micromachines

Article
Numerical Study on the Fluid Flow and Heat Transfer
Characteristics of Al2O3-Water Nanofluids in Microchannels
of Different Aspect Ratio
Huajie Wu 1, * and Shanwen Zhang 2

1 College of Traffic Engineering, Yangzhou Polytechnic Institute, Yangzhou 225127, China

2 College of Mechanical Engineering, Yangzhou University, Yangzhou 225127, China; swzhang@yzu.edu.cn
* Correspondence: whj_china@163.com

Abstract: The study of the influence of the nanoparticle volume fraction and aspect ratio of mi-
crochannels on the fluid flow and heat transfer characteristics of nanofluids in microchannels is
important in the optimal design of heat dissipation systems with high heat flux. In this work, the
computational fluid dynamics method was adopted to simulate the flow and heat transfer character-
istics of two types of water-Al2 O3 nanofluids with two different volume fractions and five types of
microchannel heat sinks with different aspect ratios. Results showed that increasing the nanoparticle
volume fraction reduced the average temperature of the heat transfer interface and thereby improved
the heat transfer capacity of the nanofluids. Meanwhile, the increase of the nanoparticle volume
fraction led to a considerable increase in the pumping power of the system. Increasing the aspect
ratio of the microchannel effectively improved the heat transfer capacity of the heat sink. Moreover,





 increasing the aspect ratio effectively reduced the average temperature of the heating surface of the
Citation: Wu, H.; Zhang, S.
heat sink without significantly increasing the flow resistance loss. When the aspect ratio exceeded 30,
Numerical Study on the Fluid Flow the heat transfer coefficient did not increase with the increase of the aspect ratio. The results of this
and Heat Transfer Characteristics of work may offer guiding significance for the optimal design of high heat flux microchannel heat sinks.
Al2 O3 -Water Nanofluids in
Microchannels of Different Aspect Keywords: microchannel; nanofluid; heat transfer enhancement; numerical simulation
Ratio. Micromachines 2021, 12, 868.
https://doi.org/10.3390/mi12080868

Academic Editors: Junfeng Zhang 1. Introduction

and Ruijin Wang
With the continuous miniaturization and integration of electronic devices, the heat
flux density continues to increase, and conventional cooling methods are no longer effec-
Received: 17 June 2021
tive. Therefore, the heat dissipation problem of high heat flux density has been a research
Accepted: 22 July 2021
hotspot in the field of heat transfer [1,2]. In 1981, Tuckerman and Pease [3] proposed a
Published: 24 July 2021
silicon microchannel heat sink, which uses deionized water as a liquid cooling medium
and whose heat dissipation capacity can be as high as 790 W/cm2 . In 1995, Choi and
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
Eastman [4] proposed the concept of nanofluids; a proportion of solid particles with di-
published maps and institutional affil-
ameters less than 100 nm were added into a base fluid with low thermal conductivity,
iations.
and the resulting suspension with high thermal conductivity was found to be relatively
stable. This special liquid can significantly improve the convective heat transfer capacity
of cooling media. The addition of nanoparticles to high-Prandtl number liquids signif-
icantly increases the heat transfer performance of micro heat-sinks [5]. An increase in
nanoparticle concentration can lead to an increase in thermal conductivity and viscosity
Copyright: © 2021 by the authors.
and an increase in nanoparticle size [6]. Nanofluids contribute to the improvement of
Licensee MDPI, Basel, Switzerland.
This article is an open access article
heat transfer processes and reduce and optimize thermal systems. Different properties,
distributed under the terms and
such as wettability and thermal conductivity, can be adjusted by altering the nanoparticles’
conditions of the Creative Commons
concentration, thereby making nanofluids suitable for a wide range of applications [7].
Attribution (CC BY) license (https:// Nanofluids contain metal or nonmetal particles with nanometer sizes and exhibit much
creativecommons.org/licenses/by/ greater thermal conductivity. M. Goodarzi’s [8] expression for calculating the enhanced
4.0/). thermal conductivity of nanofluids has been derived from the general solution of the

Micromachines 2021, 12, 868. https://doi.org/10.3390/mi12080868 https://www.mdpi.com/journal/micromachines

Micromachines 2021, 12, 868 2 of 14

heat conduction equation in spherical coordinates and the equivalent hard-sphere fluid
model representing the microstructure of particle/liquid mixtures. The cooling method of
microchannel heat sinks combined with nanofluids has become one of the effective ways
to solve the heat dissipation problem of high heat flux. By comparing the heat transfer
characteristics of trapezoidal, semicircular and rectangular cross-section microchannels,
Vinoth et al. [9] found that, compared with rectangular and semicircular cross-section
microchannels, trapezoidal cross-section microchannels have the best heat transfer effect
because of their larger wall area and effective inlet length, but a larger pressure drop. For
some special-shaped cross-sections, Alfaryjat et al. [10] used numerical simulation methods
to study hexagonal, circular and rhombic cross-section microchannels. The results show
that the heat transfer coefficients of hexagonal cross-section microchannels are the highest,
followed by circular and rhombic cross-section microchannels. Ahmed et al. [11] used
the three-dimensional numerical simulation method to optimize the microchannel with
triangular, trapezoidal and rectangular grooves. The results showed that the optimized
microchannel with trapezoidal groove had the best heat transfer effect, the Nusselt number
increased by 51.59% and the friction coefficient increased by 2.35% compared with the
optimization. Kumar [12] used the finite volume method to simulate and optimize the
trapezoidal microchannel. The results show that the heat transfer performance of trape-
zoidal microchannels with semicircular grooves is 16% higher than that of trapezoidal
microchannels with a rectangular groove, but the friction coefficient is 18%. At present, the
structure optimization of microchannels has some limitations; namely, the processing is
very difficult, and the cost is high. Researchers gradually try to improve the heat transfer
by changing the flow medium in the microchannel. Farsad et al. [13] numerically simulated
the heat transfer performance of Al2 O3 , CuO and Cu-H2 O nanofluids in copper rectangular
microchannels. The results show that the thermal conductivity of metal nanofluids is
higher than that of metal oxide nanofluids. Researchers have conducted numerous studies
to obtain the optimized aspect and states in order to improve the heat transfer of equipment
and use various nanofluids. Shi Xiaojun et al. [14] carried out a multi-objective optimization
design on a single-layer nanofluid rectangular microchannel. The results show that the
pump power and thermal resistance are more sensitive to the channel width and spacing
ratio than the aspect ratio. Naphon and Khonseur [15] used air as a cooling medium to con-
duct an experimental study on the flow and heat transfer characteristics of microchannels
with different heights and widths in the Reynolds number range of 200–1000. The results
showed that the height and width of rectangular microchannels exert a significant impact
on their heat exchange effect and resistance loss. Studies of this research indicate that the
fluid in the indented sections has a higher heat transfer with the heated wall. Karimipour
et al. [16] numerically studied a two-dimensional indented rectangular microchannel. They
concluded that by increasing the volume fraction of nanoparticles, the thermal efficiency
of the nanofluid is enhanced. Yari Ghale et al. [17] numerically studied the laminar and
forced flow of a Water/Al2 O3 nanofluid in an indented microchannel by using two-phase
or single-phase methods. Their results showed that the Nusselt numbers and friction
factors in an indented microchannel are higher compared to the smooth microchannel,
and therefore, these parameters can improve fluid flow efficiency by increasing the width
of the rib. A segmental analysis pertaining to the heat exchanger takes place to evaluate
the influence of nanofluid usage on the heat transfer coefficient, the exchanger’s length
and its pressure drop. When the volume fraction of Al2 O3 nanofluids is 5%, the heat
transfer coefficient is increased by 10% compared with pure water, and the pressure drop
is significantly reduced [18]. T Raghuraman [19] used pure water as a cooling medium to
study the effect of the microchannel aspect ratio on heat transfer performance; the study
showed that the microchannel aspect ratio influences the heat transfer coefficient, pumping
power, pressure drop and heat transfer performance at different Reynolds numbers. A large
aspect ratio can enhance heat transfer, but it can also increase power consumption. Based
on the computational fluid dynamics (CFD) method, Mohamadpour et al. [20] carried out
a numerical simulation study on the heat transfer efficiency of the cooperative jet in the
Micromachines 2021, 12, 868 3 of 14

microchannel. It was found that increasing the jet frequency and pulse amplitude can
significantly improve the heat transfer ability of the microchannel.
In the current work, a rectangular microchannel heat sink is used as the research object,
and a three-dimensional flow and heat transfer numerical simulation study is conducted
on the basis of the computational fluid dynamics method. Water/Al2 O3 nanofluids of
different concentrations are utilized as the cooling medium, and their influence on the
heat transfer performance of microchannels is analyzed. This work also focuses on the
evaluation of the flow resistance characteristics and heat transfer laws of microchannels
with different aspect ratios. The purpose of the study is to provide theoretical guidance for
the optimal design of high heat flux density microchannel heat sinks.

2. Numerical Method and Model Description

2.1. Mathematical Model
The single-phase fluid model is commonly used to study the flow and heat transfer
of nanofluids in microchannels. The nanoparticles in nanofluids are considered to be
uniformly distributed in base fluids, and they are in thermal equilibrium. No relative
slip velocity exists between nanoparticles and base liquids, and the flow is regarded as an
incompressible steady laminar flow.
On the basis of these assumptions, the governing equations of nanofluid flow and
heat transfer can be expressed as:

∂

ρnf uj = 0 (1)
∂xj

" !#
∂ ρnf ui uj ∂p ∂ ∂ui ∂uj
=− + µ + (2)
∂xj ∂xi ∂xj nf ∂xj ∂xi
" #
∂
∂ ∂T
ρ Cp Tuj = λ (3)
∂xj f f ∂xj nf ∂xj

where ui and uj are velocity components, xi and xj are Cartesian coordinate components,
p is the pressure in the flow field, T is the temperature, ρnf is the density of nanofluids,
and µnf is the dynamic viscosity. Cpf is the specific heat capacity, and λnf is the thermal
conductivity. The calculation formula is as follows [21,22]:

ρnf = (1 − α)ρw + αρp (4)

 
µnf = 1 + 0.025α + 0.015α2 µw (5)
h
i
Cpf = (1 − α)(ρCp )w + α ρCp p /ρnf (6)

λp + ( n − 1) λw − ( n − 1) α ( λw − λp )
λnf = λp (7)
λp + ( n − 1) λw + α ( λw − λp )
The subscripts w and p denote the corresponding thermophysical properties of the
base fluid and nanoparticles, respectively. α represents the volume fraction of nanoparticles,
and n is the shape factor of nanoparticles. In this work, nanoparticles are regarded as
regular spheres with a value of n = 3.
The heat distribution in the solid area of the heat sink can be calculated by the
following formula, Ts is the temperature of the solid region; λs is the thermal conductivity
of the solid region:  
∂ ∂Ts
λs =0 (8)
∂xi ∂xi
 The heat distribution in the solid area of the heat sink can be calculated by the fol-
lowing formula, Ts is the temperature of the solid region; λs is the thermal conductivity
of the solid region:

∂  ∂Ts 
Micromachines 2021, 12, 868  λs =0 (8)
4 of 14
∂xi  ∂xi 

2.2. D Model and Boundary Conditions

2.2. D Model and Boundary Conditions
A microchannel heat sink is usually composed of more than 10, or even dozens, of
A microchannel heat sink is usually composed of more than 10, or even dozens, of
microchannels, and complete modeling and simulation require a large amount of compu-
microchannels, and complete modeling and simulation require a large amount of comput-
ting resources. As a result of the symmetry of the model, a typical microchannel heat sink
ing resources. As a result of the symmetry of the model, a typical microchannel heat sink
unit can be extracted for simulation. The structure and size of the microchannel heat sink
unit can be extracted for simulation. The structure and size of the microchannel heat sink
used
usedininthis
thisstudy
studyare areshown
shownin inFigure
Figure1.1.The
Thewidth
widthW Wofofthe
theheat
heatsink
sinkunit
unitisis11mm,
mm,the the
length L is 50 mm, and the microchannel width W c is 0.5 mm. The microchannel height
length L is 50 mm, and the microchannel width W c is 0.5 mm. The microchannel height W h
W h values are 5, 10, 15, 20 and 25 mm; they correspond to five different aspect ratios, that
values are 5, 10, 15, 20 and 25 mm; they correspond to five different aspect ratios, that is,
is,
HW HW = =WW h/Wc with values of 10, 20, 30, 40 and 50, respectively. The bottom height of the
h /W c with values of 10, 20, 30, 40 and 50, respectively. The bottom height of the
heat
heat sink isset
sink is setatat66mm.
mm.Symmetrical
Symmetricalboundary
boundaryconditions
conditionsare areset
seton
onboth
bothsides
sidesof ofthe
theheat
heat
sink, and a uniform heat flux q = 0.8 MW/m 2 2is set at the bottom. The top wall is set as the
sink, and a uniform heat flux q = 0.8 MW/m is set at the bottom. The top wall is set as the
adiabatic
adiabaticboundary.
boundary. The The inlet of the
inlet of the microchannel
microchannelfluid fluidadopts
adoptsthethevelocity
velocityinlet
inletboundary,
bound-
ary,
the the inlet
inlet fluid
fluid temperature
temperature is set
is set at at 300K,K,the
300 theoutlet
outletof ofthe
the microchannel
microchannel fluid
fluid isis set
setasas
pressure
pressureoutlet,
outlet, the
the upper
upper wall
wall ofof the
the fluid
fluiddomain
domainadopts adoptsan anadiabatic
adiabaticsolid
solidwall
wallboundary
bound-
ary condition
condition andandthethe interface
interface between
between fluid
fluid and and solid
solid adopts
adopts a coupled
a coupled heat
heat flow
flow bound-
boundary
ary condition. In this paper, ICEM CFD software is used for 3D modeling
condition. In this paper, ICEM CFD software is used for 3D modeling and mesh generation, and mesh gen-
eration, and the general CFD software FLUENT 15.0 is used for
and the general CFD software FLUENT 15.0 is used for numerical simulation. According to numerical simulation.
According
the geometric to the geometric
models models with
with different aspectdifferent aspect
ratios, three ratios,
sets three
of grids aresets of grids
divided are di-
to analyze
vided to analyze
the grid independencethe grid independence
under the maximum under the maximum
Reynolds number. Reynolds number.

Figure
Figure1.1.Schematic
Schematicofofheat
heatsink
sinkwith
withsquare
squarecross-section.
cross-section.

2.3.Model
2.3. ModelValidation
Validation
Theexperimental
The experimentaldata dataofofLei
Lei[23]
[23]are
areused
usedfor
for comparison
comparisonto to verify
verify the
the prediction
prediction
performance of the mathematical model. The microchannel width W
performance of the mathematical model. The microchannel width Wc is 0.1 mm, the
c is 0.1 mm, theheight
height
W 0.5 mm, the length L is 10 mm, and the heat flux q
Wh his 0.5 mm, the length L is 10 mm, and the heat flux q at the bottom of the heat sink isis
is at the bottom of the heat sink
0.6MW/m
MW/m 2 . The cooling medium is pure water. In the calculation, the volume fraction of
0.6 2. The cooling medium is pure water. In the calculation, the volume fraction of
thenanoparticles
the nanoparticlesααisis0.0.
The characteristic scale of microchannels can be defined as:
2Wh Wc
Dh = (9)
Wh + Wc

The microchannel Reynolds number is:

ρf Uin Dh
Re = (10)
µf

where Uin is the fluid inlet velocity.

 The microchannel Reynolds number is:
ρf U in Dh
Re = (10)
μf
Micromachines 2021, 12, 868 5 of 14
where Uin is the fluid inlet velocity.
The average temperature Tc of the heat transfer surface and the average temperature
Tf of the whole microchannel fluid can be respectively defined as:
The average temperature Tc of the heat transfer surface and the average temperature
 TdA
c =
Tf of the whole microchannel fluid canTbe respectively defined as: (11)
RdA
TdA
Tc = R (11)
 f
ρ TdVdA
Tf = R (12)
Tf = R
 ρρf dV
f TdV
(12)
ρf dV
Therefore, the average heat transfer coefficient can be expressed as:
Therefore, the average heat transfer coefficient can be expressed as:
qAb
h= (13)
A (TqA −T )
h = c c bf (13)
Ac ( Tc − Tf )
where Ab is the heating area of the bottom of the thermal sink and Ac is the heat exchange
where
area Ab is the fluid
between heating area of
domain andthethe
bottom
solid of the thermal
domain in thesink and Ac is the heat exchange
microchannel.
area
Thebetween
average the fluid domain
Nusselt numberand the solid
is defined asdomain
follows:in the microchannel.
The average Nusselt number is defined as follows:
hD
Nu = h (14)
λhD
Nu = f h (14)
λf
The comparison between the average Nusselt number predicted by the mathematical
model The
and comparison
the experimentalbetween the average
results is shown Nusselt
in Figure number predicted
2. The by thethat
results show mathematical
the pro-
model and the experimental results is shown in Figure 2. The
posed model achieves certain accuracy and reliability in the prediction of fluid laminar results show that the
proposed
flow and heatmodel achieves
transfer certain accuracy
in microchannels. Theand reliability
article in the prediction
only carried of fluid
out simulation laminar
research,
flow
and andwill
there heatbetransfer
errors in microchannels.
between The article
the simulation onlyand
results carried out simulation
the experiment. research,
There are
and there
many sourceswillofbeerror,
errorssuch
between
as thetheflow
simulation
may not results and the experiment.
be completely There
laminar, the are many
number of
sources
grids, the of error,
finite such as
element the flow
method and may not be completely laminar, the number of grids, the
so on.
finite element method and so on.

10
8
6
Nu

4 Exp.
CFD
2
0
0 200 400 600 800 1000
Re
Figure 2. 2.
Figure Model validation
Model byby
validation comparing thethe
comparing present results
present with
results LeiLei
with et et
al.al.
experiments [23].
experiments [23].

3. Result Analysis and Discussion

3.1. Influence of Nanoparticle Volume Fraction
Take the microchannel heat sink with aspect ratio HW = 10 (HW = W h /W c ) as an
example. The three-dimensional flow and heat transfer of the nanofluids are simulated in
the Reynolds number range of 100–500, and the nanoparticle volume fractions α are 0.5%
and 5%, respectively. According to Formula (10), the Reynolds number is directly related
to density, inlet velocity, characteristic scale and viscosity, while density and viscosity are
related to the content of solid particles in nanofluids. In this paper, the physical properties
of the fluid and the characteristic scale of the channel are determined by giving the solid
particle content and aspect ratio in the nanofluid. Finally, the Reynolds number can be
changed by adjusting the inlet velocity. The temperature distributions in the middle section
of the heat sink are extracted for comparison, and the results are shown in Figure 3. The
laws of the temperature distributions in the middle sections of two types of nanofluids at
different Reynolds numbers are similar. The temperature at the bottom of the heat sink is
the highest, and the temperature in the solid region decreases gradually along the height
 as the evaluation index to reasonably evaluate the synergistic effect of inlet velocity and
resistance loss. Its physical meaning is the external work required for nanofluids to pass
through microchannels. Pumping power p is expressed as:
P = N ⋅ U in ⋅ Wh ⋅ Wc ⋅Δ p (15)
Micromachines 2021, 12, 868 6 of 14
N is the number of microchannels in the whole heat sink, and the value is N = 1.
The variation of pumping power with the Reynolds number is shown in Figure 9.
With the increase of the Reynolds number, the pumping power of the nanofluid with a
of the microchannel. The temperature of the fluid at the center of the fluid domain of
nanoparticle volume fraction of 5% is significantly higher than that of the nanofluid with
the microchannel is relatively low. At the fluid structure coupling heat transfer surface,
a nanoparticle volume fraction of 0.5%. This result shows that the heat transfer perfor-
the fluid temperature is relatively high because of the heating of the solid surface. At the
mance cannot be improved by increasing the nanoparticle volume fraction because doing
same time, with the increase of the Reynolds number, the heat sink temperature and fluid
so greatly increases the power consumption of the whole system. At the same time, a high
temperature decrease significantly. When the Reynolds number remains the same, the
volume fraction renders the nanoparticles in nanofluids unable to maintain a stable and
heat sink temperature and fluid temperature of the nanofluid with a nanoparticle volume
uniform of
fraction suspension state [23]. Therefore,
5% are significantly lower thaninthose
engineering applications,
of the nanofluid withthe nanoparticle
a nanoparticle
volume fraction
volume fraction of
in 0.5%.
nanofluids needs to be maintained at a low level.

Figure 3. Temperature distributions of

Temperature distributions of fluid
fluid and
and solid
solid domains
domains at
at the
themiddle
middleplane.
plane.

The average temperature on the heat exchange surface of the fluid and solid domains
under each working condition is obtained. The results are shown in Figure 4. With the
increase of the Reynolds number, the average temperature on the heat transfer surface of the
two types of nanofluids decreases. At the same Reynolds number, increasing the volume
fraction of nanoparticles in the nanofluid can reduce the temperature on the heat transfer
surface. In the range of Reynolds numbers studied in this work, the average temperature
difference on the heat transfer surface caused by the differences in nanoparticle volume
fraction decreases with the increase of the Reynolds number. When the Re is 100, the
average temperature difference between the nanofluid with a nanoparticle volume fraction
of 5% and that with a nanoparticle volume fraction of 0.5% is 6.3 K. When the Re is 500,
the average temperature difference of the heat exchange surface decreases to 2.6 K. This
result shows that the average temperature of the heat exchange surface can be reduced
by increasing the nanoparticle volume fraction under the condition with a low Reynolds
number.
 Micromachines 2021, 12, x FOR PEER REVIEW 8 of 14
Micromachines 2021, 12, 868 7 of 14

340

335 α=0.5%
Micromachines 2021, 12, x FOR PEER REVIEW
Micromachines 2021, 12, x FOR PEER REVIEW330
α=5% 8 of 14
8 of 14

Tc/K
325

320
340
340
315
335 α=0.5%
335
310 α=0.5%
330 100 200 300 α400
=5%
330 α=5% 500
Re
Tc/K

325
Tc/K

Figure 4.325
Figure Average temperature
4. Average distributions
temperature of the
distributions fluid-solid
of the surface.
fluid-solid surface.
320
320
Figure 5 shows the effects of the nanoparticle volume fraction on heat transfer coeffi-
315
7000
cients 315
with different Reynolds numbers. The results show that the average heat transfer
coefficient
310 increases with the increase of the Reynolds number and that the increase of the
6500
310
100
nanoparticle 200 fraction
volume 300can improve
400 the heat 500transfer ability of the nanofluids. As
100 200 300 400 500 the difference of fluid thermal
shown
6000 in Figure 6, in orderRe
to comprehensively consider
h(W/m2K)

conductivity caused by different Re volume fractions of nanoparticles, the Nusselt number

Figure 4. Average temperature distributions of the fluid-solid surface.
Figure5500
4. Average
is used temperatureindex
as the evaluation distributions of the fluid-solid
for comparative surface.
analysis, and the same conclusion can be
obtained.
5000
7000 α=0.5%
7000
4500 α=5%
6500
6500
4000
6000 100 200 300 400 500
h(W/m2K)

6000
h(W/m2K)

Re
5500
Figure 5500
5. Effects of nanoparticle volume fraction on heat transfer coefficients.
5000
5000 α=0.5%
α=0.5%
4500
9.5 α=5%
4500 α=5%
4000
9.0
4000
100 200
100 α=0.5%300300
200
400
400
500
500
8.5 α=5%
Re
Re
Figure 5. Effects of nanoparticle volume fraction on heat transfer coefficients.
Nu

Figure
Figure8.05.
5. Effects of of
Effects nanoparticle volume
nanoparticle fraction
volume onon
fraction heat transfer
heat coefficients.
transfer coefficients.

7.5
9.5
9.5
7.0
9.0
9.0
6.5 α=0.5%
8.5 100 200α=0.5% 300 400 500
8.5 α=5%
α=5% Re
Nu

8.0
Nu

Figure 6. 8.0
Influence of nanoparticle volume fraction on Nusselt number.
7.5
7.5
7.0
7.0
6.5
6.5
100 200 300 400 500
100 200 300 400 500
Re
Re
Figure 6. Influence
Figure of nanoparticle
6. Influence volume
of nanoparticle fraction
volume on Nusselt
fraction number.
on Nusselt number.
Figure 6. Influence of nanoparticle volume fraction on Nusselt number.
 Micromachines 2021, 12, 868 8 of 14

Figure 7 shows the distribution of the pressure difference between the inlet and
outlet of the microchannel with different volume fractions. The drag loss of nanofluids
through microchannels increases with the increase of the Reynolds number. However, the
drag loss of nanofluids with a nanoparticle volume fraction of 5% is greater than that of
nanofluids with a nanoparticle volume fraction of 0.5%. At the same time, because the
volume fraction of nanoparticles affects the density and dynamic viscosity of nanofluids,
the velocity of the microchannel inlet must be adjusted to keep the same Reynolds number.
As shown in Figure 8, with the increase of the Reynolds number, the inlet velocity of the
5% nanofluid is greater than that of the 0.5% nanofluid. Pumping power is introduced
as the evaluation index to reasonably evaluate the synergistic effect of inlet velocity and
resistance loss. Its physical meaning is the external work required for nanofluids to pass
through microchannels. Pumping power p is expressed as:
Micromachines 2021, 12, x FOR PEER REVIEW 9 of 14
Micromachines 2021, 12, x FOR PEER REVIEW 9 of 14
P = N · Uin · Wh · Wc ·∆ p (15)

N is the number of microchannels in the whole heat sink, and the value is N = 1.
3000
3000
α=0.5%
2500 α=0.5%
2500 α=5%
α=5%
2000
2000
p/Pa
Δp/Pa

1500
1500
Δ

1000
1000
500
500
0
0100 200 300 400 500
100 200 300 400 500
Re
Re
Figure 7. Influence of nanoparticle volume fraction on pressure difference.
Figure 7. Influence of nanoparticle volume fraction on pressure difference.
Figure 7. Influence of nanoparticle volume fraction on pressure difference.

1.00
1.00
α=0.5%
0.75 α=0.5%
0.75 α=5%
α=5%
(m/s)
UUinin(m/s)

0.50
0.50

0.25
0.25

0.00
0.00100 200 300 400 500
100 200 300 400 500
Re
Re
Figure 8. Effect
Figure of of
Effect nanoparticle volume
nanoparticle fraction
volume onon
fraction inlet velocity.
inlet velocity.
Figure 8. 8.
Effect of nanoparticle volume fraction on inlet velocity.

The variation of pumping power with the Reynolds number is shown in Figure 9.
6 increase of the Reynolds number, the pumping power of the nanofluid with a
With the
6
nanoparticle volume fraction of 5% is significantly higher than that of the nanofluid with a
5
5
nanoparticle volume fraction of 0.5%. This result shows that the heat transfer performance
cannot4be improved α by increasing the nanoparticle volume fraction because doing so
=0.5%
α=0.5%consumption of the whole system. At the same time, a high
greatly 4increases the power
P/mW

α=5%
P/mW

α=5%
volume33fraction renders the nanoparticles in nanofluids unable to maintain a stable and
α=0.0%
α=0.0%
2
2
1
1
0
0
 0.50

Uin
0.25

Micromachines 2021, 12, 868 0.00 9 of 14

100 200 300 400 500
Re
Figureuniform
8. Effect of nanoparticle
suspension volume
state [23].fraction on inlet
Therefore, in velocity.
engineering applications, the nanoparticle
volume fraction in nanofluids needs to be maintained at a low level.

5
P/mW 4 α=0.5%
α=5%
3
α=0.0%
2

0
100 200 300 400 500
Re
FigureFigure
9. Effects of nanoparticle
9. Effects volume
of nanoparticle fraction
volume on pumping
fraction power.
on pumping power.

3.2. Influence
3.2. Influence of the of the Aspect
Aspect Ratio Ratio of Microchannels
of Microchannels
This work
This work studies
studies nanofluids
nanofluids withwith a nanoparticle
a nanoparticle volume
volume fraction
fraction of 5%.
of 5%. TheThe different
differ-
ent aspect ratios (HW) of microchannels can be obtained by changing their height (Whh).). The
aspect ratios (HW) of microchannels can be obtained by changing their height (W
aspectratios
The aspect ratiosofofthe
the five
five microchannel
microchannel heat heat sink
sink models
models are
are 10,
10, 20,
20, 30,
30,40
40and
and50.
50.AAthree-
dimensional simulation of flow and heat transfer under different
three-dimensional simulation of flow and heat transfer under different Reynolds numbers Reynolds numbers is
conducted, and the flow and heat transfer characteristics of nanofluid
is conducted, and the flow and heat transfer characteristics of nanofluid microchannels microchannels with
with different
different aspect
aspect ratios
ratiosareareanalyzed
analyzedand andcompared.
compared.AsAsshown
shown in Figure
in Figure10, the
10, pressure
the
difference between the inlet and the outlet increases with the increase of the Reynolds
number under different aspect ratios of the microchannel heat sink. The pressure difference
decreases with the increase of the aspect ratio at the same Reynolds number. When the
aspect ratio is 20–50, the pressure difference is not obvious. This result shows that the
increase of the aspect ratio of the microchannel does not greatly enhance the flow resistance
loss of nanofluids under the parameters studied in this work. Figure 11 shows the variation
of the resistance coefficient f of the microchannel with the Reynolds number. The simulation
Micromachines 2021, 12, x FOR PEER REVIEW 11 of 14
results of the five different aspect ratios show that the drag coefficient decreases with the
increase of the Reynolds number. Moreover, the differences in the drag coefficients caused
by different aspect ratios decrease with the increase of the Reynolds number.

2000
HW=10
HW=20
1500
HW=30
HW=40
Δp/Pa

1000 HW=50

500

0
100 200 300 400 500
Re
Figure
Figure10.10.
Distributions ofof
Distributions pressure difference
pressure between
difference the
between inlet
the and
inlet the
and outlet
the ofof
outlet the microchannel
the microchannel
heat sink with different aspect ratios.
heat sink with different aspect ratios.

0.25
HW=10
0.20 HW=20
HW=30
0.15 HW=40
HW=50
f

0.10
 2000
0 HW=10
100 200
HW=20 300 400 500
1500 Re
HW=30
HW=40
Figure 10. Distributions of pressure difference between the inlet and the outlet of the microchannel

Δp/Pa
Micromachines 2021, 12, 868 10 of 14
1000with different
heat sink HW=50
aspect ratios.

500
0.25
HW=10
0
0.20 HW=20
100 200 300 400 500
HW=30
0.15 Re HW=40
Figure 10. Distributions of pressure difference between the inlet and the outlet of the microchannel
HW=50
f
heat sink0.10
with different aspect ratios.

0.05
0.25
0.00 HW=10
0.20 100 200 300 400
HW=20 500
Re HW=30
0.15
Figure 11. Distributions of friction coefficients. HW=40
Figure 11. Distributions of friction coefficients.
HW=50
f

0.10
Figure 12 shows the variation of the average temperature at the bottom of the mi-
355
crochannel heat sink with the Reynolds numbers at different aspect ratios. The results
show0.05that the mean bottom temperature, corresponding
HW=10 to the five microchannel heat
sinks,350
decreases with the increase of the Reynolds HW=20number and that increasing the aspect
0.00
ratio of the microchannel can reduce the bottom HW=30temperature of the heat sink. However,
345100 200 300 400 500
when the aspect ratio exceeds 20, the decreaseHW=40 of the heat sink’s bottom temperature drops,
Re
Tb/K

caused340by an increase in the aspect ratio of the HW=50

microchannel. Furthermore, the average
Figure 11. Distributions
temperature valuesofatfriction coefficients.
the bottom of the three microchannels with aspect ratios 30, 40, and
335
50 almost coincide.
330
355
325 HW=10
350 100 200 300 400HW=20 500
345
Re HW=30
Figure 12. Temperature distributions of the heat sink HW=40
bottom with different aspect ratios.
Tb/K

340 HW=50
335

330

325
100 200 300 400 500
Re
Figure 12. Temperature
Figure distributions
12. Temperature of the
distributions heatheat
of the sinksink
bottom with
bottom different
with aspect
different ratios.
aspect ratios.

The average temperature distribution of the fluid-solid surface relative to the heat
transfer in the microchannels with different aspect ratios is shown in Figure 13. With the
increase of the Reynolds number, the average temperature of the heat transfer surface
decreases. When the aspect ratio HW increases from 10 to 20, the corresponding tempera-
ture drop is 16.8 K at an Re of 100. When the aspect ratio HW increases from 20 to 30, the
corresponding temperature drop is 5.7 K. When the aspect ratio HW increases from 30 to 40,
the corresponding temperature drop decreases to 2.9 K. When the aspect ratio HW increases
from 40 to 50, the corresponding temperature drop further decreases to 1.8 K. As shown in
Figure 14, when the aspect ratio HW increases from 10 to 30, the Nusselt number increases.
When the aspect ratio further increases from 30 to 50, the Nusselt number does not increase
significantly. This result indicates that the increase of the aspect ratio does not significantly
improve the heat transfer performance of the microchannel heat sink in this range. The
above analysis shows that the change of the aspect ratio of the microchannel affects its heat
transfer performance and resistance characteristics. In this work, the comprehensive heat
 Micromachines 2021, 12, 868 11 of 14

transfer performance index is used to quantitatively evaluate the synergistic effect. It is

defined as:
( Nu/Nu0 )
Micromachines 2021, 12, x FOR PEER REVIEW η= 12 of 14 (16)
Micromachines 2021, 12, x FOR PEER REVIEW ( f / f 0 )1/3 12 of 14

where f 0 is the resistance coefficient of the microchannel with aspect ratio HW = 10 and
Nu0 is the average Nusselt number of the microchannel with aspect ratio HW = 10.
340
340 HW=10
HW=10
HW=20
330 HW=20
HW=30
330 HW=30
HW=40
HW=40
HW=50
Tc/K

320 HW=50
Tc/K

320

310
310

300
300
100 200 300 400 500
100 200 Re300 400 500
Re
Figure 13. Temperature distributions of fluid-solid surface in microchannels with different aspect
Figure 13.13.
Figure
ratios. Temperature distributions
Temperature of fluid-solid
distributions surface
of fluid-solid in microchannels
surface with
in microchannels different
with aspect
different aspect
ratios.
ratios.

10
10

9
9
Nu

8 HW=10
Nu

8 HW=10
HW=20
HW=20
HW=30
7
7 HW=30
HW=40
HW=40
HW=50
6 HW=50
6
100 200 300 400 500
100 200 Re300 400 500
Re
Figure 14. Nusselt
Figure number
14. Nusselt distributions
number of microchannels
distributions with different
of microchannels aspect
with different ratios.ratios.
aspect
Figure 14. Nusselt number distributions of microchannels with different aspect ratios.
Figure 15 shows the variations of the comprehensive heat transfer performance pa-
1.06
rameters of the microchannel heat sinks with different aspect ratios and given different
1.06 numbers. In the range of the Reynolds numbers studied in this work, the com-
Reynolds
HW20
prehensive heat transfer performance parameters
HW20 are greater than 1. The results indicate
1.05 HW30
for the microchannel heat sink with HW
that1.05 = 10, increasing the aspect ratio can improve
HW30
HW40
its comprehensive heat transfer performance. When the aspect ratio is increased to 30, the
HW40
HW50
1.04
comprehensive heat transfer performance ofHW50
the microchannel heat sink does not continue
η

1.04
η

to improve.
1.03
1.03

1.02
1.02
100 200 300 400 500
100 200 Re300 400 500
Re
Figure 15. Comparisons of comprehensive heat transfer performances of microchannels with differ-
entFigure
aspect15. Comparisons of comprehensive heat transfer performances of microchannels with differ-
ratios.
ent aspect ratios.
4. Conclusions
4. Conclusions
In this work, a microchannel heat sink is studied on the basis of computational fluid
In this
dynamics. Thework,
flow aand
microchannel heat
heat transfer of sink
two is studied on
nanofluids the different
with basis of computational fluid
volume fractions
dynamics. The flow and heat transfer of two nanofluids with different volume
and five microchannel heat sinks with different aspect ratios in the Reynolds number fractions
 HW=40
HW=50
6
100 200 300 400 500
Re
Micromachines 2021, 12, 868 12 of 14
Figure 14. Nusselt number distributions of microchannels with different aspect ratios.

1.06
HW20
1.05 HW30
HW40
HW50
1.04
η
1.03

1.02
100 200 300 400 500
Re
Figure 15. Comparisons
Figure of comprehensive
15. Comparisons heat
of comprehensive transfer
heat performances
transfer ofof
performances microchannels
microchannelswith
withdiffer-
different
ent aspect
aspect ratios.
ratios.

4. Conclusions
4. Conclusions
In this
In this work,
work, a microchannel
a microchannel heatheat
sinksink is studied
is studied on on
thethe basis
basis of computational
of computational fluid
fluid
dynamics.
dynamics. TheTheflowflow
andand
heatheat transfer
transfer of two
of two nanofluids
nanofluids with
with different
different volume
volume fractions
fractions
andandfivefive microchannelheat
microchannel heatsinks
sinkswith
with different aspect
aspectratios
ratiosininthe Reynolds
the Reynolds number
numberrange
of 100–500 are simulated. The flow and heat transfer characteristics
range of 100–500 are simulated. The flow and heat transfer characteristics of the of the microchannel
heat sinks are compared, and the optimal parameters of the aspect ratio are analyzed. The
conclusions are as follows:
(1) Increasing the volume fraction of nanoparticles can effectively reduce the average
temperature of the heat transfer surface and improve the heat transfer capability
of nanofluids. However, because of the dual increase of the inlet velocity and flow
resistance, the power consumption of the whole system increases greatly;
(2) Increasing the aspect ratio of the microchannel does not cause significant flow resis-
tance loss, and the resistance coefficient of the microchannel tends to be consistent
with the increase of the Reynolds number at different aspect ratios;
(3) Increasing the aspect ratio of the microchannel can reduce the temperature of the heat
sink. When the aspect ratio exceeds 30, the average temperature at the bottom of the
microchannel does not decrease, and the heat transfer coefficient does not increase;
(4) In the range of the parameters studied in this paper, the aspect ratio of the mi-
crochannel heat sink with a thickness of 6 mm has an optimal value. Based on the
comprehensive heat transfer performance parameters, the optimal value of the aspect
ratio of the microchannel heat sink is 30.
This study shows that the aspect ratio of the heat sink has a significant impact on
the heat transfer performance of a microchannel, and there is an optimal value in the
range of Reynolds numbers under the condition of a given thickness. Further research
will be carried out for different thicknesses of heat sinks under different Reynolds number
conditions to obtain a universal empirical formula for guiding engineering practice.

Author Contributions: Conceptualization, H.W. and S.Z.; methodology, H.W.; software, H.W.;
validation, S.Z.; formal analysis, S.Z.; investigation, S.Z.; resources, S.Z.; writing—original draft
preparation, H.W.; writing—review and editing, H.W.; visualization, H.W.; supervision, S.Z.; funding
acquisition, S.Z. All authors have read and agreed to the published version of the manuscript.
Funding: The authors acknowledge the support of the 14th batch High-level Talents Project for “Six
Talents Peak” (Grant No. XCL-092), the Province Postdoctoral Foundation of Jiangsu (1501164B),
the Technical Innovation Nurturing Foundation of Yangzhou University (2017CXJ024), China Post-
doctoral Science Foundation (2016M600447), Yangzhou Innovative Capacity Building Plan Project
(YZ2017275) and Yangzhou University Science Foundation Project (x20180290).
Conflicts of Interest: The authors declare no conflict of interest.
Micromachines 2021, 12, 868 13 of 14

Nomenclature

ui velocity components in direction i

uj velocity components in direction j
xi cartesian coordinate components
xj cartesian coordinate components
p pressure in the flow field
T temperature
ρnf density of nanofluids
µnf dynamic viscosity
Cpf specific heat capacity
λnf thermal conductivity
w corresponding thermophysical properties of the base fluid
p corresponding thermophysical properties of the nanoparticles
α volume fraction of nanoparticles
n shape factor of nanoparticles
Uin fluid inlet velocity
q heat flux
Nu Nusselt number
N the number of microchannels
Ts the temperature of solid region
λs the thermal conductivity of the solid region.

References
1. Wu, R.; Hong, T.; Cheng, Q.; Zou, H.; Fan, Y.; Luo, X. Thermal modeling and comparative analysis of jet impingement liquid
cooling for high power electronics. Int. J. Heat Mass Transf. 2019, 137, 42–51. [CrossRef]
2. Choi, T.J.; Kim, S.H.; Jang, S.P.; Yang, D.J.; Byeon, Y.M. Heat transfer enhancement of a radiator with mass-producing nanofluids
(EG/water-based Al2 O3 nanofluids) for cooling a 100 kW high power system. Appl. Therm. Eng. 2020, 180, 115780. [CrossRef]
3. Tuckerman, D.B.; Pease, R. High-performance heat sinking for VLSI. IEEE Electron. Device Lett. 1981, 2, 126–129. [CrossRef]
4. Choi, S.; Eastman, J.A. Enhancing thermal conductivity of fluids with nanoparticles. In Proceedings of the 1995 International
Mechanical Engineering Congress and Exhibition, San Francisco, CA, USA, 12–17 November 1995.
5. Koo, J.; Kleinstreuer, C. Laminar nanofluid flow in microheat-sinks. Int. J. Heat Mass Transf. 2005, 48, 2652–2661. [CrossRef]
6. Apmann, K.; Fulmer, R.; Soto, A.; Vafaei, S. Thermal conductivity and viscosity: Review and optimization of effects of nanoparti-
cles. Materials 2021, 14, 1291–1362. [CrossRef] [PubMed]
7. Malvandi, A.; Safaei, M.R.; Kaffash, M.H.; Ganji, D.D. MHD mixed convection in a vertical annulus filled with Al2 O3 -water
nanofluid considering nanoparticle migration. J. Magn. Magn. Mater. 2015, 3, 296–306. [CrossRef]
8. Goodarzi, M.; Amiri, A.; Goodarzi, M.S.; Safaei, M.R.; Karimipour, A.; Languri, E.M.; Dahari, M. Investigation of heat transfer
and pressure drop of a contour flow corrugated plate heat exchanger using MWCNT based nanofluids. Int. Commun. Heat Mass
Transf. 2015, 66, 172–179. [CrossRef]
9. Vinoth, R.; Kumar, D.S. Experimental investigation on heat transfer characteristics of an oblique finned microchannel heat sink
with different channel cross sections. Heat Mass Transfer 2018, 54, 3809–3817. [CrossRef]
10. Alfaryjat, A.A.; Mohammed, H.A.; Adam, N.M.; Ariffin, M.K.A.; Najafabadi, M.I. Influence of geometrical parameters of
hexagonal, circular, and rhombus microchannel heat sinks on the thermohydraulic characteristics. Int. Commun. Heat Mass Transf.
2014, 52, 121–131. [CrossRef]
11. Ahmed, H.E.; Ahmed, M.I. Optimum thermal design of triangular, trapezoidal and rectangular grooved microchannel heat sinks.
Int. Commun. Heat Mass Transf. 2015, 66, 47–57. [CrossRef]
12. KumaR, P. Numerical investigation of fluid flow and heat transfer in trapezoidal microchannel with groove structure. Int. J.
Therm. Sci. 2019, 136, 33–43. [CrossRef]
13. Farsad, E.; Abbasi, S.P.; Zabihi, M.S.; Sabbaghzadeh, J. Numerical simulation of heat transfer in a micro channel heat sinks using
nanofluids. Heat Mass Transf. 2010, 47, 479–490. [CrossRef]
14. Shi, X.; Li, S.; Wei, Y.; Gao, Y. Multi objective optimization of structural parameters of nanofluid rectangular microchannel heat
sink. J. Xi’an Jiaotong Univ. 2018, 52, 61–66.
15. Ghale, Z.Y.; Haghshenasfard, M.; Esfahany, M.N. Investigation of nanofluids heat transfer in a ribbed microchannel heat sink
using singlephase and multiphase CFD models. Int. Commun. Heat Mass Transf. 2015, 68, 122–129. [CrossRef]
16. Manca, O.; Nardini, S.; Ricci, D. A numerical study of nanofluid forced convection in ribbed channels. Appl. Therm. Eng. 2012, 37,
280–292. [CrossRef]
17. Xie, H.; Fujii, M.; Zhang, X. Effect of interfacial nanolayer on the effective thermal conductivity of nanoparticle-fluid mixture. Int.
J. Heat Mass Transf. 2004, 48, 2926–2932. [CrossRef]
Micromachines 2021, 12, 868 14 of 14

18. Gkountas, A.A.; Benos, L.T.; Nikas, K.S.; Sarris, I.E. Heat transfer improvement by an Al2O3-water nanofluid coolant in
printed-circuit heat exchangers of supercritical CO2 Brayton cycle. Therm. Sci. Eng. Prog. 2020, 20, 1025–1051. [CrossRef]
19. Sohel, M.R.; Khaleduzzaman, S.S.; Saidur, R.; Hepbasli, A.; Sabri, M.F.M.; Mahbubul, I.M. An experimental investigation of
heat transfer enhancement of a minichannel heat sink using Al2 O3 –H2 O nanofluid. Int. J. Heat Mass Transf. 2014, 74, 164–172.
[CrossRef]
20. Mohammadpour, J.; Salehi, F.; Sheikholeslami, M.; Masoudi, M.; Lee, A. Optimization of nanofluid heat transfer in a microchannel
heat sink with multiple synthetic jets based on CFD-DPM and MLA. Int. J. Therm. Sci. 2021, 167, 107008. [CrossRef]
21. Khanafer, K.; Vafai, K. A critical syn study of thermophysical characteristics of nanofluids. Int. J. Heat Mass Transf. 2011, 54,
4410–4428. [CrossRef]
22. Chai, L.; Xia, G.; Wang, L.; Zhou, M.; Cui, Z. Heat transfer enhancement in microchannel heat sinks with periodic expansion-
constriction cross-sections. Int. J. Heat Mass Transf. 2013, 62, 741–751. [CrossRef]
23. Lee, J.; Mudawar, I. Assessment of the effectiveness of nanofluids for single-phase and two-phase heat transfer in micro-channels.
Int. J. Heat Mass Transf. 2007, 50, 452–463. [CrossRef]