature is shown in Table 5. It can be seen that the prepared PA/
PA NDG form-stable composite PCM enhanced the thermal properties.
S1 212.93 326.31 These results indicate that the PA/NDG composites have a potential
S2 227.17 332.26 use in the thermal management materials of an electronic devices
S3 230.08 339.53
S4 233.53 343.36
or a solar energy storage systems.
S5 237.69 349.21
3.8. Thermal effusivity of PA/NDG composite PCMs

The thermal effusivity is a measure of a material’s ability to

exchange thermal energy with its surroundings. The thermal effu-
sivity is considered to be a critical physical quantity to depict the
unsteady state heat transfer in a LHTES system. The effusivity of
a material is the square root of the product of the thermal conduc-
tivity, density and heat capacity that can be calculated as follows:
qffiffiffiffiffiffiffiffiffiffiffiffi
e¼ qKC p ð3Þ
2 (0.5)
where e (kJ/(K m s ) is thermal effusivity, K is thermal conduc-
tivity (W/(m K)), q is density (kg/m3) and Cp is specific heat capacity
(kJ/(kg K)). The thermal effusivity of the PA and PA/NDG composite
PCMs were calculated at 35 °C and listed in Table 6.
The thermal effusivity of all composite PCMs was larger than
that of pure PA, which is beneficial for the LTES systems. The ther-
mal effusivity of the PA was increased about 173% by adding only
5 wt% of NDG.

Fig. 11. Thermal conductivity of PA and PA/NDG composite PCMs at 35 °C. 3.9. Electrical resistivity

The electrical resistances of the composite PCMs were mea-

Fig. 12. Charging and discharging graph of PA and PA/NDGs composite PCM.
of PA with that of composite, it is obvious that the total charging
and discharging time was decreased by NDG addition due to the
thermal conductivity enhancement.
During cooling, temperature distribution images after 10 and
150 s are shown in Fig. 13(a–f) for PA/NDG composites. The tem-
perature distribution around the PA/NDG composites after 10 s is
almost the same for all samples while after 150 s, the lowest tem-
perature was found in S5 and the highest was in S1. From the
results it can be seen that the differences between temperatures
for S2, S3 and S4 are higher than can be explained by the thermal
conductivity graph in Fig. 11. The difference between thermal con-
ductivity of these samples are higher than others so we could
expect the results from the temperature distributions. These
sured and resistivity of samples was calculated by taking the
geometry of the sample into account using the following equation:
qðX  mÞ ¼ RðA=LÞ ð4Þ
where R is the electrical resistance of a uniform specimen at 1 Hz, L
is the length and A is the cross-sectional area of the specimen. The
electrical resistivity of different composite PCMs with a different
mass fraction of NDGs are shown in Fig. 14.
The results demonstrate that NDG additives can enhance the
electrical conductivity of the PA considerably. A percolation
threshold between 2 and 3 wt% is observed for PA/NDG composite
PCM, the place where a significant reduction in resistivity from
18,500 to 4310 X m occurs. At 5 wt% loading, the electrical resis-
tivity of the PA/NDG composite is around 1910 X m. The extremely
high aspect ratio of NDGs considerably improves the probability of
contacts between neighboring flakes and thus raises the possibility
of developing percolating networks at a lower loading content. The
main increment in thermal conductivity also happened between 2
and 3 wt%, which shows that a percolating network forms between
these two mass fractions.
3.10. Thermal reliability of the PA/NDG composite PCMs
Thermal reliability of the composite PCMs is critical to evaluate
the potency of TES systems for long period-utility. Hence, a PCM
needs to maintain its TES attributes even when it was subjected
to a prolonged thermal cycling process. Thermal reliability is also
one of the essential criteria employed for the selection of the PCMs.
For this specific purpose, in this research the thermal reliability of
the composite PCMs was examined after the exposure to 1000
melting–freezing cycles. The changes of latent heats in composite
PCMs after thermal cycling were considered to determine whether
they are thermally reliable.
 M. Mehrali et al. / Applied Energy 135 (2014) 339–349 347

Fig. 13. (a) Thermal images of the composite samples during the cooling time at 10 s (b–f) thermal images at 150 s for (b) S1, (c) S2, (d) S3, (e) S4, and (f) S5.

Table 5
Thermal conductivity and latent heat of recent form-stable PCMs in literature.

Composite Thermal conductivity (W/mK) Latent heat (kJ/Kg) References

Palmitic acid/polyaniline/exfoliated graphite nanoplatelets 1.08 157.7 [37]
Nano-graphite/paraffin 0.9362 181.81 [14]
Palmitic acid/polyaniline/copper nanowires 0.455 149 [38]
Palmitic acid/graphene oxide 1.02 101.23 [23]
Palmitic acid/graphene nanoplatelets 1.84 188.98 [8]
Paraffin/hexagonal boron nitride 0.53 177 [39]
Nitrogen-doped grapheme (4 wt%)/palmitic acid 1.54 197.53 This study

Table 6
Thermal effusivity of the PA and composite PCMs.

Sample Thermal effusivity (kJ/(K m2 s0.5))

Pure PA 0.670
S1 0.758
S2 0.895
S3 1.335
S4 1.701
S5 1.836

The composite PCMs had only one phase transition curve with
particular onset temperatures as before the cycling process. In
other words, there were clearly no extra peaks related to chemical
decomposition or phase segregation within the composites.
Through comparing the latent heat values in Fig. 15, the maximum
change in latent heat capacity was determined as 4.3 kJ/kg for S1
Fig. 14. Electrical resistivity of the composite PCMs.
and the minimum was 0.99 kJ/kg for S5. These results suggested
that the changes in the latent heat capacity of the composites after
thermal cycling were less than 1% for S3, S4 and S5 demonstrate far as the practical application is concerned, it is worthwhile to
their good thermal reliability. The results indicate that the varia- transform solid–liquid PCMs into form-stable PCMs. The prepared
tions of the phase change temperatures and latent heats are negli- PA/NDG composite form-stable PCM by applying only 3 wt% of the
gible for LHTES applications. Therefore, it can be concluded that the NDG possesses the merits of high thermal energy storage density
thermal properties of PA/NDG composite PCMs keep stable for at (199.48 kJ/kg), high thermal conductivity 0.98 (W/m K) and good
least 1000 cycling. Thermal energy storage density and speed of form stability. As a result, we believe that the prepared PA/NDG
storing and releasing are two key factors when PCMs are applied form-stable PCM certainly can find its application in low tempera-
to store thermal energy such as solar thermal energy. Besides, as ture solar-thermal applications.
348 M. Mehrali et al. / Applied Energy 135 (2014) 339–349
Fig. 15. The latent heats value of the PCMs before and after thermal cycling.
4. Conclusions
Novel form-stable PCMs were successfully prepared using NDG
as a supporting material. The results of the dropping point tests
and thermal cycling show that the shape stabilization of the PA
was achieved using only 3 wt% of NDG. The DSC results indicate
that the prepared composite has a high thermal energy storage
density (199.48 kJ/kg) while it has good stability as proven by
TGA results. The specific heat capacity and thermal effusivity of
the PA are significantly increased by using NDG as an additive.
XRD patterns show that NDG sheets were sufficiently exfoliated
and dispersed homogeneously. Meanwhile, the shape-stabilized
composite PCM containing had a good thermal reliability with
respect to the changes in its thermal properties after 1000 melting
and cooling cycles. The electrical resistivity of composite PCM
decreased significantly while the thermal conductivity of the com-
posite was increased up to 6 times that of PA. On the whole, such a
shape-stabilized PCM with a high PA content of 97 wt% can be con-
sidered as a promising PCM for LHTES due to its form-stable prop-
erty, very low content of NDG, satisfying latent heat capacity and
excellent reliability. This may open a way to enhance the heat stor-
age density of PA based shape-stabilized PCM.
Acknowledgements
This work has been financially supported by Ministry of High
Education (MOHE) of Malaysia, Grants number UM.C/HIR/MOH-
ENG/21-(D000021-16001) ‘‘Phase Change Materials (PCM) for
Energy Storage System and University of Malaya research Grant
No. UMRG RP021-2012A. The author thanks to the Bright Sparks
unit (University of Malaya) for additional financial support.
References
[1] Rathod MK, Banerjee J. Thermal stability of phase change materials used in
latent heat energy storage systems: a review. Renew Sustain Energy Rev
2013;18:246–58.
[2] Amin NAM, Bruno F, Belusko M. Effective thermal conductivity for melting in
PCM encapsulated in a sphere. Appl Energy 2014;122:280–7.
[3] Castell A, Belusko M, Bruno F, Cabeza LF. Maximisation of heat transfer in a coil
in tank PCM cold storage system. Appl Energy 2011;88:4120–7.
[4] Mahmoud S, Tang A, Toh C, Al-Dadah R, Soo SL. Experimental investigation of
inserts configurations and PCM type on the thermal performance of PCM based
heat sinks. Appl Energy 2013;112:1349–56.
[5] Pitié F, Zhao CY, Baeyens J, Degrève J, Zhang HL. Circulating fluidized bed heat
recovery/storage and its potential to use coated phase-change-material (PCM)
particles. Appl Energy 2013;109:505–13.
[6] Tahan Latibari S, Mehrali M, Mehrali M, Indra Mahlia TM, Cornelis Metselaar
HS. Synthesis, characterization and thermal properties of nanoencapsulated
phase change materials via sol–gel method. Energy 2013;61:664–72.
[7] Tahan Latibari S, Mehrali M, Mehrali M, Mahlia TMI, Metselaar HSC. Facile
preparation of carbon microcapsules containing phase-change material with
enhanced thermal properties. Sci World J 2014;2014:5.
[8] Mehrali M, Latibari ST, Mehrali M, Indra Mahlia TM, Cornelis Metselaar HS,
Naghavi MS, et al. Preparation and characterization of palmitic acid/graphene
nanoplatelets composite with remarkable thermal conductivity as a
novel shape-stabilized phase change material. Appl Therm Eng
2013;61:633–40.
[9] Zeng J-L, Zhu F-R, Yu S-B, Xiao Z-L, Yan W-P, Zheng S-H, et al. Myristic acid/
polyaniline composites as form stable phase change materials for thermal
energy storage. Sol Energy Mater Sol Cells 2013;114:136–40.
[10] Zhong Y, Zhou M, Huang F, Lin T, Wan D. Effect of graphene aerogel on thermal
behavior of phase change materials for thermal management. Sol Energy
Mater Sol Cells 2013;113:195–200.
[11] Fang G, Li H, Chen Z, Liu X. Preparation and characterization of stearic acid/
expanded graphite composites as thermal energy storage materials. Energy
2010;35:4622–6.
[12] Mehrali M, Latibari ST, Mehrali M, Metselaar HSC, Silakhori M. Shape-
stabilized phase change materials with high thermal conductivity based on
paraffin/graphene oxide composite. Energy Convers Manage 2013;67:275–82.
[13] Jiao C, Ji B, Fang D. Preparation and properties of lauric acid–stearic acid/
expanded perlite composite as phase change materials for thermal energy
storage. Mater Lett 2012;67:352–4.
[14] Li M. A nano-graphite/paraffin phase change material with high thermal
conductivity. Appl Energy 2013;106:25–30.
[15] Zhang Z, Zhang N, Peng J, Fang X, Gao X, Fang Y. Preparation and thermal
energy storage properties of paraffin/expanded graphite composite phase
change material. Appl Energy 2012;91:426–31.
[16] Sarı A, Karaipekli A. Preparation, thermal properties and thermal reliability of
palmitic acid/expanded graphite composite as form-stable PCM for thermal
energy storage. Sol Energy Mater Sol Cells 2009;93:571–6.
[17] Li M, Wu Z, Tan J. Properties of form-stable paraffin/silicon dioxide/expanded
graphite phase change composites prepared by sol–gel method. Appl Energy
2012;92:456–61.
[18] Ehid R, Weinstein RD, Fleischer AS. The shape stabilization of paraffin phase
change material to reduce graphite nanofiber settling during the phase change
process. Energy Convers Manage 2012;57:60–7.
[19] Fan L-W, Fang X, Wang X, Zeng Y, Xiao Y-Q, Yu Z-T, et al. Effects of various
carbon nanofillers on the thermal conductivity and energy storage properties
of paraffin-based nanocomposite phase change materials. Appl Energy
2013;110:163–72.
[20] Teng T-P, Cheng C-M, Cheng C-P. Performance assessment of heat storage by
phase change materials containing MWCNTs and graphite. Appl Therm Eng
2013;50:637–44.
[21] Yavari F, Fard HR, Pashayi K, Rafiee MA, Zamiri A, Yu Z, et al. Enhanced thermal
conductivity in a nanostructured phase change composite due to low
concentration graphene additives. J Phys Chem C 2011;115:8753–8.
[22] Li H, Jiang M, Li Q, Li D, Chen Z, Hu W, et al. Aqueous preparation of
polyethylene glycol/sulfonated graphene phase change composite with
enhanced thermal performance. Energy Convers Manage 2013;75:482–7.
[23] Mehrali M, Latibari ST, Mehrali M, Indra Mahlia TM, Cornelis Metselaar HS.
Preparation and properties of highly conductive palmitic acid/graphene oxide
composites as thermal energy storage materials. Energy 2013;58:628–34.
[24] Li B, Liu T, Hu L, Wang Y, Nie S. Facile preparation and adjustable thermal
property of stearic acid–graphene oxide composite as shape-stabilized phase
change material. Chem Eng J 2013;215–216:819–26.
[25] Fang X, Fan L-W, Ding Q, Wang X, Yao X-L, Hou J-F, et al. Increased thermal
conductivity of eicosane-based composite phase change materials in the
presence of graphene nanoplatelets. Energy Fuels 2013;27:4041–7.
[26] Mehrali M, Sadeghinezhad E, Latibari S, Kazi S, Mehrali M, Zubir M, et al.
Investigation of thermal conductivity and rheological properties of nanofluids
containing graphene nanoplatelets. Nanoscale Res Lett 2014;9:1–12.
[27] Mehrali M, Moghaddam E, Shirazi SFS, Baradaran S, Mehrali M, Latibari ST,
et al. Synthesis, mechanical properties, and in vitro biocompatibility with
osteoblasts of calcium silicate-reduced graphene oxide composites. ACS Appl
Mater Interfaces 2014;6:3947–62.
[28] Wang Y, Shao Y, Matson DW, Li J, Lin Y. Nitrogen-doped graphene and its
application in electrochemical biosensing. ACS Nano 2010;4:1790–8.
[29] Geng D, Yang S, Zhang Y, Yang J, Liu J, Li R, et al. Nitrogen doping effects on the
structure of graphene. Appl Surf Sci 2011;257:9193–8.
[30] Mehrali M, Sadeghinezhad E, Latibari ST, Mehrali M, Togun H, Zubir M, et al.
Preparation, characterization, viscosity, and thermal conductivity of nitrogen-
doped graphene aqueous nanofluids. J Mater Sci 2014;49:7156–71.
[31] Kim S, Drzal LT. High latent heat storage and high thermal conductive phase
change materials using exfoliated graphite nanoplatelets. Sol Energy Mater Sol
Cells 2009;93:136–42.
[32] Baradaran S, Moghaddam E, Basirun WJ, Mehrali M, Sookhakian M, Hamdi M,
et al. Mechanical properties and biomedical applications of a
nanotube hydroxyapatite-reduced graphene oxide composite. Carbon
2014;69:32–45.
[33] Guo H-L, Su P, Kang X, Ning S-K. Synthesis and characterization of nitrogen-
doped graphene hydrogels by hydrothermal route with urea as reducing-
doping agents. J Mater Chem A 2013;1:2248–55.
[34] Lu Y, Zhang F, Zhang T, Leng K, Zhang L, Yang X, et al. Synthesis and
supercapacitor performance studies of N-doped graphene materials using o-
phenylenediamine as the double-N precursor. Carbon 2013;63:508–16.
[35] Panchakarla LS, Subrahmanyam KS, Saha SK, Govindaraj A, Krishnamurthy HR,
Waghmare UV, et al. Synthesis, structure, and properties of boron- and
nitrogen-doped graphene. Adv Mater 2009;21:4726–30.
 M. Mehrali et al. / Applied Energy 135 (2014) 339–349 349

[36] Aydın AA, Aydın A. High-chain fatty acid esters of 1-hexadecanol for low
temperature thermal energy storage with phase change materials. Sol Energy
Mater Sol Cells 2012;96:93–100.
[37] Zeng J-L, Zheng S-H, Yu S-B, Zhu F-R, Gan J, Zhu L, et al. Preparation and
thermal properties of palmitic acid/polyaniline/exfoliated graphite
nanoplatelets form-stable phase change materials. Appl Energy 2014;115:
603–9.
[38] Zhu F-R, Zhang L, Zeng J-L, Zhu L, Zhu Z, Zhu X-Y, et al. Preparation and thermal
properties of palmitic acid/polyaniline/copper nanowires form-stable phase
change materials. J Therm Anal Calorim 2014;115:1133–41.
[39] Fang X, Fan L-W, Ding Q, Yao X-L, Wu Y-Y, Hou J-F, et al. Thermal energy
storage performance of paraffin-based composite phase change materials
filled with hexagonal boron nitride nanosheets. Energy Convers Manage
2014;80:103–9.