International Communications in Heat and Mass Transfer 37 (2010) 1078–1086

Contents lists available at ScienceDirect

International Communications in Heat and Mass Transfer

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i c h m t

The effect of geometrical parameters on heat transfer characteristics of

microchannels heat sink with different shapes☆
P. Gunnasegaran a, H.A. Mohammed a,⁎, N.H. Shuaib a, R. Saidur b
a
Department of Mechanical Engineering, College of Engineering, Universiti Tenaga Nasional, Km7, Jalan Kajang-Puchong, 43009 Kajang, Selangor, Malaysia
b
Department of Mechanical Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

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

Available online 6 July 2010 The effect of geometrical parameters on water flow and heat transfer characteristics in microchannels is
numerically investigated for Reynolds number range of 100–1000. The three-dimensional steady, laminar
Keywords: flow and heat transfer governing equations are solved using finite volume method. The computational
Numerical simulation domain is taken as the entire heat sink including the inlet/outlet ports, wall plenums, and microchannels.
Microchannel heat sink Three different shapes of microchannel heat sinks are investigated in this study which are rectangular,
Rectangular trapezoidal, and triangular. The water flow field and heat transfer phenomena inside each shape of heated
Trapezoidal
microchannels are examined with three different geometrical dimensions. Using the averaged fluid
Triangular
temperature and heat transfer coefficient in each shape of the heat sink to quantify the fluid flow and
temperature distributions, it is found that better uniformities in heat transfer coefficient and temperature
can be obtained in heat sinks having the smallest hydraulic diameter. It is also inferred that the heat sink
having the smallest hydraulic diameter has better performance in terms of pressure drop and friction factor
among other heat sinks studied.
© 2010 Elsevier Ltd. All rights reserved.

1. Introduction characteristics of dry nitrogen gas and water in microtubes with

The amount of literature of heat transfer effects on fluid flow in
MEMS or in microchannels is vast and growing. Most of the previous
works on heat transfer characteristics of fluid flows are using air as a
working fluid to cool the chips. However, when dealing with
components that contain billions of transistors working at high
frequency, the temperature can reach a critical level where standard
cooling methods are not sufficient.
In 1981, Tuckerman and Pease [1] pointed out that decreasing
liquid cooling channel dimensions to the micron scale will lead to
increase the heat transfer rates. They have also experimentally
demonstrated that a forty-fold improvement in heat-sinking capabil-
ity in Si-based microchannels anodically bonded to Pyrex cover plates.
Qu et al. [2,3] conducted experiments to investigate flow
characteristics of water through trapezoidal silicon microchannels
with a hydraulic diameter ranging from 51 μm to 169 μm. Their results
indicate that the pressure gradient and flow friction in microchannels
are higher than those given by the conventional laminar flow theory
due to the effect of surface roughness of the microchannels. So, they
proposed a roughness–viscosity model to interpret the experimental
data. Yu et al. [4] studied the fluid flow and heat transfer
☆ Communicated by W.J. Minkowycz.
⁎ Corresponding author.
E-mail address: Hussein@uniten.edu.my (H.A. Mohammed).
diameter of 19, 52 and 102 μm. Pfahler [5] investigated experimen-
tally the apparent viscosity of isopropanol alcohol and silicon oil in
microchannels.
Steinke and Kandlikar [6] presented a comprehensive review of
friction factor data in microchannels with liquid flows. They indicated
that entrance and exit losses need to be accounted for while
presenting overall friction factors losses in microchannels. Most of
the data that accounted for friction factor loss show good agreement
with the conventional theory. They also provided a new procedure for
correcting measured pressure drop to account for inlet and outlet exit
losses. Furthermore, three-dimensional fluid flow and heat transfer
phenomena inside heated microchannels were investigated by Toh et
al. [7]. They solved the steady laminar flow and heat transfer
equations using a finite-volume method. The numerical procedure
was validated by comparing the predicted local thermal resistances
and friction factor with the available experimental data. They have
found that the heat input lowers the frictional losses and viscosity
leading to an increase in the temperature of the water, particularly at
lower Reynolds numbers.
Tiselj et al. [8] performed experimental and numerical analysis of
the effect of axial conduction on the heat transfer in microchannel
heat sink with triangular microchannels. They pointed out that the
bulk water and heated wall temperatures did not change linearly
along the channel. In the study of Lee et al. [9], experiments were
conducted to explore the validity of classical correlations based on
conventional sized channels for predicting the thermal behavior in

0735-1933/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.icheatmasstransfer.2010.06.014
 P. Gunnasegaran et al. / International Communications in Heat and Mass Transfer 37 (2010) 1078–1086 1079

experimental data. Thus, the entire microchannel heat sink should be

Nomenclature
used as the computation domain instead of a single unit cell. In
practical simulations, the fluid should be supplied to and collected
A channel flow area, m2
from the microchannel heat sink via the inlet and outlet. There are
a channel width of trapezoidal and triangular, μm
many ways to arrange the inlet/outlet locations for the heat sink and
b bottom channel width of trapezoidal, μm
the configurations and it is expected to affect the fluid flow and heat
cp specific heat, J/kg.K
transfer characteristics inside the heat sink. Detailed understanding of
Dh hydraulic diameter, μm
the inlet/outlet arrangement effect on the heat sink performance was
f friction factor
numerically analyzed by Chein and Chen [11].
fRe Poiseuille number
Peng and Peterson [12,13] performed experimental investigations
h convective heat transfer, W/m2 K
on the pressure drop and convective heat transfer for water flow in
hc channel height of trapezoidal and triangular, μm
rectangular microchannels. It was found that the cross-sectional
Hc channel height of rectangular, μm
aspect ratio had a great influence on the flow friction and convective
κ thermal conductivity, W/m.K
heat transfer both in laminar and turbulent flows. Mala and Li [14]
κs solid thermal conductivity, W/m.K
measured the pressure drop and the flow rate for flow of deionized
Lc channel length, μm
water through microscale tubes with diameters ranging from 50 to
Nu Nusselt number
254 μm. The measured pressure drop was much higher than the
n direction normal to the wall
standard values. They found that the transition flow regime started at
P channel wet perimeter, μm
Re = 650. Liu et al. [15] studied convective heat transfer in a quart
Pr Prandtl number
microtube with three different inner diameters of 242, 315, and
P pressure, Pa
520 μm. They indicated that the experimental values for Nusselt
pin inlet pressure, Pa
number matched well with the laminar flow heat transfer correlation.
pout outlet pressure, Pa
They also indicated that laminar–turbulent transition Reynolds
q heat transfer rate, W
number was in the range of 1500–5500 for the microtubes.
qw heat flux at microchannel heat sink top plate, W/m2
Harms et al. [16] presented experimental data for a single-phase
Re Reynolds number
forced convection in deep rectangular microchannels. The channels
S distance between two microchannels, μm
were fabricated in a 2 mm thick silicon substrate by means of chemical
t substrate thickness, μm
etching and covered a total projected area of 2.5 cm by 2.5 cm. A thin-
T fluid phase temperature,
film heater was deposited on the back side of the silicon substrate,
Tin fluid inlet temperature, K
corresponding to the entire projected channel area. The silicon substrate
Ts microchannel heat sink solid temperature, K
measured 2.9 cm by 2.9 cm. All tests were performed with deionized
u fluid velocity, m/s
water as the working fluid, where the liquid flow rate ranged from
uin inlet fluid velocity, m/s
5.47 m3/s to 118 m3/s. A critical Reynolds number of 1500 for laminar–
Wc channel width of rectangular, μm
turbulent transition was found for this configuration. They indicated
x,y,z cartesian coordinates
that when Reynolds number and channel width are constant, the
pressure drop is inversely proportionally to the depth of the channel.
Recently there have been several studies that focused on various
Greek
aspects of microchannel geometry to enhance heat transfer. In
β tip angle of triangular microchannel
addition to the various interfacial effects discussed above, the cross-
μ viscosity, kg.m/s
sectional shape of the channel can have a great influence on the fluid
ρ density, kg/m3
flow and heat transfer inside noncircular microchannels was
θ dimensionless temperature
experimentally confirmed by Wu and Cheng [17]. Furthermore, it is
clear from the above literature review that a very little work has been
done to investigate the effect of geometrical parameters on the heat
Subscripts sink performance particularly for triangular microchannels. Thus, the
ch channel present work attempts to full fill the existing gap by studying the
h hydraulic effect of the geometrical parameters, the Reynolds number, and the
i inlet heat flux on pressure drop and laminar convective heat transfer in
o outlet microchannels with different shapes. The results from this investiga-
s solid tion on the effect of geometrical parameters should find its use in
w wall many industrial and natural processes in which the knowledge on the
heat transfer behavior is of uttermost importance.

2. Microchannel heat sink geometric configurations

single-phase flow through rectangular microchannels. A numerical
simulation was also carried out and compared with the experimental
data. They have concluded that both fluid flow and heat transfer are in
the developing regime and cannot be neglected in the analysis.
In the numerical simulations mentioned above, the computational
domain consists of only a single channel with the corresponding slice
of wall given as symmetrical boundary conditions was used. This kind
of computational domain referred to as the reduced model in the
study of Tiselj et al. [8]. As pointed out by Tiselj et al. [8] and Hetsroni
et al. [10], complete domain including geometric configurations of the
inlet/outlet, microchannel and heat sink base plate should be included
in the simulation in order to obtain results that agree with the
In this paper, rectangular, trapezoidal, and triangular-cross section
microchannels were analyzed. The physical configuration of one from
each shape of the microchannels used in the present investigation is
schematically shown in Fig. 1. Heat, supplied to the aluminum
substrate through a top plate, is removed by flowing water through a
number of microchannels. The dimensions of three different sets for
each shape of microchannels are given in Tables 1–3. In this study, the
effect of geometrical parameters on heat transfer characteristics of
microchannels heat sink with different shapes was considered. The
dimensions of microchannels can be varied by varying the channel
width and depth. When the channel width is increased, the channel
1080 P. Gunnasegaran et al. / International Communications in Heat and Mass Transfer 37 (2010) 1078–1086

Fig. 1. (a) Schematic diagram of the microchannels heat sink (b) section of different microchannel shapes with its dimensions.

height can be decreased. However, more computational effort is 2. Fluid is in single phase, incompressible and the flow is laminar.
required since denser grids are needed. 3. Properties of both fluid and heat sink material are temperature-
independent.
3. Mathematical formulation 4. All the surfaces of heat sink exposed to the surroundings are
assumed to be insulated except the top plate of heat sink where
3.1. Governing equations constant heat flux boundary condition simulating the heat
generation from electronic chip is specified.
To focus on the effect of the geometrical parameters on the heat
The continuity, momentum and energy equations for the current
sink performance, the following assumptions are made:
problem can be written as [18]:
1. Both fluid flow and heat transfer are in steady-state and three- Continuity:
dimensional.
∂U ∂V ∂W
+ + = 0: ð1Þ
Table 1 ∂X ∂Y ∂Z
Parameters for three different sets of rectangular microchannels.

Case 1 Case 2 Case 3 Momentum:

Hc (μm) 460 430 390

Wc (μm) 180 280 380 ∂U ∂U ∂U
x‐Momentum : U +V +W ð2aÞ
Lc (μm) 10,000 10,000 10,000 ∂X ∂Y ∂Z
S (μm) 596 500 404 !
2 2 2
Dh (μm) 259 339 385 dP 1 ∂ U ∂ U ∂ U
=− + + +
Number of channels 25 25 25 dX Re ∂X 2 ∂Y 2 ∂Z 2
 P. Gunnasegaran et al. / International Communications in Heat and Mass Transfer 37 (2010) 1078–1086 1081

Table 2 Boundary conditions at the inlet:

Parameters for three different sets of trapezoidal microchannels.

Case 1 Case 2 Case 3

U = 1; θ = 1: ð6aÞ

a (μm) 180 280 380

At the outlet:
b (μm) 125 225 325
hc (μm) 460 430 390
Lc (μm) 10,000 10,000 10,000 ∂θ
P = Pout ; = 0: ð6bÞ
S (μm) 596 500 404 ∂n
Dh (μm) 229 318 370
Number of channels 25 25 25 At the fluid–solid interface:

→ ∂θs ∂θ

 U = 0; θ = θs ; −ks = −k : ð6cÞ
∂V ∂V ∂V ∂n ∂n
y‐Momentum : +V U +W ð2bÞ
∂X ∂Y ∂Z
! At the top plate:
2 2 2
dP 1 ∂ V ∂ V ∂ V
=− + + +
dY Re ∂X 2 ∂Y 2 ∂Z 2 ∂θs
qw = −ks : ð6dÞ
∂n


∂W ∂W ∂W
z‐Momentum : U +V +W ð2cÞ In Eq. (6), U and θ are the dimensionless fluid inlet velocity and
∂X ∂Y ∂Z
! dimensionless temperature, respectively, P is the dimensionless
2 2 2
dP 1 ∂ W ∂ W ∂ W pressure at the outlet, n is the direction normal to the wall or the outlet
=− + + + :
dZ Re ∂X 2 ∂Y 2 ∂Z 2 plane, and qw is the heat flux applied at the top plate of the heat sink.
The heat flux that applied at the top plate was ranged from 100–
Energy: 1000 W/m2. The properties of water and solid used in the computation
! are ρ = 998.2 kg/m3, cp = 4182 J/kg.K, μ = 0.001003 kg/m.s, κ = 0.6 W/


∂θ ∂θ ∂θ 1 ∂2 θ ∂2 θ ∂2 θ m.K, and κs = 202.4 W/m.K.
U +V +W = + + : ð3Þ
∂X ∂Y ∂Z Re:Pr ∂X 2 ∂Y 2 ∂Z 2
4. Numerical parameters and procedures

The numerical computation was carried out by solving the

3.2. Boundary conditions
Boundary conditions for all boundaries are specified for this
simplified computational domain. At the entrance of the heat sink
assembly (z = 0, from Fig. 1), two types of boundaries are encoun-
tered which are water flows through the microchannels and removes
heat conducted to the surface of the heat sink. The remainder of the
entrance is occupied by the aluminum substrate. At the microchannels
sections, the inlet water velocity and the inlet water temperature are
specified. The inlet water temperature was taken as 293 K and the
inlet water velocity is calculated using
Re μ u the continuity equation. Even though the continuity equation does
uin = ;U =
ρ Dh uin
4A 2Hc Wc
Dh = = ð5Þ
P Hc + Wc
where A is the channel flow area, P is the channel wet perimeter, Hc is
the height of the rectangular microchannel, and Wc is the width of the
rectangular microchannel.
The Reynolds number considered in this work is ranged from 100
to 1000. In calculating the velocity, the water is assumed to be evenly
distributed into all microchannels. The transverse velocities at the
inlet are assumed to be zero. On the aluminum substrate, the
velocities are zero, and it is assumed to be an adiabatic surface.
Table 3
Parameters for three different sets of triangular microchannels.
Case 1 Case 2
governing conservation equations along with the boundary condi-
tions (Eqs. 1–6). Equations for solid and fluid phase were solved
simultaneously as a single domain conjugate problem. The discretiza-
tion of governing equations in the fluid and solid regions was done
using finite-volume method (FVM) with hybrid differencing
scheme [18]. The dimensionless parameters were calculated from
the computed velocity and temperature distribution. The flow field
was solved using the SIMPLE algorithm [19]. This is an iterative
solution procedure where the computation is initialized by guessing
the pressure field. Then, the momentum equation is solved to
determine the velocity components. The pressure is updated using
ð4Þ not contain any pressure, it can be transformed easily into a pressure
correction equation as shown by Patankar [18]. The iterations were
continued until the sum of residuals for all computational cells
became negligible (less than 10− 7) and velocity components did not
change from iteration to iteration. Because of the assumption of
constant fluid properties and negligible buoyancy, the mass and
momentum equations were not coupled to the energy equation.
Therefore, the temperature field was calculated by solving the energy
equation after a converged solution for the flow field which was
obtained by solving the momentum and continuity equations.
The distribution of hexahedral cells in the computational domain
was determined from a series of tests with different number of cells.
Computational cells with 1.9 × 105, 2.8 × 105, and 3.4 × 105 grids are
used to test the grid independence of the solution. The results have
shown that almost identical results are obtained when 2.8 × 105 and
3.4 × 105 grids are used. Thus, a computational cell with 2.8 × 105 grids
Case 3
is employed throughout the computations in this study.

a (μm) 180 280 380

5. Results and discussion
hc (μm) 460 430 390
Lc(μm) 10,000 10,000 10,000
S (μm) 596 500 404 5.1. Temperature distribution
Dh (μm) 148 203 238
β 22.14° 36.07° 51.95°
The dimensionless temperature profiles of the solid part of the
Number of channels 25 25 25
heat sinks for rectangular shaped microchannels are presented in
1082 P. Gunnasegaran et al. / International Communications in Heat and Mass Transfer 37 (2010) 1078–1086

Fig. 2. Dimensionless temperature profiles for each channel for rectangular shaped Fig. 3. Dimensionless temperature profiles along the length of the channel.
microchannel heat sinks.

increases nearly linearly. This is because the Reynolds number was

Fig. 2. For all types of microchannels, it is found that the high- increased by increasing the inlet velocity. As a result, there will be an
temperature region occurs at the edge of the heat sinks since there is increment of fluid velocity inside the channel. Therefore, this
no heat dissipation by fluid convection. The low-temperature region increment disturbs the flow and the heat transfer is strengthened.
occurs in the region where microchannels are placed, especially at the Fig. 6 shows the axial distribution of heat transfer coefficient for
entrance regions of the channels due to the high heat transfer hydraulic diameter, Dh = 0.259 mm for rectangular microchannel at
coefficient. The dimensionless temperatures for each channel can be Re = 500.The velocity at the entrance of the channel was kept
defined as: constant. It can be observed from this figure that the heat transfer
coefficient is large near the entrance and decreases downstream
Tf − Ti because of the development of the thermal boundary layer for all
θ= ð7Þ
Tw− Ti cases. As expected, there is a large variation of heat transfer coefficient
near the entrance. It can also be observed that high heat transfer
where Tf is the fluid temperature, Ti is the inlet temperature, and Tw is coefficient is obtained when the hydraulic diameter of the micro-
the wall temperature. channel is smaller.
It can clearly be seen from Fig. 2 that the lower temperature profile Shown in Fig. 7 the computed averaged heat transfer coefficient in
occurs in the channels located near the center of the heat sink. The each channel of rectangular heat sinks for various hydraulic
validity of present result shows similar trend with the result obtained diameters. The magnitude of heat transfer coefficient decreases with
by Chein and Chen [11]. For fluids in channels close to the edge of the the increase in hydraulic diameters and the variation trends are the
heat sink, higher temperature profiles are observed due to the heat same for each type of shape of the heat sink. This is because of small
transfer from the high-temperature edge of the heat sink.
The velocity along the flow direction must decrease sharply to
satisfy the needs of mass conservation equation. The decrease of
velocity causes the kinetic energy along the channel to decrease in
order to satisfy the needs of energy conservation equation. So, the
temperature along the flow direction should increase. Fig. 3 shows
dimensionless temperature profiles along the channel according to
the phenomena described above.
Fig. 4 shows the predicted water temperature rise, ΔT, between the
channel inlet and outlet of the heat sink for rectangular shaped
microchannel with Wc = 180 μm, Hc = 460 μm, and qw = 500 W/m2
respectively. As expected, for constant heat flux and water inlet
temperature, the outlet water temperature should decrease with
increasing of Reynolds number. The temperature rise decreases with
Reynolds number because a larger mass of fluid is available to carry
out the same amount of heat. Having validated the numerical method,
a more detailed depiction of the heat transfer characteristics of the
heat sinks can now be discussed.

5.2. Heat transfer coefficient

The heat transfer coefficient distribution with Reynolds number

for different channels is shown in Fig. 5. It can be seen that as the Fig. 4. Fluid temperature rise versus Reynolds number for rectangular shaped
Reynolds number increases, the average heat transfer coefficient also microchannel.
 P. Gunnasegaran et al. / International Communications in Heat and Mass Transfer 37 (2010) 1078–1086
Fig. 5. Averaged heat transfer coefficient distributions for rectangular shaped
microchannels for channels #1 and #14.
pressure drop in larger hydraulic diameter corresponds to smaller
flow rate driven into the heat sink. Furthermore, due to the difference
of the channel hydraulic diameter which is caused by the difference of
channels’ area and perimeter, the averaged heat transfer coefficient in
each heat sink is also different under a given inlet velocity. For each
type of the heat sinks, the middle channel (channel 14) has the
highest averaged heat transfer coefficient value as expected. The
averaged heat transfer coefficient value for other channels is seen to
decrease depending on their distances from the wall. The averaged
heat transfer coefficient distribution for all types of the heat sink is
almost symmetrical with respect to the centerline of the heat sink. In
overall view, microchannels provide high heat transfer coefficients for
small hydraulic diameters for each type of heat sink.
Fig. 8 shows the computed heat transfer coefficient versus
Reynolds number for different shaped microchannel of the heat
sinks. The magnitude of heat transfer coefficient increases with the
increase of Reynolds number. It is apparent from this figure that the
heat transfer coefficient for rectangular microchannel heat sink is the
Fig. 6. Heat transfer coefficient along the length of the channel.
1083
Fig. 7. Computed averaged heat transfer coefficient in each channel of the heat sinks.
highest. The triangular microchannel heat sink exhibits the lowest
values of heat transfer coefficient. While, the heat transfer coefficient
for trapezoidal microchannel heat sink is between the rectangular and
triangular microchannel.
5.3. Pressure drop profile
The pressure drop profile for different types of microchannel
configurations is shown in Figs. 9–11. From these figures, it can be
seen that the pressure drop rises linearly with the increase of
Reynolds number. A similar trend was observed for all cases studied. It
should be noted that the fluid path in the numerical simulation does
not include the fittings and pipes between the pressure transducers
used in the actual design of a heat sink. These minor loses increase
with fluid kinetic energy, which it is proportional to the square of
Reynolds number. It is also observed that the variation of pressure
drop for all cases of microchannels increases with the decrease of
channel's hydraulic diameter.
Fig. 8. Averaged heat transfer coefficient for different shaped microchannels.
1084 P. Gunnasegaran et al. / International Communications in Heat and Mass Transfer 37 (2010) 1078–1086
Fig. 9. Pressure drop variations versus Reynolds number for rectangular shaped
microchannels.
5.4. Effect of heat flux
The computed pressure drop versus Reynolds number for different
heat flux values is depicted in Fig. 12. It is observed that the largest
pressure drop has occurred at qw = 100 W/m2. While, the lowest
pressure drop has occurred at qw = 1000 W/m2 for the same Reynolds
number. This is because of the higher fluid temperature at higher
constant heat flux condition decreases the water viscosity which leads
to decrease the pressure drop value. The pressure drop trend for
different heat fluxes presented in Fig. 12 shows exactly similar trend
as given by Qu and Mudawar [20].
5.5. Effect of geometrical parameters
Only a few studies have considered the effect of roughness on
laminar flow such as Moody [21], Zhang and Jia [22], and Nikuradse
[23]. They concluded that the roughness effect on the laminar flow
characteristics can be ignored if internal relative wall roughness is less
Fig. 10. Pressure drop variations versus Reynolds number for trapezoidal shaped
microchannels.
Fig. 11. Pressure drop variations versus Reynolds number for triangular shaped
microchannels.
than 5%. However, flows in microchannels are often laminar, so the
study of laminar flow in rough microchannels has become important.
Therefore, the effect of geometrical parameters on the Poiseuille
number which is product of friction factor, f and Reynolds number,
respectively, for different microchannel shapes are discussed in this
section. The friction factor is calculated using Darcy equation [21]
2Dh Δp
f = ð8Þ
ρ u 2 Lc
where Dh is the hydraulic diameter, Δp is the pressure drop, ρ is the
density of fluid, u is the velocity of fluid, and Lc is the length of channel.
The Poiseuille number at different Wc/Hc ratio for rectangular
shaped microchannels is shown in Fig. 13. It is clearly observed that
the Poiseuille number increases with the increase of Wc/Hc ratio of the
channel. Therefore, the flow resistance will increase evidently when
the ratio is increased. This is because of the increase in Wc/Hc ratio
causes a decrease in the flow area and the vortex effect becomes more
significant. The validity of the present numerical simulation was also
Fig. 12. Pressure drop versus Reynolds number with different heat fluxes.
 P. Gunnasegaran et al. / International Communications in Heat and Mass Transfer 37 (2010) 1078–1086
Fig. 13. Poiseuille number at different width–height (Wc/Hc) ratios for rectangular
shaped microchannels.
proved with Kandlikar et al. [24] results, where the Poiseuille number
increases with the increase of Wc/Hc ratio for rectangular shaped
microchannels. The Poiseuille number increases with the increase of
Reynolds number. This is because of at lower Reynolds number, the
water attains higher temperatures leading to lower viscosities. This
resulted in lower pressure drop and thus decreases in the value of fRe.
Fig. 14 shows a comparison of Poiseuille number predicted from
the present numerical simulation for trapezoidal microchannels with
the experimental work done by Wu and Cheng [17]. Very good
agreement was obtained between the two studies. The Poiseuille
number rises with the increase of Reynolds number. There are three
geometric parameters including bottom-to-top width ratio (b/a),
height-to-top width ratio (h/a), and length-to-hydraulic diameter
ratio (L/Dh), which affect the friction and heat transfer in the
trapezoidal microchannels.
The Poiseuille number with different tip angles of triangular
shaped microchannels is illustrated in Fig. 15. It is clearly observed
that the Poiseuille number increases with the increases of tip angle of
Fig. 14. A comparison of Poiseuille number with Wu and Cheng [17] for trapezoidal
shaped microchannels.
1085
Fig. 15. Poiseuille number with different tip angles of triangular shaped microchannels.
the channel. The validity of the present numerical simulation result
was also proved with the results of Kandlikar et al. [24]. The Poiseuille
number for triangular shaped microchannel also increases with the
increase of Reynolds number because at lower Reynolds numbers, the
water attains higher temperatures leading to lower viscosities. This
resulted in lower pressure drop and thus decreases in the value of fRe.
In overall comparison, for triangular shaped microchannels (with
b/a = 0), the fRe are much lower than that of rectangular and
trapezoidal microchannels. The microchannels with trapezoidal
cross-section, the Poiseuille number are lower than that of rectangular
channels (with b/a = 1) depending on the b/a values. The Poiseuille
number for trapezoidal microchannels increases with the increase of
aspect ratio (b/a). Therefore, the Poiseuille number of water flowing in
microchannels, having the same width and height but with different
cross-sectional shapes can be very much different due to the
difference in cross-sectional shape and geometrical dimensions of
the channels.
6. Conclusions
Numerical simulation on the fluid flow and heat transfer
characteristics in full scale microchannel heat sinks was performed
in this study. The effect of different geometrical parameters such as
the differences in hydraulic diameters, height, and width of three
different shapes of microchannel heat sinks was comprehensively
studied. Based on the simulated results, the following conclusions can
be made:
1. For rectangular shaped microchannel heat sinks, the smallest
hydraulic diameter has the greatest value of heat transfer
coefficient.
2. The heat transfer coefficient and Poiseuille number increase with
the increase of Reynolds number. For microchannels with
rectangular shape, the heat transfer coefficient and Poiseuille
number are the highest, for microchannels with triangular shape
are the lowest, and for microchannels with trapezoidal shape are in
between.
3. The pressure drop decreases with the increase of heat flux from top
plate of heat sink for the same Reynolds number.
4. For rectangular shaped microchannels, the width height ratio (Wc/
Hc) has a significant effect on the Poiseuille number. The Poiseuille
number is most for Wc/Hc = 0.974, less for Wc/Hc = 0.651, and least
for Wc/Hc = 0.391.
1086 P. Gunnasegaran et al. / International Communications in Heat and Mass Transfer 37 (2010) 1078–1086
5. The height-to-top width ratio (H/a), the bottom-to-top width ratio
(b/a), and length-to-hydraulic diameter ratio (L/Dh) are the
important design parameters for trapezoidal microchannels.
Poiseuille number increases evidently when H/a and L/Dh decreases
while b/a increases.
6. The tip angle of triangular shaped microchannels has a great effect
on the Poiseuille number. Poiseuille number increases evidently
with the increase of tip angle from β = 22.14°–51.95°.
7. In order to achieve overall heat transfer enhancement, the
rectangular shaped microchannels is the best followed by
trapezoidal and triangular shaped microchannels, respectively.
Acknowledgments
The authors would gratefully acknowledge the financial support
provided by the University Council through grant no. J510050203.
References
[1] D.B. Tuckerman, R.F. Pease, High performance heat sinking for VLSI, IEEE Electron.
Devices Lett. EDL-2 (1981) 126–129.
[2] W. Qu, G.M. Mala, L. Dongqing, Pressure-driven water flows in trapezoidal silicon
microchannels, Int. J. Heat Mass Transfer 43 (1999) 353–364.
[3] W. Qu, G.M. Mala, L. Dongqing, Heat transfer for water flow in trapezoidal silicon
microchannels, Int. J. Heat Mass Transfer 43 (2000) 3925–3936.
[4] D. Yu, R. Warrington, R. Barren, T. Ameel, An experimental and theoretical
investigation of fluid flow and heat transfer in microtubes, Proc. ASME/JSME
Therm. Eng. Conf. 1 (1995) 523–530.
[5] J.N. Pfahler, Liquid transport in micron and submicron size channels, Ph.D. thesis,
University of Pennsylvania, 1992.
[6] M.E. Steinke, S.G. Kandlikar, Single-phase liquid friction factors in microchannels,
Int. J. Therm. Sci. 45 (2006) 1073–1083.
[7] K.C. Toh, X.Y. Chen, J.C. Chai, Numerical computation of fluid flow and heat
transfer in microchannels, Int. J. Heat Mass Transfer 45 (2002) 5133–5141.
[8] I. Tiselj, G. Hetsroni, B. Mavko, A. mosyak, E. Pogrebnyak, Z. Segal, Effect of axial
conduction on the heat transfer in microchannels, Int. J. Heat Mass Transfer 47
(2004) 2551–2565.
[9] P.S. Lee, S.V. Garimella, D. Liu, Investigation of heat transfer in rectangular
microchannels, Int. J. Heat Mass Transfer 48 (2005) 1688–1704.
[10] G. Hetsroni, A. Mosyak, Z. Segal, Nonuniform temperature distribution in
electronic devices cooled by flow in parallel microchannels, IEEE Trans. Compon.
Packag. Technol. 24 (2001) 16–23.
[11] R. Chein, J. Chen, Numerical study of the inlet/outlet arrangement effect on
microchannel heat sink performance, Int. J. Therm. Sci. 48 (2009) 1627–1638.
[12] X.F. Peng, G.P. Peterson, Convective heat transfer and flow friction for water flow
in microchannels structures, Int. J. Heat Mass Transfer 39 (1996) 2599–2608.
[13] X.F. Peng, G.P. Peterson, The effect of thermofluid and geometrical parameters on
convection of liquids through rectangular microchannels, Int. J. Heat Mass
Transfer 38 (1995) 755–758.
[14] G.M. Mala, D.Q. Li, Flow characteristics of water in microtubes, Int. J. Heat Fluid
Flow 20 (1999) 142–148.
[15] J.T. Liu, X.F. Peng, W.M. Yan, Numerical study of fluid flow and heat transfer in
microchannel cooling passages, Int. J. Heat Mass Transfer 50 (2007) 1855–1864.
[16] T.M. Harms, M. Kazmierczak, F.M. Gerner, A. Holke, H.T. Henderson, J. Pilchowski,
K. Baker, Experimental investigation of heat transfer and pressure drop through
deep microchannels in a (110) silicon substrate, Proc. ASME Heat Transfer Division
1, ASME/HTD, 351, 1997, pp. 347–357.
[17] H.Y. Wu, P. Cheng, An experimental study of convective heat transfer in silicon
microchannels with different surface conditions, Int. J. Heat Mass Transfer 46
(2003) 2547–2556.
[18] S.V. Patankar, Numerical Heat Transfer and Fluid Flow, Hemisphere, New York,
1980.
[19] J.D. Anderson, Computational Fluid Dynamic: The Basics with Applications,
McGraw-Hill, New York, 1995.
[20] W. Qu, I. Mudawar, Experimental and numerical study of pressure drop and heat
transfer in a single-phase microchannel heat sink, Int. J. Heat Mass Transfer 45
(2002) 2549–2565.
[21] L.F. Moody, Friction factors for pipe flow, J. Heat Transfer 66 (8) (1944) 671–684.
[22] T.T. Zhang, L. Jia, Numerical simulation of roughness effect on gaseous flow and
heat transfer in microchannels, IEEE Int. Conf. on Nano/Micro Eng. Mol. Syst. 2
(2007) 136–141.
[23] J. Nikuradse, Strmungsgesetze in rauhen Rohren, V. D. I. Forschungsheft 361
(1933) 1–22.
[24] S. Kandlikar, S. Garimella, D. Li, S. Colin, M.R. King, Heat transfer and fluid flow in
minichannels and microchannels, Elsevier, USA, 2005.