International Journal of Heat and Mass Transfer 155 (2020) 119853

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

journal homepage: www.elsevier.com/locate/hmt

Effects of graphite microstructure evolution on the anisotropic

thermal conductivity of expanded graphite/paraffin phase change
materials and their thermal energy storage performance
X.L. Wang a, B. Li a, Z.G. Qu a,∗, J.F. Zhang a, Z.G. Jin b
a
MOE Key Laboratory of Thermo-Fluid Science and Engineering, School of Energy and Power Engineering, Xi’an Jiaotong University, Xi’an, Shaanxi 710049,
China
b
Aerospace Institute of Advanced Materials & Processing Technology, Beijing 100074, China

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

Article history: The thermal conductivity of paraffin phase change materials (PCMs) is greatly enhanced by filling ex-
Received 5 February 2020 panded graphite, whereas the microstructure of graphite in the PCM is intensively affected by the fab-
Revised 6 April 2020
rication method. The effect of graphite microstructure evolution on the thermophysical properties of
Accepted 22 April 2020
the expanded graphite (EG)/paraffin composite PCM (CPCM) is not considered and remains unclear. In
this paper, the microstructure evolutions of graphite during the CPCM fabrication process are evaluated
Keywords: through SEM, XRD, Raman, and TEM. Anisotropy degree is proposed from the SEM morphology to quan-
Phase change materials titatively illustrate the distribution evolution of graphite sheets in the paraffin matrix. The results show
Microstructure evolution that the microstructure integrity of graphite is deteriorated when infiltrating the liquid paraffin into the
Anisotropy degree
EG porous bulk. Notably, the interlayer spacing of graphite is expanded as inferred from the TEM pattern.
Thermal conductivity
With increasing density, the microstructure integrity of graphite in the CPCM can be gradually improved.
Thermal energy storage
Anisotropic thermal conductivity is identified in the CPCM, and the thermal conductivity in the parallel
direction reaches 20.8 W/(m•K) at 60 °C, which is almost 70 times of the paraffin wax. The high thermal
conductivity in the CPCM can be mainly attributed to the synergetic effects induced by the intrinsic high
thermal conductivity of graphite and the high anisotropy degree of graphite sheets in the paraffin matrix.
In addition, the temperature-time curve of the CPCM shows the solid region, mushy region, and liquid
region in the parallel direction. While in the normal direction, it only shows the solid region during the
same test period. The lower temperature rise reflected in the temperature-time curve confirms that the
thermal energy storage performance of CPCM is dominated by the thermal conductivity.
© 2020 Elsevier Ltd. All rights reserved.

1. Introduction (two-dimension) [13,14], short or long fibers (one dimension)

[15,16], and three dimensional (3D) porous structure materials
Phase change materials (PCMs) have been widely used in the [2,3,17]. Specifically, those 3D fillers can be further divided as
fields of thermal energy storage (TES) units, heating and cooling expanded graphite (EG) [18,19], carbon foam [20–22], and metal
systems, and thermal management due to their high phase change foam [23,24]. Among these fillers, EG appears to be more effi-
enthalpy, wide and stable phase change temperature, and low cost cient in improving the thermal conductivity without increasing the
[1–6]. Nevertheless, one of the major drawbacks in PCMs is the specific gravity due to the high melting point, high surface area,
low thermal conductivity, which reduces the heat transfer rate and high thermal conductivity, and lightweight, which is of the most
limits their further potential applications [7–9]. frequently used thermal conductivity enhancements in TES appli-
To improve the TES performance of PCMs, numerous strate- cations [25]. The EG/Paraffin composite PCM (CPCM) can tolerate
gies have been designed to improve the thermal conductivity, high temperatures while enhancing the heat transfer rate, which
one of the most popular strategies is adding highly conduc- makes it a promising candidate in the aerospace thermal protec-
tive fillers, which can be categorized in terms of the dimen- tion systems and thermal management systems.
sion [10], i.e., nanoparticles (zero dimension) [11,12], nano-sheet Reports show that the thermal energy storage performance of
EG/Paraffin CPCM is intensively affected by the EG fillers, including
∗
the EG content, microstructure, and basic thermophysical prop-
Corresponding author.
E-mail address: zgqu@mail.xjtu.edu.cn (Z.G. Qu).
erties. Previous studies have mainly focused on investigating the

https://doi.org/10.1016/j.ijheatmasstransfer.2020.119853
0017-9310/© 2020 Elsevier Ltd. All rights reserved.
2 X.L. Wang, B. Li and Z.G. Qu et al. / International Journal of Heat and Mass Transfer 155 (2020) 119853

In this study, EG precursors with various porosities are fabri-

cated by the hot-pressing process and are further utilized to pre-
pare EG/paraffin CPCM through the vacuum infiltration process.
The microstructure evolutions of graphite sheet in the EG/paraffin
CPCM are quantitatively evaluated and further utilized to illus-
trate the anisotropy in the thermophysical properties of the CPCM.
Specifically, the enhancing mechanism of thermal conductivity in
the CPCM is illustrated in terms of the microstructure evolutions.
The fabrication approach developed in this paper is expected to
fabricate paraffin/EG composite PCMs with high latent, high melt-
ing temperature, enhanced thermal conductivity, and low cost ap-
plying in the thermal management of the aerospace field.

2. Preparation of EG/Paraffin composite phase change materials

Fig. 1. (a) Schematic illustrations of the fabrication process of EG/Paraffin CPCM
and (b) the potential orientation of the graphite sheet in the paraffin matrix. (CPCMs)

2.1. Preparation of EG bulk precursors

effects of macro parameters (such as EG content and porosity) on
the thermal conductivity, latent heat, and heat transfer rate of the To improve the thermal conductivity of CPCM, the heat transfer
CPCMs [26–30]. It is generally concluded that the thermal conduc- path of the graphite sheet should be enhanced in the EG porous
tivity and heat transfer rate of the composite PCM increases while bulk material. According to this principle, the experiment is de-
the latent heat decreases with increasing EG content. Especially, signed to maintain the lamellar structure of EG in the CPCM. To be
many theoretical and numerical models are proposed to clarify specific, the natural graphite sheet (planar size L = 350 ± 50 μm,
the mechanisms in the thermal conductivity enhancement in the density ρ = 2.25 g/cm3 ) is first converted to intercalated graphite
composite PCM [27,31–34]. In those models, several assumptions through chemical oxidation in the presence of a mixture of sulfuric
are made to simplify the numerical model building and calculation and nitric acid and then dried in a vacuum oven at 65 °C for 24 h,
process, which inevitably makes it unsatisfactory to describe the which is finally heat-treated at 900 °C for 60 s. The porous graphite
relationship between the microstructure and the thermophysical sheets are further compressed into EG bulk with increasing densi-
properties of the CPCMs. For example, the thermal conductivity ties by increasing the applied pressure. The density of the EG bulk
of graphite in the EG precursor is directly used to calculate the is obtained by averaging values from five tests (the maximum and
thermal conductivity of the CPCM, neglecting the effect of mi- minimum values are removed) using density apparatus (DMF-2).
crostructure evolution induced by the infiltration of liquid paraffin. EGs with increasing density (ρ = 0.13, 0.18, and 0.21 g/cm3 ) are
Usually, crystal defects or structure destruction is induced in the named as EG-A, EG-B, and EG-C, respectively.
graphite during the infiltration process, which inevitably decreases
the thermal conductivity of EG/Paraffin CPCMs. 2.2. Preparation of EG/Paraffin composite phase change materials
To accurately evaluate the effect of EG on the thermal energy
storage performance of EG/Paraffin CPCM, the effect of microstruc- The obtained EG bulks are further utilized as precursors to pre-
ture evolution in graphite on the corresponding thermal conduc- pare EG/paraffin CPCM. To be specific, paraffin wax with a melting
tivity should be considered. What should be noted is that the mi- point of Tm = 85 °C and latent heat ranging from 240 to 260 J/g) is
crostructure evolution of EG is greatly affected by the fabrication utilized as raw material for infiltration, which is suitable for ther-
approaches of CPCM. For example, EG based PCM is usually pre- mal management in the aerospace field. Fig. 1(a) shows the overall
pared through two routes. In the first route, paraffin and EG pow- infiltration process, the paraffin wax is heated to 130 °C and then
ders are mixed thoroughly with a mass ratio of 80% paraffin, fol- infiltrated into the porous EG bulk material by applying pressing
lowed by stirring at 80 °C and finally pressed into bulks (PCM1, stress along with the vacuum process (vacuum degree is below
density ranges from 0.7 to 0.9 g/cm3 ), and the maximum ther- 0.1 Pa) in a metal bucket with infiltration time of 60 min [26].
mal conductivity is 5.7 W/(m•K) at 0.9 g/cm3 . In the second route, Fig. 1(b) schematically shows the orientation of graphite sheets in
the EG powders are firstly pressed into porous bulks and then the CPCM, in which the graphite plane is normal to the pressing
infiltrated with liquid paraffin under continuous heating (PCM2, direction.
density ranges from 1 to 1.1 g/cm3 ). It turns out that the maxi- The CPCMs are designated as CPCM-A, CPCM-B, and CPCM-C
mum thermal conductivity of PCM2 reaches 15.7 W/(m K) (den- with corresponding average density ρ = 1.26, 1.31, and 1.34 g/cm3 ,
sity, 1.1 g/cm3 ), which is much higher than that of PCM1. The en- respectively. Fig. 2 shows the CPCMs prepared for thermal conduc-
hancement in thermal conductivity is mainly attributed to net- tivity test (a diameter of 60 mm and a thickness of 10 mm). The
work interaction and percolating threshold formed in PCM [29]. parallel and normal directions are defined based on the pressing
Microstructure evolution is used to describe the change in crys- direction in the fabrication process. Specifically, the direction par-
tal structure from amorphous to crystalline accompanied by varia- allel to the pressing direction is set as normal direction and the
tions in lattice parameters and defects, which can be evaluated by direction normal to the pressing direction is set as the parallel di-
microstructure integrity, i.e., the microstructure integrity improves rection. Fig. 2 (a-c) shows CPCM samples cut from the parallel di-
when defect decreases or crystal order increases and vice versa. rection with increasing density, and Fig. 2 (d-f) is the correspond-
Apart from the EG content and porosity, microstructure evolution ing CPCM samples cut from the normal direction.
of graphite is a very important factor in tailoring the thermal con-
ductivity of CPCMs. Nevertheless, the effect of EG microstructure 3. Microstructure characterization
evolution on the thermophysical property of CPCM is not men-
tioned and remains unclear, experimental works on characterizing To establish the relationship between the microstructure and
the microstructure evolutions of EG in the CPCM should be con- the thermophysical property of EG/paraffin CPCM, the microstruc-
ducted to evaluate the effect on the thermal energy storage perfor- ture evolutions of graphite in both the EG porous bulk and the
mance of CPCM. EG/paraffin CPCM are investigated through various approaches.
 X.L. Wang, B. Li and Z.G. Qu et al. / International Journal of Heat and Mass Transfer 155 (2020) 119853 3

Fig. 2. Images of CPCM samples prepared for thermal conductivity test from the (a–c) parallel and (d–f) normal directions with increasing bulk density.

Fig. 3. SEM morphology of (a) EG-A, (b)EG-B, and (c)EG-C bulk precursors; (d-f) the corresponding magnifications; SEM morphology of (g) composite CPCM-A, (h)CPCM-B,
and(i)CPCM-C, and (j-l) the corresponding magnifications.

Specifically, the morphology of the EG precursors and the and the CPCMs are investigated using X-ray diffraction (XRD-
CPCMs are investigated using field-emission scanning electron mi- 70 0 0S) using Cu Ka radiation (λ = 1.54060 Å). The defects of
croscopy (SEM, Gemini-500). A more detailed microstructure anal- graphite in both the EG precursors and the CPCMs are inves-
ysis is carried out using transmission electron microscopy (TEM, tigated by Raman spectroscopy (HORIBA, hr800) using 633 nm
JEOL JEM-F200). The phase compositions of the EG precursors excitation.
4 X.L. Wang, B. Li and Z.G. Qu et al. / International Journal of Heat and Mass Transfer 155 (2020) 119853

graphite sheet in the EG bulk measured from the SEM image, the
anisotropy degree of graphite (θ ) is expressed in terms of the
angle distribution ρ (θ ) as presented in Eq. (1) [35,36],
ρ
(θ )cos2 θ sin θ dθ
(θ ) = ρ (1)
(θ ) sin θ dθ
the orientation angle θ (highlighted with the green arc in
Fig. 5 (a)) between the base plane and the graphite sheets is mea-
sured and employed to calculate the angle distribution ρ (θ ) and
further the anisotropy degree (θ ). To obtain the angle distribu-
tion, the angle θ (at least 500 angles are measured in each sam-
ple to ensure the reproducibility) obtained from the SEM image is
firstly processed using Origin software to count the relative fre-
quency and plotted versus θ (Figure S1, Appendix A), the obtained
angle distribution curve is further fitted by Eq. (2) to obtain the
constant value of A1 , X0 , W, and y0 , and finally determine the spe-
cific expression of ρ (θ ),

ρ (θ ) = A1 e{[−(θ −X0 )/2W ] } + y0

Fig. 4. Variations in the graphite sheet length and graphite sheet spacing in the EG 2

bulk materials. (2)

Fig. 5(b) shows the anisotropy degree of graphite sheets in
both the EG bulk and CPCMs as a function of the density, it is
3.1. Microstructure evolution from the SEM morphology obvious that the anisotropy degree increases gradually with in-
creasing density and ranges from 0.90 to 0.98 in the EG bulk.
To evaluate the effect of infiltrating paraffin on the microstruc- While in the EG/paraffin CPCM, the anisotropy degree is lower and
ture of graphite sheets, the SEM morphology of graphite sheets ranges from 0.75 to 0.82, referring to the slight degradation in
in the EG and CPCM is investigated along the direction normal the anisotropy degree of graphite from the EG to the correspond-
to the pressing direction. Fig. 3 (a-c) shows the overall morphol- ing CPCM. The variation trend obtained in this work is consistent
ogy of graphite in the EG precursors with increasing density (EG-A, with that presented in [35]. The degradation in the anisotropy de-
EG-B, and EG-C), which performs gradual higher uniformity along gree can be attributed to the deterioration of structural integrity
the parallel direction. Fig. 3 (d-f) shows the corresponding magni- caused by the infiltration of liquid paraffin, which affects the prop-
fications, in which the lamellar structure of graphite tends to be erty anisotropy of the EG and CPCM. In addition, the anisotropy
more uniform with increasing density as highlighted by the green degree increases gradually with increasing density in both the EG
lines. Specifically, the sheet spacing is outlined by the green lines. bulk and CPCMs, which is due to the inherent high distribution
Fig. 3 (g–i) shows the morphologies of the CPCMs obtained by in- uniformity of graphite.
filtrating paraffin into the corresponding EG bulks, it is indicative
that the uniformity of graphite sheet degrades in the CPCMs. No- 3.2. Microstructure evolution from the XRD pattern
tably, as shown in the corresponding magnifications in Fig. 3 (j–l),
the distribution of graphite sheet varies from disorder pattern to The phase composition of the CPCM is analyzed by XRD to de-
gradual uniformity in the parallel direction with increasing density termine whether any chemical reaction happens during the infil-
in the CPCM. The gradual increment in the uniformity of graphite tration process. Fig. 6 shows the XRD patterns of the CPCM and
in both the EG and CPCM is considered to benefit the thermal con- corresponding EG bulk with increasing density from bottom to top.
duction in the parallel direction. The diffraction peaks of graphite in the EG precursor occur at 26.5°
To quantitatively evaluate the relationship between the mi- (002) and 54° (004), respectively. In contrast, the peak intensity in
crostructure and the EG bulk property, The graphite sheet length the corresponding CPCMs decreases compared with that in the EG
and the graphite sheet spacing between the adjacent sheets are bulk materials, indicating that the crystal structure of the graphite
measured and further plotted as a function of the bulk density. is deteriorated with the liquid paraffin infiltrated in the EG bulk
Specifically, the value of graphite sheet length and sheet spacing material. The degradation in the graphite crystal structure will de-
with distinct deviations are obtained from 10 SEM images includ- teriorate the thermal conductivity of the CPCM. It is also notable
ing more than 100 measurements. Fig. 4 shows the graphite length that the intensity of the characteristic peak of graphite (002) in-
and sheet spacing as a function of density, the graphite length in- creases significantly with increasing bulk density in both the EG
creases with bulk density in the EG precursors, while the sheet and CPCM, confirming that the crystal structure of graphite im-
spacing decreases with increasing bulk density. Notably, the in- proves with increasing density, which is beneficial to the thermal
crease in the sheet length decreases the interface thermal resis- conductivity.
tance, which in turn helps to enhance the heat transfer ability of
graphite sheets in the in-plane direction. The decreased sheet spac- 3.3. Microstructure evolution from the Raman spectrum
ing shortens the gap for heat transfer along the out-plane direc-
tion, thus enhancing the thermal conductivity of EG bulk in the The defect or disorder in the graphite induced by the infiltra-
normal direction. tion of paraffin is evaluated from the Raman spectrum of the EG
The anisotropy degree of graphite, which plays a key role in tai- bulks and CPCMs. Fig. 7 shows the Raman spectrum of EG bulks
loring the physical and thermal property of the EG and CPCM, is and CPCMs, the characteristic peaks at around 1327, 1580, and
investigated by measuring the orientation angle θ of graphite sheet 2680 cm−1 are assigned as the D peak, G peak, and 2D peak, re-
with respect to the base plane of the EG bulk [35]. Specifically, the spectively [37,38]. It is worthy to note that the frequency of all
direction that perpendicular to the pressing direction (sample fab- those three peaks increases gradually with increasing density for
rication process) is set as the base plane (0° illustrated in Fig. 5(a)) both the EG bulks and CPCMs. Moreover, the frequency of D peak
to measure the orientation angle of graphite sheets in both the is relatively low for graphite in all those samples, indicating the
EG bulk and the CPCM. Fig. 5(a) shows the orientation angle of crystal structure integrity of graphite is well maintained in the
 X.L. Wang, B. Li and Z.G. Qu et al. / International Journal of Heat and Mass Transfer 155 (2020) 119853 5

Fig. 5. (a) SEM image showing the distribution of graphite in the EG bulk and the measurement of angle θ , (b) the anisotropy degree deduced from the angle θ in the EG
bulk and CPCMs.

ratio is considered to deteriorate the thermophysical properties of

the CPCMs.

3.4. Microstructure evolution from the TEM morphology

TEM morphology is investigated to shed light on the mi-

crostructure evolution of the graphite sheet. It should be noted
that the CPCM-C and the corresponding EG-C are selected for TEM
test due to the relatively high microstructure integrity in graphite.
Fig. 8(a) shows the TEM morphologies of graphite in EG-C bulk,
the surface of graphite in the EG-C is flat and clean. The selected
area diffraction pattern (SAD, upper right corner) presents clear
spot patterns of carbon with different crystal planes, indicating
that the graphite in the EG-C bulk is of good crystal integrity. The
highly ordered graphite is also confirmed from the high-resolution
TEM (HRTEM) lattice pattern in the lower right corner (with an
interlayer spacing of 0.356 nm). In contrast, the TEM morphol-
ogy of the CPCM-C performs blurred surface with paraffin stick-
ing on the graphite, as shown in Fig. 8(b), the corresponding SAD
pattern presents disordered spots of carbon with different crystal
Fig. 6. XRD patterns showing the structure variations in the EG bulk materials and planes (upper right corner). The HRTEM in the right lower cor-
the corresponding CPCMs. ner shows that the interlayer spacing of the graphite is expanded
(0.37~0.45 nm) with the infiltration of the paraffin. The disordered
crystal and the expanded interlayer spacing are considered to de-
CPCMs. Fig. 7(a) and (b) show that the ID : IG ratio decreases grad- teriorate the heat transfer efficiency of graphite in the CPCMs.
ually with increasing density for both the EG bulks and CPCMs,
confirming that the crystal structure integrity increases with in- 4. Thermophysical properties
creasing density. Moreover, the ID : IG ratio of the CPCMs is slightly
higher compared with the corresponding EG bulks, confirming that To evaluate the effect of EG on the physical property of the
the crystal structure of graphite in the CPCM is slightly deterio- CPCM, the melting point and latent heat of the CPCMs are mea-
rated with the infiltration of liquid paraffin. Similar to the phe- sured using a differential scanning calorimeter (DSC, TA-Q20) in a
nomenon performed in the XRD data, the increase in the ID : IG nitrogen atmosphere, heated from 20 °C to 200 °C with a heating

Fig. 7. Raman spectrum showing the structure integrity variations of graphite in the (a) EG bulks and (b) the corresponding CPCMs.
6 X.L. Wang, B. Li and Z.G. Qu et al. / International Journal of Heat and Mass Transfer 155 (2020) 119853

Fig. 8. TEM morphologies showing the microstructure of graphite in the (a) EG-C bulk and (b) CPCM-C, respectively.

Table. 1
Thermophysical properties of paraffin and CPCMs.

Volume fraction (Paraffin, vol.%) Density (ρ , g/cm3 ) Melting temperature (Tm , °C) Latent heat (L, J/g)

Paraffin wax 100 0.91 85 240~260

CPCM-A 92.2 1.26 85.4 225
CPCM-B 90.0 1.31 85.3 205
CPCM-C 88.7 1.34 85.3 187

Fig. 9. Thermal conductivity of (a) the EG bulk materials and (b) the CPCMs in parallel (P) and normal (N) directions, respectively.

rate of 2 °C/min (DSC curves are offered in Appendix B). Table 1 increasing temperature (ranging from 20 to 60 °C) for the EG bulks,
presents the density, melting point, and latent heat of paraffin wax especially in the parallel direction. What should be noted is the
and CPCMs, which shows the melting point keeps stable around significant anisotropic thermal conductivity performed in parallel
85.3 °C for all the CPCMs. In contrast, the latent heat of the CPCMs and normal directions. The thermal conductivity in the parallel di-
decreases slightly (6% ~ 30%) as compared with that of the paraffin rection is higher than that in the normal direction. The maximum
wax. These stable thermophysical properties make it promising to value reaches 102 W/(m K) in the parallel direction for EG-C at
apply the CPCMs as thermal management devices in the aerospace 60 °C, which is more than three times compared with that in the
field. normal direction (32 W/(m K)).
The effect of microstructure evolution on the thermal property Similar to the EG bulk material, Fig. 9(b) shows that the ther-
of the CPCM is also investigated by measuring the thermal con- mal conductivity of CPCM increases with increasing density (rang-
ductivity of CPCMs and the corresponding EG bulks using DTC- ing from 1.26 to 1.34 g/cm3 ) in the parallel direction, and is much
300 thermal conductivity meter (the accuracy is ±0.01 °C and re- higher than that in the normal direction. The thermal conductiv-
producibility is ±0.5%). To be specific, the thermal conductivity is ity increases gradually with increasing temperatures (ranging from
tested at 20 °C, 40 °C, and 60 °C along the parallel and normal di- 20 to 60 °C) and reaches 20.8 W/(m·K) at 60 °C in the parallel di-
rection with respect to the graphite plane. Fig. 9(a) and (b) show rection, which is much higher than that reported in [39,40] and is
the thermal conductivity as a function of temperature and bulk nearly 70 times of the paraffin wax (~0.3 W/(m·K)) [41]. The en-
density for EG bulk and CPCMs from the parallel and normal di- hancement in the thermal conductivity can be attributed to the
rection, respectively. The abscissa in the left corresponds to the heat conduction network provided by the EG porous bulk, it is
temperature, and the label A-P and A-N in the right abscissa repre- confirmed from the SEM morphology that the continuous length
sent the thermal conductivity of sample-A tested from the parallel of the graphite sheet increases with increasing density, which pro-
and normal direction. Fig. 9(a) shows that the thermal conductiv- vides a prolonged heat transfer path in the bulk material in the
ity increases gradually with increasing density (ranging from 0.13 parallel direction. Moreover, the anisotropy degree increases with
to 0.21 g/cm3 ) in both the parallel and normal directions for EG increasing density, which helps reduce the interfacial thermal re-
bulk materials. Moreover, the thermal conductivity increases with sistance in the parallel direction.
 X.L. Wang, B. Li and Z.G. Qu et al. / International Journal of Heat and Mass Transfer 155 (2020) 119853 7

Fig. 10. (a) Thermal energy storage test system of the CPCM, TES test at different positions inside the sample from the (b) parallel and (c) normal directions, respectively.

Fig. 11. Temperature-time curve of the (a) PCM-B at different directions, and (b) CPCMs at position 2 with different densities in the parallel direction.

While in the normal direction, the thermal conductivity keeps (thickness 10 mm), which is designed according to the practical
steady and ranges from 6 to 7 W/(m K) for CPCMs with different application requirement of CPCM in the aerospace field.
density at different temperatures. The anisotropic thermal conduc- To investigate the effect of composite thermal conductivity on
tivity of the CPCM can be mainly attributed to the synergetic ef- the TES performance, the temperature-time curve of the CPCM
fects of the anisotropic microstructure of the graphite and the in- is collected at different positions inside the sample. As schemat-
trinsic anisotropic thermal conductivity of the graphite sheet. In ically illustrated in Fig. 10(b), the thermocouples (TT-K-30, diame-
addition, the thermal conductivity of the CPCM decreases com- ter 0.25 mm) are distributed inside the CPCM sample with a grad-
pared with that of the corresponding EG bulk, which can be ex- ual increasing distance from the heating film, which is designated
plained in terms of the infiltration of paraffin, it is reasonable that as P1 (5 mm), P2 (10 mm), and P3 (15 mm), respectively. More-
the thermal conductivity of the CPCM ranges in the middle be- over, the effect of microstructure anisotropy on the TES perfor-
tween the EG and the paraffin wax after mixing process. mance is also evaluated by collecting the temperature-time curve
of the CPCM from both the parallel and normal direction (as il-
5. Thermal energy storage performance lustrated in Fig. 10(c)). The electrical heating film is tightly fixed
on the surface of the specimen to offer a heat source with an im-
The thermal energy storage (TES) performance of the posed power density of 3086 W/m2 . What should be noted is the
EG/paraffin CPCM is mainly determined by the latent heat contact area between the heating film and the sample is different
and thermal conductivity. To evaluate the TES performance of in the parallel and normal directions, as highlighted in Fig. 10(b)
the CPCMs, the temperature-time curve of the CPCM sample and (c). The temperature signal obtained from the thermocouple is
during the heating process is collected and analyzed. As schemat- recorded every two seconds by the Agilent Data Acquisition system
ically illustrated in Fig. 10(a), the test apparatus mainly consists (34972A, USA). For comparison, the temperature-time curve of the
of three parts, i.e., the power supply module, the sample test EG bulks is also measured from the parallel and normal directions.
module, and the data acquisition/analysis module. To be spe- The temperature-time curves of the CPCM-B along the paral-
cific, the sample test module includes the electrical heating film lel and normal directions are selected and shown in Fig. 11(a),
(heating source), testing sample, thermal insulation material, and the temperature response is faster in the parallel compared with
thermocouples. The TES test is carried out in a rectangular con- that in the normal direction. To be specific, the temperature rises
tainer (260 mm × 230 mm × 100 mm) made of synthetic glass quickly in the parallel direction in the solid region (0~750 s), which
8 X.L. Wang, B. Li and Z.G. Qu et al. / International Journal of Heat and Mass Transfer 155 (2020) 119853

could be attributed to the weak heat absorption from the sensi- Acknowledgments
ble heat and the relatively low thermal conductivity at low tem-
peratures. With increasing heating time, the temperature incre- This work was supported by the National Natural Science
ment slows down and tends to be steady in the mushy region Foundation of China (grant number 51906189); the National Key
(750~20 0 0 s). The decreased temperature rise is mainly attributed Research and Development Program of China (grant number
to the high heat storage ability induced by the latent heat during 2017YFB0102703); and the Basic Science Center Program for Or-
the phase change process. Moreover, the increasing thermal con- dered Energy Conversion of the National Natural Science Founda-
ductivity at elevated temperatures further enhances the heat con- tion of China(grant number 51888103).
duction and hence reduces the temperature rise. When it comes
to the liquid region, the solid-liquid phase change process finishes
and paraffin wax is completely transferred into the liquid phase, Supplementary materials
then the sensible heat continues to absorb the heat, resulting in
the rapid increase in temperature. Supplementary material associated with this article can be
In contrast, the temperature-time curve in the normal direction found, in the online version, at doi:10.1016/j.ijheatmasstransfer.
only presents a solid region during the same test period. The slow 2020.119853.
temperature response in the normal direction is due to the low
thermal conductivity of CPCM caused by the weak heat transfer References
ability of graphite in the out-plane direction. The same variation
trend is also identified in CPCMs with different density and the [1] D. Zou, X. Ma, X. Liu, P. Zheng, Y. Hu, Thermal performance enhancement
of composite phase change materials (PCM) using graphene and carbon nan-
corresponding EG bulk materials (Figure S3, Appendix C). otubes as additives for the potential application in lithium-ion power battery,
To evaluate the effect of EG content on the TES performance Int. J. Heat Mass Transf. 120 (2018) 33–41.
of the CPCMs, the temperature-time curve of CPCM-A, CPCM-B, [2] Z.A. Qureshi, H.M. Ali, S. Khushnood, Recent advances on thermal conductivity
enhancement of phase change materials for energy storage system: a review,
and CPCM-C at position P2 in the parallel direction is selected and
Int. J. Heat Mass Transf. 127 (2018) 838–856.
shown in Fig. 11(b). It is indicative that the temperature response [3] T. Rehman, H.M. Ali, M.M. Janjua, U. Sajjad, W.M. Yan, A critical review on heat
is slower for CPCM with increasing EG content. It should be noted transfer augmentation of phase change materials embedded with porous ma-
terials/foams, Int. J. Heat Mass Transf. 135 (2019) 649–673.
that CPCM with increasing density possesses higher content of EG,
[4] H. Nazir, M. Batool, F.J.B. Osorio, M.I. Ruiz, X. Xu, K. Vignarooban, P. Phelan,
and the porous EG enhances heat transfer in the CPCM, which im- A.M. Kannan, Recent developments in phase change materials for energy stor-
proves the TES performance of the CPCM. Consequently, it can be age applications: a review, Int. J. Heat Mass Transf. 129 (2019) 491–523.
concluded that the TES performance is dominated by the thermal [5] B. Xu, P. Li, C. Chan, Application of phase change materials for thermal energy
storage in concentrated solar thermal power plants: a review to recent devel-
conductivity in the CPCM. opments, Appl. Energy 160 (2015) 286–307.
[6] K. Du, J. Calautit, Z.H. Wang, Y.P. Wu, H. Liu, A review of the applications of
phase change materials in cooling, heating and power generation in different
6. Conclusion temperature ranges, Appl. Energy 220 (2018) 242–273.
[7] E.P. Barrio, A. Godin, M. Duquesne, J. Daranlot, J. Jolly, W. Alshaer, T. Kouadio,
Expanded graphite/ paraffin composite phase change materi- A. Sommier, Characterization of different sugar alcohols as phase change ma-
terials for thermal energy storage applications, Sol. Energy Mater. Sol. Cells 159
als (CPCMs) with enhanced thermal conductivity and thermal en- (2017) 560–569.
ergy storage property are designed and obtained through infiltrat- [8] W.L. Cheng, M. Ding, X.D. Yuan, B.C. Han, Analysis of energy saving perfor-
ing liquid paraffin into the expanded graphite bulk (EG). The mi- mance for household refrigerator with thermal storage of condenser and evap-
orator, Energy Convers. Manage. 132 (2017) 180–188.
crostructure integrity of graphite is deteriorated by infiltrating liq- [9] X.D. Yuan, W.L. Cheng, Multi-objective optimization of household refrigera-
uid paraffin into the EG porous bulks as confirmed from the mi- tor with novel heat-storage condensers by Genetic algorithm, Energy Convers.
crostructure characterization. Specifically, the anisotropy degree of Manage. 84 (2014) 550–561.
[10] M. Li, B.Y. Mu, Effect of different dimensional carbon materials on the prop-
graphite decreases in the CPCM compared with that in the cor- erties and application of phase change materials: a review, Appl. Energy 242
responding EG bulk. The TEM morphology shows that the inter- (2019) 695–715.
layer spacing expands with the infiltration of liquid paraffin. What [11] A.N. Keshteli, M. Sheikholeslami, Nanoparticle enhanced PCM applications for
intensification of thermal performance in building: a review, J. Mol. Liq. 274
should be stressed is that the thermal conductivity of EG/paraffin (2018) 516–533.
CPCMs reaches 20.8 W/(m K) at 60 °C in the parallel direction, [12] S. Ebadi, S.H. Tasnim, A.A. Aliabadi, S. Mahmud, Geometry and nanoparticle
which is 70 times compared with that of the paraffin wax. This loading effects on the bio-based nano-PCM filled cylindrical thermal energy
storage system, Appl. Therm. Eng. 141 (2018) 724–740.
can be attributed to the synergetic effect of the intrinsic high ther-
[13] W. Yang, L. Zhang, Y. Guo, Z. Jiang, F. He, C. Xie, J. Fan, J. Wu, K. Zhang, Novel
mal conductivity of the graphite and the high anisotropy degree segregated-structure phase change materials composed of paraffin@ graphene
in the paraffin matrix. In addition, the temperature-time curve in- microencapsules with high latent heat and thermal conductivity, J. Mater. Sci.
dicates that the thermal energy storage performance of the CPCM 53 (2018) 2566–2575.
[14] T.Y. Zhai, X.T. Li, S. Wu, Preparation and thermal performance of form-stable
is dominated by the thermal conductivity. The CPCMs with high expanded graphite/stearic acid composite phase change materials with high
melting temperature and high thermal conductivity are promising thermal conductivity, Chin. Sci. Bull. 63 (7) (2018) 674–683.
for applying in the thermal management of the aerospace field. [15] J. Xiao, Y. Xu, F. Zhang, An analytical method for predicting the effective trans-
verse thermal conductivity of nano coated fiber composites, Compos. Struct.
189 (2018) 553–559.
[16] Z. Jiang, T. Ouyang, Y. Yang, L. Chen, X. Fan, Y. Chen, W. Li, Y. Fei, Thermal
Declaration of Competing Interest
conductivity enhancement of phase change materials with form-stable carbon
bonded carbon fiber network, Mater. Des. 143 (2018) 177–184.
To the best of our knowledge and belief, neither I nor any au- [17] L. Zhang, K. Zhou, Q. Wei, L. Ma, W. Ye, H. Li, B. Zhou, Z. Yu, C.T. Lin, J. Luo,
thors have any possible conflicts of interest. Thermal conductivity enhancement of phase change materials with 3D porous
diamond foam for thermal energy storage, Appl. Energy 233 (2019) 208–219.
[18] Z. Zhang, X. Fang, Study on paraffin/expanded graphite composite phase
change thermal energy storage material, Energy Convers. Manage. 47 (2006)
CRediT authorship contribution statement
303–310.
[19] H.J. She, Y. Chao, Z. Xu, Z. Jiao, D.J. Xing, Structure and thermal properties of
X.L. Wang: Methodology, Software, Validation, Writing - orig- expanded graphite/paraffin composite phase change material, energy sources,
inal draft. B. Li: Methodology, Validation. Z.G. Qu: Conceptualiza- part A: recovery, utilization, and environmental effects41 (2019) 86–93.
[20] Y. Tao, Y. You, Y. He, Lattice Boltzmann simulation on phase change heat trans-
tion, Supervision, Resources, Writing - review & editing. J.F. Zhang: fer in metal foams/paraffin composite phase change material, Appl. Therm.
Writing - review & editing. Z.G. Jin: Writing - review & editing. Eng. 93 (2016) 476–485.
 X.L. Wang, B. Li and Z.G. Qu et al. / International Journal of Heat and Mass Transfer 155 (2020) 119853 9

[21] W. Alshaer, S. Nada, M. Rady, E.P. Barrio, A. Sommier, Thermal management of [32] Q.L. Ren, Y.L. He, K.Z. Su, C.L. Chan, Investigation of the effect of metal foam
electronic devices using carbon foam and PCM/nano-composite, Int. J. Therm. characteristics on the PCM melting performance in a latent heat thermal
Sci. 89 (2015) 79–86. energy storage unit by pore-scale lattice Boltzmann modeling, Numer. Heat
[22] S. Nada, W. Alshaer, Comprehensive parametric study of using carbon foam Transf. Part a-Appl. 72 (10) (2017) 745–764.
structures saturated with PCMs in thermal management of electronic systems, [33] Y. Liu, Z.G. Qu, J. Guo, X.M. Zhao, Numerical study on effective thermal conduc-
Energy Convers. Manage. 105 (2015) 93–102. tivities of plain woven C/SiC composites with considering pores in interlaced
[23] B.V.S. Dinesh, A. Bhattacharya, Effect of foam geometry on heat absorption woven yarns, Int. J. Heat Mass Transf. 140 (2019) 410–419.
characteristics of PCM-metal foam composite thermal energy storage systems, [34] X. Xiao, P. Zhang, M. Li, Experimental and numerical study of heat transfer per-
Int. J. Heat Mass Transf. 134 (2019) 866–883. formance of nitrate/expanded graphite composite PCM for solar energy stor-
[24] M. Esapour, A. Hamzehnezhad, A.A.R. Darzi, M. Jourabian, Melting and solidifi- age, Energy Convers. Manage. 105 (2015) 272–284.
cation of PCM embedded in porous metal foam in horizontal multi-tube heat [35] K. Chu, X.H. Wang, F. Wang, Y.B. Li, D.J. Huang, H. Liu, W.L. Ma, F.X. Liu, H.
storage system, Energy Convers. Manage. 171 (2018) 398–410. Zhang, Largely enhanced thermal conductivity of graphene/copper composites
[25] A.D. Sarbu, Review on heat transfer analysis in thermal energy storage using with highly aligned graphene network, Carbon 127 (2018) 102–112.
latent heat storage systems and phase change materials, Int. J. Energy Res. 43 [36] C.W. Nan, R. Birringer, D.R. Clarke, H. Gleiter, Effective thermal conductivity
(1) (2019) 29–64. of particulate composites with interfacial thermal resistance, J. Appl. Phys. 81
[26] A.K. Sari, Thermal conductivity and latent heat thermal energy storage char- (10) (1997) 6692–6699.
acteristics of paraffin/expanded graphite composite as phase change material, [37] L. Malard, M. Pimenta, G. Dresselhaus, M. Dresselhaus, Raman spectroscopy in
Appl. Therm. Eng. 27 (8) (2007) 1271–1277. graphene, Phys. Rep. 473 (2009) 51–87.
[27] W.L. Cheng, W.W. Li, Y.L. Nian, W.D. Xia, Study of thermal conductive enhance- [38] A.C. Ferrari, D.M. Basko, Raman spectroscopy as a versatile tool for studying
ment mechanism and selection criteria of carbon-additive for composite phase the properties of graphene, Nat. Nanotechnol. 8 (2013) 235.
change materials, Int. J. Heat Mass Transf. 116 (2018) 507–511. [39] Z. Li, W.G. Sun, G. Wang, Experimental and numerical study on the effec-
[28] R.A. Shannaq, M.M. Farid, A novel graphite-PCM composite sphere with en- tive thermal conductivity of paraffin/expanded graphite composite, Sol. Energy
hanced thermo-physical properties, Appl. Therm. Eng. 142 (2018) 401–409. Mater. Sol. Cells 128 (2014) 447–455.
[29] H. Yin, S. Gao, C. Liao, C. Li, Z. Cai, Y. Xu, J. Liu, Self-assembly of 3D-graphite [40] M. Kenisarin, K. Mahkamov, F. Kahwash, I. Makhkamova, Enhancing thermal
block infiltrated phase change materials with increased thermal conductivity, conductivity of paraffin wax 53–57 °C using expanded graphite, Sol. Energy
J. Clean. Prod. 235 (2019) 359–368. Mater. Sol. Cells 200 (2019) 110026.
[30] Y. Zhu, J. Xiao, T. Chen, A. Chen, S. Zhou, Z. Liu, Z. Xia, Experimental and nu- [41] D. Zou, X. Ma, X. Liu, P. Zheng, Y. Hu, Thermal performance enhancement
merical investigation on composite phase change material (PCM) based heat of composite phase change materials (PCM) using graphene and carbon nan-
exchanger for breathing air cooling, Appl. Therm. Eng. 155 (2019) 631–636. otubes as additives for the potential application in lithium-ion power battery,
[31] H. Russell, Principles of heat flow in porous insulators, J. Am. Ceram. Soc. 18 Int. J. Heat Mass Transf. 120 (2018) 33–41.
(1935) 1–5.