Renewable and Sustainable Energy Reviews 54 (2016) 12121239

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

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

Numerical study of convective heat transfer of nanouids: A review

Sh.M. Vanaki a, P. Ganesan a,n, H.A. Mohammed b,nn
a
Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia
b
Department of Thermouids, Faculty of Mechanical Engineering, UniversitiTeknologi Malaysia, UTM Skudai, 81310 Johor Bahru, Malaysia

art ic l e i nf o a b s t r a c t

Article history: The recent development of nanotechnology led to the concept of using suspended nanoparticles in heat
Received 18 November 2014 transfer uids to improve the heat transfer coefcient of the base uids. Specically, numerical studies
Received in revised form are reviewed in this study to get a clear view and detailed summary of the inuence of several para-
24 August 2015
meters such as type of nanoparticle and host liquid, particle volume concentration, particle size, particle
Accepted 21 October 2015
shape, Brownian diffusion and thermophoresis effect on hydrodynamic and thermal characteristics of
Available online 11 November 2015
convective heat transfer using nanouids. In addition, the paper provides detailed information about the
Keywords: most of commonly-used correlations which are utilized to predict the effective thermophysical prop-
Convective heat transfer erties of nanouids. Finally, the main aim upon which the present work is based is to give a compre-
Nanouids
hensive review on different CFD approaches employed in numerical simulation of nanouid ow, address
CFD
the pros and cons of each approach, and nd the suitable technique which gives more credible results as
Thermophysical properties
Single-phase modeling compared to experimental results.
Two-phase modeling & 2015 Published by Elsevier Ltd.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1213
1.1. The rise of compact thermal devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1213
1.2. Nanouids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1213
1.3. Objectives of the review and the focus of the paper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1213
2. Effective parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1214
2.1. Type of nanoparticle and host liquid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1214
2.2. Nanoparticle volume concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1214
2.3. Nanoparticle diameter and shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1215
2.4. Brownian diffusion and thermophoresis effects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1215
3. Thermophysical properties of nanouids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1216
4. Numerical analysis of convective heat transfer of nanouid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1221
4.1. Single-phase approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1221
4.1.1. Conventional single-phase approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1221
4.1.2. Thermal dispersion single-phase approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1222
4.2. Two-phase approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1223
4.2.1. Eulerian-Eulerian model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1223
4.2.2. LagrangianEulerian model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1232
4.3. Other CFD approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1233
4.3.1. Lattice Boltzmann Method (LBM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1233
4.3.2. Non-homogeneous two-component model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1233
5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1234

n
Corresponding author. Tel.: 6 3 7967 5204x7670; fax: 6 3 7967 5317.
nn
Corresponding author.
E-mail addresses: shayan.mv@gmail.com (Sh.M. Vanaki), poo_ganesan@um.edu.my (P. Ganesan), Hussein.dash@yahoo.com (H.A. Mohammed).

http://dx.doi.org/10.1016/j.rser.2015.10.042
1364-0321/& 2015 Published by Elsevier Ltd.
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
5.1. Prospectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1234
Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1234
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1234
1. Introduction
Over the last few years, energy consumption has increased
drastically; therefore, scientists have developed new energy-
saving strategies to overcome the threat of energy shortage [1].
Heat exchangers are widely used in industrial systems in the eld
of energy conservation, conversion and recovery. Thus, it is
obvious that providing more efcient heat exchanging systems can
mitigate the energy concerns considerably.
1.1. The rise of compact thermal devices
Better heat removal can be achieved by decreasing the area to
volume ratio of thermal devices which is one of the most important
factors in thermal design [2]. This idea motivated Tuckerman and
Pease [3] to propose microchannel heat sink (MCHS) about two
decades ago. After that, many researches [46] proved the effec-
tiveness of using micro-channel heat exchangers for thermal
enhancement purposes. Lately, due to the advancement of micro
fabrication technology, microchannels and microtubes are manu-
factured and utilized in industries such as microelectronics, aero-
space, biomedical, robotics, telecommunications and automotive [7].
The main reasons for development of miniaturized and light
weight heat exchangers are given as follows [8,9]:
 Space and size limitations.
 Energy and material savings.
 Ease of unit handling.
 Growing need for heat transfer augmentation with increasing
energy demands.
 Cooling requirement of microscale and microelectronic devices.
Several studies [1014] have proposed various methods to
improve the thermalhydraulic performance and reduce unit sizes
of conventional heat exchangers. However, in order to have more
efcient and cost-effective heat exchangers, several approaches
have been investigated over the years. Passive technique, employing
ribs or grooves on the inner surface of heat exchangers, has been
one of the frequent approaches to break the laminar sub-layer and
create local wall turbulence due to ow separation and reattach-
ment between successive corrugations, which reduces the thermal
resistance and signicantly enhances heat transfer. Although many
studies available in the open literature have proven the signicant
effect of geometry modication on heat transfer enhancement in
recent years, this technique has already reached its limit. Presently
available heat transfer working uid such as water, engine oil (EO),
ethylene glycol (EG), propylene glycol (PG) are widely used in many
industrial elds like heating or cooling processes, chemical pro-
duction, microelectronics, power generation, air-conditioning, and
transportation. These conventional heat transfer uids have inher-
ently low thermophysical properties as compared with solids.
Hence, there is a pronounced need to develop alternative heat
transfer uids with higher cooling capability which can improve the
compactness and effectiveness of heat exchangers.
1.2. Nanouids
As discussed in the previous section the issue of heat removal
from small scale devices which can generate a high amount of heat
1213
ux has attracted substantial attention from researchers. To ensure
that the heat exchangers deliver their best performance, working
uids with greater thermal conductivity are required. Since solid
particles have larger thermal conductivity than conventional uids
[15], dispersing these particles into the base uid is anticipated to
increase the thermal conductivity of the whole mixture. Therefore,
in order to overcome the limited heat transfer capabilities of
common uids, Maxwell rst proposed a theoretical work which
showed the possibility of enhancing the thermal conductivity of
liquids by mixing micron-sized solid particles [16]. Later, Ahuja
[17] experimentally analyzed the suspension containing aqueous
sodium chloride and glycerin (as the base uid) and micrometer-
sized polystyrene particles. He observed thermal conductivity
enhancement three times greater compared to the base uid.
However, problems such as rapid sedimentation caused by these
particles have kept this concept far from practical use. Recent
advances in nanotechnology resulted in appearance of a new
generation of uid called nanouid (colloidal mixture of nano-
particles smaller than 100 nm), a term proposed by Choi and
Eastman [18]. Nanouids have unique thermal transport proper-
ties and superior performance that are unavailable in traditional
heat transfer uids. Compared to conventional solidliquid sus-
pensions for enhancing heat transfer, nanouids show a higher
potential for increasing the heat transfer rates in heat exchangers
for the following reasons [19]:
 High heat transfer surface between particles and uids and
therefore high effective thermal conductivity.
 High dispersion stability with predominant Brownian motion of
particles.
 Reduced pumping power as compared to pure liquid to achieve
equivalent heat transfer intensication.
 Less particle clogging as compared to convention slurries, thus
suitable for use in microsystems.
 Adjustable properties, including thermal conductivity and sur-
face wettability, by varying particle concentrations to suit dif-
ferent applications.
Saidur et al. [20] conducted a comprehensive review of the
applications of nanouids and suggested specic areas in which
nanouids can be used as follows:
Heat exchanger [21], cooling of electronics [22,23], solar water
heating [24], nuclear reactor cooling [2527], refrigeration (domestic
refrigerator [28,29], chillers [30,31]), engine cooling/vehicle thermal
management [3234], cooling and heating in buildings [35], cooling
of diesel electric generator [36], diesel combustion [37,38], detection
of knock occurrence in a gas SI engine [39], biomedical applications
[4043], fuel cell [4446], new sensors for improving exploration,
boiler ue gas temperature reduction, nanouids in cameras, micro
devices, and displays, application as a coolant in machining [47],
cooling welding equipment, drilling speed [48,49], transformer cool-
ing [5055], space, defense and ships.
1.3. Objectives of the review and the focus of the paper
In recent years, a noticeable number of review papers have
been provided on the convection heat transfer and uid ow
characteristics of nanouid. These papers cover the following
areas: natural convection [56], forced convection [57], entropy
1214 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
generation [58], experimental investigations [59], convective heat
transfer and friction factor correlations [15], backward facing step
[60], microchannels [7], vortex generators [61], application of
lattice Boltzmann method [62,63], annular passages [64], helically
coiled tube heat exchanger [65], direct solar absorption systems
[66], heat pipes [67,68], porous media [69], enclosures [70] and
passive methods [71]. So it is evident that there is no in-depth
study available to investigate the numerical aspects of nanouid
behavior in different types of heat exchangers.
All the numerical studies of nanouids are based on some
hypotheses which make the simulation process easier and as a
result could lead to some errors. Thus the most challenging issue
for researchers had been to adopt a reliable numerical method
which has less deviation from experimental results. The aims of
the present review paper are: (1) to focus on the numerical studies
of heat transfer enhancement and uid ow behavior concerning
the effective parameters of nanouids, (2) to summarize the most
commonly-used correlations in order to calculate nanouid ther-
mophysical properties, (3) to conduct a comprehensive compar-
ison between single-phase and two-phase nanouid ows to nd
the suitable CFD approach which gives credible results as com-
pared to experimental results, (4) to provide information about the
ow property, numerical method, software code, and adopted
mesh throughout numerical studies on nanouids, and (5) to
present research gaps which need to be considered for future
research work in the area.
2. Effective parameters
To evaluate thermophysical properties of nanouids and their
thermalhydraulic behaviors numerically, the inuences of many
parameters such as type of nanoparticle and host liquid, nano-
particle volume concentration, nanoparticle size and its shape, and
particle migration should be taken into account.
2.1. Type of nanoparticle and host liquid
As discussed before, nanouids have a notable impact on the
increase of heat transfer. The main reason is the existence of highly
conductive dispersed nanoparticle in the host liquid, such asoxide
ceramics (Al2O3, CuO), nitride ceramics (AlN, SiN), carbide cera-
mics (SiC, TiC), metals (Cu, Ag, Au), semiconductors (TiO2, SiC),
carbon nanotubes, and composite materials, which in turn,
improve the heat transfer properties of nanouids. Oztop et al.
[72] carried out a numerical study to investigate the effects of
different types of nanoparticles (Al2O3 and TiO2) on uid ow and
heat transfer in an enclosure. It was seen that the addition of
nanoparticles into water affects the uid ow and temperature
distribution, particularly for higher Rayleigh numbers. Moreover, a
heat transfer augmentation was observed over the whole range of
Rayleigh numbers. A numerical investigation on natural convec-
tion cooling of an oscillating heat source tted on the left wall of
an enclosure was performed by Ghasemi and Aminossadati [73].
Cu, Al2O3 and TiO2 nanoparticles dispersed in water were used as
working uids. It was concluded that the inclusion of dispersed
nanoparticles into the base uid improves its thermal conductivity
and enhances the heat removal from the heat source. The most
effective nanoparticle was reported for Cu in the heat removal
process, followed by Al2O3 and TiO2 respectively. Kalteh et al. [74]
studied the steady laminar mixed convection of water-based
nanouid ow in a lid-driven square cavity with a triangular
heat source. It was found that suspending the nanoparticles in
pure uid leads to a signicant heat transfer increase. Although
nanouids have shown noticeable thermal performance, it should
be noted that the addition of nanoparticles increases the density
and viscosity of nanouid and consequently results in pressure
drop penalty.
On the other hand, some researchers have highlighted the role
of using other kinds of base uids instead of water. For example,
Ebrahimnia-Bajestan et al. [75] investigated numerically the heat
transfer performance and pressure drop of using various nano-
particles including Al2O3, CuO, carbon nanotube (CNT) and titanate
nanotube (TNT) dispersed in water and ethylene glycol/water with
particle concentrations of 06 vol% through a straight circular pipe
in a laminar ow regime and constant heat ux boundary condi-
tion. The results indicated that the heat transfer characteristics of
nanouids are strongly inuenced by the type of base uid.
In another paper, Kuppusamy et al. [76] conducted a numerical
study to examine the thermal and ow elds in a trapezoidal
grooved microchannel heat sink (TGMCHS) using nanouids. It
was inferred that the value of Nunf =NuH2 O for Al2O3/H2O mixture
was much higher compared to Al2O3/engine oil and Al2O3/ethy-
lene glycol. It is because the thermal conductivity of water is much
higher compared to ethylene glycol and engine oil. In contrast, f nf
=f H2 O of Al2O3/ethylene glycol and Al2O3 /engine oil are increased
extremely due to their higher viscosity and density compared to
Al2O3/H2O. Interestingly, the new concept of combined/hybrid
nanouids, (Al2O3 CNTs)-water, was introduced by NuimLabib et
al [77]. They adopted two-phase mixture model to study laminar
forced convection ow inside a uniformly heated tube. According
to the results, combining the CNTs and Al2O3 into water base uid
enhances the convective heat transfer performance signicantly.
In addition, the effect of barely-used Al2O3 nanoparticles sus-
pended in Ethylene Glycol and water as the base uids was
reported throughout this investigation. As a result, the heat
transfer coefcient enhancement appears to be more pronounced
for using Ethylene Glycol as a base uid than water.
2.2. Nanoparticle volume concentration
Volume concentration (or volume fraction) can be accounted
for as one of the most important parameters of nanouids. It is
dened as the ratio of the volume of nanoparticles to the base uid
volume. As discussed previously, the solid particles have higher
thermal conductivity than liquids, therefore as the percentage of
dispersed nanoparticles increases, the thermal conductivity of the
whole mixture increases as well. Several studies conrmed that
nanouid heat transfer coefcients could be improved by more
than 20% in the case of low nanoparticle concentrations ( o5%)
[7880].The increasing trend in forced convective heat transfer of
nanouids with increase of nanoparticle volume concentration has
been widely expressed in most experimental and numerical
works. Conversely, there is an inconsistency for the natural con-
vection heat transfer of nanouids.
Khanafer et al. [81], Lai and Yang [82], Fattahi et al. [83]
numerically investigated the natural convection of nanouid in a
2-D rectangular cavity and Abu-Nada et al. [84] did in horizontal
annuli. They showed that the heat transfer coefcient increases
with increasing nanoparticle volume fraction. In contrast to this
observation, Putra et al. [85], Wen and Ding [86, 87], Chang et al.
[88], Nnanna [89] and Ho et al. [90] experimentally declared that
there is no enhancement in natural convection heat transfer by
using nanouids.
The strongest reason for these contradictions among the
experiments and simulations is the assumption of nanouids as a
homogeneous mixture of nanoparticles in the host uid which
ignores any slip velocity between the base uid and nanoparticles
in CFD modeling [8184].
Pakravan and Yaghoubi [91] used a two-phase mixture model
to include the slip velocities for natural convection of nanouid in
a square cavity and numerically solved the governing equations.
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
They observed the decreasing behavior of Nusselt number with
increment of nanoparticle size and volume fraction. Haddad et al.
[92] used a two component non-homogenous equilibrium model
(Buongiorno's model [93]) to study the natural convection heat
transfer and uid ow of CuO/water nanouids using Rayleigh-
Bnard problem. They showed that increasing nanoparticle con-
centration has an adverse effect on heat transfer. However, when
Garoosi et al. [94] accomplished a numerical study concerning the
natural and mixed convection heat transfer of nanouid (Al2O3/
water) in a laterally-heated square cavity and applied the Buon-
giorno's model, it was illustrated that there is an optimal volume
fraction of the nanoparticles at each Rayleigh and Richardson
number for which the maximum heat transfer rate can be
obtained.
Nevertheless, Santra et al. [95] tried to incorporate the inuence
of rheology of nanouids by considering them as non-Newtonian
uids for laminar natural convection in a differentially heated
square cavity. Results showed that heat transfer is inversely pro-
portional to concentration of nanoparticles. However, Abouali and
Falahatpisheh [96] found the improper selection of correlation for
viscosity of nanouids as a reason of above-mentioned incon-
sistency for natural convection heat transfer.
It should be noted that density and viscosity are the two
properties of nanouids which are proportional to nanoparticle
volume fraction. Thus, larger amount of particle volume fraction
can inuence the pressure drop remarkably and restricts their
applications for thermal systems working under limited pumping
power condition.
2.3. Nanoparticle diameter and shape
Particle size is another effective parameter which plays an
important role in the hydraulic and thermal behaviors of nanouids.
Theoretical evidence [9799] indicates that the effective thermal
conductivity of nanouids increases with decreasing particle size.
Dehnavi and Rezvani [100] investigated numerically the natural
convection heat transfer of Al2O3/water nanouid in a shaped
enclosure operating under differentially heated walls. They used
nite volume method (FVM) for discretizing positional expressions,
and the fourth order Rung-Kuta was used for discretizing time
expressions. FVM is the spatial integration of the conservation
equation over nite control volumes to yield a set of algebraic
equations [101]. It was observed that by decreasing the mean dia-
meters of nanoparticles from 200 to 5 nm, the heat transfer increases
21% and 17% for Gr 103 and Gr 105, respectively.
Seyf and Nikaaein [102] modeled the uid ow and heat transfer
characteristic of three different nanouids in microchannel heat
sinks using the DarcyBrinkmanForchheimer and two-equation
model, respectively. It was found that with decreasing nanoparticle
size the effect of both nanolayer formation and Brownian motion
increase which lead to augmentation of thermal conductivity and
consequently heat transfer coefcient. The heat transfer enhance-
ment of employing nanouids ows in MCHS was studied by Hung
et al. [103]. According to the results, no signicant effect on cooling
performance of the MCHS was seen for nanoparticle size above
23 nm. This behavior is referred to Brownian motion which may be
inhibited for larger particles [104]. However, the overall thermal
resistance decreased with decreasing particle size. Furthermore, in
some numerical works [105108], the independency of friction
factor to the variation of particle size has been concluded.
Mirmasoumi and Behzadmehr [109] examined the effect of
nanoparticle mean diameter on mixed convection heat transfer of
Al2O3/water nanouid in a horizontal tube by means of mixture
model. The results showed that the convective heat transfer
coefcient can be signicantly increased with decreasing the
nanoparticle mean diameter while the skin friction coefcient
1215
does not change notably. The same ow regime and nanouid
were considered by Shariatet al. [110] in an elliptic duct with
uniform heat ux boundary condition on the walls to assess the
impact of nanoparticle mean diameter and the buoyancy force on
thermal and ow behaviors. It was demonstrated that there are
non-linear relations between the size of the nanoparticles and the
nanouid ow behaviors. Akbarinia and Laur [111] carried out a
three-dimensional numerical investigation using elliptic govern-
ing equations to analyze the role of solid particles in order of micro
and nano-sized diameters on laminar mixed convection of a
nanouid consisting of water and 1 vol% Al2O3. Two phase mixture
model and control volume technique were adopted to examine the
ow pattern throughout a circular curved tube. It was shown that
for a given solid volume fraction, increasing the diameter of par-
ticles boosts axial velocity and skin friction while the Nusselt
number, secondary ow as well as the temperature reduces. It was
also concluded that for nanometer-sized particles, increasing the
diameter of particles does not change the ow behaviors.
Although the most commonly-used geometric shape of the
nanoparticles is spherical, most researchers have reported the
considerable inuence of other kinds of particle shape such as
cylindrical (ber, rod, and tube), blade, platelet, brick, oblate and
prolate spheroidal. Xie et al. [112],and Murshed et al. [113]
explained that cylindrical nanoparticles are better in improving
the thermal conductivity of nanouids due to the elongated par-
ticles and the large aspect ratio of the surface area to volume that
conducts heat through the uid. Timofeeva et al. [114] reported
that the viscosity of nanouid is strongly dependent on the par-
ticle shape and they underlined higher viscosity with elongated
particles like platelet and cylinder compared to spherical-shaped
particles.
Vanaki et al. [115] performed a numerical investigation to study
the effects of different nanouids on the thermal and ow elds
through transversely wavy wall channels with different phase shifts
between the upper and lower wavy walls. Reynolds numbers were
considered in the turbulent range of 6000rRer18,000 and a
uniform temperature of 400 K was applied on the walls. It was
revealed that SiO2-EG nanouid with platelet particle shape gives
the highest Nusselt number followed by nanouids containing
cylindrical, brick, blade, and spherical nanoparticle shapes, respec-
tively. Ooi and Popov [116] simulated the natural convection ow of
the Cu/water nanouid inside a square cavity by using the radial
basis integral equation (RBIE) method. The effects of spherical and
spheroidal nanoparticles on uid ow were examined and the
numerical results indicated that the oblate spheroid with aspect
ratio of 10 produces the largest enhancement of the overall heat
transfer characteristic.
2.4. Brownian diffusion and thermophoresis effects
In order to improve our understanding of convective heat
transfer of nanouids, it is essential to understand the hydro-
dynamics of nanoparticles. Some experimental studies [117,118]
reported an anomalous heat transfer enhancement in laminar
regime. They declared that particle migration might be a reason
for the heat transfer enhancement. The effect of particle migration
on convective heat transfer of nanouids in laminar regime has
been studied in a number of works. According to the theoretical
studies ([119,120]) performed throughout a pipe, a high non-
uniformity in the particle concentration over the pipes cross
section was reported when they considered particle migration due
to the mechanisms of viscosity gradient, shear rate, and Brownian
motion (random movement of particles).
Thermophoresis (the Soreteffect), nanoparticle transport driven
by temperature gradient, is another important mechanism which
shows a substantial inuence on particle migration. Using scale
1216 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
analysis, Bahiraei and Hosseinalipour [121] proved that the parti-
cle uxes associated with Brownian diffusion and thermophoresis
have orders of 10  10 and 10  8, respectively. Hence, it can be
concluded that the thermophoresis can affect the particle migra-
tion even more signicantly in comparison with the Brownian
diffusion and must not be neglected. They also observed closer
results between the simulation and experimental values once they
took thermophoresis into account and used dispersion approach
for simulation.
Generally the total ux of particle migration contains four
terms [120,122]:
2
! have applied the Brownian motion effect of nanoparticles, which
dp d d
J  K _ 2 1
d dr
 
2 d_ d 3. Thermophysical properties of nanouids
J C  K C dp 2 _
dr dr
d by the convective heat transfer coefcient which is a function of a
J B  DB
dr
dT=dr
J T  DT
T
where J , J C , J B and J T denote the particle uxes occurred as a result
of viscosity gradient, non-uniform shear rate, Brownian motion
and thermophoresis, respectively. Moreover K and K C are con-
stants, _ represents the shear rate, denotes concentration, is
dynamic viscosity, dp represents particle diameter, DB gives the
Brownian diffusion coefcient and DT is thermophoretic diffusion
coefcient.
To nd concentration distribution inside a pipe, the following
equation should be solved:
J J JC JB JT 0 5 coefcients of nanouids are assumed to be a linear function of
Although Sohn and Kihm [123] conrmed the effective role of
thermophoresis and Brownian diffusion as the most important
mechanisms of particle migration, they showed that the effects of
viscosity gradient and non-uniform shear rate are not signicant;
therefore they can be overlooked in Eq. (5).
Signicantly, thermophoresis makes the concentration dis-
tribution more non-uniform and decreases the concentration near
the walls which would change the spatial effective thermo-physical
properties of the nanouids especially the thermal conductivity and
viscosity as they are strong functions of particle concentrations. A
reduced effective viscosity close to the wall could result in the
thinning of the boundary layer and consequently, increased nano-
uid velocity near the walls. In addition, the migration of nano-
particles could affect the development of the boundary layer so as
to affect the ow and convective heat transfer coefcient of
nanouids.
The theoretical models employed in [119121,123] do not take
convective transport of nanoparticles into account. At that time,
Buongiorno [93] introduced seven slip mechanisms and explained
that the Brownian motion and thermophoresis are the dominant
slip mechanisms in laminar forced convection of nanouids and
eventually he developed a four equation, non-homogeneous equi-
librium model. This model was successfully tested in [124] to
examine the effect of particle migration on ow and heat transfer of
nanouids owing through a circular pipe. It was argued that
thermophoresis reduces nanoparticle concentration only in the wall
boundary layer. However, at the tube center, the velocity prole is
attened in order to keep the mass ow rate constant. Meanwhile,
considering non-homogeneous model resulted in a greater heat
transfer coefcient in comparison with homogeneous model pre-
diction. Ryzhkov and Minakov [125] attained similar results in their
numerical study.
Haddad et al. [92] studied Rayleigh-Bnard problem using CuO/
water nanouid for different Rayleigh numbers and different
nanoparticle volume fractions. They considered the role of Brow-
nian and thermophoresis effects and compared to the case where
both effects are neglected. The results revealed that higher heat
transfer is formed with the presence of Brownian and thermo-
phoresis effects, while by neglecting the role of thermophoresis
and Brownian motion, deterioration in heat transfer is observed.
More discussion of the Buongiorno [93] mathematical model is
provided in section 4.3.2. It must be noted that a small number of
proposed correlations for thermophysical properties of nanouids
are discussed in detail in Section 3.
2
The effectiveness of heat transfer using nanouids is described
3
number of thermophysical properties, namely thermal con-
ductivity, specic heat, thermal expansion coefcient, viscosity
4 and density.
Many researchers have used solidliquid mixture equations for
estimating the density and specic heat capacity [104,113,126
133] of nanouids from the following equations:
nf 1  bf np 6
     
C p nf 1  C p bf C p np 7
nf 1 bf np 8
It is noted that the density, specic heat and thermal expansion
volume fraction due to lack of experimental data on their tem-
perature dependence.
The two important transport properties of nanouids, dynamic
viscosity and thermal conductivity, are not only dependent on
volume concentration of nanoparticles, but are also strongly
dependent on other parameters such as particle size, mixture
combinations and slip mechanisms, surfactant, etc. Numerous
experimental studies [134140] proved that both thermal con-
ductivity and viscosity increase by dispersing nanoparticles into a
base uid.
Thus far, various theoretical and experimental investigations
have been carried out and various correlations have been intro-
duced for estimation of dynamic viscosity and thermal con-
ductivity of nanouids. Yet, due to lack of common understanding
on mechanism of nanouid, no general empirical correlation and
theoretical model concerning the effect of particle concentration,
size and temperature has been proposed recently.
To predict the effective thermal conductivity of nanouids in
numerical modeling, Maxwell model [16], which is restricted to
spherical particles, has been broadly cited in the open literature by
researchers [141148]. Other analogous versions of Maxwell model,
Hamilton and Crosser (HC) [149] and Wasp [150] models, were then
applied into the calculation of the convective heat transfer coef-
cient in many papers published numerically such as [84,151154]
and [81,155,156], respectively. These traditional models failed to
predict the effective thermal conductivity of nanouids accurately,
likely because such models include only the effect of the nano-
particle concentration, and basically they have been developed for
composites and mixtures with micro and milli-sized inclusions
[157]. Therefore, investigators have been motivated to develop a
number of new theoretical models for the evaluation of the effec-
tive thermal conductivity of nanouids that take into consideration
the Brownian motion of the nanoparticles, which is the case of the
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
models proposed by Yu and Choi [98], Koo and Kleinstreuer [158],
Prasher et al. [159], Patel et al. [160] and Vajjha et al. [161,162],
whose models were abundantly utilized in numerical studies like
[163177,105,115, 178181], respectively.
Most of the authors [182184], who have recently proposed
empirical correlations to predict the thermal conductivity of
nanouids, have developed their correlations from their own
experimental data. However, Hassani et al. [185] developed a new
correlation for predicting the thermal conductivity of nanouids
with large varieties of nanoparticles and base uid, and the pre-
dicted values had been found in excellent agreement with the
experimental ones with a mean deviation of 3%. In addition, this
developed model is able to take into account the Brownian
motion, the variation of volume fraction, the temperature and the
size distribution of nanoparticles. Therefore, Hassani et al. [185]
correlation can be used as a general reference to predict the
thermal conductivity for a large variety of nanouids.
It should be noted that Hamilton and Crosser [149] and Timo-
feeva et al. [114] involved the particle shape effect in their models,
providing the availability of using nonspherical particles in
numerical analysis of nanouids, e.g.in [115,116].
Minea [186] studied numerically the effect of using three dif-
ferent models of thermal conductivity correlation on the forced
convection of a nanouid that consists of water and Al2O3in hor-
izontal tubes. By comparing Maxwell [16], Yu and Choi [98] and
Maiga et al. [187] models,it was observed that [98] had the
greatest increase in values for heat transfer coefcient compared
with the two other models that offer similar results.
In the numerical study of heat transfer enhancement in hor-
izontal annuli, Abu-Nada [188], derived a correlation for nano-
uids viscosity as a function of temperature and nanoparticle
concentration based on the experimental data reported by Nguyen
et al. [137], which showed that viscosity drops sharply with tem-
perature specically for high concentration of nanoparticles. This
correlation was considered in Ref. [189,190]. He also reported a
small difference in Nusselt number by implementing Maxwell [16]
and Chon et al. [191] conductivity models, while the difference
became larger as the nanoparticles volume fraction increased.
Hwang et al. [192] theoretically examined the buoyancy-driven
heat transfer of water-based nanouid containing alumina in a
rectangular cavity. They used Jang and Choi [99] model in which
thermal conductivity of nanouids involves four modes of energy
transport, as follows: thermal diffusion of base uid; thermal
diffusion in nanoparticles; collision between nanoparticles; and
nanoconvection due to Brownian motion. For effective viscosity,
Einsteins [193] model and Pak and Chos [126] correlation were
compared with one another. Their results showed that as the
volume fraction increases the ratio of heat transfer coefcient
derived with Einsteins model is increased but the ratio derived
with Pak and Chos correlation is decreased. Once again, a rec-
tangular cavity was the case study of heat transfer enhancement
using nanouid in [194], but this time with different boundary
conditions and thermophysical models. Considering Maxwell [16]
as well asYu and Choi [98] for the calculation of effective thermal
conductivity k, and Brinkman [195], Maiga et al. [187] and Pak and
Cho [126] for effective dynamic viscosity , revealed that the
convective heat transfer performance is considerably affected by
the choice of the model used to predict the effective viscosity of
the nanouid. However, the model used to predict the thermal
conductivity seems to be less signicant.
Lai and Yang [82] performed a numerical study to simulate
natural convection of alumina nanouids in a vertical square
enclosure using three categories of mathematical models for pre-
dicting k and . The rst model involved the experimental cor-
relations given in [137,196]. Two other adopted conductivity
models and two viscosity predictions were, namely, Chon et al.
1217
[191] as well as Hamilton and Crosser [98] for k, and Masoumi
et al. [197] as well as Brinkman [195] for . It was reported that
there is a great deviationof computed Nusselt numbers using dif-
ferent models associated with the physical properties of a
nanouid.
Arefmanesh and Mahmoodi [198] investigated the effects of
uncertainties on the effective dynamic viscosity of Al2O3/water
nanouid in the laminar mixed convection uid ow and heat
transfer in a square cavity. For low Richardson numbers, the heat
transfer enhancement inside the cavity using proposed correlation
by Maiga et al. [187] for the viscosity of the nanouid was greater
compared to that obtained employing the Brinkman formula [195],
while the heat transfer enhancement trend was opposite for high
Richardson numbers.
Brinkman [195] and Pak and Cho [126] viscosity models were
considered by Nasrin et al. [145] to investigate the numerical
modeling of steady laminar combined convection ow in a vertical
triangular wavy enclosure lled with CuO/water nanouid. The
value of average Nusselt number using the Pak and Chos corre-
lation was higher than the Brinkman model due to the variation of
nanoparticles volume fraction. Similar models were employed in
the study done by Chamkha and Abu-Nada [199] for steady
laminar mixed convection ow in single and double-lid square
cavities lled with an alumina nanouid and the result was the
same as the one found in Ref. [145].
It is noticed in the literature that the majority of numerical
studies on convective heat transfer of nanouids have chosen the
theories developed many years ago by Einstein [193], Brinkman
[195], and Batchelor [200]. However, these theories have been
developed for traditional colloid dispersions and they normally
underestimate the dynamic viscosity of nanouid. Corcione [157]
proposed two empirical correlations for predicting the effective
thermal conductivity and dynamic viscosity of nanouids, tted
with a high number of experimental data available in the litera-
ture. The wide range of validity of correlations in terms of nano-
particle diameter, volume fraction and temperature, made them
useful for numerical simulation purposes. Therefore, an increased
number of authors began to employ Corciones correlations in
their CFD works [180,181,201206].
Most recently, more new empirical sets of correlations have
been proposed by Esfe and Saedodin [207], Esfe et al. [182,208],
Sharma et al. [183] and Sundar et al. [184] which may be used for
calculating in the future works.
Thermal and physical features of nanouids are inuenced by
temperature. In this regard, there exist limited correlations con-
cerning the temperature effect on the thermal conductivity and
dynamic viscosity of nanouids. Regarding the modeling of
nanouid effective thermal conductivity, those theories proposed
above by Koo and Kleinstreuer [158], Chon et al. [191], Patel et al.
[160], Vajjha et al. [161,162], Corcione [157], Esfe and Saedodin
[207], and Sharma et al. [183] are temperature dependent.
For the effective dynamic viscosity of nanouids, it was already
discussed that the Abu-Nadas [188] correlation is dependent on
temperature. Moreover, the rheological behavior of nanouids are
described by Koo and Kleinstreuer [209] and Masoumi et al. [197]
correlations which can be found in numerically conducted investi-
gations by referring to [170,210212] and [82,213215], respectively.
It is worth mentioning that most of the thermophysical models
and correlations proposed for effective thermal conductivity and
viscosity have included empirical constants of proportionality
whose values have been often determined on the basis of a limited
number of experimental data, or have not been clearly dened.
A summary of effective thermal conductivity and dynamic
viscosity models for nanouids which have been employed in
numerical studies are listed in Tables 1 and 2, respectively.
 1218
Table 1
Summary of effective thermal conductivity models for nanouids.

Year Model Equation Remarks Numerical studies

 
1873 Maxwell [16] knp 2kbf  2kbf  knp Depends on the thermal conductivities of both phases and [141148,186,
keff =kbf
knp 2kbf kbf  knp volume fraction of solid 188,194]
 
1962 Hamilton and Crosser (HC) [149] knp n  1kbf  n  1kbf  knp Valid for both the spherical and cylindrical particles and [82,84,116,151154]
keff =kbf
knp n  1kbf kbf  knp n 3=
Where is the particle sphericity
 

Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
1979 Wasp [150] knp 2kbf  2kbf  knp Hamilton and Crossers model with n 3 [81,155,156]
keff =kbf
knp 2kbf kbf  knp
 
2003 Yu and Choi [98] knp 2kbf  2kbf  knp 1 3 Considered interfacial layer [163166, 186, 194]
keff =kbf
knp 2kbf kbf  knp 1 3 Maxwell modied model
is the ratio of the nanolayer thickness to the original par-
ticle radius

2004 Jang and Choi [99] keff =kbf 1 cdnp

bf d
kf Re2dnp Pr Based on convection and conduction heat transport [192]
(Cu/Diamond)/water nanouids
  q
2004 Koo and Kleinstreuer [158] knp 2kbf  2kbf  knp Taking the effects of particle size, particle volume fraction and [167170]
keff =kbf 5  104 bf C P;bf K BdTnp f T;
knp 2kbf kbf  knp np temperature dependence as well as properties of base liquid

f T;  134:63 1722:3 0:4705  6:04 TT0 and particle phase into consideration by considering Brownian
( motion
0:0137100  0:8229 ; o 0:01 CuO/water nanouids

0:0011100  0:7272 ; 4 0:01 1% r r 4%
300K o T o 325K

2005 Chon et al. [191] dbf 0:3690 kbf 0:7476 Considering the role of temperature and particle size [82,188]
keff =kbf 1 64.70:7640 dnp knp
Pr 0:9955
T Re1:2321
Al2O3/water nanouids
294K o T o 344

2005 Maiga et al. [187] keff =kbf 1 2:72 4:972 [186]

 
2005 Prasher et al. [159] knp 2kbf  2kbf  knp Included the effect of convection near the particle and inter- [171,172]
keff =kbf 1 ARem Pr 0:333
knp 2kbf kbf  knp facial resistance
A and m are constants
Nanospheres
 
2006 Patel et al. [160] knp dbf 2K B Tdnp Included the effect of nanoparticle Brownian motion, diameter, [173177]
keff =kbf 1 k dnp 1 
1 c 2
bf bf bf dnp volume concentration and nanouid temperature
1% r r 8%
293K o T o 323K
c 25; 000

2009 Timofeeva et al. [114] keff =kbf 1 C shape C surface 1 Ck Nonspherical nanoparticles [115]
k k
8
>
> 2:61; Platelet
>
>
< 2:74; Blade
Ck
>
> 3:95; Cylinder
>
>
: 3:37; Brick
   q
2009-2010 Vajjha et al. [161,162] knp 2kbf  2kbf  knp Included the effect of nanoparticle Brownian motion, diameter, [105,115,178181]
keff k 5  104 bf C P;bf K BdTnp f T;
knp 2kbf kbf  knp bf np volume concentration and nanouid temperature

f T; 2:8217  10  2 3:917  10  3 TT0

298K o T o 363K
 3:0669  10  2  3:3:91123  10  3

8
>
> 8:4407100  1:07304 ; 1% r r 10%; Al2 O3
>
>
< 9:881100  0:9446 ; 1%r r 6%; CuO
 1:4594
>
> 1:9526100 ; 1% r r 10%; SiO2
>
>
: 8:4407100  1:07304 ; 1%r r 7%; ZnO
For oxide and metal nanoparticles suspended in water or
ethylene glycol based nanouids

2011 Corcione [157] 0:2% r r 9% [202204, 206]

10

knp 0:03 0:66 10 nmo dnp o 150 nm

kef f =kbf 1 4:4Re0:4 0:66 T
np Pr bf Tfr kbf

Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
2 K T
Renp bf 2 dB np 294K o T o 324K
bf
T f r is the frizing point of base uid
2014 Sharma et al. [183]    
 0:0336
0:01737 SiO2/water nanouids
1:37 T 0:2777 dnp np
kef f =kbf 0:8938 1 100 1 70 1 150 bf
TiO2/water nanouids
0% r r 4%
20 nmo dnp o 150 nm
293K o T o 343K
is thermal diffusivity

2014 Sundar et al. [184] kef f =kbf A B Al2O3(30 nm)/water nanouids

0:3% r r 1:5%
293K o T o 333K
8 8
< 1:1236 ; 20 : 80% EG=water
> < 8:0175 ; 20 : 80% EG=water
>
A 1:0806; 40 : 60% EG=water , B 10:164 ; 40 : 60% EG=water
>
: 1:0618; 60 : 40% EG=water >
: 10:448; 60 : 40% EG=water

2015 Esfe et al. [182] kef f =kbf 1 0:26876  0:99288  dnp

 0:35106 Fe/water(37 nm; 71 nm; 98 nm)

2015 Hassani et al. [185] knp 0:33 0:0313% r r 1%

kef f =kbf 1:04 1:11 0:99288 kbf
2 3 T bf : boiling point temperature of the base uid
 0:23  0:82    0:1    7 !
6 1 262 dref bf CP T bf 7 Br : Brownian velocity
Pr  1:7 4 
0:33 135 5
Pr  1:7 knp knp dnp Br T  1 Br 2 knp dref : molecular diameter of hydrogen
kbf

r
Br 18K B T
3 , dref 2:9A_
np dnp

1219
1220 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239

Table 2
Summary of effective dynamic viscosity models for nanouids.

Year Model Equation Remarks Numerical studies

1952 Brinkman ef f =bf 1 Spherical particles [82,145,194,198,199]

1  2:5
[195]

1956 Einstein [193] ef f =bf 1 2:5, o 0:05 Innitely dilute suspension of [192]
spheres
 
1998 Pak and Cho ef f =bf 1 39:11 533:92 Al2O3/water nanouids [145,192,194,199]
[126] TiO2/water nanouids

2005 Maiga et al. ef f =bf 1 7:3 1232 Al2O3/water nanouids [194,198]

[187] Al2O3/ethylene glycol
ef f =bf 1  0:19 3062 nanouids

bf
2005 Koo and ef f Brownian CuO/water nanouid [170,210212]
1  2:5
Kleinstreuer q 1% r r 4%
4 KBT
[209] Brownian 5  10 bf np dnp
f T; 300K o T o 325K

Al2O3/water nanouids
f T;  134:63 1722:3 0:4705  6:04 T
T0
(
0:0137100  0:8229 ; o 0:01

0:0011100  0:7272 ; 40:01

2007 Nguyen et al. ef f =bf 0:904e0:148 , dnp 47 nm CuO/water nanouids [82]

[137] ef f =bf 1 0:025 0:0152 , dnp 36 nm 1% r r 13%
T 295K
ef f =bf 1:475  0:319 0:0512 0:0093 , dnp 29 nm

2009 Abu-Nada ef f  0:155  19:582 0:794 2094:47  0:1922  8:11 0:01273 1:6044
2
2:175 Considering the inuence of [189,190]
T 
27463:863
T T2 T3 T T2
[188] temperature
Al2O3(36 nm; 47 nm)/water
nanouids
CuO(27 nm)/water
nanouids
1% o o 9:4%
295K o T o 348K

2009 Masoumi V d
2
Al2O3(13 nm; 28 nm)/water [82,213215]
ef f bf bf72C
B np

et al. [197] q r nanouids

18k T
3 6 dnp , V B d1np bfdnp , C bf  1 a b
np

 
2009 Timofeeva ef f bf 1 A1 A2 2 Nonspherical nanoparticles [115]
et al. [114] ( ( ( (
A1 37:1 A1 14:6 A1 13:5 A1 1:9
Platelet , Blade , Cylinder , Platelet
A2 612:6 A2 123:3 A2 904:4 A2 471:4

2011 Corcione [157] ef f =bf 1

 0:3 1:03
For oxide and metal nano- [180, 181, 201206]
1  34:87 dnp =dbf
particles suspended in water

1=3 or ethylene glycol based
df 0:1 6M
Nbf 0
nanouids
0:2% r r 9%
10 nmo dnp o 150 nm
294K o T o 324K

2014 Esfe and Sae- ef f =bf 0:9118Exp 5:49  0:00001359T 2 0:0303LnT ZnO(18 nm)/EG -
dodin [207] 0:25% r r 5%
298K o T o 323K

2014 Esfe et al. ef f =bf 1 11:61 1092 MgO(40 nm)/water -

[208] nanouids
0:0625% r r 1%

2014 Sharma et al.    

 0:061 SiO2/water nanouids -
11:3 T  0:038 dnp
ef f =bf 1 100 1 70 1 170
[183] TiO2/water nanouids
0% r r 4%
20 nmo dnp o 150 nm
293K o T o 343K

2014 Sundar et al. ef f =bf AeB Al2O3(30 nm)/water -

[184] nanouids
8 8
< 0:9396 ; 20 : 80% EG=water
> < 24:16 ; 20 : 80% EG=water
> 0:3% r r 1:5%
A 0:9299; 40 : 60% EG=water , B 67:43 ; 40 : 60% EG=water 293K o T o 333K
>
: 1:1216; 60 : 40% EG=water >
: 77:56; 60 : 40% EG=water

2015 Esfe et al. ef f =bf 1 0:100  0:69574  dnp

0:44708 Fe/water -
[182] (37 nm; 71 nm; 98 nm)
0:0313% r r 1%
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
Generally, convective heat transfer with nanouids can be
modeled using mainly the single and two-phase approach. In the
following, different studies based on both approaches along with
the related assumptions for each model are expressed thoroughly
and an attempt is made to determine which approach gives more
accurate results compared to experimental data.
4. Numerical analysis of convective heat transfer of nanouid
Despite the noticeable number of experimental works on dif-
ferent types of heat exchangers operating with nanouid as a base
uid, most investigators have just compared their simulation cases
with the previous numerical studies or with experimental ones
which have used pure water for cooling purpose. This may not be a
proper validation for thermal and hydraulic characteristics of
nanouids ow predicted by CFD methods. In the following, dif-
ferent CFD approaches are discussed separately and it is tried to
report those numerical studies in which the numerical results
have been compared with the experimental data to observe which
approach gives more realistic results.
4.1. Single-phase approach
4.1.1. Conventional single-phase approach
Although the mixture of nanoparticles and base uid is a multi-
phase problem due to the existence of ultra-ne nanoparticles, it is
assumed that the nanoparticles can be easily uidized and as a
result the nanouid can almost be considered as a conventional
single-phase (homogenous) uid with average physical properties
of individual phases. Further, in this model, it is considered that
both the liquid and particle phases are in thermal equilibrium and
move with the same velocity [80]. The general forms of governing
equations for steady state nanouid using conventional single-
phase approach are as follows [216]:
Conservation of mass:
!
: nf V 0 9 turbulent forced convection nanouid ow. In this study, rectan-
Conservation of momentum:
!!
! model proposed by Launder and Spalding [224]was involved. The
: nf V V  P nf 2 V  nf :vv  nf T  T 0 g 10 results of CFD model by using FLUENT software were compared
The third and fourth terms on the right side of Eq. (10) indicate
the effect of turbulent ow and natural convection, respectively. It
is obvious that for a laminar forced convection ow these terms
are considered zero.
Conservation of energy:
!
: C P nf V T : knf T  C P nf vt 11
where C P nf vt is applied for turbulent ow regime in the energy
equation.
To the best knowledge of authors, based on the open literature,
there are only a few publications available in which the convective
heat transfer of nanouids ow have been investigated both
numerically and experimentally, or in which the numerical results
have been validated with the experimental studies of heat transfer
and uid ow behaviors using nanouid, particularly for turbulent
ow regime.
Akhtari et al. [217] presented numerical and experimental
analysis for a laminar forced convection nanouid ow in double
pipe and shell and tube heat exchangers. The objective of this
study was to explore the effects of hot and cold volume ow rates,
nanouid temperature, and nanoparticle concentration on the
heat transfer characteristics. The geometries were modeled by
FVM using commercial code software, AnsysFLUENT 12.1.
1221
Unstructured quadrilateral cells were adopted for computational
domain discretization in the 2D conguration of double pipe heat
exchanger and tetragonal-hybrid cells were used in the 3D con-
guration of shell and tube heat exchanger. The results indicated
that the average relative error between the single-phase CFD
predictions and experimental data in the double pipe and shell
and tube heat exchangers was 11.2% and 15.6%, respectively. It was
noticed that the overall heat transfer coefcient of both double
pipe and shell and tube heat exchangers increases with increasing
the hot and cold volume ow rates, as well as the particle con-
centrations and nanouid inlet temperature.
Izadi et al. [218] studied the two-dimensional laminar forced
convection of dilute Al2O3/water nanouid in an annulus. FVM was
employed to discretize the single-phase governing equations along
with a rst order upwind method for convective and diffusive
terms, and Semi-Implicit Method for Pressure-Linked Equations
(SIMPLE) algorithm [219] for velocity-pressure coupling. The non-
uniform mesh was ne enough near the tube entrance and the
walls to tackle the velocity and temperature gradients. The authors
reported a good concordance between the numerical results for
the predicted friction factor and the experimental results pre-
sented in [220]. It was illustrated that the inuence of nanoparticle
volume fraction on the temperature proles is considerable but
negligible on the dimensionless axial velocity prole.
Moraveji et al. [221] used CFD tools to simulate convective heat
transfer effect on the nanouid ow in the developing region of a
tube with constant heat ux. The results were presented for the
Reynolds number range of 5002500, nanoparticle concentration
of 06% and the nanoparticle diameter of 45 nm and 150 nm. The
comparison results between the experimental data [222] and the
predictions of the heat transfer coefcients of the nanouids were
at maximum error of 10%. Further, the heat transfer coefcient
enhanced with increasing the nanoparticle concentration and
Reynolds number, and decreased with increasing the axial location
and nanoparticle diameter.
Hussein et al. [223] numerically investigated the impact of var-
ious cross sectional areas of tubes (circular, elliptical and at) on the
gular cells were used to mesh the surfaces of tube wall but trian-
gular cells were used to mesh the internal space. Standard k
with experimental data available in [126,133]. It appeared that the
friction factor and heat transfer coefcient of TiO2/water nanouid
at (12.5%) volume concentration have good compatibility with
deviation not more than 4% and 6% respectively. In addition the
friction factor of circular tube was higher than others followed by
elliptical and at tube. Mohammed et al. [225] studied the impact of
using various types of water-based nanouids on heat transfer and
uid ow characteristics in triangular shaped MCHS. The three-
dimensional steady, laminar ow and heat transfer governing
equations were solved using the nite volume method and hex-
ahedral cells were utilized for the numerical simulations. The
numerical results in terms of temperature difference between inlet
and outlet (T) versus ow rate and pressure drop (P) across the
MCHS as function of ow rate were in a good agreement as com-
pared with the experimental results of Chein and Chuang [226].
Among different types of nanoparticles including Al2O3, Ag, CuO,
diamond, SiO2, and TiO2 as the coolants with volume fraction of 2%
which were suspended in water (as base uid), diamond/water and
Ag/water nanouids were preferable to attain overall heat transfer
enhancement and low pressure drop, respectively.
Rashmi et al. [227] reported detailed investigations on heat transfer
enhancement using CNTs dispersed in water and stabilised by gum
arabic (GA). Both experiment and simulation were carried out in a
laminar ow counter-current heat exchanger. Numerical simulations
1222 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
were carried out in a 2D heat exchanger using a single-phase approach
by means of Fluent v 6.3. Numerical results exhibited good agreement
with the experimental results with a deviation of 73%. Meng and Li
[228] also reported the natural heat convection of water-based alu-
mina nanouids in a horizontal cylinder. Three-dimensional compu-
tational uid dynamics procedure was performed in a completely
open-source way. Considering both original OpenFOAM solver buoy-
ant_Boussinesq_Simple_Foam and a newly developed temperature-
dependent solver buoyant_Boussinesq_Simple_TDFoam, the earlier
solver was found better for 4% Al2O3/water cases compared to corre-
sponding experimental results, while the later solver was better for %
Al2O3/water nanouid cases. Rakhsha et al. [229] also used the original
OpenFOAM solver to investigate the steady state turbulent forced
convection developing ow of a CuO/water nanouid inside helically
coiled tubes at constant wall surface temperature. The numerical
results respectively demonstrate a 67% and a 910% increase in the
convective heat transfer and pressure drop of the applied CuO/water
nanouid over pure water, whereas experimental results show a 16
17% increase in the heat transfer coefcient and a 1416% increase in
the pressure drop for different tube geometries and different Re
numbers. Therefore, this inconsistency between numerical and
experimental results of nanouid was attributed to the presence of the
centrifugal forces which for such a two-phase uid could result in a
higher concentration of particles close to the outer wall and thus
increase the thermal conductivity and viscosity of that portion of the
space and affect the accuracy of the numerical homogeneous model
which assumes constant thermo-physical properties throughout the
domain. In CFD analysis conducted by Kumar et al. [230], they
reported simulation results on the convective heat transfer coefcient
and pressure drop of a helically coiled tube heat exchanger by using
CFD Fluent software package. On comparison, the the average relative
error between experimental Nu, pressure drop results and CFD results
were found to be 8.5% and 9.5% respectively.
Many researchers have justied the validity of their numerical
works with comparing them to the conventional proposed heat
transfer and pressure drop correlations of nanouid that have
been developed according to sets of theoretical experimental data
and are valid for limited type of geometries such as smooth tube
and enclosure.
A numerical investigation on forced convection with nano-
uids, composed of water and Al2O3 nanoparticles in 2D ribbed
channels was carried out by Manca et al. [231]. The ow regime
was turbulent for a range of Reynolds numbers (20,000 to 60,000)
with a range of volume fraction (0 to 4%) and particle size of
38 nm. Assuming single-phase model, turbulent ow and heat
transfer governing equations of continuity, momentum and energy
were solved by means of FLUENT code, employing the SST K 
model [232]. In the mesh generation process, structured cells were
taken into account with grid adoption of y 1 at adjacent wall
region. A very good correspondence was seen for the simulation
results and the given Nusselt number correlations by Pak and Cho
[126] and Maiga et al. [187] for the case of smooth channel. Finally,
it was declared that average Nusselt number increases with par-
ticle volume concentration and Reynolds number but increased
pumping power was required.
Rostamani et al. [233] studied the hydrodynamic and thermal
behavior of turbulent ow (20,000rRer100,000) in a two-
dimensional duct using Al2O3,CuO and TiO2 nanoparticles suspen-
sion at various concentrations (06%) under the constant heat ux
boundary condition. All the thermophysical properties of nanouids
were temperature-dependent. Control volume approach was cho-
sen to discretize the steady two-dimensional form of the continuity,
the time-averaged incompressible Navier-Stokes equations and the
energy equation. Reynolds stresses were modeled by k turbulence
model. Their predicted results in terms of Nusselt number for water
CuO 6% and TiO2 3% were quite close to that obtained by Pak and
Cho [126], Maiga et al. [187] and Gnielinskis [234] correlations.
By referring to a critical review on proposed heat transfer and
pressure drop correlations of nanouid by Sarkar [235], the inac-
curacy of these correlations both for laminar and turbulent ow
are criticized due to the following reasons: strong inuence of
particle properties and nanouid composition on ow and heat
transfer characteristics; misleading viscosity and thermal con-
ductivity data; lack of common understanding on basic mechan-
ism of nanouid ow; and insufcient experimental data for
nanouid heat transfer.
It is very important to know that type of turbulence model such
as Standard k, Renormalized Group (RNG) k, Realizable k,
Standard K  , Shear Stress Transport (SST) K  , Reynolds
Stress (RSM) and v2  f [224,232,236241], would result in a slight
variation among the thermal and hydraulic results. This assertion
can be backed up by Bayat and Nikseresht [242] who investigated
the thermal performance and pressure drop of nanouids in tur-
bulent forced convective ows through a circular tube under a
constant heat ux boundary condition. After analyzing different
turbulence models, they observed that Realizable k model
agrees well with Gnielinski equation.To illustrate the validity and
certitude of the used code, variation of local convective heat
transfer coefcient versus dimensionless length (x/D) for water
Al2O3 3% nanouid ow showed a reasonable agreement with the
experimental data given in [243].
The literature review reveals that the conventional single-
phase model has been used in several theoretical studies of con-
vective heat transfer using nanouids. This model is used mainly
due to its simplicity and computational efciency. However,
choosing the most appropriate thermophysical property correla-
tions for single-phase nanouid ow is very important which
signicantly affects the obtained results. It is not still clear which
correlation available in the literature is the best for a given
situation. Therefore, the concerns in conventional single-phase
modeling of nanouid consist in selecting the proper effective
properties and also taking into account the chaotic movement of
ultrane particles. Due to these facts, the numerical predictions of
this modeling may, in general, not be in good agreement com-
pared with experimental results.
4.1.2. Thermal dispersion single-phase approach
To handle the thermal dispersion resulting from chaotic
movement of the nanoparticles, the thermal dispersion single-
phase approach was proposed for the rst time by Xuan and
Roetzel [244]. In this modied model, it is assumed that the
relative motion of nanoparticles with respect to the base uid
introduces a perturbation (T! and V! , respectively) to both tem-
perature and velocity of the nanouid. Therefore the intrinsic
phase averages considering analogy with turbulence are given as:
T T T! 12a
!
! ! !
V V V 12b
R ! R! R
whereT 1
TdV; V V1 V dV; V1 T! dV 0:
V
V V V
By neglecting boundary surface between the uid and the
nanoparticles which are very small and substituting the above
!
values of V and T in Eq. (11) and rearranging,
0 1
 
!  
@ ! !A
: C P nf V T : knf T  : C P nf V T 13
The second term on the right side of Eq. (13) indicates the effect
of the thermal dispersion in energy equation.
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
The heat ux induced by the thermal dispersion in nanouid
ow can be represented as:
qd C P nf V! T!  kd T 14
where kd is the coefcient of the dispersed thermal conductivity.
Finally, Eq. (13) can be rewritten as below:
 
! work, a new correlation for the determination of radial dispersed
: C P nf V T :knf kd T 15
Xuan and Roetzel [244] proposed the dispersed thermal con-
ductivity of the nanouid as:
kd C d C P nf ux R 16
kd C d C P nf ux Rdp 17
Here C d is an unknown constant which should be determined
by matching experimental data. R and dp denote pipe radius and
nanoparticle diameter, respectively.
These relations suffer from two drawbacks: (i) The two func-
tional factors of nanouid, volume fraction and nanoparticle size
have not been included in relation (16); (ii) relation (17) is not
dimensionally compatible.
Nonetheless, a few authors applied the suggested relations in
[244]. Heris et al. [245] numerically investigated laminar-ow
convective heat transfer of nanouid in a circular tube with con-
stant wall temperature boundary condition. It was concluded that
dispersion and random movement of nanoparticles inside the uid
change the structure of the ow eld and lead to heat transfer
enhancement. To assess the accuracy of the dispersion model in
predicting heat transfer coefcients of nanouids, a comparison
between numerical and experimental results for Al2O3/water, CuO/
water, and Cu/water nanouids was carried out and the model
predictions showed good agreement. Kumar et al. [246] numeri-
cally solved the transport equations involving the thermal dis-
persion effect using semi-explicit nite volume solver in a collo-
cated grid. They estimated the natural heat transfer augmentation
for copperwater nanouid in a thermally driven two-dimensional
cavity. Their results showed that Nusselt number increases with
nanoparticle concentration which are inconsistent with experi-
mental trends. Therefore, it might be inferred that dispersion
model is useful to study forced convection of nanouids, but may
not be accurate for predicting natural convection behavior.
Mokmeli and Saffar-avval [247] introduced a new dispersed
thermal conductivity relation in radial direction to study the
laminar convective heat transfer behavior of nanouids in straight
pipes. They modied Xuan and Roetzel model and implemented
the radial dispersion assumption in their correlation as below:
 
R ux
kd C d C P nf 18
dp r
It was seen that the dispersion model predicts the nanouid
heat transfer behavior more reliably as compared to experimental
data. The discrepancy in regard to volume fraction variations was
reported less than 2.5% and for the variations of Reynolds number
was less than 4%. At the same time, the homogenous model pre-
dictions generally resulted in a discrepancy in the range of 1430%.
The problem of laminar nanouid ow through helically coiled
tubes at constant wall temperature has been investigated by
Akbaridoust et al. [248], both experimentally and numerically.
Pressure drop and the convective heat transfer behavior of nano-
uids containing 68 nm average size CuO nanoparticles dispersed
in water as a base uid with 0.1% and 0.2% volume fraction were
examined. An orthogonal staggered grid was generated to solve
the three-dimensional governing equations by nite difference
method (FDM) with projection algorithm using FORTRAN pro-
gramming language. Using modied dispersion model [247] the
1223
projection algorithm diminished the signicant deviation between
the numerical and experimental results and led to a negligible
difference of 4%.
Mojarrad et al. [249] numerically analyzed the heat transfer
performance of Al2O3 (30 nm)water nanouids owing in the
entrance region of a circular tube with constant surface tem-
perature on the walls using a control volume approach. In this
thermal conductivity was suggested as:
 
R T
kd C d C P nf 18
dp r
In addition, four different CFD approaches, namely homogenous
single-phase model, dispersion single-phase model, Eulerian-
Lagrangian two-phase model and mixture two-phase model, were
applied and the results of each model were compared with
experimental data. The homogenous single-phase model under-
estimated and the mixture model overestimated. Although the
Eulerian-Lagrangian model gave a reasonable prediction for the
thermal behavior of nanouids, the dispersion single-phase model
gave more accurate prediction despite its simplicity.
4.2. Two-phase approach
Generally speaking, nanouids are by nature two-phase uids;
therefore one can expect that they probably have some features of
solid-liquid mixture. For this reason, the theory of classical two-
phase ows has been applied for nanouids. In the two-phase
approach, the nanoparticle and the base uid are considered as
two different phases with different velocities and temperatures and
the assumption of zero slip velocity between the uid and particles is
no longer valid [78]. This is due to gravity, friction between the uid
and solid particles, Brownian motion, Brownian diffusion, sedi-
mentation and dispersion. Because the two-phase approach con-
siders the movement between the solid and uid molecule, it may
get realistic results. Although this model describes an understanding
of the functions of both liquid phase and solid phase in the heat
transfer process, it needs a long time for computation and a high
performance computer. The two-phase approach is categorized in
two general models: Eulerian-Eulerian and Lagrangian-Eulerian.
4.2.1. Eulerian-Eulerian model
Eulerian-Eulerian model is recommended for the two-phase
mixture consisting of high amount of solid particles. Consequently,
for simulation purpose, it is apparent that this can be a suitable
model for nanouids with extremely large number of nano-
particles, even at low volume fractions. There are three popular
Eulerian-Eulerian models, namely VOF (volume of uid), mixture,
and Eulerian.
4.2.1.1. Mixture model. One of the most common approaches in
modeling two-phase slurry ow is mixture theory (or the theory of
interacting continua) [250253]. The popularity of this approach in
multi-phase applications correspond to the following facts:
 It is simple in both theory and implementation.
 It requires less run time and CPU usage.
 It is straight-forward to introduce a turbulence model into the
mixture model.
 It is reasonably accurate for a wide range of two phase ows.
The mixture model is based on the following assumptions
[111,254]:
 The coupling between phases is strong, and particles closely
follow the ow.
1224 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239

 A single pressure is shared by all phases. The drift velocity is related to the relative velocity:
 The interactions between different dispersed phases are X
! ! n
k k !
neglected. V dr;p V pf  V fk
m
 Each phase has its own velocity vector eld and within a given k1

control volume there is a certain fraction of each phase.

 The secondary dispersed phases are assumed to consist of The relative velocity is determined from Eq. (23) proposed by
spherical particles of uniform particle size being specied dur- Manninen et al. [252] while Eq. (24) by Schiller and Naumann
ing calculations. [257] is used to calculate the drag function f drag .
 The concentrations of the secondary dispersed phases are 2  
solved from scalar equations taking into account the correction ! p dp p  m
V pf a 23
due to phase slip. 18f f drag p

The mixture model has the following limitations and require- (

1 0:15Re0:687
p Rep r 1000
ments [254]: f drag 24
0:0183Rep Rep 4 1000
 There is no possibility for phase change.
 The ow compressibility is not accounted for. The acceleration in Eq. (23) is:
!
!
 Pressure boundary condition cannot be specied because the a g  V m : V m
ideal gas law cannot be employed.
 Turbulence generation in the secondary phase is not accounted
for, nor is the turbulence of primary phase which is directly In this model, it is necessary to know that only one set off velocity
affected by the presence of secondary phase [255]. component is solved from the differential equations for mixture
momentum conservation. In addition, the velocities of dispersed
The following formulation represents the mathematical phases are obtained from the algebraic balance equations. This cau-
description of mixture model governing equations [216,256]: ses a remarkable reduction in the computational effort. Also, it is
Conservation of mass: important to note that the primary phase impresses the secondary
!
phase via drag and turbulence, while the secondary phase in turn
: m V m 0 19 inuences the primary phase via reduction in mean momentum and
turbulence. To apply the mixture theory for modeling of nanouid,
the continuity, momentum and uid energy equations are solved for
Conservation of momentum: the whole mixture rather than utilizing the governing equations of
! each phase independently.
! !
! X
n
: m V m V m  P m m 2 V  : k k vk vk It is declared from the literature that Behzadmehr et al. [258] were
k1 the rst who implemented the mixture approach to study numerically
!
X
n
! ! turbulent heat transfer involving 1% Cu/water nanouids throughout a
: k k V dr;k V dr;k circular tube. They compared their work with the experimental data of
k1
Xuan and Li [259] and concluded that mixture model is more precise
 m m T  T 0 g 20 than the single phase model. The same theory was used by Mirma-
! soumi and Behzadmehr [260] to analyze the laminar mixed convec-
where V dr;k is the drift velocity of the kth phase. tion of dilute Al2O3/water nanouid in a horizontal tube under con-
The third and last terms on the right side of Eq. (20) are con- stant heat ux boundary condition. The results showed that with
sidered in the case of turbulent ow and natural convection heat increasing nanoparticle volume fraction, the secondary ow is
transfer, respectively. strengthened and it would enhance the Nusselt number signicantly,
Volume fraction: while the skin friction coefcient at the fully developed region does
!
!
not notably change. With the aid of this two-phase ow approach, the
: p p V m  : p p V dr;p 21
authors also could gure out that the concentration of particles is
higher at the near wall region due to the signicant effect of viscous
Conservation of energy: forces in that area and the nanoparticles are concentrated more at the
! bottom of tube. In both previous studies, the discretization grid was
Xn
!    uniform in the circumferential direction and non-uniform in the other
: k V k k H k P : kT  C p m vt 22
two directions. Saberi et al. [261] examined the laminar forced con-
k1
vective heat transfer of nanouids containing alumina/water and zir-
where C p m vt shows the effect of turbulent nanouid ow. conia/water through a vertical heated tube using both single-phase
!
In the momentum conservation Eq. (20) V dr;k is the drift approach and two-phase mixture model. The comparison of their
velocity for secondary phase k (i.e., the nanoparticles) is dened results with experimental data showed that the mixture model had
as: better prediction for the convective heat transfer coefcient with 8%
! ! ! and 5% average relative error for alumina/water and zirconia/water
V dr;k V k  V m nanouids, respectively, while these gures were 13% and 8% for
single-phase model.
Another comparison between single-phase and two-phase was
The slip velocity (relative velocity) is dened as the velocity of
accomplished by Hejazian and Moraveji [262] in turbulent regime
secondary phase (p) relative to the velocity of the primary phase
through a horizontal circular tube exposed to convection with
(f):
saturated steam at the wall. The accuracy of the obtained results
! ! ! evaluated by single and mixture models were compared with the
V pf V p  V f
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239 1225

experimental data [263] and the maximum deviation for the single
phase model was 10.85%, and 9.36% for the mixture model,
respectively. In a similar way, Bianco et al. [264] simulated nano-
uid turbulent forced convection in a circular tube. They observed a
very good accord among the predicted results of mixture model
with the correlation proposed in [126]. Sakhani and Abbassi [265]
employed the same two-phase model to simulate the convective
heat transfer of Al2O3/water nanouid throughout ve different at
tubes using the nite volume method by FORTRAN programming
language. The ow regime and the wall boundary conditions were
assumed to be laminar and constant heat ux respectively. For grid
generation purpose, because of the large hydrodynamic and ther-
mal gradients, an exponential function was used near the walls and
also at the tube inlet. The simulated results in terms of local heat
transfer coefcient agreed well with the available experimental
data. It was deduced from the results that with increasing the tube
attening, the non-uniformity of nanoparticles is strengthened,
thereby resulting in the increase of heat transfer coefcient and
wall shear stress.
In another study done by Lot et al. [266], good agreement was
obtained between the CFD simulation results and the experi-
mental data by applying the mixture model for forced convective
nanouid ow. However, the single-phase model and the two-
phase Eulerian model underestimated the Nusselt number.
Moghari et al. [213, 214] published two papers on the heat transfer
enhancement of mixed convection laminar Al2O3/water nanouid
ow in a horizontal concentric annulus implementing mixture
model. Their rst work focused on investigating the effects of solid
nanoparticle volume fraction and variation of the inner wall heat
q
Conservation equations (i.e., equations of continuity, momen-
ux to the outer wall heat ux qw;i on the hydrodynamics and
w;o
While most of the researchers have widely focused on the
forced and mixed convection of nanouid using two-phase mod-
eling, Corcione et al. [269] recently brought the laminar natural
convection of nanouids into attention by means of a double-
diffusive, two-phase approach, in the hypothesis that Brownian
diffusion and thermophoresis are the primary slip mechanisms
between solid and liquid phases. The comparison between their
numerical solutions for square enclosures differentially heated at
sides and the corresponding experimental values of the ratio Qn/Qf
was at a satisfactory degree of agreement.
4.2.1.2. Eulerian model. To fully describe and predict the ow and
behavior of complex ows via Eulerian model, different assump-
tions are made as follows [270,271]:
 The pressure is shared by all the phases.
 Nanoparticles are considered the same as base uid as a
continuum.
 Separate continuity, momentum, and energy equations are
employed for different phases including primary and secondary
phases.
 The volume of each phase is calculated by integrating its
volume fraction throughout the domain, while the summation
of all the volume fractions is equal to unity.
 The difference between the velocity and temperature for liquid
and nanoparticle phases are considered and the governing
equations for both phases are solved numerically using the
nite volume method.
tum, and energy) are written as below [216]:

thermal behavior of the working uid. They reported that at a Conservation of mass:
given Reynolds and Grashof number, the Nusselt number at the !

inner and outer walls reduces with an increase in the heat uxes : q q V q 0 25
ratio and degradation of nanoparticles volume fraction. In a later ! R Pn
where V q q dV, q 1, and q indicates the phase.
work, two thermal boundary conditions were examined, one with V q1
a uniform heat ux at the inner wall and an adiabatic at the other
Conservation of momentum (qth phase):
wall, and the other inner and outer walls were heated in a same !
heat. They showed that decrement of nanoparticle mean diameter ! !
! X
n
: q q V q V q  q P q q 2 V q : k k vk vk
enhances the Nusselt number; however it did not have any con-
k1
siderable inuence on friction factor and contours of non- !

! X!
n
! !
dimensional axial velocity. It was noticed that, for both investi- q q g R pq F collision F lif t;q F vm;q
gations, there is a signicant agreement between the calculated k1

Nusselt number and friction factor with those reported in single- 26

phase numerical studies. P
n ! P
n ! !
Shariat et al. [215] modeled the laminar mixed convection of where R pq Spq V p V q stands for the interaction
k1 p1
nanouid ow in elliptic ducts. The results revealed that the
forces between the phases,
elliptic tube with AR 0.75 have almost the maximum Nusselt  
 
2
number and the minimum friction coefcient, therefore it was Spq q p q f =p , p p dp = 18q and f indicates the drag
recommended by the authors instead of the circular pipe. By the friction, which is calculated according to Schiller and Naumann
same token, in a given Reynolds number and Richardson number, [257] as:
increment of solid nanoparticles volume concentration increased
C D Rep
the Nusselt number and reduced the non-dimensional tempera- f 27
24
ture, while causing a small diminution in the friction coefcient.
( 24 1 0:15Re0:687
Moghadassi et al. [267] compared both single- and two-phase
p
Re r1000
mixture models to investigate the effect of nanouids on laminar CD Rep 28
forced convective heat transfer inside a horizontal circular tube. 0:44 Re 4 1000
They indicated that the maximum mean deviation from experi- ! !
mental data for the single phase model and the mixture model q j V p  V q jdp
Rep 29
were 8.48% and 4.76%, respectively. Likewise, Naphon and q
Nakharintr [268] did a comparison between the same CFD models
in the three-dimensional analysis of laminar convective heat !
For nanouids, the virtual mass effect ( F vm;q ) can be neglected
transfer of nanouids in a minichannel heat sink. In this study, the
because the secondary phase (nanoparticles) density is much
maximum deviations from their experimental data for the single-
greater than the primary phase (base uid) density. Moreover, the
phase and two-phase models were 3.74% and 1.66%, respectively.
1226 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
!
inclusion of lift force ( F lif t;q ) is not appropriate for nano-sized
particles. It was shown in [272] that particleparticle interaction
!
force ( F collision ) does not have any effect on the average Nusselt
number. Therefore, it is possible to ignore the particleparticle
interaction force for nanouid in the mathematical modeling.
Conservation of energy:
X 4.2.1.3. VOF (volume of uid) model. In general, the VOF model is
!   ! n
!
: q q V q H q  : kq :T q  q : V q Q pq 30
p1
! ! !
 The volume fraction of all phases is tracked over the entire
where Q pq h V p  V q is the heat exchange coefcient and
h
6kq q p Nup
. Nup is calculated from the Ranz and Marshall model
2
dp
[273]:
Nup 2 0:6Re0:5 Pr 0:333 31
 
q
where Pr q C p;q q =kq .
Kalteh et al. [272] carried out a numerical study on laminar forced
convection heat transfer of a copper/water nanouid inside an iso-
thermally heated microchannel. An Eulerian two-uid model was
adopted to model the nanouid ow and the governing equations for
both phases were solved using the nite volume method. For the rst
time, the detailed investigation of the relative velocity and tempera-
ture of liquid and nanoparticle phases were shown. They observed
that the relative velocity and temperature between the phases is very
small and negligible and the nanoparticle concentration distribution is
uniform. That is why they concluded that the hypothesis of taking into
account the nanouid as a homogeneous solution could be reasonable.
Also, the greater heat transfer enhancement was reported for higher
Reynolds number and nanoparticle volume concentration, and lower
particle size, which was accompanied with a slight increase in pres-
sure drop. Similar study was done by the same authors [176] both
numerically and experimentally and it was seen that two-phase
results are in better agreement with experimental results with the
maximum deviation of 7.42%, compared to 12.61% for homogeneous
(single-phase) modeling.
Fani et al. [271] preformed a CFD analysis to study the nano-
particles size effects on thermal performance and pressure drop of
CuO/water nanouid in a trapezoidal MCHS. Eulerian two-phase
numerical approach was utilized for forced convection laminar,
incompressible and steady state ow. Finite volume method was used
to solve the three-dimensional governing equations for both uid ow
and heat transfer. Their numerical studies showed that two-phase
modeling resulted in greater heat transfer than homogenous modeling
with an increment of around 15% at the entrance region of MCHS.
Moreover, an increase in pressure drop and a decrease in heat transfer
were reported as a result of nanoparticle size enlargement. It was
nally declared that base uid has more effect and contribution than
nanoparticles on thermal performance. Kalteh [271] extended the
previous study presented in Ref. [272] investigating the impact of
utilizing different nanouid types on heat transfer and pressure drop.
The results were presented for the nanoparticles concentration of 1%,
diameter of 100 nm and Reynolds number of 100. Water-based
nanouid had the lowest pressure drop and the highest heat trans-
fer coefcient than ethylene glycol and engine oil-based nanouids.
Moreover, non-dimensional pressure drop was almost the same for all
the nanoparticle types and the heat transfer coefcient was the
highest for diamond/water and the lowest for SiO2/water nanouid.
Beheshti et al. [274] made a comparison between single-phase
and two-phase models (mixture and Eulerian) to nd out which
one predicts experimental data more accurately. They analyzed the
turbulent convective heat transfer ofAl2O3/water and TiO2/water
nanouids through an annular channel and found two-phase
models give closer results to experimental data comparing to
single-phase model. The predicted values by the mixture model was
concluded to be slightly more accurate than the Eulerian model.
In total, the most important advantage of Eulerian model in
comparison to homogenous modeling is that there is no need for
effective thermophysical models for the nanouid [272]. However,
as mentioned in [266,274], it may not be as precise as the
mixture model.
based on the following assumptions [270]:
domain of study by solving a continuity equation for the sec-
ondary phases.
 A single set of momentum equations is solved for all the phases
in order to nd the velocity components which are shared by all
the phases.
 The total summation of the volume fractions for all the phases is
equal to unity. Consequently, the magnitude of the primary
phase volume fraction will be calculated.
 All of the physical properties are calculated by taking a
weighted-average of different phases based on their volume
fraction throughout each control volume.
 A shared temperature is calculated from a single energy
equation.
Mass conversion for VOF model is expressed as [254]:
!
: q V q q 0 32
P
n
where q 1 and all properties are calculated like
q1
P
n
N q N q .
q1
The momentum and energy equations are identical to Eqs. (10)
and (11).
A limited number of published papers have applied the VOF
method for nanouids. Akbari et al. [270] conducted a compara-
tive analysis of single and two-phase models in a horizontal tube
with uniform heat ux. Three different two-phase models (VOF,
mixture, Eulerian) were compared in their CDF study to simulate
laminar mixed convection nanouid ow. It was found that single-
phase and two-phase models predict almost identical hydro-
dynamic elds but very different thermal ones. Nevertheless, two-
phase models were in closer agreement with the experimental
data for the convective heat transfer coefcient than the single-
phase model, while they would over-estimate the convective heat
transfer coefcient for moderate and high particle volume frac-
tions. The authors showed more tendency toward using VOF (less
expensive) model due to the relatively same predictions by the
three two-phase models. The same research group [275] carried
out a similar numerical study but in turbulent ow regime. Rea-
lizable k model with enhanced wall functions was chosen for
this analysis. According to the results, thermal eld predictions by
the three two-phase models were almost identical but very far
from experimental data and single-phase model numerical results.
However, single-phase and two-phase models had nearly the
same hydrodynamic elds. In the case of two-phase model, the
results diverged more as particle volume fraction increased. In
contrast, however, Moraveji and Ardehali [254] reported that CFD
modeling based on the two-phase approaches represents a better
approximation of the experimental data compared to the single
phase form. Also, some other aforementioned researches
[258,262,266268,274] conrm this observation. It is evident that
there are contradictory results among the recently conducted
numerical works as compared with experimental data and the
question of which model is more accurate has remained
unanswered.
Table 3
Summary of numerical studies on single-phase nanouids.

Year Author Flow properties Model Geometry Nanouid Parameters Findings

2007 Heris et al. [245] Laminar forced Dispersion Circular tube (CuO, Cu, Al2O3)/ 0 r r 5% Dispersion and random movement of nanoparticles inside the uid change
convection water 10 nm o dnp o 50 nm the structure of the ow eld and lead to heat transfer enhancement
2500 r Pe r 6500 A comparison between numerical and experimental results for Al2O3/
water, CuO/water, and Cu/water nanouids was carried out and the dis-
persion model predictions showed good agreement

2009 Izadi et al. [218] Laminar forced - Annulus Al2O3/water 0 r r 5% Inuence of on the temperature proles is considerable but on the
convection dnp 25 nm dimensionless axial velocity proleis negligible
Re 100 & 900

2010 Rostamani et al. [233] Turbulent forced Standard k Long horizontal (CuO,TiO2, Al2O3)/ 0 r r 6% Their predicted results in terms of Nusselt number for water CuO 6% and

Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
convection duct water 2  104 r Re r 1  105 TiO23% were quite close to that of obtained by Pak and Cho, Maiga et al.
and Gnielinskis correlations

2010 Mokmeli and Saffar- Laminar forced Dispersion Horizontal tube graphite/ ATF& Al2O3/ 0 r AAl2O3=water r 1:6% Dispersion model predicts the nanouid heat transfer behavior more
avval [247] convection water 0 r graphite= ATF r 2:5% reliably as compared to experimental data. The discrepancy in regard to
dnp o 500 nm volume fraction variations was reported less than 2.5% and for the varia-
19:47 r Regraphite= ATF r 56:67 tions of Re number was less than 4%. While the homogenous model pre-
dictions generally resulted in a discrepancy in the range of 1430%

2011 Moraveji et al. [221] Developing region - Tube Al2O3/water 0 r r 6% h enhanced with increasing and Re, and decreased with increasing the
dnp 45 nm& 150 nm axial location anddnp
200 r Re r 1200 The comparison results between the experimental data and the predic-
tions of the heat transfer coefcients of the nanouids are at maximum
error of 10%

2011 Mohammed et al. [225] Laminar forced - Triangular MCHS (Al2O3,Ag,CuO,dia- 2% Diamond/water and Ag/water nanouids are preferable to attain overall
convection mon,SiO2, TiO2)/ 000 r Re r 1000 heat transfer enhancement and low pressure drop, respectively
water

2012 Hussein et al. [223] Turbulent forced Standard k Tube TiO2/water 1 r r 2:5% f and h have good compatibility with deviation not more than 4% and 6%
convection dnp 30 nm respectively, compared to experimental data
hf lat 4 helliptical 4 hcircular
5000 r Re r 2000 f f lat o f elliptical o f circular

2012 Bayat and Nikseresht Turbulent forced Realizable k Circular tube Al2O3/ (EG water) 0 r r 10% Among different turbulence models, Realizable k model agrees well with
[242] convection 1  104 r Re r 2  105 Gnielinskis equation

2013 Akbaridoust et al. [248] Laminar forced Dispersion Helically coiled CuO/water 0% r r 0:2% Utilization of base uid in helical tube with greater curvature results in
convection tube dnp 68 nm more effective heat transfer compared to the use of nanouid in straight
200 r Re r 1000 Projection algorithm diminishes the signicant deviation between the
numerical and experimental results and lead to a negligible difference of
4%

2013 Akhtari et al. [217] Laminar forced - Double pipe & shell Al2O3/water 0 r r 0:5% Overall heat transfer coefcient of both double pipe and shell and tube
convection and tube heat exchangers increases with increasing the hot and cold volume ow
rates, as well as the particle concentrations and nanouid inlet
temperature
dnp 80 nm Average relative error between the single-phase CFD predictions and
experimental data in the double pipe and shell and tube heat exchangers
was11.2% and 15.6%, respectively

2013 Manca et al. [231] Turbulent forced SSTK  Ribbed channel Al2O3/water 0 r r 4% A very good correspondence is seen for the simulation results and the
convection dnp 38 nm given Nusselt number correlations by Pak and Cho and Maiga et al. for the
case of smooth channel
2  104 r Re r 6  104 Average Nu number increases with and Re number but it is accompanied

1227
with increase of required pumping power
 1228
Table 3 (continued )

Year Author Flow properties Model Geometry Nanouid Parameters Findings

2013 Bahiraei and Hosseinali- Laminar forced Dispersion Circular tube TiO2/ water 0 r r 2% Closer results between the simulation and experimental values are
pour [221] convection 20 nm o dnp o 80 nm obtained once thermophoresis effect is taken into account and dispersion
500 r Re r 2000 approach is implemented

2014 Bahiraei et al. [298] Laminar forced Buongiorno Circular tube Al2O3/water 0 r r 1:5% Higher heat transfer coefcient was obtained than that of ignoring the
convection effects of particle migration
100 r Re r 500 Nanoparticle concentration at the tube center was higher in comparison
with the wall vicinity
Benetting from an articial neural network, more accurate heat transfer
coefcient was obtained with the average difference of 2.1% as compared
with the experimental work of Anoop et al.

Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
2014 Sheikhzadeh et al. [299] Natural convection Buongiorno Cavity Al2O3/water 0 r r 4% The predicted results by Buongiorno model led to a reasonable accuracy
respect to the experimental ones reported in [90]
25nm o dnp o 105 nm When the simulations were repeated by a homogeneous model, the model
overestimated heat transfer coefcients at all the volume fractions and the
disagreement increased as the temperature difference and bulk volume
fraction increased

2014 Garoosi et al. [94] Natural & mixed Buongiorno Cavity Al2O3/water 0 r r 5% There is an optimal volume fraction of the nanoparticles at each Rayleigh
convection 25nm o dnp o 145 nm and Richardson number for which the maximum heat transfer rate can be
102 r Gr r 104 obtained
0:01 r Ri r 1000
102 r Ra r 107

2015 Rashmi et al. [227] Laminar forced - Concentric tube CNT/gum Arabic 1% Numerical results exhibited good agreement with the experimental results
convection with a deviation of 7 3 %
dnp 20 nm Addition of GA showed no signicant effect on the thermal conductivity
enhancement while CNTs played a major role, which was temperature
dependent
CNT nanouids showed Newtonian behavior, and decrease in viscosity was
observed as a factor of temperature

2015 Meng and Li [228] Laminar natural - Horizontal cylinder Al2O3/water 1% & 4 % Numerical solvers in the OpenFOAM frame begin to give better agree-
convection ments to experimental investigations since the Ra number is larger than
2  107
0:7  107 r Ra r 5  107 Due to strong three-dimensional ow features being observed, the 3D CFD
simulation is recommended for natural heat convection problems in a
horizontal cylinder

2015 Rakhsha et al. [229] Turbulent forced Realizable k Helically coiled CuO/water 0 r r 1% Numerical results respectively demonstrate a 67% and a 910% increase
convection tube dnp 68 nm in the h and P of the applied CuO/water nanouid over pure water,
1000 r Re r 90000 whereas experimental results show a 1617% increase in the h and a 14
16% increase in the P for different tube geometries and different Re
numbers

2015 Kumar et al. [230] Laminar forced - Helically coiled Al2O3/water 0:1 r r 0:8% The average relative error between experimental Nu, pressure drop and
convection tube 1650 r De r 2650 CFD results were found to be 8.5%, 9.5% respectively
Table 4
Summary of numerical studies on two-phase nanouids.

Year Author Model Flow properties Geometry Nanouid Parameters Findings

2007 Behzadmehr et al. EuEu/mixture Turbulent forced Circular tube Cu/water 1% Mixture model is more precise than the single phase model
[258] convection dnp 42 nm
10515 r Re r 22540

2008 Mirmasoumi and Beh- EuEu/mixture Laminar mixed Horizontal tube Al2O3/water 0r r 7% With increasing, the Nu number augments signicantly
zadmehr [260] convection dnp 10 nm C f at the fully developed region does not notably change
Re 300 The concentration of particles is higher at the near wall region
0:22 r Ri r 1:9 and the nanoparticles are concentrated more at the bottom of
tube

2008 Mirmasoumi and Beh- EuEu/mixture Laminar mixed Horizontal tube Al2O3/water 4% h can be signicantly increases with decreasing dnp , while C f
zadmehr [110] convection 10 nm o dnp o 120 nm does not change notably

Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
Re 300 Larger nanoparticles and/or high value of the Gr numbers
2  104 r Gr r 17  104 would cause the non-uniformity of the particles distribution at
the tube cross section

2009 Bianco et al. [276] LaEu Laminar forced Circular tube Al2O3/water 0r r 4% The maximum difference of 11% is detected for h by comparing
convection dnp 100 nm single phase and two-phase models
200 r Re r 1200 Temperature dependent models present higher values of h and
Nu numbers because the difference between wall and bulk
temperature is minimized

2009 Akbarinia and Laur EuEu/mixture Laminar mixed Circular curved tube Al2O3/water 1% Increasing dnp causes to boost axial velocity andC f , while the Nu
[111] convection dnp 10  300 nm number, secondary ow as well as the temperature reduces
and 3  30m For nanometer-sized particles, increasingdnp , does not change
Re 648 the ow behaviors
Gr 5740

2009 He et al. [279] LaEu Laminar forced Horizontal tube TiO2/ water 0r r 1:18% Since movement of nanoparticles and interactions between
convection dnp 21 nm base uid and nanoparticles are considered in Lagrangian
model, this model led to a bit higher h than single-phase model
Re 300 & 1500 There is a signicant enhancement in heat transfer of nano-
uids particularly in the entrance region, which agree reason-
ably well with the experimental results

2010 Lot et al. [266] EuEu/mixture & Laminar/Turbulent Horizontal circular Al2O3/water 1r r 7% Good agreement was obtained between CFD simulation results
Eulerian forced convection tube and experimental data by applying the mixture model
dnp 42 nm Single-phase model and the two-phase Eulerian model under-
700 r Re r 1800 estimated the Nu number

2011 Bianco et al. [264] Eu-Eu/mixture Turbulent forced Horizontal circular Al2O3/water 0r r 6% A very good agreement was seen for mixture model as com-
convection tube dnp 38 nm pared with Pak and Cho correlation [126]
1  104 r Re r 10  104 Heat transfer enhancement increases with and Re number

2011 Moghari et al. [214] Eu-Eu/mixture Laminar mixed Horizontal con- Al2O3/water 0r r 5% At a given Re and Gr, the Nu number at the inner and outer
 
convection centric annulus dnp 28 nm walls reduces with an increase in qw;i =qw;o and degradation
Re 300 & 600 of
Gr 3  105 & 6  105

2011 Shariat et al. [270] EuEu/mixture Laminar mixed Horizontal elliptic Al2O3/water 0r r 5% AR 0.75 has almost the maximum Nu number and the mini-
convection duct mum f
dnp 28 nm In a given Re and Ri, increment of enhances the Nu number
and reduces the non-dimensional temperature, while it causes
a small diminution in f
2011 Akbari et al. [270] EuEu/VOF, mix- Laminar mixed Horizontal circular Al2O3/water 0r r 2% Single-phase and two-phase models predict almost identical
ture, Eulerian convection tube hydrodynamic elds but very different thermal ones
Two-phase models are closely in a good agreement with the
experimental data for h than the single-phase model, while
they over-estimate h for moderate and high

1229
Table 4 (continued )

1230
Year Author Model Flow properties Geometry Nanouid Parameters Findings

2011 Kalteh et al. [272] EuEu/Eulerian Laminar forced Rectangular MCHS Cu/water 0r r 5% Relative velocity and temperature between the phases is very
convection small and negligible and the nanoparticle concentration dis-
tribution is uniform
30 nm o dnp o 100 nm The greater heat transfer enhancement is achieved for higher
Re and as well as lower dnp , which was accompanied with a
slight increase in pressure drop

2012 Kalteh et al. [176] EuEu/Eulerian Laminar forced Rectangular MCHS Al2O3/water 0r r 5% The greater Nu number is achieved for higher Re and as well
convection 30 nm o dnp o 100 nm as lower dnp
The maximum deviation from experimental results is 12.61% and
7.42% for single-phase and two-phase methods, respectively

2012 Akbari et al. [275] EuEu/VOF, mix- Turbulent forced Horizontal circular Al2O3/water & CuO/water 0:02 r Al2O3=water r 0:05% Thermal eld predictions by the three two-phase models are

ture, Eulerian convection tube
1r CuO=water r 2%
dnp;Al2 O3 47 nm
dnp;CuO 100 nm
11000 r ReAl2O3=water r 20000
9500 r ReCuO=water r 22000
Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
approximately identical but very far from experimental data
and single-phase model numerical results
Single-phase and two-phase models have nearly the same
hydrodynamic elds
For two-phase model, the results diverge more as increases

2012 Moraveji and Esmaeili LaEu Laminar forced Circular tube Al2O3/water 0r r 4% The maximum difference of 11% is detected for h by comparing
[277] convection 250 r Re r 1050 single phase and two-phase models
dnp 100 nm Temperature dependent models present higher values of h and
Nu numbers because the difference between wall and bulk
temperature is minimized

2013 Moraveji and Ardehali EuEu/VOF, mix- Laminar forced Mini-channel heat Al2O3/water 0r r 6% CDF modeling based on the two-phase approaches represent a
[254] ture, Eulerian convection sink dnp 33 nm better approximation of the experimental data comparing to
130 r Re r 1600 the single phase form

2013 Saberi et al. [261] EuEu/mixture Laminar forced Horizontal circular Al2O3/water &ZrO2/water 0r r 2:76% Mixture model had better prediction for the convective heat
convection tube transfer coefcient
dnp 50 nm Single phase had better prediction for uid mean bulk tem-
perature of nanouids

2013 Hejazian and Moraveji EuEu/mixture Turbulent forced Horizontal circular TiO2/water 0:05 r r 0:25% Mixture model made a more accurate prediction of the
[262] convection tube dnp 30 nm experimental data than the single phase approach
4800 r Re r 30500 Increasing Re number and intensies the Nu number

2013 Moghari et al. [213] EuEu/mixture Laminar mixed Horizontal con- Al2O3/water 2% Decrement of dnp enhances the Nu number; however it does
convection centric annulus 13 nm o dnp o 72 nm not have any considerable inuence on friction factor and
Re 600 contours of non-dimensional axial velocity
Gr 3  105 & 6  105

2013 Fani et al. [271] EuEu/Eulerian Laminar forced Trapezoidal MCHS CuO/water 1r r 4% An increase in P and a decrease in heat transfer was seen as a
convection 100 nm o dnp o 200 nm result of dnp increment
Base uid has more effect and contribution than nanoparticles
on thermal performance
2013 Kalteh [271] EuEu/Eulerian Laminar forced Rectangular MCHS (Al2O3,CuO,Cu,Fe,Au,Ag, 1% Water-based nanouid had the lowest P and the highest h
convection TiO2,SiO2,Diamond)/ dnp 100 nm than EG- and EO-based nanouids
(water, EG,EO) Re 100 Non-dimensional P was almost the same for all the nano-
particle types and h was the highest for waterdiamond and
was the lowest for waterSiO2 nanouid

2013 Pakravan and Yaghoubi EuEu/mixture Natural convection Cavity Al2O3/water 1r r 3% Nu number of nanouid decreases as increases and it is in
[94] dnp 150 nm agreement with the experimental studies
5
6  10 r Ra r 2:5  10 6 Nu number of nanouid decreases with increasing thermo-
phoresis parameter or slip velocities

2013 Mojarrad et al. [249] SPD & LaEu & Eu Laminar forced Horizontal circular Al2O3/water 0r r 1% The homogenous single-phase model underestimated and the
Eu/mixture convection tube mixture model overestimated
 dnp 30 nm Although the EulerianLagrangian model gave a reasonable
650 r Re r 2300 prediction for the thermal behavior of nanouids, the disper-
sion single-phase model gave more accurate prediction despite
its simplicity
2014 Shariat et al. [109] EuEu/mixture Laminar mixed Horizontal elliptic Al2O3/water 0r r 5% There are non-linear relations between dnp and the nanouid
convection duct 13 nm o dnp o 72 nm ow behaviors
0:5 r Ri r 2 Nanoparticles had more tendencies to distribute near the bot-
Re 300 tom wall of the duct

2014 Mirzaei et al. [177] LaEu Laminar forced Parallel plate CuO/water & Al2O3/water 0r r 4% Two-phase modeling showed a slightly greater improvement in
convection microchannel dnp 100 nm h, compared to the homogeneous singlephase nanouid
technique
Different thermal-hydraulic behavior resulted from the com-
parison of Cu/water and alumina/water nanouids
200 r Re r 1800 Particle volume fraction distribution was uniform in most
regions except near the walls

at tubes 5%

Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
2014 Sakhani and Abbassi EuEu/mixture Laminar mixed Al2O3/water The simulated results in term of h agreed well with the avail-
[265] convection dnp 40 nm able experimental data
Re 300 With increasing the tube attening, the non-uniformity of
nanoparticles is strengthened, thereby resulting in the increase
of h and

2015 Bahremand et al. [280] LaEu Turbulent forced Helically coiled tube Silver/water 0r r 4% Two-phase approach predicted much more accurate results
convection than the homogeneous model
10 nm o dnp o 50 nm Nanoparticles do not alter the axial velocity and turbulent
kinetic energy remarkably
1400 r Re r 40000 Utilization of a base uid in a helical pipe improves the heat
transfer more effectively compared to the use of a nanouid in
a straight tube

2015 Moghadassi et al. [267] EuEu/mixture Laminar forced Horizontal circular Al2O3/water & Al2O3-Cu/ 0:1% The maximum mean deviation from experimental data for the
convection tube water dnp 15 nm single phase model and the mixture model were 8.48% and
4.76%, respectively
Re o 2300 For the hybrid nanouids, the average Nu number increase was
4.73% and 13.46% in compared to Al2O3/water and pure water,
respectively

2015 Naphon and Nakhar- EuEu/mixture Laminar forced Minichannel heat TiO2/water 0:4% The maximum deviations from the experimental data for the
intr [268] convection sink dnp 21nm single-phase and two-phase models were 3.74% and 1.66%,
respectively.
80r Re r 200

2015 Corcione et al. [269] EuEu/mixture Laminar natural Cavity CuO/water, Al2O3/water & 0r r 4% The comparison between the numerical solutions for square
convection TiO2/water 25 nm o dnp o 100 nm enclosures differentially heated at sides and the corresponding
experimental values of the ratio Qn/Qf was at a satisfactory
degree of agreement

2015 Beheshti et al. [274] EuEu/mixture & Turbulent forced Annular channel Al2O3/water & TiO2/water 0:1 r r 1:5 % For both Al2O3/water and TiO2/water nanouids, the predicted
Eulerian convection 25 nm o dnp o 100 nm values by the mixture model was slightly more accurate than
the Eulerian model

2015 Mahdavi et al. [281] EuEu/mixture Laminar forced Vertical tube Al2O3/water, 0:2 r r 2:76% DPM showed the ability of capturing slip velocity and tem-
and LaEu convection perature difference between particles and liquid
Pressure loss results predicted by DPM were found to be reli-
able for volume fraction less than 3%
SiO2/water & ZrO2/water 20 nm o dnp o 50 nm DPM could not report clustering of nanoparticles.
290 r Re r 1800

2015 Behroyan et al. [282] EuEu/ mixture, Turbulent forced Horizontal circular Cu/water 0r r 2% Newtonian single-phase and LagrangianEulerian were con-
Eulerian and La convection tube 20 nm o dnp o 50 nm cluded as the credible models
Eu 10000 r Re r 25000

1231
1232 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
4.2.2. LagrangianEulerian model
In Lagrangian-Eulerian model (or discrete phase model) the
uid phase is treated as a continuum by solving the time-averaged
Navier-Stokes equations, while the dispersed phase is solved by
tracking a large number of particles. The dispersed phase can
exchange momentum, mass, and energy with the uid phase. A
fundamental assumption made in this model is that the dispersed
second phase occupies a low volume fraction [216].
The following equations represent the mathematical formula-
tion of LagrangianEulerian two-phase model [276]:
Conservation of mass:
!
: V 0 33
Conservation of momentum:
!!
! ferent thermalhydraulic behavior resulted from the comparison
: V V  P 2 V Sm 34
Conservation of energy:
!
: C p V T :kT Se 35
Source/sink terms Sm and Se represent the integrated effects of
momentum and energy exchange with base uid as particles are
moving through an element of the Eulerian phase of the base uid
with volume of V and they are equal to zero in the case of single-
phase modeling.
The momentum transfer between the particles and base uid is
calculated as [177]:
1 X!
Sm F 36
V np
In the Lagrangian frame of reference, the equation of motion of
the particles is given by Eq. (37) as [177]:
! the Eulerian approach and the discrete phase model (DPM) from
dV p ! ! ! ! !
mp F D F L F g F b F br 37
dt
! hydrodynamic behaviors of nanouids through a vertical tube in
The left hand side of Eq. (37) V p is the velocity of the particle
and the right hand side represents the forces acting on the particle
which are drag, Saffmans lift, gravity, buoyancy and Brownian
forces, respectively.
The energy source term would be the heat transfer between the
phases as [177]:
1X  
Se Nup dp kp T T p 38
V np
where Nup is calculated from the Ranz and Marshall correlation
[273]:
Nup 2 0:6Re0:5 1=3
p Pr
Bianco et al. [276],and Moraveji and Esmaeili [277] performed
single- and two-phase CFD modeling of developing laminar forced
convection ow of alumina nanouid inside a circular tube under
constant heat ux. In both studies, Lagrangian-Eulerian model was
employed with either constant or temperature-dependent prop-
erties and the maximum difference of 11% was detected for the
average heat transfer coefcient by comparing single phase and
two-phase models. Their results were also satisfactory compared
with the available correlation given by Maiga et al. [187] and the
experimental data of Heris et al. [278]. It was concluded that the
temperature dependent models present higher values of heat
transfer coefcients and Nusselt numbers because the difference
between wall and bulk temperature is minimized. Convective heat
transfer of TiO2 nanouids owing through a straight tube under
the laminar ow conditions was studied numerically by He et al.
[279]. The analysis was performed using two different models:
Model I: single-phase ow model; Model II: Lagrangian trajectory
method. Since the movement of nanoparticles and the interactions
between base uid and nanoparticles are considered in Model II,
this model led to a bit higher heat transfer coefcient than Model
I. The results implied a signicant enhancement in heat transfer of
nanouids particularly in the entrance region, which agree rea-
sonably well with the experimental results. In a study done by
Mirzaei et al. [177], LagrangianEulerian approach was chosen to
investigate the heat transfer of laminar nanouid ow in a
microchannel under constant wall temperature. The results based
on two-phase modeling showed a slightly greater improvement in
the heat transfer coefcient compared to the homogeneous single
phase nanouid technique. Further, it was found that not only
does thermal conductivity have effect on heat transfer augmen-
tation, but also the inuence of other nanouids thermophysical
properties should be considered simultaneously due to the dif-
of Cu/water and alumina/water nanouids. However, the particle
volume fraction distribution was uniform in most regions except
near the walls, which had a greater inuence on the ow and heat
transfer characteristics. Bahremand et al. [280] investigated the
turbulent convection ow of water and watersilver nanouid in
helically coiled tubes both experimentally and numerically. The
simulation was conducted by LagrangianEulerian two-phase
approach in connection with an RNG k turbulence model
accounting for four-way coupling collisions using ANSYS CFX
software. Two-phase approach predicted much more accurate
results than the homogeneous model. It was also shown that the
nanoparticles do not alter the axial velocity and turbulent kinetic
energy remarkably. Meanwhile, it was observed that the utiliza-
tion of a base uid in a helical pipe improves the heat transfer
more effectively compared to the use of a nanouid in a
straight tube.
Mahdavi et al. [281] employed both the mixture model from
the Lagrangian approach so as to estimate the thermal and
laminar ow. Three types of nanoparticles with different dia-
meters were considered, namely Alumina, Zirconia and silica
nanoparticles. The Mixture model presented the same velocity and
temperature prole for both phases, while DPM showed the ability
of capturing slip velocity and temperature difference between
particles and liquid. Therefore, they highly recommended DPM to
model heat transfer in vertical tubes with laminar nanouid ow.
Behroyan et al. [282] comprehensively compared the predictions
of ve types of CFD models, including two single-phase models
(Newtonian and non-Newtonian) and three two-phase models
(Eulerian, mixture and Lagrangian), to investigate turbulent forced
39 convection of Cu/water nanouid in a tube with a constant heat
ux on the tube wall. They checked their numerical ndings
against the experimental ones from literature and recommended
Newtonian single-phase and LagrangianEulerian as the credible
models.
It is declared from the above literature survey that not enough
studies have been done on turbulent heat transfer of nanouid
ow using LagrangianEulerian two-phase method. Therefore,
some studies are needed to be done in order to determine the
capability of the LagrangianEulerian model in simulating nano-
uids characteristics in turbulent regime.
Tables 3 and 4 summarize the reviewed single- and two-phase
numerical works done by different authors in the order of year,
providing the details about modeling approach, ow properties
and nanouid type along with some important parameters and
the ndings for different geometries.
The CFD commercial code, FLUENT and, to a much smaller extent,
OpenFOAM have been employed in [227,230,264,268,275,276,279
282] and [228,229], respectively. But the majority of investigators
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
have written a computer code for two-phase simulation of nanouid
ow and heat transfer.
To simulate nanouid ow in the turbulent ow regime,
standard k model and Realizable k model were considered in
Refs. [258,262,264,266,268,282,229,242,275], respectively.
Although most of the two-phase studies in this literature only
declared that non-uniform grid has been utilized for the dis-
cretization of computational domain, there are some authors who
have specically reported the grid type in their works. For exam-
ple, a structured non-uniform grid distribution has been used in
!
[111,177,213,227,229,254,262,265,274,280], while in the cases
performed by Shariatet al. [109,215] and by Meng and Li [228]
unstructured non-uniform mesh was utilized. However, near the
!
wall and the entrance region of analyzed geometries (e.g. tube,
duct, and microchannel) where the velocity and temperature
gradients are high, ner grids have been generated by all the
researchers.
In the most above-mentioned studies such as [177,258,266,
271,276,279,280], it was observed that two-phase modeling has higher
heat transfer magnitude in comparison to the homogeneous model-
ing. Thus, under-estimation of the heat transfer enhancement by
homogeneous modeling seems to be related to the insufcient accu-
racy of the nanouid thermophysical property models that are used in
the homogeneous modeling.
4.3. Other CFD approaches
4.3.1. Lattice Boltzmann Method (LBM)
In spite of using traditional CFD approaches, Lattice Boltzmann
Method (LBM) (or thermal lattice Boltzmann methods (TLBM) has
exhibited some attractiveness during recent years in simulating
the nanouid ow and their heat transfer behavior. LBM solves the
Boltzmann equation to simulate the ow instead of solving the
NavierStokes equations. In LBM simulations, two different forms
are utilized more than others: Typical D2Q9 (two-dimensional and
nine-velocity) square and D3Q19 (three-dimensional and 19-
velocity) cube lattice structures [283]. In addition, this method
may present some well-known advantages such as simple and
efcient implementation for parallel coding, the uniform algo-
rithm for multiphase ows, and easiness in dealing with complex
geometries [284]. Xuan and Yao [285] rst proposed LBM for
simulating ow and energy transport processes inside the nano-
uid, considering the effective forces on ow like gravity, buoy-
ancy, Brownian, and the interactive force among nanoparticles.
LBM has been widely exploited for natural and mixed convections
[82,83,286289], and to a smaller extent for forced convection
[290293]. The obtained results from these studies are in accor-
dance with the available numerical results which are based on the
conventional CFD methods. However, to our knowledge, the
researchers who adopted LBM did not bring any code validation
according to the experimentally conducted nanouid works in
order to assess the accuracy of LBM for nanouids. Therefore,
further researches might be necessary in this area to see to what
extend LBM gives reliable results for nanouid simulation.
4.3.2. Non-homogeneous two-component model
Buongiorno [93] theoretically investigated the effect of nano-
particle dispersion on the energy transfer of nanouids and sug-
gested an approach which is known as non-homogeneous two-
component model. According to [93], there are different slip
mechanisms between nanoparticles and base uid such as inertia,
Brownian diffusion, thermophoresis, diffusiophoresis, magnus
effect, uid drainage, and gravity in the absence of turbulent
effects. It was concluded that Brownian diffusion and thermo-
phoresis effects are the two most important mechanisms that
induce variations of nanoparticle concentration. He also assumed
1233
that energy transfer by nanoparticle dispersion was negligible;
hence he ignored Dufour effect. The four coupled governing
equations for Buongiornos non-homogeneous model are written
as [93]:
Conservation of mass:
!
: nf V 0 40
Conservation of momentum:
!!
: nf V V  P nf 2 V 41
Conservation of energy:
!
: C P nf V T :knf T C p;np J np :T 42
Conservation of nanoparticles:
! 1 !
V :  : J np 43
np
!
In Eqs. (42) and (43) J np stands for nanoparticles ux and can
be written as combination of the effects of Brownian motion and
thermophoresis as:
! ! !
J np J np;B J np;T 44
!
J np;B is nanoparticle ux in Brownian motion calculated based
on EinsteinStokes' model [93]:
! kB T
J np;B  np DB DB 45
3f dnp
and the second term on the right side of Eq. (44) stands for particle
ux due to thermophoresis which can be calculated using the
McNabMeisen [294] approximation for the thermophoretic
velocity of particles dispersed in a liquid:
! kf f
J np;T  np DT TDT 0:26 46
2kf knp f T
The thermophoretic velocity (  DT T) given in Eq. (46) is
proportional to the uid viscosity. However, a vast majority of
theoretical and experimental results [295,296] as well as mole-
cular dynamics simulations [297] imply that thermophoretic
velocity in colloidal suspensions is inversely proportional to the
uid viscosity. Thus, further studies are needed to determine the
inuence of thermophoresis on heat transfer in nanouids.
Bahiraei et al. [298] presented a 2-D work on the laminar con-
vective heat transfer of alumina nanouid inside a tube and took
into account the particle migration effects by employing Buon-
giorno mathematical model. Higher heat transfer coefcient was
obtained than that of ignoring the effects of particle migration.
Furthermore, the concentration at the tube center was higher in
comparison with the wall vicinity. They also used an articial neural
network to predict the convective heat transfer coefcient and
could achieve approximately accurate results with the average dif-
ference of 2.1% as compared with the experimental work of Anoop
et al. [222].
Considering the same CFD method, Sheikhzadeh et al. [299]
published a paper on laminar natural convection ofAl2O3/water
nanouid in a square enclosure. A staggered grid system together
with the SIMPLER algorithm was adopted to couple the pressure
and the velocity components. Discretization was done using con-
trol volume technique. The predicted results by Buongiorno model
led to a reasonable accuracy respect to the experimental ones
reported in [90]. However, when the simulations were repeated by
a homogeneous model, the model overestimated heat transfer
coefcients at all the volume fractions and the disagreement
increased as the temperature difference and bulk volume fraction
increased.
1234 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
5. Conclusion
The present article comprehensively reviews the convective heat
transfer of nanouids from the numerical point of view. From the
observed numerical results it is evidently seen that nanouids can
improve the heat transfer capability of conventional heat transfer
uids signicantly due to the existence of highly conductive dis-
persed nanoparticles in the host liquid. Not only does the increase
of nanoparticle volume concentration bring about an increasing
trend in forced convective heat transfer of nanouids, also it
increases the pressure drop slightly, which is widely expressed by
most numerical works. In contrast, among the experiments and
simulations there are some inconsistencies about the inuence of
particle concentration on the variation of heat transfer in natural
convection which is attributed to improper selection of CFD mod-
eling and effective viscosity correlation for nanouids as well as
ignoring their non-Newtonian rheology in the numerical simula-
tions. With decreasing nanoparticle mean diameter, the heat
transfer coefcient would experience a considerable increase, while
the skin friction coefcient may not change notably. In addition, the
available correlations for prediction of nanouid thermophysical
properties have been proposed on the basis of a limited number of
experimental data and only very few correlations are available to
predict effective thermal conductivity and viscosity of nanouids
considering the effect of volume fraction, particle shape, particle
size, nanouid temperature and Brownian diffusion. Thus, it is still a
challenging issue for researchers to develop a general empirical
correlation or a theoretical model in order to estimate the ther-
mophysical properties of nanouids accurately. Furthermore, par-
ticle shapes could affect the thermal conductivity as well as visc-
osity of nanouids and consequently, change the thermalhydraulic
characteristic of nanouids. Therefore, more investigations are
needed to be performed by applying other kinds of particle shapes
other than spherical-shaped particles.
The review also extended to a comprehensive comparison
between single-phase and two-phase nanouid ows to nd the
suitable CFD approach which gives more accurate results as com-
pared to experimental results. Conventional single-phase modeling
has been the most common approach in majority of theoretical
studies of convective heat transfer using nanouid. However, the
numerical predictions resulted from this technique may lead to
some deviations as compared with the experimental data due to
the neglect of slip mechanisms between nanoparticle and base
uid. Employing dispersion model in forced convection studies
could improve the results signicantly, whereas it might not be
accurate for predicting natural convection behavior. As the two-
phase approaches consider the two-phase ow nature of nano-
uids, they may get more realistic results for the convective heat
transfer simulation of nanouids. Most publications conrm that
two-phase modeling has higher heat transfer magnitude in com-
parison to the homogeneous modeling, but because of contra-
dictory results among the recently conducted numerical works, it is
still controversial whether two-phase approach gives more credible
results. However, LagrangianEulerian and Buongiorno models
seem to be more reliable for simulation purpose, as they incorpo-
rate Brownian motion and Soret effects as the two migration
mechanisms in the boundary layer and moreover, they are capable
to predict non-uniform particle distribution.
5.1. Prospectives
The gaps that can be drawn from the present review and the
future directions are suggested as follows:
 To reach universal models for thermophysical properties of
nanouids, more complete experiments involving a wide range
of nanoparticles and base uids need be conducted in the
future.
 Dispersion and random movement of particles, clustering, and
temperature must be taken into account in the model devel-
opment for nanouids.
 More investigations are needed to be performed by applying
other kinds of particle shapes other than spherical-shaped
particles.
 More numerical investigation is needed to be done in several
different cases to conrm that two-phase models excel the
homogeneous one in a particular geometry.
 Nanouid is highly suited for the application in practical heat
transfer processes since it has a great potential for heat transfer
enhancement. An opportunity is offered for engineers for devel-
oping highly compact and effective heat transfer equipment.
 In practical applications the uid ow is generally in the tur-
bulent regime, because higher heat transfer is achieved through
turbulent ow. Thus there is much work to be done in under-
standing the overall effect of various turbulent models on the
heat transfer results of nanouid.
Acknowledgment
This research is nancially supported by University of Malaya,
Ministry of Higher Education High Impact Research (UM.C/HIR/
MOHE/ENG/20), and Exploratory Research Grant Scheme (ERGS:
ER013-2013A).
References
[1] Abdelaziz EA, Saidur R, Mekhilef S. A review on energy saving strategies in
industrial sector. Renew Sustain Energy Rev 2011;15:15068.
[2] Jang SP. Choi SUS. Cooling performance of a microchannel heat sink with
nanouids. Appl Thermal Eng 2006;26:245763.
[3] Tuckerman DB, Pease R. High-performance heat sinking for VLSI. Electron
Device Lett IEEE 1981;2:1269.
[4] Kishimoto T, Ohsaki T. VLSI packaging technique using liquid-cooled chan-
nels. IEEE Trans Compon Hybrids Manuf Technol 1986;9:32835.
[5] Wang B, Peng X. Experimental investigation on liquid forced-convection heat
transfer through microchannels. Int J Heat Mass Transf. 1994;37:7382.
[6] Jiang L, Mikkelsen J, Koo J-M, Huber D, Yao S, Zhang L, et al. Closed-loop
electroosmotic microchannel cooling system for VLSI circuits. IEEE Trans
Compon Packag Technol 2002;25:34755.
[7] Mohammed H, Bhaskaran G, Shuaib N, Saidur R. Heat transfer and uid ow
characteristics in microchannels heat exchanger using nanouids: a review.
Renew Sustain Energy Rev 2011;15:150212.
[8] Kandlikar S, Garimella S, Li D, Colin S, King MR. Heat Transfer and Fluid Flow
in Minichannels and Microchannels. Oxford: Elsevier; 2005.
[9] Khan MG, Fartaj A. A review on microchannel heat exchangers and potential
applications. Int J Energy Res 2011;35:55382.
[10] Shah R, Webb R. Compact and enhanced heat exchangers. Heat Exch: Theory
Practice 1983:42568.
[11] Shah RK, Sekuli DP. Heat exchanger design procedures. Fundamentals of
heat exchanger design. John Wiley & Sons, Inc; 2007. p. 60172.
[12] Steinke ME, Kandlikar SG. Review of single-phase heat transfer enhancement
techniques for application in microchannels, minichannels and micro-
devices. Int J Heat Technol 2004;22:311.
[13] Webb RL, Kim NH. Principles of enhanced heat transfer. New York, NY, USA:
Taylor & Francis Group; 2005.
[14] Zimparov V. Energy conservation through heat transfer enhancement tech-
niques. Int J Energy Res 2002;26:67596.
[15] Syam Sundar L, Singh MK. Convective heat transfer and friction factor cor-
relations of nanouid in a tube and with inserts: a review. Renew Sustain
Energy Rev 2013;20:2335.
[16] Maxwell JC. A treatise on electricity and magnetism. London: Oxford Uni-
versity Press; 1904.
[17] Ahuja AS. Augmentation of heat transport in laminar ow of polystyrene
suspensions. I. Experiments and results. J Appl Phys 1975;46:340816.
[18] Choi S, Eastman JA. Enhancing thermal conductivity of uids with nano-
particles. ASME-Publications-FED 1995;231:99106.
[19] Wen D, Lin G, Vafaei S, Zhang K. Review of nanouids for heat transfer
applications. Particuology. 2009;7:14150.
[20] Saidur R, Leong K, Mohammad H. A review on applications and challenges of
nanouids. Renew Sustain Energy Reviews 2011;15:164668.
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
[21] Huminic G, Huminic A. Application of nanouids in heat exchangers: a
review. Renew Sustain Energy Rev 2012;16:562538.
[22] Turgut A, Elbasan E. Nanouids for electronics cooling. In: Proceedings of the
IEEE 20th International Symposium for design and technology in electronic
packaging (SIITME): IEEE; 2014. p. 357.
[23] Vasu V, Rama Krishna K, Kumar A. Heat transfer with nanouids for elec-
tronic cooling. Int J Mater Product Technol 2009;34:15871.
[24] Mahian O, Kianifar A, Kalogirou SA, Pop I, Wongwises S. A review of the
applications of nanouids in solar energy. Int J Heat Mass Transf.
2013;57:58294.
[25] Nematolahi M, Behzadinejad B, Golestani A. Feasibility study of using
nanouids as a neutron absorber in reactor emergency core cooling system.
Int J Hydrogen Energy.
[26] Buongiorno J, Hu LW. Nanouid heat transfer enhancement for nuclear
reactor applications. In: Proceedings of the ASME 2009 second international
conference on micro/nanoscale heat and mass transfer. American Society of
Mechanical Engineers; 2009. p. 51722.
[27] Buongiorno J, Hu L-W, Kim SJ, Hannink R, Truong B, Forrest E. Nanouids for
enhanced economics and safety of nuclear reactors: an evaluation of the
potential features, issues, and research gaps. Nuclear Technol 2008;162:8091.
[28] Jia T, Wang R, Xu R. Performance of MoFe2O4NiFe2O4/Fullerene-added
nano-oil applied in the domestic refrigerator compressors. Int J Refrig
2014;45:1207.
[29] Xing M, Wang R, Yu J. Application of fullerene C60 nano-oil for performance
enhancement of domestic refrigerator compressors. Int J Refrig 2014;40:
398403.
[30] Anandakumar J. To conduct the performance test on chiller unit by using
nanouid cooled condenser. Int J Mech Eng Robot Res 2015;4:279.
[31] Liu M, Lin M, Wang C. Enhancements of thermal conductivities with Cu, CuO,
and carbon nanotube nanouids and application of MWNT/water nanouid
on a water chiller system. Nanoscale Res Lett 2011;6:113.
[32] Cui W-Z, Bai M-L, Lv J-Z, Zhang L, Li X-J, Xu M. Basic research on ow of
nanouids in cooling system of internal combustion engine. Trans Chin Soc
Intern Combust Engines 2012;30:4955.
[33] Raja M, Vijayan R, Suresh S, Vivekananthan R. Effect of heat transfer
enhancement and NOx emission using Al2O3/water nanouid as coolant in CI
engine. Indian J Eng Mater Sci 2013;20:4439.
[34] Azimi M, Ommi F. Using nanouid for heat transfer enhancement in engine
cooling process. J Nano Energy Power Res 2013;2:1324.
[35] Kulkarni DP, Das DK, Vajjha RS. Application of nanouids in heating build-
ings and reducing pollution. Appl Energy 2009;86:256673.
[36] Kulkarni DP, Vajjha RS, Das DK, Oliva D. Application of aluminum oxide
nanouids in diesel electric generator as jacket water coolant. Appl Thermal
Eng 2008;28:177481.
[37] Basha JS, Anand RB. An experimental study in a CI engine using nanoadditive
blended waterdiesel emulsion fuel. Int J Green Energy 2011;8:33248.
[38] Basha JS, Anand R. An experimental investigation in a diesel engine using
carbon nanotubes blended waterdiesel emulsion fuel. Proc Inst Mech Eng
Part A: J Power Energy 2011;225:27988.
[39] Ollivier E, Bellettre J, Tazerout M, Roy GC. Detection of knock occurrence in a
gas SI engine from a heat transfer analysis. Energy Convers Manag 2006;47:
87993.
[40] Alexiou C. Nanomedicinebasic and clinical applications in diagnostics and
therapy. Chemotherapy 2011;57:449520.
[41] Khosroshahi ME, Nourbakhsh MS, Ghazanfari L. Synthesis and biomedical
Application of SiO2/Au Nanouid Based on Laser-Induced Surface Plasmon
Resonance Thermal Effect. J Modern Phys 2011;2:944.
[42] Buteic A, Mihaiescu D, Grumezescu A, Vasile B, Popescu A, Mihaiescu O,
et al. The anti-bacterial activity of magnetic nanouid: Fe3o4/Oleic acid/
cephalosporins core/shell/adsorption-shell proved on S. aureus and E. coli
and possible applications as drug delivery systems. Digest J Nanomater
Biostruct 2010:5.
[43] Tombcz E, Bica D, Hajdu A, Ills E, Majzik A, Vks L. Surfactant double layer
stabilized magnetic nanouids for biomedical application. J Phys: Condens
Matter 2008;20:204103.
[44] Zakaria IA, Michael Z, Mohamed W, Najmi WA. Nanouid as Cooling Medium
in Polymer Electrolyte Membrane (PEM) Fuel Cell: A Study on Potentials and
Possibilities. Adv Mater Res: Trans Tech Publ 2015:31923.
[45] Sharma T, Mohana Reddy AL, Chandra TS, Ramaprabhu S. Development of
carbon nanotubes and nanouids based microbial fuel cell. Int J Hydrogen
Energy. 2008;33:674954.
[46] Zakaria I, Azmi WH, WANW Mohamed, Mamat R, Naja G. Experimental
investigation of thermal conductivity and electrical conductivity of Al2O3
nanouid in waterethylene glycol mixture for proton exchange membrane
fuel cell application. Int Commun Heat Mass Transf 2015;61:618.
[47] Sharma AK, Tiwari AK, Dixit AR. Progress of nanouid application in
machining: a review. Mater Manuf Process 2014;30:81328.
[48] Huang W-T, Wu D-H, Chen J-T. Robust design of using nanouid/MQL in
micro-drilling. Int J Adv Manuf Technol 2015:17.
[49] Nam JS, Kim DH, Chung H, Lee SW. Optimization of environmentally benign
micro-drilling process with nanouid minimum quantity lubrication using
response surface methodology and genetic algorithm. J Clean Prod
2015;102:42836.
[50] Davidson J. Nanouid for cooling enhancement of electrical power equip-
ment. Nashville, TN: Vanderbilt University-Department of Electrical
Engineering; 2009 transfer Bulletin.
1235
[51] Hwang JG. Electron scavenging by conductive nanoparticles in oil insulated
power transformers. In: Proceedings of the 2009 joint electrostatics con-
ference. Citesser, Boston; 2009.
[52] Ramu TS, Keshavan BK, Balasubramanya Murthy KN. Application of a class of
nano uids to improve the loadability of power transformers. In: Proceed-
ings of the IEEE 10th international conference on the properties and appli-
cations of dielectric materials (ICPADM). 2012. p. 1-6.
[53] Stoian FD, Holotescu S, Taculescu A, Marinica O, Resiga D, Timko M, et al.
Characteristic properties of a magnetic nanouid used as cooling and insu-
lating medium in a power transformer. In: Proceedings of the 8th interna-
tional symposium on advanced topics in electrical engineering (ATEE). 2013.
[54] Devananda R, Prabhu KN. Determination of cooling performance of nano-
uids for use in transformers by an instrumented copper Probe. J Nanouids
2014;3:3843.
[55] Dondapati RS, Saini V, Kishore N. Enhancement of performance parameters
of transformer using nanouids. Int J Sci Eng Technol 2015;4:3836.
[56] Haddad Z, Oztop HF, Abu-Nada E, Mataoui A. A review on natural convective
heat transfer of nanouids. Renew Sustain Energy Rev 2012;16:536378.
[57] Hussein AM, Sharma K, Bakar R, Kadirgama K. A review of forced convection
heat transfer enhancement and hydrodynamic characteristics of a nanouid.
Renew Sustain Energy Rev 2014;29:73443.
[58] Mahian O, Kianifar A, Kleinstreuer C, MdA Al-Nimr, Pop I, Sahin AZ, et al. A
review of entropy generation in nanouid ow. Int J Heat Mass Transf
2013;65:51432.
[59] Gupta M, Arora N, Kumar R, Kumar S, Dilbaghi N. A comprehensive review of
experimental investigations of forced convective heat transfer characteristics
for various nanouids. Int J Mech Mater Eng 2014;9:121.
[60] Mohammed H, Al-Aswadi A, Shuaib N, Saidur R. Convective heat transfer and
uid ow study over a step using nanouids: a review. Renew Sustain
Energy Rev 2011;15:292139.
[61] Ahmed H, Mohammed HA, Yusoff M. An overview on heat transfer aug-
mentation using vortex generators and nanouids: approaches and appli-
cations. Renew Sustain Energy Rev 2012;16:595193.
[62] Sidik NAC, Razali SA. Lattice Boltzmann method for convective heat transfer
of nanouidsa review. Renew Sustain Energy Rev 2014;38:86475.
[63] Sidik NAC, Mamat R. Recent progress on lattice Boltzmann simulation of
nanouids: a review. Int Commun Heat Mass Transf 2015;66:1122.
[64] Togun H, Abdulrazzaq T, Kazi S, Badarudin A, Kadhum A, Sadeghinezhad E. A
review of studies on forced, natural and mixed heat transfer to uid and
nanouid ow in an annular passage. Renew Sustain Energy Rev
2014;39:83556.
[65] Narrein K, Mohammed H. Heat transfer and uid ow characteristics in
helically coiled tube heat exchanger (HCTHE) using nanouids: a review. J
Comput Theor Nanosci 2014;11:91127.
[66] Mirmasoumi S, Pourgol-Mohammad M. A Review on experimental and
numerical investigations on using nanouid in volumetric solar energy
collectors. In: Proceedings of the ASME 2014 international mechanical
engineering congress and exposition: American Society of Mechanical
Engineers; 2014. p. V08BT10A048-V08BT10A.
[67] Sureshkumar R, Mohideen ST, Nethaji N. Heat transfer characteristics of nano-
uids in heat pipes: a review. Renew Sustain Energy Rev 2013;20:397410.
[68] Alawi OA, Sidik NAC, Mohammed H, Syahrullail S. Fluid ow and heat
transfer characteristics of nanouids in heat pipes: a review. Int Commun
Heat Mass Transf 2014;56:5062.
[69] Mahdi RA, Mohammed H, Munisamy K, Saeid N. Review of convection heat
transfer and uid ow in porous media with nanouid. Renew Sustain
Energy Rev 2015;41:71534.
[70] ztop HF, Estell P, Yan W-M, Al-Salem K, Or J, Mahian O. A brief review of
natural convection in enclosures under localized heating with and without
nanouids. Int Commun Heat Mass Transf 2015;60:3744.
[71] Sheikholeslami M, Gorji-Bandpy M, Ganji DD. Review of heat transfer
enhancement methods: Focus on passive methods using swirl ow devices.
Renew Sustain Energy Rev 2015;49:44469.
[72] Oztop HF, Abu-Nada E, Varol Y, Al-Salem K. Computational analysis of non-
isothermal temperature distribution on natural convection in nanouid l-
led enclosures. Superlattices Microstruct 2011;49:45367.
[73] Ghasemi B, Aminossadati SM. Periodic natural convection in a nanouid-
lled enclosure with oscillating heat ux. Int J Thermal Sci 2010;49:19.
[74] Kalteh M, Javaherdeh K, Azarbarzin T. Numerical solution of nanouid mixed
convection heat transfer in a lid-driven square cavity with a triangular heat
source. Powder Technol 2014;253:7808.
[75] Ebrahimnia-Bajestan E, Niazmand H, Duangthongsuk W, Wongwises S.
Numerical investigation of effective parameters in convective heat transfer
of nanouids owing under a laminar ow regime. Int J Heat Mass Transf
2011;54:437688.
[76] Kuppusamy NR, Mohammed H, Lim C. Numerical investigation of trapezoidal
grooved microchannel heat sink using nanouids. Thermochim Acta
2013;573:3956.
[77] Nuim Labib M, Nine MJ, Afrianto H, Chung H, Jeong H. Numerical investi-
gation on effect of base uids and hybrid nanouid in forced convective heat
transfer. Int J Thermal Sci 2013;71:16371.
[78] Xuan Y, Li Q. Heat transfer enhancement of nanouids. Int J Heat Fluid Flow
2000;21:5864.
[79] Trisaksri V, Wongwises S. Critical review of heat transfer characteristics of
nanouids. Renew Sustain Energy Rev 2007;11:51223.
1236 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
[80] Godson L, Raja B, Mohan Lal D, Wongwises S. Enhancement of heat transfer
using nanouidsan overview. Renew Sustain Energy Rev 2010;14:62941.
[81] Khanafer K, Vafai K, Lightstone M. Buoyancy-driven heat transfer enhance-
ment in a two-dimensional enclosure utilizing nanouids. Int J Heat Mass
Transf 2003;46:363953.
[82] Lai F-H, Yang Y-T. Lattice Boltzmann simulation of natural convection heat
transfer of Al2O3/water nanouids in a square enclosure. Int J Thermal Sci
2011;50:193041.
[83] Fattahi E, Farhadi M, Sedighi K, Nemati H. Lattice Boltzmann simulation of
natural convection heat transfer in nanouids. Int J Thermal Sci
2012;52:13744.
[84] Abu-Nada E, Masoud Z, Hijazi A. Natural convection heat transfer enhance-
ment in horizontal concentric annuli using nanouids. Int Commun Heat
Mass Transf 2008;35:65765.
[85] Putra N, Roetzel W, Das S. Natural convection of nano-uids. Heat Mass
Transf 2003;39:77584.
[86] Dongsheng W, Ding Y. Natural convective heat transfer of suspensions of
titanium dioxide nanoparticles (nanouids). Nanotechnol IEEE Trans
2006;5:2207.
[87] Wen D, Ding Y. Formulation of nanouids for natural convective heat
transfer applications. Int J Heat Fluid Flow 2005;26:85564.
[88] Chang BH, Mills AF, Hernandez E. Natural convection of microparticle sus-
pensions in thin enclosures. Int J Heat Mass Transf 2008;51:133241.
[89] Nnanna AG. Experimental model of temperature-driven nanouid. J Heat
Transf 2006;129:697704.
[90] Ho CJ, Liu WK, Chang YS, Lin CC. Natural convection heat transfer of alumina-
water nanouid in vertical square enclosures: an experimental study. Int J
Thermal Sci 2010;49:134553.
[91] Abouali O, Falahatpisheh A. Numerical investigation of natural convection of
Al2O3nanouid in vertical annuli. Heat Mass Transf 2009;46:1523.
[92] Haddad Z, Abu-Nada E, Oztop HF, Mataoui A. Natural convection in nano-
uids: Are the thermophoresis and Brownian motion effects signicant in
nanouid heat transfer enhancement? Int J Thermal Sci 2012;57:15262.
[93] Buongiorno J. Convective transport in nanouids. J Heat Transf
2006;128:24050.
[94] Pakravan HA, Yaghoubi M. Analysis of nanoparticles migration on natural
convective heat transfer of nanouids. Int J Thermal Sci 2013;68:7993.
[95] Santra AK, Sen S, Chakraborty N. Study of heat transfer augmentation in a
differentially heated square cavity using copperwater nanouid. Int J
Thermal Sci 2008;47:111322.
[96] Garoosi F, Garoosi S, Hooman K. Numerical simulation of natural convection
and mixed convection of the nanouid in a square cavity using Buongiorno
model. Powder Technol 2014;268:27992.
[97] Keblinski P, Phillpot S, Choi S, Eastman J. Mechanisms of heat ow in sus-
pensions of nano-sized particles (nanouids). Int J Heat Mass Transf
2002;45:85563.
[98] Yu W, Choi S. The role of interfacial layers in the enhanced thermal con-
ductivity of nanouids: a renovated Maxwell model. J Nanoparticle Res
2003;5:16771.
[99] Jang SP, Choi SU. Role of Brownian motion in the enhanced thermal con-
ductivity of nanouids. Appl Phys Lett 2004;84:43168.
[100] Dehnavi R, Rezvani A. Numerical investigation of natural convection heat
transfer of nanouids in a shaped cavity. Superlattices Microstructures.
2012;52:31225.
[101] Versteeg HK, Malalasekera W. An introduction to computational uid
dynamics: the nite volume method. 2nd ed. New York: Pearson Education
Limited; 2007.
[102] Seyf HR, Nikaaein B. Analysis of Brownian motion and particle size effects on
the thermal behavior and cooling performance of microchannel heat sinks.
Int J Thermal Sci 2012;58:3644.
[103] Hung T-C, Yan W-M, Wang X-D, Chang C-Y. Heat transfer enhancement in
microchannel heat sinks using nanouids. Int J Heat Mass Transf
2012;55:255970.
[104] Lee S, Choi S-S, Li S, Eastman J. Measuring thermal conductivity of uids
containing oxide nanoparticles. J Heat Transf 1999;121:2809.
[105] Salman BH, Mohammed HA, Kherbeet AS. Heat transfer enhancement of
nanouids ow in microtube with constant heat ux. Int Commun Heat Mass
Transf 2012;39:1195204.
[106] Mohammed HA, Al-Shamani AN, Sheriff JM. Thermal and hydraulic char-
acteristics of turbulent nanouids ow in a ribgroove channel. Int Commun
Heat Mass Transf 2012;39:158494.
[107] Parsazadeh M, Mohammed HA, Fathinia F. Inuence of nanouid on turbu-
lent forced convective ow in a channel with detached rib-arrays. Int Com-
mun Heat Mass Transf 2013;46:97105.
[108] Mohammed H, Hasan HA, Wahid M. Heat transfer enhancement of nano-
uids in a double pipe heat exchanger with louvered strip inserts. Int
Commun Heat Mass Transf 2013;40:3646.
[109] Shariat M, Moghari RM, Akbarinia A, Rafee R, Sajjadi S. Impact of nanoparticle
mean diameter and the buoyancy force on laminar mixed convection
nanouid ow in an elliptic duct employing two phase mixture model. Int
Commun Heat Mass Transf 2014;50:1524.
[110] Mirmasoumi S, Behzadmehr A. Effect of nanoparticles mean diameter on
mixed convection heat transfer of a nanouid in a horizontal tube. Int J Heat
Fluid Flow 2008;29:55766.
[111] Akbarinia A, Laur R. Investigating the diameter of solid particles effects on a
laminar nanouid ow in a curved tube using a two phase approach. Int J
Heat Fluid Flow 2009;30:70614.
[112] Xie H-q, Wang J-c, Xi T-g, Liu Y. Thermal conductivity of suspensions con-
taining nanosized SiC particles. Int J Thermophys 2002;23:57180.
[113] Murshed S, Leong K, Yang C. Enhanced thermal conductivity of TiO2-water
based nanouids. Int J Thermal Sci 2005;44:36773.
[114] Timofeeva EV, Routbort JL, Singh D. Particle shape effects on thermophysical
properties of alumina nanouids. J Appl Phys 2009;106:01430410.
[115] Vanaki SM, Mohammed HA, Abdollahi A, Wahid MA. Effect of nanoparticle
shapes on the heat transfer enhancement in a wavy channel with different
phase shifts. J Mol Liq 2014;196:3242.
[116] Ooi EH, Popov V. Numerical study of inuence of nanoparticle shape on the
natural convection in Cu-water nanouid. Int J Thermal Sci 2013;65:17888.
[117] Wen D, Ding Y. Experimental investigation into convective heat transfer of
nanouids at the entrance region under laminar ow conditions. Int J Heat
Mass Transf 2004;47:51818.
[118] Hwang KS, Jang SP, Choi SU. Flow and convective heat transfer characteristics
of water-based Al2O3nanouids in fully developed laminar ow regime. Int J
Heat Mass Transf 2009;52:1939.
[119] Wen D, Ding Y. Effect of particle migration on heat transfer in suspensions of
nanoparticles owing through minichannels. Microuid Nanouidics.
2005;1:1839.
[120] Ding Y, Wen D. Particle migration in a ow of nanoparticle suspensions.
Powder Technol 2005;149:8492.
[121] Bahiraei M, Hosseinalipour SM. Particle migration in nanouids considering
thermophoresis and its effect on convective heat transfer. Thermochim Acta
2013;574:4754.
[122] Phillips RJ, Armstrong RC, Brown RA, Graham AL, Abbott JR. A constitutive
equation for concentrated suspensions that accounts for shearinduced
particle migration. Phys Fluids A: Fluid Dyn (1989-1993) 1992;4:3040.
[123] Sohn CH, D Kihm K. Nonhomogenous modeling of nanouidic energy
transport accounting for the thermophoretic migration of nanoparticles
inside laminar pipe ows. J Korean Phys Soc 2009;55:2200.
[124] Heyhat MM, Kowsary F. Effect of Particle Migration on Flow and Convective
Heat Transfer of Nanouids Flowing Through a Circular Pipe. J Heat Transf
2010;132:062401.
[125] Ryzhkov II, Minakov AV. The effect of nanoparticle diffusion and thermo-
phoresis on convective heat transfer of nanouid in a circular tube. Int J Heat
Mass Transf 2014;77:95669.
[126] Pak BC, Cho YI. Hydrodynamic and heat transfer study of dispersed uids with
submicron metallic oxide particles. Exp Heat Transfer Int J 1998;11:15170.
[127] Williams W, Hu L-W, Buongiorno J. Experimental investigation of turbulent
convective heat transfer and pressure loss of alumina/water and zirconia/
water nanoparticle colloids (nanouids) in horizontal tubes. J Heat Transf
2008;130:042412.
[128] Abbasian Arani A, Amani J. Experimental study on the effect of TiO2water
nanouid on heat transfer and pressure drop. Exp Thermal Fluid Sci
2012;42:10715.
[129] Ali M, Zeitoun O. Nanouids forced convection heat transfer inside circular
tubes. Int J Nanoparticles 2009;2:16472.
[130] Balla HH, Abdullah S, Zulkii R, Faizal WM, Sopian K. Effect of Oxides
Nanoparticle Materials on the Pressure Loss and Heat Transfer of nanouids
in circular pipes. J Appl Sci 2012:12.
[131] Huminic G, Huminic A. The cooling performances evaluation of nanouids in
a compact heat exchanger. Fuel 2012;1989:04 1.
[132] Krishna SR, Sivashanmugam P. CFD analysis of heat transfer characteristics of
nanouids in a circular tube tted with helical inserts in laminar ow. IUP J
Chem Eng 2010;2:1934.
[133] Duangthongsuk W, Wongwises S. An experimental study on the heat transfer
performance and pressure drop of TiO2-water nanouids owing under a
turbulent ow regime. Int J Heat Mass Transf 2010;53:33444.
[134] Zhang X, Gu H, Fujii M. Experimental study on the effective thermal con-
ductivity and thermal diffusivity of nanouids. Int J Thermophys
2006;27:56980.
[135] Namburu PK, Kulkarni DP, Misra D, Das DK. Viscosity of copper oxide
nanoparticles dispersed in ethylene glycol and water mixture. Exp Thermal
Fluid Sci 2007;32:397402.
[136] Murshed S, Leong K, Yang C. Investigations of thermal conductivity and
viscosity of nanouids. Int J Thermal Sci 2008;47:5608.
[137] Nguyen C, Desgranges F, Roy G, Galanis N, Mare T, Boucher S, et al. Tem-
perature and particle-size dependent viscosity data for water-based nano-
uidshysteresis phenomenon. Int J Heat Fluid Flow 2007;28:1492506.
[138] Chen H, Ding Y, Lapkin A. Rheological behaviour of nanouids containing
tube/rod-like nanoparticles. Powder Technol 2009;194:13241.
[139] Phuoc TX, Massoudi M. Experimental observations of the effects of shear
rates and particle concentration on the viscosity of Fe2O3deionized water
nanouids. Int J Thermal Sci 2009;48:1294301.
[140] Lee SW, Park SD, Kang S, Bang IC, Kim JH. Investigation of viscosity and
thermal conductivity of SiC nanouids for heat transfer applications. Int J
Heat Mass Transf 2011;54:4338.
[141] Aminossadati S, Ghasemi B. Natural convection cooling of a localised heat
source at the bottom of a nanouid-lled enclosure. Eur J Mech-B/Fluids
2009;28:63040.
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
[142] Ho C-J, Chen M, Li Z. Numerical simulation of natural convection of nanouid
in a square enclosure: effects due to uncertainties of viscosity and thermal
conductivity. Int J Heat Mass Transf 2008;51:450616.
[143] Mohammed HA, Bhaskaran G, Shuaib NH, Abu-Mulaweh HI. Inuence of
nanouids on parallel ow square microchannel heat exchanger perfor-
mance. Int Commun Heat Mass Transf 2011;38:19.
[144] Mohammed HA, Gunnasegaran P, Shuaib NH. Heat transfer in rectangular
microchannels heat sink using nanouids. Int Communn Heat Mass Transf
2010;37:1496503.
[145] Nasrin R, Alim M, Chamkha AJ. Combined convection ow in triangular wavy
chamber lled with waterCuO nanouid: effect of viscosity models. Int
Commun Heat Mass Transf 2012;39:122636.
[146] Oztop HF, Abu-Nada E. Numerical study of natural convection in partially
heated rectangular enclosures lled with nanouids. Int J Heat Fluid Flow
2008;29:132636.
[147] Sheikholeslami M, Gorji-Bandpy M, Ganji DD, Soleimani S, Seyyedi SM.
Natural convection of nanouids in an enclosure between a circular and a
sinusoidal cylinder in the presence of magnetic eld. Int Commun Heat Mass
Transf 2012;39:143543.
[148] Tiwari RK, Das MK. Heat transfer augmentation in a two-sided lid-driven
differentially heated square cavity utilizing nanouids. Int J Heat Mass Transf
2007;50:200218.
[149] Hamilton R, Crosser O. Thermal conductivity of heterogeneous two-
component systems. Ind Eng Chem Fundam 1962;1:18791.
[150] Wasp EJ, Kenny JP, Gandhi RL. Solid-liquid ow slurry pipeline transporta-
tion: Gulf Pub. Co; 1979.
[151] Akbarinia A, Behzadmehr A. Numerical study of laminar mixed convection of
a nanouid in horizontal curved tubes. Appl Thermal Eng 2007;27:132737.
[152] Roy G, Nguyen CT, Lajoie P-R. Numerical investigation of laminar ow and
heat transfer in a radial ow cooling system with the use of nanouids.
Superlattices Microstructures 2004;35:497511.
[153] Akbari M, Behzadmehr A, Shahraki F. Fully developed mixed convection in
horizontal and inclined tubes with uniform heat ux using nanouid. Int J
Heat Fluid Flow 2008;29:54556.
[154] Pathipakka G, Sivashanmugam P. Heat transfer behaviour of nanouids in a
uniformly heated circular tube tted with helical inserts in laminar ow.
Superlattices Microstruct 2010;47:34960.
[155] Popa C, Fohanno S, Polidori G, Nguyen C. Heat transfer enhancement in
mixed convection using waterAl2O3 nanouid. In: Proceedings of the 5th
European Thermal-Sciences Conference. 2008.
[156] Keshavarz Moraveji M, Hejazian M. Modeling of turbulent forced convective
heat transfer and friction factor in a tube for Fe3o4 magnetic nanouid with
computational uid dynamics. Int Commun Heat Mass Transf 2012;39:12936.
[157] Corcione M. Empirical correlating equations for predicting the effective
thermal conductivity and dynamic viscosity of nanouids. Energy Convers
Manag 2011;52:78993.
[158] Koo J, Kleinstreuer C. A new thermal conductivity model for nanouids. J
Nanoparticle Res 2004;6:57788.
[159] Prasher R, Bhattacharya P, Phelan PE. Thermal conductivity of nanoscale
colloidal solutions (nanouids). Phys Rev Lett 2005;94:025901.
[160] Patel HE, Anoop K, Sundararajan T, Das SK. A micro-convection model for
thermal conductivity of nanouids. International Heat Transfer Conference
13: Begel House Inc.; 2006.
[161] Vajjha RS, Das DK. Experimental determination of thermal conductivity of
three nanouids and development of new correlations. Int J Heat Mass Transf
2009;52:467582.
[162] Vajjha RS, Das DK, Kulkarni DP. Development of new correlations for con-
vective heat transfer and friction factor in turbulent regime for nanouids.
Int J Heat Mass Transf 2010;53:460718.
[163] Byk t E. Natural convection of water-based nanouids in an inclined
enclosure with a heat source. Int J Thermal Sci 2009;48:206373.
[164] Haghshenas Fard M, Esfahany MN, Talaie MR. Numerical study of convective
heat transfer of nanouids in a circular tube two-phase model versus single-
phase model. Int Commun Heat Mass Transf 2010;37:917.
[165] Bhattacharya P, Samanta AN, Chakraborty S. Numerical study of conjugate
heat transfer in rectangular microchannel heat sink with Al2O3/H2O nano-
uid. Heat Mass Transf 2009;45:132333.
[166] Fard MH, Talaie MR, Nasr S. Numerical and Experimental Investigation of
Heat Transfer of Zno/Water Nanouid in the Concentric Tube and Plate Heat
Exchangers. Thermal Sci 2011;15:18394.
[167] Yu Z-T, Wang W, Xu X, Fan L-W, Hu Y-C, Cen K-F. A numerical investigation of
transient natural convection heat transfer of aqueous nanouids in a differ-
entially heated square cavity. Int Commun Heat Mass Transf 2011;38:5859.
[168] Sheikholeslami M, Gorji-Bandpy M, Ganji DD. Numerical investigation of
MHD effects on Al2O3water nanouid ow and heat transfer in a semi-
annulus enclosure using LBM. Energy 2013;60:50110.
[169] Ghasemi B, Aminossadati SM. Natural Convection Heat Transfer in an
Inclined Enclosure Filled with a Water-CuO Nanouid. Numerical Heat Transf
Part A: Appl 2009;55:80723.
[170] Yu Z-T, Xu X, Hu Y-C, Fan L-W, Cen K-F. Numerical study of transient
buoyancy-driven convective heat transfer of water-based nanouids in a
bottom-heated isosceles triangular enclosure. Int J Heat Mass Transf
2011;54:52632.
[171] Ting TW, Hung YM, Guo N. Effects of streamwise conduction on thermal
performance of nanouid ow in microchannel heat sinks. Energy Convers
Manag 2014;78:1423.
1237
[172] Ghazvini M, Akhavan-Behabadi MA, Esmaeili M. The effect of viscous dis-
sipation on laminar nanouid ow in a microchannel heat sink. Proc Inst
Mech Eng Part C: J Mechl Eng Sci 2009;223:2697706.
[173] Shahi M, Mahmoudi AH, Talebi F. Numerical study of mixed convective
cooling in a square cavity ventilated and partially heated from the below
utilizing nanouid. Int Commun Heat Mass Transf 2010;37:20113.
[174] Talebi F, Mahmoudi AH, Shahi M. Numerical study of mixed convection ows
in a square lid-driven cavity utilizing nanouid. Int Commun Heat Mass
Transf 2010;37:7990.
[175] Mahmoudi AH, Shahi M, Raouf AH, Ghasemian A. Numerical study of natural
convection cooling of horizontal heat source mounted in a square cavity
lled with nanouid. Int Commun Heat Mass Transf 2010;37:113541.
[176] Kalteh M, Abbassi A, Saffar-Avval M, Frijns A, Darhuber A, Harting J. Experi-
mental and numerical investigation of nanouid forced convection inside a
wide microchannel heat sink. Appl Thermal Eng 2012;36:2608.
[177] Mirzaei M, SaffarAvval M, Naderan H. Heat transfer investigation of laminar
developing ow of nanouids in a microchannel based on Eulerian
Lagrangian approach. Can J Chem Eng 2014.
[178] Kherbeet AS, Mohammed HA, Salman BH. The effect of nanouids ow on
mixed convection heat transfer over microscale backward-facing step. Int J
Heat Mass Transf 2012;55:587081.
[179] Ahmed HE, Mohammed HA, Yusoff MZ. Heat transfer enhancement of
laminar nanouids ow in a triangular duct using vortex generator. Super-
lattices Microstruct 2012;52:398415.
[180] Vanaki SM, Mohammed HA. Numerical study of nanouid forced convection
ow in channels using different shapedtransverse ribs. Int. Commun Heat
Mass Transf 2015. http://dx.doi.org/10.1016/j.icheatmasstransfer.2015.07.014.
[181] Yarmand H, Gharehkhani S, Kazi SN, Sadeghinezhad E, Safaei MR. Numerical
investigation of heat transfer enhancement in a rectangular heated pipe for
turbulent nanouid. Sci World J 2014;2014:19.
[182] Esfe MH, Saedodin S, Wongwises S, Toghraie D. An experimental study on the
effect of diameter on thermal conductivity and dynamic viscosity of Fe/water
nanouids. J Thermal Anal Calorimetry 2015;119:181724.
[183] Sharma KV, Sarma PK, Azmi WH, Mamat R, KadirgamaK. Correlations to
predict friction and forced convection heat transfer coefcients of water
based nanouids for turbulent ow in a tub, vol. 3. 2012. p. 125.
[184] Sundar LS, Ramana EV, Singh MK, Sousa AC. Thermal conductivity and
viscosity of stabilized ethylene glycol and water mixture Al2O3 nanouids for
heat transfer applications: an experimental study. Int Commun Heat Mass
Transf 2014;56:8695.
[185] Hassani S, Saidur R, Mekhilef S, Hepbasli A. A new correlation for predicting
the thermal conductivity of nanouids; using dimensional analysis. Int J Heat
Mass Transf 2015;90:12130.
[186] Minea AA. Uncertainties in modeling thermal conductivity of laminar forced
convection heat transfer with water alumina nanouids. Int J Heat Mass
Transf 2014;68:7884.
[187] Maiga SEB, Palm SJ, Nguyen CT, Roy G, Galanis N. Heat transfer enhancement
by using nanouids in forced convection ows. Int J Heat Fluid Flow
2005;26:53046.
[188] Abu-Nada E. Effects of variable viscosity and thermal conductivity of Al2O3
water nanouid on heat transfer enhancement in natural convection. Int J
Heat Fluid Flow 2009;30:67990.
[189] Lelea D. The performance evaluation of Al2O3/water nanouid ow and heat
transfer in microchannel heat sink. Int J Heat Mass Transf 2011;54:38919.
[190] Sebdani SM, Mahmoodi M, Hashemi SM. Effect of nanouid variable properties
on mixed convection in a square cavity. Int J Thermal Sci 2012;52:11226.
[191] Chon CH, Kihm KD, Lee SP, Choi SU. Empirical correlation nding the role of
temperature and particle size for nanouid (Al2O3) thermal conductivity
enhancement. Appl Phys Lett 2005;87:153107 3.
[192] Hwang KS, Lee J-H, Jang SP. Buoyancy-driven heat transfer of water-based
Al2O3 nanouids in a rectangular cavity. Int J Heat Mass Transf
2007;50:400310.
[193] Einstein A, Frth R. Investigations on the Theory of the Brownian Movement.
Dover Publications; 1956.
[194] Alloui Z, Guiet J, Vasseur P, Reggio M. Natural convection of nanouids in a
shallow rectangular enclosure heated from the side. Can J Chem Eng
2012;90:6978.
[195] Brinkman HC. The viscosity of concentrated suspensions and solution. J
Chem Phys 1952;20:57181.
[196] Mintsa HA, Roy G, Nguyen CT, Doucet D. New temperature dependent
thermal conductivity data for water-based nanouids. Int J Thermal Sci
2009;48:36371.
[197] Masoumi N, Sohrabi N, Behzadmehr A. A new model for calculating the
effective viscosity of nanouids. J Phys D: Appl Phys 2009;42:055501.
[198] Arefmanesh A, Mahmoodi M. Effects of uncertainties of viscosity models for
Al2O3 water nanouid on mixed convection numerical simulations. Int J
Thermal Sci 2011;50:170619.
[199] Chamkha AJ, Abu-Nada E. Mixed convection ow in single-and double-lid
driven square cavities lled with waterAl2O3 nanouid: effect of viscosity
models. Eur J Mech-B/Fluids 2012;36:8296.
[200] Batchelor G. The effect of Brownian motion on the bulk stress in a suspen-
sion of spherical particles. J Fluid Mech 1977;83:97117.
[201] Esmaeilpour M, Abdollahzadeh M. Free convection and entropy generation of
nanouid inside an enclosure with different patterns of vertical wavy walls.
Int J Thermal Sci 2012;52:12736.
1238 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
[202] Roy G, Gherasim I, Nadeau F, Poitras G, Nguyen CT. Heat transfer perfor-
mance and hydrodynamic behavior of turbulent nanouid radial ows. Int J
Thermal Sci 2012;58:1209.
[203] Garoosi F, Bagheri G, Talebi F. Numerical simulation of natural convection of
nanouids in a square cavity with several pairs of heaters and coolers (HACs)
inside. Int J Heat Mass Transf 2013;67:36276.
[204] Bianco V, Manca O, Nardini S. Entropy generation analysis of turbulent
convection ow of Al2O3 water nanouid in a circular tube subjected to
constant wall heat ux. Energy Convers Manag 2014;77:30614.
[205] Kashani S, Ranjbar AA, Mastiani M, Mirzaei H. Entropy generation and nat-
ural convection of nanoparticle-water mixture (nanouid) near water den-
sity inversion in an enclosure with various patterns of vertical wavy walls.
Appl Math Comput 2014;226:18093.
[206] Mohammed HA, Bhaskaran G, Shuaib NH, Saidur R. Numerical study of heat
transfer enhancement of counter nanouids ow in rectangular micro-
channel heat exchanger. Superlattices Microstruct 2011;50:21533.
[207] Esfe MH, Saedodin S. An experimental investigation and new correlation of
viscosity of ZnOEG nanouid at various temperatures and different solid
volume fractions. Experimental Thermal and Fluid Sci 2014;55:15.
[208] Esfe MH, Saedodin S, Mahmoodi M. Experimental studies on the convective
heat transfer performance and thermophysical properties of MgOwater
nanouid under turbulent ow. Exp Thermal Fluid Sci 2014;52:6878.
[209] Koo J, Kleinstreuer C. Impact analysis of nanoparticle motion mechanisms on
the thermal conductivity of nanouids. Int Commun Heat Mass Transf
2005;32:11118.
[210] Seyf HR, Feizbakhshi M. Computational analysis of nanouid effects on
convective heat transfer enhancement of micro-pin-n heat sinks. Int J
Thermal Sci 2012;58:16879.
[211] Aminossadati SM, Ghasemi B. Natural convection of waterCuO nanouid in
a cavity with two pairs of heat sourcesink. Int Commun Heat Mass Transf
2011;38:6728.
[212] Hosseini SS, Abbassi A. Numerical investigation of aspect ratio effect on
thermal parameters in laminar nanouid ow in microchannel heat sink. In:
Prceedings of the 3rd Micro and Nano Flows Conference. Thessaloniki,
Greece, 22-24 August; 2011.
[213] Moghari RM, Mujumdar AS, Shariat M, Talebi F, Sajjadi SM, Akbarinia A.
Investigation effect of nanoparticle mean diameter on mixed convection
Al2O3-water nanouid ow in an annulus by two phase mixture model. Int
Commun Heat Mass Transf 2013;49:2535.
[214] Mokhtari Moghari R, Akbarinia A, Shariat M, Talebi F, Laur R. Two phase
mixed convection Al2O3-water nanouid ow in an annulus. Int J Multiph
Flow 2011;37:58595.
[215] Shariat M, Akbarinia A, Nezhad AH, Behzadmehr A, Laur R. Numerical study
of two phase laminar mixed convection nanouid in elliptic ducts. Applied
Thermal Eng 2011;31:234859.
[216] Fluent 6.2 User Manual: Fluent Incorporated; 2006.
[217] Akhtari M, Haghshenasfard M, Talaie M. Numerical and Experimental
Investigation of Heat Transfer of -Al2O3/Water Nanouid in Double Pipe and
Shell and Tube Heat Exchangers. Numerical Heat Transf Part A: Appl
2013;63:94158.
[218] Izadi M, Behzadmehr A, Jalali-Vahida D. Numerical study of developing
laminar forced convection of a nanouid in an annulus. Int J Thermal Sci
2009;48:211929.
[219] Anderson JD. Computational uid dynamics. New York: McGraw-Hill; 1995.
[220] Lu G, Wang J. Experimental investigation on heat transfer characteristics of
water ow in a narrow annulus. Appl Thermal Eng 2008;28:813.
[221] Moraveji MK, Darabi M, Haddad SMH, Davarnejad R. Modeling of convective heat
transfer of a nanouid in the developing region of tube ow with computational
uid dynamics. Int Commun Heat Mass Transf 2011;38:12915.
[222] Anoop KB, Sundararajan T, Das SK. Effect of particle size on the convective
heat transfer in nanouid in the developing region. Int J Heat Mass Transf
2009;52:218995.
[223] Hussein AM, Sharma K, Bakar R, Kadirgama K. The effect of cross sectional
area of tube on friction factor and heat transfer nanouid turbulent ow. Int
Commun Heat Mass Transf 2013;47:4955.
[224] Launder BE, Spaulding DB. Mathematical models of turbulence. New York,
NY: Academic Press; 1972.
[225] Mohammed H, Gunnasegaran P, Shuaib N. The impact of various nanouid
types on triangular microchannels heat sink cooling performance. Int Com-
mun Heat Mass Transf 2011;38:76773.
[226] Chein R, Chuang J. Experimental microchannel heat sink performance studies
using nanouids. Int J Thermal Sci 2007;46:5766.
[227] Rashmi W, Khalid M, Ismail AF, Saidur R, Rashid A. Experimental and
numerical investigation of heat transfer in CNT nanouids. J Exp Nanosci
2015;10:54563.
[228] Meng X, Li Y. Numerical study of natural convection in a horizontal cylinder
lled with water-based alumina nanouid. Nanoscale Res Lett 2015;10:110.
[229] Rakhsha M, Akbaridoust F, Abbassi A, Safar-Avval M. Experimental and
numerical investigation of turbulent forced convection ow of nano-uid in
helical coiled tubes at constant surface temperature. Powder Technol 2015.
[230] Kumar PM, Palanisamy K, Kumar J, Tamilarasan R, Sendhilnathan S. CFD
analysis of heat transfer and pressure drop in helically coiled heat exchan-
gers using Al2O3/water nanouid. J Mech Sci Technol 2015;29:697705.
[231] Manca O, Nardini S, Ricci D. A numerical study of nanouid forced convection
in ribbed channels. Appl Thermal Eng 2012;37:28092.
[232] Menter FR. Two-equation eddy-viscosity turbulence models for engineering
applications. AIAA J 1994;32:1598605.
[233] Rostamani M, Hosseinizadeh S, Gorji M, Khodadadi J. Numerical study of
turbulent forced convection ow of nanouids in a long horizontal duct
considering variable properties. Int Commun Heat Mass Transf
2010;37:142631.
[234] Bejan A. Heat transfer. New York: John Wiley & Sons, Inc; 1993.
[235] Sarkar J. A critical review on convective heat transfer correlations of nano-
uids. Renew Sustain Energy Rev 2011;15:32717.
[236] Yakhot V, Orszag SA. Renormalization group analysis of turbulence. I. Basic
theory. J Sci Comput 1986;1:351.
[237] Shih T-H, Liou WW, Shabbir A, Yang Z, Zhu J. A new k- eddy viscosity model
for high reynolds number turbulent ows. Comput Fluids 1995;24:22738.
[238] Wilcox DC. Turbulence modeling for CFD: DCW industries La Canada, CA;
1998.
[239] Spalart P, Allmaras S. A one-equation turbulence model for aerodynamic
ows. 1992.
[240] Ridluan A, Eiamsa-ard S, Promvonge P. Numerical simulation of 3D turbulent
isothermal ow in a vortex combustor. Int Commun Heat Mass Transf
2007;34:8609.
[241] Spall RE, Richards A, McEligot DM. An Assessment of K and V2F Turbu-
lence Models for Strongly Heated Internal Gas Flows. Numerical Heat Transf
Part A: Appl 2004;46:83149.
[242] Bayat J, Nikseresht AH. Thermal performance and pressure drop analysis of
nanouids in turbulent forced convective ows. Int J Thermal Sci
2012;60:23643.
[243] Kim D, Kwon Y, Cho Y, Li C, Cheong S, Hwang Y, et al. Convective heat
transfer characteristics of nanouids under laminar and turbulent ow
conditions. Curr Appl Phys 2009;9:11923.
[244] Xuan Y, Roetzel W. Conceptions for heat transfer correlation of nanouids.
Int J Heat Mass Transf 2000;43:37017.
[245] Heris SZ, Esfahany MN, Etemad G. Numerical investigation of nanouid
laminar convective heat transfer through a circular tube. Numerical Heat
Transf Part A: Appl 2007;52:104358.
[246] Kumar S, Prasad SK, Banerjee J. Analysis of ow and thermal eld in nano-
uid using a single phase thermal dispersion model. Appl Math Model
2010;34:57392.
[247] Mokmeli A, Saffar-Avval M. Prediction of nanouid convective heat transfer
using the dispersion model. Int J Thermal Sci 2010;49:4718.
[248] Akbaridoust F, Rakhsha M, Abbassi A, Saffar-Avval M. Experimental and
numerical investigation of nanouid heat transfer in helically coiled tubes at
constant wall temperature using dispersion model. Int J Heat Mass Transf
2013;58:48091.
[249] Mojarrad MS, Keshavarz A, Shokouhi A. Nanouids thermal behavior analysis
using a new dispersion model along with single-phase. Heat Mass Transf
2013;49:133343.
[250] Ishii M. Thermo-uid dynamic theory of two-phase ow. NASA STI/Recon
Tech Rep A 1975;75:29657.
[251] Crowe C, Troutt T, Chung J. Numerical models for two-phase turbulent ows.
Annu Rev Fluid Mech 1996;28:1143.
[252] Manninen M, Taivassalo V, Kallio S. On the mixture model for multiphase
ow. 1996.
[253] Xu J, Rouelle A, Smith K, Celik D, Hussaini M, Van Sciver S. Two-phase ow of
solid hydrogen particles and liquid helium. Cryogenics. 2004;44:45966.
[254] Moraveji MK, Ardehali RM. CFD modeling (comparing single and two-phase
approaches) on thermal performance of Al2O3/water nanouid in mini-
channel heat sink. Int Commun Heat Mass Transf 2013;44:15764.
[255] El-Batsh H, Doheim M, Hassan A. On the application of mixture model for
two-phase ow induced corrosion in a complex pipeline conguration. Appl
Math Model 2012;36:568699.
[256] Ghaffari O, Behzadmehr A, Ajam H. Turbulent mixed convection of a nano-
uid in a horizontal curved tube using a two-phase approach. Int Commun
Heat Mass Transf 2010;37:15518.
[257] Schiller L, Naumann Z. A drag coefcient correlation. Vdi Zeitung 1935;77:51.
[258] Behzadmehr A, Saffar-Avval M, Galanis N. Prediction of turbulent forced
convection of a nanouid in a tube with uniform heat ux using a two phase
approach. Int J Heat Fluid Flow 2007;28:2119.
[259] Xuan Y, Li Q. Investigation on convective heat transfer and ow features of
nanouids. J Heat Transf 2003;125:1515.
[260] Mirmasoumi S, Behzadmehr A. Numerical study of laminar mixed convection
of a nanouid in a horizontal tube using two-phase mixture model. Appl
Thermal Eng 2008;28:71727.
[261] Saberi M, Kalbasi M, Alipourzade A. Numerical Study of Forced Convective Heat
Transfer of Nanouids inside a Vertical Tube. Int J Thermal Technol 2013;3:105.
[262] Hejazian M, Moraveji MK. A comparative analysis of single and two-phase
models of turbulent convective heat transfer in a tube for TiO2 nanouid
with CFD. Numerical Heat Transf Part A: Appl 2013;63:795806.
[263] Sajadi A, Kazemi M. Investigation of turbulent convective heat transfer and
pressure drop of TiO2/water nanouid in circular tube. Int Commun Heat
Mass Transf 2011;38:14748.
[264] Bianco V, Manca O, Nardini S. Numerical investigation on nanouids tur-
bulent convection heat transfer inside a circular tube. Int J Thermal Sci
2011;50:3419.
[265] Sakhani H, Abbassi A. Effects of tube attening on the uid dynamic and heat
transfer performance of nanouids. Adv Powder Technol 2014;25:113241.
 Sh.M. Vanaki et al. / Renewable and Sustainable Energy Reviews 54 (2016) 12121239
[266] Lot R, Saboohi Y, Rashidi A. Numerical study of forced convective heat
transfer of nanouids: comparison of different approaches. Int Commun
Heat Mass Transf 2010;37:748.
[267] Moghadassi A, Ghomi E, Parvizian F. A numerical study of water based Al2O3
and Al2O3Cu hybrid nanouid effect on forced convective heat transfer. Int J
Thermal Sci 2015;92:507.
[268] Naphon P, Nakharintr L. Numerical Investigation of Laminar Heat Transfer of
Nanouid-Cooled Mini-Rectangular Fin Heat Sinks. J Eng Phys Thermophys
2015;88:66675.
[269] Corcione M, Cianfrini M, Quintino A. Enhanced natural convection heat
transfer of nanouids in enclosures with two adjacent walls heated and the
two opposite walls cooled. Int J Heat Mass Transf 2015;88:90213.
[270] Akbari M, Galanis N, Behzadmehr A. Comparative analysis of single and two-
phase models for CFD studies of nanouid heat transfer. Int J Thermal Sci
2011;50:134354.
[271] Fani B, Abbassi A, Kalteh M. Effect of nanoparticles size on thermal perfor-
mance of nanouid in a trapezoidal microchannel-heat-sink. Int Commun
Heat Mass Transf 2013;45:15561.
[272] Kalteh M, Abbassi A, Saffar-Avval M, Harting J. EulerianEulerian two-phase
numerical simulation of nanouid laminar forced convection in a micro-
channel. Int J Heat Fluid Flow 2011;32:10716.
[273] Oosthuizen PH, Carscallen WE. Introduction to compressible uid flow. 2nd
ed . Boca Raton, FL: Taylor & Francis; 2013.
[274] Beheshti A, Moraveji MK, Hejazian M. Comparative Numerical Study of
Nanouid Heat Transfer through an Annular Channel. Numerical Heat Transf
Part A: Appl 2015;67:10017.
[275] Akbari M, Galanis N, Behzadmehr A. Comparative assessment of single and
two-phase models for numerical studies of nanouid turbulent forced con-
vection. Int J Heat Fluid Flow 2012;37:13646.
[276] Bianco V, Chiacchio F, Manca O, Nardini S. Numerical investigation of nanouids
forced convection in circular tubes. Appl Thermal Eng 2009;29:363242.
[277] Keshavarz Moraveji M, Esmaeili E. Comparison between single-phase and
two-phases CFD modeling of laminar forced convection ow of nanouids in
a circular tube under constant heat ux. Int Commun Heat Mass Transf
2012;39:1297302.
[278] Zeinali Heris S, Nasr Esfahany M, Etemad SG. Experimental investigation of
convective heat transfer of Al2O3/water nanouid in circular tube. Int J Heat
Fluid Flow 2007;28:20310.
[279] He Y, Men Y, Zhao Y, Lu H, Ding Y. Numerical investigation into the con-
vective heat transfer of TiO2 nanouids owing through a straight tube
under the laminar ow conditions. Appl Thermal Eng 2009;29:196572.
[280] Bahremand H, Abbassi A, Saffar-Avval M. Experimental and numerical
investigation of turbulent nanouid ow in helically coiled tubes under
constant wall heat ux using EulerianLagrangian approach. Powder Technol
2015;269:93100.
[281] Mahdavi M, Sharifpur M, Meyer J. CFD modelling of heat transfer and pres-
sure drops for nanouids through vertical tubes in laminar ow by Lagran-
gian and Eulerian approaches. Int J Heat Mass Transf 2015;88:80313.
1239
[282] Behroyan I, et al. Turbulent forced convection of Cuwater nanouid:
CFDmodel comparison. Int Commun Heat Mass Transf 2015. http://dx.doi.
org/10.1016/j.icheatmasstransfer.2015.07.014.
[283] Zhang J. Lattice Boltzmann method for microuidics: models and applica-
tions. Microuid Nanouidics 2011;10:128.
[284] Jeong HK, Yoon HS, Ha MY, Tsutahara M. An immersed boundary-thermal
lattice Boltzmann method using an equilibrium internal energy density
approach for the simulation of ows with heat transfer. J Comput Phys
2010;229:252643.
[285] Xuan Y, Yao Z. Lattice Boltzmann model for nanouids. Heat Mass Transf
2005;41:199205.
[286] Sheikholeslami M, Gorji-Bandpy M, Seyyedi SM, Ganji DD, Rokni HB, Solei-
mani S. Application of LBM in simulation of natural convection in a nanouid
lled square cavity with curve boundaries. Powder Technol 2013;247:8794.
[287] Nemati H, Farhadi M, Sedighi K, Fattahi E, AAR Darzi. Lattice Boltzmann
simulation of nanouid in lid-driven cavity. Int Commun Heat Mass Transf
2010;37:152834.
[288] Nabavitabatabayi M, Shirani E, Rahimian MH. Investigation of heat transfer
enhancement in an enclosure lled with nanouids using multiple relaxation time
lattice Boltzmann modeling. Intl Commun Heat Mass Transf 2011;38:12838.
[289] Kefayati GR, Hosseinizadeh SF, Gorji M, Sajjadi H. Lattice Boltzmann simu-
lation of natural convection in tall enclosures using water/SiO2 nanouid. Int
Commun Heat Mass Transf 2011;38:798805.
[290] Xuan Y, Yu K, Li Q. Investigation on ow and heat transfer of nanouids by the
thermal Lattice Boltzmann model. Prog Comput Fluid Dyn Int J 2005;5:139.
[291] Zarghami A, Ubertini S, Succi SF. lattice Boltzmann modeling of thermal
transport in nanouids. Comput Fluids 2013;77:5665.
[292] Yang Y-T, Lai F-H. Numerical study of ow and heat transfer characteristics of
alumina-water nanouids in a microchannel using the lattice Boltzmann
method. Int Commun Heat Mass Transf 2011;38:60714.
[293] Mohammadi Pirouz M, Farhadi M, Sedighi K, Nemati H, Fattahi E. Lattice
Boltzmann simulation of conjugate heat transfer in a rectangular channel
with wall-mounted obstacles. Sci Iran 2011;18:21321.
[294] McNab GS, Meisen A. Thermophoresis in liquids. J Colloid Interface Sci
1973;44:33946.
[295] Piazza R, Parola A. Thermophoresis in colloidal suspensions. J Phys: Condens
Matter 2008;20:153102.
[296] Wrger A. Thermal non-equilibrium transport in colloids. Rep Prog Phys
2010;73:126601.
[297] Galliero G, Volz S. Thermodiffusion in model nanouids by molecular
dynamics simulations. J Chem Phys 2008;128:064505.
[298] Bahiraei M, Mostafa Hosseinalipour S, Hangi M. Prediction of convective heat
transfer of Al2O3-water nanouid considering particle migration using
neural network. Eng Comput 2014;31:84363.
[299] Sheikhzadeh GA, Dastmalchi M, Khorasanizadeh H. Effects of nanoparticles
transport mechanisms on Al2O3water nanouid natural convection in a
square enclosure. Int J Thermal Sci 2013;66:5162.