International Journal of Heat and Mass Transfer 127 (2018) 67–75

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

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

Natural convection of a nanofluid between two eccentric cylinders

saturated by porous material: Buongiorno’s two phase model
Mohsen Izadi a, Sara Sinaei b, S.A.M. Mehryan c, Hakan F. Oztop d,e,⇑, Nidal Abu-Hamdeh e
a
Mechanical Engineering Department, Faculty of Engineering, Lorestan University, Khorramabad, Iran
b
Mechanical Engineering Department, Bushehr Branch, Islamic Azad University, Bushehr, Iran
c
Young Researchers and Elite Club, Yasooj Branch, Islamic Azad University, Yasooj, Iran
d
Department of Mechanical Engineering, Technology Faculty, Fırat University, Elazig, Turkey
e
Department of Mechanical Engineering, King Abdulaziz University, Jeddah, Saudi Arabia

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

Article history: This paper concerns with a numerical study on heat transfer by natural convection of different nanofluids
Received 17 June 2018 inside a porous medium between two horizontal eccentric cylinders. The Buongiorno’s model was uti-
Received in revised form 9 July 2018 lized to track the nanoparticles concentration. Heat transfer by natural convection of water-copper,
Accepted 11 July 2018
water-diamond and water–silicon dioxide nanofluids inside a porous medium between two cylinders
was studied. A set of characteristic equations were solved using finite element method. Range of param-
eters which are studied are Ra = 103-106, Da = 10
6-10
3, e = 0.1–0.9 and volume fractions of 0.0–0.1,
Keywords:
Nr = Nt = Nb = 0.1–0.5, and Le = 1–10. The porous medium was analyzed by Darcy-Brinkman model.
Natural convection of nanofluid
Porous media
The results showed that thermophoresis parameter and Brownian parameter have no significant effect
Two eccentric cylinders on isothermal lines and the streamlines. The probability of collision of nanoparticles decreases with
Buongiorno’s model increasing the Lewis number, which consequently decreases the average Nusselt number. The maximum
and minimum enhancement in natural convection was detected in water-diamond and water–silicon
dioxide respectively.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction the other hand, a porous medium with a solid section has a higher
conductivity coefficient compared to an empty space. Overall
Movement of fluid inside the porous medium can be detected conductivity also increases in the former medium. Heat transfer
both in natural and industrial area. Behavior of the fluid flowing is enhanced as the nanofluid passes through a porous medium
through the porous medium was considered in various studies [19–21]. The flow of nanofluids in a porous medium is a combined
due to its unique structure and its application in various sections problem that should be studied further in order to investigate
of mechanical engineering applications including solar heating thermal behavior of the nanofluid in this medium [22–26].
systems, energy storage units, fuel cells, geophysical system and Use of nanofluids as cooling agents in a porous medium was
catalytic converter. Scholars and engineers performed many proposed by Tsai and Chein [27] and Ghazvini and Shokouhmand
analytical, numerical and empirical studies to examine this [28]. Modeling such thermal wells requires a basic knowledge on
complex phenomenon. convective heat transfer of the nanofluid in a porous medium.
There are many methods for enhancing the heat transfer rate. Hyperthermia nanoparticles for treatment of cancer is another
Addition of nanoparticles to the base fluid is one of the methods discussion on use of nanofluids for convective heat transfer
used to enhance heat transfer [1–18]. Nanofluid is a new genera- [29,30]. In this case, use of hyperthermia nanoparticles aims to
tion of fluids obtained by distribution of particles in nano-scale impose a maximum damage to the tumor (this requires a
diameters from 1 to 100 nm inside a base fluid. Use of nanofluid temperature at least 90% above temperature of the tumor
in heat transfer systems was so remarkable that it was the subject (43 °C)) with minimal damage to body tissues. They developed
of many extensive studies by many academics in a short time. On optimal protocols for this treatment, which requires basic
knowledge on convective heat transfer in a porous medium.
Hajipour et al. [31] were the first academics who studied
⇑ Corresponding author at: Department of Mechanical Engineering, Technology combined convective heat transfer of nanofluid through a vertical
Faculty, Fırat University, Elazig, Turkey.
channel, partly filled with metal foam both numerically and
E-mail address: hfoztop1@gmail.com (H.F. Oztop).

https://doi.org/10.1016/j.ijheatmasstransfer.2018.07.066
0017-9310/Ó 2018 Elsevier Ltd. All rights reserved.
68 M. Izadi et al. / International Journal of Heat and Mass Transfer 127 (2018) 67–75

Nomenclature

C dimensional nanoparticles concentration u, v components of velocity in x and y directions, respec-

C0 reference nanoparticles concentration tively, m s
1
Cp specific heat (J/kg K) U,V dimensionless velocity in x and y directions
Da Darcy number T dimensional temperature
DB Brownian diffusion coefficient, m
2 s
1
DT thermophoresis diffusion coefficient, m
2 s
1 Greek symbols
e absolute eccentricity, m a thermal diffusivity, m
2 s
1
~e dimensionless eccentricity b thermal expansion coefficient, K–1
g gravitational acceleration l dynamic viscosity, kg m
1 s
1
L characteristic length (L = r0-ri), m # kinematic viscosity, m2 s
1
Le Lewis number q density, kg m
3
K permeability, m2 e porosity
k thermal conductivity, W m
1 K
1 u normalized nanoparticles concentration
N normal direction to the surface h nondimensional temperature
Nuavg average Nusselt number s parameter defined by s ¼ ðqcÞp =ðqcÞnf
Nb Brownian parameter
Nt thermophoresis parameter Subscripts
Nr buoyancy ratio
f base fluid
p dimensional pressure, Pa c cold
P nondimensional pressure h hot
Pr Prandtl number nf nanofluid
Ra Rayleigh number
p nanoparticles
x, y Cartesian coordinates, m
X,Y dimensionless Cartesian coordinates

experimentally. The results showed that nanofluid flow in porous 2. Governing equations and boundary conditions
metal foam can increase the heat transfer rate. This can be also
used in the industry. Nazari et al. [32] studied heat transfer by Fig. 1 shows the schematic view of the desired geometry. The
forced convection in a nanofluid in a circular tube filled with alu- external cylinder was kept at Tc temperature and the internal
minum particles in a porous medium with open cavities with cylinder at Th temperature. The space between the two cylinders
50% porosity and constant wall temperature in a laboratory. They is porous and filled with a nanofluid. In this study, the cases of
showed that use of a porous medium inside a tube considerably three different nanofluids including water-copper, water-
enhances the heat transfer compared to an empty tube. Increase diamond and water-silicon dioxide were studied. The radius of
in volume fraction of aluminum nanoparticles increases the Nus- the outer and inner cylinder was ro and ri respectively. The radial
selt number. Umouthy et al. [33] performed a numerical study ratio (RR ¼ ro =r i ) was considered as 2.5.
on heat transfer by natural convection of a nanofluid inside a rect- The distance between centric points of the cylinders is called
angular enclosure using the Darcy-Forchheimer-Brinkman Model. eccentricity shown as ‘‘e” which has a positive value. Nanofluid is
The heat transfer by natural convection of a nanofluid was investi- considered as Newtonian fluid and flow are also two-
gated in the former study in a porous medium. Armaghani [34] dimensional. Thermophysical properties of the nanofluid are
compared forced convection of heat transfer using single-phase
and two-phase nanoparticles inside a porous channel. Numerical
results of these two methods were compared. The developed
Darcy- Brinkman–Forchheimer Model was used to solve the equa-
tion. The numerical results showed an increase in local Nusselt
number by decreasing the Lewis number. A similar trend was
detected in the Schmidt number. Other related works can be found
in [41–42].
A review of literature showed that heat transfer by natural con-
vection was not studied in a nanofluid inside a porous medium
between two eccentric cylinders using two-phase method. Given
that such geometry is used extensively in heat exchangers, it is
necessary to study the effect of different parameters of porous
medium and properties of the nanofluid for enhancing heat trans-
fer in such geometries. Buongiorno method [35] was used to ana-
lyze the nanofluid behavior in a heterogeneous manner. Hence,
there are some changes in the equations of momentum and heat
transfer in the heterogeneous method compared to homogeneous
method and the fluid base. Distribution of particles and variations
in distribution of particles were studied by changing various
parameters. The effects of Brownian movement and thermophore-
sis on thermal and hydrodynamic conditions were also studied. Fig. 1. Schematic geometry of examined problem.
 M. Izadi et al. / International Journal of Heat and Mass Transfer 127 (2018) 67–75 69

Table 1
Thermodynamic properties of different nanoparticles and the base fluid [37].

Pure water Silicon dioxide Diamond Copper Thermos-physical properties

997.1 2200 3510 8933 q ðkg=m3 Þ
0.613 1.2 1000 401 k ðW=m KÞ
4179 703 497.26 385 C P ðJ=kg KÞ
21  10
5 5.5  10
6 1  10
6 1.67  10
5 b ð1=KÞ
1.47  10
7 – – – a (m2 =sÞ

assumed as constant parameters in this problem except for vari- porous medium. Porous medium was assumed Homogeneous
able density defined by the Boussinesq approximation [36]. The and isotropic. Two phases of porous medium also kept in the ther-
term on movement of nanoparticles under the influence of various mal equilibrium condition. Thermodynamic properties of the
factors is also added to the heat transfer equation. Also, nanoparti- nanoparticles and the base fluid are also shown in Table 1.
cle continuity equation is also added to the governing equations. The governing equations consist of the mass Eq. (1), the
The Darcy-Brinkman Model was used without term inertia in the momentum equation in the horizontal direction (2), in the vertical
direction (3), the energy conservation Eq. (4) and the continuity
equation of the nanoparticles (5). These equations are obtained
from relations (1)–(6) as follows:
8
4514 @u @ v
þ ¼0 ð1Þ
7 6232 @x @y
9008
 !
qnf @u @u @p lnf @ 2 u @ 2 u lnf
6 6535 u þv ¼
þ þ
u ð2Þ
14506 e2 @x @y @x e @x2 @y2 K
5

 !
qnf @ v @v @p lnf @ 2 v @ 2 v lnf
Nuavg

u þ v ¼
þ þ

4 e2 @x @y @y e @x2 @y2 K
h i
þ
ðqp
qf ÞðC
C 0 Þ þ ð1
C 0 Þqf bf ðT
T c Þ g
3
ð3Þ
2 !


1 @T @T @2 T @2 T
u þv ¼ anf þ
1 e @x @y @x2 @y2
(

2
2 !)
@C @T @C @T DT @T @T
0 1 2 3 4 5 6 þ s DB þ þ þ
@x @x @y @y Tc @x @y
Reversed arc length
ð4Þ
Fig. 2. Grid independent study.

 ! !
1 @C @C @2C @2C DT @2T @2T
u þv ¼ DB þ þ þ ð5Þ
e @x @y @x2 @y2 Tc @x2 @y2

s ¼ ðqC P Þp =ðqC P Þnf

Dimensionless governing equations are achieved using the fol-
lowing dimensionless variables (1)–(5).
x y uL vL e
~e ¼ ; u ¼
C
X¼ ; Y¼ ; U¼ ; V¼ ;
L L af af L C0
ð6Þ
T
Tc
DC ¼ C h
C c ; DT ¼ T h
T c ; h¼ ; L ¼ ro
ri
Th
Tc

@U @V
þ ¼0 ð7Þ
@X @Y

 !
1 @U @U qf @P #nf Pr @ 2 U @ 2 U #nf Pr
U þ V ¼
þ þ
U
e2 @X @Y qnf @X #f e @X 2 @Y 2 #f Da
ð8Þ

 !
1 @V @V qf @P #nf Pr @ 2 V @ 2 V #nf Pr
U þ V ¼
þ þ
V
e2 @X @Y qnf @Y #f e @X 2 @Y 2 #f Da
qf
þ Ra  Pr  Nr  ð1
uÞ þ  Ra  Pr  h ð9Þ
Fig. 3. Comparison of average Nusselt number in the present and Besak et al. [39] qnf
works.
70 M. Izadi et al. / International Journal of Heat and Mass Transfer 127 (2018) 67–75

 !
 ! !
@h @h anf @2h @2h @ u @h @ u @h @u @u e @2u @2u e  Nt @ 2 h @ 2 h
U þV ¼e þ þ Nb  e þ U þV ¼ þ þ þ ð11Þ
@X @Y af @X 2 @Y 2 @X @X @Y @Y @X @Y Le @X 2 @Y 2 Le  Nb @X 2 @Y 2
"

2 #
2
@h @h where Rayleigh number Ra, Prandtl number Pr, Darcy number Da,
þ Nt  e þ ð10Þ
@X @Y Lewis number Le, buoyancy ratio Nr, thermophoresis parameter
Nb and Brownian parameter Nb are calculated as follows:

Table 2
ð1
C 0 Þqf gbf DTL3 vf K af
Ra ¼ ; Pr ¼ ; Da ¼ ; Le ¼ ;
Comparison between the results of present study and the study carried out by Habibi lf af af 2
L DB
and Pop [40]. ð12Þ
ðqp
qf ÞC 0 sDB C 0 sDT DT
Nr ¼ ; Nb ¼ ; Nt ¼
Volume Present Habibi and Pop’s Relative error
qf bf DTð1
C 0 Þ af af T C
fraction work work percentage
0.01 5.60 5.66 1.07 Dimensionless boundary conditions are expressed using the
0.015 5.78 5.81 0.51 following equation given geometry of the problem:
0.02 5.86 5.89 0.51
0.025 6.04 5.97 1.15 @u @h
h ¼ 1; U ¼ V ¼ 0; Nb þ Nt ¼ 0 On the internal cylinder
0.03 6.07 6.02 0.82 @N @N

Fig. 4. Streamlines, isothermal-lines and distribution of nanoparticles at u = 0.05, Ra = 105, Le = 10, e = 0.65, Nb = Nt = Nr = 0.3 and different Darcy numbers (a) Da = 10
5, (b)
Da = 10
4, (c) Da = 10
3.
 M. Izadi et al. / International Journal of Heat and Mass Transfer 127 (2018) 67–75 71

Z
@u @h knf @T
h ¼ 0; U ¼ V ¼ 0; Nb þ Nt ¼ 0 On the external cylinder Nuav g ¼
ð15Þ
@N @N kf @N
ð13Þ
N in the above relation is the perpendicular orientation to the
Thermophysical relationships of the nanofluid can be obtained surface.
as follows [38]:

qnf ¼ ð1
CÞqf þ C qp 3. Numerical method, grid testing and validation

bnf ¼ ð1
CÞbf þ Cbp
knf The natural convection heat transfer rate was assessed using
anf ¼ the average Nusselt number for water-copper nanofluid at
ðqC P Þnf
u = 0.05, Re = 106, Pr = 6.21, Da = 10
3, e = 0.65, Le = 10,
ðqC P Þnf ¼ ð1
CÞðqC P Þf þ CðqC P Þp ð14Þ
Nr = Nb = Nt = 0.1, RR = 2.5 and ~e ¼ 0:5 in order to examine
lf independence of results from the grid number. For this purpose,
lnf ¼
ð1
CÞ2:5 the average Nusselt numbers for five different cells of the grids
  were 4514, 622, 6350, 9008 and 14506. Fig. 2 shows variation in
ðkp þ 2kf Þ
2Cðkf
kp Þ
knf ¼ kf the average Nusselt number for different grids number. As shown,
ðkp þ 2kf Þ þ Cðkf
kp Þ
a grid with 6232 cells can guarantee the independence of numeri-
In order to evaluate the heat transfer rate through the hot cal results on the grid size. The Simulation results were compared
boundary, the average Nusselt number defined as with those of the previous studies in order to verify the results of

Fig. 5. Streamlines, isothermal-lines and distribution of nanoparticles at u = 0.05, Da = 10
3, e = 0.65, Nb = Nt = Nr = 0.3, Ra = 105 and different Lewis numbers (a) Le = 1, (b)
Le = 5, (c) Le = 10.
72 M. Izadi et al. / International Journal of Heat and Mass Transfer 127 (2018) 67–75

numerical simulations. Fig. 3 shows comparison of average Nusselt and isoconcentrations of nanoparticles with respect to Darcy num-
number in the present work and the simulation carried out by ber for water-copper nanofluid. Two symmetric clockwise (on the
Besak et al. [39]. It is worth mentioning that u is the angle of trape- right with a negative sign) and counter-clockwise (on the left with
zoidal enclosure studied in Besak et al.’s work. Simulations results a negative sign) vortices are formed on both sides of the internal
are consistent in both studies. Table 2 shows comparison in aver- cylinder. This shows rotational direction of the nanofluid based
age Nusselt number between the present study and the that car- on the buoyancy force with respect to the boundary conditions.
ried out by Habibi and pop [40]. The results of the present study The nanofluid flows more easily by increasing the Darcy number
were consistent with the results of Habibi and Pop’s work. and the fluid friction decreases in the porous medium due to
increased permeability. In this case, the resistant force against
4. Results and discussion the nanofluid flow in the porous medium reduces and the strength
of the fluid flow greatly increases.
In this section, numerical study of natural convection heat Natural convection heat transfer is less enhanced at Darcy num-
transfer is evaluated using different nanofluid types consisting of bers of 10
4 and 10
5. In this case, heat is transferred by conductiv-
water-copper, water-diamond and water-silicon dioxide. the range ity. As a result, isothermal lines are parallel without any distortion.
of key studied parameters are such these 103 < Ra < 106, 10
6 < Da The strength of streamlines increases as Darcy number increases,
< 10
3, 1 < Le < 10, Nr = Nt = Nb = 0.1–0.5, 0 < u < 0.1, 0.1 < e < 0.9. which leads to further mixing of the nanofluid throughout the cav-
The other parameters are constant so that Pr = 6.21, RR = 2.5, ity that enhances uniformity and homogeneity of the nanofluid
~e ¼ 0:5. Fig. 4 shows variation of the streamlines, isothermal lines (third column). In addition, thickness of thermal boundary layer

Fig. 6. Streamlines, isothermal-lines and distribution of nanoparticles u = 0.05, Le = 10, Da = 10
3, e = 0.65, at Ra = 105, e = 0.65, Nr = 0.1, (a) Nb = Nt = 0.1 (b) Nb = 0.1 and
Nt = 0.5 (c) Nb = 0.5 and Nt = 0.1.
 M. Izadi et al. / International Journal of Heat and Mass Transfer 127 (2018) 67–75 73

decreases and heat transfer rate by natural convection increases nanoparticles. The thermophoresis parameter depends on temper-
with increasing Darcy number to 10
3. As a result, isothermal lines ature gradient. Comparison of the first and second rows in the fig-
are bended and distorted (second column). ure shows that the flow strength decreases (the first column),
Fig. 5. pictures the effect of Lewis number on the streamlines, thickness of the boundary layer increases and density of isotherms
isothermal lines and isoconcentration of nanoparticles. A slight in vicinity of the heater decreases by increasing the thermophore-
increase in the Lewis number increases the flow strength. This is sis parameter. The isotherms move towards outer cold cylinder
because the flow strength increases by decreasing penetration of (the second column), which reduces heat transfer. Concentration
nanoparticles and decreasing their collisions. The figure shows no of nanoparticles increases in lower part of the external cylinder
considerable change in temperature contours with increasing by increasing the thermophoresis parameter. In other words, more
Lewis numbers. On the other hand, the boundary layer thickness nanoparticles move towards lower part of the cylinder where the
of nanoparticles decreases by increasing the Lewis number. As a temperature is lower as thermophoresis parameter increases. As
result, the flows with higher Lewis number prevent from further a result, nanoparticle migration increases and nanofluid uniformity
dispersion of nanoparticles. Therefore, distribution of the nanofluid decreases as the thermophoresis parameter increases. Comparison
is more homogeneous. Fewer nanoparticles can be transferred of the first and third rows shows that the strength of streamlines
from high energy zones (hot walls) to low energy zones (cold slightly increases by increasing the Brownian parameter. An
walls) by increasing the Lewis number and thermal penetration. increase in the Brownian parameter does not considerably change
Fig. 6 demonstrates the effect of Brownian and thermophoresis the isotherms. Finally, nanoparticle movement increases and
parameters on the streamlines, isothermal-lines and distribution of nanoparticles are distributed more homogenously by increasing
the Brownian parameter.
Fig. 7 confirms that the average Nusselt number enhances by
increasing the Darcy number. As mentioned in the previous sec-
tion, the flow strength increases and heat transfer by natural con-
5
vection considerably increases by increasing the Da. The effect of
3 Da increases at higher Rayleigh number as natural convection
Ra=10
dominates thermal conductivity.
Ra=104
The effects of the thermophoresis parameter and volume frac-
4 Ra=105
tion of nanoparticles at different Ra are pictured in Fig. 8. Nuavg
Ra=106
is reduced by increasing the thermophoresis parameter. As seen
in Fig. 8b, the flow strength decreases as this parameter increases.
Nuavg

As density of isotherms decreases near the internal cylinder, the

3 Nusselt number declines. The average Nusselt number is enhanced
by increasing volume fraction of nanoparticles (Fig. 8a).
Fig. 9 displays the effect of Brownian and the buoyancy param-
eters on the Nusselt number The Figure shows that an increase in
2 Brownian parameter does not considerably change the average
Nusselt number. As mentioned earlier, increase in the Brownian
parameter slightly increases the flow strength but does not signif-
icantly change the isotherms. Therefore, the Brownian parameter
does not change the average Nusselt number. Consequently, the
10-6 10-5 10-4 10-3
Da average Nusselt number remains constant. Fig. 9a also shows that
an increase in the porosity increase contribution of pores and the
Fig. 7. The effect of Darcy number on average Nusselt number at u = 0.1, Le = 10,
volume of the fluid and as a result it decreases resistance against
e = 0.65, Nb = Nt = Nr = 0.3 and at different values Ra.
movement of nanofluid in the porous medium. Therefore, the flow

5
5

4 Ra=103 Ra=103
Ra=104 4 Ra=104
5
Ra=106 Ra=10
Nuavg
Nuavg

Ra=10
5
Ra=106
3
3

2
2

1
0 0.02 0.04 0.06 0.08 0.1 0.1 0.2 0.3 0.4 0.5
φ Nt
(a) (b)
Fig. 8. Variations Nu according (a) to volume fraction for different values of Ra when Nb = Nt = Nr = 0.3 and different values (b) thermophoresis parameter for different values
of Ra when Nb = Nr = 0.3, u = 0.1 at constant parameters Le = 10, e = 0.65, Da = 10
3.
74 M. Izadi et al. / International Journal of Heat and Mass Transfer 127 (2018) 67–75

Fig. 9. Variations Nu according (a) to Brownian parameters for different values of porosity coefficient when Nt = Nr = 0.3 (b) to buoyancy parameters for different values of
porosity when Nt = Nb = 0.3 at constant parameters, Le = 10, e = 0.65, Da = 10
3, u = 0.1.

Table 3
Average Nusselt number at different Darcy numbers for different nanofluids and pure
water at u=0.1, Le = 10, e=0.65, Nb = Nt = Nr = 0 and Ra = 106.
Nr=0.1 Nanofluid Darcy number
2.19 Nr=0.2
10
5 10
4 10
3
Nr=0.3
Water 1.1714 2.1794 4.6415
Nr=0.4 Water-cooper 1.3750 2.4600 4.9604
Water-diamond 1.3875 2.4793 5.0314
Water- Silicon dioxide 1.2427 2.3905 4.6601
Nuavg

2.16

Table 3 represents variations of the average Nusselt number

with respect to the Darcy number at Ra = 106. The water-
diamond nanofluid has the highest Nusselt number (5.0314) at
Da = 10
3 and the lowest Nusselt number (1.1714) at Da = 10
5
2.13 in water base fluid. It seems that distribution of diamond nanopar-
ticle (due to its higher thermal conductivity) has significantly
increased average Nusselt number of the base fluid. Subsequently,
2 4 6 8 10 copper and silicon dioxide nanoparticles less increased the average
Le Nusselt number compared to the base fluid.

Fig. 10. The effect of Lewis number on average Nusselt number at different
buoyancy parameters at u = 0.1, e = 0.65, Nb = Nt = 0.3 and Ra = 105. 5. Conclusion

In this paper, natural convection heat transfer in a nanofluid

strength increases, which considerably enhances the heat transfer. was studied in a porous medium of two eccentric cylinders. The
As mentioned in the previous section, higher porosity has a greater applications are in various sections of mechanical engineering
contribution to increase in the Nusselt number at higher Ra with including solar heating systems, energy storage units, fuel cells,
increasing the buoyancy forces Fig. 9(b) shows that variations in and catalytic converter. Effects of porous medium parameters
the buoyancy parameter do not considerably change the Nusselt and nanoparticle properties on heat transfer by natural convection
number. were investigated. Buongiorno’s two phase model is utilized to
As presented in Fig. 10, the average Nusselt number decreases track the nanoparticles concentration. The most important results
by increasing the Lewis number. Lewis number is defined as ratio can be summarized as follows:
of thermal diffusion coefficient to Brownian dispersion. Nanoparti-
cles collision in the base fluid decreases by increasing the Lewis 1. An increase in the Da increases natural convection heat transfer
number. As a result, less heat is transferred by decreasing in move- due to increased flow permeability and ease of fluid flow.
ment and transfer of nanoparticles. As mentioned in the previous 2. The fluid flows more easily by increasing the porosity due to the
sections, the Nusselt number also decreases by increasing the presence of more pores in the porous medium, which conse-
buoyancy parameter. quently increases natural convection heat transfer.
Inclusion effect of three copper, diamond and silicon dioxide 3. Heat is more transferred by reducing the thermophoresis
nanoparticles in water base fluid were assessed in this section. In parameter due to reduced thickness of thermal boundary layer,
the previous section, the water-copper nanofluid was studied in which consequently increases the average Nusselt number.
detail. In this section, these three types of nanofluids are compared 4. Increase in the Brownian parameter has no considerable effect
based on water base fluid. on the average Nusselt number.
 M. Izadi et al. / International Journal of Heat and Mass Transfer 127 (2018) 67–75 75

5. The thermophoresis and Brownian parameter have no consider- [19] M.H. Toosi, M. Siavashi, Two-phase mixture numerical simulation of natural
convection of nanofluid flow in a cavity partially filled with porous media to
able effect on isothermal lines and the flow strength.
enhance heat transfer, J. Mol. Liq. 238 (2017) 553–569.
6. The probability of collisions between nanoparticles decreases [20] M.A. Sheremet, T. Grosan, I. Pop, Free convection in a square cavity filled with a
by increasing the Lewis number. As a result, the average Nusselt porous medium saturated by nanofluid using Tiwari and Das’ nanofluid model,
number decreases. Transp. Porous Media 106 (2015) 595–610.
[21] M.A. Sheremet, I. Pop, M.M. Rahman, Three-dimensional natural convection in
7. The highest and lowest natural convection heat transfer belongs a porous enclosure filled with a nanofluid using Buongiorno’s mathematical
to water-diamond and water–silicon dioxide nanofluid, model, Int. J. Heat Mass Transf. 82 (2015) 396–405.
respectively. [22] M. Izadi, G. Hoghoughi, R. Mohebbi, M. Sheremet, Nanoparticle migration and
natural convection heat transfer of Cu-water nanofluid inside a porous
undulant-wall enclosure using LTNE and two-phase model, J. Mol. Liq. (2018).
Conflict of interest [23] M. Izadi, S. Mehryan, M.A. Sheremet, Natural convection of CuO-water
micropolar nanofluids inside a porous enclosure using local thermal non-
equilibrium condition, J. Taiwan Inst. Chem. Eng. (2018).
We confirm that there is no any conflict associated with this [24] S. Mehryan, M. Izadi, M.A. Sheremet, Analysis of conjugate natural convection
manuscript. within a porous square enclosure occupied with micropolar nanofluid using
local thermal non-equilibrium model, J. Mol. Liq. 250 (2018) 353–368.
[25] S.A. Mehryan, M. Ghalambaz, M. Izadi, Conjugate natural convection of
References nanofluids inside an enclosure filled by three layers of solid, porous medium
and free nanofluid using Buongiorno’s and local thermal non-equilibrium
[1] Yuan Ma, Rasul Mohebbi, Mohammad Mehdi Rashidi, Zhigang Yang, Numerical models, J. Therm. Anal. Calorim. (2018) 1–21.
investigation of MHD effects on nanofluid heat transfer in a baffled U-shaped [26] G. Hoghoughi, M. Izadi, H.F. Oztop, N. Abu-Hamdeh, Effect of geometrical
enclosure using lattice Boltzmann method, J. Therm. Anal. Calorim. (2018). parameters on natural convection in a porous undulant-wall enclosure
[2] Yuan Ma, Rasul Mohebbi, Mohammad Mehdi Rashidi, Zhigang Yang, Effect of saturated by a nanofluid using Buongiorno’s model, J. Mol. Liq. 255 (2018)
hot obstacle position on natural convection heat transfer of MWCNTs-water 148–159.
nanofluid in U-shaped enclosure using lattice Boltzmann method, Int. J. [27] T.-H. Tsai, R. Chein, Performance analysis of nanofluid-cooled microchannel
Numer. Meth. Heat Fluid Flow (2018). heat sinks, Int. J. Heat Fluid Flow 28 (2007) 1013–1026.
[3] Monireh Asadi, Rasul Mohebbi, Omid S. Fard, Lattice Boltzmann simulation of [28] M. Ghazvini, H. Shokouhmand, Investigation of a nanofluid-cooled
nanofluid natural convection heat transfer in a channel with a sinusoidal microchannel heat sink using Fin and porous media approaches, Energy
obstacle, Int. J. Mod. Phys. C (2018). Convers. Manage. 50 (2009) 2373–2380.
[4] R. Mohebbi, M.M. Rashidi, Numerical simulation of natural convection heat [29] M. Salloum, R.H. Ma, D. Weeks, L. Zhu, Controlling nanoparticle delivery in
transfer of a nanofluid in an L-shaped enclosure with a heating obstacle, J. magnetic nanoparticle hyperthermia for cancer treatment: experimental
Taiwan Inst. Chem. Eng. 72 (2017) 70–84. study in agarose gel, Int. J. Hyperth. 24 (2008) 337–345.
[5] R. Mohebbi, H. Lakzayi, N.A.C. Sidik, Japar, Wan Mohd Arif Aziz, Lattice [30] M. Salloum, R. Ma, L. Zhu, An in-vivo experimental study of temperature
Boltzmann method based study of the heat transfer augmentation associated elevations in animal tissue during magnetic nanoparticle hyperthermia, Int. J.
with Cu/water nanofluid in a channel with surface mounted blocks, Int. J. Heat Hyperth. 24 (2008) 589–601.
Mass Transfer 117 (2018) 425–435. [31] M. Hajipour, A.M. Dehkordi, Mixed-convection flow of Al2O3–H2O nanofluid in
[6] S.U. Chol, Enhancing thermal conductivity of fluids with nanoparticles, ASME- a channel partially filled with porous metal foam: Experimental and numerical
Publ.-Fed 231 (1995) 99–106. study, Exp. Therm Fluid Sci. 53 (2014) 49–56.
[7] S. Lee, S.-S. Choi, S.A. Li, J.A. Eastman, Measuring thermal conductivity of fluids [32] M. Nazari, M. Ashouri, M.H. Kayhani, A. Tamayol, Experimental study of
containing oxide nanoparticles, J. Heat Transfer 121 (1999) 280–289. convective heat transfer of a nanofluid through a pipe filled with metal foam,
[8] Z. Haddad, H.F. Oztop, E. Abu-Nada, A. Mataoui, A review on natural convective Int. J. Thermal Sci. 88 (2015) 33–39.
heat transfer of nanofluids, Renew. Sustain. Energy Rev. 16 (2012) 5363–5378. [33] J.C. Umavathi, O. Ojjela, K. Vajravelu, Numerical analysis of natural convective
[9] H.F. Oztop, E. Abu-Nada, Numerical study of natural convection in partially flow and heat transfer of nanofluids in a vertical rectangular duct using Darcy-
heated rectangular enclosures filled with nanofluids, Int. J. Heat Fluid Flow 29 Forchheimer-Brinkman model, Int. J. Thermal Sci. 111 (2017) 511–524.
(2008) 1326–1336. [34] T. Armaghani, M.J. Maghrebi, M. Nazari, Comparison between single and two
[10] M. Izadi, A. Behzadmehr, D. Jalali-Vahida, Numerical study of developing phase nanofluid convective heat transfer in a porous channel, J. Model. Eng. 14
laminar forced convection of a nanofluid in an annulus, Int. J. Thermal Sci.nces (2016) 11–20.
48 (2009) 2119–2129. [35] J. Buongiorno, Convective transport in nanofluids, J. Heat Transf. 128 (2006)
[11] M. Izadi, M.M. Shahmardan, M.J. Maghrebi, A. Behzadmehr, Numerical study of 240–250.
developed laminar mixed convection of Al2O3/water nanofluid in an annulus, [36] A. Bejan, Convective heat transfer in porous media, Handbook Single-Phase
Chem. Eng. Commun. 200 (2013) 878–894. Convect. Heat Transf. 16 (1–16) (1987) 34.
[12] M. Izadi, A. Behzadmehr, M.M. Shahmardan, Effects of discrete source-sink [37] W.M. El-Maghlany, M.M. Sorour, O. Hozien, Experimental study of natural
arrangements on mixed convection in a square cavity filled by nanofluid, convection in an annulus between two eccentric horizontal square ducts, Exp.
Korean J. Chem. Eng. 31 (2014) 12–19. Thermal Fluid Sci. 65 (2015) 65–72.
[13] M. Izadi, A. Behzadmehr, M.M. Shahmardan, Effects of inclination angle on [38] H. Hashemi, Z. Namazian, S. Mehryan, Cu-water micropolar nanofluid natural
laminar mixed convection of a nanofluid flowing through an annulus, Chem. convection within a porous enclosure with heat generation, J. Mol. Liq. 236
Eng. Commun. 202 (2015) 1693–1702. (2017) 48–60.
[14] K. Khanafer, K. Vafai, M. Lightstone, Buoyancy-driven heat transfer [39] T. Basak, S. Roy, A. Singh, A.R. Balakrishnan, Natural convection flows in porous
enhancement in a two-dimensional enclosure utilizing nanofluids, Int. J. trapezoidal enclosures with various inclination angles, Int. J. Heat Mass Transf.
Heat Mass Transf. 46 (2003) 3639–3653. 52 (2009) 4612–4623.
[15] R. Mohebbi, M.M. Rashidi, M. Izadi, N.A.C. Sidik, H.W. Xian, Forced convection [40] M.H. Matin, I. Pop, Natural convection flow and heat transfer in an eccentric
of nanofluids in an extended surfaces channel using lattice Boltzmann method, annulus filled by Copper nanofluid, Int. J. Heat Mass Transf. 61 (2013) 353–
Int. J. Heat Mass Transf. 117 (2018) 1291–1303. 364.
[16] R. Mohebbi, M. Izadi, A.J. Chamkha, Heat source location and natural [41] O. Mahian, I. Pop, A.Z. Sahin, H.F. Oztop, S. Wongwises, Irreversibility analysis
convection in a C-shaped enclosure saturated by a nanofluid, Phys. Fluids 29 of a vertical annulus using TiO2/water nanofluid with MHD flow effects, Int. J.
(2017) 122009. Heat Mass Transf. 64 (2013) 671–679.
[17] M. Izadi, R. Mohebbi, D. Karimi, M.A. Sheremet, Numerical simulation of [42] F. Selimefendigil, H.F. Öztop, A.J. Chamkha, MHD mixed convection and
natural convection heat transfer inside a\ shaped cavity filled by a MWCNT- entropy generation of nanofluid filled lid driven cavity under the influence of
Fe3O4/water hybrid nanofluids using LBM, Chem. Eng. Process. Process inclined magnetic fields imposed to its upper and lower diagonal, J. Magn.
Intensif. 125 (2018) 56–66. Magn. Mater. 406 (2016) 266–281.
[18] Y. Ma, R. Mohebbi, M.M. Rashidi, Z. Yang, Study of nanofluid forced convection
heat transfer in a bent channel by means of lattice Boltzmann method, Phys.
Fluids 30 (2018) 32001.