International Journal of Heat and Mass Transfer 115 (2017) 580–588

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

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

Overall heat transfer coefficient improvement of an automobile radiator

with graphene based suspensions
C. Selvam a, R. Solaimalai Raja a, D. Mohan Lal a, Sivasankaran Harish b,⇑
a
Refrigeration & Air-Conditioning Division, Department of Mechanical Engineering, Anna University, Chennai 600 025, Tamil Nadu, India
b
International Institute for Carbon-Neutral Energy Research Kyushu University (WPI-I2CNER), 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan

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

Article history: In the present work, we report the enhancement in overall heat transfer coefficient of an automobile radi-
Received 18 April 2017 ator using graphene nanoplatelets based nanofluid as the coolant. Water-ethylene glycol mixture (70:30
Received in revised form 20 July 2017 by volume) was used as the base fluid and stable nanofluids were synthesized by non-covalent function-
Accepted 20 August 2017
alization method with volume concentrations of graphene nanoplatelets varying from 0.1% to 0.5%.
Experiments were performed in an automobile radiator for varying nanofluid mass flow rates viz.
12.5 g/s, 25 g/s, 37.5 g/s, 50 g/s, 62.5 g/s and two nanofluid inlet temperatures viz. 35 °C, 45 °C and air
Keywords:
velocity. For each condition the ambient air velocity was varied from 1 m/s to 5 m/s in steps of 1 m/s.
Nanofluid
Graphene
The convective heat transfer coefficient of nanofluid and overall heat transfer coefficient are found to
Convection increase with respect to mass flow rate, inlet temperature of nanofluid and graphene nanoplatelets load-
Overall heat transfer coefficient ing. The improvement in convective heat transfer coefficient of nanofluid play a significant role in the
Radiator enhancement of overall heat transfer coefficient. The maximum enhancement in OHTC with respect to
Pressure drop concentration is found to be
104% at 35 °C while it is found to be
81% at 45 °C for 0.5 vol%, 62.5 g/s
flow rate and 5 m/s air velocity. Further, the pressure drop of nanofluids increases with increase in mass
flow rate and graphene loading. The increase in pressure drop is significantly influenced by the mass flow
rate than by graphene nanoplatelets loading.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction conductivity of conventional heat transfer fluids due to their high

Radiators are used in automobiles to transfer heat from the
engine coolant to ambient air. Despite the name, most radiators
transfer bulk of their heat via convection. The demand to enhance
the heat transfer rate can be addressed either by increasing the
surface area or by increasing the convective heat transfer coeffi-
cient (CHTC). With regard to higher surface area many types of
compact heat exchangers have been proposed. In order to increase
the CHTC innovative heat transfer fluids are required in which con-
ventional heat transfer fluids have lower thermal conductivity. In
this regard, Choi et al. [1] coined a new class of heat transfer fluid
named ‘nanofluid’ in which solid nano-sized (<100 nm) materials
are seeded in the conventional heat transfer fluids. Nanofluids
show an enhanced thermal conductivity than that of the conven-
tional heat transfer fluids which yields a higher CHTC [2–5].
Over the past few decades, carbon based nano materials such as
carbon nanotubes (CNTs), graphene nanoplatelets (GnP), graphene
oxide, nano diamond were utilized in order to increase the thermal
⇑ Corresponding author.
E-mail address: harish@i2cner.kyushu-u.ac.jp (S. Harish).
intrinsic thermal conductivity [6–9]. Recently many research
works have been performed with graphene based nanofluids due
to its low interfacial thermal resistance and high aspect ratio
resulting in higher thermal conductivity enhancement of nanoflu-
ids [10–12]. Several researchers have conducted experiments on
convective heat transfer coefficient of nanofluids seeded with
GnP in various types of heat exchangers [13–19]. The published lit-
erature reveals that the GnP nanofluids significantly increases the
CHTC as compared to other nanofluids. The experiments on CHTC
and overall heat transfer coefficient (OHTC) of an automobile radi-
ator using various nanofluids as a coolant are reported by several
researchers [20–29].
Vermahmoudi et al. [24] studied the OHTC of an automobile
radiator using Fe2O3/water (u = 0.15%, 0.4% & 0.65%) nanofluid as
the coolant for the flow rate ranging from 0.2 to 0.5 m3/h. The
enhancement in OHTC was reported to be 13% at 0.65 vol% of
Fe2O3. Amiri et al. [25,26] performed the experiments on CHTC of
GnP/H2O-EG (H2O-EG = 40:60 by volume) nanofluid as a coolant
(w = 0.01%, 0.05%, 0.1%, and 0.2%) in an automobile radiator for
the flow rate ranging from 3 to 7 LPM. The highest enhancement
in CHTC was reported to be 130% for 0.2 wt% at 7 LPM. It is also

http://dx.doi.org/10.1016/j.ijheatmasstransfer.2017.08.071
0017-9310/Ó 2017 Elsevier Ltd. All rights reserved.
 C. Selvam et al. / International Journal of Heat and Mass Transfer 115 (2017) 580–588 581

Nomenclature

Symbols Subscript
Cp specific heat capacity (kJ/kgK) a air
Dh hydraulic diameter (m) bf basefluid
h convective heat transfer coefficient (W/m2 K) b bulk mean
k thermal conductivity (W/mK) nf nanofluid
L length (m) N1 nanofluid inlet
m_ mass flow rate (kg/s) N2 nanofluid outlet
Q heat transfer rate (W) W wall
T temperature (°C)
V velocity (m/s) Abbreviation
U overall heat transfer coefficient (W/m2 K) CHTC convective heat transfer coefficient
w weight fraction of the nanoparticles (%) EG ethylene glycol
H2O water
Greek symbols H2O-EG water-ethylene glycol mixture
g efficiency GnP/H2O-EG graphene/water-ethylene glycol mixture nanofluid
d uncertainty GnP graphene nanoplatelets
m kinematic viscosity (cSt) OHTC overall heat transfer coefficient
u volume fraction of nanoparticles (%) vol% volume fraction of the nanoparticles

reported that the increase in pressure drop is not too high with In this work CHTC and OHTC of an automobile radiator were inves-
respect to concentration of GnP. Ali et al. [27] studied the heat tigated experimentally by varying mass flow rates, GnP loading
transfer characteristics of ZnO-water (u = 0.01%, 0.08%, 0.2% and and air velocity.
0.3%) nanofluids in an automobile radiator for the flow rates rang-
ing from 7 to 11 LPM. They reported that the heat transfer rate
increases by 46% up to 0.2 vol% then it is decreases beyond 2. Materials and methods
0.2 vol%. Sandhya et al. [28] investigated the CHTC of TiO2/H2O-
EG (H2O-EG = 60:40 by volume) nanofluid (u = 0.1%, 0.3% and 2.1. Nanofluid preparation
0.5%) in automobile radiator for the Reynolds number ranging from
4000 to 15,000. The highest increase in CHTC is reported to be 37% In the present work, graphene nanoplatelets (XG sciences, USA,
at 0.5 vol%. Selvam et al. [29] studied the CHTC of GnP/H2O-EG Grade M) and water-ethylene glycol mixture (70:30 by volume)
(H2O-EG = 70:30 by volume) in an automotive radiator for the were used as the nanoparticle and basefluid respectively. Scanning
mass flow rate ranging from 10 to 100 g/s. The enhancement in (SEM) electron microscope visualization of few-layered graphene
CHTC of GnP nanofluid for the highest mass flow rate was reported nanoplatelets (GnP) is shown in Fig. 1. In this study, the non-
to be
51% at 0.5 vol% and 3 m/s air velocity. covalent functionalization method was used to synthesize of
Thus, it is inferred that nanofluids significantly enhances the nanofluids. To synthesize stable nanofluid dispersions, the sodium
CHTC and OHTC of an automobile radiator and limited works have deoxycholate (SDC) surfactant was added in the basefluid due to
been carried out with GnP nanofluids. From the above literature, the hydrophobic nature of GnP. The SDC surfactant (0.75 vol%)
the higher improvement of CHTC in automobile radiator is found was added to the basefluid and stirred well by magnetic stirrer
with the GnP based nanofluids at ultra-low concentration. The for 30 min. The basefluid along with 0.75 vol% SDC was considered
nanofluids containing spherical nanoparticles shows higher as the 0 vol% fluid in the convection experiments. GnPs were dis-
increase in pressure drop than that of carbon based nanostructures persed in the 0 vol% fluid by intensive ultrasonication (QSonica,
(ie. GnP and CNT) due to its higher density. Nowadays anti-freeze USA, Power Rating: 700 W, Frequency: 20 kHz) for 2 h. The GnP/
fluids (ethylene glycol and water-ethylene glycol mixture) are H2O-EG nanofluids with volume concentrations of 0.1%, 0.2%,
widely used in automobile radiators in order to avoid freezing dur- 0.3%, 0.4% and 0.5% were synthesized. There was no particle sedi-
ing winter. The general recommendation is water-ethylene glycol mentation observable during visual inspection of the nanofluid
mixture containing 30% (by volume) of ethylene glycol [30]. Till
now no work has been reported using GnP dispersed in water-
ethylene glycol mixture at this concentration. Lack of experimenta-
tion using specific appliances has lead to the manufacturers being
skeptical about the performance of such heat transfer fluids in real
time scenarios. The improvement in OHTC of an automobile radia-
tor using GnP nanofluid has not been reported even though a few
papers on CHTC are published. Thus this study has significance in
the context of considering GnP/H2O-EG nanofluid as a futuristic
coolant for automotive applications.
In the present work, H2O-EG (70:30 by volume) seeded with
GnP was used as the coolant. Few-layered GnP with an average
thickness of 4–8 nm was selected due to the (1) high intrinsic ther-
mal conductivity (2) high surface area (3) 2D structure (4) low
interfacial thermal resistance (5) lower size nanoplatelets can carry
more atoms in their surface which in turn higher heat transfer rate. Fig. 1. SEM visualization of GnP.
582 C. Selvam et al. / International Journal of Heat and Mass Transfer 115 (2017) 580–588

samples even for three months. The detailed stability analyses such
as UV–Vis absorption spectroscopy and zeta potential distribution
of prepared nanofluids were reported in our previous investiga-
tions [12,29,31].
2.2. Experimental set up
30 m/s, Accuracy: ±2%) respectively. To measure the pressure drop
in nanofluid a differential pressure transmitter was used. All con-
necting tubes were insulated to avoid any heat gain or loss from
or to the surrounding. Detailed view of the experimental set up
was reported in our previous work [29]. The essential geometrical
details of the radiator are listed in Table 1.

Schematic layout of the test facility with radiator to investigate 2.3. Experimental procedure
the overall heat transfer coefficient is shown in Fig. 2. Flat tube
with louvered fin is the configuration of the radiator used in this Nanofluid side and air side convective heat transfer coefficients,
study. The radiator consists of vertical aluminum tubes with ellip- overall heat transfer coefficient and pressure drop in an automo-
tical cross section. Thin perpendicular aluminum fins fill the entire bile radiator were studied for various test conditions listed in
space between the vertical tubes. An induced draft axial fan draws Table 2. For each test condition the required flow rate and temper-
air across the radiator that cools the nanofluid. Hence, air and ature were fixed and the experiment was allowed to attain steady
nanofluid flow arrangement constitutes an indirect cross-flow heat state. After attaining the steady state condition, the temperatures
exchanger configuration. This test facility has two flow circuits were logged into a data-logger and the pressure drop in the nano-
namely cold fluid (air) and hot fluid (GnP/H2O-EG nanofluid). Hot fluid was noted. The nanofluid side and airside convective heat
fluid circuit consists of a constant temperature bath with temper- transfer coefficients and overall heat transfer coefficient were esti-
ature controller, a pump and a flow meter respectively. Two tem- mated using the measured temperatures for the considered test
perature sensors TN1 and TN2 (PT 100 RTD with ±0.15 °C accuracy, conditions.
range: 200 to +850 °C) were fixed on the tube wall for measuring
inlet and outlet temperatures of nanofluid. The filament type tem-
perature sensors were fixed on the surface of the copper tube and Table 1
were perfectly insulated to prevent external heat loss or heat gain Geometric details of the automobile radiator (compact heat exchanger).
during the measurement. To measure the tube wall temperature in Sl. No Parameter value
the radiator matrix, four filament type temperature sensors were
1 Length of the heat exchanger 35 cm
fixed and properly insulated to avoid influence from external air 2 Height of the heat exchanger 30 cm
flow. The upstream and down-stream air temperatures were mea- 3 Width of the heat exchanger 1.8 cm
sured using four temperature sensors (PT 100 RTD with ±0.15 °C 4 Length of the fin 1 cm
accuracy, range: 200 to +850 °C, dimensions: outer sheath 5 Thickness of the fin 0.01 cm
6 Number of tubes 40
OD = 3 mm SS tube, length 40 mm) which were hung across the
7 Hydraulic diameter of the tube 0.373 cm
heat exchanger in the air flow path. The temperature sensors were 8 Gap between fins 0.1 cm
connected to a data-logger (Agilent, Model No: 34970 A). The mass 9 Thickness of tube 0.008 cm
flow rate of nanofluid and air velocity were measured using a cori- 10 Length of the tube 30 cm
olis mass flow meter (±0.1% accuracy) and Digital Anemometer 11 Inner surface area 4000 cm2
12 Outer surface area 28,000 cm2
(Lutron Digital Anemometer AM-4201, air velocity Range: 0.4–

Fig. 2. Schematic view of the experimental set up [29].

 C. Selvam et al. / International Journal of Heat and Mass Transfer 115 (2017) 580–588 583

 
Table 2 hDh
Test matrix of experiments. Nunf ¼ ð7Þ
k nf
Parameters Variation in parameters
Thermo-physical properties such as thermal conductivity (knf),
GnP loading (vol%) 0, 0.1, 0.2, 0.3, 0.4 & 0.5
viscosity (mnf), density (qnf) and specific heat capacity (Cpnf) of
Inlet temperature of the nanofluid (°C) 35 & 45
Mass flow rate of nanofluid (g/s) 12.5, 25, 37.5, 50 & 62.5 GnP/H2O-EG nanofluids were measured respectively using KD2
Ambient air velocity (m/s) 1, 2, 3, 4 & 5 pro thermal properties analyzer, cannon fenske viscometer, weigh-
ing balance and differential scanning calorimeter for the tempera-
tures ranging from 30 to 50 °C as reported in our previous
2.4. Data reduction investigation [19]. Initially the thermophysical properties of pure
(without surfactant) water-ethylene glycol mixture (H2O-
Heat transfer rate of nanofluid (hot fluid) was calculated as EG = 70:30 by volume) were experimentally measured for the tem-
given in Eq. (1) perature ranging from 30 to 50 °C and compared with ASHRAE
standard values. The experimental values are in good agreement
_ nf Cpnf ðT N1  T N2 Þ
Q ¼m ð1Þ with the ASHRAE standard values within ±2% deviation [19]. In this
where TN1 and TN2 correspond to the inlet and outlet temperature of study, the convective heat transfer coefficient of nanofluids was
the nanofluid respectively. determined using the reported thermophysical properties data.
Nanofluid side convective heat transfer coefficient (hnf) was cal- The reported thermophysical properties are listed in Table 3.
culated using the following Eq. (2)
3. Uncertainty analysis
Q
hnf ¼ ð2Þ
Anf ðT b  T W Þ
The convective heat transfer coefficient and overall heat trans-
where Tb and Tw correspond to the bulk mean temperature of nano- fer coefficient were estimated using the measured parameters such
fluid and wall temperature respectively. Anf is the inner surface area as temperature, mass flow rate and thermophysical properties etc.
of the tube. The uncertainty in the experiment depends on the accuracies of
Air side convective heat transfer coefficient (ha) was calculated the measuring instruments used. Uncertainty in the experiment
using the following Eq. (3) was estimated considering the accuracies of the measuring instru-
ments as reported by Moffat [32]. Table 4 shows the uncertainties
Q
ha ¼ ð3Þ in the measuring instruments along with estimated uncertainty.
Aa ðT W  T a Þ The uncertainty in the heat transfer rate (dQ) was estimated
where Ta and Tw correspond to the bulk mean temperature of air using the following Eq. (8)
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
and wall temperature respectively. Aa is the outer surface (air side)  2  2  2  2
area of radiator. @Q @Q @Q @Q
dQ ¼ dm þ dCp þ dT N1 þ dT N2
Overall heat transfer coefficient (U) based on air side surface @m @Cp @T N1 @T N2
area was calculated using the following Eq. (4) ð8Þ
1 where
U¼ ð4Þ
1
þ
Anf1
go ha hnf @Q @Q @Q @Q
Aa
¼ cp ðT N2  T N1 Þ; ¼ mðT N2  T N1 Þ; ¼ mcp ; ¼ mcp
@m @C P @T N1 @T N2
Total efficiency (ɳo) of the heat exchanger was determined as
ð9Þ
given in Eq. (5)
    rffiffiffiffiffiffiffiffi
Af Af tanhðmLÞ 2ha
go ¼ gþ1 ; g¼ ; m¼ ð5Þ Table 4
Aa Aa mL kt
Uncertainty in the measuring instruments and estimated uncertainties.
where ɳ – fin efficiency, k – thermal conductivity of the aluminum Measuring instruments/Parameter Accuracy/uncertainty
fin, t – thickness of fin, Af – finned area and L – length of the fin.
Temperature sensor (PT100-RTD) (dT) ±0.15 °C
Reynolds number of the GnP/H2O-EG nanofluid was calculated Differential pressure transducer (dDP) ±0.1%
using the following Eq. (6) Coriolis type mass flow meter (dm) ±0.1%
  Digital Anemometer (dv) ±2%
U max Dh
Renf ¼ ð6Þ Uncertainty in heat transfer rate (dQ) ±5–10%
v nf Uncertainty in CHTC of nanofluid (dhnf) ±5–10%
Uncertainty in air side CHTC (dha) ±5–10%
Nusselt number of the GnP/H2O-EG nanofluid was calculated Uncertainty in OHTC (dU) ±3–10%
using the following Eq. (7)

Table 3
Thermophysical properties of GnP/H2O-EG nanofluid [19].

Temperature/vol 30 °C 40 °C 50 °C
% 3 3
k (W/mK) m (cSt) q (kg/m ) Cp (kJ/kgK) k. (W/mK) m (cSt) q (kg/m ) Cp (kJ/kgK) k. (W/mK) m (cSt) q (kg/m3) Cp (kJ/kgK)
0 0.465 1.622 1045 3.430 0.473 1.281 1039 3.449 0.478 1.065 1033 3.481
0.1 0.483 1.720 1063 3.317 0.51 1.395 1054 3.333 0.525 1.239 1046 3.371
0.2 0.505 1.731 1070 3.267 0.533 1.429 1059 3.281 0.556 1.284 1052 3.319
0.3 0.525 1.749 1078 3.219 0.543 1.498 1067 3.229 0.572 1.360 1055 3.253
0.4 0.534 1.809 1085 3.185 0.566 1.544 1071 3.194 0.591 1.423 1059 3.227
0.5 0.557 1.845 1088 3.151 0.591 1.599 1076 3.166 0.615 1.488 1063 3.181
584 C. Selvam et al. / International Journal of Heat and Mass Transfer 115 (2017) 580–588

where dm, dCp, dT1 and dT2 are the uncertainties in mass flow rate, and various air velocities at two nanofluid inlet temperatures. The
specific heat capacity, inlet temperature of nanofluid and outlet effects of mass flow rate, inlet temperature, GnP loadings and ther-
temperature of nanofluid respectively. mophysical properties on the CHTC and OHTC were discussed. The
The uncertainty in the CHTC of nanofluid (dhnf) was estimated effects of air side heat transfer coefficient and nanofluid side heat
using the following Eq. (10) transfer coefficient on the OHTC were discussed.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2  2  2
@hi @hi @hi
dhi ¼ dQ þ dT N þ dT w ð10Þ
@Q @T N @T w 4.1. Convective heat transfer coefficient of nanofluid

where The experiments were performed as per the test conditions

@hi 1 @hi Q @hi Q given in Table 2. The variation in CHTC of nanofluid as a function
¼ ; ¼ ; ¼ ð11Þ of Reynolds number for different loadings of GnP and nanofluid
@Q Ai ðT N  T W Þ @T N Ai ðT N  T W Þ2 @T W Ai ðT N  T W Þ2
inlet temperature at 5 m/s air velocity is illustrated in Fig. 3
where dTN, dTW are the uncertainties in the mean temperature of (a) and (b). It is inferred that the CHTC is found to increase with
nanofluid and wall temperature respectively. increase in Reynolds number and GnP loading. In general the
The uncertainty in the air side CHTC (dhnf) was estimated using higher mass flow rate causes more convection and thinner bound-
the following Eq. (12) ary layer is formed which enhances the heat transfer. For a given
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi mass flow rate as particle concentration increases the heat transfer
 2  2  2
@ha @ha @ha coefficient increases even though the Reynolds number decreases.
dha ¼ dQ þ dT a þ dT w ð12Þ
@Q @T a @T w Thus the CHTC increases with particle concentration as well as
Reynolds number. The CHTC of nanofluid is higher than that of
where the basefluid (0 vol%) for all flow rates as well as particle concen-
@ha 1 @ha Q @ha Q tration. In this study the mass flow rate of nanofluid is varied from
¼ ; ¼ ; ¼ 12.5 g/s to 62.5 g/s. The corresponding increase in CHTC with
@Q Ao ðT W  T a Þ @T W Ao ðT W  T a Þ2 @T N Ao ðT W  T a Þ2
respect to mass flow rate falls in the range of
10–45% which is
ð13Þ maximum at 0.5 vol%.
where dTa is the uncertainty in mean temperature of air. As the concentration of GnP increases the CHTC of nanofluid
The uncertainty in the OHTC (U) was estimated using the fol- increases significantly. The increment in CHTC of nanofluid is sig-
lowing Eq. (14) nificant up to 0.3 vol% while it is limited beyond 0.3 vol%. For every
s
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi increase of 0.1 vol% the average CHTC increases in the range of 10%
2  2  2ffi to 30% for the considered range of flow rate. This can be attributed
@U @U @U
dU ¼ dha þ dhnf þ dg ð14Þ to the significant enhancement in thermal conductivity of nano-
@ha @hnf @ go o
fluid with respect to GnP loadings. Clustering of particles to form
where chain-like networks, Brownian motion and Brownian motion
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi induced convection and liquid layering effects is considered to be
 2
@m primary reason behind the thermal conductivity enhancement as
dgf ¼ ðUmÞ2 ð15Þ
@ho reported by several researchers [8,33–37]. Recent experiments of
Zheng et al. [38] and Harish et al. [39] using phase change compos-
ites show that Brownian motion and convection is not responsible
4. Results and discussion for the mechanism behind heat conduction enhancement. Their
results show that when the suspensions are frozen, thermal con-
In the present work, the CHTC and OHTC of an automobile radi- ductivity enhancement is higher in the solid state primarily due
ator using GnP/H2O-EG nanofluids as the coolant for various GnP to the clustering of nanoparticles. Recent effective medium calcu-
loadings (u = 0.1% to 0.5%) were investigated experimentally. The lations of Harish et al. [11] show that the interfacial thermal
experiments were carried out for various nanofluid mass flow rate boundary conductance is an order of magnitude higher in the solid

Fig. 3. (a and b) CHTC with respect to Re for various GnP loadings at 5 m/s.
 C. Selvam et al. / International Journal of Heat and Mass Transfer 115 (2017) 580–588
state due to the high crystalline nature of the host matrix com-
pared to the liquid state. Our previous model calculations [31] on
GnP based water-ethylene glycol nanofluids show similar order
of magnitude of interfacial thermal boundary conductance as
reported by Harish et al. [11]. This also shows that despite the high
intrinsic thermal conductivity of the 2D nanostructure, the interfa-
cial characteristics predominantly govern the thermal conductivity
enhancement in nanoparticle based suspensions. The interfacial
characteristics significantly depend on the sample preparation
techniques, defect concentration and the aspect ratio of the mate-
rial used. Hence it may be possible that this could be a major rea-
son for the lack of consensus on thermal conductivity
enhancement results reported in the literature.
The improved thermal conductivity of the dispersion and 2D
structure of the GnP can be the probable reasons for enhanced heat
transfer between fluid and tube wall leading to improvement in
CHTC [40,41]. Further the CHTC is generally denoted by k/dt
(k-thermal conductivity, dt- boundary layer thickness). The
improvement in thermal conductivity and reduction in boundary
layer thickness lead to increase in CHTC with respect to GnP loading
[25,42]. Particle aggregation, Brownian motion and viscosity gradi-
ent can be the probable reasons for the reduction in boundary layer
thickness [43,44]. The highest increment in CHTC is found to be

136% at 0.5 vol%, 45 °C inlet temperature and 62.5 g/s flow rate.
As the inlet temperature of nanofluid increases the CHTC
increases significantly due to increase in thermal conductivity
and reduction in viscosity. The average increase in CHTC of the
nanofluid is found to vary in the range of 14–40% when the inlet
temperature increases from 35 to 45 °C. As the inlet temperature
increases the absolute viscosity of nanofluid decreases by
10%
while the CHTC increases by
35% at 0.5 vol%. The enhancement
in CHTC at highest concentration and highest mass flow rate is
found to be 118% at 35 °C inlet temperature while it is found to
be 136% at 45 °C and 62.5 g/s.
The increase in CHTC of nanofluid with respect to Reynolds
number at 0.5 vol% for various air velocities and nanofluid inlet
temperatures are illustrated in Fig. 4(a) and (b). It is inferred that
the CHTC of nanofluid is found to increase as a function of Reynolds
number and air velocity. The average increase in CHTC is varying in
the range of 5–63% for every increase in 1 m/s air velocity. This can
be attributed to the increase in external heat flux that leads to
increase in CHTC of nanofluid. As the air velocity increases from
1 to 5 m/s the average rise in CHTC is found to be
120% for
62.5 g/s. The absolute value of CHTC is higher at the nanofluid inlet
Fig. 4. (a and b) CHTC respect to Re for various air velocities at 0.5 vol%.
585
temperature of 45 °C due to higher absolute value of thermal con-
ductivity and reduction in viscosity.
4.1.1. Nusselt number of nanofluid
The variation in Nusselt number of GnP/H2O-EG nanofluid as a
function of Reynolds number for various GnP loadings and nano-
fluid inlet temperatures is illustrated in Fig. 5(a) and (b). It is
observed that the Nusselt number of GnP/H2O-EG nanofluid is
found to increases as a function of Reynolds number, concentration
of GnP and nanofluid inlet temperature. For the considered range
of Reynolds number, the average increase in Nusselt number of
nanofluid is falls in the range of
10 to 23% at 0.5 vol%.
The enhancement in Nusselt number for the highest volume
percent of GnP (u = 0.5 vol%) and highest mass flow rate
(m = 62.5 g/s) is found to be
88% at 35 °C while it is found to be

90% at 45 °C. The average increase in Nusselt number is found
to be varying in the range of
7 to 54% for every increase in
0.1 vol% of GnP for the considered mass flow rate. The enhance-
ment in Nusselt number is significant when the particle concentra-
tion increases up to 0.3 vol% while it is limited beyond 0.3 vol%.
The absolute value of Nusselt number is found to increase with
increase in nanofluid inlet temperature. As the nanofluid tempera-
ture increases from 35 °C to 45 °C, the average enhancement in
Nusselt number is varying in the range of 7 to 28% (u = 0.5%) for
the considered range of mass flow rate.
4.2. Overall heat transfer coefficient
The variation in OHTC of automobile radiator by using
GnP/H2O-EG nanofluid as a function Reynolds number for various
GnP loadings and nanofluid inlet temperatures at 5 m/s air velocity
is illustrated in Fig. 6(a) and (b).
It is clearly seen that the OHTC of nanofluid is found to increase
with respect to nanofluid Reynolds number and volume percent of
GnP. As the nanofluid mass flow rate increases from 12.5 to
62.5 g/s the OHTC of automobile radiator increases from 14
to 41 W m2 K1 at 0 vol% while it is increases from 52 to
84 W m2 K1 at 0.5 vol% and 35 °C. The average increment in
OHTC is varying in the range of 10–50% for every increase in
0.1 vol%. The enhancement in OHTC is significant when the particle
loading increases from 0.1 to 0.3 vol% while beyond 0.3 vol% the
enhancement is limited. The maximum enhancement in OHTC
with respect to concentration is found to be
104% at 35 °C while
it is found to be
81% at 45 °C for 0.5 vol%, 62.5 g/s flow rate and
586 C. Selvam et al. / International Journal of Heat and Mass Transfer 115 (2017) 580–588

Fig. 5. (a and b) Nusselt number respect to Re for various GnP loadings at 5 m/s.

Fig. 6. (a and b) OHTC with respect to Re for various GnP loadings at 5 m/s.

5 m/s air velocity. The dispersion of nanoparticles in the basefluid
tremendously enhances the thermal conductivity and heat flux
which increases the CHTC resulting in increase of OHTC. The
absolute value of OHTC rises with respect to temperature due to
enhancement in CHTCs. The average increment in OHTC is
observed to be varying in the range of
12 to 52% when the nano-
fluid temperature increases from 35 °C to 45 °C.
The variation in OHTC of automobile radiator with respect to
Reynolds number for various air velocities and nanofluid inlet
temperatures at 0.5 vol% is illustrated in Fig. 7(a) and (b).
The OHTC rises with increase in Reynolds number of nanofluid
and air velocities due to increase in external heat flux with respect
to air velocity. The increment in air velocity causes more
convection which increases the external CHTC. The average
increment in OHTC is found to be varying in the range of
7 to
50% for every increase in 1 m/s. The highest enhancement in
OHTC is found to be
125% at 62.5 g/s nanofluid mass flow rate
and 5 m/s as compared to 1 m/s air velocity. The increment in
OHTC is limited beyond 4 m/s air velocity. The increment in
external and internal CHTC significantly increases the absolute
value of OHTC when the nanofluid temperature increases from
35 °C to 45 °C.
4.3. Pressure drop in the nanofluids
The variation in pressure drop with respect to Reynolds number
for various GnP loadings nanofluid inlet temperatures is illustrated
in Fig. 8(a) and (b). The pressure drop of nanofluid is found to
increase with respect to GnP loading (vol%) and Reynolds number.
In the range of the considered typical rise in mass flow rate and
GnP loading, it is observed that the pressure drop is significantly
influenced by the increase in mass flow rate than that of GnP load-
ing. The average increase in pressure drop is 4.61 to 7.36 kPa when
the mass flow rate increases from 12.5 to 62.5 g/s.
The increase in pressure drop is not too high with respect to
GnP loading and the maximum absolute value of the pressure drop
is found to be around
8 kPa only. The increase in pressure drop
can be attributed to the increase in density and viscosity of nano-
fluid. The pressure drop increases from 5.31 to 7.36 kPa at 35 °C
while it increases from 5.2 to 7.15 kPa at 45 °C when the concen-
tration increases from 0.1 to 0.5 vol% in which pressure drop rise
is not too high with respect to GnP loading. With regard to GnP
loading, the increment in pressure drop is higher up to 0.2 vol%
and it is limited beyond 0.2 vol%. The pressure drop decreases with
respect to temperature due to decrease in viscosity. The average
 C. Selvam et al. / International Journal of Heat and Mass Transfer 115 (2017) 580–588 587

Fig. 7. (a and b) OHTC with respect to Re for various air velocities at 0.5 vol%.

Fig. 8. (a and b) Pressure drop respect to Re for various GnP loadings at 5 m/s.

decrease in viscosity is found to be
10% resulting in pressure drop
decreases by
5% at 0.5 vol%. The pressure drop in the nanofluid is
almost constant with respect to air velocity.
5. Summary and conclusions
An experimental study was performed on the overall heat
transfer coefficient of GnP/H2O-EG nanofluid in an automobile
radiator for varying GnP loading, nanofluid mass flow rate, nano-
fluid inlet temperature and air velocity. The effects of internal
and external convective heat transfer coefficients on the overall
heat transfer coefficient were investigated. The dispersion of GnP
in the basefluid tremendously enhances the thermal conductivity
which in turn resulting in higher convective heat transfer coeffi-
cient of nanofluid. The external convective heat transfer coefficient
increases with increase in air velocity. Enhancing the internal and
external heat transfer coefficient by use of nanofluids and higher
air velocity results in higher overall heat transfer coefficient. The
highest enhancement in overall heat transfer coefficient is found
to be
104% at 0.5 vol%, 62.5 g/s flow rate and 5 m/s air velocity
as compared to 0 vol%. The maximum increase in pressure drop
is only 39% at a 62.5 g/s mass flow rate. The present graphene
based nanofluids with improved heat transfer characteristics could
be a replacement for conventional heat transfer fluids in an auto-
mobile radiator. A study on enhancing the overall heat transfer
coefficient for higher nanofluid temperature, mass flow rate and
air velocity could be a topic for future research.
Acknowledgement
CS and DM acknowledge the financial support (F.NO. 42-
896/2013 (SR), Date- 25.3.13) received from University Grants
Commission, New Delhi, India for conducting this study. SH
acknowledge the support of JSPS Kakenhi JP16H07043.
Conflict of interest
The authors declare no competing financial interest.
References
[1] S.U.S. Choi, J.A. Eastman, Enhancing thermal conductivity of fluids with
nanoparticles in developments and applications of non-Newtonian flows,
ASME FED 231/MD 66 (1995) 99–103.
[2] V. Trisaksri, S. Wongwises, Critical review of heat transfer characteristics of
nanofluids, Renew. Sustain. Energy Rev. 11 (2007) 512–523.
588 C. Selvam et al. / International Journal of Heat and Mass Transfer 115 (2017) 580–588
[3] S. Kakac, A. Pramuanjaroenkij, Review of convective heat transfer
enhancement with nanofluids, Int. J. Heat Mass Transf. 52 (2009), 3187–
3196.3.
[4] L. Godson, B. Raja, D. Mohan Lal, S. Wongwises, Enhancement of heat transfer
using nanofluids – an overview, Renew. Sustain. Energy Rev. 14 (2010) 629–
641.
[5] M.B. Bigdeli, M. Fasano, A. Cardellini, E. Chiavazzo, P. Asinar, A review on the
heat and mass transfer phenomena in nanofluid coolants with special focus on
automotive applications, Renew. Sust. Energy Rev. 60 (2016) 1615–1633.
[6] N. Koratkar, Graphene in composite materials: synthesis, characterization and
applications, Carbon 61 (2013) 650–651.
[7] S.M. Sohel Murshed, C.A. Nieto de Castro, Superior thermal features of carbon
nanotubes-based nanofluids – a review, Renew. Sustain. Energy Rev. 37 (2014)
155–167.
[8] E. Sadeghinezhad, M. Mehrali, R. Saidur, M. Mehrali, S. Tahan Latibari, A. Reza
Akhiani, H.S. Cornelis Metselaar, A comprehensive review on graphene
nanofluids: recent research, development and applications, Energy Convers.
Manage. 111 (2016) 466–487.
[9] A.K. Rasheed, M. Khalid, W. Rashmi, T.C.S.M. Gupta, A. Chan, Graphene based
nanofluids and nanolubricants – review of recent developments, Renew.
Sustain. Energy Rev. 63 (2016) 346–362.
[10] D. Konatham, A. Striolo, Thermal boundary resistance at the graphene-oil
interface, Appl. Phys. Lett. 95 (2009), 163105-1–163105-3.
[11] S. Harish, D. Orejon, Y. Takata, M. Kohno, Enhanced thermal conductivity of
phase change nanocomposite in solid and liquid state with various carbon
nano inclusions, Appl. Therm. Eng. 114 (2017) 1240–1246.
[12] C. Selvam, D. Mohan Lal, S. Harish, Thermal conductivity enhancement of
ethylene glycol and water with graphene nanoplatelets, Thermochim. Acta
642 (2016) 32–38.
[13] A. Ghozatloo, A. Rashidi, M. Shariaty-Niassar, Convective heat transfer
enhancement of graphene nanofluids in shell and tube heat exchanger, Exp.
Therm. Fluid Sci. 53 (2014) 136–141.
[14] M. Mehrali, E. Sadeghinezhad, M.A. Rosen, S.T. Latibari, M. Mehrali, H.S.C.
Metselaar, S.N. Kazi, Effect of specific surface area on convective heat transfer
of graphene nanoplatelet aqueous nanofluids, Exp. Therm. Fluid Sci. 68 (2015)
100–108.
[15] M. Mehrali, E. Sadeghinezhad, M.A. Rosen, A.R. Akhiani, S.T. Latibari, M.
Mehrali, H.S.C. Metselaar, Heat transfer and entropy generation for laminar
forced convection flow of graphene nanoplatelets nanofluids in a horizontal
tube, Int. Commun. Heat Mass Transf. 23–31 (2015).
[16] E. Sadeghinezhad, H. Togun, M. Mehrali, P. Sadeghi Nejad, S.T. Latibari, T.
Abdulrazzaq, S.N. Kazi, H.S.C. Metselaar, An experimental and numerical
investigation of heat transfer enhancement for graphene nanoplatelets
nanofluids in turbulent flow conditions, Int. J. Heat Mass Transf. 81 (2015)
41–51.
[17] H. Yarmand, S. Gharehkhani, S.F.S. Shirazi, A. Amiri, M.S. Alehashem, M. Dahari,
S.N. Kazi, Experimental investigation of thermo-physical properties,
convective heat transfer and pressure drop of functionalized graphene
nanoplatelets aqueous nanofluid in a square heated pipe, Energy Convers.
Manage. 114 (2016) 38–49.
[18] H.A. Zanjani, M. Saffar-Avval, M. Mansourkiaei, F. Sharif, M. Ahadi,
Experimental investigation of laminar forced convective heat transfer of
Graphene water nanofluid inside a circular tube, Int. J. Therm. Sci. 100 (2016)
316–323.
[19] C. Selvam, T. Balaji, D. Mohan Lal, S. Harish, Convective heat transfer coefficient
and pressure drop of water-ethylene glycol mixture with graphene
nanoplatelets, Exp. Therm. Fluid Sci. 80 (2017) 67–76.
[20] K.Y. Leong, R. Saidur, S.N. Kazi, A.H. Mamun, Performance investigation of an
automotive car radiator operated with nanofluid-based coolants (nanofluid as
a coolant in a radiator), Appl. Therm. Eng. 30 (2010) 2685–2692.
[21] S.M. Peyghambarzadeh, S.H. Hashemabadi, S.M. Hoseini, M. SeifiJamnani,
Experimental study of heat transfer enhancement using water/ethylene glycol
based nanofluids as a new coolant for car radiators, Int. Commun. Heat Mass
Transf. 38 (2011) 1283–1290.
[22] S.M. Peyghambarzadeh, S.H. Hashemabadi, M. Seifi Jamnani, S.M. Hoseini,
Improving the cooling performance of automobile radiator with Al2O3/water
nanofluid, Appl. Therm. Eng. 31 (2011) 1833–1838.
[23] S.M. Peyghambarzadeh, S.H. Hashemabadi, M. Naraki, Y. Vermahmoudi,
Experimental study of overall heat transfer coefficient in the application of
dilute nanofluids in the car radiator, Appl. Therm. Eng. 52 (2013) 8–16.
[24] Y. Vermahmoudi, S.M. Peyghambarzadeh, S.H. Hashemabadi, M. Naraki,
Experimental investigation on heat transfer performance of Fe2O3/water
nanofluid in an air-finned heat exchanger, Eur. J. Mech. B/Fluids 44 (2014)
32–41.
[25] A. Amiri, R. Sadri, M. Shanbedi, G. Ahmadi, S.N. Kazi, B.T. Chew, M.N.M. Zubir,
Synthesis of ethylene glycol-treated Graphene Nanoplatelets with one-pot,
microwave-assisted functionalization for use as a high performance engine
coolant, Energy Convers. Manage. 101 (2015) 767–777.
[26] A. Amiri Mehdi Shanbedi, B.T. Chew, S.N. Kazi, K.H. Solangi, Toward improved
engine performance with crumpled nitrogen-doped graphene based water–
ethylene glycol coolant, Chem. Eng. J. 289 (2016) 583–595.
[27] H.M. Ali, H. Ali, H. Liaquat, H.T. Bin Maqsood, M.A. Nadir, Experimental
investigation of convective heat transfer augmentation for car radiator using
ZnO-water nanofluids, Energy 84 (2015) 317–324.
[28] D. Sandhya, M.C. Sekhara Reddy, V.V. Rao, Improving the cooling performance
of automobile radiator with ethylene glycol water based TiO2 nanofluids, Int.
Commun. Heat Mass Transf. 78 (2016) 121–126.
[29] C. Selvam, D. Mohan Lal, S. Harish, Enhanced heat transfer performance of an
automobile radiator with graphene based suspensions, Appl. Therm. Eng. 123
(2017) 50–60.
[30] M.J. Nunney, Automotive Technology, third ed., SAE International, USA, 1998.
[31] C. Selvam, D. Mohan Lal, S. Harish, Thermal conductivity and specific heat
capacity of water–ethylene glycol mixture-based nanofluids with graphene
nanoplatelets, J. Therm. Anal. Calorim. 129 (2017) 947–955.
[32] R.J. Moffat, Describing the uncertainties in experimental results, Exp. Therm.
Fluid Sci. 1 (1988) 3–17.
[33] S.S. Gupta, V. Manoj Siva, S. Krishnan, T.S. Sreeprasad, P.K. Singh, T. Pradeep, S.
K. Das, Thermal conductivity enhancement of nanofluids containing graphene
nanoplatelets, J. Appl. Phys. 110 (2011), 084302-1–084302-6.
[34] S. Harish, K. Ishikawa, E. Einarsson, S. Aikawa, S.C.J. Shiomi, S. Maruyama,
Enhanced thermal conductivity of ethylene glycol with single-walled carbon
nanotube inclusions, Int. J. Heat Mass Transf. 55 (2012) 3885–3890.
[35] P. Keblinski, S.R. Phillpot, S.U.S. Choi, J.A. Eastman, Mechanisms of heat flow in
suspensions of nano-sized particles (nanofluids), Int. J. Heat Mass Transf. 45
(2002) 855–863.
[36] J. Fan, L. Wang, Review of heat conduction in nanofluids, J. Heat Transf. 133 (4)
(2011), 040801-1-14.
[37] C. Kleinstreuer, Y. Feng, Experimental and theoretical studies of nanofluid
thermal conductivity enhancement: a review, Nanoscale Res. Lett. 6 (2011)
229–241.
[38] R. Zheng, J.W. Gao, J.J. Wang, G. Chen, Reversible temperature regulation of
electrical and thermal conductivity using liquid-solid phase transitions, Nat
Commun. 2 (2011) 1–6.
[39] S. Harish, K. Ishikawa, S. Chiashi, J. Shiomi, S. Maruyama, Anomalous thermal
conduction characteristics of phase change composites with single-walled
carbon nanotube inclusions, J. Phys. Chem. C 117 (2013) 15409–15413.
[40] S. Zeinali Heris, S.Gh. Etemad, M. Nasr Esfahany, Experimental investigation of
oxide nanofluids laminar flow convective heat transfer, Int. Commun. Heat
Mass Transf. 33 (2006) 529–535.
[41] B.A. Bhanvase, M.R. Sarode, L.A. Putterwar, Intensification of convective heat
transfer in water/ethylene glycol based nanofluids containing TiO2
nanoparticles, Chem. Eng. Process. 82 (2014) 123–131.
[42] Y. Ding, H. Alias, D. Wen, R.A. Williams, Heat transfer of aqueous suspensions
of carbon nanotubes CNT nanofluids, Int. J. Heat Mass Transf. 49 (2006) 240–
250.
[43] K.H. Solangi, S.N. Kazi, M.R. Luhur, A. Badarudin, A. Amiri Rad Sadri, M.N.M.
Zubir, Samira Gharehkhani, K.H. Teng, A comprehensive review of
thermophysical properties and convective heat transfer to nanofluids,
Energy 89 (2015) 1065–1086.
[44] R.J. Phillips, R.C. Armstrong, R.A. Brown, A.L. Graham, J.R. Abbott, A constitutive
equation for concentrated suspensions that accounts for shear-induced
particle migration, Phys. Fluids A: Fluid Dyn. 4 (1992) 30–40.