Proceedings of CHT-15

ICHMT International Symposium on Advances in Computational Heat Transfer

May 25-29, 2015, Rutgers University, Piscataway, USA
CHT-15-259

NATURAL CONVECTION FLOWS IN A POROUS NANOFLUID-FILLED

TRIANGULAR ENCLOSURE WITH WAVY LEFT WALL
Saurabh Bhardwaj, Amaresh Dalal and Gautam Biswas*
Dept. of Mechanical Engineering
Indian Institute of Technology Guwahati
Guwahati-781039, India
*
Correspondence author: gtm@iitg.ernet.in

ABSTRACT The present study numerically analyses the natural convection heat transfer and fluid
flow behaviour having nano-fluids in a two-dimensional porous right-angled triangular enclosure
having undulation on the left wall. The nano-fluids taken in this study are Cu-water and Al2O3water. Darcy-Forchheimer model is used to simulate the momentum transfer in the porous medium.
The stream function-vorticity equations are solved using finite-difference technique. The
computations are done on a structured non-orthogonal body fitted mesh. The non-uniform heating
of the enclosure has been considered for the study. The results are obtained to analyse the effect of
Rayleigh number ( 103 Ra 105 ), Darcy number (10-4 Da 10-2 ) on the heat transfer and fluid
flow. Solid volume fraction parameter of nanofluids is taken in the range of 0 0.2. The
results are presented in the form of streamlines, isotherms, local Nusselt number and average
Nusselt number. It is found that the effective thermal conductivity of the nanofluid plays a huge role
in increasing the heat transfer rate inside the enclosure. It is also observed that the more heat
transfer can be achieved by using high thermal conductivity nanoparticles in the base fluid with
large solid volume fraction.
NOMENCLATURE
J
K
U,V

Jacobian
Permeability (m2)
Dimensionless velocity components
in , directions
Volumetric expansion coefficient
Wave amplitude
Dimensionless curvilinear coordinates
Dimensionless time in curvilinear system
Solid volume fraction
General variable

Subscripts and Superscripts

avg
c

Average
Cold
1

f
h
nf
p
*

Base fluid
Hot
Nanofluid
Solid particles
Dimensional quantities

Abbreviation
QUICK Quadratic Upstream Interpolation for Convective Kinetics

INTRODUCTION
Several attempts have been made by many researchers to increase the heat transfer rate in natural
convection heat transfer and fluid flow inside an enclosure. Natural convection inside an enclosure
saturated with porous media is well known phenomenon due to its wide application in engineering
field. Many applications can be found out in several books: Nield and Bejan [2006], Vafai [2005]
and Bejan et al. [2004]. In most of the engineering application, natural convection inside porous
enclosure saturated with porous media having undulation can be seen e.g. regenerative heat
exchanger. The other applications of natural convection heat transfer in porous media are also found
in solar power collectors, air saturated fibrous insulation material surrounding a heated pollutant
dispersal in aquifiers, storage of nuclear waste in deep geological repositories and heat loss from
underground energy storage systems. A new technology to improve heat transfer is to use solid
particles in the base fluid, termed as nanofluids. Nanofluid refers to fluids in which nano scale
particles are suspended in the base fluid. Choi [1995] first introduced this concept and suggested
that the presence of nanoparticles in the nanofluid increases its thermal conductivity and therefore
enhances the heat transfer characteristics of the nanofluid.
Abu-Nada and Oztop [2009] have analysed the effects of inclination angle on natural convection
heat transfer and fluid flow in a two dimensional enclosure filled with Cu-nanofluid. Inclination
angle can be a controlling parameter for nano-fluid filled enclosure. The problem of steady natural
convection heat transfer in a differentially-heated enclosure filled with a CuO-EG-Water nanofluid
was investigated using different variable thermal conductivity and variable viscosity models by
Abu-Nada and Chamkha [2010]. Results for a wide range of Rayleigh numbers, volume fractions of
nanoparticles and enclosure aspect ratios were presented. Nasrin et al. [2012] have studied the
influence of Prandtl number and tilted angle of a partially heated inclined rectangular cavity on free
convection flow phenomena with Soret and Dufour coefficients. Water-Al2O3 nanofluid has been
taken in this study. It is found that the structure of the fluid streamlines, isotherms and isoconcentrations within the enclosure is found to significantly depend upon the Prandtl number, Pr
and Streamfunction, and Al2O3 nanoparticles with the highest Pr and the lowest are established
to be most effective in enhancing performance of heat transfer rate than the rate of mass transfer.
Basak and Chamka [2012] have analysed natural convection of nanofluid with uniform or nonuniform heating of bottom wall using Galerkin finite element method. It is observed that flow
strength as well as convective heat flow are dominant for water at the center of the cavity. Steadystate free convection heat transfer behavior of nanofluids is investigated numerically inside a rightangle triangular enclosure filled with a porous medium by Sun and Pop [2011]. The flush mounted
heater with finite size is placed on the left vertical wall. The maximum value of average Nusselt
number is obtained by decreasing the enclosure aspect ratio and lowering the heater position with
the highest value of Rayleigh number. Natural convection is studied in an isosceles triangular
enclosure with a heat source located at its bottom wall and filled with an Ethylene Glycol Copper
nanofluid by Aminossadati and ghasemi [2011]. Natural convection from a protruding heater
located at the bottom of a square cavity filled with a copper-water nanofluid is analysed by Guiet et

al. [2012]. It is observed that the heat transfer is improved by enhancing both the Rayleigh number
and the nanoparticle volume fraction.
The natural convection inside triangular enclosure saturated with porous media has also been
analysed by many researchers. They have considered different types of triangular enclosures for the
study. Varol et al. [2008] have analysed the natural convection in porous non isothermally heated
triangular cavity. Three boundary conditions were used. It was observed that heat transfer enhances
when vertical and inclined walls were isothermal while bottom wall was at non-uniform
temperature. Varol et al. [2006b] have investigated the steady state free convection heat transfer in a
porous media filled right-angled triangular enclosure. Heat transfer increases with the decreasing
aspect ratio. Natural convection in porous triangular enclosure with centered conducting body is
analysed by Varol [2011]. Anandalakshmi et al. [2011] have analysed the heat distribution and
thermal mixing during steady laminar convection flow inside right-angled triangular enclosure
filled with porous media. The heat transfer distribution and thermal mixing enhances due to
isothermal heating of walls. Basak et al. [2007] have analysed the natural convection flow in an
isosceles triangular enclosure with various thermal boundary conditions. It is analysed that at low
Darcy number the heat transfer is primarily due to conduction irrespective of Rayleigh number and
Prandtl number. Natural convection heat transfer has been analysed numerically in a triangular
enclosure with flush mounted heater on vertical wall by Varol et al. [2006a]. It is found that the
most important parameter on heat transfer and flow field is the position of heater. The effect of
undulations on the heat transfer in square cavity with sinusoidal temperature boundary condition
has been analysed by Dalal and Das [2005a, 2007].
In the present problem, a two dimensional right-angled triangular enclosure having undulations on
the left wall is considered to analyse the natural convection heat transfer and fluid flow. The
enclosure is filled with nanofluid and saturated with porous media. The two types of nanoparticles,
i.e., Cu and Al2O3, are considered for the present study and the water has been considered as base
fluid. The primary objective of this work is to investigate the thermal and fluid characteristics of
water based nanofluids within a porous triangular enclosure. The results are shown in the form of
streamlines, isotherms and Nusselt number.
PROBLEM SPECIFICATION
The problem considered in the present study is a two-dimensional right-angled triangular cavity
with fluid saturated porous media with nanofluids having wavy left wall. The left and right inclined
walls are considered to be of isothermally cold wall temperature, T and the bottom horizontal wall
c

is considered to be of spatially varying sinusoidal temperature, T ( x ) .The dimensional form of

the sinusoidal temperature distribution on the bottom wall is given as (e.g., Sarris et al. 2002)
*

Tw ( x ) Tc

T
2

2 x
1

c
o
s

(1)

where T is the temperature difference between the maximum and minimum temperatures of the
bottom walls. The non-dimensional temperature distribution on the bottom wall can be written by
using scale parameters (e.g., De and Dalal 2006). The non-dimensional form of the above equation is
as follows:
Tw ( x)

(1 c o s ( 2 x ) )

The expression for sinusoidal wavy left wall is given by

(2)

(3)

f ( y ) [1 ( c o s 2 n y ) ]

where, n is the number of undulations and is the amplitude (e.g., Adjlout et al. 2002). The amplitude
for all cases in this study has been taken as 0.05. The Rayleigh number is varied from 1 0 to 1 0 ,
Darcy number is varied from 1 0 to 1 0 and solid volume fraction is taken as =0, 0.1and 0.2. The
Prandtl number based on based fluid (water) for all simulations is fixed, Pr = 6.2.
3

NUMERICAL PROCEDURES
Governing Equations The equations governing fluid flow and heat transfer considering laminar,
incompressible and two dimensional flows are written below in stream function-vorticity
formulation. Thermo-physical properties of the fluid in the flow field are assumed to be constant
except the density variations causing a body force term in the invoked vorticity equation.
Boussinesq approximation is invoked in the fluid properties involving the density variation.
Temperature of the solid phase and fluid phase is assumed to be equal everywhere in the porous
medium and local thermal equilibrium is applicable. Darcy-Forchheimer model is used to simulate
the momentum transfer in the porous medium neglecting inertia term. In the present problem, water
has been used as a base fluid and Cu and Al2O3 are considered as nanoparticles. The different nondimensional parameters used to write governing equation are as follows.

ft
L

y
y

x
x

u L

f
*

T
f

Tc

Th Tc

The non-dimensional form of streamfunction-vorticity equations are given below:

(u )
x

( v )
y

nf

2
2
( ) nf

nf P r
T
Pr

RaPr
2
2
y f D a
nf f
x
x

Porous medium
nanofluid

Figure 1. Schematic of physical domain

(4)

xx

T
t

Where,

Pr

Da

K
L

and

(u T )
x

Ra

, v

(vT )

yy

nf

(T h T c ) L

(5)
x

T
T

2
2
y
x
2

(6)

are nondimensional Prandtl number, Darcy

number and Rayleigh number. The effective properties of nanofluids are considered in this study
which is a combination of base fluid properties and the nano-particle properties. The effective
density of the nanofluid ( ) is given as,
nf

n f (1 )

(7)
p

where , and are the solid volume fraction of nanoparticles, density of base fluid and density
of nanoparticles, respectively. The thermal diffusivity of the nanofluid can be defined as:
f

nf

k nf
( C

(8)
) nf

where, k is the effective thermal conductivity which can be calculated by the well-known
Maxwell formula as
nf

k nf k

k p 2 k f 2 ( k f k p )

k p 2 k f ( k f k p )

(9)

Here, kf and kp are the thermal conductivities of the base fluid and dispersed particles, respectively.
The use of this Eq. 9 is restricted to spherical nanoparticles where it does not account for other
shapes of nanoparticles. It should be noted that macroscopic modelling of nanofluids is restricted to
spherical nanoparticles with small temperature gradients (e.g., Ho et al. 2008, Oztop and Abu-Nada
2009). Therefore, the present formulation is limited to nanoparticles with a spherical shape.
The heat capacity of the nanofluid can be written as
( C

) n f (1 ) ( C

( C

(10)

In a similar way, the thermal expansion coefficient of the nanofluid can be determined by
( ) n f (1 ) ( )

( )

(11)

Brinkmann model has been considered to write effective viscosity of the nanofluid which can
defined as
nf

(1 )

2 .5

(12)

Where, is the viscosity of the base fluid. The thermophysical properties of the base fluid and
various solid particles (Copper, Cu and Alumina, Al2O3) are given in Table 1.
f

Boundary Conditions In addition to the governing equations, the temperature boundary conditions
take the below given form. Velocities in two directions are zero at all solid boundaries ( u v 0 )
and the value of stream function has been taken as unity ( 1) at all solid boundaries.
0 x L

On the bottom wall,

,T

(1 c o s ( 2 x ) )

On the Left wall,

0 y H ,T 0

On the inclined wall,

T 0

(13)

The vorticity boundary condition can be calculated by its basic definition,

v
x

(14)

Discretization Procedure For solving the above given governing equations, the physical plane (x, y)
is transformed into the computational plane ( , ) by using boundary-fitted coordinate system. The set
of non-dimensional governing equations in the transformed non-orthogonal boundary-fitted coordinate
system are.

1
J

( (U ) (V ) )

nf

P r

nf

Pr
Da

( (U T ) (V T ) )

1 ( ) nf

nf

nf

R a P r ( y T y T )

(16)

(17)

1
J

( x

The Laplacian of the generic scalar

),

1
J

( y

(18)

in transformed plane is given by

Table 1
Thermophysical Properties of Base Fluid and Nanoparticles
Properties
C

(J kg

k (W m

(kg m

(K

-1

-1

-1

-3

K
K
)

-1

-1

Pure water
4179

Cu
385

Al2O3
765

0.613
997.1
21 x 10-5

401
8933
1.67 x 10-5

40
3970
0.85 x 10-5

(15)

1
J

[(

y A x B )

( y A x B )

(19)

where,
x

x x y y ,

A x x x , B y y y

(20)

U y u x v ,V x v y u , J x y y x

The governing equations are discretised on a non-orthogonal body fitted grid using finite difference
method. The finite difference method has been used for solving the governing equations. Second-order
central difference scheme is used for all the spatial discretization and convective terms are discretised
using QUICK scheme . A semi-implicit time integration scheme is adopted in the present study to
advance the vorticity and temperature values in time. The grid size used for all computations in the
present study is 81 x 81.
Nusselt Number Calculation The heat transfer rate in an enclosure can be obtained from the Nusselt
number calculation. Once the temperature distribution is calculated, the local Nusselt number along
any surface can be determined from its definition
Nu

hL
k

where

k nf

k
f

(21)

is the dimensionless derivative along the direction of the outward drawn normal to that

surface. Since all the calculations have been performed on the computational domain, this derivative

should be expressed in terms of the , independent variables. The local Nusselt numbers in

different walls are expressed as

k nf
N ul
k
f

Bottom Wall,

1
( T T )

(22)

Table 2
Comparison of present results with the results of Basak and Chamkha [2012] for average Nusselt
number on the heated wall, Nuavg

% deviation

0.1

0.2

Ra

(a) Basak and Chamkha [2012]

(b) Present results

a b

*100
a

103

1.381

1.425

-3.12

104

2.471

2.514

-1.735

105

5.304

5.404

-1.87

103

1.762

1.826

-3.64

104

2.621

2.644

-0.935

5.685

5.752

-1.178

10

Left Wall,

k nf
N ul
k
f

1
( T T )

(23)

The average Nusselt number is the average of local Nusselt number along a wall and is defined as
N u avg

1
L

(24)

N uldl

Code Validation For the validation of the present code, two test cases have been performed
considering natural convection inside square enclosure with nanofluid and natural convection inside
porous triangular enclosure. In first test case, a differentially heated square enclosure has been
considered having nanofluid, i.e., Cu-Water. In this study, left wall is uniformly heated and right wall
is kept isothermally cold. The top and bottom walls are considered to be adiabatic. All the
computations are done on 57 x 57 grid size. The average Nusselt on left wall has been calculated for
different values
of Rayleigh number ranging from 103 to 105 and for different values of solid volume fraction, =0,
0.1and 0.2 considering Prandtl number equal to 6.2. The present results are compared with the
results of Basak and Chamkha [2012] and are shown in Table 2.
The second test case models as heat transfer through a two-dimensional right-angled triangular cavity
with flat walls saturated with porous media. In this problem, left wall is taken as isothermally hot and
inclined wall is taken as isothermally cold with horizontal bottom wall as adiabatic. The computations
are carried out for 49 x 49, 53 x 53 and 57 x 57 grid. The average Nusselt on left wall has been
calculated for Rayleigh number, Ra = 5 x 105, Darcy number, Da = 10-3 and Prandtl number, Pr =
1000. The results obtained in this study has been compared with the results of Basak et al. [2012] and
shown in Table 3.

RESULTS AND DISCUSSION

A parametric study has been carried out to determine the effect of Rayleigh number, Darcy number and
solid volume fraction on the heat transfer and flow field. The results are presented for Rayleigh number
103 to 105, Darcy number 10-4 to 10-2, solid volume fraction, = 0, 0.1, 0.2 and the undulation
amplitude of 0.05. For all cases, the Prandtl number is fixed at Pr = 6.2 which is for water at standard
conditions. Isotherms, streamline distribution, local Nusselt number and the average Nusselt number
on the walls are discussed here on the basis of the following results. In this study, Water is taken as the
base fluid and Cu and Al2O3 are taken as nano-particle. The reason for choosing Cu and Al2O3 as
nanoparticle is their high thermal conductivity.
Table 3
Comparison of present results with the results of Basak et al. [2012] for average Nusselt number on
the heated wall, Nuavg
% deviation
Grid size

(a) Basak et al. [2012]

(b) Present results

49 x 49

8.870

8.292

6.51

53 x 53

8.918

8.377

6.06

a b

*100
a

57 x 57

8.965

8.459

5.64

Isotherms and Streamlines Figures 2-3 show the streamlines and isotherms in a porous right-angled
triangular enclosure with sinusoidally heated bottom wall and isothermal cold left and inclined walls. It
can be seen that double circulation cells are formed in different rotating directions in which left
circulation cell rotates in anticlockwise direction and right circulation rotates in clockwise direction.
This outcome is due to enclosure is heated sinusoidally from the bottom having maximum
temperature at the centre of the bottom wall. As a result, the fluids near to the bottom wall have
lower density as compared to the fluid near to the cold walls. Figures 2(a)-(c) show the distribution
of isotherms and streamlines for Cu-water based nanofluid at solid volume fraction, = 0.1 and
Darcy number, Da = 10-2 for different values of Rayleigh number. Here, for comparison point of
view, the distribution for pure water are also shown with dotted lines. The Darcy number is kept
constant for this case to show the effect of Rayleigh number on the isotherms and streamlines. It is
observed that the streamlines and isotherms for all values of Ra are substantially affected by
nanofluid. It can be seen that for low values of Ra (i.e., 103, 104), isotherms are almost smooth and
the shape does not change much for the different Rayleigh number but their intensity increases as
shown in Figs. 2(a)-(b). This is due to the dominance of viscous force in conduction dominant
regimes. As Ra increases to 105, the isotherms are unable to maintain its smoothness and have
feather like distribution. It is also observed that at Ra = 105, the isotherms start distributing
throughout the enclosure.
The distribution of streamlines show that the fluid rises from the middle portion of the heated
bottom wall and flows down along the left and inclined walls due to isothermal cold temperature of
the walls. Here, two counter rotating circulations are observed in which left anticlockwise rotating
circulation dominates over right clockwise rotating circulation and covers the major part of the
enclosure. As Ra increases keeping Da and constant, the strength of the circulation also increases
which can be shown in Figs. 2(a)-(c). The strength of the vortex for Rayleigh number, 103, 104 and
105 is found to be equal to 1.0378, 1.4221 and 7.2228 respectively. This is due to the fact that as Ra
increases, the heat added to the system increases which in turn increases the buoyancy forces in the
cavity. This increase in the buoyancy forces strengthen the circulation and hence increase the
strength of the vortex. The comparison of streamlines of Cu-water nanofluid (shown with the solid
lines) with the pure water (shown with the dotted lines) show that the flow intensity of pure water
circulation is more than the nanofluid due to the fact that the diameter of specific streamline cell for
water is larger than that for nanofluid at a constant value of streamfunction. It is also observed that
the diameter of flow circulation cell for pure water based fluid is bigger than that for nanofluid
corresponding to a specific streamfunction. It may be noted that at low Rayleigh numbers, fluid
flow is due to viscous effect and addition of nanoparticles makes the fluid more viscous resulting in
less intense flow for nanoparticles.
The effect of Darcy number on the isotherms and streamlines are shown in Figs. 3(a)-(c) at
constant Rayleigh number, Ra = 105 and for solid volume fraction, = 0.1 for Cu-water nanofluid.
It may be noted that Darcy number shows the ease of flowability to the fluid flow. For low values of
Darcy number, hydraulic resistance to fluid flow is more and for high values of Darcy number,
hydraulic resistance decreases. At low values of Darcy number, Da = 10-4, 10-3, the fluid flow intensity
is weak enough due to the high hydraulic resistance which can be seen by the strength of the vortex in
Fig. 3(a)-(b). The strength of the vortex for Da = 10-4, 10-3 is found to be 1.1169 and 2.1457
respectively. In this case the isotherms are also smooth and distributed only in a small part of the
enclosure. As Darcy number increases to 10-2, the fluid flow intensifies and the strength of the vortex
increases to 7.2228 as shown in Fig. 3(c). This is due to the low hydraulic resistance at large Rayleigh
number which creates the buoyancy forces to take effect completely on the fluid flow.

(a)

(b)

(c)
Figure 2. Isotherms (left) and Streamlines (right) at Da = 10-2 and solid volume fraction =0.1 for
different values of Rayleigh number (a) Ra = 103, (b) Ra = 104 and (c) Ra = 105,where solid

10

lines(-) show contours for Cu-water nanofluid and dotted lines (..) show contours for pure
water

(a)

(b)

(c)
Figure 3. Isotherms (left) and Streamlines (right) at Ra = 105 and solid volume fraction =0.1 for
different values of Darcy number (a) Da = 10-4, (b) Da = 10-3 and (c) Da = 10-2,where

11

solid lines(-) show contours for Cu-water nanofluid and dotted lines (..) show contours for
pure water
Nusselt Number The parameter which predicts the heat transfer rate in the natural convection analysis
is Nusselt number. The variation of local Nusselt number along the distance of the bottom and left
walls has been shown in Figs. 4-5. In this case, the Rayleigh number is kept constant, i.e., Ra = 105 and
three different variations are shown for different solid volume fractions, i.e., =0, 0.1and 0.2. Figure
4(a) shows the variation of local Nusselt number on heated bottom wall at Darcy number, Da = 10-4. It
can be observed that as fraction of nanofluid increases, the maximum Nusselt number also increases.
But only single peak is observed, this is due to the fact that the Darcy number is very low which creates
the large hydraulic resistance. Although, the Rayleigh number is high enough but due to low Darcy

(a)

(b)

Figure 4. Variation of local Nusselt number on bottom wall at Rayleigh number, Ra = 105 and Darcy
number, (a) Da = 10-4, (b) Da = 10-2

(a)

(b)

Figure 5. Variation of local Nusselt number on left wavy wall at Rayleigh number, Ra = 105 and Darcy

12

number, (a) Da = 10-4, (b) Da = 10-2

number, the buoyant forces are not able to evolve much and conduction heat transfer takes place in the
cavity. As Darcy number increases to 10-2, two peaks observed for all values of solid volume fraction
which is due to sinusoidal heating of the bottom wall as shown in Fig. 4(b). The maximum value of
Nusselt number is found in the first half of the wall because of the suppression of the fluid due to
inclined wall and also the undulations present on the left wall which creates squeezed isotherms. Figure
5 shows the variation of local Nusselt number on the left wavy wall. Here, minus (-) sign with Nusselt
number values indicates the heat loss from the fluid to the wall. The higher the Rayleigh number and
Darcy number, the larger the amount of heat rejected from the fluid to the left wall. This pattern of
Nusselt number is due to the undulations present on the left wall. The examination of isotherms and
local Nusselt number distribution on the left wall reveals that the thermal boundary layer thickness on
the side of the undulated wall increases and decreases just before a crest or just after a trough. These
undulations introduce boundary layer interruptions and therefore improve the heat transfer. For low
Darcy number, Da = 10-4, the maximum value of Nusselt number, |Nul| is found for the first undulation
at the bottom side of the left wall due to weak isotherms and fluid flow as heat transfer occurs due to
conduction as shown in Fig. 5(a). But as Darcy number increases to 10-2, the maximum value of
Nusselt number, |Nul| is observed for second undulation on the left wall due to intensified fluid flow
because of the dominance of buoyancy forces. It is also found that the maximum values of Nusselt
number increases as solid volume fraction, increases. This is due to the fact that addition of the
nanoparticles influences the heat transfer properties of the fluid considerably. Since the thermal
conductivity of the base fluid (water) is low, addition of nanoparticles improves the thermal
conductivity of the resulting nanofluid. In fact, several researchers have reported that even with very
low nanoparticles concentration, the thermal conductivity of the resulting fluid can increase largely.
Figure 6 shows the average Nusselt number against the logarithmic Darcy number at Rayleigh number
equal to 105 along bottom and left wavy wall for different solid volume fraction. It can be observed that
as the Darcy number increases, the value of average Nusselt number also increases and also, average
Nusselt number further increases with the increase of solid volume fraction on both the walls as shown
in Figs. 6(a)-(b). It is known that at low Darcy number, the hydraulic resistance is more and the
addition of fraction of nanofluid makes the fluid more viscous. This is why the values of average
Nusselt number are lower than the values at higher Darcy number (i.e., Da = 10-2) where hydraulic

(a)

(b)

Figure 6. Variation of average Nusselt number at Rayleigh number, Ra = 105 on, (a) bottom wall,

13

(b) left wavy wall

resistance is very low and fluid easily flows throughout the cavity. But addition of nanofluid increases
the effective thermal conductivity of the fluid. Since the Nusselt number is the product of ratio of
effective thermal conductivity of nanofluid to thermal conductivity of base fluid, and nondimensional
temperature gradient as shown in Eq. 21. Hence, the Nusselt number for nanofluid is always more than
the Nusselt number for base fluid which can be seen in Figs. 6(a)-(b). The effect of fraction of
nanofluid, i.e., solid volume fraction, and Darcy number can also be seen in Fig. 7(a)-(c). Here the
average Nusselt number on left wavy wall has been plotted against logarithmic Rayleigh number. In
this case also, it can be seen that the magnitude of average Nusselt number on left wavy wall increases
with the increase in the Rayleigh number, Darcy number and also in the solid volume fractions. The
percentage increase in the magnitude of average Nusselt number on the left wall at Ra = 105 and Da =
10-2 for solid volume fraction, = 0.1 and = 0.2 from the base fluid is found to be 17.78% and
48.22% respectively.
To assess the effect of undulation on the heat transfer, a comparison of maximum Nusselt number
between wavy left wall and straight left wall (i.e., no-undulation) has been made. For this case, Ra

(a)

(b)

(c)
Figure 7. Variation of average Nusselt number on left wavy wall for solid volume fraction, (a) = 0,

14

(b) = 0.1, (c) = 0.2

is kept constant and maximum Nusselt has been plotted against Darcy number for different as
shown in Figs. 8(a)-(b). It is found that for low values of Da, the maximum Nusselt number is almost
equal for = 0.1 but as Darcy number increases to a higher value, the maximum Nusselt number for
wavy wall case is much greater than straight left wall. This increase in maximum Nusselt number
further exaggerated as increases to 0.2. The percentage increase in the magnitude of the maximum
Nusselt number on wavy wall compared to straight wall is found to be equal to 55.83% and 56.10% for
-2
= 0.1 and = 0.2 respectively at Da = 10 . Another analysis which has been done in the present
study is to quantify the effect of nature of nanoparticle on the heat transfer in a porous enclosure. For
this analysis, average Nusselt number on the heated bottom wall has been calculated for two different
nanoparticles, e.g., Cu and Al2O3 at Ra = 105 and Da = 10-2. Since the thermal conductivity of Cu is
much larger than Al2O3, the difference in average Nusselt number is found for heated bottom wall.
This difference in average Nusselt number is almost negligible for = 0 case due to base fluid (i.e.,
water) properties because there is no nanoparticle involved. But as the solid volume fraction
increases, the difference in average Nusselt number for Cu and Al2O3 also increases because Al2O3
has much lower thermal conductivity than Cu which ultimately lower down the thermal conductivity of
nanofluid.
Conclusions A parametric study has been performed to analyse the thermal and fluid behaviour with
nanofluid in a two dimensional right-angled triangular enclosure having undulations on the left wall
and saturated with porous media. For this analysis, streamfunction-vorticity equations are used
considering nanofluid properties. A thorough investigation has been done to validate the present inhouse code and the results are compared with the available published literature. The results for all cases
are in very good agreement with the compared solutions. In the present problem, the Rayleigh number
have been varied in the range of 103Ra 105, Darcy number have been taken in the range of 10-4 Da
10-2 and solid volume fraction, 0 0.2. Two types of nanoparticles have been used for this study,
i.e., Copper (Cu) and Alumina (Al2O3). The following conclusions can be made based on present
analysis.
(a) The heat transfer has been found to be an increasing function of Ra and Da. For low values of
Rayleigh number and Darcy number, the heat transfer is dominated by conduction and after a critical
value Ra, convective heat transfer dominates inside the enclosure.

(a)

(b)

Figure 8. Comparison of maximum Nusselt number on left wavy wall with no-undulation on the left

15

wall case at Ra = 105 for solid volume fraction, (a) = 0.1, (b) = 0.2
(b) The intensity of fluid circulation is found to be more for base fluid compared to nanofluid case
because the addition of nanoparticles in base fluid makes the fluid more viscous resulting in less
intense flow for nanoparticles.
(c) The effective thermal conductivity of nanofluid plays an important role in increasing the heat
transfer rate inside the enclosure. As the solid volume fraction increases, the average Nusselt
number on heated bottom wall as well as left wavy wall increases.
(d) The presence of undulations on the left wall increases the velocity of the fluid inside the
enclosure and hence creates more deviation in the isotherms. An increase of 56.10% has been found
in the maximum Nusselt number on left wavy wall compared to straight left wall case using Cuwater nanofluid for Ra = 105, Da = 10-2 and = 0.2.
(e) It is observed that Cu-water nanofluid gives more heat transfer rate compared to Al2O3-water
nanofluid due to its large effective thermal conductivity.

Figure 9. Average Nusselt number for different nanoparticles and for various solid volume fractions

REFERENCES
Abu-Nada, E. and Oztop, H.F. [2009], Effects of inclination angle on natural convection in
enclosures filled with cu water nanofluids, Int. J. Heat Fluid Flow, Vol. 30, pp. 669678.
Abu-Nada, E. and chamkha, A.J. [2010], Effect of nanofluid variable properties on natural
convection in enclosures filled with a cuo-EG-water nanofluids, Int. J. Therm. Sci., Vol. 49, pp.
23392352.
Adjlout, L., Imine, O., Azzi, A. and Belkadi, M. [2002], Laminar natural convection in an inclined
cavity with a wavy wall, Int. J. Heat Mass Trans., Vol. 45, pp. 21412152.

16

Aminossadati, S. and Ghasemi, B. [ 2011], Enhanced natural convection in an isosceles triangular

enclosure filled with a nanofluids, Comput Math Appl, Vol. 61, pp. 17391753.
Anandalakshmi, R., Kaluri, R. and Basak, T. [2011], Heatline based thermal management for natural
convection within right-angled porous triangular enclosures with various thermal conditions of walls,
Energy, Vol. 36, pp. 48794896.
Basak, T., Roy, S. and Thirumalesha, C. [2007], Finite element analysis of natural convection in a
triangular enclosure:Effects of various thermal boundary conditions, Chem. Eng. Sci., Vol. 62, pp.
26232640.
Basak, T., and Chamkha, A.J. [2012], Heatline analysis on natural convection for nanofluids
confined within square cavities with various thermal boundary conditions, Int. J. Heat Mass Trans.,
Vol. 55, pp. 5526 5543.
Basak T., Gunda P. and Anandalakshmi R. [2012], Analysis of entropy generation during natural
convection in porous right-angled triangular cavities with various thermal boundary conditions, Int. J.
Heat Mass Trans., Vol. 55, pp. 45214535.
Bejan, A., Dincer, I., Lorente, S., Miguel, A. F. and Reis, A.H. [2004], Porous and Complex Flow
Structures in Modern Technologies., Springer, New York.
Choi, S.U.S. [1995], Enhancing thermal conductivity of fluids with nanoparticles, in: Proc. 1995
ASME Int. Mech. Engng. Congress and Exposition, San Franciscos, USA., 99-105.
Dalal, A. and Das, M. K. [2005], Laminar natural convection in an inclined complicated cavity with
spatially variable wall temperature, Int. J. Heat Mass Trans., Vol. 48, pp. 38333854.
Dalal, A. and Das, M. K. [2007], Numerical study of laminar natural convection in a complicated
cavity heated from top with sinusoidal temperature and cooled from other sides, Computers Fluids,
Vol. 36, pp. 680700.
De, A.K. and Dalal, A. [2006], A numerical study of natural convection around a square, horizontal,
heated cylinder placed in an enclosure, Int. J. Heat Mass Trans., Vol. 49, pp. 46084623.
Guiet, J., Reggio, M., and Vasseur, P. [2012], Natural convection of nanofluids in a square
enclosure with a protruding heater, Adv. Mech. Eng., Vol. 2012, pp. 167296(1-11)
Ho, C. J., Chen, M.W. and Li, Z. W. [2008], Numerical simulation of natural convection of
nanofluid in a square enclosure: effects due to uncertainties of viscosity and thermal conductivity,
Int. J. Heat Mass Transfer, Vol. 51, pp. 45064516.
Nasrin, R., Parvin, S. and Alim, M.A. [2012], Free convective phenomena in a tilted enclosure
filled with water-al2o3 nanofluid, Int. J. Eng. Sci. Tech., Vol. 4, pp. 114.
Nield, D., and Bejan, A. [2006], Convection in Porous Media, third ed. Springer, New York.
Oztop, H., F., Mobedi, M. and Pop, I. [2012], A heatline analysis of natural convection in a square
inclined enclosure filled with a cuo nanofluid under non-uniform wall heating condition, Int. J.
Heat Mass Trans., Vol. 55, pp. 50765086.

17

Sarris, I. E., Lekakis, I., Vlachos, N. S. [2002], Natural Convection in a 2d Enclosure with Sinusoidal
Upper Wall Temperature, Numer. Heat Trans. PART A, Vol. 42, pp. 513530
Sun, Q., and Pop, I. [2011], Free convection in a triangle cavity filled with a porous medium
saturated with nanofluids with flush mounted heater on the wall, Int. J. Therm. Sci., Vol. 50, pp.
21412153.
Varol, Y., Koca, A. and Oztop, H. F. [2006a], Natural convection in a triangle enclosure with flush
mounted heater on the wall, Int. J. Heat Mass Trans., Vol. 33, pp. 951958
Varol, Y., Oztop, H. F. and Varol, A. [2006b], Free convection in porous media filled right-angle
triangular enclosures, Int. J. Heat Mass Trans., Vol. 33, pp. 11901197
Varol, Y., Oztop, H., Mobedi, M. and Pop, I. [2008], Visualization of natural convection heat transport
using heatline method in porous non-isothermally heated triangular cavity, Int. J. Heat Mass Trans.,
Vol. 51, pp. 50405051
Varol, Y. [2011], Natural convection in porous triangular enclosure with a centered conducting body,
Int. Commun Heat Mass Trans., Vol. 38, pp. 368376
Vafai, K. [2005], Handbook of Porous Media, Taylor & Francis, New York.
Yu, Z., Xu, X., Hu, Y. and Cen, K. [2011], Numerical study of transient buoyancy driven
convective heat transfer of water-based nanofluids in a bottom-heated isosceles triangular
enclosure, Int. J. Heat Mass Trans., Vol. 54, pp. 526532.

18