European Journal of Mechanics B/Fluids 44 (2014) 32–41

Contents lists available at ScienceDirect

European Journal of Mechanics B/Fluids

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

Experimental investigation on heat transfer performance of

Fe2 O3 /water nanofluid in an air-finned heat exchanger
Y. Vermahmoudi a , S.M. Peyghambarzadeh a,∗ , S.H. Hashemabadi b , M. Naraki a
a
Department of Chemical Engineering, Mahshahr Branch, Islamic Azad University, Mahshahr, Iran
b
Computational Fluid Dynamic (CFD) Research Laboratory, School of Chemical Engineering, Iran University of Science and Technology, Narmak, Tehran,
16846, Iran

highlights
• Overall heat transfer coefficient in the car radiator measured experimentally.
• Nanofluids showed greater heat transfer performance compared with water.
• Increasing liquid and air Re increases the overall heat transfer coefficient.
• Increasing the inlet liquid temperature decreases the overall heat transfer coefficient.

article info abstract

Article history: In this paper, the overall heat transfer coefficient of water based iron oxide nanofluid in a compact air-
Received 2 July 2012 cooled heat exchanger has been measured experimentally under laminar flow conditions. The concentra-
Received in revised form tions of 0.15, 0.4 and 0.65 vol.% of stabilized Fe2 O3 /water nanofluid have been examined with variation of
31 July 2013
flow rates in the range of 0.2–0.5 m3 /h. For better dispersion of iron (III) oxide nanoparticles in water, 0.8
Accepted 23 October 2013
Available online 31 October 2013
wt% polyethylene glycol has been added and pH has been adjusted to 11.1. The air-cooled heat exchanger
is consisted of 34 vertical tubes with stadium-shaped cross section and air makes a cross flow through the
Keywords:
tube bank with variable flow rates ranging from 740 to 1009 m3 /h. Also, hot working fluid enters the heat
Overall heat transfer coefficient exchanger at different temperatures including 50, 65, and 80 °C. The results demonstrate that increasing
Iron (III) oxide the nanofluid flow rate and concentration and the air Reynolds number can improve the overall heat trans-
Nanofluid fer coefficient and heat transfer rate whereas enhancing the inlet temperature has a negative effect on the
Heat transfer rate overall heat transfer coefficient and a positive effect on the heat transfer rate. Meanwhile, the maximum
Laminar flow enhancements of the overall heat transfer coefficient and heat transfer rate compared with base fluid (dis-
tilled water) are respectively equal to 13% and 11.5% which is occurred at the concentration of 0.65 vol.%.
© 2013 Elsevier Masson SAS. All rights reserved.

1. Introduction transfer rate [5,6], although a major consideration when using sus-
pended millimeter or micrometer sized particles is that they have
Heat transfer is one of the most important and most applicable the potential to cause some severe problems such as abrasion, clog-
engineering sciences; its applications become very important with ging, high pressure drop and sedimentation of particles. Recent
respect to the subject of energy crisis and energy consumption op- progresses in nanotechnology help us to produce nanometer sized
timization in various industrial processes. For decades, efforts have particles that their mechanical and thermal properties are com-
been done to enhance the heat transfer rate, reduce heat transfer pletely different from the millimeter or micrometer sized particles.
time, minimize size of heat exchangers, and finally increase en- Choi [7] was the first to employ the nanometer sized particles in
ergy and fuel efficiencies. These efforts include passive and active conventional fluids (water or ethylene glycol) and showed consid-
methods such as using fins [1], compact heat exchangers [2], chan- erable increase in the nanofluid thermal conductivity. Since then
nels with non-circular cross sections [3], microchannels [4] etc. some studies have been done on nanofluid properties that found
The addition of solid particles into heat transfer media has long nanofluids have better properties such as heat transfer improve-
been known as one of the useful techniques for enhancing the heat ment and stability, reduction in energy for pumping the fluid, re-
duce clogging and eventually decrease in costs compared to the use
of the suspended micro particles.
∗ Corresponding author. Tel.: +98 9123241450. There are several published studies on the convective and over-
E-mail address: [email protected] (S.M. Peyghambarzadeh). all heat transfer coefficients for nanofluids and most of them show
0997-7546/$ – see front matter © 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.euromechflu.2013.10.002
 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41 33

0–2 vol.% Cu/ethylene glycol nanofluids in an automobile radiator.

Nomenclature Their results showed at a constant Reynolds number the overall
heat transfer coefficient of nanofluid is 15% higher than that of the
A Total heat transfer area (m2 ) base fluid and also about 3.8% heat transfer rate enhancement was
Cp Specific heat (J/kg K) achieved with the addition of 2% copper nanoparticles. Fotukian
Dh Hydraulic diameter (m) et al. [10] examined the turbulent convective heat transfer coeffi-
F Logarithmic temperature correction factor (dimen- cient and pressure drop of alumina/water nanofluid in a circular
sionless) tube. They showed that the maximum increase in the convective
G Gap between two fins (m) heat transfer coefficient of Al2 O3 /water nanofluid compared with
h Convective heat transfer coefficients (W/m2 K) pure water was 48% for 0.054 vol.% at Reynolds number of 10,000.
H Height (m) Jwo et al. [11] performed an investigation to analyze the effects of
k Thermal conductivity (W/m K) concentration, inlet flow temperature, and flow rates on the over-
L Length (m) all heat transfer coefficient of Al2 O3 /water nanofluid in a multi-
ṁ Liquid mass flow rate (kg/s) channel heat exchanger (MCHE). They observed that the overall
n Shape factor heat transfer coefficient ratio was higher at higher nanoparticle
Nu Nusselt number (dimensionless) concentrations and mass flow rates but the effect of temperature
P Perimeter (m) on the above mentioned coefficient was reverse. Guo et al. [12]
Pe Peclet number (dimensionless) obtained 60% convective heat transfer coefficient enhancement
Pr Prandtl number (dimensionless) compared to (60/40) ethylene glycol/water base fluid with 2 vol.%
q Heat transfer rate (W) γ -Fe2 O3 nanofluid under laminar flow conditions in a circular tube.
Q Volume flow rate (m3 /s) Peyghambarzadeh et al. [13] investigated the effect of Al2 O3 /water
Re Reynolds number (dimensionless) nanofluid in concentration of 0.1–1 vol.% on the performances of
t Thickness (m) turbulent flow convective heat transfer coefficient of an automo-
T Temperature (K) bile radiator. Authors have found that the convective heat transfer
U Overall heat transfer coefficient (W/m2 K) coefficient has been improved in comparison with the base fluid
W Width (m) up to 45%. They also have determined that with the enhancement
in nanofluid Reynolds number and temperature, convective heat
Greek letters transfer coefficient increases. In other work, Peyghambarzadeh
et al. [14] used different concentrations of water and ethylene gly-
ϕ Volume fraction of nanoparticles col as a base fluid which is conventionally used in the car’s ra-
µ Viscosity (kg/m s) diators. They have figured out that the convective heat transfer
ρ Density (kg/m3 ) coefficient of Al2 O3 /ethylene glycol nanofluid showed an increase
ψ Particle sphericity of about 40% compared with the base fluid in the best conditions.
ηf Efficiency of a single fin Zamzamian et al. [15] investigated experimentally the effect of
ηo Overall surface efficiency of a finned surface Al2 O3 and CuO nanofluids in ethylene glycol base fluid on con-
vective heat transfer coefficient in two kinds of heat exchangers,
Subscripts plate and double pipe heat exchangers at different nanofluid con-
air Air flow centrations and various temperatures under turbulent flow. They
bf Base fluid concluded that the enhancements of forced convective heat trans-
c Cold fer coefficients of 1 wt% alumina nanofluid compared to the base
exp Experimental fluid were 26.2% and 38.3% for double pipe and plate heat ex-
f Fin changers at 75 °C, respectively. These enhancements for 1 wt%
h Hot Cu nanofluid in the mentioned heat exchangers were 37.2% and
i Inside 49.33% at 75 °C. Leong et al. [16] studied the convective heat
in Inlet transfer coefficient and overall heat transfer coefficient of copper
LMTD Logarithmic mean temperature difference nanofluid in a shell and tube heat recovery exchanger. It was ob-
nf Nanofluid served that about 16.9% and 9.5% enhancements were recorded for
o Outside ethylene glycol with 1% copper nanoparticles compared with the
base fluid, respectively. For 2 vol.% water based copper nanofluid,
out Outlet
33.4% and 10.11% enhancements for convective heat transfer coef-
p Particle
ficient and overall heat transfer coefficient in laminar flow were
s Surface
recorded compared with the base fluid. Sundar et al. [17] pre-
t Tube of the air cooler
sented an experiment in plain circular tube in turbulent flow of
th Theoretical
Fe3 O4 /water nanofluid at the concentrations of 0–0.6 vol.% and
w Water
concluded the enhancement of 30.96% and 10.01% in convective
heat transfer coefficient and friction factor compared with wa-
that these coefficients are enhanced compared with that of the ter, respectively at Reynolds number of 22,000. Pandey et al. [18]
base fluid. Farajollahi et al. [8] conducted an experiment for heat reported a study of turbulent flow heat transfer characteristics
transfer characteristics of γ -Al2 O3 /water and TiO2 /water nanoflu- in a corrugated plate heat exchanger using nanofluid containing
ids in a shell and tube heat exchanger under turbulent flow con- Al2 O3 /water nanofluid at different concentrations (0–4 vol.%). The
ditions. They observed that the overall heat transfer coefficient maximum enhancements of convective and overall heat transfer
at a constant Peclet number increases with nanoparticle concen- coefficients for 2 vol.% alumina nanofluid compared with base fluid
tration for both nanofluids. The maximum enhancement of the are more than 11% and 10%, respectively. Hung et al. [19] pre-
overall heat transfer coefficients for γ -Al2 O3 /water and TiO2 /water sented an experimental investigation of thermal characteristics of
nanofluids compared with the base fluids were approximately 20% Al2 O3 /water nanofluid with 0.5 to 1.5 wt% as coolant in an air-
and 24%, respectively. Leong et al. [9] studied the effect of air cooled heat exchanger. Their results show that the maximum en-
Reynolds number, coolant Reynolds number in the application of hancement of heat exchange is of 40% which was obtained at the
34 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41

highest weight fraction of nanoparticle (1.5 wt%), and the lowest

nanofluid temperature (30 °C). Lotfi et al. [20] demonstrated that
using multi-walled carbon nanotube (MWCNT)/water nanofluid
with low concentration of 0.015 wt% in a horizontal shell and tube
heat exchanger could enhance the overall heat transfer coefficient
compared with water as base fluid. Saeedinia et al. [21] obtained
the maximum enhancement of convective heat transfer coefficient
of 12.7% with 2 wt% CuO/oil nanofluid under laminar flow con-
ditions in a smooth tube under constant heat flux at the high-
est Reynolds number, compared to that of the base oil. Recently,
Peyghambarzadeh et al. [22] studied the heat transfer performance
of the automobile radiator by calculating the overall heat transfer
coefficient (U) according to the conventional ε -NTU technique us-
ing copper oxide (CuO)/water nanofluids. They [23] also performed
a statistical analysis based on the Taguchi design of experiment
to find the effect of each operating parameter on the overall heat
transfer coefficient and the optimum values of each parameter.
There are also some reviews about nanofluids heat transfer
and nanofluid stability published by Wang and Mujumdar [24–26], Fig. 1. TEM photograph of Fe2 O3 nanoparticles.
Kakaç and Pramuanjaroenkij [27], Li et al. [28], Chen and Ding [29],
Godson et al. [30], Murshed et al. [31], Ghadimi et al. [32], Table 1
Characteristics and physical properties of Fe2 O3 nanoparticle.
Sarkar [33], and Huminic and Huminic [34].
According to our knowledge, the overall heat transfer coeffi- Mean size (nm) 40
cient of Fe2 O3 /water nanofluid in an air-cooled heat exchanger has Purity +99%
Density (kg/m3 ) 5250 [35]
not been reported experimentally yet. In this research, an exper-
Specific heat (J/kg K) 650 [35]
imental study has been performed to evaluate effective parame- Thermal conductivity (W/m K) 20 [35]
ters on the overall heat transfer coefficient of stabilized iron oxide/ Specific surface (m2 /g) 60
water nanofluid (Fe2 O3 /water) in the laminar flow regime with Morphology Spherical
using a compact air-cooled heat exchanger in different concen- Water adsorption <0.2%
Ca 0.003%
trations of the nanofluid, several operating temperatures, various Cr 0.015%
nanofluid flow levels and different air flow rates. Mn 0.139%
Al 0.052%
2. Experimental work SiO2 0.091%
SO24− 0.629%

2.1. Nanofluid preparation and stabilization

flow meter, a forced draft fan, an air flow channel, a temperature
In this study, iron (III) oxide (Fe2 O3 ) nanoparticle of approxi- controller, four thermocouples and a cross flow air-cooled finned
mately 40 nm in diameter and +99% purity has been used. The heat exchanger. Fig. 5 presents schematic of the examined heat ex-
TEM photograph of Fe2 O3 nanoparticles is shown in Fig. 1. The changer with 38.4 cm length and 33 cm width where nanofluid
physical properties and some characteristics of the nanoparticle passes through the 34 vertical finned-tubes with stadium-shaped
are listed in Table 1. Since the density difference between metal- cross section. The geometrical characteristics of the heat exchanger
lic oxide nanoparticles and water is large, two kinds of techniques with aluminum finned tubes are given in Table 2. For cooling the
have been suggested to disperse well and prevent the sedimen- working fluids, a forced fan (Techno Pars 1400 rpm) which is capa-
tation of nanoparticles in the nanofluids. One is to change pH of ble of adjusting the air flow rate from low to high (500 < Reair <
700) was installed close and face to face with the air cooler at
the nanofluid and the other is using of surfactants. Some samples
the beginning of the air flow channel which has a 1.5 m length.
were made in order to prepare the stable nanofluid. Small amount
Consequently, the air flow and working fluid have indirect cross
of Fe2 O3 nanoparticles (0.02 vol.%) were added in 100 ml distilled
flow contact and there is heat exchange between hot working fluid
water and then the pH of the suspension was adjusted at differ-
flowing in the tubes and the air across the tube bundle. A centrifu-
ent values as shown in Fig. 2. The stationary nanofluid had the
gal pump gives a constant flow rate of 10–37 l/min, the inlet flow
best stability in pH = 11.1 after two days in a stationary state.
rate is regulated by appropriate adjusting of a globe valve on the
In this pH, different concentrations of polyethylene glycol as a sur-
recycle line to make the laminar flow regime (200 < Re < 1000).
factant were added and after magnetic stirring for approximately
The working fluid fills 35% of the feed tank whose total volume is 20
one hour, the dispersion stability of the nanoparticles was analyzed
l (height of 30 cm and diameter of 30 cm). The total volume of the
by visual observation of nanoparticles sedimentation. As shown in
circulating liquid is constant in all the experiments and is equal to 7
Fig. 3, 0.8 wt% polyethylene glycol was the best concentration to
l. The flow loop was made with five layers insulated tubes (Isopipe
guarantee the dispersion of the Fe2 O3 nanoparticles in distilled wa-
0.75 in diameter) which covered with glass wool to reduce the heat
ter for seven days in a stationary state. Since the times of the ex-
loss to the surrounding. A flow meter (Technical Group LZM-15Z
periments in the present investigation are limited to maximum six
Type) was used to control and manipulate the flow rate with the
hours, so, the nanofluid is assumed to be well stabilized in all of the precision of 0.1 l/min. For heating the working fluid, two electrical
experiments. heaters (6000 W) and a PID controller were used to vary the liq-
uid inlet temperature to the heat exchanger between 50 and 80 °C.
2.2. Experimental setup and procedure Four resistance temperature detectors RTDs (Pt-100 ) were im-
plemented on the flow lines to record the inlet and outlet tempera-
As shown in Fig. 4, the experimental setup used in this research tures of the air and liquid flows. The installed thermocouple on the
includes flow lines, a feed tank, two heaters, a centrifugal pump, a heat exchanger inlet was connected to a PID controller to set the
 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41 35

Fig. 2. Stability photograph of Fe2 O3 nanofluids in different pH after two days. (a) pH = 4.1, (b) pH = 6, (c) pH = 8.35, (d) pH = 9.37, (e) pH = 10.4, (f) pH = 11.1 and
(g) pH = 12.1.

Fig. 3. Stability photograph of Fe2 O3 nanofluids in different polyethylene glycol concentrations after seven days. (a) water, (b) 0.2, (c) 0.3, (d) 0.4, (e) 0.6, (f) 0.7, (g) 0.8 and
(h) 1.0 wt% surfactant.

Fig. 4. The schematic of experimental setup.

inlet temperature of the liquid at a specific set point. The temper- Table 2
Geometrical characteristics of the heat exchanger.
atures were measured by three digital multimeters (SU-105PRR,
SAMWON ENG) and the other one by PID controller and indicator # Parameter Value (m)
(TC-4 Series), respectively with an accuracy of ±0.1 °C. 1 Length of the heat exchanger, LR 0.384
2 Width of the heat exchanger, WR 0.33
3 Height of the heat exchanger, HR 0.022
2.3. Nanofluid physical properties 4 Width of the heat exchanger tube, Wt 0.002
5 Length of the fin, Lf 0.01
6 Gap between two fins, Gf 0.001
Addition of the nanoparticles changes the density, specific heat, 7 Thickness of the fin, tf 0.00008
thermal conductivity and viscosity of the base fluid. For better 8 Perimeter of the tube, Pt 0.045
understanding, Fig. 6 depicts variations of dimensionless physical
properties of Fe2 O3 nanofluid, i.e. the ratios of physical properties
of the nanofluid to those of pure water as a function of nanoparti- Fe2 O3 nanoparticle can change more or less all the physical proper-
cle concentration. It is obvious that the addition of small amount of ties of the base fluid. The thermophysical properties are calculated
36 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41

Fig. 5. The schematic of the air-cooled heat exchanger.

So far, various theoretical and experimental studies have been

conducted and various correlations have been proposed for the dy-
namic viscosity of nanofluids. However, any general correlations
have not been established due to the lack of common understand-
ing on the mechanism of nanofluid. Almost all such formulas have
been derived from the pioneering work of Einstein [39], which
is based on the assumption of a linearly viscous fluid containing
dilute, suspended and spherical particles. Also, it must be men-
tioned that at the low concentrations of nanofluid used in this
study the result of Einstein’s formula is very close to the predic-
tions of some other general relations like Saito [40], Brinkman [41],
Lundgren [42], Batchelor [43], and Drew and Passman [44]. Some
previous studies showed the aforementioned correlations are not
made for nowadays nanofluids. They are weak approaches which
can be employed to characterize the nanofluids viscosity. There-
fore, in this study, the viscosity of nanofluid has been used from
the experimental data [12].

2.4. Calculation of the overall heat transfer coefficient

Fig. 6. The effect of nanoparticle addition on the physical properties of Fe2 O3 /water
nanofluid.
The experimental data were used to calculate the heat transfer
based on the mean bulk temperature of the nanofluid. The follow- rate and the overall heat transfer coefficient in various Reynolds
ing correlation proposed by Pak and Chou [36] is used to estimate numbers. The heat transfer rate of the working fluid is:
the nanofluids density as follows:
 
q = ṁCP Th, in − Th, out (4)
ρnf = ϕρp + (1 − ϕ) ρbf . (1) where ṁ is the mass flow rate, Th,in and Th,out are the inlet and outlet
Cpnf is the effective specific heat of the nanofluid which can be cal- temperatures of the working fluid, respectively.
culated from Xuan and Roetzel [37] relation: The overall heat transfer coefficient, Uexp , can be calculated by
the following equation [45]:
(1 − ϕ) ρbf Cpbf + ϕρp Cpp q
Cpnf = . (2) Uexp = (5)
ρnf Ao · F · ∆TLMTD
Hamilton and Crosser [38] proposed a model for liquid–solid mix- where
tures of non-spherical particles. They introduced a shape factor, n,    
to account for the effect of the shape of the particles. The thermal Th, in − Tc , out − Th, out − Tc , in
∆TLMTD = . (6)
(Th, in −Tc , out )
 
conductivity for the suspension in which the ratio of conductivity Ln
of the solid to liquid is larger than 100 (kp /kbf > 100) can be ex- (Th, out −Tc , in )
pressed as follows: F is the logarithmic temperature correction factor for the given
flow arrangement taken from [45]. ∆TLMTD is the logarithmic mean
kp + (n − 1) kbf − ϕ (n − 1) kbf − kp
 
knf = kbf (3) temperature difference.
kp + (n − 1) kbf + ϕ kbf − kp
 
Theoretically, the overall heat transfer coefficient can be ex-
pressed based on the inside tube area as Eq. (7), where the wall
where n is the empirical shape factor given by n = 3/ψ and ψ is
resistance and fouling factors are neglected.
the particle sphericity, that is defined as the ratio of the surface area
of a sphere with volume equal to that of the particle, to the surface 1 1 1
area of the particle, and in this paper n is considered to be 3. = + (7)
Uth η o ho hi (Ai /Ao )
 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41 37

Table 3
The range of operating conditions and their measurement uncertainty.
Conditions Range Unit Uncertainty [19]

Liquid flow rate 0.2–0.5 m3 /h ±0.1

Liquid velocity in each flat tube 0.03–0.1 m/s –
Air flow rate 740–1009 m3 /h ±0.1
Air velocity in each minichannel 2–3 m/s –
Pure water: 300–1000 – 5.2%
Liquid Reynolds number
Fe2 O3 /water: 200–1000 – 5.2%
Air Reynolds number 500–700 – 8.1%
Liquid inlet temperature 50–80 °C ±0.1
Air inlet temperature 35–40 °C ±0.1
Nanoparticle concentration 0–0.65 vol.% –

Fig. 8. Comparison between the measured overall heat transfer coefficient and
Fig. 7. The overall heat transfer coefficient of distilled water with the error bars.
predicted values for distilled water at Qair = 740 m3 /h.

where hi and ho are the individual convective heat transfer coeffi- 2.5. Uncertainty analysis
cients of the fluids inside and outside of tubes, respectively. Ai and
Ao are the total surface area of heat transfer inside and outside of The range of the operating variables and their relevant uncer-
the tubes, respectively. The subscripts o and i denote the air side tainty in the measurement which were calculated according to
and the tube side, respectively. ηo is overall surface efficiency of a Moffat [47] are shown in Table 3. Furthermore, maximum uncer-
finned surface and can be calculated as follows [45]: tainty in the measurement of the overall heat transfer coefficient
is ±15% and most of this uncertainty related to the temperatures
Af  measurement. Fig. 7 shows a set of data for the overall heat transfer
ηo = 1 − 1 − ηf

(8)
Ao coefficient of water with the corresponding error bars.
where Af is total surface area of the fins. The efficiency of a single
fin, ηf , can be expressed as [45]: 3. Results and discussions
 
tanh mLf 3.1. Heat transfer to distilled water
ηf = (9)
mLf
To evaluate the accuracy of the measurements, the experimen-
tal system was tested with distilled water before measuring the

2ho
m= (10) overall heat transfer coefficient of nanofluid. Fig. 8 shows the com-
kf tf parison between the measured overall heat transfer coefficient
based on Eq. (5) and the prediction of Eq. (7), in which hi was
where ho is the air side heat transfer coefficient, kf = 238 W/m K is
evaluated by two well-known empirical correlations, one of them
the thermal conductivity of aluminum fin, Lf is the fin length, and
suggested by Sieder and Tate [48] and the other developed by De-
tf is the fin thickness.
hghandokht et al. [49] for laminar flow through a tube. These two
The air side heat transfer coefficient can be calculated according
relations were shown in Eqs. (13) and (14), respectively.
to Vithayasai et al. [46] correlation:
1/3  0.14
µ

Re · Pr
Nuair = [10.145 × Ln (Reair − 46.081)] × Pr0air.33 (11) Nu = 1.86 (13)
Lt /Dh µs
where Reair and Prair is calculated at bulk temperature.
The heat transfer coefficient can be expressed as Eq. (12): Nu = 0.951 × Re0.173 × Pr1/3 (14)

Nu k where µ and µs are viscosity of working fluid at bulk and surface

h= (12) temperature of tubes and Lt is the tube length.
Dh
As shown in Fig. 8, the experimental results show good agree-
where k is the thermal conductivity of the working fluid and Dh is ments with these empirical correlations over the Reynolds number
the hydraulic diameter of the tube. range used in this study (200 < Re < 1000). The experimental
38 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41

Fig. 9. Overall heat transfer coefficient of Fe2 O3 /water nanofluid as a function of Fig. 11. Overall heat transfer coefficient of Fe2 O3 /water nanofluid as a function of
Renf at different nanofluid concentrations (Qair = 1009 m3 /h, T = 50 °C). Reair at different nanofluid concentrations (Qnf = 0.5 m3 /h, T = 50 °C).

Fig. 10. Heat transfer rate of Fe2 O3 /water nanofluid as a function of Renf at different Fig. 12. Heat transfer rate of Fe2 O3 /water nanofluid as a function of Reair at differ-
concentrations (Qair = 1009 m3 /h, T = 50 °C). ent nanofluid concentrations (Qnf = 0.5 m3 /h, T = 50 °C).

overall heat transfer coefficients at different inlet water temper- at 0.65 vol.% and Re = 600 which is 90 W/m2 K. The same graph
atures including 50, 65, and 80 °C and at different air flow rates obtained for the heat transfer rate which is demonstrated in Fig. 10.
including 740, 870 and 1009 m3 /h show 7% and 17% absolute av- As can be seen in Fig. 10, the heat transfer rate increases when the
erage errors with respect to Sieder–Tate [48] and Dehghandokht Reynolds number and the nanofluid concentration increase. The
et al. [49] correlations, respectively. enhancement of about 11.5% in the heat transfer rate is observed
compared with the base fluid. Increasing the heat transfer rate is
3.2. Heat transfer to nanofluid the most important reason for using nanofluids.

The analysis was performed based on the conditions of working 3.2.2. Effect of air Reynolds number
fluid shown in Table 3. However, nanofluid and air flow rates, The effects of air Reynolds number on the overall heat transfer
nanoparticles concentration and nanofluid inlet temperature to coefficient and the heat transfer rate of the air cooler are discussed
the heat exchanger were varied in order to determine the overall by varying the air Reynolds number in the range of 500–700.
heat transfer coefficient and the heat transfer rate. Figs. 11 and 12 show the overall heat transfer coefficient and the
heat transfer rate of Fe2 O3 /water nanofluid versus air Reynolds
3.2.1. Effect of the nanofluid Reynolds number number at different nanofluid concentrations while the nanofluid
The nanofluid Reynolds number is varied from 200 to 1000 flow rate (0.5 m3 /h) and inlet temperature (50 °C) are constant.
which shows the flow regime is laminar. Fig. 9 presents the As shown in Figs. 11 and 12, the overall heat transfer coefficient
measured overall heat transfer coefficient of the Fe2 O3 /water and the heat transfer rate of the nanofluid in all the concentrations
nanofluid as a function of the nanofluid Reynolds number at dif- increase as the flow rate of the air and consequently air Reynolds
ferent nanoparticle concentrations and at a fixed air flow rate number increase. This enhancement is greater at higher nanofluid
(1009 m3 /h) and liquid inlet temperature (50 °C). From the results, concentrations. When the air flow rate increases, more energies
the overall heat transfer coefficient of nanofluids increases signifi- transfer from the nanofluid due to bulk movements of the air.
cantly with Reynolds number. The overall heat transfer coefficient
at a constant Reynolds number increases with the nanoparticle 3.2.3. Effect of nanofluid inlet temperature
concentration compared with the base fluid. As shown in Fig. 9, In order to find the temperature effect on the overall heat trans-
the maximum value of the overall heat transfer coefficient occurs fer coefficient and also on the heat transfer rate, the nanofluids are
 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41 39

Fig. 13. Overall heat transfer coefficient of Fe2 O3 /water nanofluid as a function
of inlet temperature at different nanofluid flow rates (Qair = 1009 m3 /h, vol.%= Fig. 15. Ratio of the overall heat transfer coefficient of nanofluid to that of water
0.65). as a function of nanofluid concentration at various nanofluid flow rates (Qair =
1009 m3 /h, Tnf = 50 °C).

Fig. 14. Heat transfer rate of Fe2 O3 /water nanofluid as a function of inlet temper- Fig. 16. Comparison between the experimental results and the calculated values
ature at different nanofluid flow rates (Qair = 1009 m3 /h, vol.% = 0.65). from the correlation suggested by Xuan and Li.

tested at three different temperatures including 50, 65 and 80 °C. 3.3. Effect of nanoparticle concentration
The variation of the overall heat transfer coefficient with nanofluid
inlet temperature at different nanofluid flow rates and at a fixed The effect of Fe2 O3 nanoparticle concentration on the ratio of
air flow rate (1009 m3 /h) and nanoparticle concentration (0.65 the overall heat transfer coefficient of nanofluid to the base fluid at
vol.%) is shown in Fig. 13. It is found that the overall heat trans- various nanofluid flow rates and fixed air flow rate (1009 m3 /h)
fer coefficient decreases with the increase in the nanofluid inlet and nanofluid inlet temperature (50 °C) is shown in Fig. 15.
temperature. As shown in Fig. 13, the overall heat transfer coeffi- Nanofluids with higher concentration and higher flow rates create
cient reaches to its lowest value, 72 W/m2 K at Tnf = 80 °C and higher overall heat transfer coefficient than that of the base fluid.
Qnf = 0.2 m3 /h. The overall heat transfer coefficient decreases About 13% improvement in the overall heat transfer coefficient can
about 5% with the increase in nanofluid temperature from 50 to be achieved with the addition of 0.65 vol.% Fe2 O3 nanoparticles at
80 °C at the nanofluid flow rate of 0.5 m3 /h. Two mechanisms may Qnf = 0.5 m3 /h. So, when the nanofluid is used as the cooling fluid
cause this reduction in the overall heat transfer coefficient include: in the air-cooled heat exchanger instead of water, reduction in the
(1) rapid alignment of nanoparticles in lower viscosity fluids, lead- heat transfer area can be obtained.
ing to less contact between nanoparticles and (2) the depletion of Generally, the enhancement of heat transfer coefficient de-
particles in the near-wall fluid phase [50] leading to an intrinsi- pends on increasing of the fluid thermal conductivity and decreas-
cally lower thermal conductivity layer at the wall. The effect of in- ing of thermal boundary layer thickness. Thermal conductivity of
let temperature on the heat transfer rate is shown in Fig. 14. It is the nanofluids increases with the increase in the nanoparticle con-
clear that unlike the effect of temperature on the overall heat trans- centration. Decreasing the thickness of the thermal boundary layer
fer coefficient, the effect of temperature on the heat transfer rate can be due to the mobility of particles near the wall, migration of
is positive at different nanofluid flow rates and at a fixed air flow them to the center of tube, and the reduction of viscosity at the wall
rate (1009 m3 /h) and nanoparticle concentration (0.65 vol.%). Heat region [51].
transfer rate increases significantly with the enhancement of the
inlet temperature. This increase in the heat transfer rate becomes 3.3.1. Comparison with available correlations
greater at higher nanofluid flow rates. The heat transfer rate in- In Fig. 16 the experimental results of the overall heat transfer
creases about 116% with increasing in the nanofluid temperature coefficient of Fe2 O3 /water nanofluid are compared with the predic-
from 50 to 80 °C at the nanofluid flow rate of 0.5 m3 /h. tion of Xuan and Li [52] correlation. This correlation was provided
40 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41

for the laminar flow of nanofluids inside a tube as follows: [9] 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
0.754
Nu = 0.4328 1 + 11.285ϕ Pe0nf.218 Re0nf.33 Pr0nf.4 .
 
(15) a coolant in a radiator), Appl. Therm. Eng. 30 (2010) 2685–2692.
[10] S.M. Fotukian, M. Nasr Esfahany, Experimental investigation of turbulent
The results show that at 0.65 and 0.4 vol.% of Fe2 O3 nanopar- convective heat transfer of dilute γ -Al2 O3 /water nanofluid inside a circular
ticles concentrations and at a fixed air flow rate (1009 m3 /h) tube, Int. J. Heat Fluid Flow 31 (2010) 606–612.
[11] Ching-Song Jwo, Lung-Yue Jeng, Tun-Ping Teng, Chien-Chih Chen, Perfor-
and nanofluid inlet temperature (80 °C) a good agreement ex- mance of overall heat transfer in multi-channel heat exchanger by alumina
ists between the experimental results and the predicted values by nanofluid, J. Alloys Compd. 504S (2010) S385–S388.
Eq. (15) specially at higher Reynolds numbers. For example, the av- [12] Shou-Zhu Guo, Yang Li, Ji-Sen Jiang, Hua-Qing Xie, Nanofluids containing
γ -Fe2 O3 nanoparticles and their heat transfer enhancements, Nanoscale Res.
erage deviation of the experimental data and the predicted val- Lett. 5 (2010) 1222–1227.
ues for 0.4 vol.% of the nanoparticles is 3.5%, while this average [13] S.M. Peyghambarzadeh, S.H. Hashemabadi, M.S. Jamnani, S.M. Hoseini,
difference for 0.65 vol.% of the nanoparticles is 3%. Therefore, for Improving the cooling performance of automobile radiator with Al2 O3 /water
nanofluid, Appl. Therm. Eng. 31 (10) (2011) 1833–1838.
Fe2 O3 /water nanofluid the experimental results are very close to
[14] S.M. Peyghambarzadeh, S.H. Hashemabadi, S.M. Hoseini, M.S. Jamnani,
the predicted values of Xuan and Li correlation at the low nanopar- Experimental study of heat transfer enhancement using water/ethylene glycol
ticle concentrations. based nanofluids as a new coolant for car radiators, Int. Commun. Heat Mass
Transfer 38 (9) (2011) 1283–1290.
[15] Amirhossein Zamzamian, Shahin Nasseri Oskouie, Ahmad Doosthoseini,
4. Conclusion Aliakbar Joneidi, Mohammad Pazouki, Experimental investigation of forced
convective heat transfer coefficient in nanofluids of Al2 O3 /EG and CuO/EG in a
In this paper, the overall heat transfer coefficient of Fe2 O3 /water double pipe and plate heat exchangers under turbulent flow, Exp. Therm Fluid
Sci. 35 (2011) 495–502.
nanofluid in a compact air-cooled heat exchanger has been mea- [16] K.Y. Leong, R. Saidur, T.M.I. Mahlia, Y.H. Yau, Modeling of shell and tube heat
sured experimentally according to the LMTD technique under lam- recovery exchanger operated with nanofluid based coolants, Int. J. Heat Mass
inar flow regime. The experiments were conducted at wide ranges Transfer 55 (2012) 808–816.
of Reynolds number for nanofluid and air flow, nanoparticle vol- [17] L. Syam Sundar, M.T. Naik, K.V. Sharma, M.K. Singh, T.Ch. Siva Reddy,
Experimental investigation of forced convection heat transfer and friction
ume concentrations and at different nanofluid inlet temperatures. factor in a tube with Fe3 O4 magnetic nanofluid, Exp. Therm Fluid Sci. 37 (2012)
The following conclusions can be achieved from this study: 65–71.
[18] Shive Dayal Pandey, V.K. Nema, Experimental analysis of heat transfer and
(a) the measured overall heat transfer coefficients for the base friction factor of nanofluid as a coolant in a corrugated plate heat exchanger,
fluid (water) and nanofluids (Fe2 O3 /water) exhibit a very good Exp. Therm Fluid Sci. 38 (2012) 248–256.
[19] Yi-Hsuan Hung, Tun-Ping Teng, Tun-Chien Teng, Jyun-Hong Chen, Assessment
agreement with correlations of Sieder–Tate and Dehghandokht of heat dissipation performance for nanofluid, Appl. Therm. Eng. 32 (2012)
et al. with average error of 7% and 17% for distilled water, 132–140.
respectively and with the correlation of Xuan and Li with [20] Roghayeh Lotfi, Ali Morad Rashidi, Azadeh Amrollahi, Experimental study on
the heat transfer enhancement of MWNT-water nanofluid in a shell and tube
average error of 3% for nanofluid; heat exchanger, Int. Commun. Heat Mass Transfer 39 (2012) 108–111.
(b) increasing the nanofluid inlet temperature from 50 to 80 °C [21] M. Saeedinia, M.A. Akhavan-Behabadi, P. Razi, Thermal and rheological
decreases the overall heat transfer coefficient which is due to characteristics of CuO–Base oil nanofluid flow inside a circular tube, Int.
Commun. Heat Mass Transfer 39 (2012) 152–159.
the large increase in the logarithmic mean temperature differ- [22] S.M. Peyghambarzadeh, S.H. Hashemabadi, M. Naraki, Y. Vermahmoudi,
ence comparing with the less increase in the nanofluid tem- Experimental study of overall heat transfer coefficient in the application of
perature difference. Conversely, it is found that the increase in dilute nanofluids in the car radiator, Appl. Therm. Eng. 52 (2013) 8–16.
[23] M. Naraki, S.M. Peyghambarzadeh, S.H. Hashemabadi, Y. Vermahmoudi,
the nanofluid inlet temperature, creates the heat transfer rate
Parametric study of overall heat transfer coefficient of CuO/water nanofluids
enhancement of about 116% which is achieved at the higher in a car radiator, Int. J. Therm. Sci. 66 (2013) 82–90.
nanofluid flow rates; [24] Xiang-Qi Wang, Arun S. Mujumdar, Heat transfer characteristics of nanofluids:
(c) the experimental results indicate that the overall heat transfer a review, Int. J. Therm. Sci. 46 (2007) 1–19.
[25] X.Q. Wang, A.S. Mujumdar, A review on nanofluids—part I: theoretical and
coefficient and the heat transfer rate of nanofluid have been numerical investigations, Braz. J. Chem. Eng. 25 (4) (2008) 613–630.
improved with the enhancement in the nanofluid and air flow [26] X.Q. Wang, A.S. Mujumdar, A review on nanofluids—part II: experiments and
Reynolds numbers; applications, Braz. J. Chem. Eng. 25 (4) (2008) 631–648.
[27] S. Kakaç, A. Pramuanjaroenkij, Review of convective heat transfer enhance-
(d) heat transfer performance of the nanofluid is improved with ment with nanofluids, Int. J. Heat Mass Transfer 52 (2009) 3187–3196.
the increase in volume concentration of nanoparticles (ranging [28] Yanjiao Li, Jingén Zhou, Simon Tung, Eric Schneider, Shengqi Xi, A review on
from 0 to 0.65 vol.%). Application of nanofluid with 0.65 vol.% development of nanofluid preparation and characterization, Powder Technol.
concentration can enhance the overall heat transfer coefficient 196 (2009) 89–101.
[29] Haisheng Chen, Yulong Ding, Heat transfer and rheological behavior of
up to 13% and the heat transfer rate up to 11.5% comparing with nanofluids—a review, Adv. Transport Phenom. (2009) 135–177. ADVTRANS 1.
the distilled water. [30] L. Godson, B. Raja, D. Mohan Lal, S. Wongwises, Enhancement of heat transfer
using nanofluids-an overview, Renew. Sustainable Energy Rev. 14 (2010)
629–641.
References [31] S.M. Sohel Murshed, C.A. Nieto de Castro, M.J.V. Lourenco, M.L.M. Lopes,
F.J.V. Santos, A review of boiling and convective heat transfer with nanofluids,
[1] N. Sahiti, F. Durst, A. Dewan, Heat transfer enhancement by pin elements, Int. Renew. Sustainable Energy Rev. 15 (2011) 2342–2354.
J. Heat Mass Transfer 48 (2005) 4738–4747. [32] A. Ghadimi, R. Saidur, H.S.C. Metselaar, A review of nanofluid stability
[2] R.K. Shah, S. Kakac, A.E. Bergles, F. Mayinger (Eds.), Classification of Heat properties and characterization in stationary conditions, Int. J. Heat Mass
Exchangers, in: Thermal Hydraulic Fundamental and Design, Hemisphere/ Transfer 54 (2011) 4051–4068.
McGraw-Hill, New York, 1981, p. 9. [33] Jahar Sarkar, A critical review on convective heat transfer correlations of
[3] S.Gh. Etemad, F. Bakhtiari, General equations for fully developed fluid flow and nanofluids, Renew. Sustainable Energy Rev. 15 (2011) 3271–3277.
heat transfer characteristics in complex geometries, Int. Commun. Heat Mass [34] Gabriela Huminic, Angel Huminic, Application of nanofluids in heat exchang-
Transfer 26 (1999) 229–238. ers: a review, Renew. Sustainable Energy Rev. 16 (2012) 5625–5638.
[4] M. Barkhordari, S.Gh. Etemad, Numerical study of slip flow heat transfer of [35] D.R. Lide, Handbook of Chemistry and Physics, eightyfifth ed., CRC Press, Boca
non-newtonian fluids in circular microchannels, Int. J. Heat Fluid Flow 28 (5) Raton, FL, 2005.
(2007) 1027–1033. [36] B.C. Pak, I.Y. Cho, Hydrodynamic and heat transfer study of dispersed fluids
[5] A.E. Bergles, Recent development in convective heat transfer augmentation, with sub-micron metallic oxide particles, Exp. Heat Transfer 11 (1998)
Appl. Mech. Rev. 26 (1973) 675–682. 151–170.
[6] A.S. Ahuja, Augmentation of heat transport in laminar flow of polystyrene [37] Y. Xuan, W. Roetzel, Conceptions for heat transfer correlation of nanofluids,
suspension, I. Experimental and results, J. Appl. Phys. 46 (8) (1975) 3408–3416. Int. J. Heat Mass Transfer 43 (2000) 3701–3708.
[7] S.U.S. Choi, Developments and application of non-Newtonian flows, ASME [38] R.L. Hamilton, O.K. Crosser, Thermal conductivity of heterogeneous two
FED-V.231/MD-V.66, New York, 1995, pp. 99–105. component systems, Ind. Eng. Chem. Fundam. 1 (1962) 187–191.
[8] B. Farajollahi, S.Gh. Etemad, M. Hojjat, Heat transfer of nanofluids in a shell [39] A. Einstein, A new determination of the molecular dimensions, Ann. Physics
and tube heat exchanger, Int. J. Heat Mass Transfer 53 (2010) 12–17. 19 (2) (1906) 289–306.
 Y. Vermahmoudi et al. / European Journal of Mechanics B/Fluids 44 (2014) 32–41 41

[40] N. Saito, Concentration dependence of the viscosity of high polymer solutions, [47] R.J. Moffat, Describing the uncertainties in experimental results, Exp. Therm.
J. Phys. Soc. Japan 5 (1950) 4–8. Fluid Sci. 1 (1988) 3–17.
[41] H.C. Brinkman, The viscosity of concentrated suspensions and solution, [48] E.N. Sieder, G.E. Tate, Heat transfer and pressure drop of liquids in tubes, Ind.
J. Chem. Phys. 20 (1952) 571–581. Eng. Chem. 28 (1936) 1429–1436.
[42] T. Lundgren, Slow flow through stationary random beds and suspensions of [49] M. Dehghandokht, G. Mesbah, A. Fartaj, S. Sanaye, Flow and heat transfer
spheres, J. Fluid Mech. 51 (1972) 273–299. characteristics of water and ethylene glycol-water in a multi-port serpentine
[43] G.K. Batchelor, The effect of Brownian motion on the bulk stress in a meso-channel heat exchanger, Int. J. Therm. Sci. 50 (8) (2011) 1615–1627.
suspension of spherical particles, J. Fluid Mech. 83 (1) (1977) 97–117. [50] Y. Yang, Z.G. Zhang, E.A. Grulke, W.B. Anderson, G. Wu, Heat transfer properties
[44] D.A. Drew, S.L. Passman, Theory of Multi Component Fluids, Springer, Berlin, of nanoparticle-in-fluid dispersions (nanofluids) in laminar flow, Int. J. Heat
1999. Mass Transfer 48 (2005) 1107–1116.
[45] F.P. Incropera, D.P. DeWitt, Introduction to Heat Transfer, third ed., John Wiley [51] C. Jwo, L. Jeng, T. Teng, C. Chen, Performance of overall heat transfer in
& Sons, Inc., New York, 1996. multichannel heat exchanger by alumina nanofluid, J. Alloys Compd. 504S
[46] S. Vithayasai, T. Kiatsiriroat, A. Nuntaphan, Effect of electric field on heat (2010) 385–388.
transfer performance of automobile radiator at low frontal air velocity, Appl. [52] Y. Xuan, Q. Li, Investigation on convective heat transfer and flow features of
Therm. Eng. 26 (2006) 2073–2078. nanofluids, J. Heat Transfer 125 (2003) 151–155.