Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 Contents lists available at ScienceDirect

2
3
4 Journal of Magnetism and Magnetic Materials
5
6
journal homepage: www.elsevier.com/locate/jmmm
7
8
9
10
11
12
Flow and heat transfer characteristics of magnetic nanofluids: A review
13
14 Q1 Mehdi Bahiraei a, Morteza Hangi b
15 Q4 a
Mechanical Engineering Department, School of Energy, Kermanshah University of Technology, Kermanshah, Iran
b
16 Q3 School of Mechanical Engineering, Iran University of Science and Technology, Tehran, Iran
17
18
19 art ic l e i nf o a b s t r a c t
20
21 Article history: Magnetic nanofluids (MNFs) are suspensions which are comprised of a non-magnetic base fluid and
22 Received 6 April 2014 magnetic nanoparticles. In this modern set of suspensions which can be called smart or functional fluids,
Received in revised form fluid flow, particles movement and heat transfer process can be controlled by applying magnetic fields.
23
29 July 2014
24 Regarding unique characteristics of MNFs, studies in this field have witnessed a phenomenal growth.
This paper reviews and summarizes recent investigations implemented on MNFs including those
25
Keywords: conducted on thermophysical properties, natural convection, forced convection, boiling as well as their
26 Magnetic nanofluids practical applications. Moreover, this review identifies the challenges and opportunities for future
27 Natural convection research.
28 Forced convection
& 2014 Elsevier B.V. All rights reserved.
29 Experimental
Numerical
30
Applications
31
32
33
34 1. Introduction fluid flow, heat transfer and particles movement by applying
35 magnetic fields and consequently, they have a great potential for
36 Recent technological advancements in the field of electronics being used in various fields such as bioengineering, electronics as
37 and thermal systems have led to an ever increasing demand for well as thermal engineering [11–13]. Moreover, there is a specific
38 heat transfer systems with higher efficiencies. Thus far, numerous type of MNFs, called temperature-sensitive magnetic fluids, whose
39 studies have been carried out on heat transfer augmentation using magnetization is considerably dependent on temperature and
40 both active and passive methods. In active methods such as hence, they are considered as a promising fluid in miniature
41 mechanical agitating, rotating and vibration, there is a need to energy conversion and heat transport systems [14,15].
42 apply external energy, while passive ones include methods such as In the field of heat transfer, using MNFs has attracted the
43 improvement of fluid thermal properties and surface geometry. attention of many researchers in recent years because of their
44 One of the passive methods that has attracted the attention of capability to be controlled under magnetic fields. The objective of
45 many researchers in recent years is application of nanofluids. After this paper is to present an overview of literature dealing with
46 Choi [1] who introduced the novel idea of adding nanoparticles to recent investigations on MNFs in different fields of thermal
47 base fluid in order to improve thermal characteristics, many engineering. It is expected that this in-depth review will provide
48 researchers applied nanofluids for heat transfer enhancement in a framework for not only investigating the current status but also
49 various thermal systems [2–8]. specifying the future direction of studies on MNFs.
50 Among investigations in the field of nanofluids, some studies
51 have focused on MNFs. MNFs or ferrofluids are suspensions of a
52 non-magnetic base fluid and magnetic nanoparticles which are
53 coated with surfactant layers such as oleic acid to provide proper
54 stability [9,10]. Magnetic nanoparticles used in MNFs are usually 2. Properties of MNFs
55 prepared in different sizes and forms from metal materials
56 (ferromagnetic materials) such as iron, nickel, cobalt, as well as Beyond doubt, flow and heat transfer characteristics of a fluid
57 their oxides such as spinel-type ferrites, magnetite (Fe3O4), and are profoundly affected by its properties such as thermal conduc-
58 so forth. tivity and viscosity. In this regard, numerous studies in the field of
59 The main feature of this type of nanofluids is that apart from nanofluids assessed their properties [16–18]. As far as MNFs are
60 improvement of thermal properties, they possess both magnetic concerned, some investigations evaluated their rheological behavior
61 properties similar to other magnetic materials and flowability like and thermophysical properties in the absence and presence of
62 other fluids. Such unique characteristic makes it possible to control magnetic fields, which are assessed in this section.
63
64 http://dx.doi.org/10.1016/j.jmmm.2014.08.004
65 0304-8853/& 2014 Elsevier B.V. All rights reserved.
66

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 2 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2.1. In the absence of magnetic field They observed that thermal conductivity of nanofluids gradually
2 approaches the value predicted by the Maxwell equation by increas-
3 Investigations on MNFs in the absence of magnetic fields ing the viscosity of the base fluid. This means that at high viscosities,
4 demonstrate that the their thermophysical properties are affected the experimental results obtained for thermal conductivity of nano-
5 by various parameters such as particle size, volume fraction of fluid become very close to those predicted by the Maxwell model. In
6 nanoparticles, base fluid properties, chemical composition of mag- fact, Maxwell model does not consider the role of Brownian motion
7 netic nanoparticles, temperature, particle coating layer, and so forth. and therefore, at low viscosities, due to high intensity of Brownian
8 In a great deal of the conducted studies, the effects of tempe- motion, the results obtained from experiments on nanofluids show
9 rature and concentration of nanoparticles on the properties of higher values in comparison with those obtained from the Maxwell
10 MNFs have been assessed. Syam Sundar et al. [19] investigated the model. However, with viscosity increment, due to decrease in the
11 effective thermal conductivity and viscosity of Fe3O4/water MNF effect of Brownian motion, the role of Brownian motion in the value
12 experimentally. Experiments were conducted in the concentration of thermal conductivity diminishes. It was observed that the viscosity
13 range of 0 to 2% and the temperature range of 20 1C to 60 1C. They around 100 cP is a critical value of demarcation. In other words,
14 concluded that the thermal conductivity enhances with increase in below 100 cP of viscosity, the Brownian motion is active, so the
15 particle concentration and temperature. It was also demonstrated thermal conductivity of nanofluids increases and consequently, the
16 that the nanofluid exhibits Newtonian behavior under the tested Maxwell model underpredicts the experimental results. Over 100 cP
17 concentration range. They proposed models of effective thermal of viscosity, the Brownian motion becomes inactive and the thermal
18 conductivity and viscosity of the MNF as below: conductivity of nanofluids obeys the Maxwell theory.
19 In Fig. 1, a comparison between the values of enhancement in
knf ¼ kbf ð1 þ 10:5φÞ0:1051 ð1Þ
20 thermal conductivity obtained by different researchers for various
21  MNFs has been depicted in the absence of magnetic field. The
φ 6:356
22 μnf ¼ μbf 1 þ ð2Þ enhancement parameter is defined as below:
23 12:5
knf  kbf
24 where subscripts nf and bf respectively refer to nanofluid and base Enh: ¼  100 ð3Þ
kbf
25 fluid, k is thermal conductivity, μ is viscosity and φ represents
26 volume concentration. In order to consider the effect of tempera-
27 ture, thermal conductivity and viscosity of the base fluid should be It can be observed from the figure that adding magnetic nano-
28 substituted in the correlations at that particular temperature. particles increases the thermal conductivity of the base fluid and
29 Thermal conductivity of Fe3O4 nanoparticles in ethylene glycol in some cases, the value of enhancement has non-linear relation
30 was determined experimentally as a function of temperature and with volume fraction of particles. The reason is that clustering is
31 concentration by Pastoriza-Gallego et al. [20]. They found that the more intense at higher concentrations and therefore, the average
32 enhancement of the thermal conductivity increases almost line- size of clusters increases with concentration increment and this
33 arly with the concentration while it is nearly temperature- causes a change in the increasing trend of thermal conductivity.
34 independent. It means that with enlargement of clusters, the slope of variations
35 Abareshi et al. [21] prepared MNFs by dispersing the Fe3O4 in thermal conductivity in terms of concentration decreases.
36 nanoparticles in water in the presence of tetramethyl ammonium Other studies in which the relation between thermal conductivity
37 hydroxide as a dispersant and measured the thermal conductivity and concentration is linear even at high concentrations indicate
38 of the nanofluid as a function of volume fraction and temperature. that the prepared nanofluids have proper stability even at great
39 The highest value of enhancement in thermal conductivity was concentrations such that clustering occurs lately in them. This can
40 reported to be 11.5% for the nanofluid with 3 vol% of nanoparticles be due to applying appropriate surfactants or correct use of
41 at 40 1C. ultrasonic waves. Therefore, the value of concentration above
42 Regarding the importance of the size of suspended magnetic
43 nanoparticles in the base fluid on the MNFs properties, some
44 researchers have investigated the effect of nanoparticle size as
45 well as cluster size of the nanoparticles. The result which has been
46 reported in most of such studies is the reduction of the thermal
47 conductivity of ferrofluids with particles enlargement. Wang et al.
48 [22] performed an experimental study to examine the effect of
49 Fe3O4 nanoparticle size on the thermal conductivities of heat
50 transfer oils. The results showed that the MNFs exhibit higher
51 thermal conductivity than heat transfer oils, and the values of
52 enhancement increase with reducing the particle size. In addition,
53 they found for the first time that the viscosities of all nanofluids
54 are remarkably lower than that of the base fluid.
55 Hong et al. [23] focused on the effect of the clustering of
56 nanoparticles on the thermal conductivity of MNFs. The thermal
57 conductivity increased nonlinearly as the volume fraction of
58 nanoparticles increased. The nonlinearity was attributed to the
59 rapid clustering of nanoparticles in condensed nanofluids.
60 In addition, the effect of base fluid on ferrofluids properties has
61 been evaluated in some studies. For instance, Tsai et al. [24]
62 performed an experimental study to investigate the effect of viscosity
63 of the base fluid on the thermal conductivity of nanofluids in which
64 Fe3O4 nanoparticles are suspended in the base fluid composed of
65 diesel oil and polydimethylsiloxane. Viscosity of the base fluid was Fig. 1. Comparison between the values of enhancement in thermal conductivity
66 varied by changing the volumetric fractions between both fluids. obtained by different researchers for various MNFs in the absence of magnetic field.

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

1 which the effect of clustering becomes important is profoundly
2 dependent on the process of nanofluid preparation. For instance,
3 in the study conducted by Abareshi et al. [21], clustering nearly
4 occurs at the concentrations above 1 vol%, while in the investiga-
5 tion performed by Zhu et al. [25], this phenomenon nearly takes
6 place at the concentrations above 2 vol%.
7 In the studies which have been conducted thus far on the
8 properties of MNFs in the absence of magnetic fields, due to the
9 limited number of measurements, it is not possible to extract
10 general correlations for their properties, particularly thermal con-
11 ductivity and viscosity, in which the effects of different parameters
12 such as concentration and size of the magnetic nanoparticles as
13 well as temperature are considered simultaneously. A review on the
14 literature shows that for nanofluids containing nanoparticles such
15 as Al2O3 and TiO2, there are comprehensive correlations to calculate
16 their effective properties such that they have been widely used in
17 studies in which nanofluids behavior has been simulated and with
18 respect to the generality of these correlations, relatively precise
19 results have been obtained in the conducted simulations. Having
20 this in mind and also the ever increasing improvements in the field
21 of MNFs, deriving such correlations which can be properly used by
22 other researchers in this field is strongly felt.
23
24 2.2. In the presence of magnetic field
25
26 Regarding the fact that applying magnetic fields can affect the
27 thermophysical properties of MNFs, some researchers have eval-
28 uated their properties in the presence of magnetic fields. It means
29 that in addition to the aforementioned parameters, strength and
30 orientation of the applied magnetic field can also have effects on
31 the properties of MNFs.
32 In some studies, the effect of applying magnetic fields with
33 different directions and strengths has been assessed on thermo-
34 physical properties. Results reveal that it is possible to alter the
35 thermophysical properties of MNFs by changing the direction and
36 strength of the magnetic field. Krichler and Odenbach [28] applied
37 an improved measuring device based on the plane heat source to
38 measure the thermal conductivity of a magnetite-based ferrofluid
39 by altering strength and direction of an external magnetic field
40 relative to heat flux. They observed that for parallel alignment of
41 the magnetic field, thermal conductivity increases with increasing
42 magnetic field strength while for perpendicular alignment, the
43 thermal conductivity has an inverse relationship to the magnetic
44 field strength.
45 Chiu et al. [29] evaluated the dependency of specific heat of a
46 water-based MNF on temperature and magnetic field. The results
47 demonstrated that the specific heat is almost independent from
48 temperature in the presence of lower-field strengths, but
49 decreases significantly upon increasing temperature under greater
50 fields. They reported that the reason of such an observation is due
51 to the reduction in the degrees of freedom of the ferrite particles
52 in the magnetic fluid under greater magnetic fields.
53 Shima and Philip [30] studied the effect of magnetic field on
54 thermal and rheological properties of MNFs and demonstrated
55 that these properties can be tuned from low to very high values by
56 varying the magnetic field strength and its orientation.
57 In some investigations, the thermophysical properties of MNFs
58 have been measured in the presence of magnetic fields and
59 compared with those in the absence of magnetic fields. The results
60 reported in such studies show that by applying magnetic fields, the
61 thermophysical properties of ferrofluids can be changed consider-
62 ably. Baby and Ramaprabhu [31] dispersed iron oxide coated multi-
63 walled carbon nanotubes (Fe2O3/MWNT) in water to prepare MNF.
64 The thermal conductivity of the nanofluid was investigated with
65 and without magnetic field and it was demonstrated that without
66 magnetic field, there is only a small enhancement in the thermal
conductivity, while a sudden increase in thermal conductivity is
observed in the presence of magnetic field. They stated that thermal
conductivity enhancement in the presence of magnetic field is due
to the aligned nature of Fe2O3/MWNT in the direction of magnetic
field which gives more contact between the nanotubes.
Parekh and Lee [32] synthesized two different MNFs compris-
ing of magnetite (i.e. Fe3O4) and Mn–Zn ferrite nanoparticles in
hydrocarbon oil. They evaluated the thermal conductivity of these
two nanofluids as a function of volume fraction under transverse
magnetic field. The results demonstrated that in the absence of
magnetic field, the enhancement in thermal conductivity of Mn–
Zn nanofluid is higher than that of magnetite nanofluid, but in
presence of magnetic field, the magnetite nanofluid shows further
enhancement in thermal conductivity while no change has been
observed for Mn–Zn ferrite.
Gavili et al. [33] experimentally investigated the thermal con-
ductivity of ferrofluid containing Fe3O4 nanoparticles suspended in
de-ionized water. Their results showed that in the absence of
magnetic field, enhancement in the thermal conductivity is not
noticeable while applying a magnetic field increases the thermal
conductivity and they reported more than 200% enhancement at
maximum value under the effect of magnetic field.
Li et al. [34] carried out experimental investigations to measure
the viscosity and thermal conductivity of the aqueous Fe and Fe3O4
MNFs in the absence and presence of the magnetic field. They
evaluated the effects of the particle concentration, concentration
of surfactants and the magnetic field strength as well as its
orientation. It was found that the viscosity first increases with
the magnetic field and finally approaches a constant value due to
the magnetization saturation of the magnetic fluid. In addition,
they observed that the thermal conductivity of the magnetic fluid
increases with the strength of the magnetic field being parallel to
the temperature gradient.
In some studies conducted on the thermophysical properties of
MNFs, structural changes caused by applying magnetic fields, such
as formation of chain-like structures, have been introduced as one
of the effective factors. Nkurikiyimfura et al. [35] investigated the
effect of chain-like magnetic nanoparticles aggregates on the
thermal conductivity of MNFs under the effect of magnetic field.
They indicated that for a magnetic field parallel to temperature
gradient, the thermal conductivity component along the magnetic
field significantly enhances due to the chain-like magnetic nano-
particles aggregates induced by the magnetic field.
Fu and Gao [36] established the two-step homogenization
method to study the anisotropic thermal conductivity of MNFs
theoretically by taking into account the magnetic field effect and
the physical anisotropy. The numerical results revealed that the
aspect ratio of chain-like aggregated clusters plays an important
role in enhancement of anisotropic thermal conductivity. Further-
more, there was a good agreement between their predictions and
the experimental results.
Philip et al. [37] evaluated the thermal conductivity of kerosene–
Fe3O4 and hexadecane–Fe3O4 nanofluids in the presence of an
external magnetic field. They observed that when the magnetic
field is parallel to temperature gradient, the thermal conductivity
starts to increase. The highest values of enhancement were reported
to be about 300% and 216% for kerosene–Fe3O4 and hexadecane–
Fe3O4 nanofluids, respectively, due to chain-like structures of
magnetic particles in the presence of the magnetic field.
Parekh and Lee [38] experimentally studied the thermal con-
ductivity of magnetite nanofluid as a function of transverse
magnetic field. Almost 30% enhancement in thermal conductivity
was observed for 4.7% volume fraction under magnetic field. They
stated that this result is explained on the basis of formation of
continuous three-dimensional zipperlike structure of magnetic
nanoparticles inside the magnetic fluid.

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 4 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 As can be seen, the thermophysical properties of MNFs sig- In order to have a better understanding of the thermomagnetic
2 nificantly depend on the type and conditions of the applied convection, a more detailed description is given in the following
3 magnetic field and thus, for such type of nanofluids, models of based on Fig. 2. This figure depicts a horizontal layer of a MNF
4 properties for cases with and without application of magnetic between two walls, where the temperature of the upper wall is
5 fields can be different from each other. In addition, in most of the higher than that of the lower one, such that a vertical temperature
6 studies conducted on properties of MNFs, both in the presence and gradient (∇T) exists in the ferrofluid. In addition, the MNF is under
7 absence of magnetic field, Fe3O4 magnetic nanoparticles have been the effect of a uniform magnetic field (H) collinear with the
8 used. However, recent furtherance in the field of nanotechnology temperature gradient. Regarding the fact that the magnetization
9 has resulted in production of nano-ferrites which possess better of MNF (M) is dependent on temperature and decreases with
10 magnetic properties (e.g. saturation magnetization of Mn–Zn temperature increment, the ferrofluid magnetization near the cold
11 ferrites is about 1:2  106 A=mwhile that of Fe3O4 is about wall is higher than that near the hot wall and thus, the fluid
12 5  105 A=m) as well as higher thermal conductivity in compar- experiences a magnetization with spatial vertical variation. There-
13 ison with Fe3O4 nanoparticles (e.g. thermal conductivity of Mn–Zn fore, an inner magnetic field gradient (∇H i ) is created as below:
14 ferrites is about 29 W/mK while that of Fe3O4 is about 6 W/mK)
∇H i ¼ H  D∇M ð5Þ
15 and consequently, properties of nanofluids produced from nano-
16 ferrites can be attracted more attention in future studies. Further- where D represents the demagnetization factor. As a result, the
17 more, as mentioned above, formation of chain-like structures has fluid element with higher magnetization, which is near the cold
18 been introduced as a reason for the observed changes in ferro- wall, is displaced in the direction of the inner magnetic field
19 fluids properties when magnetic fields are applied. However, these gradient toward the hot wall. Therefore, the resulting force (Fres)
20 investigations on MNFs have been conducted in the stagnant state on the fluid element is in the direction of the displacement, which
21 while one of the most interesting and potentially most applicable has a destabilizing effect on the ferrofluid layer and can induce a
22 applications of MNFs is that their flow and movement can be flow to the fluid. In other words, when a temperature gradient is
23 controlled by applying external magnetic fields. Therefore, the applied, ferrofluid will be non-uniformly magnetized due to the
24 question which comes to mind and can be assessed in future variation in magnetization, which results in non-uniform body
25 studies is that can chain-like structures of magnetic nanoparticles force on the MNF, known as Kelvin body force. Since temperature
26 suspended in ferrofluids be formed in the presence of magnetic and magnetic field are the basis of this phenomenon, it is known
27 fields for flowing MNFs? In other words, can application of as thermomagnetic convection. It is noteworthy that two stabiliz-
28 magnetic fields affect the properties of flowing MNFs as it has ing effects of thermal diffusivity and viscosity of the ferrofluid act
29 influenced those of stagnant MNFs? In addition, among the against this destabilizing force and both together cause the
30 thermophysical properties, very much attention from researchers temperature difference between the fluid element and its sur-
31 has been given to thermal conductivity while other properties rounding to equalize faster than the element movement from one
32 such as viscosity and specific heat, as reported in a few studies wall to the other and with temperature equalization, the magne-
33 [29,34], can also be altered in the presence of magnetic fields. tization variation which is in fact the reason for the occurrence of
34 Since these properties can considerably affect the results obtained this phenomenon decreases. All these effects are combined in the
35 from simulations of convective heat transfer of MNFs, it is hoped magnetic Rayleigh number as below [39]:
36 that more attention is given to them in future studies. 3
37 μ0  β  ∇H i  ΔT  d
Ram ¼ ð6Þ
38 κμ
39 where d denotes the thickness of the ferrofluid layer, μ0 is the
40 3. Natural convection
vacuum permeability, β¼ |∂M/∂T|H symbolizes the pyromagnetic
41 coefficient describing the dependency of ferrofluid magnetization
42 With respect to the fact that gravitational acceleration, which is
on temperature and κ is the thermal diffusivity.
43 the cause of natural convection heat transfer, is almost similar
Furthermore, under the effect of a spatially non-uniform magnetic
44 everywhere, control and improvement of this phenomenon have
field, the magnetization gradient induced by temperature gradient
45 been a challenging issue for researchers. A method which has
will intensify the force imbalance in the MNF. Therefore, in a non-
46 attracted much attention is application of MNFs. In this method, a
uniformly heated medium, colder ferrofluid is attracted towards
47 body force induced by the magnetic field is applied to the fluid
regions having higher magnetic field strength with a greater force
48 which makes it possible to control the fluid motion. The thermo-
than the warmer ferrofluid and if the high-field gradient is placed
49 gravitation phenomenon means natural movement in the fluid
near the hot region, the warmer ferrofluid will be replaced by the
50 due to buoyancy force, in which fluid motion is caused by the
colder ferrofluid from the surrounding, resulting in a flow which is
51 existence of density difference as a result of temperature gradient.
continuous as long as the magnetic field and temperature field
52 Along with such phenomenon, magnetic fields can be applied in
gradients are maintained. In both of the above conditions (i.e. in the
53 order to exert a force on MNFs and hence, control the fluid motion.
presence of either uniform magnetic field or non-uniform magnetic
54 Moreover, there is a type of MNFs which is temperature-sensitive
55 such that its magnetization decreases with temperature incre-
T2
56 ment. When there is a temperature gradient in such MNFs in the
57 presence of magnetic field, a non-equilibrium state in the magne-
58 tization of the fluid and consequently, a net magnetic driving force M-∆M
59 occurs, which induces flow in the fluid. Such phenomenon is
known as thermomagnetic convection. T M Fres Hi H
60
61 When both thermogravitational and thermomagnetic phenom- M
62 ena exist, the Rayleigh number (Ra) is obtained as below:
63 Ra ¼ RaT þ Ram ð4Þ
64 T1<T2
65 where RaT and Ram represent thermogravitational Rayleigh num- Fig. 2. Schematic principle of the situation in which thermomagnetic convection
66 ber and magnetic Rayleigh number, respectively. can occur.

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

1 field), the momentum conservation equation is as below:

2 ∇  ðρvvÞ ¼  ∇P þ ∇  ðμ∇vÞ þ μ0 ðM  ∇ÞH ð7Þ
3
4 where ρ, v and P denote density, velocity and pressure, respectively.
5 The last term in the above equation is the Kelvin body force. When
6 uniform magnetic field is applied, magnetization gradient caused by
7 temperature gradient is the source of the body force, while in the
8 presence of non-uniform magnetic field, both magnetic and tem-
9 perature field gradients are the sources of the body force.
10 Apart from being dependent on the magnetic field and tem-
11 perature distribution, the intensity of thermomagnetic convection
12 is affected by the properties of MNF and the pyromagnetic
13 coefficient [40]. With respect to the above-mentioned points, in
14 order to have more intense thermomagnetic convection, a ferro-
15 fluid with high dependency of its magnetization on temperature is
16 preferred; i.e. a MNF with high pyromagnetic coefficient. The
17 pyromagnetic coefficient can be determined from the slope of
18 the M–T curve obtained from the experimental data.
19 Ferrofluids with various pyromagnetic coefficients have been
Fig. 3. M–T diagram of the temperature-sensitive MNF reported by Yamaguchi
20 applied in different studies. It is noteworthy that only near the
et al. [42].
21 Curie temperature, variations in temperature can cause a con-
22 siderable change in the magnetization and in order to reduce the
23 Curie temperature close to the operating range (considering the
24 boiling point and the instability of the fluid at elevated tempera- temperature of 156 1C and the relative pyromagnetic coefficient
25 tures) and production of temperature-sensitive MNFs, Mn–Zn of 0.0073 1/K. In the investigations performed by Lian et al. [14,50],
26 ferrites can properly be used. For instance, Parekh et al. [41] Xuan and Lian [51] and Li et al. [52], a commercial kerosene-based
27 experimentally evaluated the pyromagnetic coefficients of three Mn–Zn ferrite MNF (BORON RUBBERS, India) was employed in
28 samples of MNFs containing Mn–Zn and Fe–Zn nanoferrites automatic cooling devices. The Curie temperature and pyromag-
29 dispersed in a non-magnetic base fluid of diester and Table 1 netic coefficient of this ferrofluid with the average particles size of
30 presents their reported results. about 6.8 nm and the concentration of 4.5% are 80 1C and 1050 A/
31 In order to assess the thermomagnetic convection, Yamaguchi mK, respectively.
32 et al. [42] applied a temperature-sensitive ferrofluid whose M–T In recent years, numerous studies have been conducted in the
33 diagram, which is representative of its pyromagnetic coefficient, is field of natural convection of MNFs, in most of which, the applied
34 illustrated in Fig. 3. This MNF which is a suspension of Mg–Zn magnetic fields have been uniform. In these investigations, the
35 ferrite nanoparticles in a kerosene solution with Curie tempera- formulation of the magnetic Rayleigh number is similar to Eq. (6).
36 ture of 477.35 K has also been used in other studies [43,44]. The The results from these studies reveal the possibility of enhance-
37 ferrofluid which has been employed in the studies conducted by ment and controlling the heat transfer phenomenon by applying
38 Blums et al. [45] and Zablotsky et al. [46] was a tetradecane based uniform magnetic fields. Yamaguchi et al. [42] experimentally and
39 temperature-sensitive MNF stabilized by oleic acid with Mn0.5Zn0.5 numerically studied the heat transfer characteristics of magnetic
40 Fe2O4 particles of about 10 nm, known as DF-67K. Magnetic fluid of Mg–Zn ferrite nanoparticles suspend in a kerosene solution
41 granulometry measurements performed by Blums et al. [45] at within a cubic container with a heat generating square cylinder
42 different temperatures and applied magnetic fields have shown stick inside and under a uniform magnetic field. Their experimental
43 almost linear dependency of the DF-67K magnetization on tem- results demonstrated that the heat transfer is enhanced with
44 perature in a wide range between 20 and 250 1C with a large value applying the magnetic field. In addition, the numerical simulation
45 of the pyromagnetic coefficient. They have reported
 the value of presented a proper consistency with the experimental findings.
1 ∂M 
46 the relative pyromagnetic coefficient (i.e. M ∂T H ) of this ferrofluid Natural convection of a temperature-sensitive MNF in a porous
47 to be approximately 0.0028 1/K. Suslov [47] applied a MNF media was investigated by Roussellet et al. [53] using lattice
48 containing Fe3O4 nanoparticles in kerosene with the pyromagnetic Boltzmann method. The cubic enclosure was under the effect of
49 coefficient of about 100 A/mK to investigate the thermomagnetic uniform magnetic fields with different magnitudes. They con-
50 convection in a vertical layer of ferrofluid. In an experimental cluded that the heat transfer is enhanced when the magnitude of
51 study conducted by Engler et al. [48], a commercial ferrofluid the magnetic field is increased.
52 which consists of magnetite nanoparticles (6.3 vol%) and a syn- Jin and Zhang [43] developed a lattice Boltzmann method to
53 thetic ester as the base fluid, called APG 513 A (Ferrotec), was simulate the laminar convection of temperature-sensitive magnetic
54 applied having the pyromagnetic coefficient of 35.3 A/mK at the fluids in a porous square cavity in the presence of uniform magnetic
55 magnetic field strength of 25 kA/m. To experimentally analyze the field. The upper and lower walls of the cavity were respectively
56 performance of a miniature thermomagnetic pump, Pal et al. [49] maintained at higher and lower temperatures while the side walls
57 used a temperature-sensitive ferrofluid with the Curie were adiabatic. It was demonstrated that the magnetic force is the
58 main driving force which enhances the average Nusselt number.
59 Table 1 Based on the observed results, they stated that using magnetic fluid
Pyromagnetic coefficients of three samples of MNFs reported by Parekh et al. [41].
60 and porous medium can be advantageous in designing energy
61 Sample Volume fraction (%) Pyromagnetic c storage equipment.
62 oefficient (A/mK) In some investigations in the field of thermomagnetic convec-
63 tion, non-uniform magnetic fields have been employed. In these
64 Diester-Mn0.5Zn0.5Fe2O4 8.91 5125 studies, apart from the force which has been mentioned in the
Diester-Fe0.5Zn0.5Fe2O4 4.94 3125
65 Diester-Fe0.3Zn0.7Fe2O4 6.99 2128
description of Fig. 2, a force due to the magnetic field with gradient
66 is applied to the MNF such that in this case, ∇H i is replaced with

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 6 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 ∇Hin the formulation of magnetic Rayleigh number. Banerjee et al. with a higher susceptibility moves down the cold wall toward
2 [54] studied thermomagnetic convection in a square cavity filled regions with a stronger magnetic field. They concluded that
3 with ferrofluid with localized heat sources and under the external thermomagnetic convection occurs due to both magnetic field
4 magnetic field provided by two line dipoles. They applied the and temperature gradients. Fig. 5 shows the relevant configuration
5 Langevin’s law to consider the effects of magnetization saturation together with their results for the average Nusselt number in
6 of the ferrofluid on heat transfer enhancement. They also assessed terms of magnetic dipole moment at different values of tempera-
7 the positions of the dipoles and effects of enclosure dimensions to ture difference (i.e. temperature difference between the hot and
8 reach maximum heat transfer. cold walls of the cavity). It can be seen that greater temperature
9 Bahiraei and Hangi [55] applied the two-phase Euler–Lagrange difference (ΔT) and higher magnetic dipole moment (m) lead to
10 method to evaluate natural convection of the water based Mn–Zn greater Nusselt number.
11 ferrite MNF in a square cavity under non-uniform magnetic field. A summary of some other studies conducted in the field of
12 The magnetic field was such that upward magnetic force was natural convection in the presence of uniform and non-uniform
13 applied to the nanoparticles near the hot wall and vice versa near magnetic fields is presented in Table 2.
14 the cold wall. The results indicated that applying the magnetic Among the MNFs, more researchers have focused on the
15 field can improve the convection of the nanofluid inside the cavity temperature-sensitive ferrofluids, which is mostly due to their
16 and enhance the heat transfer rate. Fig. 4 depicts the geometry and magnetization dependency on temperature and consequently, the
17 the applied magnetic field in their study together with the local existence of thermomagnetic phenomenon. Regarding such unique
18 Nusselt number along the hot wall obtained for the MNF with capability of this type of MNFs, it can be hoped that further in-
19 particle size of 30 nm and concentration of 4 vol% at different depth investigations in this field can not only give us more detailed
20 magnitudes of magnetic field gradient (|dB/dy|). insight into the thermomagnetic phenomenon, but also prepare the
21 Ganguly et al. [56] simulated thermomagnetic convection in a
22 differentially heated square cavity filled with ferrofluid under the
23 effect of a magnetic field produced by a line source dipole. It was
24 demonstrated that warmer fluid with lower magnetic suscept-
25 ibility moves upward and away from the dipole while colder fluid
26 Adiabatic
27
28
29
30
31
32
33 Th Tc
34
35
36
37
38
39
40 Adiabatic y0
41
42 Magnetic dipole
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64 Fig. 4. (a) Geometry and applied magnetic field, (b) local Nusselt number along the Fig. 5. (a) Schematics of the geometry and applied magnetic field, (b) average
65 hot wall of the square cavity for particle size of 30 nm and concentration of 4 vol% Nusselt number in terms of magnetic dipole moment for different values of
66 at different magnitudes of magnetic field gradient [55]. temperature difference [56].

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

1 Table 2
2 Some conducted studies in the field of natural convection of MNFs.
3
Refs. Geometry Approach Magnetic field Findings
4
5 Wen and Su Rectangular Experimental Electromagnet (uniform The heat transfer measurements suggest that the magnetic field has a destabilizing
6 [57] Hele-Shaw magnetic field) hydrodynamic effect on the flow
7 cell
Jafari et al. Cylinder Numerical Uniform magnetic field Heat transfer can be increased by applying magnetic field perpendicular to the
8 [58] (mixture temperature gradient
9 model)
10 Jin et al. [44] Porous cavity Numerical Uniform magnetic field Heat transfer on heated and cooled walls is enhanced with increasing the intensity of
11 (lattice external magnetic field
Boltzmann)
12
Engler et al. Horizontal Experimental Two pairs of Helmholtz coils The magnetoviscous effect increases the viscosity of ferrofluid which has a stabilizing
13 [48] cell (uniform magnetic field) effect on the thermomagnetic convection
14 Nanjundappa Horizontal Numerical and Uniform horizontal Increase in the viscosity dependent on the magnetic field and decrease in the magnetic
15 et al. [59] ferrofluid analytical magnetic field field stabilize the ferrofluid motion against the convection
16 layer
Yamaguchi Cubic cavity Experimental Uniform magnetic field A weak flow roll occurs inside the cavity in the presence of magnetic field which
17 et al. [60] and numerical enhances the heat transfer inside the cavity
18 Krakov and Porous Numerical Uniform magnetic field Strengthening the magnetic field can both enhance and reduce heat transfer which is an
19 Nikiforov enclosure effective method to control heat transfer
20 [61]
Suslov [47] Two vertical Numerical Uniform magnetic field Two distinct mechanisms, thermogravitational and magnetic, are responsible for the
21
plates appearance of different types of instabilities
22 Zablotsky Rectangular Experimental Permanent magnets (non- The observed enhancement in heat transfer under the magnetic field is significantly
23 et al. [46] cell and numerical uniform magnetic field) higher than that in the case of simple thermogravitational convection
24 Ashouri et al. Square cavity Numerical Permanent magnet (non- Convective flow and heat transfer increase with intensifying the magnetic field. A
25 [62] (single-phase) uniform magnetic field) general correlation is introduced for the average Nusselt number considering a wide
range of effective parameters
26 Hadavand Square cavity Numerical Line dipole (non-uniform Application of the external magnetic field modifies the temperature and velocity fields
27 and Sousa (lattice magnetic field) significantly
28 [63] Boltzmann)
29 Blums et al. Cylinder Experimental Uniform and non-uniform The results confirmed the additive action of thermogravitation and thermomagnetic
[45] magnetic fields body forces on the heat transfer intensity in ferrofluids
30
31
32
33
34 ground for achieving practical models to be used for cooling 4. Mixed and forced convection
35 objectives in near future.
36 In most of the studies which have been conducted thus far in In the field of mixed and forced convection, controlling the
37 the field of thermomagnetic convection, it has been assumed that fluid flow and improving heat transfer in a specific section of the
38 the thermophysical properties of MNFs are constant and indepen- flow can be achieved by using MNFs and application of magnetic
39 dent from the applied magnetic field. However, some theoretical fields in that section. Application of different magnetic fields can
40 investigations on MNFs have shown that applied magnetic fields not only affect the thermophysical properties of MNFs, but also
41 can undeniably affect the properties under certain circumstances. influence the convective heat transfer which is due to both
42 One of the most known effects in MNFs is the change of the changes in the thermophysical properties of the MNFs and motion
43 viscosity in the presence of a magnetic field, known as magneto- and migration of magnetic nanoparticles. The convective heat
44 viscous effect. Under the effect of a magnetic field, the magnetic transfer of MNFs has been assessed both in the absence and
45 moments of the particles align with the direction of the magnetic presence of magnetic fields.
46 field and if the alignment reaches a certain value, the magnetic With respect to the fact that factors such as formation of chain-like
47 interaction of the particles become strong enough so that the structures, magnetoviscous effect, Brownian and thermophoresis
48 particles may build chain-like structures which can cause a drastic phenomena as well as the interaction between the magnetic nano-
49 increase in the viscosity of the MNF. Having this fact in mind, it is particles and the base fluid have not been completely identified,
50 expected that in future studies, these two phenomena, i.e. ther- considering their precise effects in a numerical investigation seems to
51 momagnetic convection and magnetoviscous effect, are assessed be rather difficult. Therefore, experimental studies in this field can
52 together and not separately so that more precise insight into the present more accurate results. Some researchers have experimentally
53 mechanism of the occurrence of thermomagnetic convection as evaluated the hydrothermal characteristics of MNFs. Syam Sundar
54 well as the effective factors and conditions can be obtained. et al. [64] experimentally evaluated the convective heat transfer
55 Furthermore, a review on the studies conducted in this field coefficient and friction factor of Fe3O4 MNF flowing in a circular tube
56 shows that uniform magnetic fields have been used in most of in the range of 3000oReo22000. The amount of heat transfer in case
57 them. Even though application of such fields is necessary and of using nanofluid was higher compared to water and increased with
58 beneficial in order to have more precise insight into the mechan- volume concentration. They developed a correlation based on the
59 ism of thermomagnetic convection, providing uniform magnetic experimental data to estimate the Nusselt number and friction factor.
60 fields which is done by different means such as electromagnets Moreover in another experimental investigation by Syam Sundar et al.
61 and Helmholtz coils seems to be uneconomical due to electrical [65], turbulent convective heat transfer and friction factor of Fe3O4
62 energy consumption. Therefore, it is anticipated that with applica- MNF flowing through a uniformly heated horizontal circular tube with
63 tion of magnetic fields such as those provided by permanent and without twisted tape inserts were studied. Their results indicated
64 magnets, the phenomenon of thermomagnetic convection can be that both heat transfer and friction coefficient increase when twisted
65 used in a more practical and economical fashion, since such fields tape inserts are utilized. However, no magnetic field was applied in
66 are not only more accessible, but also do not consume any energy. both studies conducted by Syam Sundar et al. [64,65].

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 8 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 Li and Xuan [66] experimentally investigated the convective attention. Production of monodisperse nanofluids with tunable
2 heat transfer of Fe3O4–water nanofluid flow over a fine wire under microstructures is necessary both for better understanding of
3 the influence of uniform and non-uniform magnetic fields. They microstructures inside nanofluids and for experimentally investi-
4 found that in the absence of magnetic field, the MNF presents gating the effect of self-assembly configurations on the MNFs
5 higher heat transfer in comparison with water. However, the characteristics. It is well known that surfactants are necessary to
6 external magnetic field significantly affects the convective heat produce the evenly distributed and long stabilized MNFs. How-
7 transfer performance of the magnetic fluid and heat transfer ever, addition of the surfactants will affect the nanoparticle
8 process can be controlled by applying a magnetic field, such that distribution and therefore, affect the features of MNF. Moreover,
9 heat transfer is remarkably enhanced when the orientation of the it is certain that the properties of a MNF depend upon the
10 magnetic field gradient is along with the main flow of magnetic properties and concentration of the surfactants. Therefore, great
11 fluid while a magnetic field gradient opposite to the main stream attention should be given to the use of surfactants and their
12 suppresses the heat transfer. amounts in experimental studies. Despite using surfactants, some
13 Ghofrani et al. [67] performed an experimental investigation to particles may combine to each other caused by the van der Waals
14 study laminar forced convection heat transfer of a ferrofluid forces and dipole–dipole interactions, which causes the formation
15 flowing through a circular tube in the presence of an alternating of aggregations. Generally, the complexities in the preparation of
16 magnetic field. The effects of magnetic field, volume concentration MNFs and the issues pertaining to the aggregation of nanoparticles
17 and Reynolds number on the convective heat transfer were due to the large surface area can lead to nonreproducible and
18 assessed in their study. It was found that increasing the alternating inconsistent results in experiments. However, precise character-
19 magnetic field frequency and the volume fraction lead to better ization of MNFs before performing experiments can prevent the
20 heat transfer enhancement and the effect of the magnetic field is occurrence of errors to a great extent. In addition, reviewing the
21 higher at low Reynolds numbers. They reported a maximum of experimental investigations shows that studies on forced convec-
22 27.6% enhancement in the convection heat transfer. tive heat transfer of MNFs containing Mn–Zn ferrite nanoparticles
23 Lajvardi et al. [68] performed an experimental study to inves- are rather sparse and regarding superior magnetic properties of
24 tigate the convective heat transfer of water–Fe3O4 ferrofluid such nanofluids, it is expected that more research is conducted on
25 flowing through a tube in laminar regime in the presence of them in future.
26 magnetic field perpendicular to the flow direction. They evaluated Beside the experimental studies, some researchers have inves-
27 the effect of ferrofluid concentration and magnet position in the tigated mixed and forced convection of MNFs using numerical
28 experiments. Fig. 6 depicts the relevant configuration and their methods (both single- and two-phase approaches). Aminfar et al.
29 results for convective heat transfer coefficient (h) versus dimen- [69] applied the two-phase mixture model to evaluate mixed
30 sionless length of the tube for water and ferrofluid at different convection of a ferrofluid (kerosene and 4 vol% Fe3O4) flow in a
31 magnetic fields (B). They reported that the observed enhancement vertical tube under the effects of non-uniform magnetic field with
32 in heat transfer can be attributed to the improved thermophysical positive and negative gradients. The magnetic field was applied in
33 properties of the ferrofluid in the presence of magnetic field. a specific section of the tube. It was concluded that in the section
34 It should be noted that in experimental investigations, there of applied magnetic field, the magnetic field with negative
35 are major challenges to which researchers should give more gradient acts similar to buoyancy force and enhances the heat
36 transfer rate, while the magnetic field with positive gradient
37 decreases it. Fig. 7 shows the geometry and applied magnetic field
38 Cooling part as well as their results for Nusselt number at different magnetic
39 field gradients (G) for Re¼20. In addition, they concluded that
40 Coils using the magnetic field with negative gradient needs a high
Pump
41 pumping power while magnetic field with positive gradient acts as
42 a pump. In another study conducted by Aminfar et al. [70], in
Test section Reservoir tank
43 addition to the ferrohydrodynamics, the magnetohydrodynamics
44 principles were also taken into account due to electrical conduc-
T9 T8 T7 T6 T5 T4 T3 T2 T1
45 tivity of the ferrofluid (sea water and 4 vol% Fe3O4).
46 Keshavarz Moraveji and Hejazian [71] used the single-phase
47 Data acquisition numerical approach to study the effects of nanoparticle concen-
48 tration and flow rate on the convective heat transfer and friction
49 factor of Fe3O4 MNF flowing through a tube and demonstrated that
DC power supply
50 the obtained results are in a proper agreement with the experi-
51 mental results of Syam Sundar et al. [64].
52 Flow and heat transfer characteristics of water based Mn–Zn
53 ferrite magnetic nanofluid through an annulus were evaluated by
54 Bahiraei et al. [72] under the effect of non-uniform magnetic field
55 using the two-phase Euler–Lagrange method. They observed that
56 the concentration distribution is non-uniform at the annulus cross
57 section, with its value being lower near the walls. Due to this non-
58 uniformity, velocity profile becomes flatter at the cross section of
59 the annulus by applying the magnetic field. The applied magnetic
60 field was such that a force was applied to the magnetic particles in
61 opposite direction of the flow and due to the concentration non-
62 uniformity, greater force was applied to the fluid in central
63 regions. Thus, the fluid velocity decreased there and increased
64 Fig. 6. (a) The configuration of experimental system, (b) convective heat transfer
near the walls which resulted in heat transfer enhancement. They
65 coefficient versus dimensionless length of the tube for water and ferrofluid at also investigated the effects of concentration, size of particles
66 different magnetic fields [68]. and magnitude of magnetic field gradient and performed an

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 9

1 parallel plates under the influence of a magnetic dipole located

2 outlet g outlet below the channel. The upper and lower plates were kept at
3 constant but different temperatures. It was demonstrated that the
4 cooler ferrofluid flows in the direction of the magnetic field
5 gradient and displaces the hotter ferrofluid which can result in
6 D heat transfer enhancement.
7 L/4 In most of the numerical investigations [69,70,73], the correla-
8
Positive magnetic tions which had been proposed for suspensions containing parti-
9 field gradient cles larger than nanometric dimensions have been used to
10 calculate the effective properties of the MNFs. However, with
11 respect to the new studies conducted on measuring and modeling
12 the properties of MNFs, applying models which are specific to
13 them can present more accurate results. In addition, in case of
14 using models of properties in numerical analyses, more attention
15 L/2 should be given to the effect of magnetic fields on thermophysical
16 properties of MNFs. It means that applying models of properties
17 which are obtained in the absence of magnetic fields may cause
18 error in the numerical simulation.
19 Furthermore, in most of the numerical studies conducted on
20 the convective heat transfer of MNFs, single- and two-phase
21 approaches have been used. Single-phase method is considerably
Negative magnetic
22 dependent on the models of the effective properties and the
23 field gradient L/4 obtained results may change by using other properties models.
z
24 Moreover, with respect to the limitations on the computational
25 tools, two-phase methods are limited to a volume containing finite
26 particles. As a newer numerical approach, the lattice Boltzmann
27 method can be more beneficial to evaluate features of the MNFs.
inlet inlet
28 This approach discretizes the flow region into a number of grids
29 and solves the simplified Boltzmann equation on the regular
30 lattices. In the field of mixed and forced convection of MNFs, very
31 few studies have employed this approach. This method links the
32 macroscopic phenomena to the microscopic world. It is a useful
33 numerical method for simulating flow and energy transport
34 processes of a flow system without particles limitation and can
35 be properly used to simulate the hydrothermal characteristics of
36 MNFs in the presence of magnetic fields.
37 In addition, in the numerical studies which have been con-
38 ducted thus far, in order to reduce the computational volume,
39 magnetic fields applied on the MNFs flows have been simplified
40 fields. However, practical magnetic fields such as field around
41 permanent magnets, field provided by a current carrying wire and
42 so forth, are three-dimensional fields with complicated correla-
43 tions. Therefore, in future studies, in order to better understand
44 how magnetic fields affect the flowing ferrofluids, 3D analyses of
45 flow and heat transfer of MNFs under the effect of the mentioned
46 magnetic fields (which are practically more accessible) should be
47 given more attention.
48
Fig. 7. (a) Geometry and applied magnetic field, (b) Nusselt number for different
49
magnetic field gradients at Re¼ 20 [69].
50 5. Boiling
51
52 optimization to achieve the conditions with maximum heat transfer Boiling is an effective heat transfer mechanism which has a vast
53 and minimum pressure drop. range of applications, such as in cooling systems of nuclear
54 Aminfar et al. [73] applied the mixture model to investigate the reactors and thermal power plants, as well as other machines
55 hydrothermal characteristics of a ferrofluid in a vertical rectangu- and devices which require removal of high heat flux to guarantee
56 lar duct under a non-uniform magnetic field generated by an efficiency, performance, and safety. For boiling heat transfer, there
57 electric current going through a parallel wire located under the is an upper limit value in the heat transfer performance, called the
58 duct. They used the models of Hamilton-Crosser [74] and Einstein Critical Heat Flux (CHF). Therefore, equipment in which boiling
59 [75] for the effective thermal conductivity and viscosity, respec- heat transfer plays a pivotal role should be designed such that the
60 tively. The results showed that applying this magnetic field causes CHF is never reached. The reason is due to the fact that the surface
61 the velocity gradient to increase near the walls so that the local temperature suddenly increases at CHF, which can lead to
62 Nusselt number is enhanced. They also demonstrated that the mechanical failure as a result of thermal stresses, or even burnout
63 magnetic field creates a pair of vortices which can improve the of the heater.
64 heat transfer rate and prevent sedimentation of nanoparticles. For the purpose of achieving higher values of CHF, using nano-
65 Strek and Jopek [76] numerically investigated the laminar heat fluid as the working fluid is one of the promising options [77,78].
66 transfer of a ferrofluid flowing through a channel between two Investigations carried out on nanofluids indicate a considerable

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 10 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎
1 enhancement in CHF, from 30 to 200%, at different surface condi-
2 tions and nanoparticles types. Therefore, application of nanofluids
3 can prevent undesired effects such as local hot spots and conse-
4 quently, lead to reliability enhancement in components which need
5 to be cooled. You et al. [79] were the first ones who assessed the CHF
6 phenomenon in pool boiling of nanofluids. They reported astonish-
7 ing results such that the experiments demonstrated 50 to 200%
8 enhancement in CHF as compared to the base fluid.
9 Due to the fact that the magnetic nanoparticles can be affected
10 by the external magnetic fields, some researchers have investi-
11 gated MNFs in boiling heat transfer. Most of the conducted
12 investigations have reported that appropriate application of mag-
13 netic fields can cause improvement in the boiling heat transfer. Lee
14 et al. [80] performed an experimental study to investigate flow
15 boiling CHF for the Fe3O4 MNF at various concentrations. They
16 reported that in the absence of magnetic field, the CHF enhance-
17 ment mechanism is similar to that of other nanofluids due to the
18 deposition of nanoparticles on the surface. However, when the
19 magnetic field was applied, they observed even more CHF
20 enhancement since the magnets attract the magnetic nanoparti-
21 cles and intensify the local concentration of nanofluid near the
22 tube wall in comparison with the bulk region.
23 Sesen et al. [81,82] experimentally studied the pool boiling
24 heat transfer of a MNF under the effect of magnetic field provided
25 by rotating magnetic stirrers. They reported that in the presence of
26 the magnetic field, the nanofluid experiences more mixing which
27 increases convection from the surface such that the rotating
28 magnetic field intensifies the nanoparticles motion and their
29 contact with the heated surface which can result in heat transfer
30 enhancement from the heated surface to the nanoparticles.
31 Lee et al. [83] evaluated the effect of different nanofluids
32 including water–Fe3O4 magnetic nanofluid on the CHF enhance-
33 ment in pool boiling. It was observed that the MNF has the highest
34 value of CHF enhancement among all applied nanofluids in all
35 concentration ranges. In addition, they reported that in the
36 presence of magnetic field, local concentration of the MNF can
37 be increased in the high heat flux area.
38 In some investigations conducted in this field, the effects of
39 magnetic fields on the bubbles growth and their release have been
40 introduced as one of the effective factors on the boiling heat
41 transfer of MNFs. Pool boiling heat transfer of water-based MNFs
42 in the absence and presence of a magnetic field was analyzed
43 experimentally by Junhong et al. [84]. They placed a permanent
44 magnet at the bottom of the boiling vessel to create a downward
45 magnetic field gradient. It was demonstrated that the boiling heat
46 transfer enhancement of a ferrofluid in the presence of the
47 magnetic field is primarily caused by the strong influence of
48 magnetic field on the bubbles since the magnetic field makes
49 the bubbles accelerate and have faster growth.
50 Habibi Khoshmehr et al. [85] investigated the effects of ferro-
51 fluid concentration and magnetic field implementation on the
52 fluid throughout the boiling process. Their results showed that
53 ferrofluid boiling heat flux increases by 50% in all phases com-
54 pared with ferrofluid boiling without implementing magnetic
55 field. In addition, as a result of the magnetic field implementation,
56 a secondary upward current was produced, which accelerates the
57 release of the bubbles of steam.
58 The results of the conducted investigations indicate that
59 application of magnetic fields can significantly affect the boiling
60 heat transfer of MNFs. Furthermore, even though almost all
61 studies have reported enhancement in CHF for MNFs in compar-
62 ison with the base fluid, the presented enhancement values are
63 rather different, which can be due to differences in nanofluids
64 characterization, size of nanoparticles and the type of heaters used
65 in various studies. As a result, for better understanding of boiling
66 phenomenon in MNFs and their CHF, further investigations in this
Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
field are strongly felt to be implemented, particularly evaluating
various effective parameters such as surface geometry and fea-
tures, nanoparticle properties, type and intensity of the applied
magnetic field, pressure, mass flow rate, and so forth.
6. Applications
MNFs can have numerous applications thanks to their unique
characteristics. In this section, the most recently developed appli-
cations of them will be reviewed.
6.1. Thermomagnetic convection
Thermomagnetic phenomenon is the fundamental principle of
an automatic energy transport device (AETD) or automatic cooling
device (ACD) using a temperature-sensitive MNF as a coolant. In a
thermomagnetic convection loop (e.g. AETD and ACD), the MNF
continuously circulates in the conduit without additional energy
consumption. The thermomagnetic convection can be more ben-
eficial, particularly in cases which conventional pressure-driven
and thermo-gravitational flows are not able to meet the desired
demand, such as in micro-sized geometries.
Numerical investigations of thermomagnetic convection [48,86]
have revealed that the thermomagnetic effects are substantial,
particularly at smaller length scales. This finding is also supported
through scale analysis [87].
The most important features of AETD or ACD are as below:
(1) The magnetic force, induced by the dependency of magnetiza-
tion on temperature in a temperature-sensitive MNF, causes
motion in the fluid and thus, transports heat from hotter to
colder regions.
(2) These systems have simple structures without any moving
components.
(3) These systems are self-control devices. It means that the
higher heat load, the greater driving force, such that the
magnetic fluid circulates at a greater velocity and transports
the heat away more quickly from the heat source.
In general, these devices without any moving components
make a silent, self-regulating, self-driven, and maintenance-free
fluid control system which directly use the heat from the heat
source for driving the fluid and have a significant reliability.
Lian et al. [14,50] developed an AETD by using a temperature-
sensitive MNF as a coolant, in which the magnetic force due to the
effects of both magnetic field gradient and the fluid temperature
variation causes the fluid to move in a loop. They stated that since
no mechanical moving part is needed in such devices, they are
highly applicable in a variety of branches of fluid and energy
transport systems including lab-on-a-chip, energy conversion
devices, electronic cooling devices, and so forth. One of the devices
developed by them is presented schematically in Fig. 8. As can be
seen from the figure, this device is composed of two permanent
magnets, two heaters, two heat sinks and a temperature-sensitive
MNF in a loop. They concluded that by changing the external
magnetic field and/or temperature variation inside the magnetic
fluid, it is possible to control the energy transport process of such
devices.
Pal et al. [49] fabricated a thermomagnetic pump without
application of any external pressure gradient, in which the applied
temperature and magnetic field gradients were the driving forces of
the ferrofluid. Such pumps can have many applications in cooling of
state-of-the-art microelectronic devices. They also investigated the
performance of the thermomagnetic pump to characterize the pump
pressure head and discharge under different working conditions
 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 11

1 magnetic field, the magnetic fluid flow and the temperature

2 distribution of the magnetic fluid can be achieved.
3 In addition, below suggestions can be beneficial for more
4 effective use of the thermomagnetic convection phenomenon in
5 the mentioned applications:
6
7  Future furtherance in the field of nanoferrites (for instance,
8 substitution of appropriate elements in the structures of the
9 existed ferrites) can lead to production of MNFs with higher
10 pyromagnetic coefficients, which in turn, the ferrofluid depen-
11 dency on temperature increases and consequently, more effec-
12 tive thermomagnetic convection can be induced to the fluid.
13  When a MNF reaches its saturation magnetization, using
14 stronger magnetic fields has no effect on the magnetic force
15 applied to the fluid and this is one of the factors limiting the
16 intense of thermomagnetic convection. Therefore, production
17 of ferrofluids with higher saturation magnetizations in future
18 can prepare the ground for having more intense thermomag-
19 netic convection by applying stronger magnetic fields and
20 hence, increasing the performance of the above-mentioned
21 systems.
22  Thermophysical properties of MNFs, such as viscosity, specific
23 heat and boiling point can profoundly affect the performance of
24 the mentioned systems. Production of MNFs with lower visc-
25 osity can, for instance, result in more effective flow induced by
26 thermomagnetic convection phenomenon in AETD, ACD and
27 thermomagnetic pump and consequently, higher perfor-
28 mances. The boiling point of the MNF is also one of the factors
29 limiting the performance of these systems. The reason is that
30 the basic of the thermomagnetic convection on which these
31 systems rely is the temperature gradient existed in the fluid
32 Fig. 8. AETD system developed by Lian et al. [14]. and if in the future, MNFs with higher boiling points are
33 produced, greater temperature gradients can be applied and
34 such as the magnetic field strength, heating power and ferrofluid thus, more effective thermomagnetic convection will be prac-
35 properties. They reported that the pump offers precise flow control tically achieved.
36 but high flow rate compared to similar pumps in its category.
37 Li et al. [52] established a prototype of a miniature ACD. Such
38 devices can be very applicable in various fields such as electronics
39 since they are maintenance-free because of no moving parts in the 6.2. Heat pipes
40 loop. They performed an experimental study to investigate
41 the performance capacity of the cooling device and evaluated Superior thermal characteristics of MNFs in comparison with
42 the effects of several factors such as the magnetic field and the base fluids, as well as their capability of being controlled by
43 temperature distribution of the fluid on the cooling performance applying different kinds of magnetic fields have caused some
44 of the device. It was demonstrated that since the magnetic field researchers to use them in heat pipes. The effect of using a MNF
45 and the temperature gradient are the driving forces of the was experimentally studied by Mohammadi et al. [88] in a
46 magnetic fluid flow, the flow velocity increases with the increase pulsating heat pipe which is one of the most reliable and efficient
47 in the external heat load, and vice versa, which indicates that heat transfer devices. They found that using the ferrofluid in the
48 operation of the cooling device is automatically controlled by the absence of magnetic field enhances the thermal performance
49 heat load. compared to water and this enhancement becomes more signifi-
50 Xuan and Lian [51] used thermomagnetic convection loop in cant in the presence of the magnetic field.
51 electronic cooling. Such a device utilizes an earth magnet and the Ming et al. [89] experimentally studied the effect of using a
52 waste heat generated from a chip or other sources to maintain the MNF as the working fluid of a flat plate heat pipe. It was shown
53 flow of working fluid which transports heat to a far end for that the MNF can be attracted by magnetic field and the circulation
54 dissipation. In their cooling device, no additional energy other of working fluid can be improved and thus, the flat plate heat pipe
55 than the waste heat dissipated is consumed for driving the cooling with magnetic fluid can achieve more uniform heat flux distribu-
56 system and the device can be considered as completely self- tion on the condensation surface as compared to heat pipe
57 powered. Application of this system to the hot chip resulted in with water.
58 an obvious temperature drop of the chip surface. As the heat load Jahani et al. [90] evaluated the thermal performance of a micro
59 increases, a higher heat dissipation rate can be realized due to a pulsating heat pipe using silver nanofluid and ferrofluid. They
60 stronger thermomagnetic convection, which indicates a self- observed that in the absence of magnetic field, silver nanofluid
61 regulating feature of such devices. presents higher improvement in thermal performance, but when
62 Although some researchers have applied the concept of thermo- the magnetic field is applied, the thermal performance of the heat
63 magnetic convection in some practical applications, more in-depth pipe with MNF is higher than that with silver nanofluid. They also
64 studies are felt to be conducted in order to gain a comprehensive reported that the controllability of ferrofluids in micro pulsating
65 insight into different aspects of such automatic devices as well as heat pipes is very beneficial for evaporator temperature regulation
66 their reliability so that more conscious relation between imposed in electronic cooling applications.

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 12 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 Chiang et al. [91] developed a closed-loop heat pipe using a heat transfer medium in energy conversion systems such as heat
2 MNF with magnetically enhanced thermal properties. The results exchangers and processes including boiling. Even in cooling of
3 demonstrated that an optimal thermal conductivity exists in the loud speakers [9], MNFs have been used properly. Pislarun-
4 applied field of 200 Oe. They stated that this magnetically Danescu et al. [96] evaluated the use of a MNF as a coolant in a
5 enhanced heat transfer can be applicable in promising heat- transformer and demonstrated that in the vertical configuration of
6 dissipation device and the energy architecture integrated with the transformer, the magnetization forces act concurrently with
7 an additional energy system. the thermal flow which can enhance the heat transferred from the
8 Taslimifar et al. [92] performed an experimental study to transformer to the ambient.
9 evaluate the effect of using ferrofluid on thermal performances Bahiraei and Hangi [97] investigated the performance of water
10 of open loop pulsating heat pipes. They investigated different based Mn–Zn ferrite MNF in a counter-flow double-pipe heat
11 parameters such as heat input, ferrofluid concentration and exchanger under quadrupole magnetic field. Their findings showed
12 magnets location. It was found that thermal performance of heat that increasing each of the parameters of nanoparticle concentra-
13 pipe with ferrofluid enhances under the influence of magnetic tion, particle size and magnitude of the magnetic field will lead to a
14 field and thermomagnetic convection has an important role on greater pressure drop and also higher heat transfer improvement.
15 this enhancement. Furthermore, at higher Reynolds numbers, the effect of magnetic
16 force was diminished.
17 6.3. Mini-scale applications Apart from the points mentioned above, MNFs have widely
18 been used in bioengineering applications one of which is the
19 The growing importance of mini-scale heat transfer devices has thermal treatment of some kinds of cancer and tumor (i.e.
20 initiated a great deal of research that addresses heat exchange in hyperthermia). The technique involves dispersing magnetic nano-
21 miniaturized configurations. In general, due to very low Reynolds particles in the target tissue, and then adopting a magnetic field
22 numbers, the transport of momentum and energy in miniaturized with adequate strength and frequency to heat the nanoparticles.
23 devices is diffusion limited. Using MNFs in these applications and The heat conducts into the surrounding diseased tissue whereby, if
24 controlling the flow by external magnetic fields can be a useful the temperature can be kept above the therapeutic threshold for a
25 alternative to increase convection in these devices. In addition, a given time, the cancer is destroyed [98–100].
26 great number of electronic cooling devices are miniature. There-
27 fore, some studies have focused on assessing the hydrothermal
28 characteristics of MNFs in mini-scale equipment. 7. Conclusion
29 Lee and Chang [13] used a MNF to enhance the thermal
30 performance of a miniature thermal fluidic system in the presence Obviously, MNFs have great potential for heat transfer
31 of magnetic field. Due to simple construction, low weight, low cost enhancement and are highly suited to application in practical heat
32 and excellent environmental adaptability, the thermal fluidic transfer processes. This paper presents a review of the studies
33 systems have been introduced in thermal management issues. conducted on MNFs and their various applications.
34 They reported that based on the design of local electromagnetic The thermophysical properties of MNFs are significantly depen-
35 fields, a transient Lorentz force can be induced within the ferro- dent on the type and conditions of the applied magnetic field and
36 fluid flow, and thermal convection in MNF can then be actively hence, models of properties for cases with and without application
37 enhanced. of magnetic fields can be different from each other. Therefore,
38 Xuan et al. [93] applied the lattice Boltzmann method to more attention should be given when models of properties are
39 simulate flow and thermal processes of a MNF flowing through a applied in numerical analyses. In most of the numerical investiga-
40 micro channel. They concluded that the hydrothermal character- tions, the correlations proposed for suspensions containing parti-
41 istics of the MNF can be controlled by altering orientation and cles larger than nanometric dimensions have been used. Regarding
42 magnitude of the magnetic field. It was demonstrated that for heat the new studies implemented on modeling the properties of
43 transfer enhancement, the most remarkable effect is achieved MNFs, applying models which are specific to them can offer more
44 when the orientation of the magnetic field gradient is parallel to precise results.
45 the fluid flow. Among the MNFs, more investigations have been conducted on
46 Fumoto et al. [94] carried out an experimental study on a mini- the temperature-sensitive ferrofluids, which is mostly due to the
47 heat transport device incorporating a temperature-sensitive MNF. thermomagnetic phenomenon. Although some researchers have
48 The parameters which were assessed in the experiments were the applied this concept in some practical applications, more detailed
49 magnetic force, the position of the magnet and the fluid tempera- studies are felt to be conducted in order to gain a comprehensive
50 ture. It was found that the velocity of MNF could be controlled by insight into automatic devices working on the basis of thermo-
51 varying the magnetic force and the position of the magnet. magnetic phenomenon so that the proper relation between
52 Moreover, the maximum velocity of the fluid produced without imposed magnetic field, the magnetic fluid flow and the tempera-
53 using a pump was approximately 20 mm/s. ture distribution of the magnetic fluid can be achieved. Moreover,
54 Philip et al. [95] developed a multifunctional device in which it is expected that in future studies, the magnetoviscous effect on
55 MNF was used and the ratio of thermal conductivity to viscosity thermomagnetic convection is assessed more comprehensively.
56 was tunable. This system played the role of both heat dissipater In addition, application of MNFs along with magnetic fields can
57 and damper. They stated that such devices could have applications lead to control and improvement of boiling heat transfer. However,
58 in micro-fluidic devices, Micro Electromechanical Systems in order to better characterize such phenomenon, further assess-
59 (MEMS), Nano Electromechanical Systems (NEMS) and other ments on boiling of MNFs under the effects of various parameters
60 miniature devices. such as surface geometry and features, nanoparticles properties,
61 type and intensity of the applied magnetic field, pressure, mass
62 6.4. Other applications flow rate and so forth are needed to be done.
63 Eventually, practical applications of MNFs were reviewed, such
64 With respect to the great potential of MNFs, they can be applied as in AETD and ACD, miniaturized devices and heat pipes, and
65 in various applications related to heat transfer. For instance, they directions for prospective studies were presented. Such devices
66 can be used as a coolant in thermal management devices or as a which operate based on the thermomagnetic phenomenon can

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 13

1 have considerable application potentials in the field of thermal [22] B. Wang, B. Wang, P. Wei, X. Wang, W. Lou, Controlled synthesis and size-
2 management, particularly in cooling of modern electronic devices dependent thermal conductivity of Fe3O4 magnetic nanofluids, Dalton Trans.
41 (2012) 896–899.
3 in which proper heat dissipation is the key to success in the [23] K.S. Hong, T.K. Hong, H.S. Yang, Thermal conductivity of Fe nanofluids
4 current industry trend of miniaturizing and integrating electronic depending on the cluster size of nanoparticles, Appl. Phys. Lett. 88 (2006)
5 components. With further advancement in ferromagnetic materi- 031901.
[24] T.H. Tsai, L.S. Kuo, P.H. Chen, C.T. Yang, Effect of viscosity of base fluid on
6 als and producing better MNFs with greater magnetization and thermal conductivity of nanofluids, Appl. Phys. Lett. 93 (2008) 233121.
7 higher pyromagnetic coefficient, MNFs will find considerable [25] H. Zhu, C. Zhang, S. Liu, Y. Tang, Y. Yin, Effects of nanoparticle clustering and
8 applications in a wide variety of engineering fields, especially alignment on thermal conductivities of Fe3O4 aqueous nanofluids, Appl.
Phys. Lett. 89 (2006) 023123.
9 those in which there is a pressing need for reducing the energy [26] W. Yu, H. Xie, L. Chen, Y. Li, Enhancement of thermal conductivity of
10 consumption. MNF research could lead to a key breakthrough in kerosene-based Fe3O4 nanofluids prepared via phase-transfer method,
11 developing next generation coolants for various medical and Colloids Surf., A 355 (2010) 109–113.
[27] C.L. Altan, A. Elkatmis, M. Yuksel, N. Aslan, S. Bucak, Enhancement of thermal
12 engineering applications.
conductivity upon application of magnetic field to Fe3O4 nanofluids, J. Appl.
13 Phys. 110 (2011) 093917.
14 [28] M. Krichler, S. Odenbach, Thermal conductivity measurements on ferrofluids
15 Q2 with special reference to measuring arrangement, J. Magn. Magn. Mater. 326
Uncited references (2013) 85–90.
16 [29] Y.P. Chiu, Y.F. Chen, S.Y. Yang, J.C. Chen, H.E. Horng, H.C. Yang, W.S. Tse,
17 [26], [27]. C.Y. Hong, Specific heat of magnetic fluids under a modulated magnetic field,
18 J. Appl. Phys. 93 (2003) 2079.
[30] P.D. Shima, J. Philip, Tuning of thermal conductivity and rheology of
19 nanofluids using an external stimulus, J. Phys. Chem. C 115 (2011)
20 References 20097–20104.
21 [31] T.T. Baby, S. Ramaprabhu, Magnetic nanoparticles decorated multiwalled
[1] S.U.S. Choi, Enhancing thermal conductivity of fluid with nanoparticles, in: carbon nanotubes dispersed nanofluids, AIP Conf. Proc 1349 (2011) 965.
22 [32] K. Parekh, H.S. Lee, Experimental investigation of thermal conductivity of
Proceedings of the ASME International Mechanical Engineering Congress,
23 ASME FED-231, 1995, 99–105. magnetic nanofluids, AIP Conf. Proc 1447 (2012) 385.
24 [2] N. Kannadasan, K. Ramanathan, S. Suresh, Comparison of heat transfer and [33] A. Gavili, F. Zabihi, T. Dallali Isfahani, J. Sabbaghzadeh, The thermal con-
pressure drop in horizontal and vertical helically coiled heat exchanger with ductivity of water base ferrofluids under magnetic field, Exp. Therm. Fluid
25 Sci. 41 (2012) 94–98.
CuO/water based nanofluids, Exp. Therm. Fluid Sci. 42 (2012) 64–70.
26 [34] Q. Li, Y. Xuan, J. Wang, Experimental investigations on transport properties of
[3] R. Lotfi, A.M. Rashidi, A. Amrollahi, Experimental study on the heat transfer
27 enhancement of MWNT–water nanofluid in a shell and tube heat exchanger, magnetic fluids, Exp. Therm. Fluid Sci. 30 (2005) 109–116.
[35] I. Nkurikiyimfura, Y. Wang, Z. Pan, Effect of chain-like magnetite nanoparticle
28 Int. Commun. Heat Mass Transfer 39 (2012) 108–111.
[4] W.S. Han, S.H. Rhi, Thermal characteristics of grooved heat pipe with hybrid aggregates on thermal conductivity of magnetic nanofluid in magnetic field,
29 Exp. Therm. Fluid Sci. 44 (2013) 607–612.
nanofluids, Therm. Sci. 15 (2011) 195–206.
30 [5] M. Bahiraei, A comprehensive review on different numerical approaches for [36] H.L. Fu, L. Gao, Theory for anisotropic thermal conductivity of magnetic
31 simulation in nanofluids: traditional and novel techniques, J. Dispersion Sci. nanofluids, Phys. Lett. A 375 (2011) 3588–3592.
Technol. 35 (2014) 984–996. [37] J. Philip, P.D. Shima, B. Raj, Evidence for enhanced thermal conduction
32 through percolating structures in nanofluids, Nanotechnology 19 (2008)
[6] G. Huminic, A. Huminic, I. Morjan, F. Dumitrache, Experimental study of the
33 thermal performance of thermosyphon heat pipe using iron oxide nanopar- 305706.
34 ticles, Int. J. Heat Mass Transfer 54 (2011) 656–661. [38] K. Parekh, H.S. Lee, Magnetic field induced enhancement in thermal
[7] C.J. Ho, W.C. Chen, An experimental study on thermal performance of Al2O3/ conductivity of magnetite nanofluid, J. Appl. Phys. 107 (2010) 09A310.
35 [39] Th. Volker, E. Blums, S. Odenbach, Heat and Mass Transfer Phenomena in
water nanofluid in a minichannel heat sink, Appl. Therm. Eng. 50 (2013)
36 516–522. Magnetic Fluids, GAMM-Mitt 30 (2007) 185–194.
37 [8] M. Bahiraei, S.M. Hosseinalipour, Particle migration in nanofluids consider- [40] A.O. Kuzubov, O.I. Ivanova, Magnetic liquids for heat exchange, Journal de
ing thermophoresis and its effect on convective heat transfer, Thermochim. Physique III (4) (1994) 1–6.
38 [41] K. Parekh, R.V. Upadhyay, R.V. Mehta, Magnetocaloric effect in temperature-
Acta 574 (2013) 47–54.
39 [9] S. Odenbach, Ferrofluids, Springer, Berlin, 2002. sensitive magnetic fluids, Bull. Mater. Sci. 23 (2000) 91–95.
40 [10] R.E. Rosensweig, Ferrohydrodynamics, Cambridge University Press, Cam- [42] H. Yamaguchi, X.R. Zhang, X.D. Niu, K. Yoshikawa, Thermomagnetic natural
bridge, 1985. convection of thermo-sensitive magnetic fluids in cubic cavity with heat
41
[11] M. Lin, D. Zhang, J. Huang, J. Zhang, W. Xiao, H. Yu, L. Zhang, J. Ye, The anti- generating object inside, J. Magn. Magn. Mater. 322 (2010) 698–704.
42 hepatoma effect of nanosized Mn–Zn ferrite magnetic fluid hyperthermia [43] L. Jin, X.R. Zhang, Analysis of temperature-sensitive magnetic fluids in a
43 associated with radiation in vitro and in vivo, Nanotechnology 24 (2013) porous square cavity depending on different porosity and Darcy number,
Appl. Therm. Eng. 50 (2013) 1–11.
44 255101.
[12] A. Miaskowski, B. Sawicki, Magnetic fluid hyperthermia modeling based on [44] L. Jin, X. Zhang, X. Niu, Lattice Boltzmann simulation for temperature-
45 sensitive magnetic fluids in a porous square cavity, J. Magn. Magn. Mater.
phantom measurements and realistic breast model, IEEE Trans. Biomed. Eng.
46 60 (2013) 1806–1813. 324 (2012) 44–51.
47 [13] Y.W. Lee, T.L. Chang, Novel perturbations between magnetic nanofluid and [45] E. Blums, A. Mezulis, G. Kronkalns, Magnetoconvective heat transfer from a
the thermal fluidic system at heat dissipation, Microelectron. Eng. 111 (2013) cylinder under the influence of a nonuniform magnetic field, J. Phys.
48 Condens. Matter 20 (2008) 204128.
58–63.
49 [14] W. Lian, Y. Xuan, Q. Li, Characterization of miniature automatic energy [46] D. Zablotsky, A. Mezulis, E. Blums, Surface cooling based on the thermo-
50 transport devices based on the thermomagnetic effect, Energy Convers. magnetic convection: numerical simulation and experiment, Int. J. Heat
Manage. 50 (2009) 35–42. Mass Transfer 52 (2009) 5302–5308.
51 [47] S.A. Suslov, Thermomagnetic convection in a vertical layer of ferromagnetic
[15] Y. Iwamoto, H. Yamaguchi, X.D. Niu, Magnetically-driven heat transport
52 device using a binary temperature-sensitive magnetic fluid, J. Magn. Magn. fluid, Phys. Fluids 20 (2008) 084101.
53 Mater. 323 (2011) 1378–1383. [48] H. Engler, A. Lange, D. Borin, S. Odenbach, Hindrance of thermomagnetic
[16] J.H. Lee, K.S. Hwang, S.P. Jang, B.H. Lee, J.H. Kim, S.U.S. Choi, C.J. Choi, Effective convection by the magnetoviscous effect, Int. J. Heat Mass Transfer 60 (2013)
54 499–504.
viscosities and thermal conductivities of aqueous nanofluids containing low
55 volume concentrations of Al2O3 nanoparticles, Int. J. Heat Mass Transfer 51 [49] S. Pal, A. Datta, S. Sen, A. Mukhopdhyay, K. Bandopadhyay, R. Ganguly,
56 (2008) 2651–2656. Characterization of a ferrofluid-based thermomagnetic pump for microflui-
[17] S.M.S. Murshed, K.C. Leong, C. Yang, Investigations of thermal conductivity dic applications, J. Magn. Magn. Mater. 323 (2011) 2701–2709.
57
and viscosity of nanofluids, Int. J. Therm. Sci. 47 (2008) 560–568. [50] W. Lian, Y. Xuan, Q. Li, Design method of automatic energy transport devices
58 [18] M. Kole, T.K. Dey, Thermal conductivity and viscosity of Al2O3 nanofluid based on the thermomagnetic effect of magnetic fluids, Int. J. Heat Mass
59 based on car engine coolant, J. Phys. D: Appl. Phys. 43 (2010) 315501. Transfer 52 (2009) 5451–5458.
[51] Y. Xuan, W. Lian, Electronic cooling using an automatic energy transport
60 [19] L. Syam Sundar, M.K. Singh, A.C.M. Sousa, Investigation of thermal con-
ductivity and viscosity of Fe3O4 nanofluid for heat transfer applications, Int. device based on thermomagnetic effect, Appl. Therm. Eng. 31 (2011)
61 1487–1494.
Commun. Heat Mass Transfer 44 (2013) 7–14.
62 [20] M.J. Pastoriza-Gallego, L. Lugo, J.L. Legido, M.M. Pineiro, Enhancement of [52] Q. Li, W. Lian, H. Sun, Y. Xuan, Investigation on operational characteristics of
63 thermal conductivity and volumetric behavior of FexOy nanofluids, J. Appl. a miniature automatic cooling device, Int. J. Heat Mass Transfer 51 (2008)
Phys. 110 (2011) 014309. 5033–5039.
64 [53] V. Roussellet, X. Niu, H. Yamaguchi, F Magoules, Natural convection of
[21] M. Abareshi, E.K. Goharshadi, S.M. Zebarjad, H. Khandan Fadafan, A. Youssefi,
65 Fabrication, characterization and measurement of thermal conductivity of temperature-sensitive magnetic fluids in porous media, Adv. Appl. Math.
66 Fe3O4 nanofluids, J. Magn. Magn. Mater. 322 (2010) 3895–3901. Mech. 3 (2011) 121–130.

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i
 14 M. Bahiraei, M. Hangi / Journal of Magnetism and Magnetic Materials ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 [54] S. Banerjee, A. Mukhopadhyay, S. Sen, R. Ganguly, The effects of magnetiza-
2 tion saturation on thermomagnetic convection in a locally heated square
enclosure, Numer. Heat Transfer, Part A 59 (2011) 693–718.
3 [55] M. Bahiraei, M. Hangi, Natural convection of magnetic nanofluid in a cavity
4 under non-uniform magnetic field: a novel application, J. Supercond. Novel
5 Magn. 27 (2014) 587–594.
[56] R. Ganguly, S. Sen, I.K. Puri, Thermomagnetic convection in a square
6 enclosure using a line dipole, Phys. Fluids 16 (2004) 2228.
7 [57] C.Y. Wen, W.P. Su, Natural convection of magnetic fluid in a rectangular Hele-
8 Shaw cell, J. Magn. Magn. Mater. 289 (2005) 299–302.
[58] A. Jafari, T. Tynjala, S.M. Mousavi, P. Sarkomaa, Simulation of heat transfer in
9 a ferrofluid using computational fluid dynamics technique, Int. J. Heat Fluid
10 Flow 29 (2008) 1197–1202.
11 [59] C.E. Nanjundappa, I.S. Shivakumara, K. Srikumar, Effect of MFD viscosity on
the onset of ferromagnetic fluid layer heated from below and cooled from
12
above with constant heat flux, Meas. Sci. Rev. 9 (2009) 75–80.
13 [60] H. Yamaguchi, X.D. Niu, X.R. Zhang, K. Yoshikawa, Experimental and
14 numerical investigation of natural convection of magnetic fluids in a cubic
15 cavity, J. Magn. Magn. Mater. 321 (2009) 3665–3670.
[61] S. Krakov, I.V. Nikiforov, Thermomagnetic convection in a porous enclosure
16 in the presence of outer uniform magnetic field, J. Magn. Magn. Mater. 289
17 (2005) 278–280.
18 [62] M. Ashouri, B. Ebrahimi, M.B. Shafii, M.H. Saidi, M.S. Saidi, Correlation for
Nusselt number in pure magnetic convection ferrofluid flow in a square
19 cavity by a numerical investigation, J. Magn. Magn. Mater. 322 (2010)
20 3607–3613.
21 [63] M. Hadavand, A.C.M. Sousa, Simulation of thermomagnetic convection in a
cavity using the Lattice Boltzmann model, J. Appl. Math. 2011 (2011) 538637.
22 [64] L. Syam Sundar, M.T. Naik, K.V. Sharma, M.K. Singh, T.Ch. Siva Reddy,
23 Experimental investigation of forced convection heat transfer and friction
24 factor in a tube with Fe3O4 magnetic nanofluid, Exp. Therm. Fluid Sci. 37
(2012) 65–71.
25 [65] L. Syam Sundar, N.T. Ravi Kumar, M.T. Naik, K.V. Sharma, Effect of full length
26 twisted tape inserts on heat transfer and friction factor enhancement with
27 Fe3O4 magnetic nanofluid inside a plain tube: an experimental study, Int. J.
Heat Mass Transfer 55 (2012) 2761–2768.
28
[66] Q. Li, Y. Xuan, Experimental investigation on heat transfer characteristics of
29 magnetic fluid flow around a fine wire under the influence of an external
30 magnetic field, Exp. Therm. Fluid Sci. 33 (2009) 591–596.
31 [67] A. Ghofrani, M.H. Dibaei, A. Hakim Sima, M.B. Shafii, Experimental investiga-
tion on laminar forced convection heat transfer of ferrofluids under an
32 alternating magnetic field, Exp. Therm. Fluid Sci. 49 (2013) 193–200.
33 [68] M. Lajvardi, J. Moghimi-Rad, I. Hadi, A. Gavili, T. Dallali Isfahani, F. Zabihi,
34 J. Sabbaghzadeh, Experimental investigation for enhanced ferrofluid heat
transfer under magnetic field effect, J. Magn. Magn. Mater. 322 (2010)
35 3508–3513.
36 [69] H. Aminfar, M. Mohammadpourfard, Y. Narmani Kahnamouei, A 3D numer-
37 ical simulation of mixed convection of a magnetic nanofluid in the presence
of non-uniform magnetic field in a vertical tube using two phase mixture
38 model, J. Magn. Magn. Mater. 323 (2011) 1963–1972.
39 [70] H. Aminfar, M. Mohammadpourfard, F. Mohseni, Two-phase mixture model
40 simulation of the hydro-thermal behavior of an electrical conductive ferro-
fluid in the presence of magnetic fields, J. Magn. Magn. Mater. 324 (2012)
41
830–842.
42 [71] M. Keshavarz Moraveji, M. Hejazian, Modeling of turbulent forced convective
43 heat transfer and friction factor in a tube for Fe3O4 magnetic nanofluid with
computational fluid dynamics, Int. Commun. Heat Mass Transfer 39 (2012)
44
1293–1296.
45 [72] M. Bahiraei, S.M. Hosseinalipour, M. Hangi, Numerical study and optimiza-
46 tion of hydrothermal characteristics of Mn–Zn ferrite nanofluid within
47 annulus in the presence of magnetic field, J. Supercond. Novel Magn. 27
(2014) 527–534.
48 [73] H. Aminfar, M. Mohammadpourfard, S. Ahangar Zonouzi, Numerical study of
49 the ferrofluid flow and heat transfer through a rectangular duct in the
50 presence of a non-uniform transverse magnetic field, J. Magn. Magn. Mater.
327 (2013) 31–42.
51 [74] R.L. Hamilton, O.K. Crosser, Thermal conductivity of heterogeneous two-
52 component systems, Ind. Eng. Chem. Fundam. 1 (1962) 187–191.
53 [75] D.A. Drew, S.L. Passman, Theory of Multi Component Fluids, first ed.,
Springer, Berlin, 1998.
54
55
[76] T. Strek, H. Jopek, Computer simulation of heat transfer through a ferrofluid,
Phys. Status Solidi B 244 (2007) 1027–1037.
[77] N.G. Prakash, K.B. Anoop, G. Sateesh, S.K. Das, Effect of surface orientation on
pool boiling heat transfer of nanoparticle suspensions, Int. J. Multiphase Flow
34 (2008) 145–160.
[78] S. Soltani, S.G. Etemad, J. Thibault, Pool boiling heat transfer performance of
Newtonian nanofluids, Int. Commun. Heat Mass Transfer 37 (2010) 29–33.
[79] S.M. You, J.H. Kim, K.M. Kim, Effect of nanoparticles on critical heat flux of
water in pool boiling of heat transfer, Appl. Phys. Lett. 83 (2003) 3374–3376.
[80] T. Lee, J.H. Lee, Y.H. Jeong, Flow boiling critical heat flux characteristics of
magnetic nanofluid at atmospheric pressure and low mass flux conditions,
Int. J. Heat Mass Transfer 56 (2013) 101–106.
[81] M. Sesen, Y. Teksen, B. Sahin, K. Sendur, M.P. Menguc, A. Kosar, Boiling heat
transfer enhancement of magnetically actuated nanofluids, Appl. Phys. Lett.
102 (2013) 163107.
[82] M. Sesen, Y. Teksen, K. Sendur, M.P. Menguc, H. Ozturk, H.F. Yagci Acar,
A. Kosar, Heat transfer enhancement with actuation of magnetic nanopar-
ticles suspended in a base fluid, J. Appl. Phys. 112 (2012) 064320.
[83] J.H. Lee, T. Lee, Y.H. Jeong, Experimental study on the pool boiling CHF
enhancement using magnetite-water nanofluids, Int. J. Heat Mass Transfer
55 (2012) 2656–2663.
[84] L. Junhong, G. Jianming, L. Zhiwei, L. Hui, Experiments and mechanism
analysis of pool boiling heat transfer enhancement with water-based
magnetic fluid, Heat Mass Transfer 41 (2004) 170–175.
[85] H. Habibi Khoshmehr, A. Saboonchi, M. Behshad Shafii, N. Jahani, The study
of magnetic field implementation on cylinder quenched in boiling ferrofluid,
Appl. Therm. Eng. 64 (2014) 331–338.
[86] R. Ganguly, S. Sen, I.K. Puri, Heat transfer augmentation in a channel with a
magnetic fluid under the influence of a line-dipole, J. Magn. Magn. Mater.
271 (2004) 63–73.
[87] A. Mukhopadhyay, R. Ganguly, S. Sen, I.K. Puri, Scaling analysis to characterize
thermomagnetic convection, Int. J. Heat Mass Transfer 48 (2005) 3485–3492.
[88] M. Mohammadi, M. Mohammadi, M.B. Shafii, Experimental investigation of a
pulsating heat pipe using ferrofluid (magnetic nanofluid), J. Heat Transfer
134 (2011) 014504.
[89] Z. Ming, L. Zhongliang, M. Guoyuan, C. Shuiyuan, The experimental study on
flat plate heat pipe of magnetic working fluid, Exp. Therm. Fluid Sci. 33
(2009) 1100–1105.
[90] K. Jahani, M. Mohammadi, M.B. Shafii, Z. Shiee, Promising technology for
electronic cooling: nanofluidic micro pulsating heat pipes, J. Electron.
Packag. 135 (2013) 021005.
[91] Y.C. Chiang, J.J. Chieh, C.C. Ho, The magnetic-nanofluid heat pipe with
superior thermal properties through magnetic enhancement, Nanoscale
Res. Lett. 7 (2012) 322.
[92] M. Taslimifar, M. Mohammadi, H. Afshin, M.H. Saidi, M.B. Shafii, Overall
thermal performance of ferrofluidic open loop pulsating heat pipes: an
experimental approach, Int. J. Therm. Sci. 65 (2013) 234–241.
[93] Y. Xuan, Q. Li, M. Ye, Investigations of convective heat transfer in ferrofluid
microflows using lattice-Boltzmann approach, Int. J. Therm. Sci. 46 (2007)
105–111.
[94] K. Fumoto, H. Yamagishi, M. Ikegawa, A mini heat transport device based on
thermo-sensitive magnetic fluid, Nanoscale Microscale Thermophys. Eng. 11
(2007) 201–210.
[95] J. Philip, P.D. Shima, B. Raj, Nanofluid with tunable thermal properties, Appl.
Phys. Lett. 92 (2008) 043108.
[96] L. Pislarun-Danescu, A.M. Morega, M. Morega, V. Stoica, O.M. Marinica,
F. Nouras, N. Paduraru, I. Borbath, T. Borbath, Prototyping a ferrofluid-
cooled transformer, IEEE Trans. Ind. Appl. 49 (2013) 1289–1298.
[97] M. Bahiraei, M. Hangi, Investigating the efficacy of magnetic nanofluid as a
coolant in double-pipe heat exchanger in the presence of magnetic field,
Energy Convers. Manage. 76 (2013) 1125–1133.
[98] O. Celik, M.M. Can, T. Firat, Size dependent heating ability of CoFe2O4
nanoparticles in AC magnetic field for magnetic nanofluid hyperthermia,
J. Nanopart. Res., article in press (2014) 10.1007/s11051–014–2321–6.
[99] A.A. Golneshan, M. Lahonian, Diffusion of magnetic nanoparticles in a multi-
site injection process within a biological tissue during magnetic fluid
hyperthermia using lattice Boltzmann method, Mech. Res. Commun. 38
(2011) 425–430.
[100] I. Sharifi, H. Shokrollahi, S. Amiri, Ferrite-based magnetic nanofluids used in
hyperthermia applications, J. Magn. Magn. Mater. 324 (2012) 903–915.

Please cite this article as: M. Bahiraei, M. Hangi, Journal of Magnetism and Magnetic Materials (2014), http://dx.doi.org/10.1016/j.
jmmm.2014.08.004i