8/12/2019 MS Defence Presentation Elham
1/51
Eastern Mediterranean UniversityMechanical Engineering Department
Jan 19, 2010
8/12/2019 MS Defence Presentation Elham
2/51
IntroductionIn most natural convection studies, the base fluid in the
enclosure has a low thermal conductivity, which limits the
heat transfer enhancement.An innovative technique, which uses a mixture of
nanoparticles and the base fluid, was first introduced byChoi.
The resulting mixture of the base fluid and nanoparticleshaving unique physical and chemical properties isreferred to as a nanofluid.
The presence of the nanoparticles in the nanofluidincreases the thermal conductivity and thereforesubstantially enhances the heat transfer characteristics ofthe nanofluid.
8/12/2019 MS Defence Presentation Elham
3/51
Solids have Thermal Conductivities that are orderof magnitude larger than those of conventional
heat transfer fluids.
8/12/2019 MS Defence Presentation Elham
4/51
Increased thermal conductivity will result in higher heattransfer than that of the base (pure) fluid withoutdispersed nanoparticles. Measurements of the heattransfer coefficients of fluids have show that the heattrasfer capability of water increased by 15% with adispersion of less than 1 vol% copper oxide nanoparticles
Natural convection heat transfer is affected by nanofluid
properties such as viscosity and thermal conductivity.These parameters are affected by both liquid and solidphases in nanofluid, also volume fraction of nanoparticlesaffect these thermophysical properties. Some ideal
assumptions about properties of nanofluid were made insimulation of natural convection of nanofluid by someresearchers. Because of these assumptions there is aparadox in results of numerical and experimental studies.
8/12/2019 MS Defence Presentation Elham
5/51
Types of Nanofluids
The base fluids which are used in nanofluids are common
heat transfer fluids such as water, engine oil, Ethylene
glycol and ethanol.
Some nanoparticle materials that have been used in
nanofluids are oxide ceramics (Al2O3, CuO, Cu2O), nitride
ceramics (AIN, SiN), carbide ceramics (Sic, TiC), metals(Ag, Au, Cu, Fe), semiconductors (TiO2), single, double or
multi-walled carbon (SWCNT, DWCNT, NWCNT), and
composite materials such as nanoparticle core-polymer
shell composites.
8/12/2019 MS Defence Presentation Elham
6/51
Thermopysical Properties of Nanofluids
Some important thermophysical properties ofnanofluids are density, viscosity, thermal conductivity
and thermal diffusivity. These properties cause that a
nanofluid act very different from its base fluid in
cooling applications.
8/12/2019 MS Defence Presentation Elham
7/51
Density of nanofluids (nf)
The effective density of a fluid containing
suspended particles at a reference temperatureis given by
sfnf )1(
8/12/2019 MS Defence Presentation Elham
8/51
Viscosity of nanofluids(eff )
It is believed that viscosity is as critical as thermalconductivity in engineering systems that employ fluid
flow. Pumping power is proportional to the pressure
drop, which in turn is related to fluid viscosity.
In laminar flow, the pressure drop is directly
proportional to the viscosity. All reported results show
that the viscosity of nanofluids is increased
anomalously and cannot be predicted by classicalmodels .
8/12/2019 MS Defence Presentation Elham
9/51
In the most experimental studies of nanofluid,Brinkman formula were used. The effective viscosityof a fluid of viscosity f containing a dilute
suspension of small rigid spherical particles is givenby Brinkman as:
)1( 5.2
f
eff
8/12/2019 MS Defence Presentation Elham
10/51
ThermalconductivityofnanofluidsIn the most experimental studies of nanofluid, when
the samples are dilute (< 5%) and the shape ofnanoparticles are spherical , the effective thermal
conductivity of the solidliquid mixture is calculated
using Maxwell equitation which depends on thethermal conductivities of both phases and volume
fraction of solid as:
keff
)(2
)(22
kkkk
kkkk
k
k
sffs
sffs
f
eff
8/12/2019 MS Defence Presentation Elham
11/51
Thermal diffusivity of nanofluidsnf
very little work has been performed on the effectivethermal diffusivity of nanofluids, which is especially
important in convective heat transfer applications.
Generally nf
can calculated as follow:
nf=(keff)stagnant/(cp)nf
8/12/2019 MS Defence Presentation Elham
12/51
Applications of Nanofluids(1) Improved heat transfer and stability: Because
heat transfer takes place at the surface of the
particles, it is desirable to use particles with largersurface area.
(2) Microchannel cooling without clogging: The
combination of microchannels and nanofluids willprovide both highly conducting fluids and a largeheat transfer area.
(3) Miniaturized systems: Miniaturized systems will
reduce the inventory of heat transfer fluid and willresult in cost savings.
(4) Reduction in pumping power
8/12/2019 MS Defence Presentation Elham
13/51
Engineering Applications of Nanofluids
(a) Nanofluids in transportation: Nanofluids would
allow for smaller, lighter engines, pumps, radiators,and other components.
(b) In micromechanics and instrumentation:Microelectromechanical systems (MEMS) generate alot of heat during operation. Since nanofluids canflow in microchannels without clogging, they wouldbe suitable coolants.
(c) In heating, ventilating and air-conditioning(HVAC) systems: Nanofluids could improve heattransfer capabilities of current industrial HVAC andrefrigeration systems.
8/12/2019 MS Defence Presentation Elham
14/51
Natural Convection of Nanofluids
There are two general categories for study the
natural convection of fluid: Experimental studies,Numerical studies.
8/12/2019 MS Defence Presentation Elham
15/51
ExperimentalStudiesIn natural convection processes, the thermal and the
hydrodynamic are coupled and both are stronglyinfluenced by the fluid thermophysical characteristics, thetemperature differences and the system geometry.
Differentially heated enclosures are extensively used tosimulate natural convection heat transfer within systems
using nanofluids. Experimental results have beenreported in the literature that dispersion of nanoparticlesin the base fluid may result in considerable decrease inheat transfer. Putra et al and Wen and Ding found a
systematic and definite decline in the heat transfer for aparticular range of Rayleigh numbers and density andconcentration of nanoparticles. Similar results were alsoobtained by Santra et al. who modeled the nanofluid as a
non-Newtonian fluid.
8/12/2019 MS Defence Presentation Elham
16/51
In figure the experimentalresults of Wen and
Ding can be seen. The
Nusselt number decreases
with the Rayleigh number
during the transient heating
period. Similar phenomena
were also observed by Tsoet al. and Bhowmik and
Tou for cooling of electronic
chips through natural
convection in a verticalrectangular channel.
8/12/2019 MS Defence Presentation Elham
17/51
NumericalStudiesResearch studies consider a two-dimensional
enclosure of height H and width L filled with a
nanofluid as shown in Figure.
8/12/2019 MS Defence Presentation Elham
18/51
According to Khanafer et al.numerical results theaverage Nusselt number along the hot vertical wall is
correlated in terms of the Grashof number and the
particles volume fraction.The correlation of Khanaferet al. can be expressed as :
For 103Gr 105 and 0 0.25
GrNu 3123.00809.1 )4436.0(5163.0
8/12/2019 MS Defence Presentation Elham
19/51
Santra et al. (2007) presented the numerical resultsof the simulation of natural convection in a
differentially heated square cavity using copper
water nanofluid for 10
4
Ra 10
7
and 0.05 5according to figure as:
8/12/2019 MS Defence Presentation Elham
20/51
Ho et al. (2008) also did a simulation of naturalconvection in the enclosure filled with alumina
water Nanofluid and found another correlation as:
For 104Ra 106 and 0 0.04
And they found parameters C, m and n in different
four models of simulation according to Table as:
RaCNu nm)1(
8/12/2019 MS Defence Presentation Elham
21/51
Values of coefficient C and exponents m,n for different models according to
numerical results of Ho et al.
8/12/2019 MS Defence Presentation Elham
22/51
DESIGNINGTHEEXPERIMENTThe experiments have been done in an enclosure
having the dimensions of 11 cm (length) and 11 cm(height) and 8 cm (width) which has the aspect ratio
(H/L) of 1. The cavity was full of nanofluid.
The front, back, bottom and top sides of the
enclosure are made of Plexiglas which has 1 cm
thickness. Left and right sides of the enclosureconsist of brass plates (Heat Exchanger) in which
water passages are engraved.
8/12/2019 MS Defence Presentation Elham
23/51
A top view of cavity: [A] Plexiglas Plates, [B] Heat exchangers(right
side is hot wall and left side is cold wall), [C] Insulation Material, [E]
Output Pipes,[D]Input Pipes, [T10 to T13] Pipe Thermocouples
8/12/2019 MS Defence Presentation Elham
24/51
A photo of cavity
8/12/2019 MS Defence Presentation Elham
25/51
Brass plates of heat exchangers
8/12/2019 MS Defence Presentation Elham
26/51
Picture(a)and(b)showtubesandpicture(c)showscavityafterinsulation.
8/12/2019 MS Defence Presentation Elham
27/51
Schematic diagram of the experimentalapparatus
8/12/2019 MS Defence Presentation Elham
28/51
Nanofluid Used in Experiment
The candidate for this project is a dispersion of Cu2O
nanospheres in ethanol as base fluid.
8/12/2019 MS Defence Presentation Elham
29/51
Making Nanofluid Ready to Use
Pure ethanol is added to 10, 15 and 25 ml of main
nanofluid till the total volume of solution reached 968
ml (volume of cavity: 11118) after that we have 3
different solution of main nanofluid S1, S2and S3.
8/12/2019 MS Defence Presentation Elham
30/51
Experimental Measurement Method
The unsteady experiments were performed for 3
different samples of nanofluid and for each of them
in 5 different (T) bath = 10C, 15 C, 20 C, 25 C
and 30 C.
Temperature data were collected ten seconds by ten
seconds by data acquisition system in one Microsoft
Excel file during each test.
8/12/2019 MS Defence Presentation Elham
31/51
RESULTS AND DISCUSSIONS
First of all we need some thermo physical properties
of both solid and fluid at the base temperature.
8/12/2019 MS Defence Presentation Elham
32/51
Step 1: Calculate Nu number
= (Cp)w[T(13)- T(10)]
TH= Temperature on the hot wall: T (6)TC=Temperature on the cold wall: T (5)
L = Length of cavity
H =Height of cavity W =Width of cavity
=Mass flow rate of hot water
)()( kTT
HW
LQNu
effstagnant
CH
m
m
Q
8/12/2019 MS Defence Presentation Elham
33/51
Step 2:
A) calculate volume fraction of samples (%)
OCu
sample
lesnanopartic
mv
2
v
v
cavity
lesnanopartic
sample (%)
OCu
sample
lesnanopartic
mv
2
8/12/2019 MS Defence Presentation Elham
34/51
B) Calculate the effective stagnant thermal conductivity ofSamples: As our samples are dilute ( < 5%) and theshape of nanoparticles are spherical , the effectivestagnant thermal conductivity of the solidliquid mixture
is calculated using Maxwell equitation which depends onthe thermal conductivities of both phases and volumefraction of solid as:
)(2
)(22)(
kkkk
kkkk
k
k
sffs
sffs
f
eff stagnant
8/12/2019 MS Defence Presentation Elham
35/51
Step 3:
According to Raleigh number formula:
We have to calculate some parameters such as nf,
nfand nf.
nfnf
CHnfnf HTTgRa
3
)(
8/12/2019 MS Defence Presentation Elham
36/51
Transienttemperatureandheattransfercoefficient
At t=0 the temperature at the right side of the cavity
is suddenly raised by applying a temperature
difference (T) bath.
8/12/2019 MS Defence Presentation Elham
37/51
TransientNusseltnumberasafunctionoftimeforthethreesamplesat( T)bath=10 C.
8/12/2019 MS Defence Presentation Elham
38/51
TransientRaleighnumberasafunctionoftimeforthethreesamplesat( T) bath=10 C.
8/12/2019 MS Defence Presentation Elham
39/51
TransientNusselt number
versus
Rayleigh
number
for
thesampleS3at( T)bath=10 C.
8/12/2019 MS Defence Presentation Elham
40/51
Using results from
Figures, we are able tofind constants c and n in
correlation Nu= c Rafor
3 samples atT = 10C
8/12/2019 MS Defence Presentation Elham
41/51
ConstantscandnincorrelationNu=cRa
ncsample
-1.548E+15S1
-1.978E+19S2
-1.848E+18S3
8/12/2019 MS Defence Presentation Elham
42/51
Temperature Distribution in Cavity atVarious Times
Numerical studies were performed by krane and Jesse (1983),
Khanafer et al. (2003) and Abu-Nada and Oztop (2009) to analyzeTemperature distribution of nanofluid in cavity for various mesh sizes:
31x31, 41x41, 61x61 and 81x81.
8/12/2019 MS Defence Presentation Elham
43/51
Temperature distribution in cavity at t =3000 s for
the three Samples at (T) bath= 10C.
0
0.2
0.4
0.6
0.8
1
1.2
0 0.2 0.4 0.6 0.8 1 1.2
X
S1
S2
S3
8/12/2019 MS Defence Presentation Elham
44/51
Temperature distribution in cavity at t = 3000 s for thethree Samples at (T) bath= 15C.
0
0.2
0.4
0.6
0.8
1
1.2
0 0.2 0.4 0.6 0.8 1 1.2
X
S1
S2
S3
8/12/2019 MS Defence Presentation Elham
45/51
Temperature distribution in cavity at t = 3000 s for
the three Samples at(T) bath= 20C.
0
0.2
0.4
0.6
0.8
1
1.2
0 0.2 0.4 0.6 0.8 1 1.2
X
S1
S2
S3
8/12/2019 MS Defence Presentation Elham
46/51
Temperature distribution in cavity at t = 3000s for thethree Samples at (T) bath= 25C.
0
0.2
0.4
0.6
0.8
1
1.2
0 0.2 0.4 0.6 0.8 1 1.2
X
S1
S2S3
8/12/2019 MS Defence Presentation Elham
47/51
Temperature distribution in cavity at t = 3000 s for thethree Samples at (T) bath= 30C.
0
0.2
0.4
0.6
0.8
1
1.2
0 0.2 0.4 0.6 0.8 1 1.2
X
S1
S2
S3
8/12/2019 MS Defence Presentation Elham
48/51
CONCLUSION In the case of temperature distribution of nanofluid in
cavity, both numerical and experimental analyzing areagreeing each other.
The present results and also other experimental resultsare opposing numerical results in the case of thetransient Ra and Nu numbers. Present experimentalresults show that using nanofluid in cavity cause
decreasing heat transfer coefficients because Nu numberdecreased during the time.
Nu = hL/k
According to the all expermental results and also presentresults the parameter n in the correlation Nu= c Raisnegative.
8/12/2019 MS Defence Presentation Elham
49/51
The numerical results are not accurate to find acorrelation like Nu= c Rabecause the simulations
were based on some ideal assumptions of (a)
nanofluid was Newtonian, incompressible and theflow was in the laminar regime; (b) nanoparticles
were uniform in shape and size; (c) there was no slip
between liquid and particle phases in terms of bothvelocity and temperature; and (d) nanofluids had
constant thermophysical properties except for
density variation that gave rise to the buoyancy.
8/12/2019 MS Defence Presentation Elham
50/51
The assumption of (c) and (d) are very difficult to be
satisfied for real nanofluids. Although nanofluids
behave more like pure fluids than suspensions of
large particles and also some thermophysicalproperties like viscosity , thermal conductivity and
density of nanofluid cannot be constant in various
times.
8/12/2019 MS Defence Presentation Elham
51/51